Metabolomic Approach to Understand the Effect of Lairage on the Quality of Pork Loin

Animals

The experiment was approved by the Institutional Animal Care and Use Committee (IACUC) at Sunchon National University, Suncheon, Republic of Korea (SCNU IACUC-2020-12). The castrated commercial Landrace × Yorkshire × Duroc (LYD) pigs were used in this study. The animals were reared at two similar commercial animal-friendly farms where the essential environmental conditions were set according to the legislation of the Ministry of Agriculture, Food and Rural Affairs in Korea enforced from February 2, 2018. The pigs were raised in an animal-friendly farm, Cheil breeding stock farm (Icheon, Korea), until their live weight reached 30 kg (stocking density of >0.3 m² per animal) and then were transported to Yuunwoo animal-friendly farm (Hwasung, Korea) until their live weight reached 110 kg (stocking density of >1 m² per animal). Forty pigs from four rearing pens (n = 10 per pen) were used for the study, and the pigs from the same pen were randomly divided in half and classified into two groups: the pigs that were slaughtered immediately after unloading (CON; n = 20) and those that were provided with 24 h lairage (LRG; n = 20). Prior to transport, the live weight of the pigs was measured, and the average live weight of CON was 115.31 ± 0.91 kg and that of LRG was 118.11 ± 5.01 kg. The pigs were transported to an abattoir (Dodram, Anseong, Korea) for 2 h, which was about 90 km away from the farm. During transportation, the temperature was maintained at 15°C–18°C in an enclosed truck to minimize stress. The stocking density of the animals during the lairage period was >0.83 m², and the pigs could access water freely. The temperature and relative humidity were set to 15°C–18°C and 75%, respectively. After slaughter, the carcasses were cooled at 4°C for 24 h. Then, the carcass weight and backfat thickness of each carcass were measured, and loin cuts from the left half carcasses were obtained. The backfat thickness of the carcass was determined by measuring the average value of the backfat thickness between the last thoracic vertebra and the first lumbar vertebra and that between the 11th and 12th thoracic vertebrae of the left half carcass of the pig according to the Pig Carcass Grade System in Korea. The loin samples were vacuum-packaged and quickly transferred to the laboratory, and M. longissimus thoracis at 48 h postmortem from the loin samples was used for metabolomic profiling and physicochemical quality analyses.

Metabolomic analyses

Extraction of metabolites. The extraction process proceeded according to Kim et al. (2021). Five grams of loin muscle was added into 20 mL of 0.6 M perchloric acid and homogenized at 16,000 rpm for 1 min using a homogenizer (T25 digital ULTRA-TURRAX^®, Ika Works, Staufen, Germany). The homogenates were centrifuged at 2,265 × g for 20 min (Continent 512R, Hanil Co., Ltd., Daejeon, Korea). Then, the supernatant was neutralized and centrifuged at the same condition above. After filtration (No. 1, Whatman International Ltd., Kent, UK), the filtrates were lyophilized (Freezer dryer 18, Labco Corp., Kansas City, MO). The lyophilized extracts were reconstituted with the addition of 1 mL of 20 mM deuterium oxide-based phosphate buffer (pH 7.0) containing 1 mM of 3-(trimethylsilyl) propionic-2, 2, 3, 3-d₄ acid (TSP). The samples were centrifuged at 2,265 × g for 20 min (Continent 512R, Hanil Co., Ltd.), and the supernatant was further centrifuged at 17,000 × g for 10 min (HM-150IV, Hanil Co., Ltd.). The supernatant was transferred into NMR tubes and used for NMR analyses. The metabolite extraction process was conducted twice for each animal.

NMR analyses. NMR spectra were recorded using an 850 MHz cryo-NMR spectrometer (Bruker Biospin GmbH, Rheinstetten, Germany). One-dimensional ¹H NMR for quantitative analysis of metabolites was performed using a zg30 pulse program with a spectral width of 17006.803 Hz with 128 scans, and acquisition time was 4.20 s. Additionally, two-dimensional NMR spectra were obtained for peak identification. ¹H-¹³C heteronuclear single quantum coherence (HSQC) was performed with 11 ppm for the F2 and 233 ppm for the F1 axis, respectively, and 2 k data points in the t2 domain and 256 increments in the t1 domain with 16 scans. Correlation spectroscopy (COSY) was conducted with 11 ppm of spectral width, 2 k data points in the t2 domain and 128 increments in the t1 domain with 16 scans. The experimental condition of total correlation spectroscopy (TCOSY) was 2 k data points in the t2 domain and 256 increments in the t1 domain with 16 scans. Finally, heteronuclear multiple-bond correlation spectroscopy (HMBC) was recorded with 4 k data points in the t2 domain and 512 increments in the t1 domain with 16 scans. During the process, lock, tune, and shimming were performed automatically by ICON-NMR automation (Bruker Biospin GmbH).

The identification of peaks in the two-dimensional NMR spectra was accompanied using Topspin 4.0.8 (Bruker Biospin GmbH). The phase of each one-dimensional ¹H NMR spectra was first automatically corrected, and then manually corrected based on the signal of lactate (1.35 ppm), and the baseline of the spectra was corrected automatically. The concentration of identified peaks was quantified using Chenomx NMR suite 8.6 (Chenomx, Inc., Edmonton, Alberta, Canada) with the resonance of the TSP as a standard reference for both identification and quantification. An average value of metabolite concentration from two measurements in each animal was taken as one replicate.

Multivariate and pathway analysis. To identify the change in the metabolic process of pork loin by lairage, principal component analysis (PCA) coupled with orthogonal partial least squares-discriminant analysis (OPLS-DA) and pathway analysis were performed with the dataset of metabolite contents in pork samples using MetaboAnalyst 5.0 (www.metaboanalyst.ca). Prior to multivariate and pathway analysis, a log transformation and a Z-score normalization were applied to the metabolite data.

Physicochemical meat quality

pH. Two grams of loin muscle were homogenized with the addition of 18 mL of deionized distilled water at 11,000 rpm for 1 min (Polytron PT 10-35 GT, Kinematica AG, Luzern, Switzerland). The homogenates were filtered (No. 4, Whatman International Ltd.), and the pH of each filtrate was measured using a pH meter (Seven Excellence^™, Mettler Toledo International Inc., Schwerzenbach, Switzerland) at room temperature.

Moisture content. The moisture content of pork samples was measured according to the AOAC official methods (2000) with minor modifications. Three grams of loin muscle was placed on the aluminum dish and dried in the dry oven at 104°C to a constant weight. The moisture content (%) was determined by the percentage of weight loss after drying.

Water holding capacity. Five grams of the loin muscle was put into the centrifuge tube and centrifuged at 1200 × g for 10 min (Combi-514R, Hanil Co., Ltd.). The water holding capacity (WHC) of the sample was determined by the percentage of weight loss after centrifugation as follows:

Cooking loss. The pork loin was cut into 30 mm × 50 mm × 10 mm and was cooked at 100°C using an electric grill with double heating surfaces at 1,400 W (Nova EMG-533, Evergreen Enterprise, Yongin, Korea) for 90 s until the internal temperature of the sample reached 75°C. Cooking loss (%) was calculated by the weight difference of the sample before and after cooking.

Meat color. The meat color of pork samples was measured using a colorimeter (CR-410, Minolta Co., Osaka, Japan). The colorimeter was pre-calibrated with a standard white plate (Y = 86.8; x = 0.3156; y = 0.3225). The instrumental color was expressed as Commission International d’Eclairage (CIE) L* (lightness), a* (redness), and b* (yellowness) values using Spectra Magic Software (Minolta Co.). An average value of 3 measurements from each sample was taken as one replicate.

Warner-Bratzler shear force. The pork loin was cut into a meat block of 40 mm × 50 mm × 10 mm and heated until the core temperature of the meat block reached 75°C. Then, each sample of 10 mm × 20 mm ×10 mm was obtained from the cooked meat block, and used for shear force measurements using a texture analyzer (TA-XT2, Stable Micro Systems, Surrey, UK) equipped with a Warner-Bratzler blade. The muscle fiber direction was perpendicular to the blade, and the condition of the instrument was as follows: pre-test speed of 2.0 mm/s, test speed of 2.0 mm/s, and post-test speed of 5.0 mm/s.

Statistical analysis

The results were statistically analyzed by Student’s t-test (SAS 9.4, SAS Institute Inc., Cary, NC). Prior to t-test, F-tests were employed for metabolite concentrations, carcass and physicochemical quality parameters to analyze the homoscedasticity of each variable. Welch-Satterthwaite t-test was applied for 4 metabolites which showed heterogenous variances including betaine, glycine, tyrosine, and β-alanine and 3 carcass characteristic parameters, while pooled t-test was applied for the other variables with homogenous variances (Tables S1 and S2). The significant differences between the mean value of the 2 groups were determined at P < 0.05. A volcano plot was performed using MetaboAnalyst 5.0, and metabolites with the fold change (FC) of >1.2 or <0.8 and P < 0.05 were considered as regulated by lairage.

Carcass characteristics

The live weight of LRG was significantly higher than that of CON; on the other hand, there was no significant difference in carcass weight between the two groups (Table 1). Previous studies reported that the weight loss of the carcass showed an increasing trend with lairage time from 0 h to 24 h or overnight lairage (Warriss et al., 1998; Díaz et al., 2014). However, LRG showed significantly higher backfat thickness compared to CON, which was the opposite phenomenon from the research conducted by Warriss et al. (1998) and Nanni Costa et al. (2002), wherein the pigs that were provided with an overnight lairage showed a reduction in the backfat thickness compared to unrested pigs or pigs with 2 h of lairage. The authors suggested that the loss of water from the fat tissue led to a decrease in the amount of fat. Nevertheless, it was reported that the water supply during lairage could rehydrate the animals (Díaz et al., 2014). On the other hand, Cˇobanović et al. (2020) observed that cortisol positively correlated with backfat thickness but negatively correlated with lean carcass content, and they supposed that cortisol might stimulate fat storage to compensate for the increased protein degradation in a stressful condition. The mechanism of the effect of lairage on the increase in backfat thickness of pigs remains unclear; however, the carcass characteristics of both CON and LRG correspond to 1+ carcass grade, which is the best carcass grade in the Korean pig carcass grade system with 83–92 kg of carcass weight and 17–24 mm of backfat thickness (Park et al., 2022).

Identification of metabolites and multivariate analysis

We identified a total of 32 metabolites in every sample through NMR analyses (Table 2). Those were classified as (1) free amino acids, dipeptides, and derivatives; (2) nucleotide-related products; (3) organic acids; and (4) others (Table S1). Free amino acids, dipeptides, and derivatives included 19 metabolites (alanine, anserine, aspartate, betaine, carnitine, carnosine, creatine, glutamate, glutamine, glycine, isoleucine, leucine, lysine, methionine, phenylalanine, taurine, tyrosine, valine, and β-alanine); nucleotide-related products included 6 metabolites (hypoxanthine, inosine monophosphate [IMP], inosine, nicotinamide adenine dinucleotide [NAD], uridine monophosphate [UMP], and xanthine); organic acids included 5 metabolites (acetate, fumarate, lactate, nicotinate, and succinate); and others included glucose and glycerol, respectively. Free amino acids, dipeptides, and derivatives comprised a large proportion of the metabolic profiles identified in the present study due to (1) that the extraction process conducted here mainly focused on the polar metabolites and (2) a large content of amino acids present in the skeletal muscle (Gondret et al., 2021; Kim et al., 2021).

Based on the metabolomic profiles of pork loins, PCA and OPLS-based multivariate analyses were performed. PCA is an effective analysis for the identification of the overall difference between two groups (Kim et al., 2021). On the other hand, OPLS-DA is a supervised methodology that can maximize the between-class difference by using both the covariate matrix and response matrix compared to unsupervised PCA (Debik et al., 2021). In Figure 1, the red and green dots indicate the metabolomic profiles of CON and LRG, respectively. The red and green circled area indicate the 95% confidence regions for CON and LRG, respectively. It was clearly observed that CON and LRG were hardly separated by PCA as shown in Figure 1a, implying that both CON and LRG had similar metabolomics profiles in common. On the other hand, with OPLS-DA, a better distinction was found between CON and LRG compared with a PCA-based result as shown in Figure 1b. It suggests that the overall metabolomic profiles between the two classes were similar as shown in the PCA result, yet certain metabolite contents were different by the effect of lairage. Similarly, Gao et al. (2020) utilized multivariate analysis to elucidate the effect of heat stress on pork metabolomic profiles and found that the classes were clearly distinguished under the OPLS-DA method but overlapped under the PCA method.

Figure 1.

Principal component analysis (a) and orthogonal partial least squares-discriminant analysis (b) plot for the discrimination of metabolomic profiles of pork loin in relation to lairage. Pigs in the CON group were slaughtered immediately after unloading, whereas those in the LRG group were held in lairage for 24 h before slaughter.

Metabolic pathway analysis

Among the metabolites in the pork loin, the amount of lactate was the highest, followed by creatine and carnosine. However, we did not observe changes in lactate or carnosine contents after lairage (Figure 2a; P > 0.05). Instead, isoleucine and valine were up-regulated, whereas glutamate, glycerol, glycine, lysine, and methionine were down-regulated in pork loin by lairage (Figure 2b; P < 0.05). Creatine, although its FC was below 1.2, increased in LRG compared to CON (Figure 2a; P = 0.027).

Figure 2.

Heatmap analysis (a) and volcano plot (b) of metabolites from pork loin of CON and LRG. Pigs in the CON group were slaughtered immediately after unloading, whereas those in the LRG group were held in lairage for 24 h before slaughter. The red-blue color system was used in heatmap analysis to represent the relative abundance of each metabolite in the pork loin, and the * mark indicates significant differences in metabolite content between the two groups. The scattering of metabolites in the volcano plot includes up-regulated metabolites (red circle) on the right side and down-regulated metabolites (blue circle) on the left side in relation to 24 h of lairage. Metabolites with a fold change of >1.2 or <0.8 and P < 0.05 were considered as regulated by lairage.

To understand the effect of lairage on the metabolic profiles of pork loin, pathway analysis was conducted. Pathway analysis could provide valuable information about the role of each metabolite in biological reactions, and therefore the data of differentiated metabolomic profiles become interpretable in terms of biological context. Among 33 pathways identified, 13 pathways with P < 0.05 and pathway impact value > 0 were considered significantly differentiated between the two groups (Table 3). These included amino acid metabolisms (alanine, aspartate and glutamate metabolism, arginine and proline metabolism, arginine biosynthesis, cysteine and methionine metabolism, D-amino acid metabolism, glycine, serine and threonine metabolism, and histidine metabolism), citrate cycle, glutathione metabolism, glyoxylate and dicarboxylate metabolism, pantothenate and CoA biosynthesis, and primary bile acid biosynthesis. Most differentiated pathways by lairage belonged to amino acid metabolisms, which may explain the difference in free amino acids (glutamate, glycine, isoleucine, lysine, methionine, and valine) in pork loin by lairage (Figure 2b).

The results imply the beneficial effect of lairage on pigs in terms of the stability of (1) energy supply through amino acid and glycerolipid metabolisms and (2) antioxidant system via glutathione metabolism. Previous studies reported the relationship between stress and the metabolism of animals and suggested that the metabolic changes occurred to compensate for the energy needs induced by the physiological responses to stressful conditions such as increased heart rate and body temperature, impaired metabolism and hormonal regulation, and behavioral abnormalities (Zou et al., 2020; Lee et al., 2022). In this study, the up-regulation of isoleucine and valine after lairage may be related to energy metabolism as these branched-chain amino acids were reported to be utilized as nitrogen donors to produce alanine, which contributes to glycolysis (Jung et al., 2022). On the other hand, the increase of free amino acids and glycerol contents in CON may derived from the stimulated proteolysis and lipid metabolism by the action of corticosterone and norepinephrine to produce free amino acids with the action of proteases and the degradation products of lipids including triacylglycerols with glycerol and esterified three fatty acids, respectively (Zhang et al., 2021). These aforementioned metabolites can also be utilized as energy sources via glycolysis (Figure 3). Furthermore, the increase in creatine after lairage supports the idea that lairage could alleviate energy usage in response to environmental changes (Antonelo et al., 2020). Creatine is an important energy source and is involved in ATP synthesis and metabolism, so it might be possibly assumed from the result of the lower content of creatine in CON compared to LRG that creatine was utilized as an energy source (Yu et al., 2021). Other differentiated pathways, such as the citrate cycle and pantothenate and CoA biosynthesis, further support the hypothesis that the muscle required additional energy when the pigs were immediately slaughtered.

Figure 3.

A schematic figure of the metabolic pathways of the metabolites. The name of the identified metabolites in this study was expressed in bold. AICAR, 5-aminoimidazole-4-carboxamide ribonucleotide; AMP, adenosine monophosphate; G6P, glucose 6-phosphate; Hx, hypoxanthine; IMP, inosine monophosphate; NAD, nicotinamide adenine dinucleotide; NADP, nicotinamide adenine dinucleotide phosphate; PEP, phosphoenolpyruvate.

The changes in glycine, glutamate, lysine, and methionine contents might be further related to the antioxidant system. Glutathione metabolism was one of the differentiated pathways between CON and LRG, and glutathione can act as an antioxidant compound against reactive oxygen species and free radicals (Yu et al., 2021). Interestingly, glutamate, glycine, and methionine are constitutive amino acids of glutathione, and lysine plays an important role in the biosynthesis of glutathione; therefore, these 4 free amino acids participated in the glutathione metabolism pathway (Gondret et al., 2021; Zhang et al., 2022). Therefore, the increased glycine, glutamate, lysine, and methionine content in CON compared to LRG may result from the necessity of additional glutathione to protect the cellular components from reactive substances that were formed during stress-induced lipid and protein oxidation (Xing et al., 2019).

In brief, lairage led to the up-regulation of isoleucine and valine and the down-regulation of glutamate, glycerol, glycine, lysine, and methionine in pork loin, which suggested that lairage might help the animals to recover by lowering the energy need and reduce the stress-induced oxidative damage by stimulating glutathione metabolism. On the contrary, without lairage, the muscle proteins and lipids in pork loin would be degraded into energy substrates to fulfill the energy requirements as a result of physiological responses to stressful conditions.

Physicochemical meat quality

pH. The pH of CON and LRG did not differ significantly (5.71 ± 0.10 vs. 5.74 ± 0.13; P = 0.337), and both treatments showed a normal pH range (Table 4). On the other hand, previous studies in which the ultimate pH of pork with a long-term lairage for 8 h or more was higher than that with or without a short period of lairage for about 1 to 3 h (Warriss, 2003; Díaz et al., 2014; Dokmanović et al., 2014). The degree of pH decline is mainly affected by the muscle glycogen reserves, and the rate of pH change is dependent on postmortem muscle metabolism (Dalla Costa et al., 2019; Acevedo-Giraldo et al., 2020). Specifically, the glycogen after slaughter is converted into glucose which is used for glycolysis, and the accumulated lactate by glycolysis decreases the muscle pH (Lee et al., 2022). Further, the depletion of creatine phosphate and the accumulation of hydrogen ions from the hydrolysis of ATP also contribute to the decrease of meat pH (Yu et al., 2021). Among the metabolites identified in this study, the contents of glucose, lactate, and acetate (which participates in the glycolysis pathway) were not significantly different, which could explain our unexpected result (Figure 2a).

It is known that stress can affect meat pH. High levels of stress can induce the secretion of cortisol as a stress response of the animal, and cortisol stimulates postmortem glycogen degradation to produce energy (Dokmanović et al., 2014; Gonzalez-Rivas et al., 2020). In this regard, many studies reported that lairage can decrease the extent of glycolysis early postmortem because the animals can recover from pre-slaughter stress (Zhen et al., 2013). However, it was also known that under a prolonged lairage period, pigs face other stressors such as food deprivation, mixing, and fighting with unfamiliar pigs (Rey-Salgueiro et al., 2018). The increased muscular activity as a result of stressors during lairage leads to the consumption of muscle glycogen (Zhen et al., 2013). Therefore, the result of the present study might result from both the positive and negative effects of lairage. The lairage period was set to 24 h in this experimental design with the consideration of the commercial abattoir system in Korea; however, the pH of pork after a shorter lairage may exhibit a different result compared with that of the pork that was slaughtered immediately.

Moisture content, water holding capacity, and cooking loss. There were no significant differences between CON and LRG in the moisture content (74.11 ± 0.55 vs. 73.61 ± 1.08), WHC (69.20 ± 3.63 vs. 69.32 ± 2.73), or cooking loss (20.74 ± 0.93 vs. 21.06 ± 1.35; Table 4). In general, it is known that pigs can have a chance to rehydrate during the lairage period, which may affect moisture content and reduce carcass weight loss (Díaz et al., 2014). Previous studies reported that overnight lairage led to the increase of WHC with a pH rise in pork loin compared to short-term lairage due to the high ultimate pH or delayed muscle metabolism (Nanni Costa et al., 2002; Warriss, 2003; Dokmanović et al., 2017). Furthermore, Zhen et al. (2013) reported a decrease in the cooking loss of pork after 24 h lairage compared to pork without lairage. Our result was not in accordance with the results from the previous literature, possibly because these attributes can be influenced by pH, structural changes of myofibril, or postmortem storage conditions. Here, pH and the amounts of pH-associated metabolites such as lactate and acetate did not vary significantly by lairage, which might explain the present results.

Meat color. There was no significant difference in pork color between CON and LRG including lightness (55.73 ± 2.05 vs. 54.85 ± 1.65), redness (15.89 ± 1.03 vs. 15.80 ± 0.81), and yellowness (5.40 ± 0.58 vs. 5.56 ± 0.61) as shown in Table 4 (P > 0.05). It is well known that the appearance of color is important in assessing meat quality. It was reported that the lightness of pork loin showed a positive correlation with consumer preference in South Korea (Lee et al., 2021). The lightness of meat is highly attributed to the meat pH or the structure of muscles, such as myofibril diameter, sarcomere length, and distribution of sarcoplasmic proteins, which all affect light scattering (Purslow et al., 2020; Ali et al., 2021). Higher L* values were observed in pork loin that was assigned to lairage groups compared to groups without rest (Zhen et al., 2013). However, compared to a long-term lairage of 8 h or longer, a small lairage period of within 3 h was more effective in the increase in the lightness of pork meat (Dokmanović et al., 2014; Acevedo-Giraldo et al., 2020).

Similarly, the a* and b* values of pork loin were higher after a short lairage instead of a long lairage in previous studies (Dokmanović et al., 2017), whereas other studies reported an insignificant change in the b* values during the lairage period (Warriss, 2003). The myoglobin redox form can affect a* and b* values, and the relative composition of myoglobin changes depending on the mitochondrial oxygen consumption rate, metmyoglobin reducing activity, the content of NAD(H) and pH (Aroeira et al., 2017). In the present study, we could not find any significant differences in meat pH or NAD content between CON and LRG, which might be one reason why lairage did not lead to the change in meat color.

Warner-Bratzler shear force. Similar to other physicochemical traits, the shear force of pork loin was not significantly different between CON (5.70 ± 1.20) and LRG (6.01 ± 1.11) groups (Table 4; P = 0.413). In accordance with the results of physicochemical analyses, the trained sensory panels did not find any significant differences in the eating quality between CON and LRG pork (data not shown). On the other hand, Zhen et al. (2013) reported the increase in the shear force of pork loin was in accordance with the increase in the lairage time. Meat tenderization is attributed to muscle proteolytic activity, and this activity is regulated by pH, temperature, oxidation or nitrosylation of proteins, etc. (Xing et al., 2019; Ali et al., 2021). In pork and chicken meat, the degradation of muscle protein could occur by the cascade effect as a consequence of the increased corticosterone under stressful conditions (Lee et al., 2022). However, although the lairage had an effect on the relief of pre-slaughter stress that animals suffer, including transportation, loading, unloading, and dehydration, whether these kinds of stress induce change in meat tenderness or not, and related mechanisms, have been little known. Therefore, the relationship between lairage and shear force remains unclear and requires further study.

Overall, there was no considerable change in physicochemical quality parameters between CON and LRG in this study. As stated earlier, these attributes could be influenced much by pH; however, the content of organic acids including lactate between CON and LRG were not significantly different, which may lead to similar pH values between the two groups (Figure 2a). While previous studies reported the effect of overnight lairage on livestock such as the increase in pH and WHC as well as the decrease in drip loss and incidence of PSE meat (Warriss, 2003; Dokmanović et al., 2017), some studies stated that the influence of other stressors like fighting and mixing with unfamiliar animals could possibly be increased as the period of lairage increased (Rey-Salgueiro et al., 2018; Dalla Costa et al., 2019). Possibly, the effects of the changes in metabolite contents and metabolic pathways such as energy metabolism and antioxidant system on meat quality become more evident during storage as enzymatic and non-enzymatic degradation of taste compounds and the protein and lipid oxidation process also influence the meat quality.