Ad

FPa

b

c

a

ARRA

KECA(

1

w[rEtwd(yAtsaati

T

1d

Spectrochimica Acta Part A 93 (2012) 164– 168

Contents lists available at SciVerse ScienceDirect

Spectrochimica Acta Part A: Molecular andBiomolecular Spectroscopy

j our na l ho me p age: www.elsev ier .com/ locate /saa

competitive chemiluminescence enzyme immunoassay for rapid and sensitiveetermination of enrofloxacin

ei Yua, Yongjun Wua,∗, Songcheng Yua, Huili Zhanga,b, Hongquan Zhanga, Lingbo Qub,eter de B. Harringtonc

College of Public Health, Zhengzhou University, Zhengzhou 450001, PR ChinaSchool of Chemistry and Chemical Engineering, Henan University of Technology, Zhengzhou 450001, PR ChinaDepartment of Chemistry and Biochemistry, Center for Intelligent Chemical Instrumentation, Ohio University, Athens, OH 45701, USA

r t i c l e i n f o

rticle history:eceived 28 November 2011eceived in revised form 7 February 2012ccepted 3 March 2012

eywords:

a b s t r a c t

With alkaline phosphatase (ALP)–adamantane (AMPPD) system as the chemiluminescence (CL) detectionsystem, a highly sensitive, specific and simple competitive chemiluminescence enzyme immunoassay(CLEIA) was developed for the measurement of enrofloxacin (ENR). The physicochemical parameters,such as the chemiluminescent assay mediums, the dilution buffer of ENR–McAb, the volume of dilu-tion buffer, the monoclonal antibody concentration, the incubation time, and other relevant variables

nrofloxacinhemiluminescence enzyme immunoassaylkaline phosphatase (ALP)–adamantane

AMPPD)

of the immunoassay have been optimized. Under the optimal conditions, the detection linear range of350–1000 pg/mL and the detection limit of 0.24 ng/mL were provided by the proposed method. The rel-ative standard deviations were less than 15% for both intra and inter-assay precision. This method hasbeen successfully applied to determine ENR in spiked samples with the recovery of 103%–96%. It showedthat CLEIA was a good potential method in the analysis of residues of veterinary drugs after treatment ofrelated diseases.

. Introduction

Enrofloxacin (ENR) is a fluoroquinolone antibiotic, which isidely used in the treatment and prevention of animal diseases

1] because of its very broad antimicrobial spectrum. However, inecent years, many adverse reactions and drug resistance aboutNR have been reported. The drug residues in animal muscle andissue will be potential risk for human health, which has arousedide concern. Now, most of analytical methods for residues ofrugs were based on high-performance liquid chromatographyHPLC) [2], radioimmunoassay (RIA) [3], chemiluminescence anal-sis [4–6] and enzyme-linked immunosorbent assay (ELISA) [7,8].lthough these methods are reliable and accurate, they suffer from

he following problems, which restrict their use for many sub-tances. For HPLC, the steps of correct extraction and clean up

re vital, which are time consuming. Sample preparation can takepproximately two-thirds of the total HPLC analysis time. Addi-ionally, in the HPLC analysis procedure, using organic solventsncreases the cost for purchase and disposal. The problems of RIA

∗ Corresponding author at: 100 Kexue Avenue, Zhengzhou City, PR China.el.: +86 37167781450.

E-mail address: wuyongjun@zzu.edu.cn (Y. Wu).

386-1425/$ – see front matter © 2012 Elsevier B.V. All rights reserved.oi:10.1016/j.saa.2012.03.001

© 2012 Elsevier B.V. All rights reserved.

are often associated with the radioisotope, which is dangerous tothe operators and the environment, so this method is well regulatedand must be used in specialized laboratories. Chemiluminescenceanalysis is a sensitive method usually used for detection of manytypes of drug residues, but it always should be coupled with a sep-aration method because of the low selectivity of the CL reaction.ELISA is a specific method, but the sensitivity of the conventionalcolorimetric detection is relatively low [8]. Therefore, developinga simple, rapid, sensitive, and specific method to detect the drugresidues in animal food has practical significance. Chemilumines-cence immunoassay (CLIA) satisfies these requirements, becauseit is a combination of sensitive chemiluminescence detection andspecific immunosorbent assay. More and more people have a pref-erence for CLIA, and its application prospects have been involvedin medicine [9–11], environment [12,13], food [14,15] and so on[10,16]. In chemiluminescence enzyme immunoassay (CLEIA), var-ious enzymes, such as alkaline phosphatase (ALP) and horseradishperoxidase (HRP), are labeled on an antibody (or antigen) whichcan catalyze luminescent regents to produce an enhanced chemi-luminescence after the competitive immunoreaction. Because ofthe high sensitivity and specificity, CLEIA has been widely used in

drug residues assays [17–19].

In this paper, we have developed a competitive chemilu-minescence enzyme immunoassay based on the ALP–AMPPDchemiluminescence system to detect ENR residues in

Acta Part A 93 (2012) 164– 168 165

ao

2

2

tEfJPAbwt

sb0aopsw

2

Gmutwco

2

2

adca0amfAsftE

2

Esmtibsf

50µL ENR Calibrators (0-1000p g/mL)

Micr otiter plate coa ted with ENR- McAb

Washing with PBST bu ffer for 5 ti mes

25µL AMPPD

Chemi lum inesce nence assa y

37ºC Incuba ting for 60min

50µL ENR-AL P

experimental conditions, according to the basic steps of Fig. 1, therelative luminescence intensity units (RLU) were measured andcompared. From Fig. 2, one can see that the RLU of the EDC method

30000

60000

90000

12000 0

15000 0

18000 0

21000 0

RL

U

EDC method

mixed an hydride method

F. Yu et al. / Spectrochimica

nimal-derived food, which will help to realize the rapid screeningf ENR residues and curb effectively the drug abuse of ENR.

. Experimental

.1. Chemicals

Enrofloxacin standard was provided by the National Insti-ute for the Control of Pharmaceutical and Biological Products.nrofloxacin monoclonal antibodies (ENR–McAb) were purchasedrom Santa Cruz Biotechnology. ALP was from pharmaceutical Co.,erry (Japan). BSA was purchased from Sigma. Cane sugar, as wellroCell A and B of ALP–AMPPD system were obtained from Sino-merican Biotechnology Company. Deionized water was purifiedy Milli Q system (Waters, Milford, MA). Other chemical reagentsere supplied by the Tianjin or Beijing Chemical Agents, and all of

hese were of analytical grade.The buffers used in this paper were PBS buffer (0.01 mol/L

odium phosphate buffered saline, pH of 7.2); washing and dilutionuffer (PBST, PBS buffer with 0.05% Tween 20); coating buffer (CB,.05 mol/L carbonate and bicarbonate buffer solution, pH of 9.6)nd blocking buffer (PBS buffer with 1% BSA and 4% cane sugar, pHf 7.2). ENR stock solutions with the concentration of 1 mg/mL wasrepared by dissolving 50 mg ENR in 0.03 mol/L sodium hydroxideolution (NaOH) and stored at 4 ◦C. The standard concentrationsere from the stock solution by serial dilutions of 0.03 mol/L NaOH.

.2. Instruments

Chemiluminescence immunoassay analyzer (MP280 Beijing Taieke Letter Biological Technology Co., Ltd.) was performed to deter-ine the luminescence. The 96-well polystyrene microtiter plates

sed in the assay were purchased from Shenzhen Jincanhua Indus-rial Co., Ltd. The UV–vis spectrophotometer (UV-2450 Shimadzu)as used to identify the coupling object. Electrical thermostatic

ultivation cabinet was from DHG-9146A Shanghai Jing Hong Lab-ratory Instrument Co. Ltd.

.3. Procedures

.3.1. Synthesis of ALP–ENR conjugatesThe linkage of 2 imine-carbon (1-ethyl-3-(3-dimethyl-

minopropyl)-carbodiimide, EDC) follows the proceduresescribed in reference [20] with a small change was performed toouple ALP with ENR in this experiment, and the specific steps weres follows: a 20 mg ENR was weighed accurately and dissolved in.1 mol/L NaOH solution, then 10 mg N-hydroxysuccinimide (NHS)nd 12.5 mg EDC were added to the solution. The mixture wasixed with 1 mL dimethylformamide (DMF) and then incubated

or 24 h at 25 ◦C, which was solution A. A 4.0 mg/mL solution ofLP was prepared in 0.01 mol/L PBS buffer (pH of 7.2), which wasolution B. Solution A was added to solution B slowly and stirredor 4 h at 25 ◦C in the dark, then dialyzed for 6 days with 200 timeshe volume of PBS buffer solution (v/v). The final product was theNR–ALP conjugate.

.3.2. Preparation of antibody coated tubesCarbonate buffer solution was added into 25 �g/mL of

NR–McAb solution and then mixed well for coating. The mixedolution was added to the bottom of the wells of polystyreneicrotiter plates (100 �g/well) and incubated at 37 ◦C for 150 min,

hen the microtiter plates were coated with ENR–McAb by phys-

cal adsorption. The plates were subsequently washed with PBSTuffer for 3 times to eliminate the free antibodies, and the uncoatedites were blocked by adding 100 �g/well blocking buffer at 4 ◦Cor 30 min. Then, the block buffer solutions were poured out and

Fig. 1. The procedure of CLEIA.

the wells were washed with PBST buffer 3 times again. Finally, thecoated microtiter plates were dried up and stored at 4 ◦C for furtheruse.

2.3.3. Procedure of CLEIAThe procedure of CLEIA is presented in Fig. 1. First, 50 �L ENR

calibrators and ENR–ALP (desirable ratios) were added to thepolystyrene microtiter plate coated with ENR–McAb, respectively,then incubated for 60 min with gentle shaking at 37 ◦C. After thecompetitive reaction, the polystyrene microtiter plate was emp-tied, washed 5 times with PBST buffer and blotted dry against tissuepaper. A 25 �L volume of chemiluminescence substrate solutionwas added to each well and the luminescence was measured afterincubating for 3 min at room temperature in the dark.

3. Results and discussion

3.1. Optimization and identification of ALP–ENR conjugates

3.1.1. Optimization of parameters in ALP–ENR conjugates stepsBecause of the carboxyl in ENR, two methods were chosen to

couple ALP with ENR in this experiment, which were linkage of 2imine-carbon (EDC) and mixed anhydride methods. For the same

0 20 0 40 0 60 0 80 0 1000 120 0

0

conc entration of EN R( pg/mL)

Fig. 2. The final RUL of two synthesis methods.

166 F. Yu et al. / Spectrochimica Acta Part A 93 (2012) 164– 168

wtaiaimo4atiwbgt

3

uw2oibr

B

3

idt

3

t((sFtlw

3

co

Fig. 4. The effects of coating buffer. The four curves correspond to four different

was better, so the hot-pack of 150 min conditions was chosen forthis experiment.

Fig. 3. The ultraviolet spectroscopy of ALP, ENR and ALP–ENR conjugate.

as greater than that of the mixed anhydride method. Thereforehe EDC method was adopted. Then the dialysis time, preservationnd the diluted solution of conjugate were optimized. The exper-ment created dialysis times of the coupling for 3 d, 4 d, 5 d, 6 d,nd 7 d. With the time of dialysis increasing, the RLU of the systemncreased slowly until the sixth day. Therefore, 6 days was the opti-

al dialysis time. Then the conjugate was saved for 1 month usingne of the following three ways: saving directly at 4 ◦C, saving at◦C after the addition of 1% sodium azide, and saving at −20 ◦C afterdding an equal volume of 60% glycerin. The results showed thathe RLU was similar. Considering sodium azide and glycerin maynterfere with the results of the experiment, the coupling objects

ould best be stored at 4 ◦C without modification. In addition, twouffers were used to be as the diluted solution of ALP–ENR conju-ates, and they were 0.05 mol/L CB and 0.01 mol/L PBS. It was clearhat the RLU was more stable using PBS as diluted solution.

.1.2. Identification of ALP–ENR conjugatesTo identify whether ENR was labeled with ALP successfully, the

ltraviolet spectra of ALP, ENR, and ALP–ENR conjugates spectraere given in Fig. 3. There was one absorption peak for ALP at

80 nm, three for ENR at 280 nm, 316 nm, and 328 nm and threethers for ENR–ALP conjugate at 266 nm, 322 nm, and 324 nm. Fig. 3ndicated that ENR–ALP conjugate was prepared successfully. Theinding ratio B was calculated to be 8:1 using Eq. (1). The bindingatio showed that the conjugate was effective [21].

= εENR−ALP − εALP

εENR= (AENR−ALP/CENR−ALP) − (AALP/CALP)

AENR/CENR(1)

.2. Optimization of the parameters in ENR–McAb coating steps

The preparation of the ENR–McAb microtiter plates was verymportant for the assay. Several relevant parameters, including theilution buffer of ENR–McAb, the concentration of ENR–McAb, andhe incubation conditions were studied and optimized.

.2.1. Optimization of the dilution bufferBefore coating, ENR–McAb was diluted to the same concentra-

ions of 2.0 �g/mL by different buffers which were 0.05 mol/L CBpH of 9.6), 0.01 mol/L PBS (pH of 7.2), phosphate salt–citric acidPSCA, pH of 5.0) and 0.1 mol/L Tris–HCl buffer (pH of 8.5). In theame experiment conditions, according to the basic procedure ofig. 1, the immunoreaction efficiency was compared. It gave a bet-er correlation between the log RUL and the log CENR and a strongeruminous intensity when CB buffer was used (Fig. 4). So CB buffer

as selected for dilution of ENR–McAb.

.2.2. Optimization of the concentration of ENR–McAbThe concentration of ENR–McAb could impact the coating effi-

iency, so the ENR–McAb solution was diluted to concentrationsf 0.5 �g/mL, 1 �g/mL, 2 �g/mL, 3 �g/mL, and 5 �g/mL, and the

buffers which were phosphate salt-citric acid (PSCA, pH 5.0), Tris–HCl (pH 8.5), PBS(pH 7.2) and CB (pH 9.6).

immunoreaction efficiency of the different concentrations wasstudied with the same polystyrene microtiter plate. A good lin-ear relationship between log RLU and log CENR appeared at theENR–McAb concentration of 2.0 �g/mL in Fig. 5. And consider-ing the test cost, the concentration of ENR–McAb was selected as2.0 �g/mL.

3.2.3. Optimization of the incubation conditionsThe hot-pack (i.e. 37 ◦C) in the time of 60–180 min and the cold-

pack (i.e. 4 ◦C) in the time of 16–24 h were studied in this paper. Inthe course of hot-pack, the luminous intensity of different concen-tration solutions increased with increasing time, but the regularitywas the best at 150 min, suggesting that the reaction equilibriumwas achieved at 150 min, and longer time resulted in more nonspe-cific adsorption. Therefore, coating time of 150 min was chosen. Inthe course of cold-pack, the results indicated that at 22 h the reac-tion reached equilibrium. Comparing hot-pack with cold-pack, itwas found the efficiency and the linear relationship of the former

Fig. 5. The effects of the concentration of ENR–McAb. The five curves correspond todifferent concentrations of ENR–McAb (�g/mL).

F. Yu et al. / Spectrochimica Acta Part A 93 (2012) 164– 168 167

Fr

3

3

emdFkfrwetmoo

3

(uttsta

3

3

scrilc1

TT

Table 2The comparison of the four regression equations and R2.

Fitted curve Regression equation R2

C–RLU Y = −13.789x + 18068 0.9565log C–RLU Y = −20180x + 65236 0.9945log C–log RLU Y = −0.9961x + 6.6992 0.9955log C–A/A0 Y = −0.8839x + 2.8577 0.9946

y = -0.9961 x + 6.699 2

R2 = 0.99 55

63.

83.

4

24.

23.382.62.42.

logC(ENR)

logR

LU

Fig. 7. The calibration curve of CLEIA.

Table 3The precision of the method at different concentration of ENR (n = 5).

CENR (pg/mL) The intra-assay RSD (%) The inter-RSD (%)

350 9.1 13.3650 8.1 12.21000 7.3 9.5

Table 4The average recoveries (n = 5).

Theoreticalconcentration (pg/mL)

Detectableconcentration (pg/mL)

Averagerecovery (%)

350 360.7 103.1400 401.2 100.3500 489.4 97. 9

ig. 6. The kinetics curves of the chemiluminescence reaction. The six curves cor-espond to different concentrations of ENR (pg/mL).

.3. Optimization of chemiluminescence assay

.3.1. Optimum chemiluminescence timeAt room temperature, this study performed integration on the

mitted photons every 0.1 s to detect the final RUL of the chemilu-inescence system. The RLU was plotted with respect to time for

ifferent quantities of ENR to model the enzyme kinetics (Fig. 6).rom Fig. 6, it can be seen that the classical single substrate enzymeinetics were observed with the faster rate in the initial periodollowed by a decline in rate afterwards chemiluminesence. Thiseaction followed Michaelis–Menten kinetics. The initial periodas the region of greatest sensitivity. Because greater there is inter-

st in a fast assay, the initial period was selected for integrationime for measuring the chemiluminescence. An 8 min measure-

ent time was chosen. Of course by integrating over longer periodsf time will improve the sensitivity of the method with a trade-offf a longer analysis time.

.3.2. Optimum quantity of chemiluminescent substrateThe chemiluminescent substrate was diluted with CB buffer

VAMPPD:VCB of 1:4), then 25 �L/wells and 50 �L/wells were eval-ated, and the RLU were determined respectively. From Table 1he signals obtained from the 50 �L/wells were significantly bet-er than the 25 �L/wells, but the 25 �L/wells furnished adequateignal at the upper limits of the analysis. Because the cost of usinghe 50 �L/wells was significant as well and 25 �L/wells furnisheddequate sensitivity, the smaller well size was selected.

.4. Methodology evaluation

.4.1. Calibration curve and the linear rangeUnder the optimal reaction conditions, a series of standard ENR

olutions were prepared and determined by CLEIA, and four fittedurves of the RLU and the concentration of ENR were obtained. Theegression equation and the R2 of calibration curves were reported

2

n Table 2. In view of the R , log C–log RLU (base 10) was the best. Soog C–log RLU was chosen as the calibration curve. The calibrationurve was given in Fig. 7, and the linear range was from 350 to000 pg/mL.

able 1he effects on the RUL of the volume of the chemiluminescent substrate.

Volume/well(�L)

CENR (pg/mL)

0 100 300 500 700 1000

25 23, 500 19,036 16,210 9326 6210 493650 31, 952 28,541 26,427 20,372 16,337 13,956

600 578.0 96.3850 842.6 99.11000 1032.6 103.3

3.4.2. Detection limit, precision, and recoveryThe limit of detection (LOD) was the minimum amount of ENR

which could be markedly distinguished from the background sig-nal S0. It was calculated to be 0.24 ng/mL by subtracting twice thestandard deviation (SD) from the background signal and convert-ing to units of concentration using the sensitivity obtained formthe calibration line. The background signal was the average RLU ofthe blank solutions from 5 replicates and the SD was the standarddeviation. Three different concentrations of ENR in one group weremeasured for 5 times to obtain the intra-assay precision. Inter-assay precision was calculated by measuring the ENR of differentgroups in the assays. The intra- and inter-assay relative standarddeviations varied from 7.3% to 13.3%, which were given in Table 3.Known amounts of ENR were added to blank sample solutions, andthen the content of ENR was determined. Each sample was analyzedwith 5 replicates, and the recoveries ranged between 96% and 103%were given in Table 4.

4. Conclusions

In modern farming practice, to increase production and sat-isfy animal health requirements, many veterinary drugs are givento food-producing animals to prevent and treat diseases, and to

1 Acta

srApfltqzvm

Anoi(mEttiaotd

A

R

[

[

[

[

[

[

[

[[[

[

68 F. Yu et al. / Spectrochimica

horten feeding time. But an illegal or unsuitable use increases theisk of introducing harmful residues into the human food chain.

series of legislative measures have been defined to assure a highrotection level for consumers in many countries. For example, U.S.ederal regulations have set standards of fluoroquinolone residuesimits in catfish to 5 �g/kg, and in Alabama and Mississippi ashe main area for processing catfish, have stipulated that fluoro-uinolone residues in catfish cannot be detected, namely set to aero limit. So the techniques of detecting the residues in foods areery important to supervise the implementation of these legislativeeasures.In this paper, a highly sensitive and specific CLEIA using

LP–AMPPD system was devised and applied to the determi-ation of ENR residue in food. Through optimizing the dosagef immunoreagents and the parameters of the competitivemmunoreaction, the method improved the limit of detectionLOD) to picograms/milliliter (pg/mL). Enrofloxacin had been deter-

ined by HPLC method which furnished a 1 ng/mL of LOD. AnLISA method for ENR had a linear range of 1–1000 ng/mL andhe detection limit was 4 �g/kg [22]. The results showed thathe immunoreaction efficiency for ENR was positive. Howevermproved accuracy, precision, and lower detection limits are desir-ble. The CLEIA method yielded a good application for the analysisf ENR and could be expanded to other antibiotics. In the future,he CLEIA kits will be developed to realize the real-time, on-lineetection of ENR residue.

cknowledgment

This work was supported by Henan Province Major Publicesearch Project.

[

[

Part A 93 (2012) 164– 168

References

[1] F.J. Angulo, K.R. Johnson, R.V. Tauxe, M.L. Cohen, Microb. Drug Resist. 6 (2000)77–83.

[2] A.L. Cinquina, P. Roberti, L. Giannetti, F. Longo, R. Draisci, A. Fagiolo, N.R. Brizioli,J. Chromatogr. A 987 (2003) 221–226.

[3] S. Collins, M. O’Keeffe, M.R. Smyth, Analyst 119 (1994) 2671–2674.[4] S. Morandi, C. Focardi, M. Nocentini, M. Puggelli, G. Caminati, Food. Anal. Meth-

ods 2 (2009) 271–281.[5] G.H. Wan, H. Cui, H.S. Zheng, Y.Q. Pang, L.J. Liu, X.F. Yu, Luminescence 21 (2006)

36–42.[6] W. Liu, Z.J. Zhang, Z.Q. Liu, Anal. Chim. Acta 593 (2007) 187–192.[7] M.A. Burkin, I.A. Galvidis, J. Agric. Food Chem. 58 (2010) 9893–9898.[8] C.A. Marquette, P. Hezard, A. Degiuli, L.J. Blum, Sens. Actuators B: Chem. 113

(2006) 664–670.[9] W. Dungchai, W. Siangproh, J.M. Lin, O. Chailapakul, S. Lin, X.T. Ying, Anal.

Bioanal. Chem. 6 (2007) 1965–1971.10] M. Tudorache, I.A. Zdrojewska, J. Emneus, Biosens. Bioelectron. 22 (2006)

241–246.11] M.X. Zhou, C.Y. Guan, G. Chen, X.Y. Xie, S.H. Wu, J. Zhejiang Univ. Sci. B 6 (2005)

1148–1152.12] R.Q. Zhang, H. Nakajima, N. Soh, K. Nakano, T. Masadome, K. Nagata, K.

Sakamoto, T. Imato, Anal. Chim. Acta 600 (2007) 105–113.13] T.B. Xin, X. Wang, H. Jin, S.X. Liang, J.M. Lin, Z.J. Li, Appl. Biochem. Biotechnol.

158 (2009) 582–594.14] K. Kloth, M. Rye-Johnsen, A. Didier, R. Dietrich, E. Martlbauer, R. Niessner, M.

Seidel, Analyst 134 (2009) 1433–1439.15] M.H. Yang, Y. Kostov, H.A. Bruck, A. Rasooly, Anal. Chem. 80 (2008)

8532–8537.16] H.M. Zhang, W.T. Li, Z.H. Sheng, H.Y. Han, Q.G. He, Analyst 135 (2010)

1680–1685.17] C.L. Xu, D.H. Yu, X.G. Chu, C.F. Peng, Z.Y. Jin, Anal. Lett. 39 (2006) 709–727.18] J. Choi, Y.T. Kim, J.H. Lee, Analyst 135 (2005) 2445–2450.19] S.X. Zhang, Z. Zhang, W.M. Shi, S.A. Eremin, J.Z. Shen, J. Agric. Food Chem. 54

(2006) 5718–5722.20] H. Watanabe, A. Satake, Y. Kido, A. Tsuji, Analyst 127 (2002) 98–103.

21] Y.J. Wu, S.C. Yu, F. Yu, N.L. Yan, L.B. Qu, H.Q. Zhang, Spectrochim. Acta A 81

(2011) 544–547.22] Z.Y. Li, Development of One-step Enzyme-linked Immunoassay for Rapid Detec-

tion of Enrofloxacin Residues in Seafood, Ocean University of China, Qingdao,Shandong, 2009.