Four Specific Hapten Conformations Dominating Antibody Specificity

Subscriber access provided by UNIVERSITY OF MICHIGAN LIBRARY

Article

Four specific hapten conformations dominating antibody specificity: quantitative structure-activity relationship analysis for quinolone immunoassay Jiahong Chen, Lanteng Wang, Lanlan Lu, Xing Shen, Xinan Huang, Yingju Liu, Xiulan Sun, Zhanhui Wang, Sergei Alexandrovich Eremin, Yuanming Sun, Zhenlin Xu, and Hongtao Lei Anal. Chem., Just Accepted Manuscript • Publication Date (Web): 17 May 2017 Downloaded from http://pubs.acs.org on May 18, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

Four specific hapten conformations dominating antibody

2

specificity: quantitative structure-activity relationship analysis

3

for quinolone immunoassay

4 5

Jiahong Chena#, Lanteng Wanga#, Lanlan Lua, Xing Shena, Xin-an Huangb, Yingju Liuc,

6

Xiulan Sund, Zhanhui Wange, Sergei Alexandrovich Ereminf, Yuanming Suna, Zhenlin

7

Xua, Hongtao Leia*

8 9

a

Guangdong Provincial Key Laboratory of Food Quality and Safety, South China Agricultural

10

University, Guangzhou 510642, China

11

b

12

Guangzhou University of Chinese Medicine, Guangzhou 510405, China

13

c

14

University, Guangzhou 510642, China

15

d

16

University, Wuxi, Jiangsu 214122, China

17

e

18

Agricultural University, Beijing 100094, China

19

f

20

119991 Moscow, Russia

Tropical Medicine Institute &South China Chinese Medicine Collaborative Innovation Center,

Department of Applied Chemistry, College of Materials and Energy, South China Agricultural

State Key Laboratory of Food Science and Technology, School of Food Science of Jiangnan

Department of Veterinary Pharmacology and Toxicology, College of Veterinary Medicine, China

Faculty of Chemistry, M.V. Lomonosov Moscow State University, Leninskie gory 1, Building 3,

21 22

* Corresponding author. Phone: +8620-8528 3448. Fax: +8620-8528 0270. E-mail: [email protected] (Hongtao Lei).

23 24

#

Equal contribution

25 1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

26

Notes

27

The authors declare no conflict of interest.

2


Page 2 of 28

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


28

Abstract: Antibody-based immunoassay methods have been important tools for

29

monitoring drug residues in animal foods. However, due to limited knowledge of the

30

quantitative structure-activity relationship between a hapten and its resultant antibody

31

specificity, it is still a huge challenge for antibody production with a desired

32

specificity. In this study, the three-dimensional quantitative structure-activity

33

relationship (3D QSAR) was analyzed in accordance with the cross-reactivity of

34

quinolone drugs reacting with the antibody raised by pipemidic acid as the

35

immunizing hapten, as well as comparing with the reported cross-reactivity data and

36

their hapten structures. It was found that the specificity of quinolone antibody was

37

strongly related to the conformation of the used hapten, and the hapten conformations

38

shaped like “I”, “P” and “ф” were essential for the desired a high specificity with low

39

cross-reactivity, but the hapten conformation shaped like “Y” led to a broad

40

specificity antibody with high cross-reactivity. Almost all the antibodies against

41

quinolones could result from these four hapten conformations. It was firstly found that

42

the concrete conformations dominated the specificity of the antibody to quinolone,

43

which will be of significance for the accurate hapten design, predictable antibody

44

specificity and better understanding the recognition mechanism between the hapten

45

and the antibodies for immunoassay.

46

3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

47 48

INTRODUCTION Quinolones are a class of antimicrobial drugs widely used in the prevention and

49

treatment of animal diseases. However, their residues in animal foods have raised a

50

series of health issues, including skin reactions, phototoxicity, and hyperglycemia1-3.

51

In order to effectively monitor the quinolones abuse in animal-derived food products,

52

it is important to develop a rapid screening method for monitoring quinolone residues.

53

Compared with the traditional instrumental methods4-8, immunoassays relying on

54

antigen−antibody interaction are favored by analytical chemists due to their

55

convenient manipulation, simple sample treatment, low-cost and easy automation.

56

Some immunoassays have been developed for the detection of quinolone residues9-11.

57

It is well-known that an antibody with the desired broad or high specificity (high

58

or low cross-reactivity) is crucial to develop an immunoassay12. High specificity

59

means low cross-reactivity to structure-related compounds13,14, which is traditionally

60

favored by a single analyte analysis in one test. On the contrary, broad specificity

61

means that the high cross- reactivity to structure-related compounds is useful for the

62

monitoring a series of compounds to finish multianalyte analysis in one test15. Now

63

the broad-specificity, multianalyte recognition in one assay, seems to be a focus of

64

antibody production and immunoassay development for quinolone drugs16-18. There

65

have been a few of reports about the broad specificity of quinolone antibodies based

66

on molecule modeling. Wang et al developed a generic immunoassay using

67

ciprofloxacin antibodies for 12 fluoroquinolone antibiotics, and the obtained

68

cross-reactivity data showed the ethyl group of the piperazinyl ring exerted a limited

69

effect on antibody binding but appeared to be important during antibody production18.

70

In our previous study, pazufloxacin was employed as hapten to produce a broad

71

specific antibody recognizing 23 quinolones19, it was found that the quinolones could

72

interact with the antibody with different binding positions, and cross-reactivity was

73

mainly positively correlated with the bulky substructure containing an electronegative

74

atom at position 7, while it was negatively associated with the large bulky

75

substructure at position 1 of quinolones19. However, the topological properties of

76

haptens provided a rich structural information which could be helpful for 4


Page 4 of 28

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


77

understanding the specificity of antibody19, the reasonably predicting and designing a

78

resultant antibody specificity are still a great challenge due to the unclear

79

structure-activity relationship and the limited knowledge of the recognition

80

mechanism19.

81

In this study, pipemidic acid, a quinolone drug with a flat conformation shaped

82

like “I” at position 7, was used as a hapten to produce a polyclonal antibody, a highly

83

sensitive competitive indirect ELISA (ciELISA) was successfully constructed in a

84

heterologous coating format. Using the comparative molecular field analysis (CoMFA)

85

based on the obtained cross-reactivity of the antibody raised by pipemidic acid, the

86

three-dimensional quantitative structure-activity relationship (3D QSAR) was

87

constructed among quinolone hapten structures and the specificity of pipemidic acid

88

antibody. Moreover, comparing with the reported cross-reactivity data and typical

89

quinolone hapten structures including clinafloxacin, ofloxacin, pazufloxacin and

90

ciprofloxacin, the optimal hapten conformations for the corresponding antibody

91

specificity were investigated for the first time.

92 93

EXPERIMENTAL SECTION

94

Reagents

95

Pipemidic acid, 3,3’,5,5’-tetramethylbenzidine (TMB), bovine serum albumin

96

(BSA), ovalbumin (OVA), complete and incomplete Freund’s adjuvants,

97

1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC), Tween-20,

98

N,N-dimethylformamide (DMF) were purchased from Sigma (St. Louis, MO, USA).

99

HRP-conjugated goat-anti-rabbit IgG was obtained from Boster Biotech Corporation

100

Limited. (Wuhan, China). Rufloxacin, prulifloxacin, norfloxacin, pefloxacin,

101

enrofloxacin, oxolinic acid, racemic ofloxacin, ciprofloxacin, lomefloxacin,

102

danofloxacin, garenoxacin, pazufloxacin, clinafloxacin, gatifloxacin, marbofloxacin,

103

difloxacin, sarafloxacin, sparfloxacin and tosufloxacin were purchased from

104

Veterinary Medicine Supervisory Institute of China (Beijing, China). S-(-)-ofloxacin

105

and R-(+)-ofloxacin were purchased from Daicel Chiral Technologies Company

106

(Figure 1). All of the chemicals and organic solvents, which were analytical grade or 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

107

better, were obtained from a local chemical supplier (Yunhui Trade Co., Ltd.,

108

Guangzhou, China). Coating buffer, washing solution, blocking solution, substrate

109

buffer, stopping reagent and TMB solution used in this study were prepared as

110

previous work in our laboratory20. Standard stock solution (1 mg/mL) was prepared

111

by dissolving an appropriate amount of each standard in 0.03 mol/L sodium hydroxide

112

solution and kept at 4 °C until use. Working standard solutions (0.038, 0.31, 2.44,

113

19.53, 156.25, 1250, 10000 ng/mL) were prepared by diluting the stock solution in

114

Phosphate-buffered saline with 0.1% Tween-20 (PBST).

115

Instruments

116

UV-visible absorption measurement was performed on a UV-3010

117

spectrophotometer (Hitachi, Japan). ELISA plates were washed by a microtiter plate

118

washer DEM-3 (Tuopu, China). Absorbance was measured at a wavelength of 450 nm

119

using a Multiskan MK3 microplate reader (Thermo Labsystems, USA).

120

Preparation of hapten-protein conjugates

121

Pipemidic acid was coupled to BSA via EDC for the immunogen, and both

122

pipemidic acid and quinolone drugs (norfloxacin-OVA, pazufloxacin-OVA,

123

ciprofloxacin-OVA, gatifloxacin-OVA, lomefloxacin-OVA, sarafloxacin-OVA and

124

garenoxacin-OVA) were coupled with OVA via EDC for the use of the coating

125

antigens according to previous work with modification21.

126

Antibody production

127

Animal treatments were conducted in accordance with the guidelines of the

128

Chinese Association for Laboratory Animal Sciences. Two New Zealand rabbits

129

(1.5-2.0 kg), supplied by the Guangdong Medical Laboratory Animal Center, were

130

immunized using pipemidic acid-BSA as the immunogen to generate the polyclonal

131

antibody against pipemidic acid according to our previous work with modification20.

132

The obtained antisera from rabbits were purified by caprylic acid-saturated

133

ammonium sulfate precipitation, and the purity was confirmed by sodium dodecyl

134

sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), and then divided into

135

aliquots, labeled, and stored at -20 °C until use22.

136

ELISA procedure 6


Page 6 of 28

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


137

The ELISA was established on the basis of the common procedure of competitive

138

indirect enzyme linked immunosorbent assay (ciELISA)20. Calibration curves were

139

obtained by plotting the normalized signal B/B0 against the logarithm of analyte

140

concentration. The logarithm of pipemidic acid concentration served as X-axis,

141

whereas B/B0 (B, the average absorbance of the wells in the presence of a competitor;

142

B0, the average absorbance of the well without analyte) served as Y-axis. The 50%

143

inhibition value (IC50) were obtained using a four-parameter logistic equation that was

144

used to fit the sigmoidal curve by OriginPro 8.5 software (OriginLab Corporation,

145

Northampton, MA)23 . The equation is Y = (A–D)/[1+(X/C)E]+D

146 147

where A is the maximum response at high asymptotes of the curve, D is the

148

minimum response at low asymptotes of the curve, C is the concentration of the

149

analyte that results in 50% inhibition and E is the slope of sigmoidal curve. The limit

150

of detection (LOD) was defined as the concentration of analyte that inhibited 10%

151

(IC10)24. The dynamic working range was defined as the lower and upper

152

concentration that inhibited 20% ~ 80%25.

153

Specificity

154

Specificity of the antibody was evaluated by measuring the cross-reactivity (CR)

155

using a group of structurally related quinolone drugs. 22 compounds were selected for

156

this test (Figure 1), the obtained IC50 values were used to calculate cross-reactivities

157

as follows:

158

CR (%) = [IC50 (pipemidic acid) / IC50 (structurally related quinolone drugs)] × 100%

159

QSAR

160

CoMFA for pipemidic acid immunoassay

161

The molecular modeling was conducted using SYBYL-X 2.1 program package.

162

These 22 dataset molecules were constructed using the “SKETCH” option function;

163

then they were energy minimized using the Powell method with MMFF94 force field

164

and MMFF94 charges. The criteria of the termination and max iterations were set to

165

be 0.005 kcal/(mol×Å) and 1000, respectively. The other parameters were the defaults.

166

The molecular alignment was carried out using pipemidic acid as the template 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

167

molecule and its C-4a, C-5, N-8 and C-8a as the common core structure. Pipemidic

168

acid, prulifloxacin, rufloxacin, norfloxacin, pefloxacin, enrofloxacin, lomefloxacin,

169

danofloxacin, garenoxacin, S-(-)-ofloxacin, clinafloxacin, gatifloxacin, marbofloxacin,

170

difloxacin, sarafloxacin, sparfloxacin and tosufloxacin were classified into the

171

training set, and the others of oxolinic acid, R-(+)-ofloxacin, ciprofloxacin and

172

pazufloxacin were treated as the test set. The converted pIC50 (-log IC50) values were

173

used in analysis.

174

CoMFA steric and electrostatic interaction fields of each molecule were

175

calculated on a 3D cubic lattice. A sp3 carbon probe atom with Van der Waal radius of

176

1.52 Å and +1 charge was used to generate the steric and electrostatic filed energies.

177

The cross-validated correlation coefficient R2 (q2) and the optimum number of

178

components (ONC) were obtained using the partial least square (PLS) method with

179

leave one out (LOO) option. Using the obtained ONC, the final non-cross-validated

180

model was created. Except for using the MMFF94 charges, the other parameters or

181

options of the CoMFA were the defaults.

182

Conformation comparing analysis

183

Previously, we raised a clinafloxacin polyclonal antibody and developed a rapid,

184

specific immunoassay for clinafloxacin21. In this work, the converted pIC50 (-log IC50)

185

values of cross-reactivity data of clinafloxacin were directly used for molecular

186

modeling. The molecular alignment was carried out using clinafloxacin as the

187

template molecule and its C-4a, C-5, N-8 and C-8a as the common core structure.

188

clinafloxacin, gatifloxacin, ciprofloxacin, S-(-)-ofloxacin, rufloxacin, enrofloxacin,

189

marbofloxacin, lomefloxacin, prulifloxacin, pefloxacin, nalidixic acid tosufloxacin

190

and difloxacin were classified into the training set, and the others of danofloxacin,

191

R-(+)-ofloxacin, sparfloxacin and sarafloxacin were treated as the test set. The

192

molecular modeling was conducted using the same method and parameter as above.

193

The CoMFA of other quinolones, including pazufloxacin, ciprofloxacin and ofloxacin,

194

were conducted by literature and used to discuss in this study18,26,27.

195 196

RESULTS AND DISCUSSION 8


Page 8 of 28

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


197

Immunoreagent preparation

198

The pipemidic acid bearing a carboxylic acid group at the end of the spacer was

199

covalently coupled with a carrier protein (BSA or OVA) by the direct EDC method.

200

UV spectra were measured to assure the successful conjugation. Figure S1 shows the

201

UV spectra of BSA, OVA, pipemidic acid, pipemidic acid-BSA and pipemidic

202

acid-OVA. BSA/OVA showed a characteristic absorption peak at 280 nm, while

203

pipemidic acid had the peak at 332 nm. The absorption spectra of pipemidic acid

204

-BSA/OVA conjugates contained both absorption peaks of pipemidic acid and

205

BSA/OVA, but with somewhat blue shift. This indicated that the coupling of

206

pipemidic acid to BSA and OVA was successful28. Similar results were obtained with

207

other quinolone drugs-OVA conjugate (data not shown). The SDS-PAGE result

208

showed that the purified pipemidic acid antibody had a heavy chain observed at 50

209

kDa and a light chain at about 25 kDa. There was no superfluous band (Figure S2).

210

Thus, the purified antibody was ideal for the further investigation.

211

Optimization of ciELISA

212

To obtain the optimal sensitivity of ciELISA, the concentration of coating

213

antigen and antibody dilution time were optimized to obtain a maximum absorbance

214

(Amax) for the zero standard concentration (blank) in the range of 1.0-1.5 and the best

215

sensitivity (minimum IC50 value)29.

216

For homologous assay format (pipemidic acid-OVA for the use of the coating

217

antigen) , a poor binding affinity with a low inhibition rate of 48.05% was obtained.

218

In addition, the standard curve for pipemidic acid (Figure 2) was constructed in the

219

concentration range from 6.53 to 144.87 ng/mL and the values of LOD at 10%

220

inhibition was within 2.6 ng/mL. The results indicated that the sensitivity had 14.4

221

times lower than that of the UPLC–MS/MS method7 and the inhibition rate (48.05%)

222

was undesirable, which failed to meet the purpose of this study.

223

In order to obtain the better sensitivity, a series of coating antigens

224

(norfloxacin-OVA, pazufloxacin-OVA, ciprofloxacin-OVA, gatifloxacin-OVA,

225

lomefloxacin-OVA, sarafloxacin-OVA and garenoxacin-OVA) were compared at the

226

same coating concentration (1µg/mL) (Table S1). Surprisingly, although homologous 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

227

assay format demonstrated a better titer (titer=32000) than any one of heterogeneous

228

formats (norfloxacin-OVA, pazufloxacin-OVA, ciprofloxacin-OVA,

229

gatifloxacin-OVA, lomefloxacin-OVA, sarafloxacin-OVA, garenoxacin-OVA for the

230

use of the coating antigens, respectively), it was found that lomefloxacin-OVA

231

showed the highest inhibition rate (86.7%) with a titer of 8000.

232

The calibration curve for pipemidic acid in homologous assay and heterogeneous

233

assay format were constructed(Figure 2). Compared with the IC50 (30.76 ng/mL) and

234

LOD (2.64 ng/mL) in homologous assay format (Table S1 and Figure 2), the IC50 and

235

LOD in heterogeneous assay format (lomefloxacin-OVA as the coating antigen) were

236

5.99 ng/mL and 0.31 ng/mL, respectively, indicating that the sensitivity was better

237

than that of the homologous combination and that of the reported physio-chemical

238

method7. As a result, the combination of pipemidic acid antibody and

239

lomefloxacin-OVA was used for the further investigation due to its better sensitivity.

240

Specificity

241

The cross-reactivities of 22 kinds of quinolones were determined using the

242

developed ELISA with pipemidic acid as the reference compound. Except for

243

rufloxacin and prulifloxacin, the cross-reactivity values of other quinolones are lower

244

than 15% (Table 1). Only 9% of the quinolones in this study displayed that their

245

cross-reactivities were over 15%, thus, it could be regarded as high specificity of the

246

pipemidic acid antibody.

247

Generally, the high cross-reactivity of haptens could be directly ascribed to the

248

similarity of the molecular structure, and the low cross-reactivity are possibly related

249

to their structure difference of the cross-reactants30. In this study, norfloxacin had ever

250

been proved to be a good hapten candidate resulting to a broad-specificity antibody

251

that could recognize 13 quinolones31. The structure difference between norfloxacin

252

and pipemidic acid was at position 6 and 8, the atoms of norfloxacin were carbon at

253

position 6 and C-F group at position 8, however, the atoms of pipemidic acid were

254

nitrogen. Such difference seems to be a little, but their cross-reactivities were

255

significantly different (14.1% for norfloxacin, 100% for pipemidic acid). This

256

suggested that the position 6 and 8 were likely crucial to the desired hapten 10


Page 10 of 28

Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


257

conformation for antibody recognizing. Besides, the structures of rufloxacin and

258

ofloxacin are similar except for the ring formed by the group at position 1 and 8, this

259

implied that the ring also played an important role in the hapten-antibody binding.

260

Additionally, comparing with the structure of pipemidic acid, difloxacin had one

261

supplemented benzene ring at position 1, however, the cross-reactivity of difloxacin

262

dramatically dropped from CR 100% for pipemidic acid to CR 0.5%. Moreover, the

263

cross-reactivities of sarafloxacin (CR=0.3 %) and tosufloxacin (CR=0.3%) were also

264

very low due to added benzene ring at position 1. These dramatic changes in

265

cross-reactivity induced the importance of the groups at position 1, 6 and 8 for a

266

desired binding conformation. To better understand the high specificity of the

267

obtained pipemidic acid antibody, CoMFA was used for the further investigation.

268

CoMFA analysis

269

The CoMFA models exhibited the rational q2 values greater than 0.5 and the S

270

and E fields offered 58.2% and 41.8% contribution to the affinity (Table S2). The

271

pIC50 values of the molecules in the test set had been perfectly predicted in the

272

CoMFA model (Figure S3).

273

The contour maps reflected the desired/undesired steric/electrostatic features for

274

the molecular binding affinity in the CoMFA model, in which the green and yellow

275

contours represent the favorable and unfavorable steric regions, while the blue and red

276

contours represent the favorable and unfavorable electropositive regions, respectively.

277

The piperazine ring at position 7 was nearly in the same plane with the basic structure,

278

1,4-dihydro-4-oxo-3-quinolinecarboxylic acid shared by almost all quinolones (Figure

279

3a) and the conformation of pipemidic acid shaped like a letter “I”. Thus, from the

280

view of looking down, the piperazine ring formed a flat antibody binding cavity that

281

could not accommodate a large group.

282

And the green contour is near the position 4’ of pipemidic acid (Figure 3a), this

283

implied that the antibody could not recognize the group at both sides of position 7.

284

However, the piperazine ring at position 7 of norfloxacin was almost perpendicular to

285

the basic structure (Figure 3b). As a result, in order to accommodate a large group, it

286

seemed to be essential for hapten owning a large antibody binding cavity at this 11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

287

position. The piperazine ring at position 7 in other quinolones was also basically

288

perpendicular to the basic structure and was consistent with norfloxacin (Figure 3 c-f).

289

These could contribute to the main reason that the resultant antibody against

290

pipemidic acid exhibited low cross-reactivity to quinolones.

291

The hydrogen atom at the oxazine ring of S-(-)-ofloxacin interacted with the little

292

yellow contour and the methyl group were near the big yellow contour, however, the

293

hydrogen atom at the oxazine ring of R-(+)-ofloxacin interacted with the green

294

contours and the methyl group deviated from the little yellow contour due to their

295

opposite conformations (Figure 3c, d). This could be the reason that S-(-)-ofloxacin

296

(experimental pIC50=9.585) showed a lower binding activity to the pipemidic acid

297

antibody than that of R-(+)-ofloxacin (experimental pIC50= 10.000) (Table S3).

298

Compared the conformation of S-(-)-ofloxacin with that of norfloxacin, the group at

299

position 8 in S-(-)-ofloxacin was bigger and closer to yellow contours than that of

300

norfloxacin, this could be ascribed to the lower activity of S-(-)-ofloxacin

301

(experimental pIC50=9.585) (Figure 3b, d). In addition, comparison of the CoMFA

302

maps of pefloxacin and difloxacin (Figure 3e, f), the yellow contours near the benzene

303

ring of difloxacin indicated the bulky group at position 1 decreased the binding

304

activity, therefore, the activities of difloxacin (experimental pIC50=8.827) were lower

305

than that of pefloxacin (experimental pIC50=10.252) (Table S3).

306

Besides the major contribution of steric field, the electrostatic field also played

307

an important role in the quinolone-antibody binding conformation (Figure 3g, h). The

308

difference between the norfloxacin and sarafloxacin was the group at position 1, and

309

the group was interacted with blue contour of a rhombus. However, the

310

electronegative fluorine atom of sarafloxacin was near these regions, this additionally

311

led to lower binding activity of sarafloxacin. It indicated that electronegative groups

312

at position 1 decreased the binding activity, which was consistent with the conclusion

313

of steric effect above.

314

Conformation comparing analysis

315

Table 2 showed 22 kinds of models of the minimum energy conformations of the

316

quinolones from the immunoassay for pipemidic acid in this study and the previously 12


Page 12 of 28

Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


317

reported immunoassay for clinafloxacin21, they were divided into four groups in

318

accordance with their conformation features at minimum energy. The first group

319

contained pipemidic acid, nalidixic acid and oxolinic acid, their conformations shaped

320

like a letter “I”. The second group contained clinafloxacin, danofloxacin and

321

tosufloxacin, and their conformations shaped like a letter “P”. The third group

322

contained difloxacin, enrofloxacin, gatifloxacin, lomefloxacin, pefloxacin, rufloxacin,

323

R-(+)-ofloxacin, S-(-)-ofloxacin, garenoxacin, prulifloxacin, sparfloxacin and

324

marbofloxacin, and their conformations shaped like “ф”,. The fourth group contained

325

ciprofloxacin, norfloxacin, pazufloxacin and sarafloxacin, and their conformations

326

formed “Y” shape (or a lollipop).

327

In the first group, the conformation features of nalidixic acid and oxolinic acid

328

were also shaped like “I” and consistent with pipemidic acid. Thus, the antibody

329

derived from nalidixic acid or oxolinic acid as a hapten could be of high specificity.

330

In the second group, CoMFA was used for the further investigation of

331

clinafloxacin immunoassay. The CoMFA models exhibited the rational q2 values was

332

0.587 and the S and E fields offered 53.4% and 46.6% contribution to the affinity

333

(Table S2), respectively. The pIC50 values of the molecules in the test set had been

334

perfectly predicted (Figure S3). Therefore, the contour analysis of the model was

335

further carried out.

336

The group at position 7 of clinafloxacin was on the side of the basic structure

337

plane and the conformation of clinafloxacin shaped like a letter “P” (Figure 4a). And

338

the green contour near the amino group at position 3’ in clinafloxacin (Figure 4a), this

339

implied that the antibody to clinafloxacin could not recognize the group on the side

340

near the position 5’. However, the piperazine ring at position 7 in ciprofloxacin was

341

basically perpendicular to the basic structure (Figure 4b). And other quinolones with

342

piperazine ring at position 7 were also basically perpendicular to the basic structure

343

and were consistent with that of ciprofloxacin (Figure 4c-f). Moreover, the amino

344

group and 5-membered ring at position 7 seems to form a small antibody binding

345

cavity that would not accommodate a large group. For example, a piperazine ring is a

346

big group that could not allow to be accommodated in the small antibody binding 13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

347

cavity formed by the amino group and 5-membered ring of clinafloxacin. However,

348

the ciprofloxacin shaped like “Y” and the piperazine ring in ciprofloxacin could form

349

a bigger antibody binding cavity, and thus the bigger antibody binding cavity should

350

contribute to the broad specificity of the corresponding antibody to cipfloxacin, and

351

then the high specificity of clinafloxacin antibody would also be understandable. The

352

previous study of clinafloxacin also showed that all the CRs for structure-related

353

quinolones were lower than 15% (Table S5) 21.

354

In addition, when comparing the contour maps of gatifloxacin, lomefloxacin and

355

S-(-)-ofloxacin (Figure 4c, d, f), it could be found that the group at position 8 of

356

gatifloxacin was bigger and closer to the green contour over the structure than those

357

of S-(-)-ofloxacin. But the group at position 8 in S-(-)-ofloxacin was bigger and closer

358

than those of lomefloxacin, this could be ascribed to the highest activity of

359

gatifloxacin (experimental pIC50=6.076), followed by S-(-)-ofloxacin (experimental

360

pIC50=5.440), and lomefloxacin (experimental pIC50=4.828) (Table S4).

361

The green contour at the map bottom interacted with the methyl at the oxazine

362

ring of S-(-)-ofloxacin, however, the methyl at the oxazine ring of R-(+)-ofloxacin

363

were brought near the yellow contour in the upper portion of map due to their

364

opposite conformations (Figure 4e, f). It revealed the mechanism that S-(-)-ofloxacin

365

had higher binding affinity than R-(+)-ofloxacin. When the antibody distinguishing

366

the chiral isomer of quinolones, it showed that the bulky group near the green contour

367

at position 8 played a critical role. Comparison of the CoMFA maps of sarafloxacin

368

and ciprofloxacin , the yellow contours near the benzene ring of sarafloxacin

369

indicated the bulky group at position 1 decreased the binding activity (Figure 4b, g),

370

therefore, the activities of sarafloxacin (experimental pIC50=4.628) were lower than

371

that of ciprofloxacin (experimental pIC50=5.640) (Table S4). Since the yellow

372

contours near the benzene ring of sarafloxacin, it is understandable that the activity of

373

sarafloxacin (experimental pIC50=4.628) was lower, and the predicted pIC50 value of

374

sarafloxacin in the test set was lower than the experimental value (Table S4).

375

The electrostatic contour maps also reflected the molecular binding affinity in the

376

CoMFA model (Figure 4). The amino group at position 7 in clinafloxacin near the 14


Page 14 of 28

Page 15 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


377

blue contour was a strong electron donor group, while the groups of gatifloxacin and

378

lomefloxacin were weak electron donor groups. As a result, the binding activity of

379

gatifloxacin or lomefloxacin was much lower than that of clinafloxacin, this was

380

consistent with the steric effect above. The conformations of danofloxacin and

381

tosufloxacin in the second group also like “P” and similar with clinafloxacin, as a

382

result, their antibodies also would show high-specificity as clinafloxacin. This had

383

been proved by a previous study using anti-danofloxacin antibody (Table S5)32. The

384

conformation of tosufloxacin demonstrated high similarity with that of clinafloxacin

385

and could be a suitable hapten for the production of highly specific antibody.

386

The third group consisted of 12 kinds of quinolones in this study, they all shaped

387

like “ф”, and it could be predicted that their antibodies would be predicted to

388

accordingly be high-specific, this have also been confirmed by experiments (Table 2

389

and Table S5)10,11,27,33-37. For example, ofloxacin contains a substituent on the

390

piperazine ring at position 7, and the previous QSAR study showed that the green

391

contour was near the substituent, while two yellow contours distributed in both sides

392

of piperazine ring27. This implied that the antibody could not recognize quinolones

393

without a substituent on the piperazine ring or with a substituent on other site of

394

piperazine ring. And the experimental cross-reactivity had already proved that only

395

rufloxacin, garenoxacin and marbofloxacin had high cross reactivity values

396

(cross-reactivity>15%)27. Therefore, it could be concluded that hapten quinolone

397

shaped like “I” or “P” or “ф” would result in an antibody with low cross-reactivity

398

and high specificity.

399

In the fourth group (Table 2), different from the structures shaped like “I”, “P”

400

and “ф”, all the proposed compounds shaped like “Y” or a lollipop, their

401

conformations could be predicted to produce a large antibody binding cavity to

402

accommodate the piperazinyl group at position 7 (Table 2). For example,

403

pazufloxacin contained a 1-aminocyclopropyl group at position 7 but not a

404

piperazinyl group shared by many quinolones, both the amino and cyclopropyl groups

405

around position 7 of pazufloxacin formed “Y” shape or a lollipop, and the green

406

contour also lay on both sides of the position 726, it implied that the resultant antibody 15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

407

to pazufloxacin would broadly recognize other quinolones. In fact, the experimentally

408

raised pazufloxacin antibody with high sensitivity and broad specificity could

409

recognize 24 quinolones as expected (Table S5)26, this is a broad specificty, but not a

410

high specificity, because pazufloxacin shaped unlike any one of three conformations

411

“I” or “P” or “ф”.

412

Ciprofloxacin is another good example to prove the three essential conformations

413

for a high specificity18. The piperazine ring at position 7 in ciprofloxacin was almost

414

perpendicular to the basic structure (Figure 4b and Table 2), the conformation shaped

415

liked a lollipop, and thus the ciprofloxacin antibody should be able to accommodate a

416

large group and demonstrated a broad specificity according to the QSAR analysis

417

above in this study. Actually this has also been verified again by a previously reported

418

monoclonal antibody and a developed broad-specificity immunoassay for 12

419

quinolones (Table S5)18. The conformations of norfloxacin and sarafloxacin in the

420

fourth group also like “Y” and were similar with ciprofloxacin, as a result, their

421

antibodies also would show broad specificity as ciprofloxacin. The structure-related

422

quinolones of norfloxacin and sarafloxacin whose cross-reactivity values were over

423

15%, were 13 kinds and 10 kinds, respectively (Table S5)9,31.

424

Therefore, it could be concluded that the shape like “I”, “P” and “ф” was the

425

essential conformation feature of quinolone hapten used to produce antibody with

426

high specificity, while the conformation of quinolone hapten like “Y” shape (or a

427

lollipop) could be used to produce antibody with broad specificity.

428 429

CONCLUSIONS

430

In summary, 3D QSAR studies were performed to investigate the molecular

431

recognition between the antibody and quinolone drugs, the CoMFA model revealed

432

that the quinolones shaped like “I”, “P” and “ф” formed a flat conformation, which

433

could not accommodate a large group, and thus the resultant antibody would be high

434

specificity. But the hapten conformation shaped like “Y” will lead to a broad

435

specificity antibody with the high cross-reactivity. Almost all the antibodies against

436

quinolone could result from these four hapten conformations. This finding is the first 16


Page 16 of 28

Page 17 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


437

time to systemically and effectively elucidate and predict the high specificity of

438

quinolone antibodies, it will be of significance of the accurate hapten design,

439

predictable antibody specificity and better understanding of the recognition

440

mechanism of hapten and the antibody specificity.

441 442 443

ASSOCIATED CONTENT Supporting Information. Description of the optimized ciELISA conditions, the

444

summary of the calculated parameters of the CoMFA model, the predicted values

445

derived from the CoMFA models of pipemidic acid, the predicted values derived from

446

the CoMFA models of clinafloxacin, a review of immunoassay for the determination

447

of quinolones in recent years, UV spectra of BSA, OVA, pipemidic acid, pipemidic

448

acid-BSA and pipemidic acid-OVA, SDS-PAGE pattern using purified pipemidic acid

449

polyclonal antibody, the scatter plots of predicted versus experimental pIC50.

450 451 452

ACKNOWLEDGMENTS This work was supported by Natural Science Foundation of China (U1301214,

453

30700663, 21475047), Guangdong and Guangzhou Planned Program in Science and

454

Technology (S2013030013338, 2016201604030004, 2014TX01N250,

455

2013B051000072, and 2014A030306026). Guangdong Natural Science Foundation

456

(2015A030313366), the support of 2016KYDT04 and 2016KYDT06 by Guangzhou

457

University of Chinese Medicine.

458 459

REFERENCES

460

(1) Van Bambeke, F.; Michot, J. M.; Van Eldere, J.; Tulkens, P. M. Clin. Microbiol.

461

Infec. 2005, 11, 256-280.

462

(2) Andriole, V. T. Clin. Infect. Dis. 2005, 41, S113-S119.

463

(3) Mehlhorn, A. J.; Brown, D. A. Ann. Pharmaco. 2007, 41, 1859-1866.

464

(4) Hernández-Arteseros, J.; Barbosa, J.; Compano, R.; Prat, M. J. Chromatog. A

465

2002, 945, 1-24.

17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

466

(5) Espinosa-Mansilla, A.; Jiménez Girón, A.; Muñoz de la Peña, A. Czech J. Food

467

Sci. 2012, 30, 74-82.

468

(6) Pecorelli, I.; Galarini, R.; Bibi, R.; Floridi, A.; Casciarri, E.; Floridi, A. Anal. Chim.

469

Acta 2003, 483, 81-89.

470

(7) Chang, C. S.; Wang, W. H.; Tsai, C. E. J. Food Drug Anal. 2010, 18, 87-97.

471

(8) Dorival‐García, N.; Zafra‐Gómez, A.; Cantarero, S.; Navalón, A.; Vílchez, J.

472

Microchem. J. 2013, 106, 323-333.

473

(9) Huet, A. C.; Charlier, C.; Tittlemier, S. A.; Singh, G.; Benrejeb, S.; Delahaut, P. J

474

Agr. Food Chem. 2006, 54, 2822-2827.

475

(10) Lu, S. X.; Zhang, Y. L,; Liu, J. T.; Zhao, C. B.; Liu, W.; Xi, R. M. J Agr. Food

476

Chem. 2006, 54, 6995-7000.

477

(11) Sheng, W.; Xia, X. F.; Wei, K. Y.; Li, J.; Li, Q. X.; Xu, T. J Agr. Food Chem.

478

2009, 57, 5971-5975.

479

(12) Wang, Z. H.; Kai, Z. P.; Beier, R. C.; Shen, J. Z.; Yang, X. L. Int. J. Mol. Sci.

480

2012, 13, 6334-6351.

481

(13) Cui, J. L.; Zhang, K.; Huang, Q. X.; Yu, Y. Y.; Peng, X. Z. Anal. Chim. Acta 2011,

482

688, 84-89.

483

(14) Zhang, H. Y.; Wang, S.; Fang, G. Z. J. Immunol. Methods 2011, 368, 1-23.

484

(15) Li, Y. F.; Sun, Y. M.; Beier, R. C.; Lei, H. T.; Gee, S.; Hammock, B. D.; Wang, H.;

485

Wang, Z.; Sun, X.; Shen, Y. D.; Yang, J. Y.; Xu, Z. L. TrAC. Trends Anal. Chem. 2017,

486

88, 25-40.

487

(16) Leivo, J.; Lamminmäki, U.; Lövgren, T.; Vehniäinen, M. J Agr. Food Chem.

488

2013, 61, 11981-11985.

489

(17) Cao, L. M.; Kong, D. X.; Sui, J. X.; Jiang, T.; Li, Z. Y.; Ma, L.; Lin, H. Anal.

490

Chem. 2009, 81, 3246-3251.

491

(18) Wang, Z. H.; Zhu, Y.; Ding, S. Y.; He, F. Y.; Beier, R. C.; Li, J. C.; Jiang, H. Y.;

492

Feng, C. W.; Wan, Y. P.; Zhang, S. X. Anal. Chem. 2007, 79, 4471-4483.

493

(19) Xu, Z. L.; Shen, Y. D.; Zheng, W. X.; Beier, R. C.; Xie, G. M.; Dong, J. X.; Yang,

494

J. Y.; Wang, H.; Lei, H. T.; She, Z. G. Anal. Chem. 2010, 82, 9314-9321.

495

(20) Luo, L.; Xu, Z. L.; Yang, J. Y.; Xiao, Z. L.; Li, Y. J.; Beier, R. C.; Sun, Y. M.; Lei, 18


Page 18 of 28

Page 19 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


496

H. T.; Wang, H.; Shen, Y. D. J Agr. Food Chem. 2014, 62, 12299-12308.

497

(21) Chen, J. H.; Lv, S. W.; Wang, Q.; Xu, Z. L.; Yang, J. Y.; Shen, Y. D.; Wang, H.;

498

Sun, Y. M.; Lei, H. T. Food Anal. Methods 2015, 8, 1468-1476.

499

(22) Page, M.; Thorpe, R. In The Protein Protocols Handbook; Humana Press: 2002,

500

p 983-984.

501

(23) Oubiña, A.; Barceló, D.; Marco, M. P. Anal. Chim. Acta 1999, 387, 267-279.

502

(24)Yuan, Y. L.; Hua, X. D.; Li, M.; Yin, W.; Shi, H. Y.; Wang, M. H. RSC Adv. 2014,

503

4, 24406-24411.

504

(25)Yu, F. Y.; Vdovenko, M. M.; Wang, J. J.; Sakharov, I. Y. J Agr. Food Chem. 2011,

505

59, 809-813.

506

(26) Chen, J. H.; Lu, N.; Shen, X.; Tang, Q. S.; Zhang, C. J.; Xu, J.; Sun, Y. M.;

507

Huang, X. A..; Xu, Z. L.; Lei, H. T. J Agr. Food Chem. 2016, 64, 2772-2779.

508

(27) Mu, H. T.; Lei, H. T.; Wang, B. L.; Xu, Z. L.; Zhang, C. J.; Ling, L.; Tian, Y. X.;

509

Hu, J. S.; Sun, Y. M. J Agr. Food Chem. 2014, 62, 7804-7812.

510

(28) Zhang, H. T.; Jiang, J. Q.; Wang, Z. L.; Chang, X. Y.; Liu, X. Y.; Wang, S. H.;

511

Zhao, K.; Chen, J. S. J. Zhejiang Univ. Sci. B 2011, 12, 884-891.

512

(29) Han, D.; Yu, M.; Knopp, D.; Niessner, R.; Wu, M.; Deng, A. P. J Agr. Food Chem.

513

2007, 55, 6424-6430.

514

(30) Lei, H. T.; Su, R.; Haughey, S. A.; Wang, Q.; Xu, Z. L.; Yang, J. Y.; Shen, Y. D.;

515

Wang, H.; Jiang, Y. M.; Sun, Y. M. Molecules 2011, 16, 5591-5603.

516

(31) Jiang, W. X.; Wang, Z. H.; Beier, R. C.; Jiang, H. Y.; Wu, Y. N.; Shen, J. Z. Anal.

517

Chem. 2013, 85, 1995-1999.

518

(32) Liu, Z. Q.; Lu, S. X.; Zhao, C. H.; Ding, K.; Cao, Z. Z.; Zhan, J. H.; Ma, C.; Liu,

519

J. T.; Xi, R. M. J. Sci. Food Agric. 2009, 89, 1115-1121.

520

(33) Li, C.; Jiang, J.; Qi, X. H. In 2011 Eighth International Conference on Fuzzy

521

Systems and Knowledge Discovery (FSKD) 2011; Vol. 3, p 1446-1449.

522

(34) Zeng, H. J.; Yang, R.; Liu, B.; Lei, L. F.; Li, J. J.; Qu, L. B. J. Pharmaceut. Anal.

523

2012, 2, 214-219.

524

(35) Zhao, C. B.; Liu, W.; Ling, H. L.; Lu, S. X.; Zhang, Y. L.; Liu, J. T.; Xi, R. M. J

525

Agr. Food Chem. 2007, 55, 6879-6884. 19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

526

(36) Zhao, Y. L.; Zhang, G. P.; Liu, Q. T.; Teng, M.; Yang, J. F.; Wang, J. H. J Agr.

527

Food Chem. 2008, 56, 12138-12142.

528

(37) Zhi, A. M.; Li, B. B.; Liu, Q. T.; Hu, X. F.; Zhao, D.; Hou, Y. Z.; Deng, R. G.;

529

Chai, S. J.; Zhang, G. P. Food Agr. Immunol. 2010, 21, 335-345.

530

20


Page 20 of 28

Page 21 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


531

Figure Captions

532

Figure 1 The structures of quinolones

533

Figure 2 ciELISA calibration curves for pipemidic acid with different coating

534

antigens. Each point represented the mean ± standard deviation of three replicates.

535

Figure 3 CoMFA contour maps of pipemidic acid. (a-f) CoMFA streic contour maps

536

together with (a) embedded pipemidic acid, (b) embedded norfloxacin, (c) embedded

537

R-(+)-ofloxacin, (d) embedded S-(-)-ofloxacin, (e) embedded pefloxacin and (f)

538

embedded difloxacin. (g-h) CoMFA electrostatic contour maps together with (g)

539

embedded norfloxacin and (h) embedded sarafloxacin. The energies of all fields were

540

calculated with the weight of the standard deviation and the coefficient. Green, yellow,

541

blue, and red contours represent steric bulk desirable, steric bulk undesirable, positive

542

charge desirable and negative charge desirable, respectively.

543

Figure 4 CoMFA contour maps of clinafloxacin. (a-f) CoMFA streic contour maps

544

together with (a) embedded clinafloxacin, (b) embedded ciprofloxacin, (c) embedded

545

gatifloxacin, (d) embedded lomefloxacin, (e) embedded R-(+)-ofloxacin, (f)

546

embedded S-(-)-ofloxacin and (g) embedded sarafloxacin. CoMFA electrostatic

547

contour maps together with (h) embedded clinafloxacin, (i) embedded gatifloxacin, (j)

548

embedded lomefloxacin. The energies of all fields were calculated with the weight of

549

the standard deviation and the coefficient. Green, yellow, blue, and red contours

550

represent steric bulk desirable, steric bulk undesirable, positive charge desirable and

551

negative charge desirable, respectively.

552

21



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 28

553

Table 1 Cross-reactivity of pipemidic acid and 22 quinolone drugs based on coating

554

antigen lomefloxacin –OVA. IC50

CR

IC50

CR

Name

Name (nmol/mL)

(%)

(nmol/mL) (%)

Pipemidic acid

0.0076

100.0

Danofloxacin

0.21

3.6

Rufloxacin

0.026

29.5

Garenoxacin

0.23

3.3

Prulifloxacin

0.026

28.8

S-(-)-ofloxacin

0.26

2.9

Norfloxacin

0.054

14.1

Pazufloxacin

0.27

2.8

Pefloxacin

0.056

13.4

Clinafloxacin

0.49

1.5

Enrofloxacin

0.071

10.7

Gatifloxacin

0.50

1.5

Oxolinic acid

0.082

9.2

Marbofloxacin

0.96

0.8

Ofloxacin

0.090

8.4

Difloxacin

1.49

0.5

R-(+)-ofloxacin

0.10

7.3

Sarafloxacin

2.32

0.3

Ciprofloxacin

0.14

5.4

Sparfloxacin

2.53

0.3

Lomefloxacin

0.21

3.7

Tosufloxacin

2.91

0.3

22


Page 23 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

555 556 557


Table 2 Models of the minimum energy conformations of the quinolones from the immunoassay for pipemidic acid and for clinafloxacin. The elements are represented in the following manner: oxygen, red; nitrogen, navy blue; hydrogen, baby blue; fluorine, green; carbon, white.; sulphur, yellow. Shape Characteristic Models of the minimum energy conformations of the quinolones a I High-specificity Nalidixic acid P

ф

Y

Oxolinic acid

Pipemidic acid

High-specificity Clinafloxacin

Danofloxacin

Tosufloxacin

Difloxacin

Enrofloxacin

Gatifloxacin

Lomefloxacin

Pefloxacin

Rufloxacin

R-(+)-ofloxacin

S-(-)-ofloxacin

Garenoxacin

Prulifloxacin

Sparfloxacin

Marbofloxacin

Ciprofloxacin

Norfloxacin

Pazufloxacin

Sarafloxacin

High-specificity

b

Broad-specificity

558

a

High-specificity means the structurally related quinolones whose cross-reactivity values over 15% were less than 3 kinds.

559

b

Broad-specificity means most of the structurally related quinolones whose cross-reactivity values over 15%.

23



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

560

TOC

561

24


Page 24 of 28

Page 25 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1 The structures of quinolones 209x120mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2 ciELISA calibration curves for pipemidic acid with different coating antigens. Each point represented the mean ± standard deviation of three replicates. 296x209mm (300 x 300 DPI)


Page 26 of 28

Page 27 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3 CoMFA contour maps of pipemidic acid. (a-f) CoMFA streic contour maps together with (a) embedded pipemidic acid, (b) embedded norfloxacin, (c) embedded R-(+)-ofloxacin, (d) embedded S-(-)ofloxacin, (e) embedded pefloxacin and (f) embedded difloxacin. (g-h) CoMFA electrostatic contour maps together with (g) embedded norfloxacin and (h) embedded sarafloxacin. The energies of all fields were calculated with the weight of the standard deviation and the coefficient. Green, yellow, blue, and red contours represent steric bulk desirable, steric bulk undesirable, positive charge desirable and negative charge desirable, respectively. 147x147mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4 CoMFA contour maps of clinafloxacin. (a-f) CoMFA streic contour maps together with (a) embedded clinafloxacin, (b) embedded ciprofloxacin, (c) embedded gatifloxacin, (d) embedded lomefloxacin, (e) embedded R-(+)-ofloxacin, (f) embedded S-(-)-ofloxacin and (g) embedded sarafloxacin. CoMFA electrostatic contour maps together with (h) embedded clinafloxacin, (i) embedded gatifloxacin, (j) embedded lomefloxacin. The energies of all fields were calculated with the weight of the standard deviation and the coefficient. Green, yellow, blue, and red contours represent steric bulk desirable, steric bulk undesirable, positive charge desirable and negative charge desirable, respectively. 198x286mm (300 x 300 DPI)


Page 28 of 28

Four Specific Hapten Conformations Dominating Antibody Specificity

Recommend Documents