2016 Volume 22 Issue 3 Pages 337-347
In order to identify objective parameters for quality evaluation, physical dimensions, pH, color parameters, proximate composition, water activity, lipid oxidation, free amino acids, mechanical properties, protein composition and differential scanning calorimetry of commercial Japanese spicy pollack roe products of differing quality were analyzed. Low-priced roe products showed significantly higher (p < 0.05) average thiobarbituric acid reactive substances content than high-priced roe products from the same company. More bitter amino acids were detected in the low-priced, compared to the high-priced roe products from the same company. Ovary membrane and eggshell proteins in low-priced roe products were comprised of a relatively greater amount of low-molecular-weight components. Compared to low-priced roe products, high-priced ones showed higher values for the mechanical properties and thermal transition enthalpy of fish egg.
Alaska pollack (Theragra chalcogramma) is a semi-pelagic schooling fish widely distributed in the temperate and arctic waters of the North Pacific Ocean. It is a member of the cod family and supports one of the largest single-species fisheries in the world (Smith 1981). It is harvested by Japan, Canada, Korea, Russia, and the USA, and its aggregate biomass was an estimated 18 million metric tons in the Gulf of Alaska, Bering Sea, and Sea of Okhotsk in 2010-2012 (Seafish, 2013)i).
The roe obtained from Alaska pollack is commercially important, since it occupies only 5% of the fish weight but contributes about 31% of the commercial value of pollack products (Balaban et al., 2012). It is an important export product of the USA, accounting for 14,966 metric tons of total fishery products exported in 2013 (National Marine Fisheries Service (NMFS)ii), 2013). The majority of these are exported to Japan, with limited quantities exported to Korea (Alaska Seafood Marketing Institute (ASMI), 2011)iii), where it has a long history of use as a food material. Manufacturers of roe products in Japan have their own expertise in classifying raw Alaska pollack roe into gamuko, mako, and mizuko by maturity stage. Usually, only mako is accepted by manufacturers as a standard roe material for manufacturing spicy pollack roe product which is a famous delicacy known as karashi mentaiko in Japanese.
Generally, prices for spicy pollack roe products fluctuate widely according to the quality of different manufacturers. However, determination of quality is based on the empirical judgment of manufacturers. In recent years, the average price of roe products has fallen in Japan (Teikoku Databank, Ltd., 2011)iv). The lack of an objective quality evaluation system may have negatively impacted consumer perceptions of roe products and is considered as one of the reasons for the downward pressure on the average price of spicy pollack roe products. Therefore, it is necessary to develop a clear standard for the quality evaluation of roe products.
In this study, a total of 31 parameters were assessed in high-and low-priced roe products as follows: ovary physical dimensions (ovary weight, ovary length, ovary condition factor, ovary membrane thickness, egg diameter), pH, color parameters (L*, a*, and b*), proximate composition (moisture, ash, lipid, protein, carbohydrate), salt content, water phase salt content, peroxide value (PV), thiobarbituric acid reactive substances (TBARS), free amino acids (sweet amino acids, umami-tasting amino acids, and bitter amino acids), mechanical properties (firmness, gumminess, stickiness, tensile strength, and breaking strength), protein composition (ovary membrane and eggshell), and differential scanning calorimetry (denaturation temperature and thermal transition enthalpy) in order to provide basic information for the establishment of evaluation standards for spicy pollack roe products.
Materials High-priced (about 1000 yen/100 g or about 8.3 dollars/100 g) and low-priced (about 400 yen/100 g or about 3.3 dollars/100 g) spicy pollack roe products from 5 companies (marked A, B, C, D, and E) in Japan were sampled randomly.
Condition factor, egg diameter, and ovary membrane thickness After measuring the weight and length of the ovary, the condition factor was calculated by the method of Osako et al. (2003) using the following equation:
 |
Egg diameter and ovary membrane thickness were measured by a RHEONER II creep meter (RE2-33005B; Yamaden Co., Ltd., Tokyo, Japan) equipped with a cylindrical plunger (diameter = 8 mm), with a raising rate of the sample table at 1 mm/s.
pH The determination of pH value was conducted according to the method of Kung et al. (2008) with slight modifications. Each sample (10 g) was homogenized in 90 mL deionized water using a homogenizer (DIAX600; Nikko Hansen & Co., Ltd., Japan), and pH of the obtained homogenate was measured using a pH meter (pH211; Hanna Instruments Co., Ltd., Japan).
Color parameters The surface color tone of the roe products was measured by a color reader (CR-13; Konica Minolta Sensing Inc., Tokyo, Japan), and was expressed as L* (lightness), a* (redness / greenness), and b* (yellowness / blueness) values (CIE Laboratory System). Six ovaries were selected randomly for each sample lot of roe product, and the surface color was measured at four arbitrary points for each sample.
Proximate composition Moisture and crude ash contents were analyzed according to the methods of AOAC (1999). Total nitrogen content was determined by the Kjeldahl method (Bradstreet et al., 1954) and a nitrogen conversion factor of 6.25 was used for calculating the protein content. Crude lipid was extracted according to the method of Folch et al. (1957). The crude carbohydrate content was obtained by subtracting the total composition ratios (%) of other chemical components from 100%.
Salt content, water phase salt content, and water activity The same homogenate used for pH determination was used for salt content measurement with a salt meter (Atago Co., Ltd., Tokyo, Japan). Water phase salt content was calculated from the values of salt content and moisture content (Cornu et al., 2006) as follows:
Water phase salt content = Salt content / Moisture content
Water activity was measured by the method of Liu et al. (2013) with a Novasina water activity meter (model ms1; Novasina AG Ltd., Lachen, Switzerland) at 25.0 ± 0.3°C.
PV and TBARS The PV and TBARS of crude lipid were determined according to the method described by Weng et al. (2009).
Free amino acids Free amino acid analysis was conducted according to the method of Kaneko et al. (2012) with a slight modification. To extract free amino acids, 10 g of roe sample was homogenized in 10 mL deionized water; 10 mL of 10% trichloroacetic acid (TCA) was then added and the sample was homogenized. The mixed solution was centrifuged at 10000 × g for 15 min. The precipitate was washed with 10 mL 5% TCA to extract remaining free amino acids, and the same procedure was repeated 3 more times. The supernatant was filtered through filter paper (No. 5A) and its volume was adjusted to 100 mL. After filtering through a cellulose membrane filter (0.45 µm; Toyo, Roshi Kaisha, Ltd., Tokyo, Japan), its free amino acid composition was determined by an amino acid analysis system (Prominence; Shimadzu, Kyoto, Japan) equipped with a column (Shim-pack Amino-Li, 100 mm × 6.0 mm i.d.; column temperature, 39.0°C; Shimadzu) and pre-column (ShimpackISC-30/S0504 Li, 150 mm × 4.0 mm i.d. Shimadzu). A fluorescence detector (RF-10AXL, Shimadzu) was used for detecting amino acids.
Mechanical properties Sample preparation for assessing mechanical properties is shown in Figure 1. The firmness of roe products was assessed by measuring the strength at 20% strain using a RHEONER II creep meter (RE2-33005B; Yamaden Co., Ltd.) equipped with a cylindrical plunger (diameter = 16 mm). The raising speed of the sample table was 1 mm/s.
Preparation of samples for mechanical properties measurement.
Ovary membranes were cut into a rectangular (1.5 cm × 3.0 cm) film. A Tensipresser (TTP-50 BX II 2006; Taketomo Electric Inc., Tokyo, Japan) was used to tear the rectangular film into two pieces. The tensile speed was set at 1 mm/s, and the maximum strength was recorded as tensile strength.
The food textural profile analysis developed by Friedman et al. (1963), Szczesniak et al. (1963), and Bourne (1978) was used to assess the gumminess and stickiness of ovary round slice. The ovary round slice (height = 15 mm, diameter > 16 mm) was pressed twice on its center by the RHEONER II creep meter equipped with a cylindrical plunger (diameter = 16 mm). The plunger penetration was 20% and the raising speed of the sample table was 1 mm/s. The force-time curve is shown in Figure 2, and consists of 2 positive peaks and 2 negative peaks. According to Bourne (1978), the peak force during the first bite is defined as hardness, the ratio of the positive force area during the second bite to that during the first bite (Area 2 / Area 1) is defined as cohesiveness, “hardness × cohesiveness” is defined as gumminess, and stickiness (adhesiveness) is defined as the negative force area for the first bite (Area 3).
Typical force-time curve of the ovary round slice of spicy pollack roe products.
The breaking strength of fish egg was measured by the RHEONER II creep meter equipped with a cylindrical plunger (diameter = 8 mm). The raising speed of the sample table was 1 mm/s. Each measurement was conducted with 18 repeats, and the top 5 values and the bottom 5 values were deleted.
Sample preparation for SDS-PAGE of ovary membrane protein and its solubility Ovary membrane materials were taken from six roe products with the same grade randomly, and were then chopped and mixed. The mixed sample (0.5 g) was homogenized in 5 mL SDS solution (2% SDS, 8 M urea, 2% 2-mercaptoethanol in 20 mM Tris-HCl, pH = 8.0) in a centrifuge tube and heated in boiling water for 2 min. After cooling, the homogenate was stirred at room temperature for 12 h, and then centrifuged at 10,000 × g for 20 min. The supernatant thus obtained was used for SDS-PAGE of ovary membrane proteins. The protein concentration was measured by the method of Lowry et al. (1951). The ovary membrane solubility in SDS solution was measured according to the method of Chen et al. (2012).
Sample preparation for SDS-PAGE of eggshell protein and its solubility Preparation of eggshell protein for SDS-PAGE was according to the method of Bekhit et al. (2009) with slight modifications. Fish eggs from each company were taken from six roe products with the same grade randomly and mixed. The mixed sample (2.0 g) was homogenized in 6 mL pre-cooled 20 mM Tris-HCl buffer (pH = 8.0) containing 10 mM EDTA and 2 mM DTT, and then centrifuged at 45,000 × g for 30 min at 4°C. The obtained pellet was washed by repeating the procedure 3 more times to remove any remaining soluble protein. The washed pellet was collected as eggshell protein. The eggshell protein was homogenized in 5 mL SDS solution (2% SDS, 8 M urea, 2% 2-mercaptoethanol in 20 mM Tris-HCl, pH = 8.0), and heated in boiling water for 2 min. After cooling, the homogenate was stirred at room temperature for 12 h, and then centrifuged at 10,000 × g for 20 min. The obtained supernatant was collected for SDS-PAGE of eggshell protein.
The eggshell protein was not completely solubilized in SDS solution. The insoluble eggshell protein was visible and its solubility could not be measured by the same method used for measuring the solubility of ovary membrane protein. However, eggshell protein was completely solubilized in 2 N NaOH as reported by Iuchi et al. (1991). Therefore, eggshell protein solubility in SDS solution was conducted according to the method of Benjakul and Visessanguan (2003) with slight modifications. TCA solution was added to the homogenate or its supernatant to a final concentration of 10% to precipitate protein in the SDS solution. The mixed solution was centrifuged at 10,000 × g for 30 min. The obtained pellet was washed with 10% TCA 3 times, and then solubilized in 2 N NaOH. The protein concentration in 2 N NaOH was measured by the method of Lowry et al. (1951). Protein solubility in the SDS solution was calculated by the ratio of the amount of protein solubilized in the supernatant and total protein in the homogenate.
SDS-PAGE SDS-PAGE was performed according to the method of Laemmli et al. (1970) using a 15% polyacrylamide gel (AE-6000, NPU-7.5L PAGEL; Atto Co., Tokyo, Japan). Each sample (10 µg) was applied to the polyacrylamide gel and subjected to electrophoresis at a constant current of 20 mA using a compact- PAGE apparatus (Atto Co.). The polyacrylamide gel was stained with 0.05% Coomassie Brilliant Blue R-250 in methanol/acetic acid/water (5:10:85% v/v), and then destained with methanol/acetic acid/water (30:10:60% v/v). A protein ladder (Thermo Fisher Scientific Inc., Massachusetts, USA) with molecular weights ranging from 10 to 200 kDa was used.
Differential scanning calorimetry The measurements were conducted with a Shimadzu DSC-50 following the method of Schubring (2004) with slight modifications. Samples (30 – 50 mg fish eggs or ovary membrane) were weighed (±0.1 mg) and sealed in aluminum pans. Triplicate aluminum pans were heated from 10°C to 145°C at a scanning rate of 5°C/min with an empty aluminum pan as reference. Results were displayed as average curves in the figures. The transition temperature (Td, °C) and transition enthalpy (ΔH, mJ/mg) were estimated from the peak of the DSC transition curve.
Statistical analysis Data analysis was conducted using Microsoft Excel Statistical Analysis 2012 (Social Survey Research Information Co., Ltd., Tokyo, Japan). Results were compared based on the multiple comparison test of Tukey–Kramer. In all cases, the criterion for statistical significance was set at p < 0.05.
Ovary physical dimensions, pH, and color parameters Ovary physical dimensions of commercial spicy pollack roe samples are presented in Table 1. The ovary weight of high-priced spicy pollack roe products ranged from 33.3 g to 96.2 g, comparable to that of the low-priced products (33.6 – 54.1 g). Companies B, C, and E provided high-priced roe products with higher weight values than those of low-priced products. Only the products from company C showed significant differences (p < 0.05) in ovary weight and length values between high- and low-priced products. Ovary weight and length are not only affected by maturity stage, but also by fish age (Oosthuizen and Daan, 1974; Katsiadaki et al., 1999), which may explain why the weight and length values, especially ovary weight values, showed large variability. The ovary membrane thickness value (about 0.18 mm) did not differ between high- and low-priced products, regardless of the producer. Therefore, measurements of ovary weight, ovary length, condition factor, and thickness of the ovary membrane may not be applicable parameters for the objective quality evaluation of roe products.
| A | B | C | ||||
|---|---|---|---|---|---|---|
| H | L | H | L | H | L | |
| Ovary weight | 33.3 ± 2.0c | 33.6 ± 1.7c | 52.7 ± 6.7bc | 46.1 ± 2.0bc | 96.2 ± 8.9a | 40.0 ± 1.7c |
| (g, n=6) | ||||||
| Ovary length | 97.5 ± 0.8c | 9.9 ± 0.7bc | 11.0 ± 0.6b | 10.6 ± 0.6bc | 12.5 ± 0.8a | 10.6 ± 0.4bc |
| (cm, n=6) | ||||||
| Condition factor | 40.1 ± 9.5b | 35.1 ± 7.6b | 39.9 ± 5.1b | 39.0 ± 5.8b | 50.9 ± 11.1ab | 34.1 ± 3.8b |
| (kg/cm3, n=6) | ||||||
| Egg diameter | 0.53 ± 0.06ab | 0.51 ± 0.05b | 0.53 ± 0.05ab | 0.46 ± 0.07b | 0.61 ± 0.05a | 0.34 ± 0.08c |
| (mm, n=8) | ||||||
| Ovary membrane | 0.17 ± 0.01a | 0.16 ± 0.02a | 0.18 ± 0.05a | 0.18 ± 0.03a | 0.19 ± 0.03a | 0.20 ± 0.04a |
| thickness (mm, n=6) | ||||||
| pH (n=4) | 5.98 ± 0.08ab | 5.95 ± 0.03b | 6.17 ± 0.03a | 6.15 ± 0.04ab | 5.77 ± 0.01b | 5.75 ± 0.01b |
| Color parameters | ||||||
| L* (n=6) | 44.5 ± 3.8a | 36.9 ± 1.8c | 41.3 ± 0.9bc | 40.4 ± 2.0bc | 37.7 ± 3.3bc | 40.0 ± 2.2bc |
| a* (n=6) | 11.6 ± 3.3cd | 17.1 ± 2.4bc | 6.4 ± 1.2d | 6.6 ± 1.3d | 11.1 ± 0.6cd | 11.6 ± 2.2cd |
| b* (n=6) | 13.6 ± 2.6ab | 13.2 ± 2.2ab | 7.4 ± 0.6b | 10.1 ± 1.4b | 7.6 ± 1.7b | 10.8 ± 0.9b |
| D | E | All examined companies | ||||
|---|---|---|---|---|---|---|
| H | L | H | L | H | L | |
| Ovary weight | 55.9 ± 2.6b | 54.1 ± 7.5bc | 50.9 ± 4.9bc | 45.9 ± 1.2bc | 57.8 ± 23.2b | 43.9 ± 7.6bc |
| (g, n=6) | ||||||
| Ovary length | 10.1 ± 0.7bc | 9.9 ± 0.4bc | 11.1 ± 0.7b | 11.0 ± 0.6b | 10.8 ± 1.2bc | 10.4 ± 0.5bc |
| (cm, n=6) | ||||||
| Condition factor | 56.6 ± 12.8a | 56.6 ± 9.3a | 38.1 ± 6.5b | 35.4 ± 5.3b | 44.1 ± 8.2ab | 40.1 ± 9.5ab |
| (kg/cm3, n=6) | ||||||
| Egg diameter | 0.54 ± 0.05ab | 0.49 ± 0.0b | 0.53 ± 0.05ab | 0.50 ± 0.06b | 0.55 ± 0.04ab | 0.46 ± 0.07b |
| (mm, n=8) | ||||||
| Ovary membrane | 0.19 ± 0.03a | 0.18 ± 0.03a | 0.19 ± 0.04a | 0.18 ± 0.03a | 0.18 ± 0.01a | 0.18 ± 0.01a |
| thickness (mm, n=6) | ||||||
| pH (n=4) | 6.12 ± 0.12ab | 6.06 ± 0.04ab | 6.08 ± 0.03ab | 6.08 ± 0.02ab | 6.02 ± 0.16ab | 6.00 ± 0.15ab |
| Color parameters | ||||||
| L* (n=6) | 36.5 ± 1.2c | 37.8 ± 1.8bc | 41.6 ± 1.9b | 39.7 ± 1.1bc | 40.3 ± 3.2bc | 39.0 ± 1.5bc |
| a* (n=6) | 24.8 ± 3.2b | 34.4 ± 2.1a | 6.2 ± 0.7d | 6.7 ± 0.8d | 12.0 ± 7.6cd | 15.3 ± 11.5c |
| b* (n=6) | 13.5 ± 2.8ab | 15.9 ± 2.1a | 6.2 ± 0.8b | 9.9 ± 1.4b | 9.8 ± 3.4b | 11.9 ± 2.6ab |
The spicy pollack roe products from 5 companies are marked as A, B, C, D, and E.
“H” and “L” denote high-level and low-level roe products, respectively.
Different superscripts in the same row indicate statistical differences (p < 0.05).
Data are expressed as mean ± standard deviation.
The egg diameter of high-priced products occupied a narrow range (0.53 – 0.61 mm) compared to the low-priced products (0.34 – 0.51 mm). The average egg diameter of high-priced roe products from all 5 companies was higher than that of the low-priced products. This is most likely due to companies B and C using ovaries of a low maturity degree to produce low-priced roe products, since the diameter of cod eggs increases with the degree of maturity (Katsiadaki et al., 1999). As maturity is closely related to the taste of roe products (Uchiumi et al., 2009a), egg diameter, an indicator of ovary maturity, may be applicable as one of the quality parameters.
The pH of the commercial roe products examined ranged from 5.75 to 6.17 (Table 1). Roe products from the same company showed no significant differences in pH, but pH values differed between companies. This is most likely due to individual companies adjusting the pH values of roe products according to their own processing standards. Moreover, no correlation was found between the pH value and quality of roe products from the same company.
The surface color parameters of the roe products examined are presented in Table 1. The L* values (lightness) of high-priced products ranged from 36.5 – 44.5, and those of low-priced products were from 36.9 to 40.4. The a* values (redness) of high-priced products ranged from 6.2 to 24.8, and those of low-priced products were from 6.6 to 34.4. The b* (yellowness) values of high-priced products ranged from 6.2 to 13.6, while those of low-priced products were from 9.9 to 15.9. According to Ueda et al. (2009), the optimum values of L*, a*, and b* of cod roe for consumer preference are 45.22, 8.68, and 9.15, respectively. The average color values of all examined high-priced products (L* = 40.3, a* = 12.0, and b* = 9.8) were closer to the reported optimum values than those of low-priced roe products (L* = 39.0, a* = 15.3, and b* = 11.9). Processing of spicy pollack roe products can lead to changes in the proximate composition and structure of roe (Hayabuchi et al., 1997), which may consequently affect the color of the final products. According to our interviews with roe product manufacturers, pigments or color-fixing agents, such as sodium nitrite, are sometimes used to adjust the color tone of the final roe products. Therefore, the color tone of roe products may not be a suitable parameter for quality evaluation.
Proximate composition Few significant differences in proximate composition between high- and low-priced roe products were observed. Respective average values of moisture, protein, ash, lipid, and carbohydrate contents of high-priced roe products were 68.1%, 22.6%, 5.6%, 3.3%, and 0.4%, and those of low-priced products were 68.2%, 22.4%, 5.6%, 3.4%, and 0.4% (Table 2). Although the moisture content of roe material is dependent on the maturity stage (Uchiumi et al., 2009a), the proximate composition of processed roe, especially moisture content, changes during manufacturing (Hayabuchi et al., 1997). Therefore, the quality of processed roe may be difficult to evaluate using proximate composition parameters.
| A | B | C | ||||
|---|---|---|---|---|---|---|
| H | L | H | L | H | L | |
| Moisture content | 67.6 ± 2.0a | 69.0 ± 2.4a | 67.1 ± 1.4a | 68.2 ± 2.1a | 70.0 ± 1.1a | 69.5 ± 1.7a |
| (%) | ||||||
| Protein content | 24.0 ± 1.6a | 22.8 ± 2.0ab | 23.7 ± 1.5ab | 23.2 ± 1.5ab | 20.2 ± 1.0b | 20.5 ± 0.7ab |
| (%) | ||||||
| Ash content | 5.1 ± 0.1a | 5.1 ± 0.2a | 5.5 ± 0.8a | 5.2 ± 0.6a | 5.9 ± 1.1a | 6.6 ± 0.6a |
| (%) | ||||||
| Lipid content | 2.8 ± 0.6a | 2.9 ± 0.5a | 3.2 ± 0.6a | 3.0 ± 0.7a | 3.5 ± 0.2a | 3.1 ± 1.3a |
| (%) | ||||||
| Carbohydrate content | 0.5 ± 0.1a | 0.3 ± 0.1a | 0.4 ± 0.1a | 0.4 ± 0.1a | 0.4 ± 0.1a | 0.4 ± 0.2a |
| (%) | ||||||
| Salt content | 4.5 ± 0.0bc | 4.5 ± 0.0bc | 4.5 ± 0.1bc | 4.4 ± 0.1c | 4.3 ± 0.1c | 6.0 ± 0.1a |
| (g/100g) | ||||||
| Water phase salt content | 6.6 ± 0.1bc | 6.5 ± 0.2bc | 6.7 ± 0.1bc | 6.5 ± 0.3bc | 6.2 ± 0.0c | 8.7 ± 0.3a |
| (g/100ml) | ||||||
| Water activity | 0.94 ± 0.00b | 0.94 ± 0.00b | 0.94 ± 0.00b | 0.94 ± 0.00b | 0.95 ± 0.01a | 0.92 ± 0.00c |
| PV | 13.6 ± 1.5b | 17.2 ± 0.6ab | 8.8 ± 1.5b | 23.5 ± 5.8a | 13.2 ± 4.6b | 19.7 ± 0.7ab |
| (meq/kg) | ||||||
| TBARS | 71.2 ± 5.3c | 95.1 ± 9.0bc | 53.9 ± 9.8c | 151.4 ± 7.7ab | 67.5 ± 10.8c | 99.9 ± 7.2bc |
| (mg MDA/kg) | ||||||
| D | E | All examined companies | ||||
|---|---|---|---|---|---|---|
| H | L | H | L | H | L | |
| Moisture content | 68.0 ± 1.2a | 68.3 ± 0.7a | 67.7 ± 0.9a | 66.3 ± 2.8a | 68.1 ± 1.1a | 68.2 ± 1.2a |
| (%) | ||||||
| Protein content | 22.4 ± 1.1ab | 21.8 ± 1.2ab | 22.7 ± 1.4ab | 23.8 ± 2.5ab | 22.6 ± 1.5ab | 22.4 ± 1.3ab |
| (%) | ||||||
| Ash content | 5.8 ± 0.7a | 5.7 ± 0.9a | 5.8 ± 1.5a | 5.6 ± 0.9a | 5.6 ± 0.3a | 5.6 ± 0.6a |
| (%) | ||||||
| Lipid content | 3.5 ± 0.3a | 3.9 ± 0.3a | 3.5 ± 0.4a | 4.0 ± 0.3a | 3.3 ± 0.3a | 3.4 ± 0.5a |
| (%) | ||||||
| Carbohydrate content | 0.3 ± 0.2a | 0.3 ± 0.1a | 0.3 ± 0.1a | 0.4 ± 0.1a | 0.4 ± 0.1a | 0.4 ± 0.0a |
| (%) | ||||||
| Salt content | 5.5 ± 0.1ab | 5.5 ± 0.3ab | 4.4 ± 0.2c | 5.3 ± 0.1ab | 4.6 ± 0.5bc | 5.2 ± 0.7b |
| (g/100g) | ||||||
| Water phase salt | 8.1 ± 0.1ab | 8.1 ± 0.4ab | 6.4 ± 0.3bc | 8.0 ± 0.3ab | 6.8 ± 0.8bc | 7.5 ± 1.0b |
| content (g/100ml) | ||||||
| Water activity | 0.92 ± 0.01bc | 0. 92 ± 0.01bc | 0.94 ± 0.00b | 0.92 ± 0.00c | 0.94 ± 0.01b | 0.93 ± 0.01bc |
| PV | 13.6 ± 1.5b | 24.8 ± 8.4a | 9.0 ± 0.6b | 13.1 ± 1.4b | 11.6 ± 2.5b | 19.7 ± 4.7ab |
| (meq/kg) | ||||||
| TBARS | 75.5 ± 3.2c | 176.8 ± 12.8a | 54.9 ± 1.5c | 67.3 ± 1.9c | 64.6 ± 9.7c | 118.1 ± 44.7b |
| (mg MDA/kg) | ||||||
Abbreviations are the same as given in Table 1.
Different superscripts in the same row indicate statistical differences (p < 0.05).
Data are expressed as mean ± standard deviation (n = 4).
Salt content, water phase salt content, and water activity Salt content and water phase salt content, and water activity of roe products are presented in Table 2. The average salt content and water phase salt content of the high-priced roe products were 4.6% and 6.8%, while those of low-priced products were 5.2% and 7.5%. Companies C and E provided low-priced roe products containing significantly higher salt content (p < 0.05) than the high-priced products. It is possible that the roe material of the low-priced products suffered greater damage from longer frozen storage than the high-priced products, resulting in increased NaCl absorbance during salting. Another possible reason for the high salt content of the low-priced roe products, especially those from company C, is an attempt to reduce the high moisture content of roe material with a low maturity degree, since it was reported that roe material with a low maturity degree contained high moisture content (Uchiumi et al., 2009a).
The average water activity values of high- and low-priced roe products were 0.94 and 0.93, respectively, not significantly different. Roe products from company D and low-priced roe products from companies C and E showed low water activity (0.92), since they showed comparative high water phase salt content compared to other roe products.
PV and TBARS The crude lipid extracted from roe products was analyzed for PV and TBARS (Table 2). Both PV and TBARS of low-priced roe products (13.1 – 24.8 meq/kg for PV, and 67.3 – 176.8 mg MDA/kg for TBARS) were higher than those of high-priced roe products (8.8 – 13.6 meq/kg for PV, and 53.9 – 75.5 mg MDA/kg for TBARS) from the same company. A comparatively longer storage period may be the reason for the higher values of PV and TBARS of low-priced roe products compared to the high-priced roe products.
Free amino acids The 17 typical free amino acids were detected and identified in spicy pollack roe products (Table 3). Sweet amino acids (glycine and alanine) and umami-tasting amino acids (glutamic acid and aspartic acid) were the main amino acids contained in the roe products. High-priced roe products showed slightly higher average contents of umami-tasting amino acids (2057.7 mg/100 g) than those (1820.3 mg/100 g) of low-priced roe products. However, umami-tasting amino acids, as well as sweet amino acids, may be derived from food additives (Ueda et al., 2009), thus they cannot be considered as parameters for the quality evaluation of spicy roe products.
| Amino acid (mg/100g) | A | B | C | ||||
|---|---|---|---|---|---|---|---|
| Taste | H | L | H | L | H | L | |
| Taurine | - | 87.1 ± 9.8bc | 96.2 ± 6.1bc | 66.4 ± 8.4c | 76.0 ± 11.1bc | 110.7 ± 9.1b | 182.3 ± 10.4a |
| Aspartic acid | U | 45.3 ± 4.2b | 54.1 ± 1.4b | 53.7 ± 2.0b | 71.4 ± 2.8b | 76.6 ± 2.1b | 119.6 ± 2.7b |
| Threonine | S | 26.3 ± 2.4c | 33.6 ± 1.5bc | 24.0 ± 1.7c | 30.3 ± 2.4c | 33.7 ± 1.8bc | 58.8 ± 2.3b |
| Serine | S | 38.1 ± 1.6c | 48.6 ± 1.1bc | 43.4 ± 1.5c | 54.2 ± 2.0bc | 65.1 ± 1.6bc | 123.2 ± 1.9ab |
| Glutamic acid | U | 1928.1 ± 252.4ab | 1134.8 ± 153.0b | 2223.6 ± 210.3ab | 2253.7 ± 266.6ab | 2430.9 ± 230.3a | 2322.9 ± 251.4ab |
| Proline | S | 27.0 ± 3.0ab | 31.3 ± 1.7ab | 30.0 ± 2.2ab | 36.0 ± 2.3ab | 38.3 ± 2.6ab | 45.3 ± 2.2a |
| Glycine | S | 159.6 ± 38.9b | 29.4 ± 1.4b | 138.0 ± 16.7b | 147.3 ± 18.3b | 90.2 ± 22.3b | 97.8 ± 58.2b |
| Alanine | S | 45.8 ± 6.9c | 52.9 ± 4.6c | 112.3 ± 6.4bc | 153.7 ± 9.1bc | 860.3 ± 87.8ab | 1210.1 ± 147.8a |
| Valine | B | 34.0 ± 7.7c | 45.4 ± 4.9bc | 33.7 ± 6.8c | 46.1 ± 9.3bc | 46.6 ± 7.2bc | 83.8 ± 8.8bc |
| Methionine | B | 15.1 ± 1.6b | 20.0 ± 0.9b | 16.8 ± 1.2b | 18.3 ± 1.2b | 18.2 ± 1.4b | 30.9 ± 1.2b |
| Isoleucine | B | 23.4 ± 5.6b | 32.5 ± 3.4b | 20.5 ± 4.6b | 29.4 ± 5.9b | 27.0 ± 5.1b | 50.1 ± 5.5b |
| Leucine | B | 56.4 ± 5.8b | 70.3 ± 3.0b | 51.0 ± 4.0b | 48.4 ± 4.6b | 84.6 ± 4.6b | 135.7 ± 4.4b |
| Tyrosine | B | 28.0 ± 0.7c | 39.7 ± 0.4c | 25.8 ± 4.9c | 43.1 ± 4.6c | 46.8 ± 6.3bc | 102.7 ± 4.5b |
| Phenylalanine | B | 25.0 ± 3.3c | 34.9 ± 1.7bc | 21.0 ± 2.3c | 28.9 ± 1.7bc | 31.0 ± 2.8bc | 52.1 ± 1.7b |
| Histidine | B | 13.4 ± 1.2c | 17.4 ± 0.7bc | 12.6 ± 0.9c | 13.6 ± 1.1c | 18.5 ± 1.0bc | 33.8 ± 1.1ab |
| Lysine | B | 45.2 ± 1.1ab | 60.1 ± 2.5ab | 35.7 ± 6.9b | 46.4 ± 7.01ab | 41.1 ± 8.1ab | 64.0 ± 6.8a |
| Arginine | B | 47.1 ± 2.9c | 59.8 ± 0.8bc | 35.9 ± 3.4c | 46.8 ± 3.0c | 47.4 ± 4.1c | 87.7 ± 2.9b |
| Total FAA | 2645.0 ± 229.5b | 1860.9 ± 123.2c | 2944.4 ± 170.7b | 3143.4 ± 219.3b | 4066.8 ± 113.9ab | 4771.9 ± 76.9a | |
| Sweetness | 296.9 ± 29.1bc | 195.7 ± 6.4c | 347.7 ± 17.2bc | 421.4 ± 29.6bc | 1087.5 ± 73.3b | 1535.1 ± 109.3ab | |
| Umami | 1973.3 ± 248.9ab | 1188.9 ± 151.6b | 2277.3 ± 208.3ab | 2325.1 ± 263.7ab | 2507.4 ± 228.3a | 2442.5 ± 248.8ab | |
| Bitterness | 287.7 ± 28.3c | 380.1 ± 17.1bc | 253.0 ± 32.9c | 320.8 ± 36.8c | 361.1 ± 39.0bc | 640.8 ± 35.1ab | |
| D | H | All examined companies | |||||
|---|---|---|---|---|---|---|---|
| H | L | H | L | H | L | ||
| Taurine | - | 99.1 ± 4.6bc | 106.3 ± 9.3bc | 84.6 ± 8.9bc | 97.7 ± 5.6bc | 89.6 ± 16.6bc | 111.7 ± 41.0b |
| Aspartic acid | U | 253.1 ± 1.0ab | 280.8 ± 2.1a | 245.5 ± 2.1ab | 221.5 ± 1.3ab | 134.8 ± 105.2b | 149.5 ± 98.1b |
| Threonine | S | 73.5 ± 1.0ab | 87.2 ± 1.8a | 70.1 ± 1.8ab | 68.9 ± 1.1ab | 45.5 ± 24.3bc | 55.8 ± 24.0bc |
| Serine | S | 171.3 ± 0.8ab | 193.4 ± 1.7a | 170.4 ± 1.6ab | 155.1 ± 1.0ab | 97.7 ± 67.6bc | 114.9 ± 63.1b |
| Glutamic acid | U | 1580.2 ± 120.3b | 1414.0 ± 237.5b | 1451.4 ± 223.4b | 1228.6 ± 141.0b | 1922.8 ± 414.8ab | 1670.8 ± 573.1b |
| Proline | S | 0.0c | 0.0c | 0.0c | 0.0c | 19.1 ± 17.9b | 22.5 ± 21.2b |
| Glycine | S | 1116.7 ± 30.8a | 931.0 ± 79.1ab | 1070.8 ± 143.0a | 845.1 ± 79.8ab | 515.1 ± 529.1b | 410.1 ± 439.3b |
| Alanine | S | 564.9 ± 5.6b | 581.6 ± 55.9b | 531.7 ± 59.5b | 492.9 ± 35.7bc | 423.0 ± 339.9bc | 498.2 ± 455.7bc |
| Valine | B | 123.5 ± 3.5ab | 141.7 ± 7.4a | 115.4 ± 7.1ab | 112.2 ± 4.4ab | 70.6 ± 44.9bc | 85.8 ± 41.9b |
| Methionine | B | 88.7 ± 0.8a | 93.7 ± 1.5a | 84.8 ± 1.3ab | 83.8 ± 0.9ab | 44.7 ± 38.4b | 49.3 ± 36.5b |
| Isoleucine | B | 71.4 ± 2.6ab | 88.1 ± 5.2a | 72.6 ± 4.9ab | 73.8 ± 3.1ab | 43.0 ± 26.6b | 54.8 ± 25.7b |
| Leucine | B | 165.9 ± 2.6ab | 196.0 ± 4.8a | 170.6 ± 4.4ab | 194.3 ± 2.8ab | 105.7 ± 58.5b | 128.9 ± 68.4b |
| Tyrosine | B | 107.5 ± 3.9ab | 155.8 ± 6.7a | 110.5 ± 6.0ab | 115.8 ± 4.0ab | 63.7 ± 42.1bc | 91.4 ± 49.7bc |
| Phenylalanine | B | 63.0 ± 1.7ab | 77.5 ± 3.0a | 62.1 ± 2.6ab | 64.0 ± 1.7ab | 40.4 ± 20.5bc | 51.4 ± 20.1b |
| Histidine | B | 33.6 ± 0.6ab | 42.2 ± 1.1a | 34.6 ± 1.0ab | 33.0 ± 0.6ab | 22.5 ± 10.8bc | 28.0 ± 12.1b |
| Lysine | B | 15.7 ± 4.8b | 15.0 ± 8.5b | 13.4 ± 7.7b | 17.7 ± 5.0b | 30.2 ± 14.7b | 40.6 ± 23.1ab |
| Arginine | B | 100.4 ± 2.5ab | 134.5 ± 4.3a | 105.4 ± 3.8ab | 89.4 ± 2.5b | 67.2 ± 32.9bc | 83.7 ± 33.8b |
| Total FAA | 4628.6 ± 101.1a | 4556.9 ± 198.5a | 4375.2 ± 186.4a | 3893.9 ± 117.9ab | 3732.0 ± 884.8ab | 3645.4 ± 1183.0ab | |
| Sweetness | 1926.5 ± 23.7a | 1793.2 ± 57.4ab | 1843.0 ± 86.5ab | 1562.0 ± 42.6ab | 1100.3 ± 782.0b | 1101.5 ± 735.1b | |
| Umami | 1833.4 ± 119.4ab | 1694.8 ± 235.4b | 1696.9 ± 221.4b | 1450.2 ± 139.8b | 2057.7 ± 330.9ab | 1820.3 ± 546.2b | |
| Bitterness | 769.6 ± 22.3ab | 944.5 ± 40.8a | 769.4 ± 37.2ab | 784.0 ± 24.1ab | 488.2 ± 259.8bc | 614.0 ± 264.3b | |
Abbreviations are the same as given in Table 1.
“Sweetness (S)”, “Umami (U)” and “Bitterness (B)” mean total amount of sweet amino acids (threonine, serine, glycine, and alanine), umami-tasting amino acids (aspartic acid and glutamic acid) and bitter amino acids (valine, methionine, isoleucine, leucine, tyrosine, phenylalanine, histidine, lysine, and arginine), respectively.
Different superscripts in the same row indicate statistical differences (p < 0.05).
Data are expressed as mean ± standard deviation (n = 4).
Bitter amino acids, including valine, methionine, isoleucine, leucine, tyrosine, phenylalanine, histidine, lysine, and arginine are generally considered as endogenous amino acids in roe materials. Comparing within the same company, all 5 companies provided low-priced products containing higher contents (320.8 mg/100 g to 944.5 mg/100 g) of bitter amino acids than those contained in high-priced roe products (253.0 mg/100 g to 769.6 mg/100 g). Uchiumi et al. (2009b) demonstrated that the immature ovaries of Alaska pollack contained greater bitter amino acids content compared to mature ovaries. Therefore, the bitter amino acids content of spicy pollack roe was partly dependent on the maturity degree. Moreover, the low-priced products from companies B and C showed small egg diameters, indicating immature ovaries, and contained higher levels of bitter amino acids. It is also possible that a longer storage period may be another reason for the higher contents of bitter amino acids. Therefore, the content of bitter free amino acids is a potential parameter for the quality evaluation of roe products.
Mechanical properties Firmness (hardness) is the force required to attain a given strain (or deformation) of a material (Szczesniak et al., 1963). As presented in Figure 3, except for roe products from company B, the ovary firmness values of high-priced products (85 – 198 gw) were higher than those of low-priced products (41 – 99 gw). Based on interviews with the manufacturers of roe products, firmness is typically evaluated by touch, and a judgment is made by personal empirical experience. Roe products with higher ovary firmness are considered as superior. The current method in this study for measuring the firmness of roe products can be presented to manufacturers as an objective, convenient, and nondestructive mechanical method to evaluate the firmness of roe products. All of the high-priced roe products showed higher tensile strength values (69 – 213 gw/cm2) of ovary membrane than those of low-priced roe products (24 – 128 gw/cm2). The stronger ovary membranes of the high-priced roe products may be considered as one of the reasons for their higher ovary firmness values compared to low-priced roe products. Except for company B, the gumminess values of ovary round slices of high-priced products (21.1 – 66.2 gw) were higher than those of low-priced roe products (11.7 – 30.5 gw); in contrast, the stickiness values of ovary round slices from high-priced products (8.2 – 43.9 J/m3) were lower than those of low-priced products (13.8 – 50.1 J/m3). The gumminess of a sample is related to the energy required to disintegrate a semisolid food to a state ready for swallowing, and the stickiness is represented by the work necessary to pull the plunger away from the food sample (Szczesniak et al., 1963).
Mechanical properties of spicy pollack roe products. Abbreviations are the same as given in Table 1.
While most commercial roe products are prepared from frozen roe material, those prepared from fresh roe are generally considered superior. Frozen storage causes physical damage to the ovaries, and improper freezing conditions due to slow freezing or extended frozen storage causes greater drip loss from ovaries during thawing (Uchiumi et al., 2009b). The relatively lower values of egg breaking strength of low-priced roe products reflect the higher degree of physical damage during frozen storage. Most drip loss from ovaries during thawing consists of yolk protein from broken fish eggs, which may increase product stickiness. Therefore, high stickiness values indicate a high ratio of broken fish eggs in the ovaries, which is considered one of the reasons for the decline in gumminess value. According to our communications with manufacturers of spicy pollack roe products, low-priced roe products are usually processed from roe materials subjected to an extended period of frozen storage. This may help explain why, within the same company, low-priced roe products showed higher values of stickiness and generally lower values of gumminess.
Almost all high-priced roe products showed significantly greater values of egg breaking strength (18.1 – 29.7 gw) than those of low-priced roe products (4.9 – 20.0 gw). Egg breaking strength, which is mainly affected by processing method, is considered to be an important texture parameter that is closely related to a grainy feel, so called “tsubukan” in Japanese; and roe products with a stronger grainy feel are usually considered to be superior. The higher ovary firmness and gumminess values of high-priced roe products may also be attributed to the greater strength values of fish eggs.
Protein composition and solubility in SDS solution SDS-PAGE patterns of ovary membrane protein composition are shown in Figure 4 (I). The SDS-PAGE patterns of ovary membrane protein differed between high- and low-priced products. Greater high-molecular-weight proteins (> 100 kDa) were observed in the ovary membrane of high-priced products compared to low-priced products, which was especially obvious in the roe products from company B. The ovary membrane proteins from high-priced products showed less solubility in SDS solution (86% – 91%) compared to the low-priced products (91% – 99%). Differences in the protein composition of ovary membranes between high- and low-priced roe products may explain the differences in tensile strength values, as mentioned previously.
SDS-PAGE of ovary membrane protein (I) and eggshell protein (II). Abbreviations are the same as given in Table 1.
SDS-PAGE analysis of eggshell protein is shown in Figure 4 (II). The major eggshell proteins from samples AH, AL, DH, and EH showed a molecular weight of about 100 kDa, while the major proteins in BH and DL were about 50 kDa. Eggshell proteins consisting of both 50 kDa and 100 kDa were mainly contained in CH, CL, and EL. The molecular weight of eggshell proteins in BL was especially low, about 50 kDa or less. In brief, the eggshell of high-priced products contained more high-molecular-weight proteins, compared to the low-priced products within the same company. The solubility values of eggshell proteins in high-priced roe products ranged from 36% to 60%, and they were slightly lower than those of low-priced roe products (38% – 72%).
Cod eggs within the ovary are soft and can be activated to undergo a hardening process when fertilized or transferred to solutions of different ionic composition from the ovary medium (Lönning and Kjörsvik, 1984). Oppen-Berntsen et al. (1990) demonstrated that nonactivated eggshell from cod primarily consists of only 3 protein monomers, α (74 kDa), β (54 kDa), and γ (47 kDa); and a transglutaminase-dependent eggshell protein crosslinking, accompanied by a decline in protein solubility, may be responsible for the egg hardening. During the processing of spicy pollack roe products, protein crosslinking of the eggshell can be activated by soaking the eggs in a salt solution, and the amount of high-molecular-weight eggshell protein may reflect the degree of protein crosslinking. The relatively higher ratio of high-molecular-weight proteins in the eggshell of high-priced roe products may be one reason for their generally higher ovary firmness, gumminess, and egg breaking strength compared to low-priced products from the same company.
Differential scanning calorimetry Thermal transition curve, denaturation temperature, and transition enthalpy of fish eggs are shown in Figure 5. Only one distinct peak was observed in the thermal transition curve for the roe products, likely attributable to the major yolk protein vitellogenin or its derivatives (Schubring, 2004). Fish egg denaturation temperatures of high-priced roe products (88.3°C to 91.2°C) and those of low-priced products (88.5°C to 92.0°C) were similar. High-priced roe products showed higher transition enthalpy (0.7 mJ/mg to 1.7 mJ/mg) than those of low-priced roe products (0.5 mJ/mg to 0.7 mJ/mg), indicating a higher degree of cold-induced yolk protein denaturation of low-priced roe products compared to high-priced roe products, since freezing and frozen storage have a minor influence on the thermal behavior of yolk proteins (Schubring, 2004).
DSC thermograms (thermal transition curve), denaturation temperature (Td) and transition enthalpy (ΔH) of commercial spicy pollack roe. Abbreviations are the same as given in Table 1.
The mechanical and biochemical properties of roe products varied obviously among different companies, indicating that quality standards for spicy pollack roe products are not uniformly applied. Typically, high-priced roe products showed higher values of ovary firmness, gumminess of round slice, tensile strength of ovary membrane, egg diameter and breaking strength, thermal transition enthalpy, and content of high-molecular-weight protein in ovary membrane and eggshell compared to low-priced products. Low-priced roe products usually showed higher values of stickiness of round slice, lipid oxidation (PV and TBARS) and bitter free amino acids content compared to high-priced products. Therefore, these 12 parameters are proposed as potential objective standards for the quality evaluation of spicy pollack roe products.
