1898
M. Fujita et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1897–1900
Scheme 1. Reagents and conditions: (a) NCCH2R1, S, Et3N/DMF or Et3N/EtOH or morpholine/MeOH, 60 ꢀC–reflux, 37–93%;(b) NaOH,
MeOH–H2O, 0 ꢀC–rt, 2b: 34%, 2c: 59%;(c) Ac 2O or AcCl, pyridine, rt, 45–56%;(d) R COX, pyridine, 27%–quant. or amines, triphosgene, Et3N,
2
CH2Cl2, 0 ꢀC–rt, 32–92%;(e) R 3X, NaH, DMF, 11–87%.
and dihydropyran ring, and N-substitution at the
6-position. In the case of the in vitro activity, dihy-
dropyran (3c), N-benzoyl (3e) and N-ethyl analogues
(3g) were more excellent than the lead compound 3d.
However, compounds 3c and 3e indicated weaker in
the in vivo activity. From these results, it is likely that
N-acetyl and N-ethyl analogues may be totally desir-
able, nevertheless that the modification of this moiety
may not play an important role for the activity.
groups showed more potent activity, whereas the other
alkyl groups were almost equipotent in both the in vitro
and in vivo activities (3d vs 3o–t). In particular, com-
pound 3r revealed excellent activity. Replacement of the
ethyl group with an ethoxy or an ethylamino group had
diminished potency (3w and 3aa). Moreover, diethyl-
amino analogue 3cc compared with compound 3aa
showed more potent activity and especially excellent
in vivo activity, while cyclohexylethylamino analogue
3dd showed similar activity. Capping the ethyl with a
4-morpholinyl group makes little difference to activity
(3w vs 3x). Furthermore, replacement of the ethylamino
(3aa) with heterocyclic amino groups showed equally or
slightly more potent activity (3y–z), whilst changing
from the diethylamino (3cc) to cyclic secondary amino
groups had diminished potency except for 3jj (see 3ff–ll).
These results suggest that neutral or basic natures in this
region are not an important factor for the activity. The
introductions of both polar and bulky groups were
found to lead to increase in the in vitro activity, but to
decrease in the in vivo activity (3bb and 3ee). As pre-
viously described, N-methylation of the amide 3r and
the urea 3cc indicated weaker activity despite slightly in
the case of 3r (see 4a–b). Further, changing the acetyl at
the 6-position to ethyl group, 3r had diminished
potency, but 3cc had increased in vitro activity despite
equally in the in vivo activity (3mm–nn).7
Next, we modified the substituents at the 3-position
(Table 2). In the in vitro activity, the esters (3j–k) and
the nitrile (3n) compared with compound 3d showed
more potent and equipotent activity, respectively,
despite the carboxylic acid (3l) and the carboxamide
(3m) were inactive. In the in vivo activity, compounds
3j–k exhibited considerably weaker activity, considering
the in vitro activity. Therefore, these suggest that the
bulky substituents of the ester moiety are less suitable
for the in vivo activity.
Finally, as shown in Table 3, we investigated N-sub-
stitution at the 2-position. The methyl, ethyl and t-butyl
Table 1. In vitro and vivo inhibitory activities of TNF-a production
for compounds 3a–i
As the target of development of arthritis, compound 3cc
which possesses excellent in vivo activity was evaluated
Compd
X
In vitro
IC50 (mM)a
In vivo
inhibition (%)
at 50 mg/kg, pob
Table 2. In vitro and vivo inhibitory activities of TNF-a production
for compounds 3d and 3j–n
3a
3b
CH2
S
28
34
1.8
21.2
3c
3d
O
AcN
10
22
17.6
49.1
3e
3f
3g
3h
3i
BzN
MeN
EtN
10
32
15
40
31.8
22.7
Compd
R1
In vitro
IC50 (mM)a
In vivo
inhibition (%)
at 50 mg/kg, pob
69.2
Me(CH2)2N
BnN
NTc
NT
3d
3j
3k
3l
3m
3n
COOEt
COOCHMe2
COOCH2CH(Et)(CH2)3Me
COOH
22
7.1
10
>100
>100
22
49.1
42.1
10.6
NTc
NT
42
FR133605
Dexamethasone
Pentoxifylline
6.0Æ4.2
0.02
40Æ2
90.3Æ5.9
NTd
NTe
CONH2
CN
aIC50 of LPS-stimulated TNF-a production in rat whole blood.
bInhibition of LPS-stimulated serum TNF-a production in the rat.
cNT, not tested.
23.2
aIC50 of LPS-stimulated TNF-a production in rat whole blood.
bInhibition of LPS-stimulated serum TNF-a production in the rat.
cNT, not tested.
dED50=0.12Æ0.06 mg/kg, po.
e74.4% at 100 mg/kg, po.