Effectiveness of Toothpony (Gazza minuta) protein hydrolysate on reducing oil uptake upon deep-frying

F U L L P A P E R Effectiveness of Toothpony (Gazza minuta) protein hydrolysate on reducing oil uptake upon deep-frying 1* Zainol, M.K., Tan, R.C., Mohd Zin, Z., Ahmad, A. and Danish-Daniel, M. Faculty of Fisheries and Food Science, Universiti Malaysia Terengganu, 21030, Kuala Nerus, Terengganu Faculty of Science and Marine Environment, Universiti Malaysia Terengganu, 21030, Kuala Nerus, Terengganu Institute of Tropical Biodiversity and Sustainable Development, Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia Institute of Marine Biotechnology, Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia


Introduction
Malaysia produced 1.75 tons of marine fishing in 2008, which accounted for 80% of total fish production, and it was estimated that 30% of caught fish were discarded as trash fish (DoF Annual Fisheries Statistics, 2008). Toothpony (Gazza minuta) (Figure 1) was reported to be one of the main trash fish harvested (Hasan and Halwart, 2009). The toothpony fish were not popular for human consumption due to its small size and low meat content. It is commonly being used as fish meal (Klomklao et al., 2013). Surprisingly it has a high protein content (19.3%) and could be suitable for the production of fish protein hydrolysate (FPH). FPH from fish proteins can be considered good dietary supplements such as bioactive compounds and can be easily absorbed and used for different metabolic activities (Nesse et al., 2011). The by-product produced by the fish processing industry contains a good amount of protein-rich material and can be processed into hydrolysates of fish protein (FPH) (Chalamaiah et al., 2012). FPH can be used as active ingredients in various foods because they have numerous essential and special properties such as water holding capacity, emulsifying properties, foaming properties and oil binding capacity (Chalamaiah et al., 2010).
Deep-fat frying is one of the oldest and most widely used methods of cooking to make delicious and crispy foods (Pedreschi and Moyano, 2005). It involves the immersing of food in edible oils heated above the water boiling point and creates the desired quality characteristics namely taste, texture, odours and colour (Pedreschi, 2012). Coating products using deep-fat frying batters is a common and popular process used in the preparation of deep-fried products. Batters should not only add value to the product by controlling the loss of moisture and thus the uptake of oil during frying (Dogan et al., 2005). The efficacy of a batter as a barrier to the absorption of lipids by food is determined by its mechanical and barrier properties, which depend on its composition and microstructure, and the substrate characteristics (García et al., 2008). A typical batter formula consists of critical ingredients of wheat flour, maize flour and a leavening agent. Proteins can be added to the batter as functional ingredients by forming lipid barriers (Kulp and Loewe, 2011).
The recent trend towards the reduction of fat content in fried foods contributes to the development of low-fat products using different formulations of batter ingredients (Dogan et al., 2005). FPH has been successfully tested for incorporation into various food systems such as cereal products, fish and meat products, desserts and crackers (Kristinsson and Rasco, 2000). This phenomenon has attracted the attention of the food industry to use the low oil-binding FPH as a potential application to minimize the consumption of deep-fried food (He et al., 2015). Nevertheless, there is little knowledge about the introduction of FPH into the batter and its effect on the absorption of oil in batters. Most of the barriers used in the commercial production of fried foods are protein-based (Hau et al., 2018). The aim of this research was to develop a new concept product containing Toothpony fish FPH in the batter and to determine the physicochemical properties of the fried product in order to visualize it as a potential good source of protein that could later be transformed into a more functional product.

Materials and methods
Toothpony (Gazza minuta) were purchased from Jeti Pulau Kambing, Kuala Terengganu, Terengganu, Malaysia. The guts and head of Toothpony were removed and the sample was cleaned. The sample was then homogenized using food processor (Panasonic, Malaysia). Enzymatic extraction was carried out using the modified method of Hau et al. (2018). A total of 50 g of fish meat was deactivated at 90°C for 10 mins. The sample was then mixed with 100 mL of buffer at pH 8. A total of 50 g of homogenized sample was mixed with 500 mL of water in a 1000 mL beaker and heated using hot plate (Favorit, Malaysia) at 90 ο C for 10 mins. Next, 50 mL of 1M potassium phosphate buffer (pH 7.5) was added to the fish (ratio of 1:1 between fish and buffer) and the mixture was mixed thoroughly. The pH of the mixture was set at 7.5. After hydrolysing the protein for 3 hrs, the mixture was placed on hot plate which operating at 90 ο C for 5 mins to inactivate the enzymes. The mixture was then cooled and centrifuged at 10000 rpm for 20 mins. Freeze drying was conducted using a Labconco Freeze Dryer -Stoppering tray (USA) operated at -54 ο C while the vacuum was set at 0.25 mbar.

Proximate composition of Toothpony fish
The moisture, ash, fat and crude protein content of the Toothpony fish were determined according to AOAC (2012) standard procedures. Analyses were carried out in triplicates and calculated on dry weight basis using AOAC methods (AOAC, 2012).

Determination of degree of hydrolysis
Degree of hydrolysis of Toothpony fish lyophilized hydrolysate was determined using trichloroacetic acid (TCA) method, which has been described by Hoyle and Merritt (1994). A total of 2 g of lyophilized hydrolysate was needed to carry out this analysis. One gram of lyophilized FPH powder was determined through Kjedahl method (AOAC, 2012). As for the 10% TCA soluble nitrogen determination, 1 g of lyophilized Toothpony fish protein hydrolysate powder was added into 10 mL distilled water, respectively. Then, 10 mL of 20% TCA was added into the sample and mixed by vortex. The sample was left to stand for 30 mins, to allow protein precipitation. The sample was then centrifuged at 8000 rpm for 5 mins (High Speed centrifuges 1580R, Cryozen Co, Ltd, Korea). Protein content in the collected supernatant by-product using Kjedahl method (AOAC, 2012). The protein content was calculated using the formula below: The degree of hydrolysis was calculated by using the formula below:

Determination of FPH molecular weight
The Toothpony FPH were analysed for sodium dodecyl sulfate polyacrylamide electrophoresis (SDS-PAGE) analysis based on the method Zainol et al. (2014) with some modification. SDS-PAGE gel used consisted of 2 layers of acrylamide gels namely resolving gel and stacking gel. Resolving gel layer consisted of 950 µL of 40% bisacrylamide: acrylamide (19:1), 1250 µl of 3M Tris 5% SDS, 1120 µL RO-water, 400 mL Glycerol, 2 µL TEMED and 19 µL of ammonium persulphate. The Toothpony FPH sample was diluted in 5% (w/v) and mixed with 1:1 (v/v) ratio loading buffer. The mixture was cooked in 90°C water bath for 10 mins and cooled immediately prior to loading. Sample and protein standard volumes of 25μL are loaded into individual wells and operated using a discontinuous tristricine buffer with 25 mA/gel constant current and 100V constant voltage for 1 hr. After electrophoresis, proteins were visualized by 0.1% (w/v) Coomassie blue G250 staining and destaining by soaking in several changes of 40% (v/v) methanol and 10% (v/v) acetic acid until a significant clear bands were obtained. Broad protein markers (11 to 245 kDa) were used for molecular weight comparison.

Determination of functional groups
The functional groups of Toothpony FPH were determined using FTIR Transmission (Nicolet iS10, Thermo Scientific, US) in which 4 mg of Toothpony fish protein hydrolysate powder was mixed with 200 mg of KBr (1:50). The mixture was homogenized using agate mortar and pestle and then it was pressed into pellet (1-2 mm thick films) with a 15-ton hydraulic press. The FTIR spectra were obtained from wavenumber of 600 to 4000 cm -1 during 64 scans with 2 cm -1 resolution. The resulting spectrum represents the molecular absorption and transmission which then create a molecular fingerprint of the sample (Durga Maha Lakshmi et al., 2014).

Determination of oil uptake
Five batter mix was formulated (Table 1) with the addition of four different amount of fish protein hydrolysate (0%, 2%, 4%, 6%, 8%) into each formulation and the other one batter formulation was kept constant in which no FPH was added in the formulation. The batters were used to coat squids (3 cm x 3 cm). The coated squids were then deep-fried in palm oil for 3 mins at 180 ο C, was drained and cooled. Then, the batter crust was peeled off from the squids and the oil was extracted using Soxhlet extraction method.
Oil-uptake was measured with a slight modification by a method described by Wasswa et al. (2007). In a 50 mL centrifuge tube, a total of 0.5 g of Toothpony FPH powder sample was added to 9 g of palm oil, mixed with a vortex mixer for 1 min, and then centrifuged at room temperature at 2000 x g for 30 mins. The weight of the extracted oil was determined from the hydrolysate. The difference between the oil weights derived from the hydrolysate and the initial palm oil 9 g was the oil weight consumed by a sample of 1 g. The oil consumption was measured as the quantity of oil consumed by the samples.

Texture profile analysis
Texture profile analysis was performed using Dogan et al. (2005) method with a slight modification. Texture of the samples' crust was analysed in terms of hardness and crispness. These were measured in the fried samples using texture analyser (TA.XT. plus Stable Micro System Ltd., US). A load cell of 50 N was used. A conical probe was attached to the instrument for the penetration test. The instrument was set to a speed of 55 mm min -1 for 25% penetration of a conical probe into the fried sample.

Colour profile analysis
The crust colour of fried squid was determined by using colorimeter (CR-300 Chroma Meter, Konica Minolta, Japan), working with cool white light (D65) and 2° to know Hunter L* (lightness), a* (redness), and b* (yellowness) values. The instrument was calibrated using white tile provided. Instrument was directly put on the surface of each sample at three different points. L*, a* and b* values of each sample was recorded.

Sensory evaluation
A fresh batch of fried squid was used for sensory studies. The fried squid was subjected to sensory evaluation for colour, odour, oiliness, crispness, and overall acceptability. The sensory evaluation was carried out by thirty panellists using 7-point Hedonic scale. The mean score for each attribute was reported.

Proximate composition of Toothpony fish
Chemical analysis was carried out to determine the chemical composition (protein, fat, moisture) of Toothpony fish raw sample ( Table 2). The raw Toothpony fish sample contained 16.08% of protein, 4.41% of fat and 77.51% of moisture. However, the chemical compositions recorded in this study were slightly lower than the data reported by Klomklao et al. (2013). They reported that Toothpony fish muscle contained 83.32% of protein (dried weight basis), 9.32% of fat and 80.01% of moisture. The difference in nutritional composition might be due to environmental factors. Nutrient profiles of fish were reported to be varied with the environment and it was more unmanageable in the wild. The nutrient profiles of wild fish could be even more variable than those of their farmer counterparts (Perschbacher and Stickney, 2017). Table 3 shows the Toothpony fish protein hydrolysate (FPH) contained 38.02% of protein. It was expected that Toothpony FPH to exhibit a higher value of protein than that of in raw sample (16.08%). Klomklao et al. (2013) have reported higher protein content (89.02%) in Toothpony FPH. This could be due to the fact that Klomklao et al. (2013) have used viscera extract from hybrid catfish to hydrolyse Toothpony FPH instead of Alcalase enzyme. However, the result was in contrary with findings reported by Venuste et al. (2013), who reported that Alcalase treated hydrolysates were significantly higher protein content because of high solubility.

Determination of degree of hydrolysis
The high protein content of FPH demonstrates its potential use as protein supplements for human nutrition (Chalamaiah et al., 2012).
Degree of hydrolysis (DH) of Toothpony FPH was determined using trichloroacetic acid (TCA) method. DH of Toothpony FPH was illustrated in Table 3. DH of Toothpony FPH was 98.02%. The high value of DH was probably due to the Alcalase enzyme used in the extraction of Toothpony FPH. Alkaline proteases have reportedly a greater capability to solubilize fish protein compared to neutral and acidic proteases (Sugiyama et al., 1991). Alcalase, which is one of the alkaline proteases, is prominently used in the hydrolysis of proteins from fish due to its high DH in a relatively short time (Kristinsson and Rasco, 2000). However, DH was reported to have an effect on the physicochemical properties of FPH (Harun et al., 2017), which could lead to a low oil uptake. Figure 2 shows the characterisation by SDS-PAGE of Toothpony FPH's molecular weights (MW). The Toothpony FPH exhibited molecular weight lower than 175 kDa which and gelatin from catfish skin (<200 kDa) (Mahmoodani et al., 2014). Differences are found between Toothpony FPH molecular weight and catfish skin gelatin may be due to the various species and part of the fish. The hydrolysate showed multiple bands that could result from larger proteins in the raw material that the enzyme does not fully hydrolyze (Souissi et al., 2007). Several studies have shown Alcalase enzyme's ability to produce peptides of low molecular weight (MW) at a high degree of hydrolysis (Liaset et al. 2000). According to Roslan et al. (2014), alcalase has shown protein cleavage, leading to the production of small peptides. Hence, the presence of low molecular weight bands in Toothpony FPH in the present study could result in the production of FPH with useful chemical properties. Mean value with same superscript letter within the same showed no significant different (p˂0.05).

Functional groups
FTIR transmission spectrum of Toothpony fish protein hydrolysate powder is demonstrated in Figure 3. Amide I and amide II bands of Toothpony FPH appeared at around 1654 and 1535cm -1 , respectively. Amide I and amide II bands at 1700-1600 and 1560-1500 cm -1 were reported by Benjakul and Morrisey (2009). The amide I vibration mode is primarily a C=O stretching vibration coupled to minor contributions from the CN stretch, CCN deformation and in-plane NH bending modes (Lavialle et al., 1982;Bandekar, 1992). The amide I vibration is hardly affected by the nature of the side chain. It depends however on the secondary structure of the backbone and is, therefore, the amide vibration that is most commonly used for secondary structure analysis (Barth, 2007). Amide II derives mainly from in-plane NH bending and CN stretching vibration and shows less protein conformational sensitivity compared with amide I (Kong and Yu, 2007). The amide II band is generally considered to be much more sensitive to hydration than to secondary structure change (Wellner et al., 1996). Amide III band of protein hydrolysate was detected at 1264 cm -1 which was associated with loss of triple-helix state of the molecules and transformation of α-helical to random coil structure due to denaturation of protein (Muyonga et al., 2004). The amide III vibration mode is the in-phase combination of the NH bending and the CN stretching vibration with small contributions from the CO in plane bending and the CC stretching vibration. In polypeptides, the composition of this mode is more complex, since it depends on side-chain structure and since NH bending contributes to several modes in the 1400-to 1200-cm −1 region (Barth, 2007). Table 3 displays the oil binding capacity information for Toothpony fish protein hydrolysate, calculated as 1.9 g oil / g protein. The low oil binding capacity obtained in Toothpony FPH can be explained by its low MW (10-25kDa) as shown in Table 3. He et al. (2016) reported that oil-binding capacity of FPH was affected by its MW properties and distributions. FPH with the molecular weight that is smaller than 10 kDa possesses the lowest oil-binding capacity compared to FPH with higher molecular weight. An interesting study that investigated the oil binding capacity of Codfish and Bluefish processing co-products protein hydrolysates found that 1.2 g oil/ g protein and 1.9 g oil/ g protein, respectively (Slizyte et al., 2005;Geirsdottir et al., 2011), which were in agreement with this study. However, the data collected in this study was lower than Yellowtail Kingfish FPH produced by Alcalase extraction (12.14 g oil/ g FPH) (He et al., 2013).

Oil uptake
The mechanism for oil binding capacity is attributed to the combination of physical entrapment of oil and the hydrophobicity of the material (He et al., 2013). Hydrophobicity of fish protein hydrolysates develops because hydrolysis cleaves the protein chain so more internal hydrophobic groups were exposed (Kristinsson and Rasco, 2000). However, the excessive hydrolysation compromises the integrity of the protein structure and results in the degradation of the protein network formed to entrap oil (He et al., 2013). The low oil binding capacity of Toothpony FPH was in agreement with its high DH (98.02%) as shown in Table 3 Oil uptake of deep-fat-fried squid batter crust incorporated with 0%, 2%, 4%, 6%, 8% Toothpony FPH were also shown in Table 3. The fat uptake of deep-fatfried squid batter crust incorporated with 4% Toothpony FPH was significantly the lowest (15.05%) among the samples. This indicated that 4% of Toothpony FPH is the optimum amount to be incorporated in batter in order to reduce oil uptake significantly. Fat uptake of batter crust incorporated with 2%, 6% and 8% of Toothpony FPH were negligible when compared with the fat content of the control sample (27.22%). However, the fat content of the control sample was found to be the highest than that of other samples. This was probably due to incorporation of Toothpony FPH in the batter helps reduce oil uptake in deep-fried food products by its low oil binding capacity property (1.9 g oil/g protein). Lower oil-binding capacity was generally associated with lower fat uptake. He et al. (2015) also observed that the fat content of deep -fried fish with batter coating was reduced using Yellowtail Kingfish FPH. Fat uptake reduction of deepfried food using other protein-based ingredients in batter coating has also been reported before. Albert and Mittal (2002), for instance, found that batter coating systems of soy protein could reduce the fat uptake of deep-fried cereal products by up to 99.8%. Garcia et al. (2002) compared the effects of various cellulose derivatives and additionally the sorbitol effect on selected cellulose derivatives in French fries and found that a fat uptake reduction of 41% could be achieved.

Texture profile analysis
Hardness (kg) and crispness (kg.sec) of deep-fatfried squid were demonstrated in Table 3. Texture of battered products involves many characteristics, foremost of which is hardness, which describes the textural attributes of the substrate. Deep-fat-fried squid incorporated with 2%, 4%, 6%, 8% of Toothpony FPH were harder than deep-fat-fried squid incorporated with 0% of Toothpony FPH. This could be due to incorporation of Toothpony FPH has increased hardness of fried samples. The final result was the increase in the hardness value. Deep-fat-fried squid without Toothpony FPH showed the least crispness (1.28 kg.sec) among all samples. Incorporation of Toothpony FPH could be the cause of increased crispness in deep-fat-fried squid batter crust. As the protein level increased, there was a gradual increase in roughness of texture and brittleness of the fried coating. This rough, brittle texture translated into a high crispness value (Kulp and Loewe, 2011). Mohamed et al. (1998) observed that the addition of protein, ovalbumin to a batter improved crispness. However, excess protein in batter may reduce crispness by emulsifying more oil and water into the fried product. This is in agreement with the decrease of crispness in sample with 8% of Toothpony FPH (3.16 kg.sec) compared to sample with 6% of Toothpony FPH (4.09 kg.sec). Figure 4 exhibits the effect of different amount of Toothpony FPH on colour development of deep-fat-fried squid was examined in terms of Hunter L* a* b* values. Different amount of Toothpony FPH was added into batter formulation to coat deep-fat-fried squid better. Deep-fat-fried squid developed a yellowish-brown colour when they were fried. The deep-fat-fried squid incorporated with 0% Toothpony FPH has significantly lower a* value compared to deep-fat-fried squid incorporated with 2%, 4%, 6% and 8% of Toothpony FPH. It was assumed that the presence of FPH in batter has caused the significantly higher a* values in deep-fatfried squid. A light golden tone was the benchmark colour for determining the end of the final frying process (Ansarifar et al., 2012). The non-significant trend of b* values among samples might be due to the consistent frying time (3 mins) carried out in this study. Ansarifar et al. (2012) have reported that b* value of deep-fat-fried crust decreased with frying time.

Sensory evaluation
The results of sensory evaluation of deep-fat-fried squid incorporated with 0%, 2%, 4%, 6%, and 8% of Toothpony FPH is depicted in Table 4. The mean score for acceptability of colour for deep-fat-fried squid incorporated with 2% of Toothpony FPH was significantly the lowest (4.13) among all samples. However, in colour analysis, deep-fat-fried squid incorporated with 2% of Toothpony FPH showed no significant difference with other samples as shown in  Table 4. The colour acceptability was increased as the amount of Toothpony FPH incorporated in deep-fat-fried squid batter crust increased. This could be explained by higher Toothpony FPH amount has increased the affinity towards colour acceptability. Golden colour, between yellow and brown developed by coated foods when they are fried, is a characteristic that consumers normally consider very attractive (Ansarifar, 2012). The acceptability of oiliness for deep-fat-fried squid incorporated with 6% of Toothpony FPH was the lowest (3.77) among all samples. Panellists perceived that high amount of Toothpony FPH (6%) in deep-fat-fried squid batter have caused undesirable oiliness. The amount of oil uptake has generally been related to the amount of moisture lost (Basuny et al., 2009). Protein can hydrate/ hold water and possess emulsifying ability due to their hydrophilic and lipophilic side chains (Mohamed et al., 1998). Food product that contains higher initial moisture content results in an increased oil uptake, since water evaporation defines the volume of the oil reservoir through the empty spaces, as it leaves the structure (Basuny et al., 2009).

Conclusion
The outcome of this study shows the potential usage of protein hydrolysate from Toothpony edible parts for the production batter formulation which is efficient in the reduction of oil uptake. The protein hydrolysate contained 16.08% of protein, 4.41% of fat and 77.51% of moisture. The deep-fat-fried squid incorporated with 0% of Toothpony FPH showed the least hardness and crispness compared to deep-fat-fried squid incorporated with 2%, 6% and 8% of Toothpony FPH. Oil uptake of batter crust incorporated with 2%, 6% and 8% of Toothpony FPH showed no significant differences when compared with the fat content of the control sample (27.22%). In sensory evaluation, deep-fat-fried squid incorporated with 8% of Toothpony FPH showed the highest acceptance in all attributes, which were colour, crispness, oiliness, taste and overall acceptability.