Integrative Identification of Chloroplast Metabolism-Related RETICULATA-RELATED Genes in Soybean

Abstract

As a globally important leguminous crop, soybean (Glycine max L.) serves as a vital source of edible oils and proteins for humans and livestock. Oils in leaves can help crops combat fungal infections, adapt to temperature changes via fatty acid modulation, and support resource recycling during leaf senescence. However, accumulating oils in leaves is a fundamental challenge due to the need to balance the inherently competing photosynthesis and fatty acid biosynthesis processes within chloroplasts. RETICULATA-RELATED (RER), known to regulate chloroplast function and plastid metabolism in Arabidopsis, plays an essential role in leaf development. Here, 14 non-redundant GmRER genes were identified in soybean and phylogenetically classified into four subclades. Most Arabidopsis RER genes were evolutionarily preserved as gene duplicates in soybean, except for GmRER5 and GmRER6. RNA secondary structures spanning the coding sequences (CDSs), the 5'- and 3'- untranslated regions (UTRs) of GmRERs, displayed exceptional structural plasticity in CDSs, while exhibiting limited conservation in UTRs. In contrast, protein structures retained conserved folds, underscoring evolutionary constraints on functional domains despite transcriptional plasticity. Notably, GmRER4a and GmRER4b represented an exceptional case of high similarity in both protein and RNA structures. Expression profiling across fourteen tissues and three abiotic stress conditions revealed a dynamic shift in expression levels between leaf-predominant and root-enriched GmRER paralogs after stress treatments. A comparative transcriptome analysis of six soybean landraces further revealed transcriptional polymorphism in the GmRER family, which was associated with the expression patterns of lipid biosynthesis regulators. Our comprehensive characterization of GmRERs may offer potential targets for soybean breeding optimization in overall plant oil production.

Keywords: RETICULATA-RELATED genes; RNA/protein structural analysis; chloroplast development; soybean; transcriptomics analysis.

Figure 1

Phylogenetic analysis and genomic distribution of RER family genes in soybean and related species. (a) Rooted maximum-likelihood (ML) phylogenetic tree of RER proteins from six species: Oryza sativa (rice), Gossypium hirsutum (cotton), Zea mays (maize), Triticum aestivum (wheat), Arabidopsis thaliana (Arabidopsis), and Glycine max (soybean). Numbers at nodes indicate percentage of replicate trees (500 replicates) in which associated taxa are clustered together, representing branch support confidence. Soybean names are labeled as GmRERs and Arabidopsis names as AtRERs, while identifiers (IDs) from rice, cotton, maize, and wheat are retained as annotated in their respective genomes. Four major subclades containing GmRERs and AtRERs were represented in colors. (b) Chromosomal distribution and regional duplication of 14 GmRER genes in Glycine max. Locations of 14 GmRER genes on soybean chromosomes (07, 09, 11, 12, 13, 16, 18) are shown according to Wm82.a4.v1 of soybean genome annotation. Color bar represents gene density distribution on chromosomes.

Figure 2

Gene structure and motif distribution of GmRERs. (a) The gene structures of GmRERs were described using the GSDS tool. The yellow rectangles represent the coding sequences (CDS), the blue rectangles indicate the untranslated regions (UTRs), and the straight lines denote the introns. (b) The motif distributions within GmRER proteins were identified using the MEME tool. The GmRER proteins are drawn to scale. The position of each colored block indicates the location of a motif with a matching sequence.

Figure 3

The structural similarity of GmRER mRNA and proteins. (a–c) The RNA secondary structure similarity of GmRER mRNA in the 5′ UTR (a), CDS (b), and 3′ UTR (c) regions. The values represent similarity scores calculated by RNAForester, with higher scores indicating greater similarity (purple) and lower scores indicating lower similarity (green). (d) The protein structure similarity of GmRERs, represented by log(RMSD) values calculated using PyMOL. Lower log(RMSD) values indicate higher similarity (blue), while higher values indicate lower similarity (red). Structure similarities which cannot be calculated are labeled in gray, and below the heatmaps are examples of some of the most conserved structure pairs.

Figure 4

Protein–protein interactions of GmRERs and GO analysis of interacting proteins. (a) Protein–protein interaction networks of GmRERs were constructed based on STRING predictions. The nodes of GmRER proteins are labeled in red, and the nodes of other interacting proteins are labeled in yellow, blue, and green, according to the clusters they belong to. Proteins with known or predicted structures are specified in the nodes. Edge width and color represent the scores for edge confidence. (b) The lollipop plot displays the GO analysis results of GmRER interacting proteins. The x-axis represents the −log(FDR) values, showing the significance level, and the y-axis shows the significantly enriched GO terms.

Figure 5

Expression patterns of GmRERs in various tissues, developmental stages, and stress conditions. (a) Expression of GmRERs in cotyledons, roots, stems, leaves, flowers, pods, and seeds at different developmental stages (PRJNA238493). (b) Expression of GmRERs in leaves and roots under salt stress at six time points (0, 1, 2, 4, 24, and 48 h) with three biological replicates (PRJNA246058). (c) Expression of GmRERs in leaves and roots under drought and submergence stress at multiple time points (drought stress: 0d, 5d, 6d, recovery; submergence stress: 0d, 1d, 2d, 3d, recovery) with three biological replicates (PRJNA574626). In heatmap, red indicates high expression, and blue indicates low expression.

Figure 6

Statistical analysis of Cis-acting elements in the promoter regions of GmRERs. Cis-elements are categorized into three main classes: abiotic and biotic stress, phytohormone-responsiveness, and plant growth and development. The numbers represent the occurrence of each cis-element in the promoter regions of the genes. The deeper the red color, the higher the frequency of occurrence.

Figure 7

Expression levels of GmRER family genes in six northern spring soybean landraces and their seedling phenotypes. (a) The seedling phenotypes of the six landraces at the soybean V1 stage. The representative images of accessions labeled L1 to L6 correspond to the following landraces (left to right): Xiaoli Moshidou (L1), Baimaoshuang (L2), Tiejia Silihuang (L3), Heidadou (L4), Zhouye (L5), and Aqi Manjinhuang (L6). Scale bars = 1.2 cm. (b) The expression patterns of five known oil biosynthesis-associated genes (GmSWEET10b, GmB1, GmFAD2-2, GmFAD3-2a, and GmDof4) across the six landraces. The data represent Log2-transformed normalized counts derived from DESeq2. (c) Differential expression analysis of GmRER family genes in the six landraces. For panels (b,c), statistical significance was determined by pairwise comparisons using DESeq2’s Wald test. Lowercase letters above the boxplots indicate significant differences: landraces sharing the same letter are not significantly different, while distinct letters denote statistically divergent expression (p < 0.05).