Synthesis and β-adrenergic properties of (Z)-N-[3-(alkylamino)-2-hydroxypropylidene](aryl-methyloxy)amines: Effects of the configuration around the methyloxyiminomethyl (MOIM) double bond on the biopharmacological properties of MOIM-type β-blocking agents

Article abstract of DOI:10.1016/S0968-0896(98)00172-2

The N-isopropyl- (3a-g) and N-tert-butyl-substituted (4a-g) (Z)-N-(3-(amino)-2-hydroxypropylidene)(arylmethyloxy)amines were synthesized in order to compare their β₁- and β₂-adrenergic properties with those of their previously studie

Full text of DOI:10.1016/S0968-0896(98)00172-2

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 2151±2160
Synthesis and ꢀ-Adrenergic Properties of (Z)-N-[3-
(Alkylamino)-2-hydroxypropylidene](aryl-methyloxy)amines:
E€ects of the Con®guration Around the Methyloxyiminomethyl
(MOIM) Double Bond on the Biopharmacological Properties of
MOIM-type ꢀ-Blocking Agents
Aldo Balsamo,a* Maria C. Breschi,b Grazia Chiellini,b Annalina Lapucci,a
Nicola Lazzeri,b Marco Macchia,a Adriano Martinelli,a Eugenio Micali,a
Susanna Nencettia and Armando Rosselloa
a
Á
Dipartimento di Scienze Farmaceutiche, Universita di Pisa, via Bonanno 6, 56126, Pisa, Italy
b
Á
Dipartimento di Psichiatria Neurobiologia Farmacologia e Biotecnologie, Universita di Pisa, via Bonanno 6, 56126, Pisa, Italy
Received 21 May 1998; accepted 15 July 1998
AbstractÐThe N-isopropyl- (3a±g) and N-tert-butyl-substituted (4a±g) (Z)-N-(3-(amino)-2-hydroxypropylidene)-
(arylmethyloxy)amines were synthesized in order to compare their ꢀ1- and ꢀ2-adrenergic properties with those of their
previously studied corresponding analogues with the E con®guration (1a±g and 2a±g). Compounds 3 and 4 were tested
for their anity for ꢀ1- and ꢀ2-adrenoceptors by radioligand binding experiments, and the compounds with the high-
est anity were also assayed for their activity towards the same types of ꢀ-adrenoceptors by functional tests on
isolated preparations. The Z-methyloxyiminomethyl (Z-MOIM) compounds 3 and 4 proved to possess, on the whole,
anity (Ki) and activity (pIC50) indices similar to those of the E isomers 1 and 2, thus indicating that for the MOIM-
type ꢀ-adrenergic antagonists 1±4, the type of con®guration around the MOIM double bond does not have any
appreciable e€ect either on the anity or on the activity towards ꢀ-adrenoceptors. These results are rationalized on the
basis of the steric and electronic analogies existing between the MOIM groups of 1±4 in the two types of con®gurations
(E and Z). # 1998 Elsevier Science Ltd. All rights reserved.
Introduction
MAOMM) of type B compounds can act as a bioisoster
of the Ar of type A b-blocking agents.1 This result was
explained by a comparison of the chemical reactivity of
the two types of compounds (A and B). In a subsequent
study, it was found that the replacement of the
MAOMM of type B compounds with the methylox-
yiminomethyl moiety (CH2ON=C, MOIMM) with the
E con®guration leads to compounds (C) possessing b-
adrenergic properties similar to those of compounds
B.3,5 In the design of compounds C, the type of con®g-
uration (E) of the MOIMM was chosen because in
this con®guration, the atomic sequence CH2ON=C
appeared to present greater steric and electronic analo-
gies with an Ar than in the Z con®guration.3 Theoretical
As part of an on-going study1±8 on the stereoelectronic
requirements for the interaction of adrenergic drugs
with b-receptors, we examined the possibility of e€ec-
tively replacing the aryl (Ar) of type A b-blocking drugs
with non-aromatic substituents which were believed to
be capable of simulating an Ar. It was thus found
that the methyleneaminoxymethyl moiety (C=NOCH2,
Key words: adrenergic drug; b-blocking agent; [(methyloxy)-
imino]methyl moiety; (Z)-N-[3-(amino)-2-hydroxypropylidene]-
(arylmethyloxy)amine.
*Corresponding author. Tel.: 39 50 500 209; fax: 39 50 40517.
0968-0896/98/$ - see front matter # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(98)00172-2

2152
A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
studies carried out on model compounds of types B and
C showed clear spatial correspondences and electronic
analogies between the E-MOIMM present in com-
pounds C and the MAOMM of compounds B, and
therefore the Ar of drugs of type A.3
Chemistry
The (Z)-N-[3-(isopropylamino)-2-hydroxypropylidene]-
(3a±g) and (Z)-N-[3-(tert-butylamino)-2-hydroxypropyl-
idene](arylmethyloxy)amines (4a±g) were prepared
following the synthetic procedure shown in Scheme 1.
Condensation of the O-arylmethylhydroxylamines (5a±g),9
as hydrochlorides, with acrolein yielded mixtures of the
corresponding Z (6a±g) and E (7a±g) unsaturated oxime
ethers in a ratio of about 3:7,5 from which the Z com-
pounds (6a±g) were isolated by column chromatography.
Epoxidation10 of 6a±g with potassium peroxymonosulfate
(oxone) in heterogeneous phase at pH 7.5 in the pre-
sence of 18-crown-6, a€orded almost exclusively the
corresponding Z epoxides (8a±g) which were puri®ed by
column chromatography. Subsequent aminolysis of 8a±g
with i-PrNH2 or t-BuNH2 yielded the crude Z amino-
alcohols which, after puri®cation by column chromato-
graphy, were transformed into the corresponding neutral
oxalate salts (3a±g.1/2H2C2O4, 4a±g.1/2H2C2O4).
The structure and con®guration of the intermediate (6a±
g and 8a±g) and ®nal (3a±g and 4a±g) compounds were
1
assigned by comparison of their H NMR spectral data
(for 3 and 4 see Table 2) with those obtained for the
same compounds in the crude E/Z mixtures,5 with the
only exception of the unsaturated oxime ether with the
Z con®guration 6g, which was also previously separated
from the E isomer 7g.5
In order to obtain information about the con®gura-
tional stability of the Z-MOIM derivatives, one of them
(4g‰H2C2O4) was placed in the experimental condi-
tions of the biopharmacological tests and was found to
be con®gurationally stable (1H NMR).
Radioligand binding assays
Furthermore, the results obtained with the previously
studied type C derivatives (1 and 2) do not exclude the
possibility that, at least in the ®eld of b-adrenergic
antagonists, the MOIM group may be bioequivalent to
the MAOMM, even if its atomic sequence is arranged in
a di€erent way from the one shown by the MOIMM
with the E con®guration. On the basis of this con-
sideration, and as an extension of our earlier research,
we decided to study the compounds of type D which
di€er from those of type C previously studied (1 and 2)
in their con®guration around the imino double bond of
the MOIMM, which is of type Z instead of E. This
paper describes the synthesis and complete character-
ization of the (Z)-N-[3-(amino)-2-hydroxypropylidene]
(arylmethyloxy)amines 3a±g and 4a±g (see Table 1),
together with their biopharmacological b-adrenergic
properties, evaluated by radioligand binding experi-
ments and, for the compounds with the highest anity,
by functional tests on isolated preparations.
The b-adrenergic anity of the Z-MOIM derivatives
(3, 4) and of the reference drugs dichloroisoproterenol
and propranolol (Table 3) was checked by binding tests
on rat brain and bovine lung membrane preparations
for b1- and b2-adrenoceptors, respectively. 1-[[2-(3-car-
bamoyl-4-hydroxyphenoxy)ethyl]amino]-3-[4-[1-methyl-
4-(tri¯uoromethyl)-2-imidazolyl]-phenoxy]-2-propanol
([3H]-CGP 26505)11 was used as a speci®c tritiated
ligand for rat brain b1-adrenoceptors. [3H]-Dihy-
droalprenolol ([3H]-DHA)12 was utilized in the presence
of 50 nM CGP 26505 to label bovine lung b2-adreno-
ceptors. The reason for the presence of CGP 26505 in
this latter type of test was that of displacing [3H]-DHA
binding from the bovine lung b1-adrenoceptor sub-
population (17%).13
Rat brain ꢀ1-adrenoceptors. The Z-MOIM derivatives 3
and 4 proved to possess a certain binding anity

A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
2153
Table 1. Analytical and chemical data of (Z)-N-[3-(isopropylamino)-(3a±g) and (Z)-N-[3-(tert-butylamino)-2-hydroxypropylidene]-
(arylmethyloxy)amines (4a±g) as oxalates
Compound
R
R1
mp (ꢀC)
Yielda (%)
Formulab
3a‰H2C2O4
4a‰H2C2O4
3b‰H2C2O4
4b‰H2C2O4
3c‰H2C2O4
4c‰H2C2O4
3d‰H2C2O4
4d‰H2C2O4
3e‰H2C2O4
4e‰H2C2O4
3f‰H2C2O4
4f‰H2C2O4
3g‰H2C2O4
4g‰H2C2O4
H
i-Pr
t-Bu
i-Pr
t-Bu
i-Pr
t-Bu
i-Pr
t-Bu
i-Pr
t-Bu
i-Pr
t-Bu
i-Pr
t-Bu
144±146
168±170
113±118
162±164
148±150
156±158
140±143
170±172
121±124
170±172
145±148
170±173
146±148
178±180
43
48
48
46
48
46
47
36
39
43
37
45
40
46
C14H21N2O4
C15H23N2O4
C15H23N2O5
C16H25N2O5
C15H23N2O5
C16H25N2O5
C15H23N2O5
C16H25N2O5
C14H20N2O4Cl
C15H22N2O4Cl
C14H20N2O4Cl
C15H22N2O4Cl
C14H20N2O4Cl
C15H22N2O4Cl
H
o-MeO
o-MeO
m-MeO
m-MeO
p-MeO
p-MeO
o-Cl
o-Cl
m-Cl
m-Cl
p-Cl
p-Cl
aFor epoxide aminolysis and sali®cation; no e€orts were made to optimize yields.
bAll compounds were analyzed for C, H, and N.
Scheme 1. A: R=H; b: R=o-MeO; c: R=m-MeO; d: R=p-MeO; e: R=o-Cl; f: R=m-Cl; g: R=p-Cl.

2154
A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
Table 2. 1H NMR data of (Z)-N-[3-(isopropylamino)-(3a±g) and (Z)-N-[3-(tert-butylamino)-2-hydroxypropylidene](arylmethylox-
y)amines (4a±g) as free bases
Compd
R
R1
H1
H2
H3a
H3b
CH2O
3a
4a
3b
4b
3c
4c
3d
4d
3e
4e
3f
H
H
i-Pr 6.80 d (J=4.9)
t-Bu 6.81 d (J=4.9)
4.72 ddd (J=8.0, 4.9, 3.9)
4.67 ddd (J=8.0, 4.9, 4.1)
4.71 ddd (J=8.1, 4.8, 4.1)
4.66 ddd (J=8.0, 4.7, 4.1)
4.73 ddd (J=8.1, 5.0, 4.1)
4.67 ddd (J=8.1, 5.1, 4.1)
4.69 ddd (J=8.2, 4.9, 4.2)
4.63 ddd (J=8.0, 5.0, 4.0)
4.75 ddd (J=8.1, 4.9, 3.9)
4.69 ddd (J=8.2, 4.9, 4.0)
4.73 ddd (J=8.2, 5.0, 3.9)
4.67 ddd (J=8.2, 4.9, 4.1)
4.74 ddd (J=8.1, 4.9, 3.9)
4.65 ddd (J=8.0, 5.1, 4.0)
2.96 dd (J=12.2, 3.9)
2.93 dd (J=12.0, 4.1)
2.96 dd (J=12.2, 4.1)
2.93 dd (J=12.0, 4.1)
2.97 dd (J=12.2, 4.1)
2.94 dd (J=12.0, 4.1)
2.93 dd (J=12.2, 4.2)
2.90 dd (J=12.1, 4.0)
3.00 dd (J=12.2, 3.9)
2.98 dd (J=12.1, 4.0)
2.97 dd (J=12.2, 3.9)
2.94 dd (J=12.1, 4.1)
2.98 dd (J=12.2, 3.9)
2.92 dd (J=12.1, 4.0)
2.63 dd (J=12.2, 8.0) 5.1 s
2.57 dd (J=12.0, 8.0) 5.09 s
2.65 dd (J=12.2, 8.1) 5.16 s
2.59 dd (J=12.0, 8.0) 5.15 s
2.63 dd (J=12.2, 8.1) 5.07 s
2.57 dd (J=12.0, 8.1) 5.07 s
2.61 dd (J=12.2, 8.2) 5.02 s
2.55 dd (J=12.1, 8.0) 5.02 s
2.65 dd (J=12.2, 8.1) 5.21 s
2.59 dd (J=12.1, 8.2) 5.21 s
2.63 dd (J=12.2, 8.2) 5.06 s
2.57 dd (J=12.1, 8.2) 5.06 s
2.64 dd (J=12.2, 8.1) 5.15 s
2.55 dd (J=12.1, 8.0) 5.05 s
o-MeO i-Pr 6.82 d (J=4.8)
o-MeO t-Bu 6.82 d (J=4.7)
m-MeO i-Pr 6.80 d (J=5.0)
m-MeO t-Bu 6.81 d (J=5.1)
p-MeO i-Pr 6.78 d (J=4.9)
p-MeO t-Bu 6.79 d (J=5.0)
o-Cl i-Pr 6.82 d (J=4.9)
o-Cl t-Bu 6.83 d (J=4.9)
m-Cl i-Pr 6.81 d (J=5.0)
m-Cl t-Bu 6.81 d (J=4.9)
p-Cl i-Pr 6.82 d (J=4.9)
p-Cl t-Bu 6.80 d (J=5.1)
4f
3g
4g
towards b1-adrenoceptors, with Ki values ranging from
the 10 mM of 3b to the 0.2 mM of 4c. Among the N-iso-
propyl derivatives 3, while the m-substituted compounds
(3c,f) exhibited Ki values similar to that of the unsub-
stituted oxime ether 3a, the o-(3b) and p-MeO-sub-
stituted (3d) compounds showed anity indices
substantially higher than that of 3a; the Ki values of the
o-(3e) and p-Cl-substituted (3g) were found to be a bit
lower or higher, respectively, than the Ki value of 3a.
For all the N-isopropyl-substituted compounds, the
ratio between the Ki values obtained with the com-
pounds with the Z con®guration and those obtained
with the corresponding E con®guration isomers5 was
between 2.0 (i.e. the value indicating half the anity
with respect to that of the E isomer) and 0.5 (i.e. the
value indicating twice the anity of the E isomer), with
the only exception of 3g, for which the Z/E ratio was
slightly higher than 2. As regards the N-tert-butyl deri-
vatives 4, the o-Cl-substituted compound (4e) showed a
Ki value slightly higher than that of 4a, while the o-
MeO-compound (4b) and both the p-substituted ones
(4d,g) were among the compounds with the highest a-
nity indices; the m-MeO-(4c) and the m-Cl-substituted
(4f) compounds appeared to possess Ki values slightly
lower than or analogous to that of 4a, respectively.
Among the N-tert-butyl-substituted compounds, while
4b,d,f exhibited Z/E ratios between 2.0 and 0.5, 4c,g
showed values appreciably outside this range and 4a,e
values slightly below 0.5.
Bovine lung ꢀ2-adrenoceptors. All the Z-MOIM com-
pounds (3,4) showed a good degree of anity towards
this type of receptor, with Ki values which, with the only
exceptions of 3b and 3c, were in the nanomolar range.
Within the series of N-isopropyl-substituted compounds
(3a±g), the p-Cl-derivative (3g) appeared to possess the
lowest Ki value, while the o-(3b) and the m-MeO-(3c)
compounds exhibited the highest anity indices; the
phenyl-unsubstituted derivative (3a) and its p-MeO-(3d)
and o-(3e) and m-Cl (3f) analogs showed very similar Ki
values. The Z-MOIM compounds 3a,b,d,g showed Ki
values similar to those obtained for the corresponding
E-MOIM isomers,5 with Z/E ratio values between 2.0
and 0.5, while 3c,e,f exhibited Z/E ratio values slightly
lower than 0.5 (3e) or slightly higher than 2.0 (3c,f). As
for the N-tert-butyl-substituted compounds, (4) the one
unsubstituted on the phenyl ring (4a) and the p-MeO-
(4d) and m-Cl-substituted (4f) analogues showed practi-
cally identical anity indices, while the o-Cl-(4e) and
p-Cl-substituted (4g) compounds exhibited Ki values
slightly lower or slightly higher, respectively, than those
of 4a,d,f; the o-(4b) and the m-MeO (4c) derivatives were

A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
2155
Table 3. Radioligand binding anity of Z-MOIM derivatives 3 and 4
b-Adrenergic binding anity
Rat brain (b1)
Bovine lung (b2)
Compd
R
R1
Ki (nM)a
Z/Eb
Ki (nM)a
Z/Eb
3a‰H2C2O4
3b‰H2C2O4
3c‰H2C2O4
3d‰H2C2O4
3e‰H2C2O4
3f‰H2C2O4
3g‰H2C2O4
4a‰H2C2O4
4b‰H2C2O4
4c‰H2C2O4
4d‰H2C2O4
4e‰H2C2O4
4f‰H2C2O4
4g‰H2C2O4
dichloroisoproterenol
propranolol
H
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
3800 (3400±4200)
11000 (9300±12000)
3000 (2700±3300)
9200 (8200±10300)
2100 (1800±2300)
3100 (2700±3500)
4900 (4300±5600)
690 (630±760)
1.86
0.71
0.64
1.42
0.84
1.44
2.13
0.33
1.91
0.11
0.87
0.36
0.54
7.74
680 (610± 760)
1700 (1500±1800)
1500 (1300±1600)
460 (400±530)
480 (430±530)
940 (820±1100)
310 (270±350)
120 (100±130)
490 (440±540)
490 (430±560)
130 (120±140)
77 (76±78)
0.83
0.70
3.42
1.10
0.42
2.21
1.00
0.32
0.70
1.32
0.97
0.31
0.67
2.14
o-MeO
m-MeO
p-MeO
o-Cl
m-Cl
p-Cl
H
o-MeO
m-MeO
p-MeO
o-Cl
1900 (1700±2200)
220 (200±230)
5800 (5100±6400)
1200 (1100±1300)
760 (700±830)
m-Cl
p-Cl
140 (130±160)
210 (190±230)
150 (140±170)
2.0 (1.7±2.2)
2400 (2200±2600)
58 (49±67)
5.1 (3.9±5.9)
aGeometric means of ®ve separate determinations with con®dence limits in parentheses.
bRatio between the Ki value shown by the Z (3 and 4) and by the corresponding E isomer (1 and 2) (see ref. 5).
found to possess the lowest anity. Among the N-tert-
butyl-substituted compounds, 4b,c,d,f exhibited anity
indices similar to those of the corresponding E-MOIM
isomers,5 with Z/E ratio values between 2.0 and 0.5,
while 4a,e,g showed slightly di€erent Ki values, with Z/
E ratio values slightly lower (4a,e) or higher (4g) than
0.5 or 2.0, respectively.
Guinea pig atria ꢀ1-adrenoceptors. All the Z-MOIM
compounds examined (4a,c,e±g) antagonized the stimu-
lating e€ects of isoprenaline to a lower degree than the
reference drugs. The highest pIC50 values were shown
by the m-MeO-(4c) and o-Cl-(4e) derivatives, while the
m-Cl-(4f) and the p-Cl-substituted (4g) compounds
exhibited a slightly lower pIC50 value. For the Z-MOIM
compounds for which also the corresponding E isomers
were submitted to the same functional test (4a,c,g), the
activity indices were somewhat lower (4a,g) or slightly
higher (4c) than those of the corresponding analogs
(2a,c,g).5 None of the Z-MOIM compounds 4 shown in
Table 4 exhibited any stimulating activity on this b-
adrenoceptor.
Functional tests
The b-adrenergic activity of the Z-MOIM derivatives
that proved to possess the highest anity (Ki<1500 nM)
for b1-adrenoceptors was evaluated by functional tests
on guinea-pig atria and guinea-pig tracheal strips for b1-
and b2-adrenoceptors, respectively. Table 4 shows the
pIC50 values obtained on these receptors for Z-MOIM
compounds 4a,c,e,f, for the Z-MOIM derivative 4g
which is the analogue of the E-MOIM compound pos-
sessing the best anity on both b-adrenoceptors (2g),5
and for dichloroisoproterenol and propranolol.
Guinea pig tracheal strip ꢀ2-adrenoceptors. The Z-
MOIM compounds 4a,c,e±g showed a good b2-blocking
activity, with similar pIC50 indices which in some cases
(4a,e,g) were higher than that of dichloroisoproterenol.
The highest activity index was shown by the o-Cl

2156
A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
Table 4. b-Adrenergic activity of selected Z-MOIM derivatives (4)
b
b-Adrenergic activitya pIC50
Compd
R
Isolated guinea pig atria (b1)
Isolated guinea pig tracheal strips (b2)
4a‰H2C2O4
4c‰H2C2O4
4e‰H2C2O4
4f‰H2C2O4
H
m-MeO
o-Cl
4.39‹0.16c
4.78‹0.11e
4.80‹0.02
4.52‹0.07
4.31‹0.14g
6.82‹0.18
7.42‹0.15
6.51‹0.02d
6.04‹0.18f
6.72‹0.02
6.03‹0.14
6.53‹0.24h
6.09‹0.28
7.60‹0.12
m-Cl
p-Cl
4g‰H2C2O4
dichloroisoproterenol
propranolol
aThe values represent the mean of three to ®ve experiments for each drug‹standard error.
bpIC50 is the negative logarithm of the molar concentration that reduces the response to isoprenaline by 50%.
cFor 2a (ref. 5): pIC50 5.10‹0.17.
dFor 2a (ref. 5): pIC50 6.89‹0.10.
eFor 2c (ref. 5): pIC50 4.60‹0.01.
fFor 2c (ref. 5): pIC50 6.48‹0.26.
gFor 2g (ref. 5): pIC50 5.13‹0.08.
hFor 2g (ref. 5): pIC50 6.77‹0.31.
derivative (4e), while the analogue unsubstituted on the
phenyl ring (4a) and the p-Cl-derivative (4g), and, in
decreasing order, the m-substituted compounds (4c,f)
appeared to be slightly less active. For Z-MOIM com-
pounds 4a,c,g, the activity indices were found to be
slightly lower than those shown in the same type of
test by the corresponding E analogs 2a,c,g.5 The Z-
MOIM compounds of type 4 shown in Table 4
(4a,c,e±g) were devoid of any agonistic activity also on
b2-adrenoceptors.
As regards model compound 9 (the one possessing the E
con®guration), a systematic search, performed by means
of molecular mechanics calculations, showed5 that in its
preferred conformation, its C(a)-C-O-NˆC moiety is
planar, and the torsion angles t1 and t2 have values of
90ꢀ and 180ꢀ, respectively; however, it had also been
found that the benzylic portion possesses a large con-
formational freedom.
Theoretical calculations
In order to gain a better understanding of the experi-
mental results obtained for Z-MOIM derivatives (D) in
comparison with those previously obtained for E-
MOIM derivatives (C), the conformational and reactiv-
ity properties of model compound 10 were studied and
compared with those of the previously investigated5
model compound 9. Model compounds 9 and 10 corre-
spond to E-MOIM (1a and 2a) and Z-MOIM (3a and
4a) drugs respectively, the only di€erence being the lack
of the N-substituent, in accordance with a simpli®cation
already tested and used by us in previous papers.1a,2,3
An analogous conformational analysis was carried out
on model compound 10 (the one possessing the Z con-
®guration). Also for this compound in its preferred
conformation, the torsion angles t1 and t2 have values
of 90ꢀ and 180ꢀ, respectively, and the C(a)-C-O-NˆC

A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
2157
moiety is planar; furthermore, 10 proves to have a large
degree of freedom like 9.
alcoholic oxygen and the iminic nitrogen, whereas the
less extended one is generated by the iminic nitrogen in
9 and by the ethereal oxygen in 10.
The superimposition of model compounds 9 and 10 in
their preferred conformations (Fig. 1) shows that there
is an optimal spatial correspondence of the NˆCH-
CHOH-CH2NH2 portions, and that the remaining
benzyloxy moieties occupy very di€erent spatial regions;
however, the MOIMM of both molecules are almost
coplanar. As regards the phenyl-substituted Z-MOIM
derivatives (3b±g, 4b±g), no in¯uence of the phenyl sub-
stituent on the conformation was observed; this is in
agreement with previous ®ndings5 for the corresponding
phenyl-substituted E-MOIM derivatives (1b±g, 2b±g).
It was also found that the MEP in the region of the
CH2-O-NˆCH-CHOH-CH2-NH2 region of Z-MOIM
derivatives 3b±g, 4b±g is not signi®cantly in¯uenced by
the phenyl substituent, in agreement with previous ®nd-
ings for E-MOIM compounds 1b±g, 2b±g.5
Discussion and Conclusions
An examination of the data reported in Table 3 shows
that the Z-MOIM compounds studied (3, 4) possess
anity indices towards b1-adrenoceptors ranging from
0.2 to 11 mM. A very similar range of Ki values (0.3±
14 mM) was found for the corresponding analogues with
the E con®guration previously studied (1 and 2). This
indicates that, on the whole, the MOIM derivatives of
types 1±4 possess an ability to interact with b1-receptors
which seems to be independent of the con®guration
around the double bond of the MOIM group. This
result is con®rmed by the values of the ratio (Z/E)
between the Ki of the pairs of isomers with opposite
con®gurations, which, in most cases, are between 2.0
and 0.5, i.e. within the range indicating an anity of Z
compounds for b1-receptors equal to half or twice that
of the corresponding E isomers, respectively.
Figure 2 shows the molecular electrostatic potential
(MEP) of model compounds 9 and 10 in their preferred
conformations. A very similar MEP trend is found in
the region of the MOIMM in spite of the di€erent spa-
tial arrangements of the moieties.
For both 9 and 10, the MEP trend is characterized by
three negative regions. One of them is isolated and gen-
erated by the common aminic nitrogen atom in both
compounds; the other two regions are close to each
other: the more extended one is generated in 9 by the
alcoholic and the ethereal oxygens and in 10 by the
As regards the Z-MOIM derivatives submitted to the
functional tests on b1-adrenoceptors (4a, c, e±g), the
trend of their activity indices (pIC50) appeared to be in
line with that of their anity indices (Ki), with the
exception of the compound not substituted on the phe-
nyl (4a), which was among the best in terms of anity
and among the worst in terms of activity. The activity
indices for b1-adrenoceptors obtained for the Z-MOIM
compounds studied are within a range of values (4.31±4.80)
which is not very di€erent from the one (4.35±5.13)
shown by the analogues with the E con®guration pre-
viously submitted to the same type of test.
Figure 1. Compounds 9 (thinner line) and 10 (thicker line) in
their preferred conformations.
Figure 2. Compounds 9 and 10 in their preferred conformations. The MEP contours corresponding to values of 10 (thinner net-
work) and 20 kcal/mol (thicker network) are shown.

2158
A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
As regards b2-adrenoceptors, the results obtained in the
binding tests indicate that the Z-MOIM compounds 3
and 4 possess anity indices ranging from 0.08 to
1.7 mM, that is, a range of Ki values very similar to those
shown by the corresponding isomers with the E con®g-
uration 1 and 2 (0.1±5.6 mM). This indicates that for
these types of aminoalcohols (1±4), the con®guration
around the double bond of the MOIM does not sub-
stantially in¯uence their ability to interact also with b2-
adrenoceptors. Con®rmation of these ®ndings is o€ered
by an examination of the values of the ratio between the
Ki of the couples of con®gurational isomers (Z/E),
which are all between 2 and 0.5, also for this type of
b-adrenoceptor.
linked to the di€erent spatial arrangement of the
benzylic moiety with respect to the remaining portion of
the molecule. This fact could be explained either by
excluding the possibility of a direct interaction of the
aryl with the appropriate receptor sites, or by hypo-
thesizing the presence of two di€erent binding sites for
the aryl groups of the two types (Z or E) of MOIM
compounds.
Experimental
Chemistry
Melting points were determined on a Ko¯er hot-stage
apparatus and are uncorrected. IR spectra for compar-
ison of compounds were taken as paran oil mulls or as
liquid ®lms on a Mattson 1000 Series FTIR Spectro-
meter. 1H NMR spectra of all compounds were
obtained with a Bruker AC-200 instrument in ca. 2%
solution of CDCl3 (for the neutral compounds) or D2O
(for the salts), using Me4Si or Me3Si(CH2)3SO3Na as
the internal standard, respectively. Analytical TLCs
were carried out on 0.25 mm layer silica gel plates con-
taining a ¯uorescent indicator (Macherey±Nagel Alu-
gram1 SilG/UV254 Art. Nr. 81813); spots were detected
under UV light (254 nm). Column chromatographies
were performed using 230±400 mesh silica gel
(Macherey±Nagel Silica Gel 60 Art. Nr. 81538). Glc
analyses were performed on a Carlo Erba model 4200
apparatus with a ¯ame ionization detector, using a
1.6 mÂ2.6 mm column packed with neopentylgly-
colsuccinate 10% on chromosorb W silanised 80/100
mesh. Evaporations were made in vacuo (rotating eva-
porator); MgSO4 was always used as the drying agent.
Elemental analyses were performed in our analytical
laboratory and agreed with the theoretical values to
within ‹0.4%.
The activity indices obtained for the Z-MOIM deriva-
tives 4a, c, e±g in the functional tests carried out on tra-
cheal b2-adrenoceptors appear to be in good agreement
with the anity indices obtained in the binding tests,
apart from the m-chloro-substituted compound (4f)
which exhibits the lowest pIC50 value, even if it pos-
sesses an anity practically identical to that of 4a, one
of the more active Z-MOIM derivatives on this adre-
noceptor. The pIC50 values of 4a, c, e±g are between 6.03
and 6.72, a range of values that is very similar to those
obtained for the analogues with the opposite con®gura-
tion previously evaluated for b2 antagonist activity in
the same kind of test (6.01±6.89).
A comparison of the results obtained for the Z-MOIM
derivatives 3 and 4 with those previously obtained for
their corresponding isomers with the E con®guration (1
and 2) demonstrates that for the MOIM-type ami-
noalcohols 1±4 studied, the b-adrenergic properties, as
far as both the anity and the activity are concerned,
are substantially independent of the con®guration
around the MOIM double bond (E or Z).
A comparison of the conformational and electronic
properties of the model compounds 9 and 10 of the E-
MOIM (1,2) and Z-MOIM (3,4) analogues respectively,
determined by means of molecular calculations, reveals
that, in spite of the di€erences in the steric character-
istics due to the Z or E con®guration, at least at the level
of the spatial position of the arylmethyloxy group
(ArCH2O), the two kinds of compounds present some
important analogies in their molecular reactivity. In
particular, the reactivity pattern of the two con®gura-
tional isomers is very similar not only in the ethanola-
minic portion, but also at the level of the O-NˆC
atomic sequence. This fact suggests that for the interaction
of the MOIM-type compounds 1±4 with b-adrenoceptors,
the fundamental role may be played by the atomic
sequence O-NˆCH(OH)CH2NH. Moreover, the E- and
Z-MOIM analogues possess analogous adrenergic prop-
erties, even if they present di€erent steric characteristic
Synthesis of (Z)-N-Propenylidene(arylmethyloxy)amines
(6a±g). A heterogeneous mixture of the appropriate
O-(arylmethyl)hydroxylamine hydrochloride9 5a±f
(0.094 mol) and acrolein (6.3 mL, 0.094 mol) in a 1:1
H2O/CHCl3 mixture (600 mL), was treated as previously
reported5 to yield a crude residue consisting of the Z
(6a±f) and E (7a±f) unsaturated oxime ethers in a ratio
of about 3/7, which was chromatographed on silica gel,
eluting with a 7/1/0.15 hexane/CH2Cl2/methyl ethyl
ketone mixture and collecting 15 mL fractions. The ®rst
fractions a€orded pure 7a±f, whereas the subsequent
1
ones yielded pure 6a±f as oils, whose H NMR spectral
data were in agreement with those previously detected
by us for the same compounds in the crude E/Z mix-
tures.5 6a (3.8 g, 25%). Anal. C10H11NO (C, H, N). 6b
(4.3 g, 24%). Anal. C11H13NO2 (C, H, N). 6c (4.1 g,
22%). Anal. C11H13NO2 (C, H, N). 6d (4.1 g, 22%).

A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
2159
Anal. C11H13NO2 (C, H, N). 6e (4.3 g, 23%). Anal.
C10H10NOCl (C, H, N). 6f (3.1 g, 17%). Anal.
C10H10NOCl (C, H, N). Compound 6g was synthesized
and puri®ed as previously reported.5
56%). 3g (0.30 g, 58%). For 1H NMR spectral data, see
Table 2.
The appropriate aminoalcohols 3a±g (0.001 mol) were
dissolved in anhydrous Et2O (2 mL) and treated in por-
tions at 0 ꢀC, under stirring, with a solution of oxalic
acid (0.072 g, 0.0008 mol) in anhydrous MeOH (1 mL).
Addition of anhydrous Et2O gave a solid precipitate
consisting of pure 3a±g‰H2C2O4. For analytical and
General procedure for the preparation of (Z)-N-(2,3-
epoxypropylidene)(arylmethyloxy)amines (8a±g). A vig-
orously stirred heterogeneous mixture of the appro-
priate unsaturated oxime ether 6a±g (0.015 mol), 18-
crown-6 (0.680 mg, 0.0026 mol) and acetone (17 mL) in
freshly distilled benzene (85 mL) and saturated phos-
phate aqueous pH 7.5 bu€er solution (45 mL) contain-
ing an undissolved quantity of solid KH2PO4 (prepared
by adding portionwise with stirring KH2PO4 to 100 mL
of freshly distilled water until saturation and then
treating with solid KOH until the pH was 7.5, being
careful to maintain an undissolved quantity of solid
KH2PO4), was treated dropwise with a freshly prepared
solution of 0.4 M potassium peroxymonosulfate (oxone)
(51 mL), being careful to keep the pH constant (7.5) by
the addition of solid KOH. The mixture was left to react
at rt under vigorous stirring, further adding 10 mL of
0.4 M potassium peroxymonosulfate (oxone) and 5 mL
of acetone every 24 h (again keeping the pH at 7.5 with
solid KOH) until the ratio of the amounts of unsatu-
rated oxime ether and (Z)-epoxide was constant (glc)
(from 4 to 9 days). The ®nal percentage of epoxide in
the reaction mixture varied from 50 to 82% (GLC).
The two phases were then separated and the aqueous
one was extracted with benzene (2Â100 mL). The
organic layers were collected, dried and evaporated to
give a crude residue which after puri®cation by column
chromatography on silica gel, eluting with hexane/Et2O
mixture (40:1 for 8a, 8e±g and 40:3 for 8b±d), yielded the
pure (Z)-epoxides 8a±g as oils, whose 1H NMR spectral
data were in agreement with those previously detected
for the same compounds in the crude mixtures of 8a±g
together with their corresponding E isomers.5 8a (0.89 g,
33%). Anal. C10H11NO2 (C, H, N). 8b (1.5 g, 48%).
Anal. C11H13NO3 (C, H, N). 8c (1.3 g, 42%). Anal.
C11H13NO3 (C, H, N). 8d (1.0 g, 32%). Anal. C11H13NO3
(C, H, N). 8e (1.3 g, 41%). Anal. C10H10NO2Cl (C, H,
N). 8f (1.3 g, 41%). Anal. C10H10NO2Cl (C, H, N). 8g
(1.6 g, 51%). Anal. C10H10NO2Cl (C, H, N).
1
chemical data, see Table 1. H NMR data for N ˆ CH
proton (d, Hz). 3a‰H2C2O4: 6.90 (d, J=4.8); 3b‰
H2C2O4: 6.88 (d, J=4.8); 3c‰H2C2O4: 6.92 (d, J=4.9);
3d‰H2C2O4: 6.89 (d, J=4.7); 3e‰H2C2O4: 6.90 (d,
J=4.8); 3f‰H2C2O4: 6.92 (d, J=4.8); 3g‰H2C2O4:
6.92 (d, J=4.8).
General procedure for the preparation of (Z)-N-[3-(tert-
butylamino)-2-hydroxypropylidene](arylmethyloxy)amine
oxalates (4a±g‰H2C2O4). A stirred solution of the
appropriate epoxide 8a±g (0.0019 mol) in anhydrous
benzene (7 mL) and t-BuNH2 (1.2 mL, 0.0114 mol)
was kept at 90 ꢀC for 8 days and then treated, as
described in the preparation of 3a±g, to yield pure 4a±g
as free bases. 4a (0.28 g, 59%). 4b (0.30 g, 56%). 4c
(0.31 g, 58%). 4d (0.33 g, 62%). 4e (0.35 g, 65%). 4f
1
(0.32 g, 59%). 4g (0.30 g, 56%). For H NMR spectral
data, see Table 2.
The appropriate aminoalcohols 4a±g (0.001 mol) were
dissolved in anhydrous Et2O (2 mL) and treated with
oxalic acid (0.072 g, 0.0008 mol) as described in the pre-
paration of 3a±g‰H2C2O4, to give a solid precipitate
consisting of pure 4a±g‰H2C2O4. For analytical and
1
chemical data, see Table 1. H NMR data for N ˆ CH
proton (d, Hz). 4a‰H2C2O4: 6.91 (d, J=4.7);
4b‰H2C2O4: 6.92 (d, J=4.6); 4c‰H2C2O4: 6.90 (d,
J=4.8); 4d‰H2C2O4: 6.91 (d, J=4.7); 4e‰H2C2O4:
6.90 (d, J=4.7); 4f‰H2C2O4: 6.92 (d, J=4.8); 4g‰
H2C2O4: 6.91 (d, J=4.8).
Stability test for 4g‰H2C2O4. The Z aminoalcohol
4g‰H2C2O4 (50 mg) was incubated for 60 min with
25 mL of either 50 mM Tris-HCl bu€er at pH 8 (used in
the radioligand binding experiments) or Tyrode solution
(used in the pharmacological tests), at the temperature
used in the biopharmacological tests. After alkaliniza-
tion with 5% aqueous K2CO3 solution, the usual work-
up made it possible to recover the unaltered Z-MOIM
oxime 4g (1H NMR).
General procedure for the preparation of (Z)-N-[3-
(isopropylamino)-2-hydroxypropylidene](arylmethyloxy)-
amine oxalates (3a-g‰H2C2O4). A stirred solution of
the appropriate epoxide 8a±g (0.0019 mol) in anhydrous
benzene (7 mL) and i-PrNH2 (0.97 mL, 0.0114 mol) was
kept at 90 ꢀC for 4 days. The resulting mixture was eva-
porated to give a residue, which, after puri®cation by
column chromatography eluting with a 60:5:4 hexane/
AcOEt/i-PrNH2 mixture, yielded pure 3a±g as free
bases. 3a (0.25 g, 56%). 3b (0.29 g, 57%). 3c (0.30 g,
59%). 3d (0.31 g, 61%). 3e (0.31 g, 60%). 3f (0.29 g,
Theoretical calculations: conformational analysis. All
calculations were made by means of the molecular
mechanics program Discover,14 using the CVFF force®eld,
and a dielectric constant equal to 4 and distance-depen-
dent. The results were reported for molecules considered
as free bases because no signi®cant di€erence was

2160
A. Balsamo et al./Bioorg. Med. Chem. 6 (1998) 2151±2160
observed with respect to the N-protonated forms. The
starting conformation of 10 was built from the preferred
one of 9;5 the rotational freedom of its hydroxyl group
was limited in order to prevent the formation of an
intramolecular H bond between this group and the
ethereal oxygen, which should be unfavoured in the
biological environment.15
Acknowledgements
This work was supported in part by a grant from the
Ministero dell'Universita e della Ricerca Scienti®ca e
Tecnologica.
References
A full geometry optimization was performed on the
compounds 3a±g and 4a±g, starting from the preferred
geometry of 10.
1. (a) Macchia, B.; Balsamo, A.; Lapucci, A.; Martinelli, A.;
Macchia, F.; Breschi, M. C.; Fantoni, B.; Martinotti, E. J.
Med. Chem. 1985, 28, 153. (b) Balsamo, A.; Gentili, D.;
Lapucci, A.; Macchia, M.; Martinelli, A.; Orlandini, E.; Ferni,
G.; Pinza, M. Il Farmaco 1994, 49, 759. (c) Gentili, D.; Lapucci,
A.; Macchia, B.; Macchia, M.; Martinelli, A.; Nencetti, S.;
Orlandini, E.; Ferni, G.; Pinza, M. Il Farmaco 1995, 50, 519.
Molecular electrostatic potential. The MEP was calcu-
lated by using the STO3G wave functions of 9 and 10
considered in the conformations of Figure 2 in a three-
dimensional grid with a step of 0.5 A. The grid was then
contoured at levels of 10 and 20 kcal/mol using the
molecular modelling program InsightII.14
2. Balsamo, A.; Breschi, M. C.; Chini, M.; Domiano, P.;
Giannaccini, G.; Lucacchini, A.; Macchia, B.; Macchia, M.;
Manera, C.; Martinelli, A.; Martini, C.; Martinotti, E.; Nieri,
P.; Rossello, A. Eur. J. Med. Chem. 1992, 27, 751.
3. Macchia, B.; Balsamo, A.; Breschi, M. C.; Chiellini, G.;
Macchia, M.; Martinelli, A.; Martini, C.; Nardini, C.; Nencetti,
S.; Rossello, A.; Scatizzi, R. J. Med. Chem. 1994, 37, 1518.
An analogous MEP calculation was performed on
compounds 3b±g and 4b±g considered in their preferred
conformations.
4. Balsamo, A.; Breschi, M. C.; Chiellini, G.; Lucacchini, A.;
Macchia, M.; Martinelli, A.; Martini, C.; Nardini, C.; Orlan-
dini, E.; Romagnoli, F.; Rossello, A. Eur. J. Med. Chem. 1994,
29, 855.
Radioligand binding methods: rat brain ꢀ1-receptors. b1-
Receptors were assayed in rat cortical membranes, as
previously described,3 using [3H]CGP 2650511 (1-[[2-(3-
carbamoyl-4-hydroxyphenoxy)ethyl]amino]-3-[4-[1-meth-
yl-4-(tri¯uoromethyl)-2-imidazolyl]phenoxy]-2-propanol)
as the speci®c ligand (DuPont de Nemours, New Eng-
land Nuclear Division; speci®c activity 2.5 Ci/mmol).
5. Balsamo, A.; Breschi, M. C.; Chiellini, G.; Favero, L.;
Macchia, M.; Martinelli, A.; Martini, C.; Rossello, A.; Scatizzi,
R. Eur. J. Med. Chem. 1995, 30, 743.
6. Breschi, M. C.; Macchia, M.; Manera, C.; Micali, E.; Nar-
dini, C.; Nencetti, S.; Rossello, A.; Scatizzi, R. Eur. J. Med.
Chem. 1996, 31, 159.
Bovine lung ꢀ2-receptors. b2-Receptor binding was stu-
died in bovine lung, as previously described,3 using
[3H]dihydroalprenolol (DHA)12 as the ligand (DuPont
de Nemours, New England Nuclear Division; speci®c
activity 48.1 Ci/mmol), in the presence of CGP 26505.
7. Balsamo, A.; Breschi, M. C.; Chiellini, G.; Macchia, B.;
Macchia, M.; Manera, C.; Sacca, P.; Scatizzi, R. Eur. J. Med.
Chem. 1996, 31, 199.
8. Balsamo, A.; Breschi, M. C.; Chiellini, G.; Cozzini, P.;
Domiano, P.; Macchia, M.; Manera, C.; Martinelli, A.; Nen-
cetti, S.; Rossello, A.; Sacca, P.; Scatizzi, R. Eur. J. Med. Chem.
1996, 31, 291.
Pharmacological methods: guinea pig atria and guinea
pig tracheal strips. The activity of compounds 4a, c, e±g
on b-adrenoceptors was evaluated on isolated prepara-
tions obtained from adult male Dunkin±Hartley guinea
pigs, weighing 300±350 g.
9. Balsamo, A.; Bel®ore, M. S.; Macchia, M.; Martini, C.;
Nencetti, S.; Orlandini, E.; Rossello, A. Eur. J. Med. Chem.
1994, 29, 787.
10. Curci, R.; Fiorentino, M.; Troisi, L. J. Org. Chem. 1980,
45, 4758.
The ecacy of the compounds tested on b1- and b2-
adrenoceptors was experimented on preparations of
isolated guinea-pig atria and of tracheal smooth mus-
culature respectively, following the methods previously
described.3 For both b1 and b2 preparations, the antag-
onistic activity of the compounds towards b1- and b2-
adrenoceptors was expressed as pIC50, (i.e. the negative
log of the molar concentration that reduced the
response to isoprenaline by 50%).16 All compounds
were tested at a concentration ranging from 10 9 M to
10 3 M. Each antagonistic activity index was obtained
from at least ®ve active concentrations.
11. Dooley, D. J.; Bittiger, H.; Reymann, N. C. Eur. J. Phar-
macol. 1986, 130, 137.
12. Nahorski, S. R.; Richardson, A. Br. J. Pharmacol. 1979,
66, 469.
13. Minneman, K.; Hegstrand, L. R.; Molino€, P. B. Mol.
Pharmacol. 1979, 16, 34.
14. Insight II Version 2.3; Discover Version 2.9.5 Biosym
Technologies, San Diego. USA.
15. Macchia, B.; Macchia, F.; Martinelli, A. Eur. J. Med.
Chem. 1983, 18, 85.
16. Hernauder, M.; Prieto, D.; Simonsen, V.; Rivera, L.; Bar-
abona, M. V.; Garcia, S. Br. J. Pharmacol. 1992, 107, 924.