Abscisic Acid (ABA) is an important phytohormone involved in abiotic stress resistance in plants. A group of bZIP transcription factors play important roles in the ABA signaling pathway in Arabidopsis. However, little is known about the function of their orthologs in rice, where they may hold a great potential for developing drought resistant food crops. In this study, our phylogenetic analysis showed that this group of bZIPs was evolutionarily conserved between Arabidopsis and rice, which implies that they may share similar functions. We demonstrated with quantitative RT-PCR that the expressions of most of these OsbZIPs were signiWcantly induced by ABA, ACC, and abiotic stresses. OsbZIP72, a member of this group, was proved to be an ABRE binding factor in rice using the yeast hybrid systems. We showed that it could bind to ABRE and transactivate the downstream reporter genes in yeast, and the transactivity was depending on its N-terminal region. Transgenic rice overexpressing OsbZIP72 showed a hypersensitivity to ABA, elevated levels of expression of ABA response gene such as LEAs, and an enhanced ability of drought tolerance. These results suggest that OsbZIP72 plays a positive role in drought resistance through ABA signaling, and is potential useful for engineering drought tolerant rice.

The functional complexity of the rice transcriptome is not yet fully elucidated, despite many studies having reported the use of DNA microarrays. Next-generation DNA sequencing technologies provide a powerful approach for mapping and quantifying the transcriptome, termed RNA sequencing (RNA-seq). In this study, we applied RNA-seq to globally sample transcripts of the cultivated rice Oryza sativa indica and japonica subspecies for resolving the whole-genome transcription profiles. We identified 15,708 novel transcriptional active regions (nTARs), of which 51.7% have no homolog to public protein data and >63% are putative single-exon transcripts, which are highly different from protein-coding genes (<20%). We found that~48% of rice genes show alternative splicing patterns, a percentage considerably higher than previous estimations. On the basis of the available rice gene models, 83.1% (46,472 genes) of the current rice gene models were validated by RNA-seq, and 6228 genes were identified to be extended at the 59 and/or 39 ends by at least 50 bp. Comparative transcriptome analysis demonstrated that 3464 genes exhibited differential expression patterns. The ratio of SNPs with nonsynonymous/synonymous mutations was nearly 1:1.06. In total, we interrogated and compared transcriptomes of the two rice subspecies to reveal the overall transcriptional landscape at maximal resolution.