Processing of marine foods

PII:

SO309-l740(96)00071-X

Meat Science, Vol. 43, No. S, 3265-S275, 1996 Copyright Q 1996 Elsevier Science Ltd Printed in Great Britain. Ail rights reserved 0309-1740/96 $15.00+0.00

ELSEVIER

Processing of Marine Foods E. 0. Elvevoll,” N. K. Smensen,a B. Osterud,b R. Ofstad” & I. Martinez’ aNorwegian Institute of Fisheries and Aquaculture Ltd, Tromser, Norway *Department of Biochemistry, Institute of Medical Biology, University of Tromw, Tromsa, Norway

ABSTRACT For the Norwegian fish industry, it is an objective to increase the production of value added products in order to improve profitability. This paper will briefly present four areas of important research tasks in this$eld. To aid in the identification of the species present in a product, we have applied the method called Random Amplification of Polymorphic DNA (RAPD). This technique is used to produce a fingerprint of DNA contained in the sample. The application of DNA typing for species identt$cation in jish products is presented. The nutritional aspects of foods are important. Although the low death rate from coronary heart disease among the Eskimos of Greenland has been suggested to stem in large part from their consumption of fish, one should keep in mind that the daily diet of Eskimos living in the traditional way consists of substantial quantities of meat and fat (blubber) from seals and whales. A recent study as to whether seal and whale oils are more effective than cod liver oil in changing biological parameters that might be important in explaining low incidence of coronary heart disease, asthma and psoriasis among Greenland Eskimos will be presented. Commercial processing offsh must take the development of rigor mortis into consideration since it affects yield and fish flesh quality. Infiuence of early processing (pre-rigor) on jish quality and yield is also discussed. There are sign$cant differences among fish species in gross chemical composition and morphological structure. Depending on the properties of the flesh and the way it is treated, it may gain or lose water. The relationship between structure and liquid-holding properties of cod and salmon muscle as a function of temperature is presented. Copyright 0 1996 Elsevier Science Ltd

INTRODUCTION Norwegian fish supplies are dominated by Atlantic cod (Gadus morhua L.) and Atlantic salmon (Sulmo s&r), being available from fisheries and imports, and aquaculture (approx. 450.000 and 230.000 tonnes respectively in 1995). For the Norwegian fish industry, it is an objective to increase the production of processed products, based on demands from the market, in order to improve profitability. Chilled products are one priority area in this respect. Such products can be broadly interpreted to include fresh, lightly salted, frozen/thawed, partly cooked or dried/rehydrated fish. The consumer prefers products that are perceived as fresh, convenient, and most importantly, provide a good sensory experience. S265

S266

E. 0. Ehevoll et

al.

The species used to prepare most ready-to-eat dishes can not be easily identified due to both processing conditions and since only parts of the fish may be used fillet, liver, roe, etc. This makes it easy for some manufacturers to use species of low value and sell them as if more expensive species had been used, for example haddock instead of cod in products. To aid in the identification of the species present in a product, we have applied the method called Random Amplification of Polymorphic DNA (RAPD) (Williams et al., 1990) or Arbitrary Amplification of Polymorphic DNA (AP-PCR) (Welsh & McClelland, 1990). This technique is used to produce a fingerprint of DNA contained in the sample. The application of DNA typing for species identification in fish products will be presented. The nutritional aspects of foods are important. Although the low death rate from coronary heart disease among the Eskimos of Greenland has been suggested to stem in large part from their consumption of fish, one should keep in mind that the daily diet of Eskimos living in the traditional way consists of substantial quantities of meat and fat (blubber) from seals and whales (Bang et al., 1976). The fat composition of seals and especially whales differs significantly from that of fish (Padley et al., 1986). Hence, one may ask whether oil from these animals may be better candidates than fish oil in explaining the apparent protective effect of the Greenland Eskimo diet. So far no controlled studies pertinent to any effects of these oils have been performed. In a recent study (Osterud et al., 19954) has raised the question as to whether seal and whale oils are more effective than cod liver oil in changing biological parameters that might be important in explaining low incidence of coronary heart disease, asthma and psoriasis among Greenland Eskimos (Kromann & Green, 1990). The processor must be aware of the need for fresh raw material and how the process influences product quality, including yield. For the processor it is desirable to preserve freshness by starting production as early as possible after slaughter, in order to gain time for distribution. Commercial processing of fish must take the development of rigor mortis into consideration since it affects yield and fish flesh quality. Influence of early processing @e-rigor) on fish quality and yield are discussed in this paper. There are significant differences among fish species in gross chemical composition and morphological structure. Furthermore, fluctuations in the water, protein, and fat contents of the flesh, caused by long migration, spawning and starvation, affect the muscle quality. Lipid and water together make up about 80% of the fish muscle. Depending on the properties of the flesh and the way it is treated, it may gain or lose water. This is important economically since fish is sold mainly by weight. The texture and liquid-holding properties of fish muscle depend also on the parameters of processing or culinary preparation. The most important of the technological parameters are those which influence the structure and conformation of myofibrillar and connective tissue proteins, i.e. temperature, ionic strenght and pH (Ofstad et al., 1993; Ofstad & Hermansson, 1996). The relationship between structure and liquid-holding properties of cod and salmon muscle as a function of temperature will be presented. THE APPLICATION

OF DNA TYPING FOR SPECIES IDENTIFICATION FISH PRODUCTS

IN

To aid in the identification of the species present in a product, we have applied the method called RAPD (Random Amplification of Polymorphic DNA) (Williams et al., 1990) or AP-PCR (Arbitrary Amplification of Polymorphic DNA) (Welsh & McClelland, 1990). This technique is used to produce a fingerprint of the DNA contained in the sample. It can be used to differentiate individuals or breeding stocks within a given species but also to differentiate among different species. Since all individuals belonging to a given species

S261

Processing of marine foods

have more of their genetic material in common than with individuals from other species, the conditions of the analysis can be optimised to reveal species-specific differences only. We have used this approach to analyse several products including fresh, frozen, dried and salted cod, cod-roe based products, fresh and smoked Atlantic salmon and Arctic charr and ready-to-eat fishgratin (Fig. l., Martinez, 1996). Then, the pattern obtained from the product has to be compared to the patterns obtained from known species to identify the species contained in the product. Hence, the method showed not to be adequate for canned products (Fig. 1.) A great advantage of using techniques based on the analysis of DNA instead of in the analysis of proteins, is that the DNA contained in all the cells of most somatic tissues in an individual is identical. Therefore, the same analysis can be applied independently of the tissue or organ used in the manufacture of the product. The most commonly used methods to identify species in products are based on the analysis of the proteins contained in the sample (Sotelo et al., 1993; Rehbein et al., 1995). The protein pattern, however, is tissue and developmental-stage specific (Martinez et al., 1991), which means that different s123

4

5

6

7

S

6

9

10

11

12

13SB

Fig. 1. Genomic fingerprints obtained with Operon (Operon Technologies Inc. CA) primer M-01 (5’-GTT-GGT-GGC-T-3’). The genomic DNA was prepared essentially according to (Miller et al.,

1988), with the modifications described by Martinez (1996). Amplifications, were as described by Williams et al. (1990) and Welsh & McClelland (1990) in 30 ~1volumes containing 50 mM Tris-HCl, pH 9.1, 16 mM ammonium sulphate, 6.5 mM MgC12 (Ellesworth et al., 1993), 100 mg/ml BSA, 100 mM each of dATP, dCTP, dGTP and dTTP, 0.4 mM lo-mer primer, 1.5 units Klen Taql DNA polymerase (Ab Peptides Inc. St Louis Missouri) and 20 pl of genomic DNA. The reaction mixtures were overlaid with 20 ~1 of Chill Out liquid wax (MJ Research Inc., Watertown, Massachusetts) and amplification was performed on a PTC-100 programmable thermal controller (MJ Research, Inc.) programmed as follows: an initial step of 94”C, 2 min, followed by 40 cycles of 94°C 20 set, 35°C 20 set, 72°C 1 min, one step of 72°C 5 min, and a final step of 15°C 10 sec. Fifteen ~1 of the products obtained were loaded into the wells of 2% agarose gels (3:l Seakem LE:Nusieve, FMC products), and the fragments were resolved by electrophoresis in O.SxTBE. The samples are arranged in groups of two. The first and second lane in each group differ in the amount of DNA by a factor of 5. Samples are cod: 1) fresh muscle, 2) fresh liver, 3) frozen muscle, 4) bacalao, 5) Mills caviar, 6) Kavli caviar, 7) Stabburet cod roe; Atlantic salmon: 8) fresh and 9) cold smoked fillet; Arctic charr: 10) fresh and 11) warm smoked fillet; 12) fish gratin and 13) cod liver canned. S, lkb ladder standard showing the fragments, from top to bottom, of size: 1,636; 1,018; 517; 506; 396; 344; 298; 220; 201 154 and 134bp. Smaller fragments are not discernible. B, is the negative control tube in the amplification reaction that contains all reagents except DNA. (From Martinez, 1996).

E. 0. Elvevoll et al.

S268

tissues in the same individual produce different protein patterns. In this case, one has to have both species- and tissue-specific protein patterns to compare to that of the unknown in order to find a match. Another important advantage of this method is that it does not produce false results. False positives and false negatives are not possible, since the DNA of one species does not transmute into the DNA of another species.

EFFECT

OF MARINE OILS ON COAGULATION AND CELLULAR ACTIVATION IN WHOLE BLOOD

In 1978 Dyerberg and Bang (Dyerberg et al., 1978) suggested that the high content of n-3 polyunsaturated fatty acids present in the traditional Greenland Eskimo diet offers protection against atherosclerosis and thrombosis. This inspired a number of scientists throughout the world to carry out epidemiological studies on the importance of dietary intake of fish and fish oil and to explore the effects of n-3 polyunsaturated fatty acids on various important biological systems. Although several recent studies have shown that a large intake of n-3 fatty acids is beneficial with regard to lowering blood pressure (Bsnaa et al., 1990), reducing triacylglycerols (Harris et al., 1988), and modulation of cell function and cell reactivity to external stimuli (Lee et al., 1985; Endres et al., 1989; Hansen et al., 1989; Fox & DiCorleto, 1988) the final conclusion depends on more controlled studies (Goodnight, 1993). The putative role of n-3 fatty acids from fish oil on the development of coronary heart disease has been questioned (Kromhout et al., 1985). In a resent study (Osterud et al., 1995a,b) has raised the question as to whether seal and whale oils are more effective than cod liver oil (CLO) in changing biological parameters that might be important in explaining the low incidence of coronary heart disease, asthma and psoriasis among Greenland Eskimos (Kromann & Green, 1990). In order to address this question, we have measured some related parameters that are probably associated with, or mediators of, inflammatory reactions, atherogenesis and thrombus formation before and after 10 weeks of daily dietary supplemention with oils from fish, seal and whale. In addition to determining serum lipids, fatty acid, total cholesterol, high density lipoprotein (HDL) cholesterol and triacylglycerol levels, we measured the coagulation factors fibrinogen and factor VII, both of which have been proposed to be risk factors in ischemic heart disease (Wilhelmsen et al., 1984; Meade et al., 1986). It has been established that monocytes play a central role in atherosclerosis through their adhesion to a damaged endothelium, thus inducing and sustaining atherosclerosis. Healthy men and women (134) were randomly selected to consume 15 ml/day of oil from blubber of seal, cod liver, seal/cod liver, blubber of Minke whale or no oil for 10 weeks. Total cholesterol was unchanged in the oil groups, whereas HDL-cholesterol increased 7% in the seal/CL0 group (p < 0.05) and 11% in the whale oil group (p < 0.005). Triacylglycerol was significantly reduced in the CL0 group only. The concentration of prothrombin fragment 1 + 2 was reduced 25% (p < 0.05) after whale oil supplementation. No change in fibrinogen or factor VIIc was detected. Tumour necrosis factor generation in lipopolysaccharide (LPS) stimulated blood was 30% reduced after whale oil (p <0.05), but unaffected by intake of seal or CLO. The LPS-induced tissue factor activity in monocytes was reduced to a significant degree only in the seal/CL0 group (34%) and whale oil group (35%) (p < 0.05). The most dramatic change in thromboxane B2 in LPS stimulated blood was seen after whale oil intake with 44% reduction (p < 0.01). In conclusion, daily dietary supplementation with whale oil alone or seal/cod liver oil in combination appeared to have the most pronounced effects on some selected biological

Processing of marinefoods

S269

parameters thought to play a role in cardiovascular disease. Although we did not perform any analysis of diet or dietary effect of supplementation, we are quite confident that the changes observed in our test model are caused by the oil supplementation. Pilot studies and an extensive experience with single individuals on various marine oils have all given results in accordance with the present study. The effect of whale oil is probably not mediated by n-3 fatty acids alone as the content of these fatty acids is relatively low in whale oil and correspond to an intake of only 1.8 g n-3 fatty acids pr day. Our study may indicate that in addition to n-3 fatty acids, there may be another component present in the Greenland Eskimo diet that might explain its apparent protective effect against atherosclerosis and thrombosis. As this was the first controlled study to survey a number of parameters potentially influenced by components of the Eskimo diet, we realise that more studies are required to clarify the role of oils from marine blubbers as a beneficial supplement to the diet in the prevention of coronary heart disease and thrombosis. Such studies are already in progress in our laboratory. INFLUENCE

OF EARLY PROCESSING

(PRE-RIGOR)

ON FISH QUALITY

Availability of live cod is seasonal, but small quantities of fish are also kept alive in intermediate storage, to supply a fresh fish market. In aquaculture the fish can be harvested according to market demand and capacity in the factory, offering fresh fish at any time. The availability of high quality chilled, fresh fish products to the market, relies on early processing. Commercial processing of fillets is usually started after the resolution of rigor mortis which often delays production for 24 days. Onset of rigor mortis has implications for processing because handling or filleting fish pre rigor or in-rigor can change the product properties and quality. In Norway, the processing of pre-rigor fish is well known to the industry, filleting “live” saithe (Polluchius virens). The plants often keep the fish alive in pens, allowing a steady supply of fresh raw material. The main products are iced, gutted fish for the fresh trade and frozen fillets in different sized blocks. When saithe is processed pre rigor, the frozen blocks may cause problems when used as raw material for fried products. This is because the fish goes into rigor mortis when heated, and changes shape, and the batter and breading can fall off. These problems are the result of the fish being too fresh when processed. Rigor mortis is dependent on the fish species, temperature when caught and of pre-rigor storage, handling in the factory and the biological status of the fish. This paper presents data from experiments assessing onset of rigor after slaughter, strength of rigor and how the state of rigor effects salt uptake of lightly salted fish fillets, focusing on the effect of temperature (Sorensen et al., 1996). Measurement of fillet length gives some information on the rigor state, but the individual variation are usually high. The fillets are left untouched on a smooth surface while rigor lasts and the fillets must be protected from surface drying (Fig. 2). Maximum contraction of fillets was reached first after 12 hours, for the fillets stored at 20°C while at the two other temperatures (O’C, 1oOC) a maximum was reached close to two days (40 hours) after slaughter. After reaching maximum contraction, the fillets was handled in order to determine whether the fillet would return to its original length. This was not the case. Shortening is not affected by handling of the fillets after resolution of rigor mortis. The original fillet lengths was 45 f 5cm and 15% reduction in length is approx. 7 cm. The deviation between samples was in the range of f 2-3 cm, being approximately 30%. The reduction in length of the fillets gave a product with different characteristics of appearance, being more dull, opaque (i.e. less shiny), than the normal post-rigor filleted product. This was confirmed by the sensory analysis.

S270

E. 0. Elvevoll et al. 0

g

-5

E ._ tT d -10 e c f

-1s

k I 20

-201 0

l

0°C

l

10°C

0 20°C

I 60

I 40

I 80

I loo

Hours

Fig. 2. Reduction in length of single fillets of Atlantic salmon stored at different temperatures; 0°C 10°C and 2O”C, as a function of time. Measurements at each storage temperature represent an average of 10 fillets. (From Sorensen et al., 1996).

Slightly different approaches in processing may result in large differences in the quality and yield of the finished product. The quality of a fillet that has been prepared pre-rigor, is different from a traditional post-rigor fillet, especially in appearance, colour, shape, texture and technological properties such as salt uptake. The yield e.g. is very different: post-rigor fillets increase 6% in weight to reach 3% salt, while pre-rigor fillets loose 7% weight in 3 hours to reach 3% salt in the fillet (Fig. 3). A difference of 13% in yield between pre-rigor and post-rigor salting of the fillets is important for the processors when preparing salt cured products. Hence, the importance of freshness in relation to yield must be taken into consideration. Attributes of quality such as colour, discoloration and gaping must also be considered. Pre-rigor salting also resulted in changes in the texture and appearance of the products.

l

-6 -

Pre rigor . In rigor A Post rigor

Fig. 3. Average weight changes during salting for 1, 2 or 3 hours in saturated brine at 4°C of pre-rigor, in-rigor and post-rigor cod fillets. The temperature was 18°C at slaughter. (From Sorensen et al., 1996).

S271

Processing of marine foods

Hours Fig. 4. Average salt content in pre-rigor, in-rigor and post-rigor salting in saturated brine at 4°C. The temperature was 8°C at

cod fillets after

1, 2, or 3 hours

slaughter. (From Sorensen et al.,

1996). A salt content of 3-3.5% in the fillet (w/w-basis) is regarded suitable for poaching. It takes less than one hour to reach a salt level of 3% in the fillet with post-rigor fillets and approximately 3 hours with pre-rigor fillets. The in-rigor fillets did not reach this level in three hours (Fig. 4). The pre-rigor salted fillets did enter rigor during the salting process and the appearance of the fillets changed from being smooth and shiny to having a rough and dry surface. It is known that pre-rigor and in-rigor fish absorb salt more slowly than post-rigor fish. A common explanation of the salting process is that salt is absorbed in the first stage and, at pH above the isoelectric point, the Cl- will increase the negativenegative repulsion of the proteins. Water is absorbed and the weight increases, due to swelling of the muscle (Honikel, 1989; Offer & Trinick, 1983). The pre-rigor fish fillets in this experiment do not swell and they lose weight. This can be explained by water loss due to contractions squeezing water out when rigor mortis is triggered by salting in brine (Fennema, 1990). The practical result is that it is far more economical to salt a post-rigor fillet than a pre-rigor one, if a lightly salted fillet is the desired product.

HEAT-INDUCED

STRUCTURAL

CHANGES

AND LIQUID

LOSS

The content of water and its distribution within the flesh give important contributions to the overall quality. The liquid-holding properties of the muscle tissue are thus of major importance to commercial value and consumer acceptance. Fish and fish products exhibit a wide range of functional properties, even before any effects of processing are considered. The structure of fish flesh is complicated and the interrelationships between all the components are as yet obscure. Undoubtedly, the differences in structure do have some influence upon the liquid-holding properties as well as the texture. Even within a species the functional properties may vary due to a number of biological factors. One important factor seems to be the nutritional status of the fish at capture influencing the post-mortem pH and the subsequent time-dependent muscle degradation. The free water in muscle, about 90% of the total water, is held by capillary and tension forces mainly within the intracellular location (Honikel, 1989; Morrisey et al., 1987). Hence, liquid-holding capacity of muscle is greatly influenced by structural changes in the proteins; fibril swelling-contraction and the distribution of fluid between intra- and extracellular locations (Offer & Trinick, 1983; Offer et al., 1989; Fennema, 1990).

E.0. Elvevoll et al.

S272 50 r

Temperature (“C)

Fig. 5. Liquid loss (% by weight) as a function of heating temperature of coarsely chopped; salmon (-•-), wild cod (-I-) and fed cod (-+ -) muscle according to the net test. (Means f standard deviation of 18 salmon, 15 wild cod and 30 fed cod). The temperature of the fish

muscle at slaughter was 47°C. (From Ofstad & Hermansson, 1996). Knowledge about relationships between microstructure and liquid-holding properties can be used both in order to choose raw materials and to optimise processing conditions and, hence, lead to enhanced quality of the end product (Hermansson, 1986, 1987, 1989). Results concerning the relationship between microstructure and liquid-holding properties of whole fish muscle as effected by temperature in relation to species and biological variations in the raw material are presented in Fig. 5 (Ofstad et al., 1993, 1996a). Cod and salmon, both of great economic importance for Norway, were used for the experimental work. To elucidate the effects of the nutritional status wild cod harvested in different seasons were compared with net pen fed cod. Figure 5 shows the liquid loss measurements of coarsely chopped muscle of fed cod (FC), wild cod (WC) and salmon (S) as a function of heating temperature. The liquid loss in salmon refers to both water and fat loss. The liquid loss was almost constant for both fish species between 5 and 20°C. At higher temperatures, the liquid loss increased rapidly as a function of temperature, attained a maximum when pre-heated to 45/5O”C, thereafter the liquid loss was constant or decreased slightly. Most of the liquid lost from salmon was water. Easily separable fat in the liquid fraction was first observed at 40°C. At 70°C about 16% of the liquid lost was fat. Considerable differences were found in liquid loss between cod and salmon (Fig. 5). Correction of the liquid loss due to the different protein and water contents of these muscles does not essentially affect this result. Species-specific differences in structural features between whole muscle of cod and salmon were the main reason for the different liquid-holding properties. The morphological features of the salmon muscle, the higher fat content and the better stability of the myosin/actomyosin fractions are probably of major importance with regard to the better liquid-holding capacity of the salmon muscle compared to the cod muscle (Ofstad et al., 1996a,b). Wild cod possessed much better liquid-holding properties than fed cod (Fig. 5). The average muscle-pH of wild and fed cod was 6.8 and 6.3, respectively. A pH closer to the iso-electric point of the myofibrillar proteins will increase the protein-protein attraction since the actomyosin has less net charge (Honikel, 1989; Morrisey et al., 1987). Accord-

Processing of marine foods

(b)

S273

(4

presentation of the heat-induced changes as observed in whole muscle (a) unheated, and pre-heated to (b) 3O”C,(c) 45/5O”C,and (d) 60/7O”C.(Drawing: Gunn Berit Lskken). (From Ofstad & Hermansson, 1996).

Fig. 6. Schematically

ingly, the intermyofibrillar spaces were wider in the fed than in the wild cod. Furthermore, the degree of structural degradation was more pronounced in the fed fish than in the wild cod. The higher liquid loss of fed cod compared to wild cod is mainly caused by the low pH induced denaturation and shrinkage of the myofibres and thus widening of the intermyofibrillar spaces. Thus, the nutritional status at the point of capture seems to play arole by influencing the post-mortem fall in pH, and subsequently the rate and extent of both myofibrillar shrinkage and muscle degradation; factors with detrimental effect upon liquid-holding properties. The main traits of the heat-induced changes, as observed by light microscopy (Ofstad et al., 1993) in transversally sectioned muscle samples, are schematically illustrated in Fig. 6. In the unheated sample (Fig. 6a) the separation between cells is seen as narrow spaces. At low heating temperatures (2&35”C), the main structural changes appeared in the endomysium surrounding the muscle cells (Fig. 6b). The endomysial layer swells and melted collagen fills the widened extracellular spaces. The denaturation and melting temperatures of collagen are near 20 and 40°C respectively (AlmHs, 1982; Sikorski et al., 1984). In mammalian meat, thermal shrinkage of the collagen fibres causes extrusion of intracellular water. A similar shrinkage of the collagenous layer compressing the cells was not observed in the fish muscle. Hence, the liquid loss between 20 and 35”C, is probably mainly due to denaturation of collagen altering the physical properties of the pericellular layer which represents a physical barrier to the release of fluid. The influence of collagen will probably be determined not only by the quantity present, but also the quality. This includes the nature of the cross-links present, the thermal stability of the cross-links, and the relative amounts of the collagen types. Severe shrinkage of the myofibres occurred at 45/5O”C (Fig. 6c), corresponding with maximum liquid loss. The loss of water occurs by the expulsion of water from the

S274

E. 0. Elvevoll et al.

myofibrils as they shrink when the filaments get closer together due to denaturation of myosin (Offer & Trinick, 1983; Ofstad et al., 1993; Hastings et al., 1985). The capillary forces are reduced and the expelled water being less firmly held can be pressed out more easily upon centrifugation. At elevated temperatures, when an extracellular granulated material became visible (Fig. 6d), the liquid loss decreased. The cell membranes were ruptured during heating. The reason for the observed reduced water loss at the higher temperatures may be that these sarcoplasmic proteinous aggregates mixed with gelatine (Ofstad et al., 1993, 1996b) are able to hold water and/or plug the intracellular capillaries, thus reducing the amount of liquid being released during centrifugation.

REFERENCES Almbs, K. A. (1982). Muskelcellehylstret hos torsk: ultrastruktur og biokjemi. Ph.D. Thesis, NTHUniversity of Trondheim. Bang, H. O., Dyerberg, J. & Hjorne, N. (1976). Acta Med. Stand., 200, 69. Bsnaa, K. H., Bjerve, K. S., Straume, B., Gram, T. T. 8z Thelle, D. (1990). N. Engf. J. Med., 322, 795.

Dyerberg, J., Bang, H. O., Steffensen, E., Moncada, S. & Vane, J. R. (1978). Lancet, 2, 117. Ellesworth, D. L., Rittenhouse, D. & Honeycutt, R. L. (1993). Biotechniques, 14,214. Endres, S., Ghorbani, R., Kely, V. E., Georgilis, K., Lonneman, G., van der Meer, J. W., Cannon, J. G., Rogers, T. S., Klempener, M. SWeber, P. C.Schaefer, E. J.Wolff, S. M. & Dinarello, C. A. (1989). N. Engl. J. Med., 320, 265. Fennema, 0. R. (1990). J. Muscle Fd., 1, 363. Fox, P. L. & DiCorleto, P. E. (1988). Science, 241, 453. Goodnight, S. H. (1993). Arch. Pathol. Lab. Med., 117, 102. Hansen, J. B., Olsen, J. O., Wilsgiird, L. & Osterud, B. (1989). J. Int. Med. Res., 225(l), 133. Harris, W. S., Zucker, M. L. & Dujovne, C. A. (1988). Am. J. Cfin. Nutr., 48, 992. Hastings, R., Rodger, G. W., Park, R., Matthews, A. D. & Anderson, E. M. (1985). J. Food Sci., 50, 503.

Hermansson, A. M. (1986). In Functional Properties of Food Macromolecules, ed. J. R. Mitchell & D. A. Ledward. Elsevier, London, 273. Hermansson, A. M. (1987). In Proc. 33th International Congress of Meat Science and Technology, Helsinki, 290. Hermansson, A. M. (1989). In Proc. 35th International Congress of Meat Science and Technology, Copenhagen, 776. Honikel, K. 0. (1989). In Water and Food Quality, ed. T. M. Harmann. Elsevier, London, 277. Kromann, N. & Green, A. (1990). Acta Med. Stand., 208, 401. Kromhout, D., Bosschieter, E. B. & Coulander, CdeL. (1985). N. Engl. J. Med., 312, 1205. Lee, T. H., Hoover, R. L., Williams, J. D., Sperling, R. I., Ravalese, J.111, Spur, B. W., Robinson, D. R., Corey, E. J., Lewis, R. A. & Austen, K. F. (1985). N. Engl. J. Med., 312, 1217. Martinez, I. (1996). In Seafoodfrom Producer to Consumer, Integrated Approach to Quality. Proceedings of the International Seafood Conference, ed. J. B. Luten. Noordwijkerhout, In press. Martinez, I., Christiansen, J. S., Ofstad, R. & Olsen, R. L. (1991). Eur. J. Biochem., 195, 743. Meade, T. W., Brozovic, M., Chakrabarti, R. R., Hannes, A. P., Imeson, J. D., Mellows, S., Miller, G. J., North, W. R. S., Stirling, Y. & Thompson, S. G. (1986). Lancet, 2, 533. Miller, S. A., Dykes, D. D. & Polesky, H. F. (1988). Nucleic Acids Res., 16, 1215. Padley, F. B., Gunstone, F. D. & Harwood, J. L. (1986). In The Lipid Handbook, ed. F. D. Gunstone, J. L. Harwood & F. B. Padley. Chapman and Hall, London-New York, 130. Morrisey, P. A., Mulvihill, D. M. & O’Neill, E. M. (1987). In Developments in Food proteins-J, ed. B. J. F. Hudson. Applied Sci. Publ., London, 195. Offer, G., Knight, P., Jeacocke, R., Almond, R., Cousins, T., Elsey, J., Parsons, N., Sharp, A., Starr, R. & Purslow, P. (1989). Food Microstruc., 8, 151.

Processing of marine foods

s215

Offer, G. & Trinick, J. (1983). Meat Sci., 8, 245. Ofstad, R., Kidman, S., Myklebust, R. & Hermansson, A. M. (1993). Food Structure, 12, 163. Ofstad, R., Egelandsdal, B., Kidman, S., Myklebust, R., Olsen, R. L. & Hermansson, A. M. (1996a). J. Sci. Food Agric., In press. Ofstad, R. & Hermansson, A. M. (1996). In Seafood from Producer to Consumer, Integrated Approach to Quality. Proceedings of the International Seafood Conference, ed. J. B. Luten. Noordwijkerhout, (in press). Ofstad, R., Kidman, S. & Hermansson, A. M. (19966). J. Sci. Food Agric., In press. IZlsterud, B., Elvevoll, E. O., Brox, J., Anderssen, T., Eliassen, L. T., Halvorsen, H., Hsgmo, P., Kvernmo, H., Lia, K., Lund, T., Olsen, J. O., Olsen, R. L., Sissner, C. & Vognild, E. (1995). Presentation at XII Znt. Symp. on Drugs affecting Lipid Metabolism. Houston, November 7-10. IZlsterud, B., Elvevoll, E. O., Barstad, H., Brox, J., Halvorsen, H., Lia, K., Olsen, J. O., Olsen, R. L., Sissener, CRekdal, 0. & Vognild, E. (1995). Lipids, 30, 1111. Rehbein, H., Etienne, M., Jerome, M., Hattula, T., Knudsen, L. B., Jessen, F., Luten, J. B., Bouquet, W., Mackie, I. M.Ritchie, A. H.Martin, R. & Mendes, R. (1995). Food Chem., 52, 193. Sikorski, Z. E., Scott, D. N. & Buisson, D. H. (1984). Crit. Rev. Fd. Sci. Nutr., 20, 301. Sotelo, C. G., Pineiro, C., Gallardo, J. M. & Perez-Martin, R. I. (1993). Trends Food Sci. Technol., 4, 395.

Sorensen, N. K., Brataas, R., Nyvold, T. E. & Lauritzen, K. (1996). In Seafood from Producer to Consumer, Integrated Approach to Quality. Proceedings of the International Seafood Conference, ed. J. B. Luten. Noordwijkerhout, In press. Welsh, J. & McClelland, M. (1990). Nucleic Acids Res., 18, 7213. Wilhelmsen, L., Svardsudd, K., Korsan-Bengtson, K., Larsson, B., Walin, L. & Tibblin, G. (1984). N. Engl. J. Med., 311, 501. Williams, J. G. K., Kubelik, A. R., Livak, K. J., Rafalski, J. A. & Tingey, S. V. (1990). Nucleic Acids Res., 18, 6531.