Synthesis of a β-CCT-lanthanide conjugate for binding the dopamine transporter

Article abstract of DOI:10.1039/c4ob02165g

The development of a β-CCT-lanthanide conjugate that binds the dopamine transporter (DAT) with high affinity (K_d = 303 nM) is described. Contrast agents such as the one described herein could be used as molecular probes to directly study the binding of small molecules to receptors such as DAT via MRI, PET or SPECT. This journal is

Full text of DOI:10.1039/c4ob02165g

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Synthesis of a β-CCT-lanthanide conjugate for
binding the dopamine transporter†
Cite this: Org. Biomol. Chem., 2015,
3, 2537
1
a
a
b
a
Gregory R. Naumiec, Grace Lincourt, Jeremy P. Clever, Michael A. McGregor,
Received 9th October 2014,
Accepted 7th January 2015
b,c
a
Abraham Kovoor and Brenton DeBoef*
DOI: 10.1039/c4ob02165g
www.rsc.org/obc
The development of a β-CCT-lanthanide conjugate that binds the mitters in the living brain. The current state-of-the-art method
1
23
dopamine transporter (DAT) with high affinity (K
d
= 303 nM) is uses radiolabeled ligands such as TRODAT-1™ and [ I]IBZM
described. Contrast agents such as the one described herein could to bind to DAT and dopamine receptors. The radiochemical
1,2
be used as molecular probes to directly study the binding of small signal from these agents is then detected by PET or SPECT.
molecules to receptors such as DAT via MRI, PET or SPECT.
Herein we report our initial efforts towards the development of
a novel probe for studying the DAT in vitro that does not rely
on radiolabels.
Dopamine, because of its role in reinforcement learning, is
one of the most significant neurotransmitters in the human
brain. Disrupted levels of this biogenic amine have been
linked to many psychological disorders and neurodegenerative
diseases, such as depression, schizophrenia, and Parkinson’s
In essence, our hypothesis was to develop a molecular
probe by conjoining a selective DAT ligand to a lanthanide ion,
namely gadolinium(III). Lanthanide ions are often complexed
by cyclen-derived ligands such as DO3A (1,4,7,10-tetraazacyclo-
dodecane-1,4,7- triacetic acid). The resulting complexes are
stable and may be used biomedical applications. For example,
gadolinium(III)-DO3A complexes are used as contrast agents
1
disease. Additionally, amphetamine stimulants cause the
release of increased amounts of dopamine, and addictive
drugs such as cocaine compete with dopamine for binding
sites on the dopamine transporter (DAT), thus preventing
reuptake of the neurotransmitter. Molecular probes that could
bind and/or image DAT could be used to quantify dopamin-
ergic neurons and drug binding, thus allowing for better diag-
nostics and treatments of neurologic diseases.
3,4
for magnetic resonance imaging. The cyclen-derived ligands
serve as a means to prevent heavy metal toxicity in patients.
Additionally, near infrared probes are often synthesized by
ligating Eu(III), Dy(III) or Tb(III) ions to the DO3A moiety.
Similar lanthanide-DO3A complexes can also be used as mole-
cular probes that can be detected by chemical exchange satur-
ation transfer (CEST) NMR. DO3A is also one of the most
common ligands for the radionuclides gallium-67 and indium-
Herein, we describe the synthesis of a molecular probe that
binds to the dopamine receptor. The Gd(III)-complex of this
probe can be used for in vitro binding studies without the
hassle and danger associated with radiolabeled molecular
probes. However, the probe could be easily used to chelate a
transition metal radioisotope for subsequent in vivo studies
employing either positron emission tomography (PET) or
single photon emission computed tomography (SPECT).
The role of DAT in neurobiology has been extensively
studied using histological techniques in post-mortem brain
111, which are increasingly popular nuclei for molecular
imaging due to their long half-life (3.3 and 2.8 days, respect-
ively) and their ready availability in North America.
The relaxation rate of a given Gd(III)-containing contrast
agent, as described by the Solomon–Bloembergen–Morgan
theory, may be modulated by the rate of molecular rotation,
5
τ . In general, the slower a molecule rotates in solution, the
R
1
tissues. Recently, imaging technology has augmented this
more pronounced its relaxation rate will be. Thus, the binding
of a relatively small contrast agent to a macromolecular target
is characterized by an increase in observed spin-latice relax-
work by allowing researchers to study the role of neurotrans-
6
,7
a
ation rate, r1.
Department of Chemistry, University of Rhode Island, 51 Lower College Road,
Kingston, Rhode Island 02881, USA. E-mail: bdeboef@chm.uri.edu;
Fax: +1-401-874-5072; Tel: +1-401-874-9480
We hypothesized that a CA that selectively binds DAT could
be developed by conjugating cocaine analogs to ligated Gd(III)
ions. These single molecule contrast agents should produce a
quantifiable relaxation rate increase via τ -modulation upon
R
binding to the transmembrane protein DAT. Cocaine and
many of its synthetic analogs have proven abilities to success-
b
Kovogen, LLC, 168 Long Wharf Drive, Mystic, Connecticut 06355, USA
c
Department of Biomedical and Pharmaceutical Sciences, University of Rhode Island,
41 Lower College Road, Kingston, Rhode Island 02881, USA
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4ob02165g
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 2537–2540 | 2537

View Article Online
Communication
Organic & Biomolecular Chemistry
addition of catalytic or stoichiometric amounts of Cu(I) salts
such as CuBr and CuCN. The most important factors for con-
trolling the regiochemistry of the addition were solvent and
temperature. The origin of the selectivity favoring the 1,4-
addition even in the absence of Cu(I) salts is likely the result of
an intermediate in which the amine of 3 coordinates with the
Grignard reagent’s cation to direct the addition to the
9
a
β-carbon of the α,β-unsaturated ester (4).
The coupling of the carboxylic acid derivative of 5 with
trityl-protected 2-amino-ethanethiol failed to proceed to com-
pletion, even when strong coupling agents like PyBOP (benzo-
triazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate)
or HATU (1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo
[4,5-b]pyridinium 3-oxid hexafluorophosphate) were used
(
Scheme 1). However, the starting material (5) was easily recov-
ered and resubmitted to the reaction conditions, eventually
leading to usable quantities of 6, following deprotection of the
trityl group. Finally, conjugation of the thiol to a maleimide-
containing DO3A ligand, deprotection of the carboxylic acids
and insertion of Gd(III) provided the desired CA (1).
Fig. 1 Common ligands for DAT and two proposed dynamic contrast
agents (CAs).
The arecoline-derived CA (2) was synthesized in an analo-
gous manner (Scheme 2). As observed for the synthesis of the
fully cross the blood–brain–barrier (BBB) and to bind DAT. phenylptrophane, 5, the addition of a phenyl Grignard reagent
Specifically, the tropane (−)-2β-carbomethoxy-3β-(4-chloro- to the α,β-unsaturated ester of arecoline resulted in 1,4-addition,
phenyl)tropane (β-CCT, also known as RTI-31) and the areco- presumably via a cyclic intermediate (7). The installation of the
line derivative shown in Fig. 1 selectively bind the dopamine 2-amino-ethanethiol linker and subsequent conjugation with
transporter 100× and 30× better than natural (−)-cocaine, the maleimide-containing DO3A ligand, followed by deprotec-
8
respectively. We planned to take advantage of this pheno- tion and metalation with Gd(III), provided the CA (2).
menon to selectively direct a Gd(III) ion to the DAT using
To analyze the ability of 1 and 2 to enhance their observed
contrast agents 1 and 2 (Fig. 1).
relaxation rate upon binding to DAT, we stably transfected
The synthesis of the prototype CAs, 1 and 2, was accom- HEK293 cells with a plasmid vector driving expression of the
plished via addition of Grignard nucleophiles to the α,β-un- DAT, and we quantified their ability to bind cocaine analogues
saturated methyl esters of ecgonidine (3) and arecoline via orthodox radioligand binding studies (Fig. 2A and B). We
9
(
Schemes 1 and 2). We hypothesized that the regioselectivity then used these cells for relaxometry and binding studies.
of the 1,4-conjugate addition of the para-chlorophenyl
Solutions of CAs 1 and 2 in PBS buffer (95 : 5 D O–H O)
2 2
Grignard reagent could be enhanced by addition of catalytic or were placed in NMR tubes, and inversion-recovery experiments
stoichiometric Cu(I) salts (3 → 5). Interestingly, we found no were performed in a 7 T (300 MHz) NMR spectrometer. Longi-
significant improvement in the yield of the reaction upon the tudinal relaxation times (T ) were analyzed as a function of
1
Scheme 1 Synthesis of β-CCT-Derived CA, 1. (TBME = tert-butyl methyl ether, EDC = 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide, HOBT =
hydroxybenzotriazole, DIPEA = diisopropyl ethylamine, TFA = trifluoroacetic acid.
2538 | Org. Biomol. Chem., 2015, 13, 2537–2540
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Communication
−
1
s−1, and this value
2
5 °C in PBS buffer was 6.32 mM
increases by 38% (r1obs = 8.71 mM− s−1) upon binding to the
DAT-transfected HEK293 cells. Relaxivity amplification result-
ing from non-specific binding of 1 to non-transfected HEK293
cells was not observed. No significant change in relaxivity was
observed with the arecoline-derived probe 2, which is indica-
tive of a failure to bind the DAT.
The relaxation data in the presence and absence of DAT
constitutes straightforward dose–response experiment,
1
10
a
which can be used to measure the affinity of the CA for the
protein (Fig. 2C). From this data, we calculated the binding
constant for 1, K
that of natural cocaine.
d
, as 303 ± 138 nM, which is comparable to
1
1
It is unlikely that a molecular probe such as 1 could be
used for in vivo studies of DAT via MRI due to the inherent
insensitivity of magnetic resonance techniques and the nano-
molar conentrations DAT in vivo. The indium-111 or copper-64
derivatives of 1, however, of these compounds could be easily
prepared via an analogous method and employed for in vivo
SPECT and PET imaging studies, respectively. While large
polar molecules like 1 often do not cross the BBB, the tran-
sition metal-tropane conjugate, TRODAT-1™ (see Fig. 1 for
structure), does enter the brain in sufficient quantities that
enable SPECT imaging. This provides precedent that neutrally
charged, metal-containing cocaine analogs can be transported
across the BBB, presumably via a passive diffusion mechan-
ism. However, in the event that a probe similar to 1 does not
cross the BBB, it could still be used for in vitro imaging or for
in vivo imaging studies with knockout mice that have per-
Scheme 2 Synthesis of Arecoline–Derived CA, 2.
1
2
meable BBBs or with healthy animals in combination with
1
3
high-intensity focused ultrasound. Additionally, numerous
methods for transporting both drugs and imaging agents
across the BBB are currently being developed and could be
employed to deliver 1, or variants bearing radionuclei, to the
1
4,15
dopaminergic regions of the brain.
In conclusion, we have synthesized a β-CCT-DO3A conju-
gate, 1, that bound the DAT with an affinity comparable to
that of cocaine. The affordable and readily synthesized areco-
line-derived probe, 2, failed to bind the DAT. When ligated to
Gd(III), 1 behaved as a dynamic contrast agent, whose longi-
tudinal relaxation rate increased dramatically upon binding
R
the DAT due to a slowing of the rate of molecular rotation, τ .
While indium-111 and copper-64 derivatives of 2 may be suit-
able for in vivo SPECT and PET imaging, the CA described
Fig. 2 (A) Immufluorescence of HEK293 cell line stably expressing DAT
or (B) untransfected HEK293 cells stained with an anti-DAT primary anti- herein is suitable for in vitro studies of the DAT.
body and FITC labeled secondary. (C) Binding of 1 to DAT as measured
The authors thank Kevin Clark for editorial and compu-
tational assistance. This work was funded by the University of
Rhode Island (URI) and the URI Foundation.
by relaxometry. ΔR1 = R1obs − R1free, where R1free was measured before
DAT was added. (Green lines show K ). 250 000 DAT-transfected cells
0.55 nmol DAT) were added as a solution in 50 uL of buffer.
d
(
probe concentrations in the presence and absence of the DAT-
transfected HEK293 cells. Sufficient quantities of the DAT-cells
were used to saturate the CAs. The dilution of the protonated
water with deuterated water allowed for a lengthening of the
observed relaxation time, thus facilitating the detection. The
observed spin-lattice relaxivity, r1free, of 2 at 300 MHz and
References
1 (a) M. L. Trudell, S. Izenwasser and B. Wang, Dopamine
Transporters: Chemistry, Biology, and Pharmacology, Wiley,
Hoboken, New Jersey, 1st edn, 2008; (b) P. Cumming,
Imaging Dopamine, Cambridge University Press, New York,
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 2537–2540 | 2539

View Article Online
Communication
Organic & Biomolecular Chemistry
1
st edn, 2009; (c) A. Sidhu, M. Laruelle and P. Vernier,
Chem., Int. Ed., 2001, 40, 2903–2906; (c) P. L. Anelli,
I. Bertini, M. Fragai, L. Lattuada, C. Luchinat and G. Parigi,
Eur. J. Inorg. Chem., 2000, 2000, 625–630.
Dopamine Receptors and Transporters: Function, Imaging and
Clinical Implication (Neurological Disease and Therapy),
Informa Healthcare, 2nd edn, 2003; (d) P. Seeman, Pharma-
col. Rev., 1980, 32, 229–313.
For selected examples of radioligands targeting the DAT,
see: (a) J. S. Stehouwer, L. M. Daniel, P. Chen, R. J. Voll,
L. Williams, S. J. Plott, J. R. Votaw, M. J. Owens, L. Howell
and M. M. Goodman, J. Med. Chem., 2010, 53, 5549–5557;
8 (a) P. A. Petukhov, J. Zhang, A. P. Kozikowski, C. Z. Wang,
Y. P. Ye, K. M. Johnson and S. R. Tella, J. Med. Chem., 2002,
45, 3161–3170; (b) A. P. Tamiz, J. Zhang, J. L. Flippen-
Anderson, M. Zhang, K. M. Johnson, O. Deschaux, S. Tella
and A. P. Kozikowski, J. Med. Chem., 2000, 43, 1215–1222;
(c) A. P. Kozikowski, G. L. Araldi, J. Boja, W. M. Meil,
K. M. Johnson, J. L. Flippen-Anderson, C. George and
E. Saiah, J. Med. Chem., 1998, 41, 1962–1969.
2
3
4
(
b) S. K. Meegalla, K. Plössl, M.-P. Kung, S. Chumpradit,
D. A. Stevenson, S. A. Kushner, W. T. McElgin, P. D. Mozley
and H. F. Kung, J. Med. Chem., 1997, 40, 9–17;
9 (a) B. I. Mobele, T. Kinahan, L. G. Ulysse, S. V. Gagnier,
M. D. Ironside, G. S. Knox and F. Mohammadi, Org. Process
Res. Dev., 2006, 10, 914–920; (b) M.-F. Zou, T. Kopajtic,
J. L. Katz, S. Wirtz, J. B. Justice Jr. and A. H. Newma, J. Med.
Chem., 2001, 44, 4453–4461; (c) J. W. Boja, F. Ivy Carroll,
M. Abdur Rahman, A. Philip, A. H. Lewin and M. J. Kuhar,
Eur. J. Pharmacol., Mol. Pharmacol. Sect., 1990, 184, 329–332.
(
c) D. R. Elmaleh, A. J. Fischman, T. M. Shoup, C. Byon,
R. N. Hanson, A. Y. Liang, P. C. Meltzer and B. K. Madras,
J. Nucl. Med., 1996, 37, 1197–1202.
For selected recent reviews, see: (a) S. S. Kelkar and
T. M. Reineke, Bioconjugate Chem., 2011, 22, 1879–1903;
(
b) J. L. Major and T. J. Meade, Acc. Chem. Res., 2009, 42,
8
93–903; (c) D. Le, L. M. on-Rodriguez, A. J. M. Lubag, 10 See supporting information for r1 plots and control exper-
C. R. Malloy, G. V. Martinez, R. J. Gillies and A. D. Sherry,
Acc. Chem. Res., 2009, 42, 948–957; (d) I. Rodríguez,
S. Pérez-Rial, J. González-Jimenez, J. Pérez-Sánchez,
F. Herranz, N. Beckmann and J. Ruíz-Cabello, J. Pharm.
Sci., 2008, 97, 3637–3665; (e) T. J. Meade, A. K. Taylor and
S. R. Bull, Curr. Opin. Neurobiol., 2003, 13, 597–602.
iment data. It should be noted that due to the use of a
mixed protonated and deuterated solvent, the observed
relaxation rates should not be directly compared to relax-
ation rates that are commonly reported for CAs in water,
saline or plasma.
11 R. P. Hammer, The neurobiology of cocaine: cellular and
molecular mechanisms; CRC Press, 1995.
Only two examples of a contrast agent targeting a neuro-
receptor in the dopaminergic system have been published: 12 A. H. Schinkel, J. J. M. Smit, O. van Tellingen, J. H. Beijnen,
(
a) I. Zigelboim, D. Offen, E. Melamed, H. Panet, M. Rehavi
and Y. Cohen, J. Inclusion Phenom. Macrocyclic Chem., 2007,
9, 323–329; (b) I. Zigelboim, A. Weissberg and Y. Cohen, J.
E. Wagenaar, L. van Deetmer, C. A. A. M. Mol, M. A. van
der Valk, B. C. Robanus-Maandag, H. P. J. te Tiele,
A. J. M. Berns and P. Borst, Cell, 1994, 77, 491–502.
13 N. McDannold, N. Vykhodtseva and K. Hynynen, Ultra-
sound Med. Biol., 2008, 34, 834–840.
5
Org. Chem., 2013, 78, 7001–7012.
N. Bloembergen and L. O. Morgan, J. Chem. Phys., 1961, 34,
5
6
7
8
42.
14 For selected recent reviews, see: (a) M. Malakoutikhah,
M. Teixidó and E. Giralt, Angew. Chem., Int. Ed., 2011, 50,
7998–8014; (b) Y.-Y. Wang, P. C. W. Lui and J. Y. Li,
Immunotherapy, 2009, 1, 983–993; (c) W. M. Pardridge,
Bioconjugate Chem., 2008, 19, 1327–1338.
P. Caravan, J. J. Ellison, T. J. McMurry and R. B. Lauffer,
Chem. Rev., 1999, 99, 2293–2352.
For select examples of CAs that operate by a τ
R
modulation
mechanism, see: (a) D. Le, L. M. on-Rodriguez, A. Ortiz,
A. L. Weiner, S. Zhang, Z. Kovacs, T. Kodadek and 15 For a recent example of transporting imaging agents across
A. D. Sherry, J. Am. Chem. Soc., 2002, 124, 3514–3515;
b) A. L. Nivorozhkin, A. F. Kolodziej, P. Caravan,
M. T. Greenfield, R. B. Lauffer and T. J. McMurry, Angew.
the BBB and retaining them in the brain, see: W. Pham,
B.-Q. Zhao, E. H. Lo, Z. Medarova, B. Rosen and A. Moore,
NeuroImage, 2005, 28, 287–292.
(
2540 | Org. Biomol. Chem., 2015, 13, 2537–2540
This journal is © The Royal Society of Chemistry 2015