The commitment of barley microspores into embryogenesis correlates with miRNA-directed regulation of members of the SPL, GRF and HD-ZIPIII transcription factor families

Abstract

Microspore embryogenesis is a model for developmental plasticity and cell fate decisions. To investigate the role of miRNAs in this development, we sequenced sRNAs and the degradome of barley microspores collected prior to (day 0) and after (days 2 and 5) the application of a stress treatment known to induce embryogenesis. Microspores isolated at these timepoints were uniform in both appearance and in their complements of sRNAs. We detected 68 miRNAs in microspores. The abundance of 51 of these miRNAs differed significantly during microspore development. One group of miRNAs was induced when the stress treatment was applied, prior to being repressed when microspores transitioned to embryogenesis. Another group of miRNAs were up-regulated in day-2 microspores and their abundance remained stable or increased in day-5 microspores, a timepoint at which the first clear indications of the transition toward embryogenesis were visible. Collectively, these miRNAs might play a role in the modulation of the stress response, the repression of gametic development, and/or the gain of embryogenic potential. A degradome analysis allowed us to validate the role of miRNAs in regulating 41 specific transcripts. We showed that the transition of microspores toward the embryogenesis pathway involves miRNA-directed regulation of members of the ARF, SPL, GRF, and HD-ZIPIII transcription factor families. We noted that 41.5% of these targets were shared between day-2 and day-5 microspores while 26.8% were unique to day-5 microspores. The former set may act to disrupt transcripts involved in pollen development while the latter set may drive the commitment to embryogenesis.

Keywords: Hordeum vulgare; degradome; gametic embryogenesis; hc‐siRNA; miRNA; phasiRNA.

FIGURE 1

The phenotype of these microspores was captured on phased‐contrast (left) and with DAPI‐excited (right) microscopy for microspores on days 0, 2, and 5 using a Zeiss Apoptome microscope under UV laser illumination (excitation of 390/22 nm and emission of 460/50 nm) at a 40x magnification

FIGURE 2

Dynamics of variation in sRNA abundance through barley microspore development. Microspores were analyzed in the early stages of development in androgenesis for differentially expressed (at logFC ≥ |2.0| and q‐value ≤ 1.0E‐03) hc‐siRNAs (a) 24‐nt phasiRNAs (b) and miRNAs (c). Heatmaps were generated from a read‐count matrix normalized for library size and transformed in reads per million. Heatmaps were row normalized

FIGURE 3

Abundance (y‐axis) and distribution (x‐axis) of reads aligned to two PHAS loci, 24PHAS_31814 (left) and 24PHAS_34365 (right), that were up‐regulated in day‐5 microspores and overlapped with protein‐coding regions. Positive and negative values on the y‐axis refer to the two DNA strands on which phasiRNAs map

FIGURE 4

Box plots give the abundance of miRNA (left), PARE (center) cleavage product and mRNA (right) data for miR1127‐HvHCT (a) and novel.3‐HvAHP4 (b) regulatory modules. Abundance of target RNA comes from an RNA‐seq experiment published by Bélanger et al. (2018)