Evaluation of pyrrolidin-2-imines and 1,3-thiazolidin-2-imines as inhibitors of nitric oxide synthase

Article abstract of DOI:10.1016/j.bmcl.2004.06.033

Syntheses and evaluation of pyrrolidin-2-imines and 1,3-thiazolidin-2- imines as inhibitors of nitric oxide synthase (NOS) are discussed. An extensive SAR was established for pyrrolidin-2-imines class of compounds. The amidines came out as the most potent inhibitors in addition to displaying selectivity.

Full text of DOI:10.1016/j.bmcl.2004.06.033

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4539–4544
Evaluation of pyrrolidin-2-imines and 1,3-thiazolidin-2-imines as
inhibitors of nitric oxide synthaseq
K. Shankaran,a,* Karla L. Donnelly,a Shrenik K. Shah,a Ravindra N. Guthikonda,a
Malcolm MacCoss,a John L. Humes,b Stephen G. Pacholok,b Stephan K. Grant,c
T. M. Kellyc and K. K. Wongc
aDepartment of Medicinal Chemistry, Merck Research Laboratory, PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Immunology and Inflammations, Merck Research Laboratory, PO Box 2000, Rahway, NJ 07065, USA
cDepartment of Enzymology, Merck Research Laboratory, PO Box 2000, Rahway, NJ 07065, USA
Received 23 April 2004; revised 10 June 2004; accepted 10 June 2004
Available online 19 July 2004
Abstract—Syntheses and evaluation of pyrrolidin-2-imines and 1,3-thiazolidin-2-imines as inhibitors of nitric oxide synthase (NOS)
are discussed. An extensive SAR was established for pyrrolidin-2-imines class of compounds. The amidines came out as the most
potent inhibitors in addition to displaying selectivity.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
NOS3) isoform are involved in neurotransmission, and
vasodilatation, respectively.3 The inducible form of
nitric oxide synthase (iNOS or NOS2) is concerned with
host defense, septic shock, and inflammatory diseases.4
Thus, selective inhibition of iNOS over the other
isoforms may offer a new therapy for these diseases.
Pursuit of NOS inhibitors has lead to amino acid ana-
logs5 as well as nonamino acid based inhibitors includ-
ing aminoguanidines,6 isothioureas,7 and amidines,8
benzoxazolones,9 and pyridines.10 A recent communi-
cation11a disclosed that the analogs of 1,3-oxazolidin-2-
imine derivatives (2) were shown to be inhibitors of the
NOS enzymes prompted us to communicate our findings
in these area.
Nitric oxide synthases (NOS) are a group of oxidative
enzymes, which produce nitric oxide (NO) from
-arginine.2 These enzymes bring about a five electron
L
oxidation of
shown below.
L
-arginine to nitric oxide and
L
-citrulline as
FAD, FMN, Ca/CaM
Tetrahydrobiopterin (H 4B)
NH
O
COOH
NH2
L-Arginine
COOH
H2N
N
H2N
N
O2, NOS
H
H
NH2
L-Citrulline
•NO
We had previously reported that piperidin-2-imines (1)
are potent inhibitors12 of NOS enzymes displaying
varying degree of selectivity. In parallel efforts we1 and
others studied pyrrolidin-2-imines as NOS inhibitors
and toward this end Hagen and co-workers has already
disclosed their results.11b In addition to pyrrolidin-2-
imines we also probed the related thiazolidin-2-imines
and it was hoped that these classes (3) might provide
selective iNOS inhibitors. The results of such a study are
disclosed herein. While work related to the 1,3-thiazo-
lidin-2-imines lead class was minimal in scope, a detailed
structure–activity relationship was established for the
pyrrolidin-2-imines.
Nitric Oxide
To date three isoform of NOS enzymes have been dis-
covered. Of these, the two constitutive versions namely,
neuronal (nNOS or NOS1) and endothelial (eNOS or
Keywords: NOS; Pyrrolidin-2-imine; 1,3-Thiazolidin-2-imine.
q See Ref. 1.
* Corresponding author. Tel.: +1-732-594-3979; fax: +1-732-594-3007;
e-mail: kothandaraman_shankaran@merck.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.06.033

4540
K. Shankaran et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4539–4544
a, b
COOMe
OH
3
c, d
R
R
R2
R3
( )n
O
X
1
NBoc
12
( )n
NBoc
NH
NH
H
13
5
N
H
NH
N
H
N
H
e
X = S, CHR1
1
( )n
15, n =1
16, n =2
2
( )n
3
N
N
14
O
NH
Scheme 3. Reagents and conditions: (a) Oxalyl chloride, DMSO,
DCM, NEt3; (b ) PhP@CHCOOMe, DCM, rt; (c) Pd/C, H2, MeOH;
3
(d) HCl and then K2CO3; (e) Method B, steps (a), (b), and (c) (Scheme
1).
2. Chemistry
Scheme 1 shows the preparation of mono- and disub-
stituted pyrrolidin-2-imines essentially in three steps.
The first two steps consists of Michael addition of var-
ious nitroalkanes to the acrylates 4 followed by the
catalytic hydrogenation of the intermediate nitroesters
to give the lactams 5.13 The lactams 5 were transformed
to the desired amidines following one of the several
conditions shown below. Method A, which involved the
conversion of 5 to the iminoether 6 followed by reflux
with ammonium chloride in ethanol to furnish 8, as a
mixture of stereoisomers was the first choice. In some
incidences, the iminoether 6 was too volatile to be iso-
lated and for these systems method B was preferred.
This required the initial conversion of the amides 5 to
the thioamides 7 by Lawesson’s reagent.14 The thio-
amides 7 were then quaternized with excess iodo-
methane followed by reaction with methanolic ammonia
to afford the amidines 8 as a hydroiodide salt. Alterna-
tively, the thioamides 7 could be directly converted to
the amidines 8 in a procedure utilizing mercuric chloride
and gaseous ammonia as reported in the literature.15
OH
S
S
a
b
R
R
S
NH
R
NH2
N
H
N
H
17
19
18
Scheme 4. Reagents and conditions: (a) KOH, CS2, EtOH reflux;
(b) Method A, steps (a) and (b) (Scheme 1).
a, b
c
( )n
OH
S
( )n
S
( )n
NH
N
N
S
NH
20, n =1
21
22, n =1
23, n =2
Scheme 5. Reagents and conditions: (a) SOCl2, CHCl3, reflux;
(b) NEt3, CS2, CH2Cl2; (c) Method B, steps (b) and (c) (Scheme 1).
The fused amidine 15 was prepared from N-Boc pro-
tected aminol 12 as shown in Scheme 3. In a sequence of
reaction that involved Swern oxidation of 12 followed
by the Wittig–Horner reaction gave a mixtures of esters
13. Catalytic reduction of 13 followed by N-Boc removal
led to a rapid cyclization affording 14. The amide 14 was
transformed to 15 by the method B as in Scheme 1.
Likewise, starting from the corresponding piperidinol
and following the same procedure for 15 gave 16 (n ¼ 2).
The preparation of fused amidine 11 from the known16
iodolactam 9 is shown in Scheme 2. The reduction of 9
to 10 was efficiently accomplished under tinhydride
conditions. The transformation of 10 to the amidine 11
was analogous to the transformation 5 to 8 (method B)
shown in Scheme 1.
A general synthesis of 1,3-thiazolidin-2-imines is shown
in Scheme 4. This involved the reaction of aminol 17 and
carbon disulfide under basic conditions17 to afford 18.
The thiazolidine-2-thiones 18 was transformed to
1,3-thiazolidin-2-imines 19 in two steps that involved
initial reaction with Meerwein salt (to afford thioimino-
ether) that was followed by refluxing with ammonium
chloride (Scheme 1, method A).
R2
R3
R1
R2
R1
OMe
R2
R1
COOMe
a, b
Method A
a
R3
O
N
N
4
H
5
6
b
Method B
a
R2
R1
NH
R2
R1
b, c or d
The bicyclic 1,3-thiozol-3-imine analogs 22 and 23 were
accessed by a similar approach shown above for 19, b ut
with a slight variation as shown in Scheme 5. A two step
reaction that involved the transformation of aminol 20
to the intermediate chloro compound followed by
reaction with CS2/triethylamine gave 21. The thione 21
was transformed to 22 by method B. Similarly, starting
from 2-piperidinemethanol and following the aforesaid
steps gave 23.
R3
R3
S
N
H
N
H
7
8
Scheme 1. Reagents and conditions: (a) R3CH2NO2, DBU, MeOH;
(b) PtO2, H2, EtOH; (c) Method A (a) Me3OBF4, CH2Cl2, rt; (b)
NH4Cl, EtOH, reflux; Method B (a) Lawesson’s reagent, toluene,
90 °C; (b) MeI, rt; (c) NH3, MeOH; (d) NH3, HgCl2, THF, 60 °C.
a
b
O
O
NH
N
N
N
H
H
H
I
3. Results and discussion
9
11
10
The hydrochloride/hydroiodide salts of pyrrolidine-2-
imines and 1,3-thiazolidin-2-imines were evaluated for
Scheme 2. Reagents and conditions: (a) n-Bu3SnH, AIBN, PhH reflux;
(b) Method B, steps (a) and (d) (Scheme 1).

K. Shankaran et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4539–4544
4541
NOS activity by an assay protocol18 that used re-
combinant human version of all three isoform of NOS.
The binding results for both the pyrrolidin-2-imines and
1,3-thiazolidin-2-imines classes are compared with the
selectivity for iNOS over eNOS. Finally, the compound
with phenyl at C-4 (8h) was inactive.
An observation (from Table 1) that the amidine 8e was
not only potent (iNOS) but also displayed modest
selectivity encouraged us to prepare analogs of 8e with
additional substituents on pyrrolidine core to improve
its profile further. The results of binding data from such
a study compared with 8e are shown in Table 2.
standard (L)-N-iminoethyl-lysine (L-NIL). The SAR
studies on the pyrrolidin-2-imines analogs 8 centered on
the placement of alkyl substituents at all the available
positions of pyrrolidine core. The results from such a
study, compared to the parent amidine (8a) are dis-
played in Table 1. The analog (8c) with the methyl group
at C-4 of the pyrrolidine ring lead to significant
improvement in the binding potency for the iNOS (20-
folds) and nNOS (12-folds) as compared to 8a. Com-
pounds 8b and 8d bearing the methyl at C-3 and C-5
showed a modest 2- to 3-folds improvement in potency
for iNOS, however these compounds (8b–d) displayed
no selectivity. In probing other alkyl residues at C-4, it
became clear that the amidine analog 8e (with ethyl at
C-4) led to somewhat less potent compound compared
to 8c. More importantly, 8e displayed better selectivity
for iNOS (over eNOS) while the selectivity over nNOS
was not as good. The gain in potency for 8e led to the
preparation of other analogs 8f–h. The n-propyl analog
8f had lost significant activity compared to either 8c or
8e. This was also true for 8g but it still maintained good
The placement of alkyl substituents at C-3 and C-4
leading to 8i resulted in an analog that was less potent
for iNOS but surprisingly was more potent for nNOS.
This was also reflected to some extent in 8j, whose
selectivity for iNOS over eNOS was very good. Despite
slight loss of potency, the selectivity displayed by 8j
(eNOS/iNOS) was the best we have seen for these ana-
logs. The C-3/C-5 disubstituted compound 8k had lost
significant binding for the iNOS as a result this substi-
tution pattern was not further pursued. The C-4/C-5
substituted compounds were explored in greater depth
(8l–p). The data for 8l,m, and 8o clearly indicates that
small alkyl substituents at C-4 or C-5 were the most
preferred as it lead to potent analogs (compared to 8e) in
addition to maintaining good selectivity. The amidines
Table 1. Inhibition of NOS by monosubstituted pyrrolidin-2-imines (IC50 ¼ lM)a
Entry
Structureb
iNOS
eNOS
Selectivity
eNOS/iNOS
nNOS
Selectivity
nNOS/iNOS
––
L
-NIL
2.1
4.1
7.9
4.6
3.7
17.5
9.6
8.3
8a
1.1
2.3
NH
NH
N
H
8b
8c
1.5
0.2
1.8
2.6
1.2
13
0.9
0.8
0.6
0.3
N
H
NH
N
H
8d
8e
1.3
0.5
1.9
1.4
27
3.1
2.6
2.3
4.8
NH
N
H
14.5
NH
N
H
8f
38.7
3.5
53
1.4
25
53
1.4
3
NH
N
H
8g
87.8
10.6
NH
NH
N
H
Ph
8h
>50
>50
ND
>50
ND
N
H
ND ¼ not determined.
a IC50s are obtained from 10 point titration using sigma plot, where each individual point is an average of duplicate determination at that
concentration.
b The pyrrolidin-2-imines were synthesized according to Scheme 1 and the IC50s are reported for racemic mixture.

4542
K. Shankaran et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4539–4544
Table 2. Inhibition of NOS by disubstituted pyrrolidin-2-imines (IC50 ¼ lM)a
Entry
Structureb
iNOS
eNOS
Selectivity
eNOS/iNOS
nNOS
2.6
Selectivity
nNOS/iNOS
8e
0.5
14.5
27
4.8
0.5
NH
N
H
8i
1.2
8.7
7.2
0.61
NH
N
H
8j
0.93
23
60.6
5.5
65
0.61
2.4
0.65
0.1
NH
NH
N
H
8k
0.24
N
H
8l
0.3
0.02
2
0.6
0.52
49
17
26
24
0.13
0.08
4.8
3.8
4.0
2.4
NH
NH
N
H
8m
8n
N
H
NH
N
H
8o
0.16
6.3
38
0.14
0.8
N
H
NH
8p
8q
13.1
10.2
104
227
8
27
2
NH
NH
N
H
22
17.3
1.7
N
H
a IC50s are obtained from 10 point titration using sigma plot, where each individual point is an average of duplicate determination at that
concentration.
b The pyrrolidin-2-imines were synthesized according to Scheme 1 and the IC50s are reported for the diastereomeric and racemic mixture.
8n,p, and 8q with n-propyl group at both C-5 and C-4
led to significant loss in iNOS activity. This result was
akin to the one seen for 8f (Table 1) as discussed pre-
viously. It is also worth noticing that both 8n and 8q still
retained good selectivity for iNOS over eNOS despite
loss of potency.
behavior was noticed for 15 and 16. The bicyclic analogs
17 and 18 were considerably more potent for nNOS
while the potency for the iNOS was unimpressive.
While we had modest success in the pyrrolidin-2-imines
lead class, the results of the binding assay for the
1,3-thiazolidin-2-imines analogs as shown below (Table
4) were far from satisfactory and stand in sharp contrast
with the results of pyrrolidin-2-imines discussed above.
While we had achieved significant potency going from
8e (Table 1) to 8m (Table 2) the selectivity remained
unchanged. We felt that perhaps the preparation of rigid
analogs might overcome the mediocre selectivity (over
nNOS) seen for 8m. The results of binding for the select
fused amidines are displayed in Table 3. In general the
fused amidines displayed binding activity that was
drastically less for the all three isoforms and more so for
the iNOS compared to 8m. In a dramatic reversal, all
fused analogs consistently maintained better binding
potency for the nNOS over iNOS. Thus, the amidine
analog 11 (fused version of 8m) was practically inactive
and fared much worse than we had hoped. Similar
To start with, the lead compound 19a was equipotent in
all three isoform, and showed lack of selectivity. Partial
attempts to improve the potency and selectivity specifi-
cally for the iNOS enzyme suggests that this lead class
was quite sensitive to the substituents on the thiazole
ring as exemplified by the analogs 19b,c, and 19d. All
three analogs had lost significant activity in binding
assay compared to 19a. Notice also that the fused ana-
log 22 and 23 were less impressive and share the trait
similar to 15 and 16 (Table 3). But for the fused analogs,

K. Shankaran et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4539–4544
4543
Table 3. Inhibition of NOS by fused pyrrolidin-2-imines (IC50 ¼ lM)a
Entry
Structureb
iNOS
eNOS
Selectivity
eNOS/iNOS
nNOS
0.08
Selectivity
nNOS/iNOS
8m
0.02
0.52
26
4.0
NH
NH
N
H
11
15
>50
>50
>50
ND
ND
9.9
9.6
ND
ND
N
H
N
33
NH
16
17
18
>50
3.1
>50
4.7
24
ND
1.5
20
>50
0.31
0.66
ND
0.1
N
NH
NH
N
H
1.2
0.55
NH
N
H
ND ¼ not determined.
a IC50s are obtained from 10 point titration using sigma plot, where each individual point is an average of duplicate determination at that
concentration.
b The amidine 11 was synthesized according to Scheme 2, 15 and 16 were prepared according to Scheme 3 and 17 and 18 by Scheme 1. The IC50 s are
reported for the racemic mixture.
Table 4. Inhibition of NOS by disubstituted 1,3-thiazolidin-2-imines (IC50 ¼ lM)a
Entry
Structureb
iNOS
eNOS
Selectivity
eNOS/iNOS
nNOS
Selectivity
nNOS/iNOS
––
L
-NIL
2.1
0.9
7.9
1.1
3.7
17.5
1.8
8.3
S
NH
19a
1.2
2
N
H
S
NH
19b
19c
>50
>50
>50
ND
ND
>50
ND
2.5
N
H
S
NH
5.2
13.2
N
H
S
NH
19d
22
>50
>50
>50
>50
ND
ND
>50
ND
ND
N
Ph
H
N
14
NH
S
23
N
>50
>50
ND
>50
ND
NH
S
ND ¼ not determined.
a IC50s are obtained from 10 point titration using sigma plot, where each individual point is an average of duplicate determination at that
concentration.
b The 1,3-thiazolidin-2-imines were synthesized according to Schemes 4 and 5 and the IC50s are reported for the racemic mixture.
the pyrrolidin-2-imines, which were investigated in
parallel with 1,3-thiazolidin-2-imines had more
encouraging outcome. For these reasons 1,3-thiazolidin-
2-imine as a lead class was not pursued further.
a

4544
K. Shankaran et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4539–4544
M.; Marshall, S.; Harmon, M. F.; Paith, J. E.; Fufine, E.
S. J. Biol. Chem. 1994, 269, 26669.
8. Moore, W. M.; Webber, K. R.; Fok, K. F.; Jerome,
G. M.; Conner, J. R.; Manning, P. T.; Wyatt, P. S.;
Misko, T. P.; Tjoeng, F. S.; Misko, T. P.; Currie, M. G.
J. Med. Chem. 1996, 39, 3886.
9. Shankaran, K.; Donnelly, K. L.; Shah, S. K.; Humes,
J. L.; Pacholok, S.; Green, B. G.; Grant, S. K.; MacCoss,
M. Bioorg. Med. Chem. Lett. 1997, 7, 2887.
10. Hagmann, W. K.; Caldwell, C. G.; Chen, P.; Durette, P.;
Esser, C. K.; Lanza, T. J.; Kopka, I. E.; Guthikonda, R.;
Shah, S. K.; Green, B. G.; Humes, J. L.; Kelly, T. M.;
Luell, S.; Meurer, R.; Moore, V.; Pacholok, S. G.; Pavia,
T.; Williams, H. R.; Wong, K. K. Bioorg. Med. Chem.
Lett. 2000, 10, 1975.
11. (a) Ueda, S.; Terauchi, H.; Yano, A.; Ido, M.; Matsu-
moto, M.; Kawasaki, M. Bioorg. Med. Chem. Lett. 2004,
14, 313; (b) Hagen, T. J.; Bergmanis, A. A.; Kramer,
S. W.; Fok, K. F.; Schmelzer, A. E.; Pitzele, B. S.;
Swenton, L.; Jerome, G. M.; Kornmeier, C. M.; Moore,
W. M.; Branson, L. F.; Conner, J. R.; Manning, P. T.;
Currie, M. G.; Hallinan, E. A. J. Med. Chem. 1998, 41,
3675.
12. Guthikonda, R. N. Abstract #90, National Medicinal
Chemistry Symposium, Ann Arbor, MI, 1996.
13. Moffett, R. B. Org. Synth. 1963, 4, 357.
14. Scheibye, S.; Pedersen, B. S.; Lawesson, S.-O. Bull. Soc.
Chim. Belg. 1978, 87, 229.
In conclusion, the synthesis and evaluation of pyrrol-
idin-2-imines and 1,3-thiazolidin-2-imines as iNOS
inhibitor have been demonstrated. Among the pyrrol-
idin-2-imines, 8l and 8m displayed superbpotency for
iNOS and was the best with in this class. Our limited
efforts in the 1,3-thiazolidin-2-imines suggest that they
were far less significant both in potency and selectivity
compared to the pyrrolidin-2-imines or for that matter
1,3-oxazolidin-2-imines class of compounds that was
reported recently.11a
In the final analysis, it has been our experience that, in
general, the pyrrolidin-2-imines lead class was less
potent and considerably less selective compared to the
piperidin-2-imines as iNOS inhibitors.12
References and notes
1. The part of results of the work described herein was
presented at the 214th ACS National Meeting, Las Vegas,
NV, 1997 (Abstract #037).
2. Several excellent reviews have appeared on this subject,
see: (a) Kerwin, J. F.; Lancaster, J. R.; Feldman, P. L.
J. Med. Chem. 1995, 38, 4343; (b) Marletta, M. A. Cell
1994, 78, 927; (c) Feldman, P. L.; Griffith, O. W.; Stuehr,
D. J. Chem. Eng. News 1993, 26; (d) Snyder, S. H.; Bredt,
D. S. Sci. Am. 1992; (May), 68.
15. Foloppe, M. P.; Rault, S.; Robba, M. Tetrahedron Lett.
1992, 33, 2803.
16. Knapp, S.; Gibson, F. S. Org. Synth. 1991, 70, 101.
17. Nagao, Y.; Hagiwara, Y.; Kumagai, T.; Ochiai, M.;
Inoue, T.; Hasimoto, K.; Fujita, E. J. Org. Chem. 1986,
51, 2391.
3. Moncada, S.; Higgs, A. N. Engl. J. Med. 1993, 329, 2002.
4. White, K. A.; Marletta, M. A. Biochemistry 1992, 31,
6627.
5. (a) Cochran, F. R.; Selph, J.; Sherman, P. Med. Chem.
Res. 1996, 16, 547; (b) Nussler, A. K.; Billiar, T. A.
J. Leukoc. Biol. 1993, 54, 171.
18. The NOS inhibitory activity of the compound was deter-
mined by comparing the conversion of 3H-
L
-arginine to
3H-citrulline in the presence of inhibitor with control. The
assay mixture containing 1 lM of 3H-
-arginine, cofactors
6. (a) Corbett, J. A.; Tilton, R. G.; Chang, K.; Hasan, K. S.;
Ido, Y.; Wang, J. L.; Sweetland, M. A.; Lancaster, J. R.;
Willamson, J. R.; McDaniel, M. L. Diabetes 1992, 41, 552;
(b) Joly, G. A.; Ayres, M.; Chelly, F.; Kilbourn, R. G.
Biochem. Biophys. Res. Commun. 1994, 199, 147.
7. (a) Nakane, M.; Klinghofer, V.; Kuk, J. E.; Donnelly,
J. L.; Budzik, G. P.; Pollock, J. S.; Basha, F.; Carter, G. E.
Mol. Pharmacol. 1995, 47, 831; (b) Garvey, E. P.;
Oplinger, J. A.; Tanoury, G. J.; Sherman, P. A.; Fowler,
L
and the inhibitor or aqueous DMSO (control) was incu-
bated for 30 min at room temperature. The reaction was
quenched by adding a slurry of Dowex 50W-K8 resin to
complex and remove the unreacted substrate. The concen-
tration of the 3H-
fluid was determined on a scintillation counter. Under the
assay conditions the production of -citrulline was linear
with time for the duration of the experiment.
L-citrulline product in the supernatant
L