Plant materials and mapping populations
Initial linkage mapping was performed with a recombinant inbred line (RIL) population (n = 47) from the cross between Gy14 and WI7221 (PI 183967) which we used early for cucumber linkage map development and QTL analysis of horticulturally important traits (Ren et al. 2009; Sheng et al. 2020). Gy14 is an inbred line that is susceptible to JRKN while WI7221 was an inbred line selected from the wild cucumber HARD accession PI 183967 with high resistance to JRKN (Walters and Wehner 1997, 1998).
To narrow down the region for the mj locus, backcross populations were developed to identify recombinants defined by flanking markers from the initial linkage analysis. Since HARD is late flowering, a resistant RIL, RIL271 was used to cross with Gy14. RIL271 had normal flowering time similar to Gy14, and carried most Gy14 alleles at 995 SSR marker loci. First backcross (BC1) (n = 190) plants were grown in 50-cell trays. DNA was extracted from each plant and screened with flanking markers to identify recombination events within the mj interval. Individual plants (n = 14) found to have recombination events within the mj interval were transplanted to 12” pots. Their genotypes were confirmed (see genotyping methods), and each was self-pollinated to produce BC1S1 seed. We selected true breeding S1 plants, homozygous for the mj introgression, which were again confirmed by genotyping and self-pollinated to produce the BC1S2 plants used for phenotyping for mj resistance. Each line was phenotyped with at least 2 replicates in independent screens.
Histological analysis of nematode development in roots
Histological examinations were performed to investigate the development of JRKN within the susceptible Gy14 and resistant RIL271 lines as previously described (Hajihassani et al. 2020). Briefly, three galled root segments were randomly collected from each infected genotype. The galled roots were fixed in a formaldehyde solution for 1–2 days at room temperature. The fixed roots were dehydrated in a series of ethanol solution (70 to 100%) using a tissue processor (Excelsior AS; Thermo Scientific, Thermo Shandon Limited, Runcorn, Cheshire, U.K.) and embedded in paraffin. Cross-sections (4 to 5 µm thick) were prepared using a rotary microtome (RM2125 RTS; Leica, Leica Biosystems, Nussloch, Germany) and transferred to glass slides. Cross-sections were stained in hematoxylin-eosin, examined under a microscope (Olympus BX43; Olympus, Tokyo, Japan), and photographed using a digital camera (Olympus DP73; Olympus).
Root penetration assays were conducted to compare the number of juvenile nematodes entering the root tips of susceptible and resistant control lines. Plants from each line were individually planted in D40L Deepot© containers (Stuewe &Sons, Tangent Oregon) filled with an autoclave sterilized sand:soil mixture and arranged in a randomized complete block design. Each plant was inoculated with 300 infective stage 2 juvenile (J2) JRKN. Root systems of 4 inoculated plants from each line were collected at 2, 4, and 6 days post-inoculation (DPI). At each time point, root systems were washed to remove soil, soaked in a 1.5% NaOCl solution for 3 min, and rinsed with tap water. The J2 nematodes within the roots were stained by boiling for 30 sec in red food coloring (Thies et al. 2002). Roots were destained by soaking in lactophenol for 48 hours, then pressed between two glass slides as previously described (Hajihassani et al. 2019b). The number of stained J2 within each root tip was counted under a stereomicroscope. J2 counts were recorded and tested for significant differences using the 2-way student t-test in Excel.
Screening for JRKN resistance
Phenotyping of 47 RILs and 14 recombinant backcross lines was performed in a greenhouse in eight independent screening tests. Each RIL was phenotyped at least 3 times across independent screens. In each experiment, seeds from each line were germinated under moist filter paper for 24 hours at 28°C before being planted directly into D40L Deepot™ containers (Stuewe &Sons, Tangent Oregon) filled with an autoclave sterilized 1:1 sand:potting soil mix. Plants were arranged into a randomized block design, watered and fertilized using a soluble 20-20-20 fertilizer as necessary throughout the duration of each screen. Each block included the susceptible control line (Gy14) and a resistant control line (HARD or RIL271).
Two weeks after planting, each plant was inoculated with 3,000 JRKN eggs. The JRKN isolate ‘VW4’, originally isolated by Professor Valerie Williamson at the University of California, was used as inoculum for all the screens in this study. The nematode culture was maintained in the greenhouse on susceptible ‘Rutgers’ tomato grown in autoclave sterilized media. On the day of inoculation, nematode eggs were extracted from galled tomato roots by agitating them in a 0.5% NaOCl solution for 2 minutes, and rinsing the mixture through stacked 8” sieves ( No. 270 over No. 500) as previously described (Hussey and Barker 1973). The eggs were counted under a dissecting microscope, diluted in tap water, and 4mL of a 750 egg/mL mixture was pipetted into the soil at the base of each plant.
Five weeks post-inoculation, cucumber roots were washed free of soil and stained for the presence of egg masses using red food coloring as previously described (Thies et al. 2002). The total number of visible egg masses was counted for each root system. The distinction between susceptible and resistant lines was determined by the average number of egg masses on susceptible and resistant controls. Cucumber lines averaging less than 15 nematode egg masses were classified as resistant and lines averaging more than 30 nematode egg masses per root system were classified as susceptible.
The JRKN resistance in the 47 RIL lines was treated as either a qualitative or quantitative trait. Linkage analysis was performed with JoinMap 3.0 and QTL analysis was performed with the mean egg mass per root system for each RIL using Composite Interval Mapping (CIM) approach as previously described (Wang et al. 2016). Genotypic data for these 47 RILs were from a linkage map developed previously that contains 995 SSR marker loci (Ren et al. 2009). The analysis was conducted using R/qtl (Broman et al. 2003). Significance of the mj resistance QTL was determined using a 1,000 permutation test, and an initial 1.5-LOD support interval was used to identify the flanking SSR markers (SSR00160 and SSR22110) on either side of the initial 1.23Mb interval. These flanking markers were used to screen BC1 populations and identify 14 independent recombination events that were used to further narrow down the mj interval.
Marker development and analysis
To refine the map locations of the mj locus, SNPs (single nucleotide polymorphisms) were identified by aligning genomic DNA sequences in the target region between the two parental lines Gy14 (v2.0 chr1: 4,031,065 − 5,261,376) and PI 183967 (Chr1:4,043,287-5,270,956), which were available on the Cucurbit Genetics Database (CuGenDB, http://cucurbitgenomics.org/). Primer sequence information for all SNP markers developed in this study is provided in Supplemental Table S1. DNA sequences harboring SNPs were sent to C3r Bioscience (Essex, UK) for PCR Allele Competitive Extension (PACE™) assay design, or LGC Biosearch Technologies (Middlesex, UK) for Kompetitive Allele Specific PCR (KASP™) assay design (Table S1). The primers for each PACE assay were synthesized by Integrated DNA Technologies (Research Triangle Park, NC) and were reconstituted and mixed according to the manufacturer’s instructions.
Total DNA used for genotyping was extracted from leaf samples using the Plant DNAeasy Mini kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocols. PACE and KASP genotyping assays were run in 5 µl reactions in 96-well plate format on a Light Cycler 480™ (Roche Diagnostics Corporation, Indianapolis, IN). Each reaction contained 1.5 µl of DNA template [10ng/uL] along with 2.5 µl 2x KASP Master Mix (LGC Biosearch Technologies, Middlesex, UK), 0.07 µl reconstituted PACE or KASP assay, and 0.93 µl ddH2O. The thermal cycler protocol consisted of (in the sequence presented): 94°C-15 min, 10 cycles of 94°C-20 sec and 61°C[-0.6°C/cycle]-1min, 26 cycles of 94°C-20 sec and 55°C-1min, and cooled to 37°C for fluorescent signal quantification in the HEX (533-580nM) and FAM (465-510nM) wavelengths. Raw signals for each well were exported for analysis using custom scripts in R. All genotyping plates contained at least two no template control reactions as well as 2 reactions containing DNA from each parental line (Gy14 and RIL271) to facilitate visual end-point genotype clustering.
Each PCR reaction for SSR markers contained 50ng of template DNA, 0.5 µM of each primer, 1 U Taq DNA polymerase (Thermo Fisher Scientific), 2.5 µl 10× PCR buffer, 1.5 mM MgCl2, and 0.2 mM dNTPs in a total volume of 25 µl. A one-fit-all, touch-down PCR program was used for all SSR primer sets and consisted of (in the sequence presented): 3-min at 95°C; six cycles of 45 sec at 94°C, 5 min at 68°C, 1 min at 72°C, with the annealing temperature being reduced by 2°C per cycle; eight cycles of 45 sec at 94°C, 2 min at 58°C, 1 min at 72°C, with the annealing temperature reduced by 1°C per cycle; a final 25 cycles of 45 sec at 94°C, 2 min at 50°C and 1 min at 72°C. PCR products were separated and imaged in a 4% MetaPhor™ (Lonza, Rockland ME) agarose gel.