Exploring the active site of phenylethanolamine N-methyltransferase with 3-hydroxyethyl- and 3-hydroxypropyl-7-substituted-1,2,3,4- tetrahydroisoquinolines

Article abstract of DOI:10.1016/j.bmcl.2004.12.013

3-Hydroxyethyl- and 3-hydroxypropyl-7-substituted-tetrahydroisoquinolines (9, 10, 16, and 17) were synthesized and evaluated for their phenylethanolamine N-methyltransferase (PNMT) inhibitory potency and affinity for the α₂-adrenoceptor. Althou

Full text of DOI:10.1016/j.bmcl.2004.12.013

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1143–1147
Exploring the active site of phenylethanolamine
N-methyltransferase with 3-hydroxyethyl- and
3-hydroxypropyl-7-substituted-1,2,3,4-tetrahydroisoquinolines
Gary L. Grunewald,a,* F. Anthony Romero,a Mitchell R. Seim,a Kevin R. Criscione,a
Jean D. Deupree,b Christy C. Spackmanb and David B. Bylundb
aDepartment of Medicinal Chemistry, University of Kansas, 1251 Wescoe Hall Drive, Lawrence, KS 66045, USA
bUniversity of Nebraska Medical Center, 985800 Nebraska Medical Center, Omaha, NE 68198, USA
Received 9 August 2004; revised 30 November 2004; accepted 6 December 2004
Available online 28 December 2004
Abstract—3-Hydroxyethyl- and 3-hydroxypropyl-7-substituted-tetrahydroisoquinolines (9, 10, 16, and 17) were synthesized and
evaluated for their phenylethanolamine N-methyltransferase (PNMT) inhibitory potency and affinity for the a2-adrenoceptor.
Although a2-adrenoceptor affinity decreased for these compounds, selectivity was not gained over the parent 3-hydroxymethyl
compounds (1, 2) due to a loss in PNMT inhibitory potency.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
useful pharmacological tool to help define the role of
CNS Epi.
Epinephrine (Epi) in the central nervous system
(CNS) constitutes approximately 5% of the catechol-
amine content in the mammalian brain,1,2 yet its role
therein remains unclear. It has been implicated in
some of the neurodegeneration seen in Alzheimerꢀs
disease,3,4 and in the regulation of blood pressure,5
respiration,6,7 body temperature,6,7 as well as a1-8 and
a2-adrenoceptors.9,10
3-Hydroxymethyl-7-nitro-THIQ (1) and 7-bromo-3-
hydroxymethyl-THIQ (2) are two highly potent PNMT
inhibitors, but both lack significant selectivity for PNMT
versus the a2-adrenoceptor (Table 2).14 Examination of
the recently solved crystal structures of hPNMT co-crys-
tallized with S-adenosyl-L-homocysteine (AdoHcy) and
either 7-iodo-THIQ (3)15 or SK&F 29661 (4; 7-amino-
sulfonyl-THIQ)16 indicated that the active site should
be able to accommodate longer 3-hydroxyalkyl groups.
Docking of 3-hydroxyethyl- or 3-hydroxypropyl-7-
substituted-THIQs into the PNMT active site indicated
that the longer 3-hydroxyalkyl chains could bind favor-
ably (Fig. 2 shows 3-hydroxyethyl-7-nitro-THIQ).
To elucidate the role of CNS Epi, we have targeted
phenylethanolamine N-methyltransferase (PNMT: EC
2.1.1.28),1 the terminal enzyme in the biosynthesis of
Epi. Some problems associated with early PNMT inhibi-
tors are 1) their lack of selectivity for PNMT versus the
a2-adrenoceptor, or 2) their inability to penetrate into
the CNS. A compound possessing a calculated logP11,12
value greater than 0.5 is likely required for 1,2,3,4-tetra-
hydroisoquinoline-type (THIQ-type) PNMT inhibitors
to achieve significant CNS penetration.13 A potent
inhibitor of PNMT that is both selective and capable
of crossing the blood–brain barrier (BBB) would be a
A previously reported a2-adrenoceptor comparative
molecular field analysis model on a set of THIQs showed
that there was steric bulk intolerance around the 3-posi-
tion of THIQ.17 Thus, longer 3-hydroxyalkyl substitu-
ents (9, 10, 16, 17) should be disfavored at the a2-
adrenoceptor and should increase selectivity for PNMT.
2. Chemistry
Keywords: Phenylethanolamine N-methyltransferase; Enzyme inhibi-
tors; Tetrahydroisoquinoline; Structure-based design.
*
The synthesis of 9 and 10 is shown in Scheme 1. The syn-
thesis of 518 has been reported previously. Treatment of
Corresponding author. Tel.: +1 785 864 4497; fax: +1 785 864
5326; e-mail: ggrunewald@ku.edu
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.12.013

1144
G. L. Grunewald et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1143–1147
tion of both the acid and lactam moieties yielded THIQ
16. The nitro groupof 15 was reduced to the aniline,
converted to the bromide under Sandmeyer conditions,
and reduction of the acid and lactam yielded THIQ
17. Compounds 9, 10, 16, and 17 were purified by flash
chromatography before conversion to their HCl salts
that were purified by recrystallization from EtOH/
hexanes.19,20
3. Biochemistry
In this study, human PNMT (hPNMT) with a C-termi-
nal hexahistidine tag was expressed in E. coli.21,22 The
radiochemical assay conditions, previously reported
for the bovine enzyme,23 were modified to account for
the high binding affinity of some inhibitors.22,24 Inhibi-
tion constants were determined using four concentra-
tions of phenylethanolamine as the variable substrate,
and three concentrations of inhibitor.
Scheme 1. Reagents and conditions: (a) ClCO2Et; (b) PPA, 145 °C; (c)
H2SO4, KNO3; (d) H2, PtO2/C; (e) HBr, NaNO2; (f) CuBr, HBr; (g)
BH3ÆTHF.
5 with ethylchloroformate and K2CO3 in THF produced
ethylcarbamate 6 that was easily purified by Kugelrohr
distillation and was cyclized with polyphosphoric acid
heated to 145 °C to yield lactam 7. Nitration of 7 with
potassium nitrate afforded 8 that was recrystallized in
EtOAc/hexanes. Reduction of lactam 8 with BH3ÆTHF
yielded THIQ 9. The nitro moiety of 8 was reduced to
the aniline, Sandmeyer chemistry was used to transform
the aniline to a bromide, and subsequent reduction
of the lactam yielded THIQ 10.
a2-Adrenergic receptor binding assays were performed
using either cortex obtained from male Sprague Dawley
rats,25 with [3H]clonidine (2 nM) as the radioligand and
phentolamine to define nonspecific binding, or mem-
brane preparations from CHO cells stably transfected
with
human
a
2A-adrenoceptors
{radioligand:
[3H]RX821002 (0.2 nM)/nonspecific binding: (À)-nor-
epinephrine}.26,27 A comparison of the binding data
for a related series of compounds (Table 1) shows a
good correlation (r2 = 0.90) between the inhibition
of [3H]clonidine binding to rat cortex a2-adrenoceptors
compared to the inhibition of [3H]RX821002 binding
to cloned human a2A-adrenoceptors. This correlation
Compound 7 was used to synthesize 16 and 17 (Scheme
2). Compound 7 was reduced to alcohol 11, which was
transformed to nitrile 13 by conversion to mesylate 12
followed by displacement with cyanide (KCN). Hydro-
lysis of nitrile 13 with sulfuric acid afforded 14 in good
yield. Compound 14 was nitrated and subsequent reduc-
Table 1. a2-Adrenoceptor affinity comparison
Compd R3
R7
R8
a2-Adrenoceptor affinity
(pKi)
Rat cortexa Human-a2A
b
18
4
H
Cl
Cl 7.68c
6.52
3.40
5.20
4.84
3.90
3.20
3.60
3.40
4.35
H
SO2NH2
NO2
NO2
H
H
H
H
H
H
H
H
4.00d
5.37e
4.51d
4.12d
3.00f
2.82
19
20
1
H
CH3
CH2OH NO2
21
22g
23
24
CH2F
CHF2
CF3
NO2
NO2
NO2
Br
3.00h
4.72h
CF3
a Inhibition of [3H]clonidine binding to rat cortex preparation.
b Inhibition of [3H]RX821002 to cloned human a2A-adrenoceptors.
c Ref. 28.
d Ref. 14.
e Ref. 29.
f Ref. 13.
Scheme 2. Reagents and conditions: (a) LiBH4; (b) CH3SO2Cl; (c)
KCN, DMF; (d) H2SO4/H2O; (e) KNO3, H2SO4; (f) H2, PtO2/C; (g)
HBr, NaNO2; (h) CuBr, HBr; (i) BH3ÆTHF.
g Unpublished results with M. R. Seim. The synthesis and hPNMT
inhibitory data for 22 will be published separately.
h Ref. 30.

G. L. Grunewald et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1143–1147
1145
supports our continued use of [3H]clonidine to define a2-
adrenoceptor affinity.
4. Results and discussion
Compounds 9, 10, 16, and 17 (Table 2) were synthesized
and evaluated for PNMT inhibitory potency and a2-
adrenoceptor affinity. Although the a2-adrenoceptor
affinity of these compounds decreased, their PNMT
inhibitory potency also decreased such that no selecti-
vity was gained.
Docking studies (AutoDock 3.0)31 were performed to
investigate the binding of these inhibitors in the PNMT
active site, using the recently reported X-ray crystal
structures of PNMT co-crystallized with AdoHcy and
either SK&F 29661 (4)16 or 7-iodo-THIQ (3).15 The
docking of inhibitors into the hPNMT active site was
performed on the R-enantiomer as a previous study on
3-hydroxymethyl-7-substituted-THIQs indicated that
the R-enantiomer is preferred over the S- in the hPNMT
active site.32 Residues Lys57 and Met258 move within
the PNMT active site depending on the hydrophobicity
of the 7-substitutent.15 Inhibitors that contain a 7-nitro
substituent were docked into the crystal structure of
hPNMT co-crystallized with SK&F 29661 (4),16
whereas inhibitors that possess a 7-bromo substituent
were docked into the crystal structure of hPNMT co-
crystallized with 7-iodo-THIQ (3).15 The amino acid resi-
dues surrounding the 3-position of the co-crystallized
ligands SK&F 29661 (4) and 7-iodo-THIQ (3) remain
in similar positions in both crystal structures.
Figure 1. Compound 2 docked into the active site of hPNMT15 and
the amino acid residues that could interact with 2, showing that the 3-
hydroxymethyl moiety may hydrogen bond with Glu219. Yellow lines
indicate possible hydrogen bonds. A Connolly (solvent accessible)
surface of the active site is also shown. Carbon is white, nitrogen is
blue, oxygen is red and bromine is green. Hydrogens are not shown for
clarity.
The docking of 1 (3-CH2OH, 7-NO2) or 2 (3-CH2OH, 7-
Br) into the hPNMT crystal structure shows that the hy-
droxy moiety appears to hydrogen bond to Glu219 (Fig.
1 shows 2). The docking studies also show that the 7-ni-
tro moiety of 1 (similar to Fig. 2) may hydrogen bond to
Lys57 as does the sulfonamide substituent of SK&F
29661 (4) and the 7-bromo substituent of 2 (Fig. 1)
may make hydrophobic contacts with Met258 and
Val53 as does the iodo groupof 7-iodo-THIQ ( 3). The
docking studies indicated that a 3-hydroxyethyl (9) or
a 3-hydroxypropyl (16) substituent could take advan-
tage of hydrogen bonds to Tyr35 and/or the main chain
Figure 2. Compound 9 docked into the active site of hPNMT16 and
the amino acid residues that could interact with 9, showing that the 3-
hydroxyethyl moiety could hydrogen bond with Tyr35 and/or the main
chain of Phe182. Yellow lines indicate possible hydrogen bonds. A
Connolly (solvent accessible) surface of the active site is also shown.
Carbon is white, nitrogen is blue, and oxygen is red. Hydrogens are not
shown for clarity.
Table 2. In vitro activities of ( )-3-hydroxyethyl- and ( )-3-hydroxypropyl-7-substituted-THIQs
Compd
R
n
hPNMT Ki (lM SEM)
Selectivity a2/PNMT
C logPb
a
a2 Ki (lM SEM)
1c
2c
9
NO2
Br
0
0
1
1
2
2
0.047 0.006
0.012 0.001
0.51 0.03
0.35 0.01
1.4 0.1
19
1.0 0.1
46
1.3 0.2
59
4.8 0.4
1
400
83
0.65
1.77
0.91
2.03
0.93
2.05
NO2
Br
4
90
10
16
17
3.7
42
NO2
Br
6
1.7 0.2
2.8
a Inhibition of [3H]clonidine binding to rat cortex preparation.
b Calculated logP.
c The syntheses and PNMT (bovine adrenal enzyme) data for these compounds has been previously reported.14

1146
G. L. Grunewald et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1143–1147
gress; Lipton, M. A., DiMascio, A., Killam, K. F., Eds.;
Raven: New York, 1978; p261.
8. Stone, E. A.; Grunewald, G. L.; Lin, Y.; Ashan, R.;
Rosengarten, H.; Kramer, H. K.; Quartermain, D. Syn-
apse 2003, 49, 67.
9. Rosin, D. L.; Zeng, D.; Stornetta, R. L.; Norton, F. R.;
Riley, T.; Okusa, M. D.; Guyenet, P. G.; Lynch, K. R.
Neuroscience 1993, 56, 139.
10. Stolk, J. M.; Vantini, G.; Guchait, R. B.; UꢀPrichard,
D. C. Science 1983, 221, 1297.
of Phe182 (Fig. 2 shows 9), rather than hydrogen bond
to Glu219 as proposed for the 3-hydroxmethyl group
(Fig. 1), and thus PNMT inhibitory potency should be
retained. However, the biochemical data for these com-
pounds indicates that no alternative interaction is gained
in the hPNMT active site. Extension of the 3-hydroxy-
alkyl chain by just one carbon (9, 10) resulted in a signif-
icant loss in PNMT inhibitory potency. The 3-
hydroxymethyl substituent and THIQ nitrogen of 1 (3-
CH2OH, 7-NO2) and 2 (3-CH2OH, 7-Br) are in optimal
positions to interact with Glu219 (Fig. 1 shows 2). The
loss of the hydrogen bond between Glu219 and the hy-
droxyl groupof 9, 10, 16, and 17 without the formation
of a new hydrogen bond may be the reason why a signifi-
cant loss in PNMT inhibitory potency is observed.
Structure-based design is an iterative process and an
X-ray crystal structure of PNMT co-crystallized with
one of these inhibitors will be required to use these re-
sults for future structure-based design.
11. Calculated logP values were determined using the C logP
function in SYBYLÒ.
12. SYBYLÒ 6.9 Tripos Inc., 1699 South Hanley Rd., St.
Louis, Missouri 63144, USA.
13. Grunewald, G. L.; Caldwell, T. M.; Li, Q.; Slavica, M.;
Criscione, K. R.; Borchardt, R. T.; Wang, W. J. Med.
Chem. 1999, 42, 3588.
14. Grunewald, G. L.; Dahanukar, V. H.; Teoh, B.; Criscione,
K. R. J. Med. Chem. 1999, 42, 1982.
15. McMillan, F. M.; Archbold, J.; McLeish, M. J.; Caine,
J. M.; Criscione, K. R.; Grunewald, G. L.; Martin, J. L.
J. Med. Chem. 2004, 37.
16. Martin, J. L.; Begun, J.; McLeish, M. J.; Caine, J. M.;
Grunewald, G. L. Structure 2001, 9, 977.
17. Grunewald, G. L.; Caldwell, T. M.; Dahanukar, V. H.;
Jalluri, R. K.; Criscione, K. R. Bioorg. Med. Chem. Lett.
1999, 9, 481.
In conclusion, although the rationale for decreasing
a2-adrenoceptor affinity proved correct by extending
the 3-hydroxyalkyl chain of 1 or 2 (9, 10, 16, 17), these
compounds showed no increase in selectivity. Molecular
modeling suggested that longer 3-hydroxyalkyl chains
should bind favorably in the PNMT active site, but
the significant decrease in PNMT inhibitory potency
indicates that an alternative hydrogen bond did not
form. The 3-hydroxymethyl substituent of 1 or 2 ap-
pears to be an optimal group to interact with the PNMT
active site.
18. Zablocki, J. A.; Foe, S. T.; Bovy, P. R.; Miyano, M.;
Garland, R. B.; Williams, K.; Schretzman, L.; Zupec, M.
E.; Rico, J. G.; Lindmark, R. J.; Toth, M. V.; McMackins,
D. E.; Adams, S. P.; Panzer-Knodle, S. G.; Nicholson, N.
S.; Taite, B. B.; Salyers, A. K.; King, L. W.; Campion, J.
G.; Feigen, L. P. Bioorg. Med. Chem. 1995, 3, 539.
1
19. Compound 9ÆHCl: mp236–237 °C; H NMR (500 MHz,
DMSO-d6) d 9.67 (s, 2H), 8.23 (s, 1H), 8.12 (d, J = 8.5 Hz,
1H), 7.52 (d, J = 8.5 Hz, 1H), 5.94 (s, 1H), 4.50–4.41 (m,
2H), 3.63 (m, 3H), 3.30–3.26 (m, 1H), 3.05–2.99 (m, 1H),
2.04 (m, 1H), 1.79 (m, 1H). Compound 10ÆHCl: mp235–
236 °C; 1H NMR (400 MHz, DMSO-d6) d 9.20 (s, 2H),
7.51–7.46 (m, 2H), 7.19 (d, J = 8.2 Hz, 1H), 5.01 (s, 1H),
4.38–4.28 (m, 2H), 3.75–3.57 (m, 3H), 3.12–3.06 (m, 1H),
2.84–2.77 (m, 1H), 1.99–1.90 (m, 1H), 1.78–1.70 (m, 1H).
Compound 16ÆHCl: mp237–243 °C dec; 1H NMR (400
MHz, DMSO-d6) d 9.26 (s, 2H), 8.22 (s, 1H), 8.12 (d, J =
8.7 Hz, 1H), 7.52 (d, J = 8.5 Hz, 1H), 4.50–4.37 (m, 2H),
3.46 (m, 3H), 3.29–3.24 (m, 2H), 2.93–2.86 (m, 1H), 1.79 (m,
1H), 1.67–1.59 (m, 3H). Compound 17ÆHCl: mp203–
204 °C; 1H NMR (400 MHz, DMSO-d6) d 9.26 (s, 2H), 7.51
(s, 1H), 7.46 (d, J = 8.7 Hz, 1H), 7.19 (d, J = 8.5 Hz, 1H),
4.29 (m, 2H), 3.45 (m, 3H), 3.22 (m, 1H), 3.10–3.05 (m, 1H),
2.79–2.72 (m, 1H), 1.87–1.76 (m, 1H), 1.68–1.50 (m, 3H).
20. All compounds gave acceptable ( 0.4) C, H, N elemental
analyses.
Acknowledgements
This research was supported by NIH Grant HL 34193.
We thank David VanderVelde and Sarah Neuenswander
of the University of Kansas Nuclear Magnetic Reso-
nance Laboratory for their assistance. The 500 MHz
NMR spectrometer was partially funded by NSF Grant
CHE-9977422. We also 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.
References and notes
21. Caine, J. M.; Macreadie, I. G.; Grunewald, G. L.;
McLeish, M. J. Protein Expr. Purif. 1996, 8, 160.
22. Romero, F. A.; Vodonick, S. M.; Criscione, K. R.;
McLeish, M. J.; Grunewald, G. L. J. Med. Chem. 2004,
47, 4483.
23. Grunewald, G. L.; Borchardt, R. T.; Rafferty, M. F.;
Krass, P. Mol. Pharmacol. 1981, 20, 377.
24. Wu, Q.; Criscione, K. R.; Grunewald, G. L.; McLeish,
M. J. Bioorg. Med. Chem. Lett. 2004, 14, 4217.
25. UꢀPrichard, D. C.; Greenberg, D. A.; Snyder, S. H. Mol.
Pharmacol. 1977, 13, 454.
1. Axelrod, J. J. Biol. Chem. 1962, 237, 324.
2. Vogt, M. J. Physiol. 1954, 123, 451.
3. Burke, W. J.; Chung, H. D.; Strong, R.; Mattammal, M.
B.; Marshall, G. L.; Nakra, R.; Grossberg, G. T.; Haring,
J. H.; Joh, T. H. In Central Nervous System Disorders of
Aging: Clinical Intervention and Research; Strong, R., Ed.;
Raven: New York, 1988; p41.
4. Kennedy, B. P.; Bottiglieri, T.; Arning, E.; Ziegler, M. G.;
Hansen, L. A.; Masliah, E. J. Neural. Transm. 2004, 111,
547.
5. Saavedra, J. M.; Grobecker, H.; Axelrod, J. Science 1976,
191, 483.
6. Rothballer, A. B. Pharmacol. Rev. 1959, 11, 494.
7. Goldstein, M.; Lew, J. Y.; Matsumoto, Y.; Hokfelt, T.;
Fuxe, K. In Psychopharmacology: A Generation of Pro-
26. Bylund, D. B.; Blaxall, H. S.; Iversen, L. J.; Caron, M. G.;
Lefkowitz, R. J.; Lomasney, J. W. Mol. Pharmacol. 1992,
42, 1.
27. Deupree, J. D.; Hinton, K. A.; Cerutis, D. R.; Bylund,
D. B. J. Pharmacol. Exp. Ther. 1996, 278, 1215.

G. L. Grunewald et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1143–1147
1147
28. Grunewald, G. L.; Dahanukar, V. H.; Ching, P.; Crisci-
one, K. R. J. Med. Chem. 1996, 39, 3539.
29. Grunewald, G. L.; Dahanukar, V. H.; Jalluri, R.; Crisci-
one, K. R. J. Med. Chem. 1999, 42, 118.
30. Grunewald, G. L.; Caldwell, T. M.; Li, Q.; Criscione,
K. R. J. Med. Chem. 1999, 42, 3315.
31. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Hart,
W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem. 1998,
19, 1639.
32. Grunewald, G. L.; Caldwell, T. M.; Li, Q.; Dahanukar,
V. H.; McNeil, B.; Criscione, K. R. J. Med. Chem. 1999,
42, 4351.