Citation
Bioactive and Functional Properties of Catfish Protein Hydrolysates and Catfish Protein Isolates

Material Information

Title:
Bioactive and Functional Properties of Catfish Protein Hydrolysates and Catfish Protein Isolates
Copyright Date:
2008

Subjects

Subjects / Keywords:
Amino acids ( jstor )
Antioxidants ( jstor )
Catfish ( jstor )
Enzymes ( jstor )
Gels ( jstor )
Hydrolysis ( jstor )
Lipids ( jstor )
pH ( jstor )
Solubility ( jstor )
Surimi ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Embargo Date:
7/30/2007

Downloads

This item has the following downloads:


Full Text

PAGE 1

BIOACTIVE AND FUNCTIONAL PROPERTIES OF CATFISH PROTEIN HYDROLYSATES AND CATFISH PROTEIN ISOLATES By ANN ELIZABETH THEODORE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Ann Elizabeth Theodore

PAGE 3

To all who have participated in my educational journey

PAGE 4

ACKNOWLEDGMENTS I would like to thank Dr. Hordur G. Kristinsson for inviting me to be one of his first undergraduate research assistants and for encouraging me to continue onward with my education and having me as a graduate student. The knowledge, skills, and confidence gained from working with Dr. Kristinsson will always play a key role in my professional life. I would also like to thank my other committee members, Dr. Charles Sims and Dr. Bruce Welt, for their valuable suggestions and ideas. Other faculty who were especially supportive, encouraging and helpful throughout my graduate school experience include Dr. Ross Brown, Dr. Steve Talcott, and Dr. Doug Archer. I would like to thank my fianc, Tom Ballesteros, for his patience, kindness, and love throughout my graduate experience. I would like to thank my two best friends, Swapna Mony and Bergros Ingadottir, for their constant support, friendship, and humor. I would like to thank my parents and grandparents for always supporting my educational endeavors. Special thanks go to the Ballesteros family for their love and encouragement. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT.......................................................................................................................xi CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................4 Catfish Industry Overview............................................................................................4 Raw Material Inherent Complications.......................................................................5 Protein Extraction Techniques......................................................................................7 Surimi....................................................................................................................7 Acid and Alkali-Aided Protein Isolation...............................................................9 Comparing the Processes.....................................................................................10 Protein Solubility........................................................................................................13 Quality Measurements................................................................................................14 Lipid Oxidation...................................................................................................14 Color....................................................................................................................14 Protein Gelation..........................................................................................................15 Setting..................................................................................................................16 General Mechanisms...........................................................................................16 Enzymatic Hydrolysis.................................................................................................19 Enzyme Source....................................................................................................20 FPH Antioxidant Properties........................................................................................21 Peptide Concentration.........................................................................................21 Peptide Size.........................................................................................................22 Peptide Composition and Structure Effects on Antioxidant Activity Mechanisms.....................................................................................................23 FPH Physiological Advantage....................................................................................24 Renin-angiotensin-aldosterone system................................................................24 Angiotensin converting enzyme (ACE) inhibition in medicine..........................25 ACE Inactivation by Hydrolysates......................................................................26 v

PAGE 6

Objectives...................................................................................................................27 Significance................................................................................................................27 3 MATERIALS AND METHODS...............................................................................29 Raw Material Acquisition and Handling....................................................................29 Protein Solubility........................................................................................................29 Protein Isolation..........................................................................................................30 Protein Isolate Preparation..................................................................................30 Surimi Processing................................................................................................31 Product Composition..................................................................................................31 Proximate Analysis..............................................................................................31 Protein Electrophoresis: SDS-PAGE..................................................................33 Product Quality Assessments.....................................................................................34 Color Analysis.....................................................................................................34 Lipid Oxidation Products Measurements............................................................34 Gel Forming Ability of Surimi and Protein Isolate Pastes.........................................34 Hydrolysate Preparation.............................................................................................35 Protein Isolation...................................................................................................35 Enzymatic Hydrolysis.........................................................................................35 Sample Preparation..............................................................................................37 Hydrolysate Composition...........................................................................................37 SDS-PAGE Protein Electrophoresis....................................................................37 Amino Acid Analysis..........................................................................................38 Antioxidant Activity...................................................................................................39 ,-diphenyl--picrylhydrazyl (DPPH) Radical Scavenging.............................39 Hydroperoxide System........................................................................................39 Metal Ion Chelating Activity...............................................................................40 Oxygen Radical Absorbance Capacity (ORAC).................................................41 Reducing Power...................................................................................................42 Physiological Activity................................................................................................43 Statistics......................................................................................................................44 4 CATFISH HOMOGENATE SOLUBILI TY AS A FUNCTION OF pH................45 5 SURIMI AND ISOLATE COMPOSITION AND QUALITY..................................49 Composition and Recoveries......................................................................................49 Lipid Reduction...................................................................................................49 Protein Recovery.................................................................................................50 SDS-PAGE..........................................................................................................52 Oxidation Stability and Color.....................................................................................56 Thiobarbituric Acid Reactive Substances (TBARS)...........................................56 Color....................................................................................................................59 vi

PAGE 7

6 THERMAL GELATION OF THE EXTRACTED PROTEINS................................62 Gelation of Extracted Proteins at Different pH Values in the Presence of NaCl.......62 Gelation of Extracted Proteins as a Function of NaCl Concentration........................64 Possible Molecular Changes during Heating and Cooling.........................................69 7 HYDROLYSATE AND SUPERNATANT COMPOSITION...................................71 Hydrolysis of Catfish Proteins....................................................................................71 SDS-PAGE Analysis..................................................................................................71 Amino Acid and Protein Composition.......................................................................75 8 HYDROLYSATE AND SUPERNATANT BIOACTIVE PROPERTIES................78 DPPH Radical Scavenging.........................................................................................78 Peroxidation System...................................................................................................81 Metal Ion Chelating Activity......................................................................................84 Oxygen Radical Absorbance Capacity (ORAC)........................................................86 Reducing Power..........................................................................................................87 9 CONCLUSIONS........................................................................................................93 LIST OF REFERENCES...................................................................................................95 BIOGRAPHICAL SKETCH...........................................................................................107 vii

PAGE 8

LIST OF TABLES Table page 5-1 Protein recovery and lipid reduction for the surimi and protein isolate processes...52 5-2 Color values of surimi and protein isolates..............................................................61 7-1 Amino acid composition (%) and protein content of the hydrolysates at varying degrees of hydrolysis................................................................................................76 7-2 Amino acid composition (%) and protein content of the hydrolysate supernatants at varying degrees of hydrolysis...............................................................................76 viii

PAGE 9

LIST OF FIGURES Figure page 2-1 General surimi processing scheme.............................................................................8 2-2 Acid and alkali-aided protein isolation method.......................................................10 3-2 Hydrolysis reaction system......................................................................................37 4-1 The solubility of homogenized catfish muscle as a function of pH.........................46 5-1 The protein composition of the different fractions obtained by the two versions of the (a) acid-aided and (b) alkali-aided protein solubilization/precipitation processes as assessed with SDS-PAGE...................................................................55 5-2 The protein composition of washed ground catfish muscle and wash water as assessed with SDS-PAGE........................................................................................56 5-3 Secondary lipid oxidation products (TBARS) in minced catfish muscle, surimi, acid PI, and alkali PI over a 12 day time period at 4C...........................................59 6-1 Gel rigidity as measured by storage modulus (G), of 8% protein pastes at 500 mM NaCl and pH 6.5 made from surimi and acid and alkali PI..............................65 6-2 Gel rigidity as measured by storage modulus (G)..................................................68 6-3 Gel formation mechanism of the pastes at pH 6.5 with 500 mM NaCl...................70 7-1 Effect of hydrolysis time on the degree of hydrolysis (%DH) of the catfish protein isolates (3%) using 1% (w/v) Protamex for 15% and 30% DH samples and 0.08% (w/v) Protamex for 5% DH samples...................................................72 7-2 The protein composition of the (a) hydrolysates and (b) supernatants from the hydrolysates as assessed with pre-cast Tris-HCl 4-20% linear gradient SDS-PAGE gels................................................................................................................73 7-3 The protein composition of the (a) hydrolysates and (b) supernatants from the hydrolysates as assessed with pre-cast Tris-Tricine 10-20% linear gradient SDS-PAGE gels................................................................................................................74 ix

PAGE 10

8-1 DPPH radical scavenging activity of the hydrolysates at varying degrees of hydrolysis at 0.15% protein concentration...............................................................79 8-2 DPPH radical scavenging activity of the hydrolysate supernatants at varying degrees of hydrolysis at 0.15% protein concentration.............................................80 8-3 Relative hydroperoxides remaining before and after heating the hydrolysates at varying degrees of hydrolysis in a hydroperoxide generating system (0.15% protein concentration)..............................................................................................82 8-4 Relative hydroperoxides remaining before and after heating the hydrolysate supernatants at varying degrees of hydrolysis in a hydroperoxide generating system (0.15% protein concentration)......................................................................83 8-5 Metal ion chelating activity of the hydrolysates at varying degrees of hydrolysis at 0.15% protein concentration.................................................................................85 8-6 Metal ion chelating activity of the hydrolysate supernatants at varying degrees of hydrolysis at 0.15% protein concentration...........................................................86 8-7 The antioxidant activity (Trolox equivalence) of the hydrolysate supernatants at varying degrees of hydrolysis (1.5% protein concentration) as assessed by the ORAC method..........................................................................................................88 8-8 The antioxidant activity (Trolox equivalence) of the hydrolysate supernatants at varying degrees of hydrolysis at (1.5% protein concentration) as assessed by the ORAC method..........................................................................................................89 8-9 Reducing power of the hydrolysates at varying degrees of hydrolysis at 0.15% protein concentration................................................................................................89 8-10 Reducing power of the hydrolysate supernatants at varying degrees of hydrolysis at 0.15% protein concentration.................................................................................90 8-11 ACE inhibition of the hydrolysates at varying degrees of hydrolysis at 0.15% protein concentration................................................................................................91 8-12 ACE inhibition of the hydrolysate supernatants at varying degrees of hydrolysis at 0.15% protein concentration.................................................................................92 x

PAGE 11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science BIOACTIVE AND FUNCTIONAL PROPERTIES OF CATFISH PROTEIN HYDROLYSATES AND CATFISH PROTEIN ISOLATES By Ann Elizabeth Theodore August 2005 Chair: Hordur G. Kristinsson Major Department: Food Science and Human Nutrition There is a large demand for fish and fish products in the world. Unfortunately, this high demand for fish has stressed several popular fish species. Therefore, it is important to explore the utilization of alternative, less desirable species or byproducts of desirable species. Isolating proteins from undesirable sources of fish is a conscientious effort in using the aquatic resources wisely and adds economic value to raw materials of little to no value. Enzymatic hydrolysis of fish byproducts is an excellent means to reduce the waste refining costs and to create profitable products from low value byproducts. However, this method is difficult to use on many raw mate rials, partly because of microbiological problems, and partly because of color and oxidation problems (since lipids and pro-oxidants ar e not effectively removed). A hi ghly successful process was developed recently in which proteins can be extracted fr om fish byproducts using either high pH (alkali-aided isolation process) or low pH ( acid-aided isolation process). In the process, the proteins are xi

PAGE 12

separated from undesirable components of byproducts (e.g., bones, scales, connective tissue, and lipids). Applying enzymatic hydrolysis to the isolated pure proteins is one promising technology that can result in a wide variety of value-added products from byproducts. These hydrolyzed proteins are known as fish protein hydrolysates (FPH). The objective of this study was to investigate the production of functional protein ingredients and hydrolysates from acid and alkali processing of catfish muscle. For both processes, protein recovery was greater than 70%, and lipid reduction was greater than 85%. For the alkali process, lipid reduction was significantly higher (p < 0.05). Color analysis showed higher whiteness and lower yellowness in proteins from the alkali process as compared to the acid process. The alkali-aided process also led to significan tly less development of thiobarbituric acid reactive substances as compared to the aci d-aided process during refrigerated storage (4C) over 12 days. Ability to form gels was also greater in protei ns from the alkali process as compared to the acid process. The properties and quality of the isolated proteins from the alkali-aided process led to the choice of this process to recover proteins for the production of FPH. Protein and peptide SDS-PAGE analysis showed a significant breakdown of the proteins into smaller peptides. Protein hydrolysates and supernatants at different degrees of hydrolysis (5, 15, and 30%) showed antioxidant activity by measuring DPPH radical scavenging activity, metal ion chelating activity, peroxidation products, reducing power, and Trolox equivalents. Also, hydrolysates and supernatants effectively inactivated the Angiotensin I Converting Enzyme which plays an important role in hypertension xii

PAGE 13

CHAPTER 1 INTRODUCTION There is a large demand for fish and fish products in the world. Unfortunately, this high demand for fish has stressed several popular fish species. Therefore, it is important to explore the protein extraction of alternative, less desirable species or byproducts of desirable species. Pioneering the use of unpopular fish species and byproducts may reduce governmental intervention into the fishing industry. Isolating proteins from undesirable sources of fish is a conscientious effort in using the aquatic resources wisely and adds economic value to previously useless species (Hultin & Kelleher, 2000a). It is estimated that about 15% of the global catch is marked useless and is discarded. Furthermore, only a portion of the harvested aquatic animal is normally utilized. It is common to see upwards of 70% of the weight of fish being discarded or used for animal feed after processing. This leads to enormous waste that could be utilized for human consumption, provided proper economic methods exist. In the past few years, pressure by environmental groups prompted governmental bodies to raise fish waste disposal costs, thus causing fishing industries to refine their waste (Kristinsson & Rasco, 2001). Refining fish offal is costly and some fisheries have closed due to a lack of funds (Ferreira & Hultin, 1994). Recently there has been a surge of interest to utilize underutilized and unutilized sources of fish. Many processes have been proposed but most have been unsuccessful, partly due to economics and poor functionality of the final products. A very successful process was developed recently where proteins can be 1

PAGE 14

2 extracted from fish and fish byproducts using either high pH (alkali-aided isolation process) or low pH (acid-aided isolation process) (Hultin et al., 2004). In the process the proteins are separated from the lipids increasing the stability of the product. The final product has been found to have excellent functionality as a gelling agents and water-binder. There is interest in extending the applications of the isolated proteins. There are several worldwide industrial applications for protein isolates and surimi-based products. Currently, Japan consumes about 70 % of the surimi produced worldwide. In Japan, the popular surimi based products include: satsuma-age, chikuwa, kamaboko, flavored kamaboko, hanpen/naruto, and crab sticks. Imitation crab chunks, flakes, and sticks are the most popular form of surimi consumed in the United States. Korea favors a fried version of surimi, called ah-mook. France (20,000 metric tons), Italy (6,000 metric tons), and Spain (17,000 metric tons) led the consumption of finished surimi products in 1996 in the European Union. China, Russia, and South America have also recently discovered surimi-based crab sticks (Park, 2000). Proteolytic enzymes added to fish muscle break down the muscle proteins into peptides of varying molecular weight. These products are generally called fish protein hydrolysates (FPH). The hydrolysates vary based on particular enzymes used and hydrolyzing conditions, such as pH, temperature, and time. The same protein breakdown can be accomplished by chemical means (e.g. very high/low pH and temperature) but the enzyme treatments are gentler and produce fewer toxic substances. Furthermore, enzymes have a range of activity and can be chosen based on which type of hydrolysate is desired. Research has shown that some FPH have excellent food functionality, such as emulsification, foaming, water-binding, and lipid-binding (Kristinsson & Rasco, 2001).

PAGE 15

3 Recent data also suggest that FPH may have biological activity. FPH may act as scavengers of iron, heme, and potent pro-oxidants, and may have reactive groups limiting lipid oxidation, which would give them antioxidative properties (both in foods and living systems). Another recent finding is their potential for inhibiting angiotensin I converting enzyme, an enzyme related to hypertension (Kristinsson and Rasco, 2001). Intact fish proteins do not appear to have the same function that would make hydrolysate production an attractive process.

PAGE 16

CHAP TER 2 LITERATURE REVIEW Catfish Industry Overview In the late 1950s, the catfish industry originated in the southeastern United States. Between 1960 and 1991, the industrys size grew from 400 acres to 161,000 acres and 410 million pounds processed (Swan, 1992). By 1998, the channel catfish production was worth over $450 million dollars, contributing to over 46% of the national aquaculture profit in the United States. Mississippi, Louisiana, Alabama, and Arkansas dominate the industry by holding over 95% of all farm-raised catfish acreage with Mississippi generating 75% of total catfish production (Hargreaves, 2002; Swann, 1992 United States Department of Agriculture/National Agriculture Statistics Services, 2000;). There are several grounds to support the fact that channel catfish is an ideal finfish species for pond aquaculture. Breeding many fish species in captivity is often a difficult feat due to the need for complex life cycle manipulations in order to induce ovulation and spawning. The catfish species does not require such reproductive manipulations. The specifications needed for catfish reproduction is simply stocking a nesting cavity simulator-laden pond with male and female catfish. The fertilized eggs are removed and agitated and the hatchlings are transferred to a hatchery for a short period before being introduced to a fingerling pond. Caring for catfish is quite simple as well. Since catfish are omnivores, they easily accept prepared diets and adjust well to variations in formulation. The ponds are stocked with food only once a day. Managing the water quality and temperature is also required, 4

PAGE 17

5 but is not demanding because catfish have broad water quality tolerance levels. Native to the southeastern United States, catfish are not a threat to indigenous species of the ecosystem. Consumer acceptability is also high regarding catfish. Although catfish is traditionally consumed in the southeastern United States, other areas of the country are embracing the fishs mild flavor and firm texture (Hargreaves, 2002; Teichert-Coddington et al., 1997). A significant demand for seafood and seafood products has increased in the United States and beyond. Between 1980 and 1990, the per capita consumption of seafood increased by 19 percent. The aquaculture industry is expected to grow rapidly due to an increase in consumer health awareness and the limited availability of wild seafood because of over-fishing, pollution, and habitat loss (Swan, 1992). Considering all the above information, it is evident that the aquaculture fish industry, particularly farm-raised catfish, is a substantial and thriving component of the seafood trade. Therefore, exploring new and ingenious methods to utilize as much of the fish as possible is a lucrative and environmentally responsible avenue. Raw Material Inherent Complications Fresh fish from the sea has distinct quality characteristics. Shimmering skin reflects light, the belly is firm, and there are no bloodspots on the gills. The eyes are clear and are normally shaped. Red gills have a fresh, seaweed or metallic aroma unfortunately, maintaining these fresh quality characteristics for extended time periods is not possible for many reasons. Chemical reactions such as lipid oxidation, protein degradation, and various enzyme reactions cause the familiar undesirable quality changes in fish. Initially, post-harvest fish muscle is soft but eventually stiffens through rigormortis then regains softness. As time passes, post-harvest fish myofibrillar proteins

PAGE 18

6 denature, decreasing their solubility and extractability. Several factors affect the degree of post-harvest myofibrillar protein denaturation extent, such as: a drop in pH after death, high temperature exposure, and fish stress during the slaughter process. Also, as myofibrillar proteins aggregate or dissociate, the viscosity decreases due to the decrease in particle axis ratio (Suzuki, 1981). Proteolysis is a major quality problem in post-mortem fish. Protein degrading enzymes convert firm fish to supple, limp fish. In most fish species, these enzymes are located in the kidney or stomach. Rapid gutting prevents these proteolytic enzymes from leaching into the muscle. Even though most of these damaging enzymes are water soluble, it is suggested that they attach to the myofibril matrix and are not completely removed (Lanier, 2000). Since fish oil is mainly composed of polyunsaturated fats, it has high iodine values and a low melting point as compared to lard. Also, fish oil is highly susceptible to oxidation, which causes rancid odors and undesirable brown color changes. This tendency towards rapid oxidation results in major quality problems during processing and storage (Suzuki, 1981). The process of mincing fish muscle, which is a prerequisite in the surimi process, incorporates air, thus exposing the lipids to large quantities of oxygen making it more susceptible to lipid oxidation (Lanier, 2000).

PAGE 19

7 Protein Extraction Techniques Surimi The Japanese term surimi once referred to the ground fish paste formed during the manufacturing of the surimi-based product kamaboko. Surimi now describes mechanically de-boned then washed fish muscle. It is important not to confuse fish mince with surimi. Fish mince is the starting material of surimi, not surimi itself. In surimi processing, the fish mince is first washed with water to remove fat and water-soluble compounds. The remaining myofibrillar protein concentrate demonstrate enhanced functional properties, such as gel-forming ability, water holding capacity, and fat-binding (Okada, 1992). Surimi is often modified for long-term storage or further processed into other seafood products, such as imitation crab meat, by incorporating additional components such as flavoring agents, sugars, and salts. The primary fish species used to make surimi in Japan and in the United States is Alaskan pollock. However, other species such as menhaden (Brevooritia tyrannus), red hake (Urophycis chuss), Pacific whiting (Merluccius productus), and spiny dogfish (Squalus acanthias) are being used in the surimi industry (Gwenn, 1992). Surimi manufacture is a multi-step process, as shown in Figure 2-1. Fish heads are removed, guts are cleaned, and bones are removed with large amounts of water to separate the waste material from the muscle tissue. The muscle is then minced by passing the material through a perforated screen and collecting the mince. During the mincing process, tough cartilage, skin, and bones do not pass through the mesh screen, thus further removing undesirable material from the muscle. By removing blood, skin,

PAGE 20

8 membranes, and other materials, the muscle is more stable and yields a higher quality product (Park & Morrissey, 2000). Remove head, guts, and bones Mince muscle Wash with cold water Remove water to reach ~85% moisture Add cryoprotectants Freeze at -25C Figure 2-1. General surimi processing scheme. The next phase in surimi processing is the washing step. The number of washing cycles and water volume depends on many factors, such as fish species, facility type and capacity, initial fish quality, and desired final surimi quality. Generally, the fish to water ratio is between 1:5 and 1:10. The product after washing primarily consists of myofibrillar protein, with a significant decrease in amount of blood, proteins, fat (which are mostly removed during washing) as compared with the starting material. By removing or at least decreasing the amount of undesirable compounds in the fish, the surimi texture, color, flavor, and storage quality is increased. After the muscle is sufficiently washed, the fish is dewatered two or three times by centrifugation, screening,

PAGE 21

9 or pressing. Adding a 0.1-0.3% NaCl/CaCl2 mixture to the fish also facilitates water removal (Park & Morrissey, 2000). Since surimi is generally frozen after dewatering, it is important to protect the functional properties of the product during storage. By adding cryoprotectants to the raw material prior to freezing, protein denaturation and aggregation are reduced. The most common cryoprotectants used in the surimi industry are sorbitol and sucrose at ~9%, along with a 1:1 mixture of sodium tripolyphosphate, and tetrasodium pyrophosphate at ~0.3%. These compounds are uniformly distributed throughout the surimi by using a silent cutter (Park & Morrissey, 2000). There are two primary methods of freezing surimi. The first method involves forming the surimi into 10 kg blocks and freezing these blocks to reach a core temperature of -25C. The other popular freezing method is drum freezing which yields more convenient surimi chips or chunks (Park & Morrissey, 2000). Acid and Alkali-Aided Protein Isolation The acid and alkali-aided protein isolation method (Figure 2-2) is a relatively new approach used for protein extraction. Both treatments rely on either alkali (pH 10.5-11.0) or acid (pH 2-3) protein solubilization, separation of undesirable constituents by centrifugation, and finally isoelectric (pH 5.5) precipitation of proteins. The major protein in the remaining isolate is myosin and the lipid content is greatly reduced (Hultin et al. 2000; Hultin &Kelleher 2000). In this process, the fish muscle is first ground and homogenized in water to increase surface area, allowing for maximum protein solubility and decrease protein loss during centrifugation. Maintaining the homogenate temperature below 10C throughout the

PAGE 22

10 process is critical in preventing unwanted functional changes in the protein. Diluting the muscle tenfold with water, during the homogenization and high or low pH adjustment phase, allows for a large protein to water contact ratio, enabling a large final protein extraction yield. Figure 2-2. Acid and alkali-aided protein isolation method. The centrifugation steps have two purposes. The first centrifugation step separates undesirable materials such as bone, connective tissue, and lipids from the soluble proteins. The first centrifugation step is critical because it removes significant amounts of membrane lipids, which are the most susceptible to lipid oxidation. The membrane lipids are not removed in conventional surimi processing. The second centrifugation acts as a dewatering aid. The acid and alkali processes also lead to significantly higher yields than the conventional surimi process. This higher yield is attributed to part of the water soluble sarcoplasmic proteins are recovered in the acid and alkali process but not in conventional surimi processing (Hultin et al., 2005). Comparing the Processes There are several differences between the products of conventional surimi processing and the acid and alkali-aided protein isolations, as well as in the processes themselves. A disadvantage to surimi processing is the long time periods required to wash the tissue, about 20 min to extract the soluble proteins and collect the insoluble myofibrillar proteins. Since the acid and alkali-aided isolation processes use homogenation and low or high pH, protein solubility is almost instantaneous, and therefore there is no need to wash the proteins (Hultin & Kelleher 2000a; Opiacha et al. 1999). An advantage of the isolation processes is that undesirable compounds, like skin, bones, microorganisms, cholesterol, membrane lipids, and other contaminating materials

PAGE 23

11 are removed during the first centrifugation step (Hultin & Kelleher 2000a).The acid and alkali-aided protein isolation method also does not require out of the ordinary equipment and a modern surimi processing line can be easily adapted to the process (Hultin & Kelleher 2000a).Conventional surimi processing promotes sarcoplasmic protein loss during the washing steps. Multiple washings lead to myofibrillar protein solubilization and consequent loss as well (Lin & Park 1996). The loss of these proteins during conventional surimi processing is responsible for the decrease in yield. The sarcoplasmic proteins are however recovered in the acid and alkali-aided processes thus substantially increasing yield. As a testament to this, using Pacific whiting fillets as the starting material, a conventional three-washing cycle surimi processing yielded only 40% recovery, compared with 60% recovery using the acid-aided processing (Choi & Park 2002). The isoelectric precipitation step in the acid and alkali-aided process results in higher protein yields as compared to conventional surimi processing due to the protein having a zero net charge, thus aggregation and precipitation of the proteins occur. Surimi processing does not involve reducing the native pH, allowing a slightly negative charge on the protein molecules which give them more solubility and thus more proteins are leached out during washing. Removal of most lipid components (particularly the susceptible membrane lipids) in the acid and isolation procedures should lead to greater oxidation stability and decreased off-odor development as compared with conventional surimi where membrane lipids mostly remain (Hultin & Kelleher 2000a). There is also evidence that the acid and alkali-processes have a different effect on proteolysis compared to surimi. A study showed that the cathepsin B and L activity was higher in an acid-treated Pacific whiting as compared with a 3-cycle washed surimi,

PAGE 24

12 leading to poorer gel forming ability for the isolate. The differences between the cathepsin activity levels are due to cathepsin B and H removal during repeat washing and pH 5.5 (which is the pH where proteins are recovered in the acid-aided process) being the optimum pH for cathepsin L activity (An et al. 1994; Choi & Park 2002). However, cathepsin H was removed from the surimi and inactivated in the acid isolate. Therefore, this enzyme did not contribute to decreased gel-forming ability in either sample. Proteolytic problems during the acid-aided process have also been found with Spanish mackerel and mullet isolates but not for the alkali-aided process so far (Kristinsson & Demir 2003). Ingadottir (2004) also noted some proteolysis at low pH during the acid extraction of tilapia proteins but not at high pH. Proteolysis is a major problem for any muscle protein extraction process as it will lead to adverse effects on protein functionality, particularly gelation and water-binding. The main functional property of extracted fish proteins is the ability to form strong and elastic gels with high water-holding capacity. In a recent study it was found that Pacific whiting surimi from a 3-cycle washing method made stronger gels than gels from the acid-aided process, likely due to less proteolysis in the surimi process (Choi &Park, 2002). However, Hultin and Kelleher (2000b) showed that acid-aided isolates made from Atlantic cod and mackerel produces very good gels. Nimomiya and others (1990) reported that alkali-treated mackerel made stronger gels than regular surimi and acid-treated mackerel. Another study by Yongsawatdigul and Park (2001), demonstrated that rockfish protein isolates produced from the alkali-aided process had better gel forming ability as compared to the acid-aided and conventional surimi processes. Demir and Kristinsson (2003) tested several warm-water species and found the alkali-aided process

PAGE 25

13 to give superior gels over conventional surimi and the acid-aided process. Davenport and Kristinsson (2004) did also report using rheology that protein isolates from the acid-aided process have significantly lower gel forming ability than isolates from the alkali-aided process, and hypothesize that this is due to some very different effects on protein structure at acid vs. alkali pH. With respect to quality, the acid and alkali-aided processes have some advantages over surimi processing. Whiteness was lower in acid-treated Pacific whiting isolates as compared with conventional surimi (Choi & Park 2002). This lower whiteness of the acid isolates was attributed to higher b* values, indicating a more yellow appearance. Demir and Kristinsson (2003a) demonstrated that color was generally better for isolates made from several warm water species using the acid and alkali-aided processes as compared to surimi processing. It has also been found that the protein isolates have significantly less amount of aerobic microorganisms compared to surimi, and growth is reduced (Demir & Kristinsson, 2003b; Ruechel & Kristinsson, unpublished findings). Protein Solubility Solubility characteristics influence many functional properties of muscle foods, such as gelation, water-binding, emulsification, adhesive strength, and thermal properties (Asghar et al., 1985; Barbut & Findlay 1991; Gillette et al., 1977; Hultin et al., 1995; Kenny et al., 1992; Richards & Jones 1987; Samejima et al., 1985;). Solubility is defined as the remaining protein fraction suspended in a well-characterized solution after pre-determined centrifugal forces are applied over a set time period (Hultin et al.. 1995). Changes in solubility during storage has been used as a quality assessment tool for muscle foods (Stefansson & Hultin, 1994)

PAGE 26

14 Muscle proteins can be divided into three solubility groups. Sarcoplasmic proteins comprise the water soluble group. Myofibrillar proteins make up the salt-soluble group, requiring salt concentrations of greater than 0.3M, possible pH adjustments, along with magnesium and/or ATP. The third group, or the insoluble protein fraction, is mainly comprised of connective tissue proteins (Hultin et al., 1995). Quality Measurements Lipid Oxidation Lipid oxidation in fish and seafood products results in the development of off-flavors and odors, indicative of quality loss. Postmortem fish are highly susceptible to lipid oxidation, due to a high concentration of polyunsaturated fatty acids. Lipid oxidation is the result of glycerol-fatty acid esters hydrolysis, catalyzed by lipases, releasing fatty acids. The extent of lipid oxidation is dependent on the quantity of pro-oxidants such as iron, specific fatty acid composition, and enzyme activity (Foegeding et al., 1996). To determine the amount of lipid oxidation products that develop over time in the surimi, protein isolates, and minced catfish, the TBARS quantity was calculated. Demir and Kristinsson (2003c) studied the TBARS development of surimi and acid and alkali-treated catfish, mullet, Spanish mackerel, and croaker over six days. When comparing the TBARS development of each treatment for each fish species, the alkali-treated isolates had lower TBARS values than the acid treated isolates for the mullet and Spanish mackerel. Also, similar final TBARS values were reported for the surimi and alkali-treated isolates prepared from each fish species. Color The main contributing color compound in fish is myoglobin, a globular heme protein located in red muscle fibers. Hemoglobin, a tetrameric heme protein found in the

PAGE 27

15 blood, is the major heme protein in white muscle. Heme proteins can be removed during washing steps in fish processing, leading to an increase in whiteness. However, as fish deteriorate, the heme becomes less water soluble and tends to oxidize into metmyoglobin, giving a discolored product (Chaijan et al., 2005; Chen 2003). Decreasing the heme protein concentration in the surimi and isolates should cause them to appear whiter in color and increase shelf-life. In order for the protein isolates to be used in value-added products, they must be white in color, with most of the heme molecules removed. It has been shown that heme proteins are affected differently by the high and low pH values used in the acid and alkali process. Kristinsson and Hultin (2004) reported that hemoglobin is readily denatured at low pH and only partially denatured when pH is adjusted to 5.5 (the pH where proteins are recovered in the process). Because of this the heme proteins are in part co-precipitated with the muscle proteins at pH 5.5 and lead to a yellowish-brown color which develops as they oxidize on denaturation. These heme proteins can also oxidize any lipids in the protein isolate and thus cause major quality problems. The pH values of the alkali-aided process do however not lead to hemoglobin denaturation, and most of the heme proteins thus remain soluble at pH 5.5 and does not co precipitate with the muscle proteins. The result is a whiter protein isolate and less lipid oxidation in the protein isolate. Protein Gelation The main functional property of surimi and protein isolates is their ability to form strong, stable and elastic gels on heating and cooling and have the ability to bind and stabilize water in the gel matrix. The gel forming ability and gel quality has been extensively studied for surimi but less for the protein isolated. The process of gelation is rather complex and involves several different stages and mechanisms.

PAGE 28

16 Setting Prior to producing a heat induced fish protein gel, the protein paste can be conditioned to set. The normal process is to hold the protein paste at a certain temperature (species dependent) to allow for enhanced protein-protein interaction, then followed by heat induced gelation. This normally creates a stronger and more elastic gel compared to non-set gels. Connell (1960) was the first to study the setting process mechanisms by observing cod myosin gelation at various pH and ionic strength values and 0C. The results of this study showed that inactivation of ATPase caused protein denaturation coupled with side-to-side protein aggregation. Many research teams looked into this setting phenomenon more in depth and discovered that the unraveling of the -helix myosin heavy chains caused non-covalent protein-protein interactions, leading to aggregation (Gill & Conway 1989; Liu et al., 1982; Niwa, 1975; Sano et al., 1990). Covalent dipeptide linkages also play a stabilization role in low temperature setting (Lanier, 2000). Dissociation and solubilization of the myosin filaments (influenced by pH, temperature, ionic strength, and Ca2+ concentrations) were responsible for driving the myosin confirmation change (Autio et al., 1989; Liu et al., 1982; Nishimoto et al., 1987; Saeki et al., 1988;). Ultimately, the paste setting phenomena is attributed to hydrophobic interactions of unraveled adjacent myosin tails, forming the initial gel matrix, as confirmed by the Raman spectroscopy of set surimi gels (Bouraoui et al., 1997; Stone & Stanley, 1992). General Mechanisms Muscle protein gelation is a complex process involving protein unraveling and reassembling into three-dimensional arrangements. The process of thermal gelation cause

PAGE 29

17 alterations in myosin and salt-soluble myofibrillar proteins that result in changes in their rheological properties and is highly dependent on pH, temperature, and ionic strength (Egelandsdal et al., 1986; Wang et al., 1990; Xiong 1993; Xiong & Blanchard 1993). The muscle components primarily responsible for heat-induced gelation are actin and myosin. These myofibrillar proteins are readily soluble in 1-3% NaCl solutions and also when the ionic strength approaches zero (Hennigar et al., 1988; Kristinsson & Hultin, 2003b; Lanier, 2000; Lin & Park 1996; Steffansson & Hultin 1994). Muscle pH also influences myofibrillar protein solubility. As the pH is adjusted above or below the isoelectric point of the proteins, the solubility increases (Choi & Park, 2000; Ingadottir 2004; Kim et al., 2003). This increase in solubility is due to an increase in net positive or negative charge on the protein (Park & Morrissey, 2000) In surimi processing, salt is added to protein pastes (starting gel material) before cooking to increases protein solubility by assisting in protein dispersion. When salt is added to protein (with physical agitation), ionic linkages are broken due to the attraction of the Na+ ions to the carboxyl (COO-) groups of glutamic and aspartic acids while the Clions are attracted to the amino (NH2+) groups of lysine and arginine. Also, salt addition is necessary for creating elasticity in the heated gel (Niwa, 1992). During heating, the proteins begin to unfold, exposing reactive surfaces to other protein molecules. These unfolded proteins are linked together by hydrophobic interactions, forming a protein network. When the proteins are exposed to temperatures above 40C, the primary covalent bonds between proteins responsible for gel formation are disulfide (-S-S-) bonds (Lanier, 2000). A gel matrix is not composed of protein alone. Rather, it is a complex arrangement of water molecules coupled with heat-denatured exposed amino

PAGE 30

18 acid residues held together by hydrogen bonds. As the gel cools the hydrogen bonds increase in number, thus creating a stiffer gel (Howe et al., 1994). Initial muscle pH also is an important factor in protein gelation. After harvest, the pH in postmortem fish muscle drops due to the conversion of glycogen to lactic acid. This drop in pH accelerates protein denaturation, leading to undesirable changes in gelation properties. Light-flesh fish species postmortem pH generally does not drop below pH 6.2, while darker-flesh fish may decrease below pH 6.0 (Lanier, 2000). Hashimoto and Arai (1978) showed that Pacific mackerel myofibrils denatured twice as fast at pH 5.8 compared to pH 6.5. Favorable gelation depends on adjusting the degree of charge repulsion to achieve good protein dispersion and structure formation (Foegeding et al., 1996). Gel heating rate and final temperature affect the gel forming mechanisms of muscle proteins and the physical properties of the gels such as gel rigidity and breaking strength (Foegeding et al., 1986; Xiong & Brekke, 1991). Favorable gelation depends on adjusting the degree of charge repulsion to achieve good protein dispersion and structure formation (Foegeding et al., 1996). Texture is an important functional characteristic in surimi and surimi-like products. At low temperatures, fish myofibrillar proteins form highly cohesive gels. When temperatures are raised, these proteins form even stronger gels. Since surimi can be used as a gluing agent or in forming imitation seafood products, gelation properties have a great impact on consumer acceptability. The three classes of rheological tests available for texture analysis of foods by instrumentation include fundamental, empirical, and imitative. Rheology is defined as the study of material deformation and flow.

PAGE 31

19 Fundamental tests are objective and the results are measured in terms of weight, time, and distance. Unfortunately, measuring foods in rheological terms is complicated and fundamental tests can become tedious and yield complicated results. Empirical tests are quicker and more practical than fundamental test. A few examples of empirical testing include puncture, shear, and extrusion. Another type of rheological test is imitative. Imitative tests mimic the mouth, focusing on material changes due to applied forces or mechanical properties (Kim & Park, 2000). The rheology methods most often using in testing gelation properties in fish are performed using dynamic tests measuring creep and stress relaxation, categorized under the umbrella of fundamental testing. Dynamic testing allows for calculating elastic modulus and mechanical damping (associated with gel toughness) over time at various frequencies. Oscillatory sheer, represented at G (storage or elastic modulus), is calculated as a function of temperature at a constant frequency of less than 1 Hz and strain amplitude between 0.003 and 0.1 (Kim & Park, 2000), as observed in Chapter 3. Common heating and cooling rates are 1C/min and 0.5C/min (Kim & Park, 2000). Enzymatic Hydrolysis Hydrolyzing muscle is a relatively simple process if appropriate controls are in place. Specific enzymes are added to a homogenized, diluted muscle system. The slurry is mixed at a constant speed and the pH and temperature are adjusted to maintain desired values, normally close to the optimal working conditions of the enzymes. When the mixture has reached the predetermined hydrolysis endpoint, normally determined by calculating the percent degree of hydrolysis (%DH), the enzymes are inactivated by high temperature treatment or sometimes by a combination of pH and heat treatments. The

PAGE 32

20 hydrolysates are then normally collected by centrifugation or filtration, dehydrated, and then can be frozen for storage (Kristinsson & Rasco, 2001). A major limitation in making hydrolysates is the properties of the starting raw material. Usually the raw materials are seafood byproducts which contain a complex mix of components, thus yielding an impure hydrolysate. The presence of pro-oxidants such as hemoglobin or myoglobin along with lipids in the final hydrolysate can lead to major color and oxidation problems, thus producing an inedible product. The advent of the acid and alkali-aided processes previously discussed, which extracts pure proteins from byproducts, is a breakthrough for hydrolysate production. With the processes, a pure protein substrate can be economically produced and used as the raw material for hydrolysis, thus yielding pure protein hydrolysates with increased stability and presumably better qualities. Enzyme Source The enzyme source is of major importance for any protein hydrolysis process. Several studies have reported the use of enzymes naturally occurring in the substrates (e.g. pepsin from the gut or trypsin from fish pyloric ceacea) with decent results. The problem however is that the activity and level of these enzymes can vary greatly thus giving you inconsistent end products. For this reason, commercial enzymes are the choice for a well controlled hydrolysis. A number of commercial enzymes are available to hydrolyze proteins, as detailed in a review by Kristinsson (2005). One of the most recent and increasingly popular enzyme used for fish proteins is Protamex, a Bacillus protease complex produced by Novozymes (Bagsvaerd, Denmark). Protamex is an endoprotease that produces non-bitter protein hydrolysates at low degrees of hydrolysis. This is important, as bitterness development is a major problem for some enzymes during

PAGE 33

21 hydrolysis of fish proteins (Hoyle & Merritt, 1994; Lalasidis et al., 1978; Yu & Fazidah, 1994). This water-soluble microgranulate enzyme is light brown in color and is free-flowing as well as non-dusting. According to the manufacturers sheet, Protamex has a declared activity of 1.5 Anson Units per gram (AU/g). The optimum usage conditions for Protamex are pH 5.5-7.5 and 35-60C. This enzyme is inactive at temperatures greater than 90C. FPH Antioxidant Properties Most studies on FPH have centered on their use as functional food ingredients or use as nutritional supplements for growing animals (Kristinsson & Rasco, 2000).Using FPH as antioxidative agents is a relatively new application, and has received limited attention. Since many seafood species have a high lipid oxidation potential, FPH could be used to decrease oxidation. Several factors influence FPH abilities to act as antioxidants, such as amino acid composition and molecular size (Kristinsson & Rasco 2001). Another postulation by Chen and Decker (1994) states that peptides with key amino acids accept free radical electrons formed during the oxidation process. Peptide Concentration Research performed on porcine hydrolysates provides evidence that muscle hydrolysates have antioxidative effects in model systems. According to Saiga et al. (2003b), prorcine hydrolysates exhibited antioxidant activity in a linolenic acid peroxidation system. As hydrolysate concentration increased, the higher the antioxidative effect was. When using -tocopherol as a positive control radical scavenger compared to the porcine hydrolysates, the porcine hydrolysates were active over a longer time period. Also, these hydrolysates showed metal chelating abilities. Wu

PAGE 34

22 et al. (2003) also demonstrated that as mackerel hydrolysate peptide level increased, their ability to inhibit lipid peroxidation of linoleic acid increased (r2=0.98). Peptide Size The molecular characteristics of peptides have been shown to have an influence on their antioxidative properties. Work by Amarowicz and Shahidi (1996) showed that the -carotene bleaching ability of capelin FPH increased as peptide size increased. The capelin FPH was separated into four peptide fractions using gel filtration column chromatography. Peptide fractions I, II, and IV showed antioxidative activity by delaying the bleaching of -carotene. However, fraction I proved to posses the highest effectiveness. A study by Wu et al. (2003) determined that mackerel FPH peptides at higher molecular weights (1.4 KDa vs. 900 KDa and 200 KDa) were more effective inhibitors of linoleic acid autoxidation, as well as better DPPH radical scavengers, and had higher reducing power than lower molecular weight peptides. However, a study by Rajapakse et al. (2005) showed that FPH prepared from mussels showed higher superoxide radical scanvenging for lower molecular weight (>962 KDa) peptides as compared to larger peptides. These low molecular weight mussel FPH peptides performed as effective DPPH radical scavengers, metal ion chelators, and lipid peroxidation inhibitors. Since the studies by Amarowicz and Shahidi (1996) and Wu et al. resulted in opposing conclusions as compared to Rajapakse et al. (2005), assigning a standard peptide size as being the rule for optimum antioxidant activity for all assays may not determine true antioxidant potential. Therefore, it is important to investigate the

PAGE 35

23 antioxidant potential of peptides with varying lengths and concentrations using a multitude of antioxidant activity assays based on different antioxidant mechanisms. Peptide Composition and Structure Effects on Antioxidant Activity Mechanisms Amino acids, such as HIS, TYR, MET, and CYS, have proven antioxidant activity (Marcus, 1960; Marcus, 1962; Karel et al., 1966). For instance, histidine shows strong radical scavenging activity because of the decomposition of its imidazole ring (Yong & Karel, 1978). In addition to amino acids, muscle protein peptides such as carnosine (-alanyl-L-histidine) have shown antioxidant activity and its mechanisms are possibly due to a combination of radical scavenging activity (Decker et al., 1992; Kohen et al., 1988) and the quenching of the singlet oxygen species (Dahl et al., 1988) by HIS. Also, hydrophobic amino acids like LEU, VAL, and ALA have been reported to correlate with radical scavenging activity (Marcus, 1962; Rajapakse et al., 2005). Chen and others (1998) reported that peptides containing histidine at their C-termini possibly act as efficient radical scavengers towards various radicals. Hydrophobic amino acids, such as PHE and GLY, are highly soluble in lipids. Soluble amino acids are able to gain closer access to the radicals as compared to their neutral or hydrophilic counterparts (Rajapakse et al., 2005). Proper positioning of GLU, LEU, and HIS on the peptide backbone has shown to improve radical scavenging in antioxidant peptide sequences (Chen et al., 1996; Suetsuna et al., 2000). Also, histidine is implied to have strong lipid radical trapping abilities as well as hydrogen donating due to its imidazole ring (Chen et al., 1998). Evidently, aromatic amino acids are considered effective radical scavengers because they readily

PAGE 36

24 donate protons to electron-needy radicals while keeping their stability through resonance structures (Rajapakse et al., 2005). According to Rajapakse et al. (2005) Lipid peroxidation presumably occurs though peroxide arbitrated extraction of hydrogen atoms from methylene carbons in polyunsaturated fatty acids. Generally speaking, peroxidation inhibition by a peptide occurs when the peptide interferes with the propagation cycle of lipid peroxidation, slowing radical formation. Several proteins exhibit antioxidant activity against the peroxidation of lipids and fatty acids. For example, Kawashima and others (1979) observed that some of their synthesized peptides with branched-chain amino acids (VAL, LEU, and ILE) possessed antioxidative activity. Chen and Decker (1994) postulated that free radicals formed during lipid oxidation of unsaturated fatty acids were accepted by electron-taking peptides containing basic amino acids. FPH Physiological Advantage Hypertension is considered a serious medical condition due to its close association with cardiovascular disease. Food components are suggested to affect hypertension. One possibility in controlling hypertension is to incorporate antihypertensive peptides into hypertension treatments (Kim et al.., 2001). A compound known as angiotensin converting enzyme (ACE) plays a major part in the renin-angiotensin system. Renin-angiotensin-aldosterone system Renin is synthesized, stored, and released in the smooth muscle cells of the afferent arterioles. There are three important factors that stimulate the release of renin. The afferent arterioles act as high-pressure baroreceptors, releasing renin when the perfusion pressure in the kidneys decreases. Also, activation of the sympathetic nerve fibers that innervate the afferent arterioles causes renin release. Another mechanism that

PAGE 37

25 causes renin secretion occurs when NaCl delivery to the macula densa is decreased (as when vascular volume is reduced), which acts through angiotensin II (a potent vasoconstrictor) to increase blood pressure (Koeppen & Stanton, 2000). Isolated renin does not play a role in physiological function; it merely acts as a proteolytic enzyme. The substrate that renin acts upon is angiotensinogen, produced by the liver and circulates throughout the body. The product of renin-cleaved angiotensinogen is the amino acid peptide angiotensin I. Like angiotensinogen, angiotensin I has no known physiological function. However, when angiotensin I is further cleaved by the angiotensin converting enzyme (ACE) (present on the surface of vascular endothelial cells), an important product involved in many important physiological functions if formed. This product is angiotensin II, which is responsible for the following: stimulation of aldosterone secretion; arteriolar vasoconstriction, which increases blood pressure; stimulation of antidiuretic hormone and thirst; and improved NaCl absorption by the proximal tube. The conversion of angiotensin I to angiotensin II primarily occurs in the pulmonary and renal endothelial cells (Koeppen & Stanton, 2000). The secretion of aldosterone, a steroid hormone produced by the glomerulosa of the adrenal cortex, is in part stimulated by angiotensin II. Aldosterone is involved in regulating extracellular fluid (ECF) volume, by reducing NaCl excretion. NaCl excretion is reduced by stimulating the Na+ resorption. Therefore, angiotensin II is one of the most potent hormones responsible for water and NaCl uptake (Koeppen & Stanton, 2000). Angiotensin converting enzyme (ACE) inhibition in medicine Patients presenting conditions such as renal artery constriction by atherosclerotic plaque, expanded extracellular fluid volume, or elevated blood pressure are often treated with drugs that inhibit angiotensin converting enzyme. Examples of drugs that inhibit

PAGE 38

26 angiotensin converting enzyme include captopril, enalapril, quinapril, and zofenopril (Kim et al., 2001; Ondetti, 1977; Patchett et al., 1980). These drugs block the conversion of angiotensin I to angiotensin II, which in turn lowers plasma angiotensin II levels. The decrease in angiotensin II levels in the body has a fourfold consequence: reduction in water and NaCl retention; lower aldosterone secretion; diminished arterial blood pressure; and a decrease in extracellular fluid volume. Also, bradykinin (a potent vasodilator) is degraded by ACE. Therefore, a decrease in ACE contributes to an increase in bradykinin, lowering blood pressure (Koeppen & Stanton, 2000). ACE Inactivation by Hydrolysates As shown by Kim et al. (2001), low molecular weight (<1000 Da) bovine gelatin hydrolysates that were hydrolyzed by the enzymes Alcalase and Pronase E expressed high ACE inhibitory activity. Saiga et al. (2003a) also showed that chicken hydrolysates at low molecular weight (<1000 Da) showed higher ACE inhibitory activity as compared to the lower molecular weight hydrolysates. According to Jung et al. (2004), yellowfin sole frame hydrolysates at molecular weights below 5 KDa showed the greatest ACE inhibitory activity. Wu and Ding (2001) observed that soy ACE inhibitory protein acted more aggressively against ACE as the dose increased. As compared to the drug Captopril, the soy ACE inhibitory peptides did not reduce systolic blood pressure as dramatically as Captopril (Wu and Ding 2001). In the aquatic life category, Fahmi et al. (2004) have demonstrated the efficacy of collagen peptides in sea bream scales to inhibit ACE activity. According to their study, a 300mg dose of these purified peptides administered to spontaneously hypertensive rats yielded a significant decrease in blood pressure. Also, in vitro data concerning ACE

PAGE 39

27 inhibitory showed a two-fold increase in ACE IC50 value for the hydrolysate as compared to the commercial hypertension drug enalapril maleate. The ACE IC50 value represents the concentration of material needed to give a 50% reduction in ACE activity. A small the ACE IC50 correlates with high ACE inhibitory activity. Several species of FPH, such as sardine and tuna, have shown to inactivate ACE, which in turn reduces hypertension (Kim et al., 2000; Kristinsson & Rasco, 2001). According to Cheug et al. (1980), the C-terminal amino acid of peptides made the most important contribution to substrate binding at the ACE active site. The most favorable C-terminal amino acids were TPR, TYR, PHE, and PRO. Jung et al. (2003) also suggest that hydrophobic amino acids at the C-terminal tripeptide sequence of yellowfin sole hydrolysates are potent ACE inhibitors. Objectives The overall objectives were to investigate how the acid and alkali-aided processes on catfish muscle compared to a conventional surimi process and analyze the bioactive properties of catfish protein hydrolyates. The specific objectives were the following: Study the protein yield and lipid reduction of the acid and alkali process compared to a conventional surimi process Investigate the color and oxidative quality of the protein products generated from the different processes Study the physical and functional properties of the protein products generated from the different processes Analyze the alkali isolates for antioxidant activity and ACE inhibition Significance The significance of this research is twofold. Providing evidence that nontraditional fish species can be used in the surimi-making process will reduce the strain on traditional

PAGE 40

28 species as well as increase the use and economic value of fish species. Exploring new protein recovery technology, such as the acid and alkali-aided protein isolation processes, will provide new processing aids for the fish industry that may be used in the first step in manufacturing value-added seafood products from a variety of fish species.

PAGE 41

CHAPTER 3 MATERIALS AND METHODS Raw Material Acquisition and Handling Cleaned, de-boned, and skinned channel catfish fillets were obtained from a local retail establishment within one day of harvest. When the fillets arrived to the laboratory, they were cut into approximately 1 inch cubes and directly minced using a Scoville grinder (Hamilton Beach, Washington, NC) with 6mm holes. The fillets were constantly kept on ice or held in a cold room to maintain the temperature range of 0-4C throughout all experiments and preparations, but were never frozen. Protein Solubility One part minced muscle was combined with nine parts cold, deionized water and homogenized for 1 min at speed 10 using a hand-held Tissue Tearor (Biospec Products Inc., Bartlettsville, OK). A small portion of the fish homogenate was removed and used to determine the protein concentration by employing the Biuret method (Torten & Whitaker, 1964). The fish muscle homogenate was then split into two aliquots. The first aliquot was adjusted down to pH 1.5 by lowering the pH with 2N HCl. Samples were taken at 0.5 pH unit intervals for solubility measurements. The second fish homogenate aliquot was adjusted to pH 12 by increasing the pH with 2N NaOH. Samples were taken at 0.5 pH unit intervals for solubility measurements. The fish homogenates were placed in 50 mL centrifuge tubes and centrifuged for 20 min at 10,000 x g (4C). During the centrifugation, the fish muscle homogenate separated into three distinct layers. The middle layer was then collected for protein concentration determination using the Biuret 29

PAGE 42

30 method (Torten & Whitaker, 1964). The total protein content in this layer was also calculated. Percent soluble protein was calculated by multiplying the protein concentration (mg/mL) with the total volume of the middle layer (mL). Percent soluble protein was calculated by dividing the total protein content of the middle layer by the total protein content of the initial homogenized material before centrifugation. Protein Isolation Protein Isolate Preparation Protein isolates were prepared using both the acid and alkali-aided process (Figure 2-2). The pH values used in the isolation process were determined based on the solubility results, that is, the pH values that gave the highest solubility. The minced catfish was combined with cold, deionized water (1:9 dilution). The mixture was then homogenized in a Waring blender for 30 sec at 50% electrical output in a cold room (4C). The homogenate was either gradually lowered to pH 2.5 using 2N HCl (acid-aided process) or raised to pH 11 (alkali-aided process) using 2N NaOH by slow, constant stirring with a plastic spatula. The homogenate was then transferred into centrifuge bottles and centrifuged for 20 min at 10,000 x g in a Sorval RC-5B centrifuge using a GS-3 rotor (Kendro Laboratory Products, Newton, CT) with the temperature set between 0-4C. During centrifugation, the fish homogenate separated into three distinguishable layers: bottom sediment, a middle solubilized protein phase, and a top neutral lipid layer. The centrifuge bottles were then passed through a double-layer of cheesecloth (supported by a straining device) covering the top quarter of a 5 L plastic beaker, covered with ice to maintain a low temperature. The cheesecloth allowed only the soluble proteins to pass and collect into the plastic beaker. The soluble protein layer was then adjusted to pH 5.5

PAGE 43

31 by gradual addition of 2N HCl or 2N NaOH with slow but constant stirring. The aggregated fish proteins were then transferred to clean centrifuge tubes and centrifuged under the same conditions and using the same equipment as previously stated. This second (and final) centrifugation step yielded two layers, supernatant of mostly water and the protein isolate (PI) bottom layer. Several batches of raw material were subjected to this process to determine variations in the process. Surimi Processing Surimi, or washed catfish muscle, was prepared using minced catfish, cold, deionized water, and salt (Figure 3-1). The minced catfish and cold, deionized water (1:3 dilution) were added into a large plastic beaker sitting in ice (Stefansson & Hultin, 1994). The mixture was lightly stirred for 15 min with a plastic spatula then allowed to settle for an additional 15 min. The mixture was strained through a double layer of cheesecloth and then squeezed to remove excess water. This process was repeated twice more with the exception of the last water wash, which contained 0.2% NaCl in order to decrease swelling and aid in dewatering. As in the protein isolate preparation procedure, this process was repeated with several different raw material batches to determine process variation. Product Composition Proximate Analysis Proximate analyses was performed on the surimi, minced catfish (starting material), and the protein isolates from the acid and alkaline-aided processes. Protein analysis was performed using the Biuret method (Torten & Whitaker, 1964), with the modification of sodium deoxycholate (0.5 mL per 1 mL protein analyzed) was added to minimize cloudiness due to lipids. Protein recoveries were calculated using the differences in the

PAGE 44

32 protein content in the surimi or PI versus the minced catfish (starting material). Lipid extraction and quantification were determined using the chloroform: methanol extraction method described by Lee et al. (1996). Reduction of total lipids was calculated by comparing the difference in PI or surimi lipid content with those of the starting material. Moisture percentages of the three samples were determined using a Cenco moisture balance (CSC Scientific Company Inc., Fairfax, VA). Mince catfish Combine catfish mince and cold, DI water (1:3) Stir 15 min, sit 15 min Dewater with cheesecloth Repeat steps 2-4 with dewatered mince Combine twice washed catfish mince with a 0.2 % NaCl solution Stir 15 min, sit 15 min Dewater with cheesecloth Figure 3-1. The laboratory scale surimi process used for this study.

PAGE 45

33 Protein Electrophoresis: SDS-PAGE To determine the protein composition of surimi, minced catfish, and the protein isolates along with their fractions collected through the isolation process, SDS-PAGE was employed. The specific gel used was a pre-cast 4-20% linear gradient SDS-PAGE gels produced by Bio-Rad Laboratories (Hercules, CA). Each protein sample (0.33 mL) was added to Laemmli Buffer (0.67 mL) (bio-Rad Laboratories, Hercules, CA) and then adding 50 L of reducing agent -mercaptoethanol in order to achieve a 3 mg/mL protein concentration. This sample was heated at 90C for 5 min to denature the proteins. A wide range protein molecular weight standard (5 L) (Sigma Chemicals, St. Louis, MO) was added to the first gel well followed by 20 L of each sample solution to the adjacent 11 wells. The electrophoresis was run for approximately 45 min at 200V in a Mini-PROTEAN 3 Electrophoresis Cell (Bio-Rad Laboratories, Hercules, CA). The gels were then removed from the electrophoresis cell and separated from the gel plates. The gels were then placed into a 12% trichloroacetic acid protein fixing solution and agitated for 1 hr. The gels were then transferred into Sigma Brilliant Blue perchloric acid staining solution (Sigma Chemicals, St. Louis, MO) and agitated overnight, followed by destaining with deionized water for several hours. After desired destaining was achieved, the gels were scanned with an Epson Stylus CX5400 scanner and images were processed and protein bands quantified using the software Scion Image 4.0.2 (Scion Corporation, Frederick, MD). The muscle protein bands were identified from a standard curve created by using the wide range molecular SDS-PAGE standard and compared results by Hultin and Stefansson (1994).

PAGE 46

34 Product Quality Assessments Color Analysis In order to determine the color of the surimi, minced catfish, and the PI, a hand held Minolta colorimeter (Minolta, Ltd., Osaka, Japan) was used. The colorimeter was calibrated using a white standard plate. At least five Hunter L*, a*, and b* values were taken for each sample batch and then the values were averaged. The following formula was used to calculate whiteness (Fujii et al., 1973; Park, 1994): 222100100baLWhiteness Lipid Oxidation Products Measurements Immediately after production of surimi, minced catfish, and PI, small pieces of these samples were vacuum packed in bags of very low oxygen permeability (<20 cc/in*24h) and frozen at -70C in order to cease chemical reactions. In order to test the samples for secondary lipid oxidation products, these samples were thawed under cold running water then analyzed using a modified thiobarbituric reactive substances (TBARS) method by Lemon (1975). Gel Forming Ability of Surimi and Protein Isolate Pastes Pastes were prepared at 8% protein concentration by adding cold, deionized water buffered with 20 mM sodium phosphate (monobasic) to surimi or PI. The mixture was lightly homogenized with a hand-held Tissue Tearor (Biospec Products Inc., Bartlettsville, OK) at speed 5 until buffer was incorporated. After allowing the pastes to set overnight (Suzuki 1981; Lanier et al.. 1982), the pastes were tested for gel forming ability at pH 6, 6.5, 7, 7.5, and 8 in 500 mM NaCl. In addition, samples at pH 6.5 were tested at the following ionic strengths: 0, 200, 400, and 600 mM NaCl.

PAGE 47

35 The gel formation mechanisms of the surimi paste and isolates were studied in an AR2000 Advanced research rheometer (TA Instruments, New Castle, DE) operating in oscillation mode. The strain was set at 0.01 and frequency was set at 0.1 Hz. Changes in storage modulus (G`) on heating (5-80C) and cooling (80-5C) at 2C/min were monitored at 2C intervals. Storage modulus (G`) readings measured gel rigidity during and after the heating and cooling phases. Hydrolysate Preparation Protein Isolation The catfish muscle proteins were isolated as described in the U.S Patent Application Number 20040067551, which is outlined in Part I, Chapter 3 (Hultin et al., 2004). For the hydrolysate studies the protein isolate was vacuum packed and frozen at -80C until the material was needed for the hydrolysis process. Based on the results presented in Part I, the alkali-aided protein isolation process was chosen to isolate proteins for the hydrolysis experiments. This decision was primarily based on the lower lipid content, better oxidation stability, and color of the alkali isolates. Enzymatic Hydrolysis A variety of the enzymatic hydrolysis procedure by Kristinsson and Rasco (2000) was used to hydrolyze the protein isolates to 5, 15, and 30% degrees of hydrolysis (%DH). After determining the moisture content of the isolates using the Cenco moisture balance (CSC Scientific Company Inc., Fairfax, VA), deionized water was added to protein isolates to achieve 3% protein concentration. The deionized water and the protein isolates were added to a large mixing bowl where the ingredients were homogenized with

PAGE 48

36 a biological tissue homogenizer (Ultra Turrax T18, IKA, Germany) until a uniform consistency was achieved. The temperature used in this study was ambient temperature (~ 22C). Enzyme (Protamex, Novozymes A/S, Denmark) was added at 1.0% (w/w) for the 15% and 30% DH, 0.08% (w/w) for 5% DH, and 0% (w/w) for the control sample. With the aid of a pH meter, a stirrer, and a burette filled with 1 N sodium hydroxide, the pH was maintained at 7.5, as observed in Figure 10-1. The DH% was calculated at select time intervals during the reaction using the following equation: %DH = [(B*NB)/(*htot*MP)] 100 where B = mL sodium hydroxide added, NB = normality of sodium hydroxide, = average degree of dissociation of the NH groups, htot mEq/kg protein from amino acid composition of the protein isolate, MP = g protein (%N X 6.25). The degree of dissociation was calculated by: = (10pH pK)/(1 + 10pH pK) When the desired DH was reached, the mixture was then heated to 90C for 10 minutes to inactivate the enzyme. The slurry was then cooled and packed in freezer bags and frozen at -20C until further use. The hydrolysates were adjusted to 0.15% protein concentration (unless otherwise mentioned) with deionized water and homogenized on speed 5 with a hand-held Tissue Tearor (Biospec Products Inc., Bartlettsville, OK) before all analyses. Supernatants were also prepared from the hydrolysate samples by centrifuging the hydrolysates using a model 5415D Eppendorf centrifuge (Eppendorf Int., Hamburg, Germany) at maximum speed for 10 min, then adjusting to 0.15% protein concentration as previously mentioned.

PAGE 49

37 Figure 3-2. Hydrolysis reaction system. Sample Preparation The hydrolysates and supernatants were subjected to the Kjeldahl procedure to determine crude protein content, as outlined by the Official Methods of Analysis of AOAC (1995) at ABC Research Corp (Gainesville, FL). Based on the Kjeldahl protein analyses, the hydrolysates and supernatants were adjusted to 0.15% protein concentration (unless otherwise mentioned) with deionized water and homogenized on speed 5 with a hand-held Tissue Tearor (Biospec Products Inc., Bartlettsville, OK) before all analyses. Hydrolysate Composition SDS-PAGE Protein Electrophoresis To determine the protein composition of the hydrolysates and the hydrolysate supernatants, SDS-PAGE was employed. The use of SDS-PAGE provides good small peptide resolution (Anderson et al.. 1983; Smith 1998). The specific gels used were a preReaction Vessel pH Meter Burette and Stirrer and Stand

PAGE 50

38 cast Tris-HCl 4-20% linear gradient SDS-PAGE gels and pre-cast Tris-Tricine 10-20% linear gradient SDS-PAGE peptide gels produced by Bio-Rad Laboratories (Hercules, CA). Each protein sample (0.33 mL) was added to Laemmli Buffer or Tricine Buffer (0.67 mL) (Bio-Rad Laboratories, Hercules, CA) and then adding 50 L of reducing agent -mercaptoethanol in order to achieve a 3 mg/mL protein concentration. This sample was heated at 90C for 5 min to denature the proteins. A wide-range protein molecular weight standard (5 L) or a polypeptide SDS-PAGE standard (Sigma Chemicals, St. Louis, MO) was added to the first and ninth gel well followed by 20 L of each sample solution to the adjacent 8 wells. The electrophoresis was run for approximately 45 min at 200V in a Mini-PROTEAN 3 Electrophoresis Cell (Bio-Rad Laboratories, Hercules, CA). The gels were then removed from the electrophoresis cell and separated from the gel plates. The gels were then placed into a 12% trichloroacetic acid protein fixing solution and agitated for 1 hr. The gels were then transferred into Sigma Brilliant Blue perchloric acid staining solution (Sigma Chemicals, St. Louis, MO) and agitated overnight, followed by destaining with deionized water for several hours. After desired destaining was achieved, the gels were scanned with an Epson Stylus CX5400 scanner and images were processed using the software Scion Image 4.0.2 (Scion Corporation, Frederick, MD). The muscle protein bands were identified from a standard curve created by using the wide range molecular SDS-PAGE standard and compared to results by Hultin and Stefansson (1994). Amino Acid Analysis The amino acid composition of the hydrolysates and their supernatants was analyzed by ICBR/Protein Core (Gainesville, FL). One hundred L hydrolysate or

PAGE 51

39 supernatant sample was subjected to gas phase hydrolysis using 6N HCl for 24 hrs. The amino acid compositions were determined by derivitization with phenylisothiocyanate (PITC) using an Applied Biosystem 420A derivitizer (Foster City, CA) according to the manufacturers instructions. Antioxidant Activity ,-diphenyl--picrylhydrazyl (DPPH) Radical Scavenging The DPPH scavenging activity method that was used is a combination of Bersauder et al. (1998) and Saiga et al. (2003b). The benefit of performing this assay is that it measures the decomposition of the DPPH radical, which directly correlates with the hydrolysates ability to act as a free radical scavenger. The hydrolysates or deionized water (blank) were vortexed with a mixture of ethanol (99%) and 0.02% DPPH. After sitting at room temperature (~22C), the absorbance was read at 517nm using in an Agilent diode array spectrophotometer (Agilent Technologies, Palo Alto, CA). The remaining radicals were calculated by using the following equation: Residual DPPH radicals (%) = 100100xzyx where x = abs DPPH blank, y = abs control sample, and z = abs DPPH sample. Hydroperoxide System One method used to assess antioxidant activity of the hydrolysates involves measuring the development of hydroperoxides in a lipid peroxidation system (Chen et al., 1995; Osawa & Namiki, 1981; Saiga et al., 2003b; and Wu et al., 2003). A reaction mixture was formulated containing 0.1 M K-Phosphate (pH 7.0) with the following added ingredients: 40% (w/v) oleic acid; 0.5% (w/v) Triton X-100; 0.05 mM FeCl2 tetrahydrate (Sigma Chemical Co., St. Louis, MO). After homogenizing the reaction mixture at low

PAGE 52

40 speed for1 minute (on ice), 4 mL reaction mixture was combined with 100 L of the protein solution or deionized water. The combined solutions were further homogenized at low speed for 1 minute on ice. Four solutions were prepared to measure the hydroperoxide content in the reaction mixture by combining the following ingredients: a) 315 mL 75% ethanol (236 mL ethanol + 79 mL deionized water); b) 7 mL 30% ammonium thiocyanate (4.9 mL deionized water + 2.1 g NH4SCN); c) 14 mL 1 N HCl; d) 7 mL 20 mM FeCl2 tetrahydrate in 3.5% HCl (0.0278 g FeCl2 tetrahydrate + 0.245 mL HCl + 6.755 mL deionized water). After the hydroperoxide content determination solutions were prepared, the following solutions were added to empty glass tubes and vortexed: 100 L reaction mixture/protein solution or water (control); 4.5 mL 75% ethanol; 100 L 30% ammonium thiocyanate; 200 L 1 N HCl; and 100 L 20 mM FeCl2 tetrahydrate in 3.5% HCl. The contents of the glass tubes were analyzed at 500 nm in a spectrophotometer (Agilent Technologies, Palo Alto, CA), then heated at 80C for 60 min, and read again at 500 nm. The purpose of heating the mixtures was to accelerate the onset of oxidation. The relative hydroperoxides in the samples before and after heating were calculated using the following equation: Relative hydroperoxides (%) = yx100 where x = sample abs at 500 nm, y = control abs at 500 nm. Metal Ion Chelating Activity Transition metals, such as Fe3+ and Cu2+, can cause the production of reactive oxygen species that accelerate the lipid peroxidation chain reaction. Therefore, by

PAGE 53

41 exposing the hydrolysates to copper and iron, their antioxidative activity can be measured. The methods by Saiga et al. (2003b) and Cheng et al. (1982) were combined with slight modifications to assess the hydrolysates metal chelating ability. The following four solutions were prepared for this assay: 0.1% pyridine (pH 7); 4 mM CuSO4; 20 mM pyrochatechol violet (PV); and 0.045% EDTA. Each solution had a specific function in this metal ion chelating activity assay. For example, the 0.1% pyridine solution acted as the control. EDTA was chosen as the standard metal ion chelating agent, while the PV and CuSO4 formed the initial blue color complex. In the presence of a metal ion chelating agent, the Cu2+ ions dissociate from the PV and cause the color to change from blue to yellow. The hydrolysate, supernatant, or EDTA samples (1 mL) were combined with 2 mL 0.1% pyridine (pH 7), 100 L 4 mM CuSO4, 400 L deionized water, and 20 L 20 mM PV, while the control sample (representing no treatment) simply omitted the 1 mL hydrolysate, supernatant, or EDTA. The color change resulting from metal chelation was measured spectrophotometrically at 620 nm. The results were reported as absorbance at 620 nm. Oxygen Radical Absorbance Capacity (ORAC) The oxygen radical absorbance capacity (ORAC) assay used was a modification of the methods by Cao et al. (1993) and Ehlenfeldt and Prior (2001). This assay monitors the decay inhibition of fluorescein in the presence of AAPH, a peroxyl radical generator. The fluorescein decay rate is tracked by calculating the area under the decay curve over 70 min using SoftMax Pro Standard Edition software (Molecular Devices Corp., Sunnyvale, CA) while the fluorescein decay products are quantified using a standard curve of Trolox.

PAGE 54

42 Before executing the ORAC assay, four reagents were prepared. A phosphate buffer was prepared by mixing 0.75 M K2HPO4 and 0.7 M NaHPO4 at a 61.6: 38.9 (v/v) ratio. This phosphate buffer was then further diluted to 1:9 and adjusted to pH 7. Fluorescein (Sigma Chemical, St. Loius, MO) was prepared by dissolving 100 mg fluorescein in 25 ml methanol. Another reagent, 2,2-Azobis(2-amidinopropane) dihydrochloride (AAPH), was prepared by dissolving 355.5 mg AAPH in 5 mL previously prepared phosphate buffer. The final solution, 6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), was made by dissolving 5 mg Trolox into 100 mL phosphate buffer. A standard curve of Trolox was then constructed using the following concentrations: 0, 3.125, 6.25, 12.5, 25, and 50 M Trolox. Fifty L of supernatant samples, hydrolysate samples, standard, or blank were transferred into the appropriate number of wells in a 96-well microfluorometer plate (Dynex Technologies, Inc., Chantilly, VA). When the temperature reached 37C, 100 L fluorescein solution was added. An initial relative fluorescence (RF) reading was then taken using a Perkin-Elmer HTS-7000 Microplate Reader (PerkinElmer, Inc., Boston, MA) (485 nm excitation and 538 nm emission). Next, 50 L AAPH were added to each well with sample and another RF reading was taken immediately. RF readings were taken every 2 min for 70 min. Results are expressed as micromoles of Trolox Equivalents (TE) per gram of sample tested. Reducing Power Reducing power of the FPH was measured using a modified method of Oyaizu (1988) and Wu et al. (2003). Two mL protein samples or deionized water (control) were combined with 2 mL of each: 0.2M potassium phosphate buffer (pH 6.6) and 1%

PAGE 55

43 potassium ferricyanide. The solutions were then heated at 50C for 20min. Next, 2 mL 10% TCA solution was added to each incubated tube. In clean, glass test tubes, 0.4 mL 0.1% ferric chloride, 2 mL DI water, and 2 mL of the incubated protein/ TCA solution were combined. After the reaction proceeded for 10 min, the solutions absorbance was measured at 700 nm. The higher the absorbance, the greater the reducing power. Physiological Activity The method used to measure the Angiotensin I converting enzyme (ACE) inactivation activity was a combination of methods by Cheung et al. (1980), Kim et al. (2001), Saiga et al. (2003a). First, a 25 milliunit/mL ACE solution was prepared. The substrate was then made by combining 8.3 mM Hip-His-Leu, 50 mM sodium borate, and 0.5 M NaCl adjusted to pH 8.3. After preparing the substrate and ACE reagents, four variations of sample mixtures were prepared and placed into glass tubes. These sample mixtures were: 50 L ACE + 50 L deionized water; 100 L deionized water; 50 L ACE + 50 L hydrolysate or supernatant; and 50 L hydrolysate or supernatant + deionized water. The glass tubes with the combined solutions were incubated for 5 min at 37C, treated with 150 L substrate solution, then incubated again at 37C for 60 min. The reaction was stopped by adding 250 L 1 N HCL to each glass tube. An additional 1.5 mL ethyl acetate was then added into each tube. The solution was next centrifuged at 800 g for 15 min. After centrifugation, 1 mL of the supernatant was removed and transferred into clean glass tubes. The ethyl acetate was then evaporated by placing the glass tubes into a water bath (80C) for 30 min, leaving behind hippuric acid. The hippuric acid was then diluted with 3 mL deionized water and the absorbance was measured at 228 nm.

PAGE 56

44 The samples absorbance values were placed in the following formula, which calculates percent ACE inhibition: ACE Inhibition (%) = BADC100100 where A = ACE + deionized water, B = deionized water, C = ACE + hydrolysate or supernatant, and D = deionized water + hydrolysate or supernatant. Statistics Each experiment and each assay regarding the isolates and surimi was performed in at least triplicate. Reported results represent an average of each experiment and assay. Analysis of variance was used to determine significant differences between samples using Microsoft Excel (Microsoft Corp., Redmond, WA, U.S.A) The protein isolate used for the hydrolysis experiments was made from a several batches of catfish fillets. Catfish protein hydrolysates were prepared in duplicate from the protein isolate batch. Each assay with the hydrolysates was then performed in at least triplicate. Results are expressed as means SD. Tukeys test was used to determine significant differences (p< 0.05) between samples by using the Statistical Analysis System (SAS).

PAGE 57

CHAPTER 4 CATFISH HOMOGENATE SOLUBILITY AS A FUNCTION OF p H The basis for the acid and alkali-aided process is to first solubilize the muscle proteins, which then can be separated and recovered from other components in the raw material. The solubility of homogenized catfish muscle as a function of pH is demonstrated in Figure 4-1. The muscle proteins showed a typical U-shaped solubility curve, where the maximum solubility of the proteins occurred between pH 2-3 and pH 11-12, with solubility percentages ranging between 80-90%. At these pH ranges, the viscosity was low enough to allow cellular membrane removal during centrifugation (Hultin & Kelleher, 2000a). The homogenate between pH 10-12 was bright pink in color and had a strong ammonia odor, possibly due to deamidation of protein amino groups at high pH. The bright pink color indicated that the heme proteins were still native. However, the muscle protein homogenate between pH 2 and 3 was grey in color and had no distinct aroma. The gray color is an indication that heme proteins were denatured. The lowest protein solubility readings occurred around pH 5.5, yielding approximately 10% solubility. This reduction in solubility resulting in aggregation was caused by adjusting the pH values around the isoelectric point of the main muscle proteins, causing a decrease in repulsive forces. The proteins that remained soluble at pH 5.5 were sarcoplasmic proteins, and had a light pink color (more for the alkali-aided process than the acid-aided process) likely due to the presence of heme. At pH 5.5, the slurry displayed visible aggregated proteins that appeared white in color. 45

PAGE 58

46 Lowering the pH from 5.5 to 3, the solubility sharply increased from 10% to about 85%. These proteins at low pH are highly soluble due to their net positive charge causing repulsion between proteins. Increasing the pH from 5.5 to pH 10 increased the protein solubility from 10% to approximately 20%, indicating a solubility plateau. At this solubility plateau, the proteins were highly viscous and visibly white in color. Since proteins have a slightly negative charge at neutral pH, repulsive forces keep protein molecules from interacting, thus explaining an increase in solubility as pH is increased. Also, the second steep solubility increase occurred from about pH 10 (20%) up to pH 11 (~85%) where the proteins gain a high net negative charge mainly due to ionization of residues such as tyrosine, and as a result strong repulsive forces develop between the proteins. Demir and Kristinsson (2003c) reported similar solubility trends in croaker at acid, alkaline, and isoelectric point pH conditions. Solubility of Homogenized Catfish Muscle012345678910111213pH0102030405060708090100Solubility (%) Figure 4-1. The solubility of homogenized catfish muscle as a function of pH.

PAGE 59

47 The solubility pattern in the catfish homogenate shares some similarities to that of cod myofibrillar proteins, which have been extensively investigated. Stefansson and Hultin (1994) showed that at extremely low ionic strength, cod myofibrillar proteins were most soluble below pH 4.5 (~85%) and between pH 7.0 and pH 7.5 (~85%). The solubility was essentially 0% at pH 5.5. Therefore, the major solubility parallel between catfish homogenate and cod myofibrillar proteins occurs at pH 5.5. The slight solubility seen at pH 5.5 for the catfish proteins is because the whole muscle was used to make the homogenate, thus water-soluble sarcoplasmic proteins were in the system and many of these are soluble at pH 5.5. Interestingly, the catfish proteins displayed a larger solubility plateau (pH 5.5-pH 9.5) as compared with cod myofibrillar protein (pH 5.5-pH 7.0). This could have been due to a higher ionic strength in the catfish homogenates, which would have screened the negative charges and favored more protein-protein interactions rather than protein-water interactions. In the study by Stefansson and Hultin (1994), special care was taken to reduce the ionic strength as low as possible before solubility studies were conducted. This would not be practical for the acid and alkali-aided process. Ingadottir (2004) performed solubility studies using light muscle tilapia homogenate between pH 1.5-12. The solubility curve from the light muscle tilapia homogenate was somewhat similar to that of the catfish homogenate curve. Both solubility curves displayed minimum solubility at pH 5.5, showed similar solubility plateaus between pH 6-10, and gave a nearly identical sharp solubility increase when lowering the pH from 5.5 to 4. Interestingly, the acid solubilized light muscle tilapia homogenate showed maximum solubility between pH 2-4 (~90%), while in the same pH range the catfish homogenate solubility varied between 80-95%. The alkaline sections of

PAGE 60

48 the two solubility curves showed a distinct maximum solubility for light muscle tilapia homogenate (~99%) to be at pH 12, while the catfish homogenate leveled between pH 11 and 12 (~90%). The dark muscle was included in the catfish homogenate which could account for some of these differences. Choi and Park (2000) and Kim et al. (2003) also determined the solubility of Pacific whiting by subjecting the fish muscle to a wide pH range. These studies showed that the minimum solubility of the fish proteins occurred at pH 5.5, as observed in the catfish homogenate and the cod myofibrillar protein. The Pacific whiting solubility curve shared more similarities with the catfish results than with the cod myofibrillar protein study. A similar sharp solubility increase as pH is lowered from pH 5.5 to pH 3 and when the pH is increased from pH 9.5 to pH 12 is observed for both the catfish and Pacific whiting samples. Also, the wide solubility plateau between pH 5.5 and pH 9.5 is present in both studies, with the solubility increasing more in the same pH region for the Pacific whiting proteins (10%-30%) as compared with the catfish proteins (10%-20%). According to Kim et al. (2003), the large solubility plateau was caused by the various buffering capacities of the protein molecules and amino acids. The primary amino acids found in fish are glutamic acid, aspartic acid, lysine, leucine, and arginine. The pKa values of these amino acids correlate with both the pacific whiting and catfish homogenate curves.

PAGE 61

CHAPTER 5 SURIMI AND ISOLATE COMPOSITION AND QUALITY Composition and Recoveries Proximate composition of fish varies within and between species. Factors such as environmental conditions, the season, age, sexual maturity, diet, harvest source/location, genetic factors, and a multitude of other variables determine fish composition (Silva & Chamul, 1999). Therefore, it is important to analyze several different batches of catfish mince, surimi, and protein isolate batches for proximate composition. The constituents of interest are lipid content, total protein quantity, and specific protein composition. According to Silva and Chamul (1999), the edible portion of catfish contains 18.18% crude protein and 4.26% lipids. Lipid Reduction One important goal of the acid and alkali-aided protein isolation process is to recover proteins of significantly reduced lipid content from undesirable raw materials. By reducing the lipid concentration, better oxidative stability is expected. As shown in Table 5-1, the lipid reduction was greater than 85% for the two isolation processes (85.4% for acid and 88.6% for alkali process). The surimi had significantly lower lipid reduction (58.3%) as compared to the protein isolates (Table 5-1). Several other studies using a variety of fish species have also shown to decrease lipid content when using the acid or alkali-aided processes. These studies include mackerel (Hultin & Kelleher 2000a), herring (Undeland et al., 2002), mullet (Kristinsson & Demir, 2003ac), Spanish mackerel (Kristinsson & Demir, 2003ac), and croaker (Kristinsson & Demir, 2003a). Kim and 49

PAGE 62

50 others (1996) demonstrated a decrease in lipid content in surimi made from catfish frame meat. The reason the acid and alkali-treated processes remove significantly larger quantities of lipid than the surimi process is mainly due to the solubilization action by the extreme pH values causing the proteins and lipids to separate. When the mixture is centrifuged, the lipid and protein fractions separate based on density and solubility. The first centrifugation step leaves the fatty acids and storage fats floating at the surface above the solubilized proteins and the membrane lipids, settling to the bottom (Hultin, 2002). Protein Recovery Since the acid and alkali-aided processes are protein extraction and recovery processes, it was of interest to compare how much protein they were able to recover from the catfish mince compared to the surimi process. The protein recoveries in the acid (71.5%) and alkali-aided (70.3%) processes were not significantly different (p<0.05) (Table 5-1). Undeland and others (2002) compared the recoveries of herring light muscle using the acid (pH 2.7 pH 5.5) and alkali (pH 10.8 pH 5.5) isolation processes. This group found that the acid-aided process (74%) produced a higher yield than the alkali-aided process (68 %). Hultin and Kelleher (2000a) demonstrated a 94% recovery of mackerel light muscle using the acid-aided process. Acid-treated Pacific whiting fillets by Choi and Park (2002) gave 60% recovery yield. Kim and others (2003) also used the acid and alkali-aided processes with pacific whiting and the yields were 70% (pH 2 pH 5.5) and 75% (pH 11 pH 5.5), which is similar to the recoveries of the catfish proteins.

PAGE 63

51 Kristinsson and Demir (2003) obtained the following yields for the subsequent acid and alkali-treated fish samples: Spanish mackerel 73.6% (pH 2 pH 5.5) and 69.3% (pH 11 pH 5.5); croaker 78.7% (pH 2 pH 5.5) and 65.0% (pH 11 pH 5.5); mullet 81.2% (pH 2 pH 5.5) and 58.9% (pH 11 pH 5.5). It is therefore evident that there were significant species to species variations for these processes. As demonstrated by the previous results from several studies, the recovery yields for the acid and alkali-aided processes differ when using various species of fish. However, the mean recovery yield among the studies for the acid and alkali-aided processes is above 70%, which is significantly greater than the average recovery of surimi. One very likely explanation for the larger recovery in the acid and alkali-aided processes is the retention of sarcoplasmic proteins and other soluble proteins during the centrifugation steps (Choi & Park, 2002). As observed in Table 5-1, the protein recovery for the surimi was 62.3%, significantly less than in the acid and alkali-aided processes. Choi and Park (2002) used the traditional 3-washing cycle surimi process with Pacific whiting and obtained a yield of 19.2%, based on whole fish yields. Kristinsson and Demir (2003ac) recovered 57.7% yield using croaker, 54.1% for Spanish mackerel, and 59.3% with mullet surimi. Kim and others (1996) showed only a 33% recovery yield when preparing surimi from catfish frame meat. Since Hultin and Kelleher (2000) reported that the average yield from surimi production is between 55-70%, the catfish surimi (62.3%) did fall within the expected range. In conventional surimi processing, the sarcoplasmic proteins are removed during the washing steps (Lin & Park, 1996). In addition to the loss of sarcoplasmic proteins, partial solubilization of the myofibrillar proteins cause a decrease in yield.

PAGE 64

52 Table 5-1. Protein recovery and lipid reduction for the surimi and protein isolate processes. COMPONENT SURIMI ACID PI ALKALI PI Protein 62.3%a 71.5%b 70.3%b Lipid 58.3%b 85.4%a 88.6%c Means within one row having different superscript letters are significantly different (P<0.05) SDS-PAGE To determine the difference in protein composition among the surimi process and the isolates from the acid and alkali-aided processes, SDS-PAGE was employed. Standard wide range molecular weight markers were used to determine the presence, absence, or composition of several proteins throughout each process phase. The key proteins of most interest that were monitored though each processing technique include: myosin heavy chains (205 KDa); F-protein (116 KDa); -actinin and H-protein (66 KDa); desmin and vimentin (~54 KDa); actin (45 KDa). The starting material, as observed in Lane 7 of Figure 5-1a and Figure 5-1b, the proteins in the ground catfish muscle include: myosin heavy chains (205 KDa); -actinin subunit (97 KDa); desmin and vimentin (55 KDa); actin (45 KDa); paratropomyosin, -actinin subunit, and -actinin subunit (36 KDa). Comparing Lane 2 (upper lipid layer of the 1st centrifugation step) in Figure 5-1a and 5-1b, the two proteins that are both present are myosin heavy chain (205 KDa) and actin (45 KDa). However, the paratropomyosin, -actinin subunit, and -actinin subunit (36 KDa) were present only in the alkali-aided upper fat layer. Lane 3 (in both Figure 5-1a and Figure 5-2b) represents the soluble protein layer from the 1st centrifugation step of the isolation processes. The proteins most prevalent in both sets of Lane 3 include myosin heavy chain (205 KDa), actin (45 KDa), and tropomyosin, -actinin subunit, and

PAGE 65

53 actinin subunit (36 KDa). In addition to these proteins, troponin and myosin light chains (~24 KDa) were present in the acid-treated middle layer. The bottom layer resulting from the first centrifugation step of the isolation steps is represented in Lane 4 of Figure 5-1a and Figure 5-1b. The common proteins in this layer represented in Lane 4 include myosin heavy chain (205 KDa) and actin (45 KDa). Interestingly, F-protein (116 KDa) is present in only the acid-treated bottom layer. The second centrifugation step of the isolation processes acts to precipitate most of the solubilized proteins and also to serve as a de-watering aid. The only common protein found between the acid and alkali-aided process supernatant (Lane 5) was a 14.2 KDa, which may be myoglobin, a hemoglobin subunit or possibly parvalbumin. The supernatant in the acid-aided process appears to contain amorphin subunits and creatine kinase (84 KDa) and myosin and troponin subunits (24 KDa), while the supernatant from the alkali-aided process had significantly more soluble proteins, and appears to have desmin and vimentin (55 KDa), actin (45 KDa), tropomyosin, -actinin subunit, and -actinin subunit (36 KDa), troponin T (29 KDa), and myosin and troponin subunits (24 KDa). There were thus significant differences found between the two processes with respect to what protein classes were found in the three layers after the first centrifugation as well as the two phases in the second centrifugation. It is also evident that some important proteins, namely myosin and actin, were lost during the first centrifugation to the top and bottom layer. When comparing the final products (surimi and isolates) it can be seen that surimi recovered less protein types than both isolates, likely since the surimi process washed out

PAGE 66

54 some proteins the acid and alkali-aided processes recover. It is also shown that that the acid-aided process recovered more proteins than the alkali-aided process, since less protein bands were found in the supernatant. This could be attributed to more denaturation at low pH, and thus subsequently more aggregation (and precipitation on centrifugation) of the denatured proteins at pH 5.5 for the acid process vs. the alkali-aided process. (a) 1 2 3 4 5 6 7 8 14.2 KDa 20 KDa 29 KDa 24 KDa 55 KDa 45 KDa 36 KDa 66 KDa 116 KDa 97 KDa 84 KDa 205 KDa

PAGE 67

55 (b) 1 2 3 4 5 6 7 8 Figure 5-1. The protein composition of the different fractions obtained by the two versions of the (a) acid-aided and (b) alkali-aided protein solubilization/precipitation processes as assessed with SDS-PAGE. Lane1 and 8: Molecular weight standards (same as in Figure 5-1). Lanes 2-6 represent the different fractions obtained by using the full processes with two centrifugation steps. Lane 2: Upper fat layer after 1st centrifugation. Lane 3: Soluble fraction after 1st centrifugation. Lane 4: Sediment after 1st centrifugation. Lane 5: Supernatant after 2nd centrifugation (isoelectric precipitation). Lane 6: Protein isolate (sediment after 2nd centrifugation).Lane 7: Ground catfish muscle (starting raw material). Kristinsson and Hultin (2004b) found that low pH leads to more protein denaturation than high pH and thus more protein aggregation when pH was readjusted to 5.5 for hemoglobin. Hence, a similar effect may be expected for other proteins. Different proteins were found in the acid and alkali-treated protein isolates and surimi (e.g. more protein types recovered for the acid vs. alkali process and surimi process). These findings are important since this can translate to variations in functionality of the isolates.

PAGE 68

56 1 2 3 4 5 6 7 Figure 5-2. The protein composition of washed ground catfish muscle and wash water as assessed with SDS-PAGE. Lane 1: Molecular weight standards (same as in Figure 5-1). Lane 2: Wash 1 wash water. Lane 3: Wash 1 washed muscle. Lane 4: Wash 2 wash water. Lane 5: Wash 2 washed muscle. Lane 6: Wash 3 wash water. Lane 7: Wash 3 washed muscle. Oxidation Stability and Color Thiobarbituric Acid Reactive Substances (TBARS) As previously mentioned, the protein isolation process can lead to a reduction in lipid oxidation (Hultin and Kelleher, 2000). There are several ways to measure the development of lipid oxidation in foods. Thiobarbituric acid reactive substances (TBARS) are one of the best indicators of secondary oxidation products, which are responsible for off-odors in seafood and seafood products. The initial TBARS value for each sample was very low, or below 1 mol MDA/kg (Figure 5-2). The first noticeable increase in TBARS occurred between Day 2 and 6, with

PAGE 69

57 all samples reporting values above 1 mol MDA/kg. The greatest increase in TBARS during the first six days occurred in the acid-treated isolate, reaching approximately 4 mol MDA/kg. Even though these values are not statistically significant (p>0.05), between Day 6 and 9 the TBARS values dropped in both the acid and alkali-treated isolates, while increasing in the ground muscle and surimi. However, all samples increased in TBARS from Day 9 to 12. At the end of the 12 day study, the acid-treated isolate showed the largest overall increase in TBARS, ultimately reaching over 5.5 mol MDA/kg. The catfish protein isolates extracted with the alkali-aided process resulted in significantly lower developments of TBARS as compared with the ground muscle, surimi, and acid-treated isolate (Figure 5-2). The final TBARS values for the remaining samples in decreasing order were as follows: surimi (3.8 mol MDA/kg); ground muscle (2.9 mol MDA/kg); alkali-treated isolate (1.2 mol MDA/kg). Demir and others (2003a) showed a significant difference (p<0.05) in the oxidative stability with the alkali-treated catfish protein isolate being more stable than the acid-treated catfish protein isolate. Very few studies have been conducted on the development of oxidation and oxidative stability of proteins isolates from the acid and alkali-aided processes. Kristinsson and Demir (2003) studied the TBARS formation during the acid and alkali-aided extraction and also during storage of the isolates for different species. In all cases, the acid-aided process lead to significantly more TBARS formation after processing compared to the alkali-aided processing and the surimi processing. The alkali-aided processing and surimi processing demonstrated similar results. This is in agreement with the catfish results. The possible reason the acid-treated catfish isolates produced more

PAGE 70

58 TBARS than the other samples could be due to denatured heme proteins which remained suspended in the soluble protein layer from the first centrifugation step and then precipitated during the final centrifugation step of the acid-aided protein isolation procedure (see Figure 5-1). According to Kristinsson (2002ab), heme proteins subjected to low pH conditions rapidly denature and transform into highly active catalysts in lipid oxidation. The heme proteins however were likely not denatured at high pH and thus did not aggregate when refolded at pH 5.5. These heme proteins were greatly removed in the supernatant of the centrifugation step of alkali-aided protein isolation procedure. The presumably lower level of heme proteins in the alkali isolate and also the more native structure (since they did not unfold) would be expected to lead to less lipid oxidation and thus a low final TBARS value, as was seen here. The final TBARS value was lower for surimi than that of the acid-treated isolates because the washing steps in surimi processing removes heme-rich blood from the fish (Park and Morrissey, 2000). Other pro-oxidant compounds besides heme are also removed from the ground fish during surimi processing, such as sarcoplasmic proteins and some fat (Park and Morrissey, 2000). Therefore, it could be assumed that the surimi would have the lowest TBARS development over any given time period as compared with the ground material. This assumption is valid from Day 0 to 4. However, beyond Day 4 this rationale does not hold. The washing steps also remove endogenous antioxidants from the muscle, which may explain why the oxidative stability was less after Day 4 compared to the ground muscle.

PAGE 71

59 TBARS Development over 12 Days012345678024681012DaysTBARS (umol MDA/kg) Ground muscle Surimi Acid PI Alkali PI Figure 5-3. Secondary lipid oxidation products (TBARS) in minced catfish muscle, surimi, acid PI, and alkali PI over a 12 day time period at 4C. Huang et al. (1998) examined the differences in TBARS development in tilapia muscle surimi and ground tilapia muscle. The final TBARS value after 5 days for the surimi was 1.3 mol MDA/kg while the unwashed muscle had 2.4 mol MDA/kg. Hence, the TBARS development was greater in tilapia surimi as compared with the ground muscle, the same pattern found in catfish muscle. Huang et al. (1998) postulated that the difference is due to the varying susceptibilities of the lipid components to oxidation. The higher concentration of polyunsaturated-rich membrane lipids in the surimi lends the surimi more susceptible to higher oxidation rates (Huang et al., 1998). Color Color is an important quality measurement tool when comparing fish muscle and ingredients from fish muscle. Fish protein ingredients are most desirable when white in color. Whiteness or lightness of catfish can be determined by comparing L values

PAGE 72

60 given by a colorimeter. Generally, the higher the L value, the lighter the fish (Hultin & Kelleher, 2000a). Therefore, According to Table 2, the Alkali PI had the lightest color of the three materials and also had the highest calculated whiteness. The alkali-aided isolate had also less yellowness (b value) than both surimi and acid-aided isolate, which is desirable for a protein isolate. Demir and Kristinsson (2003bc) and Demir and others (2003b) also showed lower b values and higher L values for the alkali-treated catfish protein isolates as compared with the catfish surimi and acid-treated catfish protein isolates. The significantly higher level of yellowness for the acid-aided isolate over the other protein sources is likely linked to a higher level of denatured heme proteins which co-precipitated with the muscle proteins in the second centrifugation step (Demir et al., 2003b). The surimi had a significantly (p>0.05) higher redness than both isolates, which could be due to a higher retention of native heme proteins, which leads to a reddish pink color. Not only were alkali isolates found to have a more desirable color, but their color stability of the alkali-isolates was also better than the acid-aided isolates (results not shown). A comprehensive investigation by Demir and Kristinsson (2003c) reported the L, a, b, and whiteness values for surimi and acid and alkali-treated Spanish mackerel, mullet, and croaker gels. The surimi gels with the whitest score was prepared from Spanish mackerel (82.8), followed by mullet surimi (76.8) and croaker surimi (71.8). The surimi whiteness scores from these fish species are similar to that of catfish surimi (76.8). In general, the whiteness scores for the acid-treated isolate gels made from Spanish mackerel, mullet, and croaker were lower than the alkali-treated isolate gels. This whiteness trend is also present in the catfish isolate gels (Table 5-2). Also, when

PAGE 73

61 comparing the yellowness (b values) of the four fish species isolate and surimi gels, the acid-treated gels scored consistently higher than the surimi or alkali-treated gels. As observed in the catfish surimi gels, the Spanish mackerel and croaker gels had higher redness values than their acid and alkali-treated counterparts. A study by Undeland et al. (2002) compared many characteristics of herring surimi and acid and alkali-treated herring proteins, including whiteness. The whiteness values for each treatment were: water washed light muscle herring surimi (69.9), acid-treated proteins (63), and alkali-treated proteins (65.5). As observed in Table 5-2, the whiteness values for each treatment are higher than the whiteness values in the herring study. This is possibly due to a difference in the level of heme proteins in catfish vs. herring, which have a higher level of very unstable heme proteins. Alkali-treated Pacific whiting gels have been reported to have whiteness values around 61.5 (Kim et al., 2003), which were also lower than the alkali-treated catfish isolates. Cortez-Ruiz et al.(2001) reported L-values below 50 for brisling sardine surimi and acid-treated sardine gels, likely due to high levels of retained denatured heme proteins. The lower whiteness values for acid-treated fish muscle as compared to regular surimi can also be attributed to the poor light reflection from the protein-protein aggregates of the acid processed proteins (Cortez-Ruiz et al., 2001; Undeland et al., 2002). Table 5-2. Color values of surimi and protein isolates SAMPLE L A B WHITENESS Surimi 70.4 1.1a -0.9 0.2a 0.7 0.4b 70.4a Acid PI 73.8 0.4b -3.6 0.2c 5.7 0.3c 72.9b Alkali PI 75.0 0.7c -3.0 0.2c 0.2 0.4a 74.8d Means within one species having different superscripts are significantly different (P<0.05)

PAGE 74

CHAPTER 6 THERMAL GELATION OF THE EXTRACTED PROTEINS Gelation of Extracted Proteins at Different pH Values in the Presence of NaCl The main functional property of extracted fish proteins is their ability to form gels. The gel forming ability of the three different protein preparations were therefore investigated, using dynamic oscillatory rheology. The gel forming study involved following changes in storage modulus (i.e. the stiffness of the protein paste/gel) as a function of temperature. To evaluate the different protein systems gel forming ability, the storage modulus (G) after heating to 80C (Figure 6-1a) and after cooling (Figure 6-1b) was compared. The paste with the highest gel strength after heating (1600 Pa) and cooling (5500 Pa) was the alkali-treated isolate paste adjusted to pH 6.5 (Figure 6-1a and b). Interestingly, the alkali-treated isolate pastes and the surimi isolates had very similar G values upon heating for each pH value tested (Figure 6-1a). According to Dublan-Garcia and others (2003), myofibrillar proteins from giant squid (Dosidicus gigas) treated with 0.5% phosphate buffer, 3.5% NaCl, and adjusted to pH 6.0 produced the maximum gel strength on heating to 80C as compared to gels within the pH range 5.5-8. However, after the heating phase the alkali-treated isolate pastes had the highest G (gel rigidity), with surimi significantly lagging, followed by the acid-treated isolate which exhibited poor final gel rigidity (Figure 6-1b). This suggests enhanced protein-protein interaction on cooling for the alkali-treated gels compared to the other protein systems. It was shown by Kristinsson and Hultin (2003b) that muscle proteins treated to a low or high pH partially unfold and only partially refold 62

PAGE 75

63 when pH is adjusted to pH 5.5. This can lead to enhanced protein-protein interaction when the proteins are heated to form a gel, since they have exposed hydrophobic areas. From the results with the catfish proteins, it appears that the alkali-treatment led to favorable changes in protein structure with respect to gelation, while the acid process did not. The general gelation trend after cooling was stronger gels (G values above 4000 Pa) formed at pH 6 and 6.5 as compared with the pastes at pH 7, 7.5, and 8 (G values below 3000 Pa). This is interesting because muscle protein gels are normally believed to have higher gel strength as pH increases. The rheology tests were performed at protein concentrations somewhat below those found in common muscle protein products, since very high protein concentrations cannot reliably be tested in the rheometer. At higher pH values, the decrease in G is likely due to less protein-protein interactions given that proteins have more negative charges and this more repulsion compared to the lower pH values (Kristinsson & Hultin, 2003a). The higher G values at the lower pH values are most likely due to more protein-protein interactions given that repulsion is reduced. Overall, the alkali-treated isolate at pH 6.5 had the highest final G, followed by the surimi and lastly the acid-treated isolate (Figure 6-1b). Another interesting observation at pH 7.5 and 8 after cooling was the alkali (2500 Pa; 1800 Pa) and acid-treated (1900 Pa; 1500 Pa) pastes formed stronger gels than the surimi paste (1700 Pa; 1300 Pa) (Figure 6-1b). These observations are supported by findings from Kristinsson and Hultin (2003b), where the gelation of isolated cod myosin and myofibrillar proteins was tested. The results of that study were myosin proteins treated with acid (1500 Pa) or alkali (1900 Pa) pH then brought to pH 7.5 formed stronger gels than the native myosin at pH 7.5 (1250 Pa). A similar trend occurred when

PAGE 76

64 the myofibrillar proteins were treated under the same conditions. Therefore, the acid and alkali treatment of the protein isolates positively affects the ability of the myosin and myofibrillar proteins in the pastes to form firmer gels at pH 7.5 and 8 compared to surimi. As observed in Figure 6-1b, the alkali-treated and surimi pastes have higher final G values than the acid-treated pastes at pH 6, 6.5, and 7. Demir and Kristinsson (2003c) also reported that the surimi and alkali-treated protein gels at 12% protein concentration with 500 mM NaCl and adjusted to pH 7 made from catfish, Spanish mackerel, and mullet had significantly higher (p< 0.05) G values than the acid-treated isolates. Interestingly, the catfish acid isolate pastes had relatively consistent final G values of about 2000 Pa at pH 6.5, 7, and 7.5, then declined in G at pH 8, while the alkali-treated and surimi pastes showed a gradual decrease in G as pH increased from 6.5 to 8. This suggests that the three various extraction treatments have dissimilar effects on protein conformation. It is known that low and high pH treatments can unfold proteins to varying extents, and the diverse structures formed may yield different functions. Since the three processes extracted a variety of proteins in dissimilar proportions (in addition to the partial unfolding they undergo at low and high pH treatments), it is possible that these variations may contribute differently to the gel formation of a muscle protein system (Xiong, 1997). Gelation of Extracted Proteins as a Function of NaCl Concentration Just as the gelation of fish proteins are sensitive to pH; their function is also greatly affected by changes in ionic strength. To investigate the effect of ionic strength, the three fish protein systems were tested pH 6.5 (where G was the highest) at the following NaCl concentrations: 0 mM, 200 mM, 400 mM, and 600 mM (Figure 6-2)

PAGE 77

65 Storage modulus (G') after heating to 80C (a)02004006008001000120014001600180066.577.58Gel pHG' ( Pa) Surimi Base PI Acid PI Storage modulus (G') after cooling to 5C (b)010002000300040005000600066.577.58Gel pHG' (Pa) Surimi Base PI Acid PI ; Figure 6-1. Gel rigidity as measured by storage modulus (G), of 8% protein pastes at 500 mM NaCl and pH 6.5 made from surimi and acid and alkali PI. Protein pastes were heated from 5-80C (a) and then cooled from 80-5C (b) at 2C/min. Interestingly, the three different systems (grouped together as one unit Figure 6-2b) at increasing salt concentrations, displayed a uniform trend for both the heating (Figure 6-2a) and the cooling phase (Figure 6-2b). For example, in Figure 6-2a the magnitudes (heights) of the three bars at 0 mM NaCl closely corresponds to the bar magnitudes (heights) in Figure 6-2b. When each protein system is compared it can be seen that the

PAGE 78

66 G of the surimi paste increased as salt concentration increased, peaking at 400 mM NaCl. However, for the acid-treated fish paste and the alkali-treated paste, the gel strength decreased from 0 to 200 mM NaCl but regained strength at 400 mM NaCl (Figure 6-2). The strongest gel upon heating (1600 Pa) and after cooling (5000 Pa) was the alkali-treated isolate paste at 400 mM NaCl, followed closely by surimi and lastly the acid-treated isolate paste. A similar study by Ingadottir (2004) compared the final storage modulus (G) after heating (80C) and cooling (5C) of tilapia washed muscle and acid and alkali-treated tilapia muscle protein isolates with and without the addition of 2% NaCl at pH ~7. Results from this investigation show that salt concentrations of washed tilapia muscle give varying viscoelastic behavior on heating. These findings suggest that the proteins without added salt dissociate rather than form cross-linkages with the salt. However, on cooling, the G increase for both samples was similar. Overall, the alkali-treated proteins with and without salt had higher final G values than the acid-treated proteins. Another study by Kim and Park (2003) compared the gelation of surimi and alkali and acid treated Alaska pollock muscle precipitated at pH 5.5 with and without the addition of salt. This investigation demonstrated that as the NaCl concentration increased in the acid and alkali-treated protein paste, the textural properties of the gels decreased. However, an increase in NaCl concentration in the surimi gave improved texture properties, thus suggesting the isolates form gels via a different mechanism than surimi. An interesting observation regarding the gelation behavior of the acid and alkali-treated isolate pastes without the addition of salt is the higher than expected gel strengths. This unexpected gel strength could possibly be due to a shift in solubility below 200 mM

PAGE 79

67 salt. Kristinsson and Hultin (2003b) demonstrated that the acid and alkali treated cod muscle proteins displayed significantly higher solubility as compared to untreated muscle proteins below 200 mM NaCl. This higher solubility may have aided in the gel formation of the protein isolates as higher solubility would translate to better protein dispersion which is though to be an important prerequisite of good gelation (Feng and Hultin, 2001). Stefansson and Hultin (1994) demonstrated that cod muscle proteins were soluble at <0.3 mM ionic strength, if washed many times. Furthermore, Feng and Hultin (2001) and Kristinsson and Hultin (2003b) found the same conditions give excellent thermally set protein gels, provided pH is slightly alkaline to provide a good dispersion (due to electrostatic repulsion) of the proteins, and thus a regular protein gel network is achieved. The surimi was only washed three times, which is the industry standard (Stefansson and Hultin, 1994). Three washing cycles is not enough to bring the salt level sufficiently low to make a good gel, which is why the catfish surimi at low salt concentrations performed the worst of the three protein systems. For this reason, adding salt is needed for surimi-like products to solubilize the proteins to obtain good gel formation (Lanier 2000). It is interesting however that gel strength for the isolates was reduced at 200 mM NaCl while that of surimi increased. This highlights that there are some major differences with the proteins in the two systems. A major difference between the systems is that the surimi still has much of the myofibrillar structures intact at 200 mM NaCl while some of the complex proteins begin to solubilize, hence setting the stage for increased gel formation. On the other hand, the isolates contain proteins that have already been released from the myofibrillar element (during the low or high pH solubilization stage). At pH 6.5 and 200

PAGE 80

68 mM NaCl, there may be some charge neutralization which would decrease the solubility and repulsion of the proteins, resulting in decreased gel forming ability. This hypothesis is yet to be tested. Storage modulus (G') of protein pastes at pH 6.5 after heating to 80C (a)02004006008001000120014001600180066.577.58Gel pHG' ( Pa) Surimi Base PI Acid PI Stora g e modulus ( G' ) of protein pastes at pH 6.5 after cooling to 5C (b)010002000300040005000600066.577.58Gel pHG' (Pa) Surimi Base PI Acid PI ; Figure 6-2. Gel rigidity as measured by storage modulus (G), of 8% protein pastes at 0, 200, 400, and 600 mM NaCl made from surimi and acid and alkali PI at pH 6.5 upon heating from 5-80C (a) and on cooling from 80-5C (b).

PAGE 81

69 Possible Molecular Changes during Heating and Cooling The rheograms generated during the gelation experiments provide and important insight into the gelation mechanisms of the three different protein treatments. Figure 6-3 is an example of the rheograms obtained for the proteins. The first noticeable change in storage modulus (G) on heating in the surimi and alkali-treated paste was around 35C, while the acid-treated paste G increased at 43C (Figure 6-3), suggesting different molecular properties of the proteins. This initial change in G signified the beginning of protein-protein interactions, leading to aggregation (Xiong and Blanchard, 1994). Dissociation of the myofibrillar proteins has be reported to begin around 40C (Xiegler and Acton, 1984), although this is highly species and solution dependent. Such dissociations include myosin dissociation into its light and heavy chains and F-actin helix unraveling and dissociating into single chains. When proteins are heated above 40C disulfide bonds begin to form between the muscle proteins, contributing to gel formation (Lanier, 2000). During the heating phase, as observed in Figure 6-3, the greatest increase in G for each of the three samples occurred between 50-60C. At approximately 60C, hydrophobic interactions between proteins form. These hydrophobic interactions are thought to be the primary mechanism for protein gelation during heating (Lanier, 2000). In all three gelation treatments, the G leveled between 70 and 80C, which is an indication that maximal heat induced interactions were achieved at that point. Kim and others (2003) also observed that Pacific whiting treated with a 70% sorbitol solution showed that gelation on heating was completed between 75C and 80C. Similar results were reported for tilapia surimi and protein isolates (Ingadottir, 2004).

PAGE 82

70 As the temperature cooled from 80-10C, the G values for all pastes increased dramatically. This increase in gel strength is due to an increase in the number of hydrogen bonds linking proteins as temperature decreases (Lanier, 2000) and possibly interactions between myosin rods (Kristinsson and Hultin, 2003b). Hydrogen bonds also stabilize the -helix of native and denatured proteins as well as the structures that form on heating and cooling (Bouraoui et al., 1997; Lanier, 2000). The significantly larger increase in the G of the alkali isolates indicates that stronger interactions develop between the proteins. This could be due to a more partially unfolded nature of these proteins, which would allow for more contact between them. Storage Modulus (G') change During Heating and Cooling at pH 6.5, 500 mM NaCl05001000150020002500300035004000450050000102030405060708090Temperature (C)G' (Pa) Surimi Base PI Acid PI Cooling Heating Figure 6-3. Gel formation mechanism of the pastes at pH 6.5 with 500 mM NaCl.

PAGE 83

CHAPTER 7 HYDROLYSATE AND SUPERNATANT COMPOSITION Hydrolysis of Catfish Proteins After adding the enzyme preparation to the protein isolate homogenate an initial rapid hydrolysis rate was observed, followed by a more gradual and steady hydrolysis pace and eventual discontinuing of hydrolysis, as observed in Figure 7-1. During the rapid initial phase peptide bonds are rapidly broken by the enzyme, eventually slowing and finally ceasing. As hydrolysis progresses over a long period of time, product inhibition as well as total substrate exhaustion may occur (Kristinsson & Rasco, 2000). Several aquatic species show similar hydrolysis curves, such as salmon muscle mince (Kristinsson and Rasco, 2000), Atlantic cod viscera (Ampo et al., 2005), herring (Sathivel et al., 2003), and Nile tilapia myofibrillar proteins (Candido & Sgarbieri, 2003). SDS-PAGE Analysis The major trend observed in Figure 7.2 is as the degree of hydrolysis increased, the lower quantity of high molecular weight proteins decreased and a higher number of molecular weight proteins below 20 KDa were present. This trend was also observed in protein hydrolysates made from salmon (Kristinsson and Rasco, 2000), sardines (Quaglia & Orban, 1990), and Atlantic cod viscera (Aspmo et al., 2005). The highest molecular weight peptide present in all samples in Figure 7.2a was ~84 KDa. Interestingly, myosin heavy chains were only present in the 0% DH sample, indicating that it was fully hydrolyzed at very low degrees of hydrolysis. 71

PAGE 84

72 The high and low molecular weight proteins present in the hydrolysate supernatants are observed in Figure 7-2b. The 0% DH (unhydrolyzed protein) supernatant contains the same peptides as observed for the 0% DH (non-centrifuged) sample, indicating no hydrolysis. As expected, the peptide size decreased as degree of hydrolysis increased for all hydrolyzed supernatants. Degree of Hydrolysis Over Time0.005.0010.0015.0020.0025.0030.000.0020.0040.0060.0080.00Time (min)% D H 30% DH 15% DH 5% DH Figure 7-1. Effect of hydrolysis time on the degree of hydrolysis (%DH) of the catfish protein isolates (3%) using 1% (w/v) Protamex for 15% and 30% DH samples and 0.08% (w/v) Protamex for 5% DH samples. Since by definition enzymatic hydrolysis cleaves proteins into smaller units, the hydrolysates at their supernatants were also analyzed for smaller peptides by using Tris-Tricine 10-20% linear gradient SPS-PAGE gels and peptide standard markers. As observed in Figure 7-3, the hydrolysates and supernatants from the 0% DH samples (Lane 2) showed the presence of higher molecular weight proteins as compared to the

PAGE 85

73 hydrolyzed samples. Also observed in Figure 7-3, as the degree of hydrolysis increased the higher the quantity of lower molecular weight peptides were present. 205 KDa 116 KDa 97 KDa 45 KDa 20 KDa 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Figure 7-2. The protein composition of the (a) hydrolysates and (b) supernatants from the hydrolysates as assessed with pre-cast Tris-HCl 4-20% linear gradient SDS-PAGE gels. Lane1 and 9: Molecular weight standards (same as in Figure 5-1). Lanes 2-6 represent the hydrolysates or supernatants. Lanes 2 and 3: 0%. Lanes 4 and 5: 5%. Lanes 6 and 7: 15%. Lane 8: 30%.

PAGE 86

74 26.6 KDa 1.4 KDa 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Figure 7-3. The protein composition of the (a) hydrolysates and (b) supernatants from the hydrolysates as assessed with pre-cast Tris-Tricine 10-20% linear gradient SDS-PAGE gels. Lane1 and 9: Molecular weight standards (same as in Figure 5-1). Lanes 2-6 represent the hydrolysates or supernatants. Lanes 2 and 3: 0%. Lanes 4 and 5: 5%. Lanes 6 and 7: 15%. Lane 8: 30%.

PAGE 87

75 Amino Acid and Protein Composition Functional differences between hydrolysates can be largely influenced by their amino acid composition. For that reason, the amino acid composition of the hydrolysates and their supernatants were analyzed (Tables 7-1 and 7-2). Not surprisingly, the amino acid composition for all the hydrolysates was similar. This is because there was no centrifugation or filtration step used, and thus the hydrolysates are expected to be very similar. Interestingly, the hydrolysate supernatant samples were very similar to the hydrolysates, even though different peptide fractions were expected to be recovered in that fraction (mostly soluble peptides). There were some minor differences found between some of the samples, particularly for ASX and LYS, which is more likely due to analysis error or variation rather than actual differences. The amino acids that were found to be present in the highest abundance were ASX (which represents both asparagine and aspartic acid), GLX (which represents both glutamine and glutamic acid), ALA (alanine), and LEU (leucine). On the other hand, the amino acids present in the lowest quantities include SER (serine), CYS (cystine), HIS (histidine), and THR (threonine). A value was not obtained for TRP (tryptophan) since it does not survive the amino acid analysis process. The amino acid distribution of the catfish isolate hydrolysates and supernatants show a similar profile as observed in hydrolyzed porcine myofibrillar proteins (Saiag et al., 2003b) and hydrolyzed whole herring byproducts (Sathivel et al., 2003). Since the amino acid composition area % values between the hydrolysates and the supernatants are very similar, the differences in antioxidant potential and ACE inhibition (shown in the following sections) are unlikely due to amino acid composition, but rather peptide composition and the arrangement of the amino acids within the peptides.

PAGE 88

76 Table 7-1. Amino acid composition (%) and protein content of the hydrolysates at varying degrees of hydrolysis. 0% DH (control) 5% DH 15% DH 30% DH Protein Conc. (mg/mL) 2.13 1.79 0.18 2.34 0.01 2.33 0.02 ASX 9.540 8.150 1.16 9.505 0.10 9.181 0.41 GLX 13.316 9.847 3.19 11.759 0.30 11.730 0.67 SER 0.958 1.257 0.34 1.328 0.02 1.287 0.09 GLY 5.907 6.643 1.12 6.109 0.08 6.139 0.16 HIS 1.441 2.145 0.65 1.746 0.09 1.650 0.10 ARG 3.901 4.892 0.83 4.502 0.17 4.357 0.11 THR 1.748 2.513 0.89 2.625 0.19 2.261 0.04 ALA 9.310 9.147 0.44 8.991 0.04 8.915 0.22 PRO 3.469 4.972 1.19 4.202 0.03 4.313 0.07 TYR 2.494 3.731 0.70 3.093 0.16 3.199 0.16 VAL 6.817 7.129 0.01 7.079 0.07 7.217 0.12 MET 2.891 4.545 0.95 3.631 0.51 3.917 0.51 CYS 0.246 0.253 0.07 0.240 0.10 0.217 0.08 ILE 5.175 10.350 5.97 6.080 0.01 6.264 0.21 LEU 9.677 10.295 0.00 9.945 0.06 9.980 0.16 PHE 4.139 4.880 0.96 4.053 0.21 4.149 0.06 LYS 17.022 11.460 3.40 13.380 0.29 13.435 0.67 Table 7-2. Amino acid composition (%) and protein content of the hydrolysate supernatants at varying degrees of hydrolysis. 0% DH (CONTROL) 5% DH 15% DH 30% DH Protein Conc. (mg/mL) 0.23 1.23 0.01 1.78 0.08 1.93 0.10 ASX 9.956 9.721 0.07 8.383 1.66 9.761 0.22 GLX 13.598 14.338 0.09 9.666 3.33 11.872 0.18 SER 0.981 0.945 0.03 1.695 0.61 1.255 0.01 GLY 5.977 5.448 0.01 6.717 0.93 6.078 0.05 HIS 0.955 1.243 0.06 2.185 1.12 1.519 0.15 ARG 3.760 4.112 0.01 4.512 1.21 3.872 0.30 THR 1.653 1.597 0.22 3.390 1.45 2.458 0.11 ALA 10.060 9.344 0.09 9.535 0.11 9.070 0.03 PRO 3.556 3.781 0.12 5.104 1.05 4.326 0.09 TYR 1.764 2.310 0.02 3.375 1.15 2.901 0.14 VAL 6.527 6.576 0.04 6.986 0.27 7.100 0.06 MET 3.339 4.115 0.20 4.190 0.00 3.778 0.54 CYS 0.000 0.323 0.00 0.345 0.08 0.248 0.08

PAGE 89

77 Table 7-2. Continued. 0% DH (CONTROL) 5% DH 15% DH 30% DH ILE 4.868 5.390 0.04 5.499 0.65 6.085 0.11 LEU 9.593 9.842 0.11 9.417 1.17 9.946 0.08 PHE 3.930 3.282 0.02 4.542 1.22 3.994 0.07 LYS 17.411 16.122 0.13 11.523 3.60 14.038 0.20

PAGE 90

CHAPTER 8 HYDROLYSATE AND SUPERNATANT BIOACTIVE PROPERTIES DPPH Radical Scavenging As observed in Figure 8-1, the 5% DH hydrolysate sample showed the lowest residual DPPH radicals (68.8%), indicating the highest antioxidant activity with respect to radical scavenging. Since the 0% DH hydrolysate showed the highest level of residual DPPH radicals (84.4%) (p<0.0001), this sample was least effective in quenching the DPPH radicals. The 15% and 30% DH hydrolysate samples had between 75% and 85% residual DPPH radical scavenging activity, and were thus intermediate to the control and 5% DH. There was a significant difference (p<0.0001) for DPPH radical scavenging activity between the 0%, 5% and 30% DH samples. These results therefore suggest that although intact catfish proteins have a potential to scavenge radicals, partial hydrolysis of these proteins leads to more effective quenching. A study by Rajapakse and others (2005) showed peptides purified from blue mussel sauce decreased the DPPH radicals by 35.4 0.87%, which falls within the catfish protein hydrolysate data in Figures 8-1 and 8-2. Contrary to the hydrolysate data in Figure 8-1, the 0% DH supernatants (i.e. soluble intact proteins) in Figure 8-2 display the lowest residual DPPH radicals (75.6%) or most radical scavenging. The general trend in the supernatants was as the %DH increased, the residual DPPH radical activity decreased, which was contrary to what was expected. The 30% DH supernatant sample showed the lowest DPPH quenching ability (p<0.0001) of only 4.9%. There was a significant difference (p<0.0001) in DPPH radical scavenging activity between the following supernatant samples: 0% and 30% DH; 0% and 15% DH; 78

PAGE 91

79 and 5% and 30% DH. Overall, the hydrolysates performed better as a group at scavenging DPPH radicals than the supernatants. This is interesting as it was expected that smaller peptides would have more scavenging activity that intact proteins in the supernatant. However, the data does suggest that the scenario is more complex than this and intermediate size proteins/peptides may be the most active when it comes to scavenging. The current available data cannot conclude the reasons for these interesting differences. DPPH Residual Radicals-Hydrolysates 01020304050607080901000%5%15%30%Degree of HydrolysisResidua lDPPH Radical (%) a c bc b bc Figure 8-1. DPPH radical scavenging activity of the hydrolysates at varying degrees of hydrolysis at 0.15% protein concentration. Results are mean SD. Different letters indicate significant difference (p<0.0001) for treatments. Several studies have hydrolysate DPPH scavenging. A few such studies include hydrolysates prepared from porcine myofibrillar proteins (Saiga et al., 2003b), mackerel (Wu et al., 2003), casein (Suetsuno et al., 2000), egg yolk (Sakanaka & Tachibana, 2005), fish skin gelatin (Mendis et al., 2005), and shrimp hepatopancreas (Diaz et al., 2004).

PAGE 92

80 DPPH Residual Radicals-Supernatants0204060801001200%5%15%30%Degree of HydrolysisResidual DPPH Radical (%) a ab bc c Figure 8-2. DPPH radical scavenging activity of the hydrolysate supernatants at varying degrees of hydrolysis at 0.15% protein concentration. Results are mean SD. Different letters indicate significant difference (p<0.0001) for treatments. As %DH increased, there was an increase in residual DPPH radical percent. A possible explanation for the differences in radical scavenging between the hydrolysates and supernatants may possibly be due the presence of aromatic amino acids. Evidently, aromatic amino acids are considered effective radical scavengers because they readily donate protons to electron-needy radicals while keeping their stability through resonance structures (Rajapakse et al., 2005). Amino acids such as HIS, TYR, MET, and CYS have proven antioxidant activity (Karel et al., 1966; Marcus, 1960; Marcus, 1962). Since the hydrolysates and supernatants (Tables 7-1 and 7-2) all contain HIS, TYR, MET, and CYS, the radical scavenging activity of the hydrolysates and supernatants could be attributed to these amino acids in the hydrolysate peptide. Also, hydrophobic amino acids like LEU, VAL,

PAGE 93

81 and ALA, present in Tables 7-1 and 7-2 have been reported to correlate with radical scavenging activity (Marcus, 1962; Rajapakse et al., 2005). To be able to understand why the different catfish hydrolysates had so different activities, it will be necessary to isolate and characterize individual peptides in the hydrolysates. Peroxidation System As observed in Figure 8-3, the hydrolysates did not inhibit hydroperoxide formation. Rather, the hydrolysates appeared to promote hydroperoxide formation. Possibly some transition metals from hydrolyzed heme proteins may have been responsible for this increase. The hydrolysate supernatants however, showed a reduction in hydroperoxides (Figure 8-4). The 5% and 30% DH supernatants showed the greatest reduction in hydroperoxides, with 55% and 59% remaining hydroperoxides, respectively. Overall, the supernatants performed about the same in reducing the hydroperoxide content. Interestingly, the supernatant from the unhydrolyzed catfish proteins were equally as effective in reducing the level of hydroperoxides (p>0.0001). Since the hydrolysates were diluted significantly more than the supernatants to reach 0.15% protein content, it is possible that there are key compounds (peptides or other chemical compounds) in the isolates which contribute to this reduction in hydroperoxides. Studies have shown that protein hydrolysates prepared from porcine myofibrillar proteins (Saiga et al., 2003b), herring (Sathivel, 2003), and mackerel (Wu et al., 2003) reduce hydroperoxide content.

PAGE 94

82 Relative Hydroperoxides-Hydrolysates 020406080100120140Control0%5%15%30%Degree of HydrolysisRelative Value of Hydroperoxides (%) Before heating After Heating a b ab A b A a A A A b b Figure 8-3. Relative hydroperoxides remaining before and after heating the hydrolysates at varying degrees of hydrolysis in a hydroperoxide generating system (0.15% protein concentration). Results are mean SD. Different capital letters indicate significant differences (p<0.05) in hydroperoxide content values for treatments before heating. Different small letters indicate significant differences in hydroperoxide content values (p<0.05) after heating. There were no significant differences between the treatments before heating. There was a significant difference between the control samples all the samples, except for the 0% DH supernatant. Since the amino acid content of the hydrolysates and supernatants are very similar in composition, another explanation for the differences in hydroperoxide content must be explored. One reason for the differences may be traced to the molecular weight of the proteins in the samples. According to Wu and others (2003) peptides from mackerel protein hydrolysates with molecular weights of less than 1.4 KDa possessed greater antioxidant activity than the larger molecular weight proteins. As observed in Figures 7.2 and 7.3, the supernatants are mainly composed of molecular weight proteins falling below 14.4 KDa, many below 1.4 KDa. Therefore, the lower molecular weight proteins present in the supernatants may be responsible for the decrease in hydroperoxide content.

PAGE 95

83 Relatively more of the small peptides are expected in the supernatants compared to the larger peptides (which were centrifuged out), which further strengthens this theory. Relative Hydroperoxides-Supernatants 020406080100120Control0%5%15%30%Degree of HydrolysisRelative Value of Hydroperoxides (%) Before Heating After Heating A a B b B b B b B b b Figure 8-4. Relative hydroperoxides remaining before and after heating the hydrolysate supernatants at varying degrees of hydrolysis in a hydroperoxide generating system (0.15% protein concentration). Results are mean SD. Different capital letters indicate significant differences (p<0.0001) in hydroperoxide content values for treatments before heating. Different small letters indicate significant differences in hydroperoxide content values (p<0.0001) after heating. There was a significant difference between the control and the supernatants before and after heating. Supernatants initially lowered the hydroperoxide content, but raised the levels after heating. A second possible explanation regarding the differences between the hydrolysates and supernatants in hydroperoxide content may be the arrangement of GLU, LEU, and HIS on the peptide backbones. Proper positioning of these amino acids has shown to improve radical scavenging in antioxidant peptide sequences (Chen et al., 1996; Suetsuna et al., 2000). Also, histidine is implied to have strong lipid radical trapping abilities as well as hydrogen donating due to its imidazole ring (Chen et al., 1998). Hydrophobic amino acids, such as PHE and GLY, are highly soluble in lipids. Soluble amino acids are

PAGE 96

84 able to gain closer access to the radicals as compared to their neutral or hydrophilic counterparts (Rajapakse et al., 2005). Metal Ion Chelating Activity Since chelating metal ions retards lipid peroxidation, the chelating activities of the hydrolysates and supernatants were measured using pyrochatechol violet (PV) and Cu2+ (Saiga et al.. 2003b), As observed in Figure 8-5, the 15% and 30% DH hydrolysates (abs below 0.500) showed greater metal ion chelating activity as compared with the 0% and 5% DH (abs above 0.500). This was in line with what was expected, i.e. that smaller peptides are able to bind metals more effectively. However, the well known metal chelator EDTA (abs 0.022) showed the best metal ion chelating activity (p<0.0001). Other studies showed similar metal ion chelating results as compared with the catfish protein hydrolysates. For example, protein hydrolysates from porcine myofibrillar proteins (Saiga et al., 2003) also chelate metal ions, showing similar absorbance values (~0.5). As observed in Figure 8-6, the hydrolysate supernatants were even more effective than the hydrolysates in chelating Cu2+. The 15% DH supernatant (abs 0.184) had the highest metal ion chelating activity of the supernatant samples, followed by 30% DH (abs 0.229), 5% DH (abs 0.376), and 0% DH (abs 1.035) (p<0.0001). Although the amino acid composition of the hydrolysates and supernatants is similar, the chelating activities differed significantly. This difference may be caused by variations in the structure and size of the peptides in the hydrolysates (Saiga et al., 2003b) and supernatants. These results strongly suggest that smaller peptides do indeed have a better ability to chelate metals, as the supernatants contained a proportionally higher level of small peptides than the hydrolysates. Presumably the amino acid histidine, especially at the N-terminal of the

PAGE 97

85 peptide sequence, is often present in peptides with high metal ion chelating activity (Chen et al., 1998). According to Suetsuna and others (2000) amino acids with acidic or basic side chains play a role in metal ion chelation due to their respective carboxyl and amino groups. This remains to be analyzed for the catfish peptides. Metal Ion Chelating Activity-Hydrolysates 0.0000.5001.0001.5002.0002.500ControlEDTA0%5%15%30%Degree of HydrolysisAbsorbance (620nm) a b c f e d Figure 8-5. Metal ion chelating activity of the hydrolysates at varying degrees of hydrolysis at 0.15% protein concentration. EDTA was added as the standard chelating agent. The control did not contain a metal chelating agent. Results are mean SD. Different small letters indicate significant differences in absorbance values (p<0.0001). As the DH% increased, absorbance decreased. It is interesting to note that the supernatant of the unhydrolyzed proteins (0% DH sample) did not perform better than the whole protein fraction, contrary to what has been seen in the previous assays. This shows that the activities of these different fractions can be quite different depending on what activity is being measured. The central idea of the ORAC assay is decreasing the intensity of a fluorescent probe by heavily supplying it with free radicals. A change in fluorescent intensity indicates the degree of free radical damage. When an antioxidant is added to an ORAC

PAGE 98

86 system, the antioxidant acts to protect the fluorescent probe against free radical damage, thereby measuring its antioxidant capacity against the radical. The ORAC method is unlike other antioxidant capacity assays. This particular method allows for representing inhibition time and percentage of free radical damage using area under the curve calculations to combine into one unit (Huang et al., 2002). Metal Ion Chelating Activity-Supernatans0.0000.5001.0001.5002.0002.500ControlEDTA0%5%15%30%Degree of HydrolysisAbsorbance (620nm) a b f c e d Figure 8-6. Metal ion chelating activity of the hydrolysate supernatants at varying degrees of hydrolysis at 0.15% protein concentration. EDTA was added as the standard chelating agent. The control did not contain a metal chelating agent. Results are mean SD. Different small letters indicate significant differences in absorbance values (p<0.0001). Oxygen Radical Absorbance Capacity (ORAC) Since solubility of tested compounds is a prerequisite for the ORAC method, it was only possible to assay the supernatant samples. The ORAC results for the supernatants are represented in micromole Trolox equivalents per gram (Figures 8-7 and 8-8). The 30% DH supernatants demonstrated the highest ORAC value for 1.5% protein

PAGE 99

87 concentration (16.00 mol/g) and for the 0.15% protein concentration (3.55 mol/g), indicating the highest antioxidant activities (p<0.0001). Also shown in Figures 8-7 and 8-8, as the degree of hydrolysis decreases, the antioxidant activity also decreases. Interestingly, the 0% DH supernatants at 0.15% protein concentration (1.29 mol/g) showed higher antioxidant activity than the 0% DH supernatants at 1.5% protein concentration (1.00 mol/g). According to Hernandez-Ledesma and others (2005), hydrolysates prepared from bovine -lactalbumin and -lactoglobulin showed that only the TRP (4.649 mol Trolox/mol amino acid), TYR (1.547 mol Trolox/mol amino acid), MET (1.547 mol Trolox/mol amino acid), CYS (0.149 mol Trolox/mol amino acid), HIS (0.073 mol Trolox/mol amino acid), and PHE (0.003 mol Trolox/mol amino acid) containing amino acids and peptide fragments possessed antioxidant activities as measured by the ORAC method. The remaining amino acids (ARG, ASN, GLN, ASP, PRO, ALA, VAL, LYS, ILE, THR, LEU, GLU, and GLY) did not display antioxidant activity using ORAC. Reducing Power Another way to assess the potential role hydrolysates or proteins play in preventing oxidation is to assess their reducing power. In this particular assay, outlined in the Materials and Methods, an increase in absorbance indicates an increase in reducing power of the sample. Therefore, as observed in Figure 12-9, the hydrolysates with the highest reducing power were the 5% DH samples (~0.3 abs), followed by 15% DH (~0.27 abs), 30% DH (~0.19 abs), and lastly 0% DH (~0.15 abs) (p<0.0001). The supernatants of the hydrolysates did show the same declining trend with increased %DH,

PAGE 100

88 but values were lower than that of the intact hydrolysates. Interestingly, the supernatant of the unhydrolyzed proteins had increased reducing power compared to the whole protein fraction. This could possibly be explained by the dilution of the whole fraction, as mentioned before, and thus a lower level of certain critical compounds in the isolates and hydrolysates. ORACSupernatants 1.5% Protein0.002.004.006.008.0010.0012.0014.0016.0018.000%5%15%30%Degree of Hydrolysismiscromole Trolox equivalents/g b a c d Figure 8-7. The antioxidant activity (Trolox equivalence) of the hydrolysate supernatants at varying degrees of hydrolysis (1.5% protein concentration) as assessed by the ORAC method. Results are mean SD. Different small letters indicate significant differences in absorbance values (p<0.0001). Antioxidant activity increased as the %DH increased. According to a study by Kim and others (2001), a low molecular weight fraction (< 1 KDa) of bovine skin gelatin hydrolysates had higher ACE inhibitory activity than a higher molecular weight fraction. Another study involving hydrolyzed chicken breast muscle also suggests that peptides with molecular weights below 1 KDa possess ACE inhibition activity (Saiga et al., 2003a).

PAGE 101

89 ORAC-Supernatants 0.15% Protein00.511.522.533.544.50%5%15%30%Degree of Hydrolysismicromol Trolox Equivalents/g a a a b Figure 8-8. The antioxidant activity (Trolox equivalence) of the hydrolysate supernatants at varying degrees of hydrolysis at (1.5% protein concentration) as assessed by the ORAC method. Results are mean SD. Different small letters indicate significant differences in absorbance values (p<0.0001). Antioxidant activity was significantly higher in the 5%, 15%, and 30% DH samples as compared with the 0%DH samples. Reducing Power-Hydrolysates 00.050.10.150.20.250.30.35Control0%5%15%30%Degree of HydrolysisAbsorbance (700nm) a b c d e Figure 8-9. Reducing power of the hydrolysates at varying degrees of hydrolysis at 0.15% protein concentration. Deionized water was added as the control. Results are mean SD. Different small letters indicate significant differences in absorbance values (p<0.0001). All hydrolysates had higher reducing power than the control.

PAGE 102

90 Reducing Power-Supernatants00.050.10.150.20.250.30.350.40.45Control0%5%15%30%Degree of HydrolysisAbsorbance (700nm) a b c d e c Figure 8-10. Reducing power of the hydrolysate supernatants at varying degrees of hydrolysis at 0.15% protein concentration. Deionized water was added as the control. Results are mean SD. Different small letters indicate significant differences in absorbance values (p<0.0001). All supernatants had higher reducing power than the control. Yet another study by Je and others (2004) show Alaska pollack hydrolysates with molecular weight peptides below 1 KDa as having high ACE inhibition activity. Junk and others (2005) found that yellowfin sole hydrolysates with molecular weights below 5 KDa had higher ACE inhibition activity than the higher molecular weight fractions. These findings support that the hydrolyzed samples as having high ACE inhibition, but fails to explain the success of the 0% DH samples. The unhydrolyzed sample could have contained some small naturally occurring peptides which can explain in part its high inhibition ability. Several studies have also shown a link between ACE inhibition and peptides containing the amino acid proline (PRO). Such studies include hydrolysates from whey protein (Abubakar et al., 1998), bonito bowels (Matsumura et al., 1993), synthetic human -casein (Kohmura et al., 1990), and sour milk (Nakamura et al., 1995). Other amino

PAGE 103

91 acids associated with high ACE inhibition include TRP, TYR, and PHE (Cheung et al., 1980). Saiga and others (2003a) demonstrated that the amino acid sequence at the N-termini of peptide chains seem to significantly influence ACE inhibitory activity. Also, Fahmi and others (2004) noted that sea bream scale hydrolysates with the most ACE inhibition activity contained peptides with TYR in the C-terminal amino acid residue. Jung and others (2005) determined that yellowfin sole hydrolysates with molecular weights below 5 KDa as well as hydrophobic C-terminal amino acids had high ACE inhibition activity. ACE Inactivaiton-Hydrolysates0204060801001200%5%15%30%Degree of Hydrolysis% ACE Inhibitio n a a a a Figure 8-11. ACE inhibition of the hydrolysates at varying degrees of hydrolysis at 0.15% protein concentration. Different letters indicate significant difference (p<0.05) for treatments.

PAGE 104

92 a a a a ACE Inactivation-Supernatants0204060801000%5%15%30%Degree of Hydrolysis% ACE Inhibitio n a a a a Figure 8-12. ACE inhibition of the hydrolysate supernatants at varying degrees of hydrolysis at 0.15% protein concentration. Different letters indicate significant difference (p<0.05) for treatments.

PAGE 105

CHAPTER 9 CONCLUSIONS The alkali-treated isolates overall performed better than the acid-treated isolates and the conventional surimi when comparing whiteness, lipid reduction, lipid oxidation, and gel formation. The acid-treated isolates gave the highest protein recovery and a high level of lipid reduction and improved whiteness compared to surimi, yet these isolate showed the poorest lipid stability over time and weak gel forming ability. The surimi performed relatively well during the gelation studies, lipid reduction, and in TBARS development over time, while showing the lowest protein recovery and lipid reduction of the three treatments. The hydrolysates and supernatants at varying degrees of hydrolysis displayed antioxidant and ACE inhibition properties. Overall, the hydrolysates performed better as a group at scavenging DPPH radicals than the supernatants. The hydrolysates did not inhibit hydroperoxide formation. Rather, the hydrolysates appeared to promote hydroperoxide formation. The hydrolysate supernatants however, showed a reduction in hydroperoxides. The hydrolysate supernatants were even more effective than the hydrolysates in chelating Cu2+. As the degree of hydrolysis decreases, the antioxidant activity as measured by ORAC also decreases. Regarding reducing power, the hydrolysates and supernatants showed a declining trend in reducing power with increased %DH, but values were lower for the supernatants than that of the intact hydrolysates. The hydrolysates and the supernatants showed high ACE inhibition activity 93

PAGE 106

94 There are numerous studies to be performed comparing the functional properties of the acid and alkali-treated isolates, surimi, hydrolysates, and hydrolysate supernatants. Such functional properties to be explored include foaming, emulsification, and viscosity. More in-depth gelation studies, such as torsion, punch, and fold tests should also be performed. Microbial growth and identity, color change over time, and gel strength at various time intervals are also useful tests for shelf-life stability. .Additional studies need to be performed to determine the exact components of the hydrolysates and supernatants that are responsible for ACE inactivation as well as specific antioxidative action mechanisms.

PAGE 107

95 LIST OF REFERENCES Abubakar A., Saito T., Kawai Y., Itoh T. (1998). Structural analysis of new antihypertensive peptides derived from cheese whey protein by protienase K digestion. Journal of Dairy Science, 81, 3131-3138 Amarowicz R., & Shahidi F. (1996). Antioxidant activity of peptide fractions of capelin protein hydrolysates. Food Chemistry, 58, 355-359 Anderson B. L., Berry R. W., Telser A. (1983). A sodium dodecyl sulfate-polyacrylamide gel electrophoresis system that separates peptides and proteins in the molecular weight range of 2500 to 90,000. Analytical Biochemistr,y 132, 365-375 An H., Weerasinghe V., Seymour T. A., Morrissey M. T. (1994). Cathepsin degradation of Pacific whiting surimi proteins. Journal of Food Science, 59, 1013-1033 Association of Official Analytical Chemists [AOAC]. (1995). Official Methods of Analysis. 16th ed. Gaithersberg, MD: AOAC. Aspmo S. I., Horn S. J., Eijsink V. G. H. (2005). Enzymatic hydrolysis of Atlantic cod (Gadus morhua L.) viscera. Process Biochemistry 40, 1957-1966 Autio K., Kiesvaara M., Polvinen K. (1989). Heat-induced gelation of minced rainbow trout (Salmo gairdneri): Effects of pH, sodium chloride and setting. Journal of Food Science, 54, 805 Bersuder P., Hole M., Smith G. (1998). Antioxidants from a heated histidine-glucose model system. I. Investigation of the antioxidant role of histidine and isolation of antioxidants by high-performance liquid chromatography. Journal of the American Oil Chemists Society, 75, 181-187 Beuge J. A. & Aust S. D. (1978). Microsomal Lipid Peroxidation. In S. Fleischer & L. Packer (Eds). Methods in Enzymology. (pp 302) New York, NY: Academic Press. Bouraoui M., Nakai S., Li-Chan E. (1997). In situ investigation of protein structure in Pacific whiting surimi and gels using Raman spectroscopy. Food Research International 30, 65-72 Candido L. M. B. & Sgarbieri V. C. (2003). Enzymatic hydrolysis of Nile tilapia (Oreochromus niloticus) myofibrillar proteins: effects on nutritional and hydrophobic properties. Journal of Agricultural and Food Chemistry, 83, 937-944

PAGE 108

96 Cao G., Alesso H. M., Culter R. G. (1993). Oxygen-radical absorbance capacity assay for antioxidants. Free Radicals Biological Medicine, 14, 303-31 Chaijan M., Benjakul S., Visessanguan W., Faustman C. (2004). Changes of pigments and color in sardine (Sardinella gibbosa) and mackerel (Rastrelliger kanagurta) muscle during iced storage. Food Chemistry, In Press Chen H. H. (2003). Effect of cold storage on the stability of chub and horse mackerel myoglobin. Journal of Food Science, 68, 1416-1419. Chen H. M., Muramoto K., Yamauchi F., Fujimoto K., Nokihara K. (1998). Antioxidative properties of histidine-containing peptides designed from peptide fragments found in digests of a soybean protein. Journal of Agricultural and Food Chemistry, 46, 49-5 Chen H.M., Muramoto K., Yamauchi F., Nokihara K. (19960. Antioxidant activity of design peptides based on the antioxidative peptide isolated from digests of a soy protein. Journal of Agricultural and Food Chemistry, 44, 2619-2623 Chen K.M. & Decker E. M. (1994). Endogenous skeletal muscle antioxidants. Critical Reviews in Food Science and Nutrition, 34, 3423-343 Cheng K.L., Ueno K., Imamura T. (1982). Handbook of Organic Analytical Reagents. (pp 503-509) Boca Raton, FL: CRC Press. Cheung H. S., Wang H. L., Ondetti M. A., Sabo E. F., Cushman D. W. (1980). Binding of peptide substrates and inhibitors of angiotensin-converting enzyme: Importance of the COOHterminal dipeptide sequence. Journal of Biological Chemistry, 255, 401-407 Choi Y. J. & Park J. W. (2002). Acid-aided protein recovery from enzyme-rich Pacific whiting. Journal of Food Science, 67, 2962-2967 Connell J. J. (1960). Studies on the proteins of fish skeletal muscle. Biochemistry, 57, 530 Cortez-Ruiz J. A., Pacheco-Aguilar R., Garcia-Sanchez G., Lugo-Sanchez M. E. (2001). Functional characterization of protein concentrate from bristly sardine under acidic conditions. Journal of Aquatic Food Product Technology 10, 5-23 Dahl T. A., Midden W. R., Hartman P. E. (1988). Some prevalent biomolecules as defense against singlet oxygen damage. Photobiochemical Photobiology, 47, 357-362 Davenport M. & Kristinsson H. G. 2004. Effect of different acid and alkali-treatments on the molecular and functional properties of catfish muscle proteins. IFT Annual Meeting, Las Vegas, 12-16 July 2004. Abstract 49G-14.

PAGE 109

97 Decker E. A., Crum A. D., Calvert J. T. (1992). Differences in the antioxidant mechanism of carnosine in the presence of copper and iron. Journal of Agricultural and Food Chemistry, 40, 756-759 Demir N., Balaban M. O., Kristinsson H. G. (2003a).Objective quality assessment of fish protein isolates and surimi using color machine vision and measurements of lipid oxidation. IFT Annual Meeting, Chicago, 12-16 July 2003. Abstract 40-7. Demir N., Balaban M. O., Kristinsson H. G. (2003b). Quality changes in catfish protein isolates and surimi as assessed by color machine vision and lipid oxidation. IFT Annual Meeting, Chicago, 12-16 July 2003. Abstract 92D-15. Demir N. & Kristinsson H. G. (2003a). Comparison of surimi with acid and alkali produced protein isolates made from warm and temperate water fish species. IFT Annual Meeting, Chicago, 12-16 July 2003. Abstract 76A-9. Demir N. & Kristinsson H. G. (2003b). Composition, quality and physicochemical properties of catfish surimi compared to catfish protein isolates from acid and alkali-aided processing. IFT Annual Meeting, Chicago, 12-16 July 2003. Abstract 40-8. Diaz A. C., Gimenez A. V. F., Mendiara S. N., Fenucci J. L. (2004). Antioxidant activity in hepatopancreas of the shrimp (Pleoticus muelleri) by electron paramagnetic spin resonance spectrometry. Journal Agricultural and Food Chemistry, 52, 3189-319 Dublan-Garcia O., Cruz-Camarillo R., Guerrero-Legarreta I., Ponce-Alquicira E. (2003). Viscoelastic and gel strength parameters of heat-induced giant squid protein gels. 2003 IFT Annual Meeting, Chicago, 12-16 July 2003. Abstract 76F-27 Egelandsdal B., Fretheim K., Samejima K. (1996). Dynamic rheological measurements on heat-induced myosin gels: effect of ionic strength, protein concentration and addition of adenosine triphosphate or pyrophosphate. Journal Agricultural and Food Chemistry, 37, 915-926 Ehlenfeldt M. K., Proir R. L. (2001). Oxygen Radical Absorbance Capacity (ORAC) and phenolic and anthocyanin concentrations in fruit and leaf tissues of highbush blueberry. Journal Agricultural and Food Chemistry, 49, 2222-222 Elliot E. L. (1987). Microbiological Quality of Alaska Pollock Surimi. In D. E. Kramer & J. Liston (Eds.), Seafood Quality Determination (pp 269-281) Amsterdam: Elsevier Fahmi A., Morimura S., Guo H. C., Shigematsu T., Kida A., Uemura Y. (2004). Production of Angiotensin I converting enzyme inhibitory peptides from sea bream scales. Process Biochemistry, 39, 1195-2000 Feng Y. & Hultin H. O. (2001). Effect of pH on the rheological and structural properties of gels of water-washed chicken-breast muscle at physiological ionic strength. Journal of Agricultural and Food Chemistry, 49, 3927-3935

PAGE 110

98 Ferreira N. G. & Hultin H. O. (1994). Liquefying cod fish by-product under acidic conditions with a fungal enzyme. Journal of Food Processing and Preservation, 18, 87-101. Foegeding E. A., Allen C. E., Dayton W. R. (1986). Effects of heating rate on thermally formed myosin, fibrinogen, and albumin gels. Journal of Food Science, 51, 104-108 Foegeding E. A., Lanier T. C., Hultin H. O. (1996). Characteristics of Edible Muscle Tissue. In O. R Fennema (Ed.), Food Chemistry 3rd Edition. (pp 897-937). New York, NY: Marcel Dekker, Inc. Fonkwe L. G. & Singh R. K. (1996). Protein recovery from mechanically de-boned turkey residue by enzymatic hydrolysis. Process Biochemistry, 31, 605-616 Fujii Y., Watanabe K., Maruyama Y. (1973). Relation between the ATP-breakdown in ice-stored Alaska pollack meat and the quality of frozen surimi. Bulletin of the Tokai Regional Fisheries Research Laboratory, 75, 7-11 Gill T. A. & Conway J. T. (1989). Thermal aggregation of cod (Gadus morhua) muscle proteins using 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide as a zero length cross-linker. Agricultural and Biological Chemistry, 53, 2553 Gwinn S. E. (1992). Development of Surimi Technology in the United States. In T. C. Lanier & C. M. Lee (Eds.), Surimi Technology (pp 23-39). New York, NY: Marcel Dekker, Inc. Hargreaves J. A. (2002). Channel catfish farming in ponds: Lessons from a maturing industry. Reviews in Fisheries Science, 10, 499-528 Hennigar C. J, Buck E. M., Hultin H. O., Peleg M., Vareltzis K. (1988). The effect of washing and sodium chloride on mechanical properties of fish muscle gels. Journal of Food Science, 53, 963-964 Hernandez-Ledesma B., Davalos A., Bartolome B., Amigo L. (2005). Preparation of antioxidant enzymatic hydrolysates from -lactalbumin and -lactoglobulin. Identification of active peptides by HPLC/MS. Journal of Agricultural and Food Chemistry 53, 588-593 Hoyle N. & Merritt J. H.(1994). Quality of fish protein hydrolysates from herring (Clupea herrengus). Journal of Food Science, 59, 76-79 Huang C-H., Lai H-T., Weng Y-M. (1998). Suitability of hybrid tilapia (Oreochromis niloticus x Oreochromis aureus) muscle for gel formation. International Journal of Food Science and Technology, 33, 339-334

PAGE 111

99 Huang D., Ou B., Hampsch-Woodill M., Flanagan J. A., Prior R. L. (2002). High-throughput assay of oxygen radical absorbance capacity (ORAC) using a multichannel liquid handling system coupled with a microplate fluorescent reader in 96-well format. Journal of Agricultural and Food Chemistry, 50, 4437-4444 Hultin H. O. (2002). Recent Advances in Surimi Technology. In M. Fingerman & R. Nagabhushanam (Eds.), Recent Advances in Marine Biotechnology, Vol.7. (pp 241-251). Enfield, NH: Science Publishers, Inc. Hultin H. O. & Kelleher S. D. (1999). Process for isolating a protein composition from a muscle source and protein composition. U.S. Patent 6,005,073: USA Hultin H. O. & Kelleher S. D. (2000a). Surimi Processing from Dark Muscle Fish. In J. W. Park (Ed.), Surimi and Surimi Seafood (pp 59-77). New York: Marcel Dekker. Hultin H. O. & Kelleher S. D. (2000b). Acid-aided demonstration. Presented at the 8th OSU Surimi Technology School; April 11-13; Astoria, Oreg. Hultin H. O., Kelleher S. D., Feng Y., Kristinsson H. G., Richards M. P., Undeland I. A., Ke S. (2004). High efficiency alkaline protein extraction. U.S. Patent Applications No. 20040067551. Hultin H. O., Kristinsson H. G., Lanier T. C. (2004). Process for Recovery of Functional Proteins by pH Shifts. In J. W. Park (Ed), Surimi and Surimi Seafood 2nd Edition (pp.148-167). New York: Marcel Dekker. Ingadottir B. (2004). The use of acid and alkali-aided protein solubilization and precipitation methods to produce functional protein ingredients from tilapia. MS thesis. University of Florida. Ingham S. C. (1992). Microbiology of Mince, Surimi, and Value-Added Seafoods. In D. R. Ward & C. R. Hackney (Eds.), Microbiology of marine food products (pp 89-100). New York, NY: Van Nostrand Reinhold. Ingham S. C. & Potter N. N. (1987). Microbial growth in surimi and mince made from Atlantic pollock. Journal of Food Protection, 50, 312-315 Je J. Y., Park P. J., Kwon J. Y., Kim S. K. (2004). A novel angiotensin I converting enzyme inhibitory peptide from Alaska pollack (Theragra chalcogramma) frame protein hydrolysate. Journal of Agricultural and Food Chemistry, 52, 7842-7854 Jimenez-Colmenero F. & Borderias A. J. (1983). A study of the effect of frozen storage on certain functional properties of meat and fish protein. Journal of Food Technology, 18, 731

PAGE 112

100 Jung W. K., Mendis E., Je J. Y., Park P. J., Son B. W., Kim H. C., Choi Y. K., Kim S. K. (2004). Angiotensin I-converting enzyme inhibitory peptide from yellowfin sole (Limanda aspera) frame protein and its antihypertensive effect in spontaneous hypertensive rats. Journal of Food Chemistry, In Press Karel M., Tannenbaum S. R., Wallance D. H., Maloney H. (1966). Autoxidation of methyl linoleate in freeze-dried model systems. III. Effects of added amino acids. Journal of Food Science, 31, 892-896 Kawashima K., Itoh H., Miyoshi M., Chibata I. (1979). Antioxidant properties of branched-chain amino acid derivatives. Chemical Pharmacology Bulletin, 27, 1912-191 Kim F. Y. & Park J. W. (2000). Rheology and Texture Properties of Surimi Gels. In J. W. Park (Ed), Surimi and Surimi Seafood (pp.267-321). New York: Marcel Dekker. Kim J. M., Liu C. H., Eun J. B., Park J. W., Oshimi R., Hayashi K., Ott B., Aramaki T., Sekine M., Horikita Y., Fujimoto K., Aikawa T., Welch L., Long R. (1996). Surimi from fillet frames of channel catfish. Journal of Food Science, 61, 428-432. Kim S-K., Byun H-G., Park P-J., Shahidi F. (2001). Angiotensin I converting enzyme inhibitory peptides purified from bovine skin gelatin hydrolysate. Journal of Agricultural and Food Chemistry, 49, 2992-2997 Kim Y. S. & Park J. W. (2003). Gelation of special fish muscle proteins without salt. IFT Annual Meeting, Chicago, 12-16 July 2003. Abstract 76A-16 Kim Y. S., Park J. W., Choi Y. J. (2003). New approaches for the effective recovery of fish proteins and their physicochemical characteristics. Fisheries Science, 69, 1231-1239 Koeppen B. M. & Stanton B. A. (2000). Part VII Renal Systems. In R. M. Berne & M. N. Levy (Eds.), Principles of Physiology 3rd Edition. (pp 408-483). St. Louis, MO: Mosby, Inc. Kohen R., Yamamoto Y., Cundy K. C., Ames B. N. (2000). Antioxidant activity of carnosine, homocarnosine, and anserine present in muscle and brain. Processing National Acadamy of Science, USA, 85, 3175-3179 Khomura M., Nio N., Ariyoshi Y. (1990). Inhibition of angiotensin-converting enzyme by synthetic peptide fragments of human casein. Agricultural and Biological Chemistry, 54, 835-836 Kristinsson H. G. (2002a). Conformational and functional changes of hemoglobin and myosin induced by pH: Functional role in fish quality. PhD dissertation, University of Massachusetts at Amherst.

PAGE 113

101 Kristinsson H. G. (2002b).Acid-induced unfolding of flounder hemoglobin: An evidence for a molten globular state with enhanced pro-oxidative activity. Journal of Agricultural and Food Chemistry, 26, 7669-7676 Kristinsson H. G. & Demir N. (2003a). Comparison of surimi with acid and alkali produced protein isolates made from warm and temperate water fish species. IFT Annual Meeting, Chicago, 12-16 July 2003. Abstract 76A-9. Kristinsson H. G. & Demir N. (2003b). Functional Protein Isolates from Underutilized Tropical and Subtropical Fish Species and Byproducts. In Bechel, P. Ed. Advances in Seafood Byproducts (pp 227-298). Anchorage, AK: Univ. Alaska Press Kristinsson H. G. & Hultin H. O. (2003a). Changes in conformation and subunit assembly of cod myosin at low and high pH and after subsequent refolding. Journal of Agricultural and Food Chemistry, 51, 7187-7196 Kristinsson H. G. & Hultin H. O. (2003b). Effect of low and high pH treatment on the functional properties of cod muscle proteins. Journal of Agriculture and Food Chemistry, 51, 5103-5111 Kristinsson H. G & Hultin H. O. (2004). The effect of acid and alkali unfolding and subsequent refolding on the pro-oxidative activity of trout hemoglobin. Journal of Agricultural and Food Chemistry, 52, 5482-5490. Kristinsson H. G. & Rasco B. A. (2000). Fish protein hydrolysates: production, biochemical, and functional properties. Critical Reviews in Food Science and Nutrition, 40, 43-81 Kristinsson H. G. & Rasco B. A. (2001). Fish Protein Hydrolysates and their Potential Use in the Food Industry. In M. Fingerman & R. Nagabhushanam (Eds.), Recent Advances in Marine Biotechnology (pp 157-181). Enfield, NH: Science Publishers, Inc. Kudo G., Okada M., Miyauchi D. (1973). Gel-forming capacity of washed and unwashed flesh of some Pacific coast species of fish. Marine Fish Review, 32, 10-15 Laemmli U. K. (1970). Cleavage of structure proteins during the assembly of head bacteriophage T4. Nature, 277, 680-68 Lalasidis G., Bostrom S., Sjoberg L-B. (1978). Low molecular weight enzymatic fish protein hydrolysates: Chemical composition and nutritive value. Journal of Agricultural and Food Chemistry, 26, 751-756 Lanier T. C. (2000). Surimi Gelation Chemistry. In J. W. Park (Ed.), Surimi and Surimi Seafood (pp 237-259). New York, NY: Marcel Dekker, Inc.

PAGE 114

102 Lanier T. C., Lin T. S., Liu Y. M., Hamann A. (1982). Heat gelation properties of actomyosin and surimi prepared from Atlantic croaker. Journal of Food Science, 47, 1921-1924 Lee C. M., Trevino B., Chaiyawat M. (1996). A simple and rapid solvent extraction method for determining total lipids in fish tissue. Journal of the American Oil Chemists Society, 79, 487-492 Lemon, D. W. (1975). An Improved TBA Test for Rancidity. In A. D. Woyewoda, P. J. Ke, B. G. Burns (Eds.), New Series Circulator (pp 65-72). Halifax, Nova Scotia: Department of Fisheries and Oceans. Lin T. M. & Park J. W. (1996). Extraction of proteins from Pacific whiting mince at various washing conditions. Journal of Food Science, 61, 432-438 Liu Y. M., Lin T. S., Lanier T. C. (1982). Thermal denaturation and aggregation of actomyosin from Atlantic croaker. Journal of Food Science, 47, 1916 Matsumoto J. J. (1980). Chemical Deterioration of Muscle Proteins during Freezing and Storage. In J. R. Whitaker & M. Fujimaki (Eds.), Chemical Deterioration of Muscle Proteins, Advances in Chemistry Series, No. 123 (pp 95-124). Washington, DC: American Chemical Society. Matsumoto J. J. & Arai T.(1952). Studies in fish meat jellies II. A method for determining jelly strengths for fish meat jellies. Bulletin of the Japanese Society of Scientific Fisheries, 17, 377-384 Matsumura N., Fujii M., Tekeda Y., Sugita K., Shimizu T. (1993). Angiotensin Iconverting enzyme inhibitory peptides derived from bonito bowels autolysate. Biotechnology Biochemistry, 57, 695-697. Marcuse R. (1960). Antioxidative effect of amino acids. Nature, 186, 886-887 Marcuse R. (1962). The effect of some amino acids of the oxidation of linolenic acid and its methyl ester. Journal of the American Oil Chemists Society, 39, 97-103 Mendis E., Rajapakse N., Kim S. K. (2005). Antioxidant properties of a radical-scavenging peptide purified from enzymatically prepared fish skin gelatin hydrolysate. Journal of Agricultural and Food Chemistry, 53, 581-587 Nakamura Y., Yamamoto N., Sakai K., Takano T., Okubo A., Yamazaki S. (1995). Purification and characterization of angiotensin I-converting enzyme inhibitors from sour milk. Journal of Dairy Science, 78, 777-783 Nickelson R., Finne G., Hanna M. O., Vanderzant C. (1980). Minced fish flesh from nontraditional Gulf of Mexico finfish species: bacteriology. Journal of Food Science, 45, 1321-1326

PAGE 115

103 Ninomiya K., Ookawa T., Tsuchiya T., Matsumoto J. J. (1990). Concentration of fish waste soluble protein and its gelation properties. Nippon Suisan Gakkaishi, 56, 1641-1645 Nishimoto S., Hashimoto A., Seki N., Kimura I., Toyoda K., Fujita T., Arai K. (1987). Influencing factors on changes in myosin heavy chain and jelly strength of salted meat paste from Alaska pollock during setting. Nippon Susian Gakkaishi, 53, 2011 Niwa E. (1975). Role of hydrophobic bonding in gelation of fish flesh paste. Bulletin of the Japanese Society of Scientific Fisheries, 41, 377-384 Novozymes. (2001). Novozymes Product Information Sheet no. 2001-08284-01. ProtamexTM. Okada M. (1963). Elastic property of kamaboko. Bulletin of the Tokai Regional Fisheries Research Laboratory, 36, 75-76 Okada, S. (1992). History of Surimi Technology. In T. C. Lanier & C. M. Lee (Eds.), Surimi Technology (pp 3-20). New York: Marcel Dekker, Inc. Opiacha J., Hultin H. O., Huang C-H., Kelleher S. D. (1999). Improvements in Stability and Functionality of Fatty Fish Surimi. In Z. U. Haquw (Ed.), Food Funcitonality: Physicochemical and Biological (pp 127-169) Lancaster, PA: Technomic Publishing. Oyaizu, M. (1988). Antioxidative activities of browning products of glucosamine fractionated by organic solvent and thin-layer chromatography. Nippon Shokuhin Kogyo Gakkaishi, 35, 771 Park J. W. (1994). Functional protein additives in surimi gels. Journal of Food Science, 59, 525-527 Park J. W. (2000). Surimi Seafood, Products, Market, and Manufacturing. In J. W. Park (Ed.), Surimi and Surimi Seafood (pp 201-233). New York: Marcel Dekker, Inc. Park J. W. & Morrissey M. T. (2000). Manufacture of Surimi from Light Muscle Fish. In J. W. Park (Ed.), Surimi and Surimi Seafood (pp 23-54). New York: Marcel Dekker, Inc. Quaglia G. B. & Orban E. (1990). Influence of enzymatic hydrolysis on structure and emulsifying properties of sardine (Sardina pilchardus) protein hydrolysates. Journal of Food Science, 55, 1517-1573 Raccach M. & Baker R. C. (1978). Microbial properties of mechanically de-boned fish flesh. Journal of Food Science, 43, 1675-1677.

PAGE 116

104 Rajapakse N., Mendis E., Jung W. K., Je J. Y., Kim S. K. (2005). Purification of a radical scavenging peptide from fermented mussel sauce and its antioxidant properties. Food Research International, 38, 175-182 Richards M. P. & Feng Y. (2000). Choosing the proper organic solvent for the determination of lipid hydroperoxide by the ferric thiocyanate assay. Analytical Biochemistry, 278, 232-235 Robinson H. W. & Hodgen C. G. (1940). The Biuret reaction in the determination of serum protein.1. A study of the conditions necessary for the production of the stable color that bears a quantitative relationship to the protein concentration. Journal of Biological Chemistry, 135, 707-725 Saiga A., Okumura T., Makihara T., Katsuta S., Shimizu T., Yamada R., Nishimura T. (2003a). Angiotensin I-converting enzyme inhibitory peptides in a hydrolyzed chicken breast muscle extract. Journal of Food Chemistry, 51, 1741-1745 Saiga A., Tanabe S., Nishimura T. (2003b). Antioxidant activity of peptides obtained from porcine myofibrillar proteins by protease treatment. Journal of Food Chemistry, 51, 3661-3667 Sarkar B.(1987).Metal protein interactions. Progressive Food and Nutritional Science, 11, 363-400 Shantha N. C. & Decker E. A. (1994). Rapid, sensitive, iron-based spectrophotometric methods for determination of peroxide values of food lipids. Journal of the American Oil Chemists Society, 77, 421-424 Shih M. C., Yang K. T., Kuo S. J. (2002). Quality and antioxidative activity of black soybean tofu as affected by soybean cultivar. Journal of Food Science, 67, 480-484 Shimada K., Fujikawa K., Yahara K., Nakamura T. (1992). Antioxidative properties of xanthan on the antioxidation of soybean oil in cyclodextrin emulsion. Journal of Agricultural and Food Chemistry, 40, 945-948 Silva J. L & Chamul R. S. (1999). Composition of Marine and Freshwater Finfish and Shellfish and their Products. In R. Martin, E. K. Payne-Carter, G. J. Flick Jr., L. Davis (Eds.), Marine and Freshwater Products Handbook Lancaster, PA: Technomic Publishing Co. Sonu S. (1986). Surimi. National Oceanic and Atmospheric Administration Technical Memorandum NMFS-SWR-013 Smith D. M. (1998). Protein Separation and Characterization Procedures. In S. Nielsen (Ed.), Food Analysis 2nd Edition (pp 257-258) Gaithersburg, MD: Aspen Publishers.

PAGE 117

105 Stefansson G. & Hultin H. O. (1994). On the solubility of cod muscle proteins in water. Journal of Agricultural and Food Chemistry, 42, 2656-2664 Stohs S. J. & Bagchi D. (1995). Oxidative mechanisms in the toxicity of metal ions. Free Radical Biological Medicine, 18, 321-336 Stone A. P, & Stanley D. W. (1992). Mechanism of fish muscle gelation. Food Research International, 25, 381-388 Suetsuna K., Ukeda H., Ochi H. (2000). Isolation and characterization of free radical scavenging activities peptides derived from casein. Journal of Nutritional Biochemistry, 11, 128-131 Suzuki T. (1981).Fish and Krill Protein: Processing Technology (pp 1-31) Essex, England: Applied Science Publishers Ltd. Swan L. (1992). A Basic Overview of Aquaculture. Series editor: J. E. Morris. Technical Bulletin Series # 102. Iowa State University. www.ces.perdue.edu/extmedia/AS/AS-457.html Last accessed February 15, 2005. Tiechert-Coddington D. R., Popma T. J., Lovshin L. L. (1997). Attributes of Tropical Pond-Cultured Fish. In H. S. Egna and C. E. Boyd (Eds.), Dynamics of Pond Aquaculture (pp 183-198).Boca Raton, FL:CRC Press Theodore A., Kristinsson H. G. (2003). The effect of pH and salt on the gelation properties of catfish surimi and catfish protein isolates from acid and alkali-aided processing. IFT Annual Meeting, Chicago, 12-16 July 2003. Abstract 76A-10. Torten J., Whitaker J. R. (1964). Evaluation of the Biuret and dye-binding methods for protein determination in meats. Journal of Food Science, 29, 168-174 Toyoda K., Kimura I., Fujita T., Noguchi S. F., Lee C. M. (1992). The Surimi Manufacturing Process. In T. C Lanier & C. M. Lee (Eds.), Surimi Technology (pp 79-112). New York: Marcel Dekker, Inc. Undeland I., Kelleher S. D., Hultin H. O. (2002). Recovery of functional proteins from herring (Clupea haerangus) light muscle by an acid or alkaline solubilization process. Journal of Agriculture and Food Chemistry, 50, 7371-7379 United States Department of Agriculture/National Agriculture Statistics Service (USDA/NASS) Census of Aquaculture (1998). United States Department of Agriculture, Natural Statistics Service. Washington, DC. Wang S. F., Smith D. M., Steffe J. F. (1990). Effect of pH on the dynamic rheological properties of chicken breast salt-solution proteins during heat-induced gelation. Poultry Science, 69, 2220-2227

PAGE 118

106 Witte V. C., Krause G. F., Bailey M. E. (1970). New method for determining 2-thio-barbituric acid values of pork and beef during storage. Journal of Food Science, 35, 582-585 Wu J. & Ding X. (2001). Hypotensive and physiological effect of Angiotensin converting enzyme inhibitory peptides derived from soy protein on spontaneously hypertensive rats. Journal of Agricultural and Food Chemistry, 49, 501-506 Wu H., Chen H., Shiau C. (2003). Free amino acids and peptides as related to antioxidant properties in protein hydrolysates of mackerel (Scomber austriasicus). Food Research International, 36, 949-957 Xiong Y. L. (1993). A comparison of the rheological characteristics of different fractions of chicken myofibrillar proteins. Journal of Food Biochemistry, 16, 217-227 Xiong Y. L. (1997). Structure-function Relationship of Muscle Proteins. In S. Damodaran & A. Paraf (Eds.), Food Proteins and Their Applications (pp. 341-392). New York: Marcel Dekker, Inc. Xiong Y. L. & Blanchard S. P. (1993). Functional properties of myofibrillar protein from cold-shortened and thaw-rigor bovine muscles. Journal of Food Science, 58, 720-723 Xiong Y. L. & Blanchard S. P. (1994). Myofibrillar protein gelation: Viscoelastic changes related to heating procedures. Journal of Food Science, 59, 734-738 Xiong Y. L. & Brekke C. J. (1991). Protein extractability and thermally induced gelation properties of myofibrils isolated from preand post-rigor chicken muscle. Journal of Food Science, 56, 210-215 Yu S. Y. & Fazidah S.(1994). Enzymatic hydrolysis of proteins from Aristichthys nobilis by protease PAmano. Tropical Science, 34, 381-391 Yong S. H. & Karel M. (1978). Reaction of histidine with methyl linoleate: Characterization of the histidine degradation product. Journal of the American Oil Chemists Society, 55, 352-356 Yongsawatdigul J. & Park J. W. (2001). Gelation characteristics of alkaline and acid solubilization of fish muscle 2001 June 23-27; New Orleans, La. Chicago, Ill: Institute of Food Technologists. Abstract number100-1 Zeigler G. R. & Acton J.C. (1984). Mechanisms of gel formation by proteins of muscle tissue. Food Technology, 38, 77-82

PAGE 119

BIOGRAPHICAL SKETCH Ann Elizabeth Theodore was born and raised in Brooksville, Florida. She received her Bachelor of Science degree in food science and human nutrition from the University of Florida in 2003. Ann was accepted to the Graduate School at the University of Florida for the Fall 2003 semester in the Food Science and Human Nutrition Department. Under the guidance of Dr. Hordur G. Kristinsson, Ann pursued her Master of Science degree, specializing in seafood biochemistry. As an undergraduate at the University of Florida, Ann was awarded the IFAS summer internship of 2002 and participated in the IFT Annual Meetings in 2003. In graduate school, Ann shared her research at two additional IFT annual meetings and was invited into the American Chemical Society and the food science honor fraternity, Phi Tau Sigma. Upon graduation, Ann will begin her doctoral degree in food science at the University of Massachusetts-Amherst under Dr. Herbert O. Hultin. 107