New Analysis Tools for Detecting Food Fraud

NEW ANALYSIS TOOLS FOR DETECTING FOOD FRAUD 1 Listicle New Analysis Tools for Detecting Food Fraud Article Published: September 8, 2023 | Joanna Owens, PhD Credit: iStock Food and beverages are characterized throughout the production chain to ensure safety, authenticity and quality. A wide range of analytical techniques are used for this, depending on the parameter being measured. But the increasing threat of food fraud is driving the development of new and more powerful combined approaches that can “fingerprint” food constituents to stay ahead of the game. The scale of food fraud Food fraud is not new either. “Even dating back to Roman times, there were reports of fraudulent wine that had lead added to it to make it taste sweeter, and reports of spices being adulterated,” said Hellberg. “Historically, the main foods vulnerable to fraud were those that were widely traded. But any time you have foods changing hands multiple times and traveling long distances from their source, there’s an increased chance of fraud.” Today, fraud is commonly associated with dairy, seafood, meat, spices, olive oil, honey, wine and other alcoholic beverages.1 Also, the use of certifications, such as ‘organic’, can increase the risk of fraud, because these claims add value. Historically, most tests for food fraud were targeted techniques, which specifically test for one constituent, such as melamine. “Now there is a shift towards using a combination of targeted techniques and untargeted methods looking at the overall profile of the food and comparing that to a reference standard,” said Hellberg. “If something looks different, then you can move on and conduct targeted tests to see why. Those are difficult because there's always going to be some natural variation in foods. So untargeted testing can be more expensive and more datarich, but it’s very informative when there might be some unknown adulterants we're not yet aware of the need to test for.” How do you test for authenticity? So how do you test the authenticity of complex natural products such as milk, honey and olive oil? Professor Stephan Schwarzinger at Bayreuth University in Bavaria specializes in developing new methods for authenticating food. “The definition of authenticity depends on the type of food and the claims stated on it,” he explained. “A carton of milk, for example, may include claims about the product’s provenance, quality and health benefits – from its organic status to the type of diet the cows grazed, to its percentage fat content. If any one of those claims is not matched by the contents, then this is an issue with authenticity. For each of the facets that play a role in authenticity, there are different reference analysis methods.” Testing for some claims on the packaging of products is relatively straightforward, such as analyzing fat content, but Schwarzinger explains that a major challenge in food and beverage analysis is testing for new properties, such as hay-fed milk, or for food fraud where the method of adulteration is unknown. Article NEW ANALYSIS TOOLS FOR DETECTING FOOD FRAUD 2 Article “This is where advanced technologies that enable us to take one measurement but get a glimpse of many different parameters are useful,” Schwarzinger said. “For example, nuclear magnetic resonance (NMR) spectroscopy can identify 100-200 compounds within a sample and mass spectrometry (MS) coupled to liquid or gas chromatography (LC-MS or GC-MS) can provide a readout on hundreds or thousands of compounds.” Each technology has its strengths and weaknesses when it comes to food and beverage analysis. The best analytical approach depends on the authenticity issue you are looking at and what type of food you need to study. To use LC-MS, for example, you need to do an extraction and decide whether to analyze the lipophilic or hydrophilic fraction. NMR, while not as sensitive, offers an advantage here; foods such as juice, wine, edible oils (including olive oil) and honey (in its dissolved state) can be subjected to analysis directly and as a whole – without the need for fractionation. “One of the merits of NMR is its broad dynamic range, so it can measure major constituents in foods such as the sugars in honey, (which sometimes exceed 300-400 g/kg for one kind of sugar) right down to a detection level of 2-3 mg/kg,” said Schwarzinger, “But if you want to assess compounds at µg/kg levels, such as many tasterelevant constituents, then that would be the domain of MS and its more complex sample preparation.” In Hellberg’s lab, research focuses on developing DNA-based methods to detect the mislabeling of foods, predominantly seafood, but also encompassing meats, pet food and dietary supplements. “The most common approach we use is DNA barcoding, using universal primers and amplifying a standardized genetic fragment in the test food and then comparing this to a reference database to identify the species,” Hellberg explained. “This works very well with raw, lightly processed products, but it can be challenging to use DNA barcoding in heavily processed products. So we develop methods to detect shorter genetic fragments or use other types of DNA-based methods like real-time PCR to enable identification even in these heavily processed foods.” Emerging tests and tools Hellberg’s work is informed by market surveys that highlight difficult to identify products. When they found a lot of difficulties with identifying species within dietary supplements, such as ginseng, they started to explore this area further. “Developing ways to identify species in supplements is a big area of work, but when you're using DNA to detect something, there's a lot of opportunity to interpret the results incorrectly,” said Hellberg. “For example, let's say you're testing for a certain species that's claimed to be present in the product and your result comes back negative, there's a temptation to say the product was falsely labeled because it wasn't identified as ginseng, but it could just be the test isn't able to identify ginseng in such a degraded processed product.” A similar issue can occur with small amounts of contaminants, which could be unintentional rather than an act of adulteration. “It’s important to work on how we interpret results in a heavily degraded product, as well as improving the technological aspects of the tests,” said Hellberg. “This will enable us to know whether the contaminant affects a very small portion of the product or if it was intentionally added and is being used as a filler.” Another technique in the food analysis toolkit is infrared spectroscopy. “While infrared spectroscopy may not have the sensitivity of NMR or MS, you can make handheld spectrometers that are the size of a credit card and this raises the possibility of carrying out quick analytical checks at source, for example during or close to harvest,” said Schwarzinger. “I can foresee that we have those kinds of analytical devices in the future of harvesting machines. Despite lower sensitivity and resolution this technology will contribute to an even narrower network of traceability data.” NEW ANALYSIS TOOLS FOR DETECTING FOOD FRAUD 3 Article In Professor Schwarzinger’s lab, they are developing “analytical fingerprints” of the constituents of foods most affected by food fraud on a global scale, such as honey. Traditionally, analytical chemists would optimize a method to look for quality parameters such as the water content and sugar content in honey, but increasingly consumers and food regulators want assurances about the variety and provenance of honey – for example, whether it comes from a specific type of blossom, as claimed. “A unique feature of NMR is its reproducibility compared to other methods, such as MS, where there is variation between instruments and due to sample preparation. With NMR we can make a fingerprint of the constituents of a sample in a very reproducible manner – even on different instruments – and establish reference databases of authentic foods,” Schwarzinger said. “It is also possible to deconvolute the NMR spectrum into concentrations of single constituents, such as specific amino acids, and quantify these compounds, providing robust single markers you can use to compare different samples. If there’s a deviation in the spectral profile or you don’t find the markers characteristic of a specific variety or origin, then that would raise an alert flag.” Schwarzinger’s group is also working on combining different methods into an analytical ecosystem as an emerging technology. “The resulting complex analytical fingerprints might help when no single method can capture an adulteration,” he said. The future outlook on food fraud You might expect that with new technologies and as more reference databases for honey and other foods and drinks are developed, instances of food fraud would decline. Sadly, this is not the case. As the technology in the analysis laboratory improves, so does the technology used by food fraudsters. “Obviously when you have a new technology, you capture a lot of fraud initially, but this is recognized by those who make the alterations to the product, and they bring less of the product to market, or change their approach,” said Schwarzinger. “It’s important to realize adulteration of food is a moving target.” “It's really important that people don't just rely on testing,” added Hellberg. “You need to have a comprehensive food fraud mitigation plan where there's also auditing, anti-counterfeiting technologies that can be incorporated, and documentation throughout the supply chain.” “My hope is that, in the end, our new analysis technologies will raise the bar to make it economically unviable for food fraud to continue,” said Schwarzinger. “If you need to spend more money than you gain, there will be no point in adulterating food.” Reference 1. Everstine K, Hellberg RS, Sklare SA. Introduction to food fraud. Elsevier; 2021:1-7. doi: 10.1016/B978-0-12-817242- 1.00010-5