Effects of Dietary Yeast (Saccharomyces cerevisia) Supplementation in Practical Diets of Tilapia (Oreochromis niloticus)

Simple Summary World communities are concerned about the increasing impact of the aquaculture activities on fisheries resources. Aquaculture sector uses 2–5 times more fishmeal to feed farmed species than what is supplied by the farmed product. Therefore, the reduction of fishmeal dependency may provide more economic and environmentally friendly aquaculture. By identifying alternative protein sources, the authors find that brewer’s yeast is a suitable raw material as fishmeal replacement in feed of tilapia. The 15% inclusion may promote growth without affecting the end-product quality. Abstract A 51-day feeding trial was carried out to determine the effects of various dietary levels of brewer’s yeast, Saccharomyces cerevisiae, in the growth performance, body composition and nutrient utilization in Nile tilapia, Oreochromis niloticus, juveniles. Fish (7.6 ± 0.3 g) were stocked into eighteen 1,000-L tanks (100 fish per tank; n = 3) and fed to apparent satiation six isonitrogenous (27% crude protein) and isoenergetic (19 kJ/g) diets, formulated to contain different dried yeast levels (0%, 10%, 15%, 20%, 30% or 40% diet) in substitution to fishmeal. Body weight tripled at the end of the feeding trial for fish fed up to 20% dietary yeast incorporation. Daily growth coefficient (DGC, % body weight/day) decreased with increasing dietary yeast level (P < 0.0001). Voluntary feed intake (VFI, %BW/day) did not vary significantly with increasing yeast level. Fish fed 40% yeast showed significant reduction in protein efficiency rate, protein retention and nitrogen gain. Increasing levels of dietary yeast did not significantly affect protein or lipid digestibility. Dietary dried yeast was seemingly palatable to tilapia juveniles and was suitable up to 15% inclusion to promote growth and efficient diet utilization, without affecting body composition.


Introduction
Aquaculture systems currently use 2-5 times more fishmeal to feed farmed species than what is supplied by the farmed product [1]. The scarcity of high quality fishmeal and the wide gap in demand and supply for this resource are boosting its price and may eventually hamper further development of aquaculture [2,3]. Therefore, efforts have been made worldwide to find locally available protein sources to replace fishmeal from aquafeeds and relieve the pressure on fisheries resources.
A potential fishmeal surrogate, dried surplus yeast Saccharomyces cerevisiae from fermentation plants utilizing sugar cane, is commonly available in Brazil but its use in aquafeeds is still not fully optimized. Dried yeast has antinutritional-factors (e.g., nucleic acids), which if supplemented at high concentration may hamper the performances of monogastric animals including fish [4]. The use of brewer's yeast at probiotic levels (up to 2%) has proven to have a positive effects on the performance and welfare in several fish species, such as African catfish Clarias gariepinus [5] and hybrid striped bass Morone chrysops × M. saxatilis [6]. However, in studies where brewer's yeast was tested as protein source, i.e., when high levels are required, the results are not entirely clear. Rainbow trout (Oncorhynchus mykiss) and tilapia (Oreochromis niloticus) fed with diets containing moderate to high yeast levels were reported to have reduced feed intake and growth, reduced liver glycogen and increased total liver lipids [7][8][9][10]. Conversely, the inclusion of 30-50% brewers yeast in the diet improved feed efficiency of European seabass [11] and Pacu, Piaractus mesopotamicus [12]. The aim of this study was to investigate the effects of partial and total fishmeal replacement by brewer's yeast on growth, body composition, feed utilization and digestibility of juvenile tilapia.
Dietary ingredients were homogeneously ground to 500 µm, thoroughly mixed, and humidified to 25-50%, depending on the diet formulation. Moist mixtures were cold-extruded through a 1.0 mm die mincer, dried overnight in a forced-air oven (45 °C), crumbled and sieved into 0.5-3.0 mm pellets. All diets were kept frozen (−20 °C) until they are distributed.

Sampling, Analytical Procedure and Measurements
Fish were sampled at the beginning (15 fish·tank −1 ) and at the end (50 fish·tank −1 ) of the trial, sacrificed by an overdose of benzocaine (0.2 g·L −1 ) and weighed to the nearest 0.1 g. A pool of fish sample from each tank was then ground and immediately freeze-dried and stored at −20 °C pending analyses.
The digestibility study was conducted using the settling column approach (Guelph system) for faeces collection as described by Cho et al. [14]. Fish were fed once per day with diets containing Cr 2 O 3 as inert marker, and then transferred to 200-L conical-bottomed tanks, with continuous water flow (0.5 L·min −1 ; 25 ± 1 °C) and continuous aeration (DO > 6.0 mg·L −1 ). Faeces were collected into refrigerated containers for 15 hours after the afternoon meal during the last week of the feeding trial. Apparent digestibility coefficients (ADCs) of dietary protein and lipid were calculated by the formula ADC (%) = [1 -(F/D × DC/FC)] × 100; where F is the percent of nutrient in faeces, D is the percent of nutrient in diet, DC is the percent of chromic oxide in diet and FC is the percent of chromic oxide in faeces [15].

Statistical Analyses
Data were submitted to one-way ANOVA followed by Tukey's multiple range test; normality assumptions were ensured through the Kolmogorov-Smirnov test; homogeneity of variance was established through Bartlett's test [16]. Data were expressed as the mean ± SD of the replicates, each tank representing one experimental unit. Differences were considered significant if P was less than 0.05. All statistical analyses were performed using SAS software (SAS User's Guide: Statistics, Version 9.0; SAS Institute, Cary, NC, USA).

Results and Discussion
Dried yeast is an ingredient that has been increasingly used in fish and shrimp feeds in countries with ethanol or other distillation plants. In the current study the partial fishmeal replacement by brewers yeast, Saccharomyces cerevisiae, have improved the nitrogen (N) gain and protein efficiency ratio in Nile tilapia (Oreochromis niloticus) when supplemented at a maximum of 10%, but had no significant effect on growth performance (Table 2). A linear depression on growth performance and nutrient retention were observed when fish were fed diets with more than 10% yeast (equivalent to ≥50% fishmeal substitution). Nevertheless, feed conversion ratio were statistically different from control group only in fish fed at 40% yeast, while voluntary feed intake did not differ among the dietary groups.
Nitrogen (N) gain decreased linearly from 515 to 315 mg·kg −1 ABW/day (P < 0.001), while protein digestibility remained unaltered (Table 3). Such combination of results may indicate an increase in mobilization of dietary protein for catabolic purposes and/or a reduction of intestinal absorptive ability when fish is fed increasing dietary yeast levels, both causing a decrease in the efficiency of dietary protein deposition.
Contrary to our findings, Oliva-Teles and Gonçalves [11] and Ozorio et al. [12] reported that yeast can replace 50% of fishmeal protein in seabass (Dicentrarchus labrax) and pacu diet, respectively, with no negative effects on the growth performances. In addition, Lara-Flores et al. [17] found that 40% dietary inclusion of yeast positively stimulated growth performance in tilapia.  2 Feed conversion ratio (g·g −1 ): wet weight gain / dry feed intake. 3 Voluntary feed intake (g·kg −1 ABW day −1 ): crude feed intake/ABW/trial duration, where ABW, average body weight: (IBW + FBW)/2/1,000 (kg). 4 Protein efficiency ratio (%): wet weight gain/crude protein intake. 5 Nitrogen gain: (final carcass nitrogen content − initial carcass nitrogen content)/ABW/54 days. 6 Nutrient retention (% intake): 100 × (nutrient gain/nutrient intake). The reduced N gain obtained in the current study could be explained if the experimental diets had sub-optimum amino acid patterns, as lower N retention corresponds to higher N excretion and lower rates of protein growth. Since the amino acid levels did not vary significantly among the experimental diets (Table 1), it seems that the high dietary yeast level did not cause amino acid deficiencies.
The utilization of dried yeast at reduced levels may effectively improve growth [8,11,18] and non-specific immune responses [19][20][21][22] in a variety of fish species. As inasmuch, dried yeast is a source of nucleic acids and non-starch polysaccharides, including β-1,3 glucan, which in high concentrations may play a role of antinutritional factors. At high concentrations, such compounds are known to hamper nutrient digestion and/or absorption. In avian species, β-glucans may affect the absorption of nutrients, possibly by increasing gut viscosity [23], while high concentration in nucleic acids may affect nutrient metabolism in humans and most monogastric animals [4].
The carcass composition and the digestibility coefficients of the dietary nutrients are shown in Table 3. Carcass composition was not significantly affected by increasing levels of dietary yeast, with exception of ash content. Ash content was significantly higher in fish fed 30% and 40% yeast (P < 0.05) than the other dietary groups. The digestibility coefficients of protein (78-82%) and lipid (87-94%) were not significantly different with increasing dietary yeast ( Table 3), suggesting that protein from yeast was well digested by tilapia. Our results are in agreement with Olvera-Novoa et al. [24] which observed similar digestibility values in tilapia fed with diet having similar yeast incorporation.

Conclusions
Overall, this study showed a linear decrease in growth performance and efficiency in nutrient utilization when fish were fed above 15% yeast, providing additional information for future studies interested in the optimization of a more economic and low-pollution non-fishmeal based diets.