Selective inhibitors of monoamine oxidase (MAO). 5.1,2 1-Substituted phenoxathiin inhibitors containing no nitrogen that inhibit MAO a by binding it to a hydrophobic site

Article abstract of DOI:10.1021/jm970862j

It is believed that a monoamine oxidase (MAO) inhibitor specific for MAO A, which is reversibly bound to this enzyme and displaceable by tyramine, will be an antidepressant which will not cause a rise in blood pressure when tyramine-containing foods are ingested. Some linear tricyclic compounds with a larger and a smaller group forming the central ring and with a lipophilic group ortho to the larger group (here mostly the SO₂ function of phenoxathiin 10,10-dioxide) are reported to have the sought properties. Potency appears to require short length and relatively small cross section for the substituent. The 1-ethyl (13), 1-vinyl (22), 1-trifluoromethyl (27), and 1-iodo (76) phenoxathiin dioxides had the best profiles. Structure- activity relationships, syntheses, and a possible rationale for the selectivity of these compounds and related tricyclics are given. Compound 13 was selected for further development. A summary of pharmacological data for 13 is given.

Full text of DOI:10.1021/jm970862j

2118
J . Med. Chem. 1998, 41, 2118-2125
Selective In h ibitor s of Mon oa m in e Oxid a se (MAO). 5.1,2 1-Su bstitu ted
P h en oxa th iin In h ibitor s Con ta in in g No Nitr ogen Th a t In h ibit MAO A by
Bin d in g It to a Hyd r op h obic Site
Morton Harfenist,*,† Daniel P. C. McGee,† Mark D. Reeves,† and Helen L. White§,‡
Divisions of Organic Chemistry and Pharmacology, The Wellcome Research Laboratories,
Research Triangle Park, North Carolina 27709
Received December 29, 1997
It is believed that a monoamine oxidase (MAO) inhibitor specific for MAO A, which is reversibly
bound to this enzyme and displaceable by tyramine, will be an antidepressant which will not
cause a rise in blood pressure when tyramine-containing foods are ingested. Some linear
tricyclic compounds with a larger and a smaller group forming the central ring and with a
lipophilic group ortho to the larger group (here mostly the SO2 function of phenoxathiin 10,-
10-dioxide) are reported to have the sought properties. Potency appears to require short length
and relatively small cross section for the substituent. The 1-ethyl (13), 1-vinyl (22),
1-trifluoromethyl (27), and 1-iodo (76) phenoxathiin dioxides had the best profiles. Structure-
activity relationships, syntheses, and a possible rationale for the selectivity of these compounds
and related tricyclics are given. Compound 13 was selected for further development. A
summary of pharmacological data for 13 is given.
Ch a r t 1
MAO (EC 1.4.3.4, amine oxidase, flavin-containing)
consists of two differing forms distinguishable by their
substrate specificity3 and their amino acid sequences.4
Serotonin (5-HT) is specifically deaminated by MAO A,
while 2-phenethylamine is relatively specifically deami-
nated by MAO B; these properties are used for analysis
of the enzymes. Tyrosine is deaminated by both forms
with similar efficiencies. Early MAO inhibitors were
found to have clinically significant antidepressant and
antiphobic properties. However ingestion especially of
tyramine-containing foods by patients taking early MAO
inhibitors led to a significant and sometimes serious
increase in blood pressure (“cheese effect”), apparently
due to release of norepinephrine by displacement by
undestroyed tyramine and consequent vasoconstriction.
Therefore clinical use of these early MAO inhibitors
became extremely limited.
More recently, it has been found that inhibition of
MAO A (MAO A-I) gives clinically useful antidepressant
activity, while MAO B-I does not.5 An MAO inhibitor
that selectively inhibits MAO A should then have
antidepressant activity while leaving the MAO B to
destroy ingested tyramine, thereby minimizing the
possibility of a clinical “cheese effect.” Inhibitors such
as moclobemide and brofaromine are reported to have
such specificity.18 Further, since the human intestinal
wall, especially the duodenal mucosa, is reported6 to
have MAO A as the predominant form of its MAO, an
important additional safety feature would be to have
the MAO A inhibitor displaceable from the inhibited
enzyme by any ingested tyramine, thus reinstating the
first line of defense against dietary tyramine. Such an
inhibitor, selective for MAO A inhibition, reversibly
bound to the enzyme, and displaceable by tyramine, was
the target of our research.
We have reported in previous publications,1,7 and
earlier work cited there, the high potency and selectivity
for inhibition of MAO A of a variety of tricyclic hetero-
cyclics characterized by specific structural features: (1)
The tricyclic structure included a central ring made up
of a larger and a smaller moiety at least one of which
was a heteroatom. In the few examples examined in
which the central moieties were identical, loss of
specificity for MAO A was seen. (2) A hydrophilic group
was present on one outer ring para to the smaller
central function. (3) In most potent compounds, a single
pendant methyl group had to be present on the second
or third element of this hydrophilic group.
Unfortunately, despite the lack of significant acute
or chronic toxicity in animals, when the best compound
resulting from the previous work (Chart 1; S ) carbonyl,
L ) SO2, and R ) CONHMe, compound 8 of ref 1) was
evaluated in humans, it showed unacceptable toxicity
at high dosage. This compound was also found to have
its carbonyl group almost completely reduced to the
corresponding secondary alcohol function by metabolism
in humans, making the thioxanthen-9-one 10,10-dioxide
system seem a poor choice for further work even though
the reduction appeared to be reversible in vivo. How-
ever, the desired combination of potency, selectivity for
MAO A inhibition, and lack of tyramine-induced blood
pressure rise found8 for this compound and the realiza-
tion that the obvious hydrophilic substituents in the
* Address for correspondence: 6406C Farrington Rd, Chapel Hill,
NC 27514.
† Division of Organic Chemistry.
§ Division of Pharmacology.
‡ Deceased.
S0022-2623(97)00862-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 05/13/1998

Selective Inhibitors of Monoamine Oxidase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 12 2119
Ta ble 1. MAO Inhibition by Phenoxathiins Monosubstituted on an Aromatic Ring
IC50 (µM) or % inhib at µM
no.
1b
substituent
n
MAO A
MAO B
formula
C13H10OS
anal.
mp (°C)
recrysta
1-Me + 3-Me
0
1
2
0
1
2
0
1
2
2
0
1
2
2
0
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
0.61
9% @ <10
0% @ 10
6% @ 0.1
12% @ 1
8% @ 1
9% @ 1
13% @ 1
7% @ 1
11% @ 1
6% @ 1
0% @ 0.1
7% @ 1
0% @ 0.1
0% @ 0.01
8% @ 1
CHN
CHS
CHS
64-66
2
1-Me
1-Me
2-Me
2-Me
2-Me
3-Me
3-Me
3-Me
4-Me
1-Et
1-Et
1-Et
2-Et
3-Et
3-Et
3-Et
1
C
C
13H10O2S
13H10O3S
140-142
EA-H
A
3
0.03
155-156.4
4c
15% @ 1
11% @ 1
30% @ 1
23% @ 1
7% @ 1
0.2
5d
6c
7c
8
C
13H10O2S
CHS
123.5-124.0
107-109
102-103; 114-116 EA-H
77-80
A
9c
10c
11
12
13e
14
15
16
17
18
19
20
21
22e
23
24
25
26
27
28
29
30
31
32
33e
34
35
36f
37
38
39
40
41
42e
43
44
45g
0.8
15% @ 0.1
64% @ 1
0.07
0.007
0.3
C
C
14H12OS
14H12O2S
CHS
CHS
CHS
CHS
CH
CHS
CHS
CHS
CHS
CHS
CHS
CHS
CHS
CHS
CHS
CH
liq
from oil
C14H12O3S
C
C
C
C
C
C
14H12O3S
14H12OS
14H12O2S
14H12O3S
14H12O3S
15H14O3S
M
oil
0.3
0.006
2% @ 1
91-93
EA-H
EA-H
EA-H
EA-P
H
EA-P
EA-P
EA-P
EA-P
EA-P
A
EA-P
EA-H
EA-P
A-EA
EA-H
A
10% @ 0.1
9% @ 0.1
16% @ 0.1
4% @1
0% @ 0.1
0% @ 0.1
15% @1
0% @ 0.1
0% @ 1
0% @ 0.1
0% @ 0.1
5% @ 0.1
0% @ 0.1
0% @1
86-88
4-Et
1-Pr
1
85-88
67% @ 0.1
42% @1
0% @ 0.1
0.04
62% @1
0.1
20% @ 1
19% @ 0.1
0.08
8% @ 0.1
5% @ 0.1
24% @1
66% @1
0.14
0.4
2.5
0.25
26% @ 1
ca. 2
0% @ 0.1
0% @ 0.1
0% @ 0.1
40% @ 1
0.02
143-144.5
44-46
1-C5H11
C17H18O3S
C15H14O3S‚0.25H2O
C14H10O3S
1-CHMe2
1-CHdCH2
1-CMedCH2
1-CtCH
1-CH2C6H4Cl(m)
1-CHFMe
1-CF3
113-115
143-145
136-138
180-181 dec
104-106
129-131
174-176
137-139
152-154
226-228
87-88
C15H12O3S
C
14H8O3S
C19H13O3ClS
14H11O3FS
C13H7O3F3S
C
CHS
2-CF3
1-C2F5
C13H7O3F3S‚0.25H2O CHS
C14H7O3F5S
C17H19O3NS‚HCl
C13H9O3N3S
C13H10O4S
CHS
CHS
CHNS
CHNS
CHS
1-CH2NEt2
1-CH2N3
1-CH2OH
(+-)-1-CHMeOH
(+)-1-CHMeOH
(-)-1-CHMeOH
2-CHMeOH
1-CHMe(OMe)
1-CHMe(OAc)
1-CHEtOH
1-CMe2OH
1-CHOHC6H4Cl(m)
1-CH2CH2OH
1-CH(OH)CH2NO2
(+-)-1-CHOHCH2OH
(+)-1-CHOHCH2OH
0% @1
8% @ 1
0% @ 1
0%
142-144
177-179
113-115
114-116
C14H12O4S
HOAc
C
C
14H12O4S
14H12O4S
7% @ 3
14% @ 1
0% @ 1
0% @ 0.1
0% @ 0.1
0% @ 0.1
0% @ 1
7% @ 0.1
2% @ 0.1
30
C14H12O4S
C
C
C
15H14O4S
16H16O5S
15H14O4S
CHS
CH
140-141
141.5
93-95
EA-H
EA-H
EA-P
EA-H
A
EA-P
EA-H
EA-H
see text
see text
A
CHS
CHS
CHS
CHS
CHNS
CHS
C15H14O4S
C19H13O4SCl
C14H12O4S
C14H11NO6S
C14H12O5S
C14H12O5S
C14H12O5S
C17H16O5S
129-131
144-146
85-87
160-162
125-127
137-139
135-137
105-109
0.2
1.2
2.0
30
45Ag (-)-1-CHOHCH2OH
46
CHS
CHS
0% @ 0.1
0% @ 0.1
47
48
49f
50
51
52
53h
54
55
56
57
58
59
60i
61j
62k
63j
64
1-CHO
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
0
2
2
45% @ 0.1
ca. 2
1.0
0% @ 0.1
0% @ 1
5% @ 1
0% @1
5% @ 1
C
C
13H8O4S
14H10O4S
CHS
CHS
163-165
142-144
EA-H
1-C(O)Me
2-C(O)Me
1-C(O)CF3
1-C(O)CH2OAc
1-C(O)COOMe
1-COOH
1-CH2COOH
1-C(dNH)C2H4NEt2‚HCl
1-CHFMe
1-CH2PO(OEt)2
1-CH2Br
1-NH2
2-NO2
1-OH
2-OH
1-OMe
2-OMe
A
37% @1
14% @ 1
0.6
13% @ 1
22% @10
0% @ 0.1
19% @ 0.1
0% @ 0.1
0.1
58% @ 0.1
80% @1
11% @ 0.1
58% @ 1
60% @ 0.1
0.2
C14H7O4F3S
C16H12O6S
C
CHS
CHS
CHS
129-131
139-142
204-206
EA-H
A
EA
0% @ 1
5% @ 1
15H1006S
0% @10
0% @ 0.1
0% @ 0.1
0% @ 0.1
0% @ 0.1
0% @ 0.1
8% @1
0% @ 0.1
7% @ 1
0% @ 0.1
10% @ 1
C14H10O5S
C19H22N2O3S‚HCl
C
C17H19O6PS
C13H9BrO3S
C12H9NO3S
CHS
CHN
CHS
CHS
CHSBr 142-144
CHN
221-223
150-152
129-131
91-93
M-W
Ac-EA
A
EP-H
EA-H
A
14H11FO3S
166-167.5
C12H8O4S‚H2O
C
13H10O4S
CHS
125-127
EA-H

2120 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 12
Harfenist et al.
Ta ble 1 (Continued)
IC50 (µM) or % inhib at µM
no.
substituent
n
MAO A
MAO B
formula
C14H12O4S
14H12O4S
anal.
mp (°C)
recrysta
65
66
67
68
69
70
71
72
73
74
75
76
1-OEt
2-OEt
2
2
2
2
2
2
2
2
2
2
2
2
13% @ 0.1
0.04
27% @1
10% @ 0.1
0.5
0% @ 0.1
31% @ 0.1
0% @ 0.1
0% @ 0.1
0.06
0% @ 0.1
2% @ 0.1
0% @1
CHS
CHS
CHN
CHS
CHS
CH
CHS
CHS
CHS
CHBrS
CH
145-147
123-125
211-213
149-151
125-127
118-119
197-199
141-143
167-169
207-208.5
195-197
129-131
EA-P
EA-P
A-EA
EA-P
EA-H
EA-P
HOAc
EA-H
HOAc-W
EA-H
EA-H
A
C
1-OC2H4NMe2
1-OAc
2-OAc
1-SiMe3
1-SMe
1-SPh
1-SO2Ph
1-Br
3-Br
C16H17NO4S‚HCl‚0.5H2O
0% @ 0.1
19% @ 1
0% @ 0.1
0% @ 0.1
5% @ 0.1
6% @ 0.1
0% @ 0.1
4% @ 0.1
0% @ 0.1
C
C
14H10O5S
14H10O5S
C15H16O3SSi
C
C
13H10O3S2
18H12O3S2
C18H12O5S2
C
C
C
12H7O3SBr
12H7O3SBr
12H7O3SI‚0.5H2O
0.02
0.05
1-I
CHS
a
Recrystallization solvents: A ) EtOH, C ) CHCl3, E ) Et2O, EA ) EtOAc, H ) hexanes, HOAc ) acetic acid, TEP ) triethyl phosphate.
b
An unseparated mixture by NMR of 30% 1-methylphenoxathiin and 70% 3-methyl isomer. See Results and Discussion section of the
text. c Suter, C. M.; Green, F. O. Phenoxathin. II. Extension of the Ferrario Reaction. J . Am. Chem. Soc. 1937, 59, 2578-2580. Gruenhagen,
R. H. U.S. Patent 3,071,506, J an. 1, 1963. e See refs 26 and 25. f Paget, C. J .; Dennis, E. M.; Nelson, J .; DeLong, D. C. Antiviral Phenoxathiins
and Their Analogues. Study of the Structure-Activity Relationships for Antiviral Activity and the Replacement Ability in Poliovirus
d
g
Type III Dependent Variants. J . Med. Chem. 1970, 13, 620-623. Each enantiomer had TLC, mass spectrum, and NMR identical to the
racemic 44 and was prepared by chiral chromatography from 44, so elemental analysis was not done. See Results and Discussion section
of text. Reference 22. Krishna, S. Synthesis of Derivatives of Phenothioxin. J . Chem. Soc. 1923, 123, 2872-2876. j Kemp, D. S.; Buckler,
D. R. Synthesis of 1-Methoxy-9-mercaptophenoxathiin and the Resolved 1-(N-(benzyloxycarbonyl)-L-alanyl)oxy)-9-(methoxycarbon-
yl)dithio)phenoxathiin 10-Oxide Diastereoisomers. Comments on Improved Methods for Sulfone Reduction. J . Org. Chem. 1989, 54, 3647-
h
i
k
3663. Chem. Abstr. Registry No. 99971-39-8; Ger. Pat 913177, J une 10, 1954.
position para to the smaller central group had already
been evaluated encouraged us to study other positions
of ring substitution. The effect of substitutions ortho
to the large group (here nearly always sulfone) was
selected for initial study and is the subject of this report.
The greatest portion of this work was done with phe-
noxathiins, which our earlier reports had shown were
frequently comparable in activity to the correspondingly
substituted thioxanthones.7 Note that in Table 5 of ref
7 the top right heading structure labeled “Phenox”
should be a phenoxathiin; the phenoxathiin in that
table, 40, is of high potency as indicated in the table.
droxyl function near the 2-hydrophilic site shown by our
earlier work.
An important observation is that none of the potent
compounds reported here have a nitrogen atom. It has
been stated in earlier reviews that a nitrogen atom in
the molecule was necessary for MAO I activity,9 and
indeed all of our potent compounds reported in earlier
work had a nitrogen at the putative binding site, as have
all commercially significant MAO I. However, since
these nitrogens were of widely differing basicities and
anticipated charge distributions and water-binding abil-
ity, we felt that the presence of nitrogen was not
necessary to allow MAO inhibition (other than for the
hydrazine phenelzine, the propargyl- (pargyline), cyclo-
propyl-, and haloalkenylamines, and the aminoethyla-
mides (e.g., Ro41-1049), where it is mechanistically
necessary). The potency of many of the tricyclic inhibi-
tors reported here supports our view.
One earlier generalization that applies only partially
in the study reported here was that for the hydrophilic
substituents conferring activity when para to the smaller
central group, adding a duplicate group in the sym-
metrical position in the other outer ring completely
destroyed activity. However, one symmetrical disub-
stituted compound here, 77, the 1,9-dimethylphenox-
athiin dioxide (Table 2), retained nearly as high potency
as the monomethyl 3. All activity was lost for the other
1,9-disubstituted analogues (compare 78 and 84 with
13 and 32). The thioxanthone corresponding to 77, 92
(4,5-dimethylthioxanthen-9-one dioxide), while not as
active as its monomethyl analogue 87, also did have
appreciable activity.
Resu lts a n d Discu ssion
Table 1 shows the effect on selectivity for MAO A and
on potency of monosubstitution on the phenoxathiin ring
system primarily ortho to the sulfur. It is evident that
in contrast to the substitution meta to sulfur, which
required1 a hydrophilic substituent for maximum po-
tency (however, note an exception, the potency of 14
reported in Table 1), the greatest potency with substitu-
tion ortho to sulfur requires a lipophilic substituent
(compare the potent 3, 13, 27, 74, and 76, which have
small lipophilic substituents, with, for example, 47, 48,
53, 59, 61, and 63, which have small but hydrophilic
substituents). Comparing the relative potency due to
differing lipophilic substituents in the same position
ortho to sulfur, there appears to be a stringent upper
limit to the linear size and possibly the cross section
conferring activity and potency. Thus, 3 (1-methyl) and
13 (1-ethyl) are very potent, while their higher homo-
logues 19 (1-propyl), 20 (1-pentyl), and 21 (1-isopropyl)
have progressively lower activity. Similarly, the 1-vinyl
function of 22 confers high potency, while 23, which can
be considered a methylated vinyl or methenylated ethyl
analogue, is less active than either 13 or 22. Compari-
son of the potent 1-trifluoromethyl (27) with its es-
sentially inactive pentafluoroethyl analogue (29) shows
a similar effect. We have no good explanation for the
potency of the 1-(2-hydroxyethyl) compound 42. It is
conceivable that it extends to bring the terminal hy-
A point of similarity between the structure-activity
relationship (SAR) in these compounds substituted
ortho to the central ring sulfur and the SAR reported
in our earlier papers1,7 for compounds substituted meta
to the central ring sulfur is that as in the earlier cases,
potency increases from the sulfide and sulfoxide to the
sulfone form of the central sulfur for the few cases where
our data allows comparison. For example, the activity
attributable to the 1-methyl fraction of the sulfide

Selective Inhibitors of Monoamine Oxidase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 12 2121
Ta ble 2. MAO Inhibition by Phenoxathiins with Two or More Substituents on the Aromatic Rings
IC50 (µM) or % inhib at µM
no.
substituent
n
MAO A
MAO B
formula
C14H12O3S
C16H16O3S
C18H20O3S
C22H30N2O3S‚2HCl‚2H2O
C12H10N2O3S
anal.
mp (°C)
recrysta
77
78
79
80
81
82
83
84
85c
1,9-Me2
1,9-Et2
1,9-Pr2
1,9-(CH2NEt2)2
1,9-(NH2)2
1-Et-7-OH
1-Et-2-OMe
1,9-(CH2OH)2
2,8-Br2
2
2
2
2
2
2
2
2
2
0.05
8% @ 0.1
0% @ 0.1
0% @ 0.1
8% @1
0% @ 0.1
4% @ 1
15% @ 1
7% @1
17% @1
CH
259-261
159-161
135-137
245-247
268-270
169-171
137-139
218-222
EA
0% @ 0.1
0% @ 0.1
18% @1
0% @ 0.1
0.11
0.6
15% @1
13% @1
CHS
CHS
CNb
CHN
CHS
CHS
CHS
HOAc
EA-P
A-EA
A
EA-P
A
C
C
14H12O4S
15H14O4S
C14H12O5S
C12H6O3Br2S
A
a
b
See Table 1, footnote a. Anal. CN; H: calcd, 7.10; found, 6.52. c Suter, C. M.; Mckenzie, J . P.; Maxwell, C. E. Phenoxthin. I. A
Comparison of the Directive Influences of Oxygen and Sulfur. J . Am. Chem. Soc. 1936, 58, 717-720.
Ta ble 3. MAO Inhibition of Some Thioxanth-9-ones
IC50 (µM) or % inhib at µM
MAO A MAO B
1.4 0% @ 1
no.
substituent
n
formula
anal.
mp (°C)
146.5
182.6
114.6
159.1
b
recrysta
86
87
88
89
90
91
4-Me
4-Me
4-Et
4-Et
4-CMe3
4,5-Me2
4,5-Me2
4-Cl-1-Me
1-Cl-4-Me
1-Cl-4-Me
2-Cl
0
2
0
2
2
0
2
0
0
2
2
2
C14H10OS
CH
CH
A-W
HOAc-W
A-W
0.06
0% @ 1
C
C
C
14H10O3S
15H12OS
15H12O3S
C17H16O3S
C15H12OS
C15H12O3S
C14H9OClS
C14H9OClS
2.0
0.4
0% @ 0.1
8% @1
0.3
10% @ 1
2% @ 0.1
5% @ 0.1
9% @1
CHS
CHS
CHS
CHS
CHS
HOAc-W
EA-H
A-W
104.2
234-235.5
92
9% @ 1
HOAc
93c
94c
95
34% @ 1
55% @ 1
40% @ 1
23% @ 1
1
0% @ 1
0% @ 1
0% @ 1
15% @ 0.1
11% @ 1
C
14H9O3ClS
CHS
191.6
HOAc-W
96d
97d
2-Br
a
b
See Table 1, footnote a. Mp: uncertain, sintered 33.5 °C, clarified 138.3 °C. c Archer, S.; Suter, C. M. The Preparation of Some
d
1-Alkylamino- and Dialkylaminoalkylaminothiaxanthones. J . Am. Chem. Soc. 1952, 74, 4296-4309. Gilman, H.; Diehl, J . W. Orientation
in the Thioxanthenone Nucleus. J . Org. Chem. 1959, 24, 1914-1946.
mixture 1 (7 is essentially inactive) seems greater than
that of the 1-methyl sulfoxide 2, and both are far less
potent than the sulfone 3. Similarly, the 1-ethyl sulfone
13 is far more potent than either the sulfide 11 or
sulfoxide 12. It is interesting that in the 3-ethyl series,
the sulfone 17 is also much more potent than the sulfide
15 or the sulfoxide 16 and that in the thioxanthen-9-
one series of Table 3 the corresponding (note the
differing ring-numbering systems) 4-methyl and 4-ethyl
sulfones 87 and 89 are again more potent than the
sulfides 86 and 88. It is noteworthy that none of the
variety of compounds studied in this series ortho-
substituted to the central ring sulfur had appreciable
MAO B inhibitory activity.
Selection of which of the several potent compounds
tabulated should be further studied as a clinical can-
didate was based on several criteria. The monomethyl
sulfone 3 was ruled out since it was partly irreversibly
bound to MAO A, i.e., not completely dissociated from
the enzyme in 24 h of dialysis. This irreversibility has
been shown to be associated with tyramine-induced
blood pressure rise,1 and indeed, after pretreatment
with this compound, the blood pressure of tyramine-fed
rats rose to 36% of that induced by phenelzine pretreat-
ment under our standard test conditions.10 The 3-bromo
sulfone 75, while potent in vitro, showed no activity in
rat brain MAO 3 h after oral administration at 20 mg/
kg, the ED50 cutoff selected. The 1-ethylphenoxathiin
dioxide 13 was preferred to the 1-vinyl (22), the 1-tri-
fluoromethyl (27), and the 1-iodo (76) phenoxathiin
dioxides on pharmacokinetic and other grounds, since
all showed low acute toxicity, all showed negligible MAO
B inhibition, and none caused significant tyramine-
induced blood pressure rise above vehicle control values
in rats given 15 mg/kg tyramine orally after 80%
inhibition of brain MAO had been established by the
required dosage of inhibitor. This is in contrast to blood
pressure increases found with phenelzine and moclobe-
mide. The 2-ethyl sulfone 14 showed a slight increase
in blood pressure under the test conditions.
Compound 13 had an ED80 of 8 mg/kg in inhibition
of rat brain MAO A in vivo and was active in standard
antidepressant models (Porsolt test in rats, potentiation
of 5-HT activities in rats and mice, and a Rhesus model
of borderline personality disorder). More comprehensive
data on the pharmacology of 13 have been presented

2122 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 12
Harfenist et al.
and published11-13 (synonyms for 13 were 1370U and
Sch em e 1
BW1370U87).
The detected metabolites of 13 were found to be the
vinyl analogue 22, the product of R hydroxylation 33,
the acetyl compound 48, the product of â hydroxylation
42, the product of 1,2-dihydroxylation 44, and the
carboxylic acid 54 produced by terminal oxidation of the
ethyl group. Structures were determined both by
physical methods14-16 and by comparison with synthe-
sized samples. The chirality of the potentially optically
active metabolites corresponding to 33 and 44 was not
determined. As can be seen from Table 1, 22 and 42
could contribute somewhat to activity, but the chiral 34
and 35, although differing 10-fold in in vitro potency,
were not sufficiently potent to affect total activity
measurably at their in vivo concentrations. Inciden-
tally, both were reversibly bound to MAO A. The
difference in potency between the racemic 44 and its
dextrorotatory isomer 45 seemed to make determination
of the levorotatory isomer’s in vitro potency pointless;
the small sample was used to determine that 10 mg/kg
caused 43% inhibition of rat brain MAO in vivo and that
it was reversibly bound to MAO A in vitro. Since both
enantiomers of 45 not only were synthesized by chiral
reagent dihydroxylation27 but also were separated from
44 by chiral chromatography (see Experimental Sec-
tion), elemental analysis was only done on 45A.
situ. MAO B is believed to be near the mitochondrial
membrane, while MAO A is associated with lipids and,
unlike MAO B, loses its enzyme activity when separated
from its lipid matrix.20 Thus conceivably, the specific
MAO A inhibition may be due to our tricyclic inhibitors
separating MAO A from its associated lipids, a factor
not important to retention of MAO B activity. It may
be pertinent that in the aminoethylamide series, the
literature states that “the presence of a more bulky
aromatic part led to a shift of the [inhibitory] selectivity
from MAO B to MAO A.”21
Our work was not designed to study these questions
and cannot cast more light on them. It was designed
to produce clinical trial candidates and in 13 offered a
compound which seems to meet the target of a potent
and selective inhibitor of MAO A whose enzyme binding
is reversed by tyramine, with no significant potentiation
of tyramine-induced increase in blood pressure over
control values.
A procedure used to make 13 pentadeuterated on the
ethyl group is given in the Chemistry section. The
potency of the deuterated and of the proton forms of 13
was identical, but unfortunately no statistically signifi-
cant results are available showing whether deuteration
increased the half-life in vivo.
Su m m a r y
Linear tricyclic compounds whose central ring is
composed of a larger and a smaller heteroatom (here
mostly phenoxathiin 10,10-dioxides) substituted ortho
to the sulfone function by a lipophilic group include
compounds which are potent selective inhibitors of
monoamine oxidase A and therefore are potential
antidepressants. Monosubstitution by a group of rela-
tively short length and small cross section (Me, Et, CF3,
I) is most likely to yield potent compounds. Reversibil-
ity of the formation of the enzyme-inhibitor complex
by dialysis was used to predict lack of tyramine-induced
blood pressure rise (the “cheese effect”) which had halted
past clinical use of older, nonselective MAO inhibitors.
As predicted, no significant increase in blood pressure
over control values has been found in rats dosed with
tyramine after treatment with the candidate MAO A
inhibitor 13 (1-ethylphenoxathiin 10,10-dioxide).
Two questions remain unanswered by this and our
previous publications: The first is the question of the
origin of the extremely specific inhibition of MAO A by
these tricyclic inhibitors despite their noncovalent bind-
ing (as shown by the examples of complete restoration
of activity on dialysis of the enzyme-inhibitor complex).
Our working hypothesis has been that the activity
involves noncovalent binding to the flavin prosthetic
group by these inhibitors.17 This tentative explanation
was invoked to account for the fact that many different
tricyclics with differing electron densities led to active
compounds, a result rationalized most simply by as-
suming that the binding is a polarizability phenomenon.
It is certain that the mechanism, even if not this, differs
from that responsible for the MAO inhibitory activity
of the propargylamine inhibitors, which involves cova-
lent bonding at the flavin moiety,31 and for the MAO
inhibition of the aminoethylamide inhibitors such as
Ro41-1049, which are apparently deaminated by the
appropriate MAO to the aldehyde which then forms an
adduct that initially reacts reversibly (aldehyde-am-
monia or Schiff base?) with the enzyme.18 However, a
rationale for the specificity for MAO A of our tricyclics
is more difficult to suggest, especially when the short
“reach” of the substituents on these inhibitors is con-
sidered in connection with the fact that probably 18
amino acids, including the five amino acid sequence
containing the cysteine whose sulfur binds the flavin
of the FAD prosthetic group of MAO, are identical for
both MAO A and MAO B.19 A possible answer is based
on the reported difference in position of the enzymes in
Exp er im en ta l Section
Ch em istr y. Melting points were determined using an
electrically heated oil bath and are uncorrected. All reported
compounds showed a single spot on UV-fluorescent silica gel
plates (MK6F, Whatman International, Ltd.) using EtOAc-
hexanes or CH2Cl2-hexanes unless otherwise stated. Products
were identified by elemental analyses and by NMR and in most
cases mass spectra. Preparative conditions were not opti-
mized.
Most of the 1-substituted phenoxathiin 10,10-dioxides were
made by the action of electrophiles on the lithio derivative of
phenoxathiin 10,10-dioxides22 as shown in Scheme 1, followed
if necessary by further reactions. The monocarboxylic acid
made in this way by Shirley and Lehto22 had a low melting
point raised markedly on recrystallization. Later studies
showed that it was contaminated by 1,9-disubstituted prod-
uct.23 Our preparations of 13 by direct ethylation of lithiophe-

Selective Inhibitors of Monoamine Oxidase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 12 2123
MeCN. From 1.06 g of starting 48, 0.41 g of 34 was isolated,
with [R]20 ) +13.8° and [R]20 ) +40.5°, both in CHCl3. Its
Sch em e 2
D
365
optical purity was found to be 90% by chromatography using
a Chiralcel OJ column.
(-)1-(1-Hyd r oxyeth yl)p h en oxa th iin 10,10-d ioxid e (35)
was made similarly using (R)-Alpine-Borane and was 75% pure
as determined with the Chiralcel OJ column.
1-Eth ylp h en oxa th iin 10,10-Dioxid e (13) fr om 33.
A
solution of 590.2 g (2.14 mol) of 33 in 5.4 L of acetic acid
containing 250 mL of 70% perchloric acid was treated under
nitrogen with 65 g of Pearlman’s catalyst24 (Aldrich Chemical
Co.) and then hydrogenated under pressure slightly above
atmospheric. Catalyst was then removed by filtration, and
the filtrate was diluted with 4 volumes of water. After
remaining overnight, the resulting white solid was removed
by filtration, rinsed with water, dried, and recrystallized. Two
forms differing in melting point were obtained, as tabulated.
1-P en ta d eu ter ioeth ylp h en oxa th iin 10,10-Dioxid e. a .
1-Tr id eu ter ioa cetylp h en oxa th iin 10,10-d ioxid e was pre-
pared by stirring 15.0 g of 1-acetylphenoxathiin 10,10-dioxide
(48) with 30 mL of D2O, 150 mL of dry DMF, and 0.6 g of
potassium tert-butoxide under argon for 5 h. The reaction
mixture was then transferred to a round-bottom flask with
DMF and 25 mL more of D2O, and solvents were removed in
vacuo. The resulting orange solid was transferred with 40 mL
of D2O to a fritted glass funnel, filtered, and dried in vacuo.
NMR showed 0.12 methyl proton by integration and compari-
son with the aromatic multiplet. Recrystallization from ethyl
acetate-hexanes after Darco G60 treatment gave 12.44 g of
pearly platelets.
b. Red u ction to P en ta d eu ter io-13. The above trideu-
terioacetyl material, 5.48 g, was reduced with Pearlman’s
catalyst24 (Aldrich Chemical Co.) and deuterium in DOAc made
by cautiously mixing 150 mL of acetic anhydride and 30 mL
of “98.8% D” D2O and, when no reaction was evident, adding
3 drops of CF3COOH followed by stirring with exclusion of
moisture for 1 h at room temperature and then overnight at
50 °C. The product of the usual workup showed 0.1 alkyl H
by NMR, mp 115.2 °C. Anal. (C14H7D5O2S). Use of HOAc as
solvent in this reduction gave nondeuterated 13.
noxathiin 10,10-dioxide showed some 1,9-diethyl compound 78
and small amounts of additional ethylation products presum-
ably ethylated in the 4- or 6-position ortho to oxygen as well
as in the 1-position. Chromatography to yield large amounts
of 13 suitable for clinical trial seemed less efficient than the
alternative of preparing the more readily purified 1-(1-hy-
droxyethyl)phenoxathiin 10,10-dioxide (33) by reaction of
lithiophenoxathiin dioxide with acetaldehyde and subsequently
removing the benzylic hydroxy group by reduction with
hydrogen and Pearlman’s palladium hydroxide on carbon
catalyst.24 This procedure produced a very small proportion
of dihydro derivatives of 13, identified as such by mass spectral
molecular weight. Detailed preparations of 13, 22, 33, 42, 44,
and 48 have been published;25,26 some slightly modified
procedures are given below.
An alternative method of synthesis of phenoxathiins is given
in Scheme 2. This is especially useful for preparing halophe-
noxathiins substituted in other than the 1-position and their
10,10-dioxides which could be converted by lithium exchange
with lithium alkyls at low temperatures to derivatives sub-
stituted in other than the 1-position. No migration of lithium
to the 1-position was noticed under our conditions. Potassium
tert-butoxide was usually used as base to form the thiolate for
the preparation of the sulfide in the first step, especially for
convenience on small-scale preparations. Careful exclusion of
air was necessary to maximize yields.
Mon olith ia tion of P h en oxa th iin 10,10-Dioxid e. A 12-L
flask equipped with a stirrer, thermometer, addition funnel,
and nitrogen demand system was purged with nitrogen, and
a solution of 414 g of phenoxathiin 10,10-dioxide in 3.7 L of
peroxide-free tetrahydrofuran (previously dried over 4A mo-
lecular sieves) was prepared and cooled in a dry ice-acetone
bath to below -60 °C. A 1.6 M solution of butyllithium in
hexane (1181 mL) was added at a rate which allowed the
internal temperature to remain at or below -55 °C (temper-
atures above about -35 °C led to rapid degradation of yields).
These conditions were also satisfactory on a smaller scale. All
subsequent reactions were (also) run under an inert atmo-
sphere.
1-(1-Hyd r oxyeth yl)p h en oxa th iin 10,10-Dioxid e (33).
Dropwise addition of 20.37 g (0.46 mol) of chilled acetaldehyde
to lithiophenoxathiin dioxide solution prepared from 44 g (0.19
mol) of phenoxathiin 10,10-dioxide at -50 °C during 45 min
was followed by allowing spontaneous warming to room
temperature. After removal of solvent under reduced pressure,
the resulting yellow-orange residue was stirred overnight with
360 mL of 0.5 N HCl and filtered, and the solid was washed
with water and then 1.5 L of ethanol. The resulting solid was
sufficiently pure for synthetic use; an analytical sample was
prepared (Table 1).
1-Eth ylp h en oxa th iin (11) was made by reduction of 1.07
g (4.11 mmol) of 13 with 20.5 mL of 1 M DIBAL (Aldrich
Chemical Co.) in toluene and an additional 25 mL of toluene
under reflux for 5 h, followed by decomposition of excess
DIBAL with acetone (exotherm) and purification by column
chromatography using hexanes as solvent and silica gel as
stationary phase. TLC (hexanes) and proton NMR with correct
integration values as well as elemental analysis were used as
criteria of purity for the 220 mg of liquid product obtained as
the faster moving eluate fraction.
1-Acetylp h en oxa th iin 10,10-Dioxid e (48) fr om 33.
A
mixture of 29.83 g (0.14 mol) of pyridinium chlorochromate
(Aldrich Chemical Co.), 25.2 g of 4A molecular sieves, 13.92 g
(0.05 mol) of 33, and 580 mL of CH2Cl2 was stirred for 21 h
and then filtered through a 4-cm deep layer of Kieselguhr
which removed color. The apparatus and Kieselguhr were
rinsed with EtOAc and the solutions combined and chromato-
graphed on silica gel, using EtOAc-hexanes and collecting
arbitrary fractions. Evaporation of solvent gave solids melting
between 124.5 and 142 °C. The higher-melting residues were
recrystallized from hot EtOAc by adding hexanes to incipient
turbidity, yielding 9.70 g melting at 144.0 °C, and the lower-
melting residues gave an additional 1.16 g melting at 142.4
°C. Recrystallization from EtOAc-pentane gave material
giving a single spot on TLC using 1:1 EtOAc-pentane and the
correct elemental analysis, proton NMR, and mass spectrum
molecular ion.
(+)1-(1-Hyd r oxyeth yl)p h en oxa th iin 10,10-d ioxid e (34)
was made by reducing 48 with excess of (S)-Alpine-Borane
(Aldrich Chemical Co.), decomposing the product with 1 N HCl,
and chromatographing a CH2Cl2 extract of the product with
CH2Cl2 containing MeCN increased in 1% increments to 10%.
The slower material was an oil with an odor resembling
R-pinene, which was again chromatographed using 500-mL
portions of CH2Cl2-MeCN increasing by 1% increments to 8%
1-Vin ylp h en oxa th iin 10,10-Dioxid e (22) fr om 33.
A
slurry of 1.06 g (3.84 mmol) of 33, 60 mL of dried CH2Cl2, and
0.730 mL (10 mmol) of SOCl2 was heated under reflux
protected from atmospheric moisture for 3.5 h. Volatile
materials were then distilled off on a hot water bath at water
aspirator pressure, and then 10 mL of ethanol was added and
similarly removed. The residual solid was recrystallized from

2124 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 12
Harfenist et al.
ethanol, the resulting mother liquors were evaporated down,
and the residue was recrystallized from EtOAc-pentane. The
combined products were again recrystallized fron EtOAc-
pentane to yield 0.310 g of 22.
ibility which would minimize tyramine-induced blood pressure
rise, was described earlier.7 The whole animal antidepressant
tests mentioned followed literature methods.28-30
Ack n ow led gm en t. The authors thank Krenitsky
Pharmaceuticals, Inc., the licensee of 13 and its conge-
ners, for permission to publish data used in this paper.
1-(1-Meth ylvin yl)p h en oxa th iin 10,10-d ioxid e (23) was
made by an analogous method.
1-Eth in ylp h en oxa th iin 10,10-d ioxid e (24) was made by
adding excess Br2 to 4.2 g (16.3 mmol) of the 1-vinyl compound
22 in 100 mL of CH2Cl2, evaporating off solvent after 10 min,
taking up the residue in dried THF, and stirring that solution
with 7.6 g (67 mmol) of potassium tert-butoxide with exclusion
of air. After 20 min the black reaction mixture was poured
into water, acidified with 1 N HCl, and extracted with CH2-
Cl2. Evaporation of solvent and two recrystallizations of the
resulting solid from ethyl acetate by addition of pentane gave
1.42 g of light-yellow solid.
Refer en ces
(1) The previous paper was: Harfenist, M.; J oseph, D. M.; Spence,
S. C.; McGee, D. P. C.; Reeves, M. D.; White, H. L. Selective
Inhibitors of Monoamine Oxidase. 4. SAR of Tricyclic N-
Methylcarboxamides and Congeners Binding at the Tricyclics'
Hydrophilic Binding Site. J . Med. Chem, 1997, 40, 2466-2473.
(2) A preliminary publication contains
Harfenist, M.; McGee, D. P. C.; White, H. L. A selective,
Reversible Competitive Inhibitor of Monoamine Oxidase
a portion of this work:
A
1-(Tr iflu or om eth yl)p h en oxa th iin 10,10-d ioxid e (27)
was made by heating a mixture of 2.06 g (5.75 mmol) of
1-iodophenoxathiin 10,10-dioxide hemihydrate (76), 3.13 g (4
equiv) of sodium trifluoroacetate, 2.19 g (2 equiv) of cuprous
iodide, and 45 mL of N-methylpyrrolidone. After a brief
incursion to 220 °C, the reaction mixture was kept at 160 °C
for 2 h and poured onto 500 mL of ice-water. The resulting
solids were filtered off and extracted with first acetone and
then CH2Cl2. The solution was extracted with 1 N NaOH and
then 1 N HCl and dried, and solvent was removed to leave a
dark solid. Chromatography using 3:1 hexanes-EtOAc gave
140 mg of a faster moving fraction whose mass spectrum
indicated that it was recovered 76, 235 mg of desired product,
and traces of what appeared to be 29.
2-(Tr iflu or om eth yl)p h en oxa th iin 10,10-d ioxid e (28). A
mixture of 0.87 g (3.11 mmol) of 2-bromophenoxathiin, 1.18 g
(6.23 mmol as CuI) of cuprous iodide, 1.69 g of sodium
trifluoroacetate, 25 mL of DMF, and 10 mL of toluene was
heated, removing vapor to 152.4 °C. After 1.5 h the reaction
mixture was cooled and partitioned between CHCl3 and water.
The organic portion was dried under high vacuum to remove
DMF, oxidized with excess H2O2 in HOAc at 80 °C, and
extensively chromatographed first with 1:1 CH2Cl2 and petro-
leum ether and then with 1:1 EtOAc and hexanes. The final
product still had a minor additional TLC spot.
Containing No Nitrogen, With Negligible Potentiation of
Tyramine-Induced Blood Pressure Rise. J . Med. Chem. 1991,
34, 2931-2933.
(3) J ohnston, J . P. Some Observations on
a New Inhibitor of
Monoamine Oxidase in Brain Tissue. Biochem. Pharmacol. 1968,
17, 1285-1297. Squires, R. F. Additional Evidence for the
Existence of Several Forms of Mitochondrial Monoamine Oxi-
dase in the Mouse. Biochem. Pharmacol. 1968, 17, 1401-1409.
(4) Bach, A. W. J .; Lan, N. C.; J ohnson, D. L.; Abell, C. W.;
Bembenek, M. E.; Kwan, S.-W.; Seeburg, P. H.; Shih, J . C. cDNA
Cloning of Human Liver Monoamine Oxidase A and B: Molec-
ular Basis of Differences in Enzymatic Properties. Proc. Natl.
Acad. Sci. U.S.A. 1988, 85, 4934-4938.
(5) Murphy, D. L.; Lipper, S.; Pickar, D.; J imerson, D.; Cohan, R.
M.; Garrick, N. A.; Alterman, I. S.; Campbell, I. C. Selective
Inhibition of Monoamine Oxidase Type A: Clinical Antidepres-
sant Effects and Metabolic Changes in Man. In Monoamine
Oxidase Inhibitors - The State of The Art; Youdim, M. B. H.,
Paykel, E., Eds.; J ohnWiley and Sons: New York, 1981; pp 189-
205.
(6) Saura, J .; Nadal, E.; van den Berg, B.; Vila, M.; Bombi, J . A.;
Mahi, N. Localization of Monoamine Oxidase in Human Periph-
eral Tissues. Life Sci. 1996, 59, 1341-1349.
(7) Harfenist, M.; Heuser, D. J .; J oyner, C. T.; Batchelor, J . F.;
White, H. L. Selective Inhibitors of Monoamine Oxidase. 3.
Structure-Activity Relationship of Tricyclics Bearing Imidazo-
line, Oxadiazole, or Tetrazole Groups. J . Med. Chem. 1996, 39,
1857-1863.
(8) White, H. L.; Harfenist, M.; Beek, O.; Soroko, F.; Cooper, B.;
Maxwell, R.; BW A616U - A Selective Inhibitor of MAO-A With
Negligible Tyramine-potentiating Activity and Potential Anti-
depressant Activity. Soc. Neurosci. Abstr. 1985, 11, 442.
(9) J ohnson, C. L. Quantitative Structure-Activity Studies on
Monoamine Oxidase Inhibitors. J . Med. Chem. 1976, 19, 605.
(10) Carroll, M.; Beek, O. Blood Pressure Effects of Monoamine
Oxidase Inhibitors In Response to Orally Administered Tyramine
in the Rat. Drug Dev. Res. 1992, 25, 215-218.
(11) Cooper, B. R.; White, H. L.; Beek, O.; Norton, R. M.; Rigdon, G.
C.; Howard, J . L.; Kraemer, G. W.; Ferris, R. M. Overview of
the CNS Pharmacology of 1370U87: A Chemically Novel,
Reversible, Selective MAO-A Inhibitor With Potential to Be a
New Antidepressant Drug. Drug Dev. Res. 1992, 25, 181-190.
(12) White, H. L.; Scates, P. W. Mechanism of Monoamine Oxidase-A
Inhibition by BW 1370U87. Drug Dev. Res. 1992, 25, 191-199.
(13) Burchall, C. J .; Soroko, F. E.; Rigdon, G. C. Potentiation of the
Behavioral Effects of 5-Hydroxytryptophan by BW 1370U87, a
Selective Monoamine Oxidase-A Inhibitor. Drug Dev. Res. 1992,
25, 209-213.
1-(P en ta flu or oeth yl)p h en oxa th iin 10,10-d ioxid e (29)
was made by the method used to prepare 27 from 76 and
sodium pentafluoropropionate in poor yield.
1-(1,2-Dih yd r oxyeth yl)p h en oxa th iin 10,10-Dioxid e (44)
fr om 22. The product of heating under reflux of 10.1 g (0.0366
mol) of 33, 200 mL of CH2Cl2, and 20 mL of SOCl2 for 12 h
followed by addition of 6 mL more of SOCl2 and another 6 h
of reflux had solvent removed in a hot water bath under
reduced pressure and then had two 88-mL portions of 97%
formic acid added and removed in vacuo sequentially. The
residue was dissolved in 100 mL of 97% formic acid and stirred
at room temperature for 12 h after addition of 4.5 mL of 30%
hydrogen peroxide and for an additional 5 h after 5 mL more
of 30% hydrogen peroxide had been added. The reaction was
then heated on a steam bath until the solids were dissolved,
cooled to room temperature, and diluted with 1 L of water,
and the supernatant liquid was decanted from an oily residue.
The separate enantiomers of 44, 45, and 45A were prepared
using OsO4 and H2O2 with the Corey chiral ethylenediamine
(14) Harrelson, J . C.; J ohnson, T. E.; White, H. L.; Welch, R. M.
Metabolism of BW 1370U87 in Rat, Dog, and Man. Drug Dev.
Res. 1992, 25, 219-227.
(15) Taylor, L. C. E.; J ohnson, R. L.; J ohn-Williams, L. S.; J ohnson,
T.; Chang, S. Y. The Use of Low Energy, Collisionally Activated
Dissociation Negative-Ion Tandem Mass Spectrometry for the
Characterization of Dog and Human Urinary Metabolites of the
Drug BW 1370U87. Rapid Commun. Mass Spectrom. 1994, 8,
265-273.
(16) Harrelson, J . C.; Wintermute, S.; Kanics, L.; Welch, R. M.;
J ohnson, T. E. High Performance Liquid-Chromatographic Assay
With Ultraviolet Detection For The Analysis of BW 1370U87
and Its Two Major Metabolites in Human Plasma. J . Liq.
Chromatogr. 1993, 16, 2387-2398.
(17) Sanchez-Marin, J .; Orti, E.; Aulio, J . M.; Tomas, F. Molecular
Associations Between Lumiflavine and Some Beta-Carbolines.
Theochem 1985, 21, 21-27.
catalyst,27 giving for the levorotatory isomer 45A [R]20
)
D
-56.3° found (second peak) to be 93% optically pure by
chromatography on Chiralcel OJ . Chromatography of racemic
44 on a semipreparative Pirkle column Regis 1010DPG gave
99% material with [R]20 ) -63.4°.
D
Biologica l Meth od s. The methods used were given in
detail in previous papers.1,2 They involved a radiometric
procedure using [3H]serotonin and [14C]phenethylamine. The
MAO assays were performed in triplicate at each concentration
of the putative inhibitor. The percent inhibition showed SEM
variation within 5% of mean values. IC50 values were obtained
by plotting mean values vs log of inhibitor concentration and
estimating visually from these plots. Dialysis of MAO-
inhibitor complexes, used as a screening test for likely revers-
(18) Cesura, A. M.; Bos, M.; Galva, M. D.; Imhof, R.; Da Prada, M.
Characterization of the Binding of [3H]Ro 41-1049 to the Active
Site of Human Monoamine Oxidase-A. Mol. Pharmacol. 1990,
37, 358-366.

Selective Inhibitors of Monoamine Oxidase
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 12 2125
(19) Cesura, A. M.; Pletscher, A. The New Generation of Monoamine
Oxidase Inhibitors. Prog. Drug Res. 1992, 38, 171-297.
(20) White, H. L.; Glassman, A. T. Multiple Binding Sites of Human
Brain and Liver Monoamine Oxidase: Substrate Specificities,
Selective Inhibitions, and Attempts To Separate Enzyme Forms.
J . Neurochem. 1977, 987-997.
(27) Corey, E. J .; J ardine, P. da S.; Virgil, S.; Wuen, P. W.; Connell,
R. D. Enantioselective Vicinal Hydroxylation of Terminal and
E-1,2-Disubstituted Olefins by a Chiral Complex of Osmium
Tetroxide. An Effective Controller System and a Mechanistic
Model. J . Am. Chem. Soc. 1989, 111, 9243-9244.
(28) Smith, L. M.; Peroutka, S. J . Differential Effects of 5-Hydrox-
ytryptamine-Selective Drugs on the 5-HT Behavioral Syndrome.
Pharmacol. Biochem. Behav. 1986, 24, 1513-1519.
(29) Porsolt, R. D.; Anton, G.; Blavet, N.; J alfre, M. Behavioral
Despair in Rats: A New Model Sensitive to Antidepressant
Treatments. Eur. J . Pharmacol. 1978, 47, 379-391.
(30) Kraemer, G. W.; Ebert, M. H.; Schmidt, D. E.; Mckinney, W. T.
A Longitudinal Study of the Effect of Different Social Rearing
Conditions on Cerebrospinal Fluid Norepinephrine and Biogenic
Amine Metabolites in Rhesus Monkeys. Neuropsychopharma-
cology 1989, 2, 175-189.
(21) Reference 18, p 216.
(22) Shirley, D. A.; Lehto, E. A. The Metalation of Phenoxathiin 10-
Oxide and 10,10-Dioxide with n-Butyllithium. J . Am Chem. Soc.
1955, 77, 1841-1843.
(23) Gilman, H.; Eidt, S. H. Dimetalation of Phenoxathiin and
Phenoxathiin 10-Dioxide by Organometallic Compounds. J . Am.
Chem. Soc. 1956, 78, 2633-2367.
(24) Pearlman, W. M. Noble Metal Hydroxides on Carbon. Nonpy-
rophoric Dry Catalysts. Tetrahedron Lett. 1967, 1663-1664.
(25) McGee, D. P. C.; White, H. L.; J ohnson, T. E.; Harrelson, J . C.;
Harfenist, M.; Reeves, M. D.; Chandrasurin, P. 1-Alkylphenox-
athiin 10,10-Dioxides, A Method For Their Preparation And
Their Use For the Manufacture of Antidepressant Pharmaceu-
ticals. PCT Int. Appl. WO 9204897, April 2, 1992.
(31) Hellerman, L.; Erwin, V. G. Mitochondrial Monoamine Oxidase.
II. Action of Various Inhibitors for the Bovine Kidney Enzyme.
Catalytic Mechanism. J . Biol. Chem. 1968, 243, 5234-5243.
(26) Harfenist, M.; McGee, D. P. C.; White, H. L.; Cooper, B. R. U.S.
Patent 5420156, May 30, 1995.
J M970862J