Digestibility and haematological indices in Nile tilapia (Oreochromis niloticus) fed diets based on food processing by-products and plant-derived ingredients

L Niyibizi^1,2, S Rukera Tabaro¹, T Lundh² and A Vidakovic²

¹ Department of Animal Production, School Veterinary Medicine, College of Agriculture Animal Science & Veterinary Medicine (UR-CAVM), P.O. Box 210 Musanze-Rwand
l.niyibizi2@ur.ac.rw (L.N.)
² Department of Animal Nutrition and Management, Swedish University of Agricultural Sciences, P O Box 7024, SE-75007 Uppsala, Sweden

Abstract

In order to determine the nutritional quality of food processing by-products and plant-derived ingredients in the diet of Nile tilapia (Oreochromis niloticus), six ingredients were studied for apparent digestibility and blood indices in 360 tilapia fingerlings (30.2 ± 1.54 g). Each experimental diet was prepared using 70% reference mash and 30% test ingredient, on an `as is´ basis. Ingredients assessed were fishmeal (Rastrineobola argentea) (reference diet, RD), kidney bean leaf meal (KBLM), spent brewer’s grain (SBG), spent brewer’s yeast (SBY), sweet potato leaf meal (SPLM) and wheat middlings (WM). In this six weeks experiment, apparent digestibility (AD) of diets and ingredients was assessed for crude protein (CP), crude lipid (CL), gross energy (GE) and dry matter (DM). Apparent digestibility of crude protein (AD_CP, %) was highest for diet SPLM (83.1%) and SBG (83.0%), followed by RD (81.6%), SBY (77.7%), KBLM (73.1%) and WM (69.7%). Mean AD for indispensable amino acids (AD_IAA), including lysine and methionine, followed a similar pattern, i.e. it was highest for diet SPLM (87.1%), followed by SBG (85.3%) and RD (fishmeal) (83.9%) and lowest in WM (72.9%). The four main haematological indices (white blood cell count (WBC), red blood cell count (RBC), haemoglobin (Hb), haemocrit (Hct)) differed somewhat between the diets. RD displayed lower RBC than diet SPLM. Diet SBY showed the lowest WBC and KBLM and WM the highest. Diet SPLM had high (p<0.05) RBC and Hb, whereas KBLM had the lowest values for RBC, Hct and Hb, indicating possible negative effects of kidney bean leaf meal on blood physiology in tilapia. Mean corpuscular volume (MCV), mean concentration of haemoglobin (MCH) and mean corpuscular haemoglobin concentration (MCHC) did not differ between the diets. Overall, the assessment of nutritional quality of the ingredients showed that SBY, SBG and SPLM had acceptable protein content, displayed high AD of nutrients and indispensable amino acids and revealed acceptable haematological indices in fish. Hence SBY, SBG and SPLM can be considered valuable ingredients in tilapia diets that may contribute to sustainable domestic fish production. Diets containing KBLM and WM displayed low digestibility and high WBC values, thus not recommended for tilapia diets.

Keywords: agro-industrial by-products, amino acids, crop leaves, fishmeal, tilapia, nutrient digestibility

Introduction

Nile tilapia (Oreochromis niloticus) was the third most common farmed fish species worldwide in 2018, after two carp species (FAO 2020). Nile tilapia is widely farmed in tropical, subtropical and temperate regions of the world, with annual growth in production of around 12.2% (El-Sayed 2020). The future fish supply will depend on aquaculture, to meet growing demand for animal protein as a result of the growing human population. Considering unique characteristics of tilapia, such as ease of reproduction, rapid growth rate, significant tolerance to environmental stress and high market demand, it is the species of choice for next-generation sustainable aquaculture (Yue et al 2016).

For sustainable intensification of fish production, feeds based on circular, sustainable alternative ingredients should be prioritised in aquaculture, but around 68% of all fish feed currently used in the aquaculture industry is based on fish products Tacon 2020. Worldwide, availability of resources for aquafeed production is a major constraint that is expected to become more severe in the rapidly expanding aquaculture sector. In addition, the most important ingredients used in fish feed (fishmeal and fish oil) are becoming increasingly limited as the world's fish stocks are already either fully exploited or depleted (FAO 2018; Tacon et al 2008). Thus, there is a critical need for suitable and sustainable feed ingredients for aquaculture globally and especially in land-locked countries such as Rwanda where Nile tilapia is the most farmed species and where, as in many other developing countries, aquaculture is expanding. Thus, attention can be turned to locally available food processing by-products and plant-derived ingredients, such as wheat mill by-products, brewery by-products and various plant-derived material including leaves (sweet potato, kidney bean). These ingredients are available in many inland areas, but due to limited information on their nutritive value they are currently left unused (Niyibizi et al 2022). At the same time, plant protein sources such as soy and other legumes are still the main alternative to fishmeal in most commercial fish diets (Gatlin et al 2007).

Aquafeeds can be formulated using ingredients from multiple protein sources, permitting complementary nutritional profiles to be created (Tacon et al 2008). Compared to many other cultured species, tilapia has relatively low requirements for protein and highly unsaturated fatty acids (HUFA) and can efficiently use high levels (up to 48% inclusion in the diet) of dietary carbohydrate as a primary energy source, saving protein for growth (Kamalam et al 2017). Thus, this species can easily be grown on feed containing less high-value dietary components, especially in ponds. Earlier results indicate that among available grains, maize, extruded maize, wheat middlings and rice meal have the highest apparent digestibility coefficient (Pezzato et al 2002). In addition, wheat bran (Maina et al 2002), cereal grain products/by-products such as corn, broken rice, wheat middlings and broken rice and rice bran (Guimarães et al 2008), have been used as important feedstuffs especially in ponds farming, even though the later are not protein source “per se”. Commonly available low-cost meals made from the leaves of vegetables such as sweet potato and cassava and agro-industrial by-products such as spent brewer’s grain and spent brewer’s yeast have all previously been tested in fish diets with varying degree of success (El-Sayed 1999; Nhi et al 2018). To our knowledge, kidney bean leaf meal, another readily available leaf meal in Rwanda has previously not been tested in diets for tilapia.

Spent brewer’s yeast (Saccharomyces cerevisiae), also known as baker’s yeast contains around 40-50% of protein and has relatively low lipid content, high ash content and moderate levels of carbohydrates (Agboola et al 2021; Vidakovic et al 2020; Øverland et al 2013). Whole dried baker’s yeast has an amino acid profile characterised by slight methionine deficiency (Langeland et al 2016). Yeast also contains various immune-stimulating compounds, including β-glucan, nucleic acids, mannan oligosaccharides and chitin (Abdel-Tawwab et al 2020). Spent brewer’s grain, the major by-product generated during beer production, is available in large quantities throughout the year, representing ~85% of total by-products generated (Mussatto 2014). Spent brewer’s grain has various use globally, including chiefly as a feedstuff rich in protein and fibre (Mussatto et al 2006) and as a raw material in food texture enhancers due to its emulsifying properties (Celus et al 2006). Sweet potato is one of the most important food crops worldwide, is drought-resistant and has high yield of leaves, which provide a dietary source of nutrients and various bioactive compounds (Nguyen et al 2021). Sweet potato can be grown in different climates and farming systems and the leaves contain various essential minerals (Fe, Ca and Mg) and essential trace elements (Cr, Co, Ni, Cu and Zn) (Taira et al 2013).

Sustainability concerns about use of fishmeal as a protein source in fish feed make it important to keep searching for novel sustainable alternatives, but it is also important to investigate the feed properties of these alternatives, as they must not cause health problems in the reared fish. The blood system of reared fish reacts directly or indirectly to changes in the environment, including diet. Therefore, blood analyses can provide substantial diagnostic information on fish fed e.g. alternative diets under rearing conditions. This information reflects the individual physiological state of a fish and aspects of fish welfare, including the immune system and acute and long-term impacts due to adverse husbandry conditions (Seibel et al 2021). Haemoglobin (Hb) level determines the oxygen-holding capacity of the blood and hence fish endurance, with erythrocytes carrying Hb to all body tissues (Abdel-Tawwab 2015). A low level of erythrocytes (and thus Hb) leads to hypoxia stress, which negatively affects metabolism of nutrients, growth performance and immunity in Nile tilapia (Abdel-Tawwab 2015). Blood analysis may also indicate elevated Hb levels, which is associated with possible haemolytic anaemia from e.g. metabolism of high levels of nucleic acid in fish (Huyben et al 2017; Mahfouz et al 2015). Haematological index values vary depending on fish species, age, sex, health status and nutritional status (Ahmed et al 2020; Hrubec et al 2000; Blaxhall et al 1973). Various haematological parameters can be analysed; of which some are fundamental for presumptive diagnosis of organism welfare. The most commonly evaluated are red blood cell count (RBC), mean cell haemoglobin concentration (MCHC), mean cell volume (MCV), mean cell haemoglobin (MCH) and other primary indices such as Hb concentration, total erythrocyte count, packed cell volume, white blood cell count (WBC) and haematocrit (Hct) (Grant  2015; Fazio et al 2019). Extensive research on farmed fish has used these parameters to assess animal welfare, in studies involving optimisation of husbandry conditions, stress avoidance and improvement of fish quality of life (Lugert et al 2020).

Wheat mill by-products, brewery by-products, sweet potato leaves and kidney bean leaves, which are widely available, have never been assessed as alternative feed ingredients in Rwanda (Niyibizi et al 2022). National identification of available feed ingredients started a decade ago and 31 ingredients have been identified as potential feed ingredients for tilapia at only proximate chemical composition analysis level (Niyibizi et al 2022; Munguti et al 2012). The nutritional value of feed and the effect of diet composition on nutrient absorption can be accurately assessed based on digestibility studies (Koushik et al 2021), as the amount or proportion of nutrients digested are presumably available to the organism for growth and metabolism (NRC 2011). In Rwanda, a fish digestibility study has been conducted only on African catfish (Nyina-Wamwiza et al 2009), but no such trial has been conducted to date on tilapia, despite it being by far the most farmed species in Rwanda.

The aim of this study was to determine the nutritional quality of food processing by-products and plant-derived ingredients in the diet of Nile tilapia, based on assessment of apparent digestibility of dietary components, energy and amino acids and effects on fish haematology.

Materials and methods

Study area and facilities

The digestibility trial was conducted at the fish farming and research station hatchery at Rwasave, part of the University of Rwanda (UR), Huye campus, located in Southern Province, Rwanda (2°40´S, 29°45E). The experiment was conducted in a recirculating aquaculture system consisting of 18 fibreglass tanks, each 100 L in volume, installed above 4480 L concrete tanks equipped with a mechanical and biological water filtration system.

Fish, rearing conditions and water parameters

A total of 360 Nile tilapia (mixed sex) with an average weight of 30.2 ± 1.54 g were used in the six weeks’ experiment. The fish were acclimated to the experimental conditions for one week, fed a same commercial diet. Following the acclimation period, the fish were weighed using an electronic balance (Mettler PM4000, Hampton, NH 03842, USA) and were randomly distributed in the rearing tanks (100 L capacity, filled to 60 L). Six diets (one reference and five test diets) were assigned at random to the 18 tanks, with three replicate tanks per diet. All the tanks had a common water supply and were handled in the same way throughout the trial. Each tank had a plastic mesh top cover to prevent fish from escaping and the fish were kept at a natural photoperiod of 12 h light: 12 h dark. Ingredients used in the reference diet and test diets are listed in Table 1. The filtration system was continuously supplemented with fresh well water, at a flow rate of 2 L min^-1. The water was constantly aerated using an electric air pump connected to stone diffusers in each tank, to ensure adequate oxygen supply. Water parameters (pH, temperature (°C), dissolved oxygen in mg L ^-1 were recorded twice daily (at 08.00 h and 15.00 h) in each experimental tank, using a portable multiparameter probe (Hanna HI 11310, Hanna Instruments Ltd., USA). Water temperature was kept around 28 °C using aquarium heaters (Aquazonic AZ-LED 100, YI HU FISH FARM Trading. Sungei Tengah, Singapore). Concentrations of nitrite (mg L^-1) and ammonia (mg L^-1) were monitored on a bi-weekly basis, using a HACH water analysis kit (DR/890 Colorimeter, Hach Company. Colorado, USA).

Experimental diets

Ingredients used were purchased from local markets, obtained from food and beverage industries or freshly harvested in local fields. Some collected ingredients were pre-treated as follows: Full fat soybean grain was autoclaved in a pressure cooker for 60 min. The treated soybean/ blood was then sundried for 2-3 days. All dried feed ingredients were milled to flour (approx. 120 µm particles) using a Grain Hammer Mill Crusher (GMEC-280 Zhengzhou Runxiang Machinery Equipment Co. Ltd, Zhengzhou, Henan, China). A vitamin and mineral premix was added to the dried feed ingredients. Titanium (IV) dioxide, a non-toxic inert marker, was used as an indirect digestibility assessment method that involved collection of fish faeces samples from settlement (which is reported not to affect experimental fish) (Glencross et al 2007; Mmanda et al 2020; Maynard et al 1974). It was added at a rate of 0.5% to all experimental diets (Table 1). The ingredients were then mixed mechanically (Santos 10Ltr Dough Mixer, Lyon, France) for 5 min until apparent homogeneity, sunflower oil was added and mixing was continued for an additional 5 min. A small amount of clean water was then added and mixed in for 10 minutes, to form a homogenous dough, which was pelleted using a Meat Grinder machine (FAMA FTS107, Brugnera, Italy). The wet pellets (2 mm) were sun-dried for 2-3 days and stored at -20 °C until use (maximum 30 days). A small portion (enough to be offered within five days) was regularly taken from main diet batch and kept at 5 °C in sealed food-grade plastic bags. Six iso-nitrogenous and iso-energetic diets with different levels of fishmeal replacement were formulated to fulfil the nutrient requirements of Nile tilapia (NRC 2011). The reference diet (RD) was fishmeal-based, while the five test diets were formulated with the maximum possible fishmeal replacement without affecting dietary crude protein and energy content (Table 1). The test diets contained seven parts reference ingredients and three parts of test ingredient (70:30), based on previous recommendations (Cho et al 1979) (Table 1). Test ingredients were included in the experimental diets on an `as is´ basis (Table 1). Fish were hand-fed (for 45 days) a pre-determined ration (approximately 4.5% of their body weight per day), served twice daily at 09.00 h and 13.00 h.

Faeces and blood sampling

Faeces sample collection started six days after changing from commercial to the experimental diets, to allow evacuation of all the commercial feed previously ingested. Uneaten feed was siphoned out within 30 min post-feeding and faeces samples were collected through siphoning (Mmanda et al 2020; NRC 2011), from each experimental tank twice daily (11.00 h and 15.00 h) within 2 hours post-feeding, using a 2 cm pipe. During faeces sampling, care was taken to ensure maximum recovery of a relatively unbroken string of faeces (intact). The siphoned faeces were collected on a 100 µm nylon filter mesh and samples (average 40 g) were transferred to an appropriate plastic container, placed on ice and stored at -20 °C until proximate composition and amino acids analysis of faeces samples.

For evaluation of haematological parameters, at the end of the six weeks’ digestibility experiment, three fish were randomly collected from each tank (n=54 fish) and anesthetised with a solution (50 mg L^-1) of tricaine methane-sulphonate (MS-222), (Topical Anesthetics; MS-222, chemical Inc., USA). Blood samples (1.0 mL) were collected from the caudal vein of the fish (one of the least traumatic and most widely used blood sampling procedures (Seibel 2021; Congleton et al 2001), using heparinized syringes (2 mL) and immediately transferred into heparinized vials and placed on ice until further analyses. The haematological parameters measured were red blood cell count (RBC), white blood cell count (WBC), haematocrit (Hct), haemoglobin (Hb), mean cell volume (MCV), mean cell haemoglobin (MCH) and mean cell haemoglobin concentration (MCHC). RBC and WBC were counted using an improved Neubauer haemocytometer (Reichert, Inc., Depew, NY, USA) after blood dilution with phosphate-buffered saline (pH 7.2), according to the haemocytometer manufacturer’s instructions (Rusia 1961). Haematocrit values were determined by placing blood samples in glass capillary tubes and centrifuging for 5 min at 12,000 rpm in a micro-haematocrit centrifuge (Nelson et al 1989). Haemoglobin concentration was determined colorimetrically by measuring formation of cyanomethaemoglobin using a spectrophotometer at wavelength 540 nm according to Van Kampen and Zijlstra (van Kampen et al 1961). Erythrocyte indices (mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH), mean corpuscular haemoglobin concentration (MCHC) were calculated using standard formulae (Lavanya et al 2010; Bain 2017; Stoskopf 1993): (MCV = Hct/ RBC x10), MCH = Hb/RBC x 10, (MCHC = Hb/Hct x 100).

Chemical analyses of test ingredients

Ingredients used to formulate control and experimental diets were analysed for their proximate chemical composition (Table 2) at the food science laboratory at the University of Rwanda, Busogo campus. Dry matter (DM) was determined by oven-drying at 105 °C for 24 h. Ash content was determined by incineration of samples at 550 °C for 4 h. Total nitrogen (N) content was determined using the Kjeldahl method (BSWKJ, Bharat Scientific World. Bengaluru, Karnataka, India) (Rusia 1992) and crude protein (CP) was calculated as N x 6.25. Crude lipid content (EE) was measured using the Soxhlet method, after acid hydrolysis of the sample (124, NSK Scientific Company, Chennai, Tamil Nadu, India) and crude fibre (CF) content was analysed using standard methods (Chai et al 1998). Nitrogen-free extract (NFE) was calculated as NFE (%) = 100 - (CP+CL+CF+Ash), according to (AOAC 2000) and gross energy (GE) as GE = 5.72*CP + 9.50*EE + 4.79*CF + 4.17*NFE (g kg^-1 DM) (Castell et al 1980).

Chemical analysis of diets and faeces

Proximate analysis of the final test diets, amino acids and faeces samples was performed at the Swedish University of Agricultural Sciences feed laboratory in Uppsala, Sweden. Dry matter was determined by freeze-drying all frozen feed and faeces samples, oven-drying at 103 °C for 16 h and then cooling in a desiccator before weighing. Ash content was determined by incinerating in a muffle furnace at 550 °C for 3 h and cooling in a desiccator before weighing. Total N was determined by the Kjeldahl method and CP was calculated as Kjeldahl-N x 6.25 (Nordic Committee on Food Analysis, 1976 (Schiemann et al 1966), with Kjeldahl-N determined using a 2520 Digestor, Kjeltec 8400 Analyser unit and 8460 sampler unit (FOSS Analytical A/S Hilleröd, Denmark). Crude lipid content was determined according to the Official Journal of the European Communities (1998), using a Hydrotec 8000 and Soxtec 8000 Extraction Unit, both from Foss, Denmark. Neutral detergent fibre (NDF) was determined according to (Chai et al 1998). Amino acid content of the diets was analysed by high-performance liquid chromatography according to (Vázquez-Ortiz et al 1995). Gross energy content was determined by isoperibol bomb calorimeter (calorimeter Parr 6300, Parr Instrument Company, Moline, IL, USA). Titanium dioxide concentration was measured according to (Short et al 1996).

Digestibility calculations

The following calculations were performed:

Apparent digestibility coefficient (ADC_diet) of the diets was calculated as (Cho et al 1982):

ADC_diet (%)= [1- (F/D x Di/Fi)]x100

where F = % nutrient (or kJ g^-1 gross energy) of faeces, D = % nutrient (or kJ g^-1 gross energy) of diet, Di = % digestion indicator of diet and Fi = % digestion indicator of faeces.

ADC of the test ingredients was calculated as (Bureau et al 1999):

ADC_{test ingr.} = ADC_{test diet.}+ [(ADC_{test diet}-ADC_{ref. diet}) x (0.7 x D_ref/0.3 x D_{test ingr})]

where D_ref = % nutrient (or kJ g^-1 gross energy) of reference diet (as is) and D_{test ingr} = % nutrient (or kJ g ^-1 gross energy) of the test ingredient (as is).

Statistical analysis

Values obtained in the apparent digestibility assessment for DM, CP and GE were encoded into Microsoft Excel worksheets and imported into IBM SPSS STATISTIC (2021) program version 27 software for statistical analysis. The data were subjected to one-way analysis of variance (ANOVA). When appropriate, Duncan’s multiple range test was applied to evaluate differences (p<0.05) between means. All means were recorded, with standard error of the mean (SEM). Tank was considered a fixed effect, while diet was considered a random effect.

Results

Experimental diets and chemical composition of ingredients

The experimental diets were all accepted and consumed by the fish. The DM content ranged between 861 and 925 g kg^-1 with the highest value for sweet potato leaves and the lowest for fishmeal (Table 2). The CP content varied between 178 g kg^-1 DM (Wheat middlings) and 548 g kg¹ DM (fishmeal). The CL values ranged between 48 in SBY and 170 g kg¹ DM, in FM. CF ranged between 21 and 130 g kg¹ DM and ash content between 51 and 164 g kg¹ DM, NFE ranged between 108 and 612 g kg¹ DM (Table 2).

In terms of proximate composition, the DM content of the experimental diets was highest (947 g kg^-1) in the diet with spent brewer’s yeast (SBY) and lowest (924 g kg^-1) in the diet with spent brewer’s grain (SBG) (Table 3). Crude protein content was highest (371.2 g kg^-1) in diet with SBG, followed by the reference diet (343.9 g kg^-1 DM) and lowest (295.1g kg^-1 DM) in the diet with sweet potato leaf meal (SPLM). Crude lipid content was highest (110.9 g kg^-1) in SBY and lowest (81.7 g kg^-1 DM) in the diet with wheat middlings (WM), while GE was highest (21.9 MJ kg^-1) in SBY and lowest (20.4 MJ kg^-1) in SBG. Ash content was highest (164 g kg^-1) in KBLM and lowest (51 g kg^-1) in RD. Total amino acids content was highest in RD and SBG (274.5 g kg^-1) and lowest (233.4 g kg^-1) in SPLM. Sum of indispensable amino acids was highest (131 g kg⁻¹) in RD, high in SBG, moderate in the diet with kidney bean leaf meal (KBLM) and SBY and lowest (110.4 g kg^-1) in SPLM (Table 3).

Apparent digestibility (%) of CP (AD_CP), CL (AD_CL) and GE (AD_GE) differed significantly (p<0.05) between the reference diet (fishmeal) and the test diets (Table 4). AD _CP was highest (p<0.05) in KBLM (90.0%) and WM (89.0%), followed by RD and SBG (87.7%) and SBY (87.6%). ADC _CL was highest in RD (fishmeal) and SPLM and lowest (p<0.05) in KBLM. The reference diet generally showed the highest (p<0.05) AD values (except for ADC_CP) and diets KBLM and WM diets the lowest (p<0.05).

Apparent digestibility of indispensable amino acids was higher (p<0.05) in SPLM than in RD and significantly lower in both the WM and KBLM diets. Apparent digestibility of lysine and methionine was highest in SPLM (p<0.05), followed by RD (89.7-87.5 and 88.8-84.7 %, respectively)  and was significantly lowest in diet WM (75.0-77.3 respectively) (Table 4).

In general, three of the test ingredients (SPLM, SBG and SBY) performed better or almost as well as the reference diet.

Dry matter AD was statistically similar for all five test ingredients (Table 5). Three ingredients (spent brewer’s yeast and grain, sweet potato leaf meal) had high (p<0.05) CP digestibility (70.3-88.2%), whereas the two remaining ingredients, wheat middlings and kidney bean leaf meal, displayed low CP digestibility (23.0-47.6%). The AD _CL value was highest for sweet potato leaf meal and spent brewer’s grain (86.9-81.0%) followed by wheat middlings and spent brewer’s yeast (60.5-60.6%) and lowest in kidney bean leaf meal (43.7%) (p<0.05). Except spent brewer’s grain, which showed the lowest digestibility (p<0.05) (Table 5).

The AD _GE values differed significantly between the test ingredients, with the value ranging from 63.9% in sweet potato leaf meal to 14.6% in wheat middlings (p<0.05) (Table 5).

Haematological indices

Overall, WBC, Hb and Hct levels showed significant differences (p<0.05) between the diets, whereas the MCV, MCH and MCHC indices showed no differences (p>0.05) (Table 6). Diet SPLM gave significantly higher RBC in fish, while other diets, including RD, were not significantly different. Diet SBY gave the highest Hb concentration (7.54 g dL^-1). Diet KBLM gave the highest WBC values (107.6×10³ mL^-1), followed by WM (88.2×10³ mL^-1), while SBY gave the lowest WBC level in fish (54.2×10³ mL^-1). Diet KBLM have significantly lower Hct level than the other diets, while SPLM gave the highest Hct value, followed by SBY and SBG (Table 6).

Water quality

In this study, water quality presented no differences between dietary treatments (p>0.05). Mean water temperature (°C) range was 27.1±0.33-, pH was 7.60±0.11- and dissolved oxygen content was 4.81±0.30 mg L^-1. The concentration for total Ammonia-N and for nitrite-N was 0.23±0.02 mg L ^-1 and 0.11±0.01 mg L^-1, respectively.

Discussion

Our dietary treatments formulation agrees with the general concept as most aquafeeds used today are made with multiple ingredients instead of a single protein source, permitting creation of complementary nutritional profiles from multiple protein sources (Tacon et al 2011). Digestibility values are crucial for obtaining accurate matrix values for different ingredients in feed formulation, as diets are formulated based on digestible nutrients rather than the chemical composition of ingredients (Glencross et al 2021).

By the end of our trial, AD _DM was over 72% and the value for the test feed ingredients and the complete test diets did not differ (p>0.05). All test diets were consumed by the fish and water quality in the rearing tanks were the same and did not affect treatments. In addition, test diets showed acceptable AD _CP (range 69.7-83.1%), which is equivalent to or higher than most AD _CP values previously reported for Nile tilapia (Mannda et al 2020; Hanley 1987), however lower than the range (88.4-92%) reported in one study (El-Shafai et al 2004).

Overall, diet SPLM had the highest AD (%) of amino acids AAs (AD_AA), higher than for RD and the other test diets. The AD _AA for all amino acids in diet SPLM was equivalent to, or higher than, that of the amino acids in RD. For instance, the AD (%) values obtained for lysine and methionine were highest in SPLM, followed by RD and then the other test diets. Previous studies have reported a high correlation between protein digestibility and average AA availability (Bureau et al 1999; Wilson et al 1981; Gaylord et al 2004), so present results indicate that SPLM can be a valuable ingredient in tilapia diets. Furthermore, the sweet potato leaf meal itself had the second highest AD_CP,following SBG but also the highest AD _CL of the test ingredients. Our results agree with previous studies, which reported that sweet potato leaves contain a large amount of protein (26-33%) with a high amino acid score (Ekenyem et al 2006; Ishida 2000). Lysine content in our experimental diets (16.6-19 g kg^-1) exceeded the dietary lysine requirement reported for Nile tilapia (13.0-14.4 g kg⁻¹ of diet) (NRC 2011). It was also slightly higher than the 14.6 g kg^-1 suggested as the optimum requirement for fillet content in finishing Nile tilapia (Michelato et al 2016). Similarly, methionine levels in the test diets exceeded the levels (1.3 and 2.68 g kg^-1) previously reported (Santiago et al 1988). Methionine levels in the test diets were comparable (5.3-6.5 g kg^-1), but slightly below the recommended minimum concentration of 7 g kg^-1 according to NRC (2011). Dietary protein quality is the most important factor affecting digestibility, in terms of nutrient availability and growth performance in fish (El-Sayed 2020).

Apparent digestibility of CP and CL in the spent brewer's yeast ingredient and of CP, CL and GE in diet SBY, was higher than in the kidney bean leaf and wheat middling ingredients, but lower than in diets RD, SPLM and SBG. A similar trend was observed for amino acid digestibility of the spent brewer's yeast ingredient. These results agree with previous findings of a particularly rich and favourable amino acid profile content (>20 g kg ^-1 DM) for leucine, aspartic acid and glutamic acid in brewer’s yeast, although slightly lower lysine (19-15.4) content than found in this study (Agboola et al 2021; Vidakovic et al 2020; Øverland et al 2013). Those studies reported slightly higher proximate CP content (about 40–55%) for brewer’s yeast than in the present study (38%), but this can depend on the yeast strain, substrate and conditions. Sulphur-containing methionine is often reported as limiting when yeast is used as the major protein ingredient in fish feeds (Agboola et al 2021; Oliva-Teles et al 2001), while digestibility for methionine is reported to be low to moderate (Agboola et al 2021). Low digestibility of this ingredient can be due to the rigid cell wall in yeast (Vidakovic et al 2020; Rawling et al 2019).

The wheat middlings and kidney bean leaf meal ingredients showed significantly lower AD values for amino acids such as lysine, isoleucine, methionine and cysteine compared with RD and the other test ingredients. Despite the ability of tilapia to utilise nutrients from animal and plant sources and from agro-industrial by-products, wheat middlings and kidney bean leaf meal consistently displayed low indispensable amino acid (IAA) digestibility (71-79%). According to Teodósio et al 2022), absorption and utilisation of AA such as methionine in tilapia juveniles are influenced by the dietary source. Non-conventional ingredients such as kidney bean leaf meal contain a high concentration of dietary fibre (DF), which lowers the nutritional value of the diet (Jha et al 2013). Relatively high levels of crude fibre and low levels of lysine in plant ingredients are reported to be the most limiting factors in diets for fish (Ogello et al 2017). Amino acid and protein digestibility could be low in wheat middlings as a result of the high crude fibre and non-starch polysaccharide content, which increase digesta flow rate, thus reducing enzyme to substrate contact time. Additionally, the drying process may reduce the lysine content of wheat grain (Dexter et al 1984). Nutrient and energy digestibility may also differ from one fish species to another and even within same species depending on age, sex, species, water temperature and diet composition (NRC 2011). In this study, crude fibre content in the experimental diets ranged from 39 to 51 %, which may have had different effects on digestibility in fish. However, using the appropriate level of carbohydrates in aquafeed is of great importance, as inappropriate levels may have negative effects on nutrient utilisation, growth, metabolism and health (Kamalam et al 2016). For instance, the maximum recommended level of dietary carbohydrate inclusion is 15-25% for salmonids and marine fish, while it can be as high as 50% for herbivorous and omnivorous species (NRC 2011). As an omnivorous fish, tilapia can efficiently use high levels (30-70%) of dietary carbohydrates as a primary energy source, which saves protein for growth (Kamalam et al 2016).

Blood contains easily accessible information about the individual physiological state of a fish (Seibel et al 2021). In general, however, interpretation of blood-derived information requires caution, since particular physiological perturbations do not necessarily depend on a given experimental protocol. As rearing conditions were the same in all treatments in this study and the fish were not exposed to any acute stressor, any difference/deviation observed in haematological indices was most likely a response to dietary treatment. Most blood sampling techniques are considered minimally invasive for fish above a given size and do not activate stress responses (Lugert et al 2020).

Overall, four main haematological indices (WBC, RBC, Hb, Hct) differed between fish fed the different diets. Diets SBY, SBG and RD gave lower RBC than SPLM, while diet KBLM gave the lowest. Diet SBY gave the lowest WBC and KBLM and WM the highest. Diet SPLM gave the highest (p>0.05) Hb values, while diet KBLM gave the lowest Hct and Hb values. Thus, analysis of blood parameters indicated possible negative effects of diet KBLM on fish physiology in this study. However, no differences were detected between the diets for MCV, MCH or MCHC, the RBC indices that are most valuable in morphological classification of anaemia (Grant 2014). Red blood cells contain Hb, which supplies oxygen to all body tissues, so Hb levels determine fish endurance (Qiang et al 2012). The normal Hb range in tilapia blood is reported to be 5.05-8.33 g dL^-2 and the values obtained were within that range. In fact, diets SPLM, SBY and SBG gave good oxygen-holding capacity, as evidenced by higher levels of the cell type that dominates in the blood of the vast majority of fish species (Rey Vázquez et al 2007). Diet SPLM gave favourable haematological indices in the fish, in some cases better than brewer’s by-products and always better than the other plant-based ingredients tested. Theoretically, sweet potato leaves can improve the nutritional properties of the diet, as they have a high nutritive value, contain about 78 g sugars kg^-1 DM, 34 g starch kg^-1 DM and 25.6-32.4% CP (Ishida 2000) and have a good amino acid profile and relatively low fibre content. They also contain some essential minerals and vitamins such as Vitamin A, Vitamin B2, Vitamin C and Vitamin E, various bioactive compounds and high-function components such as lutein, β-carotene and total chlorophyll, although the amounts vary between different cultivars (Li et al 2017; Taira et al 2013).

Previous work has found that yeast-based feed ingredients have a beneficial influence on fish haematological indices. Most of the immunostimulants in fish diets are polysaccharides derived from bacteria, fungi or yeasts  and plants (Stolen et al 1994). Other compounds that may have an immunostimulatory effect include polysaccharides containing sugars other than glucose (glucans), with yeast glucans being the most commonly used immunomodulators in aquaculture (Pilarski et al 2017). In addition, spent brewer’s yeast is a rich source of protein, minerals, vitamins and nutraceuticals (Puligundla et al 2020). Spent brewer’s grain is reported to contain substantial amounts of yeast cells and to act as rich source of protein, betaglucan, nucleotides and B-complex vitamins, with some of these compounds reported to have a beneficial effect on health (Robertson et al 2010). It is moderately rich in CP and in oven-dried form contains approximately 23-24.2% CP, 3-9.4% CL, 50% cellulose and 3-4.1% ash (Santos et al 2003; Yu et al 2020). In this study, the two yeast-based ingredients and sweet potato leaves resulted in good values of haematological indices and thus potentially had good physiological and health effects on the fish fed diets containing these ingredients. Diets SBY, SBG and SPLM resulted in acceptable protein content and showed high AD for nutrients and for IAA. Diet KBLM had the lowest Hb level (4.24%), while other dietary treatments displayed an acceptable Hb range and KBLM (and WM) also gave significantly elevated WBC values.

While no differences were detected between the diets with regard to common RBC indices used in detection of anaemia (MCV, MCH, MCHC), fish blood constituents are not stable within known ranges but vary widely depending on a number of environmental factors (Muhamed et al 2019), including fish age, gender, environment, nutrition and oxygen deficiency. Due to absence of reference ranges for fish haematological indices (Karimi et al 2013), interpretation of blood data for fish can be difficult (Makaras et al 2020; Fazio 2019) and must consider the specific experimental conditions. Concentrations of glucose, total protein and albumin were not analysed in this study, but could provide valuable additional information when evaluating stress and the nutritional condition of fish (Azaza et al 2008). The haematological data obtained suggest that feeding tilapia fingerlings with diets based on some plant-derived ingredients and food processing by-products locally available (SPLM, SBY and SBG) did not affect their nutrition and health status at the inclusion levels tested. In contrast, despite being potentially low-cost and abundant ingredients, diets KBLM and WH are not good candidates as tilapia feeds in Rwanda, at least at the levels evaluated here.

In general, water quality parameters in this study remained stable along the experiment and were acceptable for a good performance of the fish reared in RAS. The temperature and dissolved oxygen were in the optimum level for tilapia. In this study, water temperature was thermostatically controlled and stabilized at around 28 degrees, whereas, nitrogenous compounds stayed at the minimum levels and thus did not affect the performance of either fish fed reference diet or the fish fed test diets. Any difference across treatments was not influenced by the rearing water quality parameters.

Conclusion

• Three (SPLM, SBY, SBG) of the five tilapia diets based on plant-derived ingredients and food processing by-products assessed in this study performed similarly to a reference diet based on fishmeal. Apparent digestibility was high for CP, CL and GE in diets SBG, SPLM and SBY, which suggests that sweet potato leaf meal and spent brewer’s by-products are suitable ingredients for tilapia diets. Spent brewer’s yeast also appears to be a suitable ingredient, as diet SBY had adequate protein content and high AD for nutrients, energy, IAA and overall amino acids.
  
  
• In addition, the diets based on brewer’s by-products and sweet potato leaf meal gave acceptable haematological indices in tilapia, essentially in terms of WBC, RBC and Hb status, confirming their potential as ingredients in tilapia aquafeed.
  
  
• Growth performance assessments are recommended in future studies to determine the optimum inclusion level of brewer’s by-products and sweet potato leaf meal as protein sources for Nile tilapia in sustainable production in Rwanda. Kidney bean leaf meal and wheat middlings performed poorly as ingredients and had low digestibility, so their use in tilapia aquafeed is not recommended.

References

Abdel-Tawwab M, Hagras A E, Elbaghdady H A M and Monier M N 2015 Effects of dissolved oxygen and fish size on Nile tilapia, Oreochromis niloticus (L.): growth performance, whole-body composition and innate immunity, Aquacult Int, vol. 23, no. 5, pp. 1261–1274. https://doi.org/10.1007/s10499-015-9882-y

Abdel-Tawwab M, Adeshina I and Issa Z A 2020 Antioxidants and immune responses, resistance to Aspergilus flavus infection and growth performance of Nile tilapia, Oreochromis niloticus, fed diets supplemented with yeast, Saccharomyces serevisiae’, Animal Feed Science and Technology, vol. 263, p. 114484. https://doi.org/10.1016/j.anifeedsci.2020.114484

Agboola J O, Øverland M, Skrede A and Hansen J Ø 2021, Yeast as major protein-rich ingredient in aquafeeds: a review of the implications for aquaculture production, Reviews in Aquaculture, vol. 13, pp. 949–970. https://doi.org/10.1111/raq.12507

Ahmed I, Reshi Q M and Fazio F 2020 The influence of the endogenous and exogenous factors on hematological parameters in different fish species: a review, Aquacult Int, vol. 28, no. 3, pp. 869–899. https://doi.org/10.1007/s10499-019-00501-3

AOAC 2000 Official Methods of Analysis. 17th Edition, The Association of Official Analytical Chemists, Gaithersburg, MD, USA. Methods 925.10, 65.17, 974.24, 992.16.

Azaza M S, Dhraïef M N and Kraïem M M 2008 Effects of water temperature on growth and sex ratio of juvenile Nile Tilapia Oreochromis niloticus (Linnaeus) reared in geothermal waters in southern Tunisia. Journal of Thermal Biology, 33, 98-105. https://doi.org/10.1016/j.jtherbio.2007.05.007.

Bain B J, Bates I, Laffan M A and Lewis S M 2017 Eds., Dacie and Lewis practical haematology, Twelfth edition. Philadelphia: Elsevier, 2017. https://doi.org/10.1111/bjh.14872

Blaxhall P C and Daisley K W 1973 Routine haematological methods for use with fish blood, J Fish Biology, vol. 5, no. 6, pp. 771–781. https://doi.org/10.1111/j.1095-8649.1973.tb04510.x

Bureau D P, Harris A M and Cho C Y 1999 Apparent digestibility of rendered animal protein ingredients for rainbow trout (Oncorhynchus mykiss), Aquaculture, vol. 180, no. 3–4, pp. 345–358. https://doi.org/10.1016/S0044-8486(99)00210-0

Castell J D and Tiews K 1980 Eds., Report of the EIFAC, IUNS and ICES Working Group on Standardization of Methodology in Fish Nutrition Research, Hamburg, Federal Republic of Germany, 21-23 March 1979.

Celus I, Brijs K and Delcour J A 2006 The effects of malting and mashing on barley protein extractability, Journal of Cereal Science, vol. 44, no. 2, pp. 203–211. https://doi.org/10.1016/j.jcs.2006.06.003

Chai W and Udén P 1998 An alternative oven method combined with different detergent strengths in the analysis of neutral detergent fibre, Animal Feed Science and Technology, vol. 74, no. 4, pp. 281–288. https://doi.org/10.1016/S0377-8401(98)00187-4

Cho E Y and Slinger S L 1979 Apparent Digestibility Measurement in Foodstuff for Rainbow Trout. In: Halver, J.O. and Tiews, K., Eds., World Symposium on Fin Fish Nutrition and Fish Feed Technology, 239-247.

Cho C Y, Slinger S J and Bayley H S 1982 Bioenergetics of salmonid fishes: Energy intake, expenditure and productivity, Comparative Biochemistry and Physiology Part B: Comparative Biochemistry , vol. 73, no. 1, pp. 25–41. https://doi.org/10.1016/0305-0491(82)90198-5

Congleton J L and LaVoie W J 2001 Comparison of Blood Chemistry Values for Samples Collected from Juvenile Chinook Salmon by Three Methods, Journal of Aquatic Animal Health, vol. 13, no. 2, pp. 168–172, Jun. 2001. https://doi.org/10.1577/1548-8667(2001)013<0168:COBCVF>2.0.CO;2

Dexter J E, Tkachuk R and Matsuo R R 1984 Amino Acid Composition of Spaghetti: Effect of Drying Conditions on Total and Available Lysine, J Food Science, vol. 49, no. 1, pp. 225–228. https://doi.org/10.1111/j.1365-2621.1984.tb13713.x

Ekenyem B U and Madubuike F N 2006 An Assessment of Ipomoea asarifolia Leaf Meal as Feed Ingredient in Broiler Chick Production, Pakistan J. of Nutrition, vol. 5, no. 1, pp. 46–50. https://scialert.net/abstract/?doi=pjn.2006.46.50

El-Sayed A-F M 2020 Tilapia culture, Second edition. London; San Diego, CA: Elsevier/Academic Press. https://doi.org/10.1016/C2017-0-04085-5

El-Sayed A-F M 1999 Alternative dietary protein sources for farmed tilapia, Oreochromis spp., Aquaculture, vol. 179, no. 1–4, pp. 149–168. https://doi.org/10.1016/S0044-8486(99)00159-3

El-Shafai S A, El-Gohary F A, Verreth J A J, Schrama J W and Gijzen H J 2004 Apparent digestibility coefficient of duckweed (Lemna minor), fresh and dry for Nile tilapia (Oreochromis niloticus L.), Aquac Research, vol. 35, no. 6, pp. 574–586. https://doi.org/10.1111/j.1365-2109.2004.01055.x

FAO 2018 The State of World Fisheries and Aquaculture 2018 - Meeting the sustainable development goals. https://www.fao.org/3/i9540en/i9540en.

FAO 2020The State of World Fisheries and Aquaculture 2020. Sustainability in action. Rome. https://doi.org/10.4060/ca9229en

Fazio F 2019 Fish hematology analysis as an important tool of aquaculture: A review, Aquaculture, vol. 500, pp. 237–242. https://doi.org/10.1016/j.aquaculture.2018.10.030

Gatlin D M, Barrows F T, Brown P, Dabrowski K, Gaylord T G, Hardy R W, Herman E, Hu G, Krogdahl S, Nelson R, Overturf K, Rust M, Sealey W, Skonberg D, Souza E J, Stone D, Wilson R and Wurtele E 2007 Expanding the utilization of sustainable plant products in aquafeeds: a review, Aquaculture Res, vol. 38, no. 6, pp. 551–579. https://doi.org/10.1111/j.1365-2109.2007.01704.x

Gaylord T G, Rawles S D and Gatlin D M 2004 Amino acid availability from animal, blended and plant feedstuffs for hybrid striped bass (Morone chrysops x M. saxatilis), Aquac Nutrition, vol. 10, no. 5, pp. 345–352. https://doi.org/10.1111/j.1365-2095.2004.00310.x

Glencross B D, Booth M and Allan G L 2007 A feed is only as good as its ingredients? a review of ingredient evaluation strategies for aquaculture feeds, Aquaculture Nutrition, vol. 13, no. 1, pp. 17–34. https://doi.org/10.1111/j.1365-2095.2007.00450.x

Glencross B, Grobler T and Huyben D 2021 Digestible nutrient and energy values of corn and wheat glutens fed to Atlantic salmon (Salmo salar) are affected by feed processing method, Aquaculture, vol. 544, p. 737133. https://doi.org/10.1016/j.aquaculture.2021.737133

Grant K R 2015 Fish Hematology and Associated Disorders, Veterinary Clinics of North America: Exotic Animal Practice, vol. 18, no. 1, pp. 83–103. https://doi.org/10.1016/j.cvex.2014.09.007

Guimarães I G, Pezzato L E, Barros M M and Tachibana L 2008 Nutrient Digestibility of Cereal Grain Products and By-products in Extruded Diets for Nile Tilapia, Journal of the World Aquaculture Society, vol. 39, no. 6, pp. 781–789. https://doi.org/10.1111/j.1749-7345.2008.00214.x

Hanley F 1987 The digestibility of foodstuffs and the effects of feeding selectivity on digestibility determinations in tilapia, Oreochromis niloticus (L), Aquaculture, vol. 66, no. 2, pp. 163–179. https://doi.org/10.1016/0044-8486(87)90229-8

Hrubec T C, Cardinale J L and Smith S A 2000 Hematology and Plasma Chemistry Reference Intervals for Cultured Tilapia (Oreochromis Hybrid), Veterinary Clinical Pathology, vol. 29, no. 1, pp. 7–12. https://doi.org/10.1111/j.1939-165X.2000.tb00389.x

Huyben D, Vidakovic A, Nyman A, Langeland M, Lundh T and Kiessling A 2017 Effects of dietary yeast inclusion and acute stress on post-prandial whole blood profiles of dorsal aorta-cannulated rainbow trout, Fish Physiol Biochem, vol. 43, no. 2, pp. 421–434. https://doi.org/10.1007/s10695-016-0297-0

Ishida H 2000, Nutritive evaluation on chemical components of leaves, stalks and stems of sweet potatoes (Ipomoea batatas poir), Food Chemistry, vol. 68, no. 3, pp. 359–367. https://doi.org/10.1016/S0308-8146(99)00206-X

Jha R, Htoo J K, Young M G, Beltranena E and Zijlstra R T 2013 Effects of increasing co-product inclusion and reducing dietary protein on growth performance, carcass characteristics and jowl fatty acid profile of growing–finishing pigs1, Journal of Animal Science, vol. 91, no. 5, pp. 2178–2191. https://doi.org/10.2527/jas.2011-5065

Kamalam B S and Panserat S 2016 Carbohydrates in fish nutrition

Kamalam B S, Medale F and Panserat S 2017 Utilisation of dietary carbohydrates in farmed fishes: New insights on influencing factors, biological limitations and future strategies, Aquaculture, vol. 467, pp. 3–27. https://doi.org/10.1016/j.aquaculture.2016.02.007

Karimi Sh, Kochinian P and Salati A P 2013 The effect of sexuality on some haematological parameters of the yellowfin seabream, Acanthopagrus latus in Persian Gulf, IJVR, vol. 14, no. 1.

Koushik R and Mraz J 2021 Digestibility of protein feeds in Tilapia. https://researchgate.net/publication/353365512

Langeland M, Vidakovic A, Vielma J, Lindberg J E, Kiessling A and Lundh T 2016 Digestibility of microbial and mussel meal for Arctic charr (Salvelinus alpinus) and Eurasian perch (Perca fluviatilis), Aquacult Nutr, vol. 22, no. 2, pp. 485–495. https://doi.org/10.1111/anu.12268

Lavanya S, Ramesh M, Kavitha C and Malarvizhi A 2010 Hematological, biochemical and ionoregulatory responses of Indian major carp Catla catla during chronic sublethal exposure to inorganic arsenic, Chemosphere, vol. 82, no. 7, pp. 977–985, Feb. 201. https://doi.org/10.1016/j.chemosphere.2010.10.071

Li M, Yeong Jang G, Hoon Lee S, Young Kim M, Gu Hwang S, Man Sin H, Sig Ki H m, Lee J., Sang Jeong H. 2017 Comparison of functional components in various sweet potato leaves and stalks, Food Sci Biotechnol, vol. 26, no. 1, pp. 97–103. https://link.springer.com/article/10.1007/s10068-017-0013-6

Yu L, Chen W H, Sheen H K, Chang J S, Lin C S, Chyuan Ong H Loke Show, P , Chuan and Ling T 2020 Bioethanol production from acid pretreated microalgal hydrolysate using microwave-assisted heating wet torrefaction, Fuel, vol. 279, p. 118435, Nov. 2020. https://doi.org/10.1016/j.fuel.2020.118435

Lugert V, Steinhagen D and Reiser S 2020 Lack of knowledge does not justify a lack of action: the case for animal welfare in farmed fish, Landbauforschung: journal of sustainable and organic agricultural systems , no. 70(2020)1, pp. 31–34, Jun. 2020. https://doi.org/10.3220/LBF1592499937000

Mahfouz M E, Hegazi M M, El-Magd M A and Kasem E A 2015, Metabolic and molecular responses in Nile tilapia, Oreochromis niloticus during short and prolonged hypoxia, Marine and Freshwater Behaviour and Physiology , vol. 48, no. 5, pp. 319–340. https://doi.org/10.1080/10236244.2015.1055915

Maina J G, Beames R M, Higgs D, Mbugua P N, Iwama G and Kisia S M 2002 Digestibility and feeding value of some feed ingredients fed to tilapia Oreochromis niloticus (L.): Digestibility of feed ingredients fed to O. niloticus (L.), Aquaculture Research, vol. 33, no. 11, pp. 853–862. https://doi.org/10.1046/j.1365-2109.2002.00725.x

Makaras T, Razumienė J, Gurevičienė V, Šakinytė I, Stankevičiūtė M and Kazlauskienė N 2020 A new approach of stress evaluation in fish using β-d-Glucose measurement in fish holding-water, Ecological Indicators, vol. 109, p. 105829. https://doi.org/10.1016/j.ecolind.2019.105829

Mmanda F P, Lindberg J E, Norman Haldén A, Mtolera M S P, Kitula R and Lundh T 2020 Digestibility of Local Feed Ingredients in Tilapia Oreochromis niloticus Juveniles, determined on Faeces Collected by Siphoning or Stripping, Fishes, vol. 5, no. 4, p. 32. https://doi.org/10.3390/fishes5040032

Maynard L A and Loosli J K 1974 Animal nutrition, 6th ed., 10th print. New York [etc.]: McGraw-Hill.

Michelato M, Vítor de Oliveira Vidal L, Orlandi Xavier T, Batista de Moura L, Losi Alves de Almeida F, Breno Pedrosa V, Rossetto Barriviera Furuya V and Massamitu Furuya W 2016 Dietary lysine requirement to enhance muscle development and fillet yield of finishing Nile tilapia, Aquaculture, vol. 457, pp. 124–130. https://doi.org/10.1016/j.aquaculture.2016.02.022

Muhamed A, Refaey M M, Salem M F and El-Kattan M A M 2019 Factors Affecting Fish Blood Profile: A- Effect of Nutritional Treatments, Egypt. J. of Aquatic Biolo. and Fish., vol. 23, no. 2, pp. 517–526, Jun. 2019. https://doi.org/10.21608/ejabf.2019.33741

Munguti J,Charo-Karisa H, Opiyo M, Ogello E, Marijani E, Nzayisenga L and Liti D 2012 Nutritive value and availability of commonly used feed ingredients for farmed nile tilapia (Oreochromis niloticus L.) and African catfish (Clarias gariepinus, Burchell) in Kenya, Rwanda and Tanzania, AJFAND, vol. 12, no. 51, pp. 6135–6155.

Mussatto S I 2014 Brewers spent grain: a valuable feedstock for industrial applications: Brewers spent grain and its potential applications, J. Sci. Food Agric., vol. 94, no. 7, pp. 1264–1275. https://doi.org/10.1002/jsfa.6486

Mussatto S I, Dragone G and Roberto I C 2006 Brewers spent grain: generation, characteristics and potential applications, Journal of Cereal Science, vol. 43, no. 1, pp. 1–14, Jan. 2006. https://doi.org/10.1016/j.jcs.2005.06.001

Nguyen H C, Chen C-C, Lin K-H, Chao P-Y, Lin H-H and Huang M-Y 2021 Bioactive Compounds, Antioxidants and Health Benefits of Sweet Potato Leaves, Molecules, vol. 26, no. 7, p. 1820. https://doi.org/10.3390/molecules26071820

Nhi N H Y, Da C T, Lundh T, Lan T T and Kiessling A 2018 Comparative evaluation of Brewers yeast as a replacement for fishmeal in diets for tilapia (Oreochromis niloticus), reared in clear water or biofloc environments, Aquaculture, vol. 495, pp. 654–660.

Niyibizi L, Vidakovic A, Norman Haldén A, Rukera Tabaro S and Lundh T 2022 Aquaculture and aquafeed in Rwanda: current status and perspectives, Journal of Applied Aquaculture, pp. 1–22. https://doi.org/10.1080/10454438.2021.2024315

NRC 2011 Nutrient Requirements of Fish and Shrimp. Washington, D.C.: National Academies Press, 2011, p. 13039.

Nyina-Wamwiza L, Wathelet B, Richir J, Rollin X and Kestemont P 2009 Partial or total replacement of fish meal by local agricultural by-products in diets of juvenile African catfish (Clarias gariepinus): growth performance, feed efficiency and digestibility: Locals by-products in diets of C. gariepinus, Aquaculture Nutrition, vol. 16, no. 3, pp. 237–247. https://doi.org/10.1111/j.1365-2095.2009.00658.x

Oliva-Teles A.and Gonçalves P 2001 Partial replacement of fishmeal by brewers yeast (Saccaromyces cerevisae) in diets for sea bass (Dicentrarchus labrax) juveniles, Aquaculture, vol. 202, no. 3–4, pp. 269–278. https://doi.org/10.1016/S0044-8486(01)00777-3

Ogello E O, Kembenya E M, Githukia C M, Aera C N, Munguti J M and Nyamweya C S 2017 Substitution of fish meal with sunflower seed meal in diets for Nile tilapia (Oreochromis niloticus L.) reared in earthen ponds, Journal of Applied Aquaculture, vol. 29, no. 1, pp. 81–99. https://doi.org/10.1080/10454438.2016.1275074

Øverland M, Karlsson A, Mydland L T, Romarheim O H and Skrede A 2013 Evaluation of Candida utilis, Kluyveromyces marxianus and Saccharomyces cerevisiae yeasts as protein sources in diets for Atlantic salmon (Salmo salar), Aquaculture, vol. 402–403, pp. 1–7. https://doi.org/10.1016/j.aquaculture.2013.03.016

Pezzato L E, de Miranda E C, Barros M M, Pinto L G Q, Furuya W M and Pezzato A C 2002 Digestibilidade Aparente de Ingredientes pela Tilápia do Nilo (Oreochromis niloticus), R. Bras.Zootec., vol. 31, no. 4, pp. 1595–1604. https://doi.org/10.1590/S1516-35982002000700001

Pilarski F, Ferreira de Oliveira C A, Darpossolo de Souza F P B and Zanuzzo F S 2017 Different β-glucans improve the growth performance and bacterial resistance in Nile tilapia, Fish & Shellfish Immunology, vol. 70, pp. 25–29. https://doi.org/10.1016/j.fsi.2017.06.059

Puligundla P, Mok C and Park S 2020 Advances in the valorization of spent brewers yeast, Innovative Food Science & Emerging Technologies , vol. 62, p. 102350. https://doi.org/10.1016/j.ifset.2020.102350

Qiang J, Yang H, Wang H, Kpundeh M D and Xu P 2012 Interacting effects of water temperature and dietary protein level on hematological parameters in Nile tilapia juveniles, Oreochromis niloticus (L.) and mortality under Streptococcus iniae infection, Fish & Shellfish Immunology , vol. 34, no. 1, pp. 8–16. https://doi.org/10.1016/j.fsi.2012.09.003

Rawling M D, Pontefract N, Rodiles A, Anagnostara I, Leclercq E, Schiavone M, Castex M and Merrifield D L 2019 The effect of feeding a novel multistrain yeast fraction on European seabass (Dicentrachus labrax) intestinal health and growth performance, J World Aquacult Soc, vol. 50, no. 6, pp. 1108–1122. https://doi.org/10.1111/jwas.12591

Rey Vázquez G and Guerrero G A 2007 Characterization of blood cells and hematological parameters in Cichlasoma dimerus (Teleostei, Perciformes), Tissue and Cell, vol. 39, no. 3, pp. 151–160. https://doi.org/10.1016/j.tice.2007.02.004

Robertson J A, K I'Anson JA, Treimo J, Faulds C B, Brocklehurst T F, Eijsink V GH and Waldron K W 2010 Profiling brewers spent grain for composition and microbial ecology at the site of production, LWT - Food Science and Technology, vol. 43, no. 6, pp. 890–896. https://doi.org/10.1016/j.lwt.2010.01.019

Rusia V and Sood S K 1992 Routine hematological tests. In: Mukerjee Kanai, L. (Ed.), Medical Laboratory Technology, vol. 1. Tata McGraw Hill Publishing Company Limited, New Delhi, pp. 252–258 (fifth reprint).

Santiago C B and Lovell R T 1988 Amino Acid Requirements for Growth of Nile Tilapia, The Journal of Nutrition, vol. 118, no. 12, pp. 1540–1546. https://doi.org/10.1093/jn/118.12.1540

Santos M, Jiménez J J, Bartolomé B, Gómez-Cordovés C and del Nozal M J 2003 Variability of brewers spent grain within a brewery, Food Chemistry , vol. 80, no. 1, pp. 17–21. https://doi.org/10.1016/S0308-8146(02)00229-7

Schiemann R, Jentsch W, Hoffmann L and Nehring K 1966 Die energetische Verwertung der Futterstoffe, Archiv für Tierernaehrung , vol. 16, no. 2–3, pp. 173–198. https://doi.org/10.1080/17450396609421407

Seibel H, Baßmann B and Rebl A 2021 Blood Will Tell: What Hematological Analyses Can Reveal About Fish Welfare, Front.Vet. Sci., vol. 8, p. 616955. https://doi.org/10.3389/fvets.2021.616955

Short F J, Gorton P, Wiseman J and Boorman K N 1996 Determination of titanium dioxide added as an inert marker in chicken digestibility studies, Animal Feed Science and Technology, vol. 59, no. 4, pp. 215–221. https://doi.org/10.1016/0377-8401(95)00916-7

Stolen J S and Fletcher T C 1994 Eds., Modulators of fish immune responses. Fair Haven, NJ: SOS Publications.

Stoskopf M K 1993 Ed, Fish medicine. Philadelphia: W.B. Saunders Co. https://www.jstor.org/stable/1447277

Tacon A G J 2020 Trends in Global Aquaculture and Aquafeed Production: 2000–2017, Reviews in Fisheries Science & Aquaculture, vol. 28, no. 1, pp. 43–56. https://doi.org/10.1080/23308249.2019.1649634

Tacon A G J and Metian M 2008 Global overview on the use of fish meal and fish oil in industrially compounded aquafeeds: Trends and future prospects, Aquaculture, vol. 285, no. 1–4, pp. 146–158. https://doi.org/10.1016/j.aquaculture.2008.08.015.

Tacon A G J, Hasan M R and Metian M 2011 Demand and supply of feed ingredients for farmed fish and crustaceans. Rome: Food and Agriculture Organization of the United Nations.

Taira J, Taira K, Ohmine W and Nagata J 2013 Mineral determination and anti-LDL oxidation activity of sweet potato (Ipomoea batatas L.) leaves, Journal of Food Composition and Analysis, vol. 29, no. 2, pp. 117–125.https://doi.org/10.1016/j.jfca.2012.10.007

Teodósio R, Engrola S, Cabano M, Colen R, Masagounder K and Aragão C 2022 Metabolic and nutritional responses of Nile tilapia juveniles to dietary methionine sources, Br J Nutr, vol. 127, no. 2, pp. 202–213. https://doi.org/10.1017/S0007114521001008

van Kampen E J and Zijlstra W G 1961 Standardization of hemoglobinometry II. The hemiglobincyanide method, Clinica Chimica Acta, vol. 6, no. 4, pp. 538–544. https://doi.org/10.1016/0009-8981(61)90145-0

Vázquez-Ortiz F A, Caire G, Higuera-Ciapara I and Hernández G 1995 High Performance Liquid Chromatographic Determination of Free Amino Acids in Shrimp, Journal of Liquid Chromatography, vol. 18, no. 10, pp. 2059–2068. https://doi.org/10.1080/10826079508013960

Vidakovic A, Huyben D, Sundh H, Nyman A, Vielma J, Passoth V, Kiessling A and Lundh T 2020 Growth performance, nutrient digestibility and intestinal morphology of rainbow trout (Oncorhynchus mykiss) fed graded levels of the yeasts Saccharomyces cerevisiae and Wickerhamomyces anomalus, Aquacult Nutr, vol. 26, no. 2, pp. 275–286. https://doi.org/10.1111/anu.12988

Wilson R P, Robinson E H and Poe W E 1981 Apparent and True Availability of Amino Acids from Common Feed Ingredients for Channel Catfish, The Journal of Nutrition, vol. 111, no. 5, pp. 923–929. https://doi.org/10.1093/jn/111.5.923

Yue G H, Lin H R and Li J L 2016 Tilapia is the Fish for Next - Generation Aquaculture. Int J Marine Sci Ocean Technol. 3(1), 11-13. https://doi.org/10.19070/2577-4395-160003