Forensic Science International 280 (2017) 2534

Contents lists available at ScienceDirect

Forensic Science International

journal homepage: www.elsevier.com/locate/forsciint

Determination of cross-reactivity of poly- and monoclonal antibodies

for synthetic cannabinoids by direct SPR and ELISA
Nico Langera , Franziska Steinickeb , Rainer Lindigkeita , Ludger Ernstc , Till Beuerlea,*
a
Institute of Pharmaceutical Biology, Technische Universitt Braunschweig, Mendelssohnstr. 1, 38106 Braunschweig, Germany
b
Institute of Pharmaceutical Chemistry, Technische Universitt Braunschweig, Mendelssohnstr. 1, 38106 Braunschweig, Germany
c
Chemistry Department, Central NMR Laboratory, Technische Universitt Braunschweig, Hagenring 30, 38106 Braunschweig, Germany

A R T I C L E I N F O A B S T R A C T

Article history:
Received 16 May 2017 One of the main reasons for the rise in popularity of synthetic cannabinoids (SCs) is their ability to remain
Received in revised form 30 August 2017 unrecognized in conventional drug screenings. Due to their structural diversity, caused by the constant
Accepted 8 September 2017 introduction of new substances to circumvent legal regulation, antibodies with a wide range of cross-
Available online 18 September 2017 reactivity are necessary for the establishment of a reliable immunological based drug test. Therefore,
high-quality binding data are needed to select promising antibody candidates for further development.
Keywords: In this study, we carried out a direct surface plasmon resonance (SPR) method and evaluated its
Surface plasmon resonance (SPR) suitability for the characterization of antibodySC interactions. The cross-reactivity of 22 SCs with three
Enzyme-linked immunosorbent assay
polyclonal antibodies, raised against JWH-018 haptens with different attachment positions of the linker,
(ELISA)
and two commercial available monoclonal antibodies were determined. These results were compared
Synthetic cannabinoids
Cross-reactivity with the commonly used competitive enzyme-linked immunosorbent assay (ELISA). It could be
demonstrated, that direct SPR and competitive ELISA show comparable specicity results for the majority
of the measured compounds. However, the reduced manual labor, the real-time analysis and the high
information content about the binding events of SPR compared to ELISA, showed that SPR is a valuable
tool during the development of antibodies against synthetic cannabinoids, currently the largest group of
new psychoactive substances.
2017 Elsevier B.V. All rights reserved.

1. Introduction appearing compounds, followed by a quick replacement of the

Synthetic cannabinoids (SCs) were rst detected as active
ingredients in herbal smoking mixtures in the end of the year 2008
[1] and swiftly became the predominant group of compounds
within the new psychoactive substances (NPS) worldwide [2]. In
less than a decade, the number of synthetic cannabinoids listed by
the United Nations Ofce on Drugs and Drug Addiction (UNODC)
increased from 31 in 2009, over 154 in 2013 to 241 in 2016 [24].
The popularity and motivation for the consumption of SC can be
explained by several reasons: (i) psychoactive effects similar to
cannabis, (ii) curiosity and easy availability via internet shops, (iii)
new SCs were usually not covered by national narcotic laws, which
in turn was widely misinterpreted that these compounds are legal,
and (iv) in case of misuse, the high likelihood of remaining
undetected by common drug screenings [5]. These reasons have
contributed to a fast spinning spiral of national banning of new
* Corresponding author.
E-mail address: t.beuerle@tu-bs.de (T. Beuerle).
existing SCs by slightly modied successor compounds by the
manufacturers.
Hence, at least 28 (including USA, Japan, UK, China and
Germany) countries have recently modied the common prac-
tice of individually banning synthetic cannabinoids by adopting
new laws (or modifying existing laws) towards a more generic
regulation of structurally related compound classes. These changes
will allow the control of new and so far unknown SCs based on
structural similarity to a dened SC subclass and a set of dened
structural modications [4]. The future will show how this legal
action is suitable to avoid the constant circumvention of laws,
through successive structural modications of the SCs.
The identication and quantication of SCs and their metab-
olites in biological matrices, including serum [6], blood [7], plasma
[8], urine [9], oral uid [10] and hair [11] is mainly performed by
chromatographic techniques coupled to mass spectrometric
detection, such as GCMS, LCMS/MS or LCHRMS.
Currently, urine is the most widely used matrix for the
detection of SCs in human specimen. The main advantages of
urine analysis over blood samples are the non-invasive sample

http://dx.doi.org/10.1016/j.forsciint.2017.09.011
0379-0738/ 2017 Elsevier B.V. All rights reserved.
26 N. Langer et al. / Forensic Science International 280 (2017) 2534
collection, usually higher concentrations of the analytes and a
longer detection window [12]. There are, however, some dis-
advantages with urine as sample for drug screenings, for instance
the inability to determine a very recent SC use, the possibility of
adulteration, the knowledge of the drugs metabolites and the need
of enzymatic or acidic hydrolysis of the conjugated hydroxylated
metabolites prior to analysis [1317]. Currently, little attention is
given to oral uid as another promising matrix for workplace and
roadside drug testing for SCs. Due to the facts that oral uid is fast
and easy to obtain, it is suitable for the identication of a recently
consumed drug without the invasive sample collection of blood
and the possibility to detect unchanged SCs instead of metabolites.
However, fast and simple workplace or roadside testing usually
requires immunological based test systems.
Immunoassays are cost-effective, high-throughput methods to
rapidly distinguish positive from negative specimens in an initial
screening. Hence, immunoassay based methods could represent a
simple, cheap and fast screening approach to reduce the number of
SC samples, which have to be conrmed by labor intensive
techniques in clinical, forensic or workplace screening programs.
So far, only a modest number of immunoassay based methods
are described in literature for the detection of SCs in urine [1824]
and so far only one for detection in oral uids [25]. The most
comprehensive study was undertaken by Castaneto et al. [20].
20,017 authentic urine specimens were screened with the drugs of
abuse V (biochip array DoA-V; Randox Laboratories Ltd., Crumlin,
UK), which combines three polyclonal antibodies against JWH-018,
one against JWH-250 and seven antibodies against other designer
drugs. The initial biochip screening was accompanied by LCMS/
MS conrmatory analyses of 1432 presumptive positive and
1069 negative samples. The DoA-V showed promising results of
high cross-reactivity (for 22 of 33 SCs and 37 of 42 metabolites) and
>85% sensitivity, specicity and efciency, respectively. Evalua-
tions of the Immunalysis K2 Spice homogeneous enzyme
immunoassay (HEIA) for identifying SCs in authentic urine samples
by Barnes et al. [19] and Kronstrand et al. [21] gave high sensitivity
(92/98%) and specicity (87/82%), but did not show cross reactivity
against a wide range of SC metabolites or more recent SCs.
Despite the advantage to detect parent SCs in oral uid instead
of a multitude of possible metabolites in urine, only one evaluation
of an ELISA method with a polyclonal antibody raised against
conjugated JWH-018 for the detection of SCs in oral uid was
performed. There, 21 out of 26 LCMS/MS conrmed SC positive
oral uid specimen were identied by ELISA with a cut-off
concentration of 0.25 ng/mL [25].
To establish an efcient and reliable immunoassay for SC
screenings, it is necessary to develop antibodies that show a high
degree of cross-reactivity, to ensure that a set of only a few
antibodies ideally can cover a wider range of the structural
modications across several SC subclasses. During the develop-
ment of such antibodies it is necessary to analyze the cross-
reactivity of newly developed anti-sera or antibodies against a
larger number of structurally divers SCs. These cross-reactivity
data of developed antibodies is most widely obtained by enzyme-
linked immunosorbent assays (ELISA). However, a number of
alternative biophysical techniques have emerged potentially
capable to analyze the binding of low-molecular weight com-
pounds to macromolecular targets, including X-ray crystallogra-
phy, ligand-observed NMR, isothermal titration calorimetry (ITC),
differential scanning calorimetry (DSC), differential scanning
uorimetry (DSF), microscale thermophoresis (MST), small-angle
X-ray scattering (SAXS) and surface plasmon resonance (SPR) [26].
SPR-based biosensors combine the benets of high-throughput
screening, high information content about the binding events and
the possibility to directly observe the antigen-antibody-binding
without the need of labeling or labeling reactions. In most SPR
applications biomolecules (e.g. antibodies) are bound to a gold
surface of a sensor chip. Binding interactions of added compounds
(e.g. antigens) are monitored in real-time by measuring the change
of refractive index using polarized light [27]. Although, SPR is
mainly used for measuring binding events of high molecular
weight macromolecules, it was recently demonstrated that SPR
can also be applied to monitor the binding of lower molecular
weight compounds [28,29].
To evaluate the suitability of SPR for the characterization of
antibody SC binding, we conducted a direct binding analysis of
22 SCs and monoclonal or newly raised polyclonal antibodies
based on surface plasmon resonance (SPR) to determine cross-
reactivity of this major group of new psychoactive substances. To
evaluate these ndings, the results were compared with data
obtained by competitive enzyme-linked immunosorbent assay
(ELISA) which was carried out in parallel.
2. Material and methods
2.1. Experimental
2.1.1. Chemicals
All chemicals were of analytical grade purity and were
purchased from Fluka (Buchs, Switzerland), SigmaAldrich (Stein-
heim, Germany) or Acros Organics (Geel, Belgium). Solvents were
of high-performance liquid chromatography grade purity and were
used without further purication. Monocrotaline (25), used as a
negative control in the immunoassays and terephthalaldehyde,
used as reference in quantitative 1H NMR analysis and 3,30 ,5,50 -
tetramethylbenzidine for ELISA detection (TMB, T8665, Ready-to-
Use Kit), were obtained from SigmaAldrich (Steinheim,
Germany). TLC plates (Polygram SIL G/UV 40  80 mm) and silica
gel (mesh 0.040.063) were obtained from Macherey-Nagel
(Dren, Germany).
0.1 M phosphate buffered saline (PBS, 81 mM Na2HPO4, 19 mM
NaH2PO4, 135 mM NaCl, 2.7 mM KCl, pH 7.4) was used as dialysis
buffer for the haptenprotein conjugates.
PBSEtOH (8.1 mM Na2HPO4, 1.9 mM NaH2PO4, 13.5 mM NaCl,
0.27 mM KCl, pH 7.4, modied with 10% ethanol (v/v) and 0.05%
polysorbate 20 (v/v)) was used in SPR experiments as running
buffer, sample dilution buffer and ELISA wash buffer, unless
otherwise mentioned.
HBS-EP+ (HEPES 10 mM, NaCl 150 mM, EDTA 3 mM, polysorbate
20 0.05% (v/v) at pH 7.4) was used in SPR as immobilization buffer.
Monoclonal anti-K2 antibody 1 (MAb1, 1.26 mg/mL) was
obtained from Arista Biologicals, Inc. (Allentown, PA) and
monoclonal anti-K2-antibody 2 (MAb2, 6.0 mg/mL) from CalBiore-
agents, Inc. (San Mateo, CA). Beside general information on
specicity against metabolites of JWH-018 and JWH-073 no further
information on SC-specicity was available for these commercially
available antibodies.
Peroxidase-conjugated antibodies for competitive ELISA were
anti-sheep IgG (SAB3700702), purchased from SigmaAldrich
(Steinheim, Germany) used for the polyclonal antisera (PAb1-3)
and anti-mouse IgG (610-103-121) purchased from Rockland Inc.
(Limerick, PA) used for both monoclonal antibodies (MAb1 and
MAb2).
Synthetic cannabinoids and SC-fragments (1-24) were previ-
ously synthesized by standard chemical procedures or isolated
from herbal smoking mixtures with subsequent silica gel column
purications. Structures and purities were conrmed by NMR [30
34].
2.1.2. Nuclear magnetic resonance spectroscopy
1
H and 13C NMR spectra of the haptens 1A3B were recorded at
600.1 and 150.9 MHz, respectively, on a Bruker Avance II
 N. Langer et al. / Forensic Science International 280 (2017) 2534 27

600 spectrometer with a TCI cryo probehead (Bruker BioSpin,

Rheinstetten, Germany). Compounds 1A and 1B were measured in
CDCl3, compounds 2A, 2B, 3A and 3B in [D6]dimethylsulfoxide
(DMSO-d6). Chemical shifts are reported in d units (ppm) and were
measured relative to tetramethylsilane (dH = 0.00 ppm) and to the
solvent, CDCl3 (dC = 77.01 ppm) and DMSO-d6 (dC = 39.50 ppm),
respectively. Assignment techniques used were DEPT-135 (dis-
tortionless enhancement by polarization transfer), H,H-COSY
(correlation spectroscopy), H,C-HSQC (heteronuclear single quan-
tum correlation) and H,C-HMBC (heteronuclear multiple-bond
correlation). Digital resolutions in the two-dimensional NMR
spectra were chosen small enough to permit the distinction of
cross-peaks with similar chemical shifts (if possible). The HSQC
and HMBC experiments were optimized for C,H-coupling con-
stants of 145 and 8 Hz, respectively. Iterative full-bandshape
analyses of the 1H NMR spectra of 2A and 3B were performed with
the program TopSpin 2.1pl5 (Bruker).The structures of the six
haptens were conrmed by fully assigning their 1H and 13C NMR
spectra. The NMR data and their assignments are given in the
Supplementary material.
The purities of the six haptens were determined by quantita-
tive 1H NMR analysis using terephthalaldehyde as the reference
(r) because its CHO signal does not interfere with the signals of
the haptens (h). A weighed amount mr of the reference was mixed
with a weighed amount mp of the product (p) to be analysed. The
molar ratio R = xh/xr of the hapten relative to the reference equals
the ratio of the 1H NMR integral (divided by the number nh of
associated protons), Ih/nh, of a signal or group of signals of the
hapten relative to the integral of the CHO signal of the reference
Fig. 1. Chemical structures of the six synthesized haptens (Hapten 1A3A were
(divided by 2), Ir/2. With xh = R xr the purity of the product in mass
used for immunization; Hapten 1B3B were used for antisera purication,
per cent equals xh Mh/mp where Mh is the molar mass of the respectively).
hapten.

2.1.4. Preparation of haptenprotein conjugates

2.1.3. Electrospray-mass spectrometry and electrospray-tandem-mass
spectrometry (ESI-MS and ESI-MS/MS) parameters
Two different systems were used. For general ESI-MS and ESI-
MS/MS characterization the haptens 1A,B3A,B (Fig. 1) were
dissolved in methanol at concentrations of approx. 10 mg/mL.
These solutions were directly infused into a 3200 QTrap mass
spectrometer (Applied Biosystem/SCIEX, Darmstadt, Germany)
using the integrated syringe pump of the 3200 QTrap instrument
(syringe: 1000 mL, i.d. 2.3 mm; Hamilton, Reno, NV, USA) at a ow
rate of 5 mL/min. The 3200 QTrap mass spectrometer was
equipped with an ESI interface (TurboV) ion source. The
instrument was operated in electrospray positive mode and
spectra were recorded in enhanced MS mode (EMS) and in
enhanced product ion mode (EPI). Ionization and EPI conditions
were: source voltage 5.5 kV, ambient temperature, curtain gas
10 psi, GS1 10 psi, GS2 0 psi, declustering potential 20 V, and
collision energies between 34 and 42 V. Nitrogen was used as the
curtain and auxiliary gas.
For high resolution electrospray mass spectrometry (ESI-HR-
MS) accurate mass measurements of the haptens 1A,B3A,B a LTQ-
Orbitrap Velos (ThermoFisher Scientic, Bremen, Germany) was
used. Electrospray measurements in positive mode were per-
formed in direct infusion mode using a custom-made microspray
device mounted on a Proxeon nanospray ion source. The micro-
spray device allowed sample infusion without sheath gas through a
stainless steel capillary (90 mm ID) at a ow rate of approximately
1 mL/min. Spray voltage was set between 2.0 and 2.3 kV. The
resolution of the Orbitrap mass analyzer was set to 100,000 FWHM
at m/z 400. Accurate mass measurements were performed using
the lock mass option of the instrument control software using the
cation of tetradecyltrimethylammonium bromide (256.29988 u) as
internal mass reference.
Haptens 1A, 2A, and 3A were covalently coupled to keyhole
limpet hemocyanin (KLH) as immunogens and 1B, 2B, and 3B to
bovine serum albumin (BSA) for purication of crude polyclonal
antisera using afnity chromatography. The coupling was per-
formed by standard EDC/NHS crosslinking [35]. In our case, the EDC-
activated carboxylic acids of the haptens reacted with N-hydrox-
ysuccinimide, which were subsequently bound to primary amines
of the desired proteins. The general NHS-ester activation procedure
is described below using 1AKLH-conjugate as an example.
2.1.5. Hapten 1AKLH-conjugate
Hapten 1A (1.9 mg, 4.3 mmol) was activated by incubation with
N-hydroxysuccinimide (3 eq., 2.8 mg, 12.9 mmol) and N-(3-dime-
thylaminopropyl)-N'-ethylcarbodiimide hydrochloride (EDC)
(1.2 eq., 1.0 mg, 5.2 mmol) in dimethylformamide (1 mL) and stirred
for 30 min. The resulting active ester was added slowly and
dropwise to a stirred solution of 20 mg KLH (2 mg/mL) in 0.1 M PBS
(pH 7.4) and stirring was continued at 4  C for 16 h. To separate the
conjugated product from uncoupled haptens, conjugates were
dialyzed using a Membra-Cel1 dialysis tubes ((MWCO 14 kDa)
Viskase, Darien, IL, USA) against 300 mL 0.1 M PBS at 4  C for 48 h
and a three time buffer exchange. Protein concentrations were
determined using Pierce BCA Protein Assay Kit (Thermo Fisher
Scientic, Rockford, IL). The molar density of 1A/KLH was
determined by spectrophotometric technique [36] using Varian
Cary 300 Bio (Varian, Darmstadt, Germany) at 320 nm and a 1A
calibration curve recorded at identical conditions. The nal
solution was stored at 4  C and used for immunization.
2.1.6. Sheep antisera
Polyclonal sheep sera raised against the KLH conjugates of 1A, 2A
and 3A were obtained from Seramun Diagnostica GmbH (Heidesee,
28 N. Langer et al. / Forensic Science International 280 (2017) 2534
Germany) or ProteoGenix (Schiltigheim, France) by standard
immunization protocols of the respective service provider.
Concentration of the nal polyclonal antibodies were:
Polyclonal antibody 1 (PAb1): 0.8 mg/mL.
Polyclonal antibody 2 (PAb2): 0.8 mg/mL.
Polyclonal antibody 3 (PAb3): 1.0 mg/mL.
2.1.7. Hapten synthesis
Haptens (1A,B3A,B) were synthesized by standard chemical
procedures. The chemical synthesis of the individual haptens are
detailed experimental procedures are given in the Supplementary
material. Intermediates were used without nal purication for
the next reaction step in the syntheses and nal products used for
immunization (haptens 1A,B3A,B) were puried and purities of
90% were veried by quantitative NMR.
The haptens used for immunization and antisera purication
are presented in Fig. 1.
2.1.8. Competitive ELISA methodology
96-at-bottom well plates (Nunc MaxiSorpTM, Thermo Fisher
Scientic, Rockford, IL) were coated with 100 mL/well of 3B-BSA-
conjugate, at a concentration of 0.3 mg/mL in 0.1 M PBS, pH 7.4,
incubated 1 h at RT and overnight at 4  C. The wells were washed
with 200 mL PBSEtOH each. Then, 50 mL/well of the antibody
dilutions of PAb1-3 or MAb1-2 (0.8 mg/mL) in PBSEtOH or PBS
EtOH (blank) were added. After the addition of SC 125 (200 mM,
50 mL/well) in PBSEtOH or PBSEtOH as control, the immuno-
reaction was allowed to proceed for 1 h at RT. The plates were
washed with PBSEtOH and 0.01 M PBS (pH 7.4) and incubated
with peroxidase-conjugated anti-sheep IgG for the polyclonal
antisera (0.15 mg/mL, 100 mL/well), peroxidase-conjugated anti-
mouse IgG (0.1 mg/mL, 100 mL/well) for the both monoclonal
antibodies or 0.01 M PBS as control. After another washing with
0.01 M PBS (pH 7.4), the colorimetric substrate TMB (50 mL/well,
Ready-to-Use Kit SigmaAldrich (Steinheim, Germany)) was added
and the plate was incubated for 1030 min. After stopping the color
development with 0.5 M sulfuric acid (50 mL/well), the optical
density was measured at 450 nm using Emax precision microplate
reader (Molecular Devices, Sunnyvale, CA). All incubations unless
otherwise mentioned were performed at RT under gentle rocking
and each washing step involved a three time buffer exchange
(200 mL/well).
2.1.9. SPR methodology
Experiments were performed using a Biacore X100 from GE
Healthcare (Uppsala, Sweden). Biosensor data produced were
evaluated with Biacore evaluation software 2.0. CM5 sensor chips,
the amine coupling kit containing 0.1 M N-hydroxysuccinimide
(NHS), 0.4 M 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
(EDC) and 1.0 M ethanolamine-HCl pH 8.5 and the maintenance
products were obtained from GE Healthcare (Uppsala, Sweden).
The regeneration solution mixtures of ethylene glycol/water (50/
50 wt%) pH 8.512.0, running buffer PBSEtOH and immobilization
buffer HBS-EP+ were prepared with ultrapure water and ltered
prior to use by a nylon 0.2 mm lter. The immobilization and all
following binding measurements were performed at 25  C.
Maintenance steps were performed to desorb and to sanitize
the tubings and the ow cells of the instrument further system
check and pump calibration to monitor and maintain system
suitability. Regular maintenance of Biacore X100 was performed
according to the manufacturers specications [37]. When running
maintenance the chips were stored in HBS-EP+ at 4  C.
2.1.10. Antibody immobilization to the CM5 sensor chip surface
Immobilization pH scouting for covalent coupling of the
antibodies to the CM5 sensor surface was carried out according
to Immobilization pH Scouting Wizard in the Biacore X100 control
software. Antibodies were diluted in 10 mM sodium acetate pH 4.0,
4.5, 5.0 and 5.5 to a nal concentration of 30 mg/mL in each sample.
For the polyclonal antisera a high immobilization level was
achieved at pH 4.5 and for the monoclonal antibodies at pH 5.0.
Therefore, these conditions were chosen for the immobilization.
The carboxymethylated dextran matrix of both ow cells were
activated by a mixture of equal volumes of 400 mM EDC and
100 mM NHS with a contact time of 7 min at a ow rate of 10 mL/
min with HBS-EP+ buffer as running buffer. 6 mL of antibody stock
solution were mixed with 194 mL 10 mM sodium acetate and were
injected over ow cell 2, until the target immobilization level was
reached or the instrument was running out of ligand at a ow rate
of 5 mL/min; Flow cell 1 was left blank for double-referencing the
sensorgrams. The remaining active sites on the dextran matrix for
both ow cells were blocked with 1 M ethanolamine-HCl solution
pH 8.5 for 7 min at a ow rate of 10 mL/min. Levels of
immobilization for all ve antibodies were the following:
7000 RU (PAb1), 7747.1 RU (PAb2), 9013.6 RU (PAb3), 7689.3 RU
(MAb1) and 6517.8 RU (MAb2).
2.1.11. Regeneration scouting and evaluation
Individual regeneration scoutings for all ve antibodies were
conducted right after immobilization according to the Regenera-
tion Scouting Wizard in the Biacore X100 Control Software under
the following conditions. 200 s contact time of JWH-018 solution
(200 mM), followed by one regeneration after each cycle with a 30 s
contact time. Four binding-regeneration cycles per condition have
been performed. To evaluate the regeneration conditions for long
measurement, JWH-018 (200 mM) was injected 20 times with
followed regeneration after each analyte binding. This evaluation
showed stable baseline and analyte response under the respective
regeneration conditions.
2.1.12. SPR assay conditions
2 mM stock solutions of synthetic cannabinoids (19, 1120,
2224), two synthetic cannabinoid fragments (10, 21) and control
(non-SC compound monocrotaline) (25) were prepared in ethanol
and diluted 10 fold in PBSEtOH before measurement. In the assay
set up the ow system was primed with ow buffer at the start of
the run followed by three start-up cycles to clear the chip surface. A
blind cycle was carried out after ve sample cycles, respectively.
Each sample was injected at a ow rate of 10 mL/min with a contact
time of 200 s. Report points were recorded right before injection
(baseline) and at the end of the association phase after 200 s
(binding response). The antibody surfaces were regenerated with a
30-s pulse of ethylene glycol/water (1:1) pH 8.5 to 11, specically
evaluated by the regeneration scouting for every antibody. For each
antibody an individual CM5 sensor chip was used.
For a rst scouting, SPR measurements were initially performed
as single point determinations and subsequently repeated in
triplicates for selected substances.
To allow a direct comparison of the SPR binding response with
competitive ELISA absorbance values, blank and negative control
(monocrotaline) were subtracted for both assay types and the
binding activities were expressed in relation to the best binding
substrate (100%), hence representing a relative binding response
(RBR), respectively. In the corresponding gures the obtained
readings for the control monocrotaline and blank are given as an
uncorrected value for comparison.
3. Results and discussion
Synthetic cannabinoids are small molecules (molecular weight
< 500 Da) and usually do not induce an independent immune
response by themselfs. To allow this, the SCs must be coupled to an
 N. Langer et al. / Forensic Science International 280 (2017) 2534
immunogenic protein, which is able to induce an immune reaction.
However, most SCs do not provide an appropriate functional group
for chemical coupling reactions, thus the SCs had to be structurally
modied to haptens containing a spacer including a terminal
functional group. The linkers of all three haptens had a length of 3
4 atoms with a terminal carboxylic acid, to allow the formation of
amid bonds with primary amino groups of alkaline amino acids of
the carrier protein of choice. The concept of facial recognition
developed by Matyas et al. [38] explains the connection between
the hapten design and specicity of gained antibodies. Through
hindrance at the linker site of the hapten mainly the opposed site is
exposed to the immune system and involved in the immune
reaction. For our study we developed three hapten structures for
immunization based on the core of JWH-018, one of the rst
synthetic cannabinoids in herbal smoking mixtures. The cross
linkers are attached at three different positions of the JWH-
018 molecule, offering the possibility to obtain antisera against
different structural features of many SCs in smoking mixtures. The
linkers were attached at position 5 of the indole core (1A), position
4 of the naphthoyl ring (2A) and at the indol nitrogen (3A) to, in
theory, receive antibodies against the opposite site of the
naphthoyl ring (1A), the alkylindole (2A) and against the 3-(1-
naphthoyl)indole, respectively. For our measurements we were
able to test 3 puried polyclonal antisera against SCs by
immunization of sheep with KLH-conjugates of the haptens 1A,
2A, 3A and two commercially available monoclonal antibodies
MAb1 and MAb2. The two commercial antibodies were anti-JWH-
018 and JWH-073 metabolite specic antibodies, but no further
cross-reactivity information was available.
The aim of our study was to evaluate the potential of a direct
SPR method for a fast and sensitive determination of cross-
reactivities against a group of structurally divers SCs using the set
of ve different antibodies/sera and to compare the results with
commonly used competitive ELISA results for the same set of
compounds.
Fig. 2 shows the full set of structures of the 22 SCs, 2 SC-
fragments and monocrotaline (control) which were used in both
assay types.
Initial experiments showed that regeneration was crucial for
the SPR binding studies and was one of the most important steps in
the development of a successful SPR assay. Regeneration is the
process of removing all analyte molecules from the ligand on the
surface, after each analysis cycle. The conditions should not be too
weak, which result in an incomplete removal of the analyte and a
non-specic decrease in analyte response over time. However,
conditions should not be too harsh either, since the ligand should
maintain its analyte-binding capacity, which would lead to a
decreased active-surface life time. Therefore, several SPR regener-
ation conditions scoutings were conducted, which can be divided
in acidic (10 mM glycine-HCl pH 1.5, pH 2.0 and pH 2.5), basic-
hydrophobic (ethanol/water (50:50 v/v) pH 12, ethylene glycol/
water (50:50 wt%) pH 9, 10, 10.5, 11, 11.5, 12) and hydrophobic
(ethanol/water (70:30 v/v)), ethylene glycol, ethylene glycol/water
(50:50 wt%) conditions. The best results were obtained with
ethylene glycol/water mixtures (50:50 wt%) at different pH values
for each individual antibody (MAb1 pH 11.0, MAb2 pH 10.5, PAb 1
pH 8.5, PAb 2 pH 8.5 and pH 9.0 for PAb 3).
Figs. 37 show the comparison of the SPR vs. ELISA results for
the two (commercial available) monoclonal anti-K2 antibodies
(MAb1 and MAb2) and the three polyclonal antisera (PAb1-3). The
relative binding responses (RBR) of each SC/-fragment is plotted as
blue bar for the SPR measurement and as red bar for the ELISA
measurement. To evaluate and compare the acquired antibodies
specicity data, we dened an existence of cross-reactivity for a
RBR  50%, no binding for a RBR <30% and comparability of these
two methods between an inter-assay variability of 25% RBR.
29
Negative RBR values for some SCs in the diagrams reect a
higher absorbance of the SCs than the PBSEtOH control in the
competitive ELISA or a lower binding response than the control
(monocrotaline) in the SPR assay.
Fig. 3 shows the comparison of the results for the rst
(commercial available) monoclonal anti-K2 antibody (MAb1). The
majority of the measured substances, 16 of 24 (66.7%), showed
similar results in both assay types. AM-1220 was the best binding
substrat in SPR and ELISA measurements. MAb1 demonstrated
cross-reactivity for 9 SCs in the SPR measurement and for 7 SCs in
the ELISA measurement. But only JWH-018, JWH-019, AM-1220,
AM-2201 and STS-135 showed reliable cross-reactivity in both
assays. These are three naphthoylindoles (AM-2201, JWH-018 and
JWH-019) with modied indole N-alkyl chains, one naphthoylin-
dole (AM-1220) with a cyclic amine instead of the alkyl chain and
one adamantoylindole (STS-135). However, the comparison
showed high accordance between both assays, in particular
consistency in both assay types was observed for compounds
with no cross-reactivity (e.g. JWH-081, JWH-203).
As expected, MAb1 showed good binding with the SCs JWH-
018 and its N1-hexyl homologue JWH-019 in the SPR and ELISA. But
the results for the N1-butyl homologue, JWH-073, are contradic-
tory, with highest values in the ELISA and low values in the SPR.
Additional cross-reactivity was shown for AM-2201, the 5-
uoropentyl homologue of JWH-018 and STS-135. However, best
binding responses in both assays were obtained with AM-1220,
which differs from JWH-018 by an N-methyl-piperidine-2-methyl
group instead of the pentyl side chain. These measured cross-
reactivities of MAb1 with pentyl, hexyl, 5-uoropentyl and N-
methyl-piperidine-2-methyl analogues suggested, that changes in
the alkyl side chain had less inuence for the antibodies specicity.
Furthermore, high cross-reactivity for AM-1220, JWH-018, JWH-
019 and STS-135 showed, that a 3-carbonylindole with an
unsubstituted naphthoyl moiety or an indole-3-carboxamide with
an adamantyl ring seems to be necessary for a sufcient cross-
reactivity.
A higher degree of deviation for the two test systems was
observed for the second monoclonal anti-K2 antibody (MAb2,
Fig. 7) with only 6 out of 24 SCs (25.0%) in comparable accordance.
Compared to MAb1, SPR measurements for MAb2 showed a
higher binding response for the control (monocrotaline), also
reected by the occurrence of negative blue bars in Fig. 4 and an
overall lower binding response for the best binding substrate
(JWH-203).
In addition, larger deviations were also observed in the ELISA
measurements (demonstrated by larger error bars of the red
columns in Fig. 4). Since MAb2 was a commercially available
product, we were not able to dene the underlying cause for these
ndings. But maybe a possible explanation could be the presence
of accompanying compounds besides the antibody in the
preparation and so consequently more unspecic bindings.
Nevertheless, 3 SCs (JWH-019, JWH-122 and JWH-210) showed
cross-reactivity in both assays, suggesting that modications at the
naphthoyl ring of the SCs have probably a reduced inuence on the
afnity of MAb2. This was demonstrated through JWH-122 and
JWH-210, which have an additional methyl or ethyl group at
position 4 of their naphthoyl ring.
Only general information about existing specicity of the
monoclonal antibodies MAb1 and MAb2 against the metabolites of
JWH-018 and JWH-073 were given by the manufacturer. The
evaluation of our results indicated a clear difference in the range of
detectable SCs with these two monoclonal antibodies.
The distinction of SC cross-reactivity obtained for MAb2 was
not nearly as pronounced as it was for MAb1. This was reected by
a higher non-specic binding, a general lower binding response in
the SPR and higher standard deviations in the ELISA. However, high
30 N. Langer et al. / Forensic Science International 280 (2017) 2534

Fig. 2. Structures of the 22 synthetic cannabinoids, 2 SC-fragments and monocrotaline used in both assays.

binding values of MAb2 were conrmed for JWH-019 and JWH-
203, but again also not for JWH-073. The existing cross-reactivity
with JWH-122 and JWH-210 suggests, that a small methyl or ethyl
substituent on the naphthalene ring is probably tolerated by this
monoclonal antibody. This is also shown by the possible cross-
reactivity with MAM-2201, with a RBR in the SPR just below the
50%-RBR-level (47%).
Comparison of the rst polyclonal antiserum against hapten 1A
(PAb1) demonstrated a good correlation between the SPR and
ELISA cross-reactivity data (Fig. 5), with comparable results for
 N. Langer et al. / Forensic Science International 280 (2017) 2534 31

Fig. 3. Comparison of relative binding response (RBR) for the rst monoclonal anti-K2 antibody (MAb1) against 22 SC and 2 SC-fragments determined by direct SPR (blue bar,
n = 1 or n = 3 indicated by error bars, SD) and competitive ELISA (red bar, n = 3, SD). SPR binding response: monocrotaline 1.02 RU, blank 6.02 RU. ELISA OD (450 nm):
monocrotaline 858, blank 703. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)

Fig. 4. Comparison of relative binding response (RBR) for the second monoclonal anti-K2 antibody (MAb2) against 22 SCs and 2 SC-fragments determined by direct SPR (blue
bar, n = 1 or n = 3 indicated by error bars, SD) and competitive ELISA (red bar, n = 3, SD). SPR binding response: monocrotaline 17.23 RU, blank 3.40 RU. ELISA OD (450 nm):
monocrotaline 727, blank 611. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)

Fig. 5. Comparison of relative binding response (RBR) for the polyclonal antiserum 1 (PAb1) against 22 SC and 2 SC-fragments determined by direct SPR (blue bar, n = 1 or n = 3
indicated by error bars, SD) and competitive ELISA (red bar, n = 3, SD). SPR binding response: monocrotaline 10.50 RU, blank 3.25 RU. ELISA OD (450 nm): monocrotaline
259, blank 237. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)
32 N. Langer et al. / Forensic Science International 280 (2017) 2534

Fig. 6. Comparison of relative binding response (RBR) for the polyclonal antiserum 2 (PAb2) against 22 SC and 2 SC-fragments determined by direct SPR (blue bar, n = 1 or n = 3
indicated by error bars, SD) and competitive ELISA (red bar, n = 3, SD). SPR binding response: monocrotaline 8.19 RU, blank 1.05 RU. ELISA OD (450 nm): monocrotaline 388,
blank 399. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)

10 of the 24 measured SCs (41.7%). For SPR JWH-018, JWH-210 and
XLR-11 were observed as best binding substrates. Cross-reactivity
was shown for 12 SCs in the ELISA assay and for 10 SCs in the SPR
assay. In comparison PAb1 showed cross-reactivity in both assays
for the SCs: JWH-018, JWH-073, JWH-081, JWH-210 and XLR-11.
The linker of hapten 1A is attached to position 5 of the indole ring.
Our assumption was, that these would led to a specic antisera
against the opposite naphthoyl ring and pentyl side chain. But
PAb1 shows cross-reactivity for SCs with different N-alkyl chain
lengths (JWH-018 and JWH-073), methoxy (JWH-081) or ethyl
substitution (JWH-210) at position 4 of their naphthoyl moiety and
a tetramethyl cyclopropyl moiety instead of the naphthalene ring
(XLR-11).
Fig. 6 shows a high deviation between the two methods for the
second polyclonal antiserum against hapten 2A (PAb2). Data of the
SPR assay showed a broad range of cross-reactivity for 12 SCs, but
just for 3 SCs in the ELISA. Four SCs (16.7%) were in good
accordance, with a lower inter-assay variability than 25% RBR.
AM-1220 alone showed consistently high afnities for the SPR and
ELISA measurements. Slight changes of the ELISA absorbance for
the SCs compared with the control, suggest that PAb2 is an
antiserum with an overall lower afnity. This is most clearly shown
by the highest absorbance reduction of the best binding substrate
(AM-1220) by only 24% compared with the control. However, it
was difcult to draw nal conclusions for the cross-reactivity of
this antiserum, because of the contradicting results for both
analytical methods.
The last tested polyclonal antiserum against Hapten 3A (PAb3)
again showed less deviations for the comparison of the direct SPR
and competitive ELISA binding activity (Fig. 7). 9 SCs (37.5%)
showed comparable results for both assays. Cross-reactivity was
shown for 4 SCs by the ELISA and for 6 SCs by the SPR measurement
(JWH-019, being the best binding substrate in SPR). The polyclonal
antiserum showed a similar pattern like the commercial MAb1,
with highest relative binding activities for STS-135, AM-1220,
JWH-018 respectively, suggesting, that MAb1 maybe was raised by
a similar hapten attached via the alkyl-indol side chain.
Fig. 8 summarizes the bandwidth of SCs which can theoretically
be detected with MAb1, MAb2, PAb1 and PAb3, on the basis of high
RBR in both assays. The results for the second polyclonal antiserum
PAb2 were not considered here, due to the strong discrepancy in
the ELISA and SPR results.
It was demonstrated, that 10 of the 24 measured SCs could be
detected by at least one of these four antibodies. As expected, most
of the antibodies are capable of binding specically the SC JWH-
018, because the polyclonal and most likely the monoclonal
antibodies are developed against haptens derived from the JWH-
018 structure. Eight of the cross-reacting SCs belong to the group of

Fig. 7. Comparison of relative binding response (RBR) for the polyclonal antiserum 3 (PAb3) against 22 SC and 2 SC-fragments determined by direct SPR (blue bar, n = 1 or n = 3
indicated by error bars, SD) and competitive ELISA (red bar, n = 3, SD). SPR binding response: monocrotaline 14.00 RU, blank 4.63 RU. ELISA OD (450 nm): monocrotaline
406, blank 505. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)
 N. Langer et al. / Forensic Science International 280 (2017) 2534
Fig. 8. SCs with high relative binding response in both assays for the antibodies MAb1,MAb2, PAb1, and PAb3. SCs in the overlapping areas had high binding responses for the
antibodies representing these areas. SCs on the right side had no sufcient binding activity for one of the four antibodies. Grey shaded areas indicate cross-reactivity in one of
the two assays for at least one antibody.
3-naphthoylindoles (AM-1220, AM-2201, JWH-018, JWH-019,
JWH-073, JWH-081, JWH-122 and JWH-210), one is an indole-3-
carboxamide with an adamantyl ring (STS-135) and another is a 3-
carbonylindole with a tetramethylcyclopropan residual (XLR-11).
The remaining 14 substances showed no sufcient binding with
at least one of the tested antibodies. Five of these SCs (5F-AB-
PINACA, 5F-AMB, AB-CHMINACA, AB-FUBINACA and MDMB-
CHMICA) belonged to the relatively new generation of synthetic
cannabinoids, based on indole-/indazole-3-carboxamides contain-
ing a valine-amide, valine-methyl ester or the methyl ester of the
tert-butyl homologue of valine as side chain. The lack of cross-
reactivity could be explained by the distinct structural changes
compared with previous synthetic cannabinoid generations, like
the replacement of the carbonyl group and aromatic ring by a
carboxamide group and residues structurally based on polar amino
acid derived structures. Further structural changes (also mainly of
polar nature) also resulted in loss of binding activity, for example a
cyanobutyl side chain (EAM-2201), a 2-chlorophenylacetyl (JWH-
203), a methoxybenzoyl moiety (RCS-4) and structures based on a
3-naphthoylpyrrole (JWH-307). These results revealed that ap-
proximately 60% of the tested substances would not be detectable
by a test system using combinations of these antibodies/sera.
According to the concept of facial recognition [38], one
objective was to generate polyclonal antisera with recognizable
differences in specicity through different linker positions of the
haptens. However, the polyclonal antisera PAb1 and PAb3 against
the KLH-conjugates of 1A and 3A resulted in no specicity changes,
which would have been expected in accordance to this concept.
This is probably due to the fact, that the target substances are too
small and therefore may not represent different epitopes for
different antibodies recognition.
4. Conclusions
The indirect measurement of the competitive ELISA method
depends on numerous factors: (i) a suitable coating antigen, that
has to be strongly bound by the antibody of investigation, (ii) the
competition of antibody binding sites between the analyte and the
coated antigen, (iii) the binding of the peroxidase labeled antibody
33
to the Fc region of the antibody of investigation and (iv) the
visualization through the peroxidase reaction. The main advan-
tages of the SPR method are the label-free and direct measurement
of the ligand-analyte binding. Furthermore, SPR measurements
allowed a deeper insight in the binding event of different synthetic
cannabinoids with the tested antibodies. Kinetic characterization,
like the determination of the association rate constant that reects
the speed of antigen-antibody complex formation or of the
dissociation rate constant, that allow conclusions on the strength
of an antigen-antibody interaction, could be the basis for future
developments in this area. Such kinetic data of developed
antibodies against SC structures would allow lead optimization
of the used immunogens. In the case of this SC binding study, it was
necessary to scout for optimal antibody binding and chip
regeneration conditions for each antibody/antisera. However,
these procedures were highly automated by the available software
and could be obtained by programmed sequences which were
processed continously unattended. It could be demonstrated that
SPR is a sensitive and highly automated analytical technique,
which was able to generate specicity data for newly developed
antibodies/sera quickly for a large group of new psychoactive
substances.
The comparison of the cross-reactivity for both assays showed
similar results for most of the synthetic cannabinoids.
The results showed that two of the tested antibodies were able
to detect ve SCs with sufcient sensitivity (MAB1 and PAB1).
Furthermore, there was no difference between the detection range
provided by monoclonal and polyclonal antibodies. The assump-
tion to achieve a higher range of detection of SCs by the higher
variability of polyclonal sera could not be conrmed. The
consequence of the fast changing SC market and the inability to
cross-react with newer SC analogues would mean to develop new
antibodies against different structures in very short intervals. Due
to the fact, that the classical antibody development is time-
consuming, it is difcult for manufactures of immunoassays to
keep up with the rapidly changing market. Using SPR we were able
to generate specicity data for newly developed antibodies quickly
against a large group of new psychoactive substances in a highly
automated way.
34 N. Langer et al. / Forensic Science International 280 (2017) 2534
Acknowledgments
This work was supported by Zentrales Innovationsprogramm
Mittelstand (ZIM) des Bundesministeriums fr Wirtschaft und
Energie (BMWI)FKZ: KF31565015K3. The authors express their
gratitude to Dr. Luma Baydoun and the Fraunhofer ITEM for
providing the Biacore X100 instrument used in this study and GE
Healthcare Life Sciences for excellent technical support.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.
forsciint.2017.09.011.
References
[1] V. Auwrter, S. Dresen, W. Weinmann, M. Mller, M. Ptz, N. Ferreirs, Spice
and other herbal blends: harmless incense or cannabinoid designer drugs? J.
Mass Spectrom. 44 (2009) 832837.
[2] United Nations Ofce on Drugs and Crime (UNODC), Global SMART Update
2015, Volume 13, March 2015, https://www.unodc.org/documents/scientic/
Global_SMART_Update_13_web.pdf. (Accessed 29 August 2017).
[3] United Nations Ofce on Drugs and Crime (UNODC), World Drug Report 2015,
June 2015, https://www.unodc.org/documents/wdr2015/World_Drug_Re-
port_2015.pdf. (Accessed 29 August 2017).
[4] United Nations Ofce on Drugs and Crime (UNODC), Early Warning Advisory
(EWA) on New Psychoactive Substances (NPS), https://www.unodc.org/LSS/
NpsFinding/List?pageNumber=0. (Accessed 14 May 2017).
[5] G. Loefer, E. Delaney, M. Hann, International trends in spice use: prevalence,
motivation for use, relationship to other substances, and perception of use and
safety for synthetic cannabinoids, Brain Res. Bull. 126 (2016) 828.
[6] M.A. Neukamm, T.E. Mrdter, C. Knabbe, H.-D. Wehner, F. Wehner, Quantita-
tive detection of the active spice ingredient JWH-018 in serum by means of
liquid chromatographytandem mass spectrometry, Blutalkohol 46 (2009)
373379.
[7] S.L. Kacinko, A. Xu, J.W. Homan, M.M. McMullin, D.M. Warrington, B.K. Logan,
Development and validation of a liquid chromatographytandem mass
spectrometry method for the identication and quantication of JWH-018,
JWH-073, JWH-019, and JWH-250 in human whole blood, J. Anal. Toxicol. 35
(2011) 386393.
[8] L.J. Schep, R.J. Slaughter, S. Hudson, R. Place, M. Watts, Delayed seizure-like
activity following analytically conrmed use of previously unreported
synthetic cannabinoid analogues, Hum. Exp. Toxicol. 34 (2015) 557560.
[9] T. Sobolevsky, I. Prasolov, G. Rodchenkov, Detection of JWH-018 metabolites in
smoking mixture post-administration urine, Forensic Sci. Int. 200 (2010) 141
147.
[10] C. Coulter, M. Garnier, C. Moore, Synthetic cannabinoids in oral uid, J. Anal.
Toxicol. 35 (2011) 424430.
[11] M. Hutter, S. Kneisel, V. Auwrter, M.A. Neukamm, Determination of
22 synthetic cannabinoids in human hair by liquid chromatographytandem
mass spectrometry, J. Chromatogr. B 903 (2012) 95101.
[12] S. Hegstad, A.A. Westin, O. Spigset, Detection times of carboxylic acid
metabolites of the synthetic cannabinoids JWH-018 and JWH-073 in human
urine, J. Anal. Toxicol. 39 (2015) 280286.
[13] A. Grigoryev, A. Melnik, S. Savchuk, A. Simonov, V. Rozhanets, Gas and liquid
chromatographymass spectrometry studies on the metabolism of the
synthetic phenylacetylindole cannabimimetic JWH-250, the psychoactive
component of smoking mixtures, J. Chromatogr. B 879 (2011) 25192526.
[14] C.L. Moran, V.H. Le, K.C. Chimalakonda, A.L. Smedley, F.D. Lackey, S.N. Owen, P.
D. Kennedy, G.W. Endres, F.L. Ciske, J.B. Kramer, A.M. Kornilov, L.D. Bratton, P.J.
Dobrowolski, W.D. Wessinger, W.E. Fantegrossi, P.L. Prather, L.P. James, A.
Radominska-Pandya, J.H. Moran, Quantitative measurement of JWH-018 and
JWH-073 metabolites excreted in human urine, Anal. Chem. 83 (2011) 4228
4236.
[15] T. Sobolevsky, I. Prasolov, G. Rodchenkov, Detection of JWH-018 metabolites in
smoking mixture post-administration urine, Forensic Sci. Int. 200 (2010) 141
147.
[16] K.E. Moeller, K.C. Lee, J.C. Kissack, Urine drug screening: practical guide for
clinicians, Mayo Clin. Proc. 83 (2008) 6676.
[17] M. Hutter, S. Broecker, S. Kneisel, V. Auwrter, Identication of the major
urinary metabolites in man of seven synthetic cannabinoids of the amino-
alkylindole type present as adulterants in herbal mixtures using LCMS/MS
techniques, J. Mass Spectrom. 47 (2012) 5465.
[18] A. Arntson, B. Ofsa, D. Lancaster, J.R. Simon, M. McMullin, B. Logan, Validation
of a novel immunoassay for the detection of synthetic cannabinoids and
metabolites in urine specimens, J. Anal. Toxicol. 37 (2013) 284290.
[19] A.J. Barnes, S. Young, E. Spinelli, T.M. Martin, K.L. Klette, M.A. Huestis,
Evaluation of a homogenous enzyme immunoassay for the detection of
synthetic cannabinoids in urine, Forensic Sci. Int. 241 (2014) 2734.
[20] M.S. Castaneto, N.A. Desrosiers, K. Ellefsen, S. Anizan, T.M. Martin, K.L. Klette,
M.A. Huestis, Method validation of the biochip array technology for synthetic
cannabinoids detection in urine, Bioanalysis 6 (2014) 29192930.
[21] R. Kronstrand, L. Brinkhagen, C. Birath-Karlsson, M. Roman, M. Josefsson, LC
QTOFMS as a superior strategy to immunoassay for the comprehensive
analysis of synthetic cannabinoids in urine, Fresenius J. Anal. Chem. 406 (2014)
35993609.
[22] M. Jang, W. Yang, H. Choi, H. Chang, S. Lee, E. Kim, H. Chung, Monitoring of
urinary metabolites of JWH-018 and JWH-073 in legal cases, Forensic Sci. Int.
231 (2013) 1319.
[23] A.L. Mohr, B. Ofsa, A.M. Keil, J.R. Simon, M. McMullin, B.K. Logan, Enzyme-
linked immunosorbent assay (ELISA) for the detection of use of the synthetic
cannabinoid agonists UR-144 and XLR-11 in human urine, J. Anal. Toxicol. 38
(2014) 427431.
[24] E. Spinelli, A.J. Barnes, S. Young, M.S. Castaneto, T.M. Martin, K.L. Klette, M.A.
Huestis, Performance characteristics of an ELISA screening assay for urinary
synthetic cannabinoids, Drug Test. Anal. 7 (2015) 467474.
[25] W.C. Rodrigues, P. Catbagan, S. Rana, G. Wang, C. Moore, Detection of synthetic
cannabinoids in oral uid using ELISA and LCMSMS, J. Anal. Toxicol. 37 (2013)
526533.
[26] J.-P. Renaud, C.-W. Chung, U.H. Danielson, U. Egner, M. Hennig, R.E. Hubbard, H.
Nar, Biophysics in drug discovery: impact, challenges and opportunities, Nat.
Rev. Drug Discov. 15 (2016) 679698.
[27] M. Couture, S.S. Zhao, J.-F. Masson, Modern surface plasmon resonance for
bioanalytics and biophysics, Phys. Chem. Chem. Phys. 15 (2013) 1119011216.
[28] S.V. Surwase, S.A. Patil, S. Srinivas, J.P. Jadhav, Interaction of small molecules
with fungal laccase: a surface plasmon resonance based study, Enzyme
Microb. Technol. 82 (2016) 110114.
[29] B.J. Yakes, K.M. Kanyuck, S.L. DeGrasse, First report of a direct surface plasmon
resonance immunosensor for a small molecule seafood toxin, Anal. Chem. 86
(2014) 92519255.
[30] N. Langer, R. Lindigkeit, H.M. Schiebel, U. Papke, L. Ernst, T. Beuerle,
Identication and quantication of synthetic cannabinoids in spice-like
herbal mixtures: update of the German situation for the spring of 2015,
Forensic Toxicol. 34 (2016) 94107.
[31] N. Langer, R. Lindigkeit, H.M. Schiebel, L. Ernst, T. Beuerle, Identication and
quantication of synthetic cannabinoids in spice-like herbal mixtures: a
snapshot of the German situation in the autumn of 2012, Drug Test. Anal. 6
(2014) 5971.
[32] L. Ernst, K. Krger, R. Lindigkeit, H.M. Schiebel, T. Beuerle, Synthetic cannabinoids
in spice-like herbal blends: rst appearance of JWH-307 and recurrence of
JWH-018 on the German market, Forensic Sci. Int. 222 (2012) 216222.
[33] K. Simolka, R. Lindigkeit, H.M. Schiebel, U. Papke, L. Ernst, T. Beuerle, Analysis
of synthetic cannabinoids in spice-like herbal highs: snapshot of the German
market in summer 2011, Anal. Bioanal. Chem. 404 (2012) 157171.
[34] L. Ernst, H.M. Schiebel, C. Theuring, R. Lindigkeit, T. Beuerle, Identication and
characterization of JWH-122 used as new ingredient in spice-like herbal
incenses, Forensic Sci. Int. 208 (2011) e31e35.
[35] G.T. Hermanson, Bioconjugate Techniques, 2nd ed., Elsevier Academic Press,
Amsterdam, 2008.
[36] B.F. Erlanger, F. Borek, S.M. Beiser, S. Lieberman, Steroidprotein conjugates. I.
Preparation and characterization of conjugates of bovine serum albumin with
testosterone and with cortisone, J. Biol. Chem. 228 (1957) 713727.
[37] GE Healthcare, Biacore X100: Handbook, http://www.iqm.csic.es/wp-content/
uploads/2013/equipamiento/spr/BR-1008-10AB%20Biacore%20x100%
20Handbook%20(Web).pdf. (Accessed 29 August 2017).
[38] G.R. Matyas, K.C. Rice, K. Cheng, F. Li, J.F.G. Antoline, M.R. Iyer, A.E. Jacobson, A.
V. Mayorov, Z. Beck, O.B. Torres, C.R. Alving, Facial recognition of heroin vaccine
opiates: type 1 cross-reactivities of antibodies induced by hydrolytically stable
haptenic surrogates of heroin, 6-acetylmorphine, and morphine, Vaccine 32
(2014) 14731479.