Assessment of the protein quality of fourteen soybean [Glycine max (L.) Merr.] cultivars using amino acid analysis and two-dimensional electrophoresis

www.elsevier.com/locate/foodres

Food Research International 40 (2007) 129–146

Assessment of the protein quality of fourteen soybean[Glycine max (L.) Merr.] cultivars using amino acid analysis

and two-dimensional electrophoresis

Constantinos G. Zarkadas a,*, Christine Gagnon a, Stephen Gleddie a,Shahrokh Khanizadeh b, Elroy R. Cober a, Ron J.D. Guillemette a

a Eastern Cereal and Oilseed Research Centre (ECORC), Central Experimental Farm, Research Branch, Agriculture and Agri-Food Canada,

960 Carling Avenue, Ottawa, Ont., Canada K1A 0C6b Horticulture Research and Development Center, Agriculture and Agri-Food Canada, St. Jean sur-Richelieu, PQ, Canada J3B 3E6

Received 7 August 2006; accepted 23 August 2006

Abstract

The protein quality of commercial soybeans varieties can be determined from their total protein content, their amino acid composi-tion and from the ratio of glycinin to b-conglycinin, the major seed storage protein components. In this study 14 commercial soybeancultivars were assessed. There were significant differences in storage protein composition (P < 0.05) and in their valine, proline and phen-ylalanine contents (P < 0.01 to P < 0.001). Mean protein values among these varieties ranged from 29.8% to 36.1%. The total amino acidnitrogen (AAN) ranged from 89.6 to 95.1 g AA/16 g of nitrogen, corresponding to nitrogen values from 16.5 to 17.9 g AAN/100 g pro-tein. All varieties contained a good balance of essential amino acids (EAA9), limited only in methionine. Two-dimensional gel electro-phoretic (2-DE) separations, led to the establishment of high-resolution proteome reference maps, enabling polypeptide chainidentification and calculation of the ratio of the constituent glycinin and b-conglycinin storage proteins of soybean. This method enablesthe assessment of the genetic variability of the soybean cultivars, which can then be correlated with their protein quality and food pro-cessing properties. These three methods can be used as very effective tools for assisting plant breeders in their selection of high qualitysoybean varieties.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Soybean proteins; Amino acid composition; Amino acid score; Protein quality; Soybean proteome; Glycinin/b-conglycinin ratio

0963-9969/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.foodres.2006.08.006

Abbreviations: NH4HCO3, ammonium bicarbonate; CHAPS, 3-[(3-C-HAPS cholamidopropyl)dimethyl ammonio]-1-propanesulfonate; DDT,1,4-dithiothreitol; EDTA, ethylene diamine tetra-acetate; IEF, isoelectricfocusing; IPG, imobilized pH gradient; LO, lipoxygenase; Mr, molecularmass; 1-DE, one-dimensional gel electrophoresis; PAGE, polyacrylamidegel electrophoresis; pI, isoelectric point; PMF, peptide mass fingerprinting;SDS, sodium dodecyl sulfate; 2-DE, two-dimensional electrophoresis;sLO-1, soybean lipoxygenase; TCA, trichloroacetic acid; V h, volts–hours

* Corresponding author. Tel.: +1 613 759 1642; fax: +1 613 759 1701.E-mail address: [email protected] (C.G. Zarkadas).

1. Introduction

Soybeans [Glycine max (L.) Merr.] are becoming animportant economic crop in Canada as a major source ofprotein, energy, polyunsaturated fats, fiber, vitamins, min-erals, and other nutrients, for both humans and livestock(Krishnan, 2000, 2005; Nielsen, Bassuner, & Beaman,1997). Soybean is an annual leguminous plant originatingfrom a small area of the warm climate of northeasternChina, which resulted in a very narrow genetic base (Singh& Hymowitz, 1999). Soybean is a relatively new agricul-tural crop in Canada, and until recently most of the Cana-dian production of soybeans was confined to the more

mailto:[email protected]

130 C.G. Zarkadas et al. / Food Research International 40 (2007) 129–146

temperate regions of southwestern Ontario. Because highprotein seeds are of more commercial value, considerableefforts have been made to identify the most productive highprotein cultivars that will grow in areas with longer day-lengths (>16 h), shorter growing seasons (2300–2700 cropheat units), and more northern latitudes (latitude >45�N).As a result of new breeding initiatives, several new cultivarswith improved yields, seed quality, protein content andquality, and disease resistance, are now being widely grownin the more northern regions of Canada, from Manitoba,through Ontario and Quebec, and into the Atlantic Prov-inces (Zarkadas, Yu, Voldeng, & Minero-Amador, 1993;Zarkadas, Voldeng, Yu, & Choi, 1999).

Soybean seeds contain an average of 36–38% proteinand 19% oil, on a dry weight basis (Brumm & Hurburgh,2002; Zarkadas et al., 1993, 1999), but both genetic andenvironmental factors can strongly affect the seed composi-tion (Krishnan, 2000). Soybean proteins are a complexpolymorphic mixture of polypeptides, and like many spe-cies of the Leguminosae, soybeans contain four groups ofproteins: the enzymes involved in metabolism, which donot exceed 1% of the total protein; structural proteinsincluding both ribosomal and chromosomal; membraneproteins; and the storage proteins (Krishnan, 2000; Nielsenet al., 1997).

The two major multi-subunit storage proteins of soy-beans are the salt soluble glycinin and b-conglycinin, whichaccount for 40% and 25% of the total seed endosperm pro-tein, respectively. The mean glycinin to b-conglycinin ratiohas been reported to be 1.6:1 (Nielsen et al., 1997), whileothers have reported even more extreme ratios (Fehret al., 2003; Nielsen et al., 1989). The genetic backgroundof each variety controls seed composition and how muchof the total seed protein of soybeans is contributed by thesetwo proteins (Nielsen et al., 1989). The composition ofstorage proteins also varies with maturity, nutrient supplyfrom the soil, fertilizer treatment (Krishnan, Bennett, Kim,Krishnan, & Mawhinney, 2005) and with environmentalfactors (Fehr et al., 2003). Several studies have shown thatthe accumulation of the sulfur-poor b-subunit of b-con-glycinin is promoted by excess application of nitrogen orby sulfur deficiency, while the application of sulfur fertiliza-tion increases the synthesis of glycinin (Krishnan, 2000;Krishnan et al., 2005). In addition to these two storage pro-teins, there are other known bioactive proteins, includingthe Kunitz trypsin, Bowman–Birk and related proteaseinhibitors, b-amylases, the lipoxygenases, urease and seedlectins, which each account for about 2–5% of the totalseed protein in soybeans (Liener, 1994).

Because glycinin and b-conglycinin impact greatly onthe nutritional value and quality of soybean products, theseproteins have been studied extensively, and their primaryand crystal structures have been determined (Adachiet al., 2003; Adachi, Takenaka, Gidamis, Mikami, & Uts-umi, 2001; Maruyama et al., 2001; Staswick, Hermodson,& Nielsen, 1984; Thanh & Shibasaki, 1978). Glycinin is ahexamer, with a molecular mass (Mr) � 320–380 kDa. Five

major subunits of glycinin have been identified and classi-fied into two groups, based on the identity of their aminoacid sequences (Staswick et al., 1984), namely Group I(A1aB1b, A2B1a, and A1bB2) and Group II (A3B4 andA5A4B3) (Adachi et al., 2001, 2003). Each of the subunitscharacterized so far is encoded by one of five structuralgenes, Gy1 to Gy5 (Nielsen et al., 1989). Each constituentsubunit of glycinin is composed of an acidic (A) polypep-tide with an Mr of �32 kDa and a basic (B) polypeptidewith a Mr of �20 kDa, which are linked together by aninterchain disulfide bond (Adachi et al., 2001, 2003; Stas-wick et al., 1984). Two disulfide bonds were clearly con-firmed in the crystal structure of both A1aB1b and A3B4

subunits (Adachi et al., 2001, 2003). One is an intrachaindisulfide bond, between Cys12 and Cys45 in A1aB1b sub-unit, and the other is an interchain disulfide bond, betweenCys88 in the acidic chain and Cys298 in the basic chain inthe case of A1aB1b subunit (Adachi et al., 2001, 2003; Stas-wick et al., 1984). The sequence identity in each group isabout 80%, and that between groups is about 45% (Adachiet al., 2001). Other minor protein components have beenidentified in soybeans including two new genes, namely aglycinin pseudogene, gy6, and a functional gene, Gy7,which sythesizes the glycinin subunit named G7, whichconsists of an acidic chain of Mr � 37 kDa, a basic peptidechain of 20 kDa and has an NH3-terminal valine (Beilinsonet al., 2002).

The second major storage protein, b-conglycinin, with amolecular mass (Mr) of �150 kDa, is composed of threedifferent subunits, referred to as a, a 0 and b, with Mr of76, 72 and 53 kDa, respectively, which contain no disulfidebonds (Thanh & Shibasaki, 1978) and are practicallydevoid of methionine (Maruyama et al., 2001). The b-con-glycinins are glycoproteins, containing about 5% carbohy-drate by weight, mostly of the mannose type, which iscovalently bound to asparagine residues. In addition to acommon core region which is shared by all three subunits,only the a and a 0 subunits contain an extended region intheir primary sequence, which appears to vary among thesesubunits.

Soybean storage proteins, like many globulins fromother species, characteristically have a good balance ofthe essential amino acids required by humans and animals.However, compared with meat, soybeans are deficient insulfur-containing amino acids, with methionine being con-sidered as the most significant limiting amino acid, fol-lowed by cyst(e)ine and possibly threonine (Zarkadaset al., 1993, 1999). Human nutritional studies, focusingon the utilization of various types of soybean productsfor adults, children, and full-term and premature infants,reported values for protein quality of processed soybeanprotein ranging from 62% to 92% of casein (Krishnan,2000). The nutritionally important sulfur amino acid con-tent of glycinin accounts for 3–4.5% of the amino acid res-idues (Krishnan, 2000; Liener, 1994; Nielsen et al., 1989,1997). According to Beilinson et al. (2002) the G1, G2,and G7 glycinin subunits contain higher amounts of methi-

C.G. Zarkadas et al. / Food Research International 40 (2007) 129–146 131

onine (6–7 residues per subunit) compared to G3, G4 andG5 glycinin subunits, which contain 5, 2, and 4 methionineresidues per subunit, respectively. By comparison the sulfuramino acids of b-conglycinin account for less than 1% ofthe amino acid residues (Burton, Purcell, & Walter,1982). As a result, the overall methionine content of soy-bean protein is only 1.4% of the total amino acids (Krish-nan, 2000, 2005; Nielsen et al., 1989), corresponding to�0.78% of the dry weight value reported in the 1998National Research Council nutrient requirements for swine(Steinke, 1992).

Characterizing the molecular weights, iso-electric points,charges and disulfide-bonded subunit associations, and fin-gerprinting and identifying variants of storage proteins insoybean cultivars is complex (Krishnan et al., 2005; Yaga-saki, Takagi, Sakai, & Kitamura, 1997; Yaklich, 2001).Because soybean genotypes frequently differ in the numberand properties of their storage protein polypeptides, theextent of this complexity has only recently been revealedby the application of high-resolution two-dimensional(2-DE) sodium dodecyl sulfate–polyacrylamide gel electro-phoresis (SDS–PAGE). The major advantage of using2-DE SDS–PAGE, is that it provides the highest resolutionin analytical separations of several hundred gene products(Agrawal, Yonekura, Iwahashi, Iwahashi, & Rakwal, 2005;Krishnan et al., 2005; Gorg, Weiss, & Dunn, 2004; Yak-lich, 2001). Spot displacements in 2-DE patterns may beattributed to variations in protein coding genes (Fukudaet al., 2005; Hajduch, Ganapathy, Stein, & Thelen, 2005).

To take advantage of the high resolution of 2-DE, a pre-fractionation step is usually carried out to reduce the com-plexity of the sample, and to minimize protein degradationdue to the presence of proteases, and to remove interferingcompounds such as salts, lipids, nucleic acids, polyphenols,alkaloids, pigments, terpenes, organic acids and other com-pounds (Gorg et al., 2004). TCA/acetone extraction is veryuseful for minimizing protein degradation and removinginterfering compounds (Gegenheimer, 1990). The introduc-tion of immobilized pH gradient (IPG) strips for isoelectricfocusing (IEF) followed by 2-DE PAGE has improved theseparation of proteins up to pH 12 (Agrawal et al., 2005;Gorg et al., 2004). This methodology has also been usedin genomic and proteomic analyses (Gorg et al., 2004;Mooney & Thelen, 2004). It is apparent that to identifyall proteins in soybeans, sample solubilization, and frac-tionation according to protein isoelectric point (pI) usingmultiple narrow range pH gradient (IPG) strips is animportant step (Agrawal et al., 2005; Fukuda et al., 2005;Gorg et al., 2004) in developing a data base for assessingsoybean protein quality, and for enhancing variety identifi-cation, genetic analyses, and proteomics (Mooney & The-len, 2004).

The aims of the present study were, first, to develop arapid method for determining tryptophan, and to quantita-tively measure and compare the levels and variation oftotal protein, and the individual amino acids, includingtryptophan and sulfur amino acids, in 14 soybean cultivars;

second, to assess their protein quality from digestibility andamino acid compositional data according to Food andAgricultural Organization/World Health Organization(FAO/WHO, 1991), and the US Food and Drug Adminis-tration (USFDA, 1993); third, to screen for differences ingenetic variability in storage protein subunit compositionamong these genotypes, by high-resolution 2-DE gel elec-trophoresis; and fourth, to assess the effectiveness of thesemethods as a tool for assisting plant breeders in their selec-tion of new high protein quality soybean varieties.

2. Materials and methods

2.1. Materials

The amino acid standards were obtained as follows: nor-leucine from Pierce Chemical Co., Rockford, IL; 3-nitroty-rosine from Aldrich Chemical Co., Milwaukee, WI; andthe standard amino acid calibration mixture from Beck-man Coulter Canada, Inc., Mississauga, ON. Hydrochloricacid (Analar), hydrobromic acid (Aristar), formic acid(88.0%), and hydrogen peroxide (30.0%) were purchasedfrom BDH Inc., Toronto, ON. Carrier ampholytes Biolyte3/10 (pH 3.3–9.4), IPG linear Ready Strips�, pH 4–7, 4.7–5.9 cm, carrier ampholytes Biolyte 3/10 (pH 3.3–9.4), Bro-mophenol Blue, low and broad range protein markers wereobtained from Bio-Rad Laboratories, Richmond, CA.Immobiline DryStrip gels, pH 6–11, were from AmershamBioscience, Piscataway, NJ. Agarose (low melting prepara-tion grade) SeaPlaque� was obtained from BioWhittakerMolecular Applications, Rockland, ME. Acrylamide/bis,N,N 0-methylenebisacrylamide solution, urea, glycine, glyc-erol, sodium dodecyl sulfate (SDS), 3-[(3-cholamidopro-pyl)dimethyl ammonio]-1-propanesulfonate (CHAPS),and Tris (base) were obtained from BioShop CanadaInc., Burlington, ON. N,N,N 0,N 0-tetramethylenediamine(TEMED), 1,4-dithiothreitol (DDT), Brilliant (Coomassie)Blue R-250, iodoacetamide (IA), trichloroacetic acid(TCA), b-mercaptoethanol, and ethylene diamine tetra-acetate (EDTA) were from Sigma–Aldrich Canada Ltd.,Oakville, ON. Thiourea was from J. T. Baker, Phillipsburg,PA.

2.2. Selection of plant materials

The 14 soybean genotypes selected for this investigationoriginated from the Eastern Cereal and Oilseed ResearchCentre (ECORC), Agriculture and Agri-Food Canada,Ottawa, ON, the University of Guelph, and the King AgroInc. Chatham, ON. Their pedigrees are as follows:

The commercial source (King Agro Inc.) lines wereKG20 = McCall/2S11; KG 30 = McCall/Maple Arrow;KG 41 = B220/Maple Arrow; and McCall = M433/Hark.

The University of Guelph lines were OACLibra = FH31-3/Evans; OAC Scorpio = McCall/Bicenten-nial; OAC Frontier = Brand1677/Maple Arrow; andBicentennial = Harosoy 63/Fiskeby V.


The public sector (AAFC) lines were: AC Bra-vor = Maple Arrow/Wayne; AC Harmony = Maple Pre-sto/Williams/Weber; QS 3011.24 black seed coatedgenotype that carries the self-colour seed coat allele; MapleAmber = Amsoy/Portage//840-7-3 and Maple Arrow =Harosoy 63/840-7-3.

One additional sample is Early Hakucho, an unverifiedname for a PI 342437 pure-line genotype. This rather mod-est green-podded edamame soybean variety is consumed asan hors d’oeuvre vegetable soybean in many Asian cul-tures, rather than as a dry seed flour product.

2.3. Experimental procedures for amino acid analyses

Dried seeds of the four replicate samples taken fromeach of the 14 soybean cultivars selected for this investiga-tion were pulverized in a standard electrically driven endrunner mill (Cyclone Sample Mill, U. D. Corp., Fort Col-lins, CO.), passed through a 1.0 mm mesh sieve, lyophi-lized, and then stored at �20 �C in polypropylene bottlesuntil used.

Duplicate soybean samples (0.05 g) were hydrolyzed inPyrex (No. 9860) test tubes (18 cm · 150 mm) under vac-uum (below 10 mm Hg) with triple glass-distilled con-stant-boiling HCl (6.0 M) at 110 ± 0.5 �C for periods of24, 48, 72 and 96 h with the usual precautions describedpreviously (Zarkadas et al., 1993, 1999).

Tryptophan in soybeans was determined separately afteralkaline hydrolysis by the method described by Hugli andMoore (1972). Triplicate samples (0.05 g) were hydrolyzedin 3.5 mL polypropylene (Nalgene, No. 3110-0035) centri-fuge tubes (8 · 1 cm) sealed in Pyrex (No. 9860) test tubes(18 cm · 150 mm) under vacuum (below 10 mm Hg) in1.2 mL of 4.2 M NaOH (sample plus 0.2 mL water plus1.0 mL of 5.0 M NaOH, freshly prepared from 50% NaOHcontaining 5 lL of 1% 1-octanol in toluene), at110 ± 0.5 �C for periods of 24 h with the usual precautionsdescribed earlier (Hugli & Moore, 1972; Zarkadas, Roche-mont, Zarkadas, Karatzas, & Khalili, 1987).

Amino acid analyses were carried out on a BeckmanSystem 6300 fully automated high performance amino acidanalyzer (Beckman Instrument, Inc., San Ramon, CA)using a cation-exchange microcolumn (12 · 0.40 cm), oper-ated with the highly purified Beckman sodium citrate (E, Fand D) buffers. The analyzer was equipped with the 7.1system Gold chromatographic data reduction system forWindows 3.1, which was upgraded with a Dell (PentiumIII-series) compatible personal computer with MicrosoftWindows-98, to improve its sensitivity and data reductioncapability.

Analyses were carried out on each of the three replicatesoybean samples (50.0 mg) in duplicate as described previ-ously (Zarkadas et al., 1987, 1993, 1999). Norleucine wasused as the internal standard. The data reported for serineand threonine in Tables 1 and 2 represent the average val-ues of 48 determinations extrapolated to zero time ofhydrolysis by linear regression analysis of the results. The

values for valine, isoleucine, leucine, and phenylalanineare the average of 36 values obtained from the 48, 72 and96 h of acid hydrolysis. All others are reported as the aver-age values of 48 determinations from 24, 48, 72, and 96 h ofhydrolysis.

Methionine and cyst(e)ine were determined separately ineach cultivar (50.0 mg samples) according to the performicacid procedure of Moore (1963). Norleucine was added inthe hydrolysate as an internal standard. The data were nor-malized relative to alanine, valine, leucine and isoleucinepresent in the sample, and represent the average of 24determinations. The yields obtained following performicacid oxidation were: cysteic acid 105.9% and methionineS,S-dioxide 89.0%.

2.4. Rapid method for determining tryptophan

This paper also describes the development of a rapidand highly sensitive chromatographic method for the deter-mination of tryptophan, using 3-nitrotyrosine as the inter-nal standard. This method is a modification of thetryptophan method reported earlier (Zarkadas et al.,1987), and as shown in Fig. 1. It is designed to be used withthe Beckman Spinco model 7300/6300 amino acid analyzer,using a 12 · 0.4 cm microcolumn, eluted with 0.21 Msodium citrate buffer (pH 5.125) at a flow rate of14.0 mL/h and 52 �C (5425 kN/m2; 150 psi). The ninhydrinreagent was pumped into the effluent stream at 6.9 mL/h.This rapid method has been successfully applied to thedetermination of tryptophan at picomole levels in a varietyof proteins and plant tissues, including soybeans. For ana-lytical work, the microcolumn was overloaded about 10times (12.5–50 lg of sample alkaline hydrolysate), whichyielded picomole levels of tryptophan. They were appliedto the microcolumn using 3-nitrotyrosine as the internalstandard. The data for tryptophan, presented in Tables 1and 2, represent the average of 24 determinations.

2.5. Protein determination

Precise quantitation of the protein mass in each soybeanacid hydrolysate was carried out (Zarkadas et al., 1987,1993, 1999) as follows:

WE ¼X18

i¼1

ðaibiÞ; ð1Þ

where WE (weight equivalent, in micrograms per nano-mole) is the mean residue weight, ai is the mole fractionof an amino acid i found in the analyzed aliquot and bi isthe molecular weight of amino acid residue i (inmicrograms).

P ¼WEX18

i¼1

ðaibiÞ: ð2Þ

Table 1

Comparison of the amino acid (AA) composition and protein contents (grams of AA per kilogram of total protein; meanh ± SEM) of 14 soybean cultivars

Amino acid Source of soybean cultivarsf

Commercial University of Guelph Public sector (AAFC)

KG 20 KG 30 KG 41 OAC Libra OAC Scorpio OAC Frontier AC Harmony AC Bravor QS 3011.2

Aspartic acid 108.55 ± 1.73b,c 110.12 ± 2.37a,b,c 109.01 ± 0.2b,c 109.23 ± 0.70b,c 111.21 ± 0.9a,b,c 109.68 ± 0.56b,c 114.70 ± 0.33a 111.66 ± 0.43a,b 111.57 ± 0.9a,b

Threonine 49.75 ± 0.19c 50.13 ± 0.18a,b,c 50.05 ± 0.6b,c 49.96 ± 0.02b,c 50.27 ± 0.3a,b,c 49.92 ± 0.03a,b,c 51.45 ± 0.9a 51.13 ± 0.33a,b 48.40 ± 2.2a,b,c

Serine 54.05 ± 0.28b 54.58 ± 0.19a,b 54.50 ± 0.07a,b 54.40 ± 0.02a,b 54.73 ± 0.43a,b 54.37 ± 0.04a,b 56.03 ± 3.1a 55.67 ± 0.35a,b 58.98 ± 5.6a,b

Glutamic acid 182.72 ± 2.01a,b 176.06 ± 4.83b,c 179.67 ± 0.62a,b 176.34 ± 0.85b,c 177.39 ± 0.9a,b,c 176.43 ± 1.12b,c 174.79 ± 1.3b,c 176.37 ± 1.48b,c 175.17 ± 0.4a,b

Proline 46.59 ± 0.58a,b,c 45.29 ± 1.36c 46.76 ± 0.81a,b,c 48.92 ± 2.8a,b 49.25 ± 0.61a 46.57 ± 0.61a,b,c 48.21 ± 0.36a,b,c 45.96 ± 0.34c 46.76 ± 1.1a,b

Glycine 35.92 ± 1.18a,b,c 34.43 ± 0.6a,b,c 34.41 ± 0.67a,b,c 36.69 ± 0.22a,b,c 33.59 ± 2.80a,b,c 36.69 ± 029a,b,c 36.91 ± 0.04a,b,c 36.78 ± 0.29a,b,c 36.51 ± 0.9a,b,c

Alanine 39.15 ± 0.15 39.11 ± 0.29 39.03 ± 0.22 39.36 ± 0.20 40.13 ± 0.83 39.71 ± 0341 40.61 ± 0.4 39.75 ± 0.38 39.58 ± 1.0b

Cysteine 21.77 ± 0.81a,b,c 19.92 ± 0.69a,c,d 23.96 ± 1.84a 22.07 ± 0.88a,b 18.88 ± 0.45c,d 21.73 ± 0.32a,b,c 19.82 ± 085b,c,d 20.66 ± 1.55b,c,d 20.87 ± 0.5b,c,d

Valine 51.09 ± 1.21a,b 50.56 ± 0.50a,b 51.81 ± 0.72a,b 53.89 ± 0.32a 52.56 ± 0.14a,b 53.83 ± 0.33a 52.33 ± 0.8a,b 53.32 ± 0.55a 51.53 ± 1.5b

Methionine 20.96 ± 0.81a,b,c,d,e 20.07 ± 0.71a,b,c,d 21.57 ± 0.41a,b 22.22 ± 0.85a 19.43 ± 0.97c,d,e 21.63 ± 0.35a,b 20.98 ± 0.2a,b,c,d,e 20.07 ± 0.5a,c 20.92 ± 0.7a,b,c

Isoleucine 49.31 ± 0.05a,b 47.78 ± 0.43a,b,c 48.79 ± 0.59a,b 49.11 ± 0.21a,b 48.82 ± 0.7a,b 49.81 ± 0.51a 48.82 ± 0.37a,b 49.32 ± 0.41a 47.70 ± 1.1e

Leucine 78.53 ± 0.14a,b 77.31 ± 0.4a,b,c 78.16 ± 0.30a,b 78.51 ± 0.45a,b 75.23 ± 0.9a 78.81 ± 0.52a 77.25 ± 017a,b,c 78.89 ± 0.55a 75.94 ± 1.7c

Tyrosine 38.30 ± 0.21a,b 40.02 ± 0.76a 39.79 ± 0.37a 38.76 ± 0.42a,b 38.47 ± 0.7a,b 38.90 ± 0.27a,b 38.93 ± 0.46a,b 39.37 ± 0.91a,b 38.35 ± 1.2a,b

Phenylalanine 51.08 ± 1.10 51.23 ± 1.21 52.70 ± 1.58 54.44 ± 0.16 54.45 ± .8a 54.90 ± 0.34 52.96 ± 0.18 53.13 ± 0.55 52.24 ± 1.56

Histidine 28.22 ± 0.31a,b 27.09 ± 0.31a,b,c 26.28 ± 0.82b,c 28.10 ± 0.47a,b 28.92 ± 0.2a 27.12 ± 033a,b,c 27.01 ± 0.19a,b,c 28.11 ± 0.4a,b 26.75 ± 0.4a,b,c

Lysine 63.82 ± 1.37a,b 63.71 ± 0.87a,b 64.89 ± 0.28a,b 64.83 ± 0.74a,b 66.52 ± 1.68a 63.66 ± 043a,b,c 63.83 ± 0.23a,b 65.49 ± 2.24a,b 63.83 ± 1.2b,c

Arginine 62.13 ± 5.4c 71.65 ± 0.61a,b 74.51 ± 0.81a,b 74.73 ± 0.6a 75.57 ± 1.02a,b 75.17 ± 0.91a,b 71.57 ± 0.67a,b 77.41 ± 3.15a 77.87 ± 1.8a

Tryptophan 13.90 ± 0.63a,b,c 11.62 ± 0.47c,d 14.12 ± 0.98a,b 13.45 ± 0.84a,b,c,d 11.92 ± 1.13c,d 14.41 ± 057a 11.75 ± 0.77c,d 13.69 ± 084a,c,d 11.51 ± 2.3d

Ammonia 16.22 ± 2.14b,c 20.07 ± 0.71e 10.09 ± 1.36e 11.54 ± 1.99d,e 13.20 ± 1.35c,d 13.20 ± 1.16d,e 19.73 ± 0.02a 20.07 ± 0.59a 19.27 ± 1.3a

Basich 161.11 ± 5.36 171.17 ± 1.76 171.17 ± 1.68 162.46 ± 2.41 164.11 ± 3.68 169.64 ± 1.51 154.18 ± 0.30 168.64 ± 6.50 168.64 ± 1.02

Acidich 308.33 ± 2.53 291.25 ± 6.57 291.25 ± 0.57 284.52 ± 1.54 284.04 ± 1.91 291.28 ± 1.66 304.40 ± 1.44 283.97 ± 1.87 294.28 ± 4.23

Chargedh 469.43 ± 7.79 462.42 ± 7.25 462.42 ± 2.24 446.98 ± 2.82 448.16 ± 4.24 460.93 ± 1.65 458.58 ± 1.21 452.62 ± 7.30 461 ± 3.71

Hydrophilich 558.79 ± 8.09 556.56 ± 6.91 556.56 ± 2.36 536.04 ± 2.81 553.97 ± 4.24 556.49 ± 1.72 549.15 ± 1.21 565.85 ± 6.63 567.75 ± 3.77

Hydrophobich 306.33 ± 2.08 305.14 ± 3.29 305.14 ± 3.70 297.18 ± 1.24 299.69 ± 2.41 305.96 ± 1.35 308.69 ± 0.96 303.74 ± 1.34 284.77 ± 11.22

Apolarh 266.36 ± 1.99 265.34 ± 3.49 265.34 ± 3.74 258.22 ± 1.14 261.22 ± 2.37 266.91 ± 1.59 269.69 ± 0.85 265.39 ± 0.75 247.39 ± 10.28

WEi 0.11286 ± 3 · 10�4 0.11276 ± 3 · 10�4 0.11337 1 · 10�4 0.11360 ± 3 · 10�4 0.11351 ± 3 · 10�4 0.11338 ± 1 · 10�4 0.11285 ± 1 · 10�4 0.11321 ± 1 · 10�4 0.11371 ± 2 · 10�4

CFi 0.11576 ± 3 · 10�4 0.11529 ± 4 · 10�4 0.11622 1 · 10�4 0.11651 ± 4 · 10�4 0.11603 ± 4 · 10�4 0.11629 ± 2 · 10�4 0.11547 ± 1 · 10�4 0.11659 ± 7 · 10�4 0.11672 ± 7 · 10�4

Total proteini g/kg

dry matter

332.31 ± 5.7b,c 331.01 ± 9.5b,c 313.13 ± 4.7d,e,f,g 300.10 ± 7.9f,g 321.70 ± 3.8c,d,e 303.06 ± 4.3e,f,g 297.81 ± 2.3g 310.11 ± 3.2e,f,g 342.73 ± 3.2e,f,g

Amino acid Soybean cultivarsf Edamame Significant levels among cultivarsg Orthogonal polynomial contrastsg

Overall Public sector vs.

others

Commercial vs.

University

Maple Amber Maple Arrow Bicentennial McCall Early Hakucho Mean CV F F

Aspartic acid 115.16 111.38 ± 2.29b,a,c 106.44 ± 4.01c 108.13 ± 2.4b,c 111.54 ± 0.38a,b 110.64 3.14 13.19*** 0.72ns

Threonine 42.89 46.12 ± 1.49d 49.87 ± 0.13b,c 50.07 ± 0.1b,c 50.97 ± 0.34a,b,c 50.41 1.88 54.91*** 0.16ns

Serine 46.50 55.72 ± 2.21b,c 54.29 ± 0.14a,b 54.52 ± 0.19a,b 55.24 ± 0.45a,b 54.84 2.37 39.02*** 0.29ns

Glutamic acid 193.16 186.89 ± 4.87a 169.01 ± 6.35c 176.36 ± 2.3b,c 182.71 ± 3.37a,b,c 178.05 3.31 0.01ns 1.83ns

Proline 50.47 48.46 ± 1.01a,b,c 47.55 ± 2.69a,b,c 46.27 ± 0.41a,b,c 48.78 ± 0.89a,b,c 47.37 4.82 0.40ns 7.21**

Glycine 37.63 37.12 ± 0.54a,b,c 34.35 ± 1.52a,b,c 36.52 ± 0.341a,b,c 37.16 ± 0.64a,b,c 35.84 3.45 4.97* 0.75ns

Alanine 40.91 40.61 ± 0.53 36.50 ± 1.54b 38.96 ± 0.86 36.83 ± 1.12 39.19 3.45 2.02ns 1.42ns

Cysteine 20.85 22.2 ± 0.8a,b 20.96 ± 1.92a,b,c,d 18.49 ± 0.44d 19.55 ± 0.70b,c,d 20.93 10.30 3.10ns 1.19ns

Valine 51.14 52.33 ± 097a,b 51.64 ± 3.36a,b 51.91 ± 2.45a,b 49.36 ± 2.07a,b 52.08 5.32 0.72ns 7.35**

(continued on next page)

C.G

.Z

ark

ad

as

eta

l./

Fo

od

Resea

rchIn

terna

tiona

l4

0(

20

07

)1

29

–1

46

133

Table 1 (continued)

Amino acid Soybean cultivarsf Edamame Significant levels among cultivarsg Orthogonal polynomial contrastsg


others

Commercial vs.

University

Maple Amber Maple Arrow Bicentennial McCall Early Hakucho Mean CV F F

Methionine 21.30 21.49 ± 0.80a,b,c 16.14 ± 0.73f 19.38 ± 0.47d,e 19.27 ± 0.54a,b,c,d 20.92 7.01 0.07ns 0.10ns

Isoleucine 50.36 50.36 ± 0.85a 45.76 ± 3.38b,c 47.16 ± 0.5a,b,c 44.73 ± 2.75a,b,c 48.50 5.70 2.80ns 0.47ns

Leucine 79.82 80.67 ± 0.52a 73.09 ± 5.21b,c 75.77 ± 2.75a,b,c 71.75 ± 3.68a,b,c 77.47 5.12 4.67* 0.13ns

Tyrosine 39.9 38.33 ± 0.38a 36.35 ± 1.87a,b 37.93 ± 0.87a,b 37.14 ± 0.65b 38.84 13.76 1.31ns 1.59ns

Phenylalanine 51.59 53.21 ± 1.33a,b 54.65 ± 3.46a,b 52.21 ± 1.93a,b 51.21 ± 1.07b 52.85 5.31 0.95ns 14.3***

Histidine 28.21 28.03 ± 0.47a,b 25.70 ± 1.11c 26.15 ± 0.53a 27.24 ± 0.35a 27.57 5.59 0.13ns 2.44ns

Amino acid Recurrent parents Edamame Significant levels among

cultivarsgOrthogonal polynomial

contrastsg


others

Commercial vs.

University

Maple Amberj Maple Arrow Bicentennial McCall Early Hakucho Mean CV F F

Lysine 66.56 ± 1.0 63.12 ± 1.3a,b,c 59.98 ± 2.64c 62.12 ± 1.32b,c 62.12 ± 0.71a,b 64.32 4.08 0.88ns 0.87ns

Arginine 66.33 ± 6.3 73.75 ± 1.83a,b 69.42 ± 04.16b 70.34 ± 1.25b 77.87 ± 0.75a,b 73.41 6.46 2.78ns 6.38ns

Tryptophan 12.14 ± 0.3 13.61 ± 0.88a,b,c,d 13.43 ± 0.39a,b,c,d 14.46 ± 1.35a 11.29 ± 0.51a,b,c,d 12.91 12.51 2.50ns 0.00ns

Ammonia 13.12 ± 4.6 19.04 ± 1.6a,b 19.56 ± 0.15a,b 19.64 ± 0.07a 9.62 ± 0.02a 14.36 14.87 48.05*** 0.03ns

Basicj 161.11 ± 161.11 ± 1.04 171.17 ± 7.7 158.60 ± 3.01 168.64 ± 1.97 165.31 4.62 0.89ns 6.68**

Acidicj 308.33 ± 4.6 308.33 ± 6.14 291.25 ± 10.36 291.25 ± 4.68 283.97 ± 1.04 288.69 3.02 1.77ns 0.61

Chargedj 469.43 ± 3. 469.43 ± 5.86 462.42 ± 17.74 462.42 ± 7.69 452.62 ± 3.23 454.01 3.14 2.25ns 1.89ns

Hydrophilicj 558.79 ± 3.67 558.79 ± 4.68 556.56 ± 17.51 556.56 ± 7.38 560.02 ± 3.68 559.25 2.51 4.98ns 2.11ns

Hydrophobichhh 306.33 ± 10.14 306.33 ± 1.85 305.14 ± 16.74 305.14 ± 10.92 303.74 ± 11.5 303.58 4.87 2.87ns 1.63ns

Apolarj 266.36 ± 3.56 266.36 ± 2.16 265.34 ± 15.17 265.34 ± 10.14 265.39 ± 3.68 264.74 4.62 2.76ns 2.36ns

CFi 0.11632 ± 3 · 10�4 0.11632 ± 3 · 10�4 0.116607 ± 2 · 10�4 0.11603 ± 3 · 10�4 0.11661 ± 5 · 10�4 0.11609 0.76 0.44ns 2.02n

WEi 0.1136 ± 3 · 10�4 0.11286 ± 2 · 10�4 0.11421 ± 2 · 10�4 0.11351 ± 9 · 10�4 0.11336 ± 2 · 10�4 0.113215 0.78 0.44ns 6.44**

Total proteinj g/kg

dry matter

321.91 ± 3.52 318.36 ± 11.6c,d,e,f 360.67 ± 6.23a 325.23 ± 3.5b,c,d,e 342.73 ± 3.3e,f,g 317.61 5.48 0.03ns 4.47*

a–eMeans along a row with different superscripts are significantly different (Duncan, 1955).

fMean values and standard error of measurements (SEM) for four replicates (N = 4) and 64 determinations. The values for cyst(e)ine, methionine and tryptophan represent the average of 24 determinations.

g Significance: F, values from analysis of variance among cultivars, ***P < 0.001; **P < 0.01; *P < 0.05; ns, not significant; CV, coefficient of variation.h Calculated according to Zarkadas et al. (1993).i Computed according to Zarkadas et al. (1993, 1999) using Eqs. (1) and (2).

134C

.G.

Za

rka

da

set

al.

/F

oo

dR

esearch

Intern

atio

na

l4

0(

20

07

)1

29

–1

46

able 2omparison of the amino acid (AA) composition and protein contents (grams of AA per 16 gram of total nitrogen; mean ± SEMf) of 14 commercial soybean cultivars

mino acid Source of soybean cultivars

Commercial University of Guelph Public sector (AAFC)


spartic acid 10.02 ± 0.01c,d 10.17 ± 0.02c,d 10.03 ± 0.03c,d 10.00 ± 0.05c,d 10.97 ± 0.07a 10.09 ± 0.04c,d 10.61 ± 0.03a,b,c 10.19 ± 0.21b,c,d 9.57 ± 0.07d

hreonine 4.58 ± 0.02 4.63 ± 0.02 4.61 ± 0.003 4.59 ± 0.004 4.746 ± 0.15 5.87 ± 0.02a 4.65 ± 0.003 4.67 ± 0.08 4.48 ± 2.2erine 4.99 ± 0.04a,b 5.04 ± 0.02a,b 5.01 ± 0.004a,b 5.00 ± 0.004a,b 5.171 ± 0.6a 5.00 ± 0.001a,b 5.06 ± 0.004a,b 5.08 ± 0.09a,b 15.78 ± 0.04a,b

lutamic acid 16.87 ± 0.04a 16.26 ± 0.04a 16.49 ± 0.02a 16.23 ± 0.08a 15.73 ± .01a 16.23 ± 0.09a 16.16 ± 0.13a 16.29 ± 0.27a 15.78 ± 0.04a

roline 4.30 ± 0.05b 4.18 ± 0.02b 4.37 ± 0.01b 4.49 ± 0.07b 4.64 ± 0.03b 4.06 ± 0.03b 4.45 ± 0.01,b 4.33 ± 0.07b 4.20 ± 0.018b

lycine 3.31 ± 0.01a,b 3.18 ± 0.06a,b 3.29 ± 0.05a,b 3.32 ± 0.04a,b 3.12 ± 0.02,b 3.52 ± 0.09a,b 3.41 ± 0.007a,b 3.34 ± 0.11a,b 36.51 ± 0.9a,b

lanine 3.56 ± 0.04a,b,c,d 3.51 ± 0.07c,d,e 3.59 ± 0.02b,c,d 3.63 ± 0.02b,c,d 3.76 ± 0.08a,b 3.68 ± 0.01a,b,c 3.67 ± 0.01a,b,c 3.71 ± 0.06a,b,c 3.19 ± 0.06e

ysteine 1.92 ± 0.01a,b 1.76 ± 0.02b,c,d 1.93 ± 0.03a,b 1.85 ± 0.01a,b 1.86 ± 0.04a,b 1.47 ± 0.02,c 1.82 ± 0.07a,b 1.88 ± 0.01a,b 1.81 ± 0.04a,b

aline 5.15 ± 0.06a 4.51 ± 0.08a,b 4.64 ± 0.06a,b 4.78 ± 0.03a,b 5.15 ± 0.06a 4.83 ± 0.01a,b 4.73 ± 0.03a,b 4.55 ± 0.02a,b 4.56 ± 0.04a,b

ethionine 1.83 ± .0.05a,b,c,d 1.74 ± 0.03d 1.745 ± 0.08d 1.77 ± 0.04c,d 1.77 ± 0.07c,d 1.81 ± 0.04a,b,c,d 1.94 ± 0.02a,b,c 1.96 ± 0.01a,b 1.73 ± 0.09d

soleucine 4.51 ± 0.09a,b,c 4.11 ± 0.06a,b,c 4.37 ± 0.03b,c,d 4.38 ± 0.02b,c,d 4.56 ± 0.02a,b 4.47 ± 0.02a,b,c 4.01 ± 0.03b,c,d 4.32 ± 0.07b,cd 4.23 ± 0.03b,c,d

eucine 6.90 ± 0.08b,c 6.66 ± 0.02a,b,c 6.99 ± 0.02b,c 6.99 ± 0.04a,b,c 6.56 ± 0.09a,b,c 7.08 ± 0.02a 6.98 ± 0.01b,c 7.03 ± 0.04b,c 6.71 ± 0.05b,c

yrosine 3.54 ± 0.02a,b 3.46 ± 0.01b 3.63 ± 0.02a,b 3.62 ± 0.02a,b 3.77 ± 0.03c 3.58 ± 0.02a,b 3.66 ± 0.04a,b 3.59 ± 0.05a,b 3.49 ± 0.01a,b

henylalanine 4.68 ± 0.05b,c 4.53 ± 0.04b,c 4.72 ± 0.04b,c 4.84 ± 0.011a,b,c 5.28 ± 004a 4.93 ± 0.001a,b,c 4.788 ± 0.01b,c 4.51 ± 0.02a,b,c 4.66 ± 0.02b,c

istidine 2.57 ± 0.03b 2.32 ± 0.08b 2.52 ± 0.007b 2.48 ± 0.02b 2.32 ± 0.05a 2.52 ± 0.01b 2.49 ± 0.02b 2.56 ± 0.07b 2.56 ± 0.07b

ysine 5.89 ± 0.2a,b,c 5.46 ± 0.08d 5.83 ± 0.01a,b,c,d 5.84 ± 0.04a,b,d 5.51 ± 0.03a,b,c 5.79 ± 0.01a,b,c,d 5.90 ± 0.02a,b,c 5.98 ± 0.01a,b 5.53 ± 2.2c,d

rginine 6.87 ± 0.08a 6.51 ± 0.06a 7.04 ± 0.08a 6.95 ± 0.04a 6.57 ± 0.04a 6.96 ± 0.09a 6.62 ± 0.05a 7.07 ± 0.02a 7.07 ± 0.04a

ryptophan 1.29 ± 0.03a,b 1.20 ± 0.07a,b,c,d 1.23 ± 0.07a,b,c,d 1.19 ± 0.01a,b,c,d 1.03 ± 0.07d 1.06 ± 0.02c,d 1.08 ± 0.07b,cd 1.25 ± 0.08a,b,c 1.02 ± 0.06d

mmonia 1.80 ± 0.007a,b 1.81 ± 0.01a,b 1.81 ± 0.01a,b 1.80 ± 0.001a,b 1.86 ± 0.06a 1.80 ± 0.00a,b 1.82 ± 0.001a,b 1.83 ± 0.03c 1.75 ± 0.004b

otal AANk

gAAN/kgof protein

173.79 ± 0.52a,b 172.87 ± 0.16b,c 173.89 ± 0.21a,b,c 173.59 ± 0.35b,c 171.72 ± 2.18b,c 173.88 ± 0.09a,b 173.04 ± 0.13b,c 175.05 ± 1.86a,b 179.01 ± 5.25a

AA/16 g ofN

92.19 ± 0.24a,b,c 92.55 ± 0.08a,b,c 92. 01 ± 0.11a,b,c 92.04 ± 0.058a,b,c 92.99 ± 0.99b,c 92.015 ± 0.05a,b,c 92.46 ± 0.07a,b,c 91.44 ± 0.96a,b 89.60 ± 2.48a

omparison of the amino acid (AA) composition and protein contents (grams of AA per kilogram of total protein; mean ± SEMf) of 12 commercial soybean cultivars, a black seed coated cultivar and an Edamamevariety

mino acid Soybean cultivars developed by Significant levels among cultivarsg Orthogonal polynomial contrastsg

Recurrent parents Edamame Overall Public sector vs.others

Commercial vs.University

Maple Arrow Biocentennial McCall Maple Amberi Early Hakucho Mean CV F F

spartic acid 10.96 ± 0.05a,b 9.99 ± 0.24c,d 10.00 ± 0.02c,d 11.11 10.27 ± 0.01a,b,c 10.22 5.28 0.15ns 1.07ns

hreonine 4.04 ± 0.3b 1 4.68 ± 0.08a 4.63 ± 0.01a,b 4.66 4.59 ± 0.01a,b 4.76 15.97 0.62ns 1.79ns

erine 4.59 ± 0.05a,b,c 5.10 ± 0.09a,b,c 5.04 ± 0.02a,b,c 50.1 5.01 ± 0.01a,b,c 5.03 7.03 0.24ns 0.33ns

lutamic acid 17.97 ± 0.08a 11.85 ± 0.39b 16.07 ± 0.04a 18.63 16.40 ± 0.08a,b,c 16.03 14.2 0.71ns 2.50ns

roline 4.77 ± 0.02b 7.29 ± 0.29a 4.22 ± 0.05bc 4.86 4.28 ± 0.015a,b,c 4.58 36.3 0.02ns 0.98ns

lycine 3.54 ± 0.02a 3.57 ± 0.38a 3.35 ± 0.08a,b 3.63 3.39 ± 0.12a,b,c 3.35 8.72 0.10ns 0.36ns

lanine 3.86 ± 0.03b,c,d,e 3.42 ± 0.09d,e 3.6 ± 0.08b,c,d 3.95 3.64 ± 0.03a,b,c,d 3.61 4.42 0.47ns 3.12ns

ysteine 2.07 ± 0.06a 1.936 ± 0.17a,b 11.70 ± 0.03b,c 2.01 1.86 ± 0.05a,b,c 1.83 12.39 0.25ns 1.87ns

aline 4.94 ± 0.06ac 4.73 ± 0.24c,d 54.78 ± 0.07a,b,c,d 4.94 4.73 ± 0.012a,b,c 4.77 9.15 1.83ns 0.62ns

(continued on next page)

C.G

.Z

ark

ad

as

eta

l./

Fo

od

Resea

rchIn

terna

tiona

l4

0(

20

07

)1

29

–1

46

135

TC

A

ATSGPGACVMILTPHLATAT

g

C

A

ATSGPGACV

Table 2 (continued)

Amino acid Soybean cultivars developed by Significant levels among cultivarsg Orthogonal polynomial contrastsg

Recurrent parents Edamame Overall Public sector vs.others

Commercial vs.University

Maple Arrow Biocentennial McCall Maple Amberi Early Hakucho Mean CV F F

Methionine 1.81 ± 0.08a 1.51 ± 0.07e 11.79 ± 0.04c,d 2.05 1.97 ± 0.05a,b 1.81 7.27 4.12* 0.03ns

Isoleucine 4.79 ± 0.06a,b,c 4.18 ± 0.25c,d 4.26 ± 0.05b,c,d 4.86 4.34 ± 0.06b,c,d 4.37 5.26 1.12ns 2.81ns

Leucine 7.59 ± 0.02a 6.69 ± 0.38b,c 7.07 ± 0.0.02c 7.69 6.88 ± 0.08b,c 6.94 4.99 0.13ns 0.04ns

Tyrosine 3.68 ± 0.04b,c 3.41 ± 0.12b 3.55 ± 0.09a,b 4.98 3.57 ± 0.07a,b 3.59 5.47 0.03ns 2.00ns

Phenylalanine 4.981 ± 0.03a,b,c 5.01 ± 0.25a,b 4.70 ± 0.06b,c 4.98 4.66 ± 0.014b,c 4.79 7.13 2.27ns 7.33ns

Histidine 2.68 ± 0.01b 2.41 ± 0.07b 2.42 ± 0. 2.72 2.56 ± 0.12b 2.88 2.70 0.43ns 1.52ns

Lysine 6.07 ± 0.05a 5.62 ± 0.16c,d 5.75 ± 0.02a,b,c,d 6.42 2.557 ± 0.02a,b,c 5.77 4.98 0.48ns 0.01ns

Arginine 5.19 ± 0.09b 6.51 ± 0.29a 6.50 ± 0.08a 6.42 5.91 ± 0.01a 6.67 6.19 0.43ns 0.02ns

Tryptophan 1.32 ± 0.05a 1.26 ± 0.03a,b,c 1.32 ± 0.03a 1.17 6.85 ± 0.03a,b,c,d 1.18 12.77 0.83ns 5.32*

Ammonia 1.81 ± 0.11a,b 1.84 ± 0.03a 1.82 ± 0.005a,b 1.25 1.13 ± 0.06a,b,c 1.81 4.32 0.20ns 0.41ns

Total AANk

gAAN/kgof protein

168.27 ± 1.20c 170.32 ± 2.71b,c 172.95 ± 0.12b,c 164.94 173.71 ± 0.47b,c 169.89 2.76 2.94ns 0.08ns

gAA/16 g ofN

95.07 ± 0.65a 94.02 ± 1.53a,b 92.51 ± 0.06a,b 88.98 92.11 ± 0.25a,b,c 92.38 6.28 2.94ns 0.02ns

a–e Means along a row with different superscripts are significantly different (Duncan, 1955).f Mean values and standard error of measurements (SEM) for four replicates (N = 4) and 64 determinations. The values for cyst(e)ine, methionine and tryptophan represent the average of 24 determinations.g Significance: F, values from analysis of variance among cultivars, ***P < 0.001; **P < 0.01; *P < 0.05; ns, not significant; CV, coefficient of variation.h Calculated as described previously (Zarkadas et al., 1993, 1999).i Taken from Zarkadas et al. (1999).

136C

.G.

Za

rka

da

set

al.

/F

oo

dR

esearch

Intern

atio

na

l4

0(

20

07

)1

29

–1

46

Fig. 1. Chromatographic separation of the aromatic amino acids tyrosine and phenylalanine from 3-nitrotyrosine and tryptophan on an analytical12 · 0.4 cm microcolumn eluted with 0.21 M sodium citrate buffer (pH 5.125) at a flow rate of 14.0 mL/h and 52 �C: (a) separation of synthetic amino acidcalibration mixture; (b) typical separation of a 24 h alkaline hydrolysate of a soybean sample, i.e., cv Maple Arrow. The curve shows absorbance at570 nm. Peak identification: Tyr, tyrosine; Phe, phenylalanine; Tyr(NO2), 3-nitrotyrosine (internal standard); Trp, tryptophan.


The protein content P (in micrograms) of each samplewas calculated by multiplying WE by the nanomoles oftotal amino acids found, as described previously (Zarkadaset al., 1993, 1999). Proteins for electrophoresis wereassayed according to Bradford (1976).

2.6. Statistical analysis

Data processing of the results was carried out by anEXCEL version 5 for Windows-XP spread-sheet computerprogram developed for this purpose. Analysis of variance,conducted on the amino acid data, for a completely ran-domized block design (factorial) was done by the generallinear model procedure using SAS statistical softwareunder the Windows operating system, release 8.2 (SASInstitute 2001), and represents the average values fromeight sub-samples per genotype.

2.7. One-dimensional electrophoresis (1-DE)

Dry soybean seed meals (100 mg) were suspended in2.0 mL of prechilled Tris buffer (50 mM Tris–HCl, pH7.6; 1 mM DDT, 150 mM NaCl, and 1 mM EDTA), for

2 h (20 �C) in an orbital (nutator) shaker, and centrifugedfor 10 min, at 35,500g and 20 �C (JA-20 Beckman rotor,19,000 rpm). The supernatant was diluted with an SDS-sample buffer at a final concentration of 60 mM Tris–HCl, pH 6.8, 2% w/v SDS, 3.33% v/v b-mercaptoethanol,10% glycerol and a trace of bromophenol blue, heated at98 �C for 5 min, and centrifuged at 14,500g for 1 min priorto application.

Sodium dodecyl sulfate–polyacrylamide gel electropho-resis (1-DE SDS–PAGE) was performed on a vertical slabgel (160 · 200 · 1 mm; Model Protean II xi Cell) accordingto Laemmli (1970), with an 8–18% gradient separating(acrylamide to bis acrylamide ratio 29:1) gels and a 4%(w/v) stacking gel. Aliquots (50–100 lL) containing 50–75 lg of protein, were applied onto each of the 14 wellsused for samples and one outside well was used for a stan-dard. 1-DE SDS–PAGE was performed at constant volt-age, first at 100 V for 1 h, followed by 125 V for about14 h. At the end of the run, the gels were stained overnightwith Brilliant (Coomassie) Blue R-250 (0.05%. w/v) inmethanol–acetic acid–water (25:10:65 v/v/v), destainedovernight with methanol–acetic acid–water (25:7:68 v/v/v), and washed in deionized water. Detection of protein


bands were performed on 16-bit TIFF images acquiredwith a scanning densitometer (ScanWizard Pro scanner,Microtek Lab, Inc., Carson CA), and dried between twosheets BioDeisgn GelWrap� drying film (BioDesign Inc.,New York, NY).

2.8. Two-dimensional electrophoresis (2-DE)

The soybean seed powders (5.0 g) used for sodium dode-cyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) were first extracted by the TCA–acetone procedure(Gegenheimer, 1990; Granier, 1988; Natarajan, Xu,Caperna, & Garrett, 2005; Wang, Vignani, Scali, & Cresti,2006), using 50 mL solvent (10% TCA, 0.05% DDT in coldacetone). The mixture was kept for 1 h at �20 �C, and thencentrifuged (Model J2-21 Beckman centrifuge) for 30 minat 43,700g (JA-20 Beckman rotor; 19,000 rpm). The pelletwas washed three times, each for 1 h at �20 �C with50 mL acetone containing 0.05% (w/v) DTT, dried in a des-iccator under vacuum, ground and stored at �20 �C, untilused.

Total protein was solubilized from the dry pellet (0.5–1.0 g) in 1.5 mL IEF rehydration urea thiourea buffer con-taining 8.0 M deionized urea, 2.0 M deionized thiourea, 2%3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulf-onate (CHAPS), 0.05 M DDT, 0.2% Ampholytes pH 3–10 (Bio-Rad, Inc.), and a trace (0.001%) of BromophenolBlue, and kept for 1 h at 20�C on a nutator with constantshaking. Insoluble matter was removed by centrifugation(Model 65 Beckman ultracentrifuge) at 250,000g

(55,000 rpm; 75 Ti Beckman rotor), and the supernatantwas removed and placed into a separate Eppendorf centri-fuge tube and stored at �20 �C, until used.

Isoelectric focusing (IEF) using IPG strips was per-formed on 7 cm pH 4–7, or 4.7–5.9 linear IPG ReadyStrips(Bio-Rad Hercules, CA) and on 7 cm pH 6–11 linear IPGImmobiline DryStrip gels (Amersham Bioscience, Piscata-way, NJ) as recommended by Gorg et al. (2004) and Agra-wal et al. (2005). The strips were rehydrated in a plastictray for 1 h in 125 lL rehydration urea–thiourea buffercontaining 50–100 lg of protein. They were then coveredwith mineral oil, and allowed to rehydrate overnight(16–18 h). Strips were then focused at 250 V for 20 min,4000 V for 2 h and continued at 4000 V to reach a totalof 14,000 V h at 20 �C (Herman, Helm, Jung, & Kinney,2003), using the protean IEF electrophoretic chamber(Bio-Rad).

The IPG strips were then equilibrated in a 2 mL equili-bration buffer I consisting of 0.375 M Tris–HCl (pH 8.8),6.0 M deionized urea, 20% (w/v) glycerol, 2% (w/v) SDSand 2% (w/v) DTT for 10 min with gentle agitation at roomtemperature (Mooney & Thelen, 2004). The sulphydrylgroups formed were subsequently alkylated for 10 min at20 �C in a 2.0 mL equilibration Tris-base buffer II (pH8.8) as before containing 2.5% iodoacetamide. After equili-bration, each IPG strip was rinsed briefly with SDS–PAGErunning buffer (Gorg et al., 2004), and then embedded on

top of a 12.5% SDS–PAGE (84 · 74 · 1 mm, Model MiniProtean II) gel. The strips were overlayed with 0.5% (w/v)molten agarose containing 0.025 M Tris-base (pH 8.5),0.192 M glycine, 0.1% SDS, and a trace (0.001%, w/v) ofBromophenol Blue, to track the ion front during the run,and allowed to cool for 5 min. The second dimension sepa-ration was carried out at 100 V for approximately 2.0 h,until the Bromophenol Blue dye front reached the end ofthe gel. The 2-DE gels were stained for at least 12 h withBrilliant (Coomassie) Blue R 250 (0.05% w/v) in metha-nol–acetic acid–water (25:10:65 v/v/v), destained withmethanol–acetic acid–water (25:10:65 v/v/v), washed withdeionized water overnight, scanned as before, and driedbetween two sheets Bio-Design GelWrap� drying film (Bio-Design Inc., New York, NY).

2.9. Image analysis

Electronic image acquisition of the gels (84 · 74 · 1 mm)was carried out with an AGFA ‘‘Snapscan 50’’ scanningdensitometer (AGFA-Gevaert, N.V. Mortsel, Germany)with a resolution of 300 dpi and 16-bit TIFF grayscalepixel depth. Image analysis was performed using the Pho-retic 2-DE Advanced Expression (version 2005) gel analysissoftware (Nonlinear Dynamics, Newcastle, UK). Thisinvolves determining the number of spots (proteins)observed by the Phoretic 2-DE Advanced Expression soft-ware. The volume of each spot was expressed as relativevolume, which is the ratio of a given individual spot vol-ume to the sum of spot volumes for all analyzed spots,excluding from the analysis the volumes of the spots whichwere not considered (Hajduch et al., 2005; Mooney & The-len, 2004). A computer generated reference gel was used tomatch automatically the spots from individual gels. Back-ground subtraction on detected spots was performed usingthe mode-of-non-spot function. Each of the acidic proteincomponents on the 2-DE maps (pH 4.7–5.8) was matchedindividualy to the reference gel, as recommended by Haj-duch et al. (2005), and matched spots were grouped as fol-lows: for b-conglycinin, the percentages of the relativevolumes of a, a 0, and b subunits were added together; totalacidic peptide chains consists of the sum of the percentagesof the relative volumes of A3, A4, the sum of A1a, A1b andA2 protein spots. Similarly, each of the basic spots in the 2-DE maps (pH 6–11) was matched individually, andmatched spots were grouped as the sum of the B1a, B1b,B2, B3 and B4 polypeptide chains. For direct comparisonbetween the pH from 4.7 to 5.9 and from pH 6 to 11 datasets, normaliztion was carried out as recommended by Haj-duch et al. (2005). A simplified method, for calculating thesum of the relative volumes of the basic polypeptide chainof glycinin, i.e., pH 4.7–5.9 data set, was to multiply thetotal volume of the acidic polypeptide chains of glycinin,by the mean molecular mass (Mr) ratio of the basic(Mr � 21 kDa) to the acidic (Mr � 35 kDa) polypeptidecomponents of glycinin, and the results obtained gave an


estimate of the total relative volume for the basic polypep-tide chains of glycinin.

3. Results and discussion

3.1. Protein, nitrogen, and amino acid composition of

selected soybean cultivars

Results of the amino acid analyses carried out on 14 typ-ical soybean cultivars, and levels of statistical significanceobtained from analysis of variance, are summarized inTables 1–3. This study included the development of a rapidmethod for the accurate determination of tryptophan, andthe results are illustrated in Fig. 1. The data are expressedas grams of amino acids per kilogram of anhydrous fat-and ash-free protein and are compared with reference aminoacid patterns for humans recommended by FAO/WHO(1991) and USFDA (1993). Knowing the amino acid compo-sition of the protein or protein mixture enables accurate esti-mates of the amount of protein present (Eqs. (1) and (2)).The constants, weight equivalent (WE, lg/nmol) and con-version factor (CF, lg/nmol) (Zarkadas et al., 1993, 1999)for each of the soybean cultivars presented in Table 1 canbe used in all subsequent quantitations of these cultivars.

The data in Tables 1 and 3 indicate that the protein con-tent among these soybean cultivars differs significantly(P < 0.05). For example, Bicentennial, QS 3011.2, and theedamame Early Hakucho genotypes were consistentlyhigher in total protein (34.3–36.1%) than any of the other12 soybean cultivars, which varied from 29.8% to 30.3%(AC Harmony, OAC Libra, OAC Frontier), 32.5% (KG41, AC Bravor, Maple Arrow, McCall, and Maple Amber)to 33.2% (for KG20 and KG 30). These results are inaccord with those of Zarkadas et al. (1993, 1999), whoreported soybean values ranging from 33.67% to 42.11%for Maple Arrow and AC Proteus cultivars, respectively,

Table 3Comparison of performance of 14 commercial northern adapted soybean

Soybean cultivars Methionine Protei

g/100 g of protein g/100 g of dryseed weight

g/100seed m

KG 20 2.09a,b,c,d,e 0.70 33.23b

KG 30 2.01a,b,c,d 0.67 33.10b

KG 41 2.16a,b 0.67 31.31d

OAC Libra 2.22a 0.67 30.01f

OCA Scorpio 1.94c,d,e 0.63 32.17c

OAC Frontier 2.16a,b 0.66 30.31e

AC Harmony 2.10a,b,c,d,e 0.63 29.78g

AC Bravor 2.01a,c 0.62 31.01e

QS 3011.2 2.09a,b,c 0.72 34.27e

Maple Amberj 2.13d,e 0.69 32.19d

Maple Arrow 2.15a,b,c 0.69 31.84c

Bicentennial 1.61f 0.56 36.07a

McCall 1.94d,e 0.63 32.52b

Early Hakucho 1.93a,b,c,d 0.66 34.27e

a–i Means along a horizontal row with different superscripts are significantly dj Taken from Zarkadas et al. (1999).

and 30.1–33.1% for nine yellow and brown seed coatedsoybean cultivars. These results are also in accord withthose of Brumm and Hurburgh (2002), who have reportedthat the US commercial soybean varieties contain on theaverage 36% protein. The present data, however, are con-siderably lower than those protein values reported by Bur-ton et al. (1982), Poysa, Woodrow, and Yu (2006) andGrieshop et al. (2003) who have reported soybean proteincontents determined either by the conventional micro-Kjel-dahl or by infrared nitrogen procedures.

In the present study, the best estimate of the protein andnitrogen contents of soybean cultivars was made by thesummation of the weights of the amino acid residues oramino acid nitrogen of each cultivar, as described previ-ously (Zarkadas et al., 1993, 1999). For purposes of com-parison, the data for the total nitrogen contents of theselected soybean cultivars, expressed as grams of aminoacids per 16 g of total nitrogen, are presented in Tables 2and 3. The total nitrogen of these samples ranged from16.5% to 17.9%, which is considerably higher than the16.0% value frequently assumed for proteins. The calcu-lated protein conversion factors among these samples var-ied from 5.58 in QS 3011.2 to 6.06 in Maple Amber.Although many of the amino acid values appeared to bevery similar, cultivar to cultivar variations in amino acidcontent from the weighted mean were significant withrespect to 10 amino acids. The variation in amino acidcomposition showed statistically highly significant differ-ences (P < 0.001) for four amino acids, namely asparticacid, threonine, serine and phenylalanine. The other vari-ables that showed significant effects were glycine and leu-cine at the P < 0.05 level. The values reported in Table 1,are in accord with the value of Steinke (1992) and Zarkadaset al. (1993, 1999). There was also a significant variation intheir contents of proline, valine and basic amino acids atthe P < 0.01 level. The values reported for proline were

n Nitrogen (N)

g of dryatter

g amino acid N/100 g protein g amino acid/16 g N

,c 17.4a,b 92.19a,b,c

,c 17.3b,c 92.55a,b,c

,e,f,g 17.4a,b,c 92.01a,b,c

,g 17.4b,c 92.04a,b,c

,d,e 17.2b,c 92.99b,c

,f,g 17.4a,b 92.02a,b,c

17.3b,c 92.46a,b,c

,f,g 17.5a,b 91.44a,b

,f,g 17.9a 89.60a

16.5c 88.98,d,e,f 16.8b,c 94.02a

17.0b,c 92.51a,b

,c,e 17.2b,c 88.98a,b

,f,g 17.4a,b,c 92.11a,b,c

ifferent at the P < 0.001 level by Duncan’s (1955) new multiple-range test.


slightly lower than those of Steinke (1992) and Zarkadaset al. (1993, 1999).

These results (Table 1) show that all soybean cultivarsinvestigated were high in several amino acids including glu-tamic acid. Mean glutamic acid values ranged from 16.9 g/kg of protein in Bicentennial to 19.3% in Maple Amber,followed by aspartic acid (10.6–11.5%), leucine (7.2–8.1%), arginine (6.2–7.7%), lysine (6.0–6.7%), phenylala-nine (5.1–5.5%), and valine (7.5–8.1%). Thus, seven ofthese amino acids account for 59.5–63.7% of the total pro-tein. The total acidic amino acids when added together rep-resented 27.5–30.8%, compared to the basic amino acids,which constituted 15.4–17.1%. The amino acids withhydrophobic side chains accounted for a further 30.4%.Soybean genotypes are also a good source of aromaticamino acids, tyrosine and phenylalanine, which accountedfor 8.9–9.3%. These results were in close agreement withthose of Steinke (1992) and Zarkadas et al. (1993, 1999).

Among the amino acid residues present in lowestamounts are methionine (1.6–2.2%), cysteine (1.9–2.4%)and tryptophan (1.2–1.5%). In the present study, it wasfound (Table 3) that genotypes with the higher protein con-tent had the lowest sulfur amino acid content. The levels ofmethionine found in this study, which correspond to 0.56–0.72% of the dry seed weight, are in accord to thosereported by Burton et al. (1982) and Krishnan et al.(2005) from their studies on increasing both percent proteinand methionine in soybeans. These authors found that theoverall methionine content of soybean protein is only1.4–1.6% of the total amino acid content, correspondingto 0.56–0.77% methionine on a dry weight basis. Theseresults indicate that the methionine content of soybeanswould need to be increased to 3.0% of the total proteinto provide the methionine equivalent of egg protein.Karr-Lilienthal, Grieshop, Spears, and Fahey (2005)reported that the methionine levels for soybean meals from55 US soybean processing plants varied from 0.78% formeals produced in zones I, II, V, VI and VII to 0.84%for zones III and IV, which were higher than the 0.74%value reported in the 1998 National Research Councilnutrient requirements for swine (Steinke, 1992).

Although traditional recurrent selection, and backcross-ing soybean breeding methods have successfully increasedthe protein content in soybean seeds without significantlyreducing yield, this procedure has not increased methioninelevels (Burton et al., 1982; Steinke, 1992). There is aninverse correlation between each of the following: proteincontent and yield, protein and methionine, protein andoil, and carbohydrate and protein accumulation. Accord-ing to Krishnan et al. (2005) the negative correlationbetween total protein and sulfur amino acid content ismostly mediated by the differential accumulation of theBowman–Birk protein inhibitor. By contrast both oil andcarbohydrate appear to be positively related to one anothergenetically (Burton et al., 1982; Fehr et al., 2003; Krishnanet al., 2005; Steinke, 1992). Improvements in the quality ofsoybean proteins, therefore, will necessitate either a reduc-

tion in the proportion of sulfur-poor storage proteins or ahigh level expression of genes encoding an increase inmethionine-rich endogenous proteins or a combination ofthe two. Tanksley and McCouch (1997) predicted thatthe ceiling in soybean yield has been reached, and thatbecause of the complexity of this trait, little scope existsfor increasing the harvest index beyond the present valueof 50%, unless the rate of photosynthesis and regulationof the input of the stomata can be altered (Mann, 1999).Nielsen et al. (1989, 1997) suggested that the flux of carboninto protein, oil, and carbohydrate is tightly regulated inthe developing seed, and that at least some of the biochem-ical control is based on feedback regulatory mechanisms.

Gayler and Sykes (1985) have shown that the composi-tion of soybean storage proteins is also affected bynutritional availability and environmental conditions.Methionine supplementation of an in vitro soybean cotyle-don or of intact soybean plants was reported to affect theamounts of several amino acids including a 21.9% increasein methionine and a decrease of 11.4% in arginine content.These authors suggested that the observed changes weredue to a dramatic decrease in the synthesis of the b-subunitof b-conglycinin, and an increase in the ratio of glycinin tob-conglycinin. According to Gayler and Sykes (1985) theratio of b-conglycinin to glycinin influences the sulfuramino acid content and protein quality of soybean seeds.Krishnan et al. (2005) have shown that nitrogen applica-tion to soybean plants favoured the accumulation of b-sub-unit of b-conglycinin while decreasing the accumulation ofBowman–Birk protease inhibitor, a protein rich in cysteine.These authors indicated that the negative correlationbetween total protein and sulfur amino acid content ismostly mediated by the differential accumulation of theBowman–Birk protease inhibitor. Krishnan (2005) con-cluded that a two-prong approach which involves meta-bolic engineering of sulfur assimilation pathway genesand high level expression of genes encoding sulfur-richstorage proteins, appears to be the most promising wayof designing soybeans with the desirable amount of sulfuramino acids.

Tryptophan in soybeans was determined by the rapidmethod described in this study. As may be seen in the chro-matogram of Fig. 1, operating the microcolumn at 52 � C,with a single eluting buffer, 0.21 M Na citrate pH 5.125, 3-nitrotyrosine is completely separated from phenylalanine,and tryptophan emerges as a discrete peak. The data pre-sented in Table 1 indicate that the levels of tryptophan inall commercial cultivars investigated are similar, rangingfrom 1.1% to 1.4% of the total amino acids. Tryptophanis an important biological constituent of numerous pro-teins in animals and plants, and an essential amino acidin human and animal diets. However, it is readily destroyedduring acid hydrolysis (6.0 M HCl) of proteins, and itsquantitative determination has remained a difficult analyt-ical problem. The most effective method, which is applica-ble to both purified proteins and plant and animal tissues,is the alkaline procedure of Hugli and Moore (1972), which


allows quantitative recoveries (96–100%) of tryptophan inprotein and tissues using the presently reported rapid tryp-tophan column chromatographic method.

3.2. Evaluation of protein quality of soybeans

The essential amino acid profiles and protein ratings ofthe 14 soybean cultivars investigated are compared withthose of the FAO/WHO (1991) reference pattern of hen’swhole egg for the diet of a 2- to 5-year-old child, and theresults are summarized in Table 4. The proposed methodfor calculating the protein digestibility corrected aminoacid score (PDCAAS) of soybeans can be defined (Eq.(3)) as follows:

PDCAAS

¼ AA contentðmg=g of proteinÞof food protein�digestibility

AA content of FAO=WHO ½29� pattern for 2-to5-year-old child:

These authors have defined the amino acid score as theconcentration of the limiting amino acid in the food pro-tein, which is expressed as a proportion or percentage ofthe concentration of the same amino acid in a standardor reference pattern, such as for the diet of the 2- to5-year-old child. Digestibility is included in this amino acidscoring procedure (Eq. (3)) to allow for differences in thedigestibility between plant and animal sources. The trueprotein digestibility values for soybeans quoted in thisstudy (Table 4) were taken from the US Federal Register’s

Appendix B, pp. 2193–2195 (USFDA, 1993).The calculated amino acid scores for the 14 soybean cul-

tivars are very similar in their EAA9 contents (Table 4).The essential amino acid pattern (EAA9) ranged from45% to 47% compared to the 33.9% reference protein pat-tern value given by FAO/WHO (1991) for the diet of a 2-to 5-year-old child. These results indicate that commercialsoybeans provide a good balance of total essential aminoacids, limited only in methionine, and have a proteindigestibility-corrected amino acid score of 91% for all cul-tivars, compared to the value of egg protein (97%).

Therefore, for practical human nutrition, under condi-tions of normal usage of soybean proteins in the diet,methionine supplementation is likely unnecessary, exceptfor the feeding of the newborn, where modest supplemen-tation of soybean based formulas with methionine maybe beneficial (Steinke, 1992). From the results presentedin Table 4, it is evident that this method is a more conciseand useful means for evaluating the protein quality of dif-ferent soybean cultivars (FAO/WHO, 1991; Krishnan,2005; USFDA, 1993).

3.3. One-dimensional electrophoresis of seed storage

proteins from soybean

To identify variants of storage protein subunits in com-mercial soybean cultivars, protein extracts were analyzed

by SDS–PAGE. Typical electrophoretic patterns obtainedfrom total proteins from 14 selected soybean cultivars onlarge gradient (8 to 18%) gels, are illustrated in Fig. 2.The patterns among cultivars are similar, with good resolu-tion over the entire gel, including the most basic polypep-tides. The identity of the major storage protein subunitspresent in soybeans was confirmed in separate 2-DE refer-ence maps and tandem mass spectrometric experiments,using null-allele glycinin, b-conglycinin and lipoxygenasegenotypes (Zarkadas et al., accepted for publication).

The protein bands (Fig. 2), have Mr ranging from 7 to116 kDa, with the subunits of the major storage proteins,glycinin and b-conglycinin, comprising the bulk storageproteins. The first single broad band corresponds to lipoxy-genase (LO). Soybeans have three LO isoforms (sLO-1,sLO-2, and sLO-3), which have homologous sequencesand a Mr of �96 kDa (Schenk, Neidig, Zhou, Holman, &Solomon, 2003). The subunits of b-conglycinin separatedon SDS–PAGE into three major bands of Mr � 78, 75and 47 kDa, corresponding to the a, a 0, and b subunitsof this storage protein, respectively.

The glycinin subunits separated into five acidic (A) andfive basic (B) bands. The first two glycinin bands with Mr

of �37 and 35 kDa correspond to the acidic polypeptidechains A3 and A4, respectively. The next group of acidicpolypeptide chains of glycinin migrated as a broad bandwith Mr of �34 kDa, corresponding to the A1a, A1b andA2 polypeptide chains. The smaller acidic polypeptide ofglycinin in the A5A4B3 subunit, which is designated A5,and has an Mr of �10 kDa, was not well resolved. Thepolypeptide that separated on the SDS–PAGE gels withan Mr of �23 kDa is the basic polypeptide, designated asB3. The cluster of polypeptides that migrated as a broadband with Mr of �22 kDa is composed of the combinedbasic components of glycinin, designated B1a, B1b, B2,and B4. These results are consistent with earlier reports(Adachi et al., 2003; Maruyama et al., 2001). In additionto these components of glycinin and b-conglycinin, thereare three other proteins that separated as single bands inthe SDS–PAGE gel (Fig. 2), including the Gly m Bd 30K(P34) protein with Mr of �30 kDa, which is one of thethree known food allergens of soybeans (Ogawa, Samoto,& Takahashi, 2000; Schenk et al., 2003), seed lectins andthe Kunitz trypsin inhibitor isoenzymes with Mr � 33,and 21 kDa, respectively. These results are comparable tothose reported previously by Yagasaki et al. (1997) andYaklich (2001) for soybean isogenic lines having differentglycinin subunit composition.

3.4. 2-DE references maps and identification of seed

storage proteins from soybeans

For good resolution of soybean proteins by two-dimen-sional electrophoresis, it was found that a 10% TCA–0.05%DTT in acetone precipitation step for removing salts, lip-ids, phenolic compounds, and other non-protein interfer-ing compounds from soybean seed powders, followed by

Table 4

Comparison of the essential amino acid (EAA) scores of 14 soybean cultivars with hen’s whole egg, and the EAA requirements of a 2- to 5-year-old child

EAA EAAa requirements for a preschool child (2- to 5-year-old) Source of soybean cultivars

Commercial University of Quelph Public sector (AAFC)


Milligrams of amino acid per gram of total proteinb

Histidine 19 28 27 26 28 29 27 27 28 27

Isoleucine 28 49 48 49 49 49 50 49 49 48

Leucine 66 78 77 78 78 75 79 77 79 76

Lysine 58 64 64 65 65 66 64 64 65 64

Methionine + cyst(e)ine 25 43 40 45 44 38 43 41 41 41

Phenylalanine + tyrosine 63 89 91 92 93 93 94 92 92 91

Threonine 34 50 50 50 50 50 50 51 51 48

Tryptophan 11 14 12 14 13 12 14 12 14 12

Valine 35 51 50 52 54 52 54 52 53 51

% Total protein

EAA9b 33.9 46.6 45.9 47.1 47.4 46.4 47.5 46.5 47.2 45.8

Percent true protein digestibioityc in man 91 91 91 91 91 91 91 91 91

Protein digestibility corrected amino acid scored 91 91 91 91 91 91 91 91 91

EAA EAAa requirements preschool child (2- to 5-year-old) Source of soybean cultivars Animal product

Maple Amberd Maple Arrow Biocentennial McCall Edamame

Early Hakucho Eggb

Milligrams of amino acid per gram of total proteinb

Histidine 19 29 28 26 26 27 22

Isoleucine 28 50 50 46 47 45 54

Leucine 66 80 81 73 76 72 86

Lysine 58 66 63 60 62 62 70

Methionine + cyst(e)ine 25 42 44 37 37 39 57

Phenylalanine + tyrosine 63 91 91 91 90 88 93

Threonine 37 43 46 50 50 51 47

Tryptophan 11 12 14 13 14 11 17

Valine 54 51 52 52 52 49 66

% Total protein

EAA9b 33.9 46.3 46.9 44.8 45.4 44.4 51.2

Percent true protein digestibilityd in man 91 91 91 91 91 97

Protein digestibility corrected amino acid scored 91 91 91 91 91 97

a Data from FAO/WHO (1991).b Calculation of protein ratings of the 14 northern adapted commercial soybean cultivars was carried out by comparison of the amino acid composition of hen’s whole egg with that of the reference pattern established by FAO/

WHO (1991) for a preschool child (2- to 5-year-old).c True protein digestibility values were taken from the USFDA (Federal Register, Appendix B) (1993).d Taken from Zarkadas et al. (1999).

142C

.G.

Za

rka

da

set

al.

/F

oo

dR

esearch

Intern

atio

na

l4

0(

20

07

)1

29

–1

46

Fig. 2. Comparison of one-dimensional SDS–PAGE profile of total protein extracts, of 14 Canadian commercial soybean (Glycine max L. Merr.) seedvarieties. Total protein (50–75 lg) was resolved on a linear 8–18% linear SDS–PAGE. The sample lanes are: (1) McCall; (2) OAC Scorpio; (3)Bicentennial; (4) KG 20; (5) KG 30; (6) KG 41; (7) Maple Arrow; (8) AC Bravor; (9) OAC Frontier; (10) OAC Libra; (11) AC Harmony; (12) QS 3011.24;(13) Maple Amber; and (14) Early Hakucho. Mass ranges of the protein markers are shown on the left side of the gel. The principal storage proteinsubunits of b-conglycinin (a 0, a and b subunits), and of the acidic (A1–A4), and basic (B1–B4) glycinin subunits are indicated on the right side. The othermajor bands identified are lipoxygenase, lectins, Kunitz trypsin inhibitor, and Gly m Bd 30KD which is a major thiol protease-like 34 kDa glycoproteinand a major allergen of soybean.


protein solubilization in a 8 M urea 2 M thiourea buffer,effectively partitioned the protein sample from interferingcontaminants or artifacts due to hydrolytic action of prote-ases and peroxidases (Gegenheimer, 1990; Granier, 1988;Natarajan et al., 2005; Wang et al., 2006).

Initial 2-DE studies with IPG strips that ranged frompH 3 to 10 showed that the proteome region from pH 4to 8 was highly dense, and for this reason all subsequent2-DE studies (Fig. 3) were carried out using narrow rangeIPG strips. The 2-DE reference maps shown in Fig. 3, wereobtained using IPG strips that ranged from pH 4.7 to 5.9for the acidic components of seed storage proteins, andfrom pH 6 to 11 for the separation of the basic chain of gly-cinin. Comparison of 2-DE reference maps, shown inFig. 3, isolated from seed proteins (50 lg) of five typicalsoybean varieties, namely McCall, KG 30, KG41, OACFrontier and AC Bravor, using narrow range pH 4.7–5.9IPG strips, revealed strong similarities in the overall distri-bution pattern of the acidic polypeptides of glycinin and b-conglycinin. The 2-DE reference maps (Fig. 3a) showedthat the acidic protein spots were well-resolved, free ofstreaking and smearing. Analyses of the whole soybeanseed proteins from the rest of the soybean cultivars alsorevealed strong similarities on their 2-DE protein spot pat-terns, indicating good reproducibility of the proceduresused for both protein extraction and electrophoretic sepa-ration. Additional 2-DE maps of the basic polypeptidechains isolated from Maple Arrow, KG 30, KG41, OACFrontier and AC Bravor, using medium-range pH 6–11IPG dry strips, and 15% SDS–PAGE gels, showed

improved resolution of the basic polypeptide maps of gly-cinin, as shown in Fig. 3b.

This study demonstrates that the use of narrow pHrange IPG strips for the acidic polypeptide chains in thefirst dimension followed by SDS–PAGE size separationin the second dimension, provided significantly improvedreproducibility of the subunits of b-conglycinin and gly-cinin. The number of protein spots reproducibly detectedfrom the pH 4.7 to 5.9 2-DE gels (Hajduch et al., 2005;Mooney & Thelen, 2004) is over 223 acidic soybean pro-teins, Fig. 3a, compared to the results reported in Fig. 3bwhere fewer than 100 basic protein spots were detected inthe pH 6–11 2-DE gels. Since these proteins are encodedby multigene families, the 40 most abundant protein spots,which have similar masses but differ in charge, were identi-fied as being constituent polypeptide chains of glycinin andb-conglycinin. As expected for proteins encoded by multi-gene families, several spots with similar masses but differingin charge were identified as being the same storage protein.The identity of the variant storage protein subunits of gly-cinin and b-conglycinin shown on the 2-DE reference mapswas confirmed in separate experiments by peptide mass fin-gerprint data, using null-allele soybean genotypes (Schenket al., 2003; Zarkadas et al., accepted for publication).

A comparison of the results, presented in Table 2, showsa broad range of variability among cultivars for all subunitcomponents of b-conglycinin, and for both the acidic andbasic polypeptide chains of glycinin. The data presentedin Table 5 indicate that the range among cultivars for a,and a 0 subunits was 20–28 percentage units, respectively,

Fig. 3. Two-dimensional proteome maps of total protein (55 lg) from 14 soybean cultivars. (a) The acidic peptide chains were separated using narrow pHrange 4.7–5.9 IPG strips followed by 12.5% 2-DE SDS–PAGE. The reference gel is McCall, and four typical genotypes, KG 30, KG 41, OAC Frontier andAC Bravor are included for comparison. (b) The basic storage protein peptide chains were resolved on pH 6–11 linear IPG strips followed by 15% 2-DESDS–PAGE. The reference gel is Maple Arrow. Mass (Mr) ranges and pH are noted. The Brilliant (Coomassie) Blue R-250 stained gels were imaged byAGFA ‘‘Snapscan 50’’ scanning densitometer, and proteins were detected using Phoretic 2D advanced software. The most dense spots, noted by arrows,indicate the reference storage protein spots used for computerized gel matching. Forty-one spots were selected for analysis based upon intensity, circularityand resolution.


compared with the range of 40 percentage units for the b-subunit of b-conglycinin. The values for b-conglycinin indi-cate that the cultivars Maple Arrow and Early Hakuchohad the lowest values for the b-subunit compared toMcCall, KG 30 and KG 41. The acidic A1a,1band2 polypep-tides of Maple Arrow and Early Hakucho had the highestvalues compared to those of Bicentennial and OAC Libra.The range among cultivars for A1a+1b+2 and A3 acidic pep-tide chains was only 17 percentage units compared with arange of 40 percentage units for the A4 peptide chain ofglycinin. These results are in accord with those of Fehret al. (2003) and Yaklich (2001), who also investigatedthe relationship between the individual components of gly-cinin and b-conglycinin of multiple cultivars.

There was a significant difference in the glycinin to b-conglycinin ratio among the soybean varieties investigated

ranging was from 2.71 for Maple Arrow, to 1.27 for KG 30(Table 5). The data for the glycinin to b-conglycinin ratiofor Maple Arrow, AC Harmony and QS 3011.2 are similarto those reported for Amsoy (2.77), Prize (2.16), Weber(2.13) and Vinton (2.41) by Murphy and Resurrection(1984), and more recently for Vinton 81 (2.04) by Fehret al. (2003). These results, however, were considerablyhigher than the glycinin to b-conglycinin ratios reportedby Yagasaki et al. (1997) for null soybean cultivars lackingdifferent subunits of glycinin. These authors indicated thatdecreasing the glycinin to b-conglycinin ratios due to lackof specific subunits, had a negative effect on tofu qualityand the food processing properties of soybeans.

From these results, it is evident that the protein qualityof soybeans varieties can be determined from their totalprotein content, their amino acid composition, corrected

Tab

le5

Co

mp

aris

on

of

the

sub

un

itco

mp

osi

tio

no

fb

-co

ngl

ycin

inan

dgl

ycin

inse

edst

ora

gep

rote

ins

fro

m2-

DE

gels

(exp

ress

edas

%re

lati

vevo

lum

eo

fea

chsp

ot(

s)o

nth

ege

lim

age)

amo

ng

14so

ybea

ncu

ltiv

ars

scan

ned

by

the

AG

FA

‘‘S

nap

scan

50’’

scan

nin

gd

ensi

tom

eter

and

imag

edb

ya

Ph

ore

tic

2-D

EA

dva

nce

dE

xpre

ssio

n(2

005)

gel

anal

ysis

soft

war

e

So

ybea

ncu

ltiv

ars

%R

elat

ive

volu

mea

Rat

iob

glyc

inin

/

b-c

on

glyc

inin

b-C

on

glyc

inin

esu

bu

nit

sG

lyci

nin

po

lyp

epti

de

chai

ns

aa0

bP

a,a0 ,b

Aci

dic

Bas

icP

Aci

dic

and

bas

icch

ain

sP

A1

+2

A3

A4

PA

1+

2+

3+

4

PB

1+

2+

3+

4

KG

2014

.10

8.91

10.7

833

.32

28.6

66.

286.

5941

.52

25.1

766

.34

1.99

KG

3015

.46

11.6

217

.01

44.0

923

.68

5.55

5.58

34.8

121

.10

55.9

11.

27

KG

4118

.72

10.4

513

.41

42.5

725

.62

5.89

4.25

35.7

621

.67

57.4

31.

35

OA

CL

ibra

18.0

712

.46

12.4

442

.97

24.7

87.

246.

7838

.73

23.4

762

.21.

44

OA

CS

corp

io17

.23

10.7

812

.42

40.4

322

.82

7.32

6.96

37.0

922

.48

59.5

71.

47

OA

CF

ron

tier

18.8

69.

5613

.68

41.8

325

.69

6.16

4.37

36.1

721

.95

58.1

21.

39

AC

Har

mo

ny

14.7

58.

238.

9731

.95

30.1

36.

535.

7042

.37

25.5

267

.90

2.13

AC

Bra

vor

16.5

711

.20

11.1

838

.95

25.6

45.

836.

5538

.03

19.7

157

.74

1.48

QS

3011

.215

.22

8.14

10.0

833

.44

26.8

97.

507.

0741

.47

25.1

366

.60

1.99

Map

leA

mb

er20

.03

12.3

68.

5641

.25

26.9

65.

584.

0336

.58

22.1

758

.75

1.42

Map

leA

rro

w13

.11

8.83

5.02

26.9

732

.35

7.16

5.92

45.4

727

.63

73.0

92.

71

Bio

cen

ten

nia

l22

.11

11.2

410

.23

43.5

824

.38

6.02

4.73

35.1

321

.29

56.4

31.

29

McC

all

15.4

810

.01

14.9

640

.46

26.2

59.

015.

4037

.08

22.4

759

.55

1.47

Ear

lyH

aku

cho

16.8

910

.68

4.76

32.3

428

.92

6.44

6.74

42.1

225

.53

67.6

52.

09

aT

he

per

cen

tre

lati

vevo

lum

eo

fea

chp

rote

insp

ot

on

the

gel

isth

era

tio

of

agi

ven

pro

tein

spo

tto

the

sum

of

all

pro

tein

spo

tsan

alyz

ed.

bT

his

rati

ow

asca

lcu

late

das

the

sum

of

the

bas

ican

dac

idic

sub

un

its

of

glyc

inin

ove

rth

esu

mo

fth

ea

,a0

and

bsu

bu

nit

so

fb

-co

ngl

ycin

in.


for protein digestibility, as recommended by the FAO/WHO (1991), and by the USFDA (1993), and from theratio of the major seed storage protein components of gly-cinin and b-conglycinin. The 2-DE reference maps make itpossible to assess the genetic variability of the soybean cul-tivars, which can then be correlated with their protein qual-ity and food processing. These three methods can be usedas very effective tool for assisting plant breeders in theirselection of high quality soybean varieties.

Acknowledgments

The authors thank Dr. Stephen J. Molnar and Dr. Yu-zuki Manabe for their critical review of the manuscript,Mrs. Judith McCarthy and Mrs. Susan Flood for theirexcellent graphic design work, and Miss Victoria Changfor her excellent technical support. They also gratefullyacknowledge Mr. Marc Lefebvre of the Eastern Cerealand Oilseed Research Centre for modifying the 7.1 systemGold chromatographic data reduction system.

Contribution No. 06-687 (MS 100 August 7, 2006) fromEastern Cereal and Oilseed Research Centre.

References

Adachi, M., Kanamori, J., Masuda, T., Yagasaki, K., Kitamura, K.,Mikami, B., et al. (2003). Crystal structure of soybean 11S globulin:glycinin A3B4 homohexamer. Proceedings of the National Academy of

Science of the United States of America, 100, 7395–7400.Adachi, M., Takenaka, Y., Gidamis, A. B., Mikami, B., & Utsumi, S.

(2001). Crystal structure of soybean proglycinin A1aB1b homotrimer.Journal of Molecular Biology, 305, 291–305.

Agrawal, G. K., Yonekura, M., Iwahashi, Y., Iwahashi, H., & Rakwal, R.(2005). System trends and perspectives of proteomics in dicot plants.Part I. Technologies in proteome establishment. Journal of Chroma-

tography B, 815, 109–123.Beilinson, V., Chen, Z., Shoemaker, R. C., Fischer, R. L., Goldberg, R.

B., & Nielsen, N. C. (2002). Genomic organization of glycinin genes insoybean. Theoretical and Applied Genetics,104, 1132–1140.

Bradford, M. M. (1976). A rapid sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dyebinding. Analytical Biochemistry, 72, 248–254.

Brumm, T. J., & Hurburgh, C. R. Jr., (2002). Quality of the 2002 soybean

crop from the United States. Louis, MO: American SoybeanAssociation.

Burton, J. W., Purcell, A. E., & Walter, W. M. Jr., (1982). Methionineconcentration in soybean protein from populations selected forincreased percent protein. Crop Science, 22, 430–432.

Duncan, D. B. (1955). Multiple range and multiple F test. Biometrics, 11,1–42.

FAO/WHO. (1991). Expert Consultation. Protein Quality Evaluation,FAO/WHO Nutrition Meetings, Report Series 51. Rome, Italy: Foodand Agriculture Organization/World Health Organization.

Fehr, W. R., Hoeck, J. A., Johnson, S. L., Murphy, J. D., Nott, P. A.,Padilla, G. I., et al. (2003). Genotype and environmental influence onprotein components of soybean. Crop Science, 43, 511–514.

Fukuda, T., Maruyama, N., Kanazawa, A., Abe, J., Shimamoto, Y.,Hiemori, M., et al. (2005). Molecular analysis and physicochemicalproperties of electrophoretic variants of wild soybean Glycine soja

storage proteins. Journal of Agricultural and Food Chemistry, 53,3658–3665.

Gayler, K. R., & Sykes, G. E. (1985). Effects of nutritional stress on thestorage proteins of soybeans. Plant Physiology (1985), 582–585.


Gegenheimer, P. (1990). Preparation of extracts from plants. Methods

Enzymology, 182, 174–193.Gorg, A., Weiss, W., & Dunn, M. J. (2004). Current two-dimensional

electrophoresis technology for proteomics. Proteomics, 4, 3665–3685.Granier, F. (1988). Extraction of plant proteins for two-dimensional

electrophoresis. Electrophoresis, 9, 712–818.Grieshop, C. M., Kadzere, C. T., Clapper, G. M., Flickinger, E. A., Bauer,

L. L., Frazier, R. L., et al. (2003). Chemical and nutritionalcharacteristics of United States soybeans and soybean meals. Journal

of Agricultural and Food Chemistry, 51, 7684–7691.Hajduch, M., Ganapathy, A., Stein, J. W., & Thelen, J. J. (2005). A

systematic proteomic study of seed filling in soybean. Establishment ofhigh-resolution two-dimensional reference maps, expression profiles,and an interactive proteome database. Plant Physiology, 137,1397–1419.

Herman, E. M., Helm, R. M., Jung, R., & Kinney, A. J. (2003). Geneticmodification removes an immunodominant allergen from soybean.Plant Physiology, 132, 36–43.

Hugli, T. E., & Moore, S. (1972). Determination of the tryptophancontent of proteins by ion exchange chromatography of alkalinehydrolysates. Journal of Biological Chemistry, 247, 2828–2834.

Karr-Lilienthal, L. K., Grieshop, C. M., Spears, J. K., & Fahey, G. C.(2005). Amino acid, carbohydrate, and fat composition of soybeanmeals prepared at 55 commercial US soybean processing plants.Journal of Agricultural and Food Chemistry, 53, 2146–2150.

Krishnan, H. B. (2000). Biochemistry and molecular biology of soybeanseed storage proteins. Journal of New Seeds, 2, 1–25.

Krishnan, H. B. (2005). Engineering soybean for enhanced sulfur aminoacid content. Crop Science, 45, 454–461.

Krishnan, H. B., Bennett, J. O., Kim, W.-S., Krishnan, A. H., &Mawhinney, T. P. (2005). Nitrogen lowers the sulfur amino acidcontent of soybean [Glycine max (L.) Merr.] by regulating theaccumulation of Bowman–Birk protease inhibitor. Journal of Agricul-

tural and Food Chemistry, 53, 6347–6354.Laemmli, U. K. (1970). Cleavage of structural proteins during the

assembly of the head of bacteriophage T4. Nature, 227, 680–686.Liener, I. E. (1994). Implications of antinutritional components in soybean

foods. Critical Review of Food Science and Nutrition, 34, 31–67.Mann, C. C. (1999). Genetic engineers aim to soup up crop photosyn-

thesis. Science, 283, 314–316.Maruyama, N., Adachi, M., Takahashi, K., Yagasaki, K., Kohno, M.,

Takenaka, Y., et al. (2001). Crystal structures of recombinant andnative soybean b-conglycinin homotrimers. European Journal of

Biochemistry, 268, 3595–3604.Mooney, B. P., & Thelen, J. J. (2004). High-throughput peptide mass

fingerprinting of soybean seed proteins: automated workflow andutility of UniGene expressed sequence tag databases for proteinidentification. Phytochemistry, 65, 1733–1744.

Moore, S. (1963). On the determination of cystine and cysteic acid. Journal

of Biological Chemistry, 238, 235–237.Murphy, P. A., & Resurrection, A. P. (1984). Varietal and environmental

differences in soybean glycinin and b-conglycinin content. Journal of

Agricultural Food Chemistry, 32, 911–915.Natarajan, S., Xu, C., Caperna, T. J., & Garrett, W. M. (2005).

Comparison of protein solubilization methods suitable for proteomicanalysis of soybean seed proteins. Analytical Biochemistry, 342,214–220.

Nielsen, N. C., Bassuner, R., & Beaman, T. (1997). The biochemistry andcell biology of embryo storage proteins. In R. A. Larkins & I. K. Vasil

(Eds.), Cellular and molecular biology of plant seed development

(pp. 151–220). Dordrecht: Kluwer Academic Publishers.Nielsen, N. C., Dickinson, C. D., Cho, T.-J., Thanh, V. H., Scallon, B. J.,

Fischer, R. L., et al. (1989). Charcterization of the glycinin genefamily in soybean. The Plant Cell, 1, 313–328.

Ogawa, T., Samoto, M., & Takahashi, K. (2000). Soybean allergens andhypoallegenic soybean product. Journal Nutritional Science and

Vitaminology, 46, 271–279.Poysa, V., Woodrow, L., & Yu, K. (2006). Effect of soy protein subunit

composition on tofu quality. Food Research International, 39, 309–317.Schenk, G., Neidig, M., Zhou, J., Holman, T. R., & Solomon, E. I. (2003).

Spectroscopic characterization of soybean lipoxygenase-1 mutants: therole of second coordination sphere residues in the regulation of enzymeactivity. Biochemistry, 42, 7294–7302.

Singh, R. J., & Hymowitz, T. (1999). Soybean genetic resources and cropimprovement. Genome, 42, 605–616.

Staswick, P. E., Hermodson, M. A., & Nielsen, N. C. (1984). The aminoacid sequence of the A2B1a subunit of glycinin. Journal of Biological

Chemistry, 259, 13424–13430.Steinke, F. H. (1992). Nutritional value of soybean protein foods. In F. H.

Steinke, D. H. Waggle, & M. N. Volgarev (Eds.), New protein foods in

human health: Nutrition, prevention, and therapy (pp. 59–66). BocaRaton, FL: CRC Press.

Tanksley, S. D., & McCouch, S. R. (1997). Seed banks and molecularmaps: unlocking genetic potential from the wild. Science, 277,1063–1066.

Thanh, V. H., & Shibasaki, K. (1978). Major proteins of soybean seeds.Subunit structure of b-conglycinin. Journal of Agricultural and Food

Chemistry, 26, 692–695.USFDA, (1993). Department of Health and Human Services (HHS),

Food and Drug Administration (FDA), 21 CFR Parts 1 and 101. Foodlabeling: mandatory status of nutrition labeling and nutrient contentrevision, format for nutrition label. Federal Register, 58, 2079–2195.

Wang, W., Vignani, R., Scali, M., & Cresti, M. (2006). A universal andrapid protocol for protein extraction from recalcitrant plant tissues forproteomic analysis. Electrophoresis, 27, 2782–2786.

Yagasaki, K., Takagi, T., Sakai, M., & Kitamura, K. (1997). Biochemicalcharacterization of soybean protein consisting of different subunits ofglycinin. Journal of Agricultural and Food Chemistry, 45, 656–660.

Yaklich, R. W. (2001). b-Conglycinin and glycinin in high-protein soybeanseeds. Journal Agricultural and Food Chemistry, 49, 729–735.

Zarkadas, C. G., Gagnon, C., Poysa, V., Khanizadeh, S., Cober, E. R.,Chang, V., et al. (accepted for publication). The protein quality andidentification of the storage protein subunits of new null soybean andtwo tofu genotypes, using amino acid analysis, two-dimensionalelectrophoresis and tandem mass spectrometry. Food Research Inter-

national, doi:10.1016/j.foodres.2006.08.005.Zarkadas, C. G., Rochemont, J. A., Zarkadas, G. C., Karatzas, C. N., &

Khalili, A. D. (1987). Determination of methylated basic, 5-hydrox-ylysine, elastin crosslinks, other amino acids, and the amino sugars inprotein and tissues. Analytical Biochemistry, 160, 251–266.

Zarkadas, C. G., Voldeng, H. D., Yu, Z.-R., & Choi, V. (1999).Assessment of the protein quality of nine northern adapted yellowand brown seed coated soybean cultivars by amino acid analysis.Journal Agricultural and Food Chemistry, 47, 5009–5018.

Zarkadas, C. G., Yu, Z.-R., Voldeng, H. D., & Minero-Amador, A.(1993). Assessment of the protein quality of a new high-proteinsoybean cultivar by amino acid analysis. Journal of Agricultural and

Food Chemistry, 41, 616–623.

http://dx.doi.org/10.1016/j.foodres.2006.08.005

Assessment of the protein quality of fourteen soybean [Glycine max (L.) Merr.] cultivars using amino acid analysis and two-dimensional electrophoresis

Documents