Free Radical Production and Quenching in Honeys with Wound Healing Potential

Objectives: Honey-impregnated wound dressings are now available on drug tariff in the UK, though the modes of action of honeys with antibacterial and wound healing properties are not entirely clear. The action of some but not all of these honeys is linked to the production of hydrogen peroxide on dilution of the honey with wound exudate. The present study investigates both free radical production and the antioxidant potential of some honeys, properties which may have a role to play in wound healing.

Methods: Free radical production and quenching of three honey types (Manuka, antibacterial but non-peroxide-producing; pasture, antibacterial peroxide-producing; commercial heat processed, non-antibacterial) was investigated by electron paramagnetic resonance (EPR) spectroscopy; quenching was also examined using a superoxide quenching assay.

Results: All honeys tested had antioxidant potential, with Manuka able to completely quench added radicals within 5 min of spiking. Only the peroxide-producing honey (pasture PS9) was found to form radicals on dilution.

Conclusions: The ability to modulate production and quenching of free radicals may contribute to the demonstrated ability of some honeys to help in resolving the state of inflammation typifying chronic wounds.

Introduction

Traditionally, carefully selected honey has been used for the treatment and prophylaxis of many disorders. Honey is enjoying a renaissance in wound care because it offers broad-spectrum antimicrobial properties and it promotes rapid wound healing, though the mechanisms by which these effects are achieved have not been fully elucidated.

Honey is a complex product; it has low water activity and acidity, and high sugar content but this does not contribute significantly to the antibacterial activity of those honeys recommended for clinical use. Some of these honeys become more potent upon dilution due to production of hydrogen peroxide generated by the action of glucose oxidase deposited in honey by bees. Addition of catalase to diluted honeys of this type destroys the activity of hydrogen peroxide and reduces or eliminates antibacterial activity. A survey of New Zealand honeys demonstrated that antimicrobial activity of a few honeys (mostly derived from Leptospermum scoparium, or manuka shrub) was retained on dilution in the presence of catalase. These honeys therefore rely on components other than hydrogen peroxide for their potency on dilution, and these have been postulated to be phytochemicals (though it has been suggested that peroxide still has a role, even in these ‘non-peroxide’ honeys).

The production of peroxide could lead to the formation of free radicals such as hydroxyl and superoxide. While these may play a part in antimicrobial activity (they are produced by human immune cells), they are potentially damaging to tissues and are also produced by bacteria to enable easier access to human cells.

In addition to antibacterial activity, honeys are known to have antioxidant capacity, which may act to modulate production of free radicals. The object of the present study was to investigate free radical production of peroxide and non-peroxide honeys using electron paramagnetic resonance (EPR) spectroscopy and to assess their antioxidant potential using EPR and a superoxide quenching assay.

Materials and methods

Three samples of natural honey were used. A clear, uncrystallized honey (Gale's) obtained from a supermarket was included to represent a heat-processed honey (heat processing destroys antimicrobial activity), an antibacterial peroxide-generating honey from mixed floral source (pasture PS9) and monofloral non-peroxide-generating antibacterial manuka M109 honey (not heat processed). An artificial honey that reflected the main components of honey was prepared by dissolving 1.5 g of sucrose, 7.5 g of maltose, 40.5 g of fructose and 33.5 g of glucose in 17 mL of deionized water. This solution was included in the study to evaluate the contribution of the predominant sugars and water to the assayed activities.

EPR spectroscopy was employed in two series of experiments. First the detection of hydroxyl radical (OH•) formation from hydrogen peroxide that may be produced in the diluted honey samples, second to test for any antioxidant activity in the honey by quenching free radicals deliberately introduced into diluted honey samples. To detect highly reactive radicals, such as OH•, the spin trap 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) (Sigma) was used. This compound reacts with OH• to form a stable paramagnetic species (the ‘spin-adduct’ DMPO-OH•) that gives a characteristic 1:2:2:1 quartet of lines in the EPR spectrum.

To determine the potential for OH• production by the different honey solutions, Fe2+ was added to the honey to generate OH• by Fenton chemistry from any peroxide present. As a positive control for trapping this radical, Fe2+ was used to generate the hydroxyl radical from H2O2.

To determine OH• production 40 µL of 0.1 M DMPO was added to 100 µL of a fresh 50% (w/v) solution of honey in an Eppendorf tube and mixed gently by pipetting. To initiate the reaction, 20 µL of 0.05 M FeSO4 was added and the solution immediately pipetted into a sample tube and placed in an EPR spectrometer (Varian E104 operating at 9.2 GHz microwave frequency and 20 mW power) and EPR spectra recorded. Typical spectrometer conditions were 100 Gauss (G) scan range, 120 s scan time and 1.0 G field modulation. Experiments were repeated in the presence of 1 mg/mL catalase.

In experiments to detect antioxidant activity (radical quenching) by EPR, the honey samples were also ‘spiked' with 30 µL of 30% (v/v) H2O2 prior to addition of the FeSO4 and the decay time of the resultant EPR spectrum was used to determine the OH• radical ‘quenching' or antioxidant potential of each honey.

To further investigate antioxidant activity a superoxide quenching assay was used; superoxide radicals were generated by the xanthine/xanthine oxidase (X/XO) system following a method described previously. Each assay contained 290 µL of 5 mM xanthine (sodium salt), 100 µL of the test solution (positive control, 50% honey solution or negative control) and 100 µL of 1 mM lucigenin. Positive control was superoxide dismutase (Sigma) 1000 U/mL, and negative control was sterile deionized water. After mixing, 150 µL of this mixture was transferred to wells in a 96-well microtitre plate using at least five replicates per sample. To each well 10 µL of xanthine oxidase (0.5 U/mL) was added and the plate was incubated for 15 min in the dark. The plate was read in a luminomenter (MLX Microtiter Plate Luminometer; Dynex Technologies) and results expressed in relative light units (RLU).

Results

A total of 58 EPR spectra were examined. The EPR spectra for positive and negative controls and the spectra for pasture honey with and without the addition of catalase. The only honey observed to generate radicals was the pasture honey.

The EPR spectra for Manuka honey before adding Fe2+, H2O2 and DMPO (a), immediately after addition (b) and 5 min after addition (c), demonstrating complete quenching of the OH• spectrum. Speed of signal quenching varied between honeys. Gale's also produced complete quenching 5 min after the additions; artificial honey produced only partial quenching after 5 min. The pasture honey was the slowest quencher as it actively produces radicals on dilution, but after 1 h the EPR spectrum was reduced to a similar level to that generated by the addition of catalase in the experiments to determine OH• production.

Variation in quenching activity was detected, with Manuka honey similar to the positive control, followed by pasture, Gale's, artificial and negative control.

Discussion

The only honey found to generate radicals in the present study was pasture PS9. In those honeys that do produce peroxide, bactericidal activity could be due to hydroxyl and possibly other radical products. In support of this statement catalase treatment reduces antibacterial activity, also in this study spectra of the hydroxyl radical and other radical species were observed on activation of pasture honey with ferrous ion. The additional radicals could not be unequivocally identified; their spectra suggest that they may arise from secondary reactions between hydroxyl radicals and organic compounds present in the honey. These secondary organic radicals may also have antibacterial properties. Any antibacterial properties of a peroxide-generating honey will thus be a balance between the opposing activities of radical generation and quenching.

It has been shown for peroxide honeys that the optimal dilution at which the honey will produce maximal amounts of hydrogen peroxide is between 40% and 60%. The fact that no radical production was seen from Manuka honey on dilution to 50% confirms this honey as non-peroxide, relying on other mechanisms for its antibacterial and wound healing properties, contrary to previous suggestion. These mechanisms are as yet unknown, but are possibly derived from the range of phytochemicals present in Manuka honey.

All the honeys tested by EPR and superoxide quenching assay had some antioxidant capability when compared with controls. The Manuka honey, in particular, showed complete quenching of the EPR signal generated by peroxide within 5 min. Even the pasture honey, which actively generates radicals, showed quenching after 1 h. Manuka honey has been shown to promote synthesis of cytokines by monocytes that have the potential to mediate the immune response. In responding to challenges such as infected wounds, neutrophils and macrophages utilize the presence of free radicals such as superoxide and hydroxyl to modulate the activity of other cells such as monocytes and platelets. These in turn produce specific cytokines to signal activation of other cells. However, in a prolonged insult, such as a chronically infected wound, such stimuli may give rise to an excessive response and the ability to dampen free radicals may therefore contribute to the complex interaction that helps to resolve the state of chronic inflammation typifying these wounds.

When Pseudomonas aeruginosa invades human tissue it uses free radical formation as a means to enhance invasion, and this can lead to cellular damage. Manuka honey is bactericidal to P. aeruginosa in vitro and has been suggested as a topical treatment for wounds infected with this bacterium. The present study shows that the quenching properties of Manuka honey in particular, which have been attributed to methyl syringate, could minimize the invasive effects of such an infection as well as help to resolve chronic inflammation.