Oxidative stability of tilapia feeds containing Saccharomyces cerevisiae and Spirulina platensis . Oxidative stability of tilapia feeds containing Saccharomyces cerevisiae and Spirulina

: The aim of this study was to evaluate the effect of the inclusion of microbial biomass on the oxidative rancidity of tilapia rations stored for 12 months. Treatments included a control diet and diets supplemented with either 0.01% vitamin E, 0.25 and 0.5% of Saccharomyces cerevisiae and 0.25 and 0.5% of Spirulina platensis. Experimental diets were stored in the dark inside plastic bags at room temperature (25 °C) for 12 months. The oxidative rancidity was measured as thiobarbituric acid reactive substances (TBARS). It was concluded that the inclusions of Spirulina platensis at 0.25% (1.734 ± 0.206) and 0.5% (1.629 ± 0.181) and Saccharomyces cerevisiae at 0.5% (1.459 ± 0.305) minimized the oxidative rancidity in comparation to control diet (2.843 ± 0.109) of Nile tilapia until 12 months of storage.

), thus giving rise to a growing demand for antioxidants of natural origin. Two microbial biomasses (MB) with antioxidant potential stand out as fish feed ingredients. Saccharomyces cerevisiae yeast is used in the preparation of various products (e.g. ethanol) and marketed as a coproduct at the end of processes. The yeast biomass contains proteins, vitamins such as E and complex B, enzymes, volatile fatty acids, chelated minerals, natural antibiotics and peptides (RAN et al., 2016). Spirulina (Arthrospira) platensis biomass contains proteins, vitamins, essential amino acids, minerals, essential fatty acids and pigments, such as carotenoids and phycocyanins, which impart it high nutritional value and antioxidant properties (MORIST et al., 2001;LI et al., 2003). The microalga is able to grow on different substrates, including industrial effluents and simple and lowcost culture media, making its production totally viable (ARRUDA et al., 2006).
The purpose of this study was to evaluate the effect of microbial biomasses on the oxidative rancidity of tilapia rations stored for 12 months.
The study adopted a completely randomized design with six treatments and three replicates. Treatments included a control basal diet (CD) and five diets supplemented with either 0.01% vitamin E (VE), or S. cerevisiae at 0.25% (SC25) and 0.5% (SC50) or S. platensis at 0.25% (SP25) and 0.5% (SP50). The calculated and analyzed values of the nutrients in the diets are presented in table 1. The proximate composition of the biomasses were: 4% moisture, 45% protein, 0.2% lipid and 5% ash for S. cerevisiae; and 6% moisture, 46% protein, 3% lipid and 8% ash for S. platensis. The biomasses were dissolved in soy oil and incorporated to the basal diet inside a Y-shaped mixer. The same amount of soy oil was added to the control group. The experimental diets were stored in the dark inside plastic bags (polypropylene) at room temperature (25 °C) for 12 months.
For the thiobarbituric acid reactive substances (TBARS) quantification, 1 g of feed was homogenized with 25 ml of 7.5% trichloroacetic acid using turrax for 1 min (TBARS extraction). Then, the mixture was filtered and 5 ml of the extract were transferred to a tube containing 5 ml 0.02 M thiobarbituric acid. The tubes were heated in boiling water bath (100 °C) for 40 min and cooled in running water for 10 min. The TBARS were measured at 538 nm and the values were expressed as mg malonaldehyde/kg (GRASSI et al., 2016(GRASSI et al., , 2018. All data collected were subjected to analysis of variance and the significance of differences between means was tested by Tukey's test, using Action version 3.1 ® . Broken line regression was used to estimate the biomass requirements to avoid the oxidative rancidity as a function of MB inclusion, using PPM (Practical Program for Modeling, https://sites.google. com/site/programapraticodemodelagem/). The significance level was set at 0.05.
The SC50, SP25 and SP50 had lower TBARS than the control group while SC25 and VE were not able to prevent the lipid oxidation during the storage for 12 months ( Table 2). The antioxidant activity of biomass minimized the oxidative rancidity, increasing the shelflife of Nile tilapia feed. This effect may be assigned to carotenoids and phycocyanins in Spirulina platensis and to rutin, hesperidin, sakuranetin, quercetin, naringin, naringenin and vitamin E in Saccharomyces cerevisiae (SOARES et al., 2005). The inclusion of S. cerevisiae at 0.25% (SC25) had insufficient antioxidant activity to reduce the oxidative rancidity. The broken line regression indicated that 0.4392% S. cerevisiae ( Figure.  The need to seek alternatives to the synthetic antioxidant substances has increased the interest for the production of natural antioxidants through biotechnology. For instance, microbial biomasses of yeast Phaffia rhodozyma and microalgae Haematococcus pluvialis and Dunaliella salina have been used as sources of carotenoids in the feed of aquatic animals (PONCE-PALAFOX et al., 2004). Vitamin E produced by microorganisms also has been used in fish diets to prevent the oxidative rancidity and protect the tissue integrity of several fish species (PENG et al., 2009). Although, the antioxidant activity of vitamin E in feed is recognized, in this experiment, it was not able to prevent the lipid oxidation like the biomasses did.
More than benefits to the preservation of the ration, S. platensis and S. cerevisiae biomasses can improve fish health and meat quality. Results presented herein indicated that S. cerevisiae and S. platensis biomasses can be considered as natural alternatives to replace the synthetic antioxidants currently in use. It was concluded that the inclusions of Spirulina platensis at 0.25 and 0.5% and Saccharomyces cerevisiae at 0.5%      1.900 ± 0.019 ab SC50 1.459 ± 0.305 b SP25 1.734 ± 0.206 b SP50 1.629 ± 0.181 b P 0.0155 CV (%) 19.6 * Thiobarbituric acid reactive substances; CD -control diet, VE -CD + 0.01% vitamin E, SC25 -CD + 0.25% S. cerevisiae, SC50 -CD + 0.5% S. cerevisiae, SP25 -CD + 0.25% S. platensis and SP50 -CD + 0.5% S. platensis. CV -coefficient of variation; ab Means followed by different letters are significantly different (P<0.05).