Determination of Aflatoxins in Food by

LC/MS/MS

Application

Food Safety

Authors disease from AF exposure and to advance world-

wide surveillance of food. Analysis of AFs in food
Masahiko Takino products is routinely performed by thin-layer chro-
Agilent Technologies matography (TLC) and liquid chromatography
9-1 Takakura-Cho (LC) with fluorescence detection (FD) in combina-
Hachiouji-Shi, Tokyo tion with both precolumn derivatization and post-
Japan column derivatization. The LC/FD technology is
often used due to the high selectivity and sensitiv-
Toshitsugu Tanaka
ity of these methods. Furthermore, hyphenated
Kobe Institute of Health
techniques such as LC coupled to mass spectro-
Department of Food Chemistry metric (MS) detection have been developed and
4-6 Minatojima-nakamachi applied in residual analysis of foods. The high
Chuo-ku, Kobe selectivity and sensitivity of MS detection methods
Japan associated with the resolution of LC provide deci-
sive advantages to perform qualitative as well as
quantitative analysis of a wide range of molecules
Abstract at trace levels. Several papers describing different
kinds of MS methods for the analysis of AFs have
A sensitive and selective analytical method for the deter- been published [2-4.]
mination of aflatoxin G1, G2, B1, and B2 residues in cere-
als using the Agilent G6410AA LC/MS Triple Quadrupole
Mass Spectrometer was developed. This method uses Experimental
simple sample preparation methods followed by LC/MS/
MS. The limits of detection for all aflatoxins were less Sample Preparation
than 1 ng/mL in cereals.
The samples analyzed (peanuts, corn, nutmeg, and
red pepper) were obtained from a local market and
Introduction did not include any AFs. The extraction and
cleanup steps for AFs were carried out according
Aflatoxins (AFs) belong to a closely related group to validated methods reported by Tanaka [5].
of secondary fungal metabolites. These mycotoxins Briefly, 20 g fine ground sample was poured into a
are severely toxic metabolites produced mainly by 200-mL Erlenmeyer flask, followed by adding
Aspergillus flavus and A. parasiticus, and expo- 40 mL acetonitrile-water (9:1, v/v) for corn and
sure to them can cause cancer in humans and live- cereals. After shaking for 30 min, the mixed
stock [1]. Based on epidemiological evidence, AFs solution was centrifuged for 5 min at 1,650 g. The
have been classified as human liver carcinogens by supernatant obtained was filtered through a glass
the World Health Organization and by the U.S. microfiber GF/B grade filter (Whatman Interna-
Environmental Protection Agency. Thus, accurate
determination of AFs is required to avoid human
tional Ltd, Maidstone, UK). A 5-mL portion of the pump, well-plate autosampler, thermostatted
filtrate was applied to a MultiSep number 228 car- column compartment, the Agilent G6410 Triple
tridge column for the cleanup. After passing Quadrupole Mass Spectrometer with an electro-
through at a flow rate of 1 mL/min, 2 mL of the spray ionization (ESI) source. The objective of the
first eluate was collected. The eluate was evapo- method development was to obtain a fast and sen-
rated to dryness at 40 °C under a gentle stream of sitive analysis for quantifying AFs in foods. For
nitrogen. The residue was reconstituted in 1 mL chromatographic resolution and sensitivity, differ-
methanol-water (4:6 v/v) containing 10 mM ent solvents and columns were optimized. It was
ammonium acetate. found that a simple solvent system using water,
methanol, ammonium acetate, and a 1.8-µm
Standard Preparation particle size C18 column worked very well.

Each of the standard reagents, aflatoxin G2 (AFG2),

aflatoxin G1 (AFG1), aflatoxin B2 (AFB2) and afla- LC Conditions
toxin B1 (AFB1), was dissolved in acetonitrile at 1 Instrument: Agilent 1200 HPLC
mg/mL and was stored at 4 °C in the dark until Column: ZORBAX Extend C18, 100 mm × 2.1 mm,
1.8 µm (p/n 728700-902)
use. To prepare the working standard for LC/MS
Column temp: 40 °C
analysis, each AF stock solution was equally pipet-
Mobile phase: A = 10 mM ammonium acetate in water
ted and transferred to a vial, and it was then B= Methanol
diluted with the mobile phase. The final concentra- 40% A/60% B
tion of each AF was 1 ng/mL. Flow rate: 0.2 mL/min
Injection volume: 5 µL
Chemicals
MS Conditions
The standards AFG2, AFG1, AFB2, and AFB1 were Instrument: Agilent 6410 LC /MS Triple Quadrupole
obtained from Sigma Aldrich Japan (Tokyo, Japan). Source: Positive ESI
Drying gas flow: 10 L/min
The purity of these compounds was greater than
Nebulizer: 50 psig
99%. Ammonium acetate, toluene, HPLC-grade ace-
Drying gas temp: 350 °C
tonitrile, and HPLC-grade methanol were obtained Vcap: 4000 V
from Wako Chemical (Osaka, Japan). Water was Scan: m/z 100 – 550
purified in-house with a Milli-Q system (Millipore, Fragmentor: Variable 100 V
Tokyo, Japan). The cartridge column of MultiSep MRM ions: Shown in Table 1
number 228 was purchased from Showa Denko Collision energy: Shown in Table 1
(Kanagawa, Japan).

LC/MS Instrument LC/MS/MS Method

The LC/MS/MS system used in this work consists Quantitative analysis was carried out using MRM
of an Agilent 1200 Series vacuum degasser, binary mode. The parameters for MRM transitions are
shown in Table 1.

Table 1. Data Acquisition Parameters of MRM Transitions for Each Aflatoxin

RT Molecular Precursor Product Collision

No Mycotoxins (min) weight ion (m/z) ion (m/z) energy (V)
1 Aflatoxin G2 5.21 330 331 245 30
2 Aflatoxin G1 6.61 328 329 243 30
3 Aflatoxin B2 8.44 314 315 259 30
4 Aflatoxin B1 10.89 312 313 241 30

2
Results and Discussion mode. The optimum collision voltage is compound
dependent. To establish the optimum collision volt-
Optimization of MRM Transitions age, this parameter was varied from 5 to 40 V
using a step size of 5V. As shown in Figure 2, a dis-
Determination of the optimal MRM transitions for tinct optimum in the intensity of the product ion of
each aflatoxin was carried out using single-MS full- each AF was observed at 30 V. The product ions
scan mode followed by product ion scan mode that indicated the highest intensity were m/z 245
using taflatoxin standard mixtures at 1 ug/mL. (AFG2), 243 (AFG1), 259 (AFB2), and 241 (AFB1),
Mass spectra of these standard mixtures in full respectively. On the basis of the above results, the
scan mode and product ion scan mode are shown collision voltage was set to 30 V.
in Figures 1 and 2. The mass spectrum of each
aflatoxin by full-scan mode exhibited the proto- Table 1 shows the parameters of MRM mode of
nated molecule [M+H]+ as the base peak ion. These each aflatoxin.
ions were selected as precursor ions for MRM

×10 5 ×10 5
331(M+H) + 315(M+H) +
2.4
2.4
(A) (C)
2.0 2.0

1.6
Abundance

Abundance
1.6

1.2 1.2

0.8 0.8
353
0.4 313 0.4 287 337

×10 5 329(M+H) + ×10 5 313(M+H) +

1.8 1.8
(B) (D)
1.4 1.4
Abundance

Abundance

1.0 1.0

0.6 0.6
351 285
243 311
0.2 0.2 241 335

100 140 180 220 260 300 340 100 140 180 220 260 300 340
m/z m/z

Figure 1. Mass spectra of four aflatoxins standard in single-MS full-scan mode at 1 µg/mL (A): aflatoxin G2, (B): aflatoxin G1,
(C): aflatoxin B2, and (D): aflatoxin B1.

3
 ×10 3 ×10 3
189 6 259
(A) 245
(C)
5.5
4
5
4.5
Abundance

Abundance
3 4
217 115
141 3.5
3 287
2
285
2.5
2 203
1 331(M+H) + 1.5 315(M+H) +
1
0.5
0

×10 3 200 ×10 3 241

7 (D)
(B) 9
128
8
128 243
7
5 185
Abundance

Abundance
215 6 213
5
3 171 4
269
3 157.1
2
1
283 329(M+H) + 313(M+H) +
1
0
120 160 200 240 280 320 120 160 200 240 280 320
m/z m/z

Figure 2. Mass spectra of four aflatoxins standard in product ion scan mode at 1µg/mL (A): aflatoxin G2, (B): aflatoxin G1,
(C): aflatoxin B2, and (D): aflatoxin B1.

The MRM chromatogram of each aflatoxin at

0.1 ng/mL is shown in Figure 3. These show excel-
lent signal-to-noise (S/N) ratios for all aflatoxins.
The limit of detection (LOD) for each aflatoxin was
determined using an S/N ratio of 3 with this MRM
chromatogram and is shown in Table 2. To evalu-
ate the linearity of the calibration curves, various
concentrations of aflatoxin standard solutions
ranging from 0.1 ng/mL to 100 ng/mL were ana-
lyzed. As shown in Figure 4 and Table 2, the linear-
ity was very good for all aflatoxins with
correlation coefficients (r2) greater than 0.999.

4
 1. Aflatoxin G2 3. Aflatoxin B2
5000
5000

4000
4000
Abundance

Abundance
3000 3000

2000 2000

1000
1000

6000
8000 4. Aflatoxin B1
2. Aflatoxin G1

6000

Abundance
Abundance

4000

4000

2000
2000

1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 5 6 7 8 9 10 11 1
Retention time (min) Retention time (min)

Figure 3. MRM chromatograms of four aflatoxin standards at 0.1 ng/mL in MRM mode.

80000

40000 1. Aflatoxin G2 3. Aflatoxin B2

60000
30000
Abundance
Abundance

40000
20000

20000
10000

0 50 100 0 50 100
100000
90000

80000
2. Aflatoxin G1 70000 4. Aflatoxin B1

60000
Abundance

50000
Abundance

40000 30000

20000 10000

0 50 100 0 50 100
Conc. (ng/mL) Conc. (ng/mL)

Figure 4. Calibration curves of four aflatoxins ranged from 0.1 ng/mL to 100 ng/mL.

5
Table 2. Linearity and LODs of Four Aflatoxins

No Mycotoxins r2 LOD (ng/mL)

1 Aflatoxin G2 0.9999 0.025
2 Aflatoxin G1 0.9992 0.020
3 Aflatoxin B2 0.9999 0.025
4 Aflatoxin B1 0.9993 0.020

The matrix effect of this method was investigated

by using cereal and corn extracts spiked with
mycotoxin standards at 0.2 ng/mL. Typical MRM
chromatograms of cereal and corn extract are
shown in Figures 5 and 6, respectively. There were
no additional peaks from sample matrix in either
food when compared with the mycotoxin standard
mixture. These results indicate that MRM mode
has very high selectivity.

Aflatoxin G2 10000 Aflatoxin B2

10000

8000
8000
Abundance

Abundance

6000 6000

4000 4000

2000
2000

16000
12000
Aflatoxin G1 Aflatoxin B1
10000
12000
Abundance

Abundance

8000

8000
6000

4000
4000
2000

1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 5 6 7 8 9 10 11
Retention time (min) Retention time (min)

Figure 5. MRM of four aflatoxins in cereal extract spiked at 0.2 ng/g.

6
 Aflatoxin G2 10000 Aflatoxin B2
10000

8000
8000
Abundance

Abundance
6000 6000

4000 4000

2000
2000

16000
12000
Aflatoxin G1 Aflatoxin B1
10000
12000
Abundance

Abundance
8000

8000
6000

4000
4000
2000

1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 5 6 7 8 9 10 11
Retention time (min) Retention time (min)

Figure 6. MRM of four aflatoxins in corn extract spiked at 0.2 ng/g.

Furthermore, the change on the peak intensity of References

each aflatoxin by the sample matrix was investi-
gated by comparison with the peak intensity of 1. K. K. Sinha and D. Bhatnagar, 1998, “Mycotox-
aflatoxin standards. As these results show in ins in Agriculture and Food Safety,” 1998 (New
Table 3, the relative intensity of each pesticide York: Marcel Dekker)
ranged from 88 to 96%. Thus, matrix effects such 2. W. J. Hurst, R. A. Martin, and C. H. Vestal, 1991,
as ion suppression may be insignificant and it is “The Use of HPLC/Thermospray MS for the Con-
possible to use external standards instead of firmation of Aflatoxons in Peanuts,” J. Liq.
matrix-matched standards. Chromatogr., 14, 2541-2540.
Table 3. Relative Intensity of Each Aflatoxin in Sample 3. A. Cappiello, G. Famiglini, and B. Tirillini, 1995,
Extracts “Determination of Aflatoxins in Peanut Meal by
No Mycotoxins Relative intensity (%) LC/MS with a Particle Beam Interface,” Chro-
Cereal Corn matographia, 40, 411-416.
1 Aflatoxin G2 88 91
4. M. Vahl and K. Jorgensen, 1998, “Determination
2 Aflatoxin G1 92 94
of Aflatoxins in Food Using LC/MS/MS,”
3 Aflatoxin B2 93 96 Z Lebensm Unters Forsch A., 206, 243-245.
4 Aflatoxin B1 97 95
5. T. Tanaka, A. Yoneda, Y. Sugiura, S. Inoue,
Conclusions M. Takino, A. Tanaka, A. Shinoda, H. Suzuki,
H. Akiyama, and M. Toyoda, 2002, “An Applica-
The multi-aflatoxin method by LC/MS/MS tion of Liquid Chromatography and Mass Spec-
described here was suitable for the determination trometry for Determination of Aflatoxins,”
of four aflatoxins in cereal and corn extract due to Mycotoxins, 52, 107-113.
its high sensitivity and high selectivity. Another
advantage of this method is that ion suppression
was not observed for all food samples studied.
Thus, it may eliminate the need for matrix-
matched standards, which makes analysis more
tedious for samples from different origins.
7
www.agilent.com/chem

For More Information

For more information on our products and services,
visit our Web site at www.agilent.com/chem.

For more details concerning this application,

please contact masahiko_takino@agilent.com.

Agilent shall not be liable for errors contained herein or for incidental or consequential
damages in connection with the furnishing, performance, or use of this material.

Information, descriptions, and specifications in this publication are subject to change

without notice.

© Agilent Technologies, Inc. 2008

Printed in the USA

January 4, 2008
5989-7615EN