Research Article: Using Association Mapping in Teosinte to Investigate the Function of Maize Selection-Candidate Genes

Date Published: December 9, 2009

Publisher: Public Library of Science

Author(s): Allison L. Weber, Qiong Zhao, Michael D. McMullen, John F. Doebley, Pär K. Ingvarsson.

Abstract: Large-scale screens of the maize genome identified 48 genes that show the putative signature of artificial selection during maize domestication or improvement. These selection-candidate genes may act as quantitative trait loci (QTL) that control the phenotypic differences between maize and its progenitor, teosinte. The selection-candidate genes appear to be located closer in the genome to domestication QTL than expected by chance.

Partial Text: Past natural or artificial selection leaves its signature on the genome by altering the levels and pattern of nucleotide diversity in a population or species. Advances in high-throughput genotyping and sequencing have enabled large-scale and genomic-wide scans for the signature of selection which identify sets of candidate genes that were putative targets of selection during the history of a population or species [1], [2]. This approach represents a promising way of identifying genes controlling traits important for adaptation in natural species or agronomic traits in crops. An attractive feature of selection scans is that they may identify genes controlling traits that investigators may not have considered important a priori[3].

We used a mixed linear model to test for association between SNP variation in 35 selection-candidate genes and trait variation in teosinte [11], [12]. Our teosinte sample includes two previously described panels: Panel A consists of 584 plants sampled from 74 local populations [13], and Panel B consists of 817 plants from 34 local populations [14]. Twenty-four of the 35 selection-candidate genes were those identified by Wright et al. (2005) [5], eight were identified by Yamasaki et al. (2005) [6], and three were identified using sequence data from the Maize Functional Diversity Project ( Traits assayed included those measuring flowering time, plant architecture, inflorescence architecture, vegetative morphology and kernel composition (Table 1; Table S1, Supplementary section). Most of these traits measure aspects of morphology that were altered during maize domestication and/or improvement. The two association panels were analyzed separately. For Panel A, 56 SNPs from 26 selection-candidate genes were tested for association with 17 traits, and for Panel B, 75 SNPs from 35 selection-candidate genes were tested for association with 31 traits (Table S2, Supplementary section). Overall, 82 SNPs in 35 selection-candidate genes and 32 traits were assayed.

Our association mapping study involving 35 selection-candidate genes, 32 traits and two teosinte association mapping panels found seven significant associations between selection-candidate genes and trait variation in teosinte. These associations should be regarded with caution and additional molecular work will be necessary to validate them. Interesting associations were detected between the maize homologs of two Arabidopsis genes, ARF1 and ARF2, and days to pollen (POLL). Mutant analysis in Arabidopsis indicates that both these genes affect flowering time [22]. Curiously, flowering time was not a trait hypothesized to have been under selection during maize domestication, and thus these putative associations may be independent of the selective history of these genes. Associations were also detected between AY106616, an ankyrin-repeat like protein, and derived starch content (DSCT) and oil content (OLCT). Previous studies in Arabidopsis and tobacco indicate that ankyrin-repeat like proteins are involved in carbohydrate metabolism and allocation [23], [24]. These significant associations, as well as those involving less well-annotated genes, provide information that can now be used as a starting point for further experimentation that will validate the function of these genes.



0 0 vote
Article Rating
Notify of
Inline Feedbacks
View all comments