Food Integrity with New Analytical Technologies: Unlocking the Truth

Date: November 14 2018, 13:45-14:15
Speaker: Amanda Manolis
Thermo Fisher Scientific, Austin, USA

Fraud in food and beverage products include misrepresentation or tampering with packaging and labelling; adulteration, normally replacing a higher quality, original material with one of lesser quality one or extending a product by adding an adulterant; and misrepresentation of product origin. Increased complexity in the food and beverage supply chain has provided greater opportunity for economically motivated food and beverage fraud. Consequently, legislation has been enacted globally to protect food and beverage products with respect to production processes and product labelling.

The combination of legislation and food fraud practices demand a reliable, high throughput and cost effective analytical techniques that can identify food and beverage products that are not what they are claimed to be. Detecting food and beverage fraud can be achieved using next generation sequencing and stable isotope fingerprints because these technologies can differentiate between food and beverage samples which otherwise share identical chemical or similar genetic composition. We briefly explore how these technologies really detect food and beverage fraud based on the unique problem you are trying to solve.

NMR-based Tools for Food Authenticity and Quality Control

Date: November 14 2018, 13:45-14:15
Speaker: Andrea Steck
Bruker BioSpin GmbH, Ettlingen, Germany

Cases of food adulteration hit the headlines worldwide on a regular basis. Mislabelling of content or origin, substitution of high value ingredients by low-cost imitates, and additions of fillers or even potentially harmful components are typical fraudulent practices. By now, a majority of them is elaborate in order to circumvent conventional (targeted) analytical methods, and/or veiled within complex and global supply chains.

Analytical tools, suitable for effective prevention and enforcement of food fraud beyond single spot tests, optimally combine non-targeted fingerprint analysis and targeted quantification in one run, and are rapid, cost-effective and reliable likewise.

With its high-resolution 1H-NMR based food screening solutions, Bruker BioSpin GmbH provides unique solutions for authenticity and quality control developed in a joint effort with experts in the fruit juice, wine, and honey sector. The methodology uses comprehensive 1H-NMR spectra databases of authentic samples and multivariate statistics to generate models e.g. for geographical origin, botanical variety and different type of adulterations. Simultaneously, absolute quantification of a multitude of parameters (e.g. sugars, amino acids, fermentation markers) relevant for quality and authenticity are performed.

We invite you to learn more how this sophisticated technology is transformed into an automated push-button solution, and how it is applied in real life of food screening.

Testing of Flavour Inducing Volatiles and Off-Smells in Food and Beverages without Sample Pre-Treatment

Date: November 14 2018, 18:10-18:40
Speaker: Thomas Wortelmann
G.A.S. mbH, Dortmund, Germany

Respecting the dramatically increasing demand in food quality control same as customer demands in a straightforward workflow same as in rugged analytical equipment with a high throughput; this presentation highlights the unique and enhanced capability of combining gas chromatography with ion mobility spectrometry in static headspace as hyphenated analytical metrology.

The applied physical working principles of both technologies assure maximum reliability and the second dimension of separation realized through the TOF-IMS allows full orthogonality so that co-eluting compounds and even isomers can be separated. The sensitivity of the detector lies in the low ppb-range for (off-) flavour inducing volatiles facilitating a direct sampling. By this sensory panels can be supported by a machine that works 24/7 providing impartial flavour documentation. This presentation gives an introduction into the technological set-up and insight into successful exemplary applications in the field of food & beverages quality control, authenticity and integrity.

Targeted Proteomics to Tackle the Difficult Ones: Authentication of Closely Related Species and Semi-Untargeted MS Approaches

Date: November 15 2018, 08:15-08-45
Speaker: Jens Brockmeyer
Sciex, Villebon-sur-Yvette, France

Species authentication using DNA-based methods often still shows limitations when closely related species and mixtures thereof have to be analyzed. A relevant example is the differentiation of tuna species that, despite their close phylogenetic relationship, show significant differences in price and are therefore especially prone to fraud. To showcase the potential of food proteomics, we employed a bottom-up approach to identify specific marker peptides covering all eight relevant Thunnus and Neothunnus species as well as marker for Katsuwonus pelamis (bonito). In addition, several “group marker” covering different branches of the phylogenetic tree were identified to allow further confirmation.

Besides the identification of marker peptides specific for different species we detected peptides that are widely distributed in numerous different mammalian, bird or fish species (“global marker”). The relative quantitation of specific marker peptides and global markers allows for the identification of fraudulent blending e.g. in meat products without knowledge of the species used for adulteration and is in our opinion a very promising approach for future testing routines.

Metabolomic Profiling of Manuka Honey

Date: November 15 2018, 08:15-08:45
Speaker: Emmanuel Sauvard
Agilent Technologies, France

Consumers are increasingly concerned about the quality, authenticity and impact on their health of the food they consume. Manuka honey is very popular in the UK because of its many properties, such as its antimicrobial power, and is not immune to consumer concerns. The quantity consumed today is not consistent with the production, hence growing concerns about its authenticity. There is also no methodology to characterize its benefits and there is no evidence that its integrity is preserved between production and consumption.

The work carried out aims to perform multivariate statistical comparisons of data from reference Manuka honeys and from different production sites. A Principal Component Analysis -PCA- was selected to conduct the study. Metabolomic profiles are taken into account as a source of data. The goal is to highlight the differences and establish a relevant predictive model.
A first selection of honeys is made -Gales, Manuka 24 and Nelson 100- in consideration of the observed antimicrobial activity, after a test in 10 replicates. The advantage of this approach lies in the fact that sample preparation is extremely limited and therefore of minimal inherent bias.

A method of analysis is developed, considering the complexity of the matrix, both in the number of its constituents and in its variability. The choice was HILIC stationary phase chromatography separation and time-of-flight mass spectrometry detection for the non-targeted and quadrupole tandem detection for the targeted approach. The separation tests in reverse phase chromatography prove to be insufficient for the characterization of the polar fraction and fewer than 30 characteristics unique to a group are found under these conditions.
Similarly, it does not make it possible to highlight all known markers of honeys as reported in the literature.

The data acquired using High Performance Liquid Chromatography / Quadrupole Mass Spectrometry and Time of Flight (HPLC MS QTOF) are processed automatically and locally with the tools of the Mass Hunter platform. The « Pro-finder » algorithm allows to isolate 2300 unique entities to honeys. Chemometrics then allows the relevant extraction of information and leads to the realization of a model. 40 tests in positive and negative mode were done. Honeys are then differentiable via the « Mass Profiler Professional » software platform.

However, the study must be completed to determine the « true » honeys of the « false », to say those that have undergone alteration and / or modifications. Targeted analysis is used to search for known markers of honeys. Methyl-gyloxal, 5-Hydroxymethyl Furfural, Methyl-Syringate and Leptosperin are analyzed using High Performance Liquid Chromatography / Tandem Quadrupole Mass Spectrometry (HPLC / MS TQ). The data are added in the established model. Their treatment shows a correlation between the presence of the markers and some of the honeys tested. Honeys are differentiated, and their quality checked against markers.

The predictive character of the extended model – targeted and untargeted data – is evaluated on 9 sets of samples. The results show that:

  • The authenticity of honey is established with a risk of error of 2%;
  • The origin of the places of production is differentiated between New Zealand, Europe,  Argentina and Mexico;
  • The risk of error on the brand of honey is 32%.

This first study should be completed with a larger population of samples to test and validate the model and increase its accuracy by taking into account a larger number of replicates.

 

 

 

--> -->