Effects of Bioactive Packaging Films Incorporated with Bifidocin A onMicrobial Reduction and Quality Parameters of Chill-Stored Spanish Mackerel (Scomberomorus niphonius) Fillets

To evaluate the potential of bidocin A for preservation of fresh Spanish mackerel llets, the bioactive packaging lms incorporated with bidocin A (1×minimal inhibition concentration (MIC), 2×MIC and 4×MIC) were developed, and their eects on the microbiological and physicochemical properties and sensory prole of mackerel llets at refrigerated storage were investigated. Results showed that the incorporation of bidocin A in cellulosic matrix lms did not aect the thickness and elongation of the lms, but reduced slightly the tension strength. e lms incorporated with 2×MIC and 4×MIC bidocin A presented a broad spectrum of activity against most tested bacteria, including some sh-borne specic spoilage bacteria such as Pseudomonas uorescens, Shewanella putrefaciens, Brochothrix thermosphacta, andMicrococcus luteus, andmaintained their 100% activity for 28 days during storage at 4°C.e bioactive packaging lms incorporated with bidocin A could generally suppress the growth of microora, especially Pseudomonas and Enterobacteriaceae, as well as substantially inhibit the accumulation of total volatile basic nitrogen (TVB-N), thiobarbituric acid reactive substances (TBARS) and hypoxanthine, during chilled mackerel llets storage. Overall, from a microbiological and physicochemical point of view, a much more eective treatment was achieved with 4×MIC bidocin A, extending the shelf life to 12 days and maintaining the relatively low TVB-N value (≤13.2mg/100 g), TBARS value (≤0.45mg MDA/kg), and K-value (≤32.8%), as well as a relatively high sensory score (≥7.1) during the whole storage. Hence, the bioactive packaging lms incorporated with bidocin A could be a promising hurdle technology and alternative to conventional processes used for improving the safety and quality of chilled mackerel llets.


Introduction
Spanish mackerel (Scomberomorus niphonius) is a major marine sh and is widely distributed in the Bohai Sea and East China Sea. e sh constitutes desirable components of a healthy diet, including polyunsaturated fatty acids, essential amino acids, vitamins, and minerals and is deeply loved by consumers and producers in China. However, because of its high nutritive content, fresh mackerel llets are highly perishable even under refrigerated storage [1]. e various techniques, including addition of chemical preservatives [2,3], modi ed atmosphere packaging [4], high hydrostatic pressure [5], vacuum packing [6], and dielectric barrier discharge [7], have been introduced to preserve sh llets. With regard to the microbial contamination of sh llets occurring primarily at the surface, due to postprocessing handling, the use of bioactive packaging lms containing antimicrobial agents could be more ecient, by slow migration to the food surface, thus helping to maintain high concentrations where they are needed [8].
Bacteriocins are safe, natural antimicrobial peptides for food preservation, and some of them present high potential to inhibit the growth of sh-borne spoilage or pathogenic bacteria [9][10][11][12][13][14]. During the last decades, several innovative bioactive packaging lms enriched with bacteriocins have been developed and have received a considerable attention for improving the safety of food products, such as goat milk, fresh beef, raw sliced pork, white semifirm cheese, frozen shrimp, sliced ham, and sausage [15][16][17][18][19][20][21][22]. However, at present there are no studies available that evaluated the antimicrobial packaging film containing bacteriocin on microbial reduction and quality parameters during the storage of a marine fish.
Bifidocin A, a novel 1198.68 kDa bacteriocin, is produced by Bifidobacterium animalis BB04, isolated from a centenarian's faeces. e potential of bifidocin A to control food-borne spoilage and pathogens has been reviewed and underlined in our recent work [9]. It exhibits inhibitory activity against a broad range of Gram-positive and Gramnegative bacteria in the genera Escherichia, Listeria, Salmonella, Pseudomonas, Staphylococcus, Bacillus, Clostridium, Brochothrix, Shewanella, and Micrococcus, especially Pseudomonas fluorescens, Shewanella putrefaciens, Brochothrix thermosphacta, and Micrococcus luteus, which were fish-borne specific spoilage bacteria. e current study was designed to develop a bioactive packaging film incorporated with bifidocin A and evaluate its application feasibility in chilled Spanish mackerel (Scomberomorus niphonius) fillets. e physical properties and antimicrobial activity of the films incorporated with bifidocin A was evaluated, and their effects on the microbiological and physicochemical properties and sensory profile of mackerel fillets at refrigerated storage for up to 15 days were investigated.

Preparation of Partially Purified Bacteriocin and
Quantification of Bacteriocin Activity

Bacteriocin Production in 5 L Fermentor. A 5-liter
Biostat B fermentor (B. Braun Biotech International, Melsungen, Germany) containing 3.5 liters of MRS-C medium with the constant temperature (37°C), and pH 6.5 was used to study the kinetics of B. animalis BB04 growth and bacteriocin production. Aerating nitrogen and slow agitation (100 rpm) was maintained to keep the anaerobic environment and the fermentation broth homogenous, respectively. At appropriate time intervals, samples were taken for determining cell growth by the viable cell count and bacteriocin production by the antimicrobial activity. e antimicrobial activity was defined as the reciprocal of the highest serial two-fold dilution producing a distinct inhibition of P. fluorescens CGMCC 1.55 and expressed in terms of arbitrary units per milliliter (AU/mL).

Partial
Purification of Bacteriocin. Partially purified bifidocin A was obtained from the above 24 h old culture of B. animalis BB04 by pH-mediated adsorption and desorption onto/from producer cells, as described previously [9]. Firstly, the cultures were adjusted to pH 8.0 and stirred overnight at 4°C. After being harvested (10,000 × g, 10 min) and washed twice with sterile 20 mM phosphate buffer (pH 8.0), the cells were resuspended in 200 mL of 100 mM·NaCl (pH 3.0) and agitated for 24 h at 4°C. en, the supernatant including bacteriocin was obtained by centrifugation (10,000 × g, 15 min). Finally, the bacteriocin samples were determined for antimicrobial activity as described above, and then they were lyophilized and stored at −20°C before further experiment.

Minimal Inhibition Concentration (MIC) Determination.
e MIC of partially purified bifidocin A against P. fluorescens CGMCC 1.55 was determined as described by Liu et al. [9]. e MIC was defined as the lowest concentration of partially purified bifidocin A at which the growth of the indicator strain was inhibited completely.

Preparation of Bioactive Films Incorporated with Bifidocin
A. Cellulose acetate, acetone, and partially purified bifidocin A were employed on film production. e films were prepared by the casting process, according to Santiago-Silva et al. [21], with some modifications. Briefly, a cellulose acetate solution was firstly prepared by adding the flakes of cellulose acetate to acetone in a proportion of 1 : 10 (10% w/ v) and left to stand for about 12-18 h. And then, the bifidocin A was incorporated in the cellulosic emulsion with the final concentrations ranging from 1 × MIC to 4 × MIC. After that, the mixture was homogenized and spread on glass plates, previously cleaned with acetone. Finally, the films were removed from the plates, after evaporation of the solvent under ambient conditions, and were stored in packages of polyethylene (PE) at 4°C for further analysis. A film without bifidocin A incorporation was used as the negative control.

Physical Properties of Bioactive Films Incorporated with
Bifidocin A. Films were characterized in terms of their thickness and mechanical properties. e thickness of the film samples was determined using a thickness gauge (CH-1-ST, China) with a precision of 0.01 mm. Five thicknessmeasured values were randomly taken from each film, and the average of all measurements was used in the calculations of mechanical properties. e tensile strength and elongation at break were determined with a slight modification using a Universal Testing Machine (UTM2502, China) as described by Woraprayote et al. [17]. e initial grip length and cross-head speed of ten film samples (10 × 150 mm 2 ) were set at 100 mm and 5 mm/min, respectively. Tensile strength value was calculated by dividing the maximum stress by cross-sectional area of the specimen, and elongation values were expressed as percent units, with the ratio of extended length at break point of initial length. e spectrum of activity of the antimicrobial films incorporated with/without bifidocin A (1 × MIC, 2 × MIC and 4 × MIC) was determined, respectively, against the selected food-borne spoilage and pathogenic strains (Table 1) by using an adaptation of the agar diffusion method [23]. Fragments of 1 cm 2 of each film were cut, sterilized under UV light, and placed over solidified LB/TSYEB/TSB agar. e agar was previously inoculated with each bacteria (Table 1) in order to obtain a lawn (10 7 -10 8 cfu/mL) of the strain. After 24 h incubation at 30°C, the presence of clear zone of inhibition was evaluated.

Stability of Antimicrobial Activity.
To evaluate the stability of antimicrobial activity, the films, incorporated with bifidocin A (2 × MIC and 4 × MIC), were sterilized under UV light for 15 min each side and incubated at 4°C and 25°C for up to 28 days. e residual activity was then assayed against the indicator strain P. fluorescens CGMCC 1.55 at two days interval and reported as a percentage of antimicrobial activity of the samples after storage compared with that of time zero.

Chilled Mackerel Fillets Preparation and Sampling.
Fresh mackerel was purchased from an aquatic product wholesale market in Beijing, China, in June 2017, and was immediately transported to the laboratory alive. Fish were killed by a blow to the head, scaled, gutted, filleted, and washed within 2 h. After washing, fish fillets were left to drain on sterile stainless steel wire mesh for 10 min. e average weight for each fillet was 100 g. And then, these fish fillets were divided into three groups. e treatment groups were packed by the bioactive films incorporated with 2 × MIC and 4 × MIC bifidocin A, respectively. e fish fillets samples packed using the films without bifidocin A incorporation were used as the negative control. Packed fish fillets were stored at 4°C in refrigerators for up to 15 days. ree fillets were selected randomly from each group for analysis every three days.

Microbiological Analyses.
At each selected time, 10 g of each sample was taken aseptically and homogenized in 90 mL 0.1% (w/v) sterile peptone water containing 1% (v/v) Tween 80 and 0.85% (w/v) NaCl. After disintegration for 2 min in a Stomacher (LAB-BLENDER, 400, London, UK), the resulting suspension was serially diluted in 10-fold and then plated in triplicate on total count and selective agar plates. Total viable counts (TVCs) were determined on plate count agar (PCA, Difco), incubated at 30°C for 48 h. Enterobacteriaceae were enumerated on violet red bile glucose agar (VRBGA, Difco), incubated at 37°C for 48 h. Pseudomonads were enumerated on cephaloridine-fucidincetrimide agar (CFC, Difco), incubated at 30°C for 48 h. H 2 Sproducing bacteria survival was monitored by enumeration on iron agar (IA, Difco), incubated at 25°C for 3 days. Photobacteria were enumerated on conductance medium (BCM, Difco), incubated at 15°C for 5 days.

Physicochemical
Analyses. At each selected time, 10 g of each sample was blended with 20 mL of distilled water in a blender for 30 s, and the pH value of homogenate filtrate was determined using a HI 8420 Microcomputer pH meter (Hanna instruments, Germany). Total volatile basic nitrogen (TVB-N) was measured using a microtitration methodology including distillation in a Foss 2300 Kjeldahl Analyzer Unit (Foss Tecator AB, Sweden) and titration with hydrochloric acid. TVB-N value was determined according to the consumption of hydrochloric acid and was expressed as mg/ 100 g flesh [24].
iobarbituric acid reactive substances (TBARS) value was determined according to the method of Shi et al. [25] and was expressed as mg malondialdehyde (MDA)/kg flesh. K-values were determined according to the method of Huang et al. [26].

Sensory Evaluation.
Sensory analysis was carried out by nine trained panelists from the laboratory staff according to the method of Duman andÖzpolat [27] with minor modification. Appearance, odor, texture, and overall quality were scored on a 9-point descriptive scale as follows: 8.0-9.0 � good, 6.0-7.9 � acceptable, 4.0-5.9 � unacceptable, and 1.0-3.9 � completely spoiled.

Statistical Analysis.
All experiments were conducted at least in triplicate (except microbiological analyses, which were performed in duplicate). Data were expressed as means ± standard deviation, analysis of variance was used to test for differences on SPSS 20.0, least significant difference (LSD) test for multiple comparison at the 0.05 level of significance.

Partially Purified Bacteriocin Preparation and Bacteriocin
Activity Assay. Cell growth and bacteriocin production properties of B. animalis BB04 were measured at 37°C in 5 L fermentor with controlled pH. As shown in Figure 1, strain BB04 started to produce bacteriocin (70 AU/mL) at 12 h during the early exponential growth phase. e maximal bacteriocin production (640 AU/mL) was observed after  [28], which indicated this bacteriocin has great potential as e ective and natural preservative in food industry.

Physical Properties of Bioactive Films Incorporated with
Bi docin A. Bioactive lms must possess adequate properties to maintain durability and withstand the external stresses that prevail during food processing, handling, and storage. As shown in Table 1, the incorporation of bidocin A in cellulosic matrix lms did not a ect the thickness and elongation values of the lms (P ≥ 0.05). However, there was a slight di erence (P < 0.05) in tensile resistance between the lms with bi docin A and the lm control. e lms incorporated with 1 × MIC, 2 × MIC, and 4 × MIC bi docin A showed a reduction in the tension strength value in relation to the control lm. is may be ascribed to low interaction between bi docin A and the polymer, allowing the formation of aggregates. Similar results were obtained by Gouvêa et al. [29] on basis of acetate cellulose lm with added bacteriophages, which showed reductions in the tensile resistance as compared to the control.

e Antimicrobial Activity of the Bioactive Films Incorporated with Bi docin A.
e lms incorporated without bi docin A did not exhibit the inhibitory activity against tested bacteria (data not shown). e antimicrobial spectrum of the lms incorporated with bi docin A is presented in Table 2. e lms incorporated with 2 × MIC and 4 × MIC bi docin A showed a broad spectrum of activity against most tested bacterial strains, which is in consistence with the antimicrobial activity of bacteriocin only as described previously [9]. However, for the treatments at 1 × MIC, the lms were not e ective against  [30]. is probably related to the relatively low migration level of bacteriocin transferred from the packaging material to the solidi ed medium. e lms incorporated with 2 × MIC and 4 × MIC bidocin A maintained their 100% activity against P. uorescens CGMCC 1.55 for 28 days during storage at 4°C. However, the bacteriocin activity could not be detected after incubation for 18 days at 25°C. e residual activity of the treatments with 2 × MIC and 4 × MIC bi docin A was 36% and 41%, respectively, after 15 days of incubation at 25°C. ese results indicated the bioactive lms incorporated with bi docin A are more suitable for preservation of chilled food.
Based on these result, additional studies assessing the e ect of bioactive lms incorporated with bi docin A on chilled mackerel llets were performed at 2 × MIC and 4 × MIC concentration. Figure 2(a), the initial TVC in control and treatment groups were 3.36, 2.54, and 2.43 log CFU/g, respectively. e TVC increased with storage time and reached up to 7.93 log CFU/g on the 9th day for the control samples, which exceeded the maximum acceptable level of 7.0 log CFU/g for freshwater and marine sh [6,26,27]. Compared with the control, all the treatments with bi docin A signi cantly reduced the initial bacteria counts (P < 0.05) and caused inhibition of the growth of bacteria in mackerel llets and extension of the shelf life under storage at 4°C. e treatment with 4 × MIC bi docin A was much more e ective and prolonged the chilled shelf life of mackerel llets to 12 days. Lv et al. [14] reported that the addition of bacteriocin DY4-2 produced by Lb. plantarum DY4-2 in turbot (Scophthalmus maximus) llets could achieve a microbiological shelf life extension of 4 days. And Gui et al. [11] reported that the addition of paraplantaricin L-ZB1 produced by Lb. paraplantarum L-ZB1 could extend the shelf life of chilled rainbow trout llets to 10 days. ese results demonstrated that bacteriocin from lactic acid bacteria may be considered as a suitable candidate for its application as fresh sh biopreservative.

Microbiological Analyses. As shown in
Compared with the control, all treatments with bi docin A signi cantly inhibited the growth of Pseudomonas bacteria during the whole storage (P < 0.05) (Figure 2(b)). Moreover, Pseudomonas bacteria were rst detected on the 3rd day in bi docin A treatment groups. ese results suggested that the lms incorporated with bi docin A had a better inhibitory e ect on Pseudomonas bacteria than others, which correlated well with the high inhibition activity of bi docin A for P. uorescens as described previously. However, there were no signi cant di erences in Pseudomonas bacteria  counts between the treatments with 2 × MIC and 4 × MIC bifidocin A during the first 6 days (P ≥ 0.05).
Another more evident effect of bifidocin A treatment group was shown on Enterobacteriaceae bacteria (Figure 2(c)). Compared to the control, a significant decline in Enterobacteriaceae counts was observed in bifidocin A-treated samples (P < 0.05). is may be attributed to the high inhibition activity of bifidocin A for Salmonella and Escherichia. Samples treated with 4 × MIC bifidocin A had a lower Enterobacteriaceae counts of 3.02 log CFU/g on the 15th day compared to 3.58 (2 × MIC bifidocin A), which indicated that the treatment with 4 × MIC bifidocin A more efficiently inhibited the growth of Enterobacteriaceae bacteria than did the 2 × MIC bifidocin A treatment (P < 0.05).
In addition, the growth rate of other tested bacteria, including Photobacterium and H 2 S-producing bacteria, was slightly lower in bifidocin A-treated groups when compared with the control group (Figures 2(d) and 2(e)). is may be related to the wide range of antimicrobial activity of bifidocin A used in this study. e potential of the addition of bacteriocin alone to control undesirable microbes in fish product has been evaluated by a number of research groups in recent years [10][11][12][13][14]. However, in order to comply with actual tendency of consumers of searching for healthier foods, minimally processed and free of additives, the addition of preservatives directly to foods should be avoided as far as possible [8,21]. In this study, we developed the antimicrobial potential of packaging films incorporated with bifidocin A to improve preservation of chilled mackerel fillets.
is bioactive packaging could be a more promising preservation technology than bacteriocin alone, due to its action not only as a barrier against external agents but also by maintaining high concentrations bacteriocin on the food surface with a low migration [21]. Figure 3(a), the pH value showed a rapid decrease in control and reached a minimum value of 6.04 in the first 6 days, due to the generation of an acidic substance by glycogen degradation. However, the pH increased in the further storage, which should result from alkaline substances such as ammonia and trimethylamine generated from protein degradation [10]. e change rate of pH value in the treatment groups was significantly lower than that in the control group (P < 0.05), which indicated the treatments with bifidocin A can effectively inhibit the degradation of glycogen and protein in chilled mackerel fillets.

Physicochemical Analyses. As shown in
TVB-N is the degradation product produced by proteins under the action of microorganisms and enzymes [24,27]. As shown in Figure 3(b), the initial TVB-N level of all the fish fillets samples was 9.4-9.7 mg/100 g and no significant difference was observed after 3 days (P ≥ 0.05). e TVB-N values increased gradually with the storage time, and the value of control had increased from 9.5 to 33.2 mg/100 g of fish fillets sample by day 15, had exceeded the maximum acceptable limit of 30 mg/100 g [6,26,27]. Compared with control, TVB-N values in all the treatments with bifidocin A could be controlled in a relatively low level (≤25.7 mg/100 g) during the whole storage. Besides, a higher decrease was observed in the treatment with 4 × MIC bifidocin A. is finding was similar to that of Ananou et al. [13] who observed sardine fillets treated with enterocin AS-48 produced by Enterococcus faecalis A-48-32 reduced the TVB-N content under refrigerated storage.  TBARS is used widely to assess the degree of secondary lipid oxidation by measuring the content of MDA, which is the product formed by the reaction of polyunsaturated fatty acids with oxygen [25]. TBARS values of 1 mg MDA/kg are usually regarded as the limit for normal odor [10,26]. e e ect of the lms incorporated with bi docin A on TBARS values of chilled mackerel llets is shown in Figure 3(c). Lower TBARS values were obtained for all the samples (<1 mg MDA/kg), which indicated low oxidative rancidity.
e treatments with bi docin A did not produce a significant e ect on TBARS during the rst 6 days. Compared to the control, the treatments with bi docin A delayed the increase of TBARS (P < 0.05) after the llets were stored for 9 days. However, no signi cant di erences in TBARS values were observed between the treatments with 2 × MIC and 4 × MIC bi docin A (P ≥ 0.05).
e K-values increased among all samples over time in this experiment. During almost the entire storage, the K-values of the treatments with bi docin A continued below 70%, although control samples exceeded this limit on the 12th day (70.6%).
ese results indicated that the treatments with bi docin A can e ectively delay the degradation of inosine monophosphate (IMP) and the production of hypoxanthine (Hx) in chilled mackerel llets.
From a technological point of view, it is important to determine the in uence of bioactive packing lms incorporated with bacteriocin on physicochemical parameters in the actual food system. Until now, nothing has been reported on the effect of the films incorporated with bacteriocin on quality parameters of fish product. In this study, we described the influence of packaging films incorporated with bifidocin A on the pH, TVB-N, TBARS, and K-value of chilled mackerel fillets in detail. e results indicated that the bioactive packaging with bifidocin A effectively inhibits the accumulation of TVB-N, TBARS, and Hx during chilled mackerel fillets storage, which is basically consistent with the microbiological change, and achieve the similar effect as the addition of bacteriocin alone.
3.6. Sensory Evaluation. As shown in Table 3, sensory parameters (appearance, odor, texture, and overall quality) decreased during storage. e control group was able to maintain the acceptable quality at 4°C for approximately 6 days and spoiled completely in 12 days. However, the samples in treatments with 2 × MIC bifidocin A denoted as unacceptable quality by the sensory panel on the 12th day.
e best organoleptic quality of mackerel fillets was observed in the treatment group with 4 × MIC bifidocin A, which could maintain a better sensory profile throughout the whole storage. Overall, the sensory score showed that the films incorporated with bifidocin A significantly improved the preservation of chilled mackerel fillets and prolonged their shelf life. A similar observation has been made by Fu et al. [10], who reported the large yellow croaker (Pseudosciaena crocea) fillets treated with coagulin L1208 could achieve a high sensory score and prolong the period of grade-A freshness up to 8 days at 4°C.

Conclusion
In this study, a novel antimicrobial packaging film by the incorporation of bifidocin A in the cellulosic matrix was successfully developed. e films incorporated with 2 × MIC and 4 × MIC bifidocin A presented a broad spectrum of activity in vitro against most of tested food-borne spoilage and pathogenic bacteria, and maintained their 100% activity for 28 days during storage at 4°C. e bioactive packaging films incorporated with bifidocin A could generally suppress the growth of microflora, especially Pseudomonas and Enterobacteriaceae, as well as substantially inhibit the accumulation of TVB-N, TBARS, and Hx, during chilled mackerel fillets storage. Overall, from a microbiological and physicochemical point of view, a much more effective treatment was achieved with 4 × MIC bifidocin A, extending the shelf life to 12 days and maintaining a better sensory profile during the whole storage. In summary, the present work demonstrates for the first time the potential of bifidocin A as biopreservatives for fresh marine fish during refrigerated storage. e bioactive packaging films incorporated with bifidocin A could be a promising hurdle technology and alternative to conventional processes used for extending the safety and quality of refrigerated mackerel fillets. In our laboratory, a detailed study on the mechanism of bifidocin A on the inactivation of fish-borne specific spoilage bacteria is currently underway.

Conflicts of Interest
e authors declare that they have no conflicts of interest.

Authors' Contributions
Mrs. Xue Li carried out preparation of partially purified bacteriocin and bacteriocin activity quantification guided by Dr. Guorong Liu. Ms. Xu Hao carried out the preparation of bioactive films incorporated with bifidocin A and evaluation of the antimicrobial activity of the films guided by Dr. Chengtao Wang. Mrs. Xue Li carried out the experimental design and the investigation of the effects of bioactive films on the microbiological and physicochemical properties and sensory profile of mackerel fillets at refrigerated storage guided by Dr. Baoguo Sun. e manuscript was written by Dr. Guorong Liu.