1952
G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1951–1955
4-nitro- and 4-cyanoanilines with methyl malonyl chlo-
ride gave the corresponding malonamic acid methyl
esters 17 (91%) and 18 (87%). The nitro compound 17
and the nitrile 18 were hydrogenated to afford the corre-
sponding aniline and benzylamine derivatives, which
were reacted without further purification with metha-
nesulfonyl chloride using triethylamine as a base to give
the disulfonated and monosulfonated products 19 (82%)
and 20 (15%), respectively. The sulfonamides 19 and 20
were treated with lithium hydroxide to give the N-(4-
methanesulfonylamino)- and N-(4-methanesulfonylami-
nomethyl)malonamic acids 21 (88%) and 22 (95%),
which were converted to the sulfonylamino-bearing tar-
get compounds 5 (48%) and 6 (80%) in the same manner
described above.
O
Cl
Cl
OH
O
O
N
N
N
X
N
CH3
n
CH3
ICI 1994415
Ki (κ) = 0.043 nM
μ/κ: 1233; δ/κ: 558
112
n = 0: 2,3-Dihydrobenzofuran class
n = 1: Chromane class
O
H
O
O
OH
OH
N
N
N
N
Ar
N
X
N
n
H
CH3
CH3
213
Ki (κ) = 0.17 nM
μ/κ: 5900; δ/κ: 510
I
In order to investigate the importance of the chain
length linking the two amide functional groups in the
molecules for the j affinity, homologs of the malona-
mides were prepared. The succinamide analogs 7
(76%) and 10 (37%) were prepared via the malonamic
acids 23 (100%) and 24 (90%) by following the same
reaction sequence as for the synthesis of 3 and 9 except
that methyl succinyl chloride replaced methyl malonyl
chloride in the first step. The pyridine succinamide ana-
log 12 (25%) was synthesized via the intermediate acid
25 (66%) in the same manner as malonamide 11 except
that benzyl succinic chloride19 replaced benzyl malonyl
chloride as the starting material. The urea analog 8 with
one methylene group (CH2) in succinamide 7 replaced
by amino group (NH) was prepared via a three-step
reaction sequence: reaction of the glycine methyl ester
with phenyl isocyanate (100%) followed by treatment
with lithium hydroxide (96%), and conversion of the
resulting acid 26 to the target compound 8 (83%) using
the standard procedure.
(this work)
The synthesis of the target compounds 3–12 is summa-
rized in Scheme 1. Reaction of aniline and benzylamine
with methyl malonyl chloride followed by hydrolysis of
the resulting methyl esters with lithium hydroxide gave
the corresponding N-phenyl and N-benzyl malonamic
acids 13 (85%) and 14 (83%). These intermediates were
coupled with 1-(2-methylamino-(S)-2-phenyl-ethyl)-
pyrrolidin-(S)-3-ol15 using Mukaiyama reagent16 as the
acylating reagent to furnish the target compounds 3
(79%) and 4 (79%). The analogs with the replacement
of the phenyl group with the nitrogen-containing hetero-
aromatics to increase hydrophilicity and polarity were
also prepared. By following the same reaction sequence,
the thiazole analog 9 (64%) was synthesized using
2-aminothiazole as the starting material via the interme-
diate acid 15 (38%). For the synthesis of the pyridine
analog 11, the synthetic approach was modified using
a benzyl ester instead of a methyl ester protection strat-
egy for the carboxyl group. Reaction of 3-aminopyri-
dine with benzyl malonyl chloride17 gave the
corresponding malonamic acid benzyl ester (71%),
which was then converted to the N-pyridine-malonamic
acid 16 (50%) by hydrogenation. This reaction sequence
was convenient, avoiding an aqueous work-up of the
water soluble zwitterionic pyridine carboxylic acid
product. The resulting acid was converted to 11 (60%)
using the above-mentioned method.
Malonamide derivatives 3–1220 were evaluated in the
in vitro opioid receptor binding assays and the results
are shown in Table 1.21 Malonamide 3 showed sub-
nanomolar j binding affinity (Ki = 0.27 nM) compara-
ble to the phenylamino acetamide 2 (Ki = 0.17 nM), high
selectivity over l, and moderate selectivity versus d
receptor. In contrast, compound 4 with the replacement
of aniline with benzylamine showed more than a 50-fold
loss of j binding affinity. Compounds 9 and 11, in which
the phenyl was replaced with the heteroaromatic rings,
thiazolyl and pyridinyl, had a slightly decreased j bind-
ing affinity, but increased selectivity over the l receptor
compared with compound 3. Compounds 5 and 6, with
sulfonylamino groups substituted on the phenyl ring,
groups that were well tolerated in the previous series,
12,13 had a significantly decreased j affinity and also lost
l opioid receptor selectivity. The sulfonylamino substi-
tution was not tolerated in the malonamide series. The
homologs of the malonamides with a two-carbon tether,
succinamides 7, 10, 12, were 20- to 200-fold less active at
the j receptor than their corresponding malonamides 3,
9, 11. The urea analog 8 showed a similar 20-fold loss of
j affinity in comparison to the malonamide 3. These
data indicate that the chain length linking the two amide
functionalities in the molecules is important for optimal
j binding, with a one-atom tether (CH2) preferred over
In our previous studies of the constrained chroman-2-
carboxamides and 2,3-dihydrobenzofuran-2-carboxa-
mides (compounds 1)12 as potent j agonists, we found
that the sulfonylamino groups, which are para to the
oxygen, are the most preferred substituents at the phenyl
ring for obtaining both high j affinity and low inhibitory
activity at CYP2D6.12,18 In the phenylamino acetamide
series,13 sulfonylamino-bearing agonists were well toler-
ated by the j receptor, and weak CYP2D6 inhibitors,
with para-substitution having highest receptor selectivi-
ty. We incorporated this SAR information into the
structure of our new chemical series of malonamide
derivatives I. Thus, compounds 5 and 6 which contain
para sulfonylamino groups were prepared. Reaction of