Synthesis and β-adrenoceptor agonist properties of (±)-1-(3',4'- dihydroxyphenoxy)-3-(3',4'-dimethoxyphenyl) ethylamino-2-propanol hydrochloride, (±)-RO363.HCl, and the (2S)-(-)-isomer

Article abstract of DOI:10.1016/S0223-5234(99)80103-2

The synthesis of (±)-1-(3',4'-dihydroxyphenoxy)-3-(3',4'- dimethoxyphenyl)ethylamino-2-propanol hydrochloride, (±)RO363.HCl, and the (2S)-(-)-isomer is described for the first time. The binding affinities for (±)-RO363.HCl, (2S)-(-)-RO363.HCl and a number of well known β-adrenoceptor agonists for transfected humanβ₁,-, β₂- and β₃-adrenoceptors expressed in Chinese hamster ovary cells have been determined and compared with the functional potencies in rat atria (β₁) and trachea (β₂). The results indicate that both (±)-RO363 and (2S)-(-)-RO363 are selective for the human and rat β₁-adrenoceptors. The (2S)-(-)-isomer of RO363, as expected, has a higher binding affinity for the human and functional potency for rat β- adrenoceptor subtypes than the racemate. However, in contrast to the catecholamines and formoterol, the functional potency of the racemic mixture and its (-)-enantiomer are not significantly different from their binding affinity, suggesting that they are examples of partial agonists with sufficient intrinsic activity to produce full agonist responses.

Full text of DOI:10.1016/S0223-5234(99)80103-2

Eur. J. Med. Chem. 34 (1999) 539−548
© Elsevier, Paris
539
Laboratory note
Synthesis and â−adrenoceptor agonist properties
of (±)−1−(3≠,4≠-dihydroxyphenoxy)−3−(3″,4″−dimethoxyphenyl)
ethylamino−2−propanol hydrochloride, (±)−RO363.HCl, and the (2S)−(−)−isomer
Dimitri Iakovidis, Simon N. S. Louis, Linda A. Rezmann, Felicia Colagrande,
Tracy L. Nero, Graham P. Jackman, William J. Louis*
University of Melbourne, Clinical Pharmacology and Therapeutics Unit, Department of Medicine,
Austin and Repatriation Medical Centre, Heidelberg 3084, Victoria, Australia
(Received 01 September 1998; accepted 20 January 1999)
Abstract – The synthesis of (±)-1-(3≠,4≠-dihydroxyphenoxy)-3-(3″,4″-dimethoxyphenyl)ethylamino-2-propanol hydrochloride, (±)-
RO363.HCl, and the (2S)-(-)-isomer is described for the first time. The binding affinities for (±)-RO363.HCl, (2S)-(-)-RO363.HCl and a
number of well known â-adrenoceptor agonists for transfected human â1-, â2- and â3-adrenoceptors expressed in Chinese hamster ovary cells
have been determined and compared with the functional potencies in rat atria (â1) and trachea (â2). The results indicate that both (±)-RO363
and (2S)-(-)-RO363 are selective for the human and rat â1-adrenoceptors. The (2S)-(-)-isomer of RO363, as expected, has a higher binding
affinity for the human and functional potency for rat â-adrenoceptor subtypes than the racemate. However, in contrast to the catecholamines
and formoterol, the functional potency of the racemic mixture and its (-)-enantiomer are not significantly different from their binding affinity,
suggesting that they are examples of partial agonists with sufficient intrinsic activity to produce full agonist responses. © Elsevier, Paris
(±)-RO363 / (2S)-(-)-RO363 / â1-adrenoceptor agonist / chiral synthesis / CHO cells
1. Introduction
â1- and â3-adrenoceptors. In addition, they report that
(2S)-(-)-RO363 is able to activate a novel human atrial
â-adrenoceptor, which they have designated as a putative
â4-adrenoceptor [4, 5].
The first pharmacological reports on (±)-1-(3≠,4≠-
dihydroxyphenoxy)-3-(3″,4″-dimethoxyphenyl)ethyl-
amino-2-propanol ((±)-RO363, figure 1), a synthetic cat-
echolamine, appeared in 1978 [1] and it has been
described as a potent, selective â1-adrenoceptor agonist
in the guinea-pig, rat, rabbit and cat [1–3]. More recently,
work carried out by Molenaar et al. [4, 5] suggests that
(2S)-(-)-RO363 is a potent partial agonist for both human
The original synthesis of (±)-RO363, and the sub-
sequent resolution of the (2S)-(-)- and (2R)-(+)-
enantiomers, was carried out by D. Iakovidis as part of
his doctoral studies [6] and has never been formally
reported. In light of the renewed interest in RO363 as a
research tool, we describe here the synthesis of (±)-
RO363 (7, figure 2) using a shorter (5 steps) and simpler
procedure than the original 9 step process [6]. We also
report the chiral synthesis of (2S)-(-)-RO363 (11, fig-
ure 3). Both (±)-RO363 7 and (2S)-(-)-RO363 11 have
been prepared as the hydrochloride salts.
*Correspondence and reprints
Abbreviations:
thoxyphenyl)ethylamino-2-propanol, RO363; Chinese hamster
ovary, CHO; m-chloroperbenzoic acid, MCBA; 1-(1H-2-cyano-3-
iodoindol-4-yloxy)-3-[(1-methylethyl)amino]-2-propanol,
cyanopindolol, ICYP; sodium-4-[2-[2-hydroxy-2-(3-chlorophenyl)
ethylamino]propyl]phenoxyacetate, BRL37344; 2-(t-butylamino)-
1-(4-hydroxy-3-hydroxymethylphenyl)ethanol, salbutamol; N-[2-
hydroxy-5-[1-hydroxy-2[[2-(p-methoxyphenyl)-2-propyl]amino]-
ethyl]phenyl]formamide, formoterol; 1-(3,4-dihydroxyphenyl)-2-
isopropylaminoethanol, isoprenaline; Kd, dissociation constant;
Bmax, maximal density of binding sites.
1-(3≠,4≠-dihydroxyphenoxy)-3-(3″,4″-dime-
iodo-
The binding affinity of (±)-RO363 7 and (2S)-(-)-
RO363 11 for human â1-, â2- and â3-adrenoceptors
transfected and expressed in Chinese hamster ovary
(CHO) cells and their functional potency in rat atria (â1)
and trachea (â2) have been determined and compared to

540
Figure 1. Structures of (a) RO363, (b) noradrenaline, (c) isoprenaline, (d) salbutamol, (e) formoterol and (f) BRL37344. The asterisks
denote a chiral centre.
other well known â-adrenoceptor agonists. In addition,
we have examined the functional potency of (-)-
isoprenaline and (2S)-(-)-RO363 in guinea-pig atria and
trachea.
assigned structure of (±)-RO363.HCl was fully confirmed
by MS and H-NMR spectra and elemental analysis.
1
To synthesize the (2S)-(-)-isomer of RO363 11, epi-
chlorohydrin was replaced with (2S)-(+)-glycidyl-3-
nitrobenzenesulfonate 8 (figure 3) to produce the (2S)-
(+)-epoxide 9. The optical purity of (2S)-(+)-glycidyl-3-
nitrobenzenesulfonate 8 is > 99% and is commonly used
in the homochiral synthesis of â-adrenoceptor agonists
and antagonists [8]. The optical purity of the protected
benzyl derivatives was confirmed to be > 99% by chiral
HPLC. The assigned structure of (2S)-(-)-RO363.HCl 11
2. Chemistry
The synthetic routes used to prepare (±)-RO363.HCl 7
and (2S)-(-)-RO363.HCl 11 are outlined in figures 2 and
3. Commercially available 3,4-dibenzyloxybenzaldehyde
1
1
was oxidized with m-chloroperbenzoic acid
was fully confirmed by MS and H-NMR spectra and
(MCBA) [7] to produce the corresponding formate 2,
which after mild hydrolysis produced the desired phenol
3.
elemental analysis.
The phenol 3 reacted with epichlorohydrin under
alkaline conditions to produce the epoxide 4 (figure 2).
The ring opening addition reaction of the epoxide 4 with
the amine 5, followed by the conversion of the free base
to the hydrochloride salt gave the protected form of the
final compound 6. The benzyl protecting groups were
removed by hydrogenation to give (±)-RO363.HCl 7. The
3. Pharmacology
The binding affinities for transfected human â1-, â2-
and â3-adrenoceptors expressed in CHO cells were de-
termined for the endogenous catecholamine (-)-
noradrenaline, the non-selective agonist (-)-isoprenaline,
the â1-selective agonist (±)-RO363 7 and the isomer

541
Figure 2. Synthesis of (±)-RO363.HCl 7. Reagents: (a) MCBA, CH2Cl2; (b) KOH, MeOH; (c) epichlorohydrin, KOH, EtOH; (d)
EtOH; (e) EtOH, ethereal HCl; (f) H2, Pd/C.
(2S)-(-)-RO363 11, the â2-selective agonists (±)-
salbutamol and (±)-formoterol and the â3-selective ago-
nist (±)-BRL37344 (the structures of all agonists are
given in figure 1). The functional potencies of all seven
agonists were also determined in rat spontaneously beat-
ing atria (where responses are due to â1-adrenoceptor
stimulation) and rat trachea previously contracted with
1 µM carbachol (where responses are due to â2-
adrenoceptor stimulation). The pharmacological results
are listed in table I. In addition, we have examined the
functional potency of (2S)-(-)-RO363 11 and (-)-
isoprenaline for guinea-pig spontaneously beating atria
and guinea-pig trachea previously contracted with
300 nM carbachol.
(ICYP) for the transfected human â1-, â2- and â3-
adrenoceptor subtypes were determined by saturation
binding experiments. For human â1-adrenoceptors the Kd
was 4.99 ± 0.48 pM and the Bmax was 7 127 ± fmol mg
protein–1; for the â2-adrenoceptor the Kd was 8.00 ± 1.10
pM and the Bmax was 3 914 ± 583 fmol mg protein–1; and
for the â3-adrenoceptor the Kd was 313 ± 93.8 pM and the
Bmax was 2 325 ± 322 fmol mg protein–1. For all cell
lines, Scatchard analysis gave a straight line (nH ≈ 1.0)
consistent with the presence of a single â-adrenoceptor
subtype.
The binding affinities for the seven compounds were
determined for the human â1-, â2- and â3-adrenoceptors
in competition with the radioligand ICYP. As expected all
compounds displayed monophasic displacement curves
(nH ≈ 1.0) and of the compounds tested, (2S)-(-)-
RO363 11 had the highest binding affinity for human
â1-adrenoceptors (pKi = 8.71; table I), while (±)-
formoterol displayed the highest affinity for human â2-
adrenoceptors (pKi = 6.98, table I) and (±)-BRL37344
had the highest binding affinity for human â3-
4. Results and discussion
4.1. Binding at human â-adrenoceptors
The dissociation constants (Kd) and maximal density of
binding sites (Bmax) of (-)-[125I]-iodocyanopindolol

542
Figure 3. Chiral synthesis of (2S)-(-)-RO363.HCl 11. Reagents: (a) MCBA, CH2Cl2; (b) KOH, MeOH; (c) DMF, NaH; (d) EtOH; (e)
EtOH, ethereal HCl; (f) H2, Pd/C.
adrenoceptors (pKi = 5.84, table I). Overall, the results
are consistent with the reported selectivities of these three
agonists. The relative affinity order for human â1-
adrenoceptors was (2S)-(-)-RO363 11 > (±)-RO363 7 >
(-)-isoprenaline > (-)-noradrenaline > (±)-formoterol >
(±)-salbutamol (±)-BRL37344; for human â2-
adrenoceptors it was (±)-formoterol > (2S)-(-)-RO363 11
> (±)-RO363 7 > (-)-isoprenaline = (±)-salbutamol > (±)-
>
Table I. Agonist binding affinity for human â1-, â2- and â3-adrenoceptors and functional potency at rat â1- and â2-adrenoceptors.
Compound
Human binding affinity
Rat functional potency
â1
â2
â3
atria (â1)
trachea (â2)
pKi
pKi
pKi
pD2
pD2
(-)-isoprenaline
(-)-noradrenaline
(±)-RO363
(2S)-(-)-RO363
(±)-formoterol
(±)-salbutamol
(±)-BRL37344
6.89 ± 0.06
6.02 ± 0.11
8.22 ± 0.03
8.71 ± 0.13
5.97 ± 0.15
5.86 ± 0.25
5.76 ± 0.04
6.24 ± 0.07
4.41 ± 0.15
6.63 ± 0.09
6.87 ± 0.04
6.98 ± 0.25
6.24 ± 0.11
6.10 ± 0.03
5.06 ± 0.07
4.64 ± 0.05
4.23 ± 0.17
4.48 ± 0.12
4.71 ± 0.14
4.20 ± 0.08
5.84 ± 0.05
8.53 ± 0.16
7.25 ± 0.33
8.23 ± 0.10
8.88 ± 0.16
7.10 ± 0.07
5.70 ± 0.01
5.84 ± 0.12
7.05 ± 0.05
5.55 ± 0.29
6.56 ± 0.08
6.59 ± 0.11
7.70 ± 0.25
6.45 ± 0.20
NDa
aValue not determined since (±)-BRL37344 produced variable responses in this tissue.

543
BRL37344 > (-)-noradrenaline; and for â3-adrenoceptors
was (±)-BRL37344 > (-)-isoprenaline > (±)-formoterol >
(-)-noradrenaline > (2S)-(-)-RO363 11 > (±)-RO363 7 ≥
(±)-salbutamol (table I).
4.2. Functional studies in rat tissues
The potency of the agonists for stimulating chronotro-
pic effects in isolated rat atria (â1-mediated), and in
relaxing tracheal chain preparations previously con-
tracted with 1 µM carbachol (â2-mediated), were deter-
mined (table I). (2S)-(-)-RO363 11, (-)-isoprenaline and
(±)-RO363 7 were the most potent compounds for rat
â1-adrenoceptors, whereas (-)-isoprenaline and the â2-
selective agonist (±)-formoterol were the most potent at
rat â2-adrenoceptors. (±)-BRL37344 produced variable
responses in the rat trachea making estimations of the
functional potency (pD2) difficult. In these preparations,
(±)-RO363 7 and (2S)-(-)-RO363 11 were more â1-
selective than (-)-isoprenaline (≈ 1.5-fold and ≈ 6.5-fold,
respectively).
These studies were conducted in the absence of neu-
ronal and extra neuronal uptake inhibitors which might
explain the relatively low potency of (-)-noradrenaline.
However, the literature is unclear on this issue [9, 10] and
suggests that the presence of neuronal and extraneuronal
uptake inhibitors has little or no effect on the potency of
catecholamines in rat atrial preparations [9, 10]. This
mechanism is not important for (-)-isoprenaline, the other
ethanolamine agonists studied, and phenoxypropanola-
mines like (±)-RO363 7 [11].
The relative order of potency of the seven agonists at
rat â1-adrenoceptors was (2S)-(-)-RO363 11 > (-)-
isoprenaline > (±)-RO363 7 > (-)-noradrenaline > (±)-
formoterol > (±)-BRL37344 > (±)-salbutamol; at â2-
adrenoceptors the order was (±)-formoterol > (-)-
isoprenaline > (2S)-(-)-RO363 11 ≥ (±)-RO363 7 > (±)-
salbutamol > (-)-noradrenaline (table I).
The functional potency of (2S)-(-)-RO363 11 for the
rat â1-adrenoceptor was 4.5-fold greater than the potency
of the racemic compound, whereas both the racemate and
the (2S)-(-)-isomer had similar rat â2-adrenoceptor poten-
cies. The agonist efficacy of both (±)-RO363 7 and
(2S)-(-)-RO363 11 were equivalent to (-)-isoprenaline in
the rat atria (â1-mediated) and the concentration-response
curves of the three compounds were parallel (figure 4a),
ie., both (±)-RO363 7 and (2S)-(-)-RO363 11 produce an
equivalent response to a full agonist on rat â1-
adrenoceptors. However, in the rat trachea (â2-mediated)
(±)-RO363 7 and (2S)-(-)-RO363 11 were only able to
achieve 35 ± 6% and 34 ± 7% respectively of the
maximum agonist effect obtained with (-)-isoprenaline.
Figure 4. Concentration-response curves for: a. chronotropic
effects on isolated rat atria; and b. relaxation of rat tracheal
chain previously contracted with 1 µM carbachol by (-)-
isoprenaline (V), (-)-noradrenaline (d), (2S)-(-)-RO363 (M),
(±)-RO363 (j), (±)-salbutamol (∆), (±)-formoterol (▲) and
(±)-BRL 37344 ( ). Values given represent mean ± s.e.m. of
4–9 individual experiments.
The slopes of the (±)-RO363 7 and (2S)-(-)-RO363 11
concentration-response curves were different to that ob-
tained with (-)-isoprenaline (figure 4b), ie., both (±)-
RO363 7 and (2S)-(-)-RO363 11 were acting as partial
agonists on rat â2-adrenoceptors.

544
4.3. Functional studies in guinea-pig tissues
the rat compared to the guinea-pig, so that while 10 nM
RO363 gives a full atrial response in the rat, it produces
virtually no tracheal response. Traditionally, for com-
pounds of the phenoxypropanolamine class, the (2S)-
isomer is more potent than the (2R)-isomer. For RO363,
this is confirmed by our pharmacological results where
the (2S)-(-)-isomer 11 acts as a potent selective â1-
adrenoceptor agonist at the rat â1-adrenoceptor, with a
pD2 of 8.88 (cf the racemate pD2 of 8.23) and a
selectivity ratio â1/â2 of 195 (cf the racemate of 47).
(2S)-(-)-RO363 11 also binds selectively, with high affin-
ity, to the human â1-adrenoceptor with a selectivity ratio
â1/â2 of 69 and â1/â3 of 17 000. By contrast, other
â1-adrenoceptor agonists reported in the litera-
ture [13–15] lack the â1-potency of (2S)-(-)-RO363 11,
and most are either partial â1-adrenoceptor agonists with
low intrinsic efficacy (eg. xamoterol [13, 14]) or possess
no appreciable selectivity (eg. dobutamine [12, 13, 15]).
In the present studies, the rat tracheal chain model
displayed relatively low sensitivity to â2-adrenoceptor
mediated relaxation, which resulted in the non-selective
â-adrenoceptor agonist (-)-isoprenaline displaying ≈ 30-
fold â1-selectivity. We therefore repeated our functional
studies for (-)-isoprenaline and (2S)-(-)-RO363 11 in
guinea-pig atrial and tracheal preparations. In guinea-pig
tissues, both (-)-isoprenaline and (2S)-(-)-RO363 11 had
similar potencies and efficacies for â1-mediated chrono-
tropic effects in atrial preparations (pD2 values = 8.27 ±
0.08 and 8.30 ± 0.07, respectively). In guinea-pig trachea,
precontracted with 300 nM carbachol, (-)-isoprenaline
was ≈ 2.3-fold more potent than (2S)-(-)-RO363 11 at
stimulating â2-mediated relaxation (pD2 values = 7.23 ±
0.04 and 6.86 ± 0.09, respectively). Thus in guinea-pig
tissue (-)-isoprenaline had a â1-selectivity of ≈ 10 com-
pared with ≈ 28 for (2S)-(-)-RO363 11, moreover, (2S)-
(-)-RO363 11 was only able to achieve 70 ± 3% of the
maximum agonist effect obtained with (-)-isoprenaline
and behaved as a partial agonist at â2-adrenoceptors.
6. Experimental protocols
6.1. Chemistry
6.1.1. General
5. Conclusion
Melting points were determined using a manual Gal-
lenkamp electrothermal apparatus (range 0–400 °C) in
glass capillary tubes and are uncorrected. IR spectra were
recorded on a Perkin-Elmer FT/IR 1600 spectrometer.
1H-NMR spectra were recorded on a Varian EM 360
spectrometer. Chemical ionisation (C.I., methane gas)
mass spectra were recorded on either a Finnigan GCQy
or a Finnigan 4000 series GC/MS mass spectrometer. All
spectra were consistent with the assigned structures.
Optical rotation values were obtained using a Bellingham
and Stanley Ltd. Model D polarimeter. Analyses indi-
cated by the symbols of the elements or functions were
within ± 0.4% of theoretical values.
(2S)-(-)-RO363 11 is unusual in that, in spite of its
phenoxypropanolamine structure, it possesses high intrin-
sic efficacy with approximately 66-fold greater affinity
for transfected human â1-adrenoceptors than (-)-
isoprenaline and approximately 2-fold greater potency at
stimulating chronotropic responses in rat atria (â1-
mediated) than (-)-isoprenaline. Interestingly the phenoxy-
propanolamine agonists ((±)-RO363 7 and its (2S)-(-)-
enantiomer 11), in contrast to (-)-isoprenaline, (-)-
noradrenaline and formoterol, display a functional
potency which is not significantly different from their
binding affinity. Although it is difficult to compare
binding affinity in one species with functional results in
another, the results raise the possibility that (±)-RO363 7
and (2S)-(-)-RO363 11 do not exhibit evidence for a
receptor reserve, which is consistent with the suggestion
that they are examples of partial agonists with intrinsic
efficacy at â1-adrenoceptors sufficient to produce maxi-
mal effects similar to isoprenaline [12]. The action of
(±)-RO363 7 and (2S)-(-)-RO363 11 at rat â2-
adrenoceptors, are more in keeping with classical partial
agonists with a level of intrinsic activity about 3-fold less
than (-)-isoprenaline and although in guinea-pig trachea
the efficacy of (2S)-(-)-RO363 11 was higher, it still could
only stimulate 70% of the maximal â2-mediated effect
seen with (-)-isoprenaline. The relative lack of efficacy at
rat â2-adrenoceptors increases the apparent selectivity in
6.1.2. Materials
3,4-Dibenzyloxybenzaldehyde, 3,4-dimethoxyphen-
ethylamine (homoveratrylamine), (2S)-(+)-glycidyl-3-
nitrobenzenesulfonate, and dimethylformamide (DMF)
were purchased from Aldrich Chemicals. 10% Palladium
on carbon, epichlorohydrin and sodium hydride were
obtained from Fluka Chemicals. m-Chloroperbenzoic
acid (MCBA) was obtained from Merck. Inorganic re-
agents were supplied by Ajax Chemicals. Analytical
grade solvents were purchased from Rhone Poulenc
Australia. Silica plates (5 × 10 cm, Silica F254) were
purchased from Merck. Silica for column chromatogra-
phy (100 Å, 50 µm) was supplied by Amicon Inc., MA
01923, USA.

545
6.1.3. 3,4-dibenzyloxyphenol 3
6.1.5.
(±)-1-(3≠,4≠-Dibenzyloxyphenoxy)-3-(3″,4″-
dimethoxyphenyl)ethylamino-2-propanol hydrochloride 6
(±)-1-(3≠,4≠-Dibenzyloxyphenoxy)-2,3-epoxypropane
4 (2.90 g, 8 mmol) and 3,4-dimethoxyphenylethylamine
5 (1.45 g, 8 mmol) in EtOH (20 mL) were heated to
reflux for 6 h while stirring. The reaction mixture was
evaporated to dryness and the residue was purified by
MCBA (5.06 g, 29 mmol) was added in small portions,
over 30 min, to
a
stirred solution of 3,4-
dibenzyloxybenzaldehyde (6.36 g, 20 mmol) in anhy-
drous CH2Cl2 (100 mL) at 0 °C. The resulting mixture
was stirred overnight, by which time no aldehyde proton
could be detected by 1H-NMR. The solids were dissolved
by the addition of ether and the organic layer was washed
three times with 10% NaHCO3 solution, once with brine
and then dried over anhydrous Na2SO4. Evaporation of
the solvent furnished the crude formate ester 2 which was
hydrolysed without further purification. Yield (crude) =
column
chromatography
(silica,
eluent:
CH2Cl2/MeOH/NH4OH 28%, 90:9:1). The fractions con-
taining the desired product were combined, freed from
solvents and the residue was dissolved in EtOH and
treated with excess ethereal HCl. The precipitated hydro-
chloride 6 was recrystallized from EtOH/ether. Yield =
3.56 g, 82%; m.p. = 184.5–186.5 °C; MS m/e 546 (M +
1
5.81 g, 87%; MS m/e 335 (M + 1); H-NMR (CDCl3) δ
5.28 (4H, s), 6.35–7.18 (3H, m), 7.54 (10H, s), 8.26 (1H,
s).
1
1); H NMR (CDCl3) δ 1.96–2.16 (4H, m), 2.81–2.94
(2H, m), 3.88–4.02 (2H, m), 3.92 (3H, s), 3.96 (3H, s),
4.73–4.86 (1H, m), 5.19 (4H, s), 6.52–7.06 (6H, m), 7.48
(10H, s).
The formate 2 was dissolved in the minimum amount
of MeOH and a slight molar excess of KOH was then
added while stirring under nitrogen. Half an hour later,
water was added, followed by dilute aqueous HCl to
bring the pH of the mixture to 6 and the solution was
extracted with ether. The ethereal solution was washed
with 10% aqueous NaHCO3 solution, then washed with
water and dried over anhydrous MgSO4. The phenol was
recovered after evaporation of the solvent and small
traces of 3,4-dibenzyloxybenzoic acid were removed by
column chromatography (silica, eluent CH2Cl2). The
protected phenol 3 was then recrystallized from CCl4.
Yield (overall) = 4.84 g, 79%; m.p. = 113.6–114.4 °C
(Lit. m.p. = 110.5–111.5 °C [16], Lit. m.p. = 101 °C [17],
Lit. m.p. = 106–108 °C [18]); MS m/e 307 (M + 1);
1H-NMR (CDCl3) δ 5.08 (1H, s), 5.23 (4H, s), 6.31–7.07
(3H, m), 7.49 (10H, s).
6.1.6.
(±)-1-(3≠,4≠-Dihydroxyphenoxy)-3-(3″,4″-
dimethoxyphenyl)ethylamino-2-propanol hydrochloride 7
(±)-1-(3≠,4≠-Dibenzyloxyphenoxy)-3-(3″,4″-dimethoxy-
phenyl)ethylamino-2-propanol hydrochloride
6 (3.0 g,
5 mmol) in EtOH (50 mL) was hydrogenated at ambient
temperature and pressure over Pd/C 10% (0.1 g) until
hydrogen absorption ceased. The reaction mixture was
filtered in a dark, dry box, under nitrogen and the solvent
was evaporated under vacuum. The residue (7) was
recrystallized from EtOH/ether, in the dark, under nitro-
gen and filtered under the same conditions. Yield =
1.75 g, 88%; m.p. = 69.5–71.5 °C; MS m/e 364 (M + 1);
1H NMR (CDCl3) δ 1.92–2.14 (4H, m), 2.80–2.92 (2H,
m), 3.86–4.00 (2H, m), 3.92 (3H, s), 3.96 (3H, s),
4.72–4.85 (1H, m), 6.21–7.12 (6H, m). Anal.
[C19H26ClNO6][H2O]0.5, ie. elemental analysis is consis-
tent with the presence of 0.5 mol of water of crystalliza-
tion, % C calculated 55.81%, found 55.28%; % H calcu-
lated 6.61%, found 7.00%; % Cl calculated 8.69%, found
9.09% and % N calculated 3.43%, found 3.56% (Note:
the oxalate salt of 8 was free of the water of crystalliza-
tion [6]).
6.1.4.
propane 4
(±)-1-(3≠4≠-Dibenzyloxyphenoxy)-2,3-epoxy-
3,4-Dibenzyloxyphenol 3 (3.06 g, 10 mmol) in EtOH
(30 mL) was treated with aqueous KOH (0.62 g,
11 mmol, in 5 mL of water) and epichlorohydrin (2.78 g,
30 mmol) and stirred overnight. The reaction mixture was
evaporated to dryness under reduced pressure and ex-
tracted with ethyl acetate. The extract was washed with
water, then with 5% aqueous NaOH, again with water and
then dried over anhydrous Na2SO4. The filtered solution
was evaporated under reduced pressure and used without
further purification. The epoxide 4 was homogenous by
TLC (silica, eluent CH2Cl2). Yield (crude) = 3.12 g, 86%;
MS m/e 363 (M + 1); 1H NMR (CDCl3) δ 2.68–3.02 (2H,
q), 3.20–3.47 (1H, m), 3.78–4.42 (2H, m), 5.03 (2H, s),
5.10 (2H, s), 6.28–7.02 (3H, m), 7.37 (10H, s).
6.1.7. (2S)-(+)-1-(3≠4≠-dibenzyloxyphenoxy)-2,3-epoxy-
propane 9
Sodium hydride, 60% suspension in mineral oil
(0.52 g, 13 mmol), was washed with dry ether to remove
the oil and suspended in anhydrous DMF (10 mL). While
stirring, 3,4-dibenzyloxyphenol 3 (3.06 g, 10 mmol) in
anhydrous DMF (7 mL) was added dropwise under
nitrogen. Stirring continued for 30 min. The resulting
mixture was cooled to 5 °C with the aid of an icebath and
treated with
a
solution of (2S)-(+)-glycidyl-3-

546
nitrobenzenesulfonate 8 (2.59 g, 10 mmol) in DMF
(7 mL). The mixture was stirred for 3 h under nitrogen
whereupon it was cautiously treated with distilled water
(2 mL). The mixture was diluted with twice its volume of
water and extracted twice with 25 mL of ethyl acetate.
The combined organic extracts were washed with 5%
aqueous NaOH solution, followed by brine, dried over
anhydrous Na2SO4 and evaporated. The residue was
dried overnight under reduced pressure. The epoxide 9
was homogenous by TLC (silica, eluent CH2Cl2). Yield
4.72–4.85 (1H, m), 6.21–7.12 (6H, m). [α]20 = –11.6°
(2.88% solution, MeOH). Anal. C19H26ClNO6D(C, H, N,
Cl).
6.1.10 Optical purity of (2S)-(-)-RO363
The immediate precursor of (±)-RO363, the di-O-
benzyl derivative, was completely resolved into the R-
and S-isomers on a column of Chiracel OD (4 × 250 mm,
Daicel Chemical Industries Ltd) using a mobile phase of
hexane/isopropanol/diethylamine (60:40:0.1) at a flow
rate of 1 mL/min. Detection was by UV absorbance at
283 nm. The R-isomer eluted at 17 min and the S-isomer
at 25 min, each isomer producing a similar peak area. The
(2S)-(-)-RO363 derivative showed a major peak of the
S-isomer with a trace of the R-isomer from which we
calculate the optical purity of > 99%. (±)-RO363 7 was
partially resolved on the same column with a mobile
1
(crude) = 2.99 g, 83%; MS m/e 363 (M + 1); H-NMR
(CDCl3) δ 2.68–3.02 (2H, q), 3.20–3.47 (1H, m),
3.78–4.42 (2H, m), 5.03 (2H, s), 5.10 (2H, s), 6.28–7.02
(3H, m), 7.37 (10H, s).
6.1.8.
(2S)-(-)-1-(3≠,4≠-dibenzyloxyphenoxy)-3-(3″,4″-
dimethoxyphenyl)ethylamino-2-propanol hydrochloride
phase
of
hexane/isopropanol/ethanol/diethylamine
10
(30:45:25:0.1), the two isomers eluting at 8.9 and
10.2 min. (2S)-(-)-RO363 11 produced a single peak at
10 min, consistent with a high optical purity. Because of
the incomplete resolution of the two isomers, an exact
figure for the optical purity could not be estimated,
however the chromatograms suggest a similar purity to
that of the precursor.
(2S)-(+)-1-(3≠,4≠-dibenzyloxyphenoxy)-2,3-epoxypro-
pane
9
(2.90 g, 8 mmol) and 3,4-dimethoxy-
phenylethylamine 5 (1.45 g, 8 mmol) in EtOH (20 mL)
were heated to reflux for 6 h while stirring. The reaction
mixture was evaporated to dryness and the residue was
purified by column chromatography (silica, eluent:
CH2Cl2/MeOH/NH4OH 28%, 90:9:1). The fractions con-
taining the desired product were combined, freed from
solvents and the residue was dissolved in EtOH and
treated with excess ethereal HCl. The precipitated hydro-
chloride 10 was recrystallized from EtOH/ether. Yield =
3.40 g, 78%; m.p. = 185.5–188.0 °C; MS m/e 546 (M +
6.2. Pharmacology
6.2.1. Drugs and chemicals
(-)-Isoprenaline, (-)-noradrenaline and (±)-propranolol
were purchased from Sigma Chemical Co. (St. Louis,
MO, USA). (-)-[125I]-ICYP was purchased from Amer-
sham (Buckinghamshire, UK) and (±)-BRL37344.HCl
was from Tocris Cookson Ltd. (Bristol, UK). (±)-
Formoterol was a gift from Professor J. Angus of the
Department of Pharmacology of the University of Mel-
bourne. (±)-Salbutamol and (±)-bupranolol were synthe-
sized within our department by Dr D. Iakovidis. All other
chemicals were of reagent grade from BDH Chemicals
(Kilsyth, Australia).
1
1); H-NMR (CDCl3) δ 1.96–2.16 (4H, m), 2.81–2.94
(2H, m), 3.88–4.02 (2H, m), 3.92 (3H, s), 3.96 (3H, s),
4.73–4.86 (1H, m), 5.19 (4H, s), 6.52–7.06 (6H, m), 7.48
(10H, s). [α]20D = –7.2° (2.05% solution, DMSO:CH3CN
1:1).
6.1.9.
(2S)-(-)-1-(3≠,4≠-dihydroxyphenoxy)-3-(3″,4″-
dimethoxyphenyl)ethylamino-2-propanol hydrochloride
11
(2S)-(-)-1-(3≠,4≠-dibenzyloxyphenoxy)-3-(3″,4″-dime-
thoxyphenyl)ethylamino-2-propanol hydrochloride 10
(3.0 g, 5 mmol) in EtOH (50 mL) was hydrogenated at
ambient temperature and pressure over Pd/C 10% (0.1 g)
until hydrogen absorption ceased. The reaction mixture
was filtered in a dark, dry box, under nitrogen and the
solvent was evaporated under vacuum. The residue (11)
was recrystallized from EtOH/ether, in the dark, under
nitrogen and filtered under the same conditions. Yield =
1.86 g, 93%; m.p. = 71.0–72.5 °C; MS m/e 364 (M + 1);
1H-NMR (CDCl3) δ 1.92–2.14 (4H, m), 2.80–2.92 (2H,
m), 3.86–4.00 (2H, m), 3.92 (3H, s), 3.96 (3H, s),
6.2.2. Isolated tissue preparations
All studies were performed in agreement with and
according to the Prevention of Cruelty to Animals Act
(1986), the NH & MRC/CSIRO/AAC Australian Code of
Practice for the Care and Use of Animals for Scientific
Purposes (1990) and with the approval of the Animal
Welfare Committee at the Austin and Repatriation Medi-
cal Centre. Male and female Sprague-Dawley rats,
200–300 g, and guinea-pigs (400–450 g) were used and
killed by a blow to the neck followed by cervical
dislocation and/or decapitation.

547
Studies were carried out on rat isolated atria and
tracheal rings as we described previously [19]. All tissues
were allowed to equilibrate for 45 min with Krebs Ringer
physiological salt solution; the composition of which in
mmol L–1 was NaCl, 120; KCl, 5.6; MgSO4, 1.2; CaCl2,
2.5; KH2PO4, 1.4; NaHCO3, 25; glucose 11.2 and EGTA,
0.0025. Cumulative concentration-response curves were
obtained for agonists in each preparation.
the dark. Saturation studies were performed with 0.5–500
pM [125I]-ICYP for the â1- and â2-adrenoceptor cell lines
and 1–3 000 pM for the â3-adrenoceptor cell line. Non-
specific binding was determined in the presence of 2 µM
(±)-propranolol for the â1- and â2-adrenoceptor cell lines
and 100 µM (±)-bupranolol for the â3-adrenoceptor cells
as described by Blin et al. and Gros et al. [20, 21].
Incubations were terminated by rapid filtration onto GF/C
Whatman glass fibre filters presoaked with 0.01% poly-
ethyleneimine and washing with 4 × 5 mL ice-cold
phosphate buffer (pH 7.4). Filter-bound radioactivity was
counted for 1 min using an LKB Multigamma counter.
6.2.2.1. Rat isolated spontaneously beating atria
Rat hearts were removed from adult animals and
placed in Krebs Ringer salt solution (pH 7.4) aerated with
5% CO2 in O2. The atria were dissected free of the
ventricles and overlying tissue and placed in a 20 mL
bath maintained at 37 °C and connected to an isotonic
transducer. A tension of 1 g was applied and chronotropic
activity was amplified and recorded on a Grass Poly-
graph.
6.2.4. Data analysis
6.2.4.1. Functional studies
All concentration-response curves were expressed as a
percentage of the maximum response by the agonist and
plotted against the negative log (–log) molar concentra-
tion of agonist ([agonist]). All data were computer-
fitted [22] using the sigmoidal fit function of the graphi-
cal package Origin (Version 3.01; Micro Cal Origin,
Micro Cal Software Inc., USA). The –log [agonist esti-
mated to give 50% of the maximal response (ie. EC50)]
gave the pD2 value for the agonist [23]. Values given
represent mean ± s.e.m. of 3–5 individual experiments.
6.2.2.2. Rat isolated tracheal chains
Tracheas were excised from adult rats, dissected free of
overlying tissue and cut transversely into segments about
2 mm wide. Five segments were mounted in a 20 mL bath
maintained at 37 °C at a tension of 1 g. Agonist-induced
relaxation of the segments was recorded by an isotonic
transducer connected to a Grass Polygraph after tone had
been established by administration of 1 µM carbachol
(45 min prior to concentration-response curves).
6.2.4.2. Binding studies
6.2.2.3. Isolated guinea-pig preparations
Studies were conducted as outlined above, however,
tracheal chain tone was established by administration of
300 nM of carbachol (45 min prior to concentration-
response curves).
Binding data were analysed using the iterative curve
fitting programs EBDA Version 4.0 which incorporates
LIGAND Version 4.0 [24, 25]. Dissociation constant (Kd)
(saturation studies), inhibition constant (Ki) (drug inhibi-
tion studies) and binding density (Bmax) values, are shown
as mean ± s.e.m. of individual analyses of binding
isotherms using LIGAND. Pseudo Hill coefficients (nH)
were obtained from analysis of binding data using the
sigmoidal fit of the EBDA program. Values represent
mean ± s.e.m. of 3–5 individual experiments.
6.2.3. Radioligand binding studies
Three CHO cell lines, each transfected with an indi-
vidual human â-adrenoceptor subtype were kindly pro-
vided by the Institut Cochin de Genetique Moleculaire,
Paris, France. Binding studies with CHO cell membranes
were conducted as described by Blin et al. [20]. Precon-
fluent cells were washed twice with ice-cold phosphate
buffer solution and harvested with 25 mM Tris-HCl, pH
7.5, 1 mM EDTA buffer. Harvested cells were centrifuged
at 10 000 g for 15 min and pellets were resuspended in
Tris-HCl-EDTA homogenization buffer and stored at
–80 °C until needed.
Cells were thawed as required and suspended in
Hank’s balanced salt solution supplemented with 1 mM
ascorbic acid, pH 7.4 at 200–500 µg protein mL–1 for all
studies. Aliquots of cells were incubated with 100 pM
(-)-[125I]-ICYP in the absence or presence of competitor,
in a 200 µl final volume of buffer, for 45 min at 37 °C in
Acknowledgements
The authors would like to thank Professor A. Donny
Strosberg (Institut Cochin de Genetique Moleculaire,
Paris, France) for providing us with the three CHO cell
lines transfected with the human â1-, â2- and â3-
adrenoceptors and Ms Leanne Styan for her technical
assistance. This work was supported by a grant from the
National Health and Medical Research Council of Aus-
tralia.

548
[13] Hieble J.P., in: Ruffolo R.R. Jr., â-Adrenoceptors: Molecular Biol-
ogy, Biochemistry and Pharmacology. Prog. Basic Clin. Pharmacol.,
vol. 7, Karger, Basel, 1991, pp. 105–172.
References
[1] Raper C., McPherson G.A., Iakovidis D., Eur. J. Pharmacol. 52
(1978) 241–242.
[14] Snow H.M., Br. J. Clin. Pharmacol. 28 (1989) 3S–13S.
[15] Maccarrone C., Malta E., Raper C., J. Cardiovasc. Pharmacol. 6
(1984) 132–141.
[2] Iakovidis D., Malta E., McPherson G.A., Raper C., Br. J. Pharmacol.
68 (1980) 677–685.
[16] Wenke M., Trcka V., Blaha L., Hodrova J., Weichet M., Czech., 131
(1969) 157 in: Chem. Abstr. 72 (1970) 111007.
[3] McPherson G.A., Malta E., Molenaar P., Raper C., Br. J. Pharmacol.
82 (1984) 897–904.
[4] Molenaar P., Sarsero D., Arch J.R.S., Kelly J., Henson S.M.,
Kaumann A.J., Br. J. Pharmacol. 120 (1997) 165–176.
[17] Ablad B., Brogard M., Corrodi H., Acta Pharm. Suesica 7 (1970)
551–558.
[5] Molenaar P., Sarsero D., Kaumann A.J., Clin. Exp. Pharmacol.
Physiol. 24 (1997) 647–656.
[18] Casagrande C., Ferrini R., Miragoli G., Ferrari G., Boll. Chim.
Farm. 112 (1973) 445–454.
[6] Iakovidis D., Structure Activity Relationships in Medicinal Chem-
istry, Ph D Thesis Deakin University, Victoria, Australia, 1984.
[19] Tung L., Jackman G., Campbell B., Louis S., Iakovidis D., Louis
W.J., J. Cardiovasc. Pharmacol. 21 (1993) 484–488.
[7] Godfrey I.M., Sargent M.V., Elix J.A., J. Chem. Soc. Perkin Trans.
I (1974) 1353–1354.
[20] Blin N., Camoin L., Maigret B., Strosberg A.D., Mol. Pharmacol. 44
(1993) 1094–1104.
[8] Klunder J.M., Onami T., Sharpless K.B., J. Org. Chem. 54 (1989)
1295–1304.
[21] Gros J., Manning B.S., Pietri-Rouxel F., Guillaume J.L., Drumare
M.F., Strosberg A.D., Eur. J. Biochem. 251 (1998) 590–596.
[9] Bryan L.J., Cole J.J., O’Donnell S.R., Wanstall, J.C., J. Pharmacol.
Exp. Ther. 216 (2) (1981) 395–400.
[22] Zaborowsky B.R., McMahan W.C., Griffin W.A., Norris F.H.,
Ruffolo R.R. Jr., J. Pharmacol. Methods 4 (1980) 165–178.
[10] Dyke A., Anugs J.A., J. Auton. Pharmacol. 8 (3) (1988) 219–228.
[23] Van Rossum J.M., Hurkmans J.A.T.H.M., Wolters C.J.J., Arch. Int.
Pharmacodyn. Therap. 143 (1963) 299–330.
[11] Goodman A.G., Goodman L.S., Rall T.W., Murad F. (Eds.), The
Pharmacological Basis of Therapeutics, Macmillan Publishing Com-
pany, NY, USA, 1985.
[24] McPherson G.A., Comput. Programs Biomed. 17 (1983) 107–113.
[25] Munson P.J., Rodbard D., Anal. Biochem. 109 (1980) 220–239.
[12] Malta E., McPherson A.G., Raper C., TIPS Reviews 400–403
(1985).