Biochemical and physicochemical analysis of fish protein isolate recovered from red snapper (Lutjanus sp.) by-product using isoelectric solubilization/precipitation method

The effect of acid- and alkali-process on biochemical and physicochemical characteristics of fish protein isolate from red snapper (Lutjanus sp) by-product was evaluated. Protein recovered by alkali process (16.79%) was higher compared to acid process (13.75%). Reduction of lipid content and total volatile basic nitrogen (TVB-N) exhibited in both treatments indicated both process improved fish protein isolate recovered from red snapper by-product. In addition, the increasing of water holding capacity and oil binding capacity were observed. However, high peroxide value of fish protein isolate was showed in both treatment. This finding indicated that acid and alkali process can be used as a useful method to recover proteins from red snapper by-product. Alkali process gave a protein isolate with better overall quality compared to acid process.


Introduction
Consumer demand forfunctional food has beenpredicted to increase to a value of USD100 million in 2013 and has showed a positive trend from year to year [1]. Seafood-based functional food isdeveloped from various sources such as krill, squid, and seaweed. Red snapper (Lutjanus sp.) is important for fillet production in Indonesia for its high quality and the demand for the fillet. During the filleting process, a large amount of low-value by-product is produced, and it is used as fish mealfor animal feed. Head byproduct fromfish processing is about 20 % of the fish weight [2], while the total by-product consisting of head, frame flesh and scale comprises 60 % of the fish weight [3]. Due to the high value offish head, many restaurants in Indonesia use red snapper head as traditional food. Meanwhile, the frame, skin, and scale are discharged or used as fish meal, plant fertilizer or human value-added foods.
Fish frame, scale, and skin contain valuable proteins such as myofibrillar proteins, stromata,sarcoplasmic proteins, and others that can be recovered forhuman consumption [4]. There have been some reports about the development of functional proteins and peptone using hydrolysis method from red snapper by-product forsilage or fish feed [5,6]. Currently, protein recovery with pH-shift method or isoelectric solubilization/presipitationhas beenstudied. Protein recovery of bigeye snapper (Priacanthus spp.) with pH-shift method was evaluated to have a yield of 30% and acceptable characteristics [4].
Isoelectric solubilization/precipitation (ISP) processing is selective, pH-induced water solubility and removal of lipid, bone, scales and skin [7]. This method was proposed by Hultin and Kelleher [8]. The basic concept of ISP is about protein side-chains that have different electrostatic charges depending on the pH condition. Atcertain pH levels, fish protein will solubilize (on), and atisoelectric point(pI), fish protein will precipitate (off) [9]. The pH for protein solubilizationislow (acid) at around 2-3.5 or high (alkali) at around 11-13 [10].
Fish protein isolate produced with ISP can be applied as a raw material in the making of Frankfurtertype fish sausages [11], functional fish protein gel [12], or fish ball [13]. The yield of the fish protein isolate rangesat 50-87 % [4]. Therefore, ISP process is an effective way for recovering proteins from the by-product of fish processing industry. The aim of this study was to isolate and characterize the fish protein isolate recovered from red snapper (Lutjanus sp.) filleting industry by-product.

2.1.Materials
The fish by-product was purchased from a filleting factory located in Sidoarjo, East Java, Indonesia. The sample was wrapped in apolyethylene bag under frozen condition (-18 o C) and transported to the laboratory. The sample was directly stored in the laboratory freezer (-18 o C) prior to further process. The hydrochloric acid and natrium hydroxide were of p.a.quality.

2.2.Fish by-product preparation
The frozen fish by-product sample was thawed at 4 o C overnight prior to processing. The skin, bone, flesh, and frame of thethawed sample were then separated from each other. The frame and flesh were homogenized with laboratory food processor (Bosch, Germany) three times for 3 minutes. The homogenized sample was stored in a polyethylene bag and frozen until further process.

2.3.Protein recovery with isoelectric solubilization/precipitation (ISP)
The thawed red snapper by-product (RSB) homogenate was processed to recover protein inacid and alkaline process. The solubilization was conducted at pH 2.5 and 11.5 for acid process and alkaline process,respectively. The recovery process followed Gehring et al. [3] method with a modification in thecentrifugation speed. A total of 100 g of RSB was mixed with 900 mL of distilled water (4 o C) and homogenised for 2x30 seconds using Food Processor (Bosch, Germany). The pH of the homogenate was then adjusted to either 2.5 or 11.5 using 1N HCl or 1N NaOH with constant stirring (400 rpm). The pH was monitored with calibrated pH meter (Labortechnik, Germany). The homogenate, the pH of which has been adjusted, was centrifuged at 6.000 rpm and a temperature of 4 o C for 25 minutes. The supernatant was collected by filtering usingthree layers of cotton sheet. The pH of the filtrate was then adjusted to 5.5 using 1N NaOH or 1N HCl. Second centrifugation was performed to separate the pellet (recovered protein/fish protein isolate). The recovered protein was then collected in apolyethylene bag and weighed.

Proximate analysis
The proximate analysis of fish protein isolate was performed following AOAC (2002). The moisture, protein, fat, and ash content were evaluated. The total nitrogen of the fish protein isolate was determined with Kjeldahl method. The total crude protein was measured by multiplying thetotal nitrogen of the sample by 6.26.

Total volatile basic nitrogen (TVB-N)
The TVBN of the fish protein isolate was measured with Conway microdilution method. Briefly, 2 g of sample was added to 8 mL of 4 % (w/v) TCA and homogenized at 10,000 rpm for 2 minutes. The homogenate was centrifuged at 3. 000 rpm for 15 minutes at room temperature. The supernatant as sample (1 mL) was placed in the outer ring of Conway apparatus. The inner ring consisted of 1 % boric acid solution. The reaction was initiated with an addition of 1 mL of K2CO3 to the sample at the outer ring. The Conway unit was closed and incubated at room temperaturefor 16 hours. The solution at the inner ring was then titrated with 0.02 N HCl until the green colour turned pink. The TVB-N value was expressed as mg of nitrogen released/100 gram of sample.

2.6.Peroxide value
The peroxide value was measured using Panpinat and Chaijan [4] method. A total of 2 g of sample was treated with 25 mL of organic solvent (with ratio of chloroform to acetic acid of 2:3). The mixture was shaken and added with 1 mL of saturated KI solution. The mixture was stored in the darkness for 5 minutes, then 75 mL of distilled water was added. A total of 0.5 mL of starch solution (1 % w/v) was added as an indicator. The PV was measured by titrating the iodine liberated from KI with standardized 0.01 N sodium thiosulfate solution. The PV was expressed as milliequivalents of free iodine per 100 g of lipid.

2.7.Water holding capacity
The water holding capacity was measured usingmethod by Panpinat and Chaijan [4]. A total of 100 mg of fish protein isolate was homogenized with 10 mL of distiled water and centrifuged at 10.000 rpm for 30 minutes, then decanted. The difference between the initial weight and the final weight was measured as the water holding capacity of the fish protein isolate.

2.8.Oil binding capacity
The oil binding capacity was determined by adding 10 mL of coconut oil into100 mg of fish protein isolate, and then the mixture was homogenized. The homogenate was then centrifuged at 2.500 rpm for 30 minutes. The free oil was decanted, and the weight difference was stated as oil binding capacity [14] with modification.

2.9.Amino acid profile
The amino acid profile was measured with a HPLC method [15].

Statistical analysis
All treatments except the amino acid profile weretriplicated, and the differences between treatments were analyzed with analysis of variance (ANOVA), while the comparison of means was made by conducting Duncan's multiple-range test for stating the significant difference.

3.1.Protein recoveryof red snapper by-product
The protein recovery consisting of acid and alkaline process had different yields based on the protein content. The yield of fish protein made by alkali process was higher (77.30 %) than that of fish protein made by acidprocess (63.28 %). These results were in line with Chen and Jaczynski [10], who reported protein isolates recovered from trout processing by-product. High yields of alkaline process of salmon and pollock head of 50 % and 87 % were also reported by Bechtel et al. [16]. However, theseresultswere contrary to Panpina and Chaijan [4], who reported that the recovery protein from head by-product of big eye snapper hadhigher yield by acid process than byalkaline process. These results indicated that different raw materials had different suitability at different pH levels of the processing.

3.2.Proximate analysis
The protein, lipid, and moisture contents of fish protein isolate areexhibited in table 1.  Based on the proximate analysis, somereduction of protein occurredin the acid and alkaline process. This result was contrary to Chomnawang and Yongsawatdigul [17], who reported that 85-90% of protein content in fish protein isolate was produced from tilapia by-product. The reduction of protein might bedue to the denaturation of muscle proteins induced by pH shift, which caused the aggregation of protein and led to separation of bone, skin, and debris during centrifugation [4]. Another reason might be due to the use of centrifugation speedthat was lower in this study than in the previous study. Some significant reduction of lipid content showed in acid and alkaline process compared to the raw material. Lipids are susceptible to oxidation, which leads to rancidity or fishy odour [10].Thus, the removal of lipid is an importantcharacteristic. Both treatments (acid and alkali) were effective in lipid removal. The effectivenessof alkaline process in lowering lipid content might bedue to the saponification reaction of alkali and lipid, whichproducedsoap that precipitated during centrifugation. This result was in accordance with [17], who reported that alkaline process was more effective than acidprocessinlipid removal.

Total volatilebasic-nitrogen (TVB-N)
The total volatile basic nitrogen of fish protein isolate and raw material is exhibited in table 2. There wasno significant difference between acid process and alkaline process (P <0.05) in terms of TVB-N value. Both processes improved the stability of cellular structure and preventedthe spoilage of fish protein isolate produced. TVB-N value indicates the measurement of dimethylamine (DMA), trimethylamine (TMA), ammonia, and another compound, which indicates seafood spoilage due to the degradation of muscle proteins. The decreaseinTVB-N might bedue to the washing mechanism of the centrifugation and precipitation process when the structural protein at pH 5.5 precipitated, while other compounds (hydrophilic compounds such as DMA and TMA) leached out into the supernatant [4].

Peroxide value
Peroxide value depicts the formation of unstable compound (peroxide) due to lipid oxidation process. Fish muscle contains high amount of unsaturated fatty acids, particularly n-3 fatty acid that is susceptible to oxidation [4]. The peroxide value of this study (table 2) exhibited that the fish protein wassusceptibletooxidation. Both acid process and alkaline process resulted inhigh peroxide values. Rapid lipid oxidation was reported in thefish protein isolaterecovered by acid process [9]. However, this study showedthat there wasno significant differenceinlipid oxidation between acid and alkaline process. Thesefindingswere in accordance with Panpipat and Chaijan [4],who reported that both processes(acid and alkaline process) enhanced the oxidative instability of lipid in bigeye snapper head by-product.

3.5.Water holding capacity
Water holding capacity indicates the ability of the food product to store water during processing. Sarcoplasmic protein has low water holding capacity, which does not allow water for a well-developed gel matrix formation [18]. The WHC of fish protein isolate produced by acid and alkaline process in this study was significantly higher than that of the raw material (P < 0.05), butthere was no difference between acid and alkaline process. These results were in accordance with Foh et al. [19], who reported that the fish protein isolate produced from tilapia by acid process and alkaline process was around 2.63 and 2.51 mL/g, respectively. This finding indicated that both methods reduced the sarcoplasmic proteins and isolated the muscle/myofibrilar proteins. The isoelectric solubilisation/precipitation method (ISP) employing pH shift during processing was believed to affect the water holding capacity of the fish protein isolate produced [19].

Oil binding capacity
The oil binding capacity of fish protein isolate from red snapper by-product produced byacid and alkaline process was significantly higher than that of the raw material used (table 3). The acidprocess showed a higher score (0.17 mL/g), but the score was not significantly different from that ofalkaline process (0.15 mL/g). This finding was in accordance with Foh et al. [19], who reported that the fish protein isolate of tilapia produced by acid and alkaline process had high oil binding capacity (3.38 mL/g). High oil absorption of fish protein is an important characteristic of formulation of food system such as cake, sausage, mayonnaise, butter, and saladdressing [19].

Amino acid profile
The amino acid profile of the fish protein recovered withISP was affected by the pH treatment. Chen, Tou, and Jaczynski [20] reported that extreme pH during ISP process improvedthe essential amino acid profile of the FPI isolated from whole antarctickrill (Euphausiasuperba). Glutamic acid and lysine were higher than other amino acids. Meanwhile, threonine was the lowestamino acid in acidprocess, and histidine was the lowest in alkaline process. Thesefindingswerein accordance with [20], who reported that glutamic acid was the highestamino acid contained in krill protein isolate produced with isoelectric solubilisation/precipitation method.

Conclusion
Isoelectric solubilisation/precipitation with acid and alkali process can be an alternative process to recover proteins from red snapper by-product. This study demonstrates that alkali process gave a protein isolate with better overall quality compared to acid process.