Pathogen lifestyle determines host genetic signature of quantitative disease resistance loci in oilseed rape (Brassica napus)

Abstract

Using associative transcriptomics, our study identifies genes conferring resistance to four diverse fungal pathogens in crops, emphasizing key genetic determinants of multi-pathogen resistance. Crops are affected by several pathogens, but these are rarely studied in parallel to identify common and unique genetic factors controlling diseases. Broad-spectrum quantitative disease resistance (QDR) is desirable for crop breeding as it confers resistance to several pathogen species. Here, we use associative transcriptomics (AT) to identify candidate gene loci associated with Brassica napus constitutive QDR to four contrasting fungal pathogens: Alternaria brassicicola, Botrytis cinerea, Pyrenopeziza brassicae, and Verticillium longisporum. We did not identify any shared loci associated with broad-spectrum QDR to fungal pathogens with contrasting lifestyles. Instead, we observed QDR dependent on the lifestyle of the pathogen-hemibiotrophic and necrotrophic pathogens had distinct QDR responses and associated loci, including some loci associated with early immunity. Furthermore, we identify a genomic deletion associated with resistance to V. longisporum and potentially broad-spectrum QDR. This is the first time AT has been used for several pathosystems simultaneously to identify host genetic loci involved in broad-spectrum QDR. We highlight constitutive expressed candidate loci for broad-spectrum QDR with no antagonistic effects on susceptibility to the other pathogens studies as candidates for crop breeding. In conclusion, this study represents an advancement in our understanding of broad-spectrum QDR in B. napus and is a significant resource for the scientific community.

Fig. 1

Brassica napus genotypes have common GEMs associated with resistance pathogens of the same lifestyle. a Brassica napus disease resistance to fungal pathogens is quantitative. Resistance phenotype (arbitrary units (a.u.)) of different B. napus genotypes to Alternaria brassicicola (A.b), Botrytis cinerea (B.c), Pyrenopeziza brassicae (P.b), or Verticillium longisporum (V.l). Arbitrary units signify resistance values obtained through the reciprocal of disease susceptibility scores, derived from estimated marginal means and normalized to a 0–1 range for comparative data visualization. The total number of genotypes used for each pathogen assay and the position of reference genotypes Quinta (Q), Tapidor (T), Westar (W), and Zhongshuang 11 (Z) are indicated. b Correlation between the resistance phenotype of B. napus lines to the fungal pathogens. Positive correlations (green), negative correlations (red), and no correlation (n.s.); strengths of correlation (sizes of circles) are indicated between pairwise comparisons of resistance responses. c Manhattan plot of B. napus genome showing marker-trait association of statistically significant GEMs for resistance to each fungal pathogen. The x-axis indicates GEM location along the chromosome; the y-axis indicates the − log₁₀(p) (P value). d The numbers of resistance and susceptibility gene expression markers (GEMs) shared between pairwise comparisons of pathogens. e The numbers of GEMs associated with resistance to one pathogen and susceptibility to another pathogen. f Venn diagrams showing the overlap between A.b and B.c susceptibility GEMs and P.b and V.l resistance GEMs (right) and the overlap between A.b and B.c resistance GEMs and P.b and V.l susceptibility GEMs (left). g Linear regression analysis of gene expression (RPKM) of Cab032851.1 relative to resistance to fungal pathogens (arbitrary units (a.u.), normalized values between zero and one)

Fig. 2

Shared GEMs for QDR and chitin-induced ROS are associated with resistance to hemibiotrophic pathogens but susceptibility to necrotrophic pathogens. a Chitin-induced ROS response (arbitrary units (a.u.)) of the B. napus panel. Arbitrary units signify resistance values obtained through the reciprocal of disease susceptibility scores, derived from estimated marginal means and normalized to a 0–1 range for comparative data visualization. The positions of reference genotypes Quinta (Q), Tapidor (T), Westar (W), and Zhongshuang 11 (Z) are indicated. b Linear regression analysis of chitin-induced ROS response relative to quantitative disease resistance to Alternaria brassicicola (A.b), Botrytis cinerea (B.c), Pyrenopeziza brassicae (P.b), or Verticillium longisporum (V.l) (arbitrary units (a.u.), normalized values between zero and one). c Venn diagrams showing the overlap between GEMs associated with chitin-induced ROS and QDR to the fungal pathogens, Alternaria brassicicola (A.b), Botrytis cinerea (B.c), Pyrenopeziza brassicae (P.b), and Verticillium longisporum (V.l). The percentage of QDR-related GEMs which are also related to chitin-induced ROS is indicated. d The number of gene expression markers (GEMs) whose expression was positively or negatively associated with chitin-induced ROS and the number which was also associated with resistance (R) or susceptibility (S) to the B. napus fungal pathogens. Venn diagram indicating the overlap between GEMs associated with several pathogen interactions. e Linear regression analysis of gene expression (RPKM) of BnaC04g2850D and Cab022449.1 relative to chitin-induced ROS responses and resistance to fungal pathogens

Fig. 3

Resistance to Verticillium longisporum is associated with a genomic deletion on chromosome A09. a Manhattan plots showing marker-trait association resulting from GEM and GWA analysis of resistance to V. longisporum in 193 Brassica napus genotypes. The x-axis indicates GEM or SNP location along the chromosome; the y-axis indicates the − log₁₀(p) (P value). Gray line indicates the FDR > 0.05 cut-off value, red line highlights the shared region. b Segregation of resistance to V. longisporum (V.l), Pyrenopeziza brassicae (P.b), Alternaria brassicicola (A.b), or Botrytis cinerea (B.c) (arbitrary units (a.u.)) with the highest associating marker, Cab013526.1.651.A, showing the marker effect between A and G (R = A/G). P values were determined by a Student’s t test. Arbitrary units signify resistance values obtained through the reciprocal of disease susceptibility scores, derived from estimated marginal means and normalized to a 0–1 range for comparative data visualization. c Linkage decay plot from marker Cab013526.1.651.A as a function of genetic distance (MB). The gray line indicates an R² value of 0.2, red lines indicate the area of linkage disequilibrium. d Resistance to V. longisporum (left-hand y-axis) and gene expression (RPKM) of the five highest associating GEMs Cab013522.1, Cab013524.1, Cab013523.1, Cab013526.1, and Cab013517.1 (right-hand y-axis) in reference genotypes Quinta (Q), Tapidor (T), Westar (W), and Zhongshuang 11 (Z). e Heatmap indicating query coverage (compared to the B. napus pan-transcriptome) in Q, T, W, and Z for the 107 genes predicted to be in linkage disequilibrium with Cab013526.1.651.A on chromosome A09