Phenotyping Ornamental Plants: Case Studies in Fruit Quality Attributes from Vaccinium Ovatum and Disease Resistance of Cotoneaster Spp

AN ABSTRACT OF THE THESIS OF

Kristin E. Neill for the degree of Master of Science in Horticulture presented on March 16, 2020

Title: Phenotyping Ornamental Plants: Case Studies in Fruit Quality Attributes from Vaccinium ovatum and Disease Resistance of Cotoneaster spp.

Abstract approved:

______Ryan N. Contreras

Landscape plants are highly diverse and nursery producers often grow thousands of taxa to satisfy the varied needs and aesthetic tastes of consumers. Related to this diversity, ornamental plant breeders must be nimble and diverse in their approach. Two seemingly disparate studies were conducted to investigate improving ornamental shrubs – specifically, fruit quality and disease resistance. Vaccinium ovatum Pursh. (evergreen huckleberry) is an evergreen shrub native to the Pacific Northwest. With the notable exception of cultivated blueberry (Vaccinium corymbosum) species in Vaccinium are primarily diploids (2n = 2x = 24) but polyploids have been reported from experiments with chemical mitotic inhibitors as well as naturally occurring tetraploids in cultivated blueberry (Vaccinium corymbosum). There is interest in this species, but it requires improvement of the fruit and plant qualities for an eventual cultivar release. To obtain variation in plant qualities polyploidy was induced in a collection of plants in 2013. The purpose of this study was to assess the impacts of polyploidy on the fruit and plant qualities of Vaccinium ovatum. This fruit and plant quality study provides a contribution to the scientific knowledge base that is currently lacking on evergreen huckleberries. Plant qualities were determined by

measuring plant height and width, obtained in fall 2017. The fruit volume (mm3) and brix (°Bx), for soluble solids content, were measured using a digital caliper and a digital refractometer respectively. Measurements were taken on diploid, mixoploid, and tetraploid (2x, 2x + 4x, 4x) cytotypes, once in 2017, five times over nine weeks in 2018, and three times over nine weeks in

2019. Fruit volume in 2017 increased from diploids to tetraploids (P < 0.0001), brix also increased in these cytotypes. Data from 2017 suggested there was a “gigas” effect from of polyploidy in evergreen huckleberries. However, 2018 and 2019 fruit volume of tetraploid fruit was smaller than that of diploid and mixoploid. Differences were observed in diploid fruit volume among all years (P < 0.0001). In tetraploids, brix was statistically significant across all years (P =

0.0002). The variation in data from the various years suggests that using ploidy as a way to produce larger sweeter fruit is not a plausible method. The genus Cotoneaster Medik. is composed of around 400 species with a wide variety of growth habits and form. These hardy landscape shrubs use to be commonplace because of their low maintenance and landscape functionality. However, the interest in and sales of Cotoneaster have declined for a variety of reasons, the largest being its susceptibility to a bacterial disease fire blight, caused by Erwinia amylovora. The resistance of 15 different genotypes of Cotoneaster was tested by inoculating leaves with a wildtype (Ea153) and an avrRpt2 strain (LA635) of Erwinia amylovora. Four studies took place in climate-controlled growth chambers and one study took place in a greenhouse in Corvallis, OR. Fire blight resistance was assessed by calculating the percent shoot necrosis (PSN = 100*(lesion length/total branch length)) once a week for six to eight weeks after inoculation. Across all studies, genotypes H2011-01-002 and H2011-02-001 consistently had the lowest levels of percent shoot necrosis. Plants inoculated with different isolates were directly compared from growth chamber studies done in 2019. Genotype H2011-02-005 was significantly

more resistant to EA153 than to LA635 while C. splendens was significantly less resistant to

EA153 than to LA635. Genotype H2011-02-001 has already been released as a new ornamental cultivar that has high fire blight resistance. Several other genotypes could be released as highly fire blight resistant for areas where the disease is widespread.

Phenotyping Ornamental Plants: Case Studies in Fruit Quality Attributes from Vaccinium ovatum and Disease Resistance of Cotoneaster spp.

Kristin E. Neill

A THESIS

Submitted to

Oregon State University

in partial fulfillment of the requirements for the degree of

Master of Science

Presented March 16, 2020 Commencement June 2020

Master of Science thesis of Kristin E. Neill presented on March 16, 2020.

APPROVED:

Major Professor, representing Horticulture

Head of the Department of Horticulture

Dean of the Graduate School

I understand that my thesis will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my thesis to any reader upon request.

Kristin E. Neill, Author

ACKNOWLEDGEMENTS

I would like to express my appreciation to Dr. Ryan Contreras for his support, advising, and encouragement during the course of this research. Ryan, you were an incredible resource and mentor throughout my entire time at Oregon State, thank you for challenging me and giving me the resources to go out and be successful in the real world. The invaluable support of my committee members is also sincerely appreciated, including Dr. Virginia Stockwell, Dr. Chad Finn, and Dr. Ken Johnson. A huge thank you to Dr. Virginia Stockwell for her invaluable advisement and massive contributions to helping me learn plant pathology and so much more. Thank to Dr. Finn for being such a great mentor and working with me during my time here. You left an impact on everyone you ever spoke to and you are sincerely missed by all who knew you. For stepping in at the last second and taking over Dr. Finn’s role, I would sincerely like to thank Dr. James Myers.

I would also like to thank my former advisors. Thank you, Bryce Lane, for inspiring me to join the world of horticulture and facilitating this love throughout my time at NC State and Oregon State. Thank you, Dr. Dennis Werner, for providing mentorship as an advisor and as my professor, thank you for inspiring me to go into ornamental plant breeding. Thank you to Lis Meyer for being a wonderful mentor.

None of my research would have been possible without the support of technicians, undergraduate students and fellow graduate students to help maintain plant material, do inoculations, and collect data. I would like to thank the technician in the Ornamental Plant Breeding Program, Tyler Hoskins, without whom many of my plants would not have survived. I would also like to thank the hardworking undergrads, Patrice Woodworth, Amy Kessler, Josh Ashcraft, Andrew Baker, and Adigun McLeod. To all the fellow students and friends, I’ve met in Oregon, I can easily say that I wouldn’t have made it if it weren’t for you. So, a huge thank you to Hsuan Chen, Bryan Webber, Ryan Hill, Larissa Larocca, Katerina Graham, and many more. I cannot imagine having taken this journey without all of you supporting me. Best of luck to each and every one of you.

I must thank my family for their love and support. Thank you to my mom, Karen Neill, thank you for your unconditional love and for your constant words of encouragement. You have

always encouraged me to chase my dreams even when I have been ready to give up. You are the biggest reason that I am in horticulture and my love of plants will always start and end with you.

Most importantly, I must especially thank Harrison Stierwalt. I’m so thankful for my time at Oregon State bringing us together. I can honestly say that I would not have made it through this challenge had it not been for your constant unwavering support through everything. I love you and cannot wait to continue our adventures following your dreams in Kansas City. A small thank you to our pup Oliver; you were there for the worst of it (the last 5 months of this Masters) and you still always managed to make me smile at the end of the day.

For all those mentioned above and for all those I have forgotten, you were all so instrumental in making this thesis a reality. Endless thank yous to every single one of you.

CONTRIBUTION OF AUTHORS

Dr. Ryan Contreras was involved in the planning, design, analysis and interpretation of data in all chapters of the thesis. Dr. Virginia Stockwell was involved in the preparation of bacterial strains and interpretation of results in Chapter 3.

TABLE OF CONTENTS Page CHAPTER 1: GENERAL INTRODUCTION ...... 1 Nursery Industry ...... 1 Vaccinium ovatum ...... 2 Cotoneaster spp...... 4 Literature Cited ...... 8 CHAPTER 2: ASSESSING THE IMPACT OF CHROMOSOME DOUBLING ON FRUIT AND PLANT QUALITIES OF VACCINIUM OVATUM ...... 13 Abstract...... 13 Introduction ...... 14 Materials and Methods ...... 19 Plant Material...... 19 Flow Cytometry ...... 20 Plant Measurements...... 21 Statistical analysis...... 22 Results ...... 22 Genome Size and Plant Size...... 22 Fruit measurements on a single date...... 23 Fruit measurements in 2018 and 2019...... 24 Discussion ...... 26 Literature Cited ...... 32 Figures ...... 37 Tables ...... 46 CHAPTER 3. SCREENING COTONEASTER SPP. FOR RESISTANCE TO FIRE BLIGHT USING FOLIAR INOCULATION WITH TWO STRAINS OF ERWINIA AMYLOVORA ...... 52 Abstract...... 52 Introduction ...... 53 Materials and Methods ...... 56 Plant Material...... 56 Bacterial strains...... 57 Preparation of freeze-dried inoculum...... 58 Inoculation using leaf bisection and data collection...... 58 Koch’s Postulates...... 59 Experimental design...... 60 Statistical analysis...... 61

TABLE OF CONTENTS (cont.) Page Results ...... 62 2018 Growth chamber studies 1 and 2...... 62 2019 Growth chamber studies 3 and 4...... 63 2019 Study 5...... 64 Discussion ...... 65 Literature Cited ...... 70 Figures ...... 75 Tables ...... 84 CHAPTER 4: GENERAL CONCLUSION ...... 89 Comprehensive Literature Cited ...... 92 Appendices ...... 102

LIST OF FIGURES Figure Page

Fig. 2.1. Flow cytometry histograms of three Vaccinium ovatum cytotypes. Nuclei > 4,000 were used for each analysis and peaks with Coefficient of variance (CV) < 10. (A) Peaks represent diploid V. ovatum (2C = 1.22 pg) and the internal standard Pisum sativum ‘Ctirad’ (2C = 8.76 pg). (B) Peaks represent tetraploid V. ovatum (2C = 2.45 pg) and the internal standard. (C) Peaks represent mixoploid V. ovatum with both a diploid and a tetraploid peak...... 37

Fig. 2.2. Fruit volume (mm3) and brix (Bx) measurements collected on a single date 08/30/17 from twenty-eight accessions of Vaccinium ovatum over three different cytotypes in 2017; 2x, n=21; 2x+4x, n=4; 4x, n=3...... 38

Fig. 2.3. Fruit volume (mm3) and brix (°Bx) measurements taken on a single date of collection from three different cytotypes of Vaccinium ovatum in each year 2017, 2018, and 2019. Mean separation done by Tukey’s (HSD) (α = 0.05); Comparisons across years within ploidy, Means followed by same letter were not statistically different based on: ...... 39

Fig. 2.4. 2018 Fruit volume(mm3) and brix(Bx) measurements collected over 9 weeks from three different cytotypes of Vaccinium ovatum; Bars represent fruit volume over weeks, Marker represents brix over time with a linear trendline fitted. A. Measurements from diploid accessions (2x, n= 24) B. Measurements from mixoploid accessions (2x+4x, n=4). C. Measurements from tetraploid accessions (4x, n=6)...... 40

Fig. 2.5. 2019 Fruit volume (mm3) and brix (Bx) measurements collected over 9 weeks from three different cytotypes of Vaccinium ovatum. A. Measurements from diploid accessions (2x, n=17) B. Measurements from mixoploid accessions (2x+4x, n=5). C. Measurements from tetraploid accessions (4x, n=3)...... 41

Fig. 2.6. Fruit volume (mm3) and brix (Bx) measurements collected on five separate sampling dates over 9 weeks in 2018 from three different cytotypes of Vaccinium ovatum; 2x, n= 24; 2x+4x, n=4; 4x, n=6...... 42

Fig. 2.7. Fruit volume (mm3) and brix (Bx) measurements collected on three separate sampling dates over nine weeks in 2019 from three different cytotypes of Vaccinium ovatum; 2x, n=17; 2x+4x, n=5; 4x, n=3...... 43

Fig. 2.8. Scatterplot of all 2018 fruit volume (mm3) and brix (Bx) measurements taken over 9 weeks in Vaccinium ovatum. Legend: 2x- blue circle; 2x+4x red triangle; 4x yellow square...... 44

Fig. 2.9. Scatterplot of all 2019 fruit volume (mm3) and brix (Bx) measurements taken over 9 weeks in Vaccinium ovatum. Legend: 2x- blue circle; 2x+4x red triangle; 4x yellow square...... 45

LIST OF FIGURES (cont.) Figure Page

Fig. 3.1. Study 1. Percent Shoot Necrosis of Cotoneaster genotypes inoculated in growth chambers with Erwinia amylovora strain LA635 (108 colony-forming units/mL) in 2018 with a foliar bisection assay. Mean percent shoot necrosis from fire blight from 12 plants. Means and standard errors are reported in percent shoot necrosis. Letter above bars represent a Fisher’s protected Least Significant Difference (LSD) test ( < 0.05) with a least significant difference value of 8% shoot necrosis...... 75

Fig. 3.2. Study 2. Percent Shoot Necrosis of Cotoneaster genotypes inoculated in growth chambers with Erwinia amylovora strain Ea153 (108 colony-forming units/mL) in 2018 with a foliar bisection assay. Means and standard errors from 9 plants are reported in percent shoot necrosis. Letter above bars represent a Fisher’s protected Least Significant Difference (LSD) test ( < 0.05) with a least significant difference value of 17% shoot necrosis...... 76

Fig. 3.4. Study 5. Percent Shoot Necrosis of Cotoneaster genotypes inoculated in greenhouse with Erwinia amylovora strain Ea153 (109 colony-forming units/mL) in 2019 with a foliar bisection assay. Means and standard errors from 9 plants are reported in percent shoot necrosis. Letter above bars represent a Fisher’s protected Least Significant Difference (LSD) test ( < 0.05) with a least significant difference value of 12% shoot necrosis...... 78

Fig. 3.5. Area under the disease progresss curve (AUDPC) of 3 different genotypes from the 2019 Cotoneaster app.-fire blight growth chambers studies. Genotypes H2011-02-005 and C. splendens displayed significantly different percent shoot necrosis between the two strains. Genotype H2017-005-01 displayed similar effects with both strains of the bacteria...... 79

Fig. 3.6. Confirmation of Koch’s postulates of Erwinia amylovora, causal agent of the bacterial disease fire blight in Cotoneaster genotypes, A. KB amended with cycloheximide and streptomycin, lack of growth on KB amended with streptomycin plates confirm the presence of Ea153. B. Strain LA635 were spread on both KB media amended with cycloheximide and streptomycin and LB media. C. Mucoid, domed, white colonies on the streptomycin amended media confirmed the presence of the streptomycin-resistant strain LA635 in the inoculated plant. D. Semi-selective medium for E. amylovora called CCT (Ishimaru and Klos, 1984) growth of domed colonies on CCT confirmed the presence of Ea153...... 80

Fig. 3.7. Erwinia amylovora, fire blight symptoms in Cotoneaster spp. A. Fire blight causing necrosis of stems below inoculation site (marked by red tape), showing the common shepherd’s crook symptom in a Cotoneaster shoot B. Necrosis of vascular tissue in the lesion border of Cotoneaster dammeri...... 81

LIST OF FIGURES (cont.) Figure Page

Fig. 3.8. Cotoneaster genotypes H2011-01-002, H2011-02-001, and H2011-02-005, after study 5 using Ea153 in the glasshouse...... 82

Fig. 3.9. Genotypes H2011-02-005 and Cotoneaster xsuecicus ‘Coral Beauty’ compared to each other after study 5, inoculation with strain Ea153...... 83

LIST OF TABLES Table Page

Table 2.1. Vaccinium ovatum accessions, genome size analyzed by DAPI flow cytometry, plant height, and width measured on plants grown in 21 L containers at Lewis Brown Research farm in Corvallis, OR...... 46

Table 2.2. Mean height and width measurements from three different cytotypes of Vaccinium ovatum...... 48

Table 2.3. Fruit volume (mm3) and brix (°Bx) measurements taken on a single date of collection from three different cytotypes of Vaccinium ovatum in each year 2017, 2018, and 2019...... 49

Table 2.4. Fruit volume (mm3) and brix (°Bx) measurements collected over nine weeks in 2018, and 2019 from three different cytotypes of Vaccinium ovatum. Sample collections began on 31 Aug 2018/ 11 Sept 2019 and concluded on 26 Oct. 2018/6 Nov. 2019...... 50

Table 3.1. Study 1. Percent shoot necrosis of Cotoneaster genotypes inoculated in growth chambers with Erwinia amylovora strain LA 635 (108 colony-forming units/mL) in 2018 with a foliar bisection assay...... 84

Table 3.4. Study 5. Percent shoot necrosis of Cotoneaster genotypes inoculated in greenhouse with Erwinia amylovora strain Ea153 (109 colony-forming units/mL) in 2019 with a foliar bisection assay...... 87

LIST OF APPENDICES

Appendix Page

Vaccinium SAS code ...... 102

Vaccinium Supplemental Information ...... 104

Cotoneaster SAS and R Code ...... 107

Media and Buffers for Bacterial Strains ...... 110

Cotoneaster Germination Study ...... 111

LIST OF APPENDIX TABLES Table Page

Table 1. 2018 Germination of Vaccinium ovatum seedlings from 5 different experiments. 104

Table 2. 2019 germination of Vaccinium ovatum from crosses made in 2018...... 104

Table 3. 2018 atLEAF Chl measurements of all Vaccinium ovatum accessions in pot, under polyhouse conditions at the Lewis Brown Farm in Corvallis, OR...... 105

Table 4. Germination percentage of five Cotoneaster genotypes based on 60 seed of each genotype sown per treatment...... 111

CHAPTER 1: GENERAL INTRODUCTION

Nursery Industry

Ornamental plants are a very important part of the agricultural world. Ornamental plants serve a different function than many commercial crops, but they still hold both monetary and environmental value. In the horticulture industry, ornamental landscape plants are considered nursery stock. Nursery stock are finished deciduous shade and flowering trees, coniferous and broadleaf evergreens, shrubs, bushes, ground covers, fruit and nut trees, grapevines, small fruit plants, and vines (USDA, 2016). The value of landscape plants does not end with environmental and economic benefits. Landscape plants are also beneficial in the creation of green spaces and gardens. There is growing interest in the positive effects that green spaces can have on public health and wellbeing (Matsuoka and Kaplan, 2008; Lee et al., 2015). Ornamental plants do not need to be limited to aesthetics, they also can be an edible plant in the landscape. Bryan and

Castle (1976) published a book describing ornamental plants that could fill that role for a kitchen garden.

The total US nursery plant industry sales were ~$27 billion in 2009 (Hodges et al., 2010).

In a recent USDA Census on Horticultural Specialties, nursery stock accrued over $4.26 billion in total sales (USDA, 2016). Nationwide broadleaf evergreens accounted for upwards of $807 million in total sales. Falling in the broadleaf evergreen section, Cotoneaster, a genus used in this thesis, account for nearly a million units sold at a value of over $7 million in total sales (USDA,

2016). Oregon is one of the top three nursery production states, right behind California and

Florida. Oregon's greenhouse and nursery products represent the largest commodity in Oregon agriculture, rising steadily from $745 million in 2012 to $995 million in 2018, based on production value (ODA, 2020). Cotoneaster sales in Oregon accounted for over 151,447 units 2

sold at a value of almost $1 million in total sales in Oregon (USDA, 2016). There is a lack of information available for evergreen huckleberries due to their history of being wild gathered and not being used as ornamental plants.

Vaccinium ovatum

Plants in the genus Vaccinium have been an important food crop for centuries. Existing all around the world with around 40 species in North America. Vaccinium ovatum, evergreen huckleberry, is a native evergreen commonly found as an understory plant in the Pacific

Northwest (Hichcock and Cronquist, 1973). Evergreen huckleberries are commonly found growing in mid-alpine regions up to 3,500 m above sea level, mountain slopes, forests, or lake basins in combination with red huckleberries (Vaccinium parvifolium) and snowberries

(Symphoricarpos albus) in acidic soils.

There are several species of huckleberry that are native to the Pacific Northwest. All produce edible fruit, but only two species are harvested in any substantial of quantity, mountain or black huckleberry (V. membranaceum) and blue leaf huckleberry (V. deliciosum) (Barney,

2003). Other species are primarily hand-harvested in small stands. Huckleberries have a long history of being collected on hidden sites managed by Native American tribes (Richards and

Alexander, 2006). In more recent years these berries are starting to be picked a little more frequently for various jams and local products. Evergreen huckleberries are not grown by consumers for the intent of harvesting massive amounts of berries. They are popular in the florist industry, where their foliage is often grown for floral arrangements (Vander Kloet, 1988).

Evergreen huckleberries that are grown in the sun produce sought after long spikes of stems with bright red leaves that arrange themselves spirally along the stem (Schlosser et al., 1992). These stems are great for florists but pose a challenge when breeding for a superior ornamental form. 3

There have been few to no studies characterizing traits of V. ovatum. This study will be the first of its kind to evaluate the plant and fruit characteristics of ploidy manipulated collection of V. ovatum plants. In Chapter 2, I will start by exploring the differences in genome size and ploidy of a collection of V. ovatum plants at the Lewis Brown Farm in Corvallis, OR. Plants underwent a ploidy manipulation, in 2013 with a chemical application of oryzalin when they were seedlings. The dinitroaniline herbicide oryzalin (4-(Dipropylamino)-3,5- dinitrobenzenesulfonamide) has been used effectively as a doubling agent and is far less dangerous to human DNA than colchicine (Morejohn et al., 1987; van Tuyl, 1992). There are many examples of oryzalin being used to manipulate ploidy for enhanced ornamental quality

(Contreras et al., 2010; Lehrer et al., 2008)

Flow cytometry is a fast and easy method for determining genome size through measuring the mass of genetic material inside the nuclei. For this study the genome size will be determined using a flow cytometer with 4’, 6-diamidino-2-phenylindole (DAPI) staining buffer.

The genome size is calibrated to chromosome number by using cytology to count chromosomes, however we can determine the ploidy of other samples by using the monoploid (one set of chromosomes: 1Cx) number to calculate ploidy (Rounsaville and Ranney, 2010; Shearer and

Ranney, 2013).

Blueberries primarily are composed of the sugar’s glucose and fructose (Barker et al.,

1963; Kader et al., 1993). Brix is used to express what the level of soluble solids are in a berry; this being the sugars and all other dissolved solids as amount per unit of the solvent water. Brix is one of the best tools to quickly measure sugar content in berries and is widely used for small edible fruits. Brix is extremely useful in viticulture where winemakers use brix when aging the grapes and many other studies have used brix to determine harvest and postharvest quality 4

(Barker et al., 1963; Kleinhenz and Bumgarner, 2012). Brix measurementis a good start to predict consumer acceptability, however acids in fruits also affects taste. A brix: acid ratio was shown to be a more effective measurement for consumer acceptability (Jayasena and Cameron,

2008).

There have been few papers published on how berries in Vaccinium develop on the shrub over the course of a season. This is a significant interest in evergreen huckleberries as the fruit stay on the shrub from August until as late as March provided the conditions are amenable

(Vander Kloet, 1988). In chapter 2, I will present preliminary data involving tracking brix and berry sizes throughout a growing season. These measurements were collected at varying times over three years, to determine whether brix and fruit volume change between years or change over time in a season.

The ornamental breeding program at OSU is exploring new possibilities of doing interploidy and interspecific breeding with V. ovatum. The most important goal for the breeding project is to get plants with a more superior form than the wild type. Fruit characteristics are second priority to the ornamental qualities in this program. However, an edible ornamental is a strong selling point for working with V. ovatum. The research done will provide some answers as to how berries mature on the plants over a season. These data will give the program enough information to recommend an optimum picking time for consumers. All of the data collected will go towards an increase in knowledge about how Vaccinium ovatum fruits mature over time and how ploidy affects these changes.

Cotoneaster spp.

Cotoneaster (Medik.) L. is both the genus and common name for a large group of ornamental ground covers, shrubs and trees. Cotoneaster is a member of the Rosaceae family 5

and falls in the subfamily Maloideae. It is closely related to both Pyracantha (M.Roem.) and

Crataegus L., however it is an unarmed, or thornless, species with leaves that have entire margins (Robertson et al., 1991). Taxonomically, the genus Cotoneaster is divided into 2 subgenera: Chaenopetalum (Koenhe) and Cotoneaster (Medik.). Subgenus Chaenopetalum has mildly fragrant white flowers with reflexed petals. Subgenus Cotoneaster has pink to red cuplike flowers that open successively over a couple of weeks. The subgenera can be further divided up into 11 sections and 37 series based on different morphological characteristics and place of origin (Dikore and Kasperk, 2010; Fryer and Hylmö, 2009; Robertson et al., 1991).

Cotoneaster are valued for being rugged and tolerant landscape plants. They are tolerant of poor soils, respond well to pruning and can tolerate pollution, making them ideal for urban landscapes. In addition to being robust plants, Cotoneaster also have good ornamental value with the much coveted multi season interest with attractive flowers, showy autumn color or green foliage that is persistent, abundant bright fruit than can persist through the winter, and a variety of architectural forms from prostrate ground covers to small trees. Despite their highly desirable ornamental qualities, very few Cotoneaster species are planted in the North American landscape

(Dirr, 2011)

Compared to other members of the Rosaceae, Cotoneaster have very few pest and pathogen problems in the landscape with only lace bugs, mites, and fire blight being most significant (Dirr, 2011). Their greatest shortcoming is their susceptibility to a bacterial plant disease called fire blight. The bacterium Erwinia amylovora is the causal agent of fire blight, which is a progressive necrotic disease prevalent in the family Rosaceae. Fire blight is native to

North America, but disease pressure is highest in environments that favor the bacterium, especially those with heat, rain, and high humidity of the South, Southeast, and Midwest United 6

States. The bacterial disease has since spread to many countries throughout Europe, into Russia and Kazahstan, around countries in the Mediterranean Sea and New Zealand (van der Zwet and

Keil, 1979).

E. amylovora survive by overwintering in stem cankers and then spreading in the spring through insect vectors, rain splash, birds, spiders, or through human operated equipment

(Wöhner et al., 2018). The natural site of infection is via the nectarthodes on flowers or through wounds created on the stem or leaves. The infected tissues begin to die showing necrotic stems that form a shepherd’s crook (Oh and Beer, 2005). The easiest method of managing fire blight is to remove infected branches through pruning, to decrease the inoculum source the next year.

There are registered chemical controls like copper and streptomycin sprays, but these are limited by regulation and gardeners and farmers are now having to contend with streptomycin resistant bacterial strains (Vanneste, 2003). The most effective form of control will be to breed new resistant species. Identifying and characterizing resistance genes and then breeding to incorporate those genes into desirable cultivars is the future of fire blight management (Norelli et al., 2003)

In Cotoneaster, there have been efforts underway to breed for resistance since 1978.

Persiel and Zeller (1978) began observations in Germany focusing on differences in resistance of many different species of Cotoneaster. They continued this research into the 1990’s, focusing on interspecific diversity of sexual cross progenies of Cotoneaster. Results from their studies and breeding, led to the release of some resistant Cotoneaster cultivars that are unfortunately not readily available in the US nursery industry. Their study paved the way for creating a more uniform method of doing resistance studies. Previous disease screenings for Cotoneaster largely relied on observations from natural infections in the field. Persiel and Zeller determined that a 7

better way to do screens was to incorporate an artificial inoculation method performed in controlled environments (Persiel and Zeller, 1978, 1981, 1990).

Breeding and screening for disease resistance will continue to be an important part of the development of disease resistant cultivars. Breeding is a multistep process that involves identifying traits of interest and dealing with pathogens overcoming resistance. An example of the bacterial disease overcoming resistance have been observed in the apple breeding program of

Peil et al; (2009). That research group had found wild resistant apple species that were being integrated into the breeding program. The most important of these being a rootstock named

Malus xrobusta 5 (Mr5). Mr5 contains an important quantitative trait locus (QTL) for fire blight resistance on linkage group (LG) 3 that has been used in the development of several new resistant apple cultivars. Relatively recently, Vogt et al. (2013) found that a strain of E. amylovora lacked the effector AvrRpt2 and was able to overcome the resistance to Mr5.

Recently it has been suggested there is a gene-for-gene relationship in the Mr5-E. amylovora pathosystem (Wöhner et al., 2018). This would make sense as it was a naturally occurring single nucleotide polymorphism that substituted a cysteine with a serine that was able to overcome this resistance (Vogt et al, 2013).

Chapter 3 reports on resistance of previously screened Cotoneaster genotypes to strain

Ea153 as well as screening of new genotypes and an additive new strain of the bacterium,

LA635, that carries a SNP in avrRpt2. The results from the studies will aid in the ornamental breeding program’s release of new resistant cultivars. The data will also tell us whether there are sources of resistance to the SNP mutant strain of E. amylovora. The information found should also be investigated further to determine the mechanisms of resistance offered by those genotypes resistant to both strains of the disease. 8

Literature Cited

Barker, W.G., F.A. Wood, and W.B. Collins. 1963. Sugar levels in fruits of lowbush blueberries

at four physiological ages. Nature 198:810-811

Barney, D.L. 2003. Prospects for Domesticating Western Huckleberries. Small Fruits Review

2:15-29

Bryan, J.E. and C. Castle. 1976. The edible ornamental garden. Pitman Ltd. London, UK

Contreras, R.N., J.M. Ruter and B.M. Schwartz. 2010. Oryzalin-induced tetraploidy in

Cryptomeria japonica (Cupressaceae). J. Amer. Soc. Hort. Sci. 45:316-319

Dickoré, W.B., and G. Kasperk. 2010. Species of Cotoneaster (Rosaceae, Maloideae) indigenous

to, naturalising or commonly cultivated in Central Europe. Willdenowia 40:13-45

Fryer, J., and B. Hlmyo. 2009. Cotoneasters; a comprehensive guide to shrubs for flowers, fruit

and foliage. Timber Press, Portland, OR

Hitchcock, C.L. and A. Cronquist. 1973. Flora of the Pacific Northwest. University of

Washington Press. Seattle

Hodges, A., M. Palma, and C. Hall. 2010 Trade flows and marketing practices within the U.S.

nursery industry, 2008, South. Coop. Ser. Bull. 411: 63.

Jayasena, V. and I. Cameron. 2008. Brix/acid ratio as a predictor of consumer acceptability of

crimson seedless table grapes. J. Food Quality. 31:736-750

Kader, F., B. Rovel, and M. Metche. 1993. Role of Invertase in Sugar Content in Highbush

Blueberries (Vaccinium corymbosum, L.). Food Sci. and Techol. 26:593-595

Lee, A.C.K., H.C. Jordan, and J. Horsley. 2015. Value of urban green spaces in promoting

healthy living and wellbeing: Prospects for planning. Risk Management and Health.

Policy. 8:131-137. 9

Matsuoka, R.H. and R. Kaplan. 2008. People needs in the urban landscape: Analysis of

Landscape and Urban Planning contributions. Landscape and Urban Plan. 84:7-19.

Morejohn, L.C., T.E. Bureau, J. Molè-Bajer, A.S. Bajer, and D.E. Fosket. 1987. Oryzalin, a

dinitroaniline herbicide, binds to plant tubulin and inhibits microtubule polymerization in

vitro. Planta 172:252-264

Norelli, J.N., A.L. Jones, and H.S. Aldwinke. 2003. Fire blight management in the twenty-first

century: using new technologies that enhance host resistance in apple. Plant Disease

87:756-765

Oh, C., and S.V. Beer. 2005. Molecular genetics of Erwinia amylovora involved in the

development of fire blight. FEMS Microbiology Letters 253:185-192

Oregon Dept. of Agr. (ODA). 2020. Oregon Agricultural Statistics & Directory. Accessed: 19

February 2020.

Piel, A., V.G.M. Bus, K. Geider, K. Richter, H. Flachowsky, and M.V. Hanke. 2009.

Improvement of Fire Blight Resistance in Apple and Pear. Int. J.Plant Breeding 3:1-27

Persiel, F. and W. Zeller. 1978. Differences in susceptibility to fireblight, Erwinia amylovora

(Burr.) Winslow et al., in non apomictal species, varieties and oecotypes of Cotoneaster.

Acta Hort. 86:45-50

Persiel, F. and W. Zeller. 1981. Some progress in breeding Cotoneaster for resistance to

fireblight, Erwinia amylovora (Burr.) Winslow et al. Acta Hort. 117:83-88

Persiel, F. and W. Zeller. 1990. Breeding upright growing types of Cotoneaster for resistance to

fire-blight, Erwinia amylovora (Burr.) Winslow et. al. Acta Hort. 273, 297-302 10

Richards, R.T. and S.J. Alexander. 2006. A social history of wild huckleberry harvesting in the

pacific northwest. U.S. Dept. Agr. General Tech. ReportPNW-GTR-657,

doi:10.2737/pnw-gtr-657.

Robertson, K.P., J.B Phipps, J. R Rohrer, and P.G Smith. 1991. A synopsis of genera in

Maloideae (Roseaceae). Systematic Bot. 16:376-394

Rounsaville, T.J. and T.G. Ranney. 2010. Ploidy levels and genome sizes of Berberis L. and

Mahonia Nutt. species, hybrids, and cultivars. HortScience 45:1029-1033

Shearer, K. and T.G Ranney. 2013. Ploidy levels and relative genome sizes of species, hybrids,

and cultivars of dogwood (Cornus spp.) HortScience 48:825-830

U.S. Dept. Agr. (USDA). 2016. Census of horticultural specialties for 2014. Accessed: 19

February 2020.

https://www.agcensus.usda.gov/Publications/2012/Online_Resources/Census_of_Horticu

lture_Specialties/

Vander Kloet, S.P. 1988. The genus Vaccinium in North America. Publication 1828, Research

Branch, Publication, Agr. Canada

Vanneste, J. L. 2000. Fire blight: The disease and its causative agent, Erwinia amylovora. CABI

Publishing. ISBN 0 85199 294 3 van Tuyl, J.M., B. Meijer, and M.P. van Diën. 1992. The use of oryzalin as an alternative for

colchicine in in-vitro chromosome doubling of Lilium and Nerine. Acta Hort. 325:625-

629.

Vogt, I., T. Wöhner, K. Richter, H. Flachowsky, G.W. Sundin, A. Wensing, E.A. Savory, K.

Geider, B. Day, M. Hanke, and A. Peil. 2013. Gene‐for‐gene relationship in the host–

pathogen system Malus×robusta 5–Erwinia amylovora. New Phytologist 197:1262-1275 11

Wöhner T.W., K. Richter, G.W. Sundin, Y. Zhao, V.O. Stockwell, J. Sellmann, H. Flachowsky,

M.V. Hanke, and A. Peil. 2018.Inoculation of Malus genotypes with a set of Erwinia

amylovora strains indicates a gene-for-gene relationship between the effector gene eop1

and both Malus floribunda 821 and Malus ‘Evereste’. Plant Pathol. 67:938-47

CHAPTER 2: ASSESSING THE IMPACT OF CHROMOSOME DOUBLING ON FRUIT AND PLANT QUALITIES OF VACCINIUM OVATUM

Kristin E. Neill1 and Ryan N. Contreras2

Department of Horticulture, 2017 Agriculture and Life Sciences Building, Oregon State

University, Corvallis, OR 97331-7304

1Graduate Research Assistant

2Associate Professor

CHAPTER 2: ASSESSING THE IMPACT OF CHROMOSOME DOUBLING ON FRUIT AND PLANT QUALITIES OF VACCINIUM OVATUM

Kristin E. Neill1 and Ryan N. Contreras2

Additional index words: Evergreen Huckleberry, Soluble Solids Content, Brix, Polyploidy

Abstract.

Vaccinium ovatum Pursh. (evergreen huckleberry) is an evergreen shrub native to the Pacific

Northwest. With the notable exception of cultivated blueberry (Vaccinium corymbosum) species in Vaccinium are primarily diploids (2n = 2x = 24) but polyploids have been reported from experiments with chemical mitotic inhibitors as well as naturally occurring tetraploids in cultivated blueberry (Vaccinium corymbosum). There is interest in this species, but it requires improvement of the fruit and plant qualities for an eventual cultivar release. To obtain variation in plant qualities polyploidy was induced in a collection of plants in 2013. The purpose of this study was to assess the impacts of polyploidy on the fruit and plant qualities of Vaccinium ovatum. This fruit and plant quality study provides a contribution to the scientific knowledge base that is currently lacking on evergreen huckleberries. Plant qualities were determined by measuring plant height and width, obtained in fall 2017. The fruit volume (mm3) and brix (°Bx), for soluble solids content, were measured using a digital caliper and a digital refractometer respectively. Measurements were taken on diploid, mixoploid, and tetraploid (2x, 2x + 4x, 4x) cytotypes, once in 2017, five times over nine weeks in 2018, and three times over nine weeks in

was smaller than that of diploid and mixoploid. Differences were observed in diploid fruit volume among all years (P < 0.0001). In tetraploids, brix was statistically significant across all years (P = 0.0002). The variation in data from the various years suggests that using ploidy as a way to produce larger sweeter fruit is not a plausible method.

Introduction

The genus Vaccinium includes blueberries, huckleberries, cranberries, lingonberries, whortleberries, bilberries and cowberries. The genus has two subgenera: Oxycoccus, which contains cranberries, and subgenus Vaccinium that has 21 sections. Evergreen huckleberry

(Vaccinium ovatum Pursh) falls into the section Pyxothamnus, while the section Cyanococcus contains many of the typical North American blueberries. There are about 450 species of

Vaccinium worldwide, but only around 40 in North America, and more specifically around 15 in the Pacific Northwest, where V. ovatum is native (Richards and Alexander, 2006; Vander Kloet,

1988). In the Eastern US, the common name “huckleberry” often refers to plants in the genus

Gaylussacia, which contains over fifty species, of which at least four species produce edible berries. In the Pacific Northwest, all huckleberries are members of the Vaccinium genus as they are all 5 chambered fruit as opposed to the 10 chambers in Gaylussacia (Nicholson, 2011;

Richards and Alexander, 2006). Traditionally huckleberries were collected on sites hidden and managed by Native American Tribes for eating and cultural use, but since 1948, many stands have begun to be commercially harvested, and shipped (Richards and Alexander, 2006). Coastal

Native Americans picked the berries in late fall and even during December, when all other fresh fruit had disappeared (Turner and Long, 1975). 15

Vaccinium ovatum is native to understories in the Pacific Northwest and found in abundance west of the Cascade Mountains from Canada (British Columbia) to the northern coast of California (Hitchcock & Cronquist, 1973; Postman, 2004; Vander Kloet, 1988). Evergreen huckleberries are often found growing in conjunction with red huckleberry (Vaccinium parvifolium) and common snowberry (Symphoricarpos albus). Although naturally a native woodland plant, evergreen huckleberries have become more commonly cultivated over the years and become a novelty food item in the Pacific Northwest. While still not a crop that is mass- produced, huckleberries remain a local product that has extensive history in the Pacific

Northwest.

Vaccinium ovatum is an erect evergreen shrub that varies in height from 0.5 to 3 m tall. It has leathery, alternately arranged ovate leaves, with serrate margins that are 2-5 cm long. The foliage is evergreen with new growth emerging in colors from bright pink to a burnt orange. The inflorescence emerges from the leaf axil in early spring through early summer, bearing bright white to pink urceolate flowers. The fruit ripens throughout summer into the fall, continuing to hang on the plant into late winter. Berries are rarely stripped from plants, despite being eaten by various bird species. The berries of V. ovatum tend to develop asynchronously and when ripe often persist for a month or longer (Vander Kloet, 1988). The berries hang on the shrub far longer into the fall and winter than blueberries, extending into early spring in landscape plantings, making this an ideal ornamental and edible species for landscape use. Like most other

Vaccinium, evergreen huckleberries thrive in either the sun or shade and require acidic soil

(Tamanda, 2002).

Harvesting of evergreen huckleberries is not as common as the red huckleberries

(Vaccinium parviflorum), black huckleberries (V. membranaceum), and blue leaf huckleberries 16

(V. deliciousum) that are increasingly harvested for production uses in the Pacific Northwest

(Richards and Alexander, 2006). However, for plant breeding purposes, evergreen huckleberries have the potential to become popular garden plants for their evergreen foliage and edible fruit.

For many years evergreen huckleberries have been popular among florists for use of the foliage in their bouquet’s designs (Schlosser et al., 1992; Vander Kloet, 1988). There are at least 6 cultivars, although they are not widely found at garden centers. Poor growth forms of this shrub that presumably originate from unselected seed sources, have been widely planted at Oregon

State University, Corvallis campus and other gardens. Observations of these plants with poor habit stimulated an impetus to create improved forms that fit better in cultivation than unselected native forms.

Polyploidy is a condition of having more than two sets of chromosomes that is common among plants and exists in over 80% of the world’s angiosperms. This phenomenon occurs naturally through unreduced gametes and through manipulation of the mitotic cycle using chemicals (Sleeper and Poehlman, 2006). Tetraploids contain four chromosome copies, whereas mixoploids, or cytochimeras (when occurring as periclinal chimeras), have different ploidy levels in different histogenic layers. Polyploid plants are selected and sought out by plant breeders to improve agricultural traits like organ sizes, blooming time, and pest resistance. In ornamentals crops where very little breeding has been done, making polyploids is often used to develop novel crops and increase the number of genetic variants (Manzoor et al., 2019). There is an increase in cell size associated with polyploidy; this effect is known as the “gigas effect,” which results in increasing organ size, including flowers and fruit, as well as altering plant shape

(Sattler et al., 2016). There are two common chemical methods used to induce polyploidy in ornamentals, oryzalin or colchicine. Oryzalin [3,5-dinitro-N4, N4-dipropylsulfanilamide] is the 17

active ingredient of the pre-emergence herbicide Surflan® (Southern, 1998) , which is a formulation commonly used to induce polyploids in plants. In many cases, inducing polyploidy results in sterility through disruption of the reproductive systems, or complications during meiosis (Herbert et al., 2010).

Though Vaccinium ovatum has undergone very little breeding, there has been enough to explain why polyploidy is important in the Genus Vaccinium. Blueberry breeders like James

Ballington have studied ploidy changes and crossing in the genus Vaccinium, while using V. ovatum as a parent. The North Carolina State University breeding program crossed ‘NC 2267’ (a diploid hybrid which is 1/4 V. corymbosum, 3/4 V. darrowi) with a selection of diploid V. ovatum. The seedling ‘NC 3048’, was the only one that grew from 275 pollinations and it is tetraploid. According to Ballington et al. (1997) this selection was at least partly fertile and was successfully backcrossed to tetraploid section Cyanococcus genotypes, proving successful as a male parent in producing seedlings (Ballington, 2001). Luby et al. (1991) produced a tetraploid

F1 between a tetraploid V. corymbosum, and V. ovatum in the 1970s. This tetraploid sectional hybrid was successfully backcrossed to both standard and southern highbush. Vaccinium ovatum is potentially useful as a parent for habitat adaptation and drought resistance (Ballington, 2001).

Viable but largely sterile diploid hybrids have been produced between V. ovatum and V. darrowi, and V. ovatum and V. crassifolium, and one partially fertile tetraploid hybrid between ‘Coville’

(V. corymbosum) and V. ovatum was produced through an unreduced gamete (Ballington et al.,

1986; Luby et al., 1991).

Brix (°Bx) is a measure of the mass ratio of soluble solids to water, is a widely used approximation for sugar content. Winemakers, vegetable processors, and many other members of the food industry use Brix to express the level of soluble solids content (SSC), which is sugars 18

plus all other dissolved solids as amount per unit of the solvent water (Coombe, 1991; Kleinhenz and Bumgarner, 2012). In many crops, SSC mainly measures sugars and acids making the refractometer a good predictor of sucrose content. Brix has been utilized in many studies including as a rapid postharvest method to determine soluble solids in sweet corn (Hale et al.,

2005) and grape sugar content (Song et al., 2015). It is important to note that sugar content or sweetness can be overwhelmed by other aspects of flavor (such as acidity), therefore a high Brix is not a guarantee of sweet flavor (Kleinhenz and Bumgarner, 2012). Jayasena and Cameron

(2008) used a °Brix/Acid ratio as a good prediction for consumer acceptability of grapes.

Berries contain a number of phenolic compounds including tannins, anthocyanins, and flavanols that all contribute to color and taste or flavor of fresh fruit (Seymour et al., 1993). Very little research has been done on the development of V. ovatum berries over the course of a season, though several studies have looked at anthocyanin development in several species of huckleberries (Ballington et al., 1988, Lee et al. 2004). Lee et al. (2004) determined that V. ovatum had greater total anthocyanin, higher pH and Brix than Vaccinium membranaceum.

Blueberry fruit development in two species was monitored at three different ripening stages to determine the effect of fruit maturation on the development of sugars and acid composition of the berries (Ayaz et al., 2001). In blueberries glucose and fructose, in approximately equal concentrations in the berry, were found to be the main sugars (Ayaz et al. 2001; Barker et al.,

1963; Hirvi and Hokanen, 1983; Kader et al., 1993; Kalt and McDonald, 1986). There have been several studies in both highbush and lowbush blueberries measuring various berry composition changes in different physiological stages of the berry (Ballinger and Kushman, 1970; Kader et al., 1993; Kalt et al. 1995). Ballinger and Kushman (1970) found that in general, percent soluble solids and berry weight both increased over berry development stages. Kalt et al. (1995) found 19

that sugar content correlated more strongly with berry surface color than with berry size. For a better sampling technique, methods from Hancock et al. (2000), including using hoop counts on shrubs, should be used in the future. Berries are generally measured over time because the different stages of berry fruit development are highly influenced by temperature (Darnell et al.,

1992). To have an accurate ripening date, heat units or degree-day accumulation are used in fruits to estimate developmental time (DeJong and Goudriaan, 1989; Godoy et al, 2007)

Though there are several blueberry studies using various physiological ages of berries and skin color as indices, no studies were found measuring sugar levels and volume in V. ovatum specifically, over a season. The overall goal of this study was to increase knowledge of

Vaccinium ovatum for breeding purposes, the aim of the present work was to analyze volume and brix of three cytotypes of containerized evergreen huckleberry over the course of a season to assess the impact of ploidy and harvest date.

Materials and Methods

Plant Material.

Tetraploids and mixoploids of normally diploid Vaccinium ovatum were developed by

Contreras and Friddle (unpublished data) in 2012-2013. Fruit was collected from shrubs on

Oregon State University campus in October of 2012, macerated in a blender with dulled blades and then spread to dry for 24 hours. Seeds were sown uniformly using a salt shaker, in 80% douglas fir bark, 20% peat growing media with a control release fertilizer (19-13-7.8) in trays.

Trays were watered, covered with a plastic dome to retain humidity and placed into cold stratification at 4 °C for 90 d. Trays were removed from stratification in February and placed into a glasshouse with day/night set temperatures of 24/18° C with a 16-hour daylength. Seedlings, 20

starting at the cotyledon stage, were sprayed daily for 20 d with 150 µm Oryzalin plus 0.1%

Triton X-100. Following treatment, seedlings were grown to a larger size and tested via flow cytometry during March 2014 to determine ploidy. All plants used in this study were maintained in 21 L containers filled with 100% unaged douglas fir bark (Lane Forest Products, Eugene, OR) under am unheated polyhouse with regular overhead irrigation watering.

Flow Cytometry

Flow cytometry was used to confirm ploidy level of study plants. We calculated holoploid (2C) genome size of individual accessions of Vaccinium ovatum by comparing mean relative fluorescence of the sample against an internal standard, Pisum sativum ‘Ctirad’, with a known genome size of 8.76 picograms (pg) (Greilhuber et al., 2007). A total of 57 accessions were evaluated using flow cytometric analysis of nuclei stained with 4',6'-diamidino-2- phenylindole (DAPI) (CyStain ultraviolet Precise P; Partec). For each sample, three young, fully expanded leaves were collected to represent a random sample of nuclei. Each sample was prepared by co-chopping 1to 2 cm2 of tissue from both V. ovatum and the internal standard with a double-sided razor blade in a polystyrene petri dish containing 400μL of nuclei extraction buffer (Cystain Ultraviolet Precise P Nuclei Extraction Buffer; Sysmex, Görlitz, Germany).

Buffer containing chopped leaf tissue was passed through a 50-μm gauze filter (Partec Celltrics,

Münster, Germany) into a 3.5 mL plastic tube (Sarstedt Ag & Co.; Nümbrecht, Germany). Next,

1.6 mL of DAPI stain was added to the nuclei suspension. All samples were analyzed using a

Partec PA II (Partec) flow cytometer. A minimum of 3000 nuclei were analyzed per sample with average coefficient of variation (CV) for each fluorescence histogram less than 10. Relative 2C genome size was calculated as:

mean fluorescence value of sample 2C = DNA content of standard × . mean fluorescence value of standard

Monoploid genome sizes were calculated by dividing each sample’s 2C genome size by inferred ploidy.

Plant Measurements.

Plant height and width were measured during fall of 2017 to the nearest centimeter. Fruit size was measured once during 2017 (August 31), biweekly over 9 weeks starting August 30 during 2018, and three times over 9 weeks starting September 11 during 2019. To compare volume and brix among years, the first week of testing during each year were compared.

Measurements taken in 2018 and 2019 over the course of 9 weeks were compared in a separate analysis. Two perpendicular diameters of three berries were measured using a steel digital caliper

(General Tools, UltraTech, Secaucus, NJ). These replicates were used to calculate mean berry volume. Using the mean of the two diameters, the following formula was used to get relative volume:

4 Volume of a sphere = πr2 . 3

In 2017 brix was taken only once using a combination of three berries, the juices were strained through a fine mesh filter into the sample well of a digital refractometer (Atago, Japan).

In both 2018 and 2019 sugar concentration was assessed using by squeezing the juice of three berries, individually through a fine mesh filter onto a digital refractometer. These replicates were used to calculate mean berry brix. Berry sampling was initiated when, through a visual inspection, over half of the fruit on the plant appeared to be at peak ripeness. Twenty-eight 22

accessions were tested in 2017 (2x, n=21; 2x+4x, n=4; 4x, n=3). Thirty-four accessions were tested in 2018 (2x, n= 24; 2x+4x, n=4; 4x, n=6). Twenty-five accessions were tested in 2019 (2x, n=17; 2x+4x, n=5; 4x, n=3). Due to fruiting differences across the years, not all accessions were tested each year. A particularly bad fruit set in 2019 limited the number of plants that were tested.

Statistical analysis.

All response variables were evaluated for normality (PROC UNIVARIATE, SAS 9.4).

Analysis of variance (ANOVA) was conducted using a general linear model (PROC GLM using

SAS Version 9.4, SAS Institute Inc., Cary, NC) to determine if ploidy had a significant impact on either volume or brix and means were separated using Tukey’s Honestly Significant

Difference (α=0.05), where appropriate. A t-test (PROC TTEST, SAS 9.4) was used to compare

2018 to 2019 at each ploidy and each collection time (weeks). Bar graph, scatterplot, regression equations, and multiple correlation coefficients (R2) were prepared in Excel (Microsoft Corp.,

Redmond, WA) to visualize relationships among independent and response variables. Code for all statistical models are included in the appendix.

Results

Genome Size and Plant Size.

Clear ploidy differences were revealed by flow cytometry data. Genome size of the diploid Vaccinium ovatum accessions in our collection ranged from 1.13 ± 0.04 to 1.29 ± 0.00.

Tetraploid accessions in our collection ranged from 2.39 ± 0.02 to 2.62 ± 0.04 (Table 2.1). All diploids (2x) and tetraploids (4x) had similar genome sizes and all mixoploids (2x+4x) had very distinct diploid and tetraploid peaks in the flow cytometer data. Peaks for mixoploids had close 23

to equal amounts of cells of each ploidy type (Fig. 2.1). Tetraploids were significantly shorter than both diploids and mixoploids in height (P < 0.0001). Diploids were significantly wider than mixoploids and tetraploids (P < 0.0001) (Table 2.2).

Fruit measurements on a single date.

Brix and volume were normally distributed in all three years. There were no consistent trends for brix across years for all ploidy levels. As data was only collected on one date, we could not assess change over time during 2017. In 2017, tetraploid fruit volume was larger than diploid and mixoploid with a minimum significant difference of 28.1 mm3 (P = <0.0001) (Table

2.3; Fig. 2.2). There was a difference of 55 mm3 between diploid and tetraploid fruit volume.

Volume and brix showed a trending increase over ploidy levels however for brix, ploidy levels were not significantly different from each other (P = 0.3431). Although not significantly different, brix increased from 15.3°Bx in diploids to 18.0°Bx in tetraploid plants.

On a single collection date of 8/31/18, volume (P = 0.3984) and brix (P = 0.1427) measurements were not different among ploidy levels. Although not statistically significant, volume did differ among the ploidy levels with tetraploid average volume being the smallest at

246.6 mm3 and mixoploid volume being the largest at 279.9 mm3. There were smaller variations in brix across ploidy levels with the biggest difference being between diploids with 15.5 °Bx and mixoploids with 14.1 °Bx (Table 2.3).

In 2019 fruit volume was not significant (P = 0.2237) across ploidy levels during the first week of testing. Tetraploids were smaller, when assessed based on biological significance, with a volume of 264.8 mm3, while mixoploids had a volume of 321.2 mm3 and diploids had a volume of 301.2 mm3. Brix measurements in 2019 were significant (P = 0.0163). Diploids and mixoploids had a very similar average of around 14.0 °Bx while tetraploids had significantly less 24

sugar content pf 11.0 °Bx. Brix exhibited a decreasing trend across ploidy levels for a single date measurement (Table 2.3).

Volume was significantly different across all years in diploid plants (P < 0.0001).

Volume of mixoploid plants in 2017 was significantly different from both 2018 and 2019 mixoploid plants (P < 0.0001), while volume of tetraploid plants in 2019 was significantly different from tetraploid plants in 2017 (P = 0.0045). Brix was only significantly different according to Tukey’s HSD across all years in tetraploid plants (P = 0.0002), however when testing with an overall ANOVA, brix was significantly different in diploids as well (P = 0.0173)

(Table 2.3, Fig. 2.3).

Fruit measurements in 2018 and 2019.

Across all ploidy levels brix shows an evident decrease over time in 2018 (Fig. 2.4).

Volume showed a trend of increase but then dipped mid-way through the season. Volume of diploids (P = 0.0391) and tetraploids (P = 0.0222) in 2018 showed significance between week 5 and week 7. Although statistically there was no difference in mixoploids due the large minimum significant difference, there is a clear biological difference between a fruit volume of 262.68 mm3 and 338.58 mm3. Brix values in 2018, unlike volume, showed a decrease over time within ploidy levels (Fig. 2.4). Although there was no statistical significance in the decrease in brix in diploids and tetraploids the decreasing trend is biologically significant. In mixoploids there was a significant reduction in brix (P = 0.0055) from 14.1 °Bx in week 1 to 11.0 °Bx in both week 7 and week 9 (Table 2.4).

Figure 2.3 shows no clear trend for brix over time however volume decreased across weeks consistently in all ploidy levels in 2019. Volume showed a significant decrease from week

1 to week 9 in both diploid (P = 0.0021) and tetraploid (P = 0.0135). Diploids showed a 25

significant difference in brix values between week 5 and week 9 in (P = 0.0371) (Table 2.4; Fig

2.3). Mixoploids showed an increase in brix from week 5 to week 9 but it was not significant.

Brix for all ploidy levels decreased in the fifth week of testing and then increased again in the ninth week of testing (Fig. 2.5).

There was a significant difference in diploid volume in weeks 1 (P = 0.0044), 5 (P =

0.0078), and 9 (P = 0.0252) between 2018 and 2019. Mixoploids in week 5 also displayed a significantly different volume between 2018 and 2019 (P = 0.0477). However, fruit size was not objectively larger in 2019 over 2018. In week 1 diploid fruit in 2019 had a significantly larger volume than that in 2018, but in week 9 fruit volume in 2019 is significantly smaller than in

2018. Diploid and tetraploid brix values were significantly different between 2018 and 2019 in both week 1 (2x P = 0.0079; 4x P < 0.0001) and week 5 (2x P =0.0258; 4x P = 0.0010). Brix values of mixoploids (P = 0.001) were significantly different between 2018 and 2019 in week 9

(Table 2.4).

In 2018, there was a significant difference in brix values in week 3 (P = 0.0053), 7 (P =

0.0022), and 9 (P = 0.0002) between mixoploids and diploids (Fig 2.4.). Diploids had higher degree brix than mixoploids and tetraploids across all weeks. Mixoploids had consistently larger volumes and lower brix across the weeks compared to the other ploidy levels (Table 2.4; Fig.

2.6). Meanwhile diploids had intermediate volumes and consistently the highest brix. Across the weeks diploids and tetraploids had relatively similar volumes. Similar to 2018, mixoploids in

2019 had a consistently higher volume across weeks, while tetraploids had the smallest volume and lowest degree brix (Fig. 2.7). There was a significant difference in brix values of diploids and tetraploids in week 9 (P = 0.0049). In both week 5 (P = 0.0028) and week 9 (P = 0.0048), tetraploids had significantly lower volumes than both diploids and mixoploids. Tetraploids had 26

significantly lower brix than both diploids and mixoploids all three weeks (Fig. 2.7). Mixoploids had the largest volume but diploids and tetraploids were not different across the weeks in both years, though diploids appeared slightly larger than tetraploids. Brix did not follow a similar pattern for mixoploids between years. Diploids and tetraploids brix followed a similar pattern where brix for diploids is just slightly higher than that of tetraploids over all the weeks. In 2018, there was a clear trend of increasing volume being associated with a lower brix (Fig. 2.8). In

2019 there is also a trend of brix decreasing with volume increasing (Fig. 2.9)

Discussion

The primary purpose of this study was to evaluate the fruit of the Pacific Northwest native species, Vaccinium ovatum over the course of a season and compare among ploidy levels.

In this study fruit volume and sugar content in the form of degree Brix (°Bx) were measured on fully ripe berries of three different ploidy levels. These data were then compared within and across years, and across a nine-week period in both 2018 and 2019. Overall, there were highly variable measurements across years, with some consistency within years. In 2017 on a single measurement, fruit volume increased significantly from diploid to tetraploid, while brix increased but not significantly. In 2018 on a single date there was not a clear significant decrease in volume from diploid to tetraploid and brix remained relatively stable. However, over the course of a season, there was a significant decrease in brix, with fluctuating volume in all ploidy types. In 2019, once again there was a difference in the way the data changed from both 2017 and 2018. On a single date there was a decrease in both volume and brix from diploid to tetraploid, only the trend for brix was significant. Over the course of a season in 2019, there was a clear trend of volume decreasing in all ploidy types over the 9 weeks. Mixoploids had the 27

largest volume of all cytotypes in both 2018 and 2019. Brix did not have a significant trend within ploidy in 2019, but there was a higher brix for all ploidy types in week 9 compared to week 1.

There are numerous examples where polyploidy has a “gigas” effect on fruit and plant qualities in increasing ploidy levels (Sattler et al., 2016; Wu et al., 2012). Plant height and width were statistically significant, tetraploids were found to be the shortest in height, and narrowest in width. This indicates that ploidy does affect the plant form qualities of Vaccinium ovatum. The data found in our study indicates that there are differences in the way that genome size affects fruit volume and sugar content. There was no consistent increase in fruit size from diploid to tetraploid. In 2017 measurements were taken on a single date in an early but unknown timepoint in the fruit maturity. These measurements were able to be compared to the first week of measurements from the following two years. The minimums and ranges provided show that there was much variability in measurements across accessions within ploidy levels. This should be considered when reviewing these numbers as variation among individual plants was high.

Overall tetraploids were significantly different in fruit size from both diploids and mixoploids, however nothing else was statistically significant in that year. Although the brix values were not statistically significant there is a biologically detectable difference in the tetraploid degree of

15.0 °Bx and both diploid and mixoploids with a degree below 11.0 °Bx. The results from 2017 indicated that higher ploidy increased fruit size and sugar content of Vaccinium ovatum. The data followed the “gigas effect” described by Manzoor et al. (2019). The fruit generally followed the effect that a doubled genome resulted in a larger fruit and subsequently one with a higher sugar content. 28

In 2018 and 2019, the data did not follow the same trend revealed by data from 2017.

Fruit volume did not show any trends in either 2018 or 2019. The “gigas effect” that was seen in

2017 did not appear in these data sets. Volume in diploids was statistically significant across years (P < 0.0001), and brix degree in tetraploids was significantly different across all years (P =

0.0002). In 2018 and 2019 fruit volume of tetraploid fruit was biologically smaller than that of diploid and mixoploid. In both years mixoploids were larger than the two other cytotypes, but the brix levels varied each year in their relation. Although we can say that there is some effect of genome size on fruit volume, it is not the normal “gigas” effect, with a higher ploidy levels resulting in higher values. These indicate that there were factors other than just genome size, affecting the fruit development.

Our second question asked, investigated development of the fruit size and sugars over the growing season. Biologically many brix values were very similar to each other, which resulted in a lack of statistical significance within ploidy across weeks. There was no clear trend for volume development over the season in 2018, though it is unclear why. There are no studies that show similar results in either grapes or blueberries. The only statistically significant volume measurements in 2018, were between week 5 and week 7, this could be inaccurate due to the lack of explanation for the dip in berry volume development in the fifth week of testing. In contrast to 2018, brix increased across the weeks during 2019. This indicates that ploidy level alone cannot be used to select plants. There were inconsistent results in measurements over years, indicating there may be more affecting these measurements than just genome size. Brix did not follow a similar pattern for mixoploids between years, but for diploids and tetraploids brix follows a similar pattern where brix for diploids is just slightly higher than that of tetraploids over all the weeks. Similar to 2018, mixoploids in 2019 had a consistently higher volume across 29

weeks, while tetraploids had the smallest volume and lowest degree brix (Fig 2.5). There was a significant difference in brix values of diploids and tetraploids in week 9.

For our study, knowing which accessions had the largest berries was of importance but it should be noted that is a poor indicator of ripeness in evergreen huckleberries. As displayed in several studies on highbush blueberry (Woodruff and Dewey, 1959) and lowbush blueberry

(Barker et al., 1963; Collins et al, 1966; Ismail and Kender, 1974) a better determinant of a berry’s ripeness could be color change rather than the diameter/volume measurement. If fruit had been tested from an earlier period through the later part of the season a typical double-sigmoid curve of growth would have been observed (Eck, 1988; Godoy et al., 2007; Tamanda, 2002).

This double sigmoid growth is fruit growth occurring in three different stages: stage 1 immature fruit that are growing rapidly after flowering, stage 2 fruit grow very little, and stage 3 a period of rapid growth where fruit hit maturity and maximum growth. This double sigmoid growth pattern could explain some of the variability in our data, as we did not test early enough to see the stage 1 phase, samples could have been taken in different stages of fruit growth between the years.

In several previous studies in cherries, kiwifruit, and peaches, a high brix has been connected with consumer acceptability (Crisosto et al., 2003; Gorini and Lasorella, 1990;

Robertson et al., 1988). Brix values alone are not the best indicator of consumer acceptability and preferences, Jayasena and Cameron (2008) found that acids are important in human perceptions of sweetness. They determined that a °Brix/acid ratio was highly associated with consumer acceptability of ‘Crimson’ seedless table grapes. Berries in this study were not evaluated at immature, mid-ripe, and ripe like other papers (Ayaz et al, 2001). Berries were also being evaluated to determine correct time of ripeness for home consumers and, since this 30

timeline did not need to be exact, rather a general guide to ripening would suffice. As determined from other sources and our own data inconsistencies we now know that heat unit or degree-day accumulation should be used to measure the developmental time of the fruits and determine the optimum picking time for home gardeners (Godoy et al., 2007). Fruit developmental time can also be measured using the method described by Hancock et al. (2000). At the beginning of the study and at peak harvest, fruit maturity was estimated using a hoop-count following the technique described by counting the number of mature fruit as a proportion of total fruit number.

The results of this study have added greatly to the wealth of knowledge of Vaccinium ovatum. The information gathered over the last three years will be valuable when choosing how to best evaluate evergreen huckleberry for potential cultivar releases. One of the goals of the

Ornamental Plant Breeding program, for V. ovatum was to produce a shrub that had a nice compact form, with large, sweet edible fruit. With the data collected the program will need to further evaluate accession for potential cultivar release on more than just ploidy and fruit measurements. There are several studies that provide exceptional methods for further testing of these shrubs, one of them being testing for a °Brix/acid ratio, if a sensory panel is not used

(Jayasena and Cameron, 2008).

These results are limited by our sampling technique and small number of plants tested.

Destructive sampling was a huge limitation that contributed to our limited sampling number.

Another limit that will be hard to overcome is in the determination of the brix value in single fruits, which showed a rather high standard error (S.E.) These limitations could be mitigated in the future through different sampling techniques such as near-infrared spectroscopy (NIRS), high performance liquid chromatography (HPLC), or a nondestructive technique that allows for repeated measures as described by Coombe (1991) (Godoy et al., 2007; Ventura et al., 1998). 31

Along with using different methods for testing, Jayasena and Cameron (2008) proposed a method for testing over a season in grapes that should be used for future evaluation of accessions in the Ornamental Plant Breeding Program.

Literature Cited

Ayaz, F.A., A. Kadioglu, E. Bertoft, C. Acar, and I. Turna. 2001. Effect of fruit maturation on

sugar and organic acid composition in two blueberries (Vaccinium arctostaphylos and V.

myrtillus) native to Turkey. N.Z. J. Crop Hort. Sci. 29:137-141.

Ballinger, W. E.; and L.J. Kushman. 1970: Relationship of stage of ripeness to composition and

keeping quality of highbush blueberries. J. Amer. Soc. Hort. Sci. 95:239-242.

Ballington, J.R., W.B., Kirkman, W.E. Ballinger, and E.P. Maness. 1988. Anthocyanin,

aglycone, and aglycone-sugar content in the fruits of temperate North American species

of four sections in Vaccinium. J. Am Soc Hort Sci. 113:746-749.

Barker, W.G., F.A. Wood, and W.B Collins. 1963. Sugar levels in fruits of lowbush blueberries

at four physiological ages. Nature. 198:810-811

Collins, W. B., K. H. Irving and W. G. Barker. 1966. Growth substances in the flower bud and

developing fruit of Vaccinium angustifolium Ait. Proc. Amer. Soc. Hort. Sci. 89:243-247

Coombe, B.G. 1991. Research on development and ripening of the grape berry. Am. J. Enol.

Vitic. 43:101-110

Crisosto, C.H., G.M. Crisosto, and P. Metheney. 2003. Consumer acceptance of ‘Brooks’ and

‘Bing’ cherries is mainly dependent on fruit SSC and visual skin color. Postharvest Biol.

Technol. 28:159-167

Darnell, R.L., G.W. Stutte, G.C. Martin, G.A. Lang, and J.D. Early. 1992. Developmental

physiology of rabbiteye blueberry. Hort. Rev. 13:339-405

DeJong, T.M., and J. Goudriaan. 1989. Modelling peach fruit growth and carbohydrate

requirements: reevaluation of the double sigmoid growth pattern. J. Am. Soc. Hort. Sci.

114: 1037-1040 33

Godoy, C., G. Monterubbianesi, and J. Tognetti. 2007. Analysis of highbush blueberry

(Vaccinium corymbosum L.) fruit growth with exponential mixed models. Scientia

horticulturae. 115:368-376

Gorini, F. and M. Lasorella. 1990. Sensory and objective evaluation of kiwi fruit. Acta Hort.

282:309-314

Greilhuber, J., E.M. Temsch, and J.M. Loureiro. 2007. Nuclear DNA content measurement, p.

67–101. In: J. Doleẑel, J., J. Greilhuber, and J. Suda (eds.). Flow cytometry with plant

cells: Analysis of genes, chromosomes and genomes. Wiley-VCH, Weinheim, Germany.

Hale, T.A., R.L. Hassell, and T. Phillips. 2005. Refractometer measurements of soluble solid

concentration do not reliably predict sugar content in sweet corn. HortTechnology.

15:668-672.

Hancock, J. F., P. Callow, R. Keelser, D. Prince, and B. Bordelon. 2000. A crop estimation

technique for highbush blueberries. J. Am. Pomol. Soc.. 54:123-129

Hirvi, T. and E. Honkanen. 1983: The aroma of some hybrids between highbush blueberry

(Vaccinium corymbus L.) and bogblueberry (Vaccinium uliginosum L.). Lebensmittel

Untersuchung Forschung.176:346-349.

Hitchcock, C.L. and A. Cronquist. 1973. Flora of the Pacific Northwest. University of

Washington Press. Seattle.

Ismail, A.A., and W.J. Kender. 1974. Physical and chemical changes associated with the

development of the lowbush blueberry fruit Vaccinium angustifolium Ait. Life Sciences

and Agr. Exp. Sta. Tech. Bul. 70.

Jayasena, V. and I. Cameron. 2008. Brix/acid ratio as a predictor of consumer acceptability of

crimson seedless table grapes. J.Food Quality 31:736-750 34

Kader, F., B. Rovel, and M. Metche. 1993. Role of invertase in sugar content in highbush

blueberries (Vaccinium corymbosum, L.). Food Sci. and Technol. 26:593-595.

Kalt, W. and J.E. McDonald. 1996: Chemical composition of lowbush blueberry cultivars. J.

Amer. Soc. Hort. Sci. 121:142- 146.

Kalt, W.; K. B. McRae, and L.C. Hamilton,. 1995: Relationship between surface color and other

maturity indicators in wild lowbush blueberries. Canadian Journal of Plant Science

75:485-490.

Kleinhenz, M.D. and N.R. Bumgarner. 2012. Using °brix as an indicator of vegetable quality: an

overview of the practice. Fact Sheet: Agriculture and Natural Resources: The Ohio State

University 1-4.

Lee, J., C.E. Finn, and R.E. Wrolstad. 2004. Comparison of anthocyanin pigment and other

compounds of Vaccinium membranaceum and Vaccinium ovatum Native to the Pacific

Northwest of North America. Am. Chem. Soc. 52:7039-7044.

Manzoor, A., T. Ahmad, M.A. Bashir, I.A. Hafiz, and C. Silvestri. 2019. Studies on colchicine

induced chromosome doubling for enhancement of quality traits in ornamental plants.

Plants 8:194

Nicholson, Rob. 2011. Little big plant, box huckleberry (Gaylussacia brachycera). Arnoldia:

The Magazine of the Arnold Arboretum. 11-18.

Postman, Joseph. 2004. An Evergreen huckleberry industry near the Oregon coast early in the

20th Century. J. Am. Pomol. Soc.58:147-151.

Robertson, N.A., F.I. Meredith, and R. Scorza. 1988. Characteristics of fruit from high and low

quality peach cultivars. HortScience. 23: 1032-1034 35

Richards, R.T. and S.J. Alexander. 2006. “A Social history of wild huckleberry harvesting in the

Pacific Northwest.” U.S. Dept. Agr. General Technical Report PNW-GTR-657,

doi:10.2737/pnw-gtr-657.

SAS Institute.2013. SAS.STAT Software: Release 9.4. SAS Institute Inc., N.C., USA

Sattler, M.C., Carvalho, C.R., Clarinso, W.R. 2016. The polyploidy and its key role in plant

breeding. Planta. 243:281-296

Schlosser, W.E., K.A. Blatmer, and B. Zamora. 1992. Pacific northwest forest lands potential for

floral greenery production. Northwest Sci. 66:44-55

Seymour, G.B., J. Taylor, and A. Tucker. 1993. Biochemistry of fruit ripening. Chapman and

Hall, London.

Sleeper D.A. and J.M. Poehlman. 2006. Breeding Field Crops. Blackwell Publishing, Ames,

Iowa.

Song, J., R. Smart, H. Wang, B. Dambergs, A. Sparrow, and C.Q. Qian. 2015. Effect of grape

bunch sunlight exposure and UV radiation on phenolics and volatile composition of Vitis

vinifera L. cv. Pinot noir wine. Food Chemistry 173:424-431

Tamanda, T. 2002. Stages of rabbiteye and highbush blueberry fruit development and the

associated changes in mineral elements. Acta Hort. 574:129-137

Turner, J. and J.N. Long. 1975. Accumulation of organic matter in a series of douglas-fir stands.

Long Canadian J. Forest Res. 5:681-690

Vander Kloet, S.P. 1988. The genus Vaccinium in North America. Publication 1828, Research

Branch, Publication, Agr. Canada. 36

Ventura M., A. de Jager, H. de Putter, and F.P.M.M. Roelofs.1998. Non-destructive

determination of soluble solids in apple fruit by near infrared spectroscopy (NIRS).

Postharvest Biol. and Technol. 14:21-27

Woodruff, R. E. and D. H. Dewey. 1959. A possible harvest index for Jersey blueberries based

on the sugar and acid content of the fruit; Mich. Agr. Exp. Sta. Quart Bui. 42:340-349

Wu, J., A.R. Ferguson, B.G. Murray, Y. Jia,, P.M. Datson and J. Zhang. 2012. Induced

polyploidy dramatically increases the size and alters the shape of fruit in Actinidia

chinensis, Ann. of Bot. 109:169-179

Figures A a a. Vaccinium ovatum 2n=2x b. Pisum sativum ‘Ctirad’

B a a. Vaccinium ovatum 2n=4x b. Pisum sativum ‘Ctirad’

C a a. Vaccinium ovatum 2n=2x b b. Vaccinium ovatum 2n=4x c. Pisum sativum ‘Ctirad’

Volume 250 25 y = 1.3333x + 14.247 Brix R² = 0.8827 200 20

Linear ) 3 (Brix)

150 15

Bx) °

100 10

Brix ( Brix Volume (mm Volume

50 5

0 0 2x 2x+4x 4x

Volume Brix

2x 2017 2x a 2018 20 300 b a a

) 2019 a 3 15

200 Bx) c ° 10

100 ( Brix

Volume (mm Volume 5

0 0 2017 2018 2019 2017 2018 2019

2x+4x a 2x+4x a 20 a 300 a a 15

b Bx)

200 ° 10

100 ( Brix

5 Volume (mm3) Volume

0 0 2017 2018 2019 2017 2018 2019 4x 4x 20 a 300 a ab b b 15

200 Bx) c ° 10

100 ( Brix 5 Volume (mm3) Volume

0 0 2017 2018 2019 2017 2018 2019 Fig. 2.3. Fruit volume (mm3) and brix (°Bx) measurements taken on a single date of collection from three different cytotypes of Vaccinium ovatum in each year 2017, 2018, and 2019. Mean separation done by Tukey’s (HSD) (α = 0.05); Comparisons across years within ploidy, Means followed by same letter were not statistically different based on: (2x) Volume minimum significant difference of 23.7 mm3; Brix minimum significant difference of 1.6 °Bx. (2x+4x) Volume minimum significant difference of 62.8 mm3; Brix minimum significant difference of 3.8 °Bx. (4x) Volume minimum significant difference of 42.3 mm3 ; Brix minimum significant difference of 3.4 °Bx

A 0 2 4 6 8 10 2x Volume 400 y = -0.1208x + 15.502 16

R² = 0.83 15 Brix )

3 300

14 Linear (Brix ) Bx) 200 13 °

12 ( Brix

Volume (mm Volume 100 11 0 10 1 3 5 7 9 Weeks

B 0 2 4 2x+4x 6 8 10 400 16 y = -0.365x + 13.86 15 ) R² = 0.8074 3 300

14 Bx) 200 13 ° 12 100 Brix ( Volume (mm Volume 11 0 10 1 3 5 7 9 Weeks

C 0 2 4 4x 6 8 10 400 16 y = -0.2428x + 14.567

R² = 0.8175 15 ) 3 300

14 Bx) 200 13 °

12 ( Brix

Volume (mm Volume 100 11 0 10 1 3 5 7 9 Weeks

A -1 1 3 5 7 9 11 avg vol 400 2x 16 15 avg brix

) 300 3 14

13 Bx) 200 °

12 Brix ( Brix

Volume (mm Volume 100 11 10 0 9 1 5 9 B Weeks -1 1 3 2x+45 x 7 9 11 400 16

15 )

3 300 14 Bx)

13 ° 200 12

11 Brix ( Volume (mm Volume 100 10 0 9 1 5 9 Weeks -1 1 3 5 7 9 11 C 400 4x 16 15

) 300 3 14

13 Bx) 200 ° 12 11 ( Brix Volume (mm Volume 100 10 0 9 1 5 9 Weeks Fig. 2.5. 2019 Fruit volume (mm3) and brix (Bx) measurements collected over 9 weeks from three different cytotypes of Vaccinium ovatum. A. Measurements from diploid accessions (2x, n=17) B. Measurements from mixoploid accessions (2x+4x, n=5). C. Measurements from tetraploid accessions (4x, n=3). 42

avg volume avg brix 400 18 Aa Ab Ab 350 Aa ABa Aa Aa Aa 16 Aa Bab Ab Aa Aab 14 300 Aa Aab

) 12 3 250

10 Bx) 200 ° 8

150 ( Brix Volume (mm Volume 6 100 4 50 2 0 0

Week 1 Week 3 Week 5 Week 7 Week 9

All within week analysis done by Tukey’s (HSD) (α = 0.05); Capital letters indicate significance between volume, lowercase letters indicate significance between brix: Week 1: Volume minimum significant difference of 50.5 mm3; Brix minimum significant difference of 2.1 °Bx Week 3: Volume minimum significant difference of 56.9 mm3; Brix minimum significant difference of 2.5 °Bx Week 5: Volume minimum significant difference of 50.9 mm3 ; Brix minimum significant difference of 2.8 °Bx Week 7: Volume minimum significant difference of 67.1 mm3 ; Brix minimum significant difference of 2.7 °Bx Week 9: Volume minimum significant difference of 60.1 mm3; Brix minimum significant difference of 2.3 °Bx

avg vol avg brix 400 Aa Aa ABa 16 Aa 350 Aa Aa 14 300 Ab Bb

Bb 12 ) 3 250

10 Bx) 200 ° 8

150 ( Brix

Volume (mm Volume 6

100 4

50 2

0 0

Week 1 Week 5 Week 9

All within week analysis done by Tukey’s (HSD) (α = 0.05); Capital letters indicate significance between volume within week, lowercase letters indicate significance between brix within week: Week 1: Volume minimum significant difference of 66.6 mm3; Brix minimum significant difference of 2.5 °Bx Week 5: Volume minimum significant difference of 58.1 mm3; Brix minimum significant difference of 2.0 °Bx Week 9: Volume minimum significant difference of 54.2 mm3; Brix minimum significant difference of 2.2 °Bx

Fig. 2.8. Scatterplot of all 2018 fruit volume (mm3) and brix (Bx) measurements taken over 9 weeks in Vaccinium ovatum. Legend: 2x- blue circle; 2x+4x red triangle; 4x yellow square.

Fig. 2.9. Scatterplot of all 2019 fruit volume (mm3) and brix (Bx) measurements taken over 9 weeks in Vaccinium ovatum. Legend: 2x- blue circle; 2x+4x red triangle; 4x yellow square.

Tables Table 2.1. Vaccinium ovatum accessions, genome size analyzed by DAPI flow cytometry, plant height, and width measured on plants grown in 21 L containers at Lewis Brown Research farm in Corvallis, OR. Accession Ploidy Genome Size Height Width 2C (pg) ±SEM (cm) (cm) 12-0018-002 2x 1.22 ± 0.01 76.2 66.0 12-0018-003 2x 1.21 ± 0.01 53.3 55.9 12-0018-005 2x 1.25 ± 0.01 33.0 66.0 12-0018-007 2x 1.18 ± 0.01 40.6 76.2 12-0018-011 2x 1.25 ± 0.01 30.5 55.9 12-0018-012 2x 1.22 ± 0.01 33.0 50.8 12-0018-015 4x 2.40 ± 0.03 30.5 30.5 12-0018-020 2x 1.23 ± 0.00 50.8 45.7 12-0018-021 2x 1.23 ± 0.02 35.6 43.2 12-0018-023 4x 2.48 ± 0.01 15.2 35.6 12-0018-026 2x + 4x 1.22 ± 0.01 | 2.49 ± 0.03 33.0 30.5 12-0018-027 2x + 4x 1.25 ± 0.05 | 2.52 ± 0.08 12.7 35.6 12-0018-028 2x 1.19 ± 0.04 33.0 55.9 12-0018-030 4x 2.59 ± 0.04 17.8 35.6 12-0018-031 2x + 4x 1.26 ± 0.01 | 2.59 ± 0.02 25.4 38.1 12-0018-035 2x + 4x 1.26 ± 0.00 | 2.56 ± 0.02 60.9 45.7 12-0018-040 2x 1.22 ± 0.00 35.6 38.1 12-0018-052 2x + 4x 1.13 ± 0.01 | 2.28 ± 0.03 43.2 63.5 12-0018-058 2x + 4x 1.26 ± 0.02 | 2.60 ± 0.03 38.1 25.4 12-0018-061 2x + 4x 1.21 ± 0.00 | 2.46 ± 0.01 30.5 35.6 12-0018-063 4x 2.53 ± 0.02 15.2 48.3 12-0018-064 4x 2.52 ± 0.02 38.1 35.6 12-0018-065 4x 2.48 ± 0.04 22.9 22.9 12-0018-066 4x 2.47 ± 0.03 38.1 53.3 12-0018-070 2x 1.25 ± 0.01 60.9 40.6 12-0018-072 2x 1.20 ± 0.05 53.3 38.1 12-0018-073 4x 2.56 ± 0.03 25.4 27.9 12-0018-078 4x 2.55 ± 0.06 22.9 30.5 12-0018-084 4x 2.58 ± 0.03 33.0 33.0 12-0018-088 2x 1.25 ± 0.00 22.9 45.7 12-0018-089 4x 2.51 ± 0.04 25.4 40.6 12-0018-095 4x 2.51 ± 0.02 38.1 45.7 12-0018-096 4x 2.51 ± 0.02 30.5 30.5 12-0018-098 2x 1.25 ± 0.01 55.9 30.5 12-0018-100 2x 1.24 ± 0.00 55.5 53.3 12-0018-101 4x 2.51 ± 0.04 30.5 25.4 12-0018-107 2x 1.29 ± 0.00 50.8 55.9 12-0018-108 4x 2.62 ± 0.04 27.9 40.6 12-0018-112 4x 2.39 ± 0.02 12.7 20.3 47

Table 2.1 (continued). Vaccinium ovatum accessions, genome size analyzed by DAPI flow cytometry, plant height, and width measured on plants grown in 21 L containers. Accession Ploidy Genome Size Height Width 2C (pg) ±SEM (cm) (cm) 12-0018-121 2x 1.14 ± 0.03 38.1 55.9 12-0018-124 4x 2.52 ± 0.03 27.9 40.6 12-0018-128A 2x 1.21 ± 0.03 40.6 53.3 12-0018-128B 4x 2.54 ± 0.03 22.9 40.6 12-0018-130 4x 2.48 ± 0.02 33.0 60.9 12-0018-137 4x 2.57 ± 0.02 27.9 38.1 12-0018-144 4x 2.40 ± 0.02 25.4 33.0 12-0018-145 2x 1.20 ± 0.01 45.7 43.2 12-0018-147 4x 2.44 ± 0.02 20.3 35.6 12-0018-148 4x 2.52 ± 0.03 25.4 27.9 12-0018-159 4x 2.48 ± 0.02 38.1 38.1 12-0018-163 2x 1.21 ± 0.02 55.9 55.9 12-0018-165 2x 1.23 ± 0.02 45.7 60.9 12-0018-171 2x 1.22 ± 0.01 48.3 66.0 12-0018-173 2x 1.21 ± 0.02 35.6 63.5 12-0018-176 2x 1.25 ± 0.01 55.9 58.4 12-0018-177 2x + 4x 1.23 ± 0.03 | 2.57 ± 0.03 76.2 50.8 12-0018-178 2x 1.13 ± 0.04 22.9 43.2 Base chromosome number is x = 12 Ploidy inferred from 2C genome size divided by mean 1Cx genome size (0.5) Mean ± SEM; means based on average of taxa means, three samples of the accession were used to calculate the mean | splits the 2x and 4x values

Table 2.2. Mean height and width measurements from three different cytotypes of Vaccinium ovatum. Ploidy Height (cm) Width (cm) 2x 44.4 a 52.7 a 2x+4x 40.0 a 40.6 b 4x 26.9 b 36.3 b Means within columns followed by different letters are significant based on Tukey’s HSD (α = 0.05). Diploid, n = 25 accessions. Mixoploid, n = 8 accessions. Tetraploid, n = 24 accessions. zMinimum sig diff 10.9 cm for height yMinimum sig diff 9.3cm for width

Table 2.3. Fruit volume (mm3) and brix (°Bx) measurements taken on a single date of collection from three different cytotypes of Vaccinium ovatum in each year 2017, 2018, and 2019. Year Ploidy Volume (mm3) Min Range Brix (°Bx) Min Range

2x 148.5 Bc 68.0 190.0 15.3 Aa 10.4 8.4

2017z 2x+4x 159.3 Bb 72.74 168.9 17.5 Aa 10.9 9.0

4x 203.5 Ab 144.4 126.1 18.0 Aa 15.0 9.1

2x 262.9 Ab 121.1 369.9 15.5 Aa 8.0 15.1

2018y 2x+4x 279.9 Aa 197.4 287.7 14.1 Aa 12.1 5.3

4x 246.6 Aab 176.4 135.4 14.5 Ab 11.7 8.4

2x 301.2 Aa 155.9 301.2 14.0 Aa 7.9 10.6

2019x 2x+4x 321.2 Aa 176.1 318.0 14.3 Aa 9.2 12

4x 264.8 Aa 174.2 189.8 11.0 Bc 9.3 2.7 All within year analysis done by Tukey’s (HSD) (α = 0.05); Comparisons within year across ploidy, Means followed by same uppercase letter within columns were not statistically different based on: zVolume minimum significant difference of 28.1 mm3; Brix minimum significant difference of 4.8 °Bx yVolume minimum significant difference of 50.4 mm3; Brix minimum significant difference of 2.1 °Bx xVolume minimum significant difference of 66.6 mm3; Brix minimum significant difference of 2.5 °Bx

All across year analysis done by Tukey’s (HSD) (α = 0.05); Comparisons across years within ploidy, Means followed by same lowercase letter within columns across years were not statistically different based on: (2x) Volume minimum significant difference of 23.7 mm3; Brix minimum significant difference of 1.6 °Bx (2x+4x) Volume minimum significant difference of 62.8 mm3; Brix minimum significant difference of 3.8 °Bx (4x) Volume minimum significant difference of 42.3 mm3; Brix minimum significant difference of 3.4 °Bx

Ploidy Week 1 Week 3 Week 5 Week 7 Week 9

Volume (mm3) 2x 262.9 ab 273.3 ab 254.3 b 290.9 a 283.3 ab 2x+4x 279.9 a 332.4 a 262.7 a 338.6 a 308.1 a 4x 246.6 ab 264.1 ab 215.8 b 286.0 a 254.8 ab 2018z Brix (°Bx) 2x 15.5 a 15.1 a 14.7 a 14.5 a 14.6 a 2x+4x 14.1 a 12.0 ab 12.1 ab 11.0 b 11.0 b 4x 14.5 a 14.0 a 13.0 a 12.5 a 12.8 a Volume (mm3) 2x 301.2 a** - 289.9 a* - 252.5 b* 2x+4x 321.3 a - 307.6 a* - 289.9 a 4x 264.8 a - 211.8 ab - 200.9 b** 2019y Brix (°Bx) 2x 14.0 ab* - 13.4 b* - 14.7 a 2x+4x 14.3 a - 13.2 a - 14.8 a*** 4x 11.0 a*** - 10.4 a*** - 11.7 a 2018 both week and ploidy were significant (P < 0.01) in comparison to volume and brix. All within year analysis done by Tukey’s (HSD) (α = 0.05); means followed by same lowercase letter within rows were not statistically different based on comparisons within ploidy across week: z2x Volume minimum significant difference of 36.1 mm3; 2x Brix minimum significant difference of 1.6 °Bx; 2x+4x Volume minimum significant difference of 80.0 mm3; 2x+4x Brix minimum significant difference of 2.5 °Bx; 4x Volume minimum significant difference of 58.3 mm3 ; 4x Brix minimum significant difference of 2.4 °Bx y2x Volume minimum significant difference of 33.7 mm3; 2x Brix minimum significant difference of 1.2 °Bx ; 2x+4x Volume minimum significant difference of 61.6 mm3; 2x+4x Brix minimum significant difference of 2.8 °Bx; 4x Volume minimum significant difference of 53.1 mm3; 4x Brix minimum significant difference of 1.4 °Bx. The * in 2019 indicate significance with the corresponding value in 2018.Significance within column comparing within ploidy within week across years -* Significant at P < 0.05; ** significant at P < 0.005; *** significant at P < 0.001.

CHAPTER 3. SCREENING COTONEASTER SPP. FOR RESISTANCE TO FIRE BLIGHT USING FOLIAR INOCULATION WITH TWO STRAINS OF ERWINIA AMYLOVORA

Kristin E. Neill1, Ryan N. Contreras2, and Virginia O. Stockwell3

1,2Department of Horticulture, 2017 Agriculture and Life Sciences Building, Oregon State

University, Corvallis, OR 97331-7304; 3United States Department of Agriculture, Agricultural

Research Service, Horticultural Crops Research Unit, Corvallis, OR 97330

1Graduate Research Assistant

2Associate Professor

3Research Plant Pathologist

CHAPTER 3. SCREENING COTONEASTER SPP. FOR RESISTANCE TO FIRE BLIGHT USING FOLIAR INOCULATION WITH TWO STRAINS OF ERWINIA AMYLOVORA

Kristin E. Neill1, Ryan N. Contreras2, and Virginia O. Stockwell3

Additional index words: Necrosis, Rosaceae, Maloideae, Disease Resistance, AvrRpt2 Mutant

Abstract. The genus Cotoneaster Medik. is composed of around 400 species with a wide variety of growth habits and form. These hardy landscape shrubs use to be commonplace because of their low maintenance and landscape functionality. However, the interest in and sales of Cotoneaster have declined for a variety of reasons, the major reason is because of its susceptibility to the bacterial disease called fire blight, which is caused by Erwinia amylovora. The resistance of 15 different genotypes of Cotoneaster was tested by inoculating leaves with a wild type strain of E. amylovora (Ea153) and an avrRpt2 mutant of Erwinia amylovora (LA635). Four studies took place in climate-controlled growth chambers and one study took place in a greenhouse in

Corvallis, OR. Fire blight resistance was assessed by calculating the percent shoot necrosis (PSN

= 100*(lesion length/total branch length)) once a week for six to eight weeks after inoculation.

Across all studies, genotypes H2011-01-002 and H2011-02-001 consistently had the lowest levels of percent shoot necrosis. Plants inoculated with different isolates were directly compared from growth chamber studies done in 2019. Genotype H2011-02-005 was significantly more resistant to EA153 than to LA635 and C. splendens was significantly less resistant to EA153 than to LA635. Genotype H2011-02-001 has already been released as a new ornamental cultivar that has high fire blight resistance. We identified other genotypes with fire blight resistance, which would be suitable for areas where the disease is especially damaging. 53

Introduction

Cotoneaster Medik. is a genus of hardy, ornamental shrubs in the family Rosaceae sub family Maloideae. The genus Cotoneaster originated in the Himalayas and China and is composed of approximately 400 species that are widely distributed throughout the Northern

Hemisphere (Dickore and Kasperk, 2010; Fryer and Hylmö, 2009). Cotoneaster is divided into 2 subgenera, Chaenopetalum which flowers all at once with white spreading petals, whereas subgenus Cotoneaster which flowers successively over a long period of time with cuplike pink flowers (Fryer and Hylmö, 2009). The subgenera are further divided into series based on botanical characteristics and geographic origins (Fryer and Hylmö, 2009).

Cotoneaster is an ideal urban plant due to its ability to tolerate pollution, grow in poor soils, and withstand harsh pruning. The shrub has multi-season interest in the form of autumn color or a dense, consistent green foliage, and berries that persist through the winter. The size of leaves, their color and texture ranges greatly among species and varieties. The fruit are mostly bright orange-red, but yellow and black-fruited species exist. The berries attract wildlife including birds like robins and waxwings, which have aided in the spread of Cotoneaster from cultivation into the wild. Cotoneaster are easy to grow and are commonly used as foundation plants, hedges, in parking lots, or in a mass planting along roadsides. Cotoneaster are attractive and utilitarian for landscaping, but most available cultivars have a severe shortcoming; that is, they are susceptible to the bacterial plant disease called fire blight caused by Erwinia amylovora.

Bacterial cells that overwinter in stem cankers likely serve as primary inoculum in the spring (Khan et al., 2011). Stem cankers produce bacterial ooze that is then disseminated by birds, insects, wind, and rain to flowers or wounds (Vogt et al., 2013; Wöhner et al., 2018).

Subsequently, the bacteria are spread from colonized flowers to newly opened flowers by 54

pollinating insects. After tissue invasion, the pathogen produces effector proteins that cause necrosis and can eventually kill the plant. The blighted shoot tips take on the appearance of a necrotic shepherd’s crook, which is a noticeable symptom of fire blight (Oh and Beer, 2005).

Fire blight periodically causes heavy losses in the Northwest for apple (Malus x domestica Borkh.) and pear (Pyrus communis L.) production. With losses involved in control methods, in the United States of around US$100 million and upward of US$9 million in

Switzerland, it is one of the most economically devastating diseases in pome fruit production worldwide (Norelli et al., 2003; Vogt et al., 2013). In 1998, a fire blight outbreak in Washington and northern Oregon resulted in losses of over $68 million US to pear and apple growers

(Timothy Smith, Washington State University). Fire blight also infects ornamentals, like

Cotoneaster, causing a massive decrease in sales of the shrub due to disease pressure in the landscape and loss of interest by consumers. Copper and streptomycin are the only registered chemicals for fire blight control on Cotoneaster (Vanneste, 2003). In some regions, streptomycin is no longer effective due to resistance of the pathogen to the chemical and in many countries, streptomycin is not permitted for fire blight control on Cotoneaster (McGhee et al., 2011; Peil et al., 2019; Vanneste, 2000; Vogt et al., 2013). With the lack of materials for disease control, developing host plant resistance would the most effective way to control the disease (Norelli et al., 2003). Crossing sensitive plant species with those that show resistance can be an effective method to introduce resistance genes and potentially create new cultivars. Without new, disease- resistant cultivars, fire blight management on Cotoneaster will remain challenging.

The Cotoneaster currently being planted as ornamentals are not representative of the whole genus, this is illustrated by Dirr (2009) who listied only 14 species. In Oregon, out of the

400 species of Cotoneaster, only 12 species are grown in Oregon Nurseries (Oregon Association 55

of Nurseries). The Ornamental Breeding program at Oregon State University is identifying species with resistance to fire blight (Rothleutner et al., 2014). The species that show fire blight resistance have been crossed with other species with horticultural attributes, to develop new cultivars with increased resistance to the bacterial disease. Currently, Oregon State University

Ornamental Plant Breeding Lab has developed a number of experimental Cotoneaster including four genotypes for which advanced testing has been conducted and two cultivars have been released. Genotypes H2011-01-002, H2011-02-001, H2011-02-005, and H2017-005-01 have a wide variety of foliar and floral characteristics and growth habits. Two of these genotypes,

H2011-02-001 and H2011-02-005, exhibit tolerance to fire blight and were released as the cultivars ‘Emerald Sprite’ and ‘Emerald Beauty.’

Breeding plants for disease resistance often is a multistep process as more germplasm sources carrying desirable traits are identified, and/or if pathogens evolve and adapt to overcome resistance in plants. This latter case has been observed in an apple breeding program focused on fire blight resistance. The research group of Peil and colleagues examined wild apple species for sources of resistance to fire blight (Aldwinkle et al., 1999; Peil et al., 2007; Peil et al. 2009; van der Zwet and Keil, 1979; Vogt et al., 2013), one of importance being Malus ×robusta 5 (Mr5)

(Peil et al., 2007). Mr5 carries a major quantitative trait locus (QTL) for resistance to fire blight on linkage group (LG) 3 (Peil, 2007, 2009, 2011). This resistance has been of particular interest to breeders and has been used to introduce fire blight resistance to many newly released apple cultivars (Durel et al., 2009). Unfortunately, Vogt et al. (2013) found that strains of E. amylovora lacking the effector AvrRpt2 or harboring avrRpt2 with a mutation was able to overcome the resistance of Mr5. The mutation in avrRpt2, among strains able to cause disease on Mr5, was a naturally-occurring single nucleotide polymorphism (SNP) which resulted in substitution of 56

cysteine with serine (Smits et al., 2014; Vogt et al., 2013; Wöhner et al, 2014). Wöhner et al.

(2014) reported there were major resistance genes on LG 3 of Mr5 and suggested that there is likely a gene-for-gene relationship in the Mr5–E. amylovora pathosystem (Fahrentrapp et al.,

2013; Wöhner et al, 2014, 2018). Subsequently, avrRpt2 mutant strains were used to aid in the development a QTL map of apple that will be immensely helpful for breeders (Wöhner et al.

2014).

The objective of this study is to evaluate fire blight resistance of Cotoneaster genotypes in repeated studies under controlled conditions of growth chambers. Previously, such studies were conducted in greenhouses and resistance rating results varied from study to study

(Rothleutner, 2014). The second objective of this study is to evaluate severity of fire blight on

Cotoneaster inoculated with a wild type strain of the pathogen and a strain with the avrRpt2C156S mutation, a strain which was capable of overcoming fire blight resistance in Mr5 apple.

Cotoneaster is related to apple and potentially may have some resistance determinants in common. If the strain with the avrRpt2C156S mutation cannot overcome fire blight resistance in

Cotoneaster, then the resistance may be unique from MR-5 in apple

Materials and Methods

Plant Material.

Plants were obtained through stem cuttings collected from mother plants kept in containers at the Oregon State University, Dept. of Horticulture, Lewis Brown Farm in Corvallis,

OR and rooted under intermittent mist. Cuttings were taken of each genotype, stuck in 2:1,

Perlite: Sunshine Mix Professional Growing Mix (Sun Gro Horticulture, Agawam, MA) and placed under mist for up to 3 months in a glasshouse. 57

For growth chamber studies, plants were grown in 10 cm, square pots in a 2:1 mixture of

Metro-Mix Professional Growing Mix (Sun Gro Horticulture, Agawam, MA) and Perlite

(Supreme Perlite Company, Portland, OR). Plants for the growth chamber studies were hand- watered using municipal water on an as-needed basis and liquid-fed weekly with Jack’s

Professional 20N-20P-20K (75 ppm fertilizer) (General Purpose; J.R. Peters Laboratory,

Allentown, PA) twice a week for 6 to 8 weeks. Substrate solution pH and electrical conductivity

(EC) were monitored at the beginning and end of the study, using the pour-through nutrient extraction procedure (Wright, 1986).

In the greenhouse study, plants were potted in a 100% unaged douglas-fir bark (Lane

Forest Products, Eugene, OR) in 2.5 L containers. The plants were hand-watered as needed by greenhouse staff and fertigated with a Dosatron with Jack’s Professional 20N-20P-20K (75 ppm fertilizer) at least twice a month.

Genotypes H2011-02-005 and H2011-02-001 were cuttings from seedlings obtained from open pollinated (OP) Cotoneaster xsuecicus ‘Coral Beauty’ plants in Corvallis, OR.

Bacterial strains.

Two strains of Erwinia amylovora were used to test Cotoneaster in these studies for fire blight resistance. Erwinia amylovora strain Ea153 (Ea153), was isolated from fire blight strike on ‘Gala’ apple in eastern Oregon. Ea153 has been used in numerous field trials and is sensitive to streptomycin, oxytetracycline, copper, and kasugamycin, chemicals used for fire blight management (Johnson et al. 1993; Stockwell et al. 1998). Strain LA635 (LA635) has the avrRpt2C156S mutation and causes fire blight disease on Malus xrobusta 5 (Wöhner et al., 2018).

LA635 also is resistant to streptomycin due to the commonly detected rpsLK43R mutation; the 58

mutations in rpsL and avrRpt2 were confirmed with genome sequence analysis (Smits et al.,

2014). The strains were stored in nutrient broth amended with 15% glycerol at -80 °C.

Preparation of freeze-dried inoculum.

Strains Ea153 and LA635 were recovered from storage and cultured on solidified

Lysogeny broth (LB) and LA635 also was cultured on LB amended with streptomycin (100

µg/ml). After confirming that colony morphology was uniform for each strain, Ea153 colonies were transferred to solidified King’s medium B (KB, King et al., 1954) amended with cycloheximide (50 µg/ml). Colonies of LA635 were transferred to solidified KB amended with cycloheximide and streptomycin. Following the procedure of Stockwell et al. (1998), the bacteria were harvested from plates after 5 days, mixed with a skim milk cryoprotectant, frozen at -80 °C and lyophilized with a FreeZone 6 L Freeze Dryer (Labconco Co. Kansas City, MO). Freeze dried bacterial preparations were stored in sealed tubes at -80 C. A week before inoculation, a sample of each bacterium was retrieved from storage, weighed, and suspended in 100 ml of sterile 10 mM phosphate buffer, pH 7.0. The suspension was mixed thoroughly, incubated at room temperature for 30 min, then, dilutions were spread on KB (Ea153) and KB amended with streptomycin (LA635). Colonies were counted after three days and converted to colony forming units per milligram.

Inoculation using leaf bisection and data collection.

Plants were inoculated with the respective strains at a concentration of 108 colony‐ forming units (CFU)/ml in 2018 and 109 CFU/ml in 2019. Inoculations were done using a foliar assay, bisecting the two youngest leaves with a pair of scissors dipped in inoculum before each cut. Control plants were cut with scissors dipped in sterile water to visualize wounding response from the inoculation procedure. The inoculated site was marked with tape. Lesion length on 59

infected shoots were measured in cm once a week for 6-8 weeks. Disease severity was quantified by calculating the percent shoot necrosis (PSN) using this equation:

푙푒푠𝑖표푛 푙푒푛𝑔푡ℎ P. S. N. = [100 ( )] 푡표푡푎푙 푏푟푎푛푐ℎ 푙푒푛𝑔푡ℎ

Area under the disease progress curve (AUDPC) was calculated for species in 2019 growth chamber experiments and used to compare disease progression between strains using R (R

CoreTeam, 2019) and figures were produced using the package agricolae (Mendiburu, 2019).

Using AUDPC. Disease severity across strains were compared among genotypes. AUDPC was calculated for all species according with the following function (Paraschivu and Cotuna, 2013):

푁𝑖−1 (푌 + 푌 ) 퐴. 푈. 퐷. 푃. 퐶. = ∑ [{ 𝑖 𝑖+1 } (푡 − 푡 )] 2 𝑖+1 𝑖 𝑖

in which Yi= disease severity on the ith date; ti = ith day; N = total number of observations or dates on which fire blight lesion length was recorded (Paraschivu and Cotuna, 2013).

Koch’s Postulates.

Koch’s postulates were performed to verify that the symptoms were due to infection by

E. amylovora. Clean pruners were used to collect tissue spanning necrotic areas and asymptomatic green tissue. A 1 cm stem or leaf section encompassing the lesion border was cut using a sterile razor blade and then diced on a sterile paper towel. This material was transferred into 2 ml of phosphate buffer and allowed to incubate at room temperature for 1 h, vortexing occasionally. Two dilution series were made for each sample and spread on media to recover the 60

pathogen. Dilutions of plant tissue samples inoculated with Ea153 were spread on KB amended with cycloheximide and streptomycin and on the semi-selective medium for E. amylovora called

CCT (Ishimaru and Klos, 1984). Lack of growth on KB amended with streptomycin plates and growth of domed colonies on CCT confirmed the presence of Ea153 in the plant sample.

Dilutions of suspensions of plant tissues inoculated with strain LA635 were spread on both KB media amended with cycloheximide and streptomycin and LB media. Mucoid, domed, white colonies on the streptomycin amended media confirmed the presence of the streptomycin- resistant strain LA635 in the inoculated plant.

Experimental design.

Studies were conducted in growth chambers and in a greenhouse. All plants were grown in a glasshouse with set temperatures of 24 °C day/17 °C night with a 14-h photoperiod until they reached adequate size where they were then moved into the growth chambers.

Studies were conducted in Percival model LED-30HL1 (Percival Scientific, Inc., Perry,

IA) growth chambers with an IntellusUltra real time controller, were used for the studies. The growth chambers are a woodless design, with programmable temperature, LED lighting, and relative humidity. Growth chambers were programmed for all four studies with a ramping cool white LED light for a 16 hour photoperiod (8 hours of 50% light, 8 hours of 100% light), with daytime temperatures of 25 C ( 0.5 C) and nighttime temperatures of 20 C ( 0.5 C), and constant relative humidity ~70% to 80% (10%).

All studies were setup as randomized complete block designs. In Study 1, plants of seven genotypes were inoculated with LA635 in growth chambers. Plants were distributed among four growth chambers, with each chamber being representing a block, with three subsamples and a control of each genotype in each block. In Study 2, plants of fourteen genotypes were inoculated 61

with EA153 in growth chambers. There were nine growth chambers, with three chambers comprising a block. Each block had three subsamples and a control of each genotype.

In 2019, eight genotypes were used in Study 3 and Study 4. In Study 3, plants were inoculated with LA635 and in Study 4, plants were inoculated with Ea153. Four growth chambers were used, and the blocking was similar to that described for Study 1. In all growth chamber studies, the terminal shoot was inoculated. The total shoot length (mm) and lesion length (mm) was measured weekly for 6 weeks in 2018 and 8 weeks in 2019.

Study 5 was conducted in the glasshouse with set temperatures of 24 °C day/17 °C night with a 14-h photoperiod. The same plants from Study 2, with diseased tissue removed, plus the addition of C. dammeri. The plants were moved into the glasshouse and potted into 2.84 L containers with the unaged douglas fir media. Plants were pruned three separate times to maintain an adequate size, with the final pruning at least three weeks before inoculation to allow for new growth. Fertigation with Jack’s Professional 20N-20P-20K (75 ppm fertilizer) was done using a Dosatron (Dosatron International, Inc. Clearwater, FL.) at least twice a month. Plants were placed away from the cooling fans in the greenhouse to provide a more consistent temperature range. Study was arranged with three blocks, each with three subsamples and a water-inoculated control for each genotype. On greenhouse grown Cotoneaster, three terminal shoots were inoculated. The three measurements per plant, then the means of the three plants were later averaged to calculate a mean percent shoot necrosis for each genotype. Total shoot length and lesion length on inoculated shoots were measured in mm once a week for 8 w.

Statistical analysis.

Data were analyzed within experiment by analysis of variance (ANOVA) by PROC

MIXED and PROC GLM in SAS (SAS Version 9.4, SAS Institute Inc., Cary, NC). Means were 62

separated using Fischer’s protected least significant difference ( < 0.05). The PROC GLM model was used to get a general linear model ANOVA. The PROC MIXED model was used to get a Type 3 Analysis of Variance, LS-means were used to compare the least squares means and determine genotype, treatment, and block interactions (α=0.05). A full model for the two 2019 growth chamber studies was used to test species for strain interaction. Where appropriate, means were separated using Fisher’s protected least significant difference ( < 0.05). LS-means were used to compare the least squares means (α=0.05) and determine species interactions within genotypes across strains in the 2019 growth chamber studies.

Results

2018 Growth chamber studies 1 and 2.

The PSN for all plants in Studies 1 and 2, inoculated with LA635 and Ea153, respectively, was under 26%. In both Studies, genotypes and treatment were significant (P <

0.05), while block was not (P > 0.1). In both studies, some genotypes showed no fire blight symptoms and other genotypes showed levels of quantitative resistance, where only a section of the stem had developed necrosis (Table 3.1& 3.2). Mean separation of genotypes in Study 1 did not distinguish between genotypes with PSN from 0 to 8% (Fig. 3.1). Mean separation of genotypes in Study 2 did not distinguish between genotypes with PSN from 0 to 17% (Fig. 3.2).

In Study 1, the PSN of plants inoculated with LA635 ranged from 0 to 21%. Three genotypes, C. splendens, H2011-01-002, and H2011-02-001, had less than 5% shoot necrosis.

Cotoneaster xsuecicus ‘Coral Beauty’ exhibited the highest percent shoot necrosis at 21%. In

Study 2, the PSN of plants inoculated with wild type Ea153 ranged from 0 to 26%, with

Cotoneaster acutifolius exhibiting the highest PSN at 26%. In Study 2, seven different genotypes 63

showed less than 5% shoot necrosis; these plants include C. daliensis, C. dielsianus, C. sikangensis, C. xsuecicus ‘Coral Beauty’, H2011-01-002, H2011-02-001, and H2011-02-005. C. acutifolius was significantly more susceptible in the Study 2 than ten other genotypes tested (P <

0.05) (Table 3.2). In both studies, inoculated with two different strains of E. amylovora, H2011-

01-002 showed complete resistance with a PSN of 0%. Cotoneaster xsuecicus ‘Coral Beauty’ had significantly higher PSN than five other genotypes tested in Study 2 (Fig. 3.1). Control plants showed no symptoms of stem necrosis after leaf bisection with scissors dipped in sterile deionized water.

2019 Growth chamber studies 3 and 4.

In Studies 3 and 4, genotypes and treatment were significant (P < 0.0001), while block was not significant (P > 0.1). Mean separation of genotypes in Study3 did not distinguish between genotypes showing less than 9% difference in shoot necrosis. Mean separation in Study

4 did not distinguish between genotypes showing less than 12% difference in shoot necrosis

(Table 3.3; Fig. 3.3).

In Studies 3 and 4, severity of symptoms ranged from 3% to 59%. Only two genotypes showed statistically significant different PSN to the two different bacterial strains: C. splendens

(P = 0.0003) and H2011-02-005 (P = 0.0012). Cotoneaster splendens was resistant to LA635 with only 4% shoot necrosis, the species was sensitive to the wild type strain Ea153 with 33% shoot necrosis. Genotype H2011-02-005 exhibited greater resistance the wild type strain Ea153, with 5% shoot necrosis and exhibited sensitivity to LA635 with 27% shoot necrosis. Genotype

H2011-01-002 had the lowest percent shoot necrosis in Studies 3 and 4 with 3% shoot necrosis when inoculated with LA635 and 3% shoot necrosis when inoculated with the wild type strain

Ea153. Genotype H2011-02-001 showed relatively high rates of resistance to both strains with a 64

percent shoot necrosis of 8% shoot necrosis when inoculated with LA635 and 3% shoot necrosis when inoculated with wild type strain Ea153. Cotoneaster dammeri in Studies 3 and 4 was significantly less resistant than all other genotypes tested with the highest PSN in both studies.

The H2017-005-01 hybrid, a C. dammeri progeny, had a significantly lower PSN than the C. dammeri parent, but the PSN was significantly greater than five genotypes inoculated with

LA635 (P < 0.05) and four other genotypes inoculated with Ea153 (P < 0.01) (Table 3.3).

Control plants showed no symptoms of stem necrosis after leaf bisection with scissors dipped in sterile deionized water.

AUDPC was calculated for all plants from Studies 3 and 4. There is a clear difference in progression of disease symptoms of the two strains of the pathogen over time in genotypes

H2011-02-005 and C. splendens. Figure 3.5 shows how fire blight progressed over the 8 weeks compared to genotype H2017-005-01, which had similar infection in both Studies (Table 3.5 Fig.

3.5).

2019 Study 5.

In Study 5 there was a range of sensitivity top fire blight among the genotypes ranging from very high levels of resistance to almost complete mortality (Table 3.4, Fig. 3.4). Both genotypes and treatment were significant (P < 0.0001), and block was also significant (P =

0.043). Statistically, there was not a significant difference between genotypes with 0% shoot necrosis and 12% shoot necrosis.

Genotypes C. sikangensis and H2011-02-005 showed complete resistance with no shoot necrosis on any of the plants. Genotypes H2011-01-002 and H2011-02-001 exhibited relatively high resistance both still below 5% with 4% shoot necrosis and 3% shoot necrosis (Fig 3.4).

Several species showed high susceptibility with over 50% shoot necrosis on the following 65

genotypes, C. acutifolius, C. daliensis, C. dammeri, C. dielsianus C. salicifolius var. floccosus, and C. xsuecicus ‘Coral Beauty’. Genotype H2017-05-001 showed a PSN of 49%, while its two parents, C. dammeri and C. apiculatus, had 73% and 43% shoot necrosis, respectively.

Cotoneaster splendens exhibited 11% shoot necrosis which was greater than that of its progeny

H2011-01-002 at 4%. Control plants showed no symptoms of stem necrosis after leaf bisection with scissors dipped in sterile deionized water.

Discussion

In this series of studies, we characterized the resistance of several genotypes of

Cotoneaster to fire blight. This is the third study done at Oregon State University to rate disease resistance among Cotoneaster genotypes in the program, but the first to examine sensitivity to an avrRpt2 mutant strain of this bacterial pathogen. This study was a continuation to determine the level of resistance of these genotypes to the same strain tested before and an avrRpt2 mutant strain that overcame resistance in a prominent wild apple rootstock called Mr5.

Although trees and shrubs are infected by E. amylovora naturally through flowers in field conditions, artificial shoot inoculation through leaf bisection is a common method that has been shown to be a justified way to study resistance to fire blight (Harshman et al., 2017; Peil et al.,

2019; Persiel and Zeller, 1981). As determined from Rothleutner et al. (2014), a threshold tolerance of 5% should be used as a selection tool for breeding purposes. The titer of our inoculum used in this study varied from 108 CFU/ml in 2018 to 109 CFU/ml in 2019. But this was not biologically significant, and other studies have used titers that ranged from 106 to 109

(CFU)/ml (Persiel and Zeller, 1978; Bellenot-Kapusta, 2002). The relatively small size of 66

Cotoneaster also makes it a good plant to use in growth chambers where we can have consistent environmental conditions.

We compared our results to those of previous studies done with the wild type strain

Ea153. The results both agreed with and conflicted with previous results. From our results collected in the greenhouse study there is a similar pattern of fire blight-sensitivity of genotypes,

C. salicifolius var. floccosus, C. dammeri, C. dielsianus, C. apiculatus, and C. xsuecicus ‘Coral

Beauty’ with the results of Lecomte and Cadic (1993). There is high degree of variability in resistance within species between years on the same plants, and this is clear in our results as well as those of others (Persiel and Zeller, 1978; Rothleutner et al. 2014). For example, Rothleutner et al. (2014) reported, C. salicifolius var. floccosus as highly susceptible one year and asymptomatic the following year. This could be due to a variety of causes from inoculation titer to the temperature of the greenhouses, and plant placement in the greenhouses. The confounding results also could be due to both sexual and apomictic reproduction (Persiel and Zeller, 1981,

1990).

In a similar study done in Davis, CA, Giffei et al. (unpublished) performed field inoculations on five of the same genotypes, including the three new genotypes bred for resistance at Oregon State, using an isolate of E. amylovora from California. They found that there were far lower rates of infection in the field with some of the same genotypes that had high rates of infections in this study. There are many explanations for this, but perhaps the most plausible is that the environmental conditions were hot and dry and not optimal for fire blight infections.

Genotypes H2011-01-002 and H2011-02-001 in this study consistently had low PSN values in all studies and Giffei et al. (unpublished) showed 0% infection in their study. These two genotypes were under 9% shoot necrosis in all of our inoculation studies. Genotype H2011-02- 67

001 has already been released as the new resistant cultivar ‘Emerald Sprite’ from the Oregon

Agricultural Experiment Station.

Apple breeders have incorporated fire blight resistant wild apple material with sensitive cultivars as a way to develop resistant cultivars. The resistance found in Malus xrobusta 5 (Mr5) has been studied and a major QTL found on the linkage group 3. This resistance was mapped only through the use of artificial shoot inoculation and not through the natural floral inoculation.

Since then more studies have been done and it has been found that Mr5 resistance also holds for flowers (Peil et al., 2019). However, there is a gene-for-gene interaction in the host-pathogen system of Mr5 and E. amylovora whereby a single nucleotide polymorphism in both the apple and the pathogen allowed an avrRpt2 mutant strain of the pathogen to overcome the resistance that genotype once offered (Vogt et al., 2013; Wöhner et al., 2018). The same avrRpt2 mutant strain of the pathogen was used in this study to help determine if any Cotoneaster genotypes are able to maintain resistance.

There were several genotypes that maintained resistance to both strains of the bacterial pathogen. There were only two genotypes that were significantly different when comparing percent shoot necrosis resulting from inoculation of the two strains. Genotype H2011-02-005 exhibited a high resistance to the wild type strain Ea153 and a reduction in resistance to the avrRpt2 mutant LA635. The interesting result is that C. splendens in 2019 exhibited high resistance to LA635, but a significant reduction in resistance to Ea153. These differences are very clearly seen in Fig. 3.5, where the disease either rapidly progresses in the first couple of weeks or the progression of necrosis stopped early. These results suggest that the different genotypes may possess different methods of mechanisms for disease resistance. More testing would need to be done to determine the types of resistance that Cotoneaster possesses. 68

Cotoneaster dammeri in all 2019 studies was significantly less resistant than all other genotypes tested, indicating that it is a poor parent to use in crosses. This is shown as one if its progeny,

H2017-005-01 had over 30% shoot necrosis, suggesting susceptibility was inherited.

From the data we cannot say that we have absolute resistance to the avrRpt2 mutant strain of E. amylovora. We also have no way of interpreting how the avrRpt2 mutant strain would perform in a field trial with natural floral infections. Growth chambers were used in our study to provide an optimal environment for fire blight. Optimal conditions for fire blight are temperatures between 23-30°C with high relative humidity (Schroth et al., 1974). A greenhouse study was added to provide more information on how the plants would fare under non-optimal conditions. The data for our study is important for breeding new sources of resistance for

Cotoneaster but may also provide insight into other resistance determinants that may be used for apple and pear breeding.

Fire blight is part of the problem affecting sales of ornamental Cotoneaster. Decline in interest and sales of Cotoneaster is partly because of fire blight. The more common planted genotypes tested in this study, showed high levels of infection. This is one of the issues that the ornamental breeding program was hoping to solve with continued resistance breeding with

Cotoneaster. It’s promising to see that several genotypes in our study show high levels of resistance to both wild type and an avrRpt2 mutant of the fire blight pathogen. These new genotypes spell a bright future for reviving interest in Cotoneaster as a landscape shrub.

Studies on fire blight resistance in ornamentals will continue to be important if new mutants of the pathogen emerge. Several studies involving ornamental trees and shrubs like pears and quince have noted the importance of developing fire blight resistant cultivars (Bell et al.,

2005; Fare et al., 1991; Postman, 2008). Our study found a varying range of susceptibility in 69

Cotoneaster to the wild type and a mutant strain of the fire blight pathogen. Several genotypes were observed to have consistent high levels of resistance to the disease. Our data does suggest that the two strains seem to cause similar levels of disease on some Cotoneaster genotypes at around the same rate. This study and previous fire blight disease resistance studies have provided insight to potential sources of resistance to develop fire-blight resistant cultivars of Cotoneaster.

Literature Cited

Aldwinkle, H.S, and J.L. Preczewski, 1979. Reaction of terminal shoots of apple cultivars to

invasion by Erwinia amylovora. Phytopathology 66:1439-1444

Bell, A.C., T.G. Ranney, T.A. Eaker, and T.B. Sutton. 2004. Resistance to fire blight among

flowering pears and quince. HortScience 40:413-415

Bellenot-Kapusta, V., R. Chartier, M.N. Brisset, and J.P. Paulin. 2002. Selection of a genotype

of Cotoneaster with a high level of resistance to fire blight. Acta Hort. 590:385-387

Dickoré, W.B., and G. Kasperk. 2010. Species of Cotoneaster (Rosaceae, Maloideae) indigenous

to, naturalising or commonly cultivated in Central Europe. Willdenowia 40:13-45

Dirr, M.A. 2009. Manual of woody landscape plants: Their identification, ornamental

characteristics, culture propagation and uses. Stipes Publishing, Champaign, IL.

Durel C., C. Denancé, and M. Brisset. 2009. Two distinct major QTL for resistance to fire blight

co-localize on linkage group 12 in apple genotypes ‘Evereste’ and Malus floribunda

clone 821. Genome 52:139-147

Fare, D.C., C.H. Gilliam, and H.G. Ponder.1991. Fire blight susceptibility, growth, and other

characteristics in ornamental pears in Alabama. J. Arboricult. 17:257-260

Fryer, J., and B. Hlmyo. 2009. Cotoneasters; a comprehensive guide to shrubs for flowers, fruit

and foliage. Timber Press, Portland, OR

Giffei, B.L., R.N. Contreras, V.O. Stockwell, J.A. Sisneroz, S.K. Reid, and L.R. Oki.

unpublished. Assessing fire blight resistance of new Cotoneaster genotypes inoculated

with Erwinia amylovora in the field. 71

Harshman, J.M., K.M. Evans, H. Allen, R. Potts, J. Flamenco, H.S. Adwinkle, M.E. Wisneiwski,

and J.L. Norelli. 2017. Fire blight resistance in wild accessions of Malus sieversii. Plant

Disease 101:1738-1745

Ishimaru, C. and E.J. Klos. 1984. New medium for detecting Erwinia amylovora and its use in

epidemiological studies. Phytopathology 74:1342-1345

Johnson, K.B., V.O. Stockwell, D.M. Burgett, D. Sugar, and J.E. Loper. 1993. Dispersal of

Erwinia amylovora and Pseudomonas fluorescens by honey bees from hives to apple and

pear blossoms. Phytopathology 83:995-1002.

Khan, M.A., Y. Zhao, and S.S. Korban. 2012. Molecular mechanisms of pathogenesis and

resistance to the bacterial pathogen Erwinia amylovora, causal agent of fire blight disease

in Rosaceae. Plant Molec. Biol. Rep. 30:247-260

Lecomte, P., and A. Cadic. 1993. Further results on shoot susceptibility of Cotoneaster to fire

blight. Acta Hort. 338:407-412

McGhee GC, J. Guasco, L.M. Bellomo, S.E. Blumer‐Schuette, W.W. Shane, A. Irish‐Brown, and

G.W. Sundin. 2011. Genetic analysis of streptomycin‐resistant (SmR) strains of Erwinia

amylovora suggests that dissemination of two genotypes is responsible for the current

distribution of SmR E. amylovora in Michigan. Phytopathology 101:182-191

Mendiburu, F. 2019. Agricolae: Statistical Procedures for Agricultural Research. National

Engineering University (UNI), Lima-Peru. https://CRAN.Rproject.org/package=agricolae

Norelli, J.N., A.L. Jones, and H.S. Aldwinke. 2003. Fire blight management in the twenty-first

century: using new technologies that enhance host resistance in apple. Plant Disease

87:756-765 72

Oh, C., and S.V. Beer. 2005. Molecular genetics of Erwinia amylovora involved in the

development of fire blight. FEMS Microbiology Letters 253:185-192

Paraschivu, M. and O. Cotuna. 2013. The use of the area under the disease progress curve

(AUDPC) to assess the epidemics of Septoria tritici in winter wheat. Res. J. Agr. Sci.

45:193-201

Persiel, F. and W. Zeller. 1978. Differences in susceptibility to fire blight, Erwinia amylovora

(Burr.) Winslow et al., in non apomictal species, varieties and oecotypes of Cotoneaster.

Acta Hort. 86:45-50

Persiel, F. and W. Zeller. 1981. Some progress in breeding Cotoneaster for resistance to

fireblight, Erwinia amylovora (Burr.) Winslow et al. Acta Hort. 117:83-88

Persiel, F. and W. Zeller. 1990. Breeding upright growing types of Cotoneaster for resistance to

fire-blight, Erwinia amylovora (Burr.) Winslow et. al. Acta Hort. 273, 297-302

Peil, A., T. Garcia-Liberos, K. Richter, F.C. Trognitz, M.V. Hanke, and H. Flachowsky.2007.

Strong evidence for a fire blight resistance gene of Malus robusta located on linkage

group 3. Plant Breeding 126:470-475

Piel, A., V.G.M. Bus, K. Geider, K. Richter, H. Flachowsky, and M.V. Hanke. 2009.

Improvement of fire blight resistance in apple and pear. Int. J. Plant Breeding 3:1-27

Peil, A., H. Flachowsky, M.V. Hanke, K. Richter, and J. Rode. 2011. Inoculation of Malus Å~

robusta 5 progeny with a strain breaking resistance to fire blight reveals a minor QTL on

LG5. Acta Hort. 869:357-362

Peil, A., C. Hübert, A. Wensing, A. Horner, O.F. Emeriewen, K. Richter, T.W. Wöhner, D.

Chagne, C. Orellana-Torrejon, M. Saeed, M. Troggio, E. Stefani, S.E. Gardiner, M.V. 73

Hanke, H. Flachowsky, and V.G.M. Bus. 2019. Mapping of fire blight resistance in

Malus ×robusta 5 flowers following artificial inoculation. BMC Plant Biol 19:532

Postman, J.D. 2008. The USDA quince and pear genebank in Oregon, a world source of fire

blight resistance. Acta Hort. 793:357-362

R Core Team (2019). R: A language and environment for statistical computing. R Foundation for

Statistical Computing, Vienna, Austria. URL https://www.R-project.org/

Rothleutner, J.J., R.N. Contreras, V.O. Stockwell, and J.S.Owen. 2014. Screening Cotoneaster

for resistance to fire blight by artificial inoculation. HortScience 49:1480-1485

Schroth, M.N., S.V. Thomson, D.C. Hildebrand, and W.J. Moller. 1974.Epidemiology and

control of fire blight. Ann. Rev. Phytopathol 12:389-412

Smits T.H.M., V.M. Guerrero-Prieto, G. Hernández-Escarcega, J. Blom, A. Goesmann, F.

Rezzonico, B. Duffy, and V.O. Stockwell. 2014. Whole-genome sequencing of Erwinia

amylovora strains from Mexico detects single nucleotide polymorphisms in rpsL

conferring streptomycin resistance and in the avrRpt2 effector altering host interactions.

Genome Announc. 2:1229-1213

Stockwell, V. O., K.B. Johnson, and J.E. Loper. 1998. Establishment of bacterial antagonists of

Erwinia amylovora on pear and apple blossoms as influenced by inoculum preparation.

Phytopathology 88:506-513. van der Zwet T. and H. Keil. 1979. Fire blight: a bacterial disease of rosaceous plants. U.S Dept.

Agr.

Vanneste, J. L. 2000. Fire blight: The disease and its causative agent, Erwinia amylovora. CABI

Publishing. ISBN 0 85199 294 3 74

Vogt, I., T.W. Wöhner, K. Richter, H. Flachowsky, G.W. Sundin, A. Wensing, E.A. Savory, K.

Geider, B. Day, M. Hanke, and A. Peil. 2013. Gene‐for‐gene relationship in the host–

pathogen system Malus×robusta 5–Erwinia amylovora. New Phytologist 197:1262-1275

Wöhner T.W., H. Flachowsky, K. Richter, T. Garcia-Libreros, F. Trognitz, M.V. Hanke, and A.

Peil, 2014. QTL mapping of fire blight resistance in Malus ×robusta 5 after inoculation

with different strains of Erwinia amylovora. Mol. Breeding 34:217-230

Wöhner T.W., K. Richter, G.W. Sundin, Y. Zhao, V.O. Stockwell, J. Sellmann, H. Flachowsky,

M.V. Hanke, and A. Peil. 2018.Inoculation of Malus genotypes with a set of Erwinia

amylovora strains indicates a gene-for-gene relationship between the effector gene eop1

and both Malus floribunda 821 and Malus ‘Evereste’. Plant Pathol. 67:938-47

Figures

30% a 25%

20% ab 15% bc bc 10%

5% c c c Percent Shoot Necrosis Shoot Percent 0%

Genotypes

ab 40%

30% a abc abc

20% bc bc bc bc 10% bc c

Percent Shoot Shoot PercentNecrosis c c c c 0%

Genotypes

70% a mean shoot infection 60% a # # LA 635 50% b mean shoot 40% b b b b infection 30% Ea153 c c 20% d d 10% c c d d c

Percent Shoot Shoot PercentNecrosis 0%

Genotypes

Fig. 3.3. Study 3 and 4. Percent Shoot Necrosis of Cotoneaster genotypes inoculated in growth chambers with Erwinia amylovora strain LA635 (109 colony-forming units/mL) in 2019, then in a following experiment with Ea153 (109 colony-forming units/mL) in 2019 with a foliar bisection assay. Means and standard errors from 12 plants are reported in percent shoot necrosis. Letter above bars represent a Fisher’s protected Least Significant Difference (LSD) test ( < 0.05) with a least significant difference value of, LA635 9% shoot necrosis and Ea153 12% shoot necrosis.

# indicates significance between the different strains within genotype H2011-02-005 (P = 0.0012) & C. splendens (P = 0.0003)

100% a b 80% bc cd de de 60% ef f 40% g gh 20% hi i i i i Percent Shoot Shoot PercentNecrosis 0%

Genotypes

Fig. 3.5. Area under the disease progress curve (AUDPC) of 3 different genotypes from the 2019 Cotoneaster app.-fire blight growth chambers studies. Genotypes H2011-02-005 and C. splendens displayed significantly different percent shoot necrosis between the two strains. Genotype H2017-005-01 displayed similar effects with both strains of the bacteria.

B A

C D

A B

H2011-01-002

H2011-02-005 ‘Emerald Beauty’

H2011-02-001 ‘Emerald Sprite’

Fig. 3.8. Cotoneaster genotypes H2011-01-002, H2011-02-001, and H2011-02-005, after study 5 using Ea153 in the glasshouse.

H2011-02-005 C. xsuecicus ‘Coral Beauty’ ‘Emerald Beauty’

Fig. 3.9. Genotypes H2011-02-005 and Cotoneaster xsuecicus ‘Coral Beauty’ compared to each other after study 5, inoculation with strain Ea153.

Tables Table 3.1. Study 1. Percent shoot necrosis of Cotoneaster genotypes inoculated in growth chambers with Erwinia amylovora strain LA 635 (108 colony-forming units/mL) in 2018 with a foliar bisection assay. Taxon Accession Percent shoot necrosisz C. xsuecicus 'Coral Beauty' 17-0006 20 a C. xsuecicus ‘Coral Beauty’ OP Seed H2011-02-005 14 ab C. dammeri x C. apiculatus H2017-005-01 8 bc C. apiculatus 16-0027 7 bc C. splendens 09-0024 1 c C. xsuecicus ‘Coral Beauty’ OP Seed H2011-02-001 0.5 c C. xsuecicus ‘Coral Beauty’ x C. splendens H2011-01-002 0 c zMean percent shoot necrosis from fire blight from 12 plants; letters represent Fisher’s protected Least Significant Difference (LSD) test ( < 0.05) with a least significant difference value 8% shoot necrosis.

Table 3.2. Study 2. Percent shoot necrosis of Cotoneaster genotypes inoculated in growth chambers with Erwinia amylovora strain Ea153 (108 colony-forming units/mL) in 2018 with a foliar bisection assay. Taxon Accession Percent shoot necrosis z C. acutifolius 10-0126 26 a C. frigidus 11-0065 20 ab C. salicifolius var. floccosus 09-0022 13 abc C. dammeri x C. apiculatus H2017-005-01 13 abc C. simonsii 09-0023 8 bc C. splendens 09-0024 6 bc C. apiculatus 16-0027 6 bc C. xsuecicus 'Coral Beauty' 17-0006 4 bc C. xsuecicus ‘Coral Beauty’ OP Seed H2011-02-005 4 bc C. dielsianus 09-0013 0.6 c C. daliensis 10-0129 0 c C. sikangensis 11-0057 0 c C. xsuecicus 'Coral Beauty' x C. splendens H2011-01-002 0 c C. xsuecicus ‘Coral Beauty’ OP Seed H2011-02-001 0 c zMean percent shoot necrosis from fire blight from 9 plants; letters represent Fisher’s protected Least Significant Difference (LSD) test ( < 0.05) with a least significant difference value of 17%shoot necrosis.

Table 3.3. Studies 3 and 4. Percent shoot necrosis of Cotoneaster genotypes inoculated in growth chambers with Erwinia amylovora strain LA 635 (109 colony-forming units/mL) in 2019, then in a following experiment with Ea153 (109 colony-forming units/mL) in 2019 with a foliar bisection assay. Taxon Accession Percent shoot Percent shoot necrosis - strain necrosis - strain LA 635z Ea153z C. dammeri 19-0036 49 a 59 a C. dammeri x C. apiculatus H2017-005-01 32 b 30 b C. xsuecicus ‘Coral Beauty’ OP Seed H2011-02-005 27 b 5 c # C. apiculatus 16-0027 18 c 25 b C. xsuecicus ‘Coral Beauty’ OP Seed H2011-02-001 9 d 3 c C. xsuecicus 'Coral Beauty' 17-0006 9 d 13 c C. splendens 09-0024 4 d 33 b # C. xsuecicus ‘Coral Beauty’ x C. splendens H2011-01-002 3 d 3 c z Mean percent shoot necrosis from fire blight from 12 plants; letters represent Fisher’s protected Least Significant Difference (LSD) test ( < 0.05) with a least significant difference value of LA635: 9% shoot necrosis and Ea153: 12% shoot necrosis.

# indicates significance between the different strains within genotype H2011-02-005 (P = 0.0012) & C. splendens (P = 0.0003)

C. acutifoliusy 10-0126 79 b

C. dammeriy 19-0036 73 bc C. daliensis 10-0129 66 cd C. dielsianus 09-0013 59 de C. xsuecicus 'Coral Beauty' 17-0006 58 de C. dammeri x C. apiculatus H2017-005-01 49 ef C. apiculatus 16-0027 43 f C. simonsii 09-0023 24 g C. frigidus 11-0065 20 gh C. splendens 09-0024 11 hi C. xsuecicus 'CB' x C. splendens H2011-01-002 4 i C. xsuecicus ‘Coral Beauty’ OP Seed H2011-02-001 3 i C. sikangensis 11-0057 0 i C. xsuecicus ‘Coral Beauty’ OP Seed H2011-02-005 0 i zMean percent shoot necrosis from fire blight from 9 plants; letters represent Fisher’s protected Least Significant Difference (LSD) test ( < 0.05) with a least significant difference value of 12% shoot necrosis. yIndicates that several subsamples had complete mortality

Table 3.5. Disease measured as area under the disease progress curve (AUDPC) for Cotoneaster genotypes evaluated for fire blight susceptibility over 8 weeks with either a normal, Ea153, strain or a mutant, LA635, strain of E. amylovora. Taxon Accession Ea153 AUDPC LA 635 AUDPC C. apiculatus 16-0027 977.23 806.15 C. dammeri 19-0036 2309.16 2184.56 C. dammeri x C. apiculatus H2017-005-01 1129.17 1435.24 C. splendens 09-0024 1367.55 136.99 C. xsuecicus ‘Coral Beauty’ 17-0006 544.74 396.41 C. xsuecicus 'CB' x C. splendens H2011-01-002 139.47 115.43 C. xsuecicus ‘Coral Beauty’ OP H2011-02-001 116.69 352.8 C.Seed xsuecicus ‘Coral Beauty’ OP H2011-02-005 247.52 1231.61 AllSeed AUDPC values are from data taken once a week for 8 weeks. All AUDPC values were determined through R coding.

CHAPTER 4: GENERAL CONCLUSION

Nursery crops in the United States are a highly profitable crop and in 2009, the US nursery plant industry brought in around $27 billion in sales. While there is no information on crop sales of evergreen huckleberry, Cotoneaster cultivars have a large economic impact through sales in both the US and in Oregon (ODA, 2020; USDA, 2016). Ornamental plants offer a lot more than just revenue. Aesthetic appeal and increased green spaces leading to better wellbeing is another reason to continue the push to breed for better landscape plants. As with all horticultural and agricultural crops, there is always interest in improving the crop. Improvement can range from different forms from aesthetic appeal or environmental purpose to disease resistance. In this thesis I focused on evaluating fruit volume and sugar content in Vaccinium ovatum, evergreen huckleberry, and completed a number of disease resistance screens using the bacterial disease fire blight in Cotoneaster.

Genome size and ploidy variation can affect a variety of different phenotypic and genotypic traits in plants. It has long been thought that an increased ploidy will lead to what is known as the “gigas effect”. This effect includes increased cell size, leading to larger flowers and fruit, thicker leaves, and increased variation in plant shape. It was unknown if ploidy manipulation would change fruit size or brix in V. ovatum. As such these studies were set up to gather a variety of different information ranging from understanding how ploidy affects these traits to seeing whether year or time in a season affect these traits. Determining how ploidy generally affects fruit volume and sugar content in V. ovatum is vital information that is helpful for starting a breeding plan for that species. By measuring these traits over three different years, I was able to compare within ploidy levels across years. By measuring over the course of 9 weeks in both 2018 and 2019, I was able to determine how these fruit characteristics change over the 90

course of the season while attached to the plants. I found that there is no consistent correlation between fruit volume and brix between ploidy or across years. However, there was consistency across years from 2017 through 2019, where volume steadily increased across years within ploidy and brix steadily decreased across years within ploidy. Mixoploids in both 2018 and 2019 were large in fruit volume in all weeks. These results help us learn that there are more factors to consider than just year and ploidy when measuring fruit volume and sugar content. Future directions for this study should keep in mind several different testing methods that were described in the discussion.

Due to lack of consistent data, there should be more data collected over the next couple years. This was the first study collecting fruit measurements over a set period of time in a collection of ornamental Vaccinium ovatum accessions. This data is extremely valuable to the increase in knowledge in the area of this native edible shrub, as well as useful for the breeding program with the eventual release of an ornamental cultivar. The information gathered in these studies will help the ornamental breeding program develop an optimum picking time for consumers who purchase the cultivars.

Disease resistance breeding is extremely important as pathogens are continuing to evolve and overcome resistance. Erwinia amylovora is a bacterial disease that hits plants in the

Rosaceae particularly hard. The pathogen enters the plants through flowers or wounds and progresses into the vascular system of stems and reduces water flow, resulting in wilting and further necrosis which is unsightly on landscape plants. Recently there has been a SNP avrRpt2 mutant of the bacterial disease identified that overcame resistance in an important apple rootstock. Through several growth chamber studies and a greenhouse study, I found that there was variable disease resistance in fifteen Cotoneaster genotypes when using two different strains 91

of E. amylovora. Disease severity was determined by calculating the percent shoot necrosis at the end of either a 6- or 8-week study. AUDPC was also calculated for several genotypes that showed significant differences among them for resistance to the two different strains.

This information could be useful on a large scale in future studies as researchers continue to evaluate the best methods for control, treatment, and prevention of fire blight in apple orchards. The information gathered will be useful to the ornamental breeding program for determining which potential new cultivars of Cotoneaster will offer resistance to fire blight. Our studies were important in determining if resistance remains consistent over multiple studies. The studies are also helpful in determining if there is resistance in eight tested genotypes to LA635.

For future directions the genotypes that were resistant to LA 635 can be analyzed more to determine if there is a similar gene conveying resistance. These studies will be used to more definitively describe the resistance of several cultivars of Cotoneaster that have recently been released from the ornamental plant breeding program at Oregon State University.

Comprehensive Literature Cited

Aldwinkle, H.S, and J.L. Preczewski, 1979. Reaction of terminal shoots of apple cultivars to

invasion by Erwinia amylovora. Phytopathology 66:1439-1444

Ayaz, F.A., A. Kadioglu, E. Bertoft, C. Acar, and I. Turna. 2001. Effect of fruit maturation on

sugar and organic acid composition in two blueberries (Vaccinium arctostaphylos and V.

myrtillus) native to Turkey. N.Z. J. Crop and Hort. Sci. 29:137-141.

Ballinger, W. E.; and L.J. Kushman. 1970: Relationship of stage of ripeness to composition and

keeping quality of highbush blueberries. J. Amer. Soc. Hort. Sci. 95: 239-242.

Ballington, J.R., W.B., Kirkman, W.E. Ballinger, and E.P. Maness. 1988. Anthocyanin,

aglycone, and aglycone-sugar content in the fruits of temperate North American species

of four sections in Vaccinium. J. Am. Soc. Hort. Sci. 113: 746-749.

Barker, W.G., F.A. Wood, and W.B Collins. 1963. Sugar levels in fruits of lowbush blueberries

at four physiological ages. Nature 198:810-811

Barney, D.L. 2003. Prospects for domesticating western huckleberries. Small Fruits Review.

2:15-29

Bell, A.C., T.G. Ranney, T.A. Eaker, and T.B. Sutton. 2004. Resistance to fire blight among

flowering pears and quince. HortScience 40:413-415

Bellenot-Kapusta, V., R. Chartier, M.N. Brisset, and J.P. Paulin. 2002. Selection of a genotype

of Cotoneaster with a high level of resistance to fire blight. Acta Hort 590:385-387

Bryan, J.E. and C. Castle. 1976. The edible ornamental garden. Pitman Ltd. London, UK

Collins, W. B., K. H. Irving and W. G. Barker. 1966. Growth substances in the flower bud and

developing fruit of Vaccinium angustifolium Ait. Proc. Amer. Soc. Hort. Sci. 89:243-247 93

Contreras, R.N., J.M. Ruter,.and B.M. Schwartz. 2010. Oryzalin-induced tetraploidy in

Cryptomeria japonica (Cupressaceae). J. Amer. Soc. Hort. Sci. 45:316-319

Coombe, B.G. 1991. Research on development and ripening of the grape berry. Am. J. Enol.

Vitic. 43: 101-110

Crisosto, C.H., G.M. Crisosto, and P. Metheney. 2003. Consumer acceptance of ‘Brooks; and

‘Bing’ cherries is mainly dependent on fruit SSC and visual skin color. Postharvest Biol.

Technol. 28: 159-167

Darnell, R.L., G.W. Stutte, G.C. Martin, G.A. Lang, and J.D. Early. 1992. Developmental

physiology of rabbiteye blueberry. Hort. Rev. 13: 339-405

DeJong, T.M., and J. Goudriaan. 1989. Modelling peach fruit growth and carbohydrate

requirements: reevaluation of the double sigmoid growth pattern. J. Am. Soc. Hort. Sci.

114: 1037-1040

Dickoré, W.B., and G. Kasperk. 2010. Species of Cotoneaster (Rosaceae, Maloideae) indigenous

to, naturalising or commonly cultivated in Central Europe. Willdenowia 40:13-45

Dirr, M.A. 2009. Manual of woody landscape plants: Their identification, ornamental

characteristics, culture propagation and uses. Stipes Publishing, Champaign, IL.

Durel C., C. Denancé, and M. Brisset. 2009. Two distinct major QTL for resistance to fire blight

co-localize on linkage group 12 in apple genotypes ‘Evereste’ and Malus floribunda

clone 821. Genome 52:139-147

Fare, D.C., C.H. Gilliam, and H.G. Ponder.1991. Fire blight susceptibility, growth, and other

characteristics in ornamental pears in Alabama. J. Arboricult. 17:257-260

Fryer, J., and B. Hlmyo. 2009. Cotoneasters; a comprehensive guide to shrubs for flowers, fruit

and foliage. Timber Press, Portland, OR 94

Giffei, B.L., R.N. Contreras, V.O. Stockwell, J.A. Sisneroz, S.K. Reid, and L.R. Oki.

unpublished. Assessing fire blight resistance of new Cotoneaster genotypes inoculated

with Erwinia amylovora in the field.

Godoy, C., G. Monterubbianesi, and J. Tognetti. 2007. Analysis of highbush blueberry

(Vaccinium corymbosum L.) fruit growth with exponential mixed models. Scientia

horticulturae. 115:368-376

Gorini, F. and M. Lasorella. 1990. Sensory and objective evaluation of kiwi fruit. Acta Hort

282:309-314

Greilhuber, J., E.M. Temsch, and J.M. Loureiro. 2007. Nuclear DNA content measurement, p.

67-101. In: J. Doleẑel, J., J. Greilhuber, and J. Suda (eds.). Flow cytometry with plant

cells: Analysis of genes, chromosomes and genomes. Wiley-VCH, Weinheim, Germany.

Hale, T.A., R.L. Hassell, and T. Phillips. 2005. refractometer measurements of soluble solid

concentration do not reliably predict sugar content in sweet corn. HortTechnology

15:668-672.

Hancock, J. F., P. Callow, R. Keelser, D. Prince, and B. Bordelon. 2000. A crop estimation

technique for highbush blueberries. J. Am. Pomol. Soc. 54:123-129

Harshman, J.M., K.M. Evans, H. Allen, R. Potts, J. Flamenco, H.S. Adinkle, M.E. Wisneiwski,

and J.L. Norelli. 2017. Fire blight resistance in wild accessions of Malus sieversii. Plant

Dis. 101:1738-1745

Hitchcock, C.L. and A. Cronquist. 1973. Flora of the Pacific Northwest. University of

Washington Press. Seattle

Hodges, A., M. Palma, and C. Hall. 2010 Trade flows and marketing practices within the U.S.

nursery industry, 2008, South. Coop. Ser. Bull. 411:63. 95

Hirvi, T. and E. Honkanen. 1983: The aroma of some hybrids between highbush blueberry

(Vaccinium corymbus L.) and bogblueberry (Vaccinium uliginosum L.). Lebensmittel

Untersuchung Forschung.176:346-349.

Ishimaru, C. and E.J. Klos. 1984. Newmedium for detecting Erwinia amylovora and its use in

epidemiological studies. Phytopathology 74:1342-1345

Ismail, A.A., and W.J. Kender. 1974. Physical and chemical changes associated with the

development of the lowbush blueberry fruit Vaccinium angustifolium Ait. Life Sciences

and Agr. Exp. Sta. Tech. Bul.70.

Jayasena, V. and I. Cameron. 2008. Brix/acid ratio as a predictor of consumer acceptability of

crimson seedless table grapes. J. Food Qual. 31:736-750

Johnson, K.B., V.O. Stockwell, D.M. Burgett, D. Sugar, and J.E. Loper. 1993. Dispersal of

Erwinia amylovora and Pseudomonas fluorescens by honey bees from hives to apple and

pear blossoms. Phytopathology 83:995-1002.

Kader, F., B. Rovel, and M. Metche. 1993. Role of invertase in sugar content in highbush

blueberries (Vaccinium corymbosum, L.). Food Sci.Technol.. 26:593-595.

Kalt, W. and J.E. McDonald. 1996: Chemical composition of lowbush blueberry cultivars. J.

Amer. Soc. Hort. Sci. 121:142-146.

Kalt, W.; K. B. McRae, and L.C. Hamilton. 1995: Relationship between surface color and other

maturity indicators in wild lowbush blueberries. Canadian J. Plant Sci. 75:485-490.

Khan, M.A., Y. Zhao, and S.S. Korban. 2012. Molecular mechanisms of pathogenesis and

resistance to the bacterial pathogen Erwinia amylovora, causal agent of fire blight disease

in Rosaceae. Plant Mol. Biol. Rptr. 30:247-260 96

Kleinhenz, M.D. and N.R. Bumgarner. 2012. Using °Brix as an indicator of vegetable quality: an

overview of the practice. Fact Sheet: Agriculture and Natural Resources: The Ohio State

Uni.1-4.

Lecomte, P., and A. Cadic. 1993. Further results on shoot susceptibility of Cotoneaster to fire

blight. Acta Hort 338:407-412

Lee, J., C.E. Finn, and R.E. Wrolstad. 2004. Comparison of anthocyanin pigment and other

compounds of Vaccinium membranaceum and Vaccinium ovatum Native to the Pacific

Northwest of North America. Am. Chemical Soc.52:7039-7044.

Lee, A.C.K., H.C. Jordan, and J. Horsley. 2015. Value of urban green spaces in promoting

healthy living and wellbeing: Prospects for planning. Risk Management and Health.

Policy. 8:131-137.

Manzoor, A., T. Ahmad, M.A. Bashir, I.A. Hafiz, and C. Silvestri. 2019. Studies on colchicine

induced chromosome doubling for enhancement of quality traits in ornamental plants.

Plants. 8:194

Matsuoka, R.H. and R. Kaplan. 2008. People needs in the urban landscape: Analysis of

Landscape and Urban Planning contributions. Landscape and Urban Plan. 84:7-19.

McGhee GC, J. Guasco, L.M. Bellomo, S.E. Blumer‐Schuette, W.W. Shane, A. Irish‐Brown, and

G.W. Sundin. 2011. Genetic analysis of streptomycin‐resistant (SmR) strains of Erwinia

amylovora suggests that dissemination of two genotypes is responsible for the current

distribution of SmR E. amylovora in Michigan. Phytopathology 101:182-191

Mendiburu, F. 2019. Agricolae: Statistical Procedures for Agricultural Research. National

Engineering University (UNI), Lima-Peru. https://CRAN.Rproject.org/package=agricolae 97

Morejohn, L.C., T.E. Bureau, J. Molè-Bajer, A.S. Bajer, and D.E. Fosket. 1987. Oryzalin, a

dinitroaniline herbicide, binds to plant tubulin and inhibits microtubule polymerization in

vitro. Planta 172:252–264

Nicholson, Rob. 2011. little big plant, box huckleberry (Gaylussacia brachycera). Arnoldia: The

Magazine of the Arnold Arboretum. 11-18.

Norelli, J.N., A.L. Jones, and H.S. Aldwinke. 2003. Fire blight management in the twenty-first

century: using new technologies that enhance host resistance in apple. Plant Disease

87:756-765

Norelli, J.N., A.L. Jones, and H.S. Aldwinke. 2003. Fire blight management in the twenty-first

century: using new technologies that enhance host resistance in apple. Plant Disease

87:756-765

Oh, C., and S.V. Beer. 2005. Molecular genetics of Erwinia amylovora involved in the

development of fire blight. FEMS Microbiology Letters 253:185-192

Oregon Dept. Agr. (ODA). 2020. Oregon Agr. Stat. & Directory. Accessed: 19 February 2020.

Paraschivu, M. and O. Cotuna. 2013. The use of the area under the disease progress curve

(AUDPC) to assess the epidemics of Septoria tritici in winter wheat. Research J. Agr.

Sci. 45:193-201

Persiel, F. and W. Zeller. 1978. Differences in susceptibility to fire blight, Erwinia amylovora

(Burr.) Winslow et al., in non apomictal species, varieties and oecotypes of Cotoneaster.

Acta Hort 86:45-50

Persiel, F. and W. Zeller. 1981. Some progress in breeding Cotoneaster for resistance to

fireblight, Erwinia amylovora (Burr.) Winslow et al. Acta Hort 117:83-88 98

Persiel, F. and W. Zeller. 1990. Breeding upright growing types of Cotoneaster for resistance to

fire-blight, Erwinia amylovora (Burr.) Winslow et.al. Acta Hort 273:297-302

Peil, A., T. Garcia-Liberos, K. Richter, F.C. Trognitz, M.V. Hanke, and H. Flachowsky.2007.

Strong evidence for a fire blight resistance gene of Malus robusta located on linkage

group 3. Plant Breeding 126:470-475

Piel, A., V.G.M. Bus, K. Geider, K. Richter, H. Flachowsky, and M.V. Hanke. 2009.

Improvement of fire blight resistance in apple and pear. Int. J. Plant Breeding 3:1-27

Peil, A., H. Flachowsky, M.V. Hanke, K. Richter, and J. Rode. 2011. Inoculation of Malus Å~

robusta 5 progeny with a strain breaking resistance to fire blight reveals a minor QTL on

LG5. Acta Hort 869:357-362

Peil, A., C. Hübert, A. Wensing, A. Horner, O.F. Emeriewen, K. Richter, T.W. Wöhner, D.

Chagne, C. Orellana-Torrejon, M. Saeed, M. Troggio, E. Stefani, S.E. Gardiner, M.V.

Hanke, H. Flachowsky, and V.G.M. Bus. 2019. Mapping of fire blight resistance in

Malus ×robusta 5 flowers following artificial inoculation. BMC Plant Biol. 19:532

Postman, Joseph D. 2008. The USDA quince and pear genebank in Oregon, a world source of

fire blight resistance. Acta Hort. 793: 357-362

Postman, Joseph D. 2004. An evergreen huckleberry industry near the Oregon coast early in the

20th Century. J.Am. Pomol. Soc. 58:147-151.

R Core Team (2019). R: A language and environment for statistical computing. R Foundation for

Statistical Computing, Vienna, Austria. URL https://www.R-project.org/

Richards, R.T. and S.J. Alexander. 2006. “A social history of wild huckleberry harvesting in the

Pacific Northwest.” United States Department of Agriculture. General Tec. Report PNW-

GTR-657, doi: 10.2737/pnw-gtr-657. 99

Robertson, K.P., J.B Phipps, J. R Rohrer, and P.G Smith. 1991. A synopsis of genera in

Maloideae (Roseaceae). Systematic Bot. 16:376-394

Robertson, N.A., F.I. Meredith, and R. Scorza. 1988. Characteristics of fruit from high and low

quality peach cultivars. HortScience. 23:1032-1034

Rothleutner, J.J., R.N. Contreras, V.O. Stockwell, and J.S.Owen. 2014. Screening Cotoneaster

for resistance to fire blight by artificial inoculation. HortScience 49:1480-1485

Rounsaville, T.J. and T.G. Ranney. 2010. Ploidy levels and genome sizes of Berberis L. and

Mahonia Nutt. species, hybrids, and cultivars. HortScience. 45:1029-1033

SAS Institute.2013. SAS.STAT Software: Release 9.4. SAS Institute Inc., N.C., USA

Sattler, M.C., Carvalho, C.R., Clarinso, W.R. 2016. The polyploidy and its key role in plant

breeding. Planta. 243:281-296

Seymour, G.B., J. Taylor, and A. Tucker. 1993. Biochemistry of fruit ripening. Chapman and

Hall, London.

Schlosser, W.E., K.A. Blatmer, and B. Zamora. 1992. Pacific northwest forest lands potential for

floral greenery production. Northwest Sci. 66:44-55

Schroth, M.N., S.V. Thomson, D.C. Hildebrand, and W.J. Moller. 1974. Epidemiology and

control of fire blight. Ann. Rev. Phytopathol 12:389-412

Shearer, K. and T.G Ranney. 2013. Ploidy levels and relative genome sizes of species, hybrids,

and cultivars of dogwood (Cornus spp.) HortScience 48:825-830

Sleeper D.A. and J.M. Poehlman. 2006. Breeding Field Crops. Blackwell Publishing, Ames,

Iowa.

Smits T.H.M., V.M. Guerrero-Prieto, G. Hernández-Escarcega, J. Blom, A. Goesmann, F.

Rezzonico, B. Duffy, and V.O. Stockwell. 2014. Whole-genome sequencing of Erwinia 100

amylovora strains from Mexico detects single nucleotide polymorphisms in rpsL

conferring streptomycin resistance and in the avrRpt2 effector altering host interactions.

Genome Announc. 2:1229-1213

Song, J., R. Smart, H. Wang, B. Dambergs, A. Sparrow, and C.Q. Qian. 2015. Effect of grape

bunch sunlight exposure and UV radiation on phenolics and volatile composition of Vitis

vinifera L. cv. Pinot noir wine. Food Chemistry. 173:424-431

Stockwell, V. O., K.B. Johnson, and J.E. Loper. 1998. Establishment of bacterial antagonists of

Erwinia amylovora on pear and apple blossoms as influenced by inoculum preparation.

Phytopathology 88:506–513.

Tamanda, T. 2002. Stages of rabbiteye and highbush blueberry fruit development and the

associated changes in mineral elements. Acta Hort. 574:129-137

Turner, J. and J.N. Long. 1975. Accumulation of organic matter in a series of douglas-fir stands.

Long Canadian Journal of Forest Research. 5:681-690

U.S. Department of Agriculture (USDA). 2016. Census of horticultural specialties for 2014.

Accessed: 19 February 2020.

https://www.agcensus.usda.gov/Publications/2012/Online_Resources/Census_of_Horticu

lture_Specialties/

Vander Kloet, S.P. 1988. The genus Vaccinium in North America. Publication 1828, Research

Branch, Publication, Agr. Canada. van der Zwet T. and H. Keil. 1979. Fire blight: a bacterial disease of rosaceous plants. U.S. Dept.

Agr.

Vanneste, J. L. 2000. Fire blight: The disease and its causative agent, Erwinia amylovora. CABI

Publishing. ISBN 0 85199 294 3 101

van Tuyl, J.M., B. Meijer, and M.P. van Diën. 1992. The use of oryzalin as an alternative for

colchicine in in-vitro chromosome doubling of Lilium and Nerine. Acta Hort 325:625-

629.

Ventura M., A. de Jager, H. de Putter, and F.P.M.M. Roelofs.1998. Non-destructive

determination of soluble solids in apple fruit by near infrared spectroscopy (NIRS).

Postharvest Biol. and Technol.14:21-27

Vogt, I., T.W. Wöhner, K. Richter, H. Flachowsky, G.W. Sundin, A. Wensing, E.A. Savory, K.

Geider, B. Day, M. Hanke, and A. Peil. 2013. Gene‐for‐gene relationship in the host–

pathogen system Malus×robusta 5–Erwinia amylovora. New Phytologist 197:1262-1275

Wöhner T.W., H. Flachowsky, K. Richter, T. Garcia-Libreros, F. Trognitz, M.V. Hanke, and A.

Peil, 2014. QTL mapping of fire blight resistance in Malus ×robusta 5 after inoculation

with different strains of Erwinia amylovora. Molec. Breeding 34:217-230

Wöhner T.W., K. Richter, G.W. Sundin, Y. Zhao, V.O. Stockwell, J. Sellmann, H. Flachowsky,

M.V. Hanke, and A. Peil. 2018. Inoculation of Malus genotypes with a set of Erwinia

amylovora strains indicates a gene-for-gene relationship between the effector gene eop1

and both Malus floribunda 821 and Malus ‘Evereste’. Plant Pathol 67:938-47

Woodruff, R. E. and D. H. Dewey. 1959. A possible harvest index for Jersey blueberries based

on the sugar and acid content of the fruit; Mich. Agr. Exp. Sta. Quart Bui. 42:340-349

Wu, J., A.R. Ferguson, B.G. Murray, Y. Jia, P.M. Datson, and J. Zhang. 2012. Induced

polyploidy dramatically increases the size and alters the shape of fruit in Actinidia

chinensis, Ann. of Bot. 109:169-179

102

Appendices

Vaccinium SAS code

Univariate/Normality title x; data x; input ploidy$ volume brix; cards;

(DATA) ; proc univariate plots normal; run; quit;

Proc GLM Table 2.2 title x; data x; input ploidy$ height width; cards;

(DATA) ; proc glm; class ploidy; model height width = ploidy means ploidy/ lsd tukey duncan; run; quit;

Table 2.3 title x; data x; input ploidy$ volume brix; cards;

DATA ; proc glm; class ploidy; model brix volume = ploidy means ploidy/ lsd tukey duncan; run; quit; 103

across years separated by ploidy title x; data x; input year$ volume brix; cards;

(DATA) ; proc glm; class year; model brix volume = year means year/lsd tukey duncan; run; quit;

Table 2.4 Ran 2019 and 2019 separately, ran ploidies individually title x(diploid)(tetraploid)(mixoploid); data x; input week$ volume brix; cards;

(DATA) ; proc glm; class week; model brix volume = week means week/lsd tukey duncan; run; quit;

T-test Ran each week separately title x; data x; input year$ ploidy$ week$ volume brix; cards;

(DATA) ; proc ttest; class year; var brix; (var volume); run; quit;

104

Vaccinium Supplemental Information Tables

Table 1. 2018 Germination of Vaccinium ovatum seedlings from 5 different experiments. Study # Mean # seedlings Germination Percent Exp. 1 61.25 7.14% Exp. 2 168.71 19.66% Exp. 4 6.60 0.77% Exp. 3 11.50 3.75% Exp. 5 1.80 1.80% Exp. 1: 2 months warm stratification, then 1 month cold stratification; 30 berries sown Exp. 2: 2 months warm stratification, then 3 months cold stratification; 30 berries sown Exp. 3: Hot water treatment; 300 seed sown Exp. 4: Sown damp; 30 berries sown Exp. 5: 4 months cold stratification; 100 seed sown Germination % came from calculating how many seeds are in a berry respectively.

Table 2. 2019 germination of Vaccinium ovatum from crosses made in 2018. Number Number of Germination Crosses berry sown seedlings rate OP Seed 30 each 30.714 3.58% 12-0018-020 x Pollen Mix 4 9 7.87% 12-0018-020 x Pollen Mix 1 1 3.50% 12-0018-005 x Pollen Mix 1 1 3.50% 12-0018-005 x Pollen Mix 2 1 1.75% 12-0018-145 x Pollen Mix 1 6 20.98% 12-0018-052 x 18-0072 (V. membranaceum) 1 1 3.50% 12-0018-145 x 18-0072 (V. membranaceum) 1 4 13.99% All seed sown damp/ no treatment; germination % came from calculating how many seeds are in a berry respectively.

105

Table 3. 2018 atLEAF Chl measurements of all Vaccinium ovatum accessions in pot, under polyhouse conditions at the Lewis Brown Farm in Corvallis, OR. Accession Ploidy Average SPAD Total chl (mg/cm2) atLEAF Chl 12-0018-148 4x 50.0 39.4 0.0370 12-0018-002 2x 66.8 56.0 0.0627 12-0018-003 2x 57.0 46.3 0.0471 12-0018-005 2x 62.0 51.3 0.0548 12-0018-007 2x 69.4 58.6 0.0671 12-0018-011 2x 58.1 47.4 0.0488 12-0018-012 2x 71.9 61.1 0.0714 12-0018-015 4x 67.7 56.9 0.0642 12-0018-020 2x 66.4 55.6 0.0620 12-0018-021 2x 73.3 62.5 0.0739 12-0018-023 4x 59.3 48.6 0.0506 12-0018-026 2x+4x 74.4 63.6 0.7590 12-0018-027 2x+4x 62.0 51.3 0.0548 12-0018-028 2x 57.3 46.6 0.0475 12-0018-030 4x 59.7 49.0 0.0512 12-0018-031 2x+4x 66.3 55.5 0.0618 12-0018-035 2x+4x 61.1 50.4 0.0534 12-0018-040 2x 73.1 62.3 0.0736 12-0018-052 2x+4x 70.5 59.7 0.0690 12-0018-058 2x+4x 62.5 51.8 0.0556 12-0018-061 2x+4x 64.7 54.0 0.0592 12-0018-063 4x 66.9 56.1 0.0628 12-0018-064 4x 68.6 57.8 0.0657 12-0018-065 4x 67.1 56.3 0.0632 12-0018-066 4x 63.2 52.5 0.0568 12-0018-070 2x 60.3 49.6 0.0521 12-0018-072 2x 64.9 54.2 0.0595 12-0018-073 4x 68.6 57.8 0.0657 12-0018-078 4x 61.3 50.6 0.0537 12-0018-084 4x 59.5 48.8 0.0509 12-0018-088 2x 63.8 53.1 0.0577 106

Table 3 (cont.) 2018 atLEAF Chl measurements of all Vaccinium ovatum accessions in pot, under polyhouse conditions at the Lewis Brown Farm in Corvallis, OR. Accession Ploidy Average SPAD Total chl (mg/cm2) atLEAF Chl 12-0018-089 4x 61.4 50.7 0.0539 12-0018-095 4x 69.5 58.7 0.0673 12-0018-096 4x 72.4 61.6 0.0723 12-0018-098 2x 61.2 50.5 0.0536 12-0018-100 2x 65.6 54.8 0.0607 12-0018-101 4x 51.9 41.3 0.0369 12-0018-107 2x 65.9 55.1 0.0612 12-0018-108 4x 65.2 54.4 0.0600 12-0018-112 4x 71.1 60.3 0.0700 12-0018-121 2x 59.5 48.8 0.0509 12-0018-124 4x 65.9 55.1 0.0612 12-0018-128A 2x 67.2 56.4 0.0633 12-0018-128B 4x 62.2 51.5 0.0551 12-0018-130 4x 62.7 52.0 0.0559 12-0018-137 4x 59.3 48.6 0.0506 12-0018-144 4x 63.8 53.1 0.0577 12-0018-145 2x 61.7 51.0 0.0544 12-0018-147 4x 61.9 51.2 0.0547 12-0018-159 4x 60.7 50.0 0.0528 12-0018-163 2x 67.7 56.9 0.0642 12-0018-165 2x 59.4 48.7 0.0507 12-0018-171 2x 64.6 53.9 0.0590 12-0018-173 2x 71.7 60.9 0.0711 12-0018-176 2x 67.6 56.8 0.0640 12-0018-177 2x+4x 71.0 60.2 0.0699 12-0018-178 2x 60.9 50.2 0.0531 Morris 2 2x 74.1 63.3 0.0754 Morris 3 2x 74.8 64.0 0.0766 Morris 4 2x 69.7 58.9 0.0676 CHL measurements taken with a digital handheld chlorophyll leaf meter on the fully expanded lamina avoiding the midrib (atLEAF Chl, FT Green LLC, Wilmington, DE). Conversions to SPAD and total chlorophyll numbers were done via a conversion tool on the atLEAF website. https://www.atleaf.com/SPAD 107

Cotoneaster SAS and R Code Imported data through filee, filenames changed each time: filename gcla520 'C:\Users\neillkr\Documents\2018 GC Ea153 Book1.csv';

Univariate/Normality data experiment1; infile gcla520 firstobs = 2 dlm = ','; input block $ taxa $ PSN treatment $; run; quit; proc univariate data=experiment1 plots normal; run; quit;

Proc mixed/Differences of Least Squares Means data experiment1; infile gcla520 firstobs = 2 dlm = ','; input block taxa $ PSN treatment $; run; quit; proc mixed data =experiment1 method = type3; class taxa treatment; model PSN= taxa treatment block; lsmeans taxa / diff cl; run; quit;

Proc GLM data experiment1; infile gcla520 firstobs = 2 dlm = ','; input block taxa$ PSN; run; quit; proc glm data =experiment11; class taxa; model PSN = taxa; means taxa/lsd duncan tukey; means taxa/tukey; run;

2019 Comparison Difference of Least Squares Means data experiment6; infile gcla520 firstobs = 2 dlm = ','; 108

input block taxa $ PSN treatment $; run; quit;

; proc mixed data =experiment6 method = type3; class taxa treatment; model PSN= taxa treatment; lsmeans taxa/ diff cl; lsmeans treatment/ diff cl; run; quit;

R code for AUDPC ```{r} ds0<-0 gs0<-0 ds1<-13.55 gs1<-2.08 ds2<-21.34 gs2<-5.06 ds3<-24.17 gs3<-5.01 ds4<-23.98 gs4<-5.31 ds5<-25.33 gs5<-5.10 ds6<-27.40 gs6<-5.12 ds7<-26.78 gs7<-5.12 ds8<-26.79 gs8<-5.12 disease.severity1<-c(ds0,ds1,ds2,ds3,ds4,ds5,ds6,ds7,ds8) disease.severity2<-c(gs0,gs1,gs2,gs3,gs4,gs5,gs6,gs7,gs8) t0<-0 t9<-0 t1<-7 t10<-1 t2<-14 t11<-2 t3<-21 t12<-3 t4<-28 t13<-4 t5<-35 t14<-5 t6<-42 t15<-6 t7<-49 t16<-7 t8<-56 t17<-8 time.period<-c(t0,t1,t2,t3,t4,t5,t6,t7,t8) #time in days for calculation time.periodweeks<- c(t9,t10,t11,t12,t13,t14,t15,t16,t17) #time in weeks for x axis plot(time.periodweeks, disease.severity1, ylim=c(0,(ds8 + 1)), xlim=c(0,(t8 + 0.5)), xlab="Weeks", 109

ylab="Disease Severity (%)", type="o", pch=19, col="midnightblue") points(time.periodweeks, disease.severity2, col="coral2", pch=19) lines(time.periodweeks, disease.severity2, col="coral2", pch=19) title(main= "AUDPC of Cotoneaster H2011-02-005", sub="Figure 7") audpc<- function(disease.severity,time.period){ n <- length(time.period)

meanvec <- matrix(-1,(n-1)) intvec <- matrix (-1, (n-1)) for(i in 1:(n-1)){ meanvec[i] <- mean(c(disease.severity[i], disease.severity[i+1])) intvec[i]<- time.period[i+1] - time.period[i] } infprod <- meanvec * intvec sum(infprod) } audpc(disease.severity1, time.period) -> AUDPCexample text(1.05,(ds5+0.8), "AUDPC LA635", col="midnightblue") text(1.05, (ds5+-2), AUDPCexample, col="midnightblue") audpc(disease.severity2, time.period) -> AUDPCexample text(3.5,(gs1+7), "AUDPC LA 520", col="coral2") text(3.5, (gs1+5), AUDPCexample, col="coral2") ```

110

Media and Buffers for Bacterial Strains

King's B medium (King et al., 1954) Proteose peptone No. 3 20 g Glycerol 10ml K2HPO4 2.0 g MgSO4·7H2O 1.5 g Agar 15 g Deionized water 1.0 L CCT medium (Ishimaru & Klos, 1984) Sucrose 100 g D-Sorbitol 10.0 g Tergitol 7 300 l 0.2 %Crystal violet 2.0 ml Nutrient agar 23.0 g Deionized water 1.0 L Lysogeny broth medium (LB Media) Glycerol 10 ml Difco Nutrient Broth 8.0 g Agar 15 g Distilled water to 1 L

https://onlinelibrary.wiley.com/doi/10.1111/epp.12019

111

Cotoneaster Germination Study Tables

Table 4. Germination percentage of five Cotoneaster genotypes based on 60 seed of each genotype sown per treatment. Taxon Direct Scarification Stratification Scar and Strat C. apiculatus 0.0% 0.0% 1.1% 5.0% C. simonsii 16.1% 16.7% 15.0% 14.4% C. xsuecicus 'Coral Beauty' 5.0% 8.3% 9.4% 5.0% H2011-02-001 0.00% 0.00% 0.00% 0.00% H2011-02-005 0.00% 0.00% 0.00% 0.00% Genotypes H2011-02-001 and H2011-02-005 failed to produce any viable seed in all trials.

Figures

20%

15% Direct

Scarification 10% Stratification

Germination Germination 5% Scar and Strat 0%

Genotypes Figure. 1. Germination percentage of five different genotypes of Cotoneaster based on 60 seed of each genotype sown per treatment. Seed were collected in early January, macerated and allowed to dry, then stored in petri dishes on dehydrant until experiment. Seeds were scarified in sulfuric acid for 60 minutes. Seeds were stratified for 60 days in 2:1 P:SS at 4 °C.