Genome-wide association studies in a diverse strawberry collection unveil loci controlling agronomic and fruit quality traits

Abstract

Strawberries (Fragaria sp.) are cherished for their organoleptic properties and nutritional value. However, breeding new cultivars involves the simultaneous selection of many agronomic and fruit quality traits, including fruit firmness and extended postharvest life. The strawberry germplasm collection here studied exhibited extensive phenotypic variation in 26 agronomic and fruit quality traits across three consecutive seasons. Phenotypic correlations and principal component analysis revealed relationships among traits and accessions, emphasizing the impact of plant breeding on fruit weight and firmness to the detriment of sugar or vitamin C content. Genetic diversity analysis on 124 accessions using 44,408 markers denoted a population structure divided into six subpopulations still retaining considerable diversity. Genome-wide association studies for the 26 traits unveiled 121 significant marker-trait associations distributed across 95 quantitative trait loci (QTLs). Multiple associations were detected for fruit firmness, a key breeding target, including a prominent locus on chromosome 6A. The candidate gene FaPG1, controlling fruit softening and postharvest shelf life, was identified within this QTL region. Differential expression of FaPG1 confirmed its role as the primary contributor to natural variation in fruit firmness. A kompetitive allele-specific PCR assay based on the single nucleotide polymorphism (SNP) AX-184242253, associated with the 6A QTL, predicts a substantial increase in fruit firmness, validating its utility for marker-assisted selection. In essence, this comprehensive study provides insights into the phenotypic and genetic landscape of the strawberry collection and lays a robust foundation for propelling the development of superior strawberry cultivars through precision breeding.

FIGURE 1

Genetic diversity and population structure of 124 strawberry accessions. (A) Principal component analysis (PCA) based on 44,002 SNPs describing the relationships among accessions using TASSEL (Bradbury et al., 2007). Hybrids with Fragaria Chiloensis (clustering on the right side of panel [C]) have been enclosed on a green circle. (B) Determination of the number of subpopulations using the Delta K method (Evanno et al., 2005). (C) Dendrogram of similarity by neighbor‐joining (NJ). Accessions have been color‐coded according to their membership to the six STRUCTURE subpopulations. NJ‐tree implemented in TASSEL (Bradbury et al., 2007). For each accession, the color bar represents the proportion of ancestry using K = 6 ancestral groups inferred with the STRUCTURE program (Pritchard et al., 2000).

FIGURE 2

Agronomic and fruit quality traits in the strawberry collection evaluated in the 2020–2021 season. (A) Correlation matrix among the 25 strawberry traits. Each circle indicates Pearson correlation coefficient value (r) in a scale from red (negative correlation) to blue (positive correlation). *p‐value ≤ 0.05; **p‐value ≤ 0.01; *** p‐value ≤ 0.001. Principal component analysis (PCA) of strawberry accessions showing trait vectors (B) and accession distribution along the first two principal component axes (C). Strawberry accessions are color‐coded in C according to their population structure.

FIGURE 3

Significant genome‐wide association study (GWAS) associations for presence of an achene‐free band on the fruit (A), anthocyanins in runners (B), leaf glossiness (C), and leaf Powdery Mildew (D). Manhattan plots of −log₁₀(p) versus chromosomal position for the four models, phenotypic distribution of traits, and QQ plots of the model(s) with the highest number of associations. Different chromosomes are shown in different colors, which follow the order of ‘Camarosa’ v.1.0 (Edger et al., 2019): chromosome 1 (1‐1) to chromosome 28 (7‐4). Green horizontal line represents the significance threshold following the Bonferroni correction method (−log₁₀ [0.01/total SNPs]).

FIGURE 4

Significant genome‐wide association study (GWAS) associations for fruit acidity (A) and external color (B). Manhattan plots of −log₁₀(p) versus chromosomal position for the four models, phenotypic distribution of traits, and QQ plots of the model with significant associations. Different chromosomes are shown in different colors, which follow the order of ‘Camarosa’ v.1.0 (Edger et al., 2019): chromosome 1 (1‐1) to chromosome 28 (7‐4). Solid and dashed green horizontal lines represent the significance thresholds following the Bonferroni correction method (−log₁₀ (0.01/total SNPs) and (−log₁₀(0.05/total SNPs), respectively).

FIGURE 5

Significant genome‐wide association study (GWAS) associations for runnering time (A) and runner number (B) in 2020–2021 and 2019–2020. Manhattan plots of −log₁₀(p) versus chromosomal position for the four models, phenotypic distribution of traits, and QQ plots of the model(s) with the highest number of associations. Different chromosomes are shown in different colors, which follow the order of ‘Camarosa’ v.1.0 (Edger et al., 2019): chromosome 1 (1‐1) to chromosome 28 (7‐4). Solid and dashed green horizontal lines represent the significance thresholds following the Bonferroni correction method (−log₁₀ (0.01/total SNPs) and (−log₁₀(0.05/total SNPs), respectively). Significant SNPs detected in two seasons or in the two related traits are highlighted.

FIGURE 6

Significant genome‐wide association study (GWAS) associations for fruit firmness in 2020–2021. Manhattan plots of −log₁₀(p) versus chromosomal position for the four models (A), phenotypic distribution of fruit firmness (B), and QQ plots of the model(s) with the highest number of associations (C). The plots of the different chromosomes are shown in different colors, which follow the order of ‘Camarosa’ v.1.0 (Edger et al., 2019): chromosome 1 (1‐1) to chromosome 28 (7‐4). Solid and dashed green horizontal lines represent the significance thresholds following the Bonferroni correction method (−log₁₀ (0.01/total SNPs) and (−log₁₀(0.05/total SNPs), respectively).

FIGURE 7

Fruit firmness and FaPG1 expression in contrasting strawberry accessions. (A) Strawberry accessions in the high‐ and low‐firmness pools. (B) Difference in fruit firmness between pools. (C) Relative expression of FaPG1 by qRT‐PCR in the two pools. Statistical significance by t‐test is shown (***p‐value < 0.005; ****p‐value < 0.001). ACC, accession.

FIGURE 8

Development of a Kompetitive Allele Specific PCR (KASP) assay for fruit firmness prediction. (A) Allele effects of KASP‐184242253 marker on fruit firmness in a collection of 138 accessions evaluated in season 2022–2023. (B) Example of genotype clusters for KASP‐184242253 marker. The reference C allele is labeled with HEX and the alternative T allele is labeled with FAM. Heterozygous samples are represented in green and non‐template controls in black. Boxes span the 25th and 75th percentiles and the middle line represents the median. Whiskers (T‐bars) are the minimum and maximum values. Letters show significant differences by one way analysis of variance (ANOVA) and Tukey test (*p‐value < 0.0001). n, number of accessions. RFU, relative fluorescence units for HEX and FAM fluorescent dyes.