Brewer’s yeast as a diet supplement in carp aquaculture: impact on production coefficients and haematological and biochemical plasma parameters

Abstract While brewer’s yeast (Saccharomyces cerevisiae), a rich source of proteins and B vitamins, is commonly used as an animal feed supplement, its use in aquaculture has been limited. Here, we assess its impact on 192 individuals of common carp (Cyprinus carpio) production and haematological and biochemical parameters from blood plasma using a control group fed commercial feed with 20% soybean meal and three test groups fed commercial feed with (i) 17.50% soybean + 2.50% brewer’s yeast (group B-2.5), (ii) 15% soybean + 5% yeast (group B-5) and (iii) 10% soybean and 10% yeast (group B-10). At the end of the experiment, all fish were weighed and measured and subsequently blood samples were taken from eight fish in each tank. There was a significant increase in total length (3%) and weight (5.37%) in group B-5. While fish muscle nitrogenous matter and protein efficiency ratio increased slowly across experimental groups, Fulton’s and Clark’s condition factors changed minimally (range 2.73–2.89%). Apparent net protein utilisation increased to 17.63% in B-10, suggesting that increasing yeast levels result in improved nitrogen utilisation. All fish muscle fat levels were within optimal range (4.16–4.68%) and, while there was a significant increase in erythrocytes in B-5 (up 0.40 T.L−1), haemoglobin levels increased only slightly in B-5 and B-10 (average 4.50 g.L−1) and B-10 had the lowest number of leukocytes (6.8 G.L−1). No significant changes were recorded in blood plasma biochemical parameters. Our results suggest that a 5% brewer’s yeast feed supplement represents a useful protein source for carp aquaculture. Highlights Replacement of soybean meal with brewer’s yeast increases carp production rates and reduces the feed coefficient ratio Brewer’s yeast can be an important source of protein for carps


Introduction
In terms of production, members of the carp family are some of the world's most important inland aquacultural species.In Asia and Africa, for example, rohu (Labeo rohita) and grass carp (Ctenopharyngodon idella) are both important species, while common carp (Cyprinus carpio), the dominant species utilised in Central Europe, has a global production rate of over 4.20 million tons, i.e. 8.60% of global fish production in 2020 (FAO 2021).As such, carp are one of the most economically important species of farmed fish, with the result that breeders are constantly trying to make carp breeding/production as efficient as possible.
In an effort to minimise the environmental burden and pharmaceutical or chemical residues in the food chain as much as possible, farmers are trying to minimise the industrial additives and focus on the use of natural additives in intensive farming.Feed additives such as herbal extracts, oils, acidifiers, prebiotics and probiotics now play an important role in commercial aquaculture worldwide (Oliva-Teles 2012;Ghafarifarsania et al. 2022).The last of these, probiotics, are live microorganisms such as yeasts, lactobacilli, bifidobacterium that are intended to have health benefits when consumed (or applied to the body) and, while they are found in some foods naturally, are often now added to the diet as supplements.Some probiotics compete with pathogenic bacteria stemming from the diet and their addition to feed can help in feed utilisation, boost animal growth and weight gain (Ghosh et al. 2003) and have a positive impact on the animal's immune system (White et al. 2002).Furthermore, they are relatively inexpensive as they are already produced on an industrial scale.One of the most popular probiotics is brewer's yeast (Saccharomyces cerevisiae), which contains a high proportion of dietary protein, making it a possible substitute for the main protein source in dietary feeds (Olvera-Novoa et al. 2002).Yeasts in general are a rich source of proteins and B vitamins and have been used as a feed complement to improve levels of amino acids (Egel-Mitani et al. 2000).Moreover, the cell wall of brewer's yeast contains a high number of bioactive compounds, including ß-glucans, which enhance resistance against bacterial and viral infections by improving the strength of the immune system (Chang et al. 2003;Lin et al. 2011).
While brewer's yeast has been widely used for improving dairy cattle performance (Rossow et al. 2018), fattening pigs (van Heugten et al. 2003) and increasing poultry production (Dixon et al. 2022), its use in aquaculture has been limited and mainly aimed at aquaculture of carnivorous fish such as Nile tilapia (Oreochromis niloticus; Trosvik et al. 2012) or rainbow trout (Oncorhynchus mykiss; Adel et al. 2017).The aim of this study, therefore, was to whether substituting the standard protein source in fish feed (soybean meal) with brewer's yeast has any positive impact on body weight gain and haematological and biochemical parameters in one-year old common carp.

Fish
One hundred and ninety-two clinically healthy common carp (Ropsin and Zdarsky scaled crossbreeds; originating from Pohorelice Fish Farming) were divided into four groups comprising three replications, with 64 carp per replication.For each test run, 16 weight balanced carp were placed into 180 l tanks fed by a Nexus Easy 210 recirculation system (Evolution Aqua s.r.o., Czech Republic), mean weights in each tank being 63.36 ± 5.79 g; 62.16 ± 5.83 g and 62. 74 ± 6.38 g in group C; 63.42 ± 7.07 g; 62.56 ± 5.42 g and 61.11 ± 4.51 g in group B-2.5; 61.81 ± 5.26 g; 62.82 ± 6.40 g and 62.51 ± 5.31 g in group B-5 and 62.59 ± 5.41 g; 63.34 ± 5.81 g and 62.76 ± 4.81 g in group B-10.Prior to each replication, the fish were given a two-week adaptation period, during which they were fed with standard soybean enriched KP1 feed mixture.

Dietary feed
Four diet variants were used in the experiment.The control (C) group were fed a commercially prepared carp feed (KP1) containing wheat, wheat flour, rapeseed cake, wheat bran, soybean meal, barley, corn and calcium carbonate enriched with 20% soybean meal (i.e.200 g of soybean meal per 1 kg of feed mixture) to increase content of nitrogenous matter (see Table 1).In addition, three test variants were prepared by the feed producer, where in part of the soybean meal in the same mixture above was substituted with brewer's yeast in the following proportions: 17.50% soybean meal and 2.50% brewer's yeast (group B-2.5); 15% soybean meal and 5% brewer's yeast (group B-5) and 10% soybean meal and 10% brewer's yeast (group B-10).Brewer's yeast was produced by firm Agro Mraz.

Feeding during the experiment
All test fish were fed three times a day at 8 a.m., noon and 4 p.m. every day over the 10-week experiment, with the daily feeding ratio corresponding to 3% of tank stock weight.Control weighing took place every 14 days (i.e.five times per experiment), after which the daily feed ration was recalculated based on the actual weight of the fish.The control days also served for examination of fish health.Note that no fish died during the experiment.

Water quality
Water temperature, dissolved oxygen content, oxygen saturation and pH were measured once a day before first feeding in each tank using a HACH HQ40D Multiparameter (HACH, Germany).Every second day, ammonium nitrogen (N-NH 4 þ ), nitrogen dioxide (N-NO 2 À) and chloride (Cl -) were determined using a PhotoLab 6600 UV-VIS 112 Spectrophotometer (WTW, Germany).There were no significant differences in water quality parameters throughout the experiment (Table 2).

Fish body parameters
At the end of the experiment fish were removed from water, stunned by a blow to the head and after the collection of the blood they were killed by cutting the branchial venae.This way of killing is in accordance with the legislation 'The prevention of cruelty to the animals act' (Law number 246/1992 Collection).Each fish was measured for total length, standard length, body height, body width, body weight, eviscerated body weight, liver weight, pancreas weight and gonad weight.Fulton's condition factor, Clark's condition factor, highbackedness index and widebackedness index were then calculated from these basic data (see Gela and Linhart 2000), along with the feed conversion ratio, specific growth rate, weight gain, hepatosomatic index, viscerosomatic index, splenosomatic index, gonadosomatic index, protein efficiency ratio, lipid retained and apparent net protein utilisation.Note that not all parameters are listed in the tables as some were only used for calculation of the indices.

Blood examination
Blood samples (2 mL) were taken from eight stunned fish in each tank at the end of the experiment by puncturing the vena caudalis and storing the sample in a heparinised syringes, rinsed out with heparin sodium salt to avoid coagulation.Each blood sample was divided into two parts, one being used for haematological examination (haemoglobin, number of erythrocytes, number of leukocytes, haematocrit, mean corpuscular haemoglobin concentration, mean corpuscular volume and mean corpuscular haemoglobin, while the other part was centrifuged (252 x g, 5 min) using a MPW 140 350 R cooling centrifuge (MPW Med.Instruments, Poland) and the obtained plasma was frozen and stored in a freezer (Arctiko ULTF 80, Denmark) at À75 C until further analysis.Blood smears, haemoglobin and haematocrit determination (Svobodov a et al. 2012) were made immediately after blood collection.Blood smears were stained using the Hemacolor Rapid staining kit (Merck, Darmstadt, Germany.The biochemical profile of blood plasma (alanine aminotransferase (ALT), aspartate aminotransferase (AST), lactate dehydrogenase (LDH), alkaline phosphatase (ALP), albumin (ALB), cholesterol (CHOL), creatine (CREA), glucose (GLUC), urea (UREA), total protein (TP), triacylglycerol (TAG), calcium (Ca) and inorganic phosphorous (PI) and magnesium (Mg)) was measured using the Konelab 20i kit and commercially available kits (BioVendor, Czech Republic).

Statistical analysis
Production parameters (i.e.feed conversion ratio, specific growth rate, weight gain), condition (Fulton's condition factor, Clark's condition factor, highbackedness index and widebackedness index) and biochemical parameters (mentioned above) calculated for each fish and mean levels calculated per tank, allowing comparisons between test treatments.One-way ANOVA with Tukey post-hoc tests was used to determine differences between experimental variants, with all data being first log (x þ 1) transformed to meet the assumptions of ANOVA.All analyses were performed in Statistica 13 (TIBCO Software Inc. 2018).In all cases, differences were considered significant at p < .05.

Fish body parameters
Carp in groups B-5 and B-10 had a significantly greater body length than those in groups C and B-2.5 (F ¼ 26.0, d.f.¼ 3, p < .001).A similar trend was also observed for body weight (F ¼ 5.0, d.f.¼ 3, p ¼ .03);however, unlike body length it was not possible to determine significant differences between dietary treatments (Table 3).Values for both Fulton's condition factor and Clark's condition factor tended to decrease with increasing feed yeast content, with values for the B-10 group being significantly lower than the control and B-2.5 and B-5 groups (Fulton: 3).While the hepatosomatic and viscerosomatic indices showed no discernible trend in relation to feed yeast content, the splenosomatic index increased slightly in groups B-5 and B-10, and the gonadosomatic index decreased slightly in all three treatments, though not significantly (Table 3).While the widebackedness index showed a decreasing trend with increasing feed yeast content, with B-10 being significantly lower than the other treatments and control (F ¼ 7.0, d.f.¼ 3, p ¼ .015), the highbackedness index showed the opposite trend, with values in B-5 and B-10 being higher than those in B-2.5, though not significantly so (F ¼ 2.8, d.f.¼ 3, p ¼ .11;Table 3).

Production parameters
None of the production parameters examined displayed statistically significant changes against the control with increasing feed yeast levels (Table 4).Overall trends were variable, however, particularly those for fat in fish muscle and liver.While lipids and fat in viscera (except for liver) showed a continuous decreasing trend with increasing feed yeast, ash in fish muscle showed a continuous increasing trend.Nitrogenous matter and fat in fish muscle both showed an increasing trend to B-5 that dropped slightly in B-10.Feed conversion ratio showed a slight decreasing trend to B-5 that increased slightly at B-10, while specific growth rate showed the opposing trend, increasing to B-5 and then decreasing slightly, resulting in an overall declining trend in FCR/SGR.A similar trend to SGR was also visible in weight gain and the protein efficiency ratio.Finally, apparent net protein utilisation values increased with increasing feed yeast levels, levelling off between B-5 and B-10 (Table 4).Note that the same superscript letter ( a ) across all experimental diets indicates that, while there was a significant effect in the treatment overall, we were unable to distinguish between diets.

Haematological and biochemical parameters
While the number of erythrocytes dropped slightly below those in the control in groups B-2.5 and B-10, numbers in group B-5 were significantly higher than all other groups (F ¼ 5.57, d.f.¼ 3, p ¼ .02,Table 5).None of the other haematological parameters examined displayed statistically significant changes against the control with increasing feed yeast levels (Table 5).Haematocrit and mean corpuscular haemoglobin concentration values showed no change with increasing feed yeast levels.Haemoglobin and number of leucocytes showed opposing trends, with haemoglobin levels increasing slightly against the control and leucocyte numbers decreasing; in both cases groups B-5 and B-10 had equally high and low values, respectively (Table 5).Both mean corpuscular volume and mean corpuscular haemoglobin showed similar trends, with levels increasing in B-2.5, decreasing sharply in B-0.5 and again increasing in B10, particularly for mean corpuscular haemoglobin (Table 5).
None of the biochemical parameters examined displayed statistically significant changes against the control with increasing feed yeast levels (Table 6).Both ALB and GLUC declined slightly at B-2.5, increased above control levels at B-5 and dropped below again at B-10; in comparison, ALP and CHOL showed the opposing trend, with levels increasing at B-2.5, decreasing at B-5 and increasing again at B-10 (Table 6).Both Ca and CDH levels initially declined but increased above the control by B-10, while AST initially increased at B-2.5, but then declined continuously below the control (Table 6).While Mg and PI both showed a very slight increasing trend with feed yeast levels, ALT, TP and TAG levels all remained close to the control (Table 6).In contrast, both UREA and, especially, CREA showed a continuous decreasing trend with increasing feed yeast levels (Table 6).

Discussion
Substituting soybean for brewer's yeast in carp feed had a positive impact on growth of the fish in this study, carp in the B-5 and B-10 increased their length about 2.80% and weight about 5.40% comparing to fish in the control group, apparently due to improved conversion of protein from the yeast.Zhang et al. (2020) recorded similar results to ours in a 63-day trial using the brewer's yeast additive in the diet of gibel carp (Carassius auratus gibelio), with the group having a 3% addition having the highest final body weight (7% higher comparing to the control group) at the end of their experiment.On the other hand, Korkmaz and Cakirogullari (2011) failed to record any significant increase in koi carp fingerling growth performance after adding dried baker's yeast to the diet; indeed, they recorded a significant reduction in weight gain after a 40% addition.Likewise, in the present study, Fulton's and Clark's condition factors, both calculated from fish length and weight, showed very little change over the course of the study, though there was a very slight decrease in body condition in the B-10 group.There was also little or no evidence of any trend in the hepatosomatic, viscerosomatic or splenosomatic indices.In a separate study, Dob s ıkov a et al. ( 2013) also found no significant change in biometric indices (i.e.hepatosomatic and splenosomatic index and Fulton's condition factor) after feeding carp with a mixture enriched with beta-1.3/1.6-D-glucan which are the main component of yeast cell walls.
Overall, there were no obvious trends in the levels of fat in fish muscle, with values fluctuating between 4.16 and 4.68% among groups, i.e. within the optimal range (Taheri Mirghaed et al. 2017).Muscle fat content is an important parameter in fish as it affects the nutritional, technical and sensory qualities of flesh (Zheng et al. 2016), with retention of unsaturated omega-3 and À6 fatty acids being especially important in relation to human health (Bohm et al. 2014).Consequently, the decreasing trend in fat retention with increasing brewer's yeast concentration noted in this study can be seen as positive as increased fat retention in the abdominal cavity is considered undesirable comparing to fat retention in the muscle.Fat content in the muscle is the most important bute of flesh quality because it is associated with flesh texture, flavour and juiciness (Johansson et al. 2000).
Nitrogenous matter in fish muscle, together with protein efficiency ratio values, showed a gradual increasing trend in all experimental groups except B-10, a similar result to those of Zhang et al. (2020), who also noted increases in these parameters with a 5% addition of yeast, with no further increase at higher dose rates.A similar pattern in apparent net protein utilisation as yeast concentration in the diet increased (i.e.increasing then levelling off at B-5 to B-10) suggests that carp are better able to convert nitrogenous matter and retain it within the body at higher yeast concentrations.This confirms the findings of previous studies that have found that addition of yeasts to the diet both improves feed utilisation and enhances fish growth (Wang and Xu 2006;Sealey et al. 2009).As with the results outlined above, the values for feed conversion ratio and specific growth rate both indicated best results at 5% yeast addition (B-5); thus, a 5% brewer's yeast in carp feed would appear to be both the most efficient proportion and the most economic from a production point of view.
Significant changes to the diet can be a source of stress for fish, which can be measured through changes in plasma constituent concentrations, changes in the size and number of blood cells or through functional changes in vital organs such as the gills, kidney or intestines (Ahmed and Jaffar 2013).
Based on previous studies (e.g.Seibel et al. 2021), we chose to monitor haematological changes to evaluate the influence of dietary stress (i.e.increasing proportions of yeast in the diet) on the body's state, and particularly, any increase in haemoglobin concentrations (Kopp et al. 2014;Palikova et al. 2015).As our results showed a slight, though non-significant, increase in haemoglobin values in groups B-5 and B-10 (average 4.50 g.L À1 ), this would suggest that the addition of yeast has no negative impact on carp stress levels and therefore health.This is also supported by the significant increase in erythrocytes in group B-5 and the decrease in number of leukocytes in all experimental groups, with lowest counts in group B-10.These findings agree with those of Bozorgnia et al. (2011), who recorded increased numbers of red blood cells and thus an increase in haemoglobin level, in carp under stress conditions.As the number of leukocytes, i.e. white blood cells, tends to increase in fish under stress (Roberts 2012), the decrease in leukocytes in all experimental groups with increasing yeast concentration observed in this study would also suggest dietary yeast supplements are beneficial for carp.
Plasma ALT, AST and ALP are cytosolic enzymes found in many tissues and consequently are used as blood plasma indicators of tissue health (Haschek et al. 2009).ALT is found at high concentrations in fish liver and any elevation in levels in the blood is indicative of hepatocyte damage.ALP, on the other hand, is found at high concentration in erythrocytes and cases of haemolysis will result in blood ALP elevation.Finally, increased levels of AST in the blood will indicate damage to fish tissues (Taheri Mirghaed et al. 2017).AST also plays a role in glucose production from amino acids; thus, can measurement of blood AST, along with a stress marker such as blood glycaemic state, can be used to diagnose stress in fish (Tejpal et al. 2009).In the present study, none of these indicators showed any significant increase in blood plasma, and glucose levels fluctuated only slightly between groups, indicating that there was no change in carp health/stress status as yeast levels increased.Similarly, we observed no significant changes in plasma LDH, ALB, CHOL, TP, TAG, Ca, PI and Mg.On the other hand, there was a decreasing trend in UREA and, especially, CREA levels with increasing yeast concentration, thou in both cases, the changes were not significant.Concentration of UREA and CREA in the blood is dependent on the protein content of food and healthy excretion by the kidneys and metabolic functioning of the liver (Schrama et al. 2018).As differences in values between test groups were non-significant, it can be assumed that there was no decline in protein availability in the feed and that partial replacement of soybean meal with brewer's yeast had no negative effect on fish health.

Conclusion
Replacement of soybean meal with 5% brewer's yeast resulted in increased carp production rates, with an invariable fat retention in the muscles and increased protein retention, alongside a reduction in the feed coefficient ratio.The addition of yeast had no negative impact on any haematological or biochemical plasma parameters.Addition of 5% of yeast into the diet decreased the number of erythrocytes and values of mean corpuscular volume and mean corpuscular haemoglobin.The reason for decrease in these parameters will be the subject of further study.Our results indicate that brewer's yeast can be an important source of available protein for carp and, therefore, can be used to partly replace soybean in aquacultural fish feed.

Table 1 .
Analytical composition of the feed mixture used in this study.

Table 2 .
Mean water quality parameter values for each test treatment over the course of the experiment.
Notes: Water temperature, dissolved oxygen content, oxygen saturation and pH were measured once a day.Every second day, ammonium nitrogen (N-NH 4 þ ), nitrogen dioxide (N-NO 2-) and chloride (Cl-) were determined.

Table 3 .
Length-weight and condition indices of common carp used in this study (mean ± SD, n ¼ 12).Significant differences (p < .05;one-way ANOVA with Tukey post-hoc tests) indicated by different superscript letters.

Table 4 .
Production parameters for the common carp used in this study (mean ± SD, n ¼ 3).
Note: No statistically significant difference was demonstrated in the parameters.

Table 5 .
Haematological parameters for the common carp used in this study (mean ± SD, n ¼ 12).MCHC: mean corpuscular haemoglobin concentration; MCV: mean corpuscular volume; MCH: mean corpuscular haemoglobin.Significant differences (p < .05;one-way ANOVA with Tukey post-hoc tests) indicated by different superscript letters.No statistically significant difference was demonstrated in the parameters.

Table 6 .
Biochemical parameters for the common carp used in this study (mean ± SD, n ¼ 7).