Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/24/2001
Publication Date: N/A
Citation: N/A Interpretive Summary: The hybrid maize industry is founded on the use of heterosis to recover the performance lost in producing the inbred lines used to make the hybrids. The loss of performance of inbreds is an effect of inbreeding (the mating of relatives) and is called inbreeding depression. Inbreeding depression is often considered to be the converse of heterosis. Despite theoretical and empirical research spanning nearly 60 years, we still know very little about the genetic basis of heterosis. This is partly because heterosis is difficult to study empirically. Inbreeding depression is easier to study empirically and detailed knowledge of the genetic basis of inbreeding depression will also provide information about heterosis. In a theoretical and empirical study of inbreeding depression, we found that dominance deviations are much more important at the inbred level than previously thought. Furthermore, these dominance deviations are heritable, which means that selection can be effective for decreasing the amount of inbreeding depression observed in a population. Unfortunately, we also found a large correlation between inbred genotypic value and these dominance deviations. This means that selection on inbred performance, as is often done by breeders, will improve ultimate inbred performance but will not result in an increase in the performance of the noninbred (open- pollinated) population which is the desired result. This research represents the most detailed study of inbreeding depression in maize to date and will be useful to maize breeders and geneticists in understanding the genetic basis of heterosis and inbreeding depression.
Technical Abstract: The average effects of inbreeding depression have been measured extensively in maize (Zea mays L.), but the influence of inbreeding on genetic variance has not been well studied. We developed 200 random inbred lines in the BS13(S)C0 maize population and a set of 200 related half-sib families. The lines and half-sib families were evaluated in replicated yield trials for six agronomic characters. Three of the traits in our study were influence by dominance, while three traits exhibited negligible effects of dominance. For traits not influenced by dominance, little inbreeding depression was observed, and the genetic variance nearly doubled with inbreeding, as expected under an additive model. For traits influenced by dominance, we found that dominance deviations had a much larger variance in inbred individuals than in noninbred individuals and were negatively correlated with breeding values in inbred individuals. The total genetic variance increased by 1.5 times or less for traits influenced by dominance. An estimator of the degree of dominance in a population with arbitrary allele frequencies was developed using the covariance components for inbred relatives. For the three traits influenced by dominance, the average degree of dominance was suggestive of overdominance, which we attributed to linkage disequilibrium in the population.