This article was downloaded by: [Monash University Library] On: 07 December 2014, At: 03:28 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Food Additives & Contaminants Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tfac19

Reproducibility of a commercially available kit utilizing enzyme‐linked immunosorbent assay for determination of aflatoxin in peanut butter a

C. M. Ward & M. R. A. Morgan

a

a

AFRC Institude of Food Research , Colney Lane, Norwich, NR4 7UA, UK Published online: 10 Jan 2009.

To cite this article: C. M. Ward & M. R. A. Morgan (1991) Reproducibility of a commercially available kit utilizing enzyme‐linked immunosorbent assay for determination of aflatoxin in peanut butter, Food Additives & Contaminants, 8:1, 9-15, DOI: 10.1080/02652039109373951 To link to this article: http://dx.doi.org/10.1080/02652039109373951

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sublicensing, systematic supply, or distribution in any form to anyone is expressly

Downloaded by [Monash University Library] at 03:28 07 December 2014

forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

FOOD ADDITIVES AND CONTAMINANTS, 1991, VOL. 8, NO. 1, 9 - 1 5

Reproducibility of a commercially available kit utilizing enzymelinked immunosorbent assay for determination of aflatoxin in peanut butter C. M. WARD and M. R. A. MORGAN AFRC Institude of Food Research, Colney Lane, Norwich NR4 7UA, UK

Downloaded by [Monash University Library] at 03:28 07 December 2014

(Received 9 March 1990; revised 29 June 1990; accepted 24 July 1990) The reproducibility of a commercially available kit utilising enzyme-linked immunosorbent assay (ELISA) for determination of total aflatoxins has been analysed. Peanut butter samples contaminated with aflatoxin at three concentrations (mean 5.8, 11.7 and 24.8 ng/g) were analysed on different occasions using four kits. Coefficients of variation at all concentrations ranged from 2.6% to 13.6% within replicate, 0.3% to 16.2% within plate and 6.7% to 10.2% between kits. Standard curve precision profiles showed a minimum coefficient variation (CV) of 2-4% and a working range (below 10% CV) covering almost the entire standard curve. Keywords: Aflatoxin, ELISA, reproducibility, peanut butter, kit.

Introduction

There is considerable evidence for the human toxicity and carcinogenicity (Van Rensburg et al. 1985, IARC 1987) of the group of secondary fungal metabolites known as afiatoxins. Produced by strains of Aspergillus flavus and A. parasiticus under conditions of high temperature and humidity, aflatoxins are of concern in the UK primarily in imported commodities such as cereals and nuts intended for animal and human consumption (Pohland and Wood 1987). The potential risk to health has led to almost world-wide governmental imposition of voluntary and statutory tolerance levels for these mycotoxins in foodstuffs (Van Egmond 1989). Regulations exist requiring aflatoxin analysis of a wide range of foods. However, uneven distribution of the toxin necessitates the assessment of large numbers of samples obtained in statistically derived sampling plans in order to obtain meaningful values for contamination. Many techniques have been used to measure aflatoxin in food (Shepherd et al. 1987), including thin-layer chromatography (TLC) and high-performance liquid chromatography (HPLC), but these can require laborious clean-up procedures and, in the case of HPLC, expensive equipment. Techniques based on immunoassay can be applied rapidly to single samples, or can enable the simultaneous analysis of large numbers of samples with simple equipment which can also be used outside the specialized laboratory. Many kits using the immunoassay principle to determine aflatoxin are now commercially available (Fukal and Kas 1989) and acceptance by analysts is increasing. The simplest assay format is a card or spot test suitable for analysis of small numbers of samples. Such assays give a visual, semi-quantitive indication of the presence of aflatoxin. Other types of kit apply the enzyme-linked immunosorbent assay 0265-203X/91 $3.00 © 1991 Taylor & Francis Ltd.

10

C. M. Ward and M. R. A. Morgan

Downloaded by [Monash University Library] at 03:28 07 December 2014

(ELISA) microtitration plate technology to quantitative analysis suitable for application to large sample numbers. One study (Mortimer et al. 1988) estimated that, even given the less than strenuous rate of analysis of two plates per day, ELISA was at least 6 times quicker than HPLC in the analysis of peanut butter samples. Many studies have shown the high degree of correlation between immunologically based tests and conventional procedures, including a recent report where 300 samples were analysed by highperformance thin-layer chromatography (HPTLC), HPLC and ELISA. Agreement was excellent (Dell et al. 1990). The purpose of the present study was to determine the reproducibility and variability of a commercially available kit, both within and between kits, when applied to the analysis of peanut butter. Methods Peanut butter Naturally contaminated nuts (the kind gift of Dr Tony Taylor, Humberside College of Higher Education, UK) were roasted for 20 min at 180°C before removing the testas and grinding the nuts to a paste in an Ika-Werk (Janke and Kunkel, West Germany) grinder in 5 X 1 min bursts. The peanut butter was finely sieved and diluted in stages by thorough mixing with an aflatoxin-free retail sample of smooth peanut butter to give three samples (3 x 200 g) with approximate concentrations of 5, 10 and 25 ng/g. As the oil tends to separate from the butter on storage each contaminated sample was immediately aliquoted into approximately 6 g lots to ensure continued homogeneity. Prior to analysis, 5 g was weighed from each aliquot after stirring thoroughly. HPLC analysis of one of each of the samples (M. Sharman, MAFF Food Science Laboratory, Norwich, UK) gave values for total aflatoxin content of 5*6, 13 and 28 ng/g respectively. Assay procedure Four Biokits Total Aflatoxin Assay Kits (Cortecs Diagnostics, UK) were purchased from the manufacturer and were representative of normal commercial production. The kits include a 96-well plate in 6 X 2 row strips and all reagents required for the assay either ready to use or in a concentrated solution. The instructions supplied were closely followed, except that, as plates were not shaken during incubation, times were increased as described. The monoclonal antibody used in the kit was produced using an aflatoxin Bi-BSA immunogen synthesized after formation of the Bi-oxime derivative (Kang, Bramham and Morgan, unpublished data). Four replicate samples (5 g) of each of the three concentrations of peanut butter, together with a single replicate of the blank peanut meal provided, were each weighed into 50 ml stoppered round-bottomed flasks. To each was added 25 ml of acetonitrile:water (1:1, v:v), the stoppers were taped on and the flasks shaken for 30 min on a wrist-action shaker. Extracts were filtered through Whatman No. 1 paper and the filtrate diluted 1:25 (100 ^1 + 2-4 ml) in diluent solution. Both negative and positive control samples are provided with the kit. The negative control peanut meal extract was also diluted as directed 1:25 (200 /tl + 4-8 ml) in dilutent solution. The positive control was prepared by the addition of 100 id of

Downloaded by [Monash University Library] at 03:28 07 December 2014

Reproducibility of an ELISA kit

11

the aflatoxin control concentrate provided to 2-4 ml of the diluted negative control peanut meal extract. Samples (five replicates), controls (three) and standards (six) were pipetted (50 /tl/well) into the microtitration plate according to the plan (figure 1). Standards were supplied ready diluted at 16, 40, 160, 400 and 1600 pg/ml. To each well, 50 /tl of rat anti-aflatoxin monoclonal antibody was added before the plate, fitted with a lid, was incubated at 4°C overnight. The plate was hand-washed five times with wash buffer from a plastic wash bottle. Following the addition of anti-rat horseradish peroxidase conjugate (100 /tl/well) the plate was incubated at room temperature for 1 h. Then after a further wash cycle, 100 /il of ABTS substrate solution (2, 2-azinobis (3ethylbenzthiazoline) sulphomc acid) was added to each well and the plate left for 1 h at room temperature with occasional agitation. The enzymic reaction was stopped with 50 /tl/well of 1*5% (w/v) sodium fluoride solution (provided with the kit) and the plate read at 414 nm on a Titertek MCC (Flow Laboratories, UK) plate reader. The analysis of the three samples was repeated completely (including extraction) on four different days, using a new kit each day. Data analysis Data were processed using Immunofit EIA/RIA software (Beckman, USA) on a Vig II personal computer (Viglen, UK). Standard curves with 95% confidence limits were fitted using the four-parameter logistic equation: Y= . (A7D\

1

A

2

NC

3

4

A3

STD

B C

5

B1 A2

+D

6

7

8

9

NC

PC

B4

C1

10

11

12

PC

STD

STD

C3

A4 C2

D E F G

B3 A1

B2

C4

H Figure 1. Plate layout of controls, standards and samples. NC = Negative control; PC = positive control; standards were arranged in three duplicate rows of 0-8, 2, 8, 20 and 80 pg/well, and samples in five replicate wells for each of four extracted aliquots of three peanut butter samples.

C. M. Ward and M. R. A. Morgan

12

Downloaded by [Monash University Library] at 03:28 07 December 2014

where A = maximum absorbance; B = slope; C= midpoint and D = minimum absorbance. Coefficients of variation (CV) were calculated for each standard and replicate sample. The use of a precision profile, a plot of °7oCV against concentration, provided an assessment of assay performance and standard curve working range. Results and discussion Immunoassays for both quantitative and semi-quantitative analysis of mycotoxins have been shown to offer a number of advantages over conventional procedures. The combination of antibody specificity and assay sensitivity allows the use of rapid, simple toxin extraction procedures without the need for sample clean-up. As well as the savings in time and cost, increased simplicity can also be expected to reduce assay variation and the need for high levels of operator skill. In the present example the aflatoxins in the peanut butter samples are solubilized and analysed with only a dilution step in between. Immunoassays also result in time savings, either by employing very short assay times, of the order of minutes, or by analysing large numbers of samples in a batch-wise manner. The manipulations and measurements required are simple, and utilize comparatively inexpensive equipment, or alternatively can be found automated if required. The results for the determination of aflatoxin in each of the peanut butter samples measured with different kits on each of four days are given in table 1. The mean correlation coefficient (r) for the four standard curves was 0-9979 ± 0-0003. The within-replicate CV was generally below 10%, although between-replicate CV tended to be more variable with some values above 10%. The between-kit CV was 10% or below. The mean aflatoxin content of the three samples was 5-8, 11-7 and Table 1. Aflatoxin content of three peanut butter samples measured by total aflatoxin ELISA kit. Sample

Kit 1

Al A2 A3 A4

4-4t (5-Dt 5-3 4-6 5-5

(2-0) (2-8) (8-7)

X

5-0

(10-6)

Bl B2 B3 B4

10-9 11-5 14-0 10-2

(6-0) (1-6) (5-4) (6-9)

X

11-7 (14-1)

Cl C2 C3 C4

24-1 24-4 26-0 23-8

X

24-6

Positive control Blank peanut meal

Kit 2 4-9 7-3 6-5 6-0

(2-6) (7-4) (5-4) (6-6)

6-2 (16-2)

Kit 4

(6-4) (8-4) (6-0) (4-9)

6-3(6-4) 6-1(3-2) 5-7(2-6) 7-0(3-4)

5-8 (15-2)

6-3(8-6)

4-8 6-8 6-2 5-4

5-8 (10-2)

12-2(7-1) 11-5(4-3) 12-1(7-2) 11-6(5-7)

12-6 (11-7)

10-7 (3-7)

11-9(0-3)

(1-9) (7-9) (5-3) (6-2)

26-1 24-2 20-0 25-2

25-6 21-3 22-7 27-8

(6-3) (5-4) (7-4) (4-8)

26-1(3-4) 27-7(4-6) 27-4(5-9) 29-6(6-3)

(3-9)

23-9 (11-2)

23-1 (7-8)

27-7(5-2)

24-8 (8-1)

15-5

(8-9)

15-4 (6-3)

14-9

(3-D

13-0(1-6)

14-7 (1-2)

1-2

(2-2)

1-4 (4-4)

0-7(0-9)

1-4 (0-7)

2-4

(6-8) (6-0) (6-5) (3-5)

(11-5) (13-6) (7-2) (4-5)

0-9)

10-0 11-1 10-2 10-6

X

(3-5) (11-1) (6-0) (8-4)

tMean of 5 readings.

11-3 14-7 12-4 12-0

Kit 3

11-7 (6-7)

Downloaded by [Monash University Library] at 03:28 07 December 2014

Reproducibility of an ELISA kit

13

24-8 ng/g, as determined by the ELISA (corresponding HPLC values were 5-6, 13 and 28 ng/g). All readings were entered into the analysis so that the CV values include accumulated errors from all sources. Thus the errors due to inhomogeneity of sample, variation in efficiency of extraction, bias due to differences between wells as provided, plate-reader accuracy and inconsistency of reagents between kits, in addition to operator errors in weighing, diluting and pipetting samples, will contribute to the observed variations. One of the four standard curves is shown in figure 2 with the 95°7o confidence limits and a superimposed precision profile. Using the confidence limits at an aflatoxin concentration of 5 pg/well would produce a value between 4-3 and 7-0pg/well on this standard curve (1 pg/well = 2-5 ppb). The minimum %CV from the precision profile was 3% at approximately 5 pg/well. The assay's working range, accepting a maximum CV of 10%, can be seen form the precision profile as being between 1 and 50 pg/well. Working ranges for the remaining curves were 1-80 pg/well, 1-70 pg/well and 1-5-80 pg/well, with minimum CV values of 3%, 4% and 2% respectively. The kit instructions recommend that the absorbance for zero aflatoxin be between 1-4 and 2-0, with a target optical density of 1-7. The mean absorbance for the blank peanut meal extracts for the four kits were 1-29, 1 • 71, 2-02 and 1 -89 respectively. Thus the first kit fell outside the preferred range but without apparent deleterious effects on performance. The inclusion of blank peanut meal and positive controls also allow a further

Aflatoxin (pg/well)

Figure 2. An aflatoxin standard curve with 95% confidence limits (dotted line) and precision profile, for one of the four kits.

Downloaded by [Monash University Library] at 03:28 07 December 2014

14

CM.

Ward and M. R. A. Morgan

check on assay performance (table 1). The former, measured on the extrapolated standard curve, gave apparent mean aflatoxin content of 1 -4 ng/g—a value below the limit of detection (2 ng/g) of the kit. The positive controls were consistent with a mean of 14-7 ng/g. This is higher than the range of values (8-0-12-4, mean 10-05 ng/g) given by the kit manufacturers in the instructions. Use of a plate layout which was different from that suggested by the manufacturers of the kit permitted an increase in standard and sample replication. This was necessary to allow for detailed analysis of within- and between-replicate and plate variation. Thus standards were placed in three duplicate sets across the plate and each replicate sample was assayed in five wells reducing numbers of samples assayed from a maximum of 37 to 12. The mean results for total aflatoxin content of the three contaminated samples A, B and C were 5-8, 11*7 and 24-8 ng/g respectively. The UK Ministry of Agriculture, Fisheries and Food has proposed that nuts and nut products destined for human consumption should not contain more than 10 ng/g total aflatoxin. It is of interest, therefore, to look at the results of table 1 relative to the likely level of statutory interest. Sample A, analysed 16 times with four kits, gave results ranging from 4-4 to 7-3 ng/g—all values below 10 ng/g. The results for sample B ranged from 10-0 to 14-7 ng/g, and for sample C from 20-0 to 29-6 ng/g. All of the analyses of B and C would have classified these samples as being 'positive', a particularly interesting observation since the level of contamination of B (mean value 11-7 ng/g) is so close to the 10-0 ng/g level of interest. The use of microcomputers in the analytical laboratory is allowing complex statistical analysis of data, thus increasing the assurance of the analyst in the data. The most appropriate statistical assessment of the reliability of the performance of an assay is the precision profile. This takes into account the slope of the standard graph and the consequent standard deviation on the dose and response axes. The characteristic profile is U-shaped and any deteriorations in kit reagents, assay conditions or operator error are likely to be reflected in the shape or position of the curve. Comparison of successive precision profiles will reveal any slight but consistent drift in any of these parameters. By selecting an acceptable level of %CV the working range of the assay can be determined. We strongly suggest that the use of precision profiles should increase for all quantitative analytical procedures in the food area. In conclusion, the present observations have shown that there is a high degree of reproducibility between assays over the range of concentrations of peanut butter tested. Variation measured by CV and precision profile is well within the accepted limits, and compares favourably with that from the best HPLC techniques available. Acknowledgements The financial support of the UK Ministry of Agriculture, Fisheries and Food is gratefully acknowledged. We also thank Matthew Sharman of the MAFF Food Science Laboratory for the HPLC analyses. References DELL, M. P. K.,

GRZESKOWIAK, R.,

HASWELL, S. J., JEWERS, K., COKER, R. D., MEDLOCK, V.

P.,

TOMLINS, K., and ROCH, O. G., 1990, Analytical methodology for the determination of aflatoxins in peanut butter. Part A: The comparison of HPTLC, ELISA and HPLC methods. Analyst (In press).

Reproducibility of an ELISA kit

15

FUKAL, L., and KÁS, J., 1989, The advantages of immunoassay in food analysis. Trends in Analytical Chemistry, 8, 112-116. IARC 1987, Aflatoxins. IARC Monographs on the evaluation of the carcinogenic risk of chemicals to humans, Suppl. 7 (Lyon: International Agency for Research on Cancer), pp. 83-87. MORTIMER, D. N., SHEPHERD, M. J., GILBERT, J., and MORGAN, M. R. A., 1988, A survey of the

occurrence of a aflatoxin B1 in peanut butter by enzyme-linked immunosorbent assay. Food Additives and Contaminants, 6, 139-188. POHLAND, A. E., and WOOD, G. E., 1987, Occurrence of mycotoxins in food. Mycotoxins in Food, edited by P. Krogh (London: Academic Press), pp. 35-64. SHEPHERD, M. J., MORTIMER, D. N., and GILBERT, J., 1987, A review of approaches to the rapid

analysis of aflatoxins in food. Journal of the Association of Public Analysts, 25, 129-142. VAN, EGMOND, H. P., 1989, Current situation on regulations for mycotoxins. Overview of tolerances and status of standard methods of sampling and analysis. Food Additives and Contaminants, 6, 139-188.

Downloaded by [Monash University Library] at 03:28 07 December 2014

VAN RENSBURG, S. J., COOK-MOZAFFARI, P., VAN SCHALKEWYK, D. J., VAN DER WATT, J.

J.,

VINCENT, T. J., and PURCHASE, I. F., 1985, Hepatocellular carcinoma and dietary aflatoxins in Mozambique and Transkei. British Journal of Cancer, 51, 713-726.

Reproducibility of a commercially available kit utilizing enzyme-linked immunosorbent assay for determination of aflatoxin in peanut butter.

The reproducibility of a commercially available kit utilising enzyme-linked immunosorbent assay (ELISA) for determination of total aflatoxins has been...
392KB Sizes 0 Downloads 0 Views