Vernalization-induced flowering in cereals is associated with changes in histone methylation at the VERNALIZATION1 gene

Abstract

Prolonged exposure to low temperatures (vernalization) accelerates the transition to reproductive growth in many plant species, including the model plant Arabidopsis thaliana and the economically important cereal crops, wheat and barley. Vernalization-induced flowering is an epigenetic phenomenon. In Arabidopsis, stable down-regulation of FLOWERING LOCUS C (FLC) by vernalization is associated with changes in histone modifications at FLC chromatin. In cereals, the vernalization response is mediated by stable induction of the floral promoter VERNALIZATION1 (VRN1), which initiates reproductive development at the shoot apex. We show that in barley (Hordeum vulgare), repression of HvVRN1 before vernalization is associated with high levels of histone 3 lysine 27 trimethylation (H3K27me3) at HvVRN1 chromatin. Vernalization caused increased levels of histone 3 lysine 4 trimethylation (H3K4me3) and a loss of H3K27me3 at HvVRN1, suggesting that vernalization promotes an active chromatin state at VRN1. Levels of these histone modifications at 2 other flowering-time genes, VERNALIZATION2 and FLOWERING LOCUS T, were not altered by vernalization. Our study suggests that maintenance of an active chromatin state at VRN1 is likely to be the basis for epigenetic memory of vernalization in cereals. Thus, regulation of chromatin state is a feature of epigenetic memory of vernalization in Arabidopsis and the cereals; however, whereas vernalization-induced flowering in Arabidopsis is mediated by epigenetic regulation of the floral repressor FLC, this phenomenon in cereals is mediated by epigenetic regulation of the floral activator, VRN1.

Fig. 1.

The effect of vernalization on histone modifications at HvVRN1, HvVRN2, and HvFT1 in barley seedlings. (A) Diagram of the 5′ end of HvVRN1 showing the regions (1–6, short dashed lines) analyzed by ChIP, followed by quantitative real-time PCR. (B) HvVRN1 expression in nonvernalized (NV) and vernalized (V) seedlings harvested immediately at the end of vernalization. Data represents the mean ± SEM from 6 biological replicates. (C–F) Relative abundance of H3K4me3 (C and D) and H3K27me3 (E and F) at HvVRN1, HvVRN2, and HvFT1 in nonvernalized (NV) and vernalized (V) seedlings from the varieties Sonja (C and E) and Morex (D and F). Note that Morex contains a 5.2-kb deletion in the first intron of HvVRN1 (regions 4 and 5 are deleted) and lacks HvVRN2. Data represents the mean ± SEM from at least 3 biological replicate experiments. An asterisk (*) indicates significantly different to NV (P < 0.05).

Fig. 2.

The effect of seed vernalization on histone modifications at HvVRN1, HvVRN2, and HvFT1 in leaves postvernalization. (A–C) HvVRN1 (A), HvVRN2 (B), and HvFT1 (C) expression in nonvernalized (NV) and postvernalized (PV) leaves from the barley varieties Sonja and Morex. PV leaves were taken from plants at the third leaf stage grown in long days from vernalized seed. Data represents the mean ± SEM from 6 biological replicates. (D–G) Relative abundance of H3K4me3 (D and E) and H3K27me3 (F and G) at HvVRN1, HvVRN2, and HvFT1 in nonvernalized (NV) and postvernalized (PV) leaves from Sonja (D and F) and Morex (E and G). HvVRN1 regions analyzed are shown in Fig. 1A. Note that Morex contains a 5.2-kb deletion in the first intron of HvVRN1 (regions 4 and 5 are deleted) and lacks HvVRN2. Data represents the mean ± SEM from at least 3 biological replicate experiments. An asterisk (*) indicates significantly different to NV (P < 0.05).