Segregation distortion and overdominance for heterosis in an elite maize single cross

Hong Lu, Purdue University

Abstract

There has been no satisfactory explanation of the genetic mechanisms for heterosis. Molecular markers enable the study of heterosis at the single-marker level. But molecular markers have often been found to have segregation distortion. The objectives of this study were to (1) identify the simple sequence repeat (SSR) markers closely linked with quantitative trait loci (QTL) for grain yield and for related agronomic traits in an elite maize (Zea mays L.) single cross, (2) elucidate the importance of partial-to-complete dominance versus overdominance or pseudo-overdominance for heterosis at the molecular marker level, (3) identify chromosomal regions associated with segregation distortion in maize, and (4) examine the effects of segregation distortion on linkage maps and on QTL mapping. Eighteen chromosomal regions on the 10 maize chromosomes were associated with segregation distortion. Three known gametophyte factors (Ga) are possible genetic causes of these segregation distortion regions (SDR). Segregation distortion induced by Ga factors is demonstrated not to bias the estimates of linkage map and the position and effect of the linked QTL. For a marker or a QTL linked with a Ga, the homozygotes are expected to be more likely and severely distorted than the heterozygotes. Phenotypic data analyses showed overdominance for grain yield and partial dominance for plant height, grain moisture, and stalk lodging. The total number of QTL identified in two backcross populations was 28 for grain yield, 11 for plant height, 16 for grain moisture, and 8 for stalk lodging. Eighty-six percent (86%) of QTL for grain yield showed overdominance. In contrast, the majority of QTL for plant height, grain moisture, and stalk lodging showed partial to complete dominance. No evidence was found for two-locus epistasis. Six QTL were identified to have effects on two to four traits. Pure additive gene action within each gene could cause variable-dominance in the end. The variable-dominance, proposed in this thesis, suggests the level of dominance may vary from partial to over dominance depending on specific crosses. A bottleneck model may explain the genetic mechanisms for heterosis.

Degree

Ph.D.

Advisors

Romero-Severson, Purdue University.

Subject Area

Agronomy|Molecular biology

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS