THE USE OF CASSAVA LEAVES AS A DIETARY COMPONENT FOR AFRICAN CATFISH

An experiment was carried out to evaluate the effect of graded levels of cassava leaf meal (CLM) with varying cyanide content on growth performance of African catfish (Clarias gariepinus) fry. There were four groups of fish, each comprising 12 fish per group. One group was fed on a control diet (0 mg HCN/kg diet) with fish meal (FM) and soybean meal (SBM) as protein source (30% CP). Groups two, three and four were fed graded levels of CLM to replace 10, 20, and 30% of the total diet at the expense of SBM which contains 4.19, 7.47 and 11.96 mg HCN/ kg diet respectively. After 12 weeks, fish fed with increasing level of CLM showed a significant growth depression and poor feed utilization compared to those fed the control diet. Apparent dry matter digestibility (ADMD) and apparent protein digestibility (APD) were also significantly (P<0.05) affected by cassava leaf meal inclusion, with the best values were found in fish fed the control diet (52.26+0.24% and 83.11+1.17% for ADMD and APD respectively). Body composition of fish fed higher CLM tended to have less protein and fat but more ash. The poor methionine availability as a result of low protein digestibility and the presence cyanide content limits the use of CLM in African catfish fry diet.


INTRODUCTION
One approach that can be used to reduce the cost of fish feed is the use farm waste as components of the feed ingredient.In tropical countries, large quantities of farm waste are produced annually because of a lack of knowledge concerning their nutritive value.Old cassava leaves generally go to waste after harvesting of cassava root and young cassava leaves for human consumptions.Cassava leaves are rich in protein, minerals and vitamin (Muller et al., 1974as quoted by Eruvbetine et al., 2003).Cassava leaves have been reported to have a good amount of indispensable amino acids such as lysine, phenylalanine, valine, tryptophan and arginine, however it's deficient in the sulfur containing amino acids methionine and cysteine (Balagopalan, 1988).Cassava leaves are known to contain cyanoglucosides with 92-98% of the cyanogenic material being in the form of linamarin (Izonkun-Etiobhio et al., 1987).This antinutritional factor can be solved by the proper processing techniques such as oven drying (Gomez and Valdivieso, 1985).Generally, over 70% of linamarin in fresh cassava is removed by enzymatic hydrolysis during processing and nearly all the cyanide which is generated is removed by volatilization (Mkpong, 1990).However, not all of the cyanoglucosides are eliminated when cassava is processed into feedstuff and there are possibilities of some residual cyanogenic glucosides remaining (Umoh et al., 1983) and the remaining non hydrolyzed linamarin present in processed cassava can present health problem because linamarin can be hydrolyzed in the body to release toxic cyanide substances (Balagopalan et al., 1988).Consumption of cassava is also known to reduce the plasma levels of sulfur amino acids in animal since these amino acids were required in cyanide detoxification process.
More studies are needed to ascertain the nutritive and antinutritive value of cassava leaf meal (CLM) as protein source for African catfish.Therefore, the objective of this experiment was to determine the effect of graded levels of cassava leaf meal (CLM) with varying cyanide content in fish meal (FM) and soybean meal (SBM) based diets at the expenses of SBM, on growth, feed utilization, and body composition of African catfish.

MATERIALS AND METHODS
Four experimental diets were prepared to contain equal levels of animal and plant protein at a 1 : 1 ratio (Table 1).The control diet contained only FM and SBM as the protein source.Diets 2, 3, and 4 contained 10, 20, and 30% CLM respectively at the expense of SBM.All diets were isonitrogenous (30% protein) and isoenergetic (432 kcal/100 g diets).Caloric levels were maintained by adjusting the levels of α-starch while α-cellulose was added as filler (Table 1).The various ingredients were thoroughly mixed in a feed mixer and added chromium oxide as an inert marker for protein digestibility determination.
This was then passed through a meat mincer and resulting a strand like string.After drying in an electric oven at 60 0 C the diets broken up and sieved into convenient pellets size and stored at -20 0 C prior to use.
Twelve African catfish with average weight 1.45 g was randomly stocked into aquaria (60 × 30 × 30) cm filled with 40 l of water.Three replicate were used for each experimental diet.Aeration was provided and water changed manually, twice a day.Fish were fed their respective diet twice daily at rate 7% body weight.After 12 weeks feeding trial, all fish were starved for 48 hours.Three of the fish were randomly selected from each aquarium, killed and individually weighed.Their muscles were kept in a freezer at -20 0 C until further analysis.Some of the fish were retained for faecal collection.The fish were fed to satiation then transferred to 19 l plastic container with a funnel shaped bottom connected to 3 cm plastic tube.The faeces were dried in oven and put in sealed polyethylene bags and placed in desiccators for chemical analysis.
Chemical analysis of ingredients, experimental diets, muscle and faeces were run in triplicates.Fish muscle was freeze dried prior to analysis.Cyanide analysis of fresh cassava leaves, dry cassava leaves and experimental diets were determined according to methods adopted from Nambisan and Sundaresan (1984).The feed ingredients, experimental diets and fish carcass were analyzed to their proximate composition following standard AOAC methods (Anonymous, 1997), while digestibility was analyzed using Furukawa and Tsukahara method's (1966).
Data were subjected to One Way ANOVA (SPSS Ver 11) to compare the differences among treatment and followed by a Duncan's multiple range test (DMRT).

RESULTS AND DISCUSSION
Whole cassava leaf meal contains 24.61, 6.65, 24.69, and 7.3% for crude protein, lipid, fiber and ash respectively.The protein content of cassava leaves used in this study was within the range of reported previously around 20 to 35% of dry weight (Montaldo, 1977).The cyanide content of fresh whole cassava leaves was 443.78 + 3.4 mg/kg and decreased to 70.55 + 5.49 mg HCN/kg after oven drying.This value was then reflected in the experimental diets in which it was noted that higher levels of CLM inclusion in the diet resulted in higher cyanide levels in the diets.Further processing of experimental diets resulted in a decrease of cyanide content in the experimental diets as can be seen in Table 1.The low cyanogenic glucoside content of the analyzed cassava as a result from drying the sample was expected and has been previously reported (Best, 1979;Gomez and Valdivieso, 1985;and Saka, 1993).The dried cassava leaves used in the present study contained cyanogenic glucoside levels that were lower than the permissible maximum level of 100 mg/kg hydrocyanic acid set by the Council of the European Community for cassava products to be used as animal feed (Ingram, 1975 as cited by Gomez and Valdivieso, 1985).Ravindran et al. (1987) found that the cyanide levels of 53 mg/kg dry matter were tolerated without adverse effects on growth performance of poultry and pigs when cassava leaf meal is incorporated at levels not more than 15 to 25% respectively.
Overall, this experiment revealed that increasing level of cassava leaf meal (CLM) in the diet for African catfish resulted in a linear depression of growth performance and also feed efficiency (Table 2).Ng and Wee (1989) reported a similar result for Nile tilapia (Oreochromis niloticus) fed with diets containing increasing levels of CLM.They found that a diet containing CLM which replaced fish protein as much as 20% was enough to depress growth performance and feed food utilization of Nile tilapia.In a related study, Pacu (Piaractus mesopotamicus) which were fed with diets containing increasing levels of cassava foliage meal in pond also showed a progressive reduction growth performance on the basis of final weight gain and specific growth rate (Padua et al., 1998).However Bureau et al. (1995) who also conducted experiment in ponds found that African catfish performance which was fed 0, 10 and 20% CLM (g/100 g diets) did not show any significant difference in biomass and final weight gain.
The difference results observed in the above studies were probably due to the difference in replacement of fish meal (FM) in Padua's experiment and Ng and Wee's experiment and replacement of partial soybean meal (SBM) in Bureau's experiment.Consequently, protein from cassava used in the latter studies was lower.In an earlier study which evaluated the nutritive values of CLM in diets for Nile tilapia (Ng and Wee, 1989), CLM was used to replace protein from FM at levels of 0, 10, 20, 40, 60, and 100%.However in the present study, lower proportions of CLM were used and to replace protein from SBM in the diet containing animal and plant protein at ratio 50 : 50 respectively.CLM inclusion levels of 30% contributed up to 24.6% of dietary protein in and these levels were similar to those used in Bureau's experiment.
The poor growth response and feed utilization values of the fish as levels of CLM in diets increase in first experiment It is possible that the relatively higher level of crude fiber in diets containing CLM inclusion which contain 24.69% compared with 7.48% in SBM.High percentage of indigestible components such as crude fiber in diet have been attributed as the possible cause of poor digestibility and lowering availability of the nutrient content of plant.The APD and ADMD data supported this assumption in which the control diet had significantly higher APD and ADMD compared to others diets.Ng and Wee (1989) also attributed the reduction of growth performance of Nile tilapia fed increasing level of CLM to low digestibility.They found that crude fiber of CLM diet was increased from 6% to 12% with increasing 20% to 60% CLM as a percentage of dietary protein replacement in the diets.Eruvbetine et al. (2003) also found similar results when they studied the effect of cassava concentrate on broiler chickens performance, and concluded that the presence of 13.5% crude fiber in the leaves was attributed to the declined body weight in broiler chickens.
Another possible reason for the depression in growth and feed utilization of fish fed CLM is could be explained by the use of sulfur containing amino acid for detoxification of cyanides.Cyanide, even found in low amounts in this study, occurred in the CLM diets.Cyanide plays its toxic role by using up body sulfur in detoxification process; this therefore increases the body demand for sulfur containing amino acids.Consequently its causes a decrease in methionine that might be already marginal in diet based on CLM (Haque and Bradbury, 1999) and will reduce growth response in African catfish in this study.The possible effect of this inhibitory factor was more pronounced in the 30% CLM dietary substitution.It was found that cyanide containing diets was responsible for weight losses observed in Giant rat (Tewe, 1984) and in goat  a, b, c, d Mean values in the same row with different superscripts show significant difference (P<0.05).
The result of fish carcass composition is presented in Table 3.The results of this study showed that muscle protein was observed to decrease as CLM increased as well as lipid content in the first experiment, however, ash content were increase with increasing of CLM in the diets.This finding is in agreement to previous findings (Ng and Wee, 1989), whereby increase in dietary level of CLM resulted in a numerical decrease in lipid and protein content.Fasakin and Balogun (1998) also found similar decrease of protein and fat content as well as increasing of ash content in African catfish with increasing level of dried water fern (Azolla pinnata) to replace SBM component in the diets.

CONCLUSION
It appears that African catfish with initial stocking size of 1.45 g could not be able to fully utilize cassava leaves meal based diet.The poor methionine availibity as a result of low protein digestibility and the presence cyanide content limits the use of CLM in African catfish fry diet.

Table 1 .
Composition and proximate analysis of experimental diets containing varying levels of cassava leaf meal (dry weight basis)

Table 2 .
Growth performances and feed utilization of African catfish (Clarias gariepinus) fed graded levels of cassava leaf meal

Table 3 .
Proximate composition of African catfish (Clarias gariepinus) muscle fed different cassava leaf meal (CLM) diet (%) on wet weight basis Note: a, b, c, d Values are the mean of triplicate groups.Mean values in the same row with different superscripts show significant difference (P<0.05).