JFS C: Food Chemistry

Extraction and Characterization of Chickpea

(Cicer arietinum) Albumin and Globulin

C: Food Chemistry
L.H. LIU, T.V. HUNG, AND L. BENNETT

ABSTRACT: Albumin and globulin fractions of 1 Desi and 2 Kabuli varieties of chickpeas (Cicer arietinum) were ex-
tracted with water and salt solutions (K 2 SO 4 and NaCl). The extractable yields and particularly the albumin-globulin
ratio varied greatly with the extraction medium and chickpea variety. Depending on the procedure employed, albu-
min could be extracted as a major fraction of chickpea proteins. Higher levels of essential amino acids and sulfur
containing amino acids were found in albumins than in globulins of all chickpeas investigated. The common struc-
tural characteristics of both Kabuli and Desi chickpea albumins and globulins were clearly identified by densitomet-
ric profiles of their sodium dodecyl sulfate polyacrylamide gel patterns. Albumins contained subunits with higher
molecular weights than those of globulins. The in vitro digestibility of the chickpea proteins by papain, pepsin, chy-
motrypsin, and trypsin indicated that globulins were more susceptible to proteolytic hydrolysis.
Keywords: albumin, characterization, chickpea protein, extraction, globulin

Introduction bumin accounted for 35% to 38% of the extractable chickpea em-

C hickpea (Cicer arietinum L.) is a highly valued commodity and

also an important source of vegetable protein for human con-
sumption. During the 2002 to 2004 period, global chickpea pro-
bryo protein.
A major water-soluble protein fraction (up to 48% of the total
protein) was obtained in our search for economical means of quan-
duction was 8 MT, a third of total grain legume productions, 2% titatively extracting chickpea proteins (Liu and others 1993). The
of which was contributed from Australia (ICRISAT 2006). Chick- different levels of chickpea albumin obtained in this study and in
pea has one of the highest nutritional compositions of any dry ed- others suggested that the extractable yield of this protein fraction
ible legume and contains only low levels of some antinutritional might vary considerably, depending on extraction conditions and
factors (Newman and others 1987; Friedman 1996). It is classi- the physicochemical procedures used (Bhatty 1982). Because of its
fied among the high-protein seeds with a mean protein content of high nutritional values and the potential applications in food uses,
22.2% (Haard and Chism 1996). the albumin fraction warrants further investigation.
Several biological studies indicated that the nutritional values This article reports the effects of extraction conditions of the
of chickpea proteins could be compared favorably with those of chickpea albumin fraction by 3 methods and the characterization
other plant proteins. Khan and others (1979) reported that chick- of these extracted proteins, including comparison with chickpea
pea proteins had a higher true digestibility, biological value, and globulin.
net protein utilization than those of cowpea and mung bean. Chan-
drasekharappa (1979) found that chickpea possessed a higher pro- Materials and Methods
tein efficiency ratio than faba bean, pigeon pea, mung bean, and
even soybean.
Materials
Like other legume proteins, chickpea albumin is a rich source of The seeds of 3 major chickpea varieties grown in Kaniva, Victo-
essential amino acids (Singh and Jambunathan 1982). Sulfurcon- ria, Australia were used: 2 Kabuli (Garnet and Kaniva) and 1 Desi
taining amino acids, tryptophan, threonine, and lysine, are higher (Dooen). Clean seeds of these varieties were ground into powder
in the albumin than in the globulin (Bhatty 1982). Chickpea al- and then passed through a sieve (850 μm). The flours were de-
bumin was also found to be readily hydrolyzed by proteolytic en- fatted twice with hexane (hexane:flour = 10:1, v/w) by stirring for
zymes (Murray and Roxburgh 1984). Despite these desirable nutri- 12 h at 5 ◦ C, filtered, and air dried. Chickpea protein content was
tional values, only a few studies of this protein fraction have been determined in triplicates of each defatted flour (0.2 g), measured
reported. This might have been because albumin was generally ac- by LECO (FP-228, St. Joseph, Mich., U.S.A.), and calculated as %N
cepted as a minor fraction while globulin and glutelin were con- × factor 6.25 (Hung and others 1993). The defatted flours from
sidered as major protein fractions. Singh and Jambunathan (1982) Kaniva, Dooen, and Garnet contained 24.1%, 25.8%, and 23.9% pro-
reported that the albumin fraction extracted from chickpea ac- tein and 0.2% fat, respectively.
counted for 12%. Similar results were reported by Bhatty (1982) and
Singh and others (1988). Schroeder and others (1988) also found an
Chemical reagents and standards
albumin–globulin ratio of 0.23 for Hyprosok, a high yield chickpea Sodium dodecyl sulfate-polyacrylamide gel electrophoresis
variety. In contrast, Murray and Roxburgh (1984) estimated that al- (SDS-PAGE) and gel filtration molecular mass standard markers
(116 to 14 kD) were purchased from BioRad Laboratories, Rich-
mond, Calif., U.S.A. Sepharose 6B and Blue Dextran 2000 were
MS 20070616 Submitted 8/8/2007, Accepted 3/19/2008. Authors are with obtained from Pharmacia, Uppsala, Sweden. Porcine pancreatic
Food Science Australia, CSIRO, 671 Sneydes Rd., Werribee, Victoria, Aus-
tralia 3030. Direct inquiries to author Liu (E-mail: [email protected]).
trypsin (type 1) EC 3.4.21.4, bovine pancreatic chymotrypsin (Type
II), porcine stomach mucosa pepsin EX 3.4.23.1, papain EC 3.4.22.2,

C 2008 CSIRO
Journal compilation 
C 2008 Institute of Food Technologists Vol. 73, Nr. 5, 2008—JOURNAL OF FOOD SCIENCE C299
doi: 10.1111/j.1750-3841.2008.00773.x
Further reproduction without permission is prohibited
 Extraction and characterization of chickpea albumin and globulin . . .

and bovine serum albumin (BSA) were supplied by Sigma Chemical Gel electrophoresis
Co., St. Louis, Mo., U.S.A. Molecular masses of chickpea proteins Analytical SDS-PAGE was carried out by a procedure similar to
were estimated using a protein standard curve of known molecular that of Weber and Osborne (1969), using an LKB-2001 vertical elec-
masses (18.5 to 106 kD). trophoresis apparatus. Stacking gel was 4.5% acrylamide in 0.125 M
Tris-HCl buffer containing 0.4% SDS at pH 6.8. Running gel was
Extraction of albumin 9.2% acrylamide in 0.43 M Tris-HCl buffer (pH 8.8) containing 0.4%
Proteins from defatted chickpea flours were extracted by the fol- SDS. Samples (0.20 to 0.30 mg/well) were dissolved in 30 μL of
C: Food Chemistry

lowing 3 methods initially at the 5-g level, then at the 100-g level to 0.05 M Tris-HCl buffer containing 2% SDS and 20% sucrose. The
confirm the results. electrophoresis was carried out at 10 mA per well in 0.025 M Tris
Method 1. The extraction and separation of albumin protein base, 0.2 M glycine buffer containing 0.1% SDS at pH 8.2 until the
were based on the method of Bhatty (1982) but with a longer ex- prestained marker reached a position of 1 cm above the glass plate
traction time. Defatted flour (100 g) was stirred for 60 min at 25 ◦ C in edge. The gel was stained with Coomassie brilliant blue R250 for 30
5% K 2 SO 4 aqueous solution (1000 mL) at pH 7. The resulting slurry min and destained in acetic acid–methanol–water (1:2:7). After be-
was centrifuged at 12000 × g for 20 min and the supernatant was ing destained, the gel was photographed and densitometric scan-
collected. The residue was extracted with a further 1000 mL of the ning of the stained gel was conducted at 650 nm using a Shimadzu
extraction medium for 30 min and separated as mentioned previ- CS-910 Tl C-scanner (Shimadzu Seisakusho Ltd., Kyoto, Japan). The
ously. The combined supernatants were dialyzed (> 10000 MWC) protein electrophoretic subunits were presented by their Rf values,
at 4 ◦ C for 48 h against 3 changes of deionized water. The dialyzed based on the distance migrated by a subunit, compared to that of
supernatant was then centrifuged for 20 min at 12000 × g to give an the fastest prestained marker.
albumin fraction (supernatant) and a globulin fraction (curd) after
freeze-drying (Bhatty 1982). In vitro digestibility assay
Method 2. The Weber’s procedure (Weber and Osborne 1969) The in vitro digestibility of chickpea protein was measured using
used by Singh and others (1981) for chickpea protein extraction the method of Hung and Ishii (1975) with some modifications. Pro-
was followed. The defatted flour samples (100 g) were extracted tein sample was dissolved in a sample buffer (0.05M Tris-HCl) to
with 0.5 M NaCl solution (1000 mL) in 0.01 M phosphate buffer form a 2% protein solution and the enzymes were dissolved in dis-
(pH 7.0) by stirring at 25 ◦ C for 60 min. The slurry was centrifuged tilled water. The pH of the samples was adjusted to maximize the
at 12000 × g for 20 min, and the residue was re-extracted by stirring reactivity of each enzyme. The optimal pH values were 2 for pepsin
for 30 min with a further 500 mL of buffer. The steps for separa- treatment and 7.5 for other enzymes. Protein solution (0.5 mL)
tion and purification of albumin and globulin were carried out as was mixed with enzyme solution in each Eppendorf centrifuge
in Method 1. tube at a concentration of 2% (protein:enzyme = 98:2). The pro-
Method 3. Proteins from the defatted flour samples (100 g) tein enzyme mixture was incubated at 37 ◦ C for various periods of
were extracted as described in method 1, but with deionized water incubation (1 to 24 h). At scheduled times, the digestion was ter-
(1000 mL × 2). minated by adding 0.4M trichloroacetic acid (TCA) (1 mL) to the
digested mixture and centrifuged for 5 min at 15000 × g. Part of
the supernatant (0.7 mL) was mixed with 20% (v/v) aqueous Folin–
Protein content and amino acid analysis
Ciocalteau reagent (0.7 mL), 0.4 M Na 2 CO 3 (2.3 mL) and held for
Samples were analyzed for nitrogen on a LECO Nitrogen Anal-
15 min. A control protein-enzyme solution was subjected to the
yser, Model FP 228, calibrated with analytical reagent grade EDTA
same treatment, and the absorbance of the control and mixture was
and a factor of 6.25 was used for protein calculation (Liu and others
measured at 650 nm with a Hitachi UV spectroscope.
1994). Amino acid content was determined as described by Fox and
The in vitro digestibility of each protein was presented by the
others (1985) except for tryptophan. Amino acids were analyzed
relative activity (RA) of each enzyme employed by measuring the
by ion exchange chromatography on a Waters HPLC system using
amount of small proteins or large peptides soluble in TCA solutions
post column derivatization with ninhydrin. Tryptophan was mea-
after the digestion of each enzyme. The relative activity was calcu-
sured following an HPLC method described by Delhaye and Landry
lated as follows:
(1986).

RA(%) = [(ODt − ODi)/ODi] × 100

Gel filtration
Qualitative characterization of the protein isolate by gel filtration where OD t was absorption measured at t time, and OD i was ab-
was carried out with Sepharose 6B at room temperature (column sorption measured at initial time.
30 × 1.5 cm) in 0.1 M Tris buffer (pH 8.2) containing 0.1% SDS. Sam-
ples (30 to 42 mg) were dissolved in a sample buffer, 0.05 M Tris- Results and Discussion
HCl buffer, containing 2% SDS, 20% sucrose (0.5 to 0.7 mL), eluted
at a rate of 11.16 mL/h, and monitored at a wavelength of 280 nm Protein extractability
(Andrews 1965). To establish a standard calibration curve for pro- The mean extraction yields of albumin and globulin frac-
tein molecular determination, a number of standards (BioRad) with tions of 3 chickpea varieties by 3 methods are summarized in
known molecular mass were used. The standards were thyroglobu- Table 1. The protein yields varied with the procedure employed
lin (670 kD), gamma globulin (158 kD), ovalbulmin (44 kD), myo- and with varieties, ranging from 43% to 77% of the total protein in
globin (17 kD), and vitamin B12 (1.35 kD). These standards were the defatted flours. More protein was extracted with either NaCl or
diluted with distilled water at a concentration of 5 to 10 mg/mL be- K 2 SO 4 solution at pH 7 (61.9% to 77.2%) than with water (43% to
fore being applied to the gel. The column void volume was deter- 49%). The difference was not significant in the protein extractabil-
mined using Blue Dextran (MW 2 × 106 ). An elution was completed ities obtained by these 2 different salt solutions. Regardless of the
by observing the complete elution of vitamin B12 with the sharp procedure employed, the highest yield of proteins was obtained
pink color. from the Dooen chickpea variety (49.9% to 77.2%). Extraction yield

C300 JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 5, 2008

Extraction and characterization of chickpea albumin and globulin . . .

of Kaniva ranged from 43.4% to 72% while that of Garnet ranged lute saline solution. A significantly higher level of albumin was ex-
from 48.8% to 61.9%. tracted with distilled water from the Dooen variety than the other
Although both salt solutions extracted similar levels of total 2 varieties. The possible different mineral composition and content
proteins, more globulin was separated from the albumin-globulin in different varieties could contribute to the different protein ex-
extracts by the K 2 SO 4 than by NaCl solutions. Using the K 2 SO 4 tractabilities in the water system, but not to those in the salt dom-
system, up to 72% of the total protein in the albumin-globulin ex- inating system. This could explain why not only albumin but also
tract of the Dooen variety was separated as globulin. In contrast, globulin was extracted with water but not comparable to the results

C: Food Chemistry
only a smaller portion (14.7%) of the total extractable protein was using saline extraction medium.
isolated as globulin using NaCl solutions. The total amount of albumin and globulin extracted with K 2 SO 4
Several investigators (Bollini and Chrispeels 1978; Youle and and NaCl in this study was similar to those reported in other studies
Huang 1978; Franco and others 1997) noted that a clear-cut dis- (Singh and Jambunathan 1982; Schroeder and others 1988; Singh
tinction between the albumin and globulin groups of plant pro- and others 1988). Although albumin was found as a major pro-
teins is not always achieved. There are some proteins showing in- tein fraction that accounted for 38% of chickpea embryo protein
termediate solubility behaviors. Albumins of castor bean and great (Murray and Roxburgh 1984), a ratio 1:4 for albumin:globulin was
northern bean behave like globulin in term of solubility (Bollini normally found not only with chickpea but also with other grain
and Chrispeels 1978; Youle and Huang 1978). Similarly, globulins legumes such as black gram (Padhye and Salunkhe 1979), pea
of lupin seed under certain circumstance were found to be soluble (Grant and others 1976), lentil, and faba bean (Bhatty 1988). In con-
in deionized water (Blagrove and Gillespie 1975; Franco and oth- trast, the ratio of albumin to globulin changed greatly in this study.
ers 1997). In an early study on albumin in pea cotyledons, Mur- The results in this study showed that under certain conditions al-
ray (1979) found a wide range of the extracted albumin (14% to bumin was isolated as a major fraction, which accounted for 44%
42% of the total protein) using K 2 SO 4 as an extract solvent. Bhatty of the total protein. The variation that was observed in this study
(1982) has also reported that wide variations between albumin has confirmed that the yield of chickpea albumin varied depending
(8.1% to 14.1%) and globulin (13.5% to 42%) fraction were found on extraction conditions, particularly the salt employed.
between different or within grain legume species. The wide vari-
ation in albumin–globulin ratio was explained by the protein ex- Amino acid profile
traction and its subsequent separation. The albumin protein could The amino acid profiles of 9 albumin and 3 globulin fractions,
be in completely extracted or coprecipitated with the globulin dur- extracted by 3 methods from Kaniva, Dooen, and Garnet, are given
ing their separation dialysis. Sodium chloride is known as a strong in Table 2. The amino acids compositions of these protein fractions
solubilize agent for various seed globulins (Soriano-Santos and showed some similarities.
others 1992; Ferreira and others 1999; Mendoza and others 2001), All 6 albumin fractions of Kaniva, Dooen, and Garnet extracted
while potassium sulfate is a globulin precipitant (Alam 2005). The by K 2 SO 4 solutions or by water showed similar contents of essen-
low yield of globulin separated from the NaCl extractable albu- tial amino acids (46.4% to 48.4%, mean = 47.4%). Compared with
min and globulin fraction found in this study suggests that in their albumin fractions isolated from NaCl solutions or the corre-
the presence of the sodium chloride, more globulins behave like sponding globulins, these 6 albumins had higher contents of lysine
albumin, which dissolved and remained in the dialyzed super- (8.1% to 8.9%, mean = 8.6%), methionine (2.1% to 2.5%, mean =
natant. In contrast, a higher portion of globulin separated from the 2.2%), and threonine (4.6% to 4.9%, mean = 4.8%) but lower con-
K 2 SO 4 extractable albumin–globulin fraction was perhaps due to tents of glutamic acids (14.2% to 14.9%, mean = 14.7%) and argi-
the globulin–albumin coprecipitating capacity of K 2 SO 4 . A detailed nine (8.4%). In contrast, the common features of globulin fractions
extractability mechanism of these salts requires further study. isolated from Dooen and Kaniva were a constant proportion of es-
Protein levels in the water extracts (method 3) of the 3 varieties sential amino acids (41.5%, 41.5%, and 41.4%) and a high level of
ranged from 43% to 50% of the total protein. After being dialyzed phenylalanine (6.8% to 6.9%) and glutamic acid (16.7% to 17.7%).
and separated, the water extracts also gave a precipitated globulin In summary, the albumin fractions have higher levels of the to-
fraction ranging from 10% to 48% and a soluble albumin fraction tal essential amino acids and particularly lysine, methionine, and
ranging from 52% to 90% of the total extractable protein. Legumes threonine than the globulin. The globulin fractions contain higher
are a source of minerals, especially a richer source of calcium and levels of phenylalanine and glutamic acid compared to the albumin
phosphorus than most cereals (Shukla and others 1983). Conse- fraction.
quently, salts in the chickpea flours may have converted the dis- Even though the fractionation was conducted under identical
tilled water, which was used as extraction medium, to a very di- conditions, it appeared to be more difficult to fractionate the globu-
lin from a mixture of albumin and globulin extracted in a NaCl solu-
tion than in a K 2 SO 4 solution. This led to a degree of mixture of both
Table 1 --- Extraction yields (percent of total protein) of
albumin and globulin fractions from Kaniva, Dooen, and albumin and globulin, which was reflected in the amino acid pro-
Garnet varieties, using K 2 SO 4 (1), NaCl (2) solutions, and file of all 3 albumin fractions extracted with NaCl solution. Conse-
water (3) at pH 7.a quently, 3 albumin fractions of Kaniva, Dooen, and Garnet varieties
Method: 1 2 3 extracted with NaCl solution had a lower amount of the total essen-
tial amino acids (mean = 44.7%) lysine, methionine, and threonine
Extraction
medium: K 2 SO 4 , pH 7 NaCl, pH 7b H2O (mean = 7.0%, 1.7%, and 4.0%, respectively) and a higher amount
of phenylalanine, glutamic acid, and arganine than those of the 6
Protein: Total Alb Glob Total Alb Glob Total Alb Glob
albumin fractions extracted with water or with K 2 SO 4 solution.
Kaniva 72.0 28.5 43.5 67.2 60.3 7.2 43.4 22.4 21.0
Dooen 77.2 21.7 55.5 77.5 66.2 11.3 49.9 44.7 5.2
Garnet 61.9 30.4 31.5 63.9 42.3 21.6 48.8 32.2 16.7
Gel filtration
a
All albumin fractions of Kaniva, Dooen, and Garnet varieties iso-
Values are means of 3 extraction yields, each in triplicate, with SD range from
2.6% to 6.5%. lated by 3 methods were subjected to gel filtration on Sepharose B6.
b
K 2 HPO 4 buffer at pH 7. The Sepharose B6 showed a similar resolving pattern for all the 9

Vol. 73, Nr. 5, 2008—JOURNAL OF FOOD SCIENCE C301

 Extraction and characterization of chickpea albumin and globulin . . .

albumin fractions (Figure 1) with 4 major fractions clearly sep- fractions extracted with NaCl solution from all chickpea varieties
arated. Based on a protein standard curve of proteins of known exhibited some globulin content.
molecular mass, 5 peaks in the albumin chromatogram were es- The electrophoretic profile of all of the globulin fractions was
timated at approximately 695, 98, 70, 19, and 15 kD, respectively. similar but it was clearly different from that of the albumin. The
The chromatographic profiles of these albumins were similar but electrophoretic profiles of all globulin fractions were similar to each
the relative proportions of individual protein were different. other but were different from the albumin both in terms of inten-
The protein eluted at the void volume (690 to 700 kD) could be sity and mobility. The globulin fraction was also characterized by 7
C: Food Chemistry

an agglomerate of several proteins and accounted for only a small distinct subunits, having Rf values of 0.35, 0.41, 0.47, 0.53, 0.73, 0.77
amount (15% of total eluted protein). The major fractions were and 0.85. Compared with the albumins, the electrophoretic profiles
those having molecular masses of 41 to 72 kD and of 15 to 19 kD, re- of globulin showed a smaller number of bands of lower molecular
spectively. The proportion of these 2 fractions varied with the pro- weight subunits with relative higher abundance.
cedure employed and with the varieties. For example, Kaniva ex- The SDS-PAGE pattern failed to distinguish the cultivar differ-
tracted with water or with K 2 SO 4 solution showed approximately ence, but albumin and globulin fractions of the chickpea proteins
equal amount of proteins between these 2 fractions. When ex- could be identified by their clear different electrophoretic patterns.
tracted with NaCl solution, the fraction (41 to 72 kD) was much In an attempt to characterize several protein fractions of the chick-
higher than another fraction (15 to 19 kD). This reflected some de- pea, Singh and others (1981) also found little differences in the SDS-
gree of albumin–globulin mixture in the sample due to the extrac- PAGE profile of Kabuli and Desi types.
tion with NaCl solution. This observation was confirmed by the
gel chromatograph of the Dooen and Garnet albumin fractions ex- Densitometric scanning profile
tracted with K 2 SO 4 solution, which showed an approximately equal The electrophoretic characteristics of chickpea albumins and
amount of protein in both fractions. globulins were examined by densitometric scanning of their SDS-
PAGE patterns. Densitometric scanning profiles of 9 albumin and
Gel electrophoresis globulin fractions of Kaniva, Dooen, and Garnet extracted by 3 pro-
The SDS-PAGE patterns of the albumin and globulin fractions cedures are given in Figure 3.
of 3 chickpea varieties extracted with water, K 2 SO 4 , and NaCl solu- The densitometric scanning profiles of 6 albumin fractions
tions are given in Figure 2. present in all 3 chickpea varieties extracted with K 2 SO 4 solution
There was no difference in SDS-PAGE patterns of all the albu- and water were almost identical, as characterized by 6 distinct
mins extracted under different conditions. The electrophoretic pro- peaks. The molecular masses of these subunits were estimated at
file of all albumin fractions was characterized by 7 distinct subunits, 96, 84, 65, 38.5, 32.5, and 20 kD, respectively. The albumin fraction
identified by the following Rf values of 0.23, 0.27, 0.38, 0.55, 0.64, from each of 3 chickpea varieties extracted by NaCl solutions also
0.87, and 0.93. For the albumin fractions extracted with NaCl, an showed the 6 distinct subunits observed previously. There were a
increase in intensity of the subunits with Rf values of 0.55 to 0.64 few extra peaks in the range of 18 to 26 kD and also at 41 to 46 kD,
and 0.87 to 0.93 was observed. Electrophoretic profiles of albumin indicating some degree of globulin contamination.

Table 2 --- Amino acid profiles (g/100 g protein) of albumin and globulin fractions of 3 chickpea varieties extracted at
pH 7 with K 2 SO 4 , NaCl solutions, and water.
Fraction: Albumin Globulin
Variety: Kaniva Dooen Garnet Kaniva Dooen
Extraction medium: K 2 SO 4 NaCla H2O K 2 SO 4 NaCla H2O K 2 SO 4 NaCla H2O K 2 SO 4 NaCla K 2 SO 4
Essential amino acid
(EAA)
Histidine 2.6 2.9 2.6 2.7 2.8 2.7 2.8 3.0 2.7 2.8 2.8 2.7
Isoleucine 4.6 4.6 4.5 4.9 5.0 5.0 4.6 4.8 4.7 4.7 4.7 4.8
Leucine 7.3 7.6 7.2 7.6 7.6 7.8 7.2 7.3 7.3 7.4 7.2 7.6
Lysine 8.8 7.4 8.9 8.1 6.6 8.8 8.1 6.9 8.9 5.6 5.2 5.6
Methionine 2.4 1.8 2.5 2.3 1.4 2.1 2.1 1.3 2.1 1.3 1.9 0.9
Cystine and cysteine 2.6 2.1 2.9 2.8 1.7 2.3 2.4 1.4 2.2 1.3 1.1 0.9
Phenylalanine 5.0 6.0 4.6 5.1 6.7 4.9 4.9 6.3 4.7 6.9 6.8 7.0
Tyrosine 3.7 3.4 3.7 3.5 3.1 3.7 3.3 3.1 3.5 2.6 2.8 2.7
Threonine 4.9 4.2 4.6 4.7 3.9 4.9 4.8 4.0 4.8 3.2 3.3 3.1
Tryptophan 0.7 0.7 0.8 0.6 0.8 0.9 0.7 0.7 0.8 0.6 0.9 0.6
Valine 5.2 5.0 5.3 5.3 5.4 5.5 5.4 5.1 5.4 5.0 5.0 5.1
Nonessential amino acid
(NEAA)
Aspartic acid 11.5 11.1 11.5 11.0 10.6 11.1 11.1 11.2 11.0 11.6 12.0 11.6
Serine 5.2 5.4 5.8 5.5 6.0 5.3 5.5 6.0 5.3 6.3 6.0 6.2
Glutamic acid 14.8 15.7 14.2 14.8 16.1 14.6 14.9 16.0 14.8 17.2 16.7 17.7
Proline 3.9 4.7 4.0 4.7 4.9 4.2 4.4 5.1 4.3 5.3 5.3 4.8
Glycine 4.5 4.1 4.3 4.2 4.0 4.5 4.2 3.9 4.2 3.7 3.8 3.8
Alanine 4.7 4.5 4.4 4.4 4.5 4.9 4.5 4.4 4.7 4.2 4.4 4.3
Arginine 7.8 8.8 9.7 7.7 9.6 7.4 8.8 9.6 9.0 10.3 10.2 10.3
EAA/total amino acid (%) 47.7 45.7 47.4 47.6 44.6 48.4 46.4 43.9 46.8 41.5 41.5 41.4
Protein recovery (%) 76.6 68.0 84.2 96.2 91.0 99.5 99.3 88.3 93.2 85.4 88.5 97.2
a
K 2 HPO 4 buffer at pH 7.

C302 JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 5, 2008

Extraction and characterization of chickpea albumin and globulin . . .

The densitometric scanning profile of all 9 chickpea globulin In vitro digestibility

fractions showed similar patterns. They were generally character- Albumin. The in vitro digestibility of Kaniva albumin by papain,
ized by 5 distinct subunits with molecular masses estimated to be trypsin, pepsin, and chymotrypsin over 24 h of digestion is given in
69, 42, 40, 24.5, and 22 kD, respectively. The profiles of globulins ex- Figure 4.
tracted with NaCl did not reveal individual peak as sharp as those The in vitro digestibility of all 3 extracted albumins by each
extracted with K 2 SO 4 or water. There was also a reduction of some of 4 proteolytic enzymes was very similar. There were only a few
subunits with molecular mass of about 65 kD. This indicated that minor differences in digestibility behavior. Albumin was more di-

C: Food Chemistry
not all albumins were separated from the globulin when extracted gestible by chymotrypsin than pepsin and trypsin, but there was
with a NaCl solution, thus forming some differences in the densito- no significant difference in the in vitro digestibility by pepsin and
metric profile of globulins extracted under different conditions. trypsin. The relative activity of papain toward the albumins was
These results indicated that chickpea albumin contained some higher than that of other 3 enzymes. After 18-h digestion, the NaCl-
subunits having higher molecular mass than chickpea globulin. extracted albumin was hydrolyzed more than that extracted with
Similarly, the densitometric profile of the albumin fraction ex- either K 2 SO 4 solution or water. An increase in relative activity of
tracted from Hyprosola and its parent—2 high yield chickpea papain found with this albumin was perhaps due to the hydroly-
varieties—showed some subunits having a higher molecular mass sis of some globulin, which failed to be removed from the albumin
(92 kD) than those of globulin (75 kD) (Schroeder and others 1988). fraction.

Figure 1 --- Gel filtration profiles of

chickpea albumins from Kaniva,
Dooen, and Garnet varieties
extracted with (1) K 2 SO 4 solution,
pH 7, (2) NaCl (K 2 HPO 4 buffer, pH 7)
solution, and (3) water, using
Sepharose B6 (30 × 1.6 cm column),
0.1 M Tris buffer (pH 8.2) containing
0.1% SDS and eluted at a rate of
11.16 mL/h.

Vol. 73, Nr. 5, 2008—JOURNAL OF FOOD SCIENCE C303

 Extraction and characterization of chickpea albumin and globulin . . .

Regardless of their different extraction conditions, the similarity

in the in vitro digestibility of the 3 extracted albumins by 4 enzymes
suggests that all these albumins have a similar structure.
Globulin. The in vitro digestibility by 4 enzymes of chickpea
globulins extracted with the 3 methods is shown in Figure 5.
The common feature of all extracted globulin fractions was that
they were more susceptible to proteolytic enzymes than their cor-
C: Food Chemistry

responding albumin fractions. The protein hydrolyzing capacity of

the 4 enzymes on the globulin fraction was, in decreasing order,
pepsin, papain, chymotrypsin, and trypsin. Pepsin was found to be
the most reactive enzyme; its relative activity increased about 12-
fold within 4 h of digestion while the relative activity of trypsin in-
creased only 4-fold after 24 h of digestion. The different rates and
degrees of hydrolysis of chickpea globulin by pepsin and trypsin
reflected the enzyme specificities. Being a nonspecific enzyme,
pepsin can catalyze the hydrolysis of a wide range of substrates
while trypsin, being more specific, can only hydrolyze lysyl and
arginyl residues (Kakade 1974).
Figure 2 --- SDS-PAGE patterns of the albumin and glob- The results confirmed the susceptible characteristic to pro-
ulin fractions. Globulin extracted with water (1 to 3), teolytic hydrolysis of chickpea globulins. In an earlier study,
with NaCl solution (4 to 6), and with K 2 SO 4 solution (7
to 9) from Garnet, Dooen, and Kaniva varieties. Albu- Clemente and others (2000) found a low peptidase–trypsin–
min extracted with water (10 to 12), with NaCl solution chymotrypsin digestibility of albumin from chickpea seeds. Simi-
(13 to 15), and with K 2 SO 4 solution (17 to 19) from Gar- lar observations were obtained with great northern bean albumins
net, Dooen, and Kaniva varieties. Marker (16), molecular
mass: 200, 116, 97, 66, 45, 22, and 14 kDa. and globulins hydrolyzed by trypsin and chymotrypsin (Sathe and
others 1981). Bhatty (1988) found that pea globulin was more sus-
ceptible to pepsin hydrolysis than pea albumin. Although all globu-
lin fractions showed a higher digestibility than their corresponding
albumins, the globulin fraction obtained by NaCl extraction exhib-
ited the highest digestibility, whereas the globulin extracted with
K 2 SO 4 solutions showed the lowest digestibility. This reflects again

Figure 4 --- In vitro digestibilities by papain, trypsin,

Figure 3 --- Densitometric scanning profiles of chickpea al- pepsin, and chymotrypsin of chickpea albumins from the
bumins and globulins from Kaniva, Dooen, and Garnet va- Kaniva variety extracted with (1) K 2 SO 4 (A1), (2) NaCl
rieties, extracted with (1) K 2 SO 4 solution, pH 7, (2) NaCl (B1) solutions, and (3) water (C1) at pH 7 are presented
(K 2 HPO 4 buffer, pH 7) solution, and (3) water at pH 7. by the relative activity (% RA) of each enzyme.

C304 JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 5, 2008

Extraction and characterization of chickpea albumin and globulin . . .

a degree of albumin content in the globulin fraction, extracted with References

K 2 SO 4 . Alam A, inventor; St. Louis, Mo., assignee. 2005. Agent for protein precipitation, a
method of protein precipitation, a method of protein assay using protein precip-
itation agent, and a kit for protein assay. Issued April 5 2005. U.S. patent 6,875,617.
Andrews P. 1965. The gel filtration behaviour of proteins related to their molecular
weights over a wide range. Biochem J 96:595–606.
Conclusions Bhatty RS. 1982. Albumin proteins of eight edible grain legume species: electrophore-

T he results of the present study suggest those relative amounts

of albumin and globulin found in chickpea depended on ex-
sis patterns and amino acid composition. J Agric Food Chem 30:620–2.
Bhatty RS. 1988. In vitro hydrolysis of pea, faba bean and lentil meals and isolated
protein fractions by pepsin and trypsin. Can Inst Food Sci Tech J 21:66–71.

C: Food Chemistry
traction conditions. Their extractable yields varied with the extrac- Blagrove RJ, Gillespie JM. 1975. Isolation, purification and characterization of the seed
tion medium and also with chickpea variety. Depending on the globulins of Lupinus angustifolius. Aust J Plant Physiol 2:13–27.
Bollini R, Chrispeels MJ. 1978. Characterization and subcellular location of vicilin and
method employed, albumin could be extracted as either a major phytohemag glutenin, the 2 major reserve proteins of Phaseolus vulgaris L. Planta
or minor protein fraction. Fractionation with K 2 SO 4 solution pro- 142:291–8.
Chandrasekharappa G. 1979. Nutritional quality of the protein of blends of wheat and
duced highly purified albumins. The structural differences of Kab- rice with Bengal grain, red grain or black grain. Nutr Reports Int 18:401–10.
uli and Desi chickpea albumins and globulins could be identified by Clemente A, Vioque J, Sanchez-Vioque R, Pedroche J, Bautista J, Millan F. 2000. Fac-
tors affecting the in vitro protein digestibility of chickpea albumins. J Agric Food
their corresponding distinct electrophoretic patterns, as the albu- Chem 80:79–84.
mins comprised subunits with high molecular masses and globulin Delhaye S, Landry J. 1986. HPLC and ultraviolet spectrometry for quantitation of tryp-
tophan in barytic hydrolysates. Anal Chem 159:175–8.
comprised subunits with low molecular masses. The level of essen- Ferreira RB, Franco E, Teixeira AR. 1999. Calcium and magnesium dependent aggre-
tial amino acids in the albumin was always found to be higher than gation of legume seed storage proteins. J Agric Food Chem 47:3009–15.
Franco E, Ferreira RB, Teixeira AR. 1997. Utilization of an improved methodology to
that of globulin. For kaniva chickpea proteins, in vitro digestibility isolate lupinus albus conglutins in the study of their sedimentation coefficients. J
of globulins was generally higher than that of albumins, and the di- Agric Food Chem 45:3908–13.
Friedman M. 1996. Nutritional value of proteins from different food sources. A review.
gestibility of globulin depended on extraction method. J Agric food Chem 44:6–29.
Fox M, Rayner C, Wu P. 1985. Amino acid composition of Australian foods. Food Tech
Aust 40:320–3.
Acknowledgments Grant DR, Summer AK, Johnson J. 1976. An investigation of pea seed albumins. Can
Inst Food Sci Tech J 9:84–91.
The authors thank Drs U. Singh (ICRISAT, India), F. Solsuski (Univ. Haard NF, Chism GW. 1996. Characteristics of edible plant tissues. In: Fennerma OR,
of Saskatchewan, Canada), B. Imison, M. Palmer, I. Batey, and the Marcel D, editors. Food chemistry. New York: Pergamon. p 952–3.
Hung TV, Ishii K. 1975. Hydrolysis of soybean protein by papain. Bull College Agric Vet
late Mr. R. Black (FSA) for their scientific discussions, and Ms. D. Med, Nihon Univ, Nihon-Daigaku-Nojoigakubu-Gakujutsu-K 32:17–27.
Womerley (FSA) for her technical assistance. This study was sup- Hung TV, Liu LH, Black RG, Trewhalla MA. 1993. Water absorption in chickpea (C.
arietinum) and field pea (P. sativum) cultivars using the Peleg model. J Food Sci
ported by the Grains Research and Development Corp., Australia. 58:848–52.
[ICRISAT] Intl. Crops Research Inst. for the Semi-arid Tropics. 2006. Available from:
http://icrisat.org/Chickpea. Accessed March 6 2007.
Kakade ML. 1974. Biochemical basis for the differences in plant protein utilisation. J
Agric Food Chem 22:550–5.
Khan MA, Jacobsen I, Eggum BO. 1979. Nutritive value of some improved varieties of
legumes. J Sci Food Agric 30:395–400.
Liu LH, Hung TV, Black R, Trewhella MA. 1993. Chickpea proteins for human con-
sumption. Proc AIFST Conv, Adelaide, Aust, May.
Liu LH, Hung TV, Black R, Trewhella MA. 1994. A study on solubility of grain legume
proteins by infrared spectroscopy. ASEAN Food J 9:24–30.
Mendoza EMT, Adachi M, Bernardo AEN, Utsumi S. 2001. Mungbean [Vigna radi-
ata (L.) Wilczek] globulins: purification and characterization. J Agric Food Chem
49:1552–8.
Murray DR. 1979. A storage role for albumin in pea cotyledons. Plant Cell Environ
2:221–6.
Murray DR, Roxburgh CM. 1984. Amino acid composition of the seed albumins from
chickpea. J Sci Food Agric 39:893–6.
Newman CW, Roth NR, Lockrman RH. 1987. Protein quality of chickpea (Cicer ariet-
inum L.). Nutr Rep Int 36:1–5.
Padhye VW, Salunkhe DK. 1979. Biochemical studies on black gram (Phaseolus mungo
L.) seeds: amino acid composition and subunit constitution of fractions of the pro-
teins. J Food Sci 44:606–10.
Sathe SK, Iyer VI, Salunkhe DK. 1981. Functional properties of the great northern bean
(Phaseolus vulgaris L.) proteins. Amino acid composition, in vitro digestibility and
application to cookies. J Food Sci 47:8–15.
Schroeder HE, Gibson AH, Oram RN, Shaikh MAQ. 1988. Seed protein characterisa-
tion and nitrogen fixation rates in the chickpea mutant Hyprosola and its parent. J
Sci Food Agric 44:31–41.
Shukla UC, Dixit ML, Arora SK. 1983. Mineral nutrition. In: Arora SK, editor. Chemistry
and biochemistry of legumes. London, U.K.: Edward Arnold. p 259–86.
Singh DK, Rao AS, Singh R, Jambunathan R. 1988. Amino acid composition of
storage proteins of a promising chickpea (Cicer arietinum L.) cultivar. J Sci Food
Agric 43:373–9.
Singh KD, Raju SM, Jambunathan R. 1981. Studies on Desi and Kabuki chickpea
(C. arietinum L.) cultivars. II. Seed protein fraction and amino acid composition.
J Food Sci Tech India 18:86–8.
Singh U, Jambunathan R. 1982. Distribution of seed protein fractions and amino acids
in different anatomical parts of chickpea (Cicer arietinum L.) and pigeon pea (Ca-
janus cajan L.). Plant Foods Hum Nutr 32:347–54.
Singh U, Jambunathan R, Gurtu S. 1981. Seed protein fractions and amino acid com-
position of some wild species of pigeon pea. J Food Sci Tech India 18:83–5.
Soriano-Santos J, Iwabuchi S, Fujimoto K. 1992. Solubility of amaranth seed proteins
in sodium sulphate and sodium chloride: the main factor in quantitative extraction
for analysis. Int J Food Sci Tech 27:337–46.
Figure 5 --- In vitro digestibilities by papain, trypsin, Weber K, Osborne M. 1969. The reliability of molecular weight determinations by do-
pepsin, and chymotrypsin of chickpea globulins from the decyl sulfate-polyacrylamide gel electrophoresis. J Biol Chem 244:4406–12.
Kaniva variety extracted with (1) K 2 SO 4 , pH 7 (A2), (2) Youle RJ, Huang AHC. 1978. Albumin storage proteins in the protein bodies of castor
NaCl (B2) solutions, and (3) water (C2) at pH 7 are pre- bean. Plant Physiol 61:13–6.
sented by the relative activity (% RA) of each enzyme.

Vol. 73, Nr. 5, 2008—JOURNAL OF FOOD SCIENCE C305