Dr. M V Manoharan, Ms Mch Plastic Surgery, Preethi Hospitals, Madurai.

Revealing AWF’s Impact on MGO Strength and UMF Activity

The non-peroxide antibacterial activity (NPA) of manuka honey was identified as methylglyoxal (MGO) and was given an activity rating on a unique manuka factor (UMF) scale, which is the same as that for NPA (1,2) (Table 1). Note that the UMF organisation state that the UMF and NPA scales are not interchangeable (https://www.umf.org.nz/unique-manuka-factor/introduction/), but for our purposes we can consider them equal. The concentration of MGO is related to both NPA% and UMF (2,3). 

Table 1. Relationship between MGO, NPA and UMF (data from Adams et al. (3)).

MGO (mg/kg)	UMF (+)	NPA (%)	Activity (UMF)
83	5	5	Low
263	10	10	Marginal
10	15	15	High (Medical grade)
514	20	20	High (Medical grade)
10	25	25	High (Medical grade)

To compare MANUKApli with competitors products, we purchased tubes of Activon (100% manuka honey, made between 9.5% to 15% NPA according to Advancis website  (https://uk.advancismedical.com/products/activon-tube), Medihoney (80% manuka honey, but grade not specified, Comvita), Mesitran (40% unspecified medical grade honey, Netherlands) and have had a 3-in-1 assay (Analytica) and a ZOI study (CaiQTest) assay carried out on each product to compare against pli.

MANUKApli  (ManukaMed ) has concentrations of MGO >700 mg/kg giving it a UMF of nearly 19 (Table 2) and 1450 mg/kg 3PLA (Table 3). In contrast, Activon has an average concentration of MGO of 195.5 mg/kg giving it a UMF of 8 and, therefore, isn’t really a medical grade manuka honey. Similarly, Medihoney, which is 80% manuka honey, has very low concentrations of MGO along with a low UMF (3.4), making it a manuka bush-honey at best. The concentrations present in Tables 2 and 3 have been corrected to a 100% honey equivalent. Therefore, the grade of those honeys is very low.

Because Medihoney and Mesitran have additives, Analytica could not run an MPI test on them. 

Table 2: Concentrations of MGO, DHA and HMF from the 3-in-1 assay (Analytica) in MANUKApli, Activon, Medihoney and Mesitran. Concentrations have been normalised to 100% for Medihoney (80%) and Mesitran (40%). Note that concentrations of MGO, DHA were undetected in Mesitran (non-manuka honey), but I have inserted the lower detection limit of the assay which is typical for reporting (concentrations may actually be zero).

	n	Honey(%)	MGO	DHA	HMF	UMF
			>500 mg/kg	Age dependent	<40 mg/kg	Expect >10%
MANUKApli	3	100	767 ± 124	651 ± 331	92 ± 24	18.8 ± 1.8
Activon	2	100	182 ± 17	196 ± 12.5	83 ± 12	8.0 ± 0.5
Medihoney	1	80	36.25	50	3	34
Mesitran	1	40	8	40	23	1.2

Table 3: Concentrations and units of the five factors distinguishing manuka honey for MPI (Analytica) and for the zone of inhibition (ZOI) from CAIQTEST. Noted in rows two and three are the cutoff concentrations and Cq (PCR) for DNA in pollen. MANUKApli meets the requirements for a monofloral manuka honey while Activon only meets most the requirements for a multifloral manuka honey (DNA was not detected). TBD = to be determined.

	n	MPI	4-HPLA	2-MBA	2’MAP	3-PLA	DNA	ZOI
		Monofloral	>1mg/kg	>1mg/kg	>5mg/kg	>400mg/kg	<36 Cq	(mm)
MANUKApli		Multifloral	>1mg/kg	>1mg/kg	>1mg/kg	>400mg/kg	<36 Cq	(mm)
ManukaMed Pli	3		7 ± 2	8 ± 1	4 ± 2	1333 ± 286	30.2 ± 0.4	10.7 ± 0.8
Activon	2		5 ± 2	4 ± 1	2 ± 1	560 ± 250	0	TBD
Medihoney	1		–	–	–	–	–	TBD
Mesitran	1		–	–	–	–	–	TBD

Concentrations of MGO are shown graphically below (Figure 1A) above the zone of inhibition (ZOI) (Figure 1B). Note that total activity (TA) and non-peroxide activity (NPA=UMF) are similar for MANUKApli, but there is no NPA for Activon which is consistent with the low-grade (<10+ UMF) MH in that product. Also note that both Eurofins and CaiQTest carry out the ZOI as per Molan (4). However, Eurofins provides the result in mm (Table 3), while CaiQTest provides the result in %phenol equivalent using a phenol standard curve, which is more consistent with Molan’s method. Measuring in mm is necessary for the dressings, whereas measuring in NPA% can be done for fluids (honey).

Figure 1. Concentrations of MGO (A) (the UMF scale is indicated to the left of the Y-axis) and the zone of inhibitions (ZOI) (B) for pli (MM0051), Activon, Medihoney and Mesitran. ZOI-TA = total activity and ZOI-NPA is the non-peroxide activity after neutralising peroxide with catalase (n=1 for each product).

Surprisingly, there is similar activity for TA and NPA in Medihoney, which is only 80% MH and of a lower-grade (<5+ UMF). The Comvita website states that the additional 20% consists of waxes and oils that act as gelling agents (https://www.comvita.co.nz/medihoney/wound-gel-25g). The TA for the ZOI is 88% of Activon, which is consistent with the lower percent in the product, but there should be a very low, or no NPA considering that Activon is 100% MH and has a higher UMF. I suspect that the gelling agents form a film into which S. aureus can’t grow and, therefore, the NPA ZOI may be an artifact. 

Not surprisingly, the lack of activity in Mesitran is related to both the low honey content (40%) and lack of MGO (European honey varieties). Mesitran claim that their product is antibacterial (https://mesitran.com/products/ointment/), but only mention that this is related to the high osmotic activity. While it may have some antibacterial activity against some strains of bacteria, the test carried out by CaiQTest is the method developed by Peter Molan and uses S. aureus, which is probably more appropriate for testing wound products on skin. The lower quality of honey in the Mesitran product wouldn’t be expected to have much effect on S. aureus.

While we endeavour to have a wound dressing with a high concentration of MGO (>700 mg/kg), the exudate from a wound can dilute the honey and reduce the amount of MGO available for antimicrobial action. Furthermore, the protein content in the exudate may sequester free MGO, thereby further reducing available MGO for antimicrobial activity. Therefore, we carried out an in vitro study to determine how much MGO is lost with dilution from exudate and, in addition, what effect the protein content has on activity (UMF).

Wound exudate contains 2.6 to 5.1% protein, with a mean of 3.9%, which is about half that of plasma (5,6). The volume of wound exudate varies with size and type of wound. Using a pressure ulcer as an example, a wound size of 16 cm2 (4×4 cm) generates between 2-6 mL/d, on a low to moderate discharge rate, or between 4-14 mL/d for a 25 cm2 pressure ulcer (5 x 5 cm) (7). Volumes of exudate can be higher in larger ulcers, or those with heavy discharge rates. 

A MANUKAhd dressing of 10 x 12.5 cm has 0.16g of MH per cm2, which equates to 20 g of MH per dressing. If we consider that 6 mL is a mean volume of exudate, then the amount of MH in a MANUKAhd dressing is diluted by 6/20 = 30%. Therefore, we carried out a study with the same batch of pli that was diluted 0, 10, 20, 30, 40 and 50% to cover a range of exudate volumes. 

To consider the additional effect of protein on sequestering MGO, we made an artificial wound fluid (AWF) without 3.9% BSA (AWF vehicle (-BSA) and with BSA (AWF + BSA).

As expected, dilution reduced concentrations of DHA and MGO and reduced the UMF activity by about 50% (Figure 1). In addition, the 3.9% protein in the AWF reduced concentrations of MGO and UMF by a further 8%. In contrast, concentrations of DHA were only reduced by dilution, which suggest that DHA is not being sequestered by protein. 

At a dilution of 50%, concentrations of MGO remained sufficiently high to maintain a UMF of about 13%, which would continue to have an antimicrobial effect. The same cannot be said for Activon, Medihoney and Mesitran, the first two of which had concentrations of MGO giving a UMF of 8% and <5%. When those percents are applied to the same conditions of dilution with exudate as we used for pli, then the activity would be reduced by about 50% and leave those products with very limited antimicrobial activity (hatched area at the bottom of Figure 2C). 

Mesitran had no MGO, no UMF and no ZOI against S. aureus. Therefore, dilution with exudate would render that product ineffective (it was only 40% honey from European sources to begin with).

Figure 2. Concentrations of DHA (A), MGO (B) and UMF grade (C) for MANUKApli that has been diluted from 100% to 50% in artificial wound fluid (AWF) without 3.9% BSA (AWF vehicle (-BSA)), or with BSA (AWF + BSA). Medical grade manuka honey should have a UMF above 12+. A 50% dilution of MANUKApli in AWF + BSA maintains a UMF above 12+ UMF, while competitors products (hatched area) are unlikely to have much UMF activity even before dilution.

Conclusion

A manuka honey wound dressing should have concentrations of MGO >700 mg/kg in order to maintain an active UMF during the course of wound healing.

References

1. Mavric E, Wittmann S, Barth G and Henle T. Identification and quantification of methylglyoxal as the dominant antibacterial constituent of Manuka (Leptospermum scoparium) honeys from New Zealand. Mol Nutr Food Res 52: 483-489, 2008.
2. Adams CJ, Manley-Harris M and Molan PC. The origin of methylglyoxal in New Zealand manuka (Leptospermum scoparium) honey. Carbohydr Res 344: 1050-1053, 2009.
3. Adams CJ, Boult CH, Deadman BJ, Farr JM, Grainger MN, Manley-Harris M and Snow MJ. Isolation by HPLC and characterisation of the bioactive fraction of New Zealand manuka (Leptospermum scoparium) honey. Carbohydr Res 343: 651-659, 2008.
4. Allen KL, Molan PC and Reid GM. A survey of the antibacterial activity of some New Zealand honeys. J Pharm Pharmacol 43: 817-822, 1991.
5. Lutz JB, Zehrer CL, Solfest SE and Walters SA. A new in vivo test method to compare wound dressing fluid handling characteristics and wear time. Ostomy Wound Manage 57: 28-36, 2011.
6. Trengove NJ, Langton SR and Stacey MC. Biochemical analysis of wound fluid from nonhealing and healing chronic leg ulcers. Wound Repair Regen 4: 234-239, 1996.
7. Iizaka S, Sanada H, Nakagami G, Koyanagi H, Konya C and Sugama J. Quantitative estimation of exudate volume for full-thickness pressure ulcers: the ESTimation method. J Wound Care 20: 453-454, 458-463, 2011.