Hapten Synthesis and Development of a Competitive Indirect Enzyme

Jul 5, 2014 - haptens for acrylamide were designed in an attempt to prepare antibodies with ... KEYWORDS: acrylamide, hapten, antibody, immunoassay...
1 downloads 0 Views 527KB Size
Download PDF
Article pubs.acs.org/JAFC

Hapten Synthesis and Development of a Competitive Indirect Enzyme-Linked Immunosorbent Assay for Acrylamide in Food Samples Jing Wu, Yu-Dong Shen, Hong-Tao Lei, Yuan-Ming Sun,* Jin-Yi Yang, Zhi-Li Xiao, Hong Wang, and Zhen-Lin Xu* Guangdong Provincial Key Laboratory of Food Quality and Safety, South China Agricultural University, Guangzhou 510642, China ABSTRACT: The high level of acrylamide in widely consumed processed foods poses a potentially significant risk to human health, which has led to an increasing demand for rapid, simple, and selective analytical methods. In the present work, several haptens for acrylamide were designed in an attempt to prepare antibodies with acrylamide affinity, but they failed their purpose. However, a polyclonal antibody was produced against 4-mercaptophenylacetic acid (4-MPA)-derivatized acrylamide, which showed high binding affinity to the derivative. As acrylamide easily reacted with 4-MPA at high derivation yield, a competitive indirect enzyme-linked immunosorbent assay (ciELISA) for acrylamide via a preanalysis derivatization was developed. The derivatization and ELISA conditions were fully optimized to produce a method for acrylamide assay that exhibited an IC50 of 2.86 μg/kg, limit of detection at 0.036 μg/kg, and linear range of 0.25−24.15 μg/kg. The results of preanalysis recovery tests of acrylamide-spiked food samples and screening of blind food samples by both ciELISA and HPLC-MS/MS indicated the proposed ciELISA’s good accuracy and reliability. This method was thus deemed suitable for routine acrylamide screening in food samples at low cost. KEYWORDS: acrylamide, hapten, antibody, immunoassay

■

INTRODUCTION Acrylamide is spontaneously formed from unsaturated amides and is mainly generated through the Maillard reaction of the amino acid asparagine with reducing sugars.1 Although since 2002 acrylamide has been discovered in a wide range of cooked foods, it poses a potentially significant risk to human health through its neurotoxicity, possible reproductive toxicity, carcinogenicity, and genotoxicity.2 In 1994, it was classified by the International Agency for Research on Cancer as probably carcinogenic to humans.3 In 2010, the European Chemicals Agency added acrylamide to the list of substances of very high concern.4 There are two established legal limits for acrylamide, one concerning acrylamide migration from packaging materials into food, with a limit of detection (LOD) of 10.0 μg/kg (European Union Commission), and the other involving drinking water, with a LOD of 1.0 μg/L (World Health Organization) .5 In food samples, risk assessments of daily acrylamide intake are still being performed and, therefore, no legal limit involving food samples has been established. A report from the Joint Expert Committee on Food Additives in 2005 indicated that the highest acrylamide contents are generated through frying potatoes, roasting coffee and cocoa beans, baking bread and cakes, thermal processing of cereals, and meat roasting, with the average content estimated from ∼0.1 to 0.5 mg/kg.6 Therefore, increasing effort is being invested in monitoring foodborne acrylamide that might pose a risk to humans and animals. As a result of acrylamide’s low molecular mass (71.08 Da), good water solubility (2155 g/L), and lack of sufficiently strong chromophore groups, its quantitation in food is difficult. The most commonly used analytical methods are high-performance © 2014 American Chemical Society

liquid (HPLC) and gas chromatography (GC) coupled with different detectors, such as mass spectrometry (MS) and electron capture detection (ECD).7,8 Methods such as capillary electrophoresis,9 proton transfer reaction MS,10 and an electrochemical biosensor11 have also been reported. Generally, these instrumental methods are accurate and sensitive, but they are laborious, time-consuming, and expensive. Owing to the high level of residue in a variety of food samples, the development of a quick, simple, inexpensive, and reliable quantitative acrylamide analysis is urgently needed. Immunoassays based on specific interactions between an antibody and corresponding antigen12 can meet the analytical requirement of a useful acrylamide assay with high sensitivity, cost effectiveness, and specificity. However, owing to its low molecular weight and the lack of strong epitope groups, there have been few reports regarding antibody production against acrylamide and the establishment of a related immunoassay. Zhou et al. synthesized an antigen for acrylamide, using Nacryloxysuccinimide (NAS) as the hapten instead of acrylamide itself, and obtained a polyclonal antibody that shows binding affinity toward acrylamide.13 This antibody was then used to develop a biotin−avidin enzyme-linked immunosorbent assay (BA-ELISA) for acrylamide. Employing the same strategy of antibody preparation, Quan et al. have developed an enhanced chemiluminescence ELISA to quantitate acrylamide in food products, such as potato chips, instant noodles, cookies, and Received: Revised: Accepted: Published: 7078

December 12, 2013 June 30, 2014 July 5, 2014 July 5, 2014 dx.doi.org/10.1021/jf5015395 | J. Agric. Food Chem. 2014, 62, 7078−7084

Journal of Agricultural and Food Chemistry

Article

Figure 1. Chemical structures of acrylamide, its haptens, and the hapten−protein conjugates.

cakes.14 These two methods have been reported to detect acrylamide directly; however, repeatability of these assays has been very poor because attempts using the same strategy to produce specific acrylamide antibodies have failed twice in our group. Preston et al. have designed a novel hapten by derivatizing acrylamide with 3-mercaptobenzoic acid and obtained a polyclonal antibody with high binding affinity for acrylamide derivative.15 They proposed a competitive indirect enzyme-linked immunosorbent assay (ciELISA) for the indirect detection of acrylamide, necessitating a derivatization of extracted acrylamide with 3-mercaptobenzoic acid. This strategy appears helpful for the rapid determination of acrylamide, but its sensitivity is insufficient, with a LOD of 65.7 μg/kg. In this study, several haptens against acrylamide were designed in attempts to obtain a specific acrylamide-affinity antibody. Although no antibodies with binding ability to free acrylamide were obtained, a polyclonal antibody was produced against 4-mercaptophenylacetic acid (4-MPA)-derivatized acrylamide (acrylamide-4-MPA) that showed high binding affinity to the derivative. A ciELISA for acrylamide via preanalysis derivatization was then developed, as acrylamide easily reacted at high yield with 4-MPA. The assay conditions were then optimized, and the assay was applied to detect acrylamide in several food samples.

■

(OVA), N-hydroxysuccinimide (NHS), 3,3′,5,5′-tetramethylbenzidine (TMB), and complete and incomplete Freund’s adjuvants were purchased from Sigma (St. Louis, MO, USA). Chromatographically pure methanol and acetonitrile were purchased from Guangzhou FanHong Trade Co., Ltd. Polystyrene ELISA plates were from Guangzhou Jiete Biotech Co. All other reagents were of analytical grade and obtained from a local chemical supplier (Yunhui Trade Co., Ltd., Guangzhou, China). Strata-X-C cartridge solid phase extraction cartridges were purchased from Phenomenex Science Apparatus Co. (Guangzhou, China). Instruments. Plates were washed in a Multiskan MK2 microplate washer (Thermo Scientific, Hudson, NH, USA). ELISA values were read with a Multiskan MK3 microplate reader (Thermo Scientific). Ultraviolet−visible (UV−vis) spectra were recorded on a UV-160A Shimadzu spectrophotometer (Kyoto, Japan). Nuclear magnetic resonance (NMR) spectra were obtained with both DRX-400 and DRX-600 NMR spectrometers (Bruker, Rheinstetten, Germany). HPLC-MS/MS analysis was carried out by using the 1200 series HPLC system (Agilent Technologies, Santa Clara, CA, USA) equipped with the Triple Quad 5500 HPLC-MS/MS System (AB SCIEX, Framningham, MA, USA). HPLC-ECD analysis was carried out on the SPD-20A HPLC system (Shimadzu, Kyoto, Japan). Buffers. The following buffers were used in this study: (1) 0.1 mol/ L phosphate-buffered saline (PBS, pH 7.4) containing 0.05% Tween20 for antibody dilution; (2) 0.05 mmol/L carbonate-buffered solution (pH 9.6) for coating; (3) 0.01 mmol/L PBS containing 0.05% Tween20 (PBST, pH 7.4) for washing; (4) 0.1 mol/L citrate and sodium phosphate for substrate buffer (pH 5.5); (5) 10 mL of substrate buffer, 150 μL of 1% (w/v) solution of TMB in DMF, and 2.5 μL of 6% (w/ v) H2O2 mixed for substrate solution; (6) 2.0 mol/L H2SO4 used for stopping reagent. Synthesis and Characterization of Haptens. Haptens were synthesized as follows (Figure 1). Hapten 1. Maleic acid monoamide was used directly as hapten 1. Hapten 2, NAS was used directly as hapten 2. Hapten 3. A 1 mL volume of acrylic acid was added to 1 mL of thionyl chloride and the mixture stirred at 50 °C for 3 h and then filtered to yield solution A. A 500 mg quantity of aminobutyric acid

MATERIALS AND METHODS

Materials and Reagents. Acrylamide (99%), methacrylamide, methyl acrylate, acrylic acid, acrolein, maleic acid monoamide, Nacryloxysuccinimide (NAS), and aminobutyric acid were purchased from J&K Scientific (Beijing, China). 4-MPA was obtained from Meryer Chemical Technology Co., Ltd. (Shanghai, China). Horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG was obtained from Boster Biotech Co., Ltd. (Wuhan, China). Keyhole limpet hemocyanin (KLH), dicyclohexyl carbodiimide (DCC), ovalbumin 7079

dx.doi.org/10.1021/jf5015395 | J. Agric. Food Chem. 2014, 62, 7078−7084

Journal of Agricultural and Food Chemistry

Article

Table 1. Characterization of Antiserum against Free Acrylamide Using Homologous and Heterologous Coating Antigens antiserum 1

antiserum 2

antiserum 3

antiserum 4

coating antigen

titera (×103)

inhibitionb (%)

titer (×103)

inhibition (%)

titer (×103)

inhibition (%)

titer (×103)

inibition (%)

H1-OVA H2-OVA H3-OVA H4-OVA

256 8 128 1024

2.0 0 6.0 6.6

1 4 4 0

−3.0 8.0 0.2 0

0 0 2 4

0 0 0 0

16 32 32 1024

−4.0 −1.3 0 0

a

Antiserum dilution considered suitable while A490 nm value approximately 1.0 at coating concentration of 1.0 mg/L. bPercentage inhibition calculated by following equation: inhibition (%) = (Ack − Aanalyte)/Ack. Ack is the A490 nm in the absence of analyte and Aanalyte is the A490 nm in the presence of analyte (100 mg/L acrylamide). was added. After 40 min in a water bath at 37 °C, the wells were washed five times with PBST and HRP-conjugated goat anti-rabbit IgG (diluted 1:5000) added with 100 μL/well. Finally, TMB solution (400 μL of 0.6% TMB−dimethyl sulfoxide and 100 μL of 1% H2O2 diluted with 25 mL of citrate−acetate buffer, pH 5.5) was added to the wells and incubated for 10 min. The reaction was quenched by the addition of 50 μL of 2 M H2SO4 and the absorbance recorded at 450 nm. Optimization of Immunoassay. Experimental parameters, such as coating concentration, antibody dilution, and working buffer, were optimized to improve the optimal sensitivity of this immunoassay. The criteria used to evaluate immunoassay performance were maximal absorbance (Amax) and IC50 (50% inhibition of antibody binding) values, with the Amax/IC50 ratio being a convenient estimate of an influence on ELISA sensitivity and higher ratios indicating higher sensitivity.18 Competitive curves were created in working buffer containing a concentration series of PO43− (from 0.1 to 0.4 M), Tween-20 (from 0.08 to 0.5 M), or methanol (from 0 to 15%, v/v) to evaluate the effects of ELISA conditions on assay sensitivity. Cross-Reactivity (CR). CR was studied using the standard acrylamide solution or its derivative and some of its analogues. CR values were calculated as follows: CR% = (IC50 of analyte/IC50 of analogue) × 100. Data Analysis. Competitive curves were obtained by plotting absorbance against the logarithm of the analyte concentration. Sigmoid curves were generated using OriginPro 7.5 software (OriginLab Corp., Northampton, MA, USA): Y = (A − D)/[1 + (x/C)B] + D, where A and D are the maximal and minimal Abs (Amax and Amin, respectively), B is the slope of the sigmoidal curve, and C is the concentration of acrylamide or its derivative that causes IC50. The LOD was defined as the standard concentration of acrylamide or its derivative that inhibited 10% of antibody binding, and the detectable concentration range was defined as the standard concentration that inhibited 20− 80% of acrylamide’s or its derivative’s standard antibody binding.19 Preparation of Samples. Samples of potato chips, cookies, and coffee were bought from a local supermarket. Samples of 4.0 g and 20.0 mL of water were placed in 50 mL centrifuge tubes and subjected to ultrasound for 1 min, followed by a defatting step using 20 mL of nhexane and then centrifugation at 8200g at 15 °C for 10 min. The 10 mL supernatant was cleaned by solid phase extraction on a strata-X-C cartridge (Phenomenex Inc., Torrance, CA, USA). The eluate was subjected to HPLC-MS/MS analysis and derivatized to obtain acrylamide-4-MPA for ciELISA analysis. For derivatization, 1 mL of eluate was adjusted to pH 10 by adding aqueous NaOH and then 10 μg of 4-MPA. This reaction mixture was stirred in darkness at 70 °C for 30 min and then acidified to pH 2.0 with HCl. The resulting mixture was then extracted using ethyl acetate and the organic phase blown dry under a nitrogen stream. Finally, the residue was dissolved in 1 mL of PBS and diluted at least 4-fold with optimized working buffer to remove matrix effects prior to ciELISA analysis. HPLC Conditions. The yield of acrylamide-4-MPA was characterized by HPLC-ECD. A 20 μL portion of sample was resolved at a 0.8 mL/min flow rate and 40 °C using an acetonitrile/acetic acid (plus 1 g of H2O/L) gradient. A linear gradient was programmed over 15 min using 30−60% (acetonitrile content, by vol). The initial acetonitrile proportion (30%) was then pumped for 5 min before

was added to 2 mL of double-distilled H2O, followed by the addition of 500 μL of 2 M NaOH and 2 mL of butylene oxide to obtain solution B. Next, solution A was slowly introduced dropwise into solution B and stirred overnight at room temperature. The mixture was then extracted with ethyl acetate and evaporated to obtain 4acrylamidobutanoic acid (hapten 3). ESI-MS analysis (negative) m/z 156 [M − H]−; 1H NMR (CDCl3, TMS, 300 MHz) δ 11.0 (s, 1H), 8.03 (brs, 1H), 6.48 (d, J = 10.2 Hz, 1H), 6.04−5.98 (m, 1H), 5.52− 5.48 (m, 1H), 3.20−3.13 (m, 2H), 2.33−2.28 (m, 2H), and 1.94−1.89 (m, 2H). Hapten 4. With stirring, 1.0 mmol of 4-MPA in 1.6 mL of methanol was added to a solution of 5 mmol of acrylamide in 0.8 mL of coating buffer. The mixture was left to stand at 37 °C in darkness for 1 h. Subsequent precipitation of the product was enhanced by the addition of 2 mL of deionized water and then dried under light vacuum in darkness for 0.5 h to obtain 2-(4-(3-amino-3-oxopropylthio)phenyl)acetic acid (hapten 4). ESI-MS analysis (negative) m/z 238 [M − H]−; 1 H NMR (CD3OD, 400 MHz) δ 7.335−7.349 (d, J = 5.6 Hz, 2H), 7.233−7.247 (d, J = 5.6 Hz, 2H), 3.579 (s, 2H), 3.144−3.168 (m, 2H), and 2.488−2.513 (m, 2H). Preparation of Hapten−Protein Conjugates. All four haptens were coupled to KLH for immunogens and to OVA for coating antigens. Haptens 1, 3, and 4 were conjugated to carrier proteins by the active ester method, as described by Sheng et al.16 Hapten 2 was coupled to carrier proteins according to the method of Zhou et al.13 The hapten−protein conjugates were then dialyzed against 0.01 M PBS at 4 °C for 24 h with six changes of PBS. UV−vis spectral data were used to confirm the structures of the final conjugates and the ratios of hapten to carrier protein determined using the 2,4,6trinitrobenzenesulfonic acid method.17 Preparation of Polyclonal Antibodies. For each immunogen, intramuscular injections were given to two New Zealand white rabbits and a total of eight rabbits used to raise antibodies. The prepared immunogen (0.5 mg of coupled protein in 1 mL of PBS) was emulsified with 1 mL of Freund’s complete adjuvant (1:1, v/v) for the first injection into a New Zealand rabbit weighing 1.2−2.2 kg, provided by Guangdong Medical Laboratory Animal Center. Then, Freund’s incomplete adjuvant was used as the booster muscle injection, at intervals of 4 weeks. Bleed samples were collected 10 days after the fourth immunization, and the obtained antiserum was purified by the caprylic acid−ammonium sulfate method and separated into 1 mL aliquots held at −20 °C until use. Negative (control) serum was obtained from an unimmunized rabbit. All animal experiments were performed in compliance with the relevant protective and administrative laws for laboratory animals of China and conducted with the approval of the Institutional Authority for Laboratory Animal Care. Competitive Indirect ELISA. Serum samples were screened for antibodies specific to the analyte (acrylamide or its derivative acrylamide-4-MPA) via a ciELISA. The 96-well plates were coated with coating antigens (1 mg/L, 100 μL/well) in carbonate buffer at 4 °C overnight. The wells were next washed two times with PBS Tween20 (PBST) and blocked with 120 μL/well of 5% skimmed milk in PBST at 37 °C for 3 h, and the plates were dried at 37 °C for 1 h. The wells were then incubated with 50 μL of diluted acrylamide or its derivative standards in PBS, and 50 μL of diluted antibody in PBST 7080

dx.doi.org/10.1021/jf5015395 | J. Agric. Food Chem. 2014, 62, 7078−7084

Journal of Agricultural and Food Chemistry

Article

the next injection was applied. The ECD quantitation channel was selected at 254 nm. HPLC-MS/MS was used for acrylamide detection, with the spectra recorded by a mass spectrometer (AB SCIEX Triple Quad 5500, AB SCIEX) coupled to a spectra series HPLC (Shimadzu-20A) equipped with a SunFire HPLC column (250 × 4.6 mm, 5 μm; Waters Corp., Milford, MA, USA). A 20 μL portion of sample was separated at a 1.0 mL/min flow rate using an isocratic eluent (methanol/diluted acetic acid, 10:90, v/v, plus 1 g of H2O/L). Analytes were determined using the above mass spectrometer equipped with a turbo-spray ion source. Quantitation was performed in multiple reaction monitoring (MRM) analysis mode. The respective MS parameters were as follows: curtain gas at 40 psi; collision gas at 6 psi; ion spray voltage at 5500 V; temperature at 650 °C for both; ion source gases 1 and 2 both at 55 psi; declustering potential at 57 V; and entrance potential at 12 V. The analyte indentities were confirmed with product/precursor ions as well as on the basis of the peak retention times corresponding to the standard. The [M + H]+ (m/z 72.0) was selected as the parent acrylamide ion. Qualitative ions were selected at m/z 43 and quantitative ions at m/z 55. The mass spectrometer was controlled and quantitation performed using Analyst 1.5.1 software from AB SCIEX.

design and synthesis of an optimal immunizing hapten. In the beginning, the goal was to produce an antibody with specificity against acrylamide, and for this purpose three haptens were designed. Maleic acid monoamide shares the common structure of acrylamide as well as a carboxyl group, which can be directly attached to carrier protein; therefore, it was first selected as a hapten (H1, Figure 1). After coupling to protein, the acrylamide amino linkage could be exposed. NAS (H2) was also employed as a hapten because it has been reported to be able to produce antibodies specific to free acrylamide.13 After reaction with protein, the conjugate preserved the intact acrylamide structure, and the double bond could be exposed. However, the H2 spacer arm might have been too short, which might have resulted in embedding the hapten in the protein surface. Therefore, another hapten, 4-acrylamidobutanoic acid (H3), was synthesized through the reaction of acrylic acid and aminobutyric acid. These haptens were coupled to KLH and used to immunize rabbits, but the resulting antisera against the three haptens proved not to contain antibodies specific for free acrylamide (Table 1). The inability of the above immunogens to stimulate antibody production possessing the desired specificity for acrylamide was probably because of their diminutive mass of 71.08 Da and the lack of sufficient epitope for virgin B or T helper cell binding. The double bond of acrylamide can easily react with sulfhydryl groups by Michael addition. Preston et al. have designed a hapten by reacting acrylamide with 3-mercaptobenzoic acid (3-MBA).15 The corresponding antiserum showed specific binding to acrylamide-3-MBA in a ciELISA; however, the sensitivity of this ciELISA was unsatisfactory. In that study, immunogens were generated by coupling the hapten to carrier proteins via a carboxylic acid spacer. However, in relation to hapten design for small molecular weight compounds, a suitable spacer length between the hapten and the carrier protein should be beneficial for producing desired antibodies.20 A previous study has indicated that hapten conformation can be altered when too short a spacer is employed for its conjugation

■

RESULTS AND DISCUSSION Hapten Design and Antibody Production. The key step during the production of an immunoassay is to consider the Table 2. Characterization of Antiserum 4 against Acrylamide-4-MPA Using Homologous and Heterologous Coating Antigensa antiserum 4

a

3

coating antigen

titer (×10 )

inhibition (%)

H1-OVA H2-OVA H3-OVA H4-OVA

16 32 32 1024

0 0 0 86

Analyte, 1 mg/L acrylamide-4-MPA.

Figure 2. Effect of physicochemical parameters on ELISA performances (n = 3). 7081

dx.doi.org/10.1021/jf5015395 | J. Agric. Food Chem. 2014, 62, 7078−7084

Journal of Agricultural and Food Chemistry

Article

Figure 3. Standard curve (A) and inhibition curve (B) of acrylamide-4-MPA assessed by ciELISA.

Table 3. Cross-Reactivity of Obtained Antibody with Acrylamide-4-MPA and Its Analogues

Table 5. Recoveries of Acrylamide from Spiked Food Samples by HPLC-MS/MS and Recoveries of Acrylamide Derivative from Spiked Food Samples by ciELISA recovery ± SD (%) food sample

no.

(B) reaction rime (min)

(C) pH

(D) molar ratio of 4-MPA/ acrylamide

yield of acrylamide4-MPA (%)

1 2 3 4 5 6 7 8 9 K1a K2a K3a Rab

37 37 37 50 50 50 70 70 70 129.2 206 231.4 102.2

15 30 50 15 30 50 15 30 50 133.4 223.5 209.7 76.3

8 9 10 9 10 8 10 9 8 193.4 169.4 203.8 34.4

2 5 10 10 2 5 5 10 2 206.9 237.1 122.6 114.5

32.1 64.2 32.9 21.6 91.2 93.2 79.7 68.1 83.6 ΣY = 566.6

spiked level (μg/kg)

ciELISA

HPLC-MS/ MS

cookies

300

50.0 200.0 1000.0

90.4 ± 19.2 107.7 ± 3.6 81.9 ± 6.4

96.1 ± 12.1 80.7 ± 9.2 96.5 ± 5.2

potato chips

1368

50.0 200.0 1000.0

104.9 ± 9.2 88.7 ± 7.9 90.5 ± 8.8

80.4 ± 11.4 90.0 ± 6.8 95.0 ± 9.2

coffee

113

50.0 200.0 1000.0

89.6 ± 8.1 73.7 ± 4.9 112.3 ± 7.2

74.5 ± 14.8 81.0 ± 7.3 78.1 ± 6.6

Although the corresponding antiserum produced here showed no binding affinity for acrylamide (Table 1), it demonstrated significant binding to acrylamide-4-MPA (Table 2). As acrylamide can easily react with 4-MPA to yield acrylamide-4MPA, an ELISA based on antiserum 4 could be developed to detect acrylamide by including a rapid preanalysis derivatization. Development and Optimization of ciELISA. Antiserum 4 and hapten 4-OVA were used to develop a ciELISA for acrylamide-4-MPA. Several physicochemical parameters, including coating concentration, antibody dilution, ionic strength, and Tween-20 concentration, which are important in assay performance, were evaluated to improve ELISA performance. For each condition, standard curves for acrylamide-4-MPA were established (n = 3), and the maximum value of the inhibition curve (Amax) as well as the analyte concentration that produced 50% decreased inhibition in Amax (IC50) was obtained from these curves. The results of these physicochemical parameter effects on ELISA performance are summarized in Figure 2. For coating concentration and antibody dilution, a preliminary checkerboard titration was applied to choose several combinations of coating concentration and antibody dilution, under which the A450 was ∼1.0. The results indicated that the combination of coating concentration and antibody dilution at 125 μg/L and 1:256000, respectively, were optimal for the highest sensitivity (Figure 2A). For the working system, three solutions (PBST, PBS, 100 mmol/L, pH 7.4, and H2O) were examined, and it was found that the Amax/IC50 ratio was highest with PBST (Figure 2B). Further study of Tween-20 concentration and ionic strength effects indicated that the

Table 4. Analysis of L9(3)4 Test Results (A) temperature (°C)

HPLC concn (μg/kg)

a Ki = Σ amount of target compounds at concentration i. bRefers to result of extreme analysis. Ra = max(Ki) − min(Ki).

to carrier protein.21 Therefore, in this study, 4-MPA instead of 3-MBA was reacted with acrylamide to obtain a novel hapten, which was then coupled to KLH to form an immunogen. 7082

dx.doi.org/10.1021/jf5015395 | J. Agric. Food Chem. 2014, 62, 7078−7084

Journal of Agricultural and Food Chemistry

Article

Table 6. Acrylamide Content in Different Food Samples food sample

typical concn4 (μg/kg)

mean concn23 (μg/kg)

food sample name

cookies

50−700

350

Chinese cookie compressed cookie hard cookie fermented cookies

potato chips

330−2300

752

Chinese potato chip French potato chip American potato chip potato crisp baked potato

coffee

100−668

220

coffee milk tea cappuccino instant coffee cafe latte instant coffee

HPLC-MS/MS concn (μg/kg)

ciELISA concn (μg/kg)

213 302 574 113

186.3 263.7 655.8 81.4

705 626 752 1000 1130

851.7 743.7 913.3 1161.4 1427.5

179 595 651 113

151.2 558.3 854.6 129.7

determinant. The highest acrylamide-4-MPA derivatization yield was A3B2C3D2 (70 °C reaction temperature, 30 min reaction time, pH 10, 1:5 acrylamide/4-MPA molar ratio). Under the best conditions, the acrylamide-4-MPA derivatization yield was 93.2%. Sample Preparation and Recovery Test. A standard HPLC-MS/MS method was developed here for acrylamide analysis. The acrylamide limit of detection by HPLC-MS/MS was 0.2 μg/L, and good linearity was observed over the acrylamide range of 0.9−1000 μg/L in each calibration, with a correlation coefficient of 0.9995. Potato chips, cookies, and coffee samples were first analyzed for their acrylamide content by HPLC-MS/MS and then spiked with three different concentrations (50, 200, and 1000 μg/kg) of acrylamide standards. The samples were extracted by water and then derivatized with 4-MPA to form acrylamide-4-MPA. Next, the acrylamide-4-MPA content was analyzed using the optimized ciELISA procedure. Recoveries were tested from 81.9 to 107.7% for cookies, from 88.7 to 104.9% for potato chips, and from 73.7 to 112.3% for coffee (Table 5). Good correlations were observed between the results from ciELISA and standard HPLC-MS/MS (Table 5). As a 20-fold dilution of the three samples was performed and good recoveries were observed from ciELISA, the IC50 of the proposed ciELISA for acrylamide in cookie, potato, and coffee samples was concluded to be 57.2 μg/L. The LOD and linear range here were 0.7 and 5.0−483 μg/kg for acrylamide, respectively, and the sensitivity almost the same as that of HPLC-MS/MS. ciELISA Screening and HPLC-MS/MS Analysis of Blind Samples. Four different kinds of cookie samples, five potato chip samples, and four coffee samples were obtained from a local supermarket. The content of acrylamide in these samples was then determined by both ciELISA and HPLC-MS/MS; the results are shown in Table 6. A correlation coefficient of 0.9828 was obtained between the results of ciELISA and HPLC-MS/ MS, indicating good reliability and accuracy for the proposed ciELISA. In acrylamide content, higher average concentrations were observed in potato samples. The results were equal to the typical concentration4 and mean concentration23 of acrylamide in these foods. In conclusion, a ciELISA was developed for acrylamide detection via the determination of 4-MPA-derivatized acrylamide. For acrylamide, the IC50 was 2.86 μg/kg, the LOD 0.036

optimal Tween-20 concentration and ionic strength were 0.08 and 0.1 mol/L, respectively (Figure 2C,D). Sensitivity and Specificity. Using the optimized reaction conditions, dose−response and inhibition curves for acrylamide-4-MPA in assay buffer were constructed (Figure 3, panels A and B, respectively). The method exhibited an IC50 of 9.61 μg/kg and an LOD of 0.12 μg/kg for acrylamide-4-MPA (239 Da), whereas the detection linear range was 0.85−81.2 μg/kg, which was equivalent for acrylamide (71.08 Da) to an IC50 of 2.86 μg/kg, an LOD of 0.036 μg/kg, and a linear range of 0.25−24.15 μg/kg. The sensitivity of this assay was significantly improved compared to that of Preston et al.,15 which might be attributed to the introduction of a suitable length spacer arm between the hapten and carrier protein. A similar conclusion, that the introduction of longer spacer arm rather than a monocarboxylic acid on benzene ring could led to antibody production with higher affinity to analyte, was also obtained by our group in antibody production and ELISA development for furaltadone metabolite 3-amino-5-morpholinomethyl-2-oxazolidinone.22 The selectivity of the present antibody and its applicability in actual acrylamide screening were investigated by examining the CRs of the antibody to a series of structural analogues and their derivatives. Acrylamide-4-MPA, 4-MPA, acrylamide, methacrylamide, methyl acrylate, acrylic acid, acrolein, maleic acid monoamide, NAS, and 4-acrylamidobutanoic acid were tested, and the results are summarized in Table 3. No CR was observed to any of the test compounds except acrylamide-4MPA, suggesting that the antibody possessed high specificity for acrylamide-4MPA. As no CR to 4-MPA was detected, there was no need to remove unreacted 4-MPA after the derivatization step. Optimization of Acrylamide-4-MPA Derivatization. The derivatization yield of acrylamide-4-MPA was improved by studying several derivatization conditions, such as the reaction’s molar ratio of acrylamide/4-MPA, temperature, pH, and time, with the acrylamide-4-MPA content determined by HPLC-ECD. An L9(34) orthogonal experiment was performed after a series of preliminary single-factor experiments. As seen in Table 4, the influences on the mean compound extraction yields by these factors were found to decrease in the order molar ratio of acrylamide/4-MPA (D) > temperature (A) > time (B) > pH (C), according to the related Ra values. The reaction temperature was found to be an important yield 7083

dx.doi.org/10.1021/jf5015395 | J. Agric. Food Chem. 2014, 62, 7078−7084

Journal of Agricultural and Food Chemistry

Article

μg/kg, and the linear range 0.25−24.15 μg/kg in the chosen assay buffer. The recoveries from acrylamide-spiked potato chips, cookies, and coffee samples were tested from 73.7 to 112.3%, and good correlations were obtained between the results of ciELISA and standard HPLC-MS/MS. Although the sample pretreatment was not simple enough, it was more inexpensive than HPLC-MS/MS analysis. Additionally, the ciELISA was easy to apply with professorial training and suitable for routine acrylamide screening in a large number of food samples.

■

Rio Grande River Basin, Texas. Water Qual. Expo. Health 2009, 1, 145−158. (13) Zhou, S.; Zhang, C.; Wang, D.; Zhao, M. P. Antigen synthetic strategy and immunoassay development for detection of acrylamide in foods. Analyst 2008, 133, 903−909. (14) Quan, Y.; Chen, M. L. Development of an enhanced chemiluminescence ELISA for the rapid detection of acrylamide in food products. J. Agric. Food Chem. 2011, 59, 6895−6899. (15) Preston, A.; Fodey, T.; Elliott, C. Development of a highthroughput enzyme-linked immunosorbent assay for the routine detection of the carcinogen acrylamide in food, via rapid derivatization pre-analysis. Anal. Chim. Acta 2008, 608, 178−185. (16) Sheng, W.; Li, Y.; Wang, S. Development of an enzyme-linked immunosorbent assay for the detection of gentamycin residues in animal-derived foods. LWT−Food Sci. Technol. 2013, 50, 204−209. (17) Tinashe, B. R.; Victor, J. J.; Paul, D. S. Production, characterization and utility of a panel of monoclonal antibodies for the detection of toluene diisocyanate haptenated proteins. J. Immunol. Methods 2011, 373, 127−135. (18) Li, M.; Wang, M. H. Development of immunoassays for detecting clothianidin residue in agricultural products. J. Agric. Food. Chem. 2013, 61, 3619−3623. (19) Mika, K.; Hiroshi, Y.; Mikiko, U.; Takeshi, K.; Toshihide, T.; Masahiro, F.; Machiko, S.; Seiji, I.; Shigekazu, I.; Shiro, M. Development of an enzyme-linked immunosorbent assay for residue analysis of the insecticide emamectin benzoate in agricultural products. J. Agric. Food Chem. 2009, 57, 359−364. (20) Ruud, M. T.; Chris, R.; Barbara, D.; Tomlinson, I. M. Antibody arrays for high-throughput screening of antibody−antigen interactions. Nat. Biotechnol. 2000, 18, 989−994. (21) Xu, Z. L.; Wang, H.; Shen, Y. D. Â .; Mikaela, N.; Lei, H. T.; Beier, R. C.; Zheng, W. X.; Yang, J. Y.; She, Z. G.; Sun, Y. M. Conformational changes of hapten-protein conjugates resulting in improved broad-specificity and sensitivity of an ELISA for organophosphorus pesticides. Analyst 2011, 136, 2512−2520. (22) Xu, Z. L.; Shen, Y. D.; Sun, Y. M.; Katrina, C.; Tian, Y. X.; Zhang, S. W.; Lei, H. T.; Jiang, Y. M. Novel hapten synthesis for antibody production and development of an enzyme-linked immunosorbent assay for determination of furaltadone metabolite 3amino-5-morpholinomethyl-2-oxazolidinone (AMOZ). Talanta 2013, 103, 306−313. (23) Javad, K.; Alain, L. B. Acrylamide in foods: chemistry and analysis. A review. Food Bioprocess Technol. 2011, 4, 340−363.

AUTHOR INFORMATION

Corresponding Authors

*(Y.-M. Sun) Mail: College of Food Science, South China Agricultural University, 483 Wushan Road, Tianhe District, Guangzhou 510642, China. Phone: +86 2085283925. Fax: +86 2085280270. E-mail: [email protected]. *(Z.-L. Xu) E-mail: [email protected]. Funding

This work was supported by the National Basic Research Program of China (973 Program, 2012CB720803) and the Science and Technology Innovation Project of Department of Education of Guangdong Province (2013KJCX0027). Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Zhang, Y.; Yi, P. R.; Zhang, Y. New research developments on acrylamide: analytical chemistry, formation mechanism, and mitigation recipes. Chem. Rev. 2009, 109, 4375−4397. (2) Heinz, F.; Aurelie, M. Dietary acrylamide intake of adults in the European Prospective Investigation into Cancer and Nutrition differs greatly according to geographical region. Eur. J. Nutr. 2013, 52, 1369− 1380. (3) International Agency for Research on Cancer,Lyon, France, 1994; pp 270. (4) Ioannis, S. A.; Niki, D. Acrylamide: formation, occurrence in food products, detection methods, and legislation. Crit. Rev. Food Sci. 2014, 54, 708−733. (5) Evrim, S.; Tekkeli, K.; Ö nal, C.; Ö nal, K. A review of current methods for the determination of acrylamide in food products. Food Anal. Methods 2012, 5, 29−39. (6) International Food Safety Authorities Network (INFOSAN). Acrylamide in Food Is a Potential Health Harzard; Geneva, Switzerland, 2005; Vol. 3, p 1. (7) Hoenicke, K.; Gatermann, R.; Harder, W.; Hartig, L. Analysis of acrylamide in different foodstuffs using liquid chromatography− tandem mass spectrometry and gas chromatography−tandem mass spectrometry. Anal. Chim. Acta 2004, 520, 207−215. (8) Zhang, Y.; Dong, Y.; Ren, Y.-P.; Zhang, Y. Rapid determination of acrylamide contaminant in conventional fried foods by gas chromatography with electron capture detector. J. Chromatogr., A 2004, 1116, 209−216. (9) Elisabet, B.; Oscar, N.; Luis, P.; Maria, T. G. Analysis of acrylamide in food products by in-line preconcentration capillary zone electrophoresis. J. Chromatogr., A 2006, 1129, 129−134. (10) Philippe, P.; Lindinger, C. Y.; Imre, B. Proton transfer reaction mass spectrometry, a tool for on-line monitoring of acrylamide formation in the headspace of Maillard reaction systems and processed food. Anal. Chem. 2003, 75, 5488−5494. (11) Nelson, A. F.; Silva, M. J.; Matos, A. K.; Maria, M. R. An electrochemical biosensor for acrylamide determination: merits and limitations. Port. Electrochim. Acta 2011, 29, 361−373. (12) Chang, N. B.; Mani, S. Pesticide impact assessment via using enzyme-linked immunosorbent assay (ELISA) technique in the Lower 7084

dx.doi.org/10.1021/jf5015395 | J. Agric. Food Chem. 2014, 62, 7078−7084