An analytical device for on-sight immunoassay: demonstration of its applicability in detection of DDT in food samples

Abstract

A simple analytical device was developed for performing non-instrumental immunoassay. The device consists of membrane with antibody-coated zones and filter pad acting as an absorbent body. Very low concentrations of DDT in food samples were estimated by using an improved catalysed reporter deposition I-CARD method. The signal amplification involved biotinylated tyramine and avidin–alkaline phosphatase (ALP) conjugate. 5-Bromo-4-chloro-3-indolyl-phosphate in conjunction with nitro blue tetrazolium was used for colour visualisation. Semi-quantitative results were obtained by visual comparison of the colour intensity of the sample spot with those of reference standards. Quantitative estimation is possible by densitometry using CCD camera. A batch of five extracted samples could be analysed in a single kit within 30 minutes. Spiked samples were analysed without sample cleanup. The possible matrix interference was analysed. The results were comparable with gas chromatography and CCD analyses. The method is well suited for visual screening of food samples for DDT under field conditions.

Keywords:

• immunoassay
• signal amplification
• DDT
• food samples
• matrix effect

Introduction

1,1,1-Trichloro-2,2-bis (p-chlorophenyl)ethane (DDT) is the most economical and versatile insecticide for both agricultural and public health applications. Due to the worldwide use of this pesticide, more than one million tonne of this chlorinated compound has been added to the environment every year. DDT and its metabolites 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene and 1,1-dichloro-2,2-bis(p-chlorophenyl)ethane are resistant to physical, chemical and biological degradation and thus get accumulated in the adipose tissues of animals and man, as well as in the environment (Kennish & Ruppel, 1996). DDT has been reported to be carcinogenic, mutagenic and teratogenic and is known to cause other toxic effects (Smith, 1991).

Thin layer chromatography and gas liquid chromatography are generally used to analyse DDT (Dikshit, 1990). But these methods need an extensive cleanup of the sample and are labour intensive, expensive, time consuming and hence are not suitable for field applications. With an increasing demand for pesticide residue analysis in various food and agriculture commodities, it is imperative to develop a simple, quick, specific and sensitive assay to detect the pesticide. Immunochemical techniques are considered as good alternatives to the more classical chromatographic methods. Immunoassays are biological techniques, which are fast, sensitive, selective and economical. They provide a variety of advantages like high throughput of samples, reduced sample preparation, versatility for target analytes and adaptability to field condition (Shan, Lipton, Gee, & Hammock, 2002).

Immunoassays have been developed for some pesticides by others (Centero, Johnson, & Sehon, 1970; Engvall & Perlmann, 1972; Ercegovich et al., 1981; Hammock & Mumma, 1980; Langone & Van Vunakis, 1975). Non-isotopic immunoassay such as enzyme-linked immunosorbent assay (ELISA) is sensitive, specific and provide high sample throughput. But the method requires well-equipped laboratory, trained personnel and takes few hours to complete a test. These methods are not suited for on-field detection. Hence, numerous membrane-based methods such as dipsticks and immunochromatographic test strips have been developed (Pal & Dhar, 2004). But in these assays also, only a single sample can be assayed at a time. For rapid analysis of a batch of samples, development of novel approach is required to provide reliable semi-quantitative results without instrumentation.

In this report, we describe a cost–effective analytical device for performing a non-instrumental Dot-ELISA technique for the detection of DDT on a batch of food samples. The easy-to-manufacture device consists of a nitrocellulose membrane mounted on filter pad. The membrane is marked into six zones, and each spot is coated with anti-DDT antibody, i.e. five immunochemical reactions can be performed simultaneously. We have used improved catalysed reporter deposition method of signal amplification termed I-CARD to increase detection sensitivity. The method is similar to tyramide signal amplification method (Bobrow, Shaughnessy, & Litt, 1991) except for the use of synthetic electron rich proteins containing multiple phenolic groups to block the vacant sites of the membrane. This markedly increases the deposition of biotinylated tyramine (B-T) molecules. The net effect is high signal amplification and increased sensitivity compared to the tyramide signal amplification method. By this method, DDT could be detected with high sensitivity under field conditions within 30 minutes.

Materials and methods

Nitrocellulose membrane, 0.45 µm was from BioRad Company, USA. Filter paper, Whatman No. 3 was from Whatman (Springfield, Maidstone, UK). Semi-rigid polyethylene circular pieces, adhesive, analytical grade buffer salts were from local market. Other chemicals were from Sigma Aldrich Co., St. Louis, USA. DDA-protein conjugates were prepared according to Deepthi (2010).

Preparation of immunoassay reagents

Polyclonal antibody was raised in New Zealand white rabbits using DDA–BSA conjugate as immunogen. The antibody was characterised by ELISA (Deepthi 2010) and was found to be highly specific for DDT and its metabolites. For the present study, the antibody was purified by ammonium sulphate precipitation (33% saturation) followed by dialysis against phosphate-buffered saline. It was passed through BSA-Sepharose column to remove anti-BSA antibodies.

The DDA-ALP conjugate was prepared according to Deepthi (2010), and was purified by extensive dialysis against phosphate buffer (50 mM, pH 7.5).

Preparation of electron rich casein

Casein was dissolved in sodium bicarbonate (150 mM). To the aliquot phenoxyacetic acid–NHS dissolved in dry dimethylformamide (DMF) was added drop wise with stirring. The reaction mixtures were stirred continuously at room temperature for three hours and then dialysed extensively against distilled water and subsequently against phosphate buffer (20 mM, pH 7.5). The casein conjugate thus obtained was centrifuged to remove slight turbidity and then lyophilised. The conjugate was characterised spectroscopically at 214 nm.

Preparation of biotinylated tyramine

A solution of biotin in a mixture of dry DMF and dimethyl sulfoxide was treated with N-hydroxysuccinimide and dicyclohexylcarbodiimide overnight at 4°C. The activated ester solution was filtered to remove urea and added to a solution of tyramine hydrochloride in DMF. The reaction mixture was stirred overnight in the presence of powdered solid sodium carbonate and filtered. The resulting solution was stored at 4°C and used in the assay directly with suitable dilution.

Membrane preparation for assay

Nitrocellulose membrane circles (50 mm diameter) were marked with a pencil to give 10 mm circles. The membrane was soaked in buffer containing 0.01% triton- X100 (Tris–HCl 20 mM, pH 8.0) for five minutes, with gentle shaking. The membrane was then semidried. Anti-DDA antibody was diluted 1: 25,000 times with Tris-buffered saline (50 mM, pH 8.0, 0.9% NaCl). The diluted antibody (5 µl) was loaded within these marked circles with the tip of a dispenser. The membrane was dried at room temperature for 15 minutes and then at 37°C for 30 minutes. The vacant sites of the membrane were blocked with 0.5% electron rich casein in carbonate–bicarbonate buffer (50 mM, pH 9.6), and then washed with washing buffer three times (Tris–HCl, 20 mM, pH 8.0 containing 2.9% NaCl and 0.05% tween 20). The strips were dried at room temperature for 15 minutes and then at 37°C for 30 minutes.

Assembly of the analytical device

The prepared membrane was mounted on a Whatman blotter using adhesive. The polyethylene card containing the holes corresponding to the spotted areas on the membrane was placed on the membrane, with the holes exactly matching the spotted area. The whole set up was then placed in the box.

Assay

The antibody-coated membrane was wetted with buffer. Standard DDT (1 µg through 1 ng) present in phosphate buffer containing DMF was loaded at the centre of the antibody-coated zones (required quantity of DDT present in 5 µl). To each spotted area, 5 µl DDA–ALP conjugate (1:25,000) in phosphate buffer (50 mM, pH 7.2) was also added. The reaction was allowed to proceed for five minutes. The spotted area was washed by dropping washing buffer using pipette/dropper. The area was then gently sponged with a tissue paper. Five microlitres of 5-bromo-4-chloro-3-indolyl-phosphate–nitro blue tetrazolium (BCIP–NBT) substrate was added and the set up was kept in the dark for five minutes. Then the reaction was stopped by adding distilled water.

For ultrasensitive assay, initial steps were the same as described earlier. After the addition of sample and the substrate-ALP conjugate, the membrane was washed by drop wise addition of washing buffer. For amplification, 5 µl of B-T (1:1000) was added drop wise and incubated for five minutes. The spots were washed again with drop wise addition of wash buffer. The area was sponged gently with tissue paper. Avidin–ALP conjugate (7 µl, diluted 500-fold) was added drop wise on to the spotted area and incubated for five minutes. The spotted area was washed with drop wise addition of wash buffer and sponged gently with tissue paper. Substrate solution (BCIP–NBT) was added (5 µl) and incubated in dark for two minutes and washed with distilled water to stop the reaction. The whole assay was performed at ambient temperature. The colour intensities of the sample spots were compared with those of reference standards or control spots either visually or by densitometry by using CCD camera.

The calibration curve for DDT was constructed by plotting percentage B/B ₀ values (x-axis) obtained by densitometric analysis against the DDT concentration (y-axis), where B ₀ is the density of the zero standard (no DDT) and B is the density of the DDT added spot. The limit of detection (LOD) was estimated at two standard deviations below the zero standard (n = 10). Double the LOD value was taken as the limit of quantification.

Absorbent pad size

Membranes were prepared, and the assays were carried out in the absence of DDT by the conventional protocol as described earlier. Whatman filter pads of different grades No. 1, No. 3, No. 20, No. 40 and hand made were studied for their efficacy in absorbing (faster flow through) and helping the faster assay. The reagents were added, and the colour intensities were measured by densitometry. Flow rate was observed by measuring the time required for the complete absorption of the applied fluid from the spotted area. The effect of the distance between the spots was also studied by applying 5 µl of copper sulphate solution (5% w/v) instead of the assay reagents and measuring the absorption areas in the filter paper.

Spiking of food samples

Food samples such as milk, fruit juices, carbonated beverages and vegetable samples were spiked with known concentration of DDT, mixed well and kept overnight at room temperature and at 4°C for stabilisation. Food samples without spiking with DDT served as control.

Extraction of food samples

The spiked and non-spiked food samples were adjusted to pH 2.0 with concentrated HNO₃, and extracted thrice with three volumes of methylenedichloride. The solvent fractions were pooled, passed over anhydrous sodium sulphate and evaporated to complete dryness. The samples were then suspended in phosphate buffer (50 mM, pH 7.5) containing DMF for immunoassay.

The samples after extraction were passed through florisil column for and the gas chromatography analyses were done as described earlier (Deepthi, Rastogi, & Manonmani, 2007).

Results and discussion

Immobilisation of antibody to membrane

We compared the assay response of nitrocellulose membranes (0.45 µm pore size) from Millipore, BioRad, Whatman and Sigma. Under identical conditions, all the membranes gave very low background colour in the unspotted areas. However, membranes from Bio-rad and Whatman produced distinctly higher spot intensities of zero standard leading to higher sensitivity. Therefore, the Bio-rad membranes were used in all subsequent experiments.

The overall performance of the assay depended on many factors such as dilutions of anti-DDT antibody, the enzyme conjugate and the B-T reagent and the incubation period in the amplification steps. Various parameters affecting the assay response were quantitatively investigated and optimised (Deepthi 2010). As the detection depends on the visualisation of the intensity of the coloured spots, the uniformity of spotting the antibody over the membrane surface is important. The spotted areas also must be sufficient to cover the volume of the reactants added subsequently. We have applied 5 µl of antibody and spread it to give a spot of approximately 5 mm diameter. Smaller or larger volumes of antibody may also be used, but we have not investigated it in detail. The minimum spot distance required between the adjacent spots from the experiments carried with copper sulphate was found to be 5 mm. It also showed the spot size on the membrane to be approximately 4 mm diameter. Decreasing the distance resulted in the merger of the absorption areas of the adjacent spots.

Membrane size

The size of the membrane was found not to be critical. The number of spotted areas could be increased depending on the handling capacity, number of samples to be analysed and the fastness with which one could do the assay. The size of the holes marked on the polyethylene card should match with that on the membrane, i.e. the size and the holes on the polyethylene card were also subject to change depending on that of the membrane. Cards of alternate material with adequate strength and compatibility could also be used. We used membrane circle with six to seven marked circles in a single test device.

Absorption pad size

The intensities of the spots on the nitrocellulose membrane obtained in the absence of DDT by using filter pads of different porosities are shown in Table 1. The results showed that the use of handmade filter sheet as filter pad below the nitrocellulose membrane gave least colour intensity as the added liquid passed through the set up very fast because the filter pad absorbed it very fast in the beginning. With progress in the reaction, the filter sheet could not absorb the material added, and this resulted in decreased colour intensity. The absorption of added reagents increased in Whatman filter pads. Whatman No. 3 and No. 40 gave almost the same colour intensities. The absorption of the added reagents was good by using these pads. Hence the reaction could be completed in short time. The adherence of the nitrocellulose membrane to the filter pads maintains the continuity of capillary channels with fluid receiving zone. However, care should be taken to see that no air bubble gets entrapped between the membrane and the filter pad because the air bubbles interfere with the fluid flow through the membrane. Both the membrane and the absorbent body should have intimate contact, so that the applied fluid is efficiently absorbed by the absorbent body through antibody-immobilised zones within few seconds without requiring any force. Because there is very little lateral diffusion, costly labelled reagents can be used efficiently.

Table 1. Influence of absorbent pad type on intensity spot without DDT.

	Spot intensity (arbitrary units) (1×10^4)
Absorbent pad (pore size)	A	B	C	D
(No. 1) 0.45 µm	10.55	10.11	11.06	11.11
No. 3	11.46	11.65	11.24	11.55
No. 20	8.04	8.12	8.30	8.25
No. 40	11.50	11.73	11.74	11.52
Hand made	9.01	8.20	7.85	7.93

Assay

To evaluate the performance of the analytical device, a calibration curve of DDT was constructed. The LOD using the conventional and ultrasensitive (signal-amplified) method was 1 ng ml⁻¹ and 1 pg ml⁻¹, respectively. The corresponding B/B ₀ values were 82.28 and 89.89%, respectively. The DDT concentrations of 1000 ng, 100 ng, 10 ng, 1 ng, 100 pg, 10 pg and 1 pg used by conventional method gave the mean percentage B/B ₀ values of 68.35±3.8, 74.68±4.1, 77.85±3.9, 82.28±4.2, 84.18±3.9, 89.24±4.4 and 94.94±4.3, respectively. In the ultra sensitive method, at the levels of 1000 ng, 100 ng, 10 ng, 1 ng, 100 pg, 10 pg and 1 pg, the mean percentage of B/B ₀ values were 62.92±3.6, 67.98±4.0, 71.91±4.6, 77.53±3.8, 80.34±4.9, 84.83±4.1 and 89.89±3.9, respectively.

Assay with in situ concentration of samples

Before performing the calibration studies of DDT for the pre-concentration assay, the effect of applied sample volume on the spot intensities was studied using zero standard. The spot intensities obtained using 2, 10 and 20 µl were approximately 60, 65 and 72% less, respectively, compared to the control spot (5 µl). A possible explanation could be that excess wetting of the absorbent body with the zero standard spreads the enzyme conjugate solution and also reduces its flow rate. The effect was more pronounced when a higher volume (50 µl) was used. When 50 µl of zero standard was added in 10 portions, the time taken was too long. But the colour intensities were comparable to that of the control. This showed that the analytical device could be used for the assay of diluted samples by pre-concentration.

Food samples analysis

Because the ultra sensitive (signal-amplified) method was sufficient to detect low quantities of DDT, the spiked samples were analysed in a batch of five extracted samples. The analytical device was prepared by coating the antibody in the marked areas, blocked and mounted on to the absorbent pad. The whole set up was fixed in the box and the reagents were provided for ready use of the device. The device was stored at 4°C until further use. The total time required for the assay including addition of sample, enzyme conjugate, washing steps and colour development steps (excluding extraction of the sample) was 30 minutes. The method has four incubation steps each of five minutes duration.

To assess the possible interference of the sample by constituents other than DDT, the hexane and the dichloromethane extracts of all food samples were used as control samples in the assay. It was observed that hexane treatment of the extracts was not necessary in any of the food samples including milk. The matrix interference of the samples was very low (Table 2). Proper dilution would prevent if any matrix effect is observed. In a study with aflatoxin detection, sample dilution was recommended to counter the matrix effect (Llamanen & Veijalainen, 1992; Pal & Dhar, 2004). The extraction efficiency of the spiked DDT from all the food samples was in the range of 92–96%.

Table 2. Matrix effect of dichloromethane extracts of non-spiked food samples.

Sample	B/B0 (%)
Milk	94.8	92.1	94.6	93.3
Water	96.5	98.4	94.4	95.1
Fruit juices	92.6	93.8	92.9	93.2
Carbonated beverages	93.3	92.8	95.2	94.5
Vegetables	96.3	95.8	95.1	95.7

The food samples spiked with DDT and extracted with dichloromethane gave recovery of spiked DDT ranging from 79 to 95% (Table 3). The extraction was most efficient with water sample. Carbonated beverage showed 79% recovery of DDT. The immunodetection done with spiked food samples indicated that the colour intensity as well as percentage B/B₀ value increased as the concentration of DDT in the sample decreased (Table 4). This trend is same to that of standard DDT, i.e. the colour intensity is inversely proportional to the concentration of the analyte.

Table 3. Recovery of DDT from spiked food samples.

Sample	DDT concentration spiked (µg ml^−1)	% Recovery
Water	1	95
Milk	1	88
Apple juice	1	85
Mango juice	1	81
Carbonated beverage	1	79
Vegetable	1	82

Table 4. Immunodetection of DDT from spiked food samples.

	B/B0 (%)
DDT concentration spiked	Water	Milk	Apple juice	Mango juice	Carbonated beverages	Vegetables
1 µg ml^−1	65.69±3.7	60.73±3.9	58.43±3.1	62.45±3.8	64.37±3.0	59.58±3.5
10 ng ml^−1	74.42±4.1	67.16±4.6	66.01±4.2	71.06±4.7	68.24±4.3	67.12±4.1
1 ng ml^−1	78.54±4.5	72.78±3.7	70.61±4.2	78.03±4.3	77.71±4.0	75.78±3.8
10 pg ml^−1	83.66±4.6	79.96±4.4	79.74±4.2	87.11±3.9	84.87±4.7	80.80±4.8
1 pg ml^−1	86.76±4.9	88.84±4.8	86.55±4.9	91.15±5.2	90.97±5.5	88.21±5.2
Unspiked sample	100.00	100.00	100.00	100.00	100.00	100.00

The stability of the reagents kept along with the analytical device at 4°C was examined up to six months by periodically assaying DDT standards. The results showed that the coated membrane and other reagents were stable for more than six months.

Conclusion

A new approach for performing immunoassay of DDT based on Dot-ELISA on a batch of samples has been demonstrated. The method is non-instrumental and does not require a pump to pull the reagents through the membrane. The absorbent pad allows the focussed absorption of the applied reagents through antibody immobilised zones in a very short time. Further amplification of the samples could be possible by using other amplifying systems. The data indicate that the method satisfactorily analyses DDT residues in a variety of food stuffs in the crude sample extracts without requiring any purification.

Acknowledgements

The authors thank the Director, CFTRI, Mysore and the Head, FTBE Department, for their constant encouragement during this work. Thanks to DBT for funding the project. One of the authors thanks CSIR for fellowship in the form of SRF.