In:
Proceedings of the National Academy of Sciences, Proceedings of the National Academy of Sciences, Vol. 109, No. 14 ( 2012-04-03)
Abstract:
The magnitude of the population subdivisions in Mesoamerica that we present in our scenario had not been recognized clearly before our study. Our work indicates the potential of exploring genetic diversity that is not incorporated into the current, domesticated germplasm (the total genetic diversity present in a species). Specifically, in our case, the wild Mesoamerican germplasm should be used in breeding programs, because it has the potential for creation of new cultivars (plant varieties produced by selective breeding). Moreover, it is crucial to consider the Mesoamerican wild germplasm to sample the largest amount of diversity for use in commercial bean varieties, especially because most improved varieties of the common bean are of Andean origin at present. Furthermore, exploration of newly recognized genetic diversity is vital for meeting future challenges posed by climate change. In summary, our study presents clear evidence of a Mesoamerican origin of P. vulgaris , most likely Mexico, and different migrations into South America ( 2 ). Thus, we suggest that P. vulgaris from northern Peru–Ecuador is a relict population that represents only a fraction of the genetic diversity of the ancestral population. Analysis of haplotype networks supported these results. Importantly, the PhI accessions showed haplotypes that were closer to the Mesoamerican accessions and often separated from the majority of the Andean accessions. These outcomes clearly are not compatible with the hypothesis of a South American origin, in which the PhI genotypes would be expected to be an intermediate. In studying the population structure, we found six genetic clusters (B1–B6) ( Fig. P1 A ). Although all the Andean and northern Peru–Ecuador genotypes were assigned clearly to single specific clusters (B6 and B5, respectively), the Mesoamerican genotypes were subdivided into four different clusters (B1, B2, B3, and B4), and they showed higher levels of admixture. The Mesoamerican accessions were found to be distributed in all the branches of the NJ tree ( Fig. P1 B ): The Andean accessions clustered with the Mesoamerican group B3, and the northern Peru–Ecuador accessions (PhI) were more related to the other Mesoamerican groups and particularly to the B4 accessions. The Mesoamerican populations showed the highest genetic diversity estimates for each of the five genes. A very strong reduction in genetic diversity was found for the Andean gene pool compared with that of Mesoamerica. Our data strongly support the Andean bottleneck suggested by Rossi et al. ( 2 ). A clear relationship between the rate of mutations (genetic alterations) and the time of diversity recovery from the occurrence of a bottleneck was seen: The higher the mutation rate, the faster is the recovery of diversity. Indeed, considering our results and those obtained with different molecular markers, we found that the bottleneck was recovered almost completely by markers with a high mutation rate (simple sequence repeats) compared with markers that show lower rates of mutation (amplified fragment-length polymorphisms and sequence data). Diversity analysis was carried out by estimating several measures of nucleotide variability for the five gene fragments. Furthermore, we used statistical methods to measure the loss of nucleotide diversity in the Andean wild versus Mesoamerican wild bean populations. A Bayesian model-based approach was used to infer the hidden genetic population structure of our sample and thus to assign the genotypes (the genetic constitution of an organism) to genetically structured groups. Finally, to define the underlying relationships between the different genetic groups thus identified, we constructed haplotype and neighbor-joining (NJ) trees. Haplotypes are DNA sequences of a chromosome or of a single locus (in our case, of the gene fragments analyzed); NJ is a widely used method for constructing trees that illustrate the evolutionary development and diversification of a species (in our case based on multilocus data). In general, two major ecogeographical gene pools for P. vulgaris have been recognized: Mesoamerica and the Andes. However, the bean's wild form includes an additional gene pool located between Peru and Ecuador ( 1 ). Kami et al. ( 1 ) suggested that the seed protein, type I (Inca) phaseolin, which is characteristic of this wild form, is ancestral and thus identifies the Andes in northern Peru and Ecuador as the bean's origin. Recently, however, the alternative hypothesis of a Mesoamerican origin was suggested by Rossi et al. ( 2 ). They drew upon genetic analyses that suggested that, before domestication, the wild bean arrived in the Andean region, where it remained isolated. This process resulted in a bottleneck that reduced the genetic diversity of the Andean bean. To address this ongoing debate, we investigated the diversity of nucleotides (the subunits of the DNA molecule) at five gene sites (or loci) in a large sample that represents the entire geographical distribution of the bean's wild forms. The individuals studied are representative of the different gene pools of this species. Studies of the evolution of crop species are aimed at highlighting the structure and organization of their genetic diversity as well as the evolutionary forces that shaped it. Such knowledge is crucial for efficient conservation and development of new, improved plant varieties. Our study investigated the origins of the common bean ( Phaseolus vulgaris ) by examining the diversity at five gene sites (or loci). We present clear evidence of a Mesoamerican origin and reveal the very complex geographical structure of its genetic diversity in Mesoamerica.
Type of Medium:
Online Resource
ISSN:
0027-8424
,
1091-6490
DOI:
10.1073/pnas.1108973109
Language:
English
Publisher:
Proceedings of the National Academy of Sciences
Publication Date:
2012
detail.hit.zdb_id:
209104-5
detail.hit.zdb_id:
1461794-8
SSG:
11
SSG:
12
Bookmarklink
