COOH
3. Analytical Methods
3.1. Sample Treatment
The quality of a chemical analysis is as trustworthy as the step that is of the least accuracy. No doubt, sample treatment procedure is a step that significantly impacts data quality but also assay through-put. Over the years, many sample treatment methods have been developed to determine polyphenolic and simple phenolic com-pounds in various sample types. This step varies a great deal from protocol to protocol and among techniques owing to the diver-sity in matrices. The basic goals of the sample preparation process in the analysis of phenolic compounds are as follows:
I. Analytes isolation from the primary matrix and the associ-ated interfering compounds present in the sample.
II. Phase switching to one suitable for the chosen analytical technique.
III. Enrichment of analytes to allow their determination by the analytical method chosen.
A parameter which should be taken into consideration when making the decision of the sample treatment technique consists
70 Stalikas
of such information as the number of samples to be analyzed.
The question is whether the planned procedure is going to be used in carrying out single, multiple, or even routine analysis.
In the latter case, the techniques facilitating automation and low cost per analysis are favorable. However, one should bear in mind that the absence of the sample preparation stage in an analytical procedure is the most convenient and preferable way of sample preparation in order to minimize changes in the sample compo-sition. Hence, reducing the number of operations performed on the primary sample should be aimed by the analyst.
It is significant to choose the optimal treatment of matrix, based on the chemical structures and properties of analyzed compounds. There are three main types of phenolic-containing matrices with respect to the treatment needs: plants, foods and liquid samples, the latter including biological fluids and beverages.
Sample preparation for the analysis of phenolics can range from the simple “filter-and-inject” procedures to the more elab-orated hydrolysis/extraction/cleanup. Because of the assorted types of phenolics and the different matrices with many inter-fering components, the choice of the technique for the treat-ment differs from one another. In some cases, only a one-step extraction and/or simple cleanup procedure are sufficient before analysis, but the described assays almost invariably include two or more steps of preparation. Obviously, each step contributes to higher sensitivity and selectivity but, as mentioned above, it could increase the number of errors through introducing interferents and artifacts and decrease the recovery.
The solid samples are usually subject combinations of sieving, milling, or grinding and homogenization before further treat-ment for the determination of the extractable analytes. Liquid samples are filtered and/or centrifuged before isolation or sep-aration and detection. In the case of wines, the alcohol is usually removed from the sample via rotary evaporation, and the residue is taken up in a small volume of the solvent for subsequent anal-ysis. Air-drying or even more freeze-drying are universal steps for drying or condensation of primary samples.
An important aspect of phenolic analysis is whether to deter-mine the target analytes in their various conjugated or free forms (e.g., aglycones). In neutraceuticals and food products, researchers are usually interested in the intact conjugates and, for the classification of plant species, intact flavonoid profiles in plants are determined. In many other instances, the knowledge of total aglycone content is required. In biological fluids, flavonoids exist as glucuronide and sulfate conjugates. Therefore, a hydrolysis-digestion step is used to disrupt glycoside or sulfur linkages. When the glycosylated form of the flavonoids is the case, digestion is omitted.
Occurrence and Analysis of Phenolic Compounds 71
3.1.1. Hydrolysis of Phenolics
Acid hydrolysis and saponification are the most common means of releasing the phenolic acids, even though it is still unclear how much of the acids decompose under these conditions. Usually, acid ranges from 1 to 2 N aqueous HCl and the reaction times from 30 min to 1 h. Krygier et al. reported that loss under acidic conditions varies with the phenolic acid fluctuating from 15 to 95% for o-coumaric and sinapic acids, respectively (28). Saponifi-cation entails treating the sample with a solution of NaOH with reported concentrations ranging from 1 to 4 M. Some investiga-tions recommend that such reacinvestiga-tions be carried out in the dark, as well as under an inert atmosphere such as argon or nitrogen gases (29). Andreasen et al. discuss and compare several different enzyme preparations for the release of phenolic acids from the cell wall of rye grains (30). Yu et al. reported that a sequential acid, α-amylase, and cellulose hydrolysis might be applicable to the release of phenolic acids from barley (31).
In flavonoids, the hydrolyses frequently used to remove the sugar moieties from glycosides are acidic, basic or enzymatic. The hydrolysis process should be a compromise to minimize degra-dation reactions of glycosides and to achieve complete release of aglycones. Phenolic extract of sunflower honey was hydrolyzed in 2 N NaOH (32) while the glycosides of flavones and flavonols were hydrolyzed in refluxing 1.2 N HCl in 50% MeOH/H2O (v/v) (33). For plasma, serum, and urine, flavonoids may be first hydrolyzed withβ-glucuronidase, sulfatase, or a mixture contain-ing both enzymes (34). To ensure that these enzymes are active in the incubates, in certain cases,13C-labeled flavonoid conjugates have been available (35).
3.1.2. Extraction – Cleanup
Analytes have to be isolated, completely or to a certain degree, from the matrices in which they exist naturally, before analysis.
Extraction serves this task along with the additional one to isolate analytes from potentially interfering sample components while bringing these analytes into a form suitable for analysis. Extrac-tion of phenolic compounds is influenced, apart from the stor-age time and conditions, by their chemical nature, the extraction method employed, sample particle size as well as the presence of interfering substances.
Liquid–liquid and solid–liquid extractions are the most com-monly used procedures prior to analysis of polyphenolics and sim-ple phenolics. Widely accepted extraction solvents are alcohols (methanol, ethanol), acetone, diethyl-ether, and ethyl acetate.
However, very polar phenolic acids (e.g., benzoic, cinnamic acids) cannot be extracted completely with pure organic solvents, and mixtures of alcohol–water or acetone–water are recommended.
Process conditions such as pH, temperature, sample-to-solvent volume ratio, and the number and time intervals of individual
72 Stalikas
extraction steps also play an important role in the extraction pro-cedure. Extractions are repeated 2–3 times and extracts are com-bined.
Extraction of flavonoids from biological matrices is one of the fastest and less time consuming tasks. In a recent attempt, rat plasma was acidified with 0.25 N HCl, mixed with ethyl acetate, vortexed, and centrifuged (36). The upper organic phase was evaporated to dryness and the residue was reconstituted in the mobile phase for HPLC analysis. To quote another example, for quercetin and kaempferol in urine, 1 mL of 25% HCl was added to 4.0 mL of human urine and mixed well (37). After the urine sample had been hydrolyzed for 30 min at 80◦C, 5.0 mL of ether was added. The hydrolyzed sample was extracted for 5 min and the 4.0 mL ethereal phase was evaporated to dryness under nitrogen stream. The residue was reconstituted with 100 L of mobile phase and an aliquot of 20 L of the resulting solution was injected into the HPLC system.
Soxhlet extraction is used less frequently to isolate flavonoids from solid samples. Various flavonoids or phenolic acids were extracted from Tilia europea, Urtica dioica, Mentha spicata, Hypericum perforatum, and Echinacea purpurea after 12 h Soxh-let extraction with methanol (38, 39).
A feature that exists with the above-mentioned conventional extractions is that they influence the integrity of flavonoid glyco-side during the prolonged extraction, thus affecting the repro-ducibility and reliability (40). Supercritical fluid extraction has increasingly gained momentum in both food and pharmaceutical industries. The intrinsic low viscosity and high diffusivity of super-critical CO2has granted higher separation speed and efficiency to this mode of extraction, providing relatively clean extracts. The solvating power of a supercritical fluid is varied by controlling the pressure or by adding organic modifiers such as methanol.
Supercritical fluid extraction was compared to Soxhlet extraction, steam distillation and maceration for the isolation of the active components present in chamomile flower heads (41). The recov-ery of the flavonoid apigenin obtained by supercritical CO2 after a 30-min extraction at 200 atm and 40◦C was 71.4% compared to Soxhlet extraction performed for 6 h and 124.6% compared to maceration performed for 3 days. However, the highly polar flavonoid apigenin-7-glucoside was not extracted by 100% CO2 and the addition of the polar modifier methanol (5%, v/v) to the CO2 fluid was indispensable. This technique is applicable to medicinal and other plant samples and it can be combined also with other sample preparation techniques.
Finally, microwave irradiation and sonication have been suc-cessfully used to enhance extraction of phenolic acids from Echinacea purpurea (42) while a newly designed dynamic microwave-assisted extraction system has been developed for the
Occurrence and Analysis of Phenolic Compounds 73
continuous and rapid extraction of flavonoids from Saussurea medusa Maxim dried cell cultures (43). By comparing the dynamic microwave-assisted extraction with dynamic solvent extraction without microwave assistance, the former showed obvi-ous advantages related to the short extraction time and high effi-ciency.
Despite the merit of extraction in sample treatment, extra steps may be required for the removal of non-phenolic sub-stances. Although the sample preparation for contemporary ana-lytical techniques (e.g., liquid chromatography–(tandem) mass spectrometry) does not need to be as elaborate as others (e.g., liquid chromatography–ultraviolet), it remains pivotal to remove matrix components that might contaminate the system, when high sensitivity is needed, or give rise to unreliable results (44).
In many cases, biological samples cannot be assayed directly but require a treatment to free from endogenous proteins, carbohy-drates, salts, lipids, etc.
Solid-phase extraction is a reasonable choice for the precon-centration and cleanup procedure for crude plant extracts and biological matrices. Protein precipitation is the simplest means of sample treatment in order to analyze natural products in bioflu-ids and should normally precede this kind of extraction. There is a consistence to the choice of sorbents for isolating the pheno-lic acids and flavonoids. The C18 bonded silica is the sorbent of choice; sample solution and solvents are usually slightly acidified to prevent ionization of the flavonoids, which would reduce their retention. In a recent study, different sample preparation methods for human plasma phenolic compounds (six phenolic acids, five flavonoids, trans-resveratrol, and tyrosol) were compared (45).
These treatments included solid-phase extraction, extraction with methanol, removal of plasma proteins with different deproteiniza-tion agents, and inhibideproteiniza-tion of enzymatic plasma activity. Aiming to quantitate the whole set of compounds, in this case, the most suit-able approach was to inhibit enzymatic activity and then depro-teinize with acidified ethanol.
Hydrolyzed and non-hydrolyzed acidified urine were ana-lyzed, elsewhere, by passing it through Amberlite XAD-2 parti-cles and stirred to retain the phenolic compounds on the sur-face of the nonionic Amberlite particles (46). Two hundred and fifty microliters of plasma were diluted and acidified with 0.5%
formic acid before application to a C18sorbent. Urine was passed through Amberlite XAD-2 particles and stirred to retain the phe-nolic compounds on the surface of the nonionic Amberlite parti-cles for hydrolyzed and non-hydrolyzed urine.
The prominent solid-phase microextraction mode was also employed to extract genistein and daidzein from human urine in combination with liquid chromatographic analysis. A Carbowax-templated poly(divinylbenzene) resin proved to be the best fiber,
74 Stalikas
with a 5-min extraction at pH 4 and a temperature of 35◦C, which accommodates the need for absence of organic solvents (47).
A relatively new solid-phase extraction method used a molec-ularly imprinted polymer as the sorbent to determine quercetin in red wine (48). The recovery was over 98% when using methanol containing 15% acetic acid or acetonitrile containing 10% aque-ous triethylamine, as eluent. The molecularly imprinted polymer was proved to be highly selective for the target analyte enhanc-ing, at the same time, the intensity of the quercetin and reducing the complexity of the chromatographic trace. Another molecu-larly imprinted polymer was evaluated toward six phenolic acids extracting selectively the analytes from Melissa officinalis (49).
Matrix solid-phase dispersion is an alternative for sample preparation workable for liquid and semi-liquid samples. Sample extraction and cleanup are carried out simultaneously with, gen-erally, good recoveries and precision. Matrix solid-phase disper-sion is frequently used to determine pesticides in, e.g., foods, but application to flavonoid analysis was reported only recently. For the determination of isoflavone aglycones and glycosides in Radix astragali, this extractive cleanup step was compared to Soxhlet and ultrasonic extraction with respect to the extraction capacity (50). For the aglycones, matrix solid-phase dispersion yielded the best extraction efficiency but for the glycosides Soxhlet proved to be more efficient.
4.