- Email: [email protected]
A selective spectrophotometric method for determination of quercetin in the presence of other flavonoids
Taianra,Vol. 39,No. 3, pp, 259-263,1992 Printedin GreatBritain.All rightslescrd
0039-9140/92 M.OO+ 0.00 copyrightQ 1992Pagmon Prua pk
A SELECTIVE SPECTROPHOTOMETRIC METHOD FOR DETERMINATION OF QUERCETIN IN THE PRESENCE OF OTHER FLAVONOIDS HASSAN F. ASKAL and GAMALA. SALEH
Departmentof Pharmaceutical Chemistry, Faculty of Pharmacy, Assiut University, Assiut, Egypt ENMM Y. BACKEEET* Departmentof Pharmacognosy, Faculty of Pharmacy, Assiut University, Assiut, Egypt (Received 13 March 1991. Revised 4 July 1991. Accepted 4 July 1991) Summary-A simple, rapid and highly selective method for determination of quercetin in the presence of other flavonoids was developed. The method is based on the oxidation reaction of quemetin in neutral aqueous solution with N-bromosuccinimide (NBS) in the presence of phenol to give a violet chromogen measurable at 510 nm. Beer’s law was valid within a concentration range of 2.5-30 p&l with a good correlation coetEcient (r = 0.9990). All variables were studied to optimize the reaction conditions. The method is highly selective for quercetin. Other investigated flavonoids do not interfere. Mixtures of tlavonoids were also analysed through native W measurements of absorbance readings at 370 mn and then at 510 mn after adopting the proposed procedure. The method could also be utilixed for the quantitative determination of quercetin in some plant extracts. Moreover, the proposed procedure could be considered as a good tool to follow the hydrolysis of quercetin glycosides.
The quantitative analysis of flavonoids by ultraviolet absorption is well known.‘” Other methods for their determination include fluorimetry,’ polarography,8 densitometry9 and HPLC.‘&” Few calorimetric methods have been reported for their determination.iG2’ Most colorimetric methods cannot be used to distinguish between the different flavonoids. When interfering flavonoids are to be present, quercetin must first be isolated chromatographically’ but this requires an empirical correction for adsorption of quercetin. Recent studies have shown that, flavonoids have mutagenic activity.“-24 Quercetin is the most mutagenic; it also acts as a carcinogenic agent towards rats.25 There has been a need for a selective colorimetric method which could best be utilized in detecting and distinguishing between quercetin and other flavonoids without the need for any pretreatments or prior separation. N-Bromosuccinimide (NBS) has been previously used for the analysis of hydroxy compounds.26*27 However, no report is available on its use for quantitative determination of flavonoids.
In the present work, the applicability of NBS for the quantitative determination of quercetin in the presence of other flavonoids has been investigated and has resulted in a selective, simple and rapid spectrophotometric method. JzxPERlMJWTNA Apparatus
Uvidec-320 (Jasco, Tokyo, Japan) and SP 1750 (Pye-Unicam) spectrophotometers were used. Reagents
A 0.2% w/v solution of freshly recrystallixed N-bromosuccinimide (NBS, obtained from Aldrich, Germany) was prepared in distilled water, the solution was freshly prepared before use and standardized iodometrically.28 Phenol, a 1% w/v solution was prepared in methanol. Flavonoids were obtained from Fltia, Koch Light, Aldrich, Eastman Organic Chemicals and Sigma. Solutions containing 0.5 mg/ml of quercetin were prepared in methanol and diluted quantitatively with the same solvent to obtain dilutions between 20-300 pg/ml. Solvents used were of analytical reagent grade.
*Author for correspondence. 259
260
HASSAN F.
ASKAL et al.
Procedures Quercetin. Transfer 1 ml of working quercetin solution into a lo-ml standard flask, add 1 ml of phenol solution and 1 ml of NBS solution and make up to volume with distilled water. Mix and measure the absorbance at 510 nm against a reagent blank. Quercetin and kaempferol. Into 2 sets of lo-ml standard flasks, transfer 0.5-ml aliquots (to the first set) and l-ml aliquots (to the second set) of a mixture containing different proportions from quercetin and kaempferol (Table 3). Dilute the first set to volume with methanol, mix and measure the absorbance at 370 nm. To the other set, add 1 ml of phenol solution followed by 1 ml of NBS solution. Make up to volume with distilled water, mix and measure the absorbance at 510 nm against a reagent blank. Stoichiometric study. A series of 5-ml quantities of mixtures containing master equimolar solutions (2 x 10v4M) in different complementary proportions (from 0: 5 to 5: 0 inclusive) were made up in lo-ml standard flasks containing 1 ml of phenol solution. After the flasks have been made up to volume with water, absorbances were measured at 510 nm against a blank treated in the same manner. Monitoring of the hydrolysis of quercetin glycosides (Rutin). Into a 50-ml round-bottom flask, weigh accurately 18 mg of rutin (equivalent to 7.5 mg quercetin), dissolve in 5 ml of methanol, add 2.5 ml of distilled water and 7.5 ml of hydrochloric acid to produce a final solution of 5N. Reflux the resulting solution and at different time intervals, withdraw a l-ml volume, transfer to a 50-ml separating funnel, extract with ethylacetate (three times, each of 10 ml). Pass the collected organic layer over anhydrous sodium sulphate and evaporate ethylacetate under reduced pressure. Dissolve the residue in methanol to produce 5 ml in a 5-ml standard flask and use 1 ml from this solution for the procedure for quercetin. Determination of quercetin in plant. A 30-g sample of either leaves and stems, or flowers (dried and ground to 40 mesh) of Cassia didymobotrya Fres. was heated under reflux with 100 ml of 70% ethanol for 2 hr. The extract was concentrated to 25 ml and extracted with ethylacetate. After evaporation of the solvent under reduced pressure, the residue was dissolved in methanol to produce 10 ml in a IO-ml standard flask. Apply the procedure for quercetin, using 1 ml of the final extract solution in the blank. __.
RESULTS AND DIscussION
Optimization of conditions Addition of an aqueous solution of NBS to a methanolic solution of different flavonoids resulted in the formation of an instantaneous violet colour only with quercetin, the other investigated flavonoids fail to give any coloured product. The violet colour is developed immediately at room temperature and in the presence of phenol remains stable for at least 15 min. Higher temperatures produce lower absorbance values. No increase in the absorbance reading (0.540 at 510 nm) was obtained upon increasing the temperature from 20 to 30” while a further loo-increase decreases the absorbance by 6.9%. After that a regular 10” increase in the temperature up to 80” produces a 4% decrease in the corresponding absorbance readings. The absorption spectra for the quercetin reaction product exhibits three absorption peaks at 305, 375 and 510 nm (Fig. 1). Quercetin and NBS both have negligible absorbances at 510 nm. A l.O-ml volume of 1S-3 mg/ml NBS was found to be optimum. Use of 1.O ml of 5-15 mg/ml phenol solution in the total volume of IOml gives maximum absorbance readings. Various reducing substances such as phenol, hydrogen peroxide, salicylic acid, salicylamide, sulphanilamide, sodium nitrite and aniline sulphate were tested to try to increase the stability of the colour formed. Phenol was found the best as it gives the highest absorbance readings as well as the higher stability time. This effect might be explained on the basis of interacting the excess NBS with phenol thus preventing its destructive effect on the formed chromogen.
0.6
0.5
9
x 0.4
1
a 0.3 0.2 0.1 0 2%
300
400 Wavdength,
450
do
“m
Fig. 1. Absorption spectra oE (---) quercetin, 12.4 &ml; c u.) kaempferol 3.2 pg/ml (-) quercetin-NBS (12.4 pg quercetin and 20 mg NBS/ml and (-*-*-*) kaempferol-NBS (3.2 pg kaempferol and 20 mg NBS/ml).
Spectrophotometric
The violet colour is not developed in acidic or alkaline media and the effect of pH was so studied. Tests with acetate and phosphate buffers, showed that, the maximum colour intensity and stability were obtained with phosphate buffer of pH 7 but this system was still inferior to the phenol procedure, with regard to colour intensity and stability. Dilution of the developed coloured product by different solvents brings about bathowith dimethylsulphoxide chromic shifts dimethylformamide, methanol, ethanol, npropanol and isopropanol relative to water whereas absorption intensity was only slightly influenced (Table 1). Water was used throughout this work as it is the cheapest. These findings are in agreement with the fact that, in n-x* transition peaks, a hypsochromic shift occurs with increasing polarity of the solvent (expressed as dielectric constant). This is due to stabilization of the ground state through hydrogen bonding.29 Stoichiometry As assessed by the continuous molar variation methodm quercetin was found to interact with NBS in a ratio of 1: 4. Reaction mechanism The reaction does not seem to depend only on the catechol function, since rutin (which has the same catechol function as quercetin but sugar moiety at C-3 and luteolin (which lacks the 3-hydroxyl group) does not form such colour with NBS. Also a hydroxyl group at C-5 is necessary, as quercetin-5-glucose gives no such colour. Kaempferol failed to give a positive response indicating the necessity for orthodihydroxyl groups at C-3’ and 4’ in addition to the 3-, 5- and 7-hydroxyl groups. Morin (which has hydroxyl groups at C-3,5,6,2’ and 4’) produces Table 1. Effect of diluting solvents on the absorption intensity of the developed colour Solvent Water Dimethylsulphoxide Dimethylformamide Methanol Ethanol n-Propanol Iso-Pronanol
Absorbance* 0.540 0.560 0.294 0.459 0.490 0.540 0.523
*Final concentration 20 pg/ml. tAverage of three determinations. SReference 33.
&_t 510 518 522 525 525 528 532
Dielectric constant (E)$ 78.3 47.0 37.0 33.0 25.0 21.8 19.9
261
determination of quercetin
no such colour under the specified reaction conditions indicating that m-dihydroxyl groups at ring B (see structure below) are not suitable for such interaction. In conclusion, hydroxyl groups at C-3, 5, 7, 3’ and 4’ are necessary for this reaction.
f&J@ 5
4
Quercetin = 3,5,7,3’,4’- pentahydroxyflavone Kaempferol = 3.5.74’ - tetrahydroxyflavone Rutin = 3- glucose-rhamnose quercetin Luteolin = 3- deoxy - quercetin Apigenin = 3-deoxy-kaempferol Isorhamnetin= quercetin-3’-methylether Morin = 3,5,6,2’ and 4’pentahydroxyflavone
Evidence for the oxidation of 3-, 5-, 3’- and 4’-hydroxyl groups in the proposed procedure is the failure of aluminium chloride alone or with hydrochloric acid to induce any bathochromic shifts when added to the reaction product indicating the absence of these hydroxyl groups (cJ quercetin gives bathochromic shifts due to the presence of 3-, 5-, 3’- and 4’-hydroxyl groups).” Unfortunately, all trials made to isolate the formed chromogen were unsuccessful as it is destroyed during different isolation steps even under reduced pressure. From the aforementioned observations as well as similar reports on the oxidation of catecholamines26*27it could be concluded that a polyquinone structure of quercetin may be formed. Determination range At 510 nm, a linear correlation was obtained between absorbance of the coloured product and concentration of quercetin over the range of 2.5-30 pg/ml with good correlation coefficient and small intercept, the regression equation was: A 5,,,= 0.019 + 0.026 Co (t = 0.9990, n = 9) The molar absorptivity mole-‘. cm-‘.
was
8.14 x 10’1.
Selectivity of the colour reaction Selectivity of the method was checked by examining the effect of NBS on various
262
HASSANF.
flavonoids having hydroxyl groups in different positions such as kaempferol, rutin, isorhamnetin, luteoline, apigenin and morin. No violet colour was produced with any of these compounds inspite of their observed interaction with NBS as shown in Fig. 1, for kaempferol as an example.
ASKAL
et al.
Table 3. Monitoring the hydrolysis of rutin and its subsequent determination as quercetin Time,
hr
Quercetin produced,
0.5 1.0 1.5 2.0 32::
Precision The mean of 10 replicate analyses of a solution of quercetin at a concentration of 20 ,ug/ml was assayed with a coefficient of variation of 1.02%. This level of precision is adequate for quality control analysis of quercetin in natural sources. Analysis of mixtures
Table 2. Determination of synthetic mixtures of kaempferol and quercetin Mixture taken, pgglml Recovery, % * SD+ Kaempferol Quercetin Kaempferol Quercetin 0.0 2.5 5.0 7.5 10.0 15.0 12.5 15.0 20.0 25.0 17.5 20.0
*Average of three determinations and calculated with reference to quercetin treated in the same manner as rutin.
the reaction, then at 510 nm after the reaction, and, using the following equation: Ak = 2A4370-2.08AS10= 0.004 + 0.086 C, from which,
Eleven synthetic mixtures containing 2.5-20 pg kaempferol/ml and 2.5-25 pg quercetin/ml with a quercetin/kaempferol ratio of 0.125-10 were subjected to analysis with the proposed procedure and the results are shown in Table 2 with a mean recovery of 98.88 & 0.88% for kaempferol and 99.54 + 0.45% for quercetin. The method depends upon the fact that, under the specified reaction conditions, only quercetin reacts with NBS to give a coloured product measurable at 510 nm. The other flavonoids give zero absorbance at this wavelength. Measurements at 370 nm determine the total flavonoids. Concentration of quercetin could be directly calculated from the regression equation derived from the NBS-reaction product (A,,, = 0.019 + 0.026 CQ). On the other hand, concentration of other flavonoid-alculated as kaempferolcould be calculated through measurement of the native absorbance at 370 nm before carrying out
20.0 17.5 15.0 12.5 10.0 10.0 7.5 5.0 5.0 5.0 2.5 0 Mean % *SD
% f SD*
22.76 f 2.32 40.08 f 2.51 66.17 f 2.13 83.36 f 1.87 98.14 f 0.93 97.34k2.17
97.28 f 97.17 f 99.81 f 99.53 f 99.32 f 99.49 f 98.72 f 99.15 f 98.79 f 99.25 f 99.18 f
*Average of five determinations.
0.71 0.92 0.67 0.78 0.61 0.53 0.60 0.68 0.91 0.53 1.22
98.88 0.88
99.69 f 1.29 100.62 f 1.38 98.97 f 1.09 99.46 f 1.07 99.49 f 0.55 99.94 f 1.18 99.37 f 0.83 99.00 f 0.75 99.57 IO.68 99.43 f 0.55 99.39 f 0.71 99.54 0.45
ck=(2A’m
-2.08 A,,,) -0.004 oo86
where C, is the concentration of flavonoids other than quercetin (calculated as kaempferol) in pg/ml, 2A370is the total absorbance of the solution at 370 nm, 2.08 is a factor relating the absorbance of quercetin at 370 and 510 nm before and after the reaction, A,,, is the absorbance of the quercetin reaction product, 0.004 and 0.086 are the intercept and slope of the regression equation of kaempferol at 370 nm respectively. Hydrolysis of quercetin giycosides The proposed procedure would be valuable as a tool for quantitative monitoring of the hydrolysis of quercetin glycosides. When rutin is subjected to hydrolysis with 5N hydrochloric acid under reflux for 4 hr a gradual increase in the concentration of quercetin was observed, Table 3. Analysis of plant extracts To check the validity of the proposed procedure, assay of quercetin in Cassia didymobotrya Fres. in which quercetin has been described3* was carried out. Quercetin could be determined quantitatively in ethylacetate extract without the need for preseparation steps and without interference from other constituents. Interference from any trace absorption from the extract could be eliminated by taking the same volume of extract in the blank and the results obtained are shown in Table 4. Accuracy of the method was confirmed by the good recovery of added quercetin to the plant extracts (Table 4).
Spectrophotometric
detmnioatioo
Table 4. Determination of quercetio in Cassia didymobofrya Fres. Quercetio Extract Leaves and stems Flowers
Found, mg 1.86 2.31
Added, mg 2.0 2.5
Recovery, % f SD* 99.59 f 1.56 99.18 f 1.99
*Results based on three determinations per sample.
In conclusion, the proposed procedure is simple, time saving and selective for quercetin, moreover, it could be considered as a new highly-selective shift reagent for quercetin in the presence of other related flavonoids as well as a stability-indicating assay for quercetin glycosides. In addition, it reports a good method for the selective analysis of quercetin in the presence of other flavonoids. REFERENCES 1. L. E. Dowd, Anal. Chem., 1959, 31, 184. 2. D. H. Charles, H. W. Margref and T. E. Weichselbaum, ibid., 1960, 32, 122. 3. L. Jurd and T. A. Geissmao, J. Org. Chem., 1956, 21, 1395. 4. N. M. Akhmedkhcdxhaeva, A. N. Svechoikova, V. A. Baodyukova and D. M. Kambarova, Farmafsiya, 1986, 35, 60. 5. Z. P. Kosteooikova, G. A. Paoova and R. Dambrauskieoe, ibid., 1984, 33, 33. 6. G. A. Fetkhullina and T. I. Buleokov, ibid., 1984, 33, 38. 7. J. Peioado and J. Florindo, Analyst, 1988, 113, 555. 8. L. Xu, A. Liu and X. Zhaog, Yaoxue Xuebao, 1987,22, 208; Anal. Abstr., 1987, 49, E16. 9. Y. Zhang, J. Cui and S. Zhao, Yaoure Fenxi Zazhi, 1984, 4, 1; Anal. Abstr., 1985, 47, E18. 10. H. Wagner, G. Tittel and S. Bladt, Dtsch. Apoth. Ztg., 1983, 123, 515.
of quercetio
263
11. D. J. Diagle and E. J. Cookertoo, J. Liq. Chromarogr., 1983, 6, 105. 12. D. J. Diagle aod E. J. Cookertoo, ibid., 1988, 11, 309. 13. K. Hostettoraoo, B. Domoo, D. Schaufelberger aod M. Hostetnnaon, J. Chromutogr., 1984, 283, 137. 14. E. Revilla, E. Alooso and M. I. Estrella, Chromatographia, 1982, 22, 137. 15. K. H. Law and N. P. Das, J. Chromatogr., 1987, 388, 225. 16. I. S. Bhatia, J. Singh and K. L. Bajaj, Mikrochimica Acta, 1974, 5, 909. 17. J. A. Delccur and D. Janssens de Varebeka, J. Inst. Brew., 1985, 91, 37; Anal. Abstr., 1985, 47, 8F77. 18. Y. Liaog and Q. Zhaog, Yaowu Fermi Zazhi, 1987, 7, 347; Anal. Absrr., 1988, 50, 7E23. 19. H. Glasl, Z. Anal. Chem, 1985, 321, 325. 20. H. Ogura, Y. Shikiba and Y. Yamaxaki, J. Pharm. Sci., 1968, 57, 705. 21. R. Neu, Z. Anal. Chem., 1961, 42, 335. 22. J. T. MacGregor and L. Jurd, hfutat. Res., 1978, 54, 297. 23. G. Tamura, C. Gold, A. Ferrc-Luxxi and B. N. Aores, Proc. Natl. Acad. Sci., 1980, 77, 4961. 24. M. Nagao, N. Morita, T. Yahagi, M. Shiorixu, M. Kuroyaoagi, M. Fukuoka, K. Yoshihira, S. Natori, T. Fujioo and T. Sugimura, Environ. Mutagen., 1981, 3, 401. 25. A. M. Paoukcu, J. Hatcher, H. Taguchi and G. T. Bryan, Proc. Am. Assoc. Cancer Res., 1980, 21, 74. 26. M. I. Walash, A. Abcu Ouf and E. B. Salem, J. Assoc. Off., Anal. Chem., 1982, 65, 1445. 21. I&m, Analyst, 1981, 106, 949. 28. M. Z. Barakat and M. F. Abdel-Wahab, Anal. Chem., 1973, 26, 1954. 29. K. A. Connors, A Textbook of Pharmaceutical Analysis, 3rd Ed., p. 206. Wiley-Ioterscieoce, New York, 1982. 30. J. Rose, Advanced Physico-Chemical Experiments, p. 54. Pittmao, London, 1964. 31. J. B. Harboroe, T. J. Mabry and H. Mabry, The Flavonoids, Chapman and Hall, London, 1975. 32. S. M. El-Sayyad, A. M. Abdel-Baky, E. Y. Backheet and K.-W. Glombitza, Bull. Pharm. Sci., Assiut Uoiversity, 1989, 12, 195. 33. J. A. Riddick, W. B. Bunger, Organic Solvents, 3rd Ed., Wiley Ioterscieoce, New York, London, 1970.
0039-9140/92 M.OO+ 0.00 copyrightQ 1992Pagmon Prua pk
A SELECTIVE SPECTROPHOTOMETRIC METHOD FOR DETERMINATION OF QUERCETIN IN THE PRESENCE OF OTHER FLAVONOIDS HASSAN F. ASKAL and GAMALA. SALEH
Departmentof Pharmaceutical Chemistry, Faculty of Pharmacy, Assiut University, Assiut, Egypt ENMM Y. BACKEEET* Departmentof Pharmacognosy, Faculty of Pharmacy, Assiut University, Assiut, Egypt (Received 13 March 1991. Revised 4 July 1991. Accepted 4 July 1991) Summary-A simple, rapid and highly selective method for determination of quercetin in the presence of other flavonoids was developed. The method is based on the oxidation reaction of quemetin in neutral aqueous solution with N-bromosuccinimide (NBS) in the presence of phenol to give a violet chromogen measurable at 510 nm. Beer’s law was valid within a concentration range of 2.5-30 p&l with a good correlation coetEcient (r = 0.9990). All variables were studied to optimize the reaction conditions. The method is highly selective for quercetin. Other investigated flavonoids do not interfere. Mixtures of tlavonoids were also analysed through native W measurements of absorbance readings at 370 mn and then at 510 mn after adopting the proposed procedure. The method could also be utilixed for the quantitative determination of quercetin in some plant extracts. Moreover, the proposed procedure could be considered as a good tool to follow the hydrolysis of quercetin glycosides.
The quantitative analysis of flavonoids by ultraviolet absorption is well known.‘” Other methods for their determination include fluorimetry,’ polarography,8 densitometry9 and HPLC.‘&” Few calorimetric methods have been reported for their determination.iG2’ Most colorimetric methods cannot be used to distinguish between the different flavonoids. When interfering flavonoids are to be present, quercetin must first be isolated chromatographically’ but this requires an empirical correction for adsorption of quercetin. Recent studies have shown that, flavonoids have mutagenic activity.“-24 Quercetin is the most mutagenic; it also acts as a carcinogenic agent towards rats.25 There has been a need for a selective colorimetric method which could best be utilized in detecting and distinguishing between quercetin and other flavonoids without the need for any pretreatments or prior separation. N-Bromosuccinimide (NBS) has been previously used for the analysis of hydroxy compounds.26*27 However, no report is available on its use for quantitative determination of flavonoids.
In the present work, the applicability of NBS for the quantitative determination of quercetin in the presence of other flavonoids has been investigated and has resulted in a selective, simple and rapid spectrophotometric method. JzxPERlMJWTNA Apparatus
Uvidec-320 (Jasco, Tokyo, Japan) and SP 1750 (Pye-Unicam) spectrophotometers were used. Reagents
A 0.2% w/v solution of freshly recrystallixed N-bromosuccinimide (NBS, obtained from Aldrich, Germany) was prepared in distilled water, the solution was freshly prepared before use and standardized iodometrically.28 Phenol, a 1% w/v solution was prepared in methanol. Flavonoids were obtained from Fltia, Koch Light, Aldrich, Eastman Organic Chemicals and Sigma. Solutions containing 0.5 mg/ml of quercetin were prepared in methanol and diluted quantitatively with the same solvent to obtain dilutions between 20-300 pg/ml. Solvents used were of analytical reagent grade.
*Author for correspondence. 259
260
HASSAN F.
ASKAL et al.
Procedures Quercetin. Transfer 1 ml of working quercetin solution into a lo-ml standard flask, add 1 ml of phenol solution and 1 ml of NBS solution and make up to volume with distilled water. Mix and measure the absorbance at 510 nm against a reagent blank. Quercetin and kaempferol. Into 2 sets of lo-ml standard flasks, transfer 0.5-ml aliquots (to the first set) and l-ml aliquots (to the second set) of a mixture containing different proportions from quercetin and kaempferol (Table 3). Dilute the first set to volume with methanol, mix and measure the absorbance at 370 nm. To the other set, add 1 ml of phenol solution followed by 1 ml of NBS solution. Make up to volume with distilled water, mix and measure the absorbance at 510 nm against a reagent blank. Stoichiometric study. A series of 5-ml quantities of mixtures containing master equimolar solutions (2 x 10v4M) in different complementary proportions (from 0: 5 to 5: 0 inclusive) were made up in lo-ml standard flasks containing 1 ml of phenol solution. After the flasks have been made up to volume with water, absorbances were measured at 510 nm against a blank treated in the same manner. Monitoring of the hydrolysis of quercetin glycosides (Rutin). Into a 50-ml round-bottom flask, weigh accurately 18 mg of rutin (equivalent to 7.5 mg quercetin), dissolve in 5 ml of methanol, add 2.5 ml of distilled water and 7.5 ml of hydrochloric acid to produce a final solution of 5N. Reflux the resulting solution and at different time intervals, withdraw a l-ml volume, transfer to a 50-ml separating funnel, extract with ethylacetate (three times, each of 10 ml). Pass the collected organic layer over anhydrous sodium sulphate and evaporate ethylacetate under reduced pressure. Dissolve the residue in methanol to produce 5 ml in a 5-ml standard flask and use 1 ml from this solution for the procedure for quercetin. Determination of quercetin in plant. A 30-g sample of either leaves and stems, or flowers (dried and ground to 40 mesh) of Cassia didymobotrya Fres. was heated under reflux with 100 ml of 70% ethanol for 2 hr. The extract was concentrated to 25 ml and extracted with ethylacetate. After evaporation of the solvent under reduced pressure, the residue was dissolved in methanol to produce 10 ml in a IO-ml standard flask. Apply the procedure for quercetin, using 1 ml of the final extract solution in the blank. __.
RESULTS AND DIscussION
Optimization of conditions Addition of an aqueous solution of NBS to a methanolic solution of different flavonoids resulted in the formation of an instantaneous violet colour only with quercetin, the other investigated flavonoids fail to give any coloured product. The violet colour is developed immediately at room temperature and in the presence of phenol remains stable for at least 15 min. Higher temperatures produce lower absorbance values. No increase in the absorbance reading (0.540 at 510 nm) was obtained upon increasing the temperature from 20 to 30” while a further loo-increase decreases the absorbance by 6.9%. After that a regular 10” increase in the temperature up to 80” produces a 4% decrease in the corresponding absorbance readings. The absorption spectra for the quercetin reaction product exhibits three absorption peaks at 305, 375 and 510 nm (Fig. 1). Quercetin and NBS both have negligible absorbances at 510 nm. A l.O-ml volume of 1S-3 mg/ml NBS was found to be optimum. Use of 1.O ml of 5-15 mg/ml phenol solution in the total volume of IOml gives maximum absorbance readings. Various reducing substances such as phenol, hydrogen peroxide, salicylic acid, salicylamide, sulphanilamide, sodium nitrite and aniline sulphate were tested to try to increase the stability of the colour formed. Phenol was found the best as it gives the highest absorbance readings as well as the higher stability time. This effect might be explained on the basis of interacting the excess NBS with phenol thus preventing its destructive effect on the formed chromogen.
0.6
0.5
9
x 0.4
1
a 0.3 0.2 0.1 0 2%
300
400 Wavdength,
450
do
“m
Fig. 1. Absorption spectra oE (---) quercetin, 12.4 &ml; c u.) kaempferol 3.2 pg/ml (-) quercetin-NBS (12.4 pg quercetin and 20 mg NBS/ml and (-*-*-*) kaempferol-NBS (3.2 pg kaempferol and 20 mg NBS/ml).
Spectrophotometric
The violet colour is not developed in acidic or alkaline media and the effect of pH was so studied. Tests with acetate and phosphate buffers, showed that, the maximum colour intensity and stability were obtained with phosphate buffer of pH 7 but this system was still inferior to the phenol procedure, with regard to colour intensity and stability. Dilution of the developed coloured product by different solvents brings about bathowith dimethylsulphoxide chromic shifts dimethylformamide, methanol, ethanol, npropanol and isopropanol relative to water whereas absorption intensity was only slightly influenced (Table 1). Water was used throughout this work as it is the cheapest. These findings are in agreement with the fact that, in n-x* transition peaks, a hypsochromic shift occurs with increasing polarity of the solvent (expressed as dielectric constant). This is due to stabilization of the ground state through hydrogen bonding.29 Stoichiometry As assessed by the continuous molar variation methodm quercetin was found to interact with NBS in a ratio of 1: 4. Reaction mechanism The reaction does not seem to depend only on the catechol function, since rutin (which has the same catechol function as quercetin but sugar moiety at C-3 and luteolin (which lacks the 3-hydroxyl group) does not form such colour with NBS. Also a hydroxyl group at C-5 is necessary, as quercetin-5-glucose gives no such colour. Kaempferol failed to give a positive response indicating the necessity for orthodihydroxyl groups at C-3’ and 4’ in addition to the 3-, 5- and 7-hydroxyl groups. Morin (which has hydroxyl groups at C-3,5,6,2’ and 4’) produces Table 1. Effect of diluting solvents on the absorption intensity of the developed colour Solvent Water Dimethylsulphoxide Dimethylformamide Methanol Ethanol n-Propanol Iso-Pronanol
Absorbance* 0.540 0.560 0.294 0.459 0.490 0.540 0.523
*Final concentration 20 pg/ml. tAverage of three determinations. SReference 33.
&_t 510 518 522 525 525 528 532
Dielectric constant (E)$ 78.3 47.0 37.0 33.0 25.0 21.8 19.9
261
determination of quercetin
no such colour under the specified reaction conditions indicating that m-dihydroxyl groups at ring B (see structure below) are not suitable for such interaction. In conclusion, hydroxyl groups at C-3, 5, 7, 3’ and 4’ are necessary for this reaction.
f&J@ 5
4
Quercetin = 3,5,7,3’,4’- pentahydroxyflavone Kaempferol = 3.5.74’ - tetrahydroxyflavone Rutin = 3- glucose-rhamnose quercetin Luteolin = 3- deoxy - quercetin Apigenin = 3-deoxy-kaempferol Isorhamnetin= quercetin-3’-methylether Morin = 3,5,6,2’ and 4’pentahydroxyflavone
Evidence for the oxidation of 3-, 5-, 3’- and 4’-hydroxyl groups in the proposed procedure is the failure of aluminium chloride alone or with hydrochloric acid to induce any bathochromic shifts when added to the reaction product indicating the absence of these hydroxyl groups (cJ quercetin gives bathochromic shifts due to the presence of 3-, 5-, 3’- and 4’-hydroxyl groups).” Unfortunately, all trials made to isolate the formed chromogen were unsuccessful as it is destroyed during different isolation steps even under reduced pressure. From the aforementioned observations as well as similar reports on the oxidation of catecholamines26*27it could be concluded that a polyquinone structure of quercetin may be formed. Determination range At 510 nm, a linear correlation was obtained between absorbance of the coloured product and concentration of quercetin over the range of 2.5-30 pg/ml with good correlation coefficient and small intercept, the regression equation was: A 5,,,= 0.019 + 0.026 Co (t = 0.9990, n = 9) The molar absorptivity mole-‘. cm-‘.
was
8.14 x 10’1.
Selectivity of the colour reaction Selectivity of the method was checked by examining the effect of NBS on various
262
HASSANF.
flavonoids having hydroxyl groups in different positions such as kaempferol, rutin, isorhamnetin, luteoline, apigenin and morin. No violet colour was produced with any of these compounds inspite of their observed interaction with NBS as shown in Fig. 1, for kaempferol as an example.
ASKAL
et al.
Table 3. Monitoring the hydrolysis of rutin and its subsequent determination as quercetin Time,
hr
Quercetin produced,
0.5 1.0 1.5 2.0 32::
Precision The mean of 10 replicate analyses of a solution of quercetin at a concentration of 20 ,ug/ml was assayed with a coefficient of variation of 1.02%. This level of precision is adequate for quality control analysis of quercetin in natural sources. Analysis of mixtures
Table 2. Determination of synthetic mixtures of kaempferol and quercetin Mixture taken, pgglml Recovery, % * SD+ Kaempferol Quercetin Kaempferol Quercetin 0.0 2.5 5.0 7.5 10.0 15.0 12.5 15.0 20.0 25.0 17.5 20.0
*Average of three determinations and calculated with reference to quercetin treated in the same manner as rutin.
the reaction, then at 510 nm after the reaction, and, using the following equation: Ak = 2A4370-2.08AS10= 0.004 + 0.086 C, from which,
Eleven synthetic mixtures containing 2.5-20 pg kaempferol/ml and 2.5-25 pg quercetin/ml with a quercetin/kaempferol ratio of 0.125-10 were subjected to analysis with the proposed procedure and the results are shown in Table 2 with a mean recovery of 98.88 & 0.88% for kaempferol and 99.54 + 0.45% for quercetin. The method depends upon the fact that, under the specified reaction conditions, only quercetin reacts with NBS to give a coloured product measurable at 510 nm. The other flavonoids give zero absorbance at this wavelength. Measurements at 370 nm determine the total flavonoids. Concentration of quercetin could be directly calculated from the regression equation derived from the NBS-reaction product (A,,, = 0.019 + 0.026 CQ). On the other hand, concentration of other flavonoid-alculated as kaempferolcould be calculated through measurement of the native absorbance at 370 nm before carrying out
20.0 17.5 15.0 12.5 10.0 10.0 7.5 5.0 5.0 5.0 2.5 0 Mean % *SD
% f SD*
22.76 f 2.32 40.08 f 2.51 66.17 f 2.13 83.36 f 1.87 98.14 f 0.93 97.34k2.17
97.28 f 97.17 f 99.81 f 99.53 f 99.32 f 99.49 f 98.72 f 99.15 f 98.79 f 99.25 f 99.18 f
*Average of five determinations.
0.71 0.92 0.67 0.78 0.61 0.53 0.60 0.68 0.91 0.53 1.22
98.88 0.88
99.69 f 1.29 100.62 f 1.38 98.97 f 1.09 99.46 f 1.07 99.49 f 0.55 99.94 f 1.18 99.37 f 0.83 99.00 f 0.75 99.57 IO.68 99.43 f 0.55 99.39 f 0.71 99.54 0.45
ck=(2A’m
-2.08 A,,,) -0.004 oo86
where C, is the concentration of flavonoids other than quercetin (calculated as kaempferol) in pg/ml, 2A370is the total absorbance of the solution at 370 nm, 2.08 is a factor relating the absorbance of quercetin at 370 and 510 nm before and after the reaction, A,,, is the absorbance of the quercetin reaction product, 0.004 and 0.086 are the intercept and slope of the regression equation of kaempferol at 370 nm respectively. Hydrolysis of quercetin giycosides The proposed procedure would be valuable as a tool for quantitative monitoring of the hydrolysis of quercetin glycosides. When rutin is subjected to hydrolysis with 5N hydrochloric acid under reflux for 4 hr a gradual increase in the concentration of quercetin was observed, Table 3. Analysis of plant extracts To check the validity of the proposed procedure, assay of quercetin in Cassia didymobotrya Fres. in which quercetin has been described3* was carried out. Quercetin could be determined quantitatively in ethylacetate extract without the need for preseparation steps and without interference from other constituents. Interference from any trace absorption from the extract could be eliminated by taking the same volume of extract in the blank and the results obtained are shown in Table 4. Accuracy of the method was confirmed by the good recovery of added quercetin to the plant extracts (Table 4).
Spectrophotometric
detmnioatioo
Table 4. Determination of quercetio in Cassia didymobofrya Fres. Quercetio Extract Leaves and stems Flowers
Found, mg 1.86 2.31
Added, mg 2.0 2.5
Recovery, % f SD* 99.59 f 1.56 99.18 f 1.99
*Results based on three determinations per sample.
In conclusion, the proposed procedure is simple, time saving and selective for quercetin, moreover, it could be considered as a new highly-selective shift reagent for quercetin in the presence of other related flavonoids as well as a stability-indicating assay for quercetin glycosides. In addition, it reports a good method for the selective analysis of quercetin in the presence of other flavonoids. REFERENCES 1. L. E. Dowd, Anal. Chem., 1959, 31, 184. 2. D. H. Charles, H. W. Margref and T. E. Weichselbaum, ibid., 1960, 32, 122. 3. L. Jurd and T. A. Geissmao, J. Org. Chem., 1956, 21, 1395. 4. N. M. Akhmedkhcdxhaeva, A. N. Svechoikova, V. A. Baodyukova and D. M. Kambarova, Farmafsiya, 1986, 35, 60. 5. Z. P. Kosteooikova, G. A. Paoova and R. Dambrauskieoe, ibid., 1984, 33, 33. 6. G. A. Fetkhullina and T. I. Buleokov, ibid., 1984, 33, 38. 7. J. Peioado and J. Florindo, Analyst, 1988, 113, 555. 8. L. Xu, A. Liu and X. Zhaog, Yaoxue Xuebao, 1987,22, 208; Anal. Abstr., 1987, 49, E16. 9. Y. Zhang, J. Cui and S. Zhao, Yaoure Fenxi Zazhi, 1984, 4, 1; Anal. Abstr., 1985, 47, E18. 10. H. Wagner, G. Tittel and S. Bladt, Dtsch. Apoth. Ztg., 1983, 123, 515.
of quercetio
263
11. D. J. Diagle and E. J. Cookertoo, J. Liq. Chromarogr., 1983, 6, 105. 12. D. J. Diagle aod E. J. Cookertoo, ibid., 1988, 11, 309. 13. K. Hostettoraoo, B. Domoo, D. Schaufelberger aod M. Hostetnnaon, J. Chromutogr., 1984, 283, 137. 14. E. Revilla, E. Alooso and M. I. Estrella, Chromatographia, 1982, 22, 137. 15. K. H. Law and N. P. Das, J. Chromatogr., 1987, 388, 225. 16. I. S. Bhatia, J. Singh and K. L. Bajaj, Mikrochimica Acta, 1974, 5, 909. 17. J. A. Delccur and D. Janssens de Varebeka, J. Inst. Brew., 1985, 91, 37; Anal. Abstr., 1985, 47, 8F77. 18. Y. Liaog and Q. Zhaog, Yaowu Fermi Zazhi, 1987, 7, 347; Anal. Absrr., 1988, 50, 7E23. 19. H. Glasl, Z. Anal. Chem, 1985, 321, 325. 20. H. Ogura, Y. Shikiba and Y. Yamaxaki, J. Pharm. Sci., 1968, 57, 705. 21. R. Neu, Z. Anal. Chem., 1961, 42, 335. 22. J. T. MacGregor and L. Jurd, hfutat. Res., 1978, 54, 297. 23. G. Tamura, C. Gold, A. Ferrc-Luxxi and B. N. Aores, Proc. Natl. Acad. Sci., 1980, 77, 4961. 24. M. Nagao, N. Morita, T. Yahagi, M. Shiorixu, M. Kuroyaoagi, M. Fukuoka, K. Yoshihira, S. Natori, T. Fujioo and T. Sugimura, Environ. Mutagen., 1981, 3, 401. 25. A. M. Paoukcu, J. Hatcher, H. Taguchi and G. T. Bryan, Proc. Am. Assoc. Cancer Res., 1980, 21, 74. 26. M. I. Walash, A. Abcu Ouf and E. B. Salem, J. Assoc. Off., Anal. Chem., 1982, 65, 1445. 21. I&m, Analyst, 1981, 106, 949. 28. M. Z. Barakat and M. F. Abdel-Wahab, Anal. Chem., 1973, 26, 1954. 29. K. A. Connors, A Textbook of Pharmaceutical Analysis, 3rd Ed., p. 206. Wiley-Ioterscieoce, New York, 1982. 30. J. Rose, Advanced Physico-Chemical Experiments, p. 54. Pittmao, London, 1964. 31. J. B. Harboroe, T. J. Mabry and H. Mabry, The Flavonoids, Chapman and Hall, London, 1975. 32. S. M. El-Sayyad, A. M. Abdel-Baky, E. Y. Backheet and K.-W. Glombitza, Bull. Pharm. Sci., Assiut Uoiversity, 1989, 12, 195. 33. J. A. Riddick, W. B. Bunger, Organic Solvents, 3rd Ed., Wiley Ioterscieoce, New York, London, 1970.