Research Collection

Doctoral Thesis

Global population genetics of Spongospora subterranea F. SP. subterranea, the plasmodiophorid pathogen causing powdery scab of potato and its impact on disease management

Author(s): Gau, Rebecca D.

Publication Date: 2012

Permanent Link: https://doi.org/10.3929/ethz-a-007593856

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.

ETH Library DISS. ETH No. 20653

GLOBAL POPULATION GENETICS OF SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA, THE PLASMODIOPHORID PATHOGEN CAUSING POWDERY SCAB OF POTATO AND ITS IMPACT ON DISEASE MANAGEMENT

REBECCA DOROTHEA GAU

Zürich, 2012

Diss. ETH No. 20653

GLOBAL POPULATION GENETICS OF SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA, THE PLASMODIOPHORID PATHOGEN CAUSING POWDERY SCAB OF POTATO AND ITS IMPACT ON DISEASE MANAGEMENT

A dissertation submitted to the

ETH ZURICH

for the degree of

DOCTOR OF SCIENCES

presented by

REBECCA DOROTHEA GAU

MSc in Genome Based Systems Biology, Bielefeld University, Germany

Born on December 28th, 1982

Citizen of Germany

Accepted on the recommendation of Prof. Dr Bruce A. McDonald, examiner Dr Ueli Merz, co-examiner Prof. Dr Richard E. Falloon, co-examiner Dr Patrick C. Brunner, co-examiner

Zürich, 2012

TABLE OF CONTENTS ABSTRACT ...... 9 ZUSAMMENFASSUNG...... 11 CHAPTER 1: GENERAL INTRODUCTION ...... 13 1.1 PREFACE...... 15 1.2 INVASIVE PLANT PATHOGENS ...... 21 1.3 THE POTATO ...... 25 1.3.1 History of the potato ...... 28 1.3.2 Economic importance ...... 33 1.3.3. Origin of the potato, major subspecies, and cultivars ...... 35 1.3.4 Potato diseases ...... 37 1.4 THE PATHOGEN: SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA ...... 43 1.4.1 Life cycle ...... 43 1.4.2 Phylogeny of Spongospora subterranea f. sp. subterranea ...... 45 1.4.3 Spongospora genetics...... 45 1.4.4 Cultivation ...... 47 1.4.5 Host range ...... 47 1.4.6 Environmental factors ...... 47 1.4.7 Propagation and economic impact of the disease ...... 48 1.4.8 Diagnostics and control methods ...... 49 1.4.9 Resistance against powdery scab in potato ...... 53 1.5 AIM OF THE PRESENTED STUDY ...... 55 1.6 REFERENCES ...... 57 CHAPTER 2: GLOBAL GENETICS AND INVASION HISTORY OF THE POWDERY SCAB PATHOGEN, SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA ...... 69 2.1 ABSTRACT ...... 71 2.2 INTRODUCTION ...... 73 2.3 MATERIALS AND METHODS...... 79 2.3.1 Samples and DNA extraction ...... 79 2.3.2 Nucleotide sequences and phylogenetic reconstruction ...... 80 2.3.3 Microsatellite library construction ...... 81 2.3.4 Microsatellite analyses and population structure ...... 81 2.3.5 Inferring migration history ...... 83 2.4 RESULTS ...... 85

2.4.1 Phylogeny ...... 85 2.4.2 Clonal and genetic diversity ...... 86 2.4.3 Population structure ...... 89 2.4.4 Inferred migration history ...... 91 2.5 DISCUSSION ...... 93 2.5.1 Role of hosts, inoculum and the potential of cryptic taxa ...... 93 2.5.2 Invasion scenario of Sss and evidence for a bridgehead effect ...... 94 2.5.3 Global trade of seed potatoes reflects recent migration patterns of the pathogen ...... 96 2.5.4 Risk assessment for Sss, and comparison with P. infestans ...... 96 2.5.5 Strategies for controlling Sss and potato breeding ...... 97 2.6 ACKNOWLEDGEMENTS ...... 99 2.7 SUPPLEMENTARY MATERIAL ...... 101 2.8 REFERENCES ...... 113 CHAPTER 3: CROSS-INFECTION POTENTIAL OF SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA ROOT GALL AND TUBER LESION INOCULUM FROM DIFFERENT POTATO (SOLANUM SPP.) HOSTS ... 119 3.1 ABSTRACT ...... 121 3.2 INTRODUCTION ...... 123 3.3 MATERIAL AND METHODS ...... 127 3.3.1 Cultivation and preparation of tissue cultured potato plantlets ...... 127 3.3.2 Nutrient solution (NS) ...... 127 3.3.3 Preparation of inoculum ...... 128 3.3.4 Cross-inoculation experiment ...... 128 3.3.5 Molecular methods ...... 128 3.4 RESULTS ...... 131 3.4.1 Cross inoculation experiment ...... 131 3.4.2 Molecular analyses ...... 132 3.5 DISCUSSION ...... 135 3.6 ACKNOWLEDGEMENTS ...... 139 3.7 REFERENCES ...... 141 CHAPTER 4: GENERAL DISCUSSION AND CONCLUSIONS ...... 145 4.1 GENERAL DISCUSSION AND CONCLUSIONS ...... 147 4.1.1 Marker development ...... 147 4.1.3 Sample clusters ...... 148 4.1.4 Influence of the inocula, hosts and environment on disease development ...... 149

4.1.5 Origin and migration patterns of Spongospora subterranea ...... 152 4.1.6 Resistance breeding and quarantine measures ...... 153 4.1.7 Future prospects for Spongospora genetic research ...... 155 4.2 REFERENCES ...... 157 ACKNOWLEDGEMENTS ...... 159 CURRICULUM VITAE ...... Fehler! Textmarke nicht definiert.

ABSTRACT

Spongospora subterranea f. sp. subterranea (Sss), a soilborne biotrophic parasite, causes two characteristic disease symptoms on potato (Solanum tuberosum): lesions on tubers (powdery scab) and root galls. Both of these are economically important diseases and occur in all major potato producing areas of the world.

The resting spores formed by this plasmodiophorid pathogen are highly resistant to environmental stresses and remain infectious for many years in contaminated soils. To reach a nearby host, biflagellate zoospores can emerge from resting spores and swim from a few centimeters up to a few meters, depending on soil type and soil water flow. Furthermore, Spongospora does not form airborne spores, thus it cannot spread over large areas itself. All introductions are likely to be human mediated, in contaminated soil or on infected seed potato tubers.

Resistant potato cultivars or completely effective control agents are not available, thus the only current effective control measure is to avoid further spread of the disease by planting healthy seed tubers into healthy soil without prior powdery scab history. This needs seed potato grading and certification systems on a global scale with high quality standards. A more promising approach would be the development of potato lines resistant to Sss, but the necessary population genetics data of the pathogen used to screen new lines during breeding have been missing.

Microsatellite analysis and DNA sequencing were applied to investigate the population structure, i.e. genetic and genotypic diversity, and invasion history of Spongospora on a global scale. South American collections consistently showed greater genetic diversity than other regions of the world. Additionally, genetic data for the collections of the pathogen from tuber lesions and root galls were clearly distinguished. Historic and recent migration analyses showed that Europe is likely to have been the recent source of Spongospora migrants, whereas the pathogen’s native region is most likely to be in South America, from where it was introduced to Europe, probably on diseased potatoes. The results are descriptive for a bridgehead effect with Europe being the established founding population from where further global introductions were outgoing.

Cross inoculation experiments were carried out with Sss inocula from native and invaded origins, obtained from three different potato host (sub-)species and two different resting spore sources (root galls or tuber lesions). The performance and root galling ability of the Sss inocula on potato plantlets of different subspecies were evaluated. All inocula were able to infect all host (sub-) species. Genotyping using the earlier developed molecular markers was performed with all inocula, harvested galls, and tuber lesions, which were only observed on the potato cultivar ‘Agria’ but from all inocula. The native inocula had two host organ specific genotypes corresponding to their ability to cause either tuber lesions or root galls, while the invasive inocula all had a third genotype in common independent of the resting spore source. Almost all harvested root galls had the European genotype. Tuber lesions were all of the South American lesion genotype. The native root gall genotype was not detected on harvested material. Results suggested that the inocula are a mixture of spores with different genotypes, and that during infection of the host, organ specific selection may take place. Some genotypes of spores are more likely than others to produce root galls or tuber lesions respectively.

The pathogen is already difficult to control and more aggressive strains could be introduced or arise from possible sexual recombination. To avoid this scenario, global genetic diversity must be kept at a low level and exchange of genetic material must be avoided between native and introduced regions. This is best achieved by continuing with appropriate quarantine measures and improving already established systems. Nevertheless, the current global genetic diversity of Sss is low which is advantageous for potato breeders. Resistant lines are likely to provide durable protection, and are likely to be effective in different countries with similar Sss strains.

ZUSAMMENFASSUNG

Spongospora subterranea f. sp. subterranea (Sss) ist ein bodenbürtiger, obligat biotropher Parasit, der zwei charakteristische Krankheitssymptome auf der Kartoffel (Solanum tuberosum) hervorruft: Läsionen auf Knollen (Pulverschorf) und Wurzelgallen. Beide sind wirtschaftlich von Bedeutung und kommen weltweit in allen wichtigen Kartoffel produzierenden Regionen vor.

Die Dauersporen dieses Pathogens, das zu den Plasmodiophoriden gehört, sind extrem resistent gegen umweltbedingte Stressfaktoren, daher bleiben kontaminierte Böden viele Jahre infektiös. In Wirtsnähe können zweifach begeisselte Zoosporen aus den Dauersporen schlüpfen und wenige Zentimeter bis maximal einige Meter schwimmen, je nach Bodentyp und Wasserfluss. Spongospora bildet ausserdem keine mit der Luft übertragbaren Sporen und kann sich nicht selbstständig über weite Flächen ausbreiten. Für Einschleppungen des Pathogens beispielsweise in kontaminierter Erde oder auf infizierten

Saatkartoffeln ist wahrscheinlich der Mensch verantwortlich.

Resistente Kartoffelkultivare oder hochwirksame Kontrollmethoden gibt es nicht, damit ist die Vermeidung der Ausbreitung durch das Anpflanzen gesunder Saatkartoffeln in gesunden Böden ohne Pulverschorfhintergrund derzeit die einzig wirksame Strategie. Dazu sind global einheitliche Saatkontrollen und Zertifizierungssysteme notwendig, die hohen Qualitätsansprüchen genügen. Ein vielversprechender Ansatz wäre die Entwicklung von Sss- resistenten Kartoffellinien, doch die notwendigen populationsgenetischen Daten des Erregers, die zur Züchtung und Auslese notwendig wären, gab es bisher nicht.

Mikrosatellitenanalyse und DNS-Sequenzierung wurden durchgeführt, um die Populationsstruktur, also die genetische und genotypische Vielfalt, und die Invasionsgeschichte von Sss global zu erforschen. Südamerikanische Proben zeigten konsistent grössere genetische Vielfalt als andere Regionen der Welt. Des Weiteren konnten in den genetischen Daten der Pathogenproben Knollenläsionen und Wurzelgallen klar unterschieden werden. Analysen zu historischer und rezenter Migration ergaben, dass Europa wahrscheinlich die rezente Quelle von Sss Migranten ist, das Pathogen jedoch höchstwahrscheinlich in Südamerika heimisch ist. Von dort aus wurde es möglicherweise auf

infizierten Kartoffeln nach Europa eingeschleppt. Die Ergebnisse beschreiben einen Brückenkopfeffekt. In Europa konnte sich die erste Gründerpopulation etablieren, die als Brückenkopf zur globalen Invasion weiterer Regionen dient.

Inokulationsexperimente wurden mit Sss Inokula von heimischen und nicht- heimischen Regionen durchgeführt, die von verschiedenen Kartoffelwirts(sub)spezies und verschiedenen Dauersporentypen (Wurzelgallen oder Knollenläsionen) stammten. Die Effizienz und die Fähigkeit zur Wurzelgallenbildung auf Kartoffelgewebekulturpflanzen verschiedener (Sub)Spezies wurden ausgewertet. Alle Inokula konnten alle Wirtspflanzen infizieren. Genotypisierung wurde mit allen Inokula, geernteten Gallen und Knollenläsionen durchgeführt. Letztere wurden nur auf dem Kultivar „Agria“ beobachtet. Südamerikanische Inokula wiesen jeweils für Knollenläsionen und Wurzelgallen organspezifische Genotypen auf. Nicht heimische Inokula hatten einen dritten Genotyp gemeinsam. Fast alle geernteten Wurzelgallen hatten den nicht heimischen (europäischen) Genotyp, alle Knollenläsionen von Agria zeigten den südamerikanischen Knollenläsionengenotyp. Der südamerikanische Wurzelgallengenotyp wurde nicht mehr beobachtet. Die Ergebnisse legen nahe, dass die Inokula eine Sporenmischung verschiedener Genotypen sind und dass während der Infektion eine organspezifische Selektion stattfindet. Dieses könnte bedeuten, dass je nach Sporengenotyp entweder die Bildung von Wurzelgallen oder Knollenläsionen wahrscheinlicher ist.

Bisher ist das Pathogen schwierig zu kontrollieren und aggressivere Stämme könnten importiert werden oder durch sexuelle Rekombination entstehen. Um dieses zu vermeiden, muss die globale genetische Diversität niedrig gehalten und der Austausch genetischen Materials zischen heimischen und invasiven Regionen vermieden werden. Dazu müssen geeignete, bereits bestehende Quarantänemassnahmen aufrechterhalten und verbessert werden. Dennoch ist die globale genetische Diversität von Sss in invasiven Regionen niedrig, was für Kartoffelzüchter von Vorteil ist. Resistente Linien würden sehr wahrscheinlich dauerhaften Schutz bieten und könnten womöglich zwischen verschiedenen Regionen mit

ähnlichen Sss-Stämmen ausgetauscht werden ohne ihre Schutzfunktion zu verlieren.

CHAPTER 1: GENERAL INTRODUCTION

1.1 PREFACE

The major part of the living environment of earth consists of plants. Directly or indirectly, all food animals or humans depend on is made up of plants. They are the only higher organisms, which can convert light energy and abiotic soil substances into biotic substances like carbohydrates, fat, and proteins (Agrios, 2005). These can then be taken up by other life forms.

To allow healthy plant development, an environment with optimal light, temperature, water, and nutrient conditions is needed. In contrast, sickness negatively influences plant growth and productivity. A plant is considered diseased, if its physiological performance is restricted in comparison to its genetic potential (Agrios, 2005).

Causes of disease in plants are very similar to those of humans or animals. They include biotic causes such as pathogenic microorganisms (viruses, bacteria, fungi, protozoa, and nematodes), but also abiotic environmental factors such as shortage or excess of nutrients, moisture, light or presence of toxic pollutants in the soil or air (Agrios, 2005;

Oerke, 2006, Fig. 1).

Boote et al. (1983) suggested classification of pests by their impact on host plants: stand reducers, photosynthetic rate reducers, leaf senescence accelerators, light stealers, assimilate sappers, or tissue consumers. This categorization illustrated that the presence of plant diseases in agro-ecosystems can either drastically reduce the yield of plants by destroying parts of, or whole plants prior to harvesting, or totally preventing plant cultivation and growth in the area they occur.

Figure 1: Abiotic and biotic factors causing crop losses (Oerke 2006).

Worldwide crop losses due to diseases, insects and weeds together are estimated between 31 % and 42 % annually (Oerke, 1994), without taking environmental factors into account. Two loss rates are usually differentiated when describing crop losses. The potential loss is calculated as the difference in loss between unprotected crops versus protected crops in a no-loss scenario. The actual loss is the loss occurring despite crop protection treatments (Fig. 2; Oerke, 2006).

Figure 2: Crop losses and yield levels (Oerke 2006).

The losses are usually greater in developing countries than developed countries and they do not inform about the consequences for large populations. Yield losses due to plant diseases can result in human malnutrition, hunger, and starvation, but also in lost income or jobs and abandoning of farms and migration to large cities to look for new occupations (Agrios, 2005; Zadoks & Schein, 1979).

Plant diseases also reduce profit margins. If diseases occur in a field or area, the choice of crops which can be grown is limited by control measures like crop rotation. To protect crops from disease, time and money has to be spent on physical, biological and pesticide chemical treatments, but these practices can have damaging effects on the environment (Oerke, 2006; Agrios, 2005).

Plant pathology is a science that studies organisms and environmental factors, which can lead to plant diseases, the manner how these factors cause disease, and of the strategies for controlling plant diseases. This science discipline is often combined with several fields or biology and natural sciences, such as entomology, weed science, ecology and genetics. The practical goal is to protect cash crops and the food available for humans and animals from the deleterious effects of plant diseases. Challenging to plant pathology is to carefully balance all factors, while improving food productivity and quality, and at the same time limiting the side effects for living organisms and the environment, in a world with an increasing population while many resources and arable land are decreasing (Agrios, 2005).

The field of molecular plant pathology is established since the 1980s, but began much earlier in 1935 when W. Stanley isolated the tobacco mosaic virus as a crystalline protein, which he believed to be infectious (Agrios, 2005). Two years later, it was shown that the protein also contained RNA, which was proven to cause disease in tobacco in 1956 by Gierrer & Schramm (Agrios, 2005). Meanwhile, the one-gene-one-protein hypothesis was formulated by Beadle and Tatum in 1941 and in the following year, the gene-for-gene concept was published by Flor. In 1953, Watson and Crick revealed their double helix model of the DNA (Agrios, 2005). Since then, several other important discoveries have been made: The mode of infection of Agrobacterium tumefaciens with its characteristic tumor inducing (Ti-) plasmid was described in the 1970s. Albersheim observed in 1984 that the cell wall proteins of Phytophthora megasperma act as elicitors in plants to induce defense

mechanisms (Agrios, 2005). In 1986 the first transgenic plants resistant to tobacco mosaic virus were developed, and in the 1990s molecular methods were finally established in plant pathology, with the discoveries of the mechanisms of systemic acquired resistance (SAR), leucine rich repeats (LRRs), and ribonucleic acid (RNA) silencing (Agrios, 2005).

The development and improvement of diagnostic tools have largely contributed to molecular plant pathology (Agrios, 2005). The availability of monoclonal antibodies, radioactive labeling, analysis methods for isozymes or fatty acid profiles, and of nucleic acid fragments, have all greatly assisted knowledge advances. Very commonly used and sensitive methods involve blotting methods with antibodies, like the enzyme-linked immunosorbent assay (ELISA) and polymerase chain reaction (PCR), with very many application possibilities for detection of plant pathogens or quantification of compounds produced during parasite- host interactions.

Detection and quantification of plant pathogens is crucial to assist development of strategies for effective disease management and control. Living in changing agro-ecosystem environments, pathogens must constantly adapt to survive (McDonald, 1997). Population genetic approaches are important for development of strategies for long term protection against evolving plant pathogens.

Population genetics have been successfully applied to plant pathogens to empirically observe and describe which factors play major roles in pathogen evolution, and how evolutionary forces interact to determine the genetic composition and evolutionary potential of pathogen populations (McDonald, 1997). These factors include the five evolutionary forces: Mutation, mating systems, gene flow or migration, population size, and selection (McDonald & Linde, 2002).

As a source of genetic variation, mutation directly leads to changes in the deoxyribonucleic acid (DNA). The process of mutation can result in the creation of new virulent strains of plant pathogens which are able to break major gene resistances, or it may create strains with greater pathogenicity that can overcome quantitative resistance (McDonald & Linde, 2002).

The distribution of gene diversity within and among individuals in a population is affected by reproduction, which may be asexual, sexual or mixed. The mating system may vary from strict inbreeding to obligate outcrossing and is relevant only for sexually reproducing species (McDonald & Linde, 2002).

Exchange of alleles or individuals among geographically separated populations, including the movement of virulent mutant alleles in different field populations, is mediated by a process called gene flow. Generally, high degrees of gene flow are thought to result in greater genetic diversity in pathogens than low degrees of gene flow (McDonald & Linde, 2002).

The presence of mutants is influenced by the size of the population. More mutants are present in large populations, than in small populations, due to generally low mutation rates (McDonald & Linde, 2002).

Finally, selection is the main force to drive changes in frequency of mutant alleles (McDonald & Linde, 2002).

To better understand and to fight global pathogens, the forces that impact on the evolutionary potential of the populations must be understood. Any interaction between populations or individuals may influence the genetic structure and the risks that particular pathogens impose in agricultural ecosystems around the world. Population genetic studies may help breeders to select for the most durable resistance in the host plants, allow prediction of the evolution of plant pathogens, assist risk assessments, and promote integrated forms of disease management.

1.2 INVASIVE PLANT PATHOGENS

Human mediated biological invasions have occurred repeatedly throughout history. Intentionally and unintentionally, organisms have been taken to new areas beyond their native regions (Valtueña et al., 2011), or ecosystems have been modified in a way that simplifies invasions (Moore, 2005). Many important crops were introduced to areas in the world which are far away from their original habitat, usuall because of their benefits for food and fiber production (Bhagwat et al., 2008; Lashermes et al., 1997; Pimentel et al., 2001). Plant yield makes up more than 98 % of the world’s food supply, and has a commercial value of 5 trillion $US/year (USBC, 1998).

Recurrent global trade with plants and their products can enhance biological invasions, because stowaway pathogens and pests in shipments can be introduced accidentally to new areas (Agrios, 2005; Dybdahl & Storfer, 2003; Montarry et al., 2010). These are considered to be invasive alien species (IAS), meaning that they are “species that are non-native to a particular ecosystem and whose introduction and spread causes, or are likely to cause, socio-cultural, economic or environmental harm or harm to human health.” (Moore, 2005).

Invasive plant pathogen species can cause severe economic losses in crop yield by the destruction of plants, which serve important purposes in forests, nutrition for humans or livestock or are of aesthetic value (González-Varela et al., 2011; Anderson et al., 2004).

The list of examples for invasive plant pathogens whose introduction to new exotic environments was due to human activities is extensive. The most infamous of these is Phytophtora infestans, the pathogen causing the potato late blight. It was first reported in Philadelphia, USA, in 1843 and introduced to Europe in the mid-nineteenth century (Fry et al., 1993). The consequences for economy and society were most devastating in Ireland, where the potato was a staple food. The emerging disease destroyed potato fields in three successive years causing the Irish potato famine (FAO, 2009). More than a million people died from starvation or famine related diseases, and millions of people emigrated from Ireland (Gavrilets et al., 2010). The international distribution of Phytophthora infestans was

most likely facilitated by trade in seed potato, because in an early stage of infection the pathogen is visually undetectable in potato tubers.

Other examples for pathogens which have been introduced my human activities are the wheat stripe rust pathogen Puccinia striiformis f. sp. tritici, which caused severe economic losses for wheat farmers in the United States about 100 years ago (Chen, 2005), and the sudden oak death pathogen Phytophthora ramorum, which has killed most of the North American oak trees (Rizzo et al., 2005).

Not all introduced species have the potential to become invasive, however. It is believed that only few of these can establish, and only a few of those which have established can spread and become pests. Accordingly, Vander Zanden (2005) formulated the “tens rule”, which says that 1 out of 10 introduced species escapes to the wild, 1 out of 10 of the escaping species is able to establish in the wild, and just 1 out of 10 of these can become pests.

Fungal or fungal-like species belong to the most successful invasive plant pathogens, in terms of ability to establish populations in new exotic environments and dispersal over large areas (Vizzini et al., 2009). Despite the importance of these biological invasions, knowledge about the frequency and means of introductions is scarce. Understanding the basic principles is crucial for risk assessment, forecast, prevention and reaction to additional introductions (Desprez-Loustau et al., 2007; González-Varela et al., 2011).

A common approach to expand knowledge is the combination of molecular genetic data, such as that obtained from studies using genetic markers, with ecology. Genetic information can contribute to detect source populations and invasion routes. Furthermore, studies of the genetic structure of founding populations can indicate the reasons for invasion success of the species. Ecological genetics and biological invasions are linked: The genotype of an invasive alien species (IAS) needs to be appropriate to overcome challenges of adaption to the ecological conditions in the new environment, which might be considerably different from the invader’s native region (Lawson Handley et al., 2011).

A long established theory is that the population size and genetic variation of IAS populations are drastically reduced compared to the source populations, which leads to

founder effects during introduction and establishment. These are in theory deleterious to invasion success, because of the genetically limited ability of the IAS to react to selective pressures and environmental stresses. Furthermore, in small populations the chance of inbreeding is increased, which can cause deleterious effects, due to recessive mutations in homozygous individuals. Before expansion and colonization of new areas, a lag phase is expected. In this phase multiple further introductions can occur, which increase the gene pool of the invasive founder population and increase the chance to evolve adaptation (Lawson-Handley et al., 2011).

The ecological concept of bridgehead effects and their detection using ABC analyses has recently been described by Lombaert et al. (2010). This phenomenon leads to rapid adaptive evolution despite strong bottlenecks after just one introduction event. Thus, genetic diversity of founding populations would play a minor role in invasion success; more important is the presence of a fit genotype which can overcome the challenges of an exotic environment. Bridgehead populations are examples of successful invasive plant pathogen species with low genetic variability.

Phytophthora infestans is an example of an invasive plant pathogen which had a clonal genetic structure at first. Striking is its ability to develop different phenotypes depending on climatic and environmental conditions (Fry & Goodwin, 1997). Hence, the thorough study of clonal invasive plant pathogen species in interaction with environmental and host factors is important.

1.3 THE POTATO

Potato (Solanum tuberosum) is a member of the economically important Solanaceae or “nightshade” family, all members of the genus Solanum which includes at least 1000 described species. Among them are tomato, eggplant, and tobacco. The potato is a herbaceous annual plant belonging to the asterid clade of eudicot plants, which represent 25% of flowering plant species (Potato Genome Consortium, 2011; FAO, 2009).

Potato plants grow up to one meter in height and produce stolons (underground stems) that under suitable environmental conditions swell to form tubers, also known as “potatoes”. These serve both as storage organs and a vegetative propagation material which are unique characteristics among the major world food crops (FAO, 2009).

The compound leaves of potato plants form starch which is transported to the ends of the stolons (underground stems). These thicken and form up to twenty tubers on each plant below the soil surface. Soil water and nutrients are necessary for tubers to reach maturity. Shape and size of the tubers vary, but the average weight of a tuber is 300 g. After the growing season, the leaves and stems die and the tubers detach from the stolons. Tubers serve as nutrient stores which help survival, and tubers later regrow and reproduce. Two to 10 buds (“eyes”) circle in a helix around the tuber surface. Under favorable conditions, these may generate shoots which grow into new plants (Fig. 3; FAO, 2009).

Temperature is the main limiting factor to potato production, because cool weather is required for cultivation. Below 10 °C and above 30 °C, tuber growth is strictly inhibited. Daily temperatures ranging from 18 to 20 °C are necessary for optimum yields. As a consequence, early spring is the time for potato planting in temperate regions and late winter in warmer climates. In hot and tropical zones, the growing season spans the coolest months of the year. Mild temperatures in some sub-tropical highlands allow potato cultivation throughout the year and harvesting of tubers within approximately 90 days, in contrast to about 150 days required for tuber maturation in temperate climates like northern Europe. The period of tuber maturation can vary between varieties (FAO, 2009).

Figure 3: A typical potato plant (FAO, 2009).

Potato ranks fourth of the world’s most important food crops after maize, wheat and rice (Potato Genome Consortium, 2011). Potato is the higherst ranked food crop for starch production per hectare, and ranked second place after soybeans for protein production (Oerke, 2006). Raw potato tubers are also rich in micronutrients, including antioxidants, vitamins and minerals that are essential to human health (Burlingame et al., 2009). A medium-size potato contains high levels of potassium and nearly half the daily adult requirement of vitamin C. Potato tubers are also good sources of B vitamins, and minerals such as phosphorus and magnesium (Table 1).

Table 1: Nutrients of one raw potato, including skin, 213 g. Source: The United States National Nutrient Database, data retrieved on 18 June 2012.

Nutrient Type Nutrient Amount Proximates Water 168.99 g Protein 4.3 g Total lipid (fat) 0.19 g Carbohydrate, by difference 37.21 g Fiber, total dietary 4.7 g Sugars 1.66 g Minerals Potassuim 897 mg Phosphorus 121 mg Magnesium 49 mg Iron 1.66 mg Vitamins Vitamin C 42 mg Niacin 2.2 mg Vitamin B6 0.62 mg Thiamine 0.17 mg

Commercially produced potato is clonally propagated, but the underlying mechanisms of tuber initiation and growth, and the evolution of tuber development have remained elusive for long time (FAO, 2009). Analysis of the recently published potato genome indicated that gene family expansion, tissue-specific gene expression, and recruitment of genes to new pathways contributed to the evolution of tuber development (Potato Genome Consortium, 2011).

In general, it was found that the potato genome is highly heterozygous, autotetraploid, and that gene presence/absence variants and other potentially deleterious mutations occur frequently and are a likely cause of inbreeding depression (Potato Genome Consortium, 2011). The potato species grown most throughout the world (Solanum species tuberosum) is believed to contain just a fragment of the genetic diversity found in the four

recognized potato species, with 200 wild species and 5,000 potato varieties still grown in the Andes (FAO, 2009).

1.3.1 HISTORY OF THE POTATO

About 15,000 years ago, humans entered the South American continent (FAO, 2009) 8,000- 10,000 years ago the first communities of hunter gatherers began domesticating wild potato plants growing in the Lake Titicaca region (Ugent, 1970; Ugent et al., 1982; Ugent et al., 1987). This region is situated at 3,800 m above sea level in the Andes on the border between Peru and Bolivia (Fig. 4).

Figure 4: Map of the Lake Titicaca region on the border between Peru and Bolivia (http://en.wikipedia.org/wiki/Lake_Titicaca, retrieved on 4 July 2012).

Although wild potato species occur throughout South America, only cultivators in the Central Andes succeeded in selecting and improving edible potatoes (FAO, 2009). Early Andean farmers cultivated many food crops including tomatoes, beans and maize. Their potato varieties proved particularly suited to the quechua or “valley” zone, which extends at altitudes of from 3,100 to 3,500 m above sea level along the slopes of the Central Andes. At higher altitudes, special traits were required from a crop potato. Bitter potatoes with high alkaloid content were widely grown and detoxified after harvesting, using a technique similar to freeze drying to make potatoes edible (Johns, 1990). Frost-resistant potato species

were also grown, which are still cultivated and survive on the alpine tundra of the puna zone at 4,300 m above sea level. The cultivators expanded into the vast variety of ecological niches in the environment of the Andes and bred several thousand potato varieties, which differed in size, colour, shape, and texture. All varieties have been selected for multiple traits, including resistance to pests, frost, and other stresses (Brush, 1992; Brush et al., 1992; FAO, 2009).

Knowledge of the benefits of specially selected crop varieties for the different zones in the Andes allowed the South American states and civilizations to develop complex crop production, distribution, and consumption systems, which were arranged in a vertical manner: At low altitudes with mild climates, seed crops (especially maize, secondarily quinoa) were cultivated, which were adapted to mild climates. In contrast, at high altitudes with more extreme climates, potatoes and other tubers were grown, which could tolerate the environmental conditions. Common means of transportation were llamas (camelids) to transfer goods between areas (Morris, 1981; Kiple & Ornelas, 2000).

Sophisticated farming techniques were applied to achieve high yields. Irrigation was used, but also hillside terracing, which conserved moisture and soils, and encouraged the selection of new cultivars. Furthermore, rigid, raised, or mounded field technology with elevated soil beds were used for potato growing in the Lake Titicaca region, a technique still applied today. This involves the lining of fields with water canals to keep moisture and offer frost protection. It is estimated, that potato yields of 10 tonnes per hectare were possible (FAO, 2009).

The farming techniques and the vertical food production systems taken together provided food security and allowed for the emergence of pre-Inca civilizations around 500 AD, which built the first large cities in the Central Andes around Lake Titicaca. At their height around 800 AD, these civilizations were able to feed a population of estimated 500,000 people (Hawkes, 1990; Hawkes & Francisco-Ortega, 1992; FAO, 2009).

Between 1000 and 1200 AD, the early civilizations collapsed, which led to a period of turmoil that ended with rise of the Inca civilization in the Cusco valley around 1400. They adopted and improved farming techniques of earlier civilizations and created the largest

state in pre- Columbian America, which reached from present-day Argentina to Colombia (FAO, 2009).

For the Incas, maize had special importance, but the potato was essential to secure food supplies of their empire. Potatoes were stored for short term in the ground, fresh or in processed forms (Morris, 1981). The most common processing technique was traditional freeze-drying: Potatoes were exposed to night frosts, alternating with sunny days, and running water at high altitudes. This technique did not just detoxify alkaloid containing potato varieties (Johns, 1990). Freeze drying also allowed potatoes to be kept over years and transporting them over long distances when they were moved from higher to lower altitudes, and for trade (FAO, 2009).

In the immense network of Incan state storehouses, potato in its freeze dried form called chuño was one of the main food items. Potato tubers were described by early Spanish chroniclers (for example de Acosta, 1590) as items for every-day cooking, and also as an emergency stock after crop failures (FAO, 2009).

With the Spanish conquistadors came the end of the Incas, but potato cultivation continued, because it was a “people’s food” (Cobo, 1653) which had long been established and played an important role in Andean life (Guaman Poma de Ayala & Guaman, 1936). Time periods were expressed by how long it took to cook a pot of potatoes (Coe, 1994), and farmers in some parts of the high Andes still measure land in topo, the area a family needs to grow their potato supply (FAO, 2009).

Potato dissemination began in the Andes in Peru between 1532 and 1572, when the Spanish conquistadors came in search for gold, but the Spanish also took a few landraces of potato back to Europe (FAO, 2009; Kiple & Ornelas, 2000, Spooner et al., 2005a). First evidence of potato growing in Europe dates in 1565 on Spain’s Canary Islands. By 1573, the potato was already grown on the Spanish mainland. About the same time, tubers were sent around Europe as gifts from the Spanish court, for example to the Pope in Rome, from Rome to the papal ambassador in Mons, and from there to Vienna. In London, potatoes were grown in 1597 and soon reached France and the Netherlands (Fig. 5; CIP, 2008). The plant

entered botanical gardens and encyclopedias, but it was admired by European aristocracy mainly for its flowers, while the tubers were considered as food for pigs or the poor.

The first people to appreciate potatoes were sailors; they took the tubers on long ocean voyages, because their consumption proved to obviate scurvy. In this way, the potato reached India, China, Japan, and Australia in the 17th and 18th Centuries (Fig. 5; CIP, 2008). Soon after that, farmers in Ireland realised that the plant grew well in the cool air and moist soils, and started extensive cultivation (Woodham-Smith, 1962; Bourke, 1993; Kinealy, 1995).

Figure 5: Human mediated dissemination of the potato from its native region in the Central Andes in South America first to Europe and later to the rest of the world. 1) South America – Canary Islands: 1567; 2) South America – Spain: 1570; 3) South America – England: 1593; 4) Portugal – India: 1600s; 5) Netherlands – China: 1600s 6) England – Bermuda: 1613; 7) Bermuda – Virginia: 1620; 8) England – Kenya: 1880s; 9) England – New Zealand: 1769; 10) England – Australia: 1787. Map by Centro Internacional de la Papa (CIP, International Potato Center), Peru, 2008.

Challenging to wide acceptance of potato were on the one hand the deep-rooted eating habits of the different nations, and on the other the adaption of an Andean plant to the temperate climate of the northern hemisphere. Only a few genotypes from the rich potato gene pool came from South America to Europe. In the first 250 years, all European potato varieties were derived from the first introductions. A 150 year adaption period was necessary, before adapted varieties began to appear. This point in time coincided with famines across Europe in the 1770s. After a period of hesitation, because of the potato’s infamous reputation of being toxic, causing human diseases, and its names such as “devil’s apple” (Russia), European farmers began to grow the potato on large scales. The potato was declared “edible” with the help of the French scientist Antoine Auguste Parmentier (1737-

1813) and Thomas Jefferson (1743-1826) in the United States who served French fries to guests of the White House. The potato suddenly became a food security crop, when Frederick the Great of Prussia ordered to grow potato as insurance against cereal crop failure (Kiple & Ornelas, 2000). During Napoleon’s wars, the potato became Europe’s food reserve and by 1815 it was already an established staple crop in northern Europe (Nef, 1950). By this time, the strong influences of industrial revolution on the agrarian society in the United Kingdom displaced millions of rural people into the large cities. In urban areas, the potato had traits for becoming the first modern convenience food, which was energy- rich, nutritious, and easy to grow on small plots. It was cheap to purchase and ready to cook without expensive processing. The increased potato consumption during the 19th Century stands in direct relation to higher birth rates, growing populations in Europe and the US and reducing diseases like scurvy and measles.

Due to the potato’s clonal propagation and it’s still reduced gene pool resulting from several bottlenecks during dissemination, the grown varieties were genetically similar, and pests and diseases had strong impacts. Once one plant was infected, diseases could easily spread to the next. In 1844-1845, late blight, caused by Phytophthora infestans, led to severe losses for farmers in continental Europe from Belgium to Russia, but it was of greater consequence for Ireland. Potato supplied a large percentage of the calorie intake, but the disease caused three complete crop losses between 1845 and 1848, which led to the Irish potato famine. As a result, many people died from starvation and famine related diseases, and millions of people emigrated, particularly to the USA (FAO, 2009; Fry et al., 1993).

During European colonialism, settlers, emigrant farmers, and missionaries took the potato to further regions (Fig. 5; CIP, 2008). Already before the 1840s, the Chilean landraces including the long-day variety S. tuberosum ssp. tuberosum were used as a breeding stock for many of the modern varieties, which became predominant in parallel with, or subsequent to, the Irish potato famine (Spooner et al., 2005a; FAO, 2009; Kiple & Ornelas, 2000).

These varieties further intensified potato production in both South America and the rest of the world in the 20th century (Kiple & Ornelas, 2000). The Soviet Union harvested 100 million tonnes/year and after the Second World War, wide areas in Germany and Britain

were used for potato production. Belarus and Poland produced more potatoes than cereals, a situation which remains unchanged (FAOSTAT).

In the 1920s, mechanical potato peelers were invented, which largely contributed to make potato crisps America’s top-selling snack. Millions of dollars were invested by the McDonald brothers to “perfect the French fry” for their restaurant chain. Frozen French fries were produced on large scale by the Canadian company McCain in 1957, which expanded and is now known throughout the world.

In the developing world, potato cultivation began in the 1960s. India and China produced 16 million tonnes in 1960, which has expanded to 100 million tonnes in 2007. Southeast Asia is producing potatoes for the increasing demands of the food industry. In Bangladesh, potato is a valuable winter cash crop, and it is suited to the highlands of Cameroon, Kenya, Malawi and Rwanda. Meanwhile, potato production in Europe is decreasing, but the per capita consumption of potato still remains much greater in developed countries than in developing countries (FAOSTAT; FAO, 2009).

In mountainous Lesotho in 2008, an FAO project for production of virus-free seed tubers motivated many farmers to neglect maize production to grow potato instead. In China, a 30 % increase in potato yields is achievable according to agricultural experts. In the native region of potato, the Andes, the Government of Peru established a national register of Peruvian native potato varieties, to conserve the large genetic diversity of potato, not least to facilitate the breeding of new varieties for future developments in the story of Solanum tuberosum (FAO, 2009).

1.3.2 ECONOMIC IMPORTANCE

Potato (Solanum tuberosum) production has been expanded over the past few decades, which placed the potato among the five most important food crops worldwide (Oerke, 2006), and also elevated it to the world’s most important non-grain food crop, which is central to global food security. Potato occupies a wide eco-geographical range (Hijmans, 2001) and is grown in more than 100 countries under temperate, subtropical and tropical conditions (FAO, 2009).

Worldwide potato production in 2010, reached 324 million tonnes (http://www.fao.org). The largest potato producer by far is China with about 75 million tonnes (Table 2), about half as much is produced in India, followed by the Russian Federation, Ukraine, and the USA. The 20 most important producers, listed in Table 2, consist mostly of European and Middle Eastern states. The worldwide importance of potato is expanding, especially within the developing world.

Table 2: Top 20 potato producers in 2010, Source: FAOSTAT, http://faostat.fao.org/

Ranking Country Tonnes 1 China 74,799,084 2 India 36,577,300 3 Russian Federation 21,140,500 4 Ukraine 18,705,000 5 United States of America 18,016,200 6 Germany 10,201,900 7 Poland 8,765,960 8 Bangladesh 7,930,000 9 Belarus 7,831,110 10 Netherlands 6,843,530 11 France 6,582,190 12 United Kingdom 6,045,000 13 Malawi 4,706,400 14 Turkey 4,548,090 15 Canada 4,421,770 16 Iran (Islamic Republic of) 4,054,490 17 Peru 3,814,370 18 Egypt 3,643,220 19 Brazil 3,595,330 20 Belgium 3,455,800

In the temperate climate of North America and Europe, tuber yields of >40 tonnes/hectare can be reached within 4 months of planting, if sophisticated agricultural practices and irrigation are applied. Average yields in developing countries are less from about 3 to 25 tonnes/hectare (Oerke, 2006; FAO, 2009). This lower production is mostly due to lack of high quality seed or improved cultivars, less application of fertilizer and irrigation, and pest and disease problems.

Harvested potatoes serve a variety of purposes. Estimates of FAO state, that less than 50 % of the world’s potato yield is consumed as fresh product. The remaining potatoes are further processed into food products and ingredients, fed to livestock, processed into industrial starch, and set aside as seed tubers for planting of new crops.

1.3.3. ORIGIN OF THE POTATO, MAJOR SUBSPECIES, AND CULTIVARS

There has been a long controversy about the origin of the first potato introductions to Europe. One group of researcher supports the theory that the origin of the potato is Chile (Chilotaum Group) based on shared morphology and long daylength adaption (Juzepczuk & Bukasov, 1929). Another group supports the theory that the first introduced potatoes came from the Andes (Andigenum Group. Salaman, 1937; Salaman & Hawkes, 1949; Hosaka & Hanneman, 1988; Grun, 1990; Hawkes, 1990). The second theory would imply that there has been a convergent rapid selection of European potato to traits which resemble those of the Chilotanum Group. A further condition of this theory is that the late blight epidemics, which started in 1845 in the United Kingdom and subsequently spread around the world, must have replaced the earlier European cultivars by germplasm from Chile or hybrids with this germplasm (Spooner et al., 2005a).

Microsatellite studies were carried out, and results suggested that early potatoes were introduced to Europe both from the Andes and Chile. However, the Chilean landraces were already used as breeding stock for modern varieties before the 1840s (Spooner et al., 2005a).

Multilocus amplified fragment length polymorphism (AFLP) genotyping of the potato was performed by Spooner et al. (2005b). Their research supported the theory that the modern potato cultivars descend from several landraces which grow in the Andes from western Venezuela to Argentina and the island Chiloé or the Chronos archipel in South Chile in general. The Chilean landraces most likely evolved from Peruvian varieties of the Andes (Solanum tuberosum ssp. andigena), which hybridized with the wild variety Solanum tarijense, found in Bolivia and Argentina.

Today, two very similar groups can be distinguished in the botanical species Solanum tuberosum. Firstly, Solanum tuberosum ssp. andigena, which is adapted to the short-day conditions in the Andes. Secondly, the subspecies Solanum tuberosum ssp. tuberosum, which is adapted to long-day conditions and is also known as the “European” potato. The latter is the potato cultivated most in the world. Other Solanum species are grown mostly in South America and the special climatic zones in the high Andes, and these include Solanum phureja, a diploid, which is grown to produce edible tubers.

The breeding and selecting by Andean farmers continues, thus preserving genetic and biodiversity. Often, several species are cultivated on a single, small plot of land (FAO, 2009). There, also wild varieties are allowed to grow and hybridize with domesticated cultivars to further improve the potato and decrease the risks of crop failures due to disease or stress

(Ugent, 1970; Brush, 1992). Some of these hybrids, which result from crossings of different Solanum species, are tri- and tetraploid and produce sterile progeny in the seed. Therefore, only vegetative propagation is possible. This technique is predominant in all other potato varieties (Oerke, 2006), because the plants and tubers have the same genotypes whereas the seed in the fruit of potatoes is a product of sexual reproduction leading to new genotypes with unknown properties when cultivated. To meet quality standards, usually seed potato tubers are planted.

Despite that the potato grown worldwide belonging to just one botanical species, there are very many different varieties, with flowers of different colors and tubers produced in many colors, shapes, sizes, textures, cooking characteristics and tastes. Varieties are distinguished into several different categories, between fresh consumption and processing varieties or varieties fed to livestock, early-season potatoes (short growing period, and late season (long growing period) varieties, varieties with tubers with robust skins (long storage) or non-robust skins (perishable), and varieties with high starch content for industrial starch production. Preferred and registered cultivars differ between countries. An overview of Swiss varieties is given in Fig. 6.

Of the total potatoes produced in Switzerland in 2011, 43.5 % were used for human consumption, 36.5 % were used for industrial purposes, 10 % were processed for livestock feed and the rest were used as either as seed or fed fresh to livestock (Swisspatat, http://www.kartoffel.ch/, information retrieved on 12 July 2012).

Figure 6: Swiss potato varieties 2011. Size of pie slices is proportional to the area of land a variety occupies corresponding to the total area designated for potato cultivation in Switzerland. Source: Swisspatat, information retrieved on 12 July 2012.

1.3.4 POTATO DISEASES

Despite its limited genetic variation, the potato is highly adaptable. Often, it produces good yields, without ideal soil and growing conditions. Nevertheless, its genetic makeup makes it vulnerable to many pests and diseases. Worldwide totals of actual losses due to pests, diseases and weeds were estimated by Oerke (2006). According to his calculations, there are losses of 14 % due to pathogens, 7 % due to virus infections, 11 % due to animal pests, and 8 % lost due to weeds. He further estimated that without protection measures, about 75 % of the world’s potato production would be lost.

For this reason, farmers do not generally grow potatoes on the same fields in successive years, to avoid the build-up of pathogens in the soil. Instead, potatoes are cultivated in rotations of three or more years, always alternating with other, not related crops. Other Solanum spp., e.g. tomato, which are susceptible to the same pathogens, are not grown to break the development cycle of potato pests (FAO, 2009).

1.3.4.1 DISEASES CAUSED BY ENVIRONMENTAL STRESSES AND NUTRIENT DEFICIENCY

Some diseases occur without the presence of pests or pathogens. Very common causes of disorders are lack of nutrients or disadvantageous environmental conditions (CIP, 1996; Agrios, 2005).

Blackening of the inner core of the tuber flesh (Black heart of potato) is caused by soil oxygen deficiency and too high or too low soil temperatures. Tubers affected by this internal heat necrosis rot quickly. Low temperature injuries show similar symptoms and should not be used as seed (CIP, 1996).

Fast tuber growth in favourable soil and nutrient conditions can cause hollow hearted tubers or cracking of tubers because of internal pressure. Tuber cracking or bruising can also occur, because of a virus infection, mechanical injury or little care during the harvest. However affected tubers do not essentially rot, but their market value is reduced (CIP, 1996).

A quick change from nutrient poor or otherwise disadvantageous growth conditions during tuber initiation to more suitable conditions while tubers are growing can cause knobbiness or irregular shapes of tubers (CIP, 1996).

Tuber development can abruptly stop and stagnate for a while. A result is often the relocation of carbohydrates from the basal parts of tubers to other parts or to other tubers situated on the same stolons. As a consequence, basal tuber parts become soft in a condition known as jelly end rot (CIP, 1996).

Chemicals such as fertilizers, herbicides, insecticides and fungicides can cause injury of the above and below ground parts of potato plants, if they are applied improperly (CIP, 1996).

Air pollution (e.g. sulfur oxides) can cause chlorosis, bleaching, and even burning of leaves, whereas photochemical air pollutants may cause early maturity and dying of potato plants. Both causes can be easily confused with nutrient lack and are difficult to diagnose (CIP, 1996).

Nonvirus leafroll occurs uniformly in a field as a consequence of intense light conditions and long photoperiods. Plants are not affected otherwise and yield well. Other

leaf disorders are caused by nutrient imbalances of nitrogen, phosphorous, and potassium, but also a soil pH below 5.0 and 7.0 can lead to deformation (CIP, 1996).

1.3.4.1 NEMATODE AND INSECT CAUSED POTATO DISEASES

Several nematodes feed on the potato (CIP, 1996; Agrios, 2005), such as the golden cyst nematode (Globodera pallida and G. rostochiensis), and nematodes causing root knot (Meloidogyne spp.), false root-knot (Nacobbus aberrans) or lesions (Pratylenchus spp.).

There is also a large variety of insects and their larvae, which feed and reproduce on the potato. They are most commonly distinguished as pests which feed on foliage or tubers or which suck plant juices. Some of these are not specific to potato or to specific potato organs, but they can also cause severe damage (Berry et al., 2000).

Foliage feeding pests cause symptoms on leaves and stems of plants, such as perforation, feeding traces on leaf edges or cause deformation and decoloration, which can reduce tuber yields because less starch is produced in the leaves and transported to the tubers. The most widespread pests are the Colorado potato beetle (Leptinotarsa decemlineata) and cutworms (larvae of moths like Agrotis spp. and other Noctuidae species; Berry et al., 2000).

Plant juice-sucking pests are mostly found on the undersides of leaves and are often best controlled by their natural enemies. Symptoms are wilting, and silvery, stippled or speckled appearance of leaves, which can lead to yield losses. Examples of such pests are aphids (Myzus persicae and other Aphididae), thrips (Frankliniella spp., Thrips spp.) leafhoppers (Empoasca spp. and other genera), and whiteflies and other Aleyrodidae. Whiteflies sit on the underside of leaves and suck on plant juices. As a consequence a black fungus grows on the honeydew produced by the nymphs and may spread over the affected plants (Berry et al., 2000).

Pests feeding on tubers cause direct damage to the tubers. They can cause surface tracking and small to large irregular tunnels, and color changes of the tuber flesh. Pests causing these symptoms are potato tuber moths (Phthorimaea operculella, Symmetrischema plaesiosema, Tecia solanivora, and Scrobipalpula absoluta), the Andean potato weevil also known as white worm, white grubs (Phyllophaga spp. and other Scarabaeidae), and

wireworms (Agriotes spp. and other Elateridae). Wireworms occur frequently in temperate climates but less often in warm climates (Berry et al., 2000).

1.3.4.2 BACTERIAL POTATO DISEASES

Several bacteria are known to cause potato diseases (CIP, 1996; FAO, 2009). These affect the above ground parts of plants, causing the diseases black leg and soft rot (Erwinia spp.), and bacterial wilt (Pseudomonas solanacearum). Bacteria can also infect the below ground parts of the potato, causing the diseases ring rot (Clavibacter michiganensis ssp. sepedonicum) and common scab (Streptomyces scabies).

1.3.4.3 VIRAL, VIROID AND MYCOPLASM POTATO DISEASES

There are many viruses and viroids causing diseases in potato (CIP, 1996). Most of these cause symptoms on the leaves of the plants. Typical symptoms are deformation of leaves, wilting, bleaching, chlorosis and necrosis. Examples are the Potato leaf roll virus (PLRV), yellow vein, Potato virus Y (PVY) and Potato virus A (PVA), Andean potato mottle virus (APMV) and Andean potato latent virus (APLV), and mosaics (caused by Potato virus X, Potato virus S, Potato virus M, and also PVY and PVA). Symptoms on tubers are caused for example by the spindle tuber disease, mycoplasma pathogens and the Potato mop-top virus (PMTV), which is transmitted by the soilborne parasite Spongospora subterranea f. sp. subterranea (CIP, 1996; see below).

1.3.4.4 FUNGAL POTATO DISEASES

Among the many fungal pathogens attacking potato crops are also several economically important parasites which infect potatoes. The most infamous disease of potato is late blight, caused by Phytophthora infestans. The disease affects above and below ground parts of the potato host. Leaves and stems show water soaked lesions which turn necrotic within a few days. Lesions on stems are brittle and often break. Tubers infected with spores have skin discoloration and tuber flesh becomes brown and necrotic. Although there are some control measures like application of fungicides, late blight remains the most serious fungal disease of potato (CIP, 1996; FAO, 2009).

There are many further important diseases. One of them is early blight, caused by Alternaria solani. Dark brown lesions are formed on leaves, which turn necrotic. On tubers, dry dark brown leathery lesions from on the surface (CIP, 1996; Agrios, 2005). Another widespread disease is powdery mildew (Erysiphe cichoracearum), which has a wide host range and infects many crops. Humidity is a prerequisite for disease development (CIP, 1996; Agrios, 2005). Stems and leaves show lesions after infection, these turn black, and the tissue dies. Stem canker and black scurf are caused by the fungal pathogen Rhizoctonia solani, which is present in almost all soils, has a wide host range, survives on host debris and disseminates by sclerotia on tubers. Lesions on the sprout tip can cause late or failing emergence. Sunken, brown cankers are formed on stolons and stems below the soil line. Stems are often deformed which might lead to aerial tuber development, plant wilt and death, and inhibition of tuber development. Hard, dark brown to black sclerotia are the resting bodies of the fungus on the tuber surface. These are able to survive long periods in the soil (CIP, 1996; Agrios, 2005). Powdery scab is a potato tuber and root disease caused by Spongospora subterranea f. sp. subterranea (Sss). Open lesions on the tuber surface are formed, which each contain a powdery mass consisting of the resting spores of the pathogen are the typical symptom of this blemish disease. Root galls, another resting spore structure of the pathogen, occur on roots and stolons of the host plant. As a vector for the Potato mop-top virus (PMTV), Sss infections may also lead to serious double infections. The disease is widespread and occurs in most of the temperate potato producing and processing regions, but also in hot and dry climates where irrigation is deployed (Merz & Falloon, 2009; Merz, 2008).

1.4 THE PATHOGEN: SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA

Spongospora subterranea f. sp. subterranea (Sss) is a soilborne pathogen causing powdery scab of potato. Taxonomically, it belongs to the plasmodiophorids and is an obligate biotrophic parasite, which needs a living host to reproduce and complete its life cycle. The pathogen cannot be cultivated on artificial media, so Koch’s postulates cannot be completed for the organism (Merz & Falloon, 2009).

1.4.1 LIFE CYCLE

The formation of lesions on tuber surfaces and root galls are characteristic symptoms on host plants after Sss infection. These galls and lesions contain sporosori (also referred to as “spore balls” or “cystosori” (Harrison et al., (1997)) which facilitate long-term survival in the soil and are highly resistant to environmental stresses. Each sporosorus contains about 700 thick-walled resting spores, but numbers can vary between 160 and 1530. The number of resting spores in one sporosorus can be calculated accurately by measuring its dimensions (Falloon, 2006; Falloon, 2007).

Figure 7 Life cycle of Spongospora subterranea f. sp. subterranea as shown in Harrison et al. (1997).

Each resting spore can release a biflagellate primary zoospore, which is able to swim short distances in moist soil to reach the host (Fig. 7; Merz & Falloon, 2009). Merz (1989) used a pulse inoculation technique and reported zoospore release after 15 min. However, in nature the release proceeds arbitrarily (Fornier, 1997, Harrison et al., 1997). Some spores are released immediately from sporosori, while others may need external stimuli to mediate germination. Additionally, it is believed that resting spores have a dormant state (Merz, 1989), but there has as yet been no evidence for this hypothesis. Subsequent to the hypothetical dormancy, a short period of maturation takes place before the zoospores are ready to emerge (Merz, 1997). When the primary zoospores reach the host, they encyst on the surface of root or stolon epidermis cells (Fig. 8 A), form the “Rohr” which acts as a microscopic tube (Fig. 8 B), and build up vacuole pressure to press the “Stachel”, a projectile for mechanical penetration of the host cell wall, through the Rohr (Fig. 8 B and C). The contents of a zoospore then enters the host cell and forms a plasmodium (Fig. 8 D).

Figure 8: The mechanical host cell penetration system Sss zoospores use to infect the host is a common and unique feature of all plasmodiophorids (Aist & Williams, 1971).

This mechanism for mechanical host cell wall penetration is common in all plasmodiophorids.

The plasmodium in the host cell can develop into a zoosporangium and release secondary zoospores, which can cause further infections. However, the plasmodia can also proliferate to resting spore structures, such as galls on roots or stolons and pustules on tubers (Fig. 7), which again contain sporosori.

Primary and secondary zoospores are formed in very different ways. Primary zoospores develop from resting spores (in sporosori) while secondary zoospores develop in zoosporangia in root cells. Merz (1997) found no differences in morphology or swimming pattern between the two types of Sss zoospores.

1.4.2 PHYLOGENY OF SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA

Many phylogenies have been used by different authors since Sss was first described and named. First considered to belong to the true fungi, Sss is now classified in the Plasmodiophoridae (Plasmodiophorida, Phytomyxea, Opalozoa, Dictyozoa (subkingdom), Protozoa (Braselton, 2001), hereafter called plasmodiophorids (Down et al., 2002). This family includes ten genera and 36 species. The genus Spongospora consists of four described members, two of which are defined as formae speciales: Spongospora subterranea f. sp. nasturtii (Ssn), the cause of crook root of water cress, and Spongospora subterranea f. sp. subterranea (Karling, 1968). This categorization is due to the morphological similarity of the sporosori of both organisms, and these large aggregations of resting spores are characteristic for Spongospora. Merz et al. (2005) confirmed the close relationship of these organisms when they found that monoclonal antibodies produced against Sss sporosori cause the same reaction with sporosori of Ssn. However, the phylogeny of the plasmodiophorids is still developing. Dick (2001) believed taht Ssn was a separate species and not a formae specialis, namely Spongospora nasturtii. This species name was also used by Down et al. (2002), but the species status of Ssn has not been officially confirmed.

1.4.3 SPONGOSPORA GENETICS

Genetic knowledge of Sss is scarce, as is knowledge about the genetic variation of the organism or its ability for sexual reproduction. However, plant breeders need this information to screen new potato lines for cultivars resistant to powdery scab.

First characterizations of Sss have considered size of sporosori as an indicator for variation in the pathogen. This way, Wuerzer (1965) distinguished two groups. Two groups were also distinguished by Bulman & Marshall (1998) and Qu & Christ (2004) using PCR primers for the amplification of the internal transcribed spacer (ITS) of Sss. From both studies, the same two sequences were published, named by Qu & Christ (2004) ITS Type I and ITS Type II.

Later, also randomly amplified polymorphic DNA (RAPD) analysis and restriction fragment length polymorphism (RFLP) analysis were applied to 24 Sss samples from North America. Genetic variation was recorded only among but not within geographic locations represented by potato fields, which lead to the assumption that field populations of Sss are clonal and not the results of sexual reproduction (Qu & Christ, 2006).

There is an alternative theory, that sporosori consisting of the resting spores are the result of sexual reproduction. Thus, each resting spore would have its own genotype, meaning that one sporosorus, which consists of many resting spores, may contain a mix of genotypes instead of clones (Braselton, 1992).

Braselton (1992) also observed the formation of senaptonemal complexes in transitional plasmodia of Sss, which are protein structures required for the pairing of chromosomes. He concluded that these transitional plasmodia containing multiple nuclei undergo meiosis prior to cleavage. Furthermore, individual resting spores are believed to be haploid, but this is still unclear, as is the occurrence and process of meiosis in this organism.

So far, only the actin, polyubiquitin, and the two ITS sequences of Sss have been published. Despite of the existence of an expressed sequence tag (EST) library established by Dr Simon Bulman (personal communication), in New Zealand, a completely sequenced genome of Sss has not been published or catalogued. Recently Bulman et al. (2011) published a set of sequences of Sss.

There is need for (population) genetic knowledge of Sss for plant breeders, because the information on strains representing the known genetic diversity is required for screening of new resistant cultivars, which play a key role in powdery scab management.

1.4.4 CULTIVATION

As an obligate biotrophic parasite, Sss cannot be cultivated in axenic culture on artificial media. Thus, DNA extraction is always performed on non-sterile potato tuber lesion scrapings or root galls, resulting in DNA mixtures of at least Sss and potato, but potentially also numerous other soil organisms.

Some approaches have been made in the past decades to cultivate Sss on media. A possible breakthrough was reported Diriwächter and Gindrat (1982), who described growing amoeboid Sss spores on media. After many years of trials and testing, however, this proved to be the result of contamination and not Sss growing on the plates (Qu et al., 2001).

It is possible to establish potato hairy root cultures of Sss (Qu & Christ, 2007) as a way of maintaining permanent cultures of Sss. Recently, Bulman et al. (2011, also mentioned in Burki et al. 2010) were able to establish callus cultures of Sss from infected potato root galls.

1.4.5 HOST RANGE

The main host of Sss is the potato, but many members of many plant families can be hosts of the zoosporangial stage of this pathogen (Harrison, 1997). The hosts most commonly recorded are members of the Solanceae and Chenopodiaceae, but the complete life cycle, with the formation of sporosori, occurs exclusively in Solanaceae.

1.4.6 ENVIRONMENTAL FACTORS

A wide temperature range, from 5 °C to 25 °C, was determined for zoospore release. However, the activity of zoospores differentiates with temperatures. For tuber infection, 12 – 13 °C are favoured, but root galling occurs at warmer temperatures around 17 °C (van de Graaf et al., 2007).

Soil moisture is required by zoospores to swim and reach host tissue, and it has been shown that constant dampness increases tuber disease (Kole, 1954). Thus, it has been suggested by Wale (2000) that overwatering should be avoided during tuber initiation, de Boer (2000a,b) suggested that withholding irrigation during this period could have similar effects.

High incidence and heavy severity of powdery scab is usually found in heavy soils with high water retention capacities, but also sandy soils and loams may increase disease, while less disease has been recorded in clay soils (van de Graaf et al., 2005).

It is widely accepted and agreed, that the most conductive conditions for powdery scab development occur in cool, moist and heavy soils.

1.4.7 PROPAGATION AND ECONOMIC IMPACT OF THE DISEASE

Powdery scab is a blemish disease of potato tuber skins with the potential to cause severe economic losses. During the past few years, “new reports” of powdery scab have been increasingly published. This suggests that the problems with this disease are increasing; alternatively this could indicate that the disease has been underestimated or did not receive wide attention (Merz, 2008).

Due to the lack of airborne spores of the pathogen and the limited swimming distance of the zoospores of only 2 m at most (with additional soil water flow), the pathogen is unlikely to spread across great geographic distances without human assistance. The exact center of origin of Sss is unkown, but it is assumed that the pathogen is native to South America (Abbott, 1931). From there the pathogen was probably introduced to the rest of the world by humans through infected seed potato tubers or contaminated soil, as it is still spread this way, today.

Other factors likely to contribute to greater disease incidence are the intensification of potato production and the more frequent use of susceptible potato cultivars. On the one hand, these often have increased value for fresh market sales or potato production and processing, but they carry the potential to accumulate inoculum or lead to crop failures. Other factors positively influencing the disease incidence are more frequent use of irrigation and the removal of mercury seed tuber treatments (Merz et al., 2004).

Infected seed potato lines may require extra grading to select for healthy tubers and discard infected ones. Depending on the severity of infection of the crops, whole lots of seed potatoes may be rejected. In Switzerland, it was estimated that the losses of seed potato tuber production in wet years rise to several hundred thousand euros (SEMAG, 2005).

High quality is expected in fresh market sale of potato, especially when tubers are displayed as washed product. In case of infection, extra grading is needed to remove infected tubers to meet market standards. Powdery scab poses additional problems in potato processing. Infected skin is more difficult to remove, requiring extra processing and producing increased waste (Merz & Falloon, 2009; Falloon, 2008).

Infection of plant roots by Sss may also have harmful effects on productivity. Falloon et al. (2005) showed that plants with infected roots take up less water and nutrients, have smaller shoots, fewer leaves and produce less dry matter.

Manuring might also have an impact on disease development. It was reported by Pethybridge (1911), that a crop was badly affected by powdery scab after applying manure of pigs, which had been fed with unboiled diseased tubers. Recently, experiments were performed with goat and cow manure (U. Merz, A. Kaiser, P.-Y.- Jaquiéry and T. Oberhänsli, unpublished data). These indicated that resting spores of Sss can survive the passage through the digestive tract of livestock and are still infectious.

Serious double infections of the potato host are possible if Sss carries PMTV, which causes further superficial and internal symptoms on tubers, making them unmarketable (Jones & Harrison, 1969).

1.4.8 DIAGNOSTICS AND CONTROL METHODS

Currently, short term control of the potato diseases caused by Sss can be achieved by avoiding soil contamination with infected seed potatoes, by planting clean seed tubers into soils without former powdery scab history. The direct control of the pathogen is difficult and not completely effective.

The tolerance limit in Switzerland for infected seed potatoes is at most 1 % of the tubers with not more than five lesions (Anonymous, 1989). This is a level of severity which corresponds to an average tuber surface infection of about 1 %. If this limit was chosen accurately is difficult to determine. Nothing is known about the level of seed tuber infection necessary to induce powdery scab, so the veracity of the infection limit has not been accurately assessed. Even symptomless tubers may cause disease of progeny tubers (Theron, 1999), probably from sporosorus contamination of symptomless seed tubers.

Nevertheless, seed certifying authorities need practical tolerance limits to prevent high levels of infection, but these vary between countries (Wale, 2000) and international coordination and standards in diagnostics is necessary. For example, visual inspection of unwashed potatoes is still a widely used practice, but then common scab can be confounded with powdery scab (Merz, 2000). Well-trained inspectors can distinguish between the two scab diseases, caused by different pathogens if they have good light conditions and tubers are washed for inspection.

Using a microscope, the sporosori of Sss are easily recognized. Further methods for detection are the Agristrip (Bioreba, CH) a rapid on site test for laymen without special diagnostics training or the immune based enzyme-linked immunosorbent assay (ELISA). The detection limit was determined by Merz (decribed in Merz & Lees, 2001) using a peeling machine on 20 tubers, 18 healthy and two tubers each with one lesion. Usually, the more sensitive detection techniques are based on (real-time (rt)) PCR (Bulman & Marshall, 1998; van de Graaf, 2003), but Boucheket et al. (2011) showed that PCR and Agristrip detection limits are about equal for identification of tuber symptoms.

If the presence of Sss has been verified, different forms of treatments can be considered. None of the currently available treatments are completely effective and offer total protection, but they can assist to reduce disease.

For seed tuber treatment, the best chemicals were mercury containing compounds, but they were banned in the late 1980s, because they are toxic to humans and animals and are harmful to the environment (Merz & Falloon, 2009). Several other treatments have been applied since then.

Promising is fluazinam (3-chloro-N-(3-chloro-5-trifluoromethyl-2-pyridyl)-alpha, alpha, alpha-trifluoro-2,6-dinitro-p-toluidine), an acaricide and fungicide. Applied on infected seed tubers one day before planting into uncontaminated soil, it was reported to control the disease in New Zealand (Falloon et al., 1996) and registered. Later, the license was withdrawn because of phytotoxicity problems. These beneficial control effects have not been confirmed in Switzerland. Thus, the control of the pathogen is not completely effective, but the disease might be reduced.

Flusulfamide is a pesticide originally developed for club root (Plasmodiophora brassicae) control in Japan, but is now registered in New Zealand against powdery scab. It has a similar control spectrum as fluazinam (Falloon, 2008).

Zinc oxide application to seed tubers is performed in Scotland and effective when disease pressure is low (Wale, 2000). The same effect was not achieved in New Zealand. Some other pesticide chemicals have (limited) effects for powdery scab control (Falloon, 2008).

An alternative control method without chemical usage is the steam treatment of seed tubers for 10 s, which was reported to have similar effects as the successful treatment with mercury compounds but without the disadvantages for environmental, animal, and human health (Afek & Orenstein, 2002).

Although some control methods exist, they only show effects when the seed tuber infection levels are low to moderate and where soil is not infested with Sss. The soil inoculum level plays a role at least as important as infected seed in disease development. Determining the presence and quantity of soil inoculum of Sss is therefore essential for studies of epidemics and powdery scab risk assessment (Falloon, 2008).

Merz (1989) developed a bioassay using tomato (Solanum lycopersicum) bait plants and inoculated them with soil samples. Zoosporangia of Sss formed in root epidermis cells (including root hairs) of the plant can be detected using a microscope and indicate a contaminated soil sample. With the modification described by Walsh et al. (1996), the detection limit as low as 100 sporosori/g soil was achieved.

Using ELISA a similar level of sporosori in soil can be detected (100 sporosori/g; (Walsh, 1996)), but the test works better starting from 2,000 sporosori/g soil. For lower levels of inoculum, Merz’s (1989) bioassay showed better discrimination, because the pathogen can multiply in plant roots. An improvement of the ELISA technique was achieved by Merz (2005) with monoclonal antiserum against sporosori, such that the test can be used as a semi-quantitative method to assess inoculum levels. In Switzerland this test has already become routine and is well established.

Bell et al. (1999) were able to detect one to ten sporosori/g soil with an adjusted deoxyribonucleic acid (DNA) extraction method. Using quantitative PCR, van de Graaf (2003) used the DNA extraction method of Bell et al. (1999) but a different set of primers for detection. Qu et al. (2006) developed new primers, purified the DNA prior to quantitative PCR, and were able to detect one sporosorus/g soil. Furthermore, they could determine sporosorus concentrations from naturally infested fields in a competitive PCR assay. Developing a standard curve from soil samples inoculated with known amounts of sporosori, they were later able to compare soil samples from infected fields and determine the degree of contamination.

The advantages of immunoassays and PCR in Sss detection are their sensitivity, the saving of time and workspace, quantification, and the possibility of achieving high throughput. There is no need for bait plants, which have to be cultivated and maintained. However, the advantages of the bioassay are that only viable resting spores are detected and false positives are reduced. To determine the disease risk, a combination of the bioassay and a PCR technique, preferably quantitative, could be used to make realistic prognoses, as suggested by Nakayama et al. (2007). All of these methods need expertise and lab experience, thus better on-site tests could be helpful for laymen.

Furthermore, the relationship between Sss soil inoculum levels and powdery scab outbreaks is still unclear. There are difficulties in establishing reliable thresholds for soils with high risk. Many factors might influence the risk of disease.

Not only the amount of sporosori but also their distribution is important. Tests have demonstrated the amounts of sporosori/g soil vary within a field (Merz et al., 2005). In Switzerland it is therefore standardized procedure to take 20 soil samples of 100 g each in a diagonal manner from one planting site and then testing them separately for inoculum.

In cases where potato must be grown in infested soil, e.g. in countries with little clean soil and long histories of powdery scab, it might be environmentally justifiable to treat soils with chemicals. Some treatments have shown efficacy. Fluazinam and flusulfamide, which are also used for seed tuber treatments, could be applied. Also the fungicide cyazofamid, also used for late blight (Phytophthora infestans) control, seems to be effective

to control Sss. In Pakistan, stable bleaching powder in combination with boric acid solution tuber treatment showed control effects. Finally, boron used as sodium tetraborate might also be used and effective (Falloon et al., 2001).

In any case, the treatment of soils against powdery scab is unlikely to be completely effective. It remains the question, if it is environmentally acceptable to treat soils using chemicals without knowing if it will be successful in controlling the disease.

Brassica crops can reduce soilborne pathogens. They produce volatile sulfur compounds toxic to soil microbes, and “biofumigation” with Indian mustard can be performed. A further biological control may be the use of Trichoderma harzinum, which has shown to reduce root infections by Sss on tomato when products were mixed with sporosori in soil (Nielsen & Larsen, 2004).

1.4.9 RESISTANCE AGAINST POWDERY SCAB IN POTATO

More efficient control than application of chemicals to seed and fields could be achieved with potato cultivars resistant to diseases caused by Sss. Additionally, the cultivars must be of high commercial value. Currently, no such cultivars exist. However, there are considerable differences in susceptibility to tuber and root infection between host subspecies and cultivars (Schwaerzel, 2002).

Screening for resistance is performed on naturally infested soils in field trials. Roots, stolons, and tubers can be infected by the pathogen. Only tubers possess a commercial value and root infection has been unattended for a long time. Nevertheless, root gall formation in cultivars resistant to tuber infection but susceptible to root infection increases inoculum of Sss in fields (Schwaerzel, 2002) and must be considered in resistance screening.

Recently, standards for scoring of tuber and root infection were established at the First Powdery Scab Workshop in La Fretaz, Switzerland (http://www.spongospora.ethz.ch/ LaFretaz/index.htm, 11.07.2012).

During the past years, the cultivar ‘Gladiator’ from New Zealand showed very good resistance to powdery scab in international trials (Merz et al., 2012). In six locations in five countries (Switzerland, United Kingdom, France, Netherlands, Germany), 10 international

varieties were screened, eight of these performed always as expected from previous tests. In all trials, tuber infection was assessed 2 months prior to harvesting and root infection was assessed 1 month before harvesting. The cultivar ‘Gladiator’ always was the most resistant, showing the least tuber and root infection.

Falloon et al. (2003) suggested that the resistance to tuber infection or root infection, respectively, is likely to be quantitative, because the spectrum of resistance across a large number of cultivars is a continuum from highly resistant to highly susceptible.

In Peru, Torres et al. (1995) screened a multitude of potato accessions in different locations over 6 years. Some of these were resistant over all years in all locations, while for others both resistance and susceptibility was found. This led to the assumption that Sss may have different pathotypes.

To reduce costly and time-consuming field trials, Merz (2004) proposed a laboratory- based micro-bioassay for screening different potato lines for susceptibility to Sss. Additionally, genetic studies on Sss can assist to clarify open questions.

1.5 AIM OF THE PRESENTED STUDY

Powdery scab of potato is difficult to manage. Control agents for seed potatoes and soil do not provide complete protection against the disease and resistant cultivars with high market value are not available. The most promising approach lies in breeding of new resistant varieties, but the genetic data of the pathogen, which is crucial for screening processes during breeding is lacking. The aim of the presented study was therefore to develop and optimize genetic markers for Sss, which causes the disease.

Findings can expand the population genetic knowledge of the pathogen, and these are likely to lead to improved powdery scab management. On the one hand, established quarantine measures can be evaluated, but also new and optimized quarantine measures could be designed and implemented. On the other hand, breeding of resistant plants might be simplified and practicable.

The second chapter of this study describes marker development and application to Sss field populations from all continents where crops are grown, and different sources of sporosori, i.e. root galls or tuber lesions. The aims of this research were to determine the population structures and characterize evolutionary units, such as haplotypes, clusters, and genetic relationships in a global phylogenetic network.

In the third chapter of the study, a cross inoculation experiment is described. Inoculum of Sss was obtained from native and invaded regions, from tuber lesions and root galls, and from different hosts. All major potato host (sub-) species, namely Solanum phureja, Solanum tuberosum ssp. tuberosum and Solanum tuberosum ssp. andigena, were inoculated with the different inocula, and subsequently root infection was scored. Inocula, harvested root galls and mini-tuber lesions were genotyped with the previously developed markers. This experiment was carried out to determine the ability of Sss sporosori to cause root galling on potato hosts using a micro-bioassay (Merz et al., 2004). Tested were sporosori collected from lesions and galls, different potato hosts, and from both the centre of origin and elsewhere in the world.

This work was supported by funding from Horticulture Australia Ltd (project number P08032 (APRP1, University of Tasmania)), the ETH Zürich, the United Kingdom Potato

Council, the New Zealand Ministry of Science and Innovation, Horticulture New Zealand, and the Swiss Federal Office of Agriculture.

Data analyzed in this work were generated in the Genetic Diversity Centre of ETH Zurich, Switzerland, and at the New Zealand Institute for Plant and Food Research Limited, Lincoln, Canterbury, New Zealand.

1.6 REFERENCES

Abbott EV, 1931. Further notes on plant diseases in Peru. Phytopathology 21, 1061-1071.

Afek U, and Orenstein J, 2002. Disinfecting potato tubers using steam treatments. Canadian Journal of Plant Pathology 24, 36-39.

Agrios GN, 2005. Plant Pathology. Department of Plant Pathology University of Florida: Elsevier Academic Press.

Aist JR, and Williams PH, 1971. Cytology and kinetics of cabbage root hair penetration by Plasmodiophora brassicae. Canadian Journal of Botany 49 (11), 2023-2034.

Anderson PK, Cunningham AA, Patel NG, Morales FJ, Epstein PR, and Daszak P, 2004. Emerging infectious diseases of plants: Pathogen pollution, climate change, and agrotechnology drivers. Trends in Ecology & Evolution 19 (10), 535-544.

Anonymous, 1989. Schweizerische Handelssubstanzen für Kartoffeln. Schweizerische Kartoffelkommission, Düdingen.

Bell KS, Roberts J, Verrall S, Cullen DW, Williams NA, Harrison JG, Toth IK, Cooke DEL, Duncan JM, and Claxton JR, 1999. Detection and quantification of Spongospora subterranea f. sp. subterranea in soils and on tubers using specifique PCR primers. European Journal of Plant Pathology 105, 905-915.

Berry RE, Reed GL, and Coop LB, 2000. Identification & management of major pest & beneficial insects in potato [online]. Publication No. IPPC E.04-00-1. Oregon State University, Department of Entomology and Integrated Plant Protection Center, Corvallis, OR. Available from Internet: http://ippc2.orst.edu/potato.

Bhagwat SA, Willis KJ, Birks HJB, and Whittaker RJ, 2008. Agroforestry: A refuge for tropical biodiversity? Trends in Ecology & Evolution 23 (5), 261-267.

Boote KJ, Jones JW, Mishoe JW, and Berger RD, 1983. Coupling pests to crop growth simulators to predict yield reductions. Phytopathology 73, 1581–1587.

Bouchek-Mechiche K, Montfort F, and Merz U, 2011. Evaluation of the Sss AgriStrip rapid diagnostic test for the detection of Spongospora subterranea on potato tubers. European Journal of Plant Pathology 131 (2), 277-287. DOI: 10.1007/s10658-011- 9807-1.

Bourke A, 1993. The visitation of God? The potato and the great Irish famine. Dublin.

Braselton JP, 1992. Ultrastructural karyology of Spongospora subterranea. Canadian Journal of Botany 70 (6), 1228-1233.

Braselton JP, 2001. Plasmodiophoromycota. In The Mycota, VII Part A. Systematics and evolution. DJ McLaughlin, EG McLaughlin & PA Lemke, eds. Springer-Verlag, Berlin and Heidelberg: 81-91.

Brush S, 1992. Reconsidering the Green Revolution: Diversity and stability in cradle areas of crop domestication. Human Ecology 20, 145—9867.

Brush S, Taylor JE, and Bellon M, 1992. Technology adoption and biological diversity in Andean potato agriculture. Journal of Development Economics 39, 365—387.

Burki F, Kudryavtsev A, Matz MV, Aglyamova GV, Bulman S, Fiers M, Keeling PJ, Pawlowski J, 2010. Evolution of Rhizaria: new insights from phylogenomic analysis of uncultivated protists. BMC Evolutionary Biology 10, 377.

Burlingame B, Mouillé B, and Charrondiére R, 2009. Nutrients, bioactive non-nutrients and anti-nutrients in potatoes. Journal of Food Composition Analysis 22, 494–502.

Bulman SR, and Marshall JW, 1998. Detection of Spongospora subterranea in potato tuber lesions using the polymerase chain reaction (PCR). Plant Pathology 47 (6), 759-766.

Bulman S, Candy JM, Fiers M, Lister R, Connera AJ, Eady CC, 2011. Genomics of Biotrophic, Plant-infecting Plasmodiophorids Using In Vitro Dual Cultures. Protist 162 (3), 449– 461.

Chen XM, 2005. Epidemiology and control of stripe rust Puccinia striiformis f.sp. tritici on wheat. Canadian Journal of Plant Pathology – Revue Canadienne de Phytopathologie 27 (3), 314-337.

Centro Internacional de la Papa (CIP, International Potato Center), 1996. Major Potato Diseases, Insects, and Nematodes — Lima, Peru.

Centro Internacional de la Papa (CIP, International Potato Center), 2008. The odyssey of the potato. Der lange Weg der Kartoffel zur weltweiten Kulturpflanze. Lima, Peru.

Cobo B, [1653] 1890—1893. Historia del nuevo mundo, ed. M. Jiminez de la Espada. 4 vols. Seville.

Coe S, 1994. America’s first cuisines. Austin, Texas. de Acosta J, [1590] 1880. Historia natural y moral de las Indias, trans. E. Grimston, London, 1604. Haklyuyt Society reprint, 1880. Seville. de Boer RF, 2000a. Research into the biology and control of powdery scab of potatoes in Australia. In: Merz U, Lees AK (eds) Proceedings of the first European powdery scab workshop, Aberdeen, Scotland, July 20–22, pp 79–83. http://www.spongospora. ethz.ch/EUworkshop/proceedings.html. de Boer RF, 2000b. Summary of the session on recognising the components of an integrated control approach to powdery scab and the potato mop top virus. In: Merz U, Lees AK (eds) Proceedings of the first European powdery scab workshop, Aberdeen, Scotland, July 20–22, pp 101–104. http://www.spongospora.ethz.ch/EUworkshop/proceedings. html.

Desprez-Loustau ML, Robin C, Buee M, Courtecuisse R, Garbaye J, Suffert F, Sache I, and Rizz DM, 2007. The fungal dimension of biological invasions. Trends in Ecology & Evolution 22 (9), 472-480.

Dick MW, 2001. Straminipilous fungi: systematics of the Peronosporomycetes including accounts of the marine straminipilous protists, the plasmodiophorids and similar organisms. Kluwer, Dordrecht.

Diriwächter G, and Gindrat D, 1982. Contribution a l'etude de la gale poudreuse de la pomme de terre. Revue Suisse Agricole 14, 265-269.

Down GJ, Grenville LJ, and Clarkson JM, 2002. Phylogenetic analysis of Spongospora and implications for the taxonomic status of the plasmodiophorids. Mycological Research 106, 1060-1065.

Dybdahl MF, and Storfer A, 2003. Parasite local adaption: Red queen versus suicide king. Trends in Ecology & Evolution 18 (10), 523-530.

Falloon RE, Wallace AR, Braithwaite M, Genet RA, Nott HM, Fletcher JD, and Braam WF, 1996. Assessment of seed tuber, in-furrow, and foliar chemical treatments for control of powdery scab (Spongospora subterranea f. sp. subterranea) of potato. New Zealand Journal of Crop and Horticultural Science 24 (4), 341-353.

Falloon RE, Merz U, Curtin D, and Butler RC 2001. Boron affects Spongospora subterranea infection of host roots; laboratory and glasshouse results. I.J. Porter et al. (eds.); Proceedings of the 2nd Australasian Soilborne Diseases Symposium, Lorne, Victoria, Australia, 5-8 March, 101-102.

Falloon RE, Genet RA, Wallace AR, and Butler RC, 2003. Susceptibility of potato (Solanum tuberosum) cultivars to powdery scab (caused by Spongospora subterranea f. sp. subterranea), and relationships between tuber and root infection. Australasian Plant Pathology 32, 377-385.

Falloon RE, Genet, RA, Lister RA, and Butler RC, 2005. Root function and growth of potato plants reduced by Spongospora subterranea infection. American Journal of Potato Research 82, 68.

Falloon RE, Merz U, Lister RA, Wallace AR Lamberts R, Hayes S, 2006. Morphology of Spongospora subterrnaea sporosori assists enumeration of resting spore inoculums. In: Falloon RE, Cromey MG, Stewart A, Jones EE (eds) Proceedings of the 4th Australasian soilborne diseases symposium, Queenstown, New Zealand, September 3–6, 2006, pp 70–71

Falloon RE, Merz U, Lister RA, Wallace AR, 2007. Numbers of resting spores determined in sporosori of Spongospora subterranea. Proceedings of the 16th biennial Australasian

Plant Pathology Society conference, Adelaide, Australia, September 24–27, 2007, p 141

Falloon RE, 2008. Control of powdery scab of potato: Towards integrated disease management. American Journal of Potato Research 85 (4), 253-260.

Food and Agriculture Organization of the United Nations, Statistics Devision (FAOSTAT) information available online at: http://faostat.fao.org/.

Food and Agriculture Organization of the United Nations (FAO), 2009. International year of the potato 2008 – New light on a hidden treasure – An end of the year review. ISBN 978-92-5-306142-8, Rome, Italy.

Fornier N, 1997. Epidemiology of Spongospora subterranea, the cause of powdery scab of potatoes. PhD thesis, Department of Agriculture, University of Aberdeen.

Fry WE, Goodwin SB, Dyer AT, Matuszak JM, Drenth A, and Tooley PW, 1993. Historical and recent migrations of Phytophthora infestans: chronology, pathways, and implications. Plant Disease 77, 653-661.

Fry WE, and Goodwin SB, 1997. Re-emergence of potato and tomato late blight in the United States. Plant Disease 81, 1349 -1357.

Gavrilets S, Giraud T, and Gladieux P, 2010. Linking the emergence of fungal plant diseases with ecological speciation. Trends in Ecology & Evolution 25, 387-395.

González-Varela G, Gonzalez AJ, and Milgroom MG, 2011. Clonal population structure and introductions of the chestnut blight fungus, Cryphonectria parasitica, in Asturias, northern Spain. European Journal of Plant Pathology 131, 67-79.

Grun P, 1990. The evolution of cultivated potatoes. Economical Botany 44 [Suppl 3], 39–55.

Guaman Poma de Ayala, and Felipe Guaman, 1936: Nueva cronica y buen gobierno. In Traveaux et Mémoires de l’Institut d’Ethnologie 23, ed. P. Rivet. Paris.

Harrison JG, Searle RJ, and Wiliams NA, 1997. Powdery scab disease of potato - a review. Plant Pathology 46, 1-25.

Hawkes JG, 1990. The potato: Evolution, biodiversity, and genetic resources. Belhaven Press, London.

Hawkes JG, and Francisco-Ortega J, 1992. The potato during the late 16th century. Economic Botany 46, 86—97.

Hijmans RJ, 2001. Global distribution of the potato crop. American Journal of Potato Research 78, 403–412.

Hosaka K, Hanneman Jr RE, 1988. The origin of the cultivated tetraploid potato based on chloroplast DNA. Theoretical and Applied Genetics 76, 172–176

Johns T, 1990. With bitter herbs they shall eat it: Chemical ecology and the origins of human diet and medicine. Tucson, Arizona.

Jones RAC and Harrison BD, 1969. The behavior of potato mop-top virus in soil and evidence for its transmission by Spongospora subterranea (Wallr.) Lagerh. Annals of Applied Biology 63, 1-17.

Juzepczuk SW, Bukasov SM, 1929. A contribution to the question of the origin of the potato (in Russian). Proceedings of the USSR Congress of Genetics of Plant and Animal Breeding 3, 592–611.

Karling JS, 1968. The Plasmodiophorales (2nd ed.). Hafner Publ. Co, N.Y. and London.

Kinealy C, 1995. The great calamity: The Irish famine 1845—52. Boulder, Colombia.

Kiple KF, and Ornelas KC, 2000. The Cambridge World History of Food. ISBN:9780521402163. Cambridge University Press. England.

Kole AP, 1954. A contribution to the knowledge of Spongospora subterranea (Wallr.) Lagerh., the cause of Powdery scab of potatoes. Tijdschrift over plantenziekten 60, 1-65.

Lashermes P, Combes MC, Trouslot P, and Charrier A, 1997. Phylogenetic relationships of coffee-tree species (Coffea L) as inferred from ITS sequences of nuclear ribosomal DNA. Theoretical and Applied Genetics 94, 947-955.

Lawson Handley L-J, Estoup A, Evans DM, Thomas CE, Lombaert E, Facon B, Aebi A, and Roy HE, 2011. Ecological genetics of invasive alien species. BioControl 56, 409–428. DOI 10.1007/s10526-011-9386-2.

Lombaert E, Guillemaud T, Cornuet J-M, Malausa T, Facon B, and Estoup A, 2010. Bridgehead Effect in the Worldwide Invasion of the Biocontrol Harlequin Ladybird. Plos One 5.

McDonald BA, 1997. The population genetics of fungi: Tools and techniques. Phytopathology 87, 448-53.

McDonald BA and Linde C, 2002. Pathogen population genetics, evolutionary potential, and durable resistance. Annual Review of Phytopathology 40, 349-379.

Merz U, 1989. Infectivity, inoculum density and germination of Spongospora subterranea resting spores: a solution-culture test system. Bulletin OEPP 19, 585-592.

Merz U, 1997. Microscopical observations of the primary zoospores of Spongospora subterranea f. sp. subterranea. Plant Pathology 46 (5), 670-674.

Merz U and Lees AK, eds., 2001. Proceedings of the First European Powdery Scab Workshop, 2000, Aberdeen, Scotland, July 20-22, 104 p./www.pa.ipw.agrl.ethz.ch/EUworkshop /proceedings.html

Merz U, Martinez V, and Schwaerzel R, 2004. The potential for the Rapid Screening of Potato Cultivars (Solanum tuberosum) for Resistance to Powdery Scab (Spongospora subterranea) using a Laboratory Bioassay. European Journal of Plant Pathology 110, 71-77.

Merz U, Walsh JA, Bouchek-Mechiche K, Oberhänsli T, and Bitterlin W, 2005. Improved immunological detection of Spongospora subterranea. European Journal of Plant Pathology 111, 371-379.

Merz U, 2008. Powdery scab of potato - Occurrence, life cycle and epidemiology. American Journal of Potato Research 85, 241-246.

Merz U and Falloon RE, 2009. Review: Powdery Scab of Potato—Increased Knowledge of Pathogen Biology and Disease Epidemiology for Effective Disease Management. Potato Research 52 (1), 17–37.

Merz U, Lees AK, Sullivan L, Schwaerzel R, Hebeisen T, Kirk HG, Bouchek-Mechiche K, and Hofferbert H-R, 2012. Powdery scab resistance in Solanum tuberosum: an assessment of cultivar — environment effect. Plant Pathology 61(1), 29-36.

Moore B, 2005. Alien invasive species: Impacts on Forests and Forestry. – A Review. Forest Resources Development Service Working Paper FBS/8E, Forest Resources Division FAO, Forestry Department, Rome, Italy. Available online: http://www.fao.org// docrep/008/j6854e/j6854e00.htm

Montarry J, Andrivon D, Glais I, Corbiere R, Mialdea G, Delmotte F, 2010. Microsatellite markers reveal two admixed genetic groups and an ongoing displacement within the French population of the invasive plant pathogen Phytophthora infestans. Molecular Ecology 19, 1965-1977.

Morris C, 1981. Tecnología y organizacion Inca del almacenamiento de víveres en la sierra. In La Tecnología en el Mundo Andino, ed. H. Lechtman and A. M. Soldi, 327—375. Mexico.

Nakayama T, Horita M, and Shimanuki T, 2007. Spongospora subterranea soil contamination and its relationship to serverity of powdery scab on potatoes. Journal of General Plant Pathology 73, 229–234. doi:10.1007/ s10327–007–0008-x.

Nef JU, 1950. War and human progress. Cambridge.

Nielsen SL and Larsen J, 2004. Two Trichoderma harzinum-based bio-control agents reduce tomato root infection with Spongospora subterranea (Wallr.) Lagerh., f.sp. subterranea, the vector of Potato mop-top virus. Zeitschrift für Pflanzenkrankheiten und Pflanzenschutz 111, 145-150.

Oerke EC, Dehne H-W, Schönbeck F, and Weber A, 1994. Crop Production and Crop Protection – Estimated Losses in Major Food and Cash Crops. Amsterdam: Elsevier Science.

Oerke EC, 2006. Centenary Review. Crop losses to pests. Journal of Agricultural Science 144, 31–43.

Qu XS, Kavangh JA, Egan D, Lahert H, 2001: Studies on amoebae and cysts associated with the isolation of Spongopora subterranea f. sp. subterranea in vitro. Plant Pathology 50 (4), 420-426.

Qu XS, and Christ BJ, 2004. Genetic variation and phylogeny of Spongospora subterranea f.sp subterranea based on ribosomal DNA sequence analysis. American Journal of Potato Research 81, 385-394.

Qu XS, and Christ BJ, 2006. Single Cystosorus Isolate Production and Restriction Fragment Length Polymorphism Characterization of the Obligate Biotroph Spongospora subterranea f. sp. subterranea. Phytopathology 96, 1157-1163.

Qu XS and Christ BJ, 2007. In Vitro Culture of the Obligate Parasite Spongospora subterranea (Cercozoa; Plasmodiophorida) Associated with Root-Inducing Transferred-DNA Transformed Potato Hairy Roots. The Journal of Eucaryotic Microbiology 54 (6), 465- 467.

Pethybridge GH, 1911. Investigations on potato diseases. Journal / Department of Agriculture and Technical Instruction Ireland 11, 442.

Pimentel D, McNair S, Janecka J, Wightman J, Simmonds C, O’Connell C, Wong E, Russel L, Zern J, Aquino T, Tsomondo T, 2001. Economic and environmental threats of alien plant, animal, and microbe invasions. Agriculture, Ecosystems and Environment 84, 1– 20.

Potato Genome Consortium, 2011. Genome sequence and analysis of the tuber crop potato. Nature 475, 189-195.

Rizzo DM, Garbelotto M, Hansen EA, 2005. Phytophthora ramorum: Integrative research and management of an emerging pathogen in California and Oregon forests. Annual Review of Phytopathology 43, 309-335.

Salaman RN, 1937. The potato in its early home and its early introduction into Europe. Journal of the Royal Horticultural Society 62, 61–77, 112–123, 153–162, 253–266.

Salaman RN, Hawkes JG, 1949. The character of the early European potato. Proceedings of Linnean Society London 161, 71–84.

Schwaerzel R, 2002. Sensibilités des racines et tubercules des variétés de pommes de terre à la gale poudreuse et quelques résultats de lutte chimique. Revue Suisse Agricole 34, 261-266.

SEMAG, Saat- und Pflanzgut AG, 2005: Lyssach, Switzerland.

Spooner DM, Nuñez J, Rodríguez F, Naik PS, Ghislain M, 2005a. Nuclear and chloroplast DNA reassessment of the origin of Indian potato varieties and its implications for the origin of the early European potato. Theoretical and Applied Genetics 110, 1020–1026.

Spooner DM, McLean K, Ramsay G, Waugh R, and Bryan GJ, 2005b. A single domestication for potato based on multilocus amplified fragment length polymorphism genotyping. In: Proceedings of the National Academy of Science of the United States of America 102, 14694-14699.

Swisspatat, information available online http://www.kartoffel.ch/.

Theron DJ, 1999. Chemical control of powdery scab: a “new” threat in South Africa. In: Abstracts of conference papers, posters and demonstrations, 14th triennial conference of the EAPR, Sorrento, Italy, May 2–7, 1999, pp 265–266.

Torres H, Pacheco MA, and French ER, 1995. Resistance of potato to powdery scab (Spongospora subterranea) under Andean field conditions. American Potato Journal 72, 355-363.

Ugent D, 1970. The potato. Science 170, 1161—1166.

Ugent D, Pozorski S, and Pozorski T, 1982. Archaeological potato tuber remains from the Cosma Valley of Peru. Economic Botany 36, 182—192.

Ugent D, Dillehay T, and Ramirez C, 1987. Potato remains from a late Pleistocene settlement in south central Chile. Economic Botany 41, 17—27.

The United States National Nutrient Database 2012. Basic Report Nutrient data for 11352, Potato, flesh and skin, raw. Available online: http://ndb.nal.usda.gov/ndb/ foods/show/3147?fg=&man=&lfacet=&format=&count=&max=25&offset=&sort=&ql ookup=potato.

USBC, 1998. Statistical Abstract of the United States. US Bureau of the Census, US Government Printing Office, Washington, DC.

Valtueña FA, Preston CD, and Kadereit JW, 2011. Evolutionary significance of the invasion of introduced populations into the native range of Meconopsis cambrica. Molecular Ecology 20, 4318-4331. van de Graaf P, Lees AK, Cullen DW, and Duncan JM, 2003. Detection and quantification of Spongospora subterranea in soil, water and plant tissue samples using real-time PCR. European Journal of Plant Pathology 109, 589-597. van de Graaf P, Lees AK, Wale SJ and Duncan JM, 2005. Effect of soil inoculum level and environmental factors on potato powdery scab caused by Spongospora subterranea. Plant Pathology 54, 22-28. van de Graaf P, Wale SJ, and Lees AK, 2007. Factors affecting the incidence and severity of Spongospora subterranea infection and galling in potato roots. Plant Pathology 56, 1005-1013.

Vander Zanden MJ, 2005. The success of animal invaders. Proceedings of the National Academy of Sciences 102 (20), 7055-7056.

Vizzini A, Zotti M, and Mello A, 2009. Alien fungal species distribution: the study case of Favolaschia calocera. Biological Invasions 11, 417-429.

Wale SJ, 2000. Powdery scab control in Scotland. In: Merz U, Lees AK (eds) Proceedings of the first European powdery scab workshop, Aberdeen, Scotland, July 20–22, 2000, p 49. http://www.spongospora.ethz.ch/EUworkshop/proceedings.html.

Walsh JA, Merz U, and Harrison HG, 1996. Serological detection of spore balls of Spongospora subterranea and quantification in soil. Plant Pathology 45 (5), 884-895.

Woodham-Smith, C. 1962. The great hunger: Ireland 1946—1949. London.

Wuerzer B, 1965. Ergaenzende Untersuchungen ueber den Pulverschorf der Kartoffel und dessen Erreger Spongospora subterranea (Wallr.) Lagerh. Verlag E.Ulmer, Stuttgart, 104 p.

Zadoks JC and Schein RD, 1979. Epidemiology and Plant Disease Management. Oxford: Oxford University Press.

CHAPTER 2: GLOBAL GENETICS AND INVASION HISTORY OF THE POWDERY SCAB PATHOGEN, SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA

Rebecca D. Gau1, Ueli Merz1, Richard E. Falloon2,3, Patrick C. Brunner1*

1 Plant Pathology, Institute of Integrative Biology, ETH Zurich, CH-8092 Zurich, Switzerland; 2 New Zealand Institute for Plant and Food Research Limited, Lincoln, New Zealand; 3 Bio- Protection Research Centre, Lincoln University, Lincoln, New Zealand

Submitted to Plant Pathology on 6 December 2012

2.1 ABSTRACT

Spongospora subterranea f. sp. subterranea (Sss) causes two characteristic symptoms on potato (Solanum tuberosum), lesions on tubers and galls on roots. Both are economically important diseases in many potato production regions worldwide. Global seed trade is likely to be responsible for successful dispersal of Sss, despite quarantine measures. Chemical control is difficult so host resistance is a key component of disease management. Knowledge of the genetic diversity of a globally distributed pathogen is essential for risk assessment and successful resistance breeding. A combination of microsatellite and DNA sequence data was used to investigate the population structure and invasion history of Sss. South American populations were consistently more diverse compared to all other regions, consistent with the hypothesis that Sss originated in South America. Estimates of past and recent gene flow suggested that Sss was likely introduced from South America into Europe. Subsequently, Europe is likely to have been the recent source of migrants of the pathogen, acting as a “bridgehead” for further global dissemination. Quarantine measures must continue to maintain low global genetic diversity and to avoid exchange of genetic material between the native and introduced regions. Aggressive strains of the pathogen, which is already difficult to control, could arise from possible sexual recombination. Nevertheless, the current low global genetic diversity of Sss allows potato breeders to select for resistance which is likely to be durable.

2.2 INTRODUCTION

There have been many deliberate or accidental anthropogenic introductions of organisms beyond their original geographical ranges (Valtuena et al., 2011). Many important crops have been intentionally taken from their regions of origin and introduced to other suitable environments around the world, because of their value in food and fiber production (Bhagwat et al., 2008; Lashermes et al., 1997; Schrader & Unger, 2003). These introductions have often been accompanied by unintentional introduction of plant pathogens. Furthermore, increasing global trade in plant products carries the danger of introduction of pathogens, providing recurrent opportunities for new invasions (Dybdahl & Storfer, 2003; Montarry et al., 2010; Agrios, 2005). These pathogens have the potential to cause severe economic losses to crops, ornamental plants, or forests and can lead to severe problems in human or livestock nutrition (Anderson et al., 2004; Gonzalez-Varela et al., 2011).

There are numerous examples of invasive plant pathogens that have been introduced and distributed into new areas beyond their native ranges through human activities. Puccinia striiformis f. sp. tritici, the causal agent of stripe rust of wheat, was introduced into the USA almost 100 years ago and caused severe economic losses (Chen, 2005). The introduction of Phytophthora ramorum was also human-mediated, leading to sudden oak death in North America (Rizzo et al., 2005). The most infamous example is Phytophthora infestans, the causal agent of potato late blight. This disease was first reported in the USA 1843, and soon after appeared in Ireland where it led to the Irish potato famine with well-recognised and documented consequences of mass human starvation and forced migration (Fry et al., 1993). Once introduced in Europe, P. infestans was distributed worldwide via the international seed potato trade.

Invasive plant pathogens often successfully establish in new regions and spread over large areas (Vizzini et al., 2009). Despite the importance of these invasions, relatively little is known about the modes, timing or frequency with which they have occurred. Knowledge of these factors is important for prediction, prevention, and response to additional introductions (Desprez-Loustau et al., 2007;

Gonzalez-Varela et al., 2011). Molecular genetic data can be useful as an adjunct to ecological knowledge, to elucidate sources and routes for invasions, to identify the patterns of dispersal and the genetic composition of founding populations, and thus evaluate the reasons for invasion success.

It has long been accepted that populations of invasive alien species have reduced genetic variation compared to their source populations (Lawson Handley et al., 2011), and founder effects occur due to small population size in the introduction and establishment phases of biological invasions. This should hamper successful invasions, due to limited ability to respond to selective pressures. Usually, lag phases are expected between colonization and expansion, and during these lags multiple further introductions are thought to be needed to allow evolution of new adaptations (Lawson Handley et al., 2011). However, there are several examples of successful invasive plant pathogens with low genetic variability, including the potato pathogens P. infestans, and the wilt-causing bacteria Burkholderia solanacearum and Ralstonia solanacearum (Smith et al., 1995, 1998). A concept from invasion theory states that some invasive species are able to establish in a new territory after only one or a few introduction events and are then cut off from the source population. Further spread is exclusively outgoing from this introduced population to other regions. This phenomenon -called the “bridgehead effect” by Lombaert et al., 2010 - suggests that genetic diversity is not essential for invasion success, and that rapid adaptive evolution is possible despite strong bottlenecks and single introduction events. Although such invasive pathogens may have a clonal genetic structure, different phenotypes can be observed under varying climatic or environmental conditions, which was observed by Fry and Goodwin (1997) for P. infestans. Therefore, it is important to thoroughly characterize even clonal invasive plant pathogen species, concerning the interaction with their environment and hosts.

Spongospora subterranea f. sp. subterranea (Cercozoa, Plasmodiophoridae; hereafter abbreviated as “Sss”), is the causal agent of powdery scab, an economically important disease complex of potato. Powdery scab usually refers to the scabby lesions caused by the pathogen on potato tuber surfaces, but also root galls on

potato roots are a symptom of infections with Sss. The main host species of Sss is potato, Solanum tuberosum ssp. tuberosum. Other important hosts are Solanum tuberosum ssp. andigena and Solanum phureja, potatoes cropped in South America, the native region of potato, in the higher altitude regions of the Andes. Alternative hosts are wild potato species (South America) and other solanaceous plants, e.g. nightshade (Solanum nigrum), a common weed in potato production (Shah et al., 2010). It has been reported that different potato cultivars show differences in susceptibility to root and tuber infection (Schwaerzel, 2002). In Colombia it is well recognized that root galls are commonly formed on Sss-infected potato plants, and that tuber lesions caused by Sss are less frequently observed (Gilchrist et al., 2011). In most of the regions, where potato was introduced, both symptoms occur. The mechanisms behind the differences in susceptibility to the two forms of disease caused by Sss remain to be elucidated.

The life cycle of this obligate soilborne biotroph prevents natural long distance dispersal, because resting spores, clustered in sporosori, are formed in the soil in lesions on the surface of potato tubers and in galls on roots (Figs. 9a-c). Furthermore, the biflagellate primary zoospores (Fig. 9d), which emerge from resting spores, are able to swim only short distances in moist soil to infect new tissue. This reduced dispersal ability should lead to genetically distinct populations due to the lack of homogenizing gene flow. However, there is a considerable global trade in potatoes, and movement of Spongospora-infected seed potatoes is therefore likely to be responsible for successful short and long distance dispersal of the pathogen (Merz & Falloon, 2009).

Powdery scab is difficult to manage because contaminated soils remain infectious for many years due to the formation of numerous, highly resistant resting spores. From about the 1950s to the 1980s, seed potato tubers were routinely treated with mercury-containing pesticides to effectively protect potato crops from the disease (Merz, 2008). These treatments were suspended for human health and environmental reasons, and have not been replaced by fully effective seed tuber

Figure 9: Disease symptoms on potato caused by Spongospora subterranea f.sp. subterranea and life cycle stages of the pathogen: a) powdery scab lesions. b) root galls. c) Sporosorus containing resting spores. d) Single, biflagellate primary zoospore. Bars = 5 µm. Pictures a) and b) taken by R. Lamberts, pictures c) and d) taken by U. Merz. treatments. Breeding of resistant potato cultivars will play an important role in controlling the disease (Merz & Falloon, 2009). Until now, plant breeders screening cultivars and lines for susceptibility to powdery scab have been doing so without knowledge of genetic variability in Sss, and little is known about the role of sexual recombination in its lifecycle. The assumption that sporosori are the product of sexual recombination (Braselton, 1995) remains to be demonstrated. Very few studies have addressed genetic variation of Sss, mainly for detection and diagnostic purposes, using variation in the internal transcribed spacer (ITS) region (Bulman & Marshall, 1998; Qu & Christ, 2004). Because ITS is a rather conserved marker, these studies detected only slight genetic diversity among a limited number of samples. Only one reported study has used randomly amplified polymorphic DNA (RAPD) analysis to evaluate restriction fragment length polymorphism (RFLP) data (Qu & Christ, 2006), but only for North American samples from eight sites with each site represented by three samples. Variation among, but not within, geographic locations

has been detected. These studies have provided little insight into the present genetic population structure of Sss on a global scale.

One goal of the present study was to provide the first broad scale population genetic study of an important plasmodiophorid pathogen, based on newly- developed microsatellite markers and sequences of the actin gene and ITS region. Data were obtained from sporosorus samples originating from six continents and many potato producing regions. It has been suggested that South America is the native origin of Sss (Lyman & Rogers, 1915). Our second goal was to test the hypotheses that the pathogen was introduced from South America to Europe and subsequently was dispersed with Europe acting as a bridgehead through colonial (CIP, 2008) and/or contemporary seed potato trade (Rabobank, 2009) to the other introduced regions.

These new insights will expand knowledge of the pathogen and are required to improve powdery scab management, particularly for designing and implementing suitable quarantine strategies and for developing plant resistance as a sustainable method for practical powdery scab management.

2.3 MATERIALS AND METHODS

2.3.1 SAMPLES AND DNA EXTRACTION

Collecting of global populations of Sss followed an hierarchical sampling scheme (McDonald, 1997). The 19 countries of sample origin, combined into six regions and sample sizes per region are listed in Table 3 and in more detail in Supplementary Table 3. To simplify further naming in this paper, South American samples are specified as ‘native’ and all others samples as ‘introduced’. Total DNA from dried lesion scrapings from potato tubers or potato root galls was extracted using the cetrimonium bromide (CTAB) method (Winnepenninckx et al., 1993) or the QIAgen DNeasy Plant mini kit. Prior to further analyses, the presence of Sss DNA in the extraction was confirmed using a Sss-diagnostic PCR that amplifies a fragment of the ITS region (Bulman & Marshall, 1998).

Table 3: Number of Spongospora subterranea f.sp. subterranea samples collected, genotyped and sequenced, sorted by geographical region.

Region Specified as Countries No. of No. of No. of samplesa genotyped sequenced samples samples Europe Introduced Switzerland, 215 (8) 215 69 Germany, Netherlands, Norway, Sweden, Scotland, Iceland Africa South Africa 57 (0) 57 12 Asia South Korea, 98 (6) 98 38 Japan, Pakistan, Sri Lanka Australasia Australia, New 170 (3) 170 40 Zealand North United States 26 (0) 26 11 America South Native Colombia, 210 (39) 127 133 America Venezuela, Peru, Ecuador Total 776 (56) 693 303 aNumber of root gall samples in parentheses.

2.3.2 NUCLEOTIDE SEQUENCES AND PHYLOGENETIC RECONSTRUCTION

The taxonomy and species diversity of the Plasmodiophoridae is poorly investigated. Since we compared samples from a broad range, including different host plants, tuber lesions and root galls, we first conducted a phylogenetic analysis to avoid the comparison of cryptic species. A subset of 303 Sss samples representing all regions was analyzed for nucleotide sequence variation at the ITS region (389 bp) and the partial actin gene (615 bp). A standard PCR protocol was used to amplify ITS using the primer pairs Spo8 and Spo9 (Bulman & Marshall, 1998). A nested PCR using newly designed “outer” and “inner” primer pairs based on the published actin sequence (AY452193.1) had to be applied to amplify the actin gene (Supplementary Table 2). Products were sequenced with an ABI 3730 xl sequencer (Applied Biosystems). Sequences were edited using SEQUENCHER (Gene Codes Corporation). BLASTN searches (Altschul et al., 1990) were carried out against the GenBank data base to verify that the sequences were not from other organisms. The sequences of the two genes were concatenated and aligned using ClustalW (Higgins et al., 1996). ITS and/or actin sequences of other Plasmodiophorids were retrieved from GenBank as outgroups, including Spongospora subterranea f. sp. nasturtii, the closest known relative of Sss.

Maximum Likelihood (ML) phylogenetic trees were constructed using MEGA version 5 (Tamura et al., 2011). The ML analyses were performed using the Kimura- 2-parameter model with a discrete Gamma with 5 rate categories. The model search algorithm implemented in MEGA selected this model as having the lowest BIC score (Bayesian Information Criterion), which is considered to describe the nucleotide substitution pattern the best. All positions with less than 70% site coverage were eliminated. Statistical node support was estimated using 500 bootstrap replications. We also constructed a parsimony-based haplotype network using TCS (Clement et al., 2000) and the haploNet function of R (http://www.r-project.org/), to better visualize haplotype relationship and frequencies across regions.

2.3.3 MICROSATELLITE LIBRARY CONSTRUCTION

An enriched microsatellite library (Glenn & Schable, 2005) was established to develop microsatellite primers for Sss. In short, Sss DNA from a Swiss tuber lesion sample was digested with RsaI and XmnI to obtain ca. 500 bp blunt-ended fragments. From these fragments an enriched microsatellite library was produced using magnetic beads (MyOne T1 Streptavidin Dynabeads, Invitrogen) and biotinylated oligonucleotides representing the microsatellite motives (AT)10, (CT)10,

(TTG)8 and (TCG)8. Ligation and cloning of enriched fragments were performed, using the TA Cloning® Kit (Invitrogen). Blue-white selected transformants were sequenced using an ABI 3730 xl sequencer (Applied Biosystems) and screened for microsatellites. A BLASTN search (Altschul et al., 1990) with all obtained sequences was performed against the GenBank database to identify non-specific Sss fragments (e.g. from potato or soil organisms). The web-based PRIMER3 program (Rozen & Skaletsky, 2000) was used to design the specific microsatellite primers.

PCR amplifications using the msat246 primers produced two fragments of 140 bp and 160 bp respectively. Cloning and re-sequencing suggested a duplication of this locus in the genome with the shorter fragment having a 20 bp deletion. Since the deletion was located outside of the microsatellite motif (CAA) and the two loci were unlinked, both were scored and analyzed.

2.3.4 MICROSATELLITE ANALYSES AND POPULATION STRUCTURE

Six polymorphic microsatellite loci yielding unambiguous PCR products were selected to genotype 693 Sss samples. Separate PCRs were carried out for each locus using fluorescent-labeled primers (Supplementary Table 2). Amplicons were separated using either an ABI 3730 xl or an ABI 3130 sequencer (Applied Biosystems). Data processing and calling of allele-sizes was performed using internal GeneScan LIZ600 standards and the GENEMAPPER software (both Applied Biosystems).

Based on observation of either one (homozygote) or two microsatellite alleles (heterozygote) among the samples, we assumed that Sss is a diploid organism, and all analyses were performed accordingly. Isolates with identical multilocus genotype

(MLG; i.e. possessing the same allele at all microsatellite loci) were considered distinct clones and only one MLG was retained per region for subsequent analyses. We used the program GENODIVE (Meirmans & Van Tienderen, 2004) assuming an infinite allele model of microsatellite evolution to calculate allele frequencies (provided in Supplementary Table 3), site specific genotypes, the clonal fraction which describes the proportion of individual samples originating from asexual reproduction, Nei’s genotypic diversity (Nei, 1987) and gene diversity for each region. Estimates were corrected for differences in sample size using the rarefaction (Petit et al., 1998). Significance of differences between regions was assessed using the implemented bootstrapping approach. Linkage equilibrium of MLGs (i.e. the random association of microsatellite loci) was assessed as a proxy to estimate the S amount of sexual recombination by estimating the index of association IA using LIAN 3.5 (Haubold and Hudson, 2000).

The population structure of Sss was explored on the individual and the region levels using the multivariate clustering approach of principle component analysis (PCA) based on the covariance matrix of allele frequencies. The optimal number of clusters on the individual level was further assessed using K-Means clustering (MacQueen, 1967) as implemented in GENODIVE. The method uses a pairwise matrix of distances between all observations and divides these observations into an a priori assigned number (k) of groups in such a way that the among-groups Sum of Squares is maximized. We used the option of simulated annealing based on a Monte Carlo Markov Chain (MCMC; one million steps) and repeated the analysis five times to ensure that the clustering did not get stuck in local optima. The optimal value of k was inferred from the Calinski & Harabasz (1974) pseudo-F-statistic, that is particularly suited when there is non-random mating, and for clustering individuals

(Meirmans, 2012). Differentiation on the region level was estimated as pairwise FST values using the method of Weir and Cockerham (1984), and the significance of genotypic differentiation was assessed using the permutation approach (10,000 iterations) implemented in GENODIVE.

2.3.5 INFERRING MIGRATION HISTORY

We inferred migration rates between regions on two temporal scales to test the two main hypothesis of Sss distribution. The first hypothesis is the historic introduction of Sss from South America to Europe through colonial trade (Lyman & Rogers, 1915; CIP, 2008). These past migration rates were estimated using the maximum-likelihood approach implemented in MIGRATE (Beerli & Palczewski, 2010). Based on the FST analyses that suggested non-significant genetic differentiation (see results below), Colombia tuber lesions and Venezuela tuber lesions were pooled into “South America tuber lesions”. The starting values for the migration rates were estimated based on pairwise FST. Markov chain settings were 10 short chains, three long chains with a burn-in of 10, 000 trees and averaging over long chains. We applied a three- temperature heating scheme and selected the Brownian mutation model for microsatellite evolution. Convergence of parameter estimates was controlled by checking the MCMC process for high acceptance ratios (>95%), for stationarity of data-likelihood estimates, and by running the entire analyses three times to ensure consistency of results.

The second hypothesis to be tested was that Europe acted as a bridgehead for subsequent dispersal of Sss to the other introduced regions, for example through contemporary seed potato trade (Rabobank, 2009). Estimates of these recent migration rates were performed using the software BayesAss v. 3 (Rannala & Yang, 2003). The Bayesian approach estimates the proportion of genotypes in a population composed of migrants over the last few generations. After checking preliminary runs that log-probability fluctuations were restricted to the burn-in phase, indicating good convergence, the final mixing parameter for allele frequencies was set to 0.3 and the mixing parameter for inbreeding coefficients was set to 0.2. We allowed for a burn-in of 1,000,000 iterations and a MCMC sampling of 10,000,000 iterations.

2.4 RESULTS

2.4.1 PHYLOGENY

We used the concatenated nucleotide data from the actin and ITS region to reconstruct the phylogenetic relationship of Sss and other Plasmodiophorids. The distribution of distinct haplotypes was extremely skewed. Of the 19 distinct concatenated haplotypes (cHap) 17 were found in the native regions and two in the introduced regions. Of these two, cHap6 was found in all introduced regions and accounted for 96% of haplotypes found in these regions. None of the cHaps was shared between introduced and native regions. A detailed list of haplotype distribution is given in Supplementary Table 4 and visualized in Supplementary Fig. 1.

The reconstructed ML tree (Fig. 10) clustered all Sss haplotypes in a well supported monophyletic clade (96% bootstrap replicates) and clearly distinct from its sister species S. spongospora f.sp. nasturtii. Average pairwise distances among cHaps were low (0.017 substitutions / site) and there was no significant substructure within the Sss clade. However, there was a tendency of cHaps to cluster according to inoculum as most haplotypes collected from tuber lesion (tu) formed one group, and most haplotypes collected from root galls (ga) clustered in a second group. CHaps are inoculum specific in the native region, i.e. there were no shared haplotypes between root galls and tuber lesions. In contrast, cHap6 that accounted for most haplotypes in all introduced regions was associated to both root galls and tuber lesions.

Figure 10: The evolutionary relationship of Spongospora subterranea f.sp. subterranea and other Plasmodiophorids was inferred by using the Maximum Likelihood method on the concatenated ITS and actin sequences. The bootstrap consensus tree inferred from 500 replicates is taken to represent the evolutionary history of the taxa analyzed. Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. The number of samples possessing a particular haplotype is given in parenthesis. The underlined haplotypes were detected in the introduced regions Europe (EU), Africa (AF), Asia (AS), Australasia (AU), and North America (NA). tu, haplotypes detected on tuber lesions; ga, haplotypes detected on root galls. GeneBank accession number are given for the outgroup species.

2.4.2 CLONAL AND GENETIC DIVERSITY

A complete list of microsatellite allele frequencies for the six regions is available in Supplementary Table 3. Measures for genetic and genotypic diversity are summarized in Table 4. All results of genotypic diversity and genetic diversity showed significantly lower values for the introduced regions compared to the native regions. The six microsatellite loci had two to 12 alleles per locus (average of six alleles per locus) and 35 alleles in total. A total of 131 different multilocus genotypes (MLG) were detected among the 693 samples analyzed in the six regions. South

America possessed most of the detected genotypes. Out of 127 genotyped samples, 82 MLGs were detected, resulting in the smallest clonal fraction of cf = 0.346 (Table 4). Of these 82 distinct MLGs, 81 were site-specific, found only in South America. Within South America, the root gall (cf = 0.226) and tuber lesion samples (cf = 0.382) had similar diversities. In contrast, all introduced regions had significantly greater clonal fractions, ranging from cf = 0.906 (Australasia) to cf = 0.632 (Africa). Most striking was the low number of site-specific genotypes ranging from 1 to 13 among the introduced regions compared to 81 detected in South America. Only one multilocus genotype was shared between introduced regions and the native region. In the 566 samples of the pooled introduced regions, 49 MLGs were detected, resulting in a significantly greater clonal fraction (cf = 0.913) compared to South America (Table 4). No significant differences in pairwise comparisons between the introduced regions were found, with exception of North America. North America had a genetic diversity of G = 0.151, which is roughly five to six times less than found in any other region.

S The standard index of association (IA ) for each regions and pooled regions was measured to test for statistical independence amongst alleles at each of the six S microsatellite loci (Table 4). Based on this method, only the IA = 0.0144 for Africa did S not significantly deviate from linkage equilibrium. In contrast, IA estimates for all other regions indicated significant linkage disequilibrium, suggesting that Sss is only very rarely undergoing sexual recombination.

Table 4: Estimates of clonal and genetic diversity parameters based on six microsatellite markers of Spongospora subterranea f.sp. subterranea sorted by sampled regions.

a b c d e f g Region N cnum ssg cf G H IA Europe 215 26 11 0.879 0.885 0.225 0.0339*** Africa 57 21 13 0.632 0.917 0.234 0.0144ns Asia 98 16 7 0.837 0.835 0.249 0.0673*** Australasia 170 16 2 0.906 0.878 0.21 0.0131*** North 26 3 1 0.885 0.151 0.013 na America Total 566 49 34 0.913 0.91 0.235 0.0205** Introduced

South 39 29 26 0.226 0.983 0.391 0.0630** America Root galls South 88 55 54 0.382 0.972 0.314 0.0269* America Tuber lesions Total Native 127 82 81 0.346 0.985 0.461 0.1138*** a: N = Sample size b: cnum = Number of multilocus genotypes c: ssg = Site specific genotypes; clones specific to a region and not shared with other regions d: cf = Clonal fraction; proportion of individual samples originating from asexual reproduction e: G = Nei’s corrected diversity (genotypic diversity) f: H = Nei’s Gene Diversity g : IA = Index of association to tests the null hypothesis of linkage equilibrium for multilocus data. Significance of deviation from equilibrium expectations are indicated by asterisks. *, p < 0.05, **, p < 0.01, ***, p < 0.001; ns = non-significant; na = not enough diversity for estimation

2.4.3 POPULATION STRUCTURE

The microsatellite data was subjected to PCA on two levels to explore the population structure of Sss (Fig. 11). On the individual level the first two PCA axes explained 46% and 17% of the total genetic variance, respectively. While no sub-clustering by region was observed, the canonical plot showed clear distinction between individuals from the introduced regions and individuals from the native region. These two global clusters were confirmed as the optimal number of groups using the K-means approach on the individual level (Fig. 11a).

On the regional level PCA-Axis 1 (72%) and Axis 2 (19%) together explained 91% of the total genetic variance. The clustering confirmed the shallow sub-structure among the introduced regions, grouping the five regions very closely together.

However, based on the pairwise FST estimates (Table 5) only Australasia and Asia were not significantly differentiated from each other. Also in contrast to the individual level analyses, PCA and FST analyses detected significant substructure among the native South American regions. Here, independent of geographic proximity, the tuber population from Colombia and the tuber population from Venezuela formed a cluster distinct from the gall population from Colombia (Fig. 11b).

Table 5: Pairwise estimates of FST between sampled regions (above diagonal) and corresponding p-values (below diagonal)

Region EU AF AS AU NA CO CO galls VE lesions lesions Europe --- 0.073 0.081 0.116 0.176 0.710 0.654 0.645 Africa < 0.001 --- 0.043 0.041 0.303 0.689 0.617 0.607 Asia < 0.001 < 0.001 --- 0.006 0.361 0.667 0.612 0.609 Australasia < 0.001 < 0.001 0.028ns --- 0.394 0.704 0.663 0.652 North < 0.001 < 0.001 < 0.001 < 0.001 --- 0.877 0.712 0.691 America Colombia < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 --- 0.507 0.071 tuber lesions Colombia root < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 --- 0.448 galls Venezuela < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.002 ns < 0.001 --- tuber lesions ns non-significant after Bonferroni correction

Figure 11: Principle Component Analyses (PCA) of Spongospora subterranea f.sp. subterranea multilocus genotypes based on covariances of allele frequencies. (a) PCA performed on individuals resulted in axis 1 explaining 46% and axis 2 explaining 17% of the genetic variance. The dashed lines encircle the two clusters identified by K-means clustering. (b) PCA performed on regions resulted in axis 1 explaining 72% and axis 2 explaining 19% of the genetic variance. The dashed lines encircle regions that are not differentiated based on FST analysis.

2.4.4 INFERRED MIGRATION HISTORY

Historical migration rates M (scaled by the mutation rate) were estimated with MIGRATE-N. Preliminary analyses including pairwise comparisons of all regions yielded inconsistent results, likely due to the shallow population structure among the introduced regions. We therefore restricted our analyses to estimating historic migration rates between Europe and the native regions, and in a second analysis between pooled introduced regions and the native regions. We found strong asymmetrical gene flow (Fig. 12). All estimates for South America tuber lesions and South America galls into Europe or into the pooled introduced regions had confidence ranges >1 and ranged from M = 2.14 to M = 7.23. In contrast, migration estimates for the opposite directions indicated no significant historical gene flow from Europe or the pooled introduced regions into South America with estimates ranging from 0.00 to 0.34 (Supplementary Tables 5a,b).

We estimated recent migration rates between all regions using BayesAss. The most striking finding was the high proportion of European migrants in all other introduced populations, ranging from 17% in North America to 29% in Australasia (Fig. 12). In sharp contrast, the proportions of migrants originating from other regions than Europe were all non-significant from zero based on confidence intervals. In contrast to our estimates of past migration rates, we did not detect any significant recent gene flow between the South American regions and the introduced regions (Supplementary Table 6).

(United (United

Confidence intervals (5% (5% intervals Confidence

Data extracted from historic documents, documents, historic from extracted Data

lanum tuberosum) among Europe (EU), Africa Africa (EU), Europe among tuberosum) lanum

n activities; 1) The conquistadors brought the potato potato the brought conquistadors The 1) activities; n

nt of the receiving population. population. receiving the of nt

alia (1787). (1787). alia

ved white arrows are estimates of historic gene flow using MIGRATE given as migrants scaled by the bythe scaled migrants as given MIGRATE using flow gene historic of estimates are arrows white ved

he potato was taken from Europe to Asia; 3) In 1613, the potato was taken from England to Bermuda and from there to Virginia Virginia to there from and Bermuda to England from taken was potato the 1613, In 3) Asia; to Europe from taken was hepotato

1593); 2). In the early 1600s, t 1600s, early the In 2). 1593);

: Global gene flow estimates for Spongospora subterranea f.sp. subterranea (Sss) and historic dissemination of the potato (So potato the of dissemination historic and (Sss) subterranea f.sp. subterranea Spongospora for estimates flow gene Global :

ure 12 ure

Fig Cur (SA). America South and (NA), America North (AU), Australasia (AS), Asia (AF), perce as given BAYESASS, by inferred regions between flow gene recent significant indicate arrows black Curved rate. mutation huma by mediated potato of steps dissemination historic the represent arrows straight Numbered parenthesis. in given 95%)are (1567 Europe to Austr and (1769) Zealand New (1880s), Africa Southern to England from potato of dissemination Further 4,5) 1620; States)in Peru. CIP, by provided

2.5 DISCUSSION

To our knowledge, the present study is the first extensive population genetic characterization of a plasmodiophorid plant pathogen. The determination of the genetic structure of Sss was challenging, because suitable genetic markers were lacking and plasmodiophorids are obligate biotrophic pathogens that cannot be grown as pure culture on artificial media (Agrios, 2005; Merz & Falloon, 2009). For this reason, sequencing of housekeeping genes and the ITS region has a longer tradition in Sss than for other gene sequencing approaches (Bulman & Marshall, 1998; Down et al., 2002; Qu & Christ, 2006). We were able to successfully genotype 693 individual samples of Sss, originating from all relevant continents, different climate zones, different potato subspecies and cultivars, and also from different sources of Sss sporosori, with six newly developed polymorphic microsatellite markers. Our results of microsatellite and sequencing data showed that South American populations were consistently more diverse compared to all other regions. Estimates of migration rates further suggested a historic gene flow from South America to Europe and recent gene flow from Europe to the other introduced regions. Consequently, we conclude that Sss is very likely to have been introduced from South America to Europe and, with Europe as a “bridgehead”, was then further globally disseminated, with no or a very limited number of new introductions from the native region.

2.5.1 ROLE OF HOSTS, INOCULUM AND THE POTENTIAL OF CRYPTIC TAXA

Although our collection included Sss samples extracted from the three most common cultivated potato hosts S. tuberosum ssp. tuberosum, S. tuberosum ssp. andigena, and S. phureja, no genetic differences in Sss were related to the different hosts. On Colombian potatoes, tuber lesion formation is rare (Gilchrist et al., 2011); for that reason, tuber lesion samples have been obtained from only one region in Colombia. However, we do not think that this sampling introduced a bias caused, for example, by regional adaptation, since FST results showed that the Colombian tuber lesion samples clustered with the Venezuelan tuber lesions (Fig. 10). On S. phureja only root galls can be found. This Solanum species may have specific resistance to

tuber infection. Among the many cultivars of S. tuberosum ssp. tuberosum, which is the potato host widely grown in the introduced regions, both disease symptoms (root galls and tuber lesions) can be observed. However, there are also cultivar differences in susceptibility to disease on tubers and roots (Schwaerzel, 2002).

South American samples collected from galls were genetically distinct from those collected from lesions. Interestingly, no such distinction could be detected among gall and lesion samples from the introduced regions. This could indicate an ecological adaptation in the native regions due to co-evolutionary processes and/or competitive exclusion. This might be especially true in Colombia, where little or no potato exchange has occurred to date with other countries (E. Gilchrist, Corporación Universitaria Lasallista, Colombia, personal communication), and only S. tuberosum ssp andigena and S. phureja hosts are cropped. Our phylogenetic analysis did not indicate a significant distinction between samples originating from root galls and tuber lesions and MIGRATE results suggested strong gene flow between root gall and tuber lesions in the native South American regions. Taken together, we conclude that Sss from root galls and Sss from tuber lesions are not distinct species or subspecies. However, they deserve further investigations that could likely result in the identification of different ecotypes.

2.5.2 INVASION SCENARIO OF SSS AND EVIDENCE FOR A BRIDGEHEAD EFFECT

Given the very restricted dispersal ability of Sss and our samples covering six continents, we expected to find considerable regional substructure due to restricted gene flow, founder effects or local adaptations. However, our PCA analyses indicated shallow population substructures, both for the introduced and native regions (Fig. 11). In contrast, we observed marked differences in genetic diversity between South America and the other regions. South American samples were genetically more variable, consistent with the hypothesis that the native region of Sss is South America – which is also the native region of potato. Reduced genetic diversity and lack of substructure in collections from introduced regions can be attributed to founder effects, and are indicative of a rapid global invasion process and/or

restricted origin of all introduced populations, combined with a lack of recurrent gene flow from the native region.

The results of our migration analyses indicated a historic gene flow from South America into Europe. In contrast, estimates of recent gene flow suggested no gene flow between South America and Europe, but all introduced regions received migrants from Europe (Fig. 12). We included historic documentation of dissemination of potato (CIP, 2008) to reconstruct the invasion scenario of Sss (Fig. 12). In combination with our results, we hypothesize the following “bridgehead” invasion scenario of Sss. The native region and source of Sss is located in South America. It is most likely that Sss was introduced from South America to Europe on contaminated potato specimens, in the second half of the sixteenth century. This could possibly be through the exploration and migratory activities associated with the conquistador era. Supporting this theory is the first documented report of powdery scab published in Germany in 1842, describing the disease as a well-known problem for farmers (Wallroth, 1842). From the bridgehead in Europe, Sss was spread subsequently due to the lack of plant quarantine or control measures to the North American and European colonies in Africa, Asia and Australasia. North America had the lowest genetic and genotypic diversity of all introduced regions (Table 4). Since potatoes were first introduced from Europe to the Bermudas and from there to Virginia (Fig. 12), this possibly resulted in a secondary genetic bottleneck in the North American populations of Sss. No significant further exchange of infected potatoes has occurred between South America and the introduced regions since the first introduction to Europe.

To pinpoint the exact location of origin of Sss, a more thorough sampling of South American Sss populations is necessary. Results may reveal the region of origin of Sss to be in Peru, if the pathogen has co-evolved with the edible potato, by far the most important host today. This plant was first collected by hunter gatherers in the Lake Titicaca region in Peru around 3,000 B.C., and was later distributed to other Andean countries and cultivated in the first agricultural societies in South America

(CIP, 2008). Supporting this hypothesis is the most basal haplotype (cHap19) of the Sss clade reconstructed in the phylogenetic tree (Fig. 10) that was collected in Peru.

2.5.3 GLOBAL TRADE OF SEED POTATOES REFLECTS RECENT MIGRATION PATTERNS OF THE PATHOGEN

Sss is a soilborne obligate biotroph with very restricted natural dispersal abilities. The most likely way that Sss has been dispersed throughout the world is, therefore, via the movement and trade of seed potatoes. Sss can be transmitted on seed tubers, either as sporosori in visible powdery scab lesions or as non-visible surface contaminants (Diriwaechter & Parbery, 1991). In this way, the pathogen might invade new regions as a contaminating organism in consignments of shipped seed potatoes.

According to the recently published potato trade map (Rabobank, 2009), Europe, mainly the Netherlands, is by far the greatest exporter of seed potatoes worldwide. In accordance with our results, this strongly suggests that Europe is the contemporary global distributor of potatoes potentially infected with Sss to other introduced regions of the world.

2.5.4 RISK ASSESSMENT FOR SSS, AND COMPARISON WITH P. INFESTANS

It is likely that several important potato pathogens were carried out of South America together with the potato, in contaminated soil or infected plant material. Parallels between Sss and other important potato pathogens can be found. Potato late blight is now considered as a re-emerging disease. Several P. infestans introduction events took place in the nineteenth and twentieth centuries outgoing from Latin America to Europe. The pathogen reproduces sexually given the presence of the compatible mating types A1 and A2, but sexual reproduction was not observed until the mid-1970s, when potatoes were imported on large scale after a drought in 1976 (Montarry et al., 2010). Before this specific event, P. infestans reproduced asexually, and studies revealed it was a worldwide clone (Fry et al., 1993). A similar situation exists with other successful globally distributed and economically important potato pathogens, including the two bacterial wilt

pathogens Burkholderia solanacearum and Ralstonia solanacearum (Smith et al., 1998; Smith et al., 1995). Like P. infestans and the bacterial wilt pathogens, Sss was able to successfully invade regions far beyond its native range in South America despite its reduced genetic diversity. Multiple introductions may greatly modify the population genetic structure of plant pathogens in their new areas and influence their evolutionary potential (McDonald & Linde, 2002; Stukenbrock & McDonald, 2008). We found a smaller proportion of clones in native South American samples S compared to the introduced regions. However, estimates of IA for both introduced and native regions suggested a predominantly asexual reproduction. This is counter- intuitive, because the DNA was extracted from sporosori, a resting structure assumed to be the product of sexual recombination (Braselton, 1995). This knowledge gap on the occurrence of sexual reproduction in the life cycle of Sss remains to be adequately elucidated.

2.5.5 STRATEGIES FOR CONTROLLING SSS AND POTATO BREEDING

The present study confirms that development of potato breeding lines and cultivars resistant to powdery scab is likely to be an efficient and sustainable way to manage the disease (Merz & Falloon, 2009). Chemical control measures available are not completely reliable and are increasingly undesirable for environmental and consumer resistance reasons. Given the great clonality of Sss in the introduced regions, resistance screening during breeding is not likely to be faced with variable virulence in pathogen populations. The risk of virulence differences within the clonal lineages (Blandón-Díaz et al., 2012) seems to be low, as screening trials during 4 years with cultivars selected for their susceptibility to powdery scab showed no differences in the performance of the genotypes (Merz et al., 2012). The cultivar ‘Gladiator’, bred in New Zealand (Genet et al., 1995) with very low Sss susceptibility to tuber and root infection, performed well in all years and locations, even in those where inoculum levels were high and severe powdery scab epidemics occurred.

The similarities between Sss and other potato pathogens must be considered, however. New introductions of Sss genotypes, particularly from South America, increase the potential of more aggressive inoculum, e.g. due to recombination. This

could lead to multiple pathotypes and additional challenges for resistance breeding. In order to prevent such introductions, strict quarantine measures for potato import need to be established, or where they exist, strictly enforced. This will help to preserve the long-term benefit of resistant cultivars and maintain low genetic variability of the pathogen on a global scale.

2.6 ACKNOWLEDGEMENTS

This work was supported by funding from Horticulture Australia Ltd (APRP1, University of Tasmania), the ETH Zürich, the United Kingdom Potato Council, the New Zealand Ministry of Science and Innovation, Horticulture New Zealand, and the Swiss Federal Office of Agriculture.

Data analyzed in this paper were generated in the Genetic Diversity Centre (GDC) of ETH Zurich, Switzerland, and at the New Zealand Institute for Plant and Food Research Limited, Lincoln, Canterbury, New Zealand.

We thank the following colleagues for providing Sss collections: Georg Babu (Sri Lanka), Kirsten Bundgaard (Norway), Wilbert Flier (Ecuador, Peru, Netherlands), Celsa Garcia (Colombia), Hans-Reinhard Hofferbert (Germany, Netherlands), Shamim Iftikhar (Pakistan), Kim Jeom-Soon (South Korea), Takato Nakayama (Japan), Xinshun Qu (USA), Dorian Rodriguez (Venezuela), Herbert Torres (Peru), Maria Sandgren (Sweden), Halldór Sverrisson (Iceland), Stuart Wale (Scotland), Tonya Wiechel, (Australia), and Jessica Wright (South Africa).

2.7 SUPPLEMENTARY MATERIAL

Supplementary Figure 1: Concatenated ITS and actin haplotype network inferred by the software TCS from sequencing data of 308 global samples of Spongospora subterranea f.sp. subterranea.

101

Supplementary Table 1: Spongospora subterranea samples examined in this study.

Origin Number Number of Solanum Cultivar Tissue Year of sites samples (N) subspecies Switzerland, Europe Wallestalden, 1 31 Tuberosum Antonia, Talent, Rosagold, Lesions 2009 Langnau Pepite, BP, GO2TT118004, Bintje, Agria, Zorba, Mirage, AR98-1196, AR99-1200, ST98-74-9, G00SC235, Erika, GO2TT118004, B00/244/51, VR-98-72, Salome, Musica, Challenge Wallestalden, 2 30 Tuberosum Lesions 2004, Langnau Estima, Ratte, Saturna and Galls 2005 Solothurn, Biezwil 1 25 Tuberosum Agria Lesions 2004 Gebr. Kobel 1 1 Tuberosum Agria Lesions 1998 Semag 7 7 Tuberosum Agria, Erntestolz, Markies, Bintje Lesions 1998, 2000, 2003 Luzern, Kägiswil 2 19 Tuberosum Agria Lesions 2004 Germany, Europe Nordrhein- 1 23 Tuberosum Agria, E 05/421/529, M Lesions 2009 Westfalen, 05/127/29, E 05/60/8, E and Galls Meinersen 05/251/215, E 05/99/45, B 05/251/115, B 05/139/34, M 05/68/34, SP 05/582/58, Seedling in peat soil Netherlands, Europe Limburg 1 3 Tuberosum Unknown Lesions 2009 Unknown 1 1 Tuberosum Bintje Lesions 2000 Norway, Europe Farmen, Kvelde 1 25 Tuberosum Redstar Lesions 2008 Rustad, Romeldal 1 25 Tuberosum Beate Lesions 2008 Iceland, Europe South Iceland, 1 22 Tuberosum Red Icelandic Lesions 2009 Thykkvibær Scotland, Europe Scotland, Aberdeen 3 3 Tuberosum Estima, Nadine, Mixture Lesions 1999 Sweden, Europe Uppsala 2 2 Tuberosum Kultivator Lesion 1994, scrapings 2004 South Africa, Africa Kwazulu-Natal & 1 25 Tuberosum Mondial Lesions 2009 Sandvelt Sandvelt, Western 1 32 Tuberosum Mondial Lesions 2009 Cape Pakistan, Asia Sharan, Kaghan 1 16 Tuberosum Cardinal, Barma Lesions 1994, Valley 2009 Sri Lanka, Asia Unknown 1 4 Tuberosum Mixture, Calwhite, Redlasoda, Galls 2010 Chieftain, Keuka Gold, Granola South Korea, Asia Hoenggye 1 25 Tuberosum Superior Lesions 2010 Wangsan 1 25 Tuberosum Superior Lesions 2010 Unknown 1 1 Tuberosum Unknown, minitubers Lesions 2005 Japan, Asia

102

Hokkaido, Kyogoku- 1 25 Tuberosum Irish cobbler Lesions 2009 Town Hokkaido, Unknown 1 3 Tuberosum Unknown (KNP, QNO, TUM) Lesions 1999 Australia, Australasia Ballarat 3 57 Tuberosum Unknown Lesions 2006 New Zealand, Australasia New Zealand, 1 106 Tuberosum Rua, Agria, and Desiree Lesions 1997, Canterbury, Lincoln 2004, 2008 New Zealand, 1 1 Solanum Bolivian Weed Lesions 2008 Canterbury, Lincoln chacosense New Zealand, 1 3 Tuberosum Shepody, Kennebec, Iwa Lesions 2008 Canterbury, Lincoln New Zealand, 1 1 Tuberosum Ranger Russet Galls 2008 Canterbury, Lincoln New Zealand, 1 2 S. phylum Unknown Galls 2008 Canterbury, Lincoln United States of America, North America Pennsylvania, Potter 1 25 Tuberosum Shepody, Yukon Gold, Lesions 2009 County B1992/106, AF2376-5, NY139, BC001357-4, FL30, W2310-3, Dakota Jewel, W2133-1, Atlantic, NYB38-40, Katahdin, Snowden, Dakota Diamond, Superior, Kennebec, Beacon Chipper, Dark Red Norland, Chieftain, NY141, BC001306-2, FL22, AF2393-7, B2452-3 Madras, Oregon 1 1 Tuberosum NDO 1496-1 Lesions 1995 Colombia, South America Villapinzón, Sonsa 1 18 Andigena or Unknown Lesions 2008 Bajo Phureja Nariño, Coba Negra 1 4 Andigena or Unknown Galls 2008 Phureja Nariño, La Marqueza 1 5 Andigena or Unknown Galls 2008 Phureja Nariño, Río Bobo 1 6 Andigena or Unknown Galls 2008 Phureja La Union, Antioquia 1 25 Phureja Chuscalito, Buena Vista, La Galls 2010 Madera Soka, Valluelitos Zuia la Cana, Valluelito Capiro, Don Juan de Jesu, Chuscalito Capiro Venezuela, South America Mérida State, El 1 24 Andigena No Name Lesions 2010 Llano Mérida State, Los 1 25 Andigena Unica Lesions 2010 Muros de Tadeo Mérida State, La 1 22 Andigena Unica Lesions 2010 Toma Mérida State 3 80 Tuberosum Granola Lesions 2011 and 1 gall sample Peru, South America Peru 1 1 Andigena Unknown Lesions 1996 Peru 1 1 Tuberosum- Mariva Lesions 1999 Andigena Hybrid Ecuador, South America Ecuador 1 1 Andigena Unknown Lesions 2000

103

Supplementary Table 2: PCR protocols and primers used to amplify microsatellite loci and partial sequences of the actin gene and ITS region in Spongospora subterranea f.sp. subterranea.

Marker Name Synthetic Fragment Motif Label Primer Sequence from 5’ Length [bp] Act ~650 Actin gene none F: ACTCCGGCCATGTATGTCG outer region R: TGCCTCGTCATACTCATCTTTG ActN 615 Actin gene none F: inner region GGCTGTTCTATCTCTATACGCATC R: CGACAACGATGACAAAATCG Msat6 210 GAC/CAC PET F: PET- GGGAGATCAGCTCCAGATCA R: GGTGTCGTTGTTAGGGTTGC Msat45 268 TCA NED F: NED- GAGCGAAACTGAGGGGATTC R: CCTTCTCCAGGATCGGGATC Msat84 291 GCT VIC F: VIC- GAGCTAGACCTACCGACGACT R: ACGTCAGTGATCCAAGCACA Msat103 193 CT HEX F: HEX- GATGATCGTCGGAATTCGTT R: CATCTCGAGCTTCGTTCAGC Msat246 150 CAA FAM F: FAM- CCAGACAACCCCTGTTCAGT R:CCAAGCGTTAACCCACTGTT Protocol Actin 94°C for 3 min initial denaturation; 45 cycles of 92°C for 45 s, 50°C for 1 min, 72°C PCRs for 1:30 min; and 72°C for 7 min final elongation Mix Actin PCRs: To obtain actin amplicons, nested PCRs consisting of two sequential PCR steps were performed. First with an outer and second with an inner primer pair. For the first step, 5 µl of genomic DNA (15 to 20 ng of final concentration) were added to 20 µl volumes of 2 µl 10 x DreamTAQ PCR buffer including 25 mM of MgCl2, 1 µl 2.5 mM dNTP, 0.5 µl 5 mM forward primer Act_F, 0.5 µl 5 mM reverse primer Act_R and 1 unit DreamTAQ polymerase (Fermentas). The PCR protocol started with an initial denaturation for 3 min at 94°C, followed by 45 cycles of denaturation for 45 s at 92°C, annealing for 50 s at 50°C and elongation for 1:30 min at 72°C, and the protocol ended with a final extension step for 5 min at 72°C. For the second step, the products resulting from the first step were diluted 1:20 and 3 µl of the dilution were added to 20 µl volumes of 2 µl 10 x DreamTAQ PCR buffer including 25 mM of MgCl2, 1 µl 2.5 mM dNTP, 0.5 µl 5 mM forward primer ActN_F, 0.5 µl 5 mM reverse primer ActN_R and 1 unit DreamTAQ polymerase (Fermentas). The PCR protocol was the same as in the first step. Protocol and Mix As in Bulman & Marshall (1998) for ITS PCRs Protocol 95°C for 2:30 min initial denaturation; 35 cycles of 95°C for 40 s, 55°C for 30 s, Microsatellite 72°C for 30 s; and 72°C for 7 min final elongation PCRs Mix Microsatellite 20 µl volumes containing 5 µl of genomic DNA (15 to 20 ng final concentration), 2 PCRs µl 10 x DreamTAQ PCR buffer (Fermentas) including 25 mM MgCl2, 0.2 mM dNTP, 20 nM labeled forward primer, 90 nM unlabeled forward primer, 100 nM unlabeled reverse primer and 1 unit DreamTAQ polymerase (Fermentas)

104

Supplementary Table 3: Allele frequencies and absolute numbers for all 693 samples, genotyped with six loci (Supplementary Table 2), estimated by GENODIVE. The total number of alleles was 35.

Msat6_GAC/GAC Population Missing 203 207 210 225 231 237 Europe 0 0 1 0 0 0 0 Africa 0 0 1 0 0 0 0 Asia 0 0 0.98 0.01 0 0.01 0 Australasia 0 0 1 0 0 0 0 North America 0 0 1 0 0 0 0 South American 0 0.026 0.039 0.039 0.013 0.882 0 root galls South American 0 0 0.079 0.888 0 0.028 0.006 tuber lesions Overall 0 0.001 0.826 0.118 0.001 0.053 0.001

Msat45_TCA Population Missing 265 268 271 274 Europe 0 0.028 0.972 0 0 Africa 0 0 1 0 0 Asia 0 0 0.974 0.005 0.02 Australasia 0 0 1 0 0 North America 0 0 1 0 0 South American root galls 0 0 0.184 0.263 0.553 South American tuber lesions 0 0 0.006 0.006 0.989 Overall 0 0.009 0.815 0.016 0.16

Msat84_GCT Population Missing 258 261 270 274 279 282 288 290 Europe 0 0 0 0 0.002 0 0 0 0 Africa 0 0.053 0.018 0.009 0 0.009 0 0.018 0.009 Asia 0 0 0 0 0 0 0 0 0 Australasia 0 0 0 0 0 0 0 0 0 North America 0 0 0 0 0 0 0 0 0 South American root 0 0 0 0 0 0 0 0 0 galls South American tuber 0 0 0 0 0 0 0.006 0 0 lesions Overall 0 0.004 0.001 0.001 0.001 0.001 0.001 0.001 0.001

105

Msat84_GCT cont.

Population 291 294 297 303 Europe 0.991 0.007 0 0 Africa 0.868 0 0.018 0 Asia 1 0 0 0 Australasia 1 0 0 0 North 1 0 0 0 America South 0.066 0.316 0 0.618 American root galls South 0 0.938 0.011 0.045 American tuber lesions Overall 0.807 0.14 0.003 0.04

Msat103_CT Population Missing 179 187 191 193 195 197 199

Europe 0 0 0 0 0.316 0 0.679 0.005 Africa 0 0.018 0 0 0.605 0 0.377 0 Asia 0 0 0.005 0 0.648 0.01 0.327 0.01 Australasia 0 0 0 0 0.732 0.003 0.262 0.003 North 0 0 0 0 0.019 0 0.962 0.019 America South 0 0 0 0 0.026 0 0.039 0.934 American root galls South 0 0 0 0.045 0.556 0 0.287 0.112 American tuber lesions Overall 0 0.001 0.001 0.006 0.493 0.002 0.427 0.07

Msat246.1_CAA Population Missing 140 142 Europe 0 0.26 0.74 Africa 0 0.202 0.798 Asia 0 0.321 0.679 Australasia 0 0.259 0.741 North America 0 0 1 South American root galls 0 0.355 0.645 South American tuber 0 0.303 0.697 lesions Overall 0 0.265 0.735

106

Msat246.2_CAA Population Missing 156 158 160 168 Europe 0 0.342 0.656 0.002 0 Africa 0 0.219 0.781 0 0 Asia 0 0.434 0.566 0 0 Australasia 0 0.4 0.6 0 0 North America 0 0 1 0 0 South American root galls 0 0.737 0.263 0 0 South American tuber 0 0.511 0.478 0 0.011 lesions Overall 0 0.39 0.608 0.001 0.001

Msat6_GAC/GAC Population Total Missing 203 207 210 225 231 237 Europe 430 0 0 430 0 0 0 0 Africa 114 0 0 114 0 0 0 0 Asia 196 0 0 192 2 0 2 0 Australasia 340 0 0 340 0 0 0 0 North America 52 0 0 52 0 0 0 0 South American 76 0 2 3 3 1 67 0 root galls South American 178 0 0 14 158 0 5 1 tuber lesions Overall 1386 0 2 1145 163 1 74 1

Msat45_TCA Population Total Missing 265 268 271 274 Europe 430 0 12 418 0 0 Africa 114 0 0 114 0 0 Asia 196 0 0 191 1 4 Australasia 340 0 0 340 0 0 North America 52 0 0 52 0 0 South American root galls 76 0 0 14 20 42 South American tuber 178 0 0 1 1 176 lesions Overall 1386 0 12 1130 22 222

107

Msat84_GCT Population Total Missing 258 261 270 274 279 282 Europe 430 0 0 0 0 1 0 0 Africa 114 0 6 2 1 0 1 0 Asia 196 0 0 0 0 0 0 0 Australasia 340 0 0 0 0 0 0 0 North America 52 0 0 0 0 0 0 0 South American root galls 76 0 0 0 0 0 0 0 South American tuber 178 0 0 0 0 0 0 1 lesions Overall 1386 0 6 2 1 1 1 1

Msat84_GCT cont. Population 288 290 291 294 297 303

Europe 0 0 426 3 0 0 Africa 2 1 99 0 2 0 Asia 0 0 196 0 0 0

Australasia 0 0 340 0 0 0 North America 0 0 52 0 0 0 South American root galls 0 0 5 24 0 47

South American tuber 0 0 0 167 2 8 lesions Overall 2 1 1118 194 4 55

Msat103_CT Population Total Missing 179 187 191 193 195 197 199 Europe 430 0 0 0 0 136 0 292 2 Africa 114 0 2 0 0 69 0 43 0 Asia 196 0 0 1 0 127 2 64 2 Australasia 340 0 0 0 0 249 1 89 1 North America 52 0 0 0 0 1 0 50 1 South American 76 0 0 0 0 2 0 3 71 root galls South American 178 0 0 0 8 99 0 51 20 tuber lesions Overall 1386 0 2 1 8 683 3 592 97

108

Msat246.1_CAA Population Total Missing 140 142 Europe 430 0 112 318 Africa 114 0 23 91 Asia 196 0 63 133 Australasia 340 0 88 252 North America 52 0 0 52 South American root galls 76 0 27 49 South American tuber 178 0 54 124 lesions Overall 1386 0 367 1019

Msat246.2_CAA Population Total Missing 156 158 160 168 Europe 430 0 147 282 1 0 Africa 114 0 25 89 0 0 Asia 196 0 85 111 0 0 Australasia 340 0 136 204 0 0 North America 52 0 0 52 0 0 South American root galls 76 0 56 20 0 0 South American tuber 178 0 91 85 0 2 lesions Overall 1386 0 540 843 1 2

109

Supplementary Table 4: Distribution of ITS and actin haplotypes across global regions. Distribution of Spongospora subterranea f.sp. subterranea ITS and actin haplotypes. The haplotypes occurring in each global region are listed for both sequences. This is visualized in Supplementary Fig. 1.

Region ITS Haplotypes Actin Haplotypes Concatenated (iHap) (aHap) Haplotypes (cHap) Europe 6 1, 3 6, 9 Africa 6 3 6 Asia 6 1, 3 6, 9 Australasia 6 3 6 North America 6 3 6 Root galls South America 3 1, 2, 3, 4, 5, 6, 7, 8 3, 7, 8, 10, 11, 12, 13, 14 Tuber lesions South America 1, 2, 4, 5, 7 1, 3, 9, 10, 11, 12 1, 2, 4, 5, 15, 16, 17, 18, 19 Total South America 1, 2, 3, 4, 5, 7 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19

Supplementary Table 5a: MIGRATE estimates of past migration rates M for Spongospora subterranea f.sp. subterranea between Europe and introduced regions. 0.05 and 0.95 percentiles indicated in parentheses. Effective population sizes theta are on the diagonal.

Recipient of migrants Source of migrants EU SA lesions SA galls Europe - 0.00 0.34 (0.00 – 0.00) (0.29 – 0.61) South America tuber 7.23 - 9.68 lesions (5.02 – 10.01) (7.62 – 12.09) South America root galls 3.09 9.27 - (1.67 – 5.09) (7.39 – 11.37)

Supplementary Table 5b: MIGRATE estimates of past migration rates M for Spongospora subterranea f.sp. subterranea between native and pooled introduced regions. 0.05 and 0.95 percentiles indicated in parentheses.

Recipient of migrants Source of migrants Introduced regions SA lesions SA galls Introduced regions - 0.00 0.13 pooled (0.00 – 0.18) (0.11 – 0.40) South America tuber 2.14 - 2.22 lesions (1.37 – 3.12) (1.37 – 3.34) South America root galls 5.07 7.52 - (3.90 – 6.51) (5.94 – 9.38)

110

Supplementary Table 6: BayesAss estimates of recent migration rates. Estimates of recent migration rates (% of immigrant origin) of Spongospora subterranea f.sp. subterranea between native and introduced regions. 95 % confidence intervals indicated in parentheses.

Recipient of migrants Introduced regions Native regions Source of EU AF AS AU NA SA lesions SA galls migrants Europe 99 26 25 29 17 0 1 (98 – (21 – 32) (20 - 30) (26 - 33) (12 - 24) (0 – 1) (0 – 3) 100) Africa 0 69 1 0 2 0 0 (0 – 1) (67 - 75) (0 - 3) (0 - 2) (0 – 4) (0 – 1) (0 – 1) Asia 0 1 71 1 2 0 0 (0 – 1) (0 - 3) (67 - 77) (0 - 2) (0 – 4) (0 – 1) (0 – 1) Australasia 0 1 1 69 2 0 0 (0 – 1) (0 - 3) (0 - 3) (67 – 70) (0 – 4) (0 – 1) (0 – 1) North 0 1 1 0 75 0 0 America (0 – 1) (0 - 3) (0 - 2) (0 - 2) (67 – 84) (0 – 1) (0 – 1) South 0 0 1 1 1 99 1 America (0 – 1) (0 - 2) (0 - 3) (0 - 3) (0 - 3) (98 – 100) (0 – 3) tuber lesions South 0 0 1 1 1 0 97 America (0 – 1) (0 - 2) (0 - 3) (0 - 3) (0 - 3) (0 – 1) (94 – 99) root galls

111

2.8 REFERENCES

Agrios GN, 2005. Plant Pathology. Department of Plant Pathology, University of Florida, USA: Elsevier Academic Press.

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ, 1990. Basic local alignment search tool. Journal of Molecular Biology 215, 403-10.

Anderson PK, Cunningham AA, Patel NG, Morales FJ, Epstein PR, Daszak P, 2004. Emerging infectious diseases of plants: pathogen pollution, climate change and agrotechnology drivers. Trends in Ecology & Evolution 19, 535-44.

Beerli P, Palczewski M, 2010. Unified framework to evaluate panmixia and migration direction among multiple sampling locations. Genetics 185, 313-463.

Bhagwat SA, Willis KJ, Birks HJB, Whittaker RJ, 2008. Agroforestry: a refuge for tropical biodiversity? Trends in Ecology & Evolution 23, 261-7.

Blandón-Díaz JU, Widmark A-K, Hannukkala A, Andersson B, Högberg N, Yuen JE, 2012. Phenotypic variation within a clonal lineage of Phytophthora infestans infecting both tomato and potato in Nicaragua. Phytopathology 102, 323-30.

Braselton JP, 1995. Current status of the Plasmodiophorids. Critical Reviews in Microbiology 21, 263-75.

Bulman SR, Marshall JW, 1998. Detection of Spongospora subterranea in potato tuber lesions using the polymerase chain reaction (PCR). Plant Pathology 47, 759-66.

Calinski T, Harabasz J, 1974. A dendrite method for cluster analysis. Communications in Statistics 3, 1-27.

Chen XM, 2005. Epidemiology and control of stripe rust Puccinia striiformis f. sp tritici on wheat. Canadian Journal of Plant Pathology 27, 314-37.

113

CIP, 2008. The Odyssey of the Potato. Der lange Weg der Kartoffel zur weltweiten Kulturpflanze. Brochure of an exhibition, International Potato Center (CIP), Lima, Peru, ISBN 978-92-9060-336-8

Clement M, Posada D, Crandall KA, 2000. TCS: a computer program to estimate gene genealogies. Molecular Ecology 9, 1657-9.

Desprez-Loustau M-L, Robin C, Buée M, et al., 2007. The fungal dimension of biological invasions. Trends in Ecology & Evolution 22, 472-80.

Diriwaechter G, Parbery DG, 1991. Infection of potato by Spongospora subterranea. Mycological Research 95, 762-764.

Down GJ, Grenville LJ, Clarkson JM, 2002. Phylogenetic analysis of Spongospora and implications for the taxonomic status of the Plasmodiophorids. Mycological Research 106, 1060-65.

Dybdahl MF, Storfer A, 2003. Parasite local adaptation: Red Queen versus Suicide King. Trends in Ecology & Evolution 18, 523-30.

Fry WE, Goodwin SB, Dyer AT, et al., 1993. Historical and recent migrations of Phytophthora infestans - chronology, pathways, and implications. Plant Disease 77, 653-61.

Fry WE, Goodwin SB, 1997. Re-emergence of potato and tomato late blight in the United States. Plant Disease 81, 1349 -57.

Genet RA, Braam WF, Gallagher DTP, Anderson JaD, Lewthwaite SL, 1995. 'Gladiator': a new potato cultivar with high resistance to potato cyst nematode and powdery scab suitable for french fries and fresh market. New-Zealand Journal of Crop and Horticultural Science 23, 105-7.

Gilchrist E, Soler J, Merz U, Reynaldi S, 2011. Powdery scab effect on the potato Solanum tuberosum ssp. andigena growth and yield. Tropical Plant Pathology 36, 350-5.

114

Glenn TC, Schable NA, 2005. Isolating microsatellite DNA loci. Molecular Evolution: Producing the Biochemical Data, Part B 395, 202-22.

Gonzalez-Varela G, Gonzalez AJ, Milgroom MG, 2011. Clonal population structure and introductions of the chestnut blight fungus, Cryphonectria parasitica, in Asturias, northern Spain. European Journal of Plant Pathology 131, 67-79.

Haubold B, Hudson RR, 2000. LIAN 3.0: detecting linkage disequilibrium in multilocus data. Bioinformatics 16, 847-48.

Higgins DG, Thompson JD, Gibson TJ, 1996. Using CLUSTAL for multiple sequence alignments. Computer Methods for Macromolecular Sequence Analysis 266, 383-402.

Lashermes P, Combes MC, Trouslot P, Charrier A, 1997. Phylogenetic relationships of coffee-tree species (Coffea L) as inferred from ITS sequences of nuclear ribosomal DNA. Theoretical and Applied Genetics 94, 947-55.

Lawson Handley L-J, Estoup A, Evans DM, et al., 2011. Ecological genetics of invasive alien species. BioControl 56, 409-28.

Lombaert E, Guillemaud T, Cornuet J-M, et al., 2010. Bridgehead effect in the worldwide invasion of the biocontrol Harlequin ladybird. Plos One 5.

Lyman GR, Rogers JT, 1915. The native habitat of Spongospora subterranea. Science 42, 940-1.

MacQueen JB, 1967. Some methods for classification and analysis of multivariate observations. In: Le Cam LM, Neyman J, eds. Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, 1965-66. Berkeley, Calif.:University of California Press, 281–97.

McDonald BA, 1997. The population genetics of fungi: Tools and techniques. Phytopathology 87, 448-53.

McDonald BA, Linde C, 2002. Pathogen population genetics, evolutionary potential, and durable resistance. Annual Review of Phytopathology 40, 349-79.

115

Meirmans PG, 2012. AMOVA-based clustering of population genetic data. Journal of Heredity 103, 744-50.

Meirmans PG, Van Tienderen PH, 2004. GENOTYPE and GENODIVE: two programs for the analysis of genetic diversity of asexual organisms. Molecular Ecology Notes 4, 792-4.

Merz U, 2008. Powdery scab of potato - Occurrence, life cycle and epidemiology. American Journal of Potato Research 85, 241-6.

Merz U, Falloon RE, 2009. Review: Powdery Scab of Potato - Increased Knowledge of Pathogen Biology and Disease Epidemiology for Effective Disease Management. Potato Research 52, 17-37.

Merz U, Lees AK, Sullivan L, et al., 2012. Powdery scab resistance in Solanum tuberosum: an assessment of cultivar x environment effect. Plant Pathology 61, 29-36.

Montarry J, Andrivon D, Glais I, et al., 2010. Microsatellite markers reveal two admixed genetic groups and an ongoing displacement within the French population of the invasive plant pathogen Phytophthora infestans. Molecular Ecology 19, 1965-77.

Nei M, 1987. Molecular Evolutionary Genetics. New York, USA: Colombia University Press.

Petit RJ, El Mousadik A, Pons O, 1998. Identifying populations for conservation on the basis of genetic markers. Conservation Biology 12, 844-55.

Qu XS, Christ BJ, 2004. Genetic variation and phylogeny of Spongospora subterranea f.sp subterranea based on ribosomal DNA sequence analysis. American Journal of Potato Research 81, 385-94.

Qu XS, Christ BJ, 2006. Single cystosorus isolate production and restriction fragment length polymorphism characterization of the obligate biotroph Spongospora subterranea f. sp. subterranea. Phytopathology 96, 1157-63.

116

Rabobank, 2009. World Potato Map. Utrecht, The Netherlands: Food & Agrobusiness Research and Advisory Department, Rabobank.

Rannala B, Yang Z, 2003. Bayes estimation of species divergence times and ancestral population sizes using DNA sequences from multiple loci. Genetics 164, 1645-56.

Rozen S, Skaletsky HJ, 2000. Primer3 on the WWW for general users and for biologist programmers. In: Krawetz S, Misener S, eds. Bioinformatics Methods and Protocols: Methods in Molecular Biology.: New York, USA:Humana Press, 365-86.

Schrader G, Unger J-G, 2003. Plant quarantine as a measure against invasive alien species: The framework of the International Plant Protection Convention and the plant health regulations in the European Union. Biological Invasions 5, 357-64.

Schwaerzel R, 2002. Sensibilités des racines et tubercules des variétés de pommes de terre à la gale poudreuse et quelques résultats de lutte chimique. Revue Suisse Agricole 34, 261-6.

Shah FA, Falloon RE, Bulman SR, 2010. Nightshade weeds (Solanum spp.) confirmed as hosts of the potato pathogens Meloidogyne fallax and Spongospora subterranea f. sp. subterranea. Australasian Plant Pathology 39, 492-8.

Smith JJ, Offord LC, Holderness M, Saddler GS, 1995. Genetic diversity of Burkholderia solanacearum (synonym Pseudomonas solanacearum) race 3 in Kenya. Applied and Environmental Microbiology 61, 4263-8.

Smith JJ, Kibata GN, Murimi ZK, et al., 1998. Biogeographic studies on Ralstonia solanacearum race 1 and 3 by genomic fingerprinting. In: Prior PH, Allen C,

117

Elphinstone J, eds. Bacterial Wilt Disease: Molecular and Ecological Aspects. Berlin, BRD: Springer, 50-5.

Stukenbrock EH, McDonald BA, 2008. The origins of plant pathogens in agroecosystems. Annual Review of Phytopathology 46, 75-100.

Tamura K, Peterson D, Peterson N, et al., 2011. MEGA5: Molecular evolutionary genetics analysis using maximum likelihood evolutionary distance, and maximum parsimony methods. Molecular Biology 28, 2731–39.

Valtuena FJ, Preston CD, Kadereit JW, 2011. Evolutionary significance of the invasion of introduced populations into the native range of Meconopsis cambrica. Molecular Ecology 20, 4318-31.

Vizzini A, Zotti M, Mello A, 2009. Alien fungal species distribution: the study case of Favolaschia calocera. Biological Invasions 11, 417-29.

Wallroth RW, 1842. Der Knollenbrand der Kartoffel. LINNEA. Ein Journal für die Botanik in ihrem ganzen Umfange. 16, 332.

Weir BS, Cockerham CC, 1984. Estimating f-statistics for the analysis of population- structure. Evolution 38, 1358-70.

Winnepenninckx B, Backeljau T, Dewachter R, 1993. Extraction of high-molecular- weight DNA from mollusks. Trends in Genetics 9, 407.

118

CHAPTER 3: CROSS-INFECTION POTENTIAL OF SPONGOSPORA SUBTERRANEA F. SP. SUBTERRANEA ROOT GALL AND TUBER LESION INOCULUM FROM DIFFERENT POTATO (SOLANUM SPP.) HOSTS

Rebecca D. Gau1, Ueli Merz1*, Richard E. Falloon2,3

1 Plant Pathology, Institute of Integrative Biology, ETH Zurich, CH-8092 Zurich, Switzerland; 2 New Zealand Institute for Plant and Food Research Limited, Lincoln, New Zealand; 3 Bio-Protection Research Centre, Lincoln University, Lincoln, New Zealand; * Author for correspondence (Phone: +41(0) 44 632 3948; Fax: +41(0) 44 632 1572; E-mail: [email protected])

To be submitted to the European Journal of Plant Pathology in 2013.

3.1 ABSTRACT

Spongospora subterranea f. sp. subterranea (Sss) causes the potato diseases powdery scab on tubers and root galls. The pathogen occurs in most potato production areas worldwide and was probably introduced to Europe from South America in the 16th century. Three different Sss genotype clusters have been found worldwide: the genetically variable root gall or tuber lesion groups from South America (native), and, in contrast, the nearly clonal group (lesions and galls) outside South America (invasive). A cross-inoculation experiment was carried out with three potato hosts (Solanum spp.) and five different sporosorus inocula from root galls or tuber lesions, to determine the virulence of native and invasive Spongospora inoculum on potato hosts from both areas. Scoring of root galls showed that all combinations led to root infection. The S. tuberosum ssp. tuberosum host inoculated with tuber lesion sporosori from the same host had the greatest amount of infection, whereas the S. tuberosum ssp. andigena host had the least infection, regardless of inoculum source. The inoculum genotypes corresponded to the three described groups. Most genotypes from harvested root galls were of the invasive group, independent of the inoculum, whereas tuber lesion samples from S. tuberosum ssp. tuberosum either matched the native lesion type (with native inoculum) or the invasive type (with invasive lesion inoculum). The native root gall type was not found. These results suggest that pathogen genotype selection is host organ-specific, and that the most successful invasive genotype is still present in native pathogen populations. The results emphasise the need for continued quarantine vigilance to prevent invasions by new virulent recombinants of the pathogen.

121

3.2 INTRODUCTION

Spongospora subterranea f. sp. subterranea (Sss: Cercozoa, Plasmodiophoridae), is a soilborne, obligate biotrophic pathogen, which causes two diseases throughout the major potato (Solanum tuberosum) growing regions of the world: Powdery scab, an important blemish disease of potato tubers, and root galls on potato roots. Both diseases can cause severe losses in potato production and processing (Merz & Falloon, 2009; Shah et al., 2012).

Resting spores of the pathogen are accumulated in sporosori (spore balls), which are formed below ground in tuber lesions and roots galls (Fig. 13a-c). From resting spores, biflagellate primary zoospores (Fig. 13d) emerge and can swim short distances in moist soil to reach and infect the host tissue (Merz & Falloon, 2009).

Figure 13: a) Potato tuber with powder scab lesions. b) Pale, immature root galls. c) Scanning electron micrograph of a Spongospora subterranea f. sp. subterranea sporosorus, consisting of numerous single resting spores, some with openings for zoospore release (bar = 10 µm). d) Scanning electron micrograph of a primary zoospore of the pathogen, with two flagella (bar = 5 µm). All pictures taken by Ueli Merz.

Plasmodia of Sss are formed in infected host cells, which may proliferate to zoosporangia from which secondary zoospores emerge to cause further infections.

123

Completing the life cycle, plasmodia can also develop into resting spore structures, which contain the primary zoospores. Although it has been suggested, that resting spores are the product of sexual reproduction (Braselton, 1995), the role of sexual and asexual reproduction in the life cycle of Sss still remains to be elucidated. The resting spores are highly resistant to environmental stresses and pesticides. Thus, once a field is contaminated it is likely to remain infectious for decades (van de Graaf et al., 2007).

The life-cycle of Sss prevents natural long-distance dispersal. An invasion scenario for Sss was suggested by Gau et al. (2012): The pathogen most likely first reached Europe carried by humans, when the conquistadors brought the potato from South America which is likely to be the native region. After a few more introduction events, Sss established in Europe and was cut off from South America. Further spread occurred only from Europe to all other regions in the world via infected seed tubers. The pathogen must have adapted successfully to new climatic conditions despite low genetic variability and strong bottlenecks.

New introductions of the pathogen to “virgin” soils occur by planting contaminated seed potatoes, because Sss is a very successful invader. This is reflected by the frequent publications of “first reports” of the diseases caused by Sss over the last 30 years (Merz, 2008). Although introductions of new Sss strains from South America might be possible, no substantial exchange between South America and the rest of the world occurs in modern seed potato trade (Rabobank, 2009). This resulted in a geographical separation of Sss populations in native and introduced regions, represented by their differences in genetic variability. Under these circumstances the risk of increasing genetic variability of Sss in introduced regions is likely to be low.

There are no resistant potato cultivars or effective pesticides against Sss, thus, the only way to control the pathogen currently is prevention in the form of planting certified seed. Further, enforcement of quarantine measures is necessary to avoid new introductions, especially from South America. Until now, there is no global quarantine concept and individual countries have definitions of the tolerance limit

124

for the amount of disease caused by Sss on seed tubers. To decide which quarantine measures are efficient and applicable, the host-pathogen interaction under different environments must be understood.

Among the edible potato varieties, there are three (sub-) species, which are of economic importance and which are hosts of Sss: In some Andean regions of South America, mainly short day potato plants (S. tuberosum ssp. andigena (An) and S. phureja (Ph)) are cultivated, whereas in other regions the long day plant S. tuberosum ssp. tuberosum (Tu) is also cultivated, which is the most grown potato subspecies worldwide. None of the commercially available varieties are completely resistant to Sss, but striking differences in susceptibility to tuber and root infection of Tu cultivars have been observed by Schwaerzel (2002) who assessed both diseases. A recently published population genetics study by Gau et al. (2012) indicated genetic substructuring in South American Sss populations according to the host tissue from which these samples were obtained. Samples from root galls, mainly collected on Ph, formed one group (A) and samples from tuber lesions, mainly on An and Tu, formed the other group (B). Samples from the rest of the world belonged to a third group with a highly clonal genetic structure (C) and no host tissue differentiation.

Climatic conditions may interfere with the infection process and development of host and pathogen, which challenges results drawn from field trials when comparing the performance of potato host species with different photoperiodic needs. Torres et al. (1995) and Rodriguez et al. (2009) performed trials in South America comparing short-day and long-day potato plants. Both of these research groups reported greater susceptibility to powdery scab of some Tu cultivars, compared to short day hosts, than previously reported. These field trials, however, were conducted at altitudes above 3000 m under short-day conditions where the performance of long-day-adapted Tu hosts is likely to be different from their performance under their preferred climate conditions. Alternatively, the presence of local pathotypes of Sss was suggested, which might differ in virulence (Torres et al. 1995, Rodriguez et al. 2009). Trials are usually conducted in naturally contaminated soil or soil or sand artificially inoculated with local Sss sporosorus material.

125

Studies of how different Sss inocula interact with the three potato host (sub-) species under controlled climate conditions have not been previously reported. Such data would help to define risk scenarios for the improvement of quarantine measures to prevent further spread and new introductions of the pathogen.

The present study aimed to elucidate the ability of different Sss sporosorus inocula of to cause disease symptoms on the main potato hosts, and to compare the genetics of the inocula with their infection products on host roots and tubers. A micro-bioassay (Merz et al., 2004) was used in a cross-inoculation experiment. Root gall and tuber lesion inoculum, both from South America and elsewhere in the world and obtained from different potato hosts, was applied to three different potato hosts. Inoculum and sporosori from infected plants were genetically analysed.

126

3.3 MATERIAL AND METHODS

3.3.1 CULTIVATION AND PREPARATION OF TISSUE CULTURED POTATO PLANTLETS

Stem cuttings of three potato hosts (Table 6) were cultivated as described in Merz et al. (2004) and sub-cultured every 4 weeks. The roots of 4-week-old plantlets were washed free of agar medium with tap water, trimmed to a length of 60 mm, and plantlets (two per dish) were transferred to black petri dishes with lids (see below).

Table 6: Spongospora subterranea sources and Solanum host treatment combinations applied in an inoculation experiment.

Inoculum source Inoculum ID Host Cultivar Colombia CPG S. tuberosum ssp. tuberosum ‘Agria’ S. phureja S. phureja ‘Shaucha galls Amarilla’ Switzerland STG S. phureja ‘Shaucha S. tuberosum ssp. Amarilla’ tuberosum galls Switzerland STL S. tuberosum ssp. tuberosum ‘Agria’ S. tuberosum ssp. S. tuberosum ssp. andigena ‘Pardo pastusa’ tuberosum tuber lesions Venezuela VTL S. tuberosum ssp. tuberosum ‘Agria’ S. tuberosum ssp. S. tuberosum ssp. andigena ‘Pardo pastusa’ tuberosum tuber lesions Venezuela VAL S. tuberosum ssp. tuberosum ‘Agria’ S. tuberosum ssp. andigena S. tuberosum ssp. andigena ‘Pardo pastusa’ tuber lesions

3.3.2 NUTRIENT SOLUTION (NS)

A nutrient solution was prepared as a concentrated stock solution. One liter contained 7.22 g calcium nitrate tetra hydrate (Ca(NO3)2), 2.53 g of potassium nitrate

(KNO3), 2.46 g of magnesium sulfate hepta hydrate (MgSO4 × 7 H2O), 400 mg of ammonium nitrate (NH4NO3), 230 mg of potassium hydrogen phosphate (KH2PO4), 37 mg of potassium chloride (KCl), 200 mg of Sequestren (Fe) and 12.5 ml of Hoagland’s B micronutrient solution (Hoagland & Snyder, 1933). One liter working solution contained distilled H2O with100 ml of the stock solution.

127

3.3.3 PREPARATION OF INOCULUM

Sporosorus inoculum (Table 6) was prepared from infected tubers (scrapings) and root galls (ground with a mortar). The material was soaked in 5 ml NS for 30 min, and mixed for 2 min with a Polytron mixer (PT 2100, Kinematik AG) at 11,000 revolutions min−1. The sporosorus concentration was estimated using a haemocytometer and the inoculum suspension adjusted to 500 sporosori ml-1 in a total volume of 360 ml.

3.3.4 CROSS-INOCULATION EXPERIMENT

The treatments applied in this experiment are outlined in Table 6. For each treatment, six Petri dishes (diameter 75 mm, depth 30 mm, painted black outside) were prepared. These each contained 60 ml of the sporosorus suspension and two plantlets, held by two opposite slits in the blackened lid. For experimental controls, six Petri dishes with 60 ml NS medium (without inoculum), containing two plantlets per potato host, were prepared. The setup of the trial is outlined in Table 6. The plants were cultivated in a controlled environment chamber set at a daily cycle of 18°C, 16 h light (5 kL) and 15 °C, 8 h dark, and 70 % relative humidity. Fresh NS was regularly added (twice per week) to the Petri dishes. The severity of root galling on test plants was assessed after 63 days using a standard scale (http://www.spongospora.ethz.ch/LaFretaz/scoringtablegalls.htm), where: 0 = no galls; 1 = 1–2 galls; 2 = 3–10 galls; 3 = >10 galls, but most in clusters; 4 = many galls, regularly distributed over the roots. A root gall index was calculated, multiplying incidence (% plants with galls) by mean severity score.

3.3.5 MOLECULAR METHODS

At least one root gall sample per treatment was taken for molecular analysis. As the cv. ‘Agria’ host produced tubers, which showed scab lesions 94 days after inoculation, lesion samples were also collected from these tubers. The samples were freeze dried and DNA was extracted using the DNeasy Mini Plant Kit (QIAGEN) following the steps outlined by the manufacturer.

Presence of Sss DNA was confirmed using the Sss-ITS-specific Spo8 and Spo9 primer pair (Bulman & Marshall, 1998). Additionally, the partial actin gene of each

128

sample was amplified, using previously described nested PCR (Gau et al. 2012). Additionally, the amplicons of both PCRs were sequenced with an ABI 3730 xl sequencer (Applied Biosystems). Based on the sequences, haplotype networks were established separately for the ITS and actin loci using TCS (Clement et al. 2000). Multilocus genotypes were determined to establish microsatellite profiles for all samples, as previously described (Gau et al. 2012).

Cloning of ITS fragments of the five inocula was performed with the TA cloning kit (Invitrogen). Inserts of the transformants were sequenced using a 3730xl sequencer (Applied Biosystems) and haplotype networks were created using TCS (Clement et al., 2000). Actin fragments of the inocula were not cloned and sequenced, because they are generated in a nested PCR (Gau et al. 2012), which is a technique vulnerable to contamination. Furthermore, the two steps required in the nested PCR protocol could bias the results.

129

3.4 RESULTS

3.4.1 CROSS INOCULATION EXPERIMENT

All inoculum-potato host combinations showed root infection (Fig. 14), and each produced at least one root gall. With sporosorus inoculum from tubers, the potato host Tu (cv. ‘Agria’) had consistently greater amounts of root galling compared to An (cv. ‘Pardo Pastusa’), even with inoculum from An tubers. A different picture was found when Ph (cv. ‘Shaucha Amarilla’) root infection levels were compared to those for cv. ‘Agria’ with sporosorus inoculum from galls (Fig. 14). The incidence of infected Ph plants and the overall severity of root infection were greater. The greatest root galling severity score resulted from the combination of Tu lesion inoculum with the host Tu, whereas the smallest score resulted from the An lesion inoculum in combination with An hosts. The non-inoculated control plants were free of infection.

200 Solanum tuberosum ssp tuberosum Solanum tuberosum ssp andigena Solanum phureja

180

160

140

120

100

Root gall Root index 80

0 STL VTL VAL CPG STL VTL VAL STG CPG Inoculum source

Figure 14: Root gall indices for different Solanum hosts inoculated with Spongospora subterranea f. sp. subterranea sporosori from five sources. Uninoculated experimental controls were free of root galls (data not shown). Inoculum source: Country: S = Switzerland, V = Venezuela, C = Colombia Host: T = S. tuberosum ssp. tuberosum, A = S. tuberosum ssp. andigena, P = S. phureja Sporosori from: L = tuber powdery scab lesions, G = root galls

131

Gall and tuber lesion formation was enhanced when the infection sites were exposed to air. Especially when NS levels in the dishes were low just before adding fresh solution, galls became more obvious on the roots and stolons located above the surface of the NS (Fig. 15a). When the cv. ‘Agria’ plants produced small tubers with white lesions (‘cauliflower’ type), the addition of NS was halted to avoid immersing the infected tubers. These lesions changed into wart-like structures (Fig 15b).

Figure 15: Aerial root galls (a) and lesions on small tubers (b) of cv. ‘Agria’ potato plants inoculated with Spongospora subterranea f. sp. subterranea sporosorus inoculum in a micro-bioassay.

3.4.2 MOLECULAR ANALYSES

Four haplotypes were found among nucleotide sequences for the actin locus and ITS region from the five inoculum sources, each of them corresponding to one of the three previously found genotype clusters for South American root gall samples

132

(group A), South American lesion samples (group B), and root gall and lesion samples from elsewhere (group C; Table 7). Analogous results were found for the multilocus genotypes in the microsatellite profile (Table 7). None of the harvested sporosori from root galls or tuber lesions clustered with group A. Root gall samples of most of the combinations clustered with group C, with two exceptions. Genotypes of the tuber lesion samples of the cv. ‘Agria’ clustered either with group B (South American gall or tuber lesion inoculum) or group C (tuber lesion inoculum from other areas).

Table 7: Genotype pattern of Spongospora subterranea from five inoculum sources, and of sporosori from three inoculated Solanum hosts (root galls and tuber lesions). Inoculum source Country: S = Switzerland, V = Venezuela, C = Colombia Host: T = S. tuberosum ssp. tuberosum, A = S. tuberosum ssp. andigena, P = S. phureja Sporosori from: L = tuber powdery scab lesions, G = root galls

Inocula Hosts Galls Tuber lesions

Haplotype Multilocus Haplotype Multilocus Haplotype Multilocus ITS Actin genotype ITS Actin genotype ITS Actin genotype CPG S. t. tuberosum S. phureja STG S. phureja STL S. t. tuberosum S. t. andigena VTL S. t. tuberosum S. t. andigena VAL S. t. tuberosum S. t. andigena

South American root gall genotype (group A)

South American tuber lesion genotype (group B)

Other region gall and tuber lesion genotype (group C)

Intermediate genotype (B+C)

Cloning of ITS fragments from inocula showed that in each inoculum several different genotypes were present. The CPG inoculum was the most diverse with 17 different ITS sequences. In contrast, only three different sequences were found for the STL inoculum. A haplotype network of all five inocula and the clones of all inocula was constructed (Fig. 16).

The inocula STL and STG and the corresponding clones were separated from the inocula VTL and VAL and the corresponding clones. The inoculum CPG and the

133

respective clones were distributed all across the network. For VAL and VTL, many sequences sharing one haplotype were found. For CPG, only two sequences shared one haplotype and for STL and STG all sequences were different.

Figure 16: Haplotype network of the ITS regions of the five Spongospora subterranea inoculum sources and their clones (indicated by numerals). Distances between haplotypes are expressed by the number of point mutations (pt). Inoculum source Country: S = Switzerland, V = Venezuela, C = Colombia Host: T = S. tuberosum ssp. tuberosum, A = S. tuberosum ssp. andigena, P = S. phureja Sporosori from: L = tuber powdery scab lesions, G = root galls

134

3.5 DISCUSSION

This paper describes the first reported cross inoculation experiment with Sss inoculum from both South America and the rest of the world, and from host plants with different photoperiod requirements. All inoculum-host-combinations caused disease on roots, but the non-South American combination gave the greatest root gall severity. The experiment was carried out under controlled long-day conditions. The bioassay used has the potential to allow root gall assessment as well as tuber production and infection, beside estimation of zoosporangium root infection (Merz et al., 2004), in a hydroponic system. This protocol is therefore a useful tool for detailed evaluation of specific host-pathogen interactions.

The long day potato host Tu showed greatest susceptibility to all inocula in the experiment, which was also observed in the field trials of Torres et al. (1995) and Rodriguez et al. (2009). The short day host An (cv. ‘Pardo Pastusa’) developed little root infection in our experiment independent of the source of inoculum.

This suggests that the host (sub-)species used in the experiment are different in the susceptibility to Sss. The cv. ‘Pardo Pastusa’ is rated as susceptible to powdery scab on tubers and roots (Ňústez López, 2011) in Colombia. No tubers were produced until root infection assessment (63 days after inoculation) by the South American hosts. Tuber initiation started when the conditions were changed to short days after the assessment. Most of the combinations with Ph and An hosts had small tubers at 115 days after inoculation. None of these were infected, however, although an increased rate of root infection was observed, especially with An (data not shown). In Colombia, Ph and An hosts mostly develop root galls when infected by Sss, and tuber lesions rarely develop on these hosts (Gilchrist et al., 2011). There has apparently been no international exchange of potato germplasm in Colombian potato production, and consequently all varieties are national cultivars. This is in contrast with Venezuela, where Tu (international cultivars) and An (from Colombia) are planted (Rodriguez et al., 2009), and root galls are rarely seen in that country (D. Rodriguez, personal communication).

135

The experiment was conducted in an environment to which the native hosts are not adapted (long-day conditions). Environmental conditions are likely to affect host physiology (Hannapel et al., 2004) and may alter resistance to pathogens (Mihovilovitch et al., 2010). In future studies, environmental conditions appropriate for the host (sub-) species should be tested.

As an obligate biotroph, Sss cannot be grown on artificial media. Therefore, it is difficult to directly observe the effects of environmental factors on this pathogen. The pathogen was studied using the recently developed genetic markers, and haplotypes and genotypes of the different inocula matched the three genotype groups described by Gau et al. (2012). These were: gall samples from the South American host Ph (group A); tuber lesion samples from the South American host An (group B); and gall and tuber lesions samples from the host Tu (group C). ‘Genotype shifts’ occurred between infection of the host plants and harvesting of galls and lesions at the end of the experiment. Only the hosts which were inoculated with group C sporosori developed root galls and tuber lesions of the same genotype group. Similarly, group B inoculum caused group B tuber lesions on the cv. ‘Agria’ host. However, in all other combinations the genotype changed.

These results suggest that all the inocula were mixtures of genotypes. In conventional PCR, just the predominant genotype is detected in a sample. Minor genotypes are detected only when the PCR products of inoculum samples are cloned, as in the present study. In this instance, host organ-specific selection of genotypes in the mixed inoculum probably occurred.

Our results suggest that the apparent clonal population structure of Sss in introduced regions may be more diverse than previously reported, but in comparison to the native region, genetic variability is much less. It is possible that sexual reproduction might be the cause of increased diversity. However, only predominant genotypes have established in introduced areas, favoured by selection.

Single genotype inocula would answer this question, but protocols to establish these have not been developed. A possible approach would be to produce

136

root galls from inoculation experiments which incorporate genetic characterization, including cloning of inocula and harvested material. The harvested galls could be maintained using the recently described Sss callus cultures (Bulman et al. 2011) to multiply the material for repeated reinfection experiments, with recurring characterization to verify that single genotypes are achieved and maintained.

The differences between the three genetic groups of Sss could be explained by their evolutionary background. Given that Colombian Sss populations are the most diverse (Gau et al. 2012) we hypothesize that they are the most ancient. Furthermore, group A was only found in Colombian root gall inoculum, which was the most diverse of the five inocula (Fig. 16), supporting the hypothesis. It is likely that non tuber-bearing solanaceous hosts pre-dated tuber bearing plants selected as food crops. Thus, group A could be the oldest Sss type. Group B inoculum was obtained from South American tuber lesions of An and Tu, and this group occurs in Colombia, Venezuela and Peru (Gau et al. 2012). With the domestication of tuber- bearing solanaceous plants group B genotypes could have descended from group A. The group C inoculum on Tu caused most disease in our experiment. We suggest, that group C is possibly more adapted to lowland conditions (long days). This group may have evolved when the edible potato was taken from its region of origin (Lake Titicaca region of Peru/Bolivia) to long-day areas by pre-Columbian sophisticated farming civilizations (FAO, 2009).

It was further suggested that Sss was introduced to Europe on infected potato tubers shipped by the conquistadors in the 16th century (Gau et al. 2012), most likely as a mixture of genotypes. Under long-day conditions, group C individuals became dominant and spread to the rest of the world. The group C genotype still occurs in South America (Gau et al. 2012) but is less competitive in the native region, where group A and B predominate. Increased knowledge of Sss evolution may help to predict the future development of the pathogen.

Our results imply that South American Sss strains are less competitive and unable to establish in introduced regions. If sexual reproduction does not occur, the risk of increasing virulence by importing variable inoculum from South America may

137

be less than previously supposed (Gau et al., 2012). Hybridization or sexual recombination between genetically different Sss strains could, however, lead to new virulent pathotypes.

Our results suggest that pathogen genotype selection is host organ-specific, and that the most successful invasive genotype is still present in native pathogen populations. For a better understanding of the predominance of the different genotypic groups in the native and invaded regions of Sss, a detailed study of South American populations is necessary. Sporosori, both from root galls and tuber lesions, should be collected from cultivated and wild solanaceous hosts. Traditional South American potato culture may introduce wild potatoes for hybridization with cultivated potatoes (FAO, 2009). This could introduce new Sss genotypes from wild potatoes to fields of cultivated potatoes. Therefore, there is a need for continued quarantine vigilance to prevent invasions of new virulent recombinants of the pathogen.

138

3.6 ACKNOWLEDGEMENTS

This research was supported by funding from Horticulture Australia Ltd (APRP1, University of Tasmania), the ETH Zürich, the United Kingdom Potato Council, the New Zealand Ministry of Science and Innovation, Horticulture New Zealand, and the Swiss Federal Office of Agriculture. IPK Gattersleben, Germany, and Agroscope, Changins, Switzerland freely supplied potato plantlets.

139

3.7 REFERENCES

Braselton JP, 1995: Current status of the Plasmodiophorids. Critical Reviews in Microbiology. 21: 263-275.

Bulman SR, Marshall JW, 1998: Detection of Spongospora subterranea in potato tuber lesions using the polymerase chain reaction (PCR). Plant Pathology 47, 759-766.

Bulman S, Candy JM, Fiers M, Lister R, Connera AJ, Eady CC, 2011. Genomics of Biotrophic, Plant-infecting Plasmodiophorids Using In Vitro Dual Cultures. Protist 162 (3), 449–461.

Clement, M, Posada, D, and Crandall, KA 2000. Tcs: A computer program to estimate gene genealogies. Molecular Ecology 9 (10), 1657-1659.

Gau R., Merz U., Falloon R. and Brunner P., 2012. Global genetics and invasion history of the powdery scab pathogen, Spongospora subterranea f. sp. subterranea. Plant Pathology (submitted)

Gilchrist E., J. Soler, U. Merz and S. Reynaldi, 2011. Powdery scab effect on the potato Solanum tuberosum ssp. andigena growth and yield. Tropical Plant Pathology, 36 (6), 350-355.

Hannapel D, Chen H, Rosin FM, Banerjee AK, Davies PJ (2004) Molecular controls of tuberization. American Journal of Potato Research 81, 263–274

Hoagland DR, Snyder WC, 1933. Nutrition of strawberry plants under controlled conditions: a) effects of deficiencies of boron and certain other elements b) susceptibility to injury from sodium salts. Proceedings of the American Society of Horticulture Sciences 30, 288-294.

141

Merz U, Martinez V, Schwaerzel R, 2004. The potential for the rapid screening of potato cultivars (Solanum tuberosum) for resistance to powdery scab (Spongospora subterranea) using a laboratory bioassay. European Journal of Plant Pathology 110, 71-77.

Merz U, 2008. Powdery scab of potato - Occurrence, life cycle and epidemiology. American Journal of Potato Research 85, 241-246.

Merz U, Falloon RE, 2009. Review: Powdery scab of potato - increased knowledge of pathogen biology and disease epidemiology for effective disease management. Potato Research 52, 17-37.

Mihovilovich E, Munive S and Bonierbale M, 2010. Inﬂuence of day-length and isolates of Phytophthora infestans on ﬁeld resistance to late blight of potato. Theoretical and Applied Genetics 120, 1265-1278.

Ňústez López C.E., 2011. Variedades colombianas de papa. Universidad Nacional de Colombia, Faculdad de Agronomia. 46p, http://papaunc.com/2011_flipbook_ Variedades_colombianas_de_papa/index.html)

Rabobank, 2009. World Potato Map. In. Utrecht, The Netherlands: Rabobank. Food & Agrobusiness Research and Advisory Department.

Rodriguez D., M. Ojeda, M.P. de Camacaro, M. Gallardo, R. Valera and F. Bittara, 2009. Production, powdery scab incidence and quality of potato advanced clones. Revista de la Facultad de Agronomia de la Universidad del Zulia 26 (4), 508-531.

Schwaerzel R, 2002. Sensibilités des racines et tubercules des variétés de pommes de terre à la gale poudreuse et quelques résultats de lutte chimique. Revue Suisse Agricole 34, 261-266.

Shah F, Falloon RE, Butler RC, Lister RA, 2012. Low amounts of Spongospora subterranea sporosorus inoculum cause severe powdery scab, root galling and reduced water use in potato (Solanum tuberosum). Australasian Plant Pathology 41, 219-228.

142

Torres H., M.A. Pacheco and E.R. French, 1995, Resistance of potato to powdery scab (Spongospora subterranea) under andean field conditions. American Potato Journal 72, 355-363. van de Graaf P, Wale SJ, Lees AK, 2007. Factors affecting the incidence and severity of Spongospora subterranea infection and galling in potato roots. Plant Pathology 56, 1005-1013.

143

CHAPTER 4: GENERAL DISCUSSION AND CONCLUSIONS

4.1 GENERAL DISCUSSION AND CONCLUSIONS

The pathogen Spongospora subterranea f. sp. subterranea (Sss) causes important diseases of potato (Solanum ssp.), manifesting as powdery scab on tubers and galls on roots. These diseases occur in all major potato producing areas of the world, and cause economic losses. No completely effective method exists to protect plants and soils from the pathogen. Additionally, the resting spores of Sss are very resistant to environmental stresses and chemical pesticides, and remain infective in soils for many years.

Breeding of resistant potato cultivars is currently the most promising approach for management of these diseases, but genetic information on the pathogen, required to develop robust resistance breeding strategies, is scarce. The research described here is the first thorough study to determine the population genetic structure of Sss on a global scale, which aims to provide knowledge to assist resistance breeding in potato, and thus provide a solid basis for improved powdery scab management.

4.1.1 MARKER DEVELOPMENT

Microsatellite markers (also referred to as short tandem repeats (STRs) or simple sequence repeats (SSRs)) have been developed specifically for Sss. Development of specific markers for this organism was challenging. The pathogen is an obligate biotrophic parasite, requiring living hosts to survive and reproduce; it cannot be cultivated on artificial media in pure culture. Thus, extracted DNA of Sss is always contaminated with host DNA and often with microorganisms living in the soil. To ensure the presence of Sss DNA in samples of the pathogen, all samples were subjected to a Spo8/Spo9-PCR (Bulman & Marshall, 1998), which amplifies a partial Sss-ITS-fragment. Furthermore, several control analyses were performed to ensure the specificity of the markers for Spongospora. During microsatellite library establishment, all STR sequences (marker candidates) were used in a BLAST (Altschul et al., 1990) search against several databases. Sequences with hits, such as microorganisms, were excluded. The remaining sequences were kept in the library.

147

Primer pairs annealing in the flanking regions of the microsatellites were designed and applied to Sss samples and potato samples. Primer pairs amplifying on Sss samples only were used for screening. A final set of five primer pairs amplifying six loci was obtained. The loci were all polymorphic, but the number of markers was too few for a detailed study. To compensate for the low number of microsatellite markers, ITS and actin sequence data of Sss were included in the present study. The markers were applied to more than 700 samples of the pathogen in the collection. These samples were obtained from geographically separate regions from all major continents, different resting spore sources of the pathogen (root galls or tuber lesions) and from different potato host (sub)species, to provide a comprehensive picture of the current population genetic structure of Sss.

4.1.3 SAMPLE CLUSTERS

In our PCA analysis, three Sss sample clusters were observed (Fig. 11b). These clusters most likely describe the three main Sss groups in the world, causing economic loss in potato cropping. Two of them occur in South America, the proposed native region of Sss. The first cluster corresponded to root galls from South America (group A) and the second cluster corresponds to tuber lesions from South America (group B). The third group includes samples from elsewhere, independent of the resting spore source (root galls or tuber lesions, group C). All clusters were clearly separated from each other.

This suggests that an ecological adaption in the native region due to co- evolutionary processes and/or competitive exclusion has taken place. The distinction between root gall and tuber lesion samples from South America, however, was not significant. This indicates that Sss from tuber lesions and Sss from root galls are not distinct species. Further investigations could likely result in the detection of different ecotypes.

The sequence data of ITS and actin were used to build haplotype networks (Supplementary Figure 1). Again, the invasive Sss samples were clearly separated from the native Sss samples. The native root gall samples did not cluster with the

148

native tuber lesion samples of Sss. Consistent with the microsatellite analyses, in the concatenated ITS and actin haplotype network, the same three clusters were observed.

In South America, the diseases caused by Sss were determined by the resting spore source. Root gall and tuber lesion samples each had a distinct genotype. Elsewhere, however, the resting spore source did not influence the Sss genotype, because all samples usually had the same genotype, distinct from the South American genotypes. There were a few exceptions; a few tuber lesion samples from Asia and Europe clustered closer to South American tuber lesion samples than to the other group C samples (cHap9). This genotype is probably less fit in the introduced climates or less aggressive on the host plants in the invaded regions, than the genotype of Sss commonly found in those regions (cHap6).

All three types of Sss have been found in South America (one sample with the group C genotype was detected in South America, data not shown). Future studies should focus on this region, because it is likely that this is the native region of Sss. With a more detailed investigation of South American strains of Sss, more knowledge about the evolution of this pathogen may be gained.

4.1.4 INFLUENCE OF THE INOCULA, HOSTS AND ENVIRONMENT ON DISEASE DEVELOPMENT

To examine the influence of different factors in the Sss-host-environment system, a cross inoculation experiment was carried out using a micro-bioassay (Merz et al., 2004). Different Sss inocula from native and invaded origins, from three different potato (sub-)species and two different resting spore sources (root galls or tuber lesions) was applied separately to potato plantlets of different subspecies. All Sss inocula led to infection on all host (sub)species, and and disease incidence and severity were assessed prior to harvesting genotyping of resting spore structures (sporosori).

The S. tuberosum ssp. tuberosum (Tu) host inoculated with sporosori from the invaded region had the greatest amount of infection. South American host

149

plantlets (Solanum tuberosum ssp. andigena (An) and Solanum phureja (Ph)) showed least infection with all Sss inocula. The controlled environment chamber used for the experiment had long-day settings, which may have influenced the susceptibility of the short-day hosts An and Ph to root infection. The host Ph is reported to be susceptible to powdery scab on tubers and roots (Ňústez López, 2011), but no tubers were produced by An and Ph at the root infection assessment. After conditions were changed to short days (after the assessment), tubers were formed by the short-day hosts. No tuber infection was found on any of these tubers, but increased rates of root infection were recorded.

These results demonstrate that the daylength conditions may influence the host plants. Still unclear is the influence of daylength on the disease development. As a biotrophic parasite, Sss depends on living hosts. Thus, the daylength conditions in the trial could have biased the results of the short-day hosts. In future studies, light conditions should be carefully adjusted to the host’s requirements to avoid any bias.

Inocula, harvested root galls, and harvested tuber lesions (observed on Solanum tuberosum ssp. tuberosum cultivar ‘Agria’ only) from this experiment were genotyped, to investigate the influence of different types of inocula on infection of host plants. Genotyping of inocula revealed two genotypes in the native inocula and one in the invasive inocula. These corresponded to the previously determined three clusters for native root gall, native tuber lesion, and invasive inoculum (from galls and lesions). Harvested root galls had the invasive genotype, with the exception of one sample. In contrast, the tuber lesions all had the South American lesion genotype in common, and the native root gall genotype was not observed in the harvested material.

These results suggested that the Sss inocula consisted of resting spore mixtures with different genotypes and that during host infection an organ specific selection may take place. Some resting spore genotypes could be more likely to produce root galls whereas others are more likely to produce tuber lesions.

150

Group A is probably the most ancient, which may also explain why the inoculum CPG (Colombian root gall inoculum of Solanum phureja) was the most diverse in the in ITS fragment cloning experiment performed on the five inocula. Later, tuber-bearing solanaceous plants were domesticated and the group B family of Sss genotypes which causes tuber infection and is found in Colombia, Venezuela and Peru, mixed with the group A. Last to evolve or join the other groups was probably group C, which remains a minority in South America. However, group C is able to reproduce both on roots and tubers and is possibly more adapted to milder climates, i.e. South American lowland conditions. The group C was then introduced to Europe, probably together in a mixture with group B tuber inoculum. In Europe with its longer days, group C individuals might have had the advantange of already being adapted to these climatic conditions and outperformed the others (i.e. were more virulent). Thus they became the predominant type of Sss in Europe and later in the world.

Still unresolved is why most of the inocula led to galls with the group C genotype and why tuber lesions showed the group B genotype. To investigate this, the influences of the host (sub)species or the infection location must be described. For that purpose, inocula and harvested material could be cloned. Probably there is selection corresponding to the host organ (roots or tubers), reducing the number of clones. Another hypothesis is that there is an ongoing speciation event. Prerequisites for this are already established: Geographic separation and different hosts which are grown in the different regions.

For future experiments, single genotype inocula could be beneficial. These could be obtained by establishing callus cultures of young root galls (Bulman et al., 2011; Burki et al. 2010) from the first inoculation experiment. These could be used for re-infection in a future experiment. After some passages the genotypes could be purified and distinguished. Using single genotype inocula in experiments may be beneficial to further examine and elucidate organ specific selection.

151

4.1.5 ORIGIN AND MIGRATION PATTERNS OF SPONGOSPORA SUBTERRANEA

The microsatellite data set was divided into seven geographical regions for further analyses: Europe, Africa, Asia, Australasia, North America and South America. To account for the two clusters found (see below) for native Sss samples, the South American sample set was divided into root gall and tuber lesion samples. The genetic and genotypic variabilities were slight overall. Greatest diversity was found in South American samples, which were significantly more diverse than samples from all other regions of the world. This suggested that South America is likely to be the native region of Sss.

The program MIGRATE (Beerli & Palczewski, 2010), was used to determine ancient migration patterns. This indicated that the first introductions of Sss came from South America to Europe, supporting the hypothesis of South America being the native region of the pathogen. Due to the limited population substructuring, time scales could not be calculated consistently. The aim of future studies should be to add time scales to evolutionary events and migration. A more variable sample set is needed for this task. Such a set could most likely be collected in South America.

The PCA analyses indicated the presence of very little population substructuring for introduced and native regions (Fig. 11). South American samples, however, were genetically more variable, consistent with the hypothesis that South America is the native region of Sss. The shallow substructure and reduced genetic diversity in sample collections from invaded regions could be due to founder effects. They furthermore indicate a rapid global invasion process and/or restricted origin of all introduced populations, probably combined with a lack of recurrent gene flow from the native region.

The PCA and migration analysis results taken together with a recent potato trade map (Rabobank, 2009) and the dissemination map for potato (Fig. 5; Fig. 12; CIP, 2008) indicated the presence of a bridgehead invasion scenario for Sss. This led to the following hypothesis: Native Sss strains were taken to Europe by the conquistadors on infected potato specimens. After only one or very few introduction events, the invasive Sss founder population was cut off from the source population

152

in South America, due to the lack of further introductions of infected potato roots or tubers to Europe. In Europe, the founder population of Sss was able to establish even though it had very limited genetic diversity and introduction of just few genotypes led to a severe bottleneck. Historic human activities, such as sending potato tubers as gifts to courts and botanical gardens in the early years of potato in Europe (Kiple & Ornelas, 2000) or settlers bringing the potato to colonies around the world (Fig. 5; Fig. 12), the pathogen was further distributed as a contaminant on shipped potatoes. Potato trade, especially of seed tubers, and similar human activities, have probably been important for exchange and dispersal of Sss material. The least diversity was detected for North American samples. This may be due to bottlenecks which have occurred, when the potato was brought first from South America to Europe, from there to the Bermudas and then further to North America (Fig. 5; Fig. 12; CIP 2008), which further supports the above hypothesis.

To exactly determine the centre of origin of Sss, more samples from South America should be collected and analysed. These samples should be collected from the known geographic range of the Solanaceae and from all known hosts which produce resting spores of the pathogen. Including all of these hosts may assist to determine if Sss coevolved with the potato or other hosts, and if host shifts have occurred.

4.1.6 RESISTANCE BREEDING AND QUARANTINE MEASURES

There has not been any economically important and recent potato trade activity between the introduced regions and the native region of Sss (Rabobank, 2009). Furthermore, the global genetic variability of Sss is very low, probably the most important finding. This could be advantageous for breeders, because varieties selected for Sss resistance in a country in the invaded regions could be grown in other invaded countries without losing resistance to the pathogen.

For some clonal pathogen lineages it has been reported that different phenotypes can occur with variable virulence (Blandón-Díaz et al., 2012). This, however, does not seem to be the case for Sss. Merz et al. (2012) assessed the root

153

galling and tuber lesion formation by Sss on cultivars selected for resistance to powdery scab used in geographically different regions with Sss history. They found that the cultivar ‘Gladiator’ bred in New Zealand (Genet et al., 1995) performed best with low susceptibility to root and tuber infection in four successive years in all regions, even where soil inoculum levels were high.

Whether with or without resistant potato cultivars, quarantine measures should continue or be enforced and improved to keep global genetic diversity levels of the pathogen low by avoiding the exchange of genetic material, especially between the native (South American) and invaded regions. The role of sexual reproduction in the pathogen’s life cycle remains unclear, but allowing for exchange of genetic material could result in sexual recombination. More aggressive strains of the pathogen could arise, which are even more difficult to control than the mostly clonal strains which currently cause disease.

After the inoculation experiment, the necessity of quarantine measures was reassessed. Apart from the possibility of sexual reproduction, new introductions of pathogen strains from the native region may not have such a strong influence on the disease severity. When Sss reached the invaded regions on infected potato specimens, probably a mixture of several genotypes was introduced to the invaded regions, only one genotype established successfully. What happened to other genotypes, after the first introduction(s) of Sss to Europe, and what will happen in case of future introductions in native Sss material, is now open to speculation. The invasive Sss genotype was probably fit enough to establish almost immediately after its arrival in Europe, whereas the other Sss types were not able to establish, and are unlikely to be able to do so in the future. In this case, quarantine measures would not have to be very strict.

Some risks still remain. Due to coincidence, it could be that genotypes other than the invasive one have not yet been introduced, but could be able to establish as well. Furthermore, the possibility of sexual reproduction between different Sss strains under field conditions cannot be excluded. South American potato breeders allow wild potato species to hybridize with commercially grown varieties of potato

154

(FAO, 2009), which is a potential further risk factor. Strains of Sss from the wild, which have previously not occurred on commercial varieties, could be introduced to agroecosystems. Thus, as a final conclusion, strict quarantine measures should be continued.

4.1.7 FUTURE PROSPECTS FOR SPONGOSPORA GENETIC RESEARCH

Future studies would benefit from annotated genome sequences of all three Sss types. The absence of genome sequences limits the possibilities and analysis methods of the pathogen’s genetic properties. However, Sss is an obligate biotrophic parasite, which cannot be grown as a pure culture. To retrieve Sss sequences, callus cultures of the pathogen (Bulman et al., 2011; Burki et al. 2010) could be used for DNA extraction prior to sequencing. This DNA would only be a mixture of potato host DNA and Sss DNA. The potato DNA sequences could be identified bioinformatically and excluded, because the potato genome has been recently published (Potato Genome Consortium, 2011). Another method to obtain pure Sss DNA could be the cultivation of Sss in sterile root cultures (Qu & Christ, 2007), after zoospore release from the root epidermis cells of the host. These could be collected and used for DNA extraction. This method has the advantage that the DNA samples would not be contaminated with host DNA.

A deeper insight into the genetics of Sss would pioneer the way for many further analyses, which could be beneficial for the understanding of pathogen-host- interactions, pathogenicity, and probably answer the open questions concerning the life cycle and mode of reproduction of Sss. This would assist in evaluating and optimizing established quarantine measures and the development of new systems. Resistance breeding of potato plants might be further simplified and practicable, using this type of knowledge.

155

4.2 REFERENCES

Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ, 1990. Basic Local Alignment Search Tool. Journal of Molecular Biology 215, 403-10.

Beerli P, Palczewski M, 2010. Unified Framework to Evaluate Panmixia and Migration Direction Among Multiple Sampling Locations. Genetics 185, 313 -U463.

Blandón-Díaz JU, Widmark A-K, Hannukkala A, Andersson B, Högberg N, Yuen JE, 2012. Phenotypic Variation Within a Clonal Lineage of Phytophthora infestans Infecting both Tomato and Potato in Nicaragua. Phytopathology 102, 323-30.

Bulman SR, Marshall JW, 1998. Detection of Spongospora subterranea in potato tuber lesions using the polymerase chain reaction (PCR). Plant Pathology 47, 759-66.

Bulman S, Candy JM, Fiers M, Lister R, Connera AJ, Eady CC, 2011. Genomics of Biotrophic, Plant-infecting Plasmodiophorids Using In Vitro Dual Cultures. Protist 162 (3), 449–461.

CIP, 2008. The Odyssey of the Potato. Der lange Weg der Kartoffel zur weltweiten Kulturpflanze. In: (Cip) IPC, ed. Lima, Peru: International Potato Center (CIP).

Cornuet J-M, 2010. Inference on population history and model checking using DNA sequence and microsatellite data with the software DIYABC (v1.0). BMC bioinformatics 11, 401.

Genet RA, Braam WF, Gallagher DTP, Anderson JaD, Lewthwaite SL, 1995. 'Gladiator': a new potato cultivar with high resistance to potato cyst nematode and

157

powdery scab suitable for french fries and fresh market. New-Zealand Journal of Crop and Horticultural Science 23, 105-7.

Kiple KF, and Ornelas KC, 2000. The Cambridge World History of Food. ISBN:9780521402163. Cambridge University Press. England.

Ňústez López C.E., 2011. Variedades colombianas de papa. Universidad Nacional de Colombia, Faculdad de Agronomia. 46p, http://papaunc.com/2011_flipbook_ Variedades_colombianas_de_papa/index.html)

Potato Genome Consortium, 2011. Genome sequence and analysis of the tuber crop potato. Nature 475, 189-195.

Rabobank, 2009. World Potato Map. In. Utrecht, The Netherlands: Rabobank. Food & Agrobusiness Research and Advisory Department.

158

ACKNOWLEDGEMENTS

I would like to thank all the people who have supported me during the time of my studies In particular I thank the following people and organisations for support during my studies:

Prof. Bruce A. McDonald gave me the opportunity to work in his group, and gave me valuable advice and guidance at critical stages of the project, especially in the project’s early stages.

Prof. Richard E. Falloon and Dr Ueli Merz gave support, guidance, supervision, and the opportunity to work one year at the New Zealand Institute for Plant and Food Research, Ltd. I appreciate their efforts and time, given beyond simple student- supervisor relationships. I am grateful for the friendship, knowledge, and patience offered me along the often bumpy road of the project, which gave me strength and many memories to treasure.

Dr Patrick Brunner, one of my co-examiners, gave valuable supervision during all computational work in this study.

Horticulture Australia Ltd (APRP1, University of Tasmania), the ETH Zurich, the United Kingdom Potato Council, the New Zealand Ministry of Science and Innovation, Horticulture New Zealand, and the Swiss Federal Office of Agriculture provided financial support for this project.

The Genetic Diversity Centre of ETH Zurich, Switzerland, especially Tania Torossi and Aria Minder, assisted with data generation and evaluation.

The New Zealand Institute for Plant and Food Research, Ltd, in Lincoln, Canterbury, New Zealand and the Bioprotection Centre at Lincoln University provided facitities for my 1 year period of work in New Zealand.

The Plant Science Graduate School supported this study, and especially Eliane Zumstein gave friendship and provided important information.

159

To the ETH crew: Mr Marcello Zala and Mrs Ulrike Rosenberger for so many nice conversations and answering all questions around supplies, methods and administration. Big thanks also to my colleagues Mrs Jana Schneider, Ms Juliane Weger, Mr Mark Lendenmann, Ms Fabienne Di Gennaro, Dr Tryggvi Stefansson, Dr Stefano Torriani, Dr Pierre-Marie Le Roux, Dr Monika Maurhofer, Dr Daniel Croll, Dr Giovanni Broggini, Mr Michele Gusberti, Dr Megan McDonald, Mr Ethan Stewart, Dr Joana Meyer, Dr Joana Bernandes de Assis, Dr Andreas von Felten, Dr Johannes Fahrentrapp, Dr Gabriella Parravicini, Dr Michelangelo Storari, and Prof. Cesare Gessler. It was great to work with these people, and to spend many nice hours in the social events associated with the Department at ETH.

Ms Lina Ramos and Prof. Masao Arakawa, left me important an impression, even although their stays at the ETH were short.

I also thank the people at Plant and Food Research in New Zealand: Mrs Monika Walter, helped the damsel in distress and placed me at Ms Lee Smit’s place, who I also want to sincerely thank for all her kindness and friendship. Dr Soonie Chng, Mr John Fletcher, and Dr Mathew Cromey, gave me support and friendly interactions. Mrs Kirsty Boyd-Wilson assisted with the MAF procedures for import of biological material, Ms Sarah Murray, and Dr Gail Timmerman-Vaughan gave valuable assistance with the use of “GeneZilla”, making the genotyping possible. Dr Farhat Shah, Mrs Sandra Visnowsky and Mr Ikram Khan provided many good ideas and a nice atmosphere in the lab. Finally, I particularly thank Ms Loreto Maldonaldo- Hernandez, Ms Kathrin Erlenbach, and Mr Subha Das, who proved to be real friends during all good and bad times, including the Christchurch earthquakes.

Ich danke ausserdem meinem Biologielehrer Herrn Peter Rausch, dessen spannender Unterricht den Grundstein für meine Faszination für die Wissenschaft des Lebendigen gelegt hat.

Meinen Freunden in und um Bielefeld danke ich für viele lustige Treffen und das Gefühl nach Hause zu kommen, egal wie viel Zeit vergangen ist, darunter besonders: Herrn Daniel Georgi, Frau Dr. Christin Klenke, Frau Dr. Eva Trost, Herrn

160

Robert Gehring, Frau Christine Niebuhr, Herrn Ralf Mertens, Herrn Dr. Burkhard Linke, Herrn Fabian Brod, Herrn Christian Maiwald, Frau Dr. Sabrina Vanessa Klein, Herrn Stephan Echterhoff, Frau Ramona Vormbrock, Frau Sonja Kleineberg, Herrn Sebastian Voges, Frau Victoria Gödde, Herrn Tim Stückemann, Herrn Niels Hübner, den Damen aus der „Kontaktstelle Wissenschaftliche Weiterbildung“ (KWW), dem „Zentrum Wissenschftliche Weiterbildung" (ZWW) und „Studieren ab 50“ in der Universität Bielefeld, besonders: Frau Claudia Dietz, Frau Sabine Böhling, Frau Bianca Gorys und Frau Dr.Ursula Bade-Becker. Ausserdem danke ich Frau Tamara Sarter, Herrn Steffen Mürdter und Herrn Benjamin Korduan.

Ich möchte auch meinen Freunden in der Schweiz danken. Besonders: Herrn Simon Waldvogel für die Hilfe bei den Umzügen. Herrn Benjamin Spiess für den Platz auf dem Dachboden. Ausserdem danke an Herrn Tobias Senn, Frau Manuela Bärenfeld, und Herrn Dirk Blom für die Freundschaften, die so unerwartet kamen. Danke, dass ihr alle eine deutsche Einwanderin so herzlich aufgenommen habt.

I would also like to thank Mr Christopher Middleton: Spending time with you has always been „coel“! I wish you heaps of success and good luck for the final stages of your own dissertation including all the gap-filling in your sequences and the aroused feelings while writing up your research.

Meinen Eltern Wolfgang und Ira Gau danke ich dafür, dass sie ihre Jüngste in die weite Welt ziehen lassen und in Bielefeld immer auf mich warten. Meinen Geschwistern Christine Wiebe, Elfriede Wiebe, Marlene Wiebe, Wilfried Wiebe und Helmut Wiebe und der jüngeren Generation - meinen Nichten und Neffen - Teresa de Simone, Juliane Wiebe, Anna Haindl, Ida de Simone, Emily Heikel, Leon Heikel und Pauline Valentin. Vor allem ein dickes Dankeschön an Elfriede für die Hilfe beim Korrekturlesen und dass ich immer willkommen bin, wenn ich zu Besuch in Bielefeld bin. Ausserdem danke an Fanny, niemand sonst begrüsst mich so aufgeregt und fröhlich, wenn ich mal wieder im Lande bin. Schön, dass es euch gibt und „Sehen wir uns nicht in dieser Welt, dann sehen wir uns in Bielefeld.“

161

Zuguterletzt danke ich meinem Freund Nicolin Ragaz und seinen Eltern Georg und Edith Ragaz, die mir alle bei der Wohnungssuche, auf den Ämtern, beim Einleben und Erleben der Schweiz geholfen haben. Ausserdem danke ich Curdin Ragaz für seinen grossartigen Humor und seiner Frau Sandra für ihre Herzlichkeit und beiden zusammen für die guten Ratschläge zur ETH und der Graduate School. Bei euch ist Zuhause und mit euch bin ich glücklich.

162