Pangenome-wide identification of chloroplast RNA splicing and ribosome maturation (CRM) genes in eight Pyrus genomes indicated their involvement in multiple stresses Authors: Asma Khalil, Muhammad Sadaqat, Kinza Fatima, Kahkashan Perveen, Jayanthi Barasarathi, Muhammad Tahir ul Qamar, Shoaib Munir Published: 2026 ABSTRACT Chloroplast RNA splicing and ribosome maturation (CRM) domain proteins are central to chloroplast intron splicing, rRNA processing, and plant stress responses. While extensively studied in model plants such as Arabidopsis, rice, and maize, the CRM gene family has not been characterized in Pyrus, a major fruit crop of global commercial agricultural productivity . This study presents the first pangenome-wide analysis of CRM genes across 8 Pyrus genomes, aiming to unravel their evolutionary patterns, structural diversity, and potential functional roles in stress adaptation. In this study, we identified 25 PyCRM pan-genes, comprising 9 core, 15 variable, and 1 species-specific members, reflecting gene presence-absence variation within the family. Phylogenetic analysis grouped these proteins into four conserved subfamilies, with the CRS1 subfamily being the largest. Gene structure and motif profiling revealed subfamily-specific conservation alongside distinct motif variations, suggesting functional divergence. Ka/Ks calculation revealed that most of the PyCRM genes evolved under purifying selection, while a few members experienced positive selection, highlighting functional diversification within the CRM gene family. Promoter analysis identified multiple cis-regulatory elements linked to the light response, hormone signaling, stress responses, and developmental processes. Functional enrichment highlighted roles in RNA binding and mRNA processing. Members of the same family share similar motifs and conserved protein 3D structures. Furthermore, expression profiling under cold, drought, and disease revealed that most PyCRM genes were downregulated under stress conditions, while a few members exhibited stress-specific upregulation, suggesting functional diversification within the gene family. These findings lay the foundation for future investigations into CRM-mediated stress tolerance and crop improvement.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-48265-0. Introduction Chloroplasts are essential organelles in plants and carry out a series of metabolic processes, drive photosynthesis, and synthesize compounds such as tetrapyrroles, terpenoids, lipids, amino acids, hormones, and stress-responsive metabolites1. Chloroplasts contain about 100–150 genes in their circular genome that encode messenger RNAs (mRNAs), ribosomal RNAs (rRNAs), and transfer RNAs (tRNAs). RNA-binding proteins have recently been recognized as the primary factors that mediate post-transcriptional steps in chloroplasts2, 3. According to the conservation of their primary and secondary structures and the RNA splicing mechanism, introns in plants can be categorized into groups I and II, with group II introns further subdivided into subgroups IIA and IIB. Among these, the chloroplast RNA splicing and ribosome maturation (CRM) domain proteins perform a crucial role in plant growth4. The CRM domain-containing proteins may carry numerous copies of CRM domains that participate in the intron-splicing mechanisms and affect rRNA processing in chloroplasts. However, the RNA recognition capabilities of CRM domain-containing proteins are associated with the conserved GxxG sequences within the loop of the CRM domains5. The CRM domain gene family of rice (Oryza sativa) comprises 14 homologous genes that target different organelles, such as chloroplasts and mitochondria6. CRM domain occurs in four characterized protein subfamilies that are associated with catalytic RNA and were named to indicate the functions of these proteins: the CRS1 subfamily, the CAF subfamily, subfamily 3, and subfamily 4. The CRS1–YhbY domain is represented as an unconnected protein in Archaea and Bacteria, and as a member of a family of single- and multiple-domain proteins in plants7. The first three subfamilies have been reported to perform regulatory functions in group I and II intron splicing, whereas subfamily 4 is associated with the assembly of large ribosomal subunits. ZmCRS1, the first CRM gene, was identified and cloned in maize8. The ZmCRS1 protein comprises three CRM domains that can directly attach to the atpF intron in chloroplasts and control its splicing. The genomes of rice and Arabidopsis contain 16 and 14 CRM domain-containing proteins, respectively. It has been reported that, in low temperature and salt stress conditions, Arabidopsis CRM-containing chloroplast-localized CFM4 (involved in 16 S and 23 S rRNA processing) acts as a positive effector in seed germination and seedling development9. Moreover, rice OsCAF1 and OsCFM2 are important for the chloroplast intron splicing as the orthologues of AtCAF1 and AtCAF210. Furthermore, the mitochondria-localized Arabidopsis mCSF1 and CFM9 participate in multiple mitochondrial introns splicing, which can affect seed development and seedling growth, respectively. Mutations in CRM domain–containing proteins have been reported to cause albino seedling phenotypes, defective chloroplast development, aborted seed production, and increased sensitivity to abiotic stresses11. Pyrus is a commercially valuable fruit crop, with a yearly worldwide production of ~ 24 million tons, and is cultivated in temperate regions worldwide12. These are excellent sources of nutritional components and have an exuberant taste. Pears, like the other pip fruit species, such as apples and quinces, are part of the sub-family Maloideae in the Rosaceae family, sharing a basic chromosome number of x = 17, which depicts a polyploidy origin. The genus Pyrus includes at least 22 recognized species, with more than 5000 accessions maintained worldwide13. Recent advances in plant genomics have led to pan-genome approaches for capturing species-level genetic diversity and understanding evolutionary processes within the genus. Li et al., (2024) proposed the first pangenome graph for pear to understand the structural variation and analyze the trait-related candidate alleles across Pyrus species. Moreover, comprehensive pan-genomic analyses have successfully characterized key gene families, such as CDPKs, across multiple pear genomes and demonstrated their functional roles in disease resistance through integrated bioinformatic and transcriptomic strategies14. In the same context, our study identified 16 CRM genes from 8 Pyrus species, comprehensively identifying their underlying evolutionary mechanisms, protein characteristics, subcellular localization, and expression patterns. Furthermore, an ML approach was employed to identify the multi-stress-responsive PyCRM genes. Hence, this study substantially broadens our current knowledge of the CRM gene family in Pyrus, demonstrating its role in conferring resistance against biotic and abiotic stresses. Our findings will lay the groundwork for future research on the functional roles of CRMs in Pyrus species. Materials and methods Genome completeness assessment and identification of PyCRMs The pear genome data were obtained from the pear genomics database (PGDB, http://Pyrusgdb.sdau.edu.cn/)15. The cultivars included d’Anjou (Py-An, P. communis)16, Bartlett v2.0 (Py-Ba, P. communis)17, Dangshansuli v1.1 (Py-Da, P. bretschneideri)18, Cuiguan v1.0 (Py-Cu, P. pyrifolia)19, Nijisseiki (Py-Ni, P. pyrifolia)20, Yunhong No.1 (Py-Yu, P. pyrifolia)21, Shanxi Duli (Py-Sh, P. betulifolia)22, and Zhongai 1 (Py-Zh, P. ussuriensis × P. communis)23. Genome assembly quality of these pear genomes was evaluated using BUSCO v5.8.0, employing the proteomes derived from the downloaded genome assemblies. Arabidopsis CRM protein sequences were retrieved from the TAIR genome database (https://www.arabidopsis.org/)24. The protein sequences of Pyrus species were utilized as subject sequences for analysis through the BLAST+ command line tool. A BLASTp search was conducted using AtCRM protein sequences as queries against 8 Pyrus proteomes. The identified matches were further filtered by removing replicates and isoforms. Furthermore, the pinpointed proteins were examined to validate the presence of the RNA-binding CRS1/YhbY (CRM) (PF01985) domain. For this objective, the NCBI conserved domain database (CDD)25 and InterProScan (https://www.ebi.ac.uk/interpro/)26 were utilized. To classify CRM gene family members at the pan-gene level, including core, variable, and unique, orthologous gene clustering was performed through the OrthoFinder v2.5.5. The pan-gene classification was illustrated in the form of presence or absence through the pheatmap27 function in R . Prediction of physicochemical properties and subcellular localization of PyCRMs Information on several physicochemical characteristics, including amino acid (aa) length, molecular weight (MW), their isoelectric point (pI), instability index (II), aliphatic index (AI), and the grand average of hydropathicity (GRAVY), was forecasted by using the ExPASy ProtParam tool (https://web.expasy.org/protparam/)26. The WoLF PSORT28 was utilized to identify the subcellular localization of CRM proteins. Comparative sequence alignment and evolutionary relationship analysis For evolutionary analysis, the PyCRMs along with their homologous sequences from other plant including Arabidopsis thaliana and Oryza sativa, were aligned using MAFFT v7.490, and a phylogenetic tree was constructed using IQTREE v2. The Interactive Tree of Life (iTOL) v7 (https://itol.embl.de/) was utilized for final graphics and editing. Synteny and Ka/Ks calculation of PyCRMs To investigate the collinearity of the CRM gene family across 8 Pyrus genomes, a synteny analysis map was constructed. Protein and corresponding gene annotation files of each genome were subjected to comparative analysis. An all-against-all BLASTP search was conducted with an E-value cutoff of 1e−5 to identify homologous gene pairs29. The resulting BLAST outputs, together with genome annotation files, were subjected to synteny detection through the MCScanX30. The syntenic relationships were subsequently visualized using TBtools to generate inter-genomic synteny maps. The CDS sequences from PyCRMs were aligned using MACSE v2,31 and the frequency of synonymous (Ks) and non-synonymous substitutions (Ka) phenomena was determined using the form of Ka/Ks ratio through KaKs Calculator v2 by using Nei–Gojobori method32. Genomic structure, conserved motif, and cis-acting element analysis The MEME tool (https://meme-suite.org/meme/tools/meme)33 was used to determine the motifs conserved among the PyCRM sequences. For every sequence, the number of conserved motifs was set at 20. Gene structures of PyCRMs were constructed using CDS coordinates present in the annotation file. After that, the conserved motifs pattern and gene structures were visualized using TBtools34. The upstream 2000 bp transcriptional control regions of the PyCRM genes were obtained through TBtools and submitted to the PlantCare database35 to study cis-regulatory elements. PPI, GO, and 3D prediction analysis Amino acid sequences of PyCRMs were submitted to the STRING (https://string-db.org/) database to examine the interactions (PPIs)36 among Pyrus and other CRM proteins. The threshold level was set at 0.4, and the 3rd level of connection was used. After that, the Cytoscape v3.10.4 was employed to visualize the PPI network37. Additionally, the PANNZER2 database (http://ekhidna2.biocenter.helsinki.fi/sanspanz/) was employed to identify the GO Constituents of enrichment profiling, such as biological processes (BPs), cellular components (CCs), and molecular functions (MFs)38. The proper functioning of a protein relies on its three-dimensional (3D) structure. Based on the expression analysis, the 3D structures of 25 PyCRM protein domains were predicted using AlphaFold339. The evaluation of predicted structures was examined through the SAVES server and MolProbity40. Lastly, the UCSF ChimeraX41 software was utilized to visualize these PyCRM structures in a rainbow layout. Expression analysis of PyCRMs under multiple stresses To investigate the expression patterns of PyCRM genes under diverse abiotic and biotic stress conditions, publicly available RNA-Seq datasets were retrieved from the NCBI SRA database (https://www.ncbi.nlm.nih.gov/sra)42. The datasets represented different tissues, cultivars, and treatments. Specifically, fruit peel tissues of ‘Bartlett’ (Py-Ba) and ‘D’Anjou’ (Py-An) subjected to cold conditioning (PRJNA527142) for 50 and 100 days, leaf blade tissues of ‘Dangshansuli’ (Py-Da) exposed to seasonal drought stress under field conditions (PRJNA655255), leaf tissues of the susceptible cultivar ‘Cuiguan’ (Py-Cu) infected with Colletotrichum fructicola (anthracnose disease) (PRJNA1078101), and callus tissues of ‘Shanxi’ (Py-Sh) treated under low-temperature stress (PRJNA1276717) were included in this study. Raw sequencing reads were subjected to quality assessment using FastQC43 to evaluate base quality, GC content, and adapter contamination. Considering the pan-genome nature of Pyrus and to reduce mapping bias, clean paired-end reads from each cultivar were aligned to their corresponding reference genome and annotation. Genome sequences and gene structure annotation (GFF) files were obtained from the Pyrus Genome Database (PGDB;http://pyrusgdb.sdau.edu.cn/). The Pyrus genome indexes were constructed using HISAT2, and the high-quality paired clean reads were subsequently mapped onto the indexed reference genome with default parameters. Transcript assembly and expression quantification were performed using StringTie44, and gene expression levels were calculated as fragments per kilobase of transcript per million mapped reads (FPKM). For heatmap visualization, FPKM values were log-transformed to normalize expression differences and better illustrate fold changes among samples. Results Pangenome-wide identification of CRM genes from 8 Pyrus species Among the 8 Pyrus genomes, Py-Yu revealed the highest genome completeness (95.8%), indicating a well-assembled genome with minimal missing content. Furthermore, Py-Cu and Py-Ni showed the genome completeness (95.6%), followed by Py-Ni (95.6%), Py-Yu (95.4%), Py-Sh (93.5%), Py-Da (93.9%), Py-An (91.5%), Py-Zh (89.6%), and Py-Ba (80.4%). The proportion of single-copy BUSCOs (S) ranged from 48.1% (Py-An) to 61.4% (Py-Yu). In comparison, duplicated BUSCOs (D) varied between 34.2% and 48.1%, revealing the presence of gene duplication events due to whole-genome duplication and evolutionary expansion. The fraction of fragmented BUSCOs (F) remained low across all genomes (1.2–4.8%), and the percentage of missing BUSCOs (M) was also limited (2.7–11.7%), further supporting the overall good quality of the assemblies (Fig. 1A; Table S1). The CRM gene family members were identified across 8 different Pyrus genomes, including wild species and interspecific hybrids. Specifically, 28 members were identified in Py-An, 19 in Py-Ba, 24 in Py-Cu, 16 in Py-Da, 22 in Py-Ni, 25 in Py-Sh, 24 in Py-Yu, and 18 in Py-Zh. After clustering homologous genes, 25 PyCRM family members were identified. These 25 PyCRMs include 9 core, 15 variable, and 1 unique gene (Fig. 1B; Table S2). These findings highlight the substantial genomic divergence at specific CRM loci among Pyrus species, suggesting that the differential gain or loss of these genes may contribute to variation in environmental adaptation and abiotic or biotic stress tolerance. For downstream analysis, PyCRM1-PyCRM24 were characterized as anchors from Py-Yu because it has highest completeness as compared to other genomes (Fig. 1A), and they also have most of the sequences present in all orthogroup clusters. PyCRM25 was uniquely present in the Py-Sh genome, so it was taken as an anchor to represent it. Fig. 1(A) Visual depiction of genome quality assessment across 8 Pyrus genomes. (B) Pangenome-wide distribution of CRM genes across 8 Pyrus genomes. Orange color indicates the presence of a gene, while blue indicates the absence of a gene. (A) Visual depiction of genome quality assessment across 8 Pyrus genomes. (B) Pangenome-wide distribution of CRM genes across 8 Pyrus genomes. Orange color indicates the presence of a gene, while blue indicates the absence of a gene. Physicochemical characteristics of PyCRMs The PyCRM proteins were in a range of 101–1118 aa, and MW 10.96–123.95 kDa. The isoelectric point (pI) indicated that the maximum number of CRM members in the pear are basic in nature. The instability index (II) predicted that only one protein, PyCRM8, is stable, whereas all other proteins showed instability index values greater than 40, suggesting that most of the proteins are likely to be unstable under in vitro conditions. Aliphatic index (AI) indicates that most of the PyCRMs are thermally stable, while the negative GRAVY values suggested that these proteins are hydrophilic. According to the Subcellular localization analysis, most of the PyCRM proteins were located in the chloroplast, followed by cytoplasm, mitochondria, and nuclear (Table 1). Table 1Prediction of physicochemical characteristics of the CRM gene family in Pyrus.Gene nameProtein length (aa)MW (kDa)Isoelectric point (pI)Instability index (II)Aliphatic index (AI)GRAVYSubcellular localization PyCRM1 77485.148.8750.2076.96− 0.54Chloroplast PyCRM2 41047.009.4146.3674.59− 0.79Mitochondrial PyCRM3 36841.948.8851.1681.52− 0.60Mitochondrial PyCRM4 41547.599.3552.3777.08− 0.77Chloroplast PyCRM5 45552.468.9151.2575.41− 0.84Chloroplast PyCRM6 83494.295.8053.6683.14− 0.67Chloroplast PyCRM7 67377.177.2155.5357.99− 1.12Chloroplast PyCRM8 10110.969.7829.1296.53− 0.07Cytoplasm PyCRM9 80791.459.4843.6284.47− 0.52Cytoplasm PyCRM10 35140.059.7658.5273.87− 0.65Mitochondrial PyCRM11 29032.019.4764.2973.00− 0.58Chloroplast PyCRM12 31736.675.5052.2681.80− 0.73Cytoplasm PyCRM13 24828.488.1049.1775.40− 0.76Chloroplast PyCRM14 56062.587.1851.3978.68− 0.55Chloroplast PyCRM15 40845.758.7649.3576.89− 0.61Chloroplast PyCRM16 23026.498.9544.4692.04− 0.54Cytoplasm PyCRM17 16518.784.7854.3975.09− 0.89Mitochondrial PyCRM18 19322.8410.2751.0170.73− 0.83Chloroplast PyCRM19 82493.026.2455.6482.50− 0.65Chloroplast PyCRM20 78887.148.7846.4072.37− 0.62Nuclear PyCRM21 983111.188.7647.1376.82− 0.70Chloroplast PyCRM22 86798.498.7451.3479.24− 0.64Chloroplast PyCRM23 67277.147.6153.9358.65− 1.10Nuclear PyCRM24 1118123.959.3245.5080.13− 0.61Chloroplast PyCRM25 10311.849.7345.0576.70− 0.70Cytoplasm Table 1Prediction of physicochemical characteristics of the CRM gene family in Pyrus.Gene nameProtein length (aa)MW (kDa)Isoelectric point (pI)Instability index (II)Aliphatic index (AI)GRAVYSubcellular localization PyCRM1 77485.148.8750.2076.96− 0.54Chloroplast PyCRM2 41047.009.4146.3674.59− 0.79Mitochondrial PyCRM3 36841.948.8851.1681.52− 0.60Mitochondrial PyCRM4 41547.599.3552.3777.08− 0.77Chloroplast PyCRM5 45552.468.9151.2575.41− 0.84Chloroplast PyCRM6 83494.295.8053.6683.14− 0.67Chloroplast PyCRM7 67377.177.2155.5357.99− 1.12Chloroplast PyCRM8 10110.969.7829.1296.53− 0.07Cytoplasm PyCRM9 80791.459.4843.6284.47− 0.52Cytoplasm PyCRM10 35140.059.7658.5273.87− 0.65Mitochondrial PyCRM11 29032.019.4764.2973.00− 0.58Chloroplast PyCRM12 31736.675.5052.2681.80− 0.73Cytoplasm PyCRM13 24828.488.1049.1775.40− 0.76Chloroplast PyCRM14 56062.587.1851.3978.68− 0.55Chloroplast PyCRM15 40845.758.7649.3576.89− 0.61Chloroplast PyCRM16 23026.498.9544.4692.04− 0.54Cytoplasm PyCRM17 16518.784.7854.3975.09− 0.89Mitochondrial PyCRM18 19322.8410.2751.0170.73− 0.83Chloroplast PyCRM19 82493.026.2455.6482.50− 0.65Chloroplast PyCRM20 78887.148.7846.4072.37− 0.62Nuclear PyCRM21 983111.188.7647.1376.82− 0.70Chloroplast PyCRM22 86798.498.7451.3479.24− 0.64Chloroplast PyCRM23 67277.147.6153.9358.65− 1.10Nuclear PyCRM24 1118123.959.3245.5080.13− 0.61Chloroplast PyCRM25 10311.849.7345.0576.70− 0.70Cytoplasm Prediction of physicochemical characteristics of the CRM gene family in Pyrus. Phylogenetic analysis of Pyrus CRM proteins To investigate the evolutionary relationships among CRM proteins, a phylogenetic tree was constructed using PyCRMs, AtCRMs, and OsCRMs. The resulting tree grouped these proteins into four distinct subfamilies, each containing a different number of members from the three species. Among them, the CRS1 subfamily contained the highest number of CRM members, followed by subfamily 3, the CAF1 subfamily, and subfamily 4. Each of the four subfamilies included at least one core PyCRM protein (Fig. 2). Notably, PyCRM25 was uniquely positioned within subfamily 3. This clustering pattern suggests functional diversification within the CRM protein family and indicates that these genes may play important roles in regulating stress tolerance mechanisms in various environmental stresses. Fig. 2Phylogenetic relationship among PyCRMs in pear genotypes and previously reported CRMs from A. thaliana and O. sativa. Different branch colors indicate the different subfamilies, while the bootstrap values are indicated as circles on the corresponding branches. Diamond, star, and square shapes represent the core, variable, and unique genes, respectively. Phylogenetic relationship among PyCRMs in pear genotypes and previously reported CRMs from A. thaliana and O. sativa. Different branch colors indicate the different subfamilies, while the bootstrap values are indicated as circles on the corresponding branches. Diamond, star, and square shapes represent the core, variable, and unique genes, respectively. Collinearity analysis of PyCRMs Synteny analysis revealed a high level of conservation of PyCRM genes across the 8 Pyrus species (Py-Yu, Py-Cu, Py-Ni, Py-Sh, Py-An, Py-Zh, Py-Da, and Py-Ba). We found dense collinear links among Py-Yu, Py-Cu, Py-Ni, Py-Sh, Py-An, and Py-Zh, suggesting the conservation of gene function and potential for similar biological processes across these species. Whereas Py-Da and Py-Ba exhibited relatively fewer collinear connections, demonstrating an enhancement of the genomic divergence. Moreover, collinear blocks were unevenly distributed across 17 chromosomes, reflecting heterogeneity in CRM gene conservation among the analyzed genomes. The extensive one-to-one and one-to-many collinearity relationships suggest that both segmental duplications and whole-genome duplication events have largely driven the expansion of the CRM gene family (Fig. 3). Fig. 3Synteny (collinearity) analysis of PyCRM genes across different pear genotypes. Chr1–Chr17 indicate the 17 chromosomes, and “Un” represents unassembled scaffolds. Synteny (collinearity) analysis of PyCRM genes across different pear genotypes. Chr1–Chr17 indicate the 17 chromosomes, and “Un” represents unassembled scaffolds. Selection pressure analysis of PyCRM genes To gain insights into the evolutionary rate across members of the Pyrus PyCRM pan-gene family revealed that different members have been subjected to varying selective forces throughout evolution. The results demonstrated a broad range of selective pressures among different PyCRM genes (Fig. 4). Most PyCRM genes had Ka/Ks ratios below or near 1, indicating they evolved under strong purifying selection. Conversely, a subset of genes, including PyCRM5, PyCRM9, PyCRM11, PyCRM14, PyCRM15, and PyCRM17, displayed Ka/Ks ratios greater than 1, suggesting positive selection acting on these genes. The distribution of Ka/Ks values across the gene family emphasizes the diversity of evolutionary pressures that have influenced the different CRM members in the Pyrus. Fig. 4Ka/Ks values for CRMs in Pyrus genomes. The Ka/Ks ratio represents the non-synonymous (Ka) to synonymous (Ks) substitution rates, reflecting evolutionary selection pressure on genes. Ka/Ks values for CRMs in Pyrus genomes. The Ka/Ks ratio represents the non-synonymous (Ka) to synonymous (Ks) substitution rates, reflecting evolutionary selection pressure on genes. Motif profiling and structural organization of PyCRM genes The motifs are concise functional sites of enzymes aiding the process of protein structural arrangements. A total of 20 distinct motifs were identified. In the PyCRM proteins, motif 1 was highly conserved in all members of the CRM subfamilies. Members of same family has similar conserved motifs (Fig. 5B). Within the CRS1 subfamily, motifs 2 and 8 are conserved across all members. Motifs 4 and 6 are also conserved in nearly all members, with the exception of PyCRM8 and PyCRM16. Similarly, motifs 1, 3, 5, 7, 15, and 17 are conserved throughout the CRS1 family, except in PyCRM8, PyCRM15, and PyCRM16. In the CAF1 subfamily, motifs 3, 10, and 12 are conserved across all members. Motifs 1, 2, 4, 18, and 20 are also conserved throughout the subfamily, with the exception of PyCRM13. In subfamily 3, motif 9 is conserved in all members. Motifs 7 and 11 are likewise conserved, except in PyCRM17. Additionally, motifs 1 and 2 are conserved in all members except PyCRM17, PyCRM18, and PyCRM25. Subfamily 4 contains only a single member, PyCRM11, which possesses conserved motifs 1, 8, and 16, suggesting that these motifs might have important functions in these families. The changes in motif organizations signified the functional diversity of the associated proteins. Gene structure analysis provides important insights into the evolutionary diversification and functional conservation of gene families. To investigate the structural organization of CRM genes, the genomic and corresponding CDS sequences from eight Pyrus species were analyzed to determine exon–intron architecture. As illustrated in Fig. 5C, variation in exon number was observed among different subfamilies. Members of the CAF1 subfamily contained 3–6 exons, whereas subfamily 4 has one gene that possess 2 exons. Subfamily III members exhibited a relatively conserved structure with 2–4 exons. In contrast, the CRS1 subfamily displayed greater structural diversity, with exon numbers ranging from 3 to 13. Overall, the observed variation in exon organization among CRM subfamilies suggests differential evolutionary trajectories while maintaining conserved structural frameworks within each subfamily. Fig. 5(A) Phylogenetic relationship (B) Conserved motifs identified through the MEME suite. (C) Gene structure representing the conserved pattern of CDS and UTR. On the right, diamond, star, and square shapes represent the core variable and the unique genes, respectively. Different colored boxes depict different motifs. The green boxes represent UTR, and the yellow boxes represent exons. (A) Phylogenetic relationship (B) Conserved motifs identified through the MEME suite. (C) Gene structure representing the conserved pattern of CDS and UTR. On the right, diamond, star, and square shapes represent the core variable and the unique genes, respectively. Different colored boxes depict different motifs. The green boxes represent UTR, and the yellow boxes represent exons. Cis-regulatory elements in the promoters of Pyrus CRM genes The 2000-bp upstream promoter regions of the PyCRM genes were analyzed to identify putative cis-acting regulatory elements. A diverse array of cis-elements associated with 15 distinct biological processes was detected. These processes were further categorized into four major functional groups: light-responsive, hormone-responsive, stress-responsive, and development-related elements. Among these, light-responsive elements were the most abundant, followed by hormone-responsive, stress-responsive, and development-related elements. The enrichment of light- and hormone-responsive cis-elements in the promoter regions suggests that PyCRM genes may play important roles in integrating environmental signals and hormonal pathways to regulate plant growth, development, and stress adaptation (Fig. 6; Table S3). Fig. 6(A) Phylogenetic relationship (B) The PyCRMs upstream promoter regions comprise cis-regulatory elements. Different colors represent the cis-elements and their position on the upstream region of genes. (A) Phylogenetic relationship (B) The PyCRMs upstream promoter regions comprise cis-regulatory elements. Different colors represent the cis-elements and their position on the upstream region of genes. PPI, GO and 3D structure prediction analysis To predict the function of PyCRM proteins, GO enrichment analysis was performed. According to BP, PyCRMs are involved in Group II intron splicing, mRNA processing, and regulation of DNA-templated transcription shows less enrichment. The CCs showed enrichment in the ribonucleoprotein complex, along with localization in the chloroplast, and mitochondria. The MFs linked with the genes involved in RNA binding, DNA binding, and mRNA binding (Fig. 7A; Table S4). PyCRM proteins were evaluated to predict interactions among them to analyze their functional relationship. The interaction network consisted of 13 Pyrus proteins, namely PyCRM2, PyCRM9, PyCRM10, PyCRM11, PyCRM12, PyCRM14, PyCRM16, PyCRM19, PyCRM21, PyCRM22, PyCRM23, and PyCRM24, interacting with 20 A. thaliana proteins. Among the predicted potentially associated proteins, the PyCRM14 exhibited the highest degree of interaction, while PyCRM24 and THA8 revealed the second and third highest degree of connectivity (Fig. 7B; Table S5). The dense interaction network suggests that PyCRM proteins may play an important role in post-transcriptional regulation, particularly within chloroplasts. Fig. 7(A) Predicted BPs, CCs, and MFs associated with PyCRM proteins. (B) Interactions among PyCRMs and other homologous proteins. The green color shows the PyCRM proteins, and the orange color represents the other associated proteins from different species. (A) Predicted BPs, CCs, and MFs associated with PyCRM proteins. (B) Interactions among PyCRMs and other homologous proteins. The green color shows the PyCRM proteins, and the orange color represents the other associated proteins from different species. All PyCRM proteins 3D structures have Alpha helices ranging from 2 to 16 and beta sheets ranging from 2 to 15, except PyCRM17 and PyCRM25 (Fig. 8). Members of the same subfamily share similar conserved motifs, which contribute to a similar structural organization. The variability in the number of alpha helices and beta sheets suggests that proteins may have undergone structural and functional divergence during evolution. Fig. 83D structures of the three PyCRM proteins predicted by Alphafold3. 3D structures of the three PyCRM proteins predicted by Alphafold3. Gene expression profiling of PyCRM genes Examining transcriptome data from various tissues, stresses, and conditions allowed us to delve deeper into the biological roles played by the CRM gene family in Pyrus. The expression patterns of PyCRM genes in Py-An and Py-Ba were analyzed under cold treatment for 50 days and 100 days using transcriptome data (Fig. 9). In both genotypes, the majority of PyCRM genes were downregulated under cold stress compared with their respective control conditions. In Py-An, PyCRM-An1 and PyCRM-An16 exhibited upregulation at 50 and 100 days, respectively (Fig. 9A; Table S6). In Py-Ba, only PyCRM-Ba14 showed slight upregulation at 50 days cold (Fig. 9B; Table S7). Fig. 9Heatmap illustrates the expression pattern of PyCRMs under cold stress at 50 and 100 days in (A) Py-An and (B) Py-Ba. Heatmap illustrates the expression pattern of PyCRMs under cold stress at 50 and 100 days in (A) Py-An and (B) Py-Ba. In Py-Cu, the majority of PyCRM genes were downregulated under disease conditions, however, PyCRM11-Cu exhibited upregulation in response to pathogen infection (Fig. 10A; Table S8). In Py-Da, PyCRM2-Da, and PyCRM4-Da were upregulated under drought stress (Fig. 10B; Table S9). In contrast, all identified PyCRM genes in Py-Sh were downregulated under the cold conditions (Fig. 10C; Table S10). The predominant downregulation of PyCRM genes under multiple stress conditions indicates that this gene family may participate in stress-responsive regulatory pathways, potentially through transcriptional repression or modulation of RNA processing during stress adaptation in Pyrus. Fig. 10Heatmap illustrates the expression pattern of PyCRMs under different stress conditions including, (A) Py-Cu in anthracnose disease, (B) Py-Da in drought, and (C) Py-Sh in Cold condition. Heatmap illustrates the expression pattern of PyCRMs under different stress conditions including, (A) Py-Cu in anthracnose disease, (B) Py-Da in drought, and (C) Py-Sh in Cold condition. Discussion Distribution of CRM genes in eight Pyrus genomes and other plant species The CRM gene family plays an indispensable role in the growth and development of chloroplasts in green plants45. Structural and biological evidence have revealed that CRM domains function as RNA-binding domains. Furthermore, researchers have identified CRS1, a CRM protein with multiple domains, as a protein that binds RNA in a sequence-specific manner46. In the present work, we identified 25 PyCRM pan-genes, including 28 CRM members in Py-An, 25 in Py-Sh, 18 in Py-Zh, 22 in Py-Ni, 24 in Py-Yu, 24 in Py-Cu, 19 in Py-Ba, and 16 in Py-Da. This diversity suggests that these genes may be involved in several physiological and biochemical processes. Other plant species also exhibit differences in the number of CRM genes, with O. sativa having 14 members47, and A. thaliana having 16 identified members48. Chloroplasts are needed for photosynthesis. When CRM domain mutations severely impair chloroplast function, plants often exhibit slow growth, pale leaves, and may not reach maturity49. In severe situations, where critical CRM proteins are non-functional, the plant may not survive beyond the seedling stage. For instance, the phenotypic analysis showed that changes to the rice gene OsCAF2 caused a chlorotic phenotype during the seeding stage, which killed the plant50. Likewise, a defective gene in the maize ZmELL1 caused seedling lethality51. Subfamily characteristics of CRM proteins The phylogenetic tree showed that Pyrus CRM proteins could be categorized into four subfamilies: CAF1, CRS1, subfamily 3, and subfamily 4. The clade of subfamily CRS1 was the largest, and subfamily 4 was the smallest. All genes within the same subfamily are closely grouped and exhibit shared homology, including those from other species. Similarly, the CRM members of O. sativa and A. thaliana are classified into four subfamilies52. Evolutionary expansion and Ka/Ks of the CRM gene family in Pyrus Synteny analysis shows that the CRM genes in all 8 Pyrus genomes are unevenly distributed on chromosomes. A similar trend was observed in CDPK14. Our synteny analysis of PyCRM genes has shown evolutionary history (Fig. 3) dense conservation among the Asian pears, Py-Yu, Py-Cu, Py-Ni, Py-Sh, and hybrid Py-Zh; whereas few collinear connections are seen with European P. communis cultivars (Py-An, Py-Ba). This pattern supports independent domestications for Asian and European pears and suggests that wild and hybrid species retain ancestral genomic content, in accordance with previous phylogenetic and domestication studies53. Most PyCRM genes exhibited Ka/Ks values close to or below 1, revealing that PyCRM genes underwent purifying selection, and only 6 members, PyCRM5, PyCRM9, PyCRM11, PyCRM14, PyCRM15 and PyCRM17, displayed Ka/Ks ratios > than 1. A similar phenomenon was observed in Gossypium hirsutum, in which 146 of the genes have undergone strong purifying selection, and only three duplicated gene pairs revealed positive selection pressure during the rapid evolution of the gene family. This may be related to the adaptability of pears to various biotic and abiotic stresses54. The conservation of motifs and gene structures resembled those identified in rice, which had fewer introns and more exons. Genomic structural analysis showed that the PyCRM genes contain more exons than introns. Similarly, proteins in the CRS1 subfamily comprise more motifs than those in other subfamilies, suggesting that this subfamily harbors genes and proteins of large sizes compared to the other three subfamilies. Despite this diversification in gene structure, most genes share similar conserved motifs, suggesting that their functions have been conserved throughout evolution55. Functional divergence and multi-stress responsive expression of PyCRMs Correlation studies have demonstrated that promoters influence temporal and spatial differences in gene expression. The promoter’s cis-regulatory elements modulate gene function by interacting with transacting factors56. In the present work, all CRM genes exhibited responsiveness to various hormones, light, and stress-related factors, with some differences in development-related elements. These variations in the specific regulatory elements suggest functional divergence among CRM genes across species. The protein-protein interaction analysis revealed key interactions among several CRM proteins and other proteins, with some showing higher connectivity, such as PyCRM14 and PyCRM24. Furthermore, GO enrichment analysis indicated involvement of these proteins in RNA binding, Group II intron splicing, transcription regulation, and ribonucleoprotein complexes, which may play roles in RNA processing. RNAs in plastids have CAF1 and CAF2 that bind to CRS2 to make CRS2-CAF1 and CRS2-CAF2 complexes, which control the splicing of intron II. Additionally, studies have demonstrated that CRM FAMILY MEMBER 2 (CFM2), present in large ribonucleoprotein complexes with CAF1 or CAF2, aids in ndhA intron splicing52. The physicochemical characteristics of PyCRM proteins revealed that the protein length, isoelectric point, and molecular weight ranged from 101 to 1118 aa, 4.78 to 10.26, and 10.95 to 123.95 kDa, respectively. In Prunus persica L, most of the proteins have a lower amino acid length, molecular weight, and isoelectric point range from 208 to 327, 5.11 to 7.85, and 22.59 to 36.16, respectively57. The physicochemical differences between these Rosaceae species highlighted the evolutionary differences in protein characteristics58. Furthermore, 3D structural predictions of PyCRM proteins revealed a high degree of conservation of the motifs, demonstrating that they have a conserved structure as well. Pears face many biotic and abiotic threats and production vulnerabilities that can hinder their growth and development59. RNA-seq expression profiling revealed that the expression of PyCRM genes varies among different tissues, abiotic stresses, and disease conditions. Under cold stress, PyCRM-An1, PyCRM-An16, and PyCRM-Ba14 showed upregulation, whereas most other PyCRM members were downregulated, indicating selective activation of specific CRM genes in response to low temperature. Similarly, PyCRM11-Cu was induced under pathogen infection, and PyCRM2-Da and PyCRM4-Da were upregulated during drought stress, while the remaining genes were largely downregulated, suggesting functional specialization of particular PyCRM paralogs in stress adaptation. According to a recent study, the CRM gene family expression in various tissues of Chinese cabbage revealed varying expression levels under biotic stress. The study also found that the CRM genes exhibited active transcription in the leaves and stems. Flg22 and Ps simulated biotic stress, which significantly downregulated the expression levels of some CRM genes, highlighting that these genes could respond to biotic stress in plants60. In Arabidopsis, AtCFM4 (belonging to the subfamily 4) is particularly sensitive to salt and cold stress61. Expression levels of rice CRM genes significantly change under abiotic stress, particularly salt stress, where all CRM genes were significantly downregulated6. Breeding and genetic engineering initiatives combining the selection of these genes may significantly benefit from the insights offered by these structural and functional views to increase agricultural stress resilience and make them valuable candidates for paving the way for future research. Conclusion This work presents a comprehensive and pangenome-wide analysis of CRM genes across 8 economically significant and nutritionally valuable Pyrus genomes. We identified 25 CRM pan-gene clusters encoded in the Pyrus 8 genomes and classified them into four subfamilies through phylogenetic analysis. Structural analysis revealed variation in α-helices and β-sheets among PyCRM proteins, while GO enrichment indicated their key roles in RNA binding, Group II intron splicing, transcription regulation, and defense-related functions in chloroplasts and mitochondria. Synteny analysis revealed strong conservation across Pyrus genomes, while the Ka/Ks analysis showed both purifying and positive selection, providing functional diversification within the gene family. Several cis-regulatory elements linked with light, hormonal responses, stress, and developmental processes were discovered in the promoter regions of PyCRMs. CRM genes demonstrated different expression under biotic and abiotic stress, providing insights into their potential roles in plant stress responses. Further research on these genes is required, as they can be applied in genetic engineering to develop multi-stress-resistant crops. Additionally, our study will help in future functional studies of Pyrus CRM genes. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1