Medicinal honey research is undergoing a substantial renaissance. From a folklore remedy largely dismissed by mainstream medicine as “alternative”, we now see increased interest by scientists, clinical practitioners and the general public in the therapeutic uses of honey. There are a number of drivers of this interest: first, the rise in antibiotic resistance by many bacterial pathogens has prompted interest in developing and using novel antibacterials; second, an increasing number of reliable studies and case reports have demonstrated that certain honeys are very effective wound treatments; third, therapeutic honey commands a premium price, and the honey industry is actively promoting studies that will allow it to capitalize on this; and finally, the very complex and rather unpredictable nature of honey provides an attractive challenge for laboratory scientists. In this paper we review manuka honey research, from observational studies on its antimicrobial effects through to current experimental and mechanistic work that aims to take honey into mainstream medicine. We outline current gaps and remaining controversies in our knowledge of how honey acts, and suggest new studies that could make honey a no longer “alternative” alternative.

Manuka honey has broad-spectrum antimicrobial activity, and unlike traditional antibiotics, resistance to its killing effects has not been reported. However, its mechanism of action remains unclear. Here, we investigated the mechanism of action of manuka honey and its key antibacterial components using a transcriptomic approach in a model organism, Pseudomonas aeruginosa. We show that no single component of honey can account for its total antimicrobial action, and that honey affects the expression of genes in the SOS response, oxidative damage, and quorum sensing. Manuka honey uniquely affects genes involved in the explosive cell lysis process and in maintaining the electron transport chain, causing protons to leak across membranes and collapsing the proton motive force, and it induces membrane depolarization and permeabilization in P. aeruginosa. These data indicate that the activity of manuka honey comes from multiple mechanisms of action that do not engender bacterial resistance. IMPORTANCE The threat of antimicrobial resistance to human health has prompted interest in complex, natural products with antimicrobial activity. Honey has been an effective topical wound treatment throughout history, predominantly due to its broad-spectrum antimicrobial activity. Unlike traditional antibiotics, honey-resistant bacteria have not been reported; however, honey remains underutilized in the clinic in part due to a lack of understanding of its mechanism of action. Here, we demonstrate that honey affects multiple processes in bacteria, and this is not explained by its major antibacterial components. Honey also uniquely affects bacterial membranes, and this can be exploited for combination therapy with antibiotics that are otherwise ineffective on their own. We argue that honey should be included as part of the current array of wound treatments due to its effective antibacterial activity that does not promote resistance in bacteria.

16Manuka honey has broad-spectrum antimicrobial activity and unlike traditional antibiotics, 17 resistance to its killing effects has not been reported. However, its mechanism of action 18 remains unclear. Here we investigated the mechanism of action of manuka honey and its key 19 antibacterial components using a transcriptomic approach in a model organism, Pseudomonas 20 aeruginosa. We show that no single component of honey can account for its total 21 antimicrobial action, and that honey affects the expression of genes in the SOS response, 22 current array of wound treatments due to its effective antibacterial activity that does not 40 promote resistance in bacteria. 41While MGO is a key antibacterial component of manuka honey, it alone cannot account for 65 its total antimicrobial activity 21-23 , as manuka honey inhibits the growth of pathogenic 66 bacteria (including Pseudomonas aeruginosa, Escherichia coli and Staphylococcus aureus) at 67 concentrations well below the minimum inhibitory concentration (MIC) of MGO alone 21-24 . 68 Additionally, many bacteria are innately equipped to detoxify MGO 25-27 , so additional 69 components in honey must also modulate its activity. From this, we hypothesise that the 70 antibacterial activity of manuka honey comes from a combination of its various constituents 71 and that its mechanism of action cannot be elucidated based exclusively on investigations of 72 the individual components. Rather, to generate a fundamental understanding of the 73 mechanism of antibacterial activity, the effects of the key components of manuka honey 74 against microorganisms must be studied in isolation and in combination with each other. 75Despite the prominent role of MGO in the antibacterial activity of manuka honey, to what 76 degree it contributes to the effect manuka honey has on bacterial gene expression and 77 physiology has not been thoroughly investigated [28][29][30][31][32][33][34] . Currently, the antimicrobial activity of 78 manuka honey is reported and marketed based on its NPA, which can be directly tested via 79 bioassays or derived from the MGO concentrations of manuka honey since MGO and NPA 80 are well correlated 18 . This is problematic since NPA is only a measure of anti-staphylococcal 81 activity and not representative of activity against other bacterial species 35 . Therefore, it is 82 important to understand how MGO alone and in combination with sugars works against 83Gram-negative microorganisms like P. aeruginosa, in order to better understand the 84 mechanism of whole manuka honey. This is critical for its use in infection control, which 85 requires killing of multiple species of bacteria present in wounds. 86Previous studies have identified a number of biological processes in bacteria that may be 87 affected by the action of honey, including cell division 22,29,32,33 , motility 28 , quorum sensing 88 (QS) 36-40 , protein synthesis 29,32,41 and responses to oxidative stress 7,41 . With the increased 89 128 MICs of all treatments were determined using the brot...