Gelatin‐starch composite coating containing cucumber peel extract and cumin essential oil: Shelf life improvement of a cheese model

Abstract In this study, the effects of gelatin‐starch (GS) composite coating containing cucumber peel extract (CPE) and cumin essential oil (CEO) were evaluated on the shelf life enhancement of ultrafiltered (UF) cheese during 56 days of storage under refrigerated conditions. The obtained hydroethanolic CPE by the microwave method showed the best results in terms of the total phenolic content, reducing power, 2,2′‐diphenyl‐1‐picrylhydrazyl (DPPH) activity, and 2,2‐azino‐bis‐3‐ethylbenzothiazoline‐6‐sulfonic acid (ABTS) radical scavenging activity compared to the immersion and ultrasound methods. The studied treatments were as follows: Control (C), GS, CPE, CEO, GS‐CPE, GS‐CEO, and GS‐CPE‐CEO. Scanning electron microscopic surface morphology of treated cheese samples showed the formation of a firm, integrated, flawless, and homogenous layer on the cheese slices of the GS‐CPE‐CEO treatment. All treatments significantly (p ≤ .05) decreased the total viable count, psychotropic bacteria, and yeast–mold population compared to the control group. Adding CEO and/or CPE to GS significantly (p ≤ .05) controlled undesirable changes in physical characteristics, such as weight, color, and hardness of the cheese slices. Throughout storage time, the coated cheese slices showed more stable chemical features in comparison to the uncoated cheese samples in terms of moisture, lipid oxidation, pH, and titratable acidity (TA). Sensory evaluation of the preparations showed that the GS coating containing CPE and CEO significantly (p ≤ .05) had pleasant effects on the sensory features (taste, odor, texture, and overall acceptability) of the cheese samples during storage time. It was concluded that composite coating of GS containing CPE and CEO could improve the microbial, physical, chemical, and sensory features of ultrafiltration (UF) cheese during refrigerated storage.


| INTRODUC TI ON
Ultrafiltration (UF) system concentrates milk fat and protein and is used in cheese manufacture to upgrade cheese yield. The UF cheese is wrapped in plastic packaging in order to prevent envi- ronmental contamination and maintain the cheese quality (Fox et al., 2017). Plastics are the most utilized materials in food packaging. Due to the environmental concerns regarding the low recycling rate of plastics, various biodegradable and natural packaging types have been developed to decrease the use of plastic packaging (Cerqueira et al., 2010;Di Pierro et al., 2011). Compared to the plastic, the edible coatings can act as the carriers of active ingredients with antimicrobial and antioxidant properties, which are able to be appropriately distributed in the whole food (Ramos et al., 2012). In addition, the edible coatings produce no waste and residual materials; since, they are biodegradable and easily decomposable in the environment. Some are even consumed with food and can improve the sensory features of it (Cerqueira et al., 2010;Ramos et al., 2012). Carbohydrate polymers are appropriate choices in making coatings and films because of their unique colloidal properties. Starch polysaccharide is an abundant and cheap natural carbohydrate, which can be used in coating production. In order to improve the mechanical properties of starch for coating such as solubility in water and fragility, it can be mixed with a protein polymer, such as gelatin. Gelatin is a collagen-derived protein that forms a soft, pliable, and elastic gel that can be a good companion to starch rigid gels. Combining hydrophobic substances, such as fatty acids, vegetable oils, resins, surfactants, and waxes in hydrocolloid-based coatings, is one approach to boost the moisture barrier properties of them. Also, starch-based films and coatings are considered as selective barriers against oxygen and the incorporation of active agents, such as essential oils (EOs), into the starch coatings can inhibit lipid oxidation, preserving the freshness of product (Fakhouri et al., 2012;Moreno et al., 2018).
The use of EOs in food protection is often limited because of their intense taste, aroma, and potential toxicity. An interesting alternative approach to reduce the release of EOs in food is incorporation of those features into the edible coatings (Ghadermazi et al., 2016). Green cumin (Cuminium cyminum L.) is an aromatic plant belonging to the Apiaceae family, which is traditionally used as a condiment in foods. In addition to flavoring and aromatic properties, cumin also has various therapeutic effects (Alizadeh Behbahani et al., 2019. Cucumber (Cucumis sativus), a member of the Cucurbitaceae family, belongs to traditional Mediterranean diets. Antioxidant and antimicrobial properties of the peel, pulp, and seed extracts of this fruit have been reported in many studies (Fatima et al., 2018;Sotiroudis et al., 2010). Due to the long-standing interest of Iranians to consume cucumber and cumin seeds along with UF cheese, we decided to evaluate the effects of gelatin-starch (GS) edible coating containing cucumber peel extract (CPE) and cumin essential oil (CEO) on the physical, chemical, microbial, and sensory characteristics of UF cheese under refrigerated (4 ± 1°C) storage.

| Extraction of CEO
Fresh cumin seeds were purchased from local markets. CEO was extracted using hydrodistillation method for 3 h by a Clevenger-type apparatus (Simax, Pyrexfan). Anhydrous sodium sulfate was added to the CEO for dehydration and was preserved in opaque airtight glass vials at 4°C (Alizadeh Behbahani et al., 2020).

| Preparation of CPE
Fresh cucumbers were purchased from the local markets, washed, and rinsed by potable water and peeled. Then, cucumber peels were dried in the shade at environment temperature for 2 weeks. The dried samples were ground using a kitchen grinder. The obtained powder was mixed to each of the concentrated ethanol (98%), aqueous ethanol (70%), and water solvents with the ratio of 1:10; then, they were extracted by maceration, ultrasound, and microwave methods. In the maceration procedure, the samples were shaken at 250 rpm (revolutions per minute) for 24 h. Ultrasound apparatus (FAPAN) was used for ultrasound-assisted extraction (UAE) with the frequency of 20 kHz, the power of 50 W, and the temperature of 25°C for 30 min. In the microwave-assisted extraction (MAE), the samples were extracted by a microwave oven (SolarDOM, LG) with the power of 360 W for 10 min. Then, the obtained solutions were filtered through filter paper and concentrated by rotary evaporator apparatus (Lab Tech) at 40°C. The remained solvent was removed under vacuum at 50°C. After drying, the extracts were preserved at −18°C until being used (Gallo et al., 2010;Ince et al., 2013).

| Total phenolic content
Total phenolic contents of CPE were measured using the Folin-Ciocalteu reagent assay with gallic acid as a standard. Briefly, 500 μl of the extracts was mixed with 2.25 ml of distilled water (DW) and then, 250 μl of the Folin-Ciocalteu reagent was added. The mixture was vortexed for 1 min and allowed to react for 5 min. Then, 2 ml of sodium carbonate (7.5%) was added. After incubation at room temperature for 120 min, the absorbance of each mixture was measured at 760 nm. The same procedure was also used to a standard solution of gallic acid and a standard curve was prepared. The total phenolic values were considered as mg of gallic acid per gram of the sample (Machu et al., 2015).

| Reducing power test
The reducing power of the extracts was measured, according to the modified method of Jemli et al. (2016). One milliliter of extracts was added to 2.5 ml of the sodium phosphate buffer (0.2 M, pH 6.6) and 2.5 ml of potassium ferricyanide (1%). After incubation at 50°C for 20 min, 2.5 ml of trichloroacetic acid (10%) was mixed with the solution and was then centrifuged at 1792 g for 10 min. Finally, 2.5 ml of the obtained solution was added to 2.5 ml of the distilled water (DW) and 0.5 ml of ferric chloride (0.1%). After 10 min, the absorbance was read at 700 nm, against blanks that contained all materials except for the samples. Higher absorbance indicated higher reducing power. Butylated hydroxytoluene (BHT) (2 mg/ml) was considered as a positive control.

| ABTS radical scavenging activity test
The ABTS radical scavenging activity of the extracts was measured, based on the description of Ozgen et al. (2006). The ABTS (7 mM) and potassium persulfate (2.45 mM) solutions were prepared. They were then mixed together; after 16 h, the obtained solution was diluted with ethanol to earn an absorbance of 0.70 ± 0.02 at 734 nm.
Then, 2 ml of the mentioned solution was added to 200 μL of the extract solutions; after incubating for 1 min at room temperature, the absorbance was read at 734 nm using a spectrophotometer (Thermo Spectronic; Helios Gamma). The ABTS radical scavenging activity was calculated as follows: where, A blank is the absorbance of the blank (containing all materials except for the extract) and A sample is the absorbance of the sample.

| DPPH free radical scavenging test
The method of Fu et al. (2010) was considered for determining the potency of the samples to scavenge DPPH radical. Fifty microliters of the extracts was mixed to 2 ml of methanol DPPH (24 µg/ml) solution. The obtained solution was stored in dark at environment temperature for 60 min and the absorbance was read at 517 nm, using a spectrophotometer (Thermo Spectronic; Helios Gamma, UK).
where, A blank is the absorbance of the blank (containing all used reagents, except for the sample) and A sample is the absorbance of the sample. Butylated hydroxytoluene (BHT) (2 mg/ml) was used as the reference control in all antioxidant activity assays.

| Preparation of UF cheese
Fresh UF cheese was purchased from a dairy plant in Hamadan, Iran. Cheeses were made at the Hamadan dairy plant according to the UF cheese-making method, proposed by the Tetra Pak Company with some modifications by Hesari et al. (2006). After bactofugation, pasteurization (72°C-15 s), ultrafiltration, homogenization, and second pasteurization (80°C-20 s) stages, the retentate with a volume concentration factor of 5.4 kg of milk to 1 kg of retentate entered the starter tank, whereby adding the starter (1 g for 50 kg of retentate), the pH of milk reached the 6.2 level. Then, in the filler, rennet was mixed with water (2 g for 100 kg of retentate) and added to each cheese container. The coagulation tunnel, which was set at 37°C for 30 min, allowed the retentate to be converted into a precheese mixture. In the sealing machine, 4% salt was added onto the parchment paper on the top of cheese; then, the container was sealed using an aluminum foil.
In the preripening stage (37°C), after decreasing the cheese pH to 4.80, cheese samples were transferred to a cold room (9 ± 1°C) for cooling and ripening for 3-60 days. Three separate batches following the above procedure were considered for the production of each treatment.
The typical composition of the obtained cheese was as follows: moisture: 60%, fat: 6.75%, protein: 12%, total ash: 6%, pH: 4.35, and TA: 2.25 g of lactic acid/100 g of cheese. Under aseptic conditions, the obtained cheese was divided into seven groups, each containing 110 slices of 10 g and stored at 4°C until being used.

| Preparation of coating solutions
Ten grams of gelatin (G) powder was hydrated in 100 ml of distilled water (DW) and agitated by the heater magnetic stirrer (Fan Azma Gostar) in 250 rpm for 1 h under environment temperature. Then, the obtained solution was heated at 70°C for 10 min. Five grams of wheat starch (S) was dissolved in 100 ml of DW and was heated at 70°C for 10 min under constant shaking by the heater magnetic stirrer (Fan Azma Gostar) in 250 rpm. Glycerol (10%) was added as a plasticizer to G and S solutions. In order to obtain the desired edible coating, the resulting solutions of G and S were mixed in the equal ratio of 1:1 by a glass stirrer at environment temperature (Fakhouri et al., 2012). coatings; next, the treated samples were packaged in low-density polyethylene (LDPE) bags.

| Preparation of the treatments
Finally, all samples were stored at 4°C and analyzed for the microbial, physical, chemical, and sensory features on days 0, 14, 28, 42, and 56 of storage (Ramos et al., 2012).

| Scanning electron microscopy (SEM)
A piece of 10 × 10 × 2 mm 3 from all groups of coated cheeses was cut and immersed overnight in 2.5% buffered glutaraldehyde at 4°C.
The samples were then dehydrated in ascending concentrations of ethanol, dried in a critical point drier, and mounted on stubs. The processed samples were sputter-coated with gold and examined in a scanning electron microscope (JEOL JSM-840), operating at an accelerating voltage of 10 kV.

| Microbiological analysis
In order to measure the microbial population, 10 g of the cheese samples was poured aseptically in a stomacher pouch. After adding 90 ml of 0.1% sterile peptone water (Merck), the mixture was homogenized by a stomacher for 60 s. After serial dilution preparation, the samples were placed in the plates containing plate count agar (PCA) (Merck) and incubated at 35°C for 3 days for the total mesophilic count. Psychotropic bacteria were enumerated on PCA and the plates were incubated at 7°C for 10 days. Rose bengal chloramphenicol (RBC) selective agar (Merck) was used to evaluate the total mold/yeast count after 5 days of incubation at 25°C. Microbiological evaluation was expressed as the log of the number of colony forming units (CFU/g) (Yousef & Carlstrom, 2003).

| Weight loss
Weight loss was determined by weighting the samples at the beginning (W 1 ) and throughout the storage period (W 2 ). Weight loss of the samples was measured, based on the following equation: where, W 1 is the weight of the cheese slice in day 0 and W 2 is the weight of the cheese slice in the considered interval (AOAC, 2010).

| Color
Using image processing method, the colors of the cheese slice samples were evaluated (Shahraki et al., 2014). After taking photographs using a single-lens reflex (SLR) camera (Canon EOS 6D Mark II) in the dark box, the color parameters including L* (brightness/ darkness), a* (red/green), and b* (yellow/blue) were determined by Adobe Photoshop Cs5 software (Adobe Systems, Inc.). The overall color difference (ΔE) with the control sample was calculated as follows: where, s is the color of the sample during the storage period and L 0 , a 0 , and b 0 are the initial values (1 day after coating application), obtained for cheese under each experimental condition.

| Moisture
Moisture content was determined by drying the cheese samples to constant weight at 70°C in a vacuum oven (Fan Azma Gostar).
where, W 1 is the weight of the cheese slice before drying and W 2 is the weight of the cheese slice after drying (AOAC, 2010).

| Ash
Total ash content was determined by incineration of the samples at 550°C in an electric furnace (Fan Azma Gostar) (AOAC, 2010). The total ash content of the samples was calculated as follows:

| Protein
Protein determination was carried out by the Kjeldahl set (Simax, Pyrexfan). The amount (1 g) of the sample was weighted and digested by concentrated sulfuric acid under heat in a Kjeldahl flask.
The obtained solution was then distilled with 50 ml of sodium hydroxide (40%). An Erlenmeyer flask containing 25 ml of boric acid (2%) and protein indicator was determined to obtain the distillate until green discoloration of boric acid. For measuring the total nitrogen index, green discolored boric acid solution was titrated by 0.1 N of the sulfuric acid solution in order to achieve the primary color. The results were mentioned as milligrams of nitrogen per 100 g of the sample. Furthermore, the total protein content was Total ash ( % ) = Weight of obtained ash Weight of the sample × 100.
calculated, multiplying the nitrogen content by the conversion factor 6.38 (AOAC, 2010).

| Lipid
Fat content of the samples was measured by the Gerber method.
Three grams of the sample was added to a butyrometer. After adding 10 ml of sulfuric acid (20%) and 1 ml of amyl alcohol to the butyrometer, it was placed in a hot water bath at 65°C for 5 min. The butyrometer was then centrifuged at 136 g for 5 min. Then, the amount of extracted fat in the butyrometer was calculated in percentage (AOAC, 2010).

| Titratable acidity
Ten grams of the sample was added to 50 ml of distilled water and homogenized by homogenizer (IKA Ultra-Turrax T8) for 5 min.
The homogenates were heated at 40°C under stirring and were diluted to the final volume of 150 ml with distilled water. After centrifugation at 3000 g for 10 min, the supernatants were fil-

| pH
Ten grams of the sample was added to 50 ml of distilled water and homogenized by a homogenizer (IKA Ultra-Turrax T8) for 5 min. The homogenates were heated at 40°C under stirring and were diluted to the final volume of 150 ml with distilled water. After centrifugation at 3000 g for 10 min, the supernatants were filtered through paper filters. Then, pH measurement was carried out on a filtrate using a pH meter (Jenway) (AOAC, 2010).

| Thiobarbituric acid reactive substances
Lipid oxidation of the cheese slices was determined by the thiobarbituric acid reactive substances (TBARS) method, according to Unalan et al. (2013). For analysis, 10 g of the sample was homogenized in a warming blender using a mini jar (Waring Commercial, CAC 134) for 3 min at high speed with 50 ml of TBA (0.38%) and trichloroacetic acid (15%), prepared in HCl solution (0.25 N). Aliquots (5 ml) obtained from the homogenate were incubated in a water bath at 95°C for 15 min for color development. The samples were cooled for 10 min and then centrifuged at 4500 g for 25 min. The absorbance of the mixture reaction was measured at the wavelength of 532 nm with a spectrophotometer. A standard curve was generated using 1,1,3,3-tetraethoxypropane (TEP) and the data were considered as mg of malondialdehyde (MDA) per kg of cheese.

| Sensory analysis
A total of 20 undergraduate students (10 females and 10 males, 20-

| Statistical analysis
The obtained data were statistically analyzed by SPSS software (IBM SPSS Statistics V. 21) and reported as mean ± standard deviations

| Weight loss
Weight loss is determined by the weight difference of initial weight (W 1 ) of the samples and secondary (W 2 ) weight of them during the storage period that can be correlated to the evaporation of water or volatile compounds, oxidation, and deformation of sample constituents such as lipid and protein during storage period. In other words, weight loss of cheese depends not only on moisture loss but also on cheese chemical reactions during storage period (Cipolat-Gotet et al., 2020;Riahi et al., 2007). Figure  protein isolate (SPI) (Zhong et al., 2014), and coated "Saloio" cheese with whey protein isolate (WPI) (Ramos et al., 2012) or galactomannan (Cerqueira et al., 2010) was more than 15% during 20 days of storage. However, the weight loss of GS-coated samples in the present study was 5% at the end of the storage period, which can be related to the lower gas (water vapor, oxygen, etc.) permeability of the GS coating in comparison to other edible coatings. Figure 4b illustrates the hardness feature of the studied samples dur-

| Color
Color is an important factor in the acceptance of fresh products. Figure 4c illustrates the color change feature of the studied F I G U R E 4 Average changes in weight loss (a), hardness (b), and color (c) of the cheese slices during storage at 4°C. The significance of the acronyms is the same as in Figure 3. Different letters within the same interval (day) (a, b, c, etc.)   (e) of the cheese slices during storage at 4°C. The significance of the acronyms is the same as in Figure 3. Different letters within the same interval (day) (a, b, c, etc.) and the same treatment (A, B, C, etc.) indicate a statistically significant difference (p ≤ .05) coating compared to the uncoated samples during 60 days of storage at 10°C.

| Ash and protein
According to the obtained results, the amounts of protein (12%) and ash (6%) of all treatments showed no changes and significant differences (p > .05) during the storage period. For this reason, these values were not presented in a graph or table in the present study; because, the protein and ash of all samples were 12% and 6%, respectively, and presented no changes during 56 days of storage. A previous study found no differences in ash and protein content of  (Zhong et al., 2014). In contrast to our findings, Ramos et al. (2012) found no statistically significant differences between fat content of coated and uncoated Saloio cheeses during storage time.

| pH and TA
The initial pH of all samples (4.35) was stable until the 14th day of storage. Then, a descending pattern was observed until the end of the storage period, especially in the uncoated slices (Figure 5c). The lowest pH value belonged to the control (4.29), and others including CPE Lactobacillus spp., free fatty acids, and acidic amino acids' production due to the lipolysis and proteolysis phenomena. In agreement with our results, they reported that TA of Ricotta cheese samples, especially the uncoated ones, increased during 30 days of storage.

| Thiobarbituric acid reactive substances
The It has been reported that as a synthetic antioxidant, LDPE films containing BHT could enhance the oxidative stability of Asadero cheese during 100 days of storage (Soto-Cantú et al., 2008).

| Sensory analysis
Changes in the sensory characteristics (taste, odor, texture, and overall acceptability) of the cheese slices during storage period are presented in Table 1. There were no significant differences be- in the loss of water during the storage period. Zhong et al. (2014) reported that the hardness of Mozzarella cheese increased dramatically with water evaporation. They observed that edible coatings generally delay the hardening process of cheese and produce the softer cheese texture compared to the uncoated ones, which may be attributed to their water retention ability. Guerra-Martínez et al. (2012) reported that the hardness of fresh cheese has a high negative correlation with moisture levels (R = −.76); thus, a decrease in moisture levels during storage time results in hardness increase of the cheese. Cheese moisture itself showed to be an important factor defining cheese attributes, especially those regarding hardness; it seems that lower moisture loss of cheese during storage period may lead to the lower hardness of it. They reported that the hardness also has high negative correlations with protein content (R = −.83) and relative humidity (R = −.53). This is understandable, considering that the hydrolysis and hydration of casein protein contribute to the disintegration of casein matrix, as a result of which the hardness decreases. It is obvious that the moisture loss of cheese can decrease these reactions, leading to the hardness increase at the end. Ramos et al. (2012) reported that uncoated regional Saloio cheese had a harder texture than the ones coated with WPI during the storage period. The researchers also reported similar sensory properties to our results in reducing odor scores in uncoated specimens. Yilmaz and Dagdemir (2012) showed that the highest overall acceptability score belonged to the coated Kashar cheese with beeswax compared to uncoated or vacuum-packed samples during 120 days of storage.

ACK N OWLED G M ENT
The present research was financially supported by the Faculty of Veterinary Science, Bu-Ali Sina University, Hamedan, Iran.

CO N FLI C T O F I NTE R E S T
The authors have declared that no conflicts of interest exist.

E TH I C A L A PPROVA L
This study does not involve any human or animal testing.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.