Efficient synthesis of highly functionalized tetrahydropyridopyrimidines by a novel three-component coupling reaction

Article abstract of DOI:10.1016/j.tetlet.2011.05.071

We report the development of a novel three-component coupling reaction (3CC) for the synthesis of alkoxy tetrahydropyridopyrimidines. Systematic optimization of reaction parameters identified 3CC conditions tolerant of a wide array of functionality at all three sites of diversity, providing densely functionalized products in moderate to excellent yields. The newly developed chemistry has since been applied to the lead optimization process on a novel drug discovery program, facilitating rapid compound advancement.

Full text of DOI:10.1016/j.tetlet.2011.05.071

Tetrahedron Letters 52 (2011) 3849–3852
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Efficient synthesis of highly functionalized tetrahydropyridopyrimidines
by a novel three-component coupling reaction
⇑
Izzat T. Raheem , John D. Schreier, Michael J. Breslin
Department of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We report the development of a novel three-component coupling reaction (3CC) for the synthesis of alk-
oxy tetrahydropyridopyrimidines. Systematic optimization of reaction parameters identified 3CC condi-
tions tolerant of a wide array of functionality at all three sites of diversity, providing densely
functionalized products in moderate to excellent yields. The newly developed chemistry has since been
applied to the lead optimization process on a novel drug discovery program, facilitating rapid compound
advancement.
Received 3 February 2011
Revised 12 May 2011
Accepted 16 May 2011
Available online 23 May 2011
Keywords:
Ó 2011 Elsevier Ltd. All rights reserved.
Three-component coupling
Tetrahydropyridopyrimidine
Pharmacophore
Reaction optimization
1. Introduction
In our laboratories, we recently became interested in synthesiz-
ing analogs with a 4-alkoxy-5,6,7,8-tetrahydropyrido[4,3-d]pyrim-
The ability to rapidly assemble densely functionalized heterocy-
clic pharmacophores provides a significant advantage within the
drug discovery lead optimization space. Multiple-component cou-
plings (MCCs) comprise a family of convergent reactions in which
three or more starting materials react to form a single product,
with nearly all of the atoms involved contributing to the structure
of the newly formed product.1 In medicinal chemistry, the applica-
tion of MCCs toward the synthesis and optimization of leading
structural series often provides facile variation of two or more
structural diversity elements and facilitates efficient development
of SAR and subsequent program advancement. Herein, we describe
a novel three-component coupling (3CC) reaction for the synthesis
of highly functionalized alkoxy tetrahydropyridopyrimidines.
The tetrahydropyridopyrimidine core is a key pharmacophore
in several reported preclinical drug candidates, including selective
P2X7 receptor antagonists,2 inhibitors of platelet aggregation,3 gas-
tric anti-lesion agents,4 and phosphodiesterase 10 inhibitors.5
While a number of stepwise approaches toward the synthesis of
this heterocycle exists, only one multi-component reaction has
been reported involving the acid-catalyzed condensation of a pipe-
ridinone with 2 equivalents of a functionalized nitrile.6 Although
highly convergent, this approach allows for the installation of only
a single functionality off of the aromatic portion of the heterocycle,
limiting product versatility.
idine scaffold. The original three-step synthesis of the core
involved (1) cyclocondensation between an amidine and a piperid-
inone ester, (2) chlorination with PPh3/NCS, and (3) nucleophilic
aromatic substitution (Fig. 1). While this synthetic route was ro-
bust and versatile, typically providing products in overall yields
ranging from 10% to 30%, we were attracted to the possibility of
employing a more convergent synthesis that harnessed the nucle-
ophilic character of the intermediate pyrimidinone oxygen. This
would allow for a potential one-step assembly of the fully func-
tionalized heterocyclic framework. We reasoned that as long as
the reaction medium remained devoid of complicating nucleo-
philic species until the formation of the pyrimidinone, we could
fine tune the reaction parameters to facilitate construction of the
O
O
O
N
R2
PPh3
NCS
R1
OEt
H2N NH
HN
R2
N
N
K2CO3, EtOH
R1
2
1
Cl
OR3
R3OH
R1
R1
NaH, THF
N
N
N
N
R2
R2
N
N
4
3
⇑
Corresponding author.
E-mail address: izzat_raheem@merck.com (I.T. Raheem).
Figure 1. First generation synthesis of 4-alkoxy tetrahydropyridopyrimidines.
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.05.071

3850
I. T. Raheem et al. / Tetrahedron Letters 52 (2011) 3849–3852
Table 1
Optimization of reaction parametersa
OCH2CH3
NH
NH2
O
Boc
CH3CH2Br
Boc
N
N
+
O
N
base
N
solvent
N
HCl
O
temperature
N
7
6
5
Entrya
Base
Solvent
Temp. (°C)
Conv.b (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
None
CH2Cl2
THF
DMF
DMSO
Toluene
EtOH
DMF
DMF
DMF
DMF
DMF
50
50
50
50
50
50
30
70
90
70
70
70
70
<5
<5
87
100c
<5
100d
40
100
100
<5
100
<5
Cs2CO3
TEA
NaHCO3
DMF
DMF
<5
a
Screening reaction conditions: 1.1 equiv 5, 1 equiv 6, 2.5 equiv CH3CH2Br, 2.5 equiv base, 0.15 M solvent, 2 h reaction time.
Determined by HPLC at 254 nm.
Significant decomposition.
b
c
d
Trace conversion to fully alkylated product; reaction stalls at the pyrimidinone intermediate.
heterocycle as a 3CC by introducing the alkoxy functionality as an
electrophile rather than a nucleophile.
group by installing the final required functionality in the first step.
Surprisingly, no conversion was observed in the presence of ben-
zylic functionality, highlighting the importance of amine basicity,
with tertiary amines too basic to be employed in the reaction (en-
try 5).
Finally, we examined the scope of the alkyl halide component
(Table 4). The chemistry was again tolerant of a wide variety of
functionality, including strain-chain and branched aliphatic (en-
tries 1–3), heteroaromatic (entry 4), as well as both cyclic and acy-
clic heteroaliphatic functionalities (entries 5–7). Not surprisingly,
the identity of the leaving group was found to be very important.
When activated electrophiles such as allylic or benzylic halides
were employed, the bromide leaving group proved too reactive
for the 3CC, providing unfavorable reaction profiles with low con-
version to the desired products. This is likely due to the competing
alkylation of the amidine species. For these substrates, the corre-
sponding chloride was found to be optimal, as in the cases of both
As a model system, we examined the 3CC of 3-pyridylamidini-
um hydrochloride (5), N-Boc-piperidinone ester 6, and 2-bromo-
propane in the presence of K2CO3 (Table 1). Not surprisingly, the
identity of the solvent proved crucial, as only polar aprotic solvents
provided conversion to the desired product, likely due in part to
the poor solubility of the intermediate pyrimidinone (Table 1, en-
tries 1–6). CH2Cl2, THF, and toluene afforded only trace product,
while EtOH furnished only the pyrimidinone intermediate. Gratify-
ingly, both DMSO and DMF provided high conversion at only 50 °C,
with DMF affording a significantly more favorable overall reaction
profile. The reaction temperature was examined next (Table 1,
entries 7–9), with full conversion achieved in reasonable time at
temperatures above 50 °C. We selected 70 °C as the optimal
temperature as it afforded the best compromise between reaction
profile and reaction rate. Finally, the identity of the base was exam-
ined (Table 1, entries 10–13). Inorganic carbonate bases provided
complete conversion, while only trace conversion was achieved
with bicarbonate and amine bases; K2CO3 was selected as optimal
for this transformation. Importantly, the use of powdered K2CO3
was crucial, as the use of granular inorganic bases provided low
and variable yields.
With a set of optimal reaction conditions in hand, we examined
the scope of the 3CC with respect to all three reaction compo-
nents.7 We were pleased to find that the reaction tolerated a wide
array of amidines, including aromatic, heteroaromatic, straight-
chain aliphatic, cyclic aliphatic, and benzylic substitution patterns,
providing the desired products in moderate to good yields (Table 2,
entries 1–8). Additionally, the 3CC was tolerant of formamidine
and dimethylguanidine, providing both the unsubstituted (entry
9) and amino-substituted (entry 10) products, albeit with compro-
mised yields.
Table 2
Scope of amidine component
OiPr
NH
NH2
X-H
O
O
Boc
2-iodopropane
N
N
Boc
R
+
O
N
K2CO3
DMF
R
N
70 °C
Entry
R
Product
X
Yield (%)
1
2
3
4
5
6
7
8
9
C6H5
4a
4b
4c
4d
4e
4f
4g
4h
4i
Cl
Cl
Cl
Cl
79
61
57
38
61
74
70
81
33
52
3-BrC6H4
4-Pyridyl
3-Thienyl
2-Methyl-1,3-thiazol-4-yl
Cl
The amino substituent was examined next with a representa-
tive set of functional groups. Apart from the carbamate group em-
ployed in the above chemistry (Table 3, entry 1), both amide and
urea functionalities were tolerated at this position (entries 2–4).
This flexibility allows one to obviate the need for a protecting
Propyl
Benzyl
Cyclopropyl
H
OAc
Br
Cl
Cl
Cl
10
N(CH3)2
4j

I. T. Raheem et al. / Tetrahedron Letters 52 (2011) 3849–3852
3851
Table 3
Scope of amine component
OCH2CH3
NH
NH2
O
R
CH3CH2Br
R
N
N
+
O
N
K2CO3
N
DMF
HCl
O
70 °C
Entry
1
R
Product
Yield (%)
Boc
4k
82
O
2
4l
62
N
CH3
Cl
O
O
3
4
4m
71
N
H
N
N
CH3
4n
4o
61
0a
N
H
5
a
No significant conversion of starting material was observed.
Table 4
Scope of alkyl halide component
OR
NH
NH2
O
O
Boc
R-X
N
N
Boc
+
O
N
K2CO3
DMF
70 °C
N
HCl
Entry
R
Product
X
Yield (%)
1
2
3
Methyla
n-Propyl
Cyclopentyl
4p
4q
4r
Br
Br
Br
53
87
73
N
4
4s
Br
28
OCH3
5
6
4t
Br
Br
54
51
NHBoc
O
4u
7
4v
Br
47
O
Benzyl
8
9
4w
4x
Cl
Cl
61
77
Allyl
10b
4y
4y
Br
Cl
87
29
11c
a
Due to volatility, 5 equiv CH3Br was added as a 2.0 M solution in TBME.
Reaction time = 2 h; reaction temp. = 70 °C.
Reaction time = 4 h; reaction temp. = 100 °C.
b
c
benzyl and allyl chloride (entries 8–9). When unactivated electro-
philes were employed, the reaction proceeded most efficiently
with a bromide leaving group. However, with elevated tempera-
tures and extended reaction times, the less reactive corresponding
chlorides could be employed (Table 3, entries 10–11).
2. Conclusion
We have introduced a novel 3CC reaction that assembles densely
functionalized tetrahydropyridopyrimidines in a single efficient
transformation. Systematic optimization of reaction parameters

3852
I. T. Raheem et al. / Tetrahedron Letters 52 (2011) 3849–3852
2. Carroll, W. A.; Donnelley-Roberts, D.; Jarvis, M. F. Purinergic Signalling 2009, 5,
63–73.
3. U.S. Patent US005405848A; filed April 11, 1995.
4. Sanfilippo, P. J.; Urbanski, M.; Williams, L.; Press, J. B.; Katz, L. B.; Shriver, D. A.;
Fernandex, J. A.; Shatynski, D.; Offord, S. J. Eur. J. Med. Chem. 1992, 27, 655–661.
5. U.S. Patent Application WO2010138430; filed December 2, 2010.
6. Herrera, A.; Martinez-Alvarez, R.; Chioua, R.; Almy, J. Tetrahedron Lett. 2006, 47,
5463–5465.
identified a protocol that is both operationally simple and tolerates a
wide array of functional groups. In the context of our own drug
discovery program, this reaction has not only streamlined our lead
optimization process, but has also significantly impacted our
scale-up efforts as the transformation has been repeatedly executed
on >25-gram scale with excellent results. Current efforts are aimed
at expanding the substrate scope at both the amine and alkoxy
positions, as well as better understanding the pKa dependence of
the reaction. Our results will be disclosed in due course.
7. Representative experimental procedure: tert-butyl-2-phenyl-4-(prop-2-en-1-
yloxy)-7,8-dihydropyrido[4,3-d]pyrimidine-6(5H)-carboxylate (4x).
A 10 mL
round bottom flask was charged with phenylamidinium hydrochloride
(31.7 mg, 0.203 mmol), 3-carbethoxy-4-piperidone hydrochloride (50 mg,
0.184 mmol), powdered K2CO3 (63.7 mg, 0.461 mmol), and a stirbar. The flask
was sealed and charged with DMF (1.23 mL) followed by allyl chloride (37.9 lL,
Acknowledgments
0.461 mmol). With vigorous stirring, the reaction mixture was heated to 70 °C
for 2 h, and reaction progress was monitored by LCMS. Upon completion of the
reaction, the reaction mixture was diluted with EtOAc (10 mL), and washed with
sat. aq NaHCO3 (10 mL) and brine (10 mL). The combined aqueous phases were
back-extracted with EtOAc (5 mL), and the combined organic phases were dried
over Na2SO4, filtered, and concentrated in vacuo. The resulting residue was
purified by gradient elution on silica gel (0–25% EtOAc/hexanes) to afford the
title compound as a colorless oil that slowly solidified upon standing (52.3 mg,
77%). 1H NMR (500 MHz, CDCl3): d 8.36–8.44 (m, 2H), 7.40–7.48 (m, 3H), 6.05–
6.17 (m, 1H), 5.44 (d, J = 17.3 Hz, 1H), 5.29 (d, J = 10.5 Hz, 1H), 5.05 (b s, 2H), 4.50
(b s, 2H), 3.75 (t, J = 5.5 Hz, 2H), 2.94 (b s, 2H), 1.51 (s, 9H) ppm; 13C NMR
(125 MHz, CDCl3): d 165.34, 161.65, 154.81, 137.64, 132.76, 130.39, 130.37,
128.38, 128.01, 118.03, 80.26, 66.89, 41.38, 40.50, 31.66, 28.46; HRMS (ES)
calculated M+H for C20H21N7O2S: 368.1963l; found:368.1969.
The authors thank the Merck West Point NMR and Mass Spec-
trometry facilities for assistance in characterizing the compounds
presented in this manuscript. Dr. Chris Cox and Mr. Thomas Steele
are also gratefully acknowledged for useful interactions.
References and notes
1. For representative reviews of MCCs see: (a) Armstrong, R. W.; Combs, A. P.;
Tempest, P. A.; Brown, S. D.; Keating, T. A. Acc. Chem. Res. 1996, 29, 123–131; (b)
Uqi, I. Pure Appl. Chem. 2001, 73, 187–191; (c) Dömling, A.; Ugi, I. Angew. Chem.,
Int. Ed. 2000, 39, 3168; (d) Dömling, A. In Multicomponent Reactions; Zhu, J.,
Bienayme, H., Eds.; Wiley-VCH: Weinheim, 2005; p 76; (e) Biggs–Houck, J. E.;
Ashkaan, Y.; Shaw, J. T. Curr. Opin. Chem. Biol. 2010, 14, 371–382.