首页   按字顺浏览 期刊浏览 卷期浏览 Gas chromatographic–mass spectrometric characterization of flavanones in citrus a...
Gas chromatographic–mass spectrometric characterization of flavanones in citrus and grape juices

 

作者: Colin S. Creaser,  

 

期刊: Analyst  (RSC Available online 1992)
卷期: Volume 117, issue 7  

页码: 1105-1109

 

ISSN:0003-2654

 

年代: 1992

 

DOI:10.1039/AN9921701105

 

出版商: RSC

 

数据来源: RSC

 

摘要:

ANALYST, JULY 1992, VOL. 117 1105 Gas Chromatographic-Mass Spectrometric Characterization of Flavanones in Citrus and Grape Juices Colin S. Creaser, Mohammed R. Koupai-Abyazani" and G. Richard Stephenson School of Chemical Sciences, University of East Anglia, Norwich NR4 7TJ, UK A method for the characterization of flavanones in fruit juices, involving solvent extraction, hydrolysis to the corresponding aglycones, trimethylsilylation and combined gas chromatography-mass spectrometry, is reported. The application of the method is demonstrated for the analysis of orange, lemon, grapefruit and grape juices. Keywords: Gas chromatograph y-mass spectrometry; trimeth ylsilyl derivatives; flavanones; fruit juices The flavonoids of citrus fruits have been extensively investi- gated because of their pharmacological activity, flavour impact on citrus juices and value as by-products of the citrus industry.1 The principal citrus flavanones, naringenin (I), hesperetin (II) and eriodictyol (111), do not occur in juices as the free aglycones, but are combined through the C-7 hydroxy group with a sugar component ,1 either P-neohesperidose (IV) or p-rutinose (V). Bitterness in some citrus fruits is attributed to the flavanone neohesperidosides, while flavanone rutino- sides are tasteless.24 2' 3kR3 I: R1, R2, R4 = OH; II: R1, R2, R3 = OH; 111: R1, R*, R3, R4 = OH R3 = H R4 = OMe CH20H OH OH IV OH W The distribution of these flavanone glycosides, and of the flavanone aglycones derived from them, is characteristic of many citrus fruits and juices, and also affects the quality of processed citrus products.5 Analysis for flavanone content has, therefore, been proposed as a means of characterizing the authenticity of lemon juice,6 of measuring the adulteration of citrus juices7 and of identifying the presence of orange juice in fruit drinks.8 * Present address: Department of Botany, University of British Columbia, Vancouver, Canada V6T 1ZK. Several techniques have been used for the determination of flavanones in citrus juices.The most widely used method is based on the spectrophotometric measurement of the yellow colour produced by flavanone glycosides in alkaline solution .9 Although more selective methods have been developed for flavanone determination, the Davis method9 is still used for those analyses where its simplicity outweighs its suscepti- bility to interference.Methods involving combinations of extraction, adsorption chromatography, paper chroma- tography, thin-layer chromatography and spectrophotometry have all been reported for the separation and identification of flavonoids in citrus and other fruit juices.10-15 High-perfor- mance liquid chromatography (HPLC) has been applied to the determination of polymethoxylated flavonoids in citrus juices16717 and citrus peel.18 The adoption of HPLC for the analysis of mixtures of these compounds greatly improves the speed and accuracy of their identification and determination, but affords poor resolution for complex mixtures of flavo- noids. Capillary gas chromatography (GC) and gas chroma- tography-mass spectrometry (GC-MS) procedures have been demonstrated to yield improved separations for flavanone aglycone trimethylsilyl (TMS) derivatives,19-*1 but they have not been applied to fruit juices.In this paper, a sensitive and selective method is reported for the identification of flavanone glycosides in fruit juices, as the corresponding aglycones, which involves use of solvent extraction, hydrolysis, trimethylsilylation and combined GC- MS. The application of the method is demonstrated for the analysis of orange, lemon, grapefruit and grape juices. Experimental Reagents and Materials The solvents ethanol and diethyl ether (Aldrich, Gillingham, Dorset, UK) were of HPLC grade and ethyl acetate (Fisons, Loughborough, Leicestershire, UK) was of Distol grade. Anhydrous pyridine (Pierce, Rockford, IL, USA) was of silylation grade, and anhydrous sodium sulfate (Fisons) was obtained at 99.5% purity.The silylating reagents, hex- amethyldisilazane (HMDS) (98% ) and trimethylchlorosilane (TMCS) (98%), were obtained from Aldrich. Flavanone standards were purchased from Apin Chemicals (Abingdon, Oxfordshire, UK). Procedure A 100 cm3 sample of hand-squeezed, clarified orange juice was extracted with 2 x 100 cm3 of diethyl ether in a 500 cm3 separating funnel. The ether extracts were discarded. The aqueous phase was then extracted with 4 X 100 cm3 of ethyl acetate, and the organic phase was evaporated to dryness under reduced pressure at 35-40 "C. The residue was taken up1106 100 50 ANALYST, JULY 1992, VOL. 117 -(el - in 10 cm3 of ethanol; 5 cm3 of 5% HCl were added to a 3 cm3 portion of this solution, which was heated at 100 "C for 2 h in an oil-bath722 cooled, and extracted with 3 x 10 cm3 of ethyl acetate.The extracts were combined, dried through a plug of anhydrous sodium sulfate ( 3 3 x 2 cm i d . ) and evaporated to dryness. The residue was dissolved in 0.2 cm3 of pyridine, 0.2 cm3 of HMDS and 0.1 cm3 of TMCS, and the mixture was heated at 60 "C overnight.20.21 Samples (100 cm3) of hand-squeezed lemon, grapefruit and grape juices were extracted and pre-treated by the same procedure as that described for orange juice. Aliquots of each fruit juice were also spiked with standard naringenin, hespere- tin and eriodictyol at levels of 2 4 pg cm-3. GC-MS Analysis The derivatized samples were analysed by use of a Varian 3400 gas chromatograph (Palo Alto, CA, USA) directly interfaced with a Finnigan-MAT (San Jose, CA, USA) ion trap mass spectrometer operated via an AT/personal computer. The GC separation was carried out on a capillary column (50 m x 0.25 mm i.d.) of RSL 200 BP (0.2 pm film thickness), from Alltech (Carnforth, Lancashire, UK), under the following conditions: helium carrier gas pressure, 9 psi (62 kPa); injector and transfer line temperatures, 300 and 280 "C, respectively; oven temperature programme, 130 "C for 0.5 min, then heated to 235 "C at 30 "C min-1 and from 235 to 290 "C at 1 "C min-1.The MS conditions were electron ionization under automatic gain control at a trap temperature of 150 "C; scan speed, 1 scan S-1. Results and Discussion The polyhydroxylated flavanone glycosides present in fruit juices are insufficiently volatile for direct GC separation, and require hydrolysis to the corresponding aglycones and derivat- ization prior to analysis.Trimethylsilylation has been used successfully for this class of compounds by several groups of workers.19JO The flavanone aglycones, unlike the other flavonoids, yield a mixture of the TMS derivatives of the flavanone (e.g., VI for eriodictyol) and the corresponding chalcone (VII) in the presence of HMDS and TMCS under mild derivatizing conditions.20721 However, heating the reac- tion mixture at 60 "C overnight results in quantitative conversion into the TMS derivative of the chalcone. (CH3)3Si0 0 VI (CH3)3Si0 (CH3),Si0 0 VII Hydrolysis of the flavanone glycosides, followed by TMS derivatization of the aglycones, provides, therefore, a conve- nient method for the characterization of the glycosides, as the corresponding TMS chalcone derivatives, by GC-MS. The orange, grapefruit, lemon and grape juice extracts were prepared by a straightforward extraction procedure and were readily hydrolysed with HCl to yield the flavanone aglycones.The aglycones were converted into the TMS derivatives of the corresponding chalcones under the derivatization conditions used. The flavanones were then identified by comparison of the GC retention times and mass spectra of the TMS chalcone derivatives with those of standards prepared from the flavan- one aglycones after derivatization under equivalent condi- tions. Juice extracts were also spiked with standards prior to hydrolysis and derivatization to confirm GC retention times and mass spectral assignments.100 (=) 50 I 1 i 3 I 400 800 1200 1600 2000 Scan number Fig. 1 Selected ion chromatograms (m/z 545-633) for (a) a standard mixture of flavanone aglycones after TMS derivatization; ( b ) the TMS derivatization of oran e juice extract; (c) the TMS derivatization of lemon juice extract; 6) the TMS derivatization of grapefruit juice extract; and (e) the TMS derivatization of grape juice extract. 1, Naringenin; 2, hesperetin; and 3, eriodictyolANALYST, JULY 1992, VOL. 117 - 575 562 267 307 369 473 147 57 105 ] 179 1-1 - I . . . I I , 1 I 1 I I 1107 loo 50 Table 1 GC-MS retention times and limits of detection for TMS chalcone derivatives of flavanone aglycones Limit of Compound time/min pg cm-3 Retention detection*/ Naringenin 23.31 0.02 Hesperetin 26.47 0.02 Eriodict yo1 27.59 0.03 * Based on a signal-to-noise ratio of 2 : 1.( C) -73 - 73 267 353 532 575620 147 95 I 633 100 200 300 400 500 600 mlz Fig. 2 EI mass spectra of the TMS chalcone derivatives of (a) peak 1, naringenin; (b) peak 2, hesperetin; and (c) peak 3, eriodictyol from lemon juice extract [Fig. l(c)] The mass spectra of the TMS chalcone derivatives of naringenin, hesperetin and eriodictyol are characterized by a prominent [M - 151' ion at mlz 545, 575 and 633, respec- tively.23 These ions are the base peaks in each spectrum. The selected ion chromatogram for the sum of ion intensities in the range mlz 545-633, for the TMS chalcone derivatives obtained from a mixture of naringenin, hesperetin and eriodictyol, is shown in Fig.l(a). Table 1 gives data on retention times and limits of detection, obtained by GC-MS, for these com- pounds. Analysis for the TMS derivatives in hydrolysed orange juice extract , by GC-MS and co-chromatography with appropriate reference compounds, showed the presence of naringenin and hesperetin , but eriodictyol was not detected [Fig. l(b)]. Typical profiles for the TMS chalcone derivatives of flavanone aglycones derived from lemon and grapefruit juices are shown in Fig. l(c) and ( d ) , respectively. These ion Table 2 Concentrations and recoveries of flavanone aglycones in orange, lemon, grapefruit and grape juices Flavanone concentration*/pg ~ m - ~ Fruit juice Naringenin Hesperetin Eriodictyol Orange 0.46 (90)t 1.04 (90) NDS (86) Lemon 0.15 (88) 0.58 (87) 0.17 (86) Grapefruit 0.61 (90) ND (88) ND (85) Grape ND (87) ND (85) ND (83) * Repeatability, 9.4% in the range 0.4-1.0 pg ~ m - ~ .t Recoveries (Yo) in parentheses, for samples spiked at 2 4 $ ND = Not detected. pg cm-3. a C m -0 3 : 50 .- a > .- + - 01 n 0 100 200 300 400 500 600 mlz Fig. 3 EI mass spectrum of trihydroxymonomethoxyflavanone TMS derivative (peak 4) in the chromatogram obtained from lemon juice extract [Fig. l(c)] chromatograms establish that naringenin, hesperetin and eriodictyol are all present in lemon juice, while the predomi- nant flavanone component of grapefruit juice is naringenin. In contrast to orange, lemon and grapefruit juices, no flavanones were detected in grape juice at a level above the GC-MS limits of detection [Fig.l(e)]. Chalcone production from free flavanone aglycones was not detected in any of the juices after TMS derivatization of unhydrolysed extracts, confirming that the flavanones are present only as the glycosides in the fruit juice extracts examined. The mass spectra of the TMS derivatives of flavanones extracted from lemon juice are shown in Fig. 2. These match the spectra of the TMS chalcone derivatives of standard samples.23 Table 2 gives the flavanone concentrations found in orange, lemon, grapefruit and grape juices expressed in terms of the corresponding free flavanone aglycone. These data, based on peak area measurements, are the average of values obtained by using external standard calibration and standard additions procedures.Good agreement was obtained for both methods. Recoveries of naringenin, hesperetin and eriodictyol were in the range 83-90% for spiked fruit juices (Table 2). The chromatogram obtained after TMS derivatization of the lemon juice extract showed an unidentified peak (4), with a retention time of 27.4 min, in addition to peaks correspond- ing to naringenin, hesperetin and eriodictyol [Fig. l(c)]. The mass spectrum of this compound is shown in Fig. 3. The presence of an ion at mlz 575 in the mass spectrum of this peak (Fig. 3), tentatively assigned to the [M - 15]+ fragment ion, indicates that this is the TMS derivative of a tetrahydroxymo- nomethoxychalcone [Scheme 1, (b)] formed from a trihy- droxymonomethoxyflavanone [Scheme 1, (a)] .23 The [M - CO]+' ion at mlz 562 is a characteristic ion for chalcones.The ion at mlz 369 indicates that three of the hydroxy groups are in the A-ring [Scheme 1, (b)], of which one must be at the C-5 position of the original flavanone [Scheme 1, (a)].23 This fragment, which contains the intact A-ring, corresponds to the A2+ ion for underivatized chalcones,24.25 and is formed by the retro-Diels-Alder cleavage of the molecular ion. The absence1108 ANALYST, JULY 1992, VOL. 117 \ OH 0 Derivatization (CH3)3Si0 0 0 0’ ‘si’ C< ‘CH, [M - 15]+ rnlz 575 mlz 369 Scheme 1 OCH3 OSi(CH3)3 [M - CO]” mlz 562 of an [M - 88]+’ ion indicates that there are no adjacent TMS groups in the A-ring. On the basis of the observed ions, the original flavanone is assigned as the trihydroxymonomethoxy- flavanone [Scheme 1, (a)].23 Conclusion Analysis by GC-MS of the TMS derivatives of hydrolysed flavanone glycosides provides a sensitive and selective method for the characterization of these compounds in fruit juices, and for the identification of flavanones for which reference standards are not available.Detection limits and recoveries for the TMS derivatives of the aglycones are satisfactory for their determination in fruit juices. References 1 Horowitz, R. M., and Gentili, B., in Citrus Science and Technology, eds. Nagy, S., Shaw, P. E., and Veldhuis, M. K., AVI Publishing Co., Westport, CT, 1977, vol. 1. Horowitz, R. M., and Gentili, B., Tetrahedron, 1963, 19, 773. 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Horowitz, R.M., in Biochemistry of Phenolic Compounds, ed. Harborne, J. B., Academic Press, New York, 1964. Horowitz, R. M., and Gentili, B., J. Agric. Food Chem., 1969, 17, 696. Kefford, J. F., and Chandler, B. V., The Chemical Constituents of Citrus Fruits, Academic Press, New York, 1970. Rolle, L. A., and Vandercook, C. E., J. Assoc. Off. Anal. Chem., 1963,46, 362. Wagner, D., and Monselise, J. J., Isr. J. Technol., 1963, 1, 33; Chem. Abstr., 1964, 60, 6140a. Koch, J., and Haase-Sajak, E., Dtsch. Lebensm.-Rundsch., 1965, 61, 199. Davis, W. B., Anal. Chem., 1947, 19, 476. Albach, R. F., and Redman, G. H., Phytochemistry, 1969, 8, 127. Nishiura, M., Esaki, S., and Kamiya, S., Agric. Biol. Chem. (Tokyo), 1969,33, 1109. Vandercook, C. E., and Rolle, L. A., J.Ass,i. Off. Agric. Chem., 1963,46, 359. Yokoyama, H., J. Assoc. Off. Agric. Chem., 1965,48, 530. Hagen, R. E., Dunlap, W. J., Mizelle, J. W., Wender, S. H., Lime, B. J., Albach, R. F., and Griffiths, F. P., Anal. Biochem., 1965, 12, 472. Dunlap, W. J., Hagen, R. E., and Wender, S. H., J. Food Sci., 1962, 27, 597.ANALYST, JULY 1992, VOL. 117 1109 16 Sendra, J. M., Navarro, J. L., and Izquierdo, L., J. Chroma- togr. Sci., 1988, 26, 443. 17 Heimhuber, B., Galensa, R., and Herrmann, K., J. Chroma- togr., 1988,439,481. 18 Park, G. L., Avery, S. M., Byers, J. L., and Nelson, D. B., Food Technol, 1983, 37, 98. 19 Greenaway, W., English, S., Wollenweber, E., and Whatley, F. R., J. Chromatogr., 1989,481, 352. 20 Creaser, C. S., Koupai-Abyazani, M. R., and Stephenson, G. R., J. Chromatogr., 1989, 478,415. 21 Creaser, C. S., Koupai-Abyazani, M. R., and Stephenson, G. R., J. Chromatogr., 1991,586, 323. 22 Coffin, D. E., and Dupont, J. E., J. Assoc. Off. Anal. Chem., 1971, 54, 1211. 23 Creaser, C. S., Koupai-Abyazani, M. R., and Stephenson, G. R., Org. Mass Spectrom., 1991, 26, 157. 24 Mabry, T. J., and Markham, K. R., in The Flavonoids, eds. Harborne, J. B., Mabry, T. J., and Mabry, H., Chapman and Hall, London, 1975. Mabry, T. J., and Ulubelen, A., in Biochemical Application of Mass Spectrometry, eds. Waller, G . R.. and Dermer, 0. C., Wiley-Interscience, New York, 1973. 25 Paper 210071 1 H Received February 11, 1992 Accepted March 6, 1992

 

点击下载:  PDF (447KB)



返 回