Determination VINCENZO of Plant Phenols by Gel Filtration LATTANZIO ABSTRACT A new method for chromatographic separation and UV spectrophotometric determination of Odiphenol active principles of plants has been developed. The method is based on a technique of gel filtration on Sephadex LH 20. This technique offers selectivity, resolution, and sensibility; minimum detectable 5 pg/ml. Interactions between the gel and substances examined, which control chromatographic separation, were studied. The method is useful for objectively assessingthe quality of plant products and suitable for automation in the caseof routine analysis. INTRODUCTION PHENOLIC COMPOUNDS present in plants are very heterogeneous organic cell components, some of which are universally found in plant cells, while others are widely distributed in some plant organs at surprisingly high concentrations (Harborne et al., 1971, 1975; Lattanzio and Morone, 1979). Phenolic compounds of the hydroxycinnamic acids and flavonoids series are responsible for the biochemical alterations occurring in damaged tissues. In fact, these compounds undergo enzymatic aerobic oxidation through various reaction stages: semiquinones, quinones, and then brown polymers or plant melanins (Marigo et al., 1975; Montedoro, 1975; Piretti, 1975). The reaction products and/or oxidative reaction intermediates may also react with proteins, in conformity with several chemical reactions (Synge, 1978; Van Sumere et al., 1975). As a whole, oxidative reactions of phenolic compounds are classified as “browning” and should generally be prevented during technological processing of plant products. Evaluation of the quality of plant products is useful for both the biochemical characterization of vegetables and greenstuffs eaten fresh, and for the rational choice of industrial technologies suitable for the preservation of fruits and vegetables. Such an evaluation requires easily performed, rapid, objective and also sensitive analytical methods. The research reported here describes a method which permits assay of nontannin phenolic compounds extracted from plant tissues. MATERIALS&METHODS Samples Fresh or plant material was extracted with refluxing 5040% aqueous methanol. The pure compounds were purchased from Fluka (Switzerland), Schuchardt Munchen (West Germany), and Carlo Erba (Italy), and used as such. Tetramethylcatechin 300 mg of catechin was dissolved in 30 ml of methanol. Diazomethane was added slowly, until bubbles ceased to develop. The and AUGUST0 MARCHESINI solution was allowed to stand overnight, and was then evaporated to dryness to remove the excess diazomethane. The product was crystallized from methylene chloride:hexane, and characterized by mass spectrometry (MS). Catechin penta-acetate 300 mg of catechin was dissolved in anhydrous pyridine, then acetic anhydride was added while cooling with ice-water. The solution was allowed to stand overnight, then ice-water was added, the product was extracted with ether, and crystalhzed from methylene chloride:hexane. The product was characterized by MS spectrometry. -Continued on next page A ‘115nm A 3’Xnm Y-III- A (26 o,3mg each dd eluhon volume(ml) 0,4 082 60 d A-180 A 120 180 240 nm 046 0,6 Q4 h 02 Author Lattanzio is affiliated with Centro di Studi sull*Orticoltura Industriale-CNR, Via Amendola, Bari. Author Marchesini is affiliated with lstitu to Sperimentale per la Nutrizione delle Piante, Sezione di Torino, Via Ormea 47, Torino, Italy. 0 Fig. 180 I-Elution of benzoic and cinnamic Volume 46 /1981)-JOURNAL eluhon volume(ml) acid derivatives. OF FOOD SCIENCE-1907 Chromatographicanalysis 20g of SephadexLH-20 resin (Pharmacia,Uppsala,Sweden) were allowed to swell in water for 12 hr (Woof, 1962; Woof and Pierce,1967). The resin was then usedto fill a column(bedvolume 30 x 2 cm), which was equilibratedby passing70 ml of an aqueous ethanol (50%) solution containingapproximitely 3 ml of concentrated H3P04 (pH N 2). Solutions containing 0.1-0.3 mg of each phenol substancein 20-100 ml of aqueousethanol (50%) at pH * 2 (H3P04) were used.The flow rate was 35 ml/hr. In order to obtain a good separation the total amount of phenol substancesusednever exceeded l-2 mg. RESULTS THE RESULTS (Table 1) obtained on eluting standard solutions of phenolic compounds with aqueous ethanol (50% v/v) at pH - 4 (H3P04/NaOH) as eluent show that Lambert Beer’s law holds true in the range of concentrations explored. The specific absorptivity of the phenol substances studied is given in Table 2. Tables 3-5 give the main parameters relative to the separation on Sephadex LH-20 of some derivatives of benzoic acid, cinnamic acid, and of some flavonoids, respectively. Figures l-5 show examples of analyses of phenolic compounds and plant extracts from horticultural products. Figure 1 illustrates elution graphs relative to the assay of benzoic and cinnamic acid derivatives, while Figure 2 gives the elution of some flavonoids. Figure 3 illustrates the analysis of chlorogenic acid and luteolin-7-O-glycoside (Cynaroside) present in Fennel. The analysis was effected spectrophotometrically by reading the absorbances at 325 nm (chlorogenic acid) and 350 nm (cynaroside), corresponding to absorption maxima of these compounds. Figure 4 shows the analysis of polyphenols present in extracts of artichokes. The concentration of each component in the solution placed onto the column can be obtained by measuring the absorbanced at the two wavelengths and applying the following relationships: ~325 = p . cl + $5 . c, A35o = ,Q5’ . Cl + ,;” . c2 where CJ represents the mg of each component in the A A I e.lutm-l vdume(d) Fig. 2-Separation columns. of A quercetin derivatives z cynorln(o,o~~%dryma~b) LH-20 tcvnaroslde(oWidrr dry maker) mobs) A A- 325 nm ----A- 350 nm acid 1 - chiorogenic 2- cynaroelde (0,62x. dry matter) Co,08%dry matter) 1.0 i1II II 0,5- JA 1 0 I-Elution 0.5 2 120 Fig. 3-Spectrophotometric cynaroside on foeniculum Table on Sephades 350nm 3% nm ---chloroqenlc aG(&lA%dry matter) t Scobmosde(o,ss?i 180 eluhon volume (ml) determination vulgare leaves. of different concentrations of chlorogenic Chlorogenic Cynarine Caffeic Scolymoside Cynaroside 0.2 0.3 Ve(ml) OD325 Va(ml) Velml) OD325 Va(ml) Ve(ml) 130 174 217 117 170 0,23 0,23 0,29 0.09 39 41 52 30 132 175 217 117 0.11 39 180 0.44 0,43 0.60 0,18 0,22 49 50 61 33 42 130 174 219 120 171 46 (1981/-JOURNAL of polyphenols present in extracts of artichokes. of standardsa a Ve(ml) = elution volume; A325 = absorbance collection of phenolic substances). 1908-Volume eluhon volume (ml) Fig. *-Analysis 0.1 Sample i )’ acid and 325 nm; Va = accumulation OF FOOD SCIENCE 00325 0.4 Va(ml) 0,69 0,70 0,82 0,28 0,33 46 54 65 45 51 volume (represents 0.5 Ve(ml) OD325 Va(ml) 128 176 214 117 173: 0,84 0,90 54 58 1,lO 71 0.38 0,43 48 51 the volume of eluate Ve(ml) 130 174 221 117 165 necessary OD325 Vatml) 1.10 57 1.14 1.40 0,45 0,52 67 81 50 55 for quantitative DETM PLANTPHENOLS solution volume and E: its specific absorptivity (the spectrophotometer cell path is 1 cm). Figure 5 shows the elution graph relative to the assay of chlorogenic acid from extracts of egg plants. It can be noted that ascorbic acid .is also present in the extract. This compound was added during the blanching of slices of egg plants to avoid browning of the product as the result of enzymatic oxidation of polyphenols. The purity of the phenolic compounds isolated by chromatography on Sephadex LH-20 was tested by HPLC, using a Perkin Elmer Model LO55 liquid-liquid chromatograph. The chromatographic column was prepacked with pBondapak/Cts ; the eluent gradient was methanol, water, and acetic acid. In each case, only one peak was visible, with referance to the pure product. Each substance was also characterized by its visible and UV spectra. DISCUSSION THE RESULTS given in Tables 1-3 show that elution of phenolic compounds is essentially regulated by electrons of the benzene ring, as well as by phenolic -OH groups, which are responsible for the hydrogen bonds between solute and matrix. Other types of interaction may be due to the presence of unsaturated carbonyl or carbonyl groups in the solute molecules. The molecular weight has less influence than the above parameters. Finally, the position of the substituent in the chromatic ring is of little importance: for instance, cathecol and hydroquinone have the same elution value. As regards the effect of the substituent on the K, value, it can be seen that the phenolic -OH groups increase the retention to a greater extent in the flavonoid series than in that of benzoic and cinnamic derivatives. The presence of an -0CHs group exerts a negative effect on retention; for instance (Table 2), ferulic acid is eluted before coumaric acid and with a Kav equal to that of cinnamic acid. Similarly (Table 3), vanillic acid has a K, equal to that of ben- . zoic acid, while syringic acid is eluted before benzoic acid. This negative effect of the methoxyl group on retention can be explained by the formation of an intramolecular hydrogen bond with a neighboring hydroxyl group, in competition with the hydrogen bond between solute and matrix, or else by increased steric hindrance. The effect of the molecular weight on retention is best seen considering the separation of glycosides of flavonoids (Table 5), which have a K, less than that of aglycones. It is -Continued Table 2-Angular coefficients of A/mg on page 19 17 curve Standard E pH 4 (325) E pH 4 (325) Chlorogenic acid Cynarine Caffeic acid Scolymoside Cynaroside 2,224 2,263 2,858 0.936 1,085 1,512 1,382 - Table 3-Separation LH-20 column? of some benzoic 1,582 1,866 acid derivatives acid Vehll Mol. wt K av Benzoic acid Vanillic acid Syringic acid Gallic acid 156 150 129 189 122 168 198 170 1.67 1.60 I,34 2.07 Phenolic on Sephadex @h/2hl) 15 18 21 20 a Kav = coefficient of Partition between the liouid and the gel phase; K,, = (Ve-Vo)/(Vt-Vo): Ve = volume for elution of the substance; Vt = total volume of chromatographic bed; Vo = void volume; @h12 = width of elution peak at half height. Table *-Separation L H-20 columnsa Phenolic acid Cinnamic acid Coumaric acid Ferulic acid Caffeic acid Chlorogenic acid Cynarine L-DOPA A BY GEL FILTRATION.. of some cinnamic acid derivatives on Sephadex V,fml) Mol. wt K av @h/2(mf) 171 192 171 223 130 175 69 148 164 194 180 354 516 197 1.85 2.10 1.85 2.48 1.30 1.90 0.83 15 21 21 24 18 20 12 aK av = coefficient of Partition between the liquid and the gel phase; K av = (Ve-Vo)/(Vt-Vo); Ve = volume for elution of the substance: Vt = total volume of chromatographic bed; Vo = void Volume; @h/2 = width of elution peak at half height. to --w-m--m A 250 nm A 325 nm I - ascorbrc 2- 015 chlorogenic Table 5-Separation Flavonoid acid acid Quercitrin Scolymoside Cynaroside Rutin tetra-OMe-Catechin Catechin Dihydroquercetin Spigenin F&tin Quercetin I Fig. 5-Analysis evolutton 110 of egg plant treated with ascorbic v&me acid. Myricetin Kaempferol Catechin pentacetate of some flavonoids on Sephadex L H-Mcolumns V,fml) Mol. wt K av dh/2(m)) 234 117 172 135 179 260 354 462 555 630 710 570 115 448 594 448 610 346 290 304 270 286 302 318 286 500 2.61 1.25 1.87 1.52 1.95 2.92 4.06 5.36 6.48 7.39 8.35 6.70 1.23 24 18 18 24 20 30 30 40 50 63 70 57 18 Volume 46 (1981)-JOURNAL OF FOOD SCIENCE- 1909 GREA T NOR THERN BEAN STARCH. wheat flour-starch blends. They partially attributed this increase in water absorption to the surface area of the starch phase and excessive dilution at high concentrations of starch, of the continuous gluten phase. Our observations suggest that, in addition to these factors, the water absorption capacity of the native starch may also have an important role in water retention by wheat flour-starch blends. REFERENCES Davies, T., Miller, D.C.. and Procter, A.A. 1980. Inclusion complexes of free fatty acids with amylose. Starke 32: 149. Goering. K.J., Jackson, L.L., and DeHaas, B.W. 1975. Effect of some nonstarch components in corn and barley starch granules on the viscosity of heated starch-water suspensions. Cereal Chem. 52: 493. Goto, F. 1972. Determination of gelatinization property of highly concentrated starch suspensions by Brabender plastographs 3. Effects of fatty acids upon plastograms. J. Jap. SIX. Starch Sci. 19: 76. Heckman, E. 1977. Starch and its modification for food industry. In “Food Colloids.” Ed. Graham. H.D. The Avi Publishing Company, Inc., Westport, CT. Jackson. G.R. and Landfried. B.W. 1965. The effect of various elvcerides on the baking properties of starch doughs. Cereal Ch&. 42: 323. Kruger, L.H. and Rutenberg, M.W. 1967. Production and uses of starch acetates. In “Starch: Chemistry and Technology,” Ed. Whistler, R.L. and Paschall, E.F. Academic Press. New York. Leach, H.W. 1965. Gelatllization of starch. In “Starch: Chemistry and Technology.” Ed. Whistler. R.L. and Paschall. E.F. Academic Press, New York. Leach, H.W., McCowen, L.D.. and Schoch. T.J. 1959. Structure of the starch granule. 1. Swelling and solubllity patterns of various starches. Cereal Chem. 36: 534. Langley, R.W. and Miller, B.S. 1971. Note on relative effects of monoglycerides on the gelatinization of wheat starch. Cereal Chem. 48: 81. Medcalf, D.G., Youngs, V.L., and Gilles, K.A. 1968. Wheat starches. 2. Effect of uolar and nonuolar liaid fractions on _nastine - characteristics. Cereal Chem. 45: 85. Melvin, M.A. 1979. The effect of extractable lipid on the viscosity characteristics of corn and wheat starches. J. Sci. Food Agric. 30: 731. Mercier, C., Charbonniere, R.. Grebaut,J., and de La Gueriviere, J.F. 1980. Formation of amylose-lipid complexes by twin-screw extrusion cooking of manloc starch. Cereal Chem. 57: 4. DETM PLANTPHENOLS BY GEL FILTRATION.. .. Ohashi. K.. Goshima. G.. Kusada. H.. and Tsuge. H. 1980. Effect of embraced lipid on the gelatinizat&of rice starch. Starke 32: 54. Orthoefer, F.T. 1976. Effect of type of fat on starch pastes containing glycerol monostearate. Cereal Chem. 53: 561. Osman, E.M. and Dix, M.R. 1960. Effects of fats and nonionic surface active aeents in starch oastes. Cereal Chem. 37: 464. Rasper, V.F. &I DeMan. J.M. 1980. Measurement of hydration capacity of wheat flour/starch mixtures. Cereal Chem. 57: 27. Sandstedt, R.M. and Abbott, R.C. 1964. A comparison of methods for studying the course of starch gelatinization. Ceral Sci. Today 9: 13. Sathe, S.K., Ponte, J.G. Jr., Rangnekar, P.D., and Salunkhe, D.K. 1981. Effects of addition of the Great Northern bean (Phaseolus vulgaris L. ) flour and protein concentrates on rheological properties of dough and baking quality of bread. Cereal Chem. 58: 97. Sathe. S.K. and Salunkhe, D.K. 1981a. Solubilization and electrophoretic characterization of the Great Northern bean (Phaseolus vulgaris L.) proteins. J. Food Sci. 46: 82. Sathe. S.K. and Salunkhe. D.K. 1981b. Functional properties of the Great Northern bean (Phaseolus vulgaris L.) proteins. Emulsion, foaming, viscosity, and gelation properties. J. Food Sci. 46: 71. Sathe, S.K. and Salunkhe, D.K. 1981c. Studies on trypsm and cnymotrypsin inhibitory activities, hemagglutinating activity, and sugars in the Great Northern beans (Phaseolus vulgaris L.). J. Food Sci. 46: 626. Sathe, S.K. and Salunkhe, D.K. 1981d. Isolation, partial characterization, and modification of the Great Northern bean (Phaseolus vulgaris L.) starch. J. Food Sci. 46: 617. Schoch, T.J. 1964. Swelling power and solubility of granular starches. In “Methods in Carbohydrate Chemistry,” Ed. Whistler. R.L.. Vol. 4. Academic Press, New York. Yamazaki, W.T. 1953. An alkaline water retention capacity for the evaluation of cookie baking potentialities of soft winter wheat fours. Cereal Chem. 30: 242. Yamazaki. W.T., Donelson, J.R., and Kwolek, W.F. 1977. Effects of flour fraction composition on cookie diameter. Cereal Chem. 54: 352. Yasumatsu. K. and Moritaka, S. 1964. Changes of characteristics of starch during gelatinization in the presence or absence of fatty acids. J. Food Sci. 29: 198. MS received 10/7/80; revised 5/21/81; accepted 6/13/81. Presented at the 41st Annual Meeting of the Institute of Food Technologists, June 7-10,1981. Atlanta, GA. Journal Paper No. 2616 of the Utah Agricultural Experiment Station and a contribution of Western Regional Project W-150. The authors thank Mr. P.D. Rangnekar, Dept. of Grain Science & Industry, Kansas State Univ., Manhattan, KS. for the kind assistance ln viscoamylographic studies. _ From page 1909 worth noting that the latter class of compounds indicates an interesting effect due to the electronic unsaturation of the molecule. For molecules of the same molecular weight, that presenting the greatest unsaturation is delayed most. For instance, comparing dihydroquercetin and catechin, the presence of a C=O group conjugated with a C=C double bond in dihydroquercetin leads to a greater retention of this compound compared to catechin. Quercetin with a further unsaturated C=C bond, has a still higher K,. Tetramethoxycatechin and catechin pentacetate are further examples of the negative effect of seric hindrance on retention. In conclusion, the methodology developed allows highly simplified analysis of phenolic compounds in plants, the chromatographic separation of these compounds proceeding in strict correlation with the molecular structure. The HPLC method is more rapid in operation and posses greater analytical sensitivity than the gel filtration method. The advantages of the latter are linked to the possibility of operating with simple equipment accessibleto all laboratories and suitable for automation in the caseof routine analyses, for instance, relative to the quality of agricultural products. REFERENCES Harborne, J.B., Boulter. D., and Turner. B.L. 1971. In “Chemotaxonomie of the Legtmiinosae.” Academic Press. London. Harborne, J.B., Mabry, T.J., and Mabry. H. ‘1975. In “The Flavonoids.” Chapman & Hall. London. Lattanzio. V. 1977. Determinhxuione dei principi attivi polifenolici de1 carciofo (Cynara Scolymus L.) per mezzo della gel cromatografia su Sephadex LH-20. Ind. Conserve, 152: 316. Lattanzio. V. and Morone, I. 1979. Variation of the orthodiphenol content of Cynara Scol~mus L. durine - the olant -~ -arowing I seasons. Experientia 35(8): 993.N., and Bouldet, A.N. 1975. Metabolisme Marlgo, G., Rossignol, des esters hydroxycinnamiaues et de l’acide quinique. Bull. Liaison “Goupe Polyphenols” 6: 308. Montedoro, G. 1975. Effets des traltements a les pates d’olives par additifs absorbants les composes phknoliques sur les processus d’exteraction mechaniques de l’huile. Bull. Liaison “Groupe Polyphenols” 6: 259 Piretti, M.V. 1975. Le sostanze fenoliche negli alimenti di origine vegetale. Ed. Cisalpino Goliardlca, Milan. Synge. R.L.M. 1978. Polyphenol proteins reaction and their signficance for agricultural practices. BulI. Liaison “Groupe Polyphenols” 8: 13. Van Sumere, C.F., Albrecht, J., Dedonder, A., De Pooter. H., and PC I 1975. “The Chemistry and Biochemistry of Plant Proteins.” Aiaademic Press, London. Woof, J.B. 1962. Investigation of phenolic components of brewing materials by gel filtration. Nature 195: 184. Woof, J.B. and Pierce, J.S. 1967. Separation of complex mixtures of polyhydrory phenols on columns of Sephadex. J. Chromat. !28:~94. MS received 7/11/80; revised 3/g/81; accepted 3124181. Volume 46 (1981kJOURNAL OF FOOD SCIENCE- 1917