2018 Volume 24 Issue 4 Pages 635-651
Metabolic changes of Tan sheep meat maintained at 0 °C for up to 8 d were identified based on gas chromatography time-of-flight mass spectrometry. Total viable counts (TVC), total volatile basic nitrogen (TVB-N) and pH were utilized as freshness indicators during meat storage. A relationship between meat freshness, metabolite accumulation and key metabolic pathways was postulated. Twenty-seven statistically significant metabolites were characterized. D-glyceric acid, phenylalanine, methionine, glucose-1-phosphate, D-(glycerol-phosphate), lysine, ribitol, asparagine and isomaltose were significantly (p < 0.01) correlated to the freshness indicators. A correlation coefficient with gluconic acid, citric acid and trans-4-hydroxy-L-proline and the freshness indicators ranged from 0.539 to 0.911. 1, 5-anhydroglucitol was significantly correlated with pH (p < 0.05) and TVC (p < 0.01). These metabolites were identified as potential spoilage biomarkers. Three major metabolic processes occurred during storage: metabolism of alanine, aspartate, and glutamate; an operational TCA cycle and metabolism of amino sugar and nucleotide sugar.
Tan sheep are becoming a staple source of animal protein in Ningxia province, China. Over the last ten years the consumption of mutton has increased due to a surge in consumer purchasing power. Preference, however, is for fresh chilled meat rather than previously frozen meat. In this respect, chilled meat has one major drawback. Even when maintained at temperatures as low as 0 °C, it will have a limited self-life. The shelf-life of raw meat is normally evaluated by visual and sensory features as well as by microbiological and biochemical characteristics and proteomic analysis (Aru et al., 2016). Raw meat spoilage during storage is mainly due to the activity of various microorganisms, primarily bacteria (Doulgeraki et al., 2012). Although freshness and or deterioration of chilled meat can be judged on the basis of sensory characteristics as a means of establishing shelf life, it cannot be done with certainty. Biochemical changes in meat due to endogenous and microbial enzymes including proteases and lipases contribute to the production of various metabolic products including biogenic amines, volatile nitrogen compounds and free fatty acids (Dave and Ghaly, 2011).
Metabolomics, which involves the study and classification of metabolites produced as a result of biochemical reactions in living cells, is a relatively new food science research tool (Dixon et al., 2006). Metabolomic techniques have been used in foodomics with respect to food quality and safety, and in monitoring genetically modified foods. Using this technique researchers have reported on the activities of bacterial populations and the volatiles or fingerprints associated with their growth contributing to meat spoilage (Castejon et al., 2015). For example, several studies regarding the application of metabolomics in identifying the spoilage status of minced beef stored at various temperatures and packaging systems have been reported (Argyi et al., 2015). In these studies the relationship between microbial growth and chemical changes that occur during meat storage have been recognized as potential indicators useful in assessing beef quality or freshness (Argyri et al., 2015) Metabolomics has also been used to evaluate the preservation and aging of beef (Castejon et al., 2015) and for the detection of mechanically recovered meat in food products (Surowiec et al., 2011).
Present analytical techniques used in metabolomics analyses include nuclear magnetic resonance (NMR), gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS) and capillary electrophoresis - mass spectrometry technologies (CE-MS). GC-MS and LC-MS are well established analytical methods. These are important techniques in food analysis due to their excellent ability in the separation and identification of metabolites in complex samples. Generally, low-molecular-weight metabolites and volatile compounds have been profiled by GC-MS, and a relatively small number of nonvolatiles showing similar chemical features have been profiled by LC-MS (Sugimoto et al., 2017). GC-MS can simultaneously measure hundreds of compounds, their derivatives and secondary metabolites including: amino acids, organic acids, sugars and sugar alcohols (Uri et al., 2014). Although various profiling techniques are used in metabolomics, GC combined with time of flight-mass spectrometry (TOF-MS) appears to be used most often due to its higher resolution and sensitivity. This method also utilizes mass spectral databases such as NISTi in the identification and quantification of numerous metabolites (Lisec et al., 2006).
In this paper, GC-TOF/MS-based metabolomics was employed to assess spoilage biomarkers in chilled Tan sheep meat. Freshness or spoilage indicators in the stored meat consisted of pH, total viable counts (TVC) and total volatile basic nitrogen (TVB-N). Overall, the relationship between freshness and specific biomarkers was investigated as a means of further revealing the metabolic pathways involved in the spoilage of chilled meat.
Materials and samples collection Freshly butchered and dressed Tan sheep hind legs (c. 2.5 kg each) were obtained from a local abattoir (Yinchuan, China), placed in portable coolers at 4 °C and transported to the laboratory where they were cooled at −20 °C for 1 h. The hinds were then placed into sterile food grade bags and maintained at 0 °C and sampled at 0, 4 and 8 d. Sampling consisted of scraping the surfaces of 5 whole hind legs over an area of c. 20 cm × 20 cm using an alcohol sterilized knife at each time period. The scraped contents from the hind legs which included surface microbial growth, blood and exudate were commingled. Five-100 mg pooled samples were then taken and placed into individual sterile capped tubes and refrigerated until analyzed. In addition, 2 cm × 3 cm × 4 cm sections from each of the previously scraped hind legs were aseptically hand cut and placed into sterile sample bags.
Total viable microbial counts Meat sample sections (20 g) from each hind leg at each sampling period were stomached with sterile saline (0.85 %; 80 mL) and serially diluted using sterile saline. TVC was determined using a spread plate method with Luria-Bertani medium (Hengbai Biotech Co., Ltd, Shanghai, China). Plates were enumerated following incubation at 35 °C for 48 h (He et al., 2016).
pH and total volatile basic nitrogen tests Meat samples (10 g) from each hind leg at each sampling period were homogenized in distilled water (90 mL, pH 7.0) for 2 min and paper filtered. The pH was determined using a digital pH meter (FE20, Mettler Toledo, Shanghai, China). TVB-N was determined using a Kjeltec TM 8100 analyzer (Foss, Sweden). Values are expressed in mg nitrogen per 100 g sample (Cai et al., 2014).
GC-MS sample preparation Pooled samples (50 mg) were placed into 2-mL EP tubes, treated with methanol-chloroform (Meryer Technologies Co., Ltd., Shanghai, China; 0.4 mL, 3:1, v/v) and L-2-chlorophenylalanine (20 µL, Hengbai Biotech Co., Ltd., Shanghai, China; 1 mg/mL stock in d H2O) was added as an internal standard. The suspension was homogenized using a ball mill (3 min, 65 Hz) and centrifuged 14,000 ×g for 15 min at 4 °C. The supernatant (c. 350 µL) was transferred to GC-MS glass vials (2 mL), vacuum dried at 37 °C for about 3.5 h and methoxylamine hydrochloride (80 µL) added. Methoxylamine hydrochloride (Meryer Technologies Co., Ltd., Shanghai, China) was initially dissolved in pyridine for a final concentration of 20 mg/mL. Samples were incubated at 80 °C for 20 min following mixing and sealing. After incubation the vials were opened and BSTFA (100 µL; Regis Technologies, Inc., Morton Grove, USA) was added into each sample. Vials were sealed again, incubated at 70 °C for 1 h and FAME (Hengbai Biotech Co., Ltd.; 5µL, standard mixture of fatty acid methyl esters, C8–C16:1 mg/mL; C18–C24:0.5 mg/mL in chloroform) was added to the mixed sample and cooled to room temperature for subsequent GC-MS analysis.
GC-MS metabolites analysis GC-TOF-MS analysis was performed using an Agilent 7890 gas chromatography system coupled with a Pegasus HT time-of-flight mass spectrometer (LECO, St. Joseph, MI, USA). The system utilized an Rxi-5Sil MS column (30 m × 250 µm inner diameter, 0.25 µm film thickness; Restek, USA). A1-µL aliquot of the analyte was injected in splitless mode. Helium was used as the carrier gas and the front inlet purge flow was 3 mL/min. The initial temperature was kept at 50 °C for 1 min, then raised to 330 °C at a rate of 10 °C/min, then kept for 5 min at 330 °C. The injection, transfer line, and ion source temperatures were 280 °C, 280 °C, and 220 °C, respectively. The energy was 70 eV in electron impact mode. The mass spectrometry data were acquired in full-scan mode with the m/z range of 30–600 at a rate of 20 spectra per second after a solvent delay of 366 s.
GC-MS data analysis The overload peaks of metabolites were removed and the GC-TOF-MS data were imported into Microsoft Excel. An area normalization method was used to assess the data. Principal component analysis (PCA) and supervised orthogonal partial least squares discriminate analysis (PLS-DA) was performed using SIMCA-P version 13.0 (Umetrics, Sweden; Zhang et al., 2015). Variables with importance in projection (VIP) values greater than 1 were further subjected to t tests in order to determine significance. Only variables with “VIP > 1.00”, “p < 0.05”, and “Similarity > 700” were selected as potential biomarkers (Lv et al., 2015).
Relative concentrations of twenty-seven differential metabolites in the chilled meat obtained from GC-MS analysis were inputted into an online analysis platform (Metaboanalysti 3.0). Metabolites were first converted by logarithm as an input to a hierarchical clustering algorithm, where the distance style is Euclidean and linkage is by average (Sun et al., 2016).
In order to further identify and visualize potential metabolic pathways resulting from the synthesis of metabolites during meat storage, the identified biomarkers were input into a KEGG database (Kono et al., 2005) and metabolic pathways were constructed and analyzed (Kastenmüller et al., 2011).
Statistical analysis Analysis of variance and the correlation between metabolites and freshness indicators were analyzed by SPSS 19.0 statistical software (SPSS Inc., Chicago, IL, U.S.A.).
TVC of chilled meat Changes in TVC during chilled storage of meat are shown in Figure 1. The TVC of the chilled meat increased c. 3.5 log during storage reaching c. 6.8 log10 CFU/g by the eighth day. According to the General Administration of Quality Supervision Inspection and Quarantine and The National Standardization Management Committee Standards (2008), the microbial level in the raw mutton exceeded a limit of 5.0 × 105 CFU/g and is therefore not considered to be fresh.
Changes in pH, TVB-N, and TVCs of meat during storage . Error bars indicated the standard deviations (n=5). Columns with different letters for the same parameter are different (p < 0.05).
pH and TVB-N of chilled meat During storage the pH of the meat gradually increased from 5.8 to 6.7 (Fig. 1). Ostensibly the increase in pH was due to the accumulation of basic compounds including ammonia as a result of enzymatic reactions. According to the General Administration of Quality Supervision Inspection and Quarantine, and The National Standardization Management Committee Standards (2008), a pH in the range of 5.7–6.2 is considered ideal with respect to freshness in meat. A pH of 6.3–6.6 is still compatible for fresh meat, however, when it exceeds pH 6.6 the meat it is regarded as being no longer fresh. This occurred on the eighth day.
A TVB-N value less than 15 mg/100 g indicates freshness in meat (General Administration of Quality Supervision Inspection and Quarantine and The National Standardization Management Committee, 2008). As shown in Fig. 1, the TVB-N of the chilled meat c. doubled during storage. The greatest increase in TVB-N occurred after day 4 of storage which also coincided with the greatest increase in microbial numbers. Overall, all the freshness indicators increased significantly (p < 0.05) over the storage period.
Metabolites profiling of chilled meat Following pretreatment procedures, the accumulated data were applied to PCA and PLS-DA in order to visualize group trends. As shown in Fig. 2, the PCA score plot (Fig. 2A) and PLS-DA (Fig. 2C and 2D) showed a clear separation of all sampling periods. It indicated that the storage time had a significant impact on the changes of metabolites. There were some differences between days 4 samples and days 8 samples that compared with days 0 samples; the spots of days 8 samples were farther than days 4 samples. It showed that metabolism changes of days 8 samples was more serious than days 4 samples. The corresponding PCA loading plot (Fig. 2B), wherein each spot represents a metabolite, reveals variables that contributed to sample separation as illustrated on the score charts. The spots of variables were father away from the origin, the more contributions to classification. As shown in Fig. 2B, these metabolites that the searching ID number was 460, 469, 408, 254, 783, 796, 145, 65, 346, 135, etc. had a great contribution to the classification.These metabolites maybe potential biomarkers during meat storage. The data set was then applied to PLS-DA module to select possible biomarkers responsible for these separations. After screening with “VIP > 1.00”, “p < 0.05”, and “Similarity > 700”, 27 metabolites were selected. Detailed information is summarized in Table 1.
(A) PCA score plot of meat during storage. The contribution ratios were 39.2 and 23.4 % for t[1] and t[2], respectively; (B) PCA loading plot of meat during cold storage;(C) PLS-DA score plot of meat at 0 and 4 d group. Contribution ratios were 53.5 and 13.4 % for t[1] and t[2], respectively; and (D) PLS-DA score plot of meat at 4 and 8 d group. The contribution ratios were 27.4 and 20.5 % for t[1] and t[2],respectively. Day 0: triangles; day 4: circles; day 8: squares.
| ID | RT (min) | Metabolite compound | VIP | p value |
|---|---|---|---|---|
| 165 | 10.6590 | D-Glyceric acid | 1.6502 | 0.001070 |
| 175 | 10.9790 | Fumaric acid | 1.1983 | 0.014203 |
| 346 | 14.3908 | Phenylalanine | 1.9107 | 0.000840 |
| 276 | 13.1135 | Methionine | 1.9091 | 0.000988 |
| 145 | 10.0434 | Acetanilide | 1.6818 | 0.011973 |
| 275 | 13.1076 | Aspartic acid | 1.8780 | 0.043087 |
| 408 | 15.9058 | Glucose-1-phosphate | 1.6560 | 0.029773 |
| 406 | 15.8602 | D-(Glycerol-phosphate) | 1.5897 | 0.007541 |
| 783 | 26.0927 | Inosine-5′-monophosphate | 1.6586 | 0.001638 |
| 268 | 12.9775 | Asparagine | 1.7594 | 0.007804 |
| 356 | 14.5668 | 1,3-Diaminopropane | 1.6575 | 0.009422 |
| 483 | 17.4977 | Lysine | 1.9906 | 0.000610 |
| 96 | 9.02759 | 2-Hydroxyvaleric acid | 1.7542 | 0.042966 |
| 429 | 16.2463 | Terephthalic acid | 1.7650 | 0.019103 |
| 750 | 24.7788 | 5′-Methylthioadenosine | 1.8403 | 0.001655 |
| 763 | 25.2606 | Isomaltose | 1.1372 | 0.000039 |
| 254 | 12.7022 | L-malic acid | 1.8123 | 0.000375 |
| 551 | 19.1173 | Myo-inositol | 1.4204 | 0.049925 |
| 387 | 15.4255 | Ribitol | 1.8677 | 0.002223 |
| 278 | 13.1917 | Oxoproline | 1.6727 | 0.025586 |
| 516 | 18.2383 | Gluconic acid | 1.9460 | 0.046719 |
| 463 | 17.1267 | Mannose | 1.1104 | 0.016090 |
| 620 | 20.9957 | Fructose-6-phosphate | 1.1060 | 0.018655 |
| 437 | 16.4430 | Citric acid | 1.4379 | 0.049834 |
| 277 | 13.1753 | Trans-4-hydroxy-L-proline | 1.6073 | 0.034945 |
| 364 | 14.9056 | Ribose | 1.6854 | 0.017564 |
| 451 | 16.7946 | 1,5-Anhydroglucitol | 1.8421 | 0.009993 |
PCA (Fig.2A and 2B) and PLS-DA (Fig. 2C and 2D) analyses of the GC-TOF/MS metabolic profiles clearly illustrated separate clusters at days 0, 4, and 8 suggesting that the GC-TOF/MS-based metabolomics model could be used to monitor freshness progress during meat storage.
Hierarchical clustering of chilled meat The concentration of 27 metabolites identified by GC-MS at 0, 4, and 8 d are illustrated in the heat map (Fig. 3). The color in each box represents the concentration of the metabolite; boxes exhibiting an increase in red or green intensity indicate a higher or lower concentration, respectively. Samples at 0 d contained much higher levels of fumaric acid, 2-hydroxyvalericacid, glucose-1-phosphate, 5′-methylthioadenosine, acetanilide, inosine-5′-monophosphate and 1, 3-diaminopropane. With storage at 4 d, the concentrations of aspartic acid, terephthalic acid, isomaltose, D-(glycerol-phosphate), lysine, phenylalanine and methionine increased in most samples while at 8 d the concentrations of ribitol, citric acid and 1,5-anhydroglucitol increased in the majority of samples. In contrast, levels of L-malic acid, myo-inositol, oxoproline, gluconic acid, mannose, glucose-1-phosphate, inosine 5′-monophosphate, fructose-6-phosphate, ribose and trans-4-hydroxy-L-proline decreased during storage. Additionally, levels of asparagine and D-glyceric acid increased in days 4 samples then decreased in days 8 samples.
Average linkage hierarchical clustering of significantly altered metabolites of meat during storage.The light blue boxes indicate an expression ratio less than the mean; the dark red boxes denote an expression ratio greater than the mean. Tree clusters and their shorter Euclidean distances indicate higher similarities. Similarity between two metabolites is represented by branch height; thus, when a node is lower vertically, the subtree is more similar.
The relative concentrations of the 27 metabolites are shown in the heat map (Fig. 3). Clusters of varying size from different storage periods appear close. The heat map illustrates the changes in the concentration of the metabolites among the groups. The metabolic activity of the individual metabolites within a group appeared dissimilar. This would suggest that the metabolite concentrations are also different. Aggregation of colored boxes is evident; this would indicate that the biomarkers and metabolic pathways based on the 27 metabolites are closely associated.
Following animal slaughter, blood circulation stops and glycogen is no longer respired. Under ensuing anaerobic conditions, residual glycogen is converted to lactic acid (glycolysis) and results in a decrease in muscle pH (Yano et al., 1995). Although the decrease in pH is beneficial with respect to preservation, activation of endogenous protease enzymes with age releases metabolites suitable for rapid microbial growth and subsequent spoilage (Pablo et al., 1989). In this study, the use of GC-MS metabolomics also confirmed that during the course of meat storage there was a gradual reduction of glycogen as evidenced by a decrease in the levels of glucose-1-phosphate and ribose. In addition, at day 8 the level of 1,5-anhydroglucitol was observed to increase ostensibly due to microbial metabolism of carbohydrates resulting in the production of organic acids and alcohols. It is recognized that upon glucose exhaustion microbial metabolism of nitrogenous based compounds such as proteins, peptides, amino acids and amines occurs very quickly resulting in the production of various metabolites including glyceric acid, asparagine, aspartic acid, acetanilide and ammonia (Ercolini et al., 2006). These compounds were also identified in the present study.
Glucose, lactic acid, and certain amino acids, followed by water-soluble proteins, are the major metabolites precursors responsible for meat spoilage (Zhou et al., 2010). The concentration and nature of these precursors can affect both the rate and degree of spoilage. Endogenous (autolytic) enzyme activity in meat muscle is also important and will contribute to spoilage; however, the endogenous contribution to spoilage has been reported to be small compared to the effects of microbial growth. Dave et al. (2011) reported that the accumulation of microbial metabolites, such as aldehydes, ketones, esters, alcohols, organic acids, amines, and sulfur (mercaptans) compounds, largely contributed to the spoilage of meat. Research regarding the qualitative and quantitative presence of metabolites as they pertain to meat spoilage is therefore valuable and highly warranted.
Correlative analysis of freshness and characteristic metabolites Results of the correlation analyses between 27 metabolites and freshness indicators (Table 2) suggested that glyceric acid, phenylalanine, methionine, glucose-1-phosphate, glycerol-phosphate, lysine, ribitol, asparagine, isomaltose were significantly (p < 0.01) correlated with the freshness indicators .The correlation coefficient between gluconic acid, citric acid, trans-4-hydroxy-L-proline and the freshness indicators ranged from 0.539 to 0.911(p < 0.01 or p < 0.05). In addition, the production of 1, 5-anhydroglucitol correlated with pH (p < 0.05) and TVC (p < 0.01). It is suggested that glyceric acid, phenylalanine, methionine, asparagine, isomaltose, ribitol, lysine, citric acid glucose-1-phosphate, g lycerol-phosphate, gluconic acid, 1, 5-anhydroglucitol, trans-4-hydroxy-L-proline may function as spoilage markers. In contrast, researchers investigating spoilage in chicken breast muscle identified several spoilage markers which included: aspartic acid, histidine, glycine, threonine, proline, and leucine (Alexandrakis et al., 2012). Aru et al. (2016) also used metabolomics to study the freshness of mussels, and explored potential metabolites and microbiological correlations. The potential spoilage biomarkers identified by these researchers included: acetate, lactate, succinate, alanine, branched chain amino acids, and trimethylamine all of which exhibited significant increased levels. It is interesting to note that the spoilage markers identified in the above studies including the present contained some common markers but also some markers that were particular to the food type. As expected, this would indicate that the composition of the substrate is important with regards to the types of markers identified and that the results obtained in this study are not unusual.
| Metabolite | pH | TVB-N | TVC |
|---|---|---|---|
| D-Glyceric acid | 0.642** | 0.662** | 0.664** |
| Phenylalanine | 0.927** | 0.887** | 0.896** |
| Methionine | 0.972** | 0.925** | 0.938** |
| Glucose-1-phosphate | 0.940** | 0.851** | 0.931** |
| D-(Glycerol-phosphate) | −0.764** | −0.796** | −0.763** |
| Asparagine | 0.910** | 0.834** | 0.897** |
| Lysine | 0.823** | 0.702** | 0.813** |
| Isomaltose | 0.823** | 0.887** | 0.762** |
| Ribitol | 0.955** | 0.907** | 0.943** |
| Gluconic acid | 0.811** | 0.639* | 0.853** |
| Citric acid | 0.911** | 0.952** | 0.872* |
| Trans-4-hydroxy-L-proline | 0.614* | 0.702** | 0.539* |
| 1,5-Anhydroglucitol | 0.794* | 0.646 | 0.843** |
* and ** indicate significant at p < 0.05 and 0.01.
Analysis of key metabolic pathways In this study 29 KEGG pathways were identified when metabolites were compared at days 0 and 4 of storage; at days 4 and 8 of storage, 21 pathways were identified. After enrichment and pathway topology analysis of the identified pathways (Table 3; Fig. 4A), only 10 pathways (comparison of days 0 and 4 of storage) exhibited an impact value at the comprehensive level. These included: alanine, aspartate and glutamate metabolism; cysteine and methionine metabolism; the TCA cycle; glycine, serine, and threonine metabolism; streptomycin biosynthesis; glyoxylate and dicarboxylate metabolism; starch and sucrose metabolism; galactose metabolism; amino sugar and nucleotide sugar metabolism; and purine metabolism. Of these 10 pathways, only alanine, aspartate and glutamate metabolism, streptomycin biosynthesis, starch, sucrose and purine metabolism exhibited the highest impact value. Adjusted p values via the Holm-Bonferroni method for multiple testing and comprehensive analyses, indicated that the impact value for pathways that had the greatest difference included metabolism of: alanine, aspartate, and glutamate (p = 0.000423, Holm adjusted = 0.0368, Impact value = 0.1383).
| Pathway | Hits | p value | Holm p | Impact value |
|---|---|---|---|---|
| Alanine, aspartate, and glutamate metabolism | 4 | 0.000423 | 0.0368 | 0.1383 |
| Cysteine and methionine metabolism | 3 | 0.034566 | 1.0000 | 0.0620 |
| TCA cycle | 2 | 0.067797 | 1.0000 | 0.0729 |
| Glycine, serine, and threonine metabolism | 2 | 0.151470 | 1.0000 | 0.0224 |
| Streptomycin biosynthesis | 1 | 0.180290 | 1.0000 | 0.2286 |
| Glyoxylate and dicarboxylate metabolism | 1 | 0.476450 | 1.0000 | 0.0339 |
| Starch and sucrose metabolism | 1 | 0.499660 | 1.0000 | 0.2172 |
| Galactose metabolism | 1 | 0.563540 | 1.0000 | 0.0464 |
| Amino sugar and nucleotide sugar metabolism | 1 | 0.610750 | 1.0000 | 0.0956 |
| Purine metabolism | 1 | 0.811160 | 1.0000 | 0.1025 |
The metabolome view map of significant metabolic enrichment pathways of meat during storage between (A) 0 and 4 d, and (B) 4 and 8 d. The x-axis represents pathway enrichment, and the y-axis represents pathway impact. Larger sizes and darker colors represent greater pathway enrichment and higher pathway impact values, respectively.
Eleven pathways were calculated to have an impact value at the comprehensive level by days 4 and 8 of storage (Table 4; Fig. 4B). These included: the TCA cycle; amino sugar and nucleotide sugar metabolism; streptomycin biosynthesis; glyoxylate and dicarboxylate metabolism; galactose metabolism; pentose phosphate pathway; pyruvate metabolism; starch and sucrose metabolism; glycine, serine, threonine and purine metabolism. A comprehensive analysis of the p value and impact value indicated that pathways that had the greatest difference included: the TCA cycle (p = 0.006097, Holm adjusted = 0.5304, impact value = 0.1968) and amino sugar and nucleotide sugar metabolism (p = 0.00769, Holm adjusted = 0.66129, impact value = 0.11111). The key pathways and metabolites are shown in Fig. 5. Accordingly, three metabolites were characterized, including asparagine (VIP = 1.7594, p = 0.007804; FC = 1.5773), citrate (VIP = 1.4379, p = 0.049834; FC = 4.9357), and D-glucosamine-1-phosphate (VIP = 1.3284, p = 0.000634; FC = 0.1563).
| Pathway | Hits | p value | Holm p | Impact value |
|---|---|---|---|---|
| TCA cycle | 3 | 0.006097 | 0.5304 | 0.1968 |
| Amino sugar and nucleotide sugar metabolism | 4 | 0.007689 | 0.6613 | 0.1111 |
| Streptomycin biosynthesis | 2 | 0.012323 | 1.0000 | 0.2286 |
| Glyoxylate and dicarboxylate metabolism | 3 | 0.017430 | 1.0000 | 0.1804 |
| Galactose metabolism | 3 | 0.033489 | 1.0000 | 0.0464 |
| Pentose phosphate pathway | 2 | 0.091007 | 1.0000 | 0.0392 |
| Cyanoamino acid metabolism | 1 | 0.148200 | 1.0000 | 1.0000 |
| Pyruvate metabolism | 1 | 0.409090 | 1.0000 | 0.0023 |
| Starch and sucrose metabolism | 1 | 0.466810 | 1.0000 | 0.2172 |
| Glycine, serine and threonine metabolism | 1 | 0.477690 | 1.0000 | 0.0224 |
| Purine metabolism | 1 | 0.779910 | 1.0000 | 0.1025 |
Key metabolic pathway integration map of meat during storage. The figure was generated using the reference map from KEGG and consisted of entry number of metabolites and pathways.
At day 4 of storage, biosynthesis of alanine, aspartate, and glutamate increased. Based on the KEGG pathway, it would appear that glycogen breakdown and amino acid metabolism increased due to enzymatic activity. For example, asparagine catabolism occurs via the action of asparaginase and results in the release of aspartic acid and ammonia. Aspartic acid is a precursor for the biosynthesis of additional amino acids, such as lysine, threonine, isoleucine, and methionine and is utilized in the biosynthesis of purine and pyrimidine bases. It also contributes to microbial growth (Gram et al., 2002). At day 8 of storage, the up-regulated pathways appeared to include the TCA cycle, biosynthesis of amino sugars and nucleotide glucose metabolism. In this respect malate is converted into oxaloacetic acid via malate dehydrogenase and citric acid levels increase by condensation of oxaloacetic acid and acetyl-CoA. Both mannose and ribose sugars are converted into amino and nucleotide sugars. These reactions are consistent with results reported by previous researchers that employed metabolic analysis to assess potential markers of spoilage food (Cheng et al., 2015).
The metabolic pathways utilized by microorganisms and the metabolites produced during their growth are dependent on a multiplicity of extrinsic and intrinsic factors including the nature of the substrate. Analysis of these pathways and metabolites is not only interesting from an academic point of view but also can provide vital information with respect to assessing the degree and or nature of spoilage and the mechanisms involved. This information can be useful especially in evaluating the quality and safety of foods.
In conclusion, GC-TOF/MS profiling and multivariate analysis were used to illustrate the significant changes in metabolites and metabolic pathways during cold storage of mutton. D-glyceric acid, phenylalanine, methionine, glucose-1-phosphate, D-(glycerol-phosphate), asparagine, lysine, isomaltose, ribitol, gluconic acid, citric acid, trans-4-hydroxy-L-proline and 1, 5-anhydroglucitol can be used as biomarkers for the identification of mutton spoilage. As the storage time increases, alanine, asparate, and glutamate metabolism, TCA cycle, and amino sugar and nucleotide sugar metabolism are the key metabolic pathways. This method could be used to further develop new food safety techniques and be helpful in gaining a further understanding of Tan sheep meat spoilage.
Acknowledgment This study was supported by grants from the National Natural Science Foundation of China (No. 31460431). and the Innovation Project of Ningxia University in China (No. ZKZD2017007).
