The pH of the meat sample was measured according to Lee et al [16]. One gram of each sample was homogenized with 9 mL of deionized distilled water using a homogenizer (T25 digital ULTRA-TURRAX; Ika Works, Staufen, Germany) at 9,600 rpm for 30 s. Then, the homogenates were centrifuged at 2,265×g for 10 min (Continent 512R; Hanil Co., Ltd., Daejeon, Korea). The supernatants were filtered (No. 4; Whatman International Ltd., Kent, UK) and the pH of each sample was measured using a pH meter. A pH meter (Seven2Go; Mettler Toledo Inc., Schwerzenbach, Switzerland) was pre-calibrated by standardized buffer solutions (pH 4.01, 7.0, and 9.21) at room temperature.

Meat color was measured using a colorimeter (CM-5; Konica Minolta Censing Inc., Osaka, Japan) [16]. Before the analysis, the meat was bloomed for 30 min. Following pre-calibration of the colorimeter using a standard white plate, the analysis was conducted under the conditions of a D65 illuminant, a 10° standard observer, and a plate with a diameter of 30 mm. The meat’s lightness, redness, and yellowness were expressed as Commission Internationale d’Eclairage (CIE) L*, a*, and b* values, respectively. Six measurements were taken for each sample, and the average was used as one replicate.

The shear force value of Chikso and Hanwoo beef was measured according to Kim et al [17] with some modifications. Samples with an average weight of 100 g were placed in polyethylene bags in a water bath at 85°C until the internal temperature of the sample reached 72°C. After cooking, the samples were cooled to room temperature. Then, round cores were taken using a cork borer (1.27 cm diam.) parallel to the direction of the muscle fiber. A texture analyzer (TA1; AMETEK Lloyd Instruments Ltd., Fareham, UK) with a cross-head speed of 200 mm/min, a test speed of 60 mm/min, and a trigger load of 0.1 N was used.

The myofibrillar fragmentation index (MFI) analysis was conducted according to Kim et al [17] with a slight modification. The meat sample (1 g) was placed in the centrifugal tube and was homogenized at 15,000 rpm by adding 19 mL of MFI buffer (pH 7.0 at 4°C, containing 100 mM KCl, 1 mM ethylenediaminetetraacetic acid, 1 mM NaN₃, and 25 mM potassium phosphate) for 30 s (T25 digital ULTRA-TURRAX; Ika Works, Germany). The homogenates were filtered through a 1-mm mesh strainer to remove connective tissues and washed with 10 mL of MFI buffer. The filtrate was centrifuged at 1,000×g for 15 min (Continent 512R; Hanil Co., Ltd., Korea). The supernatant was discarded, and the pellet was washed with 10 mL MFI buffer and vortexed. This experiment was repeated five times. Next, 10 mL of MFI buffer was added to the remaining pellet, vortexed, and the protein concentration was determined using the biuret method. The samples were then diluted with MFI buffer to reach 0.5 mg/mL of protein concentration. The absorbance of each sample was measured at 540 nm using a spectrophotometer (X-ma 3100; Human Co. Ltd., Seoul, Korea). The MFI value was calculated by multiplying the absorbance by 200.

The nuclear magnetic resonance (NMR) analysis was conducted according to Kim et al [18]. The meat samples (5 g) were homogenized with 20 mL of 0.6 M perchloric acid at 16,000 rpm for 1 min (T25 digital ULTRA-TURRAX; Ika Works, Germany). The homogenates were centrifuged at 2,265×g for 20 min (Continent 512R; Hanil Co., Ltd., Korea). The pH of the collected supernatants was adjusted to 7.0 at room temperature using perchloric acid and potassium hydroxide solutions. The samples were centrifuged under the same conditions. Afterward, the samples were filtered (No. 1; Whatman International Ltd., UK), and the filtrates were lyophilized (Freezer dryer 18; Labco Corp., Kansas City, MO, USA). Following lyophilization, 1 mL of 20 mM deuterium oxide-based phosphate buffer (pH 7.0) containing 1 mM 3-(trimethylsilyl) propionic-2, 2, 3, 3-d₄ acid (TSP) was added to the lyophilized samples. The reconstituted samples were incubated at 37°C for 10 min and centrifuged at 2,265×g for 20 min (Continent 512R; Hanil Co., Ltd., Korea). The supernatants were further centrifuged at 17,000×g for 10 min (HM-150IV; Hanil Co., Ltd., Korea). The supernatants were loaded into NMR tubes and used for further analysis.

One-dimensional ¹H NMR and ¹H-¹³C heteronuclear single quantum coherence (HSQC) spectra of the samples were recorded using a Bruker 850 MHz cryo-NMR spectrometer (Bruker Biospin GmbH, Rheinstetten, Germany). The spectra were manually corrected by Chenomx NMR suite 7.1 (Chenomx, Inc., Edmonton, AB, Canada). The peak was identified by comparing the peak spectrum of the sample from HSQC using Topspin 4.0.8 (Bruker Biospin GmbH, Germany) with that of the Human Metabolome Database (www.hmdb.ca). The identified peaks were quantified through Chenomx NMR Suite 7.1. The identification and quantification of each metabolite were referenced to the resonance of the TSP.

VOC analysis was conducted using solid-phase microextraction and gas chromatography-tandem mass spectrometry (SPME-GC-MS/MS), according to Lee et al [16] with a slight modification. Ground meat samples (5 g) were placed into a 20-mL headspace vial and sealed with a PTFE-faced silicone septum. The samples were incubated at 40°C for 10 min before the extraction. A polydimethylsiloxane/divinylbenzene fiber (Supelco Inc., Bellefonte, PA, USA) was injected into the headspace of the vial and the extraction process was continued at 40°C for 30 min. The extracted VOCs were desorbed using a GC injector (Trace 1310; Thermo Fisher Scientific, Waltham, MA, USA) for 2 min at a split ratio of 1:10. A DB-Wax column (60 m×0.25 mm i.d., and 0.50 μm film thickness; Agilent Technologies Inc., Santa Clara, CA, USA) was equipped to the GC for VOCs separation. The GC condition was as follows: helium was used as the carrier gas at the flow rate of 1.5 mL/min; initial oven temperature was held at 40°C for 2 min, increased at a rate of 4°C/min and held at 150°C for 10 min, then increased at a rate of 4°C/min and held at 200°C for 5 min, and finally, increased at a rate of 10°C/min and held at 230°C for 5 min. Mass spectra of the fragmentations were obtained using a triple quadrupole mass spectrometer (TSQ 8000; Thermo Fisher Scientific, USA) in the electron impact (EI) mode with a scan range from 35 to 550 m/z in full-scan mode at 0.2 s of scan interval. VOCs in Chikso and Hanwoo beef were identified by comparing their mass spectra with those from the National Institute of Standards and Technology (NIST) mass spectral library (version 2.0 g) and the linear retention index (LRI). For the LRI, a n-alkane standard (C₈–C₂₀) was analyzed under the same conditions for the calculation of the LRI of each VOC.

Electronic tongue analysis was performed according to Lee et al [14] with slight modifications. The ground meat samples (5 g) were homogenized with the addition of 100 mL deionized distilled water at 10,000 rpm for 30 s (T25 digital ULTRA-TURRAX; Ika Works, Germany). The samples were centrifuged at 2,265×g for 10 min (Continent 512R; Hanil Co., Ltd., Korea) and the supernatants were filtered (No. 4; Whatman International Ltd., UK). Taste attributes of the samples were analyzed using an electronic tongue (Astree; Alpha MOS, Toulous, France). Taste screening was performed using Alpha soft (Alpha MOS, France) to compare the relative taste intensities of the samples using the following equation:

where X and m are the mean sensor values for each and the total group, respectively, and σ represents the deviation of the sensor value within the total group. Relative taste intensity was measured on a 10-point scale. For the AHS and NMS sensors, the value X-mσ was subtracted from 2.5, as lower sensor values of AHS and NMS indicate higher taste intensities.

The pH of Hanwoo and Chikso rumps on day 0 was not significantly different, and they gradually decreased until day 21 (Figure 1a). However, on day 28, the Chikso rump showed a significantly higher pH than the Hanwoo rump. Generally, the meat pH drops during wet aging because of the accumulation of lactic acid by the increase in lactic acid bacteria under anaerobic conditions [11]. On the contrary, the generation of alkaline products such as ammonia and amines by protein hydrolysis during wet aging would increase the pH [19]. Meanwhile, Chikso loin had a significantly higher pH compared to Hanwoo loin on day 0. However, no significant difference in loin pH was observed between the two breeds after wet aging.

During wet aging, the L* value of the Chikso rump significantly increased on days 14 and 21 compared to day 0 (Figure 1b). The increase in L* value in wet-aged beef may be attributed to the expulsion of moisture at the meat surface, which reflects the light, modification of surface protein structure by hydrolysis, denaturation during wet aging, etc. [7,20]. On the other hand, the Hanwoo rump had not exhibited any difference in L* value during 28 days of wet aging.

During wet aging, a* values of rump from both breeds significantly decreased (Figure 1c), following the results of the study performed by Ma et al [21]. This was possibly due to the reduced color stability by the oxidation of myoglobin and lipids [20]. It was also reported that extended vacuum storage would deteriorate color stability and bloom development of meat [21]. Meanwhile, a major difference in meat color between Hanwoo and Chikso rump was found in a* value. The Hanwoo rump showed a redder surface than the Chikso rump during wet aging (p<0.05), except on day 14. The a* value is highly affected by the content and chemical status of myoglobin [22]. Furthermore, recent studies found a negative correlation between color stability and phenylalanine or tryptophan, which act as substrates for reactive oxygen species, suggesting that metabolomic changes might influence the meat color [12,22]. In the present study, Chikso rump had higher phenylalanine content during wet aging compared to Hanwoo rump (Table 1), and phenylalanine was negatively correlated with a* value (r = −0.80; Supplementary Table S3). Therefore, it can be assumed that the high phenylalanine content in Chikso rump could decrease the a* value during storage.

The b* value in the Chikso rump decreased during wet aging (p<0.05), whereas there was no significant difference in Hanwoo rump (Figure 1d). However, the Hanwoo rump had a higher b* value on days 0 and 28 than the Chikso rump. In contrast, there were fewer changes in the color of Hanwoo and Chikso loins during wet aging.

Before wet aging, Chikso beef showed a significantly higher shear force value than Hanwoo rump (Figure 2a). Similarly, the shear force value of Chikso loin was higher than that of Hanwoo loin on day 0 (p<0.05). Hanwoo cattle have been bred for a long time to enhance meat tenderness, intramuscular fat content, marbling score, etc. [23]. As a result, Hanwoo beef has been recognized as tender beef with high meat quality grade, thin muscle fibers, and connective tissues [5]. However, the muscle characteristics of Chikso beef have not been well studied. It has been reported that the differentially differentiated genes between Hanwoo and Chikso were associated with adipose tissue development, fatty acid metabolism, and muscle lipid composition [24]. Therefore, differences in the characteristics of muscle tissue of Hanwoo and Chikso beef might be a possible candidate for explaining their different tenderness.

Nonetheless, as wet aging proceeded, the shear force value of the Chikso rump decreased to almost half of the original value on day 21. No significant difference was found between the Hanwoo and Chikso rumps after day 7. The Chikso rump showed lower shear force values on days 21 and 28 than the Hanwoo rump (p<0.05). The decrease in shear force value during wet aging may be attributed to the action of proteolytic enzymes [10]. Moreover, the effect of wet aging on meat tenderization was more remarkable in Chikso rump than in Hanwoo rump. Based on the standard of Belew et al [25] Chikso and Hanwoo rump on day 0 could be classified as “tough” (>45 N) and “tender” (between 31 and 38 N), respectively. However, after wet aging process, the shear force values of rump cut from two breeds showed similar values with the same category of “very tender” (<31 N), suggesting that Chikso rump might have similar tenderness properties with Hanwoo rump after day 7 of wet aging.

Similarly, the shear force values of the Chikso loin was significantly higher than the Hawnoo loin in the early phase of wet aging (days 0 to 7). However, as the wet aging process extended to day 14, the shear force value of the Chikso loin reached a value similar to that of the Hanwoo loin (p>0.05). Then, it significantly decreased and became lower than that of the Hanwoo loin on day 28. However, unlike the Chikso loin, there was no significant change in the shear force value of the Hanwoo loin during the wet aging period. Initially, Chikso loin could be categorized as “intermediate” (between 38 and 45 N) on day 0 [25]. Nonetheless, its shear force value reduced below 31 N after day 7 and showed insignificant values with Hanwoo loin on day 14, indicating that 14 days of wet aging might lead Chikso loin to show similar tenderness with Hanwoo loin.

Meat tenderness is mainly determined by the connective tissue content, myofibrillar protein degradation after slaughter, and sarcomere length [23,26]. Among them, the degree of myofibril degradation can be assessed by MFI. Especially, MFI value is more direct than shear force value for estimating the tenderization effect of wet aging on meat product [27]. In the present study, the MFI values of Hanwoo and Chikso rumps significantly increased with the wet aging period (Figure 2b), indicating that wet aging had a positive effect on the tenderization of Hanwoo and Chikso beef. Especially, the MFI value of Chikso rump was significantly higher than that of the Hanwoo rump after day 7. In the case of loin cuts, the MFI value of Chikso loin significantly increased, whereas that of Hanwoo loin did not change during wet aging. The results showed that Chikso beef with low tenderness on day 0 was significantly improved by wet aging. Furthermore, aging was more effective for Chikso beef than for Hanwoo beef. Therefore, wet aging for 14 days or more may contribute to the value addition of Chikso beef by tenderizing it to make the tenderness similar to that of Hanwoo beef.

A total of 43 VOCs were identified in the rump cuts during wet aging, including 12 alcohols, 10 aldehydes, 5 esters, 1 furan, 7 hydrocarbons, 5 ketones, and 3 volatile fatty acids (Figure 4). Most of the VOCs detected in this study have been reported as compounds produced by lipid oxidation [3]. Interestingly, on day 0, the major VOC group in the Hanwoo rump was aldehydes, whereas that in the Chikso rump was ketones (Figure 5a, b). In general, aldehydes are known as the major contributors to beef flavor among lipid oxidation-derived VOCs [7]. However, ketones also exhibit distinct oily, fruity, and fatty flavors with low odor-thresholds [11,31]. Therefore, these compounds could contribute significantly to the volatile flavor of Chikso rump. Ketones are known to be generated by amino acid degradation, lipid oxidation, carbohydrate metabolism or β-keto acid oxidation [32]. In particular, the amount of 3-hydroxybutan-2-one was significantly higher in the Chikso rump than in the Hanwoo rump, and it was the most abundant compound in the Chikso rump on day 0 (Supplementary Table S1). Therefore, 3-hydroxybutan-2-one is considered a potential contributor to the unique flavor of Chikso rump. Furthermore, 4-methylheptan-2-one was present only in the Chikso rump until day 21. In contrast, other VOC groups, such as alcohols, aldehydes, and esters, were present at higher concentrations in the Hanwoo rump than in the Chikso rump. Hexanal was the most abundant compound in the Hanwoo rump. It is a product of linoleic acid oxidation and is a major VOC in beef that gives a pleasant flavor to cooked meat [7,8]. Moreover, the degradation products of oleic acid, such as heptanal, octanal, and nonanal, were also detected in the Hanwoo and Chikso rumps. These compounds have sweet and fatty odor characteristics and influences the beef flavor significantly [11,33]. Differences in the lipid oxidation-derived VOCs contents, such as hexanal, heptanal, octanal, and nonanal, between Hanwoo and Chikso rump might be due to differences in antioxidant activity. We found that Chikso rump had significantly higher 2,2′-azinobis-3-ethylbenzothiazoline-6-sulfonic acid (ABTS) radical scavenging activity than Hanwoo rump on days 7, 21, and 28 (Supplementary Figure S1). This might help reduce the extent of lipid oxidation, as major lipid oxidation-derived products (hexanal and heptanal) were less abundant in the Chikso rump than in the Hanwoo rump.

As the wet aging period increased, the alcohol percentage also increased, while that of ketones decreased in the Chikso rump. Ethanol can be generated by microbial fermentation during the aging process [11]. Other compounds, (E)-oct-2-en-1-ol and hexan-1-ol, in the Chikso rump also amplified during aging (p<0.05). However, these compounds have high odor thresholds, and thus contribute less to beef flavor. There was also a significant rise in the aldehyde and ester contents in the Chikso rump until day 14 and then reduced on day 28. On the other hand, volatile fatty acids in the Chikso rump significantly decreased on day 7 but increased thereafter. Wet aging enhances meat flavor by generating flavor precursors and VOCs; however, previous studies have also reported a decrease in major volatile flavor contributors, such as aldehydes, during wet aging. Xu et al [8] reported that the octanal and nonanal concentrations decreased as the wet aging period increased. Similarly, Lee et al [11] reported that the concentration of esters in beef decreased after day 28. In the present study, the total amount of VOCs in both Hanwoo and Chikso rumps decreased after day 28 (Supplementary Table S1). The overall VOC profiles of Hanwoo and Chikso rump were compared using OPLS-DA to identify universal differences between the two breeds (Figure 6a). Two breeds were clearly differentiated which meant that the VOC profiles of Hanwoo and Chikso rump were different. Wei et al [6] reported that the difference in the VOC of different breeds of cattle might be attributed to the fatty acid composition or intramuscular fat content, derived from the genetic difference in lipid metabolism. Furthermore, numerous metabolites including amino acids and nucleotides participate in flavor formation, and therefore the difference in metabolites may also affect the change in VOC profiles of beef [30]. While most VOCs with a VIP>1 were more abundant in the Hanwoo rump than in the Chikso rump, the contents of 4-methylheptan-2-one and 1, 4-xylene were higher in the Chikso rump. During wet aging, the Hanwoo rump showed more remarkable VOC changes than the Chikso rump (Figure 6b).

A total of 41 VOCs were identified in the Hanwoo and Chikso loins (Supplementary Table S2). It is composed of 14 alcohols, 9 aldehydes, 5 esters, 1 furan, 5 hydrocarbons, 4 ketones, and 3 volatile fatty acids. Most VOCs were fewer in the Chikso loin than in the Hanwoo loin. Instead, 1, 4-xylene and 2, 4-dimethylhepten-1-ene were generally more abundant in the Chikso loin throughout the wet aging period than in the Hanwoo loin (Figure 6c).

During wet aging, Hanwoo loin showed a general increase in VOCs; however, Chikso loin only showed a gradual rise in alcohols, but the amounts of aldehydes and esters decreased after 28 days of wet aging (Figure 5), indicating that these two breeds showed different trends in VOC profile changes during wet aging (Figure 6d). Considering the origin of these VOCs, the degree of lipid oxidation might influence the volatile flavor characteristics of Hanwoo and Chikso loins. Chikso loin showed higher antioxidant activities (ABTS radical scavenging activity and ferric reducing antioxidant power) than Hanwoo loin on day 0, which might explain the overall lower quantities of lipid oxidation-derived VOCs in Chikso loin than in Hanwoo loin (Supplementary Figure S1). Overall, VOCs decreased gradually in the Chikso rump during wet aging, while 14 days of wet aging led to abundant VOC content in the Chikso loin.

Electronic tongue analysis was conducted using seven sensors for Hanwoo and Chikso beef on days 0, 7, 14, and 28 (Supplementary Figure S2). The sensors AHS, CTS, NMS, ANS, and SCS respond to sour, salty, umami, sweet, and bitterness, respectively, whereas PKS and CPS represent universal taste intensity. The PLS-DA plot for the electronic tongue results showed that the taste characteristics of the Chikso rump on day 0 were clearly separated from those of Hanwoo rumps on days 0 to 28 and Chikso rumps on days 7 to 28 (Figure 7a). The characteristics of the Chikso rump on day 0 were high values of relative taste intensity for NMS and CPS, indicating a higher intensity for umami taste and universal taste, respectively. In addition, Chiko rump on day 0 showed overall high relative taste intensities, except SCS (bitterness) and PKS (universal), compared to Hanwoo rump on day 0. Umami taste is mainly derived from umami amino acids (aspartic and glutamic acid) and nucleotides (IMP and guanosine 5′-monophosphate [GMP]) [34]. Particularly, the synergistic effect of IMP and glutamic acid could improve the strength of umami taste considerably [8]. Here, the IMP content was the highest in Chikso rump on day 0 of wet aging (Table 2), and a negative correlation between IMP and sensor intensity of NMS was identified (r = −0.54; Supplementary Table S4). The lower sensor intensity of NMS indicates a stronger taste intensity for umami taste, and therefore, high content of IMP in Chikso rump on day 0 might contribute to the high umami intensity in electronic tongue analysis.

Additionally, the Chikso rump on day 14 also had different characteristics than the other groups; it showed high values of taste intensity for ANS, SCS, and PKS, which correspond to sweetness, bitterness, and universal taste, respectively. Lee et al [14] reported that dry and wet aging of beef led to the increase of umami taste intensity. However, in our study, umami intensity of Chikso rump decreased considerably from day 0 to 7, although it gradually increased afterwards. Previous studies have reported that the taste characteristics of alanine, glutamine, and glycine impart sweetness, while isoleucine, leucine, phenylalanine, tyrosine, and valine could contribute to bitterness [9]. Moreover, reducing sugars can contribute to the sweetness of the meat, and anserine, carnosine, and nucleosides such as inosine and hypoxanthine provide bitterness [14,30]. In this study, the contents of most of the above-mentioned metabolites were significantly increased during wet aging, which would lead to high intensity for sweetness and bitterness. Especially, considering that umami intensity decreased while bitterness increased dramatically during 14 days of wet aging in Chikso rump, the degradation of umami contributors such as IMP which generated inosine and hypoxanthine at the early phase of wet aging might influence the change of taste attributes of Chikso rump considerably during wet aging. Previous study also noted that PKS is used for the general purpose of taste differentiation in electronic tongues [8], and the relatively high intensity of the Chikso rump on day 14 for the PKS sensor indicates that the universal taste of the Chikso rump on day 14 may be distinguishable from others. In contrast, wet aging for 14 days in Hanwoo rump increased sour taste intensities but lowered other taste intensities. The increase of sourness in wet-aged beef might derive from the accumulation of lactate by the action of lactic acid bacteria [35]. During wet aging for 14 days, lactate content was significantly increased in Hanwoo rump (Table 2).

The PLS-DA plot for taste characteristics of Hanwoo and Chikso loins revealed that loins from the two breeds showed a clear separation on day 0 (Figure 7b). In particular, higher umami intensity was found in Hanwoo loin on day 0 than in Chikso loin. Multivariate analysis also implied that the wet aging period of 14 days for Chikso loin would lead to different taste attributes compared with days 0, 7, or 28. Chikso loin on day 14 showed higher values of ANS, SCS, PKS, and CPS sensor intensities among treatments, indicating a strong intensity of sweetness, bitterness, and universal taste. In contrast, Chikso and Hanwoo loin on day 28 led to high sour attributes, possibly owing to the increased amount of organic acids including acetate, formate or fumarate (p<0.05). This was possibly due to increased alanine, phenylalanine, tyrosine, acetic acid, and formic acid in meat wet-aged for a long period [8,30]. The results suggested that the umami taste intensity was high in Chikso rump and Hanwoo loin on day 0. In contrast, the wet aging process for 14 days was effective in enhancing overall taste intensities and taste-active compounds in the Chikso beef.