LABORATORY INFORMATION BULLETIN
DFS/ORA No 3972
ISOLATION OF MYCOTOXINS USING EXTRACTING DISK
YUHONG ZHAO, GEORGE M. WARE, SHIA S. KUAN AND
ALLEN S. CARMAN, Food and Drug Administration
Natural Toxins Research Center
Zearalenol and zearalenone are naturally occurring fungal metabolites that contaminate many food and agricultural commodities. Zearalenol and zearalenone along with other mycotoxins can present a health concern at the parts per billion level. The Food and Drug Administration requires analytical methods that are reliable and capable of measuring low levels of these toxins using techniques that minimize solvent use. In addition, the FDA favor analytical procedures that are amenable to automation. Most of the mycotoxin methods currently used by FDA to accomplish the agency's regulatory mission are methods that use liquid-liquid partition, open bed column or solid phase extraction (SPE) cleanup procedures prior to the analyte final determination.
The rapid growth in analytical technology has made, in some instances, analytical methods developed prior to 1980 antiquated (1-3). Current detection systems when compared to 20 years ago are much more sensitive and selective. Such improvements have impacted the requirements for cleanup procedures used in current analytical methods. Conventional analytical methods use large sample sizes that are cleaned up using open bed column chromatography or liquid-liquid extraction. Large sample sizes are needed because of the poor sensitivity associated with older instrumentation. Open-bed chromatography and liquid-liquid partition are excellent cleanup procedures, but required large sample sizes and large volumes of some potentially hazardous solvents.
Modern instrument detection limits are usually in the low nano to pico gram range. With such small quantities needed for actual quantitation, sample size requirements have significantly been reduced. Smaller sample sizes permit the use of a comparable reduction in the size of open-bed column used in the cleanup procedures. Around 1980, a smaller open-bed column, Solid Phase Extraction (SPE), cleanup column was introduced which could accommodate smaller sample size (4). Solid phase extraction reduces the quantity of solvent used per analysis. Recently, the solid phase extraction disks were developed and required much less solvent than the original SPE columns (5). These columns seem ideally suited to replace open bed-column chromatography and liquid-liquid partition. Numerous applications have been published that demonstrates the use of the solid phase extraction disk to separate and enrich target analytes from complex interferences found in environmental (6-8), drug (9-11) and natural product samples (12-14). The solid phase extraction disk seem to have three promising advantages: reduction in the amounts hazardous solvent used in the analysis; shorter analysis time; and amenable to automation.
The present AOAC Official Method for zearalenol and zearalenone use a liquid-liquid extraction procedure consisting of a strong base to extract the weakly acidic zearalenol and zearalenone from an organic solvent (15). The neutral and basic interferences remain in the organic layer and are discarded with the organic solvent. The basic solution is acidified and zearalenol and zearalenone are extracted from the aqueous layer with an organic solvent. The basic-acid extraction is an effective cleanup technique, however, the procedure is time consuming and cannot easily be automated. Another, drawback of the liquid-liquid extraction method is that it uses a large volume of organic solvent. The more effective organic solvents such as, chloroform, benzene, and methylene chloride use have been drastically reduced by most laboratories and will be totally eliminated within the next few years. The reduction in the use of organic solvents is due to the potential health hazards associated with the exposure of laboratory personnel to these solvents. The disposal cost of hazardous organic solvents have become an important concern for laboratory management as well. The rapid increase in the cost of disposing large amounts of hazardous organic solvents have made methods which use conventional cleanup techniques, such as open bed chromatography and liquid partition, cost prohibitive.
The prerequisite for developing techniques which can be automated will continue to have an impact on the design of analytical methods. Automated instrumentation prohibit the use of large separatory funnels and large open-bed chromatographic columns. Instead, small syringe type columns that have a total column volumes about 1 to 5 mL must be used.
These small columns can be packed with a variety of packing material which mimic separations based on liquid-liquid partition, adsorption, ion exchange and immunoaffinity chromatography. Aflatoxins and fumonisins have been isolated from interferences using Aflatest and Fumonitest immunoaffinity columns, respectively (16,17). SPE procedures have been used to isolate mycotoxins from matrix interferences (18-22). The SFE columns are packed with a selectivity packing material, such as silica gel, C-18, or ion exchange material and the sample extract is fractionated by polarity, water solubility or ionic attraction. Recently, a new type of solid phase extraction technique was introduced that has several advantages over immunoaffinity or the conventional SPE column. The SPE disks are unique solid phase extraction devices consisting of glass fibers embedded with various phases of bonded silica. The C-18 SPE disk are less selective than the typical immunoaffinity column, however, SPE disks are inexpensive relative to the immunoaffinity columns.
This Laboratory Information Bulletin describes our research on converting a zearalenol and zearalenone method using a liquid-liquid partition to the SPE disk format. The SPE disk methods use minimal amounts of hazardous organic solvents with the total solvent consumption reduced by about 95 % to that of the conventional cleanup procedure. Research on converting other methods are in progress. Six of the pharmaceutically active ergot alkaloids were isolated from wheat and rye using the SPE disk and a cleanup device and the results will be published later in another journal.
REAGENTS and SOLUTION:
Extraction Solvent -- methanol + water (50 + 50, v/v)
Condition Solvent -- methanol, followed by water
Wash Solvent -- methanol + water (30 + 70, v/v)
Elution Solvent -- methanol + water (60 + 40, v/v)
Standard Solution -- the mixed standard contained both zearalenol and zearalenone at 25, 50, 100, 250 and 500 ng/mL HPLC mobile phase
HPLC Mobile Phase -- methanol + water + acetic acid
(65 + 35 + 1, v/v)
Solid Phase Extraction Microcolumn -- SPE disk 3 mL C-18, adsorbent wt. 15 mg
Vacuum Manifold -- Visiprep™ Solid Phase Extraction vacuum manifold, Supelco, or equivalent
Liquid Chromatograph -- Beckman System Gold Liquid Chromatograph with fluorescent detector.
Detector:Mcpherson FL750 Plus Spectrofluoresecene detector with xenon-mercury lamp, High Sensitivity Accessory (HA) and 20 æL quartz flow cell. Excitation 236 nm and emission filter cutoff 418 nm
Column: C-8, 5 æ particle size, 250 mm x 4.6 mm
Injector: Beckman model 210 sample injection valve with a 20 æL loop.
Mobile Phase: Methanol + water + acetic acid (65 + 35 +1, v/v)
Flow Rate: 1.0 mL/min
Wrist Action Shaker
Filter Paper -- Pre-pleated filter #588 and gelman glass acrodisc®
Shelled corn was ground in a Comitrol Mill with a 2 mm sieve. The ground corn was mixed thoroughly. A 25 g sample was weighed into a 250 mL glass-stopper flask and 100 mL extraction solvent was added. The flask was shaken for 30 min. The extract was filtered through a pre-pleated filter paper to remove solid material. Smaller particulate matter was removed from the sample extract by filtering the extract through a glass fiber filter (Glass Fiber Acrodisc®). The filtrate was collected and saved for SPE disk cleanup.
SPE DISK CLEANUP
The SPE procedure was performed using a vacuum manifold equipped with vacuum control valve that allowed for the concurrent processing of 12 samples (Supelco's Visiprep™). The C-18 SPE disk was conditioned by sequentially passing 0.5 mL MeOH and 0.5 mL water through the disk. A 4.0 mL aliquot of the extraction filtrate was added to 25 mL Erlenmeyer flask and diluted with 6.0 mL distilled water. The diluted sample extract was added to micro column reservoir and the micro column was aspirated with vacuum (<2" Hg). The flask was rinsed with 1.0 mL water. The rinse was added to microcolumn and aspirated. The SPE disk was wash with 0.5 mL MeOH + water (30 + 70). Wash was discarded. SPE disk was dried by pulling a vacuum of 15" Hg for 3 min. Zearalenol and zearalenone were eluted from the SPE disk by gently aspirating (<2" Hg) two 0.5 mL portion MeOH + water (70 + 30). The eluate were collected, mixed and saved for HPLC determination.
Inject 20 æl of each mixed standard solution into the HPLC and prepare standard curves by plotting concentration vs peak area. Check the reproducibility of the standard curves daily by injecting 100 ng/mL of a mixed standard solution.
Inject 20 æl sample eluate into the HPLC under the same conditions used for preparing the standard curve. Identify zearalenol and zearalenone in the sample by comparing retention time with those of the standards. Calculate the concentration of the two mycotoxins in the sample by comparing the areas of the sample and standard using the following formula;
Zearalenol/zearalenone (ng/g)=A x B x C / (D X W)
where, A = peak area of zearalenol or zearalenone in sample
B = the concentration of the standard injected, ng/mL
C = Volume of the eluate, mL
D = Peak area of the standard
W = amount of sample added to the microcolumn, g
RESULTS and DISCUSSION
Our laboratory has developed an analytical protocol that uses the SPE disk to isolate mycotoxins from food extracts. We used the SPE disk as an alternative to liquid/liquid partition and open bed column cleanup procedures. The general scheme for the SPE disk cleanup procedure is to take a sufficient volume of the extract that represents about 1.0 gram of solid sample and dilute the extract with 5.0 -10. mL of a weak solvent. The diluted sample extract is eluted onto the SPE disk where the analyte is extracted from the weak solvent. Weakly adsorbed interferences are separated from the analytes by washing the SPE disk with solvent that will not elute the analytes. The ideal wash solvent is a solvent which has the strength and selectivity to extract the interferences from the disk that contains interferences and analytes. The wash solvents are found experimental and are critical to the overall effectiveness of a cleanup procedure. The analytes are eluted from the SPE disk with a strong solvent (about 0.5 mL). Generally, the analyte enriched sample extract can be injected into a GC or HPLC without further evaporation of the sample extract.
Table 1 shows the amount of zearalenol and zearalenone eluting from the disk as a function of percent methanol in the eluent. As can be seen, zearalenol and zearalenone are efficiently removed from the disk when the methanol composition exceeds 40 %. This experiment was conducted to determine the solvent strength needed to extract and elute zearalenol and zearalenone from a C-18 extraction disk.
Table #2 shows the results obtained from a study to determine the extraction disk capacity. Zearalenol and zearalenone were added in varying amounts (100 ng to 20 æg) to a 10.0 mL solution of MeOH + water (20 + 80) and were added to a C-18 extraction disk. The standard materials were eluted from the extraction disk with the elution solvent MeOH + water (60 + 40). The data indicate that all levels of standard material were efficiently recovered from the extraction disk. Table # 3 shows the effect of sample size on the recovery of zearalenol and zearalenone from a C-18 extraction disk. A constant amount (100 ng) of zearalenol and zearalenone were added to varying amounts of sample extract. Data in Table #3 indicate that the recovery of zearalenol and zearalenone are excellent for samples size ranging from 0.5 -1.25 g of sample extract. The extraction disk was over loaded when more than 2.0 grams of sample extract was applied to the C-18 extraction disk, significantly lowering the recovery of the two mycotoxins.
Table #4 shows data for the recovery of zearalenol and zearalenone added to a control corn sample. Zearalenol and zearalenone were spiked at levels ranging from 2,000 - 10ng/g. The recovery of both toxins from spiked corn samples were excellent. Both toxins were recovered from spiked samples at 106.3 and 103.8% with a percent coefficient of variation 7.6 and 13.0%, respectively.
Figures 1 and 2 show chromatograms of a control corn sample and spiked corn sample (100 ng/g) that were cleaned up with a C-18 SPE disk, respectively. The combination of a selective fluorescence detector and efficient cleanup procedure allows for low levels of zearalenol and zearalenone to be determined. The control sample does not have any interference near the time zearalenol and zearalenone elute.
Research on converting other analytical methods to the SPE disk format is in progress. Success from this project has lead our laboratory to explore the other methods that can be converted. Our goal to adapt most mycotoxin methods to a single format, such as SPE disk, that is amenable to automation.
Table 1 Effect of MeoH/Water on Analyte Retention Pattern
Column No Proportion of MeOH/ Water Analyte Recovery
1 10 + 90 0 %
2 20 + 80 0 %
3 30 + 70 0 %
4 40 + 60 0 %
5 50 + 50 40 %
6 60 + 40 > 90 %
7 70 + 30 > 90 %
Table 2 SPE Microcolumn's Capacity for Standard Material
Amount Added Amount Recovered Recovery (ng) (ng) (%)
Zearalenol Zearalenone Zearalenol Zearalenone
100 97.3 94.4 97.3 94.4
500 471.2 413.8 94.2 82.8
1000 1009.8 906.9 101.0 90.7
5000 5188.6 4863.1 103.8 97.3
5000 4426.6 4722.6 88.5 94.4
8100 8807.7 7272.5 108.7 89.8
10000 9600.6 9066.0 96.0 90.7
20000 19198.0 19273.1 95.4 96.3
mean , Rec., % 98.1 92.0
Table 3 Effect of Sample Size on Column Capacity
Sample Size Amount Added Amount Recovery Recovery (g) (ng) (ng) (%)
zearalenol zearalenone zearalenol zearalenone
0.50 100.0 97.8 98.3 97.8 98.3
0.75 100.0 83.8 87.4 83.8 87.4
1.00 100.0 108.3 107.5 108.3 107.4
1.25 100.0 93.0 84.9 93.0 84.9
2.00 100.0 64.3 62.7 64.3 62.7
2.50 100.0 52.2 42.6 52.2 42.6
Table 4 Recovery of Zearalenol and Zearalenone
Amount Added Amount Recovery Percent Recovery (ng/g corn) (ng/g) (%)
Zearalenol Zearalenone Zearalenol Zearalenone
2000 2233.0 2231.6 116.5 115.8 1964.3 1921.7 98.2 96.1
1000 928.0 892.8 92.8 89.3 1084.8 1065.1 108.5 106.5
200 195.6 196.5 97.8 98.3 221.7 221.0 110.8 110.5
40 42.7 32.3 106.8 80.7 20 22.5 23.5 112.4 117.5
10 11.3 12.0 113.0 120.0
Mean, Rec. % 106.3 103.8
SD 8.1 13.5
CV % 7.6 13.0
1. Eppley, R. M., Stoloff, L., and Campbell, A. D., JAOAC, 1968, 51, 67-73.
2. Waltking, A. E., JAOAC, 1970, 53, 104-113.
3. Ware, G. M. and Thorpe, S. W., JAOAC, 1978, 61, (5), 1058-1062.
4.Wolkoff, A. Wand Creed, C. J., J. Liquid Chromatogr., 1981, 4, 1459-1472.
5.Chiron, S., and Barcelo, D., J. Chromatogr. 1993, 645 (1), 125-134
6. Frebortova, J. and Tatarkovicova, V., Analyst, 119 (7), 1519-1523.
7. Miliadis, G. E., Bull Environ. Contam. Toxicol. 1994, 52 (1), 25-30.
8.Snyder, J. L., Grob, R. L., McNally, M. E. and Oostdyk, T. S., 1994, LC-GC, 12 (3), 230-242.
9.Cone, E. J., Hillsgrove M. and Darwin, W. D., Clin. Chem., 1994, 40 (7 Pt 1), 1299-1305.
10.Chen, X. H., Frank, J. P., Wijsbeek, J., and de Zeeuw, R. A., J. Anal. Toxicol. 1994, 18 (3), 150-153.
11.Hernandez, A., Andollo, W. and Hearn, W. L., Forensic Sci. Int. 1994, 65, (3) 149-156.
12.Stafford R. G., and Hines H. B., J. Chromatogr. B. Biomed. Appl. (Netherlands) 1994, 657 (1), 119-124.
13.Shephard G. S., Thiel, P. G., Sydenham, E. W., Vleggaar, R. and Alberts, J. F., Food Chem. Toxicol., 32 (1) 23-29.
14.Lawrence J. F., Lau, B. P., Cleroux, C. and Lewis, D., J. Chromatogr. (Netherlands), 1994 659 (1), 114-125.
15.Bennett, G. A., Shotwell, O. L., and Kwolek, W. F., JAOAC, 1985, 68, 958-961.
16.Miller, K. V. and Guerrero H. G., Laboratory Information Bulletin, 1991 7 (3), 3555.
17.Ware, G. M., Umrigar, P. P., Carman, A. S., and Kuan, S. S., Analytical Letters, 1994, 27 (4), 693-715.
18.Tosch, D., et al., JAOAC, 1984, 67 (2), 337-339.
19.Ware, G. M., Umrigar, P. P., Carman, A. S., and Kuan, S. S., Analytical Letters, 1993, 26 (8), 1751-1770.
20.Swanson, S. P. et. al., JAOAC, 1986, 69 (1),41-43.
21Lawrence J. F., Lau B. P., Cleroux, C., and Lewis, D., J. Chromatogro., 1994, 659 (1), 119-126.
22.Hatfield C. L., Wekell J. C., Gauglitz, E. J., and Barnett, H. J., Natural Toxins, 1994, 2 (4) 206-211.
CONTROL CORN SAMPLE
SPIKED CORN SAMPLE