Fire blight susceptibility of select cider apple cultivars

Abstract There is increasing interest in growing apple cultivars (Malus domestica Borkh.) of European origin for the production of hard cider in Canada; however, little is known about their susceptibility to fire blight (FB). FB can spread rapidly through apple (and pear) orchards causing extensive tree mortality and economic loss. Twenty-eight promising cider cultivars were evaluated over a 7 year period, and in their seventh year of production they were severely naturally infected by an Erwinia amylovora outbreak causing FB. Herein, we report the bloom and harvest dates and tree mortality that developed largely as secondary shoot blight in the summer of 2021. Overall, the cultivars could be classified according to relative susceptibility to FB, based on percentage tree mortality after 7 years: Enterprise (0%); GoldRush and Porter’s Perfection (<20%); Binet Rouge, Kingston Black, Cline Russet, Dabinett, Grimes Golden, Frequin Rouge, Crimson Crisp®, Cox Orange Pippin, and Muscadet De Dieppe (20%–40%); Calville Blanc d’Hiver, Bramley’s Seedling, Yarlington Mill, Michelin, Bulmers Norman, Stoke Red, Golden Russet, Breakwell, Esopus Spitzenberg (50%–90%); Brown Snout, Medaille d’Or, Michelin, Brown’s Apple, Sweet Alford, Tydeman Late, Ashmead’s Kernel, and Tolman (90%–100%). This study highlights the importance of selecting FB tolerant cider cultivars and following best management orchard practices to reduce the spread and prevent infection, which can be achieved by using FB-resistant rootstock, controlling rootstock suckers, FB prediction models, and limited use of antibiotics, biologicals, and careful nitrogen application to regulate tree vigor.


Introduction
There is an increasing interest in growing apples (Malus domestica Borkh.) for alcoholic cider production in North America because of increased consumer demand for this beverage.Traditionally, cider, often referred to as "hard cider" in North America, was produced primarily in Europe and is made from bittersweet and bittersharp apple cultivars.Specialty cider cultivars have been selected for their fruit and juice characteristics, and in some cases, for ease of management but not necessarily for yield or disease resistance.Traditional European cider cultivars were selected from areas with climates and pathogens vastly different from those found in southern Ontario's humid continental climate (Dfb Köppen Climate Classification).Consequently, several cultivars are challenging to grow because of disadvantages such as higher disease susceptibility, biennial bearing, pre-harvest drop, and excessive tree vigor (Moulton et al. 2010;Rothwell 2012).Growers must be extremely cautious about cultivar susceptibility to the fire blight (FB) pathogen Erwinia amylovora (Burrill 1882) Winslow et al. 1920, because it can spread rapidly within and between orchards, and severely impacts tree growth and often results in tree mortality.Even if trees do not fully succumb to the pathogen, the best management practice is to promptly remove trees to prevent further spread.Hence, FB is a serious disease that can have a widespread economic impact (Jones and Sutton 1997;Sutton et al. 2014).
The disease is caused by the bacterium, Erwinia amylovora, which infects other Rosaceae hosts including quince (Cydonia), crab apple (Malus), hawthorn (Crataegus), cotoneaster (Cotoneaster), mountain ash (Sorbus), and firethorn (Pyracantha).The bacteria E. amylovora is found in most areas of southern Ontario and has become a significant concern in higher density apple plantings that contain both susceptible cultivars and dwarfing rootstocks, such as M.9 and M.26.In addition, E. amylovora attacks many different parts of the tree, and FB symptoms are often named by referring to the infected location, such as the blossom, shoot, fruit, limb and trunk, and collar or rootstock blight.
Blossom blight is observed in spring when the flowers become infected.Infected blooms first appear water-soaked and the begin to wilt, shrivel, and turn brown or black.Pollinating insects that visit contaminated flowers spread the disease to noninfected flowers.Once blooms are infected, blossom blight progresses into the flower pedicel and then into the spur.If the spur on a main branch or the trunk becomes infected, the bacteria can move from into the branch or trunk eventually developing into a canker resulting in tissue death.Succulent shoots and water sprouts can also become infected later during the growing season resulting in shoot blight.Infected shoots first appear water-soaked but quickly turn brown or black.Shoot blight develops very quickly once infection occurs and can be exacerbated when the leaves are mechanically injured by wind or hail.Bacteria can move 15-30 cm or more in the shoot within a few days of infection.As the bacteria migrate down the shoot, stem tissues become discoloured and shriveled.Infected shoots form a classical shepherds' crook shape at the tip and are often associated with bacterial ooze along the midvein of infected leaves and the stem of infected shoots.The bacteria in the ooze can easily and quickly spread to other tissues or trees by visiting insects and splashing rain during wet periods.The most severe losses from both blossom and shoot blight occur when the disease progresses into older wood tissue.Erwinia amylovora progresses into the main limbs and trunk of the tree from infected spurs or shoots when warm temperatures with high humidity combine to form ideal conditions for bacterial growth.Rootstock blight can occur in susceptible rootstocks such as M.9 and M.26, and often occurs at ground level just below the graft union.This type of blight frequently results in tree mortality, often occurring within a year of infection.Infection of FB causing bacteria can also occur through suckers inoculated with bacteria washed by rain down the trunk from infected twigs and fruit, or translocation via the phloem.This can occur in the absence of symptoms in the scion, masking the extent and severity of infection.Further details and cultural methods for managing FB have been reported elsewhere (OMAFRA 2022).
There is a paucity of information on FB susceptibility of apple cultivars grown for cider in the United States and Canada.Miles et al. ( 2020) provide a informative treatise on trends and research opportunities in the United States but exclude specific details of cultivar traits and resistance to FB.The lack of data and empirical information on the FB susceptibility of cider cultivars makes it difficult for cider producers to identify which cultivars will have the greatest chance of being successful in Canada and other regions with similar climatic conditions.
The goal of this study was to report the FB susceptibility of select apple cultivars grown for cider that succumb to an extensive and devastating natural E. amylovora infection that caused widespread tree mortality in a 7 year orchard in 2021.

Materials and methods
The orchard was established in 2015 at the Horticultural Experiment Station, Simcoe (42 • 51 40 N, 80 • 16 8 W) on M.9 rootstock and trained in a vertical axis orchard system.Trees were spaced 1 m within and 4.5 m between rows (1667 trees/ha).Trees were planted in a randomized complete block, with four replications of five trees for each of the 28 cultivars, for a total of 560 trees.All trees were custom propagated on the same source of rootstock from a single commercial nursery in Ontario.Bloom and harvest data were collected from the middle three trees of each five-tree block, with the two outside trees in each block acting as guard trees.Each replicate occupied two rows, resulting in a total of eight rows.Gala trees on M.9 rootstock were planted as "guard trees" at the ends of each row.Evaulated cultivars consisted of Ashmead's Kernel, Breakwell, Brown's Apple, Bulmers Norman, Binet Rouge, Bramley's Seedling, Brown Snout, Calville Blanc d'Hiver, Crimson Crisp , Cox Orange Pippin, Cline Rus-set, Dabinett, Enterprise, Esopus Spitzenberg, Fréquin Rouge, GoldRush, Grimes Golden, Golden Russet, Kingston Black, Michelin, Muscadet de Dieppe, Medaille d'Or, Porter's Perfection, Sweet Alford, Stoke Red, Tydeman Late, Tolman Sweet, and Yarlington Mill.
Trees were provided with supplemental trickle irrigation via 2 L/h pressure-compensating emitters spaced 45 cm apart and watered daily during the growing season with an equivalent of ∼2.5 cm water weekly (adjusted for natural rainfall) on a schedule of six irrigation run-times per day every 4 h (20 min per event).The orchard soil consisted of a Brady sandy loam (Brunisolic Grey Brown Luvisol) (Presant and Acton 1984) with imperfect drainage and soil textures consisting of mainly lacustrine sand and sandy loam over glaciolacustrine clays at depths >1.5 m (Hohner and Presant 1989).Pests and diseases were managed with conventional pesticide (OMAFRA 2021).Trees were fertilized annually with muriate of potash based on leaf tissue analyses but were not fertilized with nitrogen--to reduce tree vigor and the risk of FB.Daily minimum and maximum temperatures at 1.5 m above the ground were recorded at the Simcoe Research Station.Further planting details, cultivar descriptions, early yield, and juice characteristics have been previously reported (Plotkowski 2020;Cline et al. 2021;Plotkowski andCline 2021a, 2021b).

Horticultural measurements
In the spring of 2017-2021, phenology was rated three to four times per week from late April to the end of flowering in June.The rating system used herein was developed at Washington State University, and covers the progression from the silver tip stage immediately after dormancy to petal fall (Ballard et al. 1981).Tree mortality was recorded every autumn from 2015 to 2021.Fruits were harvested when the average starch index rating of 10-fruit ranged from 6 to 8 using the Cornell Starch Index (Blanpied and Silsby 1992).

Fire blight management
FB was managed with prophylactic applications of streptomycin during bloom and prohexadione-calcium (Apogee™) pre-and post-bloom.The incidence of FB pathogen infections was extremely low to nonexistent in this orchard between 2015 and 2020.However, FB symptoms began to appear in vegetative 1 year old shoots in mid-June 2021 almost exclusively in the cider apple cultivars, while other adjacent fresh-market cultivar blocks of Gala, Honeycrisp, Ambrosia, Golden Delicious, and Cortland were largely unaffected.The FB symptoms moved quickly in the tree to subtending older tissues.An aggressive spray programme to contain its spread was immediately implemented with an initial application of streptomycin (24 June) followed by fixed copper (Cueva) on 28 June, and 5, 19, and 26 July (Table 1).Unfortunately, the FB infection continued to spread despite active removal of affected shoots using recommended sterile methods, and in some cases removal of entire trees when there were multiple strikes on a tree.The infections continued unabated, and in many instances advanced to the rootstock, causing collapse of the entire tree.By November 2021, there was widespread tree mortality in the orchard, to the extent that the difficult decision to remove the entire orchard was made to prevent further spread to adjacent research and commercial orchards.Historically, the incidence of FB infection has been low over the past 30 or more years of production.This led us to suspect that the primary infection inoculum may have migrated from adjacent commercial orchards, where FB infections have been observed in the recent years and may be related to the use of overhead irrigation for frost control at bloom.

Statistical analyses
Data were subjected to analysis of variance using the PROC GLIMMIX procedure in SAS (SAS 9.4, SAS Institute, Cary, NC).Mean separation using Tukey's HSD was used to separate treatment means (P = 0.05).Shapiro-Wilk tests were used to test the normality of the residuals.Scatterplots of studentized residuals were visually observed to test the assumption that errors were not heterogeneous.In cases where there were large deviations from assumptions, the data were transformed using log-or square-root transformation prior to analysis.
The bloom date and the period during which apple trees remain in bloom are not only important factors that determine which cultivars should be planted together to synchronize cross-pollination, but can also impact susceptibility to frost and FB.This is because a late bloom or protracted bloom period may increase the risk of blossom flight infection of Erwinia amylovora, which is of particular concern to apple growers in the humid regions of eastern North America (Grove et al. 2003).

Harvest date and fruit maturity
The 5 year harvest dates for all apple cultivars ranged over a 76 day period from 17 August (Brown's Apple) to 1 November (GoldRush) (Fig. 2).The 5 year average indicated that very early maturing cultivars included Brown's Apple (17 August) and Bulmer's Norman (27 August).Binet Rouge, Stoke Red, Kingston Black, Muscadet de Dieppe, and Michelin were harvested during the first 2 weeks of September.Breakwell's Seedling, Cox's Orange, Bramley's Seedling, Dabinett, Médaille d'Or, Ashmead's Kernel, Crimson Crisp, and Tolman Sweet matured between 15 and 30 September.The later maturing cultivars Porters Perfection, Fréquin Rouge, Cline Russet, Brown Snout, Esopus Spitzenburg, Sweet Alford, and Grimes Golden matured within the first 2 weeks of October.Lastly, Tydeman's Late Orange, Calville Blanc d'Hiver, Golden Russet, and Enterprise all matured in late October, with Gol-dRush maturing the latest on 1 November.Across the 5 years, Kingston Black, Ashmead's Kernel, Cox's Orange, and Goldrush varied the least in harvest dates (±1-1.5 days), whereas Golden Russet, Sweet Alford, Enterprise, and Fréquin Rouge harvest dates varied from 10 to 16 days across the 5 years.Therefore, there is a difference in cultivar sensitivity to environmental factors or orchard management features (e.g., crop load) that influence fruit maturity.

Cultivar sensitivity to fire blight (Erwinia amylovora)
Prior to 2021, all-cause tree mortality was relatively low, but was as high as 8% in Dabinett and Tydeman.Late in 2015, mortality was 8% in Cline Russet, Michelin, and Breakwell in 2017, and ranged from 8% to 24% in Yarlington Mill and Michelin, respectively, in 2021 (Fig. 3).All-cause tree mortality did not exceed 33% for any cultivar in 2020, and all but five cultivars (Cline Russet, Dabinett, Yarlington Mill, Michelin, and Tydeman Late) showed no tree mortality in the same year.In 2021, despite the application of pesticides to suppress and reduce the spread of E. amylovora (Table 1), FB symptoms began to appear in mid-June and continued to spread through the cultivar block.By the end of the growing season in November, 42% of all trees in the experiment (of the initial 560 trees) displayed FB symptoms sufficiently severe that the trees had or would succumb to E. amylovora bacteria.Although the cultivars were randomly located in each block, all blocks were affected, indicating the E. amylovora pressure appeared uniform throughout the block.There was a wide range but significant differences in cultivar tree mortality (P < 0.001) ranging from no tree mortality for Enterprise (n = 20 for all cultivars over four replications) and 100% tree morality for Ashmead's Kernel and Tolman Sweet.Overall, the incidence of FB symptoms was zero to very low for Enterprise, GoldRush, and Porter's Perfection (<20%); low for Binet Rouge, Kingston Black, Cline Russet, Dabinett, Grimes Golden, Frequin Rouge, Crimson Crisp , Cox Orange Pippin, and Muscadet De Dieppe (20%-40%); intermediate for Calville Blanc d'Hiver, Bramley's Seedling, Yarlington Mill, Michelin, Bulmers Norman, Stoke Red, Golden Russet, Breakwell, Esopus Spitzenberg (50%-90%), and high for Brown Snout, Medaille d'Or, Michelin, Brown's Apple, Sweet Alford, Tydeman Late, Ashmead's Kernel, and Tolman Sweet (90%-100%).The high variability in FB symptoms and tree mortality resulted in a relatively high threshold to detect significant  differences among cultivars, as indicated by the Tukey's mean separation in Fig. 3; a difference of >70% tree mortality was required to detect treatment differences of LS-means based using Tukey's HSP and α = 0.05.Notwithstanding, tolerance for E. amylovora infections in commercial orchards is extremely low.Even when low FB symptoms and/or tree mortality caused by E. amylovora exist, cider growers should be concerned about further tree mortality in subsequent years if the bacteria continue to overwinter and spread, which is likely at specific temperatures, humidity, and precipitation conditions.This study indicates that the majority of apple cultivars grown for cider in this study are susceptible to FB under ideal conditions for the causative agent, E. amylovora.
One must be careful when comparing the FB results of this study with those reported elsewhere.First, other studies which depend on natural epidemics and spread of infection may not have uniform inoculum throughout the orchard block.Second, there are several strains of E. amylovora that may differ in their transmission and virulence (Martínez-Bilbao et al. 2009).Third, apple trees and cultivars can have varying degrees of resistance, and often FB prevalence is recorded by incidence rather than by severity.Finally, many studies have been conducted by challenging the tips or leaves of actively growing shoots with specific strains of inoculation, in part to reduce the spread in orchards (Norelli et al. 1988;Harshman et al. 2017).Since this in vivo method evaluates the equivalent of secondary FB infection, the resistance response may differ from natural blossom blight.
Notwithstanding, several studies have reported the relative field resistance of apple cultivars, mostly culinary and crab apple cultivars, from orchard evaluations in natural epi-demics or following controlled inoculation (see van der Zwet and Keil 1979).There are limited data on FB susceptibility of cider apple cultivars; however, available data from several locations in Europe and North America are presented in Table 2.These data were largely consistent with those of the present study, with some exceptions.We observed that Brown's Apple, Michelin, and Stoke Red were susceptible to FB, while others have observed otherwise (Gwynne 1984;Byrde et al. 1986;Czynczyk et al. 2013).In contrast, we observed that Binet Rouge, Crimson Crisp , and Dabinett displayed varying degrees of FB resistance, whereas others reported these cultivars to be susceptible (Byrde et al. 1986;Paulin et al. 1993;Marin et al. 2017).
Controlling FB infection in orchards through improved pesticides or by improving plant resistance through genetic improvement are two approaches employed in the apple industry.For example, the Servicio Regional de Investigacion y Desarrollo Agroali-mentario in Asturias, Spain, has used Florina and Priscilla to develop new cider cultivars with lower FB susceptibility.Florina, which carries the Vf gene of resistance to apple scab (Venturia inaequalis), also shows very low susceptibility to FB.The l'Institut national de recherche pour l'agriculture, l'alimentation et l'environnement breeding program in Angers, France, has developed four cider apple hybrids named Raxina 8, Raxina 12, Raxina 16, and Raxina 30, which have shown little susceptibility to FB (Dapena and Blázquez 2004).With advances in genomics and plant breeding, the development of new cider apple cultivars with FB and additional disease and pest resistance appears to be promising.This study indicates that cider producers should take preventative measures by using more FB tolerant cultivars and resistant rootstocks when there is a history of FB within or adjacent to the planting location, or when environmental conditions are often ideal for the proliferation for E. amylovora.Many apple cider cultivars flower later than fresh-market culinary apples and have secondary bloom or protracted bloom periods (Plotkowski and Cline 2021a).These attributes increase the risk of FB incidence and severity (Gwynne 1984;Locke et al. 1993;Luby et al. 2002;Dupont 2019), indicating that selecting early flowering cultivars may be a best practice when growing cider-specific apple cultivars.

Environmental conditions
To illustrate the impact of environmental conditions on the prolific spread of E. amylovora in 2021 throughout this cider block, the daily minimum and maximum temperatures, relatively humidity, and precipitation from 1 May to 31 July are presented in Fig. 4.While "Cougarblight" and "Maryblyt" are two FB predicative models developed in North America that estimate the risk of FB infection during bloom, and hence infections' events for blossom blight (Smith and Pusey 2010;Dupont 2022), there are no FB models that predict E. amylovora infections post-bloom.It is unclear, and indeed surprising, why E. amylovora was so pronounced in 2021 and not in previous years.Streptomycin was applied during full bloom (10 May).Full bloom was protracted in 2021 and lasted approximately 7 days starting as early as 7 May for early flowering cultivars and 17 May for late flowering cultivars, which may have predisposed flowers to greater E. amylovora infection.There was no measurable rainfall between 9 and 27 May 2021 (Fig. 4), FB incidence has been historically low at the Simcoe Research Station even within adjacent trees of fresh-market FB susceptible cultivars such as Gala, Jonagold, and Fuji.Moreover, orchard management practices have been similar over the years, suggesting that migration from adjacent (within 500 m) commercial orchards with a known history of blossom flight, combined with high precipitation events on 26 June, 7 and 17 July and daily temperatures exceeding 27 • C, is the probable cause of the high incidence of FB shoot blight in this cider block.Erwinia amylovora has optimal growth at 27 • C (80 • F) and declines above 35 • C (Dupont 2019).Unfortunately, the post-bloom chemical control measures (Table 1) combined with repeated physical antiseptic removal of the affected tissue preventatively offered little benefit and cure, consistent with other reports in the literature (Anonymous 2022).The etiology and transmission of E. amylovora have been reported elsewhere (Thomson 2000;Zeng et al. 2021) and, based on extensive studies, E. amylovora transmission in orchards through rainfall and insect vectors is well recognized (Ordax et al. 2015).

Fig. 1 .
Fig. 1.Five year average bloom date (2017-2021) of select apple cultivars grown for cider at the University of Guelph Simcoe Research Station, Simcoe, ON.Error bars represent the standard deviation of the means.

Fig. 3 .
Fig. 3. All-cause tree mortality (percent) of select apple cultivars cultivars from 2015 (year of planting) to 2021 (year 7) grown for cider at the University of Guelph Simcoe Research Station, Simcoe, ON.Mean values represented by the vertical bar with the same letter are not significantly different according to the Tukey-Kramer's test at P = 0.05 (for 2021 data only).

Fig. 4 .
Fig. 4. Daily minimum (dashed line) and maximum (solid line), air temperature, and relative humidity (RH) and total daily precipitation at the University of Guelph, Simcoe (1 May-31 July 2021).

Table 1 .
Pesticides applied in 2017-2021 in an attempt to control E. amylovoa, the causal agent of fire blight (FB), in a 7 year old orchard of 28 apple cultivars grown on M.9 rootstock for cider production.
Note: University of Guelph, Simcoe, Ontario.* Pesticides were applied at the rate of 1000 L/ha, which equated to tree row volume dilute.
Five year average harvest date (2017-2021) of select apple cultivars grown for cider at the University of Guelph Simcoe Research Station, Simcoe, ON.Error bars represent the standard deviation of the means.