Effects of dietary supplementation with krill meal on pigmentation and quality of flesh of rainbow trout (Oncorhynchus mykiss)

Abstract

Effects of administration of krill meal and synthetic astaxanthin during the finisher phase of the fattening cycle of rainbow trout on flesh pigmentation and quality traits were studied. The inclusion of krill meal increased the body weight and size and decreased the peri-visceral fat and visceral weight indices. The astaxanthin diet produced the highest accumulation of total carotenoids in the fillet compared to the krill meal diet: the difference was significant after 15 days of feeding (2.50 vs 2.10 mg/kg) till the end of the trial (5.00 vs 4.80 mg/kg). The same pattern was observed for astaxanthin concentration with the highest values in the fillets of fish fed the astaxanthin diet. Fillet lightness (L*) was not affected by trout diets whereas redness (a*) and yellowness (b*) were significantly higher in fish fed the astaxanthin diet until day 30 of the trial. Hue was not affected by feeding, whereas chroma was significantly higher in the fish fed astaxanthin throughout the trial except on day 45 of sampling. Trout fed the krill meal diet had a paler pink-red colour on the SalmoFan scale than those receiving the astaxanthin diet. No significant differences emerged in proximate composition and cholesterol content of trout in the two groups. The fatty acid profile of the fillets reflected the fatty acids of the diets administered to the trout: eicosapentaenoic, docosa-hexaenoic and docosapentaenoic acids and total n-3 polyunsaturated fatty acids were significantly higher in the fish fed the krill meal.

Key Words:

• Krill meal
• Astaxanthin
• Fatty acid profile
• Proximate composition
• Rainbow trout.

Introduction

The concentration of carotenoid pigments in salmonid muscles has been studied for many years with the aim of improving the efficiency of the pink-red colouration of the flesh and comparing the colour between wild and farmed salmon (Storebakken et al., 1987; Torrissen, 1989; Tibaldi and Ballestrazzi, 1990; Torrissen et al., 1995; Nickell and Springate, 2001). In fact, in farmed salmonids, flesh colour is the most important quality trait together with freshness and nutritional value of the product (Anderson, 2000). Synthetic astaxanthin and canthaxanthin are still the most important sources of pigments used in the feeding of farmed salmonids because of their high pigmentation ability of both skin and flesh. However, in the last few years the use of synthetic pigments has been discussed in relation to the possible damage to human health deriving from the consumption of fish rich in these synthetic colouring additives (Anderson et al., 1997; Georgakopoulos and Thomson, 2005). In a review on carotenoid deposition and pigmentation regimens (Nickell and Springate, 2001), 40–60 mg/kg astaxanthin proved the most efficient dietary dose for optimal pigment retention in Atlantic salmon while in freshwater rainbow trout the level ranged from 50 to 70 mg/kg astaxanthin.

Although the European Food Safety Authority has indicated that the current approved practice of feeding canthaxanthin to layers, meat-type chickens and salmonids does not raise safety concerns for the consumer (EFSA, 2007), many farmers have decided to adopt guidelines that forbid the use of synthetic pigments in feeds. Moreover, the law currently in force regarding organic fish products (EC Regulation No. 710/2009) only recognises the use of astaxanthin as a natural pigment in the feeds to colour fish muscle instead of synthetic additives. In this situation, studies on natural sources of astaxanthin able to favour flesh pigmentation for an alternative pigmentation of salmonids have increased. Pigments such as shrimp waste (Diler and Gökoğlu, 2004), red yeast Phaffia rhodozyma (Gentles and Haard, 1990; Bjerkeng et al., 2007), green algae Haematococcus pluvialis (Choubert and Heinrich, 1993) and red pepper and marigold flower (Yanar et al., 2007) have been tested with different results. The use of krill, including around 85 species of pelagic crustaceans belonging to the order Euphausiacea, in feeds to colour different salmon species has also been investigated (Storebakken 1988; Scott et al., 1994), in addition to considering the possibility of using krill meal to replace fish meal (Moren et al., 2006; Yoshitomi et al., 2007). Analysing the nutritional value of frozen Antarctic krill, Gigliotti et al. (2008) found a content of approximately 78% protein and 8% fat, whereas n-3 polyunsaturated fatty acids (PUFA) amounted to 27% of the total fatty acids. In overweight and obese men and women, krill supplementation increased plasma concentrations of eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids, without adverse effects on tolerability and safety when it was compared with menhaden and olive oil (Maki et al., 2009). The purpose of our study was to investigate the effect of the dietary inclusion of krill meal, during the finisher phase of the fattening cycle, on the fillet pigmentation of rainbow trout. Fillet colour, astaxanthin and carotenoid concentrations, growth performance and flesh quality traits, such as proximate composition, fatty acid profile and cholesterol content, were compared with those of trout receiving a commercial feed containing synthetic astaxanthin.

Materials and methods

The trial was performed at a trout farm located in Sefro (Italian Central Apennine). A total of 4400 rainbow trouts (Oncorhynchus mykiss), weighing 453±57 g (12 months old) and measuring 34±2 cm, were stocked in two 100-m3 raceways at the same stocking density (22 fish/m3). The trial lasted 60 days during which the main physicochemical water parameters were monitored regularly and nitrogen and phosphorous compounds analysed at the laboratory following APHA standard methods (APHA, 1989).

Feeds and feeding management

Two isonitrogenous diets were used. The experimental diet (KR), formulated with GMO-free certified ingredients, included krill meal obtained from the Euphasia superba species as the protein and pigment source. The chemical composition of this ingredient is reported in Table 1. The fluoride content in the krill meal (130 mg/kg) was under the maximum limits (150 mg/kg feed) set by Annex I to Commission Directive 2003/100/EC. Commercial feed (AS) was based on soybean extracted meal, fish meal and wheat meal as the main ingredients and contained astaxanthin (90 mg/kg). The feed-stuffs, the proximate composition and the fatty acid profile of the two diets are reported in Tables 2 and 3. The feeds were administered to the trouts at the daily rate of 1% body weight, according to water temperature and fish weight.

Table 1 Chemical composition, pigment concentration and categories of fatty acids evaluated in the krill meal.

Moisture, %	10
Protein, % wet weight	57.8
Lipids, % wet weight	18.1
Ash, % wet weight	13.2
Astaxanthin, mg/kg	150
Saturated fatty acids, % of total fatty acids	34.9
Monounsaturated fatty acids,
% of total fatty acids	27.1
Polyunsaturated fatty acids,	38
% of total fatty acids
n-3 polyunsaturated fatty acids,	36.4
% of total fatty acids

Table 2 Ingredients, proximate composition of the finisher diets administered to rainbow trout in the last 60 days of the rearing cycle.

AS	KR
Ingredients
Krill meal, %	-	60
Soybean extracted meal, %	40	-
Fish meal, %	34	5
Corn gluten, %	-	10
Wheat meal, %	14	8
Gelatinised wheat starch, %	6	5
Fish oil, %	2.5	5
Soybean oil, %	2.5	-
Whey powder, %	-	2.5
Wheat, %	-	3.5
Mineral vitamin premix, %	1	1
Proximate composition
Moisture, %	8.0±0.2	7.0±0.2
Protein, % wet weight	44.1±0.9	44.3±0.6
Lipids, % wet weight	19.9±1.2	18.3±0.9
Ash, % wet weight	10±0.5	9.6±0.3
Pigment concentration,	90	90
mg/kg feed

AS, astaxanthin diet; KR, krill diet.

Table 3 Fatty acid profile of the finisher feeds administered to rainbow trout in the last 60 days of the rearing cycle.

Diets	AS	KR
n	3	3
Fatty acid profile, % of total fatty acids
Saturated
C 14:0	4.60±0.02	11.70±3.60
C 15:0	0.30±0.03	0.50±0.04
C 16:0	14.70±1.10	20.30±1.80
C 17:0	0.50±0.01	1.80±0.04
C 18:0	4.00±0.03	2.10±0.06
C 20:0	0.40±0.02	0.20±0.04
C 21:0	0.10±0.03	0.10±0.03
C24:0	0.10±0.01	0.00±0.00
Total saturated	24.80±1.30	36.70±1.60
Monounsaturated
C 14:1	0.20±0.03	0.30±0.07
C 15:1	0.05±0.01	0.10±0.02
C 16:1	5.10±0.06	8.10±0.04
C 17:1	0.80±0.02	1.00±0.02
C 18:1	19.40±0.12	19.80±0.40
C 20:1	1.10±0.02	2.10±0.03
C 24:1	0.40±0.02	0.40±0.01
Total monounsaturated	27.10±2.02	31.80±1.90
Polyenoic (n-3)
C 18:3 n-3	1.02±0.01	1.83±0.04
C 20:5 n-3, EPA	6.83±0.57	9.02±1.60
C 22:5 n-3, DPA	1.32±0.04	0.60±0.02
C 22:6 n-3, DHA	4.01±0.72	6.56±1.30
Total n-3	13.17±1.31	18.01±2.80
Polyenoic (n-6)
C 18:2 n-6	27.33±0.06	4.67±0.03
C 18:3 n-6	3.80±0.02	1.01±0.01
C 20:2 n-6	0.10±0.02	0.04±0.01
C 20:3 n-6	0.66±0.03	2.01±0.01
C 20:4 n-6	0.12±0.01	1.51±0.02
Total n-6	32.00±0.09	9.23±0.02
Total polyunsaturated	45.17±1.40	27.24±2.70
n-3/n-6	0.41±0.04	1.95±0.02
Others	2.90±0.03	4.17±0.02

AS, astaxanthin diet; KR, krill diet.

Colour measurements, carotenoid and astaxanthin content

Every 15 days, 10 fish per group were sacrificed for the colour measurements and pigment content of the flesh, whereas the proximate composition and lipid profile of the fillet were determined at the end of the trial on 10 trouts per group. Immediately post-slaughter, muscle colour was assessed by a reflectance colourimeter (Minolta mod. Chroma Meter II CR-300, Minolta Italia SPA, MI, Italy) using the illuminant source D65 and 10° standard observer, set on multimeasure mode. Measurements were carried out according to the recommendations of the International Commission on Illumination (CIE, 1978), reporting values of lightness (L*; negative value: blackness, positive value: whiteness), redness (a*; negative: greenness, positive: redness) and yellowness (b*; negative: blue-ness, positive: yellowness). Readings were obtained at four positions along the left fillet, of which two were carried out above the lateral line and the other two below the lateral line. The averages of the three measurements were calculated to determine the L*, a* and b* values. Values of chroma (C*) and hue (H*) were also computed to determine the saturation and main hue of the fillet, respectively, according to the following formulae:

In addition, the colour of the trout fillets was assessed by comparison using the DSM SalmoFan (DSM Nutritional Products, Parsippany, NJ, USA), which adopts a scale ranging from 21 (light red) to 34 (dark red), under standardised conditions, employing a light cabinet fitted with a cool, daylight fluorescent light source. Three measurements of the colour of each fillet were carried out in the same position as those measured by means of the Minolta Chroma Meter instrument. This method was used because it is the easiest to apply and the best known and used by farmers. The left fillets were immediately stored in black plastic bags at −30°C for subsequent analysis. The carotenoid concentration and astaxanthin content in the diets and salmon fillets were determined chromatographically. Skinned fillets were ground with a meat mincer and frozen. Two samples of each fillet were accurately weighed (10±0.2 g) and repeatedly extracted with acetone (40 mL) until they were colourless. The pooled extracts were filtered and an aliquot (10 mL) was centrifuged at 1800 g for 5 min). The astaxanthin content in the supernatant was determined by means of a Varian ProStar instrument, equipped with a UV/vis detector, using an external astaxanthin standard at the detection wavelength of 470 nm. Analysis was performed on a Varian Kromasil 100 C18 250×0.3 mm column, according to Bjerkeng et al. (1997).

Proximate composition, cholesterol and fatty acid profile of the fillet

At the end of the trial a portion of about 50 g of skinless, left dorsal muscle was collected from each fish (10 fish per group), homogenised and submitted to proximate analysis (moisture, protein, lipid and ash content). The percentage of moisture was determined in duplicate according to the Association of Official Analytical Chemists procedure (AOAC, 1990). Proteins were determined using the standard Kjeldahl copper catalyst method (AOAC, 1990). Ashes were determined using the procedure described by the AOAC (AOAC, 1990). Total lipids were measured using a modification of the chloroform:methanol procedure described by Folch et al. (1956). After the determination of total lipids, fatty acids were converted to methyl esters following the method described by Christopherson and Glass (1969). The separation of fatty acids was carried out by using a Carlo Erba HRGC 5160 gas chromatograph (Carlo Erba Strumentazione, Rodano, MI, Italy) with a WP-4 Shimadzu integration system (Shimadzu Corporation, Tokyo, Japan), equipped with a Supelco SP™ – 2340 capillary column (30 m×0.32 mm i.d.; 0.20 μm film thickness; Supelco, Bellefonte, PA, USA) and a flame ionisation detector. The operating conditions of the gas chromatograph were as follows: oven temperature was kept at 170°C for 15 min, increased to 190°C at a rate of 1°C/min, then increased to 220°C at a rate of 5°C/min and kept at this temperature for 17 min. The temperature of the injector was 270°C and the temperature of the detector was 300°C. Helium was used as the carrier gas at a constant flow of 1.7 mL/min. The identification of individual fatty acids was accomplished by a comparison of retention times to fatty methyl esters of standard mixtures (37 FAME Mix and C22:5 n3, Supelco). Cholesterol determination was carried out on the total lipid extract, following the method described by Manzi et al. (1996), by using a Shimadzu HPLC with a Shimadzu SPD-M10A Diode Array Detector equipped with an Alltech Nucleosil C18 column (150 mm length; 4.6 mm i.d.; 5 μm thickness; Alltech Italia Srl, Sedriano, MI, Italy). The mobile phase was methanol: water (97:3, vol:vol) at a flow rate of 1.5 mL/min. The quantification of total cholesterol content was obtained by using an external calibration curve of β-sitosterol (Sigma-Aldrich, St. Louis, MO, USA).

Statistical analysis

The effect of the diet on the colour coordinates (L*, a*, b*, C*, H*), carotenoid and astaxanthin concentration, proximate composition and fatty acid profile was analysed by one-way analysis of variance (ANOVA) using the General Model procedure of SAS (SAS, 1988). Differences were considered significant at P<0.01 and means were separated by the Student Newmann-Keuls test.

Results

The physicochemical parameters of the water controlled in the two raceways always stayed within the range considered optimal for salmonids: water temperature was 14±2°C; dissolved oxygen was maintained above 8 ppm; pH ranged around 7.98±0.5; nitrogen compounds stayed below 0.40 mg/L (N-NH3), 0.001 mg/L (N-NO2) and 0.5 mg/L (N-NO3); phosphates were around 0.08 ppm throughout the trial.

Diet composition

The proximate analysis of the two feeds is shown in Table 2: the compositions of the diets were very similar and they were in agreement with the requirements estimated by the National Research Council (Bromage and Shepherd, 1988). The fatty acid profile showed relevant variations particularly for the categories of saturated and polyunsaturated fatty acids (Table 3). The KR diet had a higher proportion of saturated fatty acids (+50%) than the AS diet owing to the higher proportions of myristic (C14:0) and palmitic acids (C16:0) (11.7 and 20.3% vs 4.63 and 14.70%, respectively). The total n-3 PUFA proportion was much higher in the KR diet owing to the higher proportions of both EPA and DHA, and consequently the n-3/n-6 PUFA ratio was more than four times higher in the KR diet compared to the AS diet (1.95 vs 0.41). The total PUFA proportion and n-6 PUFA noticeably increased in the AS diet because of the very high content of linoleic acid (C18:2 n-6): 27.33% vs 4.67%. This may be attributed to the use, in the feed formulation of the AS diet, of soybean meal and oil, notoriously rich in linoleic acid and lacking in n-3 PUFA, in the replacement of fish meal and fish oil.

Colour measurements, carotenoid and astaxanthin concentration

The AS diet produced the highest accumulation of total carotenoids in the fillet compared to the KR diet; the differences between the two groups were significant after 15 days of feeding (2.5 vs 2.1 mg/kg) till the end of the trial (5.8 vs 4.8 mg/kg). The same pattern was observed for the astaxanthin concentration with the highest values in the fillet of fish fed the AS diet (Table 4).

Table 4 Content of total carotenoids (mg/kg) and astaxanthin (mg/kg) in the fillet of rainbow trout fed the two finisher diets at different sampling times.

	Total carotenoids		Astaxanthin
	AS	KR	AS	KR
fish/sampling, n	10	10	10	10
Day 0	1.00±0.03	1.00±0.03	0.80±0.02	0.80±0.02
Day 15	2.50±0.01^A	2.10±0.02^B	2.30±0.06^A	1.90±0.02^B
Day 30	3.70±0.02^A	2.30±0.06^B	3.60±0.13^A	2.10±0.07^B
Day 45	4.70±0.01^A	3.50±0.04^B	4.20±0.06^A	3.30±0.05^B
Day 60	5.80±0.03^A	4.80±0.02^B	5.40±0.01^A	4.30±0.03^B

AS, astaxanthin diet; KR, krill diet;

A,B different letters on the same line show significant differences (P<0.01).

The colour coordinates of fillet are reported in Table 5. Fillet lightness (L*) was not affected by the trout diets whereas redness (a*) and yellowness (b*) were significantly higher in fish fed the AS diet until day 30 of the trial. Hue was not affected by the feeding treatment, whereas chroma was significantly higher in AS-fed fish throughout the trial with the exception of day 45 sampling. From the evaluation of fillet colour using the SalmoFan scale, it emerged that trout fed the diet containing krill meal had a paler pink-red colour than those receiving the diet containing synthetic astaxanthin.

Table 5 Evaluation of fillet colour of rainbow trout fed on the two feeds through the CIE system and SalmoFan at the different sampling times.

L*	a*	b*	Hue	Chroma	SalmoFan
Day 0	41.6±2.3	0.6±1.1	3.0±1.5	0.3±0.03	4.1±1.16	19.5±0.5
Day 15
AS	42.3±1.6	3.2±1.6^A	2.6±0.2^A	0.8±0.3	4.3±1.3^A	22.8±0.06^A
KR	42.1±1.2	1.0±0.6^B	1.1±1^B	0.9±0.3	1.2±0.09^B	20.1±0.05^B
Day 30
AS	41.2±1.5	4.1±0.4^A	4.9±0.8^A	0.9±0.1	6.4±0.4^A	24.7±0.04^A
KR	42.1±3.5	2.9±0.7^B	2.0±0.6^B	0.6±0.2	3.6±0.6^B	22.6±0.08^B
Day 45
AS	41.6±1.2	5.8±1.7	5.5±2.1	0.8±0.3	8.2±1.6	28.6±0.06^A
KR	42.5±2.7	4.9±1.6	5.0±1.3	0.8±0.1	6.9±1.3	26.4±0.07^B
Day 60
AS	42.4±1.9	6.7±1.5	6.2±2	0.7±0.1	10.5±1.3^A	30.8±0.03^A
KR	43.6±2.8	5.7±1.4	5.9±1.6	0.8±0.2	7.6±1.3^B	27.7±0.09^B

CIE system, L*= lightness, a* = redness and b* = yellowness;

A,B different letters on the same line show significant differences (P<0.01).

Proximate composition, cholesterol and fatty acid profile of the fillet

No significant differences emerged regarding proximate composition and cholesterol content of the trout in the two groups (Table 6). The profile of the most important fatty acids of the rainbow trout, evaluated after 60 days of feeding with the two diets, is shown in Table 7. Total saturated fatty acids, primarily represented by palmitic acid (16:0), were significantly higher in the fillets of the KR-fed fish than in the AS-fed ones. The proportion of total monounsaturated fatty acids was similar in the fillets of the two groups. The percentage of n-6 PUFA in the AS-fed trout was twice that of the KR-fed trout (19.8 vs 9.8) owing to the higher proportion of linoleic acid (C18:2) (14.6% vs 6.7%). On the contrary, the proportions of EPA, DPA, DHA and total n-3 PUFA were significantly higher in KR-fed fish. The ratio n-3/n-6 PUFA was also markedly higher in trout fed the KR diet than fish fed the AS diet. In general, the fatty acid profile of the fillets reflected the fatty acid profile of the diets administered to the trouts.

Table 6 Proximate composition and total cholesterol amount in fillets of rainbow trout fed the two finisher feeds.

AS	KR
Moisture, g/100 g	76.6±0.5	76.3±0.3
Protein, g/100 g	19.4±0.6	19.5±0.2
Lipids, g/100 g	3.3±0.5	3.5±0.4
Ash, g/100 g	1.3±0.2	1.3±0.1
Cholesterol, mg/100 g	57.0±0.8	56.7±5.8

AS, astaxanthin diet; KR, krill diet.

Table 7 Fatty acid profile of fillets of rainbow trout fed the two finisher feeds.

	AS	KR
Fatty acid profile, % of total fatty acids
Saturated
C 11:0	0.02±0.04	0.03±0.03
C 12:0	0.01±0.02	0.02±0.03
C 14:0	4.80±0.10	5.40±0.20
C 15:0	0.20±0.08	0.40±0.01
C 16:0	15.60±0.30^B	18.50±0.10A
C 17:0	0.40±0.10	0.50±0.02
C 18:0	3.10±0.06	4.30±0.02
C 20:0	0.50±0.10	0.40±0.04
C 21:0	0.90±0.03^A	0.30±0.02^B
C 24:0	0.30±0.30	0.00±0.00
Total saturated	25.80±0.30^B	29.90 ±0.10^A
Monounsaturated
C 14:1	0.10±0.01	0.10±0.10
C 15:1	0.00±0.00^B	0.09±0.01^A
C 16:1	3.60±0.60^B	6.70±0.10^A
C 17:1	0.50±0.10^B	0.90±0.05^A
C 18:1	23.60±1.50^A	19.30±0.20^B
C 20:1	1.70±0.70	1.30±0.02
C 24:1	0.20±0.10^B	0.60±0.02^A
Total monounsaturated	29.70±2.40	29.10±0.30
Polyenoic (n-3)
C 18:3 n-3	2.70±0.40^A	0.50±0.06^B
C 20:5 n-3, EPA	4.80±0.30^B	8.90±0.70A
C 22:5 n-3, DPA	1.50±0.20^B	3.50±0.40^A
C 22:6 n-3, DHA	13.50±0.80^B	16.40±1.00^A
Total n-3	22.40±1.50^B	29.40±1.02^A
Polyenoic (n-6)
C 18:2 n-6	14.60±0.50^A	6.70±0.09^B
C 18:3 n-6	2.50±0.50^A	1.00±0.06^B
C 20:2 n-6	0.70±0.00^A	0.20±0.02^B
C 20:3 n-6	1.30±0.20^A	0.90±0.02^B
C 20:4 n-6	0.70±0.20^B	1.00±0.03^A
Total n-6	19.80±0.50^A	9.80±0.10^B
Total polyunsaturated	42.20±1.80^A	39.20±0.10^B
n-3/n-6 ratio	1.10±0.07^B	3.00±0.10^A

AS, astaxanthin diet; KR, krill diet;

A,B different letters on the same line show significant differences (P<0.01).

Discussion

In our study, the use of krill meal in the feed for pigmenting the fillet of the rainbow trout was investigated in the last 60 days of the rearing cycle. Trout receiving feed with krill meal had a paler pink-red colour than those receiving commercial astaxanthin. This difference was previously reported by Storebakken (1988), who explained this discrepancy as the form of krill astaxanthin, which is mostly in the esters of fatty acids, whereas salmonids utilise esterified astaxanthin to a lesser degree than free astaxanthin. On the contrary, Suontama et al. (2007) obtained fish with a deep red flesh colouration when the fish-meal protein of the control diet was replaced with 20% of the northern krill species (Thysanoessa inermis). In our experiment, following the application of the L*, a*, b* system, the L* parameter appeared the least suitable among the chromatic indices to highlight differences in the intensification of colour in the muscle of the rainbow trout. In this salmonid species, colourimetric measurements showed large variations in fillet colour, irrespective of dietary treatment: L* values ranged from 35 to 63, a* from 5 to 26 and b* from 7 to 31. Large variations in colour measures were found by Schubring (2009) who found that the L* measure is significantly correlated with flesh lipid content. On the contrary, according to Christiansen et al. (1995), the high fat content in farmed salmon causes dilution of astaxanthin and interferes with colour perception. In our experiment, we did not find any correlation between colour measurements and fat content of the flesh. Other papers reported that muscle colour attributes were more affected by the astaxanthin source than by the oil source; this was reported by Choubert et al. (2006) who ascertained that trout fed on synthetic astaxanthine showed higher chroma (C*), redness and yellowness than fish fed on feed supplemented with astaxanthin from the green micro-algae Haematococcus pluvialis. Using other natural pigmenting compounds, such as red pepper, in comparison with commercial astaxanthin, Yanar et al. (2007) observed that the fillet colour was lighter. Similar results were obtained by Akhtar et al. (1999), who observed that paprika, compared with canthaxanthine, produced less pigmentation of the muscle with higher values in lightness and redness.

In our trial, the fillet colour measured by means of SalmoFan showed significantly higher scores in fish receiving the synthetic astaxanthine. In Atlantic salmon, SalmoFan was able to score an increase in colour when increasing proportions of canthaxanthin were added to the feed compared with astaxanthin (Buttle et al., 2001). In the same species, some interesting studies showed a significant linear correlation between the SalmoFan score and muscle astaxanthin concentration (Christiansen et al., 1995). A study carried out by Forsberg and Guttormsen (2006) modelled the relationships between muscle astaxanthin concentration and visual colour perception (SalmoFan score), using the explanatory variables dietary astaxanthin concentration and fish size, and showed that the two dependent variables gave an excellent fit to the observed data. In another experiment (Misimi et al., 2007), the comparisons between fillet colour evaluated through visual measures by a panel of human inspectors, according to the SalmoFan standard, and the colour scores generated from a computer vision algorithm confirmed that there were no significant differences between the two methods. However, the SalmoFan method proved unsuitable when farmed and wild salmon were compared, as observed by Johnston et al. (2006) who did not to find differences between farmed and wild salmon: flesh colour measured with SalmoFan ranged between 27.4 and 28.0, even if total carotenoid pigment concentrations were considerably higher in farmed (7.68–8.43 mg/kg) than in wild (5.53–6.44 mg/kg) salmon.

In relation to carotenoid pigments and astaxanthin content, although in our experiment fish receiving natural pigments exhibited a paler muscle colour than those fed a diet containing synthetic astaxanthin, the amounts of carotenoids and astaxanthin contained in the diet were considered to be adequate for the pigmentation of the rainbow trout. In other studies (Yanar et al., 2007), carried out using natural pigment compounds such as red pepper and marigold flower, the carotenoid contents found in the fish fillet ranged from 1.5 to 5.6 mg/kg depending on the application time, but synthetic astaxanthin provided the highest carotenoid accumulation in the fish followed by red pepper. In the rainbow trout, canthaxanthin was more efficiently absorbed (3.8–7.9 mg/kg) in the flesh than the paprika carotenoids (2.4–3.1 mg/kg) (Akhtar et al., 1999). In Atlantic salmon, the final total carotenoid concentration of the muscle was significantly higher in fish fed the diet supplemented with P. rhodozyma (2.56 mg/kg) compared to salmon fed a diet supplemented with a source of chemically synthesised astaxanthin (1.96 mg/kg) (Bjerkeng et al., 2007).

As regards the quality traits of the fillets, we found a very similar proximate composition between trout receiving the experimental and commercial feeds, showing it to be affected by the dietary composition as documented in the literature (Rasmussen, 2001). The protein content was nearly equal to that reported for rainbow trout at market size, whereas the lipid content appeared lower than data reported both by Yanar et al. (2007) and by Chaiyapechara et al. (2003); the latter ascertained a proportion of 8.8% of lipids in the trout fed on a commercial diet based on a 17% lipid content.

Krill meal provided a high quality product from the nutritional point of view, because it improved the fatty acid profile. The trout fed the diet based on krill meal showed a notable content of essential n-3 PUFA. The fatty acid profile showed a very high proportion of EPA and DHA, and a low rate of linoleic acid with a favourable n-3/n-6 PUFA ratio. These results appeared more satisfactory than those reported by Yanar et al. (2007) who, using a basal diet containing 45% crude protein, 10% crude fat, supplemented with commercial astaxanthin and natural pigmenting compounds, obtained EPA levels ranging from 5.73% (red pepper) to 5.95% (commercial astaxanthin). Studies carried out in rainbow trout reared in freshwater and seawater cages (Haliloĝlu et al., 2004) also reported lower EPA contents and PUFA than those obtained in our experiment. This demonstrates that krill meal has a satisfactory nutritional value and also appears to be useful for inclusion in the diet for pigmenting rainbow trout.

Conclusions

In this paper, krill meal was used for pigmenting the trout fillet; even if the pigmentation was lighter pink, this colour is appreciated by the consumers of organic products who are adverse to the use of synthetic products in animal feeding. Moreover, krill meal enhanced the flesh amount of n-3 PUFA considered beneficial for human health. Krill meal could therefore represent an alternative to synthetic sources for flesh pigmentation in organic aquaculture, as contemplated in the recent EC Regulation No 710/2009 of 5 August 2009 (European Commission, 2009) that admits the use of astaxanthin derived primarily from organic sources, such as organic crustacean shells, in the feed ration for salmon and trout within the limit of their physiological needs.

This consideration is in agreement with the guidelines regarding organic production, as reported by a survey of the European Commission in 2000, as well as with the relationship between organic fish feeds and environmental sustainability, focused on during the International Federation of Organic Agriculture Movements (IFOAM) World Congress (Bridson, 2008). Moreover, considering that the krill biomass worldwide production is estimated to be as high as 500 million tons, and still represents an unexploited source (only about 12% of it is consumed) (Ichii, 2000), it appears to be a valuable ingredient in organic fish feeding and particularly in trout production.

Acknowledgements:

the authors wish to thank Mr. Roberto Rossi and the staff of Eredi Rossi (Sefro, Macerata) for the maintenance of the trout during the trial.