Genes with a Combination of Over-Dominant and Epistatic Effects Underlie Heterosis in Growth of Saccharomyces cerevisiae at High Temperature

Abstract

Heterosis describes a phenotypic phenomenon of hybrid superiority over its homozygous parents. It is a genetically intriguing phenomenon with great importance for food production. Also called hybrid-vigor, heterosis is created by non-additive effects of genes in a heterozygous hybrid made by crossing two distinct homozygous parents. Few models have been proposed to explain how the combination of parental genes creates an exceptional hybrid performance. Over-dominant mode of inheritance is an attractive model since a single gene can potentially create the heterotic effect, but only a few such loci have been identified. To a collection of 120 hybrids, made by crossing 16 divergent Saccharomyces cerevisiae yeast strains, we applied a method for mapping heterozygous loci that non-additively contribute to heterotic growth at 37°. Among 803 candidate loci that were mapped, five were tested for their heterotic effect by analyzing backcrosses and F2 populations in a specific hybrid background. Consistently with the many mapped loci, specific analyses confirmed the minor heterotic effect of the tested candidate loci. Allele-replacement analyses of one gene, AEP3, further supported its heterotic effect. In addition to over-dominant effects, the contribution of epistasis to heterosis was evident from F2 population and allele-replacement analyses. Pairs of over-dominant genes contributed synergistically to heterosis. We show that minor over-dominant effects of multiple genes can combine to condition heterosis, similarly to loci affecting other quantitative traits. Furthermore, by finding of epistatic interactions between loci that each of them individually has an over-dominant effect on heterosis, we demonstrate how hybrid advantage could benefit from a synergistic combination of two interaction types (over-dominant and synergistic epistatic). Thus, by portraying the underlying genetic complexity, these findings advance our understanding of heterosis.

Keywords: QTL; epistasis; genetic interaction; hybrid vigor; over-dominance; quantitative traits; yeast.

Figure 1

Genomic distribution of candidate over-dominant loci. X-axis denotes the position along the chromosome (in Kb) and Y-axis denotes the chromosome numbers. Each dot marks one of the 803 windows that were found to be significant in the genome-wide scan. Green dots mark the 55 potential over-dominant windows in the SK1xS1001 background and red dots the 12 statistically significant ones.

Figure 2

Effects of five candidate loci in three analyses. (A) Reanalysis of 120 hybrids in three genotypic groups. Mean d/a (+/−StdErr) for three genotypic groups of hybrids (heterozygotes and homozygotes for one or the other allele) were compared using Tukey–Kramer HSD test. Different means in the TK test are denoted by different letters. (B) Analysis in RBC1 populations. Mean DT (+/−StdErr) at 37° for each window. Hom and Het denote homozygotes and heterozygotes for the allele, respectively. Significant differences between local heterozygotes and local homozygotes are marked by asterisks (Wilcoxon test, P < 0.05). (C) Analysis in F2 population. Mean DT (+/−StdErr) at 37° of each local genotype for each window. Different means in the TK test are denoted by different letters.

Figure 3

Effect of AEP3 allele replacement in haploid strains. Comparison of DT-values at 30° and 37° of AEP3 genotypes. At each temperature, three haploid strains in each genetic background were compared: original haploid, original after AEP3 deletion and the original after introducing the corresponding AEP3 allele from the other strain. Mean Log₂DT was compared among six strains at each temperature by a Tukey–Kramer HSD test. Bars with different letters denote genotypes with different growth rates, separately for each temperature (P < 0.05).

Figure 4

Effects of AEP3 genotypes in three diploid genetic backgrounds at 37°. Comparisons of DT between local genotypes of AEP3 (homozygotes to S1001 allele, homozygotes to SK1 allele and heterozygotes) were carried out in three diploid genetic backgrounds: heterozygous SK1xS1001 hybrid, S1001 homozygous and SK1 homozygous. Within each genetic background, the differences between three AEP3 genotypes were tested by a Tukey–Kramer HSD test. Bars show means and StdErr of DT at 37°. Different letters denote a significant difference (P < 0.05), separately for each genetic background.