Integrated Bulk Segregant Analysis, Fine Mapping, and Transcriptome Revealed QTLs and Candidate Genes Associated with Drought Adaptation in Wild Watermelon

Abstract

Drought stress has detrimental effects on crop productivity worldwide. A strong root system is crucial for maintaining water and nutrients uptake under drought stress. Wild watermelons possess resilient roots with excellent drought adaptability. However, the genetic factors controlling this trait remain uninvestigated. In this study, we conducted a bulk segregant analysis (BSA) on an F₂ population consisting of two watermelon genotypes, wild and domesticated, which differ in their lateral root development under drought conditions. We identified two quantitative trait loci (qNLR_Dr. Chr01 and qNLR_Dr. Chr02) associated with the lateral root response to drought. Furthermore, we determined that a small region (0.93 Mb in qNLR_Dr. Chr01) is closely linked to drought adaptation through quantitative trait loci (QTL) validation and fine mapping. Transcriptome analysis of the parent roots under drought stress revealed unique effects on numerous genes in the sensitive genotype but not in the tolerant genotype. By integrating BSA, fine mapping, and the transcriptome, we identified six genes, namely L-Ascorbate Oxidase (AO), Cellulose Synthase-Interactive Protein 1 (CSI1), Late Embryogenesis Abundant Protein (LEA), Zinc-Finger Homeodomain Protein 2 (ZHD2), Pericycle Factor Type-A 5 (PFA5), and bZIP transcription factor 53-like (bZIP53-like), that might be involved in the drought adaptation. Our findings provide valuable QTLs and genes for marker-assisted selection in improving water-use efficiency and drought tolerance in watermelon. They also lay the groundwork for the genetic manipulation of drought-adapting genes in watermelon and other Cucurbitacea species.

Keywords: drought tolerance; genetic resources; genomics-assisted breeding; quantitative trait loci (QTL) mapping; transcriptome; wild watermelon.

Figure 1

Drought tolerance index (DTI) based on the number of lateral roots (NLR) in 38 watermelon accessions exposed to 15% Polyethylene Glycol (PEG, drought) and distilled water (CK, control) for 4 days in pouches. The DTI was calculated as the ratio between NLR with drought and CK treatments. The tested genotypes were categorized into five groups based on their DTI. The plotted values are averages of three independent biological replications (nine plants/replication/treatment).

Figure 2

Root phenotyping and validation of drought tolerance in the parents, F₁ and F₂ individuals. (a) Root phenotypes of ZJU076 (tolerant), ZJU196 (sensitive), and F₁ after 4 days of growth with 15% PEG6000 (drought, Dr) and distilled water (control, CK). (b) Drought tolerance index (DTI) of the parents and F₁. Values are means ± S.D. (n = 18). (c) After 15 days of water withholding (the final soil water content ≈ is 2%), (d) Frequency distribution of the number of lateral roots (as drought tolerance indicator) among 484 F₂ individuals exposed to 15% Polyethylene Glycol (PEG) for four days.

Figure 3

Bulk segregant analysis (BSA) results of the drought tolerance in watermelon. (a) The single nucleotide polymorphism (SNP) index of the high pool, (b) the SNP index of the low pool, and (c) the delta SNP index values used for the association analysis. The x and y axes show the 11 chromosomes of watermelon and the SNP index, respectively. The curved line indicates the fitted SNP index or delta SNP index. The horizontal line indicates the association threshold with FFN of a 95% confidence interval. (d) Major QTLs for drought tolerance in watermelon detected by the G prime (G’) method. The QTL1, and QTL2 represent qNLR_Dr. Chr01 and qNLR_Dr. Chr02, respectively.

Figure 4

Validation of the detected QTLs through haplotype analysis using six Kompetitive allele specific PCR (KASP) markers for each QTL in 305 F₂ individuals. Green and red indicate homozygous segments from sensitive and tolerant parents, respectively. Yellow indicates heterozygous segments. The sidebars (on the left) represent the root traits of F₂ individuals: primary root length (PRL), total root system (TRS), lateral root system (LRS), and the number of lateral roots (NLR). The single nucleotide polymorphism (SNP) data are arranged in ascending order from top to bottom based on the NLR values. Zooming in will make the sample IDs and root trait values much more visible.

Figure 5

Fine mapping of the detected QTLs with 12 Kompetitive allele specific PCR (KASP) markers. (a) Parent genotypes and two F₂ recombinants in the target regions. The left part represents qNLR_Dr. Chr01, while the right part shows qNLR_Dr. Chr02. (b) The recombinant offspring individuals of F₂ are sorted based on qNLR_Dr. Chr02. (c) The recombinant offspring individuals of F₂ are sorted based on qNLR_Dr. Chr01. The F₂ offspring were classified into two groups based on the segment origins. Red indicates the homozygous ZJU076 (tolerant) segment, green indicates the homozygous ZJU196 (sensitive) segment, yellow shows the heterozygous region, and gray represents mixed (involved individuals similar to the parents or heterozygous). The average number of lateral roots (NLR) of each family was calculated from ten individuals.

Figure 6

The significant differentially expressed genes (DEGs) within the delimited region of qNLR_Dr. Chr01. Log₂ (fold change) of the 13 DEGs (p < 0.05) in CK_76 vs. DR_76 and CK_ 196 vs. D.R._ 196 within the delimited region of qNLR_Dr. Chr01.