Synthesis of two novel regioisomeric oxospiropyrazole piperidine scaffolds

Article abstract of DOI:10.1016/j.tetlet.2011.02.060

The synthesis of two unprecedented regioisomeric oxospiropyrazole piperidine scaffolds is detailed. Their potential utility as templates for drug discovery is also exemplified. Chemical stability and solid state conformation were also evaluated.

Full text of DOI:10.1016/j.tetlet.2011.02.060

Tetrahedron Letters 52 (2011) 1949–1951
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of two novel regioisomeric oxospiropyrazole piperidine scaffolds
⇑
Benjamin D. Stevens , Etzer Darout, Vincent Mascitti, Kim F. McClure
CVMED Chemistry, Pfizer Worldwide Research & Development, Groton Laboratories, Eastern Point Rd, Groton, CT 06340, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of two unprecedented regioisomeric oxospiropyrazole piperidine scaffolds is detailed. Their
potential utility as templates for drug discovery is also exemplified. Chemical stability and solid state
conformation were also evaluated.
Received 22 December 2010
Revised 8 February 2011
Accepted 14 February 2011
Available online 18 February 2011
Ó 2011 Elsevier Ltd. All rights reserved.
The preparation of structurally novel templates for use in active
pharmaceutical ingredients is one of the numerous challenges faced
by drug discovery chemists. In addition to providing opportunities
for modifying physicochemical properties (e.g., lipophilicity, solu-
bility, permeability), rationally designed cores can incorporate con-
formational restrictions that reduce systemic entropy and thereby
may lead to improved potency or selectivity.1 Such restriction can
also provide useful information that help to better define a specific
pharmacophore. Herein, we report the synthesis of two constrained
regioisomeric oxospiropyrazole piperidine scaffolds and their
application toward the discovery of small molecule agonists of
the G-protein coupled receptor 119 (GPR119).2,3 GPR119 is ex-
pressed at the surface of L-cells (gut) and b-cells (pancreas) and
plays a key role in glucose homeostasis. As such, GPR119 agonism
has recently emerged as a very promising mechanism to the treat-
ment of type 2 diabetes.
various conditions tested gave only low yield of pyrazole 5.6 Hence,
an alternative condensation was explored. Butynoate 6 was treated
with 3 under mildly basic conditions to afford the conjugate adduct
7 in 90% yield.7 Treatment of intermediate 7 under acidic condi-
tions and microwave irradiation provided the corresponding inter-
mediate pyrazole ester, which was used crude in the subsequent
reduction step (LiBH4, THF, 80 °C) to provide intermediate 8 in
65% yield after acidic methanolysis of the borate complex. One-
pot O-tosylation and intramolecular SN2 displacement promoted
effective cycloetherification to provide spirocyclic scaffold 9 in
82% yield. Approaches relying on Mitsunobu conditions to promote
Analysis of targets containing constrained oxospiropyrazole
ring systems (e.g., 1 and 2) suggested that such analogs would be
limited to two major conformations at the piperidine; however,
it was unclear whether the alkyl or pyrazole substituent would oc-
cupy the axial or equatorial positions in the ground state (Fig. 1).
Thus, we embarked on the synthesis of 1 and 2 to probe the utility
of these analogs as potential agonists and to confirm their relative
conformation. Surprisingly, substructure searches of the available
chemical literature did not reveal procedures detailing the synthe-
sis of such heterocyclic scaffolds, hence it was necessary to design
a de novo route to such structures.4
Figure 1. Targets 1 and 2 containing regioisomeric oxospiropyrazole piperidine
scaffolds.
We reasoned that both 1 and 2 could be derived from a diver-
gent route stemming from the known N-benzyl piperidyl hydra-
zine 3.5 Condensation of this building block with appropriate
regioisomeric dicarbonyls or related reagents followed by intramo-
lecular O-alkylation would readily provide access to the desired
scaffolds, primed for further analog synthesis (Fig. 2).
The synthesis of 1 is described in Scheme 1. Condensation of
hydrazine 3 with 4-benzyloxy ethyl acetoacetate 4 under the
⇑
Corresponding author. Tel.: +1 860 686 4360; fax: +1 860 686 8305.
E-mail address: benjamin.stevens@pfizer.com (B.D. Stevens).
Figure 2. Hydrazine 3 as a common precursor to regioisomeric oxospiropyrazole
scaffolds.
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.02.060

1950
B. D. Stevens et al. / Tetrahedron Letters 52 (2011) 1949–1951
Scheme 2. Synthesis of oxospiropyrazole piperidine regioisomer 2.
Scheme 1. Synthesis of oxospiropyrazole piperidine regioisomer 1.
Figure 3. Possible origin of reduction selectivity.
cycloetherification proved ineffective presumably due to the
relatively high pKa of the hydroxypyrazole. Tandem deprotection
of the N-benzyl group and isopropyl carbamate formation followed
by catalytic hydrogenolysis of the benzyl ether proceeded unevent-
fully to afford the free primary alcohol 10 in 60% overall yield. The
final target 1 was subsequently prepared in 48% yield via SNAr of
1,2-difluoro-4-(methylsulfonyl)-benzene with the sodium alkoxide
of 10 (unoptimized).8
The synthesis of 2 is described in Scheme 2. Condensation of 3
with malonyl alkylidene 11 proved sluggish under a variety of con-
ditions; of those screened only microwave irradiation in basic
methanol facilitated good conversion to intermediate 12 in which
the pyrazole and piperidine esters had been, respectively, hydrol-
ized and transesterified (Scheme 2).9 Attempts to reduce the alka-
linity of the reaction medium to afford direct access to the ester
were unsuccessful, highlighting the crucial nature of a 4:1 base-
to-reactant stoichiometry. As a result of these considerations, the
requisite ester was reformed through a standard Fischer esterifica-
tion (MeOH, HCl, dioxane). Reduction of this diester intermediate
(excess LiBH4, THF, reflux) proceeded with exclusive regiocontrol
to produce 13 (45% yield from 3). The preference for the complete
regioselective reduction of the most hindered ester is believed to
arise from the intervention of a boron-‘ate’ species involving the
oxygen in C-5 of the pyrazole that renders the pyrazole ester in
C-4 highly electron rich, and thus less reactive toward nucleophilic
attack (Fig. 3).
In contrast to the regioisomer, exposure of 13 to standard Mits-
unobu conditions resulted in rapid cycloetherification; the pres-
ence of the electron withdrawing ester likely modulates the
hydroxypyrazole pKa rendering it an effective nucleophile in this
case. A one-pot N-debenzylation and isopropylcarbamate forma-
tion gave the desired product 14 in 38% overall yield from 13.
DIBAL-H was discovered to be an optimal reducing reagent for
the preparation of spirocyclic scaffold 15.10 Scaffold 15 was easily
converted into compound 2, the regioisomer of compound 1, via
SNAr of 1,2-difluoro-4-(methylsulfonyl)-benzene.11
Although structurally similar and nearly identical in Rf by
thin layer chromatography (silica gel), 1 and 2 exhibited distinct
characteristics. While oxospiropyrazole 1 was quite stable under
Figure 4. Potential explanation for differential stability of 1 and 2.

B. D. Stevens et al. / Tetrahedron Letters 52 (2011) 1949–1951
1951
Figure 5. X-ray structure of 2 with observed NOEs indicated for axial and equatorial hydrogens.
typical workup and storage conditions, 2 proved to be quite sensi-
References and notes
tive to prolonged weak acid exposure (e.g., stirring in aq ammo-
nium chloride, standing on silica oxide); decomposition of 2 led
to recovery of 15 and the corresponding phenol.12 It is, therefore,
possible that resonance donation of the pyrazole nitrogen or alk-
oxy group assists in elimination of the phenoxide leaving group
in the case of isomer 2, while in isomer 1, cross conjugation of
the putative benzylic cation with either of these donating groups
would reduce the likelihood of this decomposition pathway
(Fig. 4).13
Single crystals of isomer 2 prepared by slow evaporation from
ethyl acetate were obtained for X-ray diffraction analysis (Fig. 5).
As pictured above, the solid state conformation of the piperidine
places the pyrazole substituent in the axial position, while the
methylene group is projected in the equatorial position; this is
likely due to the restricted rotation around the pyrazole–piperidine
C–N linkage that would be expected to dramatically reduce the
effective steric bulk for this group thereby minimizing diaxial
interactions in this conformer. Analysis of the NOESY spectra for
2 confirms that solution state conformation is identical to that ob-
served in the solid state.
When evaluated in vitro at human GPR119, both compounds 1
and 2 proved inferior to other agonists profiled in our program.
This result indicates that the conformation occupied by these con-
strained analogs is suboptimal when compared to more flexible
variants, suggesting that they are not representative of the bioac-
tive conformation.
In conclusion, we have described the synthesis and stability of
two novel constrained regioisomeric oxospiropyrazole piperidine
scaffolds. Although these compounds were not immediately suit-
able for continued pursuit within our lead discovery program,
the work described around the preparation of these constrained
spirocyclic scaffolds provided further information and understand-
ing of the GPR119 pharmacophore. We also expect that these gen-
eral synthetic strategies will find further application for other
small molecule drug discovery programs.
1. For a review of this topic, see: Mann, J.; Le Chatelier, H. Conformational
Restriction and/or Steric Hindrance in Medicinal Chemistry. In Practice of
Medicinal Chemistry; Wermuth, C. G., Ed., 2nd ed.; Elsevier: London, 2003; pp
233–250.
2. For general information regarding the GPR119 receptor as a therapeutic target,
see: (a) Semple, G.; Fioravanti, B.; Pereira, G.; Calderon, I.; Uy, J.; Choi, K.; Xiong,
Y.; Ren, A.; Morgan, M.; Dave, V.; Thomsen, W.; Unett, D. J.; Xing, C.; Bossie, S.;
Carroll, C.; Chu, Z.-L.; Grottick, A. J.; Hauser, E. K.; Leonard, J.; Jones, R. M. J. Med.
Chem. 2008, 51, 5172; (b) Shah, U. Curr. Opin. Drug Discovery Dev. 2009, 12, 519;
(c) Lauffer, L.; Iakoubov, R.; Brubaker, P. L. Endocrinology 2008, 149, 2035;.
3. For examples of known GPR119 agonists see: (a) Jones, R. M.; Leonard, J. N.
Annu. Rep. Med. Chem. 2009, 44, 149; (b) Jones, R. M.; Leonard, J. N.; Buzard, D.
J.; Lehmann, J. Expert Opin. Ther. Patents 2009, 19, 1339.
4. We were unable to find any closely related ring structures through
substructure searches in the chemical literature.
5. Jenkins, S. M.; Wadsworth, H. J.; Bromidge, S.; Orlek, B. S.; Wyman, P. A.; Riley,
G. J.; Hawkins, J. J. Med. Chem. 1992, 35, 2392.
6. For example: Et3N (5 equiv), b-ketoester (1.2 equiv), EtOH (0.1 M), 70 °C, 16 h,
conventional heating; Et3N (5 equiv), b-ketoester (1.2 equiv), EtOH (0.1 M),
130 °C, 30 min, microwave heating; NaOAc (5 equiv), b-ketoester (1.2 equiv),
EtOH (0.1 M), 70 °C, 16 h, conventional heating; NaOAc (5 equiv), b-ketoester
(1.2 equiv), EtOH (0.1 M), 130 °C, 30 min, microwave heating; Et3N (5 equiv),
b-ketoester (1.2 equiv), EtOH (0.5 M), 70 °C, 16 h, conventional heating; NaOAc
(5 equiv), b-ketoester (1.2 equiv), EtOH (1 M), 70 °C, 16 h, conventional heating.
7. For the preparation of 6 see: Marmsaeter, F. P.; Vanecko, J. A.; West, F. G. Org.
Lett. 2004, 6, 1657.
8. Spectral data for 1: 1H NMR (chloroform-d, 500 MHz): d (ppm) 7.65–7.69 (m,
2H) 7.32 (t, J = 7.80 Hz, 1H), 5.51 (s, 1H), 5.13 (s, 2H), 4.96 (septet, J = 6.10 Hz, 1
H), 4.75 (s, 2H), 3.93–3.98 (m, 2H), 3.50–3.60 (m, 2H), 3.05 (s, 3H), 2.05–2.15
(m, 2H), 1.75–1.85 (m, 2H), 1.28 (d, J = 6.35 Hz, 6H); MS (M+1): 468.0.
9. Although treatment with ethyl ethoxymethylene-malonate for extended
reaction times (ꢀ24–48 h) in refluxing EtOH with potassium carbonate gave
reasonable conversions, the reaction could not be driven to completion without
the increased formation of the carboxylic acid. The presence of base was
absolutely necessary for the reaction to proceed. Overall, the use of the
described procedure afforded rapid conversion to the acid, which immediately
recrystallized out of the reaction medium with cooling and could be recovered
conveniently by filtration. It appears as if hydrolysis only occurs at the
heteroaryl ester while transesterification only occurs at the alkyl ester.
10. Careful control of reaction stoichiometry was necessary to prevent reduction of
the carbamate to the N-methyl piperidine. This consideration led to suboptimal
yields of the desired alcohol.
11. Spectral data for 2: 1H NMR (chloroform-d, 500 MHz):
d (ppm) 7.70 (d,
J = 8.8 Hz, 1H), 7.65 (dd, J = 10.0, 2.2 Hz, 1H), 7.46 (s, 1H), 7.20 (t, J = 8.2 Hz, 1H),
4.99 (s, 2H), 4.92–4.98 (m, 1H), 4.81 (s, 2H), 3.91 (ddd, J = 13.2, 8.8, 3.7 Hz, 2H),
3.61 (br s, 2H), 3.05 (s, 3H), 2.06–2.14 (m, 2H), 1.78–1.88 (m, 2H), 1.27 (s, 3H),
1.26 (s, 3H); MS (M+1): 468.0.
Acknowledgments
12. An NMR sample of 2 was left for >5 d at ambient temperature in DMSO-d6 and
very little decomposition was observed. It therefore appears very likely that
weak acid is a critical component of the decomposition.
13. The authors thank the editors for noting that resonance donation of the
pyrazole nitrogen lone pair would also be expected to behave analogously to
the ether participation invoked in Figure 4 and that in all likelihood it is a
combination of both donating groups that results in this instability.
We would like to thank Brian Samas for the generation of the
X-ray structure for 2 and Geeta Yalamanchi for the NOESY spectra
interpretation. We also thank Kevin Filipski and Dr. Michael Green
for correction of the manuscript.