Colloidal Gold Immunochromatographic Assay for Rapid Detection of Carbadox and Cyadox in Chicken Breast.

Abused or misused carbadox (CBX) and cyadox (CYA) in animal feed may cause food safety concerns, threatening human health. Here, we describe the design of a novel hapten for preparation of a monoclonal antibody against CBX and CYA simultaneously. Using this antibody, colloidal gold immunochromatographic assay (GICA) was developed for screening of CBX and CYA residues in chicken breast. Under optimal conditions, semiquantitative analysis results were visible by eye, with a visual limit of detection of 8 μg/kg for CBX and CYA, and cut-off values of 20 μg/kg for CBX and 40 μg/kg for CYA in chicken breast. Quantitative analysis could be performed using a hand-held strip scanner, with a calculated limit of detection of 2.92 μg/kg for CBX and 2.68 μg/kg for CYA in chicken breast. Validated by liquid chromatography-MS/MS, the developed GICA provides a useful tool for rapid on-site CBX and CYA residue screening in chicken breast.

Introduction

Carbadox (CBX) and cyadox (CYA) belong to the class of compounds known as quinoxaline 1,4-dioxides, which are widely used as antibacterial growth-promoting agents in animal feed.

Because CBX has mutagenic, teratogenic, and carcinogenic properties, many countries have forbidden its use in food animals.[1] CYA is a novel species of quinoxaline and is considered to be safer than CBX, and thus, has replaced other quinoxalines in some countries.[2] However some studies recently reported that CBX might have potential mutagenicity and liver toxicities at certain doses.[3] Thus, it is necessary to establish a screening method for CBX and CYA residues for animal-origin food.

Several instrument methods have been established for detection of CBX and CYA, such as high-performance liquid chromatography with ultraviolet (UV) detection[4,5] and high-performance liquid chromatography tandem mass spectrometry (HPLC–MS/MS).[6−8] Because of its high accuracy and sensitivity, HPLC–MS/MS is used as the standard method for actual sample detection. However, such methods usually need complex sample pretreatment, expensive instruments, long detection times, and professional technicians. These disadvantages restrict their application for the rapid screening of large numbers of samples.

Compared with these instrumental methods, immunoassay methods have advantages of simple sample preparation, low cost, time-saving, and convenient operation. For this reason, immunoassays, including enzyme-linked immunosorbent assay (ELISA),[9,10] colloidal gold immunochromatographic assay (GICA),[11−18] and fluorescence immunoassays,[19−21] have been widely applied in food safety on-site detection. Recently, some research studies about immunoassays for the rapid detection of quinoxalines had been established.[22−29] As shown in Table , ic-ELSA and immunochromatographic assays have been developed to simultaneously detect five quinoxalines: CBX, CYA, olaquindox (OLA), quniocetone (QCT), and mequindox (MEQ).[30] However, no immunoassays have been reported for simultaneous detection of CBX and CYA in animal tissues.

Table 1

Immunoassays for Quinoxaline 1,4-Dioxide Detection

methods	target analytes	LOD	matrix	references
immunochromatographic assay	QCA	25 ng/g	pig tissues	Le et al. 2012[22]
ic-ELISA and time-resolved fluoroimmunoassay	QCA		porcine muscle and liver	Le et al. 2015[26]
immunochromatographic strip	CBX, MEQ, OLA, QCT, and CYA	10, 15, 15, 20 and 20 ng/mL	animal feeds	Le et al. 2015[30]
immunochromatographic assay	MQCA	0.25 ng/mL	fish	Liu et al. 2017[11]
fluorescent ELISA	OLA	0.68 μg/kg	swine feeds	Peng et al. 2019[27]
GICA	CBX and CYA	2.92 and 2.68 μg/kg	chicken breast	this work

In this work, we first designed a novel hapten for CBX and CYA monoclonal antibody (mAb) preparation. Based on this antibody and the visualization of gold nanoparticles (GNPs), a strip sensor was fabricated for rapid detection of CBX and CYA residues in chicken breast.

Results and Discussion

Hapten Design

CBX (MW = 262.23) and CYA (MW = 271.24) are both of low-molecular weight and have no immunogenicity. Thus, they need to be coupled to a carrier protein to induce an immune response by the mouse. However, CBX and CYA have no active groups (Figure a), such as −NH2 or −COOH that can react directly with a carrier protein. In order to prepare a mAb which can identify CBX and CYA simultaneously, we kept the shared structural element of CBX and CYA, (E)-2-(hydrazonomethyl)quinoxaline 1,4-dioxide, as the hapten (Figure b). Besides, the −NH2 of the hapten can conjugate with a carrier protein using the GA method. GA is a common protein coupling method and can introduce a five carbon chain as the spacer arm that is beneficial to exposure of the antigenic determinant. This results in conditions that are favorable for the mouse’s immune system to produce antibodies against CBX and CYA. The LC–MS/MS spectrum (Figure a,b) revealed a molecular ion at m/z 205.1 [M + 1]+ at a retention time of 2.287 min, which supported a molecular formula of C9H8N4O2 (MW 204.19). The structure of the hapten in this work was also further confirmed by 1H NMR spectrometry (400 MHz, DMSO-d6) (Figure c): δ 8.40–8.54 (m, 5H), 8.18 (s, 1H), 7.86–7.93 (m, 2H).

Figure 1

(a) Chemical structure of CBX, CYA; (b) synthetic route of hapten.

Figure 2

LS–MS/MS and 1H NMR spectra of hapten. (a) Positive ions LC spectrum of hapten with a retention time of 2.287 min; (b) mass spectrum of hapten with a m/z ratio of 205.1 confirmed the formula of hapten (C9H8N4O2, MW 204.19). (c) 1H NMR spectra of hapten.

Antigen Characterization

Antigens, including hapten–ovalbumin (OVA), hapten–BSA, and hapten–keyhole limpet hemocyanin (KLH), were characterized by UV spectroscopy. As shown in Figure , the characteristic UV absorption peaks of hapten and carrier proteins were at 378 and 280 nm. The antigens simultaneously had the absorption peak of hapten at 345 nm and carrier proteins at 280 nm, and the obviously shifted peaks indicated these antigens were successfully produced.

Figure 3

UV spectrogram of hapten–KLH, hapten–BSA, and hapten–OVA.

mAb Characterization

The sensitivity of a mAb determines to a great extent the sensitivity of the associated immunoassay. The assay buffer plays a vital role in immunoassay analysis. The pH value, ionic strength, and organic solvent content of assay buffer have an effect on protein configuration, which will influence the conjugation of the antibody and antigen.[31,32] Besides, different analytes have different dissolved conditions; for example, dibutyl phthalate could be sufficiently dissolved at a certain concentration of organic solvent; tetracycline could undergo hydrolysis under acidic and basic conditions, and remain stable under neutral conditions. In this work, NaCl content ranging from 0.4 to 6.4% was tested to assess the effect of ionic strength. As shown in Figure a, the absorbance value decreased significantly along with the increasing NaCl content. The maximum absorbance value (Amax) was less than 1.0 when the NaCl content was greater than or equal to 3.2%. The biggest Amax/IC50 value was obtained when the NaCl content was 0.4% in assay buffer. The methanol content and pH value of assay buffer have few effects on Amax values. The biggest Amax/IC50 value was obtained when the methanol content was 10% and the pH value was at 7.4 (Figure b,c). Thus, assay buffer with 0.4% NaCl content, 10% methanol content, and pH 7.4 was used to establish the standard curve. Under these optimum conditions, the IC50 values of CBX and CYA were 1.84 and 1.85 ng/mL (Figure d), respectively.

Figure 4

Characterization of mAb. Optimization of assay buffer for ic-ELISA, (a) NaCl content, (b) methanol content, (c) different pH; (d) standard curve for CBX and CYA with ic-ELISA; (e) subtypes determination; (f) affinity detection.

Identification of the mAb subtype benefits the selection of the mAb purification method.[33]Figure e shows that the mAb against CBX and CYA was subtype IgG2a, which can be purified using the salting out method (caprylic acid–saturated ammonium sulfate precipitation method), protein A, or protein G method.[34] We used the protein G method for ascites purification to obtain the mAb.

In general, a large Kaff value indicates high mAb affinity. A high affinity antibody, with a Kaff value between 107 and 1012 L/mol, can limit the consumption of antibody and antigen in immunoassay development. Through fitting the curve in Figure f, [Ab]t values of 2.14 × 109, 4.94 × 109, and 2.50 × 109 mol/L were obtained at the corresponding coating concentrations of 1, 0.3, and 0.1 μg/mL, respectively. Substituting into the calculating formula, the Kaff value of our mAb was 3.19 × 109 L/mol. In addition to establishment of the GICA, this antibody was also applied in immunoaffinity column development for the pretreatment of positive samples containing CBX or CYA.[35]

The cross-reactivity of our mAb is shown in Table . As shown, it has little cross-reactivity with QCA, with an IC50 value of 25.5 ng/mL and cross-reactivity (CR) of 7.3%. However, it shows no cross-reactivity with other quinolones (CR < 1.0%), including OLA, MEQ, QCT, and MQCA. This indicated that the methyl group had a significant influence on antibody generation in the mouse.

Table 2

Cross-Activity of mAb

GICA Principle and Establishment

The schematic diagram showing the GICA principle is shown in Figure . The GNP-labeled mAb was first conjugated with target analytes (CBX or CYA) to form the GNP-labeled-mAb–CBX (or CYA) complex. The complex flowed with the standard or sample solution from the sample pad to the absorption pad due to the capillary action of the absorption pad. When the solution reached the T zone on the nitrocellulose (NC) membrane, the unconjugated GNP-labeled mAb conjugated with the hapten–OVA on the T line. When it reached the C zone, the GNP-labeled mAb always conjugated with the goat antimouse IgG antibody on the C line. If the sample was negative, the GNP-labeled mAb conjugated with the hapten–OVA and generated a red line in the T zone. If the sample was weakly positive, some of the GNP-labeled mAb could also react with the hapten–OVA on the T line and produce a light red line in the T zone. If the sample was positive, all of the GNP-labeled mAb conjugated with the CBX (or CYA), and no extra GNP-labeled mAb could conjugate with the hapten–OVA sprayed on the T line. Therefore, the T zone was colorless. Meanwhile, the C zone would always appear as a red line because of the conjugation between the GNP-labeled mAb and the goat antimouse IgG antibody. If the C line was colorless, this indicated that the strip was invalid.

Figure 5

Schematic diagram of strip for GICA. (a) Planar view of the strip; (b) Schematic illustration in negative, weakly positive, and positive sample.

The coating antigen on the T line and the amount of the mAb used for GNP labeling was optimized. As shown in Figure a,b, when the coating antigen was hapten–OVA and the amount of mAb was 8 μg/mL GNP solution, the T line color was relatively lighter at a CBX concentration of 5.0 ng/mL, which showed that the inhibition to CBX was relatively better.

Figure 6

Optimization for GICA. (a) Coating antigen on the T line; (b) mAb amount for GNP labeling; (c) sample suspension solution.

The sample resuspension solution plays an important role in actual sample analysis. Three surfactant species (Tween-20, ON-870, and Triton-100) at 3% in 0.01 M pH 7.4 phosphate-buffered saline (PBS) were optimized. From Figure c, it can be seen that the GNPs clumped on the sample pad, which resulted in the T line and C line color being really light. Compared with PBS, the addition of the surfactant increased the flow of sample solution on the strip. However, it also caused visible GNP aggregation on the sample pad when the surfactant was Tween-20 or ON-870. Consequently, 0.01 M PBS pH 7.4 with 3% Triton was chosen as the chicken breast sample resuspension solution for analysis of CBX and CYA in this work.

Under the above optimal conditions, a series of fortified samples (0, 2.0, 4.0, 8.0, 10, 20, and 40 μg/kg) were tested to establish the GICA method for detection of CBX and CYA residues in chicken breast. As shown in Figure a,c, the cut-off values when the T line became colorless were 20 μg/kg for CBX and 40 μg/kg for CYA. The lowest concentrations when the T line color become too light to see with the naked eye, defined as the visual limit of detection (LOD), were 8 μg/kg for both CBX and CYA. These results can be used for semiquantitative analysis of unknown samples. If the CBX (or CYA) content of a sample was equal to or greater than 20 μg/kg (or 40 μg/kg), it was considered a positive sample. If the CBX (or CYA) content was between 8 and 20 μg/kg (or 40 μg/kg), it was a weakly positive sample, while if the CBX (or CYA) content was below than 8 μg/kg, the sample was negative.

Figure 7

GICA analysis for CBX and CYA in spiked chicken breast. (a,c) are strip image results. 1 = 0, 2 = 2.0 μg/kg; 3 = 4.0 μg/kg; 4 = 8.0 μg/kg; 5 = 10 μg/kg; 6 = 20 μg/kg; 7 = 40 μg/kg. (b,d) are standard curves established for GICA analysis.

For quantitative analysis, results were evaluated using a hand-held strip scanner. The standard curves established with the chicken matrix are shown in Figure b,d. Twenty blank chicken breast samples were tested, and the corresponding CBX and CYA concentrations were calculated using the matrix standard curve. The average plus three times the standard deviation was defined as the calculated LOD (cLOD). According to this calculation, the cLOD was 2.92 μg/kg for CBX and 2.68 μg/kg for CYA. Therefore, the proposed GICA and hand-held strip scanner could be applied to screening of CBX or CYA residues in chicken samples on site.

Validation of GICA with LC–MS/MS

To evaluate the accuracy, recovery tests were performed. Samples were spiked with CBX (or CYA) at concentrations of 4, 8, or 12 μg/kg. From Table , we can see that the recovery rate ranged from 89.5 ± 8.9 to 109.6 ± 10.3% for CBX and CYA analysis, with coefficients of variation (CVs) ranging from 9.0 to 14.6% in chicken breast. The LC–MS/MS results further confirmed the proposed GICA, with recovery rates ranging from 93.9 ± 8.9 to 114.4 ± 8.5%, and CVs ranging from 6.8 to 9.9%. These results further confirmed the feasibility of the proposed GICA method.

Table 3

Analysis of CBX and CYA in Spiked Chicken Breast by GICA and LC–MS/MS (n = 3)

		GICA		LC–MS/MS
analytes	spiked level (μg/mL)	recovery rate (%) ± SD	CV (%)	recovery rate (%) ± SD	CV (%)
CBX	4	89.5 ± 8.9	9.9	93.9 ± 8.9	9.4
	8	98.8 ± 12.1	12.2	96.1 ± 8.0	8.3
	12	107.8 ± 7.5	9.0	99.6 ± 7.3	7.4
CYA	4	98.9 ± 14.4	14.6	106.6 ± 10.4	9.9
	8	109.6 ± 10.3	9.4	114.4 ± 8.5	7.6
	12	108.1 ± 9.7	9.0	106.3 ± 7.3	6.8

Conclusions

In this work, we designed a new hapten for production of a mAb against CBX and CYA. Based on this antibody, a GICA was established for semiquantitative and quantitative screening of CBX and CYA residues in chicken breast samples. The detection process could be completed within 15 min. Recovery tests validated the accuracy, and the LC–MS/MS method further validated the results. In conclusion, the developed GICA method could be applied for rapid detection of CBX and CYA residues in chicken breast samples in the field.

Experimental Section

Materials and Reagents

Standards, including CBX, CYA, OLA, QCT, and MEQ, were purchased from J&K Scientific Ltd. (Beijing, China). The carrier proteins KLH and OVA were obtained from Sigma-Aldrich (St. Louis, MO, USA). Coupling agent (25% glutaraldehyde solution, m/v), immunologic adjuvants, including Freund’s complete adjuvant (FCA) and Freund’s incomplete adjuvant (FICA) were also purchased from Sigma-Aldrich. The horseradish peroxidase (HRP)-labeled goat antimouse immunoglobulin was obtained from Kangcheng Bioengineering Co. (Shanghai, China). The reagents used for cell fusion, including polyethylene glycol (PEG, MW 1450), RPMI-1640 medium, fetal bovine serum, and hypoxanthine–aminopterin–thymidine (HAT) and hypoxanthine–thymidine supplements were obtained from Thermo Fisher Scientific Inc. (Shanghai, China). The materials used for assembling the strip sensor, including absorption pad (SAP-Z80), sample pad (CB-SB08), NC membrane, and polyvinylchloride (PVC) sheet (DB-6), were all obtained from JieYi Biotech. Co. Ltd. (Shanghai, China). The handheld GICA analyzer was purchased from Wuxi Determine Biotech. Co. Ltd. (Wuxi, China). Chicken breast samples were bought from Auchan (Wuxi, China).

The scheme of CBX–hapten synthesis is shown in Figure b, and comprised three steps, as follows.

Synthesis of 2-Methylquinoxaline 1,4-Dioxide

The compounds 2-methylquinoxaline (5.00 g, 34.7 mmol) and 3-chlorobenzoperoxoic acid (29.9 g, 174 mmol) were dissolved in dichloromethane (50.0 mL). After stirring overnight at room temperature, this reaction mixture was poured into ice/water. The aqueous layer was extracted with dichloromethane. The combined organic layers were washed with brine, dried over Na2SO4, and concentrated. The residue was purified on a silica column to give 2-methylquinoxaline 1,4-dioxide (4.10 g) as a yellow solid with a yield of 67.2%.

Synthesis of 2-Formylquinoxaline 1,4-Dioxide

A solution of 2-methylquinoxaline 1,4-dioxide (1.00 g, 5.68 mmol) in 1,4-dioxane (10.0 mL) was added to selenium dioxide (1.26 g, 11.4 mmol) and stirred overnight at 100 °C. The reaction mixture was then poured into ice/water, and the aqueous layer was extracted with ethyl acetate. The combined organic layers were washed with brine, dried over Na2SO4, and concentrated. The residue was purified on a silica column to give 2-formylquinoxaline 1,4-dioxide (800 mg) as a yellow solid with a yield of 72.7%.

Synthesis of 2-(Hydrazonomethyl)quinoxaline 1,4-Dioxide (Hapten)

A solution of 2-formylquinoxaline 1,4-dioxide (500 mg, 2.63 mmol) in dichloromethane (5 mL) was added dropwise to hydrazine hydrate (410 mg, 7.88 mmol) and stirred overnight at room temperature. The reaction mixture was poured into ice/water; then, the aqueous layer was extracted with ethyl acetate. The combined organic layers were washed with brine, dried over Na2SO4, and concentrated. The residue was purified on a silica column to give the hapten (300 mg) as a yellow solid with a yield of 55.9%.

The structure of the hapten was characterized by 1H NMR spectrometry and LC–MS/MS analysis.

Antigen Preparation

The antigen was synthesized using the glutaraldehyde method.[36] In detail, 20 mg hapten was dissolved in 2 mL dimethylformamide and then 600 μL of glutaraldehyde, diluted ten times using PBS (0.1 M, pH 7.4), was added dropwise. Thirty minutes later, the activated solution was divided into three equal volumes and each was added drop by drop into a carrier protein solution (KLH, BSA or OVA, 5 mg/mL in 0.1 M carbonate buffer solution pH 9.6). After 2 h of stirring the reaction at room temperature, the conjugates were dialyzed six times against 0.01 M PBS (pH 7.4). Then, hapten–KLH, hapten–BSA, and hapten–OVA were obtained and characterized using an UV–visible spectrophotometer (Agilent, Santa Clara, CA, USA). The hapten–KLH was used as the immunogen, with hapten–BSA and hapten–OVA used as the coating antigens.

Immunization Schedule

Female BALB/c mice (6–8 weeks old) were immunized to produce antibodies against CBX and CYA.[37] Before immunization, the immunogen was emulsified with isometric FCA or FICA. Each mouse was then injected subcutaneously with 100 μg hapten–KLH emulsified with FCA for the first immunization. Every twenty one days, the mouse was given a booster injection of 50 μg hapten–KLH emulsified with FICA. After four immunizations, the serum of each mouse was assessed using ic-ELISA. The detailed ic-ELISA procedure has been described in our previous publication.[38] The mouse with the highest titer and the highest inhibition with CBX and CYA was sacrificed for cell fusion. Three days before cell fusion, this mouse was injected intraperitoneally with 25 μg hapten–KLH without any immunologic adjuvant.

Cell Fusion and Hybridoma Screening

As described in our previous reports,[38,39] spleen cells and sp 2/0 cells were fused at a ratio of 1:5–10 using the PEG method. Through HAT medium culture screening, the hybridoma cells would survive culture under 37 °C, 5% CO2. The cell supernatants were further screened using ic-ELISA. The wells with the highest titer and highest inhibition with CBX and CYA were subcloned using the limiting dilution method. After three rounds of subcloning and screening of cell supernatants, a single cell mass was picked out and cultured at an extended scale. Then, this cell line was frozen and stored in liquid nitrogen.

Ascites Preparation and Purification

Before being inoculated intraperitoneally with hybridoma cells, BALB/c mice were injected with sterilized paroline (1 mL/mouse). After seven to ten days, ascites were collected using an injection syringe and then centrifuged for 10 min at 8000 rpm. The AKTA pure protein purification system (GE Healthcare Little Chalfont, UK) was used to obtain pure mAb. The concentration of mAb was tested using a NanoDrop One (Thermo Fisher Scientific, Waltham, MA, USA).

The mAb characterization indicators included sensitivity (half inhibition concentration, IC50), antibody subtype, affinity (Kaff), and CR %.

ic-ELISA Assay Optimization and Sensitivity

The ic-ELISA assay buffer was optimized by testing various sodium chloride concentrations (0.4, 0.8, 1.6, 3.2, and 6.4% by mass fraction), different pH values (4.7, 6.0, 7.4, 8.8, and 9.6), and methanol concentrations in PBS (0, 10, 20, 30, and 40% by volume fraction). The IC50 at each condition was obtained using the standard curve generated using a series of CBX solutions (0, 0.14, 0.41, 1.23, 3.70, 11.1, 33.3, and 100 ng/mL). The ratio of absorption values of blank control (Amax) and IC50 value (Amax/IC50) was used to evaluate the ic-ELISA performance.

Antibody Subtype

The antibody subtype was determined using a mAb isotyping ELISA kit.[18] The different HRP-labeled secondary antibodies included IgA, IgG1, IgG2a, IgG2b, IgG3, and IgM.

Affinity

As described in our previous publications, the antibody was diluted from the 1 μg/mL stock to form a gradient of 8-fold serial dilutions. The coating antigen (hapten–OVA) concentrations were 1, 0.3, and 0.1 μg/mL. The mAb affinity was tested by ic-ELISA. By fitting the antibody concentration to the absorption values, the antibody concentration when the absorbance is half of the initial value ([Ab]t) can be calculated. The affinity constant (Kaff) was calculated using the following equation, where n is the multiple of two corresponding antigen concentrations[37]

Cross-Reactivity

Other quinoxalines, including CYA, OLA, MEQ, QCT, MQCA, and QCA, were used to evaluate the cross-reactivity of the mAb. Similarly, the IC50 values of each quinoxaline were determined. The CR % could be obtained from the following equation, as described in previous reports[40]

Gold Immunochromatographic Assay

Preparation of GNPs

GNPs with a diameter of 25 nm were synthesized using the citrate reduction method[14,41] Briefly, 100 mL of chloroauric acid solution (HAuCl4·4H2O, 0.01%, w/v) was boiled under continuous stirring; then, 2.0 mL fresh trisodium citrate solution (1%, w/v) was added quickly into the boiled solution. Five minutes after the color changed to wine red, the solution was cooled to room temperature while stirring. The solution was then stored at 4 °C for further use.

Preparation of GNP-Labeled mAb

First, the mAb was diluted to 0.2 mg/mL using 0.2 M borate buffer solution (pH 8.8); then, 40 μL of 0.1 M potassium carbonate solution was added into 10 mL of the GNP solution. Next, 400 μL mAb was added to the GNP solution, and the mixture was allowed to stand for 45 min at room temperature. Finally, 500 μL BSA solution (10%, w/v) was added to the mixture and incubated for 2 h at room temperature. Then, the solution was centrifuged at 8000 rpm, 4 °C for 45 min. The supernatant was then removed, and the sediment was resuspended in 1 mL 0.01 M PBS solution with 0.02% sucrose.

Strip Fabrication

The strip for the GICA was made up of four parts, comprising the sample pad, NC membrane, absorption pad, and PVC backing sheet.[42] The hapten–OVA and goat antimouse IgG were sprayed onto the T zone and the C zone of the NC membrane, respectively. The NC membrane was placed in the middle of the PVC backing sheet, while the sample pad and absorption pad were glued at either end of the PVC sheet with a 2 mm overlap. After drying at 37 °C for 24 h, the fabricated card was cut into strips with a width of 3.85 mm. The strips were stored in a dry environment for further use.

GICA Procedure

To perform the test, 150 μL of standard (CBX or CYA) or sample solution was mixed with 50 μL GNP-labeled mAb in a microplate well. The mixture was incubated for 3 min at room temperature. The strip was then dipped into this well, so that the solution flowed from the sample pad to the absorption pad due to capillary action. After 5 min, the results could be visualized by the naked eye. For quantitative analysis, the strip was tested using a hand-held strip scanner, which could give the T/C value (the ratio of color intensity between the T line and the C line) directly within 30 s. The standard curve was established by fitting a series of CBX or CYA standard concentration (X axis) against T/C values (Y axis) using the logistic function.

Analysis of Spiked Chicken Breast Using the Strip Test

Chicken breast samples were purchased from a local supermarket and were proven to be CBX and CYA-free by LC–MS/MS analysis. A series of spiked CBX or CYA chicken breasts were examined using the developed GICA method. Each sample was pretreated as follows: 5.0 g of minced chicken breast was weighted into a 50 mL centrifuge tube; then, 20 mL of acetonitrile was added, and the tube was vibrated for 3 min. Then, the mixture was centrifuged at 5000 rpm for 5 min, after which 4 mL of the supernatant was removed and dried using pressured blowing concentrators at 45 °C. The residue was dissolved in 4.0 mL PBS (0.01 M, pH 7.4) containing 3% Triton-100 solution, and 80 μL was used for GICA analysis. The chicken substrate standard curves were established for CBX and CYA quantitative analysis.

Recovery Test

Fortified CBX or CYA chicken breast samples at three concentrations (4, 8, and 12 μg/kg) were used to evaluate the proposed GICA protocol. To validate the GICA results, these spiked samples were tested using LC–MS/MS analysis at the same time.