The conversion of DHA to MG/NPA

Posted on July 1, 2016

Megan Grainger, PhD, Honey Team Leader and Technologist, Analytica Laboratories.

Manuka honey, derived from the Leptospermum scoparium (manuka) tree, is unique due to its non-peroxide antibacterial activity. This arises from the presence of methylglyoxal (MG) in the honey. This activity can be expressed as Non-Peroxide Activity (NPA). A conversion equation based on published data (1) is used to calculate the NPA directly from MG; hence MG and NPA are directly related. Honey with high MG/NPA can be sold for a higher price than lower MG/NPA honey. In order to achieve high NPA honey, it is beneficial to understand the steps that can be taken to maximise the MG.

MG is formed from a conversion of a compound called dihydroxyacetone (DHA) in the honey. The DHA is initially present in the nectar of manuka flowers and is taken back to the hive in the nectar. Once the honey is harvested, the chemical conversion of DHA to MG begins. The conversion of DHA to MG can be accelerated by increasing the temperature. At higher temperatures there is a larger loss of DHA (and MG) to side products. Hence heating the honey does not necessarily gain more MG in the long term (2). Hence there is considerable interest from both land owners and beekeepers on finding the highest DHA producing manuka trees so that the honey resulting from manuka sites will have high DHA in the fresh honey and ultimately high MG and NPA once it has been stored.

Nectar testing for DHA content is increasingly being used as a way to assess existing manuka sites or to find manuka trees suitable for planting. Nectar is taken directly from the flowers of the tree and sent to the laboratory for testing. DHA and sugar (fructose and glucose) are analysed. The laboratory report converts the DHA present in the one sample to a normalised DHA result; which is the approximate concentration of DHA that would be found in honey if it was solely derived from this one nectar sample. This is based on honey containing 80 % sugar.

A paper published in 2009 (3) calculated the concentration of DHA that would be present in the honey if it was solely derived from the one nectar sample. The authors reported that the DHA in the nectar was more than double what is observed in typical freshly produced honeys. This has also been observed in work carried out by Analytica Laboratories.

The lower concentration of DHA in honey compared to the nectar is attributed to the fact that the DHA to MG conversion process has already begun and also to a dilution effect; ie. a hive contains many bees which collect nectar from both manuka and non-manuka flowers. In addition, it is well known that the NPA of a batch of honey can vary both between and within regions as well as over years. This is due to not all manuka trees producing the same amount of DHA. Different manuka plants produce different levels of DHA. Furthermore, it is possible that environmental factors (such as temperature and soil conditions) also affect the expression of DHA in the nectar over time.

It is important to collect multiple nectar samples across a manuka site to get an idea about how the site will perform. This allows an average for the entire site to be calculated which will take into account the variation of DHA expression between trees.

Results for manuka sites should be compared relative, to each other – a site with higher DHA overall should produce honey with a higher average DHA level.

Other factors to take into consideration when choosing plants are their floral density and the amount of nectar they produce. If a tree has a high normalised DHA but very low floral density and nectar production, there will not be a large contribution from this plant into the honey, hence the MG/NPA of honey from this site will not be as high.

In summary, sampling nectar from a representative proportion of trees on a manuka site will help to give a good assessment of the average normalised DHA level from that site. Although the DHA is not a 1:1 conversion from nectar to honey it gives a good understanding on the potential of the site to produce honey with NPA activity.

Figure 1 shows the frequency of DHA concentrations of 1,309 manuka nectar samples from New Zealand; the average is 3,887 mg/800 g sugar (equivalent to 1 kg of honey). The highest reported nectar in this set of data is 27,070 mg/800 g sugar, but there are reports of Manuka trees with higher DHA.

Figure 2 shows the frequency of DHA concentrations found in a selection of honeys (manuka and blends) collected from various New Zealand sites which contained less than 4 mg/kg HMF (to indicate fresh honey) and over 100 mg/kg DHA (to indicate presence of manuka). In comparison to the nectar samples (3,887 mg/kg honey equivalent), the honey samples have an average DHA of 981 mg/kg. This is approximately four times lower than the reported values in the nectar.


Figure 1 Frequency plot of Normalised DHA (mg/800 g sugar) in manuka nectar samples. The average is 3,887 mg/800 g sugar.


Figure 2 Frequency plot of DHA (mg/kg) in 3,661 honey samples containing >100 mg/kg DHA and <4 mg/kg HMF. The average is 981 mg/kg DHA. The dotted vertical like depicts the average normalised DHA in nectar.


(1) Isolation by HPLC and characterisation of the bioactive fraction of New Zealand manuka (Leptospermum scoparium) honey. C. J. Adams, et al. Carbohydrate Research 343 (2008) 651-659. And, Corrigendum to ‘’Isolation by HPLC and characterization of the bioactive fraction of New Zealand manuka (Leptospermum scoparium) honey” Carbohydrate Research 343 (2008) 651. Carbohydrate Research 344 (2009) 2609. C. J. Adams, et al.

(2) Kinetics of conversion of dihydroxyacetone to methylglyoxal in New Zealand manuka honey: Part I – Honey systems. Grainger, M. N. C et al. Food Chemistry 2016

(3) The origin of methylglyoxal in New Zealand manuka (Leptospermum scoparium) honey Adams et al. Carbohydrate Research 244 (2009)