A Past Genetic Bottleneck from Argentine Beans and a Selective Sweep Led to the Race Chile of the Common Bean (Phaseolus vulgaris L.)

Abstract

The domestication process of the common bean gave rise to six different races which come from the two ancestral genetic pools, the Mesoamerican (Durango, Jalisco, and Mesoamerica races) and the Andean (New Granada, Peru, and Chile races). In this study, a collection of 281 common bean landraces from Chile was analyzed using a 12K-SNP microarray. Additionally, 401 accessions representing the rest of the five common bean races were analyzed. A total of 2543 SNPs allowed us to differentiate a genetic group of 165 accessions that corresponds to the race Chile, 90 of which were classified as pure accessions, such as the bean types 'Tórtola', 'Sapito', 'Coscorrón', and 'Frutilla'. Our genetic analysis indicates that the race Chile has a close relationship with accessions from Argentina, suggesting that nomadic ancestral peoples introduced the bean seed to Chile. Previous archaeological and genetic studies support this hypothesis. Additionally, the low genetic diversity (π = 0.053; uHe = 0.53) and the negative value of Tajima' D (D = -1.371) indicate that the race Chile suffered a bottleneck and a selective sweep after its introduction, supporting the hypothesis that a small group of Argentine bean genotypes led to the race Chile. A total of 235 genes were identified within haplotype blocks detected exclusively in the race Chile, most of them involved in signal transduction, supporting the hypothesis that intracellular signaling pathways play a fundamental role in the adaptation of organisms to changes in the environment. To date, our findings are the most complete investigation associated with the origin of the race Chile of common bean.

Keywords: Andean pool; common bean; genetic diversity; genetic structure; race Chile.

Figure 1

Population structure analysis of the 667 common bean accessions. (A) Based on the Bayesian clustering method. The inferred ancestry probability (Q_i) for each accession is on the Y axis and the accessions is on the X axis. (B) Principal component analysis based on gene pool. The first two principal components (PCs) of analysis accounting for 50.3 and 6.6% of total variation, respectively. Green and red accessions correspond to the Andean and Mesoamerican gene pool, respectively. (C) Principal component analysis of 667 accessions according to their country of origin. CHILE; PERU; BRA: Brazil; CO: Colombia; CR: Costa Rica; HO: Honduras; ES: El Salvador; GU: Guatemala; NI: Nicaragua; MX: Mexico; OTHER: advanced breeding lines from different countries.

Figure 2

Genome-wide loess curves for nucleotide diversity (π) and Tajima’s D values across all chromosomes in the P. vulgaris genome for Mesoamerican (blue), Andean (red), and Chile (black) genetic groups.

Figure 3

Neighbor-joining tree based on genetic distances between some pure accessions belonging to the race Chile of common bean and others from the Andean genetic pool. Red, green, and black clades indicate accessions belonging to the Mesoamerican, Andean, and the race Chile genetic groups, respectively.

Figure 4

Linkage disequilibrium (LD) decay across the 11 common bean chromosomes (Pv01–Pv11). LD values were corrected by population structure and kinship (r²_sv). The blue, red, and black lines indicate the LD decay for the Mesoamerican (M1 + M2), Andean (A1 + A2) and Chile (A3) genetic group, respectively.

Figure 5

Gene ontology (GO) functional classification. (A) GO classification annotated in GO categories biological process, cellular components, and molecular function. (B) Histograms of distribution in subcategories of the enriched gene ontology (GO). Only GO subcategories constituted by at least five sequences from the data set in the MF, BP, and CC categories are represented.