Structural Changes and Evolution of Peptides During Chill Storage of Pork

In this work, we investigated changes in protein structures in vacuum-packed pork during chill storage and its impact on the in vitro protein digestion. Longissimus dorsi muscles were vacuum packed and stored at 4°C for 3 days. Samples were subjected to Raman spectroscopy, in vitro digestion and nano LC-MS/MS. The 3 d samples had lower α-helix content, but higher β-sheet, β-turn, and random coil contents than the 0 d samples (P < 0.05). SDS-PAGE revealed significant protein degradation in the 3 d samples and the differences in digested products across the storage time. Proteome analysis indicated that the 3 d samples had the higher susceptibility to digestion. Increasing protein digestibility was mainly attributed to the degradation of myofibrillar proteins. Thus, exposure of more enzymatic sites in loose protein structure during chill storage could increase protein degradation in meat.


INTRODUCTION
Postmortem aging and its impact on eating quality of fresh meat has been widely concerned (1)(2)(3). Physicochemical and structural changes in meat proteins occurring in postmortem aging and processing may affect the protein digestion and nutritional value of meat (4)(5)(6). Postmortem aging for relatively long time may produce bioactive peptides in beef that exhibit DPPH radical scavenging, ACE-and renin-inhibitory activities (7). However, little is known about whether bioactive peptides are produced in a short-term aging of pork.
At early postmortem time, the dephosphorylation of several metabolic enzymes has been shown to affect the rate of glycolysis and pH decline, and subsequently the activation of µ-calpain, the release of lysosomal enzymes and meat tenderization (8). In addition, protein oxidation may occur in fresh meat during chill storage (9), which may cause the formation of disulfide bonds, and a decrease in protein hydrophilicity and water holding capacity. To avoid such changes, vacuum packaging and chill storage have been widely applied (10). However, protein oxidation and structural changes may still occur in vacuum-packed and chilled meat because radical oxygen species widely existing in living muscle tissues could be retained in postmortem muscles (11). Furthermore, protein oxidation also alters the protein structures and digestibility of meat (5). An increase in digestion can be attributed to protein unfolding and increased susceptibility to digestion (4). Previous studies have focused how processing and physicochemical changes affect protein structure and digestion (12,13). However, few data are available on how structural changes of meat proteins enhanced digestibility of fresh meat from a proteomic perspective, in terms of the evolution of peptides from the stomach to the small intestine. In this study, we investigated the structural changes of meat proteins in pork during chill storage and their impacts on protein digestibility and the release of bioactive peptides.

Sample Preparation
Longissimus dorsi muscles (size: 10 × 10 × 5 cm) were obtained from 8 native Suhuai pig carcasses at the same line in a commercial slaughterhouse. Each muscle was cut into four 2.5 × 10 × 5 cm pieces (weight: 50 to 65 g each), vacuum-packed and stored at 4 • C for 3 days. Samples were taken on 0, 1, 2, and 3 d for further analyses.

Raman Spectroscopy
Raman spectroscopy was performed to evaluate changes in protein secondary structure (LabRAM HR Evolution, Horiba/Jobin, Yvon, Longjumeau, France) as previously described (5). Laser (excitation wavelength: 785 nm, power: 100 mW) was applied and the backscattering Raman signals ranging from 400 to 3,200 cm −1 were collected. Raman spectra were normalized against the band at 1,003 cm −1 . The 1,685-1,645 and 1,309-1,229 cm −1 bands correspond to amide I and III vibrational modes of α-helix, random coil, and β-sheet. The 1,341 and 940 cm −1 bands reflect CH-bending and C-C stretching respectively. The range of 1,658-1,650 cm −1 in the amide I vibrational mode mainly reflects C=O stretching vibrations, C-N stretching, and N-H in-plane bending of peptide groups. The 1,003, 830-850, and 760 cm −1 bands correspond to phenylalanine, tyronsine, and tryptophan, respectively. The α-helix, β-sheet, β-turn, and random coil were quantified (14).

In vitro Digestion
The protein digestibility of cooked meat from 0, 1, 2 to 3 d samples was assessed as described by Zou et al. (15). Briefly, meat pieces were packed in plastic bags and cooked in a 72 • C water bath until the core temperature reached 70 • C. The cooking time was around 20 min. The core temperature was tracked by a portable thermal probe (Pt 100, Testo AG, Mönchaltorf, Schweiz, Germany). After cooking, meat samples were cooled for 2 h to room temperature (22 • C). Then, samples (1.00 g) were homogenized (9,600 rpm, 30 s, twice; 13,400 rpm, 30 s, twice) in PBS (10 mmol/L, pH 7.0). The homogenate was digested by gastric pepsin (32 mg/mL in 0.1 mol/L HCl, pH 1.0) at 37 • C for 2 h and the digestion was stopped by adjusting the pH to 7.5 with 1 mol/L NaOH. Then the resulting mixture was digested by trypsin (24 mg/mL in 0.1 mol/L PBS, pH 7.0) at 37 • C for 2 h and stopped by heating the system at 95 • C for 5 min.
The undigested proteins were precipitated by adding three volumes of ethanol at 4 • C for 12 h. Then the samples were centrifuged at 10,000 × g at 4 • C for 20 min. Protein contents in the whole meat samples and the precipitates were quantified by a BCA protein assay kit (Thermo Scientific, Rockford, IL) according to the manufacturer's instructions. Protein digestibility was calculated as follows: Where W 1 is the content (g) of undigested proteins precipitated by ethanol. W 0 is the total protein content (g) in the whole meat before digestion. The supernatant containing digested products that are ethanol soluble was subjected to nano LC-MS/MS analysis and the precipitate including undigested proteins was separated on the SDS-PAGE gels.

SDS-PAGE
The whole proteins were extracted according to the method of Zarkadas and Maloney (16) with some modifications. Briefly, cooked meat samples (0.3 g) were homogenized in 4.5 mL 2% SDS with 3 × 30 s at 10,000 rpm and centrifuged at 4,000 × g for 20 min at 4 • C. Two milliliters of the supernatant were dialyzed for 48 h in 1 L distilled water to remove SDS. The dialyzed samples were transferred to new tubes and stored at −80 • C for further analyses.
SDS-PAGE was performed to separate proteins or their fragments under reducing conditions (5). Appropriate volumes of samples were mixed with 12.5 µL sample buffer (4×) and 5 µL reducing sample agent (10×) and made up to a total volume of 50 µL with ultrapure water. The final protein concentrations were 0.50 µg/µL for all samples. The samples were heated at 70 • C for 10 min. Twelve microliters of samples were loaded in triplicate into the wells of 4-12% precast gels (GenScript, Piscataway, NJ, USA). The gels were run in a total 800 mL of SDS running buffer at 150 V till the blue dye front disappeared. The gels were stained with colloidal Coomassie brilliant blue R250 (CBB) for 30 min and then destained for 20 min three times. CBB staining solution contained 0.1% CBB/45% acetic acid/10% ethanol/45% ultrapure water. Destaining buffer contained 10% acetic acid/10% ethanol/80% ultrapure water. The gel images were acquired by an image scanner (GE Healthcare, Little Chalfont, Uppsala, Sweden) and the band intensities were quantified with the Quantity One image analysis software (Bio-Rad, Hercules, CA). The relative intensities of bands were calculated by the actual band intensity divided by that of the 150 kDa band in the calibration marker lane. FIGURE 1 | Postmortem aging caused a reduction of the α-helix content but an increase in the β-sheet, β-turn, and random coil contents, accompanying with the exposure of tryptophan. All data were presented as means and standard deviations. Asterisks denote that the contents of α-helix, β-sheet, β-turn, random coil, and the ratios of I −1 760cm /I −1 1003cm and I −1 850cm /I −1 830cm differ significantly between the two time points. *P < 0.05; **P < 0.05.

Protein Identification of Digested Products by Nano LC-MS/MS
The ethanol-soluble fractions of the digested products were identified by Nano LC-MS/MS system as previously described (17) with minor modifications. Briefly, ethanol in the supernatant was removed by a vacuum concentrator (ZXJY, Beijing, China), and dried in a freeze dryer (Christ, Osterode, Germany). The dried samples were dissolved in 0.2% formic acid in ultrapure water and then centrifuged (15,000 × g, 15 min, 4 • C) in ultra-0.5 mL filter tubes and desalted in ZipTip C18 tips (Millipore, Billerica, MA). Peptides (1.5 µg) were separated in a C18 column (2 cm × 200 µm, 5 µm, Thermo Fisher Scientific, Palo Alto, CA) and then sequentially a C18 chromatographic column (75 µm × 100 mm, 3 µm, Thermo Fisher Scientific, Palo Alto, CA). Elution was applied by running a mixture of buffer A (0.2% formic acid in 60% acetonitrile) and buffer B (0.2% formic acid in ultrapure water) at 300 nL/min. The elution buffers were changed by 97%A from 0 to 10 min, 92%A from 10 to 70 min, 62%A from 70 to 72 min, 2%A from 72 to 82 min, and 97%A from 82 to 90 min. The eluted peptides were identified by a hybrid quadrupole orbitrap mass spectrometer equipped with a nanoelectrospray ionization source (Thermo Fisher Scientific, Palo Alto, CA). A full-scan mode was selected from 300 to 1,800 amu.

Statistical Analysis
The effects of storage time on measured variables (αhelix, β-sheet, β-turn, random coil, I −1 760cm /I −1 1003cm and I −1 850cm /I −1 830cm ) were evaluated by Student's t-test using GraphPad Prism 7.0 (GraphPad Software, Inc., La Jolla). The effects of storage time on other measured variables were evaluated by analysis of variance (ANOVA) and the least significant means were compared by Tukey's posthoc t-test. Data were presented as means and standard deviations. The means were considered significantly different if the P-value was smaller than 0.05. Venn diagrams (http:// bioinformatics.psb.ugent.be/webtools/Venn/) were applied to analyze the differences of matched proteins among the four time points.

Changes in Raw Meat Protein Structures Detected by Raman Spectroscopy
Raman spectroscopy provides powerful information on secondary structural modifications in proteins (18). The α-helix content decreased from 77.41% on 0 d to 67.89% on 3 d (P < 0.05, Figure 1), while β-sheet, β-turn, and random coil increased from 3.39, 11.72, and 8.92% on 0 d to 9.75, 13.21, and 9.49% on 3 d, respectively (P < 0.05, Figure 1). This indicates that denaturation and unfolding occurred in meat proteins from 0 to 3 d. During aging, meat proteins were degraded by endogenous enzymes into fragments. Degradation of myofibrillar proteins is modulated by protein oxidation and nitrosylation (19).
The side chains of tryptophan and tyrosine residues may change under a polar microenvironment, which results in modifications in the tertiary structure of proteins. The ratios of I −1 760cm /I −1 1003cm and I −1 850cm /I −1 830cm were related to the exposed or buried status of tryptophan and tyrosine residues, respectively (20). A substantial increase in the ratio of I −1 representing changes of tryptophan residues was observed during chill storage (P < 0.05, Figure 1), indicating transformation of the tryptophan residues from a buried, hydrophobic state to a polar aqueous state (20). The ratio of I −1 850cm /I −1 830cm significantly increased from 0 to 3 d (P < 0.05, Figure 1), indicating an increasing number of exposed tyrosine residues and tryptophan residues. In addition, the ratio of I −1 850cm /I −1 830cm is also a good indicator for the microenvironment and hydrogen bonds of the ionization of the phenolic hydroxyl group (20). It could be the  exposure of polar amino acid residues at the surface of the protein molecules. The ordered and steady structure of the protein was broken into disordered and loose fragments during storage, from a buried status of tryptophan and tyrosine residues to an exposed status, which exposed cleavage sites of digestive enzymes. Thus, we postulate that this disordered and hydrophobic status would increase the degree of protein degradation by exposing more cleavage sites to digestive enzymes.

The Degree of Protein Degradation in Cooked Meat
No significant difference was observed in the degree of protein degradation after pepsin digestion among the four time points (P > 0.05, Figure 2A). However, the degree of protein degradation increased after two-step digestion, and the 3 d samples had the highest values (P < 0.05, Figure 2B). In a previous study, postmortem storage did not improve the in vitro digestion parameters of pork (4), in which the structural organization of the muscle cell and the extracellular matrix were not taken into account. To a certain extent, the storage-induced differences in the degree of protein degradation could be attributed to changes in protein secondary or tertiary structures. Such a structural change may further affect protein digestion after meat cooking. Different cooking conditions may affect the structure of meat protein, and the digestibility of meat (17). To avoid it, we cooked meat samples with a similar size for the same time in a water bath. The proteolysis during chill storage may increase protein unfolding and expose more sites to bind to pepsin and trypsin. This is in agreement with the Raman data that reflected structural changes in side chains of proteins, including hydrophobic and electrostatic interactions.

SDS-PAGE of Proteins in Cooked Meat and Their Digested Products
SDS-PAGE was applied to separate soluble meat proteins or their digested products before or after pepsin and trypsin treatments (Figure 3). Before digestion (untreated samples), a reduction in band intensity of several protein bands was observed in cooked pork as the storage time of raw meat increased, indicating fragmentation and breakdown of meat proteins. This result was in agreement with our previous study (15). The appearance of the 30 kDa component (Troponin-T fragment) has been considered In pepsin treated samples, a substantial decrease in band intensity was observed for some large-molecular-weight protein bands (Figure 3), which could be due to degradation of proteins into smaller peptides or free amino acids. In addition, some insoluble proteins could become soluble after enzymatic digestion (23). Intensity loss or even disappearance of bands was observed for pepsin and trypsin treated samples (Figure 3). Most of large molecular weight bands disappeared because proteins or peptides were further degraded into smaller ones after twostep digestion (Figure 3). Similar phenomena have been observed in different pork cuts and pork products (14,24). However, SDS-PAGE could not separate smaller peptides well that were characterized by nano LC-MS/MS (15).

Unique Peptides Could Be a Good Indicator for Chill Storage Time
Pepsin or two-step digestion products were identified by nano LC-MS/MS and matched to proteins. All matched proteins are listed in Supplementary Files 1, 2. Venn diagrams revealed that 97 peptides were matched in all pepsin digested samples and pepsin and trypsin digested samples, respectively (Figures 4A,B). Most of these peptides came from myofibrillar proteins (myosin-1, myosin-7, actin) and sarcoplasmic proteins (phosphoglycerate kinase, creatine kinase M-type). In general, peptide abundances decreased as the storage time of raw meat increased, indicating that these peptides were degraded into smaller ones during digestion. Furthermore, 9 peptides matched to myofibrillar proteins, and 52 peptides matched to sarcoplasmic proteins in pepsin-treated samples (Supplementary File 1). More peptides matched to myofibrillar proteins (48/97) than to sarcoplasmic proteins (22/97) (Supplementary File 2). Taken together, the increased protein degradation could be mainly attributed to the degradation of myofibrillar proteins.
Several common peptides appeared in 1, 2, and/or 3 days samples, which were derived from large molecular weight peptides or proteins. Specifically, 22, 5, 8, and 19 peptides were uniquely matched in pepsin-treated samples on 0, 1, 2, and 3 days respectively, and 18, 3, 33, and 6 peptides were unique for pepsin/trypsin-treated samples at the four-time points (Figures 4A,B). Unique peptides could be a good indicator for chill storage time.

Digested Products From Myofibrillar Proteins in Pork Evolved With Storage Time
All peptides matched to myofibrillar proteins were listed in Table 1. Previous studies have shown that actin, myosin-1, myosin-4, myosin-7, and many glycolytic enzymes are highly abundant proteins in pork muscle (15,25). These proteins are involved in muscle contraction and energy production. In the present study, most of differently abundant peptides are derived from these proteins (Tables 1, 2).
From the digested products, 117 peptides were selected to match the BIOPEP database showing ACE-inhibitory (angiotensin-converting-enzyme inhibitory) and dipeptidyl peptidase IV (DPPIV) inhibitory activities ( Table 2). Eighteen peptides were estimated to have ACE-inhibitory and DPPIV inhibitory activities, which have been identified in the database (26). ACE inhibitors are known to beneficial to treat arterial hypertension (27), diabetes mellitus, ischemic heart disease, chronic heart failure, angioedema, chronic kidney disease, and Parkinson's disease (28,29). DPPIV inhibitor may be beneficial for NASH subjects (30). Four peptides (TKLEQQVDDLEGSL EQEKK, LEQQVDDLEGSLEQEKK, VKLEQHVDDLEGSLEQ EKK, LEQHVDDLEGSLEQEKK) were identified, which contain sequences VDDLEGSLEQEKK and DDLEGSLEQEKK that were previously reported to have antioxidant peptides (31). It is of note that some identified peptides maintain a long sequence after the two-step digestion. However, almost all the peptides would be hydrolyzed by peptidases into smaller fragments (dipeptides and tripeptides) locating the outer layer of intestinal epithelium before reaching the blood stream (32). Thus, moderate chill storage may enhance the production of antioxidant, ACE inhibitor and DPP-IV inhibitory peptides. Table 2 illustrates dynamic evolution of 117 peptides derived from cooked meat samples. The potential bioactive peptides were predominantly originated from myosin, actin, tropomyosin and troponin. Peptides EDQVFPMNPPK (Oxidation) and IEDE QALALQLQK appeared in 0 day samples, whereas peptides EFEMSNLQSKIEDEQALAMQLQK, IEDEQALAMQLQK, and SKQLEDELVSLQK appeared only in 2 day samples. Abundances of LAQESIMDIENEK, LAQESTMDIENDKQQLDEK, and their   oxidative peptides and KPAAAAAPAPAPAPAPAPAPAPPKEEK were the lowest in 2 day samples. Abundances of SVMLQIAA TELEK, DIDDLELTLAKVEK, and TKLEQQVDDLEGSLEQEK were the highest in 2 day samples. This could be mainly attributed to the degradation of proteins and accumulation of peptides in the first stage, which would be further degraded into smaller peptides that have ACE and DPPIV inhibitory bioactivities. Thus, 2 days could be a key time point for chill storage of pork in terms of meat nutrition. Two peptides were observed to have the same sequence but their molecular weights were different, which is attributed to protein oxidation based on the database. Eighteen peptides from two-step digestion were observed to have oxidation modification.
It is known that protein oxidation occurs during storage and the extent of protein oxidation may be enhanced as meat storage time increases. In the present study, pork samples were vacuum packed and stored at low temperature (4 • C) for 3 days. In such a condition, protein oxidation could be alleviated. And thus more cleavage sites could be exposed in loose protein structure and protein degradation increased. The number of bioactive peptides may increase after the two-step digestion.
The above results could be associated with rigor mortis and subsequent aging of muscle (33). It is well-known that glycolysis occurs in skeletal muscles after slaughter (34,35). This process is accompanied by the dephosphorylation of energy metabolic enzymes, the formation of actomyosin, and the shortening of sarcomeres (6,36). Such activities may make some energy metabolic enzymes and myofibrillar proteins less susceptible to pepsin and trypsin digestion under in vitro condition on day 1. However, prolonged storage can tenderize meat, and make the myofibrillar proteins and bound energy metabolic enzymes more digestible. This is because the endogenous enzymes, e.g., µ-calpain and cathepsins, catalyze the degradation of myofibrillar proteins into smaller fragments (33), which will expose the cleavage sites for digestive enzymes. It is notable that, too long-time storage of fresh meat may have the problems of food spoilage, discoloration, and the decreased digestibility of meat proteins due to oxidation (37).

CONCLUSIONS
In this study, postmortem aging was shown to have significant impacts on the structural characteristics and in vitro digestion of pork proteins. It is observed that the 3 d samples had lower α-helix content and the peak intensity at 760 cm −1 (tryptophan residues), but higher β-sheet, β-turn, and random coil contents than the 0 d samples. This is accompanied by significant protein degradation and increased susceptibility to digestion. The structural changes of proteins altered the accessibility of proteolytic enzymes to the cleavage sites, and consequently the protein digestion. Further work is necessary to evaluate the effect of postmortem aging on in vivo digestion of pork protein.

DATA AVAILABILITY STATEMENT
The datasets generated for this study can be found in Proteome Xchange, Accession No. PXD020595.

ETHICS STATEMENT
The animal study was reviewed and approved by the Ethical Committee of Experimental Animal Center of Nanjing Agricultural University.