The relationship between myofiber characteristics and meat quality of Chinese Qinchuan and Luxi cattle

Objective The objectives of this study were to explore the expression patterns of myosin heavy chain (MyHC) genes of different skeletal muscles from Chinese cattle, and to investigate the relationship between myofiber characteristics and meat quality of M. longissimus lumborum (LL), M. psoas major (PM), and M. semimembranosus (SM) from Chinese Luxi and Qinchuan cattle. Methods Three major muscles including LL, PM, and SM from Chinese Luxi cattle and Chinese Qinchuan cattle were used in this study. The myofiber characteristics were measured by histochemical analysis. The MyHC isoforms expression was evaluated by real-time quantitative polymerase chain reaction. Quality traits including pH value, meat color, cooking loss, Warner-Bratzler shear force (WBSF) and sarcomere length were determined at day 5 postmortem. Results PM muscle had higher pH value, a* value, sarcomere length and lower WBSF value compared to LL and SM muscles (p<0.05). Numbers of type I myofiber and the relative expression of MyHC I mRNA in PM muscle were higher than those of LL and SM muscles (p<0.05). Myofiber diameter of PM muscle was lower than that of LL and SM muscles, regardless of myofiber types (p<0.05). Conclusion According to the stepwise linear regression analyses, tenderness was influenced by myofiber characteristics in all three examined muscles. Tenderness of beef muscles from Qinchuan and Luxi cattle could be improved by increasing numbers of type I myofiber.


INTRODUCTION
Breed is an important factor that can influence the characteristics of meat products. Luxi and Qinchuan cattle are two famous indigenous Chinese breeds widespread throughout the central area of China. Many local consumers believe that beef from native cattle pos sess better meat quality than imported beef. Therefore, higher price has been considered for beef from Luxi and Qinchuan cattle in central China.
Meat quality traits including tenderness, color and water holding capacity (WHC) are always crucial to consumers and consequently to the beef industry [1]. It is well known that beef quality of different muscle sources is highly variable, and numerous factors can be responsible for the variations of these quality characteristics [25].
Myofiber characteristics largely affect meat quality because muscle fiber is a primary component of skeletal muscle. Myofiber can be categorized by their morphological traits, contractile and metabolic properties [6]. Based on the histochemical reaction for myosin adenosine triphosphatase (mATPase), myofibers have been divided into three types, in cluding slowoxidative or type I, fast oxidoglycolytic or type IIa, and fast glycolytic IIb [7]. The distribution of myosin heavy chain (MyHC) based myofiber typing is another useful method for classification of myofibers [8]. There are eight isoforms of MyHC in mammalian skeletal muscles, but only MyHC I, MyHC IIa, and MyHC IIx were generally expressed in adult bovine skeletal muscle [9]. Previous studies have outlined that the expression patterns of MyHC genes varied between bovine muscles and cattle breeds [1012]. Thus, further research is needed to elucidate the mechanism on MyHC genes expression of different skeletal muscles from Chinese native cattle.
Several studies have investigated the relationship between myofiber characteristics and meat quality of individual beef muscles [4,13,14]. However, the magnitude of their influ ences on beef quality is still controversial. Moreover, little information is available on the relationship between myofi ber characteristics and meat quality of Chinese indigenous cattle. Therefore, the objectives of this study were to explore the expression patterns of MyHC genes of different skeletal muscles from Chinese Luxi and Qinchuan cattle, and to in vestigate the relationship between myofiber characteristics and meat quality of individual muscles from these cattle.

Sample preparation
A total of 20 Luxi and Qinchuan bulls were randomly se lected at a commercial abattoir. Ten animals from each breed were slaughtered at 24 months of age, according to the Na tional Standard of China (GB/T 194772004). Carcasses were not electrically stimulated. Within 45 min postmortem, two pieces of 5 g samples from three major muscles includ ing longissimus lumborum (LL), psoas major (PM), and semimembranosus (SM) were taken and frozen using liquid nitrogen for histochemical analysis and MyHC gene expres sion. At 48 h postmortem, all the muscles were collected from the carcasses, individually vacuum packaged and trans ferred on ice to the laboratory. Three pieces of 2.54 cm thick steak were obtained from central part of each muscle upon arrival. Quality traits including pH value, meat color, cooking loss, WarnerBratzler shear force (WBSF) and sarcomere length were determined.

Histochemical analyses
Histochemical analyses were carried out by the method de scribed by Brooke and Kaiser [15]. Cross sections (10 μm) were obtained using a cryostat (Cryostar NX50, Thermo Scientific, Waltham, MA, USA) at -20°C. The sections were subsequently used for histochemical analyses of mATPase following alkaline preincubation. This technique consisted of the preparation of three initial solutions: solution A, Tris base (5 g) and CaCl 2 (0.5 g) dissolved in 250 mL distilled water; solution B, the pH value of solution A adjusted to 10.4 by HCL; solution C, the pH value of Solution A adjusted to 9.4 by ATP. Samples were incubated in solution B for 5 min and solution C for 30 min. The next steps consisted of washing samples three times with solution A for 2 min each; applica tion of Co(NO 3 ) 2 on the cross section for 5 min; rinsing of samples with distilled water; addition of ammonium sulfide solution (1%) for 30 seconds. Images were captured by a mi croscope (Eclipse E100, Nikon, Tokyo, Japan). The stained sections were examined using an image analysis system (Im agePro plus 6.0, Media Cybernetics, Rockville, MD, USA). Approximately 300 myofibers per sample were counted to analyze the myofiber characteristics. Myofiber type, myofi ber number percentage, myofiber diameter, and myofiber area percentage were determined.

Quantification of mRNA
Total RNA was extracted using RNAiso Plus reagent (TaKaRa, Beijing, China) according to the manufacturer's protocol. Then, genomic DNA was removed from total RNA sample. Subsequently, the purity and integrity of RNA were evaluat ed by optical density under 260 nm and 280 nm wavelength and by 1% agarose gel electrophoresis, respectively. Reverse transcription of mRNA was done with a PrimeScript RT re agent Kit with gDNA Eraser (TaKaRa, China). The cDNA was stored in a refrigerator at -20°C for realtime quantita tive polymerase chain reaction (qRTPCR). Quantification of mRNA was performed with a realtime PCR system (CFX96, BioRad, Hercules, CA, USA) with SYBR Premix Ex Taq (TaKaRa, China). In this study, glyceraldehyde3phosphate dehydrogenase (GAPDH) was selected as the reference gene. The primer sequences are shown in Table 1. Relative gene expression was calculated by 2 -ΔCt according to the method of Zhang et al [9].

Meat quality traits
A portable pH meter with temperature compensation (Sen venGo, Mettler Toledo, Zurich, Switzerland) was used to determine pH value. The pH electrode was calibrated in buf fers at pH 4.00 and 7.00 before use at room temperature. Measurements were taken three times in different locations for each sample. Surface color was measured using a portable colorimeter (SP62, XRite, Grand Rapids, MI, USA) with 4 mm diameter aperture, Illuminant A, 10° observer angle. The instrument was calibrated with a white and a black standard plate before use. Steaks were exposed to the air at room temperature for 30 min blooming before measurements. Each sample was measured at 6 different locations.
The measurement of cooking loss was performed accord ing to Lu et al [16]. Steaks were cooked individually in a water bath at 80°C until the core temperature reached 70°C. During heating, a digital thermometer was inserted into each sample to track the core temperature. The cooked steaks were cooled to the room temperature and stored at 4°C overnight. Cook ing loss was expressed as the percentage of weight loss before and after heating.
WBSF was determined as described by Lu et al [16]. After measuring cooking loss, six columnar samples with the di ameter of 1.27 cm were removed parallel to the myofibers from each steak. Each sample was sheared once perpendicu lar to the fiber orientation using a texture analysis machine (TAXT2i, Stable Micro Systems, Godalming, UK) with a HDP/BSW blade. The average measurement of columnar samples from each steak was the value of WBSF in Newtons (N).
Sarcomere length was measured according to the method of Li et al [17] with minor modifications. Four grams of beef muscle was mixed with 35 mL of 0.25 M sucrose solution for 1 min in a homogenizer (T18, IKA, Staufen, Germany) at low speed. Then a drop of the slurry containing the fiber fragments was transferred on a microscopic slide and covered with a cover slip. The suspension was examined with the oil objective under a phasecontrast microscopy (BX41, Olympus, Tokyo, Japan), and 30 single myofibrils were photographed by an Olympus camera. All the images were analyzed by the software ImagePro Plus (6.0, Media Cy bernetics, USA). Five measurements of sarcomere length were performed at different points on each image. The aver age of 150 measurements was designated as the sarcomere length of one sample.

Statistical analysis
The experimental data were analyzed by the MIXED proce dure of Statistical Analysis System (SAS, Version 9.2, Cary, NC, USA). The MIXED procedure was used with cattle breed, muscle type, and their interaction as fixed factors, and animal as a random factor. Tests of differences between predicted means were applied using the PDIFF statement and differ ences were considered significantly different at p<0.05. To analyze the relationship between myofiber characteristics and meat quality, stepwise multiple linear regression was used (SPSS, Version 18.0, Chicago, IL, USA).

Myofiber characteristics
The myofiber types in three different muscles obtained from Qinchuan and Luxi cattle were divided into type I and type II, based on alkaline mATPase reactions (Figure 1). A clear difference in myofiber type composition was observed among three muscles. PM muscle had a distinguishably higher num ber percentage of myofiber type I compared with LL and SM muscles ( Table 2). This is consistent with Hwang et al [13] and Lang et al [14], who reported a higher proportion of type I myofiber in PM muscle than LL muscle. There were no dif ferences for myofiber number percentage between Qinchuan and Luxi cattle. Ryu et al [18] concluded that muscle loca tion was more important than breed in affecting myofiber type composition, which is in support of our study.
Results of the myofiber diameter are presented in Figure  2. There was a significant interaction effect of cattle breed× muscle type on the diameter of myofiber (p<0.05). Joo et al [1] reported that myofiber diameter varies with species, chrono logical age, nutritional state of the animal, genetic background, and composition of myofibers. As expected, the diameter of myofibers in LL and SM muscles were significantly longer than PM muscle (p<0.05). This finding is in line with Joo et al [4], who also observed smaller myofiber diameter in PM muscle than LL and SM muscles. There was no differ ence between LL and SM muscle for the myofiber diameter, irrespective of cattle breeds. However, PM muscle from Qinchuan cattle exhibited larger myofiber diameter than that of Luxi breed (p<0.05).
Differences in myofiber area percentage are shown in Table  3. Myofiber type, breed and their interaction had significant effect on myofiber area percentage (p<0.05). The area per centage of myofiber did not differ between Qinchuan cattle and Luxi cattle. However, the area percentage of type I myo fiber was significantly lower than that of type II myofiber in both cattle breeds (p<0.05), which is similar with the results of Joo et al [4].

Expression pattern of myosin heavy chain isoforms
As shown in Table 4, the expression patterns of MyHC I, MyHC IIa, and MyHC IIx were compared. The mRNA abun dances of MyHC I were significantly higher than MyHC IIa and MyHC IIx (p<0.05), irrespective of cattle breeds. The ex pression level of MyHC I in Qinchuan cattle was higher than that of Luxi cattle (p<0.05). However, there were no difference in the expression level of MyHC IIa and MyHC IIx mRNA between different cattle breeds (p>0.05). The expressions level of MyHC I in PM muscle was distinctly higher than LL and SM muscles (p<0.05), which is consistent with the his tochemical analyses above. In the present study, MyHC I, MyHC IIa, and MyHC IIx were expressed in bovine LL, PM, and SM muscles, whereas MyHC IIb expression was not de tected. Zhang et al [9] and Oe et al [19] similarly reported the absence of MyHC IIb expression in bovine skeletal mus cles. A previous study concluded that MyHC IIb expression was scarcely found in most cattle breeds except for young Blond d' Aquitaine bulls, and the MyHC IIb transcript was not translated in all beef muscles [11]. However, skeletal mus cles from small rodent animals have been observed to express the four major MyHC isoforms [20]. Therefore, the expression of MyHC IIb gene was musclespecific and speciesspecific.

Meat quality traits
In this study, the main effects of cattle breed and muscle type significantly affected pH values (p<0.05). As shown in Table  5, all the muscles had a normal pH value ranging from 5.46 to 5.68. Steaks from Qinchuan cattle possessed a higher pH value than that of Luxi cattle (p<0.05). In addition, the pH value of PM muscle was higher than that of LL and SM mus cles (p<0.05). This finding is consistent with Liu et al [21], who also reported a higher ultimate pH of PM muscle com pared to LL muscle from Chinese Luxi cattle.
Meat color is an important factor influencing purchase decision. Consumers usually identify brightred colored meat as evidence of freshness and wholesomeness at the point of sale [22]. L* values of three different muscles were only affected by cattle breed (p<0.05). There was no signifi cant interaction effect of breed×muscle for the L* and b* values (p>0.05). As shown in Table 5, steaks from Qinchuan cattle had higher L* values than that of Luxi breed. However, the present study did not observe any variation for the L* values among different muscle types. Kim et al [23] simi larly reported no significant differences in lightness values x,y Different letters indicate significant differences between fiber types for the same cattle breed and muscle type (p < 0.05).  Cooking loss is an important parameter to reflect WHC of beef muscle. There was a significant interaction effect of breed×muscle for cooking loss (p<0.05). SM muscle showed the highest cooking loss compared to LL and PM muscles (Table 5), which is in accordance with the results of Joo et al [4].
Tenderness is one of the most important attributes re garding eating quality [24]. Muscle type, cattle breed and their interactions had significant effect on WBSF value and sarcomere length (p<0.05). As shown in Table 5, the lowest WBSF value along with longest sarcomere length was ob served in PM muscles, which is consistent with the findings of Joo et al [4] and Hwang et al [13]. In addition, the WBSF values of LL muscle was higher than that of PM and SM mus cles (p<0.05). This is partly in agreement with previous studies [25,26]. Variations in the structure, composition, solubility of intramuscular connective tissues and sarcomere length are known to exist between different muscle types and are related to variations in tenderness [5]. x,y Different letters indicate significant differences between fiber types for the same cattle breed and muscle type (p < 0.05).  Different letters indicate significant differences between cattle breeds for the same gene and muscle type.

Relationships between myofiber characteristics and
x,y Different letters indicate significant differences between genes for the same cattle brees and muscle type (p < 0.05).

meat quality traits
It is generally accepted that muscle's metabolic properties af fected by total number of myofibers, crosssectional area of myofibers and myofiber type composition greatly influence meat quality [1]. Considering the variations in meat quality among LL, PM, and SM muscles from Qinchuan and Luxi cattle, the regression analyses were conducted between meat quality parameters and myofiber characteristics. Previous studies reported that muscle pH 24 was nega tively correlated with the percentage of myofiber type IIa [13,27]. If type II myofibers are predominant in the muscle, postmortem glycolysis is rapidly induced, resulting in an accelerated pH decline in the muscle [28]. However, our present study showed no significant correlation between pH values and myofiber characteristics, probably due to the delayed pH measurement.
It is well documented that the a* value and myoglobin con tent increased with a higher proportion of red myofiber, which decreased color stability with a possible shift to a brownish metmyoglobin color [1]. Ryu and Kim [29] reported that muscle with a higher percentage of type I and type IIa myo fiber had lower L* values. However, Hwang et al [13] reported that there was no significant correlation between the a* value and myofiber types, which is similar to the results in this study.
There is evidence that an incompatibility of the mATPase based and MyHCbased myofiber classification may exist [30]. It is difficult to differentiate pure myofiber types such as IIa and IIx from hybrid myofibers by using histochemical reaction intensity. In the present study, the myofiber were classified only based on the contractile property, i.e., slow (type I) and fast myofibers (type II). Thus, the correlations between pH values, meat color, cooking loss and myofiber characteristics were not significant (p>0.05).
Significant correlations were observed between tenderness and myofiber characteristics in all three muscles from both cattle breeds. Ozawa et al [27] and Ryu et al [29] reported positive correlation between myofiber type I and shear force and negative correlation between type IIb and shear force. However, the relationship between myofiber type composi tion and meat tenderness is still under debate. Hwang et al [13] found that myofiber type I was negatively correlated with WBSF. The same correlation was also observed in the present study. To predict the tenderness using myofiber characteris tics, a multiple linear regression analysis was conducted by a stepwise method (Table 6). A linear regression equation for WBSF value was deduced as Y 1 (WBSF value) = 55.45+0.74X 1 (the diameter of myofiber type I)-0.52X 2 (the number per centage of myofiber type I), Y 2 (WBSF value) = 1.62+0.82 X 3 Similarly, a linear regression equation for sarcomere length was deduced as Y 3 (sarcomere length) = 4.34-0.08X 1 (the di ameter of myofiber type I), Y 4 (sarcomere length) = 4.57-0.07X 3 (the diameter of myofiber type II). The diameters of myofiber are the major elements determining the sarcomere length, which can explain 57% and 55% of the model, respec tively. When the diameter of myofiber type I and II increased by 1 unit, the sarcomere length decreased by 0.08 and 0.07, respectively. Therefore, it is possible to quantitatively predict the tenderness using these parameters. Consequently, these findings suggest that beef tenderness could be improved by increasing type I myofiber.

CONCLUSION
A clear difference in meat quality and myofiber type com position was observed between LL, PM, and SM muscles from Chinese Qinchuan and Luxi cattle. Among the muscles, PM showed the highest a* value and lowest WBSF value. In addition, PM muscle had the highest proportion of type I myofiber and abundances of mRNA of MyHC I. SM mus cle exhibited the highest composition of type II myofiber, along with higher WBSF value. The significant differences in myofiber characteristics may influence beef tenderness. Hence, it is suggested that myofiber characteristics of beef muscle could be an important indicator of tenderness.

CONFLICT OF INTEREST
We certify that there is no conflict of interest with any financial organization regarding the material discussed in the manu script.