Ribosomal heterogeneity revealed by a spatial transcriptomics in the water fern Ceratopteris richardii

Background and Aims: Ferns are providing insight into fundamental biological processes, and spatial transcriptomics is a potent tool in unraveling complexes of plant development. This research is targeted at leaf development in the water fern Ceratopteris richardii with the use of spatial transcriptomics. Methods: A shoot tip of sporophytes under vegetative growth was investigated using spatial transcriptomics, which conducted spot clustering, marker gene identification, and functional enrichment. Ribosomal protein genes were queried using BLAST program and categorized based on multi-alignment produced by Clustalw2 program. RNA in situ tissue hybridization was performed on three ribosomal protein genes. Key Results: Leaves were mutually different at the level of transcriptome under various developmental stages. There were unexpected complexes in tissue structure for stems. A few transcription factors were identified characteristic of leaf development. Activities associated with translation were very different between tissues, and ribosomal protein genes were heterogeneous at the level of transcription. Ribosomal protein genes were specifically expressed in the shoot apical meristem, leaf primordia, and vasculature of various organs. Conclusions: Leaves are mutually different under various developmental stages in C. richardii, with contribution of transcription factors. Stems are unexpectedly complex in tissue structure. Organs/tissues are very different in translation activity from each other, partly due to transcription heterogeneity of ribosomal protein genes. Ribosomal protein genes are specifically expressed in the shoot apical meristem, leaf primordia, and vasculature. The water fern Ceratopteris richardii was employed, one sporophyte under vegetative growth with three fully expanded leaves was dissected, and the central shoot tip was retained and directly immersed into the optimal cutting temperature compound (OCT, Sakura Finetek Europe B.V.) under dry ice. The specimen was sectioned with a thickness of 20 μm, and the anchored section was fixed on slides using 100% methanol. Topuidine Blue staining was applied for histological examination. RNA was trapped by poly(T)VN to the spot array, namely BMKMANU S1000 platform (BioMarker, Beijing) having a density of more than two million spots with a diameter less than 2.5 μm and 5 μm centre-to-centre distance within 6.8 mm2 area. Double-strand cDNA was synthesized to incorporate the spatial barcode and the unique molecular index (UMI), and furthermore the sample index. The fragmented DNA was obtained via enzymatic digestion, and it was sequenced using NovaSeq PE150 (Illumina). Sequencing reads were mapped to the genome assembly with the accession number of GCA_020310875.1. BSTMatrix (BioMarker) was used to construct the spot-expression matrix under a total of eight levels. The package Seurat v5 provided tools in normalizing data, identifying clusters, and determining marker genes. The threshold for the marker gene was |log2FC| ≥ 0.5 and false discovery rate < 0.05. Spot clustering was visualized using BSTViewer V1.42 (BioMarker). Genes were functionally annotated using multiple platforms such as NR, Pfam, Swiss-Prot, and TrEMBL with the help of the program BLAST, and the subsequent enrichment analysis was conducted using clusterProfiler v3.18.1.