A note on the nutritive value of roasted velvet bean { Mucuna pruriens ) in broiler diets

The nutritive value of roasted Mucuna pruriens var. utilis (mucuna or velvet bean) was investigated in a four-week study using 336 mixed-sex three-week old broiler chickens. The birds were randomly allocated to four dietary treatments. Fifty, 100 or 150 g of mucuna beans, roasted in metal bowls over a wood-fired stove until they turned light brown was included into nutritionally balanced diets. Control diet did not contain mucuna beans. Feed and water were supplied ad libitum. Performance criteria studied included feed intake, growth rate, feed conversion ratio, carcass parameters and mortality. Roasted mucuna beans at the levels of 50 and 100 g kg" significantly (P<0.05) improved the growth and feed utilization of broilers. At 150 g kg" diet, however, mucuna beans depressed weight gains and feed utilization.


INTRODUCTION
Mucuna puriens var.utilis (mucuna or velvet bean; MP) has been grown from time immemorial in Ghana and other countries as a component of the traditional crop farming system.Among various Ghanaian tribes, the beans find use in the preparation of soups and stews (Ahenkora et al., 1994, unpublished).In recent years mucuna has gained popular acceptance among farmers as a cover crop helping to control weeds and also improve soil fertility by fixing nitrogen (Osei-Bonsu and Buckles, 1993;Buckles, 1995).The large-scale cultivation of mucuna has meant that large quantities of the beans are left un-utilized as human food.The dry matter of velvet beans MP contains 250 g kg" 1 protein (Ahenkora et al, 1999) and this makes MP attractive as a potential feed ingredient for poultry that may serve as a replacement (at least partially) for more expensive protein ingredients like fish meal and soyabean meal.There is little published literature in this regard although work using raw MP showed deleterious effects on chick and broiler performance (Vieira and Carvalho, 1996;Osei and Dei, 1998, unpublished).The negative effects of MP have been variously attributed to the presence of toxic antinutritional factors including anti-trypsin factors, tannins, phytic acid, L-DOPA, anticoagulants and others (Liener and Kakade, 1969;Sathe and Salunke, 1984;Ravindran and Ravindran, 1988;Osei-Bonsu and Buckles, 1993;Houghton and Skari, 1994).Several heat treatments have been employed to destroy anti-nutrients in various leguminous seeds (Huisman and Tolman, 1992).
This work was undertaken to investigate the beneficial effects of roasting on the nutritive value of Mucuna pruriens for broiler chickens.

MATERIAL AND METHODS
Dried mature mucuna beans (MP) were obtained either from local farmers growing the crop under a Sasakawa Global 2000-sponsored pilot project or from test plots of the Crops Research Institute, Kumasi (Ghana).The beans were roasted in quantities of approximately 1 kg for approximately 20 min in metal bowls over wood-fired stoves until they turned light brown.They were then milled through a 3 mm sieve in a hammermill.The milled beans were analysed for proximate constituents using conventional methods (AOAC, 1980).
Three hundred and thirty six mixed-sex Shaver Tropicbro commercial broiler chickens (Pomadze Poultry Enterprises, Winneba, Ghana) aged 21 days old were initially weighed in groups to obtain the mean weight for the flock.They were then individually weighed and re-allotted in equal numbers to four experimental dietary groups such that the groups were equalized for sex and weight.The control diet was formulated as a combined starter-finisher diet containing approximately 210 g kg" 1 crude protein; the three other test diets contained 50, 100 and 150 g mucuna per kg, respectively, replacing equal amounts of the control diet (Table 1).The diets were isocaloric and isonitrogenous.
The trial was conducted according to the completely randomised design and each treatment had three replicates.The experimental broilers were reared in raised wire floor coops with floor space averaging approximately 0.14 m 2 .Feed and water were supplied ad libitum.The trial lasted 28 days.
The following parameters were measured: feed intake, liveweight gain, feed conversion ratio and mortality.On the final day of the trial, three broilers were selected randomly from each treatment replicate and slaughtered for carcass analysis.The analysis of variance was used in data analysis while significant differences among treatment means were estimated using Fisher's least significant difference test (Steel and Torrie, 1980).

RESULTS AND DISCUSSION
Proximate analysis showed that roasted velvet beans contained 92.8 g kg" 1 dry matter which in turn contained per kg/g: crude protein, 246; crude fibre, 65.4; ether extract, 52.0; ash 32.0; and N-free extractives, 604.6.These values are in close agreement with those presented by Osei and Dei (1998, unpublished) and Dei Carmen et al. (1999) and indicate that heating may generally slightly increase nutrient values as a result of the evaporation of water (Ahenkora et al., 1999).The crude protein content of MP compares favourably with that of chickpea, mung bean and several cowpea varieties (Hsu et al., 1977;Abbey and Ibeh, 1988).
The effects of roasted velvet beans on broiler performance are summarized in Table 2. Roasted MP at 50 or 100 g kg" 1 diet significantly improved weight gains (2-4%) and feed conversion ratio (1.2-3.4%) over the control diet.At the 150 g kg" 1 diet level, however, MP significantly depressed weight gain (3.3%) and efficiency of feed utilization (6.7%) compared with the control.The depression in weight gain was evident by the end of the second week while impaired feed utilization efficiency was observed at the end of the third week.Del Carmen et al. (1999) have reported that heated MP when fed at 10% of the diet had deleterious effects only early weight gain while 20 and 30% heated MP progressively depressed weight gain and efficiency of feed utilization in a 42-day feeding trial.In an earlier trial, Osei and Dei (1998, unpublished) all levels of raw MP from 5 to 15% of the diet caused significant deterioration in broiler chicken performance.It is suggested that heating in this trial might have destroyed or inactivated anti-nutritional factors present in unheated MP making it possible to include up to 100 g kg" 1 in broiler diets.Data from Del Carmen et al. (1999) indicates that heat treatment destroyed anti-trypsin activity without affecting L-DOPA.Whether heating affected any other factor is not clear since these were not studied.The depressive effect of the 150 g kg" 1 MP on broiler performance is difficult to explain.It is possible that other thermostable factors were involved.Ahenkora et al. (1999) have stated that the majority of toxic effects reported in human consumption of MP bean preparations are attributable to L-DOPA.There were no significant dietary treatment effects on feed consumption, mortality and carcass dressing percentage.Del Carmen et al. (1999) observed significant decreases in feed consumption only in broiler chickens fed diets containing 20 or 30% MP.The highest level of MP in this trial was only 150 g kg 1 .The lack of significant effect of MP on mortality and carcass weights agrees with other reports (Osei and Dei, 1998, unpublished;Del Carmen et al., 1999).
Table 3 shows the effects of dietary mucuna on various organ weights.No significant effects on the kidney weight were evident from feeding MR On the other hand, there were significant depressions in the weights of gizzard, liver and small intestine at all levels of MP incorporation.MP at 150 g kg" 1 diet exerted the most serious effects.The authors are not aware of work with MP in which organ weights were studied.Most work on antinutritional factors has in fact concentrated on physiological changes (Huisman and Tolman, 1992).
The results of this study show that roasting can improve the nutritive value of MP and permit its incorporation into broiler diets up to 10%.It is however possible that if all the toxic/antinutritional factors were removed it might be possible to include higher levels of MP in broiler diets.A variety of detoxification methods including particurarly boiling and soaking have been employed to make MP beans safe for human consumption.Studies of processing technologies and treatments for improving the nutritive value of MP for chickens are recommended.

TABLE 1
Composition and nutrient content of experimental diets, g kg" 1 DM except ME 2amino acid content of mucuna fromAchinewhu (1982)

TABLE 3
c different at PO.05