Livestock Research for Rural Development 15 (12) 2003

Citation of this paper

Potential of seeds of Napoleona imperialis (p. beauv) as a source of haemolytic saponin and feed ingredients

U H Ukpabi  and U J Ukpabi **

Department of Agriculture, Abia State University, Umuahia Campus, PMB 7010, Umuahia, Nigeria
* Biochemistry Division, National Root Crops Research Institute, Umudike, PMB 7006, Umuahia, Nigeria


Seeds extracted from fruits of Napoleona imperialis in Nigeria were processed into a dry powder and analysed for nutrients, calorific value and antinutritional factors. Extracted saponin from the seed was used for erythrocyte fragility tests (haemolysis) while buffered saline extract of the seed meal was used for toxicity test with 3-week old broiler chicks.

Results of the proximate analysis showed that the dry seed meal had 4.8% moisture, 11.7% crude protein, 4.95% ether extract, 3.60% crude fibre and 3.52% ash. The mineral content of the seed meal included 5.01 g/kg Ca, 17.5 g/kg K and 16.1 g/kg Na. The values for saponin and cyanide contents were 20% and 135 mg/kg, respectively. In vitro experiment with the seed meal showed very high level of haemolysis while the in vivo experiment had no observable lysing.

The investigation therefore shows that the seeds of N. imperialis are rich sources of commercial haemolytic saponin. In addition, saponin extraction and cyanide detoxification would make the seed meal a good raw material source for livestock feed.

Key words: Feed ingredients, heamolytic saponin, Napoleona imperialis


Napoleona imperialis P. Beauv. belongs to the family Lecythidaceae which is a small tropical family that grows in all the regions of Nigeria and other parts of West Africa (Dalziel 1955; Hutchinson and Dalziel, 1958; Keay et al., 1964). The plant is commonly known as ntum in the Ikwuano dialect of Ibo language in Nigeria.

Though, N. imperialis is one of the lesser known plants, its economic importance has partially been reported by Dalziel (1955) and Irvine (1961). These include the use of the fruits sugary pulp as desserts, the roots for medicinal purposes and the twigs as traditional chewsticks. Not much is known of the chemical composition of the seeds of N. imperialis. However, the chemical compositions of the leaf, bark and roots of N. vogelii had been partially determined by Ogbonnaya (1983). His work showed the presence of up to 200mg/kg cyanide potential in the leaves of the plant. Earlier, the possible presence of saponin in the seed of N. imperialis had been reported (Dalziel, 1955; Irvine, 1961).

Saponins are glycosides containing a polycyclic aglycone moiety of either C27 steroid or C30 triterpenoid (collectively known as sapongenins) attached to a carbohydrate. They have a characteristic bitter taste, foaming properties, and can cause injuries to the digestive mucosa and haemolytic changes in blood (Kumar and D'Mello, 1995; Seifert, 1996; Radostits et al, 1997; McDonald et al., 1998).

Little or nothing had been done to know if the saponin from N. imperialis could be employed for laboratory haemolysis in medical (human and veterinary) research. There is need to identify other plant sources of haemolytic saponin in addition to known plant saponins such as digitonin from the temperate digitalis seeds (Karlson, 1968). Furthermore, there is need to know the nutrient composition of the seed in other to evaluate its use as feedstuff after the removal of antinutritional phytochemicals.

Materials and Methods

Procurement and Processing of Seeds

Ripe fruits of Napoleona imperialis were harvested at Umuahia Campus of Abia State University near Umudike, Abia State Nigeria (Latitude 50N, Longitude 70E). The fruits were broken and the seeds were put in a sealed cotton bag and fermented for 72 hours at ambient room temperature (28-320C) to make for easy washing off (with water) of the gelatinous pulp on the seed coat. The clean seeds (after washing) were dried at 450C to a moisture content of 4.8% in a convection oven (Gallenkamp Oven BS). Kett infra-red moisturemeter (Model F-IS) was used to determine the moisture content of the dried seed.

The dried seeds were then decoated and winnowed manually before milling into a fine powder (using 1mm sieve) with a Thomas milling machine (Arthur Thomas Philadelphia, P.A., USA, Cat. No. 3379-K30, Serial No. 810604).

Proximate Composition and Energy Determination

Determination of the proximate composition of the sample was based on AOAC (1990) procedures, employing the Micro-Kjeldahl method for crude protein (CP) and Soxhlet extraction method for ether extract (EE). The gross energy (GE) of the sample was determined using adiabetic oxygen bomb calorimetry technique.

Macrominerals and Hydrogen cyanide (HCN) Determinations

The dry N. imperialis seed powder was wet-digested (AOAC, 1990) for macromineral analysis. The cations content of the digested material was done with atomic absorption spectrophotometer (Unicam 929 AA) for calcium (Ca) and magnesium (Mg) and flame photometer (Jenway, Model PFP7) for sodium (Na) and potassium (k). The phosphorous (P) content of the digested solution was determined by the standard spectrophotometric Vanadomolybdate method as described by AOAC (1990).

The cyanide content was obtained by alkaline titration method involving the use of 25% NaOH, 6N NH4OH solution, 5% potassium iodide and 0.02N silver nitrate (A.O.A.C., 1990).

Saponin Content Determination

The method used was similar to that of Birk et al. (1963) as modified by Hudson and El-Difrawi (1979). Ten grammes of the N. imperialis sample was put in 100ml of 20% aqueous ethanol and agitated with a magnetic stirrer for 12 hours at 550C. Then the solution was filtered using Whatman No. 1 filter paper and the residue was re-extracted with 200ml of 20% aqueous ethanol. The extracts were combined and reduced to about 40ml under vacuum. The extract and 20ml diethyl ether were put in a 250ml separating funnel and shaken vigorously. The aqueous layer was discarded. The process of purification was continued until a colourless aqueous extract was obtained. The pH of the remaining aqueous solution was adjusted to about 4.5 by adding 4g NaCl, and the solution was then shaken successively with 60ml and 30ml portions of n-butanol.

The butanolic extract was washed twice with 10ml of 5% aqueous NaCl evaporated to dryness in a fume cupboard, to give the saponin which was weighed with a balance (Sartorius LC 1201S, Satorius AG Gottengen, Germany).

Haemolytic test with human blood

The red cell fragility test was carried out with filtered 5%, 10%, 15%, and 20% aqueous suspensions of the N. imperialis powder, and heparinized human venous blood (Baker, 1980) that was diluted to 30% (with the isotonic 0.85% saline solution).Two percent aqueous solution of crystalline haemolytic saponin (BDH Chemicals Ltd, Poole, England) and 0.85% saline solution served as controls.

Red cell pipette was used to put equal amounts (1ml) of the blood sample into small test tubes that respectively contained 1ml of the N. imperialis, saponin (BDH) and saline solution samples for visual observation for haemolysis. Disposable micropettes were used to spot the well mixed solutions on grease-free slides (Baker, 1980) and viewed under a microscope (Zeiss, Standard 25, Carl Zeiss, Germany). Observations for percentage lysed cells were done at 10 mins, 30 mins, 60 mins after the addition of the test samples to the blood (at 28-290C).

Bioassay with broiler chicks

Thirty-six day-old broiler chicks procured form P. C. Onuoha Farms Ltd., Owerri, Nigeria were used to assay for safety of the raw Napoleona imperialis seeds. The chicks were reared on commercial broiler ration for 21 days until they attained an average weight of 370g. Subsequently, they were randomly distributed into six cages (each with six birds).

Phosphate buffered saline (PBS, pH 7.0) was used to prepare seed extract (50 mg/ml concentration) to give doses and extracted quantities as shown in Table 1. The six cages were labeled T0, T1, T2, T3, T4 and T5.

Table 1. Doses of N. imperialis seed extract administered per bird.

Treatment group

Seed extract

Dosage (ml)

Soluble solids* (mg)



















*Solubles in phosphate buffered saline solution.

The extract was administered to the birds by drenching (liquid oral feeding) using a 10ml syringe. The birds were observed for 24 hours at ambient room temperature (280C) and clinical observation commenced 5 minutes after the oral administration. Observations were made at 10 minutes interval for the first 30 minutes and later at 30 minutes interval for the duration of the experiment, following the Hippocratic guide of Malone (1983).

After the clinical observation, blood was collected randomly from 3 chicks from each treatment group. The blood was collected by bleeding the birds with sterile needle under the wing. The samples were mixed thoroughly with heparin and were viewed under the microscope for haemolysis as in the case of human blood.

Results and Discussion

Table 2 shows the results of the proximate analysis of the dried milled seeds of N. imperialis. The results show that the seed meal is a possible good source of protein. This is higher than the values (10.65%, 5.29%, 10.9% and 4.88%) obtained respectively for maize, cassava (peels only), coconut seed (edible portion) and groundnut husk respectively by Oyenuga (1968). The crude protein values (11%, 10% and 10.2% ) reported by Obioha (1992) for guinea corn, millet and barley respectively are lower than that of N. imperialis. However, some recently developed maize varieties contain more than 14% crude protein (McDonald et al., 1998). The NFE value is also relative high compared to 77.84% reported for maize (Gohl, 1981). The ether extract content is low and comparable to that of cereals like maize (4.6%) and millet (4.0%) as reported by Obioha (1992).

The energy value for N. imperialis seed meal (Table 3) indicates that the meal could be used as a source of energy in livestock feed (in the absence of antinutritional factors). Table 3 also shows that N. imperialis seed is a good source of macroelements particularly Na and K. The saponin and cyanide contents of the seed meal were as presented in Table 3. Results of the red cell fragility test gave 100% lysing of the erythrocytes with the experimental aqeous suspensions of N. imperialis powder and the commercial haemolytic saponin at 10,30, and 60 minutes after the respective appplications. On the other hand, the 0.85% saline solution gave 0% haemolysis during the same time interval. Therefore, N. imperialis seed could be considered as a good source of haemolytic saponins. It is however necessary to do detailed chemical characterization of the N. imperialis saponins before large scale commercialization.

Clinical observation during the bioassay with 3-week old broiler chicks showed that all the birds (except the control, To) drenched with N. imperialis seed extract became dizzy immediately after administration of the extract (T1, T2, T3, T4 and T5). However, the intensity of dizziness was more with the chicks that received the higher dozes of 250 and 350mg.

After 1 hour, the chicks on 250 and 350mg extracts produced diarrheic droppings. The intensity was more for chicks on the 350mg dose. Ingestion of phytotoxins are known to cause certain disease conditions in animals, and Radostits et al. (1997) reported that saponins could cause gastroenteritis, manifested by diarrhea and dysentery. Generally, saponins are not absorbed by the gastrointestinal tracts of animals and need to be injected into circulatory system to cause haemolysis in the animal (Trease and Evans, 1983). In this investigation, there was no observed haemolysis in the blood of all the chicks (in the bioassay test) with a microscope and this corroborates the ascertion of Trease and Evans (1983).

The chicks on 250 and 350mg doses also visibly showed loss of appetite for about 3 hours after the administration of the extract. On the other hand, the level of feed consumption was normal in the chicks that received the lower doses of 0, 50, 100 and 150mg extract after the administration of the extract (0-24 hours).

During the 9th hour, there was observed respiratory noise in the chicks on the higher doses of 250 and 350mg, followed by gasping for air. The observed respiratory problem could be attributed to cyanide toxicity (Seifert, 1996, Radostits, et al., 1997) as HCN is a known inhibitor of the energy-giving oxygen linked respiratory activities in the cellular mitochondria (Lehninger, 1975). However, there was no mortality amongst all the chicks used for the experiment at the expiration of the bioassay at the 24th hour. Achinewhu (1997) had also noted that cyanide toxicity could lead to complications in the nervous system and other disease conditions. Oke (1975) had earlier observed that although man and animals have efficient systems to detoxify hydrogen cyanide (HCN) to thiocyanate (SCN), high ingestion of HCN overwhelms the capacity of the detoxifying systems. Achinewhu (1997) recommended that food materials with <20mg/kg HCN are safe for human consumption. Fortunately, cellular disruption (crushing) and semi-solid fermentation of cyanide rich foods (such as cassava roots) had been shown to reduce the cyanide content of food materials to safe levels, as endogenous enzymes and exogenous microbial enzymes hydrolyse the cyanogenic glycosides to release the volatile and extremely water soluble HCN. This means that an appropriate processing method could be used to reduced the cyanide content of the seed (135mg/kg) as some cassava varieties have cyanide contents as high as 263mg/kg fresh weight (Indira et al., 1994). Fermentation is also known to generally reduce or eliminate most antinutritional factors in seeds such as soyabeans (Pederson, 1979).

Table 2. Proximate composition of the seeds of Napoleona imperialis




Crude Protein (CP)

Crude Fibre (CF)

Ether Extract (EE)

Nitrogen Free Extract (NFE)








Table 3. Mineral, energy, hydrogen cyanide and saponin contents of the seeds of Napoleona imperialis on dry matter (DM) basis



Minerals (g/kg)






Energy (KJ/g)

HCN (mg/kg)

Saponin (g/kg)










Achinewhu, S.C. (1997). Toxic components of Cassava and their Mode of Removals . In: Cassava Processing for Health: Some Major Advances in Nigeria. (Ejiofor, M.A.N. and Ukpabi, U. J., Eds). Topclass, Umuahia, Nigeria. Pp 10 - 14.

A.O.A.C. (1990). Official Methods of Analysis 15th ed. Association of Official Analytical Chemists, Washingrton, D.C.

Baker, F. J. (1980). Introduction to Medical Laboratory Technology. Butterworth, London.

Birk, Y., Bondi, A., Gestetner, B. and Ishaya, I.A. (1963). Thermostable Hemolytic Factor in Soyabeans Nature. 179:1089-1090.

Dalziel, J.M. (1955). The Useful Plants of West Tropical Africa. Crown Agents For Oversea Governments and Administrations, London pp 70-71.

Gohl, B. (1981). Tropical Feeds. Feed Information Summaries and Nutritive Values FAO Animal Production and Health Series 12:344-347.

Hudson, B.J.F. and El-Difrawi, E.A. (1979). The sapogenins of the Seeds of Four Lupin Species. J. Plant Foods 3:181-186.

Hutchinson, J. and Dalziel, J.M. (1958). Flora of West Tropical Africa. Crown Agents for Oversea Governments and Administrations. London. Second ed. Vol. Part 1 pp243-245.

Indira, P., Ramanajam, T., Sinha, S.K., Maina, S.B., Nair, T.V.R. and Kurian, T. (1994). Variability in Hydrocyanic Acid Content in Cassava: Detection and Control. Central Tuber Crops Research Institute, Thiruvanathapuran, Kerela, India. Tech. Bull. Series 10.

Irvine, F.R. (1961). Woody Plants of Ghana. Oxford University Press, London pp 106- 108.

Karlson, P. (1968). Introduction to Modern Biochemistry 3rd Edition. Academic Press, New York.

Keay, R.W.J., Onochie, C.F.A. and Stanfield, D.P. (1964). Nigerian Trees. University of

Ibadan Press, Ibadan Vol. 1. pp 134-139.

Kumar, R. and D'Mello, J.P.F. (1995). Anti-nutritional Factors in Forage Legumes. In: Tropical Legumes in Animal Nutrition. D'Mello, J.P.F. and Devendra, C. (Eds). CAB International Publishers, Wallingford. pp 95-133.

Lehninger, A.L. (1975). Biochemistry. 2nd Edition. Worth Publishers, New York.

Malone, M.H. (1983). Hippocratic Screening in Animals. J. Ethnopharmacology, 8:177.

McDonald, P., Edwards, R.A., Greenhalgh, J.F.D. and Morgan, C.A. (1998). Animal Nutrition. 5th Edition. Longman Publishers, London.

Obioha F.C. (1992). A Guide to Poultry Production in the Tropics. ACENA Publishers Ltd. Enugu.

Ogbonnaya, S. C. (1983). Some Chemical Components of Some Woody Species. M.Sc. Thesis, University of Ibadan. pp 21-26.

Oke, O.L. (1975). The Role of Cassava in the Nutrition of Nigerian Population J. Root Crops 1 (1):1-18.

Oyenuga, V.A. (1968). Nigerian Foods and Feeding Stuffs: Their Chemistry and Nutritive Values 3rd Edition. Ibadan University Press, Ibadan.

Pederson, C.S. (1979). Microbiology of Food Fermentation. 2nd Edition. A.V.I., Westport, Connecticut.

Radostits, O.M., Blood, D. C. and Gray, C.C. (1997). Veterinary Medicine: A Textbook of the Diseases of Cattle, Sheep, Pigs, Goats and Horses. 8th Edition. Saunders Publishers London. pp 1580.

Seifert, H.S.H. (1996). Tropical Animal Health. Kluwer Academic Publishers, Dordrecht. p. 451.

Trease, G.E. and Evans, W.C., (1983). Pharmacognosy. Bailliers-Tindal, London, p.622.

Received 6 June 2003 ; Accepted 13 July 2003

Go to top