|Home | About | Journals | Submit | Contact Us | Français|
Small ribonucleic acids (smRNA) have been identified as important post-transcriptional regulators of gene expression. One important class of smRNA implicated in stress responses are natural antisense short interfering RNA (natsiRNA). These natsiRNAs are generated from two overlapping mRNA that are processed into short-interfering RNAs and target complementary mRNA sequences for degradation. A feature of natsiRNA is the inverse regulation of gene expression that is observed after environmental or developmental stimuli. Genome-wide expression analysis of hypoxia treated Arabidopsis roots in combination with the examination of gene expression in plants defective in natsiRNA processing, was used to find putative natsiRNA regulated genes. The results indicate the potential involvement of natsiRNA in regulating lipid signaling in hypoxia.
Natural antisense transcripts short interfering RNAs (natsiRNA) are a class of endogenous ribonucleic acid regulators approximately 21–24 nt in length that are generated from regions of complementarity between mRNAs. These natsiRNA can occur either in cis (transcribed from the opposite strands at the same locus) or in trans (transcribed from different genomic locations—see Fig. 1). Analysis of sequenced organisms has shown that mRNA transcripts often overlap and that there exists widespread complementarity between transcripts at different loci, indicating that natsiRNA gene regulation could be widespread in eukaryotes.1–5 Currently, natsiRNA have been implicated in plant development and environmental responses including salt tolerance,6 bacterial resistance,7 cell wall biosynthesis8 cytokinin regulation9 and catalase expression.10 The formation of natsiRNA is not completely understood and can involve different DICER proteins (DCL1 or DCL2) and RNA-dependent RNA polymerases. Other proteins shown to be involved in the biogenesis pathway include NRPD1a, RDR6, SGS3, HYL1 and HEN1; possibly carrying out functions similar to their roles in the other smRNA pathways.6,7
A common feature of natsiRNA is the inverse regulation (the expression of one increases, while the other decreases) of genes involved that are triggered by environmental or developmental cues. The rapid rise of transcripts from the normally lowly expressed or silent gene, results in increased pairing between transcripts presumably reaching a threshold whereby they are recognised and processed to form siRNA. In Arabidopsis, it is the constitutive gene that is targeted for cleavage by the natsiRNA, although in the case of salt tolerance, the gene product arising from the induced gene is also important for stress tolerance.6
To determine whether potential natsiRNA are involved in response to hypoxia, a genome-wide expression analysis of Arabidopsis genes capable of forming natsiRNA during hypoxia was performed (Fig. 2). Arabidopsis plants were subjected to 5 h of either 0.1% oxygen or normal air (21% oxygen) and RNA from roots isolated and hybridized to Affymetrix Arabidopsis ATH1 genome arrays. The experiment was repeated twice. Bioinformatic analysis of the Arabidopsis genome has identified 1,126 putative cis-natsiRNA pairs3 and 1,320 putative trans-natsiRNA pairs.5 When the expression of these two groups of gene pairs were compared, 27 putative cis-nat-siRNA and 7 putative tran-natsiRNA gene were identified as possessing an inverse expression correlation (p value of <0.005, Table 1). Quantitative reverse transcriptase PCR (QPCR) was then conducted on the same RNA samples to validate the microarray results (list of primers used listed in Suppl. Table 1). Twenty putative cis-natsiRNA and all seven of the putative trans-natsiRNA were validated as having significant inverse expression.
The two characterised Arabidopsis natsiRNA both require NRPD1a for formation, as plants defective in this gene product suppress natsiRNA formation either partly or completely.6,7 Wild type and nrpd1a plants were grown and subjected to 5 h of 0.1% oxygen stress. QPCR was then conducted only on genes that experienced downward regulation in the pairing, as a reduction in natsiRNA regulation is predicted to result in derepression under hypoxia. The results showed that three genes AT2G38480, AT3G61300 and AT1G55390 had lost repression under hypoxia in the nrpd1a mutant by at least two-fold, and therefore are candidates for natsiRNA regulation (Table 2).
In our recent work, deep sequencing was used to directly find small RNAs that change in abundance during hypoxia.11 A large list of putative small RNAs derived from regions of complementarity between genes was found, although none belonging to AT2G38480, AT3G61300 or AT1G55390. However, the deep sequencing had not reached saturation, and most of the putative natsiRNA had a low numbers of reads.1–5 Looking for changes in gene regulation between genes that share transcript complementarity is an alternative method for finding genes regulated by natsiRNA, that may be more likely to find genes actively involved in the hypoxic response. The three candidate genes identified are currently uncharacterised and require further experimentation to validate that they are important for the response to hypoxia and are regulated by natsiRNA. The three associated hypoxia upregulated genes AT3g61310 (DNA binding protein), At2g38470 (WRKY33) and At5g42270 (VAR1; ATP-dependent peptidase) are also potential candidates involved in hypoxia.
The three putative targets for natsiRNA regulation are all likely to be involved in signalling processes and AT3G61300 and AT1G55390 possess C2 and divergent C1 domain containing (DC1) regions respectively that are predicted to function in lipid signalling. It is interesting to note that another DC1 protein AT1G55430 also showed decreased downward regulation in the nrpd1a mutant, just failing to meet our cutoff (1.8-fold reduction). Therefore, it appears that natsiRNA may have a specific role in controlling lipid signalling during hypoxia.
D. Moldovan was supported by a CSIRO Office of the Chief Executive postgraduate scholarship. We would like to thank Gavin Kennedy for his expert bioinformatics support.