Metabolomic profiling to detect different forms of beef fraud using rapid evaporative ionisation mass spectrometry (REIMS)

Abstract

Organic food fraud is a significant challenge in the food testing sector—high price premiums, ease of access to produce to be relabelled and difficulties in developing testing strategies that can detect such frauds make organic foods particularly attractive and thus highly vulnerable to fraud. Samples of conventional and organic cattle taken across meat plants in Ireland and the United Kingdom, consisting of the neck (supraspinatus), rump (gluteus), and shin (flexor carpi radialis) regions of the carcass were analysed using a high resolution time-of-flight based rapid evaporative ionisation mass spectrometry (REIMS) system. The resulting untargeted lipidomic data (m/z 600–1000) was used to generate PCA-LDA models for production system and for muscle type, for these models, it was found that the production system model could differentiate organic from conventional beef with an accuracy of 84%, whilst the muscle type model could identify the cut of meat with a 98% accuracy; additionally, samples can be assessed against multiple models simultaneously, reducing analysis time and sample numbers. The use of REIMS showed considerable promise in its ability to detect different forms of meat fraud; its accuracy in differentiating organic from conventional beef is superior to stable isotope ratio mass spectrometry, with the added advantages of substantially shorter analysis times and lower sample analysis costs. The ability to rapidly confirm the cut of meat also demonstrates the potential of REIMS to concurrently determine multiple aspects of beef authenticity in a close to real time analysis.

Rapid evaporative ionization mass spectrometry: A survey through 15 years of applications

“…This sampling device produces comparatively less aerosol since it requires less sample biomass for analysis and has a lower current density because of the electrodes' parallel geometry. Subsequently, this setup configuration was used also in clinic and food applications [6,16,17]. One of the main drawbacks of the bipolar probe consists in the lack of automation, different from the monopolar probe which was successfully automated through the use of a commercially available colony picker robot equipped with an autosampler for sample positioning on suitable plates.…”

“…Similar works were performed on meat, aiming at the assessment of quality attributes through the reliable identification of breed [28,75,76], production background (conventional vs organic; different feeding) [76][77][78], and cut type (neck, rump, shin, etc.) [78], as well as the sensitive detection of offals illegally added to more precious meat products [79] or other adulterants added in meat to falsify the nitrogen content or to improve the texture [80]. The detectability of the method was evaluated in some of those works by mixing authentic meat samples with different percentages of adulterants.…”

Food metabolomics: Latest hardware—developments for nontargeted food authenticity and food safety testing

“…The resulting aerosol from analyte ions formed due to a thermal ionization process can then be analyzed directly using an MS [81]. Some food analytical issues have been addressed using REIMS, mainly for the detection of food fraud in meat [82,83], for example, to distinguish organically from conventionally produced beef or poultry to be able to assign different muscle types [84,85]. Using lamb meat, it was recently possible to differentiate between breed and sex in addition to various forage diets [86].…”

Advances in ionisation techniques for mass spectrometry‐based omics research

“…developed an LA‐REIMS methodology in complement to a UHPLC‐HRMS method for metabolomic profiling of saliva samples, where LA‐REIMS provided a rapid analysis platform much more suitable to clinical labs, demonstrating the proof of concept where the metabolomic profiles of normal weight and obese individuals were successfully differentiated [119]. However, these applications have all the same potential limitations of poor quantitation and ion suppression due to the lack of prior sample preparation and therefore matrix effects [120], which on the other hand provides simplicity, universality and speed [121].…”

“…Van where the metabolomic profiles of normal weight and obese individuals were successfully differentiated [119]. However, these applications have all the same potential limitations of poor quantitation and ion suppression due to the lack of prior sample preparation and therefore matrix effects [120], which on the other hand provides simplicity, universality and speed [121].…”