Analysis of volatile compounds in pork from four different pig breeds using headspace solid‐phase micro‐extraction/gas chromatography–mass spectrometry

Abstract Purpose The volatile compounds that contribute to the flavor of pork are unknown. Therefore, the present study aimed to determine the differences in volatile compounds from pork meats of four different pig breeds using headspace solid‐phase micro‐extraction (HS‐SPME)/gas chromatography–mass spectrometry (GC‐MS). Methods Piglets from four breeds (8/breed) (crossbred Ziwuling Sus scrofa [SUS] and purebreds Bamei pig [BAM], American Yorkshire pig [YOK], and Hezuo pig [HZP]) were selected. Characteristics of meat were measured. HS‐SPME/GC‐MS were used to analyze the volatile compounds of the meats. Results The tenderness, taste, succulence, and broth flavor of the BAM and HZP were good. One hundred and eight volatile compounds with known molecular formulas were identified in BAM, 106 in SUS, 98 in YOK, and 98 in HZP. Sixty‐four common volatile compounds were found in all four breeds. The highest relative amount of volatile compounds was found in the BAM. The compounds which may contribute to the flavor of pork were 3‐methyl‐1‐butanol, 1‐nonanal, octanal, hexanal, 2‐pentyl‐furan, 1‐penten‐3‐one, N‐morpholinomethyl‐isopropyl‐sulfide, methyl butyrate, and (E,E)‐2, 4‐decadienal. Conclusion The volatile compounds in pork belong to several classes, and the highest relative amount of volatile compounds was found in BAM.


| Animals
All animals were fed at a single farm in Zhangye City, Gansu Province, China, in sheds for pig herds with large open spaces, allowing the pigs to freely move around. Sheds were kept dry and clean, with free drinking water and good ventilation. The feeding and management conditions were maintained as consistent as possible. The basic feed was prepared according to the recommendations of the National Research Council (NRC), taking into account the feeding patterns of different pigs. All studies were performed with the approval of the Animal Use and Care Committee of Gansu Agricultural University, China. The feed composition and nutrient levels of basal diets are shown in Table 1.
A Ziwuling S. scrofa male was captured, kept in a local Guihuayuan pig farm, and crossbred with a female of the Gansu

TA B L E 1 Composition and nutrient levels of basal diets
local breed Bamei pig, to produce the F1 generation of SUS. In this study, four breeds of animals were evaluated: SUS (F1 generation crossbred), BAM, YOK, and HZP. In each group, the same paternal line was mated with 2-3 maternal lines of the same breed to obtain piglets. Then, eight weaned piglets were selected from the 2-3 maternal lines of each group. All selected animals were in the same growth conditions, and the weight difference was <0.5 kg (Table 2).
These pigs were bred for 100 days under the same conditions for determining production performances, and then were slaughtered for determination of meat texture. All visible fat was removed during sample preparation. Before analyzing for volatile components, the samples were tested for microbiological populations.

| Taste identification
Meat samples were boiled without sauce, sliced, and placed into a dinner plate. Then, 10 experts in animal by-products assessed tenderness, taste, succulence, and broth flavor. Taste was evaluated using a 10-point method: scores >8.5, 8.5-7.0, and 7.0-6.0 referred to good, intermediate, and poor taste, respectively.

| Meat color
The CR-400 type color difference meter (Hangzhou Ke Sheng Instrument Co., Ltd.) was adopted in the experiment. The flesh color in the eye muscle was determined after slaughter for 45 min, and the difference was judged according to the measured value of L, a, and b.

| PH 1 value
PH 1 value was determined within 45 min after slaughter and was directly performed by puncturing the longissimus dorsi muscle at the last but two and three thoracic vertebra. The procedures were in accordance with the instructions of pH meter (pH210) or digital pH meter (DHS-2F). Holes were punctured in the meat sample using a knife. Then, the electrode was directly inserted into the central puncture hole, at a depth to ensure that the electrode head was completely embedded in the meat sample (1.0-2.0 cm). Then, the pH 1 value was read (precision of 0.01). Normal pH 1 value was 6.0-6.6. If the meat had a pH 1 <5.9, accompanied by gray color and a large amount of exudative fluids, it was judged as a PSE meat. For hybrid swine, the lower limit of normal pH 1 was 5.6 according to the slaughtering circumstances and referring to individual stress-sensitive breeds (such as Pietrain). It is because that the hybrid swine is impatient, shy, and difficult to catch, which results in a long slaughtering duration, and is likely to cause effects on meat quality and result in a low acidity.

| Water loss percentage
Percentage of water loss was determined using the pressure method at room temperature. The longissimus dorsi muscle at the last and the last but one thoracic vertebra was collected within 45 min after the pigs were slaughtered. Circular meat samples (area of 5 cm 2 and thickness of 1 cm) were cut using a circular cutter with a diameter of 2.532 cm, and were weighted. Then, the circular meat sample was sandwiched between two layers of gauze and 18 layers of qualitative filter paper were applied to both sides. The samples were pressurized to 35 kg (stress of 138.8 kPa) for 5 min. After the pressure was removed, meat sample was stripped from the gauze and weighted. The percentage of water loss and water holding capacity was given by: Water loss percentage (%) = [(prepressure weight−post-pressure weight)/pre-pressured weight] × 100%.

| Marbling
Marbling was evaluated using US-made NCCP colorimetric plate (1991 edition). Fresh meat samples were cut from the longissimus dorsi muscle at the thoracolumbar junction (thickness not <1.5 cm). Marbling was evaluated at the same time as meat color using the same testing conditions. The meat sample was scored by comparison with the colorimetric plate: 1, 2, 3, 4, and 5 points referred to trace amount, micro amount, moderate amount, plenty amount, and excessive amount of fat.

| Cooked meat percentage
The greater psoas muscle was collected from the left carcass, from which about 500 g (W 1 ) of meat was cut off, weighted, and labeled.
Then, the meat sample was placed in an aluminum cooker, added with an appropriate amount of cold water, and steamed on a 2,000 electric stove for 45 min after the water started to boil. Subsequently, the meat sample was taken out and hung for 30 min, followed by weighting (W 2 ). Cooked meat percentage (%) = W 2 /W 1 × 100%.

| Area of eye muscle
Parchment paper was firmly attached to the cross section of the thoracolumbar junction of the hot carcass (left side), and the profile of the cross section was depicted using a panicle, which was brought back to the laboratory for measurement using a planometer. Area of the eye muscle (cm 2 ) = Length × width × 0.7.

| Meat tenderness
Loin-eye muscle was collected within 45 min after the pigs were slaughtered, then immersed into a water bath at 75-80°C until the core temperature reached 70°C. Then, the meat sample was taken out and cooled to room temperature. Meat piece with a width of 1.5 cm was cut perpendicular to the muscle fibers. Meat pieces were cut along the muscle fiber direction using a circular cutter (diameter of 1.27 cm). Ten samples were obtained from each animal. Meat tenderness was measured using a C-LM3 digital meat tenderness instrument developed by the Engineering College of the Northeast Agricultural University. The shear forces of the 10 meat pieces were recorded, and their average value was used for analysis as N or kg.

| Sample preparation and extraction procedure
The longissimus muscles at the last and second last ribs were extracted. The meat samples (120 g) from each piglet breeds were grinded, divided into four parts, and placed in sealed glass bottles.
One part was randomly selected for homogenization; the remaining parts were kept at −27°C. The sealed vials were incubated at room temperature for 40 min and then placed in an incubator at 80°C for 40 min for activation (Wang, Wang, Liu, & Chen, 2008). Then, 6 g of activated samples was placed into headspace vials. The syringe plunger was pushed to head out the fiber from the needle, and the fiber was placed into the top space (headspace mode) for extraction for 40 min and then subsequently at room temperature using the SPME technique. The fiber head was retracted, and the needle was withdrawn from the vial. The SPME needle was inserted into the GC-MS inlet, and the syringe plunger was pushed to expose the fiber for thermal desorption, followed by column chromatography analysis during which a manual SPME injector was used. The fiber was retracted, and the needle was removed. For each sample, 100 μm of polydimethylsiloxane (PDMS) fiber and 85 μm of polyacrylate fiber (Supelco, Sigma, St Louis, MI, USA) were used and pooled.

| Gas chromatography-mass spectrometry
Gas chromatography-mass spectrometry was carried out on a Beijing, China). Helium was used as carrier gas in the splitless mode at constant flow of 0.8 ml/min. The inlet and interface temperature was 250°C, and for separation, an initial column temperature of 35°C for 5 min was followed by a gradual increase of 5°C/min to 230°C, held for 8 min. MS conditions were as follows: ion source temperature, 200°C; ionization, electron impact (EI); electron energy, 70ev; mass scanning range: 33-500 atomic mass units (amu).

Identification of compounds was initially carried out by search-
ing the NIST02 spectrum library and referring to literature about the volatile compounds in pork, followed by confirmation using CI-derived (M + 1) quasi-molecular ions. Quantification was carried out by normalizing the area of an ion to the total ion chromatogram.

| Statistical analysis
All statistical analyses were performed using SPSS 13.0 (SPSS Inc., Chicago, IL, USA). Data were expressed as mean ± stand-

| Characteristics of meat
As can be seen from Tables 2 and 3, the carcass length was longer in the BAM and YOK groups compared with the SUS and HZP groups The tenderness, taste, succulence, and broth flavor of the BAM and HZP were good. In the SUS, except that succulence was moderate, the other indicators were good. Meanwhile, all the indicators of the YOK pigs were moderate. There were no differences among the four groups for these characteristics.

Crude protein content was higher in the SUS group compared
with the other groups (all p < 0.05), while intramuscular fat was higher in the YOK (all p < 0.05). All other parameters were similar among the four groups.

| Comparison of volatile compounds in pork meats from all breeds
Over 100 peaks were found in each gas chromatogram (Supporting Information Figure S1). Among them, 108 volatile compounds with known molecular formulas were identified in BAM, the largest number among all breeds, followed by 106 compounds in SUS. In YOK and HZP, only 98 compounds were identified from pork meats. The relative amounts of total volatile compounds with known molecular formulas accounted for 73.020%, 99.957%, 75.877%, and 76.996% of all detected compounds in SUS, BAM, YOK, and HZP, respectively. The highest relative amount of the unidentified compounds was found in the crossbred SUS, followed by YOK and HZP, while the lowest was found in BAM.
Sixty-four common volatile compounds were observed in the various pork meats studied, at different amounts (Table 4).
BAM showed the highest sum of relative amount ( Table 4.

| The relative amounts of other volatile compounds from specific breeds
In total, 34-44 other different volatile compounds were found in the four pork breeds (

| D ISCUSS I ON
There are two main types of compounds responsible for meat flavor: water-soluble components and lipids (Khan, Jo, & Tariq, 2015).
Carbohydrates are derived from the homolytic cleavage of alkoxy radicals in fatty acids. Alcohols are produced from oxidative degradation of fats. Alcohols and carboxylic acids are condensed to form esters, in which oil flavors are dominant (Shahidi, 2001). It was found that lamb flavor was related to heptan-2-one and oct-1-en-3one amounts, while rancid or undesirable flavors were not related to carbonyl compound abundance (Resconi et al., 2010). Miyasaki et al. assessed several fish samples using an electronic nose system and GC-MS with HS-SPME, and showed that some aldehydes and alcohols increased rapidly in jack mackerel and chub mackerel, slowly in skipjack, and slightly in red sea bream and puffer during storage (Miyasaki, Hamaguchi, & Yokoyama, 2011). Likewise, Kaskonienè, Venskutonis, and Ceksteryt (2008) analyzed honey samples of different botanical origins by HS-SPME/GC-MS, and identified 93 compounds. Interestingly, using SPME/GC-MS, it was found that  (Machiels & Istasse, 2003).
Rongchang pork meat contains volatile compounds such as aldehydes, alcohols, ketones, acids, ethers, esters, hydrocarbons, sulfur-containing compounds, amines, nitrogen-containing compounds, furans, oxygen-containing compounds, and benzene and its derivatives (Liu & Sun, 2010). It was suggested that interactions between these compounds in the muscles produced water-soluble and volatile flavor compounds, which together formed the unique flavor of Rongchang pork (Liu & Sun, 2010). It was suggested (Du, 2012) that aldehydes and furans, which are compounds found at high amounts, are the main components of pork flavor. Du, (2012) extracted the flavor compound from fermented pork by SPME These data suggest that different heredity background is an important factor affecting meat flavor.
It was demonstrated that pretreatment temperature constitutes an important factor in the analysis of volatile components of meat from swine by HS-SPME and GC-MS. Indeed, Wang et al. (Wang et al., 2008) stated that 50 flavor compounds were detected when samples were pretreated at 60°C and that 168 were found with pretreatment at 80°C. Therefore, 80°C was selected for this study in order to identify the highest number of compounds.
Some previous studies compared the nutritional and flavor compound profiles between different breeds of pork. Quaresma et al. (2011) showed that the fatty acid profile of wild Sus scrofa scrofa was similar to that of commercial pork. Sales and Kotrba (2013) reviewed the differences among wild S. scrofa and a number of domestic breeds. Of course, S. scrofa tended to be smaller than the industrial breeds and to have different meat characteristics. Another study showed the diversity of volatile compounds and that profound differences in these compounds between wild boars and domestic pigs, as shown in the present study (Lammers, Dietze, & Ternes, 2009).
These differences could be due to differentially expressed genes.
Indeed, it has been shown that 23 genes involved in fatty acid me- determine the contribution of these genes to the volatile compound profile of these breeds.
The present study is not without limitations. Indeed, some components were not identified, indicating that further research is required. In addition, SPME at 80°C might not exactly reflect the conditions at which pork is routinely processed for food. Finally, it would have been helpful to carry out such studies using a range of temperatures for SPME.
In conclusion, the tenderness, taste, succulence, and broth flavor of the BAM and HZP were good. In the SUS, except that succulence was moderate, the other indicators were good. Meanwhile, all the indicators of the YOK pigs were moderate. HS-SPME/GC-MS identified many kinds of compounds in longissimus muscle samples from different pig breeds including the crossbreed SUS (106 compounds), and the purebreds BAM (108 compounds), YOK, and HZP (98 compounds each). The volatile compounds in pork belong to several classes, and the highest relative amount of volatile compounds was found in BAM. The main volatile compounds in pork which may contribute to flavor of pork were 3-methyl-1-butanol, 1-nonanal, octanal, hexanal, 2-pentyl-furan, 1-penten-3-one, N-morpholinomethyl-isopropyl-sulfide, methyl butyrate, and (E,E)-2,4-decadienal.

This work was supported by Discipline Construction Fund Project of
Gansu Agricultural University with Project No. GAU-XKJS-2018-045.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Conception and design of the research: GC; acquisition of data: HW; analysis and interpretation of data: SS; statistical analysis: HW; obtaining funding: GC; drafting the manuscript: LH; revision of manuscript for important intellectual content: YS. All authors read and approved the final manuscript.

E TH I C A L S TATEM ENT
All studies were performed with the approval of the Animal Use and Care Committee of Gansu Agricultural University, China.