XH-14 analogues as adenosine antagonists

Article abstract of DOI:10.1016/S0968-0896(98)00093-5

Analogues of the potent adenosine antagonist 5-(3'-hydroxypropyl)-7-methoxy-2-(3'-methoxy-4'-hydroxyphenyl)benzo[b]furan-3-carbaldehyde (XH-14, 1) with alternate substituents in the 2-, 5- and 7-positions have been synthesised. The affinity of these compo

Full text of DOI:10.1016/S0968-0896(98)00093-5

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 1517±1524
XH-14 Analogues as Adenosine Antagonists
Peter J. Scammells,a,* Stephen P. Bakerb and Anthony R. Beaugleholea
aSchool of Biological and Chemical Sciences, Deakin University, Geelong, VIC 3217, Australia
bDepartment of Pharmacology, University of Florida, Gainesville, FL 32610, USA
Received 9 December 1997; accepted 5 March 1998
AbstractÐAnalogues of the potent adenosine antagonist 5-(30-hydroxypropyl)-7-methoxy-2-(30-methoxy-40-hydroxy-
phenyl)benzo[b]furan-3-carbaldehyde (XH-14, 1) with alternate substituents in the 2-, 5- and 7-positions have been
synthesised. The anity of these compounds for the A1 adenosine receptor has been evaluated using a [3H]CPX com-
petitive binding assay. This structure-activity study highlighted the importance of the 3-formyl and 5-(3-hydroxy-
propyl) moieties for high receptor anity. # 1998 Elsevier Science Ltd. All rights reserved.
Introduction
The relatively potent adenosine antagonist XH-14 (1)
was isolated from the plant, Salvia miltiorrhiza, com-
monly known as Danshen.5,6 Aqueous extracts of this
plant have long been used as a traditional Chinese
medicine for angina pectoris and acute myocardial
infarction. This compound is the ®rst known nitrogen
free adenosine receptor ligand and has the advantage of
relatively high water solubility. Since the initial reports
which focused on the isolation, total synthesis and A1
adenosine receptor anity, benzofuran adenosine
antagonists of this type have remained largely unexplored.
Therefore, the aim of our research was to investigate the
structure±activity relationships of this interesting class
of compounds with respect to their interaction with the
A1 adenosine receptor. The approach chosen towards
this end involved the modi®cation of XH-14 based on
known requirements for receptor blockade, developed
largely from the structure±activity relationships of xan-
thines. In the case of xanthines, 1,3-dipropyl and 8-
cycloalkyl/aryl substitution produced dramatic increases
in receptor anity. The e€ect of these 1,3 and 8-sub-
stituents on A1 receptor anity is exempli®ed by a
comparison of 1,3-dipropyl-8-phenylxanthine (2) and
theophylline (1,3-dimethylxanthine). 1,3-Dipropyl-8-phenyl-
xanthine exhibited over 800 times greater anity than
theophylline in a [3H]-N6-(phenylisopropyl)adenosine
competitive binding assay (Ki values of 10 and 8470 nM,
respectively).7 Since XH-14 acts at the same receptor
site as these xanthines it was thought that incorporation
of propyl and phenyl substituents in the appropriate
positions may produce benzofuran antagonists with
improved binding anity (Fig. 1).
Adenosine is an important regulatory compound which
mediates a wide range of physiological e€ects in the
cardiac, nervous and immune systems.1,2 Adenosine
interacts with extracellular receptors which are coupled
to a variety of secondary messenger systems, including
enzymes and ion channels. Four adenosine receptor
subtypes (termed A1, A2a, A2b, and A3 AR) have been
de®ned based on their pharmacological properties and
molecular cloning. Although an enormous e€ort has
been directed toward the development of therapeutic
agents based on adenosine, relatively few compounds
have entered trials or been approved for clinical use.2
Adenosine agonists typically produce side e€ects result-
ing from the widespread distribution of adenosine
receptors in the body and are also known to cause
tachyphylaxis (receptor desensitisation).3 Adenosine
antagonists may have improved therapeutic potential
since their e€ects are dependant on the level of endo-
genous purinergic tone, rather than the absolute dis-
tribution of receptors.4 It has been proposed that A1
selective antagonists with good bioavailability could be
useful agents for the treatment of ischaemic brady-
arrhythmias, cardiac arrest and renal disease.4
Key words: Adenosine antagonist; benzofuran; A1 adenosine
receptor.
*Corresponding author. Tel.: +61 3 5227 1439; Fax: +61 3
5227 1040; E-mail: scam@deakin.edu.au
0968-0896/98/$Ðsee front matter # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(98)00093-5

1518
P. J. Scammells et al./Bioorg. Med. Chem. 6 (1998) 1517±1524
Figure 1.
Results and Discussion
we attempted to brominate 5b in the hope that the
smaller halogen would give the desired regioselectivity.
The careful addition of bromine (1.1 molar equivalents)
to compound 5b at 0 ꢀC produced one major product.
The synthetic approach used to prepare a 2-phenyl
analogue of the natural product, XH-14 (1) is outlined
in Scheme 1. Brie¯y, this involved coupling ethyl 3-
methoxy-4-hydroxy-5-bromocinnamate (3) with cuprous
phenylacetylide to a€ord 2-phenyl-7-methoxy-5-(2-ethoxy-
carbonyl-E-ethenyl)benzo[b]furan (4). Reduction of the
a,b-unsaturated ester moiety using lithium aluminium
hydride followed by protection of the resultant alcohol
yielded the benzo[b]furan 5b. Introduction of a 3-formyl
substituent was attempted using the modi®ed Gatterman±
Adams reaction developed by Yang et al.6 A range of
conditions were trialed, but in each case only depro-
tected starting material (5a) was isolated. In order to
overcome this problem an indirect method involving
halogenation and formylation8,9 of the resultant aryl
halide was considered. Iodination of 5b using iodide and
silver sulphate a€orded a crystalline product. NMR and
MS spectra con®rmed that iodination had occurred, but
the position of the iodide was unable to be unambigu-
ously assigned. A crystal structure was obtained10 that
indicated the iodine substituted for H4 rather than
reacting at the 3-position as desired. Modelling studies
on the benzo[b]furan 1 and related compounds indicated
the 3-position was electronically favoured for electro-
philic substitution, but the 4-position was sterically
favoured.6 This postulation was supported by experi-
mental data in which bulky electrophiles reacted at C4
and small electrophiles at C3. Based on this information
1
Comparison of the H NMR with that of 6a indicated
the bromine had also substituted at the 4-position.
Other approaches involving halogenation-formylation-
dehalogenation of the benzofuran 5b or coupling of the
ortho-bromophenol 3 with phenylpropargyl aldehyde
under the conditions developed by Larock11 were con-
templated. Fortunately closely related benzofurans were
able to be formylated (see Schemes 2 and 3) which
enabled the importance of the 3-formyl group for ade-
nosine receptor anity to be assessed and obviated the
need to persist with these alternatives.
In order to prepare an analogue of XH-14 (1) with a
propyl (rather than 3-hydroxypropyl) side chain via the
cuprous acetylide coupling approach it was necessary to
prepare the ortho-bromophenol 9 (Scheme 2). This was
initially attempted via the bromination of 2-methoxy-4-
propylphenol, however regioselectivity problems
induced us to turn to the alternative sequence which
involved bromination of 4-propylphenol followed by
nucleophilic displacement of bromine. Once obtained, 9
was coupled with cuprous (4-benzyloxy-3-methoxyphenyl)
acetylene and the resultant benzo[b]furan deprotected to
a€ord 10b. In contrast to the 2-phenylbenzo[b]furan 5b,
this compound underwent a Gattermann±Adams reac-
tion to give the desired product 11 in 40% yield. The
Scheme 1. (a) PhC=CCu, pyridine; (b) LiAlH4, THF; (c) Ac2O, pyr; (d) see text.

P. J. Scammells et al./Bioorg. Med. Chem. 6 (1998) 1517±1524
1519
Scheme 2. (a) Br2, CH2Cl2; (b) MeOH, BaO, CuCl2, DMF; (c) (4-BnO-3-OMe)PhCꢁCCu, pyridine; (d) H2, Pd/C, AcOH/THF; (e)
Zn(CN)2, KCl, HCl, Et2O then H2O/EtOH.
Scheme 3. (a) CH3CH2COCl; (b) AlCl3, CS2, Á; (c) Zn(Hg), EtOH/aq HCl; (d) Br2, AcOH; (e) (4-BnO-3-OMe)PhCꢁCCu, pyridine;
(f) H2, Pd/C, AcOH, THF; (g) Zn(CN)2, KCl, HCl, Et2O then H2O/EtOH.
higher reactivity of 10b relative to that observed for 5b
may be a result of mesomeric factors associated with the
2-(4-hydroxyphenyl) substituent. Similarly, Yang et al.
argued that XH-14 analogues bearing this substituent at
C2 were more reactive as a result the negative charge of
the phenolate oxygen being delocalised over the 3- and
4-positions, thus making the benzofuran ring system
more electron rich.6
In order to allow a comparison of the receptor anity
of these analogues with the natural product, a supply of
XH-14 (1) was prepared using
a new synthetic
approach12 developed in our laboratory. The ®nal
precursor in this synthesis, 5-(3-hydroxypropyl)-7-meth-
oxy-2-(30-methoxy-40-hydroxyphenyl)benzo[b]furan
(17), was also evaluated at the A1 adenosine receptor.
The anity of the compounds outlined above for the A1
adenosine receptor was measured using a competitive
binding assay which employed [3H]8-cyclopentyl-1,3-
dipropylxanthine ([3H]CPX) as a reference. The results
are listed in Table 1.
A 5,7-dipropylbenzo[b]furan analogue of XH-14 (1) was
prepared using a similar synthetic approach (Scheme 3).
The key intermediate in this synthesis was 6-bromo-2,4-
dipropylphenol (14), which was prepared from 2-propyl-
phenol in four steps. The additional propyl side chain
was constructed by O-acylation, Fries rearrangement,
Clemmensen reduction all of which proceeded in good
yield. Bromination of the resultant dipropylphenol
a€orded 14, which was coupled with cuprous (4-benzyl-
oxy-3-methoxyphenyl)acetylene to form the benzo[b]-
furan ring system. After deprotection, a formyl group
was introduced into the 3-position via a Gattermann±
Adams reaction (49% yield).
Conclusions
The natural product XH-14 (1) was found to bind the
A1 adenosine receptor with relatively high anity
(Ki=50 nM). However, this value for the binding a-
nity of 1 was somewhat lower than the value originally
reported by Yang et al. (IC50=17 nM6). This disparity
maybe attributed to species di€erences between the A1

1520
P. J. Scammells et al./Bioorg. Med. Chem. 6 (1998) 1517±1524
Table 1. Dissociation constants (Ki) for the A1 adenosine
receptor in DDT1 MF-2 cell membranes
over sodium wire and either used fresh or stored over
4 A sieves. All reactions were performed under an inert
(N2) atmosphere. Merck Kieselgel 60 and 60 F254 were
used for column and thin layer chromatography,
respectively. Melting points (mp) were determined on an
Electrothermal melting point apparatus and are uncor-
rected. 1H and 13C NMR spectra were recorded on a
Varian 300 MHz Unity Plus spectrometer using CDCl3
as the solvent and TMS as an internal standard. FAB
mass spectra were measured on a Jeol JMS-DX300 mass
spectrometer and processed on an MSS data system.
Compd
Ki (mM)
1 (XH-14)
5a
0.05‹0.02
9.5‹2.1
4.9‹1.5
>50.0
6a
10b
11
15b
2.2‹90.3
7.4‹0.9
>50.0
16
17 (XH-14-CHO)
8.4‹0.4
5-(20-(Ethoxycarbonyl)-E-ethenyl)-7-methoxy-2-phenyl-
benzo[b]furan (4). Cuprous phenylacetylide (7.91 g,
48 mmol) was dissolved in dry pyridine (120 mL). Ethyl
3-methoxy-4-hydroxy-5-bromocinnamate6 (3, 14.48 g,
48 mmol) in dry pyridine (120 mL) was added and the
mixture was re¯uxed for 20 h. After cooling, the pyr-
idine was evaporated under reduced pressure and the
residue extracted with chloroform (500 mL). The
chloroform was washed with water (150 mL), dried over
magnesium sulphate, ®ltered and evaporated to give the
crude product. Column chromatography on silica gel
using dichloromethane:petroleum ether 40-60 (1:3) as an
eluent a€orded pure 4 (9.48 g, 61%); mp 159±161 ꢀC; 1H
NMR d 1.35 (t, 3H, J=7.0 Hz, CH3), 4.06 (s, 3H,
OCH3), 4.28 (q, 2H, J=7.0 Hz, CH2), 6.41 (d, 2H,
J=15.9 Hz, CH=CH-Ar), 6.98 (s, 1H, ArH), 7.00 (s,
1H, ArH), 7.34 (s, 1H, CH), 7.37 (d, 1H, J=7.3 Hz, H-
40), 7.44 (t, 2H, J=7.3 Hz, H-30,50), 7.75 (d, 1H,
J=15.9 Hz, CH=CH-Ar), 7.87 (d, 2H, J=7.3 Hz, H-
20,60); 13C NMR d 56.2, 60.4, 101.6, 105.6, 114.7, 117.1,
125.1, 128.3, 128.6, 128.9, 129.9, 130.7, 131.2, 145.3,
145.4, 145.5, 157.1.
DDT1 MF-2 cells were incubated with 2.5 nM [3H]CPX and
various concentrations of the compounds for 90 min at 25 ꢀC.
The Ki values were calculated from the concentration of the
compounds that inhibited speci®c [3H]CPX binding by 50%.
Each value is the mean ‹SE, n=3.
adenosine receptors used in the binding assays; we used
cells derived from a steroid-induced leiomyoscarcoma of
the vas deferens of an adult Syrian Hamster, while Yang
and co-workers used bovine cerebral cortex cells as a
source of A1 receptors.
From the structure±activity data it is immediately
apparent that all modi®cations made to XH-14 sig-
ni®cantly diminished receptor anity. Since analogous
modi®cations (incorporation of propyl and phenyl sub-
stituents) produced increased anity in the case of xan-
thine antagonists it is fair to conclude that benzofuran
and xanthine ring systems interact with the adenosine
receptor in a di€erent orientation (i.e. not via a `super-
imposed' model in which the 6:5 fused heterocycles bind
in the same orientation). The importance of the 3-for-
myl substituent was highlighted by a comparison of
compounds with and without this substituent. XH-14
bound the receptor with two orders of magnitude higher
anity than the corresponding analogue which lacked a
formyl group (compare 1 with 17). The 5-propylbenzo-
furan 10b possessed negligible anity for the receptor
(Ki>50 mM), but after a 3-formyl was incorporated
receptor anity increased to 2.2 mM. The hydroxyl
attached to the 5-propyl side chain of XH-14 is also
required for high anity binding. The XH-14 analogue
which was devoid of this hydroxyl (compound 11)
showed a ꢂ40 fold lower receptor anity. Analogues of
5-(3-hydroxypropyl)-7-methoxy-2-phenylbenzo[b]furan
(5a) in which this hydroxyl was lacking (10b) or acetyl-
ated (5b) also exhibited signi®cantly diminished activity.
5-(30-Hydroxypropyl)-7-methoxy-2-phenylbenzo[b]furan
(5a). A solution of lithium aluminium hydride (42.2 mL,
1.0 M, 42.2 mmol) in THF (70 mL) was cooled to
15 ꢀC on an ice bath. 5-(2-(Ethoxycarbonyl)-E-ethenyl)-
7-methoxy-2-phenylbenzo[b]furan (4, 8.0 g, 24.8 mmol)
in THF (124 mL) was added dropwise over 30 min. The
reaction mixture was then stirred at room temperature
for 24 h before being quenched with cold 5% H2SO4.
Extraction with ether (3Â120 mL), washing with brine
(2Â80 mL), drying over Na2SO4 and evaporation
a€orded a crude solid. Puri®cation was e€ected by
column chromatography using petroleum ether:CH-
Cl3:EtOH (40:10:1) followed by recrystallisation from
ethanol/water (5.93 g, 84%); mp 97±98 ꢀC; H NMR d
1
1.71 (s, 1H, OH), 1.95 (m, 2H, CH2CH2CH2OH), 2.79
(t, 2H, J=7.7 Hz, CH2CH2CH2OH), 3.72 (t, 2H,
J=6.4, CH2CH2CH2OH), 4.05 (s, 3H, OCH3), 6.66 (s,
1H, H3), 6.95 (s, 1H, H4/6), 7.01 (s, 1H, H4/6), 7.35±
7.89 (m, 5H, ArH); 13C NMR d 32.4, 34.7, 56.2, 62.3,
101.5, 107.7, 112.5, 125.0, 128.5, 128.7, 130.4, 130.9,
137.6, 142.7, 144.9, 156.21.
Experimental
Solvents were either AR grade or distilled prior to use.
Diethyl ether and pyridine were dried by distillation

P. J. Scammells et al./Bioorg. Med. Chem. 6 (1998) 1517±1524
1521
5-(30-Acetoxypropyl)-7-methoxy-2-phenylbenzo[b]furan
(5b). Compound 5a (1.52 g, 5.38 mmol) was dissolved in
dry pyridine (10 mL). Acetic anhydride (2.5 mL) was
added and the reaction was stirred at room temperature
for 24 h. The solvent was removed under vacuum and
the crude product chromatographed using dichloro-
methane:hexane (2:1) to a€ord the desired product
(1.6 g, 86%); mp 63±64 ꢀC; 1H NMR d 2.02 (m, 2H,
J=6.6, 7.7 Hz, CH2CH2CH2OAc), 2.07 (s, 3H, Ac),
2.76 (t, 2H, J=7.7 Hz, CH2CH2CH2OAc), 4.05 (s, 3H,
OCH3), 4.13 (t, 2H, J=6.6 Hz, CH2CH2CH2OAc), 6.64
(s, 1H, H3), 6.95 (s, 1H, H4/6), 6.99 (br s, 1H, H4/6),
7.31±7.88 (m, 5H, ArH); 13C NMR d 21.0, 30.7, 32.5,
56.2, 63.9, 101.4, 107.6, 112.5, 125.0, 128.5, 128.7, 130.4,
130.9, 136.9, 142.8, 144.9, 156.3, 171.2.
portion (CH2Cl2, 3Â30 mL), the organic portions were
combined, dried and evaporated to a€ord a crude oil.
Puri®cation by chromatography using DCM:hexane
1
(1:1) yielded pure 8 (7.25 g, 88%); H NMR d 0.92 (t,
3H, J=7.4 Hz, CH2CH2CH3), 1.59 (m, 2H, CH2
CH2CH3), 2.47 (t, 2H, J=7.6 Hz, CH2CH2CH3), 5.73
(s, 1H, OH), 7.25 (s, 2H, H3,5); 13C NMR d 13.6, 24.4,
36.5, 109.5, 131.8, 137.3, 147.2.
2-Bromo-6-methoxy-4-propylphenol (9). 2,6-Dibromo-4-
propylphenol (5.49 g, 18.7 mmol) and BaO (14.32 g,
93.4 mmol) were re¯uxed in dry MeOH (100 mL) for 3 h.
The methanol was evaporated under reduced pressure
and the residue was taken up in dry DMF (50 mL).
Copper(II)chloride (2.51 g, 18.7 mmol) was added and
the reaction was brought to re¯ux to ensure all solids
were immersed and temperature was maintained at
100 ꢀC for 4 h. After this period the solvent was removed
under reduced pressure and the residue partitioned
between EtOAc (100 mL) and water (100 mL). The
aqueous portion was further extracted with EtOAc
(2Â100 mL) before the organic portions were combined,
dried (MgSO4), ®ltered and evaporated to give a crude
brown oil. After column chromatography using hex-
ane:DCM (2:1) as an eluent, the desired product was
obtained in 42% yield as an orange oil (1.74 g of starting
material was also recovered from the column); mp 78±
80 ꢀC; 1H NMR d 0.94 (t, 3H, J=7.6 Hz, CH2CH2CH3),
1.61 (m, 2H, J=7.6, 7.6 Hz, CH2CH2CH3), 2.50 (t, 2H,
J=7.6 Hz, CH2CH2CH3), 3.89 (s, 3H, OCH3), 5.76 (s,
1H, OH), 6.62 (s, 1H, H3), 6.91 (br s, 1H, H5); 13C
NMR d 13.7, 24.6, 37.4, 56.2, 107.8, 110.3, 124.1, 135.4,
140.9, 146.9.
5-(30-Acetoxypropyl)-4-iodo-7-methoxy-2-phenylbenzo[b]-
furan (6a). Compound 5b (0.6 g, 1.85 mmol), silver sul-
phate (0.58 g, 1.86 mmol) and iodine (0.47 g,
1.85 mmol) were stirred in dry chloroform (10 mL) for
16 h at room temperature. During this period the col-
our of the reaction mixture was changed from reddish
brown to clear. Water (20 mL) was added and the
reaction mixture was extracted with chloroform
(3Â20 mL). Drying, ®ltration and evaporation of the
organic phase a€orded a yellowish solid. Puri®cation
by column chromatography using dichloromethane:
hexane (2:1) yielded pure 6a (0.81 g, 92%); mp 134±
135 ꢀC; 1H NMR d 1.99 (m, 2H, CH2CH2CH2OAc),
2.10 (s, 3H, Ac), 2.89 (t, 2H, J=7.8 Hz, CH2CH2
CH2OAc), 4.04 (s, 3H, OCH3), 4.17 (t, 2H, J=6.5 Hz,
CH2CH2CH2OAc), 6.72 (s, 1H, H3), 6.95 (s, 1H, H6),
7.27±7.90 (m, 5H, ArH).
5-(30-Hydroxypropyl)-4-bromo-7-methoxy-2-phenylbenzo-
[b]furan (6b). Compound 6a (126 mg, 0.388 mmol) was
dissolved in dichloromethane (8 mL) and cooled to
5 ꢀC on an ice±salt bath. Bromine (68 mg, 0.426 mmol)
was added and the reaction was stirred for 1 h. Eva-
poration of the solvent a€orded a brown solid. Pur-
i®cation via column chromatography using dichloro-
methane:hexane (1:1.25) as an eluent yielded 6b (137 mg,
87%); mp 155±157 ꢀC; 1H NMR d 2.01 (m, 2H,
CH2CH2CH2OAc), 2.09 (s, 3H, Ac), 2.89 (t, 2H,
J=7.7 Hz, CH2CH2CH2OAc), 4.04 (s, 3H, OCH3), 4.16
(t, 2H, J=6.6 Hz, CH2CH2CH2OAc), 6.68 (s, 1H, H3),
7.03 (s, 1H, H6), 7.37±7.89 (m, 5H, ArH); 13C NMR d
21.0, 29.3, 32.3, 56.4, 63.8, 102.2, 105.7, 109.0, 125.1,
128.8, 129.8, 132.5, 135.4, 142.2, 144.3, 156.6, 171.2.
2-(40-Benzyloxy-30-methoxyphenyl)-7-methoxy-5-propyl-
benzo[b]furan (10a). A solution of cuprous acetylide
(2.33 g, 7.75 mmol) in dry pyidine (40 mL) was added to
2-bromo-6-methoxy-4-propylphenol (1.9 g, 7.75 mmol)
in dry pyridine (20 mL) and the mixture was re¯uxed
for 15 h. The pyridine was removed under reduced
pressure and the residue was taken up in chloroform
(50 mL) and washed with water (50 mL). The aqueous
layer was extracted with chloroform (2Â50 mL) and the
combined chloroform portions were dried, ®ltered and
evaporated under reduced pressure to give a brown oil.
This was puri®ed by column chromatography
(DCM:hexane, 1:1) to a€ord the product as a white
solid (2.27 g, 73%); mp 115±118 ꢀC; H NMR d 0.98 (t,
1
3H, J=6.4 Hz, CH2CH2CH3), 1.70 (m, 2H, CH2
CH2CH3), 2.66 (t, 2H, J=7.6 Hz, CH2CH2CH3), 4.00
(s, 3H, OCH3), 4.04 (s, 3H, OCH3), 5.21 (s, 2H,
CH2Ph), 6.62 (s, 1H, H3), 6.83±7.48 (m, 10H,
H4,6,20,50,60, CH2Ph); 13C NMR d 13.9, 25.1, 38.4, 56.1,
56.2, 71.0, 100.4, 107.3, 108.6, 112.3, 113.9, 117.9, 124.0,
127.3, 127.9, 128.6, 130.6, 136.9, 138.4, 142.4, 144.6,
148.6, 149.7, 156.1.
2,6-Dibromo-4-propylphenol
(8).
4-Propylphenol
(27.9 mmol) was dissolved in dichloromethane (30 mL)
and cooled to 0 ꢀC. Bromine (8.92 g, 55.8 mmol) was
added dropwise and the reaction was stirred for 2 h at
0 ꢀC. Water (20 mL) was added and the organic layer
separated. After further extraction of the aqueous

1522
P. J. Scammells et al./Bioorg. Med. Chem. 6 (1998) 1517±1524
2-(40-Hydroxy-30-methoxyphenyl)-7-methoxy-5-propyl-
benzo[b]furan (10b). Compound 10a (150 mg,
propanoyl chloride (3.30 mL, 33.5 mmol) was added and
the mixture stirred for a further hour at 50 ꢀC. The
resulting oil was distilled under vacuum to a€ord the
product as a translucent liquid (9.5 g, 73%); bp 84±86 ꢀC
0.373 mmol) was dissolved in THF (10 mL). Acetic acid
(glacial, ꢂ0.1 mL) and palladium on carbon (10%,
ꢂ20 mg) were added and the reaction was placed under
an atmosphere of hydrogen. After stirring for 16 h at
ambient temperature, the reaction mixture was ®ltered
and evaporated to a€ord a colourless oil. Column
chromatography using dichloromethane:hexane (1:1) as
an eluent gave pure 10b as a colourless oil, which crystal-
(1.3±1.4 mmHg); 1H NMR
d 0.93 (t, J=7.5 Hz,
CH2CH2CH3), 1.29 (t, J=7.6 Hz, COCH2CH3), 1.59
(m, J=7.5 Hz, CH2CH2CH3), 2.48 (t, J=7.5 Hz,
CH2CH2CH3), 2.60 (q, J=7.6 Hz, COCH2CH3), 7.01±
7.24 (m, 4H, ArH); 13C NMR d 9.2, 13.9, 23.1, 27.7, 32.2,
122.2, 125.8, 126.8, 130.2, 134.2, 148.9, 172.9.
lised upon standing (105 mg, 90%); mp 106±107 ꢀC; H
1
NMR d 0.99 (t, J=7.3 Hz, 3H, CH2CH2CH3), 1.71 (m,
2H, CH2CH2CH3), 2.67 (t, 2H, J=7.6 Hz,
CH2CH2CH3), 3.99 (s, 3H, OCH3), 4.05 (s, 3H, OCH3),
5.79 (s, 1H, OH), 6.63 (s, 1H, H3), 6.82 (s, 1H, H4/6), 6.97
(s, 1H, H4/6), 6.99 (d, 1H, J=8.3 Hz, H50), 7.38 (s, 1H,
H20), 7.41 (d, 1H, J=8.3 Hz, H60); 13C NMR d 13.9, 25.1,
38.4, 56.0, 56.1, 100.1, 107.3, 107.6, 112.3, 114.7, 118.8,
123.1, 131.0, 138.4, 142.3, 144.6, 146.2, 146.7, 156.3.
40-Hydroxy-30-propylphenylpropanone. 20-Propylphenyl-
propanoate (2.3 g, 12.2 mmol) was added to a stirred
mixture of aluminium chloride (1.8 g, 13.5 mmol) in dry
carbon disul®de (12 mL) and the mixture gently re¯uxed
for 2 h. The solvent was then removed by distillation,
the mixture stirred for as long as possible with heating
at 140±150 ꢀC (for ꢂ3 h), and allowed to cool a€ording
a hard brown mass. This was dissolved in methanol
(20 mL), hydrochloric acid (1:1, 16 mL) and water
(24 mL) were added slowly, and the mixture allowed to
stand overnight. The product was extracted with diethyl
ether (3Â100 mL) and the solvent removed under
reduced pressure. The resulting brown solid was dis-
solved in 3 M sodium hydroxide (100 mL) and the solu-
tion washed with ether (3Â100 mL). The aqeous layer
was neutralised with HCl, the product extracted with
ether (3Â100 mL), and the solvent removed under
reduced pressure. The solid was puri®ed by column
chromatography (petroleum ether:ethyl acetate, 7:2) to
a€ord the product as a white solid (1.2 g, 51%); mp
2-(40-Hydroxy-30-methoxyphenyl)-7-methoxy-5-propyl-
benzo[b]furan-3-carbaldehyde (11). The benzo[b]furan
10b (91 mg, 0.291 mmol) was dissolved in dry ether
(10 mL) and cooled on an ice±salt bath in a 50 mL, three
necked round bottomed ¯ask. A HCl generator (concd
H2SO4 onto solid NH4Cl) in series with a H2SO4 wash
bottle and a safety bottle was connected to the ¯ask.
Attached to another joint was a condenser in series with
a H2SO4 wash bottle and a safety bottle connected to
the surface of aqueous NaOH. Zinc cyanide (51 mg,
0.437 mmol) and potassium chloride (15 mg) were added
and HCl gas was bubbled through the solution for 1 h.
The ethereal solution was combined with ethanol
(10 mL) and water (20 mL) and heated at 50 ꢀC for 1 h.
Extraction with ethyl acetate (3Â40 mL) followed by
drying (MgSO4), ®ltration and evaporation a€orded a
crude yellow oily solid. Puri®cation was achieved via
column chromatography with dichloromethane as an
eluent. The pure product was a pale-yellow solid weigh-
ing 40 mg (40%); 1H NMR d 0.97 (t, 3H, J=7.3 Hz,
CH2CH2CH3), 1.71 (m, 2H, CH2CH2CH3), 2.70 (t, 2H,
J=7.6 Hz, CH2CH2CH3), 4.00 (s, 3H, OCH3), 4.03 (s,
3H, OCH3), 6.01 (s, 1H, OH), 6.73 (s, 1H, H4/6), 7.07
(d, 1H, J=8.3 Hz, H50), 7.38 (s, 1H, H20), 7,40 (d, 1H,
J=8.3 Hz, H60), 7.65 (s, 1H, H4/6), 10.3 (s, 1H, CHO);
13C NMR d 13.8, 25.1, 38.5, 56.0, 56.3, 108.8, 111.0,
113.6, 115.0, 116.8, 120.7, 123.6, 127.1, 140.8, 141.5,
144.4, 146.9, 148.5, 165.9, 186.8; HR MS (FAB) calcd
341.13889, found 341.14012.
77±78.0 ꢀC; 1H NMR
d 0.98 (t, 3H, J=7.3 Hz,
CH2CH2CH3), 1.22 (t, 3H, J=7.3 Hz, COCH2CH3),
1.67 (qt, 2H, J=7.3, 7.7 Hz, CH2CH2CH3), 2.63 (t, 2H,
J=7.7 Hz, CH2CH2CH3), 2.97 (q, 2H, J=7.3 Hz,
COCH2CH3), 6.11 (br s, 1H, OH), 6.84 (d, 1H,
J=8.3 Hz, ArH), 7.75 (d, 1H, J=8.3 Hz, ArH), 7.80 (s,
1H, ArH); 13C NMR d 8.7, 13.9, 22.7, 31.4, 31.9, 115.1,
128.2, 130.9, 129.0, 129.1, 159.1, 201.5.
2,4-Dipropylphenol. A solution of mercuric chloride
(2.1 g) in water (150 mL) was added to mossy zinc
(107.4 g) and the mixture agitated occasionally, then
allowed to stand overnight. The supernatant liquid was
decanted o€ and the zinc rinsed once with water. A
mixture of water (100 mL) and concentrated hydro-
chloric acid (100 mL) and a solution of 40-hydroxy-30-
propylphenylpropanone (7.7 g, 39.7 mmol) in ethanol
were added the zinc, and the mixture stirred vigorously
with re¯uxing for 69 h. Toluene (50 mL) was added to
the resulting mixture and stirring was continued for a
few minutes. The solution was ®ltered and washed with
water (3Â50 mL). The solvent was removed by distilla-
tion and the resulting brown liquid puri®ed by column
chromatography (petroleum ether/ethyl acetate, 5:1) to
a€ord the desired product as an orange oil (5.8 g, 81%);
2,4-Dipropylphenol (13)
20-Propylphenylpropanoate. Propanoyl chloride (7.40 mL,
75.1 mmol) was added to stirred 2-propylphenol over a
period of 30 min, and the mixture stirred at room tem-
perature for 6 h. Starting material still present. Extra

P. J. Scammells et al./Bioorg. Med. Chem. 6 (1998) 1517±1524
1523
1H NMR d 0.94 (t, 3H, J=7.3 Hz, CH2CH2CH3), 0.99
(t, 3H, J=7.3 Hz, CH2CH2CH3), 1.64 (m, 4H,
J=7.3,7.6,7.7 Hz, 2ÂCH2CH2CH3), 2.51 (t, 2H,
J=7.6 Hz, CH2CH2CH3), 2.58 (t, 2H, J=7.7 Hz,
CH2CH2CH3), 4.66 (br s, 1H, OH), 6.70 (d, 2H,
J=8.0 Hz, ArH), 6.90 (d, 2H, J=8.0 Hz, ArH), 6.94 (s,
1H, ArH); 13C NMR d 13.8, 14.0, 23.0, 24.8, 32.1, 37.3,
115.0, 126.8, 128.0, 130.3, 134.9, 151.4.
and the reaction was placed under an atmosphere of
hydrogen. After stirring for 16 h at ambient tempera-
ture, the reaction mixture was ®ltered and evaporated to
a€ord a colourless oil. Column chromatography using
dichloromethane:hexane (1:1) as an eluent gave pure
1
15b as a white solid (136 mg, 87%); H NMR d 0.98 (t,
2H, CH2CH2CH3), 1.04 (t, 2H, CH2CH2CH3), 1.70 (m,
2H, CH2CH2CH3), 1.85 (m, 2H, CH2CH2CH3), 2.66 (t,
2H,CH2CH2CH3), 2.92 (t, 2H, CH2CH2CH3), 4.01 (s,
3H, OCH3), 5.78 (s, 1H, OH), 6.83 (s, 1H, H3), 6.91 (d,
1H, J=1.1 Hz, H4/6), 7.01 (d, 1H, J=8.2 Hz, H50), 7.19
(d, 1H, J=1.1 Hz, H4/6), 7.34 (d, 1H, J=1.7 Hz, H20),
7.42 (dd, 1H, J=1.7, 8.2 Hz, H60); 13C NMR d 13.8,
14.0, 23.1, 25.2, 31.7, 38.1, 56.2, 111.0, 115.1, 116.8,
119.1, 121.1, 123.4, 125.2, 125.4, 126.7, 139.6, 146.9,
148.5, 151.1, 165.5, 186.9; HR MS (FAB) calcd
325.18036, found 325.18083.
6-Bromo-2,4-dipropylphenol (14). 2,4-Dipropylphenol
(5.2 g, 29.0 mmol) was added slowly to bromine (7.0 g,
43.7 mmol) in 90% acetic acid (50 mL), and the mixture
stirred at room temperature for 10 h. Water (100 mL)
was added to the mixture, the product extracted with
ethyl acetate (3Â100 mL) and washed with water
(3Â100 mL). The solvent was removed by reduced pres-
sure and the resulting crude brown liquid was puri®ed
by column chromatography (petroleum ether:chloro-
form, 2:1) to a€ord the product as an orange oil (4.5 g,
5,7-Dipropyl-2-(40-hydroxy-30-methoxyphenyl)benzo[b]-
furan-3-carbaldehyde (16). The same apparatus descri-
bed for the preparation of 11 was used for this
reaction. The benzo[b]furan 15b (214 mg, 0.660 mmol)
was dissolved in dry ether (15 mL) and cooled on an
ice-salt bath. Zinc cyanide (116 mg, 0.988 mmol) and
potassium chloride (20 mg) were added and HCl gas was
bubbled through the solution for 1 h. The ethereal solu-
tion was combined with ethanol (10 mL) and water
(20 mL) and heated at 50 ꢀC for 1 h. Extraction with
ethyl acetate (3Â40 mL) followed by drying, ®ltration
and evaporation a€orded a crude yellow oily solid.
After puri®cation by column chromatography with
dichloromethane as an eluent the product was isolated
61%); 1H NMR
d 0.93 (t, 3H, J=7.4 Hz, CH2
CH2CH3), 0.96 (t, 3H, J=7.4 Hz, CH2CH2CH3), 1.62
(m, 4H, J=7.4,7.6,7.7 Hz, 2ÂCH2CH2CH3), 2.48 (t, 2H,
J=7.6 Hz, CH2CH2CH3), 2.62 (t, 2H, J=7.7 Hz,
CH2CH2CH3), 5.37 (s, 1H, OH), 6.87 (s, 1H, ArH), 7.12
(s, 1H, ArH); 13C NMR d 13.7, 14.0, 22.8, 24.6, 32.9,
36.9, 110.1, 128.9, 129.9, 135.7, 148.0.
5,7-Dipropyl-2-(40-benzyloxy-30-methoxyphenyl)benzo[b]-
furan (15a). A solution of cuprous (4-benzyloxy-3-
methoxy)phenylacetylide (0.64 g, 2.14 mmol) in dry pyr-
idine (5 mL) was added to 6-bromo-2,4-dipropylphenol
(0.55 g, 2.14 mmol) in dry pyridine (5 mL). The mixture
was re¯uxed for 24 h. The pyridine was removed under
reduced pressure and the residue was taken up in
chloroform (50 mL) and washed with water (50 mL).
The aqueous layer was extracted with chloroform
(2Â50 mL) and the combined chloroform portions were
dried, ®ltered and evaporated under reduced pressure to
give a brown oil. The crude product was puri®ed by
column chromatography using hexane:DCM (3:1); mp
ꢀ
1
(113 mg, 49%); mp 103±104 C; H NMR d 0.96 (t, 2H,
CH2CH2CH3), 1.01 (t, 2H, CH2CH2CH3), 1.73 (m, 2H,
CH2CH2CH3), 1.81 (m, 2H, CH2CH2CH3), 2.69 (t,
2H,CH2CH2CH3), 2.90 (t, 2H, CH2CH2CH3), 4.02 (s,
3H, OCH3), 6.06 (s, 1H, OH), 7.02 (s, 1H, H4/6), 7.10
(d, 1H, J=8.6 Hz, H50), 7.36 (d, 1H, J=2.0 Hz, H20),
7.43 (dd, 1H, J=2.0, 8.6 Hz, H60), 7.89 (s, 1H, H4/6),
10.3 (s, 1H, CHO); 13C NMR d 13.8, 14.0, 23.1, 25.2,
31.7, 38.1, 56.2, 111.0, 115.1, 116.8, 119.1, 121.1, 123.4,
125.2, 125.4, 126.7, 139.6, 146.9, 148.5, 151.1, 165.5,
186.9; HR MS (FAB) calcd 353.17529, found
353.17617.
97±98 ꢀC; 1H NMR
d 0.98 (t, 3H, J=7.5 Hz,
CH2CH2CH3), 1.04 (t, 3H, J=7.4 Hz, CH2CH2CH3),
1.70 (m, 2H, CH2CH2CH3), 1.85 (m, 2H, CH2CH2CH3),
2.66 (t, 2H, J=7.6 Hz, CH2CH2CH3), 2.92 (t, 2H,
J=7.6 Hz, CH2CH2CH3), 4.01 (s, 3H, OCH3), 5.22 (s,
2H, CH2Ph), 6.84 (s, 1H, H3), 6.91 (s, 1H, H4/6), 6.96
(d, 1H, J=9.0 Hz, H50), 7.19 (s, 1H, H4/6), 7.30±7.52
(m, 7H, H20,60,CH2Ph); 13C NMR d 13.9, 14.2, 23.1,
25.2, 32.0, 38.1, 56.1, 71.0, 100.3, 108.6, 114.1, 117.5,
117.8, 124.5, 124.9, 125.4, 127.2, 127.9, 128.6, 129.0,
136.9, 137.3, 148.5, 149.8, 152.1, 155.5.
Receptor binding determinations. DDT1 MF-2 cells were
grown and cell membranes prepared as described pre-
viously.13 The dissociation constants of the compounds
were determined from displacement of radioligand
binding assays. Brie¯y, cell membranes (0.05 mg pro-
tein) were incubated in a total volume of 0.25 mL con-
taining 50 mM Tris±HCl bu€er at pH 7.4, 5 mM MgCl2,
2.5 nM [3H]8-cyclopentyl-1,3-dipropylxanthine (CPX,
120 Ci/mmol, New England Nuclear) and without or
with varying concentrations of the compounds for
90 min at 25 ꢀC. Nonspeci®c binding was determined in
5,7-Dipropyl-2-(40-hydroxy-30-methoxyphenyl)benzo[b]furan
(15b). Compound 15a (200 mg, 0.483 mmol) was dis-
solved in THF (20 mL). Acetic acid (glacial, ꢂ0.1 mL)
and palladium on carbon (10%, ꢂ20 mg) were added

1524
P. J. Scammells et al./Bioorg. Med. Chem. 6 (1998) 1517±1524
parallel assays without the compounds but containing
10 mM 8-cyclopentyltheophylline (CPT). At the end of
the incubation, each suspension was diluted with 3 mL
of ice-cold incubation bu€er and poured onto a glass
®ber ®lter under reduced pressure. The ®lters were
washed with a further 6 mL of ice-cold incubation buf-
fer, placed in a vial with 3 mL of scintillation ¯uid and
the radioactivity counted in a liquid scintillation coun-
ter. Speci®c binding to the A1-adenosine receptor was
calculated as the di€erence between total binding in the
absence of CPT and nonspeci®c binding determined in
the presence of CPT. The concentration of compounds
which inhibited speci®c [3H]CPX binding by 50% (IC50)
was determined from a nonlinear regression analysis of
the concentration-inhibition data (Prism, GraphPad)
and the Ki for each compound was calculated from the
IC50 using the method of Cheng and Pruso€.14 Each
binding experiment was performed in triplicate and
repeated at least three times.
References
1. Daly, J. W. J. Med. Chem. 1982, 25, 197.
2. Jacobson, K. A.; van Galen, P. J. M.; Williams, M. J. Med.
Chem. 1992, 35, 407.
3. Williams, M. Drug News & Perspectives, 1991, 4, 5.
4. Collis, M. J.; Hourani, S. M. O. TIPS Reviews, 1993, 14, 360.
5. Yang, Z.; Hon, P. M.; Chui, K. Y.; Xu, Z. L.; Chang, H.
M.; Lee, C. M.; Cui, Y. X.; Wong, H. N. C.; Poon, C. D.;
Fung, B. M. Tetrahedron 1991, 32, 2061.
6. Yang, Z.; Liu, H. B.; Lee, C. M.; Chang, H. M.; Wong, H.
N. C. J. Org. Chem. 1992, 57, 7428.
7. Jacobson, K. A.; Kiriasis, L.; Barone, S.; Bradbury, B. J.;
Kammula, U.; Campagne, J. M.; Secunda, S.; Daly, J. W.;
Neumeyer, J. L.; P¯eiderer, W. J. Med. Chem. 1989, 32, 1873.
8. Baillargeon, V. P.; Stille, J. K. J. Am. Chem. Soc. 1986, 108,
452.
9. Okano, T.; Harada, N.; Kiji, J. Bull. Chem. Soc. Jpn. 1994,
67, 2329.
10. Scammells, P. J.; Tiekink, E. Z. Krist. 1998, 213, 59.
11. Larock, R. C.; Yum, E.; Doty, M. J.; Sham, K. K. C. J.
Org. Chem. 1995, 60, 3270.
Acknowledgements
Financial support from the Australian Research Coun-
cil, the Clive & Vera Ramacotti Foundation and the
American Heart Association (Florida Aliate) is grate-
fully acknowledged. The authors also wish to thank Ray
A. Olsson and Richard A. Russell for providing helpful
suggestions and discussion.
12. Hutchinson, S. A.; Luetjens, H.; Scammells, P. J. Bioorg.
Med. Chem. Lett. 1997, 7, 3081.
13. Scammells, P. J.; Baker, S. P.; Berlardinelli, L.; Olsson, R.
A. J. Med. Chem. 1994, 37, 2704.
14. Cheng, Y.; Pruso€, S. Biochem. Pharmacol. 1973, 22, 3099.