Protein Isolates from Bambara Groundnut (Voandzeia Subterranean L.): Chemical Characterization and Functional Properties

Abstract

The physicochemical, functional, and thermal properties of protein isolates obtained from two varieties of Bambara groundnut were evaluated. Proteins were isolated using alkaline extraction (isoelectric precipitation [IEP]) and micellisation techniques. IEP recorded a higher protein yield (56.3–58.2 g/100 g) than the micellised protein (MP) (14.2 – 15.6 g/100 g). A similar trend was observed for the protein content of the isolates. The isolates contained a high level of lysine, arginine, and glutamic acid compared to soy protein. Minimum solubility of the flours of the two varieties occured at pH 5. MP isolates exhibited higher solubility than the corresponding isoelectric (IEP) isolates over all pH values. The micellised protein recorded superior functional characteristics than the isoelectric isolates. The micellised isolates also showed a significantly higher (P < 0.05) foam capacity and stability, oil and water absorption properties than the isoelectric isolate. The MP of both varieties also recorded significantly higher emulsifying properties-+ than their isoelectric protein isolates. The micellised protein also had better gelation properties than the isoelectric isolate. Micellised and isoelectric isolates did not reveal major differences in the electrophoretic patterns; both isolates had three major bands at 35.0, 43.0, and 112.0 kDa. The bands in the isoelectric protein isolate however, were well defined compared with the micellised isolate. All Bambara isolates were not dissociated by 1,4-Dithiothreitol (DTT) suggesting that they do not contain subunits linked by a disulphide bond. This suggests that 7S vicilin may be the major storage protein in Bambara groundnut isolates. Differential scanning calorimetry studies (DSC) of the two varieties of bambara groundnut proteins indicated that the thermograms of the micellised isolates have a higher denaturation temperature Td (97.9–108.4°C) than their corresponding isoelectric isolates (89.5–90.6°C).

Keywords:

• Bambara groundnut
• Micellised protein
• Isoelectric protein
• Physicochemical properties
• Functional properties

INTRODUCTION

The production of plant proteins is of growing interest to developing and developed countries alike because of its increasing food and non-food applications.[1] Plant protein isolates are utilised in foods to improve nutritional quality and functionality. However, with respect to plant protein applications, their uses are almost limited to proteins from soybean seeds, neglecting other vegetable proteins. Therefore, there is the need to intensify research efforts aimed at identifying new vegetable protein sources.

Bambara groundnut (Voandzeia subterranean), is grown extensively in sub Sahara Africa and some parts of Latin America and Asia. It belongs to the family Fabaceae. According to Lacroix et al.,[2] Bambara groundnut is the third most important legume after groundnut (Arachis hypogea) and cowpea (Vigna unguiculata) in Africa. It serves as a low cost protein and has 6–12 g/100 g oil, 14–24 g/100 g protein and 28–40 g/100 g carbohydrate. It is underutilized because of lack of information on its compositional analysis and possible utilization. Apart from this, the long cooking time discourages its use for the preparation of local dishes.

There is a dearth of information on the preparation and properties of the protein isolates from Bambara groundnut. Therefore, the main objective of this study was to isolate the major protein fraction from Bambara groundnut using both isoelectric precipitation and micellisation and to characterize and determine the physicochemical properties of the various proteins using SDS-PAGE and differential scanning calorimetry. This will generate data that will provide basic information for understanding the structure-function relationship and potential applications of Bambara groundnut protein in domestic and industrial food products.

MATERIALS AND METHODS

Bambara groundnut seeds were obtained from the International institute for Tropical Agriculture, International Livestock Research Institute IITA/ILRI, Ibadan-Nigeria.

Preparation of Defatted Flour

Bambara groundnut seeds were dehulled manually, using a pestle and mortal. The seeds were ground in a Christy Laboratory Mill (Cheff Food Processor, Tokyo, Japan) and thereafter sieved through a screen of 20 mesh sizes before extraction for 9h with hexane in a soxhlet apparatus (10 g/100 g w/v hexane). The defatted flour was air-dried at room temperature (approximately 28°C) and subsequently kept in air-tight plastic containers at 4°C prior to use.

Preparation of Isolates

The Bambara groundnut isolates were prepared as previously described by other workers[3] with some modifications. When the protein isolate was obtained by isoelectric precipitation, it was termed isoelectric protein (IEP). The protein isolate produced by micellisation technique was termed micellisation protein (MP). The basic steps were as follows.

Slurry (1:20, flour-to-water ratio) of defatted flour in water was prepared at pH 6.37. This slurry was stirred for 2 h using a Gallenhamp magnetic stirrer (the pH was adjusted to the desired pH of 9.0 using 1 M NaOH), followed by centrifuging using a Sorvall RC5C automatic super speed refrigerated centrifuge at 10,000 × g for 30 min at 5°C. After centrifugation and recovery of supernatant, three additional extractions were carried out with half of the volume of the initial water as described above. The supernatants were pooled and the proteins precipitated at pH 5.0, the isoelectric point (IP). The precipitate formed was subsequently recovered by centrifugation at 10,000 × g for 15 min at 5°C. The precipitate was washed twice with distilled water adjusted to pH 5.0 with HCl. The precipitate was neutralised by the addition of 1 M NaOH. The final protein isolate was obtained by lyophilisation.

The MP isolate was obtained by extracting the protein from defatted meal with 0.8 M NaCl at pH 7.0 (1:10, w/v) for 2 h at 25°C. The suspension was centrifuged for 60 min at 10,000 × g at 4°C and the supernatant was concentrated by ultrafiltration to one half of its volume using a Minitan ultrafiltration system (Millipore Corp., Denvers, MA, USA) with four membranes of 10 kDa nominal molecular weight limit. The concentrated protein solution was then diluted 1:12 with cold distilled water. After stirring for 2 h at 25°C, the protein was recovered by centrifugation and then freeze-dried.

Analysis of Flour and Protein Isolates

Moisture, crude protein, fat, ash, and crude fibre of the flours and protein isolates were determined using the standard methods described in the AOAC.[4] The carbohydrate content was determined as the weight difference using moisture, crude protein, lipid, and ash content data.

Amino Acid Determination

Freeze-dried protein sample (10.0 to 11.0 mg) were hydrolysed in 10 ml, 6 mol/L HCl following the procedure of Henle et al.[5] Amino acid analysis was performed with an alpha plus amino acid analyser (LKB Biochrom, Freiburg, Germany) using an ion exchange chromatograph with a sodium system. The chemicals, standards, analysis, and the conditions of running were as described in Henle et al.[5, 6] The eluents were detected via derivatisation with ninhydrin. Peak integration was evaluated using Chromstar chromatographie software 32-bit. Dried samples from acid hydrolysis were dissolved in 0.2 mol/L Na citrate pH 2.2 and passed through a 0.20 μm membrane filter before injection into the amino acid analyser. Lysinoalanine (LAL) was also determined according to the method of Henle et al.[5] using the same amino acid analyser.

Functional Properties

Protein solubility

Protein solubility was determined by the method of Sathe et al.[7] with some modifications stated below. The suspension (0.2 g/100 g) of the flour or isolate in distilled water was adjusted to different pH values of between 2 and 12 using either 1 M HCl or 1 M NaOH. Percent nitrogen in each supernatant was determined by the micro Kjedahl method according to the AOAC method.[4] Percent soluble protein was calculated as percent nitrogen multiplied by 6.25 on a wet basis.

Foaming Properties

The foaming characteristics of the protein isolates were investigated using the methods of Coffmann and Garcia,[8] and Vani and Zayas.[9] The procedure involved blending of 50 ml of a protein suspension (1 g/100 g, adjusted to the required pH) in an Ultra-Turrax homogeniser (Janke & Kunkel, Staufen, FRG) at 12,000 rev min⁻¹ for 1 min at about 25°C and determining the volume of foam (in ml) that was present above the surface of the liquid contained in a glass cylinder of 100 ml. Foam expansion was expressed as follows:

Emulsifying Properties

The emulsifying stability index (ESI) and the emulsifying activity index (EAI) for the protein isolates were determined by the turbidimetric methods.[10, 11] Freshly prepared emulsions (1 ml) were pipetted out at 0, and 10 min after homogenisation and serially diluted with 99 ml distilled water (100-fold) followed by 1 ml of the diluted emulsion into 39 ml (40-fold) of 1g.kg⁻¹ SDS, resulting in a 4000-fold total dilution. Absorbance of the final dispersion was measured at 500 nm (Ultrospec 1000, UV/Visible Spectrophotometer, Pharmacia Biotech, Cambridge, England). The ESI and EAI were determined as follows:

where A ₀ is the absorbance of the diluted emulsion immediately after homogenisation, ΔA is the change in absorbance between 0 and 10 min (A ₀–A ₁₀ ), and t is the time interval, 10 min in this case.

where T = 2.303, C = weight of protein per unit volume (g ml⁻¹) of the protein aqueous phase before emulsion formation, Φ = oil volume fraction of the emulsion, and the dilution factor was 4000.

Water and Oil Absorption Capacity

Water absorption capacity was determined using the method of Sathe and Salunkhe[12] with slight modifications. Ten ml of distilled water was added to 1.0 g of the sample in a beaker. The suspension was stirred using a magnetic stirrer for 5 min. The suspension obtained was thereafter centrifuged at 4000 × g for 30 min and the supernatant measured in a 10-ml graduated cylinder. The density of water was taken as 1.0 g/cm³. Water absorbed was calculated as the difference between the initial volume of water added to the sample and the volume of the supernatant. The same procedure was repeated for oil absorption except Mazola corn oil was used instead of water.

Determination of the Gelation Concentration

The least gelation concentration was determined by the method of Sathe et al.[13] Test tubes containing suspensions of 2, 4, 6, 8, up to 20 g/100 g (w/v) flour in 5 ml distilled water were heated for 1 hr in boiling water, followed by cooling in ice and further cooling for 2 hr at 4°C. The least gelation concentration was the one at which the sample did not slip when the test tube was inverted.

SDS-PAGE Electrophoresis

The molecular weight profiles for the protein fractions were established using SDS-PAGE according to the method of Lane[14] with slight modification. Protein samples (1 mg/ml) were prepared in a buffer containing 4.84 g Tris, 0.03 g EDTA, 1.0 g SDS, 13.8 ml glycerol (87%) and 0.01 g orange G (pH 8.0). SDS-PAGE was carried out on a slab gel with 8 μl protein samples and the anode buffer (pH 8.9) contained 121.1 g Tris and cathode buffer (pH 8.2) contained 12.1 g Tris, 17.9 g tricine, and 1 g SDS. The separating gel was 20 g/100 g final acrylamide concentration while the stacking gel was 4 g/100 g concentration. For reduction of the protein samples, 0.5 ml of sample buffer was added to 1 mg of protein sample. Thereafter, 10 μl of 1,4-Dithiothreitol (DTT) (0.15 DTT/0.2 ml double distilled water) was added, and the solution was heated for 4 min in a water bath. After cooling, 5 μl of DTT solution and 53 μl of iodacetamide (0.2 g iodacetamide/1 ml distilled water) solution was added. The solution was stored at 4°C prior use.

The molecular weights of protein subunits for each sample were determined using the molecular weight marker from Carl Roth GmbH and SERVA electrophoresis GmbH containing the following proteins: myosin (from beef), 212 kDa; β-galactosidase (from E. coli) 118 kDa; serum albumin (from beef) 66 kDa; ovalbumin (from chicken) 43 kDa; carboanhydrase 29 kDa; trypsin inhibitor (from soybean) 20 kDa; lysozyme (from chicken) 14k Da; trypsin inhibitor (bovine lung) 6.5 kDa; and phosphorylase B 97 kDa.

Differential Scanning Calorimetry

Differential scanning calorimetry was performed according to the method described by Escamilla-Silva et al.[15] using a Thermal Advantage TA Instrument Universal Analysis 2000 differential scanning calorimeter (New Castle, DE, USA), calibrated with indium. Thermograms were obtained using slurries of protein prepared on a 20 g/100 g (w/v) basis. Samples of 10–15 mg were weighed accurately to the nearest 0.01 mg on a DSC stainless steel capsule and scanned at a heating rate of 10°C/min from 30–150°C. The onset temperature (T_o ), denaturation temperature (T_d ), and enthalpy of denaturation (ΔH) were determined using the universal analysis program version 1.9D (TA Instruments, New Castle, DE, USA).

Statistical Analysis

All values presented are means of three replicates. One-way analysis of variance (ANOVA) was carried out and differences in the means were determined at P < 0.05 (SAS[16]).

RESULTS AND DISCUSSION

Proximate Chemical Composition of Flour and Protein Isolates

Results of proximate composition are summarized in Table 1. Both varieties of Bambara groundnut had similar proximate composition; the protein content of Bambara groundnut isolates compared well with soybean protein isolates. Protein yield from the two techniques is presented in Table 2. The yield of protein isolate using the IEP (56.08–58.70 g/100 g) was higher than that obtained by MP (14.40–15.60 g/100 g). This might be due to differences in the method of isolation. Sefa-Dedeh and Stanley[17] reported a similar observation during the extraction of cowpea flour protein with sodium hydroxide; likewise, Okezie and Bello[18] noted that the use of sodium hydroxide for pH adjustment resulted in higher protein extractability from winged bean flour. In addition, legumes are noted for the high solubility of their nitrogenous constituents in NaOH.[19]

Table 1 Chemical composition of Bambara groundnut flour and protein isolates from three replicate analysis

	Bambara groundnut flour		Bambara groundnut protein isolates
Parameters (g/100 g)	White variety	Brown variety	White variety	Brown variety
Moisture	5.64 ± 0.10^a	6.82 ± 0.24^a	2.11 ± 0.10^b	2.72 ± 0.24^a
Dry matter	94.36 ± 0.56^a	93.18 ± 0.11^a	95.36 ± 1.56^a	96.18 ± 1.11^a
Protein	29.53 ± 0.27^a	30.37 ± 0.36^a	90.0 ± 1.27^a	91.0 ± 1.36^a
Lipid	9.65 ± 0.16^a	9.69 ± 0.47^a	ND	ND
Ash	4.50 ± 0.16^a	4.36 ± 0.06^a	5.04 ± 0.17^b	5.26 ± 0.23^b
Crude fiber	2.20 ± 0.06^a	3.00 ± 0.02^ab	0.02 ± 0.00^a	0.01 ± 0.00^a
Carbohydrates*	44.94 ± 0.25^b	46.76 ± 0.35^b	ND	ND
Reducing sugars (mg/100 g)	334 ± 3.61^c	430 ± 4.04^b	ND	ND
Non-reducing sugars (mg/100 g)	2.56 ± 0.34^b	3.78 ± 0.45^a	ND	ND
Starch	39.45 ± 0.31^a	40.76 ± 0.23^a	ND	ND
Gross energy (MJ/kg)	21.7 ± 0.23^b	21.75 ± 0.45^b	ND	ND
The results are mean ± SD of triplicate analysis. Means followed by different superscripts in each row indicates significant differences at P < 0.05. ND implies not detectable (i.e., below detectable limit); ND implies not determined. Results are expressed as mg/100 g.
*Carbohydrates by difference as 100 − (moisture + crude protein + lipid + ash).

Table 2 Protein yield and content of Bambara groundnut isolates obtained by isoelectric precipitation (IEP) or micellization (MP) and soybean isolate

	Bambara white variety		Bambara brown variety		Soybeans
Type of isolate	Protein yield	Protein content	Protein yield	Protein content	Protein yield	Protein content
IEP	56.30	90.0	58.2	91.0	56.6	92.0
MP	14.4	75.0	15.6	78.0	16.2	72.0

Amino Acid Composition of Bambara Groundnut Protein Isolate

The amino acid composition of the two varieties of Bambara groundnut is shown in Table 3. The result is presented alongside soybean protein isolates for comparison purposes. According to Pellet and Young,[20] the nutritive value of proteins depends primarily on the capacity to satisfy the needs of nitrogen and essential amino acids. Aspartic acid and glutamic acid were the most abundant amino acids found in Bambara groundnut protein isolates. The levels of some of the essential amino acids in the protein isolate were within that of the “ideal protein” (FAO[21]). Bambara groundnut also contained high levels of leucine, arginine, and lysine. The high lysine content of the Bambara groundnut protein is a very important nutritional attribute that makes the legume a good supplementary protein to cereals that are known to be deficient in lysine. However, the content of sulphur-containing amino acids, such as methionine, is low. Methionine and cysteine are considered to be limiting amino acids in Phaseolus beans.[22–24] The low level of methionine in Bambara groundnut protein isolate is similar to other legume seed protein reported by other studies.[25, 26] Protein sources rich in arginine and glutamine have recently gained popularity because of the reported effects of arginine in preventing heart disease[27] and of glutamine in supporting the immune system[28] and improving athletic performance.[29]

Table 3 Amino acid (g/100 g protein) composition of Bambara groundnut

Amino acid	Bambarra groundnut (white)*	Bambarra groundnut (brown)	Soybeans	FAO/WHO (children)	Reference pattern
Isoleucine ^a	3.98 ± 0.80	4.10 ± 0.90	4.0	2.8	1.3
Leucine ^a	6.70 ± 0.44	6.48 ± 0.14	7.80	6.6	1.9
Lysine ^a	5.93 ± 0.22	6.10 ± 0.43	6.40	5.8	1.6
Methionine ^a	1.40 ± 0.15	1.40 ± 0.75	1.40	2.5	1.7
Phenylalanine ^a	4.73 ± 0.24	5.28 ± 0.40	5.20	6.3	1.9
Tyrosine	7.12 ± 0.16	7.42 ± 0.51	3.80	Na	Na
Threonine ^a	5.10 ± 0.12	4.68 ± 0.21	4.10	3.4	0.9
Valine ^a	5.10 ± 0.89	4.98 ± 0.39	5.00	3.5	1.3
Histidine	2.42 ± 0.41	2.62 ± 0.65	2.80	1.9	1.6
Arginine ^a	8.06 ± 0.19	7.79 ± 0.51	7.70	Na	Na
Alanine	3.84 ± 0.73	3.67 ± 0.32	4.30	Na	Na
Aspartic acid	12.34 ± 0.09	11.90 ± 0.07	11.60	Na	Na
Glutamic acid	15.34 ± 0.94	15.08 ± 0.06	19.10	Na	Na
Glycine	2.38 ± 0.69	2.55 ± 0.05	4.70	Na	Na
Serine	5.56 ± 0.50	5.71 ± 0.59	5.60	Na	Na
TAA	90.00	89.76	93.50
TEAA	41.00	40.81	41.60
TNEAA	49.00	48.95	51.90
%TEAA/TAA	45.56	45.47	44.49
%TNEAA/TAA	54.44	54.53	55.51
^aEssential amino acids, FAO/WHO (children), and reference pattern are not available for non-essential amino acid.
*Mean values ± standard deviation of replicate analysis. Na: Not available.

The two varieties of Bambara groundnut protein isolates had very similar amino acid patterns. It is noteworthy that the levels of all the essential amino acids in Bambara groundnut protein isolates except methionine were higher than the FAO/WHO/UNU[30] recommended pattern and that bambara groundnut is comparable to soybean in the amino acid pattern. Protein isolation by alkali at high pH has been shown to cause various cross-linkages between amino acids; the cross-linkages of major concern are lysinoalanine formation from lysine and dehydroalanine through degradation of cystine or serine, because it reduces protein digestibility.[31] The isoelectric protein isolates were analysed for lysinoalanine and it was found that no lysinoalanine was produced during the isolation of Bambara protein isolate at pH 9.

Functional Properties of Bambara Groundnut Protein Isolate

Protein solubility

The pH-dependent protein solubility profile for the Bambara groundnut flours and the isolates are presented in Figs. 1a and b. It was observed that the solubility of the flours of the two varieties were pH dependent having their minimum solubility (isoelectric point) at pH 5. Generally, the solubility decreased as the pH increased until it reached the isoelectric point; this was followed by a progressive increase in solubility with a further increase in pH. A similar observation was reported for winged bean,[7] chickpea,[32] and cowpea.[33] The solubility profile of a protein provides some insight into the extent of denaturation or irreversible aggregation and precipitation that might have occurred during the isolation process. It is also important in foods because it modifies other properties, such as emulsification, foaming, and gelation.[34, 35]

Figure 1 (a) Protein solubility profile for two varieties of Bambara groundnut flours. (b) Protein solubility profiles of IEP isolate and MP isolate for the two varieties of Bambara groundnut.

There was a significant difference in the solubility of IEP and MP isolates. Micellised protein isolates exhibited higher solubility than the corresponding isoelectric isolates over all pH values. Previous studies had shown that micellised protein isolate was superior to isoelectric protein isolate in terms of solubility for safflower[36] and chickpea.[26] High solubility is related to the presence of a low number of hydrophobic residues, elevated charge, electrostatic repulsion, and ionic hydration occurring at pH above and below the isoelectric pH.[37]

Foaming Properties

The foaming capacity and stability of Bambara groundnut protein isolates are presented in Table 4. The micellised isolates (MP) showed a significantly higher foam capacity and stability in the two varieties of Bambara groundnut and soybean isolate than the isoelectric isolate (IEP). However, soy isolate exhibited better foaming capacity than Bambara isolate. Similar studies on micellised and isoelectric protein of safflower proteins gave similar results.[38] Differences in the foaming properties of the MP and IEP isolates might be due to the relative degree of denaturation of the isolates by the two techniques.[39] This observation was supported by results from the DSC studies of the protein reported in this study. The foaming properties of bambara groundnut isolate from this study indicates that it may serve as a potential replacement for better-known proteins in food applications requiring high foamability and stability, such as cakes, breads, marshmallow, whipped toppings, ice cream, and desserts.

Table 4 Functional properties of Bambara groundnut protein isolate in comparison with soybean protein isolate

	Bambara groundnut (white variety)		Bambara groundnut (brown variety)		Soy bean isolate
Properties	IEP	MP	IEP	MP	IEP	MP
Foaming capacity (%)	75 ± 1.2^a	78 ± 1.8^ab	72 ± 2.2^b	74 ± 1.8^bc	80 ± 1.3^ac	85 ± 1.1^a
Foaming stability (min)	52 ± 1.6^a	56 ± 2.0^b	50 ± 1.4^a	58 ± 1.2^ac	66 ± 1.2^c	68 ± 1.8^ab
Water absorption capacity (g/g)	6.0 ± 0.1^c	6.5 ± 0.2^b	5.8 ± 0.2^ab	6.7 ± 0.1^b	6.0 ± 0.3^a	6.6 ± 0.2^bc
Water absorption capacity (g/g)	6.9 ± 0.1^ac	7.2 ± 0.6^ab	6.6 ± 0.3^b	7.0 ± 0.4^a	6.2 ± 0.1^b	6.8 ± 0.2^c
Least gelation concentration (%)	8.0	6.0	8.0	6.0
Mean values ± standard deviation of replicate analysis, means followed by different superscript in each row indicates significant differences at P < 0.05. MP = Micelle protein; IEP = isoelectric protein.

Emulsifying Properties

The emulsifying properties of the two varieties of Bambara groundnut isolates are presented in Figs. 2a and 2b. The emulsifying properties were determined using the two indices of emulsifying activity index (EAI) and emulsifying stability index (ESI). The EAI reflects the ability of the sample to rapidly adsorb at the water-oil interphase during the formation of emulsion, thereby preventing flocculation and coalescence; while ESI reflects the ability to maintain a stable emulsion over a period by preventing the flocculation and coalescence of the oil globules.[40] The micellised protein (MP) of both varieties showed significantly (P < 0.05) higher emulsifying activity than their isoelectric protein isolates (IEP). For the micellised protein isolate, the EAI ranged from 130–138 m²/g compared with isoelectric protein isolate, which had EAI values between 120 and 126 m²/g.

Figure 2 (a) Emulsifying activity index of isoelectric isolate (IEP) and micellisation isolate (MP) of Bambara groundnut and soy protein isolate. (b) Emulsion stability index of isoelectric isolate (IEP) and micellisation isolate (MP) of Bambara groundnut and soy protein isolate (color figure available online).

Emulsifying properties of protein depends on the hydrophilic-lipophilic balance. At the oil-water interphase, the protein orients lipophilic residues to the oil phase and hydrophilic residues to the aqueous phase. The net charge of the lipophilic-hydrophilic interphase determines the emulsifying properties. The higher EAI and ESI in Bambara groundnut micellised protein might be attributed to higher solubility of the isoelectric isolate compared with micellised isolate. Similar studies showed that micellised faba bean, chickpea, and fenugreek protein isolates exhibited better emulsifying properties than isoelectric protein isolates.[41–44] Bambara groundnut protein isolate had higher EAI and ESI than soybean isolate. The EAI and ESI values of the isolates obtained in this study could serve as potential ingredients in many food formulations, such as salad dressing, sausages, comminuted meats, ice creams, cake batters, and mayonnaise.

Water and Oil Absorption Capacity

The water and oil absorption capacity of Bambara groundnut micellised isolate and isoelectric isolate in comparison with soy isolate is presented in Table 4. The water absorption capacity ranged from 5.8 g H₂0/g sample to 6.0 g H₂0/g sample for the isoelectric protein isolate, and 6.5 g H₂0/g sample to 6.7 g H₂0/g sample for the micellised protein isolate. The micellised protein isolate had a higher water absorption capacity than the isoelectric protein isolate. Isoelectric precipitation disrupts the structure of the proteins thereby limiting the interaction between the proteins and surrounding aqueous system and reducing the water absorption capacity. In contrast, in the micellised protein isolate there is exposure of a more polar and ionic group allowing greater interaction with the surrounding water by means of hydrogen bonding; other factors that affect the protein-water interaction lead to an increase in the water absorption capacity.[45, 46]

Apart from this, the micellised technique of extraction favours conformational changes in the protein molecules, which expose previously buried side chains, thereby making them available to interact with water.[47] Knorr[48, 49] reported that the water uptake of spray-dried potato protein concentrates was significantly affected by coagulation method, pH, and their interaction although no consistent trends could be observed.

The oil absorption capacity of the protein isolates ranged from 6.2 g oil/g sample to 6.9 g oil/g sample for the isoelectric isolate and from 6.8 g oil/g sample to 7.2 g oil/g sample for the micellised protein isolate. Micellised protein isolate (MP), therefore, exhibited higher oil absorption than the isoelectric protein isolate. This might be due to the effect of extraction technique on this property. Earlier authors[26, 50] have reported higher oil absorption capacity in pigeon pea and cowpea micellised protein isolate compared with their isoelectric isolate. The oil absorption capacity of Bambara groundnut in the present study was higher than those of micellised and isoelectric protein isolates from chickenpea,[26] cowpea isolates,[17] and adzuki bean.[42]

Oil absorption capacity is the ability of fat to bind non polar side chains of proteins.[7] However, the mechanism of oil absorption is not clearly understood, although various theories, such as physical entrapment of oil and also involvement of surface area, size of macromolecule, charge, and hydrophobicity, have been shown to affect oil absorption.[33] Thus, any change in any of these parameters would affect the oil absorption capacity. Alterations in conformation caused during protein isolation may also result in more oil binding sites in protein structure.[51]

Water absorption capacity is a useful indication of whether isolates can be incorporated into aqueous food formulations especially those involving dough formulations, while the oil absorption capacity may determine whether the protein material will perform well as meat extenders or analogues. The results obtained from the current investigation indicated that the isolates would be potentially useful in flavour retention, improvement of palatability, and extension of shelf life in meat products through reduction of moisture and fat loss.

Least Gelation Concentration

The least gelation concentration of Bambara groundnut protein isolates is presented in Table 4. The MP had better gelation properties than the IEP. The lower water absorption capacity exhibited by the isoelectric protein isolate may be responsible for the high least gelation concentration and weak gel formed by the IEP isolate compared with micellised protein isolate. Lopez de Ogara et al.[52] earlier reported the interrelationship between the gelation properties of chickpea protein and water absorption capacity. The least gelation concentration of Bambara groundnut protein isolate in the present study was comparable to those of cowpea and pigeon pea.[50]

Gel Electrophoresis (SDS-PAGE)

The electrophoretic patterns of Bambara proteins isolated by isoelectric precipitation and micellisation techniques are presented in Figure 3. It was carried out in the presence and absence of a reducing agent (1, 4-Dithiothreitol [DTT]). This is to distinguish between those polypeptide chains that are linked by disulphide bridges and free polypeptide chains. Soybean isolate was included in the study for comparison. Soybean isolate under non-reduced condition had about 19 subunits ranging from 14.0–212.0 kDa with three major bands at 24.0, 31.0, and 43.0 kDa (Lane 2). However, under reduced condition (with DTT) it revealed more prominent bands (Lane 8), the band around 130 kDa disappeared to give a band at 97 kDa. In addition, two pronounced bands appeared at 68 and 66 kDa and another three bands appeared between 14 and 16 kDa. This indicates that soy isolate aggregates were formed via disulphide bonds.[53] Derbyshire et al.[54] earlier reported that the parent subunit in soybean belongs to the legumin-like 11S type storage proteins, which are characterized by disulphide, linked α-β paired subunits.

Figure 3 Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) patterns of Bambara groundnut and soybean protein isolates: Lanes 2, 3, 4, 5, and 6 carried out without Dithiothreitol (DTT). Lanes 1, 7, and 13—Standard reference protein; Lane 2—Soy protein isolate (IEP); Lane 3—Bambarra isolate (IEP), white variety; Lane 4—Bambarra isolate (IEP), brown diversity; Lane 5—Bambarra isolate (MP), white variety; Lane 6—Bambarra isolate (MP), brown variety. Lanes 8, 9, 10, 11, and 12 with Dithiothreitol (DTT): Lane 8—Soy protein isolate (IEP); Lane 9—Bambarra isolate (IEP), white variety; Lane 10—Bambarra isolate (IEP), brown diversity; Lane 11—Bambarra isolate (MP), white variety; Lane 12—Bambarra isolate (MP), brown variety (color figure available online).

Observation of the polypeptide patterns of the Bambara protein isolates did not reveal major differences in the electrophoretic patterns of micellised and isoelectric isolates. Both isolates have three major bands at 35.0, 43.0, and 112.0 kDa. The bands in the isoelectric protein isolate, however, were well defined compared with the micellised isolate. A similar observation was reported by Ordorica-Falomir et al.[55] on electrophoretic patterns of safflower protein isolates.

Not all protein loaded appeared to enter the gel under non-reducing condition compared with the gel under reduced condition. They appeared as aggregate bands on the top of the gel. This resulted in a pronounced increase in the width of all the major bands in the reduced gel.

All Bambara protein isolates were not dissociated by DTT, suggesting that they do not contain subunits linked by disulphide bond and suggesting that 7S vicilin may be the major storage protein in Bambara groundnut isolates. Vicilin are glycoprotein devoid of disulphide bonds and are frequently non-covalently associated in trimers or even hexamers with molecular weights of 200.0 ± 50.0 kDa.[56]

The result from this study is in contrast with the work of Soottawat et al.[57] on Bambara groundnut protein extract. The authors reported an appearance of a new band under reducing condition. Differences in our observations might be due to the nature of the sample used. The authors used seed extract rather than the protein isolate, which was used for the present study. In addition, there were no differences in the electrophoretic pattern of the white and brown varieties of Bambara groundnut.

DSC Thermograms of Bambara Groundnut Protein Isolates (IEP and MP)

The DSC thermograms of the Bambara groundnut protein isolates are shown in Fig. 4a, while the DSC of the soybean isolate is shown in Fig. 4b. In both varieties, the thermograms of the micellised isolates have higher denaturation temperature (T_d ) than their corresponding isoelectric isolates. Denaturation temperature is the temperature at which transitions occur and it is a measure of thermal stability. The low T_d of the isoelectric isolate could be attributed to the extreme conditions of pH utilized in its isolation. It appeared that alkaline extraction (pH 9.0) and isoelectric precipitation at pH 5 resulted in structural rearrangements, which encompassed a wide molecular weight distribution. Cordero-de-los-Santos et al.[58] reported similar results in amaranth protein isolates produced by isoelectric precipitation or by micellisation. They reported that MP had well defined, narrow, and symmetrical peaks, which indicated a homogenous, and less denatured protein population. According to Arntfield and Murray,[59] and Biliaderis[60] high T_d and T_o indicate that a less heat stable group of proteins has been denatured during the process. If a protein isolate is already partly denatured, the heat of transition or enthalpy (ΔH) will decrease and if it is completely denatured no endothermic transition will appear. Soybean isolate also showed two endothermic peaks with T_d of 77.73°C for the first peak and 97.30°C for the second peak, and compared very well with Bambara groundnut isolates. The information on protein thermal properties is useful for food processing strategies and heat-processing design.[61]

Figure 4 (a) DSC thermograms of the Bambara groundnut protein isolates produced by: (A) isoelectric precipitation (white variety); (B) micellisation (white variety); (C) isoelectric precipitation (brown variety); (D) micellisation (brown variety) (color figure available online).

Figure 4 (b) DSC thermogram of soybean protein isolate produced by isoelectric precipitation (color figure available online).

CONCLUSION

Bambara groundnut legume flours and isolates are good sources of protein and are rich in essential amino acids, except sulphur-containing amino acids. This implies that the groundnut is useful as a supplementary protein source to most cereals and viable raw material for the food industry. Summarily, the micellised protein isolates had better functional characteristics than the isoelectric isolates. Both isolates possessed high water and oil absorption capacities. This suggests its potential usefulness in flavour retention, improvement of palatability, and extension of shelf life in meat products. The isolates also had good foaming capacity and stability. Since foam contributes to smoothness, lightness, flavour dispersions, and palatability, the results obtained from the study indicate that the isolates could serve as replacements of better-known proteins in food applications that require high foam capacity and stability, such as cakes, breads, marshmallow, whippings, toppings, ice creams, and desserts. The high emulsifying activity and stability of the Bambara groundnut isolates indicate that they could be used as ingredients in many food formulations, such as salad dressing, comminuted meats, ice creams, cake batters, and mayonnaise. The integrity of the isolates was preserved better by micellisation than by isoelectric precipitation. This effect was evident from DSC thermograms, which showed less denaturation in terms of the denaturation temperature T_d and the heat of denaturation ΔH of the micellised sample compared with the isoelectric isolate. The result on gel electrophoresis (SDS-PAGE) revealed that Bambara protein are devoid of disulphide bonds; hence, the protein might consist mainly of 7S globulin with a marginal amount of 11S globulin.

ACKNOWLEDGMENTS

YAA is grateful to the Alexander von Humboldt AvH Foundation for their support through the Georg Foster Fellowships. The assistance of Frau Dr. Yvonne Schneider and Frau Dr. Böhme Birgt is acknowledged.