Chemical composition

The moisture, protein and fat contents were determined using a Food Scan^TM Lab 78810 (Foss Tecator, Hillerod, Denmark), as described in our previous study (Seong et al., 2016). Each sample was determined in triplicates.

pH measurement

The pH of the meat samples was measured in triplicate by inserting a calibrated stainless steel pH probe of a pH*K 21 meter (NWK-Technology GmbH, Kaufering, Germany) deeply into the meat. Three readings were carried out at random locations for each the sample.

Instrumental color measurement

Transversal sections of belly or shoulder butt were taken consecutively and bloomed for 30 min before color measurement using a Minolta Chroma Meter CR-400 with a D65 illuminant*1 and 2° observer (Minolta Camera, Osaka, Japan). Care was taken to avoid scanning of intermuscular fat areas in the samples. The color was expressed according to the Commission International de l’Eclairage (CIE) system and reported as CIE L*(lightness), CIE a*(redness) and CIE b*(yellowness). The color values were measured at three random locations on each the sample.

Cooking loss determination

The cooking loss was determined by subjecting approximately 150 g meat steak (2.54-cm in thickness) of each sample to heat treatment by cooking in a pre-heated water bath (72°C) until the temperature reached 70°C as described by Van Ba et al. (2019). Following the cooking process, the cooked samples were immediately cooled for 30 min under running water and then re-weight to determine cooking loss. The cooking loss was calculated as the ratio of the cooked to the raw meat sample weight.

Sensory evaluation

The sensorial characteristics of both the pork samples were evaluated using a six-member well trained panels selected from the institution’s staffs as described in our previous study (Van Ba et al., 2019). The sensory evaluation procedure was approved by the Institutional Review Board of National Institute of Animal Science (No.11-1390744-000007-01). To minimize the variation in eating quality caused by the sampling location, for each the belly, three fixed sub-samples (Fig. 1) were collected, separately evaluated and the mean score for each sensorial trait was the average of scores obtained from these three sub-samples. Prior to use, the frozen vacuum-packed sub-samples were defrosted at 4°C for approximately 2 h, and they were then manually sliced into 7 representative slices (50×50×4 mm: W×L×D). Of which 1 strip was used for general sensorial color evaluation after 30 min cutting (blooming). The rests of strips (6 per sample) were cooked at approximately 180°C on an open tin-coated grill for about 2 min. Immediately after cooking, the samples were placed on individual dishes and served to the panelists. The panelists then handled the cooked samples with an approved odorless plastic fork and ranked on 7-point hedonic scale (1, dislike extremely; 2, dislike very much; 3, dislike moderately; 4, neither like nor dislike; 5, like moderately; 6, like very much; and 7, extremely likeand) for flavor, juiciness, tenderness and overall acceptability as described by Meilgaard et al. (1991). Between the samples, the panelists were asked to refresh their palate with drinking water and unsalted crackers.

Volatile flavor compounds

The volatile flavor compounds in cooked pork samples were determined using the method standardized by Van Ba et al. (2010) with minor modifications. Briefly, immediately after cooking, 2.0 g of each the cooked sample was taken and placed into a 20-mL headspace vial (Part No. 5188-2753, Agilent, Santa Clara, CA, USA) and 1.0 μL of 2-methyl-3-heptanone (816 mg /mL in methanol) as an internal standard (ISD) was also added. The vial containing sample was sealed with PTFE-faced silicone septum and was then extracted for volatile flavor compounds at 65°C for 60 min using the solid-phase micro-extraction technique. The extracted volatiles were then separated into a DB-5MS capillary column, 30 m×0.25 mm i.d.×0.25 μm film thickness (Agilent J & W Scientific, Folsom, CA, USA) connected to a Gas Chromatography (Model: 7890B GC) and Mass Spectrophotometry (Model: 5977B MSD, Agilent Technologies, Santa Clara, CA, USA). Conditions used for the separation and analysis of the volatiles were same as those described in the above-cited reference (Van Ba et al., 2010). The volatiles were identified by (i) comparing their mass spectra with those already present in the Wiley registry of mass spectral data (Agilent Technologies) and (ii) by comparing their retention times with those of external standards. The final concentration (μg/g meat) of each identified was calculated by comparing its peak area with the peak area of known-concentration internal standard.

Effects of QG on the volatile flavor compounds

The concentrations of the identified volatile flavor compounds in the cooked belly and shoulder butt cuts as affected by the QG are presented in Table 3. The outcome of our analysis displayed a broad range of flavor compounds (over forty compounds) comprising of 17 aldehydes, 6 alcohols, 2 ketones, 6 hydrocarbons, 2 furans and 4 nitrogen-and sulfur-containing compounds. Based on the formation pathways of flavor compounds in cooked meats (Mottram, 1998; Van Ba et al., 2013), it appears likely that most of the identified compounds were derived from the lipid oxidation/degradation, and only few were formed via the Maillard reaction between amino acids with reducing sugars. In general, both cut types had the volatile flavor profile characteristic of high fat content meats, being indicated by a greatly predominant number and amount of the fatty acids-derived compounds such as aldehydes, alcohols and hydrocarbons (Elmore et al., 2005).

Table 3.

Volatile aroma profiles in cooked belly and shoulder butt among the quality grades

	Retention time (min)	Belly			Shoulder butt			Identification method1)
		QG 1+	QG 1	QG 2	QG 1+	QG 1	QG 2
Aldehydes
Propanal	1.723	0.03±0.01	0.04±0.01	0.05±0.02	0.03±0.01	0.03±0.01	0.03±0.01	MS+STD
2-Ethylhexanal	2.167	0.01±0.01	0.02±0.01	0.03±0.01	0.02±0.01a	0.01±0.01b	0.01±0.01b	MS+STD
3-Methylbutanal	2.72	0.02±0.01b	0.02±0.01b	0.04±0.02a	0.02±0.02	0.01±0.01	0.02±0.01	MS+STD
2-Methylbutanal	2.829	0.02±0.02b	0.03±0.02b	0.08±0.04a	0.04±0.03	0.02±0.02	0.03±0.02	MS+STD
Hexanal	6.121	3.06±0.19a	2.82±0.14ab	2.72±0.26b	2.50±0.46	2.56±0.33	2.95±0.23	MS+STD
2-Methyl-4-pentenal	7.815	0.00±0.01	ND	ND	ND	0.01±0.01	0.01±0.01	MS
Heptanal	9.261	0.17±0.05	0.16±0.05	0.16±0.03	0.23±0.05	0.25±0.05	0.27±0.04	MS+STD
E,2-Heptenal	10.755	0.04±0.02	0.04±0.02	0.04±0.02	0.07±0.02	0.06±0.02	0.06±0.02	MS+STD
Benzaldehyde	10.873	0.04±0.01	0.04±0.01	0.06±0.02	0.08±0.01a	0.05±0.01b	0.05±0.02b	MS+STD
Octanal	11.915	0.18±0.11	0.21±0.03	0.21±0.09	0.24±0.06	0.26±0.04	0.27±0.06	MS+STD
Benzenacetaldehyde	12.874	0.01±0.01	0.01±0.01	0.02±0.01	0.02±0.02	0.01±0.01	0.01±0.00	MS+STD
E,2-Octenal	13.19	0.02±0.02	0.03±0.01	0.02±0.01	0.03±0.02	0.03±0.01	0.02±0.00	MS+STD
Nonanal	14.198	0.21±0.08	0.20±0.05	0.18±0.05	0.53±0.14a	0.26±0.05b	0.25±0.04b	MS+STD
E,2-Nonenal	15.33	0.10±0.06	0.06±0.03	0.14±0.13	0.05±0.04	0.10±0.06	0.10±0.07	MS+STD
E,E-2,4-Decadienal	16.229	ND	ND	ND	0.01±0.00	0.01±0.00	0.01±0.01	MS+STD
2-Undecenal	17.277	ND	ND	ND	0.01±0.00	0.01±0.00	0.00±0.00	MS+STD
2-Methylundecanal	17.471	ND	ND	ND	0.02±0.01	0.01±0.00	0.01±0.00	MS
Alcohols
1-Penten-3-ol	3.067	0.01±0.00	0.01±0.00	0.01±0.00	0.01±0.00	0.01±0.00	0.01±0.00	MS+STD
4-Amino-1-hexanol	3.302	0.15±0.09	0.20±0.05	0.20±0.05	0.22±0.04	0.14±0.08	0.17±0.02	MS
1-Pentanol	5.026	0.16±0.02a	0.13±0.02b	0.12±0.02b	0.12±0.01	0.12±0.02	0.14±0.02	MS+STD
1-Heptanol	11.112	0.02±0.01	0.01±0.01	0.01±0.01	0.04±0.01	0.02±0.01	0.03±0.01	MS+STD
1-Octen-3-ol	11.356	0.11±0.06	0.08±0.06	0.09±0.04	0.12±0.04	0.12±0.05	0.07±0.03	MS+STD
2-Ethyl-1-hexanol	12.588	0.03±0.01	0.03±0.00	0.03±0.00	0.08±0.08	0.03±0.02	0.03±0.01	MS
Hydrocarbons
Toluene	4.929	0.01±0.00	0.01±0.00	0.01±0.00	0.02±0.00a	0.01±0.00b	0.01±0.00b	MS+STD
1,3-Dimethyl benzene	7.982	0.01±0.01b	0.01±0.01b	0.02±0.00a	0.00±0.00	0.01±0.01	0.01±0.00	MS
Xylene	8.915	0.08±0.02	0.07±0.03	0.07±0.04	0.03±0.00b	0.06±0.03b	0.06±0.02a	MS
2,4-Dimethylhexane	13.029	0.03±0.01	0.02±0.01	0.03±0.01	0.03±0.01	0.03±0.00	0.03±0.01	MS
Benzoic acid	15.433	0.06±0.01	0.05±0.01	0.05±0.04	ND	ND	ND	MS+STD
Tridecane	16.101	ND	ND	ND	0.03±0.01	0.01±0.01	0.01±0.00	MS
Furans
2-Pentylfuran	11.581	0.27±0.08	0.27±0.03	0.21±0.08	0.14±0.05b	0.19±0.05ab	0.25±0.05a	MS+STD
2-Octylfuran	15.965	0.04±0.01	0.03±0.00	0.02±0.01	0.02±0.01b	0.03±0.01a	0.03±0.01a	MS+STD
Nitrogen and sulfur containing compounds
4-Methylthiazole	11.475	0.19±0.09	0.20±0.01	0.17±0.03	0.11±0.03b	0.15±0.04ab	0.16±0.03a	MS+STD
2,5-Dimethyl-pyrazine	9.558	0.01±0.01b	0.01±0.01b	0.04±0.03a	0.02±0.02	0.02±0.01	0.01±0.01	MS+STD
Carbon disulfide	1.862	ND	0.01±0.00	0.01±0.01	0.01±0.00	0.01±0.00	0.01±0.00	MS+STD
2-Ethyl-3,5-dimethyl-pyrazine	13.575	0.02±0.01	0.02±0.00	0.02±0.01	0.04±0.02a	0.02±0.00b	0.02±0.00b	MS

Open in a separate window

^a,bMeans within a row in each cut with different superscripts are different at p<0.05.

¹⁾Identification method: the compounds were identified by mass spectra (MS) from library or external standard (STD). QG, quality grade; ND, not detectable.

Regarding aldehydes, which were the most predominant flavor class found in the both cut types with a total of 15 and 17 compounds in the belly and shoulder butt, respectively. Each of the identified aldehydes was present at a level of at least 0.01 μg per 1.0 g of sample in each the QG group. Of these compounds, however, only few were influenced by the QG when examined by analysis of variance. For the belly, only 3 compounds namely 3-methyl-butanal, 2-methyl-butanal and hexanal showed statistical difference among the QGs. The 3-methyl-butanal and 2-methyl-butanal possessing cheese, nutty and salty notes in cooked pork (Dos Santos et al., 2015), were significantly higher in the QG 2 compared to the other remaining QG groups. These two compounds are originated from the degradation of isoleucine and leucine, respectively (Aaslyng and Meinert, 2017). Hexanal is known to arise from the degradation/oxidation of linoleic acid (Martín et al., 2001; Van Ba et al., 2013), our results depict that its amount was greater in the QG 1⁺ than in the QG 2. Hexanal has been reported to contribute positively to the cooked meat flavor (e.g., fatty odor), but may produce undesirable flavors at higher concentrations (Calkins and Hodgen, 2007). For the shoulder butt, three aldehydes showing the statistical difference (p<0.05) among the QGs were 2-ethylhexanal, benzaldehyde and nonanal. Of them, 2-ethylhexanal appears likely to be formed from the Strecker degradation of amino acid, and its concentration was significantly higher in the QG 1⁺ than those in the QG 1 and QG 2 (p<0.05). While, benzaldehyde and nonanal are the products derived from the oxidation/degradation process of linolenic and oleic acid, respectively (Elmore et al., 2002; Van Ba et al., 2013). The concentrations of these compounds also were higher in the QG 1⁺ than those in the QG 1 and QG 2 (p<0.05). The benzaldehyde has been reported to possess unpleasant flavors (e.g., almond oil, bitter almond and fishy odors) whereas, the nonanal was reported to possess pleasant flavors (e.g., roasted, sweet and fatty odors) in cooked meat (Aaslyng and Schäfer, 2008; Calkins and Hodgen, 2007). Thus, the results indicating the differences in amounts of these aldehydes is likely related to the variations in levels of precursors (e.g., amino acids and fatty acids) among the QGs studied because the content and nature of the precursors determine the flavors generated during cooking (Aaslyng and Meinert, 2017).

Regarding the alcohols, they partly contribute to the cooked meat flavors due to their low odor-detection threshold (Sabio et al., 1998). However, except 1-pentanol, all of the identified alcohols showed no significant differences among the QG groups for both the belly and shoulder butt cuts (p>0.05). The 1-pentanol associated with fruity and oily odors (Calkins and Hodgen, 2007), is known as the linoleic acid oxidation-derived product in meat during cooking (Elmore et al., 2002; Van Ba et al., 2013). Our result depicts that the amount of this compound was higher in the QG 1⁺ bellies (0.16 μg/g) compared to those in the other remaining QG groups. Similarly, a research conducted to examine the effect of QG on volatile flavor profiles in pork LD muscles has also shown that the QG had a minor effect on the quality and quantity of alcohol class (Van Ba et al., 2019).

Out of the identified hydrocarbons, toluene, and 1,3-dimethylbenzene and xylene were the compounds showing significant (p<0.05) differences among the QG groups in the belly and shoulder butt, respectively. Of which, toluene and 1,3-dimethylbenzene were likely derived from the Strecker degradation of amino acids (Olivares et al., 2011). In general, hydrocarbons are known as the lipid oxidation/or amino acids Strecker degradation-derived products which apparently have a minor contribution to the cooked meat flavors because of their high odor-detection thresholds (Mottram, 1998). No differences occurred in the identified furans among the QG groups for the bellies (p>0.05). For the shoulder butt, both of the furans (2-pentylfuran and 2-octylfuran) showed differences among the QG groups, with significantly higher amounts in the QG 2 (p<0.05). The 2-pentylfuran and 2-octylfuran are the products derived from the oxidation of C18:2n-6 and C18:1n-6, respectively (Van Ba et al., 2013). The furan class seems to little contribute to the flavor of cooked meat due to their high odor-detection thresholds.

Nitrogen-and sulfur-containing compounds are produced in the Maillard reaction between amino acids and a reducing carbohydrate in meat during cooking/heating (Mottram, 1998; Thomas et al., 2014). In which, the sulfur-containing amino acids such as cysteine are the main precursors for the formation of the sulfur-containing compounds which are associated with pleasant odors such as meaty and onion of cooked meats (Mottram, 1998). The other amino acids are such as; glycine and valine favor the formation of nitrogen-containing flavor compounds such as pyrazines and thiazoles which are associated with roasted and grilled flavors of cooked meats (Mottram, 1998). With respect to these Maillard compounds, the QG only affected the 2,5-dimethylpyrazine whose amount was significantly higher in the QG 2 bellies compared to those in the other QG groups (p<0.05). For the shoulder butt, the QG also did affect two compounds (4-methylpyrazole and 2-ethyl-3,5-dimethyl-pyrazine) whose amounts also were higher in the QG 2 than those in the QG 1⁺ or the QG 1 (p<0.05). Almost all of these compounds have also been reported in cooked pork and beef in literatures (Cho et al., 2020; Van Ba et al., 2020). It appears that both the belly and shoulder butt cuts in the lower QG group (e.g., QG 2) presented higher amounts of the Maillard reaction-derived flavor compounds which are associated with meaty, roasted and grilled flavors whereas, those from the higher QG groups (e.g., QG 1⁺) presented higher amounts of the fatty acids-derived compounds which are associated with the fatty and oily flavors. This could be related to the differences among the QG groups in the content and nature of precursors present in the cuts.

This study was supported by 2020-Postdoctoral Fellowship Program of National Institute of Animal Science (Project No. PJ01212502), Rural Development Administration, Korea.