Characteristics of Sarcoplasmic Proteins and Their Interaction with Surimi and Kamaboko Gel


  • A. Jafarpour,

    1. Authors are with Food Sciences, School of Applied Sciences, RMIT Univ., Melbourne, Victoria 3001, Australia. Author Jafarpour is also with Dept. of Fishery, Faculty of Animal Science and Fishery, Sari Univ. of Agriculture and Natural Resources, Iran. Direct inquiries to author Jafarpour (E-mail:
    Search for more papers by this author
  • E.M. Gorczyca

    1. Authors are with Food Sciences, School of Applied Sciences, RMIT Univ., Melbourne, Victoria 3001, Australia. Author Jafarpour is also with Dept. of Fishery, Faculty of Animal Science and Fishery, Sari Univ. of Agriculture and Natural Resources, Iran. Direct inquiries to author Jafarpour (E-mail:
    Search for more papers by this author


ABSTRACT:  This study examined the effect of adding common carp sarcoplasmic proteins (Sp- P) on the gel characteristics of threadfin bream surimi and kamaboko while maintaining constant moisture and myofibrillar levels. Based on the temperature sweep test, which is involved in heating of surimi gel from 10 to 80 °C to monitor the viscoelastic properties, at temperature range of 40 to 50 °C, the decrease level (depth of valley) in storage modulus (G′) thermograph was in proportion to the concentration of added Sp- P. Storage modulus (G′) showed greater elasticity after adding Sp- P compared with the control without Sp- P. Furthermore, the breaking force and distance and consequently gel strength of the resultant kamaboko were improved significantly (P > 0.05). Thus, added Sp- P did not interfere with myofibrillar proteins during sol–gel transition phase but associated with textural quality enhancement of resultant kamaboko; however, addition of Sp- P from the dark muscle of the carp decreased the whiteness of the resultant surimi. Furthermore, according to the SEM micrographs, the gel strength could not be associated with either the number of polygonal structures/mm2 or the area of the polygonal structures in the kamaboko gel microstructure.


The significance of sarcoplasmic proteins (Sp- P) on enhancement or reduction of surimi and kamaboko gel characteristics has been examined; however, there does not seem to be a consensus among authors as to whether the Sp- P contribute positively to the gelling characteristics of surimi. Some researchers believe that the presence of Sp- P will cause a weakening of the gel (Okada 1964; Kudo and others 1973; Nakagawa and others 1989). The increased presence of Sp- P is cited as the main cause for weaker gels from fish with a higher degree of dark muscle compared with gels from fish that has a lower degree of dark muscle (Nakagawa and others 1989). On the other hand, some other researchers believe in the enhancive effect of Sp- P on the viscoelstic properties of resultant kamaboko, either by retaining Sp- P or adding them back to the surimi gel matrix (Macfarlane and others 1977; Nakagawa and Nagayama 1988; Morioka and Shimizu 1990; Park and others 2003; Kim and others 2005a). The inhibitory or enhancive role of Sp- P on the gel characteristics of surimi and kamaboko has been reviewed by Gorczyca and Jafarpour (2008). This study was conducted to examine the role of added SP- P on the physiochemical characteristics of surimi and kamaboko gel including color, and specifically thermodynamic behavior of proteins during sol–gel transition, gel strength, protein composition, and microstructure.

Materials and Methods


Female common carp (Cyprinus carpio) was harvested by the K & C Fisheries Co. from the Gippsland Lakes in Victoria, Australia. Fresh whole fish was obtained from the Victorian wholesale fish market and transported in crushed ice to the RMIT University's pilot plant. After gutting, heading, and filleting, the dark fillets of common carp were separated and minced using a bench meat mincer with a mesh size of 3 mm. The mince was vacuum-packed and stored at 4 °C for a maximum of 24 h before Sp- P extraction. Frozen threadfin bream surimi was obtained from Austrimi Seafood Co. (Victoria, Australia) and kept at −20 °C.

Sarcoplasmic protein extraction

To extract Sp- P, fish mince (100 g) was mixed with cold (≤ 4 °C) 0.1 M phosphate buffer solution (pH 7.0) in a ratio of 1: 10, respectively, and homogenized in a Waring blender (Waring commercial, Model 32BL80 [8011], Torrington, Conn., U.S.A.) at low and high speed, each for 60 s. The paste was then centrifuged at 10000 ×g (BECKMAN J2-21M/E) for 20 min at 4 °C. The supernatant was decanted into plastic tubes (50 mL), frozen with liquid nitrogen (< −150 °C) followed by freeze-drying (DYNAVAC Freeze dryer, model FD 12, Sydney, Australia) functioning under vacuum (1.33 to 13.3 Pa) at −80 °C. The freeze-dried Sp- P powder was stored at 4 °C until added into the threadfin bream surimi.

Preparation of surimi gels with added Sp- P

To prepare a gel, frozen threadfin surimi was cut into smaller pieces (approximately 2 × 2 × 2 cm) which were then defrosted at room temperature (approximately 22 °C) for about 1.5 h to defrost. The surimi pieces were further chopped with a Sunbeam Oskar II food processor (Household Appliance Sales and Service, Niles, Ill., U.S.A.) for about 60 s to create a homogeneous paste. After addition of NaCl (2%) as powder, chilled water (2 to 4 °C [to maintain the temperature at 4 to 10 °C]) was sprinkled over the mince to adjust the moisture content of the paste to 85%. Blending continued for another 60 s.

Sp- P was added to the surimi paste at concentrations of 0%, 5%, 10%, 20%, and 35% based on the total protein content while the myofibrillar proteins (Mf-P): Sp- P ratio was maintained at 3: 1 or 65%: 35%. The moisture content was maintained at 85%.

Kim and others (2005b) have shown that sucrose does not interfere with surimi gel formation process. For treatments with added Sp- P lower than 35%, sucrose (D-Sucrose from Fluka Chemie GmbH, Zwijndrecht, The Netherlands) was used to maintain the mass balance (Mf-P: Sp- P) without altering the Mf-P or moisture content. Sp- P freeze-dried powder was sprinkled over the mince at the same time as water adjustment was made as described previously.

After removing a sample (30 g) for a temperature sweep test, the remaining paste was then compressed into stainless steel tubes (length [L] of 20 cm, and diameter [Ø] of 2.5 cm), which had previously been sprayed with canola oil to lubricate the inner surface of the tube. Both ends of each tube were then sealed with screw-thread caps. To obtain low temperature setting (Lanier 1992), the surimi in the tubes were refrigerated (4 °C) overnight (16 to 18 h). Then, for making kamaboko, the tubes were placed in a water bath (steam-jacketed kettle) preset at 90 ± 2 °C for 30 min. Immediately after heating, the tubes were cooled in ice-water to approximately 10 °C to stop any further effect of heat on the texture. Once cooled, the gels were removed from the tubes with a plunger and sliced to required dimensions for analysis of large-scale texture characteristics (puncture test).

Color evaluation

The reflected color of samples was measured using a Minolta CR-100 Chroma meter (Konica Minolta Sensing Inc., Osaka, Japan); a tristimulous color analyzer. In this system, the degree of lightness, redness, greenness, yellowness, or blueness are represented by L*, +a*, −a*, +b*, −b*, respectively, and are reported as L*, a*, and b* values (Lanier 1992). The instrument was calibrated using a standard white tile (L* value of 98.46, a* value of 0.0, and b* value of 2.18) placed under the orifice of the instrument. The color of each treatment was recorded at 3 spots per sample. The efficacy of each treatment to improve the color of surimi was determined by calculating whiteness (Park 1994; Luo and others 2004), thus:


Measurements of texture properties

Temperature sweep test Dynamic rheological properties of the surimi pastes were monitored by a Rheostress RS50 rheometer (HAAKE, Germany) using a plate (MP60 steel 18/8) and cone (60 mm Ø/2° angle) by running a temperature sweep test. Surimi paste (approximately 5 g) was loaded in the gap (0.105 mm) between the cone and plate. During the temperature sweep test, the surimi sample was heated from 10 to 80 °C at a heating rate of 1 °C/min. The stress (100 Pa) and frequency (0.1 Hz) values used for the temperature sweep test were selected from within the range of the linear viscoelastic response at both 10 and 80 °C.

Dynamic rheological parameters: storage modulus (G′) representing elastic properties, loss modulus (G″) representing viscous characteristics, and phase angle (δ) were determined. Phase angle is calculated from tan−1 (G″/G′) and can range from 0°, with the material being totally elastic, to 90° when the material is regarded as being totally viscous (Kim and others 2005a).

Puncture test After removal from the stainless steel tubes, the gels were sliced transversely into 25-mm pieces and tempered to room temperature before the puncture test. Puncture test was carried out. Puncture tests were performed using a texture analyzer (TXAT2, Stable Micro Systems Ltd., Surrey GU7 1YL, U.K.) equipped with a spherical-ended stainless steel probe (Ø= 5 mm). Breaking force (g) and breaking distance (mm) were determined (Kim and others 2005a). Five replicates were carried out for all determinations.


The protein composition of samples was analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE). Samples were collected from (1) supernatant of common carp dark-muscle homogenate after centrifugation, (2) freeze-dried Sp- P powder after solubilization, and (3) threadfin bream surimi after adding Sp- P (0% to 35%), before making kamaboko. After protein quantification procedure following BCA method (Bollag and Edelstein 1991a), by using a buffer phosphate solution (pH = 7.0), the protein content of samples was diluted to approximately 3 mg/mL and stored at −80 °C. A pre-cast gradient gel (Invitrogen, NuPAGE® 4% to 12% Bis-Tris Gel 1.0 mm × 12 well [Cat Nr NP 0322BOX]) was used. After thawing, an aliquot (20 μL) sample was mixed with 5 μL sample buffer (Bollag and Edelstein 1991b) and after heating in a hot water bath at 100 °C for 5 min, an aliquot (15 μL) of the sample was loaded by a micropipette into a slot or well onto the gel. An unstained protein ladder (PageRuler™ Unstained Protein Ladder [Fermentas ≠ SM0661]) ranging from 10 to 200 KDa was used as the marker. The gel was placed into a tank (XCell SureLock™, Invitrogen) which was filled with “MOPS” gel buffer (Bollag and Edelstein 1991), before being connected to a power pack (Bio-RAD Power PAC 3000) to supply a voltage from 85 to 120 V with 80 to 90 mA current during approximately 90 min. The gel was then stained with Coomassie Blue solution for approximately 45 min followed with destaining with an acid solution (10% glacial acetic acid, 10% methanol, and 80% Milli-Q water) for approximately 2 h.

Scanning electron microscopy (SEM)

SEM with a cryo-transfer device (Alto 2100, Gatan, OX14 1RL, U.K.) was used to study the internal microstructure of the kamaboko samples. Each specimen was formed or cut into a conical shape (approximately 2 × 2 × 2 mm, length × width × height) with a Stanley knife, fixed by an agar glue to the golden stage and then frozen (to −180 °C) by liquid nitrogen in the slushing station once the liquid nitrogen was converted to slush. To analyze the inner structure of the gel matrix, the tip of the specimen in the preparation chamber was fractured by a blade and then transferred to the adjacent chamber, called the SEM cold stage, for sublimation at −90 °C for approximately 5 min. Using Argon sputtering for 90 s, the specimen was coated with gold after being returned to the preparation chamber. Finally, the gold-coated specimen was placed back into the SEM cold stage for scanning under 30 kv and 5.0 spot value.

The electron micrographs were analyzed by ImageJ program (ImageJ 1.34s, Wayne Rasband, Natl. Inst. of Health, Bethesda, Md., U.S.A.) for the number of polygonal structures/mm2 and the area of the polygonal structures in the kamaboko gel matrix.

Statistical analysis

Texture data were analyzed by SPSS ver. 15.0 (SPSS, Inc., Chicago, Ill., U.S.A.) and significant differences between mean values at 95% confidence level were determined by one-way analysis of variance (ANOVA) and LSD test.

Results and Discussion

Effects of Sp- P on color

Addition of Sp- P powder into surimi gel decreased the L* value while increased the a* and b* values, significantly (P < 0.05), compared with the control samples (Table 1). For instance, by adding 5% Sp- P, the whiteness of threadfin bream surimi was reduced by approximately 33% with a shift in color (ΔE) from the control by approximately 8 units. As the concentration of added Sp- P increased the whiteness of the surimi decreased further. With 35% added Sp- P, the whiteness of threadfin bream surimi decreased by approximately 77%, and ΔE was recorded at approximately 20, compared with the control. As Sp- P from the dark muscle of female common carp has a dark red color (photos not shown), it was expected that after adding these proteins into the threadfin bream surimi, the whiteness of resultant surimi would decrease. Interestingly, Joo and others (1999) found that 71% of the variation in lightness of pork loin muscle can be explained by Sp- P solubility. It is widely accepted that mainly myoglobin (Mb) and partially hemoglobin (Hb) and their chemical products are responsible for muscle color (Joo and others 1999).

Table 1—.  Color parameters of threadfin bream surimi before and after adding Sp- P.a
Addition ofColor values (Mean ± SE)WhitenessColor deviation (ΔE)
  1. aSp- P was extracted from the dark muscle of common carp by homogenization and centrifugation at 10000 ×g for 20 min at 4 °C, collection of the supernatant, which was freeze dried and stored at 4 °C until required. Then Sp- P in powder form was blended with threadfin bream surimi from 5% to 35%.

  2. SE = standard error; NA = not applicable.

  3. bSamples to which sucrose (35%) was added but no Sp- P were taken as a control.

  4. Different superscripts in the same column indicate significant difference (P < 0.05) according to a one-way ANOVA and LSD test.

NilNil55.32 ± 0.22z−0.62 ± 0.16z−2.64 ± 0.09z  63.24 ± 0.36z2.97 ± 0.88z
Nil35% (control)b56.11 ± 0.46z−0.52 ± 0.23y−3.05 ± 0.11y  65.26 ± 0.26yNA
5%30%48.84 ± 0.48y0.30 ± 0.15x2.40 ± 0.18x41.64 ± 0.76x9.18 ± 0.89y
10%25%45.90 ± 0.55x1.30 ± 0.24w4.62 ± 0.17w32.04 ± 0.31w2.98 ± 0.88x
20%15%43.56 ± 0.41w4.04 ± 0.42v8.32 ± 0.28v18.60 ± 1.03v17.58 ± 0.85w
35%0%39.82 ± 0.95v5.76 ± 0.42u8.46 ± 0.39v14.44 ± 1.10u21.03 ± 1.61v

Our results are supported by Kim and others (2005b) who investigated the effect of Rockfish Sp- P on Alaska Pollock surimi, demonstrated that the negative impact of added Sp- P (2%) on the color of the resultant surimi was mainly due to an increase in b* values. Similarly, Park and Park (2007) stated that by adding extracted Pacific whiting myoglobin (Mb [0% to 1%]) into Pacific whiting surimi, the whiteness (L*− 3b*) of the resultant kamaboko was reduced from approximately 74 to 69 which the authors attributed to the added Mb pigments.

Park and others (2003) studied the effect of Sp- P (0% to 9%) on the whiteness of heated gels prepared by the alkaline-aided method from croaker. According to their results, there was no difference in whiteness between the control and the samples with 1% Sp- P as both had a whiteness value of 62. Even after adding 9% Sp- P, the whiteness of croaker surimi decreased only to approximately 52 (16% reduction), possibly due to an increase in b* values. Considering that these researchers extracted Sp- P from Jack mackerel (a species with a relatively high ratio/content of dark muscle) and added to the croaker (a species with a low content of dark muscle) surimi, more reduction in whiteness is expected especially at higher level of added Sp- P (9%).

Texture properties

Temperature sweep test Oscillatory dynamic properties of threadfin bream surimi, without and with added Sp- P were tested by a temperature sweep test (Figure 1). The control (35% sucrose) and the surimi without either added Sp- P showed very similar thermodynamic patterns during linear (1 °C/min) heating from 10 to 80 °C (Figure 1). It confirms that sucrose does not interfere with the gel forming process as shown by Kim and others (2005b).

Figure 1—.

Storage modulus (G′) of threadfin bream surimi during the temperature sweep test from 10 to 80 °C at 100 Pa stress and 0.1 Hz frequency, before and after added Sp- P.

Initial storage modulus (G′) was 2.3 to 2.4 KPa for surimi samples tested irrespective of treatment. By increasing the temperature (up to 40 °C), a slight increase in G′ was recorded from approximately 3.3 (lowest value) to 3.8 (highest value) KPa for the control surimi and surimi with 20% added Sp- P, respectively. After this stage, G′ dramatically decreased and the depth of the G′ graph was inversely associated with the concentration of added Sp- P. Surimi with added Sp- P underwent a smaller reduction in G′ at a lower temperature compared with the control treatment at this stage. For example, the lowest G′ and the corresponding temperature for the control treatment were recorded at approximately 1.4 KPa and 50 °C, respectively; whereas for surimi with added Sp- P (35%) these parameters were recorded at approximately 2.9 KPa and 46 °C (Figure 1).

Furthermore, G′ values at the end of kamaboko stage, when the proteins are mostly aggregated and surimi gel has turned to a firmer and more elastic gel matrix, also were affected by the concentration of added Sp- P. At 80 °C, G′ increased from approximately 3.3 KPa for the control sample which had no added Sp- P (but 35% sucrose) to approximately 12.3 KPa for surimi with 35% added Sp- P; approximately 273% improvement in the G′ of the resultant kamaboko gel (Figure 1). The oscillatory dynamic properties during the temperature sweep test for samples with an additional 20% or 35% Sp- P were not significantly (P > 0.05) different which indicated that about the same amount of stored energy is available in their resultant kamaboko gel networks.

These temperature sweep results are not in agreement with those of Kim and others (2005b) who reported that adding rockfish Sp- P at different concentrations (1% to 5%) into Alaska Pollock surimi paste notably lowered the G′ at the initial temperature (10 °C) as well as increased the depth of the G′ curve at the modori stage, (which occurred at 45 °C) compared with the control (no added Sp- P). Furthermore, the same researchers found that G′ value at 75 °C (gelation completed) was unaffected by the added Sp- P (1% to 5%) whereas Park and Park (2007) pointed out the adding myoglobin (Mb) into the Pacific Whiting surimi affected only the kamaboko stage, negatively. There are at least 3 possible explanations for the different trend in small-scale deformation, over an added Sp- P range of 0% to 5% for the 3 studies, namely, species, moisture content, and Mf-P differences.

The more likely explanation is the difference in the moisture and Mf-P content of the surimi gels studied: 78%, 80%, and 85% moisture levels for Kim and others (2005b), Park and Park (2007), and this study, respectively. Possibly the stronger gels (lower moisture) are less “responsive” to added Sp- P (0% to 5%). The Mf-P content appears to be handled differently in the 3 studies—in this study, it was kept constant (approximately 16.25%) whereas in 2 other studies (Nowsad and others 1995; Kim and others 2005b) it seems that the Mf-P level decreased as Sp- P increased. The decreasing Mf-P level (with added Sp- P) would result in less elasticity and therefore explain the decrease in the initial G′ value as well as the increase in the depth of the modori phase (Kim and others 2005b). In Park's study (Park and Park 2007), the approach used for addition of Sp- P was based on the surimi weight (w/w) not Mf-P content. Overall, the significance of the moisture level and the Mf-P content on the effect of added Sp- P on the gel quality is unclear. Therefore, further studies are required to clear these issues.

The temperature at which an increase and decrease occurred in G′ was not affected by the concentration of added Sp- P (Figure 1) which is in agreement with the results of Ko and Hwang (1995). These researchers used 2 setting times (20 min and 2 h) and heated the milkfish meat (with 10 mg/g Sp- P) at 90 °C to examine the trend during thermal gelation. They found no displacement in the temperatures at which suwari (40 to 50 °C) and modori (60 to 70 °C) formed with the addition of Sp- P (Ko and Hwang 1995).

Effect of Sp- P on gel strength Threadfin bream kamaboko to which common carp Sp- P had been added, had a greater puncture force, braking distance, and a significantly greater gel strength compared to the control (Table 2). In the case of control treatments, the samples were very weak and as the trigger force of texture analyzer was set at 10 g, the samples were destroyed under the plunger (approximately 9 g) before the puncture force reached the trigger point. The possible explanation was that high moisture content (85%) of the control surimi samples resulted in a very weak kamaboko gel. However, after adding Sp- P, the gel strength of the kamaboko (proportionally) increased as the added Sp- P concentration increased. For example, by adding only 5% Sp- P, the gel strength was improved by approximately 300% compared with the control (Table 2), whereas after adding 35% Sp- P, the gel strength increased to approximately 88 which was 388% greater than the gel strength of the surimi with 5% added Sp-P (Table 2). The positive effect of added Sp-P on the surimi gel was reported by Morioka and Shimizu (1990) who carried out an experiment involving the addition of Sp-P from Pacific mackerel into threadfin bream myofibril at a ratio of 1:3 (Sp-P: Mf-P). The resultant kamaboko (Table 3) had greater (ca. 54%) gel strength than that of the control (with no added Sp-P). Although these results demonstrated a positive effect on gel formation by Sp- P, whether Sp- P enhances or reduces the gel strength of kamaboko is still being debated. The evidence for the role of Sp- P in gel formation and consequently gel strength was reviewed by Gorczyca and Jafarpour (2008).

Table 2—.  Puncture test values (mean ± SE) of threadfin bream kamaboko before and after adding Sp- P.a
Addition ofBreaking force (g)Break distance (mm)Gel strength (
  1. SE = standard error.

  2. aAs per Table 1.

  3. bBreaking force of these samples was lower (observed to be 9 g) than the trigger force (set at 10 g) of texture analyzer. Thus the gel strength is an estimate and not greater than the value cited.

  4. Different superscripts in the same column indicate significant difference (P < 0.05) according to a one-way ANOVA and LSD test.

Nil35% (control)9b0.54.5
5% Sp- P30%11.55 ± 0.33z1.57 ± 0.10z18.08 ± 0.84z
10% Sp- P25%11.84 ± 0.64z1.94 ± 0.19z23.28 ± 3.22z
20% Sp- P15%17.99 ± 0.33y2.61 ± 0.16y47.14 ± 3.68y
35% Sp- P0%26.60 ± 1.20x3.30 ± 0.14x88.32 ± 7.27x
Table 3—.  Effect of native and heat-coagulated Sp- P on gel-forming of threadfin bream myofibrils (adapted from Morioka and Shimizu 1990).
AdditivesPuncture force (g)Puncture depth (cm)Gel strength (
  1. aWater was added to myofibril (84% moisture, 5% sucrose) instead of Sp- P.

  2. bNative Sp- P added to gel (Mf-P: Sp- P ratio of 3: 1).

  3. cHeat coagulated Sp- P added to gel (Mf-P: Sp- P ratio of 3: 1).

  4. Sp- P was prepared from Pacific mackerel by homogenizing the dorsal muscle with 5 volumes of phosphate buffer (I = 0.05, pH = 7.0) and concentrating the extract obtained to 85% water content. Myofibrils were prepared from threadfin bream by homogenizing the dorsal muscle with 5 volumes of buffer (0.09 M KCl, 5 mm EDTA, 0.039 M borate buffer, pH = 7.0), washing with the same buffer solution 4 times, dehydrating to 88% water content, and adding 5% sucrose.

+Sp- Pb67.20.5637.6
+DSp- Pc39.60.4317.0

SDS–PAGE profile of Sp- P

To determine the purity of extracted Sp- P from common carp dark muscle, samples were analyzed on SDS–PAGE analysis (Figure 2). Nine protein bands were observed for both fresh and freeze-dried Sp- P samples. The major bands were at approximately 10, 23, 26, 27, 38, 43, 50, 60, and 97 KDa (Figure 2, lanes 2 and 3).

Figure 2—.

SDS–PAGE of common carp Sp- P and threadfin bream myofibrilla proteins. Lane 1 is the protein marker, lanes 2 and 3 are the protein profile of common carp Sp- P, before and after freeze drying, respectively, and lanes 4 to 8 are the protein profile of threadfin bream surimi with no added Sp- P, 5%, 10%, 20%, and 35% added Sp- P, respectively.

These bands are similar to those from Nakagawa and others (1988) who analyzed Sp- P from muscle of 16 different fish species and found approximately 10 bands for carp: ≤ 23, 25, 26, 33, 35, 40, 43, 49 to 51, 60, and 94 KDa. Furthermore, in another study conducted by Kristinsson and others (2005) using channel catfish, about 11 bands were observed on the SDS–PAGE of the water from the washing cycles of the conventional method: 6.5, approximately 15, 3 polypeptides between 23 and 29, approximately 36, 43, 50, 60, 84, and 97 KDa.

Apart from some Sp- P such as those with molecular weights of 23, 27, 38, and 43 (these bands are in a similar position to that of Mf-P, see Figure 2, lanes 2 and 3 compared with lane 4), the rest of the bands were obvious on the SDS–PAGE profile especially as the percentage of added Sp- P increased. As expected, there was no major shift in the protein profiles of threadfin bream surimi before (Figure 2, lane 4) and after (Figure 2, lanes 5 to 8) adding common carp Sp- P (5% to 35%).

This result is supported by Morioka and Shimizu (1993) who identified more than 8 bands (≤ 23, 25, 26, 35, 40, 43, 55, and 94 KDa) of Sp- P for 8 fish species, including carp and noted that the distribution of protein from the wash water was not related to the fish species being examined as they all showed the same SDS–PAGE patterns. Morioka and Shimizu (1993) also examined the relationship between specific Sp- P in some fish species and the gel strength of the Sp- P gel and they found about a 0.80 correlation coefficient between greater gel strength of Sp- P gels with (1) increasing heat coagulability of specific Sp- P and (2) Sp- P proteins with molecular weights of 94, 40, and 26 KDa. However, the same researchers reported only a 0.44 correlation coefficient with Sp- P with molecular weight of 35 KDa. This was in agreement with results of a study conducted by Nakagawa and others (1989) who pointed out that some Sp- P such as glyceraldehyde phosphate dehydrogenase (GAPDH) with a molecular weight of 36 KDa as well as aldolase (ALD [160 KDa]), reduced the gel strength of kamaboko.

Although the effect of specific Sp- P on gelling was not addressed in the article, the authors recommend that further work needs to be done to confirm that Sp- P distribution is not species dependant as suggested (Morioka and Shimizu 1993) and that some Sp- P are detrimental to gelling while other Sp- P can assist gelling (Nakagawa and others 1989).

Kamaboko gel microstructure

Scanning electron microscopy (SEM × 1000) was carried out to analyze the inner microstructure of kamaboko samples (Figure 3). In our study, all treatments produced kamaboko samples that exhibited an orderly matrix with cross-linking to form polygonal structures. However, there was no discernible trend in terms of number and area of polygonal structures/mm2 for the different treatments (data not shown), other than the polygonal structures/mm2 being denser in the center of a sample than at the corners of each specimen rather than being evenly distributed throughout the specimen.

Figure 3—.

Scanning electron micrograph (SEM × 1000) of threadfin bream surimi before (B1, no Sp-P or sucrose addition; B2, only 35% sucrose) and after the addition of Sp- P (B3, 5%; B4, 10%; B5, 20%; B6, 35%).

To explain the microstructure of sardine surimi, subjected to different setting times and temperatures, Alvarez and others (1999) examined only the outer surface of the suwari and kamaboko gel. These researchers associated the optimum kamaboko texture “to a fibrillar microstructure with continuous, clearly contoured fibers or fiber bundles” observed in the surimi gel, which was set at either 35 °C for 60 min or 40 °C for 30 min. This description (Alvarez and others 1999) could also be meaningful for the SEM obtained in this study (Figure 3).

Kubota and others (2003) stated that it is difficult to expound the internal gel microstructure of fish mince using conventional SEM. Thus, they (Kubota and others 2003) suggested using VP (variable pressure)-SEM as a desirable technique for observing the broken (either with a razor or by hand) surface of a kamaboko gel to reveal 3-D microstructure. The same researchers categorized the kamaboko gel microstructure as “porous structures, network structures, and flat structures.” The authors of this article agree that obtaining meaningful micrographs using conventional SEM is difficult. In our study, the addition of cryotransfer to the SEM technique was used and resulted in continuous contoured fiber bundles creating both porous and network structure (polygonals) being observable at ×1000 magnification (Figure 3).

Based on the large-scale and small-scale texture results, binding occurred between myofibrillar proteins and Sp- P. However, no association could be found between (1) the area or (2) the number of polygonal structures/mm2 and textural characteristics such as gel strength of the resultant kamaboko. Surprisingly, kamaboko gels magnified to ×4000 showed an increase in the thickness and depth of the polygonal structures at higher concentrations of added Sp- P, specifically at 20% and 35% (Figure 4).

Figure 4—.

Scanning electron micrograph (SEM × 4000) of threadfin bream surimi before (B1, no addition of Sp-P and sucrose) and after addition of Sp-P (B2 5%, B3 20%, and B4 35%).

The changes in polygonal structures as a result of added Sp- P may be attributed to cross-linking between Sp- P and Mf-P. As mentioned previously, Morioka and Shimizu (1992) associated improved gelling in threadfin bream surimi to the binding of Sp- P to Mf-P. Nowsad and others (1995) also related the improvement in the gel strength of the suwari gels to the presence of Sp- P specifically transglutaminase which mediate (catalyzed) the cross-linking of myosin heavy chain (MHC) during setting.

Further study is required to determine how the analysis of SEM can be used to qualitatively and quantitatively explain the effect of adding Sp- P to surimi on the textural parameters (such as gel strength) of the resultant kamaboko to be able to associate the gel strength with the nature or level of cross-linking in the kamaboko gel.


The whiteness of surimi decreased significantly with added Sp- P mainly due to an increase in b* value and this was not unexpected as the Sp- P added was isolated from the dark muscle of common carp. As utilization of these proteins (Sp- P) rather than improving the gel characteristics of surimi and kamaboko gel has an environmental advantage, a process to improve the color of isolated Sp- P from dark muscle of fish needs to be developed.

The results of both small-scale and large-scale deformation showed that addition of freeze-dried Sp- P powder improved the gel properties of the resultant surimi and kamaboko. Such an enhancement of the textural quality could be associated with the added Sp- P. In addition, it was concluded that Sp- P are not interfering with kamaboko gel formation during the sol–gel transition phase; however, based on SEM photos (×1000), no specific physical interaction between the added Sp- P and Mf-P was observed. In other words, there was no correlation between added Sp- P concentrations and number of polygonal structure/mm2 in the resultant kamaboko gel network. Thus, it can be hypothesized that such an improvement in the gel properties of surimi and kamaboko gel was the consequence of enzymatic activity of endogenous transglutaminase, which catalyzes the cross-bridging of protein–proteins bounds. By understanding both the chemical nature of the Sp- P: Mf-P interaction and physical manifestation of these interactions at the SEM level, it may be possible to optimize these interactions for the benefit of the gel quality. Further studies are needed to clarify this issue.