2020 年 26 巻 3 号 p. 329-338
The amount and fatty acid composition of intramuscular fat are considered to contribute to beef tenderness and aroma; however, the effect on taste-traits remains poorly understood. This study aimed to develop a suitable beef broth sampling method using an electronic taste sensing system to clarify the influence of intramuscular free fatty acids (FFAs) on the taste of Japanese Black Wagyu beef. Japanese Black Wagyu and Holstein beef broths were prepared by four methods during postmortem aging and analyzed using the electronic taste sensing system. For comparison, the cooked samples were also subjected to sensory evaluation. The results of this study revealed that boiling is the most appropriate method for beef taste-trait estimation, as the taste-traits of boiled broth analyzed by the electronic taste sensing system coincided well with those analyzed by sensory evaluation, especially for the taste-traits of umami and sweetness. The increase in amounts of FFAs in beef broth during postmortem aging likely influences the taste-traits of Japanese Black Wagyu beef.
In beef production, postmortem aging is an essential process in converting muscle to meat and is used to improve beef palatability, such as tenderness, juiciness and flavor, to satisfy consumer preferences (McGee, 2004; Iida et al., 2015). Meanwhile, fat also plays an important role in deliciousness of beef, since it is known that intramuscular marbling contributes to the tenderness and juiciness of beef (Okumura et al., 2007; Iida et al., 2015). In addition, beef flavor derived from fat is influenced not only by the amount of intramuscular fat but also its quality, in other words, the fatty acid composition of triacylglycerol (TG). For example, it has been previously reported that intramuscular fat rich in monounsaturated fatty acids contributes to beef flavor and chemical characteristics (Westerling et al., 1979; Melton et al., 1982).
Japanese Black Wagyu beef is popular in Japan and is exported around the world, having recently been recognized for its deliciousness, which is influenced by the extensive marbling of the meat and the fatty acid composition of enriched oleic acid in the intramuscular fat (Oka et al., 2002). In terms of taste, the effect of fatty acids on taste-traits is regarded to be of increasing importance, as increases in the amount of free fatty acids (FFAs) in Japanese Black Wagyu beef during postmortem aging could contribute to its taste quality. Research on the perception of FFAs by rats revealed that rats preferred long-chain fatty acid fluids in short-term, two-bottle tests (Tsuruta et al., 1999). It was also suggested that long-chain fatty acids such as oleic acid enriched umami but inhibited bitterness and sourness tastes (Pittman et al., 2006). Also, fatty acid receptors are present in the taste buds of rat and human tongues (Matsumura et al., 2007; Galindo et al., 2012). Thus, changes in the amounts and types of FFAs during postmortem aging and cooking may affect the desirable taste of Japanese Black Wagyu beef.
Sensory evaluation by human panels is a well-recognized method used to objectively analyze beef taste-traits. However, the procedures for panel selection and training are more complex for beef evaluation as compared to other foods (Brachieri et al., 2007). In addition, beef taste-traits are composed of multiple attributes, making sensory evaluation challenging. Thus, it is vital that novel rapid and objective estimation methods are developed as substitutes for traditional sensory evaluation of beef taste-traits. The electronic taste sensing system was recently developed as a method to evaluate the taste-traits of foods and drugs (Kobayashi et al., 2010). The equipped taste sensors mimic the human tongue to convert taste signals by various substances in foods and drugs into numerical data. The taste sensors consist of artificial lipid membranes (similar to that of the human tongue) with specific electric potentials that cause electrostatic or hydrophobic interactions with various taste substances, allowing them to sense “taste” (Toko, 1996; Kobayashi et al., 2010). By employing the unique aftertaste measurement technology of the system, even taste aspects such as “richness” and “sharpness” can be numerically expressed (Chikuni et al., 2010; Kobayashi et al., 2010). Recently, reports of the analysis of taste-traits of meat and meat products using this system have been gradually increasing. Sasaki et al. (2005) applied the electronic taste sensing system to the discrimination and evaluation of pork quality. They indicated that the system was useful for the discrimination of pork extract, and there was a correlation between sensor output and umami substances by principal component scores. Chikuni et al. (2010) reported differences in beef taste-traits such as sourness and bitterness among different muscle types using the taste sensing system. Additionally, sensory evaluation, taste-trait analysis using the system and chemical analysis of pork coincidentally showed a common increase in the intensity of saltiness and bitterness during pork curing (Nodake et al., 2013). Together, these results demonstrate that taste sensors can be used as a substitute for human sensory evaluation of meat taste-traits because of its high correlation with sensory score. However, this correlation can be invalidated when samples from the same meat species but prepared according to different procedures are applied to the electronic taste sensing system. Thus, the establishment of a standard sample preparation method for beef taste-trait analysis using the system is needed.
The purpose of the present study was to establish a beef broth sample preparation method enabling high correspondence between the electronic taste sensing system and human sensory evaluation, which could then be applied to elucidate the effects of intramuscular FFAs on beef taste-traits. Therefore, we prepared beef broth samples according to four different methods in this study and estimated the taste-traits of samples using the electronic taste sensing system as well as sensory evaluation.
Chemicals and reagents All research grade chemicals used in the electronic taste sensing system and research grade reagents of tartaric acid, caffeine, monosodium glutamate (MSG) and sucrose used in the human sensory evaluation were obtained from Wako Pure Chemical Industries (Osaka, Japan).
Beef samples The rearing periods of seven Japanese Black Wagyu and six Holstein cattle were about 28 and 20 months of age, respectively. All cattle were slaughtered at local abattoirs and the dressed carcasses were stored in the refrigerator of the meat-processing center at least overnight. Beef samples taken from the longissimus thoracis muscle of the sixth and seventh ribs of the dressed carcasses were sent in vacuum packaging at 4 °C to Kobe University within 7 days after slaughter. At our laboratory, the beef samples were immediately removed from the packaging and wrapped in preservative film sprayed with 75% ethanol, and then stored at 4 °C until sampling. Samples taken from both muscles including intramuscular fat tissues at 7-, 14-, and 21-days postmortem were cooked on an electric griddle to an internal temperature of 65–70 °C, and used as the cooked samples.
Preparation of beef broth samples Beef broth samples of the two breeds at the predetermined postmortem times (see above) were prepared according to the four preparation methods as follows. Distilled water broth samples (WB) were prepared by Method 1, which mainly focused on fat removal by homogenization and low-temperature solidification according to Nodake et al. (2013), with a slight modification. Briefly, 4 g of cooked samples were homogenized in 10 mL of distilled water using an AM-T homogenizer (Nihonseiki, Tokyo, Japan) at 10 000 rpm for 2 min. The homogenates were transferred to a cylinder and adjusted to 15 mL before being cooled at 4 °C on crushed ice and then centrifuged at 5 000 rpm for 20 min at 4 °C. The supernatants were filtered and adjusted to 10 mL with distilled water. Ethanol broth samples (EB) were prepared by Method 2, which mainly focused on fat removal by homogenization in 15% ethanol solution according to Chikuni et al. (2010), with a slight modification. Four grams of cooked samples were homogenized in 10 mL of 15% ethanol solution (room temperature). The homogenates were centrifuged at 5 000 rpm for 20 min at 25 °C. The supernatants were filtered and adjusted to 10 mL with 15% ethanol solution. Boiled broth samples (BB) were prepared by Method 3, which mainly focused on fat removal by boiling and low-temperature solidification according to Sasaki et al. (2005). Four grams of raw samples were minced and added to 20 mL of distilled water in a glass beaker, which was covered with aluminium foil. The samples were boiled and stirred every 15 min, and distilled water was added to maintain a volume of 20 mL throughout the 1 hr boiling period. Then, the samples were cooled at 4 °C on crushed ice. The supernatants obtained by filtration were adjusted to 10 mL with distilled water. Deproteinized broth samples (DB) were prepared by Method 4, which mainly focused on fat removal by homogenization, centrifugation and, in contrast to the other three methods, deproteinization. Deproteinization was employed to exclude the effects of proteins other than as taste substances on taste-traits. Briefly, 4 g of cooked samples were homogenized in 10 mL of distilled water, and the homogenates were centrifuged at 5 000 rpm for 20 min at 25 °C. The supernatants were filtered and deproteinized with 80% ethanol (final concentration) solution. Then, the supernatants were filtered again and evaporated, and finally adjusted to 10 mL with distilled water.
All beef broth samples were frozen at −80 °C until used to estimate taste-traits by the electronic taste sensing system and sensory evaluation, and to measure the amounts of FFAs by commercial quantification kits.
Measurement of pH pH was determined by the automatic temperature compensation (ATC) glass electrode method using the multi-function water quality meter model MM-43X (DKK-TOA Corporation, Tokyo, Japan) at room temperature (25 °C).
Measurement of FFA amounts The amounts of FFAs in beef broths were photometrically measured using a NEFA C commercial quantification kit (Wako Pure Chemical Industries Ltd.) according to the acyl-CoA synthetase and acyl-CoA oxidase (ACS-ACOD) method, which is based on absorbance increase with the formation of quinoneimine dye (Shimizu et al., 1980). Forty microliters of color A (dissolved in solvent A) was added to 50 µL of BB and incubated for 10 min at 37 °C. Then, 80 µL of color B (dissolved in solvent B) was added to the mixture. After incubating for 10 min at 37 °C, the absorbance was measured at 550 nm and the amounts of FFAs were calculated from the calibration curve using oleic acid as a standard.
Taste-traits by the electronic taste sensing system The SA402B electronic taste sensing system (INSENT, Kanagawa, Japan) equipped with six taste sensors of saltiness (CT0), sourness (CA0), umami (AAE), bitterness (C00), astringency (AE1) and sweetness (GL1) was used for the taste-trait analysis of beef broths. The taste sensors evaluate two types of taste according to the manufacturer's manual, namely initial tastes (saltiness, sourness, umami, acid bitterness, astringency and sweetness), which are the tastes perceived when food first enters the mouth, and aftertastes (bitterness, aftertaste from astringency and richness), which are tastes that remain on the tongue after the food has been swallowed. One disadvantage of the sensors is that they cannot be applied to samples high in lipids (Fujimura and Sasaki, 2013) and this problem has persisted in recent years. However, beef broth samples in which pretreatment procedures were employed to exclude fat, described above, were used to identify a suitable preparation method for analysis by the system. Briefly, each 10-mL beef broth sample prepared by the above methods was divided in half. One half of the broth was diluted to 35 mL for the sweetness measurement and the other half was diluted to 70 mL for the other taste measurements, respectively. The measurement procedure was automatically carried out at 25 °C under the default conditions previously described by Ikezaki et al. (1997).
Taste-traits by human sensory evaluation The sensory evaluation, five basic tastes training and discrimination test were conducted according to the Japanese Guidelines for Sensory Evaluation of Meat (2005) and the Outline of Ethical Guidelines for Medical and Health Research Involving Human Subjects of the Ministry of Education, Culture, Sports, Science and Technology, Japan. All beef broth samples were evaluated by 14 trained panelists (6 males and 8 females, 21 to 26 years of age) who were students of Kobe University and selected using the five basic tastes training. Namely, the panelists were repeatedly (3 times) trained to identify the specific tastes with five standard samples that consisted of 0.13 g/dL NaCl as saltiness, 0.005 g/dL tartaric acid as sourness, 0.02 g/dL caffeine as bitterness, 0.05 g/dL MSG as umami and 0.4 g/dL sucrose as sweetness at 25 °C, respectively. One of the training sessions involved evaluation of discrimination ability between two different taste intensities of sourness, saltiness, umami and sweetness for additional selection. For each sensory evaluation session, the test room was illuminated by white fluorescent lights and air-conditioned at a constant temperature of 25 °C. The panelists were isolated from each other by desk partitions and their nostrils were closed to exclude the influence of aroma. Each 5-mL broth sample was diluted to 35 mL and randomly evaluated. Between each sample evaluation, panelists took a break and rinsed their mouths with bottled purified water. Following each sensory evaluation session, panelists recorded the results in questionnaires, which included 7 taste-traits evaluated by scoring of 1, 2, 3, 4, 5 and 6, which corresponds to equal, very weak, a little weak, a little strong, strong and very strong compared to the standard, respectively. Two taste-traits of acid bitterness and astringency analyzed by the electronic taste sensing system were not evaluated by sensory evaluation; thus, these two tastes are shown as “NA, not analyzed” in the results.
Statistical analysis All results were expressed as mean values. Statistical analysis was performed using JMP®13.2.1 software (SAS Institute Inc., Cary, NC, USA). The effects of preparation method, postmortem time and breed on the taste-traits of beef broths in Table 1 were analyzed by the NESTED procedure (Snedecor and Cochran, 1967). The amounts of FFAs in Fig. 3 were compared by one-way analysis of variance (ANOVA) with post hoc analysis using Tukey's HSD test among aging time, and by a Student's t-test between the two breeds at the predetermined postmortem times (Snedecor and Cochran, 1967).
| Taste element | Method | Postmortem time(days) | Breed | P-value | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 7 | 14 | 21 | JBW | HOL | Method | Postmortem time | Breed | |
| Sourness | −26.68b | −25.10b | −28.73c | −20.46a | −25.29 | −25.17 | −27.28 | −25.35 | −23.40 | ** | NS | NS |
| Acid bitterness | 3.54a | 3.71a | 2.23b | 1.09c | 1.87b | 2.25b | 3.62a | 2.59 | 2.31 | ** | * | NS |
| Astringency | −1.41 | −1.98 | −1.96 | −2.07 | −1.56c | −1.55c | −0.65a | −0.38a | −2.92c | NS | ** | ** |
| Umami | 6.47b | 6.20b | 7.13a | 5.28c | 6.23c | 6.49c | 7.14a | 7.23 | 6.18 | ** | ** | NS |
| Saltiness | −10.73 | −10.70 | −9.64 | −11.29 | −10.83b | −10.70b | −9.27a | −10.12 | −9.49 | NS | * | NS |
| Bitterness | −0.15 | −0.22 | −0.25 | 0.04 | −0.11 | −0.04 | −0.25 | −0.12 | −0.14 | NS | NS | NS |
| Aftertaste from astringency | −0.11 | −0.08 | −0.05 | −0.05 | −0.02 | −0.09 | −0.11 | −0.04 | −0.08 | NS | NS | NS |
| Richness | 0.92 | 0.83 | 0.94 | 0.75 | 0.75b | 0.81b | 1.17a | 0.29b | 1.50a | NS | * | * |
| Sweetness | 5.23b | 6.58a | 7.01a | 5.63b | 5.65c | 5.22c | 8.04a | 6.10 | 6.72 | * | ** | NS |
JBW Japanese Black Wagyu (n = 4); HOL Holstein (n = 4)
Taste-traits of beef broths prepared by different methods at predetermined postmortem times In order to investigate the effects of preparation method, postmortem time and breed on the characteristics of beef taste-traits, we compared all sensor outputs among the beef broth samples estimated by the system (Table 1). In the NESTED design, there were three factors of preparation method, postmortem time and breed, but where postmortem time was nested within breed, and preparation method was nested within postmortem time and breed. As a result, we determined the effect of preparation method on the taste-traits of sourness, acid bitterness, umami and sweetness, and the effect of postmortem time on acid bitterness, astringency, umami, saltiness, richness and sweetness. The effect of breed was found on both astringency and richness. In detail, there were significant differences in the taste-traits of sourness (P < 0.01), acid bitterness (P < 0.01), umami (P < 0.01) and sweetness (P < 0.05) among the preparation methods of beef broth samples, respectively. Additionally, significant differences in the taste-traits of acid bitterness (P < 0.05), umami (P < 0.01) and sweetness (P < 0.01) were found between 7- and 21-days postmortem, respectively. Significant differences in the taste-traits of astringency and richness were found in “Postmortem time” and “Breed” (P < 0.01 and P < 0.05, respectively), whereas no significant differences were found in “Method”. In other words, both the preparation method and postmortem time were considered to affect the taste-traits of acid bitterness, umami and sweetness in the beef broths from the two breeds. There were significant differences in the two taste-traits of sourness in “Method” and saltiness in “Postmortem time” (P < 0.01 and P < 0.05, respectively); however, we did not discuss those taste-traits in detail, as they are regarded as tasteless in human sensing (Kobayashi et al., 2010). From the above results, we focused on the taste-traits of acid bitterness, umami and sweetness in beef broth samples prepared by the four methods at predetermined postmortem times (Figs. 1 and 2).
Effects of different preparation methods on taste-traits of beef broth from longissimus thoracis muscles of Japanese Black Wagyu and Holstein cattle at 21-days postmortem estimated by the electronic taste sensing system
Broth samples from Japanese Black Wagyu (a) and Holstein (b) beef. Details of preparation methods of WB by Method 1 
, EB by Method 2 
, BB by Method 3 
and DB by Method 4 
are described in “Materials and Methods”. Results are expressed as mean values (Japanese Black Wagyu, n = 6; Holstein, n = 6)
Human sensory evaluation of beef broth prepared by different preparation methods from longissimus thoracis muscles of Japanese Black Wagyu and Holstein cattle at 21-days postmortem
Broth samples from Japanese Black Wagyu (a) and Holstein (b) beef. Details of preparation methods of WB by Method 1 
, EB by Method 2 
, BB by Method 3 
and DB by Method 4 
are described in “Materials and Methods”. Results are expressed as mean values (Japanese Black Wagyu, n = 4; Holstein, n = 4)
Taste-traits of beef broths prepared by different methods at 21-days postmortem We showed the results of taste-traits of broth samples prepared by four different methods at 21-days postmortem from cooked Japanese Black Wagyu and Holstein beef by the electronic taste sensing system (Fig. 1) in comparison with the results of sensory evaluation of the same samples. DB samples from Japanese Black Wagyu and Holstein beef at 21-days postmortem were analyzed to exclude the effects of proteins in our laboratory, and then used as the control for each breed. As shown in Fig. 1a, different patterns of taste-traits in WB, EB and BB prepared from Japanese Black Wagyu beef were found; namely, much lower values of sourness and higher values of acid bitterness, astringency, umami and sweetness as compared to those of the control (DB). This may be attributable to the deproteinization of control beef broth samples. On the other hand, the taste-traits of broth samples from Holstein beef are shown in Fig. 1b. Except for saltiness, the trends were similar to Wagyu beef, where much lower values of sourness and higher values of acid bitterness, astringency, umami and sweetness were found in WB, EB and BB compared to the control. Further, acid bitterness and saltiness showed greater increases, while sourness and sweetness decreased as compared to those from Japanese Black Wagyu beef. Especially, sweetness and umami showed greater output values in BB from Japanese Black Wagyu and Holstein beef, respectively. Differences in the values of umami and sweetness among preparation methods could be explained by the characteristics of AAE and GL1 sensors. The AAE sensor was designed to respond to umami substances such as amino acids (Toko, 1996). In addition, it has been reported that the umami sensor output increased with elevations in pH (Fujiwara and Kurita, 2011; Fujiwara and Ishikawa, 2012). The results of pH measurement in WB, EB and BB from Japanese Black Wagyu beef were 5.8, 5.9 and 6.0, respectively, and could explain the high umami values in BB. Similar results were also found in the broth samples from Holstein beef. On the other hand, the GL1 sensor output seems to be different from the inherent sweetness of beef, since GL1 is basically made to respond to more than 3% sugar and sugar alcohol (Toyota et al., 2011).
Sensory evaluation of beef broths prepared by different methods at 21-days postmortem Prior to sensory evaluation, the taste values shown in Fig. 1 were compared to those of the reference solution. The reference solution was a tasteless solution containing 30 mM KCl and 0.3 mM tartaric acid, which, according to the manufacturer's manual, is used as a substitute for human saliva. We found that the values of sourness and saltiness in all beef broth samples (WB, EB, BB and DB) were below the values obtained for the reference solution (data not shown), indicating that those tastes were regarded as tasteless in human sensory evaluation (Kobayashi et al., 2010). Additionally, the two taste-traits of acid bitterness and astringency analyzed by the electronic taste sensing system were difficult to evaluate by sensory evaluation. Thus, we focused on umami and sweetness among the beef broth samples from the two breeds in sensory evaluation.
In order to establish an optimal preparation method for beef taste-trait estimation by the electronic taste sensing system, sensory evaluation of each beef broth by a human panel is needed. Control samples used in sensory evaluation were the same as the DB used in the electronic taste sensing system analysis. It is conceivable that the taste-traits of broth samples from Japanese Black Wagyu and Holstein beef determined by sensory evaluation could differ from those by the electronic taste sensing system (Fig. 2). Firstly, as mentioned above, the taste-traits of sourness and saltiness were excluded because they were considered to be tasteless in sensory evaluation. Secondly, noticeable differences in the taste pattern of EB (broken line) appeared, especially for bitterness and aftertaste from astringency of both beefs. These results imply that the ethanol solution may strongly affect all taste-traits in sensory evaluation by the human panel, resulting in difficulties in comparing the results to taste-traits of EB determined by the electronic taste sensing system. High values of richness in WB and BB were probably related to the increased outputs of acid bitterness and umami analyzed by the electronic taste sensing system (Kobayashi et al., 2010), although acid bitterness was not assessed by sensory evaluation. As a consequence of individual comparisons of taste-traits between the two beef broths, the taste-traits of BB analyzed by the electronic taste sensing system coincided well with those analyzed by sensory evaluation, especially in the taste-traits of umami and sweetness. Therefore, it was concluded that BB among the four different preparation methods for beef broth was a suitable method for the estimation of beef taste-traits by the electronic taste sensing system.
Changes in FFA amounts of BB during postmortem aging As shown in Fig. 3, the amounts of FFAs in BB from Holstein beef were 5.7 mg and 6.8 mg at 7- and 14-days postmortem, respectively, and increased to 8.4 mg at 21-days postmortem. On the other hand, the amounts of FFAs in BB from Japanese Black Wagyu beef were 6.8 mg and 7.9 mg at 7- and 14-days postmortem, respectively, and increased significantly to 9.3 mg at 21-days postmortem (P < 0.05). This indicated that postmortem aging has increasing effects on the amounts of intramuscular FFAs in BB from Japanese Black Wagyu and Holstein beef. However, the amount of FFAs in Japanese Black Wagyu beef was obviously greater compared to Holstein beef, indicating that the amount of FFAs is related to the taste-traits of Japanese Black Wagyu beef.
Effects of postmortem aging on the amounts of intramuscular FFA in BB from longissimus thoracis muscles of Japanese Black Wagyu and Holstein cattle
Broth samples from Japanese Black Wagyu (■) and Holstein (□) beef. Results are expressed as mean values (Japanese Black Wagyu, n = 7; Holstein, n = 5). *Significant difference at postmortem time (P < 0.05)
Changes in taste-traits of BB during postmortem aging To clarify the effects of intramuscular FFAs on beef taste-traits, BB from Japanese Black Wagyu and Holstein beef were analyzed using the electronic taste sensing system during postmortem aging. The samples were the same as those subjected to FFA measurement as described above (Fig. 3). As shown in Fig. 4, each BB at day 7 in early postmortem was used as the control sample in assessing the effects of postmortem aging. Sourness and saltiness decreased markedly, whereas acid bitterness, astringency, umami and sweetness increased in BB from Japanese Black Wagyu beef during postmortem aging (Fig. 4a). Organic acids such as lactic acid highly affected both the CA0 sensor output and pH of the beef broth. Decreases in sourness during postmortem aging can be attributed to increases in pH of 5.8, 5.9 and 6.1 in BB. These data are consistent with the report of Spanier et al. (1997). The decrease in saltiness during postmortem aging can be explained by the reported negative effect of fat on sodium release and on the saltiness intensity of chicken sausages in sensory evaluation (Chabanet et al., 2013).
Effects of postmortem aging on taste-traits of BB from longissimus thoracis muscles of Japanese Black Wagyu and Holstein cattle at appropriate postmortem time estimated by the electronic taste sensing system
Broth samples from Japanese Black Wagyu (a) and Holstein (b) beef. Postmortem time are at 7-days 
, 14-days 
, and 21-days 
. Results are expressed as mean values (Japanese Black Wagyu, n = 4; Holstein, n = 4)
Chikuni et al. (2010) reported that the outputs of C00 and AE1 sensors indicate the complexity of taste substances, since both sensors might respond to certain hydrophobic compounds such as fatty acids and the C00 sensor responds to low concentrations of FFAs in beef. They also showed that the contents of FFA were found to be significantly higher in the slow-type muscles of beef and were correlated to the higher values of acid bitterness and astringency. In this study, acid bitterness determined by the C00 sensor showed higher values especially in broth samples from Japanese Black Wagyu beef at 21-days postmortem, which was attributed to increases in the amounts of FFAs during postmortem aging, as shown in Fig. 3. At the same time, increases in the amounts of FFAs may elevate richness values as reported previously (Yamanoue et al., 2013). In addition, the increases in free amino acid content in cooked beef broth samples during postmortem aging (data not shown) were considered to relate to the increasing values of umami, since the AAE sensor was designed to respond to umami substances such as amino acids (Toko, 1996). The increase in contents of several amino acids was suggested to be involved in the increased umami intensity of pork sausage during sensory evaluation (Nodake et al., 2013). Obviously increased sweetness may be due to high activities of total hydrolytic enzymes during postmortem aging (Spanier et al., 1990).
On the other hand, similar to the results of Japanese Black Wagyu beef, sourness and saltiness decreased markedly, whereas acid bitterness, astringency, umami and sweetness increased in BB from Holstein beef during postmortem aging (Fig. 4b). Those changes in taste-traits can be explained by the same reasons given for Japanese Black Wagyu beef. In particular, we observed greater increases in sweetness in Holstein beef than in Japanese Black Wagyu beef at 21-days postmortem; however, the reason for this is unclear.
From above results, the intensities of taste-traits such as acid bitterness, umami and sweetness in BB from Japanese Black Wagyu beef were highly similar to or greater than those from Holstein beef, even though the amounts of taste substances such as free amino acids in Holstein beef were greater compared to Japanese Black Wagyu beef (manuscript in preparation). Thus, it is suggested that increases in the amount of FFAs in broth samples from Japanese Black Wagyu beef during postmortem aging are likely associated with the intensities of those taste-traits.
The boiling method, described as Method 3, in this study is considered to be the most appropriate preparation method for estimating taste-traits by the electronic taste sensing system with high correspondence to sensory evaluation. In addition, increases in the amount of FFAs in beef broth during postmortem aging are likely associated with the taste-traits of Japanese Black Wagyu beef. Further investigation of beef taste-traits in terms of intramuscular FFAs and taste substances with the boiling method are needed.
Acknowledgements We acknowledge the financial support of the Ito Foundation (Tokyo, Japan) and Grant-in-Aid for Scientific Research (C) (16K08007) of Japan Society for the Promotion of Science for this research.
