ORIGINAL ARTICLES
Validation of Molecular Markers Linked to Scab Resistance Genes Derived from Different Pear Species for Pyramiding in Japanese Pear
2025 Volume 94 Issue 2 Pages 138-147
Details
2025 Volume 94 Issue 2 Pages 138-147
Pear scab, a fungal disease caused by Venturia nashicola, is the most serious disease in Asian pear production, leading to decreases in yield and fruit quality. Five major scab resistance genes (Rvn1–Rvn4 and Vnlf) and two QTLs (Rvn5 and Rvn6) identified from different cultivars have been used in pear breeding programs to produce resistant cultivars. Here, we validated the effectiveness of three simple sequence repeat (SSR) markers (LFA02a04, LFA02a09, and LFA02b20) linked to Vnlf (possibly allelic to Rvn2) by using 29 cultivars and four breeding populations. Since there was little discrepancy between the phenotype and genotype in any of the breeding populations, with the frequency of correct classification ranging from 89% to 100%, these markers will be useful for breeding programs. In a population segregating for Rvn1 and Vnlf, the ratio of resistant to susceptible individuals fitted a 3:1 model, confirming that both genes are dominant. Although symptoms with sporulation were observed on leaves of some individuals with only Rvn1 and on those with no resistance gene, no sporulation was observed on leaves of gene-pyramided individuals (Vnlf + Rvn1) or on those carrying only Vnlf. Moreover, we revealed that resistant Chinese pear (Pyrus bretschneideri Rehder) ‘Changxili’ is heterozygous for a previously unidentified host-resistance gene that behaves as a single dominant gene with the observed segregation ratios in four progeny populations. These segregation results in combination with marker data suggest that this newly discovered gene, designated Rvn7, is at a locus distinct from both Rvn1 (derived from ‘Kinchaku’) and Rvn4 (derived from ‘Hongli’). We expect these findings to contribute to pear breeding programs working to develop resistant cultivars with single and multiple resistance genes.
Pear scab, caused by the ascomycete fungus Venturia nashicola, decreases fruit yield and is the most serious disease affecting production of Asian pear (Pyrus pyrifolia Nakai, P. ussuriensis Maxim, and P. bretschneideri Rehder). The two main pathogens that cause pear scab worldwide, V. nashicola and V. pirina, differ significantly in morphology, culture, and pathology (Ishii and Yanase, 2000). Pears (P. spp.) are classified into two major groups: Asian and European (P. communis L.) (Dondini and Sansavini, 2012). Asian and European pears can be crossed even though they are different species, and there are no major incompatibility barriers to interspecific hybridization (Westwood and Bjonstad, 1971). Asian pears are commercially produced in central and northeastern China, Siberia, Korea, and Japan, whereas European pears are traditionally grown in Europe, North America, and other non-Asian temperate regions (Itai, 2007). V. nashicola is pathogenic only to Asian pears, while V. pirina is pathogenic only to European pears (Abe et al., 2000; Ishii and Yanase, 2000; Tanaka and Yamamoto, 1964). The Japanese pear ‘Kinchaku’ and Chinese pears (P. bretschneideri) ‘Hongli’, ‘Mili’, ‘Xiangli’, and ‘Changxili’ exhibit host (species-specific) resistance to scab caused by V. nashicola, whereas almost all other Asian pear cultivars used in breeding programs are susceptible to V. nashicola (Nishio et al., 2022). European pears ‘La France’ and ‘Bartlett’ exhibit high levels of non-host resistance with no symptoms on leaves (Abe et al., 2000; Abe and Kurihara, 1993).
The host-resistance gene Rvn1 is located on linkage group (LG) 1 of ‘Kinchaku’, and DNA markers linked to it have been developed (Gonai et al., 2009, 2012; Terakami et al., 2006). Rvn4 was identified on LG 7 of ‘Hongli’ (Terakami et al., 2023). Abe et al. (2000) found that both ‘Bartlett’ and ‘La France’ are homozygous for a dominant non-host resistance gene against V. nashicola and that F1 hybrid progeny between P. pyrifolia and P. communis exhibit high resistance to V. nashicola in phenotypic surveys. A CAPS marker, PSC217-XhoI, linked to a scab resistance gene (Rvn2) from ‘Bartlett’, was developed by Cho et al. (2009) and mapped to LG 2 of the European pear ‘Navara’ by Bouvier et al. (2012). Oh et al. (2021) identified the resistance gene Rvn3 against V. nashicola on LG 6 of ‘Greensis’, a descendant of ‘Bartlett’. The quantitative trait loci (QTLs) for pear scab resistance from ‘La France’ were located on LG 2 and LG 14 of a genetic linkage map of breeding line 282-12, which is a descendant of ‘La France’ (Yamamoto et al., 2009). Other QTLs, Rvn5 and Rvn6, were detected in P019R045T042, which has several European and Japanese pears as ancestors (Won et al., 2024). The candidate genetic region of Vnlf, derived from ‘La France’, was estimated to be located on chromosome (Chr) 2: 20,220,913–21,180,341 (959 kb) of 537-14, which is a progeny of ‘La France’ (Takeuchi et al., 2023). Additionally, ‘La France’ (Vnlf) and ‘Bartlett’ (Rvn2) share the same resistance alleles for the simple sequence repeat (SSR) markers linked to Vnlf (Takeuchi et al., 2023), suggesting that Vnlf is closely linked to Rvn2 and possibly allelic to it. However, it had not previously been verified whether these Vnlf-linked markers would be practical and versatile in pear breeding programs. Therefore, it is necessary to check the consistency between phenotype and genotype using multiple populations, including some derived from major cultivars used in pear breeding.
In addition to the importance of scab diseases in pear, scab disease caused by V. inaequalis is a serious problem in commercial apple (Malus × domestica) production (Machardy et al., 2001). Both fungicide breakdown and resistance breakdown are serious problems in apple orchards, increasing the risks to sustainable apple production (Patocchi et al., 2020). Although 20 apple scab resistance genes (Rvi1–Rvi20) have been identified in various apple cultivars (Bus et al., 2011; Khajuria et al., 2018), Rvi6 is the one most often used in apple breeding programs (Moriya et al., 2019; Sansavini and Tartarini, 2013). The continued use of a single resistance gene can often result in the emergence of a new race that can infect cultivars carrying that gene, and Rvi6 resistance has been overcome (Parisi et al., 1993). Pyramiding Rvi6 with QTLs associated with resistance was effective to control scab and alleviate the impact of Rvi6 breakdown (Lasserre-Zuber et al., 2018). Similarly, Perchepied et al. (2015) reported that pear populations containing QTLs conferring resistance against pear scab caused by V. pirina exhibited stronger resistance than a population with a single QTL, and intermediate reactions in the population were shifted towards resistance.
Venturia nashicola has developed resistance to benzimidazoles and sterol demethylation inhibitor fungicides (Ishii, 2012), which are used in pear orchards to control scab, and even effective fungicides must be applied multiple times each season. For these reasons, disease-resistant cultivars are required for environmentally friendly, labor-saving scab control. Cultivars that carry Rvn1, such as the Japanese scab-resistant cultivars ‘Hoshiakari’ (Saito et al., 2021) and ‘Hoshimaru’ (Takada et al., 2023), have been developed, but to develop cultivars with even more reliable and durable scab resistance, it is necessary to identify and pyramid multiple scab resistance genes or QTLs.
In this study, we assessed the practicality and versatility of SSR markers linked to Vnlf in four populations derived from ‘La France’ or ‘Bartlett’ to assess their usefulness for pear breeding programs. In addition, we assessed the effect of resistance gene pyramiding using a population containing both Rvn1 and Vnlf. Pyramiding multiple resistance genes or QTLs contributes to the development of new cultivars with reliable and durable scab resistance. To introduce more variation in resistance genes, we investigated the inheritance of scab resistance derived from ‘Changxili’ using four populations, including one also segregating for Rvn1.
We used 29 cultivars and breeding lines (Table 1) planted in a field at the Institute of Fruit Tree and Tea Science, National Agriculture and Food Research Organization (NIFTS, NARO), in Tsukuba, Ibaraki, Japan. For validation of SSR markers linked to Vnlf and Rvn2, four populations were used (hereinafter referred to as Pop 1–4; Tables 2 and S1–S4). These populations were each derived from a cross between a susceptible Japanese pear cultivar and a resistant breeding line. The resistant lines 537-14 (pseudo-BC3) and 282-12 (F1) carry the non-host V. nashicola resistance gene Vnlf derived from the European pear ‘La France’, and 290-36 (F1) carries the non-host V. nashicola resistance gene Rvn2 derived from the European pear ‘Bartlett’. (Since 537-14 was generated using a backcross-type strategy, i.e., each generation was produced by crossing a heterozygote for scab resistance and a homozygote for susceptibility, it is considered a pseudo-backcross [pseudo-BC] breeding line.) ‘Kanta’, ‘Shurei’, and ‘Akizuki’ are all susceptible to pear scab disease. A fifth population (Pop 5; Tables 3 and S5) was derived from a cross between breeding line Tsukuba 60 and 537-14 to evaluate the level of resistance when two resistance genes are pyramided. Tsukuba 60, derived from a cross between ‘Natsushizuku’ and ‘Hoshiakari’, is heterozygous for the host-resistance gene Rvn1 which is derived from Japanese pear ‘Kinchaku’.
Phenotypes and genotypes of SSR markers associated with Vnlf in pear cultivars and breeding lines.
Resistance assessment of Pop 1–4 in inoculation tests and through marker genotyping.
Resistance assessment of Pop 5, segregating for pyramided resistance genes Vnlf and Rvn1, tested in a field without fungicide treatment.
Three additional populations (Pop 6–8) were used to clarify the inheritance of resistance in the progeny of the Chinese pear ‘Changxili’. Pop 6–8 (Tables 4 and S6–S8) were derived from a cross between a susceptible Japanese pear cultivar or breeding line and ‘Changxili’. 515-20 was derived from a cross between ‘Natsushizuku’ and ‘Hatsumaru’, both of which are susceptible to pear scab. We used a ninth population (Pop 9; Tables 5 and S9), derived from a cross between Tsukuba 60 and ‘Changxili’, to evaluate the effect of resistance gene pyramiding. This population has Rvn1 and the newly discovered resistance gene derived from ‘Changxili’ (here named Rvn7; see Results). The trees of each population were grown in pots (9 cm diameter, 30 cm height) at NIFTS, NARO, Tsukuba.
Resistance assessment of Pop 6–8 in inoculation tests or in a field without fungicide treatment.
Resistance assessment of Pop 9 with pyramided resistance genes Rvn7 and Rvn1 in a field without fungicide treatment.
The resistance of Pop 1–4 and Pop 6 to scab disease was evaluated according to the methods of Abe and Kotobuki (1998) and Iketani et al. (2001) with slight modifications (Takeuchi et al., 2023). Pop 1–4 were developed in 2021 and inoculated in 2022 or 2023, whereas Pop 6 was developed in 2019 and inoculated in 2020 and 2021. The trees were inoculated after five to eight newly unfolded leaves had appeared. Fresh conidia of V. nashicola were collected from infected leaves of the Japanese pears ‘Chojuro’ and ‘Kosui’ grown in a NIFT field that has never been sprayed with fungicide; the samples were collected in 2017 and 2011, respectively, and stored at −80°C until use. The conidia were suspended in 0.1% sucrose solution and adjusted to 2.5 × 105 per mL before inoculation. The conidial suspension was sprayed over the entire plant surface using a mist sprayer at a rate of 200 mL per 100 plants. The inoculated plants were air dried, placed in a moist chamber at 20°C and 90%–95% relative humidity for 48 h in the dark, and then transferred to a greenhouse, where the temperature was set at 20–30°C using natural ventilation under ambient light and humidity conditions. Four weeks after inoculation, we observed 10 unfolded leaves (or as many as had unfolded if fewer than 10) from each plant. The resistance of each leaf was classified into one of four classes according to Takeuchi et al. (2023): class 0, no symptoms; class 1, chlorotic lesions; class 2, pinpoint necrotic lesions but no sporulation; class 3, symptoms with sporulation. Individuals with class 3 leaves were classified as susceptible, and all others were classified as resistant. Disease scores were calculated by averaging the class scores of 10 leaves. When the population was observed for two years or twice in the same year, the disease score for each individual was determined by calculating the average of the two observations.
Pop 5, Pop 7, and Pop 8 were evaluated for resistance to Japanese pear scab disease in the NIFT field that had never been sprayed with fungicide. These populations were developed in 2022 and planted in the field in April 2023. They were observed twice in June 2023 as above.
DNA extraction and SSR genotypingGenomic DNA was extracted from young leaves (10 mg) of all individuals using a NucleoMag Plant Kit (Macherey-Nagel, Duren, Germany) following the manufacturer’s instructions with a slight modification to buffer MC1 (2-mercaptoethanol was added to a final concentration of 2%). Pop 1–5, the cultivars, and the breeding lines were genotyped using three SSR markers linked to Vnlf (LFA02a04, LFA02a09, and LFA02b20; Takeuchi et al., 2023). Pop 5 and Pop 9 were genotyped using two SSR markers linked to Rvn1 (TsuENH157 and TsuENH101; Gonai et al., 2012). Pop 6 was genotyped using TsuENH157, TsuENH101, and a marker linked to Rvn4 (TsuENH165; Yamamoto et al., 2013; Terakami et al., 2023) to determine whether ‘Changxili’ carried Rvn1 or Rvn4. Table 6 describes the resistance genes and DNA markers used in this study.
Genes for resistance to pear scab disease (V. nashicola) and DNA markers used in this study.
PCR amplification for SSR genotyping was performed in a 10-μL reaction mixture containing 2.5 μL of 2× GoTaq Green Master Mix (Promega, Madison, USA), 0.5 μM forward primer (fluorescently labeled with FAM or VIC), 0.5 μM reverse primer, and 1.0 μL of genomic DNA (10–40 ng·μL−1). A 7-bp pigtail sequence (5′-gtttctt-3′) (Brownstein et al., 1996) was added at the 5′ end of each reverse primer to facilitate accurate genotyping. Multiplex PCR amplification was performed with up to two primer combinations per reaction. The PCR conditions were initial denaturation for 5 min at 95°C; 35 cycles of denaturation for 60 s at 94°C, annealing for 60 s at 55°C, and extension for 60 s at 72°C; and a final extension for 7 min at 72°C (GeneAmp PCR System 9700; Thermo Fisher Scientific Inc, Walthman, USA).
The PCR products were separated and detected in an Applied Biosystem 3730 xl Genetic Analyzer (Thermo Fisher Scientific Inc.) with a 36-cm capillary array and POP-7 polymer. The size of each amplified band was determined by comparison with a set of internal standard DNA fragments (GeneScan 400HD ROX Dye Size Standard; Thermo Fisher Scientific Inc.) in GeneMapper v. 5.0 software (Thermo Fisher Scientific Inc.).
The resistance or susceptibility to Japanese pear scab disease (Abe et al., 2008; Nishio et al., 2022; Takeuchi et al., 2023) and the genotypes of SSR markers linked to Vnlf of 23 cultivars and six breeding lines are shown in Table 1. The markers LFA02a04, LFA02a09, and LFA02b20 amplified clearly distinguishable fragments in all cultivars and breeding lines. The resistant European pear cultivars ‘La France’, ‘Bartlett’, and their progeny (537-14, 403-10, 282-12, and 290-36) were heterozygous for Vnlf or Rvn2 on the basis of band sizes amplified from LFA02a04, LFA02a09, and LFA02b20. In other resistant cultivars and breeding lines that do not have Vnlf or Rvn2, namely, ‘Hoshiakari’, ‘Hoshimaru’, ‘Kinchaku’, Tsukuba 60, ‘Changxili’, and ‘Hongli’, the resistance-specific bands of these markers were not detected.
Validation of SSRs linked to Vnlf or Rvn2 using four populationsOut of the 94 individuals in Pop 1 inoculated with conidia of V. nashicola, 42 were classified as resistant and 52 as susceptible (Tables 2 and S1). The segregation of resistant and susceptible individuals fitted a 1:1 ratio in the chi-squared test. In case of any discrepancy between the estimated genotypes of the three markers, we used LFA02a09, which is closest to Vnlf (Takeuchi et al., 2023), to determine the genotype. All resistant individuals were heterozygous for Vnlf, while all susceptible individuals were homozygous for vnlf; in other words, there was 100% correspondence between the LFA02a09 genotype and the resistance phenotype. Out of the 57 individuals in Pop 2, 22 were classified as resistant and 35 as susceptible (Tables 2 and S2). The segregation of resistant and susceptible fitted a ratio of 1:1 in the chi-squared test. Out of the 22 individuals with a resistant phenotype, 16 were heterozygous for Vnlf, but six were homozygous for vnlf, and the frequency of correct classification was 89.5%. All individuals classified as susceptible were homozygous for vnlf. Out of the 60 individuals in Pop 3, 25 were classified as resistant and 35 as susceptible (Tables 2 and S3). The segregation of resistant and susceptible fitted a ratio of 1:1 in the chi-squared test. Among the 25 individuals with a resistant phenotype, three were homozygous for vnlf, while the remaining 22 were heterozygous for Vnlf, and the frequency of correct classification was 95.0%. Out of the 90 individuals in Pop 4, 43 were classified as resistant and 47 as susceptible (Tables 2 and S4). The segregation of resistant and susceptible individuals fitted a ratio of 1:1 in the chi-squared test. Twenty-eight of the resistant individuals were heterozygous for Vnlf, but 15 were homozygous for vnlf, while 36 susceptible individuals were homozygous for vnlf, but 11 were heterozygous for Vnlf; thus, the frequency of correct classification in this population was 71.1%.
The effect of pyramiding Rvn1 and VnlfIn Pop 5, segregating for both Rvn1 and Vnlf, 24 individuals were classified as resistant and 13 as susceptible based on observations in the field without fungicide treatment (Tables 3 and S5). The segregation of resistant and susceptible individuals fitted a ratio of 3:1 in the chi-squared test. The genotypes of Vnlf were determined using LFA02a09, and the genotypes of Rvn1 were determined using the SSR markers TsuENH157 and TsuENH101 (Table S5). Four genotypes were identified: those carrying Vnlf + Rvn1 (n = 7), those carrying only Rvn1 (n = 14), those carrying only Vnlf (n = 11), and those with no resistance genes (n = 5). The genotypes segregated in a ratio that fitted 1:1:1:1 in the chi-squared test (χ2 = 5.2703, df = 3 and P = 0.153). Average disease scores were 0.000 for genotypes carrying Vnlf + Rvn1, 0.420 for Rvn1, 0.005 for Vnlf, and 1.390 for none (Fig. 1).
The relationship between genotype and disease score in Pop 5, which segregated for Rvn1 and Vnlf (later renamed Rvn2). The boxplots indicate the median (dark horizontal bars), interquartile values (light horizontal bars), and range for each genotype, with the red points showing disease scores for resistant individuals (no sporulation) and the green points for susceptible individuals (showing sporulation). The genotypes were determined by SSR markers linked to Rvn1 and Vnlf.
Out of the 94 individuals in Pop 6 (susceptible ‘Natsushizuku’ × resistant ‘Changxili’) that were inoculated with conidia of V. nashicola in 2 years, 49 were classified as resistant and 45 as susceptible (Tables 4 and S6). The segregation of resistant and susceptible individuals fitted a ratio of 1:1 in the chi-squared test. The genotypes of Rvn1 estimated by SSR marker TsuENH157 were 168/170 (bp) for ‘Changxili’ and 175/175 for ‘Natsushizuku’, while those detected by TsuENH101 were 132/150 for ‘Changxili’ and null/null for ‘Natsushizuku’. None of these marker genotypes matched those linked to resistance at Rvn1 in other populations, namely, the 172-bp allele at TsuENH157 and the 138-bp allele at TsuENH101. The genotypes of Rvn4 estimated by SSR marker TsuENH165 were 270/276 for ‘Changxili’ and 276/276 for ‘Natsushizuku’. Pop 6 was divided into two genotypes each by TsuENH157 (168/175 or 170/175), TsuENH101 (150/null or 132/null), and TsuENH165 (270/276 or 276/276), and the average disease score of the individuals with 168/175 was 1.1, with 170/175 was 1.2, with 150/null was 1.2, with 132/null was 1.1, with 270/276 was 1.2, and with 276/276 was 1.1, providing further evidence that neither Rvn1 nor Rvn4 is linked to the resistance caused by ‘Changxili’ (Table S6). Two additional populations (Pop 7 and Pop 8), which were produced from a cross of a susceptible parent with ‘Changxili’, were observed in the field without fungicide treatment (Tables 4, S7, and S8). Out of 99 individuals in Pop 7, 59 were classified as resistant and 40 as susceptible. Out of 87 individuals in Pop 8, 52 were classified as resistant and 35 as susceptible. The chi-squared test showed that the segregation of resistant and susceptible individuals of both Pop 7 and Pop 8 fitted a ratio of 1:1. These results suggest that ‘Changxili’ carries a resistance gene that is different from Rvn1 and Rvn4, herein named Rvn7.
Segregation of resistance in a population carrying two resistance genes, Rvn1 and Rvn7In Pop 9, segregating for both Rvn1 and Rvn7, 17 individuals were classified as resistant and 11 as susceptible on the basis of observations in the field without fungicide treatment (Tables 5 and S9). The segregation of resistant and susceptible individuals fitted a ratio of 3:1 in the chi-squared test. When the genotypes of Rvn1 were estimated using the SSR markers TsuENH157 and TsuENH101, 17 individuals were classified as Rvn1/rvn1 and 11 as rvn1/rvn1, with average disease scores of 0.1 and 0.7, respectively. Among the 17 individuals that carried Rvn1, 12 were classified as resistant and five as susceptible, with average disease scores of 0.01 and 0.40, respectively. Of the 11 individuals without Rvn1, five were classified as resistant and six as susceptible, with average disease scores of 0.02 and 1.21, respectively (Fig. 2).
The relationship between genotype and disease score in Pop 9, which segregated for Rvn1 and Rvn7. The boxplots indicate the median (dark horizontal bars), interquartile values (light horizontal bars), and range for each genotype, with the red points showing disease scores for resistant individuals (no sporulation) and the green points for susceptible individuals (showing sporulation). The genotypes were determined by SSR markers linked to Rvn1.
Three SSR markers (LFA02a04, LFA02a09, and LFA02b20) linked to the scab resistance gene Vnlf derived from ‘La France’ were used to genotype 23 cultivars, six breeding lines, and four breeding populations. The predicted genotypes of Vnlf based on the three SSR markers and resistance phenotypes fully agreed in all of the tested non-host-resistant cultivars and breeding lines, that is, ‘La France’, ‘Bartlett’, and their progeny. The resistance-specific allele linked to Vnlf was not detected in susceptible cultivars or in host-resistant cultivars that carry Rvn1, Rvn4, or Rvn7. On the basis of the segregation ratio of 1:1 of resistant to susceptible individuals in all breeding populations with Vnlf (Pop 1 and Pop 2) or Rvn2 (Pop 3 and Pop 4) in inoculation tests, we reconfirmed that the non-host-resistance gene on Chr 2 is inherited as a single dominant gene (Abe et al., 2008; Takeuchi et al., 2023).
In Pop 1, there was no discrepancy between the estimated genotypes and the phenotypes. In Pop 2 and Pop 3, most individuals had agreement between the estimated genotypes and the phenotypes, but some individuals predicted to be susceptible by SSRs were classified as resistant in an inoculation test. In Pop 4, there was some discrepancy between the genotypes and the phenotypes of both resistant and susceptible individuals. In a previous study, the parental cultivars/breeding lines of Pop1–4 were shown to be highly resistant (Abe et al., 2000; Takeuchi et al., 2023), and here, the presence or absence of disease could be clearly scored in Pop1–4. The pollen parent of Pop 1, 537-14, was also used to construct a mapping population with which the markers were developed and the pseudo-BC3 of ‘La France’, while the pollen parents of Pop 2–4, 282-12 or 290-36, were F1 progeny of ‘La France’ or ‘Bartlett’, respectively. The resistance genes in ‘Bartlett’, Rvn2 and Rvn3, were identified on LG 2 and LG 6, respectively (Bouvier et al., 2012; Cho et al., 2009; Oh et al., 2021). There was no discrepancy between the genotype and phenotype scores for Pop 1 showing that the LFA02a09 marker and the resistance gene are closely linked. Therefore, the discrepancies between genotype and phenotype in Pop 2 and Pop 3 are suggested to be influenced by other resistance genes, although recombination may also play a role. Our data, together with the pedigree of 290-36, suggest that this line may carry both Rvn2 and Rvn3, and that its progeny, which have no Rvn2, but were observed to be resistant in the inoculation test, may carry Rvn3. Also, in Pop 2, six individuals that had no detected resistance alleles showed resistance in the inoculation test, suggesting that ‘La France’ and 282-12 carry not only Vnlf, but also other dominant gene(s)/QTL(s) conferring resistance. Given that two resistance QTLs were identified on LG 2 and LG 14 of 282-12 (Yamamoto et al., 2009), other resistance loci are predicted to be located on LG 14. However, some susceptible individuals in Pop 4 had marker genotypes linked to resistance. Pop 3 and Pop 4 have a common pollen parent, but different seed parents. We therefore assume that the seed parent of Pop 4, ‘Akizuki’, contains factors that negatively regulate the non-host resistance conferred by Rvn2. In rice blast disease, the susceptibility QTL pi21 was reported to negatively regulate resistance on the basis of the observation that suppression of pi21 expression increased resistance (Fukuoka et al., 2009). ‘Kanta’ (the seed parent of Pop 1) is a descendant of ‘Akizuki’ (the seed parent of Pop 4), and ‘Shurei’ (the seed parent of Pop 2 and Pop 3) has some common ancestors with ‘Akizuki’, but the same phenomenon (i.e., possible negative regulation of resistance conferred by Vnlf or Rvn2) was not observed in Pop 1–3. It is necessary to verify whether this phenomenon occurs in other populations for which the seed parent is ‘Akizuki’, or when using other cultivars as the seed parent. Even though LFA02a09 is close to the resistance gene (Takeuchi et al., 2023), the apparent recombination rate in Pop 4 was higher than that of other populations, and we speculate that resistance in this population may be influenced by other genes, so we evaluated the practicality and versatility of the SSR markers in Pop 1–3. On the basis of the determined genotypes of the parental cultivars and of Pop 1–3, the three SSR markers (LFA02a04, LFA02a09, and LFA02b20) were useful for selecting individuals that carry Vnlf or Rvn2, which were correctly identified at a frequency of 89%–100%. ‘La France’ and ‘Bartlett’ and their progeny (breeding lines 282-12, 403–10, and 537-14; reported to carry Vnlf) shared the same resistance allele at each of the three SSR markers (Table 1; Takeuchi et al., 2023). Here, we revealed that 290-36 (F1 progeny of ‘Bartlett’; reported to carry Rvn2) had the same three alleles, which segregated in its progeny population (Pop 3), suggesting that Vnlf and Rvn2 are either alleles or tightly linked genes (Hencefortth, Vnlf will be renamed Rvn2).
To produce more durable scab-resistant pear cultivars, we constructed breeding populations containing multiple scab resistance genes in our pear breeding program. Pop 5 had two resistance genes, Rvn1 and Vnlf, derived from ‘Kinchaku’ and ‘La France’, respectively. The segregation ratio of resistant and susceptible individuals in this population fitted a 3:1 model, indicating little if any segregation distortion, even though the two genes came from different pear species. This segregation ratio fits the expected Mendelian ratio for a model in which two dominant genes are segregating. The population was classified into four genotypes: Vnlf + Rvn1, Rvn1 only, Vnlf only, and no resistance gene, using the markers TsuENH157 (for Rvn1), TsuENH101 (for Rvn1), and LFA02a09 (for Vnlf). The segregation ratio of the four genotypes fit a ratio of 1:1:1:1, which is consistent with Tsukuba 60 and 537-14 each being heterozygous for a different dominant gene. All individuals with the Vnlf + Rvn1 or Vnlf-only genotype were classified as resistant, while all individuals without either gene were classified as susceptible in the field assessment. Notably, nine of the 14 individuals with only Rvn1 were susceptible. The disease scores were generally lower for individuals with only Rvn1 than for those without resistance genes, but leaves with sporulation (class 3) were observed in all nine susceptible individuals. In apple scab disease, the breakdown of some resistance genes has occurred (Patocchi et al., 2020). It was assumed that the breakdown of Rvi6 in apple was caused by the selection and mutation of an Rvi6-virulent gene in a population of V. inaequalis maintained in a wild apple species (Malus floribunda) following the introduction of new resistant hosts (Lemaire et al., 2016). Detecting resistance gene breakdown is not always clear-cut (Patocchi et al., 2020). Since 2021, the Japanese scab-resistant pear cultivar ‘Hoshiakari’ (Rvn1) has occasionally showed sporulation in the NIFT field without fungicide (Yoshida et al., 2023). The development of pear scab disease in ‘Hoshiakari’ and its progeny needs to be continuously observed by plant pathologists to reveal the cause of disease onset. Since the gene-pyramided individuals (Vnlf + Rvn1) had no leaves with sporulation, we suggest that pyramiding of major genes reinforces resistance, even if the effect of one gene is low.
The Asian pear ‘Changxili’ (P. bretschneideri) has host resistance for the scab disease caused by V. nashicola (Abe et al., 2008). ‘Changxili’ was revealed to be a heterozygote for a single dominant gene, now called Rvn7, on the basis of a segregation ratio of 1:1 between resistant and susceptible individuals in Pop 6. Furthermore, we evaluated the phenotypes of two populations (Pop 7 and Pop 8) with ‘Changxili’ as the pollen parent in the field without fungicide treatment. The segregation ratio of both populations was 1:1, confirming that ‘Changxili’ is heterozygous for a single dominant gene. The genotypes of SSR markers linked to Rvn1 and Rvn4 were checked in Pop 6 and its parents. There was no difference in the average disease score between different genotypes at each of these loci, suggesting that Rvn7 is a different locus from both Rvn1 and Rvn4, and that it represents a newly identified host-resistance gene. Pop 9 has two host-resistance genes, Rvn1 and Rvn7, both of which are single dominant genes. The ratio of resistant to susceptible individuals was 3:1, consistent with the expected inheritance of a single dominant gene contributed by each parent. Genotyping using the SSR markers linked to Rvn1 revealed that 17 out of 28 individuals were Rvn1 heterozygotes; however, six of these 17 were classified as susceptible in the field without fungicide treatment. A similar phenomenon was observed with Pop 5 (which segregates for both Rvn1 and Vnlf), which we assume to be due to the reduced effect of Rvn1. Some individuals in Pop 9 without Rvn1 were estimated to be resistant in field observations, and we assume that these carried only Rvn7. Thus, to select individuals that definitely carry both Rvn1 and Rvn7, evaluation of the phenotype alone is not sufficient. Developing new DNA markers that enable genotyping of Rvn7 is essential to produce cultivars that have multiple resistance genes.
Development of scab-resistant cultivars is necessary to reduce the use of fungicides and improve fruit quality. However, continued use of a single resistance gene in pear orchards may cause gene breakdown due to selection and mutation of the pathogenic fungus. It is essential to reduce the risk of genetic breakdown by increasing the diversity of resistance genes in pear orchards. This study revealed that the SSR markers linked to Rvn2 (Vnlf) are useful for selecting individuals and that gene pyramiding is effective for reinforcing resistance, as evidenced by the evaluation of individuals with multiple resistance genes. Furthermore, the discovery of Rvn7 provides an additional gene for use in such pyramids. We expect these findings to contribute to breeding programs aiming to develop several resistant cultivars with single and multiple resistance genes.
We are deeply indebted to everyone involved in the Japanese pear breeding program at the Institute of Fruit Tree and Tea Science, NARO.
