Genetic characterization of cucumber genetic resources in the NARO Genebank indicates their multiple dispersal trajectories to the East

Abstract

Genotyping-by-sequencing of 723 worldwide cucumber genetic resources revealed that cucumbers were dispersed eastward via at least three distinct routes, one to Southeast Asia and two from different directions to East Asia. The cucumber (Cucumis sativus) is an economically important vegetable crop cultivated and consumed worldwide. Despite its popularity, the manner in which cucumbers were dispersed from their origin in South Asia to the rest of the world, particularly to the east, remains a mystery due to the lack of written records. In this study, we performed genotyping-by-sequencing (GBS) on 723 worldwide cucumber accessions, mainly deposited in the Japanese National Agriculture and Food Research Organization (NARO) Genebank, to characterize their genetic diversity, relationships, and population structure. Analyses based on over 60,000 genome-wide single-nucleotide polymorphisms identified by GBS revealed clear genetic differentiation between Southeast and East Asian populations, suggesting that they reached their respective region independently, not progressively. A deeper investigation of the East Asian population identified two subpopulations with different fruit characteristics, supporting the traditional classification of East Asian cucumbers into two types thought to have been introduced by independent routes. Finally, we developed a core collection of 100 accessions representing at least 93.2% of the genetic diversity present in the entire collection. The genetic relationships and population structure, their associations with geographic distribution and phenotypic traits, and the core collection presented in this study are valuable resources for elucidating the dispersal history and promoting the efficient use and management of genetic resources for research and breeding in cucumber.

Fig. 1

Phylogenetic relationships, population structure, and geographic distribution of the cucumber genetic resources. a Maximum likelihood phylogenetic tree of the 723 cucumber and two C. hystrix accessions inferred using the 61,367 filtered SNPs. Colors at the tips indicate the geographic origin of each accession. (A tree with individual tip labels and node supports is given in Fig. S1). b Model-based clustering with K from 2 to 4. The x-axes list the accessions in the same order as they appear on the phylogenetic tree, and the y-axes quantify the membership to each ancestral population, represented in different colors. c Geographic distribution of the different genetic populations. For each country, the pie represents the proportion of the different populations, and its size reflects the relative sample size. The map was created using the rworldmap package (South 2011) in R v4.1.0

Fig. 2

Genetic diversity and differentiation among the four genetic populations. Colored circles represent each population. The circle size reflects the nucleotide diversity (π) within each population, and the distance between circles represents the population fixation index (F_ST) between populations. Actual values of π and F_ST are given in parentheses and between pairs of populations, respectively

Fig. 3

Phenotypic variation in each genetic population. a Fruit length (cm). b Fruit diameter (cm). c Seed length (cm). d Sex expression. e Number of female flowers per node. f Fruit bearing position. g Parthenocarpy. h Resistance to melon yellow spot virus (MYSV). i Resistance to papaya ringspot virus (PRSV). The gray color in panels d–i represents accessions that showed intra-accession segregation

Fig. 4

Differences in fruit traits and corresponding local genetic differentiation between two subclades within the East Asian population. a Number of accessions with different mature fruit skin color, netting, and spine color in each subclade. The gray color in each pie represents accessions that showed intra-accession segregation. b Fixation index (F_ST) between the two subclades surrounding the CsMYB60/HEUKCHEEM gene responsible for fruit spine color. The position of the gene is highlighted in yellow. The horizontal dashed line indicates the genome-wide threshold for the top 1% highly divergent region

Fig. 5

Development of the World Cucumber Core Collection (WCC). a Allelic coverage of the 61,367 filtered SNPs versus collection size selected by GenoCore. b The same phylogeny as in Fig. 1a, showing only topology. Accessions selected for the WCC are highlighted in red. c Fruit morphological diversity of the WCC (partial)