AN ABSTRACT OF THE THESIS OF

Vasana C. Weerasinghe for the degree of Master of Science in Food Science and

Technology presented on April 28. 1995. Title: Characterization of Pacific

Whiting Protease and Food-Grade Inhibitors for Surimi Production

Abstract approved:

Haejung An

Cathepsin B was the most active cysteine proteinase in the Pacific whiting

(Merluccius productus) fish fillet, and cathepsin L in surimi when the activities of the most active cysteine proteinases (cathepsin L, B, and H) were compared. Cathepsin L showed maximum activity at 550C in both fish fillet and surimi, indicating its function in myosin degradation during conventional cooking of fish fillet and surimi. Washing during surimi processing removed cathepsin B and H but not cathepsin L. Autolytic analysis of surimi proteins showed that the myosin was the primary target, while actin and myosin light chain showed limited hydrolysis during 2 hr incubation. When purified Pacific whiting proteinase was incubated with various component of fish muscle, proteinase was capable of hydrolyzing purified myofibrils myosin, and native and heat-denatured collagen. The degradation pattern of myofibrils by the proteinase was the same as the autolytic pattern of surimi.

Inhibition by the food-grade proteinase inhibitors varied with the catalytic type of proteinase. Beef plasma protein (BPP) had a higher percentage of papain inhibitors. followed by whey protein concentrate (WPC), potato powder (PP), and egg white

(EW). On the other hand, EW had a higher percentage of trypsin inhibitors followed by BPP, PP, and WPC. EW inhibited trypsin activity completely at levels as low as

1%. WPC inhibited the autolytic activity of fresh surimi. Bovine serum albumin

(BSA) was not effective as WPC. WPC can be used as an inhibitor for the Pacific whiting surimi, but high concentration is required.

A limited number of inhibitory components were found, as the components in food-grade inhibitors were characterized by inhibitory activity staining. Both EW and

PP showed more serine proteinase inhibitors than cysteine proteinase inhibitors. PP showed one cysteine inhibitory component while EW did not show any. BSA in both

WPC and BPP acts as an nonspecific competitive inhibitor and reduces the enzyme activity. An unidentified high molecular weight protein (HMP) found in WPC, BPP, and BSA functions as an alternative substrate for papain while it functions as true inhibitor for trypsin. Characterization of Pacific Whiting Protease and Food-Grade Inhibitors for

Surimi Production

by

Vasana C. Weerasinghe

A THESIS

submitted to

Oregon State University

in partial fulfillment of the requirement for the degree of

Master of Science

Completed April 28, 1995 Commencement June 1995 Master of Science thesis of Vasana C. Weerasinghe presented on April 28. 1995.

APPROVED:

_^ _^ . Major Professor, representing Food Science and Technology

■f— '-^ Hea^fi

I gMl. T | V TV-M/ » *■

Dean of Graduate Schdpl ^

I understand that my thesis will become part of the permanent collection of Oregon State University libraries. My signature below authorized release of my thesis to any reader upon request.

Vasana C.r£ Weerasinghe, Author ACKNOWLEDGMENT

I wish to express my deepest gratitude to Dr. Haejung An, my major professor, for her invaluable encouragement, continuous guidance and advice. She was always a source of inspiration to me. My sincere gratitude to Dr. Michael T. Morrissey, Dr.

Thomas Seymour, and Dr. Jae Park for their valuable advice and support. Special thanks to Lewis Richardson (Daddy Lew), who was always ready to help me; Nancy

Chamberlain for her unending support and editing my thesis; and Lynne Johnson for assistance in photographic preparations and for help in numerous ways. I am also grateful to all my colleagues and others whose names are not mentioned, for their support and help.

Last but not least, I wish to thank my loving parents who continually supported my education and always encouraged me, and to my beloved husband, Channa, who encouraged me, cheered me up, and endlessly supported me during this period of my study. TABLE OF CONTENTS

Page

INTRODUCTION 1

LITERATURE REVIEW 5

Pacific whiting 5

Surimi production 6

Muscle proteins and their degradation 8

Cathepsins 9

Proteinase inhibitors 11

Beef plasma proteins 12

Egg white 13

Potato powder 14

Whey protein concentrate 15

Oj-Macroglobulin 18

MATERIALS AND METHODS 20

Materials 20

Fish and surimi samples 21

Preparation offish and surimi extracts 21

Catheptic activity assays 22

Temperature profile of cathepsins L, B, and H 22

Autolytic pattern of surimi proteins 23

Protein extraction for SDS-PAGE 23

SDS-PAGE 24 TABLE OF CONTENTS (Continued)

Page

Storage study of surimi 24

Extraction of muscle and connective proteins 24

Purification of Pacific whiting cathepsin L 25

Hydrolysis of proteins by Pacific whiting cathepsin L 25

Inhibition of autolytic activities 26

Preparation of inhibitor solutions 27

Cysteine proteinase inhibitory activity assay 27

Serine proteinase inhibitory activity assay 28

Optimization of enzyme concentration for inhibitory activity assays 29

Determination of total protein content of inhibitor solutions 29

Partial purification of the Pacific whiting proteinase 30

SDS substrate-polyacrylamide gel electrophoresis 30

Development of proteolytic activity zones by partially purified proteinase 31

Development of inhibitory activity zones 32

Specific activity of the enzymes 33

RESULTS AND DISCUSSION 35

Catheptic activities in fish fillets and surimi 35

Autolytic pattern of Pacific whiting surimi 39

Hydrolysis of myofibrils by purified Pacific whiting cathepsin L 42

Time course study on surimi autolysis 48

Effect of protein inhibitors on autolysis of surimi 48 TABLE OF CONTENTS (Continued)

Page

Optimization of enzyme concentration for proteinase inhibitory assays... 52

Screening for cysteine protease inhibitors 52

Screening for serine protease inhibitors 57

Activity staining of purified Pacific whiting proteinase 57

Protein determination of inhibitor concentrates 61

Optimization of enzyme concentration for inhibitory activity staining 62

Staining for inhibitory activity 62

CONCLUSIONS 75

BIBLIOGRAPHY 77 LIST OF FIGURES Figure Page

1. Temperature Profile of cathepsins L, B, and H activities in fish extract 36

2. Temperature Profile of cathepsins L, B, and H activities in surimi extract 37

3. Autolytic pattern of Pacific whiting surimi proteins at 550C 40

4. SDS-PAGE patterns of surimi proteins during storage in ice 41

5. Hydrolysis of Pacific whiting myofibrils by the purified Pacific whiting cathepsin L at 550C 43

6. Hydrolysis of Pacific whiting myosin by the purified Pacific whiting cathepsin L at 550C 45

7. Proteolysis of heat-denatured Pacific whiting crude stroma by the purified Pacific whiting cathepsin L at 550C 46

8. Proteolysis of Pacific whiting insoluble crude stroma by the purified Pacific whiting cathepsin L at 20oC 47

9. Autolytic activity of Pacific whiting surimi at 550C at various incubation time 49

10. Inhibition of autolytic activity of Pacific whiting surimi by WPC and BSA as analyzed at 55CC for one hr 50

11. Optimization of papain concentration for BANA assay 53

12. Optimization of trypsin concentration for BAPNA assay 54 LIST OF FIGURES (Continued)

Figure Page

13. Inhibition of papain activity by various proteinase inhibitors as analyzed by BANA assay 55

14. Inhibition of trypsin activity by various proteinase inhibitors as analyzed by BAPNA assay 58

15. Activity staining on Partially purified Pacific whiting proteinase stained for activity at 370C 59

16. Activity staining on Partially purified Pacific whiting proteinase stained for activity at 550C 60

17. SDS-PAGE pattern of food-grade proteinase inhibitors 63

18. Proteinase inhibitors stained for papain inhibitory activity at 370C 64

19. Proteinase inhibitors stained for trypsin inhibitory activity at 370C 65

20. Proteinase inhibitors stained for Pacific whiting proteinase inhibitory activity at 370C 68

21. Proteinase inhibitors stained for Pacific whiting proteinase inhibitory activity at 550C 70

22. Inhibition of trypsin activity by BPP at various concentrations at 370C 71 LIST OF TABLES

Table Page

1. Enzyme and casein concentration, and buffers used for inhibitory activity staining 33

2. Protein contents of typically prepared inhibitors at 1% level 61 CHARACTERIZATION OF PACIFIC WHITING PROTEASE AND FOOD- GRADE INHIBITORS FOR SURIMI PRODUCTION

INTRODUCTION

Utilization of Pacific whiting (Merluccius productus) has increased over the past years by the growing demand for seafood analog products. A harvest of 179,073 metric tons from the U.S. Northwest (Washington-Oregon-Califomia) coast was reported for 1994 (Talley, 1995). At present, the majority of the Pacific whiting harvest is utilized for production of surimi with the aid of food-grade inhibitors.

Pacific whiting presents a significant technical challenge because of proteolytic degradation of myofibrillar components, which inhibits optimal gel formation.

Degradation of myofibrillar components was highest at 550C, resulting in loss of surimi gel strength (Chang-Lee et al., 1989; Morrissey et al., 1993). Crude preparation of the enzyme showed maximal activity at 550C and pH 5.5 (An et al., 1994; Chang-

Lee et al., 1989). The protease responsible for tissue softening was purified from

Pacific whiting fish fillets and identified as Cathepsin L (Masaki et al., 1993; Seymour et al., 1994), which is a highly active lysosomal cysteine proteinase for a variety of protein substrates (Kirschke et al., 1977, 1982; Koga et al., 1990; Okitani et al., 1980;

Yamashita and Konagaya, 1991). The properties of purified cathepsin L from Pacific whiting (Seymour et al., 1994) were very similar to the autolytic characteristics of surimi (An et al., 1994).

Cathepsin L was previously shown to be related to muscle softening of chum salmon during the spawning migration (Yamashita and Konagaya, 1990c). A cysteine proteinase was reported to be related to muscle softening of arrowtooth flounder 2

(Atheresthes stomias) (Wasson et al., 1992a). Cysteine proteinases in Pacific whiting and arrowtooth flounder seem to have similar properties in degrading surimi myofibrillar proteins. They both hydrolyzed myosin actively and actin to a lesser degree (Morrissey et al., 1993; Wasson et al., 1992a).

Several chemical inhibitors were shown to be inhibitory towards Pacific whiting proteinase activity (Miller and Spinelli, 1982; Nagahisa et al., 1983; Seymour et al., 1994). Hydrogen peroxide and potassium bromate are currently used in the U.S. food industry to improve functional and organoleptic properties of formulated food

(Miller and Spinelli, 1982). Although these additives inhibit proteolysis in Pacific whiting fish muscle (Miller and Spinelli, 1982), food-grade proteinase inhibitors have become popular for use in surimi.

Proteinase inhibitors are used to improve the physical properties of surimi gels made from arrowtooth flounder (Reppond and Babbitt, 1993; Wasson et al., 1992b),

Walleye pollock (Reppond and Babbitt, 1993), and Pacific whiting (Morrissey et al.,

1993). Commonly used food grade inhibitors are beef plasma protein (BPP), egg white (EW) and potato powder (PP). These additives show varying degrees of inhibition towards proteinases. (Akazawa et al., 1993; Morrissey et al., 1993; Porter et al., 1993; Reppond and Babbitt, 1993).

BPP contains inhibitors effective towards Atlantic menhaden (Brevoorti tyrannus), Alaska pollock (Themgam chcdcogrammd) (Hamann et al., 1990), and

Walleye pollock (Reppond and Babbitt, 1993) surimi and arrowtooth flounder

(Reppond and Babbitt, 1993; Wasson et al., 1992b) and Pacific whiting (Akazawa et al., 1993; Morrissey et al., 1993) washed mince. EW shows proteinase inhibitory activity towards surimi made from Atlantic menhaden, Alaska pollock (Hamann et al.. 3

1990), arrowtooth flounder (Reppond and Babbitt, 1993; Wasson et al., 1992b),

Walleye pollock (Reppond and Babbitt, 1993), and Pacific whiting (Chang-Lee et al.,

1989, 1990; Morrissey et al., 1993). Potatoes are a rich source of proteinase inhibitors

(Porter et al., 1993) and show inhibitory activity against proteinase from Pacific whiting (Akazawa et al., 1993; Morrissey et al., 1993), arrowtooth flounder (Porter et al., 1993), and trout muscle (Kaiser and Belitz, 1973).

Although food grade inhibitors inhibit proteinase activities in surimi, many factors contribute to their use in surimi. With the FDA approval on BPP as a food additive, its use in the surimi industry has increased. BPP is more effective than EW and PP because it gives the highest gel strength with the least quantity added.

However, its use has been limited due to the flavor associated with it at concentrations above 1%, which is not acceptable by Japanese consumers (Akazawa et al., 1993). In addition, BPP gives off-color (Reppond and Babbitt, 1993) to surimi and is required to be labeled, both of which have a negative impact on consumers. Similarly EW has off-color, off-flavor, high cost, and an undesirable egg-like odor at the levels required for the inhibition (Porter et al., 1993). PP did not show any sensory limitations

(Akazawa et al., 1993) but had off-color. Surimi made with PP suffers low gel strength as measured by torsion test (Morrissey et al., 1993). There is a growing demand for alternative food grade inhibitors due to the limitations of the currently used inhibitors. Various researchers stated that the Pacific whiting surimi gel strength could be increased by adding Whey Protein Concentrate (WPC) at high levels

(Akazawa et al., 1993; Chang-Lee et al., 1990; Piyachomkwan, 1994). WPC is an abundant inexpensive by-product of cheese manufacturing. No negative sensory properties were reported for WPC. 4 Even though food-grade inhibitors control proteolytic activities in surimi, questions remain concerning the proteolysis mechanism. For the inhibitors currently used, inhibitory components need to be identified. Therefore, the objectives of this study are to: (1) establish contribution of cathepsin L and other active lysosomal cysteine proteinases to surimi gel weakening; (2) investigate degradation mechanism of surimi myofibrils; (3) characterize inhibitory components in the food grade inhibitors; and (4) evaluate whey protein concentrate (WPC) as a potential inhibitor for Pacific whiting surimi. LITERATURE REVIEW

Pacific whiting

Interest in Pacific whiting significantly increased in the U.S. fishing industry after the establishment of the Fishery Conservation and Management Act of 1976

(Miller and Spinelli, 1982). Pacific whiting is the most abundant natural fishery resource off the Northwest coast of the United States (Bailey and Francis, 1985). It is usually harvested from May through September, and 177,000 metric tons of Pacific whiting is predicted for 1995 (PFMC, 1992).

Little commercial interest was given to Pacific whiting in the past due to the soft texture associated with it (Erickson et al., 1983). During conventional cooking, muscle tissue turns mushy and pasty resulting in an unacceptable texture for human consumption. Prolonged cooking at low temperatures results in a liquified product.

This soft texture was due to the presence of heat-stable endogenous proteinases which breakdown the muscle proteins responsible for the characteristic texture of fresh fish (Chang-Lee et al., 1989; Patashnik et al., 1982). The extensive hydrolytic activity of the enzyme is a result of the presence of Myxosporidian parasites {Kudoa paniformis) (Kabata and Whitaker, 1985). Konagaya and Aoki (1981) reported that the proteinase is derived from the parasite. This postulate has been questionable due to the absence of enzyme-producing organelles in the parasite spores (Stehr and

Whitaker, 1986). Microscopic examination of the fish fillet has shown the presence of black and white "hair-like" cysts (Patashnik et al., 1982). The black, or darker, cysts are the more mature, while the white cysts are the younger and proteolytically more active. White cysts are less readily apparent on visual inspection because of the 6 translucent color and poor contrast with the surrounding tissue. The level of infection varies with each fish as does the proteinase activity and extent of softening (Nagahisa

et al., 1983; Patashnik et al., 1982). Studies have shown a correlation between degree of infection and proteolytic activity (Konagaya and Aoki, 1981; Patashnik et al., 1982).

The enzyme responsible for tissue softening in Pacific whiting has its maximum

activity at 550C (An et al., 1994; Chang-Lee et al., 1989). During conventional

cooking, the gradual increase in temperature can enhance the enzyme activity. One

way to avoid this is to use rapid cooking methods, such as deep-fat frying or

microwave cooking, which quickly passes the high enzyme activity temperature range

and inactivates the enzyme before it hydrolyses the muscle tissue (Patashnik et al.,

1982). These methods, however, will not completely eliminate the enzyme activity,

even though they decrease the overall enzyme activity.

The best method for utilization of this fish has been the production of surimi due to its bland taste, white color, low cost, and large availability. At present, a large part of the harvest is successfully utilized for surimi production. Pacific whiting has been a good source of surimi with strong and cohesive gels, when proteinase inhibitors were added (Chang Lee et al., 1990; Morrissey et al., 1993).

Surimi production

Surimi is a Japanese term for minced and washed fish flesh, and used as a raw ingredient for kamaboko products/seafood analogs such as crab legs, shrimp, and lobster (Lee, 1984). It is the only functional protein ingredient to be commercially produced from animal muscle (Lanier, 1986). 7

Fish species with low fat content and white flesh are best for surimi production

(Ohshima et al., 1993). Among these are Southern blue whiting (Micromesistius austnalis), threadfm bream (Nemipterus japonicus), and hoki (Macruronus novaezelandiae). Only the fish whose myofibrillar proteins have not been denatured can be used to produce high-quality surimi. Freshness of fish determines the quality of the final product (Ohshima et al., 1993).

In surimi production, the raw fish is washed, headed, and gutted with a header/gutter and cleaned thoroughly with cold water. Fish are filleted with a mechanical filleter, mechanically deboned, and separated from bones and skin. The resulting fish mince is washed and dewatered several times using a rotary screen. Fish flesh contains 65-80% myofibrillar proteins and 18-20% sarcoplasmic proteins, based on the total proteins. During the washing procedure, sarcoplasmic proteins, fat, blood, pigments, odorous substances, digestive enzymes, lipases and phospholipases, water- soluble compounds such as trimethylamine oxide, formaldehyde, and inorganic salts are removed. As a result, the myofibrillar protein level goes up. Normally, three washing and dewatering cycles are performed during the processing of Pacific whiting surimi. In the final step, about 0.3% NaCl solution is added to facilitate the removal of water.

The washed mince is passed through a refiner and a screw press to remove the connective tissue and excess water, respectively. The cryoprotectants, i.e., sugar, sorbitol and phosphates, are added and blended to achieve a stable frozen storage.

Cryoprotectants protect the denaturation of the muscle proteins during frozen storage, and their effect is greatly enhanced by the addition of phosphates. Sometimes polydextrose is added instead of sucrose to avoid browning during frozen storage and 8 also to avoid sweetness in the final product (Ohshima et al., 1993). The prepared surimi is plate-frozen and stored at -20oC.

Chopping with salt solubilizes myofibrillar proteins. The resulting surimi paste, is extruded to the desired shape before it sets to gel. Dissolved myosin, due to the action of salt and chopping, combines with actin filament to form actomyosin. Gelling characteristics of actomyosin are mainly a function of the myosin portion of the molecule (Niwa, 1992). Among the myofibrillar proteins, myosin is the most important component contributing to the formation of surimi gels. The gel network is formed due to the unfolding and association of denatured myofibrills at high temperatures. Covalent bonds are formed between myosin chains. Myosin is also essential for maintaining the water and fat binding properties of kamaboko products.

Muscle proteins and their degradation

Proteins in the muscle are classified into three groups: (1) sarcoplasmic, the metabolic proteins which are mostly globular; (2) myofibrillar, the contractile element proteins which are fibrous; and (3) stroma, the connective tissue proteins. Myofibrillar proteins exhibit excellent gel-forming ability, while sarcoplasmic proteins do not. The gelling characteristics of surimi are dependent on several factors: freshness of fish, protein concentration, degree of myofibrillar protein denaturation, cryoprotectants, water content, pH, salt, and ingredients such as starch or fillers (Chung et al., 1993).

The presence of proteinase in fish muscle can have an adverse effect on the quality and physical property of the final product. Surimi is a functional ingredient in many food applications and the price is usually based on gel strength (Lee, 1984).

The rapid breakdown of myofibrillar proteins consequently inhibits the development of 9 a three dimensional gel network during heat-setting of surimi based products.

Nishimota et al. (1987) reported a positive correlation between the cross-linking ability of the myosin heavy chain and the gel-forming ability of the salted meat paste.

Different types of proteinases are found to be responsible for the thermal degradation of fish muscle which results in a poor gel. Cysteine proteinases are responsible for the tissue softening of arrowtooth flounder (Greene and Babbitt, 1990;

Wasson et al., 1992a). Alkaline proteinases are responsible for the breakdown of muscle texture of carp (Cyprinus carpio) (Iwata et al., 1973, 1974; Makinodan, 1987a;

Makinodan and Ikedo, 1969), White croaker (Sciena schlegeli) (Makinodan et al.,

1987b), Atlantic croaker (Micmpogon undulatus) (Lin and Lanier, 1980; Su et al.,

1981), Atlantic menhaden (Brevoorti tynmnus) (Boye and Lanier, 1988), and hoki

(Lorier et al., 1991). A trypsin-like alkaline proteinase and a latent serine proteinase are responsible for the breakdown of muscle proteins of Peruvian hake (Martone et al.,

1991) and threadfin bream (Nemipterus virgatus, Kinoshita et al., 1991), respectively.

Cathepsins are also reported to be responsible for the breakdown of muscle tissue and are discussed in the next section.

Cathepsins

A number of proteolytic enzymes, which are necessary for the complete degradation of proteins, are present in the lysosomes (Kirschke and Barrett, 1987).

Only cysteine and aspartic proteinases have been detected in lysosomes (Kirschke and

Barrett, 1985). The catalytic mechanisms of cysteine and aspartic proteinases allow them to be active at the acidic pH of the lysosomes, whereas the serine and metallo- proteinases rarely show much activity below pH 7 (Kirschke and Barrett, 1987). 10

Cysteine proteinases exhibit broad specificity in the pH range 5-7 (Kirschke and

Barrett, 1985). In lysosomes, endopeptidases initiate protein breakdown. Though some exopeptidases partially degrade proteins, extensive hydrolysis can take place only after endopeptidases acted on the proteins (Kirschke et al., 1977).

The most active proteinases in the body are the lysosomal cysteine proteinases.

They play a major role in intracellular protein degradation (Barrett and Kirschke,

1981). Cathepsin L, B, and H are the major enzymes in this group (Mason et al.,

1984). Cathepsin L is known to be a true endopeptidase; cathepsin B functions as an endopeptidase as well as exopeptidase (peptidyldipeptidase); and cathepsin H is known as an endoaminopeptidase (Kirschke and Barrett, 1987). Several cathepsins are purified and characterized from many sources of mammals and fish: cathepsin B from chum salmon (Oncorhynchus keta) (Yamashita and Konagaya, 1990b), and mackerel

(Scomber austrcdasicus) (Jiang et al., 1994); cathepsin L from rat liver (Kirschke et al., 1977), rabbit liver (Mason et al., 1984), human liver (Mason et al., 1985), sheep and ox liver (Mason 1986), chum salmon muscle (Yamashita and Konagaya, 1990a), and mackerel muscle (Lee et al., 1993); cathepsin D from tilapia (Tilapia niloticaX

Tilapia aured) (Jiang et al., 1990, 1992b), banded shrimp (Penaeus japonicus), and grass shrimp {Penaeus monodon) (Jiang et al., 1992a).

Cathepsins are also involved in cellular immunity induced by pathological conditions and are commonly present in lysosomes of phagocytes. During postmortem, cathepsins are released from lysosomes, which result in hydrolysis of muscle proteins (Asghar and Bhatti, 1987).

Erickson et al. (1983) reported that cathepsins may be responsible for the accelerated muscle breakdown in whiting. They could not detect any carboxypeptidase 11

A or B, collagenase, elastase, or trypsin-like activities in Pacific whiting proteinase extract using specific substrates. An enzyme was purified from Pacific whiting by

Seymour et al. (1994) and is identified to be cathepsin L, a highly active lysosomal cysteine proteinase with sulfhydryl residues on the active site. Cathepsin L has been isolated in single (M, ~ 30,000) and two (M, ~ 25,000 and M, ~ 5,000) chain forms

(Mason, 1989; Pike et al., 1992). Mammalian cathepsin L has a heavy chain (Mf ~

23,000 - 25,000) and a light chain (M,- 4,000-5,000) (Mason et al., 1985). Cathepsin

L has higher activity than other lysosomal proteinases against a variety of substrates such as collagen and myofibrillar proteins (Kirschke et al., 1982; Mason et al., 1985;

Okitani et al., 1980).

Proteinase inhibitors

Numerous fish species have high proteinase activity, i.e.. Pacific whiting, arrowtooth flounder and Alaska pollock. Among these only Alaska pollock do not need proteinase inhibitors to produce surimi with high gel strengths. Numerous factors drive the choice of additives in the surimi market. The major determining factors are cost, domestic and international regulations, and cultural preferences.

Many different proteinase inhibitors are used in surimi production to control proteinase activity and improve surimi gel strength. These include egg white (EW) powder (Chang-Lee et al., 1990; Hamann et al., 1990; Porter, 1993; Reppond and

Babbitt, 1993; Wasson et al., 1992b), dehydrated whole potato (PP, potato powder)

(Porter, 1993; Reppond and Babbitt, 1993), beef plasma proteins (BPP, Hamann et al.,

1990; Porter, 1993; Reppond and Babbitt, 1993; Wasson et al., 1992b), and o^- 12 macroglobulin (c^M, Lorier and Aitken, 1991; Wasson et al., 1992b) which is a component of BPP.

Compounds that react with sulfhydryl groups, i.e., hydrogen peroxide (free and alkaline), potassium bromate, iodoacetate, and ethymeleimaide, inhibited proteolysis sufficiently during frozen storage, when mixed with ground parasitized pacific whiting muscle. Texture of the cooked fish which contained these compounds was comparable with nonparasitized fish (Miller and Spinelli, 1982). Early researchers reported that proteinase inhibitors such as EW, potato, and soy and lima beans were ineffective as inhibitors for Pacific whiting (Miller and Spinelli, 1982). Later, it was discovered that

BPP, EW, and PP can be used to inhibit the gel degradation of Pacific whiting

(Akazawa et al., 1993; Morrissey et al., 1993; Porter et al., 1993).

Beef plasma proteins

BPP has been reported to improve surimi gel strength of arrowtooth flounder

(Porter, 1993; Reppond and Babbitt, 1993; Wasson et al., 1992b) and Walleye pollock

(Reppond and Babbitt, 1993) surimi. Beef plasma hydrolysate was used in Atlantic menhaden and Alaska pollock surimi and resulted in a good gel strength (Hamann et al., 1990).

Plasma proteins contain mainly albumin, globulins, and fibrinogen (White et al., 1973). Fibrinogen has M, ~ 341,000. A number of globulins exists such as, a,-, ctj-, P-, and y-globulins with varying molecular weights. BPP consists of ~ 50% BSA

(Mr ~ 67,000). c^M is another protein which has been isolated from plasma and has a wide range of inhibitory activity. This will be discussed in a later section. 13

Egg white

EW contains numerous types of proteins. All the egg white proteins are globular, and most of the proteins have an acidic isoelectric point (Stevens, 1991).

Ovalbumin is the most abundant protein and consists 54% of total egg white proteins.

It is a glycoprotein with M, ~ 45,000, pi of 4.5, and has homology with a group of proteinase inhibitors know as serpins by its amino acid sequence (Stevens, 1991).

Ovotransferrin comprises 12% of the total EW proteins. It is a glycoprotein with M, ~

76,700 and functions in transporting iron. Ovomucin is responsible for the high viscosity of EW. It comprises 3.5% of the total proteins and has Mf ~ 110,000

(Stevens, 1991; Whitaker and Tannenbaum, 1977). Lysozyme (3.4%, M, ~ 14,300) has alkaline pi and is one of the more thoroughly investigated protein. It cleaves peptidoglycans and plays a role in EW to protect from invading bacteria (Stevens,

1991).

Majority of the EW proteins are proteinase inhibitors and a number of these have been reported (Whitaker and Tannenbaum, 1977). Both ovomucoid (11%, M, ~

28,000), a heat-stable glycoprotein, and ovoinhibitor (1.5%, M, ~ 49,000) are serine proteinase inhibitors. Ovoinhibitor inhibits both trypsin and chymotrypsin, but ovomucoid inhibits only trypsin and not chymotrypsin (Kassell, 1970). Chicken ovomucoid is a temporary inhibitor of trypsin. It is gradually hydrolyzed in the presence of trypsin and is readily digested by pepsin. (Kassell, 1970). Cystatin

(0.05%, M, ~ 13,000) is a non-glycoprotein and an inhibitor of thiol proteinase

(Anastasi et al., 1983; Fossum and Whitaker, 1968). It inhibits ficin, papain, cathepsin

B, and dipeptidyl peptidase I, but only weakly inhibits bromelain (Barrett, 1981). This inhibitor was first discovered by Fossum and Whitaker (1968), and they called it ficin 14 and papain inhibitor. Later, Barrett (1981) proposed the name cystatin. EW also contains ovostatin, which is formerly known as ovomacroglobulin (Stevens, 1991). It is a high molecular weight protein (M, ~ 780,000) and is homologous to ctj- macroglobulin, by its structure and the mechanism of inhibition (Nagase et al., 1985).

Ovostatin inhibits a wide variety of endoproteinases including thermolysin and collagenases, both of which are metalloproteinases (Nagase et al., 1985). EW increased the gel strength of surimi made from several fish species such as Atlantic menhaden (Hamann et al., 1990), Alaska pollock (Hamann et al., 1990), arrowtooth flounder (Porter, 1993; Reppond and Babbitt, 1993; Wasson et al., 1992b), and

Walleye pollock (Reppond and Babbitt, 1993).

Potato powder

Potatoes are a rich source of proteinase inhibitors (Porter et al., 1993). Potato tubers contains a number of protein inhibitors. Cysteine proteinase inhibitor (potato cysteine proteinase inhibitor, PCPI 8.3, M, ~ 23,000) has a broader inhibitory spectrum for endopeptidases. It inhibits stem bromelain and fruit bromelain, which are not inhibited by EW cystatin, and shows stronger inhibition of cathepsin L compared to other cystatin-related inhibitors. Effectiveness of inhibition of the PCPI differs towards various cysteine proteinases (Rowan et al., 1990). An aspartic proteinase inhibitor (PDI) (Mares et al., 1989) is the only inhibitor of protein characteristic for cathepsin D and has a M, ~ 27,000 (Laskowski and Kato, 1980). It also inhibits trypsin and is homologous with the soybean trypsin inhibitor (STI) (Kunitz) family. Soluble heat stable proteins in the Russet Burbank potato are composed of the three major inhibitors. (1) Inhibitor I with M, ~ 39,000 is a tetrameric isoinhibitor, composed of 15 two major and two minor subunits. One of the two major subunits is a powerful inhibitor of both chymotrypsin and trypsin, whereas the other strongly inhibits chymotrypsin and weakly inhibit trypsin (Melville and Ryan, 1972). (2) Inhibitor II contains four monomeric isoinhibitor of M, ~ 10,5000 and inhibits both trypsin and chymotrypsin (Bryant et al., 1976). (3) A polypeptide (doublet) with M, ~ 4,100-6,000 with three types of inhibitor is also present. Among the three inhibitors only two have been investigated. The first is an inhibitor for pancreatic carboxypeptidase A and B with M, ~ 3,100 which contains 37-39 amino acid residues of which 6 are cystine with no free sulfhydryl groups (Ryan, 1974). The second is an inhibitor for serine endopeptidases and metallocarboxypeptidases with M, ~ 5,400, and it has similar characteristics of carboxypeptidase inhibitor but with 49 amino acid residues (Hass et al., 1976).

Porter et al. (1990) developed a method to obtain two potato products which can be used to treat fish with softening problems. Potato powder (PP) is freeze-dried raw potato. Potato extract (PE), freeze dried water extract of potato, has improved color characteristics and inhibitory power than PP. Potatoes are inhibitory against proteinase from trout muscle (Kaiser and Belitz, 1973), arrowtooth flounder (Porter et al., 1993; Reppond and Babbitt, 1993), Walleye pollock (Reppond and Babbitt, 1993), and Pacific whiting (Morrissey et al., 1993; Porter et al., 1993).

Whey protein concentrate

Whey is a major protein in milk. Whey Protein Concentrate (WPC) is an inexpensive, abundant by-product of cheese and casein (Giese, 1994) manufacture.

WPC is produced as a result of the disposal problem in the cheese industry by 16 industrial-scale ultrafiltration and dialfiltration technology (Dunker, 1984). Ion exchange adsorption technology is used to produce high-protein content WPC. The resulting WPC is pasteurized and spray dried (Morr, 1989). There are about 80 million to 100 million lbs of WPC produced per year in USA (Dunker, 1984).

Different WPC products are available in the market with protein concentrations ranging from 35 to 85%. Whey protein isolates (WPI) have protein contents of 90% or higher (Giese, 1994). The WPC is Generally Recognized As Safe (GRAS) (Morr and Foegeding, 1990) and has been utilized mostly as a functional ingredient in the food industry. WPC holds a variety of food applications: (1) as an emulsifier in ham, frankfurters, and luncheon meats; (2) as a replacement for nonfat dried milk (NFDM) powder in baked goods; (3) to increase the stability to the gel and reduce syneresis in yogurt manufacturing; (4) as a water binder to the crystallization of sucrose in confections; (5) egg white replacer in confectionery such as meringues; (6) in formulated fruit-juice beverages that are high in protein, e.g., healthy children's drinks, weight management products, fruit-based sport drinks, due to the solubility of WPC at low pH; (6) in the formulation of infant formulas; and (7) in milk-replaced product

(Dimker, 1984; Giese, 1994). WPC also contributes to the flavor, texture and color of the products (Gillies, 1974). Another use of WPC is for the production of edible bio- films by polymerization with transglutaminase (Mahmoud and Savello, 1992). Besides these wide applications, WPC can also increase the nutritive value of food (Giese,

1994).

WPC consists of 45-55% p-lactoglobulin (M,- 18,000), 20-25% a-lactalbumin

(M,- 14,000), 6-12% bovine serum albumin (M,- 67,000), and 8-10% immunoglobulin

(M, > 145,000) (Morr, 1985 and 1992). Composition data shows a high degree of 17 variability in the concentration of each individual protein (Morr, 1992; Morr and

Foegeding, 1990).

p-Lactoglobulin is the most abundant and most hydrophobic whey protein

(Morr, 1985). It has a free sulfhydryl group from a cysteine residue (Webb et al.,

1974), while a-lactalbumin, the second highest constituent, contains no free sulfhydryl groups despite the high content of cysteine (Webb et al., 1974). a-Lactalbumin has a good nutritive value because of its high content of tryptophan (Webb et al., 1974).

Lactalbumin and bovine serum albumin are identical in physical properties and composition (Webb et al., 1974). WPC produces higher foam expansion than liquid egg white proteins but lower viscosity and stability values (Morr 1985). Gelling properties of WPC was similar to those of EW (Burgarella et al., 1985).

WPC from cheddar cheese manufacture contains glycomacropeptide (GMP), which is a fragment of K-casein that results from the hydrolytic action of rennet upon casein micelles during cheese-making (Morr, 1985). The functional properties of cheddar cheese WPC can be affected by this highly acidic and glycosylated GMP fragment (Morr, 1985). It was first reported that GMP can inhibit the activity of trypsin and chymotrypsin (Chemikov et al., 1978). But later Sick et al. (1990) showed that GMP is not a trypsin inhibitor and it acts as a regular substrate for trypsin. GMP showed competitive inhibition of trypsin with the BAPNA substrate and the activity was lost when GMP was incubated with the enzyme. The WPC used in this study,

ALACO Surimi Plus, does not contain GMP (Meredith, 1994).

Ryder et al. (1991) studied the use of milk protein in fish gel products using

Alaska pollock as the fish source. WPC showed similar results to those using EW and soy proteins. 18

WPC was reported to inhibit gel weakening of Pacific whiting (Akazawa et al.,

1993; Chang-Lee et al., 1990). In arrowtooth flounder it was neither a good proteinase inhibitor nor a gel enhancer (Wasson et al., 1992b). Akazawa et al. (1993) used a- lactalbumin, a component of WPC, on Pacific whiting surimi and was unable to produce a surimi gel with reasonable strength.

otj-Macroglobulin

o^-Macroglobulin (ctjM), is a high molecular weight protein found in the plasma of vertebrates. Sottrup-Jansen et al. (1990) showed the presence of a functional homologue to mammalian o^M in horseshoe crab (Limulus polyphemus).

Similarities in the amino acid sequence were observed between the two species. o^M has been purified from plasma of several animals, such as dog, rabbit, mouse, pig, rat, cow, and hedgehog (Starkey and Barrett, 1977). Hudson et al. (1987) reported that mouse a-macroglobulin is a homologue of human OjM with functional similarities.

Molecular weight of o^M is around 800,000 and is one of the highest molecular weight plasma proteins. For the human, o^M molecular weight was determined by ultracentrifugation to be 725,000 (Barrett and Starkey, 1973). c^M is a glycoprotein with 8-11% carbohydrate (Starkey and Barrett, 1977) and is composed of four identical subunits (Barrett and Starkey, 1973). Its quartenary structure is composed of two units, which are bound together by a covalent bond and each unit has two subunits covalently linked by disulfide bonds (Starkey and Barrett, 1977).

c^M has a broader reactivity than any other known inhibitor that is specific for proteinases (Barrett and Starkey, 1973). Earlier c^M has been known as an inhibitor of serine proteinases, but now it is clear that otjM binds essentially to most of the 19 proteinases from the four catalytic classes of enzymes involving in a wide range of

endopeptidases (Barrett and Starkey, 1973). Each molecule of o^M can react with only one molecule (equimolar ratio) of proteinase (Starkey and Barrett, 1973).

Reactions with exopeptidases, non-proteolytic hydrolases, or inactive forms of

endopeptidases were not observed (Barrett and Starkey, 1973).

o^M is a different type of inhibitor compared to active site-directed proteinase

inhibitor. Instead of attacking the active site, OjM binds to the proteinase (Laskowski and Kato, 1980). "Trap hypothesis", as a mechanism of binding and inhibition, was proposed by Barrett and Starkey (1973). When o^M is exposed to a proteinase, limited proteolysis occurs at a "bait" region, near the middle of the sequence, of the

180,000 subunit. This results in a conformational change which will physically and

irreversibly trap the attacking proteinase. Due to this steric hindrance, access of large molecules to the enzymes active site becomes unaccessible (Starkey and Barrett,

1977). But small substrates can still reach the active site and proteolysis occurs. This results in a more pronounced inhibition with the use of large substrate molecules than with small ones (Barrett and Starkey, 1973). When the proteinase is trapped, more than 80% is covalently bound by the reaction between e-amino group on the proteinase and an internal P-cysteinyl-y-glutamyl thiol ester present in each c^M subunit (Salvesen et al., 1981). Saturation of OjM molecule with a proteinase will prevent the subsequent binding of other proteinase molecules (Starkey and Barrett,

1977). Salvesen and Barrett (1980) reported that for the inhibitory capacity of OjM, ability to form a covalent link is not essential, though it can stabilize the complexes against dissociation or proteolysis. Data from Heumann et al. (1989) showed that c^M can inhibit both soluble and insoluble proteinases. 20

MATERIALS AND METHODS

Materials

N-Carbobenzoxy-Phe-Arg-7-amido-4-methylcoumarin (Z-Phe-Arg-NMec), carbobenzoxy-Arg-Arg-7-amido-4-methylcoumarin (Z-Arg-Arg-NMec), L-Arg-7- amido-4-methylcoumarin (Arg-NMec), Polyxyethylene 23 lauryl ester (Brij 35, 30% w/v), L-trans-epoxysuccinylleucylamido(4-guanidino)butane (E-64), ethylene glycol- bis(B-aminoethyl ether) N,N,N',N'-tetraacetic acid (EGTA), bovine serum albumin

(BSA, fraction V, 98-99% albumin), papain, a-N-benzoyl-DL-arginine-2-naphthylamide

(BANA), benzoyl-DL-arginine-p-nitroanilide (BAPNA), bovine pancreas trypsin, and phenylmethanesulfonyl fluoride (PMSF), were purchased from Sigma Chem. Co., St.

Louis, MO. Molecular weights standard (Low Molecular Weight Calibration Kit) including phosphorylase b (94,000), albumin (67,000), ovalbumin (43,000), and carbonic anhydrase (30,000) was purchased from Pharmacia, Piscataway, NJ. Sodium caseinate (SC, casein-sodium) and Mersalyl acid were purchased from U.S.

Biochemical Corp., Cleveland, OH. Whey protein concentrate (WPC, ALACO-Surimi

Plus), beef plasma protein (BPP, AMP 600N), egg white (EW, standard egg albumin), and potato powder (PP, GS-91) were obtained from New Zealand Milk Products (Pvt)

Ltd., American Meat Protein Corp. Inc. Ames LA., Monark Egg Corporation, Kansas

City, MO., and Nonpareil Corp., Black foot, ID., respectively. ALACO Surimi Plus is a product developed by New Zealand Dairy Board after the study of Ryder et al.

(1991) to modify the texture of surimi based foods. It is a soluble, 77% milk protein product which forms a firm, elastic gel when heated. Fast Gramet GBC base (4-amino-

2,3-dimethylazobenzene) was obtained from Aldrich Chem. Co., Milwaukee, WI. 21 Stopping reagent for flurometric assays consists of 100 mM sodium monochloroacetate, 30 mM sodium acetate, 70 mM acetic acid in pH 4.3. Mcllvaine's buffer, pH 5.5 contains 0.2 M phosphate, and 0.1 M citrate. Phosphate buffer (100 mM), pH 6.0 contains 88 mM KH2P04 and 12 mM Na2HP04.7H20 .

Fish and surimi samples

Pacific whiting were harvested off the Oregon coast, stored on-board in champagne ice, and off-loaded within 24 hr of capture. The round fish were mechanically butterfly-cut in a local surimi processing plant and transferred in ice to

Oregon State University Seafood Laboratory. The fish were hand-filleted, vacuum- packed, and kept frozen at -20oC until use within six months.

Pacific whiting surimi without additives was obtained from a local surimi processing plant. The surimi was collected from the processing line, kept in ice, and transferred to the OSU-Seafood Laboratory. The samples were used on the same day, unless otherwise specified.

Preparation of fish and surimi extracts

Sarcoplasmic fluid was prepared from Pacific whiting fillet as a source of proteinase by centrifugation at 5,000 xg according to the method of Erickson et al.

(1983) and was referred to as fish extract. For surimi, two volumes of cold water

(v/w) were added, and the mixture was homogenized with a Polytron (Brinkmann

Instruments, Westbury, NY). The homogenate was centrifuged at 5,000 xg for 30 min

(Sorvall, DuPont Co., Newtown, CT), and the supernatant was referred to as surimi extract. 22

Catheptic activity assays

Activity analyses of cathepsin L, B, and H were carried out using the method of Barrett and Kirschke (1981). N-Carbobenzoxy-Phe-Arg-7-amido-4-methylcoumarin

(Z-Phe-Arg-NMec), N-carbobenzoxy-Arg-Arg-7-amido-4-methylcoumarin (Z-Arg-Arg-

NMec), and L-Arg-7-amido-4-methylcoumarin (Arg-NMec) at the concentration of 5

HM were used as specific substrates for cathepsins L, B, and H, respectively. Activity was analyzed by adding 250 jiL of the assay buffer, 300 \iL of 1% Brij 35, and 250

\xL of various substrates preheated to assay temperatures. Due to the interfering metalloproteinase activity observed in fish extract at room temperature, the assay was done in the presence and absence of 40 jaM E-64, which inhibits the activity of cysteine proteinases specifically. Cysteine proteinase activities were calculated by the difference in activities with and without E-64 (Activity = Actv,/oB_64 - Actw/E.64). No interfering activity was observed in surimi extract (An and Weerasinghe, 1993).

Temperature profile of cathepsins L, B, and H

Fish and surimi extracts were incubated with or without E-64 in buffer prior to activity assays. For control groups, 50 |iL of each assay buffer for cathepsins L, B, and H (Barrett and Kirschke, 1981), 100 ^iL of 40 v-M E-64, and 50 \iL of fish extract or 100 (iL of surimi extract were mixed together. The mixture was incubated at room temperature for 10 min prior to activity assay. For the samples, water was used in place of E-64. The mixture was assayed at the following temperatures for 10 min: 0,

20, 37, 45, 55, 65, and 750C. The reaction was completed by adding 1 mL of the stopping reagent (pH 4.3). The fluorescence was measured immediately using an 23

Aminco-Bowman Spectrophotofluorometer (American Instrument Co., Silver Spring,

MD) with excitation wavelength at 370 nm and emission wavelength at 460 nm.

Autolytic pattern of surimi proteins

Autolytic patterns of surimi were studied on SDS-PAGE by incubation of fresh surimi at 550C. Autolysis was carried out by the method of Greene and Babbitt

(1990). Fresh surimi samples without cryoprotectants or additives were obtained and tested on the same day. Fresh surimi, 3 g, was incubated in a covered beaker in a

550C water bath for 0, 5, 10, 15, 20, 25, 30, 45, and 60 min. The autolyzed samples were subjected to total protein extraction as described in the following section. The protein extracts were analyzed for protein content by Lowry's assay (Lowry et al.,

1951) and applied to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-

PAGE).

Protein extraction for SDS-PAGE

Total proteins were extracted by the method of Wasson et al. (1992a) with a slight modification. A mixture of 3 g surimi and 27 mL preheated (950C) 5% (w/v) sodium dodecyl sulfate (SDS) solution was homogenized for 1 min at speed 3-4 using a Polytron (Brinkmann Instruments, Westbury, NY). The homogenates were incubated in a 80oC water bath for 1 hr for complete solubilization of total proteins. Insoluble residues were centrifuged at 7,000 xg (Sorvall, DuPont Co., Newtown, CT) for 10 min at room temperature. 24

SDS-PAGE

SDS-PAGE was run according to the method of Laemmli (1970). Protein samples were boiled for 1 min in the SDS-PAGE treatment buffer (1:1, v/v), and 50

Hg proteins were applied on 10% poly aery lamide gels. Identification of individual myofibrillar protein components was made by comparing protein banding patterns to those of Farouk et al. (1992), Robe and Xiong (1992), and Ndi and Brekke (1992).

Molecular weights (M,) of protein bands were determined by the method of Weber and

Osbom (1975) using a molecular weights standard (Low Molecular Weight Calibration

Kit). Also included in the molecular weight calculation was Pacific whiting myosin heavy chain (MHC) with M, 200,000.

Storage study of surimi

Surimi samples were collected every day and stored in ice up to 3 days. Total proteins were extracted from the stored surimi samples as described above and protein content was determined by the method of Lowry et al. (1951). Autolysis of surimi proteins was observed on SDS-PAGE with 50 ng of proteins applied on 10% polyacrylamide gel.

Extraction of muscle and connective proteins

Myofibrils were extracted from fresh Pacific whiting fillets by the method of

Kay et al. (1982). The fillets were chopped, homogenized in 0.1 M KC1, 7 mM

KHjPO,,, 8 mM EGTA, 1 mM MgCl2, 1 mM dithiothreitol, pH 7.0, containing 1%

(v/v) Triton X-100, and centrifuged at 2000 xg for 10 min. The resultant pellet was 25 resuspended in the same buffer, homogenized, and centrifuged. This procedure was repeated for a total of five cycles. In the last two cycles, Triton was not included.

Myosin was prepared from whiting fillets by the method of Martone et al. (1986).

Crude stroma extract was prepared from fish fillets by the method of Yamashita and

Konagaya (1991).

Purification of Pacific whiting cathepsin L

Proteinase was purified from previously frozen Pacific whiting muscle by the method of Seymour et al. (1994). Pacific whiting fish fillets infected with Kudoa paniformis (Kabata and Whitaker, 1985) were used because of the higher proteolytic activities. The proteinase was purified using four steps, sarcoplasmic protein preparation, heat treatment, butyl-Sepharose, and DEAE. Two activity peaks, P-I and

P-II, were separated on butyl-Sepharose, and P-I was used as a representative activity.

DEAE fractions of P-I containing proteolytic activities were pooled and concentrated by ultrafiltration using a Centri-Prep 10 cartridge (Amicon, W.R. Grrace & Co.-Conn.,

Beverly, MA) and stored at -50oC until used.

Hydrolysis of proteins by Pacific whiting cathepsin L

The extracted proteins (myofibrils, myosin, and heat-denatured stroma) were subjected to enzymatic reaction by the method of An et al. (1994). The proteinase purified from Pacific whiting was added to the reaction mixture containing 10 mg protein substrate and 3.125 mL Mcllvaine's buffer (pH 5.5). The total volume was brought to 6.25 mL with water. The mixture was incubated at 550C for 0, 5, 10, 15,

20, 25, and 30 min. At each reaction time interval, 200 nL aliquot of the reaction 26 mixture containing approximately 320 ng of total proteins was taken out. For hydrolysis of insoluble stroma, the reaction mixture was incubated at 20oC for 0, 15,

30, 45, 60, 90, and 120 min. The unit of enzyme used in each assay was equivalent to activity to give AA428 of 0.5 for 10 min on 2 mg of azocasein at the reaction conditions previously reported (An et al., 1994). The aliquots were added to 100 yiL of SDS-PAGE treatment buffer and immediately boiled for 1 min to stop the enzymatic reaction and to prepare for SDS-PAGE. The samples containing 50 and

100 ng of proteins were applied to 10% SDS-PAGE gels.

Inhibition of autolytic activities

Whey Protein Concentrate (WPC) and bovine serum albumin (BSA) were tested as proteinase inhibitors (PI) for Pacific whiting surimi to reduce autolysis.

Sodium caseinate (SC) was also included as a negative control. Autolysis assay was carried out according to the method of Morrissey et al. (1993) with slight modifications. Fresh surimi without additives or cryoprotectants was obtained and tested on the same day. Proteinase inhibitors were added at 0-4% (w/w) to fresh surimi to a final weight of 3 g. The mixtures were mixed and spread evenly in a 100 mL beaker and incubated at 550C for 1 hr, unless specified. Autolytic reaction was stopped by adding 27 mL 5% (w/v) cold trichloroacetic acid (TCA) solution. The mixture was incubated at 40C for 15 min for unhydrolyzed proteins to precipitate, followed by centrifugation at 6,100 xg for 15 min. The TCA-soluble proteins were recovered from the supernatant, analyzed for oligopeptide contents by Lowry's assay

(Lowry et al., 1951), and expressed as nmoles of tyrosine released. Sample blanks containing all components were kept on ice and analyzed to correct for the 27 oligopeptides originating from fresh surimi and proteinase inhibitors. This procedure was repeated once. The residual activity of the enzyme was expressed as follows:

Tyrosine with PI % Residual Activity = X 100 Tyrosine without PI where PI refers to proteinase Inhibitor.

Preparation of inhibitor solutions

Inhibitor solutions at 4% (w/v) were prepared as a stock for enzyme assays.

WPC, beef plasma proteins (BPP), egg white (EW), BSA, and SC were diluted in water by constant stirring, and the further dilutions were made from this solution to prepare 1%, 2%, and 3% solutions. For EW the solution has to be homogenized at a low speed with a polytron in order to obtain a homogeneous solution. In the case of potato powder (PP), 1, 2, 3, and 4% solutions were individually prepared.

Cysteine proteinase inhibitory activity assay

The inhibitory activity against papain as a model for cysteine proteinase was measured according to the method of Izquierdo-Pulido et al. (1994) using a-N- benzoyl-DL-arginine-2-naphthylamide (BANA) as the synthetic substrate (Barrett,

1972). Papain stock solution (4.5 \ig/\iL) prepared in water and diluted with 4% Brij solution to a working strength. Solutions containing 50 \xL (0.045 ng/^L) of papain and 50\iL of inhibitors were added to 100 mM phosphate buffer (pH 6.0) containing

1.33 mM EDTA and 2.7 mM cysteine to a final volume of 400 |iL. The mixture was preincubated for 5 min at 40oC, and 10 uL of 40 mg/mL BANA in dimethylsulfoxide was added and vortexed to start the reaction. The mixture was incubated at 40oC 28 precisely for 10 min for activity assays. Then 400 JAL of a mixture of Fast Gramet

GBC base (4-amino-2)3-dimethylazobenzene), sodium nitrite, and Mersalyl acid was added to terminate the reaction. After allowing 10 min for color development, resultant solution was transferred to microfuge tubes and centrifuged at 8,000 xg for 3 min (Eppendorf Micro Centrifuge, Model 5415C, Brinkmann, New York, NY). The residual papain activity was measured by the absorbance at 520 nm due to a red azo dye produced by 3-naphthylamide liberated and coupled with Fast Gramet. Reagent blanks were prepared by adding the enzyme after the color development. Standards were prepared by replacing the enzyme with 50(iL P-naphthylamide standard (20 mM) and blank with 50 (iL water, respectively. For each assay, a positive control and a negative control were carried out by adding 100 mM phosphate buffer and E-64, respectively, in place of inhibitors.

Serine proteinase inhibitory activity assay

Serine proteinase inhibitory activity was measured according to the method of

Smith et al. (1980) using trypsin and benzoyl-DL-arginine-p-nitroanilide (BAPNA) as a synthetic substrate with slight modification. A solution containing 100 \xL of inhibitor solution (prepared as described above), 200 nL (20 \xglmL) bovine pancreas trypsin and 100 ^L of water was preincubated for 10 min at 370C. Then, 500 jiL (0.4 mg/mL) of BAPNA (prewarmed to 370C) was added and vortexed immediately to start the reaction. After incubating at 370C for 10 min, 100 nL of 30% (v/v) acetic acid was added to terminate the reaction. The reaction mixture was centrifuged at 8,000 xg for 3 min (Eppendorf Micro Centrifuge). Residual activity of trypsin was measured by the absorbance at 410 nm due to ^-nitroaniline released. Sample blank was prepared 29 by adding enzyme after terminating the reaction with acetic acid. Since no coupling reactions were carried out to determine the residual activity compared to cysteine proteinase inhibitory activity assay, sample blank for each assay was also used as a standard. Absorbance measurement from the blank was used to calculate the amount of /?-nitroanilide released from each of the assay. For each inhibitor assay, a positive control and a negative control were carried out by adding water and phenylmethanesulfonyl fluoride (PMSF), respectively, in place of inhibitors in the original reaction mixtures.

Optimization of enzyme concentration for inhibitory activity assays

The amount of enzyme required for each of the above assay was determined by comparing the reactions in the presence and absence of 4% BPP for various enzyme concentrations. For both activity assays, BPP was chosen, since mammalian blood plasma contains various types of proteinase inhibitors (Laskowski and Kata, 1980). A blank was prepared for each enzyme concentration by adding the enzyme after the reaction. The enzyme concentration, which showed the largest difference between the activity with and without the added inhibitor, was used as the optimal enzyme concentration for the assay.

Determination of total protein content of inhibitor solutions

Total protein content of each inhibitor was determined by extracting the total proteins from 1% inhibitor solutions by the method of Wasson et al. (1992a) with a slight modification. In the case of PP, 10% solutions were used due to the low protein content. To 0.3 g (3 g in the case of PP) of inhibitor, 29.7 mL (27 mL in the case of 30

PP) preheated (950C) 5% (w/v) SDS solution was added and homogenized for 1 min at speed 3-4 using a Polytron (Brinkmann Instruments, Westbury, NY). After incubating the homogenates in 80oC water bath for 1 hr for complete solubilization of all proteins, the insoluble residues were pelleted by centrifugation at 7,000 xg (Sorvall, DuPont

Co., Newtown, CT) for 15 min at room temperature. The protein content of supernatant was determined by Lowry's assay (Lowry et al., 1951) using BSA as the standard.

Partial purification of the Pacific whiting proteinase

Proteinase was purified from previously frozen parasitized Pacific whiting by the method of Seymour et al. (1994) but using batch processing. Parasitized Pacific whiting fish fillets were used because of high proteolytic activities. Previous work in our lab (Wu, 1994) showed that the two activity peaks, P-I and P-II, from the butyl sepharose column has the same characteristics and they are relatively pure at this stage. Therefore for this study, the proteinase was purified from the sarcoplasmic fluid by heat treatment at 60oC and batch style with 50 mL butyl sepharose resin. The ratio of extraction was 2:1, of resin to buffer. The purity of the partially purified proteinase was checked by the activity staining as described later.

SDS substrate-polyacrylamide gel electrophoresis

The purified Pacific whiting proteinase was analyzed on SDS-PAGE according to the method of Laemmli (1970). Proteolytic and proteinase inhibitory activities were analyzed on 12% and 15% acrylamide gels, respectively, using casein as a substrate by the method developed by Garcia-Careno, et al. (1993) with slight modification. 31

Vertical electrophoresis was conducted on a BioRad Modular Mini-PROTEAN n

electrophoresis system (Bio-Rad Laboratories, Hercules, CA) with 7.3 x 8.3 x 0.75 cm

slab gels using constant voltage of 75 mV per gel for 45 min on ice. Proteinase or

inhibitor solutions were mixed with sample treatment buffer (1:1) containing SDS but

no P-mercaptoethanol (BME) and was not boiled before loading to the gel. To get an

estimate for the apparent molecular weight of inhibitory components, low molecular

weight standards (Pharmacia Biotech. Inc., NJ) containing phosphorylase b (94,000),

albumin (67,000), ovalbumin (43,000), carbonic anhydrase (30,000), trypsin inhibitor

(20,100), and a-lactalbumin (14,400) were used as molecular weight markers.

Development of proteolytic activity zones by partially purified proteinase

When tracking dye reached the bottom of the running gel, electrophoresis was

stopped and the gels were immersed in 50 mL of 2% casein in 50 mM Tris/HCl

buffer, pH 7.5, for 30 min at 0oC, in order to allow the substrate to penetrate into the

gels at reduced enzyme activity. Then, the gels were immersed in 2% casein in 200

mM Mcllvaine's buffer, pH 5.5 with 2 mM BME and incubated at 370C and 550C for

90 min for the digestion of the protein substrate by the purified Pacific whiting proteinase. Both temperatures were studied because all inhibitory activity staining was done at 370C, but the enzyme had maximal activity at 550C (An et al., 1994; Chang-

Lee et al., 1989). The gels were fixed and stained with a solution containing 50% ethanol, 10% acetic acid, and 0.125% Coomassie blue R-250 (Bio Rad Laboratories,

CA) and destained with 25% ethanol. Clear zones on the dark background of the gel indicate the bands with proteinase activity. 32

Development of inhibitory activity zones

Inhibitory activity of food grade inhibitors was analyzed on SDS substrate-

PAGE gels. Two identical gels were prepared for inhibitory activity and total protein staining. Fifteen |ig of proteins were applied on the gel and electrophoresed on the condition described above. The control gel was fixed and stained for proteins and inhibitory gel was washed in 2.5% Triton X-100 for 15 min to remove the SDS

(Izquierdo-Pulido et al., 1994). Then the gels were washed with distilled water and incubated with an appropriate proteinase and activator in the relevant buffer according to Izquierdo-Pulido et al. (1994) with slight modification. In the case of Pacific whiting proteinase, the activator was included in the substrate solution. After allowing the enzyme to diffuse into the gel at 0-5oC for 30 min, the inhibitory gel were washed again with distilled water and incubated in 1% casein solution in relevant buffer (Table

1) for 90 min at 3TC. In the case of Pacific whiting proteinase the gel was also incubated at 550C, at which the enzyme showed maximal activity (An et al., 1994;

Chang-Lee et al., 1989). Then the gels are washed again with distilled water, fixed and stained with Coomassie blue R-250. After staining both the control and the inhibitory activity gels for 2 hr, the gels were destained and the relative migration distance of the bands with proteinase inhibitory activity (dark zones on the clear background) was compared with the control gel and the molecular weight markers. 33

Table 1 - En2yme and casein concentrations, and buffers used for inhibitory activity staining

Enzyme / Buffer Activator Concentration

papain (0.20 mg/mL) 0.01 M phosphate buffer, 1 mM EDTA + 2 mM pH6.0 cysteine

trypsin (0.20 mg/mL) 0.05 M Tris buffer, 20 mM Cal2 pH8.2

Pacific whiting 0.30 M Mcllvaine's 2 mMBME proteinase (0.41 buffer, pH 5.5 mg/mL)

Specific activity of the enzymes:

Specific activity of the enzymes used for the inhibitory activity staining was analyzed by measuring the activity on casein according to An et al. (1994) with slight modification. A reaction mixture containing 2 mg of casein, 0.625 mL of relevant buffer with the activator, and water was preincubated at 370C for 5 min. Total volume of the reaction mixture with the enzyme was maintained at 1.25 mL. Then the enzyme was added and incubated at the same temperature for 5 min, and the enzymatic reaction was stopped by adding 0.2 mL of cold 50% (w/v) trichloroacetic acid (TCA) to the reaction mixture. After allowing unhydrolyzed proteins to precipitate at 40C for 5 min, samples were centrifuged at 6,000 xg for 5 min

(Eppendorf Micro Centrifuge). The TCA-soluble proteins were recovered in the supernatant and the oligopeptide content was determined by the Lowry's assay using 34 tyrosine as the standard. A unit of activity (U) of proteinase was expressed as nmoles of tyrosine released from proteins. Specific activity was expressed by U/mg. 35

RESULTS AND DISCUSSION

Catheptic activities in fish fillets and surimi

Activities of cathepsins L, B, and H were monitored to study the effect of washing during surimi processing. An et al. (1994) reported that 97% of the proteolytic activity in Pacific whiting fish extract was due to cysteine proteinase activity. Cathepsins are the most active cysteine proteinases in muscle (Barrett and

Kirschke, 1981), even though they function slightly differently. Cathepsin L is known to be a true endopeptidase, and cathepsin B functions as a endopeptidase as well as a exopeptidase (Barrett and Kirschke, 1981). Cathepsin H is known as an endoaminopeptidase (Kirschke et al., 1977).

During normal surimi processing, fish mince is washed three times with water and low ionic salt solutions (Lee, 1984). A gradual decrease in proteinase activity by washing was reported earlier (Chang-Lee et al., 1989). Our results showed there was selective removal of proteinases as shown by the difference in activity profiles of fish fillets versus fresh surimi. When activities in both fish and surimi extracts were measured, substantial changes in the composition of proteolytic activities were observed (Figures 1 and 2). In fish fillets, peak activity of cathepsin B was the highest, followed by cathepsins L and H. Peak activity of cathepsin B, observed at

20oC, was 170% of cathepsin L at 550C; and the peak activity of cathepsin H, shown at 20oC, was 75% of cathepsin L at 550C. However, after processing which included extensive washing and dewatering, cathepsin B activity in the muscle mass was substantially decreased to 22% of cathepsin L with no activity detected for cathepsin 2000

1500

CA

O a 1000

20 30 40 50 60 70 80 Temperature (0C)

Figure 1. Temperature profile of cathepsin L, B, and H activities in fish extract 1200 (»Cat. L ♦Cat. B ▼Cat. H ) 1000

ac/1 800 CD

aO 600 O

200

30 40 50 80 Temperature (0C)

Figure 2. Temperature profile of cathepsin L, B, and H activities in surimi extract 38

H. This result implies that cathepsin L is the most active cysteine proteinase in surimi and that it contributes to the hydrolysis of myosin in surimi at elevated temperatures.

In order to determine the contribution of these enzymes at different temperatures, their temperature profiles were studied with respect to cathepsin L, B, and H activities of fish and surimi extract. Cathepsins L and H in fish extract showed their highest activities at 55 and 20oC, respectively, while the maximum activity of cathepsin B was between 20 and 370C (Figure 1). Previously, it was reported that

Pacific whiting underwent severe textural degradation during prolonged baking

(Patashnik et al., 1982), and the purified proteinase showed maximum activity at 550C

(An et al., 1994; Chang-Lee et al., 1989). In this study, the total cumulative activity of the three cysteine cathepsins (L, B, and H) was found to be highest at 20oC in fish fillets. This implied that the total activity of cathepsins may not be as important as the activity of a specific proteinase in the case of tissue softening during conventional cooking. Among the three cathepsins studied, only cathepsin L showed the maximum temperature at 550C indicating that this enzyme may be the principal proteinase contributing to the textural degradation of Pacific whiting fillets during conventional slow cooking. However, at temperatures below 550C, textural degradation is most likely due to the activity of cathepsin B as shown by its high activity in fish fillets at this temperature range.

The optimum temperatures were shown to be 55 and 370C for cathepsins L and

B in surimi extract, respectively (Figure 2). No cathepsin H activity was observed.

The total cumulative activity of all three cathepsins was the highest at 550C (Figure 2).

Morrissey et al. (1993) reported that no myosin heavy chain was found in Pacific whiting surimi after cooking for 30 min at 60oC. Myosin degradation was 39 accompanied by the total loss of surimi gel strength. The predominance of cathepsin

L activity at 550C makes it clear that loss of surimi gel strength is primarily due to the activity of cathepsin L.

Autolytic pattern of Pacific whiting surimi

The degradation pattern of Pacific whiting surimi was studied by autolysis at

550C (Figure 3). Surimi has a high percentage of myofibrillar proteins, i.e., predominantly myosin and actin along with troponin and tropomyosin. Among the proteins, myosin heavy chain (MHC) was the primary target. MHC was most extensively hydrolyzed, followed by troponin-T, and a and P-tropomyosin. The intensity of the MHC band on SDS-PAGE gel was substantially reduced within 5 min incubation and completely disappeared within 20 min. Troponin-T, a-tropomyosin, and P-tropomyosin were not affected until incubation for 20 min but completely disappeared in 30 min. Both actin and myosin light chain (MLC) were the least affected, and their presence was still detected even after 2 hr. Generally, degradation of all myofibrillar proteins increased with time. The high molecular weight protein bands (MI > 94,000) disappeared during incubation with subsequent appearances of smaller molecular protein bands.

The effect of storing fresh surimi without any additives was studied with respect to hydrolysis of myosin (Figure 4). No significant hydrolysis of myosin or other myofibrillar proteins was observed on SDS-PAGE during three day storage on ice. The minor difference noticed on day 2 was thought to be due to sample variation by the processing. No pungent odor or noticeable microbial spoilage was apparent. 40

MHC

Actin

6-TM/TNT a-TM

MLC 0 5 10 15 20 30 60 90 120 L

Figure 3. Autolytic pattern of Pacific whiting surimi proteins at 550C. The numbers designate incubation time (min). L, low molecular weight standards; MHC, myosin heavy chain; TM, tropomyosin; TNT, troponin-T; and the arrow indicates the M, of 94,000. 41

0 1 2 3 L

Figure 4. SDS-PAGE patterns of surimi proteins during storage in ice. The numbers designate storage days. L, low molecular weight standards. 42

These results agreed with the textural measurements of fresh surimi by torsion that the gel strength was acceptable for commercial use up to 5 days (Lin et al., 1993).

Hydrolysis of myofibnls by purified Pacific whiting cathepsin L

The hydrolytic activity of purified Pacific whiting cathepsin L was studied by incubation with the muscle proteins purified from Pacific whiting, i.e., myofibrils, myosin, and insoluble and heat-denatured stroma. The hydrolysis products were analyzed on SDS-PAGE for their degradation patterns.

Hydrolysis of myofibnls Myofibrils extracted from Pacific whiting contained MHC and actin as major constituents and 13-tropomyosin/troponin-T and MLC as minor

(Figure 5). The general degradation pattern and sequence were similar to those of surimi autolysis. The MHC was hydrolyzed to a trace amount in 5 min with subsequent appearances of the primary hydrolysis products (Mj ranging from 74,500 to

145,000). The hydrolysis pattern of myofibrils showed protein bands with Mj between

67,000 and 145,000 as degradation products of myosin. These bands were detected in all surimi samples used in this study without storage (Figure 5), indicating a large portion of myosin was hydrolyzed during or prior to surimi processing.

The highest M/s of protein bands detected for the 5 and 10 min reaction times were 145,000 and 122,000, respectively. No increase in intensity of protein bands between M, of 200,000 (MHC) and 145,000 suggested the proteinase had endopeptidyl activity. The primary hydrolysis products of myofibrils (M,. > 67,000) were hydrolyzed to the secondary products with M, 29,300; 36,500; and 45,100.

A rapid hydrolysis of I3-tropomyosin/troponin-T was also observed but to a lesser degree than MHC. Both MHC and C-tropomyosin/troponin-T were completely 43

- MHC

"^ 145,000 _ 122,000 _ 107,000 83,100 77,400 - 74,500

Actin 45,100

36,500

29,300

LOS 10 15 20

Figure 5. Hydrolysis of Pacific whiting myofibrils by the purified Pacific whiting cathepsin L at 550C. The numbers designate incubation time (min). L, low molecular weight standards; and MHC, myosin heavy chain. 44 hydrolyzed in 10 min, and their presence was not detected on the SDS-PAGE gel.

Actin was hydrolyzed very slowly compared with MHC and 13-tropomyosin/troponin-T.

In 15 min, all the proteins including actin and MLC and the primary hydrolysis products of MHC and 6-tropomyosin/troponin-T disappeared from the SDS-PAGE gel indicating complete hydrolysis of these proteins in the reaction mixture.

Hydrolysis of myosin MHC was degraded rapidly within 5 min by the purified proteinase as shown by SDS-PAGE (Figure 6). Due to the rapid hydrolysis, the intermediate degradation products (M, between 67,000 and 145,000) were not as easily detected as the low M, products (29,300; 36,500; and 45,100) which were the secondary reaction products of myosin.

Hydrolysis of stroma Hydrolysis of both heat-denatured and native stroma by the purified proteinase was tested by incubation of a crude stroma extract at 550C and room temperature, respectively. At the optimum temperature, 550C, the proteinase was highly active against stroma. At this temperature, stroma was heat-denatured as indicated by the color change from opaque white to translucent. The collagens were degraded into smaller molecular weight proteins after 30 min as shown on SDS-PAGE

(Figure 7). The average molecular weight of the protein bands shown for each reaction time decreased with increased reaction time. Band intensity of a degradation product (M, of 30,000) was proportional with the degree of hydrolysis and increased with the duration of reaction time.

The proteinase slowly hydrolyzed native insoluble collagens at room temperature (Figure 8). The extracted stroma mainly consisted of al(I), a2(I), and

B(I) type 1 collagens (Sato et al., 1991). The proteinase had only 7% activity at room temperature compared with that at 550C on azocasein (An et al., 1994). Even though 45

MHC

-

mm L 0 5 10 15 20 25 30

Figure 6. Hydrolysis of Pacific whiting myosin by the purified Pacific whiting cathepsin L at 550C. The numbers designate incubation time (min). L, low molecular weight standards; and MHC, myosin heavy chain. 46

L 0 5 10 15 20 25 30

Figure 7. Proteolysis of heat-denatured Pacific whiting crude stroma by the purified Pacific whiting cathepsin L at 550C. The numbers designate incubation time (min). L, low molecular weight standards. The arrow indicates the protein band with ^ 30,000. 47 wn 6(1)

<*2(I) «1(I)

L 0 15 60 90120150

Figure 8. Proteolysis of Pacific whiting insoluble crude stroma by the purified Pacific whiting cathepsin L at 20oC. The numbers designate incubation time (min). L, low molecular weight standards. 48 some variation was observed in the intensity of protein bands on SDS-PAGE gel due to the insolubility of stroma, generally the intensity of the collagen bands decreased with the subsequent appearance of smaller molecular weight protein bands on the

SDS-PAGE gel (Figure 8). Yamashita and Konagaya (1991) previously reported that native collagens were degraded at 20oC by chum salmon cathepsin L but not by cathepsin B. Our results showed that Pacific whiting cathepsin L was also capable of hydrolyzing insoluble collagens at 20oC. This result confirmed that cathepsin L was involved along with collagenases in textural degradation of Pacific whiting due to breakdown of collagen and thus could contribute to overall softening of fillets or whole fish.

Time course study on surimi autolysis

Time required for the autolytic assay was determined by incubating fresh surimi without inhibitors for various time intervals. Surimi autolysis showed a lag period of 10 min at the beginning of the reaction, followed by a linear reaction up to one hr. Thus one hour was used for the subsequent studies in the measurement of autolytic activity of surimi (Figure 9). The A750 reading for positive control was -0.8 when using 50 nL of TCA-soluble supernatant without dilution.

Effect of protein inhibitors on autolysis of surimi

WPC and BSA inhibited autolytic activity of surimi (Figure 10). The activity decreased with increasing concentration of WPC and BSA. At 1% level, there was only 50% residual activity in fresh surimi with WPC, while it was higher at 89% with

BSA. Steady decrease in residual activity was observed with WPC up to 3% level. 200

20 40 60 Time (min)

Figure 9. Autolytic activity of Pacific whiting surimi at 550C at various incubation time 100

ICO

•a a:(U

% Proteinase inhibitor

Figure 10. Inhibition of autolytic activity of Pacific whiting surimi by WPC and BSA as analyzed at 550C for one hr.

U1 o 51 and the activity plateaued at the higher concentration with the residual activity of 14%.

These data indicated that there was no additional benefit of WPC at levels higher than

3%. Wasson et al. (1992b) reported that inhibitors incorporated in arrowtooth flounder surimi at levels higher than necessary to prevent proteolytic activity displayed enhanced gelling effect. Additional ingredients, such as starch in potato powder and heat-coagulable protein in egg albumin and bovine plasma, can also enhance gel strength (Porter et al., 1993). Therefore it is difficult to correlate gel strength of surimi solely with inhibitor concentration.

Dose-related response was observed with BSA. The residual activity decreased linearly with increasing concentration of inhibitor. However, BSA was not as effective as WPC. Casein was tested as a negative control to determine whether competitive inhibition by increased protein substrate concentration was the mechanism. Increasing the concentration of casein in surimi from 0 to 4% showed an approximately 20% decrease in autolytic activity. This indicated that proteins can generally reduce the activity of proteinases at high concentrations, even though the degree of inhibition is not as high as inhibitory compounds.

WPC is an abundant, reasonable, and readily available by-product of cheese manufacturing. It does not have limitations such as off-flavor, off-odor, and off-color.

Based on the autolysis data, WPC may be used as a proteinase inhibitor for Pacific whiting surimi at 3% or above. Akazawa et al. (1993) reported that WPC formed better gel in Pacific whiting than what was predicted by the autolysis data, though it exhibited only minimal inhibition of myosin heavy chain (MHC) breakdown.

Piyachomkwan (1994) reported that WPC protected MHC from degradation at 4% 52 level. The discrepancies in inhibition by WPC as observed by various studies may be due to the use of lower concentration of WPC like 0.5% and 0.25%.

Optimization of enzyme concentration for proteinase inhibitory assays

Optimum enzyme concentrations for enzyme assays using synthetic substrate were determined to measure the most effective proteinase inhibition. Using 4% BPP, the optimum enzyme concentration for papain and trypsin was 0.045 ng/nL and 0.020

\ig/\iL, respectively (Figures 11 and 12). These two enzyme concentrations in each of the respective assays showed the largest difference between the residual activity with and without the added inhibitor. At higher enzyme concentrations than the optimum, the inhibitor did not showed inhibition of the enzyme activity, indicating limited amount of inhibitor in the reaction mixture. At the lower enzyme concentrations, the amount of enzyme for substrate hydrolysis was too low to be measured by the assay.

Screening for cysteine proteinase inhibitors

All the proteinase inhibitors under study inhibited papain as shown by BANA assay (Figure 13). At 1% level, there was only 26% residual activity shown for BPP, which was lower than those of EW, WPC, and PP (88%, 84%, and 68%, respectively).

At higher levels of BPP, no drastic change occurred. The residual activity plateaued out at 3% level at the concentration above 3%. This indicates that increasing BPP higher than 3% did not enhance the efficiency of inhibition. For WPC and EW, there was a dose-related response. The residual activity decreased proportionally with increasing level of the inhibitors. At 4% level, the lowest residual activity was observed at 9% and 13% for WPC and EW, respectively. BSA, used as the negative (■W/OBPP E3W/BPP )

0.8

a o 0.6 I ■ © u 4 4 4 4 4 4 '4444444 >'4444444 ► 4 >4444444 >4444444 ► 4 ► 4 ► 4 ► 4 > 4 4 4 i 4 4^ 4 4 4 o 0.4 4 4 1 3 ► ♦♦ v.v.v. >4444444 >>4444444 0.2 4 1 1 4 4 4 4 4'* 4 4 Hv.M SOX 10X IX .IX Papain concentration (IX = 0.045 \ig/\iL)

Figure 11. Optimization of papain concentration for BANA assay 54

X

X ^—s "^ a.

o ©' II

X (4^ X g ^H •H < H ^ ^ o 5l c PQ 8 eg •S § X 1 1 c o

o X ti o o N

o (N u-> oi O m Ui 11111 Olt7 ® aoxreqiosqy (*WPC ■▼•BPP -a-EW rArPI -•-BSA~) I

•i-H M a:

0 1 2 Proteinase inhibitor concentration (%

Figure 13. Inhibition of papain activity by various proteinase inhibitors as analyzed by BANA assay

in 56 control, hardly affected the residual activity up to 3%. At 4% level, the residual activity decreased slightly to 84%. These data showed that among the four inhibitors studied BPP had the highest amount of cysteine proteinase inhibitors. Inhibitory efficiency of WPC, PP, and EW were similar, even though WPC showed more cysteine proteinase inhibition than EW or PP at 2% levels or higher. Inhibition comparable to that of 1% BPP can be achieved with other proteinase inhibitors at higher levels, such as 3% and 4%.

A cysteine proteinase is responsible for degradation of Pacific whiting surimi proteins (An et al., 1994). Morrissey et al. (1993) reported that proteinase inhibition by BPP was greater than the inhibition by EW or PP at 1% level, when measured by autolysis of fresh surimi. In the BANA assay used in this study, BPP also showed the highest cysteine proteinase inhibition, followed by PP and EW. EW was an effective proteinase inhibitor at high concentrations.

BSA was used as a negative control in this study. Initially, casein was used as the negative control, since it is a good substrate for papain with the least effect of competitive inhibition (Amon, 1970). Papain can readily hydrolyse casein, and can function as a competing alternative substrate to BANA. Casein resulted in a lower color intensity produced from the BANA assay indicating inhibition by casein. Acid- denatured haemoglobin was not used as a negative control because of the interference with the color development. Therefore, BSA was used as the negative control for the comparison purposes. 57

Screening for serine proteinase inhibitors

The content of serine proteinase inhibitors in the food-grade inhibitors was determined by BAPNA assay (Figure 14). Trypsin was completely inhibited by EW even at 1% concentration due to the presence of specific serine proteinase inhibitors in

EW. Egg white is one of the earliest known trypsin inhibitors (Kassell, 1970). It is consisted of 11% ovomucoid and 1.5% ovoinhibitor by weight (Whitaker and

Tannenbaum, 1977). Ovomucoid is a specific trypsin inhibitor, and ovoinhibitor is a general serine proteinase inhibitor which can inhibit both trypsin and chymotrypsin

(Kassell, 1970; Whitaker and Tannenbaum, 1977). At 1% concentration, there was 64,

68, and 85% residual activity by BPP, PP, and WPC, respectively. For BPP, residual activity decreased linearly up to 3% and then reached the plateau which indicated the excessive presence of inhibitor. WPC decreased the activity linearly with the increasing concentration indicating a dose-related response. The residual activity reached 41% at 4% WPC.

Activity staining of purified Pacific whiting proteinase

The catalytic purity of the proteinase was tested by activity staining using casein as the substrate. The purified Pacific whiting proteinase showed only one activity band on SDS-PAGE by activity staining at both 370C (Figure 15) and 550C

(Figure 16). This shows no other contaminating activities in the proteinase preparation on the activity staining condition. Enzyme was purified 12-fold. The specific activity of the proteinase was approximately 2000 U/mg as analyzed at 370C on 2 mg of casein. Io en -78 9 T3 M cu en

1 2 Proteinase inhibitor concentration (%

Figure 14. Inhibition of trypsin activity by various proteinase inhibitors as analyzed by BAPNA assay

in oo 59

94,000 — 67,000 —

43,000 —

30,000 —

20,100 -

14,400-

Figure 15. Activity staining of partially purified Pacific whiting proteinase stained for activity at 370C. The numbers designate the ^ of sample applied on 12% acrylamide gel. M, molecular weight standards. 60

j 94,000 — 67,000 —

43,000 —

30,000— «■»

20,100 — Ms*

14,400 —

M2345 678

Figure 16. Activity staining of partially purified Pacific whiting proteinase stained for activity at 550C. The numbers designate the \iL of sample applied on 12% acrylamide gel. M, molecular weight standards. 61

Protein determination of inhibitor concentrates

Protein contents of dry inhibitor powders were determined by preparing protein solutions at 1% (Table 2). Due to lower amount of protein in PP, 10% was used to determine the protein content. When the protein inhibitors were solubilized by heating in the presence of SDS, all inhibitors were completely solubilized except PP. Potato powder contained a large amount of starch due to the manufacturing method (Porter et al., 1990). PP is produced by grinding the peeled, dried and frozen potato, and then freeze-drying and ball-milling to a uniform powder. Even though EW showed the highest protein concentration, and SC the lowest, only slight differences in the protein concentrations were observed.

Table 2. Protein contents of typically prepared inhibitors at 1% level.

Inhibitor Protein concentration (mg/mL)*

WPC 13.46

BPP 12.56 1

EW 15.73 PP I 1.06 BSA 12.81 SC I 11.65 * - average of two data 62

Optimization of enzyme concentration for inhibitory activity staining

Initially, papain and trypsin at concentrations employed by Izquierdo-Pulido et al. (1994) were used for inhibitory activity staining. Though only limited number of components in proteinase inhibitors showed inhibition towards trypsin, several components were inhibitory towards papain. In the earlier studies on EW and PP, more number of trypsin inhibitors were reported than cysteine (Bryant et al., 1976;

Kassell, 1970; Melville and Ryan, 1972; Rowan et al., 1990; Stevens, 1991). By increasing the papain concentration, the number of inhibitory bands was reduced.

Therefore, in order to optimize the enzyme concentration for the inhibitory activity staining, specific activity of the enzymes was determined. Papain and trypsin concentrations for proteinase inhibitory activity on electrophoresis were chosen such that both enzymes have comparable specific activity on 2 mg casein at 370C.

Staining for inhibitory activity

Inhibitory components in the food-grade inhibitors were identified by the activity staining (Figure 18 and 19). The presence of inhibitors was detected as dark bands on the light background. As a control, SDS-PAGE gel was developed and stained for total proteins by Coomassie blue R-250 (Figure 17). The control gel detected all protein components present in the inhibitor compounds. In both inhibitor activity-stained gels against papain and trypsin (Figure 18 and 19), protein bands originating from casein or molecular weight standards have disappeared completely.

These lanes served as negative controls.

WPC is reported (Morr, 1985, 1989, and 1992) to compose of two major proteins, P-lactoglobulin (50% by weight, M, ~ 18,300) and a-Lactalbumin (20% by 63

94,000 -v. 67,000 — I 43,000 —

30,000 — •

20,100 —

14,400 —

M W B P S C M

Figure 17. SDS-PAGE pattern of food-grade proteinase inhibitors. W, whey protein concentrate; B, beef plasma protein; E, egg white; P, potato powder; S, bovine serum albumin; C, casein; and M, molecular weight standards. Proteins 15 fig, were applied on 15% gel. 64

<-BSA

Figure 18. Proteinase inhibitors stained for papain inhibitory activity at 370C. W, whey protein concentrate; B, beef plasma protein; E, egg white; P, potato powder; S, bovine serum albumin; C, casein; and M, molecular weight standards. Proteins 15 Hg, were applied on 15% gel. 65

^-BSA

M W B E S C M

Figure 19. Proteinase inhibitors stained for trypsin inhibitory activity at 370C. W, whey protein concentrate; B, beef plasma protein; E, egg white; P, potato powder; S, bovine serum albumin; C, casein; and M, molecular weight standards. Proteins 15 jig, were applied on 15% gel. 66 weight, M, ~ 14,000). Serum albumin (10% by weight, M, - 69,000) and immunoglobulin (10% by weight, M, > 145,000) exist as minor proteins. In the control gel major components detected in WPC are a-lactalbumin, P-lactoglobulin,

BSA, and an unidentified high molecular weight protein (HMP) with an apparent M,. >

94,000. There was also a minor band (apparent M, ~ 80,000) shown between BSA and

HMP bands. The latter two showed inhibition towards trypsin (Figure 19), but only

HMP was inhibitory towards papain (Figure 18). The HMP band with inhibitory activity showed the same intensity as the control gel when stained for trypsin inhibitory activity but less intensity when stained for papain inhibitory activity.

Akazawa et al. (1993) reported that a-lactalbumin from WPC did not participate in the inhibition of Pacific whiting proteinase which is in accordance with our findings.

BSA and the HMP, which are also present as major components in BPP, inhibited both papain and trypsin activities (Figures 18 and 19). In papain and trypsin inhibitory gel, the intensity of BSA band was decreased. The intensity of the HMP band was low compared to control in the gel stained for papain inhibitory activity, while it remained same in the gel stained for trypsin inhibitory activity. BPP also had a couple of minor bands between BSA and HMP, but they did not show any inhibitory activity towards papain or trypsin.

A number of protein band were detected in EW (Figure 17). Fossum and

Whitaker (1968) reported the presence of a papain and ficin inhibitor (cystatin) in EW.

In this study, we were unable to detect any specific cysteine proteinase inhibitors possibly due to their low concentrations (Whitaker and Tannenbaum, 1977). Fifteen jag was used for electrophoretic analysis, and the amount of cystatin in the sample is only about 7.5 ng, which might be too low to be detected. Ovalbumin, the major 67 protein in EW, did not show any papain inhibitory activity. Akazawa et al. (1993) reported ovalbumin is not the inhibitory component of chicken EW against Pacific whiting surimi. EW showed numerous trypsin inhibitors, which is in agreement with the results obtained from the trypsin inhibition assay.

Potato is a rich source of proteinase inhibitors (Porter et al., 1993). PP have been shown inhibitory against proteinases from Pacific whiting (Akazawa et al., 1993;

Morrissey et al., 1993) and trout muscle (Kaiser and Belitz, 1973). In the inhibitor activity stained gel, many serine proteinase inhibitors were detected for trypsin, but only one band (apparent Mj ~ 67,000) for papain (Figures 19 and 18). Potato tuber contains a number of protein inhibitors: (1) cysteine proteinase inhibitor (M, ~ 23,000,

Rowan et al., 1990); (2) aspartic proteinase inhibitor (M, ~ 27,000, Mares et al., 1989);

(3) inhibitor I (M, ~ 39,000, Melville and Ryan, 1972) and II (M, ~ 21,000, Bryant et al.,

1976), which are inhibitors of trypsin and chymotrypsin; (4) inhibitors for pancreatic carboxypeptidase A and B (M, ~ 3,100, Ryan et al., 1974), and (5) inhibitor for serine endopeptidases and metallocarboxypeptidases (M,. ~ 5,400, Hass et al., 1976).

According to these data, PP contains a number of serine proteinase inhibitors, which were also seen in the inhibitory gel (Figure 19). There is only one cysteine proteinase inhibitor reported from PP. These data agrees with the results of papain inhibition assay for cysteine proteinase inhibitors. PP contained more cysteine proteinase inhibitors than EW and WPC.

Figure 20 shows the inhibitory gel towards partially purified Pacific whiting proteinase at 370C. Most of the bands were visible in the inhibitory gel, even though the bands from casein are completely disappeared. From these results, it was difficult to conclude which component of the proteinase inhibitors was inhibitory towards the 68

94,000-v. 67,000 — 4-BSA 43,000 —

30,000 —

20,100 —

14,400 —

M W B P S C M

Figure 20. Proteinase inhibitors stained for Pacific whiting proteinase inhibitory activity at 370C. W, whey protein concentrate; B, beef plasma protein; E, egg white; P, potato powder; S, bovine serum albumin; C, casein; and M, molecular weight standards. Proteins 15 \ig, were applied on 15% gel. 69

Pacific whiting proteinase. Specific activity of the Pacific whiting proteinase was low at around 2000 U/mg, compared to papain and trypsin which showed specific activities of 11852 and 11436 U/mg respectively. Although the purified proteinase hydrolysed the casein bands, specific activity could have been too low to hydrolyse the other proteins. Inhibitory gel at 550C showed similar results (Figure 21). Pacific whiting proteinase is reported to have an optimum temperature of 550C (An et al., 1994), but not much difference was shown in the inhibitor activity stained gel as compared to the gel developed at 370C. It could be due to higher heat denaturation rate of the enzyme at the higher temperature, since 90 min incubation time was used.

BSA and HMP in BPP showed inhibitory activity towards papain and trypsin.

To investigate whether concentration of these components have an effect on the inhibition, a series of different concentration of BPP was applied to the gels and tested for inhibition towards trypsin. The stained band intensity decreased with decreased protein concentrations (Figure 22). The intensity of the BSA bands decreased in the inhibitory gel compared to control but not the intensity of the HMP band.

Protein inhibitors showed different degrees of proteinase inhibition towards different types of enzymes. Thus the inhibition by these food grade inhibitors seems to depend on the enzyme source, i.e., fish mince or surimi, due to different protease composition. Morrissey et al. (1993) reported PP had a higher percentage of inhibition than EW in fish mince, and the opposite in surimi. When measured by papain

(BANA) and trypsin (BAPNA) inhibition assays, BPP showed the highest cysteine proteinase inhibition and the second highest serine proteinase inhibition. BPP is reported to contain approximately 50% BSA (Akasawa et al., 1993). If BSA is the main inhibitory component in BPP, it should show higher inhibition than BPP. On 70

94,000-y. 67,000 — — 43,000- mm <-BSA

30,000 — -^°

20,100 —

14,400 —

•

MWBEPSCM

Figure 21. Proteinase inhibitors stained for Pacific whting proteinase inhibitory activity at 550C. W, whey protein concentrate; B, beef plasma protein; E, egg white; P, potato powder; S, bovine serum albumin; C, casein; and M, molecular weight standards. Proteins 15 (ig, were applied on 15% gel. 71

94,000 -\. 67,000 — 4-BSA 43,000 — H I 30,000 —

20,100 —

14,400 —

B

^-BSA

M 1 .5 .1 .05 .01 .005 B

Figure 22. Inhibition of trypsin activity by BPP at various concentrations at 370C. The number designates dilution factors of BPP amount applied, where IX is 25 ng. A, control gel; B, inhibitory gel. 72 autolytic data shown in Figure 10, BSA only had 10% inhibition at 1% level while

BPP had 90% inhibition (Morrissey et al., 1993). On the other hand, WPC containing about 10% BSA had only 50% inhibition at 1% level. In other words, inhibition by

BSA is lower than both WPC and BPP. This result suggests that BSA is not the main inhibitory component in both BPP and WPC, even though it reduces the enzyme activity.

Furthermore, the gel electrophoresis data showed that BSA component in BPP was visible on the gel stained for papain and trypsin inhibitory activities (Figures 18 and 19), but lower intensity was observed than the control gel (Figure 17). In WPC,

BSA component was not seen, which may be due to the lower amount of BSA. The component might have already been hydrolysed by the enzyme. An inhibitor can retard the enzyme activity either by binding to enzyme active site as a specific inhibitor, or binding to other sites and consequently effect the active site, or simply act as a alternative substrate/competitive inhibitor. From the autolysis and electrophoresis data, it can be concluded that BSA is not a true inhibitor for cysteine proteinase or serine proteinase but it can retard the enzyme activity by functioning as an alternative substrate/competitive inhibitor. Similar results were observed for the HMP component in WPC, BPP, and BSA in the gel stained for papain inhibitory activity, indicating

HMP's function as an competitive inhibitor for cysteine proteinase. In contrast, the

HMP component in WPC, BPP and BSA shows inhibition towards trypsin without decreasing the intensity of the band, which indicates true inhibition of serine proteinase.

BSA did not show any inhibition in cysteine and serine proteinase inhibitor assays, in contrast to inhibition of autolytic activity and inhibitory activity staining. 73

Use of synthetic substrates in the enzyme assays compared to protein substrate in the inhibitory activity staining could contribute to this effect. Synthetic substrates are smaller molecules than protein substrates and they may reach the active site even when the enzyme is complexed with a protease inhibitor. The mechanism may be similar to the "trap hypothesis" for the inhibition of proteinase by c^-macroglobulin (Barrett and

Starkey, 1973), where the access of large substrates to the catalytic site are hindered but not the small substrates. Piyachomkwan (1994) reported than WPC protected the

Pacific whiting MHC at 4% level, but BSA did not. Also Morrissey et al. (1993) showed that BPP, which is currently used as a food grade inhibitor in surimi production, protected the Pacific whiting MHC. Based on these data, it is also possible that the inhibitory component in the BPP and WPC is not BSA.

EW inhibited papain activity in BANA assays (Figure 13) and inhibited Pacific whiting proteinase in autolysis assay of fish mince and fresh surimi (Morrissey et al.,

1993). But it did not show any papain-inhibitory component on the inhibitory gel electrophoresis, which may be due to the presence of small amount of cysteine proteinase inhibitors. High molecular weight protein ovostatin (M, ~ 780,000) is also reported to be present in EW, which is homologous to Oj-macroglobulin (Nagase et al., 1985). Ovostatin inhibits a wide range of endoproteinases, including thermosylin and collagenases, which are metallo proteinases. The mechanism of inhibition is similar to the inhibition by OjM (Nagase et al., 1985). This indicates that there can be other components and mechanisms of EW for inhibition of cysteine proteinases.

Hammon et al. (1990) suggested that the inhibiting agent in EW and BPP is o^M, which is a high molecular weight (M, ~ 720,000) glycoprotein. It is known to be present in the plasma of vertebrate and some fish species (Sottrup-Jansen et al., 1990). 74

OjM can inhibit broad range of proteolytic enzymes from the four catalytic classes of enzymes (Starkey and Barrett, 1977). Modori of New Zealand hoki and alaska pollack surimi was inhibited by o^M-rich fraction of BPP (Hammon et al., 1990). Lorier et al.

(1991) also reported that surimi made from Hoki was treated with OjM and it had better gel stress and strain values compared with surimi not treated with OjM. Wasson et al. (1992b) showed that gel strength of arrowtooth flounder surimi can be increased by adding o^M. No evidence of myosin degradation was observed on electrophoresis for the surimi treated with o^M. Since o^M does not form a gel, they concluded that the increase in gel strength is primarily due to the inhibition of the proteinase.

Serum was reported (Akazawa et al., 1993) to contain another family of protein inhibitors known as kininogens, which are high molecular weight cysteine proteinase inhibitors. They are specific for cathepsin B, H, and L and belong to cystatin super-family. For BPP and EW, they can be one of the high molecular weight protein inhibitors to contribute to the inhibitory effect.

The mechanism of inhibition by WPC is not clear, and limited number of reports are available for o^M in whey. Lelievre et al. (1990) reported the inhibition of calf veil and microbial rennet action by a protein in WPC, which had properties of otjM. Rantamaki et al. (1992) reported the presence of OjM in mastitis whey. But o^M was not detected in normal whey by SDS-PAGE or Western blotting. 75

CONCLUSIONS

Cathepsin B was the most active cysteine proteinase in fish fillets but was selectively removed along with cathepsin H by washing during surimi processing.

Cathepsin L, whose highest activity was observed at 550C, was the most predominant cysteine proteinase in surimi. Autolysis of fresh surimi showed that MHC was completely hydrolyzed within 20 min at 550C. The proteinase purified from Pacific whiting actively hydrolyzed all the Pacific whiting muscle components studied, i.e., myofibrils, myosin, and heat-denatured and native collagen at room temperature. The degradation pattern of myofibrils by the proteinase was the same as the autolysis pattern of surimi.

Inhibition by the PI varies with the catalytic types of the enzyme. BPP had a higher percentage of papain inhibitors followed by WPC, PP, and EW, while EW had a higher percentage of trypsin inhibitors followed by BPP, PP, and WPC. EW inhibits trypsin completely at levels as low as 1%. WPC inhibited the autolytic activity of fresh surimi, and BSA too but to a less degree. By using high levels of WPC, inhibition comparable to 1% BPP can be achieved. WPC may be used as an inhibitor for the Pacific whiting surimi.

Both EW and PP showed more serine proteinase inhibitors than cysteine proteinase inhibitors on the inhibitory activity-stained gels. PP showed one non- specific cysteine inhibitory component (apparent M,. ~ 61,000) which is also inhibitory towards trypsin. EW did not show any cysteine proteinase inhibitors. In BPP and

WPC, BSA is not the main inhibitory component but it can reduce proteinase activity by acting as an alternate substrate. HMP found in WPC, BPP, and BSA functions as 76 an alternative substrate towards papain, but it is a specific inhibitor towards trypsin.

Specific activity of the purified proteinase was insufficient to test for the inhibitory effect. Use of a partially purified active component may increase inhibitory effects and lower the required amount of additives. 77

BIBLIOGRAPHY

Akazawa, H., Miyauchi, Y., Sakurada, K, Wasson, D.H., and Reppond, K.D. 1993. Evaluation of protease inhibitors in Pacific whiting surimi. J. Aquat. Food Prod. Technol. 2(3): 79-95.

An, H., Seymour, T.A., Wu, J.W., and Morrissey, M.T. 1994. Assay systems and characterization of pacific whiting (Merluccius productus) protease. J. Food Sci. 59: 277-281.

An, H. and Weerasinghe, V. 1993. Unpublished data. Oregon State University Seafood Laboratory, Astoria, OR.

Anastasi, A., Brown, M.A., Kembhavi, A.A., Nicklin, M.J.H., Sayers, C.A., Sunter, D.C., and Barrett, A.J. 1983. Cystatin, a protein inhibitor of cysteine proteinases. Biochem. J. 211: 129-138.

Amon, R. 1970. Papain. Methods in Enzymology 19: 226-244.

Asghar, A. and Bhatti, A.R. 1987. Endogenous proteolytic enzymes in skeletal muscle: their significance in muscle physiology and during postmortem aging events in carcasses. Adv. Food Res. 31: 343-451.

Bailey, K.M. and Francis, R.C. 1985. Recruitment of Pacific whiting, Merluccius productus, and the Ocean Environment. In Pacific -whiting. Marine Fisheries Review. 47(2): 8-15.

Barrett, A. J. 1972. A New Assay for Cathepsin Bl and other thiol proteinases. Anal. Biochem. 47: 280-293.

Barrett, A.J. 1981. Cystatin, the Egg White Inhibitor of Cysteine Proteinases. Methods in Enzymology 80: 771-778

Barrett, A. J. and Kirschke, H. 1981. Cathepsin B, Cathepsin H, and Cathepsin L. Methods in Enzymology 80: 535-561.

Barrett, A.J. and Starkey, P.M. 1973. The interaction of o^-Macroglobulin with Proteinases. Biochem J. 133: 709-724.

Boye, S.W. and Lanier, T.C. 1988. Effects of heat-stable alkaline protease activity of Atlantic menhaden {Brevoorti tyrannus) on surimi gels. J. Food Sci. 53: 1340- 1342.

Bryant, J., Grreen, T.R., Gurusaddaiah, T., and Ryan, C.A. 1976. Proteinase inhibitor II from potatoes: Isolation and characterization of its protomer components. Biochem. 15(16): 3418-3423. 78

Burgarella, J.C., Lanier, T.C., and Hamann, D.D. 1985. Effects of added egg white or whey protein concentrate on thermal transition in rigidity of croaker surimi. J. Food Sci. 50: 1588-1594,1606.

Chang-Lee, M.V., Lampila, L.E., and Crawford, D.L. 1990. Yield and composition of surimi from Pacific whiting (Merluccius productus) and the effect of various protein additives on gel strength. J. Food Sci. 55: 83-86.

Chang-Lee, M.V., Pacheco-Aguilar, R., Crawford, D.L., and Lampila, L.E. 1989. Proteolytic activity of surimi from Pacific whiting (Merluccius productus) and heat-set gel texture. J. Food Sci. 54: 1116-1119,1124.

Chemikov, M.P., Stan, E.Y., Smimova, L.I., and Usachewa, NT. 1978. Brief Communications, E, 20th Intern. Dairy Congress 917-918.

Chung, Y.C., Richardson, L., and Morrissey, M.T. 1993. Effects of pH and NaCl on Gel strength of Pacific whiting surimi. J. Aquat. Food Prod. Tech. 2(3): 19-35.

Dunker, J.M. 1984. The Future of Whey protein concentrate. Proceedings. Whey Protein Conference. Chicago, IL. Oct. 25-26. 35-39.

Erickson, M.C., Gordon, D.T., and Anglemier, A.F. 1983. Proteolytic activity in the sarcoplasmic fluids of parasitized Pacific whiting {Merluccius productus) and unparasitized True cod (Gadus marcocephalus). J. Food Sci. 48: 1315-1319

Farouk, M.M., Price, J.F., and Salih, A.M. 1992. Post-exsanguination infusion of ovine carcasses: effect on tenderness indicators and muscle microstructure. J. Food Sci. 57: 1311-1315.

Fossum, K. and Whitaker, J.R. 1968. Ficin and papain inhibitor from chicken egg white. Arch. Biochem. Biophys. 125: 367-375.

Garcia-Carreno, F., Dimes, L.E., and Haard, N.F. 1993. Substrate-gel electrophoresis for composition and molecular weight of proteinases or proteinaceous proteinase inhibitors. Anal. Biochem. 214: 65-69.

Giese, J. 1994. Proteins as Ingredients: Types, function, applications. Food Tech. 48(10): 50-60.

Gillies, M.T. 1974. Whey processing and Utilization - Economic and Technical Aspects. Noyes Data Corp., Park Ridge, NJ.

Greene, D.H. and Babbit, J. 1990. Control of muscle softening and pro tease-parasite interactions in arrowtooth flounder (Atheresthes stomias). J. Food Sci. 55: 579- 580. 79

Hamann, D.D., Amato, P.M., Wu, M.C., and Foegeding, E.A. 1990. Inhibition of Modori (Gel weakening) in surimi by plasma hydrolysate and egg white. J. Food Sci. 55: 665-669,795.

Hass, G.M., Venkatakrishnan, R., and Ryan, C.A. 1976. Homologous inhibitors from potato tubers of serine endopeptidases and metallocarboxypeptidases. Proc. Natl. Acad. Sci. USA. 73(6): 1941-1944.

Heumann, D., Burger, D., and Vischer, T.L. 1989. Human Oj-macroglobulin as an inhibitor of insoluble trypsin. Biochem. Biophys. Research Comm. 165(1): 14- 19.

Hudson, N.W., Kehoe, J. M., and Koo, P.H. 1987. Mouse a-macroglobulin. Biochem. J. 248: 837-845.

Iwata, K., Kobashi, K., and Hase J. 1973. Studies of muscle alkaline protease-I. Isolation, purification and some physicochemical properties of an alkaline protease from carp muscle. Bull. Japan. Soc. Sci. Fish. 39(12): 1325-1337.

Iwata, K., Kobashi, K., and Hase J. 1974. Studies of muscle alkaline protease-II. Some enzymatic properties of carp muscular alkaline protease. Bull. Japan. Soc. Sci. Fish. 40(2): 189-200.

Izquierdo-Pulido, MX., Haard, T.A., Hung, J., and Haard, N.F. 1994. Oryzacystatin and other proteinase inhibitors in rice grain: Potential use as a fish processing aid. J. Agric. Food. Chem. 42: 616-622.

Jiang, ST., Lee, J.J., and Chen, H.C. 1994. Purification and Characterization of Cathepsin B from Ordinary Muscle of mackerel (Scomber austmlasicus). J. Agric. Food Chem. 42: 1073-1079.

Jiang, ST., Nei, F.P., Chen, H.C, and Wang, J.H. 1992b. Comparative study on the cathepsin D from Banded Shrimp (Penaeus japonicus) and Grass shrimp (Penaeus monodon). J. Agric. Food Chem. 40: 961-966.

Jiang, ST., Wang, Y.T., and Chen, C.S. 1992a. Lysosomal enzyme effects on the postmortem changes in Tilapia (Tilapia nilotica X T.aured) muscle myofibrils. J. Food Sci. 57(2): 277-279,282.

Jiang, ST., Wang, Y.T., Gau, B.S., and Chen, C.S. 1990. Role of pepstatin-sensitive proteases on the post-mortem changes of tilapia muscle myofibrils. J. Agric. Food Chem. 38: 1664-1668.

Kabata, Z. and Whitaker, D.J. 1985. Parasites as a limiting factor in exploitation of pacific whiting (Merluccius productus). Mar. Fish. Rev. 47(2): 55-59. 80

Kaiser, K.P. and Belitz, H.D. 1973. Specificity of potato isoinhibitors towards various proteolytic enzymes. Z. Lebensm. Unters. Forsch. 151(1): 18-22.

Kassell, B. 1970. Proteinase inhibitors from egg white. Methods in Enzymology 19: 890-906.

Kay, J., Siemankowski, L.M., Siemankowski, R.F., Greweling, J.A., and Goll, D.E. 1982. Degradation of myofibrillar proteins by trypsin-like serine proteinases. Biochem. J. 201: 279-285.

Kinoshita, M., Toyohara, H., Shimizu, Y., and Sakaguchi, M. 1991. Induction of Modori- phenomenon (Thermal gel degradation) by a latent serine proteinase. Nippon Suisan Gakkaishi 57(10): 1935-1938.

Kirschke, H and Barrett, A.J. 1985. Cathepsin L - A lysosomal cysteine proteinase. In Intnacellular protein catabolism. 61-69. Alan R. Liss, Inc. NY.

Kirschke, H. and Barrett, A.J. 1987. Chemistry of Lysosomal Proteases. In Lysosomes: Their role in Protein Breakdown. Academic Press, NY.

Kirschke, H., Kembhavi, A.A., Bohley, P., and Barrett, A.J. 1982. Action of rat liver cathepsin L on collagen and other substrates. Biochem. J. 201: 367-372.

Kirschke, H., Langner, J., Wiederanders, B., Ansorge, S., and Bohley, P. 1977. Cathepsin L: A new proteinase from rat-liver lysosomes. Eur. J. Biochem. 74: 293-301.

Koga, H., Yamada, H., Nishimura, Y., Kato, K., and Imoto, T. 1990. Comparative study on specificities of rat cathepsin L and papain: Amino acid differences at substrate-binding sites are involved inthier specificities. J. Biochem. 108: 976- 982.

Konagaya, T., and Aoki, T. 1981. Jellied condition in Pacific hake in relation to Myoxosporidian infection and protease activity. Bull. Tokai Reg. Fish. Res. Lab. 105: 1-6.

Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 860-685.

Lanier, T.C. 1986. Functional properties of surimi. J. Food Sci. 40: 107-114.

Laskowski, M., Jr., and Kato, I. 1980. Protein inhibitors of proteinases. Ann. Rev. Biochem. 49: 593-626.

Lee, CM. 1984. Surimi Process Technology. Food Tech. 38(11): 69-80. 81

Lee, J.J., Chen, H.C., and Jiang, S.T. 1993. Purification and characterization of proteinases identified as cathepsin L and L-like (58kDa) proteinase from mackerel (Scomber austrcdasicus). Biosci. Biotech. Biochem. 57(9): 1470-1476.

Lelievre, J., Creamer, L.K., and Tate, K.L. 1990. Inhibition of calf veil and microbial rennet action by whey protein concentrate. Milchwissenschaft, Milk Science International 45(2): 69-136.

Lin, D., Morrissey, M.T., Richardson, L., and Aksomsri, W. 1993. Shelflife of fresh Pacific whiting surimi (Merluccius productus). Presented at the annual meeting of Pacific Fisheries Technologists in January, Oak Harbor, WA.

Lin, T.S. and Lanier, T.C. 1980. Properties of an alkaline protease from the skeletal muscle of Atlantic croaker. J. Food Biochem. 4: 17-28.

Lorier, M.A. and Aitken, B.L. 1991. Method for treating fish with alpha-2- macroglobulin. U.S. patent 5,013,568.

Lowry, OH., Rosebrough, N.J., Farr, A.L., and Randall, R.J. 1951. Protein measurement with Folin phenol reagent. J. Biol. Chem. 193: 256-275.

Mahmoud, R. and Savello, P. A. 1992. Mechanical properties of and water vapor transferability through whey protein films. J. Dairy Sci. 75: 942-946.

Makinodan, Y. and Ikeda, S. 1969. Studies on fish muscle protease-I. On the existence of two kinds of proteinases active in acid and in slightly alkaline pH range. Bull Jap. Soc. Sci. Fish 35: 672-676.

Makinodan, Y., Kitagawa, T., Toyohara, H, and Shimizu. 1987a. Participation of Muscle Alkaline proteinase in Himodori of Kamaboko. Nippon Suisan Gakkaishi 53(1) 99-101.

Makinodan, Y., Yokiyama, Y., Kinoshita, M., and Toyohara, H. 1987b. Characterization of an alkaline protease of fish muscle. Comp. Biochem. Physiol. 87B: 1041-1046.

Mares, M., Meloun, B., Pavlik, M., Kostka, V., and Bandys, M. 1989. Primary structure of cathepsin D inhibitor from potatoes and its structure relationship to soybean trypsin inhibitor family. FEBS Lett. 251: 94-98.

Martone, C.B., Busconi, L., Folco, E.J., and Sanchez, J.J. 1991. Detection of a trypsin-like serine protease and its endogenous inhibitor in hake skeletal muscle. Arch. Biochem. Biophys. 289: 1-5.

Martone, C.B., Busconi, L., Foldo, E.J., Trucco, RE., and Sanchez, J.J. 1986. A simplified myosin preparation from marine fish species. J. Food Sci. 51: 1554- 1555. 82

Masaki, T., Shimomukai, M., Miyauchi, Y., Ono, S., Tuchiya, T, Mastuda, T., Akazawa, H., and Soejima, M. 1993. Isolation and characterization of the protease responsible for jellification of Pacific hake muscle. Nippon Suisan Gakkaishi 59: 683-690.

Mason, R.W. 1986. Species variants of cathepsin L and their immunological identification. Biochem. J. 240: 285-288.

Mason, R.W. 1989. Interaction of Lysosomal cysteine proteinases with OjM: conclusive evidence for the endopeptidase activities of cathepsin B and H. Archives of Biochem. and Biophysics. 273(2): 367-374.

Mason, R.W., Green, G.D.J., and Barrett, A.J. 1985. Human liver cathepsin L. Biochem. J. 226: 233-241.

Mason, R.W., Taylor, A.J., and Etherington, D.J. 1984. The purification and properties of cathepsin L from rabbit liver. Biochem J. 217: 209-217.

Melville, J.C. and Ryan, C.A. 1972. Chymotrypsin inhibitor I from Potatoes. J. Biol. Chem. 247(11): 3445-3453.

Meredith, C. 1994. Personnel communication. New Zealand Milk Products (North America) Inc. Santa Rosa, CA.

Miller, R. and J. Spinelli. 1982. The effect of protease inhibitors on proteolysis in parasitized Pacific whiting, Merluccius productus, muscle. Fish. Bull. 80(2): 281-286.

Morr, C.V. 1985. Composition, physicochemical and functional properties of reference whey protein concentrates. J. Food Sci. 50: 1406-1421.

Morr, C.V. 1989. Whey Proteins: Manufacture. In Developments in Dairy Chemistry -4. Functional milk proteins. F'ox, P^.F. 245-284. Elsevier Applied Science, London and New York.

Morr, C.V. 1992. Improving the Texture and Functionality of Whey Protein Concentrate. Food Tech. 46(1): 110-113.

Morr, C.V. and Foegeding, E.A. 1990. Composition and functionality of commercial whey and milk protein concentrates and isolates: A status report. Food Tech. 44(4): 100-112.

Morrissey, M.T., Wu, J.W., Lin, D.D., and An, H. 1993. Effect of food grade protease inhibitor on autolysis and gel strength of surimi. J. Food Sci. 58: 1050-1054. 83

Nagahisa, E., Yasuda, S., Nishinura, S., and Aoki, S. 1983. Jellification of Myxosporidia - infected Pacific hake muscle and its protease activity. Bull. Jpn. Soc. Sci. Fish. 49(12): 15-23.

Nagase, H., Brew, K., and Harris, ED. Jr. 1985. Ovostatin: A proteinase inhibitor in egg white that is homologous to otj-macroglobulin. In Intracellular Protein Catabolism. p. 283-285. Alan R. Liss, Inc. NY.

Ndi, E.E. and Brekke, C.J. 1992. Thermal aggregation properties of duck salt-soluble proteins at selected pH values. J. Food Sci. 57: 1316-1320.

Nishimoto, S.I., Hashimoto, A., Seki, N., Kimura, I., Toyoda, K., Fujita, T., and Arai, K. I. 1987. Influencing factors on changes in myosin heavy chain and jelly strength of salted meat paste from Alaska pollack during setting. Nippon Suisan Gakkaishi 53(11): 2011-2020.

Niwa, E. 1992. Chemistry of surimi gelation. In Surimi Technology. T.C. Lanier and CM. Lee (Ed.) Marcel Dekker, Inc. Newyork, NY.

Ohshima, T., Suzuki, T., and Koizumi C. 1993. New developments in surimi technology. Trends in Food Sci. & Tech. 4: 157-163.

Okitani, A., Matsukura, U., Kato, K., and Fujimaki, M. 1980. Purification and some properties of a myofibrillar protein-degrading protease, cathepsin L, from rabbit skeletal muscle. J. Biochem. 87: 1133-1143.

PFMC. 1992. Pacific whiting allocation. Pacific Fishery Management Council (PFMC). Portland OR.

Patashnik, M., Groninger, H.S., Bamett, H., Kudo, G., and Koury, B. 1982. Pacific whiting, Merluccius productus: I. Abnormal muscle texture caused by Myxosporidian-induced proteolysis. Marine Fish. Rev. 44(5): 1-12.

Pike, N.R., Coetzer, T.H.T., and Dennison, C. 1992. Proteolytically active complexes of cathepsin L and a cysteine proteinase inhibitor; Purification and demonstration of their formation in vitro. Arch. Biochem. Biophy. 294 (2): 623-629.

Piyachomkwan, K. 1994. Apparent inhibition of Pacific whiting surimi-associated protease by whey protein concentrate. MS thesis, Dept of Food Science and Technology, Oregon State University, Corvallis, OR.

Porter, R.W., Koury, B.J., and Kudo, G. 1990. Method for treating fish meat contaminated with sporozoa and potato product for improving heat gelation of fish muscle. U.S. patent 4,935,192. 84

Porter, R., Koury, B., and Kudo, G. 1993. Inhibition of Protease Activity in Muscle Extracts and Surimi from Pacific Whiting, Merluccius productus, and Arrowtooth Flounder, A theresthes stomias. Marine Fish. Rev. 55(3): 10-15.

Rantamaki, L.K. and Muller, HP. 1992. Isolation and characterization of o^- macroglobulin from mastitis milk. J. Dairy Res. 59: 273-285.

Reppond, K. D. and Babbitt, J. K. 1993. Protease inhibitors Affect Physical Properties of Arrowtooth flounder and Walleye Pollock Surimi. J. Food Sci. 58: 96-98.

Robe, G.H. and Xiong, Y.L. 1992. Phosphates and muscle fiber type influence thermal transitions in porcine salt-soluble protein aggregation. J. Food Sci. 57: 1304-1307, 1310.

Rowan, A.D., Brzin, J., Buttle, D.J., and Barrett, AJ. 1990. Inhibition of cysteine proteinases by a protein inhibitor from potato. FEBS. 269(2): 328-330.

Ryan, C.L. 1974. Purification and properties of a carboxypeptidase inhibitor from potatoes. J. Biol. Chem. 249(17): 5495-5499.

Ryder, D.N. 1991. Study of the use of milk proteins in fish gel products. I.I.F - I.I.R. - Commission C2 - Aberdeen (UK) 1990-3: 107-113.

Salvesen, G.S. and Barrett, A.J. 1980. Covalent binding of proteinases in their reaction with OjMacroglobulin. Biochem. J. 187: 695-701.

Salvesen, G.S., Sayers, C.A., and Barrett, A.J. 1981. Further characterization of covalent linking reaction of OjMacroglobulin. Biochem. J. 195: 453-461.

Sato, K., Ohashi, C, Ohtsuki, K., and Kawabata, M. 1991. Type V collagen in trout (Salmo gcdrdneri) muscle and its solubility change during chilled storage of muscle. J. Agric. Food Chem. 39: 1222-1225.

Seymour, T.A., Morrissey, M.T., Peters, MY., and An, H. 1994. Purification and characterization of Pacific whiting proteases. J. Agric. Food Chem. 42(11): 2421-2427.

Sick, H., Kaiser, C, and Henne, A. 1990. The action of glycomacropeptide (GMP) on trypsin activity in vitro. Milchwissenschaft, Milk Science International 45 (2): 76-78.

Smith, C, Megen, W.V., Twaalfhoven, L., and Hitchcock, C. 1980. The determination of trypsin inhibitor levels in foodstuffs. J. Sci. Food Agric. 31: 341-350. 85

Sottrup-Jensen, L., Borth, W., Hall, M., Quigley, IP., and Armstrong, P.B. 1990. Sequence similarity between Oj-macroglobulin from horseshoe crab, LimuJus polyphemus, and proteins of the ctj-macroglobulin family from mammals. Comp. Biochem. Physiol. 96B(3): 621-625.

Starkey, P.M. and Barrett, A.J. 1977. Oj-Macroglobulin, a physiological regulator of proteinase activity. In proteinases in mammalian cells and tissues. Ch. 16. North-Holland Publishing Company.

Stehr, C, Whitaker, D.J. 1986. Host parasite interaction of the myxosporidian Kudoa paniformis (Kabata and Whitaker, 1981) and Kudoa thyrsites (Gilchrist, 1924) in the muscle of Pacific whiting, Merluccius productus (ayres): an ultra- structural study. J. Fish Dis. 9: 505-517.

Stevens, L. 1991. Egg white proteins. Comp. Biochem. Physiol. 100B (1): 1-9.

Su, H., Lin, T.S., and Lanier, T.C. 1981. Investigation into potential sources of heat- stable alkaline protease in mechanically separated Atlantic croaker (Micropagon unduJatus). J. Food Sci. 46, 1654-1656,1664.

Talley, K. 1995. Stats pack 1995. Pacific fishing. March: 55-79.

Wasson, D. H., Babbitt, J. K., and French, J.S. 1992a. Characterization of a heat stable protease from arrowtooth flounder, Aiheresthes stomias. J. Aquat. Food Prod. Technol. 1(3/4): 167-182.

Wasson, D.H., Reppond, K. D., Babbitt, J. K., and French, J. S. 1992b. Effects of additives on proteolytic and functional properties of arrowtooth flounder surimi. J. Aquat. Food Prod. Technol. 1(3/4): 147-165.

Webb, B.H., Johnson, AH, and Alford, J.A. 1974. Fundermentals of Dairy Chemistry. Avi Publishing Co., Westport, CT.

Weber, K. and Osbom, M. 1975. Proteins and sodium dodecyl sulfate: Molecular weight determination of polyacrylamide gels and related procedures. Ch. 3. In The Proteins, 3rd ed. H. Neurath and R.L. Hill (Ed), pp. 179-223. Academic Press, New York, NY.

Whitaker, J. R. and Tannenbaum, S. R. 1977. Food Proteins. Avi Publishing Company, INC. Westport, Connecticut.

White, A, Handler, P., and Smith, E.L. 1973. Principles of Biochemistry. McGraw- Hill Book Company.

Wu, J.W. 1994. Kinetic properties and characterization of purified protease from Pacific whiting {Merluccius productus). MS. thesis, Dept. of Food Scienceand Technology, Oregon State University, Corvallis OR. 86

Yamashita, M. and Konagaya, S. 1990a. Purification and characterization of cathepsin L from the white muscle of chum salmon, Oncorhynchu keta. Comp. Biochem. Physiol. 96B (2): 247-252.

Yamashita, M. and Konagaya, S. 1990b. Purification and characterization of cathepsin B from the white muscle of chum salmon, Oncorhynchu keta. Comp. Biochem. Physiol. 96B (4): 733-737.

Yamashita, M. and Konagaya, S. 1990c. Participation of Cathepsin L into Extensive Softening of the Muscle of Chum Salmon Caught during Spawning Migration. Nippon Suisan Gakkaishi 56(8): 1271-1277.

Yamashita, M. and Konagaya, S. 1991. Hydrolytic action of salmon cathepsin B and L to muscle structural proteins in respect of muscle softening. Nippon Suisan Gakkaishi 57(10): 1917-1922.