*2.6. Statistical Analysis*

The experimental design was a randomized complete block design, where each carcass (*n* = 6) served as a block. Each measurement was performed in triplicate. To determine the significant effect by two factors (muscles and aging period), data analysis was performed using the SAS 9.2 program (SAS Institute, Cary, NC, USA) with muscles and aging period as the main effects, using two-way analysis of variance (ANOVA). Analysis of variance was performed on all the variables using the General Linear Model (GLM) procedure. Duncan's multiple range test (*p* < 0.05) was used to determine the significance of the di fferences in the mean values for di fferent samples. Pearson's correlation coe fficients were calculated for variables.

#### **3. Results and Discussions**

#### *3.1. Changes in pH Value*

The changes in pH values of beef muscles during aging are shown in Table 1. There was a considerable variation in pH between muscles. Compared to the IN muscle, the pH values in ST, LT, PM, and BF were lower at 1 day (*p* < 0.05). The di fference in pH may be due to the rate of glycolysis between muscles [28]. Similar results were obtained in some studies of beef and lamb [29,30], indicating that the rate of decrease in the pH of PM accelerated. Aging had no significant impact on pH in PM muscles (*p* > 0.05) from 1 to 14 days. Except for IN and PM, no significant changes were observed (*p* > 0.05) in the pH of the remaining six muscles from 3 to 14 days of aging. In terms of value, the pH of IN showed the highest value (*p* < 0.05) of all the muscles in the whole aging period. IN had more type I muscle fibers and its main characteristics were higher oxidative metabolism and lower glycogen content [31]. Therefore, in this study, we can speculate that lower glycolysis capacity would result in a higher pH of IN muscle, which might influence proteolysis.


**Table 1.** Changes in pH values of di fferent beef muscles during aging.

ST, *semitendinosus*; LT, *longissimus thoracis*; RH, *rhomboideus*; GN, *gastrocnemius*; IN, *infraspinatus*; PM, *psoas major*; BF, *biceps femoris*. a–c Values within a column with different superscript are significantly different (*p* < 0.05). x–z Values within a row with different superscript are significantly different (*p* < 0.05). SEM, standard error of the mean. *n* = 6 in each muscle at each aging period.

#### *3.2. Changes in Myofibril Fragmentation Index (MFI)*

Changes in MFI of di fferent beef muscles during aging are displayed in Table 2. The highest MFI was observed in the RH muscle at 1 day (*p* < 0.05). The ST muscle had the highest MFI from day 7 onwards (*p* < 0.05). Compared with the other muscles, the MFI values of IN and LT were found to be lower from 3 to 11 days (*p* < 0.05). MFI is used as an important indicator of I band rupture and interstitial fibril connection breakage [32]. These results seem to indicate that the degree of rupture of muscle fibers in the I band is greater in ST muscles during aging, resulting in changes in the integrity and solubility of myofibrillar proteins [33,34]. The MFI of each individual muscle increased significantly with the aging period (*p* < 0.05) and this is in agreemen<sup>t</sup> with Li et al. [35]. However, the di fference in MFI change was very small in the later days of the aging period. This might be due to the fact that as the aging period increases, the degree of variation in myofibril breaks becomes smaller [36].


**Table 2.** Changes in the myofibril fragmentation index (MFI) of different beef muscles during aging.

ST, *semitendinosus*; LT, *longissimus thoracis*; RH, *rhomboideus*; GN, *gastrocnemius*; IN, *infraspinatus*; PM, *psoas major*; BF, *biceps femoris*. a–c Values within a column with different superscript are significantly different (*p* < 0.05). x–z Values within a row with different superscript are significantly different (*p* < 0.05). SEM, standard error of the mean. *n* = 6 in each muscle at each aging period.

#### *3.3. Changes in Protein Solubility*

The results of the changes in TPS, SPS, and MPS during aging are shown in Table 3. No significant changes (*p* > 0.05) were evidenced in TPS of RH, GN, and PM muscles and in MPS of IN and BF muscles during aging. In regard to SPS, there were no significant differences in SPS between different muscles at day 11 and 14 (*p* > 0.05). IN muscle exhibited a lower TPS at day 9, 11, and 14 than that at day 7 (*p* < 0.05). LT muscles had the highest TPS and MPS among all muscles at day 14 (*p* < 0.05). The SPS of RH was significantly lower than that of IN and GN at day 7 and 9 (*p* < 0.05). Protein solubility was an important indicator of protein properties, which was attributed to protein denaturation [37]. In this study, the results indicated that IN muscle had a high extent of protein denaturation after 9 days [19]. The solubility of myofibrillar protein increased with the aging period up to day 7 and then did not change significantly. This is related to the characteristics of myofibrillar proteins [38,39]. During aging, myofibrillar protein bonds were weakened and more protein hydrolysis was released, which resulted in higher MPS [40]. However, myofibrillar proteins are gradually unfolded as the aging period increases, exposing hydrophobic groups and resulting in almost no change in MPS [41].


**Table 3.** Changes in protein solubility of different beef muscles during aging.


**Table 3.** *Cont.*

ST, *semitendinosus*; LT, *longissimus thoracis*; RH, *rhomboideus*; GN, *gastrocnemius*; IN, *infraspinatus*; PM, *psoas major*; BF, *biceps femoris*. a–c Values within a column with different superscript are significantly different (*p* < 0.05). x–z Values within a row with different superscript are significantly different (*p* < 0.05). SEM, standard error of the mean. *n* = 6 in each muscle at each aging period.

### *3.4. Pearson Correlations*

Correlation coefficients among pH, MFI, TPS, SPS, and MPS of all muscles during aging are presented in Table 4. MFI, TPS, and MPS are negatively correlated with pH (*p* < 0.01). MFI is positively correlated with MPS (*p* < 0.05), whereas no correlations with TPS and SPS are observed (*p* > 0.05). The correlation coefficients of TPS with MPS are higher than those of SPS, indicating that TPS was affected to a larger extent by the denaturation of MPS than that of SPS. These results sugges<sup>t</sup> that pH is a more important determinant for protein solubility than MFI in this study. Moreover, the correlation can further explain the results of protein solubility and confirm that the variation in protein solubility of different muscles may be due to pH changes.

**Table 4.** Pearson correlation coefficients for pH, MFI, and protein solubility.


MFI, myofibril fragmentation index; TPS, total protein solubility; SPS, sarcoplasmic protein solubility; MPS, myofibrillar protein solubility. \* *p* < 0.05; *\*\* p* < 0.01. *n* = 6 in each muscle at each aging period.

#### *3.5. Changes in Muscle Microstructure*

The changes in the microstructure of muscle (ST, RH, and IN) by TEM are shown in Figure 1. In ST, RH, and IN muscles, the sarcoplasmic reticulum around the sarcomeres can be clearly distinguished and the myofibrils are tightly combined with the visible I-band and A-band and the Z-disk and M-line can be differentiated on day 1 and day 3 (Figure 1(a1,a2,b1,b2,c1,c2)). After day 7, the overall integrity of the myofibrils diminishes (Figure 1(a3,b3,c3)). The M-line located on the centre of the A-band looks vague and the Z-disk and I-band junctions are weakened after day 9. The Z-disk is distorted but undamaged, although some longitudinal splits are evident in RH muscles (Figure 1(b4,b5)). The worst myofibrillar structure is observed in IN muscle, in which the overlapping structure of thick and thin filaments is destroyed, the skeletal muscle structure is severely broken, and the Z-disk is distorted and weakened (Figure 1(c4,c5)). At day 14, IN muscle shows fractures in the Z-disk, as well as fragmentation at the junction of the I-band and Z-disk (Figure 1(c6)). Pan and Yeh [42] found that cracking of muscle fibers and shortening of sarcomere could largely reduce the tenderness of the meat. Aside from mechanical damage, muscle fiber structure was also affected by endogenous proteases during aging [43,44]. Moczkowska et al. [45] had reported that BF muscles were more sensitive to

oxidation than LL muscles and the extent of this phenomenon depends on the type of muscle examined. In this study, TEM results showed that IN muscle had a longer sarcomere length and a greater degree of rupture in the muscle fiber structure, as expected because this muscle has a higher proportion of type I fiber (approximately 79.69%), which predicted that the aging rate would be faster. These results are in accordance with the MFI described above. On the other hand, previous research shows that the concentration of coenzyme Q10, carnosine, and taurine in IN muscle was higher than that in LT muscle [46], implying a higher concentration of functional bioactive compounds in oxidized muscle. These results sugges<sup>t</sup> that oxidized muscle accelerates the improvement of meat tenderness.

**Figure 1.** Transmission electron microscopy (TEM) images of beef muscles during different aging periods. TEM images at original magnifications of ×25,000. (**a1**–**a6**): The microstructure of *semitendinosus* (ST) at day 1, 3, 7, 9, 11, and 14. (**b1**–**b6**): The microstructure of *rhomboideus* (RH) at day 1, 3, 7, 9, 11, and 14. (**c1**–**c6**): The microstructure of *infraspinatus* (IN) at day 1, 3, 7, 9, 11, and 14. Red circles indicate the fracturing zones.
