Screening for Melamine Adulteration in Protein-Based Foods by GC/MS
Melamine is a nitrogen-rich industrial chemical with many different uses; one of the most common is in the manufacturing of melamine resin, a very durable polymer. In spring of 2007, melamine became well known in North America after its presence in wheat gluten was linked with renal failure in dogs and cats.[1] Further studies led scientists to believe that a combination of melamine and its related homologues—ammeline, ammelide and cyanuric acid—caused the illness of these pets.[2] Investigation into ingredient material sources of pet food resulted in the determination that melamine was intentionally added to wheat gluten to increase the apparent protein content. The industry-standard test for protein content measures only nitrogen and correlates the result to protein; melamine is a very inexpensive nitrogen source that will dramatically increase apparent protein concentration under this type of test method.
The concern over melamine adulteration spread to includea number of different types of dry protein materials in both animal and human food sources. High visibility and the potential public health threat prompted the U.S. Food and Drug Administration (FDA) to rapidly issue a standard test method for the analysis of melamine in protein materials. This method, entitled “GC-MS Screen for the Presence of Melamine, Ammeline, Ammelide, and Cyanuric Acid”[3] is used in conjunction with nitrogen tests to verify that a material is protein-rich, rather than rich in non-protein nitrogen. The presence of any amount of melamine or its related analogues renders the product adulterated and illegal to sell. The FDA methodology is used for qualitative confirmation and semi-quantitative estimation only.
This paper presents a robust, efficient and definitive analysis of melamine and related analogues. The sample preparation, including extraction and derivatization, are discussed in addition to the optimized gas chromatography/mass spectrometry (GC/MS) method.
Experimental
The experimental portion of this paper is broken into three parts: extraction, derivatization and GC/MS analysis; this extraction, derivatization and GC/MS analysis closely follow the method published by FDA.
Extracting melamine from dry protein materials is relatively simple; a 0.5-g sample is sonicated in 20 mL of 10:40:50 (by volume) solution of diethylamine (DEA), water, acetonitrile. The sonication period is 30 minutes to allow for complete extraction. Following sonication, the sample is centrifuged for 10 minutes at 3,500 rpm to settle fine particulates from the sample matrix. The supernatant fluid is filtered through a 0.45-μm polytetrafluoroethylene (PTFE) filter; a PTFE filter is preferred because many compounds found in food-type matrices adhere to other filter materials. Once filtered, 200 μL of the solution is evaporated to dryness at 70 °C with a flow of clean, dry nitrogen gas. Once dry, the sample is ready for derivatization.
The dry sample is reconstituted in an autosampler vial with 200 μL pyridine and 100 μL of a 0.5-μg/mL internal standard (ISTD) solution. Melamine and related compounds are converted to trimethylsilyl (TMS) derivatives with the reagent Sylon-BFT (Supelco) consisting of bis(trimethylsilyl)trifluoroacetamide with 1% trimethylchlorosilane; 200 μL of this solution is added and the sample is incubated at 70 °C for 45 minutes. One important note on derivatization is to consider your matrix and how it will react with the reagent. When derivatizing an amino acid-rich matrix, less sample or more reagent may be necessary to compensate for the consumption of the Sylon-BFT by the amino-acids; some experimentation may be needed. The ISTD, 2,6-diamino-4-chloropyrimidine (DACP), monitors the derivatization process; poor ISTD recovery is indicative of incomplete derivatization, likely resulting from residual water or an active matrix.
The GC/MS system used in this paper was the PerkinElmer® Clarus® 600 GC/MS and the instrument parameters are summarized in Tables 1 and 2. The GC utilized a programmable split/splitless injector, isothermal at 280 °C. A 2-mm I.D. inlet liner was used; the 0.5-μL injection volume matched the solvent vapor-expansion volume with the inner volume of the inlet liner, an injection of larger volume will result in vapor expansion outside the liner; when the solvent expansion exceeds the liner volume, chromatographic peak shape often degrades.
The instrumental conditions presented here are optimized for the analysis of melamine in rice protein; in this case, it was necessary to extend the GC oven program to resolve ammeline and melamine from matrix peaks. The variety and complexity of sample matrices may require modifying the GC method to achieve resolution between analytes and matrix interference; modifications to achieve better resolution must be considered on a per-matrix basis. The best technique to match resolution to sample matrix is to compare the analysis of a blank spike with the analysis of the sample matrix. If the peaks associated with sample matrix elute at times different than the retention time of the analytes, the method does not need modification. However, if matrix peaks co-elute with analytes, then it is necessary to modify the temperature ramp of the GC oven to improve resolution.