Preparation and evaluation of 1,3-diaminocyclopentane-linked dihydropyrimidinone derivatives as selective α(1a)-receptor

Article abstract of DOI:10.1016/S0960-894X(00)00374-7

Several 1,3-diaminocyclopentane linked α(1a)-receptor antagonists were prepared using a divergent chemical strategy that allows for rapid analysis of all stereochemical permutations for their effect on α₁-receptor binding. (C) 2000 Published by Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00374-7

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1917±1920
Preparation and Evaluation of 1,3-Diaminocyclopentane-Linked
Dihydropyrimidinone Derivatives as Selective ꢀ1a-Receptor
Antagonists
James C. Barrow,a,* Kristen L. Glass,a Harold G. Selnick,a Roger M. Freidinger,a
Raymond S. L. Chang,b Stacey S. O'Malleyb and Carla Woydenb
aDepartment of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
bDepartment of Pharmacology, Merck Research Laboratories, West Point, PA 19486, USA
Received 30 May 2000; accepted 16 June 2000
AbstractÐSeveral 1,3-diaminocyclopentane linked a1a-receptor antagonists were prepared using a divergent chemical strategy that
allows for rapid analysis of all stereochemical permutations for their e€ect on a1-receptor binding. # 2000 Published by Elsevier
Science Ltd.
Recently, dihydropyrimidinone/piperidines such as 11
(Fig. 1) have been disclosed as selective a1a-receptor
antagonists, and these compounds have the potential to
be e€ective therapy for treatment of benign prostatic
hyperplasia.2 As part of a strategy to design potent and
selective antagonists of the a1a-receptor, constraints in
the ¯exible linker between the piperidine and dihy-
dropyrimidinone heterocycle of 1 were examined. Con-
straining a ¯exible section of an enzyme inhibitor or
receptor antagonist often results in enhanced binding
anity due to reduced entropic penalties on binding
when the constraint mimics the bioactive confor-
mation.3 While there are several conceivable ways
to constrain the 3-carbon chain of 1, 1,3-substituted
cyclopentanes such as 2 were chosen for this study. It
was hoped that, in addition to improving the potency at
the a1a-receptor, the four stereochemical permutations
of this cyclopentane system might reveal new strategies
for gaining selectivity among the a1a-, a1b-, and a1d-
receptor subtypes.4 Below is described a novel method
for synthesis of stereochemically de®ned 1,3-diamino-
cyclopentanes and evaluation of each con®guration at
the three a1-receptor subtypes.
preparation of optically active analogues of 3 was the
chiral auxillary-based approach of Miller.5 A more
¯exible approach that would provide entry into all ste-
reochemical permutations involves resolution of a 3-
amino cyclopentanol derivative 56 in analogy to the
known desymmetrization of 4a.7
Deardor€ has reported the palladium catalyzed opening
of epoxide 6 (Scheme 1) with several oxygen nucleo-
philes8 as well as displacement of 4b9 with NaN3 which
a€orded azide (+)8 in 44% yield as a 4:1 mixture with
the trans-isomer.10 It is possible to combine these two
procedures so that 6 can be opened with azide under
Pd(0) catalysis to give 8. Further, it was found that
replacing NaN3 with TMSN3 improved the yield and
selectivity of this reaction, most likely because the steric
bulk of the TMS group helps direct attack of azide
toward the distal end of the Pd-allyl system. Further-
more, the large TMS group stabilizes the allylic azide 7
from decomposition via sigmatropic rearrangements.11
Removal of the silyl protecting group from 7 (1 M aq
HCl/EtOAc, 15 min) a€orded the more labile allylic
azide 8 in racemic form. To minimize the number of
operations involving these somewhat unstable allylic
azide intermediates, a resolution procedure based on a
selective enzyme catalyzed acylation was chosen rather
than an acylation followed by selective hydrolysis. In
analogy to the work of Theil7 we employed the convenient
pancreatin/vinyl acetate system in THF.12 Using 10
equiv of vinyl acetate and 0.5 g of pancreatin/mmol of
substrate, the selective acylation of 8 proceeded to
The most rapid and convergent method for construction
of each possible stereoisomer of 2 would be a non-
selective reductive alkylation of a piperidine with enan-
tiopure ketone 3. At the outset of these studies, the only
*Corresponding author. Tel.: +1-215-652-4780; fax: +1-215-652-3971.
0960-894X/00/$ - see front matter # 2000 Published by Elsevier Science Ltd.
PII: S0960-894X(00)00374-7

1918
J. C. Barrow et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1917±1920
Figure 2.
by slow addition of NaCNBH3 a€orded excellent yields
of the alkylated products with a 2:1 ratio of cis to trans
isomers.17 Chromatographic separation followed by
removal of the Boc group (HCl, EtOAc) a€orded the
corresponding amines which could be coupled to the
resolved dihydropyrimidinone 19 under the previously
reported conditions.1 This process allowed for several
piperidines and a piperazine to be coupled with ketones
12 and 13 to give the set of products arrayed in Table 1.
Figure 1.
1
50% conversion in 3±4 h (as monitored by H NMR
analysis of aliquots) and the product was isolated by
®ltration through Celite followed by removal of the
solvent in vacuo. The unpuri®ed reaction mixture was
immediately subjected to reduction by 60 psi H2 over
10% palladium on carbon with in situ protection with
Boc anhydride.13 This procedure a€orded the stable and
easily separable acetate 10 and alcohol 9 in 88±92% ee14
at 50% conversion. The optical purity of either product
could be increased (at the expense of the other) by
modulating the resolution reaction conversion.
Each of the compounds prepared was screened for
binding anity toward the human cloned a1a-, a1b- and
a1d-receptors stably expressed in CHO, LM and HEK
cells, respectively.18 Interestingly, few distinct trends
emerge from Table 1 as there appears to be interplay
between the cyclopentane stereochemistry and the nature
of the piperidine or piperazine substituents. The most
potent compounds at the a1a-receptor subtype have
similar potency to the open chain variant 1 (Fig. 1) with
several of the stereochemical permutations. The best
con®guration for overall potency and a1a- subtype
selectivity is the (R,R) stereochemistry of 20c, 21c and
22c. The a1b- and a1d-receptor subtypes favor the (S,S)
cyclopentane con®guration (20d, 21d, 22d and 23d), but
these compounds were still somewhat a1a selective. The
piperazines 23a±d were quite potent but displayed low
subtype selectivity compared to the piperidines. Also
notable is the complementarity between the ortho-
cyanophenyl substituent on the piperidine and piperazine
with the (R,S) cyclopentane con®guration (22b, 23b).
That these compounds have sub-nanomolar potency
and good subtype selectivity for the a1a-receptor would
not have been predicted by consideration of the data for
20a±d and 21a±d alone.
Despite the sensitivity of some of the intermediates, the
overall yield of the four-step process was good (38% of 9,
30% of 10)15 due to rapid reactions and simple work up
procedures. Alcohol 9 could be oxidized directly to the
desired ketone 13 (COCl2, DMSO, Et3N), while acetate
10 was ®rst hydrolyzed (K2CO3, MeOH) then oxidized
(Scheme 1). The stereochemical course of the resolution
was con®rmed by comparison of optical rotation of
unreacted ( )8 with that reported for (+)8 by Deardor€,
and this is consistent with the analogous diol resolu-
tion.7 Interestingly, compounds (Æ)11, (Æ)14, and (Æ)15
were not good substrates for the pancreatin enzyme16
and were recovered unchanged when subjected to the
standard reaction conditions.
With the resolved ketones 12 and 13 in hand, the
attachment of the piperidine or piperazine segment was
examined (Scheme 2). Treatment of the free base of
piperidine 16 and the ketone 13 with acetic acid followed
The strategy of constraining the ¯exible linker of a1a-
receptor antagonists such as 1 into a 1,3-diaminocyclo-
pentane has resulted in the preparation of several potent
and receptor subtype selective antagonists, indicating
Scheme 1.

J. C. Barrow et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1917±1920
1919
Scheme 2.
Table 1. Comparison of in vitro binding Ki (nM) versus human cloned a1a-, a1b-, and a1d-receptorsa
a1a
a1b
a1d
a1a
a1b
a1d
a1a
a1b
a1d
a1a
a1b
a1d
A
B
C
1.0Æ0.15 1900Æ240 620Æ170 0.40Æ0.15 1300Æ680 320Æ60 0.24Æ0.10 130Æ31 16Æ5.4 1.6Æ0.45 28Æ1.5 7.5Æ3.6
2.4Æ0.45
730Æ380 660Æ400
4.7Æ1.2
660Æ330 0.38Æ0.01 350Æ110 500Æ20 0.46Æ0.18 91Æ20 31Æ6.4 0.90Æ0.21 14Æ1.5 5.4Æ3.3
95Æ20 8.3Æ2.8 130Æ5 170Æ10 2.4Æ0.90 36Æ16 11Æ3.8 4.8Æ0.50 15Æ1.5 9.0Æ3.0
290Æ95
550Æ65 0.60Æ0.35 70Æ19 97Æ5.8 0.44Æ0.09 51Æ10 39Æ20
0.37Æ0.12
210Æ75
D
2.3Æ0.25
120Æ2.5
aValues represent the mean Æ SEM (nM) for displacement of 125I-HEAT from the human cloned receptor subtypes.
2. Chapple, C. R. Eur. Urol. 1996, 29, 129.
3. Khan, A. R.; Parrish, J. C.; Fraser, M. E.; Smith, W. W.;
Bartlett, P. A.; James, M. N. G. Biochemistry 1998, 37, 16839.
4. De Benedetti, P. G.; Fanelli, F.; Menziani, M. C.;
Cocchi, M.; Testa, R.; Leonardi, A. Bioorg. Med. Chem.
1997, 5, 809.
5. Ghosh, A.; Ritter, A. R.; Miller, M. J. J. Org. Chem. 1995, 60,
5808. See also Ramesh, N. G.; Kiunder, A. J. H.; Zwanenburg, B.
J. Org. Chem. 1999, 64, 3635.
6. Since completion of this work, a similar enzymatic resolu-
tion was reported: Mulvihill, M. J.; Gage, J. L.; Miller, M. J.
J. Org. Chem. 1998, 63, 3357.
that this particular constraint mimics the bioactive con-
formation of 1. Surprisingly, many of the stereo-
chemical permutations were tolerated indicating some
receptor ¯exibility. The divergent chemical strategy uti-
lized to prepare these compounds allowed for expedient
preparation of all possible isomers from the common
starting material (Æ)8 and uses low-cost, readily available
reagents. This process is quite general and ¯exible, and
should prove useful in other contexts as 4-aryl piperidines
and piparazines are common units in bioactive molecules.
7. Theil, F.; Schick, H.; Lapitskaya, M. A.; Pivnitsky, K. K.
Liebigs Ann. Chem. 1991, 195.
8. Deardor€, D. R.; Myles, D. C. Tetrahedron Lett. 1985, 26,
5615.
Acknowledgements
The authors would like to thank Carl Homnick for ee
determination of 12 and 13, Sandor Varga for NMR
assignment of 17 and 18, and Mary Becker for assis-
tance in preparation of this manuscript.
9. Deardor€, D. R.; Windham, C. Q.; Craney, C. L. Org.
Synth. 1995, 73, 25.
10. Deardor€, D. R.; Linde, R. G.; Martin, A. M.; Shulman,
M. J. J. Org. Chem. 1989, 54, 2759.
11. Compound 7 can be stored for at least 3 months at 0 ꢀC
without signi®cant decomposition.
12. Pancreatin is exceptionally cheap ($36.40/kg Aldrich) and
the reaction does not require aqueous conditions.
13. Saito, S.; Nakajima, H.; Inaba, M.; Moriwake, T. Tetra-
hedron Lett. 1989, 30, 837.
14. The ee was determined after conversion to ketones 12 and
13 by HPLC analysis on a 4.6Â250 mm ChiracelTM AD chiral
column, 10% EtOH/Hexane 1.5 mL/min, detected by UV
absorbance at 200 nM.
References and Notes
1. Dhar, T. G. M.; Nagarathnam, D.; Marzabadi, M. R.;
Lagu, B.; Wong, W. C.; Chiu, G.; Tyagrajan, M. R.; Miao, S.
W.; Zhang, F.; Sun, W.; Tian, D.; Shen, Q.; Zhang, J.; Wetzel,
J. M.; Forray, C.; Chang, R. S. L.; Broten, T. P.; Schorn, T.
W.; Chen, T. B.; O'Malley, S.; Ransom, R.; Schneck, K.;
Bendesky, R.; Harrell, C. M.; Vyas, K.; Zhang, K.; Gilbert, J.;
Pettibone, D. J.; Pantane, M. A.; Bock, M. G.; Freidinger, R.
M.; Gluchowski, C. M. J. Med. Chem. 1999, 42, 4778.
15. The lower yield of the acetylated product is probably
due to Pd catalyzed decomposition of the allylic acetate

1920
J. C. Barrow et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1917±1920
during the reduction step. When lower H2 pressures were used
(resulting in longer reaction times), the yield of 10 was even
lower.
pounds were assigned by analogy (ratio of products, TLC Rf,
1H NMR). We thank Dr. S. Varga for performing the NMR
analysis.
16. Compound (Æ)14 was shown by Miller to be a substrate
17. 2D-COSY and 1-D NOE 1H NMR experiments on 17 and
18 con®rmed the stereochemical assignment. The other com-
18. O'Malley, S. S.; Chen, T. B.; Francis, B. E.; Gibson, R. E.;
Burns, H. D.; DiSalvo, J.; Bayne, M. L.; Wetzel, J. M.;
Nagarathnam, D.; Marzabadi, M.; Gluchowski, C.; Chang, R.
S. L. Eur. J. Pharm. 1988, 348, 287.
for several other enzymes. See ref 6.