Expanding the repertoire of microRNA target mimicry in plants.

Abstract

microRNAs (miRNAs) are master regulators of gene expression, guiding ARGONAUTE proteins to bind to and repress target RNAs. Interestingly, a special class of target RNAs, termed target mimics (TMs), can in turn trigger miRNA degradation in plants-a process genetically dependent on the F-box protein HAWAIIAN SKIRT (HWS). However, the pairing rules governing effective TM sites remain unclear. Here, we systematically investigate the pairing architectures that enable plant TM-directed miRNA degradation. Using transient expression in Nicotiana benthamiana leaves and validation in stable Arabidopsis thaliana transgenic lines, we demonstrate that effective TM sites contain a central or near-central unpaired region-classified as insertions (I-type), mismatches (X-type), or deletions (D-type)-flanked by complementary segments. I-type sites tolerate considerable variation in bulge size and position, whereas X- and D-types are more constrained. By incorporating these features into a predictive pipeline, we identified endogenous TMs with alternative pairing architectures, including EARLY NODULIN-LIKE PROTEIN5 (ENODL5), which fine-tunes miR159 levels during Arabidopsis floral development in an HWS-dependent manner. Duplex stability and non-coding context also contribute to TM efficacy. Together, our findings establish a robust empirical framework for understanding, designing, and predicting miRNA TMs in plants.