Sensitive Immunoassay for Detection and Quantification of the

Apr 11, 2017 - (9, 10, 14, 15) Its performance was evaluated in spike-and-recovery studies performed with buffer and mouse and human serum. The result...
1 downloads 6 Views 1MB Size
Subscriber access provided by HACETTEPE UNIVERSITESI KUTUPHANESI

Article

Sensitive immunoassay for detection and quantification of the neurotoxin, tetramethylenedisulfotetramine (TETS). Natalia Vasylieva, Bogdan Barnych, Amelia Ann Rand, Bora Inceoglu, Shirley J Gee, and Bruce D. Hammock Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b00846 • Publication Date (Web): 11 Apr 2017 Downloaded from http://pubs.acs.org on April 13, 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 31

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

Sensitive immunoassay for detection and

2

quantification of the neurotoxin,

3

tetramethylenedisulfotetramine (TETS).

4

Natalia Vasylieva1, Bogdan Barnych1, Amy Rand1, Bora Inceoglu1, Shirley J. Gee1, Bruce D.

5

Hammock1.

6

7

1

8

University of California Davis, Davis, California 95616. United States

Department of Entomology and Nematology and UCD Comprehensive Cancer Center

9 10 11 12 13 14

KEYWORDS: rodenticide, threat agent, antibody, mice, CounterAct, countermeasure.

15 16

Corresponding Author:

17

* Tel.: 530-752-7519. Fax: 530-752-1537. Email: [email protected]

1 ACS Paragon Plus Environment

Analytical Chemistry

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 2 of 31

18

Abstract

19

Tetramethylenedisulfotetramine (TETS, tetramine) is a formerly used and highly neurotoxic

20

rodenticide. Its lethality, recent history of intentional use for mass poisoning and the absence

21

of a known antidote raises public health concerns. Therefore, rapid, high throughput and

22

sensitive methods for detection and quantification of TETS are critical. Instrumental analysis

23

method such as GC/MS is sensitive but not rapid or high-throughput. Therefore, an

24

immunoassay selective to TETS was developed. The assay shows an IC50 of 4.5±1.2

25

ng/mL, with a limit of detection of 0.2 ng/mL, comparable to GC/MS. Performance of the

26

immunoassay was demonstrated by a recovery study using known concentrations of TETS

27

spiked into buffer, human and mouse serum matrices giving recoveries in the range of 80-

28

120%. The assay demonstrated good correlation in TETS recovery with established GC/MS

29

analysis. The immunoassay was then used to quantify TETS concentration in the serum of

30

mice exposed to 2x LD50 dose of TETS and to monitor kinetics of TETS clearance from

31

blood over a short period of time. TETS concentration in the serum reached 150 ng/mL

32

without significant change over 4h post treatment. Results obtained with the immunoassay

33

had good correlation with GC/MS analysis. Overall this immunoassay is an important tool to

34

rapidly detect and quantify levels of TETS from biological samples with high sensitivity. The

35

assay can be adapted to multiple formats including field or hospital use.

2 ACS Paragon Plus Environment

Page 3 of 31

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

36 37

Analytical Chemistry

Introduction Tetramethylenedisulfotetramine (TETS, tetramine) is an organic compound, initially

38

developed as a rodenticide in the 1950s but was subsequently banned worldwide due to

39

high neurotoxicity in all mammals. The compound is a noncompetitive blocker of chloride

40

channels on the γ-aminobutyric acid (GABA) receptor of the neuronal cell membrane,

41

leading to lack of inhibitory currents and excessive excitation in the central nervous system

42

(CNS). Symptoms of poisoning with TETS are dose dependent, including seizures, status

43

epilepticus and coma at high levels of exposure. The LD50 in mice is 0.1 mg/kg, and 5-10 mg

44

is considered to be a lethal dose for humans1. Tetramine is simple to synthesize and handle.

45

It is prepared from readily available starting materials. As predicted from its structure it

46

neither absorbs ultraviolet light nor is it fluorescent and it ionizes poorly on LC-MS. It is a

47

white odorless and tasteless powder with significant solubility in water. In addition, it is also

48

extremely stable in the environment. 1 Despite being banned globally in 1984 it is still

49

illegally available on the rural black market in China. As a result, TETS has been implicated

50

in numerous cases of mass poisoning.2,3 In the period of 2000-2012, at least 148 cases

51

were reported in China; over 3526 people were reported to be exposed to TETS, resulting in

52

225 deaths.2 Although, most of the known cases of poisoning with TETS occurred in China,

53

one infant was poisoned in New York City, USA.4 Six months after poisoning the infant

54

remained severely developmentally delayed. Poisoning with TETS has long lasting effects.

55

Even after patients are treated for seizures, there is evidence that numerous neurological

56

lesions occur.5,6 Most victims who have no seizures initially after exposure, develop them

57

later on. On average recurrent seizures stop after 2.3 years with anti-seizure treatment,

58

although cases that continued for 9 years have been reported.5 Thus, poisoning with TETS

3 ACS Paragon Plus Environment

Analytical Chemistry

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

59

leads to seizure-induced brain damage even in patients that survive acute intoxication.

60

Poisoning with TETS may be fatal within an hour. There are no generally recognized

61

antidotes for it and victims only receive supportive therapy and symptomatic treatment,

62

including gastric lavage, dialysis, blood perfusion, and administration of drugs controlling

63

seizures and acting on GABA receptors.7,8 The high toxicity, unique physical properties,

64

absence of antidote, and history of use as an agent of intentional mass poisoning make

65

TETS a serious public health concern because it may pose a serious threat to the

66

population.

67

Page 4 of 31

Currently, gas chromatography (GC) with different detection systems (MS9-13,

68

MS/MS9,11, nitrogen/phosphorus detection14, flame thermionic detection15), is the main

69

analytical technique used for TETS detection and quantification. This is due in part to poor

70

ionization on LC-MS. Quantification protocols with good sensitivity were developed for

71

tetramine detection in blood (8 - 500 ng/mL and LOD 1 ng/mL),14 urine (LOD around 3

72

ng/mL),9,10,15 food (down to 0.2 ng/g)12 and beverages (0.5 - 100 ng/mL)11,13 Based on

73

clinical reports, depending on the severity of poisoning, the concentration of tetramine in the

74

blood of victims varies widely from low ng/mL up to µg/mL.7,16,17 In urine, the concentration

75

of TETS is typically similar or slightly higher than in blood.10,17 The low detection limits,

76

achieved with GC methods require laborious sample preparation in order to reduce matrix

77

interference with analysis. In addition, instrumental methods are expensive, slow, require

78

advanced infrastructure and are inappropriate as a high throughput technique, as well as

79

requiring specific expertise and a well-trained operator. On the other hand, immunoassays

80

are widely used for detection of small molecules in environmental18-20 and biological18,20,21

81

samples, as well as in food18,22,23 and beverages.24,25 They usually have high selectivity

4 ACS Paragon Plus Environment

Page 5 of 31

1 2 3 82 4 5 6 83 7 8 84 9 10 85 11 12 13 86 14 15 16 87 17 18 88 19 20 21 89 22 23 90 24 25 91 26 27 28 92 29 30 93 31 32 94 33 34 35 95 36 37 38 96 39 40 97 41 42 43 98 44 45 99 46 47 100 48 49 50 101 51 52 102 53 54 55 103 56 57 58 59 60

Analytical Chemistry

toward their corresponding analytes with sensitivity greater or comparable to instrumental methods. Furthermore, they provide the advantage of minimal or no sample preparation, thus dramatically decreasing the labor involved in analysis. These attributes allow high throughput screening and quantification. An additional benefit of immunoassay is its potential to be packaged in a field portable platform for on-site detection.26-28 The tetramine structure appears to be an ideal candidate for an immunoassay. It is sterically constrained, lacking freedom of the rotation about its bonds and has several heteroatoms to provide recognition points for the antibody. Surprisingly, to the best of our knowledge, there is no antibody selective to TETS reported in the literature. We suppose this deficiency may be due to the challenging properties of TETS that make it inert and difficult to modify as well as potentially dangerous in order to produce appropriate haptens for immunization. We developed a simple and general chemical approach for the synthesis of TETS-like compounds and created a library of TETS analogs several of which were suitable haptens for immunoassay development.29 In this work we used the haptens and analogs of TETS to develop the first quantitative immunoassay selective to tetramine. The sensitivity of the developed immunoassay was compared to a GC/MS method, performed in conditions similar to methods published in the literature. 9,10,14,15 Its performance was evaluated in spike-andrecovery studies performed with buffer, mouse and human serum. The resulting data were compared to the data obtained with GC/MS. The developed assay was used to monitor kinetics of TETS in the blood of mice treated with tetramine. MATERIAL AND METHODS

5 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 104 4 5 6 105 7 8 106 9 10 11 107 12 13 14 108 15 16 109 17 18 19 110 20 21 111 22 23 112 24 25 26 113 27 28 114 29 30 115 31 32 33 116 34 35 36 117 37 38 118 39 40 41 119 42 43 120 44 45 46 121 47 48 122 49 50 51 123 52 53 54 124 55 56 125 57 58 59 60

Page 6 of 31

Information concerning chemicals and instruments, immunization and antiserum preparation, extraction procedure and the GC/MS method is detailed in the Supporting Information (SI). Preparation of immunogens and coating antigens. Synthesis of TETS analogs and corresponding haptens was reported by Barnych et al.29. For immunoassay development, haptens with a reactive carboxylic acid group were conjugated to proteins by a Nhydroxysuccinimide (NHS) (Haptens 2d, 2j, 2c in Table 1) method. Haptens having an aromatic amine group (–NH2) were conjugated to the protein by the diazotization method for coating antigens and through glutaraldehyde chemistry for use as immunogens. Haptens 6a and 6b were conjugated to the protein via a succinate linker or via a click chemistry approach.30 All haptens (Table 1, Table S1) were conjugated to thyroglobulin (Thy) for immunogen preparation and to bovine serum albumin (BSA) and conalbumin (CON) for coating antigen screening. The conjugation protocols are detailed in the SI. Reagent optimization. Each serum was screened against 1 homologous and 5 heterologous haptens using a three point competitive ELISA at 0, 50 and 5000 µg/L TETS concentration. Selected coating antigens were used to obtain 8-point full competition curves to determine and compare assay sensitivities (Table S2). In further assay development, the dilution of serum from rabbit #3442 was optimized in checkerboard titration with coating antigen 2c-CON. Optimization of the assay conditions. The sensitivity of the ELISA was optimized through the following parameters. (a) Protein in blocking buffer. The effect of different proteins on assay sensitivity was assessed on plates blocked with 1% solution of BSA, skim

6 ACS Paragon Plus Environment

Page 7 of 31

1 2 3 126 4 5 6 127 7 8 128 9 10 129 11 12 13 130 14 15 131 16 17 132 18 19 20 21 133 22 23 134 24 25 26 135 27 28 136 29 30 137 31 32 33 138 34 35 139 36 37 140 38 39 40 141 41 42 142 43 44 45 143 46 47 144 48 49 145 50 51 52 146 53 54 55 147 56 57 148 58 59 60

Analytical Chemistry

milk or ovalbumin in PBST. (b) Protein in antibody buffer. The effect of protein presence in the buffer was evaluated with BSA tested at 0, 0.5 and 1% concentration prior to addition to analyte on the plate. (c) Co-solvent in analyte buffer. The effect of organic solvent on assay sensitivity was studied by analyzing standard solutions of TETS prepared in PBS containing 10, 20 or 40% of methanol and DMSO (prior to addition of antibody in the well). (d) Ionic strength. The effect of ionic strength on assay sensitivity was evaluated from the assays performed in 5 mM, 10 mM, 20 mM and 30 mM PBS (pH 7.5). Indirect competitive ELISA. Plates were coated with 0.5 µg/mL of antigen (Hapten 2c/CON) in coating buffer (100 µL/well). After incubation for 1 h at room temperature (RT), the solution was replaced with blocking buffer (200 µL/well) and plates were incubated over night at 4°C or for 1-4 h at RT. Plates were washed with washing buffer (phosphate buffered saline (PBS) containing 0.05% Tween 20, PBST) 3 times prior to sample loading. TETS solutions (calibration curve or sample) in assay buffer were loaded on the coated plate (in duplicates or triplicates) at 50 µL/well. An equal volume of anti-TETS antiserum diluted in PBS containing 1% BSA was added. The plate was incubated for 1 h at RT and then washed 5 times with wash buffer. Goat anti-rabbit IgG-HRP conjugate was added at 100 µL/well in a 1:10000 dilution as instructed by the manufacturer. The plate was incubated for 1 h at RT and washed 5 times. Substrate solution was added (100 µL/well) and left to develop color for about 10 min. The reaction was stopped by addition of 2 M H2SO4 (50 µL/well) and absorbance was read at 450 nm. SigmaPlot 11.0 software was used for curve fitting and data analysis. Matrix effect. The effect of matrix on the assay performance was evaluated using mouse serum, human serum and human urine spiked with TETS. Spiked matrix was allowed to 7 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 149 4 5 6 150 7 8 151 9 10 152 11 12 13 14 153 15 16 154 17 18 19 155 20 21 156 22 23 157 24 25 26 158 27 28 29 159 30 31 32 160 33 34 161 35 36 37 162 38 39 163 40 41 42 164 43 44 165 45 46 166 47 48 49 167 50 51 52 168 53 54 169 55 56 170 57 58 59 60

Page 8 of 31

stand for at least 1 h. Competition curves were obtained by a serial dilution of the analyte in the assay buffer that is PBS containing either 20% or 40% methanol. The assay buffer also contained 0%, 20%, 40% or 100% of a neat matrix (matrix content is indicated as a concentration prior to addition of the antibody in the well). Immunoassay validation. Validation studies were performed by 1) evaluating the recovery from buffer, mouse and human serum fortified with varying TETS concentrations measured by immunoassay and GC/MS using both neat and extracted samples; 2) applying immunoassay to analyze serum samples obtained from mice treated with TETS. Sample preparation for validation studies was done in a blind fashion by another operator and analysis of samples from exposure studies are detailed in the SI. RESULTS AND DISCUSSION 1. Hapten design and immune response in rabbits. The design and synthesis of haptens is a critical step in immunoassay development. The literature data show that the most sensitive competitive immunoassays are those developed when the coating antigen is different from the immunizing antigen (heterologous assay).31-34 Not only the carrier protein should be different but also at least the handle and preferably the chemical structure of the hapten should be modified. The smaller the hapten the more important hapten design becomes. Therefore, the affinity of the antibody should be strong enough to bind to the coating antigen but remain lower than the affinity towards the target analyte. The first series of haptens (Table 1, haptens 6a and 6b) preserved most of the TETS structure except one nitrogen and one –SO2 group, which were replaced by carbon atoms. On the side of the molecule missing those functional determinants, an amine (-NH2) or a 8 ACS Paragon Plus Environment

Page 9 of 31

1 2 3 171 4 5 6 172 7 8 173 9 10 174 11 12 13 175 14 15 176 16 17 177 18 19 20 178 21 22 179 23 24 25 180 26 27 181 28 29 182 30 31 32 183 33 34 184 35 36 185 37 38 39 186 40 41 187 42 43 44 188 45 46 189 47 48 190 49 50 51 191 52 53 54 192 55 56 193 57 58 59 60

Analytical Chemistry

hydroxy (-OH) functional group was added. Each hapten was then conjugated via different chemistries to create heterology in a linker arm. The first approach included reaction of haptens with succinic anhydride, followed by conjugation of the resulting hemisuccinates to the protein via standard EDC/NHS chemistry. The second approach took advantage of the copper-catalyzed azide-alkyne cycloaddition (CuAAC) reaction, probably the most widely known example of click chemistry (Table S1). Click chemistry has not been widely used in immunoassay coupling, but this study illustrates its applicability. Haptens 6a and 6b were first transformed into an amide and an ester of the azidoacetic acid respectively which then were reacted with BSA-pentynoic acid conjugates under reported optimized conditions.30 Haptens conjugated to Thy (Table S1) were used for immunization and the titer was assessed in a homologous assay with BSA as the carrier protein. Immunogens 6bS-Thy and 6aS-Thy (Table S1) raised a strong response in rabbits giving very high titers even when tested at 0.01 µg/mL of coating antigen and 1:25 000 dilution of the serum after the first immunization. This high titer remained unchanged following boosting injections. We hypothesize that the polyclonal serum was highly reactive toward the triazole moiety present on the linker arm. In turn, immunogens 6bCC-Thy and 6aCC-Thy resulted in very low titers that slowly increased over the immunization period of 3 months. However, none of the obtained sera had sufficient affinity to TETS. It is possible that elimination of the nitrogen and -SO2 functionalities resulted in haptens differing from TETS with enough significance to decrease antibody affinity to TETS relating to the coating antigen, making the assay insensitive to the analyte. In order to better mimic the TETS structure we developed a second series of haptens using a novel approach to the synthesis of TETS analogs recently reported by our group.29

9 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 194 4 5 6 195 7 8 196 9 10 197 11 12 13 198 14 15 199 16 17 200 18 19 20 21 201 22 23 202 24 25 203 26 27 28 204 29 30 205 31 32 206 33 34 35 207 36 37 208 38 39 209 40 41 42 210 43 44 211 45 46 47 212 48 49 213 50 51 214 52 53 54 55 215 56 57 216 58 59 60

Page 10 of 31

This group of haptens (Table 1) has a structure that preserves all the heteroatoms of TETS and differs only in one of the methylene bridges that was replaced by an ethylene bridge carrying a side chain with functional group. Heterology is achieved with different linker arms including a short arm (hapten 2c) or a flexible long arm (hapten 2d), rigid arm (hapten 2j) or a long arm with a bulky aromatic ring (hapten 2k). All haptens were conjugated to Thy for immunization, while BSA and CON were used for protein heterology in coating antigens. Immunizations were performed with haptens 2j, 2k and 2d. 2. Identification of sensitive pairs of coating hapten and antibody. Sera from 6 rabbits were screened in a four-point (5000, 500, 50, 0 ng/mL TETS) competitive format against 5 coating antigens. Figure 1 shows the percentage of inhibition of antibody binding to the plate in the presence of TETS at 50 ng/mL (one point concentration shown). A colored bar shows the percentage of inhibition for selected sera tested on one of 6 coating antigens, including homologous (same haptens as immunogen) and heterologous (different hapten for immunization). Absence of the bar means that selected sera did not recognize corresponding coating antigen. Among tested combinations the following ones seemed to be the most promising for further exploration because the inhibition was above 20 % at 50 ng/mL TETS: 2165/2c, 2165/2k, 2165/6bCC, 2168/2c, 2168/2k, 2168/6bCC, 3442/2c, 3442/2k, 3443/2c, 3443/2k. These serum/coating antigen pairs were tested in an 8 point competitive format (Table S2). The combination of serum 3442 with coating antigen 2c/CON was chosen for the following studies because it had the highest sensitivity, high slope of 0.75 (target slope is 1.0 or more) and high signal-to-noise ratio. 3. Assay optimization. We then performed a series of optimization experiments to attempt to achieve better sensitivity and to characterize the assay for further applications. a. The 10 ACS Paragon Plus Environment

Page 11 of 31

1 2 3 217 4 5 6 218 7 8 219 9 10 220 11 12 13 221 14 15 222 16 17 223 18 19 20 224 21 22 225 23 24 25 26 226 27 28 227 29 30 228 31 32 33 229 34 35 230 36 37 231 38 39 40 232 41 42 43 233 44 45 46 234 47 48 235 49 50 236 51 52 53 237 54 55 238 56 57 239 58 59 60

Analytical Chemistry

effect of blocking agent on assay characteristics was evaluated first. The most common proteins of BSA, skimmed milk and ovalbumin (OVA) were used as 1% solutions in PBST. Blocking with skimmed milk resulted in significantly stronger reduction of the overall read-out signal compared to other proteins tested (Figure S1). Skim milk also gave the highest sensitivity with the IC50 value being 8 ng/mL vs 20 and 26 ng/mL obtained with OVA and BSA respectively. One probable reason for this observation is that casein, the major protein in skim milk, is smaller in molecular mass (~ 20 kDa) and thus blocks the surface of the plate more efficiently than BSA or OVA (66 kDa and 45 kDa). Better blocking then results in decreased non-specific binding of antibodies and improves assay sensitivity. Addition of the protein to the antibody assay buffer also had a significant effect on assay sensitivity. Addition of 1% BSA to the antibody solution (or 0.5% BSA on the plate at final dilution) resulted in a 2.4X increase in sensitivity (IC50 8 ng/mL) compared to the assay with no protein added to the antibody solution (IC50 19 ng/mL). This effect is indeed protein dependent because addition of a smaller amount of BSA (0.5 %) also improved the IC50 value (IC50 11 ng/mL, Figure S2). Interestingly, the overall signal of the assay also increased when some protein was added to antibody solution. b. Co-solvent selection. Organic co-solvents are often added to the immunoassay to improve analyte solubility or as a part of the sample preparation procedure. Methanol and DMSO were evaluated in the assay at final concentrations of 5, 10 and 20 % (Figure S3). The signal intensity did not change greatly when methanol was used at different concentrations. Similarly, sensitivity varied slightly giving IC50 values ranging between 3.8 ng/mL (20% methanol), 7.7 ng/mL (10% methanol), and 10 ng/mL (5 % methanol). However, DMSO had a dramatic effect on signal intensity but almost no effect on sensitivity. The signal 11 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 240 4 5 6 241 7 8 242 9 10 243 11 12 13 14 244 15 16 245 17 18 19 246 20 21 247 22 23 248 24 25 26 249 27 28 29 250 30 31 251 32 33 252 34 35 36 253 37 38 254 39 40 41 255 42 43 256 44 45 257 46 47 48 258 49 50 259 51 52 53 260 54 55 261 56 57 58 262 59 60

Page 12 of 31

suppression was concentration dependent, causing a larger decrease with higher concentrations of the solvent. These data suggest that methanol is a better co-solvent for the TETS ELISA in the current format, and it can be used at a wide range of concentrations in the sample with minimal effect on assay performance. Ionic strength of the buffer did not have a significant influence on assay sensitivity, but the maximum absorbance was negatively affected (Fig.S4). This suggests that the binding interaction between antibody and coating antigen is gradually suppressed in solutions with higher ionic strength. Therefore, the following experiments were performed in 10 mM PBS. The optimized ELISA included coating antigen-antibody pair 2c-CON/3442 with coating antigen concentration of 0.5 µg/mL. The coated plate was blocked with 1% skim milk. The analyte was loaded in assay buffer containing 20% methanol in PBS, pH 7.5. Serum dilution was 1:3000 in PBS containing 1% BSA, before addition to the plate. The heterologous assay had a linear range (IC20-80) of 0.7-30 ng/mL of TETS in assay buffer and IC50 value of 4.5±1.2 ng/mL (concentration reported as final in the well or 0.45 ng of TETS in the well, tested in triplicate for 9 days) (Fig. 2). The LOD in the buffer was determined as the IC10 value, and estimated to be 0.2 ng/mL (0.02 ng TETS in the well). To the best of our knowledge, this is the first immunoassay sensitive to TETS reported in the literature. The sensitivity of the assay is comparable to the sensitivity of reported GC/MS methods (in house and reported LOD ranging from 1 to 8 ng/mL following clean-up).9-11,14,15 4. Cross-reactivity (CR). As evidenced from the literature and based on multiple physiological and toxicological criteria, tetramine is a noncompetitive blocker of the GABA type A receptor (GABAAR).35-37 TETS competes with ethynylbicycloorthobenzoate (EBOB), a 12 ACS Paragon Plus Environment

Page 13 of 31

1 2 3 263 4 5 6 264 7 8 265 9 10 266 11 12 13 267 14 15 268 16 17 269 18 19 20 270 21 22 271 23 24 25 272 26 27 273 28 29 274 30 31 32 275 33 34 276 35 36 277 37 38 39 278 40 41 279 42 43 44 280 45 46 47 281 48 49 282 50 51 283 52 53 54 284 55 56 285 57 58 59 60

Analytical Chemistry

standard ligand of GABAAR, for binding to the receptor. A number of other organic small molecules have also been shown to have good affinity toward GABA receptors. 38 For instance, at 1 µM TETS inhibits 57% of EBOB binding, while fipronil and picrotoxinin give 69% and 51% inhibition, lindane and the cyclodiene α-endosulfan give 66% and 103% inhibition, respectively.38 Therefore, these compounds might share certain structural features that enable their binding to the same target site on the protein, and thus may also bind to the antibody developed to TETS. Figure S5 shows inhibition curves obtained with 5 tested molecules. Polychlorinated cage compounds like the cyclodiene heptachlor and the halogenated terpene toxaphene showed some inhibition in the assay, although with IC50 values were over 1000 ng/mL. This observation correlates with findings of Esser, et al. 39 who reported a monoclonal antibody (mAb), that was developed to cyclodienes to be sensitive to TETS with CR of 17% compared to α-endosulfans, or 2% compared to toxaphene. The authors supposed that the sulfamide moiety of TETS is comparable with the planar Cl-C-CCl2-C-Cl group of the cyclodienes, thus leading to ligand recognition.39 Other compounds (fipronil, picrotoxin and pentylenetetrazol) tested in the assay did not show any significant inhibition. Indeed CR was not expected since these compounds bind to a series of diverse sites on the GABA-gated chloride channels. When a series of analyte analogs are examined, their CR data may give some information about the binding preferences of the antibody. Figure S6 shows the CR pattern for serum 3442. A rabbit was immunized with hapten 2j having a benzene ring directly attached to the TETS cage. Accordingly, the serum showed a high CR toward analogs with highly similar structures (hapten 2j and analog 2i, Fig.S6). Lack of the phenyl ring (2a) significantly decreased the strength of the interaction between the antibody and TETS

13 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 286 4 5 6 287 7 8 288 9 10 289 11 12 13 290 14 15 291 16 17 292 18 19 20 293 21 22 294 23 24 25 295 26 27 28 296 29 30 297 31 32 298 33 34 35 299 36 37 300 38 39 301 40 41 42 302 43 44 303 45 46 47 304 48 49 305 50 51 306 52 53 54 307 55 56 308 57 58 59 60

Page 14 of 31

analog, however it was almost completely restored by introducing a substituent on the ethylene bridge (2c, 2b), suggesting that the nature of the side group is not important for binding efficiency. On the other hand, the bulky substituent in 2g led to less efficient interaction with the antibody and thus lower CR. Significantly lower CR to 4a and 2e compared to 2a is probably due to structural changes that resulted in a different relative position of the two sulfamide groups. In addition, the absence of CR toward 6bCC and 6aCC probably indicates that the presence of both sulfamide groups is important for antibody recognition. Finally, it is interesting to note that the selectivity of antibody to 2a and TETS is almost equal, suggesting that for antibody recognition the structure of the bridge, either methylene or ethylene does not have any influence. 5. Matrix effect on assay performance. Complex samples are challenging for accurate analysis and quantification because of the interfering components present in the sample matrix. For analytical instrumental methods like GC- or LC/MS sample clean-up is an inevitable step that is laborious and time consuming. On the other hand, even though the immunoassay performance may also be affected by the matrix, there are a number of simple methods to account for it including addition of the blank matrix to the calibration curve. Simple dilution of the sample also significantly decreases interference. In this study, commercial blank mouse and human serum, as well as human urine from healthy volunteers were tested. Mouse serum had a strong effect on assay performance. Figure S7 A shows significant decrease of the signal in the assay with 10% mouse serum. Interestingly, further increase in mouse serum content in the buffer did not give significantly higher signal suppression. It is worth noticing that as little as 1% mouse serum caused a pronounced decrease in the signal (data not shown).

14 ACS Paragon Plus Environment

Page 15 of 31

1 2 3 309 4 5 6 310 7 8 311 9 10 312 11 12 13 313 14 15 16 314 17 18 315 19 20 21 316 22 23 317 24 25 318 26 27 28 319 29 30 320 31 32 33 321 34 35 322 36 37 38 323 39 40 324 41 42 43 325 44 45 326 46 47 327 48 49 50 328 51 52 329 53 54 330 55 56 57 331 58 59 60

Analytical Chemistry

Compared to mouse serum, human serum had only a minor effect on the inhibition curve (Fig. S7C). This is a big advantage compared to instrumental methods since the analysis can be performed without any sample treatment. It is of particular importance when urgent evaluation must be performed after exposure to unknown chemicals. Interestingly, even though a urine matrix is supposed to be less complex compared to serum, it caused significant suppression of the immunoassay signal (Fig. S7E). However, case studies reported by Chau, et al. 40 and Lu, Wang, Yan, Xiao and Stephani 6 showed that concentration of TETS in human blood, urine and other biofluids is high, at least 50 ng/mL in mild poisoning cases. Accordingly, a 10-20x dilution will be necessary to adjust the concentration of TETS in samples to be within the analytical linear response range of the assay. Therefore, despite the interference of the matrix on assay performance, sample dilution significantly improves assay performance (Fig.S7). Another series of tests were performed to evaluate if adding higher concentration of methanol to the assay buffer may precipitate protein and/or salts from the samples, thus improving the assay signal. We tested two conditions where the assay buffer contained 10% methanol (Figures S7 A, C, E) or 20% methanol (Figures S7 B, D, F). Indeed, some differences were observed, in particular for the urine samples. When these samples were diluted 5-10 times (10-20% of matrix) in the buffer with 20% methanol, the signal did not differ significantly from the control, where no matrix was added. This is a significant improvement since urine samples could be diluted and tested in the assay without further sample treatment. However, this effect was not consistent with matrix. For human serum, better results were obtained with buffer containing 10% methanol, while for mouse serum the amount of co-solvent did not significantly influence the assay performance. Therefore, 15 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 332 4 5 6 333 7 8 334 9 10 335 11 12 13 336 14 15 337 16 17 338 18 19 20 339 21 22 23 340 24 25 341 26 27 28 342 29 30 343 31 32 344 33 34 35 345 36 37 346 38 39 347 40 41 42 348 43 44 349 45 46 47 350 48 49 351 50 51 352 52 53 54 353 55 56 354 57 58 59 60

Page 16 of 31

further experiments were conducted in assay buffer containing 10% methanol as the final concentration. These data also suggest that blank matrix should be added into the calibration curve samples to account for signal suppression for analysis of serum samples obtained from mice, but not for human serum (urine samples were not further analyzed). It should be noted that prior to the immunoassay analysis all biosamples should be adjusted to optimized assay conditions (pH and co-solvent). Therefore, while the assay LOD in the buffer is 0.2 ng/mL, the LOD of the method differs for different matrices and varies in the range of 0.4-0.8 ng/mL. Extraction procedure and efficiency. For analysis of samples from asymptomatic victims that may have experienced exposure to a very small amount of tetramine, such a high dilution of the sample is not applicable. Therefore, extraction of the analyte from the matrix is an inevitable sample preparation procedure. To evaluate the efficiency of extraction, we used 14C- labeled TETS previously synthesized by our group.38 Use of radioactive material for the development and validation of an extraction procedure is relatively little used and rarely appreciated in analytical chemistry. However, it is a very attractive method for extraction optimization since it allows fast and easy tracking of the distribution of spiked radioactive material within the sample and subsequent extraction fractions, by counting the number of decay events. This in turn allows more rapid optimization of the extraction conditions to achieve higher recoveries and cleaner extracted material. For TETS we used simple liquid-liquid extraction with ethyl acetate (EtOAc). Blank EtOAc, human urine, mouse and human serum were spiked with an aliquot of 14C-TETS with theoretical radioactivity of 1200 disintegrations per minute (dpm). Each matrix was spiked with TETS and extracted 3 times followed by liquid scintillation counting. To account

16 ACS Paragon Plus Environment

Page 17 of 31

1 2 3 355 4 5 6 356 7 8 357 9 10 358 11 12 13 359 14 15 360 16 17 361 18 19 20 21 362 22 23 363 24 25 364 26 27 28 365 29 30 366 31 32 367 33 34 35 368 36 37 369 38 39 370 40 41 42 371 43 44 372 45 46 47 373 48 49 374 50 51 375 52 53 54 376 55 56 377 57 58 59 60

Analytical Chemistry

for possible quenching due to the matrix, the same matrix was extracted first and then spiked with 14C-TETS. Activity of 14C-TETS was determined first in spiked EtOAc and gave 1029±36 counts per minute (cpm) (n=5, Table S3). There was no significant quenching observed in any of the tested matrixes, as can be seen from Table S3. Radioactivity measurements in extracted samples revealed that liquid-liquid extraction with EtOAc was very efficient for TETS, as evidenced by high recovery values for mouse and human serum at 93 and 90% respectively, and a 95% recovery from human urine. Immunoassay evaluation with spiked samples. To characterize the immunoassay developed to quantitatively detect TETS in biofluids, we performed a recovery evaluation from spiked matrices, including buffer, mouse and human serum all with and without an extraction step. Since we did not expect to obtain urine samples from exposed humans (not available) or mice (very low volumes), urine matrix was not included in spike-and-recovery studies. Buffer and control sera samples were spiked with TETS ranging from 2-5000 ng/mL and then quantified in a blind fashion (by a different operator). Good recoveries were obtained with the ELISA generally ranging from 80 to 120% in buffer (Table 2, part 1). Recoveries from mouse serum were slightly lower for samples spiked with higher concentrations. This is an expected observation. Samples with high concentration spikes had to be largely diluted to obtain a signal in the linear range while the calibration curve was built with a blank serum added to account for matrix effect. This dilution resulted in decreased matrix effect and slight underestimation of TETS concentration. Recoveries from human serum were consistent with theoretical spike concentrations with some insignificant overestimation. It should be noted that the calibration curve was built in the assay buffer containing no human serum. Thus, the slight overestimation might be related to a minor

17 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 378 4 5 6 379 7 8 380 9 10 381 11 12 13 382 14 15 383 16 17 384 18 19 20 21 385 22 23 386 24 25 387 26 27 28 388 29 30 389 31 32 390 33 34 35 391 36 37 392 38 39 393 40 41 42 394 43 44 395 45 46 47 396 48 49 397 50 51 398 52 53 54 399 55 56 57 400 58 59 60

Page 18 of 31

matrix effect of the human serum. Nevertheless, the overall recovery results suggest that the ELISA is an excellent tool for rapid, cheap, high throughput and very sensitive analysis of a variety of samples that does not require special sample treatment, clean-up or preconcentration. For comparison purposes, we also performed analysis of a separate series of spiked samples, where TETS was extracted and then analyzed with the immunoassay. As expected, cleaned up samples had improved recovery values making them more consistent with theoretical values and decreased recovery variation (Table 2, part 2). Similar experiments were conducted with an instrumental analytical method- GC/MS run under conditions similar to the EPA approved method41 (Table 2). Consistent correlation with theoretical values and ELISA results was observed. However, despite addition of isotope labeled 13C-TETS as an internal standard and clean-up with solid-phase microextraction (SPME), GC/MS data were variable and generally provided overestimated results showing a strong effect of the matrix on the instrumental analysis (Table 2, part1). Recoveries from serum samples were particularly affected by the matrix. Therefore, for the mouse serum series we also tried LLE as an alternative to SPME. Simple LLE improved recovery values, but produced some underestimated values (Table S4). Table 2 part 2 shows that LLE efficiently removed TETS from the matrix and provided clean extracts for ELISA resulting in excellent recoveries mostly in the range of 80-120%. However, this cleanup procedure was not sufficient for GC/MS, resulting in overestimation of 40-60%. Both methods gave highly variable results for the low range of spiked concentrations 2-5 ng/mL, indicating this concentration to be near the limit of quantification of the methods. The results obtained are in good agreement with published data. They show that indeed, compared to LLE, more sophisticated methods for sample extraction and clean-up 18 ACS Paragon Plus Environment

Page 19 of 31

1 2 3 401 4 5 6 402 7 8 403 9 10 404 11 12 13 405 14 15 406 16 17 407 18 19 20 21 408 22 23 409 24 25 410 26 27 28 411 29 30 412 31 32 413 33 34 35 414 36 37 415 38 39 416 40 41 42 417 43 44 418 45 46 47 419 48 49 420 50 51 421 52 53 54 422 55 56 423 57 58 59 60

Analytical Chemistry

are needed to provide more accurate results with the instrumental methods. However, reported modern techniques for sample preparation, including SPME, stir bar sorptive extraction or membrane assisted solvent extraction remain time consuming and may require sophisticated techniques to perform high throughput samples preparation and analysis. 9,11,12,14,15

Therefore, the immunoassay developed in this work provides an advantageous

alternative to instrumental methods. It is a fast, cheap and high throughput method with adequate accuracy, while not requiring any sample pre-treatment. Quantification of TETS in the serum of exposed mice. We sought to evaluate the ELISA in real samples from exposed subjects. Since human samples are unavailable, we generated in-house samples by exposing mice to TETS at 2 x LD50 dose (0.2 mg/kg, intraperitoneal route). Animals were protected from lethality by pre-treating them with 10 mg/kg riluzole, a sodium channel blocker. The immunoassay was used to quantify the TETS concentration in the serum and to monitor the kinetics of TETS clearing from blood over a short period of time. Table 3 gives TETS concentrations detected in groups of animals bled at 30 min, 60 min, 120 min and 240 min after treatment. Over the 240 min period, TETS remained in the blood at 150 ng/mL with only slight differences between different time points and different animals. This concentration corresponds to about 10% of the injected dose. These data were confirmed with GC/MS analysis that resulted in similar TETS concentrations in corresponding samples. The rest of the injected TETS might have distributed in other organs and tissues, with the major part being excreted with feces as reported by Radwan et al. 42 Even though limited data are available on the biology of TETS and there are no published data on TETS pharmacokinetics when administrated through injection, it is known that tetramine is slowly eliminated from the body.40 For example, from

19 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 424 4 5 6 425 7 8 426 9 10 427 11 12 13 428 14 15 429 16 17 430 18 19 20 431 21 22 432 23 24 25 433 26 27 28 434 29 30 435 31 32 436 33 34 35 437 36 37 438 38 39 439 40 41 42 440 43 44 441 45 46 47 442 48 49 443 50 51 52 444 53 54 445 55 56 57 446 58 59 60

Page 20 of 31

case studies, Chau et al. 40 reported that TETS was identified in vomitus and blood even 1 week after poisoning. They also noted that TETS has a high volume of distribution and a slow metabolism causing it to stay in the body for up to 6 months after exposure. In a separate study42, where tetramine was orally administrated to mice, TETS appeared in the blood within 10 min after dosing and its concentration remained constant for 3 days. These few published data support our findings in this work. They also support the hypothesis that the high toxicity of TETS seen with only moderate binding to the GABA A chloride channel is explained in part by slow clearance. They also caution that effective therapies need to be long lasting. There is limited knowledge on TETS distribution in the human body and more research is required to understand its pharmacokinetics. The antibodies against TETS and corresponding immunoassay provide very useful tools and are clearly a significant progress in the analysis field. Immunoassay provides a convenient way for high throughput screening of samples with high sensitivity and good accuracy. The immunoassay can be employed for monitoring TETS distribution and rates of its elimination in treated animals. The antibodies also can be used to show that 14C in tissue following 14C TETS administration is in fact TETS. In view of poisoning cases reported in China, the developed immunoassay may be formatted into a portable biosensor device to provide a rapid screening tool with simple yes/no readout when rapid identification of unknown material is needed, for instance in subway stations, airports, or governmental institutions and may aid in therapy. Recently, there have been reports of intentional poisoning of children and adults at public organizations like daycares and grocery stores in China.43-45 Reports indicate that 30 kindergarten children were poisoned and two victims died. 44 Therefore, rapid and cheap 20 ACS Paragon Plus Environment

Page 21 of 31

1 2 3 447 4 5 6 448 7 8 449 9 10 450 11 12 13 451 14 15 16 452 17 18 453 19 20 21 454 22 23 455 24 25 456 26 27 28 457 29 30 458 31 32 459 33 34 35 460 36 37 461 38 39 462 40 41 42 463 43 44 464 45 46 47 465 48 49 50 466 51 52 53 467 54 55 56 468 57 58 59 60

Analytical Chemistry

immunoassay based screening tools may allow establishment of a security routine and control in public organizations involving food consumption. Additionally, formatted as a more sophisticated biosensor it can be used for evaluation of the degree of exposure or poisoning in victims. In conclusion, this is the first report in the literature describing development of a sensitive immunoassay selective to TETS. The assay showed a detection limit around 0.2 ng/mL that is lower or in the range of the GC/MS limit of detection. As an advantage, the immunoassay does not require sample pre-treatment, clean-up or pre-concentration prior to analysis. The immunoassay has the potential for being adapted into a sensor for rapid infield screening and monitoring purposes. The advantage of the high throughout and cheap analysis may help for in-time detection of contamination with TETS or early detection of the toxin in the blood of the victims. Timely, selective and quantitative detection of the poisoning with TETS may allow fast and appropriate treatment of the victims resulting in better clinical outcomes. ACKNOWLEDGMENT This work was supported by the CounterACT Program, National Institutes of Health Office of the Director, and the National Institute of Neurological Disorders and Stroke, Grant Number U54 NS079202. The research was also supported by National Institute of Environmental Health Sciences, Superfund Research Program, P42 ES04699. SUPPORTING INFORMATION Additional information including text, four tables and seven figures are available in the supporting information.

21 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 469 4 5 470 6 471 7 472 8 9 473 10 474 11 475 12 476 13 477 14 15 478 16 479 17 480 18 481 19 482 20 483 21 484 22 485 23 24 486 25 487 26 488 27 489 28 490 29 491 30 31 492 32 493 33 494 34 495 35 496 36 497 37 38 498 39 499 40 500 41 501 42 502 43 503 44 504 45 46 505 47 506 48 507 49 508 50 509 51 510 52 511 53 54 512 55 513 56 514 57 515 58 59 60

Page 22 of 31

References (1) International Program on Chemical Safety http://www.inchem.org/documents/pims/chemical/pim982.htm#2.1. (2) Li, J. M.; Gan, J.; Zeng, T. F.; Sander, J. W.; Zhou, D. Neurotoxicology 2012, 33, 207-211. (3) Li, Y.; Gao, Y. X.; Yu, X. Z.; Peng, J. M.; Ma, F.; Nelson, L. Bmc Public Health 2014, 14. (4) Center of Disease Control and Prevention: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5210a4.htm. (5) Deng, X. J.; Li, G.; Mei, R. W.; Sun, S. G. Clin Toxicol 2012, 50, 172-175. (6) Lu, Y.; Wang, X.; Yan, Y.; Xiao, Z.; Stephani, U. Clin Neuropharmacol 2008, 31, 57-61. (7) Lu, Y.; Wang, X. F.; Yan, Y.; Xiao, Z.; Stephani, U. Clin Neuropharmacol 2008, 31, 57-61. (8) Chau, C.; Leung, A.; Tan, I. Hong Kong Med J 2005, 11, 511-514. (9) Hamelin, E.; Swaim, L.; Thomas, J.; Kobelski, R.; Johnson, R. C. J Chromatogr B 2010, 878, 2541-2547. (10) Xu, X. M.; Song, G. L.; Zhu, Y.; Zhang, J.; Zhao, Y. X.; Shen, H. T.; Cai, Z. X.; Han, J. L.; Ren, Y. P. J Chromatogr B 2008, 876, 103-108. (11) Owens, J.; Hok, S.; Alcaraz, A.; Koester, C. J Agr Food Chem 2009, 57, 4058-4067. (12) De Jager, L. S.; Perfetti, G. A.; Diachenko, G. W. J Sep Sci 2009, 32, 1081-1086. (13) Knaack, J. S.; Hamelin, E. I.; Magnuson, M.; Silvestri, E.; Ash, D.; Johnson, R. C. Anal Methods-Uk 2014, 6, 2780-2784. (14) Luan, T. G.; Li, G. K.; Zhao, M. Q.; Zhang, Z. X. Anal Chim Acta 2000, 404, 329-334. (15) Zeng, D.; Chen, B.; Yao, S. Z.; Ying, J. Y. Forensic Sci Int 2006, 159, 168-174. (16) Chau, C.; Leung, A.; Tan, I.; sdsds; sdsqd. Hong Kong Med J 2005, 11, 511-514. (17) Dehua, G.; Daxi, J.; Honglang, X.; Bin, X.; Yun, L.; Leishi, L. Blood Purificat 2006, 24, 524-530. (18) Bever, C. S.; Dong, J. X.; Vasylieva, N.; Barnych, B.; Cui, Y.; Xu, Z. L.; Hammock, B. D.; Gee, S. J. Anal Bioanal Chem 2016, 408, 5985-6002. (19) Bahlmann, A.; Carvalho, J. J.; Weller, M. G.; Panne, U.; Schneider, R. J. Chemosphere 2012, 89, 1278-1286. (20) Vasylieva, N.; Ahn, K. C.; Barnych, B.; Gee, S. J.; Hammock, B. D. Environmental science & technology 2015, 49, 10038-10047. (21) Zhang, W. Y.; Asiri, A. M.; Liu, D. L.; Du, D.; Lin, Y. H. Trac-Trend Anal Chem 2014, 54, 1-10. (22) Kantiani, L.; Llorca, M.; Sanchis, J.; Farre, M.; Barcelo, D. Anal Bioanal Chem 2010, 398, 2413-2427. (23) Meulenberg, E. P. Toxins 2012, 4, 244-266. (24) Xu, M. L.; Liu, J. B.; Lu, J. Appl Spectrosc Rev 2014, 49, 97-120. (25) Morozova, V. S.; Levashova, A. I.; Eremin, S. A. J Anal Chem+ 2005, 60, 202-217. (26) Ngom, B.; Guo, Y. C.; Wang, X. L.; Bi, D. R. Anal Bioanal Chem 2010, 397, 1113-1135. (27) Liao, J. Y.; Li, H. Microchim Acta 2010, 171, 289-295. (28) Mallat, E.; Barcelo, D.; Barzen, C.; Gauglitz, G.; Abuknesha, R. Trac-Trend Anal Chem 2001, 20, 124-132. (29) Barnych, B.; Vasylieva, N.; Joseph, T.; Hulsizer, S.; Nguyen, H. M.; Cajka, T.; Pessah, I.; Wulff, H.; Gee, S. J.; Hammock, B. D. Chemistry - A European Journal (Provisionally accepted) 2017. (30) Hong, V.; Presolski, S. I.; Ma, C.; Finn, M. G. Angew Chem Int Edit 2009, 48, 9879-9883. (31) Lee, N.; Beasley, H. L.; Skerritt, J. H. J Agric Food Chem 1998, 46, 535-546. (32) Wie, S. I.; Hammock, B. D. J. Agric. Food Chem. 1984, 32, 1294-1301. (33) Liu, Y. H.; Xie, R.; Guo, Y. R.; Zhu, G. N.; Tang, F. B. J. Environ. Sci. Health B 2012, 47, 475-483. (34) Cao, B.; Yang, H.; Song, J.; Chang, H.; Li, S.; Deng, A. Talanta 2013, 116, 173-180. (35) Bowery, N. G.; Brown, D. A.; Collins, J. F. British journal of pharmacology 1975, 53, 422-424. (36) Squires, R. F.; Casida, J. E.; Richardson, M.; Saederup, E. Molecular pharmacology 1983, 23, 326-336. (37) Ratra, G. S.; Kamita, S. G.; Casida, J. E. Toxicology and applied pharmacology 2001, 172, 233-240. (38) Zhao, C. Q.; Hwang, S. H.; Buchholz, B. A.; Carpenter, T. S.; Lightstone, F.; Yang, J.; Hammock, B. D.; Casida, J. E. P Natl Acad Sci USA 2014, 111, 8607-8612. 22 ACS Paragon Plus Environment

Page 23 of 31

1 2 3 516 4 517 5 518 6 7 519 8 520 9 521 10 522 11 523 12 524 13 525 14 15 526 16 17 527 18 528 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

(39) Esser, T.; Karu, A. E.; Toia, R. F.; Casida, J. E. Chemical research in toxicology 1991, 4, 162-167. (40) Chau, C. M.; Leung, A. K.; Tan, I. K. Hong Kong Med J 2005, 11, 511-514. (41) EPA600/R-11/091, High Throughput Determination of Tetramine in Drinking Water by Solid Phase Extraction and Isotope Dilution Gas Chromatography/Mass Spectrometry (GC/MS): Centers for Disease Control and Prevention, 2011. (42) Radwan, M.; Dodge, W. Northwest Science 1970, 44, 25-30. (43) Carney, J.; Young sisters collapse and die on their way home from school after their brother put poison in their cola, online article at dailymail.com, 2015. (44) Ramzy, A. In Sinosphere Dispatches from China, 2014. (45) Choi, S.; 1 dead, 4 hospitalized after man injects poison into Wanglaoji drinks across Dongguan, online article in Shanghaiist News, 2015.

23 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 529 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 47 530 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 31

Table 1. Selected haptens for immunoassay development. 1ST SERIES

2ND SERIES

Resulting antibodies did not sufficiently recognize TETS

Resulted in antibodies selective to TETS

24 ACS Paragon Plus Environment

Page 25 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 531 21 22 532 23 24 533 25 26 534 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

Figure 1. Screening for successful pairs of coating antigen/serum. Criteria of success is 50% of inhibition and over, at 50 ng/mL TETS. Absence of the bar indicates that selected sera did not recognize corresponding coating antigen. Conalbumin was used as a carrier protein.

25 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 535 19 536 20 21 22 537 23 24 538 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 26 of 31

Figure 2. TETS competitive curve with rabbit serum 3442 in assay buffer. Assay conditions: coating antigen 0.5 µg/mL (2c-CON, on the left), anti-TETS serum in 1/6000 dilution; goat anti-rabbit IgG-HRP (1/10000). IC50 4.5±1.2 ng/mL (n=9) (or 0.45±0.12 ng in the well).

26 ACS Paragon Plus Environment

Page 27 of 31

1 2 3 539 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 53 540 54 55 541 56 57 58 59 60

Analytical Chemistry

Table 2. Recovery of TETS from spiked samples Spike, ng/mL

ELISA, ng/mL

%*

GC/MS, %* ng/mL

1. Without extraction Buffer 4 5±1.2 10 13.4±4.9 50 55±14 500 400±68 2000 1930±330 Mouse serum 2 1.4±0.2 8 7.6±2.6 40 26±7 300 161±38 4000 3820±1220 Human serum 3 4.3±2.4 20 15.5±3.6 40 56±16 200 202±36 3000 3610±350 2. With extraction Mouse serum 4.6 11.6±4.3 6.5 9.5±2.1 23.8 28±10 455 447±17 1456 1440±390 Human serum 3.7 3±0.1 12.2 11.1±1.6 45.5 49±7 238 218±30 2381 2250±370 * % of recovery

125 134 110 80 96

6.7