Expression dynamics during metamorphosis and genome-wide analysis of the myosin gene family in the Chinese mitten crab (Eriocheir sinensis) Authors: Haiyue Wu, Sisi Chen, Xinyao An, Aobo Pang, Xian Hu, Jinqiu Sun, Chijie Yin, Wenqi Yang, Xiaoli Sun, Huabin Zhang, Ge Ding, Gang Wang, Daizhen Zhang Published: 2026 ABSTRACT BackgroundThe Chinese mitten crab (Eriocheir sinensis) undergoes periodic molting and exhibits discontinuous muscle growth during its larval metamorphosis, yet the underlying molecular drivers remain poorly understood. Myosin, a central motor protein, is crucial for muscle function and diverse cellular processes.ResultsThis study integrated transcriptomic analysis across three key developmental stages (zoea, megalopa, and early juvenile) with a systematic genome-wide characterization of the myosin gene family in E. sinensis. Transcriptomics revealed that myosin genes were dynamically regulated and were among 196 genes continuously upregulated in both the megalopa vs. zoea and early juvenile vs. megalopa comparisons. These co-upregulated genes are involved in tissue remodeling and nutrient metabolism, with functional enrichment highlighting tight junction assembly and actin cytoskeleton regulation. We identified 23 myosin genes (EsinMyos) in the E. sinensis genome, belonging to 11 subfamilies, with a notable representation of the conventional myosin II (Myo2) subfamily, which comprises nine members—a pattern consistent with Myo2 abundance in crustaceans across 42 surveyed animal species. Chromosomal clustering of EsinMyo2 genes on chromosome 35 suggests that tandem duplication event(s) contributed to this phenomenon. Phylogenetic analysis across 42 animal species confirmed the high conservation of myosin classes and revealed the absence of Myo10, Myo16, and Myo28 in all examined arthropods, indicating lineage-specific gene loss. Structural comparison further identified an atypical Myo2, truncated EsinMyo2-1. Expression profiling demonstrated distinct stage-specific patterns for different EsinMyos; for instance, EsinMyo1D and EsinMyo2-8 showed progressively increasing expression from zoea to juvenile, while EsinMyo15 was highest in zoea.ConclusionsThese findings underscore the pivotal and coordinated role of the myosin gene family in supporting the phase-specific morphological and physiological demands of muscle development and metamorphosis in E. sinensis, providing valuable genetic insights for aquaculture and developmental biology.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12864-026-13087-x. Background The Chinese mitten crab (Eriocheir sinensis) is an economically important freshwater crustacean in China [1, 2]. Its development proceeds through multiple stages, including zoea, megalopa, juvenile, and adult phases [3]. The entire process relies on periodic molting followed by rapid tissue repair and growth. While several studies have investigated signaling regulatory mechanisms during molting in E. sinensis, most have focused on the crucial roles of the central nervous system and neuroendocrine signaling [4–6]. Notably, the development of its muscle tissue exhibits a distinct discontinuous growth pattern, offering insights into the fundamental mechanisms of episodic growth in animals [7–9]. This growth pattern necessitates precise coordination between muscle protein synthesis and energy metabolism to fuel periodic rebuilding [10]. Yet, how metabolic pathways and molecular motors are integrated to support the rapid, phase-specific muscle development in crustaceans remains poorly understood. Among the key molecular players governing muscle structure and function, myosin stands out as a central component. This highly conserved motor protein forms the thick filaments of the sarcomere and, through its ATP-dependent interaction with actin, is directly responsible for generating contractile force [11]. Structurally, a typical myosin molecule consists of a highly conserved N-terminal motor domain, a neck region that binds myosin light chains, and a C-terminal tail domain. The variable tail domain determines cargo specificity and functional specialization [12]. Phylogenetic and functional analyses classify the myosin superfamily into multiple classes. Myosin II, or conventional myosin, constitutes the core of the muscle contraction machinery. In contrast, unconventional myosin participate in a broader range of cellular functions, including cytoskeletal remodeling, signal transduction, and auditory perception [13, 14]. Recent research indicates that during early mammalian embryonic development, the actin-myosin network ensures the precise segregation of genetic material by regulating preliminary chromosome positioning and spindle assembly, diverging from the traditional microtubule-centric perspective [15]. Concurrently, studies on virus-host cytoskeletal interactions have identified an interaction between the VP26 protein of duck enterovirus (DEV) and the host non-muscle myosin II (MYH9) [16]. This interaction harnesses the actin-myosin network to promote viral replication, suggesting myosin’s pivotal role in host-pathogen interactions. These studies collectively demonstrate that myosin’s functions extend beyond its traditional role in muscle contraction, making it an indispensable component in animal growth and development. Studies have shown that expansion of the myosin gene family is closely linked to skeletal muscle growth patterns. For example, researchers identified 60 myosin genes in the black rockfish (Sebastes schlegelii) and demonstrated that specific members of the Myo2 subfamily were significantly upregulated during rapid skeletal muscle growth in both juvenile and adult stages, suggesting a crucial role in myofiber synthesis [17]. In crustaceans, it has been reported that the well-developed muscular system of the Pacific white shrimp (Penaeus vannamei) may be associated with coordinated expansion of the Myo2 and actin gene families [18]. These expanded genes show specifically high expression in abdominal fast muscle, potentially as an adaptation for powerful burst swimming. Furthermore, a high-quality genome assembly of P. vannamei revealed significant expansion of myosin and actin gene families related to movement, vision, and neurotransmission providing new insights into its benthic adaptive evolution [19]. While studies in shrimp suggest a link between myosin gene family expansion and muscular adaptation, whether similar genetic mechanisms operate in crabs, which exhibit distinct growth and locomotion patterns, is unknown. E. sinensis, with its complex life history and high commercial value, is well-suited as a model for studying the evolution and functional diversity of myosin in crustaceans. In aquaculture practice, high mortality during the early developmental stages of E. sinensis, especially throughout the larval metamorphosis from zoea to megalopa to juvenile, remains a major bottleneck for stable seed production [20, 21]. This challenge is compounded by a fundamental lack of understanding of the molecular drivers governing muscle development and energy allocation during these critical transitions. To address this, the present study performed comprehensive transcriptomic analyses across three key early metamorphic stages (zoea, megalopa, and juvenile) of E. sinensis. Expression profiling and functional enrichment analysis highlighted the dynamic involvement of myosin genes throughout metamorphosis, suggesting their central role in supporting muscle remodeling and energy-dependent processes during discontinuous growth. Building on these transcriptomic insights, we further systematically identified and characterized the myosin gene family, analyzing their phylogenetic relationships, domain structures, and stage-specific expression patterns. This study aims to identify key genes and pathways involved in the early development of the E. sinensis and to characterize its myosin gene family, with the goal of uncovering potential genetic markers associated with healthy larval growth. The findings are expected to provide a scientific basis for selective breeding and the optimization of larval rearing techniques in aquaculture. Methods Sample collection The E. sinensis used in this experiment were collected from the Yancheng Jinyang Aquatic Seed Farm (Yancheng, China). Specimens at different developmental stages were selected: zoea Ⅰ, megalopa, and early juvenile (the first crab stage, sampled immediately after the megalopa molted into a crab-like form). Three biological replicates were established for each stage, with each replicate consisting of a pool of 10 whole individuals. All samples were rapidly frozen in liquid nitrogen and stored at − 80 °C for subsequent experimental use. Library construction and sequencing After homogenizing the samples, total RNA was extracted using TRIzol reagent (Vazyme, R401-01). RNA concentration was quantified using a Qubit fluorometer (Thermo Fisher Scientific), and RNA integrity was assessed with an Agilent 2100 Bioanalyzer. RNA samples meeting the required concentration, purity, and integrity criteria were selected. The mRNA transcriptome library construction kit (YEASEN, 12301ES96) was used to complete mRNA isolation, fragmentation, double-stranded cDNA synthesis, end repair, A-tailing, adapter ligation, and library amplification, converting the RNA into a library for platform sequencing. All steps were strictly performed according to the manufacturer’s instructions. Sequencing was performed using the DNBSEQ-T7 platform to generate 150 bp paired-end reads. Transcriptome analysis pipeline Raw reads were trimmed and filtered using fastp v1.0.1 [22] to obtain clean reads. Read alignment was performed with Hisat2 v2.1.0 [23] under default parameters, utilizing the available “Jianghai 21” E. sinensis genome (GCA_024679095.1) as the reference genome. Subsequently, featureCounts v2.0.2 [24] was employed to tally the count of each gene across all samples and calculate the TPM values. Sequencing quality metrics for each sample are provided in Table S1. All samples were processed and sequenced in a single batch; therefore, no batch correction was applied. Genes were filtered from the TPM matrix by removing those not expressed (TPM < 0.1) in at least three samples and two groups. Perform principal component analysis using the filtered TPM matrix to validate relationships between samples. To quantitatively evaluate replicate consistency, Pearson correlation coefficients were calculated between all samples based on log2(TPM + 0.1) values. Differentially expressed genes (DEGs) were identified using R package DESeq2 v1.48.2 [25]. Genes that satisfied the criteria of an adjusted P < 0.05 and |log2Fold Change| ≥ 2 were designated as differentially expressed. Performed time series clustering analysis using the R package Mfuzz v2.68.0 [26], setting the number of clusters to six. The fuzzifier parameter m was estimated from the data using the mestimate function, and all other parameters were set to defaults. To conduct functional enrichment analysis, the eggnog-mapper was first employed to annotate the protein sequences of the E. sinensis. Subsequently, functional enrichment analysis was performed in R using the clusterProfiler v4.16.0 package [27], with visualization achieved through the enrichplot v1.28.4 and circlize v0.4.17 packages. Identification of the myosin gene family in multiple animal species First, we downloaded reference genomes and genome annotation data (.gff) for 42 species, including the E. sinensis, from the NCBI database. Relevant information is detailed in Table S2. Protein sequences were extracted for all species, and a reference database was constructed using the makeblastdb command within BLAST v2.17.0 with default parameters [28]. A total of 872 myosin sequences from Platyhelminthes, Annelida, Arthropoda, Echinodermata, Chordata, Cnidaria, Mollusca, and Nematoda were obtained from the NCBI database as query sequences (Table S3). Sequences with E-values < 1e− 5 were selected as candidate myosin sequences following alignment using the blastp command. Additionally, the Myosin_head HMM model (PF00063) was downloaded from the PFAM database. The hmmsearch function in HMMER v3.3.1 was then employed to further screen for sequences bearing the myosin conserved domain, applying the --cut_ga option and an E-values cutoff of 1e− 5, with all other parameters kept at default settings. Phylogenetic tree construction To investigate the evolutionary and phylogenetic relationships of myosin between the E. sinensis and other species, the identified myosin protein sequences were employed for interspecies phylogenetic analysis. Employing the plant myosin Ⅺ sequences (NCBI accession numbers NP_180882.2, AAK21311.1, and AHI45151.1) as the outgroup, sequence alignment was performed using the MUSCLE algorithm within MEGA v12 software [29]. The alignment was trimmed with TrimAl to remove poorly aligned regions. Maximum likelihood (ML) phylogenetic trees were constructed in MEGA v12 using the LG substitution model, and tree reliability was assessed with 1,000 bootstrap replicates. The evolutionary tree was visualized and aesthetically enhanced using the online platform iTOL (https://itol.embl.de/). Characteristics of the myosin gene family in E. sinensis Chromosomal localization of E. sinensis myosin was performed using the ‘Show Genes on Chromosome’ plugin within TBtools [30]. Subcellular localization prediction was conducted via the online platform CELLO (http://cello.life.nctu.edu.tw/). Protein physicochemical properties of myosin were predicted using the ProtParam tool (https://web.expasy.org/protparam/). The MEME tool (https://meme-suit.org/meme/tools/meme) was employed to predict conserved motifs within the myosin sequence, whilst the NCBI CD-search tool (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi?mode=all) was utilized to identify conserved domains within the myosin protein. The secondary structure of myosin was predicted using the GOR4 website (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_gor4.html). The 3D structure of myosin was predicted using the SWISS-MODEL website (https://swissmodel.expasy.org/interactive). Real-time quantitative reverse transcription PCR Total RNA passing quality assessment was reverse transcribed using the HisyGo RT Red SuperMix for qPCR (+ gDNA Wiper) kit (Vazyme, RT101-01) to synthesize cDNA for subsequent Real-time Quantitative Reverse Transcription PCR (RT-qPCR) validation. Primers were designed using Premier 6.0 software (Table S4). qPCR experiments employed SupRealQ Purple Universal SYBR qPCR Master Mix (U+) (Vazyme, Q412) and the QuantStudio3 Fluorescence Quantitative System (Thermo Fisher Scientific), with reaction systems configured and procedures executed strictly according to manufacturer instructions. Relative gene expression levels were calculated using the 2−ΔΔCt method. A one-way Analysis of Variance (ANOVA) was performed using SPSS v21.0 software, with Duncan’s post hoc multiple comparison test employed to assess intergroup differences. Differences were considered statistically significant at P < 0.05. Results Transcriptomic analysis reveals developmental metabolic reprogramming and associated myosin expression during E. sinensis larval metamorphosis We conducted transcriptome analysis on E. sinensis across three developmental stages: zoea (Z), megalopa (M), and early juvenile (EJ). PCA validated the reliable repeatability within groups (Figure S1A). Pearson correlation analysis further confirmed high reproducibility among biological replicates (Figure S1B). Differential expression analysis identified DEGs between two consecutive stages: 2,492 DEGs in M vs. Z, predominantly upregulated in M (1,853); and 1,754 DEGs in EJ vs. M, with 828 upregulated in EJ (Fig. 1A and B). There are 196 DEGs that were up-regulated in both M vs. Z and EJ vs. M comparisons (Fig. 1A). These genes represent key genes playing a pivotal role in the developmental transition from Z to EJ in the E. sinensis, including genes encoding cuticle protein (Esine00701, Esine00705), chitin-binding protein FIBCD1 (Esine02394), chitinases Cht5(Esine11817), Cht9(Esine17681), myosin unc-54 (Esine13120) and MYO1E (Esine11002) (Table S5). Functional enrichment analysis of the 196 co-upregulated genes revealed significant enrichment in KEGG pathways including Protein digestion and absorption, Carbohydrate digestion and absorption, Starch and sucrose metabolism, Sphingolipid metabolism, and Amino sugar and nucleotide sugar metabolism (Fig. 1C). These findings correspond to the intrinsic adaptive strategy and core biological program of the crab during its development from Z to EJ, whereby it coordinately enhances nutrient metabolism and utilization to provide the material and energetic foundation for exoskeleton remodeling and muscle development, thereby adapting to shifts in life history and ecological niche. Conversely, only 29 DEGs were downregulated in both M vs. Z and EJ vs. M comparisons, including CPVL (Esine01422, Esine01423), ABCC4 (Esine09924), TMPRSS6 (Esine17249), CHST11 (Esine26226), and MYO15 (Esine01944) (Table S5). This set also included a myosin gene (MYO15), whose reduced expression may be associated with functional specialization and tissue remodeling of locomotor organs during different developmental stages. Fig. 1Differential expression analysis of three early developmental stages in the E. sinensis. A Volcano plots displaying differentially expressed genes (DEGs) in the M vs. Z comparison and the EJ vs. M comparison. B Venn diagram showing the overlap between significantly upregulated genes in the megalopa (M) vs. zoaea (Z) and those in the early juvenile (EJ) vs. M. C Venn diagram showing the overlap between significantly downregulated genes in the M vs. Z and EJ vs. M. D KEGG enrichment analysis of the 196 consistently upregulated genes. Enrichment significance was determined at P < 0.05 Differential expression analysis of three early developmental stages in the E. sinensis. A Volcano plots displaying differentially expressed genes (DEGs) in the M vs. Z comparison and the EJ vs. M comparison. B Venn diagram showing the overlap between significantly upregulated genes in the megalopa (M) vs. zoaea (Z) and those in the early juvenile (EJ) vs. M. C Venn diagram showing the overlap between significantly downregulated genes in the M vs. Z and EJ vs. M. D KEGG enrichment analysis of the 196 consistently upregulated genes. Enrichment significance was determined at P < 0.05 Based on the transcriptomic profiles across three developmental stages of the E. sinensis, Mfuzz time-series clustering analysis categorized 16,100 protein-coding genes into six clusters (Fig. 2A). Cluster 2, comprising 2,678 genes, exhibited a progressively increasing expression pattern throughout development. GO enrichment analysis of these upregulated genes indicated significant enrichment for terms related to tissue morphogenesis and functional specialization (Fig. 2B). Key biological processes included tight junction assembly and apical junction assembly, underscoring an active program for establishing polarized epithelial barriers. Concurrent enrichment of immune response-regulating cell surface receptor signaling and leukocyte migration highlighted the parallel development of immune readiness. At the molecular level, significant enrichment for calcium-dependent protein binding was identified, suggesting a critical role for calcium-mediated signaling in coordinating these developmental events, potentially in cell adhesion and exocytotic processes. KEGG pathway analysis further supported the functional themes identified by GO enrichment (Fig. 2C; Table S6). Some key genes were co-enriched in tight junction and regulation of actin cytoskeleton. These include RhoL, ITGB1, TIAM1, WASL, Mlc2, and the beta-actin gene ACTB, directly linking the assembly of cell junctions to actomyosin remodeling. The gene unc-54 (Esine13120), encoding a core component of the actin-myosin complex, exhibits specific enrichment in the tight junction pathway. This indicates that myosin plays crucial roles in stabilizing tight junction structures, promoting morphogenesis, and regulating immune responses during the developmental process of the E. sinensis. Fig. 2Time-series clustering and functional enrichment analysis of genes dynamically expressed during E. sinensis development. A Mfuzz time-series clustering of genes across three developmental stages. Genes were grouped into six distinct expression pattern clusters. B Dotplot display the GO terms significantly enriched (P < 0.05) in genes within Cluster 2. C KEGG enrichment analysis for genes in Cluster 2. Significantly enriched pathways (P < 0.05) related to morphogenesis, cytoskeletal regulation, and immune signaling are displayed Time-series clustering and functional enrichment analysis of genes dynamically expressed during E. sinensis development. A Mfuzz time-series clustering of genes across three developmental stages. Genes were grouped into six distinct expression pattern clusters. B Dotplot display the GO terms significantly enriched (P < 0.05) in genes within Cluster 2. C KEGG enrichment analysis for genes in Cluster 2. Significantly enriched pathways (P < 0.05) related to morphogenesis, cytoskeletal regulation, and immune signaling are displayed Identification and phylogenetic analysis of myosin gene families in multiple animal species Given the crucial role of myosin in the growth and development of the E. sinensis and the lack of systematic identification of its gene family, we conducted this study to address the gap. In this study, we identified myosin gene family members across 42 species, comprising 19 arthropods, 10 vertebrates, 4 mollusks, 3 echinoderms, 2 flatworms, 2 cnidarians, 1 annelid, and 1 nematode (Table S7). Within the maximum likelihood tree, distinct myosin subtypes from various species can be observed clustered together (Fig. 3A). This includes fifteen types: Myo1, Myo2, Myo3, Myo5, Myo6, Myo7, Myo9, Myo10, Myo15, Myo16, Myo18, Myo19, Myo20, Myo22, and Myo28. This demonstrates the high degree of evolutionary conservation of the myosin gene family across animal species, having undergone major diversification events early in their evolutionary history. Myo2 members are the most numerous, a fact particularly evident in vertebrates and crustaceans (Table S7). Twenty-three members of the Myo family have been identified in the E. sinensis, comprising four Myo1, nine Myo2, two Myo19, and one each of Myo3, Myo5, Myo6, Myo7, Myo9, Myo15, Myo20, and Myo22 (Fig. 3B). Members Myo10, Myo16, Myo18, and Myo28 were absent. Furthermore, Myo10, Myo16, and Myo28 were not identified in any of the arthropods examined (Table S7). Fig. 3Maximum likelihood tree of the myosin gene family. A Phylogenetic analysis of myosin genes reconstructed from 42 species (see Table S7). The tree reveals 15 conserved myosin subtypes (Myo1, Myo2, Myo3, Myo5, Myo6, Myo7, Myo9, Myo10, Myo15, Myo16, Myo18, Myo19, Myo20, Myo22, and Myo28). B Phylogenetic tree of 23 E. sinensis myosin genes Maximum likelihood tree of the myosin gene family. A Phylogenetic analysis of myosin genes reconstructed from 42 species (see Table S7). The tree reveals 15 conserved myosin subtypes (Myo1, Myo2, Myo3, Myo5, Myo6, Myo7, Myo9, Myo10, Myo15, Myo16, Myo18, Myo19, Myo20, Myo22, and Myo28). B Phylogenetic tree of 23 E. sinensis myosin genes Chromosomal distribution of the myosin gene family members in E. sinensis We further focused our analysis on the characteristics of the myosin gene family members in E. sinensis (EsinMyos). Chromosomal localization analysis revealed that 23 EsinMyos are distributed across 13 chromosomes and one scaffold in the E. sinensis genome (Fig. 4). Chromosome 35 harbors the highest number of myosin family members, exhibiting distinct gene clustering, which may reflect tandem duplication event(s) involving EsinMyo2. Fig. 4Chromosomal localization of the 23 identified E. sinensis myosin genes (EsinMyos). The genes are distributed across 13 chromosomes and one scaffold. The scale on the left denotes the relative physical distance (Mb) along each chromosome Chromosomal localization of the 23 identified E. sinensis myosin genes (EsinMyos). The genes are distributed across 13 chromosomes and one scaffold. The scale on the left denotes the relative physical distance (Mb) along each chromosome Physicochemical properties and subcellular localization of EsinMyos Physicochemical and subcellular localization analyses reveal that most predicted EsinMyos possess extended amino acid lengths and substantial molecular weights, aligning with their role as heavy chains (Table S8). However, EsinMyo1B and EsinMyo2-1 exhibit notably shorter sequences, distinguishing them from other family members. Furthermore, EsinMyo1B is uniquely predicted to localize to the cytoplasm and has an acidic isoelectric point (pI), whereas the other three EsinMyo1 members are nuclear-localized and alkaline. All nine members of the EsinMyo2 subfamily share cytoplasmic localization and an acidic pI. Genetic structure, protein motif and domain, and protein structure predictions of EsinMyos All EsinMyos possess relatively conserved motifs and domains (Fig. 5A and B). EsinMyo2-1 lacks the long tail domain present in other EsinMyo2 members, accounting for its significantly lower molecular weight. While this structural difference could imply that the myosin represents a non-traditional type II myosin that forms heterodimers exclusively with specific partners, it is important to note that this interpretation is speculative and has not been experimentally validated. Similarly, EsinMyo1B lacks the TH1 domain compared to other EsinMyo1 members, which is likely the primary reason for its reduced molecular weight and subcellular localization within the cytoplasm. The gene structure of EsinMyo2-1 reveals that, despite its short coding sequence length, it possesses a longer overall gene length (Figure S2). This further substantiates that this gene has diverged from other members within the family. Fig. 5Analysis of conserved motifs and protein domains of the EsinMyos. A Distribution of conserved motifs (Motif 1–10) among the 23 EsinMyos. B Protein domain architecture of the EsinMyos Analysis of conserved motifs and protein domains of the EsinMyos. A Distribution of conserved motifs (Motif 1–10) among the 23 EsinMyos. B Protein domain architecture of the EsinMyos The results of the protein secondary structure prediction indicate that the proportion of α-helices in EsinMyos ranges from 28.72% to 67.13%, representing the highest proportion; the proportion of disordered coils ranges from 23.54% to 54.08%; while the proportion of β-folds ranges from 9.01% to 25.31%, which is relatively low (Table S9). Furthermore, we conducted a three-dimensional structure prediction for EsinMyos, which similarly reveals the absence of the long coiled-coil tail domain in EsinMyo2-1 (Figure S3; Table S10). Expression profile of EsinMyos in E. sinensis at different developmental stages All EsinMyos were expressed (TPM > 0.1) during the Z, M, and EJ stages in E. sinensis (Fig. 6A). Among them, EsinMyo2-2, EsinMyo2-8, and EsinMyo1D exhibited significantly upregulated expression in the EJ stage. In contrast, genes such as EsinMyo9, EsinMyo15, and EsinMyo19-2 showed markedly higher expression in the Z stage. Additionally, myosin genes including EsinMyo2-3, EsinMyo2-4, and EsinMyo2-5 displayed the highest expression levels during the M stage. Fig. 6Expression profiles of EsinMyos across three developmental stages in E. sinensis. A Expression levels of all 23 identified EsinMyos across all samples. Heatmaps were plotted using Z-score normalized TPM values. B Validation of expression patterns for EsinMyos by RT-qPCR. Relative expression levels were calculated using the Livak method. Data are presented as mean ± SEM (n = 3 biological replicates). Statistical significance between stages was assessed by one‑way ANOVA followed by Duncan’s post hoc test. *, P < 0.05; **, P < 0.01; ***, P < 0.001 Expression profiles of EsinMyos across three developmental stages in E. sinensis. A Expression levels of all 23 identified EsinMyos across all samples. Heatmaps were plotted using Z-score normalized TPM values. B Validation of expression patterns for EsinMyos by RT-qPCR. Relative expression levels were calculated using the Livak method. Data are presented as mean ± SEM (n = 3 biological replicates). Statistical significance between stages was assessed by one‑way ANOVA followed by Duncan’s post hoc test. *, P < 0.05; **, P < 0.01; ***, P < 0.001 The relative expression levels of selected EsinMyos were further validated by RT-qPCR (Fig. 6B). The results demonstrated that the expression of EsinMyo1D and EsinMyo2-8 was significantly lower in the Z stage compared to the M and EJ stages (P < 0.05), with expression in EJ being significantly higher than that in M (P < 0.05). Conversely, EsinMyo15 showed the opposite pattern, with its expression significantly higher in Z than in the other two stages (P < 0.05). The remaining RT-qPCR results were consistent with the transcriptomic data (Fig. 6B and Figure S4), thus validating the reliability of the RNA-seq findings. Discussion The growth of the E. sinensis is highly dependent on periodic molting, a process during which the old exoskeleton is shed and new tissues undergo rapid regeneration and expansion [10, 31, 32]. Muscle tissue, in particular, exhibits marked discontinuous growth throughout this cycle [8]. Our transcriptomic profiling reveals that dynamic expression patterns of myosin genes are a central feature of larval metamorphosis in E. sinensis. The co-upregulation of myosin genes within the core set of 196 continuously upregulated DEGs strongly implicates myosin-driven processes as integral to the developmental program from the zoea to the early juvenile stage. This coordinated upregulation aligns functionally with the concurrent enrichment of pathways related to protein, carbohydrate, and nutrient metabolism, suggesting that myosin activity is coupled with the enhanced bioenergetic and biosynthetic demands of metamorphosis. Similar coupling between myosin gene expression and energy metabolism has been observed during insect metamorphosis, for example in Drosophila melanogaster where the upregulation of muscle-specific myosin isoforms is temporally linked to the activation of metabolic pathways during pupation [33]. Myosin, as a conserved cytoskeletal motor molecule, plays an essential role in animal development [34, 35]. Consistent with findings in other crustaceans, such as P. vannamei where myosin genes also show dynamic expression linked to morphogenesis [18], our time-series expression profiling identified multiple myosin genes among steadily upregulated transcripts. These genes were significantly enriched in pathways closely associated with myosin function, including tight junction assembly and regulation of the actin cytoskeleton, underscoring the crucial role of actin-myosin remodeling in establishing and stabilizing epithelial barriers during E. sinensis metamorphosis [36, 37]. This is consistent with previous findings that myosin plays a crucial role in cell adhesion and junctional integrity during morphogenesis [38, 39]. Given the pivotal role of myosin in the growth and development of E. sinensis, this study presents the first genome-wide systematic identification and analysis of the myosin gene family in this species. We further expanded the phylogenetic context by identifying myosin genes from 42 additional species, making this the most comprehensive comparative study of the myosin gene family in invertebrates to date. A total of 23 myosin members were identified in the E. sinensis genome, a number comparable to the 29 reported in the closely related crustacean P. vannamei [18]. Notably, we observed that the myosin Ⅱ (Myo2) subfamily is relatively abundant in crustaceans compared to other invertebrates, along with potential duplication event(s) of this subfamily on chromosome 35 of E. sinensis. As “conventional” myosin, Myo2 proteins are core components of muscle contraction and contribute to diverse cellular processes including cell division, migration, adhesion, and tissue morphogenesis [40–42]. This relative abundance likely reflects an elevated demand for cyclic tissue remodeling and rapid muscle regeneration associated with molting and discontinuous growth in crustaceans. Furthermore, we detected structural heterogeneity within the EsinMyo2 subfamily. The protein encoded by EsinMyo2-1 is considerably smaller than other members due to the absence of a long tail domain. EsinMyo2-1 may represent a functionally specialized, shorter Myo2 adapted to specific cellular or mechanical contexts. Myosin Ⅰ (Myo1) constitutes the second-largest myosin subfamily in E. sinensis, with all four conserved subtypes (A–D) identified. Myo1 is an evolutionarily conserved single-headed motor expressed in diverse tissues and organisms from ancestral eukaryotes to humans. All Myo1 isoforms can bind directly to membranes, providing a dynamic link between the actin cytoskeleton and the plasma membrane or intracellular organelles [43]. In this study, members of the EsinMyo1 subfamily exhibited distinct stage-specific expression patterns, suggesting functional diversification during metamorphosis. Notably, EsinMyo1D demonstrated a progressive increase in expression from the zoea to the early juvenile stage, while EsinMyo1C showed its highest expression during the zoea stage, followed by a decline. This divergent expression pattern likely indicates distinct temporal and functional specializations, with EsinMyo1C potentially supporting early larval motility or cytoskeletal organization, whereas the rising expression of EsinMyo1D may cater to increasing demands for membrane-cytoskeleton linkage and intracellular trafficking in later juvenile stages. Sub-functionalization within the Myo1 family has similarly been documented in vertebrates, where different Myo1 genes are deployed for distinct tasks such as microvillar formation, endocytosis, and stereocilia maintenance [44, 45]. Other EsinMyos also displayed varied expression profiles. Particularly noteworthy were genes such as EsinMyo1D and EsinMyo2-8, which exhibited a consistent upward trend across the three stages. These continuously upregulated genes are of significant interest because their expression trajectory strongly correlates with the progressive morphological and physiological complexity of the developing crab [46]. Their sustained induction likely underpins constitutive and escalating roles in fundamental processes such as sustained muscle growth, persistent tissue remodeling, and the strengthening of cellular structures—processes that are essential for the transition to a more active, benthic juvenile lifestyle [47]. The specific temporal regulation of different myosin isoforms thus appears to be a key mechanism fine-tuning cytoskeletal dynamics to meet stage-specific developmental demands. The findings of this study regarding the relationship between EsinMyos and crab development are primarily based on transcriptomic expression patterns, phylogeny, and sequence-based structural predictions, and therefore remain descriptive and correlative. Direct functional validation through gene knockdown or overexpression in E. sinensis would be necessary to establish causal roles in muscle remodeling during metamorphosis. Second, although qPCR validated the RNA-seq expression trends for selected genes, the study lacks protein-level validation, which could further clarify the dynamics of myosin protein isoforms. Future work integrating functional genetics and proteomics will be essential to translate the transcriptomic resources generated here into mechanistic understanding of muscle development in crustaceans. Conclusion In summary, this study provides a comprehensive genome-wide analysis of the myosin gene family in the E. sinensis, elucidating its dynamic role during early larval metamorphosis. Transcriptomic profiling revealed that myosin genes are integral components of a core developmental program, with their expression tightly coupled to pathways governing tissue morphogenesis, cytoskeletal remodeling, and energy metabolism. We systematically identified 23 EsinMyos and observed a notable relative abundance of the Myo2 subfamily, a feature potentially linked to the high demand for cyclic tissue reconstruction associated with molting and discontinuous growth in crustaceans. The discovery of a truncated EsinMyo2-1 suggests functional diversification within this subfamily. Expression analysis revealed distinct temporal patterns for different myosin members, indicating stage-specific functional specializations to meet shifting developmental demands, such as the sustained upregulation of EsinMyo1D and EsinMyo2-8 correlating with progressive morphological complexity. Collectively, our findings highlight the myosin gene family as a key genetic module orchestrating muscle development and adaptive remodeling in E. sinensis. This work establishes a foundational resource for understanding the molecular basis of growth in crustaceans and identifies potential candidate genes for future selective breeding and larval health management in aquaculture. Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2.