Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: February 9, 2004
Publication Date: January 4, 2005
Citation: Casler, M.D. 2005. Agricultural fitness of smooth bromegrass populations selected for divergent fiber concentration. Crop Science. 45:36-43. Interpretive Summary: During selection for a particular trait in a plant breeding program, lots of other traits may change as well. These "correlated" changes may occur because a gene controls more than one trait (pleiotropy), because genes controlling the two traits are in close physical proximity on a chromosome (linkage), or because of random segregation (drift). The distinctions among these three causes are important, because the solutions all differ. Solutions may be necessary, because some correlated changes are undesirable. Casler developed a technique to separate the effects of each causal factor, using divergent selection, simultaneous selection in the high and low directions. Dietary fiber was the selection criterion. There were four response criteria: forage yield, seed yield, lodging (standability), and survival. For forage yield, pleiotropy was the most important factor, indicating that dietary fiber and forage yield share many of the same genes. For survival and lodging, linkage was the most important factor, indicating that simultaneous selection for these traits will be successful. For seed yield, drift was the most important factor, indicating that random changes in genes (inbreeding) can have a negative effect on seed yield, as has been shown in previous studies. This technique has provided valuable insight into the nature of genetic correlations between important traits in a perennial forage grass. The study will be extremely useful in shaping and guiding future research and plant breeding for forage nutritional value. The technique should be useful for a large number of species, traits, and applications.
Technical Abstract: Genetic correlations can be caused by pleiotropy, linkage, or drift. In the case of agriculturally undesirable genetic correlations, the cause is important, because the solutions to each cause are radially different. Divergent selection in four germplasms, analyzed with a factorial ANOVA model, was used to test and estimate the effects of pleiotropy, linkage, and drift on several genetic correlations. Selection was for neutral detergent fiber (NDF) concentration, an approximate measure of cell-wall concentration in smooth bromegrass (Bromus inermis Leyss.). Four agricultural fitness traits were measured: forage (biomass) yield, survival, seed yield, and lodging. Pleiotropy was estimated as the main effect linear response, linkage was estimated as the heterogeneity in linear responses across germplasms, and drift was estimated as asymmetry of linear responses. Selection responses for forage yield were linear and homogeneous, suggesting pleiotropic effects. Growth and expansion of the cell wall appears to be essential for accumulation of forage yield, lending a certain allometry to these two traits. However, natural selection within swards appeared to regulate this response, with forage yield responses disappearing by the 3rd production year. Selection responses for survival and lodging were linear, but non-homogeneous, suggesting linkage. Selection responses for seed yield appeared to be regulated by all three phenomena, with drift (asymmetry) the most important. Seed yield is highly sensitive to inbreeding depression, which occurs as a result of drift. Short-term divergent selection experiments, analyzed by a factorial ANOVA model, provide a mechanism to identify genetic phenomena responsible for observed genetic correlations.