Nutritive value of potential feed resources used in Laos for African catfish (Clarias gariepinus) production

O Phonekhampheng, L T Hung* and J E Lindberg** 

Department of Livestock and Fisheries, Faculty of Agriculture, National University of Lao, Vientiane, Lao PDR.

*Faculty of Fisheries, Nong Lam University, Ho Chi Minh City, Vietnam.

**Department of Animal Nutrition and Management, Swedish University of Agricultural Sciences, P.O. Box 7024, SE-750 07 Uppsala, Sweden
jan-eric.lindberg@huv.slu.se 

Abstract 

The present experiment was performed to determine the digestibility of nutrients and energy in carbohydrate-rich (rice bran, broken rice, maize, cassava root meal) and in protein-rich (dried fish, snails, earthworms, frogs, termites) feed resources that are commonly used to formulate diets for the African catfish (Clarias gariepinus) in Laos.

The apparent digestibility (AD) of organic matter ranged from 78.1 to 87.5 %, and was on average 83.2 % (SD 2.7). The AD of crude protein (CP) in the test diets ranged from 86.6 to 93.2 % (average 90.6 %, SD 2.4) and the AD of crude fat (EE) ranged from 79.9 to 91.2 % (average 87.0, SD 3.4). The AD of the carbohydrate components was lower than the AD for CP and EE. However, despite a larger variation in the AD between diets and ingredients, on average 72.2 % (SD 6.7) of the nitrogen-fee extracts and 77.7 % of the total carbohydrate fraction (SD 3.7) were digested. The AD of energy in the test diets was high and ranged from 80.3 to 88.2 % (average 84.5 %, SD 2.3).

The estimated content of digestible energy (DE) ranged from 13.2 MJ/kg DM for rice bran to 18.1 MJ/kg DM for frogs. On average, the DE content of the carbohydrate-rich feed ingredients was lower than that of the protein-rich feed ingredients. The protein: energy ratio (P/E ratios) ranged from 2.2 to 6.4 g CP/MJ DE for the carbohydrate-rich feed ingredients, and from 17.4 to 26.1 g CP/MJ DE for the protein-rich feed ingredients.

To reach optimum P/E ratios in the diet for growing African catfish (25-30 g CP/MJ DE) only the protein-rich feed ingredients has the potential to meet required levels. Thus, there is a need to search for other potentially useful feed ingredients for catfish production.

Key words: catfish, digestibility, feed resources, nutrition, nutritive value

Introduction

The majority of fish producers in Laos are smallholders, most of whom are located in rural areas, and with a limited knowledge-base on fish production. The African catfish (Clarias gariepinus) has been identified as one of the fish species with the greatest potential to contribute to fish production in Laos, in particular because the technologies available can be transferred to function profitably even in a rural context (DLF 1999, MOAF 2004). However, the future development of small-scale aquaculture systems depends on available feed resources, as feeding constitutes a significant portion of the operation cost (Edwards 1997). An increased use of local resources will reduce the feeding costs (Edwards and Allan 2004).

In a recent survey, a total of thirteen feedstuffs, including both conventional and unconventional, were identified as being frequently used for feeding of the African catfish (Clarias gariepinus) in Laos (Phonekhampheng et al 2008). The use of feedstuffs varied between different agro-ecological areas and between farms of different size within each area, and there were large differences in the overall diet content of major nutrients. However, in order to formulate nutrient balanced diets for African catfish production in Laos there is a need to have access to digestibility data for the more commonly used feed ingredients.

The present experiment was performed to get some baseline data on the digestibility of nutrients and energy in carbohydrate-rich (rice bran, broken rice, maize, cassava root meal) and in protein-rich (dried fish, snails, earthworms, frogs, termites) feed resources that are commonly used to formulate diets for the African catfish (Clarias gariepinus) in Laos.

Materials and methods  

Location and climate

The experiment was carried out in facilities of the Faculty of Fisheries at the Nong Lam University, Ho Chi Minh City, Vietnam. The feedstuffs studied were collected and prepared in Laos, and transported to Vietnam for the digestibility study.

The mean ambient temperature was 34 °C in the middle of the day during the trial, which started in March 2007.

Diets and experimental design

A basal diet (B), composed of fish meal, soybean meal, rice bran, cassava root meal, fish oil, soybean oil, vitamins and minerals, a binder (carboxy-methyl cellulose) and Cr₂O₃ as an indigestible marker was formulated (Table 1). The feedstuffs used in the basal diet were purchased from the local market.

Four test diets were formulated to determine the digestibility of dietary components in the following selected carbohydrate-rich feedstuffs; rice bran, broken rice, maize and cassava root meal. A batch of each feedstuff was purchased from local farmers in Laos.

Five test diets were formulated to determine the digestibility of dietary components in the following selected protein-rich feedstuffs; raw Golden Apple Snail (Pomacea spp) (GAS), boiled GAS, dried fish, earth worms, frogs and termites (Coptotermes curvignathus). GAS, dried fish and frogs were purchased from local markets and earthworms and termites from local farmers in Laos.

The test diets were composed of the feedstuff of interest plus Cr₂O₃, making up 30% of the dietary dry matter (DM), and diet B making up 70% of the DM.

The dietary treatments were randomized to three replicate fish tanks per treatment. Faeces were pooled within each treatment to obtain sufficient material for analysis. Thus, no statistical analysis was possible to perform on the digestibility data presented.

Preparation of test feedstuffs

To prepare the raw GAS, the shell was broken and removed, and the remaining cover and flesh cleaned with water before sun-drying. To prepare the boiled GAS, snails were immersed in boiling water for 3 minutes, before shell removal and further preparation as described above. Both raw and boiled GAS was sun-dried for 2 days before milling.

The termite samples comprised mainly termites but also some of the nest. The fragments of nests were broken up into small pieces before drying and milling. Dried fish, frogs and earthworms were cleaned with water before sun-drying and milling.

After preparation, all tested feedstuffs were dried, milled through a 1 mm sieve and stored at room temperature until use.

The test diets were prepared by mixing the ingredients in the basal diet with each of the test feedstuffs, as described above, after which the mixture was cold pelleted using a 5 mm dye and then dried at 60°C for 24 hours before use.

Digestibility studies

Clarias gariepinus juveniles with an average body weight of 50±15 g were acclimatised to the experimental conditions and cultured in concrete tanks to be familiar with artificial feed for at least 10 days before faeces collecting started. The digestibility studies were performed in 120 l cylindrical tanks with stagnant water. Each tank was stocked with 16-20 fingerlings of mixed sex.  The digestibility study lasted for 14 days, comprising an adaptation period of 3-4 days followed by total faeces collection. Water quality parameters (temperature, pH, dissolved oxygen) were controlled during the experiment.

The feeding level was approximately 5% of fish body weight and day. The daily feed allowance was divided in two equal meals, which were fed at 8 a.m. and 2 p.m.. Diets were placed in the water for 1 h, after which the tanks were cleaned from residual feed and feaces. The fish were not fed until the next morning and the feaces produced during that time was settled at the bottom of the conical tanks. The faeces were collected daily at 07.00 a.m. and at 01.00 p.m. from each tank and were stored at -20°C in a freezer. At the end of the collection period faecal matter were dried in an oven at 60°C ground and preserved in airtight containers until analysis.

Chemical analyses

The samples were analyzed for dry matter (DM), ash, crude protein (CP), crude fat (EE) and crude fiber (CF) according to standard methods of AOAC (1990). Chromic oxide content in diets and faeces was analyzed according to Furukawa and Tsukahara (1996).

Calculations

The apparent digestibility (AD) of the diets was calculated using the indicator technique according the equation:

AD_D=1-(DC_F × I_D)/(DC_D × I_F)

AD_D is the apparent digestibility of the dietary component;
DC_F is the dietary component concentration in faeces, while
I_D is the indicator concentration in diet,
DC_D is the dietary component concentration in diet and
I_F the indicator concentration in faeces.

The digestibility of tested ingredient (AD_T) was calculated from the AD_D of each dietary component in the basal diet and in the test diet (Bureau and Hua 2006), as follows:

AD_T = AD_D test diet + [(AD_D test diet – AD_D basal diet) x (0.7 x DC_D basal diet/0.3 x DC_D ingredient)]

The content (g/kg) of nitrogen-free extracts (NFE) in DM was calculated as;

1000-(ash + CP + EE + CF).

The content (g/kg) of total carbohydrates (CHO_T) in DM was calculated as;

1000-(ash + CP + EE).

The content (kJ/kg DM) of gross energy (GE) in feeds and faeces was calculated from the content of CP, EE and ash as described by Ewan (1989).

Results and discussion 

Chemical composition of ingredients

The chemical composition of the more common carbohydrate-rich feed ingredients studied was comparable to other published data (Table 2) (NRC 1993; NIAH 1995).

In the more unconventional protein-rich feed ingredients, the chemical composition of raw GAS and frogs were similar to those reported by others (Keansombath 2003; Somsueb and Boonyaratpalin 2001; Olvera-Novoa et al 2007). In contrast, the chemical composition of dried fish, earthworms and termites used in the present study differed from other published data (NIAH 1995; Oyarzun et al 1996; Phounvisoul Latsamy and Preston 2008).

The CP content in dried fish was markedly lower and the content of ash markedly higher than in conventional fish meal (NRC 1993; NIAH 1995). This could be due to the use of small-sized trash fish, which have a high proportion of bone to meat in their body. Similarly, the CP content of earthworms was markedly lower than reported by Phounvisoul Latsamy and Preston (2008). In addition, the ash content was high. This indicates soil contamination in the samples studied. Moreover, the CP content of termites was markedly lower and the content of crude fiber was markedly higher than reported for samples of termites by Oyarzun et al (1996). This was due to the partial inclusion of the nests in the samples studied and reflects the common way of small-holder rural farmers in Laos to collect and feed termites to catfish.    

Digestibility of test diets and ingredients

Considering that the African catfish (Clarias gariepinus) has been described as an omnivorous scavenger (Clay 1979) it should be expected to have the potential to efficiently utilize a wide range of feed ingredients. This should apply to feed of both plant and animal origin. In general, this contention was supported by the high digestibility values of organic matter (OM) found for most dietary components in the present study for both the basal diet (Table 3) and the test diets (Table 4).

The apparent digestibility (AD) of OM in the test diets ranged from 78.1 to 87.5 %, and was on average 83.2 % (SD 2.7). Degani and Revach (1991) showed that the digestive capability of Clarias gariepinus (Burchell 1822) was comparable to that of tilapia (Oreochromis aureus x O. niloticus) and carp (Cyprinus carpio L.).

The AD of CP in the test diets ranged from 86.6 to 93.2 %, and was on average 90.6 % (SD 2.4). The latter was only slightly lower than the AD of CP in the basal diet (Table 3) and was also reflected in high AD of CP in most feed ingredients (Table 4). This was in agreement with other published data on catfish (Fagbenro 1996; 1998). However, on average the carbohydrate-rich ingredients showed lower AD-values than the protein-rich ingredients (54.8 vs. 82.7 %). This could partly be due to low CP content in the carbohydrate-rich feeds as compared with the protein-rich feeds. Moreover, this indicates that animal protein may be better utilized than plant protein by the African catfish (Clarias gariepinus). Similar data has been reported by Fagbenro (1996; 1998).  

The AD of EE in the test diets ranged from 79.9 to 91.2 %, and was on average 87.0 (SD 3.4). Although the range for the AD of EE was higher than for CP (Table 3), the data indicates a high utilization of EE in most of the feed ingredients studied (Table 4). Due to unexpectedly low AD of EE in raw and boiled Golden Apple snails (GAS), the average AD of EE in the protein-rich feeds was lower than in the carbohydrate-rich feeds.    

The AD of the carbohydrate components (NFE and CHO_T) in the diets and ingredients was lower than the AD for CP and EE (Tables 3 and 4). However, despite a larger variation in the AD between diets and ingredients, on average 72.2 % (SD 6.7) of the NFE and 77.7 % of the CHO_T (SD 3.7) were digested. This is in line with the presumed omnivorous nature of the African catfish (Clay 1979; Degani and Revach 1991) and indicates a capacity to utilize carbohydrates to some extent in their diet. This is further supported by a recent study by Leenhouwers et al (2007), who reported starch digestibilites in the range 80 to 97 % in cereal-based diets fed to African catfish. As expected, the digestibility of non-starch polysaccharides (NSP) was lower and more variable. Interestingly, up to 55 % of the NSP fraction was digested in a maize/rye-based diet.

The AD of energy in the test diets was high and ranged from 80.3 to 88.2 %, and was on average 84.5 % (SD 2.3) (Table 3). This was in agreement with other published data on catfish (Fagbenro 1996; 1998). The AD of energy for the feed ingredients showed a wider range than that of test diets (68.1-86.4 %; SD 6.9) (Table 4). On average, the AD of energy for the carbohydrate-rich feed ingredients was lower (73.8 %) and varied more (SD 6.5) between ingredients than the values for the protein-rich feed ingredients (83.6 %; SD 3.8). Similar findings were reported by Fagbenro (1996; 1998).

Nutrient supply from potential feed resources

Studies on the African catfish (Clarias gariepinus) have shown that the crude protein level in the feed should be 300-350 g per kg DM to give a high growth rate (FAO 1983). The crude protein requirements have estimated to be around 400 g per kg DM (Machiels and Henken 1985; Degani et al 1989). Given the CP content of available feed resources in this study, diets for catfish have to be based on protein-rich feeds only to provide the protein needed to support a high growth rate. The only exception is if the CP supply in the diet was based on frogs. In this case, around 30 % of the dietary DM could come from low-protein feed ingredients.

The estimated content of digestible energy (DE) ranged from 13.2 MJ/kg DM for rice bran to 18.1 MJ/kg DM for frogs (Table 5).

On average, the DE content of the carbohydrate-rich feed ingredients was lower than that of the protein-rich feed ingredients. This was mainly due to the lower AD for energy, as the content of gross energy differed less between feed ingredients.

The protein: energy ratio (P/E ratios) ranged from 2.2 to 6.4 g CP/MJ DE for the carbohydrate-rich feed ingredients, and from 17.4 to 26.1 g CP/MJ DE for the protein-rich feed ingredients (Table 5). The optimum P/E ratio for growing Clarias gariepinus has been reported to be in the range 25 to 30 g CP/MJ DE (Machiels and Henken 1985; Uys 1989; Ali 2001). Thus, to reach this level in the diet only the protein-rich feed ingredients studied have the potential to provide sufficient P/E ratios in catfish diets.

Conclusions 

The African catfish have digestive capacity to utilize most of the studied potential feed resources reasonably well. Based on chemical composition and the digestibility data presented, it should be possible to formulate diets that better meet the nutrient needs of cultured African catfish in Laos. However, mainly the protein-rich feed ingredients are useful if the diet should reach a sufficient CP: DE ratio. Thus, there is a need to search for other protein-rich feed ingredients that also could be used in diet formulation for the African catfish.

Acknowledgments 

The authors are grateful to the Swedish International Development Authority (Sida/SAREC) for funding this study and the faculty of Agriculture, National University of Lao for allowing the first author to carry out this study. Thanks are also due to the Feed Analysis Laboratory of the Department of Livestock and Fisheries of the Ministry of Agriculture and Forestry.

References  

Ali Md Z 2001 Dietary protein and energy interactions in African catfish Clarias gariepinus (Burchell, 1822). PhD-thesis, University of Stirling, Scotland, UK. 273 pp.

AOAC 1990 Official Methods of Analysis 15th edition. Association of Official Analytical Chemists. Washington, DC. Volume 1:69-90.

Bureau D P and Hua K 2006 Letter to the Editor of Aquaculture. Aquaculture 252: 103-105

Clay D 1979 Population biology, growth and feeding of African catfish (Clarias gariepinus) with special reference to juveniles and their importance in fish culture. Archives of Hydrobiology 87: 453-482.

Degani G, Ben-Zvi Y and Levanon D 1989 The effect of different dietary protein levels and temperatures on fed utilization, growth and body composition of Clarias gariepinus (Burchell 1822). Aquaculture 76: 293-301

Degani G and Revach A 1991 Digestive capabilities of three commensal fish species: carp, Cyprinus carpio L., tilapia, Oreochromis aureus x Oreochromis niloticus, and African catfish, Clarias gariepinus (Burchell 1822). Aquaculture and Fisheries Management 22: 397-403

DLF 1999 Fisheries resources and fisheries development policy framework in the Lao PDR. Department of Livestock and Fisheries, Ministry of Agriculture.Vientiane. 23 pp.

Edwards P 1997 Sustainable food production through aquaculture. Aquaculture Asia 2(1): 4-7

Edwards P and Allan G L 2004 Feed and feeding for inland Aquaculture in Mekong Region countries. ACIAR Technical Reports 56:1-37

Ewan R C 1989 Predicting the energy utilization of diets and feed ingredients by pigs. In: Energy metabolism (editors: Y van der Honing and W H Close). European Association of Animal Production, Bulletin No. 43: 215-218. Pudoc, Wageningen, the Netherlands.

Fagbenro O A 1996 Apparent digestibility of crude protein and gross energy in some plant and animal-based feedstuffs by Clarias isheriensis (Siluriformes: Clariidae) (Sydenham 1980). Journal of Applied Ichthyology 12: 67-68

Fagbenro O A 1998 Apparent digestibility of various oilseed cakes/meals in African catfish diets. Aquaculture International 6: 317-322

FAO 1983 Fish feeds and feeding in developing countries. An interim report on the Aquaculture Development and Coordination Programme (ADCP) Feed Development programme. ADCP/REP/83/18. FAO, Rome, Italy. 97 pp.

Furukawa A and Tsukahara H 1996 On the acid digestion method for the determination of chromic oxide as the index substance in the study of fish feed. Bulletin of the Japanese Society of Scientific Fisheries 32: 502–506

Keansombath L 2003 Evaluation of nutritive value of ensiled and fresh Golden Apple Snail (Pomacea spp) for the growing pigs. Department of Animal Nutrition and Management, Swedish University of Agricultural Sciences, Uppsala, Sweden. M. Sc. Thesis ISBN 91-576-6838-8.

Leenhouwers J I, ter Veld M, Verreth J A J and Schrama J W 2007 Digesta characteristics and performance of African catfish (Clarias gariepinus) fed cereal grains that differ in viscosity. Aquaculture 264: 330-341

Machiels M A M and Henken A M 1985 Growth rate, feed utilization and energy metabolism of the African catfish, Clarias gariepinus (Burchell, 1822), as affected by dietary protein and energy content. Aquaculture 44: 271-284

MOAF 2004 State Planning Committee. Poverty in the Lao PDR. Participatory poverty assessment. Ministry of Agriculture and Forestry, Lao PDR, Vientiane, Laos. 8-24.

NIAH 1995 Composition and Nutritive value of Animal feed in Vietnam. Ministry of Agriculture and food industry. 249 pp.

NRC 1993 Nutrient Requirements of Fish. National Research Council. National Academy Press, Washington, D.C.

Olvera-Novoa M A, Ontiveros-Escuita V M and Flores-Nava A 2007 Optimum protein levels for growth in juvenile bullfrog (Rana catesbeiana Shaw, 1802). Aquaculture 266: 191-199

Oyarzun S E, Crawshaw G J and Valdes E V 1996 Nutrition of the Tamandua: I. Nutrient Composition of Termites (Nasutitermes spp.) and Stomach Contents from Wild Tamanduas (Tamandua tetradactyla). Zoo Biology 15: 509-524

Phonekhampheng O, Hung L T and Lindberg J E 2008 A survey on African catfish (Clarias gariepinus) feeding and production in Laos. Asian
Fisheries Science 21, 21-33.

Phounvisoul Latsamy and Preston T R 2008: Fly larvae, earthworms and duckweed as feeds for frogs in an integrated farming system. Livestock Research for Rural Development. Volume 20, supplement. http://www.lrrd.org/lrrd20/supplement/lats2.htm

Somsueb P and Boonyaratpalin M 2001 Optimum protein and energy levels for the Thai native frog, Rana rugulosa Weigmann. Aquaculture Research 32: 33-38

Uys W 1989 Aspects of the nutritional physiology and dietary requirements of juvenile and adult sharptooth catfish, Clarias gariepinus (Pisces: Clariidae). PhD Thesis, Rhodes Univerisity, Grahamstown, South Africa.

Received 21 February 2008; Accepted 21 April 2008; Published 5 December 2008

Go to top