Synthesis of bivalent β2-adrenergic and adenosine A 1 receptor ligands

Article abstract of DOI:10.1021/jm800613s

Research in the area of simutaneously targeting more than one G protein-coupled receptor (GPCR) has increased in recent times. By exploiting the cross talk between the β₂-adrenergic (β₂AR) and adenosine A₁ receptors (A₁AR) on adenylate cyclase activity, we synthesized a series of bivalent agonists for both GPCRs to generate responses from more than one receptor. We have demonstrated a relationship between the various β₂-adrenergic and A₁ adenosine bivalent parameters of linker and bifunctionality by using data that are drawn from in vitro assays. The hexyl-linked 12e (K_i, 311 nM) and butyl-linked 12c (K_i, 863 nM) bivalent compounds displayed reasonable binding affinities for the β₂AR when compared with the control (-)isoproterenol (K_i, 136 nM), and both compounds also exhibited a persuasive bifunctional trend for both receptors at various drug concentrations. The bivalent compound 12e was also found to have significant EC₅₀ potency (6 nM) at the β₂AR in DDT cells.

Full text of DOI:10.1021/jm800613s

6128
J. Med. Chem. 2008, 51, 6128–6137
Synthesis of Bivalent ꢀ2-Adrenergic and Adenosine A1 Receptor Ligands
Peter Karellas,†,‡ Michael McNaughton,† Stephen P. Baker,*,§ and Peter J. Scammells*,†
Medicinal Chemistry and Drug Action, Monash Institute of Pharmaceutical Sciences, Monash UniVersity, 381 Royal Parade,
ParkVille VIC 3052, Australia, Starpharma Pty. Ltd., LeVel 6, Baker Heart Research Building, Commercial Road, Melbourne, VIC, Australia,
and Department of Pharmacology and Therapeutics, UniVersity of Florida College of Medicine, Box 100267, GainesVille, Florida 32610
ReceiVed May 23, 2008
Research in the area of simutaneously targeting more than one G protein-coupled receptor (GPCR) has
increased in recent times. By exploiting the cross talk between the ꢀ2-adrenergic (ꢀ2AR) and adenosine A1
receptors (A1AR) on adenylate cyclase activity, we synthesized a series of bivalent agonists for both GPCRs
to generate responses from more than one receptor. We have demonstrated a relationship between the various
ꢀ2-adrenergic and A1 adenosine bivalent parameters of linker and bifunctionality by using data that are
drawn from in vitro assays. The hexyl-linked 12e (Ki, 311 nM) and butyl-linked 12c (Ki, 863 nM) bivalent
compounds displayed reasonable binding affinities for the ꢀ2AR when compared with the control
(-)isoproterenol (Ki, 136 nM), and both compounds also exhibited a persuasive bifunctional trend for both
receptors at various drug concentrations. The bivalent compound 12e was also found to have significant
EC50 potency (6 nM) at the ꢀ2AR in DDT cells.
Introduction
A bivalent ligand is a single chemical entity that is composed
of two covalently linked pharmacophores. There are two general
types of bivalent ligands: homobivalent, where the two phar-
macophores are the same, and heterobivalent, where the two
pharmacophores are different.1-3 Heterobivalent ligands can
have pharmacophores that bind to the same or to different
molecular targets.2-4 Bivalent compounds can have enhanced
receptor subtype selectivity,1,2,5 and in the case where target
cross-linking has been implicated, they have an enhanced affinity
and can be used to estimate the distance between targets or their
spatial distribution.1,3,6,7 Furthermore, bivalent ligands have been
shown to increase biological activity through the activation of
different receptors that mediate the same effect and, in the case
of an agonist/antagonist bivalent ligand for the same receptor,
the ability to produce partial agonism.2,4 From a therapeutic
perspective, bivalent ligands may have pharmacokinetic and
efficacy advantages compared with multiple drug regimens.
The cross talk between different G protein-coupled receptors
(GPCRs)a allows the regulation of cellular responses from
several extracellular mediators. A classic example of receptor
cross talk is the bidirectional effects on adenylate cyclase
activity. There are a number of GPCRs that either stimulate or
inhibit adenylate cyclase through the activation of G proteins
Gs or Gi, respectively, and in some cases, the Gi coupled
receptors can inhibit the ability of Gs coupled receptors to
stimulate the enzyme.8 In several cell types, the activation of
ꢀ-adrenergic receptors (ꢀAR) stimulates adenylate cyclase,
which is attenuated by the simultaneous activation of adenosine
A1 receptors (A1AR).9-11 This cross talk provides fine control
of cAMP levels and may have physiological consequences. For
Figure 1. Bivalent ligands that target the ꢀ2-adrenergic and adenosine
A1 receptors.
example, cardiac ꢀ-adrenergic responsiveness is decreased
during aging, and part of this decrease appears to be due to
enhanced adenosine formation and the activation of A1ARs with
a resulting attenuation in ꢀ responsiveness.12,13 In addition, it
has been shown that certain cardiac arrhythmias that are initiated
at high concentrations of a full ꢀ-agonist can be blunted by the
activation of the A1AR without affecting the ꢀAR-mediated
increase in contractility.14
On the basis of the ꢀ2AR/A1AR cross talk, the objective of
the present study is to synthesize a series of bivalent ꢀ2AR/
A1AR agonists and to determine the effects of alkyl and ether
spacers on their ability to activate both receptors and to produce
an interactive response (Figure 1). The pharmacophores are
based upon the well-known ꢀ2-agonist formoterol and the
endogenous A1AR agonist adenosine. The spacer connection
to each pharmacophore is important for the retention of affinity
and activity. Therefore, the chosen linkage was between the
side-chain amino of the ꢀ-agonist and the N6 position of
adenosine because large substituents in these positions have been
shown to be well tolerated.15,16 We characterized the bivalent
* To whom correspondence should be addressed. E-mail: Peter.Scammells@
vcp.monash.edu.au. Tel: +61 (0)3 9903 9542. Fax: +61 3 9903 9582. (P.J.S.)
E-mail: sbaker@pharmacology.ufl.edu. Tel: +1 352 392 3519 (S.P.B.).
†
Monash University.
Starpharma Pty. Ltd.
University of Florida College of Medicine.
‡
§
a Abbreviations: A1AR, adenosine A1 receptor; ꢀ2AR, ꢀ2-adrenergic
receptor; GPCR, G protein-coupled receptor; DPCPX, 1,3-dipropyl-8-
cyclopentylxanthine; ScAMP-TME, 2′-O-monosuccinyladenosine 3′:5′-
cyclic monophosphate tyrosyl methyl ester.
10.1021/jm800613s CCC: $40.75
2008 American Chemical Society
Published on Web 09/11/2008

ꢀ2-Adrenergic and Adenosine A1 Receptor Ligands
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19 6129
Scheme 1a
a Reagents and conditions: (i) 1 with 2b in toluene/THF and LiClO4, 24 h, 110 °C; (ii) TFA in CH2Cl2, 2 h, rt (4 was isolated as the TFA salt); (iii)
6-chloropurine riboside (5), N(i-Pr)2Et in t-BuOH, 80 °C, 24 h; (iv) PtO2, H2, (1 atm) in CH3OH, rt, 9 h; and (v) Ac2O, HCOOH, 12 h, rt.
Scheme 2a
a Reagents and conditions: (i) 1 with 2 in toluene/THF and LiClO4, 24 h, 110 °C; (ii) PtO2, H2, 1 atm in CH3OH; (iii) Ac2O, HCOOH, 12 h, rt; (iv)
TFA/CH2Cl2 (product isolated as the TFA salt); (v) 6-chloropurine riboside (5), N(i-Pr)2Et in t-BuOH, reflux 24 h; (vi) 10% Pd/C, H2, 50 psi in EtOH.
compounds by using DDT1 MF-2 cells that express both receptor
types and in which the activation of the A1AR inhibits ꢀ2AR-
mediated increases in cAMP accumulation.11,17
ranged from 51-71% and were utilized as nucleophiles, as
shown in Schemes 1-3.
The reaction between N-benzyl amine 2b and epoxide 1
initially proved to be quite problematic. Several conditions19,20,22
were evaluated, and in all cases, these reactions returned
unreacted starting material. However, when the epoxide precur-
sor was treated with N-benzylamine 2b in the presence of excess
LiClO4,23 the desired ethanolamine 3b was obtained in reason-
able yield (Scheme 1).
Results and Discussion
Chemistry. The desired starting material, 4-benzyloxy-3-
nitrostyreneoxide (1), was successfully prepared on a multigram
scale according to literature methodology.18-20 Also, a series
of diamine linkers that were both (mono-) Boc-protected and
N-benzylated were synthesized via reductive alkylation of the
corresponding Boc-protected diamines.21 All linkers (compounds
2a-c) were purified by column chromatography in yields that
The t-butoxycarbonyl (Boc) group of 3b was cleaved by
treatment with trifluoroacetic acid, and the desired product 4
was isolated as the TFA salt following the evaporation of the

6130 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19
Karellas et al.
Scheme 3a
a Reagents and conditions: (i) BnNHCH2CH2(OCH2CH2)2NHBoc in toluene/THF and LiClO4, 24 h, 110 °C; (ii) PtO2, H2, 1 atm in CH3OH; (iii) Ac2O,
HCOOH, rt; (iv) TFA/CH2Cl2 (product isolated as the TFA salt); (v) 6-chloropurine riboside (5), N(i-Pr)2Et in t-BuOH, 110 °C, 24 h; (vi) 10% Pd/C, H2,
50 psi in EtOH.
excess trifluoroacetic acid. This amine was subsequently reacted
with 6-chloropurine riboside (5) in the presence of Hünig’s base
to provide N6-substituted adenosine 6. The nitro group of 6 was
reduced via catalytic hydrogenation (PtO2 and hydrogen gas at
1 atm) to yield the corresponding aniline derivative 7a. The
formylation of this aniline was attempted by treatment with an
aged mixture of formic acid and acetic anhydride.25 Although
standard catalytic hydrogenation conditions (palladium on a
carbon catalyst under an atmosphere of hydrogen gas at 50 psi).
An attempt was made to purify the targets by preparatory TLC
with various mixtures of MeOH/H2O, but the targets proved to
be difficult to isolate as a result of their high polarity. The
purification of these polar compounds was subsequently achieved
by preparatory HPLC. A reliable and reproducible HPLC
method was developed, and compounds 12a, 12c, and 12e were
successfully purified, albeit in low yield. The corresponding
deformylated compounds 12b and 12d, were also isolated in
low yield as byproducts in the synthesis of 12a and 12c. No
attempts were made to optimize these yields because sufficient
material was obtained for the analysis and the pharmacological
evaluation. The deformylated byproducts 12b and 12d possess
a ꢀ2AR component that is analogous to the known ꢀ2AR agonist
desformoterol26 and were also evaluated as bivalent ligands.
1
characteristic formyl peaks appeared in the crude H and 13C
NMR spectra, only starting material was isolated following
column chromatography. This led us to believe that formyl ester
derivative 7c formed and was further hydrolyzed on the silica
column to return the original starting material. Because the
introduction and the purification of the formyl group in the
presense of the adenosine component proved to be problematic,
an alternative strategy was adopted in which the formyl group
was introduced prior to the attachment of the adenosine
component (Scheme 2).
This modified approach allowed us to synthesize our final
targets in the same number of steps that was set out in our initial
plan and to take advantage of a number of the synthetic steps
that had already been optimized. Furthermore, the introduction
of the adenosine component later in the sequence would also
minimize the handling of polar products. A starting point for
this approach involved the preparation of ethanolamine deriva-
tives 3a-c from epoxide 1 (Scheme 2). The nitro groups of
compounds 3a-c were subsequently reduced with H2 and PtO2
to afford the corresponding anilines 8a-c, respectively. The
formyl group was successfully introduced with an excess
mixture of acetic acid and acetic anhydride (on a per aniline
basis). 1H and 13C NMR spectroscopy showed peaks at 8.3 and
162 ppm, respectively, which is characterisitic of the formyl
group for all analogs. The Boc protecting group of compounds
9a-c was then cleaved by treatment with trifluoroacetic acid
to afford the corresponding amines as TFA salts in quantitative
yield. The subsequent incorporation of the adenosine component
was achieved by the reaction of the amine moiety of compounds
10a-c with 6-chloropurine riboside in the presence of Hünig’s
base in t-BuOH. In the final step of the synthesis, the benzyl
protecting groups were concomitantly cleaved by the use of
The incorporation of polyethylene glycol units is a common
approach for improving the water solubility of small mol-
ecules.27 Accordingly, a bivalent ligand with a polyethylene
glycol linker that joined the formoterol and adenosine compo-
nents was also targeted. Compound 18 was prepared in seven
steps from 4-benzyloxy-3-nitrostyreneoxide (1) according to the
approach described above (Scheme 3).
Pharmacology. The bivalent compounds were subjected
to several pharmacological assays whereby the affinities of
the bivalent ligands were determined for the ꢀ2AR and the
A1AR by displacement of specific [125I]-(-)iodopindolol and
[3H]DPCPX binding, respectively, in DDT cell membranes.
These assays included 10 µM 5′-guanylyl-imidodiphosphate
to maintain the receptor in the agonist low affinity state. The
compounds were also tested for their ability to stimulate
(ꢀ2AR effect) and inhibit (A1AR effect) cAMP accumulation
in DDT cells.11,17 In these cells, the stimulation of cAMP
accumulation that is mediated by ꢀ-agonists is attenuated by the
activation of the A1AR. Therefore, in the present experiments, the
effects of the bivalents were determined alone and in the
presence of the A1AR antagonist DPCPX to block any inhibitory
effects. The effects of the bivalent ligands were compared with

ꢀ2-Adrenergic and Adenosine A1 Receptor Ligands
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19 6131
Table 1. Ki and EC50 Values for Bivalent Derivatives at the
ꢀ2-Adrenergic and Adenosine A1 Receptors in DDT1 MF-2 Cells
linked aniline bivalent (12d) produced a biphasic cAMP
response with stimulation that occurred in concentrations up to
1 µM and inhibition that occurred at higher concentrations. The
maximal stimulation was 50% of the (-)isoproterenol maxi-
mum. In the presence of DPCPX, the inhibitory phase was
abolished, and the maximal stimulation was similar to that of
(-)isoproterenol. The stimulation of cAMP accumulation that
was produced by 0.1 µM 12c and 12e or 1 µM 12c and 18 was
blocked by the inclusion of 0.1 µM propranolol, which indicates
that this response was mediated by the ꢀ2AR (data not shown).
Table 1 shows the potencies (EC50) of the bivalent ligands
for the ꢀ2AR stimulation of cAMP accumulation in the presence
of DPCPX, and in general, they are greater than the corre-
sponding affinities. The butyl-linked (12c) and hexyl-linked
(12e) derivatives have potencies in the low nanomolar range,
and the butyl derivative has a slightly lower and the hexyl
derivative has a slightly higher potency than that of (-)isopro-
terenol (20 nM). In contrast, the ether-linked (18) and butyl-
linked aniline (12d) derivatives have potencies that are 9 and
54 times lower than that of (-)isoproterenol, respectively. The
EC50 for the ethyl-linked (12a) and ethyl-linked aniline (12b)
bivalents could not be determined because a plateau for cAMP
accumulation was not reached at the highest concentration that
was used for each compound.
compd
Ki A1ARa (nM) Ki ꢀ2ARa (nM) EC50 ꢀ2ARb (nM)
isoproterenol
CPA
12a
12b
12c
12d
12e
18
136 ( 25(6)
20 ( 4(14)
9 ( 2(5)
1979 ( 256(6)
2749 ( 334(5)
1914 ( 190(5)
1014 ( 67(4)
436 ( 63(5)
>10 000(3)c
>10 000(3)c
863 ( 196(4)
4314 ( 1359(4)
311 ( 62(4)
NDd
NDd
32 ( 4(5)
1087 ( 51(3)
6 ( 2(4)
1907 ( 357(5)
2296 ( 355(4)
177 ( 37(3)
a Ki values were calculated from the concentration of the ligands that
inhibited specific [3H]DPCPX binding to the A1AR or [125I]-(-)iodopindolol
binding to the ꢀAR by 50%. b EC50 values are the concentration of ligands
that gave half-maximal stimulation of cAMP accumulation in the presence
of 1 µM DPCPX. c Less than 20% inhibition of radioligand binding at the
highest concentration (10 µM) used. d ND is not determined. A maximal
stimulation was not achieved with up to a 100 µM compound. Numbers in
parentheses are the number of separate experiments.
those of the classical ꢀ-agonist (-)isoproterenol and the A1AR
agonist N6-cyclopentyladenosine (CPA).
As shown in Table 1, the affinities of the ethyl-linked (12a)
and butyl-linked (12c) bivalent ligands for A1AR are similar
and are in the low micromolar range. Increasing the linker length
with the hexyl (12e) bivalent increased the affinity by about
4.5 times to 436 nM. The ether-linked bivalent (18) has an A1AR
affinity that is similar to that of the ethyl- and butyl-linked
derivatives. The ethyl-linked (12b) and butyl-linked (12d)
aniline bivalent compounds also had affinities that were in the
low micromolar range. All of the bivalents had substantially
lower affinites for the A1AR as compared with the A1AR agonist
CPA. At the ꢀ2AR, there is a linker-related effect on affinity.
Therefore, the ethyl (12a) bivalent has an affinity that is greater
than 10 µM, which is increased to 863 nM with the butyl (12c)
derivative and is further increased to 311 nM with the hexyl
(12e) derivative. Furthermore, the ethyl-linked aniline bivalent
(12b) has an affinity that is greater than 10 µM, and the butyl-
linked derivative (12d) has an increased affinity at 4.3 µM. The
ether-linked bivalent (18) also has a low micromolar affinity
for ꢀ2AR.
The effects of the bivalent ligands on cAMP accumulation
are shown in Figure 2. The ethyl-linked (12a) bivalent showed
less than 15% stimulation of cAMP accumulation at the highest
concentration that was employed (100 µM) as compared with
(-)isoproterenol, and this stimulation was increased only slightly
in the presence of DPCPX (Figure 2A). In contrast, the butyl
(12c) and hexyl (12e) derivatives produced a biphasic cAMP
accumulation response. A concentration-dependent stimulation
of cAMP accumulation occurred in the range of 1-100 nM,
after which there was a concentration-dependent inhibition of
cAMP accumulation (Figure 2A,B). The peak cAMP accumula-
tion effect for both compounds was comparable to that of
(-)isoproterenol, and the inhibitory phase of cAMP accumula-
tion was completely prevented by the inclusion of 1 µM
DPCPX, which resulted in a stimulation plateau. This indicates
that the inhibitory phase is mediated by activation of the A1AR.
The ether-linked bivalent (18) also showed a biphasic effect on
cAMP accumulation (Figure 2C) with a maximal stimulation
that was 70% the (-)isoproterenol maximum. In the presence
of DPCPX, the inhibitory phase was abolished and the maximal
stimulation was increased to that observed with (-)isoproterenol.
As shown in Figure 2D, the ethyl-linked aniline bivalent (12b)
produced less than 15% stimulation of cAMP accumulation at
100 µM as compared with (-)isoproterenol. In the presence of
DPCPX, the stimulation by this derivative increased to over
50% of the (-)isoproterenol maximum. In contrast, the butyl-
The data from the present study show that linking agonist
pharmacophores for the ꢀ2AR and the A1AR into a single ligand
can result in the retention of activity at both receptor types.
Within the limited bivalent synthesized series, the affinity of
the ligands for the A1AR was relatively insensitive to the linker
length. In contrast, for the ꢀ2AR, there was a modest increase
in affinity and potency with alkyl linker length. Interestingly,
the ether-linked bivalent (18), which has the same number of
methylene groups as 12e, had a decrease in ꢀ2AR affinity and
potency, which suggests that the length, chemical nature, and
conformational flexibility of the linker can also affect these
parameters, as has been shown for other bivalent ligands.1,6,28
The data show affinity and agonist activity of the bivalent
compounds at each receptor type; however, the data do not
directly address the notion of receptor cross-linking by a single
bivalent molecule. The lack of major changes in the affinity of
the bivalent ligands, especially with the A1AR, suggests that
receptor cross-linking is unlikely.
With the exception of the two bivalent compounds with
relatively weak affinities for the ꢀ2AR (12a and 12b), all of
the others produced the same maximal stimulation of cAMP
accumulation as the classical ꢀ-agonist (-)isoproterenol when
the inhibitory phase was blocked by DPCPX. This indicates
that they are full ꢀ-agonists. Furthermore, the maximal inhibition
of cAMP accumulation by these compounds is 75-85%, which
has been shown to be the maximum produced by the classical
A1AR agonist CPA; this suggests that they are also full agonists
at the A1AR (Figure 2). In the absence of DPCPX, two of the
bivalent compounds (12b and 18) had reduced maximal
responses compared with (-)isoproterenol. These suppressed
maxima were due to the concurrent activation of the A1AR,
which resulted in an apparent (physiological) partial ꢀ-agonist
response of the bivalent ligand. In DDT cells the A1AR-
mediated inhibition of (-)isoproterenol-stimulated cAMP ac-
cumulation is dominant because it occurs even when the ꢀ2AR
are saturated with agonist.11,24 Therefore, the bivalent concen-
tration-dependent relationship between the ꢀ2AR-mediated
stimulation of cAMP accumulation followed by the A1AR-
mediated inhibition phase is likely due to several factors
including the differential affinity and the relationship between

6132 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19
Karellas et al.
Figure 2. Effect of bivalent compounds and (-)isoproterenol on cAMP accumulation in DDT cells.
receptor occupancy and response (efficacy). Compounds 12c
and 12e have 2.2- and 1.4-fold higher affinities for the ꢀ2AR
than for the A1AR, respectively whereas 18 and 12d have 1.8-
and 4.4-fold higher affinities for the A1AR than for the ꢀ2AR,
respectively. Therefore, the compounds with higher affinities
for the ꢀ2AR may produce higher maximum for cAMP
accumulation compared with those that have slightly higher
affinities for the A1AR (Figure 2). In addition, agonists with
higher efficacies will have nonlinear relationships (hyperbolic)
between the receptor occupancy and response such that low
receptor occupancy can achieve high levels of response. This
effect will also shift the concentration response to the left of
the occupancy curve, which will give the agonist a higher
observed potency. The efficacy or shift in response from
occupancy curves can be estimated by the Ki/EC50 ratio.
According to the data from Table 1, the ratio for (-)isoprot-
erenol is 6.8, whereas the ratios for 12c and 12e are 26.9 and
51.8, respectively. The relatively high ratios for 12c and 12e
indicate that a relatively small fraction of occupied ꢀ2ARs will
produce a maximum response and may partially explain why
these two compounds can produce a maximal response (com-
pared with (-)isoproterenol) before the inhibitory phase occurs.
In contrast, the Ki/EC50 ratio for 18 is 12.9 with a maximum
response that is 70% of the (-)isoproterenol maximum, whereas
the ratio for 12d is 3.9 with a 50% maximum response. The
reduced ratios for these compounds suggest that they will need
to occupy a greater fraction of the ꢀ2ARs to achieve a maximum
response, which is suppressed because of the concurrent
occupancy and activation of the A1ARs. It should be pointed
out that the efficacy for the bivalent-mediated A1 inhibitory
response will also affect the biphasic response relationship but
the Ki/EC50 ratios for the A1AR were not calculated because of
the inability to determine an EC50 value accurately.
biphasic cAMP response through receptor cross talk. The linker
length and composition between the two receptor pharmaco-
phores affected binding affinity more at the ꢀ2AR than the
A1AR. Compound 12e, which contains a hexyl spacer, possessed
the highest affinity for both receptors and was the most potend
ꢀ-agonist. The relationship between the ꢀ2AR stimulatory and
A1AR inhibitory responses was likely dependent upon both
receptor affinity and efficacy. Certain compounds (12c and 12e)
produced potent ꢀ2AR-mediated stimulation of cAMP with the
same maximum response as the full agonist (-)isoproterenol,
whereas others (18 and 12d) had an A1AR-mediated suppression
of the maximal stimulatory response such that they appeared
as partial ꢀ-agonists. The data suggest that bivalent agonists
may be useful in the elucidation of the mechanisms that
contribute to the modulation of cellular responses through
receptor cross talk. Furthermore, bivalent ligands may hold
promise in the development of physiological partial agonists
by the differential activation of interacting receptors.
Experimental Section
Chemistry. 6-Chloropurine riboside was purchased from Toronto
Research Chemicals. The mono Boc-protected amine linkers
(structures not shown) were provided at a high degree of purity by
Starpharma Pty. and were converted to secondary amine linkers
(2a-d). Prepacked C18 columns (strata C18-E) were purchased from
Phenomenex. Unless otherwise stated, the solvents were HPLC
grade and were used without further purification. 1H NMR and 13
C
NMR spectra were recorded in MeOD on a Bruker 300 UltraShield
300 MHz NMR instrument. Chemical shifts (δ) are recorded in
parts per million (ppm) relative to MeOD, and coupling constants
(J values) are in Hertz. HPLC/ES-MS was conducted on a Waters
2795 instrument with a 2996 diode array detector (chromatograms
show total UV absorbance 200-300 nm) coupled to a Waters
ZQ4000 instrument with an ESI probe and inlet flow split to give
around 50 µL/min to the MS. The analytical chromatography
column was a Waters Xterra C18 (hydrophilic) (0.3 × 100 mm2),
and unless otherwise stated, TFA method ACN/water (0.1% TFA)
gradients at 0.4 mL/min were utilized. High-resolution electrospray
mass spectra (HRMS) studies were conducted on a Bruker Bio-
Conclusions
A series of bivalent ꢀ2AR/A1AR agonists were synthesized,
and several were shown to produce a concentration-dependent

ꢀ2-Adrenergic and Adenosine A1 Receptor Ligands
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19 6133
Apex II FTMS instrument. High-resolution ES-MS data were
obtained for all targeted compounds (12a-e and 18). Preparatory
HPLC was carried out on a Waters Xterra prep column (RP18, 10
µm, 19 × 250 mm2). We carried out thin layer chromatography
by using 20 cm plates (Merck silica gel 60 F254), and we conducted
column chromatography by using Merck silica gel 60 (particle size
0.04-0.063 nm, 230-400 mesh).
2-(Benzyl(4-(N6-adenosinyl)butyl)amino)-1-(4-(benzyloxy)-3-ni-
trophenyl)ethanol (6). To a suspension of 6-chloropurine riboside
(5) (107 mg, 0.37 mmol) in t-BuOH (40 mL) was added 4 (175
mg, 0.31 mmol), followed by DIPEA (135 µL, 0.78 mmol). The
reaction mixture was stirred for 24 h at 80 °C under nitrogen.
Evaporation of the solvent and purification of the residue by flash
silica chromatography (eluent 10% MeOH/CH2Cl2 + 1% NH3)
produced title compound 6 as a yellow oil (106 mg) in 49% yield.
1H NMR (MeOD, δ): 1.49-1.63 (m, 4H, 2 × CH2), 2.49-2.58
(m, 2H, CH2), 2.60-2.69 (m, 2H, CH2), 3.40-3.56 (m, 2H, CH2),
3.56, 3.64 (d, 2H, J ) 13.2, CH2), 3.69-3.92 (m, 2H, CH2),
4.14-4.20 (m, 1H, CH), 4.32 (dd, 1H, J ) 2.7, J ) 5.0, CH), 4.66
(t, 1H, J ) 6.6, CH), 4.74 (t, 1H, J ) 5.4, CH), 5.18 (s, 2H, CH2),
5.94 (d, 1H, J ) 6.6, CH), 7.10-7.48 (m, 12H, ArH), 7.71 (br s,
tert-Butyl 2-(Benzyl(2-(4-(benzyloxy)-3-nitrophenyl)hydroxyeth-
yl)amino)ethylcarbamate (3a). To a stirred solution of 4-benzyloxy-
3-nitrostyreneoxide (1) (343 mg, 1.27 mmol) in toluene/THF (30:
15 mL) was added 2a (396 mg, 1.58 mmol) in toluene (30 mL),
followed by LiClO4 (671 mg, 6.33 mmol). The reaction was stirred
at 110 °C for 24 h under an atmosphere of nitrogen. The solvent
was removed under reduced pressure, and the residue was purified
by flash silica chromatography (eluent 3% MeOH/CH2Cl2) to
produce the title compound as a yellow oil (310 mg, 47% yield).
1H NMR (MeOD, δ): 1.42 (s, 9H, 3 × CH3), 2.61 (t, 2H, J ) 6.3,
CH2), 2.68 (dd, 2H, J ) 4.2, J ) 6.9, CH2), 3.00-3.20 (m, 2H,
CH2), 3.66 (s, 2H, CH2), 4.68 (t, 1H, J ) 6.6, CH), 5.26 (s, 2H,
1H, ArH), 8.17 (s, 1H, CH-purine), 8.20 (s, 1H, CH-purine). 13
C
NMR (MeOD, δ): 25.4, 28.2, 41.4, 55.3, 60.3, 62.9, 63.5, 71.3,
72.2, 71.7, 75.5, 88.2, 91.4, 116.2, 121.4, 124.3, 128.1, 128.3, 129.1,
129.2, 129.6, 130.2, 133.0, 137.6, 138.1, 140.3, 141.3, 141.4, 149.0,
152.0, 153.6, 156.3. LCMS (hydrophilic): Rf (min) ) 14.87, (ESI
+ ve) found 700 [M + H]+ calcd for C36H42N7O8.
CH2), 7.16-7.52 (m, 12H, ArH), 7.75 (d, 1H, J ) 2.1, ArH). 13
C
2-(Benzyl(4-(N6-adenosinyl)butyl)amino)-1-(4-(benzyloxy)-3-ami-
nophenyl)ethanol (7a). To compound 6 (106 mg, 0.15 mmol)
dissolved in methanol (6 mL) was added PtO2 (10 mg), and the
reaction mixture was stirred for 9 h under an atmosphere of
hydrogen gas (at 1 atm). Filtration through a Millipore 0.45 µm
filter and evaporation of the solvent afforded a residue that was
purified by flash silica chromatography (eluent 5-10% MeOH/
CH2Cl2). The title compound was obtained as a yellow gum (90
NMR (MeOD, δ): 28.1, 28.8, 39.4, 55.5, 60.6, 63.2, 71.5, 72.2,
80.1, 116.4, 124.2, 128.1, 128.3, 129.2, 129.3, 129.6, 130.2, 132.9,
137.6, 138.0, 140.4, 141.4, 152.1, 158.5. LCMS (hydrophilic): Rf
(min) ) 12.17, (ESI + ve) found 522 [M + H]+ calcd for
C29H36N3O6.
tert-Butyl 4-(Benzyl(2-(4-(benzyloxy)-3-nitrophenyl)hydroxyeth-
yl)amino)butylcarbamate (3b). The title compound was prepared
using the same method that was described for 3a. After purification
by column chromatography (eluent 3% MeOH/CH2Cl2), the product
1
mg) in 89% yield. H NMR (MeOD, δ): 1.40-1.65 (m, 4H, 2 ×
CH2), 2.42-2.74 (m, 4H, 2 × CH2), 3.40-3.93 (m, 6H, 3 × CH2),
4.15-4.19 (m, 1H, CH), 4.32 (dd, 1H, J ) 2.4, J ) 4.8, CH),
4.59-4.66 (m, 1H, CH), 4.74 (t, 1H, J ) 5.7, CH), 5.04 (s, 2H,
CH2), 5.94 (d, 1H, J ) 6.3, CH), 6.70-6.90 (m, 2H, ArH),
7.10-7.50 (m, 11H, ArH), 8.18 (br s, 1H, CH), 8.21 (br s, 1H,
CH). 13C NMR (MeOD, δ): 25.2, 28.2, 41.4, 55.0, 60.1, 63.1, 63.6,
71.5, 72.2, 72.8, 75.5, 88.3, 91.4, 113.2, 115.0, 117.7, 121.5, 128.2,
128.6, 128.9, 129.3, 129.5, 130.4, 137.4, 137.8, 138.9, 140.0, 141.4,
147.6, 149.0, 153.6, 156.3. LCMS (hydrophilic): Rf (min) ) 8.01,
(ESI + ve) found 670 [M + H]+ calcd for C36H44N7O6.
1
was obtained as a yellow foam in 73% yield. H NMR (MeOD,
δ): 1.35-1.55 (m, 4H, 2 × CH2), 1.41 (s, 9H, 3 × CH3), 2.58 (t,
2H, J ) 6.9, CH2), 2.68 (d, 2H, J ) 6.9, CH2), 2.97 (t, 2H, J )
6.6, CH2), 3.69 (s, 2H, CH2), 4.69 (t, 1H, J ) 6.6, CH), 5.22 (s,
2H, CH2), 7.18-7.45 (m, 12H, ArH), 7.73 (d, 1H, J ) 1.8, ArH).
13C NMR (MeOD, δ): 25.3, 28.7, 28.9, 41.2, 55.4, 60.2, 62.9, 71.3,
72.2, 79.9, 116.3, 124.2, 128.1, 128.3, 129.1, 129.2, 129.6, 130.2,
133.0, 137.6, 138.1, 140.5, 141.3, 152.0, 158.5. LCMS (hydro-
philic): Rf (min) ) 9.05, (ESI + ve) found 550 [M + H]+ calcd
for C31H40N3O6.
tert-Butyl 2-(Benzyl(2-(4-(benzyloxy)-3-aminophenyl)hydroxy-
ethyl)amino)ethylcarbamate (8a). To compound 3a (280 mg, 0.54
mmol) dissolved in methanol (40 mL) was added PtO2 (28 mg),
and the reaction mixture was stirred for 2 h under an atmosphere
of hydrogen gas (at 1 atm). Filtration through a Millipore 0.45 µm
filter and evaporation of the solvent afforded a residue that was
purified by flash silica chromatography (eluent 5-10% MeOH/
CH2Cl2). The title compound was obtained as a yellow oil (240
tert-Butyl 6-(Benzyl(2-(4-(benzyloxy)-3-nitrophenyl)hydroxyeth-
yl)amino)hexylcarbamate (3c). The title compound was prepared
using the same method that was described for 3a. After purification
by column chromatography (eluent 3% MeOH/CH2Cl2), the product
1
was obtained as a yellow foam in 62% yield. H NMR (MeOD,
δ): 1.17 (br s, 4H, 2 × CH2), 1.38 (br s, 4H, 2 × CH2), 1.41 (s,
9H, 3 × CH3), 2.44 (t, 2H, J ) 7.1, CH2), 2.60 (d, 2H, J ) 6.6,
CH2), 2.97 (t, 2H, J ) 7.1, CH2), 3.56 (s, 2H, CH2), 4.64 (t, 1H, J
) 6.6, CH), 5.16 (s, 2H, CH2), 7.12-7.45 (m, 12H, ArH), 7.74 (d,
1H, J ) 2.1, ArH). 13C NMR (MeOD, δ): 27.7, 28.0, 28.9, 30.9,
41.3, 55.5, 60.1, 62.9, 71.0, 72.1, 79.7, 116.1, 124.2, 128.0, 128.2,
129.1, 129.2, 129.6, 130.1, 133.0, 137.4, 137.9, 140.4, 141.1, 151.9,
158.3. LCMS (hydrophilic): Rf (min) ) 19.04, (ESI + ve) found
578 [M + H]+ calcd for C33H44N3O6.
1
mg) in 91% yield. H NMR (MeOD, δ): 1.43 (s, 9H, 3 × CH3),
2.45-2.75 (m, 4H, 2 × CH2), 3.00-3.20 (m, 2H, CH2), 3.58, 3.75
(d, 2H, J ) 13.5, CH2), 4.55 (dd, 1H, J ) 4.5, J ) 8.5, CH), 5.05
(s, 2H, CH2), 6.61 (dd, 1H, J ) 1.8, J ) 8.2, ArH), 6.76 (d, 1H, J
) 1.8, ArH), 6.81 (d, 1H, J ) 8.4, ArH), 7.18-7.46 (m, 10H, ArH).
13C NMR (MeOD, δ): 28.9, 39.4, 55.2, 60.4, 63.6, 71.5, 72.4, 80.0,
113.2, 114.8, 117.6, 128.1, 128.6, 128.9, 129.3, 129.5, 130.2, 137.3,
137.8, 138.8, 140.3, 147.5, 158.5. LCMS (hydrophilic): Rf (min)
) 7.39, (ESI + ve) found 492 [M + H]+ calcd for C29H38N3O4.
tert-Butyl 4-(Benzyl(2-(4-(benzyloxy)-3-aminophenyl)hydroxy-
ethyl)amino)butylcarbamate (8b). The title compound was prepared
using the same method that was described for 8a. Following flash
chromatography (eluent 10% MeOH/CH2Cl2 + 1% NH3), the
product was isolated as a yellow oil in 47% yield. 1H NMR (MeOD,
δ): 1.30-1.50 (m, 4H, 2 × CH2), 1.43 (s, 9H, 3 × CH3), 2.38-2.70
(m, 4H, 2 × CH2), 2.97 (t, 2H, J ) 6.6, CH2), 3.57, 3.70 (d, 2H,
J ) 13.5, CH2), 4.56 (dd, 1H, J ) 5.4, J ) 7.5, CH), 5.07 (s, 2H,
CH2), 6.61 (dd, 1H, J ) 2.1, J ) 8.1, ArH), 6.75 (d, 1H, J ) 1.8,
2-((4-Aminobutyl)benzylamino)-1-(4-(benzyloxy)-3-nitrophe-
nyl)ethanol: TFA Salt (4). A solution of trifluoroacetic acid/
dichloromethane (1:1) (590 µL) was added dropwise to a stirred
suspension of 3b (210 mg, 0.38 mmol) in dichloromethane (4 mL).
The mixture was stirred at room temperature for 2 h under argon. The
solvent and excess TFA were removed under reduced pressure. The
residue was redissolved in MeOH (40 mL), was concentrated in vacuo,
and was freeze-dried to produce title compound 4 as the TFA salt
1
(200 mg, 93% yield). H NMR (MeOD, δ): 1.62-1.80 (m, 2H,
CH2), 1.83-2.09 (m, 2H, CH2), 3.00 (t, 2H, J ) 7.2, CH2),
3.24-3.33 (m, 4H, 2 × CH2), 4.42-4.60 (m, 2H, CH2), 5.25 (s,
2H, CH2), 7.26-7.64 (m, 12H, ArH), 7.83 (br s, 1H, ArH). 13C
NMR (MeOD, δ): 21.7, 25.6, 40.0, 56.4, 59.1, 59.4, 68.9, 72.2,
116.9, 123.9, 128.4, 129.3, 129.7, 130.6, 130.7, 131.4, 132.5, 132.6,
135.0, 137.3, 141.6, 152.7. LCMS (hydrophilic): Rf (min) ) 7.32,
(ESI + ve) found 450 [M + H]+ calcd for C26H32N3O4.
ArH), 6.82 (d, 1H, J ) 8.4, ArH), 7.18-7.48 (m, 10H, ArH). 13
C
NMR (MeOD, δ): 25.3, 28.6, 28.9, 41.2, 55.0, 60.1, 63.2, 71.5,
72.2, 79.8, 113.2, 115.0, 117.7, 128.1, 128.6, 128.9, 129.3, 129.5,
130.3, 137.5, 137.7, 138.9, 140.5, 147.5, 158.5. LCMS (hydro-
philic): Rf (min) ) 7.59, (ESI + ve) found 520 [M + H]+ calcd
for C31H42N3O4.

6134 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19
Karellas et al.
tert-Butyl 6-(Benzyl(2-(4-(benzyloxy)-3-aminophenyl)hydroxy-
ethyl)amino)hexylcarbamate (8c). The title compound was prepared
using the same method that was described for 8a. Following flash
chromatography (eluent 5-10% MeOH/CH2Cl2 + 1% NH3), the
product was isolated as a yellow oil in 61% yield. 1H NMR (MeOD,
δ): 1.19-1.26 (m, 4H, 2 × CH2), 1.36-1.52 (m, 4H, 2 × CH2),
1.43 (s, 9H, 3 × CH3), 2.40-2.70 (m, 4H, 2 × CH2), 2.99 (t, 2H,
J ) 7.1, CH2), 3.58, 3.71 (d, 2H, J ) 13.5, CH2), 4.55 (dd, 1H, J
) 5.4, J ) 7.8, CH), 5.10 (s, 2H, CH2), 6.61 (dd, 1H, J ) 2.1, J
) 8.4, ArH), 6.76 (d, 1H, J ) 1.8, ArH), 6.83 (d, 1H, J ) 8.4,
ArH), 7.18-7.50 (m, 10H, ArH). 13C NMR (MeOD, δ): 27.7, 27.9,
28.0, 28.9, 30.9, 41.3, 55.1, 60.0, 63.3, 71.4, 72.0, 79.7, 113.1,
114.7, 117.5, 128.0, 128.5, 128.8, 129.3, 129.5, 130.2, 137.3, 137.7,
138.8, 140.5, 147.4, 158.4. LCMS (hydrophilic): Rf (min) ) 8.16,
(ESI + ve) found 548 [M + H]+ calcd for C33H46N3O4.
tert-Butyl 2-(Benzyl(2-(4-(benzyloxy)-3-formamidophenyl)hy-
droxyethyl)amino)ethylcarbamate (9a). Acetic anhydride (1.02 mL,
10.8 mol) was added to formic acid (406 µL, 10.8 mol) at 0 °C,
and the solution was stirred at 60 °C for 15 min. After cooling to
0 °C, a solution of 8a (240 mg, 0.49 mmol) in CH2Cl2 (15 mL)
was added, and the reaction was stirred at room temperature for
1 h under argon. The solvent was removed under reduced pressure,
and the residue was redissolved in methanol (15 mL). To the
solution was added 1 M NaOH (1.47 mL, 1.47 mmol), and the
mixture was left to stir for 30 min. The solvent was removed, and
the residue was redissolved in ethyl acetate, washed with water,
and dried over MgSO4 to yield the title compound (230 mg, 91%
yield) as a yellow oil. 1H NMR (MeOD, δ): 1.41 (s, 9H, 3 × CH3),
2.49-2.75 (m, 4H, 2 × CH2), 2.98-3.20 (m, 2H, CH2), 3.60, 3.70
(d, 2H, J ) 13.8, CH2), 4.57-4.67 (m, 1H, CH), 5.19 (s, 2H, CH2),
6.99-7.09 (m, 2H, ArH), 7.16-7.50 (m, 10H, ArH), 8.19 (s, 1H,
ArH), 8.32 (s, 1H, CHO). 13C NMR (MeOD, δ): 28.9, 39.4, 55.3,
60.4, 63.6, 71.8, 72.3, 79.9, 113.3, 120.8, 123.6, 127.8, 128.0, 128.8,
129.1, 129.3, 129.6, 130.3, 137.2, 138.1, 140.3, 148.7, 158.4, 162.0.
LCMS (hydrophilic): Rf (min) ) 8.14, (ESI + ve) found 520 [M
+ H]+ calcd for C30H38N3O5.
tert-Butyl 4-(Benzyl(2-(4-(benzyloxy)-3-formamidophenyl)hy-
droxyethyl)amino)butylcarbamate (9b). The title compound was
prepared using the same method that was described for 9a. The
product was obtained as a yellow oil in 92% yield. 1H NMR
(MeOD, δ): 1.35-1.55 (m, 4H, 2 × CH2), 1.42 (s, 9H, 3 × CH3),
2.52-2.64 (m, 2H, CH2), 2.68-2.77 (m, 2H, CH2), 2.96 (t, 2H, J
) 6.6, CH2), 3.71, 3.78 (d, 2H, J ) 13.5, CH2), 4.64 (t, 1H, J )
7.1, CH), 5.20 (s, 2H, CH2), 6.99-7.08 (m, 2H, ArH), 7.19-7.51
(m, 10H, ArH), 8.18 (s, 1H, ArH), 8.33 (s, 1H, CHO). 13C NMR
(MeOD, δ): 24.6, 28.5, 28.9, 41.0, 54.9, 59.7, 62.6, 71.3, 71.6,
79.7, 113.2, 120.6, 123.5, 127.8, 128.5, 128.7, 129.1, 129.4, 129.6,
130.5, 136.7, 138.0, 138.7, 148.6, 158.3, 161.9. LCMS (hydro-
philic): Rf (min) ) 8.54, (ESI + ve) found 548 [M + H]+ calcd
for C32H42N3O5.
tert-Butyl 6-(Benzyl(2-(4-(benzyloxy)-3-formamidophenyl)hy-
droxyethyl)amino)hexylcarbamate (9c). The title compound was
prepared using the same method that was described for 9a. The
product was obtained as a yellow oil in 89% yield. 1H NMR
(MeOD, δ): 1.14-1.25 (m, 4H, 2 × CH2), 1.35-1.49 (m, 4H, 2 ×
CH2), 1.42 (s, 9H, 3 × CH3), 2.42-2.53 (m, 2H, CH2), 2.62-2.71
(m, 2H, CH2), 2.97 (t, 2H, J ) 6.9, CH2), 3.61, 3.68 (d, 2H, J )
13.5, CH2), 4.61 (t, 1H, J ) 6.6, CH), 5.12 (s, 2H, CH2), 6.98-7.07
(m, 2H, ArH), 7.17-7.50 (m, 10H, ArH), 8.21 (s, 1H, ArH), 8.32
(s, 1H, CHO). 13C NMR (MeOD, δ): 27.7, 27.9, 28.0, 29.0, 31.0,
41.3, 55.3, 60.2, 63.2, 71.8, 71.9, 79.8, 113.3, 120.8, 123.7, 128.1,
128.9, 129.2, 129.4, 129.7, 130.3, 137.2, 138.1, 140.3, 148.7, 158.4,
162.0. LCMS (hydrophilic): Rf (min) ) 9.23, (ESI + ve) found
576 [M + H]+ calcd for C34H46N3O5.
concentrated in vacuo, and freeze-dried to produce the title
compound as a yellow gum (236 mg, 100%). H NMR (MeOD,
1
δ): 3.28-3.40 (m, 2H, CH2), 3.47-3.75 (m, 4H, 2 × CH2), 4.56
(s, 2H, CH2), 4.83 (t, 1H, J ) 6.9, CH), 5.20 (s, 2H, CH2),
6.97-7.12 (m, 2H, ArH), 7.25-7.67 (m, 10H, ArH), 8.19 (s, 1H,
ArH), 8.35 (s, 1H, CHO). 13C NMR (MeOD, δ): 35.6, 52.2, 60.4,
60.5, 69.3, 71.7, 113.7, 120.2, 123.5, 128.2, 128.8, 129.2, 129.6,
130.6, 130.7, 131.4, 132.5, 134.2, 137.9, 149.5, 162.2. LCMS
(hydrophilic): Rf (min) ) 6.27, (ESI + ve) found 420 [M + H]+
calcd for C25H30N3O3.
N-(5-(2-((4-aminobutyl)(benzyl)amino)-1-hydroxyethyl)-2-(ben-
zyloxy)phenyl)formamide: TFA salt (10b). The title compound was
prepared using the same method that was described for 10a. The
product was obtained as a brown gum in quantitative yield. 1H
NMR (MeOD, δ): 1.72 (m, 2H, CH2), 1.95 (m, 2H, CH2), 3.00 (t,
2H, J ) 7.2, CH2), 3.18-3.30 (m, 4H, 2 × CH2), 4.42-4.58 (m,
2H, CH2), 5.21 (s, 2H, CH2), 7.00-7.14 (m, 2H, ArH), 7.20-7.65
(m, 10H, ArH), 8.22 (s, 1H, ArH), 8.35 (s, 1H, CHO). 13C NMR
(MeOD, δ): 21.8, 25.7, 40.0, 59.1, 59.5, 59.7, 68.8, 71.7, 113.7,
120.2, 123.4, 128.4, 128.8, 129.2, 129.6, 130.5, 130.7, 131.4, 132.5,
134.4, 137.4, 149.4, 162.2. LCMS (hydrophilic): Rf (min) ) 9.23,
(ESI + ve) found 448 [M + H]+ calcd for C27H34N3O3.
N-(5-(2-((6-aminohexyl)(benzyl)amino)-1-hydroxyethyl)-2-(ben-
zyloxy)phenyl)formamide: TFA salt (10c). The title compound was
prepared using the same method that was described for 10a. The
product was obtained as a yellow oil in quantitative yield. 1H NMR
(MeOD, δ): 1.43 (m, 4H, 2 × CH2), 1.69 (m, 2H, CH2), 1.87 (m,
2H, CH2), 2.93 (t, 2H, J ) 7.2, CH2), 3.18-3.27 (m, 4H, 2 ×
CH2), 4.42-4.60 (m, 2H, CH2), 5.24 (s, 2H, CH2), 7.01-7.64 (m,
12H, ArH), 8.22 (s, 1H, ArH), 8.35 (s, 1H, CHO). 13C NMR
(MeOD, δ): 24.3, 26.8, 27.0, 28.2, 40.4, 59.1, 59.5, 59.6, 68.8,
71.8, 113.7, 120.2, 123.5, 128.2, 128.7, 129.2, 129.6, 130.4, 130.8,
131.2, 132.4, 134.5, 137.4, 149.3, 162.3. LCMS (hydrophilic): Rf
(min) ) 9.23, (ESI + ve) found 476 [M + H]+ calcd for
C29H38N3O3.
N-(5-(2-(benzyl(2-(N6-adenosinyl)ethyl)amino)-1-hydroxyethyl)-
2-(benzyloxy)phenyl)formamide (11a). To a suspension of 6-chlo-
ropurine riboside (381 mg, 1.33 mmol) in t-BuOH (20 mL) was
added 10a (236 mg, 0.44 mmol), followed by DIPEA (694 µL,
3.98 mmol), and the reaction mixture was stirred at 80 °C for 24 h
under N2. The solvent was evaporated under reduced pressure and
the residue was purified by flash silica chromatography (eluent 10%
MeOH/CH2Cl2 + 1% NH3) to produce the title compound as a
yellow oil (134 mg, 45% yield). 1H NMR (MeOD, δ): 2.60-2.85
(m, 4H, 2 × CH2), 3.45-3.95 (m, 6H, 3 × CH2), 4.16-4.21 (m,
1H, CH), 4.31-4.36 (m, 1H, CH), 4.67 (t, 1H, J ) 6.0, CH),
4.73-4.79 (m, 1H, CH), 5.10 (d, 2H, J ) 1.8, CH), 5.96 (d, 1H,
J ) 6.3, CH), 6.60-7.45 (m, 12H, ArH), 8.10 (m, 1H, ArH),
8.19-8.26 (m, 2H, 2 × CH-purine), 8.29 (s, 1H, CHO). 13C NMR
(MeOD, δ): 39.7, 54.5, 60.7, 63.6, 63.8, 71.7, 72.6, 72.8, 75.5,
88.3, 91.4, 114.9, 120.7, 121.7, 123.8, 127.9, 128.0, 128.6, 128.8,
129.1, 129.6, 130.2, 137.1, 138.2, 140.1, 141.4, 148.8, 150.2, 153.3,
156.2, 162.0. LCMS (hydrophilic): Rf (min) ) 6.54, (ESI + ve)
found 670 [M + H]+ calcd for C35H40N7O7.
N-(5-(2-(benzyl(4-(N6-adenosinyl)butyl)amino)-1-hydroxyethyl)-
2-(benzyloxy)phenyl)formamide (11b). The title compound was
prepared using the same method that was described for 11a. After
purification by flash chromatography (eluent 5-10% MeOH/CH2Cl2
+ 1% NH3), the product was obtained as a yellow oil in 38% yield.
1H NMR (MeOD, δ): 1.50-1.63 (m, 4H, 2 × CH2), 2.48-2.65
(m, 4H, 2 × CH2), 3.40-3.55 (m, 2H, CH2), 3.60-3.94 (m, 4H, 2
× CH2), 4.15-4.19 (m, 1H, CH), 4.32 (dd, 1H, J ) 2.4, J ) 5.1,
CH), 4.59-4.65 (m, 1H, CH), 4.74 (t, 1H, J ) 5.4, CH), 5.15 (s,
2H, CH), 5.94 (d, 1H, J ) 6.6, CH), 6.96-7.49 (m, 12H, ArH),
8.18 (m, 2H, ArH and CH-purine), 8.22 (s, 1H, CH-purine), 8.31
(s, 1H, CHO). 13C NMR (MeOD, δ): 25.4, 28.1, 41.4, 55.0, 60.2,
63.3, 63.5, 71.8, 72.1, 72.7, 75.5, 88.2, 91.3, 113.2, 120.9, 121.4,
123.7, 127.8, 128.1, 128.8, 129.1, 129.2, 129.6, 130.3, 137.3, 138.1,
140.2, 141.4, 148.7, 150.3, 153.5, 156.2, 162.0. LCMS (hydro-
philic): Rf (min) ) 6.62, (ESI + ve) found 698 [M + H]+ calcd
for C37H44N7O7.
N-(5-(2-((2-aminoethyl)(benzyl)amino)-1-hydroxyethyl)-2-(ben-
zyloxy)phenyl)formamide: TFA salt (10a). A solution of trifluo-
roacetic acid/dichloromethane (1:1) (682 µL) was added dropwise
to a stirred suspension of 9a (230 mg, 0.44 mmol) in dichlo-
romethane (3 mL), and the mixture was stirred at room temperature
for 4 h under argon. After the evaporation of the solvent and excess
TFA, the residue was then redissolved in MeOH (40 mL),

ꢀ2-Adrenergic and Adenosine A1 Receptor Ligands
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19 6135
N-(5-(2-(benzyl(6-(N6-adenosinyl)hexyl)amino)-1-hydroxyethyl)-
2-(benzyloxy)phenyl)formamide (11c). To a suspension of 6-chlo-
ropurine riboside (135 mg, 0.47 mmol) in t-BuOH (25 mL) was
added 10c (213 mg, 0.36 mmol), followed by DIPEA (158 µL, 0.9
mmol), and the reaction mixture was stirred for 24 h at 80 °C for
24 h under N2. More 6-chloropurine riboside (135 mg, 0.47 mmol)
and DIPEA (158 µL, 0.9 mmol) were added, and the reaction was
stirred for an additional 24 h at 80 °C. The solvent was removed
under reduced pressure, and the residue was purified by flash silica
chromatography (eluent 5-10% MeOH/CH2Cl2 + 1% NH3) to
produce the title compound as a yellow oil (100 mg, 38% yield).
1H NMR (MeOD, δ): 1.05-1.63 (m, 8H, 4 × CH2), 2.33-2.48
(m, 2H, CH2), 2.55-2.70 (m, 2H, CH2), 3.40-3.62 (m, 4H, 2 ×
CH2), 3.80 (dd, 2H, J ) 2.7, J ) 12.6, CH2), 4.15-4.19 (m, 1H,
CH), 4.33 (dd, 1H, J ) 2.7, J ) 5.0, CH), 4.56-4.63 (m, 1H,
CH), 4.75 (t, 1H, J ) 5.4, CH), 5.09 (s, 2H, CH), 5.94 (d, 1H, J )
6.6, CH), 6.93-7.42 (m, 12H, ArH), 8.14-8.21 (m, 2H, 2 ×
CH-purine), 8.22 (d, 1H, J ) 1.5, ArH), 8.32 (s, 1H, CHO). 13C
NMR (MeOD, δ): 27.8, 27.9, 28.0, 30.4, 41.6, 55.2, 60.3, 63.1,
63.5, 71.8, 72.2, 72.7, 75.5, 88.2, 91.4, 113.1, 120.8, 121.4, 123.7,
127.9, 128.0, 128.8, 129.1, 129.2, 129.5, 130.3, 137.4, 138.0, 140.4,
141.3, 148.7, 150.2, 153.5, 156.2, 162.0. LCMS (hydrophilic): Rf
(min) ) 6.82, (ESI + ve) found 726 [M + H]+ calcd for
C39H48N7O7.
CH2), 3.62-3.72 (m, 2H, CH2), 3.81 (dd, 2H, J ) 2.4, J ) 12.6,
CH2), 4.17 (dd, 1H, J ) 2.4, J ) 4.8, CH), 4.32 (dd, 1H, J ) 2.4,
J ) 5.1, CH), 4.74 (dd, 1H, J ) 5.3, J ) 6.6, CH), 5.95 (d, 1H, J
) 6.3, CH), 6.87 (d, 1H, J ) 8.1, ArH), 7.04 (dd, 1H, J ) 2.1, J
) 8.4, ArH), 8.11 (d, 1H, J ) 1.8, ArH), 8.22 (s, 1H, CH-purine),
8.26 (s, 1H, CH-purine), 8.31 (s, 1H, CHO). 13C NMR (MeOD,
δ): 25.0, 27.9, 40.9, 49.6, 55.6, 63.6, 70.6, 72.8, 75.5, 88.3, 91.3,
116.1, 120.4, 121.5, 123.7, 127.0, 133.6, 141.6, 148.5, 149.9, 153.6,
156.5, 162.1. LCMS (hydrophilic, formate method ACN/water (10
mM NH4-formate)): Rf (min) ) 3.33, (ESI + ve) found 518 [M
+ H]+ calcd for C23H32N7O7. HR-ESMS (m/z): [M + H]+ calcd
for C23H32N7O7, 518.2358; found, 518.2370. 12d: 1H NMR (MeOD,
δ): 1.74-1.86 (m, 4H, 2 × CH2), 3.05-3.15 (m, 4H, 2 × CH2),
3.62-3.73 (m, 2H, CH2), 3.82 (dd, 2H, J ) 2.7, J ) 12.5, CH2),
4.17 (dd, 1H, J ) 2.4, J ) 5.1, CH), 4.32 (dd, 1H, J ) 2.4, J )
5.0, CH), 4.71-4.79 (m, 2H, 2 × CH), 5.95 (d, 1H, J ) 6.6, CH),
6.62 (dd, 1H, J ) 2.1, J ) 8.1, ArH), 6.70 (d, 1H, J ) 8.1, ArH),
6.78 (d, 1H, J ) 1.8, ArH), 8.23 (s, 1H, CH-purine), 8.27 (s, 1H,
CH-purine). 13C NMR (MeOD, δ): 24.4, 27.8, 40.7, 49.7, 55.3,
63.6, 70.3, 72.7, 75.6, 88.3, 91.3, 114.7, 115.5, 117.5, 121.5, 133.6,
136.8, 141.7, 146.7, 149.5, 153.6, 156.5. LCMS (hydrophilic): Rf
(min) ) 1.88, (ESI + ve) found 490 [M + H]+ calcd for
C22H32N7O6. HR-ESMS (m/z): [M + H]+ calcd for C22H32N7O6,
490.2409; found, 490.2422.
N-(2-hydroxy-5-(1-hydroxy-2-(6-(N6-adenosinyl)hexylamino)-
ethyl)phenyl)formamide (12e). Compound 11c (100 mg, 0.14
mmol) was dissolved in methanol, 10% Pd/C (123 mg) was added,
and the reaction was shaken for 2 days under an atmosphere of
hydrogen at 50 psi. The solution was filtered through a Millipore
0.45 µm filter, and the solvent was removed under reduced pressure.
The oily residue was purified by preparatory HPLC on a C18 column
with a gradient of 1-18% ACN/H2O over 50 min to produce the
desired compound 12e with traces of ammonium acetate. The
compound was subjected to a prepacked C18 column to remove
the ammonium acetate, and the title compound was produced as a
light-yellow solid (3.5 mg) in 5% yield. 1H NMR (MeOD, δ):
1.42-1.56 (m, 4H, 2 × CH2), 1.66-1.80 (m, 4H, 2 × CH2), 3.04
(t, 2H, J ) 8.0, CH2), 3.07-3.17 (m, 2H, CH2), 3.61 (m, 2H, CH2),
3.81 (dd, 2H, J ) 2.1, J ) 12.6, CH2), 4.18 (m, 1H, CH), 4.32
(dd, 1H, J ) 2.4, J ) 5.0, CH), 4.74 (t, 1H, J ) 5.7, CH), 5.95 (d,
1H, J ) 6.3, CH), 6.88 (d, 1H, J ) 8.1, CH), 7.05 (dd, 1H, J )
1.8, J ) 8.3, ArH), 8.13 (br s, 1H, ArH), 8.21 (s, 1H, CH-purine),
8.25 (s, 1H, CH-purine), 8.31 (s, 1H, CHO). 13C NMR (MeOD,
δ): 26.9, 27.3, 27.4, 30.3, 41.4, 48.9, 55.1, 63.6, 70.0, 72.7, 75.5,
88.3, 91.3, 116.1, 120.3, 121.5, 123.7, 127.0, 133.2, 141.5, 148.6,
149.1, 153.5, 156.4, 162.1. LCMS (hydrophilic): Rf (min) ) 4.53,
(ESI + ve) found 546 [M + H]+ calcd for C25H36N7O7. HR-ESMS
(m/z): [M + H]+ calcd for C25H36N7O7, 546.2671; found, 546.2673.
tert-Butyl 2-(2-(2-(Benzyl(2-(4-(benzyloxy)-3-nitrophenyl)hy-
droxyethyl)amino)ethoxy)ethoxy)ethylcarbamate (13). The title
compound was prepared using the same method that was described
for 3a. After purification by flash chromatography (eluent 5%
MeOH/CH2Cl2 + 1% NH3), the product was obtained as a yellow
N-(2-hydroxy-2-(2-(N6-adenosinyl) ethylamino)ethyl)phenyl)for-
mamide (12a) and (12b). Compound 11a (130 mg, 1.35 mmol) was
dissolved in ethanol, 10% Pd/C (173 mg) was added, and the
reaction was shaken for 3 days under an atmosphere of hydrogen
at 50 psi. The catalyst was removed by filtration, and the reaction
vessel was recharged with 10% Pd/C (173 mg) and was left to
shake for 3 days under the same catalytic hydrogenation conditions.
The solution was filtered through a Millipore 0.45 µm filter, and
the solvent was removed under reduced pressure. The oily residue
was purified by preparatory HPLC on a C18 column with a gradient
of 1-18% ACN/H2O over 45 min to produce the desired compound
12a as a pale-yellow solid (13.5 mg) in 14% yield. The deformy-
lated byproduct 12b was also isolated as a yellow-brown oil (8
mg) in 9% yield. 12a: 1H NMR (MeOD, δ): 3.20-3.49 (m, 4H, 2
× CH2), 3.72-3.95 (m, 4H, 2 × CH2), 4.18 (dd, 1H, J ) 2.4, J )
5.1, CH), 4.33 (dd, 1H, J ) 2.7, J ) 5.1, CH), 4.74 (t, 1H, J )
5.7, CH), 5.99 (d, 1H, J ) 6.3, CH), 6.87 (d, 1H, J ) 8.1, ArH),
7.06 (dd, 1H, J ) 2.1, J ) 8.4, ArH), 8.13 (d, 1H, J ) 1.8, ArH),
8.27 (s, 1H, CH-purine), 8.30 (s, 1H, CH-purine), 8.33 (s, 1H,
CHO). 13C NMR (MeOD, δ): 39.2, 49.9, 55.2, 63.5, 70.2, 72.7,
75.7, 88.2, 91.2, 116.2, 120.3, 121.8, 123.7, 127.0, 133.1, 142.3,
148.6, 149.8, 153.4, 156.6, 162.1. LCMS (hydrophilic): Rf (min)
) 3.17, (ESI + ve) found 490 [M + H]+ calcd for C21H28N7O7.
HR-ESMS (m/z): [M + H]+ calcd for C21H28N7O7, 490.2045;
found, 490.2042. 12b: 1H NMR (MeOD, δ): 3.20-3.49 (m, 4H, 2
× CH2), 3.72-3.95 (m, 4H, 2 × CH2), 4.18 (dd, 1H, J ) 2.4, J )
5.1, CH), 4.33 (dd, 1H, J ) 2.7, J ) 5.1, CH), 4.73 (t, 1H, J )
5.4, CH), 5.99 (d, 1H, J ) 6.3, CH), 6.77 (d, 1H, J ) 8.1, ArH),
6.83 (dd, 1H, J ) 2.1, J ) 8.3, ArH), 6.97 (d, 1H, J ) 1.8, ArH),
8.27 (s, 1H, CH-purine), 8.33 (s, 1H, CH-purine). 13C NMR
(MeOD, δ): 39.1, 49.9, 55.2, 63.5, 70.0, 72.7, 75.7, 88.3, 91.2,
116.0, 116.8 120.6, 121.8, 132.3, 133.6, 142.3, 148.2, 149.8, 153.4,
156.6. LCMS (hydrophilic): Rf (min) ) 1.74, (ESI + ve) found
462 [M + H]+ calcd for C20H28N7O6. HR-ESMS (m/z): [M + H]+
calcd for C20H28N7O6, 462.2101; found, 462.2101.
1
oil in 58% yield. H NMR (MeOD, δ): 1.37 (s, 9H, 3 × CH3),
2.64-2.88 (m, 4H, 2 × CH2), 3.20 (t, 2H, J ) 5.4, CH2), 3.44-3.59
(m, 8H, 4 × CH2), 3.70 (s, 2H, CH2), 4.66 (t, 1H, J ) 6.6, CH),
5.22 (s, 2H, CH2), 7.16-7.49 (m, 12H, ArH), 7.77 (d, 1H, J )
2.1, ArH). 13C NMR (MeOD, δ): 28.8, 41.4, 55.1, 61.1, 63.6, 70.6,
71.1, 71.2, 71.3, 71.4, 72.2, 80.1, 116.2, 124.1, 128.1, 128.3, 129.1,
129.3, 129.6, 130.2, 132.9, 137.5, 137.7, 140.5, 141.3, 151.9, 158.4.
LCMS (hydrophilic): Rf (min) ) 9.46, (ESI + ve) found 610 [M
+ H]+ calcd for C33H44N3O6.
N-(2-hydroxy-5-(1-hydroxy-2-(4-(N6-adenosinyl) butylamino)-
ethyl)phenyl)formamide (12c) and (12d). Compound 11b (110 mg,
0.16 mmol) was dissolved in ethanol (10 mL), 10% Pd/C (187 mg)
was added, and the reaction was shaken for 2 days under an
atmosphere of hydrogen at 50 psi. The solution was filtered through
a Millipore 0.45 µm filter, and the solvent was removed under
reduced pressure. The oily residue was purified by preparatory
HPLC on a C18 column with a gradient of 1-18% ACN/H2O over
45 min to produce the desired compound 12c as an off-white solid
(12 mg) in 15% yield. The deformylated byproduct 12d was also
tert-Butyl 2-(2-(2-(Benzyl(2-(4-(benzyloxy)-3-aminophenyl)hy-
droxyethyl)amino)ethoxy)ethoxy)ethylcarbamate (14). The title
compound was prepared using the same method that was described
for 8a. After purification by flash chromatography (eluent 5-10%
MeOH/CH2Cl2 + 1% NH3), the product was obtained as a yellow
1
oil in 45% yield. H NMR (MeOD, δ): 1.39 (s, 9H, 3 × CH3),
2.60-2.88 (m, 4H, 2 × CH2), 3.21 (t, 2H, J ) 5.4, CH2), 3.43-3.59
(m, 8H, 4 × CH2), 3.64, 3.78 (d, 2H, J ) 13.5, CH2), 4.54 (t, 1H,
J ) 6.6, CH), 5.03 (s, 2H, CH2), 6.61 (dd, 1H, J ) 2.1, J ) 8.4,
1
isolated as a pale-yellow solid (3 mg) in 4% yield. 12c: H NMR
(MeOD, δ): 1.74-1.84 (m, 4H, 2 × CH2), 3.02-3.14 (m, 4H, 2 ×

6136 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19
Karellas et al.
ArH), 6.76 (d, 1H, J ) 1.8, ArH), 6.80 (d, 1H, J ) 8.4, ArH),
7.18-7.46 (m, 10H, ArH). 13C NMR (MeOD, δ): 28.8, 41.3, 54.6,
60.8, 64.0, 70.3, 71.1, 71.2, 71.4, 72.3, 80.0, 113.1, 114.6, 117.4,
128.1, 128.6, 128.9, 129.3, 129.5, 130.2, 137.0, 137.8, 138.8, 140.5,
147.3, 158.3. LCMS (hydrophilic): Rf (min) ) 7.74, (ESI + ve)
found 580 [M + H]+ calcd for C33H46N3O6.
tert-Butyl 2-(2-(2-(Benzyl(2-(4-(benzyloxy)-3-formamidophenyl)-
hydroxyethyl)amino)ethoxy)ethoxy)ethylcarbamate (15). The title
compound was prepared using the same method that was described
for 9a. The product was obtained as a light-yellow oil in 99% yield.
1H NMR (MeOD, δ): 1.39 (s, 9H, 3 × CH3), 2.65-2.90 (m, 4H,
2 × CH2), 3.20 (t, 2H, J ) 5.4, CH2), 3.44-3.63 (m, 8H, 4 ×
CH2), 3.72, 3.81 (d, 2H, J ) 13.5, CH2), 4.61 (t, 1H, J ) 6.6, CH),
5.17 (s, 2H, CH2), 6.95-7.13 (m, 2H, ArH), 7.15-7.51 (m, 10H,
ArH), 8.19 (s, 1H, ArH), 8.32 (s, 1H, CHO). 13C NMR (MeOD,
δ): 28.8, 41.3, 54.7, 60.7, 63.6, 70.0, 71.1, 71.2, 71.3 or (2C at
71.2), 71.7, 71.9, 80.0 113.2, 120.6, 123.5, 127.9, 128.3, 128.8,
129.1, 129.4, 129.6, 130.4, 136.6, 138.1, 139.7, 148.6, 158.3, 162.0.
LCMS (hydrophilic): Rf (min) ) 8.71, (ESI + ve) found 608 [M
+ H]+ calcd for C34H46N3O7.
tericin B (2.5 µg/mL), penicillin G (100 U/mL), and 5% fetal bovine
serum at 37 °C. Cells were used in experiments at one day
preconfluence. The culture medium was aspirated, and the cells
were rinsed once with warm Hank’s balanced salt solution (HBSS).
Experiments were started by the addition of 1 mL of HBSS that
contained adenosine deaminase (0.5 U/mL) and 20 µM rolipram
and that was either with or without varying concentrations of test
compounds. After a 6 min incubation at 37 °C, the medium was
aspirated, and 0.5 mL of 50 mM HCl was added. The cAMP content
in each well was determined by radioimmunoassay. Briefly, 5 µL
from each plate was diluted with 100 µL of 50 mM HCl, and the
cAMP was acetylated by the addition of 4.5 µL of a 3.5:1 mixture
of triethylamine and acetic anhydride with vortex mixing. A 10
µL aliquot of [125I]-ScAMP-TME that contained 20 000 cpm was
added, followed by 100 µL of cAMP antibody dissolved in 50 mM
Na-acetate buffer at pH 4.75 that contained 0.125% BSA. The
samples were incubated at room temperature for 60 min, after which
50 µL of hydroxyapatite in a 1:1 suspension with water was added
for an additional 10 min. The samples were diluted with 3 mL of
ice-cold 10 mM Tris buffer at pH 7.0 and were then aspirated
through Whatman GF/B glass fiber filters by the use of a Brandell
cell harvester. The filters were washed with an additional 6 mL of
ice-cold buffer, and we determined the retained radioactivity by
using a Beckman gamma counter. The cAMP that was accumulated
was determined from a standard curve. All assays were performed
in quadruplicate.
Receptor Assays. DDT cell membranes and the displacement of
specific [3H]-8-cyclopentyl-1,3-dipropylxanthine (2.5 nM) binding
from the A1AR was determined, as described previously.29 The
interaction of test compounds with the ꢀ2AR was determined by
the displacement of specific [125I]-(-)iodopindolol binding. Briefly,
cell membranes were incubated for 60 min at room temperature in
a total volume of 0.25 mL that contained 50 mM Tris-HCl buffer
at pH 7.4, 5 mM MgCl2, 100 pM [125I]-(-)iodopindolol, and 10
µM 5′-guanylyl-imidodiphosphate and that was either with or
without varying concentrations of test compounds. Nonspecific
binding was determined in parallel assays that contained 1 µM
alprenolol. At the end of the incubation, each suspension was diluted
with 3 mL of ice-cold incubation buffer, and the membranes were
collected by retention on a Whatman GF/B glass fiber filter under
reduced pressure. The filters were washed with an additional 6 mL
of ice-cold buffer, and the radioactivity was determined in a gamma
counter. All assays were performed in triplicate.
N-(5-(2-((2-(2-(2-aminoethoxy)ethoxy)ethyl)benzyl)amino)-1-hy-
droxyethyl)-2-(benzyloxy)phenyl)formamide: TFA salt (16). The
title compound was prepared using the same method that was
described for 10a. The product was obtained as a yellow gum in
quantitative yield. 1H NMR (MeOD, δ): 2.99-3.14 (m, 2H, CH2),
3.32-3.41 (m, 2H, CH2), 3.59 (m, 2H, CH2), 3.60-3.75 (m, 6H,
3 × CH2), 3.92 (m, 2H, CH2), 4.52, 4.62 (d, 2H, J ) 13.2, CH2),
5.22 (s, 2H, CH2), 7.12-7.64 (m, 12H, ArH), 8.24 (s, 1H, ArH),
8.35 (s, 1H, CHO). 13C NMR (MeOD, δ): 40.5, 54.4, 59.7, 60.2,
66.0, 67.9, 68.6, 71.3, 71.4, 71.7, 113.7, 120.2, 123.5, 128.3, 128.6,
129.2, 129.6, 130.4, 130.9, 131.2, 132.5, 134.4, 137.9, 149.4, 162.3.
LCMS (hydrophilic): Rf (min) ) 8.10, (ESI + ve) found 508 [M
+ H]+ calcd for C29H38N3O5.
N(5-(2-((2-(2-(N6-adenosinylethoxy)ethoxy)ethyl)(benzyl)amino)-
1-hydroxyethyl)-2-(benzyloxy)phenyl)formamide (17). The title
compound was prepared using the same method that was described
for 11a (yield 50%, yellow oil, eluent 5-10% MeOH/CH2Cl2 +
1
1% NH3). H NMR (MeOD, δ): 2.55-2.85 (m, 4H, 2 × CH2),
3.40-3.92 (m, 14H, 7 × CH2), 4.17 (dd, 1H, J ) 2.4, J ) 4.7,
CH), 4.32 (dd, 1H, J ) 2.4, J ) 5.0, CH), 4.55-4.63 (m, 1H,
CH), 4.70-4.76 (m, 1H, CH), 5.09 (s, 2H, CH), 5.93 (d, 1H, J )
6.6, CH), 6.92-7.02 (m, 2H, ArH), 7.01-7.45 (m, 10H, ArH),
8.14-8.21 (m, 3H, ArH and 2 × CH-purine), 8.30 (s, 1H, CHO).
13C NMR (MeOD, δ): 41.5, 54.7, 60.7, 63.5, 63.7, 70.4, 70.7, 71.3,
71.4, 71.7, 72.1, 72.7, 75.5, 88.2, 91.3, 113.1, 120.7, 121.5, 123.7,
127.8, 128.1, 128.8, 129.1, 129.2, 129.6, 130.2, 136.8, 138.1, 140.3,
141.5, 148.7, 150.2, 153.4, 156.2, 162.0. LCMS (hydrophilic): Rf
(min) ) 6.50, (ESI + ve) found 758 [M + H]+ calcd for
C39H48N7O9.
Data Analysis. The concentration of test compounds that
stimulated cAMP accumulation by 50% the maximal response
(EC50) were determined by nonlinear regression analysis of the
concentration response using the GraphPad Prism 3.0 program
(GraphPad Software, San Diego, CA). We determined the concen-
tration of compounds that inhibited specific radioligand binding
by 50% (IC50) by nonlinear regression analysis using GraphPad
Prism. We calculated the dissociation constants (Ki) for the
compounds from the IC50 values by using the conversion described
by Cheng and Prusoff.30
N-(5-(2-((2-(2-(N6-adenosinylethoxy)ethoxy)ethyl)amino)-1-hy-
droxyethyl)-2-hydroxyphenyl)formamide (18). We prepared the title
compound by utilizing the same method that was described for 12e
(yield 16%, preparatory HPLC, C18 column with a gradient of
1-18% ACN/H2O over 45 min. 1H NMR (MeOD, δ): 2.72-2.90
(m, 4H, 2 × CH2), 3.53-3.68 (m, 6H, 3 × CH2), 3.69-3.93 (m,
6H, 3 × CH2), 4.17 (dd, 1H, J ) 2.4, J ) 5.1, CH), 4.32 (dd, 1H,
J ) 2.4, J ) 5.1, CH), 4.69 (dd, 1H, J ) 4.8, J ) 8.6, CH), 4.74
(dd, 1H, J ) 5.4, J ) 6.3, CH), 5.95 (d, 1H, J ) 6.6, CH), 6.82 (d,
1H, J ) 8.1, ArH), 6.99 (dd, 1H, J ) 2.1, J ) 8.4, ArH), 8.04 (d,
1H, J ) 2.1, ArH), 8.20-8.25 (m, 2H, 2 × CH-purine), 8.29 (s,
1H, CHO). 13C NMR (MeOD, δ): 41.6, 49.5, 57.5, 63.6, 70.5, 70.7,
71.4, 71.5, 72.8, 73.0, 75.5, 88.3, 91.3, 116.2, 120.6, 121.5, 124.0,
126.8, 135.3, 141.6, 148.3, 149.3, 153.5, 156.4, 162.0. LCMS
(hydrophilic, formate method ACN/water (10 mM NH4-formate)):
Rf (min) ) 3.75, (ESI + ve) found 578 [M + H]+ calcd for
C25H36N7O9. HR-ESMS (m/z): [M + H]+ calcd for C25H36N7O9,
578.2569; found, 578.2566.
1
Supporting Information Available: H NMR, 13C NMR, and
LCMS analyses for all target compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Portoghese, P. S.; Larson, D. L.; Sayre, L. M.; Yim, C. B.; Ronsisvalle,
G.; Tam, S. W.; Takemori, A. E. Opioid agonist and antagonist bivalent
ligands. The relationship between spacer length and selectivity at
multiple opioid receptors. J. Med. Chem. 1986, 29, 1855–1861.
(2) Perez, M.; Jorand-Lebrun, C.; Pauwels, P. J.; Pallard, I.; Halazy, S.
Dimers of 5HT1 ligands preferentially bind to 5HT1B/1D receptor
subtypes. Bioorg. Med. Chem. Lett. 1998, 8, 1407–1412.
(3) Bhushan, R. G.; Sharma, S. K.; Xie, Z.; Daniels, D. J.; Portoghese,
P. S. A bivalent ligand (KDN-21) reveals spinal δ and ꢀ opioid
receptors are organized as heterodimers that give rise to δ1 and ꢀ2
phenotypes. Selective targeting of δ-ꢀ heterodimers. J. Med. Chem.
2004, 47, 2969–2972.
Cell Culture and cAMP Determination. DDT1 MF-2 cells,
derived from leiomyosarcoma of Syrian hamster vas deferens, were
grown in 48-well plates by using Delbecco’s modified Eagle’s
medium that contained streptomycin sulfate (0.1 mg/mL), ampho-
(4) Jacobson, K. A.; Xie, R.; Young, L.; Chang, L.; Liang, B. T. A novel
pharmacological approach to treating cardiac ischemia. Binary con-

ꢀ2-Adrenergic and Adenosine A1 Receptor Ligands
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 19 6137
jugates of A1 and A3 adenosine receptor agonists. J. Biol. Chem. 2000,
275, 30272–30279.
(5) Halazy, S.; Perez, M.; Fourrier, C.; Pallard, I.; Pauwels, P. J.; Palmer,
C.; John, G. W.; Valentin, J.-P.; Bonnafous, R.; Martinez, J. Serotonin
dimers: application of the bivalent ligand approach to the design of
new potent and selective 5-HT1B/1D agonists. J. Med. Chem. 1996,
39, 4920–4927.
(6) Portoghese, P. S.; Larson, D. L.; Yim, C. B.; Sayre, L. M.; Ronsisvalle,
G.; Lipkowski, A. W.; Takemori, A. E.; Rice, K. C.; Tam, S. W.
Stereostructure-activity relationship of opioid agonist and antagonist
bivalent ligands. Evidence for bridging between vicinal opioid
receptors. J. Med. Chem. 1985, 28, 1140–1141.
(7) Krammer, R. H.; Karpen, J. W. Spanning binding sites on allosteric
proteins with polymer-linked ligand dimers. Nature 1998, 395, 710–
713.
(19) Hett, R.; Fang, Q. K.; Gao, Y.; Hong, Y.; Butler, H. T.; Nie, X.; Wald,
S. A. Enantio- and diastereoselective synthesis of all four stereoisomers
of formoterol. Tetrahedron Lett. 1997, 38, 1125–1128.
(20) Campos, F. M.; Bosch, P.; Guerrero, A. An efficient enantioselective
synthesis of (R,R)-formoterol, a potent bronchodilator, using lipases.
Tetrahedron: Asymmetry 2000, 11, 2705–2717.
(21) Muller, D.; Zeltser, I.; Bitan, G.; Gilon, C. Building units for
N-backbone cyclic peptides. 3. Synthesis of protected NR-(ω-
aminoalkyl) amino acids and NR-(ω-carboxyalkyl) amino acids. J.
Org. Chem. 1997, 62, 411–416.
(22) Alikhani, V.; Beer, D.; Bentley, D.; Bruce, I.; Cuenoud, B. M.;
Fairhurst, R. A.; Gedeck, P.; Haberthuer, S.; Hayden, C.; Janus, D.;
Jordan, L.; Lewis, C.; Smithies, K.; Wissler, E. J. Long-chain
formoterol analogues: an investigation into the effect of increasing
amino-substituent chain length on the ꢀ2-adrenoceptor activity. Bioorg.
Med. Chem. Lett. 2004, 14, 4705–4710.
(8) Gilman, A. G. G proteins: transducers of receptor-generated signals.
Annu. ReV. Biochem. 1987, 56, 615–649.
(23) Kumar, N.; Kiuchi, M.; Tallarico, J. A.; Schreiber, S. L. Small-
molecule diversity using a skeletal transformation strategy. Org. Lett.
2005, 7, 2535–2538.
(9) Elks, M. L.; Jackson, M.; Manganiello, V. C.; Vaughan, M. Effect of
N6-(L-2-phenylisoproyl)adenosine and insulin on cAMP metabolism
in 3T3-L1 adipocytes. Am. J. Physiol. 1987, 252, C342–C348.
(10) Romano, F. D.; MacDonald, S. G.; Dobson, J. G. Adenosine receptor
coupling to adenylate cyclase of rat ventricular myocyte membranes.
Am. J. Physiol. 1989, 257, H1088–H1095.
(24) Yoshida, Y.; Barrett, D.; Azami, H.; Morinaga, C.; Matsumoto, S.;
Matsumoto, Y.; Takasugi, H. Studies on anti-Helicobacter pylori
agents. Part 1: benzyloxyisoquinoline derivatives. Bioorg. Med. Chem.
1999, 7, 2647–2666.
(11) Gerwins, P.; Nordstedt, C.; Fredholm, B. B. Characterization of
adenosine A1 receptors in intact DDT1 MF-2 smooth muscle cells.
Mol. Pharmacol. 1990, 38, 660–666.
(12) Dobson, J. G.; Fenton, R. A.; Romano, F. D. Increased myocardial
adenosine production and reduction of ꢀ-adrenergic contractile
response in aged hearts. Circ. Res. 1990, 66, 1381–1390.
(13) Suteparuk, S.; Nies, A. S.; Andros, E.; Gerber, J. G. The role of
adenosine in promoting cardiac ꢀ-adrenergic subsenitivity in aging
humans. J. Gerontol. 1995, 50A, B128-B134.
(14) Song, Y.; Shryock, J. C.; Knot, H. J.; Belardinelli, L. Selective
attenuation by adenosine of arrhythmogenic action of isoproterenol
on ventricular myocytes. Am. J. Physiol.: Heart Circ. Physiol. 2001,
281, H2789–2795.
(15) Jacobson, K. A.; van Galen, P. J. M.; Williams, M. Adenosine
receptors: pharmacology, structure-activity relationships, and thera-
peutic potential. J. Med. Chem. 1992, 35, 407–422.
(16) Ruffolo, R. R.; Bondinell, W.; Hieble, J. P. R- and ꢀ-adrenoceptors:
from the gene to the clinic. 2. Structure-activity relationships and
therapeutic applications. J. Med. Chem. 1995, 38, 3681–3716.
(17) Scarpace, P. J.; Baresi, L. A.; Sanford, D. A.; Abrass, I. B.
Desensitization and resensitization of ꢀ-adrenergic receptors in a
smooth muscle cell line. Mol. Pharmacol. 1985, 28, 495–501.
(18) Kaiser, C.; Colella, D. F.; Schwartz, M. S.; Garvey, E.; Wardel, J. R.,
Jr. Adrenergic agents. 1. Synthesis and potential ꢀ-adrenergic agonist
activity of some catecholamine analogs bearing a substituted amino
functionality in the meta position. J. Med. Chem. 1974, 17, 49–57.
(25) Hutchinson, S. A.; Baker, S. P.; Scammells, P. J. New 2, N6-
disubstituted adenosines: potent and selective A1 adenosine receptor
agonists. Bioorg. Med. Chem. 2002, 10, 1115–1122.
(26) Senanayake, C. H. Desformoterol and Process for its Preparation. U.S.
Patent 5,965,622, 1999.
(27) Greenwald, R. B., Zhao, H. Poly(ethylene glycol) Prodrugs: Altered
Pharmacokinetics and Pharmacodynamics. In Prodrugs: Challenges
and Rewards; Stella, V. J., Tilley, J. W., Eds.; Springer: Boston, 2007;
pp 283-338.
(28) Portoghese, P. S.; Ronsisvalle, G.; Larson, D. L.; Takemori, A. E.
Synthesis and opioid antagonist potencies of nalterexamine bivalent
ligands with conformationally restricted spacers. J. Med. Chem. 1986,
1650–1653.
(29) Zhang, J.; Belardinelli, L.; Jacobson, K. A.; Otero, D. H.; Baker, S. P.
Persistent activation by and receptor reserve for an irreversible A1-
adenosine receptor agonist in DDT1 MF-2 cells and in guinea pig
heart. Mol. Pharmacol. 1997, 52, 491–498.
(30) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition constant
(Ki) and the concentration of inhibitor which causes 50% inhibition
(I50) of an enzymatic reaction. Biochem. Pharmacol. 1973, 22,
3099–3108.
JM800613S