Application of the goldilocks effect to the design of potent and selective inhibitors of phenylethanolamine N-methyltransferase: Balancing pKa and steric effects in the optimization of 3-methyl-1,2,3,4- tetrahydroisoquinoline inhibitors by β-fluorination

Article abstract of DOI:10.1021/jm051262k

3-Methyl-1,2,3,4-tetrahydroisoquinolines (3-methyl-THIQs) are potent inhibitors of phenylethanolamine N-methyltransferase (PNMT), but are not selective due to significant affinity for the α₂-adrenoceptor. Fluorination of the methyl group lowers the pK_a of the THIQ amine from 9.53 (CH₃) to 7.88 (CH₂F), 6.42 (CHF₂), and 4.88 (CF₃). This decrease in pK^a results in a reduction in affinity for the α₂-adrenoceptor. However, increased fluorination also results in a reduction in PNMT inhibitory potency, apparently due to steric and electrostatic factors. Biochemical evaluation of a series of 3-fluoromethyl-THIQs and 3-trifluoromethyl-THIQs showed that the former were highly potent inhibitors of PNMT, but were often nonselective due to significant affinity for the α₂-adrenoceptor, while the latter were devoid of α₂-adrenoceptor affinity, but also lost potency at PNMT. 3-Difluoromethyl-7-substituted-THIQs have the proper balance of both steric and pK_a properties and thus have enhanced selectivity versus the corresponding 3-fluoromethyl-7-substituted-THIQs and enhanced PNMT inhibitory potency versus the corresponding 3-trifluoromethyl-7-substituted- THIQs. Using the "Goldilocks Effect" analogy, the 3-fluoromethyl-THIQs are too potent (too hot) at the α₂-adrenoceptor and the 3-trifluoromethyl-THIQs are not potent enough (too cold) at PNMT, but the 3-difluoromethyl-THIQs are just right. They are both potent inhibitors of PNMT and highly selective due to low affinity for the α₂- adrenoceptor. This seems to be the first successful use of the β-fluorination of aliphatic amines to impart selectivity to a pharmacological agent while maintaining potency at the site of interest.

Full text of DOI:10.1021/jm051262k

J. Med. Chem. 2006, 49, 2939-2952
2939
Application of the Goldilocks Effect to the Design of Potent and Selective Inhibitors of
Phenylethanolamine N-Methyltransferase: Balancing pKa and Steric Effects in the Optimization
of 3-Methyl-1,2,3,4-tetrahydroisoquinoline Inhibitors by â-Fluorination1
Gary L. Grunewald,* Mitchell R. Seim, Jian Lu, Mariam Makboul, and Kevin R. Criscione
Department of Medicinal Chemistry, UniVersity of Kansas, Lawrence, Kansas 66045
ReceiVed December 19, 2005
3-Methyl-1,2,3,4-tetrahydroisoquinolines (3-methyl-THIQs) are potent inhibitors of phenylethanolamine
N-methyltransferase (PNMT), but are not selective due to significant affinity for the R2-adrenoceptor.
Fluorination of the methyl group lowers the pKa of the THIQ amine from 9.53 (CH3) to 7.88 (CH2F), 6.42
(CHF2), and 4.88 (CF3). This decrease in pKa results in a reduction in affinity for the R2-adrenoceptor.
However, increased fluorination also results in a reduction in PNMT inhibitory potency, apparently due to
steric and electrostatic factors. Biochemical evaluation of a series of 3-fluoromethyl-THIQs and 3-triflu-
oromethyl-THIQs showed that the former were highly potent inhibitors of PNMT, but were often nonselective
due to significant affinity for the R2-adrenoceptor, while the latter were devoid of R2-adrenoceptor affinity,
but also lost potency at PNMT. 3-Difluoromethyl-7-substituted-THIQs have the proper balance of both
steric and pKa properties and thus have enhanced selectivity versus the corresponding 3-fluoromethyl-7-
substituted-THIQs and enhanced PNMT inhibitory potency versus the corresponding 3-trifluoromethyl-7-
substituted-THIQs. Using the “Goldilocks Effect” analogy, the 3-fluoromethyl-THIQs are too potent (too
hot) at the R2-adrenoceptor and the 3-trifluoromethyl-THIQs are not potent enough (too cold) at PNMT, but
the 3-difluoromethyl-THIQs are just right. They are both potent inhibitors of PNMT and highly selective
due to low affinity for the R2-adrenoceptor. This seems to be the first successful use of the â-fluorination
of aliphatic amines to impart selectivity to a pharmacological agent while maintaining potency at the site of
interest.
Table 1. The Effect of Fluorine on the pKa of THIQ
Introduction
Aliphatic amines are one of the most common functional
groups in bioactive molecules2 and are often crucial in the
binding of these compounds by acting as a hydrogen bond donor
a
compd
R
pKa
or acceptor, or through electrostatic interactions when proto-
nated. Fluorine can strongly influence the basicity of aliphatic
amines. For example, the incorporation of a monofluoromethyl,
difluoromethyl, or trifluoromethyl group at the 3-position of
1,2,3,4-tetrahydroisoquinoline (THIQ, 1) lowers the pKa of the
amine by approximately 1.5 log units per fluorine (Table 1).
The decrease in pKa associated with the â-fluorination of
aliphatic amines has been shown to have a profound effect on
their biological properties. Some examples include increased
substrate activity for lung N-methyltransferase,3 decreased inhi-
bition of mitochondrial monoamine oxidase,4 increased N-
dealkylation by CYP1A25 and decreased aromatic hydroxylation
by CYP2D66,7 of fluorinated propranolol analogues, increased
oral bioavailability,8 and changes in tissue distribution.4,9
1
H
CH3
CH2F
CHF2
CF3
9.53
19d
19a
19b
19c
9.42b
7.88b
6.42
4.88b
a pKa values ((0.02) were determined by potentiometric titration by
pION, Inc., Cambridge, MA. b Ref 44.
Since the steric and pKa changes caused by fluorination can
simultaneously affect multiple biological properties, it is not
uncommon to observe that the added benefits of fluorination
are accompanied by negative effects. A decrease in pKa that
leads to enhanced oral absorption or improved metabolic
properties may also result in a decrease in potency.10 While
there are examples of varying the extent of fluorination to find
an optimal pKa at which pharmacokinetic properties are
improved and potency is maintained,8 this approach has not been
Figure 1. The terminal step in epinephrine (3) biosynthesis is the
transfer of a methyl group from AdoMet (4) to norepinephrine (2) to
form epinephrine (3) and the cofactor product AdoHcy (5).
used to impart selectivity to a pharmacological agent while
maintaining potency at the site of interest.7,11
As one means of determining the function of epinephrine (3)
in the CNS, our laboratory has targeted phenylethanolamine
N-methyltransferase (PNMT; EC 2.1.1.28). This enzyme cata-
lyzes the terminal step in the biosynthesis of epinephrine (3),
in which an activated methyl group is transferred from S-
adenosyl-L-methionine (AdoMet; 4) to the primary amine of
norepinephrine (2) to form epinephrine (3) and the cofactor
product S-adenosyl-L-homocysteine (AdoHcy; 5) (Figure 1).
* To whom correspondence should be addressed. Gary L. Grunewald,
Dept. of Medicinal Chemistry, School of Pharmacy, Malott Hall, Rm. 4060,
University of Kansas, 1251 Wescoe Hall Dr., Lawrence, KS 66045-7582.
Phone: (785) 864-4497. Fax: (785) 864-5326. E-mail: ggrunewald@
ku.edu.
10.1021/jm051262k CCC: $33.50 © 2006 American Chemical Society
Published on Web 04/14/2006

2940 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10
Grunewald et al.
Table 2. In Vitro Human PNMT (hPNMT) and R2-Adrenoceptor Affinity of Some PNMT Inhibitors
Ki (µM) ( SEMa
b
compd
R7
R8
R3
hPNMT
R2
selectivity R2/hPNMT
6c
Cl
Br
I
NO2
SO2NH2
Br
Cl
H
H
H
H
H
H
H
H
H
H
H
0.0031 ( 0.0006d
0.056 ( 0.003f
0.093 ( 0.007g
0.12 ( 0.01f
0.021 ( 0.005
0.23 ( 0.02
0.22 ( 0.04
4.3 ( 0.3
6.8
4.1
2.4
36
360
280
7e
8e
9e
10g
11ah
11ci
11df
H
0.28 ( 0.02d
0.023 ( 0.003f
3.2 ( 0.3j
100 ( 10
CH2F
CF3
CH3
6.4 ( 0.2
Br
Br
>1000
>310
0.017 ( 0.005
1.1 ( 0.1
65
a Standard error of the mean. b In vitro activities for the inhibition of [3H]clonidine binding to the R2-adrenoceptor. c Ref 15. d Ref 39. e Ref 21. f Ref 48.
g Ref 22. h Ref 44. i Ref 23. j Ref 24.
The role of epinephrine (3) as a hormone in the periphery is
well established. An acute stress response triggers the release
of epinephrine (3) into the blood stream, which initiates the
physiological changes associated with the “fight or flight
syndrome”. In contrast to the vast understanding of the functions
of epinephrine (3) in the periphery, its role in the CNS is much
less clear. Significant levels of epinephrine (3) are present in
the mammalian CNS,12-14 where it is postulated to act as a
neurotransmitter. The most thoroughly studied physiological
process to which epinephrine (3) in the CNS (central epineph-
rine) has been linked is the regulation of peripheral blood
pressure. Administration of the PNMT inhibitor SK&F 6413915
(6, Table 2) to hypertensive rats resulted in a reduction in
peripheral blood pressure.16 However, 6 also showed significant
affinity for the R2-adrenoceptor.17 Thus, some of the decrease
in peripheral blood pressure may be attributed to an R2-
adrenergic effect rather than to decreases in central epinephrine
concentrations.18-20 Selective inhibitors of PNMT would be
Figure 2. The active site of hPNMT cocrystallized with SK&F 64139
useful pharmacological tools for clearly defining the connection
(6) and AdoHcy (5).27 A Connolly (solvent accessible) surface exposing
between central epinephrine concentrations and peripheral blood
6 and 5 is also shown and indicates the presence of a lipophilic channel
adjacent to the 3-position of 6. A lipophilic potential is mapped on the
pressure.
THIQs having electron-withdrawing 7-substituents (either
lipophilic or hydrophilic) are highly potent inhibitors of
PNMT,21 and those with lipophilic electron-withdrawing 7-sub-
stituents, such as halides (7 and 8) are generally more potent
than those having hydrophilic groups [9 and SK&F 2966122
(10)]. Unfortunately, THIQs bearing lipophilic electron-
withdrawing 7-substituents also have significant affinity for the
R2-adrenoceptor. Addition of a methyl (11d) or a fluoromethyl
(11a) group to the 3-position of 7 provides a slight increase in
PNMT inhibitory potency while decreasing affinity for the R2-
adrenoceptor. While compounds 11d and 11a still have sig-
nificant affinity for the R2-adrenoceptor, the 3-trifluoromethyl
analogue (11c) shows a dramatic reduction in R2-adrenoceptor
affinity. Previous studies suggest that this is due to the large
drop in pKa of the THIQ amine (pKa ≈ 5) making it mostly
neutral at physiological pH. Unfortunately, 11c is not particularly
potent at PNMT, showing a greater than 100-fold loss of potency
versus 11d or 11a. This has been attributed to the increased
steric bulk of the trifluoromethyl group.23,24
Connolly surface whereby the areas shown in green are neutral, blue
are hydrophilic, and brown are lipophilic. Carbon is white, nitrogen is
blue, oxygen is red, chlorine is green, and sulfur is yellow. Hydrogens
are not shown for clarity.
and Asn39. The â-hydroxyl of phenylethanolamine substrates
or the amine of THIQ inhibitors interacts with Glu219 and
Asp267. The substrate â-hydroxyl group interacts through
hydrogen bonds, but it is currently unknown whether the THIQ
amine interacts through hydrogen bonds in its neutral form or
through an electrostatic interaction in its protonated form. We
have cocrystallized hPNMT with inhibitors containing either a
lipophilic (8)26 or hydrophilic (10)25 7-substituent. Lipophilic
7-subsituents interact with the side chains of Val53 and Arg44.
Hydrophilic 7-substituents interact with Lys57 through hydrogen
bonds.25 Examination of the hPNMT active site region sur-
rounding the 3-position of cocrystallized THIQ inhibitors
revealed the presence of a narrow lipophilic channel (Figure
2), which was consistent with THIQs having small 3-substituents
(e.g., methyl, ethyl, or fluoromethyl) being more potent at
PNMT than those having larger 3-substituents (e.g., isopropyl,
butyl, pentyl, or trifluoromethyl). Docking studies with 3-tri-
fluoromethyl-THIQs showed that the trifluoromethyl group did
not fit into the lipophilic channel due to negative steric
interactions.24
Analysis of the crystal structure of human PNMT (hPNMT)
cocrystallized with inhibitors25-27 and substrates,28 coupled with
site directed mutagenesis studies,27 has helped to identify the
key amino acids at the hPNMT active site, some of which are
shown in Figure 2. The aromatic rings of substrates or THIQ-
type inhibitors are positioned between the side chains of Phe182

Design of Potent and SelectiVe Inhibitors of PNMT
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10 2941
Scheme 2a
Scheme 1a
a Reagents and conditions: (a) H2, Pd/C, MeOH; then HCl, NaNO2, CuI,
KI; (b) FSO2CF2CO2Me, CuI, DMF; (c) BH3‚THF.
Scheme 3a
a Reagents and conditions: (a) P2O5; (b) PhCH2MgCl, Et2O; then NaBH4,
MeOH; (c) ClCO2Me, pyridine, CH2Cl2; (d) polyphosphoric acid; (e)
BH3‚THF; (f) H2SO4, KNO3; (g) H2, Pd/C, MeOH; then HBr, NaNO2, CuBr;
(h) H2, Pd/C, MeOH; then HCl, NaNO2; then NaOH, KCN, NiSO4‚6H2O,
benzene.
a Reagents and conditions: (a) ClSO3H; (b) NH2NH2, THF; then MeI,
NaOAc, EtOH; (c) BH3‚THF; (d) NH4OH, CH3CN.
We hypothesized that THIQs having a 3-substituent that (1)
sufficiently lowers the pKa of the THIQ amine to diminish
affinity for the R2-adrenoceptor, and (2) is small enough to
maintain potency at PNMT would be an ideal inhibitor. We
reasoned that 3-difluoromethyl-THIQs could meet these criteria
because (1) the pKa of the THIQ amine is reduced below 6.5,
and (2) docking studies indicated that the 3-difluoromethyl group
would be able to fit into the lipophilic channel. Application of
the “Goldilocks Effect”29 suggested that while the 3-fluoro-
methyl-THIQs were too potent (too hot) at the R2-adrenoceptor
and the 3-trifluoromethyl-THIQs were not potent enough (too
cold) at PNMT, the 3-difluoromethyl-7-substituted-THIQs
(11b-25b) would be just rightsboth highly potent at PNMT
and highly selective due to low affinity for the R2-adrenoceptor.
Chemistry. The synthesis of 3-difluoromethyl-7-substituted-
THIQs 11b-25b and 7-(N-substituted aminosulfonyl)-3-triflu-
oromethyl-THIQs 14c, 17c, 23c, and 25c are shown in Schemes
1-4. While the synthesis of phenylpropylamine 30 has been
reported previously (7 steps),30 a more direct route is reported
here. Difluoroacetamide (28) was reacted in a P2O5 melt to form
difluoroacetonitrile (29, Scheme 1).31 This highly volatile liquid
was immediately reacted with benzylmagnesium chloride to
form the iminium salt intermediate, which was then reduced in
situ with NaBH4 to yield 30. Phenylpropylamine 30 was treated
with methyl chloroformate in CH2Cl2 and pyridine to afford
carbamate 31. Cyclization of 31 with polyphosphoric acid
yielded lactam 32, the key intermediate in the synthesis of all
of the 3-difluoromethyl-THIQs listed in Table 3. Reduction of
32 with diborane formed 3-difluoromethyl-THIQ (19b). Nitra-
tion of lactam 32 with sulfuric acid and potassium nitrate
afforded the desired regioisomer, 33. Reduction of 33 with
diborane yielded 15b, which was reduced to the aniline and
converted to 11b via a Sandmeyer bromination reaction and to
18b by Greis cyanation.32
Lactam 33 was reduced to the aniline and then converted to
lactam 34 via a Sandmeyer iodination reaction (Scheme 2).33
Treatment of 34 with FSO2CF2CO2CH3 and CuI in DMF yielded
35.34 Compounds 34 and 35 were reduced with diborane to yield
THIQs 13b and 12b, respectively.
The sulfonyl chloride group was introduced exclusively to
the 7-position by reacting lactam 32 with chlorosulfonic acid
(Scheme 3). Compound 36b was then converted to the sulfonyl
hydrazid, and reacted in situ with methyl iodide and sodium
acetate in ethanol to form methyl sulfone 37,35 which was
subsequently reduced with diborane to afford THIQ 20b.
Sulfonyl chloride 36b was also treated with NH4OH in
acetonitrile to produce aminosulfonyl 38, which was reduced
with diborane to yield THIQ 16b.
Sulfonyl chloride 36b or 36c (which was prepared by
previously described methods)23 was treated with the requisite
amine in a either a solution of CHCl3 and pyridine for the
synthesis of sulfonamides 39b, 40b, 45b, 39c, 40c, and 45c or
a biphasic mixture of EtOAc and saturated sodium carbonate
for the synthesis of sulfonamides 41b-44b and 43c (Scheme
4). These lactams were reduced with diborane to afford the
corresponding 3-difluoromethyl-THIQs 14b, 17b, and 21b-
25b and 3-trifluoromethyl-THIQs 14c, 17c, 23c, and 25c.
Biochemistry. In the current study, human PNMT (hPNMT)
with a C-terminal hexahistidine tag was expressed in E. coli.36,37
The radiochemical assay conditions, previously reported for the
bovine enzyme,38 were modified to account for the high binding
affinity of some inhibitors.36,39 Inhibition constants were
determined using four concentrations of phenylethanolamine as
the variable substrate, and three concentrations of inhibitor.
R2-Adrenergic receptor binding assays were performed using
cortex obtained from male Sprague Dawley rats.40 [3H]Clonidine

2942 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10
Grunewald et al.
Table 3. In Vitro Activities of 3-Mono-, Di-, and Trifluoromethyl-7-substituted-THIQs
R3 ) CH2F (a)a
R3 ) CHF2 (b)
R3 ) CF3 (c)b
c,d
c,d
c,d
compd
R7
PNMTc
R 2
sele
PNMTc
R2
sele
PNMTa
R2
sele
11a-c
12a-c
13a-c
14a-c
15a-c
16a-c
17a-c
18a-c
19a-c
20a-c
21a-c
22a-c
23a-c
24a-c
25a-c
Br
CF3
I
0.023
0.030
0.054
0.13
0.15
0.15
0.27
0.80
0.82
1.1
6.4
41
7.1
1200
76
680
140
460
3.8
230
550
610
750
280
1400
130
9200
510
4500
520
570
4.6
210
390
360
290
0.094
0.067
0.20
0.25
0.17
0.68
0.90
3.1
3.4
6.0
3.2
2.1
230
190
2400
2800
3.2
0.98
1.9
9.4
2.3
8.0
15
21
23
41
60
>1000
>1000
>1000
>1000
1400
>1000
>1000
2900
>310
>1000
>530
>110
610
>125
>67
140
>1000
>1000
>1000
>1000
>1000
>1000
150
>5000
>4000
>5800
>1500
>1100
>320
44
SO2NHCH2CF3
NO2
SO2NH2
SO2NH(p-C6H4Cl)
CN
H
SO2CH3
SO2NHEt
SO2NHPr
400
3900
17
95
2500
420
1.4
1.7
2.6
3.4
>1000
>1000
>1000
>1000
>1000
>310
>480
>190
>180
>27
>1000
>17
-
13
-
110
-
-
SO2NH(CH2)3OCH3
SO2NHBu
SO2NH(p-C6H4NO2)
5.3
5.6
37
>1000
-
>77
-
260
>1000
76
>130
7.7
>1000
>9.1
a Data for compounds 11a-25a found in refs 48 and 44. b Data for compounds 11c-13c, 15c, 16c, and 18c-21c found in refs 23 and 24. c Ki values
reported in µM. Standard error of the mean was not greater than 10%. d In vitro activities for the inhibition of [3H]clonidine binding to the R2-adrenoceptor.
e Selectivity R2 Ki /hPNMT Ki.
Scheme 4a
THIQ (11a). Although the R2-adrenoceptor affinity of 7-bromo-
3-difluoromethyl-THIQ (11b) is quite low, it does have greater
R2-adrenoceptor affinity than 7-bromo-3-trifluoromethyl-THIQ
(11c, pKa ≈ 4.3)41 since a higher percentage of the former
compound is protonated at physiological pH. This trend is
consistent for all of the compounds in Table 3. At the
R2-adrenoceptor, the rank order of affinity is CH2F > CHF2 >
CF3.
At PNMT, previous studies have established that 3-fluoro-
methyl-7-substituted-THIQs are highly potent inhibitors,36,42-44
while the corresponding 3-trifluoromethyl-7-substituted-THIQs
are substantially less potent.23,24 The 3-difluoromethyl-7-
substituted-THIQs were found to have PNMT inhibitory poten-
cies that were quite similar to the corresponding 3-fluoromethyl-
7-substituted-THIQs. This is especially evident for the most
potent 3-fluoromethyl-7-substituted-THIQs (11a-17a), which
have hPNMT Ki values of less than 300 nM. For these
compounds, the corresponding 3-difluoromethyl-7-substituted-
THIQs (11b-17b) have Ki values at PNMT ranging from
statistically equipotent to 4-fold (average of 2.4-fold) less potent,
whereas the corresponding 3-trifluoromethyl-7-substituted-
THIQs (11c-17c) are between 15 and 140-fold (average of 58-
fold) less potent. This data trend is consistent for all of the
compounds in Table 3. At PNMT, the rank order of potency is
CH2F ≈ CHF2 > CF3. The following sections analyze the data
trends for the 3-mono-, di-, and trifluoromethyl-THIQs at the
R2-adrenoceptor and PNMT.
r2-Adrenoceptor. The 3-difluoromethyl-THIQs and 3-tri-
fluoromethyl-THIQs show a significant reduction in affinity for
the R2-adrenoceptor relative to the corresponding 3-fluoro-
methyl-THIQs, indicating that either an increase in the size of
the 3-substituent, a decrease in the pKa of the THIQ amine, or
a combination of these factors is responsible for the reduction
in R2-adrenoceptor affinity. To delineate between steric and pKa
effects, the 7-bromo-THIQs (11a-c) and 7-nitro-THIQs (15a-
c) were compared to 7-bromo- and 7-nitro-THIQs bearing
increasingly larger alkyl groups on the 3-position (Table 4).
Analysis of these data suggests that the steric bulk of the
3-substituent does not appear to significantly affect the affinity
of THIQs for the R2-adrenoceptor. Despite considerable differ-
ences in the steric bulk of their 3-substituents [according to the
a Reagents and conditions: (a) amines, pyridine/CHCl3 or EtOAc/
Na2CO3; (b) BH3‚THF.
was used as the radioligand to define the specific binding and
phentolamine was used to define the nonspecific binding.
Clonidine was used as the ligand to define R2-adrenergic binding
affinity to simplify the comparison with previous results.
Results and Discussion
The biochemical data for the 3-difluoromethyl- (11b-25b)
and the 3-trifluoromethyl-7-substituted-THIQs (14c, 17c, 23c,
25c) are shown in Table 3, along with previously reported data
for selected 3-mono- and 3-trifluoromethyl-THIQs. The pKa of
the 3-difluoromethyl-THIQs (pKa ≈ 5.5-6.5)41 is lowered to a
level at which affinity for the R2-adrenoceptor is drastically
reduced. 7-Bromo-3-difluoromethyl-THIQ (11b) has 1000-fold
less affinity for the R2-adrenoceptor than 7-bromo-THIQ (7,
Table 2) and 35-fold less affinity than 7-bromo-3-fluoromethyl-

Design of Potent and SelectiVe Inhibitors of PNMT
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10 2943
Table 4. In Vitro Activities of (()-7-Nitro- and -7-Bromo-3-substituted-THIQs
Ki (µM) ( SEMe
d
f
compd
R7
Br
Br
Br
Br
Br
Br
NO2
NO2
NO2
NO2
NO2
NO2
R3
CH2F
CHF2
CF3
Chartona
Es Taftb
Es′ Duboisc
pKa
PNMT
R2
selectivity R2/hPNMT
11ag
11b
0.62
0.68
0.91
0.52
0.56
0.76
0.62
0.68
0.91
0.52
0.56
0.76
-1.48
-1.91
-2.40
-1.24
-1.31
-1.71
-1.48
-1.91
-2.40
-1.24
-1.31
-1.71
1.32
1.44
1.90
1.12
1.20
1.43
1.32
1.44
1.90
1.12
1.20
1.43
7.77
6.12
4.33
9.29
9.44
9.59
7.45
5.83
4.06
8.95
9.11
9.27
0.023h
0.094
3.2j
6.4
230
>1000
280
2400
>310
65
2.5
0.89
510
11ci
11dh
11eh
11fh
15ag
15b
CH3
0.017
0.48
4.4
1.1
1.2
3.9
CH2CH3
CH(CH3)2
CH2F
CHF2
CF3
CH3
CH2CH3
CH(CH3)2
0.15i
0.17
2.3j
76
>1000
1400
31
>5900
15ci
15dh
15eh
15fh
610
0.072
0.49
4.6
430
57
7.9
28
36
a Ref 45. b Ref 46. c Scaled to H ) 0. Ref 47. d pKa values were calculated using pKaPlugin from the JChem (version 3.1) software package (see ref 41).
e Standard error of the mean. f In vitro activities for the inhibition of [3H]clonidine binding to the R2-adrenoceptor. g Ref 44. h Ref 48. i Ref 23. j Ref 24.
Charton,45 Taft (Es),46 or Dubois (Es′)47 parameters], the 7-alkyl-
3-bromo-THIQs (11d-f) and the 7-alkyl-3-nitro-THIQs (15d-
f) have similar affinity for the R2-adrenoceptor.48 This evidence
alone suggests that changes in steric bulk at the 3-position of
THIQ are not affecting R2-adrenoceptor affinity. This is also
supported by comparing the 3-fluoroalkyl-THIQs with the
3-alkyl-THIQs and is best illustrated by comparing the 3-iso-
propyl-THIQs (11f and 15f) and the 3-difluoromethyl-THIQs
(11b and 15b). Isopropyl and difluoromethyl groups are
geometrically similar, but the isopropyl group is considered to
have steric bulk that is either approximately equal to or greater
than that of the difluoromethyl group. If an increase in the steric
bulk of the 3-substituent is the determining factor in lowering
R2-adrenoceptor affinity, the 3-isopropyl analogues would be
expected to have R2-adrenoceptor affinity that is either ap-
proximately equal to or less than that of the 3-difluoromethyl
analogues. This is not the case. The 3-isopropyl-THIQs 11f and
15f have, respectively, 59 and > 28-fold more affinity for the
R2-adrenoceptor than the 3-difluoromethyl-THIQs 11b and 15b.
These data demonstrate that the reduced R2-adrenoceptor affinity
of the 3-difluoromethyl-THIQs, and by extension the 3-triflu-
oromethyl-THIQs, is likely a result of their reduced pKa values.
It appears that the most pronounced reduction of R2-adrenoceptor
affinity occurs somewhere between a pKa of approximately 8
and 6, the range between the 3-fluoromethyl-THIQs and the
3-difluoromethyl-THIQs.
The R2-adrenonergic receptor data reported in this paper was
obtained from the inhibition of [3H]clonidine binding to rat
cortex R2-adrenoceptors. We recently evaluated a series of
THIQs, which included 3-mono-, di-, and trifluoromethyl-
THIQs, for the inhibition of [3H]RX821002 binding to cloned
human R2A-adrenergic receptors.49 The binding data for this
series of THIQs showed good correlation (r2 ) 0.90) between
the inhibition of [3H]clonidine binding to rat cortex R2-
adrenoceptors and the inhibition of [3H]RX821002 binding to
cloned human R2A-adrenergic receptors. This correlation sup-
ports the use of structural models of the human R2A-adrenergic
receptor to explain the R2-adrenoceptor data trends reported in
this paper.
receptors bind in their protonated form.50-52 Recent studies with
the human HT1D receptor demonstrated that difluorination of
amine ligands significantly lowers their binding affinity, ap-
parently due to a reduction of protonated ligand at physiological
pH.8 We expect that this same phenomenon is resulting in the
low R2-adrenoceptor affinity of the 3-di- and 3-trifluoromethyl-
THIQs, which exist primarily in the neutral form at physiological
pH. Mutagenesis studies52 and rhodopsin-based homology
models53,54 of the human R2A-adrenoceptor predict that Asp113
provides an anchoring point for ligands containing a positively
charged amine group. Structural models of the R2A-adrenoceptor
predict that the positively charged amine of phenylethylamines
and the positively charged nitrogen in the imidazoline ring of
clonidine analogues interact electrostatically with Asp113.52 The
low affinity of the 3-di- and 3-trifluoromethyl-THIQs compared
to 3-fluoromethyl- and 3-methyl-THIQs provides evidence that
interaction between the protonated amine of THIQs and Asp113
is required for optimal R2A-adrenergic receptor affinity.
The binding of aliphatic amines that are primarily neutral at
physiological pH to cationic ligand binding G-protein coupled
receptors has not been thoroughly investigated. The results in
this paper coupled with the previous study of fluorinated HT1D
receptor ligands8 suggest that the fluorination of amine ligands
for other G-protein coupled receptors may also result in reduced
binding affinity. Since the key aspartate residue is highly
conserved in all cationic ligand binding G-protein coupled
receptors,50 lowering the pKa of amine ligands is likely to have
a similar effect. Fluorination of amine ligands having undesired
affinity for G-protein coupled receptors could be useful in the
design of selective pharmacological agents.
PNMT. The rank order of PNMT inhibitory potency for the
3,7-disubstituted-THIQs in Table 3 is CH2F ≈ CHF2 > CF3.
The 3-mono- and difluoromethyl-7-substituted-THIQs have
similarly high inhibitory potency while the corresponding
3-trifluoromethyl-7-substituted-THIQs have drastically reduced
inhibitory potency, which poses an interesting question. Why
does a simple hydrogen to fluorine substitution result in only a
slight change in going from the fluoromethyl series to the
difluoromethyl series, but has a profound effect when going
from the difluoromethyl series to the trifluoromethyl series?
Structure-activity relationship studies on substituted benzyl-
amines have found that electron-withdrawing aromatic substit-
uents enhance PNMT inhibitory potency.55 The inverse relation-
ship observed between the biochemical activity and the pKa of
a series of benzylamines led to the hypothesis that the
R
2A-Adrenoceptors belong to the rhodopsin-like class of
G-protein coupled receptors. Site directed mutagenesis studies
and rhodopsin-based homology models have implicated an
aspartate residue present in all G-protein coupled receptors as
a key amino acid for ammonium ion binding.50,51 Numerous
studies suggest that amine ligands for G-protein coupled

2944 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10
Grunewald et al.
Figure 3. This figure shows 7-bromo-3-fluoromethyl-, difluoromethyl-, and trifluoromethyl-THIQ (11a-c) docked into the active site of hPNMT
[from the hPNMT X-ray structure cocrystallized with AdoHcy (5) and SK&F 64139 (6)] and the amino acid residues that could interact with
11a-c. Nitrogen is blue, oxygen is red, bromine is gray, fluorine is green, carbons of amino acid side chains are cyan, carbons of 11a are orange,
carbons of 11b are yellow, carbons of 11c are red, and hydrogens on the 3-substituents of 11a and 11b are purple. Other hydrogens are not shown
for clarity. (A) This figure shows an overlay of 11b and 11c. 11a is not shown for clarity but the THIQ portion of 11a overlays almost identically
with that of 11b. 11c is shifted approximately 0.3 Å away from Tyr222 relative to 11a and 11b.57 (B) This figure shows the interaction of 11a with
Tyr222, the aromatic ring of which is shown in the plane of the paper. The hydrogens of the fluoromethyl group are positioned in the face of
Tyr222 while the fluorine is moved away from Tyr222. (C) This figure shows the interaction of 11b with Tyr222, the aromatic ring of which is
shown in the plane of the paper. The hydrogen of the difluoromethyl group is positioned in the face of Tyr222 while the two fluorine atoms are
moved away from Tyr222. (D) This figure shows the interaction of 11c with Tyr 222, the aromatic ring of which is shown in the plane of the paper.
One of the fluorine atoms is in the face of Tyr222.
nonprotonated form of benzylamine might be the active inhibitor
species at PNMT.55 However, there were two obvious problems
with this hypothesis. First, the strong correlation between the σ
value of the aromatic substituent and the pKa of the benzyl-
amines suggested that the electron-withdrawing substituents may
facilitate the binding of benzylamines to PNMT not by lowering
the pKa, but by promoting an interaction between an electron
deficient aromatic ring and the PNMT active site. Analysis of
the X-ray crystal structure of hPNMT cocrystallized with THIQ
inhibitors (which are conformationally constrained benzyl-
amines) indicates that a π stacking interaction between the
aromatic ring of THIQs (and by extension benzylamines) and
Phe182 may be source of the enhanced potency of inhibitors
having electron-withdrawing aromatic substituents. Second, even
in the most thorough study of substituted benzylamines, the pKa
range was narrow and under the assay conditions (pH ) 8.0)
all of the compounds studied existed primarily as the protonated
species.55
This study of 3-mono-, di-, and trifluoromethyl-THIQs
compares compounds having a broad range of pKa values
significantly above and below the pH of the assay and is thus
superior for determining the ionization state in which these
compounds bind. Compounds that are either protonated under
the assay conditions, such as 3-methyl-THIQs 11d and 15d
(Table 4), or are primarily neutral under the assay conditions,
such as 3-difluoromethyl-THIQs 11b and 15b are highly potent
inhibitors, indicating that THIQ inhibitors can bind to PNMT
as either the protonated or neutral species. This is not incon-
sistent with the X-ray crystal structures of hPNMT that indicates
that the THIQ amine could interact with Glu219 and Asp267
by either hydrogen bonding in its neutral form or through
electrostatic interactions in its protonated form. The observed
decrease in PNMT inhibitory potency of the 3-trifluoromethyl-
THIQs does not appear to be due to the fact that they are
primarily neutral at physiological pH, since we do not observe
a significant drop in the inhibitory potency of the 3-difluoro-
methyl-THIQs.
It is well established that an increase in steric bulk at the
3-position of THIQ results in a decrease in PNMT inhibitory
potency,48 as illustrated by the biochemical results from a study
of 7-bromo- and 7-nitro-THIQs having increasingly larger alkyl
groups in the 3-position (Table 4, compounds 11d-f and 15d-
f). However, when the 3-fluoroalkyl-7-substituted-THIQs (11a-c
and 15a-c) are compared to the corresponding 3-alkyl-THIQs
(11d-f and 15d-f), a correlation between the steric bulk of
the 3-substituent [according to the Charton,45 Taft (Es),46 or
Dubois (Es′)47 parameters] and PNMT inhibitory potency is not
observed (Table 4).
Docking studies suggest that this is because these steric
parameters measure size only, without taking the geometry of
the group or the electrostatic properties of the 3-substituent into
account. Analysis of 7-bromo-3-mono-, di-, and trifluoromethyl-
THIQ (11a-c) docked into the active site of hPNMT (Figure
3) suggests that the interaction of the 3-substituents of 11a-c
with Tyr222 dictates their relative potencies at hPNMT. The
docking of 7-bromo-3-fluoromethyl-THIQ with hPNMT shows
that the hydrogens on the fluoromethyl group are positioned in
the face of Tyr222 while the fluorine is positioned away (Figure
3B). This result is supported by the cocrystallization of 15a
and 17a with hPNMT, which show that the fluoromethyl groups
of these compounds are positioned in the same manner as
predicted by the docking study.56 A similar result is observed
when 7-bromo-3-difluoromethyl-THIQ (11b) is docked into the
active site of hPNMT (Figure 3C). The hydrogen of the
difluoromethyl group is positioned in the face of Tyr222 while
the two fluorines are positioned away.

Design of Potent and SelectiVe Inhibitors of PNMT
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10 2945
10-10 mass spectrophotometer. Flash chromatography was per-
formed using silica gel 60 (230-400 mesh) supplied by Universal
Adsorbents, Atlanta, Georgia.
When 7-bromo-3-trifluoromethyl-THIQ (11c) is docked into
the active site of hPNMT, a negative steric interaction is
observed between one of the fluorines and Tyr222 (Figure 3D),
resulting in a translational shift of 11c in the binding pocket
approximately 0.3 Å away from Tyr222,57 relative to 11a and
11b (Figure 3A). It is also likely that electrostatic repulsion
between the partially negatively charged fluorine atom and the
π cloud of Tyr222 is contributing to this move away from
Tyr222. All of the X-ray crystal structures reported to date
indicate that it is not energetically favorable for Tyr222 to
change its position in order to avoid this negative interaction
as its position in the active site is stabilized by a hydrogen bond
with Tyr27 and hydrophobic interactions with Leu229 and
Ala224.25-28 Docking studies of 3-mono-, di-, and trifluoro-
methyl-THIQs having various hydrophilic and lipophilic 7-sub-
stituents produced similar results to those obtained from the
docking studies of the 7-bromo analogues in Figure 3. The
3-trifluoromethyl-THIQs are shifted farther away from Tyr222
than the 3-mono- and difluoromethyl-THIQs. This is most likely
preventing the THIQ amine of the 3-trifluoromethyl-THIQs from
making optimal interactions with Glu219. It is also possible that
this shift in the active site could be preventing optimal
π-stacking interactions between the aromatic ring of the
3-trifluoromethyl-THIQs and Phe182. Clarification of this
hypothesis awaits cocrystallization of 3-mono-, di-, and triflu-
oromethyl-THIQs having the same 7-substituent bound to
hPNMT.
Anhydrous methanol and ethanol were used unless stated
otherwise and were prepared by distillation over magnesium. Other
solvents were routinely distilled prior to use. Anhydrous tetrahy-
drofuran (THF) and diethyl ether (Et2O) were distilled from sodium-
benzophenone ketyl. Hexanes refers to the mixture of hexane
isomers (bp 40-70 °C), and brine refers to a saturated solution of
NaCl. All reactions that required anhydrous conditions were
performed under a positive nitrogen or argon flow, and all glassware
was either oven-dried or flame-dried before use. [methyl-3H]AdoMet
and [3H]clonidine were obtained from PerkinElmer (Boston, MA).
Radiochemical Assay of PNMT Inhibitors. A typical assay
mixture consisted of 25 µL of 0.5 M phosphate buffer (pH 8.0),
25 µL of 50 µM unlabeled AdoMet, 5 µL of [methyl-3H]AdoMet,
containing approximately 3 × 105 dpm (specific activity ap-
proximately 15 Ci/mmol), 25 µL of substrate solution (phenyl-
ethanolamine), 25 µL of inhibitor solution, 25 µL of enzyme
preparation (containing 30 ng hPNMT and 25 µg of bovine serum
albumin), and sufficient water to achieve a final volume of 250
µL. After incubation for 30 min at 37 °C, the reaction mixture was
quenched by addition of 250 µL of 0.5 M borate buffer (pH 10.0)
and was extracted with 2 mL of toluene/isoamyl alcohol (7:3). A
1 mL portion of the organic layer was removed, transferred to a
scintillation vial and diluted with cocktail for counting. The mode
of inhibition was ascertained to be competitive in all cases reported
in Tables 2-4 by examination of the correlation coefficients (r2)
for the fit routines as calculated in the Enzyme Kinetics module
(version 1.1) in SigmaPlot (version 7.0).32 While all Ki values
reported were calculated using competitive kinetics, it should be
noted that there was not always a great difference between the r2
values for the competitive model versus the noncompetitive model.
All assays were run in duplicate with 3 inhibitor concentrations
over a 5-fold range. Ki values were determined by a hyperbolic fit
of the data using the Single Substrate-Single Inhibitor routine in
the Enzyme Kinetics module (version 1.1) in SigmaPlot (version
7.0). For inhibitors with apparent IC50 values less than 0.1 µM (as
determined by a preliminary screen of the compounds to be
assayed), the Enzyme Kinetics Tight Binding Inhibition routine was
used to calculate the Ki values.
R2-Adrenoceptor Radioligand Binding Assay. The radioligand
receptor binding assay was performed according to the method of
U’Prichard et al.29 Male Sprague-Dawley rats were decapitated,
and the cortexes were dissected out and homogenized in 20 volumes
(w/v) of ice-cold 50 mM Tris/HCl buffer (pH 7.7 at 25 °C).
Homogenates were centrifuged thrice for 10 min at 50 000g with
resuspension of the pellet in fresh buffer between spins. The final
pellet was homogenized in 200 volumes (w/v) of ice-cold 50 mM
Tris/HCl buffer (pH 7.7 at 25 °C). Incubation tubes containing [3H]-
clonidine (specific activity approximately 55 Ci/mmol, final
concentration 2.0 nM), various concentrations of drugs and an
aliquot of freshly resuspended tissue (800 µL) in a final volume of
1 mL were used. Tubes were incubated at 25 °C for 30 min, and
the incubation was terminated by rapid filtration under vacuum
through GF/B glass fiber filters. The filters were rinsed with three
5 mL washes of ice-cold 50 mM Tris buffer (pH 7.7 at 25 °C).
The filters were counted in vials containing premixed scintillation
cocktail. Nonspecific binding was defined as the concentration of
bound ligand in the presence of 2 µM of phentolamine. All assays
were run in quadruplicate with five inhibitor concentrations over a
16-fold range. IC50 values were determined by a log-probit analysis
of the data and Ki values were determined by the equation Ki )
IC50 /(1 + [clonidine]/KD), as all Hill coefficients were ap-
proximately equal to 1.
Conclusions
By varying the extent of fluorination on a series of highly
potent, but nonselective 3-methyl-THIQ inhibitors of PNMT,
we have dramatically increased their selectivity while maintain-
ing their potency. This increase in selectivity (decrease in R2-
adrenoceptor affinity) is apparently due to a decrease in the pKa
of the THIQ amine resulting from â-fluorination and seems to
be the first successful application of this approach for the design
of selective pharmacological agents. Affinity for the R2-
adrenoceptor is dependent on the extent of fluorination; thus,
the 3-difluoromethyl- and 3-trifluoromethyl-THIQs have much
less affinity for the R2-adrenoceptor than the 3-fluoromethyl-
THIQs. However, increased fluorination also results in a
reduction in PNMT inhibitory potency, apparently due to steric
and electrostatic factors. 3-Difluoromethyl-7-substituted-THIQs
have the proper balance of both steric and pKa properties and
thus have enhanced selectivity versus the corresponding 3-fluo-
romethyl-7-substituted-THIQs and enhanced PNMT inhibitory
potency versus the corresponding 3-trifluoromethyl-7-substituted-
THIQs. The observed superiority of the 3-difluoromethyl-THIQs
can be summed up with a simple porridge analogy from the
fairytale Goldilocks and the Three Bears.29 While the 3-fluo-
romethyl-THIQs were too potent (too hot) at the R2-adrenoceptor
and the 3-trifluoromethyl-THIQs were not potent enough (too
cold) at PNMT, the 3-difluoromethyl-THIQs are just right. They
are both highly potent at PNMT due to minimal negative steric
interaction with Tyr222 and highly selective due to low affinity
for the R2-adrenoceptor.
Experimental Section
All reagents and solvents were of reagent grade or were purified
by standard methods before use. Melting points were determined
in open capillary tubes on a Thomas-Hoover melting point apparatus
calibrated with known compounds. Proton (1H NMR) and carbon
(13C NMR) nuclear magnetic resonance spectra were taken on a
Bruker DRX-400 or a Bruker AM-500 spectrophotometer. High-
resolution mass spectra (HRMS) were obtained on a Ribermag R
Molecular Modeling. Connolly surfaces were generated in
SYBYL on a Silicon Graphics Octane workstation.58 Docking of
the various inhibitors into the PNMT active site was performed
using AutoDock 3.0.59 The default settings for AutoDock 3.0 were
used. The compound to be docked was initially overlayed with the
cocrystallized ligand SK&F 64139 (6) and minimized with the

2946 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10
Grunewald et al.
Tripos force field. The docking of inhibitors into the hPNMT active
site was performed on the R-enantiomer as a previous study on
3-substituted-THIQs indicated that the R-enantiomer is preferred
over the S-enantiomer in the hPNMT active site.60 Figure 3 was
drawn with Molscript61 and Raster3D.62
Hz); HRMS (FAB+) m/z calcd for C10H10F2NO (MH+) 198.0730,
obsd 198.0759.
General Procedure for Lactam Reduction. Synthesis of 11b-
25b, 14c, 17c, 23c, and 25c (selected procedure for 19b). Lactam
32 (130 mg, 0.66 mmol) was dissolved in THF (10 mL) and 1 M
BH3‚THF (4.4 mL, 4.4 mmol) was added. The solution was heated
to reflux for 4 h and cooled to ambient temperature, and MeOH
(15 mL) was added dropwise. The solvent was removed under
reduced pressure and to the remaining residue a solution of MeOH
(15 mL) and 6N HCl (15 mL) was added. The mixture was heated
to reflux for 3 h and the MeOH was removed under reduced
pressure. Water (25 mL) was added to the mixture, which was then
made basic (pH ≈ 10) with 10% NaOH. The basic solution was
extracted with CH2Cl2 (4 × 30 mL) and the combined organic
extracts were dried over anhydrous Na2SO4 The solvent was
removed under reduced pressure to yield the free amine, which
often required purification by flash chromatography eluting with
EtOAc/hexanes. The free amine was dissolved in CH2Cl2 or Et2O
and dry HCl(g) or HBr(g) was bubbled through the solution to form
the hydrochloride or hydrobromide salt, which was recrystallized
from MeOH/CH2Cl2, EtOH/Et2O or EtOH/hexanes.
(()-Methyl N-(1-Difluoro-3-phenyl-prop-2-yl)carbamate (31).
Difluoroacetonitrile (29) was prepared according to a procedure
reported by Swarts.31 Difluoroacetamide (28, 2.30 g, 24.2 mmol)
and P2O5 (5.00 g, 17.6 mmol) were stirred to a homogeneous
mixture in a 50 mL RB flask equipped with a short path distillation
head. Et2O (5 mL) was placed in the collection flask, which was
cooled to -78 °C. The product (29) boils at 22 °C, so a Tygon
tube was fitted to the end of the distillation head and connected to
a pipet to bubble the product directly into the Et2O in the collection
bulb. This solution of difluoroacetonitrile (29) in Et2O was diluted
with additional Et2O (15 mL) and warmed to -40 °C. A solution
of 1 M benzylmagnesium chloride in Et2O (27.0 mL, 27.0 mmol)
was added dropwise over 30 min. The solution was stirred for 30
min at -40 °C and then allowed to warm to -20 °C and stirring
was continued for 1.5 h. A gray precipitate (iminium salt) formed
and the reaction mixture was transferred to a solution of NaBH4
(1.51 g, 40.0 mmol) in MeOH (100 mL) and H2O (1 mL) at -20
°C. The reaction mixture was stirred at -20 °C for 1 h and at 0 °C
for an additional 1 h. 3 N HCl (10 mL) was slowly added to the
reaction mixture, which was then concentrated to approximately
20 mL. 3 N HCl (30 mL) was added and the acidic aqueous mixture
was washed with Et2O (2 × 100 mL). The aqueous solution was
made basic with 4 N NaOH and extracted with Et2O (4 × 75 mL).
The combined organic extracts from the basic aqueous solution were
washed with brine and dried over anhydrous K2CO3. The solvent
was removed under reduced pressure to yield crude 1,1-difluoro-
3-phenyl-2-aminopropane (30) as a dark oily residue (1.12 g, 6.54
mmol.): 1H NMR (400 MHz, CDCl3) δ 7.4-7.28 (m, 5H), 5.71-
5.41 (m, 1H), 3.35-3.15 (m, 1H), 3.00-2.59 (m, 2H). This crude
product was dissolved in dry CH2Cl2 (25 mL) and pyridine (5 mL)
and the solution was cooled to 0 °C. Methyl chloroformate (0.61
mL, 7.9 mmol) was added dropwise and the solution was stirred
overnight at ambient temperature. Thin-layer chromatography in
hexanes/acetone (1:1), with visualization by ninhydrin staining,
showed the presence of starting material so an additional aliquot
of methyl chloroformate (0.40 mL, 5.2 mmol) was added and the
reaction was stirred for 4 h at ambient temperature. Ice-cold water
(25 mL) was added and the mixture was stirred for 30 min. The
mixture was washed with 3 N HCl (3 × 30 mL) and brine (30
mL). The organic phase was dried over anhydrous Na2SO4 and the
solvent was removed under reduced pressure to yield a yellow oil.
The oil was purified by flash chromatography eluting with hexanes/
EtOAc (6:1) followed by recrystallization from EtOAc/hexanes to
yield 31 as white crystals (1.26 g, 5.50 mmol, 23% from 28): mp
(()-3-Difluoromethyl-1,2,3,4-tetrahydroisoquinoline Hydro-
bromide (19b‚HBr). The hydrobromide salt was recrystallized from
EtOH/hexanes to yield 19b‚HBr as white crystals (122 mg, 0.46
1
mmol, 70%). mp 242-244 °C; H NMR (500 MHz, CD3OD) δ
7.37-7.29 (m, 4H), 6.51-6.29 (m, 1H), 4.59-4.49 (m, 1H), 4.22-
4.11 (m, 2H), 3.34-3.18 (m, 2H); 13C NMR (500 MHz, CD3OD)
δ 129.4, 128.8, 128.1, 127.2, 127.1, 126.2, 113.6 (t, J ) 244 Hz),
54.4 (t, J ) 22 Hz), 44.9, 24.5 (t, J ) 4.0 Hz); HRMS (FAB+) m/z
calcd for C10Hl2F2N (MH+) 184.0938, obsd 184.0931. Anal.
(C10H12BrF2N) C, H, N.
(()-3-Difluoromethyl-7-nitro-3,4-dihydroisoquinolin-1-(2H)-
one (33). Lactam 32 (560 mg, 2.84 mmol) was dissolved in
concentrated H2SO4 (8 mL) and cooled to 0 °C. Potassium nitrate
(314 mg, 3.12 mmol) was added in small portions to the solution,
which was stirred overnight at ambient temperature. The mixture
was poured slowly onto ice (50 g) and extracted with CH2Cl2 (4 ×
40 mL). The combined organic extracts were washed with brine
and dried over anhydrous Na2SO4. The solvent was removed under
reduced pressure and the crude product was purified by flash
chromatography eluting with hexanes/EtOAc (3:1). Recrystallization
from CHCl3/hexanes yielded 33 as light yellow crystals (520 mg,
2.15 mmol, 76%). mp 199-201 °C; 1H NMR (500 MHz, DMSO-
d6) δ 8.70 (s, 1H), 8.45 (b, 1H), 8.25-8.23 (m, 1H), 7.46 (d, J )
8.4 Hz, 1H), 5.92-5.69 (m, 1H), 3.96-3.88 (m, 1H), 3.34-3.15
(m, 2H); 13C NMR (500 MHz, DMSO-d6) δ 162.8, 147.2, 143.2,
130.0, 129.1, 126.5, 122.5, 115.0 (t, J ) 246 Hz), 51.5 (t, J ) 24
Hz), 26.2; HRMS (FAB+) m/z calcd for C10H9F2N2O3 (MH+)
243.0581, obsd 243.0573.
(()-3-Difluoromethyl-7-nitro-1,2,3,4-tetrahydroisoquino-
line Hydrochloride (15b‚HCl). Compound 33 (490 mg, 2.02
mmol) was reduced to THIQ 15b according to the general procedure
for lactam reduction. The crude amine was purified by flash
chromatography eluting with hexanes/EtOAc (1:1). The hydrochlo-
ride salt was recrystallized from EtOH/hexanes to yield 15b‚HCl
as white crystals (382 mg, 1.44 mmol, 72%): mp 102-104 °C;
1H NMR (400 MHz, DMSO-d6) δ 10.44 (b, 1H), 8.28 (s, 1H),
8.16-8.14 (m, 1H), 7.60 (d, J ) 8.4 Hz, 1H), 6.70-6.43 (m, 1H),
4.57-4.47 (m, 2H), 4.24-4.18 (m, 1H), 3.36-3.13 (m, 2H); 13C
NMR (500 MHz, DMSO-d6) δ 146.6, 138.8, 131.1, 130.8, 122.6,
122.3, 114.1 (t, J ) 243 Hz), 52.8 (t, J ) 24 Hz), 44.4, 24.8; HRMS
(FAB+) m/z calcd for C10H11F2N2O2 (MH+) 229.0788, obsd
229.0781. Anal. (C10H11ClF2N2O2) C, H, N.
1
83-85 °C; H NMR (400 MHz, CDCl3) δ 7.36-7.23 (m, 5H),
5.83 (t, J ) 55.7 Hz, 1H), 4.84 (b, 1H), 4.26 (b, 1H), 3.65 (s, 3H),
3.05-2.82 (m, 2H); 13C NMR (400 MHz, CDCl3) δ 156.9, 136.2,
129.5, 129.2, 127.5, 114.9 (t, J ) 245 Hz), 54.2 (t, J ) 45 Hz),
52.9, 34.4; HRMS (FAB+) m/z calcd for C11H14F2NO3 (MH+)
230.0993, obsd 230.0991.
(()-3-Difluoromethyl-3,4-dihydroisoquinolin-1-(2H)-one (32).
Carbamate 31 (1.40 g, 6.11 mmol) was dissolved in polyphos-
phoric acid (PPA, 25 g) and heated to 120 °C. After stirring for 2
h, the reaction was poured onto ice water (100 mL) and stirred
vigorously for 5 min. This aqueous mixture was extracted with
EtOAc (4 × 50 mL). The combined organic extracts were washed
with water (50 mL) and brine (50 mL) and dried over anhydrous
Na2SO4. The solvent was removed to yield an off-white solid, which
was purified by flash chromatography eluting with EtOAc/hexanes
(1:1) followed by recrystallization from hexanes to yield 32 as white
(()-3-Difluoromethyl-7-bromo-1,2,3,4-tetrahydroisoquino-
line Hydrochloride (11b‚HCl). THIQ 15b‚HCl (109 mg, 0.413
mmol) in dry EtOH (20 mL) was hydrogenated over 10% Pd/C
(50 mg) for 2.5 h at 50 psi. The suspension was filtered through
Celite and washed with EtOH. This solution was evaporated to
dryness to yield the crude aniline, which was dissolved in a solution
of 48% HBr (1.0 mL) and water (3.0 mL). A solution of sodium
nitrite (32.0 mg, 0.464 mmol) and water (1 mL) was added dropwise
1
crystals (840 mg, 4.26 mmol, 70%): mp 98-100 °C; H NMR
(400 MHz, CDCl3) δ 8.10 (d, J ) 7.6 Hz, 1H), 7.53-7.49 (m,
1H), 7.42-7.38 (m, 1H), 7.26 (d, J ) 7.5 Hz, 1H), 6.91 (b, 1H),
5.88-5.59 (m, 1H), 3.96 (b, 1H), 3.31-3.06 (m, 2H); 13C NMR
(400 MHz, CDCl3) δ 166.2, 135.9, 133.3, 128.5, 128.4, 128.1,
128.0, 115.3 (t, J ) 245 Hz), 53.1 (t, J ) 25 Hz), 27.5 (t, J ) 2.4

Design of Potent and SelectiVe Inhibitors of PNMT
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10 2947
to the HBr solution. After 30 min, excess HNO2 was destroyed by
the addition of urea (25 mg). The diazonium salt solution was added
to a mixture of copper(I) bromide (180 mg, 1.25 mmol), 48% HBr
(2.5 mL) and water (5.0 mL). The reaction was warmed to 75-80
°C and was stirred for 1.5 h. The reaction was stirred overnight at
ambient temperature and then cautiously made basic with a 50%
NaOH. The formation of blue copper salts was observed at this
time. Ethyl acetate (50 mL) was added and the resulting solution
was filtered through Celite and washed with EtOAc (3 × 20 mL).
The organic phase was separated and the aqueous phase was
extracted with EtOAc (3 × 50 mL). The combined organic extracts
were washed with brine and dried over anhydrous K2CO3. The
solvent was removed under reduced pressure to yield a dark oil
which was purified by flash chromatography eluting with hexanes/
EtOAc (1:1). The free amine was dissolved in Et2O and dry HCl
(g) was bubbled through the solution to form the hydrochloride salt,
which was recrystallized from MeOH/Et2O to yield 11b‚HCl as
was cooled to 0 °C and a solution of sodium nitrite (282 mg, 4.09
mmol) in water (3 mL) was added. After stirring for 15 min, excess
HNO2 was destroyed by the addition of urea (20 mg). The
diazonium salt solution was added in small portions to a vigorously
stirred biphasic mixture of CH2Cl2 (25 mL), KI (1.30 g, 29.0 mmol),
CuI (50 mg, 0.26 mmol) and water (8 mL). The reaction mixture
was stirred overnight at ambient temperature. The resulting brown
suspension was diluted with CH2Cl2 (75 mL). The aqueous phase
was removed and the organic phase washed with 10% Na2S2O3(aq)
(3 × 40 mL). The organic phase was dried over anhydrous Na2-
SO4 and the solvent was removed under reduced pressure. The crude
product was purified by flash chromatography eluting with hexanes/
EtOAc (3:1) and recrystallized from CHCl3/hexanes to yield lactam
34 as a white solid (950 mg, 2.94 mmol, 65%) mp 144-146 °C;
1H NMR (500 MHz, CDCl3) δ 8.44 (d, J ) 8.4 Hz, 1H), 7.85-
7.83 (m, 1H), 7.03 (d, J ) 8.0 Hz, 1H), 5.84-5.61 (m, 1H), 4.02-
3.91 (m, 1H), 3.25-3.02 (m, 2H); 13C NMR (500 MHz, CDCl3) δ
163.7, 141.6, 137.0, 134.6, 129.5, 129.3, 114.7 (t, J ) 246 Hz),
92,4, 52.6 (t, J ) 25 Hz), 26.9 (t, J ) 2.8 Hz); HRMS (FAB+) m/z
calcd for C10H9F2N2O3 (MH+) 323.9697, obsd 323.9669.
(()-3-Difluoromethyl-7-iodo-1,2,3,4-tetrahydroisoquinoline Hy-
drochloride (13b‚HCl). Compound 34 (490 mg, 2.02 mmol) was
reduced to THIQ 13b according to the general procedure for lactam
reduction. The crude amine was purified by flash chromatography
eluting with hexanes/EtOAc (3:1). The hydrochloride salt was
recrystallized from MeOH/Et2O to yield 13b‚HCl as white crystals
(112 mg, 0.324 mmol, 89%): mp dec 282-284 °C; 1H NMR (400
MHz, DMSO-d6) δ 10.32 (b, 1H), 7.70 (s, 1H), 7.64 (d, J ) 8.1
Hz, 1H), 7.10 (d, J ) 8.1 Hz, 1H), 6.67-6.39 (m, 1H), 4.39-4.29
(m, 2H), 4.16-4.07 (m, 1H), 3.11-2.95 (m, 2H); 13C NMR (500
MHz, DMSO-d6) δ 136.5, 135.4, 131.6, 131.4, 130.4, 114.2 (t, J
) 243 Hz), 92.7, 53.0 (t, J ) 24 Hz), 43.8, 24.2; HRMS (FAB+)
m/z calcd for C10H11F2IN (MH+) 309.9904, obsd 309.9901. Anal.
(C10H11ClF2IN) C, H, N.
(()-3-Difluoromethyl-7-trifluoromethyl-3,4-dihydroisoquino-
lin-1-(2H)-one (35). To a solution of 34 (370 mg, 1.15 mmol) in
DMF (2 mL) was added methyl 2,2-difluoro-2-(fluorosulfonyl)
acetate (0.32 mL, 2.5 mmol) and CuI (252 mg, 1.32 mmol). The
reaction mixture was stirred at 80 °C for 5 h. The reaction mixture
was filtered through Celite, which was washed with CH2Cl2 (2 ×
25 mL). The filtrate was evaporated under reduced pressure and
the residue was purified by flash chromatography eluting with CH2-
Cl2/Et2O (3:1) to give a mixture of 34 and 35. The mixture was
separated by flash chromatography eluting with hexanes/EtOAc (3:
1) to yield 35 as a white solid (177 mg, 0.67 mmol 58%); 1H NMR
(400 MHz, CDCl3) δ 8.40 (s, 1H), 7.77 (d, J ) 8.0 Hz, 1H), 7.42
(d, J ) 8.0 Hz, 1H), 6.45 (b, 1H), 5.90-5.61 (m, 1H), 4.08-3.96
(m, 1H), 3.39-3.15 (m, 2H); HRMS (FAB+) m/z calcd for C11H9F5-
NO (MH+) 266.0604, obsd 266.0598.
1
white crystals (86 mg, 0.29 mmol, 69%): mp 240-242 °C; H
NMR (400 MHz, DMSO-d6) δ 10.35 (b, 1H), 7.56 (s, 1H), 7.50-
7.48 (m, 1H), 7.27 (d, J ) 8.3 Hz, 1H), 6.68-6.40 (m, 1H), 4.42-
4.32 (m, 2H), 4.17-4.09 (m, 1H), 3.14-2.96 (m, 2H); 13C NMR
(500 MHz, DMSO-d6) δ 131.6, 131.4, 130.8, 130.1, 129.6, 120.0,
114.2 (t, J ) 243 Hz), 53.0 (t, J ) 22 Hz), 44.1, 24.1; HRMS
(FAB+) m/z calcd for C10H11BrF2N (MH+) 262.0043, obsd 262.0035.
Anal. (C10H11BrClF2N) C, H, N.
(()-3-Difluoromethyl-7-cyano-1,2,3,4-tetrahydroisoquino-
line Hydrobromide (18b‚HBr). THIQ 15b‚HCl (150 mg, 0.548
mmol) in dry EtOH (20 mL) was hydrogenated over 10% Pd/C
(50 mg) for 2.5 h at 50 psi. The suspension was filtered through
Celite and washed with EtOH. This solution was evaporated to
dryness to yield the crude aniline, which was dissolved in a solution
of concentrated HCl (1.6 mL) and water (2 mL) and cooled in an
ice bath. A solution of sodium nitrite (41.5 mg, 0.601 mmol) in
water (2 mL) was added dropwise. After 30 min, excess HNO2
was destroyed by the addition of urea (20 mg). In a separate flask,
a solution of NaOH (66 mg) in water (1 mL) and KCN (179 mg)
water (5 mL) was prepared. Benzene (5 mL) was added to the basic
KCN solution and the suspension was cooled to 0 °C. A solution
of Ni2SO4‚6H2O (143 g, 0.55 mmol) and water (1.5 mL) was added
to the basic KCN solution and the color of the solution changed to
yellow-brown. The diazonium salt solution was added dropwise to
this yellow-brown solution. Brisk evolution of N2 was observed
and the reaction mixture was allowed to warm to ambient
temperature over a period of 2 h. The mixture was warmed to 50
°C for 1 h, cooled to ambient temperature, made basic with 10%
NaOH, and filtered through Celite. The Celite was washed with
CH2Cl2 (2 × 100 mL). The organic phase was removed and the
aqueous filtrate was extracted with CH2Cl2 (3 × 50 mL). The
combined organic extracts were washed with brine and dried over
anhydrous K2CO3. Removal of the solvent yielded a dark oil which
was purified by flash chromatography eluting with EtOAc/hexane
(5:1) to yield 18b as a pale brown solid. The free amine was
dissolved in Et2O and dry HBr(g) was bubbled through the solution
to form the hydrobromide salt, which was recrystallized from
MeOH/Et2O to yield 18b‚HBr as off-white crystals (31.1 mg, 0.107
(()-3-Difluoromethyl-7-trifluoromethyl-1,2,3,4-tetrahydroiso-
quinoline Hydrochloride (12b‚HCl). Compound 35 (490 mg, 2.02
mmol) was reduced to THIQ 12b according to the general procedure
for lactam reduction. The crude amine was purified by flash
chromatography eluting with hexanes/EtOAc (3:1). The hydrochlo-
ride salt was recrystallized from MeOH/Et2O to yield 13b‚HCl as
1
1
mmol, 20%) mp 246-248 °C; H NMR (400 MHz, DMSO-d6) δ
white crystals (125 mg, 0.435 mmol, 82%): mp 248-250 °C; H
9.76 (b, 1H), 7.83 (s, 1H), 7.79 (d, J ) 8.0 Hz, 1H), 7.54 (d, J )
8.0 Hz, 1H), 6.63-6.36 (m, 1H), 4.50-4.41 (m, 2H), 4.24-4.17
(m, 1H), 3.30-3.03 (m, 2H); 13C NMR (500 MHz, DMSO-d6) δ
136.6, 131.4, 131.0, 130.6, 130.6, 118.8, 114.3 (t, J ) 243 Hz),
53.0 (t, J ) 23 Hz), 44.5, 25.1; HRMS (FAB+) m/z calcd for
C11H11F2N2 (MH+) 209.0890, obsd 209.0875. Anal. (C11H11BrF2N2)
C, H, N.
(()-3-Difluoromethyl-7-iodo-3,4-dihydroisoquinolin-1-(2H)-
one (34). Lactam 33 (900 mg, 4.56 mmol) in dry MeOH (75 mL)
was hydrogenated over 10% Pd/C (120 mg) for 3.5 h at 50 psi.
The suspension was filtered through Celite and evaporated to
dryness to yield the aniline intermediate as an oily residue. CH2-
Cl2 (50 mL) was added to the oily residue and a white precipitate
formed. A solution of concentrated HCl (1.6 mL) and water (2 mL)
was added, which dissolved the precipitate. This biphasic mixture
NMR (400 MHz, DMSO-d6) δ 10.38 (b, 1H), 7.75 (s, 1H), 7.66
(d, J ) 8.0 Hz, 1H), 7.54 (d, J ) 8.0 Hz, 1H), 6.69-6.42 (m, 1H),
4.46 (s, 2H), 4.26-4.13 (m, 1H), 3.28-3.10 (m, 2H); 13C NMR
(400 MHz, DMSO-d6) δ 137.2, 136.2, 130.3, 129.1 (q, J ) 32.4
Hz), 124.5 (q, J ) 272 Hz), 123.6 (q, J ) 3.6 Hz), 123.4 (q, J )
3.8 Hz), 117.1 (t, J ) 243 Hz), 55.3 (t, J ) 23 Hz), 47.8, 27.7 (t,
J ) 3.7 Hz); HRMS (FAB+) m/z calcd for C11H11F5N (MH+)
252.0812, obsd 252.0809. Anal. (C11H11ClF5N) C, H, N.
(()-7-Chlorosulfonyl-3-difluoromethyl-3,4-dihydroisoquino-
lin-1-(2H)-one (36b). Lactam 32 (1.00 g, 5.07 mmol) was treated
with chlorosulfonic acid (10 mL) and heated to 50 °C for 16 h.
The solution was cooled and carefully added via pipet onto ice
(100 mL). A white precipitate formed. The mixture was extracted
with CH2Cl2 (3 × 50 mL). The combined organic extracts were
washed with brine and dried over anhydrous Na2SO4. The solvent

2948 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10
Grunewald et al.
was removed under reduced pressure to yield a light brown residue,
which was purified by flash chromatography eluting with EtOAc
to yield a white solid. Recrystallization from EtOAc/hexanes yielded
36b as white crystals (1.05 g, 3.55 mmol, 70%) mp 165-167 °C;
1H NMR (400 MHz, CDCl3 δ 8.79 (d, J ) 2.1 Hz, 1H), 8.18-
8.16 (m, 1H), 7.54 (d, J ) 8.1 Hz, 1H), 6.45 (b, 1H), 5.94-5.65
(m, 1H), 4.15-4.02 (m, 1H), 3.46-3.17 (m, 2H); 13C NMR (500
MHz, CDCl3) δ 164.0, 144.5, 137.8, 129.6, 128.4, 127.7, 125.1,
115.2 (t, J ) 245 Hz), 51.6 (t, J ) 24 Hz), 26.2; HRMS (FAB+)
m/z calcd for C10H9ClF2NO3S (MH+) 295.9960, obsd 295.9957.
(()-3-Difluoromethyl-7-methylsulfonyl-3,4-dihydroisoquino-
lin-1-(2H)-one (37). Sulfonyl chloride 36b (228 mg, 0.77 mmol)
was dissolved in THF (10 mL) and cooled to 0 °C. Hydrazine (0.072
mL, 2.32 mmol) was added dropwise to the solution, which was
stirred overnight at ambient temperature. The solution was cooled
to 0 °C and a white precipitate (hydrazidosulfone) was collected
by filtration. The precipitate was dissolved in EtOH (5 mL), then
NaOAc (0.316 g, 3.85 mmol) and iodomethane (0.240 mL, 3.85
mmol) were added. The mixture was heated at reflux overnight.
Water (30 mL) was added and the solution was extracted with
EtOAc (3 × 40 mL). The combined organic extracts were washed
with brine and dried over anhydrous Na2SO4. The solvent was
removed under reduced pressure to yield a white solid, which was
recrystallized from EtOAc/hexane to yield 37 as white needles (150
mg, 0.55 mmol, 71%) mp 215-217 °C; 1H NMR (400 MHz, CDCl3
δ 8.68 (d, J ) 2.0 Hz, 1H), 8.12-8.10 (m, 1H), 7.53 (d, J ) 8.0
Hz, 1H), 6.27 (b, 1H), 5.88-5.65 (m, 1H), 4.06-4.01 (m, 1H),
3.42-3.19 (m, 2H), 3.12 (s, 3H); 13C NMR (500 MHz, CDCl3) δ
163.0, 141.0, 140.6, 131.1, 129.2, 128.9, 127.6, 114.5 (t, J ) 246
Hz), 52.5 (t, J ) 24 Hz), 44.2, 27.3; HRMS (FAB+) m/z calcd for
C11H11F2NO3S (MH+) 276.0506, obsd 276.0492.
(()-3-Difluoromethyl-7-methylsulfonyl-1,2,3,4-tetrahydroiso-
quinoline Hydrochloride (20b‚HCl). Compound 37 (150 mg, 0.55
mmol) was reduced to THIQ 20b according to the general procedure
for lactam reduction. The crude amine was purified by flash
chromatography eluting with hexanes/EtOAc (3:2). The hydrochlo-
ride salt was recrystallized from EtOH/hexanes to yield 20b‚HCl
as white crystals (120 mg, 0.40 mmol, 74%): mp 243-245 °C;
1H NMR (400 MHz, DMSO-d6) δ 10.32 (b, 1H), 7.92 (s, 1H), 7.84
(d, J ) 8.1 Hz, 1H), 7.59 (d, J ) 8.1 Hz, 1H), 6.69-6.42 (m, 1H),
4.53-4.45 (m, 2H), 4.24-4.16 (m, 1H), 3.30-3.11 (m, 2H); 13C
NMR (500 MHz, DMSO-d6) δ 139.8, 136.8, 130.5, 130.5, 126.2,
125.8, 114.2 (t, J ) 243 Hz), 52.9 (t, J ) 23 Hz), 44.5, 43.9, 24.8;
HRMS (FAB+) m/z calcd for C11H14F2NO2S (MH+) 262.0713, obsd
262.0695. Anal. (C11H14ClF2NO2S) C, H, N.
°C; 1H NMR (500 MHz, DMSO-d6) δ 10.30 (b, 1H), 7.76 (s, 1H),
7.73 (d, J ) 8.1 Hz, 1H), 7.50 (d, J ) 8.1 Hz, 1H), 7.43 (b, 2H),
6.67-6.45 (m, 1H), 4.47 (s, 2H), 4.24-4.14 (m, 1H), 3.32-3.09
(m, 2H); 13C NMR (500 MHz, DMSO-d6) δ 143.1, 134.7, 130.0,
129.8, 125.1, 124.2, 114.2 (t, J ) 243 Hz), 53.0 (t, J ) 24 Hz),
44.5, 24.6; HRMS (FAB+) m/z calcd for C10H13F2N2O2S (MH+)
263.0666, obsd 263.0668. Anal. (C10H13ClF2N2O2S) C, H, N.
General Procedure for the Preparation of Compounds 39b,
40b, 45b, 39c, 40c, and 45c. Sulfonyl chloride 36b or 36c (between
0.25 and 1.0 mmol) was dissolved in CHCl3 (20 mL) and cooled
to 0 °C. The requisite amine (3 equiv) and pyridine (3 equiv) were
dissolved in CHCl3 (2 mL) and added dropwise to the sulfonyl
chloride solution. The reaction was stirred for 12 h at ambient
temperature and the solvent was removed under reduced pressure.
The resulting residue was dissolved in EtOAc (50 mL) and this
solution was washed with 3 N HCl (2 × 30 mL) and brine (30
mL) and dried over anhydrous Na2SO4. The solvent was removed
under reduced pressure to yield the crude product, which was
purified by flash chromatography eluting with hexanes/EtOAc (if
required), followed by recrystallization from EtOAc/hexanes or
EtOH/hexanes.
(()-3-Difluoromethyl-7-(N-2,2,2-trifluoroethylaminosulfonyl)-
3,4,-dihydroisoquinolin-1(2H)-one (39b). According to the above
procedure, sulfonyl chloride 36b (170 mg, 0.58 mmol) was
converted to 39b (140 mg, 0.39 mmol, 68%): mp 208-210 °C;
1H NMR (500 MHz, CD3OD) δ 8.43 (d, J ) 2.0 Hz, 1H), 8.01-
7.99 (m, 1H), 7.56 (d, J ) 8.0 Hz, 1H), 6.02-5.79 (m, 1H), 4.07-
4.00 (m, 1H), 3.67 (q, J ) 9.1 Hz, 2H), 3.46-3.22 (m, 2H); 13C
NMR (500 MHz, CD3OD) δ 164.3, 141.4, 140.2, 130.1, 128.9,
128.6, 125.4, 124.0 (q, J ) 278 Hz), 115.2 (t, J ) 244 Hz), 51.5
(t, J ) 24 Hz), 43.5 (q, J ) 35 Hz), 25.8 (t, J ) 3.9 Hz); HRMS
(FAB+) m/z calcd for C12H12F5N2O3S (MH+) 359.0489, obsd
359.0485.
(()-7-(N-2,2,2-Trifluoroethylaminosulfonyl)-3-trifluorometh-
yl-3,4,-dihydroisoquinolin-1(2H)-one (39c). According to the
above procedure, sulfonyl chloride 36c (100 mg, 0.32 mmol) was
converted to 39c (85 mg, 0.23 mmol, 71%): mp 203-205 °C; 1H
NMR (400 MHz, CD3OD) δ 8.60 (s, 1H), 7.98 (d, J ) 8.0 Hz,
1H), 7.57 (d, J ) 8.0 Hz, 1H), 4.36 (t, J ) 8.0 Hz, 1H), 3.64 (q,
J ) 8.0, 2H), 3.57-3.28 (m, 2H); 13C NMR (400 MHz, CD3OD)
δ 165.6, 142.2, 142.0, 131.8, 130.0, 129.9, 127.0, 126.8 (q, J )
281 Hz), 125.6 (q, J ) 281 Hz), 55.1 (q, J ) 33 Hz), 45.1 (q, J )
35 Hz), 27.1.
(()-7-(N-4-Chlorophenylaminosulfonyl)-3-difluoromethyl-
3,4,-dihydroisoquinolin-1-(2H)-one (40b). According to the above
procedure, sulfonyl chloride 36b (200 mg, 0.68 mmol) was
converted to 40b (245 mg, 0.63 mmol, 94%): mp 191-193 °C;
1H NMR (500 MHz, CD3OD) 8.36 (d, J ) 1.8 Hz, 1H), 7.86-
7.84 (m, 1H), 7.47 (d, J ) 8.0 Hz, 1H), 7.24-7.22 (m, 2H), 7.11-
7.09 (m, 2H), 5.98-5.75 (m, 1H), 4.04-3.96 (m, 1H), 3.40-3.16
(m, 2H); 13C NMR (500 MHz, CD3OD) δ 164.2, 141.6, 138.7,
136.1, 130.4, 129.8, 128.8, 128.8, 128.4, 125.7, 122.2, 115.2 (t, J
) 244 Hz), 51.4 (t, J ) 24 Hz), 25.8 (t, J ) 3.8 Hz).
(()-7-Aminosulfonyl-3-difluoromethyl-3,4-dihydroisoquino-
lin-1-(2H)-one (38). Sulfonyl chloride 36b (200 mg, 0.68 mmol)
was dissolved in acetonitrile (5 mL), concentrated ammonium
hydroxide (5 mL) was added, and the reaction was stirred overnight
at ambient temperature. The solvent was removed under reduced
pressure and the residue was partitioned between water (50 mL)
and EtOAc (50 mL). The organic layer was removed and the
aqueous layer was extracted with EtOAc (2 × 30 mL). The organic
layers were pooled, washed with brine, and dried over anhydrous
Na2SO4. The solvent was removed under reduced pressure and the
resulting oil was recrystallized from EtOH/hexanes to yield 38 as
(()-7-(N-4-Chlorophenylaminosulfonyl)-3-trifluoromethyl-
3,4,-dihydroisoquinolin-1-(2H)-one (40c). According to the above
procedure, sulfonyl chloride 36c (300 mg, 0.960 mmol) was
converted to 40c (325 mg, 0.81 mmol, 84%): mp 230-232 °C;
1H NMR (400 MHz, CDCl3) δ 8.68 (s, 1H), 8.34 (br s, 1H), 7.83-
7.81 (m, 1H), 7.32 (d, J ) 8.0 Hz, 1H), 7.19 (d, J ) 6.9 Hz, 2H),
7.09 (d, ) 6.9 Hz, 2H), 7.03 (br s, 1H), 4.33-4.22 (m, 1H), 3.48-
3.17 (m, 2H); 1H NMR (400 MHz, CDCl3) δ 164.4, 140.3, 139.3,
135.4, 132.0, 131.5, 129.8, 128.7, 128.6, 127.6, 124.8 (q, J ) 282
Hz), 123.7, 52.4 (q, J ) 32 Hz), 26.8; HRMS (FAB+) calcd for
C16H13ClF3N2O3S (MH+) 405.0288, obsd 405.0286.
1
white crystals (180 mg, 0.65 mmol, 96%): mp 205-207 °C; H
NMR (500 MHz, CD3OD) δ 8.49 (d, J ) 1.9 Hz, 1H), 8.04-8.02
(m, 1H), 7.53 (d, J ) 8.0 Hz, 1H), 6.01-5.78 (m, 1H), 4.88 (s,
1H), 4.06-4.00 (m, 1H), 3.46-3.20 (m, 2H); 13C NMR (500 MHz,
CD3OD) δ 164.5, 143.0, 140.6, 129.5, 128.6, 128.3, 124.8, 115.3
(t, J ) 244 Hz), 51.5 (t, J ) 24 Hz), 25.8 (t, J ) 3.8 Hz); HRMS
(FAB+) m/z calcd for C10H11F2N2O3S (MH+) 277.0458, obsd
277.0456.
(()-7-Aminosulfonyl-3-difluoromethyl-1,2,3,4-tetrahydroiso-
quinoline Hydrochloride (16b‚HCl). Compound 38 (150 mg, 0.55
mmol) was reduced to THIQ 16b according to the general procedure
for lactam reduction. The crude amine was purified by flash
chromatography eluting with hexanes/EtOAc (3:2). The hydrochlo-
ride salt was recrystallized from EtOH/hexanes to yield 16b‚HCl
as white crystals (120 mg, 0.40 mmol, 74%): mp dec 239-242
(()-3-Difluoromethyl-7-(N-4-nitrophenylaminosulfonyl)-3,4-
dihydroisoquinolin-1-(2H)-one (45b). According to the above
procedure, sulfonyl chloride 36b (175 mg, 0.59 mmol) was
converted to 45b (135 mg, 0.34 mmol, 57%): mp dec 192-194
°C; 1H NMR (500 MHz, CD3OD) 8.45 (d, J ) 2.0 Hz, 1H), 8.16-
8.13 (m, 2H), 8.03-8.01 (m, 1H), 7.52 (d, J ) 8.1 Hz, 1H), 7.35-
7.32 (m, 2H), 5.98-5.75 (m, 1H), 4.05-3.95 (m, 1H), 3.41-3.16

Design of Potent and SelectiVe Inhibitors of PNMT
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10 2949
(m, 2H); 13C NMR (500 MHz, CD3OD) δ.164.1, 143.7, 143.4,
142.1, 138.7, 130.4, 129.0, 128.7, 125.7, 124.7, 118.2, 115.2 (t, J
) 244 Hz), 51.4, 25.8; HRMS (FAB+) m/z calcd for C16H14F2N3O5S
(MH+) 398.0622, obsd 398.0624.
(()-3-Difluoromethyl-7-(N-butylaminosulfonyl)-3,4-dihydroiso-
quinolin-1-(2H)-one (44b). According to the above procedure,
sulfonyl chloride 36b (180 mg, 0.61 mmol) was converted to 44b
(190 mg, 0.57 mmol, 94%): mp 194-196 °C; 1H NMR (500 MHz,
CD3OD) δ 8.41 (d, J ) 1.9 Hz, 1H), 7.99-7.97 (m, 1H), 7.56 (d,
J ) 8.0 Hz, 1H), 6.02-5.79 (m, 1H), 4.07-3.99 (m, 1H), 3.46-
3.21 (m, 2H), 2.88 (t, J ) 7.0 Hz, 2H), 1.45-1.41 (m, 2H), 1.34-
1.30 (m, 2H), 0.88 (t, J ) 7.3 Hz, 3H); 13C NMR (500 MHz,
CD3OD) δ 164.5, 141.0, 140.0, 130.3, 128.8, 128.4, 125.5, 115.3
(q, J ) 244 Hz), 51.5 (t, J ) 24 Hz), 42.3, 31.3, 25.8 (t, J ) 3.8
Hz), 19.2, 12.4; HRMS (FAB+) m/z calcd for C14H19F2N2O3S
(MH+) 333.1084, obsd 333.1086.
(()-7-(N-4-Nitrophenylaminosulfonyl)-3-trifluoromethyl-3,4-
dihydroisoquinolin-1-(2H)-one (45c). According to the above
procedure, sulfonyl chloride 36c (300 mg, 0.96 mmol) was
converted to 45c (370 mg, 0.89 mmol, 93%): mp 118-120 °C;
1H NMR (400 MHz, CD3OD) δ 8.44-7.65(m, 4H), 7.54 (d, J )
8.0 Hz, 1H), 7.32 (m, 2H), 4.36 (m, 1H), 3.60-3.25 (m, 2H); 13
C
NMR (400 MHz, CD3OD) δ 164.2, 142.5, 141.9, 131.6, 131.0,
129.3, 129.0, 127.0, 126.8 (q, J ) 281 Hz), 126.1, 125.2, 118.7,
51.0 (q, J ) 31 Hz), 26.0; HRMS (FAB+) m/z calcd for
C16H12F3N3O5S (MH+) 416.0528, obsd 416.0536.
(()-3-Difluoromethyl-7-(N-2,2,2-trifluoroethylaminosulfonyl)-
1,2,3,4-tetrahydroisoquinoline Hydrochloride (14b‚HCl). Com-
pound 39b (150 mg, 0.55 mmol) was reduced to THIQ 14b
according to the general procedure for lactam reduction. The
hydrochloride salt was recrystallized from EtOH/hexanes to yield
14b‚HCl as white crystals (120 mg, 0.40 mmol, 74%): mp 202-
General Procedure for the Preparation of Compounds 41b-
44b and 43c. Sulfonyl chloride 36b or 36c (between 0.5 and 1.0
mmol) was dissolved in a biphasic mixture of EtOAc (15 mL) and
saturated Na2CO3 (10 mL). The requisite amine (3 equiv) was added
to the reaction, and the mixture was stirred for 4 h. The organic
phase was removed, washed with 3 N HCl (3 × 50 mL) and brine
(30 mL), and dried over anhydrous Na2SO4. The solvent was
removed under reduced pressure to yield the crude product, which
was purified by flash chromatography (if required) eluting with
hexanes/EtOAc, followed by recrystallization from EtOAc/hexanes
or EtOH/hexanes.
(()-3-Difluoromethyl-7-(N-ethylaminosulfonyl)-3,4-dihydroiso-
quinolin-1-(2H)-one (41b). According to the above procedure,
sulfonyl chloride 36b (180 mg, 0.61 mmol) was converted to 41b
(160 mg, 0.53 mmol, 86%): mp 199-201 °C; 1H NMR (500 MHz,
CD3OD) δ 8.42 (d, J ) 1.9 Hz, 1H), 7.99-7.97 (m, 1H), 7.56 (d,
J ) 8.0 Hz, 1H), 6.02-5.79 (m, 1H), 4.08-3.98 (m, 1H), 3.46-
3.21 (m, 2H), 2.93 (q, J ) 7.3 Hz, 2H), 1.08 (t, J ) 7.3 Hz, 3H);
13C NMR (500 MHz, CD3OD) δ 164.4, 141.0, 140.0, 130.3, 128.8,
128.5, 125.5, 115.2 (q, J ) 244 Hz), 51.5 (t, J ) 24 Hz), 37.5,
25.8 (t, J ) 3.9 Hz), 13.8; HRMS (FAB+) m/z calcd for
C12H15F2N2O3S (MH+) 305.0771, obsd 305.0753.
1
204 °C; H NMR (500 MHz, CD3OD) δ 7.84 (s, 2H), 7.84-7.83
(m, 1H), 7.56 (d, J ) 8.7 Hz, 1H), 7.43 (b, 2H), 6.53-6.31 (m,
1H), 4.63 (s, 2H), 4.26-4.18 (m, 1H), 3.69 (q J ) 9.2 Hz, 1H),
3.42-3.23 (m, 2H); 13C NMR (500 MHz, CD3OD) δ 140.3, 134.8,
129.9, 128.6, 126.0, 124.9, 124.0 (q, J ) 278 Hz), 113.4 (t, J )
244 Hz), 53.9 (t, J ) 23 Hz), 44.6, 43.5 (q J ) 35 Hz), 24.6 (t, J
) 4.0 Hz); HRMS (FAB+) m/z calcd for C12H14F5N2O2S (MH+)
345.0696, obsd 345.0687. Anal. (C12H14ClF5N2O2S) C, H, N.
(()-7-(N-2,2,2-Trifluoroethylaminosulfonyl)-3-trifluorometh-
yl-1,2,3,4-tetrahydroisoquinoline Hydrochloride (14c‚HCl). Com-
pound 39c (310 mg, 0.83 mmol) was reduced to THIQ 14c
according to the general procedure for lactam reduction. The
hydrochloride salt was recrystallized from EtOH/Et2O to yield 14c‚
HCl as white crystals (200 mg, 0.51 mmol, 61%): mp dec 230-
232 °C; 1H NMR (400 MHz, CD3OD) δ 7.80-7.77 (m, 2H), 7.52
(d, J ) 8.0 Hz, 1H), 4.72-4.60 (m, 1H), 4.63 (s, 2H), 3.61 (q, J
) 9.0 Hz, 2H), 3.50-3.32 (m, 2H); 13C NMR (400 MHz, CD3-
OD) δ 142.3, 135.6, 131.8, 130.1, 128.0, 126.7, 125.9 (q, J ) 280
Hz), 125.2 (q, J ) 280 Hz), 55.3 (q, J ) 33 Hz), 46.8, 45.4 (q, J
) 35 Hz), 26.3; HRMS (FAB+) m/z calcd for C12H13F6N2O2S
(MH+) 363.0602, obsd 363.0602; Anal. (C12H13ClF6N2O2S) C, H,
N.
(()-3-Difluoromethyl-7-(N-propylaminosulfonyl)-3,4-dihy-
droisoquinolin-1-(2H)-one (42b). According to the above proce-
dure, sulfonyl chloride 36b (180 mg, 0.61 mmol) was converted
1
to 42b (155 mg, 0.49 mmol, 80%): mp 197-199 °C; H NMR
(500 MHz, CD3OD) δ 8.42 (d, J ) 1.9 Hz, 1H), 7.99-7.97 (m,
1H), 7.56 (d, J ) 8.0 Hz, 1H), 6.02-5.79 (m, 1H), 4.07-3.99 (m,
1H), 3.46-3.21 (m, 2H), 2.84 (m, J ) 7.0 Hz, 2H), 1.52-1.45
(m, 2H), 0.88 (t, J ) 7.4 Hz, 3H); 13C NMR (500 MHz, CD3OD)
δ 164.5, 140.9, 140.1, 130.2, 128.8, 128.4, 125.5, 115.3 (q, J )
244 Hz), 51.5 (t, J ) 24 Hz), 44.4, 25.8 (t, J ) 3.8 Hz), 22.5, 9.9;
HRMS (FAB+) m/z calcd for C13H17F2N2O3S (MH+) 319.0928,
obsd 319.0931.
(()-3-Difluoromethyl-7-(N-3-methoxypropylaminosulfonyl)-
3,4-dihydroisoquinolin-1-(2H)-one (43b). According to the above
procedure, sulfonyl chloride 36b (180 mg, 0.61 mmol) was
converted to 43b (185 mg, 0.53 mmol, 87%): mp 143-145 °C;
1H NMR (500 MHz, CD3OD) δ 8.42 (d, J ) 1.8 Hz, 1H), 7.99-
7.97 (m, 1H), 7.56 (d, J ) 8.0 Hz, 1H), 6.02-5.79 (m, 1H), 4.08-
4.00 (m, 1H), 3.46-3.21 (m, 2H), 3.39 (t, J ) 6.1 Hz, 2H), 3.28,
(s, 3H), 2.96 (t, J ) 6.9 Hz, 2H), 1.73-1.68 (m, 2H); 13C NMR
(500 MHz, CD3OD) δ 164.4, 141.0, 139.9, 130.3, 128.8, 128.5,
125.5, 115.2 (q, J ) 244 Hz), 69.2, 57.3, 51.5 (t, J ) 24 Hz), 46.6,
29.2, 25.8 (t, J ) 3.8 Hz); HRMS (FAB+) m/z calcd for
C14H19F2N2O4S (MH+) 349.1033, obsd 349.1047.
(()-7-(N-3-Methoxypropylaminosulfonyl)-3-trifluoromethyl-
3,4,-dihydroisoquinolin-1(2H)-one (43c). According to the above
procedure, sulfonyl chloride 36c (320 mg, 1.02 mmol) was
converted to 43c (345 mg, 0.94 mmol, 92%): mp 146-147 °C;
1H NMR (400 MHz, CD3OD) δ 8.49 (s, 1H), 7.99 (d, J ) 8.0 Hz,
1H), 7.57 (d, J ) 8.0 Hz, 1H), 4.48-4.37 (m, 1H), 3.63-3.38 (m,
2H), 3.39-3.36 (m, 2H), 3.27 (s, 3H), 2.95 (t, J ) 6.8 Hz, 2H),
1.73-1.67 (m, 2H); 13C NMR (500 MHz, CD3OD) δ 164.1, 140.4,
140.1, 130.6, 128.6, 128.4, 125.6, 124.7 (q, J ) 283 Hz), 69.2,
57.4, 51.0 (q, J ) 31 Hz), 39.8, 29.3, 25.6; HRMS (FAB+) m/z
calcd for C14H18F3N2O4S (MH+) 367.0939, obsd 367.0943.
(()-7-(N-4-Chlorophenylaminosulfonyl)-3-difluoromethyl-
1,2,3,4-tetrahydroisoquinoline Hydrochloride (17b‚HCl). Com-
pound 40b (180 mg, 0.47 mmol) was reduced to THIQ 17b
according to the general procedure for lactam reduction. The
hydrochloride salt was recrystallized from EtOH/hexanes to yield
17b‚HCl as white crystals (150 mg, 0.37 mmol, 79%): mp dec
250-252 °C; 1H NMR (500 MHz, CD3OD) δ 7.76-7.74 (m, 2H),
7.49 (d, J ) 8.6 Hz, 1H), 7.23-7.21 (m, 2H), 7.12-7.10 (m, 2H),
6.48-6.26 (m, 1H), 4.54 (s, 2H), 4.19-4.09 (m, 1H), 3.36-3.16
(m, 2H); 13C NMR (500 MHz, CD3OD) δ 138.7, 136.1, 135.1,
129.9, 129.6, 128.8, 126.4, 125.4, 121.9, 113.4 (t, J ) 244 Hz),
53.9 (t, J ) 22 Hz), 44.6, 24.6; HRMS (FAB+) m/z calcd for C16H16-
ClF2N2O2S (MH+) 373.0589, obsd 373.0574. Anal. (C16H16-
Cl2F2N2O2S) C, H, N.
(()-7-(N-4-Chlorophenylaminosulfonyl)-3-trifluoromethyl-
1,2,3,4-tetrahydroisoquinoline Hydrochloride (17c‚HCl). Com-
pound 40c (325 mg, 0.804 mmol) was reduced to THIQ 17c
according to the general procedure for lactam reduction. The
hydrochloride salt was recrystallized from EtOH/Et2O to yield 17c‚
HCl as white crystals (288 mg, 0.67 mmol, 84%): mp 162-165
1
°C; H NMR (500 MHz, DMSO-d6) δ 7.78 (s, 1H), 7.60 (d, J )
8.2 Hz, 1H), 7.50 (d, J ) 8.2 Hz, 1H), 7.31 (d, J ) 8.8 Hz, 2H),
7.15 (d, J ) 8.8 Hz, 2H), 4.72-4.61 (m, 1H), 4.47-4.39 (m, 2H),
3.49 (b, 3H), 3.34-3.13 (m, 2H); 13C NMR (500 MHz, DMSO-
d6) δ 138.4, 137.0, 135.7, 130.8, 130.4, 129.6, 128.5, 125.8, 125.4,
124.1 (q, J ) 282 Hz), 121.8, 52.6 (q, J ) 32 Hz), 44.9, 25.2;
HRMS (FAB+) m/z calcd for C16H15ClF3N2O2S (MH+) 391.0495,
obsd 391.0468; Anal. (C16H15Cl2F3N2O2S) C, H, N.
(()-3-Difluoromethyl-7-(N-4-nitrophenylaminosulfonyl)-1,2,3,4-
tetrahydroisoquinoline Hydrochloride (25b‚HCl). Compound
45b (115 mg, 0.29 mmol) was reduced to THIQ 25b according to

2950 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10
Grunewald et al.
the general procedure for lactam reduction. The hydrochloride salt
was recrystallized from EtOH/hexanes to yield 25b‚HCl as white
crystals (83 mg, 0.20 mmol, 68%): mp dec 250-252 °C; 1H NMR
(500 MHz, CD3OD) δ 8.14-8.11 (m, 2H), 7.91 (s, 1H), 7.88-
7.86 (m, 1H), 7.53 (d, J ) 8.2 Hz, 1H), 7.35-7.33 (m, 2H), 6.49-
6.28 (m, 1H), 4.60 (s, 2H), 4.22-4.12 (m, 1H), 3.38-3.17 (m,
2H); 13C NMR (500 MHz, CD3OD) δ 143.8, 143.3, 138.6, 135.6,
130.1, 128.9, 126.4, 125.5, 124.7, 118.1, 113.3 (t, J ) 244 Hz),
53.9 (t, J ) 23 Hz), 44.5, 24.5; HRMS (FAB+) m/z calcd for
C16H16F2N3O4S (MH+) 384.0830, obsd 384.0837. Anal. (C16H16-
ClF2N3O4S) C, H, N.
(()-7-(N-4-Nitrophenylaminosulfonyl)-3-trifluoromethyl-1,2,3,4-
tetrahydroisoquinoline Hydrochloride (25c‚HCl). Compound 45c
(350 mg, 0.85 mmol) was reduced to THIQ 25c according to the
general procedure for lactam reduction. The hydrochloride salt was
recrystallized from EtOH/Et2O to yield 25c‚HCl as white crystals
(210 mg, 0.48 mmol, 57%): mp 235-236 °C; 1H NMR (400 MHz,
CD3OD) δ 8.14-8.12 (m, 2H), 7.93-7.88 (m, 2H), 7.55 (d, J )
8.0 Hz, 1H), 7.35-7.33 (m, 2H), 4.67-4.61 (m, 3H), 3.50-3.28
(m, 2H); 13C NMR (500 MHz, CD3OD) δ 143.8, 143.4, 138.8,
134.7, 130.2, 128.6, 126.6, 125.6, 124.7, 123.3 (q, J ) 279 Hz),
118.2, 53.2 (q, J ) 32 Hz), 45.0, 24.5; HRMS (FAB+) m/z calcd
for C16H15F3N3O4S (MH+) 402.0735, obsd 402.0723; Anal. (C16H15-
ClF3N3O4S‚EtOH) C, H, N.
(()-3-Difluoromethyl-7-(ethylaminosulfonyl)-1,2,3,4-tetrahy-
droisoquinoline Hydrochloride (21b‚HCl). Compound 41b (140
mg, 0.46 mmol) was reduced to THIQ 21b according to the general
procedure for lactam reduction. The hydrochloride salt was
recrystallized from EtOH/hexanes to yield 21b‚HCl as white crystals
(146 mg, 0.45 mmol, 97%): mp 198-200 °C; 1H NMR (500 MHz,
DMSO-d6) δ 7.76 (s, 1H), 7.69 (d, J ) 8.1 Hz, 1H), 7.65 (t, J )
5.7 Hz, 1H), 7.53 (d, J ) 8.1 Hz, 1H), 6.66-6.44 (m, 1H), 4.52-
4.45 (m, 2H), 4.25-4.14 (m, 1H), 3.26-3.10 (m, 2H), 2.81-2.76
(m, 2H), 1.00 (t, J ) 7.2 Hz, 3H); 13C NMR (500 MHz, DMSO-
d6) δ 139.6, 135.3, 130.3, 130.2, 125.8, 125.3, 114.2 (t, J ) 243
Hz), 53.0 (t, J ) 24 Hz), 44.5, 37.9, 24.7, 15.2; HRMS (FAB+)
m/z calcd for C12H17F2N2O2S (MH+) 291.0979, obsd 291.0956.
Anal. (C12H17ClF2N2O2S) C, H, N.
(()-3-Difluoromethyl-7-(propylaminosulfonyl)-1,2,3,4-tet-
rahydroisoquinoline Hydrochloride (22b‚HCl). Compound 42b
(140 mg, 0.44 mmol) was reduced to THIQ 22b according to the
general procedure for lactam reduction. The hydrochloride salt was
recrystallized from EtOH/hexanes to yield 22b‚HCl as white crystals
(135 mg, 0.40 mmol, 90%): mp 186-188 °C; 1H NMR (500 MHz,
DMSO-d6) δ 7.75 (s, 1H), 7.70-7.68 (m, 1H), 7.66 (t, J ) 5.9 Hz,
1H), 7.53 (d, J ) 8.1 Hz, 1H), 6.65-6.42 (m, 1H), 4.52-4.45 (m,
2H), 4.25-4.13 (m, 1H), 3.27-3.08 (m, 2H), 2.72-2.68 (m, 2H),
1.43-1.36 (m, 2H), 0.81 (t, J ) 7.4 Hz, 3H); 13C NMR (500 MHz,
DMSO-d6) δ 139.7, 135.2, 130.3, 130.2, 125.8, 125.3, 114.2 (t, J
) 243 Hz), 53.0 (t, J ) 24 Hz), 44.7, 44.6, 24.7, 22.8, 11.5; HRMS
(FAB+) m/z calcd for C13H19F2N2O2S (MH+) 305.1135, obsd
305.1137. Anal. (C13H19ClF2N2O2S) C, H, N.
from EtOH/hexanes to yield 23c as white crystals (297 mg, 0.84
mmol, 77%): mp 95-96 °C; 1H NMR (400 MHz, CD3OD) δ 7.65
(d, J ) 8.0 Hz, 1H), 7.61 (s, 1H), 7.40 (d, J ) 8.0, 1H), 4.15-
4.06 (m, 2H), 3.73-3.64 (m, 1H), 3.39 (t, J ) 6.0 Hz, 2H), 3.28
(s, 3H), 3.14-2.94 (m, 2H), 2.91 (t, J ) 6.9 Hz, 2H), 1.72-1.66
(m, 2H); 13C NMR (500 MHz, CD3OD) δ 138.4, 137.2, 136.1,
129.6, 125.9 (q, J ) 279 Hz), 124.5, 124.2, 69.3, 57.3, 54.3 (q, J
) 29), 46.3, 39.7, 29.3, 26.8; HRMS (FAB+) m/z calcd for
C14H20F3N2O3S (MH+) 353.1147, obsd 353.1136; Anal. (C14H19-
F3N2O3S) C, H, N.
(()-7-(N-Butylaminosulfonyl)-3-difluoromethyl-1,2,3,4-tet-
rahydroisoquinoline Hydrochloride (24b‚HCl). Compound 44b
(175 mg, 0.53 mmol) was reduced to THIQ 24b according to the
general procedure for lactam reduction. The hydrochloride salt was
recrystallized from EtOH/hexanes to yield 24b‚HCl as white crystals
(170 mg, 0.48 mmol, 91%): mp 206-208 °C; 1H NMR (500 MHz,
DMSO-d6) δ 7.75 (s, 1H), 7.70-7.68 (m, 1H), 7.65 (t, J ) 5.8 Hz,
1H), 7.53 (d, J ) 8.2 Hz, 1H), 6.64-6.42 (m, 1H), 4.52-4.45 (m,
2H), 4.25-4.14 (m, 1H), 3.26-3.10 (m, 2H), 2.75-2.71 (m, 2H),
1.40-1.34 (m, 2H), 1.29-1.21 (m, 2H), 0.82 (t, J ) 7.3 Hz, 3H);
13C NMR (500 MHz, DMSO-d6) δ 139.6, 135.2, 130.3, 130.2,
125.7, 125.3, 114.2 (t, J ) 243 Hz), 53.0 (t, J ) 23 Hz), 44.6,
42.6, 31.5, 24.7, 19.6, 13.9; HRMS (FAB+) m/z calcd for
C14H21F2N2O2S (MH+) 319.1292, obsd 319.1297. Anal. (C14H21-
ClF2N2O2S) C, H, N.
Acknowledgment. This research was supported by NIH
Grant HL 34193 and NIH Predoctoral Training Grant GM 07775
(M.R.S.). M.R.S. is also an American Foundation for Pharma-
ceutical Education predoctoral fellow. We thank David Vander-
Velde and Sarah Neuenswander of the University of Kansas
Nuclear Magnetic Resonance Laboratory for their assistance.
The 500 MHz NMR spectrometer was partially funded by the
National Science Foundation Grant CHE-9977422. We thank
Gerald Lushington of the University of Kansas Molecular
Graphics and Modeling Laboratory and Todd Williams of the
University of Kansas Mass Spectrometry Laboratory for their
assistance. We also thank Ernst Scho¨nbrunn for assistance with
molecular modeling.
Supporting Information Available: All elemental analyses (C,
H, N) for assayed compounds are included. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) The contents of this paper were taken in large part from the Ph. D.
dissertation (University of Kansas, 2006) of Mitchell R. Seim.
(2) Zuccotto, F. Pharmacophore Features Distributions in Different
Classes of Compounds. J. Chem. Inf. Comput. Sci. 2003, 43, 1542-
1552.
(3) Fuller, R. W.; Roush, B. W. Influence of pKa on the Methylation of
Arylalkylamines by Rabbit Lung N-Methyltransferase. Res. Commun.
Chem. Pathol. Pharmacol. 1975, 10, 735-738.
(()-3-Difluoromethyl-7-(N-3-methoxypropylaminosulfonyl)-
1,2,3,4-tetrahydroisoquinoline Hydrochloride (23b‚HCl). Com-
pound 43b (160 mg, 0.46 mmol) was reduced to THIQ 23b
according to the general procedure for lactam reduction. The
hydrochloride salt was recrystallized from EtOH/hexanes to yield
23b‚HCl as white crystals (150 mg, 0.40 mmol, 88%): mp 160-
(4) Fuller, R. W.; Molloy, B. B.; Parli, C. J. Effect of â,â-Difluoro
Substitution on the Metabolism and Pharmacology of Amphetamines.
Psychopharmacol., Sex. Disord. Drug Abuse, Proc. Symp. Congr.
Coll. Int. Neuro-Psychopharmacol., 8th, 1973; 615-624.
(5) Upthagrove, A. L.; Nelson, W. L. Importance of Amine pKa and
Distribution Coefficient in the Metabolism of Fluorinated Propranolol
Analogues: Metabolism by CYP1A2. Drug Metab. Dispos. 2001,
29, 1389-95.
(6) Upthagrove, A. L.; Nelson, W. L. Importance of Amine pKa and
Distribution Coefficient in the Metabolism of Fluorinated Propranolol
Derivatives. Preparation, Identification of Metabolite Regioisomers,
and Metabolism by CYP2D6. Drug Metab. Dispos. 2001, 29, 1377-
88.
(7) The authors report that fluorination led to a decrease in metabolism
via CYP2D6. As the potency of these compounds was not reported,
it is not possible to ascertain the net effect of fluorination on their
selectivity. Boulet, S. L.; Mathes, B. M.; Hudziak, K. J.; Boot, J. R.;
Clugery, A. H.; Findlay, J. D.; Lavis, A. A.; Mahadevan, S.; Wallace,
L.; Filla, S. A. Effect of hydroxylation and fluorination of aryloxy
1
162 °C; H NMR (500 MHz, DMSO-d6) δ 7.75 (s, 1H), 7.70-
7.67 (m, 2H), 7.53 (d, J ) 8.2 Hz, 1H), 6.65-6.43 (m, 1H), 4.52-
4.45 (m, 2H), 4.25-4.14 (m, 1H), 3.29 (t, J ) 6.1 Hz, 2H), 3.27-
3.09 (m, 2H), 3.17 (s, 3H), 2.80-2.78 (m, 2H), 1.64-1.59 (m,
2H); 13C NMR (500 MHz, DMSO-d6) δ 139.5, 135.3, 130.3, 130.2,
125.8, 125.3, 114.2 (t, J ) 243 Hz), 69.3, 58.3, 53.0 (t, J ) 24
Hz), 44.6, 29.6, 24.7; HRMS (FAB+) m/z calcd for C14H21F2N2O3S
(MH+) 335.1241, obsd 335.1245. Anal. (C14H21ClF2N2O3S) C, H,
N.
(()-7-(N-3-Methoxypropylaminosulfonyl)-3-trifluoromethyl-
1,2,3,4-tetrahydroisoquinoline (23c). Compound 43c (400 mg,
1.09 mmol) was reduced to THIQ 23c according to the general
procedure for lactam reduction. The free amine was recrystallized

Design of Potent and SelectiVe Inhibitors of PNMT
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10 2951
phenylpropanamine serotonin and norepinephrine reuptake inhibitors
on CYP450 2D6 dependent metabolism. Abstracts of Papers, 228th
American Chemical Society National Meeting, Philadelphia, PA,
August 22-26, 2004, MEDI-058.
(27) Wu, Q.; Gee, C. L.; Lin, F.; Tyndall, J. D.; Martin, J. L.; Grunewald,
G. L.; McLeish, M. J. Structural, Mutagenic, and Kinetic Analysis
of the Binding of Substrates and Inhibitors of Human Phenyletha-
nolamine-N-Methyltransferase. J. Med. Chem. 2005, 48, 7243-7252.
(28) Gee, C. L.; Tyndall, J. D.; Grunewald, G. L.; Wu, Q.; McLeish, M.
J.; Martin, J. L. Binding mode of methyl acceptor substrates to the
adrenaline-synthesizing enzyme phenylethanolamine N-methyltrans-
ferase: implications for catalysis. Biochemistry 2005, 44, 16875-
16885.
(29) The story of the Three Bears and their intruder was first published
by Robert Southey in 1837. The name Goldilocks was first used in
the book Old Nursery Stories and Rhymes, by John Hassall in 1904.
In this fairytale, Goldilocks goes into the kitchen of the three bears
where she sees three bowls of porridge. She first tastes Papa Bear’s
porridge but it is too hot. She then tastes Mama Bear’s porridge but
it is too cold. She then tastes Baby Bear’s porridge and it is neither
too hot nor too cold, but just right.
(30) Bey, P.; Schirlin, D. General approach to the synthesis of R-difluo-
romethylamines as potential enzyme-activated irrevesible inhibitors.
Tetrahedron Lett. 1978, 52, 5225-5228.
(31) Swarts, F. Sur les nitriles des acides fluor- et difluoracetiques. Bull.
Soc. Chim. Belg. 1922, 31, 364-365.
(32) McRae, J. A. Preparation of 1-Naphthoic Nitrile from 1-Naphthyl-
amine. J. Am. Chem. Soc. 1930, 52, 4550-4552.
(8) van Niel, M. B.; Collins, I.; Beer, M. S.; Broughton, H. B.; Cheng,
S. K.; Goodacre, S. C.; Heald, A.; Locker, K. L.; MacLeod, A. M.;
Morrison, D.; Moyes, C. R.; O’Connor, D.; Pike, A.; Rowley, M.;
Russell, M. G.; Sohal, B.; Stanton, J. A.; Thomas, S.; Verrier, H.;
Watt, A. P.; Castro, J. L. Fluorination of 3-(3-(Piperidin-1-yl)propyl)-
indoles and 3-(3-(Piperazin-1-yl)propyl)indoles Gives Selective Hu-
man 5-HT1D Receptor Ligands with Improved Pharmacokinetic
Profiles. J. Med. Chem. 1999, 42, 2087-104.
(9) Saleh, S.; Aboul-Enein, H. Y.; Parhar, R.; Collison, K.; Al-Mohanna,
F. Ca2+-dependent production of reactive oxygen metabolites by
human neutrophils in response to fluorinated propranolol analogues.
Biochem. Pharmacol. 2001, 61, 517-525.
(10) Bohm, H. J.; Banner, D.; Bendels, S.; Kansy, M.; Kuhn, B.; Muller,
K.; Obst-Sander, U.; Stahl, M. Fluorine in Medicinal Chemistry.
Chembiochem 2004, 5, 637-643.
(11) Fluorination of a series of propranolol analogs which act as beta
blockers and antioxidants was shown to eliminate their beta blocking
effect while maintaining their antioxidant properties. Although this
effect was attributed to the stereochemistry of these compounds, the
reduction in the pKa due to â-fluorination likely contributed to the
observed decrease in their beta blocking activity. Weglicki, W. B.;
Mak, I. T.; Aboul-Enein, H. Y. Preparation, beta blocking effects,
and antioxidant activity of fluorinated propranolol. Patent WO
9718184, 1997.
(33) Chalmers, J. R.; Dickson, G. T.; Elks, J.; Hems, B. A. Synthesis of
Thyroxine and Related substances. Part V. A Synthesis of L-
Thyroxine from L-Tyrosine. J. Chem. Soc., Abstracts 1949, 3424-
3433.
(12) Vogt, M. The Concentration of Sympathin in Different Parts of the
Central Nervous System Under Normal Conditions and After the
Administration of Drugs. J. Physiol. (London) 1954, 123, 451-81.
(13) Gunne, L. M. Relative Adrenaline Content in Brain Tissue. Acta
Biochim. Pol. 1962, 56, 324-333.
(14) Axelrod, J. Purification and Properties of Phenylethanolamine-N-
Methyltransferase. J. Biol. Chem. 1962, 237, 1657-1660.
(15) Pendleton, R. G.; Kaiser, C.; Gessner, G. Studies on Adrenal
Phenylethanolamine N-Methyltransferase (PNMT) with SK&F 64139,
a Selective Inhibitor. J. Pharmacol. Exp. Ther. 1976, 197, 623-
632.
(16) Goldstein, M.; Kinguasa, K.; Hieble, J. P.; Pendleton, R. G. Lowering
of Blood Pressure in Hypertensive Rats by SK&F 64139 and SK&F
72223. Life Sci. 1982, 30, 1951-1957.
(17) Toomey, R. E.; Horng, J. S.; Hemrick-Luecke, S. K.; Fuller, R. W.
R2-Adrenoreceptor Affinity of Some Inhibitors of Norepinephrine
N-Methyltransferase. Life Sci. 1981, 29, 2467-72.
(18) Goldstein, M.; Saito, M.; Lew, J. Y.; Hieble, J. P.; Pendleton, R. G.
The Blockade of R2-Adrenoceptors by the PNMT Inhibitor SK&F
64139. Eur. J. Pharmacol. 1980, 67, 305-308.
(19) Drew, G. M. R2-Adrenoceptor-blocking Action of the Phenyletha-
nolamine-N-Methyltransferase Inhibitor SK&F 64139. J. Pharm.
Pharmacol. 1981, 33, 187-188.
(20) Pendleton, R. G.; Hieble, J. P. Studies on the Adrenergic Receptor
Specificity of Inhibitors of Phenylethanolamine N-Methyltransferase.
Res. Commun. Chem. Pathol. Pharmacol. 1981, 34, 399-408.
(21) Grunewald, G. L.; Dahanukar, V. H.; Jalluri, R. K.; Criscione, K. R.
Synthesis, Biochemical Evaluation, and Classical and Three-
Dimensional Quantitative Structure-Activity Relationship Studies
of 7-Substituted-1,2,3,4-tetrahydroisoquinolines and Their Relative
Affinities Toward Phenylethanolamine N-Methyltransferase and the
R2-Adrenoceptor. J. Med. Chem. 1999, 42, 118-134.
(22) Pendleton, R. G.; Gessner, G.; Weiner, G.; Jenkins, B.; Sawyer, J.;
Bondinell, W.; Intoccia, A. Studies on SK&F 29661, an Organ-
Specific Inhibitor of Phenylethanolamine N-Methyltransferase. J.
Pharmacol. Exp. Ther. 1979, 208, 24-30.
(34) Chen, Q.; Wu, S. Methyl (fluorosulfonyl)difluoroacetate; A New
Trifluoromethylating Agent. J. Chem. Soc., Chem. Commun. 1989,
705-706.
(35) Ballini, R.; Marcantoni, E.; Petrini, M. A New General Synthesis of
Sulfones from Alkyl or Aryl Halides and p-Toluenesulfonyl Hy-
drazide. Tetrahedron 1989, 45, 6791-6798.
(36) Romero, F. A.; Vodonick, S. M.; Criscione, K. R.; McLeish, M. J.;
Grunewald, G. L. Inhibitors of Phenylethanolamine N-Methyltrans-
ferase that are Predicted to Penetrate the Blood-Brain Barrier: Design,
Synthesis, and Evaluation of 3-Fluoromethyl-7-(N-substituted ami-
nosulfonyl)-1,2,3,4-tetrahydroisoquinolines that Possess Low Affinity
Toward the R2-adrenoceptor. J. Med. Chem. 2004, 47, 4483-4493.
(37) Caine, J. M.; Macreadie, I. G.; Grunewald, G. L.; McLeish, M. J.
Recombinant Human Phenylethanolamine N-Methyltransferase: Over-
production in Escheria coli, Purification and Characterization. Protein
Expression Purif. 1996, 8, 160-166.
(38) Grunewald, G. L.; Borchardt, R. T.; Rafferty, M. F.; Krass, P.
Conformational Preferences of Amphetamine Analogues for Inhibi-
tion of Phenylethanolamine N-Methyltransferase; Conformationally
Defined Adrenergic Agents. 5. Mol. Pharmacol. 1981, 20, 377-
381.
(39) Wu, Q.; Criscione, K. R.; Grunewald, G. L.; McLeish, M. J.
Phenylethanolamine N-Methyltransferase Inhibition: Re-evaluation
of Kinetic Data. Bioorg. Med. Chem. Lett. 2004, 14, 4217-4220.
(40) U′Prichard, D. C.; Greenberg, D. A.; Snyder, S. H. Binding
Characteristics of a Radiolabeled Agonist and Antagonist at Central
Nervous System Alpha Noradrenergic Receptors. Mol. Pharmacol.
1977, 13, 454-473.
(41) pKa values were calculated using pKaPlugin from the JChem (version
3.1) software package. JCHEM is available from ChemAxon Ltd.,
Maramaros koz 3/a Budapest, 1037 Hungary, http://www.chemax-
on.com. The accuracy of this program was assessed by comparing
the experimentally determined pKa values for the compounds in Table
1 with those obtained from the JChem software. The pKa values of
19d, 19a, 19b, and 19c as determined by the JChem software were
found to be 9.44, 7.91, 6.25, and 4.45, respectively.
(23) Grunewald, G. L.; Caldwell, T. M.; Li, Q. F.; Criscione, K. R.,
Synthesis and Evaluation of 3-Trifluoromethyl-7-substituted-1,2,3,4-
tetrahydroisoquinolines as Selective Inhibitors of Phenylethanolamine
N-Methyltransferase Versus the R2-Adrenoceptor. J. Med. Chem.
1999, 42, 3315-3323.
(24) Grunewald, G. L.; Lu, J.; Criscione, K. R.; Okoro, C. O. Inhibitors
of phenylethanolamine N-methyltransferase devoid of R2-adreno-
ceptor affinity. Bioorg. Med. Chem. Lett. 2005, 15, 5319-5323.
(25) Martin, J. L.; Begun, J.; McLeish, M. J.; Caine, J. M.; Grunewald,
G. L. Getting the Adrenaline Going: Crystal Structure of the
Adrenaline-Synthesizing Enzyme PNMT. Structure 2001, 9, 977-
985.
(26) McMillan, F. M.; Archbold, J.; McLeish, M. J.; Caine, J. M.;
Criscione, K. R.; Grunewald, G. L.; Martin, J. L. Molecular
Recognition of Sub-Micromolar Inhibitors by the Epinephrine-
Synthesizing Enzyme Phenylethanolamine N-Methyltransferase. J.
Med. Chem. 2004, 47, 37-44.
(42) Grunewald, G. L.; Caldwell, T. M.; Dahanukar, V. H.; Jalluri, R.
K.; Criscione, K. R. Comparative Molecular Field Analysis (CoMFA)
Models of Phenylethanolamine N-Methyltransferase (PNMT) and the
R2-Adrenoceptor: The Development of New Highly Selective
Inhibitors of PNMT. Bioorg. Med. Chem. Lett. 1999, 9, 481-486.
(43) Grunewald, G. L.; Romero, F. A.; Criscione, K. R., Nanomolar
Inhibitors of CNS Epinephrine Biosynthesis: (R)-(+)-3-Fluoro-
methyl-7-(N-substituted aminosulfonyl)-1,2,3,4-tetrahydroisoquino-
lines as Potent and Highly Selective Inhibitors of Phenylethanolamine
N-Methyltransferase. J. Med. Chem. 2005, 48, 1806-1812.
(44) Grunewald, G. L.; Caldwell, T. M.; Li, Q. F.; Slavica, M.; Criscione,
K. R.; Borchardt, R. T.; Wang, W. Synthesis and Biochemical
Evaluation of 3-Fluoromethyl-1,2,3,4-tetrahydroisoquinolines as Se-
lective Inhibitors of Phenylethanolamine N-Methyltransferase Versus
the R2-Adrenoceptor. J. Med. Chem. 1999, 42, 3588-3601.
(45) Charton, M. Steric Effects. I. Esterification and Acid-Catalyzed
Hydrolysis of Esters. J. Am. Chem. Soc. 1975, 97, 1552-1556.

2952 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 10
Grunewald et al.
(46) Hansch, C.; Leo, A.; Hoekman, D. Exploring QSAR Hydrophobic,
Electronic and Steric Constants; ACS Professional Reference Book;
American Chemical Society: Washington, DC, 1995.
(47) MacPhee, J. A.; Panaye, A.; Dubois, J. E. A Critical Examination of
the Taft Steric Parameter-Es. Definition of a Revised, Broader and
Homogeneous Scale. Extension to Highly Conjested Alkyl Groups.
Tetrahedron 1978, 34, 3553-3562.
(48) Grunewald, G. L.; Romero, F. A.; Chieu, A. D.; Fincham, K. J.;
Bhat, S. R.; Criscione, K. R. Exploring the Active Site of Phenyl-
ethanolamine N-Methyltransferase: 3-Alkyl-7-substituted-1,2,3,4-
tetrahydroisoquinoline Inhibitors. Bioorg. Med. Chem. 2005, 13,
1261-1273.
(49) Grunewald, G. L.; Romero, F. A.; Seim, M. R.; Criscione, K. R.;
Deupree, J. D.; Spackman, C. C.; Bylund, D. B. Exploring the Active
Site of Phenylethanolamine N-Methyltransferase with 3-Hydroxy-
ethyl- and 3-Hydroxypropyl-7-Substituted-1,2,3,4-tetrahydroisoquino-
lines. Bioorg. Med. Chem. Lett. 2005, 15, 1143-1147.
(50) Bikker, J. A.; Trumpp-Kallmeyer, S.; Humblet, C. G-Protein Coupled
Receptors: Models, Mutagenesis, and Drug Design. J. Med. Chem.
1998, 41, 2911-2927.
(51) Hibert, M. F.; Trumpp-Kallmeyer, S.; Bruinvels, A.; Hoflack, J.
Three-Dimensional Models of Neurotransmitter G-Binding Protein-
Coupled Receptors. Mol. Pharmacol. 1991, 40, 8-15.
(52) Wang, C. D.; Buck, M. A.; Fraser, C. M. Site-Directed Mutagenesis
of R2A-Adrenergic Receptors: Identification of Amino Acids Involved
in Ligand Binding and Receptor Activation by Agonists. Mol.
Pharmacol. 1991, 40, 168-79.
(53) Nyronen, T.; Pihlavisto, M.; Peltonen, J. M.; Hoffren, A.-M.; Varis,
M.; Salminen, T.; Wurster, S.; Marjamaki, A.; Kanerva, L.; Katainen,
E.; Laaksonen, L.; Savola, J.-M.; Scheinin, M.; Johnson, M. S.
Molecular Mechanism for Agonist-Promoted R2a-Adrenoceptor Ac-
tivation by Norepinephrine and Epinephrine. Mol. Pharmacol. 2001,
59, 1343-1354.
(54) Salminen, T.; Varis, M.; Nyronen, T.; Pihlavisto, M.; Hoffren, A.-
M.; Lonnberg, T.; Marjamaki, A.; Frang, H.; Savola, J.-M.; Scheinin,
M.; Johnson, M. S. Three-Dimensional Models of R2A-Adrenergic
Receptor Complexes Provide a Structural Explanation for Ligand
Binding. J. Biol. Chem. 1999, 274, 23405-23413.
(55) Fuller, R. W.; Molloy, B. B.; Day, W. A.; Roush, B. W.; Marsh, M.
M. Inhibition of Phenylethanolamine N-Methyltransferase by Ben-
zylamines. 1. Structure-Activity Relationships. J. Med. Chem. 1973,
16, 101-106.
(56) Unpublished data from J. L. Martin, University of Queensland,
Brisbane, Australia.
(57) The shift in the binding pocket was measured by comparing the
distance between the centroid of the aromatic ring of Tyr222 and
the centroids of the aromatic rings of THIQs 11a-c.
(58) SYBYL 7.0Tripos Inc., 1699 South Hanley Rd., St. Louis, MO 63144.
(59) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Hart, W. E.; Belew,
R. K.; Olson, A. J. Automated Docking Using a Lamarckian Genetic
Algorithm and an Empirical Binding Free Energy Function. J.
Comput. Chem. 1998, 19, 1639-1662.
(60) Grunewald, G. L.; Caldwell, T. M.; Li, Q. F.; Dahanukar, V. H.;
McNeil, B.; Criscione, K. R. Enantiospecific Synthesis of 3-Fluo-
romethyl-, 3-Hydroxymethyl-, and 3-Chloromethyl-1,2,3,4-tetrahy-
droisoquinolines as Selective Inhibitors of Phenylethanolamine
N-Methyltransferase Versus the R2-Adrenoceptor. J. Med. Chem.
1999, 42, 4351-4361.
(61) Kraulis, P. J. MOLSCRIPT: a program to produce both detailed and
schematic plots of protein structures. J. Appl. Crystallogr. 1991, 24,
945-949.
(62) Merrit, E. A.; Bacon, D. J. RASTER3D: photorealistic molecular
graphics. Methods Enzymol. 1997, 277, 505-524.
JM051262K