Identification of the egusi seed trait locus (eg) and its suppressor gene associated with the thin seed coat trait in watermelon.

Egusi watermelon has a unique egusi seed type, which could be useful for breeding both edible seeds and edible flesh in watermelon. However, the genetic basis of the unique egusi seed type is not clear. In the present study, we first reported that at least two genes with inhibitory epistasis were responsible for the thin seed coat (unique egusi seed type) in watermelon. Inheritance analysis of five populations, including F2, BC, and BCF2, suggested that the thin seed coat trait was controlled by a suppressor gene together with the egusi seed locus (eg) in egusi watermelon. Based on high-throughput sequencing technology, two quantitative trait loci located on chromosome 1 and chromosome 6 were identified for the thin seed coat trait in watermelon. One of the loci, the eg locus on chromosome 6, was finely mapped to a genomic region of 15.7 kb, which contained only one candidate gene. Comparative transcriptome analysis highlighted differentially expressed genes involved in cellulose and lignin synthesis between watermelon genotypes varying in the thickness of the seed coat and provided several potential candidate genes for the thin seed coat trait. Taken together, our data suggest that at least two genes are complementarily involved in the thin seed coat trait and will be useful for cloning novel genes. The results presented here provide a new reference for uncovering egusi seed genetic mechanisms and valuable information for marker-assisted selection in seed coat breeding.

Watermelon domestication was shaped by stepwise selection and regulation of the metabolome.

Although crop domestication has greatly aided human civilization, the sequential domestication and regulation of most quality traits remain poorly understood. Here, we report the stepwise selection and regulation of major fruit quality traits that occurred during watermelon evolution. The levels of fruit cucurbitacins and flavonoids were negatively selected during speciation, whereas sugar and carotenoid contents were positively selected during domestication. Interestingly, fruit malic acid and citric acid showed the opposite selection trends during the improvement. We identified a novel gene cluster (CGC1, cucurbitacin gene cluster on chromosome 1) containing both regulatory and structural genes involved in cucurbitacin biosynthesis, which revealed a cascade of transcriptional regulation operating mechanisms. In the CGC1, an allele caused a single nucleotide change in ClERF1 binding sites (GCC-box) in the promoter of ClBh1, which resulted in reduced expression of ClBh1 and inhibition of cucurbitacin synthesis in cultivated watermelon. Functional analysis revealed that a rare insertion of 244 amino acids, which arose in C. amarus and became fixed in sweet watermelon, in ClOSC (oxidosqualene cyclase) was critical for the negative selection of cucurbitacins during watermelon evolution. This research provides an important resource for metabolomics-assisted breeding in watermelon and for exploring metabolic pathway regulation mechanisms.

Discovery of the Genomic Region and Candidate Genes of the Scarlet Red Flesh Color (Yscr ) Locus in Watermelon (Citrullus Lanatus L.).

The flesh color of watermelon (Citrullus lanatus) is an important fruit quality trait that helps to determine fruit attractiveness and is potentially beneficial to human health. Previous inheritance analyses determined that a single dominant gene, Yscr , produces the scarlet red flesh color rather than the coral red flesh color in watermelon. However, no genomic region or gene-based molecular markers for the locus Yscr have been reported thus far. In the present study, two high-density genetic maps and whole-genome variation detection aided by genome resequencing were first map the flesh color locus Yscr to a small region on chromosome 6 based on two independent populations derived from two scarlet red-fleshed lines and two coral red-fleshed lines. Two major quantitative trait loci located in the same genomic regions were identified in the F2 and BC1P2 populations and explained 90.36% and 75.1% of the phenotypic variation in flesh color, respectively. Based on the genetic variation in the two parental lines, newly developed PCR-based markers narrowed the Yscr region to 40 Kb. Of the five putative genes in this region, four encoded glycine-rich cell wall structural proteins, which implied that a new regulatory mechanism might occur between scarlet red- and coral red-fleshed in watermelon. Moreover, the genotypes of two newly developed InDel markers (InDel27_fc6 and InDel28_fc6) were completely consistent with the phenotypes in the F2 and BC1P2 populations and all 56 scarlet red-fleshed watermelon accessions. The results presented here provide valuable information for marker-assisted selection of flesh color breeding and the functional validation of candidate genes in watermelon.

A comprehensive genome variation map of melon identifies multiple domestication events and loci influencing agronomic traits.

Melon is an economically important fruit crop that has been cultivated for thousands of years; however, the genetic basis and history of its domestication still remain largely unknown. Here we report a comprehensive map of the genomic variation in melon derived from the resequencing of 1,175 accessions, which represent the global diversity of the species. Our results suggest that three independent domestication events occurred in melon, two in India and one in Africa. We detected two independent sets of domestication sweeps, resulting in diverse characteristics of the two subspecies melo and agrestis during melon breeding. Genome-wide association studies for 16 agronomic traits identified 208 loci significantly associated with fruit mass, quality and morphological characters. This study sheds light on the domestication history of melon and provides a valuable resource for genomics-assisted breeding of this important crop.

Fine mapping and discovery of candidate genes for seed size in watermelon by genome survey sequencing.

Fine mapping and discovery of candidate genes underlying seed size are important for modern watermelon breeding. Here, by using a high-resolution genetic map and whole-genome genetic variation detection aided by genome survey sequencing, we fine mapped and discovered candidate genes for seed size in watermelon. QTL (quantitative trait locus) mapping identified two pleiotropic QTLs for seed size, namely, qSS4 and qSS6, using a high-density genetic map constructed by specific length amplified fragment sequencing. qSS6 explained 93.00%, 94.11% and 95.26% of the phenotypic variation in thousand-seed weight, seed length and seed width, respectively, and was defined as a major QTL. Then, high-coverage re-sequencing of two parental lines detected a total of 193,395 SNPs (single nucleotide polymorphisms) and 45,065 indels (insertions/deletions), which corresponded to a frequency of 534 SNPs/Mb and 124 indels/Mb. Based on the genetic variation in the two parental lines, newly developed PCR-based markers allowed the region of qSS6 to be narrowed to 55.5 kb. Three potential candidates were identified, including a known seed size regulator in rice, SRS3. Taken together, our results reveal successful rapid fine mapping and discovery of candidate genes for seed size in watermelon, which could be applied to many traits of interest in plants.