Background & Summary

The grapevine belonging to the Vitis genus of the Vitaceae family, which is rich in nutrients and is one of the most important horticultural crops in the world1,2. They are classified into three groups based on their geographical distribution and ecological characteristics: North American, Eurasian, and East Asian3,4. Vitis vinifera is the only species in the Eurasian grape. Vitis vinifera and its hybrid cultivars are widely planted around the world and are absolutely dominant among the cultivars3.

Most grapevine cultivars are susceptible to many diseases, especially downy mildew (DM). Plasmopara viticola (Pv), the pathogen of DM, can infect leaves, fragile shoots, tendrils, young fruits, inflorescences, and other tender tissues of grapevine. Serious infection can cause wilting or death of the vine, resulting in a considerable decrease in production5. Fungicides are commonly utilized for the management of P. viticola, however, their extensive use can result in the development of pathogen resistance. Overreliance on fungicides not only increases costs but also contributes to environmental contamination and poses risks to human health. The exploration of Pv resistant germplasm is very important for a deeper understanding of the Pv resistance mechanism and the breeding of Pv resistant cultivars.

Pv originated in North America and was introduced to Europe in the 1870s alongside phylloxera resistant rootstocks6. It has spread worldwide to other viticultural regions through the exchange of grapevine material. DM is currently prevalent in China’s primary grape-producing regions, particularly in the east and south of China7, with grape varieties like Cabernet Sauvignon, Chardonnay, and Merlot being highly susceptible.

China is the source of most East Asian wild grapevines and has an abundance of native grape resources. Many native Chinese grape varieties, particularly V. pseudoreticulata, V. amurensis, and V. davidii are highly resistant to Pv8,9,10,11. The mechanism of how Chinese native grapevine evolved Pv resistance germplasm is very interesting, and relevant research is limited.

Currently, there is limited genomic information available for native Chinese grapevine species. We collected representative wild grapes from various regions of China and used Genotyping by Targeted Sequencing (GBTS) technology to obtain genomic information from them (Fig. 1). The reliability of marker development based on GBTS has been demonstrated in our previous study, but the purpose and data of this study were different from those of our previous study12. The purpose of this study was to provide critical genetic information for the investigation of the mechanism of resistance to Pv of native Chinese grapevine.

Fig. 1
figure 1

The workflow of the data output in this study. The light green box indicates the selection of the Pv resistant parents and the sensitive parent for the hybrid population. The dark green boxes indicate the three hybrid populations and the number of individuals. The dark green boxes indicate the three hybrid populations and the number of individuals. Blue boxes indicate the steps of data output for marker design, sample DNA detection, and phenotypic identification.

Full size image

Methods

Plant material and resistance evaluation

A total of nine accessions from East Asian grapes with different genetic backgrounds (including V. pseudoreticulata, V. davidii, V. amurensis, V. hancockii, V. piasezkii, and V. yeshanensis) and ‘Cabernet Sauvignon’ (V. vinifera, CS) were selected for identification of Pv resistance. Nine leaf discs with 1-cM diameter were taken from the fourth and fifth fully expanded leaves. The leaf discs were inoculated with Pv as described by Yu et al.9, and then incubated in a controlled environment with a 12-hour photoperiod, 21°C temperature, and 80% humidity. The sporulation density (SPD) approach outlined by Fu et al.10 was used to assess Pv resistance (Fig. 2). The average score of nine leaf discs was continuously collected from the fifth to the seventh day after inoculation. The average score from the fifth to the seventh day after inoculation indicates the resistance level of each grapevine.

Fig. 2
figure 2

Phenotypic evaluation standards for SPD. Sporulation density (SPD) is a semi-quantitative scoring system of P. viticola sporulation density10.

Full size image

Accessions with high resistance to Pv were selected to create population by -cross-fertilization and selfing. All the vines were planted at the Center for Viticulture and Enology of Shanghai Jiao Tong University (Minhang District, Shanghai City, China).

DNA extraction

Genomic DNA was isolated from young grapevine leaves using CTAB techniques, as reported by Qu et al.13 DNA quality was evaluated using 1% agarose gel electrophoresis and a λ-DNA ladder, while DNA concentration was measured using NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA).

Target location selection for GBTS and hybridization probe design

The target locations were chosen using the Illumina 20 K Chip and previous GBS sequencing data10,14,15. A total of 20,597 locations were selected according to the following criteria: minor allele frequency (MAF) >0.1, missing data points <5%, and evenly distributed loci throughout the genome. GenoBaits Designer software (MolBreeding Biotechnology Co., Ltd., Shijiazhuang city, Hebei province, China) was used to design a 110 bp probe for each target location. This method has been applied and validated in another of our studies12, but the two works have different research objectives and data.

Library construction, probe hybridization and sequencing

GBTS library construction and probe hybridization were performed as described by Yang et al.16 In short, the library construction process consists of four steps: (1) ultrasound was used to fragment DNA; (2) fragmented DNA was end-repaired and added with an A-tail; (3) adapters with barcode sequences were ligated to the A-tailed segments; and (4) the library was amplified by PCR. After the library construction was completed, probe hybridization was performed using library mixture, library hybridization, target capture, library amplification, purification, and library control. All processes were automated using various instruments to save labor and time.

The Qubit 2.0 fluorometer (Thermo Fisher Scientific, CA, USA) was used to evaluate the quality of the enriched libraries. The samples that passed quality control were loaded into the flow cell and sequenced using PE150 on the MGISEQ-2000 platform (MGI, Shenzhen, China). The sequencing depth was 100×, and the average total amount of sequencing data per sample was about 1 Gbp. mSNPs are multiple SNPs that result from a single probe, including the target SNP and adjacent regions. To make the best use of sequencing data, mSNPs that could exist in each fragment captured by a single probe were identified. The mSNP was developed as reported by Yang et al.16

Data Records

These data files could be divided into 4 categories. They were native grapevine accessions (including 9 East Asian grapevines and 1 Eurasian grape V. vinifera ‘Cabernet Sauvignon’), hybrid population of V. viniera ‘Cabernet Sauvignon’ and V. amurensis ‘Shuanghong’ (CS × SH), hybrid population of V. viniera ‘Cabernet Sauvignon’ and V. pseudoreticulata ‘Huadong1058’ (CS × HD), and self-inbred population of V. davidii ‘Huiliang’ (HL selfing). The names and category details of these data files are provided in Supplementary Table 1. The SNP information details of all samples are provided in Supplementary Table 2. All Supplementary Table files can be accessed through Figshare at https://doi.org/10.6084/m9.figshare.28504721.v117.

All the GBTS raw data of 160 grapevines have been deposited in the NCBI Sequence Read Archive (SRA) (https://www.ncbi.nlm.nih.gov/) with the accession number SRP53944918 under BioProject accession PRJNA1174865.

Technical Validation

Population construction and Pv resistance evaluation

The Pv resistance of all the selected native grapevine accessions was significantly higher than the resistance of CS, and there was also extremely resistant accession like ‘Huadong1058’ (V. pseudoreticulata, HD) that was completely unable to inoculate Pv (Fig. 3a). HD, ‘Huiliang’ (V. davidii, HL), and ‘Shuanghong’ (V. amurensis, SH) was selected to create population. HD and SH were used as male parents to cross with the susceptible variety CS. HL was self-pollinated to create a population.

Fig. 3
figure 3

Plasmopara viticola resistance statistics. (a) Disease resistance indices of 9 varieties/accessions of East Asian grape and Cabernet Sauvignon (V. vinifera). (b) Distribution of disease resistance level in the progeny of the three hybrid populations. The boxes in the boxplot are 1/4 and 3/4, respectively, and the black horizontal lines are medians. The density distribution is shown on the right side of the boxplot. The Y-axis indicates the Pv resistance level.

Full size image

Fifty offspring of each population were selected. The resistance of the progeny exhibited continuous variation and followed a quantitative trait distribution. The resistance level distribution range of the three populations was different: ‘CS × SH’ population ranging from 3.31 to 8.90 with the mean value of 6.41, ‘CS × HD’ population ranged from 6.09 to 9.00 with an average of 7.55, and ‘HL × HL’ population ranged from 2.11 to 9.00 with an average of 5.91 (Fig. 3b).

Sequence alignments and SNP calling

To quickly identify the SNP information of grapevine, CS, HD, HL, SH, and the three populations were tested by the GBTS method to obtain all types of sequencing data, fastp software (version 0.20.0, https://github.com/OpenGene/fastp) was used to perform the following operations: removal of adapter sequences, elimination of paired reads if more than 10% of the read length is comprised of N bases, and exclusion of paired reads with over 40% of low-quality bases (Q value ≤ 20) in the read length.

The paired-end reads of targeted sequencing were aligned to the Vitis vinifera 12X.2 (vitviv2) genome using Burrows-Wheeler Aligner (BWA, version 0.5.9-tpx). A total of four individuals and three populations sequencing samples were analyzed. For all samples, the average Q20 and Q30 were 97.15% and 90.24%, respectively, the GC content was 40.08%, and the average alignment rate was 99.72%. Notably, HL and its selfing population have the highest Q20 and Q30.

Samtools (v0.1.18), and Genome Analysis Toolkit (GATK, v4.2.1.0) were used for BAM file handling, base recalibration, calling variants and variants filtration, respectively. To call variants, the HaplotypeCaller module of the GATK software was employed, with the following detection parameters:

-dcov 1000000 -minIndelFrac 0.15 -glm BOTH -l INFO

Subsequently, the VariantFiltration module was utilized for filtering, applying the following filtering parameters:

–filterExpression “MQ0 >  = 4 & & ((MQ0 / (1.0 * DP)) > 0.1) “–filterName “HARD_TO_VALIDATE”–filterExpression “DP < 5 || QD < 2”–filterName “LOW_READ_SUPPORT”

To extract the genotype information for SNP positions from the above alignment and mutation detection results, the following criteria were applied:(1) If the coverage depth of a sample at a specific SNP position is less than 5X, indicating insufficient sequencing depth, that position is treated as missing and denoted as ‘NA’ in the table. (2) If a sample exhibits a mutation at a specific SNP position, the determination of genotype depends on the mutation frequency. If the mutation frequency is greater than or equal to 0.8 or less than or equal to 0.2, the position is classified as homozygous. If the mutation frequency falls within the range of 0.2–0.8, the position is considered a heterozygous mutation, provided that each allele has a minimum of 4 supporting reads. Otherwise, the position is treated as missing and represented as ‘NA’ in the table. (3) The final summary compiles the genotype results for all SNP positions across the samples (Fig. 4).

Fig. 4
figure 4

Characteristics of the sequencing data. From the outer circle to the inner circle, the percentages of homozygote (Hom), heterozygote (Het), insertion (INS), deletion (DEL), NA, and variant (1Mbp as the window size) are shown.

Full size image