Synthesis and screening of conformationally restricted and conformationally free N-(tertiary aminoalkyl)dithiocarbamic acids and esters as inhibitors of neuronal nitric oxide synthase

Article abstract of DOI:10.1016/S0968-0896(98)00132-1

N-(Tertiary aminoalkyl)dithiocarbamic acids and esters were synthesized and evaluated for their ability to inhibit neuronal nitric oxide synthase. Preliminary results show these compounds are able to act at the binding site for l-arginine and the conformationally restricted esters may have a second site of activity involving the cofactor (6R)-5,6,7,8-tetrahydro-l-biopterin. Copyright (C) 1997 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(98)00132-1

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 1759±1766
Synthesis and Screening of Conformationally Restricted and
Conformationally Free N-(Tertiary aminoalkyl)dithiocarbamic
Acids and Esters as Inhibitors of Neuronal
Nitric Oxide Synthase
Mark B. Sassaman,a,* John Giovanelli,b Virendar K. Soodc
and William C. Eckelmana
aPositron Emission Tomography Department, Warren G. Magnuson Clinical Center, National Institutes of Health, Bethesda,
MD 20892-1180, USA
bLaboratory of Neurochemistry, National Institute of Mental Health, Bethesda, MD 20892-4096, USA
cDepartment of Radiology, George Washington University Medical Center, Washington, DC 20037, USA
Received 6 February 1998; accepted 1 May 1998
AbstractÐN-(Tertiary aminoalkyl)dithiocarbamic acids and esters were synthesized and evaluated for their ability to
inhibit neuronal nitric oxide synthase. Preliminary results show these compounds are able to act at the binding site for
l-arginine and the conformationally restricted esters may have a second site of activity involving the cofactor (6R)-
5,6,7,8-tetrahydro-l-biopterin. Published by Elsevier Science Ltd.
Introduction
binding region of calmodulin acting as a hinge between
the two domains; and on the cofactor (6R)-5,6,7,8-tetra-
hydro-l-biopterin (BH4). These properties have been
reviewed by Grith and Stuehr.1
Nitric oxide synthase (NOS) catalyzes the oxidation of
l-arginine to l-citrulline and the biologically active free
radical, nitric oxide, in a two step, ®ve electron process,
involving NADPH and molecular oxygen. The enzyme
exists in two constitutive and one inducible isoform,
which bear a high degree of structural homology to one
another and to cytochrome P-450 reductase, in regions
associated with the binding of the cofactors FMN,
FAD, and NADPH. NOS contains an iron proto-
porphyrin IX (heme) prosthetic group bound to the
protein by axial coordination of the heme iron to a
cysteine thiolate and a vacant second axial position is
available to bind molecular oxygen. The protein is thus
organized into a reductase and an oxygenase domain,
the latter being associated with the binding site for the
natural substrate, l-arginine. In addition, NOS shows
dependence on the presence of Ca2+/calmodulin, the
The role of BH4 is unclear, however, a growing body of
evidence supports several complimentary functions
including stabilization of the catalytically active NOS
homodimer,2 enhancement of the binding of l-arginine
to the enzyme,3 facilitating the coupling of reductive
oxygen activation to the oxygenation of l-arginine,3,4
protecting the enzyme from product inhibition,5 and
involvement in the overall odd-electron redox process.6±8
BH4 resides in the oxygenase domain2,9 in close prox-
imity to the heme prosthetic group; binding of BH4 to
NOS may, like heme, involve a cysteine residue.10
The majority of NOS inhibitors are modi®ed arginines,
other guanidines, or bioisosteric equivalents, such as
amidines, S-alkylisothioureas, and iminoazahetero-
cycles.11,12 l-Thiocitrulline,13 obtained by replacement
of l-citrulline's ureido oxygen with sulfur, results in
potent inhibition of the enzyme. EPR studies14 have
Key words: Inhibitors; nitric oxide synthase; tetrahydro-
biopterin; dithiocarbamic acid; dithiocarbamic ester.
*Corresponding author. Tel: 301 435 2229; Fax: 301 402 3521.
0968-0896/98/$Ðsee front matter. Published by Elsevier Science Ltd.
PII: S0968-0896(98)00132-1

1760
M. B. Sassaman et al./Bioorg. Med. Chem. 6 (1998) 1759±1766
shown that l-thiocitrulline may act as a sixth axial
ligand at the heme cofactor by sulfur-iron coordination.
These studies also show that l-arginine perturbs the
heme cofactor, but does not act as a sixth ligand.
are conformationally restricted and possess a chiral
center at C-2.
Synthesis
Several other sulfur-containing compounds are known
to inactivate cytochrome P-450.15 The interaction of
cytochrome P-450 with 1-octanethiol shows a char-
acteristic (UV/VIS) spectral shift attributed to iron-
thiolate ligation, as does the interaction of BH4 de®cient
NOS with dithiothreitol.16 Pyrrolidine dithiocarbamate
(PDTC) inhibits the bacterial lipopolysaccharide (LPS)-
induced synthesis of both NOS and BH4.17 While
PDTC18 and diethylcarbamate19 suppress the induction
of NOS synthesis by blocking the activation of nuclear
factor kB (which, in turn, is stimulated by LPS), the
mechanism appears to involve binding of intracellular
iron.20
The general method of preparing the parent zwitterionic
dithiocarbamic acids is by the reaction of carbon dis-
ul®de with tertiary (aminoalkyl)amines. Thus, the reac-
tion of 2-aminomethyl-1-ethylpyrrolidine (1) with
carbon disul®de in methanol gave N-(1-ethylpyrrolidin-
2-yl)methyldithiocarbamic acid (2) in 90% yield. This
was then alkylated with methyl iodide and treated with
oxalic acid to obtain a 61% yield of S-methyl N-(1-
ethylpyrrolidin-2-yl)methyldithiocarbamate (3) as its
oxalate salt (see Scheme 1).
In order to prepare optically active analogues, l-pro-
lineamide (4) was reductively methylated with para-
formaldehyde and H2 over 10% palladium on carbon
(Scheme 2) to obtain (2S)-1-methylpyrrolidine-2-car-
boxamide (5). Recrystallization from acetone a€orded
an 83% yield as white needles. Reduction of 5 with
LiAlH4, gave the chiral diamine (2S)-2-aminomethyl-1-
methylpyrrolidine (6) in 65% yield after puri®cation by
vacuum distillation. Compound 6 was treated with car-
bon disul®de, methyl iodide, and oxalic acid, in the
same manner as above, to obtain the optically active
dithiocarbamic acid 7 in 95% yield and ester oxalate 8
in 70% yield. The noncyclic conformationally mobile
analogues were prepared according to Scheme 3. Alky-
lation of dimethylamine with 5-bromovaleronitrile (9)
gave 5-dimethylaminopentanenitrile (10), in 98% yield
after distillation. Compound 10 was hydrogenated over
Raney nickel to obtain 5-dimethylaminopentaneamine
(11) in 70% yield and 95% purity after distillation, the
remaining 5% being unreacted nitrile. Treatment of
diamine 11 with carbon disul®de gave a 93% yield of
dithiocarbamic acid 12, which was methylated and
treated with oxalic acid to obtain a 70% yield of 13.
These observations led us to explore the possibility that
other sulfur compounds, particularly functionalized
dithiocarbamates, might inhibit NOS by a mechanism
involving heme, BH4 or both. We began our study with
the synthesis of a number of N-substituted dithio-
carbamic acids and S-methyl esters. Each contained a
four or ®ve carbon linker (chain or ring), terminating in
hydroxyl, carboxyl or tertiary amino group. Biological
activity was assessed by competitive binding and func-
tional studies (see below). Only those compounds con-
taining the tertiary amino moiety showed activity. The
synthesis and results of preliminary screening of the N-
(Tertiary aminoalkyl)dithiocarbamates are reported in
this paper.
Results and Discussion
While salts of dithiocarbamic acids are stable, the par-
ent acids ordinarily are not, decomposing spontaneously
into amines and carbon disul®de. However, when a ter-
tiary amine is present in the same molecule, dithio-
carbamic acids, like amino acids, form stable
zwitterions. Preparation of the requisite ®ve-carbon
diamino compounds, dithiocarbamic acids and esters is
described below. They may be categorized in two
groups: linear compounds, which are free to rotate; and
those derived from 2-aminomethylpyrrolidines, which
Biological evaluation
Compounds 2, 3, 7, 8, 12, and 13 were assayed for bio-
logical activity by two methods: inhibition of [3H]-l-
NG-nitroarginine ([3H]-NOARG) binding, using rat
brain membranes as the source of neuronal NOS
(nNOS), and by functional studies, that is, inhibition of
Scheme 1. Reagents: (a) CS2, MeOH; (b) Me4NOH, MeI, MeOH; (c) oxalic acid, acetone-Et2O.

M. B. Sassaman et al./Bioorg. Med. Chem. 6 (1998) 1759±1766
1761
the conversion of [3H]-l-arginine ([3H]-ARG) to [3H]-l-
S-methyl ester 13, which gave 5.2% and 14% inhibition,
respectively, showed very similar characteristics to the
two sets of conformationally restricted compounds, dif-
fering by only a few percentage points. In each case, the
S-methyl esters showed stronger inhibition than the
zwitterionic parents.
citrulline ([3H]-CIT), which utilized puri®ed enzyme.
Competitive binding
Competitive binding studies were performed by
NOVASCREEN1, according to a published proce-
dure.21 Brie¯y, reactions were carried out in 50 mM
TRIS-HCl (pH 7.4) for 60 min at 25 ꢀC in the presence
of 5 nM [3H]-NOARG and 1 mM test compound. Reac-
tions were terminated by rapid vacuum ®ltration onto
glass ®ber ®lters. Radioactivity trapped onto the ®lters
was determined and compared to control values in order
to ascertain interactions of test compounds with the
enzyme.
Functional studies
Puri®cation of nNOS. Rat brain nNOS was puri®ed
from transfected human kidney 293 cells,22 by mod-
i®cation of the procedure reported by McMillan and co-
workers.23 All bu€ers, with the exception of that used in
the ®nal gel ®ltration step, contained 0.5 mM l-arginine,
5 mM BH4, and 0.5 mM dithiothreitol. The preparation
had greater than 95% purity when evaluated by sodium
dodecyl sulfate-polyacrylamide gel electrophoresis
The results of the competive binding studies are listed in
Table 1 under the heading ``Nitroarginine binding''. All
the compounds showed weak to moderate inhibition of
[3H]NOARG binding at a concentration of 1 mM, ran-
ging from 3.7% for the conformationally restricted
compound 2 to 20.2% for its S-methyl ester, compound
3. The chiral center in the structurally related optically
active dithiocarbamic acid 7, which showed 9.8% inhi-
bition, and its S-methyl ester 8, which showed 15%
inhibition, had no dramatic e€ect on binding. The
conformationally free dithiocarbamic acid 12 and its
(SDS±PAGE) and typically had a speci®c activity of
1
250±400 nmol min
mg 1. The BH4 content of the
puri®ed nNOS was approximately 0.5 mol per mol sub-
unit as determined by high performance liquid chroma-
tography of the oxidized sample.24 Protein was
determined by the bicinchonic acid method25 with the
use of a bovine serum albumin standard.
Determination of nNOS inhibition. nNOS activity was
determined by monitoring the conversion of [3H]-ARG
Scheme 2. Reagents: (a) Ch2O, H2, MeOH, 10% Pd/C; (b) LiAlH4, THF, Et2O; (c) CS2, THF; (d) MeNOH, MeI; (e) oxalic acid,
acetone-Et2O.
Scheme 3. Reagents: Me2NH, MeOH; (b) H2, MeOH satd w/NH3, RaNi; (c) CS2, THF; (d) NaOH, MeI; (e) oxalic acid, acetone-
Et2O.

1762
M. B. Sassaman et al./Bioorg. Med. Chem. 6 (1998) 1759±1766
Table 1. E€ects of dithiocarbamates on the inhibition of binding of nitroarginine and inhibition of nitric oxide synthase activity
determined with or without added (6R)-5,6,7,8-tetrahydro-l-biopterin (BH4)
Dithiocarbamate
Nitroarginine binding
% competitive inhibition
Nitric oxide synthase activity
Minus BH4 (% inhibition)
Plus BH4 (% inhibition)
2
3.7
20.2
9.8
15
39 [10, n=8]
59 [7, n=10]
24 [8, n=6]
40 [4, n=7]
68 [4, n=7]
3 [6, n=6]
33 [14, n=8]
21 [6, n=10]
22 [8, n=6]
29 [16, n=7]
64 [7, n=5]
2 [4, n=4]
3
7
8
12
13
5.2
14
Values for percent inhibition are means with standard deviations and number of determinations, respectively, shown in parentheses.
to [3H]-CIT26 over a 10 min period at 25 ꢀC. Reaction
mixtures were prepared containing 5 mM [3H]-ARG,
0.67 mM NADPH, 0.88 mM calmodulin, 0.9 mM CaCl2,
enzyme activity during the course of the run. The results
of any experiment in which the baseline controls varied
by more than 10% of the initial value were discarded.
1
0.1 mM FAD and FMN, 0.6 mM EDTA, 1 mg mL
catalase, and 33.3 mM HEPES bu€er (pH 7.4). BH4
(10 mM) and dithiocarbamate derivatives (25 mM) or l-
thiocitrulline (1 mM) were added as required (see Table 1
and text). Immediately prior to starting the reaction
with NOS, a stock solution of enzyme was diluted 100-
fold with a solution containing 50 mM Tris±HCl (pH
Control experiments on the inhibition of nNOS activity
by l-thiocitrulline (1 mM) showed no e€ect when exo-
genous BH4 was added, giving a mean value of 43%
inhibition [12, n=28] in the absence of BH4, and 45%
inhibition [14, n=26] in the presence of BH4 (data not in
Table 1).
7.5), 10% (v/v) glycerol, 0.1 mM EDTA, 100 mM NaCl,
1
1 mg mL
bovine serum albumin and the protease
The results of the functional studies are listed in Table 1
under the heading ``Nitric oxide synthase activity''. The
linear, conformationally free dithiocarbamic acid 12,
showed the strongest inhibition, with mean values of
68% in the absence of added BH4 and 64% with added
BH4. In contrast, its S-methyl ester 13 had negligible
ability to inhibit the conversion of [3H]-ARG to [3H]-
CIT, showing only 2% inhibition with added BH4 and
3% in its absence.
inhibitors: 4-(2-aminoethyl)benzenesulfonyl ¯uoride
(0.1 mM) leupeptin (0.5 mM), and pepstatin (0.5 mM).
The reaction was initiated by adding 5 mL (approxi-
mately 0.15 mg) of this dilute nNOS solution to obtain a
®nal volume of 200 mL for the reaction mixture. After
the incubation period, reactions were terminated by the
addition of 0.8 mL of a solution containing 16.7 mM
sodium HEPES (pH 5.5), 1.67 mM EDTA, 8.1 mM l-
arginine and 8.1 mM l-citrulline. [3H]-l-Arginine was
trapped on a column of Dowex 50 Na+, and [3H]-l-
citrulline in the e‚uent quanti®ed by liquid scintillation.
The racemic, conformationally restricted dithiocarbamic
acid 2 and optically active 7, performed equally well
with or without BH4 and did not di€er signi®cantly
from one another in their ability to inhibit the enzyme.
Their mean values for inhibition [2: 39% (minus BH4),
33% (plus BH4); 7: 24% (minus BH4), 22% (plus BH4)]
suggest a slight increase in potency for the racemate.
Typically, each experiment consisted of 18 separate
reaction mixtures designed to measure baseline enzyme
activity (in triplicate), activities in the presence of two
dithiocarbamate derivatives (each in duplicate) and
activities in the presence of the NOS inhibitor l-thio-
citrulline (in duplicate). As mentioned in the previous
section, puri®cation yields nNOS that is stoichiome-
trically de®cient in BH4, but still contains approximately
0.5 mol of endogenous, irreversibly bound cofactor per
mol of nNOS subunit. Each determination was, there-
fore, measured in the de®cient state obtained by pur-
i®cation and with the addition of BH4 (``Minus BH4''
and ``Plus BH4'' in Table 1). In the absence of inhibitor,
nNOS activity was stimulated approximately 25% by
the addition of BH4. Determinations of baseline activ-
ities were distributed at the beginning, middle and end
of each experiment to detect any progressive loss of
As in the case of the competitive binding studies, the
two conformationally restricted S-methyl esters, 3 and
8, were somewhat better at inhibition than their parent
acids in the absence of added BH4. This behavior was
markedly di€erent than for the conformationally free
ester 13, described earlier. One of the most interesting
and unexpected observations was the e€ect that adding
BH4 had on the conformationally restricted esters. In
both cases, their ability to inhibit nNOS activity was
reduced upon the addition of BH4, mean values falling
from 59 to 21% for 3 and from 40 to 29% for 8. A one-
tailed paired t-test on the raw data showed statistical

M. B. Sassaman et al./Bioorg. Med. Chem. 6 (1998) 1759±1766
1763
relevance, giving p<0.01 for 3 with and without BH4
and p<0.03 for 8 with and without BH4.
N-(tertiary aminoalklyl)dithiocarbamic acids and S-
methyl esters show an ability to inhibit nitric oxide syn-
thase activity at the active site of the enzyme by virtue of
their capacity to competitively inhibit the binding of
[3H]-l-nitroarginine. Functional studies, with and with-
out the addition of BH4, show the conformationally
restricted S-methyl esters can also inhibit enzyme activ-
ity at a second site, involving the BH4 cofactor.
Conclusions
The N-(tertiary aminoalkyl)dithiocarbamic acids and
esters in this study appear to have two modes of inhibi-
tion, one involving the site for substrate binding and
one involving the BH4 cofactor.
Experimental
Biology
As mentioned earlier, l-thiocitrulline exerts its inhibi-
tion by coordinating to the heme cofactor at or near the
binding site for l-arginine. Similarly, the strong inhibi-
tion of NOS by NOARG is caused by binding to the
same site as the substrate.27,28 Since all the compounds
were able to inhibit NOARG binding to some extent,
interaction with the l-arginine binding site is implied.
The anticipated manner of interaction is between the
heme iron and one or both sulfurs of the dithiocarba-
mates, but we have no direct evidence of this.
Materials used in the functional studies described earlier
were obtained from the following sources: beef liver
catalase and bovine serum albumin (fraction V, protease
free) from Boehringer Mannheim (Indianapolis, IN,
USA); high purity bovine brain calmodulin from Cal-
biochem (San Diego, CA, USA); (6R)-5,6,7,8-tetrahydro-
l-biopterin from Schircks Laboratories (Jona, Switzer-
land); [2,3,4,5-3H]-l-arginine from Amersham Corp.
(Arlington Heights, IL, USA); FAD from Sigma Chem-
ical Corp. (St. Louis, MO, USA); FMN from Fluka
Chemical Corp. (Ronkonkoma, NY, USA); l-thioci-
trulline from Alexis Corp. (San Diego, CA, USA); and
analytical grade Dowex 50 (AG 50W-X8, 200±400)
mesh from Bio-Rad Laboratories (Hercules, CA, USA).
The results of the functional studies, inhibition of the
conversion of [3H]ARG to [3H]CIT, suggest a second
mode of inhibition. While none of the zwitterionic
dithiocarbamic acids or l-thiocitrulline controls were
a€ected by the addition of BH4, both conformationally
restricted S-methyl esters (3 and 8) showed a substantial
loss of inhibitory activity when the cofactor was added.
It is likely, therefore, that a second mode of inhibition
involves the binding site for BH4 or interaction with the
cofactor itself. Dual modes of inhibition of NOS invol-
ving the binding sites for l-arginine and for BH4 are not
without precedent. 1-(2-Tri¯uoromethylphenyl)imidazole
was recently reported to exert its inhibition by a similar
process;29 and 7-nitroindazole appears to bind at a dis-
tinct site which allosterically interacts with the binding
domains for both l-arginine and BH4.30
Chemistry
THF and Et2O used for reaction solvents were freshly
distilled from sodium/benzophenone ketyl. All other
chemical reagents and solvents were purchased from
Aldrich and used without further puri®cation. 13C and
1H NMR data were obtained at 50 and 200 MHz,
respectively. Chemical shifts are reported in ppm (d)
down®eld of tetramethylsilane and coupling constants
are given in Hertz. Mass spectra were obtained by GC/
MS analysis, using a 12.0 meter HP-5 capillary (5%
diphenyl/95% dimethylpolysiloxane) or by direct injec-
tion probe (DIP) for zwitterionic compounds. Micro-
analyses were performed by Galbraith Laboratories.
The divergence of behavior of the linear, con-
formationally free dithiocarbamic acid±ester pair, com-
pounds 12 and 13, is dramatic. While acid 12, which
possesses the highest mean values for inhibition, is
similar in its inhibitory e€ect to the conformationally
restricted acids 2 and 7, and is likewise una€ected by the
addition of BH4, its ester (13) shows only minimal abil-
ity to inhibit the enzyme. The inhibition of [3H]-
NOARG binding by 13 is comparable to that of the
conformationally restricted esters 3 and 8. Therefore,
the relatively poor inhibition of enzyme activity shown
by 13, may imply a size limitation involving the BH4
binding site, the conformationally restricted, more
compact pyrrolidines having more accessibility.
N-(1-Ethylpyrrolidin-2-yl)methyldithiocarbamic acid (2).
To a rapidly stirring solution of 1-(aminomethyl)-1-
ethylpyrrolidine31 (1) (1.28 g, 10.0 mmol) in 10 mL
MeOH was added carbon disul®de (1.2 mL, 20 mmol).
An exothermic reaction gave a heavy pale yellow pre-
cipitate within several seconds. After cooling to room
temperature, 10 mL Et2O was added, the precipitate
collected, rinsed with Et2O and air-dried to obtain 1.85 g
(90%) of the zwitterionic title compound as a pale yellow
powder, mp 156±157 ꢀC (d); 1H NMR (MeOH-d4+
NaOD/D2O) d 1.15 (t, J=7.2, 3H), 1.59±2.04 (m, 4H),
2.16±2.40 (m, 2H), 2.72±3.19 (m, 3H), 3.40 (d, J=7.1 of
d, J=13.4, 1H), 3.92 (d, J=4.0, of d, J=13.4, 1H), 5.78
It is dicult to draw meaningful conclusions concerning
structure activity relationships without additional data.
However, a trend is evident from this study. The six

1764
M. B. Sassaman et al./Bioorg. Med. Chem. 6 (1998) 1759±1766
(s, 1H); 13C NMR (MeOH-d4+NaOD/D2O) d 13.9, 23.4,
29.5, 49.6, 51.3, 54.4, 63.5, 214.0; EIMS (DIP) m/z 205
(MH+), 98 (base). Anal. calcd for C8H16N2S2: C, 47.02;
H, 7.89; N, 13.71; found: C, 47.38; H, 8.11; N, 13.55.
THF, was added to 20 mL dry Et2O in a 250 mL 3-neck
¯ask equipped with condenser, magnetic stir-bar, addi-
tion funnel and inlet for argon. Compound 5 (1.28 g,
10.0 mmol) in 50 mL THF was added dropwise over
30 min. On completion, the slightly cloudy reaction
solution was stirred at room temperature 60 min, then at
re¯ux 18 h. The resulting opaque milky-white solution
was cooled to room temperature and quenched by the
careful addition of 2 mL satd Na2SO4. Inorganic salts
were removed by ®ltration and rinsed liberally with
Et2O. Solvent was removed under reduced pressure and
the residue distilled at 55±57 ꢀC (20 mbar) to obtain
0.74 g (65%) of the title compound. 1H NMR (CDCl3) d
1.44 (s, 2H), 1.54±1.99 (m, 4H), 2.12±2.28 (m, 2H), 2.32
(s, 3H), 2.63±2.81 (m, 2H), 3.01±3.10 (m, 1H); 13C
NMR (CDCl3) d 22.2, 27.9, 40.5, 43.7, 57.2, 67.2; EIMS
S-Methyl N-(1-ethylpyrrolidin-2-yl)methyldithiocarbamate
oxalate (3). Compound 2 (716 mg, 3.50 mmol) and
tetramethylammonium hydroxide pentahydrate (634 mg,
3.50 mmol) were combined in 14 mL MeOH to give a
bright yellow solution. Methyl iodide (240 mL,
3.8 mmol) was added all at once via syringe, instant-
aneously producing
a white precipitate of tetra-
methylammonium iodide. After stirring 10 min, the
mixture was diluted with 14 mL Et2O, the precipitate
removed by ®ltration and rinsed with 25 mL Et2O, and
the combined ®ltrate and rinse evaporated under
reduced pressure. The residue was redissolved in 20 mL
Et2O, washed with H2O (4Â10 mL), brine (1Â10 mL)
and dried over Na2SO4/K2CO3. This was ®ltered and
allowed to slowly drip into a solution of oxalic acid
(315 mg, 3.50 mmol) in 20 mL acetone. The resulting
mixture was warmed slightly to dispel cloudiness and
allowed to stand at room temperature several hours to
obtain 661 mg (61%) of the product as a crystalline
solid. For analytical purposes, a small portion was
recrystallized from boiling acetone, mp 111±113 ꢀC (d);
1H NMR (DMSO-d6) d 1.21 (pseudo t, 3H), 1.71±2.19
(m, 4H), 2.54 (s, 3H overlapping solvent signal), 2.94±
3.35 (m, 4H), 3.47±4.09 (m, 4H), 8.96 (bs, 2H), 10.50
(bs, 1H); 13C NMR (DMSO-d6) d 10.3, 17.4, 21.7, 27.6,
46.1, 48.6, 52.8, 63.4, 165.0, 199.1; EIMS (free base) m/z
170 (M+ HSCH3), 98 (base). Anal. calcd for
C11H20N2O4S2: C, 42.84; H, 6.54; N, 9.08; found: C,
42.86; H, 6.66; N, 9.03.
m/z 115 (MH+), 114 (M+), 84 (base); [a]21 2.04 (c 2.6,
d
MeOH). For microanalysis, the dihydrochloride was
prepared by adding excess ethereal HCl to a methanolic
solution of free base. Anal. calcd for C6H16ClN2: C,
38.51; H, 8.62; N, 14.93; found: C, 38.53; H, 8.62; N,
14.89.
N-[(2S)-1-Methylpyrrolidin-2-yl]methyldithiocarbamic
acid (7). To a solution of compound 6 (509 mg,
4.43 mmol) in 4.5 mL MeOH was added carbon dis-
ul®de (0.5 mL, 8.3 mmol) via syringe. Precipitation of
the product began within several seconds. After stirring
1 min, 4.5 mL THF was added and the zwitterionic
product collected, rinsed with Et2O and air-dried to
obtain 804 mg (95%) as a white salt, mp 165±166 ꢀC (d);
1H NMR (MeOH-d4+NaOD/D2O) d 1.65±2.19 (m,
4H), 2.22±2.32 (m, 1H), 2.42 (s, 3H), 2.62 (m, 1H), 3.09
(m, 1H), 3.39±3.52 (m, 2H), 3.98 (d, J=3.9 of d,
J=13.4, 1H), 6.43 (s, 1H); 13C NMR (MeOH-
d4+NaOD/D2O) d 23.3, 29.7, 41.4, 50.9, 58.1, 65.3,
214.5, EIMS (DIP) m/z 190 (M+), 84 (base);
(2S)-1-Methylpyrrolidine-2-carboxamide32 (5). l-Proline-
amide (4) (1015 mg, 8.89 mmol) and paraformaldehyde
(293 mg, 9.76 mmol) were dissolved in 18 mL MeOH. To
this was added 10% Pd/C (96 mg, 0.090 mmol) and the
mixture hydrogenated 14 h at 60 psig. The catalyst was
removed by ®ltration (Whatman #5 ®lter paper), rinsed
with 10 mL MeOH, and the ®ltrate evaporated under
reduced pressure to obtain a white powder. Recrystalli-
zation from a minimal volume of boiling acetone fol-
lowed by chilling to dry ice temperature gave 945 mg
(83%) of the title compound as white needles, mp 140±
141 ꢀC; 1H NMR (DMSO-d6) d 1.60±1.77 (m, 3H), 1.97±
2.26 (m [containing 2.26 (s)] 5H), 2.49±2.62 (m, 1H),
2.95±3.04 (m, 1H), 7.00 (bs, 1H), 7.11 (bs, 1H); 13C
NMR (DMSO-d6) d 23.4, 30.3, 41.2, 56.2, 69.1, 175.6;
[a]21 1.62 (c 2.0, 50% MeOH/50% 1 N NaOH). Anal.
d
calcd for C7H14N2S2: C, 44.18; H, 7.41; N, 14.72; found:
C, 44.08; H, 7.45; N, 14.55.
S-Methyl N-[(2S)-1-methylpyrrolidin-2-yl]methyldithio-
carbamate oxalate (8). Following the same procedure
as for compound 3, 717 mg (70%) product was
obtained, in three crops, as a white powder, mp 100±
101 ꢀC; 1H NMR (DMSO-d6) d 1.70±2.18 (2 over-
lapping m's, 4H), 2.54 (s, 3H), 2.78 (s, 3H), 2.93±3.06
(m, 1H), 3.46±3.61 (m, 2H), 3.80±3.90 (m, 1H), 4.04±
4.13 (m, 1H), 10.49 (bs, 1H), 11.30 (bs, 1H); 13C NMR
(DMSO-d6) d 17.7, 21.7, 28.0, 39.6, 46.1, 55.9, 65.2,
165.5, 199.2; [a]23d+0.18 (c 1.5,MeOH). Anal. calcd for
C10H18N2O4S2: C, 40.80; H, 6.16; N, 9.52; found: C,
41.18; H, 6.41; N, 9.74.
EIMS m/z 128 (M+), 84 (base); [a]21 0.83 (c 0.75,
d
MeOH). Anal. calcd for C6H12N2O: C, 56.22; H, 9.44;
N, 21.86; found: C, 56.08; H, 9.29; N, 21.50.
(2S)-2-Aminomethyl-1-methylpyrrolidine (6).
commercial solution of LiAlH4 (20 mL, 20 mmol) in
A
1.0 M
5-Dimethylaminopentanenitrile33 (10). 5-Bromovalero-
nitrile (1.62 g, 10.0 mmol) and 40% aq dimethylamine

M. B. Sassaman et al./Bioorg. Med. Chem. 6 (1998) 1759±1766
1765
(3.8 mL, 30 mmol) were dissolved in 20 mL THF and
re¯uxed 120 min under argon. After cooling to room
temperature, K2CO3 (2.76 g, 20.0 mmol) was added,
followed by 5 mL H2O. Stirring 10±15 min gave two
clear phases, which were separated and the aqueous
portion extracted with Et2O (2Â20 mL). The combined
organic portions were dried over Na2SO4/K2CO3, the
volatiles removed under reduced pressure, and the resi-
due distilled from anhydrous K2CO3 at 50±51 ꢀC (0.8
torr) to obtain 1.24 g (98%) of the title compound as a
colorless liquid; 1H NMR (CDCl3) d 1.43 (bm, 4H),
1.98±2.18 (m, 10H); 13C NMR (CDCl3) d 16.4, 22.8,
26.0, 44.9, 58.0, 119.1; EIMS m/z 127 (MH+), 58 (base).
aqueous portion was extracted with Et2O (2Â5 mL) and
the combined organic portions dried over Na2SO4/
K2CO3. This was ®ltered and allowed to drip into a
solution of oxalic acid (90 mg, 1.00 mmol) in 2 mL acet-
one and the resulting solution stored at 20 ꢀC over-
night. The product, 217 mg (70%), was obtained as a
white salt, mp 100±101 ꢀC; 1H NMR [@ 60 ꢀC (DMSO-
d6)] d 1.27±1.38 (m, 2H), 1.54±1.68 (m, 4H), 2.51 (s, 3H),
2.67 (s, 6H), 2.87±2.95 (m, 2H), 3.58±3.61 (m, 2H), 9.99
(bs, 1H), 10.90 (vbs, 2H); 13C NMR [@ 60 ꢀC (DMSO-
d6)] d 17.1, 23.2, 23.3, 26.8, 42.0, 46.2, 56.4, 165.0, 197.1;
EIMS (free base) m/z 172 (M+ HSCH3), 58 (base).
Anal. calcd for C11H22N2O4S2: C, 42.56; H, 7.14; N,
9.02; found: C, 42.90; H, 7.30; N, 8.98.
5-Dimethylaminopentaneamine33 (11). In a 60 mL Parr
bottle, compound 10 (1.63 g, 12.9 mmol) was dissolved
in 20 mL MeOH and a 50% aqueous slurry of W-2
Raney nickel (580 mg, 4.9 mmol) added. The mixture
was packed in dry ice and saturated with anhydrous
NH3 by introducing a slow stream of bubbles over
15 min. After warming to room temperature, the bottle
was sealed and hydrogenated 18 h at 65 psig. Catalyst
was removed by ®ltration and the ®ltrate evaporated
under reduced pressure using added acetonitrile to
remove residual water as an azeotrope. Distillation of
the residue at 51±52 ꢀC (0.6 torr) gave 1.18 g (70%)
product as a clear, colorless liquid. Analysis by GC/MS
showed 95% purity, the remaining 5% being unreacted
starting material. 1H NMR (CDCl3) d 1.25±1.55 (m,
6H), 1.70 (bs, 2H), 2.21±2.29 (overlapping s and t, 8H),
2.69 (pseudo t, 2H); 13C NMR (CDCl3) d 24.8, 27.5,
33.6, 42.1, 45.4, 59.8; EIMS m/z 130 (M+), 58 (base).
Acknowledgements
The authors gratefully acknowledge the NIMH/
NOVASCREEN1 Psychotherapeutic Drug Discovery
and Development Program (Contract #NO1MH20003)
for the in vitro assays of [3H]-NOARG binding and Mr.
K. Park for assisting in the NOS inhibition assays.
References and Notes
1. Grith, O. W.; Stuehr, D. J. Ann. Rev. Physiol. 1995, 57,
707.
2. Stuehr, D. J. Ann. Rev. Pharmacol. Tox. 1997, 37, 339.
3. Mayer, B.; Werner, E. R. Naunyn-Schmiedeberg's Arch.
Pharmacol. 1995, 351, 453.
4. Gorren, A. C. F.; List, B. M.; Schrammel, A.; Pitters, E.;
Hemmens, B.; Werner, E. R.; Schmidt, K.; Mayer, B. Bio-
chemistry 1996, 35, 16735.
N-(5-Dimethylaminopentan-1-yl)dithiocarbamic acid (12).
Compound 5 (1.10 g, 8.45 mmol) was dissolved in
8.5 mL dry THF and carbon disul®de (1.0 mL, 17 mmol)
was added all at once. The resulting exothermic reaction
immediately produced a white precipitate, which was
triturated with a small volume of hot MeOH, then col-
lected, rinsed with Et2O and air-dried to obtain 1.63 g
(93%) of the zwitterionic product, mp 157±159 ꢀC (d);
1H NMR (MeOH-d4+NaOD/D2O) d 1.37±1.74 (m,
6H), 2.24±2.36 (overlapping m and s, 9H), 3.58 (t,
J=7.1, 2H), 6.17 (s, 1H); 13C NMR (MeOH-
d4+NaOD/D2O) d 25.8, 27.8, 29.4, 45.5, 48.7, 60.5,
213.4, EIMS (DIP) m/z 206 (M+), 58 (base). Anal. calcd
for C8H18N2S2: C, 46.56; H, 8.79; N, 13.57; found: C,
46.60; H, 9.04; N, 13.43.
5. Griscavage, J. M.; Fukuto, J. M.; Komori, Y.; Ignarro, L. J.
J. Biol. Chem. 1994, 269, 21644.
6. Werner, E. R.; Pitters, E.; Schmidt, K.; Wachter, H.; Wer-
ner-Felmayer, G.; Mayer, B. Biochem. J. 1996, 320, 193.
7. Pfei€er, S.; Gorren, A. C. F.; Pitters, E.; Schmidt, K.; Wer-
ner, E. R.; Mayer, B. Biochem. J. 1997, 328, 349.
8. Presta, A.; Siddhanta, U.; Wu, C.; Sennequier, N.; Huang,
L.; Abu-Soud, H. M.; Ersurum, S.; Stuehr, D. J. Biochemistry
1998, 37, 298.
9. Masters, B. S. S.; McMillan, K.; Sheta, E. A.; Nishimura,
J. S.; Roman, L. J.; Martasek, P. FASEB J. 1996, 10, 552.
10. Chen, P.-F.; Tsai, A. L.; Wu, K. K. Biochem. Biophys. Res.
Commun. 1995, 215, 1119.
11. Southan, G. J.; Szabo, C. Biochem. Pharmacol. 1996, 51, 383.
12. Moore, W. M.; Webber, R. K.; Fok, K. F.; Jerome, G. M.;
Connor, J. R.; Manning, P. T.; Wyatt, P. S.; Misko, T. P.;
Tjoeng, F. S.; Currie, M. G. J. Med. Chem. 1996, 39, 669.
S-Methyl N-(5-dimethylaminopentan-1-yl)dithiocarbamate
(13). Compound 12 (206 mg, 1.00 mmol) was sus-
pended in 1 mL MeOH and neutralized with 1 mL 1 N
NaOH. Methyl iodide (65 mL, 1.0 mmol) was added all
at once (via syringe) to the resulting solution. After
stirring 2 h at room temperature, the reaction mixture
was diluted with 5±6 mL Et2O and separated. The
13. Frey, C.; Narayanan, K.; McMillan, K.; Spack, L.; Gross,
S. G.; Masters, B. S.; Grith, O. W. J. Biol. Chem. 1994, 269,
26083.
14. Salerno, J. C.; Frey, C.; McMillan, K.; Williams, R. F.;
Masters, B. S. S.; Grith, O. W. J. Biol. Chem. 1995, 270,
27423.

1766
M. B. Sassaman et al./Bioorg. Med. Chem. 6 (1998) 1759±1766
15. De Matteis, F. Mol. Pharmacol. 1974, 10, 849.
24. Fukushima, T.; Nixon, J. C. Anal. Biochem. 1980, 102, 176.
16. Gorren, A. C. P.; Schrammel, A.; Schmidt, K.; Mayer, B.
Biochemistry 1997, 36, 4360.
25. For details of this method, refer to Pierce BCA Application
Note #11. The Sorensen Method: Use of BCA for determination
of protein and reducing agents. Use of BCA to locate enzyme
conjugates. Product #23225, Pierce, P.O. Box 117, Rockford,
IL 61105±9976.
17. Hattori, Y.; Nakanishi, N.; Kasai, K.; Shimoda, S. I.;
Gross, S. S. Eur. J. Pharmacol. 1996, 296, 107.
18. Sherman, M. P.; Aeberhard, E. E.; Wong, V. Z.; Grisca-
vage, J. M.; Ignarro, L. J. Biochem. Biophys. Res. Commun.
1993, 191, 1302.
26. Bredt, D. S.; Snyder, S. H. Proc. Natl. Acad. Sci. U.S.A.
1990, 87, 682.
27. Fur®ne, E. S.; Harmon, M. F.; Paith, J. E.; Garvey, E. P.
Biochemistry 1993, 32, 8512.
19. Mulsch, A.; Schray-Utz, B.; Mordvintec, P. I.; Hauschildt,
S.; Busse, R. FEBS Lett. 1993, 321, 215.
28. Klatt, P.; Schmidt, K.; Brunner, F.; Mayer, B. J. Biol.
Chem. 1994, 269, 1674.
20. Schreck, R.; Meier, B.; Mannel, D. N.; Droge, W.;
Baeuerle, P. A. J. Exp. Med. 1992, 175, 1181.
29. Handy, R. L. C.; Moore, P. K. Life Sci. 1997, 60, PL389.
21. Michel, A. D.; Phul, R. K.; Stewart, T. L.; Humphrey,
P. P. A. Br. J. Pharmacol. 1993, 109, 287.
30. Mayer, B. In Nitric Oxide in the Nervous System; Vincent,
S. R., Ed.; Academic: New York, 1995; pp 21±42.
22. Bredt, D. S.; Hwang, P. M.; Glatt, C. E.; Lowenstein, C.;
Reed, R. R.; Snyder, S. H. Nature 1991, 351, 714.
31. Reitsema, R. H. J. Am. Chem. Soc. 1949, 71, 2041.
32. For the racemic compound, see: Renshaw, R. R.; Cass,
W. E. J. Am. Chem. Soc. 1939, 61, 1195.
23. McMillan, K.; Bredt, D. S.; Hirsch, D. J.; Snyder, S. H.;
Clark, J. E.; Masters, B. S. S. Proc. Natl. Acad. Sci. U.S.A.
1992, 89, 11141.
33. Stewart, J. M. J. Am. Chem. Soc. 1954, 76, 3228.