Data from: Dearth of polymorphism associated with a sustained response to selection for flowering time in maize

Long term selection experiments bring unique insights on the genetic architecture of quantitative traits and their evolvability. Indeed, they provide the opportunity to (i) investigate the contribution of non allelic interactions; (ii) evaluate the role of standing variation versus new mutations during adaptation; (iii) monitor changes in allele frequencies and therefore assess the number of genomic regions involved in the trait determinism, and their effects. Here we describe genetic and phenotypic evolution of two independent Divergent Selection Experiments (DSEs) for flowering time conducted during 16 years from two early maize inbred lines.

Our experimental design includes selfing as mating system and small population sizes so that two independent families evolved within each population, Late and Early. Observed patterns are strickingly similar between the two DSEs. We observed a significant response to selection in both directions during the first 7 generations of selection. Within Early families, the response is linear through 16 generations, consistent with the maintenance of genetic variance. Within Late families and despite maintenance of significant genetic variation across 16 generations, the response to selection reached a plateau after 7 generations. This plateau is likely caused by physiological limits. Residual heterozygosity in the initial inbreds can partly explain the observed responses as evidenced by 42 markers derived from both Methyl-Sensitive Amplification- and Amplified Fragment Length- Polymorphisms. Among 42, 13 markers most of which are in high linkage disequilibrium, display a strong association with flowering time variation. Their fast fixation throughout DSEs' pedigrees result in strong genetic differentiation between populations and families.

Our results reveal a paradox between the sustainability of the response to selection and the associated dearth of polymorphisms. Among other hypotheses, we discuss the maintenance of heritable variation by few mutations with strong epistatic interactions whose effects are modified by continuous changes of the genetic background through time.