First enantioselective syntheses of the dopamine D1 and D2 receptor modulators, (+)-and (-)-govadine

Article abstract of DOI:10.1016/j.bmcl.2012.01.005

There is a pressing need to find and develop new antipsychotic agents for the treatment of schizophrenia. Current drugs primarily target dopamine D2receptors and are only effective in the treatment of the positive symptoms of this indication. The tetrahyd

Full text of DOI:10.1016/j.bmcl.2012.01.005

Bioorganic & Medicinal Chemistry Letters 22 (2012) 1557–1559
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
First enantioselective syntheses of the dopamine D1 and D2 receptor
modulators, (+)- and (À)-govadine
Huimin Zhai a, James Miller b, Glenn Sammis a,
⇑
a University of British Columbia, Department of Chemistry, 2036 Main Mall, Vancouver, BC, Canada V6T 1Z1
b Panora Pharmaceuticals Inc., 506-1168 Hamilton Street, Vancouver, BC, Canada V6B 2S2
a r t i c l e i n f o
a b s t r a c t
Article history:
There is a pressing need to find and develop new antipsychotic agents for the treatment of schizophrenia.
Current drugs primarily target dopamine D2 receptors and are only effective in the treatment of the positive
symptoms of this indication. The tetrahydroprotoberberine natural product ( )-govadine has shown
unique promise for the treatment of both the positive and negative symptoms of schizophrenia as it targets
both dopamine D1 and D2 receptors. However, further clinical research has been hindered by the lack of
availability of significant quantities of enantioenriched material. A new, enantioselective synthetic route
has been developed that affords (À)-govadine in 39% overall yield over 5-steps from commercially available
dopamine and homovanillic acid derivatives. Using only minor modifications in the synthetic route, (+)-
govadine can be synthesized in comparable yields and enantioselectivities. The route is readily scalable
as every intermediate was purified by crystallization and no flash column chromatography was necessary.
Ó 2012 Published by Elsevier Ltd.
Received 19 November 2011
Revised 3 January 2012
Accepted 3 January 2012
Available online 10 January 2012
Keywords:
Govadine
Tetrahydroprotoberberine
Schizophrenia
Synthesis
Enantioselective
Schizophrenia is a chronic neuropathic disease for which there
is currently no cure.1 Studies suggest that the symptoms of schizo-
phrenia are largely due to an excess of dopamine in the subcortex2
and a deficiency of dopamine in the frontal cortex.3 Current anti-
psychotic agents are primarily dopamine D2 receptor antagonists2
and are only partial treatments of the imbalance of the dopaminer-
gic system; these drugs are effective at treating the positive symp-
toms of schizophrenia, such as delusions and hallucinations, but
have little effect on the cognitive deficits and negative symptoms,
such as apathy. There is a pressing need to find and develop new
pharmaceutical agents that effectively treat all symptoms of this
debilitating disease. Recent studies have revealed that several tet-
rahydroprotoberberines (THPBs), a bioactive4 class of tetracyclic
alkaloids (Fig. 1),5 exhibit unique pharmacological activities in
their ability to elicit activities in both the D1 and D2 receptors
in vitro and in vivo.6 In a recent preclinical assessment of the THPB
govadine,7 in vitro and in vivo studies of the racemic compound
indicate that it is a viable dopamine D1 agonist and dopamine
D2 antagonist, and thus may be effective for treating both the
positive and negative symptoms of schizophrenia.8 Further testing
of each enantiomer was restricted by access to each enantiomer.9
As only racemic syntheses of govadine have been reported,10 an
efficient and scalable route to both (+)- and (À)-govadine was
essential to continue its evaluation as an antipsychotic drug.
When designing an enantioselective synthesis of govadine, we
wanted to minimize the need for flash column chromatography as
this would significantly impede future production of either (+)- or
(À)-govadine on a multi-gram scale. In addition, a convergent syn-
thesis that minimized the total number of synthetic steps would
facilitate future analog synthesis. Based on these requirements, we
planned to synthesize govadine according to Scheme 1. Benzyl pro-
tective groups were to be used throughout the synthesis because
they increase the crystallinity of each intermediate and they should
bereadilyremovedinthefinalstep. Weplannedtoset thestereocen-
ter using an enantioselective reduction of dihydroisoquinoline 3. If
bothenantiomersof thecatalyst are readilyavailable, then thisroute
can be used to access either enantiomer of govadine. Intermediate 3
shouldbe readilyaccessedfrom a Bischler–Napieralski cyclizationof
RO
MeO
N
N
RO
HO
H
OR
OR
OH
OMe
( )-govadine(1)
_
tetrahydroprototperberine
core structure
⇑
Figure 1. Core structure of tetrahydroprotoberberines (THPB’s) and (À)-govadine
(1).
Corresponding author. Tel.: +1 604 827 4080; fax: +1 604 822 2847.
E-mail address: gsammis@chem.ubc.edu (G. Sammis).
0960-894X/$ - see front matter Ó 2012 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2012.01.005

1558
H. Zhai et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1557–1559
O
MeO
MeO
MeO
BnO
NH2
X
NH
N
HCl
HN
O
BnO
MeO
BnO
MeO
+
1
OMe
OMe
OBn
OBn
BnO
BnO
BnO
OMe
6a: X = OH
6b
3
2
4
5
: X = Cl
Scheme 1. Retrosynthetic analysis of (À)-govadine.
Table 1
Optimization of the peptide coupling between dopamine derivative 5 and homovanillic acid derivative 6
O
MeO
X
NH2
conditions
HN
O
BnO
+
OMe
OMe
OBn
OBn
6
4
5
BnO
6a
: X = OH
SOCl2
OMe
6b: X = Cl
Entry
Homovanillic acid derivative
Conditions
NEt3, DCM
DCC, N-hydroxysuccinimide, DCM
Neat, 180 °C
Yield (%)
1
2
3
4
6b
6a
6a
6a
59a
46a
90b
nd
lw, 170 °C, DMSO
a
Yield after column chromatography.
Yield after recrystallization.
b
4.11 Compound 4, in turn, can be prepared from the coupling of
dopamine derivative 5 and homovanillic acid (6a), both of which
are commercially available.
MeO
BnO
MeO
BnO
HCl
N
HN
O
POCl3
The synthesis of (À)-govadine began with an investigation of
the known peptide couplings between dopamine derivative 5
and homovanillic acid derivative 6 to determine which conditions
provided the highest yields and which could be purified by crystal-
lization (Table 1). We first explored conversion of carboxylic acid
6a to the corresponding acid chloride (6b) using thionyl chloride,
followed by coupling with amine 5 (entry 1).12 This protocol only
afforded modest yields (50–60%) of amide 4 and the reaction was
not clean, thus necessitating purification by flash column chroma-
tography. Peptide coupling using DCC and hydroxysuccinimide
provided 5 in a slightly lower yield (40–50%),13 but the product
could still not be isolated cleanly by crystallization (entry 2). We
next explored a thermal peptide coupling between amine 5 and
acid 6a (entry 3). Although these conditions have never been
explored with either 5 or 6a, they have been demonstrated to be
successful when coupling comparable dopamine and homovanillic
acid derivatives.14 Gratifyingly, heating the two species neat at
180 °C provided clean conversion to the desired coupled product,
which could readily be purified by recrystallization. Thermal
peptide couplings were also attempted using a microwave reactor,
but there was significant decomposition (entry 4).
C6H6
80 °C
86% yield
BnO
OBn
4
3
OMe
OMe
1 mol% Noyori catalyst
HCO2H/NEt3=5/2
DMF
MeO
MeO
BnO
NH HCl
NH
HCl
BnO
After
MeOH/Et2O
–20 °C
recrystallization:
85% yield
OBn
OBn
2
2
HCl
>99% ee
OMe
OMe
Noyori catalyst = RuCl[(R,R)-TsDPEN(p-cymene)]
Scheme 2. Synthesis of tetrahydroisoquinoline derivative 2ÁHCl from amide 4.
Hydrogenation of dihydroisoquinoline 3 under the standard condi-
tions described by Noyori et al. cleanly provided the desired tetrahy-
droisoquinoline 2. Tetrahydroisoquinoline 2 was unstable at room
temperature,18 so it was converted to the corresponding HCl salt
and crystallized in 85% overall yield and >99% ee. While the yield
can be improved by carrying crude material through to the next step,
later intermediates could not be efficiently enantioenriched through
recrystallization.
Treatment of coupled product 4 with phosphoryl chloride
induced a Bischler–Napieralski cyclization affording dihydroiso-
quinoline hydrochloride 3 in 86% yield (Scheme 2). We then began
our investigations into the key enantioselective reduction step with
a Noyori transfer hydrogenation (Scheme 2).15 Noyori transfer
hydrogenations are powerful methods for the hydrogenation of imi-
nes and have been demonstrated to enantioselectively hydrogenate
dihydroisoquinolines with excellent selectivities (93–95% ee).16,17
Mannich-type cyclization of tetrahydroisoquinoline 2ÁHCl using
aqueous formaldehyde and acetic acid afforded only the desired

H. Zhai et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1557–1559
1559
MeO
BnO
MeO
BnO
Acknowledgments
.
HCl
N
37% HCO2H aq.
AcOH, 90°
.
NH
HCl
This work was supported by the University of British Columbia,
the Natural Sciences and Engineering Research Council of Canada,
and Panora Pharmaceuticals.
H
then HCl
MeOH, –20 °C
81% yield
OBn
OBn
.
.
2 HCl
7 HCl
OMe
OMe
Supplementary data
Scheme 3. Mannich-type cyclization of tetrahydroisoquinoline 2.
Supplementary data (all of the full procedures and chemical
compound information) associated with this article can be found,
in the online version, at doi:10.1016/j.bmcl.2012.01.005.
H2
MeO
BnO
MeO
HO
Pd/C
MeOH
References and notes
HCl
HCl
N
N
H
1. For a review, see: Freedman, R. N. Engl. J. Med. 2003, 349, 1738.
2. (a) Carlsson, A.; Carlsson, M. L. Dialogues Clin. Neurosci. 2006, 8, 137; (b) van Os,
J.; Kapur, S. Lancet 2009, 374, 635.
3. (a) Weinberger, D. R.; Egan, M. F.; Bertolino, A.; Callicott, J. H.; Mattay, V. S.;
Lipska, B. K.; Berman, K. F.; Goldberg, T. E. Biol. Psychiatry 2001, 50, 825;
(b) Goldman-Rakic, P. S.; Castner, S. A.; Svensson, T. H.; Siever, L. J.; Williams, G.
V. Psychopharmacology 2004, 174, 3.
H
or
HCl
MeOH
64 °C
OBn
OH
1 HCl
7 HCl
OMe
OMe
Scheme 4. Completion of the synthesis of (À)-govadine (1).
4. Hsu, B.; Kin, K. C. Arch. Int. Pharmacodyn. 1962, 89, 318.
5. Mo, J.; Guo, Y.; Yang, Y.-S.; Shen, J.-S.; Jin, G.-Z.; Zhen, X. Curr. Med. Chem. 2007,
14, 2996.
6. For in vitro and in vivo studies on THPBs, see: (a) Jin, G. –Z.; Huang, K.; Sun, B. –
C. Neurochem. Int 1992, 20, 175S; (b) Jin, G. –Z.; Sun, B. –C. Adv. Exp. Med. Biol.
1995, 363, 27; (c) Jin, G. –Z.; Zhu, Z. –T.; Fu, T. Trends Pharmacol. Sci. 2002, 23, 4.
7. (a) Mehra, K.; Garg, H. S.; Bhakuni, D. S.; Khanna, N. M. Indian J. Chem., Sect B
1976, 14B, 216; (b) Mehra, K.; Garg, H. S.; Bhakuni, D. S.; Khanna, N. M. Indian J.
Chem., Sect B 1976, 14B, 844.
8. Lapish, C. C.; Belardetti, F.; Ashby, D. M.; Ahn, S.; Butts, K. A.; So, K.; Macrae, C.
M.; Hynd, J. J.; Miller, J. J.; Phillips, A. G. Int. J. Neuropsychop. in press. Doi:
10.1017/S1461145711001581.
9. Both (À)-govadine and (+)-govadine are not commercially available.
10. For syntheses of ( )-govadine, see: (a) Kametani, T.; Satoh, M. J. Pharm. Soc. Jpn
1967, 179; (b) Chiang, H.-C.; Brochmann-Hanssen, E. R. J. Org. Chem 1977, 42,
3190; (c) Chiang, H.-C.; Cheng, Y.-C.; Liu, J.-C. Taiwan Yaoxue Zazhi 1978, 54, 30;
(d) Kametani, T.; Ihara, M. J. Chem. Soc., Perkin Trans. 1 1980, 629; (e) Yongzhou,
H. Chinese J. Med. Chem 1998, 8, 190; (f) Ref. 7a These routes access the desired
tetrahydroisoquinoline core utilizing racemic reductions of the corresponding
dihydroisoquinoline.
govadine regioisomer 7 (Scheme 3). To complete the synthesis, the
benzyl group was removed through hydrogenation, and the prod-
uct was crystallized as the HCl salt to afforded the desired govadine
salt in 92% yield (Scheme 4). While it has been noted that racemiza-
tion has been observed during benzyl hydrogenolysis of other THPB
analogs, no loss of enantioselectivity in the final product was
observed when the reaction was performed on small scale. However,
racemization was observed during scale-up (79% ee) due to in-
creased reaction times.19 To avoid racemization, the benzyl groups
can be deprotected using HCl to afford govadine hydrochloride in
73% yield. Using the opposite enantiomer of the Noyori catalyst,
the same route was utilized for the synthesis of (+)-govadine.
Preliminary in vitro testing of both enantiomers of govadine ob-
tained from this synthetic route provided different antagonist prop-
erties towards the dopamine D1 and D2 receptors.8 (À)-Govadine
11. Bischler, A.; Napieralski, B. Ber. 1893, 26, 1903.
12. For a representative example of a comparable peptide coupling, see: Khorna,
N.; Markmee, S.; Ingkaninan, K.; Ruchirwat, S.; Kitbunnada, R. J.; Pullagurla, M.
R. Med. Chem. Res. 2009, 18, 231.
13. For a representative example of a comparable peptide coupling, see: Yang, Y. –
S.; Ding, Y.; Sun, P. –H.; Mo, J.; Jin, G. –Z.; Ji, R. –Y. Faming Zhuanli Shenqing
Gongkai Shuomingshu 2007. CN 1900075 A.
was found to be a nanomolar dopamine D1 antagonist (EC50
=
5.6 Â 10À9) and moderate D2 antagonist properties (EC50 = 5.6 Â
10À5). In contrast, (+)-govadine displayed weak antagonist proper-
ties at both D1 (EC50 >1.0 Â 10À4) and D2 (EC50 = 1.4 Â 10À5).
Current in vivo studies on each enantiomer of govadine are currently
underway.
Overall, we have developed an efficient 5-step synthesis of the
potential antipsychotic (+)-govadine and (À)-govadine in 39% over-
all yield from commercially available materials. This marks the first
enantioselective synthesis of govadine and the one of the most effi-
cient enantioselective synthesis of any THPB. Furthermore, every
intermediate was purified by crystallization and no flash column
chromatography was necessary. Testing of both (+)- and (À)-gova-
dine as pharmaceutical agents in the treatment of schizophrenia
and other neuropathic diseases is currently underway.
14. For
a representative example of a thermal peptide coupling between
homovanillic acid and dopamine derivatives, see: Iwasa, K.; Cui, W.;
Takahashi, T.; Nishiyama, Y.; Kamigauchi, M.; Koyama, J.; Takeuchi, A.;
Moriyasu, M.; Takeda, K. J. Nat. Prod. 2010, 73, 115.
15. Uematsu, N.; Fujii, A.; Hashiguchi, S.; Ikariya, T.; Noyori, R. J. Chem. Soc. 1996,
118, 4916.
16. (a) Bedejs, E.; Trapencieris, P.; Suna, E. J. Org. Chem. 1999, 64, 6724; (b) Tietze, L.
F.; Rackelmann, N.; Sekar, G. Angew. Chem., Int. Ed. 2003, 42, 4254;
(c) Mujahidin, D.; Doye, S. Eur. J. Org. Chem 2005, 2689; (d) Werner, F.;
Blank, N.; Opatz, T. Eur. J. Org. Chem 2007, 3911.
17. Cheng, J. –J.; Yang, Y. –S. J. Org. Chem. 2009, 74, 9225.
18. For a description of the instability of a related isoquinoline, see: Pyo, M. K.; Lee,
D. H.; Kim, D. H.; Lee, J. H.; Moon, J. C.; Chang, K. C.; Choi, H. S. Y. Bioorg. Med.
Chem. Lett. 2008, 18, 4110.
19. For examples, see: (a) Parvulescu, A. N.; De Vos, D. E.; Jacobs, P. A. Chem.
Commun 2005, 42, 5307; (b) Parvulescu, A. N.; Jacobs, P. A.; De Vos, D. E. Chem.
Eur. J 2007, 13, 2034; (c) Parvulescu, A. N.; Van der Eycken, E.; Jacobs, P. A.; De
Vos, D. E. J. Catal. 2008, 255, 206; (d) Ref. 17.