ALLEVIATION OF ANAEMIC GROWTH SUPPRESSION IN FISH

The present invention relates to the field of fisheries and aquaculture of all kinds, including the farming of fish, including saltwater or freshwater fish. The invention also relates to the treatment of fish to relieve conditions of anaemia. Further, the invention concerns composition for fish food for alleviating anaemia in fish and increasing their growth rate Diminishing ocean fish stocks is causing a shift in fishing patterns and pressure in order to realise a goal of sustainable fisheries. However, that process of shift to sustainability is not guaranteed, nor can it guarantee to supply increasing human demand for fish. Fish farming and aquaculture provides an alternative to sustainable ocean fishing and the number of species of fish, crustacean and shellfish which are being farmed around the world is on the increase.

Solea solea (the Dover sole) is a sandy and muddy bottom dwelling sea fish living down to about 300 metres that is exploited commercially by trawling. The diet of S. solea has been studied and found to include polychaete worms, small soft-shelled bivalves, small fishes and crustaceans. According to the UN Food and Agriculture Organisation (FAO) fishery statistics, global production in 2010 was about 40,000 tonnes, down from a recent peak of just over 60,000 tonnes per year in 1993 - 1995.

S. solea is also now being produced on a smaller scale in aquaculture systems and according to FAO data first records of production go back to 1980 with production in 2010 at about 120 tonnes.

Dover sole has been found to grow well in aquaculture when it is fed on a diet of fresh Nereis virens (ragworm) and mussels, but this is not at present an economically viable way of producing the fish. Ragworm is expensive to produce. Conventional fish foods therefore continue to be used. As a result, the productivity of Dover sole in aquaculture systems continues to remain below expectations of what the fish is capable of in terms of growth. Much research therefore has been carried out to try and formulate an optimal diet for Dover sole in aquaculture, but without success. Various aspects have been investigated, including attractivity, nutritional composition, probiotic effects and processing effects. So far though, no one has succeeded in formulating an improved or optimal fish food for Dover sole. No currently available fish food composition or diet is able to realise the expected growth potential of Dover sole. The 25 - 65% faster growth rate of sole fed on ragworm compared to commercially available fish food shows that there is considerable growth potential to be achieved in sole and a fish food composition has yet to be formulated which would work to achieve such a faster growth rate.

WO201 1/027279 A1 (UNIV DO ALGARVE et al) describes feed additives for aquaculture comprising 1 -methyl-L-tryptophan and/or its isomers. The compounds are disclosed as being olfactory attractants and inducers of digestion in fish and are for use at concentrations between 0.001 and 10 mg per kg of dry fish feed. The compounds were identified from seawater conditioned by the presence of 100g per liter of a polychaete worm Hediste diversicolor which was known to an important element of the natural diet of the Senegalese sole, S. senegalensis. The conditioned seawater was lyophilized to reduce its volume and then fractionated and tested on anaesthetised Senegalese sole subjected to olfactory receptor neurone electrode measurements. In the final fractionations mass spectrometry was used to identify 1 - methyl-L-tryptophan as the olfactory compound.

The condition of anaemia is known to occur from time to time in various species of fish, salt or freshwater due to disease. Serious conditions are often manifest by pale gills and surface swimming and gulping. The disease agent may be a bacteria, a virus or a parasite, such as a leech or lice. For example, there is growing concern about incidences of anaemia in farmed salmon caused by the infections salmon anaemia virus (ISAV).

CN101491669 A (UNIV ZHEJIANG OCEAN) discloses a fish protein iron-peptide capsule (tablet or granule) for preventing iron deficiency anaemia, which is prepared by turning low-value marine fish and leftover protein into composite polypeptide through protein enzymatic hydrolysis, modifying the composite polypeptide with ferrous iron so as to obtain fish protein iron-peptide and mixing the fish protein iron- peptide with safflower seed oil, beeswax, phospholipids, starch according to a certain weight ratio. The invention discloses three formulations and one optimum

formulation . The chemical nature of the capsule is alleged to be a chelated product of peptide, amino acid and ferrous iron. The protein content of the capsule is more than or equal to 30 percent, and the total iron content of each capsule is 10+/-0.5 mg. The capsule is a novel functional food for preventing iron deficiency anaemia.

CN1644080 A (MINGBO AQUATIC PRODUCT CO LTD) discloses a feed for the half-smooth tongue-sole that is prepared from pulverized clam worm, oyster meat, fish powder, cybiidae powder, vitamin C, vitamin E, cod-liver oil, yeast, etc. Its advantages are easy absorption and digestion, and promoting growth.

In the field of fish culture the problem needing to be solved is how to increase the productivity of the systems in use.

The inventors are surprised to discover from their work that by supplementing the diet of fish with annelid worm, an extract or fraction thereof, and/or a mollusc, mollusc extract or fraction thereof, that an anaemic state that holds back growth of the fish is reduced or overcome. In other words, the "scope for growth" of the fish is made accessible and the inventors refer to this as the "the worm effect". The inventors have also surprisingly discovered that the "worm effect" is due to heme. Also discovered is that the "worm effect" includes a combination of heme and vitamin Bi ₂ . Further, the inventors have discovered that best increases in fish growth rate compared to available commercial feeds arise when there is a combination of added heme and vitamin B ₂ in the diet and there is a level of taurine presence comparable to the levels generally found in fishmeal .

Accordingly in one aspect the present invention provides a method of increasing the productivity of a system for culturing fish comprising feeding the fish a diet, which diet comprises a source of heme. Advantageously, the full natural growth potential, in term of specific growth rate (SGR) in %.d ^"1 , growth in metabolic body weight (g.kg ^{" 0 8} .d ^"1 ) and/or growth in gram per day is realised compared to non-supplemented feeds. Any source of heme is suitable so long as iron is bound to heme and not free. For example, bovine haemoglobin, iron sulphate hydrate, iron proteinate, or iron methionate.

Productivity of the system of the invention may be as much as 287% of ordinary pellet fed farmed fish; optionally 243%, 197% 167% or 135%.

The diet may further comprise a source of vitamin Bi ₂ . Vitamin B12 or cobalamin belongs to a group of cobalt containing compounds known as corrinoids that contain a specific corrinring. Known forms are methylcobalamin, adenosylcobalamin, hydroxocobalamin, cyanocobalamin, sulfitocobalannin and possibly other unknown forms.

Within the term "source of heme" is heme itself, and within the term "source of vitamin B ₂ " is vitamin B ₂ itself.

A culture system for fish includes all kinds of aquaculture and fish farming, including "sea ranching" whereby fish are allowed to roam but they are preconditioned to go to a feeding area at specific times where they are rewarded with the free availability of food which may include molluscs and/or annelids.

The diet of the invention preferably has a taurine level comparable to levels found in fish meal. Particularly advantageous increases in growth rate of fish are found where there is a combination of heme and vitamin Bi ₂ and a level of taurine comparable to levels found in fishmeal or compared to commercially available fish diets based on fish meal. In preferred embodiments the diets of the invention contain taurine in the range 3.1 to 7.6. kg ^"1 .dm.

Table below shows preferred minimum and maximum desired levels of heme;

optionally vitamin B-| ₂ in accordance with the invention. These may vary to some extent depending on the species. Table

Based on maximum iron concentration not on leucine imbalances! These maxima do also depend on the species. ^" According to literature there are no known negative effect caused by high vitamin Bi ₂ intake. Therefore no max inclusion level can be given.

The source of heme may be an annelid worm or an extract or fraction thereof;

optionally wherein the annelid worm is a member of the class Polychaeta; preferably a ragworm or is a member of the family Nereidae; more preferably wherein the annelid worm is of the genus Nereis; even more preferably the species Nereis virens.

Alternatively or in addition, the source of heme may be a mollusc or an extract or fraction thereof; optionally wherein the mollusc is a species of the phylum Mollusca; preferably the class Bivalvia.

In another aspect of the invention, the source of heme is not an annelid worm or an extract or fraction thereof; and not any annelid worm being is a member of the class Polychaeta; not a ragworm or a member of the family Nereidae; also not an annelid worm of the genus Nereis; and not species Nereis virens. Also, the source of heme is not a mollusc or an extract or fraction thereof; and not a species of the phylum Mollusca; and not a member of class Bivalvia.

The source of vitamin Bi ₂ may be fermented salmon kidneys. A known preparation called "Mefun" which is a natural product with the highest known Bi ₂ content (about 3280 μg.kg ^"1 ).

The heme and vitamin B ₂ may be mixed with and thereby may form part of the food given to the fish. Advantageously this allows easier supplementation and feeding regimes in that fish may be fed from a single bulk supply of pre-prepared food.

Alternatively, the heme and vitamin B-| ₂ may be formulated to be given to the fish alongside the food as a separate supplement, but not actually mixed into the normal food to form part of it. In such supplement embodiments, the heme and vitamin B ₂ may be formulated as a mixture or as separate additives.

The fish is preferably a saltwater fish, although a freshwater fish may be the subject of the method of the invention.

A preferred fish subjected to the method of the invention may be a member of the family selected from Solidae, Achiridae or Cynoglossidae. An alternatively preferred freshwater fish is selected from the family Cyprinidae; preferably Danio rerio (zebra fish).

Where the fish is a member of the genus Solea, preferred fish are Solea solea or Solea senegalensis.

Over 50 species of commercially exploited fish may be suitably fed with a

supplemented diet in accordance with the present invention. As well as the aforementioned fish, such fish may be selected from: Seriola quinquieradiata (Japanese amberjack), Epiniphelus fusoguttatus (Tiger grouper), Onchrhynchus kisutch (Coho salmon), Sparus aurata (Giltehead sea bream) Pagrus major (Red sea bream), Gadus morhua (cod), Sander lucioperca (Pike), Oncorhynchus mykiss (Rainbow trout), Salvelinus alpines (Arctic char), Oncorhynchus tshawatcha

(Chinook), Salmo trutta (Brown trout), Lutjanus erythropterus (Red snapper),

Dipolodus sargus (White sea bream), Oreochromis niloticus (Nile tilapia),

Pseudosciaena crocea (Large yellow croaker), Pollachius pollachius (Pollack), Dicentrarchus labrix (Sea bass), Hippoglossus hippoglossus (Halibut), Lates calcarifer (Barramundi), Morone saxatilis (Striped bass), Coryphaena hippurus (Mahi mahi), Pagellus bogaraveo (Black spot sea bream), Paralichtys olivaceus (Bastard halibut), Cyprinus carpio (Carp), Scopthalmus/Psetta maximus (Turbot), Salmo salmar (Atlantic salmon), Epinephelus coioies (Green grouper), Anguilla anguilla (European eel). Acipenser sturio (Sturgeon), Clarias gariepinus (African catfish), Salvenlinus fontinalis (Brook trout), Anarhichas lupus (Wolf fish), Ictalurus punctatus (Channel catfish), Thunnus thynnus (Northern Bluefin tuna). Ornamental or aquarium species of fish may also benefit from the present invention.

The annelid worm, extract or fraction, and/or the mollusc, extract or fraction is preferably in the form of a meal and/or is freeze dried. Where the annelid worm or mollusc is taken and used as the whole animal, then it would usually be dried and processed to form a meal, but without boiling. Alternatively, the whole animals may be freeze dried and then physically processed to form flakes, granules or powder.

The average temperature of the water in which the fish are maintained is depending on the species and size of species cultured. For common sole (Solea solea) during grow-out this is preferably not less than about 15.5 °C; more preferably the average temperature is a temperature falling in the range 15.5 °C to 20 °C; even more preferably an average temperature of 17.5 °C. Growth of fish at such average temperature regimes using the diets of the present invention advantageously results in a reduction or elimination of nutritionally incurred anaemia and optimally increased growth rate compared to a normal, non-supplemented diet. Similar temperature ranges apply for other species.

Fish grown in accordance with the invention may have a growth rate (in g per day) which is at least 5% or at least 10%, or at least 15%, or at least 20%, or at least 25%, or at least 30%, or at least 35%, or at least 40%, or at least 45%, or at least 50%, or at least 55%, or at least 60%, or at least 65%, or at least 70%, or at least 75%, or at least 80%, or at least 85%, or at least 90%, or at least 95%, or at least 100% greater than the equivalent growth stage of fish fed the same fish food but lacking the essential source of heme and source of vitamin B-| ₂ .

A culture system for fish may comprises a body of marine water and associated marine bed; optionally delineated at least in part by a physical barrier, e.g.

comprising a mesh screen or net. Alternatively, the culture system may comprise a tank of seawater.

The invention also provides heme for use as an oral medicament in fish. The invention further provides heme in combination with vitamin B ₂ for use as an oral medicament for fish.

The nature of the oral medicament is such that it is in any form ingestible by fish, preferably by self-feeding from the medicament being made available in the form of a food or a supplement provided with food.

The invention includes a combination of heme and vitamin Bi ₂ for use in the prevention or treatment of anaemia in fish, wherein the heme and vitamin B ₂ are in the form of an oral composition.

Therefore, the invention includes a method of preventing or treating anaemia in fish comprising providing a combination of heme and vitamin B ₂ to the fish. The invention also includes a method of increasing haematocrit in a fish comprising feeding the fish with a fish food supplemented with heme and vitamin B-| ₂ .

In the aforementioned medical uses of the invention, such medicaments and compositions are preferably administered as a food supplement or included within food forming the general diet of the fish. The optional and preferred features of the method of the invention as hereinbefore defined apply equally to the aforementioned medical uses of the invention.

In preventing or treating anaemia in fish, preferably sole, in accordance with the invention, and/or in increasing the growth of fish in accordance with the invention, the haematocrit of the treated fish is at least about 16%; preferably the haematocrit is in the range 16 - 30 %; more preferably 16 - 20%. The haematocrit of treated fish may therefore be 17% + 1 %; or 18%; + 1 %; or 19 + 1 %; or 20% + 1 %; or 21 % + 1 %; or 22%; + 1 %; or 23% + 1 %; or 24% + 1 %; or 25%; ± 1 %; or 26% + 1 %; or 27% + 1 %; or 28% + 1 %; or 29%; + 1 %; or 30% + 1 %. Whilst specific haematocrit levels may differ in other fish species, values for sole may be considered broadly representative in many other species. Additionally or alternatively to the above, in preventing or treating anaemia in fish, preferably sole, in accordance with the invention, and/or in increasing the growth of fish in accordance with the invention, the haemoglobin level of the treated fish is at least about 22 g/l; preferably the haemoglobin is in the range 22 - 47 g/l; more preferably in the range 22 - 27 g/l. The haemoglobin level of the treated fish may therefore be 22 + 1 g/l; or 23 + 1 g/l; 24 + 1 g/l; or 25 + 1 g/l; 26 + 1 g/l; or 27 + 1 g/l; 28 + 1 g/l; or 29 ± 1 g/l; or 30 ± 1 g/l; or 31 + 1 g/l; 32 + 1 g/l; or 33 ± 1 g/l; 34 + 1 g/l; or 35 + 1 g/l; 36 + 1 g/l; or 37 + 1 g/l; 38 + 1 g/l; or 39 + 1 g/l; or 40 + 1 g/l; or 41 + 1 g/l; 42 + 1 g/l; or 43 + 1 g/l; 44 + 1 g/l; or 45 + 1 g/l; 46 + 1 g/l; or 47 + 1 g/l. Whilst specific haemoglobin levels may differ in other fish species, values for sole may be considered broadly representative in many other species.

Additionally or alternatively to the above, in treating or preventing anaemia in fish in accordance with the invention, and/or in increasing the growth of fish in accordance with the invention, the specific growth rate (SGR) of the treated fish is greater than about 0.75 % body weight per day (%bw.d ^"1 ). In preferred embodiments the SGR of treated fish is at least about 0.8 %bw.d ^"1 ; or at least 0.9 %bw.d ^"1 ; or at least 1 .0 %bw.d ^"1 ; 1 .1 %bw.d ^"1 ; or at least 1 .2 %bw.d ^"1 ; or at least 1 .25 %bw.d ^"1 ; or at least 1 .3 %bw.d ^"1 ; or at least 1 .4 %bw.d ^"1 ; or at least 1 .5 %bw.d ^"1 ; or at least 1 .6 %bw.d ^"1 ; or at least 1 .75 %bw.d ^"1 .

Additionally or alternatively to the above, in treating or preventing anaemia in fish in accordance with the invention, and/or in increasing the growth of fish in accordance with the invention, the growth rate of fish may be greater than about 0.8 g per day (g.d ^"1 ). Preferred growth rates of fish may be greater than about 0.9 g.d ^"1 ; or greater than about 1 .0 g.d ^"1 ; or greater than about 1 .0 g.d ^"1 ; or greater than about 1 .1 g.d ^"1 ; or greater than about 0.9 g.d ^"1 ; or greater than about 1 .0 g.d ^"1 ; or greater than about 1 .25 g.d ^"1 . Fish which may be subject to the various aspect of the invention may be at any stage from juvenile (e.g. from as early as weaning from yolk sac to artificial feeds) to adult. Fish as small as 0.03 g may be subjected to the methods of the invention. Mature fish such as Sole, grown or treated in accordance of the invention, may have a weight in the range 50g to 350g, for example.

In another aspect, the invention provides a method of making fish food composed of a proteinaceous material of animal and/or fish and/or vegetable origin, comprising adding a source of heme to the proteinaceous material. Preferably the method of making fish food further comprises adding a source of vitamin Bi ₂ .

The source of heme and optionally the source of vitamin B ₂ may be as set forth in relation to the method of increasing productivity invention as hereinbefore defined. The source of heme and optionally vitamin Bi ₂ may be incorporated into fish food pellets by a process of cold extrusion with the fish food composition.

In accordance with the invention a conventional fish food may be supplemented with a source of heme and optionally a source of vitamin Bi ₂ . The source of heme and optionally the source of vitamin B-| ₂ may be as set forth in hereinbefore defined. The conventional fish food may be made by grinding and mixing together ingredients such as fishmeal, fish oil, vegetable proteins and binding agents, e.g. wheat. The resultant past (water may be added to improve consistency) is extruded through a die (e.g. simple holes in a plate). The diameter of the pellets depending on the species and size (growth stage) of species cultured. The time of adding oil is depends on the process used and species specific requirements (fat requirements, or e.g. the need for floating, slow sinking or sinking pellets). In the process producing pellets e.g. for Sole, (a low fat diet) the oil is mixed with the other ingredients before forming.

Sometimes it is necessary to add the oil after the drying process (e.g. vacuum coating). The temperature at which pellets are formed and the amount of oil affects their density and therefore whether they float or sink. The supplemental ingredients of the invention may be added before, during or after pelleting. Commercial fish foods are readily obtainable from the likes of Biomar, EWOS or Skretting.

A method of feeding fish in accordance with the invention comprises making a proteinaceous fish food available to the fish and whether prior to, separately, simultaneously or subsequent to making food available, administering a source of heme; optionally also a source of vitamin B ₂ .

Also provided is a kit for improving the productivity of an aquaculture system comprising at least one container which contains a source of heme; optionally wherein a source of vitamin B ₂ is provided in the same or different container to the source of heme.

In a preferred kit of the invention, the at least one container is graduated, or the at least one container is a unit dose of heme, optionally a unit dose of vitamin B ₂ in addition. Optionally the kit may include a set of instructions for generating the correct level of supplement to the conventional fish food.

The invention also includes a fish food comprising protein of plant and/or animal origin, gluten, a binder, salt, minerals and vitamins, and supplemented with a source of heme; optionally also a source of vitamin Bi ₂ . In preferred fish foods of the invention there is a level of taurin comparable to levels found in fish meal. In diets of the invention, e.g. for sole, taurin levels vary between 3.1 and 7.6 g.kg ^"1 .dm. The supplement may be present in an amount in the range 5 - 85% by weight of the total food; preferably an amount in the range 10 - 75% by weight. Other preferred ranges include 15 - 65%, 20 - 60%, 25 - 55%, 30 - 50% or 35 - 45%. The supplement may be present in an amount in the range 5 - 10%; 10 - 20%; 20 - 30%, 30 - 40%, 40 - 50%, 50 - 60%, 60 - 70% or 70 - 80%. Optionally the supplement may be present in an amount of about 10%, or about 25%, or about 50%, or about 75% by weight of total food. .

The plant protein may be pea protein and/or soy protein, e.g. soy protein

concentrate. The animal protein may be casein.

Fish foods of the invention may also comprise one or more of the following

ingredients: an oil, e.g. a fish oil; sugar; lime; wheat gluten; salt; choline chloride; vitamin C; vitamin E; betaine; preservative, e.g. BHT; antifungal, e.g. calproprionate. In all aspects of the invention, the source of heme and optionally source of vitamin B-12 is preferably not heat treated in excess of about 55°C, and in particular is not subjected to boiling for any period of time. Fish food which is supplemented in accordance with any aspect of the invention may have an amount of one or more of the following added to it to increase concentration thereof, whether prior to, during or after pellet formation: e.g. vitamin C, folic acid, vitamin B1 , vitamin B2, vitamin B5, vitamin A, iron, zinc, selenium, and copper etc. (e.g. see premix table 7).

Also, in particular embodiments of fish food used and supplemented in aspect of the invention there is preferably reduced or no detectable level of one or more of:

phytate and polyphenols. The invention will now be described in detail by way of examples and with reference to the accompanying drawings in which:

Figure 1 shows levels of oxygen, pH and temperature as monitored during the trial of example 1 investigating growth of sole fed on diets of fresh worm, pellets of a commercial fish food with extract of worm and untreated pellets of a commercial fish food.

Figure 2a shows the average feed intake in g. dm. fish ^"1 .d ^"1 as described in example 1 . Figure 2b shows the average growth in g.d ^"1 and standard deviation of fish fed ragworm, pellet treated with worm extract and the commercial pellet, as described in example 1 .

Figure 3 shows average haematocrit values for the sample group t(0) , negative control and pellets treated with extract do not differ, as described in example 1 .

Average haematocrit values of blood from sole fed with conventional pellets are significantly lower than values of sole fed fresh worms (p=0). Sole fed fresh worm does show an increase in haematocrit level of 87% compared to sole fed a

conventional with or without worm extract pellets. Figure 4 shows the relationship found in example 1 between feed intake _d m and temperature °C of culture water for sole fed pellet, pellet with extract and fresh worm. Figure 5 shows the relationship found in example 1 between temperature °C, htc and oxygen use (in ml.min ^"1 .kg ^"1 ) calculated from the feed intake per tank (following pattern of figure 4). Real oxygen use is expected even lower as requirements for maintenance (basic metabolic oxygen demand) are not included. VO ₂ inax for the three treatments and in relation to htc, VO ₂ max calculated using the relation between VO ₂ max and htc as estimated for trout by Gallaugher and Farrell (1998) Fish

Physiology: volume 17 Fish respiration (Academic Press).

Figure 6 shows the growth in metabolic bodyweight of fish fed ragworm from example 1 ; the treated and untreated pellet in period 1 , period 2 and the average of the total trial.

Figure 7 shows the growth in metabolic bodyweight of fish fed ragworm from example 1 ; the treated and untreated pellet in period 1 , period 2 and the average of the total trial.

Figure 8 shows the relationship between haematocrit (%) and growth in metabolic body weight in g.kg ^{"0 8} .d ^"1

Figure 9 shows the result from example 1 of feeding boiled worm compared to fresh worm has a negative effect on growth of sole, especially in period 1 with the higher water temperatures.

Figure 10 shows the result from example 1 that feeding boiled worm compared to fresh worm has a negative effect on haematocrit level of sole, although still higher than fish fed the commercial pellet.

Figure 1 1 shows from example 2 the relationship between worm meal inclusion level and growth in metabolic body weight. The inclusion level of 100% worm meal is the fresh worm. Figure 12 shows from example 2 the relationship between worm meal inclusion level and haematocrit (%). The inclusion level of 100% worm meal is the fresh worm. Average haematocrit levels measured in fish fed the commercial pellet was 1 1 .2%.

Figure 13 is from example 2 and shows the relationship between haematocrit and growth.

Figure 14 shows the relationship between iron content and growth in metabolic body weight. Iron content of pc (worm), nc (Weanex) and worm meal were analysed. Iron content of diets G, B, C, D and E were calculated.

Figure 15 shows the relationship between the calculated heme content, haematocrit and growth in metabolic body weight. Heme content of fresh worm analysed was comparable as values calculated using the data of Vinogradoff et al. (1991 ) "Iron and heme contents of the extracellular haemoglobins and chlorocruorins of Annelids." Comp. Biochem. Physiol. Vol. 98B, No. 2/3, pp. 187-194.

Figure 16 is from example 3 and shows the recovery pattern of sole fed ragworm (N. virens).

Figure 17 shows the relationship between haematocrit and haemoglobin as measured in the trial of example 3. Figure 18 shows the average haematocrit level in percentage (left) and the average haemoglobin levels in g.l ^"1 (right) and the standard deviations of sole fed ragworm, mussel and the commercial pellet.

Figure 18 shows the relation between bodyweight in gram and the specific growth rate (SGR) in %.d ^"1 of sole fed live ragworm.

Figure 19 shows the weight gain (g, in time (days) of sole fed live ragworm. Fermented salmon kidneys or "Mefun" is a natural product with the highest known B ₂ content (3280 ug.kg ^"1 ). "Mefun" can therefore be substituted for RMM (raw mussel meat) and/or ragworm, with equivalent or greater effect of alleviating Sole from anaemic growth suppression.

All experiments were approved by the ethical Committee for Animal Experiments (DEC) and conducted at Wageningen Aquaculture Centre (WAC), the Netherlands. For Common sole (Solea solea) the inventors have observed growth increase of between 25% and 67% when fed ragworm compared to when fed a commercial pellet.

The inventors have also observed the following needed for sole to achieve certain (plate) size categories:

Under commercial conditions using recirculation systems (RAS) and available commercial feeds sole will grow from 50 to 150 g in ± 9 months. It will take sole fed commercial pellets ± 21 months or more than twice as long compared to sole fed worm to grow to sizes of ± 350 grams.

When using worm as feed; sole fed worm will grow from 50-350 g in ± 9 months with far less mortality.

· Growth period extrapolated using growth figures achieved by feeding pellets with wormmeal; sole are expected to grow from 50 to 300 grams in ± 9 months.

Growth period extrapolated using growth figures achieved by feeding pellets of the invention; sole are expected to grow from 50 to 250 grams in ± 9 months. The table below shows expected growth based on the rise of haematocrit when sole is fed pellets using the invention. Sole fed pellets with the invention will most likely grow from 50-250 g in ± 9 months.

Table: Achieved growth rates

1 Pellet 3.67 0.50 100

Worm 6.13 0.91 167

2 Pellet 4.81 0.66 100

Pellet + 6.13 0.86 127

invention

(wormmeal)

Worm 7.38 1 .07 161

3 Pellet 4.58 0.64 100

Pellet + 5.82 0.84 127

invention

(wormmeal)

worm 6.46 1 .04 141

4 Pellet + 4.96 1 .06 No commercial invention pellet used as

(heme+B ¹² ) reference

Figure 19 shows the combined growth curve using two different size classes: a group that grew from 50 to 235 grams in 150 days and a group that grew from 210 to 370 gram in 150 days. From this figure one can extrapolate that sole fed ragworm can grow from 50 to 350 grams in 275 days.

In an experiment the pellet of the invention achieved growth figures of 0.73%.d ^"1 , 4.96 g.kg ^{0 8} .d ^"1 and 1 .06 g.d ^"1 for SGR, Growth in metabolic body weight and growth in gram per day using sole of around 150 grams. In this experiment there was no reference included in terms of a commercial pellet or worm, but when comparing this figure with the known commercial data the results are better.

Example 1 - Identification/screening of intrinsic factors of the ragworm (Nereis virens) responsible for the excellent growth of sole (Solea solea)

Experimental design, diets and preparation: The experimental set-up contained five diets: (1 ) freshly chopped ragworm, (2) boiled chopped ragworm, (3)

frozen/thawed worm, (4) a pellet flavoured with worm extract, and (5) a commercial pellet as a negative control (Weanex, 3mnn, Biomar. Diets were tested in triplicate, except fresh worm treatment, which was tested in 4 tanks, giving 16 experimental units, 15 fish per tank, for a period of 64 - 67 days due to sampling of one block per day. Tanks within treatments were divided over 4 different blocks. Boiled worms were cooked for three minutes before being chopped. Frozen worms were stored for 24 hours at -20°C and thawed before being fed.

Extracts for flavouring pellets were produced by homogenizing cooled ragworms (Topsy Baits B.V.) using a blender. Homogenates were centrifuged (20 min,

4500rpm) to remove solids. Obtained supernatants were diluted (8g extract per 50ml) using an isotonic salt solution for proper spraying and ensuring even

distribution of extracts on pellets. Diets were produced by spraying a commercial sole pellet with diluted extracts using 8g extract per kg of feed. Treated pellets were dried for 24h at room temperature before storage.

Fish housing and husbandry: The experiment was conducted at IMARES department of aquaculture in Yerseke, the Netherlands. The experiment consisted of a 19 day acclimatization period, a 57 day growth period and a 64 - 67 day

experimental period, depending on the sampling day due to the necessity to sample per block. Sole, (S. solea, weight ± 70 g obtained from Solea B.V.) were

accommodated in 16 tanks (0.4m ² , 130 liters (I)), integrated in a flow through system using sand filtered seawater with a flow of 2.5 I. min ^"1 .tank ^"1 . Before allocated to the experimental units, fish were preventively treated. This consisted of one treatment of 12 hours using formalin (0.1 g.l ^"1 ). Furthermore, an extra group of 10 fish was sacrificed at the start of the experimental period or initial sampling. Environmental parameters were: photoperiod 12L:12D; light intensity 1 1 -15 lux (i.e. low);

temperature ±1 1 -18°C; O ₂ 6 -1 1 mg.l-1 ; pH 7.5 - 8.0; and salinity 25-30 ppt.

Temperature and O2 were measured daily. Flow, pH, TAN, NO2- were measured weekly. Mortality, date and weight of dead animals were recorded.

Feeding: Fish were fed by hand twice a day (8:30 and 16:30) and amount was recorded accordingly. Feeding level was equal for all tanks and adapted daily towards the highest feeding level possible. During acclimatization (19 days) fish were fed the commercial diet. Worms were chopped using a knife and chopping board and sieved for 1 min to drain excess fluids prior to weighing.

Feed intake and feed conversion: Feed intake in g.d ^"1 , as sum of feed given in the morning and afternoon, was estimated by number of pellets given minus number of pellets recovered, multiplied by the average pellet weight or weight worms given, minus the weight worms recovered. As dry matter content varied between

experimental diets, feed intake was calculated on dry matter basis. Feed conversion on dry matter (FCRDM) was calculated applying the following formula 1 :

Formula 1 ) FCRDM = (Fl ^* dry matter experimental diet)/growth.

FCRDM Feed conversion ratio on dry matter base

Fl Feed intake

Growth Increase in bodyweight

Growth: Fish were weighed individually at the start and at the end of the 57-day growth period. Fish were starved for one day prior to weighing. From the individual weight data, average body weight at start and at end (BW ₀ and BW _t ) of the experimental period was calculated per tank, being the experimental unit. From the mean BW ₀ and BW _t , growth rates were calculated per tank expressed in (a) g.d ^"1 , (b) specific growth rate (SGR in %.d ^"1 and in g.kg ^{"0 8} .d ^"1 ). The SGR and the mean growth expressed in growth in metabolic bodyweight (GMBW), units of g.kg ^{"0 8} .d ^"1 during the experimental period were calculated by applying formula 2 and 3 below:

Formula 2) G _M Bw=(BW _(T ).BW ₍ o)) (((BW ₍ t _R BW ₍ o) )° ⁵ /1000° ⁸ )/t

GMBW Growth in metabolic bodyweight in g.kg ^{"0 8} .d ^"1 .

BW(o) Bodyweight at start

BW(t) Bodyweight at end

T Duration of the experiment

Formula 3) SGR = (Ln(BW(t))-LN(BW(0)))/(T) ^* 100 SGR=specific growth rate in %.d

L _n Natural logarithm

BW(0) Bodyweight at start

BW(t) Bodyweight at end

T Duration of the experiment

Sampling: 20 fish at start, 5 fish of every tank at the end of the feeding trial and 10 fish of every tank at the end of the experiment (day 64 - 67) were sacrificed using an overdose of phenoxyethanol (1 :1000) to collect the samples. Upon termination of the feeding trail (day 57) all fish were harvested, weighed and counted. After weighing 5 fish per tank were frozen at -20°C for further analysis if necessary. 10 fish were put back in their experimental unit for another 7-10 days before being sampled for tissue samples (gut and liver) and physiological parameters (haematocrit, plasma). The same time difference (15 hours) was kept between last feeding and sampling of fish for every experimental unit, feeding was split up per tank the evening before sampling and sampling was split up over 4 blocks, with every block having its own sampling day; day 64, 65, 66 and 67 for block 1 , 2, 3 and 4 respectively.

Diets: Worms were sampled daily. Each day adequate quantities of ragworm were taken and stored at -20°C for further analysis. Quantities were based on the average feed intake per treatment from the previous day. Daily samples were pooled per treatment after the end of the experiment to get the average dry matter content and proximate composition. Blood: 10 fish at start and 5 fish of every tank at the end of the experiment were sacrificed to collect blood. Blood was obtained by caudal venous puncture using a heparinized syringe (0.6mm/30mm needle) as soon as fish was anaesthetized. After collection, blood was immediately transferred into labelled tubes and stored on ice until further processing. A little amount was used for determination of haematocrit by using heparinized tubes (0 1 .5mm, L 75mm) and a haematocrit centrifuge (standard rpm during 5 minutes). The remaining part of the sample was centrifuged at (10000 x g, 4°C, 10 min) to collect plasma. Plasma was transferred to clean tubes and stored at -80°C until further analysis. Statistical analysis: Data were analysed using ANOVA to test for block and treatment effects. The model assumes that the residual effects (e) are

independently normal distributed with mean (u) and constant variance σ2.

Model: Yij= u +Ti + Bj + e

with:

u = mean

Ti = effect treatment

Bj = block effect

e = residual error

When significant, means were compared using the LSD multiple comparison post hoc test using a one sided t-test to test hypothesis. For all tests a probability p<0.05 is considered significant.

General aspects of the experiment: Body weight at the start did not differ between treatments. The trial was carried out without problems related to feed intake and health, with exception of one of the four experimental units of fish fed with chopped worms. After 31 days, it appeared that both feed intake (g dm .fish ^"1 .d ^"1 ) and growth in (g.kg ^" ° ⁸ .d ^"1 ) of this particular experimental unit was 9 and 8 times the standard deviation lower for feed intake and growth respectively than the average of the other three experimental units with the same treatment. Feed intake and growth at day 31 for worm tank 1 , 2, 3 and 4 were 0.700, 0.738, 0.755 and 0.466 g dm.fish ^"1 .d ^"1 and 6.16, 6.58, 5.64 and 2.43 g.kg ^{"0 8} .d ^"1 , respectively. As the experimental units were completely independent (flow through) and both feed intake and growth are strong indicators of welfare and health of fish, it was clear that there was something wrong with the latter population and it was decided to discard this tank from the experiment. Mortality during the experimental period in the other tanks was limited to three fish and was not related to treatments. Average values for temperature, oxygen, pH, TAN, NO ² , NO ³ , and flow.tank ^"1 during the trial were 15.7 , 8.9mg.l ^"1 , 8.16, 0.07 mg.l ^"1 , 0.05 mg.l ^"1 , 1 .14 mg.l ^"1 , 2.5 l.min ^"1 , respectively. The average temperature for period 1 and period 2 were 17 ^° C and 14 ^° C respectively. The daily temperature and oxygen levels during the experiment and divided over period 1 and 2 are shown figure 1 . Dry matter content of diets were 89.1 %, 89.6% and 18.0% for the untreated pellet, treated pellet and worm respectively. P values from F test, testing for block effects show no significant block effects.

Feed intake and growth between fresh worm, pellet with extract and untreated pellet: Feed intake on dry matter base does significantly differ between diets

(p=0.014, table 1 ). Sole fed fresh worm showed the highest feed intake with 0.69g dm .fish ^"1 .d ^"1 followed by the treated and untreated pellet with a feed intake of 0.68 and 0.56g dm.fish ^"1 .d ^"1 respectively. Feed intake (Fldm) of sole fed treated pellet was 0.13g dm .fish ^"1 .d ^"1 or 22.4% higher compared to sole fed untreated pellet and did not differ from sole fed fresh worm (see figures 2a and 2b).

Growth expressed in gram per day (g.d ^"1 ) does significantly differ between diets (p=0.02). Sole fed the fresh worm showed the highest growth of 0.7 g.d ^"1 followed by the treated and untreated pellet with a growth of 0.56 and 0.41 g.d ^"1 respectively. Growth of sole fed the treated pellet was 0.15g.d ^"1 or 36.6% higher compared to sole fed the untreated pellet, but still 0.14g.d ^"1 lower compared to sole fed fresh worm (see figure 2b). Growth of sole fed fresh worm compared to sole fed commercial pellet was 0.29g.d ^"1 or 70.7% higher. Growth expressed in g.kg ^{"0 8} .d ^"1 or SGR in %.d ^{" 1} are following an equal pattern . Table 1 below shows body weight at start (BW _s t _ar t), body weight at end (BW _end ), growth, SGR, growth in metabolic body weight, hepatosomatic index (HSI), feed conversion on dry matter base (FCR _dm ) of the different treatments.

Table 1 :

Sem 1.02 0.01 3.31 0.01 0.05 0.04 0.27 0.04 0.07 p value 0.34 0.34 0.01 0.94 0.02 0.03 0.02 0.24 0.09

Means within columns with a common superscript are not significantly different (P<0.05).

Haematocrit and hepatosomatic index: For haematocrit, the effect of

heterogeneity and variance in values of haematocrit are also negligible as values of groups are not close to zero per cent (10.34 - 23.08%). Transformation of

percentages before analysis is not necessary and percentages were used for statistical analysis. Haematocrit does differ between treatments (p=0). Haematocrit values for the sample group t(0) and untreated pellet do not differ (p=0.48).

Haematocrit of sole fed worm is significantly higher than haematocrit of sole fed the untreated pellet (p=0) and treated pellet (p=0). Haematocrit of worm fed fish were on average 83-87% higher compared to the values of pellet fed fish. Haematocrit between treated and untreated pellets differed only 2.6%, which was not significant (p=0.76, see figure 3).

The effect of heterogeneity and variance in values of HSI are negligible as values only vary little between groups (1 .34 -1 .71 ). Again transformation of percentages before analysis in not necessary and percentages were used for statistical analysis. The hepatosomatic index (HSI) did not differ between treatments (p=0.24, table 1 ). Table 2 below shows the results of HSI, Htc, growth, feed intake, iron intake and expression of casp3, Hsp70, transferrin and hepcidin, for sole fed an untreated pellet, a flavoured commercial pellet, fresh ragworm and boiled worm. Preliminary results, regarding the expression of genes involved in iron homeostasis and immune response (see table 2 below), show that an anaemia caused by inflammation can be ruled out and that the anaemia observed in sole fed commercial pellets is a form of nutritional anaemia.

Table 2: Pellet 0 0 0 0 0 #NA #NA #NA #NA

Treated pellet +4.5 +2.51 +27.4 +22.4 +19.4 Normal Normal Low Low

Fresh worm -9.5 +87.3 +53.8 +24.7 +97.0 Normal Normal Normal Normal

Boiled worm -3.2 +49.5 +22.4 -26.8 +17.2 #NA #NA #NA #NA

Significant No Yes Yes Yes Yes No No Yes Yes

^a Hepatosomatic Index (HSI), "Haematocrit (Htc), 'Growth g.kg ^{0 8} .cT ¹ , ^d Feed intake g.dm ^'1 .fish ^'1 .cT ¹ , ^e lron intake mg.fish ^'1 .d ^~1 and, 'Expression of Casp3, Hsp70, transferrin & hepcidin compared to

housekeeper beta actin (Kals & Palstra, unpublished data). All data except expression data are

shown as % of difference compared to the untreated pellet.

Haematocrit, feed intake and growth in relation to temperature: The current trial can be divided in two periods (see figure 1 ). Period P1 has an average temperature of 17.2 °C and P2 has an average temperature of 14.5 °C. Due to these different periods relations between haematocrit, temperature, feed intake and growth could be made which are shown and briefly discussed below.

Feed intake in relation to temperature: Use of worm extract equalizes the

difference in long term feed intake on dry matter basis between live feed and dry feed in sole (see figure 2a) when measured over the whole 57 day experimental period.

As feed intake over the whole trial is not different between fresh worm and the

treated pellet, but haematocrit levels in fish fed worm are almost 85% higher and do not differ between the treated and untreated pellet, it can be concluded that feed

intake has no effect on haematocrit levels (see figure 3 and table 2). Fl does rise

with increasing temperature, especially for worm (see figure 4). The parabolic

relation between feed intake and temperature of the treated pellet (feed intake

decrease at temperature higher than 15.5 °C can be explained by a limited scope for growth during higher temperatures as haematocrit levels in worm fed are significantly higher compared to the pellet fed fish (see figures 3 and 4).

Growth in relation to diet and temperature: Figure 4 clearly shows the advantage of feeding worm in P1 (average temperature of 17.2°C) completely disappears in period P2 (average temperature of 14.5°C). Combining this finding with the temperatures, it can be stated that the "worm effect" disappears at lower

temperatures. The relationship between haematocrit and growth is shown in figure 8. What is clear is that low haematocrit levels (anaemia) in fish compromises O2 transport and metabolic performance. Therefore it is likely that the need for O ₂ plays a pivotal role in setting the lower limit for haematocrit in normocythemic fish. Virtually nothing is known about the effects of physiological and environmental influences on haematocrit regulation, or the links between haematocrit, haemoglobin, nutrition, and growth (Gallaugher and Farrell (1998)).

The data from trials where the nutritional treatment is controlled show a clear relationship between nutritional status, haematocrit, oxygen carrying capacity and the metabolic scope for growth. It is also clear the visibility is highly dependent on the environment as for example the temperature of the culture water. Haematocrit values and therefore haemoglobin (Hb) values of sole fed fresh worm are much higher than sole fed pellets. This affects their growth potential or "scope for growth" at higher temperatures (see figure 4 and 5). This is also reflected in the relation between feed intake and culture temperature for sole fed pellet and fresh worm (see figure 4). Optimal feed and optimal water temperature interact.

The following conclusions are drawn from the results of example 1 :

There is a positive effect of haematocrit on growth especially in period 1 (P1 ) A low haematocrit (anaemia) is limiting feed intake & growth at higher temperatures.

· The anaemia observed in sole fed commercial pellet is a form of nutritional anaemia.

Temperature is critical at about 16 °C.

Worm increases haematocrit and thereby "scope for growth" at higher temperatures.

· Boiling has a negative effect on the properties of fresh worm as shown by lower growth and lower haematocrit of fish fed with boiled worm compared to fish fed fresh worm. Example 2 - The "worm factor"; a strong dose response relation between worm meal inclusion level, haematocrit and the growth of sole (Solea solea).

Fresh worm has no added value. Experimental design, diets and preparation: The experimental set-up contained seven diets; a diet with 0, 10, 15, 50 or 75% of worm meal, fresh worm as a positive control, which is referred to as the diet with 100% worm meal inclusion level, and a negative control (Weanex3mm, Biomar). Diets were tested in triplicate, giving 21 experimental units, 15 fish.tank ^"1 , for a period of 42 days for growth and 45, 46 and 47 days for other parameters due to the necessity of sampling of one block per day. Ragworms were provided three times a week by Topsy Baits B.V. The experimental recipes are shown in table 3. Diets were prepared by cold extrusion.

Table 3:

Binder 3 ^* 1 .00 1 .00 1 .00 1 .00 1 .00

Mineral and vitamin 1 .00 1 .00 1 .00 1 .00 1 .00 premix ^**

Choline chloride 0.20 0.20 0.20 0.20 0.20 vitamin C 0.082 0.082 0.082 0.082 0.082 vitamin E 0.076 0.076 0.076 0.076 0.076

Betaine HCI 0.05 0.05 0.05 0.05 0.05

Check 100.00 100.00 100.00 100.00 100.00

Extra BHT ^*** 100

Cal propionate ^**** 1000

Calculated (macro G B C D E nutritional composition)

DS (g/kg) 918.6 921 .7 926.3 934.1 941 .9

RAS (g/kg) 142.5 144.5 147.4 152.4 157.4

RE (g/kg) 533.2 533.2 533.2 533.1 533.1

RVET (g/kg) 134.8 134.8 134.8 134.8 134.8

RC (g/kg) 27.1 28.2 30.0 32.9 35.8

OK (g/kg) 76.5 76.5 76.5 76.5 76.5

GE (MJ/Kg) 19.6 19.7 19.7 19.7 19.8

CP/GE 27.2 27.1 27.1 27.0 26.9

^** Composition according to pre-mix Borges et al. (2009) Dietary lipid level affects growth performance and nutrient utilisation of Senegalese sole (Solea senegalensis) juveniles British Journal of Nutrition 102: 1007 - 1014.

* ^** Antioxidant for prolonging the storage time.

* ^*** Added as an anti-fungal compound.

Composition diet B, C, D, E and G are equal in macronutrients, amino acids, calcium and phosphates Table 4: the vitamin and mineral premix of Borges et al (2009): Vitamin and mineral premix

(Borges et al 2009)

(per kg

Vitamins diet)

mg (8000

Vitannine A (retinyl acetate) 2.4 IU)

mg (1700

Vitannine D3 (cholecalciferol) 0.04 IU)

Vitannine K3 (menadione sodium

bisulfite) 10 mg

Thiamin Vit B1 8 mg

Riboflavin Vit B2 20 mg

Nicotinic acid Vit B3 70 mg

Pantothenic acid Vit B5 30 mg

Pyridoxin Vit B6 10 mg

Biotin Vit B7 0.7 mg

Inositol Vit B8 300 mg

Folic acid (B9/B1 1 ) 6 mg

Vit B12 (cobalamine) 0.02 mg

Vit E 300 mg

Vit C ^* 500 mg

Sum 1 .25716 g

Minerals

MnO 20 mg

Kl 1 .5 mg

Cu(SO4) 5 mg

Co(SO4) 0.1 mg

Mg(SO4) 500 mg

ZnO 30 mg

NaSe 0.3 mg Fe(SO4) 60 mg

CaCO3 2150 mg

DiCa(PO4) 5000 mg

KCI 1000 Mg

Fish housing and husbandry: The experiment was conducted at IMARES department of aquaculture in Yerseke, the Netherlands. The experiment consisted of a 14 day acclimatization period, a 42 day growth period. Sole, (S. solea, weight ± 70 g obtained from Solea B.V.) were accommodated in 21 tanks (0.4m ² , 130 I), integrated in a flow through system using sand filtered seawater with a flow of 7 l.min ^{" 1} .tank ^"1 . Before allocated to the experimental units fish were preventively treated. The applied protocol consisted of one treatment of 12 hours using formalin (0.1 g.l ^"1 ). Furthermore an extra group of 10 fish was sacrificed at the start of the experimental period or initial sampling. Environmental parameters for photoperiod, light intensity, temperature, O ₂ , pH and salinity were 12L:12D, low;1 1 -15 lux, ±18°C, 6-8mg.l ^"1 , 7.5- 8.0, and 25-30ppt, respectively. Temperature and O ₂ were measured daily. Flow, pH, TAN, NO2 were measured weekly. Mortality, date and weight of dead animals were recorded.

Feeding: Fish were fed by hand twice a day (8:30 and 16:30). Feeding level was equal for all tanks and adapted daily towards the highest feeding level possible. After an hour uneaten feed was removed from tanks randomly after every feeding period. During acclimatization (14 days) fish were fed the commercial diet. Worms were chopped using a knife and chopping board and sieved for 1 min to drain excess fluids prior to weighing.

Growth: Fish were weighed individually at the start and at the end of the 57-day growth period. Fish were starved for one day prior to weighing. From the individual weight data, average body weight at start and at end (BW ₀ and BW _t ) of the

experimental period was calculated per tank, being the experimental unit. From the mean BW ₀ and BW _t growth rates were calculated per tank expressed in g.d ^"1 , SGR in %.d ^"1 and g.kg ^{"0 8} .d ^"1 . The specific growth rate and the mean growth expressed in GMBW during the experimental period were calculated by applying formula 2 and formula 3:

Sampling: 20 fish at start, 5 fish of every tank at the end of the feeding trial and 10 fish of every tank at the end of the experiment (day 45-47) were sacrificed using an overdose of phenoxy ethanol (1 :1000) to collect the samples. Upon termination of the feeding trail (day 42) all fish were harvested, weighed and counted. After weighing 5 fish per tank were frozen at -20°C for further analysis if necessary. 10 fish were put back in their experimental unit for another 4-6 days before being sampled for tissue samples (gut and liver) and physiological parameters

(haematocrit, plasma). The same time difference (15 hours) was kept between last feeding and sampling of fish for every experimental unit. Feeding was split up per tank the evening before sampling and sampling was split up over 4 blocks, with every block having its own sampling day; day 45, 46 and 47 for block 1 , 2 and 3

respectively.

Diets: Worms were sampled daily. Each day adequate quantities of ragworm were taken and stored at -20°C for further analysis. Quantities were based on the average feed intake per treatment from the previous day. Daily samples were pooled per treatment after the end of the experiment to get the average dry matter content and proximate composition.

Blood: 10 fish at start and 5 fish of every tank at the end of the experiment were sacrificed to collect blood. Blood was obtained by caudal venous puncture using a heparinized syringe (0.6mm/30mm needle) as soon as fish was anaesthetized. After collection, blood was immediately transferred into labelled tubes and stored on ice until further processing. A little amount was used for determination of haematocrit by using heparinized tubes (0 1 ,5mm, L 75mm) and a haematocrit centrifuge (standard rpm during 5 minutes). The remaining part of the sample was centrifuged at (10000 x g, 4°C, 10 min) to collect plasma. Plasma was transferred to clean tubes and stored at -80°C until further analysis.

Statistical analysis: See Example 1 . Results -general aspects: The trial was carried out without problems related to feed intake and health. Mortality during the experimental period was limited to five fish and was not related to treatments. Average values for temperature, oxygen, pH, TAN, NO ₂ , NO ₃ , and flow.tank ^"1 during the trial were 19.2 ^° C, 7.3 mg.l ^"1 , 7.6, 0.02 mg.l ^{" 1} , 0.07 mg.l ^"1 , 4.3 mg.l ^"1 , 7 l.min ^"1 , respectively.

Growth: There is a positive relationship between the worm meal inclusion level and growth. The inclusion level of 100% worm meal is the fresh worm. Therefore fresh worm has no added value as otherwise the data point of growth of fresh worm would be laying above the regression line. The growth with the conventional pellet was 4.58 g.kg ^{"0 8} .d ^"1 comparable with 50% worm meal inclusion.

In conclusion from this trial:

There is a positive relationship between worm meal inclusion level,

haematocrit and growth.

There is a clear "worm effect".

There is no added effect of "live" worm

There is no effect of iron intake on haematocrit. Iron absorption is not known There is a strong relationship between calculated heme content of diet, haematocrit and growth.

Example 3 - Sole fed ragworm recover from their anaemia in 15 - 21 days

Experimental design, diets and preparation: The experiment consists of 13 different treatments, 10 fish.tank ^"1 ; of which12 with fish fed freshly chopped ragworm using a regression set up of which the sample times were spread over 26 days (day 0, 2, 4, 6, 8, 10, 13, 16, 19, 21 , 23 and 26) and one with fish fed a commercial diet (Pelleted feed, 3mm, from a regular producer) as a control group which got sampled at day 26. Treatments were allocated randomly over available tanks. Ragworms were provided three times a week by Topsy Baits B.V.

Fish housing and husbandry: The experiment was conducted at IMARES department of aquaculture in Yerseke, the Netherlands. The experiment consisted of a 9 day acclimatization period, a 26 day experimental period. Sole, (S. solea, weight ± 70 g, IMARES.) were accommodated in 21 tanks (0.4m ² , 130 I), integrated in a flow through system using sand filtered seawater with a flow of 4 l.min ^"1 .tank ^"1 . Before allocated to the experimental units fish were preventively treated. The applied protocol consisted of one treatment of 12 hours using formalin (0.1 g.l ^"1 ). Furthermore an extra group of 20 fish was sacrificed at the start of the experimental period or initial sampling. Environmental parameters for photoperiod, light intensity, temperature, O ₂ , pH and salinity were 12L:12D, low;1 1 -15 lux, ±18°C, 6-8mg.l ^"1 , 7.5 - 8.0, and 25-30ppt, respectively. Temperature and O2 were measured daily. Flow, pH, TAN, NO ₂ ^" were measured weekly. Mortality, date and weight of dead animals were recorded.

Feeding: Fish were fed by hand twice a day (8:30 and 16:30) and uneaten feed was removed randomly after an hour. Feeding level was equal for all tanks and adapted daily towards the highest feeding level possible. During acclimatization (9 days) fish were fed the commercial diet. Worms were chopped using a knife and chopping board and sieved for 1 min to drain excess fluids prior to weighing.

Sampling: 20 fish at start, 10 fish of every tank during the trail were sacrificed using an overdose of phenoxy ethanol (1 :1000) to collect the samples. After weighing fish were sampled for tissue samples (gut and liver) and physiological parameters (haematocrit, plasma). Tissue samples were frozen at -80°C and or stored in

RNAIater until further analysis.

Blood: 20 fish at start and 10 fish of every tank were sacrificed to collect Blood. Blood was obtained by caudal venous puncture using a heparinized syringe

(0.6mm/30mm needle) as soon as fish was anaesthetized. After collection, blood was immediately transferred into labelled tubes and stored on ice until further processing. Little amounts were used for determination of haematocrit and haemoglobin. Haematocrit was determined by using capillaries and a haematocrit centrifuge (standard rpm during 5 minutes). Haemoglobin content of blood was determined using the method described by Kampen and Zijlstra (1961 )

"Standardization of Hemoglobinometry II. The Hemiglobincyanide Method" Clinica Chimica Acta, 6 (1961 ) 538-544. The remaining part of the sample was centrifuged at (10000 x g, 4°C, 10 min) to collect plasma. Plasma was transferred to clean tubes and stored at -80°C until further analysis.

Statistical analysis: Data was analysed using best fit regression models.

Results - General aspects of the experiment: The trial was carried out without problems related to feed intake and health. No mortality occurred. Average values for temperature, oxygen, pH, TAN, NO ₂ , NO3, and flow.tank ^"1 during the trial were 19.2 , 8.3 mg.r ¹ , 7.9, 0 mg.l ^"1 , 0.01 mg.l ^"1 , 2.1 mg.l ^"1 , 4.1 I. min ^"1 , respectively. Sole do recover from their anaemic state in 15-21 days as shown in the changes of both haematocrit as haemoglobin levels. The control group fed with the commercial pellet kept both their haematocrit as haemoglobin levels comparable as levels at the start of the trial t(0). As shown in figure 16, the relationship between haematocrit and haemoglobin is very high and comparable to relation found in literature (see

Gallaugher and Farrell, (1998)).

Example 4 - The effect of mussel (Mytilus edulis) is comparable to the effect of ragworm (Nereis virens) on alleviation of anaemic growth suppression in Dover sole (Solea solea)

Growth of sole fed raw mussel meat (RMM) was found to be comparable to that of sole fed ragworm. Sole were fed three different diets. (1 ) a commercial pellet, (2) RMM and (3) ragworm. The feeding strategy was restricted, i.e. the feed intake was 0.54g.dm.fish ^"1 .d ^"1 and equal for all treatments. The Htc and Hb values of sole fed RMM and worm increased from 33% to 45% and 41 % to 79% within 23 days, respectively; both being significantly higher compared to sole fed pellets. The Htc level of sole fed pellets (13.07%) is significantly lower compared to the Htc level of sole fed RMM (17.43%) or worm (18.95%). Htc levels of sole fed worm or RMM are not significantly different. The Hb levels of sole fed pellets (18.92 g.l ^"1 ) is significantly lower compared to the Hb level of sole fed RMM (26.70 g.l ^"1 ) or worm (33.72 g.l ^"1 ). However, the Hb level of sole fed worm is higher than that of sole fed RMM. Feeding sole RMM does increase Htc and Hb to comparable levels as with ragworm. In this regard mussel and ragworm are considered substantially equivalent in the alleviation of anaemic growth suppression in Dover sole. The experimental set-up contained three diets; raw mussel meat (RMM), fresh worm and a commercial diet (extruded, 3mm, from a regular producer). The diets were tested in triplicate, giving 9 experimental units with 15 fish.tank ^"1 for a period of 23 days. Mussels were collected at the Oosterschelde, the Netherlands. The RMM was separated from their shell, all at the same day, and frozen at -80°C until needed for feeding. The frozen mussel meat was thawed slowly in a fridge the day before feeding. Ragworms were provided three times a week by Topsy Baits B.V, The Netherlands A 7-day acclimatization and a 23-day experimental period was used for Sole of weight ± 220g. The fish were distributed within 9 tanks (0.4m ² , 1301) having integrated flow through system, using sand filtered seawater with a flow of 4 l.min ^{" 1} .tank ^"1 . Before the experiment, fish were preventively treated with formalin (O.lg.l ^"1 , duration 12 hours). Environmental parameters for photoperiod, light intensity, temperature, O ₂ , pH and salinity were 12L:12D, 1 1 -15 lux, ±18°C, 6-8mg.l ^"1 , 7.5-8.0, and 25-30ppt, respectively. Temperature and O2 were measured daily. Flow, pH, TAN, NO ₂ ^" were measured weekly. Mortality, date and weight of dead animals were recorded. The feeding of fish, diets, sampling, analysis, measurements, blood measurements and calculations were carried out substantially as described in Example 1 . 30 fish at the start and 10 fish of every tank at the end of the experiment, were sacrificed then measurements as per example 1 No mortality occurred and the experiment was carried out without problems. The average water quality parameters for oxygen, temperature, pH, ammonia, nitrite and nitrate were 16.9, 8.3, 8.1 , 0.3, 0.0, 4.5 respectively and stayed within range. The hepatosomatic index (HSI), the haematocrit (Htc) and haemoglobin (Hb) values at the start of the experiment were 1 .13%, 12.48% and 19.46 g.l ^"1 , respectively.

No feed refusal or spillage was observed during the experimental period. Therefore, the feed intake was 0.54 g.dm.fish.d ^"1 and identical for all diets. The daily crude protein (CP), iron (Fe), copper (Cu) and vitamin B12 intake for sole fed the different diets are shown in table 5. Table 5: Average daily feed intake (Fl), nitrogen intake (Nl), Iron (Fel) and vitamin B ₂ (B-| ₂ I) intake for the different dietary treatments.

Both the haematocrit (P=0) and the haemoglobin levels (P=0) are different between treatments (see figure 18). The haematocrit level of sole fed the commercial pellet is lower compared to the level of sole fed mussel (P=0) or worm (P=0). The

haematocrit levels of sole fed worm and mussel are not different (P=0.131 ). The haemoglobin level of sole fed the commercial pellet is lower compared to the levels of sole fed mussel (P=0.001 ) or worm (P=0). However, the haemoglobin level of sole fed worm is higher than that of sole fed mussel (P=0.002). The hepatosomatic index did not differ between treatments (P=0.991 ). Example 5 - Heme iron in combination with high levels of B s responsible for alleviation of anaemic growth suppression in Sole

The experimental set-up contained nine diets (see table 6); 4 diets with haemoglobin as iron source, a diet with Fe(SO ) as an inorganic iron source, a diet with iron proteinate (JH Biotech, Inc), a diet with iron chelate (JH Biotech, Inc), a positive control (worm, N. Virens) and a commercial pellet. Diets enriched with haemoglobin were each different in a combination of high vs. low level of vitamin B ₂ and taurine (table 4). Two levels of taurine are included in the experiment as taurine is essential for efficient nutrient digestion. "High" vitamin B12 and taurine levels were 1 .9 mg.kg.ds ^"1 and 7.6 g.kg.ds ^"1 and equal to levels analysed in worm (Nutrilab, Boxmeer). "Low" vitamin B ₂ and taurine levels were 0.34 mg.kg.ds ^"1 and 3.5 g.kg.ds ^"1 and equal as levels analysed in the commercial pellet (Nutrilab, Boxmeer).

Diets were tested in triplicate, giving 27 experimental units, 15 fish.tank ^"1 , for a period of 23 days. Diets were prepared using cold extrusion. Table 7 shows the content of the diets.

Table 6:

D Haemoglobin ⁴ 7.6 ⁹ 0.34° Organic iron, low vitamin. Bi2, high taurine

E Haemoglobin ⁴ 3.5° 1 .90 ^y Organic iron, high vitamin. B ₁₂ , low taurine.

F Haemoglobin ⁴ 3.5° 0.34° Organic iron, low vitamin. B ₁₂ , low taurine

J Unknown 3.5° 0.34° Commercial #NA

pellet*

Live ragworms, Commercial diet. ³ Multiple types of heme as present in ragworm

(7V. Virens), ⁴ Hb (haemoglobin from Bovine, Sigma-Aldrich). ⁵ Buffermin® IFN-6-26- 150 JH Biotech, Inc, ⁶ Buffermin® Methionates JH Biotech, Inc. ⁸ Levels as analysed in commercial pellet, ⁹ Levels as analysed in fresh worm. Composition of diet B, C, D, E, F, G and H are equal in macronutrients, amino acids, calcium and phosphates.

The feeding of fish, diets, sampling, analysis, measurements, blood measurements and calculations were carried out substantially as described in Examples 1 to 3.

Table 7: Diet formulations, inclusion level of different test products and calculated composition.

Cholesterol ⁰ 0.30 0.30 0.30 0.30 0.30 0.30 0.30

Premix (Iron &

Bi2 free)† 2.00 2.00 2.00 2.00 2.00 2.00 2.00

Binder 1 ^p 2.00 2.00 2.00 2.00 2.00 2.00 2.00

Binder 2 ^q 2.00 2.00 2.00 2.00 2.00 2.00 2.00

Zout ^r 2.00 2.00 2.00 2.00 2.00 2.00 2.00

MCP ^S 1.50 1.50 1.50 1.50 1.50 1.50 1.50

Binder 3* 1.00 1.00 1.00 1.00 1.00 1.00 1.00

TMG ^U 1.00 1.00 1.00 1.00 1.00 1.00 1.00

Lucantin Pink

10% ^v 0.50 0.50 0.50 0.50 0.50 0.50 0.50

Yttrium oxide ^w 0.02 0.02 0.02 0.02 0.02 0.02 0.02

Test

ingredients

taurine ¹ 0.626 0.626 0.626 0.210 0.210 0.626 0.626

Vitamin Bi ₂ ² 0.190 0.190 0.028 0.190 0.028 0.190 0.190

Bovine

Hemoglobin ³ 6.200 6.200 6.200 6.200

Iron sulfate

hydrate

(20%) ⁴ 0.101

Iron

proteinate ⁵ 0.127

Iron

methionate ⁶ 0.106

100.0 100.0

Check 100.00 0 100.00 100.00 0 100.00 100.00

Calculated

composition

DS (g.kg- ¹ ) 934 938 938 937 937 934 934

RAS (g.kg ¹ ) 87 84 84 84 84 86 86 RE (g.kg ^"1) 599 600 600 599 599 599 598

RVET (g.kg ^"1 ) 108 108 108 108 108 108 108

RC (g.kg ^"1 ) 26 26 26 26 26 26 26

OK (g.kg ^"1 ) 76 82 83 83 84 76 76

GE (MJ.Kg- ¹ ) 19.90 19.94 19.97 19.96 19.98 19.89 19.89

CP/GE 30.08 30.07 30.03 30.03 29.99 30.09 30.07

Cholesterol

(g.kg ^"1 ) 3.46 3.46 3.46 3.46 3.46 3.46 3.46 taurine (g.kg ^"1 ) 7.27 7.27 7.27 3.13 3.13 7.27 7.27

Vitamin Bi ₂

(ug.kg ^"1 ) 1907 1907 319 1907 319 1907 1907

Iron (mg.kg-

289 290 290 290 290 290 290

Analysed

composition

DS (g.kg ^"1 ) 909.0 913.0 916.0 917.0 913.0 920.0 917.0

RAS (g.kg ¹ ) 9.4 9.2 9.2 9.1 9.3 9.3 9.4

RE (g.kg ^"1 ) 661 .0 655.0 658.0 655.0 658.0 669.0 661 .0

RVET (g.kg ^"1 ) 130.0 122.0 125.0 123.0 125.0 130.0 127.0

RC (g.kg ^"1 ) 6.6 6.6 5.6 5.6 7.7 5.4 5.5

OK (g.kg ^"1 ) 109.0 125.0 120.0 125.0 1 16.0 102.0 1 13.0

GE (MJ/Kg) 22.8 22.7 22.7 22.7 22.7 22.9 22.8

CP/GE 28.9 28.9 28.9 28.9 29.0 29.2 29.0

Vitamin B12 1254 1 150 175 1330 177 959 1075

(ug.kg ^"1 )

Iron (mg.kg ^" 355 323 316 320 340 299 349

1)TT

aAcid casein 30/60 mesh, Lactalis, Bourgbarre, France, ^b Roquette Freres, Lestrem, France, °Soycomil R ADM Eurpoort BV, ^d Cargill, Bergen op Zoom, The Netherlands, ^e Ajinomoto Parijs France, ^f Sumitomo Tokyo Japan, ⁹ Evonik, Hanau, Germany, ^h Cgel C3401 Cargill Bergen op Zoom the Netherlands, 'Danish LT fishmeal. Type LT

(Triple Nine Fish Protein Esbjerg, DK), 'Coppens International The Netherlands, ^k Sopropeche Boulogne sur Mer, France, 'Sopropeche Boulogne sur Mer, France, ^m Gluvital 21000 Cargill Bergen op Zoom The Netherlands, "Nutripur G Cargill, Hamburg, Germany, "Cholesterol SF van Dishman Netherlands BV, Veenendaal, ^p Binder1, ^q Binder2 and 'Binder3; Ingredients are not specified because of confidentiality reasons of on-going research. ^r Animalfeed salt van Kloek zout The Netherlands, ^s Tessenderlo Chemie Belgium, "Betafin van Danisco Animal Nutrition Marlborough UK, ^V BASF, Ludwigshafen Germany, ^w Sigma Aldrich, USA, Test ingredients, ² vitamin B ₁₂ (cyanocobalamin) Rosun, Krimpen a/d Ijssel The

Netherlands, ³ Bovine Hemoglobin (CAS 9008-02-0) with batchnummer 081M7002V (iron 0.31% & dm 97%), Sigma-Aldrich, USA , ⁴ Ferromel 30: Melspring International, Velp, NL, ⁵ Buffermin® Iron proteinate (I FN 6-26-150) JH Biotech, Inc. Ventura, CA 93006 USA, ⁶ Buffermin® Methionate, J H Biotech, Inc. Ventura, CA 93006 USA.. ^† Vitamins (mg or IU kg-1 diet): vitamin A (retinyl acetate); 2.7 mg, 9000 IU; vitamin D3 (cholecalciferol), 0.04 mg 1700 IU; vitamin K3 ((menadione sodium bisulfite), 10 mg; vitamin B1 (thiamine), 10 mg; vitamin B2 (riboflavin), 20 mg; vitamin B6

(pyridoxine hydrochloride), 12 mg; folic acid, 10 mg; biotin, 0.7 mg; inositol, 800 mg; niacin, 70 mg; pantothenic acid, 30 mg, choline chloride, 1500 mg; vitamin C, 500 mg; vitamin E, 300 mg; Minerals (g or mg kg ^'1 diet): Mn (manganese sulphate), 20 mg; I (potassium iodide), 1.5 mg; Cu (copper sulphate), 5mg; Co (cobalt sulphate), 2 mg; Cr (chromium sulphate), 2.6 mg; Mg (magnesium sulphate), 500mg; Zn (zinc sulphate) 60 mg; Se (sodium selenite) 0.3 mg; BHT (E300-321) 100 mg; Calcium propionate 1000 mg. ^†† / ^' n all recipes 64% of the iron is supplied by the added sources, the rest is iron from ingredients.

No mortality occurred and the experiment was carried out without problems. No feed refusal and spillage was observed during the experimental period. Therefore feed intake (0.54 g.dm.fish ^"1 .d ^"1 ) and nitrogen intake (0.35-0.36 g. dm. fish ^"1 .d ^"1 ) were identical for all diets. During the trial, average water quality parameters like oxygen, pH, ammonia, nitrite and nitrate stayed within range.

There was no difference of the haematocrit (Htc), haemoglobin (Hb) levels and the hepatosomatic index (HIS) between sole fed the diets with the different iron sources (see table 8). Therefore the source of iron has no effect on the haematocrit, haemoglobin levels and hepatosomatic index of Dover sole.

Table 8:

Htc does differ between the different diets (p=0.013, table 7). Htc is highest in the diet containing heme as iron source, high Bi ₂ and low taurine. The Haemoglobin levels follow the same pattern. The hepatosomatic index (HIS) does not differ between diets (see table 9).

There is a clear effect of B-| ₂ on the Htc of sole (p=0.013. table 9). There is also interaction between the level of B-i ₂ and level of taurine (p=0.049). A high taurine level seems to have a negative effect. The combination of heme, high B ₂ and low taurine gives the best results.

Table 9:

abc Means with different superscripts are significan tly different.

The findings are that Vitamin B ₂ has a positive effect on haematocrit in sole fed diets formulated with heme. There is an interaction between between B ₂ and taurine whereby high levels of taurine reduce the anaemia alleviating effects. Therefore a combination of heme with high B-| ₂ and low taurine appears to gives better results. Further, the presence of heme in the diet gives significantly increased copper adsorption compared to diet without heme.