Abstract

   Muscle traits are critical determinants of meat quality and
   productivity in sheep, influenced by both breed and anatomical region.
   Hanzhong sheep, an indigenous Chinese breed, are prized for tender,
   low-odor meat; yet, the molecular mechanisms underlying these traits
   remain poorly understood. In this study, we integrated meat quality
   assessment with metabolomic and transcriptomic profiling of the
   longissimus dorsi (HZ-B) and triceps brachii (HZ-T) muscles to
   elucidate biochemical and molecular bases of regional differences. The
   results, derived from metabolomic profiling, demonstrated that the
   muscle tissue of Hanzhong sheep contained abundant proteins (95 kinds)
   and fatty acids (150 kinds). The greater tenderness of HZ-B compared to
   HZ-T was associated with higher levels of dipeptides such as carnosine
   (FC = 1.07) and anserine (FC = 1.04), as well as upregulated expression
   of oxidative fiber-related genes MYH2 (FC = 2.92) and TPM1 (FC = 2.15).
   In contrast, HZ-T showed enrichment of flavor-associated metabolites,
   including acylcarnitines and glutamate, alongside higher expression of
   FBXO32 (FC = 0.35) and MYBPC1 (FC = 0.47), linked to structural
   integrity and muscle contraction. Integrated analysis revealed strong
   associations between metabolite abundance (carnosine/anserine) and key
   genes (FBXO32/GADL1), suggesting coordinated regulation of meat quality
   traits. These findings provide mechanistic insights into the metabolic
   and transcriptomic determinants of muscle quality in Hanzhong sheep,
   offering a foundation for genetic improvement and conservation
   strategies.

   Keywords: Hanzhong sheep, longissimus dorsi muscle, triceps brachii
   muscle, meat quality differences, carnosine, acylcarnitine

1. Introduction

   Lamb meat is an important component of global meat consumption due to
   its high protein content, relatively low fat levels, and distinctive
   flavor profile. These quality attributes directly influence culinary
   performance, sensory characteristics, processing suitability, and
   marketability throughout commercial supply chains [[38]1]. Meat quality
   in livestock is affected by multiple factors, with intra-species
   variation largely attributable to differences in rearing environments,
   post-mortem treatments, and anatomical regions. Such regional
   heterogeneity is recognized as a fundamental biological phenomenon, as
   reflected by substantial divergence in muscle fiber composition and
   metabolic characteristics across distinct anatomical sites in sheep and
   other livestock [[39]2,[40]3]. Contemporary studies have demonstrated
   that muscle fiber typology, structural properties, and intramuscular
   fat deposition are key determinants of regional variation in meat
   quality [[41]4,[42]5]. The influence of anatomical location is
   well-established in ovine research. Anderson et al. [[43]6] reported
   region-specific patterns of intramuscular fat deposition in different
   lamb carcass parts. Lu Dou et al. [[44]7] reported significant
   differences in malondialdehyde content, antioxidant activity, and
   volatile flavor compounds between the quadriceps femoris and
   longissimus dorsi muscles of lambs.

   In recent years, integrated multi-omics approaches have emerged as
   powerful tools in life science research, providing new insights into
   the biochemical basis of complex traits and enabling the identification
   of diagnostic and prognostic biomarkers. In meat science, these
   approaches have been widely applied to investigate muscle physiology
   and meat quality [[45]8]. Transcriptomics allows for the systematic
   characterization of gene expression profiles in muscle tissues, whereas
   untargeted metabolomics provides a comprehensive overview of metabolite
   diversity. Integrating these datasets enables the identification of
   pivotal biomarkers and the construction of regulatory interaction
   networks [[46]9,[47]10]. For instance, Qian et al. [[48]11]
   demonstrated through transcriptomic analysis that the PI3K/AKT
   signaling pathway plays a critical role in maintaining metabolic
   homeostasis in beach lamb during post-mortem storage, thereby
   supporting muscle development. Zhang et al. [[49]12] applied integrated
   transcriptomic and metabolomic analyses to compare meat quality traits
   between Hu sheep and F1 hybrids with Nanqiu sheep, revealing that
   crossbreeding significantly modulates key signaling pathways and gene
   expression. Han et al. [[50]13] further showed that
   transcriptomic-metabolomic profiling can uncover significant effects of
   gender and age on the chemical composition and flavor characteristics
   of Huishan goat meat.

   Hanzhong sheep, a short, thin-tailed breed representing an ancient
   indigenous lineage, is well recognized for its tender meat, uniform fat
   distribution, and bright muscle coloration. Previous studies on this
   breed have largely focused on resource surveys, conservation genetics,
   and nutritional management. Although its desirable meat quality
   traits—such as fine texture and rich flavor—have been well documented
   [[51]14], the molecular mechanisms underlying these traits remain
   poorly understood. In particular, the role of part-specific differences
   in shaping meat quality has not yet been elucidated using multi-omics
   strategies. Therefore, comprehensive characterization of the
   fundamental meat quality attributes of Hanzhong sheep and deciphering
   the molecular basis of region-specific variation through integrated
   omics analyses are essential. Such efforts will provide a foundation
   for genetic resource conservation, breeding improvement, meat quality
   benchmarking, and the sustainable utilization of this heritage breed.

2. Materials and Methods

2.1. Animals and Muscle Samples

   Thirty healthy male Hanzhong sheep were reared under standardized
   environmental and nutritional conditions until seven months of age.
   From this cohort, six animals with comparable body weights were
   randomly selected for slaughter and subsequent sample collection. Prior
   to slaughter, sheep were subjected to a 24 h fasting period, with water
   withheld during the final 12 h. Animals were transported to a
   commercial slaughterhouse and humanely euthanized in accordance with
   established welfare guidelines, specifically by stunning followed by
   exsanguination to ensure rapid loss of consciousness and death without
   distress.

   The longissimus dorsi (HZ-B), a stabilizing muscle primarily composed
   of glycolytic fibers and known for its tenderness, and the triceps
   brachii (HZ-T), a locomotor muscle with distinct fiber composition and
   flavor characteristics [[52]15,[53]16], were sampled from each carcass
   within 30 min postmortem (n = 6). Tissues were trimmed of visible
   connective tissue and divided into aliquots for downstream analyses.
   Histological samples were immersion-fixed in 4% paraformaldehyde (PFA,
   pH 7.4) at 4 °C. Samples designated for biochemical assays were stored
   at −20 °C, while those for omics analyses were snap-frozen in liquid
   nitrogen and stored at −80 °C in an ultra-low temperature freezer.

2.2. Slaughter Performance and Meat Quality Assessment

   Pre-slaughter live weight and body size indices (body length, height,
   cannon circumference, and cross-sectional width) were recorded with
   animals in an upright posture. Slaughter procedures were conducted in
   accordance with the Standards for Livestock and Poultry Slaughter (GB/T
   35892-2018) [[54]17], including sequential removal of the head, hooves,
   skin, hair, and viscera (excluding kidneys). Carcasses were trimmed,
   subjected to lymphadenectomy, rested for 30 min, and weighed. Backfat
   thickness and GR values were measured at the mid-dorsal spine and
   between the 12th and 13th ribs, respectively.

   Muscle pH values were determined at 45 min and 24 h postmortem by
   inserting a portable pH meter probe at a depth of 2 cm into both the
   longissimus dorsi and triceps brachii. Physicochemical indices were
   assessed within 24 h postmortem. Shear force was measured by heating
   samples in a water bath at 80 °C until an internal temperature of 70 °C
   was reached, followed by analysis with a shear force meter. Cooking
   loss was calculated as the percentage reduction in weight after cooking
   samples in a water bath at 85 °C for 20 min. Drip loss was determined
   as the percentage of weight loss after suspending samples at 4 °C for
   24 h [[55]18].

   For texture profile analysis (TPA), samples were cut into cubes (3 cm ×
   3 cm × 1 cm) and tested using a Brookfield CT3 texture analyzer with a
   cylindrical TA5 probe (5 mm diameter). Instrument calibration was
   performed according to the manufacturer’s specifications. Parameters
   were set as follows: pre-test speed 1.0 mm/s, test speed 1.0 mm/s,
   target strain 50%, and two compression cycles. Five indices were
   recorded: hardness, elasticity, cohesiveness, chewiness, and
   resilience.

2.3. Hematoxylin-Eosin (HE) Staining of Muscle Tissue

   Paraformaldehyde-fixed muscle samples were rinsed under running water
   for 30 min, dehydrated in graded ethanol, cleared in xylene, and
   paraffin-embedded. Sections (3–6 µm) were cut, dewaxed, and rehydrated.
   Staining was performed with hematoxylin for 3 min, differentiation
   solution treatment, and eosin staining for 3 min, followed by
   dehydration, clearing, and sealing with neutral resin. Muscle fiber
   morphology was observed under a light microscope. The diameters,
   cross-sectional areas, and densities of 45 randomly selected fibers per
   section were quantified using Image Pro Plus (v6.0).

2.4. Metabolite Extraction and LC-MS Analysis

   Approximately 50 mg of muscle tissue was homogenized in a 2 mL
   centrifuge tube containing 6 mm grinding beads and 400 µL extraction
   solution. Quality control (QC) samples were pre-pared by pooling
   aliquots of all samples processed within the same experimental batch;
   this included the Hanzhong sheep samples analyzed in this study, in
   addition to samples from a parallel investigation on crossbred sheep
   ([56]Figure A1). This pooled QC sample was employed to monitor
   instrumental performance and correct for technical variation during
   data acquisition. Metabolomic profiling was performed on a UHPLC-Q
   Exactive HF-X system (Thermo Fisher Company, Waltham, MA, USA).
   Chromatographic separation was achieved on a C18 column (2.1 × 100 mm,
   1.7 µm) at 40 °C, with a flow rate of 0.3 mL/min. The original data
   were peak aligned, baseline corrected, and matched to databases such as
   HMDB (versions 5.0), Metlin (Version 1.0.0r00), and in-house libraries
   using Progenesis QI software (Waters Corporation, Milford, USA).
   Multivariate analyses, including principal component analysis (PCA) and
   orthogonal partial least squares discriminant analysis (OPLS-DA) models
   were constructed using the ropls package in R (Version 1.6.2).
   Differential metabolites were identified with thresholds of VIP > 1 and
   p < 0.05. Pathway enrichment analysis was performed using Fisher’s
   exact test based on KEGG (kegg_v20230830) pathway annotation combined
   with SciPy.stats (Version1.0.0). The resulting data were analyzed and
   visualized using the online cloud platform at cloud.majorbio.com
   [[57]19,[58]20].

2.5. Total RNA Extraction and Transcriptomic Analysis

   Total RNA was extracted from muscle tissue using a TRIzol-based reagent
   kit (QIAzolLysisReagent, Qiagen, Hilden, Germany). RNA concentration
   and purity were assessed using a NanoDrop 2000 (Thermo Fisher
   Scientific, Waltham, MA, USA) spectrophotometer, while RNA integrity
   was evaluated with an Agilent 5300 Bi-oanalyzer system (Agilent, Santa
   Clara, CA, USA). Only samples with an RNA integrity number (RIN) ≥ 6.5
   were utilized for subsequent library construction. To ensure the
   robustness and reliability of the statistical models against potential
   batch effects, a rigorous permutation test (200 iterations) was
   performed on the OPLS-DA models. Transcriptome libraries were
   constructed using the TruSeq Stranded Total RNA Li-brary Prep Kit
   (Clontech, Mountain View, CA, USA), following the manufacturer’s
   protocol. Paired-end sequencing (150 bp) was performed on an Illumina
   NovaSeq 6000 platform (Illumina, San Diego, CA, USA). Raw sequencing
   reads underwent quality control and adapter trimming using FastQC and
   Trimmo-matic, respectively, and were subsequently aligned to the sheep
   reference genome (GCF_002742125.1) using the STAR aligner. Gene and
   transcript abundance was quantified using RSEM (v1.3.3). Differential
   gene expression analysis was performed with the DESeq2 package
   (v1.42.0), with genes satisfying the thresholds of an adjusted p-value
   (FDR) < 0.05 and |log2(fold change)| ≥ 1 considered significantly
   differentially expressed genes (DEGs). Functional enrichment analysis
   of Gene Ontology (GO, Version 2023.07) terms and Kyoto Encyclopedia of
   Genes and Genomes (KEGG, Version 2023.09) pathways was conducted based
   on the identified differentially expressed genes using GOA Tools
   ([59]https://github.com/tanghaibao/GOatools (accessed on 19 September
   2025)) and KOBAS software ([60]https://www.kobas.co.uk/), respectively,
   with statistical significance determined by Fisher’s exact test. All
   data analyses were performed on the Meiji Online cloud computing
   platform [[61]21,[62]22].

2.6. Comprehensive Analysis of Transcriptome and Metabolome Data

   Pearson correlation analysis was used to evaluate associations between
   DEGs and differential metabolites (DMs). Correlation heatmaps were
   generated, and network visualizations were constructed in Cytoscape
   (Version 3.10.0) with significance set at p < 0.05. KEGG-based
   co-enrichment analyses of DEGs and DMs were performed, and
   significantly enriched pathways were identified (p < 0.05).

2.7. Validation of DEGs via Real-Time Quantitative PCR

   Seven DEGs were selected for validation using RT-qPCR, with GAPDH as
   the reference gene. Each gene was analyzed in triplicate. Reactions
   were conducted as follows: initial denaturation at 95 °C for 30 s;
   amplification at 95 °C for 5 s and 60 °C for 5 s (40 cycles); final
   extension at 95 °C for 15 s, 60 °C for 1 min, and 95 °C for 15 s.
   Relative expression levels were calculated using the 2^−ΔΔCt method and
   statistically analyzed in GraphPad Prism 9. The detailed information of
   the primers can be found in [63]Appendix A.

2.8. Statistics and Analysis of Data

   Data were presented as mean ± standard deviation (SD). Statistical
   differences between groups were assessed using one-way analysis of
   variance (ANOVA), followed by appropriate post hoc tests when
   necessary. A significance threshold of p < 0.05 was applied for all
   analyses. Graphical representation and statistical processing were
   conducted using GraphPad Prism 9 (Version 9.5.0, USA).

3. Results

3.1. Hanzhong Sheep Exhibit Small Body Size, Slow Growth and Superior
Tenderness

   Hanzhong sheep in this study presented a small body conformation, with
   a body height of 55.67 ± 1.53 cm, body length of 59.00 ± 4.00 cm, chest
   circumference of 77.00 ± 6.56 cm, cross-sectional width of 7.10 ± 0.17
   cm, and cannon bone circumference of 13.67 ± 2.09 cm. The mean live
   body weight was 25.07 ± 0.50 kg and the mean carcass weight was 11.34 ±
   0.45 kg, corresponding to a slaughter rate of 45.23%. Backfat thickness
   and GR value were 1.07 ± 0.46 mm and 0.74 ± 0.46 mm, respectively
   ([64]Figure 1A). Comparison of meat physical properties revealed clear
   intermuscular differences. The longissimus dorsi (HZ-B) showed
   significantly lower shear force (54.75 ± 2.64 N), hardness (143.83 ±
   3.72 g), and chewiness (3.89 ± 0.05 mJ) than the triceps brachii (HZ-T:
   67.91 ± 2.21 N, 151.40 ± 0.58 g, and 4.10 ± 0.11 mJ, respectively) (p <
   0.05), indicating superior tenderness and a more favorable texture
   profile in HZ-B. No significant differences were detected between the
   two muscles for cooking loss, pH, springiness, or cohesiveness (p >
   0.05) ([65]Figure 1B).

Figure 1.

   [66]Figure 1
   [67]Open in a new tab

   Growth performance and meat physicochemical characteristics of Hanzhong
   sheep. (A) Body size indices; (B) comparison of meat quality parameters
   between HZ-B and HZ-T; and (C) muscle fiber morphological
   characteristics (values are mean ± SD, n = 6; * 0.01 < p ≤ 0.05, **
   0.001 < p ≤ 0.01, *** p ≤ 0.001); “ns” denotes not statistically
   significant.

   Histomorphometric analysis demonstrated marked regional variation in
   muscle fiber architecture. The longissimus dorsi exhibited a 29.81%
   smaller mean fiber cross-sectional area and a 12.58% smaller mean fiber
   diameter compared with the triceps brachii (p < 0.01), alongside a
   29.02% higher fiber density (p < 0.001) ([68]Figure 1C). Collectively,
   these data indicate substantial intermuscular heterogeneity in Hanzhong
   sheep, with the longissimus dorsi characterized by finer fiber
   morphology and greater tenderness relative to the triceps brachii.

3.2. Muscle Metabolome of Hanzhong Sheep Is Compositionally Rich and
Regionally Distinct

   Untargeted metabolomic profiling detected a total of 709 metabolites
   across the longissimus dorsi and triceps brachii samples: 707
   metabolites were found in HZ-B and 702 in HZ-T, with 700 metabolites
   shared by both muscle types ([69]Figure 2A). Although the two muscles
   shared broadly similar metabolite categories, their relative abundances
   differed substantially. Principal component analysis (PCA) revealed a
   clear separation between the metabolomic profiles of HZ-B and HZ-T
   ([70]Figure 2B), indicating distinct metabolic signatures between these
   anatomical regions; QC samples clustered tightly, supporting the
   stability and reproducibility of the dataset.

Figure 2.

   [71]Figure 2
   [72]Open in a new tab

   Overview of muscle metabolite characteristics in Hanzhong sheep. (A)
   Venn diagram of metabolites detected in HZ-B and HZ-T; (B) PCA score
   plot of HZ-B, HZ-T and QC samples; (C) HMDB classification and relative
   composition of detected metabolites.

   Shared metabolites were classified into 21 categories (20 of which were
   annotated). The largest class was amino acids, peptides and their
   analogs (95 species, 16.3%), followed by fatty acid esters (58 species,
   9.95%) and glycerophospholipids (52 species, 8.92%). Other represented
   classes included fatty acids and conjugates (40 species), carbohydrates
   (30 species), carbonyl compounds, glycerophosphorylethanolamines,
   monoterpenes, amines, purines and derivatives, heterotrimeric peptides,
   linoleic acid derivatives, and a range of other metabolites (totaling
   158 species) ([73]Figure 2C). The prevalence of amino acids, lipid
   species and flavor-related compounds in the metabolome was consistent
   with the previously observed favorable organoleptic properties of
   Hanzhong sheep meat.

3.3. Significant Differences in Metabolite Content Between the Longissimus
Dorsi and Triceps Brachii Muscles

   Significant intermuscular variation in metabolite composition was
   observed between the longissimus dorsi and triceps brachii muscles of
   Hanzhong sheep. Specifically, seven metabolites were uniquely detected
   in the longissimus dorsi muscle—1,3,7-trimethyluric acid,
   PC(PGJ2/17:0), D-fructose-1,6-bisphosphate, pyrogallol, Ala-Leu,
   ozolinone, and diphosphoglucose—whereas two metabolites, Pro-Val and
   isodeoxycholic acid, were uniquely present in the triceps muscle
   ([74]Figure 3A). Several of these metabolites are functionally linked
   to meat quality. For example, 1,3,7-trimethyluric acid, a major
   caffeine metabolite, and pyrrolidine glucoside, a polyphenol
   derivative, possess antioxidant activity that may delay lipid oxidation
   during postmortem aging, thereby enhancing meat preservation. In
   addition, the accumulation of fructose-1,6-bisphosphate, a key
   glycolytic intermediate, in the longissimus dorsi may contribute to its
   superior tenderness.

Figure 3.

   [75]Figure 3
   [76]Open in a new tab

   DMs profiling and pathway enrichment analysis of HZ-B and HZ-T groups:
   (A) HMDB classification of up-regulated DMs, (B) HMDB classification of
   down-regulated DMs, (C) heatmap of the top 50 DMs between HZ-B and
   HZ-T, (D) VIP plot of DMs, and (E,F) KEGG pathway enrichment of
   upregulated (F) and downregulated DMs.

   A comparative metabolomic analysis identified 163 differential
   metabolites (DMs) between the two muscle types, with 70 upregulated in
   the longissimus dorsi and 93 upregulated in the triceps brachii (using
   the triceps muscle as the reference; [77]Figure A2). To visually
   compare the differential composition of up- and down-regulated
   metabolites between muscle types, we separately categorized them based
   on HMDB classification, as shown in [78]Figure 3A,B. These metabolites
   were classified into 19 categories, with six predominant groups shared
   between muscles: amino acids/peptides/analogs, fatty acid esters,
   carbohydrates, fatty acid conjugates, flavonoid glycosides, and
   glycerophospholipids. Among these, amino acids/peptides/analogs and
   fatty acid esters were most abundant, each contributing more than 10%
   to the total metabolite pool. Notably, the longissimus dorsi exhibited
   significantly higher levels of heteropeptides (e.g., carnosine and
   anserine), amines, and imidazole derivatives, while the triceps brachii
   contained elevated levels of carbonyl compounds, organic sulfonic acids
   and derivatives, and purines ([79]Figure 3A,B). These differences
   likely underlie the distinct tenderness of the longissimus dorsi and
   the unique flavor attributes of the triceps brachii.

   Cluster analysis of the top 50 DMs further supported these findings
   ([80]Figure 3C). The longissimus dorsi was enriched in amino acids and
   heteropeptides such as 1-methylhistidine, carnosine, and anserine, as
   well as glucose metabolism-related products including
   glycerol-3-phosphate and dihydroxyacetone. Conversely, the triceps
   brachii contained significantly higher levels of flavor-related
   compounds, such as acylcarnitines, L-glutamine, L-glutamic acid,
   inosinic acid, oxidized glutathione, and taurine. These results suggest
   that the biochemical basis of meat quality divergence lies in the
   tenderness-associated metabolites of the longissimus dorsi and the
   flavor-related metabolites of the triceps brachii. Variable importance
   in projection (VIP) analysis further highlighted this divergence,
   showing higher levels of citrulline and L-alanyl-L-histidine in the
   longissimus dorsi, whereas L-glutamine, L-glutamic acid, carnosine, and
   taurine were more abundant in the triceps brachii ([81]Figure 3D).

   To further elucidate the biochemical pathways underlying these
   metabolite differences, KEGG enrichment analysis was performed
   ([82]Figure 3E,F). In the longissimus dorsi, differential metabolites
   were significantly enriched in pathways related to histidine
   metabolism, thiamine metabolism, β-alanine metabolism, oxytocin
   signaling, and adipocyte lipolysis, which are primarily associated with
   muscle structure stabilization and energy balance. In contrast, the
   triceps brachii exhibited enrichment in pathways including glutathione
   metabolism, ABC (ATP-binding cassette) transporters, purine metabolism,
   D-amino acid metabolism, FoxO signaling, alanine/aspartate/glutamate
   metabolism, the pentose phosphate pathway, and amyotrophic lateral
   sclerosis-related pathways, which are closely linked to the
   accumulation of flavor precursors. Collectively, these results
   demonstrate that the longissimus dorsi muscle is metabolically
   predisposed to enhanced tenderness and structural stability, whereas
   the triceps brachii is enriched in pathways supporting flavor
   development.

3.4. Transcriptomic Profiling Reveals Gene Expression Differences Between
Muscles

   Gene expression patterns are strongly associated with livestock growth
   and development, and distinct transcriptional profiles are often
   observed across different anatomical regions. To investigate the
   molecular mechanisms underlying the differences between the longissimus
   dorsi (HZ-B) and triceps brachii (HZ-T) muscles in Hanzhong sheep,
   transcriptome sequencing was performed following non-targeted
   metabolomic profiling. A total of 9917 genes were commonly expressed in
   both muscles, with 323 genes uniquely expressed in HZ-B and 447 genes
   uniquely expressed in HZ-T ([83]Figure 4A). Using HZ-T as the
   reference, 185 genes were significantly upregulated and 160 genes were
   significantly downregulated in HZ-B ([84]Figure A2). Cluster analysis
   of differentially expressed genes (DEGs) revealed strong intra-group
   consistency and clear inter-group separation, highlighting substantial
   transcriptional divergence between the two muscle types ([85]Figure
   4B).

Figure 4.

   [86]Figure 4
   [87]Open in a new tab

   Differential gene expression profiles between HZ-B and HZ-T groups: (A)
   Venn diagram of shared and unique expressed genes between HZ-B and
   HZ-T, (B) heatmap of DEGs showing intra-group consistency and
   inter-group separation, (C) multi-dimensional GO-enriched circle plot
   of DEGs, (D,E) KEGG pathway enrichment of DEGs: (D) upregulated in
   HZ-B; (E) upregulated in HZ-T, (F) radar plot of the top 30 DEGs
   between HZ-B and HZ-T, and (G) RT-qPCR validation of representative
   DEGs.

   Functional enrichment analyses were conducted to further explore the
   biological roles of DEGs. Gene Ontology (GO) enrichment showed that
   DEGs were mainly involved in the Biological Process (BP) and Molecular
   Function (MF) categories, including DNA-binding transcription factor
   activity, regulation of biological processes, RNA polymerase
   II-specific transcription, and embryonic forelimb morphogenesis
   ([88]Figure 4C). KEGG pathway analysis demonstrated distinct functional
   enrichment between the two muscle regions. Upregulated DEGs in HZ-B
   were enriched in pathways related to amino acid metabolism and
   oxidative stress regulation, including Th17 cell differentiation,
   glutathione metabolism, MAPK signaling, TNF signaling, β-alanine
   metabolism, and adipocyte lipolysis ([89]Figure 4D). Conversely,
   upregulated DEGs in HZ-T were enriched in pathways associated with
   muscle growth and development, such as the cGMP–PKG signaling pathway,
   Wnt signaling, histidine metabolism, growth hormone synthesis and
   secretion, and the oxytocin signaling pathway ([90]Figure 4E). These
   findings suggest that the longissimus dorsi muscle emphasizes
   homeostasis and metabolic regulation, whereas the triceps brachii
   muscle prioritizes growth and structural development.

   To identify key candidate genes, the top 30 DEGs were screened
   according to expression levels and statistical significance. In HZ-B,
   significant upregulation was observed for MYBPH, MYH2, TPM1, and IGFN1,
   which are involved in myofiber proliferation and differentiation, as
   well as GADL1, a gene related to carnosine synthesis. In contrast, HZ-T
   showed higher expression of MYBPC1, MYOM3, ALDH1A1, and TBX15, which
   maintain myofiber structural integrity, along with atrophy-related
   genes FBXO32 and KLHL38 ([91]Figure 4F). Validation by RT-qPCR
   confirmed consistency with transcriptomic results ([92]Figure 4G).
   Collectively, these transcriptomic findings indicate that distinct sets
   of genes regulating myofiber development, differentiation, and
   structural stability contribute to the observed differences in muscle
   composition and meat quality between anatomical regions in Hanzhong
   sheep.

3.5. Correlation Between DEGs and DMs in Longissimus Dorsi and Triceps
Brachii Muscles

   Gene expression patterns and metabolite abundances jointly underpin
   phenotypic traits. Previous studies in Hanzhong sheep have demonstrated
   marked transcriptomic–metabolomic divergence across distinct anatomical
   muscle regions. To elucidate the molecular basis of phenotypic
   differences between the longissimus dorsi and triceps brachii, we
   integrated transcriptomic (345 DEGs) and metabolomic (168 DMs) datasets
   using Spearman correlation analysis. The analysis focused on the top 30
   most abundant DMs and the top 25 most highly expressed DEGs.

   In the longissimus dorsi, highly expressed structural and regulatory
   genes, including MYBPH, TPM1, and GADL1, exhibited significant positive
   correlations (p < 0.05) with antioxidant dipeptides such as carnosine
   and anserine. Conversely, in the triceps brachii, MYBPC1, TBX15, and
   FBXO32 were negatively correlated with carnosine analogs, but
   positively correlated with metabolites involved in energy and lipid
   metabolism, including inosine, carnitines, and glutamate ([93]Figure
   5A). Correlation network topology analysis further identified TBX15,
   FBXO32, and MYBPH, together with key metabolites such as
   β-alanyl-L-histidine (carnosine) and anserine, as central regulatory
   nodes, highlighting their roles in muscle function and meat quality
   ([94]Figure 5B).

Figure 5.

   [95]Figure 5
   [96]Open in a new tab

   Integrated metabolomic and transcriptomic analyses of longissimus dorsi
   and triceps brachii muscles. (A) Heatmap of Spearman correlations
   between DEGs and DMs (* 0.01 < p ≤ 0.05, ** 0.001 < p ≤ 0.01, *** p ≤
   0.001), (B) correlation network illustrating key regulatory genes and
   metabolites, (C) KEGG pathway enrichment analysis of DEGs and DMs, and
   (D) VIP diagram of DMs in amino acid metabolism and lipid
   metabolism-related pathways.

   Joint KEGG pathway enrichment analysis of DEGs and DMs revealed that
   lipolysis regulation, β-alanine metabolism, histidine metabolism, and
   glutathione metabolism were significantly enriched in both muscle types
   ([97]Figure 5C). Within these pathways, histidine, 3-methylhistidine,
   carnosine, and arachidonic acid were significantly more abundant in the
   longissimus dorsi, whereas glutamate, pyroglutamate, adenosine, and
   oxidized glutathione were significantly less abundant compared with the
   triceps brachii ([98]Figure 5D). These results suggest that
   muscle-specific expression of structural and atrophy-related genes,
   together with differential enrichment of antioxidant dipeptides and
   amino acid metabolites, constitute key determinants of meat quality
   variation between muscle regions.

4. Discussion

   Meat quality in sheep is influenced by a range of intrinsic and
   extrinsic factors, including breed, anatomical region, sex, and age.
   Hanzhong sheep are locally prized for their tender texture, uniform fat
   distribution, and bright muscle color [[99]14]. However, their
   relatively slow growth rate and small frame size impose substantial
   economic limitations. For instance, in this study the average body
   weight of 7-month-old Hanzhong sheep was only 25.07 ± 0.50 kg, markedly
   lower than the 53.32 ± 4.32 kg reported for 6-month-old Hu sheep
   [[100]23]. This discrepancy underscores the suboptimal production
   performance of Hanzhong sheep, which may be linked to inherent genetic
   characteristics, husbandry practices, or environmental factors.
   Addressing these limitations requires a clearer understanding of the
   molecular and metabolic mechanisms governing muscle development and
   meat quality, thereby informing strategies for genetic improvement.

   The metabolic profile of Hanzhong sheep muscle is characterized by a
   high abundance of nitrogenous and lipid compounds, which is consistent
   with its high nutritional value. The overall metabolite composition
   predominantly comprised amino acids, peptides, fatty acids, and
   glycerophospholipids. Furthermore, the metabolites that were
   differentially abundant between muscle types were primarily associated
   with amino acid and peptide derivatives, in addition to various
   lipid-related compounds. These differential metabolites may constitute
   the material basis underlying the quality differences between the
   longissimus dorsi and triceps brachii muscles.

   Muscle fiber structure plays a pivotal role in tenderness, a critical
   determinant of consumer acceptability. Reduced fiber diameter and
   increased fiber density are strongly associated with lower shear force
   and enhanced tenderness [[101]24]. In this study, the longissimus dorsi
   (HZ-B group) displayed superior muscle fiber characteristics compared
   with the triceps brachii (HZ-T group), consistent with previous
   findings on anatomical variation in meat quality [[102]25]. Oxidative
   fibers (Type I/IIa) are generally linked with improved pH stability,
   lower shear force [[103]26], and enhanced water-holding capacity
   [[104]27,[105]28]. Certain imidazole dipeptides, notably carnosine and
   anserine, further contribute to tenderness through their roles in pH
   buffering, antioxidant defense [[106]29], and muscle differentiation
   [[107]30,[108]31]. Dietary modulation of carnosine signaling has been
   shown to increase oxidative fiber composition and improve structural
   properties [[109]31]. In line with this, the enrichment of carnosine
   and anserine in the longissimus dorsi was accompanied by reduced shear
   force and enhanced oxidative gene expression (e.g., MYH2, TPM1),
   suggesting a mechanistic link whereby these metabolites promote
   oxidative fiber development and mitigate structural resistance.

   Flavor is another central attribute of meat quality, with mutton odor
   being a major barrier to consumer acceptance. This odor largely
   originates from branched-chain fatty acids (e.g., MOA, EOA) [[110]32],
   while lipid metabolism contributes flavor precursors that shape meat
   aroma [[111]33,[112]34]. Interestingly, no odor-associated
   branched-chain fatty acids were detected in either muscle type.
   Instead, the triceps brachii exhibited significant enrichment of
   carnitine derivatives, including isovaleryl-L-carnitine and
   2-methylbutyrylcarnitine, suggesting enhanced carnitine-mediated
   β-oxidation and lipid remodeling [[113]35]. Such pathways may expand
   the precursor pool for flavor development. Concurrently, elevated
   glutamate and oxidized glutathione (GSSG) in the triceps brachii are
   consistent with the enhancement of umami flavor [[114]36]. Conversely,
   the enrichment of carnosine in the longissimus dorsi may modulate
   flavor indirectly by preventing lipid peroxidation and aldehyde
   formation through metal ion chelation [[115]37]. Transcriptomic
   evidence further implicates PLIN5 and PIK3R3 in the triceps brachii,
   genes known to regulate lipid metabolism and fatty acid composition
   [[116]38,[117]39], suggesting a potential molecular basis for flavor
   variation across muscle regions.

   Integrated transcriptomic and non-targeted metabolomic analyses
   revealed a significant inverse relationship between FBXO32 gene
   expression and carnosine content, while GADL1 expression showed a
   positive correlation with carnosine levels. FBXO32 encodes an E3
   ubiquitin ligase, which functions as a biomarker for muscle atrophy and
   the deterioration of meat quality [[118]40,[119]41]. Carnosine has been
   shown by Rahman et al. [[120]37] to inhibit skeletal muscle atrophy by
   suppressing the expression of muscle-specific E3 ubiquitin ligases,
   Atrogin-1/MAFbx and MuRF-1, thereby preventing the activation of the
   ubiquitin–proteasome system. In this study, elevated FBXO32 expression
   in the HZ-T group, along with reduced carnosine levels relative to the
   HZ-B group, suggests that carnosine analogs enriched in the HZ-B group
   may inhibit the initiation of muscle atrophy, thereby improving meat
   quality. Significant enrichment of histidine and β-alanine metabolic
   pathways was revealed by KEGG co-enrichment analysis. GADL1, which
   functions as the rate-limiting enzyme in β-alanine synthesis, was found
   to have expression levels that directly regulate the efficiency of
   carnosine precursor supply [[121]42]. A positive correlation between
   GADL1 and carnosine was confirmed in this study, further validating the
   pivotal role of this enzyme in carnosine synthesis. Conversely, the
   significant downregulation of FBXO32, a recognized marker of muscle
   atrophy, suggests a potential role for carnosine in promoting muscle
   structural integrity. Based on its well-documented biological functions
   [[122]43,[123]44], we hypothesize a dual regulatory mechanism whereby
   carnosine may inhibit proteolytic pathways while simultaneously
   stabilizing structural proteins, thereby potentially contributing to
   the observed reduction in FBXO32 expression and the enhancement of
   structural integrity. This proposed mechanism warrants further
   experimental validation.

   This study provides new insights into the metabolic and molecular
   mechanisms underlying muscle heterogeneity in Hanzhong sheep.
   Nonetheless, certain limitations must be acknowledged. First, the
   flavor implications inferred from metabolite abundance were not
   validated by sensory evaluation or electronic tongue analysis. Given
   the complexity of flavor formation and its modulation by cooking,
   direct validation in future studies is necessary. Second, as the
   animals raised under free-range conditions, environmental variables
   such as diet seasonality and exercise may have influenced the metabolic
   profile. While this reflects realistic farming practices, controlled
   trials with standardized feeding and rearing systems will be essential
   to disentangle genetic versus environmental contributions. Addressing
   these limitations will strengthen the translation of
   metabolomic–transcriptomic insights into practical breeding strategies
   for improving meat quality.

5. Conclusions

   Distinct metabolite profiles underlie the superior meat quality,
   tenderness, and low odor intensity of Hanzhong sheep. The longissimus
   dorsi was enriched in 1-methylhistidine, carnosine, anserine, and
   glucose metabolism intermediates, whereas the triceps brachii contained
   higher levels of acylcarnitines, L-glutamine, L-glutamate, inosinic
   acid, and taurine. These differences were closely linked to
   tissue-specific gene expression, including GADL1, MYH2, MYBPC1, MYBPH,
   TPM1, and FBXO32. Cross-breeding Hanzhong ewes with Suffolk or Nanqiu
   rams may enhance growth performance while preserving key
   traits—tenderness, flavor, and low odor—supporting the development of
   new meat sheep breeds adapted to the Qinba Mountain ecosystem.

Abbreviations

   TPA, texture-profile analysis; QC, Quality control sample; UHPLC-Q,
   Ultra-High Performance Liquid Chromatography–Quadrupole; HMDB, Human
   Metabolome Database; PCA, Principal Component Analysis; OPLS-DA,
   Orthogonal Partial Least Squares Discriminant Analysis; DEG,
   Differentially expressed gene; DM, Differential metabolites; MYBPH,
   myosin-binding protein H; TPM1, tropomyosin 1; MYBPC1, myosin-binding
   protein C1; MYH2, myosin heavy chain 2; IGFN1, immunoglobulin-like and
   fibronectin type III domain containing 1; MYOM3, myomesin 3; GADL1,
   glutamate decarboxylase-like 1; FBXO32, F-box protein 32; TBX15, T-box
   transcription factor 15.

Appendix A

Table A1.

   The primer information involved in RT-qPCR.
   Gene Symbol Gene Name Forward Primer (5′ to 3′) Reverse Primer (5′ to
   3′) Amplicon Length (bp) Annealing Temp (°C)
   GAPDH glyceraldehyde-3-phosphate dehydrogenase ACTGGCAAAGTGGACATCGT
   TTCCCGTTCTCTGCCTTGAC 124 58
   MYBPH myosin-binding protein H GGAACCTGGCTGTAGGTGAC
   CGATGGTCTCCTGGATGTGG 110 58
   TPM1 tropomyosin 1 GGCATGAAAGTCATTGAGAGCC CTTCTGAGAGCTCAGCCCG 186 60
   MYBPC1 myosin-binding protein C1 AAAAGCCTCAAGGCGGAAGT
   TCGAATGTGTACACCCGGC 191 58
   MYH2 myosin-binding protein H AAAAGCGCCAGGAGGC TTTGCCAGATTCCTCCTTCTT
   198 57
   CSRNP1 cysteine- and serine-rich nuclear protein 1 ATGCCAGCTGCTTCCTCAAT
   GCAAGAGGACAAGCTGAGGT 124 59
   LMOD1 leiomodin 1 AAGGTAGCCAAGTATCGCCG CTCCAGCTCCTCCATCTCCT 93 59
   TBX15 T-box transcription factor 15 GCTCAGCCCTCTGAAACCTT
   TTGCCTGCCTGCATGACATA 160 59
   [124]Open in a new tab

Table A2.

   Transcriptome quality control data. sample: sample name; raw reads: the
   total number of entries in the original sequencing data; clean bases:
   total sequencing data volume after quality control; total mapped: the
   number of clean reads that can be located on the genome; error rate:
   the average error rate of sequencing bases corresponding to the quality
   control data; Q20, Q30: Q20 and Q30, respectively, refer to the
   percentage of bases with sequencing quality above 99% and 99.9% among
   the total bases; GC content (%): the percentage of the total G and C
   bases corresponding to the quality control data in the total number of
   bases.
   Sample Raw Reads Clean Reads Total Mapped Error Rate (%) Q20 (%) Q30
   (%) GC Content (%)
   HZ-B1 50637628 50280308 49297209 (98.04%) 0.0122 98.65 95.6 51.38
   HZ-B2 50451316 50076994 48893073 (97.64%) 0.0121 98.69 95.71 51.81
   HZ-B3 48301048 47920416 46882259 (97.83%) 0.0122 98.66 95.62 52.65
   HZ-T1 50634370 50266244 49270916 (98.02%) 0.0123 98.6 95.44 51.67
   HZ-T2 49530508 49215030 48152042 (97.84%) 0.0122 98.64 95.57 53.44
   HZ-T3 46321530 45951396 44966118 (97.86%) 0.0123 98.62 95.51 52.05
   [125]Open in a new tab

Figure A1.

   [126]Figure A1
   [127]Open in a new tab

   Muscle tissue PCA and volcano plot: (A) PCA of metabolic products of
   the longissimus dorsi and triceps brachii muscles of Hanzhong sheep and
   their F1 hybrid with Suffolk sheep, and (B) differential metabolite
   volcano plots of the longissimus dorsi and triceps brachii muscles.

Figure A2.

   [128]Figure A2
   [129]Open in a new tab

   The quality control of the transcriptomes of the longissimus dorsi and
   triceps brachii muscles of Hanzhong sheep, the scatter plot of
   differentially expressed genes, and the agarose gel electrophoresis
   patterns of qPCR primers. (A), the horizontal axis represents the
   percentage of the base length of a single gene relative to the total
   base length, with 0 indicating the 5′ end of the gene and 100
   indicating the 3′ end of the gene; the vertical axis represents the
   total number of sequence segments within the corresponding intervals on
   the horizontal axis that are aligned to all genes, (B) Scatter plot of
   differentially expressed genes, with the x-axis and y-axis representing
   the expression levels of genes/transcripts in the two samples
   respectively, and (C) The agarose gel electrophoresis pattern of 2%
   concentration qPCR primers: Using TAE electrophoresis buffer, with a
   voltage of 85 V, and the electrophoresis duration of 45 min.

Author Contributions

   Z.L.: conceptualization, methodology, investigation, formal analysis,
   writing—original draft preparation. M.Z.: writing—review and editing,
   data curation, validation, investigation. W.L.: supervision,
   writing—review and editing, formal analysis. J.L.: resources,
   investigation. L.W.: resources, investigation. S.W.: formal analysis,
   investigation. H.L.: conceptualization, writing—review and editing,
   resources, supervision. T.Z.: conceptualization, methodology,
   writing—review and editing, supervision, funding acquisition. All
   authors have read and agreed to the published version of the
   manuscript.

Institutional Review Board Statement

   All animal experiments were conducted in strict accordance with the
   protocols approved by the Animal Use Committee of the Chinese Ministry
   of Agriculture and adhered to the ARRIVE guidelines. The experimental
   protocol was approved by the Institutional Animal Care and Use
   Committee of Shaanxi University of Technology (Approval ID: 2024032001,
   1 September 2024). Euthanasia procedures followed the AVMA Guidelines
   for the Euthanasia of Animals (2020), and all efforts were made to
   minimize animal suffering.

Informed Consent Statement

   Not applicable.

Data Availability Statement

   The original contributions presented in the study are included in the
   article, further inquiries can be directed to the corresponding
   authors.

Conflicts of Interest

   The authors declare no competing interests or financial relationships
   that could be construed as a potential conflict of interest.

Funding Statement

   This study was supported by Research Project of Shaanxi Science and
   Technology Department (grant number: 2024NC-YBXM-101).

Footnotes

   Disclaimer/Publisher’s Note: The statements, opinions and data
   contained in all publications are solely those of the individual
   author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI
   and/or the editor(s) disclaim responsibility for any injury to people
   or property resulting from any ideas, methods, instructions or products
   referred to in the content.

References