Main content area

Genetic relationships among selected crested wheatgrass cultivars and species determined on the basis of AFLP markers

Mellish, A., Coulman, B., Ferdinandez, Y.
Crop science 2002 v.42 no.5 pp. 1662-1668
Agropyron cristatum, Agropyron desertorum, Agropyron fragile, Agropyron, cultivars, genetic distance, genetic markers, species diversity, genetic techniques and protocols, outcrossing, genotype, Agropyron mongolicum, amplified fragment length polymorphism, species differences, population
The crested wheatgrass complex (Agropyron spp. Gaertn) consists of persistent, drought tolerant, tussock grasses, which originate from the steppes of central Asia and are of importance as range grasses on the Great Plains of North America. Amplified fragment length polymorphism (AFLP) markers were used to examine the interpopulation relationships and to compare variances within and among 12 populations from four Agropyron spp. [Agropyron cristatum (L.) Gaertn, Agropyron desertorum (Fisch ex Link) Schlut, Agropyron fragile (Roth) Candargy and Agropyron mongolicum Keng.] and two interspecific hybrid populations. Euclidean distances were calculated on the basis of marker polymorphism and a cluster analysis was performed on these populations. Fairway, Parkway, and S9240 grouped closely, and it is likely that these are the only true A. cristatum populations included in this study. Hycrest, CD-II, and Kirk grouped loosely between A. cristatum and A. desertorum. This confirms the hybrid nature of Hycrest and CD-II but may indicate that Kirk is incorrectly classified as A. cristatum, but is rather a hybrid. Agropyron desertorum cv. Nordan clustered loosely between A. cristatum and A. mongolicum, which is in agreement with the suggestion that A. desertorum is an alloploid of A. cristatum and A. mongolicum. The cultivars Vavilov, Douglas, and Ephraim clustered with A. desertorum. An analysis of molecular variance (AMOVA) was conducted on 15 individuals from six of the populations (Fairway, Parkway, S9240, Kirk, Hycrest, and Nordan). The majority (88%) of the AFLP variance was within populations, which was expected in these outcrossing populations; however, significant differences were still detected among all six populations. Relationships among the six genotypes were the same for the bulk and individual plant (AMOVA) analysis, suggesting that bulked samples can be used for phylogenic studies of different populations.