GWAS Reveals a Novel Candidate Gene CmoAP2/ERF in Pumpkin (Cucurbita moschata) Involved in Resistance to Powdery Mildew

Abstract

Pumpkin (Cucurbita moschata Duchesne ex Poir.) is a multipurpose cash crop rich in antioxidants, minerals, and vitamins; the seeds are also a good source of quality oils. However, pumpkin is susceptible to the fungus Podosphaera xanthii, an obligate biotrophic pathogen, which usually causes powdery mildew (PM) on both sides of the leaves and reduces photosynthesis. The fruits of infected plants are often smaller than usual and unpalatable. This study identified a novel gene that involves PM resistance in pumpkins through a genome-wide association study (GWAS). The allelic variation identified in the CmoCh3G009850 gene encoding for AP2-like ethylene-responsive transcription factor (CmoAP2/ERF) was proven to be involved in PM resistance. Validation of the GWAS data revealed six single nucleotide polymorphism (SNP) variations in the CmoAP2/ERF coding sequence between the resistant (IT 274039 [PMR]) and the susceptible (IT 278592 [PMS]). A polymorphic marker (dCAPS) was developed based on the allelic diversity to differentiate these two haplotypes. Genetic analysis in the segregating population derived from PMS and PMR parents provided evidence for an incomplete dominant gene-mediated PM resistance. Further, the qRT-PCR assay validated the elevated expression of CmoAP2/ERF during PM infection in the PMR compared with PMS. These results highlighted the pivotal role of CmoAP2/ERF in conferring resistance to PM and identifies it as a valuable molecular entity for breeding resistant pumpkin cultivars.

Keywords: CmoAP2/ERF; Cucurbita moschata Duchesne ex Poir.; Podosphaera xanthii; genome-wide association study; powdery mildew; powdery mildew resistance; pumpkin; single nucleotide polymorphism.

Figure 1

Visual scoring of PM resistance. At an experimental farm, 407 pumpkin genotypes were grown to first true leaf stage (~14 d after germination). Seedlings were infected twice at 5 d intervals, with PM suspensions collected from the heavily infected plants. After PM infection, leaf samples were gently removed, and the resistance index was scored on a scale of 1–9. (A) 1–4: susceptible; 5 and 6: moderately resistant; 7–9: resistant. (B) The PMS and PMR were compared for their resistance index. Values are means ± SE (n = 4), and different letters on the histograms indicate that the values differ significantly (p < 0.05).

Figure 2

Multivariate GWAS models BLINK, FarmCPU, and MMLM predicted PM resistance in pumpkins. Over a two-year study (2018 and 2019), different multi-locus based GWAS models, i.e., BLINK, FarmCPU, and MMLM, were used for the prediction of potential SNPs associated with PM resistance in 407 pumpkin genotypes. The black arrow denotes most significant association. In the Manhattan plot, the horizontal solid line indicates the Bonferroni-corrected significance threshold value at –log10(p) of 4.61. Different colors in the Manhattan plots represents different chromosomes in pumpkin. For different models, the Manhattan and quantile–quantile plots are presented on the left and right panels, respectively.

Figure 3

Candidate gene identification and allelic diversity assessment. (A) The gene structure of the CmoAP2/ERF gene and a map of the location of the missense mutations in the exon region. (B) A comparison of the PMS and PMR line Sanger-sequencing chromatograms and the resultant amino acid changes. The inverted black triangle in the upper panel indicates the PMS line or reference genome, whereas the inverted red triangle in the lower panel indicates the nucleotide changes in the PMR line. The asterisk indicates the SNP variation, but no amino acid change.

Figure 4

Profiles of designed SNPs, differentiating PMS and PMR lines. (A) Amplification profile of pumpkin lines showing polymorphism of the CmoAP2/ERF dCAPS marker. The PMS, PMS/PMR, and PMR lines amplified using the CmoAP2/ERF dCAPS marker (right) and band patterns after restriction digestion with Pst1 (left). (B) Coding sequence alignment of PMS/PMR lines, along with three PM sensitive and three PM resistant lines for SNP verification. Sequence alignments of selected lines and the red box indicate SNP variation at SNP C3. 7419839. (C) Coding sequence alignment of PMS/PMR lines, along with three PM sensitive and three PM resistant lines, for SNP verification. Sequence alignments of selected lines and the red box indicate SNP variation at SNP C3. 7419891.

Figure 5

Quantification and comparison of diseased leaf area by visual assessment and Image J. (A) The performance of the visual assessments and ImageJ for the same leaves of various disease severities of powdery mildew. The red color implies disease severity whereas the white color means tolerance (B) Assortment of ImageJ quantified PM infected PMS/PMR-F₂ population based on the genotyping results. The box plots depict the diseased leaf area (DAL) percentage, and the dotted line indicates the threshold of resistance.

Figure 6

Expression of the CmoAP2/ERF gene in PMS and PMR lines at different time intervals post-PM infection. Relative expression of CmoAP2/ERF gene in PMS and PMR lines in response to PM pathogen infection. The CmoActin gene was used as an internal control for normalization. For each experiment, the expression level of the PMS sample prior to pathogen treatment was used as a calibrator for quantification and was assumed as 1. Values are means ± SE (n = 3), and different letters on the bars indicate that the values differ significantly (p < 0.05).