African Journal of Food, Agriculture, Nutrition and Development Rural Outreach Program ISSN: 1684-5358 EISSN: 1684-5374 Vol. 9, Num. 3, 2009, pp. 859-872

African Journal of Food Agriculture Nutrition and Development, Vol. 9, No. 3, May, 2009, pp. 859-872

Adaptation of Glass Columns For Clean-up In RP-HPLC Determination Of Aflatoxins With Post-column Derivatisation With Bromine And Fluorescence Detection

Victor Ochieng Owino^1* 

¹Valid Nutrition, P.O. Box 17191, 00100-Nairobi, Kenya.
*Corresponding author email: vowino@hotmail.com

Code Number: nd09030

ABSTRACT

Consumption of aflatoxin-contaminated foods in developing countries may not only cause cancer in the general population, but is also associated with stunting and immune suppression in children. Accurate determination of aflatoxins in foods is  important for the prevention of the above health risks. This study aimed to evaluate the possibility of using cheap florisil-packed glass columns instead of relatively complex and expensive pre-packed cartridges for afltoxin clean-up prior to HPLC analysis. Additionally, effects of aflatoxin extract storage temperature, and the moisture content of florisil on aflatoxin recovery were assessed. Results showed that the standard curve was highly linear as depicted by the high correlation coefficient for all four aflatoxins [0.99, 0.99, 0.97 and 0.98, respectively for AFG₁, AFB₂ and AFB₁ and AFG2. The retention times over a total run time of 20 minutes were 14.8 (SD 0.4), 12.4 (SD 0.1), 10.7 (SD 0.3) and 9.1 (SD 0.3) minutes, respectively for AFB₁, AFB₂, AFG₁ and AFG₂.  Florisil powder that had been dried for 12 hours at 100 ^oC, then hydrated with water at 5% (w/v) before packing on glass columns gave slightly higher recovery for all aflatoxins (100.3% AFB₁, 93.1% AFB₂, 83.8% AFG₁ and 71.1% AFG₂) compared to the pre-packed cartridges (96.4% AFB₁, 91.5% AFB₂, 80.6% AFG₁ and 66.9% AFG₂). There was 10% and 20% aflatoxin loss at 4 ^oC and 22 ^oC, respectively. After 6 hours of storage, no aflatoxin degradation was observed at –18 ^oC. There was 0-5% aflatoxin loss at –18 ^oC after 12 hours, while at 4 ^oC and 22 ^oC, the loss range was 20-50% and 30-55%, respectively. Aflatoxin recoveries ranged from 120-160%, 90-110%, 80-98%, and 55-74% for florisil which had been hydrated with water at 2.5%, 5% and 10%, respectively. Glass columns are cheap and can be used many times and are a feasible and sustainable alternative to pre-packed florisil cartridges for aflatoxin clean up to ensure food safety, especially in resource poor settings.

Key words: Aflatoxins determination, glass column, HPLC

INTRODUCTION

Aflatoxins, secondary metabolites produced by moulds, are known for their acute toxicity and carcinogenicity [1-3].The most common aflatoxins are B₁, B₂, G₁ and G₂. They are associated with immune function modification [4] and disorders in protein and enzyme synthesis and lipid metabolism [5]. Maize and groundnuts, the main ingredients used for complementary foods in most African countries, are most vulnerable to aflatoxin contamination [6, 7]. Application of relatively cheap methods for aflatoxin determination in foods in developing countries may help in the evasion of adverse health effects associated with aflatoxin exposure.

The stability of aflatoxins in reversed phase high performance liquid chromatography (RP-HPLC) solvents such as methanol and acetonitrile is of concern, as samples may spend several hours in these solvents prior to analysis [8]. Aflatoxin detection can be affected by extract storage conditions such as the temperature and length of storage. Further, derivatisation reagent factors such as flow rate may also influence aflatoxin detection. Even though the solubility of many mycotoxins in water is low wet apparatus and reagents may result in significant losses of aflatoxins [9].

Most aflatoxin clean up procedures prior to high performance liquid chromatography (HPLC) determination employ pre-packed commercial cartridges, followed by C18 clean up. These include aluminium oxide-packed minicolumns [10], phenyl-packed minicolumns [11], and Sep-Pak Florisil cartridges [12, 13, 14]. Other studies [15, 16] have reported the use of immunoaffinity columns for clean up in RP-HPLC using post-column derivatisation with bromine. A major disadvantage of pre-packed columns is that they are expensive and must be disposed of after single clean up procedures. In order to efficiently promote optimum public health, the application of relatively inexpensive glass columns may be a feasible and more sustainable alternative as this may help in the evasion of adverse health effects associated with aflatoxin exposure.

This paper presents the assessment of the applicability of a florisil clean up procedure using glass columns (laboratory burettes) for aflatoxin determination by RP-HPLC with post-column derivatisation using pyridinium bromide perbromide and fluorescence detection. The paper further evaluates effect of different analytical conditions including, extract storage temperature; length of extract storage prior to analysis; PBPB flow rate; the moisture content of florisil on aflatoxin recovery.

The findings of this study are likely to inform the choice of most cost-effective methods of aflatoxin determination, especially in developing countries, where a great proportion of the population is likely to be exposed to risk of immune dysfunction and cancer attributable to aflatoxin contamination of staple foods

MATERIALS AND METHODS

Raw peanuts were purchased from Gikomba Market, Nairobi and transported by air to the Department of Food Technology & Nutrition, Ghent University, Belgium. The peanuts were held at room temperature until analysis. Black pepper and nutmeg spices were purchased from a local supermarket in Ghent, Belgium.  Ampoules containing 5mg of pure AFB₂, AFG₁ and AFG₂, respectively, were purchased from Makor Chemical Limited (Israel). A-5mg ampoule of pure AFB₁ was obtained as a gift from the RIKILT laboratory, Netherlands. All solvents used were of HPLC grade.

RP-HPLC determination of aflatoxins

The mobile phase consisted of water: acetonitrile; methanol (55:18:27, v/v) and was filtered through 0.45mm HPLC-certified cellulose filter paper (Watman). The mobile phase was delivered at 1ml/min by a Gilson pump (model 307). The injection solution was water: acetone (85:15, v/v). Aflatoxin solution samples were injected using a Rheodyne 7725I manual injector via a microlitre syringe (Hamilton 1ml 1001LTN, 22GA). The RP-HPLC column was a Chrompack C18RP 290*4.6mm (3m). PBPB solution was delivered to the effluent by a Waters HPLC pump (Model 510) via a T-junction with a low dead volume. The combined flow passed through a Teflon coil (50cm x 0.5mm i.d) before entering a Gilson (Model 122) fluorescent detector with excitation and emission at 360nm and 425nm, respectively.

Effect of PBPB flow rate on aflatoxin detection

The post-column derivatisation reagent was prepared by dissolving 25mg of pyridinium hydrobromide perbromide (PBPB) in 500ml distilled water and shaking vigorously. The solution was stored in an amber bottle wrapped in aluminium foil. A fresh solution was prepared every three days. To study the effect of PBPB on aflatoxin detection, samples were injected with or without PBPB. The derivatisation pump was set at different flow rates i.e., 0.05ml/min, 0.1ml/min and 0.3ml/min and the effect on aflatoxin recovery evaluated.

Aflatoxin standard curves

Contents (5mg) of each aflatoxin ampoule were dissolved in 50ml chloroform to give a stock solution of 0.1mg/ml for each aflatoxin. Oneml of the stock solution was diluted 1000-folds to a 0.1ng/ml from which 2000ml were drawn and evaporated under a gentle stream of nitrogen in a fume hood. The residue was dissolved in 100ml water: acetone (85:15) (v/v) solution resulting in aflatoxin concentration of 2ng/ml. Subsequent dilutions were prepared from this solution namely, 0.25ng/ml, 0.5ng/ml, 1.0ng/ml and 1.5ng/ml.

Spiking of food samples

Fifty grams of ground food sample were weighed out into a conical flask. 50ml was drawn from the 0.1ng/ml stock solution of each aflatoxin and transferred to the conical flask. The contents were shaken thoroughly by hand before shaking on a mechanical shaker (Certomat) for 30 minutes before extraction. The resulting aflatoxin-spiking rate was 100mg/kg.

Aflatoxin extraction and clean-up

Aflatoxins were extracted by adding 250ml chloroform and 25 ml water and shaking on a mechanical shaker (Certomat) for 30 minutes. The suspension was filtered through fluted filter paper (Watman) and 50 ml filtrate collected.

Effect of type of clean up column on aflatoxin recovery

Pre-packed commercial Sep-Pak Florisil cartridges were compared with glass columns (Pyrex 50ml/0.1ml division volumetric burettes with PTFE (Teflon) stop-cock) packed with 1g Florisil powder hydrated with water at 5% (w/v). The Florisil powder for glass column packing was dried overnight for 12 hours in an air oven at 100 ^oC and cooled in a dessicator.

Effect of water addition to Florisil powder on aflatoxin recovery

To study the effect of water on aflatoxin recovery, florisil powder was hydrated by adding distilled demineralised water at 2.5%, 5% and 10% ensuring uniform water dispersion and drying overnight at 100 ^oC.

Florisil Clean-up procedure

Pre-packed cartridges were primed by rapidly passing 8ml chloroform via a stopcock attached to the shorter stem and draining by gravity. A glass column was attached to the longer stem of the cartridge and 2ml chloroform added via the glass column into the cartridge. The florisil-packed glass columns were primed by passing 8ml chloroform by gravity. The 50ml filtrate obtained from extraction was introduced into either the pre-packed or glass columns and drained by gravity. The column was rinsed with 5ml chloroform followed by 20ml methanol. The eluates were discarded. Aflatoxins were eluted with 40 ml acetone: water (98:2, v/v) and collected in a 250ml round bottomed flask and evaporated on a rotary evaporator (Buchi RE11) at 45 ^oC under vacuum.

C18 Sep Pak clean-up

A rubber stopper was attached to the longer stem of a C18 cartridge. An air-tight syringe with teflon plunger was connected to the stopper. The cartridge was primed and air removed by rapidly passing 10ml methanol from the syringe followed by 10ml distilled water. The evaporated residue of aflatoxin eluates from Florisil clean-up was re-dissolved in 1ml methanol and 4ml water and transferred quantitatively to the C18 column, twice rinsing the flask with 5ml water: methanol (80:20, v/v). 25ml water: methanol (80:20, v/v) was passed through the cartridge and eluates discarded. Aflatoxins were eluted with 50ml water: acetone (85:15, v/v) and the whole eluate collected in a volumetric flask.

Effect of extract storage temperature and storage duration on aflatoxin recovery

Different 50ml samples of 1.0ng/ml mixed aflatoxin standard solution were subjected to the Florisil clean-up, followed by C18 clean-up. The extracts were stored in water: acetone (85:15, v/v) in amber bottles wrapped with aluminium foil at –18  ^oC, 4  ^oC and 22  ^oC for 6, 12 and 24 hours in the dark before HPLC determination.

RESULTS

Effect of derivatisation with PBPB on afaltoxin detection

To study the effect of PBPB on aflatoxin detection at a concentration of 10 ng/ml, samples were injected with and without derivatisation by setting the PBPB delivery pump flow rate at 0.1ml/min and 0ml/min, respectively. Table 1 shows the increment or decrement factors in detection with and without PBPB. The detection of AFB₁ and AFG₁ was enhanced 15 and 20 times, respectively in the presence of PBPB, but both peaks disappeared in the absence of PBPB. The addition of PBPB had little effect on the detection of both AFB₂ and AFG₂.

Table 1:Increment or decrement factors of aflatoxin peaks with or without PBPB at a standard mixed aflatoxin solution concentration of  10 ng/ml.

        ND – not detected, (+) =increment, (-) = decrement

Effect of PBPB flow rate on aflatoxin detection

To study the effect of PBPB flow rate on aflatoxin recovery from standard aflatoxin solution prepared at 1.0 ng/ml, the PBPB delivery pump was adjusted to 0.05, 0.1 and 0.3ml/min. Peak areas were calculated for 10 runs at each flow rate. Table 2 gives the average peak areas and the standard deviations for aflatoxins at different flow rates. Aflatoxin peak areas tended to decrease with increasing flow rate. Reproducibility as measured by coefficient of variation was highest at 0.1ml/min. The CV% ranged from 2.7%-6.1%, 0.8%-7.9% and 5.5%-23.8% at 0.05ml/min, 0.1ml/min and 0.3ml/min, respectively as presented in table 3.

Table 2:Average peak areas of aflatoxins at different PBPB flow rates obtained by injecting standard aflatoxin solutions at 1.0 ng/ml¹   

 ¹Figures in parentheses are standard deviations for 10 determinations

Table 3:Percent coefficient of variation for aflatoxins at different PBPB flow rates for standard solutions prepared at 1.0 ng/ml.

Calibration curve

Calibration solutions were prepared with mixed aflatoxins and ranged from 0.25ng/ml-1.5ng/ml for AFG₁, AFB₂ and AFB₁ and 0.5ng/ml-2.0ng/ml for AFG2. The standard curve was highly linear as depicted by the high correlation coefficient for all four aflatoxins [0.99, 0.99, 0.97 and 0.98, respectively for AFG₁, AFB₂ and AFB₁ and AFG2. The retention times over a total run time of 20 minutes were 14.8 (SD 0.4), 12.4 (SD 0.1), 10.7 (SD 0.3) and 9.1 (SD 0.3) minutes, respectively for AFB₁, AFB₂, AFG₁ and AFG₂.

Effect of type of clean up column on aflatoxin recovery

Table 4 shows a comparison of percent aflatoxin recovery obtained by using glass columns packed with florisil powder dried overnight and hydrated with water at 5% (w/v) and pre-packed cartridges. There was a trend towards greater recovery of aflatoxins with glass columns compared with pre-packed florisil cartridges.

Table 4:Comparison of percent aflatoxin recovery obtained by using glass columns packed with florisil powder dried overnight water addition at 5% (w/v) and pre-packed cartridges¹.

¹Figures in parentheses represent standard deviations of ten determinations

Effect of water addition to Florisil powder on aflatoxin recovery

Florisil powder for glass column packing was dried in an air oven at 100 ^oC overnight and moistened by adding water at 2.5%, 5% and 10% (w/v) and shaken vigorously for 30 minutes and allowed to rest for 30 minutes. Figure 1 shows the influence of addition of water to florisil powder on aflatoxin recovery from standard solutions. It was observed that the addition of 2.5% water resulted in increased recovery (90 -120%) for all the aflatoxins. However, above 2.5% water aflatoxin recovery was reduced as the amount of water increased. The recoveries ranged from 80-98% and 55-74% for florisil with water at 5% and 10%, respectively. Addition of 5% water resulted in 10-20% loss in recovery, while at 10% water, the loss of recovery ranged from 27-66%.

Effect of extract storage temperature and storage duration on aflatoxin recovery

Even though HPLC method for aflatoxin determination is accurate and sensitive, it requires time-consuming sample clean up and several hours elapse between extraction and analysis [9, 17]. Dilute solutions of aflatoxins are not particularly stable, especially to light and must be renewed regularly [10]. All standard solutions of mixed aflatoxins that had been subjected to the clean-up procedure were stored in water: acetone (85:15, v/v) in amber bottles wrapped with aluminium foil at –18  ^oC, 4  ^oC and 22  ^oC for 6, 12 and 24 hours in the dark before HPLC determination. Aflatoxin recovery was evaluated against the temperature and duration of storage. Figure 2 shows the influence of holding time at -18 ^oC on the stability of AFB1, AFB2 and AFG1 standard solutions. There was 10% and 20% aflatoxin loss at 4 ^oC and 22 ^oC, respectively. After 6 hours of storage, no aflatoxin degradation was observed at –18 ^oC. There was 0-5% aflatoxin loss at –18 ^oC after 12 hours, while at 4 ^oC and 22 ^oC, the loss range was 20-50% and 30-55%, respectively. It was observed that AFB₂ was more stable at all temperatures followed by AFB₁. AFG₁ was the least stable. At 4 ^oC, the losses were 20%, 30% and 50% for AFB₂, AFB₁ and AFG₁, respectively. At 22 ^oC only 30% AFB₂ had been lost compared to about 50% for both AFB₁ and AFG₁ after 12 hours. After 24 hours the loss ranges were 25-55%, 40-50%, and 40-50%, respectively for –18 ^oC, 4 ^oC and 22 ^oC with AFB₂ still showing the lowest degradation.

DISCUSSION

This study was designed to assess the applicability of glass columns (laboratory burettes) for aflatoxin determination by RP-HPLC and, to evaluate effect of extract storage temperature; length of extract storage prior to analysis; PBPB flow rate; and the moisture content of florisil on aflatoxin recovery.

The results showed that the addition of PBPB enhanced the detection of AFB₁ and AFG₁, but had little effect on the detection of AFB₂ and AFG₂. Aflatoxin peak areas tended to decrease with increasing flow rate. There was a trend towards greater recovery of aflatoxins with glass columns compared with pre-packed florisil cartridges. The addition of water to florisil was beneficial in peak enhancement up to 2.5% w/w of water. There was lesser aflatoxin detection as storage temperature and time increased, but this observation was more pronounced for AFB₁ and AFG₁.

The enhancement of AFB₁ and AFG₁ peaks could be explained by the fact that these two aflatoxins possess a double bond at position 8,9 on the terminal furan ring [18]. The first step of derivatization is the bromination of the 8,9-double bond to form a high fluorescent product. This bond does not exist in AFB₂ or AFG₂ and hence they do not react with bromine [18]. Thus reduction in AFB₂ and AFG₂ detection is as a result of a dilution effect of PBPB solution.  The disappearance of AFB1 and AFG1 without PBPB and their appearance with PBPB is a good confirmatory test for the presence of these aflatoxins [1].

That aflatoxin peak areas decreased with increasing flow rate can be explained by the fact that when an excess of bromine is present, the first reaction (bromination of 8,9-double bond of AFB₁ and AFG₁) is followed by a second reaction (bromination of the phlorogucinol nucleus), which yields a low fluorescent product [18]. The decrease in peak areas for AFB₂ and AFG₂ at higher PBPB flow rate could be attributed to dilution effect.

The relatively higher recovery with the glass column can be explained by the fact that the overnight drying of florisil powder resulted in air and moisture expulsion. The CV% for recoveries on pre-packed columns ranged from 17-41% compared with 11-22% for glass columns. The presence of air bubbles in florisil could have resulted in the large variations observed in the recoveries. This effect was partially eliminated by drying glass columns overnight at 100 ^oC. Secondly, gravity elution through the pre-packed cartridges took longer time than with glass columns. This could have resulted in aflatoxin degradation. A higher solvent head in the glass columns resulted in faster elution. The average aflatoxin recoveries, 83.9% and 87.1% for Sep-Pak Florisil cartridges and glass columns, respectively are within the range of average recovery of 84% reported previously [12]. Another study [13] found an average recovery of 91% for pre-packed florisil column clean up. Studies using immunoaffinity columns have reported recoveries of the range 71-92% total aflatoxin, 82-109% for AFB₁  and 92-101% for AFB₁ [15, 16].

The observed decrement of aflatoxin peaks when more than 2.5% water was added to florisil shows that hydration of aflatoxin molecules to some extent may enhance detection; however, the mechanism of this effect cannot be explained in this paper.

The existence of a double bond on position 8,9 at the terminal furane ring of AFG₁ and AFB₁ may explain their more reactivity during storage compared to AFB₂ and AFG₂ both of which lack the double bond [10, 18].

CONCLUSIONS

Glass columns are cheap and can be used many times and are a feasible and sustainable alternative to pre-packed florisil cartridges for aflatoxin clean up. The glass columns also present a more rapid clean up alternative. Even though the method described in this study was originally meant for the determination of AFB₁, this study has shown that it is also applicable to AFB₂, AFG₁ and AFG₂ between concentrations from 0.25ng/g  to at least 10 ng/g. It is recommended that all aflatoxin extracts be analysed within 6 hours of clean up. However, extracts can be held for up to 12 hours at –18 ^oC. All apparatus used for aflatoxin analysis need to be dried to remove air and water that would otherwise cause variation in results and loss of aflatoxins. PBPB flow rate has an effect both on aflatoxin detection and method reproducibility and 0.1ml/min seems to be the optimum value. These findings may be used as the basis for the choice of most cost-effective methods of aflatoxin determination, especially in developing countries, where a great proportion of the population is likely to be exposed to risk of immune dysfunction and cancer attributable to aflatoxin contamination of staple foods.

ACKNOWLEDGMENTS

I acknowledge the technical input of Andre Hughebaert and Bruno de Meuleaner in study design, data analysis and manuscript review. I am indebted to the Flanders Inter-University Council (VLIR), Belgium for providing funds for the study and to Dr. Rachel Gitau for reviewing the manuscript. Many thanks to RIKILT, Netherlands for the provision on AFB₁ stock solution.

REFERENCES

1. Jaimez J, Fente CA, Vazquez BI, Franco CM, Cepeda A, Mahuzier G and Prognon P Application of the assay of aflatoxins by liquid chromatography with fluorescence detection in food analysis. J. Chromtogr.A.2000; 882: 1-10.
2. Hall AJ and Wild CP. Liver cancer in low and middle income countries: prevention should target vaccination, contaminated needles, and aflatoxins. BMJ. 2003; 326: 994-995.
3. Nasir MS and Jolley ME Development of a fluorescence polarization assay for the determination of aflatoxins in grains. J. Agric. Food Chem.2002; 50: 3116-3121.
4. Turner PC, Moore SE, Hall AJ, Prentice AM and Wild CP Modification of immune function through exposure to dietary aflatoxin in Gambian children. Environ. Health Perspect. 2003; 111 (2): 217-220.
5. Hendrickse RG The influence of aflatoxins on child health in the tropics with particular reference to kwashiorkor. Trans. R. Soc. Trop. Med. Hyg. 1984; 78 (4): 427-435.
6. Gong YY, Egal S, Hounsa A, Turner PC, Hall AJ, Cardwell KFand Wild CP Determinants of aflatoxin exposure in young children from Benin and Togo, West Africa: the critical role of weaning. Int J Epidemiol. 2003; 32 (4): 556-562.
7. Paulsch WE, Sizoo EA and Van Edmond HP Liquid chromatographic determination of afaltoxins in feedstuffs containing citrus pulp. J. AOAC. 1988; 71 (5): 957-961.
8. Shepherd JM High performance liquid chromatography and its application to the analysis of mycotoxins. In: Modern methods in the analysis and structural elucidation of mycotoxins, Cole, J.R., ed. Academic Press, Inc. Orlando, San Diego, New York, Boston, London, Sydney, Tokyo, Toronto. 1986; p 293
9. Beaver RW Determination of aflatoxins in corn and peanuts using high performance liquid chromatography. Arch. Environ. Contam. Toxicol.1989; 18 (3): 315-8.
10. Leitao J, Blanquat SDG, Bailly JR and PaillasCH Quantitation of aflatoxins from variuos strains of Aspergillus in foodstuffs. J. Chromatogr. A. 1988; 435: 229-234.
11. Sobolev VS and Dorner JW Cleanup procedure for determination of aflatoxins in major agricultural commodities by liquid chromatography. J. AOAC Int.2002; 85 (3): 642-5.
12. Dorner JW and Cole RJ Rapid determination of aflatoxins in raw peanuts by liquid chromatography with postcolumn iodination and modified minicolumn cleanup. J. AOAC Int.1988; 71 (1): 43-7.
13. Kamimura H, Nishijima M, Yasuda K, Ushiyama H, Tabata S, Matsumoto S and Nishima T Simple, rapid cleanup method for analysis of aflatoxins and comparison with various methods. J. AOAC Int.1985; 68 (3): 458-61.
14. Stroka J, Anklam E, Jorissen U and Gilbert J Immunoaffinity column cleanup with liquid chromatography using post-column bromination for determination of aflatoxins in peanut butter, pistachio paste, fig paste, and paprika powder: collaborative study. J. AOAC Int.2000; 83 (2): 320-40.
15. Stroka J, von Holst C, Anklam E and Reutter M Immunoaffinity column cleanup with liquid chromatography using post-column bromination for determination of aflatoxins in peanut butter, pistachio paste, fig paste, and paprika powder: collaborative study. J. AOAC Int.2003; 86 (6): 1116-23.
16. Strachan CJN, John GP and Miller GI Application of an automated particle-based immunosensor for the detection of aflatoxin B₁ in foods. J. Food Agric. Immunol.1997; 9: 177-184.
17. Kok TW Determination of aflatoxins in cattle feed by liquid chromatography and post-column derivatisation with electrochemically generated bromine. J. Chromatogr. 1986; 367: 231-236.   
18. Dollear GF Detoxification of aflatoxins in foods and feeds. In: Aflatoxins, Goldblatt, A.L., ed. Academic Press, New York and London. 1969;  p359

© Copyright 2009 - Rural Outreach Program

The following images related to this document are available:

Photo images