Optimizing oxidative stability and quality in fish balls: the synergistic effects of cooking methods and turmeric powder addition

Abstract

This study examined the effects of cooking methods (boiling, steaming, pan-frying, oven-roasting, and microwave cooking) and turmeric powder (TP) addition on the physicochemical properties and oxidative stability of fish balls. Higher cooking temperatures increased moisture evaporation. Pre-cooking before freezing reduced hydroperoxide and malondialdehyde levels in fish balls over 4 months of storage, especially noticeable in boiled, pan-fried, and oven-roasted fish balls. Additionally, TP addition significantly improved the total reducing capacity of the fish balls, an effect that persisted even after cooking. Consequently, hydroperoxide and malondialdehyde levels decreased during storage, with a more pronounced effect observed in steamed and microwave-cooked fish balls. Principal component analysis identified distinct clusters based on cooking methods and TP addition, categorizing the fish balls into three groups, and highlighting the complex interplay between these two factors. These findings offer valuable insights into extending the shelf life of fish balls through optimized antioxidant and cooking methods.

Introduction

Culinary practices and dietary habits significantly influence a diverse range of food products. Among these, fish-based products have gained considerable attention due to their nutritional value, offering high-quality proteins and fats. Consequently, they are widely consumed across various cultures (Chen et al., 2022). Fish and fish products are unique sources of essential ω-3 polyunsaturated fatty acids (PUFAs), such as EPA and DHA. While PUFAs offer significant nutritional benefits for human health, their double-bond structure makes them prone to oxidation, mainly due to the inherent weakness of the bond (Barden and Decker, 2016). Lipid oxidation, initiated by the formation of alkyl radicals, ultimately leads to deterioration in food quality and reduced shelf life of lipid-rich foods (McClements and Decker, 2007).

In recent years, growing interest has focused on innovative approaches to overcome these challenges and develop products that meet consumer demand for healthier and more appealing food choices (Shah et al., 2014). Ingredient supplementation and cooking methods have emerged as key strategies for enhancing the attributes of food products. For example, antioxidant-rich ingredients like curcumin-rich turmeric powder (TP) have been added to fried fish balls (Lee and Surh, 2022a), while lutein-rich pistachio powder and anthocyanin-rich purple sweet potato powder were integrated into deep-fried doughnuts (Lee and Surh, 2022b, 2023). Fresh herbs abundant in polyphenolic compounds have been used in deep-frying or air-frying sardine fillets (Ferreira et al., 2017), and tomato and garlic extracts have been added to frozen fish croquettes (Gokoglu et al., 2012). However, the impact on oxidative stability has shown conflicting results, possibly due to alterations in the food matrix resulting from ingredient supplementation and cooking. This is especially relevant when antioxidant-rich ingredients are traditionally added before cooking. For instance, hydrophilic components may leach into boiling water, while hydrophobic components are more likely to dissolve in cooking oil, potentially reducing the anticipated antioxidant activity (McGee, 2007). Although cooking processes offer benefits like microbial safety and sensory improvements, they can also result in adverse effects, such as oxidation of PUFAs, protein denaturation, and degradation of heat-sensitive pigments or antioxidants (Serpen et al., 2012). Furthermore, the choice of cooking techniques affects factors like moisture and fat composition, due to distinct heat transfer mechanisms associated with each method (McGee, 2007). Parameters including water activity, heat intensity, and O₂ concentration, all closely related to cooking techniques, can influence the rate constant of lipid oxidation (Barden and Decker, 2016). A previous study on TP-added fish balls revealed varying patterns of hydroperoxide accumulation over time depending on the cooking method used. For instance, a power function was observed for air-frying, while a logarithmic progression following a lag phase was evident in deep-frying (Lee and Surh, 2022a). This underscores the importance of studying the antioxidant activity of incorporated ingredients in conjunction with the cooking process, especially for high-lipid foods.

This study aims to investigate the effects of commonly used cooking methods (boiling, steaming, pan-frying, oven-roasting, and microwave cooking) on the oxidative stability of PUFA-rich mackerel fish balls. Additionally, we examine the potential inhibitory impact of TP when added to fish balls subjected to each cooking method, considering its likely effects on both quality attributes and shelf life. By exploring these interactions, this study sheds light on how these factors collectively shape the final product.

Materials and methods

Chemicals and food materials

Folin-Ciocalteu’s phenol reagent, gallic acid monohydrate, 2,2′-diphenyl-1-picrylhydrazyl (DPPH), FeSO₄·H₂O, barium chloride, ammonium thiocyanate, thiobarbituric acid (TBA) and 1,1,3,3-tetraethoxypropane (TEP) were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). Cumene hydroperoxide (80%) was purchased from Thermo Fisher Scientific (Chicago, IL, USA). NaOH, ethanol, ethyl ether, chloroform, methanol, butanol, and sulfuric acid were purchased from Daejung Chemicals & Metals Co., LTD (Siheung, Korea). Boric acid and Na₂CO₃ were purchased from Showa Chemical Industry Co. (Tokyo, Japan). Turmeric powder (TP) (100%, India) was purchased from Daeseong (Seoul, Korea). Mackerel (Andong, Korea) consisting of 98.998% mackerel, 1% sea salt, and 0.002% bamboo salt, was stored in a freezer at − 20 °C until use. Wheat flour and soybean oil were purchased from Cheiljedang (Seoul, Korea) and Sajo (Seoul, Korea), respectively.

Fish ball preparation

Individually packaged frozen eviscerated mackerel were thawed under running tap water for 30 min. The fins and bones were removed, and the mackerel was homogenized using a homogenizer (SHMF-3500SS, Hanil Electric, Seoul, Korea). For the control fish balls, fish homogenates were mixed with wheat flour at a 5:1 (w/w) ratio to ensure proper cohesion. In the case of fish balls with added TP, half of the wheat flour was substituted with TP, since kneading and shaping proved difficult without the presence of wheat flour. After kneading, 15 g portions of the fish paste were hand-shaped into spheres with diameters of approximately 2.5 cm.

Cooking method

Raw fish balls were divided into six groups and subjected to five different cooking methods: microwave cooking, steaming, boiling, pan-frying, and oven-roasting. One group remained uncooked. 12 fish balls were prepared for each cooking method. For boiling, the fish balls were cooked in 2.5 L of boiling water at 98 °C without a lid for 4 min. For steaming, the fish balls were placed in a steamer containing 2.5 L of boiling water and steamed for 9 min. In the case of pan-frying, the fish balls were cooked in a frying pan coated with 20 g of soybean oil, preheated to 220 °C, and rolled continuously for 2 min and 30 s. For oven-roasting, the fish balls were cooked in a preheated convection oven (DHC3-A, Softmill, Gwangju, Korea) at 200 °C for 4 min. For microwave cooking (MW209QB, AC 220 V, 60 Hz, Jinyang, Jinju, Korea), the fish balls were cooked intermittently for 1 min, with a 20-second interval between each cycle. Cooking times were determined based on preliminary trials, wherein the fish balls were cooked until their internal temperature reached 70 °C, as measured by a Daihan IP67 waterproof digital thermocouple thermometer (Daihan Scientific, Wonju, Korea). After cooking, the fish ball samples were cooled to room temperature, frozen at − 80 °C for 24 h, and then stored at − 20 °C for 4 months. Prior to analysis, the frozen fish balls were thawed in a refrigerator at 4 °C and homogenized (Hanil Electric).

Proximate composition and color property

The moisture and fat content of the fish balls were determined using standard AOAC methods (AOAC, 2010). Specifically, the moisture content was ascertained using a 105 °C oven-drying method (OF-12, Jeio Tech, Daejeon, Korea), while the fat content was assessed through Soxhlet extraction with ethyl ether (E-816, Buchi, Flawil, Switzerland). The surface color of the fish balls was measured using a CR400 chroma meter (Konica Minolta Sensing, Osaka, Japan). The resulting data were expressed in terms of lightness (L), redness (a), and yellowness (b).

Antioxidant activity

The antioxidant activity of the fish balls was assessed by measuring their total reducing capacity (TRC) and DPPH radical scavenging activity. For the purpose, 1 g of fish ball homogenates was extracted with ten volumes of ethanol in a shaking water bath (BS-21, Jeio Tech) at 37.5 °C, at a speed of 180 rpm, for 12 h. Ethanol was chosen as the extraction solvent based on previous research indicating it yielded the highest antioxidant activity when TP was extracted using various solvents with different polarity indices (Lee and Surh, 2022a). Following extraction, the homogenates were centrifuged at 4 °C and 3,091×g for 20 min (5810R, Eppendorf, Hamburg, Germany) and then filtered (Qualitative Filter Paper No. 2, Whatman, Maidstone, UK). The resulting supernatants were utilized for further analyses.

TRC was determined using the Folin-Ciocalteu reagent method, as described by Singleton et al. (1999). 1 mL of the extract was combined with 1 mL of 10-fold diluted Folin-Ciocalteu’s reagent and allowed to stand for 5 min. Subsequently, 1 mL of 10-fold diluted N₂CO₃ was added to the mixture, which was then shaken. The reaction mixture was stored in a dark room for 1 h to allow the reaction to proceed. Absorbance was measured at 700 nm using an EON microplate spectrophotometer (Biotek Instruments, Winooski, VT, USA). TRC was quantified using a standard curve (R² = 0.9931), which was constructed with gallic acid as the standard, and the results were expressed as gallic acid equivalents (GAE).

DPPH radical scavenging activity was assessed following the method described by Brand-Williams (1995). 100 µL of the extract were mixed with 1 mL of a DPPH solution (0.058 g/100 mL). The absorbance of the mixture was measured at 525 nm, at 5-min intervals, for a total of 5 h using an EON microplate spectrophotometer (Biotek Instruments). A blank solution containing ethanol and a positive control solution containing 0.2 mM gallic acid were used for comparison. DPPH radical scavenging activity was calculated as the percentage of absorbance reduction attributable to the sample extract.

Oxidative stability

The oxidative stability of fish balls prepared using different cooking methods was assessed by monitoring changes in primary and secondary lipid oxidation products over a 4-month storage period. To prepare the sample extract, 1 g of homogenized fish balls was mixed with 20 mL of a chloroform:methanol solution (2:1, v/v) for 30 s. The mixture was then centrifuged at 4 °C and 3,091×g for 20 min (Eppendorf). The supernatant was filtered using Qualitative filter paper No. 2 (Whatman) and served as the sample extract for the analysis of hydroperoxides as primary lipid oxidation products and malondialdehyde (MDA) as a secondary lipid oxidation product.

Hydroperoxides content was determined using the ferric thiocyanate method, as described by Chapman and Mackay (1949). This method relies on the ability of peroxides to oxidize ferrous ions (Fe²⁺) to ferric ions (Fe³⁺). The sample extract was diluted with the same extraction solvent to ensure that the absorbance of the final reaction mixture fell within the range of 0.2 to 0.8. The volume was adjusted to 3 mL using a methanol:butanol solution (2:1, v/v). 15 µL of FeCl₂ and an ammonium thiocyanate solution (7.5 g in 25 mL water) were added to the sample solution, followed by shaking. The mixture was incubated in a dark room for 20 min. The absorbance of the resulting reaction mixture was measured at 510 nm using an EON microplate spectrophotometer (BioTek Instruments). Hydroperoxide content was calculated using a standard curve (R² = 0.9932) constructed with cumene hydroperoxide as the standard.

MDA content was quantified using the thiobarbituric acid (TBA) method (Pegg et al., 2004). 200 µL of the sample extract were diluted with 2.3 mL of butanol and mixed with 2.5 mL of a 50-fold diluted TBA solution (in butanol). The mixture was placed in a water bath (Jeio Tech) at 95 °C for 2 h to allow the formation of a complex between MDA and TBA. After the reaction, the resulting solution was cooled under running tap water, and the absorbance was measured at 532 nm using an EON microplate spectrophotometer (BioTek Instruments). MDA content was calculated using a standard curve (R² = 0.9528) constructed with TEP.

Statistical analysis

All experiments were conducted in triplicate, and the results are expressed as the mean ± standard deviation of the measurements. Statistical analyses, including analysis of variance (ANOVA), Duncan’s multiple range test, and principal component analysis (PCA), were performed using IBM SPSS Statistics software (ver. 24.0, IBM Corp., Armonk, NY, USA). The t-test was employed to compare the data between the fish balls with and without TP using Microsoft Office Excel (Redmond, WA, USA).

Results and discussion

Influence of cooking method and turmeric powder (TP) addition on the physicochemical properties of fish balls

The moisture content of the fish balls is detailed in Table 1. For the control group, fish balls cooked by boiling displayed the highest moisture content (p < 0.001), exceeding even that of the uncooked fish balls. This elevation in moisture content can be ascribed to the direct contact of the fish balls with water during boiling. Similarly, fish balls cooked by steaming showed no significant difference in moisture content compared to the uncooked group. This suggests that the moisture content in fish balls cooked with water or steam as a heat transfer medium is either maintained or increased. In contrast, for methods other than boiling and steaming, a significant decrease in the moisture content was observed as the cooking temperature rose (p < 0.001). This decline was most evident in high-temperature cooking methods such as oven-roasting (200 °C) and pan-frying (220 °C). When TP was added, there was no significant alteration in the moisture content for the uncooked, boiled, steamed, and microwave-cooked groups compared to the control. However, in the high-temperature pan-frying (p < 0.05) and oven-roasting (p < 0.01) methods, TP-added groups showed higher moisture content. Lee and Surh (2022a) reported that TP possesses higher water-holding capacity than wheat flour at elevated temperatures. Therefore, in cooking methods requiring high temperatures, TP-treated fish balls retained more moisture. Unlike moisture content, the fat content did not show significant differences based on cooking methods or the addition of TP (Table 1).

Table 1.

Moisture and fat content of fish balls depending on the cooking method and addition of turmeric powder (TP)

	Moisture (%)			Fat (%, dry matter)
	Control	With TP	p-valuea	Control	With TP	p-valuea
Uncooked	54.4 ± 0.4B	53.9 ± 0.1B	NS	18.2 ± 1.1	25.0 ± 3.3	NS
Boiling	55.1 ± 0.1A	55.2 ± 0.4A	NS	23.8 ± 3.4	22.3 ± 5.3	NS
Steaming	54.2 ± 0.1B	54.3 ± 0.0B	NS	19.8 ± 4.4	20.2 ± 6.2	NS
Pan-frying	45.9 ± 0.3E	48.0 ± 0.2E	*	20.8 ± 4.8	27.2 ± 2.5	NS
Oven-roasting	50.3 ± 0.0D	51.3 ± 0.1D	**	22.4 ± 1.8	22.5 ± 1.3	NS
Microwave cooking	53.4 ± 0.4C	52.8 ± 0.0C	NS	19.4 ± 1.6	22.3 ± 2.7	NS
p-valuea	***	***		NS	NS

Data was expressed as mean ± standard deviation (n = 2). Different uppercase letters within the same column indicate significant differences at the p-value

^a p-value: ‘*’, p < 0.05; ‘**’ p < 0.01; ‘***’ p < 0.001; ‘NS’ not significant (p > 0.05)

The color properties of the fish balls are presented in Table 2. The addition of TP significantly decreased the lightness of the raw fish balls (p < 0.01) while noticeably increasing their yellowness (p < 0.001). This change is attributable to TP’s inherent color properties, which are less light but more yellow and red than wheat flour (Lee and Surh, 2022a). Cooking led to a significant decrease in both lightness and yellowness for TP-added fish balls (p < 0.001). Among the cooking methods, pan-frying, conducted at the highest temperature (220 °C), yielded the most notable decline. This could result from the low thermal stability of pigments like carotenoids and curcumin present in TP, or from the acceleration of the Maillard browning reaction at higher temperatures (Kocaadam and Şanlier, 2017; Giovanelli et al., 2017). Despite both being high-temperature methods, distinct color attributes were observed between oven-roasted and pan-fried samples. This discrepancy can be linked to the differing heat transfer mechanisms, with oven-roasting utilizing air convection and pan-frying relying on oil conduction (McGee, 2007). During pan-frying, the direct contact of the fish balls with hot oil might have facilitated the release of fat-soluble pigments from TP into the oil, contributing to the variation in color properties. Conversely, microwave cooking had a lesser impact on the color changes in TP-added fish balls, suggesting the retention of most non-polar pigments in the TP. This can be explained by microwave′s inefficiency at inducing vibrations and generating heat in non-polar molecules during cooking (McGee, 2007).

Table 2.

Color property of fish balls depending on the cooking method and addition of turmeric powder (TP)

	Lightness (L)			Redness (a)			Yellowness (b)
	Control	With TP	p-valuea	Control	With TP	p-valuea	Control	With TP	p-valuea
Uncooked	56.7 ± 0.9A	51.0 ± 0.9A	**	8.7 ± 0.1AB	9.22 ± 1.2C	NS	17.4 ± 0.8	44.6 ± 2.8AB	***
Boiling	48.9 ± 0.3C	43.3 ± 0.3BC	***	7.6 ± 0.7B	10.0 ± 0.5BC	**	15.9 ± 1.1	40.2 ± 1.1C	***
Steaming	53.2 ± 2.0B	49.9 ± 1.1A	NS	7.2 ± 0.7B	7.0 ± 0.7D	NS	15.1 ± 1.5	47.1 ± 2.8A	***
Pan-frying	39.6 ± 0.5D	33.9 ± 0.9D	**	8.2 ± 0.5AB	9.7 ± 0.2BC	*	15.9 ± 0.7	19.0 ± 0.9E	**
Oven-roasting	49.2 ± 0.6C	42.0 ± 0.2C	***	7.9 ± 0.6B	12.4 ± 0.1A	***	17.7 ± 0.8	36.2 ± 0.9D	***
Microwave cooking	54.0 ± 0.7B	45.4 ± 2.4B	**	9.4 ± 1.4A	10.9 ± 0.9B	NS	17.7 ± 1.8	41.1 ± 2.6BC	***
p-valuea	***	***		*	***		NS	***

Data was expressed as mean ± standard deviation (n = 3). Different uppercase letters within the same column indicate significant differences at the p-value

^a p-value: ‘*’, p < 0.05; ‘**’ p < 0.01; ‘***’ p < 0.001; ‘NS’ not significant (p > 0.05)

Influence of cooking method and TP addition on the antioxidant capacity of fish balls

In the absence of TP, the total reducing capacity (TRC) of cooked fish balls was not significantly different from that of uncooked fish balls (Fig. 1A). However, fish balls cooked by pan-frying and oven-roasting displayed a somewhat lower TRC compared to other cooked groups. This was previously explained by Serpen et al. (2012), as a result of heat denaturation exposing reactive protein sites, followed by degradation of endogenous antioxidants in the cooked fish meat. Adding TP significantly elevated the TRC of fish balls (p < 0.001), and this increase persisted even after cooking. Notably, TP-added fish balls cooked by steaming and microwave showed a higher TRC than the uncooked group (p < 0.001). This trend may be attributed to a combination of factors. First, a higher concentration of antioxidants remained in the fish balls cooked by steaming and microwaving, as evidenced by their color closely resembling that of the uncooked group (Table 2). This color similarity strongly suggests the retention of antioxidant pigments without decomposition. Second, the heat intensity and duration of the cooking method may also be factors. Steaming took the longest time (9 min), while microwave cooking was the quickest (1 min). Serpen et al. (2012) observed that the total antioxidant capacity initially increased during the first few minutes of cooking due to the release of endogenous antioxidants from destroyed cell membranes. In the final heating stage, further increases occurred from the formation of Maillard reaction products (MRPs), which have antioxidant properties. The addition of TP enhanced the antioxidant response triggered by heat, particularly in methods like steaming and microwave cooking, consequently raising the TRC in TP-added fish balls compared to the uncooked group. Specifically, without TP, the TRC of fish balls cooked by steaming and microwave heating did not significantly differ from that of the uncooked ones. However, with TP, the TRC of steamed and microwaved fish balls significantly increased. This enhancement may be attributed to the synergistic interaction between endogenous antioxidants released by heat in fish balls or MRPs formed during extended heat and the intrinsic antioxidants in TP. This interaction underscores the interplay between cooking methods and TP addition. The TRC of TP-added fish balls cooked by pan-frying and oven-roasting was comparable to that of uncooked fish balls despite their reduced color intensity. This implies that the loss of endogenous or TP-derived antioxidants during cooking was counterbalanced by the accelerated formation of MRPs at higher temperatures, supported by their lower lightness values (Table 2).

Fig. 1.

Total reducing capacity (A) and DPPH radical scavenging activity of fish balls in the absence (B) and presence (C) of turmeric powder (TP) depending on cooking method. Bars with different uppercase letters indicate significant differences depending on the cooking method at a significance level of p < 0.001. ** and *** indicate that the values were significantly different depending on the addition of TP, with a significance level of p < 0.01 and p < 0.001, respectively

In the absence of TP, the DPPH radical scavenging activity of cooked fish balls initially exceeded that of the uncooked group when reacting with DPPH radicals. However, no significant differences were observed after an extended 5-h reaction (Fig. 2B). With TP addition, the cooked fish balls demonstrated lower DPPH radical scavenging activity than uncooked fish balls (Fig. 2C). The decline can be attributed to the decomposition of curcuminoids, abundantly present in TP. Prathapan et al. (2009), in their study on the antioxidant activity of curcuminoids under different heat treatment conditions, reported a significant decrease in the antioxidant activity of TP in a 100 °C environment. These results on DPPH radical scavenging activity contrast with the TRC findings, particularly for fish balls prepared by steaming. This discrepancy could be due to the varying sensitivities of the assays. The DPPH radical is hydrophobic, whereas the MRPs formed during cooking are hydrophilic (Serpen et al., 2012). The reduced DPPH radical scavenging activity of steamed fish balls is mainly attributable to the higher contribution of MRPs to TRC rather than to DPPH radical scavenging activity. Cooking methods that take longer, such as steaming, result in a greater presence of MRPs, while shorter methods like microwaving contribute less.

Fig. 2.

Changes in the content of hydroperoxides of fish balls with or without TP during 4-month storage. Graphs A to F present results for different cooking methods: uncooked (A), boiling (B), steaming (C), pan-frying (D), oven-roasting (E), and microwave cooking (F). The same style of symbols, with different uppercase or lowercase letters, indicates significant differences in the storage period at p < 0.05. *, **, and *** indicated the values at the same storage period were significantly different at p < 0.05, p < 0.01, and p < 0.001, respectively

Influence of cooking method and TP addition on the oxidative stability of fish balls

The hydroperoxide content in the fish balls was monitored over a 4-month storage period (Fig. 2). Without TP, the hydroperoxide content in the uncooked fish balls increased 20-fold after 4 months (p < 0.001). However, the inclusion of TP led to a significant reduction in hydroperoxide content (p < 0.05) (Fig. 2A). This reduction can be attributed to the enhanced TRC and DPPH radical scavenging activity of TP, as shown in Fig. 1. These findings suggest that TP acts as an effective antioxidant, inhibiting hydroperoxide formation in the fish balls over time.

For control fish balls without TP, the cooking process significantly impacted hydroperoxide content. Immediately after cooking, these levels were higher than in uncooked fish balls, with higher cooking temperatures leading to even higher levels. However, this trend reversed after a 4-month storage period, hydroperoxide content in cooked fish balls was lower than in the uncooked ones. Yet, fish balls prepared via steaming and microwave cooking exhibited a more pronounced increase in hydroperoxide content compared to other cooking methods (Fig. 2B and F). This may be partially due to their lower DPPH radical scavenging activity (Fig. 1B).

The addition of TP led to a reduction in hydroperoxide content over the 4-month storage period, most notably in fish balls prepared by steaming and microwave cooking (Fig. 2C and F). In contrast, TP had little impact on fish balls prepared by boiling, pan-frying, and oven-roasting (Fig. 2B, D, and E). This suggests that these cooking methods themselves had a more significant impact on reducing hydroperoxide levels than did the addition of TP. The observed differences in oxidative stability depending on the cooking method could be attributed to the unique environments or food matrices that each method creates, affecting lipid oxidation reactions differently. Factors such as cooking temperature, duration, and the distribution of heat and antioxidants during cooking influence the formation and degradation of hydroperoxides, ultimately affecting the overall lipid oxidation stability of the fish balls.

MDA content, which forms upon hydroperoxide decomposition, exhibited a similar pattern to that of hydroperoxide content (Fig. 3). In the absence of TP, MDA levels in uncooked fish balls significantly increased over the 4-month storage period (p < 0.01). However, this rapid increase was mitigated when the fish balls were cooked prior to storage, particularly when using boiling, pan-frying, or oven-roasting methods. TP addition also led to a significant decrease in MDA levels during storage for uncooked fish balls and those cooked by steaming and microwave cooking (Fig. 3A, C, and F). However, the impact of TP was minimal on fish balls prepared by boiling, pan-frying, and oven-roasting (Fig. 3B, D, and E), mirroring observations related to hydroperoxides (Fig. 2).

Fig. 3.

Changes in the content of malondialdehyde of fish balls with or without TP during 4-month storage. Graphs A to F present results for different cooking methods: uncooked (A), boiling (B), steaming (C), pan-frying (D), oven-roasting (E), and microwave cooking (F). Please refer to the legend of Fig. 2 for more details

These findings suggest that while TP does play a role in reducing hydroperoxide and MDA formation, especially in uncooked fish balls and those cooked via steaming or microwave, cooking methods like boiling, pan-frying, and oven-roasting have a more substantial impact on these oxidative markers, often overshadowing the effects of TP addition.

Influence of cooking method and TP addition on the overall characteristics of fish balls

Principal component analysis (PCA) was conducted to classify fish ball samples based on their cooking methods (Fig. 4). In the absence of TP, the total variation in the characteristics of the fish ball samples, captured by the first two significant dimensions, was 69.4%. Here, PC1 and PC2 accounted for 41.6 and 27.7%, respectively (Fig. 4A). The PCA plot shows that the cooking process induced changes in the physicochemical properties and oxidative stability of the fish balls. Uncooked fish balls were situated in the fourth quadrant, while cooked fish balls were positioned in different quadrants and clearly separated from the uncooked samples. Specifically, steaming and microwave cooking were closely related, whereas boiling, pan-frying, and oven-roasting were positioned in close proximity to each other. This trend was consistent with the results on antioxidant capacity and oxidative stability, where cooking methods of shorter or longer durations produced similar responses.

Fig. 4.

Principal component analyses of uncooked and cooked fish balls using water and fat contents, color property, total reducing capacity, DPPH radical scavenging activity, and hydroperoxides and malondialdehyde contents of fish balls in the absence (A) and presence (B) of TP

In the presence of TP, PC1 and PC2 explained 34.0 and 32.9% of the total variation in fish ball samples, respectively (Fig. 4B). The addition of TP led to a notable shift of the uncooked fish balls from the fourth quadrant to the second quadrant, indicating a significant alteration in the fish balls′ characteristics due to the addition of TP. For the cooked fish balls, the positions associated with each cooking method in the PCA plot changed upon the addition of TP. The classification of fish balls with TP, based on cooking methods, closely resembled that of the fish balls without TP. However, the distances in the PCA dimensions between steaming and microwave cooking, as well as between boiling, pan-frying, and oven-roasting, were relatively greater for the fish balls with TP than for those without. This observation suggests that both the cooking method and the addition of TP play crucial roles in altering the physicochemical properties and oxidative stability of fish balls. It also indicates that certain cooking methods may have similar effects, whereas the addition of TP may interact with the cooking methods to influence these changes in distinct ways.

The result of the present study revealed an interplay between cooking and TP addition in fish balls. The addition of TP led to a remarkable enhancement in the total reducing capacity of fish balls, an effect that persisted even after cooking. Consequently, TP reduced the hydroperoxide and MDA contents in the fish balls during storage. However, the effects of TP addition were less pronounced in fish balls cooked by boiling, pan-frying, and oven-roasting, likely due to the dominant impact of these cooking methods. PCA revealed distinctive clusters based on cooking method and TP addition, highlighting a complex interplay between these two factors. These findings offer valuable insights into extending the shelf life and improving the nutritional quality of fish-based products through the use of antioxidants and optimized cooking techniques. Further research is required to understand the specific mechanisms underlying these differences and to optimize cooking methods for achieving oxidative stability in fish balls.

Declarations

Conflict of interest

The authors declare no conflicts of interest.