The animal study protocol was approved by the Animal Welfare Committee of the Anhui Academy of Agricultural Sciences (AAAS2023-01). Eight male newborn Oura-type Tibetan F1 lambs (O) and seven male newborn three-way crossbreed AHO F1 lambs (25% Australian White × 25% Small-tail Han × 50% Oura-type Tibetan) were selected under the same natural grassland (Gonghe County, Qinghai Province, China) grazing conditions. The initial average body weight of O and AHO lambs was 3.85 ± 0.42 kg and 4.32 ± 0.11 kg, respectively, with no significant difference between groups (p > 0.05). All animals were clinically healthy, having received routine vaccinations and antiparasitic treatments prior to the trial. After a 2-week adjustment period (lambs completely transitioned from breast milk to pasture), the lambs were grazed on natural pasture from 07:00 to 18:00 daily, with supplemental concentrate feeding (300 g/lamb/day, concentrate feeding nutritional composition, moisture ≤14%, crude protein ≥15%, crude fiber ≤25%, crude ash ≤12%, calcium 0.5–1.5%, phosphorus ≥0.3%, salt 0.3–2.0%, lysine ≥0.4%), and the fattening phase lasted 5.5 months. The body weight of lambs in both the O and AHO groups was recorded at 5.5 months old.

Upon completion of the 5.5 months fattening period, lambs were slaughtered. Body conformation traits such as body height, body length, chest circumference, and cannon circumference were measured for the O and AHO lambs, as outlined in Yang et al. (14). The traits were determined and recorded by the same person to minimize random errors in this experiment. After an overnight fast with unrestricted water access, the live weight (fasted) of all lambs were measured. All lambs were euthanized by electric shock prior to exsanguination and slaughter. After evisceration, the weight of the carcass and tare were measured, respectively. The dressing percentage was determined by dividing the carcass weight by the fasted live weight. Following the full dissection of the carcass, the organ indices (organ weight relative to fasted live weight) of the heart, liver, spleen, lungs, kidneys, pancreas, and thymus were documented. The loin eye contour between the 12th and 13th ribs on the left side of each carcass was traced onto paper, and the area was measured using a scaled grid. After slaughter, muscle samples are collected and subsequently stored at −80°C until required for analyses. Before determining composition of fatty acids and amino acids of LTL muscle, the samples were removed from the −80°C refrigerator, placed in sealed bags, and slowly thawed at 4°C for 24 h to ensure that the final stable content of metabolites in the samples is detected.

After slaughter, carcasses were immediately transferred to a controlled aging chamber (0–2°C, 85% relative humidity, 0.5 m/s air velocity) for 24 h. During this period, the pH of the LTL muscle was measured at 45 min and 24 h postmortem using a portable pH meter (Testo 205, Testo Instrument Co. Ltd., Germany) supplied with electrodes with internal temperature sensor, with pH resolution of 0.01, and automatic calibration with two-point buffer set standard for pH 4.01 and 7.01. The color of the fresh LTL muscle was measured following a 45-min rest at 4°C. A color difference reader (CR-10, Minolta, Japan) was used to measure three positions on the cut surface of each sample. The illuminant, standard observer and aperture used were: light source (illuminant) - D65 (daylight at noon); degree of observer - 2° and size of measuring port - 8 mm. The average of the three measurements was recorded as a pH (or color) coordinate value for each sample. The percentage cooking loss of the meat pieces, which were after aging, was determined by measuring the weight difference before and after cooking, as described by Jiao et al. (15). The protein and fat content in the samples was determined using AOAC12 methods (2005) (16). The protein content (%) in the LTL muscle was measured using a carbon/nitrogen analyzer, while total lipid content (%) was determined via the ether extraction method (AOAC 920.39).

Amino acids were analyzed using a SYKAM S-433D amino acid analyzer (Germany). A 0.15 g sample of LTL muscle was placed in a hydrolysis tube with 15 mL of 6 M hydrochloric acid and 4 drops of phenol solution, then subjected to a 5-min freeze and vacuum treatment. Nitrogen was introduced, and the mixture underwent hydrolysis for 22 h before cooling to room temperature. The hydrolysate was diluted with 1 mL of ammonium citrate buffer (pH 2.2), collected as filtrate, vacuum-dried, and filtered through a 0.22-μm membrane. The hydrolysate was filtered to maintain a consistent volume in a 50 mL container. Amino acids were eluted with a sodium buffer using an ion-exchange chromatographic column (650–0042, Bodenheim, Germany). The compounds formed upon reaction with ninhydrin were concurrently measured at 570 nm and 440 nm wavelengths. The detailed procedure was conducted according to the Chinese standards for the measurement of amino acid composition within foods (GB 5009.168–2016).

A 3 g meat sample was powdered, homogenized with a chloroform-methanol mixture (2:1, v/v), and the solvent was evaporated. The internal standard, methyl non-adecanoate was added during extraction process. A reflux condenser was attached to the fat extract after adding 8 mL of 2% sodium hydroxide methanol solution for saponification and methyl esterification of fatty acids. N-heptane (3 mL) was added and shaken for 2 min. Subsequently, 2 mL of saturated sodium chloride solution was added, followed by the absorption of the upper layer. Anhydrous sodium sulfate was then introduced and thoroughly mixed. The upper solution was extracted into an injection vial for gas chromatographic analysis using an Agilent 7890A system (California, CA, USA). The experimental setup included a chromatographic column measuring 100 m × 0.25 mm × 0.2 μm. The inlet and detector temperatures were set at 250°C and 280°C, respectively. High-purity helium was used as the carrier gas at a flow rate of 1 mL/min. The injection volume was 1.0 μL with a split ratio of 20:1. The heating program was following the previous study (17). Fatty acids were identified by comparing retention times and quantified based on peak areas. The absolute quantity of each fatty acid was represented as a percentage of the total fatty acid methyl esters. MUFA represent the total of all monounsaturated fatty acids, while PUFA denote the total of all polyunsaturated fatty acids. n-3 PUFA consists of C18 3n-3, C20 3n-3, C20 5n-3, and C22 6n-3, while n-6 PUFA includes C18 2n-6c, C18 3n-6, and C20 3n-6. Nutritional indices, including the index of atherogenicity (IA), index of thrombogenicity (IT), and hypocholesterolemic/hypercholesterolemic ratio (H/H), were used to assess the nutritional quality of fatty acids (18, 19). The indices were computed as follows:

The LC–MS/MS method was employed to quantify IMP, GMP, 6-hypoxanthine, and inosine in muscle samples. Approximately 200 mg of fully ground sample powder was placed in a 2 mL centrifuge tube. An extraction solution (1.5 mL) composed of methanol and water in a 10:90 volume ratio with 0.1% formic acid was subsequently added. The mixture was rotated and mixed, and the sample was ultrasonically extracted for 30 min. Then, 400 μL of n-hexane was added. Centrifuge the tube at 12,000 rpm for 5 min, dilute the middle liquid phase 100-fold, filter through a 0.22 μm organic membrane, and transfer to a sample vial for LC–MS analysis. The liquid phase parameters are as follows: a Thermo HYPERSIL GOLD C18 chromatographic column (2.1 × 100 mm, 3 μm) is used with a column temperature of 35°C. The injection volume is 5 μL and the flow rate is set at 0.3 mL/min. The mobile phase consists of 0.1% formic acid in water (phase A) and acetonitrile (phase B). The elution gradient is as follows: from 0 to 2 min, phase A is 90%; from 2 to 6 min, phase A decreases to 10%; from 6 to 8 min, phase A remains at 10%; from 8 to 8.1 min, phase A increases back to 90%; and from 8.1 to 10 min, phase A is maintained at 90%. The mass spectrometry parameters include an ESI (Turbo Spray) ion source with negative polarity, a spray voltage of −4,500 V, a curtain gas pressure of 30 psi, a collision gas pressure of 9 psi, and a nebulization temperature of 550°C. The Multiple Reaction Monitoring method was used for quantitative acquisition of data from mass spectrometry.

Total RNA was extracted from the LTL muscle tissues using the RNAiso Plus according to the manufacturer’s instructions (Thermo Fisher Scientific). The RNA samples were assessed for integrity, concentration, and purity. Following the RNA-Seq sample preparation kit instructions (Illumina, San Diego, USA), the purified mRNA was fragmented and reverse transcribed to generate the final cDNA library. The libraries were sequenced using the Illumina Novaseq™ 6,000 platform by LC Bio Technology Co., Ltd. in Hangzhou, China. Bioinformatic analysis utilized data produced by the Illumina platform. Gene differential expression analysis was completed by DESeq2 software and edgeR. The genes with the parameter of false discovery rate (FDR) below 0.05 and absolute fold change (FC) ≥ 2 were considered DEGs. Then, all DEGs were mapped to the Gene Ontology (GO) database¹ and Kyoto Encyclopedia of Genes and Genomes database (KEGG)² for recognizing the main biological functions.

RNA-reverse transcription and quantitative PCR were performed as described in our previous publications (13). The primer sequences for the genes analyzed in this study are provided in Supplementary Table S1. The Ct values of target DEGs were normalized based on the Ct values of GAPDH, which were found to be stably expressed in different conditions. Subsequently, the 2^-ΔΔCt method was used to determine the relative mRNA expression of the target DEGs. All trial samples had three biological replicates.

Data were statistically analyzed using SPSS software (Version 23.0, SPSS Inc., Chicago, IL, USA), with results expressed as means ± standard error of the mean (SEM). A Student’s t-test or Mann–Whitney U-test was used to compare the two groups. A p-value below 0.05 signifies a statistically significant difference. All figures were made using the OmicStudio tools,³ GraphPad Prism (Version 9.0, Graph Pad Software Inc., San Diego, CA, USA) and Adobe Photoshop (Adobe Photoshop CC 2018, Inc., San Jose, California, USA).

First, we evaluated and compared the body conformation traits of lambs within two groups at 5.5 months of age, as shown in Table 1. The AHO group exhibited significantly greater live weight, body length, and cannon circumference compared to the O group (p < 0.05). However, the body height and chest circumference were not significantly different across the AHO and O groups (p > 0.05). Carcass traits were assessed post-sacrifice, as detailed in Table 2. The AHO group exhibited a significantly higher carcass weight compare to the O group (p < 0.05), with a tendency for increased tare weight (p = 0.091). There was no significant difference in dressing percentage between the O and AHO groups (p > 0.05). The AHO lamb exhibited a significantly larger LTL muscle loin eye area compared to the O lamb (p < 0.05). For organ weight and indices of lamb, the weight and index of spleen, and the lung weight were higher in the AHO group than those in the O group (p < 0.05). There were no significant differences in the weights and indices of the heart, liver, kidney, pancreas, and thymus between the two groups (p > 0.05). These results showed that AHO lamb showed better body conformation and carcass traits than that of O lamb.

Given that the crossbred AHO show better growth performance than O lamb, we were then headed to see whether the crossbreeding influences the meat quality including meat appearance and physical properties. The characteristics of pH, objective color (a*, b*, and L* values), crude protein, total lipid content, and cooking loss of the LTL muscle were evaluated after the lambs were sacrificed. The meat pH (45 min and 24 h) and the instrumental color values of a*, b*, and L* were not significantly different across the AHO and O groups (p > 0.05). There were no significant differences in cooking loss, crude protein, and total lipid content between the AHO and O meat groups (p > 0.05) (Table 3).

Recognizing meat flavor as a crucial aspect of meat quality and a significant factor in consumer purchasing decisions, we analyzed and compared the fatty acids, amino acids, and ribonucleotides composition and content in lambs from the two groups. First, we detected the LTL muscle fatty acid composition of lamb from the two groups. Table 4 indicates a significant increase in capric acid (C10:0), dihomo-γ-linolenic acid (C20:3n-6), and nervonic acid (C24:1 n-9) levels in the AHO group compared to the O group (p < 0.05). The AHO group exhibited a tendency for reduced oleic acid (C18:1n-9c) levels compared to the O group (p = 0.065). In the AHO group, levels of linoleic acid (C18:2n-6c), α-linolenic acid (C18:3n-3), γ-linolenic acid (C18:3n-6), and eicosadienoic acid (C20:2) tended to be higher compared to the O group, with p-values of 0.089, 0.086, and 0.069, respectively. No significant differences were observed in the levels of other fatty acids between the O and AHO groups (p > 0.05). The AHO group exhibited a tendency for reduced MUFA content compared to the O group (p = 0.056). PUFA and n-6 PUFA levels tended to increase compared to the O group (p = 0.082 and 0.086, respectively). There was no significant difference in n-3 PUFA content between the O and AHO groups (p > 0.05). The PUFA/SFA ratio tended to be higher in the AHO group than in the O group (p = 0.098). The IT value tended to decrease in the AHO group compared to the O group (p = 0.074). The O and AHO groups showed no significant differences in the n-6/n-3 ratio, IA value, and H/H rate (p > 0.05).

Subsequently, we analyzed the amino acid composition of the lambs’ LTL muscle. Results were represented in Table 5. The AHO group exhibited a significant increase in glutamic (Glu) and methionine (Met) levels compared to the O group (p < 0.05). There was no significant difference in the levels of other amino acids between the O and AHO groups (p > 0.05).

We also analyzed the nucleotide content responsible for taste in the LTL muscle of lamb (Table 6). The LTL muscles in the AHO group exhibited a trend towards higher inosine monophosphate (IMP) levels compared to the O group (p = 0.072). There were no significant differences (p > 0.05) in the levels of guanosine monophosphate (GMP), 6-hypoxanthine, and inosine between the O and AHO groups.

In depth, we were intrigued about investigating the underlying molecular process that accounts for the differences in performance and meat quality between the two groups. The transcriptomic analysis showed that a total of 664 DEGs were identified (q < 0.05 and fold change ≥ 2), of which 385 were upregulated and 279 were downregulated (Figure 1A). Figure 1B highlighted the top 50 DEGs with significant upregulation and downregulation between the two groups of lambs, including PPARGC1A, CSRP3, LMCD1, and UCP3.

GO and KEGG pathway enrichment analyses were conducted to clarify the functions of the DEGs. Figure 1C highlighted the top 20 significantly enriched GO terms. The top 5 enriched GO terms were “DNA-binding transcription factor activity,” “negative regulation of apoptotic process,” “protein binding,” “RNA polymerase II cis-regulatory region sequence-specific DNA binding,” and “positive regulation of cell migration.” Terms of “positive regulation of cell migration,” “skeletal muscle cell differentiation,” and “negative regulation of myoblast differentiation,” which associated with muscle development were also involved. Figure 1D showed the top 20 significantly enriched KEGG pathway terms. The top 5 enriched terms were “bile secretion,” “antifolate resistance,” “cytokine-cytokine receptor interaction,” and “ovarian steroidogenesis,” and “steroid hormone biosynthesis.” The “purine metabolism” pathway were also included.

We further summarized all differential GO terms related to skeletal muscle development and their related DEGs, as shown in Figure 2A. A cluster of 20 hub DEGs was identified based on 10 related GO terms. These DEGs include ANKRD1, ANKRD2, ATF3, CSPR3, CXCL9, CXCL10, PPARGC1A, FER1L5, FOS, GJA1, IFRD1, METRNL, MYF5, MYL3, MYMK, NMRK2, PITX2, TNFSF14, LOC121817478, and LOC101116972. These genes were further analyzed using gene-action network analysis using Cytoscape software. After analysis with the network analyzer plugin, 20 DEGs were ranked according to their importance in the DEGs cluster, with genes having a higher degree of correlation among the candidate DEGs represented by a darker hue (Figure 2B). To illustrate the connection between DEGs and pathways, we also annotated the changes in DEGs in the KEGG pathway diagram. Figure 2C summarized the purine metabolism pathway, the hub of 3.5.4.6 (adenosine monophosphate deaminase, AMPD), which is upstream of IMP, showed significantly increased.

To confirm the reliability of transcriptomic results, five hub DEGs were selected for RT-qPCR validation, including CSPR3, ANKRD1, IFRD1, PPARGC1A, and AMPD3. Notably, transcriptome profiling demonstrated that these DEGs were significantly upregulated in the AHO group relative to the O groups. Consistently, the RT-qPCR results also showed that the expression levels of these genes were all significantly increased in the AHO group compared to the O group, confirming the accuracy of the transcriptome results (Figure 3).