54

E. T. Contis et al. (Editors) Food Flavors: Formation, Analysis and Packaging Influences © 1998 Elsevier Science B.V. All rights reserved

597

Role of phenolics inflavorof rapeseed protein products M. Naczk*, R. Amarowicz^, and F. Shahidi'' ""Department of Human Nutrition, St. Francis Xavier University, P.O. Box 5000, Antigonish, Nova Scotia, B2G 2W5, Canada ^Division of Food Science, Institute of Animal Reproduction and Food Research of Polish Academy of Sciences, Olsztyn, Poland ''Department of Biochemistry, Memorial University of Newfoundland, St. John's, NF, AlB 3X9, Canada Abstract Utilization ofrapeseed/canola as a source offood-grade proteins is still limited by the presence of glucosinolates, phytates, hulls, and phenolics. Phenolic acids and condensed tannins are the predominant phenolic compounds found in rapeseed. The content of phenolic compounds in rapeseed/canola products is 4-3 0 times higher than that found in corressponding products obtained from other oleginous seeds. Contribution of free phenolic acids, sinapines and condensed tannins to the bitter taste and astringency of rapeseed products is important. In addition, both phenolic acids and condensed tannins may form complexes with proteins. Better understanding of factors influencing the interactions between phenolics and proteins would be beneficial in developing more efficient technological procedures for the production of phenolic-free rapeseed/canola protein products. 1.

INTRODUCTION

Rapeseed/canola are conventionally processed to oil and feed-grade meal by employing an extraction process which is an adaptation of soybean technology adjusted to small seed size, high oil content, and the presence of myrosinase [ 1 ], an enzyme which degrades glucosinolates to toxic products such as isothiocyanates, nitriles, and thiocyanates [2-4]. The use of rapeseed meal as a source of protein for human consumption has been considered for many years because the meal obtained after oil extraction contains approximately 40% protein (Nx6.25), up to 10% non-protein nitrogen, and about 7% ash [5, 6]. Futhermore, rapeseed meal has a reasonably well balanced amino acid content [7] which is not affected by the handling and processing conditions [8]. In addition, the protein efficiency ratio (PER) ofrapeseed meal was found to be 2.64 as compared to 2.19 for soybean meal [9]. The acceptability of oilseed protein products such as meal, protein isolate, and protein concentrate depends not only on their nutritive value but also on theirfiinctionalproperties and sensory quality. The functional properties ofprotein products include their water absorption (WA), oil absorption (FA), nitrogen solubility index (NSI), emulsifying capacity, emulsion stability (ES),

598 whippability and foam stability. The WA of commercial canola meal was found to be similar to that of soybean meal, but it was 15-37% lower than that of laboratory prepared canola meal. Moreover, canola meals absorbed up to 80% more soybean oil than soybean meals, but exhibited similar emulsifying properties to those of soybean meals (Table 1). These functional properties suggest that rapeseed/canola meals could be used as binders in meat products and as extenders for meat proteins [6]. Although the proximate composition, nutritive value and functional properties ofrapeseed products are comparable to corresponding soybean products, the use of rapeseed/canola protein products as a food component is still limited by the presence of undesirable components such as glucosinolates, phytates, hulls (fiber) and phenolics.

Table 1 Functional properties of canola and soybean meals Meal Canola Varieties Tower Regent Candle Altex Commercial canola Soybean I Soybean II " oil is released Adapted from: M. Naczk,

NSI (%)

WA (%)

FA (%)

ES (%)

25.3 23.5 17.5 27.6 13.7 16.3 15.5

370 343 383 377 278 311 301

188 203 190 219 134 105 96

108 102 106 93 79a 103^

99a

L.L. Diosady, and L.J. Rubin, J. Food Sci. 50, 1985, p. 1687.

Over the last two decades the composition of rapeseed has been significantly altered by both Canadian and European breeders who developed new low glucosinolate and low erucic acid rapeseed varieties. Glycosinolates upon hydrolysis produce nitriles, hydroxynitrites, isothiocyanates and thiocyanates which are responsible for goitrogenic effects. Erucic acid in the oil may cause heart lesion in certain experimental animals. In Canada, these new varieties are known as canola [10]. Nonetheless, these improved varieties of rapeseed still contain too high levels of glucosinolates to be considered as a suitable source of protein for human consumption. A number of chemical, microbial and physical as well as combination of treatments have been developed to reduce the content of glucosinolates in meals or seeds to negligible levels [2, 11, 12]. The rapeseed/canola products contain up to 4.1 % phytates [13]. Phytates are responsible for the decrease in the bioavailability of mono- and divalent cations due to complex formation [1418]. In addition, phytates are reported to delay the digestion of starch [19]. Because of this, a number of methods have been developed to remove phytic acid from rapeseed products and these have been thoroughly discussed by Thompson [20]. On the other hand, some recently published studies indicate that phytates, at low concentrations, may possess antioxidative [21] and anticarcinogenic effects [19]. Rapeseed meal contains 20-3 0% indigestible hulls, on a dry weight basis [22-24]. The rapeseed hulls consist of low-molecular-weight carbohydrates, polysaccharides, pectins, cellulose, lignin,

599 as well as proteins, polyphenols, glucosinolates and minerals. High levels of hulls limit the use of meal as feed due to lowering of its metabolizable energy [25-27] and cx)ntribute to the unattractive appearance of products containing rapeseed meal. A number of procedures for dehuUing rapeseed/canola have been proposed [28-33]. However, these methods are still not very efficient and therefore dehuUing is not a standard practice in canola processing and crushing industries. The content ofphenolic compounds in rapeseed/canola products is much higher than that found in corresponding products from other olegineous seeds. Therefore, phenolics are thought to be responsible for the dark color, undesirable flavor and lower nutritional value of rapeseed products. However, some published data also implicate residual solvent [34], residual oil [35], free amino acids [36, 37] and glucosinolates and their breakdown products [38] as contributors to the objectionable taste of rapeseed/canola products. This contribution discusses possible roles of phenolics in the flavor attributes of rapeseed/canola products.

2.

PHENOLIC ACIDS

Rapeseed/canola phenolic acids are derivatives of benzoic and cinnamic acids (Figure 1). Of these, sinapic acid is the predominant phenolic acid found in rapeseed/canola varieties. Phenolic acids are present in seeds and in corresponding protein products in the free, esterified and bound forms. The content of phenolic acids in rapeseed/canola meals is up to five times higher than those found in soybean meals. The content of phenolic acids in rapeseed flours is 10-30 times higher than that found in flours obtained from other oleginous seeds. Free and esterified phenolic acids are considered to be the principal contributors to the objectionable taste of rapeseed/canola products.

w HO-/

\-COOH

X

Y

HO - \ Z

V - C H =: CH — COOH

Acid

W

Protocatechuic

H

OH

Vanillic

OCH3

H

Syringic

OCH3

OCH3

Gallic

OH

OH

p-Hydroxybenzoic

H

H

Acid

Y

Z

p-Coumaric

H

H

Caffeic

H

OH

Ferulic

H

OCH3

Sinapic

OCH3

OCH3

Figure 1. Structures of phenolic acids found in canola and rapeseed.

600 2.1 Free Phenolic Acids Free phenolic acids constitute up to 24% ofthe total phenolic acids present in rapeseed/canola meal andfloursofdifferent cultivars [3 9,40] and approximately 15 % ofthe total phenolics present in rapeseed/canola meals (Table 2) [41]. Rapeseed protein products contain sinapic,/?-hydroxybenzoic, vanillic, gentisic, protocatechuic, syringic,/?-coumaric, cis- and trans-iQmXxc, caffeic and chlorogenic acids in thefreeform (Table 3) [40,42]. Ofthese, sinapic acids constitute 70.2 - 85.4% of the total phenolic acids present [41].

Table 2 Content of free and esterified phenolic acids of some rapeseed product (mg/lOOg of product, on dry basis) Product

Free

Esterified

Total

244 1542 1202 Tower meal' 262 1470 1837 Regent meal' 1807 1458 Altex meal' 248 Tower flour*' 98.2 1080.2 982 1196 84.5 1280.5 Candle flour^ 71.8 700 776.5 Start flour^ ' adapted from: M. Naczk and F. Shahidi, Food Chemistry, 31, 1989, p. 162. ^ adapted from: K. Krygier, F. Sosulski, and L. Hagge, J. Agric. Food Chem. 30, 1982, p. 335. "" adaptedfrom:H. Kozlowska, D. A. Rotkiewicz, and R. Zademowski, J. Am. Oil Chem. Soc. 60, 1983, p. 1121. Table 3 Free phenolic acids in rapeseed flours (ppm) Phenolic acid

Candle'

Tower'

Start^

Bronowski*'

22 6 5 trace p-hydroxybenzoic 6 8 3 3 vanillic 2 8 trace 4 gentisic 5 14 6 protocatechuic 3 24 15 6 syringic 8 39 31 11 coumaric 47 18 68 33 ferulic 18 6 4 3 caffeic 517 523 739 891 sinapic ' adapted from: K. Krygier, F. Sosulski, and L. Hogge, J. Agric. Food Chem. 30, 1982, p. 335. ^ adaptedfrom:H. Kozlowska, D. A. Rotkiewicz, and R. Zademowski, J. Am. Oil Chem. Soc. 60, 1983, p. 1121. Rackis et al. [43] suggested that mixtures of phenolic acids found in soy products in trace amounts possess typical soyflavorbut pure phenolic acids were, according to them, almost tasteless.

601 On the other hand, Arai and co-workers [44] described the flavor of phenolic acids as sour, astringent, bitter and phenol-like. Later, Maga and Lorenz [45] determined the taste thresholds of 30 individual phenolic acids soluble in distilled water. The taste thresholds of phenolic acids present in oilseeds, including rapeseed, were in the range of 30 to 240 ppm. Taste threshold of sinapic acid was not determined as it was insoluble in water at concentrations required for testing. Combination of two to four phenolics resulted in much lower flavor thresholds (Table 4). These authors also sugested that combinations of seven or more phenolic acids found in oilseeds should result in lower flavor thresholds. The taste thresholds of phenolic acids are also affected by the solvent used. Dadic and Belleau [46] reported thatflavorthresholds ofphenolic acids in 5% aqueous ethanol ranged from 2 to 10 ppm, while thresholds for phenolic acids dissolved in beer ranged from 10 to 50 ppm.

Table 4 Flavor thresholds ofindividual phenolic acids and their combinations used at a 1:1 (w/w) ratio (ppm) Phenolic Acids

Individual Thresholds

Combination Thresholds

30 Protocatechuic 40 Gallic 30; 40 Vanillic + p-hydroxybenzoic 90; 40 Ferulic + p-coumaric 90; 90 Ferulic + gentisic 90; 90; 90 Ferulic + gentisic + caffeic 90; 90; 90; 240 Ferulic + gentisic + caffeic + syringic Adapted from: J.A. Maga and K. Lorenz, Cereal Sci. Today, 18, 1973, p. 328.

_ 10 25 80 60 95

Rapeseed/canola meals contain over 2000 ppm offree phenolic acids [41,47], while for flours, the reported values range from trace to about 1000 ppm [40,42]. The published data indicate that individual phenolic acids are present in rapeseed flours at subthreshold levels (see Table 4). According to Maga and Lorenz [45] all phenolics acids present in oilseeds possess astringent flavor characteristics and their combination results in synergism. In addition, Bartoschuk and Cleveland [48] demonstrated the existence of synergism for bitter stimuli present in quaternary mixtures at both subthreshold and suprathreshold levels. Thus, the results of these studies strongly suggest that although the presence of individual free phenolic acids in rapeseed products is at subthreshold levels, their contribution to the objectionable taste of rapeseed meals, due to a synergistic effect, can not be ignored. 2.2 Esterified Phenolic Acids Esterified phenolic acids constitute up to 80% of the total phenolic acids. Total content of phenolic acids liberated from esters ranged from 520 to 1196 mg per lOOg of flours [42] and up to 1458mgper 100 g ofmeal [41]. Sinapic acid was the predominant phenolic acid found in alkaline hydrolyzates of soluble esters extracted from Tower and Candle canola flours [40]. It constituted 70.9 - 96.7% ofthe phenolic acids liberated from esters [47]. Small quantities of/?-hydroxybenzoic.

602 vanillic, protocatechuic, syringic, /7-coumaric, cis- and trans-fQvwXic and cafFeic acids were also reported in soluble phenolic esters fraction [40]. Fenton et al. [49] demontrated the presence of at least seven phenolic esters soluble in 70% acetone extracts ofrapeseed meals ofMidas and Echo varieties that upon hydrolysis yielded sinapic acid. Moreover, a number of phenolic esters and glycosides have been isolated and identified in rapeseed/canola. Ofthese, the sinapines, choline esters of phenolics, were the predominant soluble esters ofphenolic acids found in rapeseed/canola varieties [50]. The chemical structure of sinapines found in rapeseed/canola is shown in Figure 2. In addition, the presence of methyl esters o^cisand trans-^QwXic acids was confirmed by mass spectroscopy [51]. Flours ofPolish rapeseed varieties contained 0.16 - 0.65 ^mol phenolic acid glycosides/g flour [52]. According to Wanasundara et al. [53], Amarowicz and Shahidi [54] and Amarowicz et al. [55], 1-0-13-D glucopyranosyl sinapate was the predominant glycoside in rapeseed/canola varieties. Moreover, presence of flavonoid glycosides containing sinapic acid bound to aglycone was reported by Durkee and Harbome [56]. Two such glycosides, namely 3-(0-sinapoyl sophoroside)-7-0-glucoside of kaempferol and 3-(0-sinapoyl glucoside)-7-0-sophoroside of kaempferol have been identified in rapeseed meal [57].

z Y —^

\ -

^CH« + / 3 OH = CH — c — CHp — CHg — N — CH3

II X

0

Sinapines

X

Y

Coumaroylcholine

H

OH

H

Feruloylcholine

H

OH

OCH3

Isoferuioylcholine

H

OCH3

OH

Sinapine

OCH3

OH

OCH3

Sinapine glucoside

OCH3

0-Glu

OCH3

Z

Y ^CH« X —^

\ - C — 0 — CHg — CHg -- N — CH« CH3 0

Sinapines

\ "^

X

Y

4-Hydroxybenzoylcholine O H

H

Hesperaiin

OCH3

0CH3

Figure 2. Structures of sinapines found in canola and rapeseed.

603 Sinapine, a choline ester of sinapic acid, the most abundant phenolic ester present in rapeseed/canola products, is both an astringent and a bitter tasting phenolic compound. Because of this, it is considered to be a major contributor to the objectionable taste of rapeeseed/canola protein products [58-60]. According to Mueller etal. [61] the sinapine content of 5. napus cultiwavs was 1.65 - 2.26%, while that ofB. campestris cultivars was 1.22 - 1.54%. These authors found a statistically significant (P<0.01) difference in sinapine content between these two rapeseed cultivars. Later, Clausen [62] reported much lower levels of sinapine in both B. napus and B. campestris cukivars. The diversity in the reported sinapine contents may not only be affected by the differences in cultivar and growing conditions, but possibly due to the existing differences in solvent systems employed for extraction and methods used for their quantification. Larsen et al. [63] suggested that other sinapines (see Figure 2) may also contribute to objectionable taste of rapeseed/canola protein products. According to Durkee and Thivierge [64], sinapine is susceptible to both enzymatic (B-glucosidase) and alkaline hydrolysis, thus producing choline and sinapic acid. Free choline was found in both rapeseed flour and rapeseed protein concentrate [36]. This indicates that some hydrolysis of sinapine takes place during processing and handling of rapeseed. It has been reported that 0.1% choline solution was sligthly bitter in taste [65]. Therefore, the free choline may also contribute to undesirable taste of rapeseed/canola products. Ismail et al. [36] used the magnitude estimation test to evaluate the bitterness of solutions containing a mixture of sinapic acid and choline chloride. They demonstrated that sinapic acid and choline chloride accounted for about 80% of the bitterness of sinapine when used at equimolar concentrations. Futhermore, these authors reported that only 50 - 94% of the bitterness perceived by tasting of water slurries of rapeseed flours and concentrates can be derived from the bitterness evoked by sinapine and free choline present in rapeseed products examined. Thus, these results suggest that not only free choline, sinapic acid and sinapine, but other rapeseed components present in rapeseed may contribute to the objectionable taste of rapeseed protein products.

3.

CONDENSED TANNINS

Condensed tannins are dimers, oligomers and polymers offlavan-3-ols.The consecutive units of condensed tannins are linked through interflavanoid bonds between C-4 and C-8 or C-6 atoms [66]. Condensed tannins upon acidic hydrolysis produce anthocyanidins and therefore are also known as proanthocyanidins. According to Salunkhe et al. [67] seed coats of cereals and legumes are the primary locations of tannins in seeds. Presence of condensed tannins in rapeseed hulls was reported by Bate-Smith and Ribereau-Gayon [68]. Durkee [69] verified thisfindingand reported the presence of cyanidin, pelargonidin and an artefact, n-butyl derivative of cyanidin in hydrolytic products ofrapeseed hulls. Later, Leung et al. [70] reported that leucocyanidin was the basic unit of condensed tannins isolated from rapeseed hulls (Figure 3 ). According to Clandinin and Heard, [71] rapeseed meals contain approximately 3% tannins assayable by the AO AC method for determination oftannins in cloves and allspice [72], but Fenwick and Hogan [73] demonstrated that this value included sinapine. Later, Blair and Reichert [74] reported that the content of tannins, as determined by the vanillin assay, was 0.09-0.39% in the defatted rapeseed cotyledons and 0.23-0.54% in the defatted canola cotyledons. In addition, Shahidi and Naczk [75] found that canola meals contained 0.68 - 0.77% condensed tannins. They also

604

OH

OH

Leucocyanidin

Figure 3. Stmctures of basic units of condensed tannins of rapeseed.

reported that the high glucosinate Midas and Hu You 9 Chinese cultivars contained 0.56% and 0.43% of tannins, respectively. The discrepancies in the reported data on tannin contents may be due to the existing differences in solvent systems employed for the extraction and assays used for quantification of tannins. The total content oftannins in selected canola and rapeseed varieties, as determined by vanillin assay, are shown in Table 5. The content of tannins in samples of canola hulls ranged from 48 to 1717 mg tannins per 1 OOg ofhulls, while hulls ofEuropean varieties contained 5 7 -1508 mg tannins per 100 g sample. These results indicate that rapeseed/canola hulls may contain up to eight times more tannins than previously reported by Mitaru et al. [76] and Leung et al. [70]. The data shown in Table 5 also demonstrate that the differences in the tannin levels within canola varieties may range from nine- to fifleen-fold. Price et al.[77] and Butler [78] reported that tannin content in mature sorghum seeds may rangefrom3 % to 93 % ofthe maximum tannin found in immature seeds. On the other hand, Radhakrishnan and Sivaprasad [79] demonstrated that the variation in tannin content of sorghum varieties grown in different locations may range up to eight-fold. Thus, these differences in tannin content within the canola varieties may be due to growing location as well as the stage of seed maturation. Recently, Amarowicz et al. [80] isolated and fractionated the tannins in canola hulls using the method of Strumeyer and Malin [81 ]. In this study, a lyophilized sample of crude tannin extracts isolatedfromCyclone canola hulls was dissolved in 95% ethanol and applied onto a SephadexLH-20 column (2.3 x 40 cm) equilibrated with 95% ethanol. The column was exhaustively washed with 95% ethanol at a flow rate of 60 mL/hr; 6 mL fractions were collected and their absorbance at 280nm was recorded. The column was then eluted with 50% acetone-water at a flow rate of 60

605 Table 5 Sinapine content ranges in some rapeseed/canola varieties (%) Cultivar

Content

Candle 0.39-0.76 Tobin 0.57-0.69 Altex 0.62-0.77 Line 0.79-1.06 Karat 0.81-0.98 Adapted from: J. Pokomy and Z. Reblova, Potrav. Vedy, 13, 1995, p. 157.

<

30

E c g20 —

1—

if)

to

d O 10 -

6

.1

(iijiinn'nm niir 10 i i i i20

^

III

IV

-

^ * i * * * ^

30

40

50

60

Tube number ( 6 ml / tube ) Figure 4. Sephadex LH-20 chromatography ofCyclone canola hull tannins using 50% (v/v) acetone: water.

mL/hr and 6 mLfractionswere collected. Each collected fraction was assayed for tannins by the modified vanillin assay ofPrice and Butler [82]. Approximately 98% ofthe original material applied on to the column was recovered. Crude tannin extract contained about 47% tannins. Figure 4 shows the elution profile ofpurified canola condensed tannins using a Sephadex LH-20 and 50% acetone as eluent. Canola tannins were separated into fourfractions.Similar fractionations of condensed tannins, eluted from Sephadex LH-20 using 70% acetone, were reported by Czochanska et al. [83 ] and Kumar and Horigome [84]. These authorsfractionatedtannins according to their molecular size. The IR spectrum of purified canola tannins obtained with KBr pellets is

606 similar to that reported by Foo [85] for class A procyanidins. Class A procyanidins are polymers that are mainly of procyanidin type with monomers having the cis configuration, i.e. are of the epicatechin type. Each fraction of canola tannins separated on a Sephadex LH-20 column using 50% acetone was examined by the TLC methodology on silica-gel (Sigma) as described by Lea [86]. The TLC chromatogram (Figure 5) revealed the presence of a number of oligomeric proanthocyanidins in Fractions III and IV and the presence of more polymerized (less retained) proanthocyanidins in Fractions I and II. Futhermore, catechins were not detected in canola tannin fractions. Absence of catechins in Cyclone canola tannin extracts was also confirmed by HPLC methodology [87]. The catechin standards used were isolated as described by Amarowicz and Shahidi [88].

1.0

0.5J

Figure 5. TLC chromatogram of Cyclone canola hull tannin fractions separated on a Sephadex LH-20 column using 50% (v/v) acetone: water. Mobile phase: toluene: acetone: formic acid (3:1:1, v/v/v). 1, (-) epigallocatechin; 2, (-) epicatechin-3-gallate; 3, (-) epigallocatechin; and 4, (-) epigallocatechin-3 -gallate.

Condensed tannins with molecular weights of 500 to 3000 Da may bring about the astringent sensation [89] because their phenolic groups are oriented into 1,2-dihydroxy and/or 1,2,3 -trihydroxy configurations [90]. Molecular interpretation offormation of astringency has recently been reported [91,92]. According to Lea [86] and Lea and Arnold [93], the bitterness and astringency of cider procyanidins was a fiinction of their molecular weights. The maximum bitterness was observed for tetramers, while the maximum of astringency corresponded to octamers. Delcour et al. [94] determined the astringency threshold values for solutions of tannic acid (14.1 ppm),(+)catechin (46.1 ppm),procyanidinB-3 917.3 ppm);quercetindihydrateplustetrameric procyanidins (8.9) and a mixture of trimeric and tetrameric proanthocyanidins with (+) catechin (3.6 ppm) in deionized water. A mixture of trimeric and tetrameric procyanidins as well as combination of this mixture with catechin resulted in three to ten times lower threshold values. The contents of total and oligomeric tannins in Cyclone canola hulls (17170 and 3434 ppm, respectively) and corresponding meals (3434 and 687 ppm, respectively) have been determined.

607 In performing this experiment, the content of oligomeric tannins was determined by extraction of aqueous crude tannin solutions with ethyl acetate. According to Porter [95] only monomeric and dimeric proanthocyanidins are highly soluble in ethyl acetate. The content of oligomeric tannins soluble in ethyl acetate was assayed by the modified vanillin method [82]. The crude extracts of Cyclone canola tannins contained 20% of tannins soluble in ethyl acetate [96]. The data shows that the content ofoligomeric tannins in canola meal is a hundred times the threshold value reported for a mixture of trimeric and tetrameric proanthocyanidins. Because of this, condensed tannins present in hulls should be considered as one of the important contributors to the objectionable taste of rapeseed products.

4.

TANNIN-PROTEIN INTERACTIONS

Proteins, one ofthe macrocomponents offood-systems, may interact withflavoringcompounds. Such interactions will influence the flavor release and perception [97- 99]. Phenolic compounds may form soluble and insoluble complexes with proteins. The phenol-protein complexes may be stabilized by covalent bonds, ionic bonds, hydrogen bonding and/or hydrophobic interactions [ 100]. It is, however, believed that phenol-protein complexes are usually the result offormation ofhydrogen bondings and hydrophobic interactions [101], particularly under acidic conditions [103]. Studies on the complexations ofpolyphenols with proteins mainly concentrated on the evaluation of factors influencing these interaction(s) and on the impact of formation of phenol-protein complexes on nutritive value of proteins. The phenol-protein interactions are affected both by the size, conformation and charge of protein molecules and also by the size, length andflexibilityof phenol molecule [100, 101, 103]. Proteins with compact globular conformation like lysozyme andribonucleaseexhibit low affinity for phenols. On the other hand, proteins with conformational^ open structure, like gelatin, readily form complexes with phenols [104]. In addition, it has been demonstrated that proanthocyanidins should have at least threeflavanolssubunits to be an effective protein precipitating agent [ 105-107]. Phenol-protein complexes precipitate from solution only when sufficient hydrophobic surface is formed on the surface of complex [108]. The exact role ofpolyphenol-protein interactions in generating objectionable taste ofrapeseed products is still unclear. Kozlowska and Zademowski [109] and Sosulski and Dabrowski [110] reported that at least seven extractions of rapeseed flour with aqueous ethanol was needed to produce bland protein concentrates containing trace amounts of phenolics. Later, Kozlowska and Zademowski [39] demonstrated the formation of phenolic-protein complexes during preparation of protein isolates. They found that the amount of soluble matter in 80% ethanol, rich in phenolics, increased as the pH of solution used for extraction of proteins increased. Also, Zademowski [52] found that alkaline hydrolyzates of albumin and globulin fractions ofrapeseed contained 6.68 ^imol sinapic acid/g albumin and 0.49 ^mol sinapic acid/g globulin. Recently, Amarowicz and Kmita-Glazewska [111] reported that phenolic acids were bound only to selected low molecular weight rapeseed proteins. Protein precipitating capacities ofcrude condensed tannin extracts isolated from selected canola hulls are shown in Table 6. The protein precipitation assay [112] measures the amount of phenolic bound with protein, whereas the dye labeled BSA assay [113] determines the amount of proteins precipitated by phenols. Protein precipitating values for canola tannins are comparable to those reported by Hagerman and Butler [112] and Asquith and Butler [113] for sorghum tannins.

608 Table 6 Protein precipitating capacities of canola tannins as determined by two assays Sample*

Westar sample 1^ sample 2** sample S""

Protein Precipitation Assay (A5,o) 1.1 4.0

Dye-Labeled BSA Assay (mgBSA/g hulls) 30.7 23.7 58.6

Cyclone 4.5 52.7 sample 1*' 4.9 44.2 sample 4** sample 5^ 52.2 Excel'^ 2.0 33.2 Ebony*^ 5.0 49.2 3.4 45.4 PR3113'^ * Samples codes same as in Table 5 ^ Adapted from: M. Naczk, R. Amarowicz, A. Sullivan, and F. Shahidi, Food Chemistry, submitted for publication. ^ Adapted from: M. Naczk, T. Nichols, D. Pink, and F. Sosulski, J. Agric. Food Chem. 42, 1994, p. 2198 ** Adapted from: M. Naczk, unpublished data Figure 6 shows the effect of pH on the amount of tannins precipitating with selected proteins. BSA, fetuin, gelatin and pepsin were precipitated markedly at pH values between 3.0 and 5.0, but maximum precipitation of lysozyme occurred at pH >8.0. The pH optimum for precipitation was found to be 0.3 -3.1 pH units below the isoelectric points ofproteins [114]. The results demonstrate that each protein has characteristic pH optimum for precipitation by phenolics. Therefore, phenolics may be present in food system in the free or bound form depending on the pH and the kind of proteins present in this system. This in turn will influence the contribution of condensed tannins to the undesirable taste of rapeseed products.

5.

CONCLUSIONS

The bitterness and astringency of rapeseed/canola products may result from both additivity and synergism among the different stimuli present in rapeseed/canola at subthreshold and suprathreshold levels. More detailed research is needed to determine the contribution of each stimulus to the objectionable taste of rapeseed/canola products. The influence of protein-tannins interactions on the perception of objectionable taste of rapeseed/canola product is still not well understood. More detailed research is needed to determine the contribution of these interactions to the perception of the flavor of rapeseed products. Model systems to be used in such studies

609

3

o CO

4

5

6

7

8

9

10 11 12 PH

Figure 6. pH dependence of complex formation between canola tannins and several proteins. Adapted from: M. Naczk, D. Oickle, D. Pink, and F. Shahidi, J. Agric. Food Chem. 44 (1996) 1444. should consist of rapeseed proteinfractionsto whichfreephenolic acids, esterified phenolic acids, tannins or their combinations are added. Phenolic compounds form complexes not only with proteins but also with carbohydrates and minerals [100]. However, the influence of these interactions on the flavor release and perception is still not well understood. Better understanding of factors influencing the interactions between phenolics and other rapeseed/canola components would help in the development of more efficient procedures for production of bland rapeseed protein isolates and concentrates. Acknowledgment This work was supported, in part, by a research grant (to M.N.) from the Natural Sciences and Engineering Research Council (NSERC) of Canada. 6.

REFERENCES E.H. Unger, in: Canola and Rapeseed: Production, Chemistry, Nutrition and Processing Technology, F. Shahidi (ed) pp 235-250. An AVI Book, New York 1990. K. Nishie and M.E. Daxenbichler, Food Cosmet. Toxicol, 18 (1980) 159.

610 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

G.R. Fenwick, E.A. Spinks, A.P. Wilkinson, R.K. Heaney, and MA. Legoy, J. Sci. Food Agric, 37 (1986) 735. S. Rabot, C. Guerin, L. Nugon-Baudon, and O. Szylit, in: Proc. 9th Intern. Rapeseed Congress, Cambridge, UK. Volume 1 (1995) 212. F. Shahidi, M. Naczk, L.J. Rubin, and L.L. Diosady, J. Food Prot., 51 (1988) 743. M. Naczk, L.L. Diosady, and L.J. Rubin, J. Food Sci., 50 (1985) 1685. R. Ohlson, in: Proc. 5th Intern. Rapeseed Congress, Malmo, Sweden (1978) 152. F. Shahidi, M. Naczk, D. Hall, and J. Synowiecki, Food Chemistry, 44 (1992) 283. J. Delisle, J. Amiot, G. Goulet, C. Simard, G.J. Brisson, and J.D. Jones, Qual. Plant.-Plant Foods Human Nutr., 34 (1984) 243. R.K. Downey and J.M. Bell, in: Canola and Rapeseed: Production, Chemistry, Nutrition and Processing Technology, F. Shahidi (ed) pp 37-47. An AVI Book, New York 1990. L. Rubin, L. Diosady, M. Naczk, and M. Halfani, Can. Inst. Food Sci. Technol. J., 19 (1986) 57. F. Shahidi and M. Naczk, in: Canola and Rapeseed: Production, Chemistry, Nutrition and Processing Technology, F. Shahidi (ed.) pp 291-301, An AVI Book, New York 1990. M. Naczk, L.L. Diosady, and L.J. Rubin, Lebensm.-Wiss.u.Technol., 19 (1986) 13. J.W. Erdman, Jr., J. Am. Oil Chem. Soc, 56 (1979) 736. M. Cheryan, CRC Critical Rev. Food Sci. Nutr., 13 (1980) 297. J.A. Maga, J. Agric. Food Chem., 30 (1982) 1. E.R. Morris, in: Phytic Acid: Chemistry and Applications, E. Graf (ed) pp 57-76. Pilatus Press, Minneapolis 1986. L. Hallberg, Scan. J. Gastroenterol., 22 (1987) 73. S.E. Rickard and L.V. Thompson, in: Antinutrients and Phytochemicals in: Food, F. Shahidi (ed) pp 294-312, ACS Symposium Series 662, ACS Washington, DC, 1997. L.U. Thompson, in: Canola and Rapeseed: Production, Chemistry, Nutrition and Processing Technology, F. Shahidi (ed) pp 173-192. An AVI Book, New York 1990. F. Shahidi, unpublished data. L. A. Applequist and R. Ohlson, Rapeseed Cultivation, Composition, Processing and Utilization. Elsevier Publ. Co., Amsterdam 1972. J.M. Bell, Can. J. Animal Sci., 73 (1993) 679. S.K. Nensen, Y.G Liu, andB.O. Eggum, in: Proc. 9thlntern. Rapeseed Congress, Cambridge, UK (1995) 188. J.D. Jones and I.R. Sibbald, Poultry Sci., 58 (1979) 385. A.J. Liesle, J.D. Summers, and J.D. Jones, Can. J. Animal Sci., 53 (1973), 153. G. Sarwar, J.M. Bell, T.F. Sharby, and J.D. Jones, Can. J. Animal Sci., 61 (1981) 719. N.W. Tape, Z.J. Zabry, and K.I. Eapen, Food Sci. Technol., 3 (1970) 78. L.L. Diosady, L.J. Rubin, C.G Tar, and B. Etkin, Can. J. Chem. Eng., 64 (1986) 768. F.W. Sosulski and R Zademowski, J. Am. Oil Chem. Soc, 58 (1981) 96. J.D. Jones and J. Holme, Oilseed Processing, Canadian Patent, No. 1 117 134 (1982). J.J. Baudet, B. Greilsamer, and G. Vermeersch, in: Proc. 6thlntem. Rapeseed Congress, Paris, France (1983) 1381. F.H. Schneider, Method of Shelling Oil- and Protein-Containing Grains, Canadian Patent, No. 1062 118(1979). D.H. Honig, A. Kathleen, L. Selke, and J.J. Rackis, J. Agric. Food Chem., 6 (1979) 1363. J.J. Rackis, D.J. Sessa, and D.H. Honig, J. Am. Oil Chem. Soc, 56 (1979) 262.

611 36 F. Ismail, M. Vaisey-Genser, and B. Fyfe, J. Food Sci., 46 (1981) 1241. 37 M. Donaldson, M. Vaisey-Genser, B. Hoehn, and M. Latta, Cereal Foods World, 23 (1978) 495. 38 W. Boyd, Science, 112 (1950) 153. 39 H. Kozlowska and R. Zademowski, In: Abstracts, Third Chemical Congress ofNorth America, Toronto, June 5-10, 1982. 40 K. Krygier, F. Sosulski, and L. Hogge, J. Agric. Food Chem., 30 (1982) 334. 41 M. Naczk and F. Shahidi, Food Chemistry, 31 (1989) 159. 42 H. Kozlowska, D. A. Rotkiewicz, and R. Zademowski, J. Am. Oil Chem. Soc, 60 (1983) 119. 43 J.J. Rackis, D.J. Sessa, andD.H. Honig, Soybean protein foods, U.S. Dept. Agric, ARS-71-35, 1957, p. 100. 44 S. Arai, H. Suzuki, M. Fujimaki, and Y. Sakurai, Agric. Biol. Chem. (Tokyo), 30 (1966) 364. 45 J.A. Maga and K. Lorenz, Cereal Sci. Today, 18 (1973) 326. 46 M. Dadic and G. Belleau, Proc. Am. Soc. Brew. Chem. (1973) 107. 47 M. Naczk, J.P. Wanasundara, and F. Shahidi, J. Agric. Food Chem., 40 (1992) 444. 48 L.M. Bartoschuk and C.T. Cleveland, Sensory Processes, 1 (1977) 177. 49 T.W. Fenton, J. Leung, and D.R. Clandinin, J. Food Sci., 45 (1980) 1702. 50 J. Pokomy and Z. Reblova, Potrav. Vedy, 13 (1995) 155. 51 R. Amarowicz, F. Shahidi, and R. Pegg, J. Am. Oil Chem. Soc, submitted for publication. 52 R. Zademowski, Acta Acad. Agric Techn. 01st., 21(F) (1987) 1-55; (in polish). 53 U. Wanasundara, R. Amarowicz, and F. Shahidi, Food Chemistry, 42 (1994) 1285. 54 R. Amarowicz and F. Shahidi, J. Am. Oil Chem. Soc, 71 (1994) 551. 55 R. Amarowicz, M. Karamac, B. Rudnicka, and E. Ciska, Fat Sci. Technol., 97 (1995) 330. 56 A.B. Durkee and J.B. Harbome, Phytochemistry, 12 (1973) 1085. 57 B. Tantawy, J.P. Robin, and M.T. ToUier, in: Proc 6th Intern. Rapeseed Congress, Paris, France (1983) 1313. 58 D.R. Clandinin, Poultry Sci., 40 (1961) 484. 59 F.W. Sosulski, J. Am. Oil Chem. Soc, 56 (1979) 711. 60 F. Shahidi and M. Naczk, J. Am. Oil Chem. Soc, 69 (1992) 917. 61 M.M. Mueller, E. Ryl, T.W. Fenton, and DR. Clandinin, Can. J. Animal Sci., 58 (1978) 579. 62 S. Clausen, L.M. Larsen, A. Ploeger, and H. Sorensen, World Crops, 11 (1985) 61. 63 L.M. Larsen, O. Olsen, A. Ploeger, and H. Sorensen, in: Proc 6th Intern. Rapeseed Congress, Paris, France. (1983) 1577-1582. 64 A.B. Durkee and P.A. Thivierge, J. Food Sci., 40 (1975) 820. 65 D.J. Sessa, K. Wamer, and DiH. Hontg, J. Food Sci., 39 (1974) 69. 66 R. Hemingway in: Chemistry and Significance of Condensed Tannins, R. Hemingway and J.J. Karchesy (eds) pp 83-108, Plenum Press, New York, 1989. 67 D.K. Salunkhe, J.K. Chavan, and S.S. Kadam, Dietary tannins: consequences and remedies, CRC Press Inc., Boca Raton, FL, 1990. 68 E.C. Bate-Smith and P. Ribereau-Gayon, Qual. Plant Mater. Vegetable, 5 (1959) 189. 69 A.B. Durkee, Phytochemistry, 10 (1971) 1583. 70 J. Leung, T.W. Fenton, M.M. Mueller, and D.R. Clandinin, J. Food Sci., 44 (1979) 1313. 71 D.R. Clandinin and J. Heard, Poultry Sci., 47 (1968) 688. 72 AOAC Official Methods of Analysis. 10th ed. Association of Official Agricultural Chemists: Washington, DC, 1965. 73 R.G. Fenwick and S.A. Hoggan, Br. Poultry Sci., 17 (1976) 59.

612 74 75 76 77 78 79 80 81 82 83

R. Blair and R.D. Reichert, J. Sci. Food Agric, 35 (1984) 29. F. Shahidi and M. Naczk, J. Food Sci., 54 (1989) 1082. B.N. Mitaru, R. Blair, J.M. Bell, and R.D. Reichert, Can. J. Animal Sci., 62 (1982) 661. M.L. Price, A.M. Stromberg, and L.G. Butler, J. Agric. Food Chem., 28 (1979) 1214. L.G. Butler, J. Agric. Food Chem., 30 (1982) 1087. M.R. Radhakrishnan and J. Sivaprased, J. Agric. Food Chem., 28 (1980) 55. R. Amarowicz, M. Naczk, and F. Shahidi, unpublished data. D.H. Strumeyer and M.J. Malin, J. Agric. Food Chem., 23 (1975) 909. M.L. Price, S. Van Scoyoc, and L.G. Butler, J. Agric. Food Chem., 26 (1978) 1214. Z. Czochanska, L.Y. Foo, R.H. Newman, and L.J. Porter, J. Chem. Soc. Perkin Trans., 1 (1980) 2278. 84 R. Kumar and T. Horigome, J. Agric. Food Chem., 34 (1986) 487. 85 L.Y. Foo, Phytochemistry, 20 (1981) 1397. 86 A.G.M. Lea, J. Sci. Food Agric, 29 (1978) 471. 87 A.C. Hoefler and P. Coggan, J. Chromat., 129 (1976) 460. 88 R. Amarowicz and F. Shahidi, Food Res. International, in press. 89 E. Haslam, in: TheFlavonoids, J.B. Harbome, T.J. Mabry and E. Mabry (eds), Academic Press, New York, 1975. 90 E. Haslam, in: The Biochemistry of Plants, P.K. Stumpf and E.E. Conn (eds) Volume 7, pp 527-556. Academic Press, London, 1981. 91 E. Haslam, T.H. Lilley, and T. Azawa, Bull, de Liason du Groupe Polyphenols, 13 (1986) 352. 92 E. Haslam and T.H. Lilley, CRC Critical Rev. Food Sci. Nutr., 27 (1988) 1. 93 A.G.H. Lea and G.M. Arnold, J. Sci. Food Agric, 29 (1978) 478. 94 J. A. Delcour, M.M. Vandenberghe, P.F. Corten, and PC. Dodeyne, Am. J. Enol. Vitic, 35 (1984) 134. 95 L.J. Porter, in: Methods in Plant Biochemistry, J. Harbome (ed), Volume 1, pp 389-420. Academic Press, San Diego, 1989. 96 M. Naczk, D. Oickle, D. Pink, and F. Shahidi, J. Agric Food Chem., 44 (1996) 2144. 97 J. Bakker, in: Ingredient Interactions — Effect of Food Quality, A. G. Gaonkar (ed) pp. 411 439, Mercel Dekker, Inc., New York, 1995. 98 W. Pickenhagen, Food Technol. Europe, 2 (1996) 60. 99 N. Fisher and S. Widder, Food Technol.., 51 (1997) 68. 100 F. Shahidi and M. Naczk, Food Phenolics: Sources, Chemistry, Effects and Applications, Technomic Publishing Co. Inc., Lancaster, PA, 1995. 101 A.E. Hagerman and L.G. Butler, J. Biol. Chem., 256 (1981) 4494. 102 J.P. McManus, K.G Davis, J.E. Beart, S.H. Gaffney, T.H. Lilley, and E. Haslam, J. Chem. Soc. Perkin Trans., 2 (1985) 1429. 103 M. Naczk and F. Shahidi, in: Antinutrients and Phytochemicals in Food, F. Shahidi (ed) pp 186-208, ACS Symposium Series 662, ACS, Washington, DC, 1997. 104 L.G. Butler, in: Food Products, J. Kinsella and W.G. Soucie (eds), AOCS Press, Champaign, IL, 1989. 105 D.G. Roux, Phytochemistry, 11 (1972) 1219. 106 E.C. Bate-Smith, Phytochemistry, 12 (1973) 907. 107 W.E. Artz, P.D. Bishop, A.K. Dunker, E.G. Schanus, and B.G. Swanson, J. Agric. Food Chem., 35 (1987) 417.

613 108 J.P. McManus, K.G. Davis, T.H. Lilley, and E. Haslam, J. Chem. Soc. Chem. Comm. (1981) 309. 109 H. Kozlowska and R. Zademowski, Proc. 6th International Rapeseed Congress, Paris, France (1983), 1412. 110 K. Dabrowski and F. Sosulski, J. Agric. Food Chem., 32 (1984) 128. 111 R. Amarowicz and H. Kmita-Glazewska, Pol. J. Food Nutr. Sci., 43 (1995) 88. 112 A.E. Hagerman and L.G. Butler, J. Agric. Food Chem., 26 (1978) 809. 113 T.N. Asquith and L.G. Butler, J. Chem. EcoL, 11 (1985) 1535. 114 M. Naczk, D. Oickle, D. Pink, and F. Shahidi, J. Agric. Food Chem., 44 (1996) 1444.