Marimon et al. Horticulture Research (2020)7:175 https://doi.org/10.1038/s41438-020-00396-9 Horticulture Research www.nature.com/hortres ARTICLE Open Access Fine mapping and identification of candidate genes for the peach powdery mildew resistance gene Vr3 Neus Marimon1,2,3, Jordi Luque3, Pere Arús 1,2 and Iban Eduardo 1,2 Abstract Powdery mildew is one of the major diseases of peach (Prunus persica), caused by the ascomycete Podosphaera pannosa. Currently, it is controlled through calendar-based fungicide treatments starting at petal fall, but an alternative is to develop resistant peach varieties. Previous studies mapped a resistance gene (Vr3) in interspecific populations between almond (‘Texas’) and peach (‘Earlygold’). To obtain molecular markers highly linked to Vr3 and to reduce the number of candidate genes, we fine-mapped Vr3 to a genomic region of 270 kb with 27 annotated genes. To find evidence supporting one of these positional candidate genes as being responsible of Vr3, we analyzed the polymorphisms of the resequences of both parents and used near-isogenic lines (NILs) for expression analysis of the positional candidate genes in symptomatic or asymptomatic leaves. Genes differentially expressed between resistant and susceptible individuals were annotated as a Disease Resistance Protein RGA2 (Prupe2G111700) or an Eceriferum 1 protein involved in epicuticular wax biosynthesis (Prupe2G112800). Only Prupe2G111700 contained a variant predicted to have a disruptive effect on the encoded protein, and was overexpressed in both heterozygous and homozygous individuals containing the Vr3 almond 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; allele, compared with susceptible individuals. This information was also useful to identify and validate molecular markers tightly linked and flanking Vr3. In addition, the NILs used in this work will facilitate the introgression of this gene into peach elite materials, alone or pyramided with other known resistance genes such as peach powdery mildew resistance gene Vr2. Introduction infects the fruits, leaves, buds, and shoots, where myce- Peach [Prunus persica (L.) Batsch] is one of the best lium develops as white-grayish spots on the surface, and characterized species among Rosaceae1–3 and an impor- heavy infections on fruit and leaves may induce their tant stone fruit crop in temperate regions: more than 24 premature fall7,8. PPM can be controlled effectively million tons of peaches, nectarines and flat fruits pro- through foliar fungicide applications, applied regularly duced worldwide in 20184. Most commercial peach cul- every 7–14 days during the year9 from prebloom to the tivars are susceptible to different pests and diseases. One end of harvest5. Recently, a predictive model for disease of the most important being peach powdery mildew progress has been described10, which included a threshold (PPM)5,6, caused by the ascomycete Podosphaera pan- to initiate fungicide programs at early infection set. nosa7. To our knowledge, all peach commercial cultivars An environmentally safe alternative to fungicide appli- are susceptible PPM to a variable degree. The pathogen cations is the development of resistant varieties through plant breeding6. Little information is currently available on breeding for resistance to pests and pathogens in stone Correspondence: Iban Eduardo (iban.eduardo@irta.cat) 1 fruit crops11, probably due to the length of time required Centre for Research in Agricultural Genomics (CRAG) CSIC-IRTA-UAB-UB, to introduce genes from exotic sources in perennial Campus UAB, 08193 Bellaterra, Barcelona, Spain 2 IRTA (Institut de Recerca i Tecnologia Agroalimentàries), Barcelona, Spain plants. Two descriptions of PPM major resistance genes Full list of author information is available at the end of the article © The Author(s) 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Marimon et al. Horticulture Research (2020)7:175 Page 2 of 9 have been published. Pascal et al.5,6 described a mono- Breeding peach cultivars resistant to PPM will be more genic dominant locus in linkage group 8 (G8), named Vr2, efficient with the availability of molecular markers tightly from the peach rootstock cultivar ‘Pamirskij 5’. In the linked to genes of resistance or based on the poly- peach cross-compatible Prunus species almond (P. dulcis), morphism responsible for the resistance. Therefore, our Donoso et al.12 mapped a monogenic powdery mildew objectives were to fine map the Vr3 gene responsible for resistance gene in G2 in two interspecific populations PPM resistance to obtain a reduced number of candidate between almond ‘Texas’ and peach ‘Earlygold’, with the genes and to characterize them by analyzing the poly- dominant resistance allele from ‘Texas’. The gene, named morphisms of parent resequences and performing an Vr3, was located in a genomic region of 2.7 cM and expression analysis. The outcome of this study would 1.8 Mb, where 187 genes were annotated in the peach provide valuable information on the Vr3 candidate genes reference genome. Other resistance sources are quantita- and better markers for marker assisted selection in peach tive trait loci (QTLs) controlling PPM tolerance. Pacheco- breeding programs. Cruz et al.13 described a source of tolerance from peach ‘OroA’ in G7 that could explain up to 8% of the phenotypic Results variation, and several QTLs have been identified in Prunus Fine mapping and identification of Vr3 candidate genes davidiana7,8. Furthermore, Dabov14 found that in Prunus A total of 729 descendants derived from individuals ferganensis, Vr and Sr alleles conferred high and low carrying Vr3 in heterozygosis were genotyped using two resistance, respectively. Moreover, a lower susceptibility to SSR markers (CPDCT044 and BPPCT004) known to PPM in peach has also been found to be associated with include Vr312. These were from nine populations shown the presence of leaf glands, this character having been in Table 1 (TxE, T1E, E2T-031, 11P15, 15P15, 19P15, mapped at the E/e locus on chromosome 715. 25P15, and T1BT). Table 1 Individuals used in fine mapping of the Vr3 PPM resistance gene Year Population type Family code Female parent Male parent Individuals Recombinant individuals Location – F1 MB1.37 ‘Texas’ ‘Earlygold’ 1 - Caldes de Montbui – F2 TxE ‘MB1.37’ ‘MB1.37’ 111 3 Cabrils/Gimenells – BC1 T1E ‘MB1.37’ ‘Earlygold’ 189 3 Cabrils/Gimenells 2014 BC2 E2T-031 ‘Earlygold’ T1E-031 26 2 Caldes de Montbui/Mollerussa 2015 F2 TxE ‘MB1.37’ ‘MB1.37’ 150 5 Caldes de Montbui 2015 BC2 11P15 ‘Earlygold’ T1E-031 14 1 Caldes de Montbui 2015 BC3 15P15 E2T-031-005 OP 26 11 Caldes de Montbui 2015 BC3 19P15 E2T-092-002 OP 127 3 Caldes de Montbui 2015 BC3 25P15 E2T-092-021 OP 64 1 Caldes de Montbui 2015 BC2 – T1E-042 ‘Nectatop’ 4 1 Caldes de Montbui 2015 BC2 – ‘Nectatop’ T1E-03 1 1 Caldes de Montbui 2015 BC2 – ‘Sweetlove’ T1E-03 2 1 Caldes de Montbui 2016 BC2 E2T-092 ‘Earlygold’ T1E-092 11 1 Caldes de Montbui/Mollerussa 2016 BC3 14P16 ‘Nectatop’ E2T-092-025 22 6 Mollerussa 2016 BC3 1114 P01F002A054 E2T-092-025 28 11 Mollerussa 2017 BC3 44P17 ‘Nectatop’ E2T-092-025 54 2 Gimenells 2017 BC2 51P17 ‘MB1.37’ OP 218 6 Caldes de Montbui 2018 BC2 72P18 T1E-021 OP 33 2 Caldes de Montbui 2018 BC2 74P18 T1E-024 OP 12 1 Caldes de Montbui 2018 BC2 84P18 T1E-040 OP 25 3 Caldes de Montbui 2018 BC2 93P18 T1E-064 OP 259 2 Caldes de Montbui – BC1 T1BT ‘MB1.37’ ‘Big Top’ 21 1 Mollerussa OP open pollination Marimon et al. Horticulture Research (2020)7:175 Page 3 of 9 Table 2 Phenotypes and genotypes of individuals with a recombinant breakpoint (dashed lines) near Vr3 Marker Position Genotype CPDCT044 16,847,924 b b b b b b b h h h h Indel16883 16,883,671 b b b b b b b h h h h Indel16912 16,912,811 b b b b b b b h h h h SNP_16932290 16,932,290 b b b b b b h h h h b SNP_16940264 16,940,264 b b b b b b h h h h b Indel16949 16,948,818 b b b b b b h h h h b Indel17019 17,019,668 b b b b b b h h h h b Indel17048 17,048,260 b b b b b b h h h h b Indel17050 17,050,734 b b b b b b h h h h b Indel17061 17,061,201 b b b b b b h h h h b SNP_17180556 17,180,556 b b b b b b h h h h b SSR_17181256 17,181,256 b b b b b b h h h h b SSR_17182435 17,182,435 b b b b b b h h h h b SNP_17184692 17,184,692 b b b b b b h h h b b SSR_17184920 17,184,920 b b b b b b h h h b b 6620 17,166,620 b b b b b h h h h b b Indel17186 17,186,620 b b b b b h h h h b b indel17229 17,229,285 b b b b h h h h h b b Indel17242 17,242,814 b b b b h h h h h b b 2031 17,262,031 b b b b h h h h h b b Indel17272 17,272,322 b b b b h h h h h b b Indel17479 17,479,459 b b b h h h h h h b b Indel17909 17,909,204 b b h h h h h h b b b Indel18610 18,610,981 b h h h h h h b b b b BPPCT004 18,641,408 h h h h h h h b b b b Phenotype S S S S S S R R R R S N 3 4 13 2 3 1 2 3 1 2 1 b allele from the susceptible parent ‘Earlygold’, h allele from heterozygote individuals, R resistant. S susceptible, N number of recombinant individuals The recombination between the two markers in 30 of between Indel16912 and SNP_16932290, defining the these 729 individuals was observed, with 16% phenotyped lower limit of the Vr3 region. Two resistant individuals as resistant and 84% as susceptible. Based on the rese- with a recombination between SSR_17182435 and quences of the parental lines, we designed new molecular SNP_17184692, corresponding to families E2T-031-06 and markers, including four SSRs, 14 Indels and four SNPs (SM 51P17, defined the upper limit. In this region, spanning Tables 1 and 2). Recombinant individuals (Table 1) were ~270 kb, 27 annotated genes (Table 3) were found in the P. genotyped using these markers to narrow down the geno- persica Genome Annotation v2.116 retrieved from the mic region where Vr3 was located. After phenotyping the Genome Database for Rosaceae17. recombinant individuals, and using the new genotyping Among the 27 candidate genes, five were annotated as information, we located Vr3 in the region between markers involved in plant defense, an additional five encoding for Indel16912 and SNP_17184692 (Table 2), corresponding to structural function, ten genes were predicted to be involved physical positions 16,912,811 and 17,184,692, respectively. in plant metabolism, and seven were annotated as Five individuals had the nearest recombination to the unknown (Table 3). Among the five candidate genes Vr3 gene, so determining the Vr3 region. Two resistant described as involved in plant defense, Prupe.2G110900 was individuals from TxE and 44P17 families and one suscep- predicted to function as a germin-like protein. The other tible individual from 14P16 family showed a recombination four (Prupe.2G111700, Prupe.2G111800, Prupe.2G112700, Marimon et al. Horticulture Research (2020)7:175 Page 4 of 9 Table 3 Vr3 resistance candidate genes to peach powdery mildew Gene Position Predicted function Function classification Prupe.2G110900 Pp02:16913576..16914676 Agamous-like MADS-box protein (Arabidopsis thaliana) Metabolism Prupe.2G111000 Pp02:16920605..16921195 Germin-like protein (Oryza sativa subsp. Japonica) Plant defense Prupe.2G111100 Pp02:16922483..16922915 n/a Unknown Prupe.2G111200 Pp02:16923107..16925825 n/a Unknown Prupe.2G111300 Pp02:16926230..16929679 26S protease (Arabidopsis thaliana) Metabolism Prupe.2G111400 Pp02:16930138..16934333 ABC transporter (Arabidopsis thaliana) Metabolism Prupe.2G111500 Pp02:16936869..16945304 ABC transporter (Arabidopsis thaliana) Metabolism Prupe.2G111600 Pp02:16993968..16994329 n/a Unknown Prupe.2G111700 Pp02:16996435..17001837 Disease resistance protein RGA2 (Solanum bulbocastanum) Plant defense Prupe.2G111800 Pp02:17003896..17010678 Putative disease resistance protein RGA3 (Solanum bulbocastanum) Plant defense Prupe.2G111900 Pp02:17011424..17014545 Hydrolase domain-containing protein Sgpp (Arabidopsis thaliana) Metabolism Prupe.2G112000 Pp02:17015308..17018250 Endoglucanase 12 (Arabidopsis thaliana) Metabolism Prupe.2G112100 Pp02:17020262..17023645 Riboflavin biosynthesis protein PYRD (Arabidopsis thaliana) Metabolism Prupe.2G112200 Pp02:17024102..17036171 DNA replication helicase (Arabidopsis thaliana) Metabolism Prupe.2G112300 Pp02:17039404..17042658 Zinc ion binding (Arabidopsis thaliana) Metabolism Prupe.2G112400 Pp02:17049419..17050149 n/a Unknown Prupe.2G112500 Pp02:17050202..17050814 n/a Unknown Prupe.2G112600 Pp02:17061213..17068962 Protein ECERIFERUM 1 (Arabidopsis thaliana) Structural Prupe.2G112700 Pp02:17073807..17075686 TMV resistance protein N (Nicotiana glutinosa) Plant defense Prupe.2G112800 Pp02:17099320..17103427 Protein ECERIFERUM 1 (Arabidopsis thaliana) Structural Prupe.2G112900 Pp02:17113525..17117895 Protein ECERIFERUM 1 (Arabidopsis thaliana) Structural Prupe.2G113000 Pp02:17138061..17139410 Protein ECERIFERUM 1 (Arabidopsis thaliana) Structural Prupe.2G113100 Pp02:17141136..17142354 n/a Unknown Prupe.2G113200 Pp02:17142564..17145425 TMV resistance protein N (Nicotiana glutinosa) Plant defense Prupe.2G113300 Pp02:17151739..17152569 n/a Unknown Prupe.2G113400 Pp02:17166049..17166711 RING-H2 finger protein ATL3 (Arabidopsis thaliana) Metabolism Prupe.2G113500 Pp02:17168568..17172597 Protein ECERIFERUM 1 (Arabidopsis thaliana) Structural and Prupe.2G113200) were predicted to be plant resistance Variant calling and effect prediction of polymorphisms genes (R genes). Moreover, Prupe.2G112700 and Pru- A total of 3510 variants including 3073 SNPs, 222 pe.2G113200 were specifically included in the TIR-NBS- insertions, and 215 deletions were identified (details shown LRR class of plant R genes. Five genes (Prupe.2G112600, in SM Table 3). These variants were predicted to cause Prupe.2G112800, Prupe.2G112900, Prupe.2G113000, and 11,958 effects on the sequences (SM Table 3). Most of Prupe.2G113500) were predicted to encode protein Ecer- them (93.7%) were considered noncoding variants or var- iferum 1, involved in epicuticular wax biosynthesis. Finally, iants affecting noncoding genes, while 13 (0.11%) were of the ten genes predicted to be involved in plant meta- predicted as high-impact variants, 350 (2.9%) as moderate, bolism, three were annotated with hydrolase function and 392 (3.28%) as low-impact variants. The 13 high- (Prupe.2G111900, Prupe.2G112000, and Prupe.2G112100), impact variants producing a disruptive effect on the coded three related to DNA binding (Prupe.2G110900, Pru- protein were detected in six candidate genes (SM Table 4). pe.2G112300, and Prupe.2G113400), three ATP-related Genes annotated as RGA2 resistance protein (Pru- genes (Prupe.2G111300, Prupe.2G111400, and Pru- pe.2G111700), RGA3 resistance protein (Prupe.2G111800), pe.2G111500) and one predicted as a multifunctional DNA replication helicase (Prupe.2G112200), and three enzyme (Prupe.2G112200). genes with no available annotation (Prupe.2G111100, Marimon et al. Horticulture Research (2020)7:175 Page 5 of 9 7 Prupe.2G111700 5 Prupe.2G112800 (RGA2) (ECERIFERUM 1) 6 Normalized expression Normalized expression 4 5 4 3 3 2 2 1 1 0 0 Asymptomatic Symptomatic Asymptomatic Symptomatic leaves leaves leaves leaves Fig. 1 Relative normalized expression of candidate genes with significant differences in symptomatic and asymptomatic leaves (p < 0.01). Solid and dashed lines correspond to homozygous (Vr3Vr3) and heterozygous (Vr3vr3) individuals for the Vr3 allele from ‘Texas’, respectively. Dotted lines correspond to individuals with Vr3 peach alleles. Bars indicate standard error of the mean Prupe.2G112400 and Prupe.2G112500) presented one high- Vr3vr3 individuals: the normalized expression was 4.91 ± impact variant, whereas genes annotated as Eceriferum 1 0.84 (mean ± SE) and 1.85 ± 0.36 respectively in asymp- (Prupe.2G113500) and another with no available annota- tomatic leaves, and 5.68 ± 0.50 and 4.51 ± 0.16, in symp- tion presented three and four high-impact variants, tomatic leaves. In addition, the normalized expression for respectively. susceptible individuals with no Vr3 almond introgression (1.15 ± 0.11) was significantly lower compared with Expression analysis of candidate genes resistant individuals (p < 0.01). Eceriferum 1 (Pru- Relative normalized expression profiles of the 27 can- pe.2G112800) was upregulated for individuals containing didate genes annotated in the Vr3 region in this study the Vr3 allele, and again gene expression in Vr3Vr3 dif- were analyzed to describe the effect of the infection status fered significantly from heterozygous individuals. Never- and the presence of Vr3 introgression. One gene, Pru- theless, no significant differences were detected among pe.2G111600, was excluded from the analysis because we Vr3vr3 individuals and susceptible individuals not carry- could not obtain a regular amplification signal and it was ing the Vr3 allele. Finally, genes encoding for Agamous- considered inappropriate for qPCR expression analysis. like MADS-box (Prupe.2G110900) and Germin-like pro- No variants with high or medium impact were detected in tein (Prupe.2G111000) were significantly underexpressed this gene. From the 26 candidate genes that could be in individuals containing the Vr3 allele (p < 0.05), and no successively analyzed, Eceriferum 1 (Prupe.2G112600) interaction between factors was detected. was the only gene with significant interaction (p < 0.01) between infection status and the Vr3 almond allele. Seven Discussion candidate genes were found to be significantly differen- The PPM resistance gene Vr3 was located in a 1.8 Mb tially expressed for one or both factors (p < 0.01). For the genomic region of chromosome 2 where 187 genes were infection status factor, three differentially expressed genes annotated in the peach reference genome12. In our study, were identified, namely Eceriferum 1 (Prupe.2G113000), through a fine mapping approach, we narrowed the region RING-H2 finger protein (Prupe.2G113400) and an down to 270-kb (between Pp02:16,912,811 and unknown gene (Prupe.2G113100). Their expression Pp02:17,184,692), with 27 genes annotated that were increased in all three cases when infection occurred considered as a first set of Vr3 positional candidate genes. regardless of the Vr3 allele presence (data not shown). Additional evidence in support of some of these genes Regarding the allelic status, the genes RGA2 (Pru- being responsible for PPM resistance was gathered pe.2G111700) and Eceriferum 1 (Prupe.2G112800) were through expression analysis and prediction of the effect of overexpressed in individuals homozygous (Vr3Vr3) and variants in the coding sequences of the candidate genes. heterozygous (Vr3vr3) for Vr3 (Fig. 1). Among the variants detected in the region, only those In both symptomatic and asymptomatic leaves, the predicted to have a high or moderate impact on the RGA2 annotated gene (Prupe.2G111700) had higher protein encoded were considered candidates for the Vr3 relative expression for Vr3Vr3 individuals compared with resistance gene. As defined by SNPEff software, high- Marimon et al. Horticulture Research (2020)7:175 Page 6 of 9 impact predicted variants were assumed to have dis- process through genetic transformation is required, but ruptive impact in the protein, causing protein truncation this is currently difficult due to the recalcitrant character or loss of function; and moderate predicted variants might of peach24. Another possibility could be the use of a alter protein effectiveness18. In the current study, 23 heterologous system such as plum for which an efficient candidate genes were predicted to have a moderate effect transformation approach has been described25, although a on the protein, and six variants also had a high-impact limitation of this approach is that species causing powdery effect. mildew in peach differ from that in plum (P. tridactyla). From the two candidate genes that were differentially This would only be successful if our RGA2 candidate gene overexpressed in resistant individuals, encoding for RGA2 conferred broad-spectrum resistance to powdery mildew, (Prupe.2G111700) and Eceriferum 1 (Prupe.2G112800), as it has been described for the Pm21 RGA gene in only RGA2 (Prupe.2G111700) had a high-impact variant wheat26. that is producing a stop codon. This gene also presented Functional validation of these genes is the main bot- 35 moderate variants. Regarding gene expression, RGA2 tleneck in Prunus due to its recalcitrant regeneration (Prupe.2G111700) was the only gene significantly over- behavior ‘in vitro’24. Until efficient peach transformation expressed in resistant Vr3Vr3 and Vr3vr3 individuals strategies are available, a feasible alternative to integrate compared to susceptible individuals (vr3vr3) indepen- the Vr3 gene in peach breeding programs could be Mar- dently of the infection status. Therefore, RGA2 (Prupe2. ker Assisted Introgression (MAI)27. For that, a near- G111700) was considered our strongest candidate gene isogenic line carrying a unique introgression from almond for Vr3. Moreover, as the expression of RGA2 (Pru- containing the Vr3 gene needs to be developed to cross pe.2G111700) did not differ significantly with respect to with the parents from a specific breeding program and the infection status, it is assumed to be constitutively then resistant individuals can be selected using the mole- expressed, as previously reported for RGA genes involved cular markers described in this work. This strategy is in fungal resistance in Rosaceae, such as for crown rot in currently in progress in our laboratory to introgress Vr3 octoploid strawberry19 and powdery mildew in apple20. resistant alleles from ‘Texas’ almond into high quality RGA genes are involved in the recognition and prevention peach commercial cultivars. Finally, we propose to pyramid of plant pathogens21, with highly conserved amino acid these lines with other PPM resistance genes such as Vr26 to domains that are already known in P. persica22. Vr3 has increase PPM resistance durability, and with other peach been described as a monogenic resistance gene12, and is biotic resistance genes to increase crop sustainability. thought to show completely dominant gene action, being the heterozygous and homozygous plants equally resis- Materials and methods tant. When comparing the allelic status on the expression Plant material of RGA2, homozygous individuals significantly over- From 2013 to 2018, several progenies of different gen- expressed Vr3 as compared to heterozygous individuals erations derived from ‘Texas’ and ‘Earlygold’ crosses were despite of the infection status. Conversely, when in het- screened for a PPM resistance fine mapping approach: F2 erozygosity, gene expression differed significantly between (named TxE, with 111 individuals), BC1 with ‘Earlygold’ asymptomatic and symptomatic individuals. This could as the recurrent parent (named T1E, with 189 individuals) have implications in the resistance mechanisms and and BC2 also with ‘Earlygold’ (with 51 screened indivi- should be borne in mind if this gene is used for breeding duals, from E2T-031, E2T-092 and 11P15 families). Other purposes. individuals used in the fine mapping approach were Another candidate gene differentially overexpressed in obtained from the open pollination of different individuals resistant individuals was Eceriferum 1 (Prupe.2G112800), and families: 218 individuals from ‘MB1.37’ (the ‘Texas’ × an ortholog of an Arabidopsis thaliana gene related to ‘Earlygold’ F1 individual used for the construction of the fungal recognition, based on cuticle wax components23. TxE population), 329 individuals from T1E progeny This gene had variants with moderate effect, and was families (named 72P18, 74P18, 84P18, and 93P18), and overexpressed only in homozygous individuals containing 217 individuals from BC2 progeny families (including the Vr3 almond allele as compared to heterozygous and 15P15, 19P15, and 25P15 families). In addition, some susceptible individuals. As no significant differences in recombinant individuals were obtained from crosses with expression were detected between the susceptible and the several peach commercial cultivars and individuals from Vr3vr3 individuals, phenotyped as resistant, Pru- different ‘Texas’ × ‘Earlygold’ generations. This included pe.2G112800 was not considered as a candidate gene for 81 individuals derived from ‘Nectatop’ crossed with dif- Vr3. ferent BC1 and BC2 individuals, All the trees described Results obtained in this study provide important infor- (Table 1) were planted at IRTA facilities located in Cabrils mation to identify a limited number of genes as Vr3 (41° 31’ 7’ N, 2° 22’ 34’ E), Caldes de Montbui (41° 36’ 47’ candidates, responsible for PPM resistance. A validation N, 2° 10’ 12’ E), Gimenells (41° 39’ 22’ N, 0° 23’ 26’ E), and Marimon et al. Horticulture Research (2020)7:175 Page 7 of 9 Mollerussa (41° 37’ 07’ N, 0° 51’ 60’ E). Orchards were not obtained by Serra29 using HiSeq2000 sequencer (Illumina treated with fungicides to allow natural pathogen Inc.). High-quality 220–480 bp size fragmented DNA was infections. ligated to Illumina paired-end adapters. Adapter removal Regarding the gene expression analysis, three groups of was done using AdapterRemoval v1.5.230. Only reads with four individuals from near-isogenic lines coming from a minimum size of 35 bp and a mean quality of 25 were open pollination of a BC2 individual were used. A first kept. High quality reads were mapped to the peach group contained only one introgression from ‘Texas’ reference genome using BWA v0.7.531 with default almond in homozygosis in the Vr3 genomic region, parameters. The SAM file was converted to BAM using another with one introgression in heterozygosis in the BAMTools v0.1.1932 and reads mapping to more than one same region, and the final one with no almond intro- position or reads from PCR duplication events were gression in the Vr3 genomic region, but including an excluded from the alignment. Raw Illumina data for almond introgression in G3. The four individuals of each ‘Texas’, ‘Earlygold’ and ‘MB1.37’ are available at the Eur- group were considered as independent biological repli- opean Nucleotide Archive under the accession numbers cates for each case, and two technical replicates of three ERS4540423, ERS3508161, and ERS4540424, respectively. young leaves measuring 3–4 cm were sampled from sight- Polymorphisms in these resequencing data were heighted and sun-exposed branches. The samples col- detected using Integrative Genomics Viewer software33. lected were symptomatic and visually asymptomatic SSRs and InDels (Table 2) were designed from the leaves. In addition, the presence of the pathogen in the flanking sequences of the polymorphisms using Primer 3 field was assessed through detection of airborne P. pan- (http://primer3.ut.ee), v4.1.034 with the default para- nosa propagules captured with a volumetric spore sam- meters. PCR reactions were in a final volume of 10 μL pler VPPS 2000 (Lanzoni, Bologna, Italy) and using a containing 200 ng of genomic DNA, 1 × NH4 reaction specific qPCR-based protocol developed in a previous buffer, 1.5 mM MgCl2, 0.2 mM dNTPs (10 mM), 0.2 μM study (manuscript in preparation). of each marker and 1 U of BIOTaq (Bioline, London, UK) and HPLC H2O to reach the final volume. PCRs were Phenotypic evaluation performed in a GeneAmp PCR System 9700 thermal All recombinant individuals used in this study were cycler (Applied Biosystems, CA, USA), with the following phenotyped for PPM susceptibility every year between conditions: initial denaturation at 94 °C for 1 min, 35 2016 and 2019. Each year, PPM was phenotyped twice, cycles of denaturation at 94 °C for 15 s, primer annealing first in May or June (corresponding with the developing at specific temperature for each primer for 15 s, extension stage of infection) and then in September (corresponding at 72 °C for 30 s, and a final extension at 72 °C for 5 min. with the end of infection but with symptoms still For InDels less than 40 bp and for SSRs, forward primers noticeable). Young leaves from a minimum of four dif- were designed with a generic fluorochrome sequence at ferently oriented branches were examined for PPM the 5′ ends (FAM, VIC, NED, or PET), named ‘tag pri- symptoms. A given individual was scored as resistant mers’35. PCR reaction conditions for these ‘tag primers’ when total absence of PPM symptoms on leaves was were the same as described above, with the following confirmed throughout the monitoring period. In contrast, modifications: 0.4 μM of each marker and 0.20 μM for trees showing PPM symptoms in at least 1 year were each ‘tag’ primer pair. PCR amplifications were with an considered susceptible. Trees for all the experimental initial denaturation at 94 °C for 1 min, followed by a total orchards evaluated for PPM resistance were not treated, of 60 cycles with the profile: 20 cycles for 15 s at 94 °C, to ensure infection and serve as positive controls. 15 s at 63 °C, and 30 s at 72 °C, followed by 40 cycles for 15 s at 94 °C, 15 s at 54 °C, and 30 s at 72 °C, followed by a Vr3 fine mapping final extension step of 5 min at 72 °C. PCR products were Genomic DNA from the individuals described in Table added to 12 μL of deionized formamide containing 1 was extracted from young leaves using a modification of 0.35 μL of GeneScan500 LIZ size standard (Applied Bio- the CTAB protocol28 omitting the final RNAse step. DNA systems, CA, USA). The mixture was heated at 94 °C for quality and concentration were checked and quantified 3 min and capillary electrophoresed using an ABI Prism using a DNA spectrophotometer (Nanodrop Technolo- 3130xl automated sequencer (Applied Biosystems, CA, gies, Wilmington, USA). USA). GeneMapper v5.0 software (Applied Biosystems) New markers (Table 2), including SSRs (Simple- was used for SSR allele sizing. For InDels larger than Sequence Repeat Markers), InDels (Introgression and 40 bp, standard primers were designed flanking the Deletion markers), and SNPs (Single Nucleotide Poly- polymorphism and results were observed in ethidium morphism) were designed using resequencing data of bromide-stained agarose gels (1.8%) under UV light. From ‘Texas’, ‘Earlygold’, and ‘MB1.37’. Library preparation and all the SSRs and Indels designed, only those showing clear 2 × 100 bp pair-end genome sequencing data were segregation among the parents were kept for the fine Marimon et al. Horticulture Research (2020)7:175 Page 8 of 9 mapping approach, avoiding those with preferential Time PCR (Fluidigm, CA, USA). Prior to the high- amplification for peach alleles or that did not amplify. throughput qPCR, a pre-amplification of the cDNA Otherwise, primers for SNPs detection (Table 3) were samples was performed. Diluted (1:3) pre-amplified designed using the Primer Picker Lite tool from KASPar cDNA samples were loaded according to Fluidigm’s SNP Genotyping System (KBiosciences, Herts, UK). SNP EvaGreen DNA-binding dye protocols. Negative controls genotyping was performed by qPCR through a Light- were used in the assay to detect possible DNA con- Cycler 480 device (Roche Diagnostics, Spain) using uni- tamination. Four reference genes used in previous versal KASPar MasterMix (LGC, Teddington, UK) expression analysis were evaluated: actin (Act), expansin following the supplier’s technical instructions. (Exp1)39, pre-mRNA splicing factor 7 (SLU7)40 and translation elongation factor 2 (TEF2)41. The stability of Prediction of variants effect of candidate genes sequences each reference gene was defined with the SATqPCR sta- Almond and peach resequences of candidate genes tistical analysis tool42 based on the geNorm method43, defining the Vr3 region, located between Pp02:16,912,811 considering the lowest gene variability. Considering the and Pp02:17,184,692 physical positions of the P. persica use of at least two reference genes, as described in MIQE v.2.0 reference genome16, were compared to predict the rules44, Act, Exp1, and SLU7 were finally chosen for variants in the region. Their effect on annotated genes of normalizing relative quantities for each candidate gene. the region was determined using SNPEffect software The effects of two factors in the relative expression of 4.3p18. Variant effect was defined by the impact on the candidate genes were considered: (i) disease status, i.e., protein in three categories: (i) high impact, by impairing symptomatic or asymptomatic leaves, and (ii) the pre- protein function, i.e., affecting splice-sites or start and sence of the Vr3 almond alleles, either in homozygosis or stop codons, (ii) moderate impact including nondisruptive heterozygosis. Considering normal distributions and variants, and (iii) low impact including synonymous independence of the observations, two-way analysis of variants. variance (ANOVA) was used to assess the independent effect of each of these two factors in normalized expres- Gene expression analysis sion of all candidate genes. ANOVA tests were performed RNA isolation and cDNA synthesis using the ‘RqPCRAnalysis’ R-package45 included in the The sampled leaves were immediately frozen in liquid SATqPCR statistical analysis tool43. Orthogonal contrasts nitrogen after collection and RNA was isolated with the were used to detect differences in different levels of the Spectrum Plant Total RNA kit (Sigma Aldrich, Munich, factor describing the disease status. The first contrast was Germany), according to the manufacturer’s instructions. among individuals with Vr3 introgression (including RNA concentration and purity were checked with a heterozygous and homozygous individuals) and indivi- Nanodrop ND-1000 spectrophotometer. Samples were duals without the introgression from ‘Texas’, and the only further processed for cDNA synthesis, if the 260/280 second among heterozygous and homozygous individuals. ratio was between 1.9 and 2.1, and the 260/230 ratio > 2.036. Statistical significance of these tests was set at α < 0.01. cDNA was synthetized from 1 µg of total RNA for each sample using the PrimeScript RT-PCR Kit (Takara, Otsu- Acknowledgements We acknowledge financial support from the Spanish Ministry of Economy, shi, Japan) according to the manufacturer’s instructions. Industry and Competitiveness (INIA grants RTA2015-00050-00-00 and RTA2013- 00004-C03-01), the Spanish Ministry of Economy and Competitiveness, Primer design of candidate genes through the “Severo Ochoa” Programme for Centres of Excellence in R&D 2016–2019 (SEV-2015-0533), grant AGL2015-68329-R, from the CERCA Primer pairs (SM Table 5) were designed for the 27 Programme (Generalitat de Catalunya) and from the project RIS3CAT (COTPA- candidate genes defined in the region of interest after Vr3 FRUIT3CAT) financed by the European Regional Development Fund through fine mapping. Coding DNA sequences of candidate genes the FEDER frame of Catalonia 2014–2020. Neus Marimon was supported by INIA with a predoctoral grant. We would like to thank Konstantinos Alexiou were obtained from the Genome Database for Rosaceae17. from IRTA-CRAG for the analysis of the polymorphisms in the Vr3 region. Sequences were analyzed by BLAST alignment for spe- cificity checking37, primer pairs suitability (GC content, Author details 1 self-complementarity and dimer formation) was checked Centre for Research in Agricultural Genomics (CRAG) CSIC-IRTA-UAB-UB, Campus UAB, 08193 Bellaterra, Barcelona, Spain. 2IRTA (Institut de Recerca i using Oligoanalyzer 3.1(Integrated DNA Technologies, Tecnologia Agroalimentàries), Barcelona, Spain. 3Plant Pathology, IRTA Cabrils, URL: https://eu.idtdna.com), and mfold38 (URL: http:// Carretera de Cabrils km 2, 08348 Cabrils, Spain unafold.rna.albany.edu) was used to predict secondary structure formation. Data availability The data that support the findings of this study are available from the corresponding author upon reasonable request. qPCR expression analysis qPCR assays were performed using a Fluidigm 48.48 Conflict of interest dynamic array chip on the BioMark HD System Real- The authors declare that they have no conflict of interest. Marimon et al. Horticulture Research (2020)7:175 Page 9 of 9 Supplementary Information accompanies this paper at (https://doi.org/ 21. Kim, J. et al. A genome-wide comparison of NB-LRR type of resistance gene 10.1038/s41438-020-00396-9). analogs (RGA) in the plant kingdom. Mol. Cells 33, 385–392 (2012). 22. Lalli, D. A. et al. Identification and mapping of resistance gene analogs (RGAs) in Received: 23 May 2020 Revised: 18 August 2020 Accepted: 30 August 2020 Prunus: a resistance map of Prunus. Theor. Appl. Genet. 111, 1504–1513 (2005). 23. Dhanyalakshmi, K. H., Soolanayakanahally, R. Y., Rahman, T., Tanino, K. K. & Nataraja, K. N. In Abiotic and Biotic Stress in Plants (eds. Bosco de Oliveira, A.) Leaf cuticular wax, a trait for multiple stress resistance in crop plants (Inte- chOpen, 2019). References 24. Zong, X., Denler, B. J., Danial, G. H., Chang, Y. & Song, G.- Adventitious shoot 1. Arús, P., Verde, I., Sosinski, B., Zhebentyayeva, T. & Abbott, A. G. The peach regeneration and Agrobacterium tumefaciens-mediated transient transfor- genome. Tree Genet. Genomes 8, 531–547 (2012). mation of almond × peach hybrid rootstock ‘Hansen 536’. Hort. Sci. 54, 2. Verde, I. et al. The high-quality draft genome of peach (Prunus persica) 936–940 (2019). identifies unique patterns of genetic diversity, domestication and genome 25. Petri, C., Scorza, R., Srinivasan, C. In Transgenic Plants, Vol. 847 (eds. Dunwell J., evolution. Nat. Genet. 45, 487–494 (2013). Wetten A.) Highly efficient transformation protocol for plum (Prunus 3. Peace, C. DNA-informed breeding of rosaceous crops: promises, progress and domestica L.) methods and protocols (Humana Press, 2012). prospects. Hortic. Res. 4, 17006 (2017). 26. He et al. Pm21, encoding a typical CC-NBS-LRR protein, confers broad- 4. FAOSTAT. FAO Corporate statistical database. Food and Agriculture Organi- spectrum resistance to wheat powdery mildew disease. Mol. Plant. 11, zation of the United Nations. http://www.fao.org/faostat/en/#data/QC (2018). 879–882 (2018). 5. Pascal, T., Pfeiffer, F. & Kervella, J. Powdery mildew resistance in the peach 27. Serra, O. et al. Marker-assisted introgression (MAI) of almond genes into the cultivar Pamirskij 5 is genetically linked with the Gr gene for leaf color. peach background: a fast method to mine and integrate novel variation from HortScience 45, 150–152 (2010). exotic sources in long intergeneration species. Tree Genet. Genomes 12, 96 (2016). 6. Pascal, T. et al. Mapping of a new resistance (Vr2, Rm1) and ornamental (Di2, 28. Doyle, J. J. & Doyle, J. L. Isolation of plant DNA from fresh tissue. Focus 12, pl) Mendelian trait loci in peach. Euphytica 213, 132 (2017). 13–15 (1990). 7. Dirlewanger, E., Pascal, T., Zuger, C. & Kervella, J. Analysis of molecular markers 29. Serra, O. Towards increasing genetic variability and improving fruit quality in associated with powdery mildew resistance genes in peach (Prunus persica peach using genomic and bioinformatic tools (Doctoral dissertation, Universitat (L.) Batsch) Prunus davidiana hybrids. Theor. Appl. Genet. 93, 909–919 (1996). Autònoma de Barcelona, Spain). www.tdx.cat/handle/10803/460882 (2017). 8. Foulongne, M., Pascal, T., Pfeiffer, F. & Kervella, J. QTLs for powdery mildew 30. Lindgreen, S. AdapterRemoval: Easy cleaning of Next Generation Sequencing resistance in peach x Prunus davidiana crosses: consistency across generations reads. BMC Res. Notes 5, 337 (2012). and environments. Mol. Breed. 12, 33–50 (2003). 31. Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows- 9. Grove, G. G. In Compendium of stone fruit diseases (eds Ogawa, J. M. et al.) Wheeler Transform. Bioinformatics 25, 1754–1760 (2009). Powdery Mildew (APS Press, 1995). 32. Li, H. et al. The sequence alignment/map format and SAMtools. Bioinformatics 10. Marimon, N., Eduardo, I., Martínez-Minaya, J., Vicent, A. & Luque, J. A decision 25, 2078–2079 (2009). support system based on degree-days to initiate fungicide spray programs for 33. Robinson, J. T. et al. Integrative genomics viewer. Nat. Biotechnol. 29, 24–26 peach powdery mildew in Catalonia, Spain. Plant Dis. https://doi.org/10.1094/ (2011). PDIS-10-19-2130-RE (2020). 34. Untergasser, A. et al. Primer 3—new capabilities and interfaces. Nucleic Acids 11. Aranzana, M. J. et al. Prunus genetics and applications after de novo genome Res. 40, e115 (2012). sequencing: achievements and prospects. Hortic. Res. 6, 58 (2019). 35. Hayden, M. J., Nguyen, T. M., Waterman, A. & Chalmers, K. J. Multiplex-ready 12. Donoso, J. M. et al. Exploring almond genetic variability useful for peach PCR: a new method for multiplexed SSR and SNP genotyping. BMC Genom. 9, improvement: mapping major genes and QTLs in two interspecific almond x 80–85 (2008). peach populations. Mol. Breed. 36, 16 (2016). 36. Román, B., Gómez, P., Picó, B., López, C. & Janssen, D. Candidate gene analysis 13. Pacheco-Cruz, I., Eduardo, I., Rossini, L., Vecchietti, A., Bassi, D. QTL mapping for of tomato leaf curl New Delhi virus resistance in Cucumis melo. Sci. Hortic. 243, peach (Prunus persica L. Batsch) resistance to powdery mildew and brown rot. 12–20 (2019). In Proc 53rd Italian Society of Agricultural Genetics Annual Congress (2009). 37. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. Basic local 14. Dabov, S. Inheritance of powdery mildew resistance in the peach. IV. Data alignment search tool. J. Mol. Biol. 215, 403–410 (1990). supporting the hypothesis about the main role of 2 loci controlling the 38. Zuker, M. Mfold web server for nucleic acid folding and hybridization pre- reaction to the pathogen. Genet. Sel. 16, 349–355 (1983). diction. Nucleic Acids Res. 31, 3406–3415 (2003). 15. Saunier, R. Contribution to the study of relationships between certain char- 39. Rubio, M. et al. Analysis of gene expression changes in peach leaves in acteristics of simple genetic determination in the peach tree and susceptibility response to Plum pox virus infection using RNA-Seq. Mol. Plant Pathol. 16, of peach cultivars to oidium, Sphaerotheca pannosa (Wallr.) Lev. Ann. Amelior. 164–176 (2015). Plant. 23, 235–243 (1973). 40. Zúñiga, E., Luque, J. & Martos, S. Lignin biosynthesis as a key mechanism to 16. Verde, I. et al. The Peach v2.0. release: high-resolution linkage mapping and repress Polystigma amygdalinum, the causal agent of red leaf blotch disease deep resequencing improve chromosome-scale assembly and contiguity. in almond. J. Plant Physiol. 236, 96–104 (2019). BMC Genom. 18, 225 (2017). 41. Tong, Z., Gao, Z., Wang, F., Zhou, J. & Zang, Z. Selection of reliable reference 17. Jung, S. et al. 15 years of GDR: New data and functionality in the Genome genes for gene expression studies in peach using real-time PCR. BMC Mol. Biol. Database for Rosaceae. Nucleic Acids Res. 47, D1137–D1145 (2019). 10, 71 (2009). 18. Cingolani, P. et al. A program for annotating and predicting the effects of 42. Rancurel, C., van Tran, T., Elie, C. & Hilliou, F. SATQPCR: Website for statistical single nucleotide polymorphisms, SnpEff. Fly 6, 80–92 (2012). analysis of real-time quantitative PCR data. Mol. Cell. Probe 46, 101418 (2019). 19. Chen, X. R., Brurberg, M. B., Elameen, A., Klemsdal, S. S. & Martinussen, I. 43. Vandesompele, J. et al. Accurate normalization of real-time quantitative RT- Expression of resistance gene analogs in woodland strawberry (Fragaria vesca) PCR data by geometric averaging of multiple internal control genes. Genome during infection with Phytophthora cactorum. Mol. Genet. Genom. 291, Biol. 18, 7 (2002). 1967–1978 (2016). 44. Bustin, S. A. et al. The MIQE guidelines: minimum information for publication 20. Calenge, F. & Durel, C. E. Both stable and unstable QTLs for resistance to of quantitative real-time PCR experiments. Clin. Chem. 55, 611–622 (2009). powdery mildew are detected in apple after four years of field assessments. 45. R Core Team. R: A Language and Environment for Statistical Computing Mol. Breed. 17, 329–339 (2006). (R Foundation for Statistical Computing, Vienna, Austria, 2020).