Integrated transcriptomic and proteomic analyses reveal muscle fiber-type transformation in meat rabbits Authors: Cheng Jia, Wu Qianwen, Ji Zhengwei, Yao Zhongqiang, Wang Ling, Lu Hongzhao, Zeng Wenxian Published: 2026 ABSTRACT BackgroundSkeletal muscle fiber composition plays a critical role in determining muscle growth and meat quality in livestock species. However, the molecular mechanisms underlying muscle fiber-type transformation in meat rabbits remain poorly understood.ResultsHere, we integrated transcriptomic and label-free proteomic analyses to investigate skeletal muscle development in Ira rabbits at three developmental stages (D0, D35, and D70). Immunofluorescence staining revealed a gradual increase in slow-twitch muscle fibers in the gastrocnemius during postnatal growth. Enzyme activity assays showed decreased lactate dehydrogenase (LDH) activity and increased succinate dehydrogenase (SDH) activity, indicating a metabolic shift toward oxidative metabolism. Transcriptomic analysis identified thousands of differentially expressed genes across developmental stages, whereas proteomic profiling revealed numerous differentially expressed proteins associated with metabolic pathways. We identified ten candidate genes (NOS1, CASQ2, ANKRD1, ANKRD2, CXCL10, GPC3, NEDD4, MYORG, MYLK2, EZH2) and nine key proteins (TRAF3IP3, APBB2, PGK1, TMCO5A, PKM, PVALB, PDLIM7, PHKG1, PHKA1) that may be associated with muscle fiber-type transformation during postnatal skeletal muscle development. Functional enrichment analysis indicated that these molecules were associated with multiple metabolic pathways including glycolysis/gluconeogenesis and butanoate metabolism.ConclusionsThese findings provide a multi-omics perspective on skeletal muscle development in rabbits and contribute to a better understanding of the molecular basis of muscle fiber-type transformation during postnatal growth.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12864-026-12906-5. Background In recent years, rabbit meat has gained increasing popularity among consumers because of its high protein and polyunsaturated fatty acid content, as well as its low fat and cholesterol levels [1]. The meat rabbit industry has emerged as an important component of the global livestock sector, owing to its efficient reproductive cycle and desirable meat quality traits. The formation of meat quality traits is closely associated with skeletal muscle development. The biological characteristics of skeletal muscle not only regulate muscle growth but also directly influence the sensory attributes and physicochemical properties of meat [2, 3]. Skeletal muscle development is a complex multistage biological process that involves myocyte proliferation, differentiation, and muscle fiber maturation. A key aspect of this process is muscle fiber-type transformation, which is associated with the balance between fast-twitch (glycolytic) and slow-twitch (oxidative) muscle fibers. Fast-twitch fibers are characterized by high ATPase activity and abundant glycogen stores, which promote rapid glycolysis and lead to a rapid decline in muscle pH after slaughter. As a result, an acidic environment is generated, which contributes to the development of poor-quality meat, such as pale, soft, and exudative (PSE) meat [4–7]. In contrast, slow-twitch fibers possess a high mitochondrial density, which enhances water-holding capacity. They also contain abundant myoglobin and hemoglobin that associated with a bright red coloration. These characteristics promote higher meat color scores and improved meat stability through a more gradual metabolic process [4, 5, 8–10]. Nevertheless, the molecular mechanisms underlying muscle fiber-type transformation remain poorly understood. The gastrocnemius muscle was selected for analysis because of its important role in locomotion and its representative composition of both fast- and slow-twitch muscle fibers. Therefore, it serves as an ideal tissue for investigating the dynamics of muscle fiber-type transformation [11–13]. Integrated multi-omics approaches provide powerful tools for elucidating the molecular networks underlying muscle development. Transcriptomic analyses of bovine longissimus dorsi muscle have identified key regulators of myofiber differentiation, such as ATPase sarcoplasmic reticulum Ca²⁺ transporting 1 (Atp2a1) and tropomodulin 4 (Tmod4) [14]. In addition, proteomic profiling of porcine muscle has revealed DEPs involved in fatty acid metabolism and glycolysis [15]. However, existing studies on rabbit musculature mainly employ methods such as histological examination, mechanical testing, and gene sequencing. These studies primarily describe phenotypic changes in rabbit muscle tissue under various conditions, including muscle atrophy, adipose infiltration, and structural alterations [16–20]. The molecular regulatory mechanisms governing skeletal muscle development in rabbits remain relatively underexplored. A key biological process underlying skeletal muscle development is muscle fiber type transformation. This process not only plays a crucial role in determining meat quality in livestock species, but is also closely associated with various muscle-related diseases and metabolic disorders in mammals. Because meat rabbits show translational relevance to human myopathies, they represent a valuable model organism for muscle research [21–23]. Therefore, meat rabbits were selected as the experimental subjects in the present study. Based on biological relevance and economic feasibility, three key sampling time points were selected: birth (D0), weaning (D35), and slaughter (D70). In this study, the Ira rabbit was selected as the experimental model because of its stable genetic background, high reproductive efficiency, and low maintenance requirements, which together ensure experimental reliability. Integrated transcriptomic and proteomic approaches were employed to systematically analyze the key factors and pathways involved in muscle fiber-type transformation in the gastrocnemius across three developmental stages. The resulting multi-omics dataset provides molecular insights into skeletal muscle development and provides a molecular basis for future studies on meat quality regulation. Materials and methods Reagents and animals Ira rabbits used in this study were obtained from the Xifeng Rabbit Breeding Base (Xixiang County, Hanzhong City, Shaanxi Province, China). All selected Ira rabbits were raised under identical environmental and management conditions. Three biological replicates were included for each developmental stage (D0, D35, D70). Three biological replicates were used as the minimum standard for exploratory analyses in livestock tissues, acknowledging that this sample size may limit statistical power. Anesthesia was induced using CO₂, followed by euthanasia through CO₂ asphyxiation after tissue collection. All experimental procedures were conducted in accordance with the ARRIVE guidelines and the Guidelines for the Care and Use of Experimental Animals, and were approved by the Institutional Animal Care and Use Committee of Shaanxi University of Technology (approval number: SLGQD/09/2021). Immunofluorescence analysis Gastrocnemius muscle samples were placed in 5 ml centrifuge tubes and fixed with paraformaldehyde at 4 °C for 72 h. Subsequently, gradient dehydration was performed using 10%, 20%, and 30% sucrose solutions, with each step lasting 12 h. The tissues were subsequently embedded in optimal cutting temperature compound (OCT) and allowed to solidify. The embedded samples were transferred to a − 20 °C cryostat, and sections with a thickness of 7–8 μm were prepared. The sections were collected on glass slides and stored until further analysis. The sections were fixed in acetone at − 20 °C for 10–11 min. The sections were then washed three times with phosphate-buffered saline (PBS) for 5 min each. Permeabilization was then performed using 0.3% Triton X-100 for 10 min, followed by rinsing with PBS. Non-specific binding sites were blocked with 5% bovine serum albumin (BSA) at room temperature for 5 h. Excess BSA was removed before antibody incubation. Primary antibodies against myosin heavy chain 4 (MYH4; 20140-1-AP, Proteintech) and myosin heavy chain 7 (MYH7; 22280-1-AP, Proteintech) were applied at dilutions of 1:200-1:300. The sections were incubated overnight at 4 °C in a humidified chamber. The following day, after equilibration to room temperature for 30 min, the sections were washed three times with Tris-buffered saline (TBS; 5 min per wash). Fluorescent secondary antibodies (1:1000 dilution) were then applied, including fluorescein isothiocyanate (FITC; F-2765, Thermo Fisher) for MYH4 and Alexa Fluor 546 (A-11035, Thermo Fisher) for MYH7. The sections were incubated at 37 °C for 30 min under light-protected conditions. After washing with TBS, nuclei were stained with DAPI (1:1000 dilution) for 5 min in the dark. Finally, the sections were mounted with 80% glycerol and imaged using a Leica STELLARIS LAS X confocal laser scanning microscope. Protein extraction and enzyme activity detection The lysis buffer was prepared at a ratio of Radioimmunoprecipitation Assay (RIPA) buffer, 50× phosphatase inhibitor, and phenylmethylsulfonyl fluoride (PMSF) of 100:2:1. The mixture was thoroughly vortexed and maintained on ice until use. Approximately 0.1 g of fresh gastrocnemius tissue was weighed and placed in the buffer zone of a glass homogenizer, and the tissue was minced using ophthalmic scissors. Subsequently, 1 ml of lysis buffer was added, and the sample was homogenized until no visible tissue fragments remained. All procedures were conducted under low-temperature conditions. The homogenate was transferred to a 1.5 ml centrifuge tube and centrifuged at 12,000 × g for 30 min at 4 °C. The supernatant containing proteins was then collected for proteomic analysis. For enzyme activity measurements, separate tissue aliquots were processed according to kit instructions. Enzyme activities were detected using kits from Nanjing Jiancheng Biotechnology Co., Ltd. for succinate dehydrogenase (SDH; A022-1-1), malate dehydrogenase (MDH; A021-2-1), and lactate dehydrogenase (LDH; A020-1) according to the manufacturer’s instructions. One unit (U) of enzyme activity was defined as the amount of enzyme catalyzing the conversion of 1 µmol of substrate per minute under assay conditions. Enzyme activities were normalized to total protein concentration and expressed as U/mg protein. All measurements were performed using three independent biological replicates for each developmental stage. Transcriptome analysis Total RNA was extracted from gastrocnemius muscle using TRIzol reagent, followed by purification and quality assessment. RNA libraries were constructed and sequenced on the Illumina HiSeq™ 2000 platform (150 bp paired-end reads) by Shanghai Applied Protein Technology Co., Ltd. Raw sequencing data were processed to remove adapter sequences, low-quality bases (average quality score < Q20), reads shorter than 50 bp, and reads containing ambiguous bases (“N”), thereby generating high-quality clean reads. Approximately 6.1–7.1 Gb of clean data (~ 44–51 million reads) were generated per sample. Clean reads were aligned to the Oryctolagus cuniculus reference genome (assembly um_NZW_1.0) using HISAT2 (version 2.2.1) with default parameters. After quality filtering, 88–92% of clean reads were successfully mapped. Sample quality was evaluated using principal component analysis (PCA) and correlation analysis; no significant batch effects were detected among biological replicates. Gene expression levels were quantified as raw read counts using feature Counts (version 2.0.3). DESeq2 (version 1.34.0) was applied to raw count data to identify differentially expressed genes (DEGs) with thresholds of |log2 fold change| > 1 and adjusted p < 0.05 (Benjamini-Hochberg correction). A volcano plot was generated to visualize DEG distribution. Functional annotation of DEGs was performed using Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Clusters of Orthologous Groups (COG/KOG) databases. Enrichment analysis was conducted using GOseq, and terms with p < 0.05 were considered statistically significant. Proteome analysis Proteins were extracted from gastrocnemius muscle using SDT lysis buffer and quantified with a BCA protein assay kit (Bio-Rad, USA). Protein samples were digested using the filter-aided sample preparation (FASP) method with trypsin. The resulting peptides were desalted using a C18 column, freeze-dried, and re-dissolved in 0.1% formic acid. Peptide concentration was determined by OD280 measurement. Peptides were separated using an Easy-nLC system equipped with a C18 column (10 cm × 75 μm, 3 μm) at a flow rate of 300 nl/min. Mobile phase A (0.1% formic acid in water) and B (84% acetonitrile with 0.1% formic acid) were used for gradient elution. MS analysis was performed on a Q-Exactive mass spectrometer in positive ion mode. Full MS scans (300–1800 m/z) were acquired at 70,000 resolutions, followed by HCD fragmentation of the top 20 precursor ions at 30 eV (MS2 resolution: 17,500). Dynamic exclusion was set to 60 s. Raw data were searched against the Oryctolagus cuniculus protein database using MaxQuant (v1.5.3.17) with label-free quantification (LFQ) and a 1% FDR threshold at both peptide and protein levels. Proteins detected in at least two biological replicates per group were retained. LFQ intensity values were log2-transformed and normalized. Missing values were imputed using a normal distribution to simulate low-abundance signals. DEPs were defined as proteins with |log2 fold change| > 1 and adjusted p < 0.05 (Benjamini-Hochberg correction). Functional annotation was performed using GO, KEGG, and COG/KOG databases. Subcellular localization was predicted using the CELLO online tool (http://cello.life.nctu.edu.tw/). mRNA and protein correlation analyses Pearson correlation coefficients were calculated between mRNA and protein expression levels for genes detected in both datasets. Concordantly regulated genes (differential expression in the same direction) and discordantly regulated genes (opposite directions or single-omic differential expression) were identified using a nine-quadrant plot, with log₂ fold changes of mRNA (x-axis) and protein (y-axis) defining the nine categories. GO biological process terms and KEGG pathway enrichment results from the transcriptome and proteome analyses were compared to identify functionally concordant pathways. The degree of overlap in functional annotations between the two datasets was quantified to assess the biological consistency of mRNA and protein-level changes. Quantitative PCR (qPCR) Total RNA was extracted from muscle samples as described in Sect. 2.4. RNA samples with OD260/280 ratios between 1.8 and 2.0 and intact ribosomal bands (28 S/18S ≈ 2:1) were selected for cDNA synthesis. Reverse transcription was performed using the TransScript First-Strand cDNA Synthesis SuperMix (TransGen Biotech, China) in a 20 µl reaction containing 1000 ng total RNA, 1 µl anchored oligo(dT)18 primer (0.5 µg/µl), 10 µl 2× TS reaction mix, 1 µl RNase inhibitor (RI) enzyme mix, 1 µl gDNA remover, and RNase-free water. The reaction was incubated at 42 °C for 15 min, followed by enzyme inactivation at 85 °C for 5 s. The synthesized cDNA was diluted 1:2 with ddH₂O and stored at − 20 °C. qPCR was performed using the UltraSYBR High Rox Mix (CWBIO, China) in a 20 µl reaction containing 10 µl 2× mix, 0.5 µl forward primer, 0.5 µl reverse primer (Table 1), 1 µl diluted cDNA template, and 8 µl ddH₂O. Amplification was conducted on a ABI QuantStudio 3 with the following program: initial denaturation at 95 °C for 10 min; 40 cycles of 95 °C for 15 s, 60 °C for 30 s, and 72 °C for 30 s; followed by a melting curve analysis (95 °C for 15 s, 60 °C for 1 min, 95 °C for 15 s). Relative expression was calculated using the 2^−ΔΔCt method with GAPDH as the reference gene. Table 1Sequences of the forward and reverse primers used for qPCR analysisGene speciesPrimer pairs (5'―3')Length size (bp)ANKRD23F:GCCCGTAACCCTAGTTTGCT183R:GACAGAGTGTGTGGGGTGAGARG2F:GGGCTGCCGTCTAAAAGACT139R:CACCGCTCTGTTAACCACCTDUSP13F:ATCGATGGGTCTTCCCCAAC117R:CTCGGGACAGGAAGTAAGGCLDHBF:CCTGGTGCAGAGAAACGTGA144R:TGCTTGGGCAGTCCACTTAGMEOX1F:TGGAAACGTGTGAAGGGAGG165R:GAGGGTGAAAGTGGGACTGGMYOCF:ATGGCGTGTGGATGAGAGAC119R:CTGGCTGATGCGGTCATACTPLAGL1F:CGGTTCTACCTCATACCCGC181R:CAGCTCCTTGGGCAGGTTTAPRIMPOLF:CCAGCCCTTCAGTAGGAACC184R:CCAATGTTCTCACACCAGCGGAPDHF: AAGCTGACTGGTAACCACGG164R: ATTGGAGCGCCAAGATTCCA Table 1Sequences of the forward and reverse primers used for qPCR analysisGene speciesPrimer pairs (5'―3')Length size (bp)ANKRD23F:GCCCGTAACCCTAGTTTGCT183R:GACAGAGTGTGTGGGGTGAGARG2F:GGGCTGCCGTCTAAAAGACT139R:CACCGCTCTGTTAACCACCTDUSP13F:ATCGATGGGTCTTCCCCAAC117R:CTCGGGACAGGAAGTAAGGCLDHBF:CCTGGTGCAGAGAAACGTGA144R:TGCTTGGGCAGTCCACTTAGMEOX1F:TGGAAACGTGTGAAGGGAGG165R:GAGGGTGAAAGTGGGACTGGMYOCF:ATGGCGTGTGGATGAGAGAC119R:CTGGCTGATGCGGTCATACTPLAGL1F:CGGTTCTACCTCATACCCGC181R:CAGCTCCTTGGGCAGGTTTAPRIMPOLF:CCAGCCCTTCAGTAGGAACC184R:CCAATGTTCTCACACCAGCGGAPDHF: AAGCTGACTGGTAACCACGG164R: ATTGGAGCGCCAAGATTCCA Sequences of the forward and reverse primers used for qPCR analysis Statistical analysis Statistical analyses were performed using SPSS version 22.0 (SPSS Inc., Chicago, IL, USA) and R software when appropriate. Prior to statistical testing, data from enzyme activity assays, muscle fiber quantification, and qPCR experiments were evaluated for normality and homogeneity of variance using the Shapiro-Wilk test and Levene’s test, respectively. For transcriptomic and proteomic data, which follow specific distributional assumptions (negative binomial and log-normal distributions, respectively), normalization and statistical modeling were performed using dedicated algorithms (DESeq2 and MaxQuant LFQ) without traditional parametric assumptions. Differences among developmental stages were assessed using one-way ANOVA followed by Tukey’s HSD test for pairwise comparisons. P-values were adjusted using the Bonferroni method for multiple comparisons across all low-dimensional datasets (enzyme activity, muscle fiber quantification, and qPCR). For transcriptomic data, differential gene expression analysis was performed using the DESeq2 package. P-values were adjusted using the Benjamini-Hochberg FDR method, and genes with |log2 fold change| > 1 and adjusted p < 0.05 were considered significantly differentially expressed. For proteomic data, differential protein abundance was assessed using the MaxQuant LFQ statistical framework (permutation-based FDR) as described in Sect. 2.6, with thresholds of |log2 fold change| > 1 and adjusted p < 0.05 (consistent with Sect. 2.5). For qPCR validation, relative gene expression levels were calculated using the 2^−ΔΔCt method. Statistical significance was evaluated using one-way ANOVA followed by Student’s t-test with Bonferroni correction for pairwise comparisons. Results Increase in slow-twitch fibers during skeletal muscle growth in meat rabbits Given the established relationship between muscle fiber type conversion and metabolic pathways, it was hypothesized that muscle fiber conversion occurs during skeletal muscle growth. This hypothesis was supported by observations showing that the proportion of fast-twitch fibers in the gastrocnemius muscle remained consistently higher than that of slow-twitch fibers across different developmental stages (Fig. 1A-B). During the postnatal development of the gastrocnemius muscle, immunofluorescence analysis revealed a shift in the composition ratio of fast to slow muscle fibers. The proportion of fast muscle fibers was highest at the D0 stage, subsequently decreasing and stabilizing at approximately 40% in the following D35 and D70. In contrast, the proportion of slow muscle fibers remained relatively consistent during the initial two stages, but increased to over 20% by D70(Fig. 1B). This change suggests that the development of slow muscle fibers occurs between D35 and D70, indicating that a delayed conversion from fast to slow muscle fibers during gastrocnemius development. Fig. 1Composition analysis of muscle fiber types during gastrocnemius development in Ira rabbits. A Immunofluorescence staining of slow-twitch fibers (MYH7, red fluorescence) and fast-twitch fibers (MYH4, green fluorescence). Nuclei were stained with DAPI (blue fluorescence), and white arrows indicate the corresponding muscle fibers. B Quantitative analysis of the ratio of slow-twitch to fast-twitch fibers in three groups (D0, D35, and D70). C LDH activity in the gastrocnemius at three developmental stages. D SDH activity in the gastrocnemius at three developmental stages. E MDH activity in the gastrocnemius at three developmental stages Composition analysis of muscle fiber types during gastrocnemius development in Ira rabbits. A Immunofluorescence staining of slow-twitch fibers (MYH7, red fluorescence) and fast-twitch fibers (MYH4, green fluorescence). Nuclei were stained with DAPI (blue fluorescence), and white arrows indicate the corresponding muscle fibers. B Quantitative analysis of the ratio of slow-twitch to fast-twitch fibers in three groups (D0, D35, and D70). C LDH activity in the gastrocnemius at three developmental stages. D SDH activity in the gastrocnemius at three developmental stages. E MDH activity in the gastrocnemius at three developmental stages To further explain muscle fiber transformation, classical enzymatic activity markers were used to evaluate changes in the metabolic characteristics of gastrocnemius muscle fibers. The results showed that the activities of LDH and MDH significantly decreased (p < 0.05), whereas SDH activity increased (Fig. 1C-E). Enzyme activity analysis indicated a gradual reduction in glycolytic metabolism and an enhancement of aerobic oxidation. However, differences associated with the more complex steps of aerobic metabolism were not clearly defined. Identification of DEGs associated with muscle fiber growth To elucidate the core genes involved in the transition from fast-twitch to slow-twitch fibers, transcriptome analysis was performed to identify significantly differentially expressed genes. After rigorous data quality control (Fig. S1), DEGs were systematically identified. The volcano plot (Fig. 2A) provides an overview of the distribution of DEGs. A total of 5,466 DEGs were identified in the D35 vs. D0 comparison, including 2,322 upregulated genes and 3,144 downregulated genes. In the D70 vs. D0 comparison, 5,963 DEGs were identified, including 2,703 upregulated and 3,260 downregulated genes. In the D70 vs. D35 comparison, 863 DEGs were upregulated and 859 were downregulated (Fig. 2B-C). Notably, although muscle fiber size increased from D0 to D70, the number of DEGs exhibited an opposite trend. Fig. 2Identification of DEGs through transcriptome analysis. A Volcano plots showing DEGs in the comparisons D35 vs. D0, D70 vs. D0, and D70 vs. D35. B Venn diagrams of DEGs from nine samples across three developmental stages (all DEGs on the left, upregulated DEGs in the middle, and downregulated DEGs on the right). C Numbers of DEGs in each comparison group, including upregulated and downregulated genes. D Heatmap of significant DEGs across all samples from the three developmental stages. E Overview of co-expressed DEGs among D0, D35, and D70 Identification of DEGs through transcriptome analysis. A Volcano plots showing DEGs in the comparisons D35 vs. D0, D70 vs. D0, and D70 vs. D35. B Venn diagrams of DEGs from nine samples across three developmental stages (all DEGs on the left, upregulated DEGs in the middle, and downregulated DEGs on the right). C Numbers of DEGs in each comparison group, including upregulated and downregulated genes. D Heatmap of significant DEGs across all samples from the three developmental stages. E Overview of co-expressed DEGs among D0, D35, and D70 Subsequently, functional analysis was performed on 239 DEGs that were co-expressed across all three comparison groups and associated with muscle structure development and function (Fig. 2B). This analysis identified six functional clusters (Fig. 2D), including muscle system processes, negative regulation of striated muscle cell differentiation, positive regulation of signal transduction, muscle structure development, tissue development, and negative regulation of molecular function (Fig. 2E). Differential expression was observed for nitric oxide synthase 1 neuronal (NOS1), calsequestrin 2 (CASQ2), ankyrin repeat domain 1 (ANKRD1), ankyrin repeat domain 2 (ANKRD2), C-X-C motif chemokine ligand 10 (CXCL10), glypican 3 (GPC3), neural precursor cell expressed developmentally downregulated protein 4 (NEDD4), myogenesis-regulating glycosidase (MYORG), myosin light chain kinase 2 (MYLK2), and enhancer of zeste homolog 2 (EZH2). These genes were involved in biological processes including myocyte growth, myotube differentiation, and the regulation of potassium ion transport (Fig. S2). These ten genes are closely associated with skeletal muscle growth and differentiation and may represent potential candidate genes associated with skeletal muscle development. These genes were designated as potential core regulators based on their significant differential expression, their involvement in muscle development-related biological processes, and their enrichment in pathways associated with muscle structure and metabolism. Specifically, the ten candidate genes were selected according to the following criteria: (1) significant differential expression across developmental stages based on transcriptome analysis (|log2 fold change| > 1 and adjusted p < 0.05); (2) inclusion among the 239 co-expressed DEGs shared across all comparison groups (D35 vs. D0, D70 vs. D0, and D70 vs. D35); and (3) functional annotation indicating involvement in biological processes related to muscle development and differentiation. These genes were therefore considered potential candidate regulators that may be involved in skeletal muscle development. It should be noted that these candidate genes were identified from the transcriptomic dataset and were not required to overlap with DEPs in the proteomic analysis. In the final stage of this analysis, eight representative DEGs were selected for qPCR validation primarily to confirm the reliability of the RNA-seq results, rather than to experimentally validate candidate core regulatory genes. The selected genes included ankyrin repeat domain 23 (ANKRD23), arginase 2 (ARG2), dual specificity phosphatase 13 (DUSP13), lactate dehydrogenase B (LDHB), mesenchyme homeobox 1 (MEOX1), myocilin (MYOC), PLAG1-like zinc finger 1 (PLAGL1), and primase and DNA-directed polymerase (PRIMPOL). qPCR results showed that gene expression patterns across the three growth stages were highly consistent with the transcriptome sequencing data (Fig. 3), indicating that the RNA-seq results reliably reflected gene expression changes across developmental stages. Fig. 3Gene expression levels validated by qPCR. The analyzed genes included ANKRD23, ARG2, DUSP13, LDHB, MEOX1, MYOC, PLAGL1, and PRIMPOL. GAPDH was used as the housekeeping gene. The bar chart represents qPCR results, whereas the line graph represents RNA-seq data Gene expression levels validated by qPCR. The analyzed genes included ANKRD23, ARG2, DUSP13, LDHB, MEOX1, MYOC, PLAGL1, and PRIMPOL. GAPDH was used as the housekeeping gene. The bar chart represents qPCR results, whereas the line graph represents RNA-seq data Proteomic profiling of DEPs associated with muscle fiber transformation Compared with DEGs, changes in protein abundance are more directly associated with the growth status of skeletal muscle. Therefore, proteomic analysis was performed to identify DEPs in the gastrocnemius muscle of Ira rabbits at different developmental stages. The proteomic data met the quality requirements for subsequent analyses (Fig. S3). The numbers of upregulated and downregulated proteins in each comparison group are listed in Table 2. The results showed that the number of downregulated DEPs was substantially higher than that of upregulated DEPs, and this difference was more pronounced than that observed for DEGs (Figs. S4 and S5). Nine DEPs were consistently identified across all comparison groups, including TRAF3 interacting protein 3 (TRAF3IP3), amyloid beta precursor protein binding protein 2 (APBB2), phosphoglycerate kinase 1 (PGK1), transmembrane and coiled-coil domain protein 5 A (TMCO5A), pyruvate kinase M1/2 (PKM), parvalbumin (PVALB), PDZ and LIM domain protein 7 (PDLIM7), phosphorylase kinase γ1 (PHKG1), and phosphorylase b kinase regulatory subunit alpha (PHKA1). The expression levels of these DEPs initially decreased and subsequently increased. Subcellular localization analysis showed that most DEPs were primarily distributed in the nucleus and cytoplasm (Fig. 4A-C). The majority were identified in the D70 vs. D0 group, whereas fewer were detected in the D70 vs. D35 group. These findings suggest that DEPs were more abundant during the early developmental stages in Ira rabbits, which is consistent with the trend observed for DEGs. The nine proteins were selected from the proteomic dataset based on their consistent differential abundance across comparison groups (|log2 fold change| > 1 and adjusted p < 0.05) and their functional association with metabolic pathways related to muscle fiber transformation. Table 2Number of proteins showing significant changes in abundanceComparisonsUpregulatedDownregulatedAllD35 vs D036310811444D70 vs D027811841462D70 vs D3588479567 Table 2Number of proteins showing significant changes in abundanceComparisonsUpregulatedDownregulatedAllD35 vs D036310811444D70 vs D027811841462D70 vs D3588479567 Number of proteins showing significant changes in abundance GO annotation analysis was performed to characterize the biological functions of DEPs during skeletal muscle growth. Overall, the DEPs were mainly associated with cellular processes, metabolic processes, and biological regulation. Quantitative analysis indicated that DEPs in the D35 vs. D0 group were mainly associated with RNA processing, whereas those in the D70 vs. D0 group were primarily involved in protein binding. In contrast, DEPs in the D70 vs. D35 group were mainly related to mitochondrial components (Fig. 4D‒F). These findings suggest that DEPs during the early and middle stages of gastrocnemius development in Ira rabbits are mainly involved in molecular processes such as gene transcription and protein synthesis. In later stages, DEPs appear to participate in various biochemical reactions within muscle tissue. Fig. 4Comparative proteomic analysis of DEPs. A Subcellular localization of DEPs in D35 vs. D0. B Subcellular localization of DEPs in D70 vs. D0. C Subcellular localization of DEPs in D70 vs. D35. D Top 20 enriched GO terms for DEPs in D35 vs. D0. E Top 20 enriched GO terms for DEPs in D70 vs. D0. F Top 20 enriched GO terms for DEPs in D70 vs. D35. Histogram colors range from light to dark, representing increasing statistical significance of enrichment. Darker colors indicate higher statistical significance. Numbers above the bars represent the rich factor, and the y-axis indicates the number of proteins enriched in the corresponding pathway Comparative proteomic analysis of DEPs. A Subcellular localization of DEPs in D35 vs. D0. B Subcellular localization of DEPs in D70 vs. D0. C Subcellular localization of DEPs in D70 vs. D35. D Top 20 enriched GO terms for DEPs in D35 vs. D0. E Top 20 enriched GO terms for DEPs in D70 vs. D0. F Top 20 enriched GO terms for DEPs in D70 vs. D35. Histogram colors range from light to dark, representing increasing statistical significance of enrichment. Darker colors indicate higher statistical significance. Numbers above the bars represent the rich factor, and the y-axis indicates the number of proteins enriched in the corresponding pathway Although both upregulated and downregulated DEPs were enriched in different KEGG pathways, several notable patterns were identified (Fig. 5). GO and KEGG enrichment analyses indicated significant enrichment in carbohydrate metabolism pathways, including butanoate metabolism, propanoate metabolism, and the pentose phosphate pathway. For example, the expression levels of succinic semialdehyde dehydrogenase (ALDH5A1) and 3-hydroxybutyrate dehydrogenase (BDH1) were elevated in the butanoate metabolism pathway. In contrast, the expression levels of medium-chain acyl-CoA ligase/lipoate-activating enzyme (ACSM1), 2-hydroxyglutarate dehydrogenase (L2HGDH), and hydroxymethylglutaryl-CoA synthase (HMGCS) were higher at D35 than at D70 (Fig. S5). Fig. 5KEGG enrichment analysis of DEPs in the rabbit gastrocnemius. Bubble size represents the number of genes involved in each pathway. Larger bubbles indicate a greater number of enriched genes. Bubble color ranges from light to dark, indicating increasing statistical significance. Darker colors correspond to higher enrichment significance. A Bubble plot of KEGG pathways enriched for DEPs in D35 vs. D0. B Bubble plot of KEGG pathways enriched for DEPs in D70 vs. D0. C Bubble plot of KEGG pathways enriched for DEPs in D70 vs. D35. D KEGG pathways enriched for upregulated and downregulated DEPs in D35 vs. D0. E KEGG pathways enriched for upregulated and downregulated DEPs in D70 vs. D0. F KEGG pathways enriched for upregulated and downregulated DEPs in D70 vs. D35 KEGG enrichment analysis of DEPs in the rabbit gastrocnemius. Bubble size represents the number of genes involved in each pathway. Larger bubbles indicate a greater number of enriched genes. Bubble color ranges from light to dark, indicating increasing statistical significance. Darker colors correspond to higher enrichment significance. A Bubble plot of KEGG pathways enriched for DEPs in D35 vs. D0. B Bubble plot of KEGG pathways enriched for DEPs in D70 vs. D0. C Bubble plot of KEGG pathways enriched for DEPs in D70 vs. D35. D KEGG pathways enriched for upregulated and downregulated DEPs in D35 vs. D0. E KEGG pathways enriched for upregulated and downregulated DEPs in D70 vs. D0. F KEGG pathways enriched for upregulated and downregulated DEPs in D70 vs. D35 To further explore the relationship between transcriptional and translational regulation during muscle fiber transformation, an integrative analysis of the transcriptomic and proteomic datasets was performed. Genes that were significantly differentially expressed at both the transcriptomic (|log2 fold change| > 1, adjusted p < 0.05) and proteomic (|log2 fold change| > 1, adjusted p < 0.05) levels were considered overlapping molecules. The correlation between mRNA and protein expression levels was evaluated using Pearson correlation analysis. Although several genes showed consistent expression trends at both levels, the overall overlap between DEGs and DEPs was limited. This phenomenon is commonly observed in multi-omics studies and may be attributed to post-transcriptional regulation, differences in protein stability, translational efficiency, and time delays between mRNA transcription and protein synthesis. Functional enrichment analysis revealed that overlapping molecules were mainly involved in energy metabolism pathways, suggesting that metabolic regulation plays a central role in muscle fiber-type transformation during rabbit skeletal muscle development. Biological differences in muscle fiber transformation revealed by DEGs and DEPs The overlap between DEGs and DEPs may reflect coordinated regulation at the transcriptional and translational levels during muscle development. The correlation between DEGs and DEPs was further investigated through integrated analysis of the two omics datasets. The results showed that the number of associated DEPs and DEGs varied among comparison groups, with the D70 vs. D0 group exhibiting the highest number of associated DEPs/DEGs (Fig. 6A-C). When both proteins and their corresponding mRNAs were detected, correlated DEGs and DEPs were classified into positive and negative correlation groups. Pearson correlation coefficients for significant DEPs/DEGs with consistent expression trends were approximately 0.84, whereas those with opposite expression trends showed lower or inconsistent correlations (Table 3). Among the nine DEPs, TMCO5A, PHKG1, and TRAF3IP3 showed negative correlations between protein and mRNA levels, whereas other DEPs (such as PVALB) exhibited positive correlations. Subsequent cluster analysis revealed that the expression patterns of corresponding DEPs and DEGs were largely consistent across the different comparison groups (Fig. 6D-F). Fig. 6Association analysis between transcriptomic DEGs and proteomic DEPs. A Venn diagram showing the overlap of genes, proteins, DEGs, and DEPs in D35 vs. D0. B Venn diagram showing the overlap of genes, proteins, DEGs, and DEPs in D70 vs. D0. C Venn diagram showing the overlap of genes, proteins, DEGs, and DEPs in D70 vs. D35. D Cluster analysis comparing expression patterns between the proteome and transcriptome in D35 vs. D0. E Cluster analysis comparing expression patterns between the proteome and transcriptome in D70 vs. D0. F Cluster analysis comparing expression patterns between the proteome and transcriptome in D70 vs. D35. Proteomic data are shown on the left, whereas transcriptomic data are shown on the right. Red indicates upregulated expression of DEPs, whereas blue indicates downregulated expression Association analysis between transcriptomic DEGs and proteomic DEPs. A Venn diagram showing the overlap of genes, proteins, DEGs, and DEPs in D35 vs. D0. B Venn diagram showing the overlap of genes, proteins, DEGs, and DEPs in D70 vs. D0. C Venn diagram showing the overlap of genes, proteins, DEGs, and DEPs in D70 vs. D35. D Cluster analysis comparing expression patterns between the proteome and transcriptome in D35 vs. D0. E Cluster analysis comparing expression patterns between the proteome and transcriptome in D70 vs. D0. F Cluster analysis comparing expression patterns between the proteome and transcriptome in D70 vs. D35. Proteomic data are shown on the left, whereas transcriptomic data are shown on the right. Red indicates upregulated expression of DEPs, whereas blue indicates downregulated expression Table 3Pearson correlation analysis between transcriptomic and proteomic dataTypesD35 vs D0D70 vs D0D70 vs D35DEPs_DEGs0.790.750.62DEPs_DEGs_SameTrend0.840.840.84DEPs_DEGs_Opposite-0.23-0.25-0.68DEPs_NDEGs0.220.170.16NDEPs_DEGs0.470.370.24 Table 3Pearson correlation analysis between transcriptomic and proteomic dataTypesD35 vs D0D70 vs D0D70 vs D35DEPs_DEGs0.790.750.62DEPs_DEGs_SameTrend0.840.840.84DEPs_DEGs_Opposite-0.23-0.25-0.68DEPs_NDEGs0.220.170.16NDEPs_DEGs0.470.370.24 Pearson correlation analysis between transcriptomic and proteomic data GO functional annotation revealed that 294, 314, and 39 associated DEPs/DEGs were significantly enriched in metabolic processes in the three comparison groups, respectively (Fig. S6). These results indicate that skeletal muscle metabolic activity in Ira rabbits differed significantly at D35 compared with D0. Subsequently, the metabolic differences between D70 and D35 gradually diminished. Integrated KEGG pathway analysis of the two omics datasets revealed that five pathways, including glycolysis/gluconeogenesis and pyruvate metabolism, were significantly enriched in the D35 vs. D0 group. In contrast, only the starch and sucrose metabolism pathway was significantly enriched in both omics datasets in the D70 vs. D0 group. In the D70 vs. D35 group, thermogenesis and oxidative phosphorylation pathways were significantly enriched (Fig. 7). Notably, glycolysis/gluconeogenesis, starch and sucrose metabolism, and oxidative phosphorylation are key pathways involved in energy metabolism and occur at different metabolic stages. These findings suggest that metabolic differences during development may influence mitochondrial metabolic states in skeletal muscle, thereby affecting overall muscle energy metabolism. Fig. 7KEGG pathways associated with DEGs and DEPs. A KEGG enrichment correlation between transcriptomic and proteomic datasets in D35 vs. D0. B KEGG enrichment correlation between transcriptomic and proteomic datasets in D70 vs. D0. C KEGG enrichment correlation between transcriptomic and proteomic datasets in D70 vs. D35. Red dots represent KEGG pathways significantly enriched in both transcriptomic and proteomic datasets. Green dots represent pathways enriched in either the transcriptome or proteome. Black dots represent pathways not significantly enriched in either dataset. D Five KEGG pathways significantly enriched in both datasets in D35 vs. D0. E One KEGG pathway significantly enriched in both datasets in D70 vs. D0. F Four KEGG pathways significantly enriched in both datasets in D70 vs. D35. The x-axis represents KEGG pathway names, whereas the y-axis indicates the number of enriched proteins (red) or genes (blue) KEGG pathways associated with DEGs and DEPs. A KEGG enrichment correlation between transcriptomic and proteomic datasets in D35 vs. D0. B KEGG enrichment correlation between transcriptomic and proteomic datasets in D70 vs. D0. C KEGG enrichment correlation between transcriptomic and proteomic datasets in D70 vs. D35. Red dots represent KEGG pathways significantly enriched in both transcriptomic and proteomic datasets. Green dots represent pathways enriched in either the transcriptome or proteome. Black dots represent pathways not significantly enriched in either dataset. D Five KEGG pathways significantly enriched in both datasets in D35 vs. D0. E One KEGG pathway significantly enriched in both datasets in D70 vs. D0. F Four KEGG pathways significantly enriched in both datasets in D70 vs. D35. The x-axis represents KEGG pathway names, whereas the y-axis indicates the number of enriched proteins (red) or genes (blue) Discussion Skeletal muscle development is primarily driven by the hypertrophy of muscle fibers during the postnatal stage. This process can be influenced by endogenous and exogenous factors during growth, which may alter the metabolic and contractile properties of muscle fibers [24, 25]. Therefore, skeletal muscle is widely regarded as a highly plastic organ. In general, slow-twitch fibers are rich in mitochondria and exhibit high aerobic oxidative capacity. In contrast, fast-twitch fibers primarily utilize glucose through the glycolytic pathway and are therefore classified as glycolytic muscle fibers [26]. In livestock research focused on meat production, muscle fiber types are strongly associated with meat quality and flavor. Because precise standardization of rabbit samples was not feasible, gastrocnemius muscle samples were collected at three specific developmental stages according to Ira rabbit feeding protocols: birth (D0), weaning (D35), and slaughter (D70). Analysis of myosin heavy chain (MyHC) isoform composition [27] and metabolic enzyme activities [28] indicated a gradual transition from fast-twitch fibers to slow-twitch fibers in the gastrocnemius muscle of Ira rabbits during natural growth. Muscle fiber type and composition during the postnatal stage are influenced by multiple factors, including sex differences, nutrient supply, exercise, and environmental conditions. In addition, the intrinsic molecular mechanisms regulated by genes and proteins have attracted considerable research interest. To identify potential regulatory mechanisms underlying muscle fiber transformation, integrated transcriptomic and proteomic analyses were conducted to examine key genes and proteins involved in this process in Ira rabbits. Overall, transcriptomic changes were primarily observed during the early developmental stages (D0-D35). Interestingly, although the largest number of DEGs was detected during the D0-D35 transition, the most pronounced phenotypic shift in muscle fiber composition occurred between D35 and D70. This apparent discrepancy may reflect the temporal lag between transcriptional regulation and phenotypic manifestation during skeletal muscle development. The D0-D35 stage likely represents a transcriptional reprogramming phase that establishes the molecular framework for subsequent muscle fiber remodeling, whereas the D35-D70 stage corresponds to the functional implementation of these regulatory processes. Therefore, the period between D35 and D70 may represent a critical developmental window for gastrocnemius fiber-type transformation in Ira rabbits. Among the ten DEGs, CASQ2 has been reported to be predominantly expressed in slow-twitch fibers [29]. ANKRD2 has been reported to participate in the regulation of slow-twitch fiber formation [30]. In contrast, MYLK2 is specifically expressed in fast-twitch fibers and is capable of phosphorylating myosin to regulate myosin light chains [31]. EZH2 has been reported to be associated with the differentiation of mouse skeletal muscle cells during the mesenchymal stem cell stage [32]. These findings suggest that the aforementioned genes may directly influence muscle fiber type through their expression patterns or molecular interactions. The remaining DEGs may represent novel candidate genes involved in the regulation of muscle fiber type transformation. Furthermore, pathway enrichment analysis of differentially expressed genes identified the calcium signaling pathway as significantly involved in the present study. This pathway may be affected because changes in Ca²⁺ concentration within muscle fibers can activate downstream signaling cascades, thereby promoting adaptive transformation of muscle fiber types [33]. Specifically, Ca²⁺ binds to calmodulin (CaM) in muscle fibers and subsequently activates calcineurin (CaN). Activated CaN then dephosphorylates the nuclear factor of activated T cells (NFAT). After translocation into the nucleus, NFAT binds to target genes and promotes the expression of genes associated with slow-twitch muscle fibers [34, 35]. Notably, the calcium signaling pathway was enriched in both the D35 vs. D0 and D70 vs. D0 comparisons. This finding suggests that calcium signaling may continuously influence muscle fiber function during skeletal muscle development. Moreover, the glycolysis/gluconeogenesis pathway is widely recognized as a central pathway in cellular energy metabolism. Previous research has shown that DEGs in the psoas major muscle of Dezhou donkeys are significantly enriched in the glycolysis/gluconeogenesis pathway. In that study, PKM was proposed as a potential target gene regulating myofibril composition [36], which is partially consistent with the findings of the present study. Furthermore, hexokinase 2 (HK2), a key enzyme in the glycolytic pathway, was found to be upregulated in both D35 and D70 compared with D0. Metabolic genes involved in the citric acid cycle, including pyruvate dehydrogenase E1 subunit alpha 2 (PDHA2) and dihydrolipoamide dehydrogenase (DLD), were upregulated in the D35 vs. D0 comparison but downregulated in the D70 vs. D0 comparison. Therefore, these DEGs may participate in the remodeling of energy metabolism in the gastrocnemius muscle and promote muscle fiber type conversion through the glycolysis/gluconeogenesis pathway. Several unexpected findings were obtained from the proteomic analysis. For example, the protein expression levels of CASQ2, ANKRD2, and MYLK2 were positively correlated with their corresponding mRNA levels, whereas ANKRD1 showed a negative correlation. Furthermore, the protein expression of six DEGs was not detected in the proteomic analysis. In contrast, nine potential core DEPs were identified and were found to be associated with muscle fiber type transformation. The expression levels of these DEPs were positively correlated with their corresponding mRNA levels, except for TMCO5A, PHKG1, and TRAF3IP3. According to the central dogma of molecular biology, these discrepancies may primarily result from post-transcriptional regulation of mRNA, such as alternative splicing [37]. In contrast to the functional pathways enriched by the DEGs, the DEPs were mainly enriched in pathways related to butanoate metabolism. Nutrient supply during the postnatal stage provides the material basis for skeletal muscle growth and development and can directly influence muscle fiber growth and type. Zhang et al. reported that dietary supplementation with 0.3% butyrate significantly promoted mitochondrial biogenesis and oxidative muscle fiber formation in fattening pigs, ultimately improving meat quality [38]. Additionally, butyrate functions as a histone deacetylase inhibitor. It promotes dissociation of the complex formed between histone deacetylase (HDAC) and myocyte enhancer factor 2 (MEF2), leading to increased MEF2 activity and enhanced promoter activity of slow-twitch fiber genes, thereby promoting their expression [39]. Sodium butyrate can also maintain the acetylation state of forkhead box protein O1 (FoxO1) and increase its phosphorylation level, thereby promoting the expression of slow-twitch myosin heavy chain isoforms [40]. The analysis indicated that DEPs involved in the butanoate metabolic pathway, including ACSM1, L2HGDH, and HMGCS, were upregulated in the D35 vs. D0 comparison but downregulated in the D70 vs. D0 comparison. These changes may be associated with muscle fiber type transformation. Although subtle differences were observed in the functional enrichment pathways between DEGs and DEPs, these differences reflect the complexity of regulatory mechanisms in biological processes in vivo, where gene and protein expression patterns do not always correspond. However, the enrichment results from the integrated analysis were largely consistent with metabolic pathways. Muscle fibers utilize glucose generated through glycolysis. In addition, nutrients such as starch in animal diets can influence energy storage in vivo and further regulate muscle fiber metabolism. Furthermore, variations in dietary composition may alter mitochondrial oxidative activity, thereby influencing the basal metabolism of skeletal muscle [41, 42]. These observations are supported by previous findings showing that PGK1 and PKM, identified among the DEPs, function as key genes in the glycolysis pathway and exert direct effects on meat quality [43]. Recent studies using human subjects and mouse models have helped elucidate that muscle fiber type is associated not only with molecular regulation [44] but also with dietary composition [45]. In conclusion, DEGs and DEPs identified in skeletal muscle across different postnatal growth stages associated with the molecular understanding of muscle development and meat quality traits. These effects are likely mediated through alterations in muscle fiber energy metabolism and an increased proportion of slow-twitch fibers. These findings provide a theoretical basis for improving rabbit meat quality through genetic selection or dietary regulation. It should be noted that the present study primarily identifies correlations between gene/protein expression and muscle fiber transformation, rather than establishing causal relationships. Although our transcriptome-proteome integration provides valuable insights, the analysis remains largely correlation-based and descriptive. Additionally, three biological replicates were included for each developmental stage. In transcriptomic and proteomic studies of livestock tissues, three biological replicates are commonly considered the minimum acceptable number for exploratory analyses. Nevertheless, this relatively small sample size may limit the statistical power and reduce the robustness of the results, potentially increasing the risk of false negatives. To mitigate these limitations, strict statistical thresholds were applied for the identification of differentially expressed genes and proteins, and key findings were further validated using qPCR. Moreover, no functional validation experiments were conducted to confirm the roles of the identified candidates. Future studies employing more advanced network-based integrative approaches and functional experiments are warranted to elucidate causal mechanisms and further clarify the regulatory networks underlying muscle fiber transformation. Conclusion The results indicate that ten core genes and nine key proteins are associated with muscle fiber type transformation in the gastrocnemius muscle of Ira rabbits, implicating energy metabolism pathways such as glycolysis/gluconeogenesis and butanoate metabolism. These findings suggest that modulation of energy metabolism and an increased proportion of slow-twitch fibers may significantly improve rabbit meat tenderness. This provides potential molecular targets and a theoretical basis for improving meat quality. Supplementary Information Supplementary Material 1. Supplementary Material 1.