Fluorescent mimics of 5-hydroxytryptamine based on N-alkylated derivatives of 6-hydroxycarbostyril

Article abstract of DOI:10.1039/b917578d

Fluorescent probes based on a 6-hydroxycarbostyril core accumulate inside neurons and astroglia in the absence of a serotonin uptake inhibitor.

Full text of DOI:10.1039/b917578d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Fluorescent mimics of 5-hydroxytryptamine based on N-alkylated
derivatives of 6-hydroxycarbostyrilw
Teresa L. Micotto, Adrienne S. Brown and James N. Wilson*
Received (in Cambridge, UK) 26th August 2009, Accepted 27th October 2009
First published as an Advance Article on the web 9th November 2009
DOI: 10.1039/b917578d
Fluorescent probes based on a 6-hydroxycarbostyril core
accumulate inside neurons and astroglia in the absence of a
serotonin uptake inhibitor.
6-Methoxycarbostyril, 5, was synthesized as described by
Fabian et al.10 Reaction of 5 with 1.2 equiv. of 2-azidoethyl
tosylate in DMF with K2CO3 as base produced intermediate
azido compound 6. Reduction with PMe3 in wet THF afforded
the aminoethyl carbostyril, 7. Finally, removal of the methyl
protecting group was achieved by reacting with 1.5 equiv. of
thiophenol with K2CO3. The synthesis of analogs 2–4 followed
a similar route: reaction of 5 with chloroethylamine followed
by deprotection of the 6-hydroxy group produced 2; 3 and 4
resulted from reaction of 5 with bromoethanol and bromo-
propylbenzene, respectively. In all cases the N-linked isomer
was isolated.w All compounds are freely soluble in methanol;
1–3 are easily dissolved in PBS as well, while 4 is soluble only
at dilute concentrations (o50 mM).
Since the development of the epifluorescent microscope nearly
a century ago, fluorescent probes have played a central role in
imaging and biotechnology applications.1,2 In most, but not
all cases, a fluorescent label is covalently appended to a
recognition unit which imparts specificity for a target. Notable
examples of this construct include fluorescently labeled anti-
bodies used in immunohistochemical assays3 and fluorescent
fusion proteins which are invaluable reporters of protein
expression or localization.4 Fluorescent analogs of bio-
molecules offer an alternative strategy to fluorescently labeled
biomolecules. By employing an inherently fluorescent
structure that closely mimics the native molecule, it is possible
to avoid additional steric bulk, changes in shape, or ionic
character that some fluorophores impart. Fluorescent nucleo-
bases have been successfully demonstrated in structural
studies5 and enzymatic assays6 while fluorescent analogs
of neuroactive compounds have been employed in binding
assays7 and as imaging agents of subcellular processes.8
We report the synthesis of a series of fluorescent probes
designed to mimic 5-hydroxytryptamine (5HT, serotonin).
5HT is a classic neurotransmitter implicated in multiple
emotional and behavioral disorders including depression.
The serotonergic system is the subject of intense research
aimed at understanding the basic mechanisms of disease and
developing pharmaceutical interventions.9 Therefore, new
tools that enable detailed, molecular level investigations of
the serotonergic system would be of great interest. Probes 1–4
(Fig. 1) were designed around a carbostyril core. Compared to
structurally related coumarins, the carbostyril lactam enables
N-alkylation as a convenient route for modification. While not
identical to the indole of 5HT, 6-hydroxycarbostyril preserves
a bicyclic structure and key hydroxy functionality while
exhibiting attractive optical properties. The pendant ethylamine
of 5HT is preserved in 1, while 2 is N,N-dimethyl analog of 1.
Most ligands targeting neurotransmitter receptors and
transporters possess aliphatic amines, but a few examples do
not. Thus, we also wanted to explore the properties of
carbostyrils lacking an ethylamine group with probes 3 and 4.
We investigated the optical properties of 1–4 in order to
determine their suitability as fluorescent probes. Based on
previous reports of related carbostyrils,11,12 we anticipated
that the phenol functionality would be sensitive to pH and
would strongly influence the absorption and emission wave-
lengths. Absorption spectra were obtained in neutral, acidic
and basic solutions. In DPBS (pH 7.2) 1–4 exhibit absorption
maxima between 350 and 355 nm with two prominent
shoulders at 340 nm and 370 nm (Fig. 2). Addition of acetic
acid effected a hypsochromic shift (lmax = 335 nm); the
addition of triethylamine results in pronounced bathchromic
(lmax = 370 nm) and hyperchromic shifts. The linear
Department of Chemistry, University of Miami, 1301 Memorial
Drive, Coral Gables, FL 33124, USA. E-mail: jnwilson@miami.edu;
Fax: +1 305 284 4571; Tel: +1 305 284 2619
w Electronic supplementary information (ESI) available: Detailed
description of synthesis, MS, HRMS, 1H, 13C, optical spectra,
additional confocal images and bleaching experiments. See DOI:
10.1039/b917578d
Fig. 1 Top, the structure of 5-hydroxytryptamine (5HT) and fluorescent
mimics 1–4. Bottom, the synthetic route to produce carbostyril
probe 1. Full experimental details for 1–4 are available in the ESI.w
ꢀc
This journal is The Royal Society of Chemistry 2009
7548 | Chem. Commun., 2009, 7548–7550

View Article Online
only, then rinsed). We anticipated that several modes of
interaction are possible between the fluorescent probes and
the cultured cells including specific interactions such as active
transport via neurotransmitter transporters, or binding to
neurotransmitter receptors, as well as non-specific interactions
such as insertion into the cell membrane. Pre-treatment
of the cell cultures with solutions of clomipramine, a
serotonin reuptake inhibitor13 (SRI), significantly reduced
the uptake of 1 and 4. This suggests that at least one mode
of fluorophore–cell interaction involves serotonin transporters.
The lack of response of 2 relative to 1 is somewhat surprising
as they differ only by two methyl groups. However, a survey of
compounds that function as SRIs reveals that many bear
secondary and tertiary aliphatic amines.14 Specifically, SRIs
citalopram, clomipramine and zimelidine possess tertiary
dimethylamines suggesting that 2 may not be an effective
substrate for serotonin transporters (SERT). The observed
uptake of 4 was also unexpected as it lacks the amine
functionality typical of most neuroactive compounds.
However, several serotonin mimics bearing large aromatic
appendages are known including roxindole, a 5HT1A agonist,
and BRL 15 572, a 5HT1D partial agonist. More importantly,
computed 3D molecular models of SERT (and related neuro-
transmitter transporters) reveal that the binding pocket and
pore contain multiple aromatic residues.15 It is plausible that
p–p interactions may allow 4 to be a substrate for SERT. We
are currently investigating the uptake of 1–4 against a broad
range of SRIs, DRIs and NRIs in order to determine the exact
mechanism of transport.
Fig. 2 Absorbance, excitation and emission spectra of 1 compared
with 5HT.
combination of the absorption spectra with excess acid
and excess base nearly perfectly overlaps with the spectrum
obtained at pH 7.2 (Fig. S5, ESIw). Therefore, we conclude
that near neutral pH, both protonated and unprotonated
species exist at roughly equal concentrations.
The emission maxima for 1–4 were found to be ca. 480 nm
in DPBS, representing a Stokes shift of 125 nm. The addition
of excess base did not significantly shift the emission maxima.
Thus, we can conclude that in neutral and basic solutions, 1–4
emit from a deprotonated excited state species. Emission in
acidic solutions is dominated by the deprotonated species as
well. In the case of 3 and 4 a small peak at 380 nm was
observed, which is likely due to photoemission from the
protonated form of these molecules. While 5HT itself is
fluorescent, single photon excitation of 5HT requires specialized
optics and a UV excitation source (lmax,abs = 277 nm).
Furthermore, the autofluorescence of biological samples is
also quite high in the UV region. The longer wavelength
absorption of 1–4 avoids excitation of aromatic amino acid
residues or nucleobases and the overall brightness of
these probes enables their imaging in the presence of other
endogenous fluorophores (see below). The excitation and
emission wavelengths of 1–4 are comparable to the commonly
employed probes DAPI and Hoechst 33 258 enabling the use
of commercially available filter sets. In addition, the absorption
spectrum possesses a broad shoulder extending past 400 nm
which facilitates excitation by standard 405 nm diode lasers
as well.
While in the microwell plate assay above, 1 showed
differential affinities for cultured brainstem cells in the absence
and presence of an SRI, we wanted to determine the fate of the
fluorescent probe. Was it localized to the cell membrane,
implying binding to transporters or receptors, or was it
internalized, implying active transport into the cytosol?
Neurons and astroglia isolated from E14 chick brainstems
were grown on poly-L-lysine and laminin treated coverslips
affixed to 35 mm culture dishes enabling imaging of live cells in
culture media. Excitation was via a 405 nm diode laser;
emission wavelengths were selected by adjusting the window
of a prism to 450 through 600 nm. Cells were exposed to the
probe solution for one minute, the media was then changed
and additional images were captured. As is evident in Fig. 4,
1 bound to and illuminated most healthy cells. We obtained
z-stacks of the cells (Fig. 4, bottom row) and found that 1 was
indeed localized internally after only one minute of exposure.
During the course of our imaging experiments, some photo-
bleaching was observed. However, 1 was sufficiently stable to
capture multiple images over 10 min without the use of
antifade reagents or solutions.
Mixed cell cultures of neurons and astroglia isolated from
the brain stems of E14 chick embryos were exposed to
solutions of 1–4. Serotonergic cells are clustered in the raphe
nucleus (RN) located along the midline of the brain stem;
they project into other regions of the brain including the
hypothalamus, the olfactory bulb, cortex and cerebellum.9
We hypothesized that if 1–4 were effective as serotonin mimics,
they would exhibit affinities for the serotonergic cells present
in the RN. The inherent fluorescence of the carbostyril core
enabled direct detection of their cellular uptake on a microwell
plate reader (lex = 355 nm, lem = 444 nm) (Fig. 3). A high
emission response was observed for 1 and to a lesser extent, 4,
while 2 and 3 did not differ significantly from the controls
(control A: cells alone, control B: wells treated with compound
Progress in understanding the molecular basis of neuro-
logical diseases and disorders will benefit from the development
of tools that enable imaging of subcellular processes.
Fluorescent analogs of neurotransmitters offer a complementary
strategy to commercially available probes for imaging neurons
and astroglia. For example, fluorescently tagged antibodies or
fluorescent fusion proteins can specifically label receptors and
transporters; they do not enable direct monitoring of neuro-
transmitter localization, transport or release. Additionally,
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 7548–7550 | 7549

View Article Online
optical properties and behavior in vitro. In screening their
activity against cultured neurons and astroglia in the presence
and absence of the SRI clomipramine, we have determined
that two probes, 1 and 4, may function as fluorescent analogs
of 5HT. Confocal images of cells exposed to 1 reveal that this
probe is localized to the interior of both astroglia and neurons.
While these results implicate SERT as a possible pathway,
we are currently working to determine the biomolecular
machinery responsible for uptake.
We thank Prof. Kathryn Tosney and Linda White for their
advice and assistance with cell cultures as well as Dr James
Baker for assistance with confocal imaging. T.L.M. was
supported by a UM research fellowship.
Fig. 3 A microwell plate assay using the inherent fluorescence of the
carbostyril probes demonstrates the affinity of 1 and 4 for cultured
neurons and astroglia isolated from E14 chick brainstems. Clomipramine,
a serotonin reuptake inhibitor, reduces the response of both 1 and 4.
Notes and references
1 K. Reichert, Phys. Z., 1912, 12, 1010; O. Heimstadt, Z. Wiss.
Mikrosk., 1912, 28, 330.
2 M. S. T. Goncalves, Chem. Rev., 2009, 109, 190; A. Mayer and
S. Neuenhofer, Angew. Chem., Int. Ed. Engl., 1994, 33, 1044.
3 A. H. Coons, H. J. Creech, R. N. Jones and E. Berliner,
J. Immunol., 1942, 45, 159.
4 B. N. G. Giepmans, S. R. Adams, M. H. Ellisman and R. Y. Tsien,
Science, 2006, 312, 217.
5 D. C. Ward, E. Reich and L. Stryer, J. Biol. Chem., 1969, 244,
1228; R. S. Coleman and M. L. Madaras, J. Org. Chem., 1998, 63,
5700.
6 Y. Cho and E. T. Kool, ChemBioChem, 2006, 7, 669.
7 J. W. Schwartz, R. D. Blakely and L. J. DeFelice, J. Biol. Chem.,
2003, 278, 9768.
8 N. G. Gubenator, H. Zhang, R. G. W. Staal, E. V. Mosharov,
D. B. Pereira, M. Yue, V. Balsanek, P. A. Vadola, B. Mukherjee,
R. H. Edwards, D. Sulzer and D. Sames, Science, 2009, 324, 1441.
9 J. G. Hensler, in Basic Neurochemistry, ed. G. J. Siegel,
R. W. Albers, S. T. Brady and D. L. Price, Academic Press,
Burlington, 7th edn, 2006, ch. 13, pp. 227–239.
10 W. M. F. Fabian, K. S. Niederreiter, G. Uray and W. Stadlbauer,
J. Mol. Struct., 1999, 477, 209.
11 G. Uray, K. S. Niederreiter, F. Belaj and W. M. F. Fabian,
Helv. Chim. Acta, 1999, 82, 1408; N. S. Badgujar, M. Pazicky,
P. Traar, A. Terec, G. Uray and W. Stadlbauer, Eur. J. Org.
Chem., 2006, 2715.
12 N. Y. Vasilyeva and O. V. Vusovich, Proc. SPIE-Int. Soc. Opt.
Eng., 2002, 4747, 230.
13 T. Suhara, A. Takano, Y. Sudo, T. Ichimiya, M. Inoue, F. Yasuno,
Y. Ikoma and Y. Okubo, Arch. Gen. Psychiatry, 2003, 60, 386.
14 National Center for Biotechnology Information, PubChem
Database. http://pubchem.ncbi.nlm.nih.gov/.
15 L. Celik, S. Sinning, K. Severinsen, C. G. Hansen, M. S. Møller,
M. Bols, O. Wiborg and B. Schiøtt, J. Am. Chem. Soc., 2008, 130,
3853; A. W. Ravna, I. Sylte and S. G. Dahl, J. Pharmacol. Exp.
Ther., 2003, 307, 34.
16 H. Corrodi and G. Jonsson, J. Histochem. Cytochem., 1967, 15, 65.
17 R. Gutzmer, C. Dietsel, S. Mommert, B. Kother, H. Stark,
M. Wittman and T. Werfel, J. Immunol., 2005, 174, 5224.
18 V. V. Bezuglov, N. M. Gretskaya, S. E. Esipov, N. B. Polyakov,
L. A. Nikitina, G. A. Buznikov and J. Lauder, Russ. J. Bioorg.
Chem., 2004, 30, 512.
Fig. 4 E14 chick brainstem cells exposed to 1 rapidly uptake the
fluorescent probe. Top row: A, phase contrast image showing the
viewing area; B, control image before addition of 1 showing minimal
cellular autofluorescence (laser intensity and gain were not altered
during the acquisition of subsequent images); C, silhouetted cells as
the fluorophore solution is added; D, 90 s after addition, virtually all
healthy cells have accumulated the probe. Bottom row: E–H, selected
z-stack sections (9 mm top to bottom) clearly demonstrating the
internalization of the fluorophore. Complete z-stack and 3D rotations
available in the ESI.w
histochemical techniques such as fluorogenic reactions of
neuronal monoamines with formaldehyde can only be per-
formed on fixed or dried tissue samples preluding their use in
imaging of dynamic processes.16 Fluorescently labeled neuro-
transmitters have been reported, yet they either are limited to
reporting the presence of cell surface receptors17 or exhibit
activity markedly different from the natural substrate.18
We have synthesized a series of fluorescent analogs of 5HT
as novel probes for the serotonergic system, and explored their
ꢀc
This journal is The Royal Society of Chemistry 2009
7550 | Chem. Commun., 2009, 7548–7550