plants Article Susceptibility Evaluation to Fire Blight and Genome-Wide Associations within a Collection of Asturian Apple Accessions Belén García-Fernández 1, * , Ramon Dolcet-Sanjuan 2 , Diego Micheletti 3 , María José Antón-Díaz 1 , Cristina Solsona 2 , Mercedes Fernández 1 , Xavier Abad 4 and Enrique Dapena 1, * 1 Regional Service for Agrofood Research and Development (SERIDA), Ctra AS-267, PK 19, 33300 Villaviciosa, Spain; mjanton@serida.org (M.J.A.-D.); mercedfr@serida.org (M.F.) 2 IRTA Fruitcentre, PCiTAL, Parc AgroBiotech-Gardeny, 25003 Lleida, Spain; ramon.dolcet@irta.cat (R.D.-S.); cristina.solsona@irta.cat (C.S.) 3 Research and Innovation Center, Edmund Mach Foundation, 38098 San Michele all’Adige, Italy; diego.micheletti@fmach.it 4 IRTA-CReSA, Centre de Recerca en Sanitat Animal, 08193 Bellaterra, Spain; xavier.abad@irta.cat * Correspondence: bgarcia@serida.org (B.G.-F.); edapena@serida.org (E.D.) Abstract: Fire blight, caused by Erwinia amylovora, is one of the most devastating apple diseases. The selection of cultivars of low susceptibility and the study of the genetic mechanisms of the disease play important roles in fire blight management. The susceptibility level to fire blight was evaluated in 102 accessions originating from Asturias, a cider-producing region located in the north of Spain with a wide apple germplasm. Evaluations took place under quarantine conditions using artificial inoculations of grafted plants. The results revealed wide variation in susceptibility responses and low- susceptible cultivars were identified. In addition, 91 cultivars were genotyped using the Affymetrix Axiom® Apple 480 K SNP array to conduct genome-wide association studies (GWAS). A statistically significant signal was detected on chromosome 10 using the multi-locus mixed model (MLMM). Two genes were identified as major putative candidate genes: a TIR-NBS-LRR class disease protein and a protein containing a development and cell death (DCD) domain. The outcomes of this study provide Citation: García-Fernández, B.; a promising source of information, particularly in the context of cider apples, and set a starting point Dolcet-Sanjuan, R.; Micheletti, D.; for future genetic and breeding approaches. Antón-Díaz, M.J.; Solsona, C.; Fernández, M.; Abad, X.; Dapena, E. Keywords: Malus domestica; Erwinia amylovora; cider apple; apple bank germplasm; phenotypic Susceptibility Evaluation to Fire evaluation; local cultivars; artificial inoculations; genome-wide association studies; apple breeding; Blight and Genome-Wide putative candidate genes Associations within a Collection of Asturian Apple Accessions. Plants 2023, 12, 4068. https://doi.org/ 10.3390/plants12234068 1. Introduction Academic Editor: Juan A. Salazar The domesticated apple (Malus domestica Borkh.) is one of the most widely cultivated Received: 9 November 2023 tree fruits in temperate regions. Apple trees can be easily propagated and they bear Revised: 29 November 2023 quality fruit, which is considered one of the top functional foods [1,2]. Accordingly, global Accepted: 29 November 2023 production for the year 2021 was estimated at 93 million metric tons [3]. Of the more than Published: 4 December 2023 500,000 tons produced in Spain that year, approximately 15% were cider apples and 85% were dessert ones [4]. The Gram-negative bacterium Erwinia amylovora [5,6] is the causative agent of fire Copyright: © 2023 by the authors. blight, which is one of the most damaging apple diseases. Since its detection in the Licensee MDPI, Basel, Switzerland. United States more than 200 years ago [7], it has been detected in many parts of the This article is an open access article world. The consequences of fire blight infections can be devastating to apple orchards, distributed under the terms and as a severe outbreak can disrupt their production for years [8]. In 2007, European apple conditions of the Creative Commons production suffered particularly severe economic losses due to elevated inoculum pressure Attribution (CC BY) license (https:// and favorable warm temperatures during the flowering period [9]. creativecommons.org/licenses/by/ The control of fire blight is a challenge due to the number of tissues that are susceptible 4.0/). to infection by the bacteria and the reduced number of management tools to control it [10]. Plants 2023, 12, 4068. https://doi.org/10.3390/plants12234068 https://www.mdpi.com/journal/plants Plants 2023, 12, 4068 2 of 13 Moreover, many of the varieties that were introduced in the last few decades (e.g., ‘Fuji’, ‘Gala’, ‘Ginger Gold’ or ‘Pink Lady’) are more susceptible to the bacteria than most older cultivars [10]. The application of antibiotics, mainly streptomycin, was one the most effective chemical control against E. amylovora [10,11]. However, the identification of streptomycin-resistant strains has reduced its effectiveness [10–17]. On the other hand, due to the risks of using antibiotics, their application was banned in many countries for sustainability and consumer-friendly issues [18]. In this sense, breeding for host resistance is an essential component of the integrated fire blight management strategy [19]. Several screenings have been performed in Malus spp. to determine host suscepti- bility of different accessions [20–25]. These evaluations are usually carried out by field assessments or through artificial inoculations under controlled conditions in areas where the bacteria is widely spread [26]. Nevertheless, they must be conducted under quaran- tine conditions and, consequently, through artificial inoculations in areas categorized as protected zones due to the quarantine status of the disease [26,27]. In this situation, the most commonly used inoculation methods are based on artificial terminal shoot or leaf infection [25]. There are also numerous works focused on unraveling the genetic basis of fire blight resistance. QTL mapping approaches using wild species revealed loci associated with a strong resistance located on linkage group 3 (LG3; LG number also corresponds to chromosome number) of Malus × robusta 5, on LG10 of Malus fusca and on LG12 of Malus ‘Evereste’, Malus floribunda 821 and Malus × arnoldiana [28–31]. Regarding domesticated cultivars, the most important one corresponds to a major QTL located on LG7 of the cultivar ‘Fiesta’, which was reported by two independent studies developed by Calenge et al. [32] and Khan et al. [33]. A complementary study carried out by Khan et al. validated this QTL and checked its application in marker-assisted selection [34]. Furthermore, other QTLs have also been reported in different QTL mapping studies [32,35–38]. Additionally, a recent genome-wide association study (GWAS) developed by Thapa et al. [39] also reported significant marker-trait associations by rating both shoot and blossom fire blight infection severity. In addition to being the Spanish region with the highest cider production and a large cider-making tradition [40,41], Asturias, located in the north of the country, also has a rich apple diversity, especially of cider cultivars [42,43]. The majority of this diversity is represented by the more than 500 local accessions that are preserved in the Apple Germplasm Bank of Asturias, maintained by the SERIDA (Regional Service for Agrofood Research and Development, Villaviciosa, Spain). The characterization of this plant material has enabled the selection of cultivars of interest for the cider sector [44–50]. Indeed, out of the 76 varieties included in the Protected Designation of Origin (PDO) “Sidra de Asturias”, 58 are Asturian cultivars and 18 are selections from the cider-apple breeding program developed by the Fruit Research Unit of the SERIDA [51–54]. Favored by wet and mild springs, Asturias remains free of fire blight and maintains the status of a protected zone [55]. Despite this, it is important to make efforts in the evaluation, selection and breeding of local apple cultivars as prevention and management strategies against this disease. Taking this context into account, the goals of this work were to evaluate the suscepti- bility level against E. amylovora among local accessions and to conduct GWAS in order to determine phenotype–genotype associations. 2. Results 2.1. Phenotypic Evaluation 2.1.1. Pathogenicity Test ANOVA did not reveal differences in virulence among isolates (F2,69 = 1.02; p-value = 0.31). Nonetheless, the isolate FB74 was selected for the phenotypic evaluation since it showed the highest mean area under the disease progress curve (AUDPC) value when compared to FB45 and EW07 (Table 1). Plants 2023, 12, x FOR PEER REVIEW 3 of 14 showed the highest mean area under the disease progress curve (AUDPC) value when Plants 2023, 12, 4068 3 of 13 compared to FB45 and EW07 (Table 1). Table 1. Isolate name, origin and area under disease progress curve (AUDPC) values of the Erwinia amylovora Table isolates. 1. Isolate name,The sameand origin letter areanext to AUDPC under values (a) disease progress means curve non-significant (AUDPC) values of differences the Erwinia based on the ANOVA analysis (p-value < 0.05). amylovora isolates. The same letter next to AUDPC values (a) means non-significant differences based on the ANOVA analysis (p-value < 0.05). Isolate Origin AUDPC Isolate Department of Engineering and Agricultural Science Origin AUDPC EW07 5.77 a (University of León, León, Spain) Department of Engineering and Agricultural Science EW07 IRTA—CReSA 5.77 a (University of León, León, Spain) FB45 (Campus of the Autonomous IRTA—CReSA University of Barcelona, 6.89 a FB45 (Campus of the Autonomous University Spain) Cerdanyola del Vallès, Cataluña, of Barcelona, 6.89 a Cerdanyola del Vallès, Cataluña, Spain) IRTA—CReSA IRTA—CReSA FB74 FB74 (Campus of the (Campus Autonomous of the AutonomousUniversity ofBarcelona, University of Barcelona, 7.54 7.54 a a Cerdanyola Cerdanyola del delVallès, Vallès,Cataluña, Spain) Cataluña, Spain) 2.1.2. Evaluation 2.1.2. Evaluation of of Plant Plant Material Material Susceptibility Susceptibility Figure11shows Figure showsthethedistribution distributionofof the the accessions accessions according according to their to their bestbest linear linear unbi- unbiased ased prediction (BLUP) ± standard deviation. They were ordered from the lowest prediction (BLUP) ± standard deviation. They were ordered from the lowest to the highest to the highest BLUP BLUP value; invalue; other in otherfrom words, words, thefrom thetolowest lowest to susceptibility highest highest susceptibility level. level. Figure Figure 1. Distributionofof 1. Distribution best best linear linear unbiased unbiased prediction prediction (BLUP) (BLUP) values values ± standard ± standard deviation. deviation. Acces- Accessions sions were ranked in increasing order, from the lowest to the highest susceptibility levels. The The were ranked in increasing order, from the lowest to the highest susceptibility levels. or- orange column ange column represents represents ‘Florina’ ‘Florina’ (FLO) (FLO) and and thethe yellow yellow columns columns correspond correspond to ‘Lagar’ to ‘Lagar’ (LG)(LG) and and ‘No ‘No Raxao’ Raxao’ (NRX) (NRX) (identical (identical genotypes). genotypes). ‘Florina’ ‘Florina’ (FLO), (FLO), international control (see international control (see Section Section 5.1.2.), 5.1.2.), was was represented representedin inorange. orange. ‘No ‘NoRaxao’ Raxao’(NRX) (NRX)and and‘Lagar’ (LG) ‘Lagar’ (LG)(identical genotypes) (identical genotypes) [56][56] hadhad BLUP BLUPvalues of 0.43 values and of 0.43 0.28, respectively and 0.28, (represented respectively in yellow (represented in Figure in yellow 1). 1). in Figure Phenotypic Phenotypicevaluation evaluationrevealed revealed aa wide wide variation variation in host responses. in host responses. Focusing Focusing onon those those cultivars included in the PDO, low susceptibility (BLUP < − 4) were detected cultivars included in the PDO, low susceptibility (BLUP < −4) were detected in varieties in varieties such such asas ‘Carrandona’, ‘Arbeya’,‘Durona ‘Carrandona’, ‘Arbeya’, ‘Duronadede Tresali’, Tresali’, ‘Xuanina’ ‘Xuanina’ andand ‘Lin’. ‘Lin’. At other At the the other end end of the distribution (BLUP > 3), highly susceptible cultivars like ‘Durón of the distribution (BLUP > 3), highly susceptible cultivars like ‘Durón Encarnado’, ‘San Encarnado’, ‘San Roqueña’, Roqueña’, ‘Limón ‘Limón Montés’, Montés, ‘Perico’, ‘Perico’, ‘Fuentes’, ‘Fuentes’, ‘Regona’ ‘Regona’ or ‘Meana’ or ‘Meana’ werewere found. found. On theOn the other other hand,hand, ‘Panquerina’, ‘Panquerina’, ‘Durón‘Durón de Arroes’, de Arroes’, ‘Solarina’, ‘Solarina’, ‘Repinaldo ‘Repinaldo de Hueso’ de Hueso’ and and ‘Ernes- ‘Ernestina’ are examples of cultivars ranked as moderately susceptible (BLUP tina’ are examples of cultivars ranked as moderately susceptible (BLUP from −0.5 to 0.5). from − 0.5 to 0.5). BLUP values, as well as the number of replicates that were evaluated per accession, are compiled in Table S1. BLUP values, as well as the number of replicates that were evaluated per accession, are Plants 2023, 12, 4068 compiled in Table S1. 4 of 13 2.2. Genotyping and Population Genetic Structure 2.2.All the diploid Genotyping and genotypes hybridized Population Genetic (90 local + ‘Florina’; Table S1) passed the quality Structure controls All the diploid genotypes hybridized (90addition, of the Axiom™ Analysis Suite. In 243,495Table local + ‘Florina’; SNPs S1)were classified passed as batch the quality alle consistency controls (BAC)-passed, of the Axiom™ Analysis PolyHighResolution Suite. In addition, 243,495andSNPs very wererobust. Of as classified these, batch28,954 alle were SNPs consistency (BAC)-passed, discarded because ofPolyHighResolution and very (MAF) a minor allele frequency robust. Of these, lower 28,954 than 5%.SNPs Genotyp- were discarded because of a minor allele frequency (MAF) lower than ing resulted in a final set of 91 genotypes and 214,541 SNPs to implement phenotype– 5%. Genotyping resulted associations. genotype in a final set of 91 genotypes and 214,541 SNPs to implement phenotype–genotype associations. Population structure analyses are plotted in Figure 2. Principal component analysis Population structure analyses are plotted in Figure 2. Principal component analysis (PCA) did not reveal separate clusters (Figure 2a). The amount of variance explained by (PCA) did not reveal separate clusters (Figure 2a). The amount of variance explained by thethe first firsttwo twocomponents components (PC1(PC1andand PC2) PC2) waswas14%14% andand 9%, respectively. 9%, respectively. BayesianBayesian cluster- clustering ing(Figure (Figure 2b) also reflects high admixture within the membership coefficients (Q), even when even 2b) also reflects high admixture within the membership coefficients (Q), when the number the number of subpopulations of subpopulations was twowas(k = two (k = 2). Similarly, 2). Similarly, the STRUCTUREthe STRUCTURE HARVESTERHAR- VESTER revealed revealed theDelta the highest K value highest K kvalue Deltafor forin-depth = 2. An k = 2. Anstudy in-depth of thestudy of thegenetic population population structure discarded strong clustering in the collection. genetic structure discarded strong clustering in the collection. Figure Figure 2. 2. Populationgenetic Population genetic structure structure analyses. analyses.(a)(a) Principal Component Principal ComponentAnalysis (PCA). Analysis Principal (PCA). Principal Component 1 (PC1) was plotted against Principal Component 2 (PC2). Component 1 (PC1) was plotted against Principal Component 2 (PC2). The percentageThe percentage of variance of variance explained explained bybyeach eachone oneis is shown shown ininparentheses; parentheses; (b)(b) Distribution of the Distribution of genotypes after dividing the genotypes the after dividing the collection into 2, 3 and 4 subpopulations collection into 2, 3 and 4 subpopulations (k). (k). 2.3. Phenotype–Genotype Associations 2.3. Phenotype–Genotype Associations A significant signal was detected using the multi-locus mixed model (MLMM) at theAbottom significant signal was 10 of chromosome detected (Figureusing the significant 3). The multi-locusmarker mixedwasmodel (MLMM) at the AX-115639581 bottom (p-value 9.92 × 10−9 ; −10 of =chromosome log(Figure 10 (p-value3).= The 8.00),significant marker was10AX-115639581 located on chromosome (p-value position 32,827,282 of the ‘Golden Delicious’ doubled-haploid genome (GDDH13) v. 1.1. = 9.92 × 10 ; −log10(p-value = 8.00), located on chromosome 10 position 32,827,282 −9 Even though noof the markers exceeded the significance threshold of 6.6 using the mixed linear ‘Golden Delicious’ doubled-haploid genome (GDDH13) v. 1.1. Even though no markers model (MLM), a convincing peak was identified in the same window region as MLMM. Indeed, the exceeded the significance threshold of 6.6 using the mixed linear model (MLM), a con- −log (p-value) of AX-115639581 in MLM was 5.9. For both MLM and MLMM, the Q-Q vincing10peak was identified in the same window region as MLMM. Indeed, the −log10(p- plot indicated a good fit of the model, as shown in Figure 3. value) of AX-115639581 in MLM was 5.9. For both MLM and MLMM, the Q-Q plot indi- cated a good fit of the model, as shown in Figure 3. Plants Plants 2023, 2023, 12, 12, x4068 FOR PEER REVIEW 55of of14 13 Plants 2023, 12, x FOR PEER REVIEW 5 of 14 Figure 3. Figure 3. Manhattanplots plots andcorresponding corresponding Q-Q plots plots for for the the mixed mixed linear linear model model (MLM) (MLM) and and the the Figure 3.Manhattan Manhattan plots and and corresponding Q-Q Q-Q plots for the mixed linear model (MLM) and the multiple multiple loci mixed mixed model model (MLMM) (MLMM) after after conducting conducting genome-wide genome-wide association association multiple loci mixed model (MLMM) after conducting genome-wide association analysis (GWAS) analysis analysis (GWAS) (GWAS) for for firefire blight. Different chromosomes are represented by different colors. The horizontal for fire blight. Different chromosomes are represented by different colors. The horizontal green linein blight. Different chromosomes are represented by different colors. The horizontal green green line line ininthe the the Manhattan Manhattan Manhattan plots plots plots indicates thethe indicates indicates the threshold at aat threshold threshold aa significant significant atsignificant level level level of of 5%of 5% 5% after after after Bonferroni Bonferroni Bonferroni multiple multiple multiple test test testcorrection. correction. Q-Q Q-Q plots plots compare compare the the expected expected value for value for the the −log (p-value) vs. −log1010(p-value) vs.the theobserved observedoneone correction. Q-Q plots compare the expected value for the −log10 (p-value) vs. the observed one for for foreach each SNP (blue point). The red line indicates aa hypothetical hypothetical distribution distributionwithout withoutassociation. association. each SNPSNP (blue (blue point). point). TheTheredred lineline indicates indicates a hypothetical distribution without association. Considering Considering the Considering the MLMM, MLMM, the MLMM, the effect the effect and effect and the and the percentage the percentage percentageof of phenotypic ofphenotypic variance phenotypicvariance ex- varianceex- ex- plained plained plained (R (R ) 222 )by by AX-115639581 AX-115639581 were were 3.22 3.22 and and 55.61%, 55.61%, respectively. respectively. Figure Figure 4 ) by AX-115639581 were 3.22 and 55.61%, respectively. Figure 4 represents 4 represents represents the the scatter scatter the plot scatter of plotplotof theofBLUP the BLUP values values the BLUP grouped grouped values by allele by allele grouped (C/A)(C/A) (C/A) by allele of AX-115639581. of AX-115639581. Seventy Seventy of AX-115639581. geno- geno- Seventy types were types werewere genotypes homozygous homozygous homozygous CC, 19 CC,CC, were heterozygous 19 19 were heterozygous CA CACA and and and2 2 2were were homozygous werehomozygous homozygousAA AA AA (MAF = 0.13) (MAF == 0.13) (MAF (Table S1). The ANOVA 0.13) (Table S1). The ANOVA test revealed significant ANOVA test revealedsignificant differences significantdifferences among differencesamong means amongmeans means (red (redrhombus; (red rhombus; p-value rhombus; p-value << 0.001), p-value 0.001), although 0.001), although aa wide phenotypic wide phenotypic variability phenotypicvariability can variabilitycan canbebe observed beobserved observed within withinthe within theheterozygous the heterozygous and heterozygous and homozygous homozygous CC groups groups (Figure (Figure 4). (Figure 4). 4). Figure 4. Scatter Figure 4. Scatter plot of the best linear unbiased unbiased prediction prediction (BLUP) (BLUP) results resultsgrouped groupedbybyallele allele(C/A) (C/A) Figure 4. Scatter plot of the best linear unbiased prediction (BLUP) results grouped by allele (C/A) of AX-115639581. AX-115639581. The Thegrey greyrhombuses rhombusesrepresent representeach each cultivar while the red ones represent the of AX-115639581. The grey rhombuses represent each cultivar while the red ones representmean of cultivar while the red ones represent the the mean of eachofgroup. each group. mean of each group. Plants 2023, 12, 4068 6 of 13 Of the five cultivars included in the PDO that showed low BLUP values (see Section 2.1.2), ‘Carrandona’ was homozygous AA, ‘Arbeya’ and ‘Xuanina’ were heterozygous CA, and ‘Durona de Tresali’ and ‘Lin’ were the other homozygous CC. 3. Discussion In accordance with the global context of fostering sustainability, the evaluation, se- lection and use of low-susceptible apple cultivars represents an important advance in fire blight management. In this sense, evaluation using grafted plants was chosen despite requiring more space since it has been reported to show more reproducibility, less variability and high resolution compared to detached leaves [22,57,58]. Moreover, Ruz et al. [26] found, while working in grafted pear plants, that the most effective inoculation method was cutting with scissors compared to pricking with clamps, local infiltration and painting onto the leaves. Despite having used the same inoculum concentration, the same inoculation pro- cedure, and trying to adjust the growth rates among replicates, the standard deviation within the same genotypes was moderate, which highlights the importance of keeping homogeneous conditions while evaluating fire blight infection. In this case, the observed results for ‘Florina’ fell between those reported by Le Roux et al. [35] (moderate suscep- tibility) and those described by Le Lezec et al. [59] (low susceptibility). Furthermore, the strong similarity between the two genetically identical accessions, ‘No Raxao’ and ‘Lagar’, provides robustness to the assessment. The phenotypic evaluation carried out in this study allowed the assessment of the susceptibility to fire blight in 102 Asturian-local accessions, leading to the identification of five low-susceptible accessions that included cultivars protected by PDO “Sidra de Asturias”. Indeed, two of them, ‘Durona de Tresali’ and ‘Xuanina’, constitute 20.5% of the total production of PDO [60]. Nonetheless, it is important to note that different strains of E. amylovora have been found to exhibit varying degrees of virulence and cultivars have shown differential suscep- tibility to these bacterial strains, as has been reported for Malus × robusta 5 [10,22,37,61–65]. The majority of the cultivars included in PDO that were previously analyzed (e.g., ‘Xu- anina’, ‘Durona de Tresali’, ‘Regona’, ‘Meana’ or ‘San Roqueña’) [66] showed consistent results between both evaluations. However, ‘Perico’ exhibited a different response level, which could be caused by this strain-dependent susceptibility. Specifically, ‘Perico’ was of low susceptibility, whereas in this study it showed a high infection level. In light of these findings, it is wise, as suggested by Martínez-Bilbao et al. [22], to continue the evaluation of cultivars by testing them against different strains. This recommendation, along with the use of multiple replicates due to variability, is also endorsed in the present study. Chromosome 10 has been associated with fire blight resistance in previous studies. In particular, Thapa et al. [39] identified QTLs in the middle region of LG10, but they are distinct from the region identified here. Emeriewen et al. [67] aligned the predicted sequence of the FB_Mfu10 gene [18,30,68] against the GDDH13 genome, revealing the best match in the region between 30,708,564 and 30,712,095, also on LG10. The marker AX-115639581 is located approximately 2 Mbp downstream from this region, which points to a different source of resistance. After exploring the window region of the marker, two putative-candidate genes were found: MD10G1232100, annotated as development and cell death domain protein (DCD); and MD10G1232400, annotated as TIR-NBS-LRR class disease protein. Both are located downstream of the marker, 12 Kbp and 29 Kbp, respectively [69]. TIR-NBS-LRR consists of a leucine-rich repeat domain (LRR), a nucleotide binding site (NBS) and an amino-terminal domain with homology to the Toll and interleukin 1 receptors (TIR) and is one of the major classes of plant resistance genes (R-genes) [70]. Overexpression of genes encoding TIR-NBS-LRR was detected in transcriptome analyses during apple infections with E. amylovora [39,71,72]. On the other hand, the DCD domain, which is composed of approximately 130 amino acid residues, can be found throughout the plant Plants 2023, 12, 4068 7 of 13 kingdom [73]. It seems that DCD could mediate signaling in plant programmed cell death by pathogens [73]. Since the MLM emerged [74], MLM-based GWAS have been widely conducted. How- ever, multi-locus models are better models when working with polygenic traits [75,76], the case of this study. Furthermore, the lack of statistical significance observed in the MLM could be partially explained by the reduced sample size or by the correction for multiple testing, which is often too conservative [76–78]. On the other hand, unexpected results between phenotype and genotype (Figure 4) can be due to recombination events in the region or strain-specific resistance. 4. Conclusions Deciphering the genomic basis of quantitative variation is crucial to understanding evolution and accelerating plant breeding [79]. Although the potential impact of strain- specific resistance and the increase in sample size should be considered for further analyses, this study sets a starting point for future genetic approaches. The identification of a genomic region associated with fire blight can be used to perform a marker-selection approach, which, in turn, can decrease both the time and money required for developing and selecting desirable genotypes [80]. Moreover, association mapping using populations of unrelated individuals can be combined with QTL analysis in bi-parental and multi-parental crosses to reduce the confidence interval of the associated regions, and complementing studies such as gene editing could validate the role of the putative candidate genes during fire blight infections [81–83]. In conclusion, the findings of this study not only offer promising information but also serve as a valuable resource for the selection of optimal cultivars for apple production and for their integration into breeding programs, especially within cider apple growing. 5. Materials and Methods 5.1. Phenotypic Evaluation 5.1.1. Plant Material and Culture Conditions Susceptibility was evaluated on potted plants in 103 accessions preserved in the Apple Germplasm Bank of Asturias (Table S1). Of them, 42 local cultivars are included in the PDO “Sidra de Asturias”, 60 accessions are representative of the local diversity and the last one corresponds to the international cultivar ‘Florina’. Phenotypic evaluation was performed by inoculation of growing shoots on potted plants over two years: 2019 and 2021. For each year, eight 1-year-old shoots per accession were whip-grafted onto ‘M7’ EMLA apple rootstock and potted on 770 mL containers with substrate Exclusive (Gebr.Brill Substrate GmbH & Co. KG, Georgsdorf, Germany). Dormant plants were placed in a cold greenhouse until they reached a length of 5 cm. Then, they were transferred to a cold chamber set at 14 ◦ C and 16/8 h light/dark photoperiod (200 µmol·m−2 ·s−1 ), when they continued to grow until they were moved to the high- biocontainment box (BSL3) of the IRTA—CReSA (Campus of the Autonomous University of Barcelona, Cerdanyola del Vallès, Cataluña, Spain). Plants that failed to achieve a growth of 10 cm were not moved to the high-biocontainment box and they were not evaluated. Moreover, the maximum number of replicates per ac- cession and year was set at six and the minimum number at three, considering those accessions with fewer than three valid replicates unsuccessful for that year. Plant material was inoculated one week after moving to the high-biocontainment box, where culture conditions were set at 23 ◦ C, high relative humidity (≈100%) and 16/8 h photoperiod with 300 mmol·m−2 ·s−1 of photosynthetically active radiation (PAR) (Figure S1). Plants were fertilized with Proturf® Active 15-5-15 (ICL Ibéria, Súria, Spain) and watered every two days with tap water at soil level, avoiding shoot or leaf watering. All plant material was evaluated in a single batch in 2019, and it was divided into two batches due to different growth rates in 2021. Plant material was randomly distributed by blocks in the three batches. Plants 2023, 12, x FOR PEER REVIEW 8 of 14 All plant material was evaluated in a single batch in 2019, and it was divided into Plants 2023, 12, 4068 two batches due to different growth rates in 2021. Plant material was randomly distributed 8 of 13 by blocks in the three batches. 5.1.2. Pathogenicity Test 5.1.2. Pathogenicity Test Pathogenicity was performed on three different isolates identified as E. amylovora by Pathogenicity nucleic was acid analysis ofperformed the 1100 pbon three16S rDNA different isolates fragment identified at the as E. Laboratory ofamylovora by Instrumental nucleic acid analysis of the 1100 pb rDNA 16S fragment at the Laboratory of Instrumental Techniques of the University of León. FB74 and FB45 were isolated from an infected apple Techniques of the University of León. FB74 and FB45 were isolated from an infected apple orchard in Catalonia (Spain) and stored at the IRTA—CreSA. EW07 was provided by the orchard in Catalonia (Spain) and stored at the IRTA—CreSA. EW07 was provided by the Department of Engineering and Agricultural Science (University of León, León, Spain). Department of Engineering and Agricultural Science (University of León, León, Spain). Eight apple cultivars, ‘Florina’ and seven local ones, were used to test the pathogenic- Eight apple cultivars, ‘Florina’ and seven local ones, were used to test the pathogenicity ity of the isolates. The local cultivars ‘Collaos’, ‘De la Riega’, ‘Limón Montés’, ‘Meana’, of the isolates. The local cultivars ‘Collaos’, ‘De la Riega’, ‘Limón Montés’, ‘Meana’, ‘Regona’, ‘Solarina’ and ‘Xuanina’ were selected because they previously showed differ- ‘Regona’, ‘Solarina’ and ‘Xuanina’ were selected because they previously showed different ent levels of susceptibility against the bacteria [66] and ‘Florina’ was used as a control for levels of susceptibility against the bacteria [66] and ‘Florina’ was used as a control for low-moderate susceptibility [35,59]. All of them, along with two genetically identical in- low-moderate susceptibility [35,59]. All of them, along with two genetically identical dividuals (‘No Raxao’ and ‘Lagar’) [56], were also used as reference in the phenotypic individuals (‘No Raxao’ and ‘Lagar’) [56], were also used as reference in the phenotypic evaluation of the collection analyzed. evaluation of the collection analyzed. Thelevel The levelofofpathogenicity pathogenicitywas wasassessed assessedfollowing followingthethesame sameprotocol protocoldescribed describedabove above for culture conditions and the same workflow described below for inoculum for culture conditions and the same workflow described below for inoculum preparation,preparation, inoculationand inoculation andevaluation. evaluation. 5.1.3.Inoculum 5.1.3. InoculumPreparation Preparationand andArtificial ArtificialInoculation Inoculation IsolateFB74 Isolate FB74stored storedatat−−80 ◦ 80 °C was sown C was sown using using the the three-phase three-phase streaking streaking pattern pattern (T- (T- streak)in streak) infresh fresh LB LB media media and incubated incubated at at2525°C◦ Cfor for4848h.h.A A suspension suspension of of 108108 cfu/mL cfu/mL was prepared was by resuspending prepared by resuspending the the isolated colonies isolated colonies in sterile water. in sterile Adjustment water. Adjustment waswas per- formed using performed usinga viable absorbance a viable absorbance relationship relationshippreviously previouslyprepared prepared(absorbance (absorbance==0.1 0.1atatλ λ==620 620nm)nm)[84]. [84].The Theinoculum inoculumwas was prepared prepared two hours before before use use and andwas storedatat44◦ °C wasstored C untilthen. until then. Inoculationwas Inoculation wasperformed performed byby cutting cutting thethe firstfirst completely completely unfolded unfolded leafleaf using using scis- scissors previously immersed sors previously in the bacterial immersed suspension in the bacterial [61]. Two suspension cutsTwo [61]. per leaf cutswere performed per leaf were per-at different formed at positions different and on eachand positions sideonofeach the mid-rip, side of the taking care not mid-rip, to break taking care notit. to break it. 5.1.4. 5.1.4.Evaluation Evaluationof ofPlant PlantMaterial MaterialSusceptibility Susceptibility Susceptibility Susceptibilitywaswasrecorded recorded7,7,14 14and and21 21days dayspost-inoculation post-inoculationby bydividing dividingnecrosis necrosis progression by shoot length. Necrosis progression was measured using progression by shoot length. Necrosis progression was measured using the scale de- the scale described by Calenge scribed by et al. [32]et(Figure Calenge al. [32]5): (i) 0—no (Figure visible 5): (i) 0—no symptoms; (ii) 0.5—necrosis visible symptoms; only affected (ii) 0.5—necrosis only veins of the inoculated leaf; (iii) 1—necrosis reached the petiole of the inoculated affected veins of the inoculated leaf; (iii) 1—necrosis reached the petiole of the inoculated leaf; (iv) 1+ leaf; necrosis (iv) length—necrosis 1 + necrosis length—necrosis reached the stem. reached the stem. Figure5.5.Representative Figure Representativeexamples examplesofofthe thescale scaleused usedtotomeasure measurenecrosis necrosisprogression: progression:0—no 0—novisible visible symptoms; 0.5—necrosis only affected veins of the inoculated leaf; 1—necrosis reached symptoms; 0.5—necrosis only affected veins of the inoculated leaf; 1—necrosis reached the the petiole of petiole the inoculated leaf; 1 + necrosis length—necrosis reached the stem of the plant. of the inoculated leaf; 1 + necrosis length—necrosis reached the stem of the plant. 5.2. Genotyping To generate the genotypic data, the genomic DNA of 91 diploid accessions (90 local + ‘Florina’; Table S1) was hybridized on the Affymetrix Axiom® Apple 480 K SNP array [85] following the manufacturer’s protocol at the genotyping platform of FEM (Fondazione Edmund Mach, San Michele all’Adige, Italy). Raw data were processed with the “Best Plants 2023, 12, 4068 9 of 13 Practices” workflow for the diploid clustering method in the Axiom™ Analysis Suite v. 5.2. A BAC analysis was performed to correct for eventual batch effects. Of the set of SNPs that were classified as PolyHighResolution by the Axiom™ Analysis Suite and passed the BAC analysis, those defined as very robust markers by Bianco et al. [85] were extracted and filtered by a MAF threshold of 0.05 using PLINK v. 1.9 [86]. The physical position of the SNPs was determined by blasting the probes against the reference genome GDDH13 v. 1.1 [87] and assigning the ambiguous-positioned markers to a fictive chromosome 0. Population Genetic Structure The population structure was studied using both PCA and Bayesian clustering. PCA was calculated using the PLINK software, which was also used to prune the set of SNPs using the “indep” option. Pruning was performed by setting a window size of 1 Mbp, a step size of 10 and linkage disequilibrium (LD) measured as r2 below 0.5. The subset of LD-pruned markers was used in the software STRUCTURE v. 2.2 [88], setting a burn-in length of 10,000, a run length of 50,000 and ranging the number of subpopulations (k) from 1 to 6. After running seven replicates for each k, STRUCTURE HARVESTER v. A.2 [89] was used to estimate the most probable number of k. 5.3. Statistical Analyses and Phenotype–Genotype Associations Susceptibility data recorded at 7, 14 and 21 days post inoculation were used to calculate AUDPC values in the ‘agricolae’ package v. 1.3-6 [90]. Pathogenicity differences were analyzed by performing a one-way ANOVA analysis at a significance level of 5% with the AUDPC values. Phenotypic results were adjusted by fitting a mixed linear model using the ‘lme4’ package v. 1.1-34 [91]. “Accession” and “batch” were considered random and fixed effects, respectively. BLUP values of the random effects were also extracted using the ‘lme4’ package. Phenotype–genotype associations were conducted using the MLM and MLMM meth- ods implemented in the ‘GAPIT’ package v. 3.1.0 [92], setting BLUP values as phenotypic- input data. Both the kinship matrix and the first three principal components (PCs) were in- cluded in the association model. The kinship matrix was calculated in TASSEL v. 5.2.86 [93] using the “Centered_IBS” method [94]. After applying the Bonferroni multiple test correc- tion, the -log10 threshold for significant associations was set to 6.6 (p-value = 2.33 × 10−7 ) at a significant level of 5%. Genes within windows of 100 Kbp in both directions of a significant marker were explored using the JBrowser tool hosted on the Genome Database for Rosaceae (GDR) website [69]. All analyses were implemented in the R software environment v. 4.3.1 [95] and the ‘ggplot2’ package v. 3.4.2 [96] was used to visualize the PCA analysis and BLUP results grouped by allele of AX-115639581. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/plants12234068/s1, Table S1: Compilation of the plant material information and phenotypic results. The name, protection status from the Protected Designation of Origin (PDO) “Sidra de Asturias” and if they were used to implement phenotype–genotype associations is reported for each accession. The number of replicates evaluated, best linear unbiased prediction (BLUP) value and genotypic information for marker AX-115639581 is also reported for each one; Figure S1: Image of potted plants 21 days after inoculation in the culture conditions of the high-biocontainment box (BSL3) at the IRTA—CReSA. Author Contributions: E.D. selected the set of apple accessions. E.D. and R.D.-S. designed the experiment with the assistance of D.M. and B.G.-F., M.J.A.-D. and M.F. collected the plant material. R.D.-S., C.S. and X.A. grafted the shoots, performed the pathogenicity test, prepared inoculum, and performed the inoculation and phenotypic evaluation. B.G.-F., R.D.-S. and E.D. organized the dataset, performed the statistical analyses and interpreted the results of the phenotypic data. D.M. and B.G.-F. handled the software, analyzed the population structure and relatedness, implemented the genome-wide association study, and interpreted the results. B.G.-F. wrote the first draft of the Plants 2023, 12, 4068 10 of 13 manuscript. B.G.-F., E.D., R.D.-S. and D.M. revised and edited the first draft. All authors contributed to manuscript revision and read and approved the submitted version. E.D. and R.D.-S. acquired funding. All authors have read and agreed to the published version of the manuscript. Funding: Funding was provided by a Ph.D. student grant to B.G.-F. (grant PRE2018-086417 funded by MCIN/AEI/10.13039/501100011033 and by ESF Investing in your future), by MCIU-INIA-AEI through the project RTA2017-00102-C03-01 and RTA2017-00102-C03-02, and through UE Horizon- CL6-2021-BIODIV-01-12–320 EU InnOBreed IA project n◦ 101061028. Data Availability Statement: Data sharing is available upon request. Acknowledgments: The authors thank Simone Larger and Massimo Pindo (FEM) for their work in the extraction and genotyping of plant material and the Department of Engineering and Agricultural Science (University of León, Spain) who provided strain EW07. It is expected that the content of this article will be part of the doctoral dissertation of B.G.-F. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results. References 1. Pereira-Lorenzo, S.; Ramos-Cabrer, A.M.; Fischer, M. Breeding apple (Malus × domestica Borkh.). In Breeding Plantation Tree Crops: Temperate Species; Springer New York: New York, NY, USA, 2009; pp. 33–81, ISBN 978-0-387-71202-4. 2. Kumar, S.; Volz, R.K.; Chagné, D.; Gardiner, S. Breeding for apple (Malus × domestica Borkh.) fruit quality traits in the genomics era. In Genomics of Plant Genetic Resources; Tuberosa, R., Graner, A., Frison, E., Eds.; Springer Netherlands: Dordrecht, The Netherlands, 2014; pp. 387–416, ISBN 978-94-007-7574-9. 3. FAOSTAT Crops. Food and Agriculture Organization of the United Nations. Available online: https://www.fao.org/faostat/es/ #data (accessed on 22 October 2023). 4. MAPA Fruta de Pepita: Informe Final Campaña 2020–21. Sep 2020–Ago 2021. Ministerio de Agricultura, Pesca y Alimentación. Available online: https://www.mapa.gob.es/es/agricultura/temas/producciones-agricolas/frutas-y-hortalizas/Informacion_ subsectorial.aspx (accessed on 17 October 2023). 5. Burrill, T.J. New species of Micrococcus. Am. Nat. 1883, 17, 319. 6. Winslow, C.E.; Broadhurst, J.; Buchanan, R.E.; Krumwiede, C.; Rogers, L.A.; Smith, G.H. The families and genera of the bacteria: Final report of the committee of the society of American bacteriologists on characterization and classification of bacterial types. J. Bacteriol. 1920, 5, 191–229. [CrossRef] [PubMed] 7. Baker, K.F. Fire blight of pome fruits: The genesis of the concept that bacteria can be pathogenic to plants. Hilgardia 1971, 40, 603–633. [CrossRef] 8. Vanneste, J.L. What is fire blight? Who is Erwinia amylovora? How to control it? In Fire Blight: The Disease and Its Causative Agent, Erwinia amylovora; Vanneste, J.L., Ed.; CABI: Wallingford, UK, 2000; pp. 1–6, ISBN 9780851992945. 9. Sehic, J.; Nyboom, H.; Garkava-Gustavsson, L.; Patocchi, A.; Kellerhals, M.; Duffy, B. Fire blight (Erwinia amylovora) resistance in apple varieties associated with molecular markers. Int. J. Hortic. Sci. 2009, 15, 53–57. [CrossRef] 10. Norelli, J.L.; Jones, A.L.; Aldwinckle, H.S. Fire Blight Management in the Twenty-first Century: Using New Technologies that Enhance Host Resistance in Apple. Plant Dis. 2003, 87, 756–765. [CrossRef] [PubMed] 11. Turechek, W.W. Apple Diseases and their Management. In Diseases of Fruits and Vegetables; Naqvi, S.A.M.H., Ed.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2004; Volume 1, pp. 1–108, ISBN 1-4020-1822-3. 12. Chiou, C.-S. The Analysis of Plasmid-Mediated Streptomycin Resistance in Erwinia amylovora. Phytopathology 1991, 81, 710. [CrossRef] 13. Loper, J.E. Evaluation of Streptomycin, Oxytetracycline, and Copper Resistance of Erwinia amylovora Isolated from Pear Orchards in Washington State. Plant Dis. 1991, 75, 287. [CrossRef] 14. McManus, P.S.; Stockwell, V.O.; Sundin, G.W.; Jones, A.L. Antibiotic use in plant agriculture. Annu. Rev. Phytopathol. 2002, 40, 443–465. [CrossRef] 15. Russo, N.L.; Burr, T.J.; Breth, D.I.; Aldwinckle, H.S. Isolation of Streptomycin-Resistant Isolates of Erwinia amylovora in New York. Plant Dis. 2008, 92, 714–718. [CrossRef] 16. McGhee, G.C.; Guasco, J.; Bellomo, L.M.; Blumer-Schuette, S.E.; Shane, W.W.; Irish-Brown, A.; Sundin, G.W. Genetic analysis of streptomycin-resistant (Sm(R)) strains of Erwinia amylovora suggests that dissemination of two genotypes is responsible for the current distribution of Sm(R) E. amylovora in Michigan. Phytopathology 2011, 101, 182–191. [CrossRef] 17. Tancos, K.A.; Cox, K.D. Exploring Diversity and Origins of Streptomycin-Resistant Erwinia amylovora Isolates in New York Through CRISPR Spacer Arrays. Plant Dis. 2016, 100, 1307–1313. [CrossRef] [PubMed] 18. Emeriewen, O.F.; Malnoy, M.; Richter, K.; Kilian, A.; Hanke, M.V.; Peil, A. Evidence of a major QTL for fire blight resistance in the apple wild species Malus fusca. Acta Hortic. 2014, 1056, 289–293. [CrossRef] Plants 2023, 12, 4068 11 of 13 19. Lespinasse, Y.; Aldwinckle, H.S. Breeding for resistance to fire blight. In Fire Blight: The Disease and Its Causative Agent, Erwinia amylovora; Vanneste, J.L., Ed.; CABI Publishing: Wallingford, UK, 2000; pp. 253–273, ISBN 9780851992945. 20. Thomas, T.M. Severity of fire blight on apple cultivars and strains in Michigan. Plant Dis. 1992, 76, 1049. [CrossRef] 21. Tsiantos, J.; Psallidas, P. Fire blight resistance in various loquat, apple and pear cultivars and selections in Greece. J. Plant Pathol. 2004, 86, 227–232. 22. Martínez-Bilbao, A.; Ortiz-Barredo, A.; Montesinos, E.; Murillo, J. Evaluation of a Cider Apple Germplasm Collection of Local Cultivars from Spain for Resistance to Fire Blight (Erwinia amylovora) Using a Combination of Inoculation Assays on Leaves and Shoots. HortScience 2009, 44, 1223–1227. [CrossRef] 23. Kostick, S.A.; Norelli, J.L.; Evans, K.M. Novel metrics to classify fire blight resistance of 94 apple cultivars. Plant Pathol. 2019, 68, 985–996. [CrossRef] 24. Dougherty, L.; Wallis, A.; Cox, K.; Zhong, G.-Y.; Gutierrez, B. Phenotypic evaluation of fire blight outbreak in the USDA Malus collection. Agronomy 2021, 11, 144. [CrossRef] 25. Sobiczewski, P.; Keller-Przybyłkowicz, S.; Lewandowski, M.; Mikiciński, A.; Maciorowski, R. Phenotypic and marker-assisted characterization of new apple genotypes with high resistance to fire blight. Eur. J. Plant Pathol. 2021, 161, 49–61. [CrossRef] 26. Ruz, L.; Moragrega, C.; Montesinos, E. Evaluation of four whole-plant inoculation methods to analyze the pathogenicity of Erwinia amylovora under quarantine conditions. Int. Microbiol. 2008, 11, 111–119. 27. EPPO Erwinia amylovora. EPPO Datasheets on Pests Recommended for Regulation. Available online: https://gd.eppo.int (accessed on 28 March 2023). 28. Peil, A.; Garcia-Libreros, T.; Richter, K.; Trognitz, F.C.; Trognitz, B.; Hanke, M.V.; Flachowsky, H. Strong evidence for a fire blight resistance gene of Malus robusta located on linkage group 3. Plant Breed. 2007, 126, 470–475. [CrossRef] 29. Durel, C.E.; Denancé, C.; Brisset, M.N. Two distinct major QTL for resistance to fire blight co-localize on linkage group 12 in apple genotypes “Evereste” and Malus floribunda clone 821. Genome 2009, 52, 139–147. [CrossRef] [PubMed] 30. Emeriewen, O.F.; Richter, K.; Kilian, A.; Zini, E.; Hanke, M.-V.; Malnoy, M.; Peil, A. Identification of a major quantitative trait locus for resistance to fire blight in the wild apple species Malus fusca. Mol. Breed. 2014, 34, 407–419. [CrossRef] 31. Emeriewen, O.F.; Peil, A.; Richter, K.; Zini, E.; Hanke, M.-V.; Malnoy, M. Fire blight resistance of Malus × arnoldiana is controlled by a quantitative trait locus located at the distal end of linkage group 12. Eur. J. Plant Pathol. 2017, 148, 1011–1018. [CrossRef] 32. Calenge, F.; Drouet, D.; Denancé, C.; Van de Weg, W.E.; Brisset, M.N.; Paulin, J.P.; Durel, C.E. Identification of a major QTL together with several minor additive or epistatic QTLs for resistance to fire blight in apple in two related progenies. Theor. Appl. Genet. 2005, 111, 128–135. [CrossRef] [PubMed] 33. Khan, M.A.; Duffy, B.; Gessler, C.; Patocchi, A. QTL mapping of fire blight resistance in apple. Mol. Breed. 2006, 17, 299–306. [CrossRef] 34. Khan, M.A.; Durel, C.-E.; Duffy, B.; Drouet, D.; Kellerhals, M.; Gessler, C.; Patocchi, A. Development of molecular markers linked to the “Fiesta” linkage group 7 major QTL for fire blight resistance and their application for marker-assisted selection. Genome 2007, 50, 568–577. [CrossRef] [PubMed] 35. Le Roux, P.M.F.; Khan, M.A.; Broggini, G.A.L.; Duffy, B.; Gessler, C.; Patocchi, A. Mapping of quantitative trait loci for fire blight resistance in the apple cultivars “Florina” and “Nova Easygro”. Genome 2010, 53, 710–722. [CrossRef] 36. Khan, M.A.; Zhao, Y.F.; Korban, S.S. Identification of genetic loci associated with fire blight resistance in Malus through combined use of QTL and association mapping. Physiol. Plant. 2013, 148, 344–353. [CrossRef] 37. Desnoues, E.; Norelli, J.L.; Aldwinckle, H.S.; Wisniewski, M.E.; Evans, K.M.; Malnoy, M.; Khan, A. Identification of Novel Strain-Specific and Environment-Dependent Minor QTLs Linked to Fire Blight Resistance in Apples. Plant Mol. Biol. Rep. 2018, 36, 247–256. [CrossRef] 38. Kostick, S.A.; Teh, S.L.; Norelli, J.L.; Vanderzande, S.; Peace, C.; Evans, K.M. Fire blight QTL analysis in a multi-family apple population identifies a reduced-susceptibility allele in “Honeycrisp”. Hortic. Res. 2021, 8, 28. [CrossRef] 39. Thapa, R.; Singh, J.; Gutierrez, B.; Arro, J.; Khan, A. Genome-wide association mapping identifies novel loci underlying fire blight resistance in apple. Plant Genome 2021, 14, e20087. [CrossRef] [PubMed] 40. Antón, M.J. Influencia del Origen y la Tecnología de Elaboración sobre el Perfil Aromático y Olfatométrico de la Sidra. Ph.D. Thesis, University of Oviedo, Oviedo, Spain, 2015. 41. González, B.; González, P.; Díaz, B.A. Análisis del sector de la sidra: Retos actuales y perspectivas de futuro. Estud. Agrosociales Y Pesq. 2002, 194, 169–188. 42. Dapena, E.; García, J.; Ma Dolores, B. La diversidad varietal del manzano en Asturias. Tecnol. Agroaliment. Boletín Inf. CIATA 1999, 1, 28–29. 43. Dapena, E.; Blázquez, M.D.; Fernández, M. Recursos fitogenéticos del Banco de Germoplasma de manzano del SERIDA. Tecnol. Agroaliment. Boletín Inf. SERIDA. 2006, 3, 34–39. 44. Mangas, J.J.; Rodríguez, R.; Suárez, B.; Picinelli, A.; Dapena, E. Study of the phenolic profile of cider apple cultivars at maturity by multivariate techniques. J. Agric. Food Chem. 1999, 47, 4046–4052. [CrossRef] [PubMed] 45. Miñarro, M.; Dapena, E. Resistance of Apple Cultivars to Dysaphis plantaginea (Hemiptera: Aphididae): Role of Tree Phenology in Infestation Avoidance. Environ. Entomol. 2007, 36, 1206–1211. [CrossRef] [PubMed] 46. Miñarro, M.; Dapena, E. Response of local and scab-resistant cultivars to woolly apple aphid. Acta Hortic. 2009, 814, 825–830. [CrossRef] Plants 2023, 12, 4068 12 of 13 47. Arias, P.; Díaz, D.; Junco, S.; Dapena, E.; Gutiérrez, M.D.; Gomis, D.B. Characterisation of Asturian cider apples on the basis of their aromatic profile by high-speed gas chromatography and solid-phase microextraction. Food Chem. 2010, 121, 1312–1318. [CrossRef] 48. Pello-Palma, J.; Mangas-Alonso, J.J.; de la Fuente, E.D.; González-Álvarez, J.; Díez, J.; Álvarez, M.D.G.; Abrodo, P.A. Characteriza- tion of volatile compounds in new cider apple genotypes using multivariate analysis. Food Anal. Methods 2016, 9, 3492–3500. [CrossRef] 49. Delgado, A.; Quinet, M.; Dapena, E. Analysis of the variability of floral and pollen traits in apple cultivars—Selecting suitable pollen donors for cider apple orchards. Agronomy 2021, 11, 1717. [CrossRef] 50. Delgado, A.; Dapena, E.; Fernandez, E.; Luedeling, E. Climatic requirements during dormancy in apple trees from northwestern Spain—Global warming may threaten the cultivation of high-chill cultivars. Eur. J. Agron. 2021, 130, 126374. [CrossRef] 51. Dapena, E.; Blázquez, M.D. Conservación, evaluación, selección y mejora de los recursos fitogenéticos del Banco de Germoplasma de Manzano del SERIDA. Frutic. Prof. Espec. Manzano II 2002, 128, 65–72. 52. Dapena, E.; Blázquez, M.D. Improvement of the resistance to scab, rosy apple aphid and fire blight in a breeding programme of cider apple cultivars. Acta Hortic. 2004, 663, 725–728. [CrossRef] 53. Dapena, E.; Blázquez, M.D. Descripción de las Variedades de Manzana de la D.O.P. Sidra de Asturias; SERIDA, Ed.; SERIDA: Asturias, Spain, 2009; p. 71. 54. Consejo Regulador de la Denominación de Origen Protegida Sidra de Asturias Variedades de manzana de la DOP “Sidra de Asturias”. Available online: https://sidradeasturias.es/en/varieties/ (accessed on 28 March 2023). 55. MAPA Fuego Bacteriano (Erwinia amylovora). Ministerio de Agricultura, Pesca y Alimentación. Available online: https://www. mapa.gob.es/es/agricultura/temas/sanidad-vegetal/organismos-nocivos/fuego-bacteriano/ (accessed on 28 March 2023). 56. Llamero García, N. Caracterización Molecular de Variedades del Banco de Germoplasma de Manzano del SERIDA. Master’s Thesis, Universidad de Oviedo, Oviedo, Spain, 2014. 57. Donovan, A. Screening for fire blight resistance in apple (Malus pumila) using excised leaf assays from in vitro and in vivo grown material. Ann. Appl. Biol. 1991, 119, 59–68. [CrossRef] 58. Donovan, A.M.; Morgan, R.; Valobra-Piagnani, C.; Ridout, M.S.; James, D.J.; Garrett, C.M.E. Assessment of somaclonal variation in apple. I. Resistance to the fire blight pathogen, Erwinia amylovora. J. Hortic. Sci. 1994, 69, 105–113. [CrossRef] 59. Le Lezec, M.; Paulin, J.P. Shoot susceptibility to fireblight of some apple cultivars. Acta Hortic. 1984, 151, 277–282. [CrossRef] 60. Consejo Regulador de la Denominación de Origen Protegida Sidra de Asturias Memoria de Actividades 2022. Sidra de Asturias Denominación de Origen Protegida. Available online: https://sidradeasturias.es/memoria-anual/ (accessed on 10 August 2023). 61. Norelli, J.L. Virulence of Erwinia amylovora Strains to Malus sp. Novole Plants Grown in Vitro and in the Greenhouse. Phytopathology 1988, 78, 1292. [CrossRef] 62. Cabrefiga, J.; Montesinos, E. Analysis of Aggressiveness of Erwinia amylovora Using Disease-Dose and Time Relationships. Phytopathology 2005, 95, 1430–1437. [CrossRef] 63. Peil, A.; Flachowsky, H.; Hanke, M.V.; Richter, K.; Rode, J. Inoculation of Malus × robusta 5 progeny with a strain breaking resistance to fire blight reveals a minor QTL on LG5. Acta Hortic. 2011, 896, 357–362. [CrossRef] 64. Vogt, I.; Wöhner, T.; Richter, K.; Flachowsky, H.; Sundin, G.W.; Wensing, A.; Savory, E.A.; Geider, K.; Day, B.; Hanke, M.-V.; et al. Gene-for-gene relationship in the host-pathogen system Malus × robusta 5- Erwinia amylovora. New Phytol. 2013, 197, 1262–1275. [CrossRef] 65. Wöhner, T.W.; Flachowsky, H.; Richter, K.; Garcia-Libreros, T.; Trognitz, F.; Hanke, M.V.; Peil, A. QTL mapping of fire blight resistance in Malus × robusta 5 after inoculation with different strains of Erwinia amylovora. Mol. Breed. 2014, 34, 217–230. [CrossRef] 66. Dapena, E.; Ma Dolores, B.; Mangas, J.J.; González, A.J.; Laurens, F.; Gómez-Bernardo, E.M.; María Pilar, C.; Marcos, M.F.; Prado, N. Obtención de Variedades de Manzana de Sidra de Calidad, Elevada Resistencia y Regularidad Productiva. Aplicación de Nuevas Técnicas de Análisis de Metabolitos y de Selección Asistida con Marcadores; Memoria de Actividades de I+D+i del SERIDA 2016; SERIDA: Asturias, Spain, 2017; pp. 161–162. 67. Emeriewen, O.F.; Flachowsky, H.; Peil, A. Characterization of genomic DNA sequence of the candidate gene for FB_Mfu10 associated with fire blight resistance in Malus species. BMC Res. Notes 2021, 14, 291. [CrossRef] [PubMed] 68. Emeriewen, O.F.; Richter, K.; Piazza, S.; Micheletti, D.; Broggini, G.A.L.; Berner, T.; Keilwagen, J.; Hanke, M.-V.; Malnoy, M.; Peil, A. Towards map-based cloning of FB_Mfu10: Identification of a receptor-like kinase candidate gene underlying the Malus fusca fire blight resistance locus on linkage group 10. Mol. Breed. 2018, 38, 106. [CrossRef] [PubMed] 69. Jung, S.; Lee, T.; Cheng, C.-H.; Buble, K.; Zheng, P.; Yu, J.; Humann, J.; Ficklin, S.P.; Gasic, K.; Scott, K.; et al. 15 years of GDR: New data and functionality in the Genome Database for Rosaceae. Nucleic Acids Res. 2019, 47, D1137–D1145. [CrossRef] [PubMed] 70. Kruijt, M.; DE Kock, M.J.D.; de Wit, P.J.G.M. Receptor-like proteins involved in plant disease resistance. Mol. Plant Pathol. 2005, 6, 85–97. [CrossRef] [PubMed] 71. Sarowar, S.; Zhao, Y.; Soria-Guerra, R.E.; Ali, S.; Zheng, D.; Wang, D.; Korban, S.S. Expression profiles of differentially regulated genes during the early stages of apple flower infection with Erwinia amylovora. J. Exp. Bot. 2011, 62, 4851–4861. [CrossRef] 72. Kamber, T.; Buchmann, J.P.; Pothier, J.F.; Smits, T.H.M.; Wicker, T.; Duffy, B. Fire blight disease reactome: RNA-seq transcriptional profile of apple host plant defense responses to Erwinia amylovora pathogen infection. Sci. Rep. 2016, 6, 21600. [CrossRef] Plants 2023, 12, 4068 13 of 13 73. Tenhaken, R.; Doerks, T.; Bork, P. DCD—A novel plant specific domain in proteins involved in development and programmed cell death. BMC Bioinform. 2005, 6, 169. [CrossRef] 74. Yu, J.; Pressoir, G.; Briggs, W.H.; Vroh Bi, I.; Yamasaki, M.; Doebley, J.F.; McMullen, M.D.; Gaut, B.S.; Nielsen, D.M.; Holland, J.B.; et al. A unified mixed-model method for association mapping that accounts for multiple levels of relatedness. Nat. Genet. 2006, 38, 203–208. [CrossRef] 75. Segura, V.; Vilhjálmsson, B.J.; Platt, A.; Korte, A.; Seren, Ü.; Long, Q.; Nordborg, M. An efficient multi-locus mixed-model approach for genome-wide association studies in structured populations. Nat. Genet. 2012, 44, 825–830. [CrossRef] 76. Wang, S.-B.; Feng, J.-Y.; Ren, W.-L.; Huang, B.; Zhou, L.; Wen, Y.-J.; Zhang, J.; Dunwell, J.M.; Xu, S.; Zhang, Y.-M. Improving power and accuracy of genome-wide association studies via a multi-locus mixed linear model methodology. Sci. Rep. 2016, 6, 19444. [CrossRef] [PubMed] 77. Gao, X.; Starmer, J.; Martin, E.R. A multiple testing correction method for genetic association studies using correlated single nucleotide polymorphisms. Genet. Epidemiol. 2008, 32, 361–369. [CrossRef] 78. Gao, X.; Becker, L.C.; Becker, D.M.; Starmer, J.D.; Province, M.A. Avoiding the high Bonferroni penalty in genome-wide association studies. Genet. Epidemiol. 2010, 34, 100–105. [CrossRef] [PubMed] 79. Alonso-Blanco, C.; Méndez-Vigo, B. Genetic architecture of naturally occurring quantitative traits in plants: An updated synthesis. Curr. Opin. Plant Biol. 2014, 18, 37–43. [CrossRef] [PubMed] 80. Larsen, B.; Migicovsky, Z.; Jeppesen, A.A.; Gardner, K.M.; Toldam-Andersen, T.B.; Myles, S.; Ørgaard, M.; Petersen, M.A.; Pedersen, C. Genome-Wide Association Studies in Apple Reveal Loci for Aroma Volatiles, Sugar Composition, and Harvest Date. Plant Genome 2019, 12, 180104. [CrossRef] [PubMed] 81. Pascual, L.; Albert, E.; Sauvage, C.; Duangjit, J.; Bouchet, J.-P.; Bitton, F.; Desplat, N.; Brunel, D.; Le Paslier, M.-C.; Ranc, N.; et al. Dissecting quantitative trait variation in the resequencing era: Complementarity of bi-parental, multi-parental and association panels. Plant Sci. 2016, 242, 120–130. [CrossRef] [PubMed] 82. Nishitani, C.; Hirai, N.; Komori, S.; Wada, M.; Okada, K.; Osakabe, K.; Yamamoto, T.; Osakabe, Y. Efficient Genome Editing in Apple Using a CRISPR/Cas9 system. Sci. Rep. 2016, 6, 31481. [CrossRef] [PubMed] 83. Urrestarazu, J.; Muranty, H.; Denancé, C.; Leforestier, D.; Ravon, E.; Guyader, A.; Guisnel, R.; Feugey, L.; Aubourg, S.; Celton, J.-M.; et al. Genome-Wide Association Mapping of Flowering and Ripening Periods in Apple. Front. Plant Sci. 2017, 8, 1923. [CrossRef] 84. Mora, I.; Cabrefiga, J.; Montesinos, E. Cyclic Lipopeptide Biosynthetic Genes and Products, and Inhibitory Activity of Plant- Associated Bacillus against Phytopathogenic Bacteria. PLoS ONE 2015, 10, e0127738. [CrossRef] 85. Bianco, L.; Cestaro, A.; Linsmith, G.; Muranty, H.; Denancé, C.; Théron, A.; Poncet, C.; Micheletti, D.; Kerschbamer, E.; Di Pierro, E.A.; et al. Development and validation of the Axiom(® ) Apple480K SNP genotyping array. Plant J. 2016, 86, 62–74. [CrossRef] 86. Purcell, S.; Neale, B.; Todd-Brown, K.; Thomas, L.; Ferreira, M.A.R.; Bender, D.; Maller, J.; Sklar, P.; de Bakker, P.I.W.; Daly, M.J.; et al. PLINK: A tool set for whole-genome association and population-based linkage analyses. Am. J. Hum. Genet. 2007, 81, 559–575. [CrossRef] [PubMed] 87. Daccord, N.; Celton, J.-M.; Linsmith, G.; Becker, C.; Choisne, N.; Schijlen, E.; van de Geest, H.; Bianco, L.; Micheletti, D.; Velasco, R.; et al. High-quality de novo assembly of the apple genome and methylome dynamics of early fruit development. Nat. Genet. 2017, 49, 1099–1106. [CrossRef] [PubMed] 88. Pritchard, J.K.; Stephens, M.; Donnelly, P. Inference of population structure using multilocus genotype data. Genetics 2000, 155, 945–959. [CrossRef] [PubMed] 89. Earl, D.A.; vonHoldt, B.M. STRUCTURE HARVESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 2012, 4, 359–361. [CrossRef] 90. de Mendiburu, F. Agricolae: Statistical Procedures for Agricultural Research. R package Version 1.3-6. 2023. Available online: https://CRAN.R-project.org/package=agricolae (accessed on 4 September 2023). 91. Bates, D.; Mächler, M.; Bolker, B.; Walker, S. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 2015, 67, 1–48. [CrossRef] 92. Wang, J.; Zhang, Z. GAPIT version 3: Boosting power and accuracy for genomic association and prediction. Genom. Proteom. Bioinform. 2021, 19, 629–640. [CrossRef] 93. Bradbury, P.J.; Zhang, Z.; Kroon, D.E.; Casstevens, T.M.; Ramdoss, Y.; Buckler, E.S. TASSEL: Software for association mapping of complex traits in diverse samples. Bioinformatics 2007, 23, 2633–2635. [CrossRef] 94. Endelman, J.B.; Jannink, J.-L. Shrinkage estimation of the realized relationship matrix. G3 2012, 2, 1405–1413. [CrossRef] 95. R Core Team R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2023. 96. Wickham, H. ggplot2: Elegant Graphics for Data Analysis; Springer: New York, NY, USA, 2016; ISBN 978-3-319-24277-4. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.