Z Lebensm Unters Forsch (1992) 195:316-319
Zeitschrift for
9 Springer-Verlag 1992
Original paper A rapid, sensitive and economic method for the detection, quantification and confirmation of aflatoxins Paul Majerus 1 and Zaki Zakaria 2 1 ChemischesUntersuchungsamt Trier, Maximineracht l 1a, W-5500 Trier, Federal Republic of Germany 2 Assiut University, Faculty of Veterinary Medicine, Veterinary Medicine - Forensic Medicine and Toxicology,Assiut, Egypt Received May 19, 1992
Eine schnelle, empfindUche und kostengiinstige Methode zum Nachweis, zur Bestimmung und Bestfitigung von Aflatoxinen Zusammenfassung. Es wird eine schnelle, empfindliche und kostengfinstige Methode zum Nachweis, Bestimmung und Best/itigung von Aflatoxinen beschrieben. Die Aflatoxine B1, B2, G1 und G2 werden mit Methanol/ Wasser (85 + 15) extrahiert und in Dichlormethan iiberffihrt. Der Dichlormethanextrakt wird auf einer mit 0,5 g Kieselgel 60 geffillten Polypropylens/iule gereinigt. Die Aflatoxine werden mit Chloroform/Aceton (90+10) eluiert und mit zweidimensionaler DC auf Kieselgel-Alufolien nachgewiesen. Die mittleren Wiederfindungsraten ffir die Aflatoxine B~, B2, G~ und G 2 in Maismehl betragen 73, 78, 80 und 64%, die Nachweisgrenzen liegen durchschnittlich bei 0,5 gg/kg. Zur Bestfitigung verdfichtiger Befunde kann auf der Platte mit Trifluoressigs/iure derivatisiert werden. Die Methode ist bisher an einer Vielzahl von verschiedenen Lebensmitteln mit gutem Erfotg getestet worden. Summary. A rapid, sensitive and economic method for the detection, quantification and confirmation of aflatoxins is described. Aflatoxins B~, B2, G1, and G2, are extracted by methanol/water (85+ 15) and partitioned into methylene dichloride. The methylene dichloride solution is cleaned up on a polypropylene column, filled with 0.5 g silica gel 60. The aflatoxins are eluted with cloroform-acetone (90:10) and are detected using bidirectional thin-layer chromatography (TLC) with aluminium silica gel foil. The mean recovery of aflatoxins B1, Bz, G~, and Gz in corn samples was 73, 78, 80, and 64%, respectively; the limit of detection was 0.5 gg/kg. The results can also be confirmed by derivative formation using trifluoroacetic acid on the TLC plate. The method has been applied to a wide range of foods with good results.
Correspondence to: P. Majerus
1 Introduction Aflatoxins are highly toxic, carcinogenic compounds. They are produced by the fungal species Aspergillus flavus and Aspergillus parasiticus and are found in many food supplies, grown in many areas of the world [14]. Interest in the development of rapid, sensitive assays for the determination of aflatoxins has been steadily increasing, because these compounds naturally occur in a wide variety of food and feedstuffs. The CB or AOAC official first action method [5] for the determination of aflatoxins in grains is time consuming and requires the use of several hundred millilitres of especially harmful solvents per sample. Different authors have reported cheap and rapid methods to determine aflatoxins, based on solid phase column extraction, high-pressure thin-layer chromatography (HPTLC) separation and quantification [6, 7]. However, most of the ready-made columns are either unavailable or too expensive for many laboratories in Third World countries. For this reason, the authors have developed a self-packed column by drying and deactivating silica gel 60. To reduce the costs of the method, they have also used normal aluminium silica gel foils in the development of bidirectional TLC, instead of the more expensive H P T L C foils.
2 Experimental procedures
2.1 Apparatus The followingapparatus were used in this study: a laboratory mixer; a mechanical shaker; a laboratory mill; a refrigerated centrifuge,capable of 4000 rpm; centrifuge bottles (250 ml) with screw caps; a rotary vacuum evaporator; a vacuum apparatus (for example Baker SPE-10 extraction system); chromatographic tubes (for example polypropylene columns, 90 x 25 mm and 65 x 10 mm with connector); an oven; a thermoplate; a long-wavelength(360 nm) ultraviolet lamp; a spectrophotometer; a development chamber; an automatic applicator for thin layer plates, with a 100 Ixlsyringe;a fluorodensitometer with an excitation wavelength of 366 nm and an emission wavelength of 429 rim, equipped with a recorder; a 250-ml-shaking
317 funnel; a 250-ml-Erlenmeyer flask; a 10-ml-conical flask; a 100-mlround bottom flask; a 100-ml-measuring cylinder; fluted filter paper (diameter 125 mm); a 100 mm i.d. funnel and a 1 ml vial.
2.2 Reagents The following reagents were used in this study: Methanol; hexane; anhydrous sodium sulphate; 10% sodium chloride solution; silica gel deactivated, in which silica gel 60 with a particle size of 0.0630.200 mm (Merck No. 7734) is first activated at 105~ C for 1 h, then shaken with 1% water (w/w) for 1 h and stored in an exsiccator; chloroform; methylene dichloride; anhydrous diethyl ether; acetone; aluminium foil silica gel 60 (20 • 60 cm, 0.25 mm thickness) without fluorescence indicator (Merck No. 5553) or plastic foil silica gel 60 (Merck No. 5748); 25% sulphuric acid; trifluoroacetic acid (TFA); solvents for TLC included diethyl ether (saturated chamber) and chloroform/acetone (90+10) (saturated chamber). For the preparation and storage of the aflatoxin standard (1 ~tg/ml), refer to sections 26.004-26.009 in [5].
2.3 Extraction and partition Depending on their consistency the samples were mixed in a laboratory mill or mixer. Twenty five grams of the prepared sample were weighed into a glass-stoppered Erlenmeyer flask or a 250-rnl centrifuge bottle with a screw cap. Methanol/water (85 + 15) (100 ml) was added and shaken for 30 rain. The mixture was filtered through medium paper or centrifuged in a refrigerated centrifuge with 4000 rpm, in order to separate the phases. A 50~ml sample of filtrate was collected in a 100-ml graduated cylinder and transfered in a 250ml separatory funnel. To this mixture 50 m110% NaC1 solution was added, which was mixed and followed by the addition of 25 ml hexane. After shaking for 1 min, the phases were separated and the lower phase was drained into a second 250 ml separatory funnel. The upper phase was discarded. The aflatoxins were extracted from aqueous phase with two 25 ml portions of methylene dichloride and shaken for 1 rain. The methylene dichloride phases were drained through filter paper, containing about 5 g anhydrous sodium sulphate, into a 100-ml round-bottom flask. The methylene dichloride extract was evaporated to dryness with a rotary evaporator (40~ C).
2.4 Silica gel column chromatography To 65 • 10 mm polypropylene column, containing a filter disc, 0.5 g anhydrous sodium sulphate was added. The sodium sulphate layer was compressed with a glass rod and 0.5 g deactivated silica gel was added, followed by compression and the addition of 0.5 g anhydrous sodium sulphate. The column was attached to the vacuum apparatus and washed with 3 ml hexane and 3 ml methylene dichloride
using vacuum (flow rate 6 ml/min). The sample extract was dissolved in 3 ml methylene dichloride, added to the column and allowed to drip freely (flow rate 3 ml/min, apply vacuum if needed). From this point on the column was never run dry. The round-bottom flask was rinsed with two 1-ml portions of methylene dichloride, which were then added to the column. The column was washed with 3 ml hexane, 3 ml anhydrous diethyl ether and 3 ml methylene dichloride (using vacuum, flow rate 6 ml/min). The vacuum was turned off, the extraction system cover was removed and 10 ml conical vials were placed under the column. The aflatoxins were eluted without vacuum with two 3-ml portions of chloroform/acetone (90 + 10). The eluate was evaporated to dryness with a rotary evaporator and the sample extract was dissolved with 0.5 ml methylene dichloride for TLC.
2.5 Thin layer chromatography The silica gel foil was marked with a pencil before spotting and was activated on a thermoplate or in an oven at 130 ~ C for 30 min [8]. Forty microlitres of sample extract were spotted alongside 10, 20 and 40 ~tl aflatoxin standard (1 ~tg/ml) 130 mm from the bottom of the aluminium silica gel foil (Fig. 1). The chromatogram was developed in the first direction with diethyl ether. The diethyl ether was evaporated and the plate was examined under an ultraviolet (UV) lamp. The top portion (about 55 mm) of the plate containing the interfering compounds was cut off. The plate was turned 180~ and developed in chloroform/acetone (90 + 10). Care was taken when putting the foil in the chamber to ensure that the aflatoxin spots were not eluted into the solvent. After the second development, the plate was dried and examined under an UV lamp.
2.5.1 Quantitative TLC. Densitometric quantification. A densitometer can be used to scan the sample and aflatoxin reference spots using fluorescence mode aid, excitation and emission wavelengths of 365 and 430 nm respectively. For each aflatoxin in the sample, the area or the height from the fluorescence chromatogram obtained from the sample solution can be calculated and the corresponding mass of aflatoxin can be read off from the graph. The content by mass (w) of each aflatoxin in Ilg/kg of sample is given by:
m, .V~ .V~ W=
mo. V~. V, '
where ml is the mass of aflatoxin read off from the area of the height from the graph in nanograms; mo is the sample mass in grams (25 g); V1 is the total volume of the sample solution, in microlitres (500 Ixl); V2 is the aliquot part of V~ in mierolitres applied to the TLC plate (40 I~1); V3 is the total volume of extraction solvent in millilitres (100 ml); V, is the volume of the filtrate in millilitres (50 ml). Substitution of the values given for too, Va, V2, V3, and 114 reduces the equation to: w=m~ (Ixg/kg).
2.6 Confirmation of aflatoxins B 1 and Ga
T culling tine
tD
1 sample and standord spots
Fig. 1. Spotting pattern for bidirectional TLC
Occasionally, suspected blue and/or green fluorescent spots with the same Rf value as the aflatoxins appeared, which could easily be mistaken for aflatoxins. Therefore, chemical confirmation of the identity of the toxin from suspected positive samples was essential. Sample extract and aflatoxins B 1 and G1 standards were spotted in 20 x 15 cm aluminium silica gel foil as shown in Fig. 2. The foil was developed in the first direction with anhydrous diethyl ether until the end of the foil. The foil was then dried for 5 rain and examined under an UV lamp. The top portion (about 4.5 cm) of the foil containing the interferences was cut. A line 10 cm from the sample spot was scribed as a solvent stop, as shown in Fig. 2. The foil was developed in the second direction with chloroform/acetone (90 + 10) until the solvent reached the stop line. The foil was dried and the site of
318 Table 1. Recovery of aflatoxins added to a sample of yellow corn
2.
Y 8
i ! I
i J ! I
iC D
A
~-45q
i
1--100
3. Fig. 2. Spotting pattern for TLC confirmation. (A) Sample extract; (B, C, D) standard
I 40q
Aflatoxin
Amount added to the sample ng/g
Amount recovered from the sample ng/g
Average recovery from the samplea ng/g
Recovery from the sample %
B1
0.5 1.0 2.0 5.0
0.30-0.48 0.64-0.80 1.46-1.75 3.40-3.60
0.35 (2) 0.7 (2) 1.5 (3) 3.7 (3)
70 70 77 75
B2
0.5 1.0 2.0 5.0
0.38 0.83-0.86 1.41-1.51 3.65-4.44
0.38 (1) 0.84 (2) 1.45 (3) 4.0 (3)
77 84 73 80
G1
0.5 1.0 2.0 5.0
0.4 0.8143.90 1.44-1.78 3.55-4.37
0.4 (1) 0.85 (2) 1.55 (3) 3.9 (3)
80 85 78 78
G2
0.5 1.0 2.0 5.0
0.3 0.66-0.73 1.31-1.36 3.1-3.35
0.3 (1) 0.65 (2) 1.32 (3) 3.2 (3)
60 65 66 64
1. the suspected spots of the sample extract and the aflatoxin standard C were located. The position of aflatoxin standard B was used as a guide to locate the suspected spots. Two microlitres of TFA was added to each of these spots and standard C and allowed to stand for 5 min at room temperature. The foil was then dried for 2 min with a hair-dryer and allowed to cool. A line about 11 cm from the sample extract, as shown in the figure, was scribed. The foil was developed in a third direction in chloroform/acetone (90 + 10) twice. The foil was dried and then examined under an UV lamp. The identity of aflatoxins B1 and G 1 with this method was confirmed by the development of new blue (aflatoxin B2a) and/or green (aflatoxin G2a) fluorescent spots in the sample extract at Rf position equal to those of aflatoxin B2, and/or G2a of aflatoxin standard C. While in false positive samples, the suspected spots have Rf values equal to those of aflatoxin standard D. Using this confirmatory method, the development of aflatoxins B2, and/or G2, in the positive sample extracts was not disturbed by interfering compounds.
a Value in parentheses is the number of analyses
Table 2. Comparison of average recoveries of aflatoxins from foods obtained using the proposed method and other official methods with samples from International Agency for Research on Cancer (IARC) International Check Sample Surveys 1989 and 1991 Sample
Ariatoxin
Overall recovery with other methods (I-tg/kg)
CB[5] BF[~5]
G2
89 4.3 4.3 n.d.
109 4.4 4.0 n.d.
110 64 4.4 2.0 5.0 2.0 n.d. n.d.
Yellow Bt corn meal Be 91-1 []0] G~ Gz
103 6.6 n.d. n.d.
83 5.6 n.d. n.d.
78 64 6.0 5.4 n.d. n.d. n.d. n.d.
26 6.0 21 5.2
19 4.0 10 3.2
16 20 3.7 3.9 9.7 9.9 3.4 3.4
3 Results and discussion
A recovery study using this m e t h o d was carried o u t with aflatoxin free yellow c o r n meal spiked with 0.5, 1, 2, a n d 5 gg/kg aflatoxin B1, B2, G1, a n d G2. Table 1 shows that recoveries o f the f o u r aflatoxins were reasonably u n i f o r m ( 6 4 - 8 5 % ) with acceptable s t a n d a r d deviations (SD) a n d coefficients o f variation (CV) at a level o f 1 g g / k g (for aflatoxin B1, B2, G1, a n d G2, S D = 0 . 0 3 2 5 , 0.084, 0.117, and 0.08 and C V = 4 . 7 , 13, 19.5, a n d 13.5% respectively). The limit o f detection was 0.5 gg/kg, the limit o f determin a t i o n 1 gg/kg. Table 2 shows the average recovery values o f aflatoxins f r o m p e a n u t butter, yellow c o r n a n d c o t t o n seed meal spiked with these c o m p o u n d s o f the W o r l d H e a l t h O r g a n i z a t i o n collaborative studies f r o m 1989 and 1991, using the p r o p o s e d m e t h o d or those described in [9, 10]. As feedstuffs contain a variety o f ingredients, c o l u m n c h r o m a t o g r a p h y clean up is n o t effective in r e m o v i n g all the interfering c o m p o u n d s that are eluted with aflatoxins. U s i n g bidirectional T L C with silica gel aluminium foil, the detection and quantification o f aflatoxins f r o m these interferences can be improved. A yellow fluorescent tailing interference c h r o m a t o g r a p h e d at the same Rf as aflatoxin G2 which was reported by Trucksess et al. [7] was n o t observed with this m e t h o d . T h e limits o f detection o f this m e t h o d were 0.5 ng aflatoxin/g. This value is below b o t h the current and p r o p o s e d legislative limits for aflatoxins in m o s t foods a n d feedstuffs [11].
Peanutbutter 89-1 [9]
Peanutbutter 91-2 [10]
B~ B2
G1
B~ B2
G1 G2
Average recovery using the proposed method (l-tg/kg)
(~g/kg)a
n.d., not detected " Recovery of aflatoxins using the method described in [5, 15]
A l t h o u g h m o r e sensitive and rapid m e t h o d s for aflatoxin analyses in foods a n d feedstuffs have been rep o r t e d [7, 12, 13], these m e t h o d s m o s t l y d e m a n d highly priced instruments, chemicals that are n o t available for all Third W o r l d laboratories or considerable experience a n d skill using highly sophisticated apparatus. F o r screening purposes, the use o f a T L C scanner is n o t absolutely necessary, because a semiquantitative visual evaluation is n o r m a l l y sufficient. The use o f a laborafory-prepared p o l y p r o p y l e n e c o l u m n with 0.5 g silica gel 60 a n d
319 normal aluminium silica gel foil will reduce the cost of analysis and improve the recovery of aflatoxins. The costs for analysis with the proposed method in comparison to those o f CB [5], Dell et al. [14] and Trucksess et al. [7] are a b o u t a quarter and a half respectively, without considering personnel costs. On the other hand, the reduction of critical solvents is becoming an important factor in environmental analysis, because of safety regulations and waste disposal costs. Different batches o f clean-up columns, as well as long storage times, can sometimes influence recovery and reproducibility. Self-made columns give a certain independence from this kind of problem. However, they m a y be replaced by readymade columns in those laboratories where large numbers of samples require automation. When the samples are not fatty, methylene dichloride m a y be used directly for extraction, saving in this way the laborious liquid partition step. This method has been successfully tested with a wide range of foodstuffs, i.e., hazelnuts, pistacchios, peanuts, peanut butter, almonds, corn and spices.
References 1. Busby WF, Wogan GN (1985) In: Searle CE (ed) Aflatoxins and chemical carcinogens, 2nd edn. American Chemical Society, Washington DC, pp 945-1136 2. Butler WH (1974) In: Purchase IFH (ed) Mycotoxins. Elsevier, New York, pp 1-28 3. Chu FS (1971) Adv Appl Microbio122:83-143 4. Zhu JQ, Zhang LS, Hu X, Chen JS, Xu YC, Fremy J, Chu FS (1987) Cancer Res 47:1848-1852 5. Official Methods Of Analysis (1980) 13th edn. AOAC Arlington, VA (secs 26,0,26-26,031)
6. Tomlins KI, Jewers K, Coker RD, Nagler MJ (1988) A bidirectional HPLC development method for the detection of low levels of aflatoxin in maize extracts. Chromatographia 27:4952 7. Trucksess MW, William CB, Stanley N (1984) Rapid quantitation and confirmation of aflatoxins in corn and peanut butter, using a disposable silica gel column, thin layer chromatography/mass spectrometry. J Assoc Off Anal Chem 67:973-975 8. Tripet FY, Riva C, Vogel J (1981) Recherche des aflatoxines et dosage de l'aflatoxine M 1 dans les produits laitiers. Trav Chim Aliment Hyg 72:367-379 9. Friesen M (1989) Aflatoxin check sample survey programme. Preliminary report on statistical analysis of results obtained for the analysis of aflatoxins B1, B2, Gx, and G 2. International Agency for Research on Cancer, Lyon, France 10. Friesen M (1991) Aflatoxin check sample survey programme. Preliminary report on statistical analysis of results obtained for the analysis of aflatoxins B~, B2, G1, and G 2. International Agency for Research on Cancer, Lyon, France 11. Van Egmond HP (1989) Current situation on regulations for mycotoxins. Overview of tolerances and status of standard methods of sampling and analysis. Food Add Cont 6:139-188 12. Patey AL, Sharman M, Gilbert J (1991) Liquid chromatographic determination of aflatoxin levels in peanut butters using immunoaffinity column cleanup method: international collaborative trial. J Assoc Off Anal Chem 74:76-81 13. Trucksess MW, Stack ME, Nesheim S, Page W, Albert RH, Hansen TJ, Donahue KF (1991) Immunoaffinity column coupled with solution fluorometry or liquid chromatography postcolumn derivatization for determination of aflatoxins in corn, peanuts and peanut butter: collaborative study. J Assoc Off Anal Chem 74:81-88 14. Dell MPK, Haswell SJ, Roch OG, Coker RD, Medlock VFP, Tomlins K (1990) Analytical methodology for the determination of aflatoxins in peanut butter: comparison of high-performance thin-layer chromatographic, enzyme-linked immunosorbent assay and high-performance liquid chromatographic methods. Analyst 115:1435-1439 15. Waltking AF, Bleffert G, Kiernan M (1969) An improved rapid physicochemical assay method for aflatoxin in peanuts and peanut products. J Am Oil Chem Soc 45:880-884
Zeitschrift for
9 Springer-Verlag 1992
Original paper A rapid, sensitive and economic method for the detection, quantification and confirmation of aflatoxins Paul Majerus 1 and Zaki Zakaria 2 1 ChemischesUntersuchungsamt Trier, Maximineracht l 1a, W-5500 Trier, Federal Republic of Germany 2 Assiut University, Faculty of Veterinary Medicine, Veterinary Medicine - Forensic Medicine and Toxicology,Assiut, Egypt Received May 19, 1992
Eine schnelle, empfindUche und kostengiinstige Methode zum Nachweis, zur Bestimmung und Bestfitigung von Aflatoxinen Zusammenfassung. Es wird eine schnelle, empfindliche und kostengfinstige Methode zum Nachweis, Bestimmung und Best/itigung von Aflatoxinen beschrieben. Die Aflatoxine B1, B2, G1 und G2 werden mit Methanol/ Wasser (85 + 15) extrahiert und in Dichlormethan iiberffihrt. Der Dichlormethanextrakt wird auf einer mit 0,5 g Kieselgel 60 geffillten Polypropylens/iule gereinigt. Die Aflatoxine werden mit Chloroform/Aceton (90+10) eluiert und mit zweidimensionaler DC auf Kieselgel-Alufolien nachgewiesen. Die mittleren Wiederfindungsraten ffir die Aflatoxine B~, B2, G~ und G 2 in Maismehl betragen 73, 78, 80 und 64%, die Nachweisgrenzen liegen durchschnittlich bei 0,5 gg/kg. Zur Bestfitigung verdfichtiger Befunde kann auf der Platte mit Trifluoressigs/iure derivatisiert werden. Die Methode ist bisher an einer Vielzahl von verschiedenen Lebensmitteln mit gutem Erfotg getestet worden. Summary. A rapid, sensitive and economic method for the detection, quantification and confirmation of aflatoxins is described. Aflatoxins B~, B2, G1, and G2, are extracted by methanol/water (85+ 15) and partitioned into methylene dichloride. The methylene dichloride solution is cleaned up on a polypropylene column, filled with 0.5 g silica gel 60. The aflatoxins are eluted with cloroform-acetone (90:10) and are detected using bidirectional thin-layer chromatography (TLC) with aluminium silica gel foil. The mean recovery of aflatoxins B1, Bz, G~, and Gz in corn samples was 73, 78, 80, and 64%, respectively; the limit of detection was 0.5 gg/kg. The results can also be confirmed by derivative formation using trifluoroacetic acid on the TLC plate. The method has been applied to a wide range of foods with good results.
Correspondence to: P. Majerus
1 Introduction Aflatoxins are highly toxic, carcinogenic compounds. They are produced by the fungal species Aspergillus flavus and Aspergillus parasiticus and are found in many food supplies, grown in many areas of the world [14]. Interest in the development of rapid, sensitive assays for the determination of aflatoxins has been steadily increasing, because these compounds naturally occur in a wide variety of food and feedstuffs. The CB or AOAC official first action method [5] for the determination of aflatoxins in grains is time consuming and requires the use of several hundred millilitres of especially harmful solvents per sample. Different authors have reported cheap and rapid methods to determine aflatoxins, based on solid phase column extraction, high-pressure thin-layer chromatography (HPTLC) separation and quantification [6, 7]. However, most of the ready-made columns are either unavailable or too expensive for many laboratories in Third World countries. For this reason, the authors have developed a self-packed column by drying and deactivating silica gel 60. To reduce the costs of the method, they have also used normal aluminium silica gel foils in the development of bidirectional TLC, instead of the more expensive H P T L C foils.
2 Experimental procedures
2.1 Apparatus The followingapparatus were used in this study: a laboratory mixer; a mechanical shaker; a laboratory mill; a refrigerated centrifuge,capable of 4000 rpm; centrifuge bottles (250 ml) with screw caps; a rotary vacuum evaporator; a vacuum apparatus (for example Baker SPE-10 extraction system); chromatographic tubes (for example polypropylene columns, 90 x 25 mm and 65 x 10 mm with connector); an oven; a thermoplate; a long-wavelength(360 nm) ultraviolet lamp; a spectrophotometer; a development chamber; an automatic applicator for thin layer plates, with a 100 Ixlsyringe;a fluorodensitometer with an excitation wavelength of 366 nm and an emission wavelength of 429 rim, equipped with a recorder; a 250-ml-shaking
317 funnel; a 250-ml-Erlenmeyer flask; a 10-ml-conical flask; a 100-mlround bottom flask; a 100-ml-measuring cylinder; fluted filter paper (diameter 125 mm); a 100 mm i.d. funnel and a 1 ml vial.
2.2 Reagents The following reagents were used in this study: Methanol; hexane; anhydrous sodium sulphate; 10% sodium chloride solution; silica gel deactivated, in which silica gel 60 with a particle size of 0.0630.200 mm (Merck No. 7734) is first activated at 105~ C for 1 h, then shaken with 1% water (w/w) for 1 h and stored in an exsiccator; chloroform; methylene dichloride; anhydrous diethyl ether; acetone; aluminium foil silica gel 60 (20 • 60 cm, 0.25 mm thickness) without fluorescence indicator (Merck No. 5553) or plastic foil silica gel 60 (Merck No. 5748); 25% sulphuric acid; trifluoroacetic acid (TFA); solvents for TLC included diethyl ether (saturated chamber) and chloroform/acetone (90+10) (saturated chamber). For the preparation and storage of the aflatoxin standard (1 ~tg/ml), refer to sections 26.004-26.009 in [5].
2.3 Extraction and partition Depending on their consistency the samples were mixed in a laboratory mill or mixer. Twenty five grams of the prepared sample were weighed into a glass-stoppered Erlenmeyer flask or a 250-rnl centrifuge bottle with a screw cap. Methanol/water (85 + 15) (100 ml) was added and shaken for 30 rain. The mixture was filtered through medium paper or centrifuged in a refrigerated centrifuge with 4000 rpm, in order to separate the phases. A 50~ml sample of filtrate was collected in a 100-ml graduated cylinder and transfered in a 250ml separatory funnel. To this mixture 50 m110% NaC1 solution was added, which was mixed and followed by the addition of 25 ml hexane. After shaking for 1 min, the phases were separated and the lower phase was drained into a second 250 ml separatory funnel. The upper phase was discarded. The aflatoxins were extracted from aqueous phase with two 25 ml portions of methylene dichloride and shaken for 1 rain. The methylene dichloride phases were drained through filter paper, containing about 5 g anhydrous sodium sulphate, into a 100-ml round-bottom flask. The methylene dichloride extract was evaporated to dryness with a rotary evaporator (40~ C).
2.4 Silica gel column chromatography To 65 • 10 mm polypropylene column, containing a filter disc, 0.5 g anhydrous sodium sulphate was added. The sodium sulphate layer was compressed with a glass rod and 0.5 g deactivated silica gel was added, followed by compression and the addition of 0.5 g anhydrous sodium sulphate. The column was attached to the vacuum apparatus and washed with 3 ml hexane and 3 ml methylene dichloride
using vacuum (flow rate 6 ml/min). The sample extract was dissolved in 3 ml methylene dichloride, added to the column and allowed to drip freely (flow rate 3 ml/min, apply vacuum if needed). From this point on the column was never run dry. The round-bottom flask was rinsed with two 1-ml portions of methylene dichloride, which were then added to the column. The column was washed with 3 ml hexane, 3 ml anhydrous diethyl ether and 3 ml methylene dichloride (using vacuum, flow rate 6 ml/min). The vacuum was turned off, the extraction system cover was removed and 10 ml conical vials were placed under the column. The aflatoxins were eluted without vacuum with two 3-ml portions of chloroform/acetone (90 + 10). The eluate was evaporated to dryness with a rotary evaporator and the sample extract was dissolved with 0.5 ml methylene dichloride for TLC.
2.5 Thin layer chromatography The silica gel foil was marked with a pencil before spotting and was activated on a thermoplate or in an oven at 130 ~ C for 30 min [8]. Forty microlitres of sample extract were spotted alongside 10, 20 and 40 ~tl aflatoxin standard (1 ~tg/ml) 130 mm from the bottom of the aluminium silica gel foil (Fig. 1). The chromatogram was developed in the first direction with diethyl ether. The diethyl ether was evaporated and the plate was examined under an ultraviolet (UV) lamp. The top portion (about 55 mm) of the plate containing the interfering compounds was cut off. The plate was turned 180~ and developed in chloroform/acetone (90 + 10). Care was taken when putting the foil in the chamber to ensure that the aflatoxin spots were not eluted into the solvent. After the second development, the plate was dried and examined under an UV lamp.
2.5.1 Quantitative TLC. Densitometric quantification. A densitometer can be used to scan the sample and aflatoxin reference spots using fluorescence mode aid, excitation and emission wavelengths of 365 and 430 nm respectively. For each aflatoxin in the sample, the area or the height from the fluorescence chromatogram obtained from the sample solution can be calculated and the corresponding mass of aflatoxin can be read off from the graph. The content by mass (w) of each aflatoxin in Ilg/kg of sample is given by:
m, .V~ .V~ W=
mo. V~. V, '
where ml is the mass of aflatoxin read off from the area of the height from the graph in nanograms; mo is the sample mass in grams (25 g); V1 is the total volume of the sample solution, in microlitres (500 Ixl); V2 is the aliquot part of V~ in mierolitres applied to the TLC plate (40 I~1); V3 is the total volume of extraction solvent in millilitres (100 ml); V, is the volume of the filtrate in millilitres (50 ml). Substitution of the values given for too, Va, V2, V3, and 114 reduces the equation to: w=m~ (Ixg/kg).
2.6 Confirmation of aflatoxins B 1 and Ga
T culling tine
tD
1 sample and standord spots
Fig. 1. Spotting pattern for bidirectional TLC
Occasionally, suspected blue and/or green fluorescent spots with the same Rf value as the aflatoxins appeared, which could easily be mistaken for aflatoxins. Therefore, chemical confirmation of the identity of the toxin from suspected positive samples was essential. Sample extract and aflatoxins B 1 and G1 standards were spotted in 20 x 15 cm aluminium silica gel foil as shown in Fig. 2. The foil was developed in the first direction with anhydrous diethyl ether until the end of the foil. The foil was then dried for 5 rain and examined under an UV lamp. The top portion (about 4.5 cm) of the foil containing the interferences was cut. A line 10 cm from the sample spot was scribed as a solvent stop, as shown in Fig. 2. The foil was developed in the second direction with chloroform/acetone (90 + 10) until the solvent reached the stop line. The foil was dried and the site of
318 Table 1. Recovery of aflatoxins added to a sample of yellow corn
2.
Y 8
i ! I
i J ! I
iC D
A
~-45q
i
1--100
3. Fig. 2. Spotting pattern for TLC confirmation. (A) Sample extract; (B, C, D) standard
I 40q
Aflatoxin
Amount added to the sample ng/g
Amount recovered from the sample ng/g
Average recovery from the samplea ng/g
Recovery from the sample %
B1
0.5 1.0 2.0 5.0
0.30-0.48 0.64-0.80 1.46-1.75 3.40-3.60
0.35 (2) 0.7 (2) 1.5 (3) 3.7 (3)
70 70 77 75
B2
0.5 1.0 2.0 5.0
0.38 0.83-0.86 1.41-1.51 3.65-4.44
0.38 (1) 0.84 (2) 1.45 (3) 4.0 (3)
77 84 73 80
G1
0.5 1.0 2.0 5.0
0.4 0.8143.90 1.44-1.78 3.55-4.37
0.4 (1) 0.85 (2) 1.55 (3) 3.9 (3)
80 85 78 78
G2
0.5 1.0 2.0 5.0
0.3 0.66-0.73 1.31-1.36 3.1-3.35
0.3 (1) 0.65 (2) 1.32 (3) 3.2 (3)
60 65 66 64
1. the suspected spots of the sample extract and the aflatoxin standard C were located. The position of aflatoxin standard B was used as a guide to locate the suspected spots. Two microlitres of TFA was added to each of these spots and standard C and allowed to stand for 5 min at room temperature. The foil was then dried for 2 min with a hair-dryer and allowed to cool. A line about 11 cm from the sample extract, as shown in the figure, was scribed. The foil was developed in a third direction in chloroform/acetone (90 + 10) twice. The foil was dried and then examined under an UV lamp. The identity of aflatoxins B1 and G 1 with this method was confirmed by the development of new blue (aflatoxin B2a) and/or green (aflatoxin G2a) fluorescent spots in the sample extract at Rf position equal to those of aflatoxin B2, and/or G2a of aflatoxin standard C. While in false positive samples, the suspected spots have Rf values equal to those of aflatoxin standard D. Using this confirmatory method, the development of aflatoxins B2, and/or G2, in the positive sample extracts was not disturbed by interfering compounds.
a Value in parentheses is the number of analyses
Table 2. Comparison of average recoveries of aflatoxins from foods obtained using the proposed method and other official methods with samples from International Agency for Research on Cancer (IARC) International Check Sample Surveys 1989 and 1991 Sample
Ariatoxin
Overall recovery with other methods (I-tg/kg)
CB[5] BF[~5]
G2
89 4.3 4.3 n.d.
109 4.4 4.0 n.d.
110 64 4.4 2.0 5.0 2.0 n.d. n.d.
Yellow Bt corn meal Be 91-1 []0] G~ Gz
103 6.6 n.d. n.d.
83 5.6 n.d. n.d.
78 64 6.0 5.4 n.d. n.d. n.d. n.d.
26 6.0 21 5.2
19 4.0 10 3.2
16 20 3.7 3.9 9.7 9.9 3.4 3.4
3 Results and discussion
A recovery study using this m e t h o d was carried o u t with aflatoxin free yellow c o r n meal spiked with 0.5, 1, 2, a n d 5 gg/kg aflatoxin B1, B2, G1, a n d G2. Table 1 shows that recoveries o f the f o u r aflatoxins were reasonably u n i f o r m ( 6 4 - 8 5 % ) with acceptable s t a n d a r d deviations (SD) a n d coefficients o f variation (CV) at a level o f 1 g g / k g (for aflatoxin B1, B2, G1, a n d G2, S D = 0 . 0 3 2 5 , 0.084, 0.117, and 0.08 and C V = 4 . 7 , 13, 19.5, a n d 13.5% respectively). The limit o f detection was 0.5 gg/kg, the limit o f determin a t i o n 1 gg/kg. Table 2 shows the average recovery values o f aflatoxins f r o m p e a n u t butter, yellow c o r n a n d c o t t o n seed meal spiked with these c o m p o u n d s o f the W o r l d H e a l t h O r g a n i z a t i o n collaborative studies f r o m 1989 and 1991, using the p r o p o s e d m e t h o d or those described in [9, 10]. As feedstuffs contain a variety o f ingredients, c o l u m n c h r o m a t o g r a p h y clean up is n o t effective in r e m o v i n g all the interfering c o m p o u n d s that are eluted with aflatoxins. U s i n g bidirectional T L C with silica gel aluminium foil, the detection and quantification o f aflatoxins f r o m these interferences can be improved. A yellow fluorescent tailing interference c h r o m a t o g r a p h e d at the same Rf as aflatoxin G2 which was reported by Trucksess et al. [7] was n o t observed with this m e t h o d . T h e limits o f detection o f this m e t h o d were 0.5 ng aflatoxin/g. This value is below b o t h the current and p r o p o s e d legislative limits for aflatoxins in m o s t foods a n d feedstuffs [11].
Peanutbutter 89-1 [9]
Peanutbutter 91-2 [10]
B~ B2
G1
B~ B2
G1 G2
Average recovery using the proposed method (l-tg/kg)
(~g/kg)a
n.d., not detected " Recovery of aflatoxins using the method described in [5, 15]
A l t h o u g h m o r e sensitive and rapid m e t h o d s for aflatoxin analyses in foods a n d feedstuffs have been rep o r t e d [7, 12, 13], these m e t h o d s m o s t l y d e m a n d highly priced instruments, chemicals that are n o t available for all Third W o r l d laboratories or considerable experience a n d skill using highly sophisticated apparatus. F o r screening purposes, the use o f a T L C scanner is n o t absolutely necessary, because a semiquantitative visual evaluation is n o r m a l l y sufficient. The use o f a laborafory-prepared p o l y p r o p y l e n e c o l u m n with 0.5 g silica gel 60 a n d
319 normal aluminium silica gel foil will reduce the cost of analysis and improve the recovery of aflatoxins. The costs for analysis with the proposed method in comparison to those o f CB [5], Dell et al. [14] and Trucksess et al. [7] are a b o u t a quarter and a half respectively, without considering personnel costs. On the other hand, the reduction of critical solvents is becoming an important factor in environmental analysis, because of safety regulations and waste disposal costs. Different batches o f clean-up columns, as well as long storage times, can sometimes influence recovery and reproducibility. Self-made columns give a certain independence from this kind of problem. However, they m a y be replaced by readymade columns in those laboratories where large numbers of samples require automation. When the samples are not fatty, methylene dichloride m a y be used directly for extraction, saving in this way the laborious liquid partition step. This method has been successfully tested with a wide range of foodstuffs, i.e., hazelnuts, pistacchios, peanuts, peanut butter, almonds, corn and spices.
References 1. Busby WF, Wogan GN (1985) In: Searle CE (ed) Aflatoxins and chemical carcinogens, 2nd edn. American Chemical Society, Washington DC, pp 945-1136 2. Butler WH (1974) In: Purchase IFH (ed) Mycotoxins. Elsevier, New York, pp 1-28 3. Chu FS (1971) Adv Appl Microbio122:83-143 4. Zhu JQ, Zhang LS, Hu X, Chen JS, Xu YC, Fremy J, Chu FS (1987) Cancer Res 47:1848-1852 5. Official Methods Of Analysis (1980) 13th edn. AOAC Arlington, VA (secs 26,0,26-26,031)
6. Tomlins KI, Jewers K, Coker RD, Nagler MJ (1988) A bidirectional HPLC development method for the detection of low levels of aflatoxin in maize extracts. Chromatographia 27:4952 7. Trucksess MW, William CB, Stanley N (1984) Rapid quantitation and confirmation of aflatoxins in corn and peanut butter, using a disposable silica gel column, thin layer chromatography/mass spectrometry. J Assoc Off Anal Chem 67:973-975 8. Tripet FY, Riva C, Vogel J (1981) Recherche des aflatoxines et dosage de l'aflatoxine M 1 dans les produits laitiers. Trav Chim Aliment Hyg 72:367-379 9. Friesen M (1989) Aflatoxin check sample survey programme. Preliminary report on statistical analysis of results obtained for the analysis of aflatoxins B1, B2, Gx, and G 2. International Agency for Research on Cancer, Lyon, France 10. Friesen M (1991) Aflatoxin check sample survey programme. Preliminary report on statistical analysis of results obtained for the analysis of aflatoxins B~, B2, G1, and G 2. International Agency for Research on Cancer, Lyon, France 11. Van Egmond HP (1989) Current situation on regulations for mycotoxins. Overview of tolerances and status of standard methods of sampling and analysis. Food Add Cont 6:139-188 12. Patey AL, Sharman M, Gilbert J (1991) Liquid chromatographic determination of aflatoxin levels in peanut butters using immunoaffinity column cleanup method: international collaborative trial. J Assoc Off Anal Chem 74:76-81 13. Trucksess MW, Stack ME, Nesheim S, Page W, Albert RH, Hansen TJ, Donahue KF (1991) Immunoaffinity column coupled with solution fluorometry or liquid chromatography postcolumn derivatization for determination of aflatoxins in corn, peanuts and peanut butter: collaborative study. J Assoc Off Anal Chem 74:81-88 14. Dell MPK, Haswell SJ, Roch OG, Coker RD, Medlock VFP, Tomlins K (1990) Analytical methodology for the determination of aflatoxins in peanut butter: comparison of high-performance thin-layer chromatographic, enzyme-linked immunosorbent assay and high-performance liquid chromatographic methods. Analyst 115:1435-1439 15. Waltking AF, Bleffert G, Kiernan M (1969) An improved rapid physicochemical assay method for aflatoxin in peanuts and peanut products. J Am Oil Chem Soc 45:880-884
