Hop Downy Mildew Management Strategies and the Etiology of Halo Blight in Michigan

HOP DOWNY MILDEW MANAGEMENT STRATEGIES AND THE ETIOLOGY OF HALO BLIGHT IN MICHIGAN

Douglas Scott Higgins

A DISSERTATION

Submitted to Michigan State University in partial fulfillment of the requirements for the degree of

Plant Pathology—Doctor of Philosophy

2020

ABSTRACT

HOP DOWNY MILDEW MANAGEMENT STRATEGIES AND THE ETIOLOGY OF HALO BLIGHT IN MICHIGAN

Douglas Scott Higgins

Effective disease management strategies for growers in Michigan and eastern U.S. growing regions that experience wet conditions are essential to support the upward trajectory of hops in these regions. Michigan experienced unusually wet weather during the 2015 growing season and plant losses resulting from hop downy mildew (HDM) infections caused by

Pseudoperonospora humuli ranged from 20 to 100%; some producers were unable to financially recover and no longer grow the crop. We evaluated the efficacy of foliar- and drench-applied fungicides against HDM and examined P. humuli isolates for point mutations linked to carboxylic acid amide (CAA; FRAC 40) resistance. Our research shows that there are four highly effective active ingredients currently registered for use on hop in Michigan and include mandipropamid (FRAC 40), fluopicolide (FRAC 43), ametoctradin (FRAC 45), and cyazofamid

(FRAC 21). The absence of resistant genotypes indicate that Michigan growers can continue to utilize CAA-containing commercial fungicides as part of an overall HDM management program.

Rootstock rot complicates foliar HDM assessments, so we selected twelve cultivars to evaluate for susceptibility (2016 and 2017). Stolons from a subset of six cultivars (2018) were used to determine rootstock rot susceptibility. Wet-rot cortex discoloration in stolons of ‘Tahoma’,

‘Newport’ and ‘Columbia’ was comparable to cultivars with more severe foliar disease symptoms (‘Nugget’ and ‘Cascade’). Differences in foliar disease among cultivars with a similar levels of rootstock rot suggests resistance to P. humuli. Asymptomatic HDM infections in planting material can escape rogueing since incubation can last 10 (leaves) or 21 days (shoots).

Real-time quantitative polymerase chain reaction (qPCR) TaqMan assays and a recombinase polymerase amplification (RPA) assay were developed and had varying success at detecting asymptomatic infections. Molecular diagnostics applied to improve HDM management in planting material should utilize the mitochondrial marker-based qPCR assay for optimal performance. In August 2018, a Michigan hop grower reported necrosis and blighting of foliage and shattering of cones resulting in yield loss. A multilocus phylogenetic analysis identified the causal agent as Diaporthe sp. 1-MI, a novel taxon that was consistently recovered from tissue and pycnidia from both leaves and cones. Pathogenicity was demonstrated in detached leaves and whole plants. Michigan hop growers face a new disease that we proposed to name “halo blight” due to a chlorotic margin of leaf lesions and browning of cone bracts.

This dissertation is dedicated to my wife, Nikita, who is a constant source of love, support, and encouragement. I am truly thankful for having you in my life. This work is also dedicated to my entire family. Especially my mother Andrea Higgins and her partner Mike Hubble, Jody and Keri Brabbit, and Kevin and Ashely Higgins whose continued encouragement and support has helped to me to pursue my passions. To my father, Jim Higgins, who unknowingly instilled in me the grit and determination needed to complete this dissertation and an appreciation of life. Finally, to beer, for giving hops a purpose that we can all appreciate.

ACKNOWLEDGMENTS

I would like to thank my advisor Dr. Mary Hausbeck. Her excellent mentorship has shown me that leadership is synonymous with compassion, inclusivity, and hard work. I appreciate her challenges to synthesize large amounts of information in a timely fashion, see the big picture, communicate effectively, and seek out impactful research. These attributes will stay with me for years to come, thank you. Thanks to Dr. Ray Hammerschmidt whom, despite his questionable beer preferences and brutal mid-term exam, has equipped me with a unique framework for understanding of pathogenesis and plant infection and a humble recognition of the vast amounts of knowledge yet to obtain. Thanks to Dr. Jan Byrne for her advice and help with the diagnostic assay development and patience in allowing me to struggle through finding countless fungal structures and then helping me with their identification. Thanks to Chris

Difonzo for providing a model teaching style, pesticide course that is invaluable to plant pathology students interested in disease management, and a glimpse into the science of entomology.

Thanks to Dr. Tim Miles for project collaboration and helping a molecular biology newbie learn the ropes. Thanks to Dr. Monique Sakalidis for helping me with the phylogenetic analysis of Diaporthe spp. Thanks to Erin Lizotte and Dr. Rob Sirrine for connecting me with hop growers, demonstrating an excellent model for Extension service, and giving me the opportunity to share my research annually at the Great Lakes Hop and Barley Conference. I thank Dr. Zack Hayden and Dr. Ron Goldy for advice on agronomic practices, Dr. Mark Nelson for recommending the non-trellised field design, Dave Francis for support with field plots, and

Filipe Couto Alves for help with the statistical analysis.

I would also like to thank my fellow graduate students and members of the Hausbeck lab.

A big thanks goes out to Sheila Linderman for her endless support, sigma plot, graphing, and editing wizardry, and her excellent taste in music. Blair Harlen for greenhouse support and a shared appreciation of hooded sweatshirts from the 90’s and unique Wisconsin cuisine. Katie

Goldenhar for teaching me how to make fungicide applications and brainstorming canoe camping trips. Dr. Kim Eang Tho for political debates, beer brewing collaborations, and sharing

Cambodian cuisine. David Perla for his practical field work ethic and joint appreciation of the torta especial. Julian Bello-Rodriguez for his thoughtful discussion, advice on molecular biology protocols, and unambiguous euchre skills. Ross Hatlen for his collaborative work with the

Diaporthe project. Celeste Dmytryszyn for keeping our lab stock with snack and caffeine and providing up lifting energy during long days. Thanks to the undergraduate research assistants,

Luis Espinosa, John Baltusis, and Ben Russell, who helped me process samples and maintain field plots.

I would like to thank the funding sources that supported this research and acknowledge the MSU Project GREEEN Grant Award Numbers GR15-021 and GR17-022, Specialty Crop

Block Grant Program Agreement Number 791AgDSC1805, the MSU SWMREC Award, and

MSU AgBioReserach. Thanks to the Hop Growers of Michigan for their support and the numerous Michigan hop growers that allowed us to collect P. humuli samples from their yards.

vii

TABLE OF CONTENTS

LIST OF TABLES ...... x

LIST OF FIGURES ...... xii

CHAPTER 1. LITERATURE REVIEW ...... 1 Introduction ...... 1 Pathogen Taxonomy and Host Range ...... 2 Symptoms and Signs ...... 6 Pathogen Biology and Epidemiology ...... 8 Disease Management ...... 14 Humulus lupulus ...... 23 LITERATURE CITED ...... 30

CHAPTER 2. FUNGICIDE EFFICACY AGAINST PSEUDOPERONOSPORA HUMULI AND POINT-MUTATIONS LINKED TO CARBOXYLIC ACID AMIDE (CAA) RESISTANCE IN MICHIGAN ...... 41 Abstract ...... 41 Introduction ...... 42 Materials and Methods ...... 45 Results ...... 53 Discussion ...... 63 APPENDIX ...... 70 LITERATURE CITED ...... 73

CHAPTER 3. SUSCEPTIBILITY OF HOP CULTIVARS AND ROOTSTOCK TO DOWNY MILDEW CAUSED BY PSEUDOPERONOSPORA HUMULI ...... 81 Abstract ...... 81 Introduction ...... 82 Materials and Methods ...... 84 Results ...... 92 Discussion ...... 98 LITERATURE CITED ...... 104

CHAPTER 4. PSEUDOPERONOSPORA HUMULI DETECTION IN ASYMPTOMATIC HOP TISSUE USING REAL-TIME MITOCHONDRIAL AND NUCLEAR MARKERS AND A RECOMBINASE POLYMERASE AMPLIFICATION (RPA) ASSAY ...... 110 Abstract ...... 110 Introduction ...... 111 Materials and Methods ...... 114 Results ...... 129 Discussion ...... 142 APPENDIX ...... 150

viii

LITERATURE CITED ...... 158

CHAPTER 5. ETIOLOGY OF HALO BLIGHT IN MICHIGAN HOP YARDS ...... 165 Abstract ...... 165 Introduction ...... 166 Materials and Methods ...... 169 Results ...... 178 Discussion ...... 197 APPENDIX ...... 204 LITERATURE CITED ...... 209

LIST OF TABLES

Table 1.1. Area under the disease progress curve (AUDPC) for downy mildew disease severity and density ratings for hop cv. Nugget when treated with foliar fungicides in 2016 and 2017 ...... 57

Table 1.2. Area under the disease progress curve (AUDPC) for downy mildew disease severity and density ratings for hop cv. Nugget when treated with drench fungicides in 2016 and 2017 ...... 60

Table 2.1: Plant vigor and disease severity and density ratings in 2016 and 2017 for downy mildew (Pseudoperonospora humuli) symptoms in the foliage of hop (Humulus lupulus) cultivars ...... 94

Table 2.2. Downy mildew (Pseudoperonospora humuli) disease incidence and severity and stolon size in 2018 for hop (Humulus lupulus) cultivars ...... 97

Table 2.3. Pearson’s correlation coefficient among stolon size and downy mildew (Pseudoperonospora humuli) symptoms in the stolons of hop (Humulus lupulus) cultivars ...... 97

Table 3.1. Primers and probes used for real-time quantitative polymerase chain reactions (qPCR) TaqMan assays to detect Pseudoperonospora humuli...... 117

Table 3.2. Recombinase polymerase amplification (RPA) primers and probes developed to detect Pseudoperonospora humuli...... 121

Table 3.4. Ten-fold serial dilution for three multiplexed assays to detect Pseudoperonospora humuli using a mitochondrial or nuclear marker and an internal plant control...... 131

Table 3.6. Results of RPA assay with the orf306 (mitochondrial) markers on healthy shoots, spikes and asymptomatic shoots collected from disease MI hop yards with diseased plants...... 141

Table A3. Quantitative PCR assays for the detection of two nuclear-based Pseudoperonospora humuli loci (c125015.3e1 and c126365.1e5) tested at a 1:30 sec extension time...... 151

Table A4. Primers and probes screened for recombinase polymerase amplification assay for the detection of open reading frame 306 in the Pseudoperonospora humuli mitochondrial genome...... 152

Table 4.1. Diaporthe spp. and outgroup D. amygdali strains used to determine the phylogenetic relationship amongst Diaporthe sp. 1-MI isolates recovered from hop leaves in Michigan ...... 175

Table 4.2. Conidial size of a Diaporthe sp. 1-MI recovered from hop cone and leaf lesions in Michigan, and previously reported pycnidia producing pathogens of hop ...... 182

Table 4.4. Diaporthe sp. 1-MI isolates obtained from symptomatic cones in Michigan hopyards in four production regions during a 2019 sampling to determine pathogen distribution ...... 190

Table 4.5. Recovery of Diaporthe sp. 1-MI and other fungal isolates from damaged hop cones (e.g. necrosis, russeting, dried bract tissue) collected from four production regions in Michigan ...... 194

LIST OF FIGURES

Figure 1.1. Symptom occurrence and disease progress of downy mildew on hop cv. ‘Nugget’ in untreated foliar fungicide field plots at two sites in Michigan. In 2016 (A and C), newly transplanted hops were inoculated with Pseudoperonospora humuli on 15, 22, and 28 June (PPF) and 21 June and 8 July (SWM). In 2017 (B and D) plots were not inoculated and disease progressed under natural conditions. The error bars represent the standard error. SWM = Michigan State University (MSU) Southwest Michigan Research and Extension Center, Benton Harbor, MI (latitude 42.6854°, longitude -84.4716°); PPF = MSU Plant Pathology Farm, Lansing, MI (latitude 42.0835°, longitude -86.3542°) ...... 54

Figure 1.2. Alignment of protein and nucleotide sequences of the cellulose synthase A3 (CesA3) gene of three downy mildew species (Pseudoperonospora humuli, Pseudoperonospora cubensis and Plasmopara viticola) linked to carboxylic acid amide resistance. Gene sequences and annotations were obtained using two sources (Blum et al., 2012 and Rahman et al., 2019). A) A protein alignment and the reported amino acid shifts for two species utilized data from Blum et al. (2012). In this plot, red regions denote putative transmembrane domains, the yellow bar in each protein sequence denotes a conserved subunit and the blue shading denotes the location of the reported amino acid shift. B) A nucleotide alignment of these three downy mildew species. Yellow within this plot denotes the exon/intron regions of the genes and the blue shading denotes the location of the PCR amplicon and primers utilized in this study to monitor for fungicide resistance. In both plots, dark vertical lines in the protein or nucleotide sequences denote amino acid shifts or single nucleotide polymorphisms, respectively ...... 61

Figure 2.1. Disease diagram developed to aid in the assessment of wet-rot cortex discoloration occurring in the stolons of hop cultivars. Stolons were split longitude with a sterile surgical blade. Disease severity was estimated directly (0-100 %) with the aid of this disease diagram as the portion of cortex tissue covered with wet-rot discoloration ...... 90

Figure 2.2. Disease diagram developed to aid in the assessment of vascular/pith discoloration occurring in the stolons of hop cultivars. Stolons were split longitude with a sterile surgical blade. Disease severity was estimated directly (0-100 %) with the aid of this disease diagram as the portion of vascular/pith tissue covered with discoloration ...... 91

Figure 3.1. Standard curves from P. humuli genomic DNA log10 dilution (from 10 ng to 10 fg) with three technical replicates for qPCR TaqMan assays based on mitochondrial marker (orf306) and nuclear markers (c125015.3e1 and c1236365.1e5). Circles and triangles are stand curves without and with the addition of plant DNA (20 ng), respectively. The linear correlation with a regression coefficient (R2) is presented for each standard curve; the * symbol corresponds to standard curves with hop DNA added...... 132

xii

time of inoculation. Control leaves (left) received only water. Leaf disks were subjective to pathogen detection with quantitative PCR and recombinase polymerase amplification assays...... 133

Figure 3.3. Quantitative real-time PCR (qPCR) assays performed with a nuclear (c125015.3e1; solid bars) or mitochondrial (orf306; patterned bars) assays to detect P. humuli in symptomatic (1 and 2 dpi) and asymptomatic (3 and 7 dpi) leaf tissue. Results were generated from two independent experiments with five replicates each. P. humuli was not detected prior to inoculation (TO) or in the negative controls for each time period post inoculation (data not shown). Bars with letters not in common represent significant pair-wise differences determined with Fisher’s protected least significant (LSD) at P = 0.001...... 134

Figure 3.4. Recombinase polymerase amplification (RPA) fluorescent signal generated for Pseudoperonospora humuli DNA (A) and non-template controls below (B) and above (C) a user- defined slope threshold established (2 mV/sec) to determine positive amplifications. A first derivative analysis applied to the same amplification curves (D-F) used to demonstrate a lack of positive amplification in non-template controls (E and F)...... 138

Figure 3.5. The onset of amplification was determined for a recombinase polymerase amplification (RPA) assay by the time the slope threshold (2 mv/sec) is crossed and the peak of the second derivative analysis is reached. Standard curves were plotted from the log of the onset of amplification for the slope threshold (A) and second derivative analysis (B) obtained from ten- fold dilution of Pseudoperonospora humuli DNA and triplet technical replicates. Circles and triangles are stand curves without and with the addition of plant DNA (20 ng), respectively. The linear correlation with a regression coefficient (R2) is presented for each standard curve...... 139

Figure 3.6. The first and second derivative analysis from real-time reverse transcription PCR (RT-PCR) and used to analyze raw recombinase polymerase amplification data to positive amplification (first derivative) and the onset of amplification (second derivative). Figure adapted from Luu-The et al. (2005) and originally credited to the LightCycler manual (Hoffam-La Roche)...... 147

Figure A1. Input algorithm for the Axxin AX0ISO desktop application to call positive Pseudoperonospora humuli amplification based on a slope threshold of 2 m/V. A threshold of 1 mV/sec was used for the internal plant control...... 154

Figure A2. Amplification beyond the LOD (100 fg) for the orf306 assay in control leaves (A) and shoots (B). Cq values during experimental runs for a 100 fg of P. humuli DNA were 33.07± 0.05 s.d. and 32.43± 0.27 s.d. (A) and Cq = 38.9± 0.60 s.d. (B)...... 155

Figure A3. Recombinase polymerase amplification from four non-template controls. The two upper line have spike with hop DNA and have exceed the slope threshold (2 m/V) for positive detection (2.99 and 3.11 mVs/sec)...... 156

xiii

Figure A4. Reaction mixtures without primers tested on non-template control (A), crude hop leaf extract (B), 10 ng P. humuli DNA (C), and 10 ng P. humuli + crude hop leaf (D)...... 157

Figure 4.1. Signs and symptoms of a Diaporthe sp. 1-MI on hop foliage. A-D) Foliar symptoms were sporadically distributed throughout the plant canopy included irregularly-shaped (13.4 cm2 ± 13.4 S.D) necrotic lesions surrounded by a chlorotic margin. C) Foliar necrosis on a 1-month old transplant of 'Cashmere' located next to an infested hopyard. E, F) Pycnidia formation on mature hop leaves prior to harvest. Coalescent alpha conidia and cirrhi from mature pycnidia of Diaporthe sp. 1-MI on hop foliage using a dissecting microscope and hand-sectioning. G, H) Diaporthe sp. 1-MI growth and conidia morphology on PDA (30 days) ...... 179

Figure 4.2. Signs and symptoms of a Diaporthe sp. 1-MI on hop cones. A,B) Dried necrotic tissue where Diaporthe sp. 1-MI has been isolated. C) A tiled image of dried bract tissue infected with Diaporthe sp. 1-MI. D,E,F) Pycnidia and pycnidial ooze observed on bracts with a dissecting microscope. G, H) Diaporthe sp. 1-MI growth and conidia morphology on PDA (27 days) ...... 180

Figure 4.3. Scanning electron micrographs of Diaporthe sp. 1-MI affecting hop leaf and cone tissue. Leaf (A-C) and cone (E) pycnidia immersed in necrotic lesions with a defined layer of fungal tissue (eustoma) separate from the outer wall of the pycnidia. Some cone pycnidia were embedded in textura angularis (D, F), but shared morphological features with leaf pycnidia including a stout pycnidia neck, conidia morphology, and no beta conidia. Coalescent cylindrical to ellipsoid conidia emerging from pycnidia in leaf (G) and cone (H) tissue; cirrhi observed in pycnidia from leaf tissue (G). Pycnidia from both tissues contained a singular ostiole (C, F, I) ...... 181

Figure 4.4. Maximum likelihood (ML) phylogram of combined HIS, ITSrDNA, TUB and TEF1 datasets. Strains in bold represent ex-type, ex-isotype, and ex-epitype. At each node ML bootstrap support values ≥ 70 and Bayesian posterior probability values ≥ 0.75 are indicated. The tree was rooted to Diaporthe amygdali ...... 187

Figure 4.5. Symptoms of cone blight on ‘Centennial’ (A-K) and ‘Centennial’ or ‘Cascade’ (G-F) hop cones. Single-spore isolates of Diaporthe 1-MI were derived directly from the pycnidia in blighted tissue of 25 cone samples. The following Diaporthe sp. 1-MI were recovered from the cones are pictured: (A) MI_119, MI_3719; (B) MI_0219, 1219; (C) MI_3019, MI_3119, MI_5319; (D) MI_0819; (E) MI_4619; (F) MI_4719; (G) MI_0719, MI_2419; (H) MI_4119; (I) MI_3919; (J) MI_3819, MI_2519; (K) MI_4119 ...... 188

Figure 4.6. Fungal recovery (%) from hop cones with necrosis, russeting, and dried bract tissue in the lower peninsula of Michigan (A) of Diaporthe sp. 1-MI, other fungal spp. (Alternaria spp. Fusarium spp. and Epicoccum spp.), and cones yielding two or more fungal isolates (co- recovery). Co- recovery (%) from hop cones (B) of Diaporthe sp. 1-MI, Alternaria spp. and Fusarium spp. with one other fungal isolate. In some cases two or more fungal isolates were co- recovered (2+) ...... 193

xiv

Figure 4.7. The effect of temperature on mycelium growth of three Diaporthe sp. 1-MI (MI_0318, MI_1418, and MI_2518) recovered from hop. Mycelium growth diameter was measured every two days. There were five replicates per isolate and the vertical bars represent standard errors of the mean ...... 196

Figure A5. Maximum likelihood (ML) phylogram of the histone H3 gene (HIS) dataset. Strains in bold represent ex-type, ex-isotype, and ex-epitype. At each node ML bootstrap support values ≥ 70 and Bayesian posterior probability values ≥ 0.75 are indicated. The tree was rooted to Diaporthe amygdali...... 205

Figure A6. Maximum likelihood (ML) phylogram of the internal transcribed spacer (ITSrDNA) dataset. Strains in bold represent ex-type, ex-isotype, and ex-epitype. At each node ML bootstrap support values ≥ 70 and Bayesian posterior probability values ≥ 0.75 are indicated. The tree was rooted to Diaporthe amygdali...... 206

Figure A7. Maximum likelihood (ML) phylogram of the beta-tubulin locus (TUB) dataset. Strains in bold represent ex-type, ex-isotype, and ex-epitype. At each node ML bootstrap support values ≥ 70 and Bayesian posterior probability values ≥ 0.75 are indicated. The tree was rooted to Diaporthe amygdali...... 207

Figure A8. Maximum likelihood (ML) phylogram of the elongation factor 1 alpha (TEF1) dataset. Strains in bold represent ex-type, ex-isotype, and ex-epitype. At each node ML bootstrap support values ≥ 70 and Bayesian posterior probability values ≥ 0.75 are indicated. The tree was rooted to Diaporthe amygdali...... 208

CHAPTER 1. LITERATURE REVIEW

Introduction

Commercial hop (Humulus lupulus L.) production returned to Michigan in 2008 after a more than 150-year hiatus (Sirrine et al. 2014). The hop cones (female strobili) are harvested for their bittering (alpha- and beta-acids) and aromatic (monoterpenoids and sesquiterpenoids) properties that are important components in beer production (Neve 1991). From 2015 to 2018, harvested hops increased from 130 to 304 ha in Michigan, making the state the fourth largest hops producer in the U.S. behind Washington, Oregon, and Idaho (George 2019). The crop has also increased in popularity in other states across the eastern U.S.; 514 ha of hops were harvested in 2018 from 19 states (George 2019).

Hop downy mildew (HDM) caused by Pseudoperonospora humuli (Miyabe & Takah.)

G.W. Wilson, (1914) is a highly destructive pathogen of hop in Michigan (O’Neal 2015) and other growing regions (Mahaffee et al. 2009; Neve 1991; O’Neal 2015; Royle and Kremheller

1981). Both above- and belowground portions of the plant are susceptible to the pathogen.

Chronic systemic infections are associated with rot and weaken susceptible rootstock (perennial underground stem organs, crown, and accompanying roots) depleting winter carbohydrate reserves needed for vigorous perennial growth (Williams et al. 1961) and reducing yield up to

56.2% (Coley-Smith 1962). Hypertrophy in diseased apical buds causes the bines to fall off the training string, requiring additional labor costs to retrain replacement bines (Ware 1926). Direct yield and quality loss occur when infected inflorescences abort, develop discolored bract, and lose alpha acid content (Neve 1991; Royle and Kremheller 1981).

Michigan experienced unusually wet weather during the 2015 growing season and HDM developed in hop yards at levels not previously observed (Lizotte 2015). Plant losses as a result

of HDM ranged from 20 to 100%; some producers were unable to financially recover and no longer grow the crop (Erin Lizotte, personal communication). While management strategies for

HDM are established for hop growers in the Pacific Northwest (Gent et al. 2009; Gent et al.

2012; Gent and Ocamb 2009; Gent et al. 2010; Gent et al. 2015; Hunger and Horner 1982;

Johnson 1991; Johnson and Anliker 1985; Johnson et al. 1994; Johnson and Skotland 1985;

Skotland and Johnson 1983; Woods and Gent 2016), similar recommendations are lacking for

Michigan and other eastern U.S. production areas where environmental conditions can be especially conducive to HDM. The research objectives of this dissertation are to develop integrated disease management strategies for Michigan including effective fungicides, resistant cultivars, and disease-free planting material.

Pathogen Taxonomy and Host Range

Pseudoperonospora humuli is an oomycete and a biotrophic plant pathogen. Oomycota are eukaryotic stramenopiles that share a similar biological lifestyle to osmotrophic members of the Kingdom Fungi (Beakes et al. 2014; Dick 2001). Both secrete enzymes to absorb nutrition and have yeast-like or hyphoid growth form (Beakes et al. 2014; Dick 2001). However, the

Oomycota are distinct from Fungi in several ways. In the Oomycota, vegetative thalli are diploid, heterogametangia (antheridia and oogonia) produce oosporangia, cellulose and β1-3 glucans form the main polysaccharide structural and carbohydrate storage components of the cell walls, lysine is synthesized from the DAP (α,ϵ-diaminopimelic acid) pathway, mitochondria contain tubular cristae, and zoosporangia flagella are heterokont possessing two types of flagella

(posteriorly directed whiplash and anteriorly directed fibrous and ciliated) (Dick 2001).

The Oomycota phylum is divided into two monophyletic class-level clades

Peronosporomycete and Saprolegniomycete and early diverging clades (classes Incertae Sedis)

(Beakes et al. 2014). Peronosporomycetes is made up of three orders including the Rhipidales,

Albuginales and Peronosporales (Beakes et al. 2014). Members of the Peronosporales include many important plant pathogenic genera such as Pythium, Phytophthora, and a group of genera that are collectively referred to as the downy mildews (Beakes et al. 2014; Thines and Choi

2016).

There are 19 downy mildew genera that collectively contain more than 700 species

(Beakes et al. 2014). The three major monophyletic groups include, 1) downy mildew with colored conidia, 2) brassicolous downy mildew, and 3) downy mildew with pyriform haustoria

(Thines and Choi 2016; Thines et al. 2009). The graminicolous lineages, parasitizing grasses, remain unresolved (Thines and Choi 2016). Two genera, Peronospora and Pseudoperonospora, comprise the group of downy mildew with colored conidia (Thines and Choi 2016). Both genera produce sporangia but the sporangium of Pseudoperonospora can only germinate with zoosporangia; Peronospora sporangia germinate with germ-tubes (Thines and Choi 2016). There are fewer than 10 species in the genus Pseudoperonospora (Thines et al. 2009) of which two, P. humuli and P. cubensis (Berk. & M.A. Curtis) Rostovzev, are economically important plant pathogens (Holmes et al. 2015).

Much attention has been directed at differentiating P. cubensis (the causal organism of downy mildew of cucurbits) and P. humuli (Choi et al. 2005; Mitchell et al. 2011; Runge et al.

2011; Runge and Thines 2012; Summers et al. 2015). However, a biological species concept based on morphology alone is difficult to apply to downy mildew identification (Hall 1996).

These sister taxa share near identical sporangiophores (Choi et al. 2005), sporangia (Choi et al.

2005; Constantinescu 2000), and haustoria (Fraymouth 1956). Further, morphological structures may differ depending on the host (Choi et al. 2005; Runge and Thines 2011, 2012).

Genetic loci can be used to distinguish between P. cubensis and P. humuli (Mitchell et al.

2011; Rahman et al. 2019; Runge et al. 2011). Early phylogenetic analysis using the internal transcribed spacer of the nuclear ribosomal DNA region (ITS) and the neighbor-joining method failed to detect genetic divergence and concluded that the sister taxa were synonymous (Choi et al. 2005). However, with improved probability methods (Bayesian and maximum likelihood analyses) congruent phylogenies were produced for individual loci including the ITS region, nuclear β-tubulin gene, and the mitochondrial cytochrome c oxidase (cox) cluster (cox2, cox2- cox1 spacer, and cox1) and multilocus (ITS and the β-tubulin gene) data that clearly distinguish

P. humuli and P. cubensis as separate species (Mitchell et al. 2011). Further, multilocus analysis including the Ras-related protein gene, cox2, and ITS can delimitate P. humuli and P. cubensis from other Pseudoperonospora species found on Urtica, Cannabis, and Celtis (Runge et al.

2011). The P. humuli nuclear genome has revealed additional divergent genetic loci, including four highly polymorphic nuclear unnamed open reading frames (c125015.3e1, c127446.1e1, c127233.5e3, and c126365.1e5), that appear to be unique to P. humuli (Rahman et al. (2019).

Pseudoperonospora humuli has a narrow host range (Hoerner 1940; Mitchell et al. 2011;

Runge and Thines 2012; Salmon and Ware 1928). Humulus lupulus is regarded as the primary host of P. humuli (Mitchell et al. 2011). Both successful (Hoerner 1940) and unsuccessful

(Salmon and Ware 1928) infection of H. scandens (Lour.) Merr. (synonym [syn]: H. japonicus

Siedold and Zucc; H. japonicus var. variegatus F. Roem) by P. humuli has been reported. P. humuli recovered from H. scandens is often genetically distinct from P. cubensis and P. humuli isolates (Mitchell et al. 2011; Runge et al. 2011) indicating it might be a separate species but

studies to date have been conducted with a limited number of isolates. The occurrence of H. yunnanensis Hu. is so rare that it has yet to be reported whether the species is able to host P. humuli (Neve 1991). Other plants also belonging to the Rosales order are potential hosts

(Hoerner 1940; Salmon and Ware 1928). P. humuli inoculated on Urtica urens L. (syn: U. lyallii; common name: annual or dwarf nettle) produced viable sporangia able to reinfect hop; limited mycelium colonization of U. dioica L. (common name: stringing or common nettle) and

Parietaria judaica L. (syn: P. ramiflora; common name: pellitory) produced little to no sporangiophores or sporangia (Salmon and Ware 1928). Several varieties of Cannabis sativa L.

(common name: marijuana) and Celtis spp. including C. occidentalis L. (common name: hackberry), closely related taxa in the Cannabaceae family (Yang et al. 2013), are reported as potential hosts of P. humuli (Hoerner 1940).

Host specificity experiments with P. cubensis and cucurbit species have produced mixed results, but in all cases of successful cross-infection, pathogen reproduction is reduced compared to the positive control (Mitchell et al. 2011; Runge and Thines 2012). P. humuli was only able to produce a single sporangiophore on Cucumis melo var. cantalupensis Ser. and no sporulation was observed on C. sativus L. (Mitchell et al. 2011). Contrasting results were observed in Europe where P. humuli infected C. sativus L. and produced sporangia at a low success rate; higher infection rates were observed on the wild cucurbit hosts Bryonia dioica Jacq. and Sicyos angulatus L. (Runge and Thines 2012). When P. cubensis was inoculated on H. lupulus L., infection success was high (100% with the hop cultivar Pacific Gem) albeit less P. cubensis sporangiophores emerged on hop compared to the positive cucumber controls (Mitchell et al.

2011). Lower rates of infection success were reported in Europe where P. cubensis sporulated on

H. lupulus in only 24% of trials. (Runge and Thines 2012). Wallace et al. (2020) presented

evidence based on genetic differentiation of two host-adapted clades for P. cubensis placing P. cubensis isolates that preferentially infect Cucurbita pepo, C. maxima, C. moschata, Citrullus lanatus, Momordica charantia, and M. balsamina in Clade 1 and P. cubensis isolates preferentially infecting C. sativus, C. melo and Lagenaria siceraria in Clade 2. It appears that previous studies investigating the infection success of P. cubensis on H. lupulus (Mitchell et al.

2011; Runge and Thines 2012) may have been conducted only with Clade 2 isolates leaving the host range of Clade 2 on H. lupulus unresolved.

Symptoms and Signs

Pseudoperonospora humuli is an airborne pathogen that overwinters primarily in the rootstock (Coley-Smith 1962; Johnson and Skotland 1985; Skotland 1961; Ware 1926). HDM symptoms in rootstock produce a wet-rot that is red brown to dark brown and black in color and can include slightly water-soaked flecks streaks (Coley-Smith 1964; Skotland 1961; Ware

1926). The darkening color of the symptoms is considered to represent the progression of the disease (Coley-Smith 1962; Skotland 1961). Coley‐Smith (1964) noted that in HDM-infected rhizomes large areas of wet-rot discoloration in the cortex was more common than the flecking symptom. In the latter stages of infection, both symptoms may occur throughout all tissue types.

Phytophthora citricola Sawada (1927) produces similar rootstock rot symptoms that are confined to the cortex but often accompanied wilting of above ground foliage (Royle 1968a). Both

Phomopsis tuberivora H.T. Güssow & W.R. Foster, the causal agent of red crown rot of hop and

Fusarium Canker caused by Fusarium sambucinum Fuckel produces a dry-rot (Gent et al. 2013;

Mahaffee et al. 2009) distinctly different from the wet-rot attributed to HDM.

Rootstock buds that are colonized after infection give rise to symptomatic basal shoots with swollen stems and chlorotic downward cupped leaves (Coley-Smith 1962; Coley-Smith

1964; Skotland 1961; Ware 1926). Secondary infections of apical meristems and lateral shoots produce symptoms similar to basal shoots but are distinguishable by one or more normally elongated internode at the shoot base (Coley-Smith 1962; Skotland 1961). Symptomatic shoots are commonly referred to as “spikes” (Royle and Kremheller 1981). Localized leaf lesions appear chlorotic, become necrotic, but remain limited by the leaf veins making them angular in shape (Mahaffee et al. 2009).

Male and female flowers are both susceptible to HDM (Hoerner 1932) and cone damage can occur at all stages of development (Ware 1926). Severely diseased cones may turn brown, dry and occasionally drop (Hoerner 1932). Cones affected early in development may cease to grow and harden; both early- and late-infected cone develop brown discoloration (Hoerner

1932). Cone discoloration is not diagnostic and may be attributed to many hop pathogens including Alternaria alternata (Fr. : Fr.) Keissl. (1912) (Darby 1984), Fusarium spp. (Bienapfl et al. 2005; Pethybridge et al. 2001a), P. humuli (Royle and Kremheller 1981), Botrytis cinerea

Pers. : Fr. (Mahaffee and Engelhard 2009), Phaeomycocentrospora cantuariensis (E.S. Salmon

& Wormald) Crous, H.D. Shin & U. Braun 2012 (analogous with Cercospora cantuariensis

(Radišek et al. 2009), Phoma exigua var. exigua (Desm.) Aveskamp, Gruyter & Verkley (2010)

(Radišek et al. 2008), and Podosphaera macularis (Wallr.) U. Braun & S. Takam. (2000)

(Twomey et al. 2015).

Pseudoperonospora humuli produces dark purple to black sporangia primarily on abaxial leaf surfaces and occasionally on stems, stipules, and cone bracts (Ware 1926). The sporangia of

P. humuli on H. lupulus are generally considered pale olivaceous to olivaceous in color,

ellipsoidal and measure 22-33 x 15-19 µm (Choi et al. 2005) with a two layered wall structure and rough to ornamented outer surface (Constantinescu 2000). Larger hyaline sporangia can occur when the sporangiophore and sporangia emerge inside water droplets (Sonoda and Ogawa

1970). Sporangia contain a lens shaped apical operculum (Shaw 1981) covering the dehiscence apparatus (also referred to as a papilla or pore) and a basal inconspicuous pedicel

(Constantinescu 2000). Sporangiophores are 120-460 µm long and straight with monopodial branches occasionally dichotomous (Choi et al. 2005). Oospores are spherical with smooth walls measuring 20 (Mahaffee et al. 2009) to 41.45±0.5 µm (Chee and Klein 1998). Haustoria are vermiform with clavate branches that occasionally appear slightly coiled (Coley-Smith 1964;

Fraymouth 1956).

Pathogen Biology and Epidemiology

Pseudoperonospora humuli was first described in Sapporo province, Japan, in 1905, on both commercially cultivated and wild hop plants (Wilson 1914). Combined with a later report in the U.S., on native hops in Wisconsin (Davis 1910), the two specimens were reasoned synonymous, despite a note (but no details) of some morphological difference (Wilson 1914).

The pathogen was first reported in English hop yards in 1920 (Salmon and Wormald 1923) and rapidly spread across the European production regions culminating in the pandemic outbreak of

1928 (Neve 1991). The sudden European pandemic was attributed to infected planting material instead of an endemic pathogen population (Neve 1991). In North America, first disease reports occurred in commercial yards in 1928 in New York, U.S. and British Columbia, Canada

(Hoerner 1932). HDM was especially prevalent in New York in severely diseased hop yards established after 1933 where reported losses of at least one-third of the crop occurred (Magie

1942). The disease appeared in western counties of Washington state and Oregon in 1929 and

1930, respectively (Hoerner 1932). The pathogen was unreported in California until 1934 and the

Yakima Valley, WA until 1937 (Skotland and Johnson 1983). In Argentina, P. humuli was first reported in 1957 (Pérez et al. 2009). Despite HDM conducive conditions, the Australian production region remains disease free (Pethybridge et al. 2003); there are no reports of HDM in

South Africa and New Zealand (Mahaffee et al. 2009).

Today in the U.S., HDM epidemics occur annually in western Oregon (Gent et al. 2010;

Skotland and Johnson 1983) and ,on average, once every three years in Washington’s semi-arid

Yakima valley (Johnson and Anliker 1985; Johnson et al. 1994; Skotland and Johnson 1983). In

Washington, seasonal epidemics (seldom longer than 4-5 weeks) rarely persist to flowering due to the onset of hot, dry weather in June (Johnson et al. 1994). In Oregon, commercial growers commonly cease HDM fungicide applications in mid-July around the time of flowering (Gent et al. 2010). HDM in Michigan has been reported annually since 2013 (Lizotte, 2013; 2014; 2015) but the duration of seasonal epidemics has yet to be characterized.

Infection and Colonization. Pseudoperonospora humuli sporangia infects indirectly, via zoosporangia, the natural openings in the leaves and at the base of shoots (Royle and Kremheller

1981). Sporangia require 1 h of wetness at 20 to 22°C to discharge zoosporangia (Royle and

Kremheller 1981). The sporangia apical operculum swells during the final stages of maturation and completely detaches just prior to the release of 4 to 8 zoosporangia (Royle and Kremheller

1981; Shaw 1981). Daylight favors zoosporangia encystment near stomata with open vestibules

(Royle and Thomas 1971b) and immature stomata (Royle and Thomas 1971a). Zoosporangia enter through the stomata opening via a germ tube (Royle and Thomas 1971b, a). In darkness, zoosporangia are ineffective at locating stomata (Royle and Thomas 1971b, a, 1973) which may

explain why disease outbreaks do not occur when wetness is provided by dew alone (Royle

1973; Royle 1970). On silicone rubber impressions of leaves exposed to light, zoosporangia successfully encysted near stomata openings, but took slightly longer to find stomata than zoosporangia released on leaf surfaces. This suggests that zoosporangia may utilize both a thigmo- and chemotaxis mechanism to select infection sites (Royle and Thomas 1973).

Both leaf and shoot infections can occur with 3 to 4 h of wetness at 19 to 23°C (Royle

1970). Even with 24 h of wetness, shoot and leaf infections fail to occur at 24 to 25°C and 30°C, respectively (Royle 1970). Temperatures 10°C and lower, limit shoot infections but leaf infection can occur at 5°C with 24 h of wetness (Royle 1970). Field observations of isolated root infections and artificial inoculation of a zoosporangia suspension applied to the soil, suggest that hop rootstock infections could occur from P. humuli zoospores washing through the soil (Coley-

Smith 1962; Coley-Smith 1965; Royle and Kremheller 1981). The process of direct infection belowground has yet to be characterized.

Nearly all host tissue can be colonized by P. humuli (Coley-Smith 1962; Skotland 1961;

Ware 1926). Mycelia growth in aboveground shoots can occur in either direction (Coley-Smith

1962; Skotland 1961; Ware 1926) and can travel up to 15 cm toward the crown in one growing season (Coley-Smith 1965). The pathogen can persist in crown tissue for up to 4 years (Royle and Kremheller, 1981). Haustoria have been observed in the cortex, phloem, pith, and the parenchyma cells of xylem of crowns, rhizomes and roots up to 12-16 in deep (Skotland 1961;

Ware 1926). The apical meristem and outer scale leaves of dormant buds are systemically colonized (Coley-Smith 1962) but not all colonized buds produce symptomatic basal shoots

(Coley-Smith 1962). Healthy and infected dormant buds were frequently present on the same node (Coley-Smith 1962). The colonization of dormant buds is the primary method of

overwintering for the pathogen (Coley-Smith 1962; Skotland 1961). Shoot symptoms, swollen shoots and chlorotic downward cupped leaves can take 7-22 days (9-20 ° C) to develop (Royle

1970). In leaves, chlorotic lesions limited by the leaf vein have a 3-10 days (7-28° C) incubation period (Royle 1970).

Dispersal and Reproduction. Symptomatic basal shoots provide the primary inoculum in established hop yards (Coley-Smith 1962; Gent et al. 2010; Johnson 1991; Johnson and

Anliker 1985; Johnson et al. 1994; Skotland 1961). In Washington, precipitation and above normal temperatures in April and May favor symptomatic basal shoot emergence (Johnson et al.

1994). The number of growing degree-days (GDD) for air (111.3) and soil (88.7) temperature can be used to predict symptomatic basal shoot emergence 4.9 days prior to the actual emergence

(Johnson 1991). GDDs were calculated by adding the maximum and minimum daily air temperatures, dividing by two, and subtracting a base temperature of 6.5° C (Johnson 1991).

However, the Washington model was not predictive in Oregon and a modified calculation of

GGD using the sine function of 6.0° C (base temperature) resulted in a GDD threshold of 115.7 and a predictive lead time of 5.3 days (Gent et al. 2010). In yards with high disease pressure the previous season symptomatic basal shoots emerge earlier than in yards with lower disease pressure the previous season (Gent et al. 2010; Johnson and Anliker 1985). Regional differences require model adjustments to predict the emergence of basal shoots (Gent et al. 2010).

In P. humuli, diploid sporangia (Dick 2001) are produced through asexual reproduction and dispersed by wind and water (Royle and Kremheller 1981). Sporangiophores emerge primarily through stoma on the abaxial leaf surface (Royle and Kremheller 1981). Profuse sporulation occurs when relative humidity is > 80% (Johnson and Skotland 1985) or 90% (Royle

1968b). Light sporulation was shown to occur at RH levels as low as 50% (Royle 1968b) and

64% (Johnson and Skotland 1985). Sporulation is altered (larger hyaline sporangia) or inhibited when the sporangiophore and sporangia emerge inside water droplets on the leaf surface (Sonoda and Ogawa 1970). Under favorable environmental conditions the alteration of light and darkness

(diurnal cycle) promotes sporulation (Yarwood 1937). Sporulation occurs overnight with sporangia maturing in the early morning (around 6:00 a.m.) (Yarwood 1937).

Airborne sporangia concentration typically peaks in mid- to late-morning as a result of declining RH (Royle and Thomas 1972; Royle 1968b; Sonoda and Ogawa 1972). Secondary peaks in airborne sporangia concentration occur when RH levels are below saturation and rainfall occurs (Royle and Thomas 1972). Airborne sporangia concentrations are unaffected by temperature (0-30°C) (Royle and Thomas 1972). Under laboratory conditions, sporangia survival is dependent on RH, with a 0% survival rate in less than 2 h at 50% RH and 70% survival rate in

24 h at 85% RH (Sonoda and Ogawa 1972).

Homothallic sexual reproduction occurs regularly in vegetative tissue, but oospore germination is fastidious, and evidence of oospore-initiated field infections is lacking (Chee and

Klein 1998; Coley-Smith 1962; Coley-Smith 1965; Gent et al. 2017; Skotland 1961). Diploid vegetative thalli produce oogonia and antheridia that combine to form a gametangium (Dick

2001). Meiosis occurs in the gametangia (Dick 2001). Karyogamy is believed to take place either in the oogonium after meiosis or in the oosphere after fertilization (Dick 2001). Chee and Klein

(1998) reported that oospore production in leaf disc assays was limited to low temperature (6 to

12° C). Gent et al. (2017) found that oospore production occurred regularly after 10-day incubation at 20° C in leaf disks inoculated with a single zoosporangium suggesting a homothallic mating system. Despite several conditioning attempts, oospore germination was unsuccessful, and viability was inferred based on microscopic observations of densely and

uniformly granular organization of oosporic cytoplasm and a plasmolysis assay (Gent et al.

2017).

In the field, sexual recombination is a source of genetic diversity (Milgroom and Peever

2015). Oospores are most often found at the end of the growing season in diseased leaves, shoots and especially cones (Royle and Kremheller 1981). Reports of oospores appear to differ based on production area (Skotland, 1961; Magie, 1942; Jones, 1934; Ware, 1926; Coley-Smith, 1962;). In the Yakima Valley, Washington, oospores were observed once in four years (Skotland, 1961) whereas in New York and England, oospore production was more readily observed (Jones, 1934;

Ware, 1926).

In an early population genetic study of P. humuli isolates from Washington and Oregon using random amplified polymorphic DNA and DNA amplification finger printing, higher levels of genetic variation was found in Oregon isolates and was attributed to sexual reproduction

(Chee et al. 2006). Gent et al. (2019) used genotyping-by-sequencing to characterize low genetic variation and linkage disequilibrium (non-random association of alleles at different loci) within populations of three Oregon hop yards located < 20 km apart, indicating high rates of inbreeding and/or asexual recombination. Genetic variation was significantly different among yards suggesting limited successful airborne migration from closely neighboring yards (Gent et al.

2019). The authors propose that a founder effect (Milgroom and Peever 2015) where populations that have already established systemic infections, either from infected planting material (Coley-

Smith 1962; Skotland 1961) or an overwhelming influx of airborne inoculum (Ojiambo et al.

2015), maintain the dominate genotype (Gent et al. 2019).

Disease Management

Cultural Controls. Cultural disease management practices include the removal or burial of primary inoculum through pruning practices (Gent et al. 2012) or hilling of soil on top of crowns mid-season (Mahaffee et al. 2009), stripping the leaves from the lowest part of the bines to reduce disease conducive conditions (Neve 1991), establishing new yard with disease-free planting material, and rouging diseased plants (Coley-Smith 1962; Skotland 1961). Pruning of spring foliage is an annual horticultural practice preformed to synchronize growth and optimize yield (Neve 1991). When the timing of pruning is delayed 5-21 days compared to grower standard practice there is some reduction in HDM severity in highly susceptible cultivars under high disease pressure (Gent et al. 2012). However, the quality of pruning, whether preformed mechanically or with a chemical desiccants (i.e. carfentrozone-ethyl), appears to be much more effective at lowering HDM disease incidence (Gent et al. 2012) by simply removing a majority of the infected basal shoots that provide a majority of the primary airborne inoculum. Pruning is typically not performed on young hops and pruning too late can negatively affect yield

(Mahaffee et al. 2009; O’Neal 2015).

A process called “stripping” typical occurs once bines have been trained to grow up the string (Neve 1991). A chemical desiccant is applied to the lower 1.5 m of leaves on the bine to remove foliage nearest to symptomatic basal shoots and increase airflow to promote drier conditions less conducive to disease (Neve 1991). However, improper stripping frequency and timing can reduce carbohydrate stores (Mahaffee et al. 2009). Around mid-season, growers may cultivate between rows deflecting excess soil up onto the base of the plant to promote the development of roots, rhizomes, and stolons (Mahaffee et al. 2009). The process is commonly

referred to as “hilling” or “hill-up” and can reduce HDM inoculum by inadvertently burying infected basal shoots (Mahaffee et al. 2009).

Disease-free planting material and sanitation efforts to rouge diseased plants from established hop yards (Coley-Smith 1962; Skotland 1961) or planting stock nurseries are cultural controls that require the accurate identification of diseased plants. In established hop yards, symptomatic basal shoots can help to identify chronically infected plants during the growing season (Coley-Smith 1962; Coley-Smith 1964; Skotland 1961; Ware 1926). In the early spring, prior to vegetative growth, when the previous year’s bines are mechanically pruned the exposed rootstock can be examined for disease symptoms (Coley-Smith 1962). However, in a nursery setting where propagated plants are rarely kept for longer than one-year, asymptomatic infection (see above) may go undetected. Additionally, propagated plants are regularly pruned to limit foliar growth and pruning may remove readily observed foliar symptoms originating from systemic infections.

Diagnostic Tests. DNA based methods can be employed as diagnostic assays to detect P. humuli (Gent et al. 2009; Patzak 2005; Summers et al. 2015). Polymerase chain reaction (PCR) based diagnostic assays for P. humuli were developed using the ITS region (Gent et al. 2009;

Patzak 2005). Yet, primers designed to amplify the ITS region also appear to lack the specificity needed to distinguish P. humuli from P. cubensis (Gent et al. 2009). Airborne outbreaks of cucurbit downy mildew are a widespread annual occurrence in Michigan’s lower peninsula

(Granke and Hausbeck 2011; Granke et al. 2014; Naegele et al. 2016) and ITS based assays are at risk of reporting false positives The mitochondrial based cox2 locus (Mitchell et al. 2011) was developed into a quantitative real-time PCR (qPCR) assay (Summers et al. 2015) to distinguish

P. humuli and P. cubensis, but the locked nucleic acid (LNA) probe used relies on the detection

of a single nucleotide polymorphism (SNP). Rahman et al. (2019) identified four highly polymorphic nuclear regions (c125015.3e1, c127446.1e1, c127233.5e3, and c126365.1e5) that appear to be unique to P. humuli and could be utilized for a diagnostic assay. It is unknown whether a single copy nuclear marker would have the same diagnostic sensitivity as a multiple copy mitochondrial marker (Rahman et al. 2017).

Resistance/Tolerance. Two main phenotypes that result from plant and pathogen interactions are resistance and tolerance (Pagán and García-Arenal 2020). While the difference between resistance and tolerance have historically been debated, the current view considers resistance to act directly on limiting pathogen multiplication; tolerance refers to the host’s ability to reduce the negative effects of infection (Pagán and García-Arenal 2020). It appears that all H. lupulus L. cultivars are susceptible to HDM but cultivars expressing low levels of disease are often termed “resistant” (Henning et al. 2018; Mahaffee et al. 2009; Woods and Gent 2016).

Differentiating between resistance and tolerance warrants attention because tolerance may reduce selection pressure on a pathogen population and may be a more stable phenotype than resistance

(Pagán and García-Arenal, 2020).

An inverse relationship between crown rot susceptibility and the incidence of symptomatic primary shoots complicates field assessments of HDM susceptibility (Neve 1991;

Woods and Gent 2016). Cultivars highly susceptible to rootstock rot may not exhibit foliar disease symptoms (Coley-Smith 1964; Neve 1991; Woods and Gent 2016). Dormant buds rapidly rot before producing symptomatic basal shoots, but uncolonized buds still produce healthy shoots (Neve 1991). While the rootstock of some susceptible hop cultivars succumb to

HDM rot, others seem to tolerate infection to produce a mix of healthy and diseased shoots

(Coley-Smith 1962; Coley-Smith 1964; Johnson and Anliker 1985; Skotland 1961; Woods

and Gent 2016). Woods and Gent (2016) proposed an indirect method to detect variation in susceptibility to rootstock rot by combined disease assessments of symptomatic HDM shoots

(incidence) with measurements of plant vigor (total number of shoots produced). High vigor cultivars with few symptomatic shoots were considered resistant. However, low vigor cultivars are difficult to classify since their growth rate might be a result of rootstock rot and susceptibility to HDM or poor adaptation of the cultivar to its environment (Woods and Gent 2016).

Breeding for resistance to HDM in diploid, dioecious H. lupulus has received limited domestic attention (Neve 1991) until recently (Henning et al. 2018; Henning et al. 2015, 2016;

Woods and Gent 2016). HDM resistance in H. lupulus appears to be under quantitative genetic control by multiple loci (Henning et al. 2015). The registration of an HDM-resistant male line

(USDA 21087M) offers new crossing opportunities with HDM-resistant female lines (Henning et al. 2018). However, disease ratings of USDA 21087M did not specify how resistance was determined and if disease ratings accounted for susceptibility to rootstock rot. Further, plant defense responses to P. humuli have not been characterized in hop. Defense responses observed in other host/downy mildew interactions, such as a hypersensitive reaction (HR) in the parenchyma (Mouzeyar et al. 1993), callose- and lignin-like encasement of hyphae and haustoria

(Cohen et al. 1989), lignification and peroxidase activity in the parenchyma (Asada and

Matsumoto 1969; Ohguchi and Asada 1975), or vascular based transcription factors that lead to an accumulation of defense signals (salicylic acid) and a resistance response (Yan et al. 2020), may act to limit systemic colonization in hop rootstock.

Fungicides. In the absence of resistant cultivars, fungicides are the main method of HDM control (Neve 1991; Woods and Gent 2016). In some cases, highly desirable brewing characteristics may incentivize planting HDM susceptible cultivars (Erin Lizotte, personal

communication). Site-specific fungicides including mefenoxam, cyazofamid, cymoxanil, and ametoctradin and the systemic fungicide fosetyl-Al are registered for use in Michigan for HDM control (Lizotte et al. 2018). In 2016, an emergency exemption from the U.S. Environmental

Protection Agency was granted to Michigan under section 18 of the Federal Insecticide,

Fungicide, and Rodenticide Act for temporary use of fluopicolide. Fluopicolide was registered in

Michigan in early 2019. Copper formulations (copper sulfate, copper hydroxide, copper oxychloride, and cuprous oxide) are the primary multi-site fungicide; folpet is also available

(Lizotte et al. 2018). The efficacy of these fungicides has been established for the PNW (Gent et al. 2020; Gent et al. 2015; Hunger and Horner 1982; Johnson and Anliker 1985; Massie et al.

2019; Nelson et al. 2011; Nelson and Grove 2006, 2007, 2008, 2009, 2010; Skotland and

Johnson 1983), but few evaluations have been performed in Eastern U.S. production areas

(Adams et al. 2018, 2019).

Phenylamide and phosphonate fungicides were the historic standards for HDM management in the PNW (Nelson et al. 2004) until regional resistance developed (Klein 1994;

Nelson et al. 2004) and became widespread (Gent et al. 2020; Gent et al. 2008; Marks and

Gevens 2019). Early trials with metalaxyl on a presumably sensitive pathogen population show a high degree of efficacy (Hunger and Horner 1982; Johnson and Anliker 1985). In Oregon, a single drench application of metalaxyl provided nearly 100% HDM control (Hunger and Horner

1982). In a semi-arid climate, a single metalaxyl application effectively reduced systemically infected shoots when applied as a banded foliar application directed at hop crowns with young shoots; injection of the fungicide into the soil with tractor driven shanks was less effective

(Johnson and Anliker 1985). While soil injected applications were not statistically different from

the untreated control, metalaxyl applied through a drip irrigation system more effective than the untreated control (Johnson and Anliker 1985).

Early reports of a single fosetyl-Al (analogous with Efosite Al) application for HDM control in Oregon showed only 56.9% control (Hunger and Horner 1982). Phosphates are still used heavily in the PNW for HDM control at increasingly higher rates via a Special Local Needs registration (Gent et al. 2020). Currently, applications of fosetyl-Al at five times (11.2 kg/ha) the

Michigan labeled rate (2.24 kg of active ingredient) applied at 14-day intervals are needed for disease control (74.7%) (Gent et al. 2020).

Carboxylic acid amides (CAA) (mandipropamid and dimethomorph) and quinone inside inhibitor (QiI) (cyazofamid) fungicides limit HDM in Washington (Massie et al. 2019; Nelson and Grove 2006, 2007, 2008, 2009, 2010). In Oregon, dimethomorph provided 100% disease control under greenhouse conditions (applied 1 day pre-infection) and approximately 80% disease control in field experiments (applied up to 7 days pre-infection) but efficacy was reduced when applied post-infection (Gent et al. 2015). In single-season fungicide evaluations in North

Carolina, either mandipropamid (Revus 2.08SC; 584.6 ml/ha) or ametoctradin + dimethomorph

(Zampro 4.33 SC; 1.0 liter/ha) were among the most effective fungicide at limiting HDM severity (Adams et al. 2018, 2019). Yet, mandipropamid-treated plots had 42% foliar necrosis

(Adams et al. 2018). In North Carolina, cyazofamid (Ranman 400SC; 153.4 ml/ha) was among the most effective fungicides (Adams et al. 2018).

Evaluations of fluopicolide for control of HDM are limited (Adams et al. 2018; Gent

2017a). Fluopicolide (Presidio 4SC; 292.3 ml/ha) reduced HDM foliar disease severity compared to an untreated control under moderate to high disease pressure (Adams et al. 2018). In P. cubensis, fluopicolide field control failures were reported in cucumber field trials five to six

years after the fungicide was registered (Adams and Quesada-Ocampo 2014; Hausbeck and

Linderman 2014; Langston and Sanders 2013).

Oxathiapiprolin is not registered for HDM in the U.S. Oxathiapiprolin + mandipropamid

(Orondis Ultra) at low (401.9 ml/ha) and high (584.6 ml/ha) rates reduced spike density compared to the untreated control (Gent 2017b). In North Carolina, oxathiapiprolin (Orondis

Opti A 0.83OD; 0.3 fl oz) and oxathiapiprolin + chlorothalonil (Orondis Opti 0.83 SC; 146.1 ml/ha [chlorothalonil is not registered for use on hop]) reduced foliar disease severity compared to the untreated (Adams et al. 2018, 2019) but oxathiapiprolin + chlorothalonil was less effective than ametoctradin + dimethomorph and fosetyl-Al (Adams et al. 2019).

The fungicide resistance action committee (FRAC) recommends no fewer than three and preferably five modes of actions to manage pathogen resistance to fungicides (Kuck et al. 2012).

Insensitivity to phenylamide (metalaxyl and mefenoxam) and phosphonate (fosetyl-Al) fungicides already occurs in some hop-growing regions (Gent et al. 2020; Gent et al. 2008;

Klein 1994; Marks and Gevens 2019; Nelson et al. 2004). P. humuli isolates collected from symptomatic HDM shoots in Washington were roughly 30 to 1,000 times more sensitive to metalaxyl than isolates collected from Oregon (Klein 1994). A follow up study confirmed that metalaxyl insensitivity persists in Oregon and had increased in Washington despite limited use of the fungicide in these areas over the past two decades (Gent et al. 2008). The presumed absence of a fitness cost associated with phenylamide insensitivity in P. humuli (Gent et al. 2008) limits the effectiveness of these fungicides in alternation or in mixtures as resistance management strategies (van den Bosch and Gilligan 2008).

The mode of action for phosphonates is currently classified as host plant defense induction (FRAC code© list 2020 [see https://www.frac.info]) but the primary target appears to

be the pathogen (Guest and Grant 1991). In hop plants treated with fosetyl-Al, higher levels of

HDM control are achieved when the fungicide is applied at inoculation or post-infection than when applied prophylactically (Gent et al. 2015). Phosphonate-treated plants invoke plant defense responses such as phytoalexin accumulation (Guest 1984) and superoxide release

(Daniel and Guest 2006) and resistance genes are expressed even in the absence of the pathogen

(Dann and McLeod 2020). However, low concentrations of the fungicide can alter oomycete metabolism by out competing phosphate as a precursor to the pentose phosphate pathway

(Barchietto et al. 1992) and release elicitors from sensitive pathogens (Perez et al. 1995) that may also trigger plant defense responses. Further, at high concentrations, fosetyl-Al can have fungicidal or fungistatic effects (Fenn and Coffey 1984) but this may be confounded in planta by increased levels of the toxic byproducts ethanol and Al3+ that also occur when fosetyl-al breaks down to phosphonate in the plant (Guest and Grant 1991). In P. humuli, fosetyl-Al toxicity appears to select for an accumulation of higher rates of phosphonate insensitivity in pathogen populations over time (Gent et al. 2020; Nelson et al. 2004). Yet, high rates of fosetyl-

Al (5X the Michigan labeled rate) are recommended in the PNW (Gent et al. 2020) despite achieving only moderate disease control and an expansion of insensitive pathogen populations. A phosphonate induced defense response has never been characterized for H. lupulus. In hop production areas without long term exposure to phosphonates, characterizing the phosphonate sensitivity of P. humuli populations in combination with an evaluation of plant defense response may help to optimize the effective dose and efficacy of phosphonate fungicides.

Insensitivity to carboxylic acid amides (CAA), benzamide, and cymoxanil (target site unknown) fungicide have been documented in other downy mildew pathogens (Blum et al. 2010;

Blum et al. 2011; Genet and Jaworska 2013; Gisi et al. 2007; Gullino et al. 1997; Pavelková et

al. 2014; Thomas et al. 2018; Toffolatti et al. 2015). CAA insensitivity is present in field isolates of P. cubensis (Blum et al. 2011; Pavelková et al. 2014) and Plasmopara viticola (Blum et al. 2010; Gisi et al. 2007). CAA sensitive oomycetes have a highly conserved glycine at position 1105 (G1105) in the cellulose synthase 3 (CesA3) gene (Blum et al. 2012). Fluopicolide insensitivity has been confirmed for P. cubensis (Thomas et al. 2018). Insensitivity to cymoxanil has been reported for the grape HDM pathogen (Genet and Jaworska 2013; Gullino et al. 1997;

Toffolatti et al. 2015). Future fungicide sensitivity monitoring in P. humuli population is needed to improve HDM management.

Disease forecasting models and surveillance of airborne sporangia have been tested to determine if the number of fungicide applications to manage HDM can be reduced (Gent et al.

2009; Gent et al. 2010). In Oregon, fungicide programs were initiated based on a modified GDD threshold used to predict the emergence of symptomatic basal shoots (Gent et al. 2010; Johnson

1991), disease risk index using weather data (Royle 1973), and plant growth stage (shoots approx.. 15 cm tall). Subsequent sprays were applied according to the disease risk index or a calendar-based program (Gent et al. 2010). Forecasting models reduced the number of fungicide applications by up to half compared to calendar-based programs (6 to 10 applications) while maintaining similar levels of disease control (Gent et al. 2010). However, there were instances where the fungicide programs failed to limit disease compared to the control (Gent et al. 2010).

Plot-to-plot variability in disease pressure (Gent et al. 2010) or the efficacy of fungicides used

(fosetyl-AL, cymoxanil, copper hydroxide, and trifloxystrobin + copper hydroxide) may have contributed to the lack of control. In fungicide programs initiated at the first detection of airborne sporangia, a single fungicide application was saved while maintaining similar levels of disease control compared to grower standard practices (Gent et al. 2009).

Humulus lupulus

Horticulture Practices. Humulus lupulus L. var lupulus is a diecious perennial climbing bine that belongs to the Cannabaceae family (Neve 1991). Clonal female plants are grown for their flowers that occur in inflorescences and develop enlarged bracts and bracteoles to produce the strobiles (referred to as cones) (Neve 1991). Peltate glandular trichomes, called lupulin glands, although found on leaves occur in abundance at the base of bracteoles (Neve 1991).

Essential oils (훼-humulene and 훽-myrcene) and terpenophenolic resins (prenylated acylphloroglucinols: humulones [훼 -acids] and lupulones [훽-acids]) biosynthesized in lupulin glands (Kavalier et al. 2011) are primarily used in beer production as a flavoring and bittering agent (Neve 1991; Van Opstaele et al. 2010) but are also of interest to the medical field for their biologically active compounds (Karabín et al. 2016).

The establishment of a new hop yard costs approximately $34,595/ha (Sirrine et al.

2014). Growers in new production regions, like Michigan, primarily source planting material from nurseries and established growers in other regions. Planting material raised in greenhouse nurseries is propagated from softwood cuttings (Howard 1965, 1967; Howard and Sykes 1966), but can also be generated through bedded sets (Neve 1991) and micropropagation (Gurriarán et al. 1999; Peredo et al. 2009). A small number of stock plants are needed for propagation by softwood cutting. The National Clean Plant Network (NCPN; nationalcleanplantnetwork.org) provides many propagators with dormant plants/rhizomes and currently offer unrooted green cuttings tested for major virus and the hop stunt viroid. However, NCPN but does not test for P. humuli or other filamentous pathogens (NCPN 2020). One- to two-node sections of softwood

(approx. 5-8 cm long) cut from stock plants are rooted under a mist system and later moved to the greenhouse floor or to an outdoor storage location (Howard 1965, 1967; Howard and Sykes

1966). Bedded sets are produced from one-node rhizomes or lateral runners (stolons) dug from the fields in the winter or early spring and planted into containers (Neve 1991). Propagated plants can remain in the nursery up to a year and are regularly pruned to limit foliar growth.

Planting occurs after the frost-free date and throughout the growing season (Mahaffee et al. 2009; O’Neal 2015). Hop yards are primarily planted by hand with 3 to 5 rhizomes or 1 to 4 plants per hop hill (O’Neal 2015). There are approximately 1,125 to 2,250 hills per hectare depending on planting pattern (Mahaffee et al. 2009). Plants are typically spaced 1.1 – 2.5 m between plants and 3.6 – 4.9 m between rows (O’Neal 2015). Many Michigan yards have been established at approx. 1.1 x 3.1 m (plant by row) spacing (Sirrine et al. 2014). A non-trellised hop yard used for HDM fungicide evaluation at the Washington State University was spaced at approx. 1.1 x 3.1 m (plant by row) (Mark Nelson, personal communication). Commercial hop yards are typically productive for 10 to 20 years (Mahaffee et al. 2009).

In an established hop yard, the previous year’s growth is removed while plants are dormant. In Michigan, perennial shoots sprout in early to mid-April followed closely by leaf development (Lizotte et al. 2018). In most commercial production, hop is grown under a high trellis (5.5 m) system to support the climbing bine (Mahaffee et al. 2009). Strings are secured to the top trellis wire and base of each plant. Three to four bines are trained dextrorse, by hand, onto the training string (Neve 1991). Some fast-growing cultivars are mechanically or chemically pruned prior to training to synchronize shoot growth, optimize yield, and prevent bines from overgrowing the trellis (Neve 1991).

As a short-day plant, H. lupulus must first outgrow a juvenile stage before temperature and daylight can induce flowering at which time the vegetative growth is abruptly suppressed

(Thomas and Schwabe 1969). Flowering only occurs on lateral growing shoots (referred to as

side arms) produced at each node of the bine (Neve 1991). In Michigan, the elongation of the bine primarily occurs from May into early June; side arm growth begins in early June (Lizotte et al. 2018). Typically, in mature Michigan hopyards vegetative growth ceases and flowering begins in early- to mid-July (Lizotte et al. 2018). However, in first-year fields (referred to as

‘baby hops’), especially in short-season production areas such as Michigan and Oregon, hops grow vegetative throughout their initial season (O’Neal 2015).

Nitrogen (N) uptake is closely related to biomass accumulation and all N fertilization is typically delivered in single or split applications (Gingrich et al. 1994) or weekly applications through drip irrigation (Iskra et al. 2019) during bine elongation (Gingrich et al. 1994; Iskra et al. 2019). Banded applications (42 to 112 kg/ha) of ammonium thiosulfate and monoammonium phosphate can be applied just prior to shoot emergence to provide the initial nutrients (Iskra et al.

2019) and dry granule N is incorporated into the soil at planting (O’Neal 2015). The amount of N required is 168 to 224 kg/ha; 84 kg/ha for baby hops (Gingrich et al. 1994). Excess N (269 kg/ha) does not improve yield and appears to have a slight negative impact on α- and β-acids and total oils in harvested cones (Iskra et al. 2019). Phosphorus (P) and potassium (K) are applied when soil tests indicate less than 24 to 40 ppm and 200 to 300 ppm, respectively (Gingrich et al.

1994; Lizotte et al. 2018). Micronutrient (calcium, magnesium, manganese, zinc) amendments made be needed in soils with a pH less than 5.7 or greater than 7.5; sulfur and boron deficiencies are also possible (Gingrich et al. 1994).

Weeds can be managed through shallow cultivation and cover crops but are primarily controlled by herbicides (O’Neal 2015). Common Michigan weeds include annual and biennial broadleaves (i.e. burdock, pigweed, common lambsquarter, kochia, mustards, and ragweed), perennial and biennial broadleaves (e.g. bindweed, Canada thistle, horse nettle, perennial

sowthistle), grasses (foxtail, quackgrass), and sedges (i.e. nutsedge) (O’Neal 2015) . Herbicide can be applied preemergence to new plantings (pendimethalin and glyphosate) and established yards (trifluralin, pendimethalin, norflurazon, flumioxazin, and indaziflam) (Lizotte et al. 2018).

Rhizomes can be planted before preemergence herbicide application, but transplants must be planted after application (O’Neal 2015). The removal of early shoots for pruning purposes with a chemical desiccant or mechanically (see above) provides some weed control (Mahaffee et al.

2009). Postemergence herbicide options include 2,4-D, clopyralid, carfentrazone, glyphosate and pelargonic acid, and clethodim (Lizotte et al. 2018).

Common insect pests include the two-spotted spider mite (TSSM) (Tetranychus urticae

Koch), damsel-hop aphid (Phorodon humuli Schrank), hop looper (Hypena humuli Harris) root weevils (Otiorhynchus spp.), potato leaf hopper (PLH) (Empoasca fabae Harris), and Japanese beetle (Popillia japonicus Newman) (Mahaffee et al. 2009; O’Neal 2015). TSSM overwinters in the soil as adult females and lay eggs early in the season (Mahaffee et al. 2009). Economic damage occurs because of outbreaks incited by warm-hot weather (Mahaffee et al. 2009).

Management of TSSM is typically achieved through chemical control including bifenazate, fenpyroximate, hexythiazox, spirodiclofen, abamectin, and bifenthrin (Lizotte et al. 2018;

O’Neal 2015). However, TSSM collected in Washington hop yards where moderate to highly sensitive to abamectin and bifenazate and insensitive TSSM populations could results in potential control failures (Piraneo et al. 2015). PLH is primarily an issue for Michigan and other eastern U.S. production region (O’Neal 2015). PLHs do not overwinter in Michigan but instead are carried in from the south on spring storms (O’Neal 2015). Chemical control options include imidacloprid, spirotetramat, and thiamethoxam; beta-cyfluthrin and bifenthrin are used as a last resort because they may incite TSSMs (O’Neal 2015).

After flowering, cone development occurs in five stages and can be recognized in the field by the size of the bracteoles (Kavalier et al. 2011). In stages I and II bracteoles enclose the ovary. Bracteoles are half the length of bracts at Stage III; terpenophenolic secretions are just beginning to fill glandular trichomes. In stage IV, the bracteoles are over half the length of bracts and the ovaries are senesced and brown. In the final stage (V), the bracteoles have fully elongated and are almost equal to the length of the bracts; glandular trichomes are filled with terpenophenolic secretions (Kavalier et al. 2011). Michigan yards are harvested from mid-

August into early October (Lizotte 2015).

To harvest hops, bines with cones attached are cut at their base, pulled or cut free from the top wire, and transported to a stationary picking machine (Mahaffee et al. 2009; Neve 1991).

In Michigan, horizontal hop pickers are most common. Bines are pulled horizontally between wire loops with springs attached to rotating drums (Neve 1991). The cones are stripped from the bines. The stripped bines are chopped and blown with forced air out of the picker. A pre- separation step of cones and debris is preformed using coarse mesh conveyors and spaced rollers that allow the hops to fall and the leaf/bines to be carried away. The cones are dropped down an inclined upward traveling belt while a cleaning fan blows off leaf material (Neve 1991). A final conveyor belt takes the cleaned cones to a forced-air kilns where they are dried for 4 to10 h at 50 to 70° C to a moisture content of 8-12% (Mahaffee et al. 2009). Improper drying through overheating or poor air circulation can result in either reduced alpha acid contents or poor cone color and aroma (Mahaffee et al. 2009). Cooled cones are packed into bales and placed into cold storage (Mahaffee et al. 2009). Hops may also be pelletized as value-added process to better preserve aroma and alpha acid content during storage (Sirrine et al. 2014).

Other pathogens. In August 2018, two Michigan commercial hopyards reported a high incidence of necrotic leaf lesions on ‘Chinook’, ‘Centennial’, and ‘Crystal’ (Erin Lizotte, personal communication). Post-harvest, growers noted a reduction in yield due to cone shatter in affected yards. An examination of samples revealed the presence of pycnidia in leaf lesions. An additional research objective on this dissertation is to determine the causal agent of foliar blight and elucidate the etiology of cone blight on hops in Michigan.

Several fungi associated with hop produce pycnidia in various tissue types (Putto et al.

1975 Gent and Radišek 2009; Radišek et al. 2008 Singh et al. 1984 Twomey et al. 2016 Gent et al. 2013; McGee et al. 2009 Allan-Perkins et al. 2020 Boerema et al. 2004; Phalip et al. 2006).

Didymellaceae pathogens including Septoria humuli Westend. (Putto et al. 1975), Phoma exigua var. exigua (Desm.) Aveskamp, Gruyter & Verkley (2010) (Gent and Radišek 2009; Radišek et al. 2008), and Ascochyta humuli Kabát & Bubák (Singh et al. 1984) produce pycnidia in aboveground hop tissue. Diplodia seriata De Not. the causal agent of black wilt produces pycnidia in hop stems (Twomey et al. 2016). Phomopsis tuberivora H.T. Güssow & W.R. Foster

(analogous with Phacidiopycnis tuberivora (H.T. Güssow & W.R. Foster) B. Sutton (1980) produces pycnidia, but the pathogen appears to be limited to tissue below ground (Gent et al.

2013; McGee et al. 2009). Diaporthe humulicola is a new pathogen recently recovered from hop leaves and cones in two research hop yards in Connecticut (Allan-Perkins et al. 2020). Although the cucumber pathogen Stagonosporopsis cucurbitacearum (Fr. : Fr.) Aveskamp, Gruyter &

Verkley (2010) (Aveskamp et al. 2010), Phoma glomerate (Corda) Qian Chen & L. Cai (2015), an endophytic mycoparasite of the powdery mildew pathogen (Sullivan and White 2000), and the opportunistic parasite Phoma aliena (Fr. : Fr.) Aa & Boerema (Boerema et al. 2004) form

pycnidia and have been isolated from diseased hop tissue with pycnidia (Boerema et al. 2004;

Phalip et al. 2006), they are not reported as pathogens of hop (Mahaffee et al. 2009).

LITERATURE CITED

Adams, M. L., and Quesada-Ocampo, L. M. 2014. Evaluation of fungicides for control of downy mildew on cucumber, Kingston 2013. Plant Dis. Manag. Rep 8:V240. Online.

Adams, M. L., Collins, H., and Quesada-Ocampo, L. M. 2018. Evaluation of fungicides for control of downy mildew on hop, Clayton 2017. Plant Dis. Manag. Rep. 12:PF082. Online.

Adams, M. L., Collins, H., and Quesada-Ocampo, L. M. 2019. Evaluation of fungicides for control of downy mildew on hop, Clayton 2018. Plant Dis. Manag. Rep. 13:V066. Online.

Asada, Y., and Matsumoto, I. 1969. Formation of lignin-like substance in the root tissues of Japanese raddish plant infected by downy mildew fungus. Jpn. J. Phytopathol. 35:160- 167.

Barchietto, T., Saindrenan, P., and Bompeix, G. 1992. Physiological responses of Phytophthora citrophthora to a subinhibitory concentration of phosphonate. Pestic. Biochem. Physiol. 42:151-166.

Beakes, G. W., Honda, D., and Thines, M. 2014. 3 Systematics of the Straminipila: Labyrinthulomycota, Hyphochytriomycota, and Oomycota. Pages 39-97 in: McLaughlin D., Spatafora J. (eds) Systematics and Evolution. The Mycota (A Comprehensive Treatise on Fungi as Experimental Systems for Basic and Applied Research), vol 7A. Springer, Berlin, Heidelberg.

Blum, M., Waldner, M., and Gisi, U. 2010. A single point mutation in the novel PvCesA3 gene confers resistance to the carboxylic acid amide fungicide mandipropamid in Plasmopara viticola. Fungal Genet. and Biol. 47:499-510.

Blum, M., Gamper, H. A., Waldner, M., Sierotzki, H., and Gisi, U. 2012. The cellulose synthase 3 (CesA3) gene of oomycetes: structure, phylogeny and influence on sensitivity to carboxylic acid amide (CAA) fungicides. Fungal Biol. 116:529-542.

Blum, M., Waldner, M., Olaya, G., Cohen, Y., Gisi, U., and Sierotzki, H. 2011. Resistance mechanism to carboxylic acid amide fungicides in the cucurbit downy mildew pathogen Pseudoperonospora cubensis. Pest Manag. Sci. 67:1211-1214.

Chee, H. Y., and Klein, R. E. 1998. Laboratory production of oospores in Pseudoperonospora humuli. Korean J. Plant Pathol. 14:618-621.

Chee, H. Y., Nelson, M. E., Grove, G. G., Eastwell, K. C., Kenny, S. T., and Klein, R. E. 2006. Population biology of Pseudoperonospora humuli in Oregon and Washington. Plant Dis. 90:1283-1286.

Choi, Y.-J., Hong, S.-B., and Shin, H.-D. 2005. A re-consideration of Pseudoperonospora cubensis and P. humuli based on molecular and morphological data. Mycol. Res. 109:841-848.

Cohen, Y., Eyal, H., Hanania, J., and Malik, Z. 1989. Ultrastructure of Pseudoperonospora cubensis in muskmelon genotypes susceptible and resistant to downy mildew. Physiol. Mol. Plant Pathol. 34:27-40.

Coley-Smith, J. R. 1962. Overwintering of hop downy mildew Pseudoperonospora humuli (Miy. and Tak.) Wilson. Ann. Appl. Biol. 50:235-243.

Coley-Smith, J. R. 1964. Persistence and identification of downy mildew Pseudoperonospora humuli (Miy. and Tak.) Wilson in hop rootstocks. Ann. Appl. Biol. 53:129-132.

Coley-Smith, J. R. 1965. Infection of hop rootstocks by downy mildew Pseudoperonospora humuli (Miy. & Tak.) Wilson and its control by early-season dusts. Ann. Appl. Biol. 56:381-388.

Constantinescu, O. 2000. The fine structure of the sporangium in Pseudoperonospora humuli (Chromista, Oomycota, Peronosporales). Cryptogam. Mycol. 21:93-101.

Daniel, R., and Guest, D. 2006. Defence responses induced by potassium phosphonate in Phytophthora palmivora-challenged Arabidopsis thaliana. Physiol. Mol. Plant Pathol. 67:194-201.

Dann, E., and McLeod, A. 2020. Phosphonic acid: A long‐standing and versatile crop protectant. Pest Management Science. Online (25 Oct): DOI 10.1002/ps.6156

Davis, J. 1910. A new hop mildew. Science, 31, 752.

Dick, M. 2001. Straminipilous Fungi: Systematics of the Peronosporomycetes Including Accounts of the Marine Straminipilous Protists, the Plasmodiophorids and Similar Organisms. Kluwer Academic Publishers, Dordrecht, Netherlands.

Fenn, M. A., and Coffey, M. 1984. Studies on the in vitro and in vivo antifungal activity of fosetyl-Al and phosphorous acid. Phytopathology 74:606-611.

Fraymouth, J. 1956. Haustoria of the Peronosporales. Trans. Br. Mycol. Soc. 39:79-107.

Genet, J.-L., and Jaworska, G. 2013. Characterization of European Plasmopara viticola isolates with reduced sensitivity to cymoxanil. Eur. J. Plant Pathol. 135:383-393.

Gent, D. 2017a. Evaluation of fungicides for hop downy mildew, Woodburn, Oregon, 2016. Plant Dis. Manag. Rep. 11:V104. Online.

Gent, D. H. 2017b. Evaluation of fungicides for hop downy mildew, Hubbard, Oregon, 2016. Plant Dis. Manag. Rep. 11:V048. Online.

Gent, D. H., and Ocamb, C. M. 2009. Predicting Infection Risk of Hop by Pseudoperonspora humuli. Phytopathology 99:1190-1198.

Gent, D. H., Nelson, M. E., and Grove, G. G. 2008. Persistence of phenylamide insensitivity in Pseudoperonospora humuli. Plant Dis. 92:463-468.

Gent, D. H., Ocamb, C. M., and Farnsworth, J. L. 2010. Forecasting and management of hop downy mildew. Plant Dis. 94:425-431.

Gent, D. H., Cohen, Y., and Runge, F. 2017. Homothallism in Pseudoperonospora humuli. Plant Pathol. 66:1508-1516.

Gent, D. H., Block, M., and Claassen, B. J. 2020. High Levels of insensitivity to phosphonate fungicides in Pseudoperonospora humuli. Plant Dis. 104:1400-1406.

Gent, D. H., Nelson, M. E., Farnsworth, J. L., and Grove, G. G. 2009. PCR detection of Pseudoperonospora humuli in air samples from hop yards. Plant Pathol. 58:1081-1091.

Gent, D. H., Twomey, M. C., Wolfenbarger, S. N., and Woods, J. L. 2015. Pre-and postinfection activity of fungicides in control of hop downy mildew. Plant Dis. 99:858-865.

Gent, D. H., Adair, N., Knaus, B. J., and Grünwald, N. J. 2019. Genotyping-by-sequencing reveals fine-scale differentiation in populations of Pseudoperonospora humuli. Phytopathology 109:1801-1810.

Gent, D. H., Mueller-Warrant, G., Woods, J. L., Putnam, M. L., and Twomey, M. C. 2013. Red crown rot of hop in oregon caused by Phomopsis tuberivora. Plant Health Prog. 14:52.

Gent, D. H., Nelson, M. E., Grove, G. G., Mahaffee, W. F., Turechek, W. W., and Woods, J. L. 2012. Association of spring pruning practices with severity of powdery mildew and downy mildew on hop. Plant Dis. 96:1343-1351.

Gingrich, C., Hart, J., and Christensen, N. 1994. Hops fertilizer guide. OSU extension catalog. Oregon State University, Extension Service FG 79.

Gisi, U., Waldner, M., Kraus, N., Dubuis, P., and Sierotzki, H. 2007. Inheritance of resistance to carboxylic acid amide (CAA) fungicides in Plasmopara viticola. Plant Pathol. 56:199- 208.

Granke, L., and Hausbeck, M. 2011. Dynamics of Pseudoperonospora cubensis sporangia in commercial cucurbit fields in Michigan. Plant Dis. 95:1392-1400.

Granke, L., Morrice, J., and Hausbeck, M. 2014. Relationships between airborne Pseudoperonospora cubensis sporangia, environmental conditions, and cucumber downy mildew severity. Plant Dis. 98:674-681.

Guest, D. 1984. Modification of defense responses in tobacco and capsicum following treatment with Fosetyl-Al [Aluminium tris (o-ethyl phosphonate)]. Physiol. Plant Pathol. 25:125- 134.

Guest, D., and Grant, B. 1991. The complex action of phosphonates as antifungal agents. Biol. Rev. 66:159-187.

Gullino, M., Mescalchin, E., and Mezzalama, M. 1997. Sensitivity to cymoxanil in populations of Plasmopara viticola in northern Italy. Plant Pathol. 46:729-736.

Gurriarán, M., Revilla, M., and Tamés, R. 1999. Adventitious shoot regeneration in cultures of Humulus lupulus L.(hop) cvs. Brewers Gold and Nugget. Plant Cell Rep. 18:1007-1011.

Hall, G. 1996. Modern approaches to species concepts in downy mildews. Plant Pathol. 45:1009- 1026.

Hausbeck, M. K., and Linderman, S. 2014. Evaluation of fungicides for control of downy mildew of cucumber, 2013. Plant Dis. Manag. Rep. 12:OT022. Online.

Henning, J. A., Gent, D., Townsend, M. S., and Haunold, A. 2018. Registration of downy mildew–resistant male hop germplasm USDA 21087M. J. Plant Regist. 12:379-381.

Henning, J. A., Gent, D. H., Twomey, M. C., Townsend, M. S., Pitra, N. J., and Matthews, P. D. 2015. Precision QTL mapping of downy mildew resistance in hop (Humulus lupulus L.). Euphytica 202:487-498.

Henning, J. A., Gent, D. H., Twomey, M. C., Townsend, M. S., Pitra, N. J., and Matthews, P. D. 2016. Genotyping-by-sequencing of a bi-parental mapping population segregating for downy mildew resistance in hop (Humulus lupulus L.). Euphytica 208:545-559.

Hoerner, G. 1932. Downy Mildew of Hops. Oregon State Agriculture College Extension Service, Bulletin 440.

Hoerner, G. 1940. The infection capabilities of hop downy mildew. J Agric. Res. 61:331.

Holmes, G. J., Ojiambo, P. S., Hausbeck, M. K., Quesada-Ocampo, L., and Keinath, A. P. 2015. Resurgence of cucurbit downy mildew in the United States: a watershed event for research and extension. Plant Dis. 99:428-441.

Howard, B. 1965. Regeneration of the hop plant (Humulus lupulus L.) from softwood cuttings. I. The cutting and its rooting environment. J. Hortic. Sci. 40:181-191.

Howard, B. 1967. Regeneration of the hop plant (Humulus Lupulus L.) from softwood cuttings: III. Treatment with root-promoting substances. J. Hortic. Sci. 42:105-107.

Howard, B., and Sykes, J. 1966. Regeneration of the hop plant (Humulus lupulus L.) from softwood cuttings. II. Modification of the carbohydrate resources within the cutting. J. Hortic. Sci. 41:155-163.

Hunger, R. M., and Horner, C. E. 1982. Control of hop downy mildew with systemic fungicides. Plant Dis. 66:1157-1159.

Iskra, A. E., Lafontaine, S. R., Trippe, K. M., Massie, S. T., Phillips, C. L., Twomey, M. C., Shellhammer, T. H., and Gent, D. H. 2019. Influence of nitrogen fertility practices on hop cone quality. J. Am. Soc. Brew. Chem. 77:199-209.

Johnson, D. 1991. Two degree-day models for predicting initial emergence of hop shoots systemically infected with Pseudoperonospora humuli. Plant Dis. 75:285-287.

Johnson, D., and Anliker, W. 1985. Effect of downy mildew epidemics on the seasonal carryover of initial inoculum in hop yards. Plant Dis. 69:140-142.

Johnson, D. A., and Skotland, C. 1985. Effects of temperature and relative humidity on sporangium production of Pseudoperonospora humuli on hop. Phytopathology 75:127- 129.

Johnson, D. A., Alldredge, J. R., and Allen, J. R. 1994. Weather and downy mildew epidemics of hop in Washington State. Phytopathology 84:524-527.

Karabín, M., Hudcová, T., Jelínek, L., and Dostálek, P. 2016. Biologically active compounds from hops and prospects for their use. Compr. Rev. Food Sci. Food Saf. 15:542-567.

Kavalier, A. R., Litt, A., Ma, C., Pitra, N. J., Coles, M. C., Kennelly, E. J., and Matthews, P. D. 2011. Phytochemical and morphological characterization of hop (Humulus lupulus L.) cones over five developmental stages using high performance liquid chromatography coupled to time-of-flight mass spectrometry, ultrahigh performance liquid chromatography photodiode array detection, and light microscopy techniques. J Agric. Food Chem. 59:4783-4793.

Klein, R. 1994. Occurrence and incidence of metalaxyl resistance in Pseudoperonospora humuli in the Pacific northwest. Plant Dis. 78:161-163.

Kuck, K. H., Leadbeater, A., and Gisi, U. 2012. FRAC Mode of Action Classification and Resistance Risk of Fungicides. Modern Crop Protection Compounds. W. Krämer, ed. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim:539-557.

Langston, D., and Sanders, F. 2013. Evaluation of fungicides for control of downy mildew on cucumber in Georgia I, 2012. Plant Dis. Manag. Rep. 7:V108: Online.

Lizotte, E. 2013. Michigan Hop Update, June 12, 2013. Michigan State University Extension. Online. https://www.canr.msu.edu/news/michigan_hop_update_june_12_2013.

Lizotte, E. 2014. Downy Mildew of Hops Already Reported in Michigan for 2014. Michigan State University Extension. Online. https://www.canr.msu.edu/news/downy_mildew_of_hops_already_reported_in_michigan _for_2014.

Lizotte, E. 2015. Management Options for Hop Yards with Downy Mildew. Michigan State University Extension. Online. https://www.canr.msu.edu/news/management_options_for_hopyards_with_downy_milde w.

Lizotte, E., Sirrine, R., Hausbeck, M. K., Jess, L., and Zandstra, B. 2018. Michigan Hop Management Guide, 2018. Michigan State University Extension, Online.

Magie, R. 1942. The epidemiology and control of downy mildew on hops. Tech. Bull. N. Y. State Agric. Exp. Stn. 267:1-48.

Mahaffee, W. F., Pethybridge, S. J., and Gent, D. H. 2009. Compendium of Hop Diseases and Pests. American Phytopathological Society (APS Press), St. Paul, MN.

Marks, M. E., and Gevens, A. J. 2019. Investigating phenylamide insensitivity in Wisconsin populations of Pseudoperonospora humuli. Plant Health Prog. 20:263-269.

Massie, S. T., Claassen, B. J., and Gent, D. H. 2019. Evaluation of fungicides for hop downy mildew and powdery mildew, Prosser, Washington, 2018. Plant Dis. Manag. Rep. 13:V027. Online.

Milgroom, M. G., and Peever, T. 2015. Population Biology of Plant Pathogens. American Phytopathological Society Press, St. Paul, MN.

Mitchell, M. N., Ocamb, C. M., Grünwald, N. J., Mancino, L. E., and Gent, D. H. 2011. Genetic and pathogenic relatedness of Pseudoperonospora cubensis and P. humuli. Phytopathology 101:805-818.

Mouzeyar, S., De Labrouhe, D. T., and Vear, F. 1993. Histopathological studies of resistance of sunflower (Helianthus annuus L.) to downy mildew (Plasmopara halstedii). J. Phytopathol. 139:289-297.

Naegele, R., Quesada-Ocampo, L., Kurjan, J., Saude, C., and Hausbeck, M. 2016. Regional and temporal population structure of Pseudoperonospora cubensis in Michigan and Ontario. Phytopathology 106:372-379.

NCPN. 2020. What is a clean plant? , Online: http://cpcnw.wsu.edu/what-is-a-clean-plant/.

Nelson, M. E., and Grove, G. G. 2006. Evaluation of fungicides for control of hop downy mildew, 2005. Fungic. Nematic. Tests 61:FC046. Online.

Nelson, M. E., and Grove, G. G. 2007. Evaluation of fungicides for control of hop downy mildew, 2006. Plant Dis. Manag. Rep. 1:FC025. Online.

Nelson, M. E., and Grove, G. G. 2008. Evaluation of fungicides for control of hop downy mildew, 2007. Plant Dis. Manag. Rep. 2:FC051. Online.

Nelson, M. E., and Grove, G. G. 2009. Evaluation of fungicides for control of hop downy mildew and powdery mildew, 2008. Plant Dis. Manag. Rep. 3:FC008. Online.

Nelson, M. E., and Grove, G. G. 2010. Evaluation of fungicides for control of hop downy mildew, 2009. Plant Dis. Manag. Rep. 5:FC002. Online.

Nelson, M. E., Garcia, M., and Grove, G. G. 2011. Evaluation of fungicides for control of hop downy mildew, 2010. Plant Dis. Manag. Rep. 5:FC076. Online.

Nelson, M. E., Eastwell, K. C., Grove, G. G., Barbour, J. D., Ocamb, C. M., and Alldredge, J. R. 2004. Sensitivity of Pseudoperonospora humuli (the causal agent of hop downy mildew) from Oregon, Idaho, and Washington to fosetyl-Al (Aliette). Plant Health Prog. 5:4.

Neve, R. A. 1991. Hops. Springer, Berlin, Germany.

O’Neal, S. 2015. Pest Management Strategic Plan for U.S. hops: Summary of a workshop held on November 4, 2014, Yakima, WA. Washington State University Irrigated Agriculture Research and Extension Center, Online: https://ipmdata.ipmcenters.org/documents/pmsps/US-hops-PMSP2015.pdf.

Ohguchi, T., and Asada, Y. 1975. Dehydrogenation polymerization products of p- hydroxycinnamyl alcohols by isoperoxidases obtained from downy mildew-infected roots of Japanese radish (Raphanus sativus). Physiol. Plant Pathol. 5:183-192.

Ojiambo, P. S., Gent, D. H., Quesada-Ocampo, L. M., Hausbeck, M. K., and Holmes, G. J. 2015. Epidemiology and population biology of Pseudoperonospora cubensis: A model system for management of downy mildews. Annu. Rev. Phytopathol. 53:223-246.

Pagán, I., and García-Arenal, F. 2020. Tolerance of plants to pathogens: a unifying view. Annu. Rev. Phytopathol. 58:77-96.

Patzak, J. 2005. PCR detection of Pseudoperonospora humuli and Podosphaera macularis (Humulus lupulus). Plant Prot. Sci. 41:141.

Pavelková, J., Lebeda, A., and Sedláková, B. 2014. Efficacy of fosetyl-Al, propamocarb, dimethomorph, cymoxanil, metalaxyl and metalaxyl-M in Czech Pseudoperonospora cubensis populations during the years 2005 through 2010. Crop Prot. 60:9-19.

Peredo, E. L., Arroyo-Garcia, R., and Revilla, M. Á. 2009. Epigenetic changes detected in micropropagated hop plants. J. Plant Physiol. 166:1101-1111.

Pérez, B. A., Martínez, E., Noetinger, F., and Wright, E. R. 2009. Hop downy mildew caused by Pseudoperonospora humuli in Argentina. Plant Dis. 93:839-839.

Perez, V., Mamdouh, A., Huet, J.-C., Pernollet, J.-C., and Bompeix, G. 1995. Enhanced secretion of elicitins by Phytophthora fungi exposed to phosphonate. Cryptogam. Mycol. 16:191- 194.

Pethybridge, S., Nelson, M., and Wilson, R. 2003. Forecasting climate suitability of Australian hop-growing regions for establishment of hop powdery and downy mildews. Australas. Plant Pathol. 32:493-497.

Piraneo, T. G., Bull, J., Morales, M. A., Lavine, L. C., Walsh, D. B., and Zhu, F. 2015. Molecular mechanisms of Tetranychus urticae chemical adaptation in hop fields. Sci. Rep. 5:1-12.

Rahman, A., Miles, T. D., Martin, F. N., and Quesada-Ocampo, L. M. 2017. Molecular approaches for biosurveillance of the cucurbit downy mildew pathogen, Pseudoperonospora cubensis. Can. J. Plant Pathol. 39:282-296.

Rahman, A., Gongora-Castillo, E., Bowman, M. J., Childs, K. L., Gent, D. H., Martin, F. N., and Quesada-Ocampo, L. M. 2019. Genome Sequencing and Transcriptome Analysis of the Hop Downy Mildew Pathogen Pseudoperonospora humuli Reveal Species-Specific Genes for Molecular Detection. Phytopathology 109:1354-1366.

Royle, D. J. 1968a. Hop Black Root‐rot (Phytophthra citricola Sawada) in England. Plant Pathol. 17:36-40.

Royle, D. J. 1968b. Plant Pathology Section the Year's Work. Rep. Dep. Hop Res., Wye, 1967.

Royle, D. J. 1970. Infection periods in relation to the natural development of hop downy mildew (Pseudoperonospora humuli). Ann. Appl. Biol. 66:281-291.

Royle, D. J. 1973. Quantitative relationships between infection by the hop downy mildew pathogen, Pseudoperonospora humuli, and weather and inoculum factors. Ann. Appl. Biol. 73:19-30.

Royle, D. J., and Thomas, G. G. 1971a. Observations with the scanning electron microscope on the early stages of hop leaf infection by Pseudoperonospora humuli. Physiol. Plant Pathol. 1:345-349.

Royle, D. J., and Thomas, G. G. 1971b. The influence of stomatal opening on the infection of hop leaves by Pseudoperonospora humuli. Physiol. Plant Pathol. 1:329-343.

Royle, D. J., and Thomas, G. G. 1972. Analysis of relationships between weather factors and concentrations of airborne sporangia of Pseudoperonospora humuli. Trans. Br. Mycol. Soc. 58:79-89.

Royle, D. J., and Thomas, G. G. 1973. Factors affecting zoospore responses towards stomata in hop downy mildew (Pseudoperonospora humuli) including some comparisons with grapevine downy mildew (Plasmopara viticola). Physiol. Plant Pathol. 3:405-417.

Royle, D. J., and Kremheller, H. T. 1981. Downy Mildew of the Hop. in: The Downy Mildews. D. M. Spencer, ed. Academic Press Inc., New York, NY.

Runge, F., and Thines, M. 2011. Host matrix has major impact on the morphology of Pseudoperonospora cubensis. Eur. J. Plant Pathol. 129:147-156.

Runge, F., and Thines, M. 2012. Reevaluation of host specificity of the closely related species Pseudoperonospora humuli and P. cubensis. Plant Dis. 96:55-61.

Runge, F., Choi, Y.-J., and Thines, M. 2011. Phylogenetic investigations in the genus Pseudoperonospora reveal overlooked species and cryptic diversity in the P. cubensis species cluster. Eur. J. Plant Pathol. 129:135-146.

Salmon, E., and Wormald, H. 1923. Three new diseases of the hop. J. Minist. Agric. 30:430.

Salmon, E., and Ware, W. 1928. Inoculation experiments with the downy mildews of the hop and nettle (Pseudoperonospora humuli (Miy. et Taka.) Wils. and P. urticae (Lib.) Salmon et Ware). Ann. Appl. Biol. 15:352-370.

Shaw, C. 1981. Taxonomy and Evolution. Pages 17-29 in: The Downy Mildews. Academic Press, New York.

Sirrine, J. R., Lizotte, E., Brown-Rylewski, D. E., O’Brien, T., and Leach, A. 2014. Estimated Cost of Producing Hops in Michigan [E3236]. East Lansing: Michigan State University, MSU Extension.

Skotland, C. 1961. Infection of hop crowns and roots by Pseudoperonospora humuli and its relation to crown and root rot and overwinting of the pathogen. Phytopathology 51:241- 244.

Skotland, C., and Johnson, D. A. 1983. Control of downy mildew of hops. Plant Dis. 67:1183- 1186.

Sonoda, R., and Ogawa, J. 1972. Ecological factors limiting epidemics of hop downy mildew in arid climates. Hilgardia 41:457-473.

Sonoda, R. M., and Ogawa, J. M. 1970. Characteristics of Pseudoperonospora humuli sporangia produced in water drops on systemically infected hop leaves. Mycologia 62:1067-1069.

Summers, C. F., Adair, N. L., Gent, D. H., McGrath, M. T., and Smart, C. D. 2015. Pseudoperonospora cubensis and P. humuli detection using species-specific probes and high definition melt curve analysis. Can. J. Plant Pathol. 37:315-330.

Thines, M., and Choi, Y.-J. 2016. Evolution, diversity, and taxonomy of the Peronosporaceae, with focus on the genus Peronospora. Phytopathology 106:6-18.

Thines, M., Voglmayr, H., and Göker, M. 2009. Taxonomy and Phylogeny of the Downy Mildews (Peronosporaceae). Pages 47-75 in: K. Lamour, S. Kamoun (eds). Oomycete Genetics and Genomics: Diversity, Interactions, and Research Tools. John Wiley & Sons, Inc., Hoboken, New Jersey.

Thomas, A., Neufeld, K., Seebold, K., Braun, C., Schwarz, M., and Ojiambo, P. 2018. Resistance to fluopicolide and propamocarb and baseline sensitivity to ethaboxam among isolates of Pseudoperonospora cubensis from the eastern United States. Plant Dis. 102:1619-1626.

Thomas, G. G., and Schwabe, W. 1969. Factors controlling flowering in the hop (Humulus lupulus L.). Ann. Bot. 33:781-793.

Toffolatti, S. L., Venturini, G., Campia, P., Cirio, L., Bellotto, D., and Vercesi, A. 2015. Sensitivity to cymoxanil in Italian populations of Plasmopara viticola oospores. Pest Manag. Sci. 71:1182-1188. van den Bosch, F., and Gilligan, C. A. 2008. Models of fungicide resistance dynamics. Annu. Rev. Phytopathol. 46:123-147.

Van Opstaele, F., De Rouck, G., De Clippeleer, J., Aerts, G., and De Cooman, L. 2010. Analytical and sensory assessment of hoppy aroma and bitterness of conventionally hopped and advanced hopped pilsner beers. J. Inst. Brew. 116:445-458.

Wallace, E. C., D'Arcangelo, K. N., and Quesada-Ocampo, L. M. 2020. Population analyses reveal two host-adapted clades of Pseudoperonospora cubensis, the causal agent of cucurbit downy mildew, on commercial and wild cucurbits. Phytopathology Online (7 Jul) https://doi.org/10.1094/PHYTO-01-20-0009-R.

Ware, W. 1926. Pseudoperonospora humuli and its mycelial invasion of the host plant. Trans. Br. Mycol. Soc. 11:91-107.

Williams, I. H., Roberts, J. B., and Coley-Smith, J. R. 1961. Studies of the dormant phase of the hop (Humulus lupulus L.). Rep. Dept. Hop Res. Wye Coll. for 1960:48-58.

Wilson, G. W. 1914. Studies in North American Peronosporales—VI. Notes on miscellaneous species. Mycologia 6:192-210.

Woods, J. L., and Gent, D. H. 2016. Susceptibility of hop cultivars to downy mildew: associations with chemical characteristics and region of origin. Plant Health Prog. 17:42- 48.

Yan, S., Ning, K., Wang, Z., Liu, X., Zhong, Y., Ding, L., Zi, H., Cheng, Z., Li, X., and Shan, H. 2020. CsIVP functions in vasculature development and downy mildew resistance in cucumber. PLoS Biol. 18:e3000671.

Yang, M. Q., van Velzen, R., Bakker, F. T., Sattarian, A., Li, D. Z., and Yi, T. S. 2013. Molecular phylogenetics and character evolution of Cannabaceae. Taxon 62:473-485.

Yarwood, C. E. 1937. The relation of light to the diurnal cycle of sporulation of certain downy mildews. J Agric. Res. 54:365-373.

CHAPTER 2. FUNGICIDE EFFICACY AGAINST PSEUDOPERONOSPORA HUMULI AND POINT-MUTATIONS LINKED TO CARBOXYLIC ACID AMIDE (CAA) RESISTANCE IN MICHIGAN

Abstract

Hops have expanded as a niche crop in Michigan and other production areas in the eastern United States, but growers in these regions face annual downy mildew outbreaks incited by Pseudoperonospora humuli, exacerbated by frequent rainfall and high relative humidity. We evaluated the efficacy of foliar- and drench-applied fungicides against downy mildew and examined Michigan isolates for point mutations linked to carboxylic acid amide (CAA) resistance. Disease severity and density were assessed weekly in 2016 and 2017 in non-trellised research hop yards in Michigan. Area under the disease progress curve values for disease severity were significantly lower for plants treated with oxathiapiprolin, ametoctradin/dimethomorph, fluopicolide, cyazofamid, or mandipropamid (90.6 to 100% control) compared to those treated with fosetyl-Al (64.3 to 93.0% control) at both locations for both years. Drench treatments of fluopicolide and oxathiapiprolin/ mefenoxam reduced disease density and severity at both locations but were only moderately effective (76.4 to 91.5% control).

To assess CAA resistance, the CesA3 gene was aligned using reference downy mildew species and primers designed to amplify the 1105 and 1109 amino acids. Point mutations conferring

CAA resistance were not detected at these loci for sporangia from 42 symptomatic shoots collected from 11 commercial hop yards. These efficacy results for hop downy mildew are needed to guide disease recommendations in this expanding Michigan industry. The absence resistant genotypes indicate that Michigan growers can continue to utilize CAA-containing commercial fungicides as part of an overall downy mildew management program.

Introduction

Commercial hop production returned to Michigan in 2008 after a more than 150-year hiatus (Sirrine et al. 2014). The hop cones (female strobili) are harvested for their bittering

(alpha- and beta-acids) and aromatic (monoterpenoids and sesquiterpenoids) properties that are important components in beer production (Neve 1991). From 2015 to 2018, harvested hops increased from 130 to 304 ha in Michigan, making the state the fourth largest hops producer in the United States (U.S.) behind Washington, Oregon, and Idaho (George 2019). The crop has also increased in popularity in other states across the eastern U.S.; 514 ha of hops were harvested in 2018 from 19 states (George 2019).

Downy mildew (DM) caused by Pseudoperonospora humuli (Miyabe & Takah.) G.W.

Wilson, (1914) is a highly destructive pathogen of hop in Michigan (O’Neal 2015) and other growing regions (Johnson et al. 2009; O’Neal 2015; Royle and Kremheller 1981). Both above- and below-ground portions of the plant are susceptible to the disease. Crown rot caused by P. humuli may result in up to 100% losses for susceptible cultivars (Coley-Smith 1962; Skotland

1961; Woods and Gent 2016). Infection of apical buds causes the bines to fall off the training string, requiring additional labor to retrain new bines (Royle and Kremheller 1981). If the hop cones become diseased, the burrs may abort, or the quality and marketability reduced from the discoloration and decreased alpha-acid content (Royle and Kremheller 1981).

P. humuli overwinters as mycelia in the crown buds (Coley-Smith 1962; Skotland 1961).

In the spring, infected crown buds produce diseased basal shoots that initiate the polycyclic disease cycle (Coley-Smith 1962; Johnson and Anliker 1985; Skotland 1961). Symptomatic basal shoots are stunted, swollen, and chlorotic with downward cupped leaves (Ware 1926).

Localized leaf lesions are chlorotic and angular in shape (Johnson et al. 2009). Secondary

infections of apical meristems and lateral shoots produce symptoms similar to basal shoots but are distinguishable by one or more normally elongated internode at the shoot base (Coley-Smith

1962; Skotland 1961). The pathogen produces dark purple to black sporangia primarily on abaxial leaf surfaces and occasionally on stems and stipules (Salmon and Ware 1925; Ware

1926). Sporangia indirectly infect through stomata by releasing zoosporangia under wet conditions (Royle 1970; Royle and Thomas 1971; Royle and Thomas 1973). Homothallic reproduction occurs regularly in vegetative tissue, but oospore germination is fastidious and evidence of oospore-initiated field infections is lacking (Chee and Klein 1998; Coley-Smith

1962; Coley-Smith 1965; Gent et al. 2017; Skotland 1961).

Michigan experienced unusually wet weather during the 2015 growing season and DM developed in hop yards at levels not previously observed (Lizotte 2015). In 2016, the United

States Environmental Protection Agency granted Michigan an exemption under Section 18 of the

Federal Insecticide, Fungicide, and Rodenticide Act for the temporary use of fluopicolide

(Presidio, Valent USA Corp., Walnut Creek, CA; FRAC 43); an oomycete fungicide and mode of action not yet registered for the control of P. humuli on hop in Michigan. Plant losses as a result of DM infection ranged from 20 to 100%; some producers were unable to financially recover and no longer grow the crop (Erin Lizotte, personal communication). While management strategies for DM are established for hop growers in the Pacific Northwest (Gent et al. 2009;

Gent and Ocamb 2009; Gent et al. 2010; Gent et al. 2012b; Gent et al. 2015; Hunger and

Horner 1982; Johnson 1991; Johnson et al. 1994; Johnson and Anliker 1985; Johnson and

Skotland 1985; Skotland and Johnson 1983; Woods and Gent 2016), similar recommendations are lacking for Michigan and other eastern U.S. production areas where environmental conditions can be especially conducive to DM.

Site-specific fungicides including mefenoxam, fosetyl-Al, cyazofamid, cymoxanil, ametoctradin, are commonly used by Michigan hop growers for DM control (Lizotte et al. 2018).

Carboxylic acid amide (CAA) fungicides (mandipropamid and dimethomorph) limit the disease in the Pacific Northwest (Gent et al. 2015; Massie et al. 2019; Nelson et al. 2011; Nelson and

Grove 2006, 2007, 2008, 2009, 2010). Yet, CAA resistance has been documented among the

Peronosporales including field isolates of P. cubensis (Blum et al. 2011; Pavelková et al. 2014) and Plasmopara viticola (Blum et al. 2010; Gisi et al. 2007). CAA sensitive oomycetes have a highly conserved glycine at position 1105 (G1105) in the cellulose synthase 3 (CesA3) gene

(Blum et al. 2012). Additionally, a valine amino acid at position 1109 (V1109) in CesA3 is conserved among Peronosporales and is found primarily in Phytophthora spp. (i.e. P. infestans,

P. melonis and P. capsici) (Blum et al. 2012). Amino acid substitutions at these loci are linked to

CAA resistance in oomycetes (Blum et al. 2012). In P. cubensis, changes at position 1105 from glycine [GGG] to valine [GTG] or tryptophan [TGG] confer resistance to CAA in field isolates

(Blum et al. 2011; Blum et al. 2012). Conserved amino acid configurations in the CAA’s target site are linked to point mutations that confer CAA insensitivity and fungicide resistance in oomycete field populations and may be detected with a polymerase chain reaction (PCR) based assay (Blum et al. 2011; Blum et al. 2012).

To develop disease management strategies for Michigan hop growers, we evaluated the efficacy of foliar- and drench-applied fungicides and assessed CAA resistance by screening

Michigan isolates for point mutations linked to resistance (Blum et al. 2012). Preliminary reports of the efficacy portion of this research have been presented previously (Higgins and Hausbeck

2017, 2018).

Materials and Methods

Field plots and experimental design. All fungicide trials were conducted on non- trellised hop plants at the Michigan State University (MSU) Southwest Michigan Research and

Extension Center (SWM), Benton Harbor, MI (latitude 42.6854°, longitude -84.4716°) and replicated at the MSU Plant Pathology Farm (PPF), Lansing, MI (latitude 42.0835°, longitude -

86.3542°). Our non-trellised hop plot was similar to that used by Nelson and Grove (2004, 2006,

2007, 2008, 2009) and Nelson et al. (2010) in Prosser WA, with modifications. Plots were established in 52.3 m rows of raised plant beds covered with black plastic and buried drip irrigation. Due to space limitations at the PPF, treatment rows for the drench treatments

(described below) were 25.9 m long. All rows were spaced 2.4 m apart. Each treatment plot consisted of 10 plants in a row with 1.1 m between plants and a 0.8 m walkway between plots.

Hop plantlets (cv. Nugget), propagated from softwood cuttings and purchased from a commercial propagator, were transplanted on 3 (PPF) and 6 Jun (SWM) 2016. All experiments contained four plots per row except the PPF drench plot that had two plots per row. A single buffer row of untreated plants bordered each experiment. Treatments were arranged in a randomized complete block design with four replications.

Plots were drip-irrigated, as necessary. Fertilizer was applied as a combination of granular fertilizer 6-24-24 (N-P2O5) and urea 46-0-0 (N) prior to bed formation and MORA-

LEAF® Plus 20-20-20 (N - P2O5 - K2O) with micronutrients (Wilbur-Ellis, San Francisco, CA) and urea 46-0-0 (N) fertilizer dissolved in water and injected through the drip line. In 2016, plots received 28.7 kg N, 11.2 kg P and 11.2 kg K (PPF) and 33.1 kg N, 11.5 kg P and 11.5 kg K

(SWM). In 2017, MORA-LEAF® Plus 20-20-20 (N - P2O5 - K2O) with micronutrients (Wilbur-

Ellis, San Francisco, CA), Liquid Re-Nforce K® 5-0-20 (N-K2O) with sulfur (Loveland

Products, Inc., Greenly, CO) and liquid urea ammonium nitrate solution 28-0-0 (N) were used to provide 23.9 kg N, 1.7 kg P and 23.9 kg K (PPF) and 28.2 kg N, 5.9 kg P and 28.2 kg K (SWM) through the drip line. Interveinal leaf bronzing was observed on 9 Jun 2017 in the SWM plots.

Nutrient tests of petiole and soil samples revealed a K deficiency and fertilizer was adjusted to deliver 3.2 kg of K each week. Weeds were managed through mechanical cultivation. In 2017, hop shoots extending into the center aisles were periodically cut back, with a mechanical hedge trimmer (STIHL, Inc., Virginia Beach, VA) to a uniform distance outside of the rating area.

Two-spotted spider mites (Tetranychus urticae C. L. Koch) and potato leaf hoppers (Empoasca fabae Harris) were managed with applications of bifenazate, bifenthrin, fenazaquin, and imidacloprid.

All plots, except the drench plot at the PPF, were inoculated in 2016 on 15, 22, and 28

Jun (PPF), 21 Jun and 8 Jul (SWM). Due to a lack of available inoculum, the drench plot was not inoculated in 2016 (PPF) but was inoculated in 2017 on 19 and 26 May. Plots receiving inoculum in 2016 were not reinoculated in 2017 and disease progressed under natural conditions.

Inoculum was prepared from symptomatic shoots collected 24 h prior to inoculation from three commercial hop yards. To induce sporulation, symptomatic shoots were placed in a 50-ml beaker of water and enclosed overnight in plastic bags (Nelson and Grove 2004). Symptomatic shoots were then placed in 3.7 liters of water, shaken, and left in water for 1 h to promote zoospore release. The zoospore suspension was strained through two layers of cheese cloth, added to 15.1 liters of water and injected into overhead mist irrigation using a CO2 pressurized container

(Nelson and Grove 2004). The final concentration of zoospore suspension was 2.6 x 104 sporangia/ml for inoculations on 15 Jun (PPF) and 2.0 x 104 for inoculations on 21 Jun (SWM) and 22 Jun (PPF). Inoculum for inoculations on 28 Jun (PPF) and 7 Jul (SWM) was prepared

from 60 to 80 symptomatic shoots. Sixty symptomatic shoots with sporulation produced an average of 4.4 x 105 sporangia per ml as measured with a hemocytometer; 15.1 liters of approximately 5.2 x 106 to 6.1 x 106 sporangia/ml were injected into the overhead mist irrigation for each inoculation. Overhead mist irrigation was established in every other row and spaced to create conditions conducive for DM. Plots were misted with 16.8 to 130.0 mm of water for periods of 3 to 6 h or overnight in 2016 on 7, 11, 21, 23 Jul and 15 Aug (SWM) and on 24, 29

Jun and 11 Jul (PPF) to create extended durations of leaf wetness conducive to disease. The foliar (both locations) and drench trials (SWM) received inoculum from all three commercial yards. The drench trial at the PPF received inoculum from two of the commercial yards. In 2017, treatments were assigned to the same treatment plots established in 2016. Foliar treatments were initiated when the first symptomatic shoots were observed at the PPF and reapplied every 10 days.

Disease ratings. The inner six out of the ten plants per replication were assessed for disease severity and density. If newly-transplanted plants died prior to inoculation or did not overwinter for the 2017 assessment, one of the remaining four plants in the plot was assessed.

When plants within each plot began to grow together, a frame (1.2 x 0.9 m) constructed of PVC pipe (1.3 cm) was placed over the center of each plant. The corners of the frame were marked in the plot with wooden stakes to ensure subsequent measurements included the same area. Frames were used to rate all experiments from 18 (SWM) and 19 Aug (PPF) in 2016 and 29 May

(SWM) and 7 Jun (PPF) in 2017. Disease density was determined by the number of symptomatic shoots per plant (Gent et al. 2012a). Disease severity, measured as a combination of symptomatic shoots and leaf lesions, was estimated visually using the Horsfall-Barratt scale of 1 to 12, where

1=0% plant area diseased, 2=>0 to 3%, 3= >3 to 6%, 4=>6 to 12 %, 5 = >12 to 25 %, 6=>25 to

50%, 7=>50 to 75%, 8=>75 to 87%, 9=>87 to 94%, 10=>94 to <100%, 12=100% plant area diseased (Horsfall and Barrett, 1945). Prior to statistical analysis, rating values from the Horsfall-

Barratt scale were converted to midpoint values.

Foliar fungicide experiments. Eight fungicides were applied singly to plant foliage and compared to an untreated inoculated plot (control). Fungicide treatments included oxathiapiprolin (Orondis 100OD; Syngenta Crop Protection, Inc., Greensboro, NC; FRAC 49) applied at 401.9 ml/ha (2016) and 350.8 ml/ha (2017), ametoctradin/ dimethomorph (Zampro;

BASF Corp., Research Triangle Park, NC; FRAC 45/40) applied at 1023.1 ml/ha, fluopicolide

(Presidio; Valent USA Corp., Walnut Creek, CA; FRAC 43) applied at 292.3 ml/ha, cyazofamid

(Ranman 400SC; Summit Agro, Durham, NC; FRAC 21) applied at 201.0 ml/ha, mandipropamid (Revus; Syngenta Crop Protection, Inc., Greensboro, NC; FRAC 40) applied at

584.6 ml/ha, dimethomorph (Forum; BASF Corp., Research Triangle Park, NC; FRAC 40) applied at 438.5 ml/ha, cymoxanil (Curzate 60DF; Corteva Agriscience, Wilmington, DE; FRAC

27) applied at 224.2 g/ha, and fosetyl-Al (Aliette WDG; Bayer CropScience LP, Research

Triangle Park, NC; FRAC P 07) applied at 2802.1 g/ha. Fungicide rates were applied at the maximum rate allowed by the product label or according to manufacturer recommendations for fungicides not currently registered. All applications were made at 467.7 liters/ha except for cyazofamid that was applied at 935.4 liters/ha (a minimum requirement listed on the product’s label). Treatments were applied with a CO2 backpack sprayer and a boom equipped with two

8003 flat fan nozzles spaced 45.7 cm apart operating at 241.3 kPa. As the hop growth expanded, a third nozzle was added to the boom to ensure optimal coverage. In 2016, treatments were applied on 13 and 23 Jun; 3, 13 and 23 Jul; 2, 12, 22 and 30 Aug; 6, 17 and 27 Sep (PPF) and on

14 and 24 Jun; 4, 14 and 24 Jul; 3, 13, 23 and 31 Aug; 7, 18 and 27 Sep (SWM). In 2017,

treatments were applied on 26 Apr, 3, 13 and 24 May; 3, 15 and 25 Jun; 3, 13, 21 and 31 Jul; 11,

21 and 31 Aug; 7 Sep (PPF) and on 27 Apr, 8, 16 and 26 May; 3, 16 and 27 Jun; 4, 11, 18 and 25

Jul, 2 and 10 Aug (SWM). In 2016, disease severity and density ratings were assessed on 20 and

28 Jul; 4, 11, 18 and 25 Aug; 1, 8, 15 and 22 Sep; 6 Oct (SWM) and 22 and 29 Jul; 5, 12, 19 and

26 Aug; 2, 9, 16 and 23 Sep; 7 Oct (PPF). In 2017, disease ratings prior to fungicide applications occurred on 24 Apr (PPF) and 26 Apr (SWM). Disease assessments occurred on 10, 17, 25 and

31 May; 7, 14, 21 and 28 Jun; 5, 19 and 26 Jul; 2 Aug (PPF) and on 8, 15, 22 and 29 May; 5 Jun

(SWM). Additional disease ratings were taken at the SWM but were excluded from analysis due to K deficiency symptoms of the plants.

Disease assessment data were summarized for each treatment level by calculating the area under the disease progress curve (AUDPC) (days) using the trapezoidal method (Madden et al. 2007). Treatments in 2016 were applied to the same plots in 2017. Statistical analysis for

AUDPC was performed with a linear mixed model:

푦푖푗푘푠 = 휇 + 푎푘 + 푏푗 + 푐푖 + 푐푎푖푘 + 푎푏푗푘 + 푐푏푖푗 + +푒푖푗푘 where, 푦푖푗푘 is the vector of observed values of AUDPC; 휇 is the intercept; 푎푘 and 푐푖 are the fixed effects associated to 푘th year and 푖th treatment respectively; 푐푎푖푘 is the two-way interaction associated to the fixed effects. Random effects are all the effects underlined in the presented model. 푏푗 is the random effect of the jth block and 푎푏푗푘 and 푐푏푖푗, are the two-way interactions between the blocks and fixed effects; 푒푖푗푘 are the residuals. All random effects were assumed to follow 푁(0, 퐈휎2), where 휎2 is the variance component associated to the random effect, and 퐈 is an identity matrix. The model accounts for repeated measures using a compound symmetry covariance structure that includes all interactions with the time variable (year). Statistical analysis was preformed using PROC GLIMMIX in SAS software, Version 9.4 (SAS Institute,

Inc., Cary, NC). Normality was evaluated using residual plots. The fourth replicate of the cymoxanil treatment at the SWM in 2017 had 8 to 19.5 times higher AUDPC values than the highest ratings for the other cymoxanil replicates and this replicate was removed as an outlier.

After data were transformed, using the square root (SWM) or a lambda value (휆 = 0.25; 푦= ((푥휆-

1)/ 휆) obtained from a box-cox analysis (PPF), the distribution of residuals was approximately normal. The homogeneity of variances for transformed data was confirmed with the Levene’s test (P > 0.05). Back transformed data are presented in the table. Fisher’s protected least significant differences (LSD) at P < 0.05 were used to make simple effect comparisons of year × treatment least squares means by year.

Drench treatments. An untreated inoculated plot (control) was compared to drench treatments including: oxathiapiprolin/mefenoxam (Orondis Gold; Syngenta Crop Protection,

Inc., Greensboro, NC; FRAC 49/4) applied at 910.0 ml/ha, mefenoxam (Ridomil Gold SL;

Syngenta Crop Protection, Inc., Greensboro, NC; FRAC 4) applied at 702.1 ml/ha, and fluopicolide (Presidio; Valent USA Corp., Walnut Creek, CA; FRAC 43). Fungicides were applied at the maximum label rate and were directed to the base of each plant. All fungicides were applied at 468.4 liters/ha, with a CO2 backpack sprayer and a single nozzle boom equipped with 8010 flat fan nozzles operating at 89.6 kPa. In 2016, treatments were applied on 13 Jun, 13

Jul, and 12 Aug (PPF) and on 14 Jun, 14 Jul, and 13 Aug (SWM). In 2017, treatments were applied on 14 Apr, 10 May, 8 Jun and 12 Jul (PPF) and on 14 Apr, 15 May, 9 Jun, and 10 Jul

(SWM). In 2016, disease density and severity were assessed on 28 Jul; 4, 11, 18, 25 Aug; 1 Sep

(SWM). At the PPF, only trace levels of foliar symptoms were observed in the control and mefenoxam-treated plots (26 Aug) and were not assessed further. In 2017, disease ratings

occurred on 10, 17, 25 and 31 May; 7, 14, 21, and 28 Jun; 5, 19, and 26 Jul; 2 Aug (PPF) and 8,

15, 22, and 30 May; 5 Jun (SWM).

Data for the SWM were analyzed with a linear mixed model and a compound symmetry covariance structure as described above. After the SWM data were transformed, using a lambda value (휆 = 0.25) obtained from a box-cox analysis (PPF), the distribution of residuals was approximately normal. The homogeneity of variances for transformed data was confirmed with the Levene’s test (P > 0.05). Fisher’s protected LSD at P < 0.05 were used to make main effect comparisons of least squares means for the factor treatment. Data for the PPF were analyzed using a linear mixed model since this experiment was performed only in 2017. Treatment was a fixed effect and block a random effect. After the PPF data were transformed, using the log

(density) or lambda value (휆 = -0.75) obtained from a box-cox analysis (severity), distribution of residuals was approximately normal. The homogeneity of variances for transformed data was confirmed with the Levene’s test (P > 0.05). Back transformed data are presented in the table.

CAA point mutations. Sixty symptomatic primary and secondary symptomatic shoots were selected from the DM susceptible cultivars Nugget, Chinook, Cascade, Centennial,

Columbus, Cashmere, and an experimental cultivar. From 23 May to 6 Jun 2018, samples were collected across nine commercial and two research Michigan hop yards located in the northwest

(Benzie, n=1; Leelanau, n=2; and Grand Traverse Counties, n=2), central (Montcalm, n=1;

Ingham, n=1; Genesee Counties, n=1), and southwest (Berrien, n=3) regions. Gloves were changed between samples. Samples were placed individually in plastic storage bags and immediately placed into a cooler for transport to the lab. Symptomatic shoots with little to no sporulation were placed in a beaker of water and incubated overnight in a humidity chamber at

20°C to induce sporulation (Gent et al. 2008). Isolates were obtained from a single symptomatic

shoot by dislodging sporangia with pressurized, sterile water from an aerosol sprayer (Preval

Complete Spray Unit, Precision Valve Corporation, Yonkers, NY). The concentration of each isolate was quantified using a hemocytometer; counts were repeated once and averaged.

Samples, under agitation to ensure uniform concentration, were sub-divided into 2 ml tubes and frozen at -80°C for later DNA extraction.

DNA was extracted using a CTAB buffer followed by phenol:chloroform steps described in Sambrook et al. (1989) with slight modifications. Sporangial samples were thawed in a 65°C water bath for 1 min and pelletized at 15,000 rpm for 5 min. The supernatant was removed without disturbing the pellet. Five glass beads (3mm; MilliporeSigma, Burlington, MA) were added to each tube and samples were frozen in liquid nitrogen for 1 min. Samples were shaken in a mechanical homogenizer (TissueLyser II, Qiagen, Hilden, Germany) for 3 min 30 sec at 30 Hz.

Next, 750 μl of CTAB extraction buffer (Sambrook et al. 1989) was added to each sample.

Samples were sonicated for 10 min, placed in a 65°C water bath for 2 h, and again sonicated for

10 min. Samples were centrifuged for 10 min at 15,000 rpm. The supernatant (500 µl) was transferred to new 1.75 ml tubes. Proteins were denatured first in an equal volume phenol:chloroform (ChCl3):isoamyl alcohol (IAA) (25:24:1) and then in an equal volume

ChCl3:IAA (24:1). DNA was allowed to precipitate overnight in isopropanol at - 20°C, then washed in 70% ethanol, and resuspended in 50 µl of TE buffer. DNA was quantified with a

Qubit Fluorometer and the Qubit dsDNA HS Assay Kit (Invitrogen, Carlsbad, CA).

Sequences of the CesA3 gene from Blum et al. (2012) and Rahman et al. (2019) were aligned using Geneious Prime (Biomatters, Ltd., Auckland, New Zealand) and primers were designed for the C terminal end. Sequences were visualized utilizing Geneious Prime software for the following reference sequences for P. humuli (contig: NQFO01001774 from Rahman et

al., 2019), P. cubensis (JF799098) and Pl. viticola (GQ258975). A Transmembrane Prediction

Tool and PANTHER (Thomas et al. 2003) were utilized to characterize general differences between function, motifs, and subunits in these three species. Primers Ph_CesA3rA (5’-

GTAGTATACTTGCCACAGCTG-3’) and Ph_CesA3fA1 (5’-

GGAATTACGTGTCAGCAATGTT-3’) were designed to amplify the region of the CesA3 gene flanking the 1105 and 1109 amino acid positions (expected product size ~188 bp). PCR consisted of 1.25 µl Ph_CesA3rA (10 µM), 1.25 µl Ph_CesA3fA1 (10 µM), 0.5 µl DNTPs (10 mM), 5.0 µl colorless buffer, 0.25 µl GoTaq DNA polymerase, 14.75 µl DNase-free water, and 2

µl template DNA (5 ng/µL) in a 25-ml reaction. Amplification was performed on a Mastercycler pro thermal cycler (Eppendorf, Westbury, NY) with initial denaturation at 95°C for 2 min, followed by 35 cycles of 94°C for 1 min, 52°C for 1 min, and 72°C for 1 min, and a final extension at 72°C for 10 min. PCR products were submitted for Sanger sequencing at the MSU

RTSF Genomics Core (East Lansing, MI). Bi-directional sequence results were aligned in

Geneious Prime with sequences of the CesA3 gene from Blum et al. (2012) and Rahman et al.

(2019). Amino acid configurations of glycine and valine were examined for a point mutation at position 1105 and 1109, respectively.

Results

Foliar treatments. In 2016, two months after the first disease assessments, the control plots had 19.4 ± 3.0 standard error (SE) symptomatic shoots and 52.1% ± 6.7 SE severity (PPF) and 23.9 ± 3.1 SE symptomatic shoots and 77.9% ± 1.8 SE severity (SWM) (Fig. 1.1). In 2017, prior to fungicide treatments, control plots had <1 (PPF; 24 Apr) and 1.5 ± 0.4 SE (SWM; 26

Apr) symptomatic shoots per plant. On 5 and 7 Jun 2017, 13.3 ± 1.9 SE symptomatic shoots and

24.4% ± 2.7 SE severity (SWM) and 8.5 ± 2.7 SE symptomatic shoots and 20.8% ± 6.1 SE severity (PPF) were observed, respectively (Fig. 1.1).

There was a significant treatment-by-year interaction for disease severity and density at both locations indicating differences in control among some foliar fungicides by year (Table

1.1). In 2016, significant differences (P < 0.0001) were detected in AUDPC values for disease severity and density among foliar treatments for both locations (Table 1.1). All treatments in

2016 reduced disease severity (70.6 to 100% [PPF] and 49.5 to 97.6% [SWM] control) and density (87.6 to 100% [PPF] and 69.2 to 99.2% [SWM] control) compared to the untreated control (Table 1.1). Oxathiapiprolin provided 97.6 to 100% control in 2016 and according to

AUDPC data for disease severity and density the efficacy was significantly better than all other treatments at both locations AUDPC values for disease severity (2016) were significantly higher for cymoxanil (70.6% [PPF] and 70.1% [SWM] control) and fosetyl-Al (75.3% [PPF] and 49.5%

[SWM] control) than other treatments. Disease density AUDPC values (2016) for fosetyl-Al were the highest among the treatments for the SWM (69.2% control) but were low for the PPF

(98.6% control) and better than some fungicide treatments. Fluopicolide, ametoctradin/dimethomorph, cyazofamid, and mandipropamid treatments were similar for both locations according to AUDPC values (2016) for disease severity and density (≥ 90.6%

[severity] and 89% [density] control) with the following exceptions; ametoctradin/dimethomorph had lower AUDPC values than fluopicolide (P = 0.0034 [severity]; P = 0.0024 [density]) and cyazofamid (P = 0.0068 [severity]; P = 0.0239 [density]) at the PPF and mandipropamid (P =

0.0324 [severity]) at the SWM.

In 2017, significant differences (P < 0.0001) were detected in AUDPC values for disease severity and density among foliar fungicides for both locations (Table 1.1). All treatments reduced disease severity (64.3 to 100.0% control) and density (73.1 to 100% control) compared to the control at the SWM (2017). For the PPF (2017), all foliar fungicides, except cymoxanil

(PPF), reduced disease severity (92.1 to 99.7% control) and density (91.6 to 100% control) compared to the untreated control (Table 1.1). For both locations, AUDPC values for disease severity (2017) were similar (> 98.8% control) for plots treated with oxathiapiprolin,

ametoctradin/dimethomorph, fluopicolide, cyazofamid, and mandipropamid; AUDPC values for fosetyl-Al (93.0% [PPF] and 64.3% [SWM] control) were significantly higher.

Table 1.1. Area under the disease progress curve (AUDPC) for downy mildew disease severity and density ratings for hop cv. Nugget when treated with foliar fungicides in 2016 and 2017. 2016 AUDPCv 2017 AUDPC PPFw SWMx PPF SWM Fungicide treatment Severityy Densityz Severity Density Severity Density Severity Density Inoculated untreated control 1887.89 a 544.35 a 3282.85 a 851.01 a 1389.99 a 356.18 a 428.08 a 212.66 a Oxathiapiprolin 0.00 f 0.00 g 80.13 g 6.72 e 5.18 c 0.00 d 0.07 d 0.00 c Ametoctradin/dimethomorph 9.16 e 2.69 f 172.84 f 37.26 d 16.48 c 0.02 cd 0.55 cd 0.08 c Fluopicolide 75.33 d 34.18 bcd 209.53 ef 54.29 cd 8.42 c 0.39 c 3.60 cd 0.14 c Cyazofamid 65.77 d 19.72 cde 265.22 ef 71.94 cd 8.32 c 0.03 cd 0.50 d 0.55 c Mandipropamid 39.40 de 15.26 def 309.69 e 87.42 cd 4.39 c 0.02 cd 0.11 d 0.04 c Dimethomorph 207.27 c 59.33 bc 558.25 d 138.79 c 109.91 b 29.91 b 23.16 c 8.54 c Cymoxanil 554.48 b 67.70 b 982.01 c 97.90 c 1113.48 a 357.03 a 18.37 cd 8.51 c Fosetyl-Al 465.38 b 7.63 ef 1658.23 b 262.34 b 97.15 b 14.17 b 152.68 b 57.24 b Year × treatment P value <0.0001 <0.0001 <0.0001 0.0054 <0.0001 <0.0001 <0.0001 0.0054 Year × treatment by year <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 v Statistical analysis for AUDPC was performed with a linear mixed model and a compound symmetry covariance structure for each location. Fisher’s protected least significant differences (LSD) at P < 0.05 were used to make simple effect comparisons of year × treatment by year. Column means with a letter in common are not significantly different. w SWM = Michigan State University (MSU) Southwest Michigan Research and Extension Center, Benton Harbor, MI (latitude 42.6854°, longitude -84.4716°). x PPF = MSU Plant Pathology Farm, Lansing, MI (latitude 42.0835°, longitude -86.3542°). y Disease severity of symptomatic leaves and symptomatic shoots was estimated visually on the Horsfall- Barratt (HB) scale. Prior to statistical analysis ratings values from the HB scale were converted to midpoint values. z Disease density was determined as the number of symptomatic downy mildew shoots per plot.

Drench treatments. At the SWM, 35-days after the first disease assessment, the untreated control plots had 9.5 ± 1.9 SE symptomatic shoots and 15.3% ± 0.6 SE severity per plant (1 Sep) (data not shown). The PPF drench plots were not inoculated in 2016 due to a lack of available inoculum and only trace levels of disease were observed in the untreated control and mefenoxam-treated plots. In 2017, 22-days after the first disease assessment, untreated control plots had 9.8 symptomatic shoots ± 4.4 SE and 15.3% ± 8.3 SE severity (SWM; 30 May). The

PPF plots were inoculated in 2017 and untreated control plots had an average of 3.6 ± 0.8 SE symptomatic shoots and 11.1% ± 2.6 SE severity (10 May) prior to inoculation. Twenty-one days later, untreated control plots had 21.1 symptomatic shoots ± 4.2 SE and 27.5% ± 7.4 SE severity (PPF) (31 May) (data not shown).

At the SWM, the interaction between year and fungicide was not significant for disease severity (P = 0.7534) and density (P = 0.1027) indicating that difference in control among drench fungicides were similar in both years (Table 1.2). Significant differences were detected among AUDPC values for main effects (SWM) by fungicide for disease severity (P = 0.0032) and density (P =0.0134) among treated plots (Table 1.2). Mefenoxam drenches were similar to the untreated control according to the AUDPC data for disease severity and density (SWM). In the fluopicolide-treated plots (SWM) AUDPC values for disease severity (85.7% disease control) and density (83.4% control) plots were similar to oxathiapiprolin/ mefenoxam treatments; both treatments reduced disease severity and density compared to the untreated control and the mefenoxam treatment (Table 1.2). At the PPF (2017 only), AUDPC data for mefenoxam drenches (74.0% [severity] and 85.4% [density] control) were significantly lower than the untreated control and similar to those for fluopicolide for disease density. Oxathiapiprolin/ mefenoxam reduced AUDPC values (PPF) for disease severity (90.1% control) and density

(91.5% control) compared to fluopicolide (82.4% [severity] and 82.1% [density] control) and the untreated control.

CAA point mutations. The annotated CesA3 gene for P. humuli was submitted to

GenBank (MN725072). Comparisons of the CesA3 gene using available genomic data revealed that P. humuli, P. cubensis and Pl. viticola have similar protein structure and predicted structural motifs (Fig. 1.2A); P. humuli and P. cubensis have the same amino acid sequence. Significant variation in nucleotide sequence was observed in CesA3 in all three species with Pl. viticola sharing 80.4 and 80.5% identity with P. humuli and P. cubensis, respectively (Fig. 1.2). Sixty P. humuli isolates were obtained from symptomatic shoots. Forty-two sequences were examined for single nucleotide polymorphisms that cause amino acid shifts in wild type V1105 and G1109; sequencing reactions for eighteen isolates were unsuccessful (Q scores <20). No evidence of amino acid shifts at V1109 or G1105 were found (Table 1.3). In 22 of the isolates obtained from five hop yards in northwest Michigan, a wild type glycine [GGG] was present at amino acid position on 1105 and an unaltered valine [GTG] was present at amino acid position 1109.

Similarly, in 13 of the isolates obtained from three central Michigan hop yards and seven isolates collected from three southwest MI hop yards, unaltered glycine [GGG] and valine [GTG] amino acids were observed at position 1105 and 1109, respectively.

Table 1.2. Area under the disease progress curve (AUDPC) for downy mildew disease severity and density ratings for hop cv. Nugget when treated with drench fungicides in 2016 and 2017. AUDPC Fungicide treatment SWMw (2016 and 2017) PPFx (2017 only) Severityy Densityz Severity Density Inoculated untreated 186.18 a 67.17 a 1307.14 a 726.98 a control Mefenoxam 260.85 a 96.19 a 340.17 b 106.24 bc Oxathiapiprolin/ mefenoxam 43.90 b 15.34 b 129.99 d 61.47 c Fluopicolide 26.57 b 11.14 b 230.58 c 129.83 b Treatment P value 0.0032 0.0134 0.0001 <0.0001 Year 0.4005 0.8134 - - Year × treatment 0.7534 0.1027 - - w SWM = Michigan State University (MSU) Southwest Michigan Research and Extension Center, Benton Harbor, MI (latitude 42.6854°, longitude -84.4716°). Statistical analysis for AUDPC was performed with a linear mixed model and a compound symmetry covariance structure for the SWM in 2016 and 2017; comparisons were made for the main effects of treatment least squares means. Fisher’s protected least significant differences (LSD) at P < 0.05 were used to make comparisons least squares means. Column means with a letter in common are not significantly different. x PPF = MSU Plant Pathology Farm, Lansing, MI (latitude 42.0835°, longitude -86.3542°). Statistical analysis for AUDPC was performed with a linear mixed model for the PPF (2017 only). Fisher’s protected least significant differences (LSD) at P < 0.05 were used to make comparisons least squares means of fungicides. Column means with a letter in common are not significantly different. y Disease severity of symptomatic leaves and symptomatic shoots was estimated visually on the Horsfall- Barratt (HB) scale. Prior to statistical analysis ratings values from the HB scale were converted to midpoint values. z Disease density was determined as the number of symptomatic downy mildew shoot per plot.

Table 1.3. Amino acid configurations at position 1105 and 1109 of wild type (fungicide sensitive) Peronosporales and Pseudoperonospora humuli isolates collected throughout Michigan and examined for a point mutation that confer resistance to carboxylic acid amide fungicide. Yard No. of Amino Acid in CesA3y CAA Species Sourcew ID isolatesx 1105 1109 Sensitivityz Phytophthora infestans Blum et al. 2012 - - Gly [GGC] Val [GTG] WT Sensitive Plasmopara viticola Blum et al. 2012 - - Gly [GGC] Val [GTT] WT Sensitive Plasmopara viticola Blum et al. 2012 - - Val [GTG] Val [GTT] WT Resistant Plasmopara viticola Blum et al. 2010 - - Ser [AGC] Val [GTT] WT Resistant P. cubensis Blum et al. 2012 - - Gly [GGG] Val [GTG] WT Sensitive P. cubensis Blum et al. 2011 - - Val [GTG] Val [GTT] WT Resistant P. cubensis Blum et al. 2011 - - Trp [TGG] Val [GTT] WT Resistant P. cubensis NW Michigan - - Trp [TGG] Val [GTG] WT Resistant P. humuli NW Michigan MI-4 4 Gly [GGG] Val [GTG] Sensitive P. humuli NW Michigan MI-5 5 Gly [GGG] Val [GTG] Sensitive P. humuli NW Michigan MI-6 3 Gly [GGG] Val [GTG] Sensitive P. humuli NW Michigan MI-7 4 Gly [GGG] Val [GTG] Sensitive P. humuli NW Michigan MI-8 6 Gly [GGG] Val [GTG] Sensitive P. humuli CEN Michigan MI-1 5 Gly [GGG] Val [GTG] Sensitive P. humuli CEN Michigan PPF 5 Gly [GGG] Val [GTG] Sensitive P. humuli CEN Michigan MI-11 3 Gly [GGG] Val [GTG] Sensitive P. humuli SW Michigan SWM 2 Gly [GGG] Val [GTG] Sensitive P. humuli SW Michigan MI-9 3 Gly [GGG] Val [GTG] Sensitive P. humuli SW Michigan MI-10 2 Gly [GGG] Val [GTG] Sensitive w Source of reference sequences and samples collected across nine commercial and two experimental hop yards (PPF and SWM) located in the Northwest (Benzie, Leelanau, Oceana (P. cubensis isolate only) and Grand Traverse counties), Central (Montcalm, Ingham, and Genesee Counties) and Southwest (Berrien County) regions of the state between 23 May and 6 Jun 2018. SWM = Michigan State University (MSU) Southwest Michigan Research and Extension Center, Benton Harbor, MI (latitude 42.6854°, longitude -84.4716°); PPF = MSU Plant Pathology Farm, Lansing, MI (latitude 42.0835°, longitude -86.3542°). x Isolates were derived from a single symptomatic shoot. y CAA sensitive Oomycetes and Peronosporales have a highly conserved glycine at position 1105 (G1105) and have a valine at position 1109 (V1109) in the cellulose synthase 3 (CesA3) gene, respectively. Amino acid substitutions at these loci are linked to CAA resistance in field isolates (Blum et al. 2012). z Isolates sensitivity is wild type (WT) sensitive (G1105 and V1109), WT resistant (V1105, S1105, and W1105), or designated sensitive if amino acid configurations matched WT sensitive isolates.

Discussion

Hops has become an important specialty crop in Michigan and supports a local craft beer industry of significant economic importance to the state (Brewers Association, 2019). In the absence of resistant cultivars, growers face annual DM outbreaks and fungicides are needed to limit disease (Higgins and Hausbeck 2017, 2018; Woods and Gent 2016). Our research shows that there are four highly effective modes of action currently registered for use on hop in

Michigan and include fluopicolide (FRAC 43; benzamide), mandipropamid (FRAC 40; CAA), ametoctradin/dimethomorph (FRAC 45; quinone outside inhibitor, stigmatellin binding site

(QoSI)/FRAC 40; CAA), and cyazofamid (FRAC 21; quinone inside inhibitor (QiI)).

Oxathiapiprolin (FRAC 49; oxysterol binding protein homologue inhibitor (OSBPI)) was among the most effective foliar fungicides but is not currently registered on hop.

In our studies, fluopicolide, cyazofamid, oxathiapiprolin, mandipropamid, and ametoctradin/dimethomorph reduced disease severity compared to fosetyl-Al and the control for both years and locations providing high levels of disease control (90.6 to 100%). These fungicides also reduced disease density (89.7 to 100% control) compared to fosetyl-Al (69.2 to

73.1% control) and the control at the SWM in both years. We found that fosetyl-Al (FRAC P07

(previously FRAC 33); phosphonates) was not as effective as most of the other fungicides. One notable exception was the low disease density AUDPC values at the PPF for fosetyl-Al in 2016 and 2017. Shoot with symptoms of DM develop slowly and infections likely occur via stomatal openings in young bud stipules; stomata are absent in immature buds (Royal 1970). Presumably, the pathogen needs to grow faster than the shoot to produce a symptomatic shoot, but this is likely confounded by factors influencing plant growth (Royal 1970). While cymoxanil (FRAC

27; unknown target site) failed to limit DM at the PPF in 2017, it was better than the untreated

control in 2016 at both sites although it was not as effective as several other treatments included in this study.

CAA-based fungicides and cyazofamid have historically controlled hop DM in

Washington (Nelson et al. 2011; Nelson and Grove 2006, 2007, 2008, 2009, 2010) but recent reports are limited (Massie et al. 2019; Adams et al. 2018). In Oregon, dimethomorph provided

>80% DM control when applied prior to pathogen infection but had limited post-infection activity (Gent et al. 2015). Mandipropamid and ametoctradin/dimethomorph limited hop DM in

North Carolina (Adams et al. 2018, 2019). Cyazofamid effectively limited DM on hop in

Washington (Nelson and Grove 2007, 2008, 2009, 2010) and North Carolina (Adams et al.

2018). Studies including fluopicolide and oxathiapiprolin have been conducted in Oregon (Gent

2017a, 2017b; Gent et al. 2020) and North Carolina (Adams et al. 2018, 2019). In North

Carolina, treatments with fluopicolide, oxathiapiprolin (Adams et al. 2018), or oxathiapiprolin/chlorothalonil (Adams et al. 2019) reduced hop DM disease severity compared to the untreated control. Oxathiapiprolin/chlorothalonil was less effective than ametoctradin/dimethomorph and fosetyl-Al but still limited disease severity to <15.0% (Adams

2019). In Oregon, oxathiapiprolin/mandipropamid reduced disease density compared to the untreated control (Gent 2017b) and provided >90% control.

Aliette WDG and Linebacker WDG are fosetyl-Al containing products labeled for DM control on hop in Michigan at 2.8 kg/ha (2.24 kg of active ingredient). In Oregon, applications of

Aliette 80 WDG at 11.2 kg/ha (five times higher than the Michigan labeled rate and two times higher than a Special Local Needs registration in Oregon) and 14-day application intervals provided 74.7% control and were less effective than oxathiapiprolin/mandipropamid and fluopicolide treatments at limiting infected hop shoots (Gent et al.2020). Rates in excess of the

product labels are not permitted in commercial production. Even prior to reports of phosphonate insensitivity in P. humuli (Nelson et al. 2004) fosetyl-Al efficacy in Oregon was marginally effective providing only 56.9% control compared to > 99% control with metalaxyl, albeit only a single application of either fungicide was applied (Hunger and Horner 1982). Phosphonates have not adequately controlled P. humuli’s sister species, P. cubensis in cucumber (Adams and

Quesada-Ocampo 2017; Hausbeck and Cortright 2012; Hausbeck and Cortright 2013; Huang and Vallad 2011). The mode of action for phosphonates has been linked to plant defense responses (i.e. phytoalexin accumulation in peppers (Guest 1984), reactive oxygen responses in

Arabidopsis (Daniel and Guest 2006), and salicylic acid signaling in potatoes (Burra et al. 2014)) that may impact efficacy, but defense responses have yet to be investigated in hop. In

Washington, cymoxanil was effective when mixed with copper or famoxadone (Nelson et al.

2003). Since cymoxanil has limited post-infection activity (Gent et al. 2015), it should be mixed with another effective fungicide (Gisi et al. 1985).

For some oomycete pathogens, fungicide drenches are more effective for plant protection than foliar sprays (Foster and Hausbeck 2010; Hausbeck et al. 2018; Meyer and Hausbeck

2013). In our trials, mefenoxam drenches reduced disease severity and density at the PPF (2017) but not at the SWM. Lower disease pressure at the SWM may have influenced our ability to detect differences at this site. In contrast, high disease pressure occurred at the PPF and fluopicolide drenches reduced disease severity and density compared to the untreated control.

The PPF drench plot was established in 2016 and inoculated in 2017 and could account for the increased disease pressure. The application method for fluopicolide for DM control on hop is restricted to foliar applications; soil applications are not currently permitted. The rate for fluopicolide drench treatments used in this study was derived from fluopicolide soil application

registered for Phytophthora control on cucurbit vegetables. Oxathiapiprolin/mefenoxam limited disease severity and density at both locations but this premix is not registered for use on hop.

However, only moderate levels of control were achieved with fluopicolide and oxathiapiprolin/mefenoxam (76.4 to 91.5% control). Additional evaluations are needed to compare fungicides applied via a drench or a foliar spray in order to develop a strategy that maximizes DM control and delays pathogen resistance.

Downy mildew pathogens can develop resistance to many of the fungicides tested in this study and management strategies are needed to delay the build-up of resistant P. humuli populations. Insensitivity to phenylamide (metalaxyl and mefenoxam) and phosphonate (fosetyl-

Al and phosphorous acid) fungicides already occurs in some hop-growing regions (Gent et al.

2008; Gent et al. 2020; Klein 1994; Marks and Gevens 2019; Nelson et al. 2004). The presumed absence of a fitness cost associated with phenylamide resistance in P. humuli (Gent et al. 2008) would limit the effectiveness of these fungicides in alternation or in mixtures as resistance management strategies (van den Bosch and Gilligan 2008). In P. humuli populations with reduced sensitivity to phosphonates, increased application rates appear to be an ineffective strategy to limit the proliferation of insensitive isolates (Gent et al.2020). Reduced efficacy of oxathiapiprolin was recently observed in P. cubensis (Hausbeck et al. 2019) and resistance management tools are needed for FRAC 49 / OSBPIs which are ranked as medium to high risk for resistance development (FRAC 2018). Control failure of fluopicolide was reported for P. cubensis within several years of registration for cucurbits (Adams and Quesada-Ocampo 2014;

Hausbeck and Linderman 2014; Langston and Sanders 2013); resistance was later confirmed

(Thomas et al. 2018). Resistance to cymoxanil has been reported for the grape DM pathogen

(Gullino et al. 1997; Genet and Jaworska 2013; Toffolatti et al.2014). CAA resistance is

reported in P. cubensis (Blum et al. 2011; Blum et al. 2012; Rahman et al. 2017) and Pl. viticola (Blum et al. 2010; Nanni et al.2016; Toffolatti et al.2018). In Pl. viticola, fungicide alternations and mixtures used as resistance management for mandipropamid, appear to be effective at lowering, but not eradicating, the frequency of resistance mutation G1105S/V in vineyards with resistant isolates (Toffolatti et al.2018) and reinforces the need to develop resistance management strategies for CAAs in hop.

While we assessed fungicide efficacy, our field trials were not designed to detect fungicide resistance in P. humuli isolates. The DM pathogens are obligate parasites that typically require time-consuming leaf bioassays for fungicide resistance monitoring (Gent et al. 2008;

Genet and Jaworska 2013; Klein 1994; Nelson et al. 2004; Thomas et al. 2018). In a limited number of DM pathogens, CAA resistance is linked to mutations in the CesA3 and detection of resistant genotypes enables expedited screening for fungicide resistance (Blum et al. 2010; Blum et al. 2011; Blum et al. 2012). Single nucleotide polymorphisms that cause amino acid shifts in

V1109 or G1105 result in reduced fungicide sensitivity in the Peronosporales (Blum et al. 2012).

CAA resistance and CesA3 have yet to be characterized for P. humuli. Yet, it is reasonable to assume that similar mutations occurring at V1109 and G1105, as reported for P. cubensis would lead to resistance (Blum et al. 2011; Blum et al. 2012; Rahman et al. 2017). Due to the large size of CesA3 (i.e. 3,514 base pairs) it would have been impractical and costly to sequence the

CesA3 gene from all isolates collected in our study. To demonstrate that the CesA3 gene is also conserved in P. humuli, we annotated the gene from the publicly available P. humuli genome

(Rahman et al. 2019) and showed it had an identical amino acid sequence compared to P. cubensis. We used PCR and a new primer set to screen for single nucleotide polymorphisms in the CesA3 gene at amino acid locations V1109 and G1105 (Blum et al. 2012).

We did not detect an amino acid shift at G1105 and V1109 in the CesA3 gene indicating that the frequency of CAA resistance in the Michigan P. humuli population is likely low. The frequency of resistant isolates in a population, ignoring any confounding factors (i.e. fitness cost, dispersal, infection, and host dynamics, etc.), can be estimated with a binomial probability

(Russell 2002). Based on our sample size (n= 43), if the frequency of resistance was 10%, we have a 98.9% chance of detecting at least one resistant isolate. With a sample size of 68, the

CesA3 resistance mutations at G1105 were detected in 47.1% of P. cubensis samples (Rahman et al. 2017). Nanni et al. (2016) sampled for G1105 based CAA resistance on bulked Pl. viticola isolates (25 to 30 leaf isolates per sampling location) and found 21 and 42% of samples were resistant or contained a mixture of resistant and sensitive isolates. With all of our P. humuli isolates genotyped as sensitive, a larger sample size is needed to detect resistant isolates that are likely present at low frequencies (1% = approx. 300 samples; 0.1% = approx. 3000 samples) and bulking multiple samples by location would be useful strategy to aid in detection. Our sampling of experimental plots and commercial hop yards, combined with field efficacy data, provides a baseline for CAA resistance. Currently, CAA fungicides can be effectively used by Michigan growers to limit DM but should be used in an integrated management program and monitored.

High disease pressure and prolonged periods of DM conducive weather are likely to impact disease management strategies in Michigan. In our untreated control plots, foliar disease severity and the number of DM infected shoots (disease density) increased 3- to 4-fold during frequent (11-14 days with > 0.3 mm) and heavy precipitation (147.7 to 211.5 mm) between 13

Aug and 10 Sep 2016 (Tables A1, A2). In North Carolina, late season disease symptoms were noted in late Sep and early Oct (Adams et al.2018, 2019). In contrast, seasonal epidemics last only 4 to 5 weeks in Washington and disease pressure rarely persists to flowering due to the

onset of hot, dry weather in Jun (Johnson et al. 1994). In Oregon, hop shoots with DM symptoms may persist throughout Jun and fungicides (commonly 8 to 10 applications) applied until mid-Jul

(Gent et al. 2010) may help protect the burr and developing flower. Under prolonged disease conditions, extended fungicide applications during the burr stage, flowering, and cone development may be needed to protect cones (typically harvested in Michigan in late Aug to

Oct). However, additional fungicide applications will likely have a long-term impact on efficacy and resistance management and more research is needed to characterize late season disease pressure and the associated risk to developing and mature cones.

Effective DM management strategies for growers in Michigan and eastern U.S. growing regions that experience wet conditions are essential to support the upward trajectory of hops in these regions. Our research provides data to support growers in these regions that can contribute to comprehensive and sustainable DM control programs. We identified four highly effective modes of action registered for hop in Michigan. The fungicide resistance action committee

(FRAC) recommends no fewer than three and preferably five modes of actions to manage fungicide resistance (Kuck et al. 2012). To maintain the efficacy of these modes of action monitoring for fungicide resistance is recommended. Molecular tools, such the G1105 CAA sensitivity marker, offer a practical approach to screen for fungicide sensitivity.

APPENDIX

Table A1. Temperature (mean and minimum), precipitation (total and days with greater than 3 mm), and relative humidity (night mean and night hours greater than 80%) at the Michigan State University Plant Pathology Farm (PPF) and Southwest Michigan Research and Extension Center (SWM) in Michigan, 2016. Temperature (°C) Precipitation Relative Humidity Date and Total Days > 0.3 Night Night Hrs. location Mean Min. (mm) mm Mean (%)y > 80%z 6/17 to 6/30 SWM 22.0 7.9 27.2 5 84.0 70 PPF 22.2 6.6 138.2 3 74.1 31 7/1 to 7/14 SWM 22.7 10.0 146.6 8 89.9 102 PPF 22.8 8.2 170.2 6 85.9 65 7/15 to 7/28 SWM 23.9 10.8 105.7 6 89.9 121 PPF 23.9 11.5 37.6 3 85.5 95 7/29 to 8/12 SWM 24.0 12.8 5.6 2 87.0 102 PPF 23.4 10.9 13.7 2 85.6 95 8/13 to 8/27 SWM 20.3 10.7 151.8 7 92.9 149 PPF 22.7 11.3 131.0 9 93.3 145 8/28 to 9/10 SWM 22.6 11.3 108.7 4 90.2 127 PPF 21.7 8.3 43.7 5 89.8 130 9/11 to 9/24 SWM 20.3 10.7 26.4 4 84.9 98 PPF 19.3 7.9 40.1 4 88.0 121 9/25 to 10/8 SWM 16.9 9.0 35.6 7 83.3 122 PPF 16.1 7.6 61.0 10 85.4 110 y Nighttime mean relative humidity from 2000 to 0600 hours. z Hours of nighttime relative humidity > 80% from 1500 to 0600 hours.

Table A2. Temperature (mean and minimum), precipitation (total and days with greater than 3 mm), and relative humidity (night mean and night hours greater than 80%) at the Michigan State University Plant Pathology Farm (PPF) and Southwest Michigan Research and Extension Center (SWM) in Michigan, 2017. Temperature (°C) Precipitation Relative Humidity Date and Total Days > 0.3 Night Night Hrs. location Mean Min. (mm) mm Mean (%)y > 80%z 5/1 to 5/14 SWM 10.1 -3.1 35.3 5 68.7 16 PPF 10.9 -2.8 34.5 4 67.0 22 5/15 to 5/28 SWM 17.2 6.3 38.6 8 73.9 63 PPF 17.5 3.9 44.2 5 71.1 45 5/29 to 6/11 SWM 19.5 6.6 2.54 2 73.2 33 PPF 19.4 7.1 2.79 2 73.6 18 6/12 to 6/25 SWM 22.6 10.4 27.94 8 78.3 62 PPF 22.2 8.3 82.8 8 80.4 53 6/26 to 7/9 SWM 21.3 8.3 70.1 5 81.5 69 PPF 21.2 9.0 45.2 5 84.9 67 7/10 to 7/23 SWM 22.4 11.4 84.1 6 91.5 128 PPF 23.1 13.4 25.1 5 89.9 115 7/24 to 8/5 SWM 21.3 11.1 9.7 3 84.4 89 PPF 21.0 8.9 9.4 3 85.3 73 y Nighttime mean relative humidity from 2000 to 0600 hours. z Hours of nighttime relative humidity > 80% from 1500 to 0600 hours.

LITERATURE CITED

Adams, M. L., and Quesada-Ocampo, L. M. 2014. Evaluation of fungicides for control of downy mildew on cucumber, Kinston 2013. Plant Dis. Manag. Rep. 8:V240. Online.

Adams, M. L., and Quesada-Ocampo, L. M. 2017. Evaluation of fungicides for control of downy mildew on cucumber, Clayton 2016. Plant Dis. Manag. Rep. 11:V100. Online.

Adams, M. L., Collins, H., and Quesada-Ocampo, L. M. 2018. Evaluation of fungicides for control of downy mildew on hop, Clayton 2017. Plant Dis. Manag. Rep. 12:PF082. Online.

Adams, M. L., Collins, H., and Quesada-Ocampo, L. M. 2019. Evaluation of fungicides for control of downy mildew on hop, Clayton 2018. Plant Dis. Manag. Rep. 13:V066. Online.

Blum, M., Gamper, H.A., Waldner, M. Sierotzki, H., and Gisi, U. 2012. The cellulose synthase 3 (CesA3) gene of oomycetes: structure, phylogeny and influence on sensitivity to carboxylic acid amide (CAA) fungicides. Fungal Biol. 116:529-542.

Blum, M., Waldner, M., and Gisi, U. 2010. A single point mutation in the PvCesA3 gene confers resistance to the carboxylic acid amide fungicide mandipropamid in Plasmopara viticola. Fungal Genet. Biol. 47:499-510.

Blum, M., Waldner, M., Olaya, G. Cohen, Y., Gisi, U., and Sierotzki, H. 2011. Resistance mechanism to carboxylic acid amide fungicides in the cucurbit downy mildew pathogen Pseudoperonospora cubensis. Pest Manag. Sci. 67:1211-1214.

Brewers Association. 2019. Michigan’s Craft Beer Sales and Production Statistics, 2018. Online. https://www.brewersassociation.org/statistics-and-data/state-craft-beer-stats/.

Burra, D. D., Berkowitz, O., Hedley, P. E., Morris, J., Resjö, S., Levander, F., Lilgeroth, E., Andreasson, E., and Alexandersson, E. 2014. Phosphite-induced changes of the transcriptome and secretome in Solanum tuberosum leading to resistance against Phytophthora infestans. BMC Plant Bio. 14:254-271.

Chee, H. Y., and Klein, R. E. Laboratory production of oospores in Pseudoperonospora humuli. Korea J. Plant Pathol. 14:618-621.

Coley-Smith, J.R. 1962. Overwintering of hop downy mildew Pseudoperonospora humuli (Miy. and Tak.) Wilson. Ann. Appl. Biol. 50:235-243.

Coley-Smith, J. R. 1965. Infection of hop rootstocks by downy mildew Pseudoperonospora humuli (Miy. and Tak.) Wilson and its control by early-season dusts. Ann. Appl. Biol. 56:381-388.

Daniel, R. and Guest, D. 2006. Defense responses induced by potassium phosphonate in Phytophthora palmivora-challenged Arabidopsis thaliana. Physiol. Mol. Plant Pathol. 67:194-201.

Foster, J. M., and Hausbeck, M. K. 2010. Managing Phytophthora crown and root rot in bell pepper using fungicides and host resistance. Plant Dis. 94:697-702.

FRAC (Fungicide Resistance Action Committee). 2018. FRAC Recommendations for OSBPI Fungicides. Online: https://www.frac.info/frac-teams/working-groups/osbpi- fungicides/recommendations-for-osbpi.

Genet, J., and Jaworska, G. 2013. Characterization of European Plasmopara viticola isolates with reduced sensitivity to cymoxanil. Eur. J. Plant Pathol. 135:383-393.

Gent, D.H. 2017a. Evaluation of fungicides for hop downy mildew, Woodburn, Oregon, 2016. Plant Dis. Manag. Rep. 11: V047. Online.

Gent, D.H. 2017b. Evaluation of fungicides for hop downy mildew, Hubbard, Oregon, 2016. Plant Dis. Manag. Rep. 11: V048. Online.

Gent, D. H., Block, M., and Claassen, B. J. 2020. High Levels of insensitivity to phosphonate fungicides in Pseudoperonospora humuli. Plant Dis. 104:1400-1406.

Gent, D. H., Cohen, Y., and Runge, F. 2017. Homothallism in Pseudoperonospora humuli. Plant Pathol. 66:1508-1516.

Gent, D. H., Farnsworth, J. L., and Johnson, D. A. 2012a. Spatial analysis and incidence-density relationships for downy mildew on hop. Plant Pathol. 61:37-47.

Gent, D. H., Nelson, M. E., and Grove, G. G. 2008. Persistence of phenylamide insensitivity in Pseudoperonospora humuli. Plant Dis. 92:463-468.

Gent, D.H., Nelson, M.E., Farnsworth, J.L., and Grove, G.G. 2009. PCR detection of Pseudoperonospora humuli in air samples from hop yards. Plant Pathol. 58:1081-1091.

Gent, D. H., Nelson, M. E., Grove, G. G., Mahaffee, W. F., Turechek, W. W., and Woods, J. L. 2012b. Association of spring pruning practices with severity of powdery mildew and downy mildew on hop. Plant Dis. 96:1343-1351.

Gent, D. H., and Ocamb, C. M. 2009. Predicting infection risk of hop by Pseudoperonospora humuli. Phytopathology 99:1190-1198.

Gent, D. H., Ocamb, C. M., and Farnsworth, J. L. 2010. Forecasting and management of hop downy mildew. Plant Dis. 94:425-431.

Gent, D. H., Twomey, M. C., Wolfenbarger, S. N., and Woods, J. L. 2015. Pre- and post- infection activity of fungicides in control of hop downy mildew. Plant Dis. 99:858-865.

George, A. 2019. USA hops 2018 Statistical Report. Hop Growers of America. Online.

Gisi, U., Binder, H., and Rimbach, E. 1985. Synergistic interactions of fungicides with different modes of action. Trans. Br. Mycol. Soc. 85:299-306.

Gisi, U., Waldner, M., Kraus, N., Dubuis, P.H., and Sierotzki, H. 2007. Inheritance of resistance to carboxylic acid amide (CAA) fungicides in Plasmopara viticola. Plant Pathol. 56:199- 208.

Guest, D. I. 1984. Modification of defense responses in tobacco and capsicum following treatment with Fosetyl-Al [Aluminum tris (o-ethyl phosphonate)]. Physiol. Plant Pathol. 25:125-134.

Gullino, M. L., Mescalchin, E., and Mezzalama, M. 1997. Sensitivity to cymoxanil in populations of Plasmopara viticola in northern Italy. Plant Pathol. 46:729-736.

Hausbeck, M. K., and Cortright, B. D. 2012. Evaluation of fungicides for control of downy mildew on pickling cucumber, 2011. Plant Dis. Manag. Rep. 6:V134. Online.

Hausbeck, M. K., and Cortright, B. D. 2013. Control of downy mildew on pickling cucumber with fungicides, 2012. Plant Dis. Manag. Rep. 7:V156. Online.

Hausbeck, M. K., and Linderman, S. D. 2014. Evaluation of fungicides for control of downy mildew of cucumber, 2013. Plant Dis. Manag. Rep. 8:V304. Online.

Hausbeck, M. K., Harlan, B. R., and Courtney, S. E. 2018. Evaluation of registered and non- registered fungicides drenches for the control of Phytophthora root rot on Calibrachoa in the greenhouse, 2016. Plant Dis. Manag. Rep. 12:OT022. Online.

Hausbeck, M. K., Perla, D. E., and Cook, A. J. 2019. Evaluation of single fungicide products for control of downy mildew of cucumber, 2018. Plant Dis. Mang. Rep. 13:V139. Online.

Higgins, D. H., and Hausbeck, M. K. 2017. Fungicides and cultivars can limit hop downy mildew (Pseudoperonospora humuli) in Michigan. (Abstr.) Phytopathology 107:S5.1.

Higgins, D. H., and Hausbeck, M. K. 2018. Effective management strategies for hop downy mildew (Pseudoperonospora humuli) in Michigan. (Abstr.) Phytopathology 108:S2.28.

Horsfall, J. G., and Barratt, R. W. 1945. An improved grading system for measuring plant disease. (Abstr.) Phytopathology 35: 655.

Huang, C. H., and Vallad, G. E. 2011. Evaluation of fungicides for management of downy mildew on cucumber, fall 2010. Plant Dis. Manag. Rep. 5:V068. Online.

Hunger, R. M., and Horner, C. E. 1982. Control of hop downy mildew with systemic fungicides. Plant Dis. 66:1157-1159.

Johnson, D. A. 1991. Two degree-day models for predicting initial emergence of hop shoots systemically infected with Pseudoperonospora humuli. Plant Dis. 75:285-287.

Johnson, D. A., Alldredge, J. R., and Allen, J. R. 1994. Weather and downy mildew epidemics of hop in Washington State. Phytopathology 84:524-527.

Johnson, D. A., and Anliker, W. L. 1985. Effect of downy mildew epidemics on the seasonal carryover of initial inoculum in hop yards. Plant Dis. 69:140-142.

Johnson, D. A., Engelhard, B., and Gent, D. H. 2009. Downy Mildew. Pages 18-22 in: Compendium of Hop Diseases and Pests. W. F. Mahaffee, S. J., Pethybridge, and D. H. Gent, eds. American Phytopathological Society, St. Paul, MN.

Johnson, D. A., and Skotland, C. B. 1985. Effects of temperature and relative humidity on sporangium production of Pseudoperonospora humuli on hop. Phytopathology 75:127- 129.

Klein, R.E. 1994. Occurrence and incidence of metalaxyl resistance in Pseudoperonospora humuli in the Pacific Northwest. Plant Dis. 78:161-163.

Kuck, K., Leadbeater, A., and Gisi, U. 2012. FRAC Mode of Action Classification and Resistance Risk of Fungicides. In: Krämer, W., Schirmer, U., Jeschke, P., Witschel, M. (eds), Modern Crop Protection Compounds, 2nd ed. Wiley‐VCH Weinheim, Germany.

Langston, D. B., Jr., and Sanders, F. H., Jr. 2013. Evaluation of fungicides for control of downy mildew on cucumber in Georgia I, 2012. Plant Dis. Manag. Rep. 7:V108. Online.

Lizotte, E. 2015. Management Options for Hop Yards with Downy Mildew. Michigan State University Extension. Online. https://www.canr.msu.edu/news/management_options_for_hopyards_with_downy_milde w.

Lizotte, E., Sirrine, R., Hausbeck, M. K., Jess, L. and Zandstra, B. 2018. Michigan Hop Management Guide, 2018. Michigan State University Extension. Online.

Madden, L. V., Hughes, G., and van den Bosch, F. 2007. The Study of Plant Disease Epidemics. American Phytopathological Society, St. Paul, MN.

Marks, M., and Gevens, A. 2019. Investigating phenylamide insensitivity in Wisconsin populations of Pseudoperonospora humuli. Plant Health Prog. Online: https://doi.org/10.1094/PHP-08-19-0057-RS.

Massie, S. T., Claassen, B. J., and Gent, D. H. 2019. Evaluation of fungicides for hop downy mildew and powdery mildew, Prosser, Washington, 2018. Plant Dis. Manag. Rep. 13:V027.

Meyer, M. D., and Hausbeck, M. K. 2013. Using soil-applied fungicides to manage Phytophthora crown and root rot on summer squash. Plant Dis. 97:107-112.

Nanni, I. M., Pirondi, A., Contaldo, N., and Collina, M. 2016. Screening of sensitivity to mandipropamid of Plasmopara viticola populations from Italian vineyards by molecular and biological methods. Lett. Appl. Microbiol. 63:267-273.

Nelson, M. E., Eastwell, K. C., and Grove, G. G. 2003. Hop downy mildew control, 2002. Fungic. Nematic. Tests 59:FC015. Online.

Nelson, M. E., and Grove, G.G. 2004. Evaluation of fungicides for control of hop downy mildew, 2003. Fungic. Nematic. Tests 59:FC017. Online.

Nelson, M. E., and Grove, G. G. 2006. Evaluation of fungicides for control of hop downy mildew, 2005. Fungic. Nematic. Tests 61:FC046. Online.

Nelson, M. E., and Grove, G. G. 2007. Evaluation of fungicides for control of hop downy mildew, 2006. Plant Dis. Manag. Rep. 1:FC025. Online.

Nelson, M. E., and Grove, G. G. 2008. Evaluation of fungicides for control of hop downy mildew, 2007. Plant Dis. Manag. Rep. 2:FC051. Online.

Nelson, M. E., and Grove, G. G. 2009. Evaluation of fungicides for control of hop downy mildew and powdery mildew, 2008. Plant Dis. Manag. Rep. 3:FC008. Online.

Nelson, M. E., and Grove, G. G. 2010. Evaluation of fungicides for control of hop downy mildew, 2009. Plant Dis. Manag. Rep. 5:FC002. Online.

Nelson, M. E., Garcia, M., and Grove, G. G. 2011. Evaluation of fungicides for control of hop downy mildew, 2010. Plant Dis. Manag. Rep. 5:FC076. Online.

Neve R. A. 1991. Hops. UK: Chapman and Hall, London.

O’Neal, S. 2015. Pest Management Strategic Plan for U.S. Hops: Summary of a workshop held on November 4, 2014 in Yakima, WA. National Integrated Pest Management Database. Online. https://ipmdata.ipmcenters.org/documents/pmsps/US-hops-PMSP2015.pdf

Pavelková, J., Lebeda, A., and Sedlá ková, B. 2014. Efficacy of fosetyl-Al, propamocarb, dimethomorph, cymoxanil, metalaxyl and metalaxyl-M in Czech Pseudoperonospora cubensis populations during the years 2005 through 2010. Crop Prot. 60:9-19.

Rahman, A., Gongora-Castillo, E., Bowman, M. J., Childs, K. L., Gent, D. H., Martin, F. N., and Quesada-Ocampo, L. M. 2019. Genome sequencing and transcriptome analysis of the hop downy mildew pathogen Pseudoperonospora humuli reveal species-specific genes for molecular detection. Phytopathology. 109:1354-1366.

Royle, D. J. 1970. Infection periods in relation to the natural development of hop downy mildew (Pseudoperonospora humuli). Ann. App. Biol. 66: 281-291.

Royle, D. J., and Kremheller, H. TH. 1981. Downy Mildew of the Hop. Pages 395-419 in: The Downy Mildews. D. M. Spencer, ed. Academic Press, New York.

Royle, D. J., and Thomas, G. G. 1971. The influence of stomatal opening on the infection of hop leaves by Pseudoperonospora humuli. Physiol. Plant Pathol. 1:329-343.

Russell, P. E. 2002. Sensitivity baselines in fungicide resistance research and management. FRAC, Crop Life International, Brussels, Monogr. 3:1-60.

Salmon, E. S., and Ware, W. M. 1925. The downy mildew of the hop and its epidemic occurrence in 1924. Ann. App. Biol. 12:121-151.

Sambrook J., Fritsch E .F., and Maniatis T., 1989. Molecular Cloning: A Laboratory Manual. 2nd edition. New York, USA: Cold Spring Harbor Laboratory Press.

Sirrine, R., Lizotte, E., Brown, D., O’Brien, T., and Leach, A. 2014. Estimated Costs of Producing Hops in Michigan. Michigan State University Extension E3236. Online.

Skotland, C. B. 1961. Infection of hop crown and roots by Psuedoperonospora humuli and its relation to crown and root rot and overwintering of pathogen. Phytopathology 51, 241- 244.

Skotland, C. B., and Johnson, D. A. 1983. Control of downy mildew of hops. Plant Dis. 67:1183- 1185.

Thomas, P. D., Campbell, M. J., Kejariwal, A., Huaiyu , M., Karlak, B., Daverman, R., Diemer, K., Muruganujan, A., and Narechania, A. 2003. PANTHER: A Library of Protein Families and Subfamilies Indexed by Function. Genome Res. 13:2129-2141.

Thomas, A., Neufeld, K. N., Seebold, K. W.., Braun, C. A., Schwarz, M. R., and Ojiambo, P. S. 2018. Resistance to fluopicolide and propamocarb and baseline sensitivity to ethaboxam among isolates of Pseudoperonospora cubensis from the eastern United States. Plant Dis. 102:1619-1626.

Toffolatti, S. L., Venturini, G., Campia, P., Cirio, L., Bellotto, D., and Vercesi, A. 2014. Sensitivity to cymoxanil in Italian populations of Plasmopara viticola oospores. Pest Manag. Sci. 71:1182-1188.

Toffolatti, S. L. Russo, G., Campia, P., Bianco, P. A., Borsa, P., Coatti, M., Torriani, S. F. F., and Sierotzki, H. 2018. A time-course investigation of resistance to the carboxylic acid amide mandipropamid in field populations of Plasmopara viticola treated with anti- resistance strategies. Pest. Manag. Sci. 74:2822-2834. van den Bosch, F., and Gilligan, C. A. 2008. Models of fungicide resistance dynamics. Annu. Rev. Phytopathol. 46:123-147.

Ware, W. M. 1926. Pseudoperonospora humuli and its mycelial invasion of the host plant. Trans. Brit. Mycol. Soc. 11:91–107.

Woods, J. L., and Gent, D. H. 2016. Susceptibility of hop cultivars to downy mildew: Associations with chemical characteristics and region of origin. Plant Health Prog. 17:42- 48.

CHAPTER 3. SUSCEPTIBILITY OF HOP CULTIVARS AND ROOTSTOCK TO DOWNY MILDEW CAUSED BY PSEUDOPERONOSPORA HUMULI

Abstract

Annual downy mildew epidemics threaten hop production throughout the eastern United

States. Rootstock rot complicates foliar disease assessments since dormant buds rot before producing a symptomatic basal shoot whereas noncolonized buds still produce healthy shoots.

We selected twelve cultivars to evaluate for downy mildew susceptibility (2016 and 2017) and examined the stolons of a subset of six cultivars (2018), that showed clear differences in foliar disease ratings, to determine rootstock rot susceptibility. Trials were conducted on nontrellised hop yards established at two research farms in 2016 and managed without fungicides. rAUDPC values for disease severity and density ratings were consistently higher for ‘Cascade’,

‘Centennial’ and ‘Nugget’ than ‘Newport’, ‘Tahoma’ and ‘Columbia’. Only ‘Centennial’ had more wet-rot cortex discoloration (CD) incidence (71.5%; P = 0.0001) and severity (31.9%; P =

0.0031) in its stolons than the other cultivars. ‘Newport’ and ‘Columbia’ were less vigorous than

‘Tahoma’, but similar for wet-rot CD (incidence = 38.5 to 46.4%; severity = 12.5 to 17.7%). The level of wet-rot CD in ‘Tahoma’, ‘Newport’ and ‘Columbia’ was comparable to cultivars with more severe foliar disease symptoms (‘Nugget’ and ‘Cascade’). Differences in foliar disease among cultivars with a similar levels of rootstock rot suggests a resistance defense mechanism to

P. humuli. Additional work is needed to differentiate tolerance from intermediate resistance, clarify rootstock infection sites, screen for compatibility with P. humuli isolates, and characterize defense responses to aid future breeding efforts.

Introduction

Niche hop (Humulus lupulus L. var lupulus) production occurs throughout the eastern

United States (George 2020). However, the wet and humid environment in these regions incite annual epidemics of downy mildew (DM) (Pseudoperonospora humuli (Miyabe & Takah.) G.W.

Wilson, (1914)) that limits production (O’Neal 2015). DM is a chronic systemic disease that rots and weakens hop rootstock (perennial underground stem organs, crown, and accompanying roots) depleting winter carbohydrate reserves needed for vigorous perennial growth (Williams et al. 1961) and reducing yield up 56.2% (Coley-Smith 1962). Hypertrophic apical buds cease to develop requiring additional labor costs to retrain replacement bines (Ware 1926). Direct yield and quality loss occur when infected inflorescences abort, develop discolored bract, and lose alpha acid content (Neve 1991; Royle and Kremheller 1981).

DM is incited by a polycyclic airborne pathogen that overwinters primarily in the rootstock (Coley-Smith 1962; Johnson and Skotland 1985; Skotland 1961; Ware 1926). The pathogen can colonize rootstock from infections that occur at the tip and base of a shoot (Coley-

Smith 1962; Coley-Smith 1965; Ware 1926; Williams et al. 1961). Zoosporangia can infect storage roots directly (Coley-Smith 1965) although the infection process is unknown. . DM symptoms in rootstock produce red-brown to dark brown and black wet-rot that can appear as water-soaked flecks and streaking but more often is observed as large discolored wet-rot areas in the cortex tissue (Coley‐Smith 1964; Skotland 1961; Ware 1926). Infected rootstock buds that are colonized give rise to symptomatic basal shoots with swollen stems and chlorotic downward cupped leaves (Coley-Smith 1962; Coley‐Smith 1964; Skotland 1961; Ware 1926). Sporangia develop primarily on abaxial leaf surfaces under high relative humidity (Johnson and Skotland

1985) and are disseminated by wind and splashing water (Royle 1970; Sonoda and Ogawa

1972). Water triggers indirect infection via swimming zoosporangia that penetrate via stomata openings (Royle and Thomas 1971a, 1973; Royle and Thomas 1971b). Symptomatic leaves, the result of secondary pathogen spread, exhibit chlorotic lesions limited by the leaf vein (Neve

1991; Royle and Kremheller 1981). Secondary shoot infections result in symptoms similar to basal shoots but are distinguishable by one or more normally elongated internode at the shoot base (Coley-Smith 1962; Skotland 1961).

Breeding for resistance to DM in diploid, dioecious H. lupulus has received limited domestic attention (Neve 1991) until recently (Henning et al. 2018; Henning et al. 2015, 2016;

Woods and Gent 2016). All hop cultivars appear to be susceptible to DM but cultivars exhibiting low levels of disease are often considered “resistant” (Henning et al. 2018; Mahaffee et al. 2009;

Woods and Gent 2016). DM resistance in H. lupulus appears to be under quantitative genetic control by multiple loci (Henning et al. 2015). The registration of a DM-resistant male line

(USDA 21087M) offers new crossing opportunities with DM-resistant female lines (Henning et al. 2018). The identification of DM-resistant female lines adapted to local growing conditions is needed (Henning et al. 2018; Woods and Gent 2016).

An inverse relationship between crown rot susceptibility and the incidence of symptomatic primary shoots complicates field assessments of DM susceptibility (Neve 1991;

Woods and Gent 2016). Cultivars highly susceptible to rootstock rot may show no foliar disease symptoms (Coley‐Smith 1964; Neve 1991; Woods and Gent 2016). Dormant buds rapidly rot before producing a symptomatic basal shoot, but noncolonized buds still produce healthy shoots

(Neve 1991). While the rootstock of some susceptible hop cultivars succumb to DM rot, others appear to tolerate infection and produce a mix of healthy and diseased shoots (Coley-Smith

1962; Coley‐Smith 1964; Johnson and Anliker 1985; Skotland 1961; Woods and Gent 2016).

Woods and Gent (2016) proposed an indirect method to detect variation in susceptibility to rootstock rot by combining disease assessments of symptomatic DM shoots (incidence) with measurements of plant vigor (total number of shoots produced). High vigor cultivars with shoots displaying minor DM symptoms were considered resistant. However, low vigor cultivars are difficult to classify since their growth rate might be a result of rootstock rot susceptibility to DM or poor adaptation of the cultivar to its environment (Woods and Gent 2016).

Local Michigan hop propagators offer 45 to 54 different cultivars to regional growers. In some cases, highly desirable brewing characteristics may incentivize the planting of DM susceptible cultivars (Erin Lizzotte, personal communication). In conjunction with the statewide association Hop Growers of Michigan we selected twelve cultivars to evaluate for DM susceptibility. We examined lateral growing stolons of a subset of six cultivars for DM symptoms; cultivars were chosen based on differences in foliar DM. Preliminary research results have been presented (Higgins and Hausbeck 2017, 2018).

Materials and Methods

Field Plots. Evaluations were conducted on non-trellised hop plants at the Michigan

State University (MSU) Southwest Michigan Research and Extension Center (SWM), Benton

Harbor, MI (latitude 42.6854°, longitude -84.4716°) and replicated at the MSU Plant Pathology

Farm (PPF), Lansing, MI (latitude 42.0835°, longitude -86.3542°). Plots were established in 52.3 m rows of raised plant beds covered with black plastic and buried drip irrigation. All rows were spaced 2.4 m apart. Each treatment plot consisted of 10 plants in a row with 1.1 m between plants and a 0.8 m walkway between plots. A single buffer row of nontreated ‘Nugget’ (Haunold et al. 1984) plants bordered each experiment. Hop plantlets, propagated from softwood cuttings

by a commercial propagator, were transplanted on 3 (PPF) and 6 Jun (SWM) 2016 and grown until stolons were destructively extracted in 2018. Hop shoots extending into the center aisles were periodically cut back, with a mechanical hedge trimmer (STIHL, Inc., Virginia Beach,

VA.) to a uniform distance outside of the rating area. Plots were drip-irrigated as necessary.

Fertilizer was applied according to soil tests in 2016 and 2017 as a combination of granular fertilizer 6-24-24 (N-P2O5) and urea 46-0-0 (N) prior to bed formation and MORA-LEAF® Plus

20-20-20 (N - P2O5 - K2O) with micronutrients (Wilbur-Ellis, San Francisco, CA), Liquid Re-

Nforce K® 5-0-20 (N-K2O) with sulfur (Loveland Products, Inc., Greenly, CO) and urea 46-0-0

(N) fertilizer dissolved in water and injected through the drip line. Weeds were managed through mechanical cultivation and fungicides were not applied.

Plots were inoculated in 2016 on 15, 22, and 28 Jun (PPF), 21 Jun and 8 Jul (SWM).

Plots received inoculum only in 2016 and disease progressed under natural conditions in 2017 and 2018. Inoculum was prepared from diseased shoots collected 24 h prior to inoculation from three commercial hop yards according to methods described by Gent et al. (2008). Zoospore suspensions were strained through two layers of cheese cloth, added to 15.1 l of water (Nelson et al 2004) and injected into overhead mist irrigation using a CO2 pressurized container. The concentration of zoospore suspension was 2.6 x 104 sporangia/ml for inoculations on 15 Jun

(PPF) and 2.0 x 104 for inoculations on 21 Jun (SWM) and 22 Jun (PPF). Inoculum for inoculations on 28 Jun (PPF) and 7 Jul (SWM) was prepared from 60 to 80 diseased shoots and approximately 5.2 x 106 to 6.1 x 106 sporangia/ml were injected into the overhead mist irrigation.

Plots were misted in 2016 on 7, 11, 21, 23 Jul and 15 Aug (SWM) and on 24, 29 Jun and 11 Jul

(PPF) to create extended durations of leaf wetness conducive to DM.

Treatments were arranged in a randomized complete block design with four replications.

Treatments were the hop cultivars including, ‘Tahoma’ , ‘Fuggle H’ (U.S.), ‘Mt. Rainier’,

‘Newport’ (Henning et al. 2004), ‘Perle’, ‘Nugget’ (Haunold et al. 1984), ‘Sorachi Ace’,

‘Cascade’ (Brooks et al. 1972), ‘Columbia’ (Haunold et al. 1976), ‘Triple Pearl’, ‘Centennial’

(Kenny and Zimmermann 1991), and ‘Comet’ (Zimmermann et al. 1975). In 2016, ‘Fuggle H.’,

‘Sorachi Ace’, and ‘Mt. Rainier’ were removed from the experiments because damage from two- spotted spidermites (Tetranychus urticae) at one or both locations in 2016 interfered with DM ratings. Subsequently, two-spotted spider mites and potato leaf hoppers (Empoasca fabae) were managed with applications of bifenazate, bifenthrin, fenazaquin, and imidacloprid. ‘Perle’ was also removed from the experiments in 2016 at both locations because it tested positive for hop latent virus, apple mosaic virus, and American hop latent virus.

Foliar Disease Ratings. In each plot, the inner six of the ten plants per replication were assessed for foliar disease severity and density (2016 and 2017), and plant vigor (2017). If newly transplanted plants died prior to inoculation or did not overwinter for the 2017 assessment, one of the remaining four plants in the plot was assessed. When plants within each plot began to grow together, a frame (1.2 x 0.9 m) constructed of PVC pipe (1.3 cm) was placed over the center of each plant. The corners of the frame were marked in the plot with wooden stakes to ensure subsequent measurements included the same area. Frames were used to rate all experiments from 18 (SWM) and 19 Aug (PPF) in 2016 and 29 May (SWM) and 7 Jun (PPF) in

2017. Disease severity, measured as a combination of symptomatic shoots and leaves, was estimated visually using the Horsfall-Barratt scale of 1 to 12, where 1=0% plant area diseased,

2=>0 to 3%, 3= >3 to 6%, 4=>6 to 12 %, 5 = >12 to 25 %, 6=>25 to 50%, 7=>50 to 75%, 8=>75 to 87%, 9=>87 to 94%, 10=>94 to <100%, 12=100% plant area (Horsfall and Barratt 1945).

Prior to statistical analysis, rating values from the Horsfall-Barratt scale were converted to

midpoint values. Disease density was determined by the number of symptomatic shoots per plant

(Gent et al. 2012). Plant vigor was the average number of shoots per plant (Woods and Gent

2016). Prior to statistical analysis, rating values from the Horsfall-Barratt scale were converted to midpoint values. In 2016, disease severity ratings were performed on 20, 28 Jul; 4, 11, 18, 25

Aug; 1, 8, 15, 22 Sep; and 6 Oct (SWM) and on 22, 29 Jul; 5, 12, 19, 26 Aug; 2, 9, 16, 23 Sep; and 7 Oct (PPF). In 2017, disease severity ratings were performed on 9, 16, 23, 30 May; 6, 13,

20, 27, 11, 18, 25 Jul; and 1 Aug (SWM) and on 11, 18, 24, May; 1, 8, 15, 22, 29 Jun; 6, 13, 20,

27 Jul; and 3 Aug (PPF). Plant vigor ratings were taken on 11, 18, 24 May; 1, 8, 15, 22 Jun

(PPF) and on 9, 16, 23, 30 May; 6, 13, 20 Jun (SWM).

Data from disease severity and density were summarized for each treatment level by calculating the relative AUDPC (rAUDPC) (Madden et al., 2007). Statistical analysis was preformed using SAS software, Version 9.4 (SAS Institute, Inc., Cary, NC) with a three-factor linear mixed model to account for repeated measures between years. Year (2016 and 2017), location (PPF and SWM), cultivar and two- (year × location; year × cultivar; location × cultivar) and three-way (year × location × cultivar) interactions were considered fixed effects. Blocks nested in location and two- and three-way interactions between blocks and fixed effects were considered random effects. Global analysis of variance (ANOVA) was conducted using PROC

GLIMMIX. The standard error and degrees-of-freedom correction were predicted using the

Kenward and Roger option. Assumptions of normality and equal variances were confirmed with residual plots and the Levene’s test, respectively. Data were transformed using the lambda value

(0.65 [severity]; 0.25 [density]) derived from a Box-Cox analysis (Box and Cox 1964). The

GLIMMIX procedure was adjusted to estimate the variances separately and account for variance heterogeneity. For simple effects, the SLICE option was used to run ANOVA separately for each

factor and the SLICEDIFF option was used to make all possible pairwise differences amongst levels of one factor. Least squared means comparisons were preformed using Tukey’s adjusted p-values. All transformed data were back transformed for presentation in tables.

Statistical analysis for plant vigor was performed with two-factor linear mixed model.

Location and cultivar were fixed factors. Blocks were nested in location and considered a random factor. The GLIMMIX procedure was adjusted to estimate the variances separately and account for variance heterogeneity. Least squared means comparisons were preformed using

Tukey’s adjusted p-values. All transformed data were back transformed for presentation in tables.

Stolon disease ratings. In 2018, stolons were collected at PPF (n = 590) from 20 Jun to 4

Jul and SWM (n = 411) from 30 Jul to 11 Aug. Stolons were extracted from the replicated plot of six hop cultivars (Nugget, Centennial, Cascade, Tahoma, Columbia and Newport) at each location. Stolons were extracted from the 3rd, 5th and 7th plant from each treatment plot. Stolons were exposed by flooding the plant with water from a garden hose or portable pressure washer.

Up to ten laterally growing stolons, identified by a pair of buds at regularly spaced internodes, were arbitrarily chosen from each plant. Stolons, approximately 14 cm in length, were cut close to the attachment point of the crown with a sterile surgical blade. If a plant had fewer than 10 horizontal stolons the maximum number of horizontal stolons present were removed. If a plant was dead, stolons from a neighboring plant were sampled. Plants in two ‘Nugget’ replicate plots at SWM where either completely dead or plants growth was so poor that only 5 stolons were present. These plots were removed from the statistical analysis. Stolons were placed in ziplock bags and then into a cooler with ice for transport back to the laboratory. One replication at a time was sampled and then processed. Samples were refrigerated and processed within 48 to 72 hrs.

Stolons were washed with tap water and their diameter recorded with a digital caliper

(Husky, Atlanta, GA). Stolons were split longitudinally with a sterile surgical blade and evaluated for symptoms typically associated with DM. Disease incidence and severity were taken separately for wet-rot discoloration located in the cortex (cortex rot = CD) or vascular or pith tissue (vascular/pith discoloration = VPD). Disease incidence was determined by the number of symptomatic stolons from the total number of stolons. Disease severity was estimated directly

(0-100 %) with the aid of a disease diagram (Figs. 2.1 and 2.2) as the portion of cortex tissue with wet-rot discoloration or vascular/pith tissue discoloration.

Statistical analysis was performed with two-factor linear mixed model. Location (PPF and SWM) and cultivar were fixed factors. Blocks were nested in location and considered a random factor. ANOVA was conducted using PROC GLIMMIX for the main effects of location and cultivar, and the interaction between location and cultivar. The standard error and degrees- of-freedom correction were predicted using the Kenward and Roger option. Assumptions of normality and equal variances were confirmed with residual plots and the Levene’s test, respectively. Data sets for VPD incidence and severity and CD severity were square root transformed to meet assumptions; data for stolon size were transformed with the lambda value (-

0.85) obtained from a Box-Cox analysis. Multiple comparisons were performed on the main effects for insignificant interactions (P = 0.05). Significant interactions were examined by cell means and comparisons among simple effects. Least squared means comparisons were performed using Fisher’s protected least significant differences (LSD). All transformed data were back transformed for presentation in tables. Phenotypic correlations were estimated from the error due to treatment effects of disease ratings and stolon size.

Results

Foliar Disease Ratings. For disease severity, the three-way interaction (year × location × cultivar; F = 2.6; P = 0.0504) and two-way interaction (location × cultivar; F = 1.71; P = 0.1467) were significant; simple effects for each cultivar were analyzed by year (Table 1). rAUDPC values were significantly lower in 2017 than 2016 for all cultivars. Cultivar had a significant treatment effect in both 2016 (F = 18.23; P < 0.0001) and 2017 (F = 43.32; P < 0.0001). In 2016 and 2017, disease severity (rAUDPC) was greater in ‘Centennial’, ‘Cascade’ and ‘Nugget’ than

‘Columbia’ ‘Tahoma’ and ‘Newport’. ‘Triple Pearl’ and ‘Comet’ rAUDPC values (disease severity) were significantly higher than ‘Columbia’, ‘Tahoma’, and ‘Newport’ in 2017, but only

‘Triple Pearl’ had a higher rAUDPC value than ‘Tahoma’ in 2016.

For disease density, the three-way interaction (year × location × cultivar; F = 0.73; P =

0.6490) and two-way interaction (location × cultivar; F = 0.94; P = 0.4910) were significant; simple effects for each cultivar were analyzed by year (Table 2.1). rAUDPC values were significantly lower in 2017 than 2016 for all cultivars except ‘Comet’ which had similar rAUDPC values in both years (F = 0.26; P = 0.6185). Cultivar had a significant treatment effect in both 2016 (F = 17.66; P < 0.0001) and 2017 (F = 52.87; P < 0.0001). In 2016 and 2017, disease density (rAUDPC) was greater in ‘Centennial’, ‘Cascade’ and ‘Nugget’ than ‘Columbia’

‘Tahoma’ and ‘Newport’. In 2016, rAUDPC values for disease density for ‘Triple Pearl’ did not differ from ‘Columbia’, ‘Tahoma’, and ‘Newport’. rAUDPC values for disease density (2016) for ‘Comet’ did not differ from all other cultivars. In 2017, rADUPC for disease density for

‘Triple Pearl’ and ‘Comet’ were higher than ‘Columbia’, ‘Tahoma’, and ‘Newport’; ‘Triple

Pearl’ had lower rAUDPC values than ‘Nugget’.

There was a significant interaction in mean plant vigor (shoots/plant; 2017 only) between locations (F = 3.69; P = 0.0205); simple effects for each cultivar were analyzed by location

(Table 2.1). Cultivar had a significant treatment effects at both PPF (F = 16.61; P < 0.0001) and

SWM (F = 9.10; P = 0.0004). At PPF, mean plant vigor was higher in ‘Tahoma’ (74.1 shoots/plant) , ‘Cascade’ (62.6 shoots/plant), ‘Centennial’ (56.0 shoots/plant), ‘Nugget’ (52.3 shoots/plant), and ‘Triple Pearl’ (52.6 shoots/plant) than ‘Comet’ (28.5 shoots/plant), ‘Columbia’

(31.4 shoots/plant) and ‘Newport’ (26.3 shoots/plant). At SWM, mean plant vigor was higher in

‘Tahoma’ (59.0 shoots/plant), ‘Cascade’ (50 shoots/plant), and ‘Triple Pearl’ (55 shoots/plant) than ‘Columbia’ (31.6 shoots/plant).

Stolon Disease Ratings. There was no interaction between location and cultivar for CD incidence (F= 1.61; P =0.1904) and severity (F = 1.66; P = 0.1769) and the main effects for each cultivar were analyzed (Table 2.2). For CD incidence, cultivar had a significant treatment effect

(F = 7.88; P = 0.0001). CD incidence for ‘Centennial’ (71.5%) was significantly higher than

‘Columbia’ (47.2%; P = 0.0024), ‘Newport’ (38.5%; P < 0.0001), ‘Tahoma’ (46.4%; P =

0.0017), ‘Nugget’ (33.7%; P < 0.0001) and ‘Cascade’ (32.3%; P < 0.0001). For CD severity, cultivar had significant treatment effects (F = 4.73; P = 0.0031). CD severity in ‘Centennial’

(31.9%) was significantly higher than ‘Columbia’ (17.7%; P = 0.0352), ‘Newport’ (13.9%; P =

0.0136), ‘Tahoma’ (12.5%; P = 0.0110), ‘Nugget’ (8.1%; P = 0.0005) and ‘Cascade’ (6.7%; P =

0.0002).

Table 2.1. Plant vigor and disease severity and density ratings in 2016 and 2017 for downy mildew (Pseudoperonospora humuli) symptoms in the foliage of hop (Humulus lupulus) cultivars. Disease severityw Disease densityx Mean plant vigory (rAUDPC) (rAUDPC) (Shoots/Plant) Cultivar 2016 2017 2016 2017 PPF (2017) SWM (2017) Cascade 0.357 az 0.118 b 0.109 a 0.042 ab 62.6 a 50.0 a Centennial 0.328 ab 0.183 a 0.112 a 0.049 ab 56.0 a 42.1 ab Nugget 0.288 abc 0.169 ab 0.080 a 0.052 a 52.3 a 44.3 ab Triple Pearl 0.254 bcd 0.116 b 0.043 b 0.023 b 52.6 a 55.5 a Comet 0.227 cde 0.112 b 0.059 ab 0.048 ab 28.5 b 48.7 ab Columbia 0.214 de 0.038 c 0.040 b 0.003 c 31.4 b 31.6 b Tahoma 0.206 de 0.019 c 0.043 b 0.000 d 74.1 a 59.0 a Newport 0.191 e 0.017 c 0.029 b 0.000 d 26.3 b 44.4 ab Treatment P < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001 0.0004 value W rAUDPC = relative area under the disease progress curve. Disease severity = percentage of symptomatic shoots and leaves estimated visually using the Horsfall-Barratt scale converted to midpoint values. Statistical analysis preformed on data from two locations and two years as a linear mixed model with unequal variances and repeated measures. No interactions amongst year × location × cultivar (F= 2.60; P =0.0504) and loc × cultivar (F= 2.60; P =0.0504); year × cultivar interaction was significant (F = 12.93; P = <0.0001). X Disease density = the number of symptomatic downy mildew shoot per plant. Statistical analysis preformed on data from two locations and two years as a linear mixed model with unequal variances and repeated measures. No interactions amongst year × location × cultivar (F= 0.73; P =0.6490) and loc × cultivar (F= 0.94; P =0.4910); year × cultivar interaction was significant (F = 12.82; P = <0.0001). Y Plant vigor = the average number of shoots per plant between assessed weekly in May and June 2017. Statistical analysis preformed on data from two locations as a linear mixed model with unequal variances. PPF = MSU Plant Pathology Farm, Lansing, MI (latitude 42.0835°, longitude -86.3542°); SWM = Michigan State University (MSU) Southwest Michigan Research and Extension Center, Benton Harbor, MI (latitude 42.6854°, longitude -84.4716°). Z Least squared means comparisons were preformed using Tukey’s adjusted p-values (P = 0.05). Column means with a letter in common are not significantly different.

There was a significant interaction between location and cultivar for VPD incidence (F=

6.62; P = 0.0004) and severity (F = 19.77; P = < 0.0001) and the simple effects for each cultivar were analyzed by location (Table 2.2). VPD incidence (F = 16.05; P = 0.0085) and severity (F =

16.23; P = 0.0072) was at least 2X higher at SWM than PFF for most cultivars; ‘Centennial’ and

‘Columbia’ had similar levels of disease symptoms both location. At PPF, cultivars had significant effects for VPD incidence (F = 4.65; P = 0.0034) and severity (F = 6.87; P = 0.0003).

VPD incidence (PPF) in ‘Tahoma’ (11.9%) was lower than ‘Centennial’ (47.8%; P < 0.0001),

‘Nugget’ (32.5%; P = 0.0057), ‘Columbia’ (33.0%; P = 0.0049), ‘Newport’ (32.6%; P = 0.0056), and ‘Cascade’ (30.2 %; P > 0.0109). VPD severity (PPF) in ‘Tahoma’ (1.2%) was lower than

‘Centennial’ (12.4%; P < 0.0001), ‘Nugget’ (4.4%; P = 0.0403), ‘Columbia’ (8.9%; P = 0.0004),

‘Newport’ (9.0%; P = 0.0003), and ‘Cascade’ (4.9 %; P = 0.0222). In ‘Centennial’, VPD severity (PPF) was higher compared to ‘Nugget’ (P = 0.005) and ‘Cascade’ (P = 0.0097), but similar to ‘Columbia’ (P = 0.2564) and ‘Newport’ (P = 0.2786).

At SWM, cultivars had significant effects for VPD incidence (F = 12.38; P < 0.0001) and severity (F = 23.97; P < 0.0001). VPD incidence (SWM) was highest in ‘Nugget’ (98.0%) and

‘Cascade’ (74.8%); ‘Cascade’ did not differ from ‘Newport’ (62.2 %; P = 0.3180) and

‘Centennial’ (56.2 %; P = 0.1337). VPD incidence (SWM) in ‘Tahoma’ (32.1%) and ‘Columbia’

(17.5%) was lower than all other cultivars. VPD severity (SWM) was the highest in ‘Nugget’

(57.8%) and the lowest in ‘Columbia’ (3.8%) compared to other cultivars. VPD severity (SWM) in ‘Newport’ (20.4%) and ‘Cascade’ (28.0%) was greater than ‘Centennial’ (12 %; P = 0.0318 and 0.0005) and ‘Tahoma’ (9.8 %; P = 0.0063 and < 0.0001).

There was a significant interaction between location and cultivar for stolon size (F =

19.05; P = 0.0001) and the simple effects for each cultivar were analyzed by location (Table

2.2). There were no significant differences in stolon size between locations (F = 2.68; P =

0.1290). There were significant differences (F = 57.01; P < 0.0001) among stolon size for cultivars at both PPF (F = 20.09; P < 0.0001) and SWM (F = 51.39; P < 0.0001). At PPF, the average stolon diameter of ‘Columbia’ (11.8 mm) was larger than ‘Nugget’ (9.9 mm; P =

0.0121), ‘Tahoma’ (9.3 mm; P < 0.0001), and ‘Cascade’ (8.1 mm; P < 0.0001). At SWM, the stolon diameters of ‘Nugget’ (10.9 mm), ‘Centennial’ (10.0 mm), and ‘Columbia’ (9.9 mm) were larger than ‘Cascade’ (8.3 mm; P = 0.0037, 0.0458, and 0.0084) and ‘Tahoma’ (7.2 mm; P =

0.0001, 0.0004, and 0.0001); Cascade was similar to Newport (9.8 mm; P = 0.4612).

Significant positive correlations were observed between symptom incidence and severity for CD (Correlation = 0.855; P < 0.0001) and VPD (Correlation = 0.748; P < 0.0001) (Table

2.3). The diameter size of stolon was positively correlated with CD incidence (Correlation =

0.403; P = 0.0055) and severity (Correlation = 0.439; P = 0.0023). No correlations were observed between VPD symptoms and stolon size.

Table 2.2. Downy mildew (Pseudoperonospora humuli) disease incidence and severity and stolon size in 2018 for hop (Humulus lupulus) cultivars. Wet-rot CDv VPDw Stolon sizex PPFy SWM Cultivar Incidencez Severity Incidence Severity Incidence Severity PPF SWM Centennial 71.5 a 31.9 a 47.8 a 12.4 a 56.2 b 12.0 c 10.7 ab 10.0 a Nugget 33.7 b 8.1 b 32.5 a 4.4 b 98.0 a 57.8 a 9.9 b 10.9 a Cascade 32.3 b 6.7 b 30.2 a 4.9 b 74.8 ab 28.0 b 8.1 b 8.3 b Columbia 47.2 b 17.7 b 32.4 a 8.9 ab 17.5 c 3.8 d 11.8 a 9.9 a Newport 38.5 b 13.9 b 32.6 a 9.0 ab 62.2 b 20.4 b 10.9 ab 9.8 ab Tahoma 46.4 b 12.5 b 11.9 b 1.2 c 32.1 c 9.8 c 9.3 b 7.2 c Treatment 0.0001 0.0031 0.0034 0.0003 < 0.0001 < 0.0001 <0.0001 <0.0001 P value v CD = cortex discoloration. Statistical analysis preformed on data from two locations as a linear mixed model. No interaction between location and cultivar for cortex incidence (F= 1.61; P =0.1904) and severity (F = 1.66; P = 0.1769). w VPD = vascular/pith discoloration; occurring as slightly-water soaked red-brown and flecking and streaking. Statistical analysis preformed on data from two locations as a linear mixed model. Significant interaction between location and cultivar for vascular/pith discoloration incidence (F= 6.62; P = 0.0004) and severity (F = 19.77; P = < 0.0001). x Stolon size = diameter of stolon (mm). Statistical analysis preformed on data from two locations as a linear mixed model. Significant interaction between location and cultivar for stolon size (F = 19.05; P = 0.0001). Y PPF = MSU Plant Pathology Farm, Lansing, MI (latitude 42.0835°, longitude -86.3542°); SWM = Michigan State University (MSU) Southwest Michigan Research and Extension Center, Benton Harbor, MI (latitude 42.6854°, longitude -84.4716°). Z Disease incidence = the number of stolons with symptoms out of the total number of stolons. Disease severity = Direct estimate (0 - 100%) with the aid of a disease diagram of the portion of cortex or vascular/pith tissue covered downy mildew symptoms. Fisher’s protected least significant differences at P < 0.05 were used to make comparisons least squares means of fungicides. Column means with a letter in common are not significantly different.

Table 2.3. Pearson’s correlation coefficient among stolon size and downy mildew (Pseudoperonospora humuli) symptoms in the stolons of hop (Humulus lupulus) cultivars. CDw incidence CD severity VPDx incidence VPD severity Diametery CD Incidence 1.000z CD Severity 0.855*** 1.000 VPD Incidence 0.132 0.257 1.000 VPD Severity 0.131 0.132 0.748*** 1.000 Diameter 0.403** 0.439** 0.125 0.038 1.000 W CD = Wet-rot cortex discoloration. X VPD = Slightly-water soaked vascular/pith discoloration; often occurring as red-brown and flecking and streaking. Y Diameter = stolon diameter. Z Disease incidence = The number of stolons with symptoms out of the total number of stolons. Disease severity = Direct estimate (0 - 100%) with the aid of a disease diagram of the portion of cortex or vascular/pith tissue covered downy mildew symptoms. *, **, and ** indicate value is statistically significant at P < 0.05, 0.01, and 0.001, respectively. N = 46.

Discussion

An inverse relationship between crown rot susceptibility and the incidence of symptomatic primary shoots complicates field assessments of DM susceptibility (Neve 1991;

Woods and Gent 2016). It is difficult to determine whether low vigor cultivars are a result of susceptibility to rootstock rot or poor adaptation to their environment. This study combined foliar

DM disease ratings (symptomatic shoots and leaf lesions) with an assessment of belowground

DM symptoms to determine cultivar susceptibility to rootstock rot regardless of plant vigor.

‘Centennial’, ‘Nugget’, and ‘Cascade’ consistently had more severe foliar symptoms than

‘Columbia’, ‘Tahoma’, and ‘Newport’, and the stolons of these cultivars were used for the assessment of rootstock rot. Only ‘Centennial’ had more wet-rot CD in its stolons than the other cultivars.

In this study, ‘Centennial’ had significantly higher amount of wet-rot CD compared to

‘Nugget’ and ‘Cascade’ despite similarly high levels of foliar symptoms. ‘Centennial’, ‘Nugget’, and ‘Cascade’ were previously reported to produce high levels of foliar disease (basal shoots) with ‘Centennial’ being less vigorous but not conclusively more susceptible to rootstock rot then the others (Woods and Gent 2016). Equivalent levels of foliar symptoms suggest these cultivars are equally susceptible to colonization that results in symptomatic shoots. If resistance acts directly on limiting pathogen multiplication, but tolerance affects the host’s ability to reduce the negative effects of infection (Pagán and García-Arenal 2020), then cultivars with similar levels of foliar symptoms but reduced rootstock rot may be expressing a tolerance response. However, it is also possible that difference in wet-rot CD is due to resistance to rootstock infection. There is circumstantial evidence to suggest that hop rootstock infections could occur from P. humuli zoospores washing through the soil (Coley-Smith 1965; Royle and Kremheller 1981) , but direct

infection has yet to be characterized for belowground organs. Tolerance warrants attention because it may reduce selection pressure on a pathogen population and may be a more stable phenotype than resistance (Pagán and García-Arenal 2020). Future studies are needed to clarify infection sites and differentiate tolerance from intermediate resistance using quantifiable pathogens loads and/or comparisons with noninfected controls (Pagán and García-Arenal 2020).

The level of wet-rot CD in ‘Tahoma’, ‘Newport’ and ‘Columbia’ was comparable to some cultivars with more severe foliar disease symptoms (e.g. ‘Nugget’ and ‘Cascade’).

’Columbia‘ was previously reported as having low vigor and low levels of foliar disease (basal shoots) but not conclusively more susceptible to rootstock rot than other cultivars (Woods and

Gent 2016). Here we show that wet-rot CD in ‘Columbia’ is lower than ‘Centennial’ and comparable to cultivars with higher vigor such as ‘Tahoma’. ‘Newport’ was noted for a lack of

DM symptoms during breeding trials in Oregon and Washington but was never inoculated

(Henning et al. 2004). ‘Newport’ parental heritage of European origin ‘Hallertauer Magnum’

(Henning et al. 2004) is linked to DM resistance (Woods and Gent 2016). ‘Tahoma’ is aroma hop and a newer release with limited information on resistance to DM (Hop Growers of

America). Reduced foliar disease symptoms in hop cultivars displaying similar levels of rootstock rot suggest that a resistance defense mechanism may be limiting foliar DM symptoms.

One possibility is that all rootstock is susceptible to infection by P. humuli, but cultivars with reduced shoot symptoms are resistant to systemic colonization. In Plasmopara halstedii, all pathogen races can infect sunflower roots but systemic growth up the hypocotyl and colonization of the cotyledon is governed by compatible pathogen/host interactions and the extent of colonization results in two distinct resistant phenotypes (Mouzeyar et al. 1993; Mouzeyar et al.

1994; Radwan et al. 2011). The formation of tyloses, as seen in interactions between

Verticillium albo-atrum and hop, can limit colonization of the vascular system (Talboys 1958).

Defense responses observed in other host/downy mildew interactions, such as a hypersensitive reaction (HR) in the parenchyma (Mouzeyar et al. 1993), callose- and lignin-like encasement of hyphae and haustoria (Cohen et al. 1989), lignification and peroxidase activity in the parenchyma (Asada and Matsumoto 1969; Ohguchi and Asada 1975), or an accumulation of saylic acid in the vascular system (Yan et al. 2020), may act to limit systemic colonization.

Screening for compatibility with P. humuli isolates and characterizing defense responses in rootstock may help explain differences in susceptibility to basal shoot infections of hop cultivars with similar tolerances to rootstock rot.

In this study, there was a significant positive correlation between stolon size and the wet- rot CD disease ratings. During ontogenesis, stolons produce a 2-6 mm ring of annual growth

(Rybaček 1991). The average diameter of stolons (> 7 mm) in all cultivars indicated that most of the stolons sampled were older than one year. As a stolon reaches maturity, typically in year four, there is an increasing rate of tissue death (Rybaček 1991) which may in part explain the strong positive correlation between stolon size and wet-rot CD. Yet, there are physiological and biochemical differences between young (1 year or less) and old stolons (Rybaček 1991) that may increase disease susceptibility as stolons mature. For instance, total monosaccharides drop 2X in young compared to older stolons (Rybaček 1991). The rare sugar monosaccharide D-tagatose can disrupt the mannose metabolism in Hyaloperonospora arabidopsidis and appears to have fungicidal effects on P. cubensis (Mochizuki et al. 2020); D-tagatose was found in root exudates of maize seedlings (Canellas et al. 2019).

It should be noted that wet-rot CD in rootstock is also caused by Phytophthora spp.

(Brien 1938; Royle 1968; Sonoda and Ogawa 1968) and may confound cultivar comparisons

100

based on rot symptoms since very little is known about cultivar susceptibility to Phytophthora spp. A molecular diagnostic tool would be helpful to distinguish wet-rot CD symptoms caused by the two pathogens. Yet, current molecular diagnostic markers for P. humuli (Gent et al. 2009;

Patzak 2005; Rahman et al. 2019; Summers et al. 2015) have not been validated in rootstock tissue that may contain PCR inhibitors leading to a false negative diagnosis (Gao et al. 2004).

The Phytophthora genus specific atp9-nad9 marker system (Bilodeau et al. 2014; Miles et al.

2017) would need to be checked for specificity against DM species to decrease the potential risk of a false positive diagnosis. Further work to validate molecular diagnostic markers would an important next step to understand the wet-rot CD response of hop cultivars.

VPD symptoms did not always follow a clear pattern among cultivars. In ‘Centennial’ where we might expect the highest VPD incidence and severity the cultivar did not separate from the others and differences among cultivars varied by location. For the most part VPD in

‘Tahoma’ was lower than ‘Centennial’; VPD severity at SWM was an exception. We observed a range of symptom color, reddish-brown, dark-brown or black, in both CD and VPD in agreement with previous observations that considered the darkening of symptoms to represent the progression of the disease (Coley-Smith 1962; Skotland 1961). Coley‐Smith (1964) noted that in DM-infected rhizomes large areas of wet-rot discoloration in the cortex was more common than the flecking symptom, but that in latter stages of infection both symptoms may occur throughout all tissue types. We devised our rating scale to distinguish between the two distinct regions, cortex or vascular/pith tissue, where symptoms occurred. Differences in the location of DM symptoms may be linked to varying defense responses among cultivars or difference in the initial site of infection. For ‘Nugget’, ‘Cascade’, and ‘Newport’ at SWM, VPD incidence was higher than CD. VPD at times appeared in the pith tissue with distinct boundary

101

between symptomatic and asymptomatic tissue. Pith symptoms might be due to natural decay

(Rybaček 1991), however, there was no significant correlation between rhizome size and either

VPD disease rating. Phomopsis tubevora, the causal agent of red crown rot of hop, produces a set similar pith symptoms (Gent et al. 2013). VPD caused by other pathogens cannot be ruled out in this study and may explain in part higher VPD incidence in some cultivars.

The high vigor and low foliar symptom response of ‘Tahoma’ in this study is similar to other cultivars that are classified as tolerant to rootstock rot and basal shoot infection (Woods and Gent 2016). Contrary to expectations, vigor did not differ between ‘Tahoma’ and

‘Centennial’ despite significant difference is wet-rot CD susceptibility. While ‘Columbia’ and

‘Newport’ showed similarly low levels of foliar symptoms, these cultivars were significantly less vigorous than ‘Tahoma’ at PPF; ‘Columbia’ was less vigorous than ‘Tahoma’ at SWM. Despite their low vigor, the wet-rot CD ratings for ‘Columbia’ and ‘Newport’ were like ‘Tahoma’ and suggests that these cultivars are less susceptible to rootstock rot than ‘Centennial’. Differences in vigor due to DM rootstock rot may be difficult to detect in short term studies. Woods and Gent

(2016) worked with cultivars established over a 25-year period and then left for seven year without the protection of fungicide (minimum age of 8 years by the last assessment date) resulting in ample time to observe the effects of rootstock rot on plant vigor. Although, commercial hop yards are typically productive for 10- to 20-years (Beatson et al. 2009) leading to a natural decline in vigor in older cultivars.

Understanding the differences in hop cultivar susceptibility has value for both breeders and growers. Highly desired brewing characteristic may continue to drive the planting of DM susceptible cultivars. The protection of desirable cultivars susceptible to rootstock rot and those that tolerate root rot but still produce foliar symptoms will require the protection of fungicides.

102

The research presented here highlights the need to explain the differences in susceptibility to basal shoot infections of hop cultivars with similar tolerances to rootstock rot to aid future breeding efforts.

103

LITERATURE CITED

104

LITERATURE CITED

Asada, Y., and Matsumoto, I. 1969. Formation of lignin-like substance in the root tissues of Japanese raddish plant infected by downy mildew fungus. Japn. J. Phyto. 35:160-167.

Beatson, R., Kenny, S., Pethybridge, S., and Gent, D. 2009. Hop Production. Pages 5-8 in: Compendium of Hop Diseases and Pests. W. F. Mahaffee, S. J. Pethybridge and D. H. Gent, eds. APS Press, St. Paul, MN.

Bilodeau, G. J., Martin, F. N., Coffey, M. D., and Blomquist, C. L. 2014. Development of a multiplex assay for genus-and species-specific detection of Phytophthora based on differences in mitochondrial gene order. Phytopathology 104:733-748.

Box, G. E., and Cox, D. R. 1964. An analysis of transformations. J. R. Stat. Soc. Series B Stat. Methodol. 26:211-243.

Brien, R. 1938. Black Root-Rot of Hops. New Zealand Department of Scientific and Industrial Research, Plant Disease Division Bulletin no. 11.

Brooks, S., Horner, C., Likens, S., and Zimmermann, C. 1972. Registration of Cascade Hop1 (Reg. No. 1). Crop Sci. 12:394-394.

Canellas, L. P., Olivares, F. L., Canellas, N. O., Mazzei, P., and Piccolo, A. 2019. Humic acids increase the maize seedlings exudation yield. Chem. Biol. Technol. Agric. 6:3.

Coley-Smith, J. R. 1962. Overwintering of hop downy mildew Pseudoperonospora humuli (Miy. and Tak.) Wilson. Ann. Appl. Biol. 50:235-243.

Coley‐Smith, J. R. 1964. Persistence and identification of downy mildew Pseudoperonospora humuli (Miy. and Tak.) Wilson in hop rootstocks. Ann. Appl. Biol. 53:129-132.

Coley-Smith, J. R. 1965. Infection of hop rootstocks by downy mildew Pseudoperonospora humuli (Miy. & Tak.) Wilson and its control by early-season dusts. Ann. Appl. Biol. 56:381-388.

Gao, X., Jackson, T., Lambert, K. N., Li, S., Hartman, G. L., and Niblack, T. 2004. Detection and quantification of Fusarium solani f. sp. glycines in soybean roots with real-time quantitative polymerase chain reaction. Plant Dis. 88:1372-1380.

Gent, D. H., Farnsworth, J. L., and Johnson, D. A. 2012. Spatial analysis and incidence–density relationships for downy mildew on hop. Plant Pathol. 61:37-47.

105

Gent, D. H., Nelson, M. E., Farnsworth, J. L., and Grove, G. G. 2009. PCR detection of Pseudoperonospora humuli in air samples from hop yards. Plant Pathol. 58:1081-1091.

Gent, D. H., Mueller-Warrant, G., Woods, J. L., Putnam, M. L., and Twomey, M. C. 2013. Red crown rot of hop in Oregon caused by Phomopsis tuberivora. Plant Health Prog. 14:52.

George, A. 2020. USA hops 2019 Statistical Report. Hop Growers of America, Online publication: usahops.org.

Haunold, A., Likens, S., Nickerson, G., and Hampton, R. 1984. Registration of Nugget hop. Crop Sci. 24:618-618.

Haunold, A., Likens, S., Horner, C., Zimmermann, C., and Roberts, D. 1976. Registration of Columbia Hop1 (Reg. No. 5). Crop Sci. 16:738-739.

Henning, J., Townsend, S., Mahaffee, W., Kenny, S., and Haunold, A. 2004. Registration of ‘Newport’ hop. Crop Sci. 44:1018-1019.

Henning, J. A., Gent, D., Townsend, M. S., and Haunold, A. 2018. Registration of Downy Mildew–Resistant Male Hop Germplasm USDA 21087M. J. Plant Regist. 12:379-381.

Higgins, D. S., and Hausbeck, M. K. 2017. Fungicides and cultivars can limit hop downy mildew (Pseudoperonospora humuli) in Michigan. Phytopathology. 107(12S):S5-78.

Higgins, D. S., and Hausbeck, M. K. 2018. Effective management strategies for hop downy mildew (Pseudoperonospora humuli) in Michigan. Phytopathology. 108(12S):S2.28.

Horsfall, J. G., and Barratt, R. W. 1945. An improved grading system for measuring plant diseases. Phytopathology 35:655.

Johnson, D., and Anliker, W. 1985. Effect of downy mildew epidemics on the seasonal carryover of initial inoculum in hop yards. Plant Dis. 69:140-142.

Johnson, D. A., and Skotland, C. 1985. Effects of temperature and relative humidity on sporangium production of Pseudoperonospora humuli on hop. Phytopathology 75:127- 129.

Kenny, S., and Zimmermann, C. 1991. Registration of ‘Centennial’ Hop. Crop Sci. 31:1092- 1092.

106

Mahaffee, W. F., Pethybridge, S. J., and Gent, D. H. 2009. Compendium of hop diseases and pests. American Phytopathological Society (APS Press), St. Paul, MN.

Miles, T. D., Martin, F. N., Robideau, G. P., Bilodeau, G. J., and Coffey, M. D. 2017. Systematic development of Phytophthora species-specific mitochondrial diagnostic markers for economically important members of the genus. Plant Dis. 101:1162-1170.

Mochizuki, S., Fukumoto, T., Ohara, T., Ohtani, K., Yoshihara, A., Shigematsu, Y., Tanaka, K., Ebihara, K., Tajima, S., and Gomi, K. 2020. The rare sugar d-tagatose protects plants from downy mildews and is a safe fungicidal agrochemical. Commun. Biol. 3:1-15.

Mouzeyar, S., De Labrouhe, D. T., and Vear, F. 1993. Histopathological studies of resistance of sunflower (Helianthus annuus L.) to downy mildew (Plasmopara halstedii). J. Phytopathol. 139:289-297.

Mouzeyar, S., De Labrouhe, D. T., and Vear, F. 1994. Effect of host‐race combination on resistance of sunflower, Helianthus annuus L., to downy mildew Plasmopara balstedii. J. Phytopathol. 141:249-258.

Neve, R. A. 1991. Hops. Springer, Berlin, Germany.

O’Neal, S. 2015. Pest Management Strategic Plan for U.S. Hops: Summary of a workshop held on November 4, 2014 in Yakima, WA. Online.

Pagán, I., and García-Arenal, F. 2020. Tolerance of plants to pathogens: A unifying view. Annu. Rev. Phytopathol. 58:77-96.

Patzak, J. 2005. PCR detection of Pseudoperonospora humuli and Podosphaera macularis (Humulus lupulus). Plant Prot. Sci. 41:141.

Radwan, O., Bouzidi, M. F., and Mouzeyar, S. 2011. Molecular characterization of two types of resistance in sunflower to Plasmopara halstedii, the causal agent of downy mildew. Phytopathology 101:970-979.

Royle, D. 1968. Hop Black Root‐rot (Phytophthra citricola Sawada) in England. Plant Pathol. 17:36-40.

Royle, D. 1970. Infection periods in relation to the natural development of hop downy mildew (Pseudoperonospora humuli). Ann. Appl. Biol. 66:281-291.

107

Royle, D., and Thomas, G. 1971a. The influence of stomatal opening on the infection of hop leaves by Pseudoperonospora humuli. Physiol. Plant Pathol. 1:329-343.

Royle, D., and Thomas, G. 1973. Factors affecting zoospore responses towards stomata in hop downy mildew (Pseudoperonospora humuli) including some comparisons with grapevine downy mildew (Plasmopara viticola). Physiol. Plant Pathol. 3:405-417.

Royle, D., and Kremheller, H. T. 1981. Downy Mildew of the Hop. in: The Downy Mildews. D. M. Spencer, ed. Academic Press Inc., New York, NY.

Royle, D. J., and Thomas, G. G. 1971b. Observations with the scanning electron microscope on the early stages of hop leaf infection by Pseudoperonospora humuli. Physiol. Plant Pathol. 1:345-349.

Rybaček, V. 1991. Hop Production. Elsevier, New York, NY.

Skotland, C. 1961. Infection of hop crowns and roots by Pseudoperonospora humuli and its relation to crown and root rot and overwinting of the pathogen. Phytopathology 51:241- 244.

Sonoda, R., and Ogawa, J. 1968. Phytophthora crown and root rot of hops in California. Plant Dis. Rep. 52:76-78.

Sonoda, R., and Ogawa, J. 1972. Ecological factors limiting epidemics of hop downy mildew in arid climates. Hilgardia 41:457-473.

Talboys, P. 1958. Association of tylosis and hyperplasia of the xylem with vascular invasion of the hop by Verticillium albo-atrum. Trans. Br. Mycol. Soc. 41:249-260.

Ware, W. 1926. Pseudoperonospora humuli and its mycelial invasion of the host plant. Trans. Br. Mycol. Soc. 11:91-107.

Williams, I. H., Roberts, J. B., and Coley-Smith, J. R. 1961. Studies of the Dormant Phase of the Hop (Humulus lupulus L.). Rep. Dept. Hop Res. Wye Coll. for 1960:48-58.

Woods, J. L., and Gent, D. H. 2016. Susceptibility of hop cultivars to downy mildew: associations with chemical characteristics and region of origin. Plant Health Prog. 17:42- 48.

108

Zimmermann, C., Likens, S., Haunold, A., Horner, C., and Roberts, D. 1975. Registration of Comet Hop1 (Reg. No. 3). Crop Sci. 15:98-98.

109

CHAPTER 4. PSEUDOPERONOSPORA HUMULI DETECTION IN ASYMPTOMATIC HOP TISSUE USING REAL-TIME MITOCHONDRIAL AND NUCLEAR MARKERS AND A RECOMBINASE POLYMERASE AMPLIFICATION (RPA) ASSAY

Abstract

In the nursery and greenhouse, hop stock plants and plantlets are exposed to environmental conditions conducive to downy mildew (Pseudoperonospora humuli) including misting during propagation, frequent overhead watering, and high relative humidity.

Management strategies include fungicide application and rogueing symptomatic plants.

However, asymptomatic plants may be infected as the incubation time may be 10 (leaves) or 21 days (shoots). Real-time quantitative polymerase chain reaction (qPCR) TaqMan assays and a recombinase polymerase amplification (RPA) assay were developed and tested for pathogen detection in asymptomatic leaf and stem tissue. The qPCR TaqMan assay based on a mitochondrial marker (open reading frame 306) was more sensitive than a nuclear (c125015.3e1) marker-based assay at detecting raw P. humuli DNA and asymptomatic infections in leaves and shoot. In DNA serial dilutions of P. humuli, linear correlations for nuclear (R2 = 0.998) and mitochondrial (R2 = 0.995) assays had limits of detection at 1000 and 100 fg, respectively.

Replicated leaf disks removed at inoculation (T0), incubation (2- and 3-days post inoculation

(dpi)), and symptomatic (3- and 7- dpi) time points were tested for pathogen detection; the experiment was repeated. The number of quantification cycles (Cq) were significantly (P =

0.001) higher (approximately 4 cycles) in the nuclear than mitochondrial assay at each time point

≥ 1 dpi; no detection in T0 or non-inoculated control leaves. In asymptomatic shoots (n=10) collected from symptomatic plants, P. humuli was detected in 100% and 40% of shoots for mitochondrial (31.4 Cq ±1.07 s.d.) and nuclear (36.0 Cq ±0.46 s.d.) based assays, respectively.

In the RPA assay, a first derivative analysis was used to call positive amplification and may help

110

reduce the number of false positive detections. The RPA assay was negatively affected by a crude DNA extraction resulting in a 10-fold loss of sensitivity. Molecular diagnostics applied to improve downy mildew management in hop nursery production should utilize the mitochondrial marker for optimal qPCR assay performance.

Introduction

In 2019, 291.4 ha were harvested in Michigan, up from 129.5 ha in 2015 (George 2020).

The establishment of a new hop yard in Michigan costs approximately $34,595/ha (Sirrine et al.

2014). To protect their investment, producers require high quality plantlets free of disease. The

National Clean Plant Network (NCPN) provides foundation (Generation 1)-level stock plants, unrooted green cuttings tested and maintained free of virus and the hop stunt viroid, to most propagators. However, the NCPN currently does not test for filamentous pathogens (NCPN

2020) and few disease certification programs and specific disease management recomendations exist for Generation-2+ propagation nurseries.

Michigan producers source their planting material from local nurseries with greenhouses, established growers, or nurseries and greenhouses in other regions. Most plantlets from nurseries and greenhouses are propagated from softwood cuttings (Howard 1965, 1967; Howard and

Sykes 1966), but they can also be generated through bedded sets (Neve 1991; Rybaček 1991) and micropropagation (Gurriarán et al. 1999; Peredo et al. 2009). Plantlets are one- to two-node sections of softwood (approximately 5-8 cm long) cut from stock plants are rooted under a mist system (Howard 1965, 1967; Howard and Sykes 1966) and later moved to the greenhouse floor or to an outdoor nursery. Bedded sets are rhizomes or stolons cut from mature hop plants in the field during the winter, planted into containers, and raised in greenhouses and outdoor nurseries

111

(Neve 1991; Rybaček 1991). Propagated plants and bedded sets are regularly pruned to limit growth and may remain in outdoor nurseries or greenhouses for up to year.

Hop downy mildew (HDM) caused by Pseudoperonospora humuli (Miyabe &

Takahashi) Wilson is a highly destructive disease; the pathogen may be disseminated via planting material (Mahaffee et al. 2009; O’Neal 2015). Chronic systemic infections deplete carbohydrate reserves needed for vigorous growth (Williams et al. 1961) and susceptible cultivars may succumb to crown rot (Coley-Smith 1962; Skotland 1961; Woods and Gent

2016). Infected apical buds fail to climb and it may take additional labor to retrain replacement bines. Secondary infections of inflorescences cause cone damage resulting in direct losses or reduced quality/ low brewing value (Royle and Kremheller 1981).

During propagation, prolonged leaf wetness and high relative humidity from misting and overhead irrigation creates conditions conducive to HDM (Coley-Smith 1963). Sporangia develop under high relative humidity (Johnson and Skotland 1985) and are moved by wind and splashing water (Royle 1970; Sonoda and Ogawa 1972). Water triggers indirect infection via swimming zoosporangia that penetrate stomata openings (Royle 1970; Royle and Thomas

1971b, a, 1973), infecting shoots and the plant crown (Coley-Smith 1962; Coley-Smith 1965).

The time from infection to chlorotic foliar lesions is 3-10 days at 7-28° C (Royle 1970). Shoot infections resulting in swollen shoots and chlorotic downward cupped leaves, can take 7-22 days at 9-20 ° C (Royle 1970). Propagators apply fungicides and remove symptomatic diseased plants to limit HDM. However, newly infected plants may be asymptomatic.

DNA-based diagnostic assays for P. humuli detection (Gent et al. 2009; Patzak 2005;

Summers et al. 2015) may be useful for propagators to detect asymptomatic infections.

Polymerase chain reaction (PCR) based diagnostic assays for P. humuli were developed using

112

the internal transcribed spacer (ITSrDNA) region (Gent et al. 2009; Patzak 2005). Yet, primers designed to amplify the ITSrDNA region appear to lack the specificity needed to distinguish P. humuli from Pseudoperonospora cubensis the causal agent of cucurbit downy mildew (Gent et al. 2009). Outbreaks of cucurbit downy mildew occur annually in Michigan (Granke and

Hausbeck 2011; Granke et al. 2014; Naegele et al. 2016) and ITSrDNA based assays may result in false positives (Gent et al. 2009). The mitochondrial cytochrome c oxidase 2 locus (cox2)

(Mitchell et al. 2011) was developed into a quantitative real-time PCR (qPCR) assay (Summers et al. 2015) to distinguish P. humuli and P. cubensis, but the locked nucleic acid (LNA) probe used relies on the detection of a single nucleotide polymorphism (SNP). Rahman et al. (2019) identified four highly polymorphic nuclear regions (c125015.3e1, c127446.1e1, c127233.5e3, and c126365.1e5) that appear to be unique to P. humuli that might be utilized for a diagnostic assay. High copy number mitochondria markers are likely more sensitive than single copy nuclear marker (Rahman et al. 2017; Rahman et al. 2020), but a side-by-side comparison has yet to be reported.

qPCR is the current gold standard for plant pathogen diagnostics but requires specialized equipment (thermal cycler) and time (numerous cycles to amplify target sequences and typically a DNA purification step to avoid reaction inhibitors in environmental samples) to process samples (Dorak 2007). Recombinase polymerase amplification (RPA) is an isothermal method that utilizes a recombinase enzyme (T4 UvsX) and a single-stranded binding protein (T4 gp32) to invade double-stranded DNA and eliminate the need for thermal cycling to denature it (Li et al. 2019; Piepenburg et al. 2006). Real-time detection of DNA targets is carried out with a portable fluorometer to detect a probe containing a tetrahydrofuran residue (THF), flanked by a fluorophore and quencher group (Li et al. 2019; Piepenburg et al. 2006). A DNA repair enzyme

113

(Exonuclease III) cleaves the THF residue when bound to its target sequence to generate the fluorescent signal (Li et al. 2019; Piepenburg et al. 2006). Lyophilized RPA kits requiring user specified primers and probes with pre-mixed enzymes and reagents are available from two manufactures (Agdia, Elkhart, IN; TwistDx, Maidenhead, UK). RPA assays using various portable fluorometers have been created to detect specific pathogens in plant tissue and soil

(Ammour et al. 2017; Burkhardt et al. 2019; Burkhardt et al. 2018; McCoy et al. 2020; Miles et al. 2015; Rojas et al. 2017). The elimination of DNA purification and the field applicability of the RPA assay would be useful to plant diagnosticians and hop propagators.

Our objective was to develop and compare qPCR TaqMan assays using species-specific mitochondrial- and nuclear-markers and test their performance in asymptomatic hop leaves and shoots. The mitochondrial based TaqMan assay was then used to develop an RPA assay and validated in asymptomatic leaves and shoots.

Materials and Methods

qPCR DNA extraction. DNA was extracted from symptomatic and asymptomatic leaf and stem tissue using the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) with modifications.

Leaf (5 to 10 mg) and stem (70 mg) tissue were physically lysed in Lysing Matrix A (MP

Biomedicals North America, Solon, OH), buffer AP1 (600 µl), and RNase A (10 mg/ml; 600 µl); stem tissue was placed between two 0.635 ceramic spheres (provided with the lysing matrix).

Tubes were shaken for 60 s (40 M/S) in a Fast Prep tissue disruptor (MP Biomedical North

America). P. humuli DNA was extracted from sporangia as described in Higgins et al. (2020).

Briefly, sporangia were mechanically homogenized (TissueLyser II, Qiagen, Hilden, Germany) for 3 min 30 sec at 30 Hz, treated with CTAB extraction buffer (Sambrook et al. 1989), and

114

subjected to a hot water bath and sonication. DNA was extracted with equal volume phenol:chloroform (ChCl3):isoamyl alcohol (IAA) (25:24:1), phenol:chloroform (ChCl3) and precipitated overnight in isopropanol at - 20°C. Pellets were washed in 70% ethanol and resuspended in 50 µl of TE buffer. All DNA was quantified on a Qbit 4 fluorometer using the

Qubit dsDNA high-sensitivity assay (ThermoFisher Scientific, Waltham, MA).

qPCR marker, probes, and primers. The mitochondrial open reading frame 306

(orf306) was identified as a potential region for marker development by aligning the mitochondrial genomes of P. humuli (NCBI accession: NC_042478.1; Rahman et al 2019) and

P. cubensis (NCBI accession: NC_027859; (Lu et al. 2016) using the Mauve whole genome aligner (Darling et al. 2010) in Geneious version 11.0.4 (Kearse et al. 2012). Sequence alignments of orf306 were examined for regions meeting basic requirements for qPCR and RPA probe development as outlined in Miles et al. (2015). Primer pairs (Ph306F [length = 26 bp;

GC% = 19.2; Tm = 52.8] and Ph306R [length = 22 bp; GC% = 31.8; Tm= 51.7]; product length

371bp) were designed with Geneious version 11.0.4 (Kearse et al. 2012) (Table 3.1). The probe

(Phorf306q [length = 35 bp]) was placed in a location close to the forward primer and labeled with 5’ fluorescein fluorescent dye (FAM) and 3’ Black Hole Quencher – 1 (BHQ-1).

Three nuclear genomic loci, c125015.3e1 (contig 125015 [marker-contig_125015- augustus-gene-0.3-mRNA-1:1]), c127233.5e3 (contig 127233 [marker-contig_127233-snap- gene-0.5-mRNA-1:3]), and c126365.1e5 (contig 126365 [contig_126365-snap-gene-0.1-mRNA-

1:5]) proposed by Rahman et al. (2019) for diagnostic markers were considered for the qPCR

TaqMan assay. Locus c127446.1e1 was omitted because of previously reported cross- amplification with P. cubensis (Rahman et al. 2019). The candidate diagnostic markers were submitted to a GenBank BLAST search (https://blast.ncbi.nlm.nih.gov) of the P. humuli genome

115

(GCA_003991265.1 [WGS Project: NQF001; WGS NQF0010000001 -NQF0010055334]) with default parameters. Search results for c125015.3e1 and c127233.5e3 returned a single target site and c126365.1e5 had multiple hits with query coverage as high as 64%. The candidate diagnostic markers were submitted to a Primer-BLAST search (Ye et al. 2012). Product size was set to 200

– 300 bp and the database option was changed to custom; a fasta file of WGS NQF0010000001 -

NQF0010055334 was uploaded; the remaining parameters were left as default. No suitable primer pairs could be identified for candidate marker c127233.5e3. For c125015.3e1, a primer pair (Ph459F [length = 20; Tm = 60.11; GC = 55%] and Ph678R [length = 20; Tm = 59.90; GC

= 55%]; product length = 220) was identified with a single target site on contig 125015 and no off-target binding sites were found (Table 3.1). For c126365.1e5, the primer-pair (Ph534F and

Ph802R [length = 20; Tm = 60.4; GC = 55%]; product length = 269) had a potential off-target binding sites with 75-80% sequence homology but was selected as the best possible option. The probes c125015.3e1 (Ph3e1n [length = 25 bp; Tm = 67.8; %GC = 56) and c126365.1e5 (Ph1e5n

[length = 27 bp; Tm = 68.0; %GC = 51.9) were placed in a location close to the forward primer and labeled with 5’ fluorescein fluorescent dye (FAM) and 3’ Black Hole Quencher – 1 (BHQ-

1).

Internal plant control (IPC) primers (FMP12b and FMP13B) and IPC probe were designed by Bilodeau et al. (2014) and are based on the cox1 gene. The IPC probe was labeled with 5’ CalFluor Red 610 dye and 3’ Black Hole Quencher-2 (BHQ-2). All TaqMan qPCR probes and primers (Table 3.1) were synthesized by Biosearch Technologies, Inc. (Novato, CA) and Integrated DNA Technologies, Inc. (Coralville, IA), respectively.

116

Table 3.1. Primers and probes used for real-time quantitative polymerase chain reactions (qPCR) TaqMan assays to detect Pseudoperonospora humuli. Primer/probesy Sequence (5’-3’) Targetz Source P. humuli - Mitochondrial Ph306F TTTTTATCAAAATGAATACCAAGATT orf306 This study Ph306R AGTATAACTGTTGGTAGTGTAT orf306 This study Phorf306q [FAM]TAAAATACAGAATAAGAAATTTGATCTTCAAAAAG[BHQ1] orf306 This study P. humuli - Nuclear Ph459F TGCTGACACAGATGCTAGGC C125015.3e1 This study Ph678R GCGCTTCAAACTGCTCTTGT C125015.3e1 This study Ph3e1n [FAM]ATCGAGTCGACCGACCACGTTTCGA[BHQ1] C125015.3e1 This study Ph534F TTCGGCATCCTTTCCAGTCC C126365.1e5 This study Ph802R ATATACAGCGCATCGCGTCA C126365.1e5 This study Ph1e5n [FAM]TGCATGCTCTAGGTGCTTCGTATCGCT[BHQ1] C126365.1e5 This study Internal Plant Control FMPl2b GCGTGGACCTGGAATGACTA coxI Bilodeau et al. 2014 FMPl3b AGGTTGTATTAAAGTTTCGATCG coxI Bilodeau et al. 2014 IPC [CALFluorRed610]CTTTTATTATCACTTCCGGTACTGGCAGG[BHQ2] coxI Bilodeau et al. 2014 y Mitochondrial = target locus located in the mitochondrial genome. Nuclear = target locus located in the nuclear genome. Nuclear and mitochondrial genomes obtained from Rahman et al. (2019). z The mitochondrial open reading frame 306 (orf306) was identified from the mitochondrial genomes of P. humuli. Nuclear primers and probes were designed for candidate diagnostic markers C125015.3e1 and C126365.1e5 proposed by Rahman et al. (2019).

117

qPCR TaqMan assay reaction. Multiplex qPCR reactions for orf306 were of 7.5 µl

PerfeCTa qPCR ToughMix 2x (Quantabio, Beverly, MA), 0.75 µl TqFPhum (10 µM), 0.75 µl

TqRFPhum (10 µM), 0.375 µl TqProbePhum (10 µM), 0.3 µl FMPl2b (1 µM), 0.3 µl FMPl3b (1

µM), 0.3 µl IPC probe (1 µM), 1.2 µl Mg (2 mM), DNase-free water, and 1 to 1.5 µl template

DNA in a 15 µl reaction. qPCR amplification was carried out on a CFX96 Touch Real-Time

PCR Detection system (Bio-Rad Laboratories, Inc., Hercules, CA) for an initial 2 minutes at

95°C, then repeat cycles of 95°C for 15 sec and 58.0°C for 1:30 min for a total of 40 cycles.

Quantitative PCR reactions for c126365.1e5 and c125015.3e1 were of 7.5 µl PerfeCTa qPCR ToughMix 2x (Quantabio), 0.75 µl forward primer (10 µM), 0.75 µl reverse primer (10

µM), 0.375 (10 µl) probe (10 µM), 0.3 µl FMPl2b (1 µM), 0.3 µl FMPl3b (1 µM), 0.3 µl IPC probe (1 µM), 1.2 µl Mg (2 mM), DNase-free water, and 1 to 1.5 µl template DNA in a 15 µl reaction. qPCR amplification was carried out on a CFX96 Touch Real-Time PCR Detection system (Bio-Rad Laboratories) for an initial 2 minutes at 95°C, then repeat cycles of 95°C for 15 sec and 57.9°C for 0:30 sec for a total of 50 cycles. An extension time of 1:30 sec did not improve amplification during optimization tests (Table S3.7).

qPCR sensitivity. The limit of detection was determined using serial dilutions of P. humuli DNA. DNA for P. humuli was extracted from sporangia with methods described above.

P. humuli was cultured on leaves according to methods described by Gent et al. (2008) with slight modification. In addition to the dampened filter paper, a cotton ball soaked in deionized water was added and petri dishes were sealed with parafilm to maintain high relative humidity and prevent drying during incubation in the growth chamber. Sporangia were harvested after 5- days of incubation and maintained on ice for approx. 15- 20 min until DNA extraction. A ten-

118

fold serial dilution of P. humuli DNA ranging from 1 ng/µl to 10 fg/µl was prepared in AE buffer (Qiagen) and quantified on a Qbit 4 fluorometer using the Qubit dsDNA high-sensitivity assay (ThermoFisher Scientific). All sensitivity tests were conducted with a single serial dilution set on the same day. To test for PCR inhibitors in plant tissue, 20 ng of DNA extracted from a hop leaf was spiked into P. humuli DNA at each dilution. A positive control for the internal plant probe (hop DNA (20 ng)) and a negative non-template control (AE buffer (Qiagen)) were included. Three technical replicates were tested and used to derive the limit of detection (LOD) that is defined as the lowest concentration of DNA at which > 95% of the reactions amplified

(Bustin et al. 2009). Standard curves were constructed with the CFX Maestro software (Bio-Rad

Laboratories). A fluorescence threshold was automatically determined according to the default software settings and fluorescence drift correction was applied as needed to the baseline settings.

Efficiency was calculated as: E = 10(–1/slope) – 1.

qPCR specificity. The specificity of the TaqMan qPCR assays was tested on P. humuli

(n = 25), P. cubensis (n = 9), other fungi and oomycetes (n = 8), H. lupulus, and a no-template control consisting of nuclease-free water (Integrated DNA Technologies, Inc.). Sporangia isolates of P. humuli were collected from primary and secondary symptomatic shoots under aseptic conditions during a 2018 survey of hop yards in Michigan’s lower peninsula (Higgins et al. 2020). The sporangial suspensions were quantified and stored at -80°C until DNA extraction

(Higgins et al. 2020). DNA or live cultures for the remaining fungi and oomycetes was provided by Dr. M Hausbeck at Michigan State University (isolate notation refers to the culture or DNA collection identification assigned by the laboratory). DNA was extracted from cultured organisms using the DNeasy Plant Mini Kit (Qiagen) and methods described above. DNA concentrations of all isolates used for specificity tests were standardized to 5 ng/µl on a Qbit 4

119

fluorometer using the Qubit dsDNA high-sensitivity assay (ThermoFisher Scientific); a single

DNA dilution set was used to test both the orf306 and c125015.3e1 assays. Two technical replicates per isolate were tested. A negative non-template control (nuclease-free water

(Integrated DNA Technologies, Inc.)) was included (three technical replicates).

RPA crude DNA extraction. A working solution of GEB2 buffer (Agdia, Elkhart, IN) was prepared according to the manufacture’s recommendations (55.8 mg/ml; stir for 30 min).

Leaf (5 to 10 mg) and stem (70 mg) tissue was added to a mesh sample bag (Agdia) and macerated with a tissue homogenizer tool (Agdia). A working solution of GEB2 buffer was added to the macerated tissue sample (1 ml/100 mg tissue) (Miles et al. 2015).

RPA markers, probe, and primers. Primers (PhR7 and PhF8) and probe (Phorf306rpa)

(Table 3.2) were designed for the orf306 region according to the guidelines in the AmplifyRP

XRT Discovery Kit (Agdia) and Miles et al. (2015). Seven reverse primers were screened against a single forward primer (PhF1) (Table S3.8); pairing with the reverse primer PhR7 produced the earliest and steepest sigmoidal amplification curve. The PhR7 primer was screened against seven forward primers (Table S3.8). The three primer sets were screened against a subset of 10 ng

DNA samples of P. humuli and P. cubensis. Primers PhR7 (length = 35; Tm = 40.9; GC =

31.4%) and PhF8 (length = 35; Tm = 60.1; GC = 25.7%) produced the steepest amplification curve and earliest onset of amplification (product length = 267) with no cross reactions to P. cubensis DNA and were used for the remaining experiments. The Phorf306rpa probe (length =

52 bp) was placed near the forward primer and labeled with 5’ fluorescein fluorescent dye

(FAM) and 3’ Black Hole Quencher – 1 (BHQ-1).

120

Table 3.2. Recombinase polymerase amplification (RPA) primers and probes developed to detect Pseudoperonospora humuli. Primer/probes Sequence (5’-3’) Target Source P. humuli PhF8 ATTTTATTCCAAAAATAATGCGTATCCATATCTAG orf306 This study PhR7 CTTGATGTTTATTACAAGTATAACTGTTGGTAGTG orf306 This study Phorf306rpa CCGCATTATTCTAGTTTTAAAATACAGAATAAGAAAT[T- orf306 This study FAM]T[THF]A[T-BHQ1]CTTCAAAAAG[Terminal C3-spacer] Internal Plant Control Cox1-IPC-F CATGCGTGGACCTGGAATGACTATGCATAGA cox1 Miles et al. (2015) Cox1-IPC-R GGTTGTATTAAAGTTTCGATCGGTTAATAACA cox1 Miles et al. (2015) Cox1-IPC GGTCCGTTCTAGTGACAGCATTCCYACTTTTATTA[ROX- cox1 Miles et al. (2015) dT]C[THF]C[BHQ2-dT]YCCGGTACTGGC[3'-C3SPACER] z Mitochondrial = target locus located in the mitochondrial genome. Nuclear = target locus located in the nuclear genome. Nuclear and mitochondrial genomes obtained from (Rahman et al. 2019). Nuclear primers and probes were designed for candidate diagnostic markers proposed by Rahman et al. (2019).

121

Probe (Cox1-IPC-P) and primers (Cox1-IPC-F and Cox1-IPC-R) for the internal plant control (Table 3.2) were designed by Miles et al. (2015). The Cox1-IPC-P probe was labeled with 5’ rhodamine fluorescent dye (ROX) and 3’ Black Hole Quencher – 1 (BHQ-1), placed near the forward primer. All RPA probes and primers were synthesized by Biosearch Technologies,

Inc. and Integrated DNA Technologies, Inc., respectively.

RPA assay reaction Amplification was performed with the AmplifyRP XRT Discovery

Kit (Agdia) and the primers/probe set described above. Using a total reaction volume of 24 µl, first a master mix was prepared with 14.75 µl rehydration buffer, 1.36 µl primer PhF8 (10 µM),

1.36 µl primer PhR7 (10 µM), 0.36 µl Phorf306rpa (10 µM), 0.94 µl primer Cox1F (10 µM),

0.94 µl Cox1R (10 µM), 0.28 µl Cox1-IPC-P (10 µM), and DNase-free water. 21.75 µl of master mix was added to 1 µl template in an 8-strip microcentrifuge tube (Bio-Rad Laboratories) and mixed. Master mix with template was transferred to a reaction tube (provided by manufacture) containing a lyophilized pellet. 1.25 µl 280 mM magnesium acetate was added to the cap of each reaction tube. The caps were closed, and tubes centrifuged briefly to uniformly add the magnesium acetate and “hot start” the reactions. The tubes were vortex briefly and recentrifuged for 20 seconds or until air bubbles were removed. Reaction tubes were placed immediately into the isothermal fluorometer (Axxin T16-ISO, Fairfield, Australia) and heated at 39 C for 25 min

(LED % = 12% FAM, 7% HEX, 60% ROX). At four minutes the reaction tubes were removed, vortexed briefly, centrifuged for 20 seconds, and placed back in fluorometer for the remaining time.

RPA positives and onset of amplification. Positive amplification was determined using a baseline threshold for the slope of raw data (Miles et al. 2015) and the fluorescence peak of the first derivative analysis (El Wahed et al. 2015; Luu-The et al. 2005; Rasmussen 2001). Data

122

were exported as an Excel file and the slope was calculated between three time points to determine the change in fluorescence signal from 320 to 1200 seconds. The max slope of non- template controls from three replicate runs (two technical replicate per run) was 1.77 mV/sec and a threshold of 2 mV/sec for a duration of 60 sec was set in the GRADIENT MODULE option in the algorithm editor of the Axxin AX0ISO desktop application; a threshold of 1 mV/sec for 60 sec was used for the internal plant control (Figure A1). The first derivative was determined using the 1st DERIVATIVE VIEW option in the Axxin AX0ISO desktop application. Reaction mixtures with and without primers were also tested on NTC, crude hop leaf extract, 10 ng P. humuli DNA, and 10 ng P. humuli + crude hop leaf extract in four replicate runs.

The onset of amplification was determined by time that the slope threshold is exceeded

(Miles et al. 2015) and the peak of the second derivative analysis (a method modified from real- time reverse transcription PCR (RT-PCR) (Luu-The et al. 2005; Meuer et al. 2012; Rasmussen

2001). The second derivative of captured data measurements was determined using the 2nd

DERIVATIVE VIEW option in the Axxin AX0ISO desktop application.

Sensitivity for RPA assay. The limit of detection for the RPA assay was determined using serial dilutions of P. humuli DNA. DNA for P. humuli was extracted from sporangia with methods described above. A ten-fold serial dilution of P. humuli DNA ranging from 1 ng/µl to

10 fg/µl was prepared in nuclease-free water (Integrated DNA Technologies, Inc.) and quantified on a Qbit 4 fluorometer using the Qubit dsDNA high-sensitivity assay (ThermoFisher Scientific).

To test for PCR inhibitors in plant tissue crude DNA extract, obtained from hop leaf as described above, was spiked into P. humuli DNA at each dilution. All sensitivity tests were conducted using the same serial dilution set. A max of 16 reaction can be analyzed on Axxin T16-ISO isothermal fluorometer and multiple runs were performed to complete the sensitivity tests. Each

123

run consisted of four non-template controls (two with and without hop DNA) and 12 reaction containing two dilution concentration with and with hop DNA (three technical replicates per concentration). Standard curves were constructed by plotting the log of the onset of amplification.

Specificity for RPA assay. The specificity of the RPA assay was tested on purified DNA of P. humuli and non-target isolates as described above (Table 3.3). DNA from all non-target isolates was combined and amplified as two bulk samples. Two technical replicates per isolate were tested. Two non-template controls (without hop DNA) were included in each run.

Symptomatic and asymptomatic leaves. A leaf disk assay to detect asymptomatic infections was designed based on methods described by Ammour et al. (2017) with modifications. The experiment was arranged in as a complete randomized design (CRD) with five replicate leaves per time treatment. Healthy hop ‘Nugget’ plantlets propagated from softwood cuttings were obtained from a local propagator. Leaves were collected from nodes three to five and immediately refrigerated for 30 min. Leaves were placed abaxial surface up into a petri dish (100 x 15 mm) with a wet Whatman filter paper (manufacturer) and a moistened cotton ball. Two 20 µl-droplets of P. humuli sporangia suspension (1x104 sporangia per ml), were obtained using methods described above, were placed onto the abaxial surfaces of leaves.

Three negative control leaves per treatment level received nuclease-free water (Integrated DNA

Technologies, Inc.). Petri dishes were sealed with parafilm and incubated in a growth chamber at

20°C with 14/10 h (light/dark) photoperiod and 95 mE of light intensity. Two leaf disks

(approximately 5-10 mg/disk) per leaf were removed at the site of inoculation with a sterile cork borer on the day of inoculation (T0) and one, two, three, and seven days post inoculation (dpi).

Leaf disks were rinsed in deionized water and placed into sterile 1.75 ml tubes. Samples were

124

frozen at -20°C for later DNA extraction. One disk was used for the qPCR assays and the other for the RPA assay. The experiment was repeated. A lower inoculum concentration (1x103 sporangia per µl) was also tested in two separate experiments but was excluded from molecular analysis due inconsistency of symptoms.

DNA extracted from a single leaf punch was analyzed with qPCR TaqMan orf306 and c125015.3e1 assays; DNA obtained from the GEB2 extraction buffer was analyzed using RPA orf306 assay. Control reactions were included for all assays with P. humuli DNA from a single dilution set. For the qPCR TaqMan assays, controls at the LOD for the orf306 qPCR (100 fg) and c125015.3e1 (1000 fg) assays were used to determine the upper threshold for Cq values for a positive reaction (three technical replicates); P. humuli DNA at 1 ng was used as a positive control (three technical replicates). For the RPA assay, no technical replicates were used.

Positive amplification was determined with the slope threshold and the first derivative analysis.

Two negative non-template controls and a positive P. humuli control were included in each run

For qPCR assays, statistical analysis was performed using SAS software, Version 9.4

(SAS Institute, Inc., Cary, NC). A three-way global analysis of variance (ANOVA) with interaction F-test was calculated using PROC MIXED to determine the fixed effects of marker

(nuclear vs. mitochondrial assay), dpi (1, 2, 3, and 7), run (experiment 1 vs. experiment 2), and the two- and three-way interactions. All data met the assumptions of normality (checked using residual plots) and homogeneity of variances (Levene’s test (P < 0.05)). Fisher’s protected least significant differences (LSD) at P = 0.001 were used to determine significant pair-wise comparisons among treatment means.

125

Table 3.3. Specificity of hop downy mildew quantitative real-time PCR (qPCR) assays containing a mitochondrial (orf306) or a nuclear (c125015.3e1) markers and positive recombinase polymerase amplification (orf306) based on a slope threshold and the first derivative analysis. Michigan Concentrationx qPCR (Cq)y RPAz Organism Isolatev Host Cultivarw State County Spor. DNA orf306 c125015.3e1 Slope 1st Der. Pseudoperonospora HYK2 Humulus Nugget MI Montcalm 3.2E+05 33.5 15.6± 0.1 25.2± 0.0 + + humuli lupulus P. humuli HYK3 H. lupulus Nugget MI Montcalm 3.4E+05 24.5 17.5± 0.3 24.1± 0.1 + + P. humuli EF2 H. lupulus Nugget MI Ingham 3.8E+05 35.0 19.5± 0.0 25.8± 0.0 + + P. humuli EF3 H. lupulus Nugget MI Ingham 6.9E+05 46.5 15.7± 0.1 26.5± 0.1 + + P. humuli King3 H. lupulus Chinook MI Grand Traverse 1.8E+05 10.8 20.6± 0.1 29.9± 0.1 + + P. humuli King2Q H. lupulus Chinook MI Grand Traverse 2.5E+05 7.1 25.8± 0.2 25.4± 0.2 + + P. humuli JB2 H. lupulus Cascade MI Grand Traverse 5.3E+05 38.0 16.3± 0.3 24.2± 0.0 +/ n.d. + P. humuli JB3 H. lupulus Cascade MI Grand Traverse 5.2E+05 39.8 15.1± 1.3 24.5± 0.0 + + P. humuli LA2 H. lupulus Chinook MI Benzie 1.3E+06 23.1 14.3± 1.4 23.7± 0.1 + + P. humuli LA3 H. lupulus Chinook MI Benzie 6.2E+05 17.6 14.6± 0.0 24.0± 0.1 + + P. humuli SB1 H. lupulus Centennial MI Leelanau 6.8E+05 22.6 12.9± 2.0 24.1± 0.0 + + P. humuli SB3 H. lupulus Centennial MI Leelanau 4.3E+05 20.9 15.1± 2.7 22.6± 0.0 + + P. humuli O2C H. lupulus Columbia MI Leelanau 6.9E+05 20.7 14.3± 0.2 23.6± 0.1 + + P. humuli O3C H. lupulus Columbia MI Leelanau 5.9E+05 20.3 12.4± 3.0 24.7± 0.2 + + P. humuli O2U H. lupulus Confidential MI Leelanau 6.8E+05 21.9 15.4± 0.2 24.7± 0.2 + + P. humuli O3U H. lupulus Confidential MI Leelanau 4.7E+05 20.1 14.8± 0.5 27.5± 0.0 + + P. humuli TopH2 H. lupulus Cashmere MI Genesee 1.9E+05 12.2 14.1± 0.2 25.9± 0.0 + + P. humuli TopH2 H. lupulus Centennial MI Berrien 7.0E+05 26.5 23.1± 0.5 27.8± 0.1 + + P. humuli TopH1 H. lupulus Centennial MI Berrien 5.3E+05 27.6 19.4± 1.6 25.0± 0.1 + + P. humuli Hol1 H. lupulus Nugget MI Berrien 1.8E+05 13.4 18.2± 1.3 26.6± 0.1 + + P. humuli Hol4 H. lupulus Centennial MI Berrien 6.0E+04 12.8 15.5± 0.1 25.9± 0.1 n.d. n.d. P. humuli Mars1 H. lupulus Cascade MI Berrien 2.5E+05 24.1 17.4± 0.4 25.4± 0.1 + + P. humuli Mars2 H. lupulus Cascade MI Berrien 3.3E+05 31.3 15.6± 0.2 25.2± 0.1 + + P. humuli SWM1 H. lupulus Cascade MI Berrien 2.0E+05 24.5 16.0± 0.8 24.1± 0.1 + + P. humuli SWM2 H. lupulus Cascade MI Berrien 3.5E+04 12.3 18.9± 0.6 26.1± 0.0 + +

126

Table 3.3. (cont’d). Concentrationx qPCR (Cq)y RPAz Organism Isolatev Host Cultivarw State County Spor. DNA orf306 c125015.3e1 Slope 1st Der. P. cubensis CDM8 Cucumis - MI Tuscola - - n.d. n.d. n.d. n.d. sativus P. cubensis CDM15 C. stativus - MI Monroe - - n.d. n.d. n.d. n.d. P. cubensis CDM62 C. stativus - MI Saginaw - - n.d. n.d. n.d. n.d. P. cubensis CDM127 C. stativus - MI Allegan - - n.d. n.d. n.d. n.d. P. cubensis CDM16 C. stativus - MI Bay - - n.d. n.d. n.d. n.d. P. cubensis CDM17 C. stativus - MI Wayne - - n.d. n.d. n.d. n.d. P. cubensis CDM35 C. stativus - OH - - - n.d. n.d. n.d. n.d. P. cubensis CDM51 C. stativus - NY - - - n.d. n.d. n.d. n.d. P. cubensis CDM68 C. stativus - ON - - - n.d. n.d. n.d. n.d. Peronospora BH3 Ocimum - MI - - - n.d. n.d. n.d. n.d. belbahrii basilicum Phytophthora DP1 Solanum sp. - MI - - - n.d. n.d. n.d. n.d. infestans P. capsici SP98 Cucurbita - MI - - - n.d. n.d. n.d. n.d. pepo P. drechsleri 7304 Euphorbia - MI - - - n.d. n.d. n.d. n.d. pulcherrima Pythium 661 Geranium sp. - MI - - - n.d. n.d. n.d. n.d. aphanidermetrum Botrytis cinerea BH001 Zinnia - MI - - - n.d. n.d. n.d. n.d. elegans Clindrocarpon BH001 Panax - MI - - - n.d. n.d. n.d. n.d. destructans quinquefolius Rhizoctonia solani BH001 Geranium sp. - MI - - - n.d. n.d. n.d. n.d. H. lupulus - - Nugget MI - - - n.d. n.d. n.d. n.d. Mean 4.5E+05 23.5 16.4 25.3 - - (95% Confidence Interval) (1.1E+05) (3.7) (1.1) (0.6)

127

Table 3.3. (cont’d). v Pseudoperonospora humuli sporangia isolates were collected from primary and secondary symptomatic shoots under aseptic conditions during a 2018 survey of hop yards in Michigan’s lower peninsula (Higgins et al. 2020). DNA or live cultures for the other isolates was provided by Dr. M Hausbeck at Michigan State University (isolate notation refers to the culture or DNA collection). w Confidential = cultivar name withheld as requested by grower. x Spor = sporangia concentration (sporangia/ml). DNA (ng/µl) from sporangia was extracted using a CTAB buffer, phenol:chloroform (Sambrook et al. 1989), mechanical homogenizer, and sonication as described in (Higgins et al. 2020). All extracted DNA was quantified on a Qbit 4 fluorometer using the Qubit dsDNA high-sensitivity assay (ThermoFisher Scientific). y qPCR = real-time quantitative polymerase chain reactions. Cq = quanification cycle. The mitochondrial open reading frame 306 (orf306) was identified from the mitochondrial genome of P. humuli Rahman et al. (2019). Nuclear markers C125015.3e1 was designed from candidate diagnostic loci proposed by Rahman et al. (2019). n.d. = no detection. Z RPA = Recombinase polymerase amplification. Positive amplification was determined using a slope threshold of 2 mV/sec (slope) and a fluorescence peak from the first derivative analysis (1st Driv.). n.d. = no detection. +/ n.d. = no detection one of the two technical replicates.

128

Symptomatic and asymptomatic shoots. Symptomatic and asymptomatic shoots were collected from ‘Cascade’, ‘Cashmere’, and ‘Nugget’ plants with a mix of symptomatic and asymptomatic shoots in Michigan hop yards. One shoot per plant was removed with a new sterile surgical blade and new gloves and placed in a clean, resealable bag. Samples were immediately placed in a cooler for transport back to the lab. For a negative control, shoots were removed under aseptic conditions from three-month old healthy-appearing ‘Cluster’ plantlets grown in a greenhouse free of HDM. Leaves were removed from all samples under aseptic conditions and stems were stored at -80 C until DNA extraction (described above). For qPCR TaqMan assays,

DNA was extracted from a single piece of tissue excised between either end of the stem. An adjacent section of stem tissue was excised and used for the crude DNA extraction for the RPA assay. For the qPCR TaqMan assays, controls at the LOD for the orf306 (100 fg) and c125015.3e1 (1000 fg) assays were used to determine the upper threshold for Cq values for a positive reaction (three technical replicates); P. humuli DNA at 1 ng was used as a positive control (three technical replicates).

Results

qPCR sensitivity. The standard curve for the TaqMan qPCR assay based on the mitochondrial marker orf306 had a linear correlation with a regression coefficient (R2) of 0.995

(P. humuli DNA only). The orf306 assay with hop DNA had a R2 of 0.994. The orf306 assay without hop DNA had slope of -3.533 and efficiency of 91.9%. The orf306 assay with hop DNA had slope of-3.728 and efficiency of 85.4%. Fewer Cqs (approx. 1 Cq) were needed to detect the orf306 probe in presence of hop DNA than when hop DNA was absent (Table 3.4). The LOD of

129

the orf306 assay was 100 fg with (Cq = 32.36± 0.39 standard deviation [s.d.]) and without (Cq =

31.31± 0.18 s.d.) hop DNA (Figure 3.1A). The internal plant control detected 20 ng of hop DNA in 17.41± 0.32 Cqs (Table 3.4). There were no differences in Cq values for the internal plant control in the presence or absence of P. humuli DNA (Table 3.4). No amplification was detected in the HEX channel (data not shown).

The standard curve for the TaqMan qPCR assay based on the nuclear marker c125015.3e1 had a linear correlation with a R2 of 0.998 (P. humuli DNA only). The

C125015.3e1 assay with hop DNA had a R2 of 0.993. The c125015.3e1 assay without hop DNA had a slope of -3.529 and efficiency of 92.0%. The c125015.3e1 assay with hop DNA had a slope of -3.611 and efficiency of 89.2%. More Cqs (approx. 1 Cq) were needed to detect the c125015.3e1 probe in presence of hop DNA than when hop DNA was absent (Table 3.4). The

LOD of the c125015.3e1 assay was 1000 fg with (Cq = 35.38± 0.54) and without (Cq = 36.84±

1.08) hop DNA (Figure 3.1B). The internal plant control detected 20 ng of hop DNA in 18.86±

0.35 Cqs (Table 3.4). There were no differences in Cq values for internal plant control in the presence or absence of P. humuli DNA (Table 3.4). No amplification was detected in the HEX channel (data not shown).

The standard curve for the c126365.1e5 nuclear marker in the TaqMan assay had a linear correlation with a R2 of 0.987 (P. humuli DNA only). The c126365.1e5 assay with hop DNA had a R2 of 0.991. The c125015.3e1 assay without hop DNA efficiency had a slope of -3.504 and efficiency of 92.9%. The c125015.3e1 assay with hop DNA efficiency had a slope of -3.477 and efficiency of 93.9% (with hop DNA). There was no difference in the Cq values needed the to detect fluorescence from the c126365.1e5 probe in presence or absence of hop DNA (Table 3.4).

The LOD of the c126365.1e5 assay was 1000 fg with (Cq = 35.03± 1.41) and without (Cq =

130

34.71± 0.84 s.d.) hop DNA (Figure 3.1C). The internal plant control detected 20 ng of hop DNA in 18.81± 0.82 Cqs (Table 3.4). There were no differences in Cq values for internal plant control in the presence or absence of P. humuli DNA (Table 3.4). No amplification was detected in the

HEX channel (data not shown).

Table 3.4. Ten-fold serial dilution for three multiplexed assays to detect Pseudoperonospora humuli using a mitochondrial or nuclear marker and an internal plant control. Mean Quantification Cycle (Cq)y P. humuli Humulus lupulus P. humuli H. lupulus orf306 c125015.3e1 c126365.1e5 orf306z c125015.3e1 c126365.1e5 10 0 14.54± 0.17 21.29± 0.07 21.07± 0.11 * * - 1 0 18.07± 0.27 24.85± 0.11 24.45± 0.04 * * - 0.1 0 21.98± 0.11 28.38± 0.08 28.40± 0.41 * * - 0.01 0 25.90± 0.25 31.95± 0.18 31.56± 0.11 - - - 0.001 0 28.28± 0.31 35.38± 0.54 35.03± 1.41 - - - 0.0001 0 32.36± 0.39 - - - - - 0.00001 0 ------10 20 13.05± 0.28 22.42± 0.06 20.99± 0.10 16.95± 0.17 18.03± 0.87 18.62± 0.44 1 20 16.57± 0.22 25.83± 0.01 24.54± 0.29 17.74± 0.30 18.86± 0.66 18.47± 1.75 0.1 20 20.63± 0.22 29.38± 0.12 28.21± 0.14 17.18± 0.29 18.30± 0.51 18.58± 1.74 0.01 20 24.91± 0.33 33.10± 0.35 31.87± 0.56 17.73± 0.28 18.58± 0.26 18.22± 0.20 0.001 20 28.21± 0.63 36.84± 1.08 34.71± 0.84 17.40± 0.29 18.70± 0.81 18.89± 1.04 0.0001 20 31.31± 0.18 - - 17.70± 0.36 19.33± 1.03 19.05± 0.47 0.00001 20 - - - 17.59± 0.57 19.39± 0.57 18.11± 0.55 0 20 - - - 17.41± 0.32 18.86± 0.35 18.81± 0.82 y Number values are the mean Cq± the standard deviation. z * = Inconsistent amplification by the internal plant control; mean Cq = 36.73± 5.31 s.d.

qPCR specificity. For the orf306 assay, no amplification was observed in non-template controls or with the DNA of off-target organisms (Table 3.3). The orf306 assay correctly identified all P. humuli sporangia isolates. The average Cq value for the orf306 assay from P. humuli sporangia isolates (DNA standardized to 5 ng/ul) was 16.4 (95% confidence interval

(C.I.) = 1.1 Cq). In specificity tests for the c125015.3e1 assay no amplification was observed in non-template controls or with the DNA of off-target organisms (Table 3.3). The c125015.3e1 marker correctly identified all P. humuli sporangia isolates. The average Cq value for the

131

c125015.3e1 marker from P. humuli sporangia isolates (DNA standardized to 5 ng/ul) was 25.3

(95% CI = 0.6 Cq). In specificity tests, DNA extraction from P. humuli sporangia suspensions using phenol/chloroform/CTAB yielded on average 23.7 ng/µl (95% C.I. = 3.7 ng/ul).

132

qPCR asymptomatic leaves. No visual DM symptoms were observed in control, TO, 1 dpi, and 2 dpi leaf disks. All inoculated leaf disks tested had light chlorosis at 3 dpi and fully developed chlorotic lesions at 7 dpi (Figure 3.2). The three-way interaction between marker, dpi and run was insignificant (F = 0.06; P = 0.9814). The two-way interaction between marker and dpi was insignificant (F= 0.20; P = 0.8967). There was no significant effect of experimental runs

(F=3.71; P = 0.0584). There were significant effects observed for amongst dpi treatments (F =

33.96; < 0.0001). The type of marker used in a qPCR assay (mitochondrial vs nuclear) had a significant effect on Cq value (F = 329.31; P < 0.0001). Cq values were approximately four cycles higher for assays with the nuclear marker than the mitochondrial marker at each time point post inoculation (P = 0.001) (Figure 3.3).

Figure 3.2. The progression of downy mildew symptoms in leaves inoculated with Pseudoperonospora humuli sporangia (1 x104 sporangia/ml). Leaf disks removed at the site of inoculation one, two, three, and seven days post inoculation (dpi). TO = leaf disks removed at the time of inoculation. Control leaves (left) received only water. Leaf disks were subjective to pathogen detection with quantitative PCR and recombinase polymerase amplification assays.

133

For the orf306 assay, there was no amplification in the non-template control. Cq values at the LOD during the assay were 33.07± 0.05 s.d. (exp. 1) and 32.43± 0.27 (exp. 2). P. humuli was not detected at TO or in the control leaves for each time post inoculation. In experiment 1, there was a slight lift in amplification beyond the LOD in one of two technical replicates for four control and two T0 leaves (Cq = 39.06± 0.58) (Figure A2). P. humuli was detected in all five replicates post inoculation and prior to symptom development. Cq values (orf306) were higher in symptomatic leaves at 1 dpi (23.2.6 Cq) than at 2 dpi (21.1 Cq) (P = 0.001). In symptomatic leaves, P. humuli was detected in all replicates. Cq values did not differ among symptomatic leaves at 3 dpi (21.1 Cq) and 7 dpi (20.5 Cq) (P = 0.001). Only asymptomatic leaves at 1 dpi had significantly higher Cq values than symptomatic leaves (P = 0.001).

For the c125015.3e1 assay, there was no amplification in the non-template control. Cq values at the LOD during the assay were 36.47± 0.58 (exp. 1; only two of three technical replicates amplified) and 35.14 (exp 2; only one of three technical replicates amplified). P. humuli was not detected (no amplification) at TO or in the negative controls for each time period. P. humuli was detected in all five replicates post inoculation and prior to symptom development. Cq values were higher in symptomatic leaves at 1 dpi (28.0 Cq) than at 2 dpi (25.6

Cq) (P = 0.001). In symptomatic leaves, P. humuli was detected in all replicates. Cq values did not differ among symptomatic leaves at 3 dpi (24.5 Cq) and 7 dpi (24.8 Cq) (P = 0.001). Only asymptomatic leaves at 1 dpi had significantly higher Cq values than symptomatic leaves (P =

0.001).

134

qPCR asymptomatic shoots. The orf306 and c125015.3e1 assays were applied to symptomatic and asymptomatic shoots collected from diseased hop yards (Table 3.5). For the orf306 assay, there was no amplification in the non-template control. Cq values at the LOD during the assay were 34.69± 0.05 s.d. P. humuli was not detected in healthy-appearing shoot tissue. There was a slight lift in amplification beyond the LOD in one or both technical replicates for three asymptomatic shoots (Figure A2). The Cq value for the internal plant control in all healthy tissue was 14.49 (95% C.I. = 0.32). P. humuli was detected in 100% of the symptomatic shoots tested (Cq = 17.82; 95% C.I. = 2.16.). P. humuli was detected in 100% of the asymptomatic shoots tested (Cq = 31.4; 95% C.I. = 0.65).

135

For the c125015.3e1 nuclear marker assay, there was no amplification in the non- template control. Cq values at the LOD during the assay were 36.90± 0.53 s.d. P. humuli was not detected (no amplification) in healthy shoot tissue. The Cq value for the internal plant control in all healthy tissue was 11.2 (95% C.I. = 1.28). P. humuli was detected in 100% of the symptomatic shoots tested (Cq = 24.0; 95% C.I. = 2.11.). P. humuli was detected in 40% (n = 4) of the asymptomatic shoots tested (Cq = 35.96; 95% C.I. = 0.44).

Table 3.5. Detection of hop downy mildew in symptomatic and asymptomatic shoots collected from diseased hop yard and tested with qPCR assays containing a mitochondrial (orf306) or nuclear (c125015.3e1) markers and an internal plant control (IPC). # Positive Detections # Negative Detections (Ave. Cq ± s.d.; P. humuli)y (Ave. Cq ± s.d.; IPC)z Hop shootx Mitochondrial Nuclear Mitochondrial Nuclear Healthy 0 (n.d.*) 0 (n.d.) 10 (14.49± 0.52) 10 (11.23± 2.06) Symptomatic 10 (17.82± 3.40 10 (24.00 ± 1.10) 0 (15.13± 0.36) 0 (11.66± 2.29) Asymptomatic from 10 (31.37± 1.07) 4 (35.95± 0.46) 0 (14.94± 0.64) 6 (12.44± 0.76) diseased hopyards AS21_Cascade (31.61± 0.03) (35.79± 0.38) (15.91± 0.10) (12.85± 0.43) AS22_Cashmere (32.22± 0.05) n.d. (13.94± 0.01) (11.10± 1.19) AS23_Nugget (30.58± 0.58) (37.39) (14.71± 0.24) (13.02± 0.56) AS24_Nugget (30.71± 0.41) (39.53) (14.45± 0.17) (12.80± 0.72) AS25_Cascade (32.81± 0.27) (35.41) (15.18± 0.41) (12.68± 0.56) AS26_Cascade (32.12± 0.06) (36.50± 0.95) (15.85± 0.17) (12.40± 0.76) AS27_Cashmere (32.42± 0.24) (36.11± 0.85) (14.76± 0.72) (12.22± 0.31) AS28_Cascade (29.61± 0.29) (37.43) (14.59± 0.08) (11.89± 0.84) AS29_Cashmere (31.38± 0.04) (37.31) (15.29± 0.05) (12.71± 0.58) AS30_Cashmere (30.21± 0.96) (35.43± 0.02) (14.76± 0.11) (12.73± 0.75) Limit of Detection (34.69± 0.05) (36.90± 0.53) - - Total 20 14 10 16 x AS##_ stands for the sample identification (i.e. asymptomatic; Sample #21) and is followed by the cultivar it was collected from. y A positive detection was defined as the amplification in both technical replicates and Cq lower than the assay’s limit of detection. * = sporadic amplification above the limit of detection. n.d. = no detection. Cq = quantification cycle. A single Cq value indicates that only one of two technical replicates was amplified. z Negative detection refers to positive amplification of the host plant DNA with the IPC, but no amplification of P. humuli DNA.

136

RPA sensitivity. Positive amplification was determined separately using a slope threshold (2 mV/sec) and a fluorescence peak from the first derivative analysis. The average slope of amplification in non-template controls was 0.56± 0.64 s.d. mV/sec without hop DNA and 1.33± 1.05 with hop DNA. The maximum slope of amplification in non-template controls was 1.77 and 2.89 with and without hop DNA, respectively. Amplification curves with no distinct second phase prior to or after the mixing step were observed above the slope threshold in

16% (n = 2) of (2.45 and 2.46 mVs/sec) of non-template controls and only in non-template control where hop DNA was added (Figure A3). No amplification was observed in non-template controls after the first derivative analysis was applied (Figure 3.4). A slight lift in the amplification curve observed in other non-template controls at 22+ min into the reaction.

Reaction mixtures with and without primers were tested on NTC, crude hop leaf extract, 10 ng P. humuli DNA, and 10 ng P. humuli + crude hop leaf extract in four replicate runs. In the absence of primers, a rise in fluoresce was observed at 22+ min (Figure A4).

The onset of amplification was determined separately by the time amplification crosses the slope threshold and the peak of the second derivative curve is reached. The R2 values were greater for sensitivity plots with P. humuli DNA constructed from the second derivative analysis

(R2 = 0.976) than the slope threshold (R2 = 0.960). The R2 values were greater for sensitivity plots spiked with hop DNA constructed from the second derivative analysis (R2 = 0.981) than the slope threshold (R2 = 0.968) (Figure 3.5). The presence of hop DNA increased the time to amplification onset and reduced the assay sensitivity (Figure 3.5). In the absence of hop DNA amplification occurred at 100 ng (P. humuli DNA) in less than 12.0 (slope threshold) and 13.5

(2nd derivative) minutes. In the presence of hop DNA amplification occurred at 1000 ng (P. humuli DNA) in less than 16 (slope threshold) and 18 (2nd derivative) minutes.

137

Figure 3.4. Recombinase polymerase amplification (RPA) fluorescent signal generated for Pseudoperonospora humuli DNA (A) and non-template controls below (B) and above (C) a user-defined slope threshold established (2 mV/sec) to determine positive amplifications. A first derivative analysis applied to the same amplification curves (D-F) used to demonstrate a lack of positive amplification in non-template controls (E and F).

138

139

RPA specificity. Positive amplification was determined using the slope threshold (2 mV/sec) and the fluorescence peak from the first derivative analysis. There were no positive detections in non-template controls (n = 10). The orf306 RPA assay correctly identified most P. humuli isolates (Table 3.3). The slope detection failed to detect P. humuli isolate JB2 in one of the two technical replicates; this isolate was detected in both technical replicates with the first derivative analysis. Both methods of calling positive amplification failed to detect P. humuli isolate Hol4. There was no amplification with the DNA of non-target organisms.

RPA asymptomatic leaves. No visual DM symptoms were observed in control, TO, 1 dpi, and 2 dpi leaf disks. All inoculated leaf disks tested had slight chlorosis at 3 dpi and fully developed chlorotic lesions at 7 dpi as described above (Figure 3.2). Asymptomatic infections were detected by both the slope threshold and first derivative analysis as early as one day post inoculation; positive amplification occurred in all replicates for both experiments (Table S3.9).

There was no positive detection in control leaves at each time point or at T0.

RPA asymptomatic shoots. Both methods of determining a positive amplification produced the same detection in symptomatic, asymptomatic, and healthy shoots (Table 3.6). No false positives detections were observed. Both methods failed to amplify P. humuli DNA in the same symptomatic shoot. P. humuli was detected in 20% of the asymptomatic shoots tested.

140

Table 3.6. Results of RPA assay with the orf306 (mitochondrial) markers on healthy shoots, spikes and asymptomatic shoots collected from disease MI hop yards with diseased plants. # Positive Detectionsz # Negative Detections Hop shooty Slope 1st Derivative Slope 1st Derivative Healthy 0 0 10 10 Symptomatic 9 9 1 1 Asymptomatic from 2 2 8 8 diseased hopyards Total 11 11 18 18 y Symptomatic downy mildew shoots are stunted, chlorotic and often swollen with downward cupped leaves. Shoots from asymptomatic plants were collected adjacent to symptomatic plants. Healthy plants were raised disease-free in a greenhouse. z Positive Pseudoperonospora humuli amplification was based on a slope threshold of 2 m/V (slope) and first derivative analysis (1st Derivative).

141

Discussion

Asymptomatic HDM infections in hop planting material can escape detection due to the incubation period of 10 or 21 days for leaves and shoots, respectively. DNA-based diagnostic assays for P. humuli may be useful for detecting asymptomatic infections (Summers et al. 2015) and help keep planting material for new hop yards disease-free. Diagnostic assays based on a high copy number mitochondria markers are likely more sensitive than single copy nuclear marker (Rahman et al. 2017; Rahman et al. 2020), but a side-by-side comparison has yet to be reported. We compared qPCR TaqMan assay based on nuclear (C125015.3e1 and C126365.1e5;

(Rahman et al. 2019)) and mitochondrial (orf306) loci and found the mitochondrial marker-based assay to be more sensitive in detecting raw P. humuli DNA and asymptomatic infections in leaves and shoot. The orf306 mitochondrial locus was then developed in an RPA assay to see if the hallmarks of the technology, a crude DNA extraction and amplification at a constant temperature, could be deployed as diagnostic tool for HDM management in the nursery. We found that using a first derivative analysis reduces the risk of a false positive detection. However, for this particular RPA assay, crude hop DNA extract appears to negatively impact RPA sensitivity. Thus, the mitochondrial qPCR assay is the best option to improve disease management by ensuring that the planting material used to establish hop yards is free of HDM.

In qPCR TaqMan sensitivity tests with P. humuli and mixed pathogen/plant DNA, assays based on a mitochondrial target (orf306) were ten times more sensitive than assay with nuclear loci (c125015.3e1 and c126365.1e5). For nuclear assays, both c125015.3e1 and c126365.1e5 assays had the same LOD (1000 fg). The c126365.1e5 assay appeared to perform slightly better than c125015.3e1 producing significantly lower Cq values at the lowest concentrations in the presence of hop DNA. However, the primers for c125015.3e1 marker had six potential for off-

142

target binding sites with sequence homology as high as 64% and were not tested further to avoid any potential downstream specificity complications. The standard curves for all assays showed the assays were not affected by the presence of hop DNA at 20- to 200-thousand times the amount of P. humuli DNA indicating each assays suitability for use with environmental samples.

In orf306 and c125015.3e1 there was a slight improvement in Cq values when hop DNA was added to the assay (approx. 1 Cq), but no cross reactions with hop DNA were detected. The Cq values for the IPC remained consistent in the presence of P. humuli DNA at 0.5 times the amount of hop DNA.

Target loci, optimization, and amplification conditions confound comparisons between qPCR assays (Bastien et al. 2008). The LODs in our qpCR assays were 1000 fg (nuclear; c125015.3e1 and c126365.1e5) and 100 fg (mitochondrial; orf306). A high mitochondria- nucleus ratio in sporangia (Constantinescu 2000; Lange et al. 1989a) may explain some difference in sensitivity. However, in sporangia undergoing zoospore formation there can be 6-8 nuclei per sporangium; a fully delineated zoospore is uninucleate with numerous mitochondria

(Lange et al. 1989a, b). The DNA in our sensitivity test was derived from sporangia suspended in water suspensions and on ice for approx. 15-20 min. and zoospore formation is unlikely to have occurred. In our assay with asymptomatic leaves and shoot, the hyphae and haustoria would be the target. Intercellular hyphae contain both nuclei and mitochondria (Cohen et al. 1989;

Constantinescu 2000). Developing haustoria lack nuclei (Cohen et al. 1989; Constantinescu

2000) but contain an abundance of mitochondria (Cohen et al. 1989). Difference in sensitivity leaf and shoot assays may in part have been influenced by haustoria with high mitochondria numbers.

143

The nuclear ribosomal ITSrDNA marker developed by Gent et al. (2009) had an LOD was 1 fg but was performed as a singleplex assay and cross reacted with P. cubensis. The multiplex nuclear marker developed to detect P. cubensis clades in spore traps had an LOD of

100 fg (Rahman et al. 2020). The LOD for the mitochondrial marker based on a SNP in the Cox2 gene was 10 fg (Summers et al. 2015), but there was high (approx. 4 Cq) and overlapping variation at the LOD and 100 fg with three technical replicates. Plant DNA contamination from biotrophic propagation of P. humuli and P. cubensis on detached may result in inaccurate DNA concentrations used in standard dilution to determine the LOD (the developmental stage of the sporangia harvested for DNA may also be a confounding factor). In our assay there was occasional amplification by the IPC in serial dilutions with only P. humuli DNA which means we probably overestimated the LOD of assays. None of the above-mentioned assay contain an

IPC so it is difficult to determine the purity of P. humuli DNA used. Even the comparison of the mitochondrial and nuclear assays tested in this study with the same DNA dilution set are not free of potential bias. For instance, the extension time was shorter for the nuclear marker-based assays (0:30 sec) than the mitochondrial assay (1:30 sec). Generally, extension times for Taq

DNA polymerase are 1000 bp per minute (anonymous). A longer extension time (1:30 sec) tested for the nuclear assays resulted in a severe reduction in assay sensitivity. For the mitochondrial assay, a longer extension time was chosen to overcome potential limitation caused by long A-T rich templates in other oomycetes (Bilodeau et al. 2014; Miles et al. 2017). However, it remains a possibility that further optimization of extension times would improve sensitivity of the mitochondrial assay.

In specificity tests, both the mitochondrial and nuclear marker-based assays were able to detect all P. humuli sporangium isolates. Overall, lower Cq values were observed in the orf306

144

assay compared to the c125015.3e1 assay are consistent with differences in sensitivity. P. humuli is considered highly clonal (Gent et al. 2019) and preliminary reports in Michigan show limited genetic diversity (Higgins et al. 2020) making this assay suitable for Michigan and possibly the surrounding great lake region. However, the assay should be validated using isolates from other production regions, such as the Pacific Northwest, to confirm specificity.

The results of the asymptomatic leaf disk assay showed that both assays could detect asymptomatic infections as early as 1 dpi. Summers et al. (2015) were able to detect asymptomatic P. humuli infections at 3 dpi with Cq values ranging from approximately 27 to

31and nearing the LOD. It remains unknown if infections caused by lower sporangia concentration are detectable, but both assays tested in this study were well below their LOD at 1 dpi and have the capacity for detection at lower pathogen levels. The mitochondrial marker- based assay (orf306) outperformed the nuclear marker-based assay by detecting 100% of asymptomatic shoots. Only 40% of asymptomatic shoots were detected by the nuclear marker- based assay (C125015.3e1) and positive reactions were at or near the LOD.

In RPA , slope thresholds are user-defined values determined from florescence background signal in non-template controls; signal above the threshold indicates a positive amplification (Burkhardt et al. 2019; Burkhardt et al. 2018; Miles et al. 2015; Rojas et al.

2017). In this study, the amplification curves of non-template controls exceeded the slope threshold in 16% (n = 2) of reactions during sensitivity tests. The shape of these curves is unusual because the positive slope remains relatively consistent prior to and after the 4-min mixing step. In RPA, continuous annealing at a single reaction temperature can produce a constant rate of fluorescence (Li et al. 2019). RPA reaction assays start outside of the fluorometer and it is conceivable that a highly efficient assay or a high starting concentration of

145

the target template could cause amplification at the start of fluorescence detection. Otherwise, it is typical to generate at least two different reaction rates (Ammour et al. 2017; DeShields et al.

2019; Miles et al. 2015) with the first consisting of relatively little fluorescence signal as polymerization begins to increase target template and accelerate the reaction rate to detectable fluorescence levels; a second agitation step at four min is typically required to prevent the localized depletion of the reaction reagents (Li et al. 2019). Strayer-Scherer et al. (2019) reported

RPA amplification curves with at least two reaction rates for Xanthomonas gardneri and X euvesicatoria, but a qualitative distinction of reaction rates for X. perforans appear to be more difficult to make.

Non-specific amplification due to indiscriminate binding sites or amplification of off- target DNA is another potential explanation for fluorescence signal in non-template controls

(Daher et al. 2015; Li et al. 2019). The slight increase in amplification rate observed in other non-template controls occurs 22+ min into the reaction. In RPA, all available ATP is believed to be consumed within 25 min preventing the formation of the recombinase-primer complex

(Lobato and O'Sullivan 2018) and amplification of target sequences occurring at the end of the incubation period is unlikely. Further a similar late increase in amplification rate was observed in non-template controls and reactions contain hop and P. humuli DNA. No amplification of DNA from non-target organism was observed and the primers selected share 54.8 % (forward) and

68.6 % (reverse) homology with closely related P. cubensis. Since the output data from the

Axxin T16-ISO isothermal fluorometer is raw fluorescence data without a baseline correction and we consider amplification below the slope threshold for negative controls to be background signal. Further, it appears that relying on the slope threshold alone could result in false positive detections.

146

In real-time reverse transcription PCR (RT-PCR) the first derivative curve is generated when there is a persistent geometric increase in fluorescence compared to a baseline signal

(Figure 3.6) (Luu-The et al. 2005; Rasmussen 2001). To detect a change in reaction rates in our

RPA assay, we applied a first derivative analysis of raw fluorescence data. The first and second derivatives can be applied directly to raw output data in the Axxin AX0ISO desktop application.

In the first derivative view, it was visually clear when a change of rate had positive amplifications. We adapted this approach from El Wahed et al. (2015) who also describe a similar development of non-specific or negative fluorescence in the raw data during the detection of dengue virus with reverse transcription RPA and concluded that a first derivative analysis was required to confirm true amplification. Aside from the non-template controls in the RPA sensitivity tests, the first derivative analysis successful detected a sporangia isolate that was not

147

detected using the slope threshold in specificity test. Increasing the slope threshold to ensure detection would not be advised because it would increase the risk of a false negative which may carry more negative consequences.

We used the time that the fluorescence signal crosses the slope threshold to determine the onset of amplification. The slope threshold has been used by others (Miles et al. 2015; Rojas et al. 2017) or combined with a user-defined fluorescence value (Burkhardt et al. 2019; Burkhardt et al. 2018) to determine the onset of amplification. We found it difficult to define a single fluorescence value since no baseline correction is applied to raw data output in the Axxin

AX0ISO desktop application. In RT-PCR, the positive peak of the second derivative amplification curve corresponds to the beginning of a log-linear phase (Figure 3.6) (Luu-The et al. 2005; Rasmussen 2001). Using the second derivative is advantageous because it does not rely on a user-dependent value. The second derivative analysis produced similar sensitivity curves with a slight improvement in R2 values. It is interesting to note the loss of sensitivity observed in the presence of crude hop DNA extract using both methods of determining amplification onset and this presents a major limitation for this assay in its current form. Another limitation of RPA assays,that generate log-linear calibration curves is that RPA is not synchronized like PCR during each thermal cycle. Since annealing is continuous at a single reaction temperature it results in a non-linear calibration curve for quantification (Li et al. 2019). The further development of digital droplet RPA (Schuler et al. 2015) will help overcome this limitation by partitioning the RPA reaction and avoiding the need for calibration curves (Li et al. 2019).

In managing HDM in greenhouse and nursery production, where the accurate detection of asymptomatic infections is important, the high sensitivity of the orf306 mitochondrial marker- based assay outperformed the c125015.3e1 nuclear -based assay. The IPC used in this study would

148

not be useful in the detection of airborne sporangia from spore traps but is essential for confirming negative detection in plant samples. Assays best suited for the detection of airborne sporangia would include markers to detect P. humuli and both P. cubensis clades (F. Martin, unpublished;

(Rahman et al. 2020). However, too much multiplexing is expensive for diagnostic labs and does not leave room for the further development of the assay to include additional markers for other pathogens that might be encountered in the nursery. While the RPA assay based on the orf306 marker requires further optimization, the locus is suitable for further exploration and improvements have been realized to reduce the number of false positives.

149

APPENDIX

150

Table A3. Quantitative PCR assays for the detection of two nuclear-based Pseudoperonospora humuli loci (c125015.3e1 and c126365.1e5) tested at a 1:30 sec extension time. Mean Quantification Cycle DNA (ng) c125015.3e1 c126365.1e5 10 26.78± 0.13 27.09± 0.11 1 30.40± 0.15 30.05± 0.32 0.1 35.33± 0.86 34.17± 0.49 0.01 - - 0.001 - - 0.0001 - - 0.00001 - -

151

Table A4. Primers and probes screened for recombinase polymerase amplification assay for the detection of open reading frame 306 in the Pseudoperonospora humuli mitochondrial genome. Primers Sequence (5’-3’) % GCy % Pairwise Identityz Forward PhF1 ATTTTTACTTTTTTATCAAAATGAATACCAAGATT 17.1 65.9 PhF2 TTTTTATCAAAATGAATACCAAGATTTTGATTTTC 20.0 68.3 PhF4 CAAGATTTTGATTTTCTTTTAACTCTTCTTCAAAA 22.9 80.5 PhF6 TTCAAAAGAATTTCTTCTTCGAATTAAAACTTTTT 20.0 77.1 PhF7 AATTTCTTCTTCGAATTAAAACTTTTTTTTCATCT 20.0 71.4 PhF8 ATTTTATTCCAAAAATAATGCGTATCCATATCTAG 25.7 88.6 PhF9 TTATTTTTTTTGTAAGTAAATCTTTTTTCCATCTA 17.1 82.9 Reverse PhR1 GATGTTTATTACAAGTATAACTGTTGGTAGTGTAT 28.6 68.6 PhR2 ATGTTTATTACAAGTATAACTGTTGGTAGTGTATT 25.7 71.4 PhR5 AAAACAGGGCTTGATGTTTATTACAAGTATAACTG 31.4 71.4 PhR6 GGGCTTGATGTTTATTACAAGTATAACTGTTGGTA 34.3 65.7 PhR7 CTTGATGTTTATTACAAGTATAACTGTTGGTAGTG 31.4 65.7 PhR8 TTGATGTTTATTACAAGTATAACTGTTGGTAGTGT 28.6 68.6 PhR9 TGATGTTTATTACAAGTATAACTGTTGGTAGTGTA 28.6 68.6 y GC = Guanine-cytosine content (%) of each primer. z Pairwise identity with P. cubensis on orf306.

152

Table A5. The detection of Pseudoperonospora humuli with a recombinase polymerase amplification assay in asymptomatic (1- and 2-days post inoculation [dpi]) and symptomatic (3- and 7-dpi) leaf disks removed at the site of inoculation. Positive/Negative Amplificationx Slope 1st Derivative Experiment Timey Controlz Inoculated Control Inoculated 1 T0 0/3 0/5 0/3 0/5 1 1 dpi 0/3 5/5 0/3 5/5 1 2 dpi 0/3 5/5 0/3 5/5 1 3 dpi 0/3 5/5 0/3 5/5 1 7 dpi 0/3 5/5 0/3 5/5 2 T0 0/3 0/5 0/3 0/5 2 1 dpi 0/3 5/5 0/3 5/5 2 2 dpi 0/3 5/5 0/3 5/5 2 3 dpi 0/3 5/5 0/3 5/5 2 7 dpi 0/3 5/5 0/3 5/5 x Positive amplification determined with a user-defined slope threshold (2 mV/sec) and a first derivative analysis applied to the same amplification curves. y Leaves (five replicates per time period) were inoculated with P. humuli sporangia (1 x104 sporangia/ml). Leaf disks were removed from inoculation sites 1, 2, 3 and 7-dpi; TO = leaf disks removed at the time of inoculation. z Control leaves (left) received only water.

153

154

155

156

A B

C D

Figure A4. Reaction mixtures without primers tested on non-template control (A), crude hop leaf extract (B), 10 ng P. humuli DNA (C), and 10 ng P. humuli + crude hop leaf (D).

157

LITERATURE CITED

158

LITERATURE CITED

Ammour, M. S., Bilodeau, G. J., Tremblay, D. M., Van der Heyden, H., Yaseen, T., Varvaro, L., and Carisse, O. 2017. Development of real-time isothermal amplification assays for on- site detection of Phytophthora infestans in potato leaves. Plant Dis. 101:1269-1277.

Bastien, P., Procop, G. W., and Reischl, U. 2008. Quantitative real-time PCR is not more sensitive than “conventional” PCR. J. Clin. Microbiol. 46:1897-1900.

Burkhardt, A., Ramon, M. L., Smith, B., Koike, S. T., and Martin, F. 2018. Development of molecular methods to detect Macrophomina phaseolina from strawberry plants and soil. Phytopathology 108:1386-1394.

Burkhardt, A., Henry, P. M., Koike, S. T., Gordon, T. R., and Martin, F. 2019. Detection of Fusarium oxysporum f. sp. fragariae from infected strawberry plants. Plant Dis. 103:1006-1013.

Bustin, S. A., Benes, V., Garson, J. A., Hellemans, J., Huggett, J., Kubista, M., Mueller, R., Nolan, T., Pfaffl, M. W., and Shipley, G. L. 2009. The MIQE Guidelines: Minimum Information for Publication of Quantitative Real-Time PCR Experiments. Oxford University Press.

Cohen, Y., Eyal, H., Hanania, J., and Malik, Z. 1989. Ultrastructure of Pseudoperonospora cubensis in muskmelon genotypes susceptible and resistant to downy mildew. Physiological and Molecular Plant Pathology 34:27-40.

Coley-Smith, J. R. 1962. Overwintering of hop downy mildew Pseudoperonospora humuli (Miy. and Tak.) Wilson. Annals of Applied Biology 50:235-243.

Coley-Smith, J. R. 1963. Hop Downy Mildew and Mist Propagation. in: Department of Hop Research Annual Report, 1962. Wye College, Ashford, Kent, UK.

Coley-Smith, J. R. 1965. Infection of hop rootstocks by downy mildew Pseudoperonospora humuli (Miy. & Tak.) Wilson and its control by early-season dusts. Ann. Appl. Biol. 56:381-388.

Constantinescu, O. 2000. The fine structure of the sporangium in Pseudoperonospora humuli (Chromista, Oomycota, Peronosporales). Cryptogam. Mycol. 21:93-101.

Daher, R. K., Stewart, G., Boissinot, M., Boudreau, D. K., and Bergeron, M. G. 2015. Influence of sequence mismatches on the specificity of recombinase polymerase amplification technology. Mol. Cell. Probes 29:116-121.

159

Darling, A. E., Mau, B., and Perna, N. T. 2010. ProgressiveMauve: multiple genome alignment with gene gain, loss and rearrangement. Plos One 5:e11147.

DeShields, J. B., Moroz, N., Braley, L. E., Mora-Romero, G. A., and Tanaka, K. 2019. Recombinase polymerase amplification (RPA) for the rapid isothermal detection of spongospora subterranea f. sp. subterranea and potato mop-top virus. American Journal of Potato Research 96:617-624.

Dorak, M. T. 2007. Real-time PCR. Taylor & Francis, New York, NY.

El Wahed, A. A., Patel, P., Faye, O., Thaloengsok, S., Heidenreich, D., Matangkasombut, P., Manopwisedjaroen, K., Sakuntabhai, A., Sall, A. A., and Hufert, F. T. 2015. Recombinase polymerase amplification assay for rapid diagnostics of dengue infection. PloS One 10:e0129682.

Gent, D. H., Nelson, M. E., and Grove, G. G. 2008. Persistence of phenylamide insensitivity in Pseudoperonospora humuli. Plant Dis. 92:463-468.

Gent, D. H., Nelson, M. E., Farnsworth, J. L., and Grove, G. G. 2009. PCR detection of Pseudoperonospora humuli in air samples from hop yards. Plant Pathol. 58:1081-1091.

Gent, D. H., Adair, N., Knaus, B. J., and Grunwald, N. J. 2019. Genotyping-by-sequencing reveals fine-scale differentiation in populations of Pseudoperonospora humuli. Phytopathology 109:1801-1810.

George, A. 2020. USA hops 2019 Statistical Report. Hop Growers of America, Online publication: usahops.org.

Granke, L., and Hausbeck, M. 2011. Dynamics of Pseudoperonospora cubensis sporangia in commercial cucurbit fields in Michigan. Plant Dis. 95:1392-1400.

Gurriarán, M., Revilla, M., and Tamés, R. 1999. Adventitious shoot regeneration in cultures of Humulus lupulus L.(hop) cvs. Brewers Gold and Nugget. Plant Cell Rep. 18:1007-1011.

Higgins, D. S., Miles, T. D., and Hausbeck, M. 2020. Fungicide efficacy against Pseudoperonospora humuli and point-mutations linked to carboxylic acid amide (CAA) resistance in Michigan. Plant Dis. (First look) Online: https://doi.org/10.1094/PDIS-01- 20-0023-RE

Howard, B. 1965. Regeneration of the hop plant (Humulus lupulus L.) from softwood cuttings. I. The cutting and its rooting environment. J. Hortic. Sci. 40:181-191.

Howard, B. 1967. Regeneration of the hop plant (Humulus Lupulus L.) from softwood cuttings: III. Treatment with root-promoting substances. J. Hortic. Sci. 42:105-107.

160

Johnson, D. A., and Skotland, C. 1985. Effects of temperature and relative humidity on sporangium production of Pseudoperonospora humuli on hop. Phytopathology 75:127- 129.

Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P., and Drummond, A. 2012. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28:1647- 1649.

Lange, L., Edén, U., and Olson, L. W. 1989a. Zoosporogenesis in Pseudoperonospora cubensis, the causal agent of cucurbit downy mildew. Nord. J. Bot. 8:497-504.

Lange, L., Edén, U., and Olson, L. W. 1989b. The zoospore of Pseudoperonospora cubensis, the causal agent of cucurbit downy mildew. Nord. J. Bot. 8:511-516.

Li, J., Macdonald, J., and von Stetten, F. 2019. A comprehensive summary of a decade development of the recombinase polymerase amplification. Analyst 144:31-67.

Lobato, I. M., and O'Sullivan, C. K. 2018. Recombinase polymerase amplification: basics, applications and recent advances. Trends Analyt. Chem. 98:19-35.

Lu, W.-J., Hu, W.-G., and Wang, G.-P. 2016. Complete mitochondrial genome of Pseudoperonospora cubensis. Mitochondrial DNA 27:3487-3488.

Luu-The, V., Paquet, N., Calvo, E., and Cumps, J. 2005. Improved real-time RT-PCR method for high-throughput measurements using second derivative calculation and double correction. Biotechniques 38:287-293.

Mahaffee, W. F., Pethybridge, S. J., and Gent, D. H. 2009. Compendium of Hop Diseases and Pests. American Phytopathological Society (APS Press), St. Paul, MN.

McCoy, A. G., Miles, T. D., Bilodeau, G. J., Woods, P., Blomquist, C., Martin, F. N., and Chilvers, M. I. 2020. Validation of a preformulated, field deployable, recombinase polymerase amplification assay for Phytophthora species. Plants 9:466.

Meuer, S., Wittwer, C., and Nakagawara, K.-I. 2012. Rapid Cycle Real-Time PCR: Methods and Applications. Springer Berlin, Heidelberg.

Miles, T. D., Martin, F. N., and Coffey, M. D. 2015. Development of rapid isothermal amplification assays for detection of Phytophthora spp. in plant tissue. Phytopathology 105:265-278.

161

Mitchell, M. N., Ocamb, C. M., Grünwald, N. J., Mancino, L. E., and Gent, D. H. 2011. Genetic and pathogenic relatedness of Pseudoperonospora cubensis and P. humuli. Phytopathology 101:805-818.

NCPN. 2020. What is a clean plant?, Online: http://cpcnw.wsu.edu/what-is-a-clean-plant/.

Neve, R. A. 1991. Hops. Springer, Berlin, Germany.

O’Neal, S. 2015. Pest Management Strategic Plan for U.S. Hops: Summary of a workshop held on November 4, 2014 in Yakima, WA. Online.

Patzak, J. 2005. PCR detection of Pseudoperonospora humuli and Podosphaera macularis (Humulus lupulus). Plant Prot. Sci. 41:141.

Peredo, E. L., Arroyo-Garcia, R., and Revilla, M. Á. 2009. Epigenetic changes detected in micropropagated hop plants. J. Plant Physiol. 166:1101-1111.

Piepenburg, O., Williams, C. H., Stemple, D. L., and Armes, N. A. 2006. DNA detection using recombination proteins. PLoS Biol. 4:e204.

Rahman, A., Standish, J., D'Arcangelo, K. N., and Quesada-Ocampo, L. M. 2020. Clade-specific biosurveillance of Pseudoperonospora cubensis using spore traps for precision disease management of cucurbit downy mildew. Phytopathology (First Look) Online: https://doi.org/10.1094/PHYTO-06-20-0231-R.

Rasmussen, R. 2001. Quantification on the LightCycler. in: Rapid Cycle Real-Time PCR: Methods and Applications. S. Meuer, C. Wittwer and K.-I. Nakagawara, eds. Springer, Heidelberg, Germany.

162

Rojas, J. A., Miles, T. D., Coffey, M. D., Martin, F. N., and Chilvers, M. I. 2017. Development and application of qPCR and RPA genus-and species-specific detection of Phytophthora sojae and P. sansomeana root rot pathogens of soybean. Plant Dis. 101:1171-1181.

Royle, D. 1970. Infection periods in relation to the natural development of hop downy mildew (Pseudoperonospora humuli). Ann. Appl. Biol. 66:281-291.

Royle, D., and Thomas, G. 1971a. The influence of stomatal opening on the infection of hop leaves by Pseudoperonospora humuli. Physiol. Plant Pathol. 1:329-343.

Royle, D., and Thomas, G. 1971b. Observations with the scanning electron microscope on the early stages of hop leaf infection by Pseudoperonospora humuli. Physiol. Plant Pathol. 1:345-349.

Royle, D., and Kremheller, H. T. 1981. Downy Mildew of the Hop. in: The Downy Mildews. D. M. Spencer, ed. Academic Press Inc., New York, NY.

Rybaček, V. 1991. Hop Production. Elsevier, New York, NY.

Schuler, F., Schwemmer, F., Trotter, M., Wadle, S., Zengerle, R., von Stetten, F., and Paust, N. 2015. Centrifugal step emulsification applied for absolute quantification of nucleic acids by digital droplet RPA. Lab Chip 15:2759-2766.

Sirrine, J. R., Lizotte, E., Brown-Rylewski, D. E., O’Brien, T., and Leach, A. 2014. Estimated Cost of Producing Hops in Michigan [E3236]. East Lansing: Michigan State University, MSU Extension.

Skotland, C. 1961. Infection of hop crowns and roots by Pseudoperonospora humuli and its relation to crown and root rot and overwinting of the pathogen. Phytopathology 51:241- 244.

Sonoda, R., and Ogawa, J. 1972. Ecological factors limiting epidemics of hop downy mildew in arid climates. Hilgardia 41:457-473.

Strayer-Scherer, A., Jones, J., and Paret, M. 2019. Recombinase polymerase amplification assay for field detection of tomato bacterial spot pathogens. Phytopathology 109:690-700.

Williams, I. H., Roberts, J. B., and Coley-Smith, J. R. 1961. Studies of the Dormant Phase of the Hop (Humulus lupulus L.). Rep. Dept. Hop Res. Wye Coll. for 1960:48-58.

163

Woods, J. L., and Gent, D. H. 2016. Susceptibility of hop cultivars to downy mildew: associations with chemical characteristics and region of origin. Plant Health Prog. 17:42- 48.

Ye, J., Coulouris, G., Zaretskaya, I., Cutcutache, I., Rozen, S., and Madden, T. L. 2012. Primer- BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC bioinformatics 13:134.

164

CHAPTER 5. ETIOLOGY OF HALO BLIGHT IN MICHIGAN HOP YARDS

Abstract

Michigan’s hop acreage ranks fourth nationally, but the state’s growers contend with unique disease challenges resulting from frequent rainfall and high humidity. In August 2018, a

Michigan hop grower reported necrosis and blighting of foliage and shattering of cones resulting in yield loss. Irregular-shaped lesions developed on leaves, surrounded by a halo of chlorotic tissue and cone bracts became brown. Pycnidia were observed in symptomatic tissue. The goal of this study was to identify and characterize the causal agent of symptoms in leaf and cone tissue.

In symptomatic leaves, 15 of 19 isolates recovered had 96.4% internal transcribed spacer

(ITSrDNA) homology with Diaporthe nomurai. Bayesian and maximum likelihood analysis were performed on a subset of isolates using ITSrDNA, histone H3, beta-tubulin, and elongation factor one alpha. Bootstrap and posterior probabilities supported a unique cluster of Diaporthe sp. 1-MI isolates most closely related to the D. arecae species complex, D. hongkongensis and

D. multigutullata. Diaporthe sp. 1-MI was pathogenic in detached leaf and whole plant assays.

Single-spore isolates from pycnidia originating from cones and leaves shared 100% ITSrDNA homology with Diaporthe sp. 1-MI obtained from the lesion margins of leaves collected in 2018.

The distribution of Diaporthe sp. 1-MI was widespread amongst cones (n = 347) collected from

Michigan hop yards (n = 15) and accounted for > 38% of fungi recovered from cones in three hop yards. Diaporthe sp. 1-MI causing halo and cone blight presents a new disease management challenge for Michigan hop growers.

165

Introduction

Commercial hop (Humulus lupulus L.) production regions have expanded significantly in recent years. Following a worldwide hops shortage and price fluctuations in 2007 new hop yards were planted to meet local demand (Sirrine et al. 2010). Production nearly tripled from 2014

(121 ha) to 2017 (327 ha) (George 2018) with new growers located throughout the lower peninsula of Michigan. In 2019, Michigan harvested 291.4 ha (George 2020) driven by the state’s $2.5 billion craft beer industry (Brewers Association 2019).

Michigan’s growers contend with unique disease challenges resulting from frequent rainfall and high humidity. Downy mildew (caused by Pseudoperonospora humuli (Miyabe &

Takah.) G.W. Wilson, (1914) and powdery mildew (Podosphaera macularis (Wallr.) U. Braun

& S. Takam. (2000) are important diseases that limit hop production in Michigan (Del Castillo

Munera et al. 2016; Higgins and Hausbeck 2017, 2018). In August 2018, two Michigan commercial hopyards reported a high incidence of necrotic leaf lesions on ‘Chinook’,

‘Centennial’, and ‘Crystal’ (Erin Lizotte, personal communication). Post-harvest, growers noted a reduction in yield due to cone shatter in affected yards. Examination of samples revealed the presence of pycnidia in leaf lesions.

Several fungi associated with hop produce pycnidia in various tissue types or cause damage to hop cones (Bienapfl et al. 2005; Darby 1984, 1988; Gent et al. 2013; Gent and

Radišek 2009; Mahaffee and Engelhard 2009; McGee et al. 2009; Pethybridge et al. 2001a;

Pethybridge et al. 2001b; Putto et al. 1975; Radišek et al. 2008; Radišek et al. 2009; Royle and

Kremheller 1981; Singh et al. 1984; Twomey et al. 2016; Twomey et al. 2015). Didymellaceae pathogens including Septoria humuli Westend. (Putto et al. 1975), Phoma exigua var. exigua

(Desm.) Aveskamp, Gruyter & Verkley (2010) (Gent and Radišek 2009; Radišek et al. 2008),

166

and Ascochyta humuli Kabát & Bubák non Lasch (Singh et al. 1984) produce pycnidia in aboveground hop tissue. Diplodia seriata De Not. the causal agent of black wilt produces pycnidia in hop stems (Twomey et al. 2016). Phomopsis tuberivora H.T. Güssow & W.R. Foster

(analogous with Phacidiopycnis tuberivora (H.T. Güssow & W.R. Foster) B. Sutton (1980) produces pycnidia, but the pathogen appears to be limited to tissue below ground (Gent et al.

Phalip et al. 2006), they are not reported as pathogens of hop (Mahaffee et al. 2009). Cone discoloration is attributed to Alternaria alternata (Fr. : Fr.) Keissl. (1912) (Darby 1984),

Fusarium spp. (Bienapfl et al. 2005; Pethybridge et al. 2001a), P. humuli (Royle and Kremheller

1981), Botrytis cinerea Pers. : Fr. (Mahaffee and Engelhard 2009), Phaeomycocentrospora cantuariensis (E.S. Salmon & Wormald) Crous, H.D. Shin & U. Braun 2012 (analogous with

Cercospora cantuariensis (Radišek et al. 2009), P. exigua var. exigua (Radišek et al. 2008), and

P. macularis (Twomey et al. 2015).

The preliminary identification of a Diaporthe species from isolates obtained from leaf lesion margins was based on pycnidia, morphology of conidia and the ITSrDNA region (Higgins et al. 2019). Results from a detached leaf assay revealed irregular necrotic lesions with light brown pycnidia developing on all inoculated abaxial leaf surfaces (Higgins et al. 2019).

167

Diaporthe spp. are pathogenic on important horticultural crops including those belonging to the families Vitaceae (Baumgartner et al. 2013), Ericaceae (Farr et al. 2002), Rosaceae (Lalancette et al. 2003), Fabaceae (Udayanga et al. 2015), Apiaceae (Bastide et al. 2017), and Asteraceae

(Herr et al. 1983; Mathew et al. 2015). Diaporthe spp. are also saprophytes and endophytes

(Gomes et al. 2013) and since many Diaporthe spp. can colonize multiple hosts (Baumgartner et al. 2013; Mathew et al. 2015; Rehner and Uecker 1994; Udayanga et al. 2014) host association is an inadequate character for delimiting species (Rehner and Uecker 1994). Morphological species delimitation in Diaporthe spp., is limited by difficulties in inducing all spore states

(alpha, beta, and gamma conidia (asexual) and sexual states) in culture (Gomes et al. 2013;

Rehner and Uecker 1994). Instead, multi-locus phylogeny is the primary criteria for the delimitation of new species and has revealed many cryptic species (Gao et al. 2017; Gomes et al. 2013; Udayanga et al. 2012). Species delimitation for Diaporthe is generally based on molecular data including the internal transcribed spacer (ITSrDNA) and the partial histone H3 gene (HIS) or partial beta-tubulin gene (TUB) (Gomes et al. 2013; Udayanga et al. 2012) in addition to the morphological description (Seifert and Rossman 2010).

The goal of this study was to confirm and expand information regarding the causal agent of foliar blight and elucidate the etiology of cone blight on hops in Michigan. The first objective was to confirm the identity of the causal agent isolated from symptomatic leaves in 2018. A preliminary study based on the ITSrDNA region identified a Diaporthe sp.; known hence forth as Diaporthe sp. 1-MI. To test if Diaporthe sp. 1-MI may be a novel species a subset of isolates was subjected to a multilocus phylogenetic analysis and a pathogenicity study. The second objective was to determine if cone and leaf symptoms might be linked to the same causal organism. The ITSrDNA region was amplified for single-spore isolates (generated directly from

168

pycnidia) recovered from diseased leaves and cones. The third objective was to determine the distribution of cone blight among four production regions in the state.

Materials and Methods

Sampling of hop yards. Symptomatic leaves and cones were collected from commercial hop yards (n = 14) located in four production regions of Michigan’s lower peninsula including counties in the north (Grand Traverse (n = 1), Leland (n = 2), and Alpena (n = 1)), central

(Genesee (n = 1) and Monroe (n =1)), west (Kent (n = 2), Calhoun (n = 1), Allegan (n = 1), and

Kalamazoo (n = 1)), and southwest (Berrien (n = 3)). A research hopyard located in the east production region at the Michigan State University’s Plant Pathology Farm, in Lansing, MI was also sampled. Cones were collected in Aug and Sep 2019 from ten rows, one plant (4-6 cones) per row, within each hopyard 1 to 3 wk prior to harvest. Leaves were collected in Aug and Sep of 2018 and 2019. Latex gloves were used to collect samples and place them in individual resealable plastic bags; gloves were changed between plants. All samples were placed into a cooler for transport to the laboratory.

Isolation, culturing and storage of fungal species. Hyphal tip isolates were obtained from symptomatic leaves and cones in 2018 and 2019, respectively. Tissue (0.5 cm2) was excised from the leaf lesion margins, surface sterilized (20% commercial bleach) for 30 sec, rinsed twice in sterile water, dried, and plated onto a water agar to recover slow-growing fungi and a potato dextrose agar (PDA) (Difco, Detroit, MI). Whole cones (n = 347) were surface sterilized in 10% commercial bleach (8.25% sodium hypochlorite) for 5 minutes under constant agitation, rinsed three times in deionized water, and dried on autoclaved paper towels. Tissue, from the margin of necrotic area on bracts, was plated onto PDA amended with ampicillin (100

169

ug/ml). Pure cultures were obtained by subculturing onto PDA. Conidia suspensions were stored in storage medium (1/10 strength yeast extract-malt extract-dextrose; 200 ml glycerol, 800 ml distilled water, 0.4 g yeast extract, 1.0 g malt extract, 0.4 g dextrose, 0.2 g K2HPO4) at -80° C

(Miles et al. 2011).

Single spore isolates were recovered directly from pycnidia in leaf (n = 4) and cone (n =

24) lesions using methods described by Choi et al. (1999). Briefly, pycnidia were removed from leaf and cone surfaces using an Entochrysis insect pin size 00 (Pardubice, Czech Republic) fixed to a glass 15 cm Pasteur pipet (VWR, Radnor, PA). Pycnidia were placed in sterile water on a sterilized microscope slide (70% ethanol for approximately 2 min or longer). The pycnidium suspension (75 µl) was spread on PDA amended (0.5 g antibiotic/l) with ampicillin and rifampicin (Sigma Aldrich, St. Louis, MO) using a plate spreader. The following day, single germinated conidia were identified using an inverted microscope, removed with a sterile hypodermic disposable needle (size 30G; Exelint International, Co., Redondo Beach, CA) and placed on PDA. A sterile alfalfa stem was added to single spore colonies growing on PDA to induce pycnidia formation (Farr et al. 2002).

DNA extraction, PCR amplification, and sequencing. DNA was extracted from leaf and pycnidia isolates using a DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) or

CTAB/phenol:chloroform (Sambrook et al. 1989). Mycelia (<100 mg) were scraped from the agar surface of up to 30-day old cultures and physically lysed using lysing matrix A (MP

Biomedicals North America, Solon, OH) and a Fast Prep tissue disruptor (MP Biomedials North

America, Solon, OH) for 60 seconds (40 M/S), or in a 1.7 ml centrifuge tube containing five glass beads (3 mm; MilliporeSigma, Burlington, MA) and 800 µl of CTAB with a TissueLyser II

(Qiagen) three times at 0:30 sec intervals (30 hz). The DNeasy Plant Mini Kit instructions were

170

followed, except 195 µl of Buffer P3, 600 µl AP1 buffer and 6 µl RNase A was used. The

CTAB/phenol/chloroform extraction procedure was adapted from Sambrook et al. (1989); DNA was precipitated in cold isopropanol at -20°C overnight.

For isolates obtained from cones, mycelia (10 mg) were placed into a 1.1 ml DNA extraction tube with a metal bead (DOT Scientific, Burton, MI). Mycelia were lyophilized for 24 h and pulverized for 15 sec at 30 Hz with a mixer mill mm 400 (Retsch, Haan, Germany). DNA was extracted using the MagMAX plant DNA kit (Applied Biosystems, Foster City, CA) and the

Kingfisher Flex (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s instruction.

For isolates obtained from leaves and directly from pycnidia in lesions of leaf and cone tissue, the internal transcribed spacer (ITSrDNA) was amplified with primers ITS1/ITS4 (White et al. 1990). Primers were synthesized by Integrated DNA Technologies, Inc. (Coralville, IA).

Polymerase chain reaction (PCR) was conducted with 1.25 µl forward primer (10 µM), 1.25 µl reverse primer (10 µM), 0.5 µl DNTPs (10 mM), 5.0 µl colorless buffer (Promega, Madison,

WI), 0.25 µl GoTaq DNA polymerase (Promega), DNase-free water, and template DNA (4-10 ng) in a 25-ml reaction. Amplifications were performed on a Mastercycler Pro (Eppendorf,

Westbury, NY) thermal cycler with initial denaturation at 95°C for 2 min, followed by 35 cycles of 94°C for 1 min, 52°C for 1 min, and 72°C for 1 min, and a final extension at 72°C for 5 min.

PCR products were then visualized on a 1.0% agarose TAE gel. Three additional nuclear DNA loci were amplified and sequenced for six Michigan isolates obtained from leaves (MI_0318,

MI_1318, MI_1418, MI _2018, MI_2218 and MI _2318). Amplification was performed with primers T1/Bt-2b (Glass and Donaldson 1995; O’Donnell and Cigelnik 1997) for β- tubulin

(TUB), CYLH3F/H3-1b (Crous et al. 2004; Glass and Donaldson 1995) for histone (HIS), and

171

EF1-728F/EF1-986R (Carbone and Kohn 1999) for the elongation factor 1 alpha (TEF1). PCR conditions were as described above for isolates obtained from leaf lesions (2018) and pycnidia

(2019). Annealing temperatures were adjusted to 58°C (HIS) and 55°C (TEF and TUB).

Isolates from cones obtained in 2019 (Table 3) were either identified morphically and/or through sequencing of the ITSrDNA region. Fungal structures were identified according to morphological characteristics (Barnett and Hunter 1998). Where identification was not possible by distinct morphology, an expanded region of ITSrDNA locus (Raja et al. 2017) was sequenced with ITS1F (Gardes and Bruns 1993) and ITS4 (White et al. 1990). Reactions were conducted in

25 µl volumes using 0.1 µl of Platinum Taq DNA Polymerase High Fidelity (Invitrogen,

Carlsbad, CA), 0.1 µl of each 10 µM primer, 1 µl of 50 nM MgSO4 (Invitrogen), 0.5 µl of 10 mM dNTPs (Thermo Fisher Scientific), 2.5 µl of 10x HiFi buffer (Invitrogen), 17.9 µl of water, and 2 µl of 2-10 ng/µl DNA template per reaction. Amplifications were performed with an initial denaturation of 94°C for two minutes, 35 cycles of 94°C for 30 sec, 52°C for 30 sec, and 72°C for 1 min, followed by 72°C for 5 min, and a hold at 4°C.

All amplicons were purified using Qiagen PCR purification kit (Qiagen) following manufacturer’s protocol and submitted for Sanger sequencing at the MSU RTSF Genomics Core

(East Lansing, MI). Sequences obtained from all primer sets were aligned using M-Coffee

(Moretti et al. 2007; Wallace et al. 2006), and manually edited in Geneious Prime (ver. 10.2.3,

Biomatters Inc., Auckland, NZ).

Preliminary GenBank BLAST search identification of isolates. Sequence similarity was determined using a standard nucleotide BLAST (BLASTn) search

(https://blast.ncbi.nlm.nih.gov) with default parameters. The BLAST search results for the

ITSrDNA locus of all leaf lesion margin isolates (2018) were used to assign preliminary

172

identification. The BLAST search results for four unaligned loci (ITSrDNA, HIS, TUB, and

TEF1) of six Diaporthe sp. 1-MI isolates (MI_0318, MI_1318, MI_1418, MI _2018, MI_2218 and MI _2318) were used with published phylogenies (Gao et al. 2017; Gomes et al. 2013) to select the taxonomic framework for the phylogenetic analysis; BLAST search results from aligned phylogenetic sequences are presented.

Phylogenetic analyses. Phylogenetic relationships were examined with the four individual loci (ITSrDNA, HIS, TUB, and TEF1) and a combined multi-locus sequence (MLS) analysis. Sequences from 19 Diaporthe spp. (15 ex-type, ex-isotype, and ex-epitype) obtained from GenBank (Du et al. 2016; Gao et al. 2016; Gao et al. 2017; Gomes et al. 2013; Huang et al. 2015; Tan et al. 2013; Thompson et al. 2015; Udayanga et al. 2014) and six Diaporthe sp. 1-

MI isolates (MI_0318, MI_1318, MI_1418, MI _2018, MI_2218 and MI _2318) (Table 4.1) were aligned with M-Coffee and manually adjusted (Hall 2011) in Geneious Prime. Diaporthe amygdali CBS 126679 was used as the outgroup. Trees were visualized using FigTree (v1.4.4).

The maximum likelihood tree is presented with ML bootstrap support values ≥ 70 and Bayesian posterior probability values ≥ 0.75.

Model selection for the Bayesian analyses (BA) was determined using jModelTest2

(Darriba et al. 2012; Guindon and Gascuel 2003) on the CIPRES Science Gateway (Miller et al.

2010). The Bayesian analysis was preformed using MrBayes (Ronquist et al. 2012). The data were partitioned with the following evolutionary models applied to each locus in the concatenated alignment: GTR + I + G (ITS), GTR + G (HIS and TUB), and HKY I + G (TEF1).

The Markov Chain Monte Carlo (MCMC) analysis of four chains started from a random tree topology and lasted 1,000,000 generations. Trees were saved every 100 generations. Burn-in was

173

set at 25% and the 50% majority rule consensus tree and posterior probabilities were calculated from the remaining trees.

The Maximum Likelihood (ML) analyses was performed using RAxML (Stamatakis

2014) on the CIPRES Science Gateway (Miller et al. 2010). The evolutionary model GTR +

GAMMA was used to estimate best scoring ML trees. 1000 bootstrap replicates were generated using the rapid bootstrapping algorithm in a single run. There was no correction for ascertainment bias.

Pathogenicity tests. For 19 isolates (15 Diaporthe sp., 3 Alternaria sp., and 1

Peyronellaea prosopidis) recovered from leaves in 2018, a detached leaf assay was used to test pathogenicity. Leaves obtained from five-month old greenhouse-grown ‘Cascade’ plantlets

(nodes three to five) were placed either adaxial or abaxial surface up in petri dishes (100 x 15 mm) with sterile filter paper (Whatman 70 mm) and a moistened cotton ball. Conidia from 21- day old cultures were harvested and inoculum containing Tween 20 (1%) was prepared at 1x104 or 1x105 conidia/ml. Inoculum or water was applied to leaves (three replicate leaves per treatment) with a handheld sprayer. Leaves were arranged in a complete randomized design

(CRD) and incubated at 20C with a 14-hr photoperiod. Plants were rated for disease severity (% necrotic tissue) on a 0-100% scale. Re-isolation was attempted from tissue at the lesion edge on all symptomatic leaves using methods described above. Diaporthe isolates were identified based on morphology and sequence homology.

174

Table 4.1. Diaporthe spp. and outgroup D. amygdali strains used to determine the phylogenetic relationship amongst Diaporthe sp. 1- MI isolates recovered from hop leaves in Michigan. Species Strainy GenBank Accession Numberz Reference ITS TEF1 TUB HIS Diaporthe sp. 1 MI_0318 MT909094 MT909100 MT909106 MT909088 This study Diaporthe sp. 1 MI_1318 MT909095 MT909101 MT909107 MT909089 This study Diaporthe sp. 1 MI_1418 MT909096 MT909102 MT909108 MT909090 This study Diaporthe sp. 1 MI_2018 MT909097 MT909103 MT909109 MT909091 This study Diaporthe sp. 1 MI_2218 MT909098 MT909104 MT909110 MT909092 This study Diaporthe sp. 1 MI_2318 MT909099 MT909105 MT909111 MT909093 This study D. betulae CFCC 50469 KT732950 KT733016 KT733020 KT732999 Du et al. 2016 D. vaccinii CBS 160.32 KC343228 KC343954 KC344196 KC343712 Huang et al. 2015 D. alleghaniensis CBS 495.72 KC343007 KC343733 KC343975 KC343491 Gomes et al. 2013 D. alnea CBS 146.46 KC343008 KC343734 KC343976 KC343492 Gomes et al. 2013 D. neilliae CBS 144.27 KC343144 KC343870 KC344112 KC343628 Udayanga et al. 2014 D. virgiliae CMW 40755 KP247573 - KP247582 - Gao et al. 2017 D. citrichinensis ZJUD96 KJ490631 KJ490510 KJ490452 KJ490573 Huang et al. 2015 D. oraccinii LC3166 KP267863 KP267937 KP293443 KP293517 Gao et al. 2016 D. charlesworthii BRIP 54884m KJ197288 KJ197250 KJ197268 - Thompson et al. 2015 D. nomurai CBS 157.29 KC343154 KC343880 KC344122 KC343638 Gomes et al. 2013 D. multigutullata ZJUD98 KJ490633 KJ490512 KJ490454 KJ490575 Huang et al. 2015 D. pseudophoenicicola CBS 462.69 KC343184 KC343910 KC344152 KC343668 Gomes et al. 2013 D. hongkongensis CBS 115448 KC343119 KC343845 KC344087 KC343603 Gomes et al. 2013 D. arecae CBS 161.64 KC343032 KC343758 KC344000 KC343516 Gomes et al. 2013 D. pseudomangiferae CBS 101339 KC343181 KC343907 KC344149 KC343665 Gomes et al. 2013 D. fraxini-angustifolia BRIP 54781 JX862528 JX862534 KF170920 - Tan et al. 2013 D. pascoei BRIP 54847 JX862538 JX862538 KF170924 - Tan et al. 2013 D. arengae CBS 114979 KC343760 KC343760 KC344002 KC343518 Gomes et al. 2013 D. amygdali CBS 126679 KC343022 KC343748 KC343990 KC343506 Gomes et al. 2013 y CBS = Centraalbureau voor Schimmelcultures, Fungal Biodiversity Centre, Utrecht, The Netherlands. CFCC = China Forestry Culture Collection Center, Chinese Academy of Forestry, Beijing, China. CWM= Collection of the Forestry and Agricultural Biotechnology Institute (FABI), University of Pretoria, South Africa. ZJUD = Zhe Jiang University, China. BRIP = Queensland Plant Pathology Herbarium, Australia. LC: Working collection of Lei Cai, housed at Institute of Microbiology, CAS, China. Strains in bold represent ex-type, ex-isotype, and ex-epitype. z TUB = partial beta-tubulin gene; HIS = partial histone H3 gene; ITSrDNA = internal transcribed spacer region; TEF1 = partial translation elongation factor 1-alpha gene.

175

Pathogenicity was tested using Diaporthe sp. 1-MI (MI_0318) on whole plants.

Greenhouse-grown seven-month old ‘Cascade’ plantlets were sprayed with MI_0318 (1x105 conidia/ml) or water. Plants were arranged in a CRD with five replicates per treatment and incubated at 100% relative humidity at 20C with a 14-hr photoperiod. Plants were rated for disease severity (necrotic foliage) on a 0-100% scale and incidence (% leaves with necrosis). Re- isolation was attempted from tissue at the lesion edge on all symptomatic leaves using the methods described above.

In vivo morphology. Pycnidia from leaf (n = 20) and cone tissue (n = 20) were removed as previously described and mounted in lactic acid. Conidia from leaf (n = 50) and cone (n = 50) tissue were measured using a Lecia DM750 microscope and Leica Application Suite X software

(Lecia Microsystems, Vienna, Austria). Additionally, leaves were scanned using a digital scanner (Epson, Suwa, Japan) and lesion size (n = 30) determined using Image J (Rasband 1997).

Leaf and cone images were also collected using a Keyence VHX-6000 (Keyence Corp., Osaka,

Japan) digital microscope.

Symptomatic leaf and cone tissue with pycnidia were examined using scanning electron microscopy. Initially, tissue was fixed at 4°C for 1-2 h in 4% glutaraldehyde buffered with 0.1M sodium phosphate at pH 7.4. Following a brief rinse in the buffer, samples were dehydrated in an ethanol series (25%, 50%, 75%, 95%) for 10-15 min at each gradation and with three 10 min changes in 100% ethanol. Samples were critical point dried in a Lecia Microsystems model EM

CPD300 critical point dryer using carbon dioxide as the transitional fluid. Tissue was hand- sectioned from slices after critical-point drying to reduce potential artifacts from the fixation process. Samples were mounted on aluminum stubs using adhesive tabs (M.E. Taylor

Engineering, Brookville, MD). Samples were coated with iridium. Samples were examined in a

176

JEOL 6610LV (tungsten hairpin emitter) scanning electron microscope (SEM) (JEOL Ltd.,

Tokyo, Japan). Observations were made on transverse cross sections of pycnidia in leaf (n = 10) and cone (n = 5) tissue.

Colony morphology and growth. Measurements of pycnidia (n = 10) and conidia (n =

50) were taken from four leaf isolates (MI_0118, MI_0318, MI_1218, MI_2518) and five cone isolate (MI_2219, MI _4119, MI _2719, MI_3019, MI_3119). Isolates were incubated at room temperature for 30+ days on PDA with and without alfalfa stems (Farr et al. 2002). Pycnidia and conidia were examined using a Lecia DM750 microscope and Leica Application Suite X software. Colony color of Diaporthe sp. 1-MI isolates on PDA was determined with the Munsell color chip system (Munsell Color, Grand Rapids, MI).

The effect of temperature on mycelium growth was characterized for three isolates of

Diaporthe sp. 1-MI (MI_0318, MI_1418, and MI_2518). Isolates were transferred from storage medium to full strength PDA in a 100 x 15m petri dish (VWR). After one week of growth at room temperature, a 5.0 mm plug of mycelium was transferred to the center of a petri dish containing PDA. Five replicates of each isolate were then grown at 10, 15, 20, 25, and 30° C.

Colony diameter was measured every two days using a digimatic caliper (Mitutoyo, Sakado,

Japan) until one isolate had a diameter of 80 mm. Growth rate over time was plotted for each isolate. Growth curve data were fitted with a polynomial inverse first order equation using the

GLOBAL CURVE FIT WIZARD option in SigmaPlot (ver 11.2, Systat Software, Inc. San Jose,

CA).

177

Results

Field symptoms and in vivo morphology. Foliar symptoms were sporadically distributed throughout the plant canopy (observations made from 1.7 m to ground level), affecting approximately 20% of leaves, included irregularly-shaped necrotic lesions surrounded by chlorotic tissue (Figure 4.1A-D). Pycnidia were observed on necrotic tissue on the adaxial surfaces of leaves (Figure 4.1E, F). Pycnidia were singular or aggregated and immersed in irregularly-shaped necrotic lesions (13.4 cm2 ± 13.4 S.D. (n=30). Pycnidia were brown to dark brown (Figure 4.1E, F), ampulliform, eustroma (Figure 4.3A, B), and with a singular ostiole

(Figure. 4.3A-C). Pycnidia were immersed, rupturing adaxial leaf epidermis with coalescent alpha conidia and occasionally light tan cirrhi (Figure 4.1E, 3G). Alpha conidia were unicellular, hyaline and cylindrical to ellipsoid (12.1 µm ± 2.1 s.d. by 3.8 µm ± 0.4 s.d. (n = 50))

(Table 4.2). No beta conidia or perithecia were observed.

Losses associated with cone shatter at harvest were estimated between 17% (‘Cascade’) and 56% (‘Chinook’) based of the average yield in previous years in two affected yards (data not shown). On cones, affected bracts displayed varying severity and patterns of brown discoloration

(Figure 4.2A-C). Pycnidia observed on cone bracts (Figure 4.2D-F) were singular or aggregated and immersed in necrotic bract tissue. Pycnidia were light brown to brown to dark brown

(Figure 4.2D-F), ampulliform to oblate with a short stout neck, large singular ostiole, eustroma, and occasionally imbedded in textura angularis (Figure 4.3D-F). Pycnidia ruptured the adaxial cone epidermis (Figure 4.3E) and produced coalescent alpha conidia (Figure 4.2F, 3H); no cirrhi were observed. Alpha conidia (in vivo) obtained from adaxial cone pycnidia were unicellular, hyaline and cylindrical to ellipsoid (11.9 µm ± 2.0 s.d. by 4.1 µm ± 0.5 s.d. (n = 50))

(Table 4.2); no beta conidia were observed.

178

179

Figure. 4.2. Signs and symptoms of a Diaporthe sp. 1-MI on hop cones. A,B) Dried necrotic tissue where Diaporthe sp. 1-MI has been isolated. C) A tiled image of dried bract tissue infected with Diaporthe sp. 1-MI. D,E,F) Pycnidia and pycnidial ooze observed on bracts with a dissecting microscope. G, H) Diaporthe sp. 1-MI growth and conidia morphology on PDA (27 days).

180

181

Table 4.2. Conidial size of a Diaporthe sp. 1-MI recovered from hop cone and leaf lesions in Michigan, and previously reported pycnidia producing pathogens of hop. Species Mean conidia size (μm)y Tissue (media)z Source Diaporthe sp. 1-MI 12.1 ± 2.1 × 3.8 ± 0.4 Leaf (in vivo) This study Diaporthe sp. 1-MI 11.9 ± 2.0 × 4.1 ± 0.5 Cone (in vivo) This study Diaporthe sp. 1-MI 8.4 ± 1.5 × 3.4 ± 0.5 Leaf (PDA) This study Diaporthe sp. 1-MI 10.2 ± 1.1 × 3.5 ± 0.5 Cone (PDA) This study Phoma exigua var. exigua 4.5-5.5 × 1.5-2.5 NA Gent and Radišek 2009 Phoma exigua var. exigua (4.5)-6.5-(8) × (2)-2.5-(4) Leaf, Cone (Var.) Radišek et al. 2008, 2009 Phomopsis tuberivora 8-12 × 4-6 Root (NA) Gent et al. 2013 Phoma sarmentella 5-6 x 2-3 Bine Grove, 1917 y Standard deviation is indicated for the Diaporthe sp. 1-MI measurements from this study. z PDA = potato dextrose agar; OA = oatmeal agar; Var. = OA, malt extract agar, and cherry decoction agar; NA = information not provided.

182

Preliminary GenBank BLAST Search identification. Hyphal-tip isolates were obtained from the margin of leaf lesions in 2018 (Table 4.3). The ITSrDNA region was amplified for 19 of these isolates, 15 were partially homologous to Diaporthe nomurai (96.6% pairwise identity to AB302238 and KC343154.1), three matched an Alternaria sp. (100% pairwise identity to MK224474) and one matched Peyronellaea prosopidis (100% pairwise identity to NR_137836).

The pairwise identities of six Diaporthe sp. 1-MI isolates (MI_0318, MI_1318, MI_1418,

MI _2018, MI_2218 and MI _2318) from individual loci of aligned phylogenetic sequences overall were low and included a 96.4% match with the ITSrDNA region of D. nomurai CBS

157.29 (KC343154; (Gomes et al. 2013), 91.5% match with the HIS partial gene region of D. citrichinensis ZJUD 38 (KJ490517; (Huang et al. 2015), 78.7 to 82.1% match with the TEF partial gene region of D. multigutullata ZJUD 98 (KJ490512; (Huang et al. 2015), and 90.1,

89.6, 89.1, 89.4, and 89.0% match with the TUB partial gene region of D. acerina (KC343974.1;

(Gomes et al. 2013), D. tibetensis (MF279873; (Fan et al. 2018), D. virgiliae CMW 40756

(KP247583; (Machingambi et al. 2015), D. hickoriae (KC244086.1; Gomes et al. 2013), and D. vawdreyi BRIP 57887 (KR936128; (Crous et al. 2015), respectively. Sequence from the four amplified gene regions for each of the six Diaporthe sp. 1-MI isolates was deposited to GenBank

(Table 4.1).

183

Table 4.3. Diaporthe sp. 1-MI obtained from symptomatic leaf margins and directly from pycnidia in symptomatic hop leaf and cone tissue and identified by the internal transcribed spacer (ITS). Isolatex Isolationy Tissue Samplez Cultivar Year Yard MI_0118 lesion margin leaf n/a Chinook or Centennial 2018 MI-1 MI_0318 lesion margin leaf n/a Chinook or Centennial 2018 MI-1 MI_1118 lesion margin leaf n/a Chinook or Centennial 2018 MI-1 MI_1218 lesion margin leaf n/a Chinook or Centennial 2018 MI-1 MI_1318 lesion margin leaf n/a Chinook or Centennial 2018 MI-1 MI_1418 lesion margin leaf n/a Crystal 2018 MI-2 MI _1518 lesion margin leaf n/a Crystal 2018 MI-2 MI _1618 lesion margin leaf n/a Crystal 2018 MI-2 MI _1718 lesion margin leaf n/a Crystal 2018 MI-2 MI _1818 lesion margin leaf n/a Crystal 2018 MI-2 MI _2018 lesion margin leaf n/a Crystal 2018 MI-2 MI _2118 lesion margin leaf n/a Crystal 2018 MI-2 MI _2218 lesion margin leaf n/a Crystal 2018 MI-2 MI _2318 lesion margin leaf n/a Crystal 2018 MI-2 MI _2418 lesion margin leaf n/a Crystal 2018 MI-2 MI _1719 pycnidia leaf 5 Chinook 2019 MI-1 MI_1819 pycnidia leaf 6 Chinook 2019 MI-1 MI_3219 pycnidia leaf 7 Chinook 2019 MI-1 MI_3319 pycnidia leaf 6 Chinook 2019 MI-1 MI _3419 pycnidia leaf 6 Chinook 2019 MI-1 MI_5219 pycnidia leaf 1 Chinook 2019 MI-1 MI _0519 pycnidia leaf 8 Chinook 2019 MI-1 MI _4019 pycnidia leaf 8 Chinook 2019 MI-1 MI _0119 pycnidia cone 1 Centennial 2019 MI-1 MI _4419 pycnidia cone 1 Centennial 2019 MI-1 MI_0919 pycnidia cone 2 Centennial 2019 MI-1 MI _1519 pycnidia cone 2 Centennial 2019 MI-1 MI _4319 pycnidia cone 2 Centennial 2019 MI-1 MI _0319 pycnidia cone 3 Centennial 2019 MI-1 MI _0619 pycnidia cone 3 Centennial 2019 MI-1 MI _0219 pycnidia cone 4 Centennial 2019 MI-1 MI _1219 pycnidia cone 4 Centennial 2019 MI-1 MI _3019 pycnidia cone 5 Centennial 2019 MI-1 MI _3119 pycnidia cone 5 Centennial 2019 MI-1 MI_5319 pycnidia cone 5 Centennial 2019 MI-1 MI _4219 pycnidia cone 6 Centennial 2019 MI-1 MI_0419 pycnidia cone 8 Centennial 2019 MI-1 MI _3319 pycnidia cone 9 Centennial 2019 MI-1 MI_2119 pycnidia cone 10 Centennial 2019 MI-1 MI_2619 pycnidia cone 11 Centennial 2019 MI-1

184

Table 4.3. (cont’d). Isolatex Isolationy Tissue Samplez Cultivar Year Yard MI_2019 pycnidia cone 13 Centennial 2019 MI-1 MI _3619 pycnidia cone 15 Centennial 2019 MI-1 MI _4719 pycnidia cone 16 Centennial 2019 MI-1 MI _0819 pycnidia cone 17 Centennial 2019 MI-1 MI _4619 pycnidia cone 18 Centennial 2019 MI-1 MI _1919 pycnidia cone 19 Centennial 2019 MI-1 MI _2719 pycnidia cone 19 Centennial 2019 MI-1 MI _4919 pycnidia cone 21 Centennial 2019 MI-1 MI _1119 pycnidia cone 22 Centennial 2019 MI-1 MI _3719 pycnidia cone 22 Centennial 2019 MI-1 MI _2919 pycnidia cone 23 Centennial 2019 MI-1 MI _4519 pycnidia cone 24 Centennial 2019 MI-1 MI _4819 pycnidia cone 24 Centennial 2019 MI-1 MI_5119 pycnidia cone 5A Centennial or Cascade 2019 MI-3 MI _3919 pycnidia cone 6A-1 Centennial or Cascade 2019 MI-3 MI_2519 pycnidia cone 6A-2 Centennial or Cascade 2019 MI-3 MI _3819 pycnidia cone 6A-2 Centennial or Cascade 2019 MI-3 MI_2219 pycnidia cone 7B Centennial or Cascade 2019 MI-3 MI _4119 pycnidia cone 8A Centennial or Cascade 2019 MI-3 MI _0719 pycnidia cone 9A Centennial or Cascade 2019 MI-3 MI_2419 pycnidia cone 9A Centennial or Cascade 2019 MI-3 x All isolates identified as Diaporthe sp. 1-MI by ITSrDNA with one exception; MI_2518 was identified by partial translation elongation factor 1-alpha (TEF1) and histone H3 (HIS) loci. MI_2518 was not used in pathogenicity testing. y Lesion Margin = A hyphal-tip isolate generated from the lesion margin of symptomatic leaf tissue. Pycnidia = A single-spore isolates generated directly from pycnidia in leaf and cone tissue. Z Isolates with the same sample code were generated from a single leaf or code sample. N/A = information not available.

185

Multi-locus phylogenetic analyses. The alignment of Diaporthe spp. and six Diaporthe sp. 1-MI isolates (MI_0318, MI_1318, MI_1418, MI _2018, MI_2218 and MI _2318) used for phylogenetic analysis contained 2066 characters (469-HIS, 481-ITSrDNA, 394-TEF1, and 722-

TUB).

In the combined locus analysis, BA and ML analyses produced near identical tree topologies (Figure 4.4). Topologies, bootstrap and posterior probabilities for Diaporthe species were in agreement with other reported phylogenies (Gomes et al. 2013; Gao et al. 2017, Huang et al. 2015). Diaporthe sp. 1-MI isolates clustered together in a single supported clade (posterior probability = 1; bootstrap = 100) within a larger clade (posterior probability = 1; bootstrap = 90) containing with D. multigutullata (ZJUD98), D. hongkongensis (CBS 115448), and the D. arecae species complex.

For each individual locus (Figures A5-8), the Diaporthe sp. 1-MI isolates clustered together with high bootstrap (100) and posterior probability (≥ 0.96) support. The TUB locus produced a tree topology most similar to the tree of combined locus analysis. Variation within the Diaporthe sp. 1-MI clade was only observed in the TEF1 locus which separated MI_2318,

MI_2018, and MI_2218 from MI_1418, MI_1318 and MI_0318 (posterior probability = 0.97; bootstrap = 98). There was relatively low support (posterior probability = 0.80; bootstrap < 70) for placing Diaporthe sp. 1-MI with another Diaporthe clade using the TEF1 locus. Instead, the

Diaporthe sp. 1-MI isolates clustered together in their own clade (posterior probability = 1; bootstrap = 100) and separate from the outgroup. Placement of Diaporthe sp. 1-MI with other

Diaporthe clades varied. The ITSrDNA locus topology placed Diaporthe sp. 1-MI in a well- supported clade (posterior probability = 0.96; bootstrap = 87) with D. nomurai. The TUB locus placed Diaporthe sp. 1-MI in a well-supported clade (posterior probability = 0.99; bootstrap =

186

94) with D. multigutullata, D. hongkongensis and members of the D. arecae species complex (D. arecae (CBS 161.64), D. pseudophoenicicola (CBS 462.69), D. pascoei (BRIP 54847), D. fraxini-angustifolia (BRIP 54781), D. pseudomangiferae (CBS 101339), and D. arengae (CBS

114979). The HIS locus placed Diaporthe sp. 1-MI in a clade with D. neilliae and D. alnea, but support for this clade was low (posterior probability < 0.75; bootstrap = 72).

187

Identification of single-spore isolates from pycnidia in cone and leaf tissue. In 2019, single-spore isolates were obtained from pycnidia found in the lesions of 25 cones and 4 leaves

(Figure 4.5; Table 4.3). ITSrDNA homology (100%) was observed amongst all isolates (n = 46) obtained from pycnidia. The ITSrDNA region of pycnidia isolates was homologous (100%; approximately 453 alignment positions) with Diaporthe sp. 1-MI isolated from the lesion margin of leaf tissue obtained in 2018.

Figure 4.5. Symptoms of cone blight on ‘Centennial’ (A-K) and ‘Centennial’ or ‘Cascade’ (G- F) hop cones. Single-spore isolates of Diaporthe 1-MI were derived directly from the pycnidia in blighted tissue of 25 cone samples. The following Diaporthe sp. 1-MI were recovered from the cones are pictured: (A) MI_119, MI_3719; (B) MI_0219, 1219; (C) MI_3019, MI_3119, MI_5319; (D) MI_0819; (E) MI_4619; (F) MI_4719; (G) MI_0719, MI_2419; (H) MI_4119; (I) MI_3919; (J) MI_3819, MI_2519; (K) MI_4119.

Distribution and identity of fungal species isolated from cones in hop yards.

Diaporthe sp. 1-MI isolated from lesion margins in cones was identified by ITSrDNA (n = 45) and matching colony morphology (n = 37) (Table 4.4; 4.5). Diaporthe sp. 1-MI was recovered alone and with other fungi (co-recovery) from cones in all four hop production regions in

Michigan (Figure 4.6A, B; Table 4.5). Diaporthe sp. 1-MI was recovered from 16.4% (n = 57) of cones (n = 347). In the northern region, Diaporthe sp. 1-MI was recovered from 5.5% (n = 3)

188

of the cones (n = 55) at four hopyards. The north region was the only region where Diaporthe sp.

1-MI was not recovered from all sampling sites. Diaporthe sp. 1-MI was the most abundant fungal species (n = 21) recovered from cones (n = 74) in the eastern region. Diaporthe sp. 1-MI was recovered from 14.1 (n = 25) and 19.5% (n = 8) of symptomatic cones collected in the west

(n = 177) and southwest (n = 41) regions, respectively. In hop yards MI-8 (n = 15 cones), MI-4

(n = 18 cones), and MI-3 (n = 39 cones) Diaporthe sp. 1-MI was recovered from 53.3 (n = 8),

50.0 (n = 9) , 38.5% (n = 15) of symptomatic cones, respectively. Diaporthe sp. 1-MI (n = 25) was co-recovered (more than one fungal isolate was recovered from a single cone) from 30% (n

= 21) of cones (n = 70).

Additional fungi isolated from lesion margins in cones were identified to genus by

ITSrDNA as Alternaria spp. (n = 119), Fusarium spp. (n = 26), and Epicoccum spp. (n = 17) and by matching colony morphology as Alternaria spp. (n = 76), Fusarium spp. (n = 31), and

Epicoccum spp. (n = 28). Some isolates contained unique phenotypes and were not classified

(Other; n = 21). These fungi were recovered alone and co-recovered from cones in all four hop production regions in Michigan (Figure 4.6A, B). Alternaria spp., Fusarium spp., and

Epicoccum spp. were recovered from 40.9 (n = 142), 9.8 (n = 34), and 6.3% (n = 22) of cones

(n=347), respectively. Fusarium spp. were recovered from 14.6% (n = 6) of cones in the southwest region (n = 41) and 10% or less of cones in other regions. Alternaria spp. were recovered from 56.4 (n = 31), 48.0 (n = 85), 25.7 (n = 19), and 17.1% (n = 7) of cones collected in the north (n = 74), west (n = 177), east (n = 74) and southwest (n = 41) regions, respectively.

Alternaria spp. (n = 53) were co-recovered from 68.5% (n = 48) of cones (n = 70).

189

Table 4.4. Diaporthe sp. 1-MI isolates obtained from symptomatic cones in Michigan hopyards in four production regions during a 2019 sampling to determine pathogen distribution. Regionx Isolate Isolationy Tissue Sample ITSz Yard East MI19S_01 lesion margin cone 7a-d ITS MI-3 East MI19S_02 lesion margin cone 4b-d ITS MI-3 East MI19S_03 lesion margin cone 9a-d ITS MI-3 East MI19S_04 lesion margin cone 9a-c ITS MI-3 East MI19S_05 lesion margin cone 6a-c ITS MI-3 East MI19S_06 lesion margin cone 5a-b ITS MI-3 East MI19S_07 lesion margin cone 5a-a ITS MI-3 East MI19S_08 lesion margin cone 3b-a M MI-3 East MI19S_09 lesion margin cone 5b-a M MI-3 East MI19S_10 lesion margin cone 6b-b M MI-3 East MI19S_11 lesion margin cone 8b-c M MI-3 East MI19S_12 lesion margin cone 10a-c M MI-3 East MI19S_13 lesion margin cone 10a-d M MI-3 East MI19S_14 lesion margin cone 6a-b M MI-3 East MI19S_15 lesion margin cone 7b-c M MI-3 East MI19S_16 lesion margin cone 8b-a M MI-3 East MI19S_17 lesion margin cone 9a-b M MI-3 East MI19S_18 lesion margin cone 9b-a M MI-3 East MI19S_19 lesion margin cone 9b-d M MI-3 East MI19S_20 lesion margin cone 10a-b M MI-3 East MI19S_21 lesion margin cone 1b-b ITS MI-9 East MI19S_22 lesion margin cone 5b-b ITS MI-9 East MI19S_23 lesion margin cone 8c-b ITS MI-9 East MI19S_24 lesion margin cone 13c-a ITS MI-9 East MI19S_25 lesion margin cone 3b-b ITS MI-9 East MI19S_26 lesion margin cone 7a-a ITS MI-11 East MI19S_27 lesion margin cone 10a-a ITS MI-11 East MI19S_28 lesion margin cone 1c-b M MI-11 East MI19S_29 lesion margin cone 6c-a M MI-11 East MI19S_30 lesion margin cone 4d-a M MI-11 West MI19S_31 lesion margin cone 2c-a ITS MI-4 West MI19S_32 lesion margin cone 1c-a ITS MI-4 West MI19S_33 lesion margin cone 6c-b ITS MI-4 West MI19S_34 lesion margin cone 3b-a ITS MI-4 West MI19S_35 lesion margin cone 7c-b ITS MI-4 West MI19S_36 lesion margin cone 7c-b M MI-4 West MI19S_37 lesion margin cone 3c-b M MI-4 West MI19S_38 lesion margin cone 10b-a M MI-4 West MI19S_39 lesion margin cone 17b-c M MI-4 West MI19S_40 lesion margin cone 17c-a M MI-4 West MI19S_41 lesion margin cone 10b-a ITS MI-5 West MI19S_42 lesion margin cone 10c-a ITS MI-5

190

Table 4.4. (cont’d). Regionx Isolate Isolationy Tissue Sample ITSz Yard West MI19S_43 lesion margin cone 3d-a ITS MI-7 West MI19S_44 lesion margin cone 10d-c ITS MI-7 West MI19S_45 lesion margin cone 2c-a ITS MI-7 West MI19S_46 lesion margin cone 2d d ITS MI-7 West MI19S_47 lesion margin cone 8d b ITS MI-7 West MI19S_48 lesion margin cone 3c-a ITS MI-7 West MI19S_49 lesion margin cone 7c-a ITS MI-7 West MI19S_50 lesion margin cone 7c-b ITS MI-7 West MI19S_51 lesion margin cone 6b-cb M MI-7 West MI19S_52 lesion margin cone 3c-c M MI-7 West MI19S_53 lesion margin cone 7c-c M MI-7 West MI19S_54 lesion margin cone 8c-d M MI-7 West MI19S_55 lesion margin cone 3a-a M MI-7 West MI19S_56 lesion margin cone 2a-b M MI-7 West MI19S_57 lesion margin cone 3d-d M MI-7 West MI19S_58 lesion margin cone 9c-d M MI-7 West MI19S_59 lesion margin cone 8c-b M MI-7 West MI19S_60 lesion margin cone 5d-b ITS MI-10 West MI19S_61 lesion margin cone 4c-a ITS MI-10 West MI19S_75 lesion margin cone 10b-a ITS MI-16 West MI19S_76 lesion margin cone 1c-b M MI-16 West MI19S_77 lesion margin cone 9a-b M MI-16 Southwest MI19S_62 lesion margin cone 7c-a ITS MI-8 Southwest MI19S_63 lesion margin cone 1a-b ITS MI-8 Southwest MI19S_64 lesion margin cone 3a-a M MI-8 Southwest MI19S_65 lesion margin cone 1b-a M MI-8 Southwest MI19S_66 lesion margin cone 1c-b ITS MI-8 Southwest MI19S_67 lesion margin cone 11a-a M MI-8 Southwest MI19S_68 lesion margin cone 5a-a M MI-8 Southwest MI19S_69 lesion margin cone 4b-a M MI-8 Southwest MI19S_70 lesion margin cone 5b-a M MI-8 Southwest MI19S_71 lesion margin cone 8b-a M MI-8 Southwest MI19S_72 lesion margin cone 2a-a M MI-8 Southwest MI19S_73 lesion margin cone 6f-c ITS MI-6 Southwest MI19S_74 lesion margin cone 1a-a M MI-17 North MI19S_78 lesion margin cone 1c-a ITS MI-12 North MI19S_79 lesion margin cone 2c-a ITS MI-13 North MI19S_80 lesion margin cone 8a-a ITS MI-13 North MI19S_81 lesion margin cone 2c-a ITS MI-14 North MI19S_82 lesion margin cone 1c-a M MI-14

191

Table 4.4. (cont’d). x Production region (East, North, Southwest, West) in the lower peninsula of Michigan. y Lesion Margin = A hyphal-tip isolate generated from the lesion margin of symptomatic leaf tissue. Z Isolates identified by ITSrDNA were 100% homologous with the Diaporthe sp. 1-MI isolates collected in 2018 (ITS). The remaining Diaporthe isolates were designated as Diaporthe sp. 1-MI based on matching colony morphology (M).

192

Figure. 4.6. Fungal recovery (%) from hop cones with necrosis, russeting, and dried bract tissue in the lower peninsula of Michigan (A) of Diaporthe sp. 1-MI, other fungal spp. (Alternaria spp. Fusarium spp. and Epicoccum spp.), and cones yielding two or more fungal isolates (co- recovery). Co- recovery (%) from hop cones (B) of Diaporthe sp. 1-MI, Alternaria spp. and Fusarium spp. with one other fungal isolate. In some cases two or more fungal isolates were co- recovered (2+).

193

Table 4.5. Recovery of Diaporthe sp. 1-MI and other fungal isolates from damaged hop cones (e.g. necrosis, russeting, dried bract tissue) collected from four production regions in Michigan. Number (%) of cones with a fungal spp. recovered Regionu Yardv Conesw Diaporthe 1-MIx Alternaria spp. Fusarium spp. Epicoccum spp. Othery Co-Recoveryz E MI-11 18 3 (16.7) 7 (38.9) 1 (5.6) 1 (5.6) 3 (16.7) 3 (16.7) E MI-3 39 15 (38.5) 9 (23.1) 4 (10.3) 2 (5.1) 0 (0.0) 9 (23.1) E MI-9 17 3 (17.6) 3 (17.6) 1 (5.9) 3 (17.6) 0 (0.0) 7 (41.2) N MI-15 15 0 (0.0) 8 (53.3) 1 (6.7) 3 (20.0) 2 (13.3) 1 (6.7) N MI-13 14 0 (0.0)* 8 (57.1) 0 (0.0) 0 (0.0) 2 (14.3) 4 (28.6) N MI-14 15 2 (13.3) 8 (53.3) 2 (13.3) 1 (6.7) 2 (13.3) 0 (0.0) N MI-12 11 1 (9.1) 7 (63.6) 1 (9.1) 1 (9.1) 0 (0.0) 1 (9.1) SW MI-6 20 0 (0.0)* 4 (20.0) 4 (20.0) 2 (10.0) 2 (10.0) 8 (40.0) SW MI-17 6 0 (0.0)* 2 (33.3) 1 (16.7) 0 (0.0) 1 (16.7) 2 (33.3) SW MI-8 15 8 (53.3) 1 (6.7) 1 (6.7) 0 (0.0) 3 (20.0) 2 (13.3) W MI-16 20 2 (10.0) 6 (30.0) 6 (30.0) 2 (10.0) 0 (0.0) 4 (20.0) W MI-5 17 1 (5.9) 12 (70.6) 1 (5.9) 0 (0.0) 0 (0.0) 3 (17.6) W MI-7 94 11 (11.7) 50 (53.2) 5 (5.3) 6 (6.4) 4 (4.3) 18 (19.1) W MI-4 18 9 (50.0) 4 (22.2) 3 (16.7) 0 (0.0) 0 (0.0) 2 (11.1) W MI-10 28 2 (7.1) 13 (46.4) 3 (10.7) 1 (3.6) 3 (10.7) 6 (21.4) Total 15 347 57 (16.4) 142 (40.9) 34 (9.8) 22 (6.3) 22 (6.3) 70 (20.2) u Region (East, North, Southwest, West) in the lower peninsula of Michigan. w All yards listed are commercial hopyards except MI-11 is a research hopyard located at the MSU Plant Pathology Farm, Lansing, MI. x In yards with * Diaporthe sp. 1-MI was only co-recovered with other fungal species. In these yards Diaporthe sp. 1-MI was co-recovered from 2 (MI-13), 1 (MI-6) and 1 (MI-17) cones. w Number of cones collected per yard. y Fungal isolates recovered that were not the Diaporthe sp. 1-MI, Alternaria spp., Fusarium spp. or Epicoccum spp. z The number of cones with two of fungal species recovered.

194

Pathogenicity tests. In the leaf assay, ten-days post inoculation, irregular necrotic lesions with light brown pycnidia (167 m x 193 m (n = 10)) and single-celled hyaline conidia (12.4

m x 3.8 m (n = 15)) were observed on all abaxial leaf surfaces inoculated with Diaporthe sp.

1-MI at either concentration; one adaxial leaf developed symptoms when inoculated at the higher concentration. Control leaves and those inoculated with Alternaria sp. or P. prosopidis remained asymptomatic. The pathogen was re-isolated from symptomatic tissue and identified as a

Diaporthe sp. 1-MI based on colony morphology and TEF sequence homology.

Similar results were obtained using whole plant inoculations. At 21 dpi, Diaporthe- inoculated and the control plants had 42.7% ± 5.5 S.E. and 13.0% ± 5.5 S.E. necrotic, senescent leaves, respectively. Re-isolation was attempted on all symptomatic leaves using the surface sterilization procedure described above. Diaporthe sp. 1-MI was re-isolated from 100% of the

Diaporthe-inoculated plants but not from the control plants.

Morphology and isolate growth on PDA. On PDA, colony morphology for Diaporthe sp. 1-MI isolates from leaves (MI_0118, MI_0318, MI_1218, MI_2518) and cones (MI_2219,

MI _4119, MI _2719, MI_3019, MI_3119) was slightly raised to raised, white (10YR 8/1) to light grey (10YR 7/1) at 18-25 days branching mycelium (Figures 4.1G, 2G). The colony underside was pale brown (2.5Y 8/2 to 8/3) or pale brown turning dark gray (10 YR 4/1) and often formed a dark brown center (10YR 2/2) from the center; no discernable growth pattern

Pycnidia formation was variable (18 to 30+ days). The addition of sterile alfalfa stems induced pycnidia formation in single-spore isolates but the number of pycnidia that formed varied

(Figure 4.2H). Alpha conidia on PDA for leaf isolates, were unicellular, hyaline and cylindrical to ellipsoid (8.4 µm ± 1.5 s.d. by 3.4 µm ± 0.5 s.d. (n = 50)); no beta conidia were observed

(Table 4.2). Pycnidia (approximately 1458.4 µm ± 861.2 s.d. by 1190.2 µm ± 438.9 s.d.) were

195

dark brown, smooth or covered in hyphae, globose to ampulliform. Alpha conidia on PDA for cone isolates, were unicellular, hyaline and cylindrical to ellipsoid (10.2 µm ± 1.1 s.d. by 3.5 µm

± 0.5 s.d. (n = 50)); no beta conidia were observed (Table 4.2). Pycnidia (approximately 1272.

µm ± 148.4 s.d. by 1091.0 µm ± 108.9 s.d.) were dark brown, smooth or covered in hyphae, globose to ampulliform.

Growth was observed for Diaporthe sp. 1-MI isolates MI_0318, MI_1418, and MI_2518 at temperatures between 10 and 30° C (Figure 4.7). All isolates had limited growth (5.5 to 7.7 mm) at 10°C. Optimal growth occurred between 20 and 25°C for MI_1418 (80.9 mm ± 1.2 standard error (SE) and 80.1 mm ± 0.9 SE) and MI_0318 (81.1 mm ± 1.5 SE and 79.7 mm ± 0.6

SE). Optimal growth for MI_2518 was 20°C (80.8 mm ± 1.7 SE).

196

Discussion

Michigan hop growers face a new disease that we propose to name “halo blight” due to a chlorotic margin of leaf lesions and browning of cone bracts. The multilocus phylogenetic analysis identified the causal agent as Diaporthe sp. 1-MI, a novel taxon. Diaporthe sp. 1-MI was consistently recovered from tissue and pycnidia from both leaves and cones. Pathogenicity was demonstrated in detached leaves and whole plants. A survey of symptomatic hop cones indicates that halo blight is widely distributed and occurs in hop yards in four Michigan production regions.

Pycnidia were observed in symptomatic leaf and cone tissue, distinguishing the disease caused by Diaporthe sp. 1-MI from that incited by Phaeomycocentrospora cantuariensis

(Radišek et al. 2009), A. alternata (Darby 1984, 1988), Fusarium spp. (Bienapfl et al. 2005;

Pethybridge et al. 2001a), Psuedoperonospora humuli (Royle and Kremheller 1981), Botrytis cinerea (Mahaffee and Engelhard 2009) and Podosphaera macularis (Twomey et al. 2015).

Conidia and pycnidia observed in vivo shared similar morphology in cone and leaf tissue. There was a defined layer of fungal tissue (eustoma) separate from the outer wall of the pycnidia. Some cone pycnidia were embedded in textura angularis but contained morphology similar to the leaf pycnidia (stout pycnidia neck and cylindrical to ellipsoid conidia; no beta conidia). Stromatic

Phomopsis and Diaporthe species are common (Wechtl 1990; Wehmeyer 1926), but the cavity shape of stromatic pycnidia can be influenced by several factors including ontogeny (Wechtl

1990). Single spore isolates obtained directly from pycnidia in both tissue types shared 100%

ITSrDNA homology with each other and leaf isolates collected in 2018, indicating that pycnidia are derived from the same or closely-related Diaporthe species.

197

In pathogenicity tests, necrotic lesions with pycnidia and conidia morphology matching field descriptions were observed in leaf assays and similar results were obtained using whole plant inoculations. Isolates from leaf margins were grown on PDA and produced pycnidia permitting pathogenicity testing by conidia suspension. In many of the isolates derived directly from pycnidia a sterile alfalfa stem helped to induced formation (Farr et al. 2002), but the number of pycnidia that formed varied. Other inoculation methods utilizing hyphae, with or without wounding (Mathew et al. 2018) may be suitable for pathogenicity testing of fastidious isolates. In the leaf pathogenicity assays of this study, infection was more successful on the abaxial than adaxial leaf surfaces indicating entry might be gained through stoma via a germ tube and/or an appressorium. Conidia of some Diaporthe spp. can invade by cuticle penetration via an infection peg (Shankar et al. 1998; Williamson et al. 1991) or by germ tube through a stoma

(Kulik 1988) originating from an appressorium (Kulik 1988) or directly from the conidium

(Shankar et al. 1998; Williamson et al. 1991).

The single-celled conidia of Diaporthe sp. 1-MI are similar to Phoma exigua var. exigua; a pathogen producing comparable symptoms on hop leaves and cones (Radišek et al. 2008).

Phoma spp. produce a single conidium type (Van der Aa et al. 1990) while Phomopsis, the anamorph of Diaporthe, forms two primary types of conidia, alpha and beta (Wehmeyer 1933).

However, beta conidia are often difficult to produce in culture or absent (Rehner and Uecker

1994). We did not observe beta conidia in vivo nor in vitro on PDA and PDA supplemented with a sterile alfalfa stem. Alpha conidia in vivo for Diaporthe sp. 1-MI were larger than those reported in P. exigua var. exigua (Radišek et al. 2008). In culture, conidia size did not differ between the species. P. exigua var. exigua is reported to produce larger alpha conidia (aseptate;

5.5 to 11 x 2-4(-6) μm) in vivo than in culture (Boerema et al. 2004). Pycnidia of Phoma sect.

198

Phyllostictoides are generally thin walled and with a pseudoparenchymatous wall structure

(Boerema et al. 2004) that appears to be in contrast to the eustroma or thicken layer of textura angularis observed in the in vivo pycnidia in this study. Combined with genetic evidence, 67.4 to

70.7% ITSrDNA identity (GenBank: EF136399, EF136400; (Radišek et al. 2008) with the

Diaporthe sp. 1-MI isolates recovered in this study, we believe that these are two different pycnidia fungi affecting hop.

The phylogenetic analysis conducted with four loci including ITSrDNA, TEF, TUB, and

HIS revealed that the Diaporthe sp. 1-MI isolates are a unique taxon. All Diaporthe sp. 1-MI isolates clustered together in a single well supported clade for the multilocus and individual loci analysis except for the TEF loci. In the multilocus analysis, Diaporthe sp. 1-MI isolates were most closely related to the clade that includes the D. arecae species complex, D. hongkongensis and D. multigutullata; these species cluster into a clade according to Gao et al. (2017). The multilocus tree topology presented in this study is well supported in both BA and ML analysis and in agreement with other Diaporthe topologies (Gao et al. 2017; Gomes et al. 2013).

In individual locus analysis, three loci were homologous amongst Diaporthe sp. 1-MI isolates (ITSrDNA, TUB and HIS). Phylogenetic analysis of the amplified TUB partial gene region produced a topology most similar to the multilogues analysis and would likely be a useful single locus marker for Diaporthe sp. 1-MI in the future. Huang et al. (2015) analyzed a similar set of Diaporthe species using the same four markers and found TEF1 and TUB to have the highest percent of parsimony informative characters. Variation within the Diaporthe sp. 1-MI clade was driven only by the TEF1 locus. Heterogenicity of individual loci, such the variability found in the ITSrDNA region within populations of the D. eres species complex, can result in an

199

overestimation of Diaporthe diversity (Udayanga et al. 2014). We would not recommend the

TEF1 region as a single locus marker for the novel clade identified in this study.

There are other reports of Diaporthe and Phomopsis spp. affecting hops. Phomopsis tuberivora (synonymous with Phacidiopycnis tuberivora) causing red crown rot is reported only in belowground plant tissue (Gent et al. 2013; McGee et al. 2009). Morphological features are similar to Diaporthe sp. 1-MI in this study including 10.7 ± 0.96 x 4-6 μm single celled, spindle shaped conidia with pycnidia found within a stroma; beta conidia rare (Güssow and Foster

1932). Phomopsis tuberivora (GenBank: JF955785; (Gent et al. 2013) shared 91.3% ITSrDNA identity with Diaporthe sp. 1-MI recovered in this study (data not shown) and suggest the two are different species. Diaporthe sarmentella (Saccardo 1878); Preserved specimen: (Carter 1973) and Phoma sarmentella (syn. Phomopsis sarmentella Grove 1917) on Humulus were reported in stolons and dead stolons (Cash 1965; Saccardo 1878). However, the morphological descriptions of either species made by Saccardo (1878) (translated with (Cash 1965) only reference the sexual reproductive state making the description for the P. sarmentella ambiguous and comparisons to the Michigan isolates uninformative. Phomopsis sarmentella recovered from hop bines (Grove 1917); Preserved specimen: New York Botanical Gardens) had smaller (5-6 x 2-3

μm) but similarly shaped (cylindric-fusoid) conidia (Grove 1917) compared to Diaporthe sp. 1-

MI. With limited genetic information from historical reports, it is difficult to draw a definitive conclusion that Diaporthe sp. 1-MI has not been previously reported on hop.

Diaporthe humulicola is a new pathogen recently recovered from hop leaves and cones in two research hop yards in Connecticut (Allan-Perkins et al. 2020). D. humulicola causes ellipsoid spots on leaves, often along the margin, and brown bract tips on cones. In contrast,

Diaporthe sp. 1-MI causes irregular-shaped lesion that result in blighting. Both pathogens

200

produce pycnidia in adaxial leaf lesion but in cones infected by D. humulicola pycnidia were not reported. In vivo SEM images of Diaporthe 1-MI infected tissue revealed pycnidia with an outer eustroma layer. Stromatic pycnidia were not noted in differential interference contrast microscopy of D. humulicola although this could have been an artifact of growth on artificial media, pycnidia ontogeny (Wechtl 1990), or microscopic resolution. Alpha conidia morphology of D. humulicola on half strength PDA are similar to Diaporthe sp. 1-MI leaf and cone isolates growing on full strength PDA; beta and gamma conidia were not observed in either species. In pathogenicity assays, both pathogens successfully infect hop leaves via conidia suspensions.

BLAST sequence results for the ITS and HIS regions share 100% homology with the Diaporthe sp. 1-MI used in the phylogenetic analysis; TEF homology ranged from 98.8 to 99.7% pairwise identity. However, the TUB region amplified for D. humulicola is not homologous with

Diaporthe sp. 1-MI nor any of the other Diaporthe spp. used in the phylogenetic analysis in this study. The TUB primer set (T1 and Bt-2b) used in this study and by Gomes et al. (2013) (T1 and

Bt-2b or CYLTUB1R) amplify a highly variable intron-rich region on the 5’ end of the TUB gene (Glass and Donaldson 1995; O’Donnell and Cigelnik 1997). The TUB primer set (T12 and

T22) used by Allan-Perkins et al. (2020) amplifies a more conserved intron-poor region on the 3’ end of the TUB gene (Glass and Donaldson 1995; O’Donnell and Cigelnik 1997) useful in determining relationships among ascomycete classes and orders (Landvik et al. 2001). Allan-

Perkins et al. (2020) also used an oomycete primer set (TUBUF2 and TUBR1) that, in

Phytophthora cinnamomi, will amplify a portion of the TUB region with no introns (Kroon et al.

2004). The amplification of non-overlapping regions accounts for the phylogenetic placement of

Diaporthe 1-MI and D. humulicola with different Diaporthe clades. Further analysis is needed to

201

complete the identification of D. humulicola using a consistent TUB region before a species can be definitely assigned to Diaporthe sp. 1-MI.

If Diaporthe sp. 1-MI is a distinct species from D. humulicola, we propose the name

“halo blight” since it is descriptive of the chlorotic margin bordering leaf lesions and blighting of cone bracts. If D. humulicola is confirmed synonymous, we propose changing the disease name from “diaporthe leaf spot” put forward (Allan-Perkins et al. 2020) to “halo blight” since disease symptoms are not confined to the leaves and as demonstrated in this research the disease results in a blighting of the cones. Furthermore, the genus Diaporthe known to be paraphyletic which may incite a phylogenetic reorganization (Gao et al. 2017) and could require future changes to any disease name containing the genus.

Diaporthe sp. 1-MI was recovered from hop cones in four production regions of

Michigan’s lower peninsula. Diaporthe sp. 1-MI was more frequently recovered in the central, eastern, and southwest regions than in the north. The presence of cirrhi and gelatinous masses of conidia suggests this pathogen may be moved by water and contaminated surfaces. Splash dispersed Diaporthe pathogens are common on other crops include monocyclic Phomopsis viticola (Anco et al. 2012) and polycyclic P. vaccinii (Parker and Ramsdell 1977). Diaporthe helianthi (Gulya et al. 1997), can generate airborne primary inoculum from ascocarps. We did not observe ascocarps, but this could be an artifact of late season sampling or a heterothallic reproductive requirement (Santos et al. 2010).

Alternaria spp. were frequently recovered from cones in each region and may be a secondary colonizer. Alternaria is a ubiquitous saprophyte, but some species are phytopathogenic (Thomma 2003). Alternaria alternata and A. humuli are reported pathogens of

Humulus lupus (Darby 1984, 1988; Pethybridge et al. 2001b; Simmons 2002) and A. humuli-

202

scandens is pathogenic on H. scandens (syn. H. japonicus) (Zhang et al. 2009). High humidity favors A. alternata infection of hop cones. Wounding is not required, but infection is exacerbated by abrasions typically caused by high winds (Darby 1984, 1988). We were not able to establish pathogenicity of three Alternaria spp. isolates on undamaged detached leaves. Pathogenicity testing and species level identification using multilocus markers (Woudenberg et al. 2013) of more isolates is needed. A. alternata is considered a cosmetic pathogen of damaged and senescent cones in the Pacific Northwest (Gent 2009; Twomey et al. 2015). The frequency of

Alternaria spp. recovered from symptomatic hop cones collected prior to harvest, combined with reports of Alternaria mycotoxins in beer (Bauer et al. 2016; Siegel et al. 2010) warrant further investigation.

Diaporthe sp. 1-MI was identified as the causal agent of halo leaf and cone blight which, to our knowledge, was previously undescribed. Field symptoms, pathogen signs, and morphological and genetic taxonomical features were characterized. The pathogen appears to be widespread in Michigan. Hop yards in other growing regions with similar environmental conditions may also be at risk. Growers require additional information regarding the epidemiology of the disease and measures needed to protect hop foliage and cones from this fungal pathogen.

203

APPENDIX

204

205

206

207

208

LITERATURE CITED

209

LITERATURE CITED

Allan-Perkins, E., Li, D.-W., Schultes, N., Yavuz, S., and LaMondia, J. 2020. The identification of a new species, Diaporthe humulicola, a pathogen causing Diaporthe Leaf Spot on common hop. Plant Dis. (First Look) Online: 30 Mar 2020:https://doi.org/10.1094/PDIS- 1008-1019-1770-RE.

Anco, D. J., Madden, L. V., and Ellis, M. A. 2012. Temporal patterns of sporulation potential of Phomopsis viticola on infected grape shoots, canes, and rachises. Plant Dis. 96:1297- 1302.

Aveskamp, M. M., De Gruyter, J., Woudenberg, J. H. C., Verkley, G. J. M., and Crous, P. W. 2010. Highlights of the Didymellaceae: a polyphasic approach to characterise Phoma and related pleosporalean genera. Stud. Mycol. 65:1-60.

Barnett, H., and Hunter, B. 1998. Illustrated genera of imperfect fungi 4th edition. APS Press, St. Paul, Minnesota.

Bastide, F., Sérandat, I., Gombert, J., Laurent, E., Morel, E., Kolopp, J., Guillermin, P.-L., Hamon, B., Simoneau, P., and Berruyer, R. 2017. Characterization of fungal pathogens (Diaporthe angelicae and D. eres) responsible for umbel browning and stem necrosis on carrot in France. Plant Pathol. 66:239-253.

Bauer, J. I., Gross, M., Gottschalk, C., and Usleber, E. 2016. Investigations on the occurrence of mycotoxins in beer. Food Control 63:135-139.

Baumgartner, K., Fujiyoshi, P. T., Travadon, R., Castlebury, L. A., Wilcox, W. F., and Rolshausen, P. E. 2013. Characterization of species of Diaporthe from wood cankers of grape in eastern North American vineyards. Plant Dis. 97:912-920.

Bienapfl, J. C., Ocamb, C. M., Klein, R., and Nelson, M. 2005. Fusarium cone tip blight of Humulus lupulus. Acta Hortic. 668:123-128.

Boerema, G. H., Gruyter, J. d., Noordeloos, M. E., and Hamers, M. E. C. 2004. Phoma identification manual: differentiation of specific and infra-specific taxa in culture. CABI Publishing, Cambridge, MA.

Brewers Association. 2019. State Craft Beer Sales & Production Statistics, 2018: Michigan. Online publication: brewersassociation.org.

Carbone, I., and Kohn, L. M. 1999. A method for designing primer sets for speciation studies in filamentous ascomycetes. Mycologia 91:553-556.

Carter, M. A. 1973. Diaporthe sarmenticia associated with Humulus lupulus. Mycological Institute (IMI) fungarium, Kew, Richmond, London, Herb IMI: 172750a.

210

Cash, E. K. 1965. A Mycological English-Latin Glossary. Mycological Memoirs (No. 1), Hafner Publishing Company, New York and London.

Choi, Y.-W., Hyde, K. D., and Ho, W. H. 1999. Single spore isolation of fungi. Fungal Divers.3:29-38.

Crous, P. W., Groenewald, J. Z., Risède, J.-M., Simoneau, P., and Hywel-Jones, N. L. 2004. Calonectria species and their Cylindrocladium anamorphs: species with sphaeropedunculate vesicles. Stud. Mycol. 50:415-430.

Crous, P. W., Wingfield, M. J., Le Roux, J. J., Richardson, D. M., Strasberg, D., Shivas, R. G., Alvarado, P., Edwards, J., Moreno, G., and Sharma, R. 2015. Fungal Planet description sheets: 371–399. Persoonia 35:264.

Darby, P. 1984. Department of Hop Research Annual Report 1983, Wye College. Invicta Press, Ashford, Kent, and London, UK.

Darby, P. 1988. Alternaria alternata infection of hop (Humulus lupulus) cones. Mycol. Res. 90:650-653.

Darriba, D., Taboada, G. L., Doallo, R., and Posada, D. 2012. jModelTest 2: more models, new heuristics and parallel computing. Nat. Methods 9:772-772.

Del Castillo Munera, J., Higgins, D. S., Harlan, B., and Hausbeck, M. K. 2016. Occurrence and management of hop (Humulus lupulus) powdery mildew (Podosphaera macularis) in Michigan. Phytopathology 106 (Suppl.):S4.187.

Du, Z., Fan, X.-L., Hyde, K. D., Yang, Q., Liang, Y.-M., and Tian, C.-M. 2016. Phylogeny and morphology reveal two new species of Diaporthe from Betula spp. in China. Phytotaxa. 269:90-102.

Fan, X., Yang, Q., Bezerra, J. D., Alvarez, L. V., and Tian, C. 2018. Diaporthe from walnut tree (Juglans regia) in China, with insight of the Diaporthe eres complex. Mycol. Prog. 17:841-853.

Farr, D. F., Castlebury, L. A., and Rossman, A. Y. 2002. Morphological and molecular characterization of Phomopsis vaccinii and additional isolates of Phomopsis from blueberry and cranberry in the eastern United States. Mycologia 94:494-504.

Gao, Y., Liu, F., and Cai, L. 2016. Unravelling Diaporthe species associated with Camellia. System. Biodivers. 14:102-117.

Gao, Y., Liu, F., Duan, W., Crous, P. W., and Cai, L. 2017. Diaporthe is paraphyletic. IMA Fungus 8:153.

Gardes, M., and Bruns, T. D. 1993. ITS primers with enhanced specificity for basidiomycetes‐ application to the identification of mycorrhizae and rusts. Mol. Ecol. 2:113-118.

211

Gent, D. H. 2009. Alternaria cone disorder. Page 15 in: Compendium of Hop Diseases and Pests. W. F. Mahaffee, S. J. Pethybridge and D. H. Gent, eds. APS press, St. Paul, MN.

Gent, D. H., and Radišek, S. 2009. Fungal Diseases and Pathogens of Minor Importance. In: Compendium of Hop Diseases and Pests. W. F. Mahaffee, S. J. Pethybridge and D. H. Gent, eds. APS press, St. Paul, MN.

Gent, D. H., Mueller-Warrant, G., Woods, J. L., Putnam, M. L., and Twomey, M. C. 2013. Red Crown Rot of Hop in Oregon Caused by Phomopsis tuberivora. Plant Health Prog. 14:52.

George, A. 2018. USA hops 2017 Statistical Report. Hop Growers of America, Online publication: usahops.org.

George, A. 2020. USA hops 2019 Statistical Report. Hop Growers of America, Online publication: usahops.org.

Glass, N. L., and Donaldson, G. C. 1995. Development of primer sets designed for use with the PCR to amplify conserved genes from filamentous ascomycetes. Appl. Environ. Microbiol. 61:1323-1330.

Gomes, R. R., Glienke, C., Videira, S. I. R., Lombard, L., Groenewald, J. Z., and Crous, P. W. 2013. Diaporthe: a genus of endophytic, saprobic and plant pathogenic fungi. Persoonia 31:1.

Grove, W. B. 1917. The British Species of Phomopsis. Royal Botanic Gardens, Kew, UK.

Guindon, S., and Gascuel, O. 2003. A simple, fast and accurate method to estimate large phylogenies by maximum-likelihood. Syst. Biol. 52:696-704.

Gulya, T., Rashid, K. Y., and Masirevic, S. M. 1997. Sunflower Diseases. Pages 263-379 in: Sunflower Technology and Production. A. A. Schneiter, ed. American Society of Agronomy, Inc. Madison, WI.

Güssow, H. T., and Foster, W. R. 1932. A new species of Phomopsis. Can. J. Res. 6:253-254.

Hall, B. G. 2011. Phylogenetic Trees Made Easy: A How to Manual. Oxford University Press, New York, NY.

Herr, L. J., Lipps, P. E., and Watters, B. L. 1983. Diaporthe stem canker of sunflower. Plant Dis. 67:911-913.

Higgins, D. S., and Hausbeck, M. K. 2017. Fungicides and cultivars can limit hop downy mildew (Pseudoperonospora humuli) in Michigan. Phytopathology 107:S5-78.

Higgins, D. S., and Hausbeck, M. K. 2018. Effective management strategies for hop downy mildew (Pseudoperonospora humuli) in Michigan. Phytopathology 108:S2.28.

212

Higgins, D. S., Byrne, J. M., Hausbeck, M. K., and Miles, T. D. 2019. An emerging pathogen (Diaporthe sp.) of hop (Humulus lupulus) associated with foliar necrosis in commercial production yards in Michigan. Phytopathology 109:S2.111.

Huang, F., Udayanga, D., Wang, X., Hou, X., Mei, X., Fu, Y., Hyde, K. D., and Li, H. 2015. Endophytic Diaporthe associated with Citrus: A phylogenetic reassessment with seven new species from China. Fungal Biol. 119:331-347.

Kroon, L. P. N. M., Bakker, F. T., van den Bosch, G. B. M., Bonnants, P. J. M., and Flier, W. G. 2004. Phylogenetic analysis of Phytophthora species based on mitochondrial and nuclear DNA sequences. Fungal Genet. Biol. 41:766-782.

Kulik, M. 1988. Observations by scanning electron and bright-field microscopy on the mode of penetration of soybean seedlings by Phomopsis phaseoli. Plant Dis. 72:115-118.

Lalancette, N., Foster, K. A., and Robison, D. M. 2003. Quantitative models for describing temperature and moisture effects on sporulation of Phomopsis amygdali on peach. Phytopathology 93:1165-1172.

Landvik, S., Eriksson, O. E. and Berbee, M. L. 2001. Neolecta a fungal dinosaur? Evidence from 훽-tubulin amino acid sequences. Mycologia 93:1151-1163.

Machingambi, N. M., Dreyer, L. L., Oberlander, K. C., Roux, J., and Roets, F. 2015. Death of endemic Virgilia oroboides trees in South Africa caused by Diaporthe virgiliae sp. nov. Plant Pathol. 64:1149-1156.

Mahaffee, W. F., and Engelhard, B. 2009. Gray Mold. In: Compendium of Hop Diseases and Pests. W. F. Mahaffee, D. H. Gent and S. J. Pethybridge, eds. APS Press, St. Paul, MN.

Mahaffee, W. F., Pethybridge, S. J., and Gent, D. H. 2009. Compendium of Hop Diseases and Pests. APS Press, St. Paul, MN.

Mathew, F. M., Jordahl, J. G., Gulya, T. J., and Markell, S. G. 2018. Comparison of greenhouse- based inoculation methods to study aggressiveness of Diaporthe helianthi isolates causing Phomopsis stem canker of sunflower (Helianthus annuus). Plant Health Prog. 19:92-96.

Mathew, F. M., Alananbeh, K. M., Jordahl, J. G., Meyer, S. M., Castlebury, L. A., Gulya, T. J., and Markell, S. G. 2015. Phomopsis stem canker: A reemerging threat to sunflower (Helianthus annuus) in the United States. Phytopathology 105:990-997.

McGee, P. A., Lawrie, A., and Porter, I. 2009. Red Crown Rot. In: Compendium of Hop Diseases and Pests. W. F. Mahaffee, D. H. Gent and S. J. Pethybridge, eds. APS Press, St. Paul, MN.

Miles, T. D., Day, B., and Schilder, A. C. 2011. Identification of differentially expressed genes in a resistant versus a susceptible blueberry cultivar after infection by Colletotrichum acutatum. Mol. Plant Pathol. 12:463-477.

213

Miller, M. A., Pfeiffer, W., and Schwartz, T. 2010. Creating the CIPRES Science Gateway for Inference of Large Phylogenetic Trees. Pages 1-8 in: Gateway Computing Environments Workshop (GCE), New Orleans, LA.

Moretti, S., Armougom, F., Wallace, I. M., Higgins, D. G., Jongeneel, C. V., and Notredame, C. 2007. The M-Coffee web server: a meta-method for computing multiple sequence alignments by combining alternative alignment methods. Nucleic Acids Res. 35:W645- W648.

O’Donnell, K., and Cigelnik, E. 1997. Two divergent intragenomic rDNA ITS2 types within a monophyletic lineage of the fungus Fusarium are nonorthologous. Mol. Phylo. Evol. 7:103-116.

Parker, P. E., and Ramsdell, D. C. 1977. Epidemiology and chemical control of Phomopsis canker of highbush blueberry. Phytopathology 67:484.

Pethybridge, S., Hay, F., Wilson, C., Sherriff, L., and Leggett, G. 2001a. First report of Fusarium crookwellense causing tip blight on cones of hop. Plant Dis. 85:1208-1208.

Pethybridge, S. J., Hay, F. S., Wilson, C. R., Sherriff, L. J., Leggett, G. W., Polizzi, G., Catara, V., Somda, I., Veena, M. S., and Mortensen, C. N. 2001b. First report of infection of hop cones by Alternaria alternata in Australia. Plant Dis. 85:804.

Phalip, V., Hatsch, D., Laugel, B., and Jeltsch, J.-M. 2006. An overview of fungal community diversity in diseased hop plantations. FEMS Microbiol. Ecol. 56:321-329.

Putto, B. L., Koul, M. L., and Saraf, R. L. 1975. Septoria leaf disease of hops (Humulus lupulus) in Kashmir. Indian Phytopathol. 27:247-248.

Radišek, S., Jakse, J., Javornik, B., and Gruyter, J. d. 2008. First report of Phoma exigua as a pathogen of hop in Slovenia. Plant Pathol. 57:381.

Radišek, S., Leskošek, G., and Javornik, B. 2009. Occurrence of Cercospora cantuariensis on hop in Austria and Slovenia. Plant Path. 58:400

Raja, H. A., Miller, A. N., Pearce, C. J., and Oberlies, N. H. 2017. Fungal identification using molecular tools: a primer for the natural products research community. J. Nat. Prod. 80:756-770.

Rasband, W. S. 1997. ImageJ. National Institutes of Health, Bethesda, MD, USA, https://imagej.nih.gov/ij/.

Rehner, S. A., and Uecker, F. A. 1994. Nuclear ribosomal internal transcribed spacer phylogeny and host diversity in the coelomycete Phomopsis. Can. J. Bot. 72:1666-1674.

214

Ronquist, F., Teslenko, M., Van Der Mark, P., Ayres, D. L., Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M. A., and Huelsenbeck, J. P. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst. Biol. 61:539- 542.

Royle, D., and Kremheller, H. T. 1981. Downy Mildew of the Hop. In: The Downy Mildews. D. M. Spencer, ed. Academic Press Inc., New York, NY.

Saccardo, P. A. 1878. Fungi veneti novi vel critici vel mycologiae venetae addendi [Translated with Cash 1965]. Michelia 1:361-445.

Sambrook, J., Fritsch, E. F., and Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

Santos, J. M., Correia, V. G., and Phillips, A. J. L. 2010. Primers for mating-type diagnosis in Diaporthe and Phomopsis: their use in teleomorph induction in vitro and biological species definition. Fungal Biol. 114:255-270.

Seifert, K. A., and Rossman, A. Y. 2010. How to describe a new fungal species. IMA Fungus. 1:109-116.

Shankar, M., Cowling, W., and Sweetingham, M. 1998. Histological observations of latent infection and tissue colonization by Diaporthe toxica in resistant and susceptible narrow- leafed lupins. Can. J. Bot. 76:1305-1316.

Siegel, D., Merkel, S., Koch, M., and Nehls, I. 2010. Quantification of the Alternaria mycotoxin tenuazonic acid in beer. Food Chem. 120:902-906.

Simmons, E. G. 2002. Alternaria themes and variations (305-309) Lewia/Alternaria revisited. Mycotaxon. 83:127-145.

Singh, G., Singh, R., and Bhandari, T. 1984. Ascochyta leaf spots on fig and hop. New records for India. Indian J. Mycol. Pl. Path. 14:100.

Sirrine, J. R., Rothwell, N., Lizotte, E., Goldy, R., Marquie, S., and Brown-Rylewski, D. E. 2010. Sustainable Hop Production in the Great Lakes Region [E3083]. Michigan State University, MSU Extension, East Lansing, MI.

Stamatakis, A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30:1312-1313.

Sullivan, R. F., and White, J. F. 2000. Phoma glomerata as a mycoparasite of powdery mildew. Appl. Environ. Microbiol. 66:425-427.

Tan, Y., Edwards, J., Grice, K., and Shivas, R. 2013. Molecular phylogenetic analysis reveals six new species of Diaporthe from Australia. Fungal Divers. 61:251-260.

215

Thomma, B. P. H. J. 2003. Alternaria spp.: from general saprophyte to specific parasite. Mol. Plant Pathol. 4:225-236.

Thompson, S., Tan, Y., Shivas, R., Neate, S., Morin, L., Bissett, A., and Aitken, E. 2015. Green and brown bridges between weeds and crops reveal novel Diaporthe species in Australia. Persoonia 35:39.

Twomey, M., Stone, J., and Gent, D. 2016. Black wilt of hop (Humulus lupulus) caused by Diplodia seriata in New York State. Plant Dis. 100:861-861.

Twomey, M. C., Wolfenbarger, S. N., Woods, J. L., and Gent, D. H. 2015. Development of partial ontogenic resistance to powdery mildew in hop cones and its management implications. PloS One 10: DOI:10.1371/journal.pone.0120987.

Udayanga, D., Castlebury, L. A., Rossman, A. Y., Chukeatirote, E., and Hyde, K. D. 2014. Insights into the genus Diaporthe: phylogenetic species delimitation in the D. eres species complex. Fungal Divers. 67:203-229.

Udayanga, D., Castlebury, L. A., Rossman, A. Y., Chukeatirote, E., and Hyde, K. D. 2015. The Diaporthe sojae species complex: Phylogenetic re-assessment of pathogens associated with soybean, cucurbits and other field crops. Fungal Biol. 119:383-407.

Udayanga, D., Liu, X., Crous, P. W., McKenzie, E. H., Chukeatirote, E., and Hyde, K. D. 2012. A multi-locus phylogenetic evaluation of Diaporthe (Phomopsis). Fungal Divers. 56:157- 171.

Van der Aa, H. A., Noordeloos, M. E., and Gruyter, J. d. 1990. Species concepts in some larger genera of the Coelomycetes. Stud. in Mycol. :3-19.

Wallace, I. M., O'sullivan, O., Higgins, D. G., and Notredame, C. 2006. M-Coffee: combining multiple sequence alignment methods with T-Coffee. Nucleic Acids Res. 34:1692-1699.

Wechtl, E. E. 1990. Phomopsis (coelomycetes) species on Compositae and Umbelliferae: a critical evaluation of characters with keys. Linz. Biol. Beitr. 22:161-173.

Wehmeyer, L. E. 1926. A biologic and phylogenetic study of the stromatic Sphaeriales. Am. J. Bot. 13:575-645.

Wehmeyer, L. E. 1933. The Genus Diaporthe Nitschke and its Segregates. University of Michigan Press, Ann Arbor, MI.

White, T. J., Bruns, T., Lee, J., and Taylor, J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. PCR protocols: a guide to methods and applications 18:315-322.

Williamson, P., Sivasithamparam, K., and Cowling, W. 1991. Formation of subcuticular coralloid hyphae by Phomopsis leptostromiformis upon latent infection of narrow-leafed lupins. Plant Dis. 75:1023-1026.

216

Woudenberg, J. H. C., Groenewald, J. Z., Binder, M., and Crous, P. W. 2013. Alternaria redefined. Stud. in Mycol. 75:171-212.

Zhang, M., Wu, H., and Li, D. 2009. A new species of Alternaria on Humulus scandens from China. Mycotaxon 108:49-52.

217