Diversity and Origin of Andean Landraces of Common Bean

Abstract

Landraces of common bean (Phaseolus vulgaris L.) pertaining to the Andean gene pool are remarkably diverse in plant and grain morphology and agroecological adaptation. The objectives of this study were to determine the genetic structure of a large sample of Andean landraces, and to establish a correspondence between Andean landraces and wild bean populations that might have served as the source of domesticated bean. A total of 182 landraces representing the three recognized races of Andean bean and including many popping bean types were analyzed using amplified fragment length polymorphism (AFLP) technology with multiple correspondence analysis (MCA) and unweighted pair group method with arithmetic mean (UPGMA). Twenty‐nine wild bean accessions representing the diversity of wild bean in South America and Middle America were also included. Two sets of primers were used to generate 189 polymorphic AFLP. The graph of the results of MCA indicated that most landrace accessions formed a single undifferentiated group, and analysis by UPGMA combined with bootstrapping confirmed this. A small number of outliers presented bands that suggested introgression from Mesoamerican beans. Among wild bean populations from South America, those from Bolivia graphed in closest proximity to the cultivated bean, suggesting that Bolivia might have been an important primary domestication site. The narrow genetic base of Andean beans emphasizes the need to broaden the genetic base of the Andean gene pool.

Microsatellite characterization of Andean races of common bean (Phaseolus vulgaris L.)

“…The introgressed genotypes with similarity to the Mesoamerican controls were easily distinguished in both germplasm sets at both K=3 or K=4. (Duarte et al 1999;Beebe et al 2001;Islam et al 2004) even though some evidence for Andean race structure was evident in RFLPs analysis ) and in our previous study with microsatellites of a limited number of Andean genotypes ). The analysis shown here is more definitive due to the large number of genotypes analyzed and the widespread distribution across the genome of loci sampled.…”

“…The majority of the genotypes in this group were landraces although a few were improved varieties (4). Most but not all of the genotypes were selected from the core collection developed at CIAT (Tohme et al 1995a) and were evaluated previously by Beebe et al (2001). The second group consisted of 60 accessions from Colombia (50 native landraces, 2 commercial varieties and 8 advanced lines from CIAT) all of which are found in the CIAT core collection and have Andean phaseolin patterns along with the 4 control genotypes described below.…”

“…Andean beans can be distinguished from Mesoamerican beans on the basis of morphological and biochemical differences such as seed size and seed proteins as well as with various types of molecular markers (Gepts 1998). Meanwhile the races within each gene pool can be distinguished somewhat by some but not all molecular markers (Beebe et al 2000(Beebe et al , 2001Diaz and Blair 2006) and to a certain extent by allelic frequency at a limited number of isozyme loci (Koenig and Gepts 1989;Singh et al 1991c). Germplasm exchange, gene flow and crossing programs between and within the two gene pools have given rise to introgression between gene pools (Beebe et al 2001;Islam et al 2004) and between races (Diaz and Blair 2006) resulting in intermediates phenotypes that do not classify well into any single race or gene pool.…”

“…Meanwhile the races within each gene pool can be distinguished somewhat by some but not all molecular markers (Beebe et al 2000(Beebe et al , 2001Diaz and Blair 2006) and to a certain extent by allelic frequency at a limited number of isozyme loci (Koenig and Gepts 1989;Singh et al 1991c). Germplasm exchange, gene flow and crossing programs between and within the two gene pools have given rise to introgression between gene pools (Beebe et al 2001;Islam et al 2004) and between races (Diaz and Blair 2006) resulting in intermediates phenotypes that do not classify well into any single race or gene pool. Furthermore, it is not known whether races originated from separate multiple domestications or were the result of a single domestication in each region with subsequent differentiation and adaptation to specific agroecological niches (Chacón et al 2005).…”

“…This was the case for Johns et al (1997) who tried to distinguish the Chile race using RAPDs, Duarte et al (1999) who attempted to differentiate various races with RAPDs and F o r P e e r R e v i e w Beebe et al (2001) who used AFLPs to characterize wild and cultivated Andean beans but could only distinguish wild accessions not cultivars. The conclusions of these three studies predict that there is less diversity among Andean cultivars than in Mesoamerican cultivars and that compared to wild accessions the cultivated Andean gene pool underwent a reduction in diversity, however none of the studies could define race structure at the molecular level or define how this structure would agree with the morphological classification made by Singh et al (1991a, b).…”