Of the many beneficial effects attributed honey, its antimicrobial properties remain perhaps the most important and widely studied  . The osmotic effect of the high sugar content of honey contributes to this antimicrobial effect, although experimental data indicate that other constituents of honey such as the mild acidity, presence of hydrogen peroxide and production of inhibins also play a role in the antibacterial activity of honey  . In recent years, there has been renewed interest in the use of honey for various therapeutic purposes, including treatment of infected wounds  -  . Current data now indicate that the beneficial effect of honey varies markedly depending on the floral source and this has stimulated the search for different types of honey with antibacterial activity  -  . Although various in-vitro and in-vivo studies have been carried out to assess the antimicrobial properties of honey from different parts of the world, there are only a few reports on honey originating from the Arabian Gulf region and the Indian subcontinent  -  . Thus this study was designed to study the antimicrobial activity of five types of natural honey available in the local markets of UAE against bacteria isolates including reference and pathogenic strains obtained from pus and wound swabs of patients.
2.1. Honey Samples
Five types of natural honey originating from three countries (UAE, Yemen and Pakistan) but locally available in the UAE market were randomly selected for the study. All honey samples were transferred into plastic sterile containers and given unique study labels. Table 1 shows the different types of honey, their provenance and assigned study labels. For the antibacterial assays, the honey samples were used undiluted and at 75% and 50% dilutions (wt/vol).
2.2. Bacterial Isolates
The bacterial isolates included in this study were three reference strains (S. aureus ATCC 29213, E. coli ATCC 25922 and P. aeruginosa ATCC 27853) and 21 clinical isolates obtained from pus and wound swabs from patients in the local hospital (Table 2). All bacterial strains were cultured and identified using standard laboratory methods. All MRSA isolates were confirmed by BD PHOENIX automated microbiology system (
Table 1. Provenance and unique study labels of the five different honey studied.
Table 2. Bacterial isolates and their clinical sources.
bacterial suspension on Muller Hinton agar plates, followed by colony counts confirmed that this was equivalent to 105 - 106 colony forming unit per ml (CFU/ml).
2.3. Assessment of Antibacterial Activity
The previously described agar well diffusion assay was used as a susceptibility screening test   . Briefly, 4 ml of bacterial suspension containing 106 CFU/ml was inoculated onto the surface of Muller Hinton agar plates and allowed to dry. Using a sterile cork-borer, 5 mm diameter wells were cut from the agar and 50 µl of the appropriate honey concentration was delivered into the well. The plates were incubated for 18 hours at 37˚C. Antibacterial activity of the honey was evaluated by measuring the zone of inhibition (ZI) against the test microorganism at the end of the incubation period. All experiments were carried out in triplicate.
3.1. Activity of Honey against Reference Bacteria Isolates
Honey E consistently gave zones of inhibition against all three reference strains at the three concentrations (Table 3) and was found to be the most effective one. The least
Table 3. Zone of inhibition of reference strains of bacteria when tested against concentrations of different honey types.
“-”: No zone of inhibition.
effective honey D showed the least antibacterial activity as it only inhibited bacterial growth of two isolates at 100% concentration with ZI of 10 - 11 mm. S. aureus ATCC 29213 was the bacterial isolate which showed the greatest susceptibility as its growth was inhibited by all concentrations of three out of the five honey types tested. E. coli ATCC 25922 showed the least susceptibility as it was completely unaffected by two of the five honey types tested (Table 3). In addition the ZI range of 10 - 15 mm was the lowest compared to 11 - 23 mm for S. aureus ATCC 29213 and 10 - 17 mm for P. aeruginosa ATCC 27453.
3.2. Activity of Honey against Clinical Isolates from Wound Swabs
Twenty one clinical isolates obtained from pus and wound swabs of patients in the UAE were tested. This comprised of nine S. aureus isolates (four of which were MRSA), six E. coli and six P. aeruginosa isolates (Table 2). All the honey types tested showed good activity against the S. aureus isolates. The least effective honey was honey D which only inhibited four of the five isolates at 100% concentration. At 100% concentration, honey E showed the best ZI (18 - 24 mm) compared to a range of 12 - 15 mm seen with other isolates (Tables 4-7). There was a trend of higher antibacterial effect of honey against the MRSA isolates as the range of ZI seen with these isolates were higher compared to those seen with the non-MRSA isolates (Table 4 and Table 5).
The antibacterial activity of all honey types against the E. coli isolates was poor. Four of the five honey types (A-D) failed to exhibit any antibacterial effect against one of the isolates (E. coli 1) and another isolate (E. coli 3) was only inhibited at 100% concentration. The highest ZI (15 mm) was seen at 100% honey E against one isolate (E. coli 2) (Table 6).
Honey E showed the best antibacterial activity against the Pseudomonas isolates as it inhibited three of the six isolates at all concentrations (Table 7). Four honey types (A-D) failed to inhibit P. aeruginosa 3 isolate at all concentrations. At all concentrations, the ZI seen with honey E was consistently higher than that seen with the other honey types. The results are comparable with the reference bacterial isolates.
Table 4. Zone of inhibition (mm) of S. aureus (5 isolates) against concentrations of different types of honey.
“-”: No Zone of inhibition.
Table 5. Zone of inhibition of MRSA (4) against concentrations of different types of honey.
“-”: No Zone of inhibition.
Table 6. Zone of inhibition of E. coli (6) isolates against concentrations of different types of honey.
“-”: No Zone of inhibition.
Table 7. Zone of inhibition of P. aeruginosa (6) isolates against concentrations of different types of honey.
“-”: No Zone of inhibition.
The variation in the antibacterial properties of honey has been linked with its provenance     . Indeed, the composition and therapeutic effects of honey tend to differ based on the floral source as well as geographical origin  . In this study, we have examined pure natural honey from two floral sources namely Zizyphys Lotus and Acacia tortilis obtained from different geographical locations within our region. The findings confirmed differences in the antibacterial efficacy of the different honey types with honey E (Acacia tortilis; UAE origin) showing the best activity against the three pathogens tested. Honey D which is also Acacia tortilis but of different geographical origin (
In terms of the antibacterial activity of the honeys on the different bacterial strains tested, S. aureus was the most inhibited bacterial strain while E. coli was the least susceptible. Both the reference strains and clinical isolates of these bacteria showed good susceptibility to all honey types tested. In recent years, treatment of MRSA infections has become a major challenge facing clinicians. As MRSA isolates are important etiological agents in skin and wound infections, it is therefore of clinical significance that all MRSA isolates showed increased susceptibility to the honey types tested. These findings are in keeping with other reported work and indicate that the use of honey may represent an effective and less expensive approach for local wound cleaning of S. aureus infected wounds   . In contrast, the overall trend was for lower susceptibility to the Gram-negative bacteria (E. coli & P. aeruginosa) tested. However although E. coli was resistant to most of the honey types tested, P. aeruginosa was relatively more susceptible with the exception of P. aeruginosa 3. P. aeruginosa is usually found in skin wounds particularly those related to burns and represents an important cause of nosocomial infection in burn patients. Hence the relatively good antibacterial effect of honey on these bacteria is of clinical significance and indicates the need for further work to identify other locally available honey with significant antibacterial activity.
There were potential limitations to this study. First, the number of clinical isolates tested against the honey samples was limited. Second, the antimicrobial test method and the choice of test organism(s), varies between publications and may be difficult to compare with other published results.
The findings from this study indicate that the use of locally available honey for the treatment of wound infections appears promising. However, differences in antibacterial activity exist among available honey depending on their provenance. Further chemical analysis is needed to identify the factors which determine the antimicrobial efficacy of different honey types.
Conflict of Interest
Nothing to declare.