Mechanisms of microRNA-mediated auxin signaling inferred from the rice mutant osaxr

Abstract

Auxin, known as the central hormone, plays essential roles in plant growth and development. In auxin signaling pathways, the tiny RNA molecules, i.e., microRNAs (miRNAs), show their strong potential in modulating the auxin signal transduction. Recently, we isolated a novel auxin resistant rice mutant osaxr (Oryza sativa auxin resistant) that exhibited plethoric root defects. Microarray experiments were carried out to investigate the expression patterns of both the miRNAs and the protein-coding genes in osaxr. A number of miRNAs showed reduced auxin sensitivity in osaxr compared with the wild type (WT), which may contribute to the auxin-resistant phenotype of the mutant. Auxin response elements (AuxREs) were demonstrated to be more frequently present in the promoters of auxin-related miRNAs. In our previous report, a comparative analysis of miRNA and protein-coding gene expression datasets uncovered a number of reciprocally expressed miRNA-target pairs. A feedback circuit between miRNA and auxin response factor (ARF) was then proposed. Here, we will discuss indepth some points raised in the previous report, in particular, the organ-specific expression patterns of miR164, the feedback regulatory model between miR167 and certain ARFs, and the potential signal interactions between auxin and nutrition or stress that are mediated by miRNAs in rice roots.

In plants, numerous signaling pathways interact with each other and cooperate to shape their hosts’ growth, development and response. One of these essential pathways originates from auxin. A number of miRNAs have been reported to be involved in auxin signaling, such as miR160 and miR167, whose targets were ARFs.^1–5 ARFs, which constitute one TF family, play an important role in early auxin response.⁶ However, few data on auxin signaling has been reported in rice compared with Arabidopsis.

In our previous study, a novel auxinresistant rice mutant was phenotypically characterized.⁷ Most of its root system, including adventitious roots, lateral roots and root caps, were misshapen. Moreover, the mutant was designated as osaxr based on its apparent insensitivity to exogenous auxin stimulus.⁷ A microarray experiment was carried out to investigate the expression patterns of the miRNAs in osaxr. The result showed that the number of auxin-sensitive miRNAs was dramatically reduced in osaxr compared with the WT, indicating the involvement of certain miRNAs in the auxin-resistant phenotype of the mutant.⁷ Subsequently, the promoters of the miRNAs abnormally expressed in osaxr were selected for further analyses. The AuxREs were found to be more frequently present in the promoters of these miRNAs which were supposed to be auxin-related.⁷ We also demonstrated that the basic characteristic of plant miRNA promoters was similar to that of RNA polymerase II-dependent gene promoters,⁷ consistent with our previous report.⁸ Microarray analysis was also performed for the protein-coding genes. Comparative analysis of the miRNA and protein-coding gene expression datasets revealed many reciprocally expressed miRNA-target pairs,⁷ which could serve as a repository for miRNA downstream analyses.

A large portion of miRNA targets are TFs such as ARFs, suggesting that miRNAs are core regulators in plants.⁹ On the other hand, the expression of miRNAs must be tightly regulated by upstream regulators although little such regulatory relationship has been uncovered in plants. One case studied in Arabidopsis was that PHR1 could regulate the transcription of miR399 possibly by binding to the imperfect palindromic cis-element GNATATNC on the promoter of miR399.^10,11 Our study suggested that most plant miRNAs may share similar transcriptional mechanisms with the RNA polymerase II-dependent eukaryotic genes.⁷

Feedback regulatory relationships between miRNAs and TFs have been validated in animals and humans^12,13 but scarcely in plants. Only a general negative feedback model was proposed in plants by Megraw et al. (2006) after discovering five cis-elements for TF binding in the miRNA promoters.¹⁴ In our study, a negative feedback circuit between miR167 and ARF6 was proposed.⁷ It is the first feedback regulatory model on auxin signaling revealed in rice. Coincidentally, our proposed model was demonstrated to be an essential part of the integrated model between ARFs and miRNAs involved in the control of adventitious rooting in Arabidopsis when this addendum was under preparation.³ Our further research also showed that miR167 was involved in adventitious root development in rice by regulating its downstream targets, i.e., ARFs (data not shown). Moreover, AuxREs were demonstrated to be frequently present in the promoters of miR167 genes (Fig. 1). Thus, the feedback regulation was likely to be prevalent in miRNA-mediated auxin signaling pathways in plants and may extend to other signal transduction or regulatory pathways.

Figure 1.

AuxRE distribution patterns in the promoters of miR167 genes in rice. All members of miR167 family (miR167a-j) were included. The length of the promoters and the miRNA precursors is measured by x axis. Nine AuxRE s (see legend at the bottom. TATA (A|T)A(T|A)(A|G): TATA-box. ‘*’ represents any nucleic acid compositions less than 100 nt. ‘|’ means ‘or’.) were analyzed and their physical positions were marked below each promoter.

For northern confirmation, we selected four miRNAs with expression patterns that have shown significant changes in osaxr. All the results supported our microarray data. However, we observed an interesting phenomenon where two signal bands of osa-miR164abf appeared in osaxr in contrast to the single band in the WT.⁷ The expression pattern of the upper band was consistent with the microarray data but the in vivo function of the lower band that was intensively repressed in osaxr remains to be elucidated. Previous hybridization results demonstrated that miR156, miR160, miR164 and miR165 occurred in two distinct size classes of 21 nucleotides (nt) and 24 nt in Arabidopsis. The 24-nt miRNAs were found to be more abundant in leaves.¹⁵ This organ-specific expression pattern of miR164 probably applies to rice. Hence, the abnormal hybridization pattern of miR164 in the roots of osaxr was likely to be caused by the disordered organ-specific regulation of miR164 processing.

In addition, certain auxin-sensitive miRNAs in the wild type rice, such as miR395 and miR169, were not involved in the auxin signaling, based on previous reports.⁷ Instead, these miRNAs or their homologs in Arabidopsis were reported to be involved in sulphate metabolism and drought response respectively.^16–19 The miR167-mediated signal interaction between auxin and nitrogen involved in lateral root development was reported in Arabidopsis.²⁰ Considering the widespread interactions between hormones and nutrition or stress in plants, we proposed a miRNA-mediated auxin—nutrition (or stress) interaction model involved in rice root development (Fig. 2). However, elaborate experiments are needed to validate the model and uncover its biological roles in plants.

Figure 2.

MicroRNA -mediated signal interactions between auxin and nutrition or stress in rice roots. The signal interactions can occur both upstream and downstream of the miRNA s (e.g., miR169 and miR395). The feedback regulation between miRNA s and TFs has been uncovered by previous studies.³,⁷ Almost all the miRNA s exert negative regulatory effects on their targets in plants, thus dashed lines were used to depict the positive effects of miRNA s on their targets. Other auxin, nutrition (or stress) signaling pathways, bypassing the miRNA—mediated ones involved in rice root development, could not be excluded.

Abbreviations

ARF

auxin response factor

AuxRE

auxin response element

miRNA

microRNA

nt

nucleotide(s)

osaxr

Oryza sativa auxin resistant

TF

transcription factor

WT

wild type

Addendum to: Meng Y, Huang F, Shi Q, Cao J, Chen D, Zhang J, et al. Genome-wide survey of rice microRNAs and microRNA-target pairs in the root of a novel auxin-resistant mutant. Planta. 2009;230:883–898. doi: 10.1007/s00425-009-0994-3.