Powdery mildew is a common disease of pea (Pisum sativum L.) with worldwide distribution and can result in yield losses up to 50% (Warkentin et al. 1996). It is particularly prevalent in areas with a warm, humid climate. The first symptoms usually appear on the lowest part of the plant, with small, but distinct, spots on the leaflets. Later, these lesions grow, covering the plant surfaces and affecting the growth of the plant and the quality of harvested seeds (Fondevilla and Rubiales 2012). Given the high cost and detrimental environmental impacts associated with fungicide use, the use of powdery mildew-resistant cultivars is a sustainable, efficient, and environmentally friendly method to control this disease.

Three monogenic sources of powdery mildew resistance were identified in pea named er1, er2, and Er3 (Harland 1948; Heringa et al. 1969; Fondevilla et al. 2007). Recessive er1 resistance alleles have been successfully cloned and shown to carry loss-of-function mutations of the pea MLO homolog PsMLO1 (Humphry et al. 2011; Pavan et al. 2011). To date, nine er1 alleles have been characterized, with each of them corresponding to a different PsMLO1 mutation (Pavan et al. 2013; Sun et al. 2015; Pavan et al. 2011; Humphry et al. 2011; Sun et al. 2016a; Sun et al. 2016b; Santo et al. 2013; Sudheesh et al. 2015). Allele names from er1-1 to er1-7 were previously assigned (Pavan et al. 2011 and 2013; Sun et al. 2016a; Sun et al. 2016b). In this study, the two alleles corresponding to the PsMLO1 mutations described by Santo et al. (2013) and Sudheesh et al. (2015) were designated as er1-10 and er1-11, respectively (the names of er1-8 and er1-9 are applied by Dr. Zhendong Zhu through personal communication with him).

Marker-assisted breeding (MAB), based on DNA markers linked to loci or genes controlling phenotypes of interest, is routinely used in many crops to select for desired characteristics in the early stages of plant growth and facilitate cultivar improvement. Among DNA marker types, SNP markers are considered the ideal choice for geneticists and breeders with advantages of co-dominance, low cost, high throughput, and automation (Gupta et al. 2001; Singh and Singh 2015). Several studies targeted the identification of markers suitable for the selection of er1 resistance, which is commonly used in pea breeding. Recently, functional markers, specifically designed on different er1 alleles, were developed (Pavan et al. 2013; Sun et al. 2015; Pavan et al. 2011; Humphry et al. 2011; Sun et al. 2016a; Sun et al. 2016b; Santo et al. 2013; Sudheesh et al. 2015). However, most of these markers are based on gel detection, which is costly and time-consuming. In addition to the limits mentioned above, gel-based markers are also associated with issues of gel-based scoring, such as blurry bands, incomplete digestion of restriction enzymes, and insufficient segregation (Holdsworth and Mazourek 2015). The kompetitive allele-specific PCR (KASPar) assay is a SNP genotyping system based on fluorescence and is flexible, accurate, and inexpensive (Semagn et al. 2014). It has been successfully applied in pepper (Holdsworth and Mazourek 2015), wheat (Neelam et al. 2013), weedy rice (Rosas et al. 2014), soybean (Rosso et al. 2011), and pea (Boutet et al. 2016). To facilitate the current genotyping system for er1 alleles, the objectives of this study were to develop eight KASPar markers, KASPar-er1-1, KASPar-er1-3, KASPar-er1-4, KASPar-er1-5, KASPar-er1-6, KASPar-er1-7, KASPar-er1-10, and KASPar-er1-11 and use them to genotype a pea collection to identify germplasm carrying the er1 alleles.

The pea germplasm accessions characterized in this study included the 246 accessions (Table S1) of the USDA PSP collection (Western Regional Plant Introduction Station, USDA-ARS, Pullman, WA, USA), nine lines carrying er1 resistance alleles, and one susceptible check. In more detail, the nine resistant lines, their er1 allele and source were as follows: JI 1559 (er1-1, John Innes Centre); JI 210 (er1-3, John Innes Centre); JI 1951 (er1-4, John Innes Centre); ROI3/02 (er1-5, University of Bari, Italy); G0001778 (er1-6, Chinese Academy of Agricultural Sciences, China); DDR-11 (er1-7, Chinese Academy of Agricultural Sciences, China); F(er1mut2) (er1-10, Universidade do Algarve, Portugal); and ps1771 and Yarrum (er1-11, Centre for AgriBioscience, Australia). The susceptible check was JI 502 (John Innes Centre).

Eight KASPar markers (KASPar-er1-1, KASPar-er1-3, KASPar-er1-4, KASPar-er1-5, KASPar-er1-6, KASPar-er1-7, KASPar-er1-10, and KASPar-er1-11) were designed by LGC KBioscience (KBioscience, Hoddesdon, UK) from the PsMLO1 sequence, based on the mutation information of the er1 alleles. Sequence information of the two allele-specific forward primers and the common reverse primer for each marker set are shown in Table 1. The KASPar markers were amplified in 10 μl reaction volumes with 5 μl of 20 ng/μl DNA, 5 μl 2× KASP master mix, and 0.14 μl KASP assay primer mix (KBioscience, Hoddesdon, UK). The PCR amplifications were performed on a GeneAmp 9700 thermal cycler (Applied Biosystems, Forster City, CA, USA) using the following cycling program: template DNAs were initially denatured at 94 °C for 15 min; followed by 10 cycles, when the annealing temperature dropped by 0.6 °C per cycle: 94 °C for 20 s, annealing for 60 s (Table 1); and followed by 26 cycles of 94 °C for 10 s and 55 °C for 60 s finally cooling to 4 °C. Fluorescent endpoint reading was performed on a Bio-Rad CFX96 Real-Time PCR System (Hercules, CA, USA) with 37 °C for 60 s, and the genotypes were called using the accompanying Bio-Rad CFX Manger 2.0 software (Bio-Rad Laboratories Inc., Hercules, CA, USA).

Table 1 Sequence information of the eight KASPar markers
Full size table

Co-segregation of the KASPar markers with the previously reported er1 functional markers, er1-1_S (er1mut1)_STS, er1-3/XbaI_dCAPS, er1-4/AgsI_CAPS, er1-5/HRM54_HRM, er1-6_SNP1121_HRM, er1-7_InDel111-120_STS, er1-10_F(er1mut2)_STS, and er1-11_SCAR, was assessed using modified protocols (Pavan et al. 2013; Sun et al. 2016a; Sun et al. 2016b; Sudheesh et al. 2015). The PCR amplifications for the er1-1, er1-3, and er1-4 functional markers were performed in 25 μl reaction volumes with 40 ng of template DNA, 1× PCR buffer, 1.5 mM MgCl2, 0.16 mM dNTPs, 0.5 μl of each forward and reverse primer (10 pmol/μl), and 1 U Taq DNA polymerase. The PCR amplifications were performed on a GeneAmp 9700 thermal cycler (Applied Biosystems, Forster City, CA, USA) using the following cycling program: template DNA was initially denatured at 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 58 °C for 30 s for STS_ S (er1mut1), 53 °C for 30 s for er1-3/XbaI_dCAPS and er1-4/AgsI_CAPS, and 72 °C for 1 min, with a final 10 min at 72 °C, finally cooling to 4 °C. The PCR products from CAPS were digested with AgsI (SibEnzyme Ltd., Novosibirsk, Russia), while the PCR products from dCAPS were digested with XbaI (NewEngland Biolabs, Beverly, MA, USA) according to the manufacturer’s instructions. The PCR products were discriminated on a 2% MetaPhor agarose gel (Lonza, Rockland, ME, USA) and visualized with ethidium bromide. For the er1-5 and er1-6 HRM markers, PCR amplifications were performed in a total volume of 10 μl including 40 ng of template DNA, 500 nM of each forward and reverse primer, and 1× Sso Fast EvaGreen Supermix (Bio-Rad). The PCR reactions for HRM curve acquisition were conducted on a Light-Cycler Roche 480 II Real-Time PCR instrument (Roche applied science, Indianapolis, Ind.) using the following cycling program: template DNA were initially denatured at 98 °C for 2 min, followed by 50 cycles of 98 °C for 10 s, 59 °C (er1-5), and 56 °C (er1-6) for 10 s; then 1 cycle of 95 °C for 10 s; followed by increasing the temperature from 65 °C to 95 °C with an increment of 0.19 °C per minute; and finally cooling down to 37 °C for 30 s. A web-based tool, uAnalyze (Dwight et al. 2012), was used for data evaluation with thermodynamic prediction from specific melting curves. For er1-7_InDel111-120_STS, er1-10_F(er1mut2)_STS, and er1-11_SCAR markers, PCR amplifications were performed with 4 ng genomic DNA; 1× PCR buffer; 1.5 mM MgCl2; 0.2 mM dNTPs; 0.05 μM forward primer; 0.25 μM reverse primer; 0.2 μM M13 primers with dyes of FAM, VIC, NED, PET, and 0.6 U BIOLASE™ DNA polymerase (Bioline); and 6.76 μl ddH2O in a total volume of 12 μl. The template DNAs were initially denatured at 95 °C for 5 min, followed by 42 cycles, each of which consisted of 95 °C for 1 min, 52 °C for 1 min, and 72 °C for 1 min. The final extension was at 72 °C for 10 min. These PCR products were analyzed on an ABI 3730 DNA analyzer (Applied Biosystems), and data were scored using GeneMarker software version 2.2.0 (SoftGenetics).

To evaluate the response of the pea collection to powdery mildew, the 246 pea accessions were planted in a naturally infected disease nursery at the Oregon State University Horticulture Farm, Corvallis, OR in 2015. The reaction to powdery mildew was rated at the flat pod stage through visual assessment according to the disease severity key with modifications of methods from the study of Falloon et al. (1995), where 1 = 0–20% of leaflets covered by lesions, 2 = 20–40%, 3 = 40–60%, 4 = 60–80%, and 5 = 80–100%. In order to determine the species of Erysiphe present, six infected plants were collected from the nursery. Isolates were developed from the powdery mildew on each plant, and the internal transcribed spacer (ITS) regions were sequenced. Total DNA was extracted from the powdery mildew conidia and mycelia using the DNeasy Mini Plant Kit according to the manufacturer’s protocol (Qiagen, Valencia, CA). PCR amplifications were performed as described in Attanayake et al. (2009) with the total genomic DNA using ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) primers (White et al. 1990) or EryF (5′-TACAGAGTGCGAGGCTCAGTCG-3′) and EryR (5′-GGTCAACCTGTGATCCATGTGACTGG-3′) (Attanayake et al. 2010). The PCR products were purified using QIAquick PCR Purification Kit (Qiagen, Valencia, CA), and purified products were sequenced using the BigDye Terminator Sequence Kit (Life Technologies, Grand Island, NY, USA) on an ABI PRISM 377 sequencer at the Genomics Core Facility, Washington State University. Pairwise comparison with the BLAST algorithm (BLASTN, E value < 1e−10) in NCBI was used to identify most similar sequences.

The KASPar assay separated homozygous resistant and homozygous susceptible accessions into different clusters, corresponding to the HEX-labeled cluster and FAM-labeled cluster (Fig. 1). They were easily visualized through Bio-Rad CFX Manger software. The eight KASPar markers developed in this study were successfully tested on 246 pea accessions. The results indicated that only one pea accession PI 142775 was homozygous for the er1-1 allele, and none carried any of the other er1-3, er1-4, er1-5, er1-6, er1-7, and er1-10 alleles. A total of 195 accessions were er1-11/er1-11 homozygotes (Table S2); however, 182 of which were susceptible based on the phenotypic evaluation (Table S2). Thus, this allele may be population specific and would need to be confirmed in the specific germplasm under investigation.

Fig. 1
figure 1

Genotypic data of eight KASPar markers on 18 pea accessions, 9 positive controls, and 1 negative control. X-axis indicates relative FAM fluorescence units, and Y-axis indicates relative HEX fluorescence units. Individuals clustered in the upper left (purple) are homozygous for HEX-labeled er1/er1 powdery mildew resistance. Individuals clustered in lower right (orange) are homozygous for FAM-labeled Er1/Er1 powdery mildew susceptibility. Individuals clustered in the lower left (black) are no template controls and samples failed to identify (color figure online)

Full size image

The KASPar markers were tested for co-segregation with the corresponding functional markers (Fig. 2) and the phenotypic characterization. All the functional markers, except for er1-6_SNP1121_HRM, er1-10_F(er1mut2)_STS, and er1-11_SCAR, showed complete co-segregation with the corresponding KASPar markers (Table S2). The er1-6_SNP1121_HRM and er1-10_F(er1mut2)_STS markers showed no polymorphism among the pea accessions used in this study. The er1-11_SCAR showed 183 er1-11/er1-11 homozygotes while KASPar-er1-11 indicated 195 er1-11/er1-11 homozygotes (Table S2). According to the phenotypic evaluation, 17 pea accessions were found to be highly resistant to powdery mildew with a score value of 1, namely Stirling, W6 17293, PI 102888, PI 116944, PI 142775, PI 179451, PI 183467, PI 207508, PI 220174, PI 220189, PI 222071, PI 222117, PI 273605, PI 274307, PI 307666, PI 486131, and PI 628276. Among them, PI 142775 was identified as an er1-1/er1-1 homozygote and 12 accessions were identified as er1-11/er1-11 homozygotes (Table S2). The identification of susceptible er1-11/er1-11 individuals indicated that both er1-11_SCAR and KASPar-er1-11 cannot be regarded as functional markers, which by definition are designed on DNA sequence polymorphisms responsible for phenotypic variation (Andersen and Lubberstedt 2003). Four pea accessions were shown to be resistant to powdery mildew in the disease nursery but were rated as susceptible from the genotypic results from the KASPar markers and the functional markers. A possible explanation is that these pea accessions might carry other er1, er2, or Er3 alleles, which were not tested in this study. To verify the species identity of the pathogen from the disease nursery, ITS sequencing was performed on six powdery mildew isolates. The sequences are listed in Table S3. From the BLAST search, all the isolates have 99 to 100% identity with Erysiphe pisi (accession number in NCBI: FJ378867.1). In conclusion, we predict that KASPar markers developed in this study might be powerful and valuable tools for use in pea breeding to help develop cultivars with resistance to powdery mildew.

Fig. 2
figure 2

Gel images of er1-1_S (er1mut1)_STS (upper left), er1-3/XbaI_dCAPS (upper middle), and er1-4/AgsI_CAPS (upper right). M1, Bioline HyperLadder™ 100 bp; M2, Bioline HyperLadder™ 50 bp; 1, PI 163126, homozygous susceptible; 2, PI 261671, homozygous susceptible; 3, PI 393488, homozygous susceptible; 4, PI 269762, homozygous susceptible; 5, PI 314795, homozygous susceptible; 6, PI 356984, homozygous susceptible; 7, JI 1559, homozygous er1-1 resistant; 8, JI 210, homozygous er1-3 resistant; 9, JI 1951, er1-4 resistant; 10, JI 502, homozygous susceptible; 11–12, no template control (NTC). The lower graphs show HRM curve profiles (er1-5/HRM54_HRM). The lower left graph indicates the normalized melting curves, while the lower right graph displays derivative melting curves. The HRM profiles show two distinct melting curves in the pea accessions, corresponding to susceptible homozygotes (curve 1) and er1-5-resistant homozygote (ROI3/02) (curve 2)

Full size image