Skip to main content

Maintenance of cellular memory by the Arabidopsis VRN1 protein

Final Report Summary - VRNEPIGEN (Maintenance of cellular memory by the Arabidopsis VRN1 protein)

Flowering at the right time is a key determinant of reproductive success for plants. As an adaptation to local climates many plant species show accelerated flowering after being exposed to cold temperatures, a phenomenon known as vernalisation. The nature of this process relies on fundamentally conserved molecular mechanisms that allow cells to memorise (cold in this example) temporal signals over many cell cycles, a process called epigenetic gene regulation. Vernalisation is mediated by a set of Polycomb (PcG) proteins highly conserved between animals and plants. Polycomb proteins cause a physical closing of the major floral repressor gene FLC turning it stably inactive upon vernalisation. During the remainder of the plants life cycle FLC will remain off. Beside the PcG mechanism at least one parallel system helps to silence FLC expression. Vernalisation1 (VRN1), which encodes a B3-domain transcription factor, acts in parallel with PcG yet very little about its function is known. Earlier work on VRN1 has shown that mutations in vrn1 alleviate FLC repression and cause failure to correctly time reproduction. The VRN1 protein has little sequence binding specificity and coats all Arabidopsis chromosomes throughout the cell cycle, an interesting characteristic shared with only a few other known DNA binding proteins. Collectively these data make VRN1 a good candidate to fulfil a role in transmitting epigenetic information over multiple cell generations via a novel mechanism. To gain a better understanding of VRN1 function we have used both biochemical and genetic approaches.

We have mapped VRN1 binding sites across the FLC locus using a high-resolution Chromatin immuno-precipitation (ChIP) technique optimised in the host lab. As stated above previous in situ analysis showed VRN1 binds all five chromosomes suggesting a wide binding profile. Indeed moderate levels of VRN1 could be found across the FLC locus and surrounding sequences. In addition, we detected peaks of VRN1 binding at some nucleosome free regions. Upon FLC silencing by vernalisation the binding profile of VRN1 remained unchanged albeit total levels dropped, implying that the physical closing of FLC upon vernalisation limits VRN1 binding. Lower VRN1 binding indeed coincided with the repressive histone mark H3K27me3, which is deposited by the PcG machinery and accumulates at FLC upon vernalisation.

An unexpected finding in vrn1 plants was a lower total nucleosome level, which we subsequently have shown to be caused by a relatively higher turnover rate of nucleosomes at the chromatin. Using a combination of ChIP and Fluorescence recovery after photobleaching (FRAP) experiments we have also shown that increased nucleosome turnover results in excessive usage of variant histone proteins throughout the genome (a finding never reported before and highly interesting for understanding fundamental aspects of chromatin dynamics). Using a genetic and molecular approach we have also shown that aberrant histone variant usage is at least partially causative for the vernalisation phenotype in vrn1 plants. Collectively our data have shed new light on the dynamics of chromatin and how disturbance of balanced nucleosome turnover can have profound effects on gene regulation. The impact on the general field of chromatin biology will be substantial and therefore this project has contributed to a further strengthening of the EU as a world leading collective in chromatin research. Beside the impact on fundamental research applicability of research on flowering time has been proven economically important and profitable before. In that perspective this project has added another piece of useful information allowing us to better understand and eventually manipulate crop flowering behaviour and help securing food production in the future.