THE UNIVERSITY OF NEW SOUTH WALES
FACULTY OF APPLIED SCIENCE
DEPARTMENT OF FOOD SCIENCE AND TECHNOLOGY
THE UTILISATION OF COWTAIL RAY VISCERA
A thesis submitted for the degree of
Doctor of Philosophy
by
ACHMAD POERNOMO
BSc (Agricultural Technology, Gadjah Mada University)
Ir (Agricultural Technology, Gadjah Mada University)
MAppSc (Food Engineering, UNSW)
Submitted
Sydney, January 1997 U N S W 1 7 JUL 1337 LIBRARY ACKNOWLEDGEMENT
I am greatly indebted to Professor Ken A. Buckle for his supervision throughout the study.
Sincere gratitude is due to the Government of the Republic of Indonesia for granting me study leave and to the Australian Agency for International Development
(AusAID) for the award of a scholarship for my study.
My thanks go to the Head and staff of The Research Station for Marine Fisheries,
Slipi, Jakarta and to the staff of The Department of Food Science and Technology, The
University of New South Wales for their assistance during my fieldwork in Indonesia and finishing of the work in Australia. I am also grateful to the dried salted ray processors in
Muara Angke, Jakarta and Labuhan Maringgai, Lampung, Indonesia, for supplying me the cowtail ray viscera used in this study.
Finally, I wish to express my indebtedness to my wife, Farida Ariyani, who has been constantly supporting me while she herself was suffering from her study creating the agony that she has patiently and strongly borne. This thesis is dedicated to her and also to my parents, Mr. and Mrs. Djarnawi Hadikusuma, who had been sending us their encouragement from a distance, but will never know that their son has finally finished the work.
All praises be to Allah, the One and Only God. TABLE OF CONTENTS
Page
ACKNOWLEDGEMENT i
TABLE OF CONTENT ii
ABSTRACT ix
1 INTRODUCTION 1
2 LITERATURE REVIEW 5
2.1 Aquatic food with special reference to fish 5
2.1.1 Classification of fish 5
2.1.2 Basic fish anatomy 11
2.1.2.1 Shape and skin 11
2.1.2.2 Skeleton and muscle 12
2.1.2.3 Digestive and internal organs 14
2.1.3 Chemical composition of fish 17
2.2 Aquatic food utilisation 21
2.2.1 Utilisation as food 21
2.2.2 Utilisation as feed 27
2.2.3 Others 28
2.3 Aquatic food processing waste 29
2.3.1 Production and characteristics 29
2.3.2 Waste treatment 30
2.3.3 Waste utilisation 33
2.3.3.1 Fin-fish waste 33
2.3.3.1.1 Fish meal 33
2.3.3.1.2 Minced fish and surimi 34 2.3.3.1.3 Fish protein preparations 35
2.3.3.1.4 Pharmaceutical products 37
2.3.3.1.5 Fertiliser and compost 37
2.3.3.1.6 Others 38
2.3.3.2 Crustacean and molluscan wastes 39
2.3.3.2.1 Animal feeds 39
2.3.3.2.2 Chitin and chitosan 40
2.3.3.2.3 Others 41
2.3.3.3 Liquid Wastes 42
2.4 Viscera and its utilisation 43
2.5 The stingrays (Dasycitididcie spp) 47
2.5.1 General description 47
2.5.2 Utilisation 49
2.5.3 Waste products 54
3 ENSILATION OF COWTAIL RAY (T. sephen) VISCERA 58
3.1 Introduction 58
3.2 Materials 63
3.2.1 Viscera 63
3.2.2 Chemicals and solvents 65
3.2.3 Equipment 65
3.3 Methods 65
3.3.1 Preparatory 65
3.3.1.1 Experiment 1 65
3.3.1.2 Experiment 2 66
3.3.2 Analytical 67 3.3.2.1 Proximate analysis 67
3.3.2.1.1 Moisture content 67
3.3.2.1.2 Protein content 67
3.3.2.1.3 Fat content 68
3.3.2.1.4 Ash content 68
3.3.2.2 pH 69
3.3.2.3 Viscosity 69
3.3.2.4 Liquefaction 69
3.3.2.5 Soluble nitrogen 69
3.3.2.6 Enzyme activity 70
3.3.3 Statistical analysis 70
3.4 Results 70
3.4.1 Experiment 1 70
3.4.2 Experiment 2 77
3.5 Discussion 81
3.5.1 pH and spoilage of silage 81
3.5.2 Proteolysis 85
3.5.3 Fish protein hydrolysis 90
4 USE OF CRUDE PEPTONES FROM COWTAIL RAY (T. sephen) VISCERA
SILAGE AS MICROBIAL GROWTH MEDIA 93
4.1 Introduction 93
4.2 Materials 98
4.2.1 Viscera 98
4.2.2 Equipment 98
4.2.3 Chemicals and microbiological media 98
IV 4.2.4 Sources of microorganisms 99
4.3 Methods 100
4.3.1 Preparatory 100
4.3.1.1 Ensilation of cowtail ray viscera and preparation of ray viscera peptones 100
4.3.1.2 Culture media 101
4.3.1.3 Microbial growth test 101
4.3.1.3.1 Preparation of cultures, inoculation and incubation 101
4.3.1.3.1.1 Bacteria and mixed population of microorganisms 101
4.3.1.3.1.2 Yeast 102
4.3.1.3.1.3 Fungi 102
4.3.1.3.2 Growth curves, growth rate and total growth 103
4.3.1.3.3 Biomass production 103
4.3.2 Analytical 104
4.3.2.1 Proximate composition 104
4.3.2.2 Amino acid composition 104
4.3.2.3 Minerals 104
4.3.3 Statistical analysis 105
4.4 Results 105
4.4.1 Protein concentration and proximate composition 105
4.4.2 Growth curves 105
4.4.3 Biomass production 114
4.5 Discussion 115
4.5.1 Proximate composition 115
4.5.2 Growth of mixed population of microorganisms 119
4.5.3 Growth of single microorganisms 121 4.5.4 General discussion 123
5 ISOLATION AND CHARACTERISATION OF COWTAIL RAY (T. Sepheri)
VISCERA PROTEASES 126
5.1 Introduction 126
5.2 Materials 130
5.2.1 Viscera 130
5.2.2 Chemicals 130
5.2.3 Equipment 131
5.3 Methods 132
5.3.1 Preparatory 132
5.3.1.1 Enzyme extraction 132
5.3.1.2 Fractionation of alkaline and acidic proteases 132
5.3.1.3 Acetone precipitation 133
5.3.1.4 Ion exchange chromatography 133
5.3.1.5 Electrophoresis 134
5.4 Analytical 135
5.4.1 General acidic and alkaline proteases assay 135
5.4.2 Trypsin activity 135
5.4.3 Chymotrypsin activity 135
5.4.4 Carboxypeptidase activity 136
5.4.5 Protein analysis 136
5.4.6 Amino acids 136
5.4.7 pH optima 136
5.4.8 pH stability 137
5.4.9 Temperature optima 137 5.4.10 Temperature stability 137
5.4.11 Effect of inhibitors 137
5.4.12 Effect ofNaCl on enzyme activity 138
5.4.13 Effect ofNaCl on enzyme stability 138
5.4.14 Milk clotting activity 138
5.4.15 Effect of substrate concentration 138
5.5 Results 139
5.5.1 Isolation and partial purification 139
5.5.1.1 Fractionation of alkaline and acidic proteases 139
5.5.1.2 Acetone precipitation and dialysis 141
5.5.1.3 Ion exchange chromatography 141
5.5.1.4 Polyacrylamide gel electrophoresis and molecular weight estimation 144
5.5.2 Characterisation of alkaline and acidic proteases 144
5.5.2.1 Optimum pH 144
5.5.2.2 Stability of alkaline and acidic proteases at different pH 144
5.5.2.3 Temperature optima of alkaline and acidic proteases 149
5.5.2.4 Stability of alkaline and acidic proteases at different temperature 149
5.5.2.5 Effect ofNaCl of enzyme activity 149
5.5.2.6 Stability of enzyme in NaCl solution 150
5.5.2.7 Effect of inhibitors 150
5.5.2.8 Substrate specificity 150
5.5.2.9 Milk clotting activity 150
5.5.2.10 Effect of substrate concentration 152
5.5.2.11 Amino acid composition 154
5.6 Discussion 154
vii 5.6.1 Isolation and partial purification of alkaline and acidic proteases 154
5.6.2 Characterisation of acidic and alkaline proteases from cowtail ray viscera 160
5.6.2.1 Acidic proteases 160
5.6.22 Alkaline protease 168
6 CONCLUSIONS AND RECOMMENDATIONS 173
6.1 Conclusions 173
6.2 Recommendations 176
7 REFERENCES 178
8 APPENDICES 227
8.1 Summary of results of statistical analysis for silages from Muara Angke and
Labuhan Maringgai 227
8.2 Raw data of cowtail ray viscera silage from Muara Angke 228
8.3 Raw data of cowtail ray viscera silage from Labuhan Maringgai 232
8.4 Raw data of fish protein hydrolysates 234
8.5 Summary of results of statistical analysis for effects of peptones on growth rate,
total growth and biomass production of test microorganisms 235
8.6 Raw data of use of cowtail ray viscera peptones as microbial growth media 236
viii ABSTRACT
The classification and utilisation of aquatic food are reviewed, including its waste treatment and utilisation. The general description of stingrays and their utilisation in Indonesia are described.
Cowtail ray viscera was preserved by acid ensilation. Parameters investigated included origin of cowtail ray viscera, levels and types of acids, temperature and time of incubation. Changes in pH, soluble nitrogen, solubilisation and viscosity were monitored. Hydrochloric acid liquefied but failed to preserve the viscera. Organic acids liquefied and preserved the viscera for at least 120 days. The pattern of nitrogen solubilisation was comparable to those of cold and temperate water fish. Propionic and formic acids (1:1 v/v; 3% v/w) are recommended.
Peptones were produced from cowtail ray viscera silages prepared with 3% organic acids and 4% HC1, and their ability to support the growth of selected microorganisms investigated. Beef meat, chicken eggs and cow's milk were used as sources of mixed populations of microorganisms, while pure cultures used were
Aspergillus flcivus, Bacillus subtilis, Escherichia coli, Sacchciromyces cerevisiae and
Staphylococcus aureus. Commercial peptones (Oxoid, Difco and BBL) were included as controls. Peptones from cowtail ray viscera silage supported strong growth of the test microorganisms. Based on growth rate, total growth and dry biomass production of the test microorganisms, the viscera silage peptones were ranked 1 or 2 compared to the controls.
Extraction of proteases from cowtail ray viscera was conducted by ensilation followed by fractionation using polyelectrolyte. Two acidic proteases were isolated and identified as pepsin-like enzymes having molecular weights of about 30,000D and one alkaline protease was identified as a trypsin-like enzyme having a molecular weight of about 45,000D.
Cowtail ray pepsins behaved similarly to cold and temperate water fish pepsins against hemoglobin in terms of temperature and pH optima, and their stability at different temperatures and pHs. Salt did not affect activity, while milk clotting activities were low. Cowtail ray trypsin had a higher temperature optimum than cold and temperate water fish trypsins, and was similar to other Indonesian fish alkaline proteases. Other characteristics such as thermal stability, pH optima and stability were similar to other fish trypsins. 1 INTRODUCTION
Fish contributes about 6% of all protein, and 18% of animal protein in human nutrition. The world production of fish is increasing yearly, and in the last decade, production has increased by 1.8 Mt annually. In 1993, the production was lOIMt, of which about 70% was for human consumption.
Indonesia is one of the major fishing countries and ranks 8th in the world in terms of production. Currently the total fish catch in Indonesia is about 4 Mt, of which 3.1 Mt is from marine fishery and 0.9 Mt from inland fishery, almost all of which is for domestic consumption. Products that are usually exported are canned fish, frozen tuna and prawns, and live ornamental fish, while the principal imported product is fish meal.
The Indonesian fishery is dominated by artisanal or small scale fishing, while fish processing activities, which are mostly traditional, are scattered throughout the country and consist mainly of backyard or home activities. Annually, about 54% of the catch is marketed and consumed fresh, while almost 40% is converted into traditional products, of which 75% is dried and salted. Fish processed into dried salted products include small pelagic fish such as sardines, scad, mackerel and anchovies, as well as demersal fish such as sharks and rays. Small fish are usually salted and dried whole, while large fish are usually gutted and butterfly-shaped, sliced or filleted.
Rays are landed in many parts of Indonesia and are seldom consumed fresh.
They are available almost the whole year and are usually processed into dried salted products. The drying and salting of rays are concentrated in Muara Angke, Jakarta and Labuhan Maringgai, Lampung, Southeast Sumatra. There are a number of ray species that are usually processed into dried salted products, but the majority is
1 cowtail ray (Trygoti sephen).
Muara Angke is a Government operated fishery area located in North Jakarta.
Here, the Government provides a fishing port where fish are landed and auctioned, stalls for selling fresh fish, a processing site mainly for salting and drying, and fishermen’s/processors’s housing. The processing site in Muara Angke is the largest of its kind in Indonesia. Each premise is provided with facilities for fish washing and salting, a drying yard with raised racks, and storage facilities. By-products from this processing site include dried tail bones and skins of rays. An area for wastes is also provided, at which the processing wastes are dumped onto an open concrete inclined floor with drainage which brings the liquid waste into a nearby pond without, currently, any treatment. In Muara Angke, about 2250 t rays are landed and processed into dried salted products annually.
Labuhan Maringgai is a small fishing village on the tip of southeast Sumatra.
Salting and drying are usually done in an extension of a fishermen’s or processor’s house. Although dried salted fish is not the only food product of this village, the quantity of dried salted fish products is high and they are marketed to large cities in
Java and Sumatra; some products are exported to Singapore. Other fish processed in the area includes anchovies, marine catfish and small shrimp. Fish processing wastes are directly dumped into a nearby river. In Labuhan Maringgai, about 2700 t rays are landed and processed into dried salted products yearly.
The two areas produce different types of dried salted products from rays.
Products from Muara Angke are thicker and brownish in colour, while those from
Labuhan Maringgai are much thinner and lighter in colour commanding a higher (3-5 times) price in the market. Each product has different market segments. Those from
Muara Angke are marketed in rural areas in West Java, while those from Labuhan
2 Maringgai are marketed in cities.
In the processing of dried salted ray products, only the wings (66-80% of total weight) are used, while the other parts of the body are wasted. During the past years, there have been a number of efforts to utilise these wastes: currently, the dried tail bones and skins are sold separately. Other wastes such as frames and offal, including viscera, are not utilised and are directly disposed of without any treatment.
The viscera is an internal organ which usually contributes about 20% of body weight and contains a number of digestive enzymes which remain active. This has made the viscera very susceptible to autolysis and spoilage takes place in a short time, especially in relatively high ambient tropical temperatures such as in Indonesia. In areas such as Muara Angke and Labuhan Maringgai, where significant amounts
(about 500 t) of viscera are produced annually, the viscera creates severe environmental problems. Moreover, as rays are elasmobranchs, an additional problem with ammoniacal smells is prevalent. The utilisation of the viscera and conversion into useable by-products not only will increase production efficiency and profitability, but also will reduce waste disposal problems.
Studies on the utilisation of the viscera of elasmobranch fish such as rays and sharks are very limited. Additionally, although research on the utilisation of fish viscera has been well documented, information on the utilisation of viscera of tropical fish is lacking.
The aims of the present studies were:
* to evaluate the ensilation of ray viscera, with particular reference to the effect of
different types and proportions of acid as well as temperature,
* to evaluate the capability of crude enzymes from ray viscera silage of hydrolysing
fish flesh, * to prepare crude peptones from ray viscera silage and analyse their proximate
composition and capability in supporting microbiological growth in comparison
with commercial peptones, and
* to isolate and characterise acidic and alkaline proteolytic enzymes from ray viscera
silage.
4 2 LITERATURE REVIEW 2.1 Aquatic food with special reference to fish
Aquatic food, especially fish, is regarded as one of the most important foods of indigenous people in their hunting stage (Cutting 1962a). Paleolithic drawings found in the Slav regions, such as those in Poland, and marine fish bones in caves inhabited by indigenous people distant from the sea 40,000-20,000 years BC indicate that fish were included in their diets and confirm that fish preservation, most probably by sun-drying, was done before transportation to other regions (Cutting 1962a,
Dembinska 1988, Horner 1992a). Evidence also shows that pond culture has been
practiced by Ancient Egyptians, in addition to fishing the Nile and the Mediterranean
(Pigott and Tucker 1990). The first known fish-like vertebrate fossil was probably from the Ordovician or even late Cambrian periods, approximately 450 to more than
500 million years ago (Pulley 1974, Bond 1979).
2.1.1 Classification of fish
Fish are the most abundant of the vertebrates and are also very diverse, comprising more than 24,000 species; almost 100 species are described each year
(Lagler et a/. 1977, Nelson 1994, Bone, Marshall and Blaxter 1995). Although this figure represents more than half of all known vertebrates (60%), not all species have been studied as extensively as other vertebrates.
Taxonomically, fish have been the subject of many in-depth studies, although there is still much disagreement. Nelson (1994) defines fish as “aquatic vertebrates that have gills throughout life, and limbs, if any, in the shape of fins”, while Gomon,
Glover and Kuiter (1994) specifically define fish as “those major groups belonging to the phylum Chordate which as adults breathe through gills and are free swimming but whose locomotory appendages have not advanced beyond primary fins”.
5 The Food and Agriculture Organization of The United Nations (FAO)
categorised aquatic species, including all aquatic animals and plants, into 9 divisions.
This system is used for statistical purposes and is named the FAO International
Standard Statistical Classification of Aquatic Animals and Plants (ISSCAAP). In this
system, all aquatic species are further classified into 51 group of species, and each is given a separate code, which includes both marine and fresh water origins (Table
2.1).
Table 2.1. International Standard Statistical Classification of Aquatic Animals and Plants (ISSCAAP)
Code DIVISION Code DIVISION Group of species Group of species
1 FRESH WATER FISH 53 Oysters 11 Carps, barbels and other cyprinids 54 Mussels 12 Tilapias and other cichlids 55 Scallops, pectens, etc. 13 Miscellaneous freshwater fish 56 Clams, cockles, arkshells, etc. 57 Squids, cuttlefish, octopuses, etc. 2 DIADROMOUS FISH 58 Miscellaneous marine molluscs 21 Sturgeons, paddlefish, etc. 22 River eels 6 WHALES, SEALS AND OTHER 23 Salmons, trouts, smelts, etc. AQUATIC MAMMALS 24 Shads, etc. 61 Blue-whales, fin-whales, etc. 25 Miscellaneous diadromous fish 62 Sperm-whales, pilot-whales, etc. 63 Eared seals, hair seals, walruses, etc. 3 MARINEFISH 64 Miscellaneous aquatic mammals 31 Flounders, halibuts, soles, etc. 32 Cods, hakes, haddocks, etc. 7 MISCELLANEOUS AQUATIC 33 Redfish, basses, congers, etc. ANIMALS 34 Jacks, mullets, sauries, etc. 71 Frog and other amphibians 35 Herrings, sardines, anchovies, etc. 72 Turtles 36 Tunas, bonitos, billfish, etc. 73 Crocodiles and alligators 37 Mackerels, snoeks, cuttlefish, etc. 74 Sea-squirts and other tunicates 38 Sharks, rays, chimaeras, etc. 75 Horseshoe crabs and other arachnoids 39 Miscellaneous marine fish 76 Sea-urchins and other echinoderms 77 Miscellaneous aquatic invertebrates 4 CRUSTACEANS 41 Freshwater crustaceans 8 MISCELLANEOUS AQUATIC ANIMAL 42 Sea-spiders, crabs, etc. PRODUCTS 43 Lobsters, spiny-rock lobsters, etc. 81 Pearls, mother-of-pearls, shells, etc. 44 Squat-lobsters 82 Corals 45 Shrimps, prawns, etc. 83 Sponges 46 Krill, planktonic crustaceans, etc. 47 Miscellaneous marine crustaceans 9 AQUATIC PLANTS 91 Brown seaweeds 5 MOLLUSCS 92 Red seaweeds 51 Freshwater molluscs 93 Green seaweeds and other algae 52 Abalones, winkles, conchs, etc. 94 Miscellaneous aquatic plants
Source: Anon. (1995a)
6 From their habitats, fish are grouped into two major groups, i.e. marine and fresh water fish. In the marine habitat, there are several further classifications such as those from the open ocean and from shallow seas and coastal regions. A complete classification of fish according to their habitat is shown in Figure 2.1.
Epipelagic fish
Mesopelagic fish
Open ocean fish Bathypelagic fish
Benthopelagic fish
Benthic fish
Marine fish
Warm water fish
Temperate and cold water fish Shallow seas and coastal region fish Estuarine fish
Intertidal fish
Freshwater fish
Figure 2.1. Fish classification based on their habitats (drawn from Bone et a/. 1995)
Classification of fish in the open ocean is mainly based on the depth of the ocean, with names following the classification of the marine environment on the basis of water depth, starting from the zone extending from shore to the continental shelf, i.e. the first 100m depth (epipelagic fish) to the depth of 4000m
7 (abyssopelagic/benthopelagic fish) down to the bottom of the sea (benthic fish).
Pelagic fish are defined as fish which are either floating or swimming in the water, while those attached or living on or in a substrate are called benthic or demersal fish
(Reish 1969).
Ocean depths influence the characteristics of the fish such as their migration, feeding habits and morphology (Bone et al. 1995). For example, many mesopelagic fish (those that live in a depth between 100m to 1000m), in order to find their foods, migrate upward to the surface (the epipelagic zone) at night and then downward before dawn following the migration of their zooplankton food (Moyle and Cech
1982, Bone et al. 1995). Fish from the benthopelagic zone usually have gas-filled swim bladders, while those from the benthic region do not. As light is not able to reach depths of 350m or more, the eye morphology of fish from those depths is different from those of deeper parts of the ocean (Sikorski and Karnicki 1990).
In the shallow seas and coastal regions, fish are classified based upon the temperature of the water (warm water fish and temperate and cold water fish) and locations where the conditions of the water may change from time to time (intertidal and estuarine waters). There are about 10,000 species of fish that live in the shallow seas, and almost 80% are in warm waters, in which the temperature of the water never falls below 18°C, even during the coldest part of the year (Bone et al. 1995).
In some places, the temperature may reach 44°C such as in an African lake where tilapia can be found (Nelson 1994). Temperate and cold waters are less rich in species, with about 1000 species in the temperate North Pacific and fewer in the temperate North Atlantic. However, many of the fish from these regions, such as
Greenland cod (Gadus ogac), can withstand extremely low temperatures, e g. to -2°C
(Simpson and Haard 1984a).
8 Estuaries and river mouths are transitional environments between freshwater
and salt water, thus they are very unstable where water salinity and temperature are
variable, and the waters are often turbulent and muddy. Fish in this area are adapted
to these conditions and are mainly salt-tolerant. There are usually five different types of fish found in this area: freshwater, diadromous, true estuarine, non-dependent marine and dependent marine fish (Moyle and Cech 1982). The relative abundance of these fish varies from season to season and locality to locality.
The intertidal zone is a harsh environment and subject to the rise and fall of tides. The fish in this area have to be able to adapt to this condition; sometimes they are alternately buffeted by waves and isolated in pools or on mudflats (Bone et al.
1995), as well as confronted by strong, fluctuating gradients of temperature and salinity, and general severity from the upper intertidal areas to the lower intertidal areas (Moyle and Cech 1982). Generally, fish in this zone are of four kinds: true residents, partial residents, tidal visitors and seasonal visitors (Moyle and Cech 1982).
From a processing or technological point of view, fish are classified as lean or fatty (Huss 1988), which is based on their fat content. Borgstrom (1962) subdivided fish into three groups according to their fat content: lean fish (fat content less then
2%, wet weight basis/wwb), semi fatty fish (fat content 2-10%, wwb) and fatty fish
(10-26%, wwb). Another classification of fish based on both fat and protein contents was proposed by Stansby (1982). According to this classification, fish are categorised into 5 groups: Group A, B, C, D and E (Table 2.2).
Based on their food, fish can be classified into 4 categories, namely detritivores (feeding on detritus food), herbivores (feeding on plant food), carnivores
(feeding on meat or other fish) and omnivores (feeding on almost all type of food)
(Moyle and Cech 1982). Under these categories, fish can be further grouped into
9 euryphagous (those having a mixed diet), stenophagous (having limited assortment of food types) and monophagous (having only one type of food). The majority of fish
are euryphagous carnivores.
Table 2.2. Classification of fish based on oil and protein contents
Category Type of fish Prototype species
A Low oil (<5%), high protein (15- Pacific cod (Gcidus macro- 20%) cephalus) B Medium oil (5-15%), high protein Mackerel (,Scomber scrombus) (15-10%) C High oil (>15%), low protein Siscowet lake trout (Cristivomer (<15%) namacush siscowet) D Low oil (<5%), very high protein Skipjack tuna (.Katsmvomis (>20%) pelamis), light meat only E Low oil (<5%), low protein Butter clams (Saxidomus nuttalli) (<15%) Source: Stansby (1982)
Fish are also classified according to their feeding habits, based on the manner the food is taken into the mouth (Lagler et al. 1977). This classifies fish into 5 different categories, i.e. predators (fish that feed on macroscopic animals), grazers
(fish that take feed by bites, or by individual small ones), strainers (fish that take feed by straining waters, and thus select the food by size not by kind), suckers (fish that suck the feed or feed-containing materials) and parasites (fish that suck body fluids from the host animal after rasping a hole in the side of the body).
According to Nelson (1994), fish belong to the phylum Chordate, subphylum
Vertebrate. They are divided into two superclasses: Agnatha (the primitive jawless fish) and Gnathostomata (jawed fish). There are 3 classes of Agnatha (one of which is extinct), and 5 classes of Gnathostomata (2 are extinct). The fish classes that still exist today are Myxini and Cephalaspidomorphi (under superclass Agnatha), and
10 Chondrichtyes, Sarcopterygii, and Actinopterygii (under superclass Gnathostomata).
Under these classes there are 482 families, 4,258 genera and 24,618 known species
(Nelson 1994). Chondrichtyes are commonly known as cartilaginous fish, while
Sarcopterygii and Actinopterygii are bony fish, or Euteleostomi. There are 846 species of Chondrichtyes and 23,689 species of Euteleostomi. Commercially, only
Cephalaspidomorphi, Chondrichtyes, and Euteleostomi are important (Huss 1988).
2.1.2 Basic fish anatomy 2.1.2.1 Shape and skin
Most fish, especially the bony fish, are torpedo-shaped or fusiformed, with ovoid cross-section and bilateral symmetry (Lagler et al. 1977). There are also variations in the shape such as compressiform (like that of sunfish and perchfish), depressiform (catfish and skate), anguiliform (eel-shaped fish), filiform (snipe eel), taeniform (ribbon-like fish), sagittiform (arrow-like shaped fish) and globbiform
(lumpsuckers) (Bond 1979). Many fish have unique forms which can not be included in the above term, such as that of sea horses or sea moth. However, in general, the fish body can be externally divided into three distinguished parts: head, trunk and tail.
An ordinary fish is covered with a relatively tough skin made up of two layers, i.e. outer epidermis and inner dermis. The epidermis is very thin and is composed of
10-30 layers of cells. Some cells are unicellular mucous glands that produce the mucus which forms the slimy coating of fish. The colouring of many fish is also due to the coloured cells of these layers. Most fish skin is armoured by scales which are dermal in origin. Some fish, such as catfish, do not have any scales, and some have scales that are modified into bony plates or scutes such as sturgeons (Bond 1979).
The fish scales range in size, thickness, ornamentation, structure as well as the strength of their attachment to the body. The skin of fish which do not have any
11 scales, is usually thicker and tougher, and hence it can be taken off from the flesh,
tanned and made into leather.
2.1.2.2 Skeleton and muscle
The skeletal system of a fish may be considered to have three main components, namely the vertebral column, the skull and the appendicular skeleton
(Moyle and Cech 1982). The vertebral column, which is usually also called the backbone, is composed of a series of segments, i.e. the vertebrae, running from the head to the tail fin. Usually there is only one vertebra per body segment, however, other fish such as shark may have two vertebrae in their body (tail) segment (Lagler et a/. 1977). The vertebral column in fish ranges in development from cartilaginous to bony. Along the trunk, the vertebrae often have lateral processes bearing ribs.
According to the location, there are two types of ribs, namely dorsal and ventral ribs, but not all fish have both ribs. The ventral ribs are reportedly limited to bony fish only, while almost all fish have dorsal ribs except those from the Cyclostomes and
Holocephali subclasess (Lagler et al. 1977). In higher bony fish, there are small, splint bones, called intermuscullar bones, of different shapes found in the layers of muscle tissue. These bones cause problems during filleting operations, and even bring about difficulty when they become lodged in the throat during eating.
The muscle of fish is basically made of blocks, called myomeres or myotomes, which are separated with connective tissues called myosepta or myocommata. The amount of connective tissue in fish is much less than that in mammalian tissue. The myomeres are shaped like a “W” on its side, and on the right and left halves of the body they are separated by a vertical septum, while a horizontal septum separates the muscle masses on the upper and lower halves of the body (Figure 2.2). Lampreys and hagfish do not have any horizontal septum (Bond 1979).
12 supracarinalis
:myosepta epaxial myotomes
vertical septum vertebra red lateral muscl ’spinal cprd horizontal septum notochord hypaxial myotomes
■oody cavity
myotomes
infracarinalis
Figure 2.2. Lateral and cross section of lamprey (A, Bond 1979) and salmon body (B, Greene and Greene 1914)
White muscles are poorer in blood supply, and have lower levels of oxygen
affinity pigments such as myoglobin. The fibers are thicker than those of dark
muscles. White muscles are reported to be used for sudden bursts of swimming such as in escape or in the capture of prey. Dogfish sharks, with almost all white muscles,
13 are able to use up to 50% of available muscle glycogen in about two minutes (Bone
1966). From a technological point of view, the presence of dark muscle is very important, as the high fat content makes it susceptible to oxidation which may lead to rancidity during processing and storage.
Fish muscle is mostly white, but pink in salmonids and tuna flesh, which is due to carotenoid pigment (FFaard 1992a). Depending upon the species, many fish also
have dark muscles of a brown or reddish colour which are located immediately under the skin. This dark muscle has many capillaries and appears red because of its high concentration of pigments possessing the ability to bind oxygen in the blood
(hemoglobin) and in the tissue (myoglobin). The proportion of dark to white muscles varies from species to species depending upon the activity the fish. The more active fish have a higher proportion of dark muscle, since they need more oxygen. The oxygen is needed by dark muscle fibres to aerobically metabolise fats as energy sources. Consequently the fat content of dark muscle is higher than that of white muscle. Typical fish with dark muscle are those from the pelagic group, with high levels of continuous swimming activity. Species such as herring and mackerel may have dark muscle up to 48% of their body weight (Love 1970).
2.1.2.3 Digestive and internal organs
Fish usually swallow their food without chewing or masticating, although many have teeth. These are used only to break up the food lumps, but not for chewing. Flowever, the structure of the mouth differs from species to species and is related to their feeding modes and habits (Moyle and Cech 1982). Catfish have powerful jaws with conical teeth in front and flat crushing teeth on each side used for smashing molluscs (Love 1982). Pikes or sharks have large mouths equipped with large sharp teeth, while other deep-sea predators have dagger-like teeth (Bond 1979).
14 The esophagus of fish is short but distensible making it able to swallow
relatively large objects. Some fish have muscular sacs connecting to the esophagus
some of which are lined with teeth, and serve various functions such as production of mucus, storage of foods or preparation of foods by pulverisation (Bond 1979).
Although the stomach is one of the important organs in the digestive system, not all fish have one, e.g. including lampreys, hagfish, chimaeras, and some bony fish which include sauries and parrotfish (Barrington 1957, Bond 1979). Normally, the stomach is shaped like a “U” or “V”, being essentially a bent muscular tube. Some variations of the shape and length may occur depending upon the type of food the fish eats. The stomach of carnivorous fish is elongated while that of omnivorous fish is sac-shaped (Lagler et al. 1977). Carnivorous fish generally have a short stomach and omnivorous fish have guts of medium length, while fish which feed on detritus and algae usually have a longer stomach (Bond 1979, Moyle and Cech 1982). This organ is essentially a food grinder and contains different types of enzymes which are produced by the stomach walls or the pancreas. Ground foods from the stomach are further digested and absorbed by the intestines, which also vary in shape and length.
Generally, the entire intestines are short, up to about twice the body length.
However, some fish such as herbivorous fish often have very long intestines, many times the length of their body, while carnivores such as the pike have shorter intestines. Other fish such as cartilaginous fish have spiral valve intestines, which increase the digestion and/or absorption, thus improving the absorption efficiency. A greater surface area of the intestines is also found in detritivorous fish which is achieved by elongation of the intestines such as those of the cyprinid Labeo (having
15-21 times the body length) or by having well-developed mucosal folds as in
Prochilodus (Wootton 1992).
15 Other internal organs attached to or associated with the intestine are the pyloric caeca, liver, pancreas and gas bladder. Pyloric caeca are mostly found in bony fish, while cartilaginous fish and other vertebrate groups lack this organ. Their number vary with species, some may have only one, such as monkfish; yellow perch has five, flatfish have no more than five, but others such as mackerel, salmon and sea- snail may have two hundred or even more (Bond 1979, Love 1982). This organ may have a digestive function as some enzymes such as proteases, lipases, lactase and invertase, have been found in it (Barrington 1957, Lagler et aL 1977, Williams 1975).
The liver is the largest internal organ in vertebrates, and its functions include digestion processes by bile secretion, as well as fat and carbohydrate (glycogen) storage. Species such as sharks and rays have extremely large livers contributing up to about 20% of body weight. The liver is important due to its high content of fat as well as fat soluble vitamins. The pancreas also plays an important role in food digestion by producing several digestive enzymes, especially carbohydrases and lipases (Moyle and Cech 1982) which are secreted into the stomach. In addition to its digestive functions, the pancreas also has the function of insulin secretion which is important in carbohydrate metabolism.
The gas bladder is another internal organ with several important functions, including as an accessory breathing organ, as a sound producer as well as a resonator in sound reception and a hydrostatic balancing/buoyancy organ (Bond 1979). It also functions as a fat-storage organ, such as in certain deep-sea mouthfish relatives, e.g. the gonostomatid Yarella (Lagler et aJ. 1977). The gas bladder is normally a thin- wall, torpedo-shaped sac typically found in the upper part of the body cavity immediately below the kidney. Fish such as those from Chondrichtyes usually lack this organ and benefit from the presence of their large livers or their much reduced
16 average density due to their cartilaginous skeleton to replace one of the gas bladder's function, i.e. buoyancy (Bond 1979, Love 1982, Moyle and Cech 1982). Bottom fish usually do not possess a gas bladder, while fish in the upper 200 m depth most frequently have it.
2.1.3 Chemical composition of fish
Fish as food offers a rich source of nutrients, providing complete and balanced protein, a variety of vitamins as well as minerals, and although having a relatively low caloric value it provides health-promoting fatty acids. The chemical composition of fish varies greatly, not only from species to species, but also from individual to individual depending upon age, sex, environment as w^ell as season. However, the most dramatic changes in the composition are due to the fat and water contents of fatty fish. The sum of water and fat in the fish flesh is roughly constant, which is estimated at about 80%, thus the greater the quantity of fat, the smaller the water content and vice versa. The other major component is crude protein which, together with fat and water, makes up about 98% of the total mass of the fish flesh. The minor components include carbohydrate, vitamins and minerals. Levanidov (1980) proposed the relationship between the major components of fish after measuring them in 153 fish species as follows:
Water (%) + Lipid (%) = 98.8 -1.01 Protein (%).
Water content of fish muscle or marine invertebrates varies from about 50 to about 85%, depending upon the species and also the nutritional status of the animal
(Sikorski, Kolakowska and Pan 1990, Pigott and Tucker 1990). The lowest water content (46.1%) was found in the flesh of eel tail, while the highest (92.9%) was in rat-tail fish (Kleimenov 1983). In an individual fish, the water content will increase during spawning due to starvation which depletes the energy reserves. Typically,
17 water content in fish is in the range of 74.8% (Stansby 1963) to 77.2% (Pigott and
Tucker 1990).
Protein in fish is a very important component since together with fat it affects both the nutritional value and the sensory properties of fish. The fish flesh usually contains crude protein in the range of 11-24% depending upon the species, the nutritional condition and the type of muscle (Sikorski 1994), but most will contain around 20 ± 2% (Sidwell 1981). The lowest protein content (5.1%) was found in lumpsucker, while the highest (30.8%) was in big-eye tuna (Kleimenov 1983).
Nutritionally, fish protein is comparable with that from meat, poultry or milk, especially due to its essential amino acid pattern having high values well compared to the FAO reference pattern (Pigott and Tucker 1990), and in vivo digestibility of 90-
98% (Sikorski et al. 1990). The proteins in fish muscle can be divided into three groups: structural proteins, which constitute 70-80% of the total protein content; sarcoplasmic proteins, 25-30% and connective tissue proteins, 3-10% (Huss 1988).
The collagen, which affects the toughness of raw fish meat, belongs to the third group.
Some non-protein nitrogen (NPN) compounds are present in fish and although considered as minor components they are important in the determination of fish meat quality. These components can be defined as water-soluble, low molecular weight, containing nitrogen of a non-protein nature, and usually have some flavour (Love
1982, Huss 1988). The amount of NPN in fish flesh is normally between 9-18% of the total nitrogen, but in cartilaginous fish, this may increase to 55% (Sikorski 1994).
The NPN contents of fish flesh vary depending upon the species, the habitat, size, season as well as the freshness of fish. The major components of NPN are volatile bases such as ammonia and trimethylamine oxide (TMAO), creatine, free amino acids,
18 nucleotides and purine bases; and in cartilaginous fish, urea. The latter affects the flavour of products from cartilaginous fish and is partly the reason why these type of fish develop a bitter taste which develops into an ammonia taste during spoilage.
The fat content of fish varies widely from species to species and is also used to classify fish, from lean fish which contains less then 2% fat to high fat fish containing up to 26% fat (Borgstrom 1962). The fat is not evenly distributed and is mostly concentrated in the subcutaneous fatty layer of marine mammals and fatty fish, in the liver of lean fish, in the muscle tissue and gonads. The amount of crude fat in fish varies according to species, age, body part and season (pre- or post-spawning).
Generally, the fat content increases with age and increases from tail to head with higher deposition in the belly flap and red/dark muscle or in the liver (Love 1970,
1980, 1982, Pigott and Tucker 1990). Unlike the protein content, the range of fat content in an individual fish varies widely during the year. Atlantic herring caught in
May contained 4.6-5.8% which increased to 19-22% in July (Kleimenov 1983).
Similarly, oil sardine from Bali Strait had a fat content of 8.7% in May and increased to 21.3% in October (Burhanuddin et al. 1984). In general, the fat content from groundfish ranged from 0.5 to 10.0%, pelagic fish from 1.2 to 21.2% and mollusc and crustaceans from 0.6 to 2.4% (Ackman 1992).
The fat depots in most fish usually consist of triacylglycerols, and in some species, these triacylglycerols are sometimes found together with other types of lipids, or even replaced by them such as in sharks or other cartilaginous fish, where a significant amount of fat may consist either of diacyl alkyl glyceryl esters or hydro carbons such as squalene. A small amount of fish fats is in the form of phospholipids and sterols which serve a role in bio-membrane structure and basic cellular functions.
These fats occur in small but relatively constant amounts of 0.2-0.3% wet weight of
19 tissue, and although principally they are not an energy reserve, some fish which do not have fat depots in the muscle tissue may use them as energy during longer starvation periods (Love 1980, Sikorski et al. 1990).
The sterols in fish are mostly cholesterol and amount to about 20-40 mg cholesterol per 100 g tissue in most fish (Sikorski et al. 1990), while in high fat fish the amount may increase to 95-100mg/100g tissue (Pigott and Tucker 1990, Huss
1988).
Generally, the fatty acids derived from fish fats are of three principal types: saturated, monounsaturated and polyunsaturated (Stansby, Schlenk and Gruger
1990). The proportion of polyunsaturated fatty acids in fish fats, especially n-3 and n-
6, are relatively higher than in mammalian lipids and has attracted much attention.
Extensive studies on these fatty acids have been well documented (Barlow and
Stansby 1982, Simopoulos, Kifer and Martin 1986, Kinsella 1987, Lees and Karel
1990, Nettleton 1995).
Fish contains an insignificant amount of carbohydrates, mostly glycogen, which is used as an energy reserve. After catching, fish are usually rendered immobile and glycogen is rapidly hydrolysed, therefore the amount of this substance in fish or fish products is very small.
Fish have both water soluble and fat soluble vitamins in their flesh or organs.
Water soluble vitamins such as the B complex and C are present in the flesh at approximately the same levels as those in other animal muscle meats. Fish is a poor source of thiamin, but it has a modest amount of biotin, folacin, niacin and pantothenic acid; some are rich in pyridoxine, such as tuna and salmon, and in Bi2, such as anchovies, herring, pilchard and sardine (Nettleton 1985, Kinsella 1988). Fat soluble vitamins are mostly found in the liver where fats are deposited, although in
20 small amounts they are also present in the flesh. The amount of these vitamins in the fish is affected by the concentration of fat. The flesh of white fish contains from 25-
50 IU of vitamin A per 100 g meat, while fatty fish may contain 100-4500 IU per 100 g meat (Sikorski et aJ. 1990). Similarly, the amount of vitamin D is also higher in fatty fish than in lean fish. Although fish liver contains concentrated amounts of vitamins A and D, it is not considered as an important source since it is not consumed as such (Kinsella 1988, Pigott and Tucker 1990). Vitamins E and K are present in smaller amounts in fish and marine invertebrates (Sikorski et a/. 1990).
Aquatic food is a good source of mineral components, especially calcium, potassium, selenium, iodine and fluorine (Gordon 1988, Sikorski et al. 1990). The total content of minerals in raw fish flesh and marine invertebrates is in the range of
0.6-1.5%. They are present as macroelements (those which occur at levels of several to hundreds of mg per 100 g meat) and microelements (which are present at levels not higher than 0.1 to a few tens of mg per 1 g meat).
2.2 Aquatic food utilisation 2.2.1 Utilisation as food
Aquatic foods, especially fish and crustaceans, play an important role in many parts of the world as the main source of animal protein. Of total world supplies, fish contributes 6% of food protein and 18% of animal protein (Mackie 1994). The total world catch of aquatic organisms for 1993 is more than 100 M t (Table 2.3), including crustaceans and molluscs. More than 70% of the total world catch was for human consumption, while the rest w;as for reduction to fish meal or fish oils (Anon. 1995b).
The edible portion of aquatic food which is usually fit for human consumption ranges from 30% for king crab to 65% for round herring (Suzuki 1981). For human consumption, aquatic food is usually marketed fresh, frozen, cured or canned.
21 Table 2.3. Production of the world aquatic organisms in 1993
Aquatic organisms according to Production (tons) Proportion ISSCAAP* of total (%)
Freshwater fish 14,546,300 14.3 Diadromus fish 3,303,300 3.3 Marine fish 68,003,900 67.1 Crustaceans 5,462,000 5.4 Molluscs 9,647,500 9.5 Miscellaneous aquatic animals 454,500 0.4
Total 101,417,500 100 * International Standard Statistical Classification of Aquatic Animals and Plants Source: Anon. (1995a)
Fresh marketing involves cold chain systems in which chilling is applied either by icing or refrigeration during distribution. These systems are applied because aquatic foods are perishable and start to deteriorate soon after catching. Although this deterioration is irreversible, reducing the food temperature as low as possible will slow down deterioration and hence maintain food quality as high as possible. Thus chilling aquatic foods as soon as possible after catching is necessary . Icing of aquatic foods during transport has been used for 200 years in Europe, e.g. to transport fresh salmon from Scotland to London, while Chinese fishermen used snow to gain similar preservative action in earlier times (Cutting 1955).
The storage life of chilled fish varies from species to species, and is affected by size, form, the method of capture and locality. In elasmobranch, the urea content is high and thus the development of ammonia during postmortem is rapid; this makes their shelf-life shorter than for bony fish. Fish from tropical waters have a longer storage life in ice than those from cooler or temperate waters, although there have been different explanations for this (Lima dos Santos 1981, Sumner and Magno-
Orejana 1985, Garthwaite 1992). In icing of tropical fish, the temperature of the fish
22 decreases significantly, by up to 30°C, and this possibly gives more pronounced effects on the microbial flora and autolytic enzymes (Huss 1988). Shewan (1977) suggested that psychrotrophic bacteria responsible for spoilage of chilled fish which are the dominant bacterial group in temperate fish are insignificant in tropical fish, thus during icing, tropical fish suffer less psychrotrophic bacterial decomposition than temperate fish do. Similarly, Mayer and Ward (1991) explained that the bacterial flora of cold-water fish were not inhibited by icing as effectively as that of tropical fish.
Natural freezing to preserve foods, including fish, was a common practice in countries with cold climates such as Russia (the previous USSR) and Canada
(Cutting 1955). Fish freezing was firstly patented in England in 1842, while commercial fish freezing began in the USA in the 1860s (Graham 1982). Since spoilage processes still take place at temperatures just below -1°C, the freezing should be completed as quickly as possible. In the UK, it is recommended that during the freezing process, the temperature of the aquatic food should be reduced from 0°C to -
5°C in 2 h or less (Garthwaite 1992). The temperature of the aquatic food during the freezing process should be reduced to -30°C before transfer to cold storage at the same temperature. Aquatic food can be frozen in blocks (for small fish such as herring mackerel as well as prawns) and also as individuals, such as tuna. The storage life of frozen aquatic food at -30°C varies greatly, depending upon the species, type and form (whole, gutted, filleted) and treatment before freezing.
Curing, which includes drying, salting or smoking, is the oldest documented form of food preservation. The preservative effect of curing is achieved by rendering the medium an unsuitable environment for microbial propagation by reducing the water available for microorganisms. This available water or water activity (aw) can be
23 reduced by increasing the concentration of soluble substances in the medium, either by reducing the water content (drying) or by diffusing soluble substances into the flesh
(salting/brining, sugar curing). In Europe, salting and drying of (white) fish including herring was probably begun in the early Middle Ages (Cutting 1955, Voskresensky
1965, Waterman 1976). In Asia or Africa, fish salting has also been practiced for centuries. Although there have been a number of changes in the way fish is processed in Asia, salted and dried fish remains a staple diet (Maynard 1983).
There are three basic methods of salting aquatic food: dry, wet and mixed salting. Dry salting is done by burying the fish in salt crystals; wet salting is achieved by immersing the fish in a concentrated brine; while mixed salting is the combination of both methods. In Indonesia, the latter is widely practiced by traditional processors to salt medium size fish such as mackerel or sardines. The brine used by traditional processors is usually saturated, but a study has shown that 15-21% (w/v) brine was sufficient to produce good quality products (Poernomo et al. 1992). For smaller fish such as anchovies, salting is usually done in boiling brine for 10-15 min. In wet and mixed salting, the brine in the salting tanks is commonly reused for several times or fresh brine is added to salt a new batch of fish (Poernomo and Utomo 1990). Dry salting is employed in the production of fermented fish such as peda (fermented Indian mackerels) and jambal (fermented marine catfish). Other fermented seafood products involving the use of salt such as fish sauce, fish and shrimp pastes
(terasi/belachan), and bekasatig are also popular in Southeast Asia (Van Veen 1965,
Adams, Cooke and Rattagool 1985, Saisithi 1987, Ijong and Ohta 1995) In European countries, fermented fish products are usually made from anchovies, herring or squid.
Another variant of salted fish products from Southeast Asia is boiled-salted fish. Fish are arranged in bamboo baskets, and cooked in boiling saturated brine for
24 15 min (in Singapore) to 3-4 h (Indonesia) (Chng, Kuang and Miwa 1991). In
Indonesia, the products are called pindang and are produced by burying the fish in salt
in a metal or earthenware container followed by heating for 3-5 h. The species usually used for pindang are mackerel, sardines, milkfish and scad. The products have a short shelf life at Southeast Asian ambient temperature, ranging from 3 to 7 days.
Drying of salted fish is usually done in the sun. In Southeast Asian countries such as Indonesia, the salted fish is usually placed on elevated bamboo racks in the sun for 6-8 h depending upon the weather. If the weather does not permit drying, usually the fish is kept in the salting tanks for a prolonged time. This affects product quality, which will also affect the price.
Sun drying, although a cheap and best means to dry fish especially in the tropics, has various disadvantages: the limited drying time, i.e. 6-7 h a day which is often interrupted by rain; the exposure of the products to various contaminating agents such as dust, sand and dirt; and, most importantly, the products are liable to insect infestation, especially by flies.
Insect infestation also causes problems during storage of dried fish. Insects commonly found in dried fish include beetles (Necrobia spp., Dermestes spp.), flies
(Chrysomyia megacephala and Piophila casei) and mites (Lardoglyphus spp and
Suidcisici spp) (Daniel and Etoh 1983, Indriati et al. 1985, Esser 1991, Gopakumar
1995, Madden, Anggawati and Indriati 1995). Some insecticides that have been tested to overcome insect infestation include pirimiphos-methyl (Minawet), deltamethrin (Decis) and cycloprothrin (Esser et al. 1990, Anggawati et al. 1992,
Madden, Anggawati and Indriati 1995). Other causes of losses of dried fish as well as methods to overcome the causes, are reported elsewhere (FAO 1981, Doe and Olley
25 1990), while a model to predict microbial spoilage was proposed by Doe and
Heruwati (1988).
A number of artificial dryers have been marketed or developed as an
alternative to sun drying during rainy season or to overcome the disadvantages of sun
drying, such as solar tent, cabinet, heat pump, oil heated tunnel, agro-waste and geothermal energy dryers (Doe et a/. 1977, Ismail, 1983, Strommen 1983, Pillai,
Prabhu and Balachandran 1986, Bandyopadhyay and Bose 1990, Souness 1990,
Arason and Arnason 1992, Wibowo et al. 1993, Souness and Wibowo 1995).
However, there have been no adoption of these dryers by traditional processors in
Indonesia.
Smoking is another process which is commonly used after salting, although there are also fish which are not salted prior to smoking. Smoking of fish was probably developed incidentally by fishermen in periods of wet or humid weather when they had to use open fires rather than the sun and wind to dry their surplus catch
(Horner 1992a). In the UK, mild-smoked fish was first produced in 1835 (Cutting
1955). Wood-generated smoke, besides preserving the fish, also adds colour and flavour that characterises smoked products. The smoke preservative effect is largely due to phenolic compounds that are active against naturally-occurring fish spoilage microorganisms on fish. In artificial fish smoking equipment, smoke flavours are used to produce flavours indistinguishable from those from traditional processes (Storey
1982, Miler and Sikorski 1990). Methods of preparing dried and smoked aquatic foods from different parts of the world, as well as their chemical composition, have been compiled (Anon. 1988), while the effects of processing on the nutritive value of such products have been well studied (Cutting 1962b, Burt 1988).
Canning of fish was started nearly as early as the development of this method
26 by Nicolas Appert in the late 18th/early 19th century using wide mouth bottles as food containers (Cutting 1955). Although fish canning is now well advanced, its principles remain the same, i.e. to destroy the microorganisms in fish of public health importance as well as those capable of causing spoilage of the packaged products by sufficient heating and to prevent recontamination by hermetic sealing. Heat treatment in canning changes the characteristics of the fish, and also changes fish components whose natures may be modified by various pre-treatment or the addition of substances such as sauce, oil or brine. Many types of containers, including glass jars, flexible containers, rigid metal containers and rigid plastic containers are now used for heat processing (Horner 1992b).
In 1993, 13% of the total world catch of more than 100 M t was converted into canned products (Anon. 1995b). Aquatic foods which are usually canned include small pelagics such as sardines and anchovies, tuna, crustaceans and molluscs.
Fish products such as fish pastes, fish spreads and fish roes are usually packed in glass jars.
2.2.2 Utilisation as feed
Fish is usually converted into meal before being incorporated into animal diets.
Fish silage, usually dried, is another form in which fish is added to feeds. The use of fish as a feed component is popular especially due to its high nutritional value. It offers high levels of amino acids such as lysine, which is often deficient in grain products that are the typical base for most animal feeds, as well as vitamins and minerals (Windsor and Barlow 1981, Ockerman 1992). Generally, 1 kg fish meal contains at least 535 g digestible proteins, compared to hydrolysed yeast that contains 401 g, sunflower oil cake 378 g, and soybean flour 360 g (Kulikov 1978).
The proximate composition of fish meal varies depending upon the raw
27 materials used as well as the process. However, typically the protein content is in the range of 66-72%, fat 4-6%, ash 13-21% and moisture 9-10% (Windsor 1982).
Raw materials used for producing fish meal are usually those unfit for human consumption or under-utilised fish such as sardines, herring and anchovies.
Traditionally, fish meal as a protein source is used in the production of swine feed and poultry feed, however, fish meal can also be used in rations for fish, cattle, sheep, mink and pet foods. Almost 60% of world fish meal production is now used in poultry feeds, 20% in swine foods, 10% in aquaculture feeds and 10% by the pet food industry (Hardy 1992). In 1993, more than 6 M t of fish was converted into meals, solubles and animal feeding-stuffs, with Chile and Peru as the main producers (Anon.
1995b). Fish solubles are a fisheries by-product containing predominantly water- soluble substances especially protein, usually in condensed solution (Lassen 1965).
2.2.3 Others
Fish also has been used for fertiliser (Cutting 1955). Nowadays, fish fertilisers are produced from either solid fish waste by methods such as composting, mealing, solubilisation and hydrolysis (Brinton 1994, Wyatt and McGourty 1990), or from liquid fish waste (Ockerman 1992). Small pelagic fish such as sardines and farmed fish such as milk fish are also used as baits in tuna fishing. These fish are used frozen and should be in prime condition, otherwise they decompose in water.
Other non-food and non-feed products from fish include pharmaceutical products such as fish oils and n-3 concentrates, adrenalin, insulin, liver oil and squalene, pearl essence, leather, fish gelatin and fish glue. However, the raw materials used for these products are fish processing wastes, and whole fish are seldom used, except fish oils which are sometimes produced from underutilised high oil content pelagic fish.
28 2.3 Aquatic food processing waste 2.3.1 Production and characteristics
In the processing of aquatic food products, especially where only the meat is
used, a large amount of waste is generated. Unutilised fish including by-catch and
underutilised fish are also considered waste. Approximately 30% of total landings is
considered as underutilised, by-catch, unconventional or unexploited resources
(Venugopal and Shahidi 1995). Aquatic food industries produce solid waste which include bones, shells, viscera, heads, eggs/roes, skin and meat trims, and liquid wastes which are usually liquid solutions and/or suspensions of water and portions of the solid waste.
The amount of solid aquatic food processing waste varies greatly depending upon the species and the type of processing, from almost nothing for fish meal processing to as high as 86% for shellfish such as crabs (Carawan, Chambers and Zall
1979, Shahidi 1994). The amount of waste in a processing area can be significant, especially if there is more than one processor. In Alaska, the fishery industry produces almost 0.5 M t waste per year, which is equivalent to 6.8% of world fish meal production (Ellis 1990). The solid processing wastes are rich in nutrients similar in proportion to those of the main products. The proximate composition of fish processing waste is crude protein 13.4-23.2%, lipid 0.7-7.8%, moisture, 70.8-81.2% and ash 1.2-6.5% (wwb) (Crawford, Law and Babbit 1972, Shahidi 1994).
The dominant solid waste from crustacean processing is shell or exoskeleton to which a significant amount of unrecovered flesh and visceral material is often attached. The solid waste from crustacean processing contains 10.7-27.2% protein,
30-57.5% chitin and 15.3-57.9 calcium carbonate (wwb) (Carawan et al. 1979).
Clean water is needed in aquatic food processing to wash, thaw, transport and
29 cook the products. Consequently some foreign matter or waste materials may be present and will pollute the original water supply. Wastewater usually contains blood, fats/oils, small pieces of flesh, visceral materials, skeletons/frames and skin.
Important pollutant parameters in liquid waste are biochemical oxygen demand
(BOD), chemical oxygen demand (COD), total suspended solids (TSS), fats, oil and grease (FOG) and the physical characteristics of the liquid such as temperature and pH. In addition, volume of water, which also correlates with the amount of pollutants, is also important.
Waste in tuna canning is produced in almost every processing step and especially from thawing, dressing, and pre-cooking steps. Pre-cooking produces the highest levels of BOD and COD, approximately 60% and 20% of total BOD and
COD in waste water (Prasertsan, Jung and Buckle 1994). Industrial fish such as menhaden are usually completely rendered, thus the waste volumes are small.
Wastewater from fish meal, fish solubles and fish oil processing can be divided into two categories: high-volume, low strength wastes and low-volume, high-strength wastes. The first consists of the water used for unloading, transporting and handling in addition to wash-down water, while the latter usually include the stickwaters that typically have high BOD values, ranging from 56,000 to 112,000 mg/L, with average solids concentrations, mainly proteinous, up to 6% (Carawan et al. 1979). Typical wastewater characteristics of some aquatic food processing industries are shown in
Table 2.4.
2.3.2 Waste treatment
Aquatic food processing waste contains high amounts of organic matter which, if not carefully handled, might produce environmental problems. There are three general alternatives in handling the waste: treat the material strictly as a waste
30 and dispose of it at least cost; treat the waste as a raw material from which a by product can be obtained; and treat the waste as a combination of both.
Table 2.4 Raw wastewater characteristics for aquatic food processing industries Flow rate BOD5 COD TSS FOG Seafood______(m3/day)_____ (mg/L)______(mg/L)_____ (mg/L) (mg/L)
Abalone2) 38-53 430-580 800-1000 200-300 250 Catfish0 116 340 700 400 200 Crab21 76-2650 600-4400 1000-6300 330-620 220 Fish meal2) 144-350 91-380 160-570 76-266 19-76 Oysters2’ 53-1216 250-800 500-2000 200-2000 22-30 Salmon2’ 220-1900 253-2600 300-5500 120-1400 20-550 Sardines2’ 304 1300 2500 920 250 Shrimp breaded)2’ 653 720 1200 800 n.r. Shrimp (canned)1’ 472 2000 3300 900 700 Shrimp (frozen)1’ 1170 1800 3700 2900 230 Tuna (USA)3 3060 710 1900 550 320 Tuna (Thailand)'^ n.r. 118744 46955 6259 2822 n.r.: no record 11 EPA (1974) Caravvan et al. (1979) 3' Prasertsan et al. (1994) 41 Ultimate BOD
Solid fish waste is seldom treated before being disposed of, either into surface waters including the ocean or on to land (landfill). However, these practices are becoming more difficult due to environmental problems. It is then necessary to utilise the solid waste as a raw material for other products.
Treatments for liquid waste involve several major steps including primary and secondary treatments. Liquid waste typically contains large amounts of insoluble suspended materials which can be removed from the waste by physical and chemical means. Primary treatments are applied for this purpose, and common treatment methods include screening and sedimentation. Screening removes solid particles from the waste flow while readily settleable coarse material can be removed in a sedimentation unit. This unit consists of a scraper to remove sludge, a sludge hopper
31 and a surface skimmer for removing oils and greases that float on the surface. A
sedimentation unit can also be used in secondary treatment to remove dissolved and colloidal particles which are not readily settleable, by addition of a coagulant such as alum, ferric chloride, lime and polymers.
Secondary treatment consists of unit operations which are based on physical, chemical or biological treatments. Dissolved air flotation is a unit operation which physically removes dissolved material and fats, grease and oils from the liquid waste stream with the aid of very small air bubbles. The air is dissolved in the waste as it flows through a pressure tank at approximately 200-350 kPa gauge. If necessary, this treatment can be combined with chemical treatment by adding coagulants to increase the particle size of the solids.
Biological treatment basically converts nitrogenous compound in the wastewater into nitrate through aerobic process. In the process, the nitrogenous compounds, after being converted into ammonium nitrogen, is utilised as an energy source by autotrophic bacteria of the genus Nilrosomoncis producing nitrite which is then converted into nitrate by bacteria of the genus Nitrobacter (Wheaton and
Lawson 1985). There are many biological treatment units which are commonly used such as activated sludge, extended aeration, aerated lagoons and trickling filters. To achieve the purposes of the treatment, it is important to properly control the process to ensure that aerobic conditions are maintained. Table 2.5 summarises the treatments commonly used in the aquatic food industries and the reduction of waste strength achieved.
32 Table 2.5. Treatments of liquid wastes from aquatic food processing
Treatment system Use Effluent reduction (%)
Dissolved air flotation without Primary treatment or by- Grease 60, BODs 30, TSS 30 flocculant aids1 product recovery
Dissolved air flotation with pH Primary' treatment or by- Grease 95-99, BOD5 50-65, control and flocculant unit1 product recovery TSS 60-97, COD 50-75
Activated sludge1 Secondary treatment BODs 90-95
Extended aeration1 Secondary treatment BOD5 95-97
Aerated lagoon1 Secondary treatment BOD5 90-95
Trickling filter1 Secondary' treatment BOD5 80-85
Anaerobic reactors2 Secondary treatment COD 40-80
1) Carawan et al. (1979) 2) Prasertsan et al (1994)
2.3.3 Waste utilisation 2.3.3.1 Fin-fish waste 2.3.3.1.1 Fish meal Fish meal is the most common and the most important product from fin-fish processing waste and is mainly used for animal feed. There have been some efforts to utilise fish meal for human consumption such as in South Africa in the 1930s where it was used as a food supplement for low income populations (Finch 1977). In addition to processing waste, certain species such as some pelagic fish, which are unacceptable for human consumption, as well as by-catch from shrimp fishery, are used as raw materials for fish meal. This has made fish meal an important commodity produced from fishing resources since its world production averages between 6 and 8
Mt per year (Hardy 1992).
Fish meal processing involves cooking, pressing, drying and grinding fish. The liquid of the pressing operation is rich in oil and soluble protein. The oil is separated
33 as another by-product, while the liquid or the stickwater is evaporated to produce protein solubles which can be returned into the processing line of fish meal before drying. The utilisation of the stickwater will save about 10-16% protein from being wasted (Etoh 1982).
Bio-fish flour is a product of biotechnological fractionation of fish protein developed by Japanese workers (Uchida et a/. 1990). The process is basically an improvement of fish meal in which proteolytic enzymes are added to hydrolyse protein and mild drying (75°C maximum) is applied. Hydrolysis replaces the cooking step in fish meal production, thus avoiding the loss of nutrients due to heating. The proximate composition of bio-fish flour is: crude protein 65-70% (minimum), crude fat 12% (maximum) and crude ash 12-16% (maximum). The protein digestibility of the bio-fish flour by swine is 95.1-95.7%, higher than that of conventional fish meal, which is 87-91%. In Japan, bio-fish flour has been commercially produced and used for feed in fish feeds, milk replacements for swine and calves, and a growth factor for
Lactobacillus.
2.3.3.1.2 Minced fish and surimi
Minced fish is produced by passing fish over a drum or meat-bone separator having small (1-7 mm diam.) perforations. By applying pressure to the fish, the soft meat portion is forced to pass through the perforations leaving the bones, skin, and scales on the exterior of the drum. The mince is frozen in blocks, and is used in fish products such as fish cakes and fish sticks and as a substitute for meat (Ingham 1991).
Minced fish can be further processed into surimi, defined as “mechanically- deboned fish flesh extracted with water and mixed with cryoprotectants for good frozen shelf-life” (Lee 1984) or “a myofibrillar protein concentrate produced by repeated washing of fish mince to remove water-soluble and odour-bearing
34 enzymes, sarcoplasmic protein, blood, inorganic salts, and some lipids” (Venugopal
and Shahidi 1995). Excessive washing in this process removes water-soluble
sarcoplasmic muscle proteins, enzymes, pigments/blood, lipids and haem compounds
(Hall and Ahmad 1994). Washing in surimi processing results in almost 60% of the total effluent water and wastes about 30-40% of the total protein (Pedersen 1990,
Jaouen and Quemeneur 1994).
Minced fish and surimi are usually used for the production of various homogenous protein gel food products such as kamaboko (if steamed), satsumaage
(if fried) and chikuwa (if broiled), fish sausages, fish ham, fish sticks and fish analogues (Suzuki 1981, Mackie 1994). Over 60 species, including under-utilised ones such as sharks, and fresh water fish such as milk fish and tilapia, have been used to produce minced fish and surimi, either on commercial scale or laboratory studies, with Alaska pollack being the main species (Suzuki 1981, Pan 1990a, Hall and Ahmad
1994).
2.3.3.1.3 Fish protein preparations
Fish protein preparations used for human consumption include fish protein concentrate (FPC) and creamy fish protein (CFP). There are two types of FPC: type
B and type A. FPC type B is a food grade fish meal which is produced by improved hygienic methods and is also called non-extracted-FPC. Some workers, such as Finch
(1977) and Mackie (1983), included crude fish meal as FPC type C. FPC type A is a more purified form than type B in which the fat content is much reduced, while the protein is extracted using a specific solvent. While FPC type B has a specific fishy flavour and odour with a protein content of 70-75% and fat content up to 10%, FPC type A is a colourless, tasteless and odourless product having a protein content at least 67.5% and up to 0.75% lipid (Windsor and Barlow 1981, Wheaton and Lawson
35 1985, Shahidi 1994).
Another variant of FPC is fish protein hydrolysate (FPH) which is produced using added proteolytic enzymes from vegetable or animal sources as processing aids.
However, there have been a number of efforts to use microbial proteases (Venugopal
1994). The hydrolysis of protein in the process should be carefully controlled for extensive hydrolysis may lead to the formation of bitter taste peptides. This has been attributed to the formation of bulky hydrophobic groups towards the C-terminal end
(Kilara 1985, Roy 1992). If sufficiently dried, FPH is quite similar to FPC but the production cost is lower. Dried FPH from cod filleting waste had a moisture content of 4.8%, crude protein 81.8%, crude fat 4.2% and ash 6.9% (Mackie 1994). Owing to the bitter taste, residual fish taints, brown colour as well as the solubility, the use of
FPH as a human food component requires careful consideration. FPC, especially type B, may be suitable for people accustomed to fish-containing diets, such as those in Africa and Asia, or can be added to some highly spiced food without any significant difference in flavour and odour (Wheaton and Lawson 1985, Ockerman and Hansen
1988).
Fish silage is another type of FPH, but no proteolytic enzymes are added. It may be defined as a liquid product made from fish or parts of fish and acid, or sometimes a base such as sodium hydroxide. The hydrolysis, which leads to liquefaction, results from the action of endogenous enzymes naturally present in the fish which are accelerated by the low pH environment. Acids, either mineral or organic, are added to reduce the pH to about 4.0. Alternatively, acids can also be produced in the process by addition of fermentable sugar and lactic acid bacteria. In addition to accelerating the enzyme reaction, the acids also act as preservatives. The process usually takes up to 40 days at ambient temperature of 15°C, while in the
36 tropics, the process time is shortened to 21 days (Kompiang 1990, Hardy 1992).
Creamy fish protein (CFP) is a relatively new product developed by Japanese workers. The process reported by Shoji (1990) is similar to those used to produce
FPH, only the hydrolysis of the protein is carefully controlled to such an extent that it produces emulsified broken down protein which will not coagulate when heated. CFP can be produced from any type of fish, including shellfish, of which the product from fish has a protein content of 14.1-16.9% and fat 0.2-10.8%, while those from snow crab have a protein content of 8.7% and fat 0.2%. CFP can be mixed with surimi to form many new types of products, and also a variety of new products such as fish tofu
(Shoji 1990).
2.3.3.1.4 Pharmaceutical products
Some pharmaceutical products such as liver oil, n-3 fatty acids, insulin, adrenalin, nucleic acids, protamin, bile salts, vitamin preparations can be obtained from fin-fish waste (Brody 1965, Kulikov 1978, Tanikawa, Motohiro and Akiba
1985, Wheaton and Lawson 1985). A tumor inhibitor has also been discovered in the cartilage of shark (Lee and Langer 1984).
n-3 fatty acids have been a subject of studies for decades especially in relation to their beneficial effects on heart disease (Stansby 1990b). This has led to the commercial production of n-3 fatty acid concentrates. Processes used in the concentration of n-3 fatty acids include esterification and molecular distillation, urea fractionation, and supercritical CO2 extraction (Hardy 1992).
2.3.3.1.5 Fertiliser and compost
Trash fish or fish waste such as offal can be converted to fertiliser by several methods, one of which is by treating the waste with sulphuric acid to digest the tissue.
37 The digestion converts the protein to ammonium sulphate and makes the bone sulphate available for plant absorption and at the same time reduces the fish odour.
Another method to solubilise the protein is by treating the waste with urea. Both methods result in protein-rich liquids which can be applied directly to the soil.
Effluents from fish processing plants are also rich in protein and thus can also be used as fertiliser in a similar manner. Since the protein in the liquids is still in the organic form, it should undergo further hydrolysis by soil bacteria before it becomes available to the plant, thus these proteins are absorbed by the plants at a slower rate than are inorganic fertilisers. Fertilisers made from fish waste contain approximately
3.5% nitrogen, 1.3% phosphorus and 0.3% potassium (Hardy 1992).
Fish waste, including fish-farm mortalities and seastar, can also be composted with carbon sources such as peat, sawdust or wood waste (Mathur et al. 1988, Hardy
1992, Brinton 1994, Liao, Vizcarra and Lo 1994, Line 1994). During composting, biodegradable solid organic matter such as that from fish waste are converted into a stable humus-like substance by a biologically controlled process (Golueke 1981).
During the process the temperature of the mixture increases to 55°C, at which temperature some pathogens are destroyed (Crawford 1985, Brinton 1994). The compost can be used as a soil conditioner or fertiliser. Compost from fish offal and peat has also been used as a substrate for production of mushroom, P leu tains ostreatus, and as a fermentation substrate for an acid-tolerant fungus, Scytalidiwn acidophihmi (Martin and Chintalapati 1989, 1990).
2.3.3.1.6 Others
Other by-products from fin-fish processing waste have also been investigated.
These include antifreeze proteins and antifreeze glycoproteins from blood of several species of polar and north temperate ocean fish hemoproteins, feed and food
38 flavourings and microbial growth media (peptones) (In 1990, Almas 1990, Strom and
Raa 1993, Shahidi 1994).
2.3.3.2 Crustacean and molluscan wastes 2.3.3.2.1 Animal feeds
Waste of crustaceans such as shrimp, crawfish and lobster consists mainly of heads, shell and viscera and is low in fat. These materials can be converted into meal involving dehydration of the waste, grinding, blending and packaging for shipment
(Wheaton and Lawson 1985). Dried crustacean meal contains 15-45% crude protein, less than 5% crude fat and up to 33% ash, mainly calcium carbonate (Hardy 1992).
The amino acid profile of crustacean (shrimp) meal is similar to that of casein
(Simpson 1978).
Crustacean meals have been used as a protein supplement for chicken, cattle and pig feeds. Crustacean meals also contain carotenoid pigments, mainly astaxanthin and its esters, thus when fed to salmonid fish produce a desirable red colouring in the fish flesh. However the astaxanthin content of the meal varies greatly depending upon the temperature used in drying, ranging from absent to 200 mg/g (Torrissen, Hardy and Shearer 1989, Hansen and Illanes 1994).
Crustacean waste can also be converted into silage before incorporated into feeds (Wheaton and Lawson 1985). Due to its high mineral content, it requires considerable acids for digestion. Ariyani and Buckle (1991) found that a mixture of formic and propionic acids (1:1, v/v) at a level of 8% produced stable shrimp head silage. However, the economic and marketing problems associated with crustacean silage are relatively unexplored (Wheaton and Lawson 1985).
Molluscan meals can be produced from squid and shellfish processing waste.
The meals are much more desirable than crustacean meals since shells are absent, and
39 thus they have a lower content of ash and chitin. In processing, the raw materials are placed in a steam-jacketed vessel that both cooks and dries. Molluscan meals usually contain 70% protein, 8% fat and 8% ash (Hardy 1992).
2.3.3.2.2 Chitin and chitosan
Chitin is the second most abundant natural biopolymer after cellulose and is widely distributed in fungi, insects and seafoods. Chitosan is a derivative from chitin, produced by deacetylation of chitin. The principal raw material for chitin production is crustacean wastes, such as shells from shrimp, crab, lobster, prawn and crawfish, which contain 14-35% chitin on a dry basis (Simpson, Gagne and Simpson 1994). It is estimated that the global annual production of crustacean waste currently reaches
1.44 M t on a dry basis (Knorr 1991). Potentially, about 200,000 t of chitin could be produced from such waste annually. At present, about 2,000 t of chitin and chitosan are produced annually (Shahidi and Synowiecki 1992), with Japan as the major user of up to 500 t (Hardy 1992). Chitin and chitosan have many uses such as for medical and pharmaceutical applications, cosmetics, food industry, animal feeds, wastewater treatment, biotechnological applications, industrial applications such as ion exchange resins, a binder for dyes, fabrics and adhesives, and a sizing and strengthening agent for paper (Wheaton and Lawson 1985, Sandford 1989, Skaugrud and Sargent 1990,
Simpson el al. 1994) .
The extraction of chitin includes steps such as grinding, deproteination, demineralisation, drying and final grinding (Wheaton and Lawson 1985, Sandford
1989, Ockerman 1994). Chitosan is produced by treating chitin with strong (40-50%) alkali such as sodium hydroxide or potassium hydroxide at high temperature (150°C) for 4 h to hydrolyse the N-acetyl linkages (Rha 1984, Sandford 1989, Shahidi and
Synowiecki 1992). Typical chitin from shrimp contains 2.5% water, 6.2% (dry basis)
40 nitrogen and 0.1% (db) ash (Shahidi and Synowiecki 1992).
2.3.3.2.3 Others
Blue crab processing wastes can be aerobically composted with the addition of a carbon source such as grain straw. Ferrous sulphate or elemental sulphur is also added to control the pH. The product of composting is a stable peat-like product with essentially no odour and can be used as a soil amendment, a potting soil or mulch
(Wheaton and Lawson 1985).
Shrimp waste to a minor extent has been used to produce fermented products such as shrimp paste, or second grade shrimp crackers (Suparno and Poernomo
1991). Food flavourings could also be produced from shrimp, clam and crab processing waste. The preparation may vary, but generally involves grinding, enzymatic hydrolysis, heating to inactive enzymes, sifting, centrifugation, filtration and concentration (In 1990, Pan 1990b).
Narkviroj and Buckle (1987) investigated the use of prawn head powder in the production of oriental prawn crackers. Prawn head powder at levels of 6, 10 and
12% (w/w) was blended with tapioca flour which was used in the making of prawn crackers. Crackers with 10% prawn head powder were the most acceptable.
Shahidi and Synowiecki (1991) isolated flavour-active ingredients from shrimp
(Pcmda/us borealis) and crab (Chinoecles opilio) processing waste by water extraction. A yield of 1.0-1.5% (w/w) of the total protein was obtained.
Flavourants from crustacean and molluscan wastes have been commercially produced (In 1990) and can be used in products such as surimi-based products, shrimp/fish feed attractants and fermented products from fish/shrimp and baits (Pan
1990b).
Microbiological media have been prepared from crustacean wastes. Stephens
41 et al. (1976) extracted peptones from shrimp heads and hulls after autolytic digestion at pH 7.0, 45°C for 2 h. The digest was filtered and freeze dried. Sophanodora and
Buckle (1988) used proteolytic enzymes (pepsin and papain), at level of 0.5% (w/w) to increase the yield of protein digest of prawn head. The chemical composition of the product was similar or better than that of autolytic digestion. Evaluation of the protein digest as microbiological media revealed that the digest was able to support the growth of the test microorganisms including bacteria (e g. E. coli, S. aureus and
B. subti/is), fungi (e.g. R. oligosporus, and A. oryzae) and yeast (e.g. S. cerevisiae).
2.3.3.3 Liquid Wastes
Seafood processing liquid wastes contain various nutrients which can be utilised as by-products. Surimi and fish meal production and fish canning produce significant amounts of liquid waste. In the processing of surimi, a large amount of water is used to wash out the sarcoplasmic proteins, representing 60% of the total effluent water (Jaouen and Quemeneur 1994) and it is estimated that about 30-40% of the total protein is lost (Pedersen 1990). In Chile, a major world fish meal producer, about 30-60 M m3 waste water containing blood, tissues and fish exudates is thrown away (Marti et al. 1994).
In the canning of fish, the liquid waste is produced from a range of processing steps, i.e. cooking of raw materials, pressing of cooked fish, and discard of viscera, tails and heads. The first major liquid waste produced is cook water, while removal of oil by centrifugation gives press water, and the mixture is stickwater. Stickwater usually contains 4-7% solids, mainly insoluble protein (0.8-1.3%), soluble proteins
(3-5%) and oil (1%), and is used as fertiliser or animal feeds after pH adjustment to
4-5 and evaporation to 50% fish solubles (Gillies 1975, Ockerman and Hansen 1988).
The washing of meat shucked from the shells of molluscs such as clams and
42 oysters removes some of the meat and although the solids content in the liquid is only
0.5-1%, it has a distinct clam-like or oyster-like flavour. Shiau and Chai (1990) utilised oyster shucking liquid in the formulation of soup. Washing water from clam processing also have been utilised to produce clam juice (Hood, Zall and Conway
1976, Hood and Zall 1980). Studies on the use of mussel processing waste water as a fermentation substrate have been reported in Spain (Murado, Gonzalez and Pasatrana
1994).
Protein loss in surimi processing is considerably high and methods to recover the proteins from surimi waste water have been attempted by many workers, especially in Japan. Membrane filtration or ultrafiltration is the most common technique investigated. Miyata (1984) reported that centrifugation as a pretreatment and ultrafiltration as the main treatment recovered 90% of protein in the washwater of red-meat fish. Lin, Park and Morrissey (1995) demonstrated the use of micro- and ultra-filtration to recover 1.7 kg protein from surimi wastewater for every 100 kg surimi produced. . -
Ninomiya et al. (1985) demonstrated that ultrafiltration recovered 90% of protein in waste water of surimi processing using 5 different fish species. Protein from surimi processing waste water was also recovered by pH adjustment of the effluent followed by heating (Niki et al. 1985).
2.4 Viscera and its utilisation
The viscera is an important digestive organ in aquatic organisms, functioning as a grinder and liquefier of ingested foods. It contains a number of enzymes which help the decomposition and absorption of foods and these are produced by the stomach wall or are secreted by the pancreas. The high enzyme level in the viscera makes it very susceptible to autolysis. Autolysis is one of the first processes that
43 takes place when an aquatic organism dies. In this process, the visceral enzymes
digest food components in the viscera. In a feedy aquatic organism, the autolysis occurs at a higher rate since large quantities of enzymes are produced. In the preservation of aquatic organisms, it is important to remove the viscera as soon as possible so that autolysis is retarded. However, visceral enzymes are partly responsible for flavour development during curing and fermentation of aquatic food products.
The proportion of viscera to body weight varies from species to species ranging from 2.5-6.0% for mackerel to 18.0-22.0% for stingray (Kleimenov 1983).
Tropical freshwater fish have from 5% to 11% (w/w) viscera (Ahmed and
Mahendrakar 1995). Sukarsa (1978a) showed that the weight percentage of viscera of marine fish species in Indonesia (9 species, 30-40 fish in each species, 40.4-182.8 g weight as samples) averaged 6%, and generally the weight of the viscera increased with the weight of the fish. The weight proportion of viscera of cod and saithe reached about 7% of the total weight (Gildberg and Almas 1986). Within species, the weight proportion of viscera also varies, according to age and sex. Dunham et al. (1985) reported that older Channel catfish had a lower weight proportion of viscera. The young male catfish also tended to have a lower proportion than the female, which was related to egg development in the female. However, as the fish aged, the difference between male and female fish disappeared.
A typical proximate composition of tuna viscera is 29.2% protein, 4.1% fat,
1.4% ash and 65.3% water, which is similar to that of the flesh, i.e. 26.4% protein,
4.7% fat, 1.2% ash and 67.5% water (Vlieg and Murray 1988). In developed countries, viscera has been considered as a waste and discarded without utilisation.
However, viscera has been utilised in countries such as Norway, freezing in blocks
44 for sale to fur animal farmers, and efforts to make full use of this resource have long been initiated (Raa, Gildberg and Strom 1983, Gildberg and Almas 1986). In developing countries, the viscera is occasionally utilised for human consumption, e g. many fermented products in Asia, as well as other cured products such as dried salted fish are consumed with intact viscera.
Although viscera generally has good nutritive value, partly spoiled viscera may cause health problems to feeding animals because of toxic products resulting from the spoilage process (Disney and Hoffman 1976, Gildberg and Almas 1986). Kompiang,
Arifiidin and Raa (1980) reported that toxic products were present in the press-liquid from 24h spoiled fish. Some studies have also shown that viscera of some aquatic organisms may contain toxins produced by marine algae or marine protozoa consumed by the fish.
Water fowl died in Monterey Bay, California, USA after eating anchovies whose viscera were contaminated with toxic domoic acid at levels up to 485 mg/kg
(Wekell et cil. 1994). Two persons in Hamamatsu, Japan were reported to be intoxicated by eating a boiled grouper (Epinephelus auvnici) whose viscera contained ciguatoxin, a toxin associated with some algae, especially Gambierdiscus toxicus
(Ragelis 1984, Hokama et al. 1993). Ciguatoxin was found in the viscera of parrotfish in Japan (Yasumoto, Nakajima and Chungue 1977). Other fish that may contain ciguatoxin are barracuda and snapper (Hall 1991, Haard, Simpson and Pan
1994), and chinaman fish (Symphorus nematophorus) and red bass (Lutjanus bohar) in Australia (Hutchins 1980). Although only a small proportion of viscera of aquatic organisms might present hazards, direct utilisation of viscera seems inappropriate and conversion of viscera into other products is necessary to minimise hazards.
Since viscera is susceptible to autolysis which leads to early onset of spoilage,
45 preservation is necessary before any effort is taken for its utilisation. One preservation method that is suitable for viscera is conversion to silage. Production of viscera silage is similar to that of fish silage. The autolysis of viscera silage is fast due to the digestive enzymes, especially if the temperature is in the range of 25-40°C.
Within one week, the protein of the stomach liquefies and is ready for further utilisation. Lipids are easily separated from the liquid products by simple centrifugation.
The incorporation of viscera silage into animal feeds has been investigated by several workers and generally, showed good results (Johnsen and Skrede 1981,
Wilson, Freeman and Poe 1984, Krogdahl 1985a, b, Ariyani, Murdinah and Putro
1986, Gildberg and Almas 1986, Basmal et al. 1988, Myer et al. 1990). Proteolytic enzymes from liquid fish viscera silage have also been extracted (Reece 1988), while as a growth substrate for microorganisms it was similar to or better than commercial peptone (Clausen, Gildberg and Raa 1985).
Upon removal of fat, viscera can be processed into meal and has a high nutritive value. Kimumaki and Arai (1987) assayed the nutritive value of meals from great blue shark {Prionake glaucus) viscera using rats, and compared this product with casein. The protein efficiency ratio of meals from shark viscera were in the range of 78-98 compared with casein.
An improved method to produce meal from viscera was investigated by Lee and Woo (1992). Skipjack tuna viscera was firstly autoclaved, and the solid materials separated from the liquid which contained viscera solubles. The liquid portion was concentrated and returned to the solids which were then fermented with Aspergillus oryzae for 72 h. This process has been reported to significantly increase the total free and essential amino acids and the content of vitamins Bj, B2 and C.
46 Joseph, Prabhu and Madhavan (1987) investigated the processing of squid
waste, which included head, tentacles, viscera and skin, into meal. The processing
method involved 2 min boiling in 2% salt solution and 2% alum, draining and drying.
Blanching in alum reduced the yield but also reduced the drying time and the level of volatile bases, while the colour was improved.
Fermented fish products in Asia are generally made from whole fish, and the presence of viscera has proved beneficial to the process. Bekasang, a fermented sauce-like fish product from Eastern Indonesia, is usually made from small whole
sardines with the addition of salt to between 30 to 60%. The mixture is packed in glass bottles and stored for 3-6 weeks (Ijong and Ohta 1995). Good quality bekasang
can also be produced from the viscera of skipjack tuna with the addition of up to
20% salt (Suparno and Poernomo 1991). In Korea, more than 30 kinds of fermented fish pastes are available in the market, and almost all include viscera in their process, some are even made from viscera alone, such as hair-tail fish viscera paste that has
12.3% protein, 6.2% lipid and 19.4% salt (Cha and Cadwallader 1995).
2.5 The stingrays (Dasyatididae spp) 2.5.1 General description
According to Nelson (1994) and Weitzman (1994) stingrays indicate any species in the order Myliobatiformes (the stingrays), the subclass Elasmobranchii (the sharks and rays), the class Chondrichtyes (the cartilaginous fish) and the superclass
Gnathosmata (the jawed fish). There are eight families in this order, i.e.
Potamotrygonidae (river rays), Dasyatididae (stingrays), Urolophidae (round stingrays), Gymnuridae (butterfly rays), Hexatrygonidae (sixgill rays), Myliobatididae
(eagle rays), Rhinopterididae (cownose rays) and Mobulidae (Mantas or devil rays).
These species usually have a flattened body from upper to lower with large
47 pectoral fins attached to the side of the head, ventral gill slits and long spikelike or whiplike tails. The width of the body or disc (from "wing" to "wing") varies greatly from species to species, but the largest is the manta or devil ray whose width may reach as large as 6.1 m (Groves and Hunt 1980), and weigh up to 350 kg (Last and
Stevens 1994). Stingrays are usually found in shallow to moderate depths in tropical, subtropical or warm waters. Some are able to change colour slightly and blend in with the bottom, especially in sheltered areas. They are basically demersal fish usually lying on the bottom or partly buried, and feed on both pelagic and benthic organisms such as small fish, molluscs and crustaceans (Devadoss 1978a). Some stingrays, including cowtail stingray, are able to adapt to freshwater conditions (Ahmad and
Sonoda 1979, Taniuchi 1979, 1991).
Stingrays are considered the most important group of venomous fish
(Halstead 1970), and stingrays' attack was recorded as early as 1608 (Russell 1965).
There are three different hazards caused by stingrays, i.e. intoxications due to ingestion of the fish flesh, envenomation by the sting and electric shock, especially by the species Torpilli (De Haro et al. 1994). Envenomation by stingrays is the most likely cause of marine puncture wounds in the US (Auerbach et al. 1994). The venom of stingrays is secreted by the sharp spine located near the base of the tail and is able to cause necrosis of tissue adjacent to the wound. The venom also contains a coagulant which prevents the wound from bleeding (Groves and Hunt 1980). If adequate first aid is not given, severe necrosis may result in prolonged pain, infection and disability. The main species causing envenomation are those from the family of
Dasyatididae (Barss 1984, De Haro 1994). Deaths from stingrays' venom have been also reported, including at least one in Australia (Russell et al. 1958, Grant 1987).
The family Dasyatididae is represented by more than 60 living species from at
48 least 5 genera (Last and Stevens 1994). Some important genera in the family
Dasyatididae are Humatiturus, Pastinachus, Dasyibatus or Trygo n and Amphotistius
(Garman 1913). These classification of genera are not confirmed. The species used in the present study is called by several names: Dasyatis sepheti, Trygon sepheti,
Pastinachus sephen ater and Raja sepheti (Garman 1913, McCulloch 1927,
Sainsbury et ai 1985, Last and Stevens 1994).
Cowtail stingray (D. sepheti) was firstly described by Forskal from the Red
Sea and is also called banana-tail ray and fantail ray (Stead 1963, Grant 1982). It is characterised by a large and uniformly dark disc with a dense band of blunt denticles over the central area (Figure 2.3). The disc width of an adult is at least 180 cm, exceeding 300 cm total length (Last and Steven 1994). It has a very long, slender and anteriorly flattened tail with a posteriorly located sting. The tail can be more than 1.7 times the length of the disc (Sainsbury et al. 1985) with one spine. This species is widely distributed in the Indo-Pacific, especially in the tropical and sub-tropical waters, from the Red Sea to Northern Australia.
2.5.2 Utilisation Cartilaginous fish such as sharks and stingrays are important primary produce in many parts of the world, especially in the oriental world. Annually, more than
700,000 t of sharks, rays and chimaeras are landed, and in 1993 this was 0.7% of the total fish landing (Anon. 1995a). Cartilaginous fish such as sharks are not fully utilised; this is usually confined to the expensive organs such as the fins. Shark fins are specialty products which are considered aphrodisiac. In 1994, dried fins in
Australia fetched prices as high as A$100 per kg (Last and Stevens 1994). In
Indonesia, sharks incidentally caught by tuna fishing fleets are usually thrown back to the ocean after the fins are taken. The fins are then dried on board and are priced
49 about 25-50 times higher than that of shark meat when landed.
In Australia sharks and rays are principally used for food, and about 7000 t are landed annually, however rays contribute only a small amount (Last and Stevens
1994). In some European countries and many parts of Asia, skates and rays are considered important foods. In India, cartilaginous fish contribute about 5% of the total catch, and currently more than 60,000 t of rays are landed (Devados 1978b, Last and Stevens 1994). In Indonesia, about 37,000 t of rays was landed in 1994 (Anon.
1996).
F. Olsen del.
Figure 2.3. Cowtail ray (after Grant 1982)
50 The nutritive value of cartilaginous fish is actually not much different from that of teleosts (bony fish). The protein content of stingrays ranges from 10.5-21.2%, while the fat content ranges from 0.3-2.5% (Licciardello and Ravesi 1988, Nasran et a!. 1992, Ariyani et cil. 1993, Fawzya et al. 1993). Although the fat content of stingrays is not high, it contains polyunsaturated fatty acids which are beneficial to health. Lytle and Lytle (1994) reported that 70% of the total fat in Mexican rays was unsaturated, and 20-33% of this was n-3 fatty acids. Gibsoon, Kneebone and
Kneebone (1984) found that the unsaturated fatty acids in stingrays' fat from tropical waters was about 62%, of which 27% and 39% were n-6 and n-3 fatty acids, respectively.
The consumption of cartilaginous fish, especially sharks and rays or stingrays, is hindered for several reasons. In many parts of the world, some people are not able to overcome the prejudice due to the unpleasant appearance of those creatures.
Stingrays are often described as "the demons of the sea" or "the devil fish" (Russel
1965). The reddish colour of the flesh meat, as opposed to the whitish colour of other fish flesh might be another factor that makes the meat unattractive to consumers. In a storage trial of frozen ray wing meat cuts, Licciardello and Ravesi
(1988) explored several cooking methods; the preferred (taste panel) product was batter-breaded and cooked by frying.
The flavour of the elasmobranchs is also often poor due to the high urea content of the flesh. Unlike other marine fish, the elasmobranchs convert waste protein into urea instead of ammonia which is then used to osmoregulate the tissue in response to changes in environmental salinity (Love 1982, Perlman and Goldstein
1988). Consequently, elasmobranch flesh contains more urea than teleost flesh and this is concentrated not only in the body but also in the blood and kidneys. While the
51 urea content of the flesh of bony fish is in the range of 0.5-50 mg%, it can be as high as 2160 mg% or 2.2% in elasmobranch flesh (Shewan 1951, Haard et al. 1994). The high concentration of urea in the muscle of the elasmobranchs affects the flavour of the meat and is partly the reason why they have a somewhat bitter taste which develops into an ammonia taste during spoilage.
A rare form of intoxication due to the consumption of stingray products has been reported in many parts of the world, especially in tropical countries (De Haro et al. 1994). It was suspected that the main cause was ciguatoxin present in the meat.
De Haro et al. (1994) further reported that in France (Marseilles) and Canada
(Quebec) the most common intoxications reported to the local Poison Centre after eating contaminated stingray meat were gastrointestinal syndromes or, more rarely, moderate neurological syndromes.
In the utilisation of stingray, only the wings (65-80% of the total weight) are used (Nasran et al. 1992). The wings are a higher proportion of the total body mass in larger fish. If they are to be marketed unprocessed, the preferred form is as frozen wing cuts (Licciardello and Ravesi 1988). In Indonesia, the wing meat is usually processed into dried salted meat slices, and small amounts are consumed fresh and smoked (Suparno 1995). Similar products are also found in Korea but there is no detailed information available regarding this product (Shung et al. 1994).
In Indonesia dried salted stingray processing is concentrated in Muara Angke,
Jakarta, Java, and Labuhan Maringgai, Lampung, Southeast Sumatra (Figure 2.4).
The method practiced in Muara Angke (Jakarta, Java) processing area is described by Nasran et al. (1992). Wings are cut from stingrays, washed and then salted in layers of solar salt, with the top layer being salt crystals, in a water-tight wooden basin to which is added water or brine at ambient temperature (28°-32°C) for about
52 30 h. The amount of salt used varies greatly depending upon the freshness of the raw materials. The fresher the fish, the less the amount of salt added. The amount of salt added is in the range of 17% to 39% of the wing weight. The salted wings are then sliced, washed and sun-dried. If the weather is good, the drying time is usually short, about 7-8 h. If the weather is poor, drying is sometimes done for 2-3 days. The products are then packed in bamboo baskets and marketed in rural areas in West Java.
Dried salted stingrays from Muara Angke are thick (3-12 mm) and have a dark colour
(Figure 2.5).
The other method is similar to the above, but the slicing is done before salting.
The slices are much thinner than those produced at Muara Angke. Salting is usually done in saturated brine with addition of salt crystals. The salting time is short, usually overnight or 10 -12 h, while drying time is about 4-5 h when the weather is good.
This method is practiced by processors in Labuhan Maringgai, Lampung, Southeast
Sumatra. The products are packed in plastic or rattan bags and marketed especially to
Jakarta and Medan, North Sumatra. Products from Labuhan Maringgai are white to yellowish in colour and very thin (3-5 mm) (Figure 2.6). Products from Labuhan
Maringgai command a higher (3-5 times) price than those from Muara Angke,
Jakarta.
Nasran et al. (1992) reported that dried salted stingrays from Muara Angke,
Jakarta, had a proximate composition as follows (ranges for 3 processors): moisture
(wet weight basis) 39 - 48%, protein 23 - 60%, fat 0.8 - 1% and salt 10 - 18%. A similar proximate composition was reported by Kikuchi et al. (1972), i.e. moisture content 24%, protein 58%, fat 1.7% and salt 14%. The proximate composition of dried salted stingrays from Labuhan Maringgai, Lampung is 42% moisture, 35% protein, 0.9% fat and 17% salt (Ariyani et al. 1993). No proximate composition data
53 are available for similar products from Korea, except for non-protein nitrogenous compounds such as dimethylamine, trimethylamine and trimethylamine oxide as reported by Shung et al. (1994).
In an effort to diversify product from stingrays, Ilyas, Nasran and Irianto
(1988) investigated the possibility of processing stingray meat into mince. After being cut from the fish, the wings were cleaned and filleted into 5-7.5 mm thick sections.
The meats were separated from the bones using a meat separator and then minced.
The minced meat was washed using ice-cold water and centrifuged. The washing was repeated 4 times and the proportion of water to fish in each washing was 3:1 (v/w).
After the addition of 0.2% sodium tripolyphosphate, the minced meat was frozen into blocks. The protein content of the mince was in the range of 75 - 83%. This process was able to reduce the urea content of the meat as well as the ammonia flavour.
Studies to make sausage from stingray minced meat have been investigated (Fawzya,
Muljanah and Nasran 1994). The acceptability of such products was high, i.e. 6.5 on a 9-scale evaluation.
2.5.3 Waste products
The processing waste of stingrays consists of skins, heads, backbones and internal organs including viscera. This waste represents from about 20% to 60% of the total body weight (Kleimenov 1983, Ilyas et al. 1988, Nasran el al. 1992). The utilisation of this waste is limited to the skin only which is converted into leather. In
Muara Angke, the meat adhering to the bones is sometimes trimmed and processed into dried salted fish cuts which are priced much less (one third) than the dried salted wing meat (Retnowati et al. 1991).
Stingray skin produces a characteristic leather. This material is obtained from the dorsal skin and weighs about 6% of the total weight (Tambunan, Tazwir and
54 Sabaruddin 1992). Deskinning takes place from tail to the head using an extremely sharp knife. After being washed, the skin is cleaned of adhering meat, since this decreases the absorption of the tanning agents or other preservatives, and thus produces lower quality leather. The upper part of the skin is also cleaned, especially between the pearly structures specific to stingray skin. The skin, if not directly tanned, is preserved either by drying or icing. At ambient tropical temperatures (28-
32°C), the skin can be kept for about only 6-8 h, while icing prolongs the shelf life up to 14 days (Tambunan et al. 1992). Stingray skin is tanned using procedures used also for marine fish such as shark (Hak et al. 1990).
No other waste utilisation methods have been practiced commercially by stingray processors. The utilisation of waste has been investigated by Basmal, Murtini and Indriyati (1995) by ensilation using 3, 4.5 and 6% (by weight) formic acid. The waste used in the investigation was stingray offal which consisted of head, bones and internal organs. The waste was cut into small pieces and acids added, and the mixtures kept at ambient temperature (28-32°C). Formic acid at 3% (by weight) was unable to preserve the waste which spoiled after 21 days. No information was provided on the bones or skulls in the silage, but it would be likely that they would remain intact, since they were unlikely to contain enzymes capable of liquefying these wastes.
55 processed
are
stingrays
where
Jakarta)
and
(Lampung
locations
and
Indonesia
of
Map
2.4. 56 Figure 2.5. Dried salted cowtail ray from Muara Angke Figure 2.6. Dried salted cowtail ray from Labuhan Maringgai 57 3 ENSILATION OF COWTAIL RAY (Trygon sep/ien) VISCERA 3.1 Introduction Fish silage can be defined as a liquid product made from whole or parts of fish as a result of acidification either by added acid or by acid produced through fermentation. Ensilation or ensiling was originally used to preserve and store wet fodder in a silo. The product, which is called a silage, has been associated with acid preserved green forage. The acid used in the process is either added or results from anaerobic fermentation of fermentable carbohydrates to lactic acid by bacteria. This process was initially developed by Artturi I. Virtanen in 1929 in Finland who preserved green fodder with a mixture of sulphuric and hydrochloric acids, which were later called A1V acids (Tatterson and Windsor 1974, Stanton and Yeoh 1977). Work on fish silage was initiated in Sweden in about 1936 using AIV acids, sulphuric acid and molasses, and formic acid. Edin (1940) and Olsson (1942) carried out more work on fish silage and suggested formulae for the acidification of the fish based on their pH, and ash and protein contents. In Scandinavian countries such as Denmark, Finland and Norway, silage from fish waste has been produced since the Second World War (Jensen and Schmidtsdorff 1977, Gildberg and Almas 1986, Strom and Raa 1992). In Southeast Asia this process is relatively unknown, although an almost similar process has been practiced traditionally, such as for producing fish sauce. Ensilation was introduced to Asia and Africa in the 1970s, especially to utilise fish waste, surplus waste and by-catch. Acidification of fish in ensilation is achieved by either direct addition of acids or by lactic acid produced by fermentation of the mixture of fish and substrate (carbohydrates). The acids added to the fish can be inorganic or organic, or a mixture of both types of acids. When using inorganic acids, the pH of the fish-acid mixture 58 should be sufficiently low to ensure preservation. Thus while pH 4.0 is sufficient in the preservation of green forage, for fish the pH should be about 2.0 or even less. The preservation of green forage using inorganic acids is not merely due to the low pH, but also to the sugars present in the forage which contribute to preservation by repressing the activity of deaminating enzymes from surviving bacteria (Raa and Gildberg 1982). Fish, however, not only contain high protein and ash contents resulting in high buffering capacity, but also insufficient sugars for conversion to acids. If organic acids are used, a higher pH of about 4.0-4.5 is sufficient due to their preservative effects. However, at this pH, the silage is not well protected against fungi, growth of which will increase the pH of the silage to the level at which spoiling bacteria are active. It has been reported that Aspergillus flcivus grew on the surface of silage produced using formic acid alone. To prevent fungal growth, the addition of potassium sorbate or propionic acid have been suggested (Raa and Gildberg 1982, Levin et al. 1989, Kompiang 1990). Acids that have been used in the production of fish silage include propionic, formic, hydrochloric, sulphuric, phosphoric and citric, either singly or in a mixture of acids. Preservation of fish by lactic acid from anaerobic bacterial fermentation involves the addition of fermentable sugars such as molasses, whey powder, malt and oat meals, cassava, and sago. During the first stage of fermentation, which is heterofermentative, many bacteria including spoilage bacteria utilise sugars and protein in the fish, producing high acid conditions. This inhibits the growth of bacteria unable to grow at low pH, but will give suitable conditions to acid-tolerant bacteria such as lactic acid bacteria which become predominant. The fermentation at this stage becomes homofermentative producing lactic acid. 59 Fish silage produced by lactic acid bacterial fermentation is relatively hygienic, and it has been reported that coliforms, faecal streptococci and pathogenic organisms such as Salmonella, enterococci, typhoid bacteria, coagulase positive staphylococci and spores of Clostridium botulinum are destroyed (Wirahadikusumah 1968, James, Iyer and Nair 1977, Raa and Gildberg 1982, Raa, Gildberg and Strom 1983). To accelerate the fermentation, it is sometimes advantageous to add suitable bacteria as a starter. As suggested by Whittenbury (1986), the starter to be added must meet several requirements, such as be able to compete with and dominate other organisms likely to occur in the silage, and be homofermentative, acid tolerant and able to ferment sugars. Food containing lactic acid bacteria, such as sauerkraut, or health drinks such as Yakult, has been used as a source of lactic acid bacteria in fish ensilation (Kompiang 1990, Suparno and Poernomo 1991, Zuberi et al. 1992). Fish silage produced using added acid and by lactic acid bacteria fermentation are comparable in stability during storage, but the first has pungent odours, while the latter has pleasant ones (James et al. 1977, Zuberi et al. 1992). Unlike acid- preserved silage, the fermented silage is not widely produced, although India has been reported to produce this type of fish silage (Raa and Gildberg 1982). The use of fermented fish silage on an industrial scale in animal feeds are doubted (Lindgren and Pleje 1983), while no commercial success on this type of product was reported (Strom and Raa 1992). Fish silage has been an object for investigation for decades and much attention has been directed to the application of this process to utilise fish waste, surplus fish or by-catch fish. Many investigations have also been directed to compare fish silage with fish meal. Windsor and Barlow (1981) and Raa and Gildberg (1982) have described some advantages and disadvantages of fish silage compared with fish meal. Some 60 advantages include: 1. Fish silage does not putrefy and retains a fresh acidic smell even after storage for weeks at tropical temperatures. The odour of fish silage is much less a problem than that of fish meal. 2. Fish silage is almost sterile, and pathogens such as Salmonella are killed, while coliforms, enterococci and spores of Clostriium botulinum are destroyed by the acid conditions (Wirahadikusumah 1968, James et at. 1977, Raa and Gildberg 1982, Raa et al. 1983). 3. The production method is easy and simple, and can be performed by unskilled workers. 4. The production as well as the capital costs are lower than those of fish meal, and sophisticated equipment is not needed. The equipment can be as simple as a homemade drum and stirrer. An oil separator may be needed when processing oily fish. The production scale can be varied as necessary without affecting the economy of processing. 5. Silage can be stored in tanks in an open area. There are, however, some disadvantages in the production of fish silage, and suggested solutions, including: * Silage is usually a liquid product and storage, as well as transport is a problem. This can be overcome by drying with an addition of carbohydrate fillers. In the tropics, drying can be done in the sun without attracting flies due to the evaporating acids. Once dried, it is less bulky and easier to transport. * Some nutritive components may be reduced in content during ensiling. Amino acids are generally stable in silage, but it has been reported that some amino acids, such as tryptophan, degrade during the process. The level of vitamin Bi (thiamin) may be also reduced due to the presence of thiaminase in fish. This can be 61 avoided by boiling the silage to deactivate the enzyme. However, most feeding experiments with fish silage show that it has a nutritional value comparable to fish meal. Nutritional deficiency can be compensated by the addition of the deficient components. Fish silage has been reported to be a good protein source, yet it has also been shown that fish silage is inferior to fish meal (Raa et al. 1983, Ockerman and Hansen 1988). These conflicting reports are probably due to the silage quality and quantity used in the experiments (Raa and Gildberg 1982). The raw materials used for ensiling also affect the quality of silage-based feeds in terms of nutritive value. Some studies have shown that different raw material freshness produced silages of different net protein utilisation, or even high mortality rate of the animal if partly spoiled raw material was used (Raa and Gildberg 1982, Gildberg and Almas 1986). Kompiang et al. (1980) reported that toxic products were present in the press-liquid of 24h spoiled fish. Strom et al. (1980) stated that there might be a risk of toxin, e g. that from Clostridium botulinum, if spoiled raw material is used for ensilation. Fat contents of feed higher than 1% dry matter will produce tainted meat, thus if fish silage is used in rations, the oil is preferably removed. Strom and Eggum (1981) found that removing oil improved the net protein utilisation of the silage, from below 60% to 70%. Fish ensilation can also be used as the first step in the utilisation of some valuable materials. Extraction of oil from ensiled shark liver, followed by squalene isolation, has been reported by Putro and Sutijana (1990), while Huei-Mei and Meyers (1983) and Torrissen et al. (1981/1982) found that ensilation improved the stability and extraction of astaxanthin pigment from crustacean waste. Isolation of enzymes and peptones from fish viscera silage has been reported (Clausen et al. 1985, Reece 1988, Vecht-Lifshitz, Almas and Zomer 1990, Almas 1990). Plastein-reactive 62 substances were obtained from silage having pH 1.7-2.0, while the optimum degree of hydrolysis in silage for plastein synthesis was 65-70% which was obtained by 3-4 days ensilation at 30°C (Raghunath and McCurdy 1991). Some studies have indicated that in the ensiling process, different parts of fish will liquefy at different rates. Silage prepared from material containing viscera and head liquefy faster than those without viscera and heads (Tatterson and Windsor 1974, Backhoff 1976, Jayawardena and Poulter 1980), and is particularly due to the high level of proteolytic enzymes contained in those body parts. Thus it is essential to include these parts in order to obtain satisfactory liquefaction during ensilation. Studies on fish or fish viscera silage from temperate waters are well documented. Tropical fish used in silage studies usually include whole fish (Hall et al. 1985, Hall, Ledward and Lawrie 1985, Hall and Ledward 1986), mixtures of fish body parts (Jayawardena and Poulter 1980), or by-catch fish (Kompiang et al. 1980). Studies on tropical fish viscera silage are limited. Recent studies on tropical fish viscera were reported by Mahendrakar et al. (1991) and Ahmed and Mahendrakar (1995). However, they used a mixture of viscera from a mixed variety of fresh water fish (mainly common carp) and included liver, intestines, swim bladder and gonads. The aims of the present study were to investigate the ensilation of tropical cowtail ray (Trygon sephen) viscera and to establish the conditions to produce stable silage. 3.2 Materials 3.2.1 Viscera Cowtail ray viscera were obtained from ray salting and drying processors at Muara Angke, Jakarta for Experiment 1 and from Labuhan Maringgai, Lampung, 63 South East Sumatra for Experiment 2. Cowtail ray landed at Muara Angke are usually caught from eastern Sumatra, and western and southern Kalimantan waters during 20-30 day fishing trips (Nasran et a/. 1992, Suparno 1995), while those landed at Labuhan Maringgai are usually from Lampung waters during 4 day fishing trips (Ariyani el a/. 1993). All are caught with bottom long liner and kept in wooden fish holds with minimal icing. Separation and washing of viscera were done at the processing sites. The viscera were road transported in ice (1:1, w/w) to the laboratory of the Slipi Research Station for Marine Fisheries, Jakarta. The time to transport the materials from Muara Angke was 1-2 h, while from Labuhan Maringgai to Slipi was 7-8 h. Upon arrival, the viscera was once more washed in fresh water, frozen and stored at -45°C until used. The specimens used in these experiments (with liver attached) are shown in Figure 3.1. In these experiments, the liver was removed. Figure 3.1 Typical cowtail ray viscera used in the experiment 64 3.2.2 Chemicals and solvents The chemicals used in the experiments are listed in Table 3.1. 3.2.3 Equipment The equipment used in the experiments are listed in Table 3.2. 3.3 Methods 3.3.1 Preparatory 3.3.1.1 Experiment 1 Ray viscera were chopped (ca. 1 cm x 1 cm) and placed into glass containers (750g). To the chopped viscera was then added a mixture of propionic and formic acids (1:1, v/v) or hydrochloric acid each at concentrations of 3 and 4% (v/w), and mixed thoroughly. Coding for acid treatments used in this study w^as Acidl and Acid2 for propionic and formic acid mixtures at concentrations of 3 and 4%, respectively; and Acid3 and Acid4 for hydrochloric acid at concentrations of 3 and 4%. The mixtures were stored at ambient (28-32°C) and 40°C. Stirring was done daily during the first week. At every predetermined storage time, pH, viscosity, soluble nitrogen (NPN) and liquefaction were measured in triplicate. The experiment was done in duplicate. 3.3.1.2 Experiment 2 Experiment 2 was conducted to confirm results of experiment 1 using materials from different locations. Methods of preparation as well as analyses of silage were similar to those of Experiment 1, except for the type of acids and their concentration. In Experiment 2, the acids used w^ere those producing stable silages in Experiment 1. The mixtures of minced viscera and acids were stored at ambient (28- 32°C) and 40°C. 65 Table 3.1 Chemicals used in the experiments Chemical Brand and specification Ammonium sulphate E. Merck, GR Borid acid E. Merck, GR Bovine serum albumin Sigma Bromocresol green indicator E. Merck, GR Buffer solutions pH 4 and 7 E. Merck Copper sulphate anhydrous E. Merck, GR Ethanol E. Merck, GR, 95% Folin & Ciocalteu’s phenol reagent Sigma Hemoglobin (bovine, lyophilised) Sigma Hydrochloric acid E. Merck, GR Methyl red indicator E. Merck Petroleum ether BDH, AR Potassium sulphate anhydrous E. Merck, GR Sodium hydroxide E. Merck, GR Sodium citrate Univar, AR Sulphuric acid E. Merck, 95-97% Trichloroacetic acid E. Merck, GR Table 3.2 Equipment used in the experiments Equipment Manufacturer and type Centrifuge Beckman Instrument Inc., J2-21 Incubator (oven) Contherm Scientific Co. pH meter 1. Orion Research Inc., Digital Ionalyzer, Model 601A 2. Gallenkamp, pH meter stick Viscometer Brookfield Eng. Lab. Inc., LVT Model To assess the ability of the enzymes of the silage to hydrolyse fish meat, a test was conducted using minced flesh from mackerel {Raslrelliger kanagurta) with proximate composition: moisture 75.1%, protein 23.3% and fat 1.5% (wet weight basis). The method to hydrolyse fish meat was adopted from Liu and Pigott (1981). A crude enzyme preparation was obtained from 5-day old 40°C silage prepared using Acidl. After centrifugation (2000xg-, 10 min), the silage liquid was mixed with ammonium sulphate, and the protein that precipitated between 25-50% saturation was redissolved into 10 mM phosphate buffer pH 7.0 and dialysed overnight against the 66 same buffer solution. After further centrifugation, the supernatant was collected and used as a crude enzyme preparation. To prepare the substrate for hydrolysis, fish fillets (15g) were mixed with one volume of distilled water (15 mL) and homogenised in a Waring blendor, then the crude enzyme added at levels of 5 and 10% (w/w with respect to protein content of both crude enzyme and minced fish). The mixture was adjusted to pH 2.0, 5.0 and 8.0 using 2N HC1 or 2N NaOH and incubated at 45°C and 60°C for 4 h with continuous stirring during which non-soluble nitrogen level was monitored. 3.3.2 Analytical 3.3.2.1 Proximate analysis 3.3.2.1.1 Moisture content Moisture content was measured by the oven method (Chng 1992a). Samples of about 5g were weighed in a tared dish, and dried in an air-forced oven at 105°C for 24h. The weight loss after drying was considered as moisture content. 3.3.2.1.2 Protein content The Kjeldahl method was used to measure the protein nitrogen content of about 2g samples (Kirk and Sawyer 1991). The catalyst used for digestion was a mixture of 9 parts (w/w) of potassium sulphate anhydrous and 1 part (w/w) of copper sulphate anhydrous. The indicator used in the titration of the distillate was a mixture of 1 part (v/v) of 0.1% methyl red in 95% ethanol and 2 parts (v/v) of 0.2% bromocresol green in 95% ethanol. The protein content was obtained by multiplying the protein nitrogen content by a factor of 6.25. Soluble protein in the crude enzyme preparation was measured by the method of Lowry et al. (1951) using the modification of Bailey (1967) in which sodium citrate 67 was used instead of sodium tartrate. Bovine serum albumin was used as the standard protein. The estimation of protein content from spectrophotometer readings was done according to the method of Coakley and James (1978) which proved sufficient to represent the non-linear relationship between absorbance (750 nm) and protein concentration in Lowry’s method, up to 150 mg/mL BSA (Peterson 1979), as follows: P0 = bAo/(l-aAo) where P0 = protein concentration of sample b = (1/A! - l/A2)/( 1/P, - 1/P2) a = A'1 - b P'1 A ’1= the mean of Af1 and A2_1 P1 = the mean of Pf1 and P2_1 Ao = absorbance (750 nm) of sample Ai and A2 = absorbance (750 nm) of protein standards Pi and P2 = concentration of protein standards. 3.3.2.1.3 Fat content Fat was extracted by the Soxhlet apparatus for 16h using petroleum ether (BP 40-60°C) as solvent, and the weight of material in the flask after evaporation of solvent was calculated as the fat content. 3.3.2.1.4 Ash content Dried material from the analysis of moisture content was ashed in a muffle furnace at 550°C overnight or until white (Chng 1992b). The white material after ashing was weighed and considered as the ash content of the material. 68 3.3.2.2 pH pH was measured using a pH meter (Digital Ionalyser Model 601 A, Orion Research Inc.) calibrated with buffer solutions of pH 4 and 7 compensated at ambient temperature and 40°C. The pH of the material was measured from the liquid dispersion of the sample (1:1, w:v), while those of silage was done by direct measurement using a calibrated pH meter stick (Gallenkamp). 3.3.2.3 Viscosity A viscometer (LVT Model, Brookfield Eng. Lab. Inc.) was used to measure the viscosity of the silage at ambient temperature. 3.3.2.4 Liquefaction Liquefaction of silage was defined as the proportion (w/w) of liquid to the total sample. Liquid silage was separated by centrifugation at 2000 g for 10 min using a Beckman J2-21 centrifuge (Beckman Instruments Inc.). 3.3.2.5 Soluble nitrogen One portion (v/v) of silage liquid was added to 3 portions of 20% TCA, mixed for 5 min and filtered (Whatman filter paper No. 3). The nitrogen was measured from the filtrate by the micro-kjeldahl method (Kirk and Sawyer 1991). In fish hydrolysates, the method of Liu and Pigott (1981) was adopted. A volume of aliquot was taken from the minced fish and crude enzyme mixture, and then mixed with 2 volumes of 20% trichloroacetic acid for 10-15 minutes, and filtered through Whatman No. 3 filter paper. The soluble nitrogen was measured as above. The results were corrected for the initial soluble nitrogen level. 69 3.3.2.6 Enzyme activity Enzyme activity was measured in the liquid silage stored at 40°C at day 5 using hemoglobin as substrate and the methods of Rick (1963a,b) for pepsin-like activity and for trypsin-like activity. One Pepsin Unit (PU) or one Trypsin Unit (TU) were defined as the amount of enzyme which hydrolysed hemoglobin at such an initial rate under the standard conditions (total volume 6 mL, containing O.lg hemoglobin, temperature 35.5°C), that the amount of tricholoroacetic acid-soluble hydrolysis products formed per min, gave the same optical density with the phenol reagent as for 1 mmole tyrosine (Rick 1963a,b). 3.3.3 Statistical analysis Experimental data were analysed in a mixed model analysis of variance (Montgomery 1991) using statistical computer package (Statistica Rel. 5, StatSoft, Inc.) in which acid, temperature and origin of raw materials were treated as the fixed factors, while storage time was the random factor. Statistical analysis was performed on data for unspoiled silage after the storage trial finished since spoiled silages were disposed off immediately. 3.4 Results 3.4.1 Experiment 1 The proximate composition of the viscera used in this study is shown in Table 3.3. In the preliminary study, an acid concentration of up to 2% was used, but this level failed to preserve the viscera, which spoiled within 1-2 days (Poernomo and Buckle 1993). Consequently, higher concentrations (3 and 4%) were then used. Upon the addition of acids, the pH of the chopped viscera sharply decreased from 7.5 70 (chopped raw material) to 3.3-5.0 depending upon the type and concentration of acids. The higher the acid concentration, the lower the pH. At the same concentration, hydrochloric acid produced a lower pH than did mixtures of organic acids. The addition of acids also changed the colour of the viscera, from reddish white to greyish white (Fig. 3.2). Table 3.3. Proximate composition of cowtail ray viscera from Muara Angke Component Proportion (% w/w)* Moisture 83.4 ± 1.53 Protein 14.5 ± 1.25 Fat 0.7 ±0.03 Ash 1.2 ±0.05 * Means ± SD from 3 determinations of 2 different raw materials. A B C D E Figure 3.2. The colour of viscera after the addition of acids (A = Raw material: B = Raw material added with 3% (v/w) mixture of propionic acid (PA) and formic acid (FA) (1:1. v/v); C = Raw material added with 4% (v/w) mixture of PA and FA (1:1, v/v); D = Raw material added with 3% (v/w) hydrochloric acid and E = Raw material added with 4% (v/w) hydrochloric acid) 71 During storage, an increase of pH was noted (Fig. 3.3). The pH of silages prepared by organic acids increased gradually. The initial pHs of these silages were 5.0 (Acidl, ambient), 4.5 (Acid2, ambient), 5.0 (Acidl, 40°C) and 4.7 (Acid2, 40°C), and became 5.2, 4.8, 5.3 and 4.8, respectively, after 120 days storage. Thus the pH of the silage changed by pH 0.1-0.3. Statistically, these increments were significant but no signs of spoilage were observed after the storage trial finished, while acid concentrations also gave significant effects on pH, while storage temperature did not (Appendix 8.1). On the contrary, pH of silages produced with hydrochloric acid increased at higher rates then all of the above and then spoiled. At ambient temperature, silage prepared using 3% hydrochloric acid spoiled at day 5 while those prepared using 4% hydrochloric acid spoiled at day 21 with pHs of 6.4 and 6.5, respectively. At 40°C, these spoiled at days 5 and 23, having pHs of 6.8 and 6.5, respectively. Data from the spoiled silage (those prepared with hydrochloric acid) were not included in the statistical analysis. The deterioration of silage was first indicated by mould growth on the surface of the silage or on the walls of the containers. This was followed by an unpleasant (faecal) smell within 1-2 days and a rapid increase in pH as shown in Figure 3.3. When advanced deterioration occurred, the unpleasant smell intensified and the colour of the silage changed to reddish to dark brown. The colour of the liquid from fresh and spoiled silage is shown in Figure 3.4. Soluble nitrogen in the liquid silage increased markedly during the first 5 days (Fig. 3.3). The level of soluble nitrogen in the raw material was about 31%, and became 70-84% after five days depending upon the acid and storage temperature. Silage stored at ambient temperature had soluble nitrogen levels of 70 and 74% for Acidl and Acid2, and 68 and 70% for Acid3 and Acid4, respectively. Thus, within 5 72 days, the level of soluble nitrogen increased by a factor between 2.3 and 2.8. During further storage time (up to 120 days), slight increases were still observed, and it reached 76 and 79% in the silages made with Acidl and Acid2, respectively. Thus, between days 5 and 120 there was only a 4-5% increase noted. Similar trends were also found in silages stored at 40°C, where the level of soluble nitrogen increased sharply during the first 5 days, i.e. from about 31% (raw material) to 80% and 84% for Acidl and Acid2, and 80% for Acid3 and 75% for Acid4, respectively. On average, the increase was by a factor of 2.5-2.8. During subsequent storage until day 120, the level gradually increased, reaching about 95% for both Acidl and Acid2. This increment was practically insignificant compared to that during the first 5 days. Statistical analysis shows that storage time and storage temperature had significant effects on soluble nitrogen, while acid concentration had no effect (Appendix 8.1). Fig. 3.3 also shows the changes in viscosity of the silages stored at ambient temperature and 40°C. At days 0 and 1 the silages were still practically solid viscera (see also Fig. 3.2) and their viscosity was not able to be measured. After 2 days, the viscosity of silages stored at ambient temperature were relatively high ranging between 1076 (Acid4) to 2048 cp (Acid2) while those stored at 40°C were very much lower, i.e. 17 (Acidl) to 24 cp (Acid4). Within 5 days of preparation, the viscosity decreased dramatically by 40% (Acid2, ambient) to 90% (Acid3, 40°C). On average, the viscosity of silages stored at ambient temperature decreased by 56% while stored stored at 40°C decreased by 84% in 3 days. Between days 2 and 5, the viscosity of silages stored at 40°C reduced by 3-7 cp/day for Acidl, Acid2, Acid3 and Acid4. After 120 days the viscosity of silages stored at 40°C did not change significantly and the silages were almost liquid, with 73 viscosity ranging between 3.1 to 3.2 cp (compared to the viscosity of distilled water of 1.6 cp). Although the silage stored at ambient temperature also showed a significant decrease in viscosity, i.e. by 91-92% from day 5 to day 120, the final viscosity was still high, i.e. 60-77 cp. On average, silages stored at ambient temperature decreased in viscosity by 315 cp/day between days 2 and 5 and by over 6 cp/day between days 5 to 120. Statistically, storage time and temperature significantly affected viscosity while acid concentration did not (Appendix 8.1). As shown in Fig. 3.3 the ratio of silage liquid to sample weight, or liquefaction level, of silages stored at ambient temperature increased markedly after the first 5 days, from about 16% (raw material) to 46% (Acidl), 50% (Acid2), 58% (Acid3) and 60% (Acid4). The liquefaction level then gradually increased reaching 83% (Acidl) and 75% (Acid2) at day 120. The liquefaction level of silages stored at 40°C sharply increased after the first 5 days, i.e. to 94% (Acidl), 93% (Acid2), 82% (Acid3) and 90% (Acid4), or by 5.7 fold, and remained relatively unchanged until the storage trial finished. Typical silages are shown in Fig. 3.5, indicating differences in liquefaction between those stored at ambient temperature and at 40°C after 30 days storage. Statistically, acid concentration did not significantly affect the liquefaction, while storage time and temperature did have a significant effect on the extent of liquefaction (Appendix 8 1). 74 Figure uefaction (%) Viscosity (cpoise) Sol. N (% of total N) 15.0 -Q 3.3 x 100 Chemical Storage stored and time □ physical at Ambient Acid ambient (days) 1; changes (28-32° temperature Acid2; C) of • cowtail Acid3; (28-3 ray A £ 2°C) ------ viscera Acid4. Storage and silage 40°C time from (days) Muara Angke 40° 75 C ABC Figure 3.4. Colour of unspoiled (A = stored at ambient temperatures, 28-32°C; B = stored at 40°C) and spoiled (C) silages A B Figure 3.5. Typical organic acid silages stored at ambient (A) and 40°C (B) after 30 days 76 3.4.2 Experiment 2 The proximate composition of raw material used in this study is shown in Table 3.4. Table 3.4. Proximate composition of cowtail ray viscera from Labuhan Maringgai, Lampung Component Proportion (% w/w)* Moisture 83.3 ± 0.14 Protein 15.5 ± 2.24 Fat 0.9 ±0.05 Ash 1.6 ±0.02 * Means ± SD from 3 determinations of 2 different raw materials. In this experiment, Acidl and Acid2 were used to confirm the results of Experiment 1 using different raw materials from Maringgai Lampung, Southeast Sumatra. Preparatory and analytical procedures were as previously described. The pH of the raw materials (pH 8.5) was slightly higher than those of Muara Angke. Upon the addition of acids, the pH decreased to 5.1, 4.7, 5.1 and 4.7 for Acidl and Acid2 at ambient temperature, and Acidl and Acid2 at 40°C, respectively, which were also higher than those of the pH of silages from Muara Angke after the addition of acids. During storage, the pH changed in a similar manner. The pH gradually increased (Fig. 3.6) reaching 5.5, 5.6, 5.5 and 5.1 for Acidl and Acid2 at ambient temperature, and Acidl and Acid2 at 40°C, respectively. On average, the pH increase was 0.3, which is higher than those for silages from Muara Angke. Results of statistical analyses are shown in Appendix 8.1. Statistically, the changes of pH due to storage time were significant, but, as with those for samples from Muara Angke, no signs of spoilage were observed. Acid concentration also significantly affected the pH, while storage temperature did not. Statistical analysis shows that different raw materials (Muara Angke and Lampung) did not significantly affect the pH of the resultant silage. 77 □ cr i Q- uefaction (%) Viscosity (cp) Sol. N (% of total N) Figure 00 100 20 10 40 60 7 30 90 10 0 0 4 - -■ ------f 0 3.6 15 Storage Chemical Maringgai 30 45 time and 60 stored Ambient (days) physical 75 at 90 ambient (28-32°C) □ changes Acidl: 105 temperature 120 AAcid2. of cowtail (28-32°C) ray *3 viscera Storage and silage time 40°C 3 from (days) Labuhan 78 □C> ] The level of soluble nitrogen of cowtail viscera from Maringgai, Lampung (24.6% of total N) was lower than from Muara Angke (30.5% of total N). As shown in Figure 3.6 the soluble nitrogen level during 120 days storage increased at a high rate during the first 5 days then remained relatively constant. Within 5 days, the soluble nitrogen level increased by a factor between 3.1 to 3.8 depending upon the acid and storage temperature to 75%, 77%, 87% and 92% for Acidl and Acid2 at ambient, and Acidl and Acid2 at 40°C, respectively. These factors were higher than for samples from Muara Angke for the same period. During 120 days storage, the level of soluble nitrogen in silages stored at ambient temperature increased slightly and the final levels were 79% and 80% for Acidl and Acid2, respectively. Similar trend were also observed for those stored at 40°C, in which the levels of soluble nitrogen at the end of storage were 91% and 95%. Thus between days 5 and 120, the level of soluble nitrogen increased by only 3.1- 3.4%. Statistically, acid concentration and the origin of raw materials (Muara Angke and Labuhan Maringgai) did not influence the level of soluble nitrogen, while storage temperature and time did (Appendix 8.1). The viscosities of silages prepared from Maringgai were not able to be measured until day 2, by which silages stored at ambient temperature had viscosities of 2100 and 1483 cp, while those stored at 40°C had viscosities of 17 and 16 cp for Acidl and Acid2, respectively. At day 5, the viscosities of silages stored at 40°C decreased by 63% and 73% for Acidl and Acid2, while those at ambient temperature decreased by 58% and 49% for the same acids. On average, the reductions after 5 days were 68% and 54% for silages stored at 40°C and ambient temperature, respectively. For the same period, the viscosity of silages from Muara Angke prepared with the same acids decreased by 79% and 50%. 79 During further storage (up to 120 days), the viscosities of silages stored at ambient temperature continuously decreased, while those at 40°C were relatively stable (Fig. 3.6). At the end of storage, the viscosities of the silages were 160, 139, 3.5 and 3.1 cp for Acidl and Acid2 at ambient temperature, and Acidl and Acid2 at 40°C, respectively. The silages stored at 40°C were practically liquid after 120 days storage. Compared to the values at day 2, the viscosities of silages stored at ambient temperature decreased by about 96% for both Acidl and Acid2, while those stored at 40°C decreased by only 81% and 75%. Storage time and temperature significantly affected viscosity, while acid concentration did not (Appendix 8.1). No significant difference observed in the viscosities of the silages from both locations (Muara Angke and Maringgai). Fig. 3.6 also shows the liquefaction of the silages during storage. During the first 5 days, the liquefaction of the silages stored at 40°C was high (87% and 88% for Acidl and Acid2), while those stored at ambient temperature increased only about 37% and 39% from the original level (16%). During subsequent storage, the liquefaction level of silages stored at ambient temperature increased continuously but at a lower rate. After 120 days storage, the levels were 84% and 83%. Silages stored at 40°C, on the other hand, showed no further significant increase in liquefaction level during subsequent storage. At day 120, the levels were 97% and 95% for Acidl and Acid2, respectively. Storage time and temperature affected the level of liquefaction. There were no significant differences in liquefaction level for silages from Muara Angke and Maringgai, and acid concentration did not affect the level of liquefaction. The enzyme activities in the silage liquid were 0.5 TU and 0.8 PU for trypsin like and pepsin-like enzymes respectively, and increased by 8 and 13 fold, respectively after ammonium sulphate precipitation and dialysis. 80 The results of fish protein hydrolysis by crude enzymes are shown in Figure 3.7. At 45°C, the hydrolysis of fish protein at all pHs and enzyme to fish meat ratios was very slow resulting in a low level of soluble nitrogen (10-30% of total nitrogen). The levels of soluble nitrogen at 60°C was relatively constant during the first 2h, sharply increased in the next hour, then leveled off. All slurries at 60°C, except that at pH 2.0 with 5% ratio of crude enzyme, had soluble nitrogen of about 60% (of total nitrogen) after 4h hydrolysis. The slurry at pH 2.0, at 60°C and 5% ratio of crude enzyme had final soluble nitrogen of about 50% (of total nitrogen). 3.5 Discussion 3.5.1 pH and spoilage of silage In preliminary studies, acids at levels up to 2% failed to preserve the viscera which spoiled within 1-2 days. This level of acid was actually higher than that (1.5%) recommended by Raa and Gildberg (1976), Gildberg and Raa (1977) and Gildberg and Almas (1986) for the ensilation of fish viscera from temperate waters. Kompiang el a/. (1980) reported a similar result to that of this study, in which a mixture of propionic and formic acids at levels of up to 1.5% failed to preserve by- catch fish from Indonesian shrimp trawlers. This was probably due to the differences in chemical composition between fish from temperate and tropical waters. Tropical fish usually have a higher mineral content and thus need more acid to reduce the pH to sufficiently low values to ensure preservation (Raa and Gildberg 1982). In the present study, all treatments using mixtures of propionic and formic acids (3% and 4%) resulted in higher pH than those produced using the same mixture of acids at 1.5% applied to viscera from temperate fish. Gildberg and Raa (1977) reported that a 1.5% mixture of propionic acid and formic acids produced cod viscera silage with an initial pH of 4.3, while in the present study, silages produced 81 with the same mixture of acids at levels of 3% and 4% had an initial pH of 4.2 - 5.0 (for viscera from Muara Angke) and 4.7-5.1 (Labuhan Maringgai). 60 - 30 - 15 - pH 5.0 60 - TO Ho— 45 - & 30 - o CO pH 8.0 60 - Hydrolysis time (hours) Figure 3. 7. Hydrolysis of fish meat by silage crude enzymes at different proportion of crude enzymes to meat, pH and temperature. ■ 5%. 45°C; • 10%. 45°C; A 5%. 60°C; ▼ 10%. 60°C 82 The freshness of the fish also might affect the initial pH of the raw materials since aged fish have a higher pH (Sikorski, Kolakowska and Burt 1990), thus more acid is needed to lower the pH to the same level than when fresh fish is used. In developing countries such as Indonesia, the quality of landed fish, especially cheap ones such as stingrays, is often not fresh. Elasmobranches such as stingrays are characterised by a high level of urea which produces a high pH in the meat. The level of non-protein nitrogen in the raw material in this experiment also reflects the characteristics of elasmobranchii fish, which contain higher levels than teleost fish. The level of non-protein nitrogen of cowtail ray viscera in the present experiment was in the range of 25-31% of total nitrogen, which is much higher than the level (20%) for temperate water cod viscera (Backhoff 1976), or for tropical silverbelly viscera of about 9.5% (Jayawardena and Poulter 1980). To produce stable fish viscera silage, Strom et a/. (1980) recommended that sufficient acids should be added to reduce the pH to approximately 4.5. In the present study, all silages produced with hydrochloric acid was between 3.3 - 4.3. However, although these treatments were able to liquefy the viscera to some extent, they failed to produce stable silages, which spoiled within 5 days (Acid3 at both storage temperatures), 21 days (Acid4 at ambient temperature) and 23 days (Acid4 at 40°C). Raa and Gildberg (1982) stated that silages produced with inorganic acids at pH 3.5 to 4.0 were not completely protected against fungi. This was also shown in the present study, where the first sign of spoilage was the occurrence of mould growth on the surface of the silage and the container wall before an unpleasant smell could be noticed. This supports the findings of other workers on fish silage. Silage from salmon farm mortalities produced with mixtures of 0.75% citric acid and 0.5% formic acid or 1% citric acid and 0.5% formic acid showed similar 83 signs of spoilage (Gao, Lo and Liao 1992). The appearance of patches of mould growth was also observed by Gildberg and Raa (1977) when mixtures of cod viscera silage produced with 1.5% formic acid and barley straw meal started to deteriorate. Jayawardena and Poulter (1980) reported that yellowish mould colonies were found in silverbelly silage produced using 3.5% formic acid (w/w) on day 55 during storage at ambient temperature (28-30°C) in Sri Lanka. Levin et al. (1989) found not only mould but also oxidative yeasts in spoiled silage from hake frames and entrails produced using phosphoric acid. It has been reported that the aflatoxin-producing fungus Aspergillus flavus was able to grow on the surface lipid of unprotected silage such as those produced with inorganic acids or formic acid alone (Kompiang et al. 1980). During autolysis, the content of ammonia, amines, amino acids and peptides increases, and this affects the buffering capacity of the fish which increases in pH (Sinell 1980). However, when moulds and yeasts are present in aerobic silage such as in the present study, their growth results in further increase in pH which initiates the growth of spoiling bacteria. In the present study, when advanced spoilage occurred, the unpleasant smell intensified and the colour of the silage turned to a reddish to dark brown, and the pH sharply increased to about 6.5. The unpleasant smell was undoubtedly due to the amines produced in the spoilage process. Changes of colour to dark brown were also observed by Gildberg and Raa (1977) in spoiled cod viscera silage and barley straw meal mixture (2:1, w/w). It is likely that the signs of spoilage of temperate and tropical fish silage are similar, and for practical purposes, these signs can be used as physical spoilage indicators without involving any measurement or equipment. Based on these results, hydrochloric acid alone can be utilised to produce viscera silage if the product is going to be used within 15 days. In the present study, 84 hydrochloric acid was able to promote liquefaction of the viscera solubilising nitrogen up to 70% at ambient temperature and up to 80% at 40°C within 15 days, which is comparable to those produced using organic acids. However, care should be taken when producing silages using inorganic acids since not only is the silage not protected against mould spoilage but also the pH is very low and needs to be neutralised before use, especially when used as an animal feed component. It is recommended to add 2- 5% (w/w) chalk to neutralise the silage produced using inorganic acids (Peterson 1953). The neutralisation step itself is laborious and the salt level resulting from neutralisation is high and nutritionally undesirable (Raa et al. 1983). 3.5.2 Proteolysis During ensilation, protein and other nitrogenous compounds are broken down due to autolysis and the increase in soluble nitrogen during ensilation (Fig. 3.3 and 3.6) shows the typical pattern of autolysis. The solubilisation of nitrogen is rapid during the first days, then slower thereafter. In the present study, during the first 5 days, the level of soluble nitrogen, on average, was about 74% and 86% for organic acid silage stored at ambient temperature and 40°C, respectively. This level is similar to viscera silage of temperate water cod stored at 27°C and 30°C, (Gildberg and Raa 1977, Backhoff 1976) and rainbow trout stored at 30°C (Raghunath and McCurdy 1990). After the same storage period, silage from fish flesh or fish frames or whole fish, either from temperate or tropical waters stored at 20-30°C, contained less soluble nitrogen than shown in the present study (Tatterson and Windsor 1974, Backhoff 1976, Hall et al. 1985, Levin et al. 1989, Tatterson 1982, Gao et al. 1992, Alwan, Buckley and O'Connor 1993, Lo, Liao and Bullock 1993). The autolysis in acid ensilation is mainly due to the action of endogenous enzymes and the difference in soluble nitrogen level in silage from viscera in this study 85 and those from fish flesh or fish frames or whole fish was mainly due to the relatively higher concentration of enzymes in the viscera. Since protein is the major structural component of fish, it is mainly the proteases that play significant roles in autolysis during ensilation. The rates of enzymic reaction during ensilation of viscera from both temperate and tropical fish are not significantly different, and the digestive enzymes might behave similarly at temperatures around 30°C. It has been reported, however, that the digestive proteases from cold water fish are relatively more active at low temperatures than those of fish from warm waters (Gildberg 1982), while Raa and Gildberg (1982) indicated that there are little differences in temperature optima of digestive enzymes from temperate and tropical fish. Hall et al. (1985) and Raghunath and McCurdy (1990) indicated that both exopeptidases and endopeptidases were active in the autolysis of viscera silage. In acidic conditions such as in silage, it is likely that endopeptidases such as pepsin-like enzymes, which have acidic pH optima, dominate the reaction. However, Raghunath and McCurdy (1990) also observed trypsin-like enzyme activity. Although it has an alkaline pH optimum, which is possibly not active in acid silage, this enzyme is reported to be quite stable in acid conditions (Hjelmeland and Raa 1982). In the present study, the activity of both acidic and alkaline proteases, which were probably pepsin- and trypsin-like enzymes, was also observed in silage prepared from Maringgai cowtail ray viscera prepared with Acidl at day 5 (other silages were not assayed for enzyme activity during ensilation). The activity was 0.8 PU and 0.5 TU for pepsin-like and trypsin-like enzymes respectively. Other enzymes present in the acidic rainbow trout viscera silage as reported by Raghunath and MacCurdy (1990) were aminopeptidase, cathepsin B and cathepsin C. 86 The effect of temperature on the soluble nitrogen level of the silage in this study clearly reflected the pattern of enzymic reaction in which higher temperature increased the reaction rate. Gildberg and Raa (1983) found that acidic digestive proteases from capelin (Mallotus villosus) showed highest activity against a hemoglobin substrate at temperatures of 38°C and 43°C. Tropical fish such as oil sardine (Scirdinella longiceps) have temperature optima for acidic digestive proteases between 40-50°C, and for alkaline digestive proteases at about 54°C (Gildberg 1982), while Sulistyani and Heruwati (1992) reported that alkaline digestive proteases from tropical grouper (Epiphenelns tauvinci F.) had temperature optima of about 45°C. Thus it is reasonable that in the present study a temperature of 40°C resulted in a faster rate of autolysis since it might be closer to the temperature optima of fish digestive enzymes. The course of nitrogen solubilisation is also reflected by the decrease in viscosity and increase in the liquefaction level. The patterns of viscosity and liquefaction level changes are thus similar to the changes in soluble nitrogen, i.e. higher rate of changes during the first days, then little change. Naturally, factors affecting the rate of nitrogen solubilisation also affect the rate of viscosity reduction and liquefaction level. This was reflected also in the present experiment where the higher temperature significantly affected the changes of viscosity and liquefaction. The low (4.2 cp, on average) final viscosity of the silages clearly shows that the silage was well liquefied, confirming that the nitrogen in the viscera had been almost completely solubilised within a short time. Hall et al. (1985) indicated that when complete solubilisation has been achieved, measurement of non-protein nitrogen in whole silage gave similar results to those found in the liquid fraction of the silage after centrifugation. This indicates that practically all breakdown products which are TCA soluble have been released into the liquid fraction, and consequently, the viscosity 87 decreased. The three dimensional plot of soluble nitrogen level, viscosity and liquefaction of all unspoiled silage after 120 days storage is shown in Figure 3.8 and can be expressed by the following equation : Z = 39.863 + 2.365X + 0.012Y - 0.01X2 - 0.00086XY + 0.00056Y2 where Z = Liquefaction (%) X = Soluble nitrogen (% of total nitrogen) Y = Viscosity (cp). The relationship between soluble nitrogen and liquefaction can be represented by the equation: Y = 0.682 X + 29.749, (n = 96 pairs, r2 = 0.90) where Y is the soluble nitrogen (% of total nitrogen) and X is liquefaction (%). Figure 3.9 shows the linear approximation of both parameters. In ensilation, it is important to monitor the progress of autolysis which produces soluble nitrogen, and using such approximation, this can be sufficiently predicted from the liquefaction level. In developing countries where the means to quickly measure the soluble nitrogen is not always available, this approximation can be used. Simple filtration using glass wool or filter cloth can easily separate solids from the liquid to measure the liquefaction level. In addition, viscosity can be a useful indicator for process control to predict the degree of liquefaction or autolysis during ensilation (Hall el al 1985). The results of liquefaction measurements show that although the ensilation has been prolonged to 120 days, complete liquefaction did not occur. The highest level of liquefaction in the present study was almost 95%, found in silage stored at 40°C, indicating an undigested fraction which was resistant to enzymatic digestion. This confirms the results of previous studies by other workers on fish silage. 88 z=-39.863+2.365x+0.012y-0.01xx-8.566e-4xy+5.624e-6yy Figure 3.8 Surface plot of parameters of cowl ail ray viscera silage ^ •• Liquefaction (%) Figure 3.9. Correlation of liquefaction and soluble nitrogen in cowtail ray viscera silage (y = 0.68x + 29.48; r2 = 0.897) 89 Visual examination of the fraction showed that the undigested fraction consisted of minute whitish rocky materials, most probably sand, and shells of crustaceans, and no insoluble meats were found. However, it is possible that insoluble protein might be present, as an investigation by Gildberg and Raa (1977) showed that the undigested fraction of cod viscera silage contained amino acids. A similar result was also observed in silage of whole silverbelly from tropical waters (Hall el al. 1985). There is no explanation in the published literatures as to why undigested fractions always remain in ensilation. It has been assumed that there are certain protein fractions in the fish that are resistant to extra-cellular proteolytic degradation at constant pH (Raa and Gildberg 1982). The level of liquefaction of silage stored at ambient temperature in this study was slightly lower than that of temperate fish viscera stored at a similar, while at 40°C the level of liquefaction was higher than that of temperate fish viscera silage stored at around 50°C. Gildberg and Raa (1977) found that autolysis of cod viscera during ensilation was highest at about 50°C, however at this temperature the proteins which were soluble due to autolysis precipitated again. This probably explains why the liquefaction of viscera at 50°C was lower than that at 40°C. Some organic silages in the present study had rather high final pH (> 4.5) but were stable. This was probably due to the high level of urea which was converted into ammonia during proteolysis. Ammonia has been reported to have preservative effects and has been tested successfully for fish preservation (Wheaton and Lawson 1985). 3.5.3 Fish protein hydrolysis The study to determine the ability of crude enzymes of viscera silage to hydrolyse fish meat shows that the enzymes in the silage were still active at least after 5 days storage at 40°C. Reece (1988) found that activity of acidic proteases in silage made from cod, mackerel and salmon viscera were almost constant during 5 days storage. Prolonged storage at up to 20°C, however, led to gradual loss of protease activity from mackerel and salmon, but this was not observed in cod viscera. Raa and Gildberg (1976) and Gildberg and Raa (1977) observed that proteases in cod viscera 90 silage were stable for at least 9 days at 27°C. The present study shows that at 60°C the crude enzymes applied were able to hydrolyse fish protein producing about 60% soluble nitrogen at pH 5.0 and 8.0, and almost 50% at pH 2.0. The similar results for different pHs were most probably due to the high buffering capacity of fish meat which tended to adjust the pH of the slurries to close to the original fish meat pH. Additionally, as protein hydrolysis is accompanied by a release or uptake of H’ (Sorensen 1908), the pH of the mixture changes, except in the region around pH 5-6 where the uptake and release of protons cancel each other (Adler-Nissen 1986). Thus since the pH of the fish meat and enzyme mixture in the present study was not maintained, pH 2.0 would eventually increased, and pH 8.0 decreased to close to the original fish meat pH. In a test of fish meat hydrolysis by bacterial proteases, Rebeca, Pena-Vera and Diaz-Castaneda (1991) found that nitrogen solubilisation was faster during pH- controlled hydrolysis than without pH control. They also observed that a decrease in pH, from 9.5 to 8.0 for the slurries using alkaline protease and from 7.5 to 6.5 for the slurries using neutral proteases, occurred during the first hour of hydrolysis, which corresponded to a 15-30% reduction of protease activity. Another possible reason why the level of soluble nitrogen in the first 2h was low is the relative resistance to proteolysis of native fish protein (Reeck 1971). Protein hydrolysis by pepsin and trypsin was easier in slightly cooked fish than in raw or fully cooked fish (Saha 1940), and small amount of denatured proteins were sufficient to influence the kinetics of proteolysis (Rupley 1967). This also explains why hydrolysis at 45°C was slower than that at 60°C since it is likely that at 45°C the protein did not suffer as much heat denaturation as at 60°C. Mackie (1982a) reported that in fish protein hydrolysate "the degree of hydrolysis as measured by the proportion of trichloroacetic acid soluble nitrogen will 91 be in excess of 50% of the total nitrogen of the fish". The hydrolysates produced at 60()C in the present study can be said to have met the above definition. In the present study the hydrolysis at 60°C seemed to cease after 3h. This was a typical hydrolysis pattern of fish muscle in which there is always about 20% of the total nitrogen remaining insoluble even when a further amount of enzyme is added (Mackie 1982b). Similar results were reported by Shahidi, Xiao-Qing and Synowiecki (1995) in the hydrolysis of capelin protein. No increase in the release of soluble protein in capelin hydrolysis was observed when a further amount proteolytic enzyme was added after the hydrolysis reached the stationary phase. They suggested that the hydrolysis was probably inhibited by the hydrolysis products or by cleavage of all susceptible peptide bonds by the enzyme. 92 4 USE OF CRUDE PEPTONES FROM COWTAIL RAY (T. sephen) VISCERA SILAGE AS MICROBIAL GROWTH MEDIA 4.1 Introduction Progress in biotechnology, both in research and in commercial products, has created an increasing demand for microbial growth substrates which constitute a significant cost in the production of microbial cell mass. Nitrogen sources are the most expensive component of microbial growth substrates, and are supplied primarily from animal products such as meats, internal organs and gelatin, milk and casein, plants and yeasts in the form of low molecular weight proteins and extracts (Power and McCuen 1988, Clausen el a/. 1985, Gillberg, Batista and Strom 1989, Vecht- Lifshitz et al. 1990). This has led to the investigation of cheaper nitrogen sources. Peptones are partially hydrolysed proteins which are soluble in water but not heat coagulable (Ockerman and Hansen 1988, Haard, Simpson and Sikorski 1994). Hydrolysis to produce peptones is achieved by either acid or alkali or by enzymic digestion of proteins. During hydrolysis parts of the proteins are broken down into low molecular weight products such as peptones. Hydrolytic enzymes in the digestion can be intrinsic, i.e. those which are already available in the raw materials, or can be added, such as papain, pepsin or trypsin. Peptones can be used as a basal ingredient in microbiological culture media, either for growing microorganisms or for the production of various microbial products such as antibiotics, enzymes and vaccines. Acid or alkali digestion usually gives a high yield but tends to destroy the vitamin content of the protein and some of the amino acids. Neutralisation steps after acid or alkaline hydrolysis result in a high ash content which is not always acceptable 93 (Gildberg etal. 1989). The first uses of fish materials as sources of nutrients for microorganisms were reported by Anderson and Fellers (1949) and Tarr and Deas (1949). Anderson and Fellers (1949) prepared sterile fish muscle juice which was able to support growth of bacteria. Tarr and Deas (1949) demonstrated the ability of a tryptic digest of fish flesh to support the growth of anaerobic bacteria. One year later, a preparation of fish peptones by enzymic hydrolysis of trout tissues using commercial pepsin in acid conditions was reported by Snieszko, Griffin and Friddle (1950). The peptones were used to prepare an enrichment medium for a fish pathogen, Hemophilus piscium. The growth of this pathogen in a medium enriched with 3-5% fish peptones was abundant. Since then, many attempts to explore the use of fish peptones as a component of microbial growth substrates have been reported. An evaluation of three papaic digests from the flesh of a fresh water fish, Barbus dubious, as microbial growth substrates was reported by Krishnaswamy and Lahiry (1963). Pure cultures of aerobic and facultative anaerobic microbes were used to test the fish peptones compared to two commercial peptones. The fish flesh digests, although inferior to the commercial peptones in supporting the growth of test microorganisms, were regarded as suitable ingredients of culture media for biochemical tests such as indole and methyl-red reactions. Green el al. (1973) reported tests of 5 fish peptones for their ability to support microbial growth. The test fish peptones included enzymic digests of menhaden and hake using commercial enzymes, autolysis digest of menhaden, menhaden fish solubles and commercial menhaden peptone which was available at the time of the experiment. A number of pure cultures of microorganisms, including Lactobacillus casei, L. plant arum, Pediococcus cerevisiae, Escherichia coli and Staphylococcus aureus, and some materials used as sources of mixed populations of microorganisms (foods, waste 94 water and soil) were used in the test using an agar plating technique. In most cases the fish peptones were able to support the growth of the test microorganisms, from both pure and mixed populations. At the time of the experiment, it cost 26-60 US cents to produce 1 kg fish peptones which was lower than the cost of commercial peptones from casein and meats. Autolytic digestion has been used by several workers to produce peptones from fish. Green, Paskell and Goldmintz (1977) prepared an autolysate of red hake (Urophycis chuss) at 50-55°C. Compared to other commercial peptones, the fish peptones were similar or better in supporting the growth of the same microorganisms with that used by Green et al. (1973). Gildberg et al. (1989) applied two step enzymic hydrolysis to produce peptones from capelin and blue whiting. The first step was autolytic digestion similar to ensilation at 25°C using 2.5% (v/w) phosphoric acid. The sediment of 3 days ensilation was separated by centrifugation, resuspended in water and added to commercial alkaline enzymes either from fish or microbes. The pooled peptones produced from both steps were comparable to commercial peptones in supporting growth of Vibrio anguillarum, V salmonicidae, and Lactobacillus sp. from fish intestinal mucosa and fermented cabbage. Fish wastes contain nutrients similar to those found in fish flesh, and have been investigated as sources for microbial peptones. Strasdine and Melville (1972) reported that salmon canning waste water supported the growth of bacteria both as a complete medium and as a supplementary nitrogen source. Beuchat (1974) demonstrated that catfish waste peptone, including heads and skins, were able to support the growth of test microorganisms, which included a range of bacterial and fungal cultures. 95 Jassim, Salt and Stretton (1988) tested extracts of enzymically digested fish waste, including heads, fins and skeletal debris, as an additional nitrogen source or as a substitute for beef extract in media for microbial growth. The digestion was at 37°C and pH 7.0-8.0 overnight, using commercial trypsin. The fish extract was able to support the growth of 15 pure cultures of test microorganisms, and together with beef extract and proteose peptones was able to increase the production of staphylococcal enterotoxin to a level higher than that on standard media containing beef peptones. Shrimp wastes also have been utilised to produce microbiological peptones. The wastes, including the heads and shells, have proved able to support the growth of a number of microorganisms, to similar or better levels than commercial peptones (Stephens et al. 1976, Rao, Dwarakanath and Saraswathi 1980, Narkviroj 1987, Sophanodora and Buckle 1988). Fish viscera contains high levels of digestive enzymes and this makes autolytic digestion of fish protein more economically viable for marine waste utilisation as enzyme costs are nil. The utilisation of fish viscera has been limited to animal feed, but the abundance of proteolytic enzymes and protein in viscera opens the possibility of further utilisation. During viscera ensilation, which is basically an autolysis in acid conditions, proteins of the viscera are broken down into soluble low molecular weight products in a complex mixture of peptides and amino acids. This product can be used as a nitrogen source in microbial growth substrates as demonstrated by some workers. Clausen et al. (1985) investigated the autolysate of cod viscera as growth substrate for V anguillarum and Proteus sp. The ensilation was performed by an addition of 3% (v/v) hydrochloric acid to reduce the pH of minced cod viscera to pH 2.5. The mixture was kept at 23°C for one week before the aqueous nitrogen-rich 96 solution was separated, filtered and concentrated (5x). The test microorganisms grew faster and produced higher yields of cell mass on media containing peptones from fish compared to those on commercial peptones (Bacto Tryptone, Difco). Further tests of the above peptones were done by Vecht-Lifshitz el al. (1990) using pure cultures of 9 microorganisms. Total growth of the test microorganisms in submerged culture after 48h was higher on media containing fish peptones than on bacto-peptone (Difco). Almas (1990) and Strom and Raa (1993) outlined the development of commercial production of fish peptones in Norway, including those produced using viscera. No further studies regarding the use of peptones from viscera to support the growth of microorganisms have been reported. Skorupa and Sikorski (1992, 1993) reviewed the use of fish and fish waste, including marine crustacean waste, as a source for microbiological peptones produced by autolysis and enzymic hydrolysis. Commercial fish peptones are now produced in Norway, and have been used by some laboratories (Gildberg el al. 1989, Gorbenko and Dzyuban 1992, Page 1992, Page and Cornish 1993, Chen and Page 1994, Cheng cl al. 1994). The Indian Council of Agricultural Research was granted a patent on the process for the preparation of bacteriological peptones from fish (Anon. 1973a). In the process, papain was used to assist the hydrolysis, while defatted casein was added as an additional protein source. The price of commercial peptones is very high in Indonesia, and with the emerging biotechnology industry and research in Indonesia, cheap but high quality peptones are necessary. The simplicity of the ensilation process and the abundance of viscera waste in Muara Angke, Jakarta, and Labuhan Maringgai, Lampung, where about 5,000 t rays are processed annually, offer an opportunity to explore the possibility of producing peptones from viscera. However, little information is available 97 regarding the utilisation of tropical viscera for microbiological media peptones, especially by ensilation. The aims of the present study are to prepare crude peptones from cowtail ray viscera and examine the suitability of the peptones as microbiological growth media compared to commercial peptones. 4.2 Materials 4.2.1 Viscera The raw material used in this study was cowtail ray (T. sephen) viscera obtained from Labuhan, Maringgai, Lampung. Proximate composition and treatments applied before processing were similar to those outlined in Section 4.2.1. 4.2.2 Equipment Equipment used in the study is listed in Table 4.1. Table 4.1 Equipment used in the study Equipment Manufacturer and type Atomic absorption spectrophotometer PYE, Unicam Ltd., SP9 Series Centrifuge Beckman Instrument Inc., J2-21 High speed liquid chromatography Shimadzu - Du Pont LC-1 Incubator(oven) Contherm Scientific Co. Incubator (orbital) Lab-line Orbit, Model No. 3527-1 pH meter Orion Research Inc., Digital Ionalyzer, Model 601A Spectrophotometer Spectronic 20, Bausch & Lomb, Inc. 4.2.3 Chemicals and microbiological media Chemicals used for silage preparation and for proximate composition analysis were as described in Section 3.2.2, while microbiological media and other chemicals used for microbiological tests are as shown in Table 4.2. 98 Table 4.2 Chemicals and media used in the study Chemical Brand and specification Ammonia E. Merck, 25% GR Ammonium molybdate E. Merck Bactopeptone Difco Bacteriological peptone Oxoid Dextrose Difco Hydrochloric acid E. Merck, 12 N Nutrient broth Difco Phenolphthalein indicator E. Merck Polyethylene glycol 20,000 E. Merck Potassium dihydrogen phosphate E. Merck Sodium citrate E. Merck Sodium hydroxide E. Merck Stannous chloride E. Merck Thiotone E peptone BBL 4.2.4 Sources of microorganisms Microorganisms used to test the ability of the peptones to support microbial growth were from two groups, i.e. mixed populations and pure cultures. Mixed populations were those from foods (beef, eggs and milk) purchased on the day of the experiment and pure cultures were obtained from different suppliers (Table 4.3). The microorganisms selected for test represented those with a wide range of growth factor requirements, including Gram positive and negative, aerobic and anaerobic facultatives, non-spore formers and spore formers, spherical and rod shaped, as well as gas formers. A. flcivus was selected as a most common toxigenic fungus found in a wide variety of dry foods and responsible for outbreaks in many countries, including tropical regions such as Southeast Asia. S. cerevisicie was selected as a fermenting yeast used for production of indigenous foods in Southeast Asia. 99 Table 4.3 Microorganisms (mixed and pure cultures) used in the study Sources of microorganisms Origin and condition at time of purchase Mixed populations from: Beef Local market, Jakarta, cut, from open display Egg (chicken) Local market, Jakarta, intact and packed in a wooden box Milk Local milk supplier, Jakarta, fresh, not pasteurised, not homogenised, and cooled (15°C) Pure cultures: Aspergillus flavus (BCC F.0073) RIVS*, lyophilised, derived from NRRL 4098 Bacillus subtilis (BCC 2034) RIVS*, lyophilised, isolated from milk Escherichia coli (ATCC 25922) BBL**, lyophilised Saccharomyces cerevisiae RIVS*, lyophilised, derived from NRRL Y-2034 (BCC F.0074) Staphylococcus aureus BBL**, lyophilised (ATCC 25923) * Research Institute for Veterinary Science, Bogor, Java, Indonesia ** Becton Dickinson Microbiology System (BBL Quali-Discs) 4.3 Methods 4.3.1 Preparatory 4.3.1.1 Ensilation of cowtail ray viscera and preparation of ray viscera peptones Cowtail ray viscera was ensiled as described in Section 3.3.1. The acids used to ensile the viscera were a mixture of propionic and formic acids (1:1, v/v) at level of 3% (v/w) and hydrochloric acid (12N) at level of 4% (v/w). The ensilation was conducted at 40°C for 5 days, after which the silage was centrifuged at 10,000xg for 10 min. The sediment was separated from the silage, while the lipid on the surface was skimmed off. Coding used in this experiment is FSHCL for silage produced using HC1, and FSPF for that produced using propionic and formic acids. Several methods to concentrate the supernatant were employed. Method 1 was by mixing the silage liquid with dry Sephadex G-25 (5:1, v/w), which was left for 100 15 min at 5°C then centrifuged at 2000xg for 10 min and the supernatant collected. In Method 2, the silage liquid was dialysed against 20% polyethylene glycol 20,000 overnight at 5°C. Method 3 was by evaporating the silage liquid at 50°C under vacuum using a rotary evaporator (Rotavapor R110; Buchi) for approximately 25 min until concentrated approximately 5 times. The concentrated supernatants, referred to as ray viscera peptones, were then frozen and kept at -45°C until used. 4.3.1.2 Culture media The liquid culture media for bacteria and mixed populations of microorganisms were made from 0.1% (w/v) dextrose and 0.5% (w/v) peptones, either commercial (Difco, Oxoid or BBL) or ray viscera peptones. The pH was then adjusted to 7.0 using IN HC1 or IN NaOH and the liquid was autoclaved (121°C, 15 min). The liquid media for fungi and yeast consisted of 1.0% (w/v) dextrose and 0.5% (w/v) test peptones as above. The pH was adjusted to 5.5 and the solution autoclaved. The ray viscera peptones were added in volume with respect to their protein content such that the proportion of the protein in the media was 0.5% (w/v). 4.3.1.3 Microbial growth test All experiments were conducted in three replicates. The protocols of the experiments are as outlined in the following sections. 4.3.1.3.1 Preparation of cultures, inoculation and incubation 4.3.1.3.1.1 Bacteria and mixed population of microorganisms Beef (25 g) was aseptically blended in 225 mL sterile distilled water in a Waring blender for 2 min at medium speed. The suspension was inoculated at a 101 dilution of 10° . Egg (whole) was whipped and diluted with sterile distilled water (1:10, v/v), and inoculated at a dilution of 10'2. Milk was diluted at 10'2 with sterile distilled water before inoculated. Lyophilised pure bacterial cultures (E. coli, Staph, aureus, and B. subtilis) were suspended in 100 mL nutrient broth (Difco) and incubated at 37°C for 24h. The cultures were centrifuged at 10,000x,g 10 min, then resuspended into 20 mL sterile 0.85% saline solution. One mL food or cell suspension was inoculated into 50 mL sterile liquid medium containing test peptones prepared according to Section 4.3.1.3 in 250 mL Erlenmeyer flasks with cotton wool plugs, and incubated at 37°C on an orbital incubator at 150 rpm. 4.3.1.3.1.2 Yeast Lyophilised yeast culture (S. cerevisiae) was suspended in 100 mL malt extract broth and incubated at 25°C for 2 days. The culture was centrifuged (10,000xg, 10 min) then resuspended in 20 mL sterile 0.85% saline solution. Cell suspension (1 mL) was inoculated into 50 mL sterile liquid medium containing test peptones prepared according to Section 5.3.1.3 in 250 mL Erlenmeyer flasks with cotton wool plugs, and incubated at 25°C in an orbital incubator at 150 rpm. 4.3.1.3.1.3 Fungi Lyophilised fungal culture (A. flavus) was suspended in 25 mL diluent containing 0.05 g agar and 0.0125 g Tween 80, then cultured on MEA agar slants for 2 days at 30°C. The surface of the culture was washed with 10 mL sterile 0.85% saline solution and 0.1 mL of the washing solution was inoculated into 50 mL sterile liquid media containing test peptones prepared as described in Section 4.3.1.3 in 250 102 mL Erlenmeyer flasks with cotton wool plugs, and incubated at 30°C in an orbital incubator at 150 rpm. 4.3.1.3.2 Growth curves, growth rate and total growth Growth curves, except that of fungi, were monitored by measuring the absorbance of the cultures. Periodically up to 36h, 2.5 mL culture samples were aseptically withdrawn and the absorbance immediately read at 600 nm against the respective original liquid media blank. Absorbance was plotted against incubation time (semi-log) to construct a growth curve, and the straight and steep segment of the curve where the correlation between time and log absorbance is linear was considered the exponential growth phase. Total growth was defined as the difference between initial and final absorbance, while growth rate was the slope (logA/h) of the exponential growth phase curve. For a pure culture or culture from a food, growth rate was ranked from 1 (highest) to 5 (lowest), and the ranking was summed for each peptone. 4.3.1.3.3 Biomass production For biomass production, except for that of fungi, 25 mL of 24h cultures were transferred into centrifugal tubes and spun at lCfOOOxg- for 15 min. The precipitates were resuspended into 0.85% saline solution and recentrifuged. The precipitates were transferred into pre-weighed porcelain dishes, dried at 105°C for 24h and weighed. For fungi, the fungal cells were collected from 24h old cultures, filtered on a pre:weighed Whatman no 1 filter paper, dried at 105°C for 24 h. and weighed. The weight of the biomass was corrected for that of blanks (respective original liquid media) treated in a similar manner and expressed as mg/100 mL liquid media. No attempt was made to determine the amount of the suspended solids broken down 103 and utilised by the microorganisms during growth. For each pure culture or culture from food, the peptones were ranked based on biomass production where rank 1 was for peptones producing the highest amount of biomass in 100 mL, while 5 was for the lowest amount, and the ranking was totaled for each peptone. 4.3.2 Analytical 4.3.2.1 Proximate composition The proximate composition of test peptones was determined by analysis for moisture, crude protein, ash and fat (Sections 3.3.2.1.1 - 3.3.2.1.3). Fat content of the ray viscera peptones was analysed by mixing 1 volume of the peptones with 4 volumes of diethyl ether, then filtering. A same amount of solvent was added to the residue, mixed and filtered. The extraction was conducted 4 times. The diethyl ether extracts were pooled, contained in a pre-weighed round flask and evaporated to constant weight. The specific weight of the ray viscera peptones was measured using a pycnometer at 25°C. 4.3.2.2 Amino acid composition Amino acid composition was determined using high speed liquid chromatography (Shimadzu - Du Pont LC-1), while acid (HC1 6N) hydrolysis was applied for sample preparation. 4.3.2.3 Minerals Minerals, including Ca, Cu, Fe, Mg, Mn, Zn, Pb and Co, were analysed in the ash using atomic absorption spectrophotometry (Pye, Unicam) and P was measured using the molybdenum blue method (Kirk and Sawyer 1991, James 1995). 104 4.3.3 Statistical analysis Statistical analysis was performed for results from each test microorganism or mixed population of microorganisms in a one way analysis of variance at p<0.05 (Montgomery 1991) using a statistical computer package (Statistica Rel. 5, StatSoft, Inc.) in which peptones were treated as fixed factors, followed by post-hoc analysis by least significant difference (LSD) method at p<0.01 if the effect was significant. 4.4 Results 4.4.1 Protein concentration and proximate composition The results of protein concentration of silages are presented in Table 4.4 which shows that protein levels were concentrated 0.6 to 5.7 fold. Vacuum evaporation at 50°C for 25 min resulted in the highest concentration of protein (260- 366 mg/mL), and reduced the volume of silage liquid by 82-86%. Based on the protein content, vacuum evaporation was used for concentrating the silage supernatant for use as ray viscera peptone. The concentrated supernatant was a very viscous solution with a specific weight of 1.2 g/mL, and brownish amber colour (Figure 4.1), while the pH was 3.5 and 5.0 for FSHCL and FSPF, respectively. The proximate composition and the amino acid profile of the ray viscera peptones and commercial peptones are shown in Tables 4.5 and 4.6, respectively. 4.4.2 Growth curves The colour of liquid media (yellowish) after sterilisation is shown in Figure 4.2. At 600 nm, the absorbance of such media were 0.002, 0.008, 0.046, 0.018 and 0.005 for liquid media containing peptones from Difco, Oxoid, FSHCL, FSPF and BBL, respectively. 105 A B Figure 4.1. Silage supernatant after concentration (A = FSHCL and B = FSPF) The growth curves of microorganisms are shown in Figure 4.3. In general, the food microorganisms showed similar growth patterns on liquid media containing different peptones. For microorganisms from beef and milk, the lag phases for growth were shorter than for microorganisms from egg, i.e. 3 h, compared to 9 h. The exponential growth phase for all microorganisms was 6h, commencing at 3h and finishing at 9h for those from milk and beef, and for those from egg commencing at 9h and finishing at 15h (Table 4.7). After this time, the growth leveled off. For pure cultures, the growth patterns for the test microorganisms shows were similar, except for S. cerevisiae, (Figure 4.4). The exponential growth phase for E. coh, Staph, aureus and B. subtil is began within 1 h of inoculation and lasted for 5h for E. coli and Staph, aureus, and 2h for B. subtilis (Table 4.7). For S. cerevisiae, the exponential growth rate did not begin until 6h after inoculation, and lasted for 18h. 106 *s C — Table 4.4. Protein concentration o C/5 2 "U T3 r\ ■r~ £ 8s _J " ^ -o (0 E X © TO © o) © C co O cz c= “ £ — E E ” 1 ^ _l .E "O ’ ^ -S ■*-» i_ X © C 0) © c E O) E x o X 05 CL U_ X 1 CM in in o T3 o o in CD CD o n © w 0 C $ TO w E o co X o ra © c - ’ o U_ X 05 o _ cl ■§ J5 CM in in o O ■o "O CD cri r-- CO CO CD to o o CD o © c w co 1 E TO > X 33 c © C o c $ CO 0) co o © to j ’ ’ CM ° CL in T- | 1 .2 CM in ? ■O c/5 x: o CO ■o “ CN •4 o in CD 1 ^ 2 'T. 0 © c D 1 c C CL (0 2 CO — ‘ » ’ FSPF o° . O O o £ ° £ c **- T - --- £ x) ^ I ^ to -*-> & E — c O ^ c "b =5 _0 5 X C CO TO $ CO to o 0 Q X 05 X o in CL 'jE T3 05 6 CM in O O ■§ T3 TO T3 CO (N o CD O > © $ d) 2 X c w o c <0 C d) X CO S -- ’ s o H o ■jE O O in o *♦ T3 CM o to © D) 5 c o 0 c 0 ra O) © c — — * - ■O to . 5 X c $ 0 o 0 ra 05 (/) o — • CM ■I) o o o O ■5 £ in CD C ••*= CD ^o° T3 00 -o to CM co in CD © O 2 ©^ © © © CO CL C w =i 0 c $ CO (/) CO CO c l ’ ’ FSPF cm g 'T ^ ,2> -C ^ +-> $ 0 > © £ 00 CM co X 05 X O 0 o' 0 0 in o _l X N- • ■5 £ (N CD T3 ^o° T3 to o 4 > © © O) O >- 0 © c « 0 =3 0 c $ TO (/) TO © TO O) CO C Q. = ’ CM 0 'T ^ — -C .!? to $ 0 > © £ c\i in O CD O' £ CM in in CD X ■E ll ■o .E ■© ■0 S 0 ra 0 FSPF .ti viz .2 jc r: C W | E © 8 c x © 5 TO TO «) § m © « >, . T- CO in co CD CO o CD o -5 £ X 05 O T3 T3 X _ X ■© T3 to 0 § © O) 0 >, X >. 13 TO © © >N TO O) TO c C TO O O C $ TO TO © J ’ ^ om previous dialysis under the same conditions. 107 Table 4.5. Proximate composition of ray viscera and commercial peptones used in this study Analysis Difco Oxoid BBL FSHCL FSPF Moisture (% db) 1.2 ± 0.3 5.2 ± 1.0 4.0 ±0.0 70.1 ±6.1 65.7 ±0.2 Dry weight (mg/mL) - - - 589 ±21 604 ± 1 Ash (% db) 5.1 ± 1.4 6.2 ±2.0 9.7 ±0.1 5.6 ±0.1 6.0 ±0.2 Total nitrogen (% db) 15.8 ±0.5 15.3 ±0.9 13.2 ±0.2 7.8 ±0.1 8.4 ±0.1 Fat (% db) 0.4 ±0.1 0.3 ±0.1 0.1 ± 0.0 0.2 ± 0.0 0.1 ± 0.0 Minerals (mg/100g, db) Ca 63.3 ± 8.3 77.2 ±4.9 52.5 ±2.2 14.6 ±0.0 12.7 ±0.0 Cu 2.3 ±0.1 2.3 ±0.3 2.0 ±0.1 tr tr Fe 3.3 ±0.0 75.0 ± 1.1 0.2 ±0.0 21.7 ±0.1 81.0 ± 0.1 Mg 61.6 ± 8.6 36.3 ± 1.3 71.8± 1.5 31.2 ± 0.1 30.9 ± 0.1 Mn 0.8 ±0.0 1.8 ±0.4 5.5 ±0.2 tr tr P 25.0 ± 4.8 30.9 ± 1.3 9.3 ± 0.3 4.3 ±0.2 4.3 ± 0.4 Pb 2.1 ±0.0 1.3 ±0.2 nd tr tr Sn 1.5 ± 0.0 1.8 ± 0.4 nd tr tr Zn 4.0 ± 0.1 2.8 ± 0.0 1.7 ±0.1 tr tr Values are means ± standard deviations of 2 determinations tr = trace, denotes value less than 0.1 mg/lOOg nd = not detected Table 4.6. Amino acid composition (g/lOOg protein) of ray viscera and commercial peptones used in this study Amino acid Difco Oxoid BBL FSHCL FSPF Alanine 7.7 ± 0.4 7.0 ± 1.2 5.1 ±0.7 6.9 ±0.3 5.6 ±0.1 Ammonia 0.3 ±0.0 0.7 ±0.0 0.4 ±0.1 0.4 ±0.0 0.6 ±0.0 Arginine 8.6 ± 0.0 6.2 ±1.1 5.7 ±0.6 6.7 ±0.3 6.3 ±0.1 Aspartic acid 7.7 ±0.1 7.5 ±1.1 6.8 ±0.5 7.2 ±0.3 7.4 ±0.1 Cystine 0.2 ±0.0 0.5 ±0.0 1.1 ±0.2 1.4 ± 0.1 1.9 ± 0.1 Glutamic acid 12.1 ±0.5 11.6 ± 1.3 10.3± 0.8 14.0 ±0.6 14.5 ±0.3 Glycine 16.4 ±0.2 5.6 ±0.1 7.3 ±0.5 9.8 ±0.4 6.9 ±0.1 Histidine 0.8 ±0.0 1.5 ±0.2 1.9 ± 0.3 3.2 ± 0.1 2.9 ±0.1 Isoleucine 1.8 ± 0.1 2.4 ±0.5 2.8 ±0.6 3.0 ±0.1 2.9 ±0.1 Leucine 4.4 ±0.3 5.0 ± 1.3 5.9 ±0.6 5.5 ±0.3 5.4 ±0.1 Lysine 4.7 ±0.1 5.8 ±1.1 5.6 ±0.8 5.7 ± 0.2 4.5 ±0.1 Methionine 0.8 ±0.0 1.4 ±0.0 1.5 ± 0.3 2.0 ± 0.1 2.3 ±0.0 Phenylalanine 2.6 ±0.6 2.3 ±0.5 . 3.3 ±0.1 3.9 ±0.2 3.8 ±0.1 Serine 3.9 ±0.1 2.9 ±0.9 3.4 ±0.6 3.7 ±0.1 3.8 ±0.1 Threonine 2.5 ±0.0 2.6 ±0.1 3.1 ±0.6 4.0 ±0.2 4.3 ±0.1 Tyrosine 0.9 ±0.3 1.8 ± 0.7 1.8 ± 0.1 2.3 ±0.1 2.5 ±0.0 Valine 2.8 ±0.1 3.6 ± 0.5 4.1 ± 0.5 4.3 ±0.2 4.5 ±0.1 Values are means ± standard deviation of 2 determinations Tryptophan can not be detected by the method used in this experiment. 108 Table 4.7 shows that for microorganisms from food, the growth rate is in the range of 0.2 to 0.3 log A (600 nm) per h. No significant difference was revealed in the growth rate of egg and milk microorganisms grown on the test peptones. The growth rate of beef microorganisms grown on ray viscera peptones, however, was significantly higher than that for commercial peptones. Both FSHCL and FSPF ranked 1 in supporting growth of beef microorganisms, 5 and 2 for egg microorganisms, and 2 and 4 for milk microorganisms. Total summed scores were, Difco 5 (ranked 1), FSPF 7 (ranked 2), FSHCL 8 (ranked 3), BBL 9 (ranked 4) and Oxoid 11 (ranked 5). A 8 Figure 4.2. Typical sterilised liquid media containing 0.5% (w/v) peptones (A = Difco; B = Oxoid; C = FSHCL; D = FSPF and E = BBL) 109 BBL V FSPF □ FSHCL ♦ O x o id ▲ D if c o uju 009 aoueqjosqv 110 Absorbance at 600 nm 0.001 Figure 0 01 . O 4.4. on Difco liquid Growth media ▲ of S. B. B. aureus subtilis Oxoid containing subtilis Incubation ♦ , E. 0.5% FSHCL coli , (w/v) time Staph, (h) D^SPF different aureus peptones and V S. S. E. cerevisiae BBL coli cerevisiae 111 Table 4.7 G row th rate* o f microorganisms (mixed population and pure cultures) on liq u id m edia** contaii £2 different peptones UJ ro £ _c -2 1 < {g CD I E 05 d 1 co c Qj o q 0 "O " to 05 in ■M- CO CO CO d CO cn CM d q CO co CM d q q t CD o in d q co d CM in CO CO d in o +i +i +i +i +1 05 0) - -N ’ CO in LU 05 05 CM co CO d CO d q 05 d CM CD d q CO d CM C\l d o cm O CM d ^ ~ -X O CM CO o o CM +i +i +1 +i +1 I CM c t c O O 05 CO n - co CO CM d q o co d q O M- CM CM LU d q o 1 +i +i +i ~ o d o CM d q CO in o O o 00 -C co d q +l +i 8? CO * 3 3 a to ro d ■M" CO d q m CO d M" CO d q co d O CM CM d d o CQ -§ § .CO M- d q +i +i +1 +i ” ro d CM d q CM q o P d 00 ■?2 CM o ° +i +1 +i Cj 03 03 <2 03 > ■<- o T_ co i- CM CO 2 C CM CM H •M CO O 03 2 c ro O - •g I -3 ■8 t i 1 ■- ! - I .1 ! a S' § g | a § S *g ^ c 2 8 ” • -2 i Zco> ■= ! I | f= ^ * » s i'll- 3 to i P r £ 2 3 — ■ 'o -5 11 g 8 3 | i S3, Ss a> § 8 g g — 'S ••3 3 l 1 S k « S * a; oo C3 Si § CO 5 1 2 | C3 | = 112 The growth rate of pure cultures ranged from 0.1 log A (600nm) per h for S. cerevisiae on all peptones to 0.5 log A (600nm) per h for B. subtilis on FSHCL and FSPF, respectively. Significant differences in growth rate were revealed among the test peptones for E. coli, Staph, aureus and B. subtilis. The growth rates of S. cerevisiae on liquid media containing test peptones were not significantly different. The growth rates of pure cultures in liquid media containing peptones from cowtail ray viscera silage were always higher than those on commercial peptones. As shown in Table 4.7, the scores for FSHCL were 4 (rank 1), FSPF 9 (rank 2), Difco 10 (rank 3), Oxoid 12 (rank 4) and BBL 18 (rank 5). If the growth rate rankings of food microorganisms and pure cultures were totaled, it can be seen the order of ranking is FSHCL, Difco, FSPF, Oxoid and BBL. The total growth of food microorganisms and pure cultures is shown in Table 4.8. For food microorganisms, the total growth ranges from 0.6 A (600 nm) for egg, beef and milk microorganisms grown on Difco peptones to 1.2 A (600 nm) for beef and egg microorganisms grown on FSPF peptones. The total growth of food microorganisms grown in liquid media containing FSPF and FSHCL was significantly higher that those on commercial peptones. This is also shown in Figure 4.2, where in general the absorbancies of food microorganisms on FSHCL and FSPF were always higher than those on other peptones. Based on the score, FSPF and FSHCL ranked first and second, followed by BBL, Oxoid and Difco, with having scores of 3, 6, 10, 11 and 15. Statistically, within the same pure culture, a significant difference in total growth was revealed among the test peptones. The total growth of microorganisms in FSHCL and FSPF was always significantly different from that for the other peptones. For FSHCL, the total growth ranked 2, 1, 3 and 1 for E. coli, Staph, aureus, B. subitilis and S. cerevisiae, respectively, while for FSPF, the ranking was 1, 113 2, 2 and 3 for the same microorganisms, respectively. If the rankings were summed, FSHCL obtained a score of 7 (ranked 1), FSPF 8 (ranked 2) and BBL 11 (ranked 3). Difco and Oxoid had a score of 17 and ranked equal 4. The total score for test peptones in total growth was 11 for FSPF (ranked 1), 13 for FSHCL (ranked 2), 21 for BBL (ranked 3), 28 for Oxoid (ranked 4) and 32 for Difco (ranked 5). 4.4.3 Biomass production Table 4.9 shows the dry biomass produced by food microorganisms and pure cultures grown on different peptones. For food microorganisms, the biomass ranged from 19.1 mg/100 mL to 171.4 mg/100 mL, which were produced by milk microorganisms grown on liquid media containing Difco peptone and FSHCL, respectively. Statistically, the test peptones significantly affected the biomass production by food microorganisms. The biomass production of food microorganisms grown in liquid media containing ray viscera peptones was significantly higher than for those grown in commercial peptones, of which FSHCL and FSPF ranked either 1 or 2. The score for each peptone in biomass production is 4 (ranked 1) for FSHCL, 5 (ranked 2) for FSPF, 10 (ranked equal 3) for Oxoid and BBL, and 15 (ranked 5) for Difco. For pure cultures, the biomass production ranged from 25.3 mg/lOOmL for Staph, aureus on BBL peptone to 257.5 mg/100 mL for S. cerevisiae on FSHCL peptone. Statistically, for the same microorganism, there were significant differences in biomass production due to the test peptones. E. coli, Staph, aureus and S. cerevisiae produced higher levels of biomass on FSHCL and FSPF than on commercial peptones. For A. flavus, FSHCL and FSPF was not able to support the production of biomass at levels as high as for the other peptones, and were ranked 4 114 and 5. Similarly, FSPF and FSHCL, although able to support the growth of B. subtilis, were ranked 3 and 4 respectively. If the scores were summed, the scores were 12, 13, 16, 17 and 18 for FSPF, FSFICL, BBL, Oxoid and Difco, respectively. The total scores for biomass production by all microorganisms are 16 (ranked 1) for FSPF, 17 (ranked 2) for FSHCL, 26 (ranked 3) for BBL, 27 (ranked 4) for Oxoid, and 33 (ranked 5) for Difco. 4.5 Discussion 4.5.1 Proximate composition Some of the characteristics of crude ray viscera peptones (FSHCL and FSPF) in this experiment, such as dry weight and specific weight, were comparable to those obtained for cod (Gcidus morhuci) using 3% hydrochloric acid (Clausen et al. 1985). The colour of the concentrated supernatants was also similar to those from cod viscera silage, which was also amber. As shown in Table 4.5, the dry weights of FSHCL and FSPF were 589 and 604 mg/mL respectively, compared to 570 mg/mL for cod viscera crude peptone. The moisture contents of these crude peptones were 41.1%, 39.6% and 43%, respectively. The specific weights were almost the same, i.e. 1.23, 1.22 and 1.19 mg/mL for FSHCL, FSPF and cod viscera peptones. Similarly, the ammonia contents were also comparable (1.4, 2.3 and 2.0 mg/mL, respectively). The ash contents of FSHCL and FSPF, however, were much lower than for the other peptones. Compared to that of cod viscera, the ash contents of FSCHL and FSPF (42 and 46 mg/mL) were 60-65% of that of cod viscera crude peptone (71 mg/mL) as reported by Clausen et al. (1985). Compared to Difco and Oxoid used in the present study, however, the ash contents of FSHCL and FSPF were similar in amount, but lower than that of BBL. 115 H jj OS Source of ______Total growth (absorbance at 600 nm) on liquid media containing peptones fr< :roorganisms Difco Oxoid BBL FSHCL FSPF q O QJ 3 -a <1) CD o m m d CD CD d o co d o co o m CO m d co in d CD CD d CD CD d d o CM o m o 0) CD -H +i -H -H -H A UJ w d oo M" d o o d o CD Tt CO d d Ln d r- m _ O CM o m -H 00 o O) O) -H H-H -H -H „ n JC ^ i= in 00 Tf d d O CM s „ o in cn o un d o co co d in d t"- co CD d o CM CM +1 -H +i — _ „ ’ co s-/ o CD < 8 c as d a. 2 CD o a) 3 o 1 m 00 O d M" d CO , d in d o CM d o co r- CM d in ■ CO rt d m o d O -H m -H -H -H _ -H - N o -C co d CD O 2 d CD M- d CM in d r^- 00 d o Tf CD co co , d o CM rt d CD d o ■M" -H Ct3 £ -H a. co -H -H _ s ■o OQ d o CD 00 ,, d o CD cp co d o CM ">t T d T o CO m d r-~ CD d d o CM CM d o co CD ■M" +1 -H •H — +i — — „ T3 CO m r- d m ' CM •3- o CM , CO CM CM CD OO St co m d o OO CO CM r~- CM d O +H -H +i — +1 - ' x CO T co h- 8 0) 03 c — 5 co. H CM CM oo CO CO W CO O O 2 c CD •O j2 .£ VO ^ < JS "g" "3 Co co CO X3 •I "3 <2 •8 ■> 'is <2 3 ro -H 3 n 3 o a C o 1) 00 £ OJ I c i'-s £ ” i 1 2 O S3 u 00 C - * 2 | | ■ G j § 1 d '55 1 IS o , -* — i A 1 a- o — > 1 116 H 3 JJ cz o CJ 6/5 Source of ______Biomass production on media containing peptones from______:roorganisms Difco______Oxoid BBL FSHCL FSPF _cn TD > To 4 O CD o E — 8 ■a ’ lu o m ^ O Tf Tf 05 o) -H to CM o co cm uo - h ^ i ^ in CO CO i- s CD UO t o CT> t — co a> +« CD CD -H co co CM ss - -H - C/3 C- UO o o E co c < o r- CM co r- CD oo co co cm s CO CO m +) CM co CO co UOM^CM- CO o > CO -H CO co M" M; h- M- -H ■H -H .O CQ -Q uo r- co M- co CD O o CD CO m o -H -H M O CM M" co -H n r*» in CD T- cm CM o T- CD - -H Hi o m co r- CM CD M- O M- +) CO r- Tf n -I- o O co v co o co +1 -H co -C -2 o Tt CO 0- uo co uo CO UO CM oo co m Q. o t m CD O 3 -H t- CO £ -H -H CD 3 co 3 cm m CM CO CO r t -H +1 — — - ’ o •2 m CM h- CD o ^2 M" CD 00 M" co O ni CO ^ CrT ^ « M- J*: ” ^ FT ^ ^ ^ o o E CO c I- g 75 r- CM CO CO 0- CM CO CM CD o o E CO c >30!l <2 1 1 • ? w 8 CZ 2 V CO « 123 - § £ :1 2 0 3 .5 f' r- £ s;s c < g § § ' n J .X .Sf£ .S o W-i8 « qj ° i q o S S “ 0 (10 117 Table 4.5 shows that peptones from cowtail ray viscera silage contained less total nitrogen and less minerals than commercial peptones, except for Fe, which was higher in FSPF. FSHCL contained a higher level of Fe than Difco and BBL but lower than the level found in Oxoid. The total amino acid contents of FSHCL and FSPF were lower than that of cod viscera, i.e. 291 and 306 mg/mL compared to 391 mg/mL. FSHCL and FSPF contained 83.6 and 79.4 g amino acids/1 OOg protein, respectively. These values are higher than for the commercial peptones used in the present study, i.e. Difco, Oxoid and BB1 which have total amino acid contents of 77.6, 67.7 and 70.0 g/lOOg protein, respectively. The lipid contents of FSHCL and FSPF were very low (0.04 and 0.1% wb) since the lipid was removed during centrifugation. The amino acid composition shown in Table 4.6 indicates that at least 16 amino acids were present in crude peptones from cowtail ray viscera silage. Some of the amino acids in ray viscera peptones were at levels higher than those in commercial peptones used in the present study. The amino acids of ray viscera peptones which were present at higher levels than those of commercial peptones were cystine, glutamic acid, histidine, isoleucine, methionine, phenylalanine, serine (lower than Difco), leucine (lower than BBL) threonine, tyrosine and valine. For the 9 essential amino acids, only tryptophan was not detected. This was due to the preparation of samples used in the analytical method. The amino acid at the highest level in FSHCL and FSPF was glutamic acid, while the lowest content amino acid was cystine. Similar results were also found in viscera silage of cod and saithe (Raa and Gildberg 1976, Gildberg and Raa 1977, Strom and Eggum 1981), salmon (Dong et a/. 1991) and catfish (Verburg and Freeman 1984). Accordingly, it can be considered that cystine is the limiting amino 118 acid in fish viscera silage supernatant. Raa and Gildberg (1976) reported that cystine and aromatic amino acids were concentrated in the sediment of cod viscera silage. Compared to viscera from Indonesian tropical fish such as red snappers (Liitjcinus spp.), scad (Caranx spp.), king mackerel (Scomberomorous sp.) and shad (CInpea sp.), some amino acids in FSHCL and FSPF were present at higher concentration (Sukarsa 1978a). Any differences in proximate composition of the test peptones in the present study were due to difference in the raw materials and methods of preparation. The commercial peptones used in the present study were produced from meat and other animal tissues (no details provided) by enzymic hydrolysis (Anon. 1973b, Power and McCuen 1988, Bridson 1990). 4.5.2 Growth of mixed population of microorganisms As shown in Figure 4.3, egg microorganisms had a longer (9h) lag and acceleration phase than those for beef and milk microorganisms (3h). This was probably due to the conditions under which these foods were obtained. Eggs were obtained fresh and intact in the shell and should have been sterile. However, if the eggs in this experiment contained microorganisms due to contamination during transport or poor handling (e g. washing), it is likely that the numbers of microorganisms was low due to physical protection by the shell, as well as the high pH of the albumen. Moreover, eggs contain antimicrobial substances such as ovotransferrin, lysozyme, avidin, ovoflavoprotein and ovoinhibitor in the albumen (Adams and Moss 1995). Beef was obtained cut into small portions from open display, while milk was fresh, cooled, contained in plastic pouches, and neither pasteurised nor homogenised. Microbial contamination very likely took place during handling and/or 119 transport of these materials and growth would occur due to the tropical temperatures. Thus when inoculated, beef and milk may have higher microbial content than egg. Once the exponential growth phase was reached, the food microorganisms grew well on the test peptones. This was indicated by the maximum growth rate during the exponential growth phase which showed rates between 0.21 to 0.39 logA/h. The average growth rate of food microorganisms on the test peptones was 0.28, 0.27 and 0.28 logA/h, for egg, milk and beef microorganisms, respectively, indicating that during this phase there was no abnormal growth. This was also supported by the length of the exponential growth phase, which was 6h for all food microorganisms on the test peptones. On average, little differences were revealed in the growth rate of food microorganisms grown on the peptones tested. The total growth of food microorganisms after 36h, however, was different among the test peptones. Higher total growth of food microorganisms was found for those grown on ray viscera peptones. Similar results were shown for the dry biomass, in which FSHCL and FSPF produced significantly higher amounts of dry biomass compared to commercial peptones. The amino acid composition might be partly the reason why FSHCL and FSPF gave better results in total growth as well as dry biomass production. From Table 4.6 it can be seen that of the 16 amino acids measured in the test peptones, FSHCL and FSPF contained 11 amino acids (6 of which are essential amino acids) that were at higher concentrations than the other peptones. There are few studies in the literature using mixed populations of food microorganisms that can be compared to the results of the present study. A study by Green et a/. (1973) showed that peptones from menhaden autolytic hydrolysate did not compete well with 2 commercial peptones (Difco Bacto-peptone, and BBL Myosate) in supporting milk microorganisms on agar growth. Other fish peptones 120 tested by Green et al. (1973), including enzymic (Alcalase) and pancreatic digests of menhaden and hake muscle, and soluble fish extract from menhaden stickwater meal and oil, were able to support growth of milk and hamburger microorganisms to a level comparable to the two commercial peptones. Although the results of Green et al. (1973) and the present study can not be directly compared, they demonstrate that fish waste peptones are more capable of supporting the growth of food microorganisms than commercial products. 4.5.3 Growth of single microorganisms The growth pattern of B. subtilis, E. coli, Staph, aureus and S. cerevisiae in this study as shown in Figure 4.4 reflected a normal pattern of microbial growth. The exponential growth phase of these microorganisms, except that for S. cerevisiae, began lh after inoculation and lasted for 5h. A similar exponential growth phase was shown by Vecht-Lifshitz et al. (1990) who tested the ability of fish viscera peptones, prepared by a similar method to the present study, to support the growth of several microorganisms including E. coli and Staph, aureus. The exponential growth phase of both microorganisms in the latter study began after 3h incubation and lasted also for 5h. The growth rate of microorganisms grown on FSHCL and FSPF was higher than or similar to those grown on commercial peptones. The differences in growth rate of E. coli and Staph, aureus in this study, however, were not as large as that in the study of Vecht-Lifshitz et al. (1990). In the latter study it was reported that the growth rates of E. coli and Staph, aureus grown on fish peptones were 4 and 2 times higher, respectively, than on bactopeptone (Difco). In the present study, the growth rate of both bacteria on fish viscera peptones was between 1-1.3 fold higher than on bactopeptone (Difco). Green et al. (1977) and Jassim et al. (1988), however, 121 indicated that the growth rate of Staph, aureus on fish waste peptones was similar to that on commercial peptones (Oxoid Proteose peptone, Difco Bactopeptone and BBL Myosate). The results of the above studies, however, were in contrast to the results of Krishnaswamy and Lahiry (1962) who demonstrated that the growth rate of E. colt and B. subtilis on liquid media containing fish hydrolysate was lower than that on commercial peptones (the brands were not disclosed). These different results were probably due to the differences in microorganism strains and/or the composition of the liquid media as well as the procedures used. Vecht-Lifshitz et al. (1990) used more complex media formulation using beef and yeast extracts as nitrogen sources in addition to fish peptone, while much simpler liquid media formulations were used in the present study. The total growth of microorganisms in the present study indicated that, in general, FSHCL and FSPF were better than for commercial peptones in supporting growth. For E. coli the total growth on ray viscera peptones was 2.6 - 2.7 times higher than on bactopeptone (Difco), and was higher than that (2.3 times) reported by Vecht-Lifshitz et al. (1990). For Staph, aureus, the total growth on fish peptones was 1.4 times higher than on bactopeptone, and this was lower (3.2 times) than that reported by Vecht-Lifshitz et al. (1990). Different procedures as well as microorganism strains might be responsible for these different results. The total growth of these two bacteria in the study by Vecht-Lifshitz et al . (1990) was obtained from 48h cultures, while the present study used 36h cultures. The average amount of dry biomass produced by the test microorganisms grown on FSHCL and FSPF was higher than that on commercial peptones, a result which was also shown by several workers. Beuchat (1974) tested the growth of 13 microorganisms (5 of which were similar to those in the present study, although he 122 used A. oryzcie instead of A. flavus) on 7 different peptones, including bactopeptone (Difco), thiotone peptones (BBL) and catfish waste (head and skin) peptones. The dry biomass produced by the 5 microorganisms on catfish waste peptones was lower than that grown on BBL, but higher than for bactopeptone. A similar result was also demonstrated by Stephens et al. (1975), where microorganisms grown on catfish peptones produced lower dry biomass than on BBL but higher than for bactopeptone. In the latter study, however, the catfish peptones produced a lower level of dry biomass by the 5 microorganisms compared to that produced on prawn head and hull peptones. The higher mineral content in the prawn waste peptones could be one of the reasons, since some minerals may act as growth factors. The ash content of prawn waste peptones was 14-33% (wb) compared to 10% (wb) for catfish waste peptones. FSHCL and FSPF in the present study produced the lowest dry biomass of fungus (A. flavus) than other peptones. This was probably due to the fat content of both peptones which was very low (0.04 - 0.09% wb), as fat might be a stimulant to increase fungal yield (Green 1974). Similar results were also noticed by Sophanodora and Buckle (1988), where peptones containing higher fat content tended to produce higher fungal (Rhizopus oligosporus) dry biomass. Beuchat (1974) and Stephens et al. (1975), however, noted that fungal biomass production was not always correlated with higher fat content of prawn waste peptones. According to Garraway and Evans (1984), fats do not always act as growth factors for fungi. 4.5.4 General discussion Peptones from fish viscera in the present study have been shown to be able to support the growth of microorganisms, either in mixed populations such as those from a variety of foods, or individually. Although the ability of these peptones to support the growth of individual samples of mixed populations or single microorganism was 123 not always better than those of commercial peptones, in general the viscera peptones ranked 1 or 2, judged from total growth and dry biomass production. The ability of viscera peptones in the present study to support the growth of microorganisms to a greater extent than commercial peptones was partly due to the composition of fish peptones. Total amino acid contents of viscera peptones were higher than that of commercial peptones. Some essential amino acids were also at higher levels in these viscera peptones, and were obviously sufficient to supply nitrogen. Staph, aureus , for example, needs up to 11 amino acids in a minimal medium (Bryan 1976). Copper and zinc were present at trace levels in viscera peptones, and this may be advantageous since copper and zinc at higher levels can possess antimicrobial activities (Foye 1977). According to Traxler and Lankford (1957) antibacterial activity of certain peptones might be caused by colloidal sulphur arising from cystine in a reaction catalysed by copper. The low content of fat in viscera peptones is also advantageous since fatty acids can have an inhibitory effect on microbial growth (Nieman 1954, Guirard and Snell 1962). The preparation of ray viscera peptones in the present study involved ensilation using hydrochloric acid and a mixture of propionic and formic acids. The use of organic acids is not recommended since they have an inhibitory effect upon many bacteria and moulds (Gildberg and Almas 1986). However, the present study has shown that organic acids at a level of 3% (w/v) for preparation of the ray viscera silage peptones did not negatively affect the growth of microorganisms. This supports the results of Vecht-Lifshitz et a/. (1990) who had demonstrated the ability of peptones, prepared from cod viscera using a mixture of propionic and formic acids at a level of 1.5% (w/v) in a way similar to the present study, to support the growth of microorganisms. Propionic and formic acids have inhibitory effects on moulds and bacteria at low pH, e g. pH 4-5 or less (Gildberg and Raa 1977, Strom et 124 ciL 1980, Raa and Gildberg 1982), while in the present study, the crude peptones produced using both acids was used at neutral pH, thus the inhibitory effects of those acids would not be as effective. The present study has shown the potential of fish waste peptones to compete with commercial peptones such as those from Difco, Oxoid and BBL. The price of these peptones in August 1996 in Sydney, Australia is AUD 208, 142 and 256 per kg for Difco, Oxoid and BBL, respectively, and these costs can be doubled in Southeast Asian countries such as Indonesia. The wastes at fish processing centres such as Muara Angke, Jakarta and Labuhan, Maringgai, Lampung, can be obtained almost at no cost. Thus the conversion of such wastes into peptones could not only reduce some environmental problems in the area, but also produce a supply of microbiological peptones that may be cheaper than imported ones. 125 5 ISOLATION AND CHARACTERISATION OF COWTAIL RAY (T. sephen) VISCERA PROTEASES 5.1 Introduction Proteases are the most important industrial enzymes and account for up to 60% of total usage, especially in the detergent, leather and food industries (Godfrey and Reichelt 1983, Knorr and Sinskey 1985). In the leather industry, the first use of a protease was reported in 1908 (Cheetam 1995), while the first commercially available enzyme (calf rennet) preparation appeared in 1874 (Adler-Nissen 1993). However, proteases have been involved in indigenous food processes in many countries for centuries, such as fish sauce processing in some Asian countries or cheese processing in European countries. Industrial enzymes used in the food industry are usually from plants, by-products from pig and cattle processing and microbial sources (Haard et al. 1982, Garcia-Carreno and Haard 1993). Proteases from aquatic organisms, especially fish, have attracted considerable interest from both fundamental and applied aspects. The marine environment is species diverse and the unique habitat conditions may have selected for particular properties of protease enzymes. The thermal denaturation temperature of fish pepsins from cold water environments have been reported to be about 20°C lower than mammalian pepsins, and unlike mammalian trypsin, fish trypsins were found to be unstable at acidic conditions (Haard 1992b). This has made the aquatic environment a potential source of enzymes with unique physical, chemical and catalytic properties. Endogenous proteases have been linked to the deterioration of fresh fish and fish products (such as deterioration of krill, belly burst, texture softening and loss of elasticity in kamaboko), blackspot development and meat degradation in crustaceans, 126 the development of desirable attributes of fish flesh such as flavour, and unique texture-forming properties of fish minces (Haard et ah 1982, Dionysius el ah 1993, Stoknes, Rustad and Mohr 1993, Kolodziejska and Sikorsksi 1995, Zotos and Taylor 1996). Visceral proteolytic enzymes are also important as agents of proteolysis and flavour development in fermented fish products such as fish sauce (bekasang in northern part of Indonesia), cured herring, matjes (salted herring) and fermented squid (Orejana and Liston 1981, Lee, Simpson and Haard 1982, Simpson and Haard 1984a, Raksakulthai, Lee and Haard 1986, Stefansson and Steingrimsdottir 1990, Ijong and Ohta 1995). Digestive enzymes from fish have been the subject of study by many workers. Stirling (1884) showed enzyme activity upon fibrin of cod and herring stomachs in the presence of HC1. The first study on the purification and characterisation of digestive proteolytic enzymes, i.e. salmon pepsin, was reported by Norris and Elam (1940), and Sprissler (1942) investigated the proteinase (pepsinogen) of the gastric mucosa of shark. Since then, many reports on the isolation and characterisation of fish digestive enzymes have appeared in the literature. Digestive proteases have been isolated and characterised from cold water fish such as Greenland, Atlantic and Polar cod (Simpson and Haard 1984b,c, Arunchalam and Haard 1985, Squires, Haard and Feltham 1986a,b, Asgeirsson, Fox and Bjarnason 1989, Simpson el ah 1989, Simpson, Simpson and Haard 1990, Amiza and Apenten 1994, 1996), arctic capelin (Gildberg and Raa 1983) and orange roughy (Xu et ah 1996). Fish from subtropical waters of which the digestive proteolytic enzymes have been isolated and characterised include anchovy (Heu et ah 1991), sardines (Murakami and Noda 1981, Noda and Murakami 1981, Noda et a/. 1982), mackerel (Ooshiro 1968, Kim and Pyeun 1986, Pyeun and Kim 1986,) and 127 skipjack tuna (Pyeun, Kim and Heu 1988), all of which were from Japanese and Korean waters. Other marine organisms of which the digestive enzymes have been investigated include shrimp (Jiang, Moody and Chen 1991), lobster (Galgani and Nagayama 1987, Zotos and Taylor 1996), crayfish (Garcia-Carreno and Haard 1993, Kim, Meyers and Godber 1996), oyster (Tsao and Nagayama 1991), starfish (Bundy and Gustafson 1973, Williams 1975), crab (Dionysius el al. 1993), seastar (Farrand and Williams 1988) and marine mammals including fin and sperm w7hales (Sasano and Ota 1964) and harp seal (Shamsuzzaman and Haard 1984, Haard 1992b). The enzymes from cold water fish usually exhibit a low Arrhenius activation energy, low temperature optimum, low thermal stability and high pH optimum compared to homologous enzymes from warm-blooded animals (Simpson and Haard 1985, Gildberg 1988). Fish digestive proteases that have been isolated are alkaline and acidic proteases such as trypsin- and chymotrypsin-like, pepsin-like and carboxypeptidase enzymes. Gildberg (1988) has reviewed the molecular properties and physiological role of aspartic proteinases which include pepsins and cathepsin D in fishes and aquatic invertebrates. The uses of these enzymes in food processing have been investigated (Shamsuzzaman and Haard 1983, Yoshinaka et aJ. 1983, Brewer, Helbig and Haard 1984, Simpson and Haard 1984a, Shamsuzzaman and Haard 1985, Stefansson 1988, Stefansson and Steingrimsdottir 1990, Haard 1992b). Fish pepsins have been industrially produced and used for the production of salmon and orange roughy caviar in Scandinavia, Canada, Japan and New Zealand (Almas 1990, Xu el aJ. 1996). Trypsin (E C. 3.1.4.21.4) from Atlantic cod viscera has been extracted and commercially marketed as a biochemical reagent. While studies on fish digestive enzymes from cold and sub-tropical waters 128 have been reported, studies on tropical fish enzymes, including digestive proteases, especially those from Indonesian fish or viscera, are very limited. Gildberg (1982) investigated the characteristics of digestive proteases of Indian oil sardines (,Sardine Ha longiceps) from the Bali Strait, Indonesia, in comparison with those from Norwegian herring and Oman lanternfish. It was reported that proteases from warm water fish have a higher temperature optimum than those from cold water fish. The digestive proteases from cold water species exhibited slightly higher activity at low temperatures than did those from warm water fish. The optimum activity of acidic and alkaline proteases from both fish was similar, i.e. slightly above pH 3 and between pH 9 and 10. Another study on digestive proteases from Indonesian fish was reported by Sulistiyani and Heruwati (1991). Alkaline proteases were isolated from the pyloric caeca of grouper (Epinephelus tauvina F.), and showed a maximum activity at pH 8.0 at a temperature around 45°C. No tests were carried out to confirm the purity of the enzyme preparation. Alkaline protease activity was also found in the pyloric caeca of Indonesian snakehead, snapper, mackerel and mullet, which at 40°C showed maximum activity at pH 8.6-8.8 (Sukarsa 1978b). The above studies have only examined the properties of digestive enzymes from Indonesian fish in term of their behaviour at different pHs and temperatures on different substrates. No further studies have been reported in the literature regarding the characterisation of such enzymes or other Indonesian fish digestive enzymes, including those from elasmobranch fish such as stingrays. The aims of this study were to isolate and characterise the principal digestive enzymes from Indonesian cowtail ray (T. sepheti). 129 5.2 Materials 5.2.1 Viscera Cowtail ray viscera used in this study were obtained from Labuhan, Maringgai, Lampung, Southeast Sumatra, Indonesia. Proximate composition and treatments applied before enzyme isolation were similar to those outlined in Section 3.2.1. 5.2.2 Chemicals Chemicals used for silage preparation and for proximate composition analysis are as described in Section 3.2.2, while those for enzyme isolation and characterisation are listed in Table 5.1. Table 5.1 Chemicals used in the experiments Chemical Brand and specification Acetic acid Ajax Chemicals, AR N-Acetyl-L-tyrosine-ethyl-ester (ATEE) Sigma Acetone Ajax Chemicals, AR Acrylamide Bio-Rad Ammonium persulphate Bio-Rad Barbitone Ajax Chemicals, AR N-Benzoyl-L-arginine-ethyl-ester (BAEE) Sigma Bis (N,N'-methylene-bis acrylamide) Bio-Rad Boric acid BDH, AR Bromophenol blue Ajax Chemicals, AR Casein (Hammarsten) BDH Calcium chloride Ajax Chemicals, AR Chloroform Ajax Chemicals, AR Citric acid Mallinckrodt, AR L - Cysteine Sigma, pure Diethylaminoethyl (DEAE) cellulose Sigma, AR Dimethylsulphoxide (DMSO) Mallinckrodt, AR Ethylene diaminetetra acetic acid (EDTA) ICN Biomedicals Glycerol Ajax Chemicals, AR Glycine Bio-Rad Hemoglobin Sigma, lyophilised Hippuryl-L-phenylalanine (HPA) ICN Biomedicals Iodoacetic acid Hopkin & Williams Li-Srxi buffer Beckman Instrument 130 Table 5.1 (continued) Chemical Brand and specification (3-Mercaptoethanol Bio-Rad, AR Pepsin (Porcine, EC 3.4.23.1) Sigma Pepstatin A Sigma Phenylmethylsulphonyl fluoride (PMSF) Sigma Potassium dihydrogen phosphate Ajax Chemicals, AR Polyacrylic acid (Carbopol C934) BF-Goodrich (MW 3-5M) Protein markers Sigma Sodium acetate BDH, AR Sodium chloride Jaegar Chemicals Sodium dodecylsulphate (SDS) Bio-Rad Sodium hydroxide Ajax Chemicals Soybean trypsin inhibitor (SBTI) Sigma N,N,N’,N’-Tetramethyl-ethylenediamine (TEMED) ICN Biomedicals N-p-Toluenesulphonyl-L-lysine chloromethyl ketone Sigma (TLCK) Sigma N-/?-Toluenesulphonyl-L-phenylalanine chloromethyl- ketone (TPCK) Sigma, AR and electro Tris(hydroxymethyl) aminomethane (Trizma Base) phoresis grade Sigma, AR Tris(hydroxymethyl) aminomethane hydrochloride (Trizma HC1) Ajax Chemicals, AR Trichloroacetic acid (TCA) 5.2.3 Equipment Equipment used in the study is listed in Table 5.2. Table 5.2 Equipment used in the experiments Equipment______Manufacturer and type Electrophoresis unit Bio-Rad, Protean II xi Centrifuge Beckman Instrument Inc., J2-MC Fraction collector Gilson Medical Electronics, MTDC/V Freeze drier Dynavac pH meter 1. TPS Pty Ltd 2. Hanna instrument, pHep Spectrophotometer Shimadzu, UV-120-02 Waterbaths Thermoline 131 5.3 Methods 5.3.1 Preparatory Unless otherwise indicated, all isolation steps were carried out at 4°C. 5.3.1.1 Enzyme extraction The enzymes of cowtail ray viscera were extracted by ensilation as described in Section 4.3.1. The extraction was conducted at the Research Station for Marine Fisheries (RSMF), Slipi, Jakarta, Indonesia. The acids used to ensile the viscera were a mixture of propionic and formic acids (1:1, v/v) at a level of 3% (v/w). The ensilation was conducted at 40°C for 5 days, after which the silage was centrifuged at 10,000xg- for 10 min. The sediment was separated from the silage, while the lipid on the surface was skimmed off. The supernatant was frozen (-45°C) in a chest freezer overnight and air transported to Sydney, Australia in a cool box (approx. 8h). Upon arrival, the supernatant was stored at -25° to -30°C until used. Before use, thawed silage supernatant was again centrifuged (10,000x,g, 10 min) and added to 1/3 volume of ice-cold chloroform to remove the remaining fat. The mixture was left for 15 min, the lower layer was separated using a separating funnel and recentrifuged to remove the remaining chloroform and fat. The lower layer after centrifugation was separated and referred to as crude enzyme extract. 5.3.1.2 Fractionation of alkaline and acidic proteases The alkaline and acidic proteases were fractionated with polyacrylic acid (PAA, MW 3-5M) through a series of preliminary trials. A volume of crude enzyme extract was mixed with 2% (w/v) PAA solution to produce various concentrations (0.0125 to 0.3%) of PAA in the final mixture. The mixtures were stirred for 30 min and then centrifuged (10,000 x g, 10 min). The supernatant (Si) was separated as 132 acidic protease extract. The precipitates were redissolved in 0.4M Tris buffer pH 8.5 with an addition of 1M CaCh to produce 50 mM in the final mixture. The mixture was stirred for 10 min and recentrifuged. The supernatant (S2) was separated as alkaline protease extract. The PAA concentration producing the highest acidic protease activity and lowest alkaline protease activity in Si and lowest acidic protease activity and highest alkaline protease activity in S2 was applied in the next trial. In the second trial, Si was adjusted to pH 4.0, pH 3.5 and pH 3.0, while S2 was adjusted to pH 8.5 and pH 9.0 using 2N HC1 or 2N NaOH, respectively. 5.3.1.3 Acetone precipitation Two volumes of cold (-30°C) acetone were slowly added to SI. The mixture was kept overnight at -30°C. The suspension was then centrifuged (lOOOOxg-, 0°C, 10 min), and the sediment dissolved in 20 mM sodium acetate buffer pH 5.0. The enzyme solution was dialysed against the same buffer for 12h with 2x buffer changes, and recentrifuged (lOOOOxg-, 4°C, 10 min). The supernatants were freeze dried (-45°C, 3.75 kPa). Similar treatment was applied to S2 except that the buffer was 20 mM Tris buffer pH 7 8 containing 5 mM CaCff 5.3.1.4 Ion exchange chromatography The procedure for ion exchange chromatography was adopted from Reece (1988). For acidic proteases, DEAE-cellulose (15-20g) was first washed with 0.5M HC1, then with 0.5M NaOH, and followed with 0.5M HC1. The final step of washing was with 20 mM sodium acetate buffer pH 5.0 which was done several times until the supernatant was clear. The washing solutions were removed each time by siphoning so that floating fine particles were also removed. The suspension of cellulose was gravity packed in a glass column (2.5 x 50 133 cm) to a height of about 30 cm. The column was washed with 20 mM sodium acetate buffer pH 5.0 until the pH of the effluent was identical to that of the applied buffer. The flow rate was adjusted to 1 mL/min. Freeze dried enzymes were redissolved in minimal volumes of 20 mM sodium acetate buffer pH 5.0 and gently loaded onto the column, and the unbound protein was washed with the buffer, while the bound proteases were eluted with 0.2 M acetic acid. Fractions (5 mL) exhibiting absorbance at 280 nm were subjected to protease assays, and those having protease activity were pooled and freeze dried. For alkaline extracts, DEAE cellulose was first washed with 0.5M NaOH, then with 0.5M HC1 and 0.5M NaOH. The remaining procedures were similar to that for acidic proteases, except that the buffer used was 20 mM Tris buffer pH 7.8 containing 5 mM CaCh. Freeze dried samples were redissolved into minimal volumes of 20 mM Tris buffer pH 7.8 containing 5 mM calcium chloride and loaded onto the column. Fractions (5 mL) containing proteolytic activity passing through the column were pooled and freeze dried. 5.3.1.5 Electrophoresis Electrophoresis was done according to the method of Laemmli (1970). The monomer concentration in the separating gel was 11%, while that in the stacking gel was 4%. The gels were run at 10 mA for 16h at ambient temperature (approx. 18°- 20°C). Staining was done using Coomassie Blue R-250 in methanol, acetic acid and distilled water for 30 min, followed by destaining in methanol, acetic acid and distilled water. The molecular weight markers were bovine albumin (MW 66,000D), egg albumin (45,000D), glyceraldehyde-3-phosphate dehydrogenase (36,000D), carbonic anhydrase (29,000D), bovine trypsinogen (24,000D), soybean trypsin inhibitor (20,100D) and oc-lactalbumin (14,200D) (Sigma). 134 5.4 Analytical 5.4.1 General acidic and alkaline proteases assay During purification, the protease activity was monitored by the method of Barret (1972) modified by Martinez, Olsen and Serra (1988). Universal buffers pH 3 and 8 (Johnson and Lindsey 1939) were used as assay buffers and enzyme dilution. Alkaline protease was monitored at pH 8 using 8% casein in 50 mM NaOH, while acidic protease was monitored at pH 3 using 8% hemoglobin in deionised water. To a mixture of 0.5 mL assay buffer and 0.25 mL substrate equilibrated at 25°C was added 0.25 mL enzyme solution. After 60 min incubation at 25°C, 5 mL 6% TCA was added to stop the reaction. Filtration was done after 15 min using Advantec filter paper no. 131 (equivalent to Whatman filter paper no. 3) . The absorbance of the filtrate was read at 280nm and was corrected for the respective blanks. Blanks were obtained in a similar manner, but TCA was added before the enzyme. One unit of enzyme activity was defined as the increase of absorbance by 1.0 at 280 nm of TCA soluble material in 60 min at 25°C. 5.4.2 Trypsin activity Trypsin activity was measured at 25°C using BAEE as a substrate by the method of Bergmeyer (1974). One unit of activity was defined as the amount of enzyme which caused an increase in extinction of 0.001/min at 253 nm under the conditions of the assay. 5.4.3 Chymotrypsin activity Chymotrypsin activity was measured at 25°C using ATEE as a substrate by the method of Bergmeyer (1974). One unit of activity was defined as the amount of enzyme which caused a decrease in extinction of 0.001/min at 237 nm under the 135 conditions of the assay. 5.4.4 Carboxypeptidase activity Carboxypeptidase activity was measured at 25°C using HPA as a substrate by the method of Bergmeyer (1974). One unit of activity was defined as the amount of enzyme which caused an increase in extinction of 0.001/min at 254 nm under the conditions of the assay. 5.4.5 Protein analysis Protein contents of the enzyme preparations were measured by the method of Lowry et cil. (1951) using bovine serum albumin as a standard, as previously described (Section 3.3.2.1.2.). 5.4.6 Amino acids Amino acids were measured from HC1 hydrolysates of samples. Samples (5- 10 mg) were contained in glass tubes, to which 4 mL 6N redistilled HC1 was added and frozen in a dry ice-acetone bath, then the tubes were sealed under vacuum. Hydrolysis was carried out at 110°C for 24h. The hydrolysates were cooled to ambient temperature, the vacuum released and the tubes kept in a vacuum desiccator with NaOH pellets to remove HC1 and moisture. The remaining particles were dissolved in 1 mL Li-S™ buffer, centrifuged in Eppendorf tubes (10,000xg, 10 min) and injected into the amino acid analyser after dilution (50-100x). 5.4.7 pH optima The effect of pH on enzyme activity was determined in the range of pH 1-7 using 8% hemoglobin substrate, and in the range of pH 6-11 using 8% casein substrate for acidic and alkaline proteases, respectively. Universal buffer (Johnson 136 and Lindsey 1939) was used to prepare the substrate at the required pHs. The procedure of assay is as described in Section 5.4.1. 5.4.8 pH stability The method of Gildberg and Raa (1983) was used to study enzyme stability at different pHs. A volume of enzyme solution (1 mg/mL) was diluted with an equal volume of universal buffer (Johnson and Lindsey 1939) of pH 1-12 and kept at 25°C for 30 min. The residual activity was assayed as described in Section 5.4.1. 5.4.9 Temperature optima To determine the temperature optima of the enzymes, hemoglobin or casein substrate with the assay buffer was equilibrated for 30 min at 25-65°C before the enzyme solution was added. The activity was assayed at the temperature range tested as described in Section 5.4.1. 5.4.10 Temperature stability Temperature stability of the enzymes was determined by incubating the enzyme solution at a temperature range of 25-65°C for 30 min, and the residual activity was assayed as described in Section 5.4.1. 5.4.11 Effect of inhibitors The effect of inhibitors on enzyme activity was studied using as substrates hemoglobin for acidic proteases or casein for alkaline proteases. An equal volume of inhibitor solution of different concentration for each inhibitor and enzyme solution was equilibrated at 25°C for 30 min before the residual activity was assayed. The assay procedure was as previously described (Section 5.4.1). TPCK, TLCK and PMSF were dissolved in DMSO; SBTI, iodoacetic acid in deionised water and 137 Pepstatin A in 70% methanol. As control, the enzymes were added with the respective solvent instead of inhibitors and treated in a similar manner. The enzyme activity without the inhibitor was taken as 100%. 5.4.12 Effect of NaCl on enzyme activity Effect of NaCl on enzyme activity was studied by assaying the enzymes with an addition of 0.5 mL 5-25% NaCl solution in the assay mixtures as described in Section 5.4.1. For control treatments, deionised water was added instead of NaCl and the activity was taken as 100%. 5.4.13 Effect of NaCl on enzyme stability To study the stability of enzyme in NaCl, a mixture of an equal volume of NaCl solution (5-25%) and enzyme solution was equilibrated at 25°C for 24h after which the residual activity was assayed as described in Section 5.4.1. For control treatment, the enzyme was mixed with an equal volume of deionised water instead of NaCl solution in a similar manner and the activity was taken as 100%. 5.4.14 Milk clotting activity Milk clotting activity of acidic proteases was measured according to the method of McPhie (1976) using skim milk. The milk was considered clotted when the absorbance increased by 0.3 unit and the time was taken. Porcine pepsin was used as control. 5.4.15 Effect of substrate concentration Optimum substrate concentration was determined using different substrate concentrations (0.2 - 1.2 %, w/v, casein for alkaline proteases and hemoglobin for acidic proteases) in a 1 h assay at 25°C, pH 8.0 or 3.0 for alkaline and acidic 138 proteases, respectively. Apparent Michaelis-Menten constant (Km') and the substrate turn over number (Vmax) were determined by the method of Lineweaver and Burk (1934). 1/V (Y axis) was plotted against 1/S (X axis). The intercept between the line and the Y axis was 1/Vmax, while that between the line and the X axis was 1/Km1. 5.5 Results 5.5.1 Isolation and partial purification 5.5.1.1 Fractionation of alkaline and acidic proteases The results of fractionation of alkaline and acidic proteases are presented in Figure 5.1 and Table 5.3. In the first trial (Fig. 5.1), acidic proteases always remained in SI at levels of more than about 80%, and precipitated at levels lower than 5%. Alkaline proteases, however, remained in SI at levels higher than those in the precipitate for PAA final concentrations below 0.1%. At final concentrations higher than 0.1%, the alkaline protease activities were higher in the precipitate than in S2. A PAA final concentration of 0.2% resulted in the highest acidic protease activity in SI (about 98%). However, at this concentration, the supernatant still contained about 30% alkaline protease activity, while in S2, the alkaline protease activity was still low (about 50%) although the acidic protease activity was also low (about 3%). At a PAA final concentration of 0.275%, highest alkaline protease activity (about 65%) in S2 was obtained. At this concentration, about 80% of acidic protease activity was retained in SI while the alkaline protease activity in SI was the lowest. This treatment was used in the subsequent fractionation step. . The pH adjustment of SI and S2 was to inactivate the alkaline protease in SI and acidic protease in S2. As shown in Table 5.3, the pH adjustment of S2 did not greatly change the alkaline protease activity. However, the acidic protease activity of 139 this fraction was eliminated. The alkaline protease activity in S2 was zero after pH adjustment, but the acidic protease activity was reduced as the pH was reduced to 3.5 (76% remained) and 3.0 (12.1% remained). At pH 4.0, the acidic protease activity increased to 99.5% of the original. c£ 100 < 80 - 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 PAA concentration (%, w/v) in final mixture Figure 5.1. Distribution of acidic and alkaline protease activity after PAA addition ▼ = Supernatant; ■ = Precipitate; A = Acidic protease; B = Alkaline protease 140 Table 5.3. Distribution (%) of acidic and alkaline proteases during fractionation with PAA and pH adjustment Supernatant Precipitate Fractionation step Acidic Alkaline Acidic Alkaline activity* activity* activity* activity * PAA ** 77.9 ±3.8 6.1 ±0.3 19.8 ±0.2 60.8 ±2.0 pH adjustment:*** 9.0 0 57.4± 1.9 8.5 - - 0 63.1 ± 1.4 4.0 99.5 ±2.0 0 - - 3.5 76.0± 1.7 0 - - 3.0 12.1 ±0.7 0 - - % of original, values are means ± standard deviation for 2 replicates, each from 3 determinations Final concentration is 0.275 Using 2N HC1 or NaOH Based on these results, the optimum fractionation condition was an addition of 2% (w/v) PAA to produce 0.275% in the final mixture, followed by pH adjustment of SI to pH 4.0, and S2 to pH 8.5. This treatment increased the specific activity of alkaline and acidic proteases by 4.7 and 1.1 fold, respectively. 5.5.1.2 Acetone precipitation and dialysis In the preliminary trials, the volume of cold (-30°C) acetone added was 1, 2 and 3 volumes of the sample, where 2x volumes produced the highest purification. Alkaline protease was purified 19.8 fold, while acidic protease was purified 3.4 fold. 5.5.1.3 Ion exchange chromatography Ion exchange chromatography of acidic proteases gave two peaks which exhibited protease activity. One acidic protease (acidic!) passed through the column, while the other was bound and released using acetic acid (cicidic2). The acidic proteases were concentrated 2.9 and 6.0 fold, respectively. Chromatography of the 141 acidic proteases is shown in Figure 5.2. Alkaline protease was not retained by the ion exchange chromatography column and was purified 24.6 fold. Chromatography of alkaline protease is shown in Figure 5.3. The summary of purification of acidic and alkaline proteases is shown in Table 5.4. 20.0 Enzyme 15.0 activity 10.0 AA (A 28 / ) L o/m 0.0 Figure 5.2. Ion exchange chromatography of acidic proteases of cowtail ray viscera ■ A280; • Enzyme activity 2.5 2.0 Enzyme 1.5 activity AA (A 1.0 2 8o/mL) 0.5 0.0 0 5 10 15 20 25 Fraction number Figure 5.3. Ion exchange chromatography of alkaline protease of cowtail ray viscera ■ A280; • Enzyme activity 142 2 > CD S Table 5.4 Summary of purification of alkaline and acidic proteases of cowtail n •c ir 1 c£ H H ■2 T75 H P "2 c: H "2 1 0 e o Q- ra o o §T*< i o 0 0 c i'.f u C 2 2 ’ w <2 •S' £ 2 sf £ 12 5 o £ fc. 2 S f c g £ 2 a. £ c « 9 ^ < 5 ^ < < 5 f oo < So „ P- fN o 00 NO o O O 0 o -f o p- cz 00 — O ra P~ fN 00 © ON NO 00 o -t 0 o r-~ CN o cz CO < 00 O fN ON 00 £ ON 0 fN CO — NO fN 00 ■n 00 00 o O fN 00 CO fN o P~ ON -+ r- fN 00 -t O cn NO <2 < 2 cz o o c 4) 0 c o c o 0 00 fN o o p- 0 ON fN o ON 00 ro o NO •Pi 73 Q LU w 2 2L < 2 2 CD a o £ 2 o o sb 00 fN fN - NO f 1 ( 1 2 < — 3 00 O o O', fN -+ fN -f fN OO fN < ‘ 2 < fN -* o _■ a NO o CD NO NO ™ O 2 NO O | _£ e-e 1 CJ g fN •n 2 £ c o 00 V-. < £ c 00 a C d JS a CO <2 T £ 2 T3 1 1 0 8 CD & 0 c 3 143 5.5.1.4 Polyacrylamide gel electrophoresis and molecular weight estimation The results of polyacrylamide gel electrophoresis (SDS-PAGE) are shown in Figure 5.4. One major discrete band was exhibited by acidicl and alkaline proteases, while more than one band was found in acidic2 protease. Due to the limited amount of acidic2, further purification of this enzyme was not done. A plot of relative mobility (Rf) and molecular weight is shown in Figure 5.5. 5.5.2 Characterisation of alkaline and acidic proteases 5.5.2.1 Optimum pH The effect of pH on the activity of alkaline and acidic proteases is presented in Figure 5.6. Both acidic proteases exhibited highest activity at almost the same pH. The pH optimum for acidicl was between pH 1.25 - 2.0, with highest activity at 1.5- 1.75, while that of acidic2 was between pH 1.75 - 2.5, with highest activity at pH 2.0. Alkaline protease showed an optimum pH around 10. 5.5.2.2 Stability of alkaline and acidic proteases at different pH Figure 5.7 shows the stability of alkaline and acidic proteases at different pH. Alkaline protease was relatively stable in the pH range 7 to 11, where the activity was from 80 to 100% of maximum activity. The activity was dramatically reduced at pHs lower than 6 and higher than 12. For acidic proteases, the activity was reduced with an increase in pH to pH 7.0, from 100% at pH 1 to approximately 30% at pH 7.0 for acidicl and 60% for acidic2. Acidicl was not as stable as acidic2, and rapidly lost its activity as pH increased. 144 A ' ! —-—r^~l r~' • t ' \ 1 i, -11 \ j 1 i i ■I 4; | • i r 1 ~\ I i H■ i, : i'' - i ' li ! i \ 1 *fM } ! !, 1 1 - 1 1 i i 1 i 4 If 1 If r u - ' i l _ . j i \ - ■1 • r ' / V-,. i j < '.^b. * ^ - vL. •- im * L \ ■ ^ i *} [ t f ) * if . f £ j :! \ > *— Figure 5.4. SDS-Page of acidic (A) and alkaline (B) proteases of cowtail ray. Middle lane in A and left lane in B are molecular weight markers. From top: bovine albumin (66.000 D). egg albumin (45.000 D). glyceraldehyde-3-P dehydrogenase (35.000 D). carbonic anhvdrase (29.000 D). bovine trypsinogen (24.000 D). soybean trypsin inhibitor (20.100 D) and oc-lactalbumin (14.200 D). Left lane in A is acidic 1. and right lane is acidic2, six replicates each. 145 Figure Molecular weight (log) 5.5. 4.6 4.8 4.2 4.4 4.6 - - - - - Estimation Bovine Bovine albumin albumin Egg Egg albumin of Glyceraldehyde-3-P ■ Glyceraldehyde-3-P dehydrogenase dehydrogenase albumin cowtail molecular Carbonic Carbonic ray Alkaline Relative anhydrase weight anhydrase viscera 0.4 Trypsinogen Soybean Trypsinogen Soybean of with mobility Acidic2 acidic trypsin trypsin Acidicl SDS-Page inhibitor inhibitor (A) (Rf) 0.6 and cr- a-lactalbumin alkaline - lactalbumin (B) proteases 146 of Relative activity (% of maximum) Figure 100 120 - 5.6. Effect ■ of Acidic pH on 1; the • activity Acidic pH of 2; proteases A Alkaline of cowtail ray viscera 147 120 100 80 60 40 20 0 120 100 80 60 40 20 0 2 4 6 8 10 12 14 PH pH stability of acidic and alkaline proteases of cowtail ray viscera ■ Acidic 1; • Acidic2; A Alkaline 148 5.5.2.3 Temperature optima of alkaline and acidic proteases The effect of temperature on the activity of alkaline and acidic proteases is shown in Figure 5.8. Alkaline protease showed maximum activity at around 50°C. At 40° and 45°C, the activity was in the range of 90-95% of the maximum activity. Both acidic proteases exhibited the highest activity at around 40°C, while at 35° and 45°C, the activity was in the range of 90-95% of that at 40°C. 5.5.2.4 Stability of alkaline and acidic proteases at different temperature The stability of acidic and alkaline proteases at different temperatures is shown in Figure 5.9. The activity of acidic proteases was not affected by temperatures up to 55°C. At temperatures higher than 55°C, the activity of the enzymes was significantly reduced, to 50% of the maximum activity. Alkaline protease showed highest activity at 35°C, exhibited 80% activity at 40° and 45°C, and about 70% at 50°C; the activity was significantly reduced at higher temperature. 5.5.2.5 Effect of NaCl of enzyme activity The effect of NaCl in the assay mixture on enzyme activity is shown in Figure 5.10. NaCl (2-3%) in the assay mixture increased the activity of acidic proteases. Above this NaCl level, the activity was reduced, and at about 8%, the activity was reduced to about 50%. For alkaline protease, the presence of NaCl in the assay mixture-gradually reduced the activity as the NaCl concentration increased. At about 8% NaCl, the alkaline protease activity was about 75%. 149 5.5.2.6 Stability of enzyme in NaCI solution The effect of NaCI on the stability of acidic and alkaline proteases is shown in Figure 5.11. The alkaline and acidic 1 proteases retained their activity after being incubated in up to 12.5% NaCI solution at 25° for 24h. The activity of acidic2, however, decreased with an increase of NaCI concentration and at level of 15%, the activity was only 50% of the control. 5.5.2.7 Effect of inhibitors The effect of various inhibitors on the activity of alkaline and acidic proteases is presented in Table 5.5. Acidic proteases were completely inhibited by Pepstatin A after 30 min incubation at 25°C. Alkaline protease was inhibited by PMSF, TLCK, SBTI, EDTA and iodoacetic acid, but was not inhibited by TPCK. 5.5.2.8 Substrate specificity Alkaline activity towards different substrates is shown Table 5.6. The alkaline protease in the present study did not have any activity against FIPA, little activity on ATEE and casein, and highest activity against BAEE. 5.5.2.9 Milk clotting activity Acidic proteases from cowtail ray viscera were able to clot skimmed milk. However, the time to clot 2 mL, milk by 1 mg of acidic proteases in the present study was much lower than that for porcine pepsin. The time to clot the milk was 40, 45 and 1 second for acidic 1, acidic2 and porcine pepsin, respectively 150 151 10 ray proteases cowtail i 8 of viscera activity ray the i 6 on Alkaline Alkaline cowtail ^ * of (%) mixture proteases NaCl i 4 stability Acidic2; Acidic2; assay the the viscera on in Acidicl;® NaCl Acidicl;# i 2 NaCl of ■ of ■ Effect Effect 0 _l - .10. 5.11. 5 0 200 150 Figure 1IAZU3 Figure Table 5.5. Effect of inhibitors on the activity of alkaline and acidic proteases1 from cowtail ray viscera Inhibitors2 Concentration'' Activity (% )4 Alkaline protease: Control - 100 PMSF 2.5 mM 22.2 TPCK 0.1 mM 115.1 TLCK 0.5 mM 50 SBTI 0.5 mg/mL 78.6 EDTA 2.5 mM 78.6 Iodoacetic acid 0.05 mM 84.7 Acidic proteasel: Control - 100 Pepstatin A 0.03 mg/mL 0 Acidic protease2: Control - 100 Pepstatin A 0.03 mg/mL 0 1 Enzyme concentration in enzyme-inhibitor mixture was 0.5 mg/mL 2 Inhibitors were incubated with a same volume of enzyme for 30 min at 25°C, before the residual activity was assayed 3 In enzyme-inhibitor mixture 4 Relative to that of control Table 5.6. Activity of alkaline protease against different substrates Substrate Specific activity (U/mg)* Casein 17.0 ATEE 1.2 BAEE 104.4 HPA 0.0 One unit activity is defined as the increase or decrease (for ATEE) of absorbance by 0.001/min at 280 nm (casein). 237 nm (ATEE). 253 nm (BAEE) and 254 nm (HPA). 5.5.2.10 Effect of substrate concentration The effect of substrate (S) on the enzymic reaction velocity'(V) is shown in Figures 5.12 and 5.13 for acidic and alkaline proteases, respectively. The correlation coefficients for each enzyme are 0.995 and 0.992 for acidicl and acidic2 proteases and 0.966 for alkaline protease. The apparent Michaelis-Menten constant (Km') and 152 turnover number (Vmax) were calculated and the results are presented in Table 5.7, which shows that acidic 1 had higher turnover number (Vmax) and Michaelis-Menten constant (Km') than acidic2, but was similar to that of porcine pepsin. Table 5.7. Kinetics of cowtail ray viscera proteases and porcine pepsin at 25°C Kinetic Acidic 1 Acidic2 Porcine pepsin Alkaline parameters Km'* 2.3 0.4 1.0 0.1 Vmax ** 34.5 12.4 37.4 3.3 Vmax/Km' 15.0 31.0 37.4 33.0 % (vv/v) casein for alkaline, hemoglobin for others Increase in absorbance by 0.001/min/mg enzyme 1 / S ■ Acidicl;T Acidic2 Figure 5.12. Plot of substrate concentration (1/S) and velocity (1/V) of acidic proteases from cowtail ray viscera 153 1 / s Figure 5.13. Plot of substrate concentration (1/S) and velocity (1/V) of alkaline proteases from cowtail ray viscera 5.5.2.11 Amino acid composition The amino acid compositions of acidic and alkaline proteases from cowtail ray viscera are shown in Tables 5.8. Porcine pepsin was analysed in the present study, while the fish enzyme and bovine trypsin data were from Noda and Murakami (1981), Asgeirsson et al. (1989), Simpson and Haard (1984a) and Murakami and Noda (1981). 5.6 Discussion 5.6.1 Isolation and partial purification of alkaline and acidic proteases The specific activity of alkaline protease in the supernatant of cowtail ray viscera silage, as shown in Table 5.4, was lower than that of salmon viscera silage as 154 reported by Reece (1988). This was probably due to the differences in species as well as the procedure of ensilation. Moreover, cowtail ray is an elasmobranch fish which does not have a distinct pyloric caeca as a source of alkaline proteases. The period of silage storage in the present study was 5 days at 40°C and this might have affected the stability of alkaline protease in acidic conditions. Reece (1988) further reported that alkaline proteases of cod and mackerel viscera were not stable under acidic conditions during ensilation and only acidic proteases were recovered. As expected, the activity of acidic proteases in the present study was high, since fish viscera is much richer in acidic proteases. Table 5.8. Amino acid composition (mol %) of cowtail ray viscera acidic proteases compared with those from other species Amino acid Cowtail ray* Capelin ** Sardine*** Porcine Acidic 1 Acidic2 Pepsm I Pepsin II Enzyme I Enzyme II pepsin* Aspartic acid 16.3 ± ().() 10.6 ±0.0 11.2 11.5 12.9 11.2 7.3 ± 0.0 Hydroxyprolme 4.7 ± 0.4 4.0 ± 0.0 na na na na 11.5 ±0.1 Threonine 3.7 ± 0.1 5.6 ± 0.1 7.1 7.2 11.2 7.5 3.0 ±0.0 Serme 4.1 ± 0.0 7.6 ± 0.0 12.5 12.4 8.8 6.8 4.3 ± 0.0 Glutamic acid 21.8 ± 0.1 10.8 ± 0.1 9.1 11.4 8.9 9.4 7.2 ±0.1 Proline 6.0 ±0.1 5.8 ±0.1 4.0 4.0 8.4 6.1 8.4 ±0.0 Glycine 21.3 ± 0.1 17.5 ± 0.1 12.3 13.0 10.4 12.6 26.8 ± 0.0 Alanine 6.0 ± 0.0 8.2 ± 0.0 7.7 7.6 8.2 7.8 6.9 ± 0.0 Citrulline 0.3 ± 0.0 1.0 ±0.0 na na na na 2.4 ±0.1 Valine 0.6 ± 0.1 3.7 ± 0.1 4.7 7.8 6.7 9.0 1.8 ±0.0 Cysteine 0.1 ± 0.1 0.6 ±0.1 1.8 1.8 2.6 2.3 0.5 ±0.1 Methionine 2.0 ± 0.1 0.1 ±0.0 0.6 0.8 2.6 2.0 0.2 ± 0.1 Isoleucine 2.0 ± 0.0 3.2 ±0.0 6.5 4.9 4.3 4.9 1.6 ±0.0 Leucine 1.7 ± 0.0 4.4 ± 0.0 5.9 5.2 5.0 5.6 2.1 ±0.0 Tyrosine 1.3 ± 0.0 2.1 ± 0.1 3.4 3.1 2.7 3.6 0.5 ± 0.0 Phenylalanine 0.8 ±0.0 2.1 ±0.0 5.1 4.8 2.4 3.5 0.8 ± 0.0 Hydroxy lysine 0.8 ± 0.0 1.4 ±0.1 na na na na 2.7 ±0.0 Lysine 3.0 ±0.0 5.1 ±0.0 2.8 2.5 2.1 2.2 5.6 ± 0.0 Histidine 0.5 ± 0.0 1.7 ±0.2 2.5 1.1 1.2 2.0 1.4 ±0.0 Arginine 2.7 ± 0.0 4.4 ±0.1 2.7 1.0 1.0 2.2 4.8 ± 0.0 Tryptophan nd nd nd nd 1.4 1.5 nd Values are means ± standard deviation for two replications Gildberg (1983) Noda and Murakami (1981) na Data not available nd Could not be detected by the method used 155 Table 5.9. Amino acid composition (mol %) of cowtail ray viscera alkaline protease compared with those from other species Amino acid Cowtail ray1 Atlantic cod' Greenland cod" Sardine4 Bovine4 Aspartic acid 8.7 ±0.0 8.4 10.6 13.4 22.0 Hydroxyproline 1.2 ±0.1 na na na na Tlireonine 7.7 ±0.1 4.4 4.6 8.1 10.0 Serine 6.2 ± 0.0 10.2 14.7 7.4 34.0 Glutamic acid 7.5 ± 0.0 12.0 8.7 7.2 14.0 Proline 4.3 ±0.0 2.7 4.6 4.1 8.0 Glycine 10.2 ±0.1 12.4 12.8 12.0 25.0 Alanine 7.5 ± 0.0 7.1 7.3 6.5 14.0 Citrulline 5.2 ± 0.0 na na na na Valine 5.7 ±0.0 8.0 7.3 7.8 15.0 Cysteine 0.1 ±0.0 5.3 3.7 3.4 12.0 Methionine 0.1 ±0.0 2.7 1.4 0.3 2.0 Isoleucine 4.8 ±0.0 3.6 3.7 4.7 13.0 Leucine 8.3 ± 0.0 7.1 6.4 4.1 13.0 Tyrosine 3.3 ±0.0 4.4 3.2 5.7 10.0 Phenylalanine 4.3 ± 0.0 1.3 1.8 1.9 3.0 Hydroxy lysine 0.5 ± 0.0 na na na na Lysine 6.0 ± 0.0 3.6 2.8 7.6 14.0 Histidine 2.6 ±0.1 2.7 3.2 1.4 .3.0 Arginine 5.8 ±0.1 2.7 2.3 2.6 2.0 Tryptophan nd 1.3 0.9 1.9 3.0 1 Values are means ± standard deviation for two replications 2 Asgeirsson et al. (1989) 3 Simpson and Haard (1984a) 4 Murakami and Noda (1981) na Data not available nd Could be detected by the method used Charged polymers have been used by several workers to precipitate proteins from solution, and at the same time fractionate the protein from other proteins or compounds. This method has been reported able to separate amyloglucosidase from transglucosidase (Sternberg 1972, 1975), lipoxygenase from (3-amylases (Sternberg 1976), alkaline from acidic proteases of salmon viscera (Reece 1988), lysozyme from egg white (Fischer and Glatz 1988), as well as in precipitating trypsin from a pepsin- trypsin mixture (Sternberg and Hershberger 1973), plant proteases (Caygill, Moore and Kanagasbapathy 1983), chitosanase (Boucher et al 1992) and lysozyme 156 (Sternberg and Hershberger 1973, Fischer and Glatz 1988, Shieh 1989). The charged polymers commonly used are carboxymethyl cellulose, methacrylate polymers, polyethyleneimine (PEI) and polyacrylic acid (PAA). In the present study, PAA was used since it was used by Sternberg and Hershberger (1973), Caygill et al. (1983) and Reece (1988) to successfully fractionate alkaline and acidic proteases, which were similar to the enzymes studied here. Moreover, in the preliminary study, the use of PEI to precipitate acidic proteases proved unsatisfactory. In the precipitation of protein by PEI, salt must be added to aid the process (Bell, Hoare and Dunnill 1983, Jendrisak 1987) and for acidic proteases from cowtail ray viscera, salt affected the activity. A concentration of 0.4M salt in the mixture increased the total acidic protease activity by almost 40%. The use of low molecular weight (90,000D) PAA was also tested in the preliminary study. However, the results were not satisfactory since the contamination of acidic proteases was high in the precipitate. Although low molecular weight (90,000-250,000D) PAA has been successful in precipitating some enzymes (Sternberg and Hershberger 1973, Sternberg 1976, Shieh and Glatz 1991, Boucher et al. 1992), it seems that low molecular weight PAA was not suitable for fractionation of alkaline and acidic proteases from cowtail ray viscera. In the fractionation of transglucosidase and amyloglucosidase (Sternberg 1975) PAA of at least 1M molecular weight was recommended. According to Sternberg (1976) the formation of an insoluble protein-polyacrylate complex depends among others on the molecular weight of PAA and the ratio of PAA'protein. Higher molecular weight PAA was then tested in the present study. Using PAA of 3-5MD molecular weight, the precipitation pattern of the alkaline protease was asymmetrically bell shaped (Figure 5.1). A similar result was 157 also shown by Caygill et al. (1983) for plant proteases (papain and ficin) and Boucher et al. (1992) in precipitating Streptomyces N174 chitosanase. It was suggested that the decrease in recovery at higher polyelectrolyte dosage was due to the excess polyelectrolyte which was incorporated into the already formed protein- polyelectrolyte complex, resulting in a soluble complex with a significant net charge, and thus hindering the aggregation of the complexes (Caygill et al. 1983, Clark and Glatz 1987, 1988, Chen and Berg 1993). In their studies, Caygill et al. (1983) employed PAA/protein ratios of 0.02-0.5, while Boucher et al. (1992) used ratios of up to 10. The optimum ratio for plant proteases was between 0.14-0.29 and 0.2-0.4 for papain and ficin, respectively, while for chitosanase it was around 4. In the present study, the final concentration of PAA in the mixture was in the range 0.025 to 0.3%, or a PAA/protein ratio of 0.006 to 0.075, and the optimum ratio to achieve the highest alkaline protease activity in the precipitate was 0.069. These ratios were much lower than those of the above studies due to the different natures of the proteases as well as the PAA. Boucher et al. (1992) used PAA with molecular weight of 250,000D, while that used in the present study was 3-5M. The molecular weight of PAA used by Caygill et al. (1983) was not reported. A similar ratio of PAA/protein (0.066) was used by Reece (1988) to separate alkaline protease from acidic protease of salmon viscera silage. No investigation of the optimum ratio was reported in Reece's study. However, it was reported that the alkaline protease activity in the precipitate was 82.7% of the original, while that of acidic proteases in the supernatant was 100%. The data for the acidic protease in Reece's study were reported to be modified to take account of the co-purification of the alkaline protease, however, no explanation of the modifications were mentioned. 158 No data on the activity of alkaline protease in the supernatant or acidic protease in the precipitates were reported. PAA precipitation in Reece's study increased the purity of the acidic and alkaline proteases by 1.6 and 6.8 fold, respectively, while in the present study, activities were 1.1 and 4.7 fold, respectively (Table 5.4). Acetone precipitation in the present study increased the purity of acidic and alkaline proteases by 3.4 and 19.8 fold, respectively. The purification of alkaline protease using acetone precipitation was about four fold compared to the preceding purification step. High purification of alkaline protease after acetone precipitation was also demonstrated by Simpson (1984) and Zotos and Taylor (1996). Bundy and Gustafson (1973) observed that after being kept overnight at 5°C the alkaline protease (trypsin) activity of an acetone extract of starfish pyloric caeca doubled. According to Simpson (1984) the increase in activity could be due to the elimination of naturally present trypsin inhibitors by the acetone treatment. Simpson (1984) further stated that the presence of trypsin inhibitor in vertebrates is a physiological control mechanism to "accommodate the otherwise devastating effects of premature activation of the zymogen". It is not known whether the cowtail ray viscera in the present study contains natural trypsin inhibitors, or if a zymogen was present. Further study is necessary to confirm this. The alkaline activity after ion exchange chromatography was increased by about 1.3 fold compared to that after acetone precipitation, and 24.6 fold compared to that of the original, which was lower to that for salmon alkaline protease (Reece 1988). The recovery of activity was also lower, i.e. 6.4% compared to 34.9%. The purification of acidic proteases in both studies cannot be directly compared since in the present study, two acidic proteases were found, while Reece (1988) found only one acidic protease. However, if acidic 1 and acidic2 are combined, the recovery of 159 acidic protease activity in the present study was 25.2%, compared to 43.7% reported by Reece (1988). Based on the above recoveries, acidic 1 was the major acidic protease in the cowtail ray viscera investigated in the present study. Pepsin 1 was the major pepsin in sardine (Noda and Murakami 1981), capelin (Gildberg 1982, 1988) and orange roughy (Xu et al. 1996), but not in cod (Martinez and Olsen 1986) and salmon (Sanchez-Chiang, Cisternas and Ponce 1987) where pepsin2 was the major pepsin. The result of gel electrophoresis indicates that the enzyme preparation in the present study exhibited one major discrete band for acidic 1 and alkaline proteases, and more than one band for acidic2 protease. Since the elution in ion chromatography for acidic2 in the present study was done using acetic acid of a single ionic strength (not gradient or step wise elution), it is possible that all bound proteins, including enzyme and some non-enzyme proteins, were eluted at the same time. Reece (1988) showed that using an almost similar purification procedure to the present study, one discrete band was found in the acidic protease preparation and more than one band in the alkaline protease for salmon viscera, while for cod and mackerel viscera, more than one band was shown in the acidic and alkaline protease preparations. It should be noted, however, that the purification procedure of Reece gave only one acidic protease. 5.6.2 Characterisation of acidic and alkaline proteases from cowtail ray viscera 5.6.2.1 Acidic proteases Acidic proteases were completely inhibited by Pepstatin A at the concentration and incubation applied. According to Garcia-Carreno (1993), Pepstatin A is a highly 160 specific inhibitor for aspartic proteases. This indicates that the acidic proteases in the present study were aspartic proteases. There are two major aspartic proteases in fish, i e. pepsins and cathepsins, and both have similar specificity and molecular structure (Gildberg 1988). However, the physiological role of both proteases is different; pepsins have extracelullar functions as the major gastric proteases, while cathepsins are lysosomal enzymes, usually found in muscle and more than a dozen types of cathepsins have been identified (McLay 1980, Asghar and Bhatti 1987, Aoki, Yamashita and Ueno 1995). Cathepsins are sometimes included in the group of non- gastric proteinases (Dunn 1991, Takahashi 1994). Cathepsin (D) has been identified in invertebrate stomachs, and although there have been reports about the availability and role of cathepsins in the digestive tracts (stomach) of fish, there is no evidence to show that cathepsins play a role as an extracellular digestive proteinase in fish stomach (Gildberg 1988). Thus, it is most likely that the aspartic proteinases in the present study were pepsin-like enzymes, and were referred to as pepsin 1 and pepsin2 in the present study. The amino acid composition of pepsins in the present study (Table 5.8) indicates that the acidic amino acids were predominant over the basic amino acids. Of the residues in acidic 1, 36.5% are acid and 5.9% are basic, while in acidic 2, 19.9% are acid and 10.5% are basic. Similar results were also reported by Noda and Murakami (1981) in acidic proteases of Japanese sardines. They also observed that, similar to the present study, the proportion of acidic amino acids reduced and basic amino acids increased in acidic protease 2. The levels of acidic amino acids in cowtail ray pepsins were higher than in other fish pepsins and porcine pepsin. As expected, aspartic acid was present in high proportion in acidic proteases since it is the active site residue of aspartic proteinases (Neurath 1989). 161 The estimation of molecular weight was based on the relative mobility of protein markers and pepsins as shown in Figure 5.5. It was estimated that the molecular weight of pepsin 1 was about 29,500 D. This was in the range of fish and marine mammal pepsins reported by several workers using SDS PAGE, such as capelin, 23,000 D (Gildberg and Raa 1983), orange roughy, 33,500 D (Xu et al. 1996), seal, 33,800 D (Shamsuzzaman and Haard 1984), sardine, 35,000 D (Noda and Murakami 1981) and Greenland cod, 36,400 D (Squires et al. 1986b). For pepsin2, however, the estimation was rather difficult, since more than one band appeared in PAGE. However, the above workers, except Noda and Murakami (1981), reported that the molecular weight of pepsin2 were slightly higher than that of pepsinl. Thus it can be estimated that the band closer to the pepsinl band in Figure 5.4 was pepsin2, having a molecular weight of about 31,600 D. Pepsins acting on hemoglobin in the present study have a wide range of temperature optima from 35° to 45°C, with peak activity at around 40°C. The temperature optima (40°C) in the present study was higher than that for Arctic and Greenland cod and American smelt proteases, which were about 30-32wC (Haard et al. 1982). However, the temperature range (35-45°C) at which cowtail ray pepsins retained 80-100% of activity brackets the optimum temperature of pepsins from cold and temperate water fish such as Arctic capelin, 38-43°C (Gildberg and Raa 1983), orange roughy, 37°C (Xu et al. 1996), sardine, 40°C (acidic protease2) (Noda and Murakami 1981), and Indonesian tropical oil sardine, 45°C (Gildberg 1982). The results of the present study support the observations by Gildberg (1982) on the temperature optima of capelin and Indonesian oil sardine, where it was shown that no large differences were observed between temperature optima of digestive acidic proteases from both fish. An investigation of the available data in the literature 162 by Gildberg (1988) also shows that temperature optima of pepsins from cold and warm water fish are not much different, except that of the acidic proteasel of sardine observed by Noda and Murakami (1981) which exhibited highest activity at 55°C, which is an anomaly. In general, the temperature optima of fish pepsins are lower than that of their mammalian counterparts, e.g. around 47-50°C for porcine pepsin (Haard et cii 1982). The thermal stability of pepsins in the present study was observed over a slightly wider range than for pepsins from cold or temperate waters. The temperatures which pepsins from cold and temperate waters can withstand are usually between 40° to 50°C, and are closely related to temperature optima (Gildberg 1988). Pepsin from dogfish was reported to have a temperature optimum of 50°C and was also stable at temperatures below 50°C (Guerard and Le Gal 1987). Similarly, Atlantic cod pepsins were rapidly inactivated at above 40°C , while the highest activity was exhibited at slightly below 40°C (Martinez and Olsen 1986). However, Japanese sardine acidic proteases were stable at 60°C when the pH of the environment was 4, and very unstable when the pH was 2, while the temperature optimum was 40-55°C (Noda and Murakami 1981). In the present study, the temperature at which cowtail ray pepsins were stable was below 55°C, while the temperature optima were between 35()-45°C. The thermal activity of pepsins of cowtail ray viscera were similar to porcine pepsin as reported by Squires et ctl. (1986b). The pH optima of pepsins of cowtail ray were between 1.5-2.5, and were slightly more acidic than some cold and temperate water fish such as cod, mackerel and salmon, pH 2.6 (Reece 1988), Arctic capelin, pH 2.5-3.7 (Gildberg and Raa 1983), Greenland cod, pH 2.5-3.0 (Squires et a/. 1986a), dogfish, 2.5 (Guerard and Le Gal 1987) and Japanese sardine, pH 2.0-4.0 (Noda and Murakami 1981). 163 However, Arunchalam and Haard (1985) observed that pH optima of polar cod pepsins were around 2.0, which were similar to that of the present study. The pH stability of cowtail ray pepsins was slightly different from that of Arctic capelin (Gildberg and Raa 1983), Greenland cod (Squires el a/. 1986b) and Japanese sardine (Noda and Murakami 1981), which were fairly stable at pH 1-6, and lost activity at pH 7.0 or higher. Cowtail ray pepsin 1 rapidly lost activity at pH higher than 4, and pepsin2 at pH higher than 7. Gildberg and Raa (1983) observed that Arctic capelin pepsins were fairly stable at pH 1-5, but at this pH, they were susceptible to autodigestion if no substrate was present. It seems that this was also true for cowtail ray pepsins, since in the pH stability test, the enzymes were incubated at different pHs without any substrate. The kinetics of cowtail ray pepsins reaction against hemoglobin at 25°C were studied by observing their Km' and Vmax. As shown in Table 5.7, pepsins of cowtail ray viscera had lower Vmax and Vmax/Km' ratio than did porcine pepsin. Pepsin 1 had higher Km', while pepsin2 had lower Km' than porcine pepsin. According to Squires el a/. (1986a), the characteristics of digestive enzymes in hydrolysing food proteins are different from those of intracelullar enzymes, and digestive enzymes are expected to hydrolyse food in the fastest and most efficient way at high substrate concentrations. In this context, a high Vmax is desirable for rapid digestion. In the present study, the Vmax of pepsins from cowtail ray viscera showed a lower value than that of porcine pepsin, indicating that cowtail ray viscera pepsins digested hemoglobin of high concentration at a lower rate than did porcine pepsin. Squires et a/. (1986a) isolated 3 pepsins from Greenland cod stomach and found that, using hemoglobin as substrate, 2 pepsins had higher Vmax, and 1 pepsin had lower Vmax than porcine pepsin. Three pepsins isolated from Atlantic cod (Gildberg, Olsen and 164 Bjarnason 1990, 1991) had higher Km' than porcine pepsin, whereas pepsin 1 had higher Km' than pepsin2. While enzymes with high Vmax can contribute more to the digestion of protein, those with lower Vmax could be a more effective catalyst at low substrate concentrations (Arunchalam and Haard 1985). Km' can be regarded as the amount of substrate that produces half-maximal velocity and enzymes with low Km' function at a faster rate with smaller amounts of food present (Squires et al. 1986a). From the value of Km', it can be deducted that pepsin2 from cowtail ray viscera had a faster rate in hydrolysing hemoglobin of low concentration, but had lower rate at higher concentration of hemoglobin than porcine pepsin. On the other hand, porcine pepsin had a faster rate in hydrolysing hemoglobin at low concentration, but had lower rate at higher concentration of hemoglobin than pepsin 1. Pollack (1965) defined Vmax/Km' as "physiological efficiency" and suggested that it can be used as an indication of the overall efficiency of enzymes. The results in the present study indicate that the physiological efficiency of pepsins of cowtail ray viscera is lower than that of porcine pepsin. Similar results for Greenland cod gastric proteases were also shown by Squires et al. (1986a). There is no explanation in the published reports why porcine pepsins had higher physiological efficiency than fish pepsins. It could be possible that the commercial porcine pepsin had higher purity than the experimental pepsins in the present study or that of Squires et al. (1986a). According to Squires et al. (1986a), the low physiological efficiency of fish pepsins may be compensated by the wider substrate specificity of fish pepsins than porcine pepsins. In addition, the relatively higher Km' of fish pepsins indicates the carnivorous nature of fish such as cod (or stingrays, in the case of the present study), which is different from pigs which usually feeds on relatively lower levels of protein 165 (Haard et al. 1982). Higher Km' also indicates a lower affinity towards the substrate used, and it makes the physiological efficiency value lower since it is used as the denominator. The acidic proteases of cowtail ray viscera have been shown to be able to clot milk in the present study. However, compared to porcine pepsin, the ability of cowtail ray pepsins to clot milk in this study was much lower. If one milk clotting unit (CU) was defined as the amount of enzyme to clot 2 mL skimmed milk in 1 s (Reece 1988), the milk clotting units of the pepsins from cowtail ray were 0.025 and 0.022 for pepsinl and pepsin2, respectively, compared to 1.7 for porcine pepsin. Cod, mackerel and salmon pepsins isolated and purified in a similar manner to that in the present study also had lower milk clotting units compared to porcine pepsin, i.e. 0.7, 0.1 and 2.1 for cod, mackerel and salmon pepsins, compared to 16.3 for porcine pepsin, respectively (Reece 1988). If a unit activity (HbU) was arbitrarily defined as an increase in absorbance at 280 nm by 1.0/h using 1% hemoglobin as a substrate, the ratio CU/HbU was 0.04, 0.06 and 1.67 for pepsinl, pepsin2 and porcine pepsin, respectively. According to Squires et al. (1986a), a high CU/HbU ratio indicates a capability of milk clotting activity of the enzyme in cheese processing. Fish pepsins from cold water species were investigated as a rennet substitute by several workers (Brewer et al. 1984, Shamsuzzaman and Haard 1985, Haard 1992b), however, in the present study it was shown that the use of cowtail ray pepsins as a rennet substitute in cheese manufacturing is restricted. Reece (1988) found that cod pepsin had a lower CU/HbU ratio than porcine pepsin and concluded that the use of this enzyme to produce cheese was not possible. However, Brewer et al. (1984) showed that cod enzymes could be employed to produce satisfactory cheddar cheese. It could be 166 possible that the experimental pepsins in the present study and of Reece (1988) was less purified. It was shown in the present study that small concentrations of NaCl increased the activity of cowtail ray pepsins. Activation of some fish pepsins by a small amount of salt has been reported by several workers. Squires et al. (1986) observed that salt concentrations up to 50 mM increased the activity of cod proteases 1 and 2, but did not affect the activity of protease3. Similar results were reported by Sanchez-Chiang et al. (1987) for salmon pepsin 1, but not for salmon pepsin2. Up to 9% NaCl (about 1 7M) slightly increased the activity of sardine pepsins (Noda et al. 1982). This might be partly the reason why the acidic protease activity increased by almost 40% in the preliminary study to precipitate acidic proteases using PEI, which involved the addition of salt. At higher concentration of NaCl, however, the activity of pepsins was inhibited. At a salt concentration of more than 5% (1.71M), the activity of all pepsins was reduced. Similar results were reported by Gildberg et al. (1991) and Xu et al. (1996) for pepsins from Atlantic cod and New Zealand orange roughy, respectively. According to Sanchez-Chiang et al. (1987), changes in the conformation of the enzyme and hemoglobin in the presence of salt in high concentration were the reason why enzyme activity was reduced. In the present study, the colour and consistency of the assay mixtures changed to a paler colour and thicker consistency as high concentrations of salt were added. The present study has also demonstrated the stability of cowtail ray pepsins in salt concentrations up to 12.5% (2.1M). Pepsin 1 was more stable in salt than pepsin2, and retained activity after 24h incubation at 25°C at all levels of salt tested. Noda et al. (1982) reported that Japanese sardine pepsin2 was stable in up to 25% salt, while 167 the activity of pepsin2 was reduced by almost 40% after 24h incubation at 30°C at the same level of salt. The stability of fish pepsins in salt solutions indicates that during traditional fish processing in some Asian and .African countries in which the viscera are not removed, the digestive enzymes are stable and still active, and must play a role in flavour development of these products. 5.6.2.2 Alkaline protease The alkaline protease in the present study was inhibited by PMSF, and this indicates that the enzyme was a serine protease. The results of the inhibitory study using TPCK, TLCK and SBTI showed that the enzyme was a trypsin-like enzyme, since the enzyme was inactivated by TLCK and SBTI but not inhibited by TPCK. TLCK and SBTI are specific inhibitors for trypsin and trypsin-like enzymes, while TPCK is an inhibitor for chymotrypsin and chymotrypsin-like enzmes (Garcia-Carreno 1993). The alkaline protease in the present study was slightly inhibited by EDTA. Similar results were also observed by Asgeirsson et al. (1989) on cod trypsin, Heu et al. (1991) on anchovy trypsin and Kim and Pyeun (1986) on mackerel trypsin. The inhibition of the alkaline protease activity by EDTA indicates that the enzyme requires divalent ions such as calcium for stability (Asgeirsson et al. 1989). The buffer used in the purification of alkaline protease in the present study contained 5 mM CaCh, which was not removed by dialysis before the enzyme was freeze dried. The results of the inhibitory study in the identification of alkaline enzymes were supported by the substrate study, which shows that the enzyme hydrolysed BAEE and did not hydrolyse ATEE or HPA at all. BAEE was a specific synthetic substrate for trypsin, as was ATEE for chymotrypsin and HPA for carboxypeptidase. In the present study, the trypsin-like enzyme of cowtail ray was referred to as trypsin. The amino acid composition of cowtail ray trypsin shown in Table 5.10 168 indicates that it contains total basic amino acids at higher levels than fish trypsins from cold and temperate waters. Compared to bovine trypsin, the level of basic amino acids in cowtail ray trypsin was lower. The total acidic amino acids of cowtail ray viscera trypsin were lower than for other fish and mammalian trypsins. Based on the similarities in the proportion of certain amino acids such as threonine, glutamic acid, proline, alanine, methionine, isoleucine, and other basic amino acids, cowtail ray trypsin was more like a temperate sardine trypsin (Noda and Murakami 1981). The molecular weight of cowtail ray trypsin estimated with SDS-Page was about 45,600 D, which was higher than for other fish trypsins, such as those from sardine, 23,000-29,000 D (Murakami and Noda 1981), anchovy, 27,000-28,000 D (Martinez el a!. 1988), Greenland and Atlantic cod, 23,000-24,000 D (Simpson 1984, Simpson el at. 1990) and skipjack, 24,000-29,000 D (Pyeun et a/. 1988). Keil (1971) stated that trypsins have a molecular weight ranging from 20,000-24,000 D. However, some studies have shown that trypsin from marine organisms has a molecular weight outside this range (Martinez et aJ. 1988). Low molecular weight has been reported for trypsin from Korean mackerel, i.e. 16,000 D (Kim and Pyeun 1986), while a high value (50,100 D) was found in shrimp, Penaeus monodon (Jiang et a/. 1991). The temperature optimum of cowtail ray trypsin against casein was 50°C. This level was higher than that of trypsin from cold and temperate water fish as well as from Indonesian oil sardines and grouper against the same substrate (Murakami and Noda 1981, Gildberg 1982, Kim and Pyeun 1986, Shin and Zall 1986, Reece 1988, Simpson et aJ. 1990, Sulistiyani and Heruwati 1991). However, this optimum temperature was similar to that of Korean skipjack and bovine trypsin using casein as substrate (Simpson 1984). The temperature optimum of cowtail ray trypsin in the 169 present study was similar to or slightly lower than that for crustaceans and shellfish such as krill, oyster, sand crab, crayfish and shrimp (Konagaya 1980, Tsao and Nagayama 1991, Jiang et al. 1991, Dionysius et al. 1993, Kim etcil. 1996). The thermal stability study showed that cowtail ray trypsin started to lose its activity above 50°C. Japanese sardine trypsin lost its activity by almost 70% at 55°C (Murakami and Noda 1981), while the activity of Korean skipjack and mackerel trypsins reduced significantly above 45°C against casein as a substrate (Kim and Pyeun 1986, Pyeun et al. 1988). Cod trypsin rapidly lost its activity against casein upon incubation at 57°C for 15 min (Shin and Zall 1986). Compared to bovine trypsin, however, the thermal stability of fish trypsins was much poorer, as bovine trypsin could withstand temperatures as high as 80°C without losing its activity against casein (Simpson 1984). The denaturation of enzyme and protein substrate was most likely the reason for the instability at high temperatures. It should be noted, however, that some studies have shown that in the presence of calcium ions, the thermal stability of trypsins might improve (Simpson 1984, Kristjansson 1991). Thus in comparing the thermal stability of such enzymes, it is important to carefully check whether calcium ions are present in the incubating mixtures. In the present study, no calcium ions were added. The pH optimum for casein digestion by cowtail ray trypsin in the present study was not much different from that of cold and temperate water fish, which were in the range of pH 8-11. Highest activity against casein for cowtail ray trypsin was exhibited at pH 10, similar to that of Japanese sardine (Murakami and Noda 1981), and slightly higher than Norwegian anchovy, pH 9.5 (Martinez et al. 1988) Greenland cod, pH 9.0-9.5 (Simpson 1984), Japanese oyster, pH 8.0 (Tsao and Nagayama 1991), American cod, pH 9.6 (Shin and Zall 1986) Korean skipjack, pH 170 9.5 (Pyeun et al. 1988) and salmon, pH 8.5 (Reece 1988). It was also higher than for bovine trypsin, i.e. pH 8.0 (Simpson 1984). Compared to alkaline proteases from other Indonesian tropical fish, the pH optimum for cowtail ray trypsin was not much different (Sukarsa 1978, Gildberg 1982, Sulistiyani and Heruwati 1991). Compared to trypsin from the above cold and temperate water fish, the pH stability of cowtail ray trypsin was similar. In general, fish trypsins from both environments were unstable at pH 7.0 or lower and at extremely high pH. The kinetic studies showed that cowtail ray trypsin had higher Km' value than those from stomachless fish, Ccircissius auratus gibelio (Jany 1976) and Japanese rotifer, Brachiomis phicalilis, (Hara, Arano and Ishihara 1984), but lower than those of Korean skipjack and mackerel (Kim and Pyeun 1986, Pyeun el al. 1988) and crayfish (Kim el al. 1996). This indicates that cowtail ray trypsin had a lower affinity for casein than proteases of stomachless fish and rotifers, but higher than those of skipjack, mackerel and crayfish. The physiological efficiency (Vmax/Km1) of cowtail ray trypsin in the present study was relatively higher than that of crayfish, which was measured at 45°C (Kim et al. 1996). Comparison to other species available in the literature can not be done since there are differences in the substrates used as well as the conditions of assay. Substrates usually used for trypsin kinetic studies in the literature were specific for trypsin such as tosyl-L-arginine methyl ester (TAME). No increase of trypsin activity was observed with the addition of salt to the assay mixtures, unlike the pepsins. As shown in Figure 5.11, the activity of trypsin gradually decreased with an increase in salt concentration. However, the rate of activity loss was lower than that of pepsins. At about 8% salt, the trypsin activity was about 80% of that without salt. Similar results were observed for sardine alkaline proteasel (Noda et al. 1982). During incubation at 25°C for 24h in the presence of 171 up to 12.5% salt no apparent loss of activity was observed in both cowtail ray and sardine trypsins. The incubation time in the present study was only 24h, and the effect of salt was not pronounced. However, from Figure 5.11, it can be deduced that higher concentration would affect the activity. Salt is an important component in the production of fish sauce and is used at very high concentration in Asian countries. Delayed addition of salt for up to 24h proved disadvantageous and led to spoilage (Ooshiro et al. 1981). However, high salt concentration inhibits the trypsin and trypsin-like activity in fish sauce production where the rate of autolysis is low (Gildberg 1989, 1993). Although not specifically identified as trypsin, alkaline protease activity in fish sauce fermentation decreased as the fermentation time increased (Thonthai et al. 1990). Gildberg (1992) and Gildberg and Xian-Quan (1994) observed a rapid decline in trypsin-like activity during fish sauce fermentation, but not with chymotrypsin. The result of this study supports the findings of Gildberg and Xian-Quan (1994) that trypsin is not the principal enzyme in fish sauce fermentation. This was probably partly the reason why a trial to use cowtail ray crude alkaline enzymes to produce fish sauce by an accelerated method of Greig and Estrella (1988) failed. However, according to Orejana and Liston (1981), trypsin and trypsin-like enzymes were the principal agents of proteolysis in fish sauce fermentation, as the trypsin activity of fish sauce made from tropical anchovies and scad was about 50% of the initial after one month storage at 37°C. It was suggested (Gildberg 1992) that an initial high concentration of trypsin is necessary before the role of trypsin in fish sauce production can be seen. 172 6 CONCLUSIONS AND RECOMMENDATIONS 6.1 Conclusions Cowtail ray processing in Muara Angke, Jakarta, West Java and Labuhan Maringgai, Lampung, Southeast Sumatra, produces a vast amount of waste, especially the viscera. The viscera is rich in protein and digestive enzymes and efforts to utilise this waste have been investigated. The preservation of cowtail ray viscera has been studied. Ensilation of cowtail ray viscera can be achieved either with organic acids such as a mixture of propionic and formic acids (1:1 v/v) or with an inorganic acid such as hydrochloric acid at a concentration of 3% and 4% (v/w). Hydrochloric acid at levels of 3% and 4% (w/v) liquefied the viscera at ambient temperature (28-32°C) and at 40°C. However, 3% HC1 was not able to preserve the viscera which spoiled within 3-5d at both temperatures. HC1 at 4% preserved the viscera for up to 15d. Mixtures of propionic and formic acids (1:1, v/v) at levels of 3% and 4% liquefied and preserved the viscera for at least 120d at both ambient temperatures (28-32°C) and at 40°C. The solubilisation pattern of nitrogen during ensilation of cowtail ray viscera with organic acids in this experiment was similar to that of temperate or cold water species indicating that there might be few differences in the behaviour of their digestive enzymes. Temperature affected solubilisation, where higher temperature promoted faster solubilisation. No significant difference due to levels of organic acids was found, therefore a mixture of propionic and formic acids (1:1, v/v) at a level of 3% (v/w) is recommended. The use of HC1 is not recommended unless the silage is used with neutralisation within 15d. 173 Digestive enzymes of the viscera were still active after 5d ensilation at 40°C. The ability of these enzymes to hydrolyse fish protein has been demonstrated and may be useful in the production of fish protein hydrolysates once the optimum conditions are achieved. Optimum hydrolysis was at 60WC for 3h at a ratio of crude enzyme to fish meat of 5%. Cowtail ray viscera silage was rich in peptones which were able to support the growth of microorganisms, either in mixed populations such as those isolated from a variety of foods, or individually. Although the ability of these peptones to support the growth of individual samples of mixed populations or single microorganisms was not always better than those of commercial peptones (Difco, Oxoid and BBL), in general the viscera peptones ranked 1 or 2, judged from total growth (increase in absorbance at 600 nm of liquid media) and dry biomass production (mg/lOOmL liquid media). Organic acids have been reported to have inhibitory effects on microorganisms, however the microorganisms grown on peptones produced by organic acid ensilation in the present study were not affected and exhibited normal growth. The present study has shown the potential of cowtail ray viscera peptones to compete with commercial peptones in supporting microbial growth. With the abundant availability of this waste in the processing sites such as Muara Angke, Jakarta and Labuhan Maringgai, Lampung, Southeast Sumatera, and the high price of microbiological peptones in Indonesia, which can be double the price in Australia (AUD 140-260 in Sydney, compared to equivalent AUD 500-600 in Jakarta, per kg, January 1997), the conversion of such waste into peptones could not only reduce some environmental problems in the area, but also supply cheap raw materials for microbiological peptones production. 174 Ensilation using organic acids has extracted both alkaline and acidic proteases from cowtail ray viscera. Two acidic proteases and one alkaline protease from tropical cowtail ray viscera have been partly purified and characterised. The acidic proteases have been identified as pepsin-like enzymes (referred to as pepsin 1 and pepsin 2), while the alkaline protease was identified as a trypsin-like enzyme (referred to as trypsin). Cowtail ray viscera pepsins had higher levels of acidic amino acids than other fish pepsins and porcine pepsin. The molecular weights of pepsins 1 and 2 were about 28,500 D and 31,600D, respectively. Both pepsins exhibited highest activity against hemoglobin at 35-45°C, and were not different from those of cold and temperate water fish. The thermal stability of both pepsins was slightly better than cold and temperate water fish pepsins. Cowtail ray pepsins had slightly more acidic pH optima than some cold and temperate water fish, and were not stable at pHs above 4. Salt at concentrations of up to 4% in the assay mixtures enhanced cowtail ray pepsins' activity while incubation with up to 12.5% salt at 25°C for 24h did not affect the activity. These results indicate that in traditional fish processing in Asian and African countries, where the viscera is not removed and salt is added, digestive enzymes such as pepsins are stable and active and might result in significant flavour development in such fish products. Cowtail ray pepsins showed no capability of substituting for rennet in cheese processing, either because tropical fish pepsins are not suitable for this purpose, or the isolation and purification procedures affect milk clotting activity. The molecular weight of cowtail ray trypsin was estimated at about 45,600D. It has an amino acid composition similar to that of sardine from temperate water. The 175 temperature optimum against casein was 50°C, which was higher than for other cold and temperate water fish, similar to that of bovine trypsin and similar or slightly lower than those of crustaceans and shellfish trypsins. Cowtail ray tiypsin lost its activity at temperatures above 50°C. The pH optimum of cowtail ray trypsin (~10) was similar to those of cold and temperate water fish, as well as to alkaline proteases from other Indonesian fish. The enzyme was unstable at pHs lower than 7.0 and higher than 11.0. Kinetic studies showed that using casein as a substrate cowtail ray trypsin had lower Km' than temperate water fish trypsin, indicating a higher affinity against casein. No increase in activity was observed due to salt at up to 8% in the assay mixture. The stability of trypsin was not much affected by an increase of salt concentration up to 12.5% upon incubation at 25°C for 24h. 6.2 Recommendations Further studies are necessary to provide better information on the ensilation of tropical fish viscera, especially those from large species such as sharks, tunas and marine catfish, from which a significant amount of viscera is wasted. It is also desirable to study the ensilation of viscera originating from different tropical regions so that a complete picture of tropical fish viscera ensilation is obtained. Studies on the characterisation of the proteases in such viscera are needed to provide more information on tropical fish enzymes in comparison with cold and temperate water fish enzymes, especially in relation to the possibility of using these enzymes as food processing aids in developing countries where the costs of commercial products are high. Investigations on the conversion of tropical fish viscera into peptones, either by ensilation or other methods such as hydrolysis using added proteases, followed by 176 further processing into peptone powder, are warranted. Such investigations should include a detailed economic feasibility of their commercial production in countries such as Indonesia, and a comparison with existing commercial products. 177 7 REFERENCES Ackman, R.G. (1992) Fatty acids in fish and shellfish. Chow, C.K. (ed.). Fatty Acids in Foods and Their Health Implications. New York: Marcel Dekker, Inc.; 169-184. Adams, M R., Cooke, R.D. and Rattagool, P. (1985) Fermented fish products of South East Asia. Trop. Sci. 25: 61-73. Adams, M.R. and Moss, M.O. (1995) Food Microbiology. Cambridge: The Royal Society of Chemistry. 398p. Adler-Nissen, J. (1986) Enzymic Hydrolysis of Food Proteins. London: Elsevier Appl. Science Publishers. 427p. Adler-Nissen, J. (1993) Proteases. Nagodawithana, T. and Reed, G. (eds.). Enzymes in Food Processing. 3rd ed. San Diego, Ca.: Academic Press, Inc.; 159-203. Ahmad, M. and Sonoda, S. (1979) Ekologi elasmobranchii air tawar: Ikan pari Sungai Indragiri, Riau (The ecology of freshwater elasmobranchs of the Indragiri River, Riau). Oceanol. Indon. 12:13-19. Ahmed, J. and Mahendrakar, N.S. (1995) Effect of different levels of molasses and salt on acid production and volume of fermenting mass during ensiling of tropical freshwater fish viscera. J. Food Sci. Technol. 32 (2): 115-118. Almas, K. (1990) Utilization of marine biomass for production of microbial growth media and biochemicals. Voit, M.N. and Botta, J.R. (eds ). Advances in Fisheries Technology' and Biotechnology for Increased Profitability. Lancaster, PA: Technomic Publ. Co., Inc.; 361-372. Alwan, S.R., Buckley, D.J. and O'Connor, T.P. (1993) Silage from fish waste: 178 Chemical and microbiological aspects. Irish J. Agric. Food Res. 32: 75-81. Amiza, M.A. and Apenten, R.K.O. (1994) Thermal inactivation for alkaline proteinase from North Sea cod (Gcidus morhua) and bovine cc-chymotrypsin. J. Sci. Food Agric. 66: 389-391. Amiza, M.A. and Apenten, R.K.O. (1996) Urea and heat unfolding of cold-adapted Atlantic cod (Gadus morhuci) trypsin and bovine trypsin. J. Sci. Food Agric. 70: 1-10. Anderson Jr., D.W. and Fellers, C.R. (1949) Preparation of sterile fish muscle press juice for chemical and bacteriological studies. J. Food Sci. 3 (8): 274. Anggawati, A. M., Indriati, N., Rahayu, S., Sabarudin and Madden, J.L. (1992) Use of dryacide, deltamethrin and cycloprothrin to control insect infestation on dried salted fish. Indo-Pacific Fishery Commission: Papers Presented at the Eighth Session of the Indo-Pacific Fishery Commission Working Party on Fish Technology and Marketing, Yogyakarta, Indonesia, 24-27 September 1991. FAOFish. Rep., (470) Suppl. Rome: FAO; 267-275. Anonymous (1973a) A Process for the Preparation of Bacteriological Peptones from Miscellaneous Fish. Indian Patent No. 137785. Anonymous (1973b) Difco Manual of Dehydrated Culture Media and Reagents for Microbiological and Clinical Laboratory Procedures. 9th ed. Detroit: Difco Laboratories Inc.; 264-265. Anonymous (1981) 1981 Annual Progi'ess Report of the ASEAN Working Group on the Management and Utilization of Food Waste Materials. Jakarta: ASEAN Committee on Science and Technology. 169p. Anonymous (1988) Dried and smoked fishery products: Preparation and composition. Burt, J.R. (ed ). Fish Smoking and Drying: The Effects of 179 Smoking and Drying on the Nutritional Properties of Fish. London: Elsevier Applied Science; 121-159. Anonymous (1995a) 1993 Fishery Statistics. Vo/. 77: Catches and Landings. Rome: Food and Agricultural Organization of the United Nations. 687p. Anonymous (1995b) 1993 Fishery Statistics. Vo/. 78: Commodities. Rome: Food and Agricultural Organization of the United Nations. 425p. Anonymous (1996) Fisheries Statistics of Indonesia 1994. Jakarta: Directorate General of Fisheries, Ministry of Agriculture. 73p. Aoki, T., Yamashita, T. and Ueno, R. (1995) Purification and characterization of a cathepsin B-like enzyme from mackerel white muscle. Fish. Sci. 61 (1): 121- 126. Arason, S. and Arnason, FI. (1992) Utilization of geothermal energy for drying of fish products. Geothermics 21 (506): 745-757. Ariyani, F. and Buckle, K.A. (1991) Ensiling of prawn heads. ASEAN Food J. 6 (2): 58-63. Ariyani, F., Fawzya, Y.N., Sari, A., Nasran, S. and Setiabudi, E. (1993) Effect of raw material handling on quality of dried-salted ray. Liang, O.B., Buchanan, A. and Fardiaz, D. (eds.). Development of Food Science and Technology’ in Southeast Asia. Bogor, Indonesia: IPB Press; 307-315. Ariyani, F., Murdinah and Putro, S. (1986) Feeding trials using feeds supplemented with fish meals and silages. Karossi, A.T., Surawidjaja, T.N. and Udin, L.Z. (eds ). Proceedings of the First ASEAN Workshop on Fish and Fish Waste Processing and Utilization, Jakarta, Indonesia, 22-24 October 1986; 292-299. Arunchalam, K. and Haard, N.F. (1985) Isolation and characterisation of pepsin from 180 polar cod (Boreogadus saida). Comp. Biochem. Physio/. 80B (3): 467-473. Asgeirsson, B., Fox, J.W. and Bjarnason, J.B. (1989) Purification and characterization of trypsin from the poikilotherm Gcidus morhta. Eur. J. Biochem. 180: 85-94. Asghar, A. and Bhatti, A.R. (1987) Endogenous proteolytic enzymes in skeletal muscle: Their significance in muscle physiology and during postmortem aging events in carcasses. Adv. Food Res. 31: 344-451. Auerbach, P.S., Burnett, J.W., Galli, R. and Neuman, T. (1994) Hook, fin, scuba, skin. Aquatic emergencies. Patient Care 28 (13): 70-74. Backhoff, H.P. (1976) Some chemical changes in fish silage. J. Food Technol. 11: 353-363. Bailey, J.L. (1967) Techniques in Protein Chemistry. London: Elsevier Pub. Co.; 340-341. Bandyopadhyay, S.M.S. and Bose, A.N. (1990) An improved solar dryer for fish drying in the coastal belt. J. FoodSci. Technol. 27 (3): 175-177. Barlow, S.M. and Stansby, M.E. (1982) Nutritional Evaluation of Long-Chain Fatty Acids in Fish Oil. London: Academic Press. 318p. Barret, A.J. (1972). Lysosomal enzymes. Dingle, J.T. (ed.). Lysosomes: A Laboratory Handbook. Amsterdam: North-Holland Publ. Co.; 46-135. Barrington, E.J.W. (1957) The alimentary canal and digestion. Brown, M.E. (ed.). The Physiology’ of Fish. Vol. 1: Metabolism. New York: Academic Press; 109-161. Barss, P. (1984) Wound necrosis caused by the venom of stingrays: Pathological findings and surgical management. Med. J. Aust. 141 (12-13): 844-845. 181 Basmal, J., Murtini, J.T. and Indriyati, N. (1995) Pengaruh konsentrasi asam formiat pada pembuatan silase dari limbah ikan pari (Trygon spp.). (Effect of formic acid concentration on the processing of silage from cowtail rays (Trygon sephen) waste) J. Penelitian Pasccipcinen Perikanan (J. Postharvest Fish. Res.) 84: 1-7. Basmal, J., Heruwati, E.S., Muljanah, I. and Kompiang, I P. (1988) Effect of substitution of fish meal with fish silage meal on the growth of Tiger prawn. Cited from Suparno and Poernomo (1991). Bell, D.J., Hoare, M. and Dunnil, P. (1983) The formation of protein precipitates and their centrifugal recovery. Fiechter, A. (ed.). Downstream Processing. Adv. Biochem. Eng. Biotechnol. 26: 1-72. Bergmeyer, H.U. (1974) Enzymes as biochemical reagents. Bergmeyer, H.U. (ed.). Methods of Enzymatic Analysis. Vo/. 1. New York: Academic Press, Inc.; 425-522. Beuchat, L.R. (1974) Preparation and evaluation of a microbiological growth medium formulated from catfish waste peptone. J. Milk Food Techno/. 37 (5): 277-281. Bond, C.E. (1979) Biology of Fishes. Philadelphia: Saunders College Publishing. 514p. Bone, Q. (1966) On the function of the two types of myotomal muscle fibre in elasmobranch fishes. J. Mar. Biol. Assoc. UK. 46: 576-588. Bone, Q., Marshall, N.B. and Blaxter, J.H.S. (1995) Biology of Fish. 2nd. ed. London: Blackie Academic & Professional. 332p. Borgstrom, G. (1962) Fish in world nutrition. Borgstrom, G. (ed.). Fish as Food. 182 Vol. II. Nutrition, Sanitation and Utilization. New York: Academic Press; 267-360. Boucher, I., Dupuy, A., Vidal, P., Neugebauer, W.A. and Brzezinski, R. (1992) Purification and characterization of chitosanase from Streptomyces N174. Appi Microbiol. Biotechno/. 38 (2): 188-193. Brewer, P., Helbig, N. and Haard, N.F. (1984) Atlantic cod pepsin: Characterization and use as a rennet substitute. Can. Inst. FoodSci. Technol. J. 17 (1): 38-43. Bridson, E.Y. (1990) The Oxoid Manual. 6th ed. Wade Rood, Hampshire: Unipath Ltd.; 3-1 to 3-19. Brinton, R. (1994) Composting research: Low cost options for fish waste. Biocycle March. 68-70. Brody, J. (1965) Fishery By-product Technology. Westport: AVI Publ. Co., Inc. 165p. Bryan, F.L. (1976) Staphylococcus aureus. Defigueiredo, M.P. and Splittstoesser, D.F. (eds.). Food Microbiology: Public Health and Spoilage Aspects. Westport, Conn.: The AVI Publ. Co. Inc., 12-128. Bundy, H.F. and Gustafson, J. (1973) Purification and comparative biochemistry of a protease from the starfish Pisaster giganteus. Comp. Biochem. Physiol. 44B: 241-251. Burhanuddin, Hutomo, M., Martosewojo, S. and Moeljanto, R. (1984) Sumberdaya Lemuru (Indian Oil Sardines Resources). Jakarta: LON-LIPI. 69p. Burt, JR. (1988) Fish Smoking and Drying: The Effects of Smoking and Dtying on the Nutritional Properties of Fish. London: Elsevier Applied Science. 166p. Carawan, R.E., Chambers, J.V. and Zall, R.R. (1979) Seafood Water and 183 Wastewater Management. Special report No. AM-18F. Greensboro, Raleigh: North Carolina Agricultural Extension Service. 126p. Caygill, J.C., Moore, D.J. and Kanagasbapathy, L. (1983) Concentration of plant proteases by precipitation with polyacrylic acids. Enzyme Microbiol. Technol. 5: 365-368. Cha, Y.J. and Cadwallader, K.R. (1995) Volatile components in salt fermented fish and shrimp paste. J. FoodSci. 80 (1): 18-24. Cheetam, P.S.J. (1995) The applications of enzyme in industry. Wiseman, A. (ed ). Handbook of Enzyme Biotechnology. London: Ellis Horwood; 419-552. Chen, G.Q. and Page, W.J. (1994) The effect of substrate on the molecular weight of poly-(3-hydroxybutyrate produced by Azotobacter vinelandii UWD. Biotechnol. Lett. 16 (2): 155-160. Chen, W. and Berg, J.C. (1993) The effect of polyelectrolyte dosage on floe formation in protein precipitation by polyelectrolyte. Chem. Eng. Sci. 48 (10): 1775-1784. Cheng, X., Ling, S., Yang, J and Zhang, S. (1994) Production and properties of exo- isomaltotriohydrolase from Alcaligenes sp. (in Chinese). Weishengwu Xuebao 34 (2): 124-130. Chng, N.M., Kuang, H.K. and Miwa, K. (1991) Southeast Asian Fish Products. Singapore: Marine Fisheries Research Department, Southeast Asian Fisheries Development Centre. 86p. Chng, N. M. (1992a) Determination of moisture. Miwa, K. and Ji, L.S. (eds ). Laboratory Manual on Analytical Methods and Procedures for Fish and Fish Products. 2nd edition. Singapore. Marine Fisheries Research Department, 184 Southeast Asian Fisheries Development Center; A-l.l - A-1.3. Chng, N. M. (1992b) Determination of ash. Miwa, K. and Ji, L.S. (eds.). Laboratory Manual on Analytical Methods and Procedures for Fish and Fish Products. 2nd edition. Singapore: Marine Fisheries Research Department, Southeast Asian Fisheries Development Center; A-2.1 - A-2.2. Clark, K.M. and Glatz, C.E. (1987) Polymer dosage considerations in polyelectrolyte precipitation of protein. Biotechnol. Prog. 3 (4): 241-247. Clark, K.M. and Glatz, C.E. (1988) Protein fractionation by precipitation with carboxymethyl cellulose. Hamel, J.F., Hunter, J.B. and Sikdar, S.K. (eds.). Downstream Processing and Bioseparation. Recovery and Purification of Biological Products. ACS Symposium Series 419. Washington, DC: American Chemical Society; 170-187. Clausen, E., Gildberg, A. and Raa, J. (1985) Preparation and testing of an autolysate of fish viscera as growth substrate for bacteria. Appl. Env. Microbiol. 50 (6): 1556-1557. Coakley, W.T. and James, C.J. (1978) A simple linear transform for the Folin-Lowry protein calibration curve to 1 mg/ml. Anal. Biochem. 85: 90-97. Crawford, J.H. (1985) Composting of agricultural waste. Cheremisinof, P.N. and Oueliette, R.P. (eds.). Biotechnology: Application and Research. Lancaster, PA: Technomic; 68-77. Crawford, D.L., Law, D.K. and Babbit, J.K. (1972) Nutritional characteristics of marine food fish carcass waste and machine-separated flesh. J. Agnc. Food Chem. 20 (5): 1048-1051. Cutting, C.L. (1955) Fish Saving: A History of Fish Processing from Ancient to Modern Times. London: Leonard Hill (Books) Ltd. 72p. 185 Cutting, C.L. (1962a) Historical aspects offish. Borgstrom, G. (ed./ Fish as Food. Vol. II. Nutrition, Sanitation and Utilisation. New York: Academic Press; 1- 27. Cutting, C.L. (1962b) The influence of drying, salting and smoking on the nutritive value of fish. Heen, E. and Kreuzer, R. (eds.). Fish in Nutrition. London: Fishing News (Books); 161-179. Daniel, D.J.E. and Etoh, S. (1983) Insect infestation of dry fish in Sri Lanka. James, D. (ed.). The Production and Storage of Dried Fish. Proceedings of the Workshop on the Production and Storage of Dried Fish. Universiti Pertanian Malaysia, Serdang, Malaysia, 2-5-November-1982. FAO Fish. Rep., (279) Suppl. Rome: FAO; 162-171. De Haro, L., Sanfacon, C., Jouglard, J. and Thomas, M. (1994) Human pathological disorders induced by stingrays (Rajiformes). Statement of the various problems brought by these fishes in a poison centre. Ichthyophysiol. Acta 17: 189-201. Dembinska, M. (1988) Methods of meat and fish preservation in the light archaeological and historical sources. Riddervold, A. and Ropeid, A. (eds.). Food Conservation: Ethnological Studies. London: Prospect Books; 13-24. Devadoss, P. (1978a) On the food of rays, Dasyatis ttarnak (Forskal), D. alcockii (Annandale) and D. sephen (Forskal). Ind. J. Fish. 25(1-2): 9-13. Devadoss, P. (1978b) A preliminary study on the Batoid fishery of Cuddalore with a note on the biology. Ind. J. Fish. 25 (1-2): 180-187. Dionysius, DA, Hoek, K.S., Milne, J.M. and Slattery, S.L. (1993) Trypsin-like enzyme from sand crab (Port units pe/agicus): Purification and 186 characterization. J. FoodSci. 58 (4): 780-784,792. Disney, J. G. and Hoffman, A. (1976) A dried fish silage product. Proc. Tony Res. Stn. Symp. Fish Silage, Torry Research Station, Aberdeen. Cited from Raa and Gildberg (1982). Doe, P.E., Ahmed, M., Muslemudin, M. and Sachithananthan, K. (1977) A polythene tent dryer for improved sun-drying of fish. Food Technol. Aust. 29 (11): 437, 439-441. Doe, PE. and Heruwati, E.S. (1988) Drying and storage of tropical fish - A model for the prediction of microbial spoilage. Seow, C.C., Teng, T.T. and Quah, C.H. (eds.). Food Preservation by Moisture Control. London: Elsevier Applied Science Publisher; 117-135. Doe, P.E. and Olley, J. (1990) Drying and dried fish products. Sikorski, Z.E. (ed ). Seafood: Resources, Nutritional Composition and Preservation. Boca Raton: CRC Press, Inc.; 125-146. Dong, F.M., Fairgrieve, W.T., Skonberg, D.I. and Rasco, B.A. (1991) Preparation and nutrient analyses of lactic acid bacterial ensiled salmon viscera. Aquaculture 109: 351-366. Dunham, R.A., Joyce, J.A., Bondari, K. and Malvestuto, S.P. (1985) Evaluation of body conformation, composition and density as traits for indirect selection for dress-out percentage of Channel catfish. Prog. Fish Cult. 47 (3): 169-175. Dunn, B.M. (1991) Structure and Function of the Aspartic Proteinases: Genetics, Structures and Mechanisms. New York: Plennum Press. 585p. Edin, H. (1940) Undersokningar angaande importavstengningens aggviteproblem. Nord. Jordbr. Forsk. 22: 142- . Cited from Raa and Gildberg (1982). 187 Ellis, K.R. (1990) Worldwide protein market development. Keller, S. (ed.). Making Profits out of Seafood Wastes. Proceedings of the International Conference on Fish By-products, Anchorage, Alaska - April 25-27, 1990. Alaska Sea Grant College Program, Report No 90-07; 15-19. EPA (1974) Development Document for Effluent Limitations Guidelines and Standards of Performance for the Catfish, Crab, Shrimp and Tuna Segments of the Canned and Preserved Seafood Processing Industry Point Source Category. EPA 440/1-74/020. US Environmental Protection Agency. Washington. D C. Cited from Wheaton and Lawson (1985). Esser, J.R. (1991) Biology of Chrysomya megacephala (Diptera: Calliphoridae) and reduction of losses caused to the salted-dried fish industry in South-East Asia. Bull. Entomol. Res. 81(1): 33-41. Esser, JR., Hanson, S.W. Wiryante, J., Sunarya and Tausin, S. (1990) Prevention of insect infestation and losses of salted-dried fish in Indonesia by treatment with an insecticide approved for use on fish. Indo-Pacific Fishery Commission: Papers Presented at the Seventh Session of the Indo-Pacific Fisheiy Commission Working Party on Fish Technology’ and Marketing, Bangkok, Thailand, 19-22 April 1988. FAO Fish. Rep., (401) Suppl. Rome: FAO; 168- 179. Etoh, S. (1982) Fishmeal from by-catch on a cottage industryt scale. Infofish Mark. Digest. 5: 24-28. FAO (1981) The prevention of losses in cured fish. Fish. Tech. Pap., (219). 87p. Farrand, A.L. and Williams, D.C. (1988) Isolation, purification and partial characterisation of four digestive proteases from the purple seastar (Pisaster 188 ochraceus). Mar. Biol. 97: 231-236. Fawzya, Y.N., Ariyani, F., Sari, A., Nasran, S. and Setiabudi, E. (1993) Quality of salted rays. Liang, OB., Buchanan, A. and Fardiaz, D. (eds ). Development of Food Science and Technology in Southeast Asia. Bogor, Indonesia: IPB Press; 300-306. Fawzya, Y.N., Muljanah, I. And Nasran, S. (1994) Pengaruh pencucian daging lumat dan konsentrasi bahan pengikat terhadap mutu dan daya awet sosis ikan pari (Trygon sephen). (Effect of minced meat washing and binding agents on the quality and storage life of sausage from cowtail ray (Trygon sephen)). J. Pene/itian Pascapanen Perikanan (J. Postharvest Fish. Res.) 77: 50-59. Finch, R. (1977) Whatever happened to fish protein concentrate? Food Technol. 31 (5): 44-53. Fischer, R.R. and Glatz, C.E. (1988) Polyelectrolyte precipitation of proteins: I. The effect of reactor conditions. Biotechnol. Bioeng. 32: 777-785. Foye, WO. (1977) Antimicrobial activities of mineral elements. Weinberg, E.D. (ed.). Microorganisms and Minerals. New York: Marcel Dekker, Inc.; 387- 419. Galgani, F. and Nagayama, F. (1987) Digestive proteinases in the Japanese spiny lobster Panulirus japonicus. Comp. Biochem. Physiol. 87B (4): 889-893. Gao, Y., Lo, K.V. and Liao, P.H. (1992) Utilization of salmon farm mortalities: Fish silage. Bioresource Technol. 41: 123-127. Garraway, M.O. and Evans, R.C. (1984) Fungal Nutrition and Physiology. New York: John Willey & Sons; 401 p. Garcia-Carreno, F.L. (1993) Proteases inhibition in theory and practice. Biotechnol. 189 Educ. 3 (4): 145-150. Garcia-Carreno, F.L. and Haard, N.F. (1993) Characterization of proteinase classes in langostilla (Pleuroncodes planipes) and crayfish (Pacifastacus ctstacus) extracts. J. FoodBiochem. 17: 97-113. Garman, S. (1913) The P/agiostomata (Sharks, Skates, and Rays). Memoirs of the Museum of Comparative Zoology at Harvard College. Vol. XXXVI. Book 1. Garthwaite, G.A. (1992) Chilling and freezing of fish. Hall, G.M. (ed.). Fish Processing Technology. London: Blackie Academic & Professional; 89-113. Gibsoon, R. A., Kneebone, R. and Kneebone, G.M. (1984) Comparative levels of arachidonic acid and eicosapentaenoic acid in Malaysian fish. Comp. Biochem. Physiol. 78C (2): 325-328. Gildberg, A. (1982) Autolysis of Fish Tissue: General Aspects. Tromso: University of Tromso. PhD thesis. Gildberg, A. (1988) Aspartic proteinases in fishes and aquatic invertebrates. Comp. Biochem. Physiol. 91B (3): 425-435. Gildberg, A. (1989) Accelerated fish sauce fermentation by initial alkalification at low salt concentration. Miyachi, S., Karube, J. and Ishida, Y. (eds.). Current Topics in Marine Biotechnology. Tokyo: Fuji Technol. Press; 101-104. Cited from Gildberg (1992). Gildberg, A. (1992). Recovery of proteinases and protein hydrolysates from fish viscera. Bioresource Technol. 39: 271-276. Gildberg, A. (1993) Enzymic processing of marine raw materials. Process Biochem. 28: 1-15. Gildberg, A. and Almas, K.A. (1986) Utilization of fish viscera. Maguer, M.L. and Jelen, P. (eds ). Food Engineering and Process Applications. Vol. 2: Unit 190 Operations. London: Elsevier Applied Science Publishers; 383-393. Gildberg, A., Batista, I. and Strom, E. (1989) Preparation and characterization of peptones obtained by a two-step enzymatic hydrolysis of whole fish. Biotechno/. Appl. Biochem. 11: 413-423. Gildberg, A., Olsen, R.L. and Bjarnason, J.B. (1990) Catalytic properties and chemical composition of pepsins from Atlantic cod (Gadus morhud). Comp. Biochem. Physiol. 96B (2): 323-330. Gildberg, A., Olsen, R.L. and Bjarnason, J.B. (1991) Characteristics and composition of pepsins from Atlantic cod. Dunn, B.M. (ed.). Structure and Function of the Aspartic Proteinoses: Genetics, Structures and Mechanisms. New York: Plenum Press; 107-110. Gildberg, A. and Raa, J. (1977) Properties of propionic acid/formic acid preserved silage of cod viscera. J. Sci. Food Agric. 28. 647-653. Gildberg, A. and Raa, J. (1983) Purification and characterization of pepsins from Arctic fish capelin (Mal/otus villosus). Comp. Biochem. Physiol. 75A: 337- 342. Gildberg, A. and Xian-Quang, S. (1994) Recovery of tryptic enzymes from fish sauce. Process Biochem. 29: 151-154. Gillies, M.T. (1975) Fish and Shellfish Processing. New Jersey: Noyes Data Corporation. 338p. Godfrey, T. and Reichelt, J.R. (1983) Introduction to enzymology. Godfrey, T. and Reichelt, J.R. (eds ). Industrial Enzymology: The Application of Enzymes in Industry. New York: The Nature Press; 1-7. Golueke, C.G. (1981) Principles of biological resources recovery. BioCycle 22: 36- 40. 191 Gomon, M.F., Glover, J.C.M. and Kuiter, R.H. (1994) The Fishes of Australia's South Coast. Adelaide: State Print. 992p. Gopakumar, K. (1995) Salted dried fish in India. Champ, B.R. and Highley, E. (eds.). Fish Drying in Indonesia. Proceedings of an International Workshop held at Jakarta, Indonesia, 9-10 February 1994. ACIAR Proceedings No. 59. Canberra: Australian Centre for International Agricultural Research; 83-88. Gorbenko, A. Yu. and Dzyuban, A N. (1992) On the possibility of oligocarbophilic (oligotrophic) bacterioplankton counting on the rich medium (in Russian). Mikrobiologiya 61 (3): 531-533. Gordon, D.T. (1988) Minerals in seafoods: Their bioavailability and interactions. Food Technol. 42 (5): 156-160. Graham, J. (1982) Freezing. Aitken, A., Mackie, I.M., Merrit, J.H. and Windsor, M.L. (eds.). Fish Handling and Processing. 2nd ed. Aberdeen: Min. of Agriculture, Fisheries & Food, Torry Research Station; 56-71. Grant, E.M. (1982) Guide to Fishes. 5th ed. Brisbane, Qld.: Dept, of Harbours and Marine. 890p. Grant, E.M. (1987) Fishes of Australia. 1st ed. Scarborough, Qld.; E.M. Grant Pty. Ltd. 480p. Green, J.H. (1974) Mushroom culture: A new potential for fishery products. Mar. Fish. Rev. 36: 27-32. Green. J.H., Goldmintz, D., Hale, M B and Flynn, D.J (1973) Exploration of experimentally produced fish peptones for growth of microorganisms. Develop. Indus. Microbiol. 14: 310-316. Green, J.H., Paskell, S.L. and Goldmintz, D. (1977) Fish peptones for microbial 192 media developed from red hake and from fishery by-product. J. Food. Protect. 40 (3): 181-186. Greene, C.W. and Greene, C.H. (1914) The skeletal musculature of the king salmon. Bull. US. Bur. Fish. 33: 21-60. Cited from Bond (1979). Greig, R.I.W. and Estrella, D C. (1988) A study on the acceleration of fish sauce production using enzymes. Maneepun, S., Varangoon, P. and Phithakpol, B. (eds.). Food Science and Technology> in Industrial Development. Vol. II. Proceedings of the Food Conference '88, Bangkok, Thailand, 24-26 October 1988. Bangkok: Institute of Food Research and Product Development, Kasetsart University; 275-281. Groves, D. G. and Hunt, L.M. (1980) Ocean World Encyclopedia. New York: McGraw-Hill Book Co.; 304-304. Guerard, F. and Le Gal, Y. (1987) Characterization of a chymosin-like pepsin from the dogfish Scyliorhinus canicula. Comp. Biochem. Physiol. 88B (3): 823- 827. Guirard, B. M. and Snell, E.E. (1962) Nutritional requirements of microorganisms. Gunsalus, I.C. and Stanier, R.Y. (eds ). The Bacteria: A Treatise on Structure and Function. Vol. IV: The Physiology of Growth. New York: Academic Press; 33-93. Haard, N.F. (1992a) Biochemistry and chemistry of color and color change in seafoods. Flick, Jr., G.J. and Martin, R.E. (eds.). Advances in Seafood Biochemistry: Composition and Quality. Lancaster, PA. Technomic Pub. Co. Inc.; 305-360. Haard, N.F. (1992b) A review of proteolytic enzymes from marine organisms and 193 their application in the food industry. J. Aqua. Prod. Techno/. 1(1): 17-35. Haard, N.F., Feltham, L A W., Helbig, N. and Squires, E.J. (1982) Modification of proteins with proteolytic enzymes from the marine environment. Feeney, R.E. and Whitaker, J.R. (eds.). Modification of Proteins: Food, Nutritional and Pharmacological Aspects. ACS Symposium Series 198. Washington: American Chemical Society; 223-244. Haard, N.F., Simpson, B.K. and Pan, B.S. (1994) Sarcoplasmic proteins and other nitrogenous compounds. Sikorski, Z.E., Pan, B.S. and Shahidi, F. (eds.). Seafood Proteins. New York: Chapman & Hall; 13-39. Haard, N.F., Simpson, B.K. and Sikorski, Z.E. (1994) Biotechnological applications of seafood proteins and other nitrogenous compounds. Sikorski, Z.E., Pan, B.S. and Shahidi, F. (eds ). Seafood Proteins. New York: Chapman & Hall; 194-216. Hak, N., Heruwati, E.S., Sari, A., and Setiabudi, E. (1990) Pengaruh kadar natrium sulfida (Na2S) dalam penyamakan kulit ikan pari terhadap mutu kulit tersamaknya. (Effect of sodium sulphide content in the tanning of rays' skin on the quality of rays' leather). J. Penelitian Pascapanen Perikanan {J. Postharvest Fish. Res.) 66: 27-36. Hall, S. (1991) Natural toxins. Ward, D.R. and Hackney, C. (eds.). Microbiology of Marine Food Products. New York: Van Nostrand Reinhold; 301-330. Hall, G.M. and Ahmad, N.H. (1994) Surimi and fish mince products. . Hall, G.M. (ed.). Fish Processing Technology. London: Blackie Academic & Professional; 72-88. Hall, G.M. and Ledward, D A. (1986). Silage from tropical fish: 3. Lipid behaviour. 194 J. Food Techno!. 21: 45-54. Hall, G.M., Ledward, D.A., and Lawrie, R.A. (1985). Silage from tropical fish: 2. Undigested fraction. J. Food Techno!. 20: 573-580. Hall, G.M., Keeble, D , Ledward, D A. and Lawrie, R.A. (1985) Silage from tropical fish: 1. Proteolysis. J. Food Techno!. 20: 561-572. Halstead, R.W. (1970) Poisonous and venomous marine animals of the world. Washington: US Government Printing Office; 615, 621, 655. (Cited from Barss 1974). Hansen, M.E. and Illanes, A. (1994) Applications of crustacean wastes in biotechnology. Martin, A.M. (ed.). Fisheries Processing: Biotechnological Applications. London: Chapman & Hall; 174-205. Hara, K., Arano, H. and Ishihara, T. (1984) Some enzymatic properties of alkaline proteinases from the rotifer, Brachionus phicatilis. Bull. Jap. Soc. Sci. Fish. 50: 1611-1616. Hardy, R.W. (1992) Fish processing by-products and their reclamation. Pearson, A.M. and Dutson, T.R. (eds.). Inedible Meat By-products: Advances in Meat Research Vo! 8. London: Elsevier Applied Science; 199-216. Heu, M.S., Pyeun, J.H., Kim, H.R. and Godber, J.S. (1991) Purification and characterization of alkaline proteinases from the viscera of anchovy, Engraulis japonica. J. FoodBiochem. 15: 51-66. Hjelmeland, K. and Raa, J. (1982) Characteristics of two trypsin type isoenzymes isolated from the arctic fish capelin. Comp. Biochem. Physiol. 7IB; 557-562. Hokama, Y., Asahina, A.Y., Titus, E., Shirai, J.L.R., Hong, T.W.P., Chun, S., Miyahara, J.T., Takata, D, Muranaka, A., Pang, E., Abbott, I.A. and Ichinotsubo, D. (1993) A survey of ciguatera; Assessment of Puako, Hawaii, 195 associated with ciguatera toxin epidemics in humans. J. Clin. Lab. Anal. 7 (3): 147-154. Hood, L.F. and Zall, R.R. (1980) Recovery, utilization and treatment of seafood processing wastes. Connell, J.J. (ed). Advances in Fish Science and Technology’. Farnham, Surrey: Fishing News Books Ltd.; 355-361. Hood, L.F., Zall, R.R. and Conway, R.L. (1976) Conversion of minced clam washwater into clam juice: Waste handling or product development? Food Prod. Devel. 10: 86-88. Horner, W.F.A. (1992a) Preservation of fish by curing (drying, salting and smoking). Hall, G.M. (ed ). Fish Processing Technology. London: Blackie Academic & Professional; 31-72. Horner, W.F.A. (1992b) Canning fish and fish products. Hall, G.M. (ed.). Fish Processing Technology. London: Blackie Academic & Professional; 114-154. Huei-Mei, C. and Meyers, S.P. (1983). Ensilage treatment of crawfish waste for improvement of astaxanthin pigment extraction. J. Food Sci. 48: 1516- 1520/1555. Hutchins, J.B. (1980) Dangerous Fish of Western Australia. Perth: Western Australia Museum. 49p. Huss, H.H. (1988) Fresh Fish: Quality and Quality Changes. Rome: Food and Agricultural Organization of the United Nations. 132p. Ijong, F.G. and Ohta, Y. (1995) Amino acid compositions of bakasang: A traditional fermented fish sauce from Indonesia. Lebensm. Wiss. U.-Techno!. 28: 236- 237. Ilyas, S., Nasran, S. and Irianto, H.E. (1988) Penelitian pendahuluan pengolahan 196 daging lumat ikan pari (Trygon sephen). (Preliminary study on the processing of mince meat from cowtail ray ( Trygon sephen)). J. Penelitian Pasca Panen Perikanan (./. Pos/harvest Fish. Res.) 60: 39-46. In, T. (1990) Seafood flavourants produced by enzymatic hydrolysis. Voit, M.N. and Botta, J.R. (eds.). Advances in Fisheries Technology’ and Biotechnology for Increased Profitability. Lancaster, PA: Technomic Publ. Co., Inc.; 425-536. Indriati, N., Sudrajat, Y., Madden, J., Anggawati, A M. and Putro, S. (1985) Insect infestation in dried-salted fish in Java. Reilly, A. (ed). Spoilage of Tropical Fish and Product Development. Proceedings of a Symposium held in conjunction with the Sixth Session of the Indo-Pacific Fishery Commission Working Party on Pish Technology and Marketing. Royal Melbourne Institute of Technology, Melbourne, Australia, 23-26 October 1984. FAO Fish. Rep., (317) Suppl. Rome: FAO; 297-301. Ingham, S.C. (1991) Microbiology of mince, surimi and value-added seafoods. Ward, D.R. and Hackney, C. (eds.). Microbiology’ of Marine Food Products. New York: Van Nostrand Reinhold; 89-110. Ismail, M.S. (1983) Solar dryers for fish. InfofishMark. Dig. 2: 31-33. James, C.S. (1995) Analytical Chemistry of Foods. London: Blackie Academic & Professional. 175p. James, M.A., Iyer, K.M. and Nair, M R. (1977) Comparative study of fish ensilage prepared by microbial fermentation and formic acid ensilage. Proceedings of the Conference on the Handling, Processing and Marketing of Tropical Fish, London, 5-9 July 1976. London: Tropical Products Institute; 273-276. Jany, K.D. (1976) Studies on the digestive enzymes of the stomachless bonefish 197 Ccircissius aurcitus gibelio (Bloch): Endopeptidases. Comp. Biochem. Physiol. 53B: 31-38 Jaouen, P. and Quemeneur, F. (1994) Membrane filtration for waste-water protein recovery. Hall, G.M. (ed.). Fish Processing Technology. London: Blackie Academic & Professional; 212-248. Jassim, S., Salt, W.G. and Stretton, R.J. (1988) The preparation and use of media based on a simple fish waste extract. Lett. Appl. Microbiol. 6: 139-143. Jayawardena, K.M. and Poulter, R.G. (1980) Studies on the preparation of fish silage: II. Rate of liquefaction in different parts of silver belly. Bull. Fish. Res. Stn. Sri Lanka. 30: 25-31. Jendrisak, J. (1987) The use of polyethyleneimine in protein purification. Burgess, R. (ed ). Protein Purification: Micro to Macro. New York: Alan R. Liss, Inc.; 75-97. Jensen, J. and Schmidtsdorff, F. (1977) Fish silage: Low fat and soluble fish protein products. IAFMM Symposium, Szczecin, Poland. Cited form Strom et al. (1980). Jiang, S.T., Moody, M.W. and Chen, H.C. (1991) Purification and characterization of proteases from digestive tract of grass shrimp (Penaeus monodon). J. FoodSci. 56 (2): 322-326. Johnsen, F. and Skrede, A. (1981) Evaluation of fish viscera as feed resource. Acta Agric. Scand. 31: 21-28. Johnson, W.C. and Lindsey, A.J. (1939) An improved universal buffer. Analyst, (Lond.) 64: 490-492. Joseph, J., Prabhu, P.V. and Madhavan, P. (1987) Utilisation of squid waste as meal. 198 Fish. Techno!. 24 (1): 41-43. Keil, B (1971) Trypsin. Boyer, P.D. (ed). The Enzymes. Vol III. 3rd edition. New York: Academic Press; 249-275. Kikuchi, R., Takayama, Y, Higashi, A. and Boedimartojo, H. (1972) Nutritive components in salted and dried fish and shellfish on the Indonesian market, (in Japanese). Bull. Tokai. Reg. Fish. Res. Lab. 70 (Aug): 95-102. Kilara, A. (1985) Enzyme-modified protein food ingredients. Process Biochem. 20 (5): 149-158. Kim, H.R. and Pyeun, J.H. (1986) The proteinases distributed in the intestinal organs of fish. 2. Characterization of three alkaline proteinases from the pyloric caeca of mackerel, Scomber japonicus. Bull. Korean Fish. Soc. 19 (6): 547- 557. Kim, H.R., Meyers, S.P. and Godber, J.H.S. (1996) Anionic trypsin from crayfish hepatopancreas: Effects on protein degradation of tail meat. J. Food Sci. 61 (1): 78-80, 96. Kimumaki, T. and Arai, K (1987) Nutritive values of great blue shark viscera meal, (in Japanese). Bull. Tokai Reg. Fish. Res. Lab. 122: 23-29. Kinsella, J.E. (1987) Seafoods and Fish Oils in Human Health and Disease. New York: Marcel Dekker, Inc. 317p. Kinsella, J.E. (1988) Fish and seafoods: Nutritional implications and quality issues. Food Technol. 42 (5): 146-150. Kirk, R.S. and Sawyer, R. (1991) Pearson's Composition and Analysis of Foods. Ninth edition. Burnt Mill, Harlow: Longman Scientific & Technical. 708p. Kleimenov, I.Ya. (1983) Nutritive value of fish. Novikov, V.M. (ed.). Handbook of 199 Fishery Technology. Vol. 1. Rotterdam: A.A. Balkema; 3-42. Knorr, D. (1991) Recovery and utilisation of chitin and chitosan in food processing waste management. Food Techno!. 45 (1): 114-122. Knorr, D. and Sinskey, A.J. (1985) Biotechnology in food production and processing. Sci. 299: 1224-1229. Kolodziejska, I. and Sikorsksi, Z.E. (1995) The properties and utilization of proteases of marine fish and invertebrates. Pol. J. Food Nutr. Sci. 4/45 (1): 5- 12. Kompiang, I.P. (1990) Fish silage and tepsil production. Indon. Agiic. Res. Dev. J. 12 (4): 55-59. Kompiang, I P., Arifudin, R. and Raa, J. (1980) Nutritional value of ensilaged by- catch fish from Indonesian shrimp trawlers. Connell, J.J. (ed ). Advances in Fish Science and Technology. Farnham, Surrey, England: Fishing News Books Ltd.; 349-352. Konagaya, S. (1980) Protease activity and autolysis of antarctic krill (in Japanese). Bull. Jap. Soc. Sci. Fish. 46: 175-183. Krishnaswamy, M.A. and Lahiry, N.L. (1963) Fish hydrolysates. IV. Microbiological evaluation. J. Food Sci. 28: 358-364. Kristjansson, M. (1991) Purification and characterization of trypsin from the pyloric caeca of rainbow trout (Oncorhynchus mykiss). J. Agric. Food Chem. 39: 1738-1742. Krogdahl, A. (1985a) Fish viscera silage as a protein source for poultry. I. Experiments with layer-type chicks and hens. Acta Agric. Scand. 35: 3-23. Krogdahl, A. (1985b) Fish viscera silage as a protein source for poultry. II. Experiments with meat-type chickens and ducks. Acta Agric. Scand. 35: 24- 200 32. Kulikov, PI. (1978) Production of Meal, Oil and Protein-Vitamin Preparations m the Fishing Industry. 3rd ed. New Delhi: Amerind Pub. Co. Pvt. Ltd. 253p. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 277: 680-685. Lagler, K.F., Bardach, J.E., Miller, R.R. and Passino, D.R.M. (1977) Ichthyology. 2nd ed. New York: John Wiley & Sons, Inc. 506p. Lassen, S. (1965) Fish solubles. Borgstrom, G. (ed ). Fish as Food. Vol. Ill: Processing: Part I. New York: Academic Press; 281-299. Last, PR. and Stevens, J.D. (1994) Sharks and Rays of Australia. Hobart: CSIRO. 513p. Lee, A. and Langer, R. (1984) Shark cartilage contains an inhibitor of tumor neovascularization. Colwell, R.R., Sinskey, A.J. and Pariser, E.R. (eds.). Biotechnology in the Marine Sciences. New York: John Wiley & Sons; 215- 218. Lee, C.M. (1984) Surimi process technology. Food Techno!. 38 (11): 69-80. Lee, S.C. and Woo, K.L. (1992) Development of an effective utilisation method for skipjack tuna viscera, (in Korean). Korean J. Food Sci. Technol. 24 (1): 86- 91. Lee, Y Z , Simpson, B.K. and Haard, N.F. (1982) Supplementation of squid fermentation with proteolytic enzymes. J. FoodBiochem. 6: 127-135. Lees, R.S. and Karel, M. (1990) Omega-3 Fatty Acids in Health and Disease. New York: Marcel Dekker, Inc. 240p. 201 Levanidov, I.P. (1980) Interrelation of the main components of the chemical composition of fish flesh. Rybn. Khoz. 8: 62- (in Russian). Cited from Sikorski, Kolakowska and Pan (1990). Levin, R.E., Witkowski, R., Meirong, Y. and Goldhor, S. (1989) Research note: Preparation of fish silage with phosphoric acid and potassium sorbate. J. Food Biochem. 12: 253-259. Liao, P.H., Vizcarra, A. and Lo, K.V. (1994) Composting of salmon-farm mortalities. Biores. Techno/. 47: 67-71. Licciardello, J.J. and Ravesi, E.M. (1988) Frozen storage characteristics of cownose ray (Rh inop (era bonassus). J. Food Oual. 11:71-76. Lima dos Santos, CAM. (1981) The storage of tropical fish - A review. Trop. Sci. 23: 97-127. Lin, T. M., Park, J. W. and Morrissey, T. (1995) Recovered protein and reconditioned water from surimi processing waste. J. Food Sci. 60 (1): 4-9. Lindgren, S. and Pleje, M. (1983) Silage fermentation of fish and fish waste with lactic acid bacteria. J. Sci. FoodAgric. 34: 1057-1067. Line, M.A. (1994) Recycling of seastar {Asterias cimurensis) waste by composting. Biores. Techno/. 49: 227-229. Lineweaver, H. and Burk, D. (1934) The determination of enzyme dissociation constant. J. Am. Chem. Soc. 56: 658-666. Liu, L.L. and Pigott, G.M. (1981) Preparation and use of inexpensive crude pepsin for enzyme hydrolysis of fish. J. Food Sci. 46: 1569-1572. Lo, K.V., Liao, PH. and Bullock, C. (1993) Silage production from salmon farm mortalities. Aquaculture Eng. 12:37-45. 202 Love, R.M. (1970) The Chemical Biology of Fish. London: Academic Press. 547p. Love, R.M. (1980) The Chemical Biology of Fish. Vol. 2. London: Academic Press. 400p. Love, R.M. (1982) Basic facts about fish. Aitken, A., Mackie, I.M., Merrit, J.H. and Windsor, M L. (eds.). Fish Handling and Processing. 2nd ed. Aberdeen: Min. of Agriculture, Fisheries & Food, Torry Research Station; 2-19. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275. Lytle, J.S. and Lytle, T.F. (1994) Fatty acid and cholesterol content of sharks and rays. J. Food Composit. Anal. 7: 110-118. Mackie, I.M. (1982a) General review of protein hydrolysates. Anim. Feed Sci. Techno!. 7: 113-124. Mackie, I.M. (1982b) Fish protein hydrolysates. Process Biochem. 17 (1): 26-31. Mackie, I.M. (1983) New approaches in the use of fish proteins. Hudson, B.J.F. (ed ). Developments in Food Proteins. Vol 2. London: Applied Science; 215- 262. Mackie, I.M. (1994) Fish protein. Hudson, B.J.F. (ed.). New and Developing Sources of Food Protein. London: Chapman & Hall; 95-143. Madden, J.L., Anggawati, A.M. and Indriati, N. (1995) Impact of insects on the quality and quantity of fish and fish products in Indonesia. Champ, B.R. and Highley, E. (eds). Fish Drying in Indonesia. Proceedings of an International Workshop held at Jakarta, Indonesia, 9-10 February 1994. ACIAR Proceedings No. 59. Canberra: Australian Centre for International Agricultural Research; 97-106. 203 Mahendrakar, N.S., Khabade, V.S., Yashoda, K.P. and Dani, N.P. (1991) Chemical and microbiological changes during autolysis of fish and poultry viscera. Trap. Sci . 31:45-54. Marti, C., Roeckel, M., Aspe, E. and Kanda, H. (1994) Recovery of proteins from fishmeal factory wastewaters. Process Biochem. 29: 39-46. Martin, A.M. and Chintalapati, S.P (1989) Fish offal-peat compost extracts as fermentation substrate. Biol. Waste 27:281-288. Martin, A.M. and Chintalapati, S.P. (1990) Composting of fish offal with peat and its use in the production of microbial biomass. Voit, M.N. and Botta, J.R. (eds.). Advances in Fisheries Technology> and Biotechnology’ for Increased Profitability. Lancaster, PA: Technomic Publ. Co., Inc.; 387-394. Martinez, A., and Olsen, R.L. (1986) Studies on pepsin from cod stomach. Abstract presented at the 22nd Annual Meeting on the Norwegian Biochemical Society. Cited from Gildberg (1988). Martinez, A., Olsen, R.L. and Serra, J.L. (1988). Purification and characterisation of two trypsin-like enzymes from the digestive tract of anchovy (Engraulis encrasicholus). Comp. Biochem. Physiol. 9IB (4): 677-684. Mathur, S.P., Daigle, J.Y., Brooks, J.L., Levesque, M. and Arsenault, J. (1988) Composting seafood wastes - Avoiding disposal problems. BioCycle 29: 44- 49. Mayer, B.K. and Ward, DR. (1991) Microbiology of finfish and finfish processing. Ward, D.R. and Hackney, C. (eds.). Microbiology of Marine Food Products. New York. Van Nostrand Reinhold; 3-17. Maynard, J. (1983) Dried fish. An Asian staple food. Infofish Mark. Dig. 2: 18-22. 204 McCulloch, A. (1927) The Fishes and Fish-like Animals of New South Wales. Sydney: Royal Zoological Society of New South Wales. McLay, R. (1980) Activities of cathepsin A and D in cod muscle. J. Sci. Food Agric. 31: 1050-1054. McPhie, P. (1976) A turbidimetric milk-clotting activity assay for pepsin. Anal. Biochem. 73:258-261. Miler, K.B.M. and Sikorski, Z.E. (1990) Smoking. Sikorsky, Z. (ed.). Seafood: Resources, Nutritional Composition and Preservation. Boca Raton: CRC Press, Inc.; 163-180. Miyata, Y. (1984) Concentration of protein from the waste water of red meat fish by ultrafiltration membrane, (in Japanese). Bull. Jap. Soc. Sci.. Fish. 50 (4): 659-663. Montgomery, D.C. (1991) Design and Analysis of Experiments. Third edition. New York: John Wiley & Sons. 649p. Moyle, P.B. and Cech, Jr., J.J. (1982) Fishes: An Introduction to Ichthyology. Englewood Cliffs, N.J.: Prentice-Hall, Inc.; 593p. Murado, M.A., Gonzalez, Ma. P. and Pastrana, L. (1994) Mussel processing wastes as a fermentation substrate. Martin, A.M. (ed ). Fisheries Processing: Biotechnological Applications. London: Chapman & Hall; 310-343. Murakami, K. and Noda, M. (1981) Studies on proteinases from the digestive organs of sardines. 1 Purification and characterization of three alkaline proteinases from the pyloric caeca. Biochim. Biophys. Acta 658: 17-26. Myer, R.O., Johnson, D.D., Otwell, W.S., Walker, W.R. and Combs, G.E. (1990) Evaluation of scallop viscera silage as a high-protein feedstuff for growing - 205 finishing swine. Anim. FeedSci. Technol. 31: 43-53. Narkviroj, P. (1987) The Utilisation of Prawn Processing Wastes. Kensington: The University of New South Wales. PhD thesis. 260p. Narkviroj, P. and Buckle, K.A. (1987) Utilization of prawn head powder in oriental prawn crackers. ASEAN Food J. 3(1): 21-23. Nasran, S., Ariyani, F., Fawzya, N. and Sari, A. (1992) Processing of dried salted ray at Muara Angke. Paper presented at the eighth session of the Indo-Pacific Commission Working Party on Fish Technology and Marketing. Yogyakarta, Indonesia, 24-27 September 1991. FAO Fish. Rep., (470) Suppl. Rome: FAO; 232-246. Nelson, J.S. (1994) Fishes of the World. 3rd ed. New York: John Wiley & Sons, Inc. 600p. Nettleton, J.A. (1985) Seafood Nutrition. New York: Osprey Book. Cited from Kinsella (1988). Nettleton, J.A. (1995) Omega-3 Fatty Acids and Health. New York: Chapman & Hall. 359p. Neurath, H. (1989) The diversity of proteolytic enzymes. Beynon, R.J. and Bond, J.S. (eds.). Proteolytic Enzymes: A Practical Approach. Oxford: IRL Press; 1-13. Nieman, C. (1954) Influence of trace amounts of fatty acid on the growth of microorganisms. Bacteriol. Rev. 18: 147-163. Niki, H., Kato, T., Deya, E. and Igarashi, S. (1985) Recovery of protein from effluent of fish meat producing surimi and utilization of recovered protein, (in Japanese). Bull. Jap. Soc. Sci.. Fish. 51 (6): 959-964. 206 Ninomiya, K., Ookawa, T., Tsuchiya, T. and Matsumoto, J.J. (1985) Recovery of water soluble proteins in waste wash water of fish processing plants by ultrafiltration, (in Japanese). Bull. Jap. Soc. Sci.. Fish. 51 (7): 1 133-1 138. Noda, M. and Murakami, K. (1981) Studies on proteinases from the digestive organs of sardines. 2. Purification and characterization of two acid proteinases from the stomach. Biochim. Biophys. Ada 658: 27-34. Noda, M., Vo Van, T., Kusakabe, I. and Murakami, K. (1982) Substrate specificity and salt inhibition of five proteinases isolated from the pyloric caeca and stomach of sardine. Agric. Biol. Chem. 46 (6): 1565-1569. Norris, E.R. and Elam, D.W. (1940) Preparation and properties of crystalline salmon pepsin. J. Biol. Chem. 134: 443-454. Ockerman, H.W. (1992) Fishery by-products. Hall, G.M. (ed.). Fish Processing Technology’. London: Blackie Academic & Professional; 155-192. Ockerman, H.W. and Hansen, CL (1988) Animal By-product Processing. Weinheim, Germany: VCH; 279-308. Olsson, N. (1942) Forsok rorande fiskeriprodukternas tillvaratagande och anvanbarheet som foder for hos och kycklingar. Lantbrukshogskolan Husdjur/drsoksanstalien (Sweden). Rep. No. 7. 55p. Cited from Raa and Gildberg (1982). Ooshiro, Z. (1968) Studies on proteinase in pyloric caeca of fishes: I. Isolation and purification of proteinase from pyloric caeca of mackerel (in Japanese). Bull. Jap. Soc. Sci. Fish. 34 (9): 847-852. Ooshiro, Z., Ok, T., Une, H., Hayashi, S. and Itakura, T. (1981) Study on the use of commercial proteolytic enzymes in the production of fish sauce. Mem. Fac. 207 Fish. Kagoshima [Jniv. 30: 383-394. Orejana, F.M. and Liston, J. (1981) Agents of proteolysis and its inhibition in patis (fish sauce) fermentation. ,J. FoodSci. 47: 198-203. Page, VV.J. (1992) Production of poly-(3-hydroxybutyrate by Azotobacter vinelandii UWD in media containing sugars and complex nitrogen sources. Appl. Microbiol. Biotechnol. 38: 117-121. Page, W.J. and Cornish, A. (1993) Growth of Azotobacter vinelandii UWD in fish peptones medium and simplified extraction of poly-(3-hydroxybutyrate. Appl. Environ. Microbiol 59 (12): 4236-4244. Pan, B.S. (1990a) Minced fish technology. Sikorsky, Z. (ed.). Seafood: Resources, Nutritional Composition and Preservation. Boca Raton: CRC Press, Inc.; 199-210. Pan, B.S. (1990b) Recovery of shrimp waste for flavouring. Voit, M.N. and Botta, J.R. (eds.). Advances in Fisheries Technology and Biotechnology for Increased Profitability’. Lancaster, PA: Technomic Publ. Co., Inc.; 437-447. Pedersen, L.D. (1990) Product recovery from surimi wash water. Keller, S. (ed.). Making Profits out of Seafood Wastes. Proceedings of the International Conference on Fish By-products, Anchorage, Alaska, 25-27 April, 1990. Alaska Sea Grant College Program, Report No 90-07; 173-176. Perlman, D.F. and Goldstein, L. (1988) Nitrogen metabolism. Shuttleworth, T.J. (ed ). Physiology’ of Elasmobranch Fishes. Berlin: Springer-Verlag; 253- 276. Peterson, G.L. (1979) Review of the Folin phenol protein quantitation method of Lowry, Rosebrough, Farr and Randall. Anal Biochem. 100: 201-219. 208 Peterson, H. (1953) Acid preservation of fish and fish offal. FAO Fish. Bull. 6(1): 18. Cited from Raa and Gildberg (1982). Pigott, G.M and Tucker, B.W. (1990) Seafood: The Effects of Technology on Nutrition. New York: Marcel Dekker, Inc. 362p. Pillai, S.A., Prabhu, P.V. and Balachandran, K.K. (1986) Design of a detachable type fish drier. Fishery Technol. 23:143-146. Poernomo, A. and Buckle, K.A. (1993) Utilization of fish viscera. Liang, O.B., Buchanan, A. and Fardiaz, D. (eds.). Development of Food Science and Technology in Southeast Asia. Bogor, Indonesia: IPB Press; 379-385. Poernomo, A., Giyatmi, Fawzya, Y.N. and Ariyani, F. (1992) Salting and drying of mackerel (Rastrelliger kanagurta). ASEAN Food J. 7(3): 141-146. Poernomo, A. and Utomo, B.S.B. (1990) Review of studies in salting and drying of sardines. Indo-Pacific Fishery Commission: Papers Presented at the Seventh Session of the Indo-Pacific Fishery Commission Working Party on Fish Technology’ and Marketing. Bangkok, Thailand, 19-22 April 1988. FAO Fish. Rep., (401) Suppl. Rome: FAO; 133-152. Pollack, MR. (1965) Purification and properties of penicillinases from two strains of Bacillus licheniformis. A chemical, physicochemical and physiological comparison. Biochem. J. 94: 666-675. Power, D A. and McCuen, P.J. (1988) Manual of BBLR Products and Laboratory Procedures. 6th ed. Cockeysville, MD.: Becton Dickinson Microbiology System; 293-295. Prasertsan, P., Jung, S. and Buckle, K.A. (1994) Anaerobic filter treatment of fishery wastewater. World J. Microbiol. Biotechnol. 10: 11-13. Pulley, K. (1974) Marine Fishes of Australian Waters. Melbourne: Lansdowne. 103 p. Putro, S. and Sutijana (1990). Study on isolation of squalene from acid ensilaged 209 shark liver oil. Reilly, P.J.A., Parry, R.W.H. and Barile, L.E. (eds.). Post harvest Technology, Preservation and Quality of Fish in South East Asia. Stockholm: Int. Found. Sci.; 183-187. Pyeun, J.H. and Kim, FI.R. (1986) The proteinases distributed in the intestinal organs of fish. 1. Purification of three alkaline proteinases from the pyloric caeca of mackerel, Scomber japonicns. Bull Korean Fish. Soc. 19 (6): 537-546. Pyeun, J.H., Kim, H R. and Heu, M.S. (1988) ) The proteinases distributed in the intestinal organs of fish. 3. Purification and some enzymatic properties of the alkaline proteinases from the pyloric caeca of skipjack, Katsnwomis vagans. Bull. Korean Fish. Soc. 21 (2): 85-96. Raa, J. and Gildberg, A. (1976) Autolysis and proteolytic activity of cod viscera. J. Food Technol. 11: 619-628. Raa, J. and Gildberg, A. (1982) Fish silage: A review. CRC Crit. Rev. Food Sci. Nutr. 16: 383-419. Raa, J., Gildberg, A. and Strom, T. (1983) Silage production - Theory and practice. Ledward, D.A., Taylor, A.J. and Lawrie, R.A. (eds.). Upgrading Waste for Feeds and Foods. London: Butterworths; 117-132. Ragelis, E.P. (1984) Ciguatera seafood poisoning: Overview. Ragelis, E.P. (ed.). Seafood Toxins. ACS Symposium Series 262. Washington: ACS; 25-36. Raghunath, M R. and McCurdy, A.R. (1990) Influence of pH on the proteinase complement and proteolytic products in rainbow trout viscera silage. J. Agiic. Food Chem. 38: 45-50. Raghunath, M R. and McCurdy, A.R. (1991) Synthesis of plasteins from fish silage. J. Sci. FoodAgric. 54: 655-658. Raksakulthai, N., Lee, Y.Z. and Haard, N.F. (1986) Effect of enzyme supplements 210 on the production of fish sauce from male capelin (Mallotns vi/losus). Can. Inst. FoodSci. Techno!. J. 19 (1): 28-33. Rao, S.V.R., Dwarakanath, C.T. and Saraswathi, C.R. (1980) Preparation and microbiological evaluation of bactopeptone from shrimp waste. J. Food Sci. Techno/. 17: 133-136. Rebeca, B.D , Pena-Vera, M.T. and Diaz-Castaneda, M. (1991) Production of fish protein hydrolysates with bacterial proteases: Yield and nutritional value. J. Food Sci. 56 (2): 309-314. Reece, P. (1988) Recovery of proteases from fish waste. Process Biochem. 23 (3): 62-66. Reeck, G.R. (1971) Phylogenetic variations among proteolytic enzymes: Studies on pancreatic enzymes of African hingfish and bovine carboxypeptidase B. Seattle: University of Washington. PhD thesis. Cited from Liu and Pigott (1981). Reish, D.J. (1969) Biology of the Oceans. Belmont, CA.: Dickenson Publishing Company, Inc. 236p. Retnowati, N., Fawzya, Y.N., Ariyani, F., Poernomo, A. and Nasran, S. (1991) Pengolahan ikan pari asin di Muara Angke: Struktur biaya, keuntungan pengolahan dan saluran pemasarannya. (Processing of dried salted rays at Muara Angke: Cost and profit analysis, and its marketing). J. Penelitian Pascapanen Peri kanan (J. Post harvest Fish. Res.) 70: 1-11. Rha, C.K. (1984) Chitosan as biomaterial. Colwell, R.R., Sinskey, A.J. and Pariser, E.R. (eds ). Biotechnology in the Marine Sciences. New York: John Wiley & Sons; 177-189. 211 Rick, W. (1963a) Trypsin. Bergmeyer, H.U. (ed). Methods of Enzymatic Analysis. New York: Academic Press; 807-818. Rick, W. (1963b) Pepsin, Pepsinogen, Uropepsinogen. Bergmeyer, H.U. (ed.). Methods of Enzymatic Analysis. New York: Academic Press; 819-823. Roy, G. (1992) Bitterness: reduction and inhibition. Trends Food Sci. Technol. 3 (4): 85-91. Rupley, J.A. (1967) Susceptibility to attack by proteolytic enzymes. Meth. Enzymol. 11: 905-917. Russell, et al. (1958) Studies on the mechanism of death from stingray venom: A report of two fatal cases. Am. J. Med. Sci. 235: 566-584. Cited from Barss (1984). Russell, F.E. (1965) Marine toxins and venomous and poisonous marine animals. Adv. Mar. Biol. 3:255-384. Saha, K.C. (1940) Effect of boiling and frying on the enzymatic hydrolysis of fish protein. J. Indian Chem. Soc. 17: 259-263. Cited from Adler-Nissen (1986). Sainsbury, K.J., Kailola, P.J. and Leyland, G.G. (1985) Continental Shelf Fishes of Northern and North Western Australia. An Illustrated Guide. Canberra: CSIRO Division of Fisheries Research. Saisithi, P. (1987) Traditional fermented fish products with special reference to Thai products. ASEAN Food J. 3 (1): 3-10. Sanchez-Chiang, L., Cisternas, E. and Ponce, E. (1987) Partial purification of pepsins from adult and juvenile salmon fish, Oncorhynus keta. Effect of NaCl on proteolytic activities. Comp. Biochem. Physiol. 87B: 793-797. Sandford, P.A. (1989) Chitosan: Commercial uses and potential applications. Skjak- 212 Braek, G., Anthonsen, T. and Sandford, P. (eds.). Chi tin and Chitosan: Sources, Chemistry, Biochemistry, Physical Properties and Applications. New York: Elsevier Applied Science; 51-69. Sasano, Y. and Ota, M. (1964) Studies on utilization of enzyme from whale pancreas: I. Proteolytic enzyme (in Japanese). Bull. Jap. Soc. Sci. fish. 30 (5): 424-429. Shahidi, F. (1994) Proteins from seafood processing discards. Sikorski, Z.E., Pan, B.S. and Shahidi, F. (eds ). Seafood Proteins. New York: Chapman & Hall; 171-193. Shahidi, F., Xiao-Qing, H. and Synowiecki, J. (1995) Production and characteristics of protein hydrolysates from capelin (Mallotus villosus). Food Chem. 53: 285-293. Shahidi, F. and Synowiecki, J. (1991) Isolation and characterization of nutrients and value-added products from snow crab (Chinoectes opilio) and shrimp (Pandalus borealis) processing discards. J. Agric. Food Chem. 39: 1527- 1532. Shahidi, F. and Synowiecki, J. (1992) Quality and compositional characteristics of Newfoundland shellfish processing discards. Brine, C.J., Sandford, P.A. and Zikakis, J.P. (eds ). Advances in Chitin and Chitosan. New York: Elsevier Applied Science; 617-626. Shamsuzzaman, K and Haard, N.F. (1983) Evaluation of harp seal protease as a rennet substitute for Cheddar cheese. J. Food Sci. 48 (1): 179-183. Shamsuzzaman, K and Haard, N.F. (1984) Purification and characterization of a chymosin-like enzyme from the gastric mucosa of harp seal, Pagophilus 213 groenlandicus. Can.Biochem. Cell Biol. 62 : 699-708. Shamsuzzaman, K and Haard, N.F. (1985) Milk clotting and cheese making properties of a chymosin-like enzyme from harp seal mucosa. J. Food Biochem. 9 173-192. Shewan, J.M. (1951) The chemistry and metabolism of nitrogenous extractives in fish. Williams, R.T. (ed.). The Biochemistry of Fish. Cambridge: Cambridge University Press; 28-48. Shewan, J.M. (1977) The bacteriology of fresh and spoiling fish and the biochemical changes induced by bacterial action. Proceedings of the Conference on the Handling, Processing and Marketing of Tropical Fish, London, 5-9 July 1976. London: Tropical Products Institute; 51-66. Shiau, C.Y. and Chai, T. (1990) Characterization of oyster shucking liquid wastes and their utilization as oyster soup. J. FoodSci. 55 (2): 374-378. Shieh, J.Y. (1989) Role of polyelectrolyte molecular weight in protein recovery by polyelectrolyte precipitation. MSc Thesis. Iowa State University, Ames, Iowa. 95p. Shieh, J.Y. and Glatz, C.E. (1991) Precipitation of proteins with polyelectrolytes: Roles of polymer molecular weight and modification of protein charge. Polymer Preprints 32 (1): 606-607. Shin, D.H. and Zall, R.R. (1986) Purification and identification of a trypsin-like enzyme from the pyloric caeca of cod. Process Biochem. 21(1): 11-15. Shoji, Y. (1990) Creamy fish protein. Keller, S. (ed.). Making Profits out of Seafood Wastes. Proceedings of the International Conference on Fish By-products, Anchorage, Alaska, 25-27 April, 1990. Alaska Sea Grant College Program, 214 Report No 90-07; 87-93. Shung, N.J., Kang, S.K., Lee, S.J. and Kim, S.H. (1994) The factors for the formation of carcinogenic N-nitrosamine from dried marine products, (in Korean). Bull. Kor. Fish. Soc. 27 (3): 247-250. Sidwell, V. (1981) Chemical and Nutritional Composition of Finfishes, Whales, Crustaceans, Mollusks and Their Products. NOAA Technical Memorandum, NMFS F/SEC-11, United States Dept. Commerce, Washington D.C. Cited from Pigott and Tucker 1990. Sikorski, Z.E. (1994) The contents of proteins and other nitrogenous compounds in marine animals. Sikorski, Z.E., Pan, B.S. and Shahidi, F. (eds.). Seafood Proteins. New York: Chapman & Hall; 6-12. Sikorski, Z.E. and Karnicki, Z. (1990) Resources and their ability. Sikorski, Z.E. (ed). Seafood: Resources, Nutritional Composition and Presentation. Boca Raton: CRC Press, Inc.; 9-28. Sikorski, Z.E., Kolakowska, A. and Pan, B.S. (1990) The nutritive composition of the major groups of marine food organisms. Sikorski, Z.E. (ed.). Seafood: Resources, Nutritional Composition and Presentation. Boca Raton: CRC Press, Inc.; 29-54. Sikorski, Z.E., Kolakowska, A. and Burt, J.R. (1990) Postharvest biochemical and microbial changes. Sikorski, Z.E. (ed ). Seafood: Resources, Nutritional Composition and Preservation. Boca Raton: CRC Press, Inc.; 55-75. Simopoulos, A.P., Kifer, R.R. and Martin, R.E. (1986) Health Effects of Polyunsaturated Fatty> Acids in Seafoods. London: Academic Press. 473p. Simpson, B .K. (1984) Isolation, Characterization and Some Applications of Trypsin 215 from Greenland Cod (Gadus morlma). PhD Thesis. St. John's, Newfoundland, Canada: Memorial University of New Foundland. 179p. Simpson, B.K., Gagne, N. and Simpson, M.V. (1994) Bioprocessing of chitin and chitosan. Fisheries Processing: Biotechnological Applications. London: Chapman & Hall; 155-173. Simpson, B.K. and Haard, N.F. (1984a) Trypsin from Greenland cod as a food processing aid. ./. Appl. Biochem. 6: 135-143. Simpson, B.K. and Haard, N.F. (1984b) Purification and characterization of trypsin from the Greenland cod (Gadus ogac). 1. Kinetic and thermodynamic characteristics. Can. J. Biochem. Cell Biol. 62: 894-900. Simpson, B.K. and Haard, N.F. (1984c) Trypsin from Greenland cod, Gadus ogac. Isolation and comparative properties. Comp. Biochem. Physiol. 79B(4): 613- 622. Simpson, B.K. and Haard, N.F. (1985) The use of proteolytic enzymes to extract carotenoproteins from shrimp waste. J. Appl. Biochem. 7: 212-222. Simpson, B.K., Simpson, M.V. and Haard, N.F. (1990) Properties of trypsin from the pyloric ceca of Atlantic cod (Gadus morhua). J. Food Sci. 55 (4): 959- 961 Simpson, B.K., Smith, J.P., Yaylayan, V. and Haard, N.F. (1989) Kinetic and thermodynamic characteristics of a digestive protease from Atlantic cod, Gadus morhua. J. Food Biochem. 13: 201-213. Simpson, K.L. (1978) The recovery of protein and pigments from shrimp and crab meals and their use in salmonid pigmentation. Muzzarelli, R.A.A. and Pariser, R.E. (eds.). Proceedings of the First International Conference on Chitin Chitosan. MIT Sea Grant Report 78-7, Cambridge, MA; 253-262. Cited from Hansen and Illanes (1994). 216 Sinnel, H.J. (1980) Interacting factors affecting mixed populations. Siliker, J.H. (ed.). Microbial Ecology of Foods. Vol 1. London: Academic Press; 251- 281. Skaugrud, O. and Sargent, G. (1990) Chitin and chitosan: Crustacean biopolymers with potential. Keller, S. (ed ). Making Profits out of Seafood Wastes. Proceedings of the International Conference on Fish By-products, Anchorage, Alaska, 25-27- April, 1990. Alaska Sea Grant College Program, Report No 90-07; 61-69. Skorupa, K. and Sikorski, Z.E. (1992) Utilization of fishery wastes for the production of peptones for microbiological purposes (in Polish). Medycyna Weterynaryjna 48 (6): 279-281. Skorupa, K. and Sikorski, Z.E. (1993) Potential for utilisation of low-grade fish material for production of microbiological peptones (in Polish). Przemysil- Spozywczy 47 (6): 159-161. Snieszko, S.F., Griffin, P.J. and Friddle, S B. (1950) A new bacterium {Hemophilus piscium N. sp) from ulcer disease of trout. J. Bacteriol. 59: 699-710. Sprissler, G.P. (1942) An Investigation of the Proteinases of the Gastric Mucosa of Shark. The Catholic University of America. PhD thesis. 52p. Sophanodora, P. and Buckle, K.A. (1988) Production of microbial peptone from prawn processing waste. Maneepun, S., Varangoon, P. and Phithakpol, B. (eds ). Food Science and Technology in Industrial Development. Proceedings of the Food Conference '88. Bangkok, Thailand, 24-26 October 1988; 744-750. Sorensen, S.P.L. (1908) Enzymstudien: I. Uber die quantitative Messung 217 proteolytischer Spaltungen. ‘Die Formoltitrierung” Biochem. Z 7: 45-101. Cited from Adler-Nissen (1986). Souness, R (1990) Development of a rice husk-fired furnace and its application to fish drying. Indo-Pacific Fishery Commission. Papers Presented at the Seventh Session of the Indo-Pacific Fishery Commission Working Party on Fish Technology’ and Marketing. Bangkok, Thailand, 19-22 April 1988. FAO Fish. Rep., (401) Snppl. Rome: FAO; 160-167. Souness, R. and Wibowo, S. (1995) Development of a rice-husk fired furnace and drying chamber for fish drying in Indonesia. Champ, B.R. and Highley, E. (eds.). Fish Drying in Indonesia. Proceedings of an International Workshop held at Jakarta, Indonesia, 9-10 February 1994. ACIAR Proceedings No. 59. Canberra: Australian Centre for International Agricultural Research; 76-82. Squires, E.J., Haard, N.F. and Feltham, L.. A. W. (1986a) Gastric proteases of the Greenland cod Gadits ogac. 1. Isolation and kinetic properties. Biochem. Cell Biol. 64: 205-214 Squires, E.J., Haard, N.F. and Feltham, L.. A. W. (1986b) Gastric proteases of the Greenland cod Gadus ogac. 2. Structural properties. Biochem. Cell Biol. 64: 215-222 . Stansby, M E. (1963) Industrial Fishery Technology. 2nd ed. New York: R.E. Krieger Pub. Co. 415p. Stansby, M.E. (1982) Properties of fish oils and their application to handling of fish and to nutritional and industrial use. Martin, R.E., Flick, G.J., Hebard, C.E. and Ward, D.R. (eds ). Chemistry and Biochemistry of Marine Food Products. Westport, Conn.: AVI Publ. Co.; 75-92. 218 Stansby, M.E. (1990a) Nutritional properties of fish oil for human consumption - Early developments. Stansby, M.E. (ed.). Fish Oils in Nutrition. New York: Van Nostrand Reinhold; 268-288. Stansby, M.E. (1990b) Nutritional properties of fish oil for human consumption - Modern aspects. Stansby, M.E. (ed ). Fish Oils in Nutrition. New York: Van Nostrand Reinhold; 289-313. Stansby, M.E., Schlenk, H. and Gruger, Jr., E.H. (1990) Fatty acid composition of fish. Stansby, M.E. (ed ). Fish Oils in Nutrition. New York: Van Nostrand Reinhold; 6-39. Stanton, W.R. and Yeoh, Q.L. (1977) Low salt fermentation method for conserving trash fish waste under SE Asian condition. Proceedings of the Conference on the Handling, Processing and Marketing of Tropica! Fish, London, 5-9 July 1976. London. Tropical Products Institute; 277-282. Stead, D.G. (1963) Sharks and Rays in Australian Seas. Sydney: Angus and Robertson. 21 lp. Stefansson, G. (1988) Enzymes in the food industry. Food Technol. 42 (3) 64-65. Stefansson, G. and Steingrimsdottir, U. (1990) Application of enzymes for fish processing in Iceland: Present and future aspects. Voit, M.N. and Botta, J.R. (eds ). Advances in Fisheries Technology and Biotechnology for Increased Profitability. Lancaster, PA: Technomic Publ. Co., Inc.; 237-250. Stephens, N.L., Bough, W.A., Beuchat, L.R. and Heaton, E.K. (1976) Preparation and evaluation of two microbiological media from shrimp heads and hulls. Appl. Environ. Microbiol. 31: 1-6. Sternberg, M. (1972) Purification of amyloglucosidases. Chem.Abstr.il. 112490f. 219 Sternberg, M. (1975) Removal of transglucosidase from amyloglucosidase. US Patent No. 3,917,513. Sternberg, M. (1976) Purification of industrial enzymes with polyacrylic acids Process Biochem. Sept. : 11-12 Stenberg, M. and Hershberger, D. (1973) Separation of proteins with polyacrylic acids. Biochim. Biophys. Acta 342: 195-206. Stirling, W. (1884) On the ferments of enzymes of the digestive tract in fishes. J. Anat. Physiol 18: 426-435. Cited from Gildberg (1988). Stoknes, I., Rustad, T. and Mohr, V. (1993) Comparative studies of the proteolytic activity of tissue extracts from cod (Gadus morhua) and herring (Clupea harengus). Comp. Biochem. Physiol 106B (3): 613-619. Storey , R.M. (1982) Smoking. Aitken, A., Mackie, I.M., Merrit, J.H. and Windsor, M.L. (eds.). Fish Handling and Processing. 2nd ed. Aberdeen: Min. of Agriculture, Fisheries & Food, Torry Research Station; 98-114. Strasdine, G.A. and Melville, J.M. (1972). Salmon-canning waste water as a microbiological growth medium. J. Fish. Res. Bd Can. 29: 1769-1771. Strom, T. and Eggum, B.O. (1981) Nutritional value of fish viscera silage. J. Sci. Food Agiic. 32: 115-120. Strom, T. and Raa, J. (1992) Ensiling of fish by lactic acid bacterial fermentation for feed production. Indo-Pacific Fishery Commission: Papers Presented at the Eighth Session of the Indo-Pacific Fishery Commission Working Party on Fish Technology> and Marketing, Yogyakarta, Indonesia, 24-27 September 1991. FAO Fish. Rep., (470) Snppl Rome: FAO; 142-147. Strom, T. and Raa, J. (1993) Marine biotechnology in Norway. J. Mar. Biotechnol. 1(1): 3-7. 220 Strom, T., Gildberg, A., Stormo, B. and Raa, J. (1980) Fish silage: Why not use propionic and formic acid? Connel, J.J. (ed). Advances in Fish Science and Technology. Farham, Surrey, England: Fishing News Books Ltd.; 352-355. Strommen, I. (1983) New equipment in fish drying. Scand. Refr. 4:208-209. Sukarsa, D. R (1978a) Pemanfaatan jeroan ikan sebagai hasil ikutan dari pengawetan ikan. (The utilisation of fish viscera as by product from fish processing). J. Penelitian Teknologi Perikanan. (J. Fish. Technol. Res.) 1: 11-52. Sukarsa, D. (1978b) Pemanfaatan pyloric caeca ikan sebagai penghasil enzim proteolitik (Utilisation of pyloric caeca as proteolytic enzyme source). J. Penelitian Hasil Perikanan. 1: 14. Cited from Sulistiyani and Heruwati (1991). Sulistiyani and Heruwati, E.S. (1992) Identification and purification of proteolytic enzymes from fish pyloric caeca. Indo-Pacific Fishery Commission: Papers Presented at the Eighth Session of the Indo-Pacific Fishery Commission Working Party on Fish Technology’ and Marketing, Yogyakarta, Indonesia, 24-27 September 1991. FAQ Fish. Rep., (470) Suppl. Rome: FAO; 320-325. Sumner, J. and Magno-Orejana, F. (1985) Do tropical fish keep longer in ice than temperate fish: The circumstantial and definitive approaches. Reilly, A. (ed.). Spoilage of Tropical Fish and Product Development. Proceedings of a Symposium held in conjunction with the Sixth Session of the Indo-Pacific Fishery Commission Working Party on Fish Technology and Marketing. Royal Melbourne Institute of Technology, Melbourne,- Australia, 23-26 October 1984. FAO Fish. Rep., (317) Suppl. Rome: FAO; 62-70. Suparno (1995) Fish salting and drying studies in Indonesia. Champ, B.R. and Highley, E. (eds ). Fish Drying in Indonesia. Proceedings of an International 221 Workshop held at Jakarta, Indonesia, 9-10 February 1994. ACIAR Proceedings No. 59. Canberra: Australian Centre for International Agricultural Research; 60-66. Suparno and Poernomo, A. (1991) Fish waste utilisation in Indonesia. ASEAN Food J. 7 (2): 67-72. Suzuki, T. (1981) Fish and Krill Protein. London: Applied Science Publishers Ltd. 260p. Suzuki, T. (1984) Increasing the use of sardine and mackerel for human consumption. Infofish Mark. Dig. Feb. (1): 34-36. Takahashi, K. (1994) Aspartic Proteinases: Structure, Function, Biology and Biomedical Implications. New York; Plennum Press. 629p. Tambunan, P.R., Tazwir and Sabaruddin (1992) Studi penurunan mutu kulit ikan pari (Trygon sephen) pada suhu kamar dan es. (Study on the quality deterioration of rays (Trygon sephen) skin at ambient temperature and chilled temperatures). J. Penelitian Pascapanen Perikanan. {J. Postharvest Fish. Res.) 73: 25-36. Taniuchi, T. (1979) Freshwater elasmobranchs from Lake Naujan, Perak River and Indragiri River, Southeast Asia. Jap. J. Ichthyol. 25 (4): 273-277. Taniuchi, T. (1991) Occurrence of two species of stingrays of the genus Dasyatis (Chondrichtyes) in the Sanaga Basin, Cameroun. Environ. Biol. Fishes. 31 (1): 95-100. Tanikawa, E., Motohiro, T. and Akiba, M. (1985) Marine Products in Japan. Rev. ed. Tokyo: Koseisha Koseikaku. 506p. Tarr, H.L.A. and Deas, C.P. (1949) Bacteriological peptones from fish flesh. J. Fish. 222 Res. Bd Can. 7: 552-560. Tatterson, IN. (1982) Fish silage - Preparation, properties and uses. Anim. Feed Sci Techno/. 7: 153-159. Tatterson, I N. and Windsor, M L. (1974) Fish silage. J. Sci. Food Agihc. 25: 369- 379. Thongthai, C., Panbangred, W , Khoprasert, C. and Dhaveetiyanond (1990) Protease activities in the traditional process of fish sauce fermentation. Reilly, P.J.A., Parry, R.W.H. and Barile, L.E. (eds.). Post-harvest Technology, Presentation and Quality of Fish in Southeast Asia. Proceedings of the Workshop on Post-harvest Technology, Preservation and Quality of Fish in Southeast Asia. Bangkok, Thailand, November 13-17, 1989. Stockholm: International Foundation for Science; 61-65. Torrissen, O.J., Hardy, R.W. and Shearer, K.D. (1989) Pigmentation of Salmonoid: Carotenoid deposition and metabolism. Rev. Aquatic Sci. 1: 209- 225. Torrissen, O.J, Tidemann, E., Hansen, F. and Raa, J. (1981/1982) Ensiling in acid - A method to stabilize astaxanthin in shrimp processing by-products and improve uptake of this pigment by rainbow trout (Salmo gaidneri). Aquaculture 26: 77-83. Traxler, R.W. and Lankford, C.E. (1957) Observation of cystine degradation and bacteriotoxicity of peptones. Appl. Microbiol. 5: 70-74. Tsao, C.Y. and Nagayama, F. (1991) Purification and characterization of proteases from oyster (Crassotrea gigas) J. FoodBiochem 15: 81-96. Uchida, Y., Hukuhara, H., Shirakawa, Y. and Shoji, Y. (1990) Bio-fish flour. Keller, S. (ed ). Making Profits out of Seafood Wastes. Proceedings of the International Conference on Fish By-products, Anchorage, Alaska, 25-27 223 April, 1990. Alaska Sea Grant College Program, Report No 90-07; 95-99. Van Veen, A.G. (1965) Fermented and dried seafood products in Southeast Asia. Borgstrom, G. (ed.). Fish as Food. Vol. Ill: Processing: Part 1. New York: Academic Press; 227-250. Vecht-Lifshitz, S.E., Almas, K.A. and Zomer, E. (1990) Microbial growth on peptones from fish industrial wastes. Left. Appl. Microbiol 10: 183-186. Venugopal, V. (1994) Production of fish protein hydrolysates by microorganisms. Martin, A M. (ed ). Fisheries Processing: Biotechnological Applications. London: Chapman & Hall; 223-243. Venugopal, V. and Shahidi, F. (1995) Value-added products from underutilized fish species. Grit. Rev. FoodSci. Nutr. 35 (5): 431-453. Verburg, G.B. and Freeman, D.W. (1984) On-farm production of liquefied catfish protein. Aquaculture 41: 171-179. Vlieg, P. and Murray, T. (1988) Proximate composition of albacore tuna, Thunnus alalunga, from the temperate South Pacific and Tasman Sea. N. Z. J. Mar. Freshwater Res. 22: 491-496. Voskresenky, N.A. (1965) Salting of herring. Borgstrom, G. (ed.). Fish as Food. Vol. II: Processing: Part l. New York: Academic Press; 107-132. Waterman, J.J. (1976) The production of dried fish. FAO Fish. Tech. Pap. 16. 52p. Weitzman, S.H. (1994) Classifying of fishes. Paxton, J.R. and Eschmeyer, W.N. (eds.). Encyclopedia of Fish Sydney: UNSW Press; 20-26. Wekell, J.C., Cauglitz, Jr., C.J., Barnett, H.J., Hatfield, C.L. and Eklund, M. (1994) The occurrence of domoic acid in razor clam (Siliqua patula), dungeness crab (Cancer magister) and anchovies (Engrau/is mordax). J. Shellfish Res. 13 224 (2): 587-593. Wheaton, F.W. and Lawson, T B (1985) Processing Aquatic Food Products. New York: John Wiley & Sons. 518p. Whittenbury, R. (1986) Microbiology of grass silage. Process Biochem. 3: 27-31 Wibowo, S., Sari, A., Suherman, M., Poernomo, A. and Souness, R.A. (1993) The development of a fish dryer using rice husk furnace as heat source. Liang, O.B., Buchanan, A. and Fardiaz, D. (eds.). Development of Food Science and Technology>in Southeast Asia. Bogor, Indonesia: IPB Press; 184-188. Windsor, M L. (1982) By-products. Aitken, A., Mackie, I.M., Merrit, J.FL and Windsor, M.L. (eds.). Fish Hatidling and Processing. 2nd ed. Aberdeen: Min. of Agriculture, Fisheries & Food, Torry Research Station; 152-160. Windsor, M. and Barlow, S. (1981) Introduction to Fishery By-products. Surrey: Fishing News Books Ltd. 187p. Williams, D.C. (1975) The occurrence and distribution of digestive enzymes in the pyloric caecum of the purple starfish Pisaster ochraceus. Comp. Biochem. Physiol. 52A: 85-90. Wilson, R.P., Freeman, D.W. and Poe, W.E. (1984) Three types of catfish offal meals for channel catfish fingerlings. Prog. Fish. Cult. 46 (2): 126-132. Wirahadikusumah, S. (1968) Preventing Clostridium botulinum type E poisoning and rancidity by silage fermentation. Lantbrukshogskolans Annaler. 34: 551-689. Wootton, R.J (1992) Fish Ecology. London: Blackie. 212p. Wyat, B. and McGourty, G. (1990) Use of marine by-products on agricultural crops. Keller, S. (ed.). ATaking Profits out of Seafood Wastes. Proceedings of the International Conference on Fish By-products, Anchorage, Alaska, 25-27 225 April, 1990. Alaska Sea Grant College Program, Report No 90-07; 187-195. Xu, R.A., Wong, R.J., Rogers, M L. and Fletcher, G.C. (1996) Purification and characterisation of acidic proteases from the stomach of the deepwater finfish orange roughy (Hop/ostethus at/anticus). J. FoodBiochem. 20: 31-48. Yasumoto, T., Nakajima, I. and Chungue, E. (1977) Toxins in the gut contents of parrotfish. (in Japanese). Bull. Jap. Soc. Sci. Fish. 43 (1): 69-74. Yoshinaka, R., Sato, M., Tsuchiya, N. and Ikeda, S. (1983) Production offish sauce from sardine by utilization of its visceral enzymes (in Japanese). Bull. Jap. Soc. Sci. Fish. 49 (3): 463-469. Zotos, A. and Taylor, K.D.A. (1996) Partial purification and characterization of proteases from Norway lobster (Nephrops norvegicus) and their role in the phenolase activation process. Food Chem. 56 (1): 61-68. Zuberi, R., Fatima, R., Shamshad, S.I. and Qadri, R.B. (1992) Preparation of fish silage by microbial fermentation. Indo-Pacific Fishery Commission: Papers Presented at the Eighth Session of the Indo-Pacific Fishery Commission Working Party on Fish Technology’ and Marketing, Yogyakarta, Indonesia, 24-27 September 1991. FAO Fish. Rep. (470) Suppl. Rome: FAO; 155-164. 226 8 APPENDICES 8.1 Summary of results of statistical analysis for silages from Muara Angke and Labuhan Maringgai Source of p-level variation pH Soluble nitrogen Solubilisation Viscosity 1 0.0000 *** 0.0000 * * * 0.0000 *** 0.0000 *** 2 0.0000 *** 0.0632 0.0516 0.1165 0.6298 0.0051 *** 0.0101 *** 0.0284 *** 4 0.0051 *** 0.6656 0.2011 0.6630 1x2 0.1946 0.1200 0.0000 *** 0.0000 *** 1x3 0.6493 0.0000 * * * 0.0000 *** 0.0000 *** 2x3 0.8911 0.0363 * * * 0.0723 0.1167 1x4 0.0001 *** 0.0001 * * * 0.0000 *** 0.0606 2x4 0.1989 0.0497 ** 0.5660 0.0825 3x4 0.6945 0.0567 0.4733 0.5832 1x2x3 0.9951 0.9936 0.0000 *** 0.0000 *** 1x2x4 0.8146 0.2979 0.0000 *** 0.4017 1x3x4 0.1764 0.2734 0.0000 *** 0.0696 2x3x4 0.5697 0.7239 0.4051 0.0802 Ix2x3x4 0.1759 0.9987 0.0001 *** 0.4397 1 = Time (0. 5. 15. 30. 60 and 120 days) 2 = Acid (propionic + formic. 3%. v/w) 3 = Temperature (ambient / 28° - 32°C and 40° C) 4 = Origin of covvtail ray (Muara Angke and Labuhan Maringgai) ** = Significant (p<0.05) *** = Highly significant (p<0.01) 227 oo V(- £ < O O £ « 3 s- a s bX) IS JS ts T3 "O 15 ro cn T3 -C _o m OO m Cv C(_ .C £ o op £ Z3 CD C/3 I TO C c/3 c TO 03 C/f TO o c/> f summarised in Appendix 8.1, have been performed -O JD H X 'Is - o3 CL O 00 03 Time ' < 12 •7 < m < -r-1 "O (N < 4 12 < < 12 < 22 CM < T3 4 V_> 0 r 0 O o o o O — ■ 1 days t O id m O < m O 4 rc 0 O 0 m 4 un (N 4 O 0 45 id O , rd m 545 45 — 0 0 4 c- ICV 45 54 45 45 45 45 0 _ 45 45 _ -H 1 , CM 4 00 4 >~r, O ' ro O O O — 4 r- O O id » — — r rc 545 45 — 4 00 O O CM 45 4 cn O 545 45 — 45 45 45 _ » ; 1 ( l/~3 id ' VO O * O 4 rd m vd O _ 4 VO O O 00 VO 0 0 — id rc 45 0 O O O 45 ■4 45 45 — 45 1 1 00 4 Ov 00 o' 00 ro O 4 vn o o' cm rd o' 0 4 - 4 00 o' 0 4 O ■H 45 ■H * - h ’ 15 0 O -cv -H rM 0 o' i | | 21 | 1 _ CO CO 00 vd 10 o' 4 45 c £ £ 52 £ 23 52 £ CO CO OC CO CO CO did id m o' CM o' CM CM 00 4 id r- o' o' 4 45 45 45 45 30 , 09 CO C0 CN 'o' o' O Ov o 0 0 co 4 o' id o' 00 CN O O 4 4 545 45 545 45 545 45 45 ’ CO CO 10 CM 0 o' 0 CO 4 r- r- 0 00 o id rc o' 45 120 ’ I T3 1 0 I O £ •o ~a V® < -t ? § < E m IT o to PA) 4 Formic acid (FA) 3% (v/w); Acid2 = PA + FA 4°/ £ 228 Table 8.2.2. NPN (% ol total N) ot cowtail ray viscera silage (M uara Angke) o o H < 15 s CN OO CO CN G CJ " O O o NO CO o' NO NO NO CO NO NO o' CO NO -H NO -H NO CO o' NO NO NO CO o' NO NO NO NO co o' NO 4t NO co o' -H NO -H 4t NO NO CO o' NO NO NO -H -H ’ NO r~- o o NO r- ' CN o' o' TT NO oo o -H CO -H NO o' O rr OO 4t o' cn CN CO CN r- r- N" o o' r- : oo o' H-H -H 4t -H CN P r- -H NO o "'3 CN 4 -H ; ’ NO NO r-' CO NO NO NO CN o' - 00 00 4t 4H NO o CO O CN 'o' o' — ^r 00 CN i CO o' r- 00 -H -H — t~-~ o i , -H ^r — ’ 1 « < ’ ’ r- NO r- NO o 1 — r- CO NO -H — -H OO OO 00 ONO O NO O CO — r-- 0o 00 oo 00 -H C" CO CO p n n « ’ 30 ' 09 r-~ CN r- NO, co r~" NO r- OO < CN -H 000 00 O O oo C o — 00 , CN 4t -H — 0oo 00 < n n ’ ’ i ’ ' r CN o r- oo 00 NO CN o r- o' oo 00 -H -H r- CO NO C NO O CN or or co ~ 4t 4t “ n 1 i s* cz 0 c J3 H _CL> 0-. .ts ’ rz: 00 27 p co ^ O CO NO CN o OO O cn ■^r CO 00 00 ■^r ^r 'O co r- co ON r NO 4 O CN o r- NO O -H CN NO O -H CN -H -H CN p CN co oo CN ■^T (N CO 44 -H -1 t ’ n NO O t-" r- OO oo o NO O o co ^r ro NO o' CN oo o -H co r- CO ON r- cn -H P -H 4 o' CO -H (N) oo n o CN -H n n t NO NO r- O r- no CN O NO CN or r- CN NO o ^r -H co NO r- NO o' r-- -H r -H o' CO -H p CN — n n 30 608.4 ± 58.9 532.4 ±81.5 S S 2.5±0.4 2.6±0.1 S 3.6±0.6 60 323.4 ± 9.4 274.6±11.2 S S 2.8±0.1 3.2±0.4 S S 120 176.9 ± 0.2 160.3 ± 0.1 S S 3.2±0.1 3.1±0.3 S S 229 co * CN CN r- Table 8.2.4. Liquefaction (%) of cowtail ray viscera silage (Muara Angke) * o H CJ ’ ' 1 NO CN of ON NO - CO OO 00 NO oo CN o 1,0 — 00 -H + +1 NO oo ri of — of ON O co of of +i O oi -H — — co no co Of 1 -H 1 co o' ON CN — n -H — NO CN ^ HH -H GO -H CN O O') CN CN tl o' ON co ^ m O NO O of HH -H n n ' O CO o — no oo' O CO 73 GO no O oo of -H -H GO 73 o' ON ON OO co O ri O — -H -H ' n n NO o O' £T> ^ -H O CN no r-' cn of' of 73 no CO NO GO on -H O n ON -H — co -H GO (/) n cn r no -H NO o CN CN of' wo GO 73 ^ cn -H O ON NO of o' OO GO O -H of (/) ' of -H 3 n " CO * *o 2 c c in 230 8.3 Raw data of cowtail ray viscera silage from Labuhan Maringgai The following tables contain original data from which Figures 3.6, 3.8 and 3.9 have been drawn; and a statistical analysis, of which the results are summarised in Appendix 1, has been performed. Table 8.3.1. pH of cowtail ray viscera silage (Labuhan Maringgai)* o Time Ambient (28 - 32° C) O C (days) Acidl Acid2 Acidl Acid2 0 5.1 ±0.0 4.8 ±0.2 5.1 ±0.0 4.7 ±0.0 5 5.1 ±0.0 4.8 ±0.0 5.1 ±0.1 4.7 ±0.0 15 5.2 ±0.1 4.8 ±0.0 5.2 ±0.0 4.8 ±0.0 30 5.2 ±0.1 4.8 ±0.0 5.3 ± 0.1 4.8 ±0.0 60 5.5 ±0.2 5.0 ± 0.0 5.4 ±0.1 5.1 ±0.0 120 5.5 ±0.1 5.6 ±0.6 5.5 ±0.1 5.1 ±0.0 * Values are means ± standard deviations from 3 determinations of 2 replications. Acidl = Propionic acid (PA) + Formic acid (FA) (3% v/\v); Acid2 = PA + FA (4% v/w). Table 8.3.2. NPN (% of total N) of cowtail ray viscera silage (Labuhan Maringgai)* o Time Ambient (28 - 32° C) O C (days) Acidl Acid2 Acidl Acid2 0 24.6 ±0.9 24.6 ±0.9 24.6 ±0.9 24.6 ±0.9 5 75.1 ±2.7 76.6 ±2.2 87.0 ±2.1 92.1± 3.8 15 78.0 ±2.3 78.6± 1.5 88.5 ±2.7 94.7 ±4.4 30 78.4 ±2.4 79.1 ± 1.8 90.5 ±5.0 94.5 ±2.6 60 78.6 ±2.8 79.2 ±3.2 90.0 ±3.3 93.4 ± 3.1 120 78.5 ±2.6 79.9 ±2.7 90.5 ±3.0 95.2 ±4.2 * See Table 8.3.1 for explanation of notes. 231 Table 8.3.3. Viscosity of cowtail ray viscera silage (Labuhan Maringgai)* o Time Ambient (28 - 32° C) O C (days) Acidl Acid2 Acidl Acid2 0 2100.0± 7.1 1482.5 ± 74.3 7.4 ± 1.3 16.2 ±0.5 5 889.5 ±85.6 752.0 ± 18.4 6.4 ± 0.9 4.3 ±0.4 15 832.5 ±71.4 560.5 ±61.5 4.8 ±0.4 3.1 ±0.1 30 519.0 ± 8.5 428.0 ± 14.1 4.0 ±0.8 2.9 ±0.8 60 339.5 ±20.5 360.0 ±21.2 3.6 ± 0.6 3.0 ± 0.3 120 160.0 ±49.5 139.0 ± 15.6 3.5 ± 0.3 3.0 ±0.1 * See Table 8.3.1 for explanation of notes. Table 8.3.4. Liquefaction (%) of cowtail ray viscera silage (Labuhan Maringgai) o Time Ambient (28 - 32° C) O C (days) Acidl Acid2 Acidl Acid2 0 15.8 ± 0.6 15.8 ± 0.6 15.8 ±0.6 15.8 ± 0.6 5 36.8± 1.9 39.0 ±2.9 87.4 ±2.5 87.7± 1.5 15 46.3 ±0.7 50.4 ±0.1 90.7 ±0.3 91.5 ± 0.4 30 44.8 ±0.9 53.1 ± 1.1 92.2 ±0.3 93.1 ±0.3 60 68.4 ±0.9 64.7 ± 1.2 94.4 ±0.4 95.0 ±0.3 120 73.8 ±0.8 82.2 ±4.7 95.3 ±0.2 95.1 ±0.3 * See Table 8.3.1 for explanation of notes. 8.4 Raw data of fish protein hydrolysates The following tables contain original data from which Figure 3.7 has been drawn. Table 8.4.1. Fish protein hydrolysis (% soluble. N/total. N) by crude enzyme from cowtail ray viscera at 45UC Time pH 2 pH 5 pH 8 (h) 5% 10% 5% 10% 5% 10% 1 7.5 ± 1.1* 5.1 ± 1.2 10.1 ±0.9 13.9 ± 1.4 3.0 ± 1.8 3.0 ±0.6 2 9.2 ±0.1 9.8 ±0.4 11.9 ±0.5 19.4 ±3.1 9.0 ±0.4 9.1 ±0.0 3 12.6 ± 1.1 18.2 ± 1.2 20.8 ±0.4 22.7 ±2.4 15.4 ± 1.2 16.3 ±0.3 4 11.4 ± 1.6 24.3 ± 1.6 25.5 ± 1.9 30.9 ±4.2 16.0 ± 3.7 23.2 ± 1.7 * Values are means ± standard deviations for 3 determinations of 2 replications. Table 8.4.2. Fish protein hydrolysis (% soluble. N/total. N) by crude < enzyme from cowtail ray viscera at 60°C Time pH 2 pH 5 pH 8 (h) 5% 10% 5% 10% 5% 10% 1 5.3 ±0.8* 5.8 ±2.2 9.9 ±0.6 8.1 ± 1.8 10.3 ±1.2 11.0 ±1.1 2 6.2 ± 1.2 6.1 ± 1.0 8.5 ±1.3 9.7 ± 1.9 13.8 ±2.5 15.7 ± 0.8 3 36.6 ±2.0 49.6 ±0.6 55.3 ±3.1 59.6 ± 1.7 53.8 ± 3.0 53.5 ±0.7 4 39.1 ±2.6 50.5 ±2.6 60.5 ±3.9 59.8 ±2.5 55.0 ± 0.8 57.1 ± 1.0 * Values are means ± standard deviations for 3 determinations of 2 replications. 233 8.5 Summary of results of statistical analysis for effects of peptones on growth rate, total growth and biomass production of test microorganisms Test p-level microorganisms Growth rate Total growth Biomass production Isolated from: * * * Egg 0.4397 0.0000 *** 0.0000 Milk 0.0851 0.0000 *** 0.0000 * * * Beef 0.0109 *** 0.0000 *** 0.0000 * * * Pure culture: E. co/i 0.0001 *** 0.0000 *** 0.0000 * * * Staph, aureus 0.0000 *** 0.0000 *** 0.0000 * * * B. suhtilis 0.0011 *** 0.0000 *** 0.0000 * * * S. cerevisiae 0.4228 0.0000 *** 0.0000 *** *** = Highly significant (p<0.01) 234 8.6 Raw data of use of cowtail ray viscera peptones as microbial growth media The following tables contain original data from which Figures 5.3 and 5.4 and Tables 4.7, 4.8 and 4.9 have been constructed and statistical analyses, of which the results are summarised in Appendix 8.5, have been performed. Table 8.6.1. Growth (absorbance at 600 nm) of egg microorganisms Time Absorbance (600 nm) (h) Difco Oxoid BBL FSHCL FSPF 0 0.000 ±0.000* 0.000 ± 0.000 0.000 ± 0.000 0.000 ±0.000 0.000 ±0.000 1 0.002 ±0.001 0.002 ±0.000 0.006 ± 0.002 0.004 ±0.001 0.006 ±0.003 3 0.004 ±0.001 0.004 ± 0.004 0.009 ± 0.000 0.010 ±0.004 0.010 ±0.001 6 0.006 ±0.003 0.005 ±0.005 0.012 ±0.002 0.016 ±0.004 0.013 ±0.004 9 0.005 ±0.002 0.002 ±0.002 0.013 ±0.004 0.016 ±0.003 0.016 ±0.003 12 0.010 ±0.006 0.091 ±0.017 0.160 ± 0.011 0.015 ±0.006 0.124 ±0.010 15 0.335 ±0.036 0.444 ± 0.022 0.576 ±0.040 0.457 ±0.014 0.827 ±0.026 24 0.552 ±0.074 0.702 ±0.050 0.797 ±0.028 0.799 ±0.044 1.052 ±0.009 36 0.588 ±0.098 0.750 ±0.062 0.836 ±0.035 0.908 ±0.098 1.175 ±0.045 * Values are means ± standard deviations for 3 replicates Table 8.6.2. Growth (absorbance at 600 nm) of milk microorganisms Time Absorbance (600 nm) (h) Difco Oxoid BBL FSHCL FSPF 0 0.000 ± 0.000* 0.000 ± 0.000 0.000 ±0.000 0.000 ±0.000 0.000 ±0.000 1 0.005 ± 0.004 0.017 ±0.003 0.008 ±0.002 0.002 ±0.002 0.011 ±0.001 3 0.006 ±0.002 0.010 ±0.003 0.008 ±0.002 0.011 ±0.002 0.018 ±0.004 6 0.110 ±0.005 0.107 ±0.007 0.083 ±0.006 0.039 ±0.017 0.329 ±0.007 9 0.310 ± 0.016 0.284 ±0.026 0.324 ±0.005 0.519 ± 0.013 0.659 ±0.042 12 0.522 ±0.020 0.386 ±0.036 0.587 ±0.005 0.636 ±0.058 0.825 ±0.048 15 0.612 ±0.004 0.509 ±0.051 0.701 ±0.009 0.748 ±0.037 0.918 ±0.037 24 0.577 ±0.014 0.682 ± 0.040 0.793 ±0.009 0.848 ±0.036 0.991 ±0.018 36 0.588 ±0.050 0.780 ±0.034 0.846 ±0.023 0.985 ±0.016 1.117 ± 0.066 * Values are means ± standard deviations tor 3 replicates 235 Table 8.6.3. Growth (absorbance at 600 nm) of beef microorganisms Time Absorbance (600 nm) (h) Difco Oxoid BBL FSHCL FSPF 0 0.000 ± 0.000* 0.000 ± 0.000 0.000 ±0.000 0.000 ±0.000 0.000 ±0.000 1 0.002 ±0.001 0.003 ±0.001 0.003 ±0.003 0.002 ±0.001 0.003 ±0.002 2 0.003 ±0.001 0.003 ±0.001 0.008 ±0.000 0.032 ±0.005 0.006 ±0.001 6 0.054 ±0.003 0.086 ±0.001 0.076 ±0.012 0.142 ±0.008 0.222 ±0.018 9 0.474 ±0.033 0.595 ±0.014 0.441 ±0.003 0.579 ±0.028 0.745 ±0.028 12 0.636 ±0.053 0.744 ±0.019 0.460 ± 0.025 0.575 ±0.046 0.876 ±0.048 15 0.711 ±0.019 0.825 ±0.016 0.468 ±0.016 0.665 ±0.034 0.950 ±0.038 24 0.671 ±0.044 0.877 ± 0.019 0.651 ±0.018 0.804 ±0.027 1.005 ±0.049 36 0.549 ±0.030 0.845 ±0.053 0.687 ±0.034 0.964 ±0.019 1.153 ± 0.035 * Values are means ± standard deviations for 3 replicates Table 8.6.4. Growth (absorbance at 600 nm) ofE. coli Time Absorbance (600 nm) (h) Difco Oxoid BBL FSHCL FSPF 0 0.000 ±0.000* 0.000 ±0.000 0.000 ±0.000 0.000 ± 0.000 0.000 ± 0.000 1 0.035 ± 0.009 0.068 ±0.013 0.105 ±0.012 0.089 ±0.009 0.096 ±0.009 3 0.209 ±0.014 0.173 ±0.005 0.325 ±0.006 0.487 ±0.026 0.570 ±0.022 6 0.318 ± 0.031 0.368 ±0.005 0.452 ±0.003 0.777 ±0.012 0.802 ±0.016 9 0.342 ±0.030 0.405 ±0.008 0.484 ±0.003 0.877 ±0.009 0.921 ±0.016 21 0.442 ±0.011 0.506 ±0.008 0.488 ±0.006 1.074 ±0.013 1.111 ±0.011 24 0.478 ±0.017 0.502 ±0.006 0.511 ±0.027 1.210 ± 0.087 1.217 ± 0.015 36 0.518 ± 0.019 0.512 ±0.007 0.837 ±0.031 1.343 ±0.052 1.407 ±0.011 * Values are means ± standard deviations for 3 replicates Table 8.6.5. Growth (absorbance at 600 nm) of Staph. aureus Time Absorbance (600 nm) (h) Difco Oxoid BBL FSHCL FSPF 0 0.000 ±0.000* 0.000 ±0.000 0.000 ±0.000 0.000 ±0.000 0.000 ± 0.000 1 0.101 ±0.002 0.115 ±0.006 0.128 ±0.003 0.118 ± 0.006 0.158 ±0.003 2 0.238 ±0.002 0.255 ±0.010 0.255 ±0.006 0.360 ±0.031 0.353 ±0.002 6 0.549 ±0.011 0.657 ±0.023 0.510 ±0.024 1.140 ±0.032 1.022 ±0.020 9 0.609 ±0.038 0.734 ±0.019 0.517 ±0.006 1.580 ±0.045 1.135 ±0.053 12 0.611 ±0.026 0.666 ±0.010 0.466 ± 0.004 4.115 ±0.038 0.947 ±0.015 21 0.598 ±0.071 0.708 ±0.008 0.530 ±0.006 1.038 ±0.039 1.000 ±0.072 24 0.680 ±0.028 0.671 ±0.005 0.492 ±0.024 0.965 ±0.036 0.941 ±0.039 36 0.697 ±0.043 0.754 ±0.020 0.506 ±0.018 0.874 ±0.028 0.944 ± 0.036 * Values are means ± standard deviations for 3 replicates 236 Table 8.6.6. Growth (absorbance at 600 nm) of S. cerevisicie Time Absorbance (600 nm) (h) Difco Oxoid BBL FSHCL FSPF 0 0.000 ± 0.000* 0.000 ±0.000 0.000 ± 0.000 0.000 ±0.000 0.000 ±0.000 1 0.010 ±0.002 0.005 ±0.002 0.008 ±0.001 0.006 ± 0.002 0.006 ±0.006 3 0.010 ±0.002 0.007 ±0.003 0.008 ± 0.002 0.002 ±0.003 0.009 ± 0.004 6 0.011 ±0.003 0.011 ±0.001 0.013 ±0.002 0.013 ±0.003 0.014 ±0.004 9 0.017 ± 0.001 0.020 ±0.002 0.027 ±0.001 0.025 ±0.002 0.034 ±0.003 12 0.037 ±0.005 0.062 ±0.001 0.090 ±0.006 0.077 ±0.007 0.114 ± 0.018 24 0.886 ±0.166 1.370 ±0.097 1.171 ±0.016 1.608 ±0.047 1.285 ±0.053 36 1.383 ±0.043 1.777 ±0.073 2.217 ± 0.104 2.625 ± 0.094 2.363 ±0.098 48 1.546 ± 0.101 2.525 ±0.300 2.686 ±0.022 2.859 ±0.051 2.619 ±0.054 72 1.981 ±0.026 2.253 ±0.078 2.742 ±0.048 3.221 ±0.038 2.715 ± 0.081 * Values are means ± standard deviations for 3 replicates Table 8.6.7. Growth (absorbance at 600 nm) of B. subtilis Time Absorbance (600 nm) (h) Difco Oxoid BBL FSHCL FSPF 0 0.000 ±0.000* 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 1 0.051 ±0.006 0.056 ± 0.003 0.083 ±0.007 0.062 ±0.003 0.077 ±0.002 3 0.415 ±0.003 0.431 ±0.018 0.520 ±0.031 0.577 ±0.014 0.700 ±0.019 6 0.440 ±0.002 0.761 ±0.002 0.618 ± 0.017 0.582 ±0.003 0.819 ±0.009 9 0.544 ± 0.012 0.931 ±0.017 0.790 ±0.006 0.566 ±0.014 0.837 ±0.002 12 0.791 ±0.007 0.985 ±0.016 0.906 ±0.025 0.644 ± 0.003 0.865 ±0.009 24 0.804 ±0.001 0.952 ±0.038 0.975 ±0.008 0.769 ±0.015 0.846 ±0.019 36 0.655 ±0.032 0.786 ±0.015 1.085 ±0.077 0.910 ± 0.016 0.939 ±0.029 * Values are means ± standard deviations for 3 replicates 237 8.7 Published paper Animal Products 379 UTILIZATION OF FISH VISCERA Achmad Poernomo and K.A. Buckle * Research Institute for Marine Fisheries, Jakarta, Indonesia 2 Dept. Food Science and Technology, The University of New South Wales, Kensington, Australia ABSTRACT Fish processing produces a significant amount of waste such as viscera, liver, bones and tails. Such waste is not fully utilized and creates potential environmental problems. Efforts to recover valuable materials from ray processing wastes are now being studied at the Research Station for Marine Fisheries, Jakarta, and the The University of New South Wales. Chopped ray viscera was acidified with a mixture of propionic and formic acids (1:1, v/v), and hydrochloric acid at concentrations of 3 and 4% (v/w), and stored at ambient temperatures (28-32° C) and 40°C. The degree of liquefaction was monitored through analyses for pFI, non-protein nitrogen (NPN) content, viscosity and ratio of liquid/sedi ment. Rapid liquefaction, indicated by a sharp increase in NPN and ratio of liquid/ sediment, and decrease in viscosity, took place within the frst 5 days followed by slower liquefaction in subsequent days. The level of acids did not affect the degree of liquefac tion, and a higher degree of liquefaction was found in silages stored at 4(PC. Only mix tures of propionic and formic acids were able to preserve the viscera for more than 30 days. INTRODUCTION In Indonesia, fish viscera, liver, bones and tails are considered as waste and uti lized only to a minor extent. Those are 41-56% of the whole body weight (Zaitsev, et al., 1969) and if not carefully handled might create environmental problems. Fish viscera contains valuable materials and can be utilized as animal feed or biochemicals (Almaas, 1990). As animal feed, fish viscera improves the general con dition of animals, but sometimes when fed in a partly putrefied form the animals may become seriously ill of even suffer form fatal toxic effects (Gildberg and Almaas, 1986). Therefore rapid and permanent preservation of fish viscera is necessary. In developing countries such as Indonesia, fish processing wastes are often scattered in many processing sites, underlining the need of such preservation methods. Fish waste is usually converted into meal before being processed into feeds. Fish viscera, however, is very susceptible to autolysis due to active digestive enzymes and is not suitable for fish meal. Ensilation, on the other hand, is able to preserve fish 238 380 Development of Food Sci. and Technol. viscera without significantly affecting the nutritive value (Gildberg and Almaas, 1986). The procedure is very simple and rapid autolysis during ensilation makes it possible to separate the aqueous protein solution, and thus recovery of valuable materials is easier (Clausen, et al., 1985; Reece, 1988). Silage from fish viscera has been extensively studied by several workers (Backhoff, 1976; Raa and Gildberg, 1976; Gildberg and Raa, 1977; Wignall and Tatterson, 1976). However, those studies were mainly on fish from temperate waters, and work on tropical fish viscera is very limited. The aims of this study were to investigate the ensilation of ray (Trygon sephen) viscera from tropical waters, and to establish the conditions to produce stable silage. MATERIALS AND METHODS Ray viscera were obtained from Muara Angke, Jakarta, and transported to the Research Station for Marine Fisheries laboratory in ice. The length of fishing as well as the period between catching and processing of the ray were not known. Upon arrival at the laboratory, the liver, gonads and viscera contents were discarded; the viscera was then washed in fresh water and chopped. The average proximate compo sition of the viscera was moisture 83.4%, protein 14.5%, fat 0.7% and ash 1.2% (wet weight basis). Ensilation was done in glass containers, each containing 750 g chopped viscera. A mixture of propionic and formic acids (1:1, v/v), and hydrochloric acid each at concentrations of 3 and 4% (v/w), were added to the viscera, and mixed thoroughly. The mixtures were then stored at ambient (28-32°C) and 40°C and stirred daily during the first week. At each predetermined storage time, pH, viscosity, non-protein nitrogen (NPN) of the liquid phase and ratio of liquid/sediment were measured. The experiments were replicated twice, with measurements done in triplicate. Methods of analysis for proximate composition were adopted from Hasegawa (1987), while pH was measured using a pH meter (Digital Ionolyser Model 601A, orion Research Inc.) calibrated with buffer solutions of pH 4 and 7. A viscometer (LVT Model, Brookfield Eng. Lab. Inc.) was used to measure the viscosity of silage. The liquid phase of silage was separated by centrifugation at 2000 g for 10 min (Beckman J2/-21/Beckman Instrument Inc.). Non-protein nitrogen was measured on a 20% TCA soluble fraction of liquid phase using a micro kjedahl method. RESULTS AND DISCUSSION Pumomo and Silvany (unpublished date) earlier used 2% acid, which was higher than recommended by other workers using fish viscera from temperate waters (Raa and Gilberg, 1976). However, this concentration was not sufficient to preserve and liquefy the viscera form tropical waters. The silages deteriorated within 1 day, probably because of the higher buffer capacity of tropical fish due to higher mineral content (Zaitsev, 1969). Therefore, higher acid concentrations 3 and 4%, were ap plied. 239 Animal Products 381 Upon addition of acids, the pH decreased to 3.25-4.95 depending upon the type and concentration of acid; the higher the concentration the lower the pH. At the same concentration, hydrochloric acid produced lower pH than a mixture of propion ic and formic acids. The colour of the viscera changed from reddish white to greyish white, probably due to oxidation of some pigments in the viscera. During storage, pH of silages produced with hydro- chloric acid increased while those produced with a mixture of propionic and formic acids were relatively stable (Fig. 1). Silages produced with 3% hydrochloric acid spoiled within 5 days at both ambient temperature and 40°, while those prepared with 4% hydrochloric acid spoiled after 21 days at ambient and 23 days at 40°C. It appears that hydrochloric acid is not able to lower the pH sufficiently. Almaas (1990) also found that if mineral apids were used, the pH should be as low as 2 to secure preservation. The deteriora tion of the silages in this experiment was initiated with the appearance of mould growth on the surface or on the walls of the containers, followed by an unpleasant smell within 1-2 days. When advanced deterioration occurred, the pH increased, reaching 6.5 or more and the colour changed from reddish to dark brown. Similar signs were also observed on deteriorated viscera silage from temperate waters (Raa and Gildberg, 1976). Thus although hydrochloric acid might be able to hydrolyze pro tein in the viscera, it failed to preserve it. Similar results were found by Levin et al. (1989) who had used phosphoric acid to ensile fish entrails and frames. PH / >• / / ------' «o oto. c Time (Jays) Praptaalc + Paralc J * Proploa Ic ♦ l:of m Ic 4% “ H yd r»< h lor I c *«I4 )% Hydrochloric «cld 4% Fig. 1. pH of ray viscera silage stored at ambient (28-32°C) and 40°C. 240 382 Development of Food Sci. and Technol. Protein in the viscera is digested during ensilation by proteolytic enzymes such as trypsin, chymotrypsin and pepsin producing lower weight molecular products which remain dis-solved in the liquid phase. Soluble nitrogen can be used to measure the extent of protein hydrolysis in ensilation (Tatterson, 1982). In this experiment, TCA soluble non-protein nitrogen in the liquid phase of the silage increased markedly (2.5- 3 fold) within 5 days and at a lower rate afterwards (Fig. 2). This demonstrates the typical pattern of nitrogen release involving rapid proteolysis during the first 5 days, and supports the findings of other workers who observed rapid proteolysis in 5-10 days for tropical fish and for temperate fish (Raa and Gildberg, 1976) viscera. The ratio of NPN to total N was in the range of 70-95% showing that proteolysis was almost complete. This was also supported by the results for viscosity and the ' amount of liquid after centrifugation. When proteolysis proceeded, the silage lique fied and the viscosity was reduced (Fig. 3), and at the same time the ratio between the liquid and sediment after centrifugation increased (Fig. 4). As shown in Figs. 2, 3 and 4, silages stored at 40°C had much lower viscosity, slightly higher NPN in the liquid phase and higher ratio of NPN/total N than those stored at ambient temperatures indicating a higher degree of liquefaction. Thus increased storage temperature slightly affected the degree of liquefaction, in support of the views of other authors (Backhoff, 1976; Raa and Gildberg, 1976). The concentrations of the propionic and formic acid mixture used in this study did not greatly affect the degree of liquefaction of ray viscera. Thus an acid concen tration of 3% (v/w) is adequate. Further studies on prolonged storage of ray viscera silage and on the isolation and characterization of valuable materials from ray viscera silage are in progress. N PN/Total N (%) NPN/Tolal N (%) Time (days) 1» ----- Proploalc + Forvlc — »cld 1* —- Mrdr.tal.rlc »cld < s Fig. 2. Non-protein nitrogen liquid phase of rav viscera silage stored at ambient (28-32°C) and 4Cr C . 241 Animal Products 383 Viscosity (cps) II 1< 21 !• II Time (days) Time (days) —— Propiaa Ic Far ■ tc J* P rapioa Ic ♦ F«» m lc 4% —H yd nac h laric mI4 J* '* ' Hy4rackl«(lc «U 4% Fig. 3. Viscosity of ray ray viscera silage stored at ambient (28-32°) and 40°C. I.iquid/Scdiment (v/w) l.iquid/.Sediment (y/w) ----- PrtflMk ♦ KoimIc J* — Kr«pUai<« Fatale 4% • H*4rocklarlc acid —— H y d roc fc I • f It acid 4* Fig. 4. Liquid/sediment ration of ray viscera silage stored at ambient (28-32°C) and 40°C. 242 384 Development of Food Sci. and Tedtnoi. CONCLUSION This study has examined the ensiiation of ray viscera by acid preservation. It is recommended to use a mixture of propionic and formic acids (1:1, v/v) at a concen tration of 3% (v/w). Viscera silage can be stored at ambient temperatures (28- 32°C) or 40°C for more than 30 days. However, the higher temperature (40°C) will increase the degree of liquefaction. REFERENCES Zaitsev, V., Kizevetter, I., Lagunov, L., Makarova, L. Minder, T. and Podsevalov, V. , 1969. Fish Curing and Processing. Translated from the Russian by DeMerindoi A. Moscow, MIR Publ, pp 37-44. Aimaas, K. A. 1990. Utilization of Marine Biomass, in Voigt MN, Biotechnology for increased Profitability. Lancaster, PA, Technomic Publ. Inc., pp 361-372. Gildberg, A., Aimaas, K.A. 1986. Utilization of fish Viscera, in Botta JR (eds): Advances in Fisheries Technology and LeMaguer M, Jelen P (eds): Food Engineering and Process Application Vol. 2. Amsterdam, Elsevier Appl. Sci. • Publ., pp 383-393. Clausen, E., Gildberg, A. and Raa, J. 1986. Preparation and testing of an autolysate of fish viscera as growth substrate for bacteris. A;;i. Environ. Microbial. 50(6), pp 1556-1557. Reece, P. 1986. Recovery of proteases from fish waste. Process Biochem, 23(3), 62- 66. Backhoff, H. 1976. Some Chemical changes in fish silage. J. Food Technol., 11, 353-363. Raa, J. and Gildberg, A. 1976. Food Technol., 11, 619-628. Gilberg, A. and Raa, J. 1977. properties of a propionic acid/formic acid preserved silage from cod viscera. J. Sci. Food Agric., 28, 647-653. Wignall, J., and Tatterson, I.N. 1976. Fish silage. Process Biochem, 11(10), 17-19 Hasegawa, H. (ed). 1987. Laboratory Manual on Analytical Methods and Proce dures for Fish and Fish Products. Singapore, SEAFDEC. Poernomo, A.,and Silvany, I. Unpublished data. 243 Animal Products 385 Levin, R.E., Witkowski, R., Meirong, Y. and Goldhar, S. 1989. Reaserch note: Preparation of fish silage with phosporic acid and potassium sorbate. J. Food Biochem., 12, 253-259. Tatterton, I. N. 1982. Fish silage. Anim. Feed Sci. Technology, 7, 153-159. 244