Research article Received: 26 August 2014,
Accepted: 27 January 2015
Published online in Wiley Online Library: 24 March 2015
(wileyonlinelibrary.com) DOI 10.1002/bio.2892
Comparison of colorimetric and chemiluminescent enzyme-linked immunosorbent assay for the detection of endosulfan in food samples T. S. Deepak, Shenoy Rashmi and Haravey K. Manonmani* ABSTRACT: Pesticides have become part of food protection since their inception. Endosulfan, an organochlorine insecticide, has been used against insect pests such as whiteflies, aphids, red spiders and mites. Methods of immunochemical assays have been devised for the determination and analysis of pesticides and commonly used for the analysis of contaminants in food, water, soil and body fluids. Chicken IgY antibodies raised against endosulfan haptens were used for the detection of endosulfan. We have compared colorimetric (CO) and chemiluminescence (CL) enzyme-linked immunosorbent assay (ELISA) techniques for the detection of endosulfan isomers in a food matrix. CL ELISA assay was found to be more sensitive than CO assay. The mean recovery was 81.2–95.6% for α- and β-endosulfan-spiked food samples with 2.8–4.6% relative standard deviation. The detection of the endosulfan isomers was linear in the range 100 μg/mL–5 fg/mL, with a limit of detection at 100 μg/mL and 5 fg/mL for the CL ELISA method and 100 μg/mL and 1 ng/mL for the CO ELISA method respectively. These methods can be used for the rapid and reliable detection of organochlorine pesticide endosulfan. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: Endosulfan; colorimetric; chemiluminescence; IgY antibodies; ELISA
Introduction
1274
Endosulfan is a potent organochlorine insecticide and acaricide consisting of two isomers, endo and exo. It has been used for crop protection around the world to resist and repel of insects and pests including aphids, Colorado potato beetles, cabbage worms and leafhoppers (1). Owing to its unique mode of action, it has found use in pest resistance. As it is not specific, it can negatively affect populations of beneficial insects (2). It is known to be moderately toxic to honey bees (3) and is less toxic to bees than organophosphate insecticides (4). Endosulfan being a derivative of hexachlorocyclopentadiene is chemically similar to aldrin, chlordane and heptachlor. Technical endosulfan is a 7 : 3 mixture of stereoisomers, designated α and β. α- and β-Endosulfan are conformational isomers arising from the pyramidal stereochemistry of sulfur. α-Endosulfan is the more thermodynamically stable of the two. Thus, β-endosulfan irreversibly converts to the α-form, although the conversion is slow (5–7). Endosulfan is one of the most toxic pesticides on the market today, responsible for many incidents of fatal pesticide poisoning around the world (8). Endosulfan is also a xenoestrogen, which is a synthetic substance that imitates or enhances the effect of estrogens, and it can act as an endocrine disruptor, causing reproductive and developmental damage in both animals and humans. Whether endosulfan is a carcinogen is debated. Consumers’ intake of endosulfan from food residues can exceed acute reference doses (9) at short-term exposures. Endosulfan is acutely neurotoxic to both insects and mammals, including humans. The US EPA classifies it as Category I: “Highly Acutely Toxic” based on the LD50 value of 30 mg/kg for female rats. It is a GABA-gated chloride channel antagonist and a Ca2+,Mg2+-ATPase inhibitor. Endosulfan is semi-volatile and persistent to degradation processes in the environment. The
Luminescence 2015; 30: 1274–1279
spectrophotometric method is based on the reaction of endosulfan with 4-(4-nitrobenzyl)pyridine in an alkaline medium to form a pink complex with an absorption maximum at 520 nm with a detection minimum of 2 mg endosulfan (10). The most used method has been gas chromatography (GC) and GC/tandem mass spectrometry with an electron capture detector. Analysis by GC requires a highly purified sample, and the number of samples that could be analyzed is limited because of the time required for each sample. Hence, it is imperative to develop a specific, sensitive, high-throughput method capable of analysis of many samples in a single run. We have addressed this and present a comparative study of both a colorimetric (CO)- and a chemiluminescent (CL)-based assay for the detection of endosulfan.
Experimental Reagents Rabbit antichicken antibody-horseradish peroxidase (HRP) (enzyme-linked secondary antibody), endosulfan isomers and other pesticides, bovine serum albumin (BSA), ovalbumin (OVA) gelatin, Freund’s complete and incomplete adjuvants, chemicals and substrates for HRP, i.e. 3,3′,5,5′-tetramethylbenzidine (TMB), * Correspondence to: Haravey K. Manonmani, Department of Fermentation Technology and Bioengineering, CSIR-Central Food Technological Research Institute, Mysore, 570 020, Karnataka, India. E-mail: [email protected] Department of Fermentation Technology and Bioengineering, CSIR-Central Food Technological Research Institute, Mysore, Karnataka, India
Copyright © 2015 John Wiley & Sons, Ltd.
Colorimetric and Chemiluminescent ELISA for detection of endosulfan were purchased from Sigma-Aldrich Chemical Company Ltd. (Bangalore, India). Chemicals of analytical grade were used that were purchased from E. Merck (Bengaluru, India), and SISCO (Chennai, India). Endosulfan stock solution (1000 p.p.m.) was prepared in acetone, and appropriate serial dilutions were made (1000 p.p.b. to 10 p.p.t.) every day in phosphate-buffered saline (PBS) containing dimethylformamide at a 1 : 1 ratio. Endosulfan–OVA conjugate dilutions were made in PBS. Synthesis of hapten for the production of IgY Hapten for endosulfan detection was synthesized as described in the literature (10). Conjugation of haptens to proteins The haptens were conjugated to three carrier proteins: BSA, OVA and alkaline phosphatase. The hapten–protein conjugates were prepared using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide for the conjugation reactions (4). Immunization of poultry White leghorn poultry birds were immunized via the intramuscular route in the breast muscle initially with the immunogen–BSA conjugate in Freund’s complete adjuvant. One-tenth of LD50 was used for immunization. Booster injections were given at 15-day intervals in Freund’s incomplete adjuvant up to 6 months. Eggs were collected daily, and antibodies were harvested from egg yolk (11). Isolation of antibodies Egg yolk that contains the required antibody (IgY) was separated from albumin and yolk was suspended in 40 mL PBS/yolk, and the yolk was mixed to form a uniform suspension. To this, 10 mL of chloroform/yolk was added, and mixing was continued for 30 min. The mixture was centrifuged at 10,000 g for 15 min at 4°C. Antibody was precipitated with the addition of polyethylene glycol 6000 at 14% (v/w) to the supernatant. Mixing was continued for another 30 min to precipitate the antibody completely. Then antibody precipitate was separated by centrifuging at 10,000 g for 15 min at 4°C. The antibody precipitate was then redissolved in PBS buffer and stored at –20°C until use. Protein and antibody titer were obtained using the isolated antibody (11). Endosulfan detection by colorimetric enzyme-linked immunosorbent assay
Luminescence 2015; 30: 1274–1279
Chemiluminescent enzyme-linked immunosorbent assay method The assay procedure was the same as that in the CO assay. Instead of the TMB-H2O2, luminol was used as a substrate. The plates used for the CL assay were analyzed by measuring luminescence in a Varioskan Flash microplate reader (Thermoscientific, Bangalore, India). Data analysis A negative cut-off value was calculated for each sample dilution tested. To calculate the negative cut-off values the standard deviation was multiplied by three and the samples at each dilution were considered positive at readings greater than the negative cut-off value at the same dilution. The highest dilution was determined to be the minimum detection limit. The three positive replicate samples were used in determining this value. The Wilcoxon test (SPSS 15.0, Chicago, IL, USA) was used statistically to compare the minimum detection limits for the CO and the CL assays. The performance of these two assays in detecting endosulfan was compared using the results of the Wilcoxon test. This method provided a way to compare the results of each test directly and provided a method for distinguishing the end-point at which the assay detected only endosulfan. CO and CL ELISA tests were performed in replicate to determine the assay precision. The Pearson correlation coefficient was calculated for these two assay methods to determine the reliability of the developed assays. Calibration curve for endosulfan isomers The calibration curve for endosulfan isomers (α and β) was prepared by plotting the concentration of endosulfan isomers on the x-axis against the B/B0 value on the y-axis where B0 is the density of the zero standards (no endosulfan), and B is the amount of endosulfan added. Cross-reactivity Organochlorine pesticides such as 1,3-dichloropropene, 1,2,3trichloropropane, 1,4-dichlorobenzene, 1,2,4-trichlorobenzene, hexachlorobenzene, methoxychlor, diuron, aldrin, endrin and endosulfan, standards were used for specificity studies. Crossreactivity was calculated as the reduction in activity compared to the control (endosulfan isomers) and expressed as a percentage of relative activity. Pesticide stocks were prepared by dissolving standards at 1 mg/mL in acetone. Assay standards were prepared by diluting the stock solutions using buffer/dimethylformamide (1 : 1) in borosilicate glass tubes and used within 30 min of preparation. Assays were carried out as described above. Food sample analysis The food samples were spiked with α- and β-endosulfan separately and were extracted with dichloromethane. The spiked and nonspiked food samples were adjusted to pH 2.0 with concentrated HNO3, and extracted three times with three volumes of
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence
1275
The OVA conjugated hapten was immobilized on to the ELISA wells (NUNC, Roskilde, Denmark) at 4°C for 16 h. The conjugate was removed, and the wells were washed with PBS-Tween20. Then the vacant sites were blocked with 1% gelatin for 1 h. After being washed with PBS-Tween20, the mixture of antibody and α-endosulfan were added. The wells were incubated at 37°C for 1 h. Antichicken rabbit secondary antibody conjugated with HRP was added to the wells. Incubation was continued for another hour. The color was developed using the TMB-H2O2 as substrate. The blue colored product that formed was measured at 450 nm after stopping the reaction using 2 N H2SO4. All assays were performed in duplicate unless otherwise stated. The samples used in the plate for the CO assay were analyzed by measuring
absorbance at 450 nm. Negative control sera were added in duplicate in the same manner as the test samples.
T. S. Deepak et al. dichloromethane. The solvent fractions were pooled, passed over anhydrous sodium sulfate and evaporated to complete dryness. The samples were then suspended in phosphate buffer (50 mM, pH 7.5) containing dimethylformamide for immunoassay as described above. The samples after extraction were passed through an activated Florisil column, and analyzed by GC. Concentrated residual substrate was resuspended in a known volume of high-performance liquid chromatography grade acetone, and GC was done using a Shimadzu (Mumbai, India) gas chromatograph. One μL of the extract suspension was injected into the BP-5 capillary column (30 mm × 0.25 mm ID) set at 180°C and programmed as 180°C for 10 min and a rise at 2°C/min up to 220°C and maintained there for 2 min. Injector was maintained at 250°C while electron capture detector (Ni63) was maintained at 280°C. Pure nitrogen gas (IOLAR grade I) was used as the carrier at 1 mL/min. Quantification of endosulfan isomers in the sample was done using the area under the peak with reference to the standard under the same conditions.
Results and discussion Optimization of parameters for the detection of endosulfan isomers Endosulfan isomers (α and β) were used in competitive ELISA by both the CO and CL methods. Initially, the assay conditions of the CO ELISA and CL ELISA for endosulfan determination were optimized. By varying the concentrations of anti-endosulfan antibody and hapten–OVA conjugate, a set of calibration curves for endosulfan determination was obtained. During optimization of the antibody concentration necessary for immobilization, different concentrations of endosulfan-specific IgY antibody were used, and 2 mg (protein) of antibody showed the best results in terms of both CO measurements and CL signals (data not shown).
Figure 2. Competitive indirect enzyme-linked immunosorbent assay (chemiluminescent). HRP, horseradish peroxidase; OVA, ovalbumin.
surface (Figs. 1 and 2). Thus, the Optical Density (OD)/CL units are inversely proportional to the concentration of free endosulfan isomers in the sample. The results were compared with a negative control that gave maximum CO/CL signals. The CO ELISA could detect 1 ng of α-isomer and 1 ng of β-isomer (Figs. 3 and 4). The CL-based ELISA technique could detect α-isomer of endosulfan at 5 fg/mL levels and β-isomer of endosulfan at 5 fg/mL levels under optimized conditions. CL ELISA could detect as low as 5 fg of α- and β-endosulfan respectively while CO ELISA could detect as low as 1 ng of α- and β-endosulfan respectively. This result indicated that CL ELISA was more sensitive and could detect very low concentrations of α- and β-endosulfan.
Detection of endosulfan isomers by colorimetric and chemiluminescent enzyme-linked immunosorbent assay The detection of endosulfan isomers by CO ELISA and CL ELISA was based on an indirect competitive method. Competition between free endosulfan isomer molecules and the conjugate (hapten–HRP) for the antigen-binding sites in an immobilized antibody was the key reaction for the sensitive detection of endosulfan isomers in water samples. Hence, for less free endosulfan isomer concentration in the sample, more CO/CL was observed, indicating maximum binding of hapten–OVA conjugate to the
1276
Figure 1. Competitive indirect enzyme-linked immunosorbent assay (colorimetric). HRP, horseradish peroxidase; OVA, ovalbumin; TMB, 3,3′,5,5′-tetramethylbenzidine.
wileyonlinelibrary.com/journal/luminescence
Figure 3. Colorimetric detection of α-endosulfan.
Figure 4. Colorimetric detection of β-endosulfan.
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015; 30: 1274–1279
Colorimetric and Chemiluminescent ELISA for detection of endosulfan Specificity of antibody
Precision and accuracy of the assay
The chicken antibody (IgY) was specific to endosulfan. It was able to detect both isomers of endosulfan. Reaction with other chlorinated aromatics was very low (Table 1) by both the assay methods.
The inter- and intra-assay precisions were determined at different endosulfan isomer concentrations, 1 ng/mL–100 μg/mL for CO ELISA and 1 fg/mL–100 μg/mL for CL ELISA. The intra-assay precision was assessed by analyzing eight replicates of each sample in a single run, and interassay precision was assessed by analyzing the same by repeated analysis of samples of endosulfan isomers (Table 2). The coefficient of variation of five assays of each standard was satisfactory, with less precision at lower analyte (endosulfan isomer) concentrations.
Table 1. IC50 values of different pesticides SL. no.
Pesticide tested
IC50 value
α-Endosulfan β-Endosulfan 1,3-dichloropropene 1,2,3-trichloropropane Dichlorobenzene Trichlorobenzene Hexachlorobenzene Methoxychlor Diuron Aldrin Endrin
1 2 3 4 5 6 7 8 9 10 11
CO ELISA
CL ELISA
79.4 ng 86.21 ng 162.3 μg 338.2 μg 265.15 μg 71.93 μg 65.5 μg 229.43 μg 0 0 294.11 μg
97.51 pg 97 pg 168.24 μg 380.14 μg 268.72 μg 77.33 μg 67.83 μg 227.43 μg 298.31 μg
SL, serial number.
Table 2. Precision of CO ELISA and CL ELISA assays Concentration (/mL)
SD (CO ELISA) α-isomer B-isomer
1 fg 10 fg 100 fg 1 pg 10 pg 100 pg 1 ng 10 ng 100 ng 1 μg 10 μg 100 μg
ND ND ND ND ND ND 0.088 0.054 0.045 0.033 0.027 0.036
ND ND ND ND ND ND 0.091 0.048 0.044 0.026 0.035 0.042
SD (CL ELISA) α-isomer
B-isomer
0.077 0.054 0.045 0.041 0.042 0.021 0.032 0.028 0.021 0.034 0.052 0.083
0.081 0.061 0.056 0.035 0.035 0.034 0.042 0.029 0.032 0.043 0.064 0.091
CL, chemiluminescence; CO, colorimetric; ELISA, enzyme-linked immunosorbent assay; ND, not detected; SD, standard deviation.
Extraction of endosulfan isomers from spiked food samples The food samples spiked with endosulfan isomers and extracted with dichloromethane gave recoveries of spiked endosulfan isomers in the range 76–99% (Table 3). The extraction was most efficient with the water sample. Extraction from spiked chili showed low recovery of β-endosulfan. The recovery of endosulfan isomers by immunodetection using IgY and GC was comparable, having significant correlation (Table 3). Simulated sample analysis for spiked α- and β-endosulfan samples α- and β-Endosulfan extracted from spiked food samples were assayed using both the CO ELISA and CL ELISA methods. The recovery of the spiked α- and β-endosulfan samples was in the range 76–98% and 97–99% respectively for α- and β-endosulfan with 2.8–4.6% relative standard deviation. For the CL ELISA, the recovery ranged from 23 to 99% and 35 to 87% for α- and β-endosulfan respectively with 2.8–4.6% relative standard deviation. In both the methods, there was no matrix interference except for the sample of chili. Thus, using the proposed CO ELISA and CL ELISA methods, α- and β-endosulfan could be detected with linearity in the range 100 μg/mL–1 ng/mL, and 100 μg/mL–5 fg/mL respectively having a limit of detection (LOD) of 1 ng/mL and 5 fg/mL respectively for CO ELISA and CL ELISA (Tables 4 and 5). Merits of the present methods Quantitative analysis of endosulfan isomers can be achieved by both CO ELISA and CL ELISA techniques at high sensitivity (LOD 1 ng/mL and 10 fg/mL for α-isomer and 1 ng/mL and 10 fg/mL for β-isomer in water). Linearity was observed in the range of 100 μg/mL–1 ng/mL (CO ELISA) and 100 μg/mL–5 fg/mL (CL ELISA) with a good regression coefficient (R2) of 0.9233 (CO ELISA) and 0.9208 (CL ELISA) for α-isomer and 0.9422 (CO ELISA) and 0.9788 (CL ELISA) for β-isomer respectively (Figs. 5 and 6). The present
Table 3. Comparison of GC and immunodetection α-endosulfan 100 ng Recovery (%)
β-endosulfan 100 ng Recovery (%)
Immunodetection
GC
Immunodetection
GC
100.28 ± 3.04 103.27 ± 0.56 90.92 ± 3.61 89.13 ± 4.10
78.59 ± 0.130 86.76 ± 0.26 773.06 ± 0.082 86.36 ± 0.123
88.96 ± 2.91 98.72 ± 0.69 87.81 ± 0.824 100.01 ± 0.030
83.41 ± 0.727 76.77 ± 0.172 91.78 ± 0.37 85.18 ± 0.212
1277
GC, gas chromatography.
Luminescence 2015; 30: 1274–1279
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence
T. S. Deepak et al. Table 4. Percentage of recovery from foods (colorimetric enzyme-linked immunosorbent assay detection) after spiking Sl no. Food sample spiked 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Carrot Lemon Cabbage Pomegranate Onion Tomato Butter Bread Ladies Finger Ginger Garlic Beet root Apple Chitranna
Recovery of α-endosulfan (%)
Recovery of β-endosulfan (%)
76.96 91.74 94.78 98.66 97.21 90.26 142.97 79.21 97.29 97.91 97.34 98.83 91.44 95.128
99.28 99.46 98.17 98.07 99.08 97.62 216.29 99.64 99.79 99.01 98.32 99.8 98.21 98.3
SL, serial number.
Table 5. Percentage of recovery from foods (chemiluminescence enzyme-linked immunosorbent assay detection) after spiking Sl. no.
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Food sample spiked Onion Carrot Butter Lemon Green chili Bread Ginger Garlic Tomato Ladies Finger Beetroot Chitranna Apple Cabbage
Recovery (%)
Figure 6. Chemiluminescent detection of β-endosulfan.
technique has many advantages over the conventional technique such as GC in terms of cost per assay, analytical time, sensitivity and specificity. A total of 48 samples could be analyzed in a 96-well plate (in duplicate). Hence, the proposed technique proves a favorable tool for the screening of endosulfan isomers because of its high sensitivity, selectivity, simplicity, reliability and applicability. This technique can detect the target accurately and selectively, i.e. femtogram levels even in food samples without food matrix interference. Both techniques are suitable to ensure rapid, reliable and sensitive detection of food contaminants in biological samples.
Conclusions
α-endosulfan
β-endosulfan
77.84 82.97 84.16 74.91 239.84 65.89 31.66 75.29 99.16 54.44 79.11 155.145 28.67 36.9
86.47 84.03 68.62 47.63 23.06 54.2 35.44 66.49 84 73.12 76.621 138.7 45.58 43.94
The CO- and CL-based immuno-techniques were employed for the sensitive detection of endosulfan isomers with linearity in the range of 100 μg/mL–1 ng/mL, and an LOD of 1 ng/mL for CO ELISA and a linear range of 100 μg/mL–5 fg/mL for and an LOD of 5 fg/mL CL ELISA. The results obtained for the CO ELISA and the CL ELISA indicated that they could be effectively used as quantitative tests for the detection of endosulfan isomers and other organochlorine pesticides. This method is suitable for rapid and reliable detection of endosulfan isomers in water, food samples and environmental samples. The method provides preliminary, quantitative data, which can be applied for the analysis of environmental samples. These techniques may help in the biomonitoring of endosulfan isomers in a given sample without significant matrix interference. Acknowledgments
SL, serial number.
The authors thank the Director, CSIR-CFTRI, for providing facilities. The authors thank Dr. Muthukumar, Animal House Facility for his help during immunization of poultry. The authors are grateful to the Department of Science and Technology (DST), India, for providing funds and supporting the research work (project no. DST/TSG/ PT/2011/204).
References
1278
Figure 5. Chemiluminescent detection of α-endosulfan.
wileyonlinelibrary.com/journal/luminescence
1. US EPA. US Environmental Protection Agency Office of Water (4303T), US EPA, Re-registration Eligibility Decision for Endosulfan. 2002. http://www.epa.gov/oppsrrd1/REDs/endosulfan_red.pdf (accessed). 2. Robert LM. Insect control. In: Ullmann’s encyclopedia of industrial chemistry. Weinheim: Wiley-VCH, 2002. DOI: 10.1002/14356007 3. Mossler M, Aerts MJ, Nesheim ON. Florida Crop/Pest Management Profiles: Tomatoes. CIR 1238. The University of Florida, IFAS Extension. 2006. http://edis.ifas.ufl.edu/pdffiles/PI/PI03900.pdf (accessed January 27, 2009).
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015; 30: 1274–1279
Colorimetric and Chemiluminescent ELISA for detection of endosulfan 4. Gershoni JM, Bayer EA Wilchek M. Blot analyses of glycoconjugates: enzyme-hydrazide – a novel reagent for the detection of aldehydes. Anal Biochem 1985; 146: 59–63. 5. Pesticide Action Network North America. Speaking the truth saves lives in the Philippines and India. PAN Magazine 2006. http://www.panna.org/magazine/fall2006/featureSpeakingTruth. html (accessed January 27, 2009). 6. Pesticide residues in food. Pesticide residues in food 2006 – Joint FAO/WHO Meeting on Pesticide Residues. Food and Agriculture Organization (FAO). 2006. http://www.fao.org/home/en/ (accessed). 7. Sowbhagya HB, Visweswaraiah K Majumder SK. A rapid spectrophotometric method for the estimation of endosulfan. Pestic Sci 1984; 15: 571.
8. Schmidt WF, Hapeman CJ, Fettinger JC, Rice CP Bilboulian S. Structure and asymmetry in the isomeric conversion of β- to α-endosulfan. J Agric Food Chem 1997; 45(4): 1023–6. 9. Schmidt WF, Bilboulian S, Rice CP, Fettinger JC, McConnell LL Hapeman CJ. Thermodynamic, spectroscopic, and computational evidence for the irreversible conversion of β- to α-endosulfan. J Agric Food Chem 2001; 49(11): 5372–6. 10. Amita Rani BE, Roshini KR. Akmal Pasha. Karanth NGK. Immunoanalysis of endosulfan residues in spices. Food Agric Immunol 2003; 15(2): 105–13. 11. Mandappa IM, Manonmani HK. Immunodetection of Bacillus cereus haemolytic enterotoxin (HBL) in food samples. Anal Methods 2014; 6: 1841–7.
1279
Luminescence 2015; 30: 1274–1279
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence
Accepted: 27 January 2015
Published online in Wiley Online Library: 24 March 2015
(wileyonlinelibrary.com) DOI 10.1002/bio.2892
Comparison of colorimetric and chemiluminescent enzyme-linked immunosorbent assay for the detection of endosulfan in food samples T. S. Deepak, Shenoy Rashmi and Haravey K. Manonmani* ABSTRACT: Pesticides have become part of food protection since their inception. Endosulfan, an organochlorine insecticide, has been used against insect pests such as whiteflies, aphids, red spiders and mites. Methods of immunochemical assays have been devised for the determination and analysis of pesticides and commonly used for the analysis of contaminants in food, water, soil and body fluids. Chicken IgY antibodies raised against endosulfan haptens were used for the detection of endosulfan. We have compared colorimetric (CO) and chemiluminescence (CL) enzyme-linked immunosorbent assay (ELISA) techniques for the detection of endosulfan isomers in a food matrix. CL ELISA assay was found to be more sensitive than CO assay. The mean recovery was 81.2–95.6% for α- and β-endosulfan-spiked food samples with 2.8–4.6% relative standard deviation. The detection of the endosulfan isomers was linear in the range 100 μg/mL–5 fg/mL, with a limit of detection at 100 μg/mL and 5 fg/mL for the CL ELISA method and 100 μg/mL and 1 ng/mL for the CO ELISA method respectively. These methods can be used for the rapid and reliable detection of organochlorine pesticide endosulfan. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: Endosulfan; colorimetric; chemiluminescence; IgY antibodies; ELISA
Introduction
1274
Endosulfan is a potent organochlorine insecticide and acaricide consisting of two isomers, endo and exo. It has been used for crop protection around the world to resist and repel of insects and pests including aphids, Colorado potato beetles, cabbage worms and leafhoppers (1). Owing to its unique mode of action, it has found use in pest resistance. As it is not specific, it can negatively affect populations of beneficial insects (2). It is known to be moderately toxic to honey bees (3) and is less toxic to bees than organophosphate insecticides (4). Endosulfan being a derivative of hexachlorocyclopentadiene is chemically similar to aldrin, chlordane and heptachlor. Technical endosulfan is a 7 : 3 mixture of stereoisomers, designated α and β. α- and β-Endosulfan are conformational isomers arising from the pyramidal stereochemistry of sulfur. α-Endosulfan is the more thermodynamically stable of the two. Thus, β-endosulfan irreversibly converts to the α-form, although the conversion is slow (5–7). Endosulfan is one of the most toxic pesticides on the market today, responsible for many incidents of fatal pesticide poisoning around the world (8). Endosulfan is also a xenoestrogen, which is a synthetic substance that imitates or enhances the effect of estrogens, and it can act as an endocrine disruptor, causing reproductive and developmental damage in both animals and humans. Whether endosulfan is a carcinogen is debated. Consumers’ intake of endosulfan from food residues can exceed acute reference doses (9) at short-term exposures. Endosulfan is acutely neurotoxic to both insects and mammals, including humans. The US EPA classifies it as Category I: “Highly Acutely Toxic” based on the LD50 value of 30 mg/kg for female rats. It is a GABA-gated chloride channel antagonist and a Ca2+,Mg2+-ATPase inhibitor. Endosulfan is semi-volatile and persistent to degradation processes in the environment. The
Luminescence 2015; 30: 1274–1279
spectrophotometric method is based on the reaction of endosulfan with 4-(4-nitrobenzyl)pyridine in an alkaline medium to form a pink complex with an absorption maximum at 520 nm with a detection minimum of 2 mg endosulfan (10). The most used method has been gas chromatography (GC) and GC/tandem mass spectrometry with an electron capture detector. Analysis by GC requires a highly purified sample, and the number of samples that could be analyzed is limited because of the time required for each sample. Hence, it is imperative to develop a specific, sensitive, high-throughput method capable of analysis of many samples in a single run. We have addressed this and present a comparative study of both a colorimetric (CO)- and a chemiluminescent (CL)-based assay for the detection of endosulfan.
Experimental Reagents Rabbit antichicken antibody-horseradish peroxidase (HRP) (enzyme-linked secondary antibody), endosulfan isomers and other pesticides, bovine serum albumin (BSA), ovalbumin (OVA) gelatin, Freund’s complete and incomplete adjuvants, chemicals and substrates for HRP, i.e. 3,3′,5,5′-tetramethylbenzidine (TMB), * Correspondence to: Haravey K. Manonmani, Department of Fermentation Technology and Bioengineering, CSIR-Central Food Technological Research Institute, Mysore, 570 020, Karnataka, India. E-mail: [email protected] Department of Fermentation Technology and Bioengineering, CSIR-Central Food Technological Research Institute, Mysore, Karnataka, India
Copyright © 2015 John Wiley & Sons, Ltd.
Colorimetric and Chemiluminescent ELISA for detection of endosulfan were purchased from Sigma-Aldrich Chemical Company Ltd. (Bangalore, India). Chemicals of analytical grade were used that were purchased from E. Merck (Bengaluru, India), and SISCO (Chennai, India). Endosulfan stock solution (1000 p.p.m.) was prepared in acetone, and appropriate serial dilutions were made (1000 p.p.b. to 10 p.p.t.) every day in phosphate-buffered saline (PBS) containing dimethylformamide at a 1 : 1 ratio. Endosulfan–OVA conjugate dilutions were made in PBS. Synthesis of hapten for the production of IgY Hapten for endosulfan detection was synthesized as described in the literature (10). Conjugation of haptens to proteins The haptens were conjugated to three carrier proteins: BSA, OVA and alkaline phosphatase. The hapten–protein conjugates were prepared using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide for the conjugation reactions (4). Immunization of poultry White leghorn poultry birds were immunized via the intramuscular route in the breast muscle initially with the immunogen–BSA conjugate in Freund’s complete adjuvant. One-tenth of LD50 was used for immunization. Booster injections were given at 15-day intervals in Freund’s incomplete adjuvant up to 6 months. Eggs were collected daily, and antibodies were harvested from egg yolk (11). Isolation of antibodies Egg yolk that contains the required antibody (IgY) was separated from albumin and yolk was suspended in 40 mL PBS/yolk, and the yolk was mixed to form a uniform suspension. To this, 10 mL of chloroform/yolk was added, and mixing was continued for 30 min. The mixture was centrifuged at 10,000 g for 15 min at 4°C. Antibody was precipitated with the addition of polyethylene glycol 6000 at 14% (v/w) to the supernatant. Mixing was continued for another 30 min to precipitate the antibody completely. Then antibody precipitate was separated by centrifuging at 10,000 g for 15 min at 4°C. The antibody precipitate was then redissolved in PBS buffer and stored at –20°C until use. Protein and antibody titer were obtained using the isolated antibody (11). Endosulfan detection by colorimetric enzyme-linked immunosorbent assay
Luminescence 2015; 30: 1274–1279
Chemiluminescent enzyme-linked immunosorbent assay method The assay procedure was the same as that in the CO assay. Instead of the TMB-H2O2, luminol was used as a substrate. The plates used for the CL assay were analyzed by measuring luminescence in a Varioskan Flash microplate reader (Thermoscientific, Bangalore, India). Data analysis A negative cut-off value was calculated for each sample dilution tested. To calculate the negative cut-off values the standard deviation was multiplied by three and the samples at each dilution were considered positive at readings greater than the negative cut-off value at the same dilution. The highest dilution was determined to be the minimum detection limit. The three positive replicate samples were used in determining this value. The Wilcoxon test (SPSS 15.0, Chicago, IL, USA) was used statistically to compare the minimum detection limits for the CO and the CL assays. The performance of these two assays in detecting endosulfan was compared using the results of the Wilcoxon test. This method provided a way to compare the results of each test directly and provided a method for distinguishing the end-point at which the assay detected only endosulfan. CO and CL ELISA tests were performed in replicate to determine the assay precision. The Pearson correlation coefficient was calculated for these two assay methods to determine the reliability of the developed assays. Calibration curve for endosulfan isomers The calibration curve for endosulfan isomers (α and β) was prepared by plotting the concentration of endosulfan isomers on the x-axis against the B/B0 value on the y-axis where B0 is the density of the zero standards (no endosulfan), and B is the amount of endosulfan added. Cross-reactivity Organochlorine pesticides such as 1,3-dichloropropene, 1,2,3trichloropropane, 1,4-dichlorobenzene, 1,2,4-trichlorobenzene, hexachlorobenzene, methoxychlor, diuron, aldrin, endrin and endosulfan, standards were used for specificity studies. Crossreactivity was calculated as the reduction in activity compared to the control (endosulfan isomers) and expressed as a percentage of relative activity. Pesticide stocks were prepared by dissolving standards at 1 mg/mL in acetone. Assay standards were prepared by diluting the stock solutions using buffer/dimethylformamide (1 : 1) in borosilicate glass tubes and used within 30 min of preparation. Assays were carried out as described above. Food sample analysis The food samples were spiked with α- and β-endosulfan separately and were extracted with dichloromethane. The spiked and nonspiked food samples were adjusted to pH 2.0 with concentrated HNO3, and extracted three times with three volumes of
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence
1275
The OVA conjugated hapten was immobilized on to the ELISA wells (NUNC, Roskilde, Denmark) at 4°C for 16 h. The conjugate was removed, and the wells were washed with PBS-Tween20. Then the vacant sites were blocked with 1% gelatin for 1 h. After being washed with PBS-Tween20, the mixture of antibody and α-endosulfan were added. The wells were incubated at 37°C for 1 h. Antichicken rabbit secondary antibody conjugated with HRP was added to the wells. Incubation was continued for another hour. The color was developed using the TMB-H2O2 as substrate. The blue colored product that formed was measured at 450 nm after stopping the reaction using 2 N H2SO4. All assays were performed in duplicate unless otherwise stated. The samples used in the plate for the CO assay were analyzed by measuring
absorbance at 450 nm. Negative control sera were added in duplicate in the same manner as the test samples.
T. S. Deepak et al. dichloromethane. The solvent fractions were pooled, passed over anhydrous sodium sulfate and evaporated to complete dryness. The samples were then suspended in phosphate buffer (50 mM, pH 7.5) containing dimethylformamide for immunoassay as described above. The samples after extraction were passed through an activated Florisil column, and analyzed by GC. Concentrated residual substrate was resuspended in a known volume of high-performance liquid chromatography grade acetone, and GC was done using a Shimadzu (Mumbai, India) gas chromatograph. One μL of the extract suspension was injected into the BP-5 capillary column (30 mm × 0.25 mm ID) set at 180°C and programmed as 180°C for 10 min and a rise at 2°C/min up to 220°C and maintained there for 2 min. Injector was maintained at 250°C while electron capture detector (Ni63) was maintained at 280°C. Pure nitrogen gas (IOLAR grade I) was used as the carrier at 1 mL/min. Quantification of endosulfan isomers in the sample was done using the area under the peak with reference to the standard under the same conditions.
Results and discussion Optimization of parameters for the detection of endosulfan isomers Endosulfan isomers (α and β) were used in competitive ELISA by both the CO and CL methods. Initially, the assay conditions of the CO ELISA and CL ELISA for endosulfan determination were optimized. By varying the concentrations of anti-endosulfan antibody and hapten–OVA conjugate, a set of calibration curves for endosulfan determination was obtained. During optimization of the antibody concentration necessary for immobilization, different concentrations of endosulfan-specific IgY antibody were used, and 2 mg (protein) of antibody showed the best results in terms of both CO measurements and CL signals (data not shown).
Figure 2. Competitive indirect enzyme-linked immunosorbent assay (chemiluminescent). HRP, horseradish peroxidase; OVA, ovalbumin.
surface (Figs. 1 and 2). Thus, the Optical Density (OD)/CL units are inversely proportional to the concentration of free endosulfan isomers in the sample. The results were compared with a negative control that gave maximum CO/CL signals. The CO ELISA could detect 1 ng of α-isomer and 1 ng of β-isomer (Figs. 3 and 4). The CL-based ELISA technique could detect α-isomer of endosulfan at 5 fg/mL levels and β-isomer of endosulfan at 5 fg/mL levels under optimized conditions. CL ELISA could detect as low as 5 fg of α- and β-endosulfan respectively while CO ELISA could detect as low as 1 ng of α- and β-endosulfan respectively. This result indicated that CL ELISA was more sensitive and could detect very low concentrations of α- and β-endosulfan.
Detection of endosulfan isomers by colorimetric and chemiluminescent enzyme-linked immunosorbent assay The detection of endosulfan isomers by CO ELISA and CL ELISA was based on an indirect competitive method. Competition between free endosulfan isomer molecules and the conjugate (hapten–HRP) for the antigen-binding sites in an immobilized antibody was the key reaction for the sensitive detection of endosulfan isomers in water samples. Hence, for less free endosulfan isomer concentration in the sample, more CO/CL was observed, indicating maximum binding of hapten–OVA conjugate to the
1276
Figure 1. Competitive indirect enzyme-linked immunosorbent assay (colorimetric). HRP, horseradish peroxidase; OVA, ovalbumin; TMB, 3,3′,5,5′-tetramethylbenzidine.
wileyonlinelibrary.com/journal/luminescence
Figure 3. Colorimetric detection of α-endosulfan.
Figure 4. Colorimetric detection of β-endosulfan.
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015; 30: 1274–1279
Colorimetric and Chemiluminescent ELISA for detection of endosulfan Specificity of antibody
Precision and accuracy of the assay
The chicken antibody (IgY) was specific to endosulfan. It was able to detect both isomers of endosulfan. Reaction with other chlorinated aromatics was very low (Table 1) by both the assay methods.
The inter- and intra-assay precisions were determined at different endosulfan isomer concentrations, 1 ng/mL–100 μg/mL for CO ELISA and 1 fg/mL–100 μg/mL for CL ELISA. The intra-assay precision was assessed by analyzing eight replicates of each sample in a single run, and interassay precision was assessed by analyzing the same by repeated analysis of samples of endosulfan isomers (Table 2). The coefficient of variation of five assays of each standard was satisfactory, with less precision at lower analyte (endosulfan isomer) concentrations.
Table 1. IC50 values of different pesticides SL. no.
Pesticide tested
IC50 value
α-Endosulfan β-Endosulfan 1,3-dichloropropene 1,2,3-trichloropropane Dichlorobenzene Trichlorobenzene Hexachlorobenzene Methoxychlor Diuron Aldrin Endrin
1 2 3 4 5 6 7 8 9 10 11
CO ELISA
CL ELISA
79.4 ng 86.21 ng 162.3 μg 338.2 μg 265.15 μg 71.93 μg 65.5 μg 229.43 μg 0 0 294.11 μg
97.51 pg 97 pg 168.24 μg 380.14 μg 268.72 μg 77.33 μg 67.83 μg 227.43 μg 298.31 μg
SL, serial number.
Table 2. Precision of CO ELISA and CL ELISA assays Concentration (/mL)
SD (CO ELISA) α-isomer B-isomer
1 fg 10 fg 100 fg 1 pg 10 pg 100 pg 1 ng 10 ng 100 ng 1 μg 10 μg 100 μg
ND ND ND ND ND ND 0.088 0.054 0.045 0.033 0.027 0.036
ND ND ND ND ND ND 0.091 0.048 0.044 0.026 0.035 0.042
SD (CL ELISA) α-isomer
B-isomer
0.077 0.054 0.045 0.041 0.042 0.021 0.032 0.028 0.021 0.034 0.052 0.083
0.081 0.061 0.056 0.035 0.035 0.034 0.042 0.029 0.032 0.043 0.064 0.091
CL, chemiluminescence; CO, colorimetric; ELISA, enzyme-linked immunosorbent assay; ND, not detected; SD, standard deviation.
Extraction of endosulfan isomers from spiked food samples The food samples spiked with endosulfan isomers and extracted with dichloromethane gave recoveries of spiked endosulfan isomers in the range 76–99% (Table 3). The extraction was most efficient with the water sample. Extraction from spiked chili showed low recovery of β-endosulfan. The recovery of endosulfan isomers by immunodetection using IgY and GC was comparable, having significant correlation (Table 3). Simulated sample analysis for spiked α- and β-endosulfan samples α- and β-Endosulfan extracted from spiked food samples were assayed using both the CO ELISA and CL ELISA methods. The recovery of the spiked α- and β-endosulfan samples was in the range 76–98% and 97–99% respectively for α- and β-endosulfan with 2.8–4.6% relative standard deviation. For the CL ELISA, the recovery ranged from 23 to 99% and 35 to 87% for α- and β-endosulfan respectively with 2.8–4.6% relative standard deviation. In both the methods, there was no matrix interference except for the sample of chili. Thus, using the proposed CO ELISA and CL ELISA methods, α- and β-endosulfan could be detected with linearity in the range 100 μg/mL–1 ng/mL, and 100 μg/mL–5 fg/mL respectively having a limit of detection (LOD) of 1 ng/mL and 5 fg/mL respectively for CO ELISA and CL ELISA (Tables 4 and 5). Merits of the present methods Quantitative analysis of endosulfan isomers can be achieved by both CO ELISA and CL ELISA techniques at high sensitivity (LOD 1 ng/mL and 10 fg/mL for α-isomer and 1 ng/mL and 10 fg/mL for β-isomer in water). Linearity was observed in the range of 100 μg/mL–1 ng/mL (CO ELISA) and 100 μg/mL–5 fg/mL (CL ELISA) with a good regression coefficient (R2) of 0.9233 (CO ELISA) and 0.9208 (CL ELISA) for α-isomer and 0.9422 (CO ELISA) and 0.9788 (CL ELISA) for β-isomer respectively (Figs. 5 and 6). The present
Table 3. Comparison of GC and immunodetection α-endosulfan 100 ng Recovery (%)
β-endosulfan 100 ng Recovery (%)
Immunodetection
GC
Immunodetection
GC
100.28 ± 3.04 103.27 ± 0.56 90.92 ± 3.61 89.13 ± 4.10
78.59 ± 0.130 86.76 ± 0.26 773.06 ± 0.082 86.36 ± 0.123
88.96 ± 2.91 98.72 ± 0.69 87.81 ± 0.824 100.01 ± 0.030
83.41 ± 0.727 76.77 ± 0.172 91.78 ± 0.37 85.18 ± 0.212
1277
GC, gas chromatography.
Luminescence 2015; 30: 1274–1279
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence
T. S. Deepak et al. Table 4. Percentage of recovery from foods (colorimetric enzyme-linked immunosorbent assay detection) after spiking Sl no. Food sample spiked 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Carrot Lemon Cabbage Pomegranate Onion Tomato Butter Bread Ladies Finger Ginger Garlic Beet root Apple Chitranna
Recovery of α-endosulfan (%)
Recovery of β-endosulfan (%)
76.96 91.74 94.78 98.66 97.21 90.26 142.97 79.21 97.29 97.91 97.34 98.83 91.44 95.128
99.28 99.46 98.17 98.07 99.08 97.62 216.29 99.64 99.79 99.01 98.32 99.8 98.21 98.3
SL, serial number.
Table 5. Percentage of recovery from foods (chemiluminescence enzyme-linked immunosorbent assay detection) after spiking Sl. no.
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Food sample spiked Onion Carrot Butter Lemon Green chili Bread Ginger Garlic Tomato Ladies Finger Beetroot Chitranna Apple Cabbage
Recovery (%)
Figure 6. Chemiluminescent detection of β-endosulfan.
technique has many advantages over the conventional technique such as GC in terms of cost per assay, analytical time, sensitivity and specificity. A total of 48 samples could be analyzed in a 96-well plate (in duplicate). Hence, the proposed technique proves a favorable tool for the screening of endosulfan isomers because of its high sensitivity, selectivity, simplicity, reliability and applicability. This technique can detect the target accurately and selectively, i.e. femtogram levels even in food samples without food matrix interference. Both techniques are suitable to ensure rapid, reliable and sensitive detection of food contaminants in biological samples.
Conclusions
α-endosulfan
β-endosulfan
77.84 82.97 84.16 74.91 239.84 65.89 31.66 75.29 99.16 54.44 79.11 155.145 28.67 36.9
86.47 84.03 68.62 47.63 23.06 54.2 35.44 66.49 84 73.12 76.621 138.7 45.58 43.94
The CO- and CL-based immuno-techniques were employed for the sensitive detection of endosulfan isomers with linearity in the range of 100 μg/mL–1 ng/mL, and an LOD of 1 ng/mL for CO ELISA and a linear range of 100 μg/mL–5 fg/mL for and an LOD of 5 fg/mL CL ELISA. The results obtained for the CO ELISA and the CL ELISA indicated that they could be effectively used as quantitative tests for the detection of endosulfan isomers and other organochlorine pesticides. This method is suitable for rapid and reliable detection of endosulfan isomers in water, food samples and environmental samples. The method provides preliminary, quantitative data, which can be applied for the analysis of environmental samples. These techniques may help in the biomonitoring of endosulfan isomers in a given sample without significant matrix interference. Acknowledgments
SL, serial number.
The authors thank the Director, CSIR-CFTRI, for providing facilities. The authors thank Dr. Muthukumar, Animal House Facility for his help during immunization of poultry. The authors are grateful to the Department of Science and Technology (DST), India, for providing funds and supporting the research work (project no. DST/TSG/ PT/2011/204).
References
1278
Figure 5. Chemiluminescent detection of α-endosulfan.
wileyonlinelibrary.com/journal/luminescence
1. US EPA. US Environmental Protection Agency Office of Water (4303T), US EPA, Re-registration Eligibility Decision for Endosulfan. 2002. http://www.epa.gov/oppsrrd1/REDs/endosulfan_red.pdf (accessed). 2. Robert LM. Insect control. In: Ullmann’s encyclopedia of industrial chemistry. Weinheim: Wiley-VCH, 2002. DOI: 10.1002/14356007 3. Mossler M, Aerts MJ, Nesheim ON. Florida Crop/Pest Management Profiles: Tomatoes. CIR 1238. The University of Florida, IFAS Extension. 2006. http://edis.ifas.ufl.edu/pdffiles/PI/PI03900.pdf (accessed January 27, 2009).
Copyright © 2015 John Wiley & Sons, Ltd.
Luminescence 2015; 30: 1274–1279
Colorimetric and Chemiluminescent ELISA for detection of endosulfan 4. Gershoni JM, Bayer EA Wilchek M. Blot analyses of glycoconjugates: enzyme-hydrazide – a novel reagent for the detection of aldehydes. Anal Biochem 1985; 146: 59–63. 5. Pesticide Action Network North America. Speaking the truth saves lives in the Philippines and India. PAN Magazine 2006. http://www.panna.org/magazine/fall2006/featureSpeakingTruth. html (accessed January 27, 2009). 6. Pesticide residues in food. Pesticide residues in food 2006 – Joint FAO/WHO Meeting on Pesticide Residues. Food and Agriculture Organization (FAO). 2006. http://www.fao.org/home/en/ (accessed). 7. Sowbhagya HB, Visweswaraiah K Majumder SK. A rapid spectrophotometric method for the estimation of endosulfan. Pestic Sci 1984; 15: 571.
8. Schmidt WF, Hapeman CJ, Fettinger JC, Rice CP Bilboulian S. Structure and asymmetry in the isomeric conversion of β- to α-endosulfan. J Agric Food Chem 1997; 45(4): 1023–6. 9. Schmidt WF, Bilboulian S, Rice CP, Fettinger JC, McConnell LL Hapeman CJ. Thermodynamic, spectroscopic, and computational evidence for the irreversible conversion of β- to α-endosulfan. J Agric Food Chem 2001; 49(11): 5372–6. 10. Amita Rani BE, Roshini KR. Akmal Pasha. Karanth NGK. Immunoanalysis of endosulfan residues in spices. Food Agric Immunol 2003; 15(2): 105–13. 11. Mandappa IM, Manonmani HK. Immunodetection of Bacillus cereus haemolytic enterotoxin (HBL) in food samples. Anal Methods 2014; 6: 1841–7.
1279
Luminescence 2015; 30: 1274–1279
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence
