Improved synthesis of C4α- and C4β-methyl analogues of 2-aminobicyclo[3.1.0]hexane-2,6-dicarboxylate

Article abstract of DOI:10.1021/ol300516y

An efficient and divergent synthesis of C4α- and C4β-methyl- substituted analogues of 2-aminobicyclo[3.1.0]hexane 2,6-dicarboxylate, which are important tools in the study of metabotropic glutamate receptor function, has been achieved. By taking advantage

Full text of DOI:10.1021/ol300516y

ORGANIC
LETTERS
2012
Vol. 14, No. 11
2662–2665
Improved Synthesis of C4r-andC4β-Methyl
Analogues of 2-Aminobicyclo[3.1.0]hexane-
2,6-dicarboxylate
Steven S. Henry,*,† Molly D. Brady,‡,§ Dana L. T. Laird,† J. Craig Ruble,† David L. Varie,‡
and James A. Monn†
Discovery Chemistry Research and Technologies and Chemical Product Research and
Development, Eli Lilly and Company, Indianapolis, Indiana 46285, United States
sshenry@lilly.com
Received March 6, 2012
ABSTRACT
An efficient and divergent synthesis of C4R- and C4β-methyl-substituted analogues of 2-aminobicyclo[3.1.0]hexane 2,6-dicarboxylate, which are
important tools in the study of metabotropic glutamate receptor function, has been achieved. By taking advantage of an unanticipated facial
selectivity of the bicyclo[3.1.0]hexane ring system, either the C4R- or C4β-methyl substituent was introduced in a highly stereoselective and high-
yielding manner.
The metabotropic glutamate (mGlu) receptors are pro-
mising targets for the treatment of CNS disorders.1 Our
laboratory has focused on the design and synthesis of
orthosteric (glutamate-site) agonists targeting metabotropic
glutamate receptors 2 and 3 (mGlu2/3 receptors). These
investigations have resulted in the identification of several
conformationally constrained glutamate analogues (1ꢀ3,
Figure 1), which exhibit high potency and excellent selectiv-
ity for mGlu2/3 receptors over other mGlu and ionotropic
glutamate receptors but which do not pharmacologically
differentiate between mGlu2 and mGlu3 receptor subtypes.2
Figure 1. mGlu2/3 receptor agonists.
To investigate the effects of ring substitution of 1 on
and (þ)-5.3 Evaluation of the functional effects of these
mGlu receptor function and selectivity, we prepared (()-4
molecules in cells expressing human mGlu2 or mGlu3
† Discovery Chemistry Research and Technologies.
receptors revealed that C4β-methyl analogue (()-4, like
compounds 1ꢀ3, acted as a full agonist at both mGlu2 and
mGlu3 receptors (EC50 = 45 and 34 nM, respectively),
whereas the C4R-methyl derivative (þ)-5 (LY541850)
exhibited an unexpected mixed mGlu2 agonist/mGlu3 an-
tagonist pharmacological profile (mGlu2 EC50 = 161 nM,
mGlu3 IC50 = 1050 nM). Despite their interesting pharma-
cologic properties, several issues associated with the originally
‡ Chemical Product Research and Development.
§ ImClone Systems, Bridgewater, NJ 08807.
(1) O’Neill, M. J.; Fell, M. J.; Svensson, K. A.; Witkin, J. M.;
Mitchell, S. N. Drugs Future 2010, 35, 307–324.
(2) (a) Monn, J. A.; Valli, M. J.; Massey, S. M.; Wright, R. A.;
Salhoff, C. R.; Johnson, B. G.; Howe, T.; Alt, C. A.; Rhodes, G. A.;
Robey, R. L.; Griffey, K. R.; Tizzano, J. P.; Kallman, M. J.; Helton,
D. R.; Schoepp, D. D. J. Med. Chem. 1997, 40, 528–537. (b) Monn, J. A.;
Valli, M. J.; Massey, S. M.; Hansen, M. M.; Kress, T. J.; Wepsiec, J. P.;
Harkness, A. R.; Grutsch, J. L., Jr.; Wright, R. A.; Johnson, B. G.; Andis,
S. L.; Kingston, A. E.; Tomlinson, R.; Lewis, R.; Griffey, K. R.; Tizzano,
J. P.; Schoepp, D. D. J. Med. Chem. 1999, 42, 1027–1040. (c) Monn, J. A.;
Valli, M. J.; Massey, S. M.; Henry, S. S.; Stephenson, G. A.; Bures, M.;
Herin, M.; Catlow, J.; Giera, D.; Wright, R. A.; Johnson, B. G.; Andis, S. L.;
Kingston, A. E.; Schoepp, D. D. J. Med. Chem. 2007, 50, 233–240.
(3) Dominguez, C.; Prieto, L.; Valli, M. J.; Massey, S. M.; Bures, M.;
Wright, R. A.; Johnson, B. G.; Andis, S. L.; Kingston, A.; Schoepp,
D. D.; Monn, J. A. J. Med. Chem. 2005, 48, 3605–3612.
r
10.1021/ol300516y
Published on Web 05/21/2012
2012 American Chemical Society

reported synthesis of (()-4 and (þ)-5 severely limited our
access to these molecules, precluding our ability to assess their
effects in vivo. First, the original synthesis as outlined in
Scheme 1 utilized conventional Strecker or BuchererꢀBergs
methodology to install amino acid functionality at the C2
center. These were low-yielding transformations and required
a selective crystallization approach to obtain the desired C2
diastereomer. Second, the synthesis of (þ)-5 was particularly
lengthy owing to a cumbersome inversion of C4β-methyl
intermediate in order to obtain the desired C4R stereochem-
istry. Third, the approach was based on a racemic precursor
necessitating chiral chromatography in order to separate the
mGlu receptor active and inactive enantiomers. These syn-
thetic issues resulted in extremely low overall yields of (()-4
(6%) and (þ)-5 (1%).
(þ)-8 was prepared in high yield and excellent diastereo-
meric purity from previously reported enone (þ)-74 via
carboxy cyclopropanation in the presence of trifluoroetha-
nol. Addition of trifluoroethanol or other proton sources
with a similar pKa is essential to achieve high selectivity for
the desired diastereomer. Under the conditions described
(10 equiv of trifluoroethanol), the desired cyclopropana-
tion product is formed as a 95:5 mixture of the desired (þ)-
8 and the undesired diastereomer with inverted configura-
tion at C-6 (“endo-ester”). In the absence of an appropriate
proton source the product ratio is reversed; the endo-ester
diastereomer is the dominant (>85%) stereoisomer in the
reaction mixture.5 Subsequent reaction of (þ)-8with methyl-
triphenylphosphonium ylide provided (ꢀ)-9 in 99% yield.
Scheme 2. Synthesis of C4β-Methyl Analogue (ꢀ)-4
Scheme 1. Published Synthetic Approach to (()-4 and (þ)-5
When (ꢀ)-9 was subjected to hydrogenation using Pd/C as
the catalyst, we observed a mixture consisting of C4β-methyl
(ꢀ)-10 and C4R-methyl (þ)-11 derivatives in a ratio of 97:3
as judged by the integration of the individual C4 methyl
1
doublets in the H NMR of the crude reaction mixture.
Stereochemical assignment of the individual isomers could
be determined based on the splitting pattern of the C4
hydrogens in (ꢀ)-10 and (þ)-11. Specifically, the C4 hydro-
genorientedontheR-face of the bicyclic system, as is the case
with (ꢀ)-10, is nearly orthogonal to both C5-H and C3β-H
resulting in coupling constants that approach 0 Hz. Hence,
C4R-H shows coupling to only the C3R-H and the hydro-
gens on the attached C4-methyl substituent leading to an
observed pentet, Figure 2. In contrast, the C4 hydrogen
oriented on the β-face of the bicyclic system, as with (þ)-11,
is coupled to C5-H (dihedral angle ≈ 40°, J ≈ 7 Hz), C3β-H
(dihedral angle ≈ 15°, J ≈ 5 Hz), C3R-H, and the methyl
To improve upon this route, we envisioned beginning
with a nonracemic precursor already possessing the C2
amino acid functionality. If successful, this would elim-
inate the loss of material during separation of both dia-
stereomeric and enantiomeric mixtures. It was originally
considered that (þ)-5 might be prepared by reduction of
exocyclic olefin intermediate (ꢀ)-9, with the hydrogena-
tion catalyst envisioned to approach the double bond from
the convex face of the bicyclo[3.1.0]hexane ring system as
was observed during reduction of the endocyclic olefin 6 in
the original route. As shown in Scheme 2, bicyclic ketone
(5) The role of the proton source in controlling cyclopropanation
stereoselectivity has not been definitively established. The diastereo-
meric esters do not interconvert under the reaction conditions. The pKa
and quantity of the proton source, the nature of the alkyl substituent on
the ylide ester, and to a lesser extent, solvent all impact the stereo-
selectivity of the reaction. A full account of the investigation of factors
controlling the stereochemical outcome of this reaction will be published
in a separate report.
(4) Enone (þ)-7 was prepared on a kilogram scale in eight steps with
an overall yield of 23% as described in the following publication: Varie,
D. L.; Beck, C. B.; Borders, S. K.; Brady, M. D.; Cronin, J. S.;
Ditsworth, T. K.; Hay, D. A.; Hoard, D. W.; Hoying, R. C.; Linder,
R. J.; Miller, R. D.; Moher, E. D.; Remacle, J. R.; Rieck, J. A., III. Org.
Process Res. Dev. 2007, 11, 546–559.
Org. Lett., Vol. 14, No. 11, 2012
2663

hydrogens leading to a much more complex, unresolved
multiplet.
Given the facial selectivity observed during hydrogena-
tion of (ꢀ)-9, we surmised that the tert-butyl ester func-
tionality at the C2-position of this molecule, not present in
enone 6, might be exerting a strong 1,3-steric influenceover
the approach of reagents to the C4-position. To further
explore this hypothesis, a series of analogues was prepared
in which the size of the carboxylic acid and amine protect-
ing groups were varied, Scheme 3.
Scheme 3. Preparation of Alternatively Protected 9
Figure 2. 1H NMR splitting patterns of the C4 proton.
The facial selectivity observed in the Pd/C-catalyzed
reduction of (ꢀ)-9 clearly indicated that hydrogenation
under these conditions occurred preferentially from the
concave face of the bicyclohexane ring system. In order to
further explore this finding, a variety of alternative hydro-
genation catalysts were surveyed. All reactions were
screened in both methanol and ethyl acetate (0.06 M of
9) at 40 °C with an initial pressure of 100 psi of H2. No
effect of solvent was observed on the resultant diastereo-
meric ratios, and therefore, only results from reactions
conducted in methanol are reported (Table 1). As can be
seen, each of the catalysts investigated, including those
employing chiral ligands, (entries 8 and 9), provided
diastereomeric mixtures of (ꢀ)-10 and (þ)-11 strongly
favoring the former, though none provided superior selec-
tivity over what was originally achieved with Pd/C.6 In this
case, the isomeric mixture was separated on silica resulting
in diastereomerically pure (ꢀ)-10 in 74% yield. Deprotec-
tion with methanesulfonic acid followed by cation ex-
change chromatography afforded (ꢀ)-4 in 69% overall
yield from (þ)-8.
Olefins9aꢀe were exposed toWilkinson’s catalyst under
1 atm of hydrogen. The results of these studies are detailed
in Table 2. Decreasing the size of either CO2R (R = t-Bu,
Et, H; entries 1ꢀ3) or NHR (R = t-BuO2C, MeO2C, H)
substituents had little impact on the ratio of methyl-
substituted products 10 and 11. This suggests that in each
case the C2-CO2R substituent likely resides in a pseudoax-
ial orientation on the β-face of the bicyclic ring system.
This particular conformation may be preferred over the
alternative (pseudoequatorial C2-CO2R, pseudoaxial
NHR) owing to a likely negative steric interaction between
the C6 hydrogen and C2-NHR substituent. Further sup-
porting this hypothesis, it is interesting to note that the
crystal structures of both 1 and a sulfoxide variant of 3
have been solved and each appears to possess conforma-
tions inwhichtheamine andcarboxylicacidfunctionalities
reside in a pseudoequatorial and pseudoaxial orientation
respectively.2a,c
Based on these data, we hypothesized that approach of a
hydride donor or other nucleophiles to the C4 carbonyl
carbon of (þ)-8 might also occur preferentially from the
side opposite to the C2-ester (i.e., from the concave face)
and that we might be able take advantage of this bias for
the preparation of (þ)-5. In this respect, we were gratified
to find that reaction of ketone (þ)-8 in THF at ꢀ5 °C with
L-Selectride provided C4β-carbinol (ꢀ)-12 with excellent
Table 1. Catalyst Effect on Ratio of 10 to 11
entry
catalyst
(ꢀ)-10/(þ)-11
1
2
3
4
5
6
7
8
9
Pd/C 5%
Pd(OH)2
PtO2
97/3
88/12
95/5
Raney Ni
94/6
(PCy3)Ir(cod)pyꢀPF6
83/17
90/10
79/21
74/26
75/25
RhCl(PPh3)3
(dippf)Rh(cod)ꢀBF4
(R,R)-(Et-DuPhos)Rh(cod)ꢀBF4
(S,S)-(Et-DuPhos)Rh(cod)ꢀBF4
(6) In some instances during the reduction with Pd/C, varying amounts
of a byproduct believed to result from reductive opening of the cyclopro-
pane ring have been observed. In contrast, Wilkinson’s catalyst showed no
evidence of byproduct under all conditions explored. Factors that lead to
the ring-opened product are not yet fully understood.
2664
Org. Lett., Vol. 14, No. 11, 2012

Table 2. Protecting Group Effect on Ratio of 10 to 11
Scheme 4. Synthesis of C4R-Methyl Analogue (þ)-5
entry
9
R1
R2
10/11
1
2
3
4
5
6
9
CO2-t-Bu
CO2-t-Bu
CO2-t-Bu
CO2CH3
CO2CH3
H
t-Bu
Et
H
86/14
82/18
77/23
84/16
85/15
80/20
9a
9b
9d
9e
9c
Et
H
Et
diastereomeric excess (>98%) and in high isolated yield
(92%, Scheme 4). Alcohol (ꢀ)-12 was smoothly con-
verted to tosylate (ꢀ)-13 (89%) under standard con-
ditions, and the tosylate subsequently underwent SN2-
mediated displacement with freshly prepared lithium
dimethyl cuprate7 to provide C4R-methyl intermediate
(þ)-11 in 71% isolated yield. Simultaneous deprotection
of the carboxy amino protecting groups with aqueous
acetic acid8 afforded zwitterion (þ)-5 in 45% overall yield
from (þ)-8.
for excellent stereochemical control of the approach of
reagents to either C4-ketone or C4-exocyclic methylene
functionalities. In addition, this methodology obviates the
need for chiral chromatography and avoids the low-yield-
ing introduction of the C2-amino acid functionality as
described in the original route.5 Hence, the synthetic routes
described in this account resulted in a significant improve-
ment in overall yield and throughput of (ꢀ)-4 and (þ)-5.
Notably, access to the latter has enabled both in vitro and
in vivo studies designed to elucidate the individual roles of
mGlu2 and mGlu3 receptors within the central nervous
system.9 Finally, the approach described herein also high-
lights the value of intermediate (þ)-8, which, owing to the
presence of the C4-ketone functionality, should enable
additional structureꢀactivity relationship studies to be
pursued. Results of these investigations will be reported
in due course.
New synthetic approaches to pharmacologically inter-
esting bicyclic amino acids from nonracemic precursor
(þ)-7 have been established. The current methodology
takes advantage of an unanticipated facial bias that allows
(7) RajanBabu, T. V.; Nugent, W. A.; Taber, D. F.; Fagan, P. J.
J. Am. Chem. Soc. 1988, 110, 7128–7135.
(8) Hao, J.; Reinhard, M.; Henry, S. S.; Seest, E. P.; Belvo, M. D.;
Monn, J. A. Tetrahedron Lett. 2012, 53, 1433–1434.
(9) (a) Sheffler, D. J.; Walker, A. G.; Conn, P. J.: Investigation of the
Group II Metabotropic Glutamate Receptor Subtype Responsible for
Increased Cyclic AMP Production and Reduced Excitatory activity
caused by co-Activation of β-Adrenergic and mGlu2/3 Receptors. In
Current Neuropharmacology, Abstracts of the 7th International Meeting on
Metabotropic Glutamate Receptors, Taormina, Italy, October 2011; Salt,
T. E., Conn, P. E., Iwata, N., Eds.; Bentham Science Publishers: Englewood,
CO, 2011; p 164. (b) Sanger, H.; Hanna, L.; Colvin, E. M.; Grubisha, O.;
Findlay, J. D.; Ursu, D.; Heinz, B. A.; Vivier, R. G.; Lodge, D.; Monn,
J. A.; Broad, L. M. Neuropharmacology 2012, submitted for publication,
2012. (c) Lucas, S.; Bortolotto, Z.; Collingridge, G.; Lodge, D. Neuropharmacol-
ogy [Online early access]. http://dx.doi.org/10.1016/j.neuropharm.2012.04.006.
Published Online April 17, 2012. (d) Hanna, L.; Ceolin, L.; Lucas, S.; Monn, J.;
Johnson, B.; Collingridge, G.; Bortolotto, Z.; Lodge, D. Neuropharmacology
[Online early access]. http://dx.doi.org/10.1016/j.neuropharm.2012.02.023. Pub-
lished Online March 15, 2012.
Acknowledgment. Special thanks to Junliang Hao (Eli
Lilly) for his assistance during the preparation of this
manuscript.
Supporting Information Available. Experimental pro-
cedures and spectroscopic data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 11, 2012
2665