Chemical and Nutritional Properties of Dry and Wet Milling Products of Red Cowpeas

Can. Inst. Food Sci. Technol. J. Vo!. 22, No. 2, pp. 147-155, 1989

RESEARCH

Chemical and Nutritional Properties of Dry and Wet Milling Products of Red Cowpeas S. Ningsanond and B. Ooraikul Department of Food Science University of Alberta Edmonton, Alberta, Canada T6G2P5

farine de niebe rouge par le procecte sec incluerent farines et proteines de niebe obtenues par l'une ou I'autre des techniques furent aussi semblables ayant pour facteur limitant les acides amines soufres. D'apres SDS·PAGE les poids moleculaires des proteines de niebe rouge furent entre 45 000 et 65 000 daltons. Le traitement thermique de I'interaction humide de proteines de niebe a semble causer la dissociation, I'agregation et la denaturation des proteines rendant celles-ci tres insolubles et de faible valeur biologique, La courbe de solubilite en regard du pH des proteines de nieM rouge pour les farines de procectes et humides ont montre le point isoelectrique au pH 404, La teneur en Tarcin des farines de niebe rouge a varie de 3 a 4.5 mg/g d'echantillon, tandis que la teneur en inhibiteur de trypsine varia entre 3.8 et 15.6 TUlImg d'echantillon. L'activite de I'inhibiteur de la trypsine dans I'isolat proteique fut environ 75% plus faible que celle de la farine. Le CEP de farines de niebe des procedes secs et humides fut environ 1.6-1.7, compare a lA pour les graines entieres et 1.2 pour I'isolat proteique. La cuisson sous pression a 120°C pendant 20 mois augmenta le CEP des graines entieres a 2.1, Ces parametres de qualite indiquerent que pour le niebe, I'echantillon a sec produisit des produits superieurs a ceux obtenus par le procecte humide.

Abstract Red cowpea flour, and starch and protein fractions from dry and wet milling processes were evaluated for their chemical and nutritional quality. Red cowpea starch contained 24.2010 amylose, slightly higher than mung bean starch with 21.2% amylose, Sugars in red cowpea flour from dry processing contained 3% stachyose, 2.1 % raffinose, 1.9% sucrose, 0.5% glucose and 1.1 % fructose, significantly higher than those in the flour from wet processing, Most of these sugars were collected in the protein fraction after air classification of dry-milled flour. Fatty acid profiles of dry- and wetdehulled flour, protein fraction, and protein isolate were similar, However, there were indications of changes in the lipids of wetdehulled flour. Amino acid profiles of the cowpea flours and proteins produced with both techniques were also similar to one another having sulfur-containing amino acids as a limiting factor, SDSPAGE showed prominent molecular weights of red cowpea proteins to be between 45,000 and 65,000 daltons. Heat treatment in the wet extraction of cowpea protein appeared to cause dissociation, aggregation and denaturation of the protein, resulting in proteins of very low solubility and low biological quality. The pattern of solubility vs pH of red cowpea proteins from both wet- and dry-dehulled flours showed the isoelectric point at pH 404. Tannin content in red cowpea flours ranged from 3 to 4.5 mg/g sample, where the trypsin inhibitor content was between 3.8 and 15.6 TUIImg samples. Trypsin inhibitor activity in the protein isolate was about 75% lower than that of the flour. PER of cowpea flours from both dry and wet dehulling was about 1.6-1,7, compared with lA for whole seeds and 1.2 for protein isolate. Pressure cooking at 120°C for 20 min increased the PER of whole seeds to 2, I. These quality parameters indicated that for red cowpeas, dry processing generally produced products superior than those obtained from wet processing.

Introduction Carbohydrates and proteins are the major nutrients in the endosperm of legume seeds. Storage proteins of legume seeds are an important source of protein for humans in tropical areas of the world, especially where roots, tubers and starchy vegetables are the primary sources of dietary calories. The most suitable method for estimating the potential nutritional value of proteins appears to be chemical analyses of the constituent amino acids. However, the analyses have limited value because they do not take into account the effect of the presence of antinutritive compounds on the digestibility and availability of amino acids. Therefore, the biological evaluation of protein quality in relation to human nutritional needs is also important. Besides proteins and carbohydrates, legume seeds provide a good source of minerals, especially phosphorus and iron, though they are less bioavailable than from animal sources. Legume lipids are highly digestible and provide both calories and essential fatty acids. However, they are present in relatively small quantities and could produce beany and off-flavors upon oxidation,

Resume De la farine et des fractions d'amidon et de proteine de niebe rouge obtenues de procectes de mouture sec et humide furent evaluees pour leurs qualites chimiques et nutritionnelles. L'amidon de niebe rouge contenait 24.2% d'amylose, legerement plus que l'amidond'amberiqueavec21 ,2% d'amylose. Les sucresdela farine de niebe rouge par le procecte sec incluerent 3% de stachyose, 2.1 % de raffinose, 1.9% de sucrose, 0.5% de glucose et 1.1 % de fructose, significativement plus eleves que ceux de la farine par le procecte humide. La plupart de ces sucres furent groupes dans la fraction proteique apres la classification par air de la farine moulue a sec, Les profils d'acides gras furent semblables pour les farines des procedes sec et humide, pour la fraction proteique et pour I'isolat proteique. Toutefois, iI y eut des indications de changement dans les lipides de la farine obtenue du procecte humide. Les profils acides amines des farines et proteines de niebe obtenues par I'une ou I'autre des

Copyright ~ 1989 Canadian INstitute of Food Science and Technology

147

particularly by the action of lipoxygenase, as soon as the cotyledons are fractured. Cowpeas (Vigna unguiculata) are widely cultivated throughout the tropics and subtropics because they can tolerate heat and relatively dry conditions, and give adequate yields even at relatively low levels of soil fertility. Much work has been done on milling of cowpeas (Dovlo et al., 1976; Vichiensanth et al., 1979; Reichert et al., 1984; Ningsanond and Ooraikul, 1989). Some chemical and nutritional properties of certain cowpea cultivars have been reported by Elias et al. (1963), Evans and Boulter (1974), Arora and Das (1976), Molina et al. (1976), Onayemi and Potter (1976), Akpapunam and Markakis (1979, 1981), Okaka and Potter (1979), Sefa-Dedeh and Stanley (1979), Longe (1980), and Ologhobo and Fetuga (1982, 1983a, 1983b). Recently red cowpea 6-1 US has been introduced to the northeastern part of Thailand as a high-protein crop due to its suitability to the growing conditions in the region and its high protein yield per hectare when compared with other available legumes. Ningsanond and Ooraikul (1989) compared the efficacy of dry and wet milling processes on cowpeas for the production of flour, starch and protein. They concluded that wet milling may be preferred to the dry process when relatively pure flour, starch and protein are desired. However, whether the wet or dry process should be recommended as a milling procedure depends on the properties and the intended uses of the products. Chemical, functional and nutritional properties of the products would determine their suitability for a specific use. This paper compares chemical and nutritional properties of cowpea products from the two milling procedures.

Materials and Methods The red cowpea 6-1 US (Vigna unguiculata) seeds were supplied by the Department of Plant Science, Khon Kaen University, Thailand. Cowpea seeds were dehulled and milled with both wet and dry processes to produce flour, starch and protein as described by Ningsanond and Ooraikul (1989). The dry process which involved abrasive dehulling, hammer milling, pin milling and air classifications produced either dry-dehulled flour (only hammer-milled) or pin-milled flour (hammer-milled and pin-milled), starch fractions I and 11 (SI, SII) and protein fractions I and 11 (PI, PII). The wet process either involved wet dehulling with a stone mill, drying and hammer milling to produce wet-dehulled flour, or wet milling of soaked whole seeds, alkaline extraction(pH 9), acid-heat precipitation (pH 4.4, 10 min at 80°C), drying and hammer milling to produce protein isolate and wet-milled starch.

Chemical analyses The moisture and lipid contents of all samples were determined using methods 14.084 and 14.088-14.089 (AOAC, 1980), respectively. 148/ Ningsanond and Ooraikul

A Fisher Accumet pH meter, model 320 (Fisher Scientific Co., Pittsburgh, PA) was used to measured pH of the samples according to method 14.022 (AOAC, 1980). The method developed by Black and Bagley (1978) was used to extract sugars in the samples, with ethanolwater (80:20) solvent. Sugars were identified and quantitated by high pressure liquid chromatography (HPLC) on a carbohydrate column, Aminex HPX87N (Bio-Rad Laboratories (Canada) Ltd., Mississauga, ON), at 85°C, using 0.015N Na2S04 as a mobile phase at a flow rate of 0.6 mL/min (SSI pump, model 300, Scientific System, Inc., State College, PA) and a differential refractometer, model R401 (Waters Associates Inc., Milford, MA). An injection volume of 20 JLLwas used. The standard sugar solution contained about 2.4 mg/mL of pure stachyose, raffinose, glucose, and fructose, and 1.2 mg/mL of sucrose. The peak area of each sugar was obtained with a Chromatopac C-R3A integrator (Shimadzu Corp., Kyoto, Japan). The amylose content of red cowpea starch was determined and compared with that of mung bean starch by the colorimetric method of Gilbert and Spragg (1964). The absorbance of amylose-iodine complex solution was measured at 680 nm. The standard curve of pure red cowpea amylose-amylopectin mixture (20:80) was used to calculate the amount of amylose in cowpea starch. Pure amylose and amylopectin from red cowpea starch were prepared using the method of Gilbert et al. (1964). For fatty acid analysis, samples of 5-20 g, containing about 200-300 mg oil, were extracted with redistilled ethyl ether for 24 h in a soxhlet extractor. The solvent was evaporated on a steam bath. Oil samples of about 300 mg were saponified and fatty acids were liberated and esterified in the presence ofBF3 catalyst according to methods 28.053-28.056 (AOAC, 1980). Methyl esters of the fatty acids were determined using a Varian model 3700 gas chromatograph (Varian Assoc. Instr., Palo Alto, CA) equipped with a 50m x 0.25mm i.d. fused silica capillary column, a flame ionization detector attached to a Hewlett-Packard integrator model 2645A (Hewlett-Packard, Palo Alto, CA) and an automatic split-type injector. To analyze for amino acids, samples of 1.00 g were hydrolyzed by refluxing with 25 mL of 6N HCI for 24 h. The solution was filtered and 5 mL of the filtrate, containing about 14 mg protein, was evaporated to dryness with pure nitrogen at 50°C. HCI solution of pH 2, containing internal standard, was added with thorough mixing. The mixture was evaluated for amino acids in a Beckman automatic amino acid analyzer model 121MB (Beckmanlnstruments, Inc., PaloAlto, CA), equipped with a Beckman model 126 data system. Electrophoresis, in sodium dodecyl sulfate (SDS) polyacrylamide gel with linear concentrations of acrylamide ranging from 7.5 to 15%, developed by Chua (1980), was carried out to estimate molecular weights of cowpea proteins. Bovine serum albumin (MW J. Inst. Can. Sci. Technal. Aliment. Vol. 22, No. 2, 1989

66,000), ovalbumin (MW 45 ,000), glyceraldehyde-3-pdehydrogenase (MW 29,000), trypsinogen (MW 24,000), trypsin inhibitor (MW 20, 100), and alpha-lactalbumin (MW 14,200) were used as molecular weight markers. Samples (1 g) were extracted with 50mLof2% (w/v) NaCI solution for 1 h at room temperature. Soluble proteins were separated using a centrifuge (Beckman Model J-21B) at 21,000 rpm for 30 min. Supernatants were dialyzed with distilled water at 4°C for 24 hand freeze-dried. The freeze-dried samples were dissolved in sample buffer solution, containing 1% SDS, 30010 glycerol, 2% beta-mercaptoethanol, 0.01 % bromophenol blue, and Tris-HCI buffer pH 6.8, to obtain a protein concentration of 4 mg/mL. Sample solutions of 20 J.tL were loaded in a Bio-Rad model 220 dual vertical slab gel electrophoresis system with upper reservoir buffer ofTris-borate, pH 8.64, and lower reservoir buffer of Tris-HCI pH 9.18. A standard solution containing the molecular weight markers was prepared by the same procedure. Electrophoresis was performed at 15 mA for 5 h. The gels were then stained with coomassie blue R250 solution for 3 hand destained with 7% acetic acid destaining solution. The gels were equilibrated overnight with destaining solution containing 2% glycerol and dried on the cellophane membrane, as described by Wallevik and Jensenius (1982). A method modified from that used by Hang et al. (1970) and Quinn and Jones (1976) was used to determine the protein solubility profile. Samples of 1g were placed in a 125 mL flask with 40 mL of distilled water and the pH of the dispersion was adjusted as desired, ranging from 2-12, with 0.5N HCl or 0.5N NaOH. The flask was then shaken at 150 rpm for 60 min at room temperature. The pH of the solution was rechecked and readjusted after 20 min of shaking. The mixture was transferred to a 50 mL centrifuge tube and centrifuged in a Beckman centrifuge (Model J-2lB) at 10,000 rpm for 20 min, then filtered through Whatman #1 filter paper. The filtrate was analyzed for nitrogen content by a micro-Kjeldahl method according to methods 47.021-47.023 (AOAC, 1980). The protein solubility profile in the ionic solution was determined by the same procedure, but distilled water was replaced with IM NaCl.

An AACC method 71-10 (AACC, 1982) previously used for the determination of trypsin inhibitor activity of soy products was used in the determination of trypsin inhibitor in cowpea samples. One trypsin unit (TU) is defined as an increase of 0.01 absorbance unit at 410 nm per 10 mL of reaction mixture under the conditions required by the procedure. Trypsin inhibitor activity was expressed in terms of the trypsin inhibitor unit (TIU). The absorbance was measured with a Beckman DU-8 spectrophotometer. Tannin contents of red cowpea samples were estimated using the UV spectrophotometric method, with HCI-ethanol (20:80) solvent, described by Sharp et al. (1978). Tannic acid was used to prepare a standard curve.

Biological evaluation ofprotein quality Red cowpea protein samples were evaluated for their biological quality in terms of protein efficiency ratio (PER) according to methods 43.213-43.216 (AOAC, 1980). Weaned, 21 day old Wistar male rats were obtained from the National Laboratory Animal Centre, Mahidol University, Thailand. Casein, salt mixture, vitamin mixture, cellulose and corn starch were obtained from the experimental animal laboratory, the Institute of Food Research and Product Development, Kasetsart University, Thailand. The diet composition was formulated according to method 43.212 (AOAC, 1980). Soybean oil was used instead of cottonseed oil and corn starch/sucrose (1:1 w/w) was used as the carbohydrate source. Ten rats were used per feeding group, individually housed in stainless steel metabolic cages and kept at 22 ± 1.5°C, with 12 h lighting period.

Results and Discussion The amylose content of red cowpea, 24.2%, was slightly higher than the 21.2% found for mung bean. It was similar to that of pea bean, faba bean, and higher than navy bean, lentil, chick pea and great northern bean, but less than that found in horse bean and pinto bean (Lineback and Ke, 1975; Naivikul and D'Appolonia, 1979; Sathe and Salunkhe, 1981). The amylose content of red cowpea was quite low when compared with that of 22 other cowpea cuItivars which ranged

Table I. Sugars in red cowpea flours, starch and protein concentrates obtained from dry and wet dehulling processes (as 070 dry basis»). Red cowpea Stachyose 2 Raffinose 2 Sucrose2 Glucose2 Fructose 2 Total 3 sugar Lot I Pin-milled flour 8.78 3.12 2.15 1.93 0.47 Ul Protein fraction I 3.95 2.96 U5 11.11 2.65 0.40 3.24 Protein fraction 11 2.61 2.61 0.47 1.33 10.26 Starch fraction I 1.90 6.68 1.58 1.65 0.49 1.06 Starch fraction 11 2.38 0.70 6.44 1.40 1.48 0.48 Lot 2 Dry-dehulled flour 3.04 2.09 Wet-dehulled flour 2.42 1.28 IAverage of duplicate or triplicate determinations. 2By HPLC method. 3Sum of detectable sugars.

Con. Inst. Food Sci. Technol. J. Vol.

22,

No. 2, 1989

1.63 0.29

0.46 0.41

1.10 0.69

8.32 5.09

Ningsanond and Ooraikul /149

from 20.9 to 48.9070, with an average of 37.5% (Arora and Das, 1976), but was quite similar to that of wheat and potatoes. This indicates that red cowpea starch could be a good substitute in products such as transparent noodles, bread, cookies and other baked foods as well as snack goods where mung bean, wheat or potato starch is traditionally used. The sugar contents of red cowpea samples are presented in Table 1. Dry-dehulled flour contained higher concentrations of the sugars than wet-dehulled flour. In general, raffinose, glucose and fructose contents of dry-dehulled red cowpea flour were higher than those of other dry-dehulled legume flours (Naivikul and D'Appolonia, 1978; Colona et al., 1980; Sosulski et al., 1982), including those in 13 American and 20 Nigerian cowpea cultivars (Akpapunam and Markakis, 1979; Longe, 1980), whereas, stachyose and sucrose were similar to those found in the others. Air classification favored accumulation of these sugars in the light fractions, i.e. PI and PH. Oligosaccharides, especially raffinose, stachyose and verbascose, have been implicated as the causative factors of flatus (Hellendoorn, 1969; Rackis et al., 1970; Wagner et al., 1976; Fleming, 1981) since the human digestive system does not have a-galactosidase. This may be a drawback in using PI and PH in food products. The problem, however, would be less serious when wet dehulling is used in flour production since much of these sugars was removed (Table 1), presumably during soaking of the peas prior to milling. Extractable lipids and fatty acid composition are presented in Table 2. Concentrations of fatty acids among the samples were quite similar, with linoleic (C:.d, palmitic (C. 6 :0 ) and linolenic (C I8 :3) acids being the major fatty acids. Saturated fatty acids in red cowpeas, comprising about 39% of total, were the same as in ten other cowpea cultivars, but higher than

those found in soybean (1~%) and lima bean (30%). The unsaturated fatty acid content of about 60% in red cowpeas was similar to that in soybean and lima bean but higher than the 52% found in ten other cowpea cultivars as reported by Ologhobo and Fetuga (1983a). Although linoleic acid was present as a major fatty acid, the cowpea samples, with the exception of the protein isolate, would not be considered as good Sources of essential fatty acids due to the rather small amounts of extractable lipids. Wet-dehulled flour had higher extractable lipids (2.1 %, dry basis) than dry-dehulled flour (1.7%, dry basis). This was because soaking water acted as a polar solvent for polar lipids such as phosphatides, resulting in a more complete extraction of lipids. However, wet-dehulled flour had slightly lower fatty acids between C 14 - C24 (97.94% of total) than dry-dehulled flour (98.93%), PI (98.82%) and protein isolate (98.67%). This may be due to enzymatic hydrolysis and/or lipoxygenase activity, followed by fatty acid fragmentation reactions which were intensified by the absorbed moisture during soaking of the cowpea seeds. These provided fatty acid chain remnants of less than C 14 , which were grouped as "unknown" (Table 2). Haydar and Hadziyev (1973, 1974) reported similar enzymatic reactions during soaking of dry pea seeds. The amino acid compositions of red cowpea flours, PI and protein isolate are presented in Table 3. The profile of red cowpea amino acids was quite similar to that of other cowpea cultivars (Elias et al., 1963; Evans and Boulter, 1974; Molina et al., 1976; Onayemi and Potter, 1976; Okaka and Potter, 1979; Ologhobo and Fetuga, 1982), in which lysine content was high and suifur-containing amino acids were limited. In general, with a few exceptions, the recovery ofamino acids was slightly higher in the PI and protein isolate than in the flours. The protein concentrate from air classifica-

Table 2. Extractable lipids and fatty acid compositions of red cowpea flours, protein fraction I (PI) and protein isolate.

Extractable lipids l Fatty acids 2

C-14:0 14:1 16:0 16:1 18:0 18:1 18:2 18:3 20:0 20:1 22:0 22:1 24:0 Total C I4 - C24

Dry-dehulled Flour

Wet-dehulled Flour

PI

Protein Isolate

I. 7

2.1

2.9

5.9

0.10 0.63 24.68 0.04 5.28 7.18 29.10 22.16 1.92 0.51 5.68 0.37 1.28 98.93

38.94 59.99 1.07 Total fatty acids 100.00 IAverage of triplicate determinations as % of sample (dry basis). 2Average of duplicate determinations as 070 of total fatty acids. Satured fatty acids Unsaturated fatty acids Unknown

150 / Ningsanond and Ooraikul

O.ll 0.34 24.69 0.05 5.28 7.15 29.45 21.16 1.92 0.58 5.63 0.21 1.37 -97.94

0.10 0.59 24.92 0.04 4.71 6.46 27.95 24.09 1.72 0.48 5.90 0.42 1.44 -98.82

0.12 0.66 26.15 0.08 5.59 7.19 29.68 19.77 2.07 0.53 5.36 0.34 1.13 98.67

39.00 58.94 2.06 -100.00

38.79 60.03 1.18 -100.00

40.42 58.25 1.33 100.00

J. Inst. Can. Sei. Technol. Aliment. Vol.

22,

No.

2, 1989

Fig. I. SDS-PAGE of protein isolate (P), wet-dehulled flour (W) and dry-dehulled flour (D) compared with standard molecular weight marker (S).

tion and the protein isolate from the alkaline extraction-acid precipitation process had chemical scores, using the FAO protein reference pattern (FAO/WHO, 1973) as standard, similar to the dry-milled flour. Wetdehulled flour had a lower score, which indicated the loss of methionine during wet processing. Molecular weights of protein subunits of cowpea flours, PI and protein isolate, as determined by SDSPAGE, showed almost identical prominent bands with molecular weights between 45,000 and 65,000 daltons (Figure 1), which were similar to those found by 0 kaka lOO tli!

~

Cl::

E-

---

60

""Q

W;l

!-o

u ..:

ll: !-o ~

W;l

80

~

80

iZ 0

100

Z

Z

!-o

and Potter (1979) in other cowpea cultivar. The highest subunit of cowpea protein in flours was about 115,300 daltons. The difference in nitrogenous subunits between cowpea protein isolate and the flours was in the absence of bands with molecular weights of 65,000 to 91,000,36,000, and 16,000 to 20,000 daltons, and the presence of bands with molecular weights of 24,000 to 29,000 daltons in the isolate. This indicated disintegration and agglomeration of the protein and its subunits, caused by wet heat processing. Dissociation and aggregation of water-soluble cowpea protein after heat treatment, reported by Sefa-Dedeh and Stanley (1979), was also evident in this study. The protein solubility of dry-dehulled red cowpea flour at various pH's is shown in Figure 2. Using water as an extraction solvent, an isoelectric point was shown at pH 4.4, which was the same as found by Sefa-Dedeh and Stanley (1979). The protein solubility increased when pH was at either side of this point; however, the solubility in the acidic range was less than that in the alkaline range. Using the isoelectric point to coagulate cowpea protein from alkaline extraction, the theoretical yield of protein isolate would be 91.1070. The protein solubility profile of red cowpea flour displayed a common pattern found in other legumes (Fan and Sosulski, 1974). Using NaCl solution as an extraction solvent, the solubility of cowpea protein was increased from 15-20% to 55-75% at pH 3.0-6.0 due to the effect of ionic strength. At pH 8.0 and higher, the solubility of protein in NaCl solution was less than in water. This was similar to the solubility of globulins from Tendergreen seeds (Phaseo/us vulgaris) reported by Sun and Hull (1975). The solubility of cowpea protein in water was about 90% at pH 8.0 and approached 100% at pH 11.0. The solubility profile of protein of wet-dehulled flour was similar to that of dry-milled flour (Figure 2). However, its isoelectric point was between pH 4.0-4.4 and the solubility at every pH value, in both water and NaCl solution, was slightly higher than that of the dry-milled flour. This indicated changes in protein and amino acids during wet dehulling.

40

20

0

Cl:

60

ll.o Dry-dehulled flour in water Dry-dehulIed flour in 1M Nael Wet-dehulled flour in water Wel-dehulled flour in IM NaCl

Cl

""EU

40

«

Cl:

E-

>
""

20

0

0 0

2

4

10

6

12

14

pH

Fig. 2. Solubility profile of protein in dry- and wet-dehulled red cowpea flours (average of triplicate determinations).

Can. Inst. Food Sei. Teehnol. J. Vol. 22, No. 2, 1989

0

2

4

8

6

10

12

14

pH

Fig. 3. Solubility profile of red cowpea protein isolate (average of triplicate determinations).

Ningsanond and Ooraikull 151

Table 3. Amino acid compositions of red cowpea flours, protein fraction I (PI) and protein isolate (g/16g N)l. Amino acid Aspartic acid Threonine Serine Glutamine Proline Glycine Alanine Cystine2 Valine Methionine 2 Isoleucine Leucine Tyrosine Phenylalanine Lysine Ammonia Histidine Arginine 070 Recovery3

Dry-dehulled Flour

Wet-dehulled Flour

PI

Protein Isolate

II.IO

11.04 3.66 4.96 17.16 6.61 3.51 4.02 0.75 4.86 1.23 4.14 7.73 2.92 6.44 6.57 1.56 3.14 6.16 96.46 1.98 57

11.44 3.93 5.65 18.25 4.79 3.69 4.21 1.01 5.05 1.38 4.23 8.11 3.38 6.04 6.88 1.60 3.25 6.97 99.86 2.39 68

11.32 3.60 5.38 17.92 4.71 3.37 4.05 0.89 5.23 1.38 4.36 8.43 3.45 6.26 6.66 1.56 3.21 7.17

3.73 5.09 17.71 7.08 3.58 4.08 0.79 4.91 1.56 4.08 7.68 2.47 5.49 6.70 1.64 3.16 6.33 97.18 2.35 67

98.95 S-containing amino acids 2.27 Chemical score4 65 I Average of duplicate determinations. 2Corrected value by factors of 100/57.27 and 100/91.88, established in a preliminary experiment for cystine and methionine, respectively. 3Recovery of amino acids based on total protein content. 40f suifur-containing amino acid as compared with FAO rer. pattern (FAO/WHO, 1973).

The solubility of protein isolate is shown in Figure 3. The effect of protein denaturation resulting from acid precipitation and heat treatment was reflected in the very low protein solubility over a wide range of pH. A dramatic increase in the solubility of cowpea protein isolate was obtained when the pH was raised to 12.0. Shen (1976) reported similar results with soy protein which indicated that alkaline conditions could effectively resolubilize the proteins. Tannins and trypsin inhibitors represent heat-stable and heat-labile antinutritive factors, respectively. The contents of the inhibitors are shown in Table 4. The tannin content in red cowpea was relatively low, about 3-4.5 mg/g samples, when compared with 4.2-7.8 mg/g (average 5.6 mg/g) in the other cowpea cultivars studied by Ologhobo and Fetuga (1983b). However, it was higher than the 1.6 mg/g in winged bean reported by de Lumen and Salamat (1980), and the 0-2 mg/g in pigeon peas found by Price et al. (1980), who also reported no tannin in chickpeas and mung beans. Fernandez et al. (1982) found twice the amount of tannins in black, red, and white common beans as in red cowpeas, with 9.0,9.3 and 7.1 mg/g whole seeds, respectively. They also found that tannins in seed coats of colored common beans were 4-5 times higher than in cotyledons. In this study, the tannin content of red cowpea was also found to concentrate more in the seed coats. The tannin content was reduced in pin milled flour, upon dry dehulling, from 4.5 to 3.5 mg/g sample. Based on the whole seeds containing 12.2% seed coat, and drydehulled cowpea flour with 300/0 seed coat remaining (Ningsanond and Ooraikul, 1989), the tannin contents in seed coat and cotyledon were estimated to be 14.8 and 3.1 mg/g, respectively. This proportion of tannins in seed coat and cotyledon of red cowpeas was similar 152 / Ningsanond and Ooraikul

to that reported by Fernandez et al. (1982) for colored common beans. By air classification, some tannin was shifted into the protein fractions; however, more than half of the total amount remained in the starch fractions. The amount found was not expected to have a serious adverse effect on the protein fractions since studies with humans on the effect of tannins in colored bean, with an average tannin content of 11.9 mg/g, indicated that polyphenols accounted for only 7% of the reduction in true protein digestibility (Bressani et al., 1982). Some tannin might be removed during soaking in wet dehulling, as wet-dehulled flour contained slightly less tannin than the dry-dehulled flour. The tannin content was the lowest in the protein isolate. This clearly showed that wet extraction could effectively reduce the amount of polyphenols in peas through solubilization in water. These relatively small amounts of tannins would be reduced even further when the protein isolate is used as an ingredient in mixed diets. The trypsin inhibitor activity in red cowpea flours and protein isolate, ranging from 9.1 to 11.3 TUl/mg sample, was higher than in the ten cowpea cultivars with 6.1 TUl/mg sample (23.7 TUl/mg protein) reported by Ologhobo and Fetuga (1983b). The trypsin inhibitor content in red cowpea was also greater than in common bean, winged bean, faba bean, lentil, field pea, mung bean, and lupin, but less than that found in soy bean, lima bean, navy bean, northern bean and chickpea (Marquardt et al., 1975; Elias et al., 1979; de Lumen and Salamat, 1980; Elkowicz and Sosulski, 1982). Trypsin inhibitor in air-classified cowpea flour (Lot 1) was most concentrated in the protein fractions (Table 4). There was no difference in the inhibitor contents found in dry- and wet-dehulled flours (Lot 2). J. Inst. Can. Sci. Technol. Aliment. Vol.

22,

No.

2,

1989

Table 4. Tannin and trypsin inhibitor contents of red cowpea sampies (dry basis). Trypsin Inhibitor2 Tannin As Sample TUIImg TUIImg Tannic Acid 1 (mg/g) Sample Protein Lot I Whole seeds Pin-milled flour Protein fraction I Protein fraction 11 Starch fraction I Starch fraction 11

4.5 3.5 4.2 4.1 3.0 2.5

Lot 2 3.0 Dry-dehulled flour 2.8 Wet-dehulled flour 1.7 Protein isolate 1Average of triplicate determinations. 2Average of duplicate determinations.

15.3 15.6 31.5 31.8 8.6 3.8

54.1 53.7 56.4 60.1 56.1 43.3

11.3 11.3 9.1

43.7 45.2 10.9

However, alkaline extraction, acid precipitation at 80°C for 10 min, and drying in the wet processing of protein isolate accounted for about 7511/0 reduction of the inhibitor, in TUl/mg protein, from the original concentration in the flour. This loss of trypsin inhibitor activity in cowpea protein isolate was 2.4 times greater than when field bean flour, having 55% moisture, was heated at 70°C for 30 min, but was the same when the flour was adjusted to the same moisture as the isolate and heated for 40 min at 90°C (Buera et al., 1984). According to human studies on the reduction of digestibility of common bean proteins by Bressani et al. (1982), trypsin inhibitors probably accounted for as much as 25% of the reduction of true protein digestibility. On the basis of this antinutritive factor, therefore, there should be little problem with the use of protein isolate in mixed diets. The protein efficiency ratio (PER) and protein quality of red cowpea seeds, flours and protein isolate are presented in Table 5. PER of protein isolate was the lowest, with protein quality about half that of casein. Protein quality of the whole seed was improved from 56.3 to 82.3% by pressure cooking at l20°C for 20 min. This was due to the effect of heat on heat-labile antinutritive factors. However, heat treatment of protein isolate, which significantly reduced the trypsin inhibitor, seemed to have adverse effects on protein quality. According to Thompson and Erdman (1981), excessive heat treatment of soy isolate caused only a 14% decrease in methionine when chemically analyzed, but its PER dropped from2.IOto 1.12as methionine availability was reduced by about 46% by the excess heat. A very low PER (1.19) of protein isolated from mung bean using a similar processing technique was reported by Bhumiratana (1977). Kon et al. (1971) reported that PER of small white bean slurry, cooked at pH 3.5 was 1.08, whereas, that of the slurry cooked at pH 6.7 was 1.37. Similarly, Chang and Satterlee (1979) found that PER of bean protein concentrate (BPC), obtained by acid precipitation at room temperature, had a higher nutritional quality than did BPC from acid precipitation at 90°C. It is possible that Can. Inst. Food Sci. Technol. J. Vol.

22,

No.

2,

1989

Table 5. Protein efficiency ratio (PER) and protein quality of cowpea fractions l . Sample Weight PER 2 PER Protein Gain (g) Corrected Quality3 Whole seeds 1.41 27.0 1.30 Cooked whole seeds4 46.1 2.06 1.90 Dry-dehulled flour 28.7 1.53 1.66 Wet-dehulled flour 39.6 1.60 1.73 Protein fraction I 1.47 35.6 1.59 Protein isolate 30.7 1.14 1.23 Casein 96.2 2.31 2.50 lAverage of 10 rats. 2pER = weight gain/protein intake. 3protein quality = 100 x PER sample/PER casein. 4Soaked seeds were cooked at 120°C for 20 min.

56.3 82.3 66.2 69.3 63.6 49.4 100.0

the low biological quality of cowpea protein isolate was caused mainly by heat treatment at the isoelectric point, not by alkali extraction. Molina et al. (1976) reported that black-eyed pea protein, extracted by the same technique used in this study without heating at the isoelectric point, had a higher PER (1.86) than peeled black-eyed pea flour (1.43). As well, alkalitreated spun soy isolate fed to rats for 90 days as the only protein source in a well-balanced diet did not result in any significant toxicological effect (Beek et al., 1974). Therefore, it was not likely that the moderate alkaline extraction per se was detrimental to the biological quality of proteins. However, the alkaline treatment probably induced subsequent changes in protein during heat treatment at the isoelectric point since more severe alkaline treatment evidently resulted in racemization of amino acids, depolymerization of protein, and formation of unusual amino acid crosslinks such as Iysinoalanine and lanthionine (Provansal et al., 1975; Masters and Friedman, 1979) which decreased protein quality. Red cowpea flours fed to rats exhibited essentially the same PER as unheated soy bean flour (Kakade et al., 1973), small white bean powders (Kon et al. ,1974), autoclave-treated field pea (Sarwar et al., 1975), and wet-dehulled and drum-dried cowpea (Onayemi and Potter, 1976). The flours showed higher PER than mung bean, pigeon peas, lentils, red grams and pinto beans (Kon et al., 1974; Liener, 1976). In most cases, besides destruction of heat-sensitive antinutritive substances by suitable heat treatment, PER of legume flours and protein concentrates or isolates could be considerably improved by supplementation with Scontaining amino acids from other sources (Longenecker et al., 1964; Kon et al., 1974; Mattil, 1974; Liener, 1976; Onayemi and Potter, 1976; Okaka and Potter, 1979; Akpapunam and Markakis, 1981). With fairly good biological quality of cowpea native protein, compared to other legume proteins, there should be no problem in using cowpea protein as a diet supplement. Furthermore, the protein quality of cowpeas may be improved by amino acid supplementation or by suitable heat treatment, as with proteins of other legumes. Ningsanond and Ooraikul /153

Conclusion Generally, red cowpea flours, protein and starch were similar to those of other legumes in chemical and biological properties. Sugars, tannins and trypsin inhibitors were accumulated into protein fractions of dry-milled cowpeas. Oligosaccharides and antinutritive factors were partially removed by wet dehulling. Amino acid and fatty acid profiles of flours and protein concentrates from both wet and dry processing techniques were similar. Wet-dehulled flour was slightly superior to dry-dehulled flour in protein quality. However, some changes in wet-dehulled flour and protein isolate occurred, as reflected in their fatty acid and protein solubility profiles. SDS-PAGE and protein solubility data showed that protein isolated by the alkaline extraction-acid precipitation technique was denatured, with disintegration and agglomeration of the protein having occurred, resulting in reduction of its biological quality. With process modification such as a milder heat treatment, the protein isolated by wet processing could have better biological quality. Nevertheless, dry milling may be preferred to wet milling due to its ease of operation, better overall chemical properties of the products, less waste disposal problems, and lower operation costs, provided that suitable facilities are available. Acknowledgements Financial support from the IDRe and the supplies of materials by Khon Kaen, Mahidol and Kasetsart Universities are gratefully acknowledged. Technical assistance in amino acid analysis from Mrs. M. Miko and fatty acid analysis from Mrs. M. Fenton, Department of Animal Science, University of Alberta, is highly appreciated. References AACC. 1982. Approved Methods of the American Association of Cereal Chemistry. The Association, St. Paul, MN. Akpapunam, M.A. and Markakis, P. 1979. Oligosaccharides of 13 American cultivars of cowpeas (Vigna sinensis). J. Food Sci.44:1317. Akpapunam, M.A. and Markakis, P. 1981. Protein supplementation of cowpeas with sesame and watermelon seeds. J. Food Sci. 46:960. AOAC. 1980. Official Methods of Analysis. 13th ed. Association of Official Analytical Chemists. Washington, DC. Arora, S.K. and Das, B. 1976. Cowpea as potential crop for starch. Die Stiirke. 28(5):158. Beek, L.V., Ferron, V.J. and de Groot, A.P. 1974. Nutritional effect of alkali-treated soy protein in rats. J. Nutr. 104:1630. Bhumiratana, A. 1977. Mung bean and its utilization in Thailand. Food (Thai). 9(4):5. Black, L.T. and Bagley, E.B. 1978. Determination of oligosaccharides in soybeans by high pressure liquid chromatography using an internal standard. J. Am. Oil Chem. Soc. 55:228. Bressani, R., Elias, L.G. and Braham, J .E. 1982. Reduction of digestibility of legume proteins by tannins. J. Plant Foods. 4:51. Buera, M.P., Pilosof, A.M.R. and Bartholomai, G.B. 1984. Kinetics of trypsin inhibitory activity loss in heated flour from bean, Phaseolus vulgaris. J. Food Sci. 49: 124

154 / Ningsanond and Ooraikul

Chang, K.C. and Satterlee, L.D. 1979. Chemical, nutritional and microbiological quality of a protein concentrate from culled dry beans. J. Food Sci. 44: 1589. Chua, N.H. 1980. Electrophoretic analysis of chloroplast proteins. Methods in Enzymology. 69(40):434. Colona, P., Gallant, D. and Mercier, C. 1980. Pisum sativum and Vicia jaba carbohydrates: studies of fractions obtained after dry and wet protein extraction processes. J. Food Sci.45:1629. de Lumen, B.O. and Salamat, L.A. 1980. Trypsin inhibitor activity in winged bean (Psophocarpus tetragonolobus) and the possible role of tannin. J. Agric. Food Chem. 28:533. Dovlo, F.E., WiIliams, C.E. and Zoaka, L. 1976. Cowpeas: home preparation and use in West Africa. IDRC-055e. 96 pp. Ind. Dev. Res. Ctr., Ottawa. Elias, L.G., Colindres, R. and Bressani, R. 1963. The nutritive value of eight varieties of cowpea (Vigna sinensis). J. Food Sci. 29:118. Elias, L.G., de Fermindez, D.G. and Bressani, R. 1979. Possible effects of seed coat polyphenols on the nutritional quality of bean protein. J. Food Sci. 44:524. Elkowicz, K. and Sosulski, F.W. 1982. Antinutritive factors in eleven legumes and their air-classified protein and starch fractions. J. FoodSci. 47:1301. Evans, M. and Boulter, D. 1974. Chemical methods suitable for screening for protein content and quality in cOwpea (Vigna unguiculata) meals. J. Sci. Food Agric. 25:311. Fan, T.Y. and Sosulski, F.W. 1974. Dispersibility and isolation of proteins from legume flours. Can. Inst. Food Sci. Techno!. J. 7:256. FAO/WHO. 1973. Energy and protein requirements. Reponrt of a joint FAO/WHO ad hoc expert committee. WHO Techn. Rep. Ser. 522. WHO, Geneva. Fleming, S.E. 1981. A study of relationships between flatus potential and carbohydrate distribution in legume seeds. J. Food Sci. 46:794. Fermindez, R., Elias, L.G., Braham, J.E. and Bressani, R. 1982. Trypsin inhibitors and hemagglutinins in beans (Phaseolus vulgaris) and their relationship with the content of tannins and associated polyphenols. J. Agric. Food Chem.30:734. Gilbert, G.A. and Spragg, S.P. 1964. lodimetric determination of amylose-iodine sorption: "blue-value". In: Methods in Carbohydrate Chemistry. Vo!. 4. R.L. Whistler (Ed.). p. 168. Academic Press, Inc., NY. Gilbert, L.M., Gilbert, G.A. and Spragg, S.P. 1964. Amylose and amylopectin from potato starch. In: Methods in Carbohydrate Chemistry. Vo!. 4. R.L. Whistler (Ed.). p. 25. Academic Press, Inc., NY. Hang, Y.D., Steinkraus, K.H. and Hackler, L.R. 1970. Comparative studies on the nitrogen solubility of mung beans, pea beans and red kidney beans. J. Food Sci. 35:318. Haydar, M. and Hadziyev, D. 1973. Pea lipids and their oxidation on carbohydrate and protein matrices. J. Food Sci. 38:772. Haydar, M. and Hadziyev, D. 1974. Pea mitochondriallipids and their oxidation during mitochondria swelling. J. Sci. Food Agric. 25:1285. Hellendoorn, E.W. 1969. Intestinal effects following ingestion of beans. Food Techno!. 23:87. Kakade, M.L., Hoffa, O.E. and Liener, I.E. 1973. Contribution of trypsin inhibitors to the deleterious effects of unheated soybeans fed to rats. J. Nutr. 103:1772. Kon, S., Wagner, J.R., Becker, R., Booth, A.N. and Robbins, D.J. 1971. Optimizing nutrient availability of legume food products. J. Food Sci. 36:635. Kon, S., Wagner, J .R., Booth, A.N. 1974. Legume powders: preparation and some nutritional and physicochemical properties. J. Food Sci. 39:897. Liener, I.E. 1976. Legume toxins in relation to protein digestibility. A review. J. Food Sci. 41: 1076. Lineback, D.R. and Ke, C.H. 1975. Starches and low molecular weight carbohydrates from chick pea and horse bean flours. Cereal Chem. 52:334.

J. Inst. Can. Sci. Techno/. Aliment. Vo!. 22. No. 2, 1989

Longe, O.G. 1980. Carbohydrate composition of different varieties of cowpea (Vigna unguiculata). Food Chem. 6: 153. Longenecker, J.B;, Martin, W.~. and Sar~tt, ~.~. 1964. Protein quality Improvement m the protem efficiency of soybean concentrates and isolates by heat treatment. J. Agric. Food Chem. 12:411. Marquardt, R.R., McKirdy, J.A., Ward, T. and Campbell, L.D. 1975. Amino acid, hemagg1utinin and trypsin inhibitor levels, and proximate analyses of faba beans (Viciajaba) and faba bean fractions. Can. J. Anim. Sci. 55:421. Masters, P.M. and Friedman, M. 1979. Racemization of amino acids in alkali-treated food proteins. J. Agric. Food Chem.27:507. Mattil, K.F. 1974. Composition, nutritional, and functional properties, and quality critena of soy protein concentrates and soy protein isolates. J. Am. Oil Chem. Soc. 51 :81A. Molina, M.R., Argueta, C.E. and Bressani, R. 1976. Protein-starch extraction and nutritive value of the black-eyed pea (Vigna sinensis) and its protein concentrates. J. Food Sci. 41:928. Naivikul, O. and D'Appolonia, B.L. 1978. Comparison of legume and wheat flour carbohydrates. I. Sugars analysis. Cereal Chem.55:913. Naivikul, O. and D'Appolonia, B.L. 1979. Carbohydrates of legume flours compared with wheat flour. H. Starch. Cereal Chem. 56:24. Ningsanond, S. and Ooraikul, B. 1989. Dry and wet milling of red cowpeas. Can. Inst. Food Sci. TechnoI. J. 22:25 Okaka, J.C. and Potter, N.N. 1979. Sensory, nutritional and storage properties of cowpea powders processed to reduce beany flavor. J. Food Sci. 44:1539. Ologhobo, A.D. and Fetuga, B.L. 1982. Chemical composition of promising cowpea (Vigna unguiculata) varieties. Nutr. Reports Inter. 25:913. Ologhobo, A.D. and Fetuga, B.L. 1983a. Varietal differences in the fatty acid composition of oils from cowpea (Vigna unguiculata) and lima bean (Phaseolus lunatus). Food Chem.12:267. Ologhobo, A.D. and Fetuga, B.L. 1983b. Investigations on the trypsin inhibitor, hemagglutinin, phytic and tannic acid contents of cowpea (Vigna unguiculata) Food Chem. 12:249. Onayemi, O. and Potter, N.N. 1976. Cowpea powders dried with methionine: Preparation, storage stability, organoleptic properties, nutritional quality. J. Food Sci. 41 :48. Price, M.L., Hagerman, A.E. and Butler, L.G. 1980. Tannin content of cowpeas, chickpeas, pigeon peas and mung beans. J. Agric. Food Chem. 28:459. Provansal, M.M.P., Cuq, J.L.A. and Cheftel, J.C. 1795. Chemical and nutritional modification of sunflower proteins due to alkaline processing: Formation of amino acid cross-links and isomerization of lysine residues. J. Agric. Food Chem. 23:938.

Can. Inst. Food Sei. Teehnol. J. Vol. 22, No. 2, 1989

Quinn, J .R. and Jones, J .D. 1976. Rapeseed protein: pH solubility and electrophoretic characteristics. Can. Inst. Food Sci. TechnoI. J. 9:47. Rackis, J.J., Honig, O.H., Sessa, O.J. and Steggerda, F.R. 1970. Flavor and flatulence factors in soybean products. J. Agric. Food Chem. 18:977. Reichert, R.D., Oomah, B.D. and Youngs, C.G. 1984. Factors affecting the efficiency ofabrasive-typedehulling ofgrain legumes investigated with a new intermediate-sized batch dehuller. J. Food Sci. 49:267. Sarwar, G., Sosulski, F.W. and Bell, J .M. 1975. Nutritive value of field pea and faba bean proteins in rat diets. Can. Inst. Food Sci. TechnoI. J. 8: 109. Sathe, S.K. and Salunkhe, D.K. 1981. Isolation, partial characterization and modification of great nortern bean (Phaseolus vulgaris L.) starch. J. FoodSci. 46:617. Sefa-Dedeh, S. and Stanley, D. 1979. Cowpea proteins. 2. Characterization of Water-extractable proteins. J. Agric. Food Chem.17:1244. Sharp, R.N., Sharp, C.Q. and Kattan, A.A. 1978. Tannin content of sorghum grain by UV spectrophotometry. Cereal Chem.55:117. Shen, J.L. 1976. Solubility profile, intrinsic viscosity, and optical rotation studies of acid-precipitated soy protein and of commercial soy isolate. J. Agric. Food Chem. 24:784. Sosulski, F.W., Elkowicz, L. and Reichert, R.D. 1982. Oligosaccharides in eleven legumes and their air-classified protein and starch fractions J. Food Sci. 47:498. Sun, S. M. and Hull, T.C. 1975. Solubility characteristics of globulins from Phaseolus seeds in regard to their isolation and characterization. J. Agric. Food Chem. 23:184. Thompson, D.B. and Erdman, J. W. Jr. 1981. Nutrient bioavailability. In: Protein Functionality in Foods. J.P. Cherry (Ed.). ACS Symposium Series 147. p. 243. Am. Chem. Soc., Washington, DC. Vichiensanth, P., Anderson, A. and Ngarmsak, T. 1979. The dehulling and milling of cowpeas. Summary paper presented at the IORC Grain Legume Workshop, 3-4 December 1979. Singapore. Wagner, J.R., Becker, R., Gumbmann, M.R. and Olson, A.C. 1976. Hydrogen production in the rat following ingestion of raffinose, stachyose and oligosaccharide-free bean residues. J. Nutr. 106:466. Wallevik, K. and Jensenius, J .C. 1982. A simple and reliable method for drying of polyacrylamide slab gels. J. Biochem. Biophys. Methods. 6:17.

Submitted June 7,1988 Revised November 4, 1988 Accepted January 10, 1989

Ningsanond and Ooraikul /155