Six 2-month-old healthy male small-tailed Han (STH) sheep, six F1 generation of Dorper and small-tailed Han crossbred sheep (Do × STH sheep), and six F1 generation of Australian white and small-tailed Han crossbred sheep (Au × STH sheep) were obtained from the Jilin Academy of Agricultural Sciences. They were raised over a period of 5 months in independent pens under a constant temperature condition and fed the same diet recommended by the “China meat sheep feed standard.” The sheep were injected with the required medicine to expel parasites, and their feeding conditions were adjusted for a week before the experiments began. The sheep were fed twice a day with concentrates and hay at a ratio of 4:6 and were given free access to water. After 5 months, the sheep were made to fast for 12 h, after which they were stunned with a pistol, hung, bled, eviscerated, and split. The muscular longissimus dorsi (MLD) was sampled to measure the meat quality traits, fatty acid composition, and amino acid composition; the semitendinosus muscle was sampled to detect the composition of volatile compounds. All procedures involving animals, as well as welfare and ethical considerations, were approved by the Committee for the Ethics of Animal Experiments under AWEC 2019A05 on 16 May 2019.

The body weight of the sheep was weighed and recorded after a 12-h fast. After slaughter, the weights of the head, hooves, carcass, longissimus dorsi muscle, tare, liver, heart, lungs, spleen, kidneys, tail fat, and gut fat were measured and recorded. Then, the separated meat and bone were used to measure the net meat weight and calculate the net meat percentage, dressing percentage, and bone/meat ratio.

After slaughter, the muscular longissimus dorsi (MLD) was taken from the left side at the 12th and 13th rib area 45 min postmortem to measure the intramuscular fat (IMF) content and other meat traits. The IMF content was measured using a Soxhlet extractor method in accordance with the Chinese standards for the measurement of fat content within food (GB 5009.3-2016). The sample included 5 g of ground meat, and petroleum ether was used as an organic solvent (Sigma-Aldrich, 32248, St. Louis, MO, USA). The meat color was described using three different color parameters, including brightness (L^*), redness (a^*), and yellowness (b^*), at 1 h and 24 h postmortem, following the method described by Costa et al. (23). A sample of 100 g of meat was hung for 24 h and 48 h under a constant temperature condition to record the remaining weight and calculate the drip loss percentage. A sample of 100 g of meat without adipose and connective tissues was cooked to calculate the cooking percentage following the method described by Mao et al. (24). The share force, moisture content, and pH value (24 h) of the sample were measured following the method described by Mao et al. (24). The measurement of protein content and the analysis of texture were performed by the Jilin Ministry of the Agriculture and Rural Quality Testing Center (Changchun, China). All trial samples had three biological replicates.

A 2-g of the sample from the MLD was homogenized with 2 mL of nonadecanoic acid, an internal standard, and then 10 mL of hydrochloric acid was added to the sample; the mixture was incubated in a water bath pot in a flask at 70°C for 40 min. The hydrolysate was placed in another flask, and 25 mL of methyl ether and petroleum ether were added to the mixture at a ratio of 1:1 (Sigma-Aldrich, 32248, St. Louis, MO, USA). Then, the mixture was concentrated into a powder using a rotary evaporator (Great Wall, R-2050Ex, Zhengzhou, China) at 40°C. The powder was treated with 8 mL of 2% sodium methoxide and 7 mL of 15% boron trifluoride to form fatty acid methyl esters (FAME), and it was extracted with 20 mL of n-heptane to analyze fatty acid composition using a capillary gas chromatograph and mass spectrometer (Agilent 7890A-5975C GC/MS, Santa Clara, CA, USA) with a column (HP-5MS, 60 m × 250 μm × 0.25 μm, Santa Clara, CA, USA). The detailed procedure was conducted according to the Chinese standards for the measurement of fatty acid composition within foods (GB 5009.168-2016) and following the method described by Kalbe et al. (25).

A 300 mg of meat sample from the MLD was homogenized and dissolved in 10 mL of 6 mol/L HCl Sigma-Aldrich, 32248, St Louis, MO, USA) and 0.1% phenol (Sigma) at 110°C for 22 h. The hydrolysate was concentrated into a powder using a rotary evaporator (Great Wall, R-2050Ex, Zhengzhou, China) at 40°C. Then, 2 mL of sodium citrate (Sigma), with a pH value of 2.2, was used to dilute the powder for analyzing amino acid composition using the liquid chromatograph and mass spectrometer (Agilent, 6500 LC/MS). The detailed procedure was conducted according to the Chinese standards for the measurement of amino acid composition within foods (GB 5009.124-2016) and following the method described by Belhaj et al. (26). Norleucine is the internal standard. All trial samples had three biological replicates.

A 1.5 g of ground meat sample from the semitendinosus muscle in a 20 mL sealed extraction flask (Supelco, 57330-U) with 5 mL of NaCl and 50 μL of 0.05 mg/mL trimethylpyridine (Sigma), an internal standard, was subjected to magnetic stirring at 500 rpm speed in an 85°C water bath pot for 40 min. The extraction catheter (Supelco, 57348-U) collected volatile flavor substances from the sample in the extraction flask into the gas chromatograph and mass spectrometer (Agilent 7890A-5975C GC/MS, Santa Clara, CA, USA) with a HP-5MS column (60 m × 250 μm × 0.25 μm, Santa Clara, CA, USA) to analyze the volatile compounds. Helium, a carrier gas, flows at a speed rate of 1.0 mL/min, at an input device temperature of 250°C, at a pressure of 14.87 Pa, and at a split ratio of 1:1. The heating conditions were as follows: 40°C for 3 min, heated to 150°C at a rate of 4°C/min and held for 1 min, further heated to 200°C at a rate of 5°C/min, and then heated to 230°C at a rate of 20°C/min and held for 5 min. The ion source temperature was set at 250°C, the MS transmission line temperature was set at 250°C, and the MASS scanning range was m/z 30 ~ 400. An odor activity value (OAV) was calculated according to the formula, OAV = Compounds concentration/its odor threshold, to reveal the vital volatile compounds in the different sheep breeds. Higher OAV values of compounds play a crucial role in affecting the meat odor.

The total RNA of the sample from the MLD was isolated using the TRIzol reagent following the manufacturer's instruction (Thermo Fisher Scientific), and then, the RNA was reversed into cDNA following the procedure described in a previous study (27). The expression of the PPARG, PDK4, and TMEM273 genes was detected by conducting a qPCR test using the Roche LightCycler^® 480 (Roche Applied Science) device, following the protocol described in a previous study (27), and considering the GAPDH gene as the reference control. The primer sequences were synthesized by GenePharma Company, Shanghai (Table 1). All trial samples had three biological replicates.

The experimental data were described as mean ± SEM and were analyzed using SPSS 17.0 software. The Student's t-test model was used to compare the data of two groups, while the one-way ANOVA model was used to compare the data of multiple groups. The correlation of the IMF content and fatty acid composition in all samples was compared first. Then, the gene expression, the IMF content, and the fatty acid composition in each group were compared using the Pearson model (paired) in SPSS 17.0 software.

The figures were illustrated using GraphPad Prism 6.0 software. Statistical significance levels were set at a p-value of < 0.05. Original images and data are shown in the Supplementary material.

A total of 18 experimental sheep, raised 5 months later, showed different body and carcass weights. The Do × STH offspring and the Au × STH offspring had higher body and carcass weights than the STH sheep (p < 0.05). However, no significant differences were observed between the Do × STH offspring and the Au × STH offspring (p > 0.05) (Figure 1). The crossbred sheep had higher net meat weight than the STH sheep (p < 0.05) (Table 2). In addition, the weight of the hooves, bones, lungs, spleen, kidneys, and gut fat was higher in the crossbred sheep than in the STH sheep, especially, in the Do × STH offspring (p < 0.05). Contrarily, the tail fat weight and the bone/meat ratio in the STH sheep were higher than those in the crossbred sheep (p < 0.05). The weight of the head, tare, liver, heart, and the longissimus dorsi muscle, the net meat weight, and the dressing percentage had no significant changes between the groups (p > 0.05). The results suggest that the crossbred sheep possessed higher meat yield and a faster growth rate compared to the STH sheep during the same feeding period and under the same feeding condition. The details are shown in Supplementary Table 1.

The crossbred sheep had higher IMF content than the STH sheep (p < 0.05), but no significant differences were observed between the Do × STH offspring and Au × STH offspring (p > 0.05) (Figure 2A). The Au × STH offspring had the highest protein content, followed by the STH sheep and the Do × STH offspring (p < 0.05) (Figure 2B). Inosinic acid influences the freshness and taste of meat (28, 29), but no significant changes were observed in the content between the groups (p > 0.05) in this study (Figure 2C). No significant differences were observed in the meat color at 1 h among the groups (p > 0.05). At 24 h, the crossbred sheep had a redder meat color than the STH sheep, and the STH sheep had a brighter meat color than the Au × STH sheep (p < 0.05) (Table 3). No significant changes were observed in drip loss (24 or 48 h) and shear force between the groups (p > 0.05). The Do × STH sheep and STH sheep had higher moisture content and pH values (24 h) than the Au × STH sheep (p < 0.05). The Do × STH sheep had a higher cooking percentage than the STH sheep (p < 0.05). The meat elasticity was higher in the STH sheep than that in the Do × STH sheep (p < 0.05). These results suggested that the Do × STH crossbred sheep group may have higher meat quality traits than the other groups due to their brighter and redder meat color, higher pH values at 24 h, higher IMF content, higher moisture content, and higher cooking percentage. The Au × STH crossbred sheep had higher IMF and protein content, along with a redder meat color; however, the remaining meat traits of the Au × STH crossbred sheep were lower than those observed in the other groups. The details are shown in Supplementary Table 2.

Fatty acids are the main components of triglycerides (TG) and play crucial biological functions in adipogenesis and TG accumulation (30). The fatty acid content and composition in mutton affect the flavor, nutrients, and meat quality (31, 32). Therefore, the fatty acid profiles in the muscular longissimus dorsi (MLD) were detected, and the results demonstrated that the crossbred sheep had a higher content of myristate, palmitic, and margaric acids compared to the STH sheep (p < 0.05) (Table 4). No significant differences were observed between the Do × STH offspring and the Au × STH offspring (p > 0.05). The Do × STH offspring had a higher content of saturated fatty acids than the STH sheep (p < 0.05). However, the STH sheep had a higher content of arachidonic acid than the Do × STH sheep (p < 0.05). No significant differences were observed in palmitoleic acid, stearic acid, oleic acid, or linoleic acid between the groups (p > 0.05). Saturated fatty acids and monounsaturated fatty acids are closely correlated with meat flavor (33); the crossbred sheep may have better meat flavor than the STH sheep. The details are shown in Supplementary Table 3.

The composition of amino acids affects the fatty acid content, meat color, and pH value (34). Additionally, these acids synthesize aromatic aldehydes through the Strecker reaction, which affects the flavor of meat (35). They also possess different tastes; for example, Glu, Arg, Asp, and Gly possess an umami taste, while Gly Ser, Thr, Lys, Pro, and Ala have a sweet taste (36). The amino acid profile was detected to identify the taste between the groups. Interestingly, the STH sheep showed a higher content of amino acids than the crossbred sheep, including Asp, Thr, Ala, Ile, Leu, and Lys (p < 0.05) (Table 5), and these amino acids may affect the taste of meat. In addition, the STH sheep had a higher content of essential amino acids than the Do × STH sheep (p < 0.05). No significant differences were observed in other amino acids between the groups (p > 0.05). The STH sheep have a more abundant amino acid content, which leads to different tastes of meat, compared to the crossbred sheep. The details are shown in Supplementary Table 4.

The content and composition of volatile compounds influence the flavor of mutton (37). To verify whether the crossbred sheep had different mutton odors compared to the STH sheep, a sample was taken from the semitendinosus muscle to compare the content and composition of the volatile chemical substances among the groups. A total of 46 volatile compounds were detected in the test, but only 29 volatile compounds had complete data in the groups. The results showed that there were no significant differences in the volatile compounds between the groups (p > 0.05) (Table 6). However, decanal was only detected in the Au × STH sheep, methyl decanoate was only detected in the STH sheep, methyl arachidonate was only detected in the Do × STH sheep, and ethyl palmitoleate was only detected in the crossbred sheep. An odor activity value (OAV) was calculated using the odor threshold (38). The results showed that heptanal, octanal, nonanal, decanal, and dodecanal volatile compounds may provide more abundant flavors in the Au × STH sheep due to a high OAV value. The OAV value of the tetradecanal compound in the STH and Do × STH sheep was more than 1, whereas in the Au × STH sheep, the value was < 1. The high OAV value of 1-octen-3-ol may affect the flavor of mutton of the STH sheep. The volatile compound composition and the OAV value may lead to different flavors of mutton between the groups. The details are shown in Supplementary Table 5.

Many studies have reported that IMF content has a close relationship with fatty acid composition in livestock (39, 40); thus, the relationship was compared in this study. The results showed that the IMF content was positively associated with palmitic acid and oleic acid and was negatively correlated with linoleic acid and arachidonic acid in all samples (p < 0.05) (Figure 3A). The results were as expected. Many studies have demonstrated that the PPARG gene regulates adipogenesis and lipid metabolism (41). Our previous experiments found that the PDK4 gene affects fat accumulation (42). The PPARG and PDK4 gene expression in the MLD was detected by conducting the qPCR test. The qPCR test results showed no significant differences in the PDK4 and PPARG gene expression between the groups (Table 7) (p > 0.05). The correlation coefficient of the gene expression and IMF content or fatty acid composition were compared. The results showed that the PPARG and PDK4 gene expression had no association with the IMF content and fatty acid content in the STH sheep (p > 0.05); in the Au × STH offspring, the correlation trends were similar to those in the STH sheep (p > 0.05). Furthermore, the PDK4 gene expression was positively correlated with the IMF content and was negatively correlated with the linoleic acid content in the Do × STH offspring (p < 0.05) (Figure 3B). Our previous experiments found that the TMEM273 gene was negatively regulated by the PDK4 gene and affected lipid metabolism. The correlation of the TMEM273 gene expression in the MLD with the IMF content or fatty acid composition was compared, and the results showed that the TMEM273 gene expression had no association with the IMF content (R = −0.783, P = 0.066) in the Do × STH offspring (Table 8). However, the TMEM273 gene expression had a positive association with the linoleic acid and arachidonic acid contents and a negative correlation with the oleic acid and palmitic acid contents in the Do × STH offspring (p < 0.05) (Figure 3C). The details are shown in Supplementary Tables 6–9.