Plants can employ three different chemical pathways during photosynthesis. Most plants, and most flowering plants, use an enzyme called ribulose bisphosphate carboxylase/oxygenase (RUBISCO) and are designated as C3 plants, which constitute about 90% of all plants, including sugar beet, wheat or rice. Alternatively, plants may also use the Hatch-Slack cycle; these plants are designated as C4 plants and typically originate from hot climates. Grasses like maize and sugar cane are C4 plants. A third group of plants, the CAM plants, has a unique metabolism called the ‘Crassulacean Acid Metabolism'. These plants generally use the C4 pathway, but can also use the C3 pathway. CAM plants are generally from very arid environments and they include pineapple, cacti and agave.

Crucially, these different pathways have all a specific preference for one of the two stable carbon isotopes (12C and 13C). Honey is made from nectar, meaning from blossoming plants, which are C3 plants. Sugar syrup made from maize or sugar cane have a different isotopic composition. Measuring the isotopic composition of honey therefore allows the detection of addition of sugar syrup from maize or sugar cane.

This method has been standardised as AOAC 998.12. This method can detect honey adulteration using C4 sugars at levels ≥ 7% but cannot detect adulteration of honey using syrups produced from C3 plants. The method was refined by adding a liquid chromatography step, hence looking at the isotopic composition of the individual sugars and so is able to detect also lower adulteration. However, sugar syrups matching the isotopic composition of honey are on the market and adulteration with these is not detectable by IRMS.

The JRC has developed and validated a method based on liquid-chromatography coupled to high-resolution mass spectrometry to quantify the following markers:

• Mannose: Mannose is a natural monosaccharide that can be obtained from both plants (e.g. jujube) and microorganisms. Mannose in blossom honey is considered as an indicator for the presence of starch-based sugar syrups or honeys treated with ion exchange resins. It is reported that certain honeys, e.g. linden, eucalyptus, jujube, or chestnut, could naturally contain mannose. Participants in the JRC technical round table agreed that a threshold of 0.02 - 0.04 g /100 g is appropriate for most blossom honeys. Higher concentration of mannose therefore indicate adulteration.   Alternatively, mannose can also be measured by nuclear magnetic resonance (NMR).
• 2-Acetylfuran-3-glucopyranoside (AFGP): AFGP is a characteristic compound found in rice syrups and so far not in honeys and therefore indicates rice syrup adulteration. AFGP is also known as the Special Marker for Rice Syrup (SMR).
• Difructose anhydride III (DFA III): DFAs are pseudodisaccharides produced by condensation of two fructose molecules by means of caramelization reaction which takes place during heating of sugars or sugar-rich foodstuffs; there are 14 DFA isomers. DFAs are easily generated with a small amount by thermal and/or acidic treatment of inulin-, sucrose-, or fructose-rich material through chemical catalysis. DFAs have been utilized as markers of adulteration with High Fructose Corn Syrup (HFCS) and Inverted Syrup (IS).
• Psicose: Psicose is a rare sugar found only in a few species in nature, such as Itea (honey-scented plants), wheat, cane molasses, and jujube. A study of various sugar syrups found psicose in 69 % of the syrups.
• Ectoine: The enzymatic preparation of glucose isomerase can contains ectoine to maintain its enzymatic activity. Such enzyme preparation can be used in the industrial production of syrups, which means that the molecule ectoine can serve as an indicator of adulteration by these syrups. By means of further measurements, the substance was also detected in high amounts in authentic South American bee feeds based on soybean and jatobá (Hymenaea spp., Fabaceae) flour. Small amounts were detected in a German feed dough whose ingredient list includes soy.
• Betaine: Betaine (also called trimethylglycine or glycine betaine) is a modified version of the simplest amino acid glycine. Betaine was identified as a marker for browned beet sugar (Beta vulgaris), which is produced mainly by adding molasses to refined (white) beet sugar
• 3-Methoxytyramine (3-MT): 3-MT is a natural pigment of the group of betaxanthins. Betalains are water-soluble pigments classified into betacyanins (red colour) and betaxanthins (yellow colour). 3-MT was previously reported as the Special Marker of Beet sugar (SMB) and its presence in honey therefore indicates adulteration with syrup from beet sugar.
• Oligo- & polysaccharides: Oligosaccharides and polysaccharides with a degree of polymerisation (DP) of six and higher in honey were used during the Coordinated Control Plan “From the Hives” as an indicator for adulteration with starch-based sugar syrups. Prescence of such oligo-/polysaccharides were assumed to be the result of incomplete enzymatic breakdown of starch and/or incomplete removal of them during syrup purification used in the adulteration of honey. However, this is only an indicator as there are indications that oligo and polysaccharides are also present in authentic honey.
• Proline: Proline is the dominant amino acid in honey and it is seen as an indicator of protein amount in honey since it constitutes 50–85% of the total amino acid content. The proline content is used as a criterion of honey ripeness and, in some cases, sugar adulteration. While there are no legislative limits, a minimum threshold of 0.018 g proline /100 g honey has been proposed below which dilution with exogenous sugars might be suspected

Nuclear magnetic resonance (NMR) is a method that can be used to quantify individual substances that indicate adulteration such as:

• Mannose: Mannose is a natural monosaccharide that can be obtained from both plants (e.g. jujube) and microorganisms. Mannose in blossom honey is considered as an indicator for the presence of starch-based sugar syrups or honeys treated with ion exchange resins. It is reported that certain honeys, e.g. linden, eucalyptus, jujube, or chestnut, could naturally contain mannose. Participants in the JRC technical round table agreed that a threshold of 0.02 - 0.04 g /100 g is appropriate for most blossom honeys. Higher concentration of mannose therefore indicate adulteration.   Mannose can also be measured by LC-HRMS
• Fructose, glucose: These are the main components of honey and deviations of their ratio from normal blossom can indicate adulteration.
• 5-Hydroxymethylfurfural (5-HMF): 5-hydroxymethylfurfural (HMF) is formed from reducing sugars in honey and various processed foods in acidic environments when they are heated through the Maillard reaction. It is typically absent in fresh honey but is an indication of excessive heating.  

Sugar syrup contains small residues of the genetic material of the plant it was produced from. This DNA can be quantified for the detection of adulteration of honey with these syrups. Methods have been described for the detection of rice and maize syrups at a level of 1 % (see, for example, here). 

Using the method in non-targeted mode can give information on adulteration with syrups form cane, corn, beet, wheat, rice, etc. If coupled with an extensive database of sugar syrups, also the kind of adulteration can be detected.

The composition of honeys also depends on the floral origin: honeys from different regions differ in their composition, so LC-HRMS can also confirm the floral origin of honey declared on the product's packaging.

Through analysing the fingerprint of NMR spectra it is possible to identify the unique pattern and markers that are characteristic of the normal composition of unadulterated honeys. Through an extensive database of authentic samples, deviations can be detected and therefore fraud. As with LC-HRMS, NMR can also confirm the floral origin of honey declared on the product's packaging. In addition, since honeys from different region also differ in their composition, NMR can also help to confirm the claims of geographical origin. 

Honey contains genetic material from the pollen of the plants, the bees and the microbes in the bees’ guts. Also this information can be used to generate fingerprints of authentic honeys to which unknown samples are compared. As is the case for NMR and LC-HRMS, this method crucially depends on an extensive database that covers ideally the complete variation of floral origin and geographic origin of commercialised honeys.

Presently, this method is offered by laboratories in Latvia and Austria, resulting in claims of very high adulteration rates. These numbers have been questioned on the basis that the database of honeys used for making these claims is comparatively small and that the method has not been validated. The German Federal Office for Consumer Protection and Food Safety therefore explicitly advised that these methods “should currently not be used as standalone methods, but in combination with standardised methods for the assessment of honey”. Read the full statement here.

We discussed challenges and the need for a full validation in a comment in the Journal Science of Food.