ABA signaling is necessary but not sufficient for RD 29 B transcriptional memory during successive dehydration stresses in A rabidopsis thaliana

• Published in: The Plant journal : for cell and molecular biology Vol. 79; no. 1; pp. 150 - 161
• Main Authors: Virlouvet, Laetitia, Ding, Yong, Fujii, Hiroaki, Avramova, Zoya, Fromm, Michael
• Format: Journal Article
• Language: English
• Published: 01.07.2014
• ISSN: 0960-7412; 1365-313X
• DOI: 10.1111/tpj.12548

Summary Plants subjected to a prior dehydration stress were seen to have altered transcriptional responses during a subsequent dehydration stress for up to 5 days after the initial stress. The abscisic acid ( ABA ) inducible RD 29B gene of A rabidopsis thaliana was strongly induced after the first stress and displayed transcriptional memory with transcript levels nine‐fold higher during the second dehydration stress. These increased transcript levels were due to an increased rate of transcription and are associated with an altered chromatin template during the recovery interval between the dehydration stresses. Here we use a combination of promoter deletion/substitutions, mutants in the trans ‐acting transcription factors and their upstream protein kinases, and treatments with exogenous ABA or dehydration stress to advance our understanding of the features required for transcriptional memory of RD 29B . ABA Response Elements ( ABRE s) are sufficient to confer transcriptional memory on a minimal promoter, although there is a context effect from flanking sequences. Different mutations in Snf1 Related Protein Kinase 2 ( Sn RK 2 ) genes positively and negatively affected the response, suggesting that this effect is important for transcriptional memory. Although exogenous ABA treatments could prime transcriptional memory, a second ABA treatment was not sufficient to activate transcriptional memory. Therefore, we concluded that transcriptional memory requires ABA and an ABA ‐independent factor that is induced or activated by a subsequent dehydration stress and directly or indirectly results in a more active RD 29B chromatin template. These results advance our knowledge of the cis ‐ and trans ‐acting factors that are required for transcriptional memory of RD 29B .