The Headspace Volatiles Generated at the Initial Stage of Spoilage in the Aerobically and Vacuum Packed Beef, Pork

The Headspace Volatiles Generated at the Initial Stage of Spoilage in the Aerobically and Vacuum Packed Beef, Pork and Chicken Yukio YANO, Tsuneaki MAEDA* and Takashi HIRATA** Central Research Institute of Itoman Foods INC., Moriya-machi, * Development Center of DKK Corporation ** Department of Agriculture Ibaraki-ken 302-01, Musashino-shi 180, Kyoto University, Sakyo-ku, Kyoto 606 (Received January 31, 1995) Abstract Headspace volatiles in aerobically and vacuum packed beef, pork and chicken meat were identified. During storage the number of headsoace components increased and some components which were not detected in fresh meat appeared. At the initial stage of bacterial spoilage, ethyl acetate, 3-methyl butanal, iso-amylalcohol and acetic acid appeared. Of those components, ethyl acetate was the predominant compound and the amount increased as the deterioration of meat progressed. The appearance of a faint fruity odor at the initial stage of spoilage was coincident with the detection of ethyl acetate. For evaluating meat freshness, ethyl acetate is thought to be a useful chemical index. Anim, Sci. Technol. (Jpn.) 66 (8):684-692, 1995 Key words: meat freshness, headspace volatiles, ethyl acetate The formation of off-odors is one of the typical symptoms for bacterial spoilage of meat. Many studies have been done to clarify the relationship between the type of odors, the bacterial species and the odor components. MCMEEKIN (1975) isolated pure cultures from spoiling chicken breast muscles and tested the ability to produce strong off odors. Three distinctive types of odors described as sulfidelike, fruity and evaporated milk were recognized. In the study of FREEMAN (1976), chicken muscle was stored aerobically and anaerobically, and the volatile compounds were analyzed. In the same study, the cultures isolated from the stored meat were inoculated to sterile meat and the volatiles produced by those cultures were identified. SUTHERLAND et al. (1975) studied the microbial flora which developed in vacuum packaged beef and also in aerobically stored beef after vacuum packing. When Pseudomonas spp. was prevalent in the additional aerobic storage, slime formation and a sweet-rotten odor generally occurred. On the other hand, when lactic acid bacteria were predominently acidic/sour spoilage odors were noticeable. According to SUTHERLAND et al. (1976), the sour/acid odor was observed on opening of vacuum-packaged beef. From the result of gas-liquid chromatography (GLC) analysis a number of lower carboxylic acids were present. In the review of DAINTY (1982) off-odors were classified into sulphur, fruity, amine/ammonia, acidic, malty, lipolytic/rancid and other miscellaneous odors. Further, these characteristic Anim. Sci. Technol. (Jpn.) 66 (8):684-692 684 1995

odors were connected with specific compounds produced by bacteria. As mentioned above, the study of volatile components in stored meat have been performed mainly at the putrefied stage and the volatiles in the putrefied meat were identified. In contrast, there have been no study of the head space volatiles at the initial stage of spoilage which is important for the estimation of shelf-life in meat. To analyze the odor components of meat, the headspace analysis method is generally thought to be most suitable. By comparing the sniffing odor of meat to the headspace analysis by gas chromatography (GC), it is possible to estimate the key component contributing to the spoilage odor. The purpose of this study was to analyze the headspace volatiles during storage of meat and to identify the volatile components produced at the initial stage of spoilage. Such components would be useful for evaluating meat freshness, and this may be useful in the development of a gas-sensor which can evaluate meat freshness nondestructively. Materials and Methods Sample preparation: Beef: Sirloin meat was obtained from the carcass of a 24-month-old Holstein steer for 2 days after slaughter. Pork: Loin meat was obtained from the carcass of a 6-month-old LWD barrow for 1 day after slaughter. Chicken: Thigh meat was obtained from the carcasses of a 2-month-old arbor-acre fowl on the day of slaughtering. All meats were cut into 2-mm thick slices. For aerobic packaging, meat was packed in bags of gas-permeable film (OPP/CPP, 0.05mm thick), and for vacuum packaging, meat was vacuum packed in bags of high barrier film The Headspace Volatiles of Meat (Nylon/Binding layer/l. LDPE, 0.07mm thick) with a vacuum-packaging machine (Model FVC-H II-G, Old Rivers Co. Ltd., 685 Satte, Japan) which pulled the vacuum of -990 hpa. The samples were stored Measurement of bacterial counts: Ten grams of meat was chopped up using sterilized surgical scissors and homogenized with 90ml of sterile 0.9% saline solution. Decimal dilutions were spread over the plates of standard method agar (Eiken Chemical Co., Tokyo, Japan) for total aerobic viable counts. Colonies on the plate were counted after incubation for 5 days at Measurements for each sample were performed in duplicate. Trapping of volatiles of meat: A schematic of the apparatus is presented in Fig. 1-(A). One hundred grams of minced sample was put in a 500ml glass bottle. The bottle was placed in a and a Tenax GC trapping tube was set on the glass bottle. Head space volatiles of meat were trapped in the Tenax GC which contained 0.8ml Tenax GC by passing a stream of helium (60 ml/min) for 1hr. The Tenax GC tube was then set on a flush heater FLS-3 (Shimadzu Ltd., Kyoto, Japan) for subsequent analysis. Co., Desorption and resorption of head space volatiles: Figure 1-(B) shows a schematic of the apparatus for desorption, resorption and GC- MS analysis. When the Tenax GC tube was with a flush heater around the tube, an adsorbing tube of the automated gas concentrator (GAS-20, DKK Co., Tokyo, Japan) adsorbing tube contained 0.2ml Tenax GC. The desorption of volatiles from the Tenax GC tube in heater was performed for 10min by passing a stream of helium (50ml/ min). At the time of GC-MS analysis, the adsorbing tube of the gas concentrator was heated to with a coiled heater around the tube and was maintained at that temperature for 4min. Helium carrier gas with a flow rate of 1.0ml/ min sweeps the trapped components into the trapping column Thikotes column Bonded MS (OV-1), film thickness 5.0

YANO, MAEDA and HIRATA (A) Collection of head space volatiles (B) GC-MS analysis Fig 1. Schematic diagram of the system for headspace analysis. (1) Tenax GC tube; (2) glass bottle; (3) meat sample; (4) cooler; (5) flow controller; (6) helium gas; (7) Tenax GC tube; (8) flush heater; (9) adsorbing tube; (10) sampling valve; (11) concentrated sample inlet valve; (12) automated concentrator ; (13) trapping tube; (14) capillary column; (15) gas chromatograph; (16) mass spectrometer; (17) helium gas; (18) liquid carbon dioxide; (19) compressor., Tokyo Kasei Kogyo Co., Ltd., Tokyo, Japan) in which the volatiles were concentrated by blowing liquid carbon dioxide for 4min, maintaining the temperature This trapping column was located at the inlet of the GC oven and connected to the capillary column. At the end of purge time, the blowing of liquid carbon dioxide was ceased, and then the temperature program of GC started automatically. GC-MS analysis: GC-MS spectra were recorded with a Shimadzu QP-1000A mass spectrometer equipped with a Shimadzu GC-15 A gas chromatograph. The ionization voltage was 70eV, and the ion source temperature was A chemical-bonded fused silica glass capillary column (ULBON HR-20M; egivalent to PEG 20M, film was held at for 10min, then programmed Helium was used as carrier gas and controlled at 1.0ml/min. The fragment ion (m/z) of over 20 were recorded using an ion chromatogram to eliminate the effect of water. Therefore, ammonia could not be determined in this study. Each volatile component was identified by matching their mass spectrum data. Dimethyl sulfide, 686

The Headspace Volatiles of Meat acetone, and acetoin which were dominant was shown in beef at 11 days, in pork at 5 days from the beginning of storage, and ethyl acetate, ethanol, 3-methyl butanal, iso-amyl- and in chicken at 4 days. On those days, the alcohol and acetic acid which increased during 107 cells/g, respectively. The type of odor was storage were identified by the retention time fruity, acidic, and fruity, respectively. On the with that of authentic compound as well as day a distinctive putrid odor appeared, the mass spectrum data. fruity odor was not recognized in the same manner of the aerobic package, and sulfide-like Results and Discussion and acidic odors became dominant except in Changes in odor and bacterial counts: Table beef. In vacuum packed beef, the acidic odor 1 shows the changes in odor and bacterial was more intense at 14 days compared with 11 counts during storage. In aerobic packaging, days. a faint putrid odor was shown in beef at 7 days, Changes in head space volatiles of meat: The in pork at 3 days and in chicken at 3 days. On ion chromatogram of the head space volatiles those days, the bacterial counts were in the aerobically packed beef is shown in Fig. 2. There were many unknown peaks because type of odor was fruity, fruity, and acidic, respectively. On the day on which a distictive nent. The number of peaks increased with of the small amount of each volatile compo- putrid odor was recognized, fruity odor was storage time. Table 2 shows the changes in not perceived and sulfide-like as well as acidic the peak area of the identified components odor became intense in all samples. during storage of beef. In fresh beef meat, In vacuum packaging, a faint putrid odor pentane, dimethyl sulfide, acetone, hexanal, 1- Table 1. Changes in odor and bacterial counts during storage of meat 687

YANO, MAEDA and HIRATA Fig. 2. The ion chromatograms of headspace volatiles in the aerobically packed beef during storage. pentanol and acetoin were recognized as the main components. In aerobically packed beef, ethyl acetate, 3-methyl butanal, 3-methyl-2- butanone, 2-methyl-1-propanol and isoamylalcohol appeared at the initial stage of bacterial spoilage. At the advanced stage of spoilage, methyl acetate, ethyl propionate and 1-hexanal appeared. The content of ethyl acetate, 3-methyl butanal, 3-methyl-2-butanone, iso-amylalcohol and acetoin increased compared with the initial stage of spoilage. As the fresh meat of vacuum packed beef was the same meat as the aerobically packed meat, only the results of initial and advanced stages of spoilage are presented in Table 2. At the initial stage of spoilage, ethyl acetate, 3- methyl butanal and 3-methyl-2-butanone appeared. In the last day of storage 1-octene, 2-methyl octane and acetic acid appeared. The content of ethyl acetate increased apparently compared with the initial stage of spoilage. Table 3 shows the changes in pork. In fresh pork meat, pentane, acetone, 3-methyl-2- butanone, 2-hexanol, 1-butanol, acetoin, 1- hexanol and nonanal were recognized as the main components. In aerobically packed pork, ethyl acetate, 3-methyl butanal, isoamylalcohol and acetic acid appeared at the initial stage of spoilage. At the advanced stage of spoilage, ethanol appeared and the content of ethyl acetate, 3-methyl butanal and iso-amylalcohol increased apparently compared with the initial stage of spoilage. In vacuum packaged pork, ethyl acetate, ethanol, 3-methyl butanal, iso-amylalcohol appeared at the initial stage of spoilage. At the advanced stage of spoilage, hexane and hexanal appeared, and the content of ethanol and iso-amylalcohol increased compared with the initial stage of spoilage. The content of ethyl acetate was the same as at the initial stage of spoilage. Table 4 shows the changes in chicken. In fresh chicken meat, only acetone was recognized as the main component. In aerobically packed chicken, ethyl acetate, 3- methyl butanal, iso-amylalcohol, acetoin and acetic acid appeared at the initial stage of spoilage. At the advanced stage of spoilage, the content of ethyl acetate, iso-amylalcohol, acetoin and acetic acid increased compared with the initial stage of spoilage. In vacuum packaged chicken, ethyl acetate, 3 -methyl butanal and acetoin appeared at the initial stage of spoilage. At the advanced stage of spoilage, 1-propanol, hexanal, isoamylalcohol and acetic acid appeared. The 688

The Headspace Volatiles of Meat Table 2. Changes in headspace volatiles during storage of beef a Peak area measured by ion chromatography Table 3. Changes in headspace volatiles during storage of pork a Peak area measured by ion chromatography 689

YANO, MAEDA and HIRATA Table 4. Changes in headspace volatiles during storage of chicken a Peak area measured by ion chromatography content of ethyl acetate increased compared with the initial stage of spoilage. Characteristics of headspace components: It was difficult to correlate the headspace components identified with the odor of the meat because many unidentified peaks with small amounts appeared, and low-threshold components can show an intense odor even in a small content. For instance, hydrogen sulfide which is known as a source of typical sulfide-like odor in putrid proteinaceous food could not be detected, although sulfide-like odor was sensible in the advanced stage of spoilage. However, dimethyl sulfide was detected in beef. Since the proportion of this component to head space volatiles was relatively large even at 0 day, it is supposed that dimethyl sulfide has a relationship to beefy flavor. In this study, ethyl acetate was the component which was detected from the initial stage of bacterial spoilage and increased dominantly according to the growth of bacteria in all samples. In our experiment of six cases, a fruity odor was sensed in the initial stage of spoilage. In the review of DAINTY (1982), the development of fruity odors in proteinaceous foods has been traced to the presence of short chain fatty acid esters. Minced beef generated the fruity odor and contained methyl acetate and ethyl acetate. Ethanol and acetate are likely to be formed in carbohydrate and/or amino acid metabolism, and these components were catalyzed by bacteria and produced ethyl acetate. Also, in the study of FREEMAN et al. (1976), methyl acetate and ethyl acetate were important components of the odor of spoiled chicken. Therefore, it was supposed that ethyl acetate contributed to the fruity odor which was sensed at the initial stage of spoilage. In spite of the increase of ethyl acetate content, fruity odor was not sensible at the advance of spoilage in all samples. This phenomenon was thought to be the masking of fruity odor by other intense odor components such as acidic and sulfide-like components. It should be clarified that bacteria produces ethyl acetate. According to FREEMAN et al. (1976), ethyl acetate was produced by P. fragi, P. fluorescens and other Pseudomonas species. Consequently, in the aerobic package, the formation of ethyl acetate can be mainly attributed to these microorganismus in this study. GILL (1983) reported that during storage of vacuum-packaged meat, volatile fatty acids were produced, and he mentioned that development of a simplified technique is necessary for estimation of total volatile acids. According to GRAD and MACFARLANE (1980), lactic acid bacteria was the predominant flora in 690

The Headspace Volatiles of Meat vacuum packaging, but also other organisms such as Brocothrix thermosphacta and Enterobacteriaceae were contained under some conditions. In our experiment, the storage temperature supposed to be a little higher to allow the selective growth of lactic acid bacteria. Thus five of six samples showed deterioration odors at the end of storage whether they were aerobically packed or vacuum packed. Consequently, a study at chill temperature is needed to confirm the usefulness of ethyl acetate as an index of bacterial spoilage. In the development of a gas sensor, a fish freshness sensor was developed in which the semiconductive metal oxides were used as a sensing device (EGASHIRA et al., 1990). In this sensor, trimethylamine was detected with high selectivity. Trimethylamine is one of the main components in volatile nitrogen component in fish meat. In the meat of domestic animals, ammonia is the main volatile basic nitrogen component. Thus, an ammonia-selective gas sensor will be applicable as a freshness sensor in meat. However, in this study, ammonia was not measured to exclude the effects of water in ion chromatography. In our previous research (YANG et al, 1990), ammonia was about 100mg/l even in fresh meat and started to increase after the bacterial counts reached 108 cells/g. Therefore, it was supposed that ammonia sensor was suitable for detecting the advanced stage of spoilage but not suitable for the detection of the initial stage of spoilage in meat. As an indicator of the initial meat spoilage, a compound which does not exist in the fresh stage is preferable. From these considerations, ethyl acetate seems to be a useful indicator for meat freshness and an ethyl acetate-selective gas sensor can be used to detect meat freshness. However, further investigation is needed to evaluate the reproducibility of the results and the identification of ethyl acetate producing bacteria, especially in vacuum packaged meat at chill temperatures. References 1) DAINTY, R. H., Biochemistry of undesirable effects attributed to microbial growth on proteinaceous foods stored at chill temperatures. Food Chemistry, 9:103-113. 1982. 2) EGASHIRA, M., Y. SHIMIZU and Y. TAKAO, Trimethylamine sensor based on semiconductive metal oxides for detection of fish freshness. Sensors and Actuators, 1:108-112. 1990. 3) FREEMAN, L. R., G. J. SILVERMAN, P. ANGELINI, C. MERRITT Jr, and W. B. ESSELEN, Volatiles produced by microorganisms isolated from refrigerated chicken at spoilage. Applied and Environmental Microbiology, 32:222-231. 1976. 4) GILL, C. O., Meat spoilage and evaluation of the potential storage life of fresh meat. J. Food Protection, 46:444-452. 1983. 5) GRAU, F. H. and J. J. MACFARLANE, The end of the storage life of refrigerated meat: why does it happen and what can be done about it? CSIRO Food Res. Q., 40:60-65. 1980. 6) MCMEEKIN, T. A., Spoilage association of chicken breast muscle. Applied Microbiology, 29:44-47. 1975. 7) SUTHERLAND, J. P., J. T. PATTERSON and J. G MURRAY, Changes in the microbiology of vacuum-packaged beef. J. Appl. Bact., 39:227-237. 1975. 8) SUTHERLAND, J. P., P. A. GIBBS, J. T. PATTERSON and J. G. MURRAY, Biochemical changes in vacuum packaged beef occuring during stor- 9) YANO, Y., Y. SHIBUTANI, F. SUZUKI, T. HADA and T. NAKAMURA, Diamines as potential indices for freshness of both aerobically packed and vacuum packed meat. Jpn. J. Zootech. Sci., 61: 16-21. 1990. 691

YANO, MAEDA and HIRATA