ADVANCES IN THE SUBSTITUTION OF FISH MEAL AND SOYBEAN MEAL BY SUNFLOWER MEAL IN DIETS OF RAINBOW TROUT (SALMO GAIRDNERI L.)

Trabajo recibido el 23 de agosto de -1984 y aceptado para su publicación el 11 de diciembre de 1984.

RESUMEN

INTRODUCCIÓN

Traditionally fish meal has been the main source of dietary protein for fish, but its high cost per tonne and its short supply on the world market, have made increasingly necessary to identify other suitable protein sources, for inclusion as partial or total replacements for fish meal. A wide variety of feedstuffs of plant and animal origin have been successfully used as partial dietary replacements for fish meal for use within commercial fish feeds.

It has been possible to substitute 50 % of the dietary fish meal component with soybean meal in trout feed, with additional exogenous amino acids to supplement the amino acid deficiencies of soybean meal (Reichle and Wunder, 1974; Runisey and Ketola, 1975; Dabrowska and Wojno, 1977).

In common with other plant oil seeds soybean does contain a wide variety of antinutritional factors that must be removed or inactivated before it is suitable as a feed for monogastric animals, including fish (Smith, 1977). Salmonids have been found to be extremely sensitive to soybean inhibitors (Sandholm et al., 1976) and cyprinids have also been shown to be affected in a similar manner (Dabrowski and Kozak, 1979). Smith (1977) and Fowler and Banks (1976) found that incorporation of 20% commercially defatted soybean meal in fish feeds resulted generally in poor growth and high mortalities. In contrast the studies of Reinitz et al. (1978) and Tacon et al. (1983) with rainbow trout have shown that soybean can adequately replace up to 75 % of the dietariy fish meal protein witn no deleterious effect on growth or feed conversion efficieney. The failure and success of soybean meal in the different published papers may be possibly due to inadequate heat treatment of the soybean meal used, and its consequent failure to destroy the heat labile antinutritional factors present (Tacon et al., 1983). Soybean meal is also deficient in avalilable energy, and certain essential amino acids, including methionine, lysine, histidine, leucine and cystine (Rumsey and Ketola, 1975; Fowler, 1980; Smith, 1977; Viola et al., 1982; Tacon et al., 1983). In conclusion soybean has disproportionate levels of amino acids which in turn may cause a reduction in growth rate, unless corrected by the addítion of specific exogenous amino acids in the feed (Harper et al., 1970). Dabrowska and Wojno (1977) has shown that the utilization of soybean enriched with cystine (1 %) and tryptophan (0.5 %) is almost equivalent to that of fish meal protein for trout. Other factor such as mineral supplementation have also, been shown to influence the nutritive value of soybean for fish (Dabrowski and Kozak, 1979; Tacon et al., 1983).

Sunflower seed meal contains 32.3 % crude protein (Jackson et al., 1982) and despite a high crude fibre content has a good amino acid profile and contains no known toxic or antinutritional substances (Daghir et al., 1979; Jackson et al., 1982). Sunflower is also a rich source of oil (14-25%) of good nutritional quality (Daghir et al., 1979).

There is a paucity of information concerning the use of sunflower meal as a dietary ingredient of fish. The only published work to date, is the feeding trial undertaken by Jackson et al. (1982) where sunflower meal was examined in relation to other plant protein sources for tilapia (Sarotherodon mossambicus).

The aim of the present study was to compare the nutritive value of oil extracted sunflower meal in five diets, as a partial replacement for soybean meal or fish meal, alone or with exogenous amino and supplementation, for fingerling rainbow trout (Salmo gairdneri).

Five isocalorific and isonitrogenous practical diets were formulated containing varying dietary concentrations of brown fish meal, solvent extracted soybean meal, sunflower meal and wheat meal (Table 1). The composition of the experimental diets is shown in Table 1. A control diet was formulated in which brown fish meal was used as the major protein source (40 % fish meal) and containing 15 % soybean. Two rations in which sunflower meal replaced 50 % and 100 % of the dietary soybean meal (11 % and 22 % sunflower meal); two rations in which the dietary sunflower inclusion level was increased to 37.5 % and 37.3 % replacing dietary soyhean meal and reducing the brown fish meal inclusion level from 40 % to 35 %, alone and with supplemental dietary L-methionine. For the determination of apparent digestibility coeficients chromic oxide was included within all diets at 0.5 % (Furukawa and Tsukahara, 1966). All diets were formulated to contain 44 % crude proteín and 14 % lipid. Experimental diets were prepared by first mixing all the dietary ingredients thoroughly in a Hobart A200 food mixer. Feed pellets were air dried by convection at 35°C. The experimental diets were then stored in air-tíght containers at ambient temperature until fed. Experimental fish were housed in 15 outdoor 40 1 circular fiberglass tanks. Each rearing tank was continously supplied with artificially aerated (Venturiaspirator) tap water at a rate or 2 l/min from a 5 m³ capacity header tank. During the experiment the water temperature varied between 4.3 to 6.3°C (± S. D. 0.58; x = 5.27°C), and fish were subjected to a natural photoperiod.

Fingerling rainbow trout (Salmo gairdneri) of mean weight 5.75 g were obtained from Howietown Fisheriers, Bannockburn, Scotland. Experimental fish were stocked at 20 fish per tank, three tanks per dietary treatment. At the start of the experimental feeding trial 20 fish were killed by an overdose of Benzocain (1:5,000) and stored at -15ºC for subsequent carcass analysis of moisture content and gross chemical constituents. A feeding regime of 3 % body weight per day (dry food/whole fish), divided in two parts, administered by hand, was fed to the experimental fish during the first two weeks of the experiment. As a result of the low temperatures encountered during this period (< 5°C) and the reduced feed intake of fish, the feeding level was subsequently reduced to 2% of the body weight and was maintained at this level for the duration of the 7 week feed trial.

Fish were batch weighed on a Mettler 440 top pan balance in dish containing preweighed water. Fish mortality was recorded daily as required. Faecal samples were collected from the bottom of the experimental tanks by manual pipeting during the last three weeks; of the experiment.

The faecal samples collected were pooled within each treatment and dried at 105ºC for 24 hours for subsequent crude protein and chromic oxide analysis.

The crude protein content of the experimental diets, fish carcass and faeces was determined using the MaeroK jeldhal Tecator/Kjeltec System 1003, distilling unit (AOAC, 1980). Fat content was determined by SoxIet method using Petroleum ether as solvent (AOAE, 1980). Ash content was determined by heating a preweighed sample within a siliea crucible in a muffle furnace at 450ºC for 12 h. Crude fibre content was determined by the digestion method dilute H₂SO₄ (0.225N) and NaOH (0.313 N) (AOAC, 1980). Moisture content was determined by drying a weighed sample in a drying oven at 1.05°C for 24 hours (AOAC, 1980). Chromic oxide was determined using the wet acid oxidation method of Furukawa and Tsukahara (1966).

TABLE 1 FORMULATION AND PROXIMATE ANALYSIS OF THE EXPERIMENTAL DIETS

Data were analysed by the analysis of variance (Hicks, 1973 and Parker, 1980) and mean values compared using DUNCAN'S multiple range tests (Duncan, 1955).

The feeding response and feed intake of fish at the start of the trial was below that observed for fish at the end of the feeding trial. This was believed to have been due to the low water temperature at the start of trial (4.5°C), and the steady rise in water temperature to 6.3°C by the end of the 7 week experiment. Fish fed diet 1 (containing the highest: dietary concentration of soybean meal, 15 %), dísplayed a reduced feeding response and feed intake compared with the remaining treatments.

The growth response and weekly weight increase of- trout over the experiment is shown in figure 1, and Table 2. The best growth response was observed for fish fed the highest dietary inclusion levels of sunflower meal (diet 3,4,5). It is interesting to note that the individual replicates within treatments showed a similar growth response, and the overall means therefore were comparable (Fig. 1). Fish fed the highest inclusion level of soybean meal (diet 1) displayed a significantly lower (p < 0.05) growth response (in terms of mean final body weight) compared with the remaining treatments (Table 2). A similar response was also observed on the basis of percentage weight gain and daily specific growth rate (Table 2). ln all cases fish fed diet 5 (containing the highest: dietary inclusion of sunflower meal 37.3%, with suplemental L-methionine) displayed the highest mean final body weight, percentage weight gain, daily weight gain and daily specific growth rate, compared with fish fed the highest dietary inclusion level of soybean meal (diet 1, 15 % soybean meal and 0 % sunflower meal) which displayed a significantly lower (p < 0.05) growth response on the basis of all the growth parameters measured.

Fig. 1. Overall expontial regressions of growth response of trout at sucessive weekly intervals over the experimental test period. The correlations were significant at the confidence levels of 95 %.

Within all treatments, FCR decreased from mean values ranging between 2.5 and 3.5 at the start of the experiments to values ranging between 1. 2 - 1.8 by the end of the experiment. The poor food conversion efficieney (FCR) observed at the start of the experiment was believed to have been due to the nervous behaviour of the fish (due to the stress imposed upon them by their new experimental surroundings), and their rejection for food fed. By the final week of the experiment the fish were noticeable more settled and aggressive in their feeding response. Although there was no significant difference (p < 0.05) between treatments on the basis of mean weekly FCR, fish fed diet 3 and diet 5 displayed the best overall FCR (Table 2, Fig. 2).

As observed with FCR, fish fed diet 3 (22 % sunflower) and diet 5 (37.3 % sunflower + L-methionine) displayed the highest Protein Efficieney Ratio (PER) values over the experimental test period, and these were found to be significantly higher (p < 0.05) than that observed for fish fed diet 1 (containing 15% soybean meal) (Table 2).

TABLE 2 MEAN GROWTH PERFORMANCE, FEED UTILIZATION EFFICIENCY AND CARCASS COMPOSITION OF TROUT FED THE EXPERIMENTAL DIETS FOR 7 WEEKS

Fig. 2. Mean weekly conversions efficieney for trout over the experimental test period.

As observed previously with PER, fish fed diet 3 and diet 5 displayed the highest: Net Protein Utilization (NPU) values, 23.16 % respectively, and these were found to be significantly higher (p < 0.05) than the low values of 16.38 observed for fish fed diet 1 and diet 2.

Apparent digestibility coefficients were calculated using theformula of Maynard and Loosi (1969) and are shown in Table 2. On the basis of apparent dry matter digestibility and appa-rent nitrogen digestibility determinations, fish fed diet 3 (22% sunflower meal) displayed the highest: digestibility coefficients ovér the experimental test period. The remaining treatments displayed similar digestibility coefficients, within the normal range of experimental variation.

The proximate composition of the whole fish carcass at the start and end of the experiment is shown in Table 2. With the possible exception of an elevated carcass moisture content in fish fed diet 3, and a slightly reduced carcass crude protein content in fish fed diet 5, chemical analysis of the fish carcass between dietary treatments remained relatively constant.

The reduced growth response and poor food conversion efficieney observed within treatments fed at the highest dietary concentration of solved extracted soybean meal may have been due to the inadequate heat: processing of the meal during manufacture and consequent destruction of the heat labile antinutritional factors present (diet 1, 15 % soybean meal) (in aceordance to the studies of Nose, 1971; Smith, 1977; Cowey et al., 1971; Dabrowski and Kozak, 1979; Lovell, 1980; Viola et al., 1982; Fowler and Banks, 1976; Davis and Stickney, 1978; Wu and Jan, 1977 and Jackson et al., 1982).

In the present trial there was a 20.4% and 11.3% reduction in the growth response of fish fed 15% and 8% soybean meal respectively compared with fish fed the highest sunflower meal inclusion level (37.5%).

Koops (1976); Rumsey and Ketola (1975); Fowler and Banks (1976); Dabrowski and Kozak (1979); Viola et al. (1981); Viola et al. (1982) and Jackson et al. (1982) have reported an increase in growth of fish, including trout, fed soybean containing rations supplemented with exogenous dietary free amino acids including methionine and lysine. Since these essential amino acids are normally limiting within soybean meal, they may in turn have been limiting within diet 1, as indicated by the low protein efficieney ratio (0.67) and nitrogen deposition observed for fish fed this ratio (Table 2). The increased growth response observed for fish fed diet 2 (11 % sunflower meal, 8 % soybean meal) may have been due to the higher methionine content of sunflower meal compared with soybean and the absence of any known nutritional factors within sunflower meal (Jackson et al., 1982).

Jackson et al. (1982) reported good growth in tilapia (Sarotherodon mossambicus) fed rations contining (35.2%) sunflower meal (replacing 50 % of the fish meal protein). During the present investigation fish fed 22% sunflower meal (diet 3) and 37.3% sunflower meal (diet 5) displayed the best weight gain response (mg/day), protein efficieney ratio and nitrogen deposition overthe experiment test period. There was no significant difference (p < 0.05) in the growth response and food conversión ratio of fish fed diet 3 (22 % sunflower meal) and diet 5 (37.3 % sunflower meal + L-methionine). In contrast, fish fed diet 4 (37.5 % sunflower with no supplemental methionine) had a significantly lower (p < 0.05) protein efficiency ratio than fish fed diet 3 (22 % sunflower meal). The importance of methionine supplementation within plant protein based fish feeds and its beneficial effect: on growth in salmonids has been reported by Poston et al. (1977). Despite the absence of known anti-nutritional factors within sunflower meal, sunflower meal does have a very high crude fibre content (24 %, Table 1). Thus the crude fibre content of the experimental diets fed ranged between 1. 52 % (diet 1, 0 % sunflower meal) to 10. 03 % (diet 4, 37.5 % sunflower meal). Although relatively few studies have been conducted on the nutritional value of dietary crude fibre for fish. Eastwood (1973) observed in chicks that the physical characteristics of fibre and its inclusion in diets can modify gut funetion. Buhler and Halver (1961) with chinook salmon (Oncorhynchus tschawytscha) reported an increase in growth and protein utilization efficieney with the addition of small quantities ( < 10%) of purified cellulose. Similarly, Dupree and Sneed (1966) tested within Channel catfish diets containing purified cellulose, ranging between 0 to 51 % of the dry diet, and found the highest weight gain for fish fed 21 % cellulose. Although no deleterious effect of crude fibre was observed on the growth and nutrient utilization efficieney of fish fed rations containing high levels of sunflower meal (diet 4, 37.5 % sunflower, 10. 03 % crude fibre). On the basis of the presente investigation sunflower meal has been demonstrated to be a good dietary replacement for solvent extracted soybean meal for trout. Sunflower has a weIl balanced amino acid profile and contains no known antinutritional toxic factors. In addition, sunflower seed meal can be used as a partial substitute for brown fish meal; however, when included in the ration at levels exceeding 37 % its deficieney in methionine is evident and exogenous supplementation is required. Finally, the high crude fibre content of sunflower seed meal may restriet its level of incorporation into fish feeds possibly causing a reduction in growth and therefore warrants further investigation.

Aviso de Privacidad

Cerrar esta ventana