Determination and detection 531 Chromatographic methods

5.3.1.1 High-performance liquid chromatography. A wide variety of methods for most mycotoxins are based on HPLC, which is the methodology of choice for aflatoxins, OTA, ZEA, FBs and patulin. The HPLC separation is followed by fluorescence detection for aflatoxins, OTA, ZEA and FBs, usually as derivative compounds. UV detection is used for DON and patulin determination. Advantages of HPLC determination include excellent performance characteristics, low detection levels (as low as 0.1 ng/g) and safety for the operator.

For aflatoxins, a large number of HPLC-based methods are available. The AOAC method 49.2.19A includes a Mycosep® clean-up procedure, followed by fluorescence detection of the trifluoroacetic acid (TFA) derivative, for aflatoxins in corn, almonds, Brazil nuts, peanuts and pistachio nuts [133]. The multifunctional column contains a mixture of reversed-phase, ion exclusion and ion exchange adsorbents. Packing retains interferences such as fats, proteinaceous compounds, pigment and carbohydrates extracted from the matrices. Methods including IAC and post-column derivatization with pyridinium bromide perbromide (PBPB) are reported for the determination of AFB1 and total aflatoxins in peanut butter, pistachio paste, fig paste and paprika powder [134], and for AFBj determination in baby food (infant formula) [135]. Other derivatization reactions for aflatoxin analysis include post-column derivatization with iodine [136], electrochemically generated bromide (Kobra Cell) [137,138], photolytic derivatization [139] and pre-column derivatization with TFA [140]. HPLC methods, without any derivatization step, are also reported for the determination of AFM1 in milk and cheese [141,142], since this aflatoxin shows native fluorescence that allows this simplification in the analytical procedure.

Methods for OTA determination by HPLC, mainly based on fluorescence detection, are available for different matrices. Official methods include OTA determination in barley, roasted coffee [143], baby food, and wine. Since OTA shows native fluorescence usually no derivatization step is required for the detection of this toxin. For confirmation purposes three methods are usually adopted, based on methylation [144,145], ammonia-derivative formation [146] or direct determination by LC-MS [147,148].

Notwithstanding, trichothecenes are frequently analysed by GC, HPLC methods are also available, such as that developed by Trucksess et al. [149] for DON in wheat-finished products. The purification step is performed by passing the extract through an SPE cartridge, and the HPLC separation is followed by UV determination.

The LC determination of ZEA and a-zearalenol (a-ZOL) in corn was collaboratively studied by Bennett et al. [150]. The determination of ZEA and a-ZOL using UV or fluorescence detection has been adopted as an official method by AOAC [126]. The extraction is based on liquid/liquid partition between CHCl3 and water with the aid of diatomaceous earth. Other methods based on HPLC and fluorescence detection are reported by Tanaka et al. [151] and Seidel et al. [152].

FB analysis is usually performed by reversed-phase LC with fluorometric detection after pre-column derivatization with o-phtaldialdehide (OPA)/ 2-mercaptoethanol [153,154]. An AOAC-IUPAC collaborative study develops an LC method, based on the previous use of SPE cartridges for the determination of fumonisins Bj, B2 and B3. This method has been adopted by AOAC International as an official first action method for corn analysis [133]; however, the method could not be successfully applied for the analysis of corn-based food products such as corn bran flour, corn bran breakfast cereals, mixed baby cereals and cornflakes due to low recoveries and inadequate clean-up. The use of IACs for the clean-up step has been reported to produce higher FB recovery [155].

LC with fluorescence detection is still the method of choice for the determination of FBs [153,154]. An AOAC-IUPAC collaborative study developed an LC method, based on the use of SPE cartridges and OPA as pre-column derivative agent for the determination of FBj, FB2 and FB3. This method was accepted as an official method of analysis [133]. A collaborative study was also preformed using IAC clean-up and pre-column derivatization with OPA. The method is suitable for the determination of FB: and FB2 [156]. The use of SAX (strong anion exchange)+C18 in the clean-up step for FBs analysis was presented by Moller and Gustavsson [157].

Also for patulin, HPLC determination is preferred. Since there is no IAC for patulin, the sample preparation step is different. According to the method reported by Mac Donald [158], the sample is extracted with ethyl acetate, employing the enzyme pectinase for ameliorating the extraction performance on fruit matrix, dried under a stream of nitrogen, redissolved in mobile phase and determined with HPLC. Patulin detection and quantification is then performed with UV detection (276 nm). A combination of Extrelut and Florisil cartridges for the patulin determination in apple cider [159] by LC method is used for extraction and clean-up steps. The use of SPE was also reported by Trucksess [160].

Minor mycotoxins analysis is also performed by HPLC, methods with fluorescence detection are reported for citrinin determination [161,162]. A method for the determination of cyclopiazonic acid in corn and rice by HPLC and UV with a photodiode array detector is reported [163]. The determination of cyclopiazonic acid in cheese is also performed by HPLC anticipated by an SPE clean-up step [123].

A wide number of HPLC methods for the analysis of various mycotoxins, tested by collaborative studies to evaluate their effectiveness, are periodically reported on Methods Committee Reports published in Journal of AOAC International. Validated methods for mycotoxins determination are also collected by CEN within WG5 Biotoxins of TC275. An overview of the official methods available for mycotoxins analysis is reported in Table 4. Mycotoxins HPLC chromatograms are shown in Figures 7 and 8.

5.3.1.2 Gas chromatography. Since trichothecenes do not strongly absorb in the UV-Visible range and are non-fluorescent, GC-based methods represent the most widely employed technique for their determination. In particular, the analytical techniques employing GC and capillary column are useful for the simultaneous determination of different trichothecenes. For the detection step heptafluorobu-tyryl (HFB), trimethylsilyl (TMS) and trifluoroacetyl derivatives are frequently used coupled with electron capture detection (ECD) [164,165]. However, MS is strongly recommended for peaks confirmation. A GC method for trichothecenes determination in barley and malt [166] has been validated and accepted by the American Society of Brewing Chemists in its Methods of Analysis, 8th ed. A method, for trichothecenes determination in barley and malt, has also been developed by supercritical fluid chromatography (SFC) [167]. Supercritical fluids can be used as the mobile phase to separate analytes with either a packed or capillary GC column. Compared with GC, capillary SFC can provide high-resolution chromatography at much lower temperatures. A GC method for the determination of DON in wheat is reported by AOAC Official methods [168].

Table 4 List of Official AOAC and CEN methods

Mycotoxins

Matrix

Reference

Method

Aflatoxins (AFBj, AFB2, AFGi, AFG2)

Aflatoxins

Aflatoxins

Aflatoxins Aflatoxins (AFBj and AFs)

Aflatoxins (AFBj,

AFB2, AFGi) AFBj

AFBi

Aflatoxins AFMi

AFMi

AFMi AFBi

OTA OTA

Peanuts, pistachios, figs, and paprika powder

Nuts and nut products

AOAC-999.07 CEN-EN i4i23:2003

AOAC-2005.08

Almonds, peanuts, AOAC-994.08 pistachio nuts, brazil nuts

Peanut butter AOAC-991.45

Cereals, shell fruits AOAC-991.31

and derived products

Corn

Feeding stuffs

Baby foods

Peanuts and corn Milk and milk powder

Milk and milk powder Milk and cheese Foods for infants and young children Corn and barley Cereals and derived products

CEN-EN i2955:i999

AOAC-993.i6

AOAC-2003.02 CEN-EN ISO i7375:2006

AOAC-2000.i6

AOAC-993.i7 AOAC-2000.08 CEN-EN ISO i450i:i999 CEN-EN ISO i4675:2003 AOAC-980.2i CEN-New itema

AOAC-99i.44 CEN-EN ISO i5i4i-i:i998

IAC-HPLC with post column derivatization and fluorescence detection HPLC with post-column photochemical derivatization Mycosep-HPLC

ELISA

IAC-HPLC with post column derivatization and fluorescence detection ELISA

IAC-HPLC with post column derivatization and fluorescence detection IAC-HPLC with post column derivatization and fluorescence detection TLC

IAC-HPLC with fluorescence detection ELISA

HPLC

Silica gel clean-up-HPLC with fluorescence detection

Mycotoxins

Matrix

Reference

Method

OTA OTA

DON DON

ZEA ZEA

Fumonisins (FBj, FB2, FB3)

Cereals and derived products

Barley

Roasted coffee

CEN-EN ISO 15141-2:1998

AOAC-2000.03

AOAC-2000.09

Barley and Roasted CEN-EN coffee 14132:2003

Wine and beer

Green coffee Barley and green coffee Foods for infants and young children Dried (vine) fruit

Wheat

Cereals (including maize) and cereal products Foods for infants and young children Wheat Corn

Cereals (including maize) and cereal products Foods for infants and young children Wheat and animal feed Corn Corn

AOAC-2001.01 CEN - EN

14133:2003 AOAC-2004.10 AOAC-975.38

CEN-New itema

CEN-New itema

AOAC-986.18 CEN-New itema

CEN-New itema

AOAC-986.17 AOAC-985.18 CEN-New itema

CEN-New itema

AOAC-994.01

AOAC-976.22 AOAC - 995.15

Bicarbonate clean-up-HPLC with fluorescence detection IAC-HPLC with fluorescence detection IAC-HPLC with fluorescence detection IAC-HPLC with fluorescence detection IAC-HPLC with fluorescence detection HPLC TLC

IAC-HPLC with fluorescence detection IAC-HPLC with fluorescence detection GC

IAC-HPLC with UV detection

TLC HPLC

IAC-HPLC with fluorescence detection

ELISA (qualitative)

TLC HPLC

Table 4 (Continued)

Mycotoxins

Matrix

Reference

Method

Fumonisins

Maize based foods

AOAC-2001.04

IAC-HPLC with

(FBlr FB2)

CEN-EN

pre column

14352:2005

derivatization

and fluorescence

detection

Fumonisins

Maize

CEN-EN

SPE-HPLC with

(FBlr FB2)

13585:2002

pre column

derivatization

and fluorescence

detection

FBs

Corn

A0AC-2001.06

ELISA

Fumonisins

Maize based foods

CEN-New itema

-

(FBlr FB2)

for infants and

young children

Patulin

Clear and cloudy

AOAC-2000.02

Liquid/liquid

apple juice and

partition clean-

puree

up-HPLC with

UV detection

Patulin

Clear and cloudy

CEN-EN

Liquid/liquid

apple juice and

14177:2004

partition clean-

puree

up-HPLC with

UV detection

Patulin

Apple juice

AOAC-974.18

TLC

Patulin

Clear and cloudy

CEN-New itema

Liquid/liquid

apple juice and

partition clean-

puree for infant

up and SPE-

and young

HPLC with UV

children

detection

aMethods of analysis ready to be mandated according to the Mandate for standardization addressed to CEN in the field of methods of analysis for mycotoxins in food. Deadline 12/2008.

aMethods of analysis ready to be mandated according to the Mandate for standardization addressed to CEN in the field of methods of analysis for mycotoxins in food. Deadline 12/2008.

GC coupled with MS could be used for the confirmation of patulin in apple juice [155]. GC/MS methods are also reported operating both without [169] and with various derivatizing agents [170,171]. Limitations such as the need to derivatize the samples before analysis have led to a prevalence of LC-MS over GC-MS approaches [172]. Figure 9 shows a chromatogram obtained after separation and detection with GC-ECD after trimethylslylation of five B trichothecenes.

5.3.1.3 Thin-layer chromatography. The employment of TLC-based methods has decreased steadily over the past few years, but, especially in developing countries, this analytical technique is still routinely used. TLC methods allow for

Milk Afm1 Hplc Graphic

Minutes

Figure 7 HPLC chromatogram of a corn sample naturally contaminated with aflatoxins. Level of contamination: AFB1 and AFG1, each at 6.03 mg/kg, and AFB2 and AFG2, each at 1.49 mg/kg.

Acdon Acdon
Figure 8 HPLC chromatogram of a wheat sample naturally contaminated with OTA. Level of contamination: 4.53 mg/kg.

a precise determination at levels not lower than 2ng/g; their disadvantages however include the use of solvents that are considered to be ecological hazard. TLC techniques have been reviewed and updated for the analysis of selected Fusarium toxins by Krska [120]. A large number of TLC-based methods have been

Gas Chromatogram

Figure 9 Gas chromatogram for wheat spiked with DON, 3-AcDON, 15 Ac-DON, Fus-X, and NIV at a concentration of about 1,000 ng/g. Source: International Programme Chemical Safety (IPCS) Inchem, www.inchem.org DON.

Time (min)

Figure 9 Gas chromatogram for wheat spiked with DON, 3-AcDON, 15 Ac-DON, Fus-X, and NIV at a concentration of about 1,000 ng/g. Source: International Programme Chemical Safety (IPCS) Inchem, www.inchem.org DON.

accepted by the AOAC as official methods including the methods for the determination of DON in barley and wheat, for aflatoxins in peanuts and corn, for AFM1 in milk and cheeses, for OTA in barley and in green coffee and for ZEA in corn. All the above-mentioned methods are reported in Table 4. The proposed use of TLC in combination with IACs represents a promising application that can improve the performance characteristics of the TLC-based methods [173]. An improvement in the technique includes the use of a microcomputer interfaced with a fluorodensitometer to simplify data handling [174]. New detection techniques for TLC have also been developed as alternatives to traditional TLC scanners; the employed principles are based on the use of a semiconductor-based detection cell (SeBaDeC) or of a modified office scanner. As the cost of commercial office scanners is very low, these are of special interest for the quantification of mycotoxins whenever other instruments are not available, such as in developing countries, in which this technique still represents the most economic method for the analysis of trichothecenes.

5.3.1.4 Liquid chromatography-mass spectrometry. LC coupled with MS and especially, LC-MS-MS have become very popular in recent years, because of the universal, selective and sensitive detection. Combining HPLC with MS/MS results in a powerful tool for characterization and identification especially for detection of the so-called conjugated mycotoxins, in which the toxin is usually bound to a more polar substance such as glucose or other sugars. These conjugated are referred to as masked mycotoxins, as these substances escape routine detection methods but can release their toxic precursors after hydrolysis [175]. Also with LC-MS there is currently a strong trend toward multimycotoxin methods for the simultaneous determination, particularly for mycotoxins belonging to different chemical families [176]. This approach has the advantage of enabling the reduction or possibly the omission of sample clean-up. Determination of DON and NIV in corn [177] and a rapid determination of FB1 are performed without any clean-up or purification step by a liquid chromato-graphic/tandem mass spectrometrer using pneumatically assisted electrospray ionization (HPLC-ESI-MS), the mass analyser being a triple-stage quadrupole equipped with a standard ESI interface [178]. Other methods described in the literature include a clean-up step, for example, graphitised carbon black for extracts from corn meal samples before LC-ESI-MS-MS multiresidue analysis [179]. An HPLC-ESI-MS method was developed for the determination of 18 mycotoxins and metabolites, OTA, ZEA, a-ZOL, b-ZOL, a-zearalanol (zeranol), b-zearalanol (taleranol), FB1, FB2, T-2 toxin, HT-2 toxin, T-2 triol, DAS, 15-MAS, DON, 3-acetylDON, 15-acetylDON, deepoxy-DON and AFM1, in milk [180]. Results confirm that, especially for multimycotoxin methods expanding over a broad range of classes, sample pretreatment is a critical step, because even within a single class of mycotoxins significant losses may occur during extraction or clean-up. A method for the determination of 39 mycotoxins in wheat and maize using a single extraction step followed by LC-ESI-triple-quadrupole MS without the need for any clean-up, has recently been developed and validated in-house [181]. The analytes included type-A and type-B trichothecenes and related derivatives including conjugated DON, FBs, ochratoxins, ZEA, aflatoxins, moniliformin, enniatins and ergot alkaloids; the limits of detection ranged from 0.05 to 220 mg/kg. The extraction solvent mixture, diluted 1:2, and injection without any clean-up, was chosen as a compromise for the extraction of the 39 analytes from both wheat and maize. In addition to the multianalyte requirement, multisubstrate methods are even more demanding. This approach, nevertheless, has the advantages of speed and ease of application with one method for multiple mycotoxin-matrix combinations over a broad range of mycotoxin classes, which is of special interest for high-throughput routine analysis. LC-MS instruments using atmospheric pressure chemical ionization (APCI) interfaces have recently been employed for the determination and identification of ZEA, trichothecenes, including DON and NIV, at trace levels [118,119,182,183].

Figure 10 shows a total ion LC-MS-MS chromatogram obtained after clean-up of a spiked maize sample containing 100 mg/kg of each mycotoxin.

0 0

Post a comment