Quantitative Genetic Dissection of Shoot Architecture Traits in Maize: Towards a Functional Genomics Approach

Authors: Lauter, Nicholas; MOSCOU, MATTHEW - Iowa State University; HABIGER, JOSH - Iowa State University; MOOSE, STEPHEN - University Of Illinois

Submitted to: The Plant Genome
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/18/2008
Publication Date: 11/20/2008
Citation: Lauter, N.C., Moscou, M.J., Habiger, J., Moose, S.P. 2008. Quantitative Genetic Dissection of Shoot Architecture Traits in Maize: Towards a Functional Genomics Approach. The Plant Genome. 01(02):99-110.

Interpretive Summary: Molecular genetic approaches to crop improvement rely on finding and favorably manipulating agriculturally important genes to improve crop productivity and utility, or to reduce inputs to farm, processing, or transport operations. The identification of such genes is currently the limiting step. Crop geneticists use three general approaches to identify agriculturally important genes: 1) introduce genetic variation through mutagenesis and try to identify favorable variants for the trait of interest, then find the causal molecular difference; 2) choose one of the approximately 40,000 genes of the crop to molecularly alter, then see if it affects the trait of interest; 3) identify natural variation in an important trait and breed an experimental population, then analyze the population to find the locations of the genes causing the observed variation. These three approaches are called forward, reverse, and quantitative genetics, respectively. While all three have utility for identifying agriculturally important genes, only the quantitative genetic approach takes advantage of natural genetic variation and allows crop breeders to proceed directly with targeted breeding approaches. In effect, "favorably manipulating" the agriculturally important genes identified in a quantitative genetic study simply requires plant breeding with assistance from genetic markers, since the actual identity of the gene causing the variation does not need to be known in order to take advantage of its effects. The research reported here utilized a quantitative genetic approach to identify genetic factors that affect the number of corn leaves produced before flowering and the position of the ear on the plant. Seven genetic factors were identified that affect these traits, which are of increasing economic value as we consider the contrasting demands of the food/feed versus bio-fuel uses for cereal crops. Each of these was mapped to a narrow region of the corn genome with a high level of certainty. These narrow regions are expected to contain an average of 65 genes, at least one of which must be responsible in each case for the variation in agronomic performance observed. Only two of the seven genetic factors identified in this report have been previously shown by other studies to affect either of these two traits, which is enough to confirm the validity of our results without detracting from their novelty. One of the genetic factors identified and mapped is thought to specifically affect the balance of leaves above and below the ear, an important measure known to affect grain-filling and lodging, which in turn affect grain quality and harvestability. Five of the seven genetic factors exert their effects on the number of leaves below the ear simply by increasing the total number of leaves (TNOL), which is a function of both the rate of leaf initiation and the duration of the leaf initiating period prior to flowering. Further experiments will be required to determine how leaf number is increased by the individual genetic factors, although 3 of the 6 genetic factors map to the same positions as genetic factors already known to affect "days till pollen shed," a widely used measure of flowering time in corn. The 6 genetic factors affecting TNOL collectively result in the difference between having 17 and 20 leaves. The parents of the experimental population have 17 and 20 leaves. The parental difference is entirely accounted for by the genetic factors characterized by this work. No single genetic factor identified causes an average increase in leaf number of more than one leaf, so all seven are good candidates for use in breeding schemes where incremental manipulation is desired. The quantitative genetic studies that have led to this report employed several novel experimental breeding and statistical genetic analysis strategies. Our study used

Technical Abstract: Quantitative trait loci (QTL) affecting the total number of leaves (TNOL) made prior to flowering and the number of leaves below the ear (NLBE) were mapped and characterized in order to dissect the genetic regulatory components of these agronomically important traits of corn. The full set of intermated B73 X Mo17 recombinant inbred lines (IBMRILs), which had previously been genotyped at more than 2,000 loci, were phenotyped in three separate years, allowing robust assessment of heritabilities as well as QTL positions and effects. B73 and Mo17 typically make 20 and 17 leaves, 14 and 11 of which are below the ear, respectively. The IBMRILs have high heritabilities for TNOL (0.81) and NLBE (0.77) and show transgression for both traits, making 15-24 leaves with 10-18 of them below the ear. Composite interval mapping identified five moderate-effect QTL for each trait, four of which appear to affect both of these highly correlated traits (r=0.86, p<0.0001). These map to chromosome bins 1.06, 8.05, 9.07, and 10.04. TNOL- and NLBE-specific QTL were detected in bins 4.08 and 3.06, respectively. QTL analysis of the proportion of leaves below the ear (PLBE) detected a single major-effect QTL in bin 3.06, supporting the inferences of pleiotropy for the other NLBE QTL. B73 alleles at all 6 QTL increase leaf numbers. Linear mixed model estimates of the collective effects of all 5 QTL for TNOL and NLBE were 40.9% and 40.7% of the heritable variation, corresponding to 103% and 77% of the mean parental differences. Thus, the parental differences between B73 and Mo17 are squarely accounted for by a handful of QTL, but the basis for the transgressive and heritable progeny phenotypes remains too complex to elucidate with this experimental design.