Fish is the most popular cheap animal protein that is consumed in Sierra Leone (Spring, 2015). Fish and other aquatic products contribute significantly to the economy of Sierra Leone through export earnings, job creation, as well as food security and nutrition. The government of Sierra Leone regards the fisheries sector as a pivot for development because of its socio-economic benefits (FAOSTAT. 2016 cited by Kassam et al., 2017). This sector is not only an important source of income and employment but also provides the most important animal-source food in the diets of Sierra Leoneans, providing about 80% of animal protein intake. Sierra Leone fisheries sector is one of the main contributors to the national economy, making up about 10% of GDP. More than 80% of the fish consumed in the country comes principally from the marine fisheries, with insignificant proportion contributed by the inland freshwater environment. Little or nothing is contributed by the ill- developed aquaculture sector. With rising population coupled   with   increasing    demand    for    fish,    recent assessment has indicated increased overfishing and a shortfall in landing from the marine environment (Hilborn et al., 2020); a pointer to the urgent imperative to develop the aquaculture sector, as an indisputable means to fill the gaps between demand and supply (WorldFish, 2016). It is believed that a sufficiently developed aquaculture sector will contribute to reducing the pressure on capture marine environments. Increased production from aquaculture is also expected to greatly complement current efforts aimed at achieving the United Nations Millennium Development Goals (MDGs), especially the eradication of extreme poverty and hunger, abatement of child mortality, improvement of maternal health, prevention of diseases, promotion of gender equity and women empowerment.

Sierra Leone, although blessed with abundant production resources, its aquaculture sector has been unfortunately dwarfed for many by the twin constraints of lack of improved fish seeds and affordable nutritionally balanced fish feed. In the recent years, Njala University has been leading the drive to solving the problems of African catfish (Clarias gariepinus) and Nile tilapia (Oreochromis niloticus) fingerlings production. Both fish species are also important in world aquaculture as good culture candidates, especially in Africa (FAO, 2018). It is unfortunately saddening to note that fish feed is still being imported at the cost that is beyond the reach of the ordinary poor farmers in Sierra Leone. The exorbitant cost of imported feed has forced fish farmers in the country to continue feeding their fish with rudimentary feed alternatives like termites, rice bran and household wastes. The result of this discordant feeding practice is the underperformance of the sector. The non-availability of cheaper and quality feeds, according to Siddhuraju and Becker (2003) is a major constraint to the development of aquaculture in developing countries. As noted by Babalola (2010), feed in fish production makes up between 60 and 75% in many African countries, which implies that would be investor has to plan diligently in order to succeed in the industry.

The quality fish feed plays a significant role in the growth, development and profitability of fish farming; and according to Tsevis and Azzaydi (2000), it is an important factor to be considered in both subsistent and commercial fish farming. Thus, the anticipated success desired for the sector will significantly depend on providing quality and least cost fish feed (Delgedo and Minot, 2003). Glencross et al. (2020) cited by Hoerterer et al (2021) posited that the challenge of feeding farmed fish with diets that are nutritious but at the same time economically and environmentally sustainable is a major issue to ponder.

The effort to resolve this challenge can only be achieved by relying on the use of cheaper, locally available feed ingredients; integrating locally available feed ingredients in feed formulation in the right proportion supplemented with crystalline Essential Amino Acids (EAAs)   based   on   the  fish  requirements  (Tacon  and Metian, 2008; Macfadyen et al., 2012). Efforts have been made in recent years, through research to identify major nutritional requirements for important farmed fish (FAO, 2020; Naylor et al., 2021). Replacement of the expensive and high in demand fishmeal and fish oil in fish feed formulation is the focus of most fish feed researchers (Sattang et al., 2021).

The African catfish is native to tropical environment, preferred by farmers and easily cultured with fast growth potential when fed with nutritionally balanced diets. Clarias spp. command good price in Sierra Leone and is relished by consumers in the country because of its taste. The staffs of the Department of Aquaculture and Fisheries Management, Njala University has been working relentlessly for more than eight years to make the culture of African catfish and Nile tilapia easy for local farmers through research on improved fish strains and quality feeds. The objective of this study was therefore to develop least-cost and nutrient-rich fish feed for the farming of African catfish (C. gariepinus) in Sierra Leone. The developed feed was compared with two imported feed on the basis of their proximate contents as a way of determining the suitability of the developed feeds for farming of African catfish in Sierra Leone.

Study area          

This study was carried out at the Department of Aquaculture and Fisheries Management Njala University, Njala Campus, Njala, Kori chiefdom, Moyamba district, Southern Sierra Leone. Njala University is located on latitude 7.25, and longitude 11.2 with the GPS coordinates of 07° 15′ 00.00″ N and 11° 12′ 00.00″ W.

Constituents of diets

Feed ingredients for the formulation of 45% crude protein (CP) least-cost and nutrient-rich catfish feed for the use of farmers in Sierra Leone were purchased from local market in the country, while fish protein concentrate (FPC booster), vitamin premix, lysine and methionine were imported. Feed ingredients used were fish meal, palm kernel cake (PKC), rice bran, methionine, lysine, vitamin premix, Yeast, salt (NaCl), starch and FPC booster (Table 1).

Diet preparation

Feedstuff were measured out in predetermined proportion, ground finely and mixed thoroughly in a clean plastic basin. Warm water was added to the ingredients and mixed to form a dough. The dough was passed through a manual pelleting machine, which gave a spaghetti-like product. The pelleted products were covered with transparent polythene sheet and kept indoor for 30 min and later sundried to obtain a constant storage weight. The well dried formulated feeds were stored in air-tight transparent polythene bags and kept in a cool dry place prior to proximate determination.

Vitamin premix

Vitamin A: 12,000,000 IU, Vitamin D3: 4,000,000 IU, Riboflavin: 8 g, d-Pantothenic acid: 24 g, Choline: 1,400 g, Niacin: 100 g, Vitamin E: 100 g, Vitamin K: 4 g, Vitamin C: 500 g, Folic acid: 1 g, Pyridoxine: 5 g, Thiamine: 5 g, BHT: 5 g.

Fish protein concentrate - liquid

Water max 65%; protein min 25%; Fat min 4%; Ash max 8% and phosphoric acid max 5%

Proximate determination of diets

Dry matter, moisture, crude protein, fat, crude fibre, ash and carbohydrate, calcium, potassium and sodium were determined for this study using procedures prescribed by AOAC (2000).

Determination of moisture content (%)

Feed samples (5 g each) were weighed into a clean pre-weighed crucible and dried in an oven at 105°C to constant weight. The moisture content was then calculated using this equation:

Determination of total ash (%)

Feed samples, 5 g each for the determination of total ash were weighed into clean pre-weighed crucibles and converted to ash in a furnace at 550°C to a constant weight. Total ash was then calculated as:

Determination of fat (%)

Feed sample (5 g) was massed into a polypropylene centrifuge bottle. Sodium acetate, aliquots of methanol, chloroform  and  water was added into the bottle and shaken for 30 min. The content of the bottle were centrifuged at 2500 rpm for 10 min, after which it was set in 25°C water bath for 15 min. The samples was evaporated to dryness under nitrogen blanket, heated in a drying oven for 30 min, and cooled in a desiccator for 30 min. Fat content was then determined as:

Where, W2 was the weight of glass tube and dried extract (g), W1 was the weight of empty dried glass tube (g), Vc was the total volume of chloroform (ml), VA was the volume of extract dried (ml), and SW was the weight of the sample in grammes.

Determination of protein (%)

Concentrated sulphuric acid (H₂SO₄) (20 ml) was carefully added into tube containing 1 g sample of feed and Cu (catalyst salt), and then shaken gently to aid digestion. Digested sample was cooled for 10 to 20 min. Distillation procedure was then performed using distillation unit and the distillate was titrated with 0.025N sulphuric acid (H₂SO₄) until the end point changed from green to pink. Volume of acid required in the titration was recorded. Blank was prepared with the exclusion of sample. The percentage of protein content was calculated using the equation below:

Where, VD is the Volume of digest; N is the normality of acid; TV is the titre value; AD is the aliquot of digest and F is the conversion factor for nitrogen to protein.

Determination of fibre

Feed sample (5 g) was weighed into a litre conical flask containing 200 ml of boiling sulphuric acid and further boiled for 30 min over a burner.  Swirling   is   done   occasionally   to   remove   solids  from  adhering to the sides of the flask. The hot solution was decanted through Buchner funnel fitted with Whatman 52 filter paper. All residues were rinsed with boiling water until there was no colour change in litmus paper, an indication that acid has been completely removed from the residue. The residue was transferred into a 200 ml of 1.25% NaOH solution in a litre flask and brought to boil. The solution was thereafter maintained at a gentle ebullition for 30 min and subsequently filtered using rapid hardened filter paper. 1% HCl and distilled water, 15 ml of alcohol and 10 ml of diethyl ether were also added. The sample was dried in an oven at 100°C for 1 h and allowed to cool in a desiccator and weighed (W1). Sample was put in a crucible in a furnace at 55°C for 3 to 4 h; it was cooled in a desiccator and weighed again (W2). Fibre was calculated thus:

Determination of carbohydrate (%)

Carbohydrate was determined as presented here:

Carbohydrate = 100% - (% moisture + % ash + % crude protein + % fat)

Statistical analysis

One-way ANOVA and simple statistical measure of mean, percentages and standard deviation were used to analyse the results of proximate determination.

Proximate and chemical analysis of formulated feed

The highest mean percentage of crude protein obtained was higher at 44.48 ± 0.40% for Diet 5, while the least CP of 40.59 ± 2.22% was obtained for Diet 3. These figures were not significantly different (P>0.05) from the 45% crude protein declared by Ranaan and Aller Aqua. The protein level obtained in this study agrees with the recommendation of National Research Council (NRC, 2011) of 35 to 40% for warm water fishes. According to Wilson (2000), most of the commercial fish feed, especially for catfish contain 32% CP. Boonyaratpalin (1988) estimated the protein requirement for tropical catfish to be 35 to 40, 25 to 35 and 28 to 32% for fry, grow-out and brood stock respectively. The result of this study agrees with that of Wantanabe et al. (1990) who reported that the utilization of protein of 35% and above inclusion in fish feed increase fish production. It has also been shown by various workers that fish growth performance is a function of the protein inclusion in the feed (Siddiqui et al., 1998; Giri et al., 2003; Ali and Jauncey, 2004a, Goda et al., 2007; Corn’elio et al. 2014; Keremah and Beregha, 2014; Olapade and Quinn, 2019).

In this study, fishmeal, amino acids and FPC booster were the sources of protein; Raanan utilized fishmeal and poultry waste meal; and Aller Aqua utilized the combination  of  animal  and  plant  sources  protein  (fish meal, soybean meal, feather meal, poultry meal, poultry blood meal and sunflower protein concentrate). The decision to utilize amino acids and FPC in this study is owing to the fact that some of the ingredients used by both Raanan and Aller Aqua are not readily available in Sierra Leone. The use of animal blood in fish feed although is cheaper required some caution especially in a country like Sierra Leone where it may be impossible to gather sufficient quantity for commercial use. The inadequacy of appropriate technology to process animal blood into meal may dispose it to contamination, which will invariably render it unusable for the culture species.  Groundnut oil cake (GNC) and soybean meal in combination with fishmeal is good in fish feed formulation (Audu et al., 2004), it is however important to mention that abnormal inclusion has been associated with reduced growth of fish as a result of poor feed utilization and digestion by the fish (NRC, 1993; Jauncey, 1998; Nyina-Wamwiza et al., 2007; Agbo et al., 2011; Olapade and Lombi, 2015; Olapade and Kargbo, 2015; Olapade and Quinn, 2019).

The result of this study showed that Diet 4 has the highest fibre value of 3.28 ± 0.15 and a least mean value of 2.81 ± 0.14 for Diet 1, slightly higher than for both Ranaan and Aller Aqua. Mean carbohydrate (CHO) value of 39.29 ± 2.24 was for Diet 3, while 36.76 ± 5.37 was the least obtained for Diet 5. These values far exceeded those of Ranaan and Aller Aqua and are beyond the recommended levels by NRC (2011). It is suggested, that a typical catfish feed should contain 25% or more soluble (digestible) carbohydrates, in addition to 3 to 6% more carbohydrates that are generally present as crude fibre (mainly cellulose). Fibre is a necessary component of plant ingredient, and they are useful in feed as binder. De Silva and Anderson (1995) opined that fibre contents of fish feed should not exceed 8 to 12%. Excess of fibre has been associated with reduced quality unusable nutrient in the diet, and has also been ascribed poor growth of C. gariepinus fingerlings (Adewolu et al., 2010; Agbabiaka et al., 2013). The implication of high level energy relative to protein in fish feed are lipid deposition, decreased feed intake and reduced weight gain and vice versa (Craig, 2017). Consequently, there is the need to balance the energy-to-protein ratio of the diets in this study, as this would help to reduce the carbohydrate and fibre level to acceptable level especially for African catfish which requires more protein than carbohydrate and fibre.

The highest fat obtained for Diet 5 (3.16 ± 0.01%), and a least value of 0.30 ± 0.07% recorded for Diet 2 in this research were significantly below those of Ranaan and Aller Aqua (Table 2). Fat values for Diet 1 (0.30 ± 0.07%), Diet 2 (0.72 ± 0.02%) and Diet 3 (0.39 ± 0.00%) are not significantly varied (P>0.05). Diet 5 was however significantly higher (P<0.05) (3.16 ± 0.01). The nature of fish tissue fatty acids is markedly influenced by dietary lipids. Fish can synthesize saturated fatty acids and members of the w9 and w7 families of unsaturated acids; however,  a dietary  source is required for the w3 (Alpha- linolenic acid - ALA, Eicosapentaenoic acid - EPA,   and Docosahexaenoic acid - DHA) and w6 (Arachidonic and Linoleic acid) families to be present in fish tissues (NRC, 1977). Cowey and Sergent (1979) reported that 10 to 20% of lipid in most freshwater fish diets gives optimal growth rates without producing an excessively fatty carcass. Wilson (2000) reported that lipid level in catfish feeds should be 5.6%.

Yildirim-Aksoy (2009) noted that dietary lipids are important sources of highly digestible energy and according to him; it is the only source of the essential fatty acids needed by fish for normal growth and development. Thus, it can be said that lipids or fatty acids is good in the diet of fish at a certain level of inclusion. This study will strive to improve on the fatty acids level by reducing the non-fatty extract (NFE) and dry matter content of this test feed.  The percentage moisture values obtained in this study, which ranged from 19.49 ± 0.29 to 21.98 ± 0.15 was higher than those of Ranaan feed and Aller Aqua feed. Moisture is of great importance in storage of feed. Improperly dried feed besides having a short shelf life will be infested by mould with possibility of developing mycotoxin.

Mineral contents of formulated diets

Aquatic animals need minerals for their physiological  and biochemical functions, and also to maintain their normal life processes (Lall and Kaushik, 2021). According to Craig (2017), minerals are inorganic elements that regulate osmotic balance and aid in bone formation and integrity. In this study, phosphorus (µg/g), sodium (µg/g) and calcium (%) were evaluated (Table 3). Phosphorus was determined for both Ranaan and Aller Aqua and the values of 1.3 and 1% reported respectively for the two feeds are below what was obtained in this study. Percentage of Na and Ca were only reported for Aller Aqua as 0.2 and 0.5%, respectively. The essentiality of macro-minerals (calcium, phosphorus, magnesium, sodium, potassium and chloride) has been confirmed in fish (NRC, 1993; NRC 2011). Lall and Kaushil (2021) however reported that gaps still exist in the knowledge of mineral nutrition of fish and shrimp with respect to their dietary requirements, physiological functions, and absorption from the gastrointestinal tract and bioavailability from feed ingredients. Even with the gaps, essential macro- and micro-minerals are required at optimum levels for growth and maintenance of normal health of farmed fish.

Cost consideration of formulated diets

The costs of producing the 45% CP diets used in the study  are  presented  in Table 4. The cost of producing a 15 kg feed was highest for Diet 5 - Le 198,883.4 ($16.72) and lowest for Diet 1 - Le 128,168.20 ($10.78). The cost of producing a 15 kg bag of Ranaan and Aller Aqua feed was Le 204,300.00 ($17.18) and Le 297, 226.59 ($23.47), respectively. It is evident that it is expensive to produce extruded feeds than local feeds, although the cost of producing Diet 5 - Le 198,883.4 ($16.72) used in this study is not significantly different from the cost of producing a 15 kg Ranaan feed.  Aller Aqua which is considered as the best fish feed by farmers in Ghana has a low FCR (feed conversion ratio) but more expensive than other feeds in the market. 

It is evident from the preliminary results obtained in this study, that it is possible to formulate least-cost and nutrient rich feeds for the African catfish (Clarias gariepinus) using locally sourced feed ingredients. Availability of least-cost and nutrient rich fish feed is one of the major problems limiting aquaculture development in Sierra Leone. Using local agricultural by-products for fish feed development will not only reduce the cost of production, but would help to convert what is termed wastes to money, and thus improving the earnings of farmers who produces them. The use of local materials will also reduce capital flight on importation of commercially extruded feeds which often is beyond the reach of local farmers.  The diets formulated in this study compared well with the imported Ranaan feed and Aller Aqua feed with respect to their respective proximate contents and costs of production. The researchers however have noted areas for improvement, and this will be worked on so that the diets can meet the recommended requirements of warmwater fishes as put forward by National Research Council (NRC, 2011).  The final diets to be produced after the adjustment of the ingredients will be tried in the laboratory to determine its suitability for good growth performance of the African catfish.  The diet level that furnishes good growth and FCR will be recommended for commercialization in Sierra Leone. The knowledge of feed development exhibited in the department will not only help to promote local aquaculture development but will contribute significantly to the development of aquaculture in Africa, particularly in the Mano River basins country. 

The authors have not declared any conflict of interests.

The authors would like to thank Dr. Alfred Seawright, Chief Executive Officer of Seawright Mining Company who provided the funding for the research. We are also thankful to the analytical laboratory of the University of Cape Coast (UCC) for helping with the chemical analysis.