Stability of Polyphenols under Alkaline Conditions and the Formation of a Xanthine Oxidase Inhibitor from Gallic Acid in a Solution at pH 7.4

Abstract

The stability of 20 different types of polyphenols that are constituents of edible plants was evaluated under different pH conditions (in buffer solutions at pH 6.8, 7.4, and 8.3). Carnosol was the most unstable, followed by myricetin, quercetin, carnosic acid, nordihydroguaiaretic acid, baicalein, gallic acid, and hydroxytyrosol. They were more unstable at a higher pH, which may be attributed to their enhanced redox properties in alkaline solutions, determined by their antioxidant and prooxidant activity. The xanthine oxidase (XO) inhibitory activity of the reaction products under alkaline conditions was evaluated. XO activity was significantly inhibited by the product produced from gallic acid in pH 7.4 solution. The structural analysis of this reaction product revealed that a gallic acid dimer, purpurogallin-8-carboxylic acid, was formed and played a role in the resulting XO inhibitory activity.

Introduction

Polyphenols are natural phenols with at least two phenolic groups in their core structures and are of great interest owing to their potential in preventing various lifestyle-related diseases (Pandey and Rizvi, 2009; Scalbert et al., 2005). A large number of functional foods and food supplements containing polyphenols as active ingredients have been developed (Gharras, 2009). Polyphenols exhibit strong antioxidant activity that is closely linked to their high reactivity towards reactive oxygen species. Interestingly, their reactivity is enhanced under alkaline conditions (Lemańska et al., 2001). The high reactivity of polyphenols also indicates their high instability; thus, alkaline treatments can convert polyphenols to other compounds depending on their stability in the solution. In a living system, the pH levels of body fluids, including the blood plasma, are balanced (weakly basic, pH 7.4), and the digestive juices from the duodenum to ileum also have a basic pH (Evans et al., 1988). In food manufacturing, alkaline processes are often used, e.g., in the preparation of Chinese noodles, freeze-dried tofu, and bread with baking soda. Therefore, the stability and reactivity of polyphenols under alkaline conditions are important fundamental properties. Although there have been several studies on the stability of polyphenols, most of these studies are limited to specific systems such as the cell culture media and gastrointestinal systems (Bermúdez-Soto et al., 2007; Xiao and Högger, 2015; Quan et al., 2017) or to a limited number of target molecules such as tea catechins and berry polyphenols (Zhu et al., 1997; Su et al., 2003; Sang et al., 2005; Narita and Inouye, 2013; Correa-Betanzo et al., 2014). Friedman and Jürgens (2000) reported the degradation of several polyphenols under different alkaline pH conditions based on changes in the UV spectra; however, the degradation products were not identified. Previously, we have assessed the reaction products from polyphenols under various biochemical conditions and reported the reaction mechanism and bioactivity of the reaction products (Honda and Masuda, 2015). In this study, we evaluated the stability (under alkaline conditions) of 20 different types of polyphenols found in edible plants. We also investigated the xanthine oxidase (XO) inhibitory activity of the reaction products produced in pH 7.4 solution, which served as the target bioactivity.

Materials and Methods

Chemicals    Caffeic acid (purity > 98 %), luteolin (purity > 98 %), myricetin (purity > 97 %), hydroxytyrosol (purity > 98 %), resveratrol (purity > 99 %), and baicalein (purity > 98 %) were purchased from Tokyo Chemical Industry (Tokyo, Japan). Gentisic acid (purity > 98 %), protocatechuic acid (purity > 97 %), quercetin (purity > 95 % as dihydrate), and gallic acid (purity 97 %) were purchased from Nacalai Tesque (Kyoto, Japan). Catechin (purity > 98 % as hydrate), rosmarinic acid (purity > 97 %), chlorogenic acid (purity > 95 %), nordihydroguaiaretic acid (purity > 97 %), sinapic acid (purity > 98 %), and dihydrocaffeic acid (purity > 98 %) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Taxifolin (purity > 96 %) was obtained from Funakoshi (Tokyo, Japan). Carnosol (purity > 99 %) and carnosic acid (purity > 99 %) were isolated from the leaves of sage using a previously reported method (Masuda et al., 2001, 2004). Curcumin (purity > 99 %) was synthesized using Pabon's method (Pabon, 1964).

Instruments Analytical HPLC was performed using the LC-10AD high-pressure gradient system (Shimadzu, Kyoto, Japan) equipped with a SPD-M10Avp photodiode array detector (Shimadzu). The 3D data were analyzed by Class-VP software (V.5.032; Shimadzu). Preparative HPLC was performed using a LC-6AD pump (Shimadzu) equipped with a SPD-6A UV detector (Shimadzu). NMR spectra were obtained with a JNM-ECZ-400S spectrometer (JEOL, Tokyo, Japan). Mass spectra were measured using a JMS-T100 spectrometer (JEOL) in the direct analysis in real time (DART) ionization mode.

Stability test of polyphenols    Polyphenol dimethyl sulfoxide (DMSO) solution (100 µL; 10 mmol/L) was added to 12.5 mmol/L phosphate buffer with different pH values (6.8, 7.4, and 8.3; 900 µL). The mixtures were well stirred, and the obtained solutions were incubated at 35 °C for 1 h, after which 100 µL of 3 % HClO₄ was added to acidify the solution. An aliquot (10 µL) of the solution was injected into the HPLC instrument to analyze the residual polyphenol peak under the following conditions: column: TSKGEL ODS-80Ts (150 × 4.6 mm i.d.; Tosoh, Tokyo, Japan); solvent A: 1 % acetic acid in water; solvent B: acetonitrile; gradient conditions: linear gradient from 0.5 % solvent B to 100 % solvent B (20 min) [for carnosic acid: from 0.5 % solvent B to 100 % solvent B (30 min)]; flow rate: 1 mL/min; detection wavelength: 280 nm. A control sample solution was prepared by mixing polyphenol DMSO solution (100 µL), 0.1 % phosphoric acid (900 µL), and 3 % HClO₄ (100 µL). The instability of polyphenols at each pH level was evaluated according to the parameter “decreased ratio”, which was calculated using the following equation: Decreased ratio (%) = [(polyphenol peak area of the control solution − polyphenol peak area of 1 h-incubated solution at a specific pH) × 100] / (polyphenol peak area of the control solution).

XO inhibitory assay    XO inhibitory assay was performed according to a previously reported method (Honda and Masuda, 2015) with slight modifications. The reaction medium consisted of 10 µL of 1 mmol/L xanthine in DMSO and 160 µL of 12.5 mmol/L phosphate buffer (pH 7.4). The solution was pre-heated at 37 °C for 5 min. Then, 10 µL of the test sample solution was added to this solution, which was prepared for alkaline degradation with or without incubation, followed by the addition of 20 µL of 0.045 unit/mL XO solution (pH 7.4 phosphate buffer) to prevent further alkaline degradation of the polyphenols. After incubation at 37 °C for 10 min, 25 µL of 3 % aqueous HClO4 was added to quench the reaction. An aliquot (20 µL) of the solution was injected into the HPLC column to quantitatively evaluate the amount of uric acid produced. HPLC analysis was performed under the following conditions: column: Mightysil RP-18 GP Aqua (250 × 4.6 mm, i.d., 5 µm; Kanto Chemical, Tokyo, Japan); solvent: methanol-0.1 % phosphoric acid in water (2.5:97.5, v/v); flow rate: 1.0 mL/min; detection wavelength: 290 nm; temperature: 35 °C. The inhibition percentage was calculated according to the following equation: Inhibition (%) = [(peak area of uric acid in the control experiment) − (peak area of uric acid in the sample experiment) × 100] / (peak area of uric acid in the control experiment). The XO inhibitory activity of the pure samples including the positive control allopurinol was measured using 10 µL of the samples in DMSO-12.5 mmol/L phosphate buffer (1:9).

Isolation and structural analysis of an XO inhibitor formed from the alkaline reaction of gallic acid    Gallic acid (500 mg) was dissolved in 250 mL of 12.5 mmol/L phosphate buffer (pH 7.4; 35 °C) with stirring. The solution was incubated at 35 °C for 2.5 h with O₂ gas bubbling every 30 min. After the addition of 3 % HClO₄ (70 mL), the solution was extracted three times with ethyl acetate (320 mL for each extraction). The ethyl acetate solution was evaporated to dryness and dissolved in DMSO (0.6 mL) and methanol (0.4 mL). The mixture was purified by preparative HPLC to separate a single product 1 (18 mg) at a retention time of 16 min under the following conditions: column: TSKGEL ODS-80Ts (250 × 20 mm i.d.); solvent: 1 % acetic acid in water-CH₃CN (75:25); flow rate: 9.6 mL/min; detection wavelength: 320 nm. Analytical data for 1: HR-MS (DART, positive): m/z 265.0355; calcd for C₁₂H₉O₇: 265.0348; ¹H NMR (400 MHz, CD₃OD): 6.92 (1H, s, H-1), 7.63 (1H, s, H-7), 8.11 (1H, s, H-9); ¹³C NMR (400 MHz, CD3OD): 113.1 (C-1), 113.6 (C-8), 115.0 (C-4a), 124.6 (C-9a), 130.9 (C-4), 136.8 (C-9), 138.1 (C-2), 150.9 (C-3), 152.0 (C-6), 153.2 (C-7), 168.6 (C-10), 182.7 (C-5); observed correlations in the HMBC spectrum: from H-1 to C-9 and C-4a, from H-7 to C-5, C-6, C-8, C-9, C-9a, and C-10, and from H-9 to C-7, C-9a, and C-10.

Results and Discussion

Polyphenol stability in different pH solutions    To determine the stability of polyphenols under different pH conditions, the decrease in polyphenol concentration in different pH buffer solutions (pH 6.8, 7.4, and 8.3) was measured. Fig. 1 shows the chemical structures of the polyphenols investigated in this study, and Fig. 2 shows the decreased ratio (%) of the polyphenols in different pH solutions (incubated at 35 °C for 1 h) to polyphenols in 0.1 % phosphoric acid (used as the control). As shown in Fig. 2, carnosol, myricetin, quercetin, and carnosic acid were highly unstable, and nordihydroguaiaretic acid, baicalein, gallic acid, and hydroxytyrosol were moderately unstable. The concentration of other polyphenols did not show significant changes in the solutions under different pH conditions. Moreover, some polyphenols showed an increasing pattern, which may be attributed to their instability in the acidic control solution. The chemical structures of the highly and moderately unstable polyphenols indicated that the presence of common substructures such as trihydroxybenzene or catechol (dihydroxybenzene) substituted by saturated alkyl groups. These substructures are important for the strong antioxidant and prooxidant properties of the polyphenolic compounds. Both properties can cause the instability of polyphenols containing these substructures under oxidative conditions including an O₂ atmosphere. Carnosol and carnosic acid are structurally similar diterpenoid antioxidants. Previous studies have indicated that the radical scavenging activity of carnosic acid is stronger than that of carnosol (Wellwood and Cole, 2004). However, Zang et al. (2012) reported that carnosol is more unstable than carnosic acid under specific conditions and that temperature could have a strong effect on the instability of carnosol. Although the effects of temperature are still unclear, the reaction temperature (35 °C) in our experiments may affect the observed instability. Myricetin and quercetin are well-known potent antioxidative flavonoids with high redox reactivity. Although myricetin has a larger number of active phenoxy groups, most studies reported that the antioxidant activity of quercetin is higher than that of myricetin (Rice-Evans et al., 1996; Pietta, 2000). Therefore, the observed weaker antioxidant activity of myricetin may be influenced by the prooxidant properties of the B ring (a trihydroxybenzene substructure) (Aruoma et al., 1993) and may depend on the different assay procedures used (Roginsky and Lissi, 2005). Recently, Atala et al. (2017) reported that the chemical oxidation rate of myricetin is higher than that of quercetin, which is consistent with our present results. Nordihydroguaiaretic acid and hydroxytyrosol have highly reactive saturated alkyl-substituted catechol structures; myricetin, baicalein, and gallic acid also have reactive 1,2,3-trihydroxybenzene substructures, which may explain their decreased concentration in the solutions.

Fig. 1.

Chemical structures of the examined polyphenols.

Fig. 2.

Decreased ratio (%) of polyphenols in solutions with different pH values (pH 6.8, 7.4, and 8.3; incubated for 1 h at 35 °C) to polyphenols in 0.1 % phosphoric acid (used as the control). Data are expressed as the mean ± SE (n = 3).

XO inhibitory activity of seven polyphenols Some polyphenols are rapidly broken down especially in solutions with an alkaline pH, forming reaction products. Owing to their difference in structure compared with that of the original polyphenols, the reaction products may have bioactivities that are different from those of the original polyphenols. In the present study, we measured the XO inhibitory activity of these products, which served as the target bioactivity. XO is an enzyme that catalyzes the final reaction of purine catabolism in humans and is mainly active in the liver. XO produces uric acid, causing gout when it is accumulated in the plasma, which is a common lifestyle-related disease. Ingested polyphenols are exposed to alkaline fluids in the intestinal tract and blood vessels; thus, polyphenols that are unstable under alkaline conditions may change their structures to reach the liver and potentially react with XO. Fig. 3 shows the XO inhibitory activity of seven unstable polyphenols and their reaction products after 1 h in pH 7.4 solution (50 µM equivalent). Although carnosic acid, carnosol, nordihydroguaiaretic acid, and gallic acid did not exhibit significant inhibitory activity at the concentration levels examined initially, baicalein, quercetin, and myricetin demonstrated strong XO inhibitory activity (over 80 %). The strong XO inhibitory activity of baicalein, quercetin, and myricetin has been reported previously with IC₅₀ values of 2.8, 2.6, and 2.4 µM, respectively (Cos et al., 1998). Considering their small IC₅₀ values, the polyphenols remaining in the pH 7.4 solution may inhibit XO activity, and their reaction products may have a similar potent inhibitory activity. On the other hand, the reaction products of carnosic acid, carnosol, and gallic acid exhibited an inhibitory activity of 8 %, 30 %, and 40 %, respectively, even though almost no activity was observed for the polyphenols initially. The results indicated that new XO inhibitory products were produced in alkaline solutions of carnosic acid, carnosol, and gallic acid when incubated for 1 h at 35 °C under aerobic conditions.

Fig. 3.

XO inhibitory activity of selected polyphenols and their reaction products at a concentration of 50 µM or equivalent (for reaction products). Data are expressed as the mean ± SE (n = 3).

Identification of an XO inhibitor produced from gallic acid The weak alkaline treatment (pH 7.4) of gallic acid resulted in enhanced XO inhibitory activity as shown in Fig. 3. Hence, we attempted to identify the XO inhibitors that were produced. Fig. 4 shows the analytical HPLC profile of the reaction of gallic acid in pH 7.4 solution. The main product peak was observed at a retention time of 22.7 min with unreacted gallic acid at a retention time of 4.5 min by PDA detection, indicating that the main product had strong XO inhibitory activity. To identify the structure of this reaction product, 500 mg of gallic acid was reacted with 250 mL of pH 7.4 buffer at 35 °C for 2.5 h. The solution was then extracted with ethyl acetate, and the product was purified by preparative HPLC to obtain the pure compound 1 (18 mg). The chemical structure of 1 was determined by HR-MS, ¹H-NMR, ¹³C-NMR, and HMBC analyses, which identified its structure as purpurogallin-8-carboxylic acid (Fig. 5). Purpurogallin-8-carboxylic acid was previously detected in the iodate oxidation of a mixture of pyrogallol and gallic acid by Crow and Haworth (1951). Also, purpurogallin-8-carboxylic acid has been identified from fermented black tea and has high anti-inflammatory activity in the TPA-induced mice ear edema assay (Sang et al., 2004). The parent structure of the compound, purpurogallin, has been reported to be a potent XO inhibitor (Honda et al., 2017). The XO inhibitory activity of product 1 was measured, and the IC₅₀ value was 4.4 µM. Although this IC₅₀ value is much lower than that of purpurogallin (0.2 µM), it is comparable to that of several XO inhibitory flavonoids (Cos et al., 1998). Gallic acid is widely distributed in edible plants mainly in the ester form (Haslam and Cai, 1994), and it can be found in human plasma after the ingestion of tea (Shanoush et al., 2001). Ellagic acid is a well-known oxidative coupling product derived from gallic acid, which also has strong XO inhibitory activity (IC₅₀ of ellagic acid: 3.1 µM; Hatano et al., 1990). However, in this study, ellagic acid was not observed in the alkaline solution, and purpurogallin carboxylic acid 1 was identified as the sole potent XO inhibitor. Therefore, product 1 might be produced in body fluids as an XO inhibitor that can regulate human liver XO activity when gallic acid and gallic acid-bearing compounds are ingested as food constituents.

Fig. 4.

Analytical HPLC data of the solution of gallic acid after 1 h of reaction. Analytical conditions: column, TSKGEL ODS-80Ts (150 × 4.6 mm i.d.); solvent A, 1 % acetic acid in H2O; solvent B, CH3CN; gradient conditions, linear gradient from 5 % solvent B (0 min) to 40 % solvent B (40 min) followed by 100 % solvent B (50 min). 100 % solvent B was used for column washout, and for the next 10 min, 5 % solvent B was eluted for reconditioning: flow rate, 1 mL/min; detection, 254, 280, and 320 nm. *indicates impurity peaks or solvent shock peaks under washout conditions.

Fig. 5.

Chemical structure of the identified XO inhibitor derived from gallic acid at pH 7.4. Position numbering was based on the IUPAC name of purpurogallin.

Acknowledgements    This study was supported by a research grant from the Mishima Kaiun Memorial Foundation (Tokyo) for S. H. and by a JSPS Kakenhi grant (Grant No. JP15H02892) for T. M.

References

•  Aruoma,  O. I.,  Murcia,  A.,  Butler,  J., and  Halliwell,  B. (1993). Evaluation of the antioxidant and prooxidant actions of gallic acid and its derivatives. J. Agric. Food Chem., 41, 1880-1885.
•  Atala,  E.,  Fuentes,  J.,  Wohrhan,  M. J., and  Speisky,  H. (2017). Quercetin and related flavonoids conserve their antioxidant properties despite undergoing chemical or enzymatic oxidation. Food Chem., 234, 479-485.
•  Bermúdez-Soto,  N.-J.,  Tomás-Barberan,  F.-A., and  García-Conesa,  M.-T. (2007). Stability polyphenols in chokeberry (Aronia melanocarpa) subjected to in vitro gastric and pancreatic digestion. Food Chem., 102, 865-874.
•  Correa-Betanzo,  J.,  Allen-Vercoe,  E.,  McDonald,  J.,  Schroeter,  K.,  Corredig,  M., and  Paliyath,  G. (2014). Stability and biological activity of wild blueberry (Vaccinium angustifolium) polyphenols during simulated in vitro gastrointestinal digestion. Food Chem., 165, 522-531.
•  Cos,  P.,  Ying,  L.,  Calomme,  M.,  Ku,  J. P.,  Cimanga,  K.,  Poel,  B. P.,  Pieters,  L.,  Vlietinck,  A. J., and  Berghe,  D. V. (1998). Structure-activity relationships and classification of flavonoids as inhibitors of xanthine oxidase and superoxide scavengers. J. Nat. Prod., 61, 71-76.
•  Crow,  W. D. and  Haworth,  R. D. (1951). Purpurogallin. Part VII. Constitution of purpurogallin-carboxylic acid. J. Chem. Soc., 1951, 1325-1327.
•  Evans,  D. F.,  Pye,  G.,  Bramley,  R.,  Clark,  A. G.,  Dyson,  T. J., and  Hardcastle,  J. D. (1988). Measurement of gastrointestinal pH profiles in normal ambulant human subjects. Gut, 29, 1035-1041.
•  Friedman,  M. and  Jurgens,  H. S. (2000). Effect of pH on the stability of plant phenolic compounds. J. Agric. Food Chem., 48, 2101-2110.
•  Gharras,  H. E. (2009). Polyphenols: food sources, properties and applications–a review. Int. J. Food Sci. Technol., 44, 2512-2518.
•  Haslam,  E. and  Cai,  Y. (1994). Plant polyphenols (vegetable tannins): gallic acid metabolism. Nat. Prod. Rep., 1994, 41-66.
•  Hatano,  T.,  Yasuhara,  T.,  Yoshida,  R.,  Agata,  I.,  Noro,  T., and  Okuda,  T. (1990). Effects of interaction of tannins with co-existing substances. VII. Inhibitory effect of tannins and related polyphenols on xanthine oxidase. Chem. Pharm. Bull., 38, 1224-1229.
•  Honda  S. and  Masuda  T. (2015). Polyphenols: functional chemicals based on their chemical reactions from antioxidation to inter-substance reactions. Kagaku to Seibutsu, 53, 442-448 (in Japanese).
•  Honda,  S.,  Fukuyama,  Y.,  Nishiwaki,  H.,  Masuda,  A., and  Masuda,  T. (2017). Conversion to purpurogallin, a key step in the mechanism of the potent xanthine oxidase inhibitory activity of pyrogallol. Free Radic. Biol. Med., 106, 228-235.
•  Lemańska,  K.,  Szymusial,  H.,  Tyrakowska,  B.,  Zieliński,  R.,  Soffers,  A. E. M. F., and  Rietjens,  I. M. C. M. (2001). The influence of pH on antioxidant properties and the mechanism of antioxidant action of hydroxyflavones. Free Radic. Biol. Med., 31, 869-881.
•  Pabon,  H. J. J. (1964). A synthesis of curcumin and related compounds. Recl. Trav. Chim. Pays-Bas., 83, 379-386.
•  Masuda,  T.,  Inaba,  Y., and  Takeda,  Y. (2001). Antioxidant mechanism of carnosic acid: structural identification of two oxidation products. J. Agric. Food Chem., 49, 5560-5565.
•  Masuda,  T.,  Kirikihira,  T.,  Takeda,  T., and  Yonemori,  S. (2004). Thermal recovery of antioxidant activity from carnosol quinone, the main antioxidation product of carnosol. J. Sci. Food Agric., 84, 1421-1427.
•  Narita,  Y. and  Inouye,  K. (2013). Degradation kinetics of chlorogenic acid at various pH values and effects of ascorbic acid and epigallocatechin gallate on its stability under alkaline conditions. J. Agric. Food Chem., 61, 966-972.
•  Pandey,  K. B. and  Rizvi,  S. I. (2009). Plant polyphenols as dietary antioxidants in human health and disease. Oxid. Med. Cell Longev., 2, 270-278.
•  Pietta,  P.-G. (2000). Flavonoids as antioxidants. J. Nat. Prod., 63, 1035-1042.
•  Quan,  W.,  Tao,  Y.,  Lu,  M.,  Yuan,  B.,  Chem,  J.,  Zeng,  M.,  Qin,  F.,  Guo,  F., and  He,  Z. (2018). Stability of the phenolic compounds and antioxidant capacity of five fruit (apple, orange, grape, pomelo and kiwi) juices during in vitro-simulated gastrointestinal digestion. Int. J. Food Sci. Technol., 52, 1131-1139.
•  Roginsky  V. and  Lissi,  E. A. (2005). Review of methods to determine chain-breaking antioxidant activity in food. Food Chem., 92, 235-254.
•  Rice-Evans,  C. A.,  Miller,  N. J., and  Paganga,  G. (1996). Structure-antioxidant activity relationships of flavonoids and phenolic acids. Free Radic. Biol. Med., 20, 933-956.
•  Sang,  S.,  Lambert,  J. D.,  Tian,  S.,  Hong,  J.,  Hou,  Z.,  Ryu,  J.-H.,  Stark,  R. E.,  Robert,  T. R.,  Huang,  M.-T.,  Yang,  C. S., and  Ho,  C.-T. (2004). Enzymatic synthesis of tea theaflavin derivatives and their anti-inflammatory and cytotoxic activities. Bioorg. Med. Chem., 12, 459-467.
•  Sang,  S.,  Lee,  M.-J.,  Hou,  Z.,  Ho,  C.-T., and  Yang,  C. S. (2005). Stability of tea polyphenol (-)-epigallocatechin-3-gallate and formation of dimers and epimers under common experimental conditions. J. Agric. Food Chem., 53, 9478-9484.
•  Scalbert,  A.,  Manach,  C.,  Morand,  C.,  Rémésy,  C., and  Jiménez,  L. (2005). Dietary polyphenols and the prevention of diseases. Crit. Rev. Food Sci. Nutri., 45, 287-306.
•  Shahrzad,  S.,  Aoyagi,  K.,  Winter,  A.,  Koyama,  A., and  Bitsch,  I. (2001). Pharmacokinetics of gallic acid and its relative bioavailability from tea in healthy humans. J. Nutr., 131, 1207-1210.
•  Su,  Y. L.,  Leung,  L. K.,  Huang,  Y., and  Chen,  Z.-Y. (2003). Stability of tea theaflavins and catechins. Food Chem., 83, 198-195.
•  Wellwood,  C. R. L. and  Cole,  R. A. (2004). Relevance of carnosic acid concentrations to the selection of rosemary, Rosmarinus officinalis (L.), accessions for optimization of antioxidant yield. J. Agric. Food Chem., 52, 6101-6107.
•  Xiao,  J. and  Högger,  P. (2015). Stability of dietary polyphenols under the cell culture conditions: avoiding erroneous conclusions. J. Agric. Food Chem., 63, 1547-1557.
•  Zang,  Y.,  Smuts,  J. P.,  Dodbiba,  E.,  Rangarajan,  R.,  Lang,  J. C., and  Armstrong,  D. W. (2012). Degradation study of carnosic acid, carnosol, rosmarinic acid, and rosemary extract (Rosmarinus officinalis L.) assessed using HPLC. J. Agric. Food Chem., 61, 9305-9314.
•  Zhu,  Q. Y.,  Zhang,  A.,  Tsang,  D.,  Huang,  Y., and  Chen,  Z.-Y. (1997). Stability of green tea catechins. J. Agric. Food Chem., 45, 4624-4628.