Roles of a glutamyl-tRNA synthetase in controlling early anther development in rice

Abstract

In seed-propagated plants, the formation and development of the male reproductive organ, the anther, is indispensable for plant propagation. In grain crops, crop productivity is highly dependent on male fertility and thereafter successful fertilisation between male and female gametophytes. Understanding the molecular mechanisms determining anther development and formation of pollen grains in important crops such as rice (Oryza sativa) and barley (Hordeum vulgare) is important for fundamental biology and agricultural practice. In flowering plants, the male development starts from the formation of anther primordia, and subsequent cell division and differentiation establish the appropriate anther cell organisation. After morphogenesis, the anther in higher plants mostly contains centrally localised germinal cells surrounded by four layers of anther wall composed by the epidermis, endothecium, middle layer and tapetum, from the outer to the inner. The mature pollen grain is formed via meiosis and mitosis supported by the degeneration of anther wall layers such as the innermost tissue tapetum as well as the middle layer. Although previous investigations uncovered that various regulators such as leucine-rich repeat receptor-like kinases (LRR-RLKs), glutaredoxins, transcription factors, hormones and small RNAs are involved in anther cell fate specification, the role of genes responsible for physiological homeostasis in early male development remains unknown. Plant aminoacyl-tRNA synthetases (aaRSs) are enzymes that catalyse the attachment of amino acids onto their cognate transfer RNAs (tRNAs), playing a central role during the translation of genetic information from messenger RNA to protein. However, little is known about their molecular characteristics and roles in plant development. In the research for this thesis bioinformatics analysis was conducted and 141 aaRSs sequences were obtained from Arabidopsis (Arabidopsis thaliana), rice and Physcomitrella patens. In these sequences, beside the conserved motifs such as HIGH, KMSKS, Motif 2, Motif 3 required for the enzymatic functions, additional new domains such as GST_C (glutathione S-transferase C terminus), WHEP and OB (oligonucleotide binding motif) fold were observed. These may function in tRNA binding or protein-protein interaction. Sequence prediction showed that these plant aaRSs have subcellular localisation in cytosol, mitochondria and chloroplasts, which may be related with evolutionary origin, signal peptide, and intra/extracellular stimulations. Notably, the aaRS genes encoding cytosolic ones are more active in reproductive tissues while genes encoding organellar ones are more expressed in vegetative tissues. Together, we provide an informative source for future research and highlight the essential roles of aaRSs in plant. To understand the function of aaRS in plant male development, one male sterile mutant called osers1 (oryza sativa glutamyl-trna synthetase 1) was isolated from a mutant library and subjected to functional analysis. Although osers1 shows normal vegetative development, approximate 50% of osers1 anthers displayed fused lobes, disarranged anther wall layers and increased number of germinal cells. The expression level of mitotic marker genes H4 and CDKB2;1 was also shown to be increased in osers1 anthers, suggesting the uncontrolled cell division during the early anther development in the mutant. These observations suggest that OsERS1 is required for early anther cell division and organisation. To identify the OsERS1 gene, map-based cloning was carried out, and an insertion of one nucleotide was discovered in the fourth exon of the gene LOC_Os10g22380, leading to a truncated protein. Complementation experiments using the wild-type genomic OsERS1 confirmed that LOC_Os10g22380 is OsERS1. OsERS1 encodes a putative glutamyl-tRNA synthetase (GluRS), containing an N terminal GST_C domain and an aminoacylation domain. Subcellular localisation analysis using GFP-OsERS1 fusion showed that OsERS1 localises in the cytosol and mitochondria. Expression analysis using RT-qPCR and in situ hybridisation revealed that OsERS1 is preferentially expressed before meiosis in anther L2- derived (L2-d) cells that undergo active cell division for the formation of anther cell layers, which is consistent with the defects of osers1 anthers. To biochemically characterise OsERS1, in vitro enzyme activity analysis was conducted. The results demonstrated that OsERS1 is able to catalyse glutamyl- tRNA synthesis using ATP-pyrophosphate (PPi) exchange and glutamylation assays. It was verified that Arg209, Glu213, His219, His222, Ser442, and Lys443 amino acid residues from three key motifs RFAPE, HIGH, LLSKR are critical for OsERS1’s function. Further, the GST_C domain mediated interaction between OsERS1 and its putative cofactor rice Aminoacyl-tRNA Synthetase Cofactor (OsARC) was verified by yeast-2-hybrid (Y2H), pull-down and bimolecular fluorescent complementation (BiFC). The biological significance is that the protein complex has a higher enzymatic activity compared with that of OsERS1 alone, or truncated OsERS1 without interacting domain. To investigate how OsERS1 affects the early anther development, non-targeted metabolomics analysis was conducted for wild-type and osers1 anthers. A total of 297 metabolites with known structures were identified, including amino acids and their derivatives, carbohydrates, lipids, cofactors, and nucleotides and secondary metabolites. These metabolites cover most of the central metabolic pathways. Notably, the abundance of glutamate and its derivatives, intermediates of TCA cycle and glycolysis chemicals differed between osers1 and the wild type. Expression changes of metabolic genes were consistent with the metabolites changes. An increased level of hydrogen peroxide (H2O2) was observed in the mutant anthers, indicating that the defects of osers1 could be caused by the altered redox status. To test this hypothesis, H2O2 was injected into wild-type anthers and it was observed that the treated plants showed disarranged anther cell wall layers and L2-d cells over-proliferation, mimicking that of osers1 anthers. Collectively, these findings provide evidence of aaRSs-mediated metabolism modulating the development of the male organ in plants, likely through affecting metabolic homeostasis and redox status. To investigate whether the function of GluRS is conserved in cereal crops, three homologs of OsERS1 were identified in barley: HvERS1, HvERS2 and HvERS3. HvERS1 and HvERS2 have high similarity with OsERS1 in catalytic motifs, preferential expression in reproductive tissues, and putative tertiary structure. To verify the possible roles of HvERS1 and HvERS2 in barley anther development, CRISPR/Cas9 vectors were designed and transformed into a barley cultivar with the objective of creating the knockout mutants by targeted gene editing. Although creating knockout mutants came to a failure, we contributed useful experiences for future gene editing applications in barley. In summary, OsERS1 was shown to have roles in modulating metabolism, redox status therefore anther cell behaviour and organ development. With OsERS1 as an entry point, the aminoacyl-tRNA synthetase family was investigated and three barley homologs were discovered. This research add new understanding of how physiological homeostasis affects early anther formation and extended noncanonical roles of housekeeping proteins.