Bookmark and Share

From: Food Quality & Safety magazine, December/January 2007

Policing Polyphenols

Detection Technology Aids Polyphenol Identification and Characterization

by Jerry Fireman

A substantial research effort is being devoted to better understand the protection against various diseases that epidemiological studies have shown to be provided by certain fruits and vegetables. Researchers generally believe that this protection comes from the activity of antioxidant compounds, particularly polyphenols, found in abundance in many fruits and vegetables.

Polyphenols are a group of chemical substances that are characterized by the presence of more than one phenol group per molecule. These compounds are responsible for providing the coloring of many plants. Various polyphenols have been shown in epidemiological studies to prevent cancers of the colon, esophagus, liver, stomach, lung, breast, pancreas and skin. The result is that a considerable amount of attention is now being focused on polyphenols, and particularly the category of polyphenols known as flavonoids which have been demonstrated in a number of studies to be powerful antioxidants.

Antioxidants are believed to help prevent disease by binding with and destroying free radicals, reducing oxidative damage to cells and biochemicals. (For example, oxidation by free radicals is a precursor to cardiovascular disease.) Therefore, it is becoming increasingly important to identify antioxidant compounds in foods, as well as in urine, tissue and plasma samples, in order to develop disease models and improve the design of epidemiological studies.

One of the more challenging aspects of this research is the identification and accurate measurement of these antioxidants amid the many types and varieties of fruits and vegetables where the compounds can be present in different quantities. For example, Dr. Ronald L. Prior of USDA has determined that the total antioxidant activity of a strawberry is 16 times that of a honeydew melon.

The standard approach for the identification of flavonoids combines the separation capabilities of high-performance liquid chromatography (HPLC) with the pinpoint molecular identification capabilities of mass spectroscopy (MS). This approach has been involved in the discovery of many polyphenols and provides the gold standard of positive identification. On other hand, the large number of polyphenols found in many fruits and vegetables and the large universe of potential plants whose antioxidant activity is of interest create the need for a method of quickly screening food, tissue and plasma samples for a large number of potential antioxidants in a minimal amount of time.

Investigation of Coulometric Array Detection

HPLC/MS is not well suited for this task because of the limited ability of HPLC to separate closely related antioxidants and the relatively large amount of time required for MS analysis.

For these reasons, researchers have looked at other detection methods, often focusing on methods that leverage the antioxidant activity of these compounds for purposes of separation and identification. Dr. Prior and his USDA colleagues Changjiang Guo, Guohua Cao, and Emin Sofic investigated the use of coulometric array detection after separation by HPLC.

Coulometric detection provides a complete voltammetric resolution of analytes as a function of their reaction potential, providing a separate capability that greatly enhances HPLC when dealing with antioxidants. Prior uses include the ESA CoulArray coulometric multi-electrode electrohemical detector in combination with HPLC. The CoulArray detector includes 16 independent electrochemical detectors set to different electronic potentials in series. A given compound can be either oxidized or reduced at a specific potential. This makes it possible to detect and quantify extremely small quantities of electrochemically active compounds in a complex matrix.

The CoulArray can simultaneously determine both retention characteristics and electrochemical potential of the compounds eluted from the HPLC column. This makes it possible to detect many peaks through differences in electrochemical potential that would otherwise be hidden because they elute at the same time. In conventional electrochemical cells, only a small proportion of the sample interacts with the electrode. CoulArray detectors are designed so that the entire sample flows through a porous graphite frit in such a way that the entire sample interacts with the electrode. This greatly improves the sensitivity of the measurement and also makes it possible to perform accurate quantitative measurements.

When using a coulometric detector with HPLC, analytes can be identified by their retention times as well as their hydrodynamic voltammograms (HDVs). The HDVs are determined by the peak height and peak area over three adjacent sensors, the lower, dominant and area sensors. Antioxidants can in most cases be positively identified by their response across these three channels. Software provided with the detector compares the voltammetric response of an unknown compound to an external standard and assigns two ratio accuracies. The two ratio accuracies compare the response of the unknown to the analyte on the lower to dominant channels and the upper to dominant channels. These ratios provide a numerical indicator of the authenticity and purity of the unknown.

Development of HPLC Procedure Using Coulometric Detection

The USDA researchers developed an HPLC procedure utilizing reverse phase chromatography coupled with a coulometric array detection system for characterizing overall antioxidant capacity and identifying and quantifying individual antioxidants such as in fruits and vegetables. The USDA researchers validated this method by analyzing standard solutions of various antioxidants as well as fruits and vegetables purchased from a local supermarket. They used a Dionex HPLC gradient pump along with the ESA coulometric detection system. Separation was performed on an octadecylsiloxane Hypersil column. Fruits and vegetables were weighed and homogenized in deionized water using a commercial blender. The identities of some of the peaks were further assessed by spiking samples of interest with the relevant standards and comparing the height or area ratios of the adjacent peaks to the dominant one between a standard and the actual sample. The compounds identified in the samples were quantified by fitting the peak area of the dominant channel to the standard calibration curve.

Figure 1 shows a 12 channel chromatogram of a 31-component standard including flavonoids and other antioxidants that shows both retention time and oxidation potential as measured by the coulometric detector. The reproducibility of the method was evaluated by repeatedly injecting the standard solution. The within-day retention time variation for each individual standard ranged from 0.09 to 1.38 percent coefficient of variation (CV). The within-day and between-day voltammetric response variability were tested by repeated injections of the standard mixture. The coefficient of variation of within-day responses ranged from 3.08 percent to 13.46 percent for eight replicates injected on a single day while the CV of between-day response ranged from 4.38 to 17.74 percent for a single injection on each of 10 days. However, the between-day voltammetric response for genistein, quercetin, and pelargonidin #2 were higher which may be due to their low solubility. The detection limit was 20 pg for all components except quercetin which was 1 ug. All compounds except for quercetin, 4-hydroxycoumarin, and gallocatechin gallate gave a linear response at concentrations from 1 to 2000 ng/mL. Hydroxycoumarin and gallocatechin gallate both gave a linear response at concentrations from 1 to 1000 ng/mL. Quercetin produced a linear response at concentrations from 50 to 1000 ng/mL.

The voltammetric response was also measured for 24 fruit and vegetable extracts as shown in Table 2. Each fruit and vegetable had a unique chromatographic and voltammetric peak distribution. The electrochemical data obtained with HPLC and coulometric detection correlated well with the automated oxygen radical absorbance capacity (ORACROO) assay with a peroxyl radical generator. The concentrations of some electroactive compounds were identified in the aqueous extracts of the fruits and vegetables as shown in Figure 2 and Figure 3. For example, catechin, rutin, naringin, vitamin C, and glutathione were detected in strawberries. Positive detection was accomplished by matching retention times and peak purity with the voltammetric response of the standard samples. Many other compounds were unidentified but well separated in fruit and vegetable extracts. The resulting chromatograms provide a fingerprint that characterizes the overall antioxidant status of these fruits and vegetables.

USDA researchers concluded that an HPLC coupled with coulometric array detection procedure provides a fast and accurate method for characterizing overall antioxidant status and identifying and quantifying specific antioxidant components in fruits and vegetables. This is important because of the large number of potential plant species as well as the large number of potential antioxidant components that need to be investigated. Of course, subsequent analysis with other tools such as MS will typically be required for positive verification but substantial time can be saved by limiting the use of the time-consuming MS method to compounds that have already been identified through their voltammetric potential. Researchers in other studies have also demonstrated the effectiveness of this method in similarly characterizing antioxidant status of urine, plasma and tissue samples.

Coulometric detection has the potential to provide a valuable tool for identifying specific polyphenols that may prevent disease and for pinpointing the types and varieties of plants that have a positive health impact.


  1. Acheson, R. M.; Williams, D. R. R. Does consumption of fruit and vegetables protect against stroke? Lancet 1983, 1, 1191-1193.
  2. Ames, B. M. Dietary carcinogens and anticarcinogens: Oxygen radicals and degenerative diseases. Science 1983, 221, 1256-1263.
  3. Hertog, M. G. L.; Feskens, E. J. M.; Hollman, P. C. H.; Katan, M. B.; Kromhout, D. Dietary antioxidant flavonoids and the risk of coronary heart disease: The Zutphen Elderly Study. Lancet 1993, 342, 1007-1011.
  4. Cao, G.; Sofic, E.; Prior, R. L. Antioxidant and prooxidant behavior of flavonoids: Structure-activity relationships. Free Radical Biol. Med. 1997, 22, 749-760.
  5. Gey, K. F. The antioxidant hypothesis of cardiovascular disease: Epidemiology and mechanisms. Biochem. Soc. Trans. 1990, 18, 1041-1045.

*All graphics in this article previously appeared in the Journal of Agricultural & Food Chemistry, 1997, 45, pgs. 1787-1796. Published by the American Chemical Society.

Jerry Fireman, president of Structured Information (Lexington, Mass.) can be reached 781-674-2300, ext. 100 or



Current Issue

Current Issue

February/March 2015

Site Search

Site Navigation