Development of determination of di-n-octyl phthalate (DOP) residue by an indirect enzyme-linked immunosorbent assay

Abstract

In this research, the hapten di-n-octyl 4-aminophthalate (DOAP) was designed and synthesised successfully. It was used to couple with carrier proteins, bovine serum albumin (BSA) and ovalbumin (OVA) by diazotization reaction for immunogen (DOP–BSA) and coating antigen (DOP–OVA), respectively. Rabbits were immunised with DOP–BSA; polyclonal antiserum was raised and determined by competitive indirect enzyme-linked immunosorbent assay (ciELISA). After optimisation, a ciELISA was established. The quantitative working range for DOP was 5–75 ng/mL with the detection limit of 1.9±0.1 ng/mL and the IC₅₀ of 19.2±1.1 ng/mL. The optimised ELISA had cross-reactivity of 22.6%, 17.6% and 21.2% with di-iso-octyl phthalate (DIOP), di-n-butyl phthalate (DBP) and di-hexyl phthalate (DHP), respectively. The result of the detection of polyvinyl chloride (PVC) samples showed that the immunoassay we developed had high-accuracy contrast with high-performance liquid chromatography electrospray ionisation tandem mass spectrometry analysis and it could be qualified to determine di-n-octyl phthalate (DOP) residue in PVC samples.

Keywords:

• phthalates
• residues
• hapten synthesis
• ciELISA
• polyvinyl chloride
• immunoassay

Introduction

Phthalates, primarily as additives to polyvinyl chloride (PVC) plastics, are widely existent environment chemical pollutants in our daily lives. Extensive use of these chemicals resulted in their appearance in various environmental matrices, such as water (drinking water, surface water and waste water), perfume, cosmetic, dust, air, sludge and soil.

Phthalates, also known as endocrine disrupting chemicals (EDC) (Castillo & Barcelo, 1997) have endocrine-disrupting properties. It showed that phthalates had estrogenic effects that disrupt the pathways of testicular differentiation in genetically male animals (Higuchi, Palmer, Gray, & Veeramachaneni, 2003; Jobling, Reynolds, White, Parker, & Sumper, 1995; Moore, Rudy, Lin, Ko, & Peterson, 2001; Mylchreest, Cattley, & Foster, 1998) and therefore had reproductive effects (Gray et al., 2000). As one of the phthalates, di-n-octyl phthalate (DOP) can lead to serious health problems. In 2005, the European Union banned the use of six phthalates including DOP in child products. In addition, DOP was also listed as priority pollutant by the World Health Organization (WHO) and China.

Several methods have been established for determination of phthalates in different environmental matrices. The analysis of phthalates is mostly performed by gas chromatography (GC) and liquid chromatography (LC) (Brossa, Marce, Borrull, & Pocurull, 2005; Luks-Betlej, Popp, Janozska, & Paschke, 2001; Penalver, Pocurull, Borrull, & Marce, 2000; Polo, Llompart, GarcIa-Jares, & Cela, 2005; Silva et al., 2004). The most common and important detector for phthalate analysis is mass spectrometric detection (Castillo, Oubina, & Barcelo, 1998; Gimeno, Marc, & Borrull, 2003; Kato, Silva, Needham, & Calafat, 2005). These traditional instrument analysis methods are often extremely time-consuming and expensive, and therefore not suitable for routine analysis of a large number of samples.

Enzyme-linked immunoassay and sensor-based immunoassay are more efficient, cost-effective, simple and reliable methods for identifying and quantitating the presence of the target analyte in samples (Chen et al., 2009; Hao et al., 2009; Li et al., 2008; Ma et al., 2009; Peng et al., 2009; Xie et al., 2009; Xu, Peng, & Xu, 2008; Xu, Yu, Chu, Peng, & Jin, 2006; Yue et al., 2009). Several immunoassays for phthalates have been developed (Goda et al., 2000; Ius, Bacigalupo, Meroni, Pistill, & Roda, 1993; Zhang, Wang, & Zhuang, 2006, 2007). In this research, we developed sensitive and quantitative indirect enzyme immunoassay for DOP. After the optimisation, we validated the immunoassay for the analysis of DOP in PVC samples and achieved a good agreement with high-performance liquid chromatography electrospray ionisation tandem mass spectrometry (HPLC-ESI-MS/MS) analysis.

Experimental

Reagents and materials

DOP, di-iso-octyl phthalate (DIOP), di-hexyl phthalate (DHP), di-n-butyl phthalate (DBP), n-octanol and reduced iron powder were obtained from Sinopharm Chemical Reagent Company (Beijing, China). 4-nitrophthalic acid (4-NPA) was obtained from Taixing Sunmy Fine Chemical Company (Taixing, China). Bovine serum albumin (BSA, MW 67,000) and Ovalbumin (OVA, MW 45,000) were purchased from Boao Biotechnology Company (Shanghai, China). Complete and incomplete Freund's adjuvant, enzyme immunoassay-grade HRP-labelled goat anti-rabbit immunoglobulin and Tween-20 were obtained from Sigma (St. Louis, MO, USA). All the other reagents were of analytical grades.

Thin-layer chromatography (TLC) was performed on 0.2 mm coated silica gel F254 on glass sheets (Qingdao, Shandong, China). The dialysis membrane (MW cutoff 12,000–14,000) was obtained from Huamei Biotechnology Company (Luoyang, Henan, China). Polystyrene 96-well microtiter plates were obtained from Costar (Corning, MA, USA).

Instruments

Multiskan microplate reader was obtained from Thermo Labsystem (Helsinki, Finland). Accela HPLC and TSQ Quantum Access mass spectrometer were obtained from Thermo Electron Corporation (Shanghai, China).

Hapten synthesis and characterisation

Since DOP does not have any suitable functional group to be coupled with carrier proteins, we designed and synthesised the hapten with amine group according to the following procedures. The synthetic route for this hapten is illustrated in Figure 1.

Figure 1.  Synthetic route for hapten DOAP: (a) di-n-octyl 4-nitrophthalate (DONP) and (b) di-n-octyl 4-aminephthalate (DOAP).

Synthesis of di-n-octyl 4-nitrophthalate (DONP): 21.1 mg (0.12 mol) of 4-NPA was added to a 100-ml volume three-neck flask and then 13.02 mg (0.1 mol) of n-octanol and 6 ml of benzene were added. After that, concentrated sulphuric acid (1.6 ml) was added slowly to the above stirred solution and it was heated to reflux (120°C) for 2 h, and the solvent was evaporated under reduced pressure. The oily residue obtained was added into cold water and the solid mixture was washed with 10% aqueous Na₂CO₃. The crude product was obtained from the upper layer. The resulting dark-red crystals of DONP were recrystallised from ethanol. TLC: R _f=0.89 (acetic acid: n-hexane 1:15). UV: λ_max=253 nm. LC-MS: positive ion scan mode, peak 1: [M + H]⁺=436; peak 2: [M + Na]⁺=458. It had a good agreement with DONP molecular weight of 435 (Figure 2).

Figure 2.  LC-MS analysis of DOAP: (a) the positive ions LC spectrum of DONP and (b) the mass spectrum of DONP.

Synthesis of hapten (di-n-octyl 4-aminephthalate – DOAP): 392 mg (7 mmol) of reduced iron powder was added to a 50-ml volume round flask and then 2 ml of water and 1 mmol of ammonium chloride were added. After heated in a water bath for 15 min, 609 mg (1.4 mmol) of DONP dissolved in 16.8 ml methanol was added to the above solution and heated to reflux for 5 h. Before the reaction solution was cooled to room temperature, it was filtered to remove the remaining iron powder. The above filtrate was distilled under reduced pressure to get the salmon pink crude product. After the crude product dissolved in hot methanol was cooled to low temperature, 473.8 mg (1.17 mmol) of DOAP was obtained. TLC: R _f=0.65 (acetic acid: n-hexane 1:15). UV: λ_max1=232 nm, λ_max2=288 nm. The LC spectrum identified that the DOAP we synthesised had a purity of more than 99%. LC-MS: positive ion scan mode, the peak of molecular weight of 406 for mass spectrum could be interpreted as molecular weight of DOAP plus one hydrogen ion, that is [M + H]⁺=406. It had a good agreement with dioctyl 4-aminephthalate molecular weight of 405 (Figure 3).

Figure 3.  LC-MS analysis of DOAP: (a) the positive ions LC spectrum of DOAP and (b) the mass spectrum of DOAP.

Antigen synthesis

DOAP of 93.2 mg (0.23 mmol) was dissolved in a mixture of 2 ml DMF with 0.2 ml of 1 mol/L hydrochloric acid and cooled to 4°C. Then, 0.14 ml of 1 mol/L NaNO₂ solution was added in drops and stirred for 1 h at 4°C. The reaction solution was monitored with potassium iodide starch paper. The NaNO₂ solution was stopped being added to the above solution when the potassium iodide starch paper turned blue-grey. After that, BSA (150 mg, 0.023 mmol) or OVA (102 mg, 0.023 mmol) in 6 ml of borate buffer (pH = 9.20) was added dropwise. The pH of the reaction solution was adjusted to 9.20 by adding 1 mol/L sodium hydroxide solution. The mixture was stirred at 4°C overnight. The conjugate was dialysed in water that was changed with fresh water twice a day for three days at 4°C.

Immunisation

Two New Zealand white rabbits (1#, 2#) weighing about 1.5 kg were chosen for raising polyclonal antibodies. Routinely, 2 mg of immunogen dissolved in 1 ml of physiological saline was emulsi?ed with 1 ml Freund's complete adjuvant (1:1, v/v) and injected subcutaneously at multiple sites for the primary immunisation. For booster immunisations, 1 mg of the immunogen emulsified with Freund's incomplete adjuvant were given (1:1, v/v) after 2 weeks intramuscularly. One week after each booster immunisation, the rabbit was bled through ear vein. The blood was allowed to coagulate for about 1 h at room temperature and then stored at 4°C overnight. The antiserum was isolated by centrifugation (4000 rpm, 15 min) and sodium azide was added as a preservative at a final concentration of 0.02%. Antiserum was then divided into aliquots and stored at −70°C until use (Peng et al., 2008; Wang, Peng, Chen, & Xu, 2008).

Screening of antiserum

Checkerboard assays were used to select appropriate concentrations of coating antigens and antibodies, and to measure affinity of antibodies. We employed the assay similar to conventional procedure (Peng, Xu, Jin, Chu, & Wang, 2006; Tian et al., 2008). Microplates were coated with DOP–OVA (1/100–1/25,600, 100 µL/well) in carbonate buffer (50 mM, pH 9.6) at 4°C overnight. The microplates were washed three times with PBST (10 mM PBS containing 0.05% Tween-20, pH 7.4) and were blocked by 0.1% gelatine in carbonate buffer (50 mM, pH 9.6, 200 µL/well) for 2 h at 37°C. After another washing step, 100 µL/well of antiserum previously diluted (1/500–1/9600) with PBST (containing 0.1% gelatine) was added. After incubation at 37°C for 30 min, the plates were washed and the goat anti-rabbit IgG–HRP conjugate (diluted 1:3000 in PBST with 0.1% gelatine 100 µL/well) was added. The mixture was incubated at 37°C for 30 min, and after one more washing step, 100 µL/well of a TMB solution (1 ml of 0.06% TMB–Glycol and 5 ml citrate–acetate buffer, pH 5.5) was added and incubated at 37°C for 30 min. The reaction was stopped by adding 100 µL of 2 M H₂SO₄ and absorbance was read at 450 nm.

Enzyme-linked immunosorbent assay (ELISA) procedure

The procedure of the competitive assay was described similarly by our group (Li et al., 2009a, 2009b; Peng, Chen, Chen, Xu, & Jin, 2008). The assay conditions were the same as the procedure described above with only a small change in the competition step: 100 µL/well of antiserum was replaced by 50 µL/well of antiserum and 50 µL/well of standard analyte. The standard analyte concentrations ranged from 0.5 to 500 ng/ml. The absorbance was measured at 450 nm and recorded. Standard curves were obtained by plotting absorbance against the analyte concentration.

Optimisation of enzyme-linked immunosorbent assay (ELISA) conditions

A set of immunoassay parameters (coating method, methanol content, ionic strength, pH of solvent, competitive time, blocking time, colouration time) were studied sequentially to improve sensitivity and stability of the immunoassay. These experiments were carried out by using the indirect protocol described above. The antibody from 1# rabbit diluted to 1/1600 in PBST and the coating antigen DOP–OVA diluted to 1/2000 was chosen as the working concentration. The criteria used to evaluate immunoassay performance were maximum absorbance (A_max), IC₅₀ and A_max/IC₅₀.

Effect of coating method

Three coating methods were employed (4°C, overnight; 37°C, 2 h; 37°C, 2 h then 4°C overnight) for the test.

Effect of the ionic strength

As to ELISA, ionic strength of the solution was an important factor. Thus, in our protocol, different PBS solutions (5, 10, 20, 40 mM PBS) were used as ELISA solution system.

Effect of the pH

The pH of the solution was another important factor for ELISA. PBS solutions ranging from pH 6.0 to 9.5 were used as the working solutions of the competitive immunoassay.

Effect of the methanol content

Various water–methanol concentration solvents were used to dissolve DOP for assay optimisation. For this test, standard DOP solutions were prepared in PBS containing methanol of varying concentration levels (5, 10, 20 and 40%) to improve the sensitivity.

Effect of competition and colouration time

As to immunoassay, the time for competitive and colouration step was an important factor that could affect the sensitivity and stability. Herein, we employed different competition time (0.5, 1 and 2 h) and colouration time (10, 20 and 30 min) for the optimisation of immunoassay reaction time.

Determination of cross-reactivity

Several phthalates were tested for cross-reactivity using the indirect ELISA procedure described above. The cross-reactivity values were calculated by using the formula: cross-reactivity%=(IC₅₀ of DOP/IC₅₀ of related compound)×100.

Analysis of real samples

The PVC sample with a thickness of 0.5–2 mm was cut into 5×5 mm fragment. PVC sample of 0.5–5.0 g of above 5×5 mm was weighed and wrapped with filter paper. Then it was put into a soxhlet extractor and extracted with organic solvent (CHCl₃:CH₃OH – 2:1) in water bath at 84±2°C for about 7.5 h. The extracted solution was transferred to a measuring flask and dissolved by methanol with a total volume of 25 ml. Then the solution was centrifuged at 17,000 rpm for 10 min and the supernatant was used for ELISA and HPLC-ESI-MS/MS analysis (Li et al., 2009a, 2009b).

Results and discussion

Hapten design and characterisation

The synthesis of appropriate hapten is the initial and critical step for the establishment of effective ELISA for low molecular weight chemicals, which affects the specificity and sensitivity of antibodies for the target analyte (Kuang et al., 2009). In designing hapten, it is preferable to retain functional groups that are unique for the target analyte. DOP has apparently two important antigenic determinants that are n-octanol group and aromatic ring. We selected the 4-NPA to synthesise DOAP as the hapten that would mostly expose the two important antigenic determinants and provide the desired amino group for protein conjugation. However, to make sure the two carboxyl groups of 4-NPA react with n-octanol completely, we employed acidic condition to heated to reflux (120°C) for 2 h to carry out esterification reaction.

Screening of antiserum

The antiserum of terminal rabbit blood was screened by using coating antigen (DOP–OVA) by a checkerboard titration method. The result showed that the titre value of pAb 1# was 1/48,000 and titre value of pAb 2# was 1/12,000. Both the antiserums obtained showed reasonably high recognition for the coating antigen. Although both lower IC₅₀ values were obtained, they showed slightly different specificity degree for the analyte DOP. The pAb 1# exhibited the better inhibition for the analyte and was selected as the assay reagent.

Enzyme-linked immunosorbent assay (ELISA) optimisation

In order to determine the DOP residue in PVC samples, it is important to develop a sensitive ELISA. For this purpose, the effects of several factors were investigated and evaluated as listed in Table 1. As shown in Table 1, all the standard deviation (SD) values are allowable which means that the results are significant and reliable.

Table 1. Effects of the factors on the performance of ELISA.

Factor	Amax	IC50	Amax/IC50
Coating method
4°C, overnight	1.58±0.06	33.88±1.0	0.047±0.003
37°C, 2 h	1.54±0.08	36.42±1.4	0.042±0.004
37°C, 2 h then 4°C, overnight	1.37±0.07	34.64±1.9	0.039±0.002
Ionic strength (mM)
5	1.54±0.08	35.49±1.3	0.043±0.003
10	1.76±0.05	34.66±0.8	0.051±0.002
20	1.78±0.11	35.72±1.5	0.050±0.002
40	1.82±0.12	36.94±3.5	0.049±0.002
pH
6.0	1.46±0.11	39.51±1.2	0.037±0.002
7.4	1.48±0.08	34.10±0.8	0.043±0.001
8.5	1.36±0.08	34.24±0.9	0.039±0.003
9.5	1.31±0.10	35.69±1.3	0.037±0.002
Methanol content (%)
5	1.52±0.14	32.21±1.0	0.047±0.006
10	1.45±0.13	28.23±0.9	0.051±0.006
20	1.44±0.11	24.12±0.9	0.060±0.004
30	1.46±0.13	26.35±1.6	0.055±0.004
Competitive time (h)
0.5	1.43±0.13	20.38±0.6	0.070±0.006
1	1.49±0.10	19.83±0.7	0.075±0.006
2	1.54±0.14	24.75±1.3	0.062±0.005
Colouration time (min)
10	1.05±0.17	21.28±1.8	0.049±0.001
20	1.54±0.11	19.92±1.1	0.077±0.004
30	1.57±0.15	21.07±1.6	0.074±0.002

Effect of coating method

Coating DOP–OVA at 4°C overnight gave lower IC₅₀ value and higher A_max and A_max/IC₅₀ value than coating at 37°C for 2 h and at 37°C, 2 h then 4 °C, overnight. So coating method of 4°C overnight was selected.

Effect of the ionic strength

The PBS ionic strength increasing from 5 to 40 mM resulted in higher A_max value. But only in 10 mM condition gave a lower IC₅₀ and a higher A_max/IC₅₀ value, indicating that only suitable ionic strength was beneficial to antigen/antibody reaction. Thus, 10 mM ionic strength of assay buffer was selected.

Effect of the assay buffer pH

To determine the effect of pH on the assay, the phosphate buffer was used in the range of pH 6.0–9.5. As shown in Table 1, the ELISA was more sensitive under neutral or slightly alkaline conditions than slightly acidic condition. Both under acidic condition (pH 6.0) and alkaline condition (pH 8.5 and 9.5) gave lower A_max, lower A_max/IC₅₀ and higher IC₅₀. Based on the sensitivity of the assay, pH 7.4 was chosen as optimised pH condition.

Effect of the methanol content

The effects of different methanol concentration on the ELISA system were evaluated. Table 1 clearly indicates that methanol concentration significantly influenced assay performance. Both under lower methanol concentration (10%) and higher methanol concentration (30%) gave lower A_max/IC₅₀ and higher IC₅₀; only 20% methanol concentration contributed to improve the sensitivity of the assay.

Effect of competition and colouration time

As we are aware, more competition and colouration time would lead to higher A_max because of more reaction time resulted in more antibody adsorbed on the microplate. As to A_max/IC₅₀ and IC₅₀, just in a suitable reaction time, we could get a higher A_max/IC₅₀ and lower IC₅₀. As to competition time, 1 h was mostly sensitive with a higher A_max/IC₅₀ and lower IC₅₀ and for colouration time, we got the best result at 20 min.

On the basis of the optimal conditions, we developed sensitive ELISA for the DOP residue as follows: the DOP–OVA conjugate as a coating antigen was coated onto the plate and placed at 4°C overnight and then the plate was blocked with 0.1% gelatine at 37°C for 2 h. The antiserum 1# was diluted 1600-fold and competed with the target analyte dissolved in PBS (10 mM, 20% methanol, pH 7.4) at 37°C for 1 h. At last for colouration step, 20 min was employed. For the ELISA solution system, pH 7.4, 10 mM PBS was used. Figure 4 shows a typical calibration curve obtained under these optimum conditions. The IC₅₀ value of the assay was 19.2±1.1 ng/ml with the detection limit of 1.9±0.1 ng/ml and quantitative working range of 5–75 ng/ml.

Figure 4.  ELISA standard inhibition curves for DOP: ‘B/B₀’ is binding ratio of antibody/coating antigen in wells.

Cross-reactivity

To determine the speci?city of the optimised ELISA, several phthalates were tested for cross-reactivity. Table 2 shows the cross-reactivity that was found by the immunoassay we developed. The assay shows a high cross-reactivity with DONP and DOAP, which was used as haptens in our research. The assay also has cross-reactivity with DIOP, DBP and DHP for 22.6%, 17.6% and 21.2%, respectively. Since these compounds have the same aromatic structure and similar alkyl groups with DOP, the cross-reactivity with DOP is understandable. To some extent, the immunoassay we developed was qualified for multi-residues detection.

Table 2. Cross-reactivity of structure related compounds to DOP determined by competitive indirect ELISA.

Analytes	IC50/ng ml^−1	CR^b (%)
Di-n-octyl phthalate (DOP)	19.2	100
Di-n-octyl 4-nitrophthalate (DONP)	35.4	54.2
Di-n-octyl 4-aminophthalate (DOAP)	29.3	65.5
Di-iso-octyl phthalate (DIOP)	84.9	22.6
Di-hexyl phthalate (DHP)	110	17.6
Di-n-butyl phthalate (DBP)	90.3	21.2
Phthalic acid	>2000	<0.1
Phthalic anhydride	>2000	<0.1
^aELISA conditions were the same as that described in Figure 4.
^bCross-reactivity (%)=(IC50 of DOP/IC50 of other compounds)×100.

Real sample analysis

To evaluate the optimised ELISA, the PVC sample analysis was employed. As shown in Figure 5, the correlative equation has a slope of 1.0299 and related coefficient of 0.998. The result identifies that the immunoassay we developed has a good agreement with HPLC-ESI-MS/MS analysis and it could be qualified to determine DOP residue in PVC samples.

Figure 5.  Comparison between ELISA and HPLC-ESI-MS/MS analysis.

Conclusion

We design and synthesise the hapten for DOP which have elicited the sensitive antibody for DOP. Furthermore, based on the sensitive antibody, an effective ciELISA was developed. Not only it can be used to screen DOP residue itself, but it can also detect DIOP, DBP and DHP residues. According to real samples analysis, the immunoassay has high accuracy contrast with HPLC-ESI-MS/MS analysis. It would be useful to monitor phthalate residues in various environmental matrices. Therefore, the ELISA we developed may become a new convenient and economical analytical tool to detect these residues in future.