SummaryPlants with winter annual life history germinate in summer or autumn and require a period of prolonged winter cold to initiate flowering, known as vernalization. In the Brassicaceae, the requirement for vernalization is conferred by high expression of orthologs of the FLOWERING LOCUS C (FLC) gene, the expression of which is known to be silenced by prolonged exposure to winter-like temperatures [1]. Based on a wealth of vernalization experiments, typically carried out in the range of 5°C–10°C, we would expect field environments during winter to induce flowering in crops with winter annual life history. Here, we show that, in the case of winter oilseed rape, expression of multiple FLC orthologs declines not during winter but predominantly during October when the average air temperature is 10°C–15°C. We further demonstrate that plants proceed through the floral transition in early November and overwinter as inflorescence meristems, which complete floral development in spring. To validate the importance of pre-winter temperatures in flowering time control, we artificially simulated climate warming in field trial plots in October. We found that increasing the temperature by 5°C in October results in raised FLC expression and delays the floral transition by 3 weeks but only has a mild effect on flowering date the following spring. Our work shows that winter annuals overwinter as a floral bud in a manner that resembles perennials and highlights the importance of studying signaling events in the field for understanding how plants transition to flowering under real environmental conditions.

Summary Flowering time is a key adaptive and agronomic trait. In Arabidopsis, natural variation in expression levels of the floral repressor FLOWERING LOCUS C (FLC) leads to differences in vernalization. In Brassica napus there are nine copies of FLC. Here, we study how these multiple FLC paralogues determine vernalization requirement as a system. We collected transcriptome time series for Brassica napus spring, winter, semi‐winter, and Siberian kale crop types. Modelling was used to link FLC expression dynamics to floral response following vernalization. We show that relaxed selection pressure has allowed expression of FLC paralogues to diverge, resulting in variation of FLC expression during cold treatment between paralogues and accessions. We find that total FLC expression dynamics best explains differences in cold requirement between cultivars, rather than expression of specific FLC paralogues. The combination of multiple FLC paralogues with different expression dynamics leads to rich behaviour in response to cold and a wide range of vernalization requirements in B. napus. We find evidence for different strategies to determine the response to cold in existing winter rapeseed accessions.

Winter, spring and biennial varieties of Brassica napus that vary in vernalization requirement are grown for vegetable and oil production. Here, we show that the obligate or facultative nature of the vernalization requirement in European winter oilseed rape is determined by allelic variation at a 10 Mbp region on chromosome A02. This region includes orthologues of the key floral regulators FLOWERING LOCUS C (BnaFLC.A02) and FLOWERING LOCUS T (BnaFT.A02). Polymorphism at BnaFLC.A02 and BnaFT.A02, mostly in cis-regulatory regions, results in distinct gene expression dynamics in response to vernalization treatment. Our data suggest allelic variation at BnaFT.A02 is associated with flowering time in the absence of vernalization, while variation at BnaFLC.A02 is associated with flowering time under vernalizing conditions. We hypothesize selection for BnaFLC.A02 and BnaFT.A02 gene expression variation has facilitated the generation of European winter oilseed rape varieties that are adapted to different winter climates. This knowledge will allow for the selection of alleles of flowering time regulators that alter the vernalization requirement of oilseed rape, informing the generation of new varieties with adapted flowering times and improved yields.