Investigation of contribution of individual constituents to antioxidant activity in herbal drugs using postcolumn HPLC method

The most important attention is paid to the search of natural antioxidants and their evaluation in medicinal and food raw materials of plant origin. A number of plants, their extracts, food products, and medicinal preparations appear to be the objects of scientific research. Effectiveness and informative character of research, undoubtedly, depend on relevance, sensitivity, and efficiency of the methods chosen. The aim of this work was to develop and validate the postcolumn high-performance liquid chromatography (HPLC)-DPPH method as well as its application in the evaluation of antioxidant activity of known and unknown compounds scavenging free radicals and existing in medicinal plant raw materials. HPLC-separated compounds were identified at the wavelength of 275 nm, and then the mobile phase with analytes flowed through a mixing tee to the reaction coil, where DPPH reagent solution was supplied. The solution flow rate was 0.4 mL/min. The reaction coil was connected with UV/VIS type detector, which measured absorption of flowing solution at the wavelength of 520 nm. It was determined that vitexin rhamnoside, the dominant compound in the leaves of Crataegus monogyna, was not a significant radical scavenger. The most active antioxidant in the leaves and flowers of Crataegus monogyna was chlorogenic acid. The most active antioxidant in Origanum vulgare raw material was rosmarinic acid. Identified analytes in the extracts of Achillea millefolium that possessed radical-scavenging properties were chlorogenic acid, luteolin-7-Oglucoside, rutin, and luteolin.


Introduction
Numerous scientific studies have investigated the role of intermediate derivatives of oxidation process, free radicals, in various (patho)physiological processes, such as development of diseases and aging, in particular (1,2).Free radicals have been established to determine the beginning of various diseases, including heart diseases, stroke, arteriosclerosis, diabetes, and cancer (3).All above-mentioned is the reason to look for effective compounds -antioxidants -that would reduce a harmful impact of free radicals.
Lately, increasing attention has been paid to the antioxidants of natural origin as the problem of safety of synthetic antioxidants is still open (4).The most important attention is paid to the search for antioxi-dants and their evaluation in medicinal and food raw materials of plant origin (5).A number of plants, their extracts, food products, and medicinal preparations appear to be the objects of research.Effectiveness and informative character of research, undoubtedly, depend on relevance, sensitivity, and efficiency of the methods chosen.
The following standard stable free radicals are often used to evaluate the antioxidant activity: DPPH, or 1,1-diphenyl-2-picrylhydrazyl radical, and ABTS, or 2,2'-azinobis(3-ethylbenzthiazoline-6-sulfonic acid) cation (6).Photocolorimetric method is used to determine the decrease in the activity of the above-mentioned radicals.Photocolorimetric method is not selective enough; therefore, to compare activity of single active compounds of the aliquot, purification of compounds or some other method is necessary.Purification of active compounds or their groups from the aliquots of plant origin is time-and labor-consuming and expensive solution.Besides, when extraction and depuration procedures are applied, some of the initial antioxidant activity is lost due to dissociation (or structural deformation) of compounds with antioxidant activity and partial yield (7).To avoid the abovementioned problems, new research technologies are required that would enable to determine the activity of different plant compounds directly in the tested aliquot and, at the same time, to maintain proportions, characteristic of that raw material or preparation.
Lately, composite methods of on-line determination have been developed.These methods are qualified as selective, short in analysis duration, and very sensitive for exact determination of activity and comparison.Separation of the investigated compounds by high-performance liquid chromatography (HPLC) together with postcolumn reaction is used for such methods.DPPH and ABTS radicals are used in postcolumn reaction, and the principle of common photocolorimetric method is followed (6).On-line postcolumn methods are employed in the rapid identification of antioxidants, determination of activity, and comparison.They are particularly informative and possess important advantages if compared with colorimetric methods (7).
The aim of this study was to develop and validate the postcolumn HPLC-DPPH method as well as its application in the evaluation of antioxidant activity of known and unknown compounds scavenging free radicals and existing in medicinal plant raw materials.

Material and methods
This paper describes optimization, validation, and application of HPLC-DPPH method on-line.Reference antioxidant trolox ((R)-6-methoxy-2,5,7,8-tetramethylchromane-2-carboxylic acid) was used for standardization of antioxidant activity.Trolox activity equivalent was calculated for individual compound and for the analyzed plant raw material.
Scheme of HPLC-DPPH system for on-line identification of free radical scavengers is presented in Fig. 1.The binary pump system Beckman solvent module 126 (Fullerton, USA) was used for the elution gradient formation.Chromatographic separation was performed at mobile phase flow rate of 0.4 mL/min.Elution gradient component A -0.1% TFA water solution, component B -0.1% TFA solution ACN.The following linear variation of gradient was used: 0 min, 95% A and 5% B; 45 min, 55% A and 45% B; 50 min, 55% A and 45% B; 55 min, 95% A and 5% B; 60 min, 95% A and 5% B. Rheodyne injector with a 20-µL injection loop was used for injection of aliquots.Analytes were separated in X-Terra RP 18 (Waters) analytical column (3.5 µm, 3.0×150 mm), where X-Terra RP 18 (3.5 µm, 3.0×20 mm, Waters) was used as precolumn.The separated compounds were identified with UV absorption detector Beckman System Gold 166 (Fullerton, USA) at the wavelength of 275 nm.From the detector, the mobile phase with analytes flowed through a mixing tee to the reaction coil, where DPPH reagent solution was supplied.HPLC pump Gilson pump 305 (Middleton, USA) was used to supply DPPH solution.The solution flow rate was 0.4 mL/min.The reaction coil was 3 m in length, made of PEEK (polyetheretherketone) tube with an inner diameter of 0.25 mm and outer diameter of 0.36 mm.The reaction coil was connected with UV/VIS type detector Gilson UV/VIS detector 118 (Middleton, USA), which measures absorption of flowing solution at the wavelength of 520 nm.The obtained data and equipment were managed with two computers, and original software System Gold (Beckman) and Unipoint (Gilson) was used.

Development and optimization of DPPH solution.
Medium pH influences kinetics of plant antioxidant reaction with DPPH radical.Koleva et al. (9) state the importance of maintaining the medium pH between 5.0 and 6.5, as this is an optimal condition for the reaction of hydrogen electron release.In the course of gradient elution of analytes, the medium pH may fluctuate.The fluctuations negatively influence the stability of the chromatogram baseline and, at the same time, S/N (ratio of analyte peak height to baseline noise) ratio.The instability of the baseline is named as the main problem in HPLC-DPPH analysis (4).
Eluents of the gradient described in the Material and methods have pH of <2.3.At low values of the medium pH, absorption of DPPH solution also decreases (9).The pH increase in the reaction solution in the reaction coil is possible only by means of adding some buffer solution to DPPH reagent solution.
The phosphatic buffer (Na 2 HPO 4 ) system, suggested by Dapkevičius et al. (5) was employed.DPPH crystals were dissolved in CAN, and 0.03 M of sodium hydrophosphate buffer solution was added in a 1:2 ratio, pH 7.6.As water content decreased during gradient analysis, sodium hydrophosphate crystals started to form in the reaction coil, which increased the instability of the baseline.The accumulation of hydrophosphate crystals might obstruct the reaction coil.The phosphatic buffer solution was refused.Pukalskas et al. (10) described the application of ammonium acetate buffer solution to DPPH solution.DPPH solution in ACN was prepared for the experiment, and 0.005 M of ammonium acetate (CH 3 COONH 4 ) buffer solution was added in a 1:2 ratio, pH 7.4.In the case of application of ammonium acetate buffer solution, no crystals were formed in the reaction coil; however, stabilization of the chromatogram baseline was unsuccessful.Therefore, ammonium acetate buffer solution was also refused.
Application of sodium citrate/citric acid buffer so-lution, described in publications by Kosar et al. (4,7), resulted in no crystals formed in the reaction coil.The baseline of chromatogram was stable.The decision was made to apply this buffer solution for the maintenance of optimal medium pH.To increase buffering capacity of solution, double concentration of sodium citrate as that suggested by Kosar was used.
For the final method, 0.1 M of sodium citrate buffer solution was prepared, and its pH was corrected with 0.05 M of citric acid solution (up to pH 7.6).The buffer solution was mixed with DPPH reagent dissolved in ACN solution, ACN being one of mobile phase components.Both solutions (buffer and reagent) were mixed in a 1:1 ratio.This proportion was chosen with the purpose to maintain sufficient solubility of DPPH in the mobile phase during gradient elution.
Choice of optimal concentration of DPPH solution is very important for qualitative evaluation of antioxidant activity.Increasing concentration of DPPH solution was shown to increase the peak area of active compound in the chromatogram.This dependence is shown in Fig. 2. Chlorogenic acid, hyperoside, and quercetin showed curves of similar character.The highest peak area was reached at the DPPH solution concentration of 0.02-0.03mg/mL, and further increase in the concentration had a slight if any impact on an increase in the peak area (quercetin curve).Trolox curve reached a plateau at the DPPH concentration close to 0.04 mg/mL.At the increasing concentration of DPPH solution, baseline noise also increased (Fig. 3).Instability of chromatogram baseline might negatively influence the accuracy of quantitative evaluation and determination of minimum concentration of active compound.DPPH solution at the concentration of 0.02-0.03mg/mL was decided to be the most suitable for analysis of aliquots of plant origin.At this concentration, chlorogenic acid, hyperoside, and quercetin (plant antioxidants) had the highest peak areas.At the concentration of 0.02 mg/mL, baseline noise was lower than that in case of 0.03 mg/mL; therefore, DPPH solution concentration of 0.02 mg/mL was chosen for further experiments.
Kinetics of plant antioxidant reaction with the
In the chosen method, the reaction coil of fixed length (3 m) was used.This length was sufficient to keep the reaction mixture in the coil for the indicated period.
Rate of DPPH solution supply to postcolumn directly determines the duration of presence of antioxidant and DPPH reagent in the reaction coil, which, in turn, influences baseline noise as well as peak height and area of analyte in a chromatogram.Therefore, choice of optimal rate of DPPH solution supply to postcolumn is necessary in order to have quantitative evaluation of antioxidant activity.
Low concentrations of known compounds were chosen for this experiment, which enables to have the most exact evaluation of the influence of rate of DPPH solution supply to postcolumn on baseline noise and peak height of analyte in the chromatogram (Fig. 4).At the increasing flow rate of DPPH solution, baseline noise and peak height of analyte decreased in the chromatogram, and on the contrary, at the decreasing flow rate -noise and peak height increased.
The flow rate of DPPH solution of 0.4-0.6 mL/min is the best for quantitative evaluation of antioxidants.For saving expensive DPPH reagent, flow rate of DPPH solution of 0.4 mL/min was chosen for further experiments.
Very little amounts of DDPH crystals were weighed for the preparation of DPPH solution; they were dissolved in a certain volume of ACN, and a particular amount of buffer solution was added.Some errors, which can distort quantitative evaluation of antioxidant activity, are possible each time when fresh reagent solution is produced.To avoid this, standardization of final DPPH solution should be performed before each run of the analysis.Standardization was carried out off-line; spectrophotometer DU-70 (Beckman) and 10-mm quartz cell were used.Wavelength was λ=520 nm; optical density of DPPH solution at 0.02 mg/mL concentration was corrected up to 0.500±0.005AU.
HPLC-DPPH method validation.For the purpose of direct validation of HPLC-DPPH method, several standard assessments of HPLC method have been done (Table 1).Limit of determination, as minimum detectable concentration, was determined for standards by signal-to-noise ratio of 7:1.The precision of present method was evaluated using repeatability and intermediate precision (day-to-day) of variation coefficients (RSD%), calculated from 5 replicated analyses.
Trolox content equivalent was used for quantitative evaluation of activity of antioxidant compounds present in extractions and pharmaceutical preparations from raw materials of medicinal plants.The equivalent (E T ) is expressed by the amount of trolox (µg), which under the same experimental conditions shows an appropriate free DPPH radical-scavenging activity, evaluated by trolox calibration curve (y=686344x+18680).
Antiradical activity equivalent (A) was used to evaluate the activity of antioxidantically active compounds in plant raw material.Antiradical activity equivalent of taken compound present in one amount of plant raw material -gram -is expressed by the amount of trolox (µg), which under the same conditions shows an appropriate free DPPH radical-scavenging activity, evaluated by the same trolox calibration curve as the taken compound: , (µg/g) Where: E T(comp) -trolox content equivalent (µg) of the taken antioxidantically active compound; V injinjected volume of the aliquot (mL); V extr -extraction amount of the analyzed plant raw material (mL); mexact amount of the analyzed plant raw material.
Application of HPLC-DPPH method on-line.Potential application of optimized HPLC-DPPH method was perfectly illustrated by the investigations aimed at quantitative evaluation of antioxidants, present in Crataegus (hawthorn), Origanum (oregano), and yarrow (Achillea) extracts.
Preparations of Crataegus are prescribed in functional heart disorders, after serious illness, in case of increased blood pressure.Commonly plant raw material is collected from Crataegus monogyna Jacq., C. rhiphidophylla, and C. laevigata.Various flavonoids and their glycosides (vitexin, vitexin-2"-O-rhamnoside, izovitexin, orientin, hyperoside, rutin, epicatechin and oligomeric procyanidins) have been determined in C. monogyna (12).UV/DPPH chro-matogram of ethanolic extract from flowers, obtained by performing evaluation of bioactive compounds of Crataegus by on-line HPLC-DPPH method, is presented in Fig. 5.The peak of hyperoside was dominant in the UV chromatogram; however, antioxidant activity of hyperoside, estimated by calculated trolox equivalent, was lower than chlorogenic acid (Table 2).
Using on-line HPLC-DPPH method, it was determined that vitexin rhamnoside, the dominant compound in leaves, was not a significant radical scavenger.The most active antioxidant in leaves of Crataegus, as well as flowers, was chlorogenic acid (Fig. 6, Table 3).
Preparations of Origanum are known for their antimicrobial, anti-inflammatory, demulcent, and painrelieving actions.A significant antioxidant determined in the herb of Origanum vulgare is phenolic acid.Rosmarinic acid is the dominant phenolic acid in Herba Origani as described by Ivanauskas et al. (13).Two main compounds influencing antioxidant activity the most were determined by on-line HPLC-DPPH method.This was rosmarinic acid and unknown compound, which peak was marked in the UV/DPPH chromatogram by number 2 (Fig. 7).UV/DPPH analysis of ethanolic extract from Origanum leaves revealed that the most active antioxidant was rosmarinic acid (Fig. 8).It was determined that the general antioxidant activity in the herb of Origanum was attributed to compounds, present in the leaves of Origanum.
Effectiveness of the designed instrumental and experimental setup was also confirmed by applying key principles of the coupled approach to the other method of chromatographic separation performed on HPLC column with different particle size, internal diameter, and chemical properties of sorbent.Separations were carried out using 0.1% TFA solution in water (solvent A) and 0.1% TFA solution in acetonitrile (solvent B) as mobile phase with the previously published ( 14) gradient elution program on a 5-µm AscentisTM RP-Amide analytical column (150×4.6 mm) guarded with

Fig. 5. HPLC-UV-DPPH coupled chromatograms of the flowers of Crataegus monogyna
For exact compound refer to Table 2.

Retention time, min
Absorbance, AU Absorbance, AU a 5-µm guard column SupelguardTM AscentisTM RP-Amide (20×4.00mm) (SUPELCO, Bellefonte, PA, USA).Since the separation conditions were optimized for a flow rate of 1.5 mL/min to ensure suffi-cient time for reaction kinetics, it was necessary to increase the length of reaction coil from 3 m to 15 m.Furthermore, preliminary experiments were performed to optimize the reaction time by setting different Medicina (Kaunas) 2009; 45( 5)

Fig. 6. HPLC-UV-DPPH coupled chromatograms of the leaves of Crataegus monogyna
For exact compound refer to Table 3. Investigation of antioxidant activity in herbal drugs using postcolumn HPLC method

Retention time, min
Absorbance, AU Absorbance, AU

Fig. 8. HPLC-UV-DPPH coupled chromatograms of the Origanum vulgare leaves
For exact compound refer to Table 5.

Table 5. Antioxidant activity of individual compounds in the extract of leaves of Origanum vulgare
namely flavonoids and phenolcarbonic acids.These bioactive compounds represent one of the most important groups of pharmacologically active substances in yarrow (15)(16)(17)(18)(19)(20).It has been reported (21,22) that antioxidant activity of extracts from several Achillea L. species might be correlated with their total phenolic and flavonoid contents.In our previous study (23), a considerable variation was observed in accumulation of phenolic compounds among the yarrow flowers, which, due to anatomical plant characteristics, constitute the most important part of the raw material of yarrow.Thereby, the investigation of contribution of individual constituents to antioxidant activity of yarrow extracts is of crucial importance for the comprehensive evaluation of pharmaceutical quality of the crude drug.

Conclusions
This study describes an application of a simple, quick, selective, and sensitive method.This method does not require any exceptional equipment, prepara-tion of complex aliquot and is available to many laboratories and scientific institutions.The optimized DPPH solution is suitable for both gradients used for analysis of different raw material samples.Using postcolumn HPLC-DPPH method, it was determined that vitexin rhamnoside, the dominant compound in the leaves of Crataegus monogyna, was not a significant radical scavenger.The most active antioxidant in the leaves and flowers of Crataegus was chlorogenic acid.It was determined that the most active antioxidant in Origanum vulgare raw material was rosmarinic acid.This study revealed that the main components among the identified analytes in the extracts of Achillea millefolium that possessed radicalscavenging properties were chlorogenic acid, luteolin-7-O-glucoside, rutin, and luteolin.