Use of novel haptens in the production of antibodies for the detection of tryptamines

Tryptamines are a group of hallucinogenic drugs whose detection in body fluids could be simplified by immunochemical assay kits. Antibodies for these assays are obtained by the immunization of laboratory animals with conjugates of a hapten similar to the target analyte and a suitable protein. Therefore we synthesized novel haptens derived from tryptamine-based drugs, with N,N-dimethyltryptamine (DMT), 5-methoxy-N,N-dimethyltryptamine (5-MeO-DMT) and N,N-diisopropyltryptamine (DiPT) selected as the target analytes. Their structures were modified with a short linker ended with a carboxylic group. The haptens were conjugated with bovine serum albumin (BSA) and rabbits were immunized with the conjugates. The obtained polyclonal antibodies showed good reactivity and the LOD of the constructed ELISAs was in the range 0.006–0.254 ng mL−1. Thus, they are suitable for the development of immunochemical assay kits.


Introduction
Tryptamines, a group of hallucinogenic drugs derived from tryptamine, include both natural and synthetic compounds. The best known tryptamine is dimethyltryptamine (DMT), the active compound of ayahuasca, the ritual beverage, which has traditionally been used in several South American indigenous cultures. 1 Other traditional natural tryptamines are psilocybin and psilocin, active compounds of the Psilocybe mushrooms also known as 'magic mushrooms'. Over the past few decades, lots of synthetic derivatives of tryptamine have been prepared. Their rise in popularity has been attributed to Alexander Shulgin, whose book TiHKAL (Tryptamines I have known and loved) described the synthesis and effects of many of them. 2 These synthetic tryptamines belong to the group of novel psychoactive substances (NPS), which are produced to give 'legal highs' and bypass legal restrictions. 3 Although there is widespread use of NPS on the drug scene, options for their detection, including for the detection of tryptamines, remain rather limited.
There are currently only two methods for the detection of tryptamines. The rst involves the use of specic colour tests, which are based on the reaction of specic reagents forming coloured products with indole-derived compounds. But, with so many different NPS now on the drug scene, colour tests are not reliable because of their low specicity. 4 The more commonly used second method involves the analysis of tryptamines by liquid chromatography coupled with mass spectrometry (LC-MS). 5,6 This method is precise, capable of determining multiple targets in one run, offers low detection limits and can be used for various matrices, especially body uids. However, it is also demanding with respect to cost, sample preparation and analysis time. Furthermore, LC-MS-based techniques cannot be used in the eld.
Because of the above-mentioned disadvantages, attention has focused on immunochemical methods, such as ELISA (enzyme-linked immunosorbent assay) and LFIA (lateral ow immunosorbent assay). Antibodies for immunochemical detection are usually produced by the immunization of laboratory animals. Because tryptamines are haptens, i.e. molecules too small to be immunogenic on their own, they must be conjugated to a carrier protein prior to immunization. To do this, molecules of the target analytes are modied with short linkers containing a suitable functional group. Using bufotenine and serotonin as haptens, Skerritt et al. reported the development of an ELISA kit for the detection of DMT and 5-MeO-DMT in Phalaris plants. 7 However, this assay is limited only to these natural tryptamines and is not selective between them. Yamaguchi et al. prepared monoclonal antibodies against psilocin for identication of magic mushrooms. 8 These antibodies showed cross reactivity with DMT, but their limit of detection was relatively high. Thus far, no other immunochemical assay has been reported that targets either natural or synthetic tryptamines.
The development of LFIA kits for the detection of synthetic and other tryptamines is needed to screen potential users. To provide such a kit, we rst synthesized novel haptens carrying a short linker with a carboxyl group for the production of antibodies selective to different tryptamines. Rabbits were immunized with conjugates of these haptens with bovine serum albumin (BSA). The antibodies obtained from the immunization showed expedient sensitivity and selectivity for individual tryptamines, and, thus, appear to provide a viable basis for LFIA development.

Materials and methods
Ethyl 2-bromoacetate was obtained from Merck and N,N 0 -dicyclohexylcarbodiimide (DCC) was obtained from Fluka. Diisobutylaluminium hydride (DIBAL-H) solution in hexane, oxalyl chloride and dry DMF were purchased from Acros. Other reagents were purchased from Sigma-Aldrich. Dry THF, dichloromethane and diethyl ether were dried with molecular sieves. All reactions were carried out under argon atmosphere. Thin layer chromatography (TLC) was performed on aluminium backed sheets coated with 60F 254 silica gel from Merck. Column chromatography was performed on silica (0.045-0.200 mm) from Merck. Reverse phase chromatography was carried out using a CombiFlash Rf 200 apparatus (Teledyne ISCO) with prepacked Redisep Rf Gold C18 columns (packed with C18reverse phase silica gel). NMR spectra were recorded on a Varian Gemini 300 (300 MHz for 1 H; 75 MHz for 13 C) or Agilent 400-MR DD2 (400 MHz for 1 H; 100 MHz for 13 C) spectrometers. High resolution mass spectra were measured on a LTQ Orbitrap XL (Thermo Fischer Scientic) spectrometer using ESI ionization technique. Mass spectra of hapten-protein conjugates were measured on a Bruker Autoex Speed MALDI-TOF/TOF spectrometer. Microplate reader uQuant BIO-TEK was from Inc. Winooski, USA and 96-well polystyrene microtiter plates Costar 9018 were purchased from Corning Inc., USA. Peroxidase labelled goat anti-rabbit antibody (GAR-Po) was obtained from Nordic Immunological Laboratories, Netherlands. Analytical standards psilocin and psilocybin were obtained from THC Pharm GmbH The Health Concept, Germany. Analytical standards of DMT, 5-MeO-DMT, DiPT and 5-MeO-DiPT were prepared according to literature 9 in purity $98% (LC). Serotonin was purchased from Sigma-Aldrich. Other tryptamines 5-MeO-DALT, 4-HO-MET, acetylpsilocin, aMT and 5-MeO-DIPT were generously donated by the Institute of Criminalistics Prague.

Preparation of conjugates
Following the procedure we described previously, 10 the prepared haptens I-IV were conjugated to BSA using activated ester method. Obtained conjugates I-IV were analyzed by MALDI-TOF to determine the number of hapten molecules bound to the protein. The average values were calculated from the peak with highest intensity.

Immunisation and antiserum preparation
Four different immunisation conjugates (conjugate I-IV) were used. Antisera were produced in rabbits and obtained according to the procedure described previously. 11 Stock solutions of antisera were prepared by dissolving 1 mg of lyophilisate in 1 ml of PBS and were stored at À20 C.

Indirect competitive ELISA
ELISA microtiter plates were coated with the coating conjugates (appropriately diluted in the carbonate-bicarbonate buffer; 100 ml per well) and incubated at 4 C overnight. The following day, the plates were washed with the PBS-Tw (4 times, 350 ml per well). The suitable standard (the concentration range of 0-500 ng mL À1 in the PBS; 50 ml per well) was added into microtiter plates followed by the solution of appropriate antiserum (diluted in the PBS-0.1% BSA (w/v); 50 ml per well) and incubated at room temperature for 1 hour. Microtiter plates were washed again (4 times, 350 ml per well) and GAR-Po (diluted 1 : 10 000 in the PBS-Tw; 100 ml per well) was added and incubated at room temperature for 1 h. Aer the washing step, the TMB substrate solution was added (100 ml per well) and incubated at room temperature for 10 min. The enzyme reaction was stopped by addition of 2 mol L À1 H 2 SO 4 (50 ml per well) and the absorbance was measured at 450 nm.

Calibration standard curve
Sigmoid calibration standard curves were obtained by plotting the mean values of absorbance against the logarithm of standard concentrations through a four-parameter logistic equation as described previously. 11 The limit of detection (LOD) was dened as the concentration of an analyte corresponding to the maximum assay signal minus 3Â standard deviation (SD) in accordance with the calibration curve (the blank was calculated from 3 parallel determinations with the absence of an analyte). The IC 50 corresponded to the concentration of analyte giving 50% inhibition of the asymptotic maximum. The linear working range corresponded to the analyte concentration causing the 20-80% inhibition of the maximum assay signal.

Specicity of antibodies
To verify the ability of the antibodies to react with similar epitopes on different tryptamines, the cross-reactivity (CR) tests were performed. The percent CR (CR (%)) was calculated from IC 50 obtained from the calibration curves: (IC 50 of target drug)/ (IC 50 of tested compound) Â 100.

Results and discussion
To design the structure of haptens, DMT, 5-MeO-DMT and DiPT were selected as target analytes and their structures were modied with a short alkyl chain ended with a carboxylic group. Four different haptens were prepared, differing in a position of the linker (Fig. 1). Hapten I (1a) with the linker on the amino group was derived from DMT. Haptens II-IV (1b-d) with the linker in position 5 of indole ring were derived from 5-MeO-DMT (haptens II and III) and DiPT, respectively (hapten IV).
We used an approach based on the Fischer reaction for the synthesis of haptens II-IV (1b-d). Haptens were obtained directly from suitable arylhydrazines and amino acetals (Fig. 3).
Because the same sequence of reactions could not be used in the synthesis of acetal 10, we developed a different approach starting from succinic anhydride (13) (Fig. 5). Reaction of 13 with diisopropylamine and subsequent reduction of acid 14 with LiAlH 4 led to the amino alcohol 15. Swern conditions were   Paper used for oxidation of 15 to aldehyde, which was immediately acetalized to obtain 10 in moderate yield.
Haptens I-IV (1a-d) were conjugated with bovine serum albumin (BSA) using the methodology previously employed in our group. 10,14 Conjugates I-IV were submitted to MALDI-TOF analysis and the number of hapten molecules was determined as follows: 13 for conjugate I, 31 for conjugate II, 28 for conjugate III and 37 for conjugate IV.
Antisera were collected by immunization of rabbits with all of prepared conjugates. 11 The indirect competitive format of ELISA was used for antiserum testing. First, checkerboard titrations were performed and suitable immunoreagent concentrations were determined when the maximum absorbance ranged from 1.2 to 1.9. Then the calibration curves with target analyte were constructed. The antibody with the highest sensitivity to the appropriate analyte was selected (on the basis of the lowest IC 50 values) for each immunisation conjugate and further tests were made. In this study, four different ELISAs (ELISA I-IV) for the detection of tryptamine-based drugs were successfully developed. Titration standard curves are shown in Fig. 6 and analytical parameters of methods are summarized in the table (Table 1).
General procedure A: preparation of arylhydrazinium chlorides. A procedure from the literature 19 was modied as follows: suspension of aniline acid 8a,b (1 eq.) in concentrated hydrochloric acid (3 ml mmol À1 ) was cooled to À5 C and a solution of sodium nitrite (1.05 eq.) in water (0.5 ml mmol À1 ) was added dropwise. Aer complete addition the mixture was stirred for 1 hour at À5 C and then it was added dropwise to a solution of SnCl 2 $2H 2 O (3 eq.) in concentrated hydrochloric acid (1 ml mmol À1 ) cooled to À20 C. Aer complete addition, the mixture was stirred for additional 2.5 hours at À20 C and then the suspension was ltered. Solids were washed with cold ethanol and diethyl ether and dried, yielding arylhydrazinium chlorides 7a,b.
4-(N,N-diisopropylamino)-4-oxobutanoic acid (14). To a solution of succinic anhydride (13) (10.07 g, 100.0 mmol) in dichloromethane (300 ml) was added diisopropyl amine (21.8 ml, 300.0 mmol) and resulting solution was stirred for 20 hours. Then the mixture was concentrated and 1 M hydrochloric acid (250 ml) was added to the residue. Aqueous layer was extracted with dichloromethane (3 Â 150 ml) and combined organic layers were dried with MgSO4. Solvent removal offered the titled compound 14 as a light brown viscous oil (16.50 g, 82%), which solidied upon standing in a fridge. 1  4-(N,N-diisopropylamino)-1,1-dimethoxybutane (10). Dimethyl sulfoxide (2.56 ml, 36.0 mmol) was added dropwise to a solution of oxalyl chloride (2.57 ml, 30.0 mmol) in dry dichloromethane (60 ml) cooled to À55 C and the mixture was stirred for additional 10 minutes. Then the solution of alcohol 15 (2.60 g, 15 mmol) in dry dichloromethane (10 ml) was added and the mixture was stirred at À55 C for additional 15 minutes. Then trimethylamine (8.9 ml, 60.0 mmol) was added and the mixture was allowed to warm to r.t. (1 hour). Reaction mixture was then poured into water (130 ml), phases were separated and the aqueous one extracted with dichloromethane (2 Â 100 ml). Combined organic phases were dried with MgSO 4 and evaporated. The residue was dissolved in methanol (50 ml), acidied with concentrated sulfuric acid (2.5 ml) and the mixture was stirred at r.t. for 18 hours. Then the mixture was diluted with dichloromethane (150 ml), cooled with ice bath and 20% NaOH solution was added (100 ml). Phases were separated, the aqueous one was diluted with water (50 ml) and extracted with dichloromethane (2 Â 75 ml). Combined organic phases were washed with brine (200 ml) and dried with MgSO 4 . Distillation under reduced pressure gave the titled amino acetal 10 as colorless liquid (1.74 g, 54%). Bp ¼ 51-55 C (0.24 torr); 1  General procedure B: preparation of haptens II-IV. A procedure from the literature 13 was modied as follows: to a 4% sulfuric acid, which was rst heated to 50 C and bubbled with argon, arylhydrazinium chloride 7a,b (1 eq.) and then amino acetal 9 or 10 (1.2 eq.) were added and the resulting mixture was heated to 80 C for 3.5 hours. Aer cooling to r.t., the mixture was neutralized with concentrated ammonia solution. Water was removed under reduced pressure and the residue was treated with ethanol (10 ml mmol À1 ) and ltered to remove most of the inorganic salts. Filtrate was evaporated, 1 M hydrochloric acid (10 ml mmol À1 ) was added and resulting solution was evaporated again. Purication of crude product by reverse-phase ash chromatography (water/methanol, gradient 5-100% of methanol) gave haptens 1b-d (haptens II-IV).

Conclusion
We successfully used haptens with novel structures to produce polyclonal antibodies against various tryptamines. The constructed ELISAs have low detection limits. Some of the antibodies show good reactivity not only with the target analytes, but also with psilocin and 5-MeO-DiPT. Although the antibodies have not yet been characterized in complex matrices, they appear to be suitable for the development of immunochemical assay kits. In our next work, we will focus on the establishment of an ELISA for the detection of tryptamines in human body uids. We believe that the outcome of our work could lead to LFIA kits designed for the on-site testing of NPS users.

Conflicts of interest
None.