Marker-Assisted Evaluation of Two Powdery Mildew Resistance Candidate Genes in Korean Cucumber Inbred Lines

: Two genes, CsLRR-RPK2 ( CsGy5G015660 ) and CsaMLO8 ( Csa5G623470 ), have been considered as powdery mildew (PM) resistance genes in cucumbers. In this study, we evaluated the involvement of the alleles of these two genes in PM resistance in 100 commercial Korean cucumber inbred lines. To achieve this, we developed cleaved ampliﬁed polymorphic sequences (CAPS) and InDel markers from CsLRR-RPK2 and CsaMLO8 . Genotyping analysis indicated that the CsLRR-RPK2 -CAPS marker showed a stronger correlation with the PM-resistant phenotype, with an 84% consistency compared to the CsaMLO8 -InDel marker. The use of the CsaMLO8 -InDel marker showed a 70% consistency between phenotype and genotype results. It was proposed that the CsLRR-RPK2 -CAPS marker successfully eliminated PM-susceptible inbred lines, since both genotype and phenotype results were 100% identical. Furthermore, the present study revealed that the introduction of one of these alleles is probably enough to confer PM resistance in cucumbers. However, seven PM-resistant inbred lines harbored either CsaMLO8 or CsLRR-RPK2 alleles, indicating that there is another PM-resistant resource(s) besides CsaMLO8- and CsLRR-RPK2 –originated resistance in the commercial Korean inbred lines. Our results provide reliable evidence conﬁrming two PM-resistant candidate genes for the detection of PM resistance resources in cucumber inbred lines.

It is necessary to understand the genetic and molecular mechanisms of PM resistance in the breeding of PM-resistant cucumber cultivars. Several efforts have been made to identify the PM resistance loci in cucumber. As a result of these efforts, six temperaturedependent QTLs (chromosomes 1, 5, 6, and 7) from PI 197088-1, five PM-resistance QTLs (PM1. 1 [1,6,[19][20][21][22][23] have been identified. Although we have deep insight into the genetic inheritance of PM resistance, it is very difficult to directly use the information on PM-resistant QTLs to select the PM-resistant germplasm. From a practical point of view, molecular markers for the selection of the PM-resistant germplasm from the above QTL regions should be developed. Several studies have reported candidate gene(s) that govern PM resistance in different cucumber germplasms. For example, Xu et al. [24] identified two genes (Csa1G064780 and Csa1G064790) encoding a putative cysteine-rich receptor-like kinase from the 41.1-kb region of the Jin5-508-derived SSSL0.7 line. Three possible candidate genes, Csa5G622800, Csa5G622830, and Csa5G623490 that encode, identified from Cucumis hystrix introgression lines, and a single recessive gene (Csa5G622830), which encodes GATA transcription factors, have been considered candidates for controlling PM resistance [25]. From PI 197088 and C. sativus cv. Anaxo, a single-recessive gene (CsaMLO8, Csa5G623470) is considered a strong candidate gene for PM-susceptibility [16,17,26,27].
To date, CsaMLO8 is the only cucumber PM-resistant gene whose function has been identified. Our previous recent study identified CsGy5G015660, which encodes a putative leucine-rich repeat receptor-like serine/threonine kinase (CsLRR-RPK2), as a strong PMresistant candidate gene derived from a Korean inbred line [28]. In the current study, we applied PM-resistant molecular markers derived from two PM-resistant candidate genes (CsaMLO8 and CsLRR-RPK2) to 100 Korean inbred lines to evaluate the accuracy of marker-assisted selection. Furthermore, we attempted to understand the origin of PM resistance in Korean cucumber inbred lines.

Plant Materials and Cultivation
In the present study, we used 100 commercial Korean cucumber inbred lines, including sixty-two PM resistant lines. The F 1 progenies were generated by crossing different PMsusceptible (GHDHW, HD, LC, CPA, RV, FPDN and MDSS) and PM-resistant parental lines (CA/794, 2K79, 64E2K9, 2JKJ, 249, 253XNH, K9AKJ, 64E3K9, 045911., KJK9, JCDS and 02789AGN). Detailed information is shown in Table 1. In this study, twelve different PM-resistant resources were used as PM-resistant donor plants (Table 1). Consecutive self-pollinations from these progenies were performed to obtain the F 3 -F 9 generation and pure lines. The seeds used in this study were kindly supplied by DONGOH SEED (Ansung, Korea). All seeds used in this study were germinated indoors in the spring of 2020 and then transplanted to DONGOH SEED greenhouse. All plants were irrigated, fertilized, and treated with pesticides according to the standard cultivation protocol of DONGOH SEED.

Evaluation of PM Response
Disease symptoms caused by natural infection of P. xanthii under greenhouse conditions became apparent 60 days after transplantation. When PM disease symptoms were clearly distinguishable, phenotypic scoring of PM resistance was conducted at 2-week intervals until disease symptoms had spread to all parts of the susceptible plants. The plants were considered to be PM-susceptible when the white spores of P. xanthii covered more than 10% of the overall cucumber leaves. Others were considered to be PM resistant. Ten individuals from each cucumber inbred line were used for the evaluation of PM response. Finally, 100 Korean cucumber inbred lines were clearly distinguishable in the PM-resistant and PM-susceptible groups. The PM resistance and PM susceptibility are presented in Figure 1 and Table 2.

Genomic DNA Isolation and Molecular Marker Design
Fresh young leaves of 100 Korean cucumber germplasms were used to isolate genomic DNA (gDNA). The DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) was used to isolate gDNA. The quality and quantity of the gDNAs were calculated using a Nanodrop ND-2000 spectrophotometer, following the manufacturer's instructions (Thermo Fisher Scientific, Waltham, MA, USA). Geneious Pro 9.1.8 (Biomatters, Auckland, New Zealand) was used to (I) design InDel and cleave amplified polymorphic sequences (CAPS) markers for CsaMLO8 and CsLRR-RPK2, respectively; and (II) identify appropriate restriction enzymes and restriction sites for the CsLRR-RPK2 gene. HinfI enzyme (New England BioLabs, Ipswich, MA, USA) was selected and used as a restriction enzyme for the CsLRR-RPK2-CAPS marker. The sequences of primers for both markers are shown in Table 3, while the gene organization is presented in Figure 2. In this study, Gy14 (version 2.0) and Chinese Long (version 2.0) genome sequences were used as the cucumber reference sequences. Table 3. Information on CsLRR-RPK2-CAPS and CsaMLO8-InDel markers.

Genotyping of Cucumber Inbred Lines
For the genotyping of CsaMLO8 and CsLRR-RPK2 alleles, we developed InDel and CAPS markers, respectively. CsaMLO8-InDel markers were composed of a combination of a forward primer (5 -TATGGCTGCCTTTCATCTCCT-3 ) and reverse primer (5 -TCCAAGCAAAGAAGGCAAGC-3 ). DNA samples were amplified using CsaMLO8-InDel markers and followed by a polymerase chain reaction (PCR) reaction program: 94 • C for 5 min, 30 cycles of 94 • C for 20 s, 65 • C for 20 s, and 72 • C for 90 s, with a final cycle at 72 • C for 5 min. CsLRR-RPK2-CAPS markers were composed of a combination of forward primer (5 -GCAACAAGTTCAATGGACCAC-3 ) and reverse primer (5 -GAATCTCTCCAGTCAAATTGTTTCC-3 ). DNA samples were amplified using CsLRR-RPK2-CAPS markers and the PCR program was as follows: 94 • C for 5 min, 35 cycles of 94 • C for 20 s, 60 • C for 10 s, and 72 • C for 20 s, with a final cycle at 72 • C for 5 min. For the PCR, Maxime PCR PreMix (iNtRON Biotechnology, Seoul, Korea) was used. PCR amplifications were performed in a Bio-Rad T100™ thermocycler (BioRad, CA, USA). HinfI, a restriction enzyme, was added to the PCR amplicons and incubated for 1 hour at 37 • C. The final reaction products were run on a 2% agarose gel with 0.5× Tris-borate-EDTA (TBE) buffer and visualized by staining with RedSafe (iNtRON Biotechnology, Seoul, Korea).

Results and Discussion
The evaluation of PM resistance and genotyping analysis using the markers for CsaMLO8 and CsLRR-RPK2 alleles in 100 Korean cucumber inbred lines is shown in Table 2 and Figure 3. Of the 100 cucumber inbred lines, 62 had a PM-resistant phenotype. In contrast, 38 inbred lines had a PM-susceptible phenotype. The two PM resistance candidate genes used in this study were CsaMLO8 (Csa5G623470) and CsLRR-RPK2 (CsGy5G015660). The CsLRR-RPK2-CAPS marker amplified the flanking region of the SNP (T/C) (Figures 2 and 3a). The SNP of the PM-susceptible line produced double bands (224 bp and 100 bp) after HinfI digestion, whereas the PM-resistant line showed an absence of restriction sites in the amplicon (Figure 3b). The CsLRR-RPK2-CAPS marker was successfully applied to detect the PM-resistant and PM-susceptible alleles using the ten phenotypically identified (five PM-resistant and five PM-susceptible lines) inbred lines (Figure 3c). The CsaMLO8-InDel marker detected 1459 bp insertion in PM-resistant lines (Figures 2 and 3d). Ten phenotypically identified inbred lines, consisting of five PMresistant and five PM-susceptible lines, were also verified for the successful application of CsaMLO8-InDel (Figure 3d).
To compare the involvement of CsaMLO8 and CsLRR-RPK2 alleles in the PM-resistant phenotype, we applied CsaMLO8-InDel and CsLRR-RPK2-CAPS markers on 100 inbred lines. Genotyping analysis using the CsaMLO8-InDel marker revealed that 30 inbred lines were inconsistent with the phenotypic evaluation results (Table 2). In particular, most of the inconsistency was observed in PM-resistant lines; 44% (27/62) of PM-resistant inbred lines had PM-susceptible CsaMLO8 alleles. Only 11% (3/28) of the PM-susceptible inbred lines (417, 420, and 429) had the PM-resistant CsaMLO8 allele. However, genotyping analysis using the CsLRR-RPK2-CAPS marker revealed fewer inconsistent results. In detail, 16 inbred lines showed inconsistency between CsLRR-RPK2-based genotyping and phenotypic evaluation results. Interestingly, all of the inconsistent results were obtained from the PM-resistant inbred lines. This indicates that analysis of the CsLRR-RPK2 allele is useful for the elimination of PM-susceptible lines during PM resistance breeding programs. More interestingly, the seven PM-resistant inbred lines (215, 219, 223, 225, 241, 394, and 424) based on phenotypic evaluation showed none of the PM resistance alleles of CsaMLO8 and CsLRR-RPK2. This indicates that the PM resistance of these seven inbred lines did not originate from CsaMLO8 or CsLRR-RPK2, but originated from different PM resistance sources. Further studies should be conducted, using these PM-resistant inbred lines, to identify PM-resistant QTL(s) and/or PM-resistant gene(s) to reveal the origin of a new PM-resistant source(s). Furthermore, we classified the sources of PM resistance used in this study (Tables 1 and 2). We detected three different types of PM resistant sources from 12 PM-resistant germplasms used for the generation of 62 PM resistance inbred lines. Firstly, a PM resistance source originated from the CsLRR-RPK2 gene, represented as A, J, and K ( Table 2; Supplementary Figure S1). Secondly, a PM resistance source originated from both the CsLRR-RPK2 and CsaMLO8 genes, represented as B, D, E, F, G, and H (  Figure S1). Thirdly, a PM resistance source originated from an unidentified PM resistance source and also contained both resistance genes (CsLRR-RPK2 and CsaMLO8), represented as C, I, and L (Table 1). These results reveal that divergent sources (at least 1-3 PM-resistant genes were introduced) of PM resistance have been introduced in PM-resistant cucumber breeding programs in Korea. Further extensive study and comparison of phenotypic evaluation and genotyping analysis using increased numbers of PM-resistant sources and their progenies should be conducted, because a limited number of PM-resistant sources and progenies were used in this study.
Comparison of genotyping analysis results between CsaMLO8 and CsLRR-RPK2 showed 32% (32 inbred lines) inconsistency ( Table 1). The three PM-susceptible lines (417, 420, and 429) showed PM-resistant CsaMLO8 alleles, but not PM-resistant CsLRR-RPK2 alleles. Twenty-nine PM-resistant inbred lines showed either PM-susceptible alleles of CsaMLO8 (21 lines) or CsLRR-RPK2 (8 lines). This indicates that PM resistance could be acquired by the introgression of either one of the alleles of CsaMLO8 and/or CsLRR-RPK2. Although current phenotypic evaluation analysis results did not strongly support the additive effect of both alleles of CsaMLO8 and CsLRR-RPK2 on the number of PM resistance, we observed a moderate correlation between the degree of PM-resistant alleles and the PM-resistant phenotype (data not shown). This fact should be verified by further extensive phenotypic evaluation of divergent PM-resistant inbred lines.
When we compared the allele types of CsaMLO8 and CsLRR-RPK2, 68 inbred lines showed alleles of both PM resistance and PM susceptibility. However, 32 inbred lines showed different allele types between CsaMLO8 and CsLRR-RPK2. This fact can be explained in different ways. Considering the high proportion (more than two-fold) of the same alleles of CsaMLO8 and CsLRR-RPK2, it may be thought that both alleles could easily co-migrate during meiosis. However, both genes (CsLRR-RPK2 and CsaMLO8) are located 8.8 Mb apart on Chromosome 5. When we consider the genetic recombination (bin) of Chromosome 5 as 0.37 cM, equivalent to 228 kb [29], it is unlikely to co-migrate during meiosis. In addition, it has been proposed that three segregation distorted regions (SDR) could exist on Chromosome1, 4, and 6 [29]. The SDRs also cannot explain the co-migration of both allele pairs, because the Chromosome 5 is not likely to contain SDRs. Considering the fact that most inconsistencies were observed before the F 4 generation, each allele of CsLRR-RPK2 and CsaMLO8 probably followed independent meiotic recombination events. Race identification of PM has been well identified in melons [30][31][32][33][34][35][36]. For example, 12 races of PM (race 1, 2, 3, 4, 5, 6, 7, N3, N4, A, O, and S) have recently been identified in melons [33]. However, race identification of PM in cucumbers has not been well identified. In Korea, Race 1 of PM has been proposed to be dominant. Therefore, we do not have a detailed population structure of the races of PM in cucumbers. This suggests that each allele of CsLRR-RPK2 and CsaMLO8 may be responsible for different races of PM in cucumbers. Breeders should continuously select the cucumber plants that show strong resistance to PM. Therefore, consecutive selections of PM-resistant lines were followed after self-pollination. The consecutive selection of PM-resistant lines from generation to generation may accumulate PM-resistant allele pairs (CsaMLO8 and CsLRR-RPK2) in PM-resistant lines. This could be a more plausible explanation than the co-migration of both allele pairs during meiotic division.

Conclusions
In this study, we applied PM-resistant molecular markers derived from two PMresistant candidate genes (CsaMLO8 and CsLRR-RPK2) to 100 Korean inbred lines to evaluate the accuracy of marker-assisted selection and to understand the origin of PM resistance in breeding inbred lines in Korea. Our results provided reliable evidence that confirms two PM-resistant candidate genes and efficiently utilized these genes for the detection of PM-resistant resources and inbred lines. Furthermore, the CsLRR-RPK2-CAPS marker in comparison to the CsaMLO8-InDel marker showed a strong correlation with the PM-resistant phenotype. In addition, our results revealed that unidentified PM-resistant sources used in breeding programs in Korea are not related to either PM-resistant candidate gene (CsaMLO8 and CsLRR-RPK2).

Data Availability Statement:
The data shown in this paper are available from the corresponding author by request.