USE OF A COMPOSITION BASED ON FISH SOFT ROE FIELD OF THE INVENTION The present invention relates to a novel use of a composition based on fresh or conserved fish soft roe useful in manufacturing a preparation enhancing the serum immunoglobulin level, especially the IgG level, and thereby strengthens the immune system. Further, the present invention relates to a novel use of a composition based on fresh or conserved fish soft roe useful in manufacturing a preparation that stimulates growth and improves an individual's general state of health.

Background of the invention The health and nutritional status of children in developing countries are far from being satisfactory. Malnutrition in all its forms and manifestations remains a global challenge, and a main public health problem in several countries in both the developed and the developing countries. The magnitude and severity of its consequence is alarming and devastating to better living conditions, living standard and economy in Africa, Asia and Latin America.

In the industrialised countries, primarily malnutrition is seen mainly among young children, the elderly who live alone and subjects addicted to drugs and alcohols.

It is well-known that malnutrition affects the immune system and cause immunodeficiency. Although considerably progress in the knowledge of the immune system and its function has been made over the last years, still little is known about the interaction and mechanism between nutrition and the immune system. Developing and understanding of nutritional needs and the role of nutrition in immune function are essential in prevention and treatment of nutrition-related immuno-compromise (Krenitsky, 1996, Nutrition and t} e immune system, AACN-Clin-Issues, 7 (3) : 359-69). Accordingly, there is a considerably need of novel supplementary diets to combat malnutrition and strengthen the immune system.

It is assumed that a good response to vaccines and other drugs depends on a functioning immune system. Also, in addition to malnutrition, there are several conditions and diseases where a strong immune system is of great

importance as regards to decrease the development and increase the chance of recovery of the condition or disease. Illustrative, but not restrictive, conditions or diseases like AIDS, different types of malignant and/or benign tumours, bacterial or viral infections, graft rejection can be mentioned, respectively. Accordingly, there is a need of supplementary diets and methods to improve the response to vaccines and to strengthen the immune system in subjects suffering of or at risk of developing conditions or diseases as mentioned herein above.

It is also a need of supplementary diets to strengthen the immune system in individuals in need of a strong immune system due to e. g. extreme stress, as top athletes.

Over the last years, a wide variety of compositions based on fish soft roe have been reported, especially composition intended for dietary supplementation of infants, see e. g. European Patent Application No. EP- A1-729710, Russian Patent No. RU 2091073, US Patent No 5,552, 889, respectively. As an example, EP 0 729 710 Al disclose a nutritive compositions for infants containing milt grinds and/or milt fraction.

However, up to date, nothing is known about how to enhance the serum immunoglobulin level, thereby e. g. providing improved state of nutrition and response to vaccines and other medications. Also, the nutritive state of an infant and its ability to respond to dietary supplement cannot be directly transferred to a child or an adult due to the fact that a lot of biochemical processes and mechanisms are not fully developed in an infant compared with a'child or an adult.

Neuman et al, (1972), investigated the quantitative levels of immunoglobulins in severely malnourished, moderately malnourished and control Ghanaian children compared with US age-matched standards. The levels of all the Igs were higher in the severely malnourished children than in the moderately malnourished and control children. Compared to the US age-matched standards, even the control Ghanaian children had higher levels of all the Igs. Similar studies reported in the literature gave comparably higher levels of immunoglobulins in malnourished children than in normal children (Bell et al, 1976,"Serum and small intestinal immunoglobulin levels in undernourished children, Am. J. Clin. Nutrl, 29 : 392-97 and Suskind et al, 1976, Immunoglobulins and antibody

response in children with protein-calorie-malnutrition", Am. J. Clin.

Nutr., 29 : 836-41).

Revillard and Cozon, 1990,"Experimental models and mechanisms of immune defencies of nutritional origih' , Food Additives and Contaminants, Vol. 7 Suppl No. 1, pages s82-S86 reports that specific humoral immunity is altered in diverse fashions by malnutrition, and that serum levels of IgM, IgG and IgA, and immunoglobulin synthesis are usually normal or increased.

Also, Revillard and Cozon, 1990, supra indicates that the catabolic rate of gamma globulins is increased during infections. However, the increased catabolic rate is compensated for by increased synthesis. Further, after nutritional rehabilitation and anti-microbial therapy the immunoglobulin levels returns to normal.

Also, in series of studies conducted with Ghanaian children, it has been shown that supplementing local cereal based diets with fish powder results in a decrease in IgG, IgM and IgA concentrations in serum (Tetteh-Tuwor, 2000; Adu-Affawuah, 1998).

Ribonucleotides are expected to be important as a dietary supplement, especially among infants and preterm neonates. Navarro et al. (1999), BioFactors, 10, 67-76 examined the influence of dietary ribonucleotides on plasma immunoglobulin levels and lymphocyte subsets of preterm infants.

More precisely, Navarro et al. indicates that dietary nucleotides may have an effect on the immature human neonate lymphocytes by showing that the level-of IgA and-IgM are=enhanced in--i-nfants=adm-inistered-a supplement- comprising CMP, AMP, UMP, GMP and IMP. Also, Navarro et al. points out that the nucleotides present in human milk mainly constitute nucleic acids and acid soluble ribonucleotides and some minor amounts of nucleosides and bases. Also, in O. Martinez-Augustin et al. (1997), Biol.

Neonate, 71 : 215-223 it is indicated that dietary nucleotides may be important for the humoral immune response against Cow's milk protein in preterm neonates. In said study, infants were given a standard diet supplemented with ribonucleotides. Another study concluded inter alia that exogenous ribonucleotides are important in maintaining optimal humoral immune responses, although a mononucleotide-nucleoside mixture had no effect on in vitro antibody production and did not increase humoral immune responses in mice fed regular chow, cf. H. Jyonouchi. (1994), American

Institute of Nutrition, 124, 138-143. The nucleotide source in the study reported by H. Jyonouchi, supra was yeast RNA.

Summary of the invention The present inventors have now surprisingly found that supplementary diets comprising a composition based on fish soft roe used according to the present invention provide enhanced serum levels of immunoglobulins.

Further, the present inventors have found that said supplementary diet yields higher serum levels of immunoglobulin G (IgG).

Accordingly, adding to the diet of malnourished people, a preparation comprising a composition based on fish soft roe used according to the present invention, provides a better immune defence shown by stronger immune response and higher serum levels of immunoglobulins like IgG.

Further, adding to the diet of malnourished people said composition according to the present invention, provides improved response to vaccines and other pharmaceutical compositions.

The increase in the immunoglobulin level after regular intake of a protein rich diet supplemented with a composition based on fish soft roe used according to the present invention is surprising, since the immunoglobulin level decreased after intake of the same diet without such supplementation, and considering that this last result is what a skilled person in the art would have expected with reference to earlier published literature (Revillard and =Cozon (19909 Definitions The term"composition based on fish soft roe"referred to herein means a composition comprising at least essentially pure DNA, fragments or derivatives thereof and physiologically and pharmaceutically acceptable salts thereof.

The term"DNA powder"referred to herein below means a composition based on fish soft roe as defined above.

Detailed description of the invention The present invention provides, in one embodiment, a novel use of a composition based on fish soft roe for the manufacture of a preparation for

enhancing the production of immunoglobulins in non-premature mammals, preferably IgG.

In another embodiment according to the present invention, a novel use is provided, wherein the mammals'general state of nutrition is improved.

In still another embodiment according to the present invention, a novel use is provided wherein the mammals'growth and development is regulated.

In still another embodiment according to the present invention, a novel use is provided wherein the mammals'response to vaccine or other pharmaceutical compositions is enhanced.

According to the present invention, it is preferred that said composition contains mainly pure DNA or fragments or derivates thereof. Further, the preparation manufactured according to the present use comprises about 0,5% DNA sodium salt. It is still more preferred that said DNA sodium salt is essentially pure and/or contains at least not more than 2 % protein.

In still another embodiment according to the present invention, a novel use is provided wherein the composition is preferably a pharmaceutical composition, a dietary supplement, a food additive or a fodder additive, respectively.

In still another embodiment according to the present invention, a novel use is provided wherein-the-mammal-is==a human.

Examples 1. Materials and Methods 1. 1 DNA sources The DNA-sodium salt is obtained from naturally occurring sources such as the milt or soft roe from fish. The DNA may be extracted from fish soft roe by well-known isolation methods in the prior art to yield a substantial pure product containing DNA-sodium salt with less than 2 % protein. The product is a white or cream coloured powder. Purity measured by spectrophotometry gives OD260/OD280 = min. 1. 7. The DNA-sodium salt has a molecular weight

distribution of 4 x 104-1 x 106 Daltons and a hyperchromicity OD260 (Denat.)/OD260 (Native) = max. 1. 05.

In the study described herein below, 0.5 % DNA-sodium salt manufactured by Bjrge Biomarin A/S, Norway was used to produce the DNA- supplemented diet according to the present use according to the invention.

1.2 Study design The study was a single blinded randomised community based study. The study site was Nima, a peri-urban slum near Accra the capital city of Ghana. The study subjects were children between the ages of 3-5 years selected by taking the weight and height measurements. Children participating in the study had weight-for-height z-scores below or equal to -1 SD. The eligible children were allocated to two groups by stratified randomisation with the help of random numbers table. Stratification was done for gender and age. The two different groups were based on the two different diets used indicated in the table below. Group one was made up of about 74 children and 70 for group two. Bjrge Biomarin A/S, Norway produced the fish protein and Na-DNA powders used according to the present invention.

Table 1: The different types of diets for the 2 groups Group 1 Group 2 'Rice and-fish protein Rice and fish-protein powder without Na-powder supplemented LU n DNA-supplementation final concentration of 0.5 % Na-DNA.

Equal portions of tomatoes, onions, pepper and spices were fried in equal portions of vegetable oil to prepare the sauce for the two different groups.

The only difference was that the first group had only fish powder while group 2 had DNA in addition to the fish powder. The diets were produced to at least provide the recommended daily intake (RDA) of protein for the children so that if they did not eat any protein at all during the day, the supplementary foods would have provided their requirements for the day.

The children were given ad libitum feeding for lunch. The feeding lasted for seven weeks.

The protein contents of the fish powder (group 1) and the mixture of fish and Na-DNA powders (group 2) were 88.96% and 84.68% respectively on dry matter basis. The protein content of the diet for group 1 was more (18.06%) than that for group 2 (17.39) because the fish powder for group 2 had lower protein content due to the addition of 5% DNA. DNA in diet adds to the nitrogen content in general but not to the protein contents of the diet.

There was no statistical significance of the difference in the intakes of the diets in the two groups.

1.3 Immunological analysis of blood samples.

Venous blood up to 10ml, were collected from the right or left arm of each child by 5 well trained and qualified phlebotomises and one medical doctor from the University Hospital, Legon and Nugouchi Memorial Research Institute, Legon, Ghana. The blood samples were immediately centrifuged and the serum samples harvested were transported under ice packs in an ice chest to the Department of Nutrition and Food Science, University of Ghana, Legon, where they were frozen. At the end of the study all the serum samples were sent to the Broegelman's Research Laboratory for Immunology and Microbiology, University of Bergen, Norway, under dried ice in an ice chest and stored at-20°C until analysed.

About 300 serum samples were sent to the laboratory for analysis. The samples were analysed for various immunoglobulins such as IgG, IgA, IgM and IgE. In all the cases, a method well-known to the persons skilled in the art, Enzyme Linked Immuno-Sorbent Assay, ELISA was used.

By means of a computer programme (Softmax for Macintosh, Version 2.3, Molecular Devices Corporation, Sunnyvale, CA, USA), it was possible to plot a standard curve in 4 parameter mode and results in optical densities were converted into immunoglobulin values. Thus the serum immunoglobulins were expressed in mg/ml. All samples were analysed in duplicates for each immunoglobulin. Samples with coefficient of variation between duplicates of 15 and above were repeated to ensure accuracy of results.

1.4 Animal experiments 1.4. 1 Animals and general experimental set-ups Rat models are the current models used in nutrition studies, in particular

studies related to protein utilisation and growth. Thus, to investigate the effect of DNA from fish roe on growth and immunity, weanling albino rats obtained from Mllengård, Denmark were used in two different experiments as described further below. All the animal studies were carried out at the Institute of Nutrition-Directorate of Fisheries, Bergen, Norway.

All animals were kept in the animal rooms of the Institute of Nutrition, Directorate of Fisheries, Bergen, Norway, and the experiments were carried out using standardized set-ups and feeding procedures. Further, all experiments were performed at a room temperature of 22°C, and with a light/dark cycle of 12/12 hrs.

In the nitrogen-balance experiments the animals were kept in metabolic cages equipped with special devices to control feed and water intake and faeces and urine output. The nitrogen balance experiments were always initiated by a 5-days preliminary period to adapt the animals to the experimental conditions ; during this period the animals also were physio- logically adjusted to the feed and feed composition. In the following 5-days experimental period all animals are subject to feed control; the amounts of feed to each animal were increased depending on whether there were left- overs or not the next day. All left-overs for each rat were collected and weighed. In addition urine and faeces were collected daily. All samples were subject to nitrogen analysis; these data were used to calculate the nitrogen balance of each animal.

In the growth experiments the animals were kept in cages made of steel <BR> <BR> <BR> without controlvof==faecesand urine. H ; =However,-theexperimental=set-sE allowed for control of individual feed intake.

Analytical methods All chemical analysis were carried out using accredited and quality assured analytical methods; they are described in detail in the quality handbook of the Institute of Nutrition-Directorate of Fisheries, and will be available on request.

The moisture and ash contents of all the samples were determined gravimetrically by the official methods of AOAC, i. e. drying for 20 hrs at 105'C, and ashing for 20 hrs at 550 °C respectively. Fat was determined gravimetrically after extraction with ethyl acetate.

Nitrogen (N) in diet, urine and faeces from the animal experiments were

determined according to Dumas'method using a Perkin Elmer 2410 Series II Nitrogen Analyzer (Perkin Elmer, USA). EDTA (containing 9.59 % N) was used to calibrate the analyzer. Crude protein was determined as N x 6.25.

Calculations Calculations were based on the analysis and content of nitrogen in the feeds, faeces and urine from the individual animal in the experiment in addition to the quantity of food eaten and the output of faeces and urine from each animal. From the amounts of nitrogen ingested and excreted, the different indices on protein metabolism were calculated using the following definitions: Apparent Digestibility: AD = (I-F)/I True Digestibility: TD = [I- (F-F')]/I Nitrogen Balance : Bal = (I-F-U)/I Biological Value: BV = [I- (F-F')- (U-U')]/I- (F-F') Net Protein Utilisation: NPU = TD x BV/100 I is the total amount nitrogen ingested by the animal, F is the total amount of faecal nitrogen, U is the total urinary nitrogen, F'is the amount of faecal nitrogen of metabolic origin and U'is the amount of urinary nitrogen on a nitrogen-free diet.

1.4. 2 Determination of the effect on growth, nitrogen metabolism and metabolic parameters The objective of the experiment was to determine the effect on growth, nitrogen metabolism and metabolic parameters of different levels of nucleic acid supplementation to nucleotide-free diets which were optimal in protein, and to compare any difference in effects of such supplementation between DNA or RNA.

The experiment comprised 48 rats which were assigned randomly to 8 groups with 6 rats in each group. Each group was fed with one of the different diets shown in Table 2. Briefly, the diets were composed of cod liver oil and soy oil as the dietary source of fat, sucrose and dextrinised potato starch as the dietary source of carbohydrate, and

cellulose as the source of fiber. Minerals and vitamins were added as separate mixtures. All diets were made isonitrogenous and isocaloric by balancing protein against dextrinised potato starch. Nucleotide-free sodium salt of bovine casein (Sigma) suppelemented with 0.3 % of L-methionine was used as the dietary source of protein. Sodium salt of DNA manufactured from fish soft roe (Bjrge Biomarin AS) and RNA isolated from torula yeast (Sigma) were used as the sources of nucleotides.

Table 2 The composition of the diets used in example 1.4. 2 with respect to protein, DNA and RNA Group Protein % Nucleic acid % No. i Càsew so ; Ol O ; 0it : salt : 2 Casein-sodiwn 20. 0 DNA 0. 5 salt . : ;'",'"S -so salt sale salut , ;.., : 50 ci sodiu 20 ; $ 0 ; 00 . salt : . v° ;... 'A, 1. 0 salt 7 Casein-sodium 20. 0 RNA 3. 0 salt : ;.. salt. :,. : : ., : :. : ' ! ! g'.' . !' salt"... ;.'.. -.', !. !. '.'. '. !'' ;.' J''.''.'''.' The animals were kept for 3 weeks with first 5 days forming the pre-testing and adjustment period where no samples of urine or faeces were collected. Four rats from each group were used in the nitrogen balance experiment for 5 days. These animals were weighed and housed each in a suspended screened-bottomed metabolic cage in a room with temperature of about 20°C. The animals were re-weighed. The five-day

urine of each rat was kept in a freezer for further analysis while the faeces were freeze-dried for 72hrs.

1.4. 3. Effects of the individual nucleotides in DNA vs. RNA The objective of the experiment was to examine the effects of the individual nucleotides found in the nucleic acids DNA and RNA on growth and nitrogen metabolism in rats fed a nucleotide-free diet, which was optimal with respect to protein, and to compare any effects of such supplementation with supplementation of Na-DNA powder from fish.

In this study, 48 rats were assigned randomly to 8 groups with 6 rats in each group. Each group was fed with one of the diets shown in Table 3. The diets were composed of cod liver oil and soy oil as dietary source of fat, sucrose and dextrinised potato starch as the dietary source of carbohydrate, cellulose as a source of fiber. Minerals and vitamins were added as separate mixtures.

All diets were made isonitrogenous and isocaloric by balancing protein against dextrinised potato starch. Nucleotide-free sodium salt of bovine casein (Sigma) supplemented with 0.3 % of L-methionine was used as the dietary source of protein. Sodium salt of DNA manufactured from fish soft roe (Bjrge Biomarin AS) ; the nucleotides (AMP, CMP, GMP, TMP and UMP) were purchased from Sigma.

Table 3 The composition of the diets used in example 1.4. 3 with respect to protein, nucleotides or DNA Group No. Protein % Nucleotide or DNA % 1 Casern-sodium salt 20. 0 Unsupplemented 0. 0 2 Casein-sodium salt 20. 0 AMP 0. 5 3 Case in-sodium salt 20. 0 4 Casein-sodium salt 20. 0 CMP 0. 5 : 50 Case 4 Casein-sodium salt 20. 0 CMP 0. 5 7 Casein-sodium salt 20. 0 DNA 0. 5 6 Casein-sodium salt 20. 0 UMP 0. 5 7 Casein-sodium salt 20. 0 DNA 0. 5 clium salt' :, . : :, 1. 0

The animals were kept for 15 days, of which the first 5 days formed the pre- testing and adjustment period where no samples of urine or faeces were collected. The experimental period lasted for 10 days. The 5 initial days of the experimental period were performed as a nitrogen balance experiment, while the remaining 5 days were performed as a growth experiment. The animals were initially weighted, at the end of the nitrogen balance period (i. e. after 5 days in experiment) and at the end of the experiment (i. e. after 10 days of experimentation).

1.4. 4 Effects ofdexyribonucleotide supplementation on growth and protein metabolism To determine the effects of deoxyribonucleotide supplementation to nucleotide-free diets on growth and protein metabolism at low, medium and high levels of dietary protein 48 rats were assigned randomly to 6 groups with 8 rats in each group and each group was fed with one of the diets shown in Table 4. The diets were composed of cod liver oil and soy oil as dietary source of fat, sucrose and dextrinised potato starch as the dietary source of carbohydrate, cellulose as a source of fibre. Minerals and vitamins were added as separate mixtures. All diets were made isocaloric by balancing protein against dextrinised potato starch. Nucleotide-free sodium salt of bovine casein (Sigma) supplemented with 0.3 % of Umethionine was used as the dietary source of protein. Sodium salt of DNA manufactured from fish soft roe was used as the source ofnucleotides (Bjjarge Biomaiin AS).

Table 4 The composition of the diets used in example 1.4. 4 with respect to protein and DNA- Group Protein level % DNA in diet % No. l Lo r ten, ;-8e03 1 : j ! ! s !' 2 : Low protem :... :, 8. 0 + DNA 0. 5' 3 Medurri protein'17. 0-DNA 4 : Medium protein 17 0 . : DNA 0 : 5 5 0.

The animals were kept for 3 weeks with first 5 days forming the pre-testing and adjustment period where no samples of urine or faeces were collected. Four rats from each group were used in the nitrogen balance experiment for 5 days. These animals were weighed and housed each in a suspended screened-bottomed metabolic cage in a room with temperature of about 20°C. The animals were re-weighed. The five-day urine of each rat was kept in a freezer for further analysis whiles the faeces were freeze-dried for 72 hrs.

2. Results 2. 1 Effect on Serum Levels of the Immunoglobulins The IgG, IgM and IgA levels in the serum samples of the children in this recent study at baseline were within the range of what has been reported for normal Ghanaian children (Adu-Affawuah, 1998 and Neuman et al, 1975,"Immunologic respons in malnourished children"; Am. J. Clin.

Nutr., 28 : 89-104). Since the Neuman's study involved children up to 6 years of age which encloses the age range used for the recent study, the results of the serum levels of the Ig's mentioned above in the study children at baseline can be said to be reliable. This is regardless of the fact that the ELISA method used, like any other laboratory analytical method, was believed to have its own inherent or associated errors but since great care was taken during the assays, any such errors were reduced to the barest minimum.

At baseline, the mean serum IgG concentration of the children in the group that received only fish powder was fund to be 12.087 mg/ml ~ 2.68, and 10.91mg/ml ht 4.23 for the children who received fish and DNA powder.

However, a student's t-test of significance showed that the difference was not statistically significant (p>0.05).

The mean serum concentrations of IgM and IgA were almost comparable for both groups.

The general health of the children in both groups improved dramatically during the period of supplementary feeding. The number of children who reported sick to the clinics, decreased as the number of weeks of supplementation increased. There was a drastic decrease in the number of children who had skin rashes and running noses by the end of the supplementary feeding. Thus, there is no doubt that supplementing the

rice with both the fish powder and the mixture of fish and DNA powders improved the health status of the children.

It is important to note that since the serum concentrations of the various immunoglobulins determined by the ELISA procedure at baseline were found to be reliable, it can be said with confidence that in the same way the results after supplementation are reliable and hence they can be used to make inferences. Also since the children in the two groups were not significantly different (statistically) with respect to their serum levels of the immunoglobulins at baseline, any difference in their levels of the immunoglobulins after supplementation can be said to be due to the difference in the supplementary diets.

As mentioned previously above, in series of studies conducted with Ghanaian children, it has been shown that supplementing local cereal based diets with fish powder results in a decrease in IgG, IgM and IgA concentrations in serum (Tetteh-Tuwor, 2000Adu-Affawuah, 1998).

In the study reported herein, supplementation with fish powder alone was found to result in a significant decrease in IgG concentration and this conforms to results of earlier studies mentioned above.

The effect of supplementation on the serum level of immunoglobulins of group 1 is summarized in Table 5 below.

Table 5 : The effect of supplementation on the serum concentrations of immunoglobulins of the children in group 1 (mean~SD) Period IgG IgM IgA IgE (mg/ml) (mg/ml) (mg/ml) (tg/ml) Baseline 12.08 0.899 1.02 0.819 ~ 2. 68 : L 0. 37 0. 39 ~ 1. 21 After suppl. 10.35 0.950 1.06 0. 848 i 2. 97 : L 0. 35 : L 0. 37 ~ 1. 43 In the group that had only fish powder (group 1), there was a significant decrease (p<0.05) in the mean IgG level, corresponding to 14,3% in terms of their baseline level, a significant increase in the mean IgM level (p=0.03)

and non-significant increases in mean IgA and IgE levels (p>0.05).

In group 2, a significant increase in the serum IgG concentration was demonstrated. This corresponds to 25% in terms of the baseline level. This significant increase in the IgG concentration, and the increase in the levels of the other Igs in the serum samples of the children (even, though not significant statistically) fed with the diet supplemented with a preparation comprising mainly DNA from fish soft roe can in no doubt be said to be due to the presence of DNA. The exact mechanism explaining this phenomenon is yet not known.

The effect of supplementation on the serum levels of immunoglobulins in group 2 is summarized in Table 6 below.

For all the immunoglobulins analyzed in this study, there were higher increases in the levels in the serum samples of the children fed with the DNA-powder supplemented diet (group 2) than those fed with the non- DNA-powder diet (group 1). A more dramatic observed difference, which was statistically very significant, was the decrease in IgG level in group one whereas there was an increase in that Ig in group 2.

The significant difference between the IgG levels in the two groups after supplementation must be due to the further supplementation of the diet for group 2 comprising DNA from fish soft roe.

Table 6: The effect of supplementation on the serum concentrations of immunoglobulins of the children in group 2 (meahSD) Period IgG IgM IgA IgE (mg/ml) (mg/ml) (mg/ml) (p g/ml) Baseline 10.91 0.821 1. 03 1.139 4.23 i 0. 39 0.36 1. 92 After suppl. 14.04 0.932 1. 11 1.393 4. 28 0. 58 i 0. 41 2. 70 In the group that got DNA-powder (group 2), there was a significant increase in the IgG level (p<0.05), and increases in IgM, IgA and IgE levels, which were not statistically significant.

Even though the increase in the IgE level in the fish/NT group was more than in the fish group, the difference between the two groups was not statistically significant (p > 0.05). However, this might be due to small sample sizes. It is worth mentioning that the very high values of standard deviation, was due to the variation between the various IgE concentrations amongst the children in both groups. Some of the children had very high values while others had very low values but this trend was evenly distributed among the two groups and hence did not in anyway affect the outcome of the results.

The normal levels of IgE in serum of children have been reported to be between 0.01-0. 31lg/ml. However in conditions of malnutrition and incidence of intestinal parasites high levels of IgE have been reported. Control African subjects have been reported to have higher levels of IgE than do US/European controls. This is probably due to a higher frequency of intestinal parasitism in the African children. Despite these high levels of IgE, allergy is uncommon particularly in children in underdeveloped areas. The children were not dewormed prior to the study. The elevated levels of IgE in the serum samples of the children in the current study at baseline was therefore not far from what should have been expected since the children were from a community at high risk of intestinal parasites considering their sanitation conditions. The very high IgE values occasionally found in some of the children may be caused by intestinal worm infections.

The lgE-lxvel-s-i-n the seru-m-of the children were expvected to be elevated after supplementary feeding with both diets since the fish used in this study was codfish. Several studies have documented the presence of a major allergen Gad cl in cod muscle tissue proteins. The consumption of fish can cause IgE-mediated reactions in allergic individuals and this can sometimes be fatal. Allergen from cod fish is different from that of herrings, mackerels and tuna eaten by most Ghanaians, however there is always an overlap of allergen specificity known to occur between different types of allergens.

Some allergens can be destroyed by heat but allergens from cod fish are heat stable. The prevalence of allergy varies with food habits. Even though the prevalence of fish allergy in Ghana is not known, the consumption of fish is not expected to be above the average levels found in Scandinavian countries,

Spain and Italy, where the incidence of fish hypersensitivity is observed to be high.

The IgE levels of the children from both groups were very high even at baseline when supplementation had not begun. The increase in the levels in both groups after supplementation even though not statistically significant, cannot be said to be due to the presence of allergens from the codfish used.

However, the contribution of this allergen to the results cannot be overlooked.

While there was a decrease in IgG concentration in serum samples of children in group 1, there was an increase in group 2. The insignificance of the increases in IgA, IgM and IgE might be because of small sample size. However the increase in the IgA concentration in the children in group 2 had a p-value of 0.0577. This means that one can say with about 94% confidence that the increase in the IgA concentration in the DNA group was significant.

Table 7 shows the mean changes in the serum immunoglobulin concentrations of the children in the two groups after supplementary feeding. It is important to note that whereas there was a significant decrease in the IgG level in the non-DNA group, there was a significant increase in the level of the same immunoglobulin in the DNA group.

Table 7: Mean changes in the serum concentrations of the various immunoglobulins- Group IgG IgM IgA IgE (mg/ml) (mg/ml) (R g/ml) (mg/ml) 1-2.2459 0.0875 0.0235 0.0458 0. 319 i 0.3153 1.075 5.1076 2 3.8152 0.0975 0.0731 0.245 0. 465 i 0.2924 i 1.236 4. 6137

A student's t-test of the difference between the two groups showed a very high significance of p-value being below 0.01. Since there was no difference between the two groups at baseline with respect to IgG, it means that the difference after supplementary feeding is due to the difference in the diets.

The increase in the concentrations of the other Igs (IgM, IgA and IgE) after supplementary feeding were still more in the group 2 than in group 1 even though the differences were not statistically significant in all the three cases (p>0.05). The increased level of IgA in the children in group 2 had a p-value of 0.0577, equivalent to a confidence level of 94%.

However, it is worth mentioning that addition of the DNA-powder to the diet caused an effect in the immunoglobulin levels and the difference between the groups could not be detected statistically in the case of the other three immunoglobulins probably because of small sample size.

Accordingly, feeding children with a fish protein powder supplemented diet for seven weeks resulted in a significant decrease in their serum IgG concentration, whereas feeding children with a fish protein powder plus a supplemented diet comprising DNA-powder isolated from fish soft roe resulted in a significant increase of their serum IgG level; the latter indicates that DNA supplementation improves the body's capability to synthesise antibodies and indicates an improved defence against infections.

Therefore, the results presented herein confers that the present use of a composition based on fresh or conserved fish soft roe according to the invention is useful in enhancing the serum immunoglobulin level according <BR> <BR> <BR> to the present invention and consequently strengthen the immune system and- the response to vaccines and/or other pharmaceutical compositions.

Especially, these results confer that the present inventions usefulness in enhancing the serum IgG level.

2. 2 Animal experiments 2.2. 1 Determination of the effect on growth, nitrogen metabolism and metabolic parameters The growth rates and feed consumption are shown in Table 8. The feed consumption between the various groups differed only slightly and was not significant (p < 0.05). Any difference between groups in growth rate and protein metabolism was therefore not caused by difference in feed

consumption; any differences should therefore be ascribed to qualities of the diet.

The weight gain and mean growth rate increased in the animals fed diets supplemented with DNA in the range of 0.5 % to 3.0 % as compared with the animals given the unsupplemented diet. Feeding the animals diets supplemented beyond 3.0 %, i. e. at a level of 5.0 % DNA leads to a reduction in the growth rate resulting in a daily weight gain equivalent that of the animals given the nucleotide-free diet. The growth rates of the animals fed the RNA supplemented diets were inferior to both the nucleotide-free and DNA-supplemented diets, and averaged 3.9 g/day as compared to 4.2 g/day in the nucleotide-free diet and 4.6 g/day in the animals fed 0.5 % to 3. 0 % DNA in the diet.

Table 8. Growth rates and feed intakes in rats fed a nucleotide-free diet and diets supplemented with different levels of DNA and RNA as the source of nucleotides (Means SD). % Total weight Daily weight Total feed Daily feed DNA/RNA gain gain intake per intake animal . 4 : 4. 2 0'. 3 : :' : . DNA :-0. 0' % 21 :. 0 , l :. 49. 9-1 0'. 1 0. 0 DNA-0. 5 % 22. 8 0. 6 4. 6 i 0. 1 49. 9 te 0. 2 10. 0 i 0. 0 DNAS 0/o i 21. 8 i'1. 4 i04 4 t DNA-3. 0 % 23. 5 zt 1. 8 4. 7 i 0. 4 49. 8 i 0. 4 10. 0 i 0. 1 INA-3. 0% 23. 51. 8 4. 70. 4 49. 80. 4 10. 0 0. 2 DNA-5 0% 20. 71. 0 4. 1. : 0 : 2 49. 61 : :. 0 : 9. 9 '02 : : I2NA=1. 0%. . 19. 00. 9. 3. 80. 2 49. 80. 4. 10. 0 01 : RNA-3. 0 % 19. 8 1. 6 4. 0 0. 3 48. 8 1. 3 9. 8 0. 3 RNA-5. 0 % 19. 0 =L 2'9'., . 8 +. 0. 6 : : :,," Table 9. Indices of protein metabolism and protein digestibility in rats fed a nucleotide-free diet and diets supplemented with different levels of DNA and RNA as the source of nucleotides (Means i SD). DNA/RNA PERop BALop BVop NPUop ADop TDop DNA-0. 0 % 1. 8 A 0. 1 0. 56 + 0. 1 0. 67 ~ 0. 1 0. 66 ~ 0. 1 0. 96 ~ 0.0 1.02 ~ 0. 0 DNA-0. 5 % 2.0 ~ 0. 1 0. 57 ~ 0.0 0. 68 0. 0 0. 67 0. 0 0. 96 0. 0 1. 02 ~ 0. 0 DNA-1. 0 % 1. 9 ~ 0.1 0. 57 0. 0 0. 68 ~ 0. 0 0. 66 i 0. 0 0. 96 ~ 0. 0 1. 02 ~ 0. 0 DNA-3. 0 % 2. 0~ 0. 1 0. 56 i 0. 1 0. 67 0. 1 0. 66 ~ 0. 1 0. 96 ~ 0. 0 1.01 0. 0 DNA-5. 0 % 1. 8 ~ 0. 1 0. 54 0. 0 0. 65 ~ 0.0 0. 64 0. 0 0. 96 ~ 0.0 1.02 ~ 0. 0 RNA-1. 0 % 1. 7 ~ 0. 1 0. 51 ~ 0.0 0. 61 ~ 0.0 0.61~ 0. 0 0. 95 0. 0 1. 01 0. 0 RNA-3. 0 % 1. 7 ~ 0. 1 0. 53 0. 0 0. 63 ~ 0. 0 0. 63 ~ 0. 0 0. 95 ~ 0. 0 1. 01 0. 0 RNA-5. 0% 1. 60. 2 0. 57 ~ 0.1 0.67 ~ 0. 1 0. 67 0. 1 0. 95 ~ 0. 0 1. 00 0. 0 Op= operational values The difference between DNA and RNA as a source of dietary nucleotides was also reflected in the Protein Efficiency Ratio (PER-values) obtained for the different groups of animals (Table 9). The PER-values for all groups fed RNA differed significantly (p < 0.05) from those fed the DNA-supplemented diets. In contrast to the growth rate the other indices for protein metabolism obtained by measuring the nitrogen balance of the animals, i. e. nitrogen <BR> <BR> <BR> balance, biological value and-net protein utilisation,-showed no difference- between diets (p < 0.05) when the animals were fed the nucleotide-free diet and diets supplemented with 0.5 %, 1.0 % and 3.0 % DNA, while the animals offered 5.0 % DNA in the diet showed reduced metabolism and utilisation of protein, corresponding with a reduced growth rate as shown in Table 8. Supplementing the nucleotide-free diet with 1.0 % and 3.0 % RNA results in a significant (p < 0.05) decrease in the nitrogen balance and net protein utilisation, while supplementing the diet with 5.0 % RNA resulted in values for the nitrogen balance and protein metabolism equal to those obtained in animals fed DNA-supplemented diets. This finding contrasts with the corresponding growth rate and PER-values, and may therefore be considered as an artefact. The different forms of nucleotide supplementation did not affect neither the apparent nor the true digestibility of protein in the animals. Consequently, the difference in protein utilisation in the animals is

not caused by a poorer digestion and absorption of protein and amino acid in the animals fed the RNA-supplemented diets, but is rather a result of an improved protein retention and deposition in the animals fed the DNA supplemented diets.

Taken together, the results from the experiment indicate that DNA from fish soft roe is far superior to RNA in supporting growth and protein metabolism in rats. In fact, the results indicate that RNA, at least RNA from torula yeast, may not be used as a source of nucleotides in nucleotide-free diets.

2.2. 2. Effects of the individual nucleotides in DNA vs. RNA The aim of the experiment to examine whether the difference in growth and protein metabolism between DNA and RNA shown in example 1.4. 2 was due to their difference in nitrogen bases ; while DNA contains thymine RNA contains uracil. Due to the highly expensive nucleotides used in the diets the experiment was only allowed to last for 15 days.

The weight gains, feed consumption and feed utilisation by the rats fed the various experimental diets are shown in Table 5. There was no difference (p<0.05) in the mean daily weight gain between animals fed the nucleotide- free diet and those fed diets containing AMP, GMP, CMP and UMP. The animals fed the diet containing UMP showed a slightly, but not significant, lower growth rate than those fed the former diets. In contrast, the animals fed the TMP supplemented diet showed significantly (p<0.05) higher growth rate than those fed the nucleotide-free diet and AMP, GMP, CMP and UMP containing diets. Further, the growth rate in the TMP-supplemented animals was identical to that of the animals fed the DNA-supplemented diets. There was no difference in growth rate between the animals fed either 0.5 % or 1.0 % DNA supplementation. Table 10. Growth rates and feed intakes in rats fed a nucleotide-free diet and diets supplemented with the nucleotides AMP, GMP, CMP, TMP and UMP, and diets supplemented with DNA as the source of nucleotides (Means i SD). % Total weight Daily weight Total feed Daily feed DNA/RNA gain gain intake per intake animal 2. 8. 37., U, 4suppl,., GMP", 27. 6 12'4 2. 8 0. 3. 3 0. 1 : : GMP j 27. 6 i 4 2. 8 i 0. 3 ; 59020 0. CMP 28. 41. 8 2. 8i0. 4 58. 8 : L 0. 4 5. 9 ztO. l TMP 3. ; ll 0 9-3 l 5 03 _ :... :.,., UMP 27. 1 0. 9 2. 7 0. 2 60. 4 0. 4 6. 0 0. 1 DNA-0. 5% 3I. 31. 6 3. 1 i 0. 3 61. 4 i 1. 3 6. 1 i 0. 3 XDMu. 0% 0 0300X50 ; ; 9 t 030 0. 6 S 9. 8 0. 3Al ; The difference between the various forms of nucleotide supplementation on the animals'protein metabolism is shown in Table 11, and verifies the results from the growth experiment. The animals fed the TMP and DNA supplemented diets showed significantly higher net protein utilisation and <BR> <BR> ni-trogen-balance-t-han-those fed-theEwnsupplemented and-A-MP, GMP-, CMP-- and UMP supplemented diets. The different forms of nucleotide supplementation did not affect neither the apparent nor the true digestibility of protein in the animals. Consequently, the difference in protein utilisation in the animals is not caused by any inferior digestion and absorption of protein and amino acids in the animals fed the AMP, GMP, CMP and UMP supplemented diets, but is rather a result of an improved protein retention and deposition in the animals fed the DNA and TMP supplemented diets.

Table 11. Indices of protein metabolism and protein digestibility in rats fed a nucleotide-free diet and diets supplemented with the nucleotides AMP, GMP, CMP, TMP and UMP, and diets supplemented with DNA as the source of nucleotides (Means i SD). Nucleotide/DNA BALop BVop NPUop ADop TDop Unsuppi. 0. 56 i ; 1 0. 1 0. 69 i 0 ;. ; 1 0. 1-. ; 01 AMP 0. 570. 1 0. 690. 1 0. 680. 1 0. 970. 0 1. 010. 0 G i ; 0 507 i 0 0 0. 0 0. 6. 0. 98 0. 0 I. 2 0*. 59" : 4 :, ., 0. 8 L 0. 0 1. 00 i 0. 0 WP. 73 0 0 0. 72 ;-. O : ; 1 01. i UMP 0. 51 0. 1 0. 65 0. 1 0. 64 0. 1 0. 97 0. 0 1. 02 il 0. 0 DNA-0. 5 % 0. 59 0. 0 0. 73 i 0. 0 0. 72 zt 0. 0 0. 98 0. 0 1. 01 0. 0 DNA 1. 0 %-i ; 0. 58 0 100 0. 720 O. 1 1 014 O 98 0 0 1. 0. 0 Op= operational values Taken together, the results may indicate that the difference between RNA and DNA in supporting growth and protein metabolism in rats as shown in Expt. No. 1.4. 2 is due to their difference in nitrogen bases, i. e. uracil versus thymine, and that the latter to a greater extent may be conditionally essential <BR> <BR> <BR> =to-animals-fed a-nuc-leotide-free-diefthan the other nitrogen-bases found-in the different types of nucleic acids. It should, however, be pointed out that the diets used in the experiment are only comparable on the nitrogen-base level. While the DNA-and TMP-diets contain deoxyribonucleotides, the AMP, GMP, CMP and UMP supplemented diets contains ribonucleotides without a reduced ribose-moiety in the nucleotide molecule. From economical reasons a study comparing the different deoxyribonucleotides with UMP was not feasible.

2.2. 3 Effects of nucleotide supplementation on growth and protein metabolism The growth rates and feed consumption are shown in Table 12. There was a slight but not significant (p < 0.05) decrease in the mean growth rate of the animals when the protein content of the diets were increased from low (LP =

8 %) to medium (MP = 17 %) to high (HP = 24 %). The slightly higher growth rate in the animals fed the unsupplemented low protein diet (LP) as compared to those fed the unsupplemented medium protein (MP) and high protein (HP) may be due to the significantly (p < 0.05) higher feed intake by the low protein group. Diets containing only 8 % protein (LP) is far inferior to the animals'requirement for protein. At this dietary protein level the limiting factor for growth and protein utilisation in the protein is its content of essential amino acids, or more specifically, its content of the first limiting essential amino acid. In the current experiment adding nucleotides in the form of DNA significantly (p < 0.05) reduce the growth rate in animals given a low protein diet (LP + DNA). However, the reduced growth rate may be explained by a significant (p < 0.05) lower feed intake in the animals given the DNA supplemented diet. In contrast, supplementing medium protein and high protein nucleotide-free diets with DNA significantly (p < 0.05) increased the animals'growth rate. Although the DNA-supplemented animals showed a higher feed intake the difference in feed intake between unsupplemented and supplemented diets do not fully account for the differences in growth rate. The highest growth per day was obtained at the medium protein level with DNA supplementation ; this protein level is optimum for rats.

Table 12. Growth rates and feed intakes in rats fed low, medium and high protein nucleotide-free diets supplemented with DNA as the source of nucleotides (Means SD). Total weight Daily weight Total feed Daily feed Diet gain gain intake per intake Dieu gain gamin animal -6 77.'0 2. 1, 15. 4 0. 4 LP + 2. 4. 9 : A 1. 50 5. 00 0 3 ; 00 9 ì ; ot-7y ; S ; MP 3 6. 9''12'.-3 : '-l-. 4 MPu + DNÅ ; 40. 5 it9. 0 ß 8-; ; 8 ; t ti 2. 2 i 8. 30 0 : 104. 4 1. 7 . 29. 0 65. 3L 4'i :'I : 13. 1, 0, : 8 : . hop

Neither the protein level in the diet nor DNA-supplementation affected the digestion and absorption of protein and amino acids in the animals, as shown by the values of true digestibility of protein (TD) in Table 13. As expected the indices of protein metabolism and utilisation showed decreasing values as the dietary level of protein increased. In contrast to the effects on growth rate supplementing DNA to low protein diets only slightly affected the indices of protein metabolism and protein utilisation. The fact that DNA supplementation to low protein diets reduce the animals growth rate but maintains a high nitrogen balance and net protein utilisation indicates that protein to a greater extent also is used for other purposes than pure growth when nucleotides is added to nucleotide-free diets.

Table 13. Indices of protein metabolism and protein digestibility in rats fed low, medium and high protein nucleotide-free diets supplemented with DNA as the source of nucleotides (Means i SD). DIET PERop BALOP BVp NPU, P, AD, p TD, p LP 4. 60. 2 0 : 81O : Ov : 1. 05. : 0 : 0. . 02. O. p p. 890 : 0. : 03p : p LP+.. 4. . 0. 5 0 : 790. 1 1. 02t0 : 1 0. 990. 0. 0. 900. 0 1. 030 : 0v DNA.. 12, DNA MP-'22 8 0, 75, t, 0.'l 0. 1 87 : 87 0. 1, 0. 95 0. 0 1. > 2. 690. 9 <087i00 O8G : : 0 D 090 : 09 1 0I 00 HP : : ; 1. 50. 2 :. : 680 : 0p. 76p. n, '75O. p. . 96°0 : 0 1 : 01O. Q : : : HP 1. 5 0. 2 0 : 68 OJO i 0. 76 0 : 0 0. 75 0 : 0 0. 96 0 : 0 1. 01, HP + 1. 6 :. 0. 1 0. 610. 1 0. 69 0. 1 0. 68 0 : 1 0. 96 0. 0 1. 01 DNA Op= operational values Taken together, 0.5 % supplementation of DNA to nucleotide-free diets at medium and high levels of protein generally improves the growth rate in rats ; while at low protein levels DNA supplementation reduce the animals growth rate. It seems as if DNA supplementation is of nutritional significance only when the diet contains adequate amounts of dietary protein to support the potential for increased growth.

According to the present invention, the results from the animal studies indicate that DNA from fish soft roe is far superior to RNA in supporting growth and protein metabolism in rats. RNA, at least RNA from torula yeast, may not be used as a source of nucleotides in nucleotide-free diets. Also, the results indicates that DNA supports the growth rate when added to nucleotide-free diets at a level of 0.5 %. The difference between RNA and DNA in supporting growth and protein metabolism in rats is due to their difference in nitrogen bases, i. e. uracil versus thymin, and that the latter to a greater extent may be conditionally essential to animals fed a nucleotide-free diet than the other nitrogen-bases found in the different types of nucleic acids. It should, however, be pointed out that the diets used in the current study only are comparable on the nitrogen-base level.

Thus, supplementation of DNA from fish soft roe at a level of 0.5 % to nucleotide-free diets at medium and high levels of protein generally improves the growth rate in rats, while at low protein levels DNA supplementation reduce the animals growth rate.