Biotic Stress

Read counts per gene were obtained from the RNA‐seq alignments using HTSeq (v0.9.1) (Anders et al., 2015) in unstranded union mode. Differentially expressed genes were identified from the gene read counts using DESeq2 (Love et al., 2014) with an alpha level of 0.01. For organ‐specific gene expression, an organ was defined similarly to Sekhon et al. (2011) with some modifications (Sekhon et al., 2011). The endosperm and embryo were grouped into a single organ (seed) and the cob, silk, tassel, and anthers were grouped into one reproductive organ. The SAM was also considered a separate organ establishing six organs for analyses: internode, leaf, reproductive, root, seed, and SAM. Each organ from the developmental gene atlas was individually contrasted against the other five organs retaining genes with an adjusted P < 0.01 and a log₂ fold change >2. A gene was considered organ‐specific if it met these criteria in the organ of interest when compared against all other organs. Tissue‐specific gene expression was similarly characterized for seed and reproductive tissues. Using read counts for genes previously identified as seed‐ or reproductive‐specific, the seed and reproductive tissues were contrasted against the other tissues in its respective organ. A gene was considered tissue‐specific if it had an adjusted P < 0.01 and a log₂ fold change >2 in all tissue comparisons. For stress‐related differential expression, each treatment was contrasted against its respective experimental control, retaining genes with an adjusted P < 0.01 and a log₂ fold change <−2 or a log₂ fold change >2. A gene was considered DE under both abiotic and biotic stress if it met these criteria in at least one treatment from both stress types. The organ‐specific heatmap was generated with the R package ‘gplots’ (Alexa and Rahnenfuhrer, 2016) ‘heatmap.2’ function and all other plots were generated with the R packages ‘ggplot2’ (Wickham, 2009) and ‘RColorBrewer’ (Neuwirth, 2014).

Substantial data comprising thousands of genes and transcripts often challenges efficient data analysis. Classical GO enrichment analyses based on a database search, though helpful for functional categorization of given gene sets, is less useful if detailed analysis is required at the level of studying genes involved in specific pathways and/or physiological functions. For this purpose, different tools have been developed and used with variable success, e.g., the TM4 suit by Saeed et al. (2003 (link)), and GoMiner^TM by Zeeberg et al. (2003 (link)).
The MapMan-omics data analyses software (Thimm et al., 2004 (link); Usadel et al., 2005 (link); Urbanczyk-Wochniak et al., 2006 (link); http://mapman.gabipd.org) allows visualization of -omics data at the process or pathway level. The software is designed and optimized to map transcriptomic data on currently available databases for many plant species, including A. lyrata, A. thaliana (Affymetrix, Agilent, TAIR 6, TAIR 7, TAIR 8, TAIR 9, and TAIR 10), Brassica napus, B. rapa, Carica papaya, Citrus, Eucalyptus grandis, Glycine max, Gossypium raimondii, Oryza sativa, Populus trichocarpa, Zea mays, and many other plant species. MapMan utilizes a hierarchical “BIN”-based ontology system. Specific bins are allocated to biological functions and sub-bins are allocated to individual steps or nodes in that particular biological function in a hierarchical order. For example, BIN number 20 is for stress, BIN number 20.1 is for biotic stress, and 20.2 is for abiotic stress. Similarly, sub-bins for abiotic stress include 20.2.1 (heat stress), 20.2.2 (cold stress), 20.2.3 (drought stress), 20.2.4 (wounding), and 20.2.5 (light). The bin and sub-bin approach minimizes the redundancy usually found in GO enrichment analyses. In addition, the software also utilizes gene expression values and displays the analyzed data as a diagram which enhances comprehension and is of a quality appropriate for presentation. Genes with increased or decreased expression levels are shown as color-coded squares in blocks. This tool has been widely used and constantly evolves to accommodate more plant species and data sets.
In order to get more meaningful information, we analyzed 6436 DEGs through MapMan version 3.6.0RC1 to visualize various genes involved in various pathways and biological functions and their expression patterns. For this purpose, all the data for the genes with significantly differential expression (P ≤ 0.05) were arranged in Microsoft Excel with their standard unique locus identifiers and their final expression value [Log2 (FPKM treated/FPKM untreated)] and saved in a tab-delimited format. These files were then mapped against the Arabidopsis “Ath_AGI_LOCUS_TAIR10_Aug2012.m02” database in MapMan. After analyzing the data, selected pathways were combined by adding their respective bins and sub-bins to custom-made images uploaded to MapMan.

According to our field observation and previous phenotype- and SSR-based studies [12 (link),16 (link)], a total of 44 accessions (Table 1) representing the major cultivars of D. alata in China were selected for whole genome sequencing. These accessions, with high phenotypic diversity, covered the main distribution area of this species in China, from west (Lijiang, Yunnan Province, geographic coordinates (gc): 26°43′ N, 100°15′ E) to east (Taizhou, Zhejiang Province, gc: 28°38′ N, 121°27′ E), and from south (Sanya, Hainan Province, gc: 18°24′ N, 109°45′ E) to north (Xuzhou, Jiangsu Province, gc: 34°39′ N, 116°35′ E). In addition, we also specifically selected accessions that have greatly contributed to the breeding and research of this species and accessions highly resistant to abiotic and biotic stress. Genomic DNA from each accession was extracted from fresh or silica gel-dried leaves using a DNAsecure Plant Kit (Tiangen Biotech, Beijing, China), according to the manufacturer’s protocol. The DNA concentration and integrity were measured on an Agilent 2100 BioAnalyzer (Agilent Technologies, Palo Alto, CA, USA), together with agarose gel electrophoresis. Paired-end libraries with an insert size of 350 bp were constructed and then sequenced on the BGISEQ-500 platform to generate raw sequences with a 150 bp read length. The library construction and sequencing were conducted at Wuhan Benagen Tech Solutions Company Limited, Wuhan, China.
Additionally, whole-genome sequencing datasets encompassing eight African accessions of D. alata (Table 1) were downloaded from the National Centre of Biotechnology Information (NCBI) Sequence Read Archive (SRA) database and were converted to the FASTQ format using the fastq-dump utility from the SRA Toolkit v.2.9.6 [43 (link)].

Two strains of P. ampelicida isolated by FEM in Trentino (Italy), and one isolated by the Julius Kühn Institute (JKI)–Institute for Grapevine Breeding Geilweilerhof (Siebeldingen, Germany), were previously genetically characterized and propagated on oatmeal agar (0.5% w/v) [22 ]. These strains were combined and used for artificial infection, following the protocol developed by Bettinelli et al. [22 ]. Briefly, fresh leaf tissues with mature BR lesions were used as the inoculum source. In the greenhouse, the cuttings of the mapping population were cultivated in a climatic chamber at 24 °C, young growing shoots with at least five fully expanded leaves were sprayed with conidia suspension adjusted to 10⁴ conidia/mL in the late afternoon and kept at 100% relative humidity overnight. In the field, clusters at susceptible phenological stage [26 ] BBCH 77 (berries beginning to touch) [89 ] and growing shoots were sprayed after the sunset to avoid direct UV irradiation. Plastic bags were used to wrap inoculated organs, ensuring a high humidity treatment overnight. Bags were removed the following early morning before sunrise. Overall, two experiments were conducted on clusters in the two growing seasons 2020 (FC1) and 2021 (FC2), while in 2022, disease progression did not occur due to unusually high temperatures during the night (>30 °C). In each field inoculation trial, the entire population was screened simultaneously, and two clusters were evaluated per genotype, generating four sub-experiments (FC1a, FC1b, FC2a, FC2b). Shoots were evaluated once in the field in 2021, distinguishing between leaves (FL) and shoot internodes (FS). A total of eight greenhouse inoculation trials were performed between 2020 and 2021 on subgroups of the population to obtain three experiments (GL1, GL2, GL3). To produce an overall dataset of the resistance trait, the replicated data of GL and FC trials were also combined by (i) retaining only the minimum resistance value per each genotype and experiment type (GL min and FC min), and (ii) calculating the median (GL median, FC median). While the minimum should exclude possible false positive resistance evaluations, the median is the most robust statistical parameter for non-normally distributed data. Resistance evaluation of leaves in the greenhouse was conducted following the 5-step scale proposed by Rex et al. [21 (link)], which resembles OIV descriptors [90 ], i.e., (1) very low, (3) low, (5) medium, (7) high, and (9) very high resistance. Contrariwise, leaves and shoot internodes in the field were evaluated with a binomial scale as susceptible (S that is 1) or resistant (R that is 9), respectively, in the presence or absence of lesions, because the occurrence of other lesions caused by multiple abiotic and biotic stresses did not permit using a more detailed scale. To screen cluster resistance in the field, a 5-step rating scheme was developed based on the percentage of infected berries (Table 3), following the example of OIV₄₅₉ descriptor for degree of cluster resistance to Botrytis bunch rot [90 ]. Berry color (white or black, coded as a 0/1 valued binary variable) was also recorded to be used as a positive control in QTL analyses, thanks to the well-known location of the responsible QTL on chromosome 2 [33 (link)]. The normality of trait distribution was determined by means of the Shapiro–Wilk normality test (p < 0.05). All statistical analyses were executed with the software PAST 3.26 [91 ].