LC-MS/MS in the hunt for ‘illegal’ dyes in foodstuffs

Foods are complex matrices: interference problems have to be minimised during analysis, and analytical methods must be constantly refined to account for the growing numbers of dyes being monitored. LC-MS/MS is used to address these problems and to prepare for future events.
Mass spectrometry (MS) is a key weapon in the armoury of modern analytical chemistry. Used in conjunction with techniques such as HPLC (LC-MS), key features of the mass spectrum of LC peaks can be used in selective detection and determination of target compounds. LC-MS has become an indispensable tool for problem solving in many fields requiring ‘information rich’ chemical analysis. The use of LC-MS technology in analysis of illegal dyes in foodstuffs is a useful illustration of the technique’s versatility.
However, even with prior chromatographic separation, the complex mass spectra produced can be difficult to interpret, with multiple peaks potentially hindering the precise identification of analytes. Tandem mass spectrometry (or MS/MS), is an established technique for analysis of target compounds in complex mixtures with minimum sample clean up. In MS/MS, two mass spectrometric analysers are used in tandem. For analysis of illegal dyes, non-scanning modes referred to as multiple reaction monitoring (MRM) or selected reaction monitoring (SRM) are used. These modes yield maximum sensitivity and selectivity for known target analytes enabling low-level quantification of compounds in food matrices with concomitant reduction in sample preparation and analysis times.

Methodology

At CSL, a method based upon electrospray (ESP)
LC-MS/MS for the determination of Sudan dyes (I-IV) has been adapted and refined in order to meet customer requirements for low detection limits (10 ppb) in chilli products1. This is currently not possible using conventional HPLC with UV-VIS detection. Over the course of 2005, this method was developed further to encompass a wider range of sample matrices especially liquid and oil-based materials such as spice oleoresins. This has been facilitated by the use of gradient rather than isocratic HPLC separation and by the addition of isotopically-labelled internal standards d5-Sudan I and d6-Sudan (IV).
Methods to detect other illegal dyes such as Para Red and Orange II have been developed, and it is now possible to determine 15 different dyes with more under consideration. At these levels of detection, the possibility of false positive results remains an important issue, so additional clean-up steps have been introduced, based upon solvent partition and solid phase extraction (SPE) to remove isobaric interferences.

All of the dyes may be separated on the basis of chromatographic retention time and MS/MS product ion spectra. However, Sudan IV and Sudan Red B for example, elute very close together chromatographically but are distinguished readily by their MS/MS characteristics. Conversely, Sudan Black comprises two isomeric forms that exhibit similar MS/MS spectra but differ in their chromatographic retention time. While the analysis of standard dyes suggests a theoretical sensitivity commensurate with detection at the 10 ppb level in samples, significant alteration of the MS signal can be caused by the presence of sample matrix co-extractives. These effects are not only related to chromatographic characteristics but also to ionisation processes in the MS itself, giving rise to both enhancement and suppression of the MS signal. These effects are highly sample dependent and make the use of external matrix-matched calibration standards very difficult.
With the fact that most samples have been found to be negative, coupled to the need to keep costs in check without compromising on quality, a screening approach for illegal dyes analysis has been developed. To this end, standard addition at a predetermined decision level (typically the analytical reporting limit) is used. This technique accounts for any reduction in response caused by matrix effects through ‘exact’ matching of standards to matrices. Confirmation of dye identity and quantity is achieved through repeat analysis using a full standard addition series.
Method performance acceptance criteria are used to confirm the identity and quantity of dyes based upon:
•  Signal to noise assessment of chromatograms.
•  Correlation of calibration standards.
•  Peak retention time and ion ratios (based on the monitoring of at least two characteristic ions per compound).

The future

LC-MS/MS methods for the analysis of foodstuffs for illegal dyes should be developed to meet the expanding needs of consumers, regulators and industry. The specificity and sensitivity of the technique will enable increased problem solving capabilities as better extraction and clean up regimes are devised, and more powerful LC-MS/MS based instruments become available.
As the list of dyes of interest and tested foodstuffs continues to grow, a non-targeted approach may be viable. 

Michael Scotter, Simon Hird, Stewart Reynolds and Simon Baty are with the Food Science Group, Central Science Laboratory, Sand Hutton, York, UK. www.csl.gov.uk

Reference: 1. Calbiani, et al. (2004).Development and in-house validation of a liquid chromatography-electrospray-tandem mass spectrometry method for the simultaneous determination of Sudan I, Sudan II, Sudan III and Sudan IV in hot chilli products. J.Chromatog. A 1042 (1-2),123-130.

Recent Issues