Microsoft Word - Brodnjak NB 9-2.doc Nova Biotechnologica 9-2 (2009) 113 HIGH-PERFORMANCE LIQUID CHROMATOGRAPHIC DETERMINATION OF SELECTED PHENOLIC ACIDS IN WINE DARINKA BRODNJAK VONČINA, MAŠA ISLAMČEVIĆ RAZBORŠEK, MARJANA SIMONIČ University of Maribor, Faculty of chemistry and chemical engineering, Smetanova 17, Maribor, Slovenia (darinka.brodnjak@uni-mb.si) Abstract: The aim of this study was to develop a method for identification and quantification of phenolic acids in different wine samples. The simple reversed-phase HPLC-UV method for simultaneous determination of p-coumaric and ferulic acid was developed. The method was validated and working range, linearity, repeatability, accuracy, limit of quantitation LOQ and limit of detection LOD were determined. The linearity of the method was tested in concentration ranges 0.1-1 mg L-1 and 1-10 mg L-1. The correlation coefficients (r2) were greater than 0.996 and quality coefficients (QC) ≤ 5%. Detection limit for both compounds was 0.01 mg L-1. The method is precise (RSD<10%) and accurate. Two different wine sorts were used for analysis, Slovene red wine Refošk and Slovene white wine Laški Rizling. The wine samples were ultrasonically extracted with ethyl acetate, centrifuged, evaporated, re-dissolved in mobile phase and filtered. The solution was analyzed directly by HPLC-UV. Phenolic compounds were separated with a C18 reversed-phase column by isocratic elution using 2 % acetic acid in water-methanol (85:12, v/v) as the mobile phase at a flow rate of 0.7 mL min-1. Acids were detected and quantitation was performed at wavelength of 320 nm. Extraction procedure was optimized and yields were calculated from internal standard recovery. O-coumaric acid was used as internal standard. From the results it is evident that the red wine Refošk contains higher concentrations of investigated compounds then the white wine Laški Rizling. Stability tests were performed with standard compounds of p-coumaric and ferulic acid. It was established that both standards occur as trans-isomers, but when they were dissolved in methanol and stored exposed to daylight and room-temperature they were gradually transformed to cis-isomers. All closely related geometrical cis- and trans-isomers were successfully separated under the same conditions as described before. The described method is simple and rapid as it involves liquid extraction with no other pre- treatments of samples. It is suitable for simultaneous determination of cis and trans isomers of p-coumaric and ferulic acid in different liquid samples. Key words: p-coumaric acid, ferulic acid, o-coumaric acid, HPLC, wine. 1. Introduction In recent years lots of studies were focused on phenolic acids due to their antioxidant antiviral, antibacterial and anti-inflammatory properties (ZGÓRKA and KAWKA, 2001). A variety of phenolic acids were separated using high-performance liquid chromatography (HPLC) as reported in literature (KALLITHRAKA et al., 2006; ROBBINS and BEAN, 2004; TARNAWSKI et al., 2006; URAKOVA et al., 2008; ZGÓRKA and KAWKA, 2001). In most studies reversed-phase HPLC method was applied using non-polar C18 column. Isocratic elution was applied (TARNAWSKI et al., 2006; URAKOVA et al., 2008; ZGÓRKA and KAWKA, 2001; YI et al., 2009) as well as a gradient method (KALLITHRAKA et al., 2006; ROBBINS and BEAN, 2004; SLADKOVSKÝ et al., 2004). Different mobile phases were prepared, using methanol–aqueous or acetonitrile–aqueous acidified mixtures. It 114 Brodnjak Vončina, D. et al. was determined that pH value is very important due to protonation of the acids. Therefore the best pH was determined to be from 1-3. GARCIÁ et al. (2007) paid a lot of attention to sample pre-treatment. They used liquid-liquid separation with diethyl ether, followed by centrifugation, and vacuum evaporation. A variety of different wines was analysed by KALLITHRAKA et al. (2006). Samples were only filtered and no other pre-treatment methods were used. HPLC-DAD method was applied for determination of phenolic acids at different wavelengths (265, 280, 320, and 365 nm, respectively, at 1 mL min-1 flow). The concentration of phenolic acids varied a lot depending on weather conditions, geographic sites, the way of wine storage and aging. These discoveries agree also with other studies (ÖZKAN et al., 2006). The aim of our work was to develop the analytical method to separate and determine two target components in wine samples, namely p-coumaric and ferulic acid. Attention was paid to the most appropriate sample pre-treatment technique. 2. Materials and methods All reagents and solvents used were of analytical grade. Milli-Q water was used. Methanol (MeOH) was purchased from Sigma-Aldrich (Germany), acetic acid (CH3COOH), ethyl-acetate (C4H8O2), 98% trans ferulic acid (FA), 98 % trans p- coumaric acid (p-CA) and 97% o- coumaric acid (o-CA) were supplied from Merck (Germany). 2.1 Preparation of Calibration Solutions and Calibration Curves Standard stock solution of trans FA, trans p-CA and o-CA was prepared by dissolving each of them with MeOH in 10 mL volumetric glass flasks to obtain concentration of 1 g L-1. o-CA was used as an internal standard (ISTD). This solution was degassed for 20 minutes by sonication. Five working calibration solutions were prepared from standard stock solutions by further dilution with mobile phase consisting of 2% acetic acid in water-methanol mixture (82:18, v/v). Concentration ranges of the final standard solutions used for calibration curves were 0.1-1 mg L-1 and 1-10 mg L-1. Calibration solutions were injected into the HPLC-UV system in five replicates. In addition, three replicate analyses of the calibration solutions were performed. Curves were constructed by linear regression of the peak-area ratio (y) of individual acids to the ISTD, versus the concentration (x). All standard solutions were kept at -18 ° C for a maximum one week. Prior to injection the solutions were filtered through 0.45 μm filter. 2.2 Sample preparation Wine samples were Refošk (red wine) and Laški Risling (white wine) from Slovenia. Wine samples undergo a liquid-liquid extraction prior to their analysis by HPLC. 5 mL of wine sample was transferred into a 50-mL round-bottomed glass- stoppered centrifuge tube, spiked with 50 µL of ISTD (0.5 mg L-1) and extracted with 5 mL of ethyl acetate by sonication (ultrasonic bath model Sonis 4, Iskra, Slovenia) for Nova Biotechnologica 9-2 (2009) 115 20 minutes. The extract was centrifuged (centrifuge model LC 321, Tehnica Železniki, Slovenia) at 1000 rpm for 10 min and the supernatant was accurately transferred into 200-mL conical glass flask using a glass Pasteur pipette. This extraction procedure was repeated three times. Supernatants were combined and concentrated to dryness by rotary evaporation (Büchi Rotavapor R-205, Switzerland). The residue was re- dissolved in 5 ml of mobile phase consisting of 2% acetic acid in water-methanol mixture (82:18, v/v). Finally, the pre-treated sample was filtered through 0.45 µm filter and analysed. The compounds in wine extracts were quantified from the corresponding calibration curves. 2.3 Stability tests Trans p-CA and trans FA standard compounds were dissolved in MeOH separately, to obtain concentrations of 1 mg L-1. The contents of trans p-CA and trans FA were determined from the corresponding calibration curves and they were used for the control. Then each solution was divided into more equal parts and transferred into graduated glass-stoppered test tubes. Four parts were placed under the daylight and room temperature for varied lengths of time (5 h, 24 h, 48 h and 1 week); another four parts of each solution were stored in the refrigerator in the darkness at -18 °C for varied lengths of time (5 h, 24 h, 48 h and 1 week). After exposure to the mentioned conditions, the content of trans p-CA and trans FA were determined from the corresponding calibration curves and compared to the control. Equipment and Chromatographic analyses A simple and rapid chromatographic separation was carried out using reverse- phase HPLC. Chromatographic analyses were performed on Varian, Prostar (USA) HPLC system, which consisted of Varian Prostar 210 HPLC pump, Varian Prostar injection valve, and UV-VIS detector. Chromatographic system was connected to the PC. Separation was carried out on 250 mm x 3mm i.d., 5 µm non-polar CHROMSPHER 5, C18 column. Elution was performed isocratically with the mobile phase consisting of 2% acetic acid in water-methanol mixture (82:18, v/v). The flow rate was 0.7 mL min-1, the pressure 180×105 Pa and the injection volume was 100 µL. Peaks were detected at 320 nm. 3. Results and discussion Based on linear regression analysis, the responses for investigated compounds were linear over concentration ranges tested. The equations of calibration curves and their correlation coefficients are presented in Table 1. The reproducibility of chromatographic analyses was evaluated by the relative standard deviation (RSD) of five replicate analyses of five calibration solutions. RSD was between 2.91 and 3.81 %. The extraction procedure was optimized regarding extraction solvent and recovery. Ethyl-acetate was used for extraction of the compounds from wine. The best recoveries (over 89%) were obtained by using three equal volumes of mentioned solvent, while using only two equal volumes gave lower recoveries. The recovery was evaluated using o–CA which was added into the wine sample before the extraction. 116 Brodnjak Vončina, D. et al. This compound is structurally related to the investigated compounds and behaves similar on the column as analytes. It does not occur in wine. Therefore it is suitable for quantitative analysis of investigated compounds. Table 1. Regression equations and correlation coefficients for investigated compounds Investigated compound Correlation coefficient (r2) Regression equation trans p–CA 0.9969 y = 0.4088 x + 0.1021 trans FA 0.9968 y = 0.3403 x + 0.0705 Chromatographic separation was optimized with respect to the mobile phase, eluent composition, and flow rate. Several experiments were carried out to resolve the acid mixture using a different mobile phases. To increase the retention behaviour acetic acid was used. Optimal peak resolution was achieved using 2% acetic acid in water-methanol mixture (82:18, v/v). In Fig. 1 chromatograms of investigated compounds, present red wine extract is presented. The separation with good resolution has been achieved for phenolic acids within 32 min. Their average contents in different wine extracts (mg L-1) are presented in Table 2. Concentrations of trans p-CA and trans FA are higher in red wine. Very high concentration of trans FA was determined in Refošk, which was even 100-times higher compared with white wine. Contents of phenolic acids are comparable to those reported in literature (ROBBINS and BEAN, 2004; KALLITHRAKA et al., 2006; GARCIÁ et al., 2007; SLADKOVSKÝ et al., 2004; OZKAN et al., 2006; CASTELLARI et al., 2002). Trans p-CA in different brands of red wine were reported in concentrations between 0.9 to 7 mg L-1 and trans FA from 0.14 to 6 mg L-1L, respectively. Concerning white wine brands lower concentrations of both acids were Fig. 1. Chromatogram of investigated compounds present in extract of red wine-Refošk. Nova Biotechnologica 9-2 (2009) 117 found in literature: trans p-CA varied from 0.2 to 1.4 mg L-1and trans FA from 0.1 to 0.7 mg L-1, respectively. Table 2. Contents of trans p–CA and trans FA in extracts of red wine-Refošk and white wine-Laški Rizling (mg L-1). Each content value is the mean of five replications. Compound red wine Refošk (n=5) white wine Laški Rizling (n=5) trans p–CA 2.95 1.82 trans FA 7.75 0.07 The results of stability tests showed that trans p–CA and trans FA gradually converted to their cis forms when they were dissolved in MeOH and stored in daylight at room temperature. The trans-cis isomerization started already after a few hours. The content of trans p–CA and trans FA has lowered during the time of storing, while the content of cis p–CA and cis FA has increased. After being stored for 1 week in the refrigerator in darkness at -18 °C, trans p–CA and trans FA dissolved in MeOH were found to be stable, and no isomer conversion occurred. Although cis and trans forms of phenolic acids are geometric isomers with very similar structures, they can be well distinguished by order of elution during HPLC (cis isomers are retained longer than trans isomers) (Fig. 2) Fig. 2. Chromatogram of compounds present in mixture of trans p–CA and trans FA in MeOH after being stored in daylight and room temperature for 24 hours. 4. Conclusion The developed isocratic HPLC method is a rapid and simple one compared with other HPLC methods. The selection of proper mobile phase is crucial for good separation of phenolic compounds. Proper storing conditions are important to ensure 118 Brodnjak Vončina, D. et al. stability of the compounds. The described method can be routinely used for separation and determination of closely related cis- and trans-isomers of phenolic acids in different wine extracts and other liquid samples with complex natural matrices, even if they are present in the minor contents. References BUIARELLI, F., CARTONI, G., COCCIOLI, F., LEVETSOVITOU, Z.: Determination of phenolic acids in wine by HPLC with a microbore column. J. Chromatogr., A, 695, 1995, 229-235. CASTELLARI, M., SARTINI, E., FABIANI, A., ARFELLI, G., AMATI, A.: Analysis of wine phenolics by high-performance liquid chromatography using a monolithic type column. J. Chromatogr., 973, 2002, 221-227. CHAMKHA, M., CATHALA, B., CHEYNIER, V., DOUILLARD, R.: Phenolic composition of Champagnes from Chardonnay and Pinot noir Vintages. J. Agric. Food Chem., 51, 2003, 3179-3184. GARCÍA, Falcón, M.S., PÉREZ, LAMELA, C., MARTÍNEZ, CARBALLO, E., SIMAL, GÁNDARA, J.: Determination of phenolic compounds in wines: Influence of bottle storage of young red wines on their evolution. Food Chem., 105, 2007, 248-259. KALLITHRAKA, S., TSOUTSOURAS, E., TZOUROU, E., LANARIDIS, P.: Principal phenolic compounds in Greek red wines. Food Chem., 99, 2006, 784- 793. ÖZKAN, G., GÖKTÜRK, BAYDAR, N.: A direct RP-HPLC determination of phenolic compounds in Turkish red wines. J. Fac. Agriculture Akdeníz University, 19, 2006, 229-234. ROBBINS, R. J., BEAN, S. R.: Development of a quantitative high-performance liquid chromatography-photodiode array detection measurement system for phenolic acids. J. Chromatogr., A, 1038, 2004, 97-105. SLADKOVSKÝ, R., SOLICH, P., URBÁNEK, M.: High-performance liquid chromatography determination of phenolic components in wine using off-line isotachophoretic pre-treatment. J. Chromatogr., 1040, 2004, 179-184. TARNAWSKI, M., DEPTA, K., GREJCIUN, D., SZELEPIN, B.: HPLC determination of phenolic acids and antioxidant activity in concentrated peat extract-a natural immunomodulator. J. Pharm. Biomed. Anal., 41, 2006, 182-188. URAKOVA I. N., POZHARITSKAYA, O. N., SHIKOV, A. N., KOSMAN, V. M.: Development and validation of an LC method for simultaneous determination of ascorbic acid and tree phenolic acids in sustained release tablets at single wavelength. J. Chromatogr., 67, 2008, 709-713. YI, C., SHI, J., KRAMER, J., XUE, S., JIANG, Y., ZHANG, M., MA, Y., POHORLY, J.: Fatty acid composition and phenolic antioxidants of winemaking pomace powder. Food Chem., 114, 2009, 570-576. ZGÓRKA, G., KAWKA, S.: Application of conventional UV, photodiode array (PDA) and fluorescence (FL) detection to analysis of phenolic acids in plant material and pharmaceutical preparation. J. Pharm. Biomed. Anal., 24, 2001, 1065- 1072.