Gas Chromatography (GC)

Gregor Kos and Rudolf Krska

Introduction

The routine use of gas chromatography (GC) is restricted to a limited number of mycotoxins, especially A- and B-trichothecenes. This is because of their non-fluorescent and weak UV-VIS absorption properties, which require derivatisation in order to facilitate detection, and makes the determination of trace levels unreliable. A more widespread use of GC methodology for other mycotoxins is mainly limited by insufficient volatility of analytes, the lack of suitable derivatisation agents, or the availability of alternative easy-to-use methods. GC methods have been reported for patulin, sterigmatocystin and zearalenone, but these are not widely used.

As it is true for HPLC methods, the main challenge in the application of GC methods is to provide comparable and accurate results: Several EC projects deal/dealt with the production of high purity calibrants and the production of (certified) reference materials as well as the organisation of intercomparison studies between different laboratories in order to study the variability of results, which is a prerequisite for the establishment and implementation of EU guidelines. The focus of these studies is on methods for the determination of trichothecenes and zearalenone.

An introduction to fundamental GC principles and applications is part of the EMAN Training Course on Separation and Detection Techniques. For details on the determination of the toxin content, please refer to the specific fact-sheet.

Basic Principles of GC Determination of Trichothecenes

Extraction and clean-up techniques have to be applied prior to separation and detection in order to enable well separated peaks without interferences from matrix components. Details are available in the internet training course "Sample Preparation Techniques for the Determination of Mycotoxins". Additionally, a derivatisation step is usually required, which enhances sensitivity and volatility of the analyte.

Coated fused silica columns are widespread and are employed according to the analyte and the type of derivatisation. (Scott and Kanhere, 1986) have compared different columns and evaluated their performance for the determination of trichothecenes. Injection takes place with a split/splitless system in splitless mode. The oven is heated with a programme running approximately from 60 Deg C to 270 Deg C, but higher initial temperatures have also been reported. Problems caused by adsorption of the analyte to the injector glass liner can be reduced significantly by silanisation.

The most sensitive and specific GC detectors are electron capture detection (ECD) and mass spectrometry (MS). Methods that skip the derivatisation step are also available and usually employ detection techniques such as flame ionisation detection (FID), which is considerably less sensitive than ECD detection or MS/MS instruments. Detection limits for ECD and MS (for electron impact ionised samples) are in the lower microgramme/kg range.

Multi-mycotoxin Methods

Zearalenone and up to 7 different trichothecenes can be identified in a single GC-MS run lasting about 37 min. A comparison with the retention time of n-alkyl-bis (trifluoromethyl) phosphine sulfide assists with the identification of individual toxins (M-series). Electron-impact (EI) mass spectra also provide valuable information for identification.

Figure 1: Chromatogram of wheat spiked with 500 microgramme/kg DON, NIV, 3-AcDON, 15-AcDON and FusX

Determination of Individual Toxins

B-Trichothecenes are commonly determined using GC methods. The limit of detection ranges from 20-50 microgramme/kg (GC-ECD) down to 5 microgramme/kg (GC-MS). Recoveries are in the 70-110% range.

The aim of the EC-SMT project (SMT4-CT96-2047) was to improve the analytical methodology for the detection of the Fusarium mycotoxins Deoxynivalenol (DON), Nivalenol (NIV), HT2-Toxin and T2-Toxin in wheat by use of GC-ECD and GC-MS. Laboratories from 21 European countries participated in several comparison studies. Participants employed an AOAC official method (Nr. 986.18) and the results of between-laboratory-CVs were 54% and 41% (at concentrations levels of 350 microgramme/kg and 750 microgramme/kg respectively). The major problems identified by Petterson were:

• Different trichothecene response for pure calibrants and calibrants in a matrix
• Non-linearity of calibration curves
• Recovery too high
• Drifting response for trichothecenes
• High variation of repeatability within a run for MS detection
• Carry over or memory effects from previous samples
• Matrix interference

A-Trichothecenes: GC methodology is also preferred for A-Trichothecenes and is based on the derivatisation of the hydroxyl groups to increase volatility and sensitivity. Trimethylsilylation and fluoroacylation are common. ECD or MS detection is recommended. Because of the poor results using GC methods for trichothecene analysis (SMT4-CT96-2047 project summarised above), possible LC methods are under consideration.

Aflatoxins: For aflatoxins, HPLC, and in some cases TLC, are clearly the methods of choice. However, there are also a small number of GC methods available. MS detection is possible with or without derivatisation. Methylsilicone-coated fused silica columns were used in a study conducted by Trucksess (1984). FID detection has also been reported, which enabled the separation of aflatoxins B1, B2 and G1, G2 on a 5% phenyl methlyl silicone column (Goto et al., 1988). MS can also be applied for detection and identification of aflatoxins.

Zearalenone (ZON) can be detected simultaneously with trichothecenes. Methods specifically designed for ZON have been reported by Scott et al. (1992) and Ryu et al. (1996). A wide range of available HPLC methods is usually preferred over GC methods.

Fumonisins: Gas Chromatography/Mass Spectrometry (GC/MS) of fumonisins, which is based on hydrolysis of the esterified side chain and derivatisation with trimethylsilyl or tri-fluoroacetate on the fumonisin backbone has been reported. Although sensitive and selective, the method involves expensive equipment and a hydrolytic pre-treatment.

Patulin: Determination of patulin by GC-MS detection and quantification of the heptafluorobutyrate derivative after treatment with heptafluorobutyrylimidazole (HFBI) was reported by Tarter and Scott (1991). A GC run lasted just over 15 min with the patulin derivative eluting at 11.5 min. Trimethylsilyl (TMS) can also serve as a derivatisation agent for the detection by ECD or MS. Detection limits were in the microgramme/l range. It should be noted that HPLC methods are generally preferred over GC methods, because they are easier to apply.

References

[1] Betina V. (1989) "Chromatographic methods as tools in the field of mycotoxins", Elsevier Science Publishers B.V., Amsterdam, The Netherlands.

[2] Goto T., Matsui M., Kitsuwa T. (1988), "Determination of Aflatoxins by Capillaty Column Gas Chromatography". Journal of Chromatography 447 (2) 410 - 414.

[3] Krska R., Baumgartner S., Josephs R., (2001) "The State-of-the-Art in the Analysis of Type A- and Type B-trichothecene Mycotoxins in Cereals", Fresenius Journal of Analytical Chemistry, 371 285-299.

[4] Krska R., Josephs R., (2001) "The State-of-the-Art in the Analysis of Estrogenic Mycotoxins in Cereals", Fresenius Journal of Analytical Chemistry, 369 469-476.

[5] Langseth W., Rundberget T., (1998) Instrumental Methods for Determination of Non-macrocyclic Trichothecenes in Cereals, Foodstuffs and Cultures, Journal of Chromatography A, 815 103-121.

[6] Ryu J.C., Yang J.S., Song Y.S., Kwon O.S., Park J., Chang I.M. (1996) "Survey of Natural Occurrence of Trichothecene Mycotoxins and Zearalenone in Korean Cereals Harvested in 1992 using Gas Chromatography Mass Spectrometry" Food Additives and Contaminants 13 (3) 333 - 341.

[7] Scott P.M., Kanhere S.R. (1986) "Comparison of Column Phases for Separation of Derivatised Trichothecenes by Capillary Gas Chromatography." Journal of Chromatography 368 (2) 374 - 380.

[8] Scott P.M. (1992) "Mycotoxins" Journal of the AOAC International 75 (1) 95 - 102.

[9] Tarter E.J., Scott P.M. (1991) "Determination of Patulin by Capillary Gas Chromatography of the Heptafluorobutyrate Derivative". Journal of Chromatography 538 (2) 441 - 446.

[10] Trucksess M.W., Brumley W.C., Nesheim S. (1984) " Rapid Quantitation and Confirmation of Aflatoxins in Corn and Peanut Butter, using a Disposable Silica Gel Column, Thin Layer Chromatography and Gas Chromatography / Mass Spectrometry" Journal of the Association of Official Analytical Chemists 67 (5) 973 - 975.