0022-3565/06/3192-739748$20.

00
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright 2006 by The American Society for Pharmacology and Experimental Therapeutics
JPET 319:739748, 2006

Vol. 319, No. 2

105981/3144005
Printed in U.S.A.

Tethered Yohimbine Analogs as Selective Human 2CAdrenergic Receptor Ligands

Supriya A. Bavadekar, Guoyi Ma, Suni M. Mustafa, Bob M. Moore, Duane D. Miller,
and Dennis R. Feller
Department of Pharmacology (S.A.B., D.R.F.) and National Center for Natural Products Research (G.M., D.R.F.), School of
Pharmacy, University of Mississippi, Oxford, Mississippi; and Department of Pharmaceutical Sciences (S.M.M., B.M.M.,
D.D.M.), University of Tennessee Health Science Center, Memphis, Tennessee
Received April 10, 2006; accepted July 20, 2006

Efforts made toward understanding the biological significance of each of the 2-adrenergic receptor (AR) subtypes
(2A, 2B, and 2C) (Bylund et al., 1998) have resulted only in
marginal success because of the lack of subtype-selective

This work was supported in part by U.S. Public Health Service Grant GM
29358 and U.S. Department of Agriculture ARS Agreement 58-6408-2-0009.
The work was part of the doctoral dissertation of Supriya A. Bavadekar at
the University of Mississippi, University, MS.
The work has been previously presented, partially or entirely, at the following meetings: 1) Bavadekar SA, Ma G, Mustafa SM, Moore BM, Liggett SB,
Miller DD, and Feller DR (2004) Tethered monomeric yohimbine analogs as
selective human 2c-adrenergic receptor ligands, in Proceedings of the Southeastern Pharmacology Society Meeting; 2004 Nov 4 5; University, MI; 2)
Mustafa SM, Bavadekar SA, Moore BM, Liggett SB, Feller DR, and Miller DD
(2004) Synthesis and selectivity studies of yohimbine and its monomeric analogs on 2C-adrenergic receptors, in Proceedings of the 228th American Chemical Society National Meeting; 2004 Aug 2226; Philadelphia; 3) Bavadekar
SA, Suni MM, Moore BM, Liggett SB, Miller DD, and Feller DR (2004)
Monomeric yohimbine analogs as selective human 2C-adrenergic receptor
ligands, in Proceedings of Experimental Biology 2004; 2004 Apr 1721; Washington DC.
Article, publication date, and citation information can be found at
http://jpet.aspetjournals.org.
doi:10.1124/jpet.106.105981.

exhibited higher binding affinities at the 2C- versus 2A- and

2B-AR subtypes. Notably, the benzyl carboxy alkyl amine and
the carboxy alkyl amine analogs exhibited 43- and 1995-fold
and 295- and 54-fold selectivities in binding to the 2C- versus
2A- and 2B-ARs, respectively. Data from luciferase reporter
gene assays confirmed the functional antagonist activities and
selectivity profiles of selected compounds from the tethered
series. The data demonstrate that the second pharmacophore
may not be essential to obtain 2C-AR subtype selectivity,
previously observed with the dimers. Further changes in the
nature of the tether will help in optimization of the structureactivity relationship to obtain potent and selective 2C-AR ligands. These compounds may be used as pharmacological
probes and in the treatment of human disorders.

ligands. In recent years, this endeavor has been greatly assisted by genetic manipulation using mice with deletions,
mutations, or overexpression of specific 2-AR subtypes. The
role of the 2C-AR, in addition to the 2A-AR, in the feedback
control of neurotransmitter release is a finding from one such
study (Hein et al., 1999). Contribution of the 2C-ARs to
2-AR opioid synergy induced by certain agonists such as
moxonidine is another finding (Fairbanks et al., 2002), suggesting that the 2C-AR may represent a better therapeutic
target for analgesic therapy than the 2A-AR, since use of
this subtype would also lead to fewer sedative effects. Peterhoff et al. (2003) used knockout mice to report that the 2Aand 2C-ARs mediate epinephrine-induced inhibition of insulin secretion in pancreatic islet cells. In the central nervous
system, the 2C-ARs seem to have a distinct inhibitory role in
various central nervous system-mediated behavioral and
physiological responses including startle reactivity, aggressive behavior, and amphetamine-induced locomotor hyperactivity (Scheinin et al., 2001). Thus, increased 2C-AR activity
may lead to or result from a constitutively stressful state,

ABBREVIATIONS: AR, adrenoceptor; CHO, Chinese hamster ovary; HEK, human embryonic kidney; CRE-LUC, cAMP response elementluciferase.
739

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

ABSTRACT
Yohimbine is a potent and relatively nonselective 2-adrenergic
receptor (AR) antagonist. In an earlier report, we demonstrated
that dimeric yohimbine analogs containing methylene and
methylene-diglycine tethers were highly selective human
2C-AR ligands. Little work has been done to examine the role
of the tether group or the absence of the second yohimbine
pharmacophore on selectivity for human 2-AR subtypes. The
goal of our study was to determine the binding affinities and
functional subtype selectivities of a series of tethered yohimbine ligands in the absence of the second pharmacophore. The
profiles of pharmacological activity for the yohimbine analogs
on the three human 2-AR subtypes expressed in Chinese
hamster ovary cells were examined using receptor binding and
cAMP inhibition assays. All of the tethered yohimbine analogs

740

Bavadekar et al.

Materials and Methods

Sources of Materials. All cell culture reagents were obtained
from Invitrogen (Carlsbad, CA). CHO cells expressing homogeneous

populations of human 2A-, 2B-, and 2C-ARs were obtained from

Drs. Marc Caron and Robert Lefkowitz (Duke University Medical
Center, Durham, NC) and Dr. Stephen Liggett (College of Medicine,
University of Cincinnati, Cincinnati, OH). HEK293 cells expressing
homogeneous populations of human 1A-, 1B-, and 1D-ARs were
obtained from Dr. Kenneth Minneman (Emory University School of
Medicine, Atlanta, GA). The cAMP response element-luciferase gene
construct (6 CRE-LUC) was provided by Dr. A. Himmler (Boehringer
Ingelheim Research and Development, Vienna, Austria). Yohimbine
and yohimbinic acid were obtained from ICN Biomedicals Inc. (Aurora, OH) and Aldrich Chemical Co. (Milwaukee, WI), respectively.
Tethered yohimbine analogs and the n 3 yohimbine dimer were
provided by Dr. Duane D. Miller (Department of Pharmaceutical
Sciences, University of Tennessee, Memphis, TN). The procedures
for synthesis of the tethered yohimbine analogs and the n 3
yohimbine dimer are as described by Mustafa et al. (2005), Zheng et
al. (2000), and Zheng (1999). Solutions of the n 3 yohimbine dimer
were prepared as described previously (Lalchandani et al., 2002).
Yohimbinic acid (2) and all the tethered yohimbine analogs, with the
exception of the alkyl amine analog (8) and the carboxy alkyl amine
analog (10), were dissolved in a mixture of water and dimethyl
sulfoxide. Yohimbine (1), the alkyl amine analog (8), and the carboxy
alkyl amine analog (10) were dissolved in water alone. Stock solutions (102 M) were prepared and diluted in water to appropriate
concentrations for the studies. [3H]Rauwolscine and [3H]prazosin
were obtained from PerkinElmer Life and Analytical Sciences (Boston, MA), and all other chemicals were obtained from Sigma-Aldrich
(St. Louis, MO).
Cell Culture. CHO cells stably expressing homogeneous populations of 2A-, 2B-, and 2C-ARs were grown in 150 cm2 Corning
flasks with Hams F-12 medium supplemented with 10% fetal bovine
serum, 2 mM glutamine, penicillin (100 units/ml), streptomycin (100
g/ml), and Geneticin (100 g/ml). The flasks were incubated at 37C
(5% CO2). Media were changed every 48 h until the cells were
confluent. Upon confluence, the cells were detached by trypsin
(0.05% trypsin EDTA, 5 min).
HEK293 cells stably expressing homogeneous populations of
1A-, 1B-, and 1D-ARs were grown in 150 cm2 Corning flasks
with Dulbeccos modified Eagles medium supplemented with 10%
fetal bovine serum, 2 mM glutamine, penicillin (100 units/ml),
streptomycin (100 g/ml), and Geneticin (100 g/ml). The flasks
were incubated at 37C (5% CO2). Media were changed every 48 h
until the cells were confluent. Upon confluence, the cells were
detached by gentle scraping.
Radioligand Binding Assays. Radioligand binding studies
were conducted in intact CHO cells expressing homogeneous populations of 2A-, 2B-, and 2C-ARs. Similar studies were performed in intact HEK293 cells expressing homogeneous populations of 1A-, 1B-, and 1D-ARs. In brief, CHO cells were
harvested using Hams F-12 media after trypsinization, whereas
HEK293 cells were detached by simple scraping. Detached cells
were washed and centrifuged with Tris-EDTA buffer, pH 7.4;
containing 50 mM Tris, 20 mM disodium EDTA, and 154 mM
NaCl, in which they were finally suspended. Competition binding
assays were performed in duplicate by incubating 50,000 cells
with [3H]rauwolscine (0.1 Ci, 0.7 nM) for human 2A-, 2B-, and
2C-ARs and [3H]prazosin (0.1 Ci, 0.7 nM) for human 1A-, 1B-,
and 1D-ARs and varying concentrations of the analogs under
investigation in a water bath at 37C. The assays were conducted
in a final volume of 2 ml. Nonspecific binding was determined in
the presence of 10 M phentolamine. Incubations were terminated at 60 min by rapid filtration over GF/C glass fiber filters
(Whatman, Maidstone, UK) using a cell harvester (Brandel Inc.,
Gaithersburg, MD). The filter discs were washed three times with
Tris-EDTA buffer, pH 7.4, at 4C. The radioactivity was quantified by using a Packard Tri-Carb 2900 TR liquid scintillation
analyzer (Packard Instrument Company, Meridian, CT), and data
were analyzed using GraphPad Prism (GraphPad Software Inc.,

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

thereby causing depression, which suggests that 2C-AR subtype-selective drugs may be useful in a variety of neuropsychiatric disorders (Scheinin et al., 2001). Besides these findings derived from gene-targeted mice, a recent study
(Chotani et al., 2000) has provided yet another potential
therapeutic use for an 2C-AR antagonist. The study showed
that at lower temperatures, the 2C-ARs are principally responsible for mediating the cold-induced augmented vasoconstrictor response. This subtype, however, did not contribute
to 2-AR-dependent vasoconstriction at 37C. A selective inhibition of the 2C-ARs in microvessels has, thus, been proposed to provide an effective treatment for cold-induced cutaneous arterial blood vessel constriction as observed in
Raynauds phenomenon.
The increasing number of potential therapeutic uses has
greatly stimulated interest in the design of ligands that interact selectively with the 2C-ARs. Lalchandani et al. (2002)
have shown that dimers of the 2-AR subtype nonselective
antagonist ligand yohimbine exhibited selectivity for the 2CAR. The n 3 and n 24 dimers exhibited the greatest
2C-AR selectivity in the series tested. Interestingly, none of
the analogs surpassed the affinity of the parent compound,
yohimbine. In addition, the exact mechanism underlying the
2C-AR selectivity observed for these dimeric compounds is
unclear. An attempt to assess the role of the second pharmacophore and the spacer arm in the potency of dimeric ligands
was recently made by Mustafa et al. (2005). To achieve this,
a monomeric tethered ligand versus a dimeric ligand approach was used. Compounds having only one pharmacophore, i.e., only one yohimbine molecule, to which side chains of
varying length and containing different patterns of hydrogen
bond donors/acceptors, rigidity, hydrophobicity and/or charge
had been appended, were evaluated for binding to the 2CARs. The data showed that even in the absence of the second
pharmacophore, the tethered yohimbine analogs displayed
high binding affinities at the 2C-AR.
In our study, our aims were to 1) determine the affinities of
the tethered yohimbine ligands at all three 2-ARs and examine subtype selectivities exhibited, if any, by the compounds and 2) elucidate the underlying physicochemical basis for the observed 2-AR subtype selectivities. To start
with, we have evaluated the standards for this study viz. the
parent compound, yohimbine, and the n 3 analog from the
dimer series at the 2-ARs. Furthermore, we have examined
tethered analogs of yohimbine, which consist of various substitutions at the C-16 carbonyl position of yohimbine and
yohimbinic acid (a nontethered analog of yohimbine possessing an acid functional group instead of a methyl ester at the
C-16 position) at 2-AR subtypes. The subtypes were stably
expressed as homogeneous populations in Chinese hamster
ovary (CHO) cells. Selected compounds were tested for binding affinities at 1-AR subtypes, stably expressed as homogeneous populations in human embryonic kidney (HEK293)
cells. Finally, functional activities of selected compounds
were determined in CHO cells expressing the 2A- and 2CARs using a cAMP response element-luciferase (CRE-LUC)
reporter gene assay.

Yohimbine Analogs as Selective 2C-Adrenoceptor Ligands

pounds, ranging from 1012 to 105 M, were added in duplicate

within each experiment, and the individual molar IC50 values were
determined using GraphPad Prism. The displacement curves were
plotted using a standard slope factor of 1.0, the Ki values of the
competing ligands were determined using the equation of Cheng and
Prusoff (1973), and the final data are presented as means S.E.M.
of n 4 or more experiments. In the functional studies, the IC50
values of the antagonists for the concentration-dependent reversal of
the action of the agonist, medetomidine, against forskolin-induced
cAMP responses at the human 2C-AR were calculated using GraphPad Prism and are expressed as the means S.E.M. of n 4
experiments. Differences between means of binding affinities and
functional responses for individual ligands on the AR subtypes were
analyzed by analysis of variance and Tukeys post hoc analysis test.
When two means were compared, statistical analyses were done
using Students t test. Values were considered to be statistically
significant when P 0.05.

Results
Radioligand Binding Assay. Radioligand binding analyses of yohimbine (1), yohimbinic acid (2), the n 3 yohimbine dimer (3), and tethered yohimbine analogs (410) were
performed in CHO cells stably expressing homogeneous populations of human 2A-, 2B-, and 2C-ARs. Structures of all
compounds are provided in Fig. 1. The binding affinities of
the compounds are presented in Table 1.
As described previously (Mustafa et al., 2005), the rank
order of binding affinities exhibited by yohimbine (1) on the
human 2-AR subtypes was 2C 2A 2B, with 2- and
7-fold higher binding affinities for the 2C- versus the 2Aand 2B-ARs. Interestingly, studies with yohimbinic acid (2)
revealed that this compound exhibited a greatly decreased
binding potency at the 2A-AR (345-fold) versus the 2B-AR
(2-fold) and the 2C-AR (20-fold), compared with yohimbine.
Furthermore, it was equipotent in binding to the 2B- and
2C-ARs, and its binding at the 2B-AR was 46-fold greater
than its binding at the 2A-AR. Figure 2 shows the binding
displacement curves for yohimbine (1) (A) and yohimbinic
acid (2) (B) at the three 2-AR receptor subtypes. The n 3
dimeric analog (3) was 18- and 68-fold selective in binding to
the 2C- versus 2A- and 2B-AR subtypes (Table 1). All of the
tethered yohimbine analogs (410) exhibited significantly
higher binding affinities at the 2C- versus 2A- and 2B-AR
subtypes (Table 1). In particular, the benzyl carbamate alkyl
amine (6), t-butyl carbamate alkyl amine (7), benzyl carboxy
alkyl amine (9), and carboxy alkyl amine (10) analogs possessed binding affinities comparable with those of the parent
molecule, yohimbine (1), at the 2C-AR (Table 1). The alkyl
amine analog (8), however, was 32-fold less potent in binding
to the 2C-AR compared with yohimbine (1). In addition, it
was 129- and 316-fold less potent in binding to the 2A- and
2B-AR subtypes in comparison with yohimbine. The benzyl
carbamate alkyl amine analog (6) and the t-butyl carbamate
alkyl amine analog (7) were 11- and 59-fold and 3- and
33-fold selective in binding to the 2C- versus the 2A- and
2B-ARs, respectively (Table 1). The benzyl carboxy alkyl
amine analog (9) and the carboxy alkyl amine analog (10)
exhibited 43- and 1995-fold and 295- and 54-fold selectivities
in binding to the 2C- versus 2A- and 2B-ARs, respectively.
Figure 2, C and D, illustrates the binding displacement
curves for the benzyl carboxy alkyl amine analog (9) and the
carboxy alkyl amine analog (10) at the 2A-, 2B-, and 2C-AR

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

San Diego, CA). The displacement curves were plotted using a

standard slope factor of 1.0; and the Ki values of the competing
ligands were determined using the equation of Cheng and Prusoff
(1973). The percentage of specific binding in the inhibition experiments was determined by dividing the difference between the
total bound (disintegrations per minute) and nonspecific bound
(disintegrations per minute) by the total bound (disintegrations
per minute).
Scatchard analyses were carried out using varying concentrations,
ranging from 0.03 to 3.6 nM, of selected radioligands to determine
their affinities (Kd) and maximal binding characteristics (Bmax). The
saturation binding of [3H]rauwolscine to human 2A-, 2B-, and 2CARs and [3H]prazosin to human 1A-, 1B-, and 1D-ARs was conducted in a final volume of 1 ml. Nonspecific binding for the 1- and
2-ARs was determined in the presence of 10 M phentolamine and
10 M yohimbine, respectively. The total and nonspecific binding for
each concentration was determined in triplicate. The specific binding, at each concentration of the radioligand, was established and
plotted as bound ligand versus bound/free ligand and the corresponding Kd and Bmax values calculated on each human -AR subtype.
Data are expressed as the means S.E.M. of n 6 to 9 experiments.
The experimentally determined Kd (nanomolar concentrations) and
Bmax (picomoles per milligram of protein) values (means S.E.M.,
n 6 9 experiments) of the radioligands on the AR subtypes were as
follows: [3H]rauwolscine: 2A 1.93 0.12 nM and 8.20 0.71
pmol/mg protein, 2B 1.45 0.08 nM and 1.64 0.10 pmol/mg
protein, and 2C 0.32 0.01 nM and 1.20 0.13 pmol/mg protein
in CHO cells; and [3H]prazosin: 1A 0.22 0.008 nM and 0.56
0.01 pmol/mg protein, 1B 0.24 0.01 nM and 1.59 0.10
pmol/mg protein, and 1D 0.14 0.01 nM and 0.46 0.04 pmol/mg
protein in HEK293 cells.
cAMP Response Element-Luciferase Reporter Gene Assay.
To verify that the observed binding affinities of yohimbine and its
selected analogs correlate with the functional responses in the 2ARs, functional responses of selected ligands were determined by
using six copies of a cAMP response element-luciferase reporter gene
construct (6 CRE-LUC, pADneo2-C6-BGL). The reporter gene assays
were conducted in CHO cells expressing the human 2A- and 2C-AR
subtypes. The cells were grown to confluence, upon which they were
isolated and electroporated in the presence of the plasmid. The
transfection procedure used was the same as that described previously in CHO cells (Vansal and Feller, 1999; Lalchandani et al.,
2002). Cells were transiently transfected with the 6 CRE-LUC plasmid (5 g/100 l of cell suspension) using electroporation at 150 V,
70 ms, single pulse. Transfected cells were plated into a 96-well
microplate at a density of approximately 50,000 cells per well and
allowed to grow for 20 h. Fixed concentrations of the nonselective
2-AR agonist medetomidine (Virtanen et al., 1988) (0.01 and 1 M
for the 2A- and 2C-ARs, respectively), which produced a submaximal inhibition of the forskolin response, were added directly to the
medium 20 min before the addition of forskolin (35 M) and then
allowed to incubate for 4 h. Selected ligands were tested for antagonist activity and added 20 min before the addition of medetomidine.
The media were then aspirated, the cells were lysed, and the luciferase activity was determined using the LucLite assay kit (Packard
Biosciences, Meriden, CT). Changes in light production were measured on a Packard Topcount Luminescence Counter (Packard Biosciences) after addition of luciferin. Medetomidine inhibited the forskolin-induced cAMP changes by 50 to 70% in each of the two
subtypes, and the antagonist effects of yohimbine and its selected
analogs were determined by their ability to reverse the medetomidine action. The IC50 values of yohimbine and its analogs for the
concentration-dependent reversal of the medetomidine action
against forskolin-induced cAMP responses at the human 2C-AR
were calculated using GraphPad Prism and are expressed as the
means S.E.M. of n 4 experiments.
Data Accumulation and Statistical Analyses. For ligand binding studies in cell lines, varying concentrations of each of the com-

741

742

Bavadekar et al.

TABLE 1
Binding affinities of yohimbine, yohimbinic acid, the n 3 yohimbine dimer, and tethered yohimbine analogs on human 2A-, 2B-, and 2C-AR
subtypes that are stably expressed in CHO cells
3HRauwolscine was used as the radioligand in the equilibrium competition radioligand binding assays for the 2-ARs, and the nonspecific binding was measured in the
presence of 10 M phentolamine. pKi log Ki (Ki was calculated according to the Cheng-Prusoff equation). See Materials and Methods. The data are means S.E.M. of
n 4 6 experiments. Statistical analyses were carried out by one-way analysis of variance followed by Tukeys test (P 0.05).
Compound

2A pKi

2B pKi

2C pKi

Yohimbine (1)
Yohimbinic acid (2)
Yohimbine n 3 dimer (3)
Yohimbine monoglycine ester (4)
Yohimbine diglycine ester (5)
Yohimbine benzyl carbamate alkyl amine (6)
Yohimbine t-butyl carbamate alkyl amine (7)
Yohimbine alkyl amine (8)
Yohimbine benzyl carboxy alkyl amine (9)
Yohimbine carboxy alkyl amine (10)

8.75 0.03a
6.21 0.12a
6.80 0.02a
7.47 0.02c
7.33 0.01a
7.91 0.07a
7.96 0.07a
6.64 0.03a
7.56 0.06a
6.03 0.09a

8.14 0.04
7.87 0.18
6.22 0.03
7.44 0.02
6.62 0.03
7.17 0.09
6.98 0.04
5.64 0.09
5.89 0.06
6.78 0.16

9.01 0.03b
7.71 0.05d
8.05 0.04b
8.07 0.03b
8.28 0.07b
8.94 0.11b
8.50 0.07b
7.50 0.04b
9.19 0.02b,e
8.50 0.07b

The mean pKi value on the 2A-AR subtype is significantly different from the mean pKi value on the 2B- and 2C-AR subtype.
The mean pKi value on the 2C-AR subtype is significantly different from the mean pKi value on the 2A- and 2B-AR subtype.
The mean pKi value on the 2A-AR subtype is significantly different from the mean pKi value on the 2C-AR subtype.
d
The mean pKi value on the 2C-AR subtype is significantly different from the mean pKi value on the 2A-AR subtype.
e
The mean pKi value of the analog is significantly different from the mean pKi value of yohimbine on the same subtype.
a
b
c

subtypes, respectively. The alkyl amine analog (8) was 7- and

72-fold selective in binding to the 2C- versus the 2A- and
2B-ARs (Table 1).

As has been observed with the yohimbine dimers (Lalchandani et al., 2002), all of the tethered yohimbine analogs, with
the exception of the monoglycine ester (4) and the carboxy

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

Fig. 1. Chemical structures of yohimbine (Yoh) and analogs.

Yohimbine Analogs as Selective 2C-Adrenoceptor Ligands

743

alkyl amine (10) analogs, displayed lower binding affinities

at the 2B- versus the 2A- and 2C-ARs. The monoglycine
ester analog (4) was equipotent in binding to the 2A- and
2B-ARs, whereas the carboxy alkyl amine analog (10) was
6-fold more potent in binding to the 2B- versus the 2A-AR
(Table 1).
The binding affinities of yohimbine (1) and the two selected
analogs, viz. the benzyl carbamate alkyl amine (6) and alkyl
amine (8) analogs, were determined in HEK293 cells stably
expressing homogeneous populations of human 1A-, 1B-,
and 1D-ARs. Yohimbine and the selected analogs were found
to bind with low affinities at all three 1-AR subtypes (Table
2). The binding affinities of yohimbine (1), the benzyl carbamate alkyl amine (6), and the alkyl amine (8) analog for
the 2C-AR subtype were at least 224-, 562-, and 100-fold

greater than those on the 1-AR subtypes, respectively (compare data in Tables 1 and 2). Taken collectively, the data
confirm the binding selectivity of these ligands for the
2C-AR subtype.
cAMP Response Element-Luciferase Reporter Gene
Assay. The functional responses of yohimbine and selected
tethered analogs in the human 2A- and 2C-ARs expressing
CHO cells were determined using 6 CRE-LUC plasmid. The
assays with both 2A- and 2C-ARs were conducted using the
non-subtype-selective 2-AR agonist, medetomidine, to block
the cAMP changes induced by the adenylyl cyclase activator,
forskolin. The concentration of forskolin (35 M) for the
assays was chosen such that it produced at least a 7- to
10-fold increase over basal levels. Basal values (solvent control) were subtracted from the forskolin values, and the re-

TABLE 2
Binding affinities of yohimbine and selected tethered yohimbine analogs on human 1A-, 1B-, and 1D-AR subtypes that are stably expressed in
HEK293 cells
3HPrazosin was used as the radioligand in the equilibrium competition radioligand binding assays for the 1-ARs, and the nonspecific binding was measured in the presence
of 10 M phentolamine. pKi log Ki (Ki was calculated according to the Cheng-Prusoff equation). See Materials and Methods. The data are means S.E.M. of n 4 6
experiments. Statistical analyses were carried out by one-way analysis of variance followed by Tukeys test (P 0.05).
Compound

1A pKi

Yohimbine (1)
Yohimbine benzyl carbamate
alkyl amine (6)
Yohimbine alkyl amine (8)

6.66 0.02
6.19 0.16

a
b

1B pKi
a

Maximum % specific
inhibition 50% at 10 M

6.03 0.08
6.16 0.20

1D pKi
b

Maximum % specific
inhibition 50% at 10 M

The mean pKi value on the 1A-AR subtype is significantly different from the mean pKi value on the 1B-AR subtype.
The mean pKi value on the 1B-AR subtype is significantly different from the mean pKi value on the 1D-AR subtype.

6.52 0.15
6.02 0.22
Maximum % specific
inhibition 50% at 10 M

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

Fig. 2. Binding displacement curves of (A) yohimbine, (B) yohimbinic acid, (C) yohimbine benzyl carboxy alkyl amine, and (D) yohimbine carboxy alkyl
amine for human 2A- F , 2B- , and 2C-ARs E stably expressed in CHO cells. Plotted values are means S.E.M. (n 4 6 experiments). Structures
of compounds are shown in Fig. 1.

744

Bavadekar et al.

Fig. 3. Concentration-dependent effects of medetomidine on forskolininduced cAMP elevations, as assessed by luciferase activity, on human
(A) 2A- versus (B) 2C-ARs stably expressed in CHO cells. Plotted values
are means S.E.M. (n 4 experiments).

forskolin-induced cAMP. Medetomidine alone (in the absence

of forskolin) did not show an increase in cAMP levels when
tested at a concentration of 1 M (data not shown). These
results may suggest a Gi (inhibitory) to Gs (stimulatory)
coupling change only for the 2A-AR at higher agonist concentrations (S. G. Lalchandani and D. R. Feller, unpublished
data; Pepperl and Regan, 1993; Eason et al., 1992).
For functional studies with yohimbine and selected yohimbine analogs at the 2A- and 2C-ARs, the concentration of
medetomidine was chosen such that it produced a submaximal (around 50 70%) inhibition of the forskolin response in
these subtypes. From our data in the previous set of experiments (Fig. 3), the medetomidine concentration was fixed at
0.01 M for the 2A-AR whereas it was fixed at 1 M for the
2C-AR. The agonist ligand, medetomidine, was added directly to the medium 20 min before the addition of forskolin
and then allowed to incubate for 4 h. Yohimbine analogs
selected for functional testing at the 2-ARs included the
benzyl carbamate alkyl amine (6) and the alkyl amine (8)
analogs. Concentrations of yohimbine (1), the benzyl carbamate alkyl amine (6), and the alkyl amine analog (8) were fixed
at 0.1, 0.1, and 1 M, respectively, for the 2A-AR assays
whereas they were varied from 0.001 to 10 M for the 2C-AR
assays. These compounds were added 20 min before the
addition of medetomidine.
On the 2A-AR, medetomidine (0.01 M) inhibited forskolin-induced cAMP changes and the mean inhibition produced
was 56 2% for n 5 experiments (Fig. 4). As noted in the
graph, the mean percent inhibition produced after preincubation with yohimbine (1) (0.1 M), the benzyl carbamate
alkyl amine analog (6) (0.1 M), and the alkyl amine analog
(8) (1 M) were 14.2, 54, and 53%, respectively. Thus, the
changes in luciferase activity produced by medetomidine
were blocked by yohimbine (1) (0.1 M). Neither of the two
analogs, however, blocked the medetomidine inhibition of
forskolin-induced luciferase activity at the chosen concentrations. Controls of yohimbine and the two tethered yohimbine
analogs were included for comparison (Fig. 4). In the pres-

Fig. 4. Effects of yohimbine and selected tethered yohimbine analogs on

medetomidine inhibition of forskolin-induced cAMP elevations, as assessed by luciferase activity, on human 2A-ARs stably expressed in CHO
cells. Plotted values are means S.E.M. (n 5 experiments). Structures
of compounds are shown in Fig. 1. F, forskolin (5 M); M, medetomidine
(0.01 M); Y, yohimbine (0.1 M); BC, yohimbine benzyl carbamate alkyl
amine analog (0.1 M); AA, yohimbine alkyl amine analog (1 M). , P
0.05 compared with F M (1) using Students t test.

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

sulting forskolin response was used as 100%. Likewise, basal

values were subtracted from all other values obtained with
ligands, and the data are expressed as a percentage luciferase response relative to that of forskolin alone.
Preliminary experiments with CHO cells stably expressing
the 2A-ARs revealed a biphasic concentration-response
curve with medetomidine (Fig. 3A). As shown, medetomidine
showed inhibition of the forskolin-induced cAMP response at
low concentrations (0.0001 0.01 M), whereas higher concentrations (0.110 M) reversed the inhibition of luciferase
activity observed at the lower medetomidine concentrations.
The maximal inhibition obtained for medetomidine was 56%
at 0.01 M. Medetomidine alone (in the absence of forskolin)
increased cAMP levels by 17% when tested at a concentration
of 10 M (data not shown).
In cells expressing the 2C-AR, however, medetomidine
caused a concentration-dependent reduction in the forskolininduced cAMP activity at all concentrations tested (0.0110
M). The maximal inhibition obtained for medetomidine was
70% at 1 M (Fig. 3B). A higher concentration of medetomidine (10 M) did not cause any further decrease in the

Yohimbine Analogs as Selective 2C-Adrenoceptor Ligands

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

ence of forskolin, none of the test antagonist ligands produced agonist activity at the concentrations tested (data not
shown).
A graphical representation of the concentration-dependent
effects of yohimbine (1), the benzyl carbamate alkyl amine
analog (6), and the alkyl amine analog (8), for the reversal of
the forskolin-induced cAMP changes by medetomidine at the
2C-ARs is provided in Fig. 5, A, B, and C, respectively. As
can be seen from these graphs, the mean percentage inhibition produced by medetomidine (1 M) at the 2C-AR was
70 2% for n 4 experiments. Controls of yohimbine and
the two tethered analogs were included for comparison (Fig.
5). In the presence of forskolin, none of the test antagonist
ligands produced agonist activity at the highest concentration tested (data not shown).
A comparison of the experimentally determined functional
antagonist activities for yohimbine and the selected tethered
yohimbine analogs at human 2A- and 2C-ARs stably expressed in CHO cells is given in Fig. 6. As shown, significant
differences were observed in the abilities of the benzyl carbamate alkyl amine analog (6) and the alkyl amine analog (8)
to reverse the medetomidine inhibition of forskolin-induced
cAMP elevations at the 2A- and 2C-AR subtypes, when used
at the same concentration. Concentrations used were yohimbine (1) (0.1 M), yohimbine benzyl carbamate alkyl amine
analog (6) (0.1 M), and yohimbine alkyl amine analog (8) (1
M). It is important to note here that the two tethered
analogs were able to reverse the medetomidine inhibition of
the forskolin response on the 2C- but not on the 2A-AR
subtype. Furthermore, despite the inhibition produced by
medetomidine being less in the 2A-AR (56 2%) versus the
2C-AR subtype (70 2%), blockade of the medetomidine
responses were only observed with yohimbine on the 2A-AR.
Taken collectively, these data indicate the 2C- versus
2A-AR selectivity of the selected tethered yohimbine analogs.
Table 3 shows the pIC50 values of yohimbine and its analogs for the reversal of the medetomidine action against forskolin-induced cAMP responses at the human 2C-AR, along
with their binding affinities (pKi values) at the same receptor
subtype. A linear regression analysis of the reversal response
for each antagonist was used to determine the IC50 values for
the antagonists. For this purpose, the luciferase response of
forskolin alone was used as 100%, whereas that of forskolin
in the presence of the agonist, medetomidine, was used as
0%. Although pKb values would have served as a better
comparison, the functional potencies (pIC50 values) of yohimbine and the benzyl carbamate alkyl amine analog at the
2C-ARs were not found to be statistically different from the
experimentally determined binding affinities (pKi values). In
addition, the rank order of functional and binding potencies
for the three compounds remained the same: yohimbine benzyl carbamate alkyl amine (6) yohimbine (1) yohimbine
alkyl amine (8).

745

Discussion

Fig. 5. Reversal of medetomidine inhibition of forskolin-induced cAMP

elevations, as assessed by luciferase activity, by yohimbine (A), yohimbine benzyl carbamate alkyl amine analog (B), and yohimbine alkyl
amine analog (C) on human 2C-ARs stably expressed in CHO cells.
Plotted values are means S.E.M. (n 4 experiments). Structures of
compounds are shown in Fig. 1. F, forskolin (3 M); M, medetomidine (1
M); Y, yohimbine (0.001, 0.01, and 0.1 M); BC, yohimbine benzyl
carbamate alkyl amine analog (0.001, 0.01, and 0.1 M); AA, yohimbine
alkyl amine analog (0.1, 1, and 10 M). , P 0.05 compared with F M
(1) using Students t test.

G protein-coupled receptors, traditionally considered to

function as monomeric proteins, have now been proposed to
exist as dimers or even higher-structure oligomers (Angers et
al., 2002). The physiological significance of such dimerization
and its implications in ligand pharmacology and, potentially,

for drug design remain to be fully understood. However,

there have been efforts to improve potency and/or subtypeselectivity by use of bivalent or dimeric ligands. Lalchandani
et al. (2002) demonstrated that dimers of the potent and

746

Bavadekar et al.

relatively nonselective 2-AR antagonist yohimbine consisting of two pharmacophores linked through a spacer, exhibited a higher degree of 2-AR subtype-selectivity than the
parent molecule. The authors used the bivalent ligand approach, which is based on the concept that a bivalent ligand
would first undergo univalent binding followed by binding of
the second pharmacophore to a recognition site on a neighboring receptor. Thus, the bivalent ligand would exhibit a
greater potency than that derived from the sum of its monovalent counterparts (Portoghese, 2001). In the yohimbine
dimer study, the addition of methylene and methylene-diglycine spacer linkages produced analogs that were potent and
selective 2C-AR ligands. It was proposed that one pharmacophore (or one yohimbine molecule) binds to the ligand
receptor site, whereas the second pharmacophore may bind
to either 1) an adjacent site of the ligand binding pocket, or
transmembrane domain, or one of the extracellular loops on
the same receptor or 2) a ligand-binding site or a recognition
site on a neighboring receptor. With shorter spacer arms,
however, an interaction of the second pharmacophore with
an adjacent receptor molecule (dimer) seems less probable.
Interestingly, none of the analogs in the yohimbine dimer series
surpassed the affinity of the parent compound, yohimbine.
Recently, Mustafa et al. (2005) demonstrated that monomeric analogs of yohimbine, even in the absence of the second
pharmacophore (or the second yohimbine molecule), with
appendages or tethers of varying nature and composition
retained high binding potencies, like that of the dimeric
analogs, at the human 2C-AR. In fact, one of the tethered
analogs, viz. the benzyl carboxy alkyl amine analog (9), displayed a greater binding affinity at the 2C-AR than the
parent compound, yohimbine (Table 1). This result suggests
that high binding affinities, previously observed with the
dimeric ligands, may not be attributable to receptor dimer-

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

Fig. 6. A comparison of the ability of yohimbine and selected tethered

yohimbine analogs to reverse the medetomidine inhibition of forskolininduced cAMP elevations, as assessed by luciferase activity, on human
2A- and 2C-ARs stably expressed in CHO cells. Plotted values are
means S.E.M. (n 4 5 experiments). Structures of compounds are
shown in Fig. 1. F, forskolin (35 M); M, medetomidine; Y, yohimbine
(0.1 M); BC, yohimbine benzyl carbamate alkyl amine analog (0.1 M);
AA, yohimbine alkyl amine analog (1 M). The concentration of forskolin
(35 M) was chosen such that it produced at least a 7- to 10-fold increase
over basal levels. The concentration of medetomidine was chosen such
that it produced at least a 50% inhibition of the forskolin-induced cAMP
response. The mean inhibition produced with the 2A-ARs was 56 2%
for n 5 experiments, whereas at the 2C-ARs, the mean inhibition was
70 2% for n 4 experiments. , response at the 2C-ARs is different
from the response at 2A-ARs (P 0.05 using Students t test).

ization and clustering, as proposed earlier (Lalchandani et

al., 2002). The main focus of our study was to determine
whether the second pharmacophore was essential for the
2C-AR selectivity of the yohimbine dimers; i.e., would truncating dimeric analogs to tethered analogs still retain the
2C-AR selectivity? For this purpose, we have extended the
evaluation of the tethered yohimbine analogs (Mustafa et al.,
2005) to all human 2-AR subtypes. The importance of the
charge of the tethers, in potency and/or subtype selectivity,
was investigated using neutral, basic, and acidic functional
groups.
In the yohimbine dimer series, the n 3 and n 24 dimers
exhibited the greatest 2C-AR selectivity (Lalchandani et al.,
2002). In functional assays, the n 3 dimeric analog was
found to be more potent as well as more 2C-AR selective
than the n 24 dimer. This finding, along with the idea that
the n 3 dimer would better fit the criteria for drug-like
status (Lipinski et al., 1997) compared with the n 24 dimer,
led us to design the structure of the tethered analogs using
the n 3 yohimbine dimer as a standard. In our study, we
have found the n 3 dimer (3) to be 18- and 68-fold selective
in binding to the 2C- versus the 2A- and 2B-ARs, respectively.
Our data for the parent compound, yohimbine (1) (Table 1;
Fig. 2A), agree with those reported in the literature (Bylund
et al., 1992; Lalchandani et al., 2002). We have also evaluated yohimbinic acid (2), a nontethered structural analog of
yohimbine, for its binding to the 2-ARs. Interestingly, there
are no reports published in the literature on the interaction
of yohimbinic acid with the ARs. Data from this study
showed that a single structural change in the yohimbine
molecule (changing the methyl ester at the C-16 carbonyl of
yohimbine to an acid functionality), yielding yohimbinic acid,
had a significant impact on its binding affinities at the 2-AR
subtypes. Yohimbinic acid exhibited greatly decreased binding potency at the 2A-AR (345-fold) versus the 2B-AR (2fold) and the 2C-AR (20-fold), compared with yohimbine
(Table 1; Fig. 2B). Furthermore, it was equipotent in binding
to the 2B- and 2C-ARs, and its binding at the 2B-AR was
46-fold greater than its binding at the 2A-AR. It is noteworthy that yohimbinic acid is a selective 2C- versus 2A-AR
ligand, with fold selectivity being 32-fold.
In the present study, as in the study reported previously
(Lalchandani et al., 2000), the n 3 dimer (3) did not surpass
the binding affinity of yohimbine. In addition, although all of
the tethered yohimbine analogs exhibited higher binding affinities at the 2C- versus 2A- and 2B-AR subtypes (Table
1), only one of the tethered analogs, the benzyl carboxy alkyl
amine analog (9), exceeded the affinity of the parent compound, yohimbine (1) at the 2C-AR. The benzyl carbamate
alkyl amine (6), the t-butyl carbamate alkyl amine (7) and
the carboxy alkyl amine (10) analogs possessed binding affinities comparable with those of the parent molecule, yohimbine, at the 2C-AR. The alkyl amine analog (8) was 32-fold
less potent in binding to the 2C-AR compared with yohimbine; however, it was 129- and 316-fold less potent in binding
to the 2A- and 2B-AR subtypes in comparison with yohimbine (Table 1).
In regards to 2C-AR selectivities of the tethered analogs, the benzyl carbamate alkyl amine analog (6) and the
alkyl amine analog (8) possessed affinities for the 2C-AR,
which were 11- and 59-fold and 7- and 72-fold greater than

Yohimbine Analogs as Selective 2C-Adrenoceptor Ligands

747

TABLE 3
IC50 values of yohimbine and selected tethered yohimbine analogs for reversing medetomidine effects on forskolin-induced cAMP elevations in
CHO cells stably expressing the 2C-ARs
IC50 and pIC50 values were calculated using GraphPad Prism and expressed as the means S.E.M. of n 4 experiments. Values were determined as the effective
concentration (IC50) and negative log (pIC50) of each compound that reversed the effect of medetomidine on the maximum cAMP response of forskolin.
Compound

pIC50

IC50

pKi

nM

Yohimbine (1)
Yohimbine benzyl carbamate alkyl amine (6)
Yohimbine alkyl amine (8)
a
b

8.90 0.15a
9.31 0.24a
6.87 0.11

1.25
0.48
134

Ki
nM

9.01 0.03b
8.94 0.11b
7.50 0.04

0.97
1.1
31.5

The mean pIC50 of the analog is significantly different from the mean pIC50 value of yohimbine alkyl amine.
The mean pKi of the analog is not significantly different from the mean pIC50 value of the analog (P 0.05 using Students t test).

ligand-receptor interactions in each 2-AR subtype, and

further optimization of the tether length and/or composition
may possibly lead to the design of 2C-AR ligands that will be
more potent than yohimbine. An additional advantage of using
the monovalent ligand approach is that, compared with the
dimeric compounds, the tethered analogs would have improved
physicochemical and pharmacokinetic parameters (Lipinski et
al., 1997).
Acknowledgments

We thank the National Center for Natural Products Research and

the United States Department of Agriculture at the University of
Mississippi.
References
Angers S, Salahpour A, and Bouvier M (2002) Dimerization: an emerging concept for
G protein-coupled receptor ontogeny and function. Annu Rev Pharmacol Toxicol
42:409 435.
Bylund DB, Blaxall HS, Iversen LJ, Caron MG, Lefkowitz RJ, and Lomasney JW
(1992) Pharmacological characteristics of 2-adrenergic receptors: comparison of
pharmacologically defined subtypes with subtypes identified by molecular cloning.
Mol Pharmacol 42:15.
Bylund DB, Bond RA, Clarke DE, Eikenburg DC, Hieble JP, Langer SZ, Lefkowitz
RJ, Minneman KP, Molinoff PB, Ruffolo RR, et al. (1998) Adrenoceptors, in The
IUPHAR Compendium of Receptor Characterization and Classification, 1st ed, pp
58 74, IUPHAR Media Company, Burlington Press, Cambridge, UK.
Cheng Y and Prusoff WH (1973) Relationship between the inhibition constant (KI)
and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an
enzymatic reaction. Biochem Pharmacol 22:3099 3108.
Chotani MA, Flavahan S, Mitra S, Daunt D, and Flavahan NA (2000) Silent 2Cadrenergic receptors enable cold-induced vasoconstriction in cutaneous arteries.
Am J Physiol 278:H1075H1083.
Eason MG, Kurose H, Holt BD, Raymond JR, and Liggett SB (1992) Simultaneous
coupling of 2-adrenergic receptors to two G-proteins with opposing effects: subtype-selective coupling of 2C10, 2C4, and 2C2 adrenergic receptors to Gi and Gs.
J Biol Chem 267:1579515801.
Fairbanks CA, Stone LS, Kitto KF, Nguyen HO, Posthumus IJ, and Wilcox GL (2002)
2C-Adrenergic receptors mediate spinal analgesia and adrenergic-opioid synergy.
J Pharmacol Exp Ther 300:282290.
Hein L, Altman JD, and Kobilka BK (1999) Two functionally distinct 2-adrenergic
receptors regulate sympathetic transmission. Nature (Lond) 402:181184.
Lalchandani SG, Lei L, Zheng W, Suni MM, Moore BM, Liggett SB, Miller DM, and
Feller DR (2002) Yohimbine dimers exhibiting selectivity for the human 2C-AR
subtype. J Pharmacol Exp Ther 303:979 984.
Lipinski CA, Lombardo F, Dominy BW, and Feeney PJ (1997) Experimental and
computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev 23:325.
Mustafa SM, Bavadekar SA, Ma G, Moore BM, Feller DR, and Miller DD (2005)
Synthesis and biological studies of yohimbine derivatives on human 2Cadrenergic receptors. Bioorg Med Chem Lett 15:2758 2760.
Pepperl DJ and Regan JW (1993) Selective coupling of 2-adrenergic receptor subtypes to cyclic AMP-dependent reporter gene expression in transiently transfected
JEG-3 cells. Mol Pharmacol 44:802 809.
Peterhoff M, Sieg A, Brede M, Chao CM, Hein L, and Ullrich S (2003) Inhibition of
insulin secretion via distinct signaling pathways in 2-adrenoceptor knockout
mice. Eur J Endocrinol 149:343350.
Portoghese P (2001) 2000 Alfred Burger Award Address in Medicinal Chemistry:
From models to molecules: opioid receptor dimers, bivalent ligands, and selective
opioid receptor probes. J Med Chem 44:2259 2269.
Scheinin M, Sallinen J, and Haapalinna A (2001) Evaluation of the 2C-adrenoceptor
as a neuropsychiatric drug target: studies in transgenic mouse models. Life Sci
68:22772285.
Vansal SS and Feller DR (1999) An efficient cyclic AMP assay for the functional

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016

their binding to the 2A- and 2B-ARs, respectively. The

benzyl carboxy alkyl amine analog (9) and the carboxy
alkyl amine analog (10) exhibited a 43- and 1995-fold and
295- and 54-fold selectivity in binding to the 2C- versus
2A- and 2B-ARs, respectively. Whether this enhanced
2C-AR selectivity observed with the benzyl carboxy alkyl
amine analog and the carboxy alkyl amine analog represents additional interactions with basic residues in or
around the ligand-binding pocket needs to be further investigated. Our binding results demonstrate that the tethered yohimbine analogs were more 2C-AR selective than
the n 3 dimer, suggesting that the second pharmacophore is not necessary to achieve the 2-AR subtype selectivity with this chemical scaffold. We have also confirmed the
functional antagonist activities of select compounds from
this series at the 2A- and 2C-ARs using luciferase reporter gene assays (Figs. 35). Data from these assays also
revealed that there is an 2C- versus 2A-AR selectivity for
all of the tethered analogs tested (Fig. 6).
In conclusion, this study has yielded tethered analogs of
yohimbine as selective and potent 2C-AR antagonists.
This series of compounds may be interpreted as a novel
series of yohimbine analogs with varying substitutions at
the C-16 carbonyl position, yielding ligands with differing
degrees of 2C-AR selectivity. The results also help to
understand the underlying basis for the 2C-AR subtypeselectivity previously observed with the dimeric analogs
and suggest the following: 1) The second pharmacophoric
group, i.e., the second yohimbinic acid molecule, may not
be essential for the 2C-AR subtype-selectivity previously
seen with the dimeric analogs. We have shown that tethered yohimbine analogs, like dimers, exhibit 2C-AR selectivity. 2) Thus, an interaction of the second pharmacophoric group with either one of the extracellular loops on the
same receptor or with a recognition site on a neighboring
receptor, an adjacent site of the ligand-binding pocket on a
single receptor, or an adjacent receptor molecule may not
be the underlying mechanism(s) required for the observed
synergy witnessed with the dimeric compounds, as proposed previously (Portoghese, 2001; Lalchandani et al.,
2002). 3) Instead, we propose that differences in the physicochemical properties of the amino acid residues (e.g.,
steric or hydrogen bonding) constituting and surrounding
the putative ligand-binding domain, are unique in each
subtype, and this distinction can be exploited to develop
receptor subtype selectivity. 4) Systematic alterations of
the nature of the tether in a ligand on the yohimbine
molecule scaffold may be crucial in maximizing unique

748

Bavadekar et al.

evaluation of -adrenergic receptor ligands. J Recept Signal Transduct Res 19:

853 863.
Virtanen R, Savola JM, Saano V, and Nyman L (1988) Characterization of the
selectivity, specificity and potency of medetomidine as an 2-adrenoceptor agonist.
Eur J Pharmacol 150:9 14.
Zheng W (1999) Novel 3- and 2-Adrenergic Receptor Ligands. Doctoral dissertation, Graduate Studies Council, University of Tennessee Health Science Center,
Memphis, TN.
Zheng W, Lei L, Lalchandani S, Sun G, Feller DR, and Miller DD (2000) Yohimbine

dimers exhibiting binding selectivities for human 2a- versus 2b-adrenergic receptors. Bioorg Med Chem Lett 10:627 630.

Address correspondence to: Dr. Dennis R. Feller, 303 Faser Hall, Department of Pharmacology and National Center for Natural Products Research,
School of Pharmacy, The University of Mississippi, University, MS 38677.
E-mail: dfeller@olemiss.edu

Downloaded from jpet.aspetjournals.org at ASPET Journals on October 13, 2016