Introduction

Hungary and Serbia are similar countries in respect of fish production and fish production capacity, in fish consumption and in cardiovascular disease (CVD) mortality rates. The area for pond culture in Hungary is around 25 000 ha and roughly 70 % of the produced fish is common carp (Pintér, 2009). In Serbia, the total area covered by warm water fish farms is about 12 000 ha and approximately 97 percent of the fish farms are located in the Northern part of Serbia, in Vojvodina. Common carp is the predominant species here, too (FAO, 2011). Compounded feeds are used now mainly in carp nursery ponds or in some carp cage culture units. In the carp rearing ponds, fish are fed mainly on cereal grains like wheat, corn and barley, depending on price and local availability. However, an increased usage of pelleted feeds in carp culture is a necessity to increase the production and fish quality. This is especially important when the human nutritional value of the carp flesh is intended to be increased.

The per capita fish consumption was 5.2 kg in 2006 in Serbia (FAO, 2011) and approx 6.0 kg in live weight in Hungary (4.16 kg fish meat/per capita in 2008 in Hungary - Pintér,2009), being among the lowest in Europe. The CVD mortalities are over 50% in both countries (Knezevic & Grozdanov, 2009; Bényi, 2008).

It is well established that consumption of fish rich in eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) favourably affects CVD mortality (e.g. Chris-Etherton et al., 2002; Simopoulos, 2008). At present the main health concern is connected to increase the consumption of omega-3 and omega-6 long chain polyunsaturated fatty acids (LcPUFA) like arachidonic acid (ARA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). The Global Omega-3 Summits (http://www.omega3summit.org/) dealt extensively with these issues (Omega-3 Summit, 2011, 2012,  2013). The experts also agreed that “Tissue concentrations of LC-Omega-3 (relative to LC-Omega-6) are the key variable for health not dietary intakes. Biomarkers need to be standardised and used as public health targets.” An Omega-3 Index of 8-11, meaning that Omega-3 in HUFA over 50% would protect 98% of population of the industrial countries (Omega-3 Summit, 2011). The Omega-3 index is the sum of 2 prominent long-chain n-3 fatty acids [i.e. eicosapentaenoic acid (EPA) and docosahexaenoic acids (DHA)] in erythrocyte membranes and is expressed as a percentage of total erythrocyte fatty acids (FAs) as suggested by Harris and von Schacky (2004).      The Global Omega-3 Summit in 2013 debated on the “Role of Omega-3 and Omega-6 in Health”, on the “Importance of Ratio and Bioavailability”, and on “Consumer Awareness of Omega-3s and its Benefits” (Omega-3 Summit, 2012).

Necessity of increasing the environmental sustainability of both feed and fish production in respect of lowering release of  waste nutrient into the environment is also a serious concern (Naylor et al., 2000; Roth et al., 2000; Bartley et al., 2007; Ellingsen et al., 2009).
Ensuring the maintaining sustainability of polyunsaturated fatty supply is attempted by applying gene manipulation plants, for instance genes for enhanced production stearidonic acid (18:4n-3) and EPA and DHA were introduced into various oilseed production plants (Nichols et al., 2010). Food grade arachidonic acid produced in Mortierella alpine is approved for trading in European Union (Nr. 2008/968/EC Commission Decision).

In the present study we planned to apply an extrusion technology for carp feed production from selected ingredients and with precise setting of the extrusion parameters for conditioning and extrusion. Efficiency was planned to be tested by feeding the new, floating type feed for carp supplemented with omega-3 and omega-6 fatty acids (FAs) with proved health benefits (Del Prado et al., 2001; Chris-Etherton et al., 2002; Harris and von Schacky 2004; Harris, 2008; Harris et al., 2009;). The health benefits of the improved quality of the produced carp meat in respect of cardiovascular diseases was planned to be studied in a mammalian animal model (rat). As common carp is widely consumed in both Hungary and Serbia, it is expected that our results – when applied in the practice of carp culture – can contribute to lowering the risk of cardiovascular deceases in the region.

Materials and Methods

Proximate analysis methods for determination of the gross nutrient contents of feed ingredients, feeds and fillets
Gross nutrient content data of feed ingredients, feeds and fillets were determined by standard methods with modifications in some cases. Shortly, dry matter content was determined by oven drying at 105ºC for 4 hours. The oven dried samples were combusted on ceramic hot plate, than the crucibles were placed into muffle furnace for crude ash determination at 550ºC for 4 hours.
Crude protein was determined by the Kjeldahl-method using digestion block (SPEED DIGESTOR K-436, Buchi, Switzerland) and distillation method (DISTILLATION UNIT K-350, Buchi, Switzerland). Ammonia was  titrated through the titration of the residual sulfuric acid (0.05 M) by NaOH (0.1 M).
Crude fat was determined by Soxhlet method using semi-automatic system (SOXTEC HT 1043, Tecator, Denmark) and petroleum ether of 40-60ºC boiling range.

Crude fiber content was determined directly from low fat samples. If the presumable fat content was higher than 10%, the determination was done after the extraction of fats with chloroform. The samples were hot digested consecutively for 30 min with sulfuric acid (0.26 mole/liter) and 30 min with potassium hydroxide (0.23 mole/liter), then filtered onto filter crucibles, dried and weighed.
The proximate composition of the experimental feeds presented in Table 1 were determined by AOAC (2005) methods and metabolizable energy contents according to Csengeri and Majoros (2004).

Feeds	CF1 Sinking control (SBO)	CF 2 Linseed oil (LSO)	CF 3 LSO+ ARA+DHA	CF 4 Fish oil (FO)	CF 5 FO+ARA
Dry matter	931	941	936	939	936
Crude protein	310	327	322	322	323
Crude fat	80	78	64	75	67
Crude fiber	28	19	19	24	21
Crude ash	91	65	68	70	66
N-free extract	422	452	463	447	459
Starch	278	217	245	218	272
NDF	104	105	99	93	118
ADF	28	21	21	24	32
ADL	7	8	11	12	9
ME (MJ/kg)	12.68	13.52	13.05	13.21	13.22

Experimental fish

The experimental fish were collected in December of 2012 then quarantined and upon arrival to the RAS system special antibacterial and antiparasitic treatments were applied with formalin and salt solutions. The carefully selected and acclimatized fish were then weighted and experimental groups distributed into the compartments rearing tanks (12 individuals/compartments). The feeding experiment was performed in 2.0 m3 plastic square-shaped tanks divided into two compartments (about 1.0 m3 each) by a screen wall equipped along the one diagonal of each tank. The tanks were covered by wide-meshed net to avoid jumping out of the fish from the rearing containers. Water inflow into the tanks was regulated to ensure oxygen concentration of the outflow water at least 80 % of saturation. This required – depending from water temperature, fish biomass and daily feeding rate – a turnover rate about 2 times per hour. During the experiment water temperature and oxygen saturation were measured twice per day (at about 8 and 19 hour), water chemistry analysis were performed twice per day.
Initially all diets was applied at a daily amount of 90 % of the previously determined maximal consumption. This value was converted to % of metabolic fish weight (kg 0.8) and further this feeding rate was be used for the calculation of the daily ration according to the new weight attained at the end of each week. Fish was weighed at weekly intervals in anesthetised condition. Feeding was performed by automatic belt feeders during 10 hours per day. Each day at morning uneaten feed residues and faeces was siphoned from each tank and tank bottoms and walls was cleaned from biofilms. Fish behaviour and health was inspected continuously and in needed cases adequate medical treatments was applied either locally on the body surface and/or in a form of bathing in the solution of the requested substances.

 

In the carp feeding experiments five soybean meal based (36%) diets were tested. The diets were formulated as listed above to have approximately 30-32 % final protein and 8-8.5% final oil contents. Each diet was supplied to duplicate groups.

The growth and feed conversion efficiency for the weekly periods, as well as for the whole duration was characterised by daily growth rate (SGR) and feed conversion rate (FCR) as follow:
SGR = 100 x (ln w_t - ln w₀) x t ^-1 (%.day ^-1)
FCR = F x (W_t – W₀) ^-1 (g.g ^-1)
where w0; wt - initial and final average weight; W0; Wt - initial and final biomass; t - duration of the investigated period in days.

The effect of different feed types on fish condition was evaluated by the condition factor (CF) as follows:
CF = 100 x w_t x L_t ^{- 3} (%)
wherelt – total length of fish (cm)

The effect of different feed types on size distribution was characterised with the proportions (H) of the coefficients of variation (CV) of the initial and final average body weights:
H = 100 x CV_t x CV₀ ^-1 (%)
Additionally the effect of different feeds on slaughtering proportion (SP) was calculated as proportion of obtained fillet weight (Fil) to life fish weight as follows:
SP = 100 x Fil x w_t ^-1 (%)
Treatment effects was examined for statistically significant differences in dw/dt; SGR, FCR, CF and H by one-way analysis of variance (one-way ANOVA).

During each week of the experimental period all groups received the same daily feeding rates, which were in the range of 2.5-3.0 % of the metabolic weight (kg ^0.8). These values were determined considering 1) the previously obtained production performance data; 2) overall health condition and behaviour of the fish; - and mainly - 3) to attain about 1 kg average body weight in each group at the end of the 8 weeks lasted treatments.

The study was conducted in accordance with the European Community guidelines on the care and use of experimental animals and according to the respective Hungarian rules (243/1998. (XII. 31.) Korm. rend. and 66/2009. (IV. 2.) Korm. rend.) and to the local institutional regulation (Regulation on Animal Experiments: Ig/258-3/2002). The experimental protocol was submitted to the Institute Committee of Animal Experiments at HAKI (Munkahelyi Állatkísérleti Bizottság – MÁB, Szarvas, Hungary).

Determination of fatty acid composition

Before lipid extraction all feed samples were grinded on laboratory mill with cooling (FOSS KnifetecTM 1095, Tecator technology, Denmark). Lipids were extracted from the samples of 5 grams by using cold extraction process, which involves mixing/homogenizing with chloroform:methanol mixture (2:1) and the extracts were purified according to the method by Folch et al. (Folch et al., 1957). Aliquots of 20 mg of total lipid samples were trans-esterified by method that uses 14% (w/w) boron trifluoride/methanol solution (Sigma Aldrich, MO, USA).
The fatty acid composition of the carp feed total lipids were determined in Folch’s extracts (Folch et al., 1957) after esterification in BF3-Methanol (Metcalfe and Schmitz, 1961). Obtained fatty acid methyl esters (FAME) samples were analyzed by gas chromatography in an Agilent 7890A system (Agilent Technologies, Santa Clara, CA, USA) equipped with flame ionization detector (FID) and auto-injection module for liquid samples. FAME samples were injected in n-heptane solution with a split ratio of 30:1 onto a fused silica capillary column (SP-2560, 100 m x 0.25 mm I.D., 0.20 µm; , SUPELCO, Bellefonte, NY, USA). Carrier gas was high purity Helium at a constant flow rate of 1.5 ml/min. Column temperature was programmed: 140 °C for 5 min - 3 °C/min up to 240 °C - hold for 10 min. Data acquisition and integration was done by using the software of Agilent 7890A system. The fatty acid methyl ester peaks were identified by comparison of retention times with retention times of peaks from standard fatty acid methyl ester mix (FAME Mix, Cat. no. 4-7885; SUPELCO, Bellefonte, NY) and with data from internal data library, based on previous experiments and fatty acid methyl ester determination on GC-MS. For quantification of the fatty acid (FA) composition, the FID detector signals were integrated and from the peak areas of FAME, the fatty acid composition was calculated and expressed as g FA per 100 g of FA according to method described by Karlović and Andrić (1996).

 

The fatty acid composition of the total lipids of the carp fillet samples were determined after esterification in t-buthyl-methyl-ether with trimethyl-sulfonium hydroxide /TMSH/. Trimethyl-sulfonium hydroxide reproducibly converts fatty acids of phospholipids and acyl-glycerols into fatty acid methyl esters (FAME) and the transesterification can be performed at room temperature (Müller et al., 1993). The obtained fatty acid methyl esters samples were analyzed by gas chromatography in a Thermo FOCUS GC system (Thermo Scientific, Part of Thermo Fisher Scientific Inc., Waltham, MA, USA) equipped with flame ionization detector (FID) and with AI 3000 autoinjector. FAME samples of 1 µl were injected with a split ratio of 100:1 onto a fused silica capillary column (TRACE TR-FAME GC Column – 60 m x 0.25 mm x 0.25 µm; Thermo Scientific). Carrier gas was high purity Helium at a constant flow of 20 cm/sec. Column temperature was programmed from 140 °C (hold for 5 min) - 4 °C/min up to 240 °C. Data acquisition and integration was done by using the software of Thermo FOCUS system.
SFA, MUFA, etc. fatty acid groups includes all of the identified FAs belonging to their groups.

Statistics

One-way analysis of variance (ANOVA Tukey-tests) was applied to explore significant differences between feed performance and fillet fatty acid composition data.

Results and discussion

Formulation and manufacturing of experimental feeds

Based on the literature data proving crucial role of long chain polyunsaturated fatty acids (LcPUFAs) in human metabolism and their relation to the prevention of cardiovascular diseases, the experimental feeds were planned to provide enriched levels of LcPUFAs in the carp flesh for human nutrition as well as for the planned animal model experiments. One sinking pellet and 4 extruded-expanded pellets supplemented with different oils were designed for using as complementary feeds in pond culture of common carp (Cyprinus carpio). Basic feed formulations were obtained by the linear programming with updated composition data using the SOLVER program of the EXCEL’s Tools (Microsoft Office Excel 2003 SP3) adapted to feed formulation (Csengeri, 1999). After evaluation of the manufacturing possibilities and commercialization aspects, a basic ingredient mixture was formulated which contained 36% of defatted soybean meal as major component (Table 2).

 

The calculated composition of basic formula was as follows: 34.25% crude protein, 2.79 % crude fat, and 3.86% crude fiber with lysine 1.92%, methionine 0.68%, meth+cys 1.2%.
Fish feed was produced on twin-screw extruder (Mu Yang MY 90, China) with a screw diameter of 85 mm, length-to-diameter ratio of 20:1, and maximum temperature of 135 °C. Extruder was equipped with differential diameter conditioner (DDC).

Ingredient	Percentage
Soybean meal	36
Corn	30
Wheat	10
Corn gluten	10
Fish meal (70 % of protein)	10
Yeast	3
Premix	1

The used linseed oil and fish oil were obtained from local Serbian market. The fortifying oils – Mortierella alpine oil, containing 40% arachidonic acid (ARA); eicosapentaenoic (EPA) and docosahexaenoic (DHA) ethyl esters – were supplied by PRAEVIDEO Kft. (Budapest, Hungary).
The amounts of the added oils were as follows: CF2: 6.0% Linseed oil (LSO); CF3: 3.5% LSO + 1.5% Arachidonic acid (ARA) + 1% DHA; CF4: 6.0% Fish oil (FO); CF5: 4.5% FO + 1.5% ARA. The control sinking feed (CF1) was supplemented with 6.0% Soybean oil (SBO). 
Addition of oil was done on laboratory vacuum coater (model F-6-RVC, Forberg International AS, Norway) with capacity of 6 L per batch.
The floating feeds were developed for inclusion of oils with high levels of long-chain polyunsaturated fatty acids. The applied extrusion technology involved a process of cooking steamed ingredients under pressure at elevated temperature.
The planned amounts of the added oils (CF2: 6.0% Linseed oil /LSO/; CF3: 3.5% LSO + 1.5% Arachidonic acid /ARA/ + 1% DHA; CF4: 6.0% Fish oil /FO/; CF5: 4.5% FO + 1.5% ARA.; control sinking feed, CF1: 6.0% Soybean oil /SBO/) were reflected in the fatty acid compositions of the developed floating feeds (Table 3). 
Raw material preparation plays an important role in the final production of extruded aquaculture feeds therefore preliminary measurement were made to determine the composition of the ingredients which were then finely grinded.
Advantages of the extrusion cooking process for aquaculture feed production include the control of pellet density, increased feed conversion rate, greater feed stability in water, and better production efficiency and versatility. Extrusion is gradually replacing the pelleting systems, where only heavy density sinking aquaculture feeds can be processed. The pellet-mill provides a very low cook on the formula such as 20-25% starch gelatinisation (gelatinisation starts at around 55-70°C (Coultate, 1989)), while the extruder has yielded cook value of more than 90% starch gelatinization (Kiang, 1999).

Feeds	CF1 Sinking control (SBO)		CF 2 Linseed oil (LSO)		CF 3 LSO+ ARA+DHA		CF 4 Fish oil (FO)		CF 5 FO+ ARA
Fatty acid	mean	±SD	mean	±SD	mean	±SD	mean	±SD	mean	±SD
14:0	0.43	±0.02	0.40	±0.03	0.47	±0.03	2.96	±0.13	2.70	±0.12
16:0	11.46	±0.14	8.06	±0.11	7.91	±0.16	12.72	±0.25	12.32	±0.20
16:1	0.59	±0.08	0.62	±0.07	0.56	±0.09	3.32	±0.09	2.97	±0.13
18:0	4.70	±0.11	3.43	±0.10	4.18	±0.04	3.08	±0.09	4.29	±0.22
18:1n9c	23.52	±0.23	22.85	±0.13	17.86	±0.20	24.06	±0.27	21.20	±0.20
18:2n6c	47.42	±0.20	32.83	±0.32	25.78	±0.23	19.85	±0.24	19.11	±0.19
18:3n3	7.08	±0.07	27.57	±0.24	17.17	±0.23	4.52	±0.14	2.62	±0.10
20:4n6	0.11	±0.04	0.16	±0.09	7.56	±0.23	0.47	±0.06	9.22	±0.19
20:5n3	0.36	±0.04	0.43	±0.09	1.09	±0.10	4.14	±0.19	3.66	±0.07
22:6n3	0.80	±0.07	0.97	±0.06	11.21	±0.23	7.32	±0.20	6.23	±0.09
SFA	17.43	±0.47	12.46	±0.36	13.76	±0.39	19.37	±0.55	20.70	±0.67
MUFA	26.14	±0.58	25.02	±0.47	20.71	±0.81	34.96	±1.03	30.39	±0.86
PUFA	56.43	±0.47	62.52	±0.91	65.53	±1.14	45.67	±0.98	48.91	±0.79
PUFA/SFA	3.24		5.02		4.76		2.36		2.36
∑ n-6	48.19	±0.29	32.99	±0.52	34.10	±0.58	26.86	±0.45	33.88	±0.53
∑ n-3	8.24	±0.18	28.97	±0.39	30.11	±0.56	17.85	±0.53	14.16	±0.26
n-6/n-3	5.85		1.14		1.13		1.51		2.39

Preconditioning at a moisture content of 25% for 150 seconds will be resulted in a degree of cook of 33%. Increasing the moisture level to 30% will be resulted in a degree of cook of 47%. These are typical moisture and associated degrees of cook levels achieved in the DDC cylinder (Kearns, 1989).  Floating feeds typically are in the density range of 320 to 400 grams per litre (Kearns, 1993). Floating feeds result in better feed conversion due to the fact that the feed consumption can be monitored and adjusted to decrease feed wastages. Sinking aquatic feeds are typically in the density range of 400 to 600 grams per litre. Sinking aquatic feeds are designed to feed slow eating bottom feeding species. Slow sinking aquatic feeds are typically in the density range of 390 to 410 grams per litre (Kearns, 1993). These types of feeds are being used in the salmon production, especially in the floating cages in the ocean. The slow sinking feeds then have a better chance to be consumed prior to their exiting the net bottom.

The 4 applied extruded feeds and the control sinking one (pellets size: 9 mm diameter) were fed to common carp specimens in parallel groups at constant water temperature with an average value of 23.9±0.3ºC, with an average oxygen saturation of 90±2 %. The measured water chemistry parameter (pH; KOI; NH4-N; NO2-N; NO3-N) remained in the optimal range for carp rearing.
The growth pattern of the fish was rather regular; continuous, non-interrupted weight increases were observed in all groups. During 7 weeks of the experiment weights were increased satisfactorily; by about 60-70 % per treatment relatively to the initial weight.

The production performance data are summarized in Table 5. By SPSS one-way analysis of variance (ANOVA test) significant differences were found between the values of total length (Lt), slaughtering proportion (SP) and daily growth rate (SGR), but not significant differences were revealed among the other parameters.

Feeds	CF1 Sinking control (SBO)	CF 2 Linseed oil (LSO)	CF 3 LSO+ ARA+DHA	CF 4 Fish oil (FO)	CF 5 FO+ ARA
wt (g)	959±36 a	984±55 a	933±21 a	1028±10 a	974±31 a
Lt (cm)	35.4±0.3 ab	36.0±0.1 ac	35.3±0.1 b	36.2±0.3 c	35.7±0.2 abc
CFt (%)	2.15±0.04 a	2.15±0.12 a	2.14±0.05 a	2.18±0.02 a	2.17±0.08 a
H (%)	197±32 a	216±60 a	200±63 a	206±23 a	204±14 a
SP (%)	39.5±1.1 ac	41.1±0.1 ab	41.9±0.5 b	38.9±0.3 c	42.2±0.6 b
SGR (%.day-1)	1.05±0.04 ab	1.06±0.09 ab	0.97±0.05 a	1.17±0.00 b	1.05±0.06 ab
FCR (g.g -1)	2.22±0.10 a	2.20±0.25 a	2.36±0.08 a	2.02±0.12 a	2.32±0.22 a

1 The initial weight, length and condition factor of the fish were 578±5 g; 32.4±0.3 cm, and 1.79±0.04 %, respectively.
2. Values are means ± SD of two replicates. Values by row with different letters are significantly different (P < 0.05).

The highest daily growth rate (SGR) was measured in group fed with fish oil based feed. However, - rather surprisingly – the slaughtering proportion (SP) was the lowest in this group. It is supposed that this phenomenon could be explained mainly with the fat deposition either in the body cavity, or in other internal organs (i.e. liver, kidney etc) – but the measurement of these parameters were not performed at the fish processing time.

There are few literature data regarding the protein content in common carp feeds for rearing them exclusively on dry diets. However, all these data suggest that feeds for intensive rearing of this species should have protein higher than 35%. Therefore, it seems that our feed feeds with around 32 % of crude protein can be applied as complementary feed in pond culture conditions in presence of natural food.
During the present experiment the fish were fed at fixed ration level at constant temperature of around 24 ºC. In the rearing season in fish ponds, in Hungarian and Serbian climate conditions, the water temperature is generally lower than this in April, May and late August and September, October.
In fish pond culture of carp usually ad libitum feeding is applied fitted to the varying water temperature. Furthermore, the natural food availability is uneven during the season (see for example Ruttkay, 2000), therefore, the beneficial effect of the floating feeds can be expected

The measured crude fat contents of the carp fillets were in the range of 6.4 to 11.1 % (Table 6) as expected from earlier observations (Csengeri et al., 2009; 2011). The fatty acid composition of fillets (Table 7) reflected that of the experimental feeds (cf. Table 3). The fatty acid composition, however, differs from the planned/expected values. The differences are probably due to the late start of carp feeding experiment: fish fed on traditional cereal feeding prepared themselves for overwintering with oleic acid deposition and the oleic acid diluted the deposited LcPUFA amounts. The protein to energy ratio of carp feeds lower then the optimal for intensive feeding, was planned for oil enrichment of carp fillets under practical pond conditions.

Feeds fed to fish	CF1 Sinking control (SBO)	CF 2 Linseed oil (LSO)	CF 3 LSO+ ARA+DHA	CF 4 Fish oil (FO)	CF 5 FO+ARA
Contents in the fillets (g/100 g)
Dry matter	29.98	30.80	30.71	29.65	30.93
Crude protein	17.66	18.31	17.23	18.24	17.33
Crude fat	8.10	11.15	10.84	6.40	7.12
Crude ash	1.16	1.09	1.10	0.97	1.10

Feed groups	CF1 Sinking control		CF 2 Linseed oil (LSO)		CF 3 LSO+ARA+DHA		CF 4 Fish oil (FO)		CF 5 FO+ARA
Fatty acid	mean	±SD	mean	±SD	mean	±SD	mean	±SD	mean	±SD
14:0	0.88	±0.02 a	0.91	±0.01 a	0.90	±0.01 a	1.73	±0.03 b	1.50	±0.08 c
16:0	16.73	±0.10 a	15.80	±0.10 a	16.33	±0.46 a	18.34	±0.25 b	17.62	±0.28 b,a
16:1	4.94	±0.05 a	5.56	±0.06 a	5.81	±0.30 a	6.95	±0.26 b	6.20	±0.59 a
18:0	5.25	±0.18	4.59	±0.18	4.74	±0.28	4.49	±0.13	4.66	±0.27
18:1n9c	41.04	±0.34	41.79	±0.12	41.89	±1.36	41.69	±0.28	42.60	±0.80
18:2n6c	21.17	±0.16 a	15.17	±0.08 b	13.96	±0.16 c	11.88	±0.04 d	11.49	±0.04 d
18:3n3	1.76	±0.04 a	7.28	±0.38 b	4.17	±0.10 c	1.22	±0.02 a,d	1.04	±0.04 d
20:4n6	0.75	±0.07 a	0.44	±0.00 b	1.85	±0.01 c	0.46	±0.03 b	2.26	±0.09 d
20:5n3 (EPA)	0.06	±0.06 a,c	0.19	±0.05 a,b	0.15	±0.00 a,c	0.31	±0.00 b,c	0.25	±0.03 c
22:6n3 (DHA)	0.21	±0.04 a	0.75	±0.07 b	2.34	±0.04 c	2.17	±0.21 c	1.88	±0.19 c
SFA	23.93	±0.04 a	22.24	±0.08 b	22.87	±0.21 a,b	25.53	±0.20 c	24.80	±0.57 c
MUFA	50.04	±0.35 a	51.61	±0.01 a,b	52.24	±0.95 b	55.06	±0.25 c	54.99	±0.34 c
PUFA	25.22	±0.27 a	25.07	±0.09 a	23.77	±0.28 b	17.46	±0.30 c	18.43	±0.47 d
PUFA/SFA	1.05	±0.01 a	1.13	±0.00 a,b	1.04	±0.02 a,c	0.68	±0.01 d	0.74	±0.04 d
∑ n-6	23.06	±0.25 a	16.36	±0.16 b	16.63	±0.14 b	13.09	±0.11 c	14.61	±0.07 d
∑ n-3	2.17	±0.02 a	8.71	±0.25 b	7.14	±0.14 c	4.38	±0.19 d	3.82	±0.40 d
n-6/n-3	10.65	±0.01 a	1.88	±0.07 b	2.33	±0.03 b	2.99	±0.11 c	3.84	±0.38 d
EPA+DHA fortification (%)	100		490		1280		750		720

1.Values are means±SD of two replicates. Values in rows with same letters are not significantly different (P<0.05).

Although, under the presented intensive rearing conditions, the expected high LcPUFA fortification levels were not well manifested, calculating the EPA+DHA content of the fillet from the data for crude fat (Table 6) and fatty acid composition (Table 7), 5 to 13 times enrichment can be seen in the fillets of fish fed on fortified feeds as compared to control feed fed group (Table 7).

When the fortified floating feeds will be applied along with sufficient levels of natural food in the fish ponds in a suitable feeding regime, the required enrichment of carp flesh will be achieved. For a suitable feeding regime we suggest to apply the complementary feeds to such rearing conditions, in which the natural food supply is high for the whole spring and summer, and the application of the long chain polyunsaturated fatty acid fortified feeds would start in mid of August and it would last until the water is “warm” enough for the fish to take food and till harvesting. The higher food conversion ratios (FCR) observed in this intensive rearing experiment (Table 5) will then be well compensated. Another improvement of the FCR would come from the control of dosage through the observable food consumption of the fish. Further experiments under pond fish conditions would confirm these expectations.

Conclusions

The formulations of the suitable ingredient mixture and details of the extrusion technology were designed to develop 4 floating carp feeds suitable for inclusion of high oil levels by vacuum infusion. The experimental feeds were manufactured and the oil supplementation was done and feed fatty acid composition well reflected the fortification. The experimental feeds were tested under intensive rearing conditions. Although the expected high LcPUFA fortification levels were not well manifested, however, the estimated EPA+DHA contents of the fillet were 5 to 12 times higher in the fillets of fish fed on fortified feeds than in those of fish fed on control feed.
Fortified floating feeds will be applied as complementary feed along with sufficient levels of natural food in the fish ponds in suitable feeding regimes. For a suitable feeding regime we suggest to apply the complementary feeds to such rearing conditions in which the natural food supply is high for the whole spring and summer, and the application of the long chain polyunsaturated fatty acid fortified feeds will start in mid of August and it would last until the autumn harvesting.

 

АCKNOWLEDGEMENTS

This work has been co-financed by the European Union through project HUSRB-1002-214-120 (“Research cooperation on developing innovative fish feed for promotion of healthy food in the region”) within the Hungary-Serbia IPA Cross-border Co-operation Programme.

 

Download full article PDF

DOWNLOAD PDF