From bees or not from bees: that is the question!

Dr Chris Buck and the next generationToday’s post is a guest post, kindly provided by Dr Chris Buck, an LC-MS Field Support Specialist for Waters Australia. He’s recently had an article published in Chromatography Today, looking at food adulteration – specifically, at adulteration of honey products. Here, he presents an overview of some of the issues that article covered and how Progenesis QI was key in the monitoring of both honey and adulterant markers. Over to you, Chris!

After recent revelations of fake honey products being sold in Australia labelled as real honey, the Australian Competition & Consumer Commission (ACCC) Commissioner, Sarah Court, declared in a statement that products sold under the label of ‘honey’ must be produced entirely by honey bees. It had been shown, however, that an imported product made from corn and sugar cane (and no honey) was being labelled as a domestic pure honey product. This sort of dishonesty and mislabelling of products can take place to substitute alternative products of lesser value, cheating consumers and potentially can be disastrous for allergy sufferers. It’s also reminiscent of scares such as the 2008 Chinese milk scandal where diluted milk was spiked with melamine to fake higher protein content. The question arises: can a food safety lab tell the difference between a good product and a bad product, be it fake or adulterated in some way?

Applying chemometrics to the problem

In my recent article in Chromatography Today, I describe an experiment to assess the application of chemometrics to a test case of honey adulteration. Food safety is of great concern at the Australian National Measurement Institute, and my goal was to demonstrate a clear and unambiguous comparison of honey, likely substitutes, and mixtures of honey with those substitute products in a very rapid manner.

Quite often, the process of identifying molecules to use as markers can require a great deal of research, but literature may have few answers as to what should be present in a product and how much should be there. Finding such molecules through a literature search, developing a multiple reaction monitoring (MRM) mass spectrometer method and testing the method with standards and different products could take months; not only to source the high purity standards, but also to optimize the methods of extraction and analysis. And that’s not to mention that a devious product counterfeiter may spike in exactly those same high purity standards, if they are available, to fool product testing.

Rapid turnaround with Progenesis QI

My approach was to apply high resolution mass spectrometric fingerprinting of honey and the adulterants, using a Waters quadrupole time of flight instrument (Xevo G2 QTof) and the Progenesis QI software to process the data. A major goal was to show the groundwork for testing a food product could be set very rapidly.

The honey, wheat glucose syrup based product, and “golden” cane sugar syrup were purchased in the morning of day one. Later that day, the samples were prepared for analysis, with the preparation of mixtures of adulterants with honey in addition to the individual food products. Before injection, sample preparation was a simple dilution in water for each sample and filtration via 0.22 micron microcentrifuge filters. The sample set was then run overnight using triplicate injections of each sample onto C18 reverse phase chromatography with positive ion mode acquisition to target basic and hydrophobic compounds. This was followed by repeat injections of all samples onto an amide HILIC column coupled to negative ion mode acquisition to target polar compounds such as simple sugars and acidic compounds.

Data was collected from the instrument on Day 2 and, by lunchtime, data had been processed using Progenesis QI. Further data analysis was done over the next few days but effectively it could be seen the experiment had worked well, and compound markers of interest were already tabulated, in less than two days.

Identifying markers

Using a UPLC for chromatography coupled to the QTof, I was able to generate an Exact Mass Retention Time (EMRT) data set that could clearly differentiate the samples of honey, its potential adulterants, and honey adulterated with the other foods. This process is also sometimes called mass fingerprinting, as a very large number of components are observed in every injection, creating a very extensive and detailed picture of what makes up each sample. The size and complexity of the data set is where Progenesis QI is invaluable in allowing comparison of groups of injections and the application of multivariate statistics to find components that are unique or clearly elevated in the different foods.

PCA plot in EZ info

Potential markers – either highly specific to the honey or found at higher levels in the adulterants – were identified.  These markers could be further used to develop targeted analysis via other instruments more plentiful in food labs, such as MRM on tandem quad mass spectrometers like the Waters TQS.  Progenesis QI can, in combination with carefully designed experimental methods, make any adulterated product stand out from the crowd of good products based upon unusual chemical components, without any preconceived idea what the tell-tale components will be.

3D montage in Progenesis QI


Christopher Buck. Streamlining the Use of High Resolution Mass Spectrometry Data to Fingerprint Adulterated Honey using Multivariate Data Analysis to Facilitate Food Product Quality Control.  (2015)  Chromatography Today 8 (3), pages 48-52.