Synthesis of the 2H-quinolizin-2-one scaffold via a stepwise acylation-intramolecular annulation strategy

Article abstract of DOI:10.1016/j.tetlet.2006.05.089

A rapid entry into the 2H-quinolizin-2-one starting from 2-alkyl pyridine has been developed. Initial deprotonation of a 2-alkyl pyridine followed by acylation with a β-TMS-propyonate derivative provides acyclic precursors that after deprotection undergoe

Full text of DOI:10.1016/j.tetlet.2006.05.089

Tetrahedron Letters 47 (2006) 5063–5067
Synthesis of the 2H-quinolizin-2-one scaffold via a
stepwise acylation—intramolecular annulation strategy
a,
a,
a
*
*
Swaminathan R. Natarajan, Meng-Hsin Chen, Stephen T. Heller,
a
a
b
b
Robert M. Tynebor, Ellen M. Crawford, Cui Minxiang, Han Kaizheng,
b
b
b
b
Jingchao Dong, Bin Hu, Wu Hao and Shu-Hui Chen
a
Departments of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
Department of Chemistry, WuXi Pharmatech, Bldg 1, #288FUTE Zhonglu, Waigaoqiao TZ, Shanghai 200131, China
b
Received 13 May 2006; accepted 15 May 2006
Abstract—A rapid entry into the 2H-quinolizin-2-one starting from 2-alkyl pyridine has been developed. Initial deprotonation of a
-alkyl pyridine followed by acylation with a b-TMS-propyonate derivative provides acyclic precursors that after deprotection
2
undergoes a 6-endo-trig cyclization to yield the desired 2H-quinolizin-2-one derivative. This synthetic strategy was found to be gen-
erally applicable as evidenced from the various examples in this letter.
Ó 2006 Elsevier Ltd. All rights reserved.
1
The 2H-quinolizin-2-one scaffold can be an attractive
8
5
-O
O
7
6
structural template for drug discovery. Calculated low
LogD (0.8 at all pH) values for this charge neutral mole-
cule were as surprising as they were encouraging. The
origin of low LogD values lies in the very large dipole
2
2
3
N+
N
4
1
3
moment, characteristic of the quinolizin-2-one scaffold
HO
HO
X
X
X
which in turn arises aromaticity driven net internal
charge separation. Compared to closely related drug dis-
covery platforms such as corresponding hydroxy naph-
thalene, hydroxy quinoline or pyrimido-pyridine
derivatives (Fig. 1), the 2H-quinolizin-2-ones were
found to be significantly more polar as judged by HPLC
and TLC retention times and greater solubilities.
X
2
X = CH or N
H
N
H
N
O
O
N
Despite the appeal offered by the 2H-quinolizin-2-one
scaffold, its use as a template for drug discovery has lar-
gely been unexplored. The paucity of synthetic method-
ology to access these structures readily is thought to be
the main obstacle. In connection with a recent medicinal
chemistry program, a 2H-quinolizin-2-one structural
motif was designed as part of a drug discovery platform.
The unique properties that manifested in the resulting
drug molecules attributable to 2H-quinolizinones cre-
ated increased interest. In order to exploit this finding,
we were faced with the task of accessing quinolizin-2-
3
4
Figure 1. 2H-quinolizin-2-one canonical structures and representative
structures of closely related platforms.
ones in a straightforward and general fashion. A search
of the literature revealed two synthetic strategies.
Fozard and Jones were able to displace bromo-quinol-
izinium salts with silver acetate followed by hydrolysis
of the corresponding acetoxy group to unravel the
desired quinolizin-2-one. Not only was the bromine
displacement a low yielding reaction, but also this
strategy demands the synthesis of corresponding bromo-
quinolizinium salt, which poses enormous limitations
4
*
Corresponding authors. Tel.: +1 732 594 0939; fax: +1 732 594
080; e-mail: ravi_natarajan@merck.com
8
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.05.089

5064
S. R. Natarajan et al. / Tetrahedron Letters 47 (2006) 5063–5067
N
N
O
X
Y
O
N
+
+
N
5
N
5
6
Ph
a
OH
O
O
N
Y
N
Y
N
Ph
7
Scheme 1. Katritzky approach to the 2H-quinolizin-2-one scaffold.
Reagents and conditions: (a) MeCN, sealed tube, 120 °C, 12 h.
O
N
1
on the utility of this synthetic approach. Kato and Ats-
umi achieved the synthesis of the quinolizin-2-one scaf-
5
Figure 2. Strategy toward the 2H-quinolizin-2-one scaffold.
fold by means of condensation of pyridylacetonitrile
derivatives with diketene. This approach is conceptually
simple but offers limited operational flexibility and
generality.
to furnish dihydro-quinolizin-2-one derivatives (Scheme
). Thus, treatment of 2-picolinyl anion, generated by
deprotonation with 2.1 equiv of LDA at À78 °C, with
.1 equiv of 3-chloro-propionyl chloride, gave the
2
In a more recent publication, Katritzky and co-workers6
have reported a one pot synthetic approach (Scheme 1)
to 2H-quinolizin-2-one core. Treatment of 2-picoline
derivatives with a prepared 1,3-bis electrophile such as
1
expected C-acylation derivative but the product isolated
after a standard work-up was the diisopropyl amine
adduct 10 and not the desired cyclized di-hydro quinoli-
zin-2-one (8). Competing displacement of pendant chlo-
ride by diisopropyl amine can be solved by the use of
LiHMDS as the base in this reaction instead. This was
not explored at the time of this publication. When 3-
ethoxy-methyl acrylate was used as the electrophile,
the C-acylated seco product was isolated in a clean reac-
tion. The cylization was induced cleanly in a separate
step by heating at 120 °C in methanol.
1
1
-benzotriazol-1-yl-3-propynone in a sealed tube at
20 °C for 12 h afforded modest yields of desired quinol-
izin-2-ones. Although this approach is relatively
straightforward, sealed tube treatment at high tempera-
tures may not be tolerant of a wide range of picoline and
propynone derivatives. Furthermore, it requires a
separate synthesis of specially activated benzotriazolyl
propynoate, suggesting that readily available alkyl
propynoates or the corresponding acid chlorides do
not participate in this reaction thus restricting general
applicability. In this communication a simple step-wise
approach toward the assembly of 2H-quinolizin-2-one
scaffold using commercially available 1,3-bis electro-
philes is described.
a
N
N
Cl
O
5
Thus, base mediated deprotonation of 2-picoline (5) at
low temperatures, followed by regioselective C-acylation
of the resulting anion with a suitable 1,3-bis electrophile,
was expected to give seco derivative (Fig. 2) which is set
up to undergo a 6-endo-cyclization to furnish the
desired quinolizin-2-one scaffold in the same reaction.
In the event the ring closure did not occur spontan-
eously as anticipated, it could be induced thermally in
a separate step. The depicted strategy is expected to be
virtually tolerant of most 2-picoline derivatives and will
represent a general approach to access all types of
quinolizin-2-ones. The scope of this approach will be
dictated by the choice of 1,3-bis electrophilic synthon
selected for the synthesis and the ease of operation.
O
b
c
N
8
N
N
iPr
Et
O
O
N
iPr
O
9
10
Scheme 2. Acylation of 2-picoline. Reagents and conditions: (a)
Initial C-acylation reactions were attempted with readily
available 3-chloro-propionyl chloride, 3-ethoxy-methyl
acrylate, and 3-trimethylsilyl methyl propynoate as the
2
.1 equiv of LDA, À78 °C, THF, 0.5 h then 1.1 equiv of 3-chloro-
propionyl chloride, À78 °C to rt, 1 h, 45%. (b) 2.1 equiv of LDA,
À78 °C, THF, 0.5 h then 1.1 equiv of 3-methoxy-methyl acrylate,
À78 °C to rt, 1 h, 53%.
1
,3-bis electrophilic synthons. The former was expected

S. R. Natarajan et al. / Tetrahedron Letters 47 (2006) 5063–5067
OMe
5065
O
In the use of 3-trimethylsilyl propynoate as the 1,3-bis
electrophile it was noted that the product of initial C-
acylation might not cyclize owing to geometric and ste-
ric considerations and that the ring closure might have
to be effected in a separate step under forcing condi-
tions. Thus treatment of 2-picolinyl anion with 1.5 equiv
of b-TMS methyl propynoate at À78 °C gave after stan-
dard aqueous work-up the C-acylated product 12 exclu-
+
N
TMS
5
11
a
sively in 85% yield. The absence of the picolinyl
1
O
OH
methylene proton in the H NMR spectrum (CD OD)
3
is conspicuous as the molecule exists predominantly in
its enolized form. The chemical shift of the methyne pro-
ton of the enolized tautomer was at 5.72 ppm. Attempts
to carry out the cyclization of 12 were largely unsuccess-
ful as anticipated. Upon removal of the silyl protecting
group the cyclization precursor 13 was obtained which
underwent a surprisingly facile ring closure upon heat-
ing to 50 °C for 0.5 h (Scheme 3). The overall yield for
annulation of a 1,3-bis electrophile such as 3-TMS-pro-
pynoate onto 2-picoline was 60%.
N
N
TMS
TMS
1
1
2
b
O
OH
N
N
H
H
3
Further discussions will involve this reaction utilizing
propynoates as electrophiles. The use of methoxy acryl-
ates and 3-chloropropionyl chlorides as electrophiles
are still under investigation and will be reported
separately.
c
O
N
Apart from 3-trimethyl silyl propynoate, the use of 3-
methyl propynoate (14) and 3-phenyl propynoate (15)
were also investigated in these annulation reactions.
As can be seen in Table 1 (Ref. ) a representative set
of six 2-picoline derivatives were used as test substrates.
These picoline substrates can be divided into two
1
Scheme 3. Assembly of the 2H-quinolizin-2-one scaffold. Reagents
and conditions: (a) 2.1 equiv of LDA, À78 °C, THF, 0.5 h then
7
1
.1 equiv of 3-TMS-CCCOOMe, À78 °C to rt, 1 h, 85%. (b) 1.1 equiv
of TBAF, THF, rt, 0.5 h, 80%. (c) 50 °C, MeCN, 100%.
Table 1. Selected examples of quinolizin-2-ones using the indicated picoline starting materials and propynoate electrophiles
O
O
O
MeO
O
MeO
O
MeO
O
Me
Ph
TMS
1
4
15
11
N
NHMe
N
NHMe
N
N
NHMe
N
N
NHMe
O
Me
O
H
H
O
Ph
Ph
O
1
6
21 (45)
23 (48)
2
2 (35)
O
O
O
O
O
O
N
N
Me
CN
5
2
2
4 (52)
5 (49)
7 (39)
1 (60)
CN
CN
CN
N
N
N
N
Me
Ph
H
1
7
26 (37)
27 (50)
(continued on next page)

5066
S. R. Natarajan et al. / Tetrahedron Letters 47 (2006) 5063–5067
Table 1 (continued)
O
O
O
MeO
O
MeO
O
MeO
O
Me
Ph
TMS
1
4
15
11
Ph
N
Ph
Ph
Ph
N
N
N
N
Me
Ph
Ph
H
H
1
8
2
3
8 (55)
29 (33)
30 (46)
O
Cl
O
Cl
O
Cl
Cl
N
N
N
Me
1
9
1 (53)
32 (23)
33 (45)
O
O
O
N
N
N
N
Cl
Me Cl
4 (41)
Ph Cl
35 (36)
H
Cl
36 (43)
2
0
3
Numbers in parentheses are uncorrected overall yields.
groups. Compounds 16, 5, 17, and 18 bear a nucleophilic
pyridine moiety while in 19 and 20 the chlorine substitu-
ents at the 2- and 4- positions attenuate the nucleophilic-
ity of the pyridine nitrogen atom. In reactions with
O
O
N
N
N
N
Me
3
7
38 (58)
3
-methyl propynoate 14, after the initial acylation ring
closure was found to occur spontaneously. In some
examples, the crude reaction had to be heated to
N
N
1
00 °C to complete the ring closure reaction. In the reac-
N
N
N
tion with 15 it was necessary to heat the reactions to
effect ring closure. The use of propynoate 11 gave seco
products that could be isolated cleanly. Removal of
TMS protecting group was carried out first in order to
effect ring closure in these examples. Weinreb amide deriv-
atives could also be used as effective acylating agents.
Me
3
9
40 (47)
O
O
N
N
N
Me
4
1
42 (52)
In addition to the examples shown in Table 1, this reac-
tion sequence was extended to encompass the synthesis
of aza-analogs of the quinolizin-2-one template. As
indicated in Figure 3 3-methyl pyridazine, 4-methyl pyr-
imidine, and 3-methyl pyrazines underwent smooth
deprotonation under the usual conditions and were suc-
cessfully acylated with 3-methyl propynoate to yield the
corresponding aza-quinolizin-2-one, respectively, in
modest to good overall yields as indicated.
N
N
N
N
Me
4
3
44 (39)
Figure 3. Aza-analogs of the quinolizin-2-one scaffold.
analogs with ease. Considering the value brought to any
drug discovery program by designing these substructures
into drug molecules, it is expected that this strategy will
get the opportunity for further improvements.
In conclusion, a straightforward strategy toward the
assembly of quinolizin-2-one template has been devised.
This method is expected to allow easy access to virtually
all kinds of quinolizin-2-ones by appropriate choice of
starting picoline derivative and 1,3-bis electrophile. The
mild reaction conditions used is expected to be tolerant
of all kinds of functional groups except carboxylic acid
esters. This methodology has also been extended to aza-
References and notes
1. For use of benzo-quinolizinones as CFTR channel inhib-
itors see Marivingt-Mounir, C.; Norez, C.; Derand, R.;

S. R. Natarajan et al. / Tetrahedron Letters 47 (2006) 5063–5067
5067
Bulteau-Pignoux, L.; Nguyen-Huy, D.; Viossat, B.; Mor-
gant, G.; Becq, F.; Vierfond, J.-M.; Mettey, Y. J. Med.
Chem. 2004, 47, 962, and references therein.
. Calculated values determined using ACD Labs. The calcu-
lated LogD values for hydroxy naphthalene, hydroxy
quinoline, and pyrimido-pyrimidine templates as indicated
in Figure 1 range from 1.5 to 2.5 at pH 7. Experimentally
determined LogD values for similar analogs are in
agreement.
. Calculated dipole moments for the 2H-quinolizinone scaf-
fold (1) was 7.5 Debye units compared to hydroxy
naphthalene (2), hydroxy quinoline (3), and pyrimido-
pyrimidine (4) at 1.7, 6.7, and 4.6 Debye units, respectively.
. (a) Fozard, A.; Jones, G. J. Chem. Soc. Abstr. 1964, 2760–
5. Kato, T.; Atsumi, T. Chem. Abstr. 1968, 68, 49422g.
6. (a) Katritzky, A. R.; Rogers, J. W.; Witek, R. M.; Nair,
S. K. ARKIVOC 2004, viii, 52–60; (b) Murthi, G. S. S.;
Gangopadhyay, S. K. Indian J. Chem. 1979, 17, 20.
2
7. Representative procedure: synthesis of 7. 2-Picoline
(1 mmol) was dissolved in 15 mL of dry THF, cooled to
À78 °C, and deprotonated with 2.1 mmol of freshly made
LDA. After 0.5 h the reaction was quenched with 1.5 mmol
of propynoate 15. The reaction was allowed to warm-up to
rt over 4 h and quenched by addition of satd ammonium
chloride. Standard aqueous work-up followed by heating
the crude in acetonitrile at 80 °C furnished the product.
After removal of solvent the residue was purified by silica
3
4
gel chromatography using 1–5% methanol in acetone to
1
2
763; (b) Sanders, G. M.; Van Dijk, M.; Vander Plas, H. C.
Heterocycles 1981, 15, 213; (c) Ihmels, H. Sci. Synth. 2005,
5, 907.
give desired product 7 in 39% yield. H NMR (CD
3
OD):
8.58 (d, J = 9.6 Hz, 1H); 8.20 (m, 1H); 8.01 (m, 1H); 7.71–
7.52 (m, 6H); 7.49 (m, 2H).
1