Identifier

etd-07082007-102140

Degree

Master of Science (MS)

Department

Plant, Environmental Management and Soil Sciences

Document Type

Thesis

Abstract

Aflatoxin contamination of maize (Zea mays L.) grain caused by Aspergillus flavus is a serious health hazard to animals and humans. Resistance to infection by A. flavus is poorly understood. The objectives of this investigation were to identify potential candidate markers associated with resistance in maize kernels and pollen grains to A. flavus using a mapping population derived from a cross between Mp313E (resistant) and SC212m (susceptible) inbred lines. The parents, F1, and F2 plants were planted in the field in 2005. Each F2 plant was selfed to produce F2:3 seed. Fresh pollen collected from each F2 plant was germinated on a growth medium in the presence of A. flavus conidia. Selfed seeds from parents, F1, and F2 plants were challenged with A. flavus conidial suspension and incubated using a medium-free method. Percent kernels uninfected (PKU) and number of pollen grains germinated (NPG) were recorded. A linkage map was constructed with JoinMap 3.0 using DNA profiles of all F2 individuals produced from amplified fragment length polymorphism (AFLP) and target region amplification polymorphism (TRAP) markers. Interval mapping and multiple-QTL model (MQM) mapping analyses were performed using MapQTL 4.0 software. Three marker-QTL associations were observed for log-transformed PKU. Potential markers associated with this trait were also identified via discriminant analysis (DA). The markers identified via DA pointed to the same genomic regions as identified via the QTL mapping strategy. For log-transformed NPG, five marker-QTL associations were detected. One QTL was associated with a TRAP marker. The DA confirmed the existence of three QTL. The QTL detected for NPG were different from the QTL detected for PKU. Resistances of pollen and kernels to A. flavus appeared to be controlled by different genetic systems/mechanisms. Correlation between pollen germination and percent kernel infection was negligible (r = 0.067), suggesting that the two traits can be improved independently.

Date

2007

Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Collins Kimbeng

Share

COinS