Effect of Extract of Perilla Leaves on the Quality Characteristics and Polycyclic Aromatic Hydrocarbons of Charcoal Barbecued Pork Patty

Perilla leaves (P. frutescens) were purchased from a local market (Chuncheon, Korea). All samples were washed under running tap water prior to extraction. The samples were lyophilized, ground, passed through a 20-mesh sieve, and stored at –20°C until extraction. The sample powders were macerated with 50%, 70%, or 90% ethanol (1:50 w/v) for 3 or 6 d at 25°C. The obtained extracts were filtered through Whatman No. 4 paper (Whatman, Clifton, NJ, USA), and the filtrates were collected and concentrated using a rotary evaporator at 40°C. The concentrated extracts were lyophilized and stored at –20°C until analysis.

DPPH radical scavenging activity was analyzed following the method of Cho et al. (2021) with slight modifications. The extract solution (100 μL) was placed in 100 μL of methanolic solution containing DPPH radicals (0.2 mM) in a 96-well microplate. The mixture was allowed to react for 30 min at 25°C in the dark. The absorbance of each extract solution was measured at 517 nm using a spectrophotometer (SpectraMax M2, Molecular Devices, San Jose, CA, USA). The standard curve was established using Trolox, and the DPPH values were expressed as mmol Trolox equivalent (TE)/g dry matter (DM).

The FRAP assay was performed as described by Kim et al. (2019) with slight modifications. The FRAP working solution was prepared with 300 mM acetate buffer, 10 mM 2,4,6-tripyridyl-S-triazine in 40 mM HCl, and 20 mM FeCl₃·6H₂O solution mixed at a ratio of 10:1:1 (v/v/v). Twenty-five microliters of the extract sample were reacted with 175 μL of FRAP working solution for 30 min at 37°C in the dark. The absorbance of the reacted solution was determined at 590 nm using a spectrophotometer (Spectra Max M2, Molecular Devices). The FRAP activity was expressed as mmol TE/g DM.

The ORAC assay was performed as described by Kim et al. (2019) with slight modifications. To measure ORAC, a mixture composed of 25 μL PLE and 150 μL fluorescein (80 nM) was mixed and incubated for 15 min at 37°C. After incubation, 25 μL 2,2′-azobis (2-amidinopropane) hydrochloride (150 mM) was added to generate peroxyl radicals, and each well contained a final volume of 200 μL. The change in the absorbance of the reacted extract sample was recorded every minute at an excitation and emission wavelength of 480 and 520 nm, respectively, at 37°C. The ORAC assay was performed using a spectrophotometer (Spectra Max M2, Molecular Devices). Trolox was used as the standard, and the results are expressed as mmol TE/g.

TPC was measured using the Folin-Ciocalteu colorimetric method as described by Cho et al. (2021) with slight modifications. The 50%, 70%, and 90% ethanol extracts were dissolved in 50%, 70%, and 90% ethanol, respectively. Each extract solution was diluted in ethanol. The diluted extract solution (0.5 mL) was mixed with 5 mL distilled water and Folin-Ciocalteu phenol reagent (Sigma-Aldrich, St. Louis, MO, USA) and kept for 3 min, after which 1 N Na₂CO₃ was added, and the mixture reacted for 90 min at 25°C in the dark. The absorbance of the reacted samples was measured at 760 nm using a spectrophotometer (Spectra Max M2, Molecular Devices). A standard curve was established using gallic acid, and the TPC was expressed as mg GAE/g.

Frozen lean pork leg and pork back fat were purchased from a local market in Chuncheon, Korea. The visible fat on the pork legs was trimmed. The defatted pork leg and fat were minced through the first 8 mm plate and then through the second 4 mm plate using a meat chopper (M-12S, Fujee, Siheung, Korea). After mincing, the defatted pork leg and fat were mixed with salt, water, and ethanol or perilla leaves ethanol extract using a mixer (5 KPM50, Kitchen Aid, Benton Harbor, MI, USA). The formulations of the pork patties are presented in Table 1. Approximately 80 g of the mixture was formed into pork patties using a Petri dish (15 mm thick×90 mm diameter). The patties were covered with polyethylene film and stored under refrigeration (4°C) for 24 h before charcoal barbecue.

The charcoal barbecue condition was performed according to Kim et al. (2021a), with slight modifications. Black charcoal with extruded charcoal (for ignition) was placed into a garden-type grill (55 cm wide, 34 cm long, and 14 cm high: Allcook, Chilgok, Korea). The temperature of the charcoal fire was measured using a laser thermometer (IR-302, Custom, Tokyo, Japan). When the temperature of the charcoal fire reached 550°C to 600°C, the barbecue started; the distance from the charcoal was 8 cm for medium-cooked patties, the barbecue time was 9 min (5 min, front side; 4 min, back side), and the internal temperature of the patty was approximately 71°C. For well-cooked patties, the barbecue time was 16 min (8 min on each side), and the internal temperature of the patties was approximately 80°C. The patty was turned once during the barbecue period.

The proximate composition was measured according to the methods of the Association of Official Agricultural Chemist (AOAC, 2012). The moisture content of the barbecued pork patties was measured by weight loss after oven-drying at 105°C for 12 h. The crude protein content was measured using the Kjeldahl method. Crude fat content was measured by solvent extraction using ether. The burned pork patties in the furnace at 550°C were analyzed for crude ash.

Cooking loss was calculated as the difference in the weight of pork patties before and after cooking. The equation for cooking loss of barbecued pork patties is as follows:

The pH was determined using a pH meter (Orion 230A, Thermo Fisher Scientific, Waltham, MA, USA). Ten grams of pork patty were homogenized with 90 mL distilled water using a homogenizer (PolyTron^® PT-2500E, Kinematica, Malters, Luzern, Switzerland).

The instrumental color of raw and barbecued (medium- and well-cooked) pork patty was determined using colorimeter (CR-400 Minolta colorimeter, Minolta, Osaka, Japan) with an aperture of 8 mm size and an illuminant-C. The color values of CIE L*, CIE a*, and CIE b* of the raw pork patties were measured after removing the polyethylene films for 10 min. The color values of the barbecued pork patties were measured after barbecue.

TBARS was analyzed using the method described by Kim et al. (2022). The pork patties of 5 g were added in 50 μL of 7.2% tert-butyl-4-hydroxyanisole and 15 mL distilled water and then homogenized for 30 s using a homogenizer (Polytron PT-2500E, Kinematica, Lucerne, Switzerland). A 1 mL homogenate was transferred to a test tube, and 2 mL of thiobarbituric acid (TBA)/trichloroacetic acid (TCA) solution (20 mM TBA/15% TCA) was added. For a blank sample, 2 mL of each patty homogenate was added to 2 mL of 15% TCA solution. The sample mixture was incubated in a water bath at 90°C for 15 min to develop color. After incubation, the samples were cooled in ice water for 10 min and centrifuged at 2,000×g at 4°C for 15 min. The absorbance of the supernatant solution was measured at 531 nm using a spectrophotometer (Spectra Max M2, Molecular Devices). The TBARS content was expressed as mg of malondialdehyde (MDA)/kg of patty, as follows:

Sensory evaluation of barbecued pork patties was performed by 15 panelists from the College of Animal Life Sciences, Kangwon National University. The sensory properties were evaluated for medium- and well-cooked pork patties. Barbecued pork patties were evaluated for color, aroma, flavor, taste, juiciness, texture, and overall acceptability using a 9-point scale as follows: Color, aroma, flavor, taste, texture, and overall acceptability (1=extremely undesirable, 9=extremely desirable) and juiciness (1=extremely low, 9=extremely high).

The PAHs in barbecued pork patties were analyzed using the methods of Kim et al. (2021a) with slight modifications. The sample (2.5 g) was placed in a 50 mL conical tube, and 5 mL ethyl acetate/acetonitrile (20:80, v/v) and 1 mL ISTD mix (naphthalene-d₈, acenaphthene-d₁₀, phenanthrene-d₁₀, chrysene-d₁₂, and perylene-d₁₂, 400 ng/mL) was added. The mixture was ultrasonicated for 20 min and centrifuged (1,968×g for 7 min at 15°C). The supernatant was transferred to a 15 mL conical tube. The remaining pellet was re-extracted using 5 mL ethyl acetate/acetonitrile (20:80, v/v) and the supernatants were combined. The combined supernatants were evaporated under vacuum conditions using a rotary evaporator (Scilab Korea, Seoul, Korea) until a volume of 2 mL was reached, and distilled water (0.5 mL) was added. Subsequently, the mixture was purified using a Captiva EMR-Lipid cartridge (Agilent Technologies, Santa Clara, CA, USA) and 0.625 mL of ethyl acetate/acetonitrile/water (16:64:20, v/v/v) was eluted through the cartridge. The 1.875 mL eluent was mixed with 2.625 mL distilled water and 1.2 mL isooctane in a new 15 mL conical tube, followed by vigorous shaking. Subsequently, the mixed samples were centrifuged (1,968×g, 7 min, 15°C), and the supernatant was transferred to a glass tube, concentrated by nitrogen gas, and analyzed by GC/MS (Agilent 8890 GC with an Agilent 5977 B GC/MSD: Agilent Technologies).

The PAHs were separated using a DB-EUPAH capillary column (20 m×0.18 mm inner diameter, 0.14 μm film thickness: Agilent Technologies). The carrier gas used pure helium (99.99%) at a constant flow rate (1.2 mL/min). The extracted samples were injected in the splitless mode (1 μL) and maintained at an initial temperature of 300°C. The source temperature was 290°C, and the mass selective detector temperature was 310°C. The initial temperature of the oven was 70°C, increased to 190°C at a rate of 30°C/min, increased to 290°C at a rate of 10°C/min, and maintained for 5 min; it was then increased to a final temperature of 320°C at a rate of 30°C/min and maintained for 1 min. The electron ionization of the mass spectrometer was operated at 70 eV, and data acquisition was conducted in the selective ion monitoring mode for the characteristic molecular ions of each PAH. The ISTD mix and 16 PAH standards (naphthalene, acenaphthylene, acenaphthene, fluorene, phenanthrene, anthracene, fluoranthene, pyrene, benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, indeno[1,2,3-cd] pyrene, dibenzo[a,h]anthracene, and benzo[g,h,i]perylene) were used. Nine-point calibration curves from 9 to 2,400 ng/mL were used to quantify the contents of the 16 PAHs in the samples. All PAHs were quantified by the relative response factors related to the list, in which the individual characterization of light PAHs, heavy PAHs, PAH4, PAH8, and a total of 16 PAHs are shown in Table 2.

All determinations were carried out in triplicate. All data were analyzed using the general linear model procedure of the SAS program (ver. 9.2; SAS Institute, Cary, NC, USA) using one-way analysis. Tukey’s test was used to determine the significance of the differences in the mean values for the different extract samples. Differences were considered significant at p<0.05.

Three different sets of antioxidant activity measurements were performed in this study, and the results are shown in Table 3. Based on these results, both ethanol concentration and extraction period significantly influenced the antioxidant activities of PLE (p<0.05). However, exceptionally, the DPPH assay showed that the highest activities were obtained when perilla leaves were extracted using 90% ethanol solution for 3 d (593.27 μmol TE/g DM), with no notable effect following extension of the extraction period (day 6; 521.92 μmol TE/g DM; p<0.05). In addition, the FRAP assay yielded similar results, wherein the highest activities were observed when perilla leaves were extracted using 90% ethanol in comparison to that by 70% and 50% (p<0.05). The extension of extraction period, however, significantly increased the antioxidant activities by 6% from 0.91 mmol TE/g DM on day 3 to 0.97 mmol TE/g DM on day 6 (p<0.05). Furthermore, the ORAC assay indicated that PLE produced with 70% and 90% ethanolic solution shared similar antioxidant activities, but notably than those of 50% (p<0.05). No remarkable effects were observed when the extraction period was extended.

Studies have shown that the antioxidant activities of certain compounds are strongly correlated with the concentration of phenolic compounds (Gong et al., 2018). Antioxidant activity may have different correlations with TPC because there are different principles depending on the different methods of antioxidant activity. The method of FRAP is based on the ferroin analog reduction and can determine the total reducing capacity (Antolovich et al., 2002). The method of ORAC is used to determine the antioxidant activity against peroxyl radicals produced by AAPH and is the most biologically relevant method for analyzing antioxidant activity (Ou et al., 2002). The assay of DPPH is based on the principle that DPPH^o is reduced to DPPH2 when it accepts a hydorgen (H) atom a scavenger molecule (Mishra et al., 2012). Additionally, each antioxidant activity may have a different level of correlation with TPC, and thus may not always have a positive correlation with TPC. A previous study (Zheng et al., 2018) showed that the linear regression (R²) correlation coefficients of the three antioxidants (FRAP, ABTS, and DPPH) and TPC differed by 0.8588, 0.7587, and 0.6344, respectively. The efficacy of solutions in extracting phenolic compounds is one of the most determining factors. Compared to aqueous solutions, organic solutions have been reported to exhibit a stronger capacity to extract major phenolic compounds, including sinapic, chlorogenic, isovanillic, caffeic, and gallic acid (Gong et al., 2018). Consistent with the results of this study on TPC, the order of phenolic compound concentration on day 3 from the highest to the lowest for 90%, 70%, and 50% ethanolic solution was 102.39, 81.97, and 79.15 mg GAE/g DM, respectively (p<0.05). Similarly, on day 6, the TPC for PLE produced using 90%, 70%, and 50% ethanolic solution were 114.93, 91.48, and 81.91 mg GAE/g DM, respectively (p<0.05), indicating a stronger capability of high ethanol concentration in extracting antioxidant compounds from perilla leaves. According to previous study (Kang and Lee, 2011), Cyanidin-3-O-(6-O-coumaroyl)-glucoside5-O-glucoside and rosmarinic acid were major phenolic compounds in the perilla leaves, and their contents showed more than 15% and 60% in total content, respectively. Additionally, regarding the essential role of TPC concentration in dictating the functional properties of foods, extraction of perilla leaves for 6 d resulted in a notably higher TPC compared to that for 3 d (p<0.05). Therefore, in this study, extraction of perilla leaves was performed using 90% ethanolic solution for 6 d.

The proximate composition of pork patties after barbecue cooking and treatment with PLE is shown in Table 4. The crude protein and crude ash compositions of barbecued pork patties were not greatly influenced by treatment with PLE. The crude fat percentage was significantly higher only when PLE were added in medium patties (p<0.05). In contrast, the moisture percentage of medium patties was lower when PLE was added in comparison to the control group (p<0.05). Its percentage was significantly higher in medium-cooked than in well-cooked patties, irrespective of PLE addition percentage (p<0.05). No significant effect was recorded with respect to the different percentages of PLE addition on the proximate composition of pork patties. Furthermore, the significant effect of PLE addition on the proximate composition of the pork patties was only seen in the medium-cooked patties, with no notable effect in the well-cooked patties. Regarding the different cooking levels, well-cooked patties had significantly higher crude protein and crude ash percentage than those of medium-cooked patties. (p<0.05). In contrast, in crude fat, the medium-cooked patties were significantly higher than the well-cooked patties only when PLE was added (p<0.05). In the case of moisture content, it was observed that well-cooked patties were lower than medium-cooked patties regardless of with or without PLE. Notable differences in some of the proximate attributes were mainly caused by the acceleration of moisture loss, which consequently changed the state of other variables (Berry, 1994). Cooking at a higher final internal temperature resulted in a lower moisture percentage but higher moisture content in the retained protein, which agrees with previous reports (Barido et al., 2022a; Berry, 1994).

The inclusion of PLE in pork patties did not significantly affect to the percentage of cooking loss, as different cooking levels did. As seen in Table 5, a markedly higher cooking loss percentage was observed in well-cooked patties than in medium-cooked patties at any given PLE treatment (p<0.05). Apart from the excessive loss of moisture at a high final internal temperature, poor fat retention in high-fat formulated patties is another important factor that causes a remarkable yield loss in a well-cooked patties. This study revealed that the fat retention ability of patties formulated with low fat at 4% was remarkably higher than that of patties formulated with 20% of fat content at high temperature cooking (Berry, 1994); the pork patty in this study formulated with 20% of fat might confirm a similar trend. Furthermore, the addition of PLE to pork patties and different final internal temperatures had no significant effect on pH. This finding agrees with a previous study, wherein no significant effect on pH value was observed when 0.4% of perilla leaf powder was added to emulsion-type pork sausage (Kim et al., 2005).

Instrumental color is critically important in influencing the organoleptic perception of consumers toward meat products; therefore, the documentation of color changes after cooking at different final internal temperatures together with the addition of PLE is of critical importance. The effects of different final internal temperatures and PLE treatments on the instrumental color of pork patties are shown in Table 6. The addition of PLE in a concentration-dependent manner significantly reduced the CIE L* of medium- and well-cooked pork patties compared to the control (p<0.05). Similarly, the CIE a* and CIE b* of the pork patties were significantly reduced following treatment with PLE at any final internal temperature, wherein the highest reduction was shown in groups treated with the highest percentage of PLE at 0.4% compared to other treatments (p<0.05). Previous studies have shown that the primary color of plant-derived phenolic extracts might be absorbed by meat during processing (Barido et al., 2021; Maqsood et al., 2015; Smaoui et al., 2019). According to research of Boles and Pegg (2010), during the cooking process, myoglobin in meat is denatured. Because it is not affected by cooking at the same temperature and intensity, it shows a red color at the end temperature, and the brown color seen in cooked meat products is due to metmyoglobin. The results on the effect of treatment with PLE on the instrumental color of pork patties are in agreement with those of a previous study by Kim et al. (2005), who found a decrease in both CIE L* and CIE a*, while maintaining a significant increase in CIE b* following inclusion of 0.4% perilla leaf powder in emulsion-type pork sausage.

The TBARS assay is one of the prominent analyses to measure the degree of lipid oxidation in meat and meat products; it is based on the concentration of MDA free radicals (Kim et al., 2020). The changes in TBARS score following treatment are depicted in Table 7. The TBARS value in the control samples was 5.88 mg MDA/kg in medium-cooked patties, while it was significantly lower at 3.39 in well-cooked patties. Moreover, this study showed a significant inhibitory effect of PLE addition, wherein at any final internal temperature, the PLE groups exhibited a markedly lower MDA concentration compared to that of the control (p<0.05). However, no further differences were observed with respect to different PLE addition percentages (p<0.05). Efforts to reduce excessive lipid oxidation in meat products using natural compounds have been carried out for decades, and certain phenolic acids are the most effective (Zhao et al., 2019). In perilla leaves, the abundant content of anthocyanin and rosmarinic acid is believed to provide strong antioxidant activities (Li et al., 2016; Zhu et al., 2014). Furthermore, the potent inhibitory effect of PLE on lipid oxidation agrees with a previous study on surimi fish balls (Zhao et al., 2019).

The effects of cooking level and addition of PLE on the sensory characteristics of pork patties are shown in Table 8. In terms of color perception, the addition of PLE at any final internal temperature promoted a lower perception compared to the control group (p<0.05). Furthermore, a higher addition percentage of PLE tended to further decrease the color perception score of the pork patties. Accordingly, the flavor perception displayed a notably lower score when well-cooked pork patties were supplemented with PLE at any percentage in comparison to the control treatment (p<0.05). However, no significant difference was observed in medium-cooked pork patties. No further significant differences were observed in the aroma, taste, juiciness, and texture perception of pork patties after treatment with PLE (p>0.05). Meanwhile, the higher final internal temperature (well-cooked) significantly lowered juiciness and texture perception compared to medium-cooked patties for all treatments (p<0.05). The acceleration of moisture loss along with excessive muscle shrinkage was assumed to be responsible for this effect (Barido et al., 2021; Viegas et al., 2012). In addition, a different trend was recorded between the control and PLE treatments regarding flavor and taste perception. In control treatment, the scores for flavor and taste were significantly higher when pork patties were well-cooked (p<0.05), whereas in the highest PLE addition group (0.4%), flavor perception received significantly higher score when pork patties were only medium-cooked. Eventually, for overall acceptability, the addition of PLE at 0.2% maintained a similar preference for pork patties compared to the control group at any cooking level, whereas the addition of PLE up to 0.4% tended to receive lower score compared to the 0.2% in medium-cooked patties and control group in well-cooked samples (p<0.05). The lower overall acceptability of 0.4% PLE in this study may be due to the low perception of color and flavor attributes.

Table 9 displays the effects of different final internal temperatures and the addition of PLE on the 16 PAHs contents of the roasted pork patties. In control samples without any PLE addition, the total concentrations of the 16 PAHs were 198.33 and 280.79 ng/g for medium- and well-cooked patties, respectively, wherein cooking to a higher final internal temperature promoted higher production of PAHs (p<0.05). However, a similar trend was not observed in pork patties treated with PLE, in which the concentration of total PAHs was significantly lower in well-cooked patties than medium-cooked patties (p<0.05). In addition, this study showed a different inhibitory effect of PLE addition at different cooking levels of pork patties. In medium-cooked patties, the addition of 0.2% PLE was not found to significantly prevent the formation of PAHs in pork patties compared to the control groups, except for light PAHs, such as phenanthrene (p<0.05). Meanwhile, when the pork patty was added with the higher concentration of PLE at 0.4%, the formation of the light PAHs (naphthalene, phenanthrene, anthracene, pyrene) was significantly suppressed; no effect on the heavy PAHs (including 4 PAH and 8 PAH) was observed compared to the control.

Furthermore, regarding the addition of PLE to pork patties under well-cooked patties, the 0.2% PLE exhibited a significant inhibitory effect on most of the light PAHs (p<0.05), except fluorene (p<0.05). Moreover, the suppressive activity of 0.2% PLE on heavy PAHs was recorded for benzo[k]fluoranthene and dibenzo[a,h]anthracene compared to the control (p<0.05). However, in well-cooked patties, the addition of 0.4% PLE showed the strongest inhibitory effect on the formation of both light and heavy PAHs (p<0.05). In addition, owing to the hazardous effect of benzo[a]pyrene in possibly increasing the risk of tumors and cancers due to the mode of action and interference with DNA replication in human cells (Park et al., 2017), the reduction of benzo[a]pyrene concentration in meat and meat products is essential. In this study, the addition of 0.4% PLE to pork patties exhibited the highest suppressive effect on the formation of the benzo[a]pyrene in well-cooked patties among the various treatments (p<0.05). Furthermore, the production of the highly toxic 4 PAHs and 8 PAHs in pork patties was remarkably inhibited by the inclusion of the 0.4% PLE. In addition, this study indicated a stronger inhibitory effect of PLE addition on the formation of both 4 PAHs and 8 PAHs in a percentage-dependent manner, wherein 0.4% addition exhibited a stronger effect than 0.2%.

At high temperatures, small organic molecules in meat undergo pyrolysis, resulting in the formation of more free radicals, and the stable polynuclear aromatic compounds in cooked meat might be a consequence of this process (Viegas et al., 2014). Owing to the involvement of free radicals in the production of PAHs, studies have inferred that the reduction of free radicals by antioxidants strongly contributes to the reduction of PAH concentration in meat products (Cordeiro et al., 2020; Gong et al., 2018; Park et al., 2017; Viegas et al., 2014; Wang et al., 2019). While the continuous application of synthetic antioxidants may detrimentally affect human health owing to their potential toxicity (Bera et al., 2006), natural compounds containing abundant polyphenols have been shown to possess potent antioxidative activities (Gong et al., 2018; Kim et al., 2021b).

In our study, the highest inhibitory effect of 0.4% PLE treatment on PAHs under well-cooked conditions might be due to the higher antioxidant activity of the phenolic compounds at high temperatures. These results are in accordance with a previous report by Wang et al. (2019), who found a higher inhibitory effect of phenolic compounds on PAHs at higher grilling temperatures (240°C and 270°C) than at low temperatures (210°C). In addition, well-cooked patties treated with 0.4% PLE showed a higher inhibitory effect on benzo[a]pyrene and 4 PAHs than medium-cooked patties. As explained by Min et al. (2018), the factors influencing PAH formation may include heating conditions, the presence of water, and antioxidants. Also, this trend was correlated with previous study, which the interaction effect between natural materials having antioxidant activity and doneness was observed for carcinogenic PAHs (Kim et al., 2021a).