Full text of DOI:10.1111/j.1365-2621.2001.tb15205.x

JFS: Food Chemistry and Toxicology
Quantitative Estimation of Dark Muscle Content
in the Mackerel Meat Paste and its Products
Using Antisera Against Myosin Light Chains
Y. OCHIAI, L. OCHIAI, K. HASHIMOTO, AND S. WATABE
ABSTRACT: Based on the high level of extractability of myosin subunits (light chains), even after prolonged heat
treatment of muscle, a new method to evaluate the dark muscle content in the fish meat and products of mackerel
is proposed. Tissue-specific rabbit antisera with myosin light chains (A1 from ordinary muscle and D1 from dark
muscle) from mackerel Scomber japonicus were obtained. Mackerel meat paste (surimi) was dissolved in 8 M urea
containing 1% SDS, and diffused on agar plates containing antiserum against A1 or D1 by single radial immunod-
iffusion (SRID). The results obtained showed that the area of halos formed in the plates was quite proportional to the
content of dark muscle.
Keywords: antiserum, dark muscle, fish, kamaboko, mackerel, myosin light chain, quantification, surimi.
Introduction
animal species identification because of its species-specific
ISH MUSCULATURE IS COMPOSED OF ORDINARY (WHITE composition and mobility on electrophoretic gels (Seki and
skeletal) muscle which generally occupies the major part, others 1980). Ordinary and dark muscle myosins of fish also
F
and dark (red skeletal) muscle. Dark muscle lies mostly along have tissue-specific light chains. Mackerel ordinary muscle
the lateral line of a fish and is used for constant cruising at myosin contains 3 kinds of light chain, namely, alkali light
low speed (Bone 1966; Tsukamoto 1981), corresponding to chains (A1, A2) and DTNB light chain in decreasing order of
slow skeletal muscle. Actually, the ATPase profiles of myo- molecular weights, while dark muscle myosin has 2 kinds, D1
fibril and myosin, as well as the oxidative function (high and D2 in decreasing order of size (Watabe and others
myoglobin content and mitochondria), are reminiscent of 1983b). As previously reported, A1 and D1 light chains from
slow muscle (Watabe and Hashimoto 1980; Altringham and ordinary muscle and dark muscle myosins, respectively, do
Johnston 1982; Watabe and others 1983a). The proportion of not cross-react with each other in immunodiffusion tests
dark muscle to the whole muscle varies less than 1 to 30% (Watabe and others 1983b). It follows then that the intensity
depending on fish species (Greer-Walker and Pull 1975). Dark of the precipitation line is solely attributable to the amount
muscle, however, after death, deteriorates faster than ordi- of homologous myosin.
nary muscle during storage and processing, because of its
high content of myoglobin and trimethylamine oxide muscle in fish mince have been proposed based on hemo-
Obatake and others 1988).
protein (myoglobin, hemoglobin) content (Takeda and oth-
In order to prepare high-quality meat paste (surimi) and ers 1980). However, these are not applicable to meat paste
Methods to estimate the proportion of contaminated dark
(
process it into a fish cake (kamaboko) of higher gel strength and kamaboko, since appreciable amounts of hemoprotein
and better whiteness, it is necessary to remove dark muscle are lost through washing (leaching) of minced muscle. This
as much as possible. However, in the case of red-fleshed fish study proposes an immunochemical method for the quanti-
such as mackerel and sardine, abundant dark muscle is unin- tative estimation of dark muscle in mackerel surimi and ka-
tentionally mixed with the product because of the technical maboko, based on a single radial immunodiffusion test using
limitations, thus causing adulteration in products. Attempts antisera against muscle-specific myosin light chains.
to remove dark muscle with a meat separator at higher levels
in processing resulted in lower yield of surimi and higher
costs of the products. Hence, it is important to estimate the
amount of mixed dark muscle in surimi in advance to con- Materials
Materials and Methods
trol the quality of product.
For myosin and myofibril preparations, live specimens of
Recently, meat species identification, including that of mackerel Scomber japonicus (520 g average body weight)
fish, has seen great developments through progress in DNA were decapitated after catch, transported to our laboratory
technology such as the polymerase chain reaction method in dry ice, and kept frozen at –80 ЊC until use. For mackerel
(
PCR) (Takagi and Taniguchi 1995; Ram and others 1996; meat products, commercially available specimens were used.
Quinteiro and others 1998; Bossier 1999; Partis and others
000). These methods have become useful tools for both raw Preparation of proteins
and cooked muscle identification. However, several prob-
Myosins were prepared according to Stafford and others
2
lems still remain; for example, the meat to be examined (1979) from ordinary and dark muscles of mackerel, except
should be from a single species, and determination of adul- that ammonium sulfate fractionation was carried out in the
terating rate is difficult.
range of 40 to 50% saturation instead of 40 to 70% saturation.
Myosin light chains were once chosen as a parameter for A1 and D1 light chains were isolated by a method that uses
©
2001 Institute of Food Technologists
Vol. 66, No. 9, 2001—JOURNAL OF FOOD SCIENCE 1301

Estimation of Dark Muscle Content in Mackerel Surimi . . .
5,5’-dithio(2-nitrobenzoic acid) (DTNB) and urea treatments, were dissolved in 10 volumes of 0.01 M sodium phosphate
and DEAE-cellulose column chromatography according to buffer (pH 7.2) containing 8 M urea, 1% SDS, 1% 2-mercap-
Gazith and others (1970), Weeds and Lowey (1971) and toethanol, and 0.1 mM phenylmethanesulfonyl fluoride
Lowey and Holt (1972) except that the slope of linear gradi- (PMSF), and allowed to stand overnight at room tempera-
ent (potassium phosphate concentration) was in the range of ture. The supernatant was exhaustively dialyzed against 0.01
50 mM to 0.4 M.
M sodium phosphate buffer (pH 7.2) containing 0.1% SDS,
Myofibrils were prepared according to Yang and others 0.1% 2-mercaptoethanol and 0.1 mM PMSF to remove urea,
(1970) and were finally suspended in 0.16 M KCl.
and dialyzed against the same buffer without SDS.
Processing procedures
Preparation of antisera
A flow chart of kamaboko processing is shown in Figure 1.
Antisera against A1 and D1 light chains were raised in
Minced meat of mackerel was sampled from a meat separator adult female rabbits as previously reported by Watabe and
and washed with cold water. Water from the washed mince others (1983b). The sera were kept frozen at –80 ЊC until they
was removed with a screw press. Then, the mince was passed were used.
through a refiner. At this step, the tough meat, together with
skin and bone, was removed. The refined meat was ground Analytical methods
with 5% sugar and 1% polyphosphates and the resulting meat
A double immunodiffusion test was carried out according
paste was used as raw material for kamaboko. At the stage of to Ouchterlony (1958). SDS-PAGE was performed in 15% poly-
mince preparation, the pressure of meat separator was manu- acrylamide slab gels containing 0.1% SDS. Single radial immu-
ally adjusted to a high, medium, or low level, although the ma- nodiffusion (SRID) was carried out according to Fahey and
chine was not equipped with a pressure gauge. However, the 3 Mckelvy (1965) and Mancini and others (1964). A 1.2% agar in
pressure levels were defined as follows: high level in which phosphate buffered saline solution (0.07% KH PO , 0.12%
2
4
meat was almost completely removed from the skin; medium Na HPO , 0.68% NaCl, and 0.1% NaN , pH 7.0) was preincu-
2
4
3
level in which a small amount of meat remained on the skin bated at 50 ЊC, and then antiserum (anti-A1 or anti-D1) was
together with tiny bones and fat layer; low level in which most added to a level of 1% and the mixture was poured onto a
of the ordinary muscle was removed and about half of the glass plate, pre-coated with a thin film of 1.2% agarose and al-
dark muscle remained on the skin. The higher the pressure, lowed to dry, to make a precisely 1-mm-thick layer. Wells of 3
the higher the yield and dark muscle content.
mm in diameter were punched at 2-cm intervals. Then, about
␮l of antigens (solubilized samples prepared as described
5
Solubilization of protein in meat paste and kamaboko above) to be tested were added to the wells. The plate was
Ordinary and dark muscles, myofibrils and the products then incubated at 37 ЊC for 48 h in a humidified box to prevent
dehydration. The diameter of the halo formed around the well
Figure 2—SDS-PAGE patterns of solubilized samples. Ab-
breviations: 1, ordinary muscle myosin light chains; 2, or-
dinary muscle; 3, ordinary muscle myofibril; 4, dark
muscle; 5, dark muscle myofibril. DTNB: DTNB light chain
Figure 1—Procedure of kamaboko processing
302 JOURNAL OF FOOD SCIENCE—Vol. 66, No. 9, 2001
of ordinary muscle myosin. 15% gel.
1

Estimation of Dark Muscle Content in Mackerel Surimi . . .
Table 1—Dark muscle content in mackerel mince prepared Table 2—Dark muscle content in mackerel surimi and
by a meat separator adjusted to different pressures
kamaboko
Dark muscle content (%)
Dark muscle content (%)
Starting
material
Surimi
Kamaboko
Pressure
Mince
Surimi
Frozen whole fish
Frozen fillet
Frozen meat paste
10.0 Ϯ 0.5
13.1 Ϯ 1.6
9.8 Ϯ 0.8
10.4 Ϯ 0.7
11.4 Ϯ 0.9
9.5 Ϯ 0.9
High
Medium
Low
9.1 (8.9)
3.7 (5.9)
Ͻ3.7 (4.5)
12.1
11.3
8.6
was measured, and the protein concentration in the sample sin light chain)-antibody (antiserum against myosin light
solution was plotted against the halo area. The quantity of chain), white turbid halos were formed at the peripheral part
dark muscle was determined using the standard curves ob- of the wells where the antigens were poured. When the area
tained with ordinary and dark myofibrils, except in the case of
minced meat where the standard curves of both muscles were
used. Protein concentration was determined according to
Lowry and others (1951). Hemoprotein concentration was de-
termined according to Takeda and others (1980). All the deter-
minations were carried out in duplicate or triplicate, and the
average values (and standard deviations, in the case of tripli-
cate experiments) were given.
Results and Discussion
IGURE 2 SHOWS THE SDS-PAGE PATTERNS OF MYOSINS AND
F
myofibril from both muscles. Light chains can be recog-
nized based on the differences in mobility. Double immun-
odiffusion patterns are shown in Figure 3 in order to prove
the tissue-specificity of light chains and the corresponding
antisera. Anti-A1 antiserum formed a precipitation line only
against the ordinary muscle and its myofibril. The cross-re-
action was also observed for the combination of anti-D1 and
its antiserum.
Attempts were then made to estimate the ratio of dark
muscle in the whole meat using dissolved muscle and myo-
fibril for the SRID test (Figure 4). As a result of antigen (myo-
Figure 3—Double immunodiffusion of the solubilized Figure 4—Halos formed on single radial immunodiffusion
samples and antisera against myosin light chains. Anti- test (SRID) plates after diffusion of ordinary muscle solu-
A1 (a) and anti-D1 antisera (b) were poured in the center tions in anti-A1 containing agar plate (a) and dark muscle
wells. Abbreviations are as in the legend of Figure 1. The solutions in anti-D1 containing agar plate (b). Antiserum
numbers in superscript represent protein concentration, concentration is 1%. Relative concentrations from left to
namely, “2” being twice as high as that of “1”. Abbrevia- right (2nd and 4th rows) and from right to left (1st and 3rd
tions: OMU, ordinary muscle; OMF, ordinary muscle myo- rows) at serial dilutions from the initial concentration,
fibril; DMU, dark muscle; DMF, dark muscle myofibril. 19-␮g (OMU, upper 2 lanes in a), 14-␮g (OMF, lower 2 lanes
Original concentration: OMU, 12.1 mg/mL; OMF, 10.6 mg/ in a), 11-␮g (DMU, upper 2 lanes in b), and 6-␮g (DMF, lower
mL; DMU, 10.3 mg/mL; DMF, 6.9 mg/mL.
2 lanes in b).
Vol. 66, No. 9, 2001—JOURNAL OF FOOD SCIENCE 1303

Estimation of Dark Muscle Content in Mackerel Surimi . . .
Table 3—Dark muscle content in mackerel mince and meat was about 12%. This difference may have been caused
surimi at different stages of processing
by partial loss of ordinary myofibrils during washing. The
Material
Dark muscle content (%)
meat remaining in the refiner was found to contain around
3% of dark muscle. This suggests that the refiner is effective
in removing dark muscle from the mince.
Species identification of muscle and meat products has a
history of a few decades. Mackie (1969) tried to identify fish
species by SDS-PAGE using solubilized products with 6 M
urea. Later, Seki and others (1980) proposed a method for fish
species identification in heated or dried products taking ad-
vantage of species-specificity of myosin light chains. However,
this method required partial purification of light chains in or-
3
Mince
Washed meat
Surimi
Meat remained
in a refiner
9.8
11.7
9.1
33.2
of halo was plotted against protein concentration of sample
solutions, these were found to be linearly proportional (Fig- der to eliminate other contaminating proteins such as tropo-
nin whose subunit molecular weights were close to light
method for the quantification of dark muscle ratio in surimi chains. As shown in Figure 1, the electrophoretic pattern was
not applicable in this case. Hayden (1977) used anti-troponin
In the case of surimi and kamaboko prepared under vari- antiserum to detect the contamination of chicken flesh in beef
sausage by the double immunodiffusion test, claiming that the
ure 5). These results strongly support the applicability of this
and kamaboko.
ous conditions, light chains were recognized only for ordinary
muscle, suggesting that estimation from light chain patterns detection level was as low as 1 to 5%. He also tried the applica-
bility of anti-myoglobin antiserum (Hayden 1979).
An excellent point about using myosin light chains as
marker proteins is that they are quite stable through muscle
cannot be applicable (data not shown). However, when the
samples were applied to agarose plates containing either 1%
anti-A1 or anti-D1 antiserum, halos were clearly observed in
both cases (data not shown). Standard curves of myofibrils processing and remain soluble. Application of immunologi-
cal specificity of light chains between ordinary and dark
Minced meat prepared by a meat separator at 3 separat- muscles combined with the SRID test facilitated the determi-
nation of dark muscle content in fish products whether raw
above), and surimi prepared from these mince were exam- or processed. This method was developed to meet the de-
mand from fisheries industries as described above. The
are shown in Table 1. Standard curves for muscle were used method is considered to be widely applicable, since antisera
to respective myosin light chains, once prepared, can be
were used to estimate the quantity of dark muscle.
ing levels (namely, high, medium and low, as described
ined for the dark muscle content. The results of the SRID test
also for mince. Mince and surimi were found to contain dark
muscle at different levels, depending on the pressure applied stored under frozen conditions without lowering the cross-
reactivity or antibody values.
at the meat separator. As shown in Table 2, surimi and kam-
aboko did not differ substantially in the dark muscle content.
About 10% of dark muscle was contained in all the samples
except for the frozen fillet products, which showed some-
what higher values. These results indicate that meat prod-
ucts, even after heating, can also be examined for dark mus-
cle content by SRID test.
Table 3 shows the content of dark muscle at each stage of
surimi processing. The mince was prepared in advance at the
medium-level pressure. The contents of dark muscle in the
mince and surimi were about 9 to 10%, while that in washed
Conclusion
N IMMUNOCHEMICAL METHOD WAS DEVELOPED TO ESTI-
mate the content of dark muscle in mackerel meat prod-
A
ucts, using specific antisera for myosin light chains (A1 from
ordinary muscle and D1 from dark muscle) of the mackerel
Scomber japonicus. Mackerel meat paste was dissolved in
urea and SDS, and diffused on agar plates containing antiser-
um against A1 or D1 by single radial immunodiffusion. The
area of halos formed in agar plates was linearly proportional
to the contents of dark muscle. It was also possible to esti-
mate the dark muscle content in mackerel fish cake (kam-
aboko).
References
Altringham JD, Johnston IA. 1982. The pCa-tension and force velocity charac-
teristics of skinned fibres isolated from fish fast and slow muscles. J Physiol
3
33: 421-449.
Bone Q. 1966. On the function of the two types of myotomal muscle fibre in
elasmobranch fish. J Mar Biol Ass U K 46(2): 321-349.
Bossier P. 1999. Authentification of seafood products by DNA patterns. J Food Sci
6
4(2): 189-193.
Fahey JL, Mckelvey EM. 1965. Quantitative determination of serum immunoglo-
bulins in antibody-agar plates. J Immunol 94(1): 84-90.
Gazith J, Himmelfarb S, Harrington WF. 1970. Studies on the subunits of myosin.
J Biol Chem 245(1): 15-22.
Greer-Walker M, Pull GA. 1975. A survey of red and white muscle in marine fish.
J Fish Biol 7(2): 295-300.
Hayden AR1977. Detection of chicken flesh in sausages. J Food Sci 42(5): 1189-
1
192.
Hayden AR. 1979. Immunochemical detection of ovine, porcine and equine flesh
in beef products with antisera to species myoglobin. J Food Sci 44(2): 494-500.
Lowey S, Holt JC. 1972. An immunochemical approach to the interaction of light
and heavy chains in myosin. Cold Spring Harbor Symp Quant Biol 37:19-28.
Lowry OH, Rowebrough NJ, Farr AL, Randall RJ. 1951. Protein measurement with
the Folin phenol reagent. J Biol Chem 193(1): 265-275.
Mackie IM. 1969. Species identification of cooked fish by disk electrophoresis.
Analyst 93(1108): 458-460.
Figure 5—Standard curves for the estimation of ordinary
and dark muscle contents by SRID test. Refer to Figure 3 Mancini G, Vaerman JP, Carbonara AO, Heremans JF. 1964. A single-radial diffu-
for abbreviations.
sion method for the immunological quantitation of proteins. In: Protein Biol
1
304 JOURNAL OF FOOD SCIENCE—Vol. 66, No. 9, 2001

Estimation of Dark Muscle Content in Mackerel Surimi . . .
Fluids, 11th Colloquium, Peeters H, editor., Oxford, UK: Pergamon Press. p.370- Takeda N, Yamaguchi K, Hashimoto K. 1980. Decoloration of dark muscle of mack-
3
73
erel and sardine. Nippon Suisan Gakkaishi 46(10):1269-1272.
Obatake A, Doi T, Ono T. 1988. Post-mortem degradation of inosinic acid and Tsukamoto K. 981. The role of the red and white muscles during swimming of
related exzyme activity in the dark muscle of fish. Nippon Suisan Gakkaishi
4(2): 283-288.
Ouchterlony O. 1958. Diffusion-in-gel methods for immunological analysis. In.
yellowtail. Nippon Suisan Gakkaishi 50(12): 2025-2030.
Watabe S, Hashimoto K. 1980. Myosins from white and dark muscles of mackerel
– Some physico-chemical and enzymatic properties. J Biochem 87(5): 1491-
1499.
5
Progress in Allergy, Vol. V., Kallos P editor. Basel, Switzerland: Karger . p.1-78
Partis L, Croan D, Guo Z, Clark R, Coldham T, Murby J. 2000. Evaluation of a DNA Watabe S, Watanabe S, Hashimoto K. 1983a. The pH-dependency of myosin AT-
fingerprinting method for determining the species origin of meats. Meat Sci
4(4): 369-386.
Quinteiro J, Sotelo CG, Rehbein H, Pryde SE, Medina I, Perez-Martin RI, Rey-
Mendez M, Mackie IM. 1998. Use of mtDNA direct polymerase chain reaction
Pases from yellowtail ordinary and dark muscles. J Biochem 94(6): 1867-1875.
Watabe S, Dinh TNL, Ochiai Y, Hashimoto K. 1983b. Immunological specificity of
myosin light chains from mackerel ordinary and dark muscles. J Biochem.
94(5):1409-1419.
5
(
PCR) sequencing and PCR-restriction fragment length polymorphism meth- Weeds AG, Lowey S. 1971. Substructure of the myosin molecule. II. The light chains
odologies in species identification of canned tuna. J Agric Food Chem 46(4):
of myosin. J Mol Biol 61(3): 701-725.
1
662-1669.
Yang, R, Okitani A, Fujimaki M. 1970. Studies on myofibrils from the stored muscle.
Part 1. Post-mortem changes of adenosine triphosphatase activity of myo-
fibrils from rabbit muscle. Biosci Biotechnol Biochem 34(12): 1765-1772.
MS 20000843
Ram JL, Ram ML, Baidoun FF. 1996. Authentication of canned tuna and bonito by
sequence and restriction site analysis of polymerase chain reaction products
of mitochondrial DNA. J Agric Food Chem 44(8): 2460-2467.
Seki N, Takayasu M, Kokuryo H. 1980. Preparation of myosin light chains as an
index for identification of fish species from raw muscle and cooked products.
Bull Japan Soc Sci Fish 46(8): 1001-1006.
Stafford III WF, Szentkiralyi EM, Szent-Gyorgyi AG. 1979. Regulatory properties
of single-headed fragment of scallop myosin. Biochemistry 18(24): 5273-5280.
Takagi M, Taniguchi N. 1995. Random amplified polymorphic DNA (RAPD) for
identification of three species of Anguilla, Anguilla japonica, A. australis and
A. bicolor. Fisheries Sci 61(5): 884-885.
Authors Y. Ochiai, L. Ochiai, Hashimoto, and Watabe are with the Labora-
tory of Aquatic Molecular Biology and Biotechnology, Graduate School of
Agricultural and Life Sciences, The University of Tokyo, Tokyo 113-8657,
Japan. Address inquiries to author Watabe (E-mail: awatabe@mail.ecc.u-
tokyo.ac.jp).
Vol. 66, No. 9, 2001—JOURNAL OF FOOD SCIENCE 1305