Novel cyclopropyl β-amino acid analogues of pregabalin and gabapentin that target the α2-δ protein

Article abstract of DOI:10.1021/jm0491086

As part of a program aimed at generating compounds with affinity for the α₂-δ subunit of voltage-gated calcium channels, several novel β-amino acids were prepared using an efficient nitroalkane-mediated cyclopropanation as a key step. Depending

Full text of DOI:10.1021/jm0491086

3026
J. Med. Chem. 2005, 48, 3026-3035
Novel Cyclopropyl â-Amino Acid Analogues of Pregabalin and Gabapentin That
Target the r2-δ Protein
Jacob B. Schwarz,* Sian E. Gibbons, Shelley R. Graham, Norman L. Colbry, Peter R. Guzzo,† Van-Duc Le,†
Mark G. Vartanian, Jack J. Kinsora, Susan M. Lotarski, Zheng Li, Melvin R. Dickerson, Ti-Zhi Su,
Mark L. Weber, Ayman El-Kattan, Andrew J. Thorpe, Sean D. Donevan, Charles P. Taylor, and
David J. Wustrow
Pfizer Global Research and Development, Michigan Laboratories, 2800 Plymouth Road, Ann Arbor, Michigan 48105, and
Albany Molecular Research, Inc., PO Box 15098, Albany, New York 12212
Received November 5, 2004
As part of a program aimed at generating compounds with affinity for the R2-δ subunit of
voltage-gated calcium channels, several novel â-amino acids were prepared using an efficient
nitroalkane-mediated cyclopropanation as a key step. Depending on the ester that was chosen,
the target amino acids could be prepared in as few as three steps. The cyclopropyl amino acids
derived from ketones proved to be potent binders of the R2-δ subunit of voltage-gated calcium
channels, but did not interact with the large neutral amino acid system L (leucine) transporter.
Anticonvulsant effects were observed in vivo with compound 34 but only after intracerebroven-
tricular (icv) administration, presumably due to inadequate brain concentrations of the drug
being achieved following oral dosing. However, pregabalin 1 was active in the DBA/2 model
after oral (and icv) dosing, supporting a hypothesis that active transport is a prerequisite for
such zwitterionic species to cross the blood-brain barrier.
Introduction
Pregabalin 1 has demonstrated potent and robust
activity in preclinical models of epilepsy, neuropathic
pain, and anxiety and is currently approved or in clinical
development for these conditions.1 Both pregabalin 1
Figure 1. Structures of pregabalin 1 and gabapentin 2.
and gabapentin 22 have been shown to bind potently to
sium tert-butoxide,18 and KF/alumina.19 However, to our
knowledge no reports concerning the conversion of these
cyclic cyanoesters to the corresponding â-amino acids
have appeared. We felt that novel compounds of this
the R2-δ subunit of voltage-gated calcium channels.3,4
As a consequence of binding to R2-δ, calcium influx is
modulated at nerve terminals5 which in turn reduces
the release of several neurotransmitters including
type, having amine and carboxylate functionalities in
glutamate,6 norepinephrine,7 and substance P.8 It is
similar proximity to those of gabapentin and pregabalin,
through the inhibition of calcium flux and subsequent
albeit with a more rigid framework, might also be
attenuation of neurotransmitter release that 1 and 2
ligands for the R2-δ subunit. The preparation and
are postulated to exert their therapeutic effects.9 Con-
biological testing of such constrained analogues thus
sistent with this hypothesis, it has been demonstrated
encompass the subject of this report.20,21
that the affinity of analogues of 1 and 2 for R2-δ (as
evidenced by the displacement of [3H]-gabapentin from
Chemistry
pig brain membranes) was proportional to their activity
It was envisioned that use of a benzyl ester in the
Knoevenagel reaction would, after cyclopropanation
with nitromethane, result in a substrate that could be
transformed in one operation to the desired amino acids.
Gratifyingly, this proved to be the case (Scheme 1).
Hence, condensation of 2-ethylbutyraldehyde 6 or 4-hep-
tanone 7 with benzyl cyanoacetate and subsequent
cyclization with nitromethane afforded cyclopropane
cyanoesters 10 and 11 in good overall yield. In the cases
where an aldehyde was employed, only a single diaste-
reomer of the cyclopropane was obtained, presumably
placing the nitrile and the alkyl group in the cis-
orientation in accordance with literature precedent.18
Although the Michael addition occurred at ambient
temperature, the aldehyde-derived substrates required
heating to effect the ring closing process. Exhaustive
reduction of the benzyl ester and nitrile was effected
by hydrogenation catalyzed by PtO2. Unfortunately,
in an in vivo seizure model.10-12 As part of a program
to generate SAR around additional R2-δ ligands,13-16 we
were interested in the possibility of installing alkyl
substituents at the position R- to the amino function of
gabapentin and examining the effect on R2-δ binding
affinity. To accomplish this task, unsaturated cyano-
ester 3 derived from cyclohexanone was treated with
nitroethane and DBU at ambient temperature. Instead
of isolating the expected Michael adduct 4, however, we
were surprised to find that a facile cyclization had
occurred, furnishing spirocyclopropane 5. Indeed, on
inspection of the literature we found several methods
available for effecting this nitroalkane-mediated cyclo-
propanation, including potassium carbonate,17 potas-
* To whom correspondence should be addressed: Phone (734)-622-
2580; Fax (734)-622-5165; Email jacob.schwarz@pfizer.com.
† Albany Molecular Research, Inc.
10.1021/jm0491086 CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/29/2005

Cyclopropyl â-Amino Acid Analogues of Pregabalin
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8 3027
Figure 2. Facile cyclopropanation of 3 with nitroethane.
Scheme 1
Finally, we wished to prepare the aldehyde-derived
cyclopropane amino acids bearing the opposite relative
configuration (i.e. alkyl group cis to carboxylate). To
achieve this objective while taking advantage of the
readily accessible intermediate cyclopropyl cyanoesters
described above, the ester would need to be converted
to an aminomethyl moiety and the nitrile to a carboxylic
acid. Such utility of the cyanoester functionality has
recently been demonstrated by D´ıaz-de-Villegas and
Ga´lvez in an enantioconvergent synthesis of (S)-R-
benzyl-R-methyl-â-alanine.25 Hence, preparation of the
i
cyclopropyl cyanoesters 43 (R ) Bu) and 44 (R )
CHEt2) by the usual method was followed by borohy-
dride reduction of the ester to provide cyano alcohols
45 and 46 (Scheme 5). Surprisingly, the alcohols were
not converted to the mesylate on treatment with meth-
anesulfonyl chloride, but rather the chloronitriles 47
and 48.26 This was of no consequence as the purified
chloronitriles could be smoothly displaced by azide anion
to afford azidonitriles 49 and 50 in nearly quantitative
yield. Finally, hydrogenation and hydrolysis of the
resultant aminoesters furnished the desired amino acids
53 and 54 after purification. Compound 53 was diaster-
eomeric with 36 and compound 54 was diastereomeric
with 12.
although the route proved expeditious, the yield of the
reduction step was quite low after purification of the
amino acids 12 and 13 (11-18%). As a result, we set
out to develop a more efficient and scalable process to
obtain the target amino acids.
The second-generation synthesis, typified as follows,
commenced with the condensation of cyclopentanone
with methyl cyanoacetate under azeotropic dehydration
conditions (Scheme 2). Nitromethane addition and
subsequent cyclization occurred smoothly at ambient
temperature, providing spirocyclopropane 16 in 62%
yield. Chemoselective reduction of the nitrile function
was effected using the sodium borohydride/Co(II) re-
agent system.22 The aminoester 17 was subsequently
hydrolyzed under basic conditions (39%) to give amino
acid 18 after acidification.
In an attempt to expand the synthetic utility of the
process, we were interested in generating differentially
protected amino acid core structures that would also be
amenable to further elaboration (i.e. insertion into a
peptide sequence). Thus, the requisite carbonyl sub-
strates (cyclohexanone 19, isobutyraldehyde 20, iso-
valeraldehyde 21) were first converted in the usual way
to the cyclopropane derivatives (Scheme 3). Although
the cyclopropanation of 22 with nitromethane proceeded
at ambient temperature (DBU, CH3CN), we found the
KF-alumina conditions19 to be most appropriate for the
aldehyde-derived unsaturated cyanoesters 23 and 24.
Following reduction of the nitrile, the amine was capped
with a tert-butoxycarbonyl group, leaving open the
possibility for further manipulation at either terminus.
To demonstrate this point, aminoesters 28-30 could be
hydrolyzed to give the N-protected carboxylic acids 31-
33. These valuable synthetic building blocks could also
be easily converted (63-88%) to the corresponding
â-amino acids 34-36 upon exposure to hydrochloric acid
in dioxane.
Pharmacology
Of the compounds disclosed in this report, in general
the ketone-derived substrates showed greater potency
for the R2-δ subunit of voltage-gated calcium channels
(as evidenced by their ability to potently inhibit [3H]-
gabapentin binding to pig brain membranes, see Table
1). For instance, the dipropyl- (13), cyclopentyl- (18), and
cyclohexyl-substituted (34) cyclopropyl â-amino acids all
bound with Ki values (equilibrium dissociation con-
stants) less than or equal to 0.2 µM, demonstrating high
potency. In contrast, none of the aldehyde-derived amino
acids containing the alkyl and carboxylate in the trans-
orientation about the cyclopropane ring showed com-
parably potent (<0.2 µM) affinity for R2-δ. However,
when the relative orientation of alkyl and carboxylate
groups in 36 was reversed as in 53 (cis-orientation), a
10-fold increase in R2-δ binding affinity was observed.
This did not prove to be the case for switching the
relative orientation of the 1-ethyl-propyl group of 12,
as the isomer 54 was found to have similar potency to
12. Noteworthy is the fact that unsubstituted 1-amino-
methylcyclopropane carboxylic acid 5527 did not bind to
R2-δ or system L, and as a point of reference, pregabalin
1 demonstrated 0.019 µM binding affinity to R2-δ (Ki)
and 158 µM to system L (IC50) as a single enantiomer.
As can also be seen from Table 1, unlike pregabalin
1, none of the cyclopropyl amino acids were able to
significantly inhibit [3H]leucine uptake by the system
L amino acid transporter.28 The ability of pregabalin and
gabapentin to be transported by system L is thought to
play a pivotal role in allowing these zwitterionic mol-
As a further test of this methodology, we set out to
prepare novel amino acids containing multiple cyclo-
propane units.23 Hence, cyclopropanecarboxaldehyde 37
underwent smooth condensation with ethyl cyanoace-
tate to provide cyanoester 38 (Scheme 4). Completion
of the sequence (under nonoptimized conditions) af-
forded the bis-cyclopropanated amino acid 40.24

3028 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8
Schwarz et al.
Scheme 2
Scheme 3
Scheme 4
or by injection into the cerebral ventricles (icv), com-
pound 34 only protected DBA/2 mice from audiogenic
seizures when the compound was introduced directly
into the brain through icv dosing. This observation
served to further implicate the system L transporter in
mediating the brain uptake of amino acid-based R2-δ
ligands. This point is further illustrated by examining
the brain and plasma levels achieved by both com-
pounds at 2 h following 30 mg/kg po dosing in Sprague-
Dawley rats (Table 2). Not only was the brain to plasma
(b/p) ratio of 34 significantly smaller than that shown
with pregabalin 1 (0.054 vs 0.22, respectively), but the
absolute plasma concentrations for 34 were also sig-
nificantly lower, which may suggest the active role of
system L (or another transporter) in improving brain
penetration and increasing the extent of absorption of
these zwitterionic amino acids by the gastrointestinal
tract. Further studies to elucidate the mechanism by
which these compounds are absorbed and distributed
to the brain are in progress.
Scheme 5
Conclusion
Using an efficient cyclopropanation reaction of electron-
deficient olefins, several novel â-amino acids were
prepared efficiently and in good overall yields. The most
expeditious route used only three steps but the efficiency
was somewhat compromised. As a result, a more reliable
method was developed, which also allowed access to
differentially substituted termini of the amino acid
products. The ketone-derived cyclopropyl amino acids
were shown to be potent substrates for the R2-δ subunit,
unlike the less hindered aldehyde-derived substrates
which in two cases were prepared with the opposite
relative stereochemistry. However, none of the com-
pounds were substrates for the system L transporter,
thought to be required for efficient brain penetration.
This was supported by in vivo anticonvulsant studies
which showed that activity could only be achieved if the
ecules to enter the systemic circulation following oral
dosing and cross the blood-brain barrier. This notion
was borne out through in vivo mouse anticonvulsant
testing, whereby the antiepileptic activity of compounds
was measured by their ability to prevent audiogenically
induced seizures in the DBA/2 strain of mice.29 Activity
in this assay is expressed as a percentage of mice
protected from seizure (Graph 1). Whereas pregabalin
showed robust anticonvulsant activity when dosed orally

Cyclopropyl â-Amino Acid Analogues of Pregabalin
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8 3029
Table 1. R2-δ and System L Transporter Affinity of Cyclopropyl-â-Amino Acids
R2-δ binding affinity
system L binding affinity
compd
R1
R2
(Ki, µM)a
(IC50, µM)b
1
- -
- -
H
0.019 ( 0.003
0.56 ( 0.11
158 ( 35
>8300
>2490
>8300
>2490
>8300
>2490
>2490
>8300
>2490
>8300
12
13
18
34
35
36
40
53
54
55
CH(CH2CH3)2
CH2CH2CH3
CH2CH2CH3
0.20 ( 0.04
-CH2CH2CH2CH2-
0.023 ( 0.007
0.013 ( 0.001
0.34 ( 0.10
0.33 ( 0.013
0.20 ( 0.055
0.037 ( 0.004
0.63 ( 0.08
-CH2CH2CH2CH2CH2-
CH(CH3)2
H
CH2CH(CH3)2
H
c-C3H5
H
H
H
H
CH2CH(CH3)2
CH(CH2CH3)2
H
>10
a See Experimental Section. Data reported as mean ( SEM (N ) 4 experiments). b IC50 is the concentration (µM) producing half-
maximal inhibition of the uptake of [3H]leucine into CHO cells.
continued for 10 min. Additional acetic acid (0.1 mL) and
piperidine (0.16 mL) were added, followed by the addition of
oven-dried 4 Å molecular sieves such that stirring was not
impeded. The mixture was stirred 1.5 h and then partitioned
between EtOAc and sat. NaHCO3 (aq). The phases were
separated, and the organic phase was washed with brine, dried
(MgSO4), and concentrated to provide 3.96 g (95%) of 2-cyano-
4-ethylhex-2-enoic acid benzyl ester 8 as a colorless oil. 1H
NMR (CDCl3) δ 7.39 (m, 6H), 5.28 (s, 2H), 2.60 (m, 1H), 1.62
(m, 2H), 1.40 (m, 2H), 0.88 (t, J ) 7.6 Hz, 6H). 13C NMR
(CDCl3) δ 168.7, 161.4, 135.0, 128.9, 128.8, 128.5, 114.1, 109.9,
68.1, 46.4, 27.3, 12.0. IR (neat) 2233, 1729 cm-1. LRMS: m/z
256.1 (M-1). Anal. (C16H19NO2) C, H, N.
Chart 1. Anticonvulsant Activity of Pregabalin 1 and
Compound 34 in DBA/2 Micea
2-Cyano-3-propyl-hex-2-enoic Acid Benzyl Ester 9.
Prepared as above for compound 8 from 4-heptanone 7 (2.0
mL, 14.3 mmol) and benzyl cyanoacetate (2.51 g, 14.3 mmol)
to provide 1.03 g (27%) of 2-cyano-3-propyl-hex-2-enoic acid
a % protection is the fraction of DBA/2 mice (N ) 10 animals)
protected from audiogenically induced tonic seizures at 2 h after
dosing by a 30 mg/kg po dose of the test compound, or a 30 µg icv
dose using saline as vehicle (see Experimental Section).
1
benzyl ester 9 as a colorless oil. H NMR (CDCl3) δ 7.35 (m,
5H), 5.24 (s, 2H), 2.71 (m, 2H), 2.52 (m, 2H), 1.60 (m, 2H),
1.49 (m, 2H), 1.00 (t, J ) 7.3 Hz, 3H), 0.95 (t, J ) 7.3 Hz, 3H).
13C NMR (CDCl3) δ 182.5, 161.7, 135.4, 128.8, 128.6, 128.2,
115.9, 104.8, 67.3, 40.8, 35.8, 22.4, 21.9, 14.5, 14.3. LRMS: m/z
272.1 (M + 1). IR (neat) 2223, 1729 cm-1. Anal. (C17H21NO2)
C, H, N.
Table 2. Comparison of Brain and Plasma Levels for
Pregabalin 1 and Compound 34 at 2 Hours Following 30 mg/kg
po Dosing in Sprague-Dawley Rats
system L
(IC50, µM)
brain levels
(ng/mL)
plasma levels
(ng/mL)
1-Cyano-2-(1-ethylpropyl)cyclopropanecarboxylic Acid
Benzyl Ester 10. To a solution of 2-cyano-4-ethylhex-2-enoic
acid benzyl ester 8 (3.76 g, 14.6 mmol) in 80 mL of acetonitrile
was added nitromethane (3.95 mL, 73 mmol), followed by DBU
(2.18 mL, 14.6 mmol), resulting in an orange solution. The
reaction was heated to 60 °C for 16 h and then cooled and
partitioned between Et2O and 1 N HCl (aq). The phases were
separated, and the organic phase was washed with brine, dried
(MgSO4), and concentrated. Flash chromatography of the
residue (5f10% EtOAc/hexanes) afforded 2.63 g (66%) of
1-cyano-2-(1-ethylpropyl)cyclopropanecarboxylic acid benzyl
ester 10 as a pale yellow oil. 1H NMR (CDCl3) δ 7.35 (m, 5H),
5.22 (m, 2H), 1.86 (dd, J ) 4.6, 9.0 Hz, 1H), 1.72 (m, 1H), 1.45-
1.57 (m, 4H), 1.42 (dd, J ) 4.6, 8.3 Hz 1H), 1.14 (m, 1H), 0.93
(m, 6H). 13C NMR (CDCl3) δ 168.2, 135.1, 128.9, 128.7, 128.2,
117.8, 68.3, 42.4, 36.7, 26.2, 26.0, 25.2, 19.3, 11.2, 10.8.
LRMS: m/z 272.1 (M + 1). IR (neat) 2245, 1734 cm-1. Anal.
(C17H21NO2) C, H, N.
1-Cyano-2,2-dipropylcyclopropanecarboxylic Acid Ben-
zyl Ester 11. To a solution of 2-cyano-3-propylhex-2-enoic acid
benzyl ester 9 (0.91 g, 3.35 mmol) in 50 mL of acetonitrile was
added nitromethane (0.91 mL, 16.8 mmol), followed by DBU
(0.50 mL, 3.35 mmol), resulting in an orange solution. The
reaction was stirred at room temperature for 16 h and then
partitioned between Et2O and 1 N HCl (aq). The phases were
separated, and the organic phase was washed with brine, dried
(MgSO4), and concentrated. Flash chromatography of the
residue (5f10% EtOAc/hexanes) afforded 0.71 g (81%) of
1-cyano-2,2-dipropylcyclopropanecarboxylic acid benzyl ester
compd
B/P
1
34
158 ( 35
>2490
4330 ( 347
385 ( 88.7
19367 ( 2021
7070 ( 572
0.22
0.054
compounds were administered by injection into the
cerebral ventricles.
Experimental Section
Unless otherwise indicated, all reagents were purchased
from Sigma-Aldrich Corporation and used without purification.
All reactions were monitored by thin-layer chromatography
on Merck glass plates precoated with 0.25 mm of silica gel.
Chromatography for purification was done with Merck silica
gel (230-400 mesh) or Biotage cartridges packed with 32-63
µM silica gel. Melting points were obtained on a Thomas-
Hoover capillary melting point apparatus and are uncorrected.
Mass spectra were obtained on a Micromass Platform LC mass
spectrometer and are chemical ionization spectra. NMR spec-
tra were obtained on either a Varian 300 or a Varian 400
spectrometer with tetramethylsilane as an internal standard.
Combustion analyses were performed by QTI Laboratories.
2-Cyano-4-ethylhex-2-enoic Acid Benzyl Ester 8. To a
mixture of 2-ethylbutyraldehyde 6 (2.0 mL, 16.2 mmol) and
benzyl cyanoacetate (2.85 g, 16.2 mmol) at 0 °C were added
acetic acid (0.1 mL, 1.62 mmol) and then piperidine (0.16 mL,
1.62 mmol) dropwise. The ice bath was removed and stirring

3030 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8
Schwarz et al.
1
11 as a pale yellow oil. H NMR (CDCl3) δ 7.35 (m, 5H), 5.21
a clear oil: 1H NMR (300 MHz, CDCl3) δ 3.82 (s, 3H), 2.1-
2.19 (m, 1H), 2.14 (d, J ) 5.0 Hz, 1H), 1.70-1.88 (m, 7H), 1.64
(d, J ) 5.0 Hz, 1H). MS (APCI) m/z 180 [M + H]+.
(dd, J ) 12.4, 18.3 Hz, 2H), 1.79 (d, J ) 5.1 Hz, 1H), 1.74 (m,
1H), 1.52 (m, 3H), 1.43 (d, J ) 5.1 Hz, 1H), 1.36 (m, 3H), 1.04
(m, 1H), 0.95 (t, J ) 7.1 Hz, 3H), 0.73 (t, J ) 7.1 Hz, 3H). 13
C
1-Aminomethylspiro[2.4]heptane-1-carboxylic Acid
Methyl Ester 17. To a solution of 1-cyanospiro[2.4]heptane-
1-carboxylic acid methyl ester 16 (3.45 g, 19.3 mmol) in
methanol (240 mL) was added cobalt chloride hexahydrate
(CoCl2-6H2O, 9.16 g, 38.5 mmol) to give a deep purple colored
solution. Sodium borohydride (7.30 g, 193 mmol) was added
portionwise over 10 min with caution to control the evolution
of hydrogen and the exothermic reaction that ensued to give
a black solution. The reaction mixture was stirred for 30 min
under nitrogen after the addition was completed and then
quenched carefully by addition of 0.5 N HCl (aq, 1.3 L). The
solution was made alkaline (pH ∼ 9) by the addition of
concentrated NH4OH (aq). The mixture was extracted with
ethyl acetate (4 × 400 mL). The combined organic phases were
dried (Na2SO4), and concentrated to afford 3.07 g (87%) of
1-aminomethylspiro[2.4]heptane-1-carboxylic acid methyl ester
17 as a yellow oil: 1H NMR (300 MHz, CDCl3) δ 3.48 (s, 3H),
3.16 (d, J ) 13.8 Hz, 1H), 2.58 (d, J ) 13.8 Hz, 1H), 1.9 (m,
1H), 1.4-1.7 (m, 10H), 0.74 (br d, 1H). MS (APCI) m/z 184 [M
+ H]+.
1-Aminomethylspiro[2.4]heptane-1-carboxylic Acid 18.
To a solution of 1-aminomethylspiro[2.4]heptane-1-carboxylic
acid methyl ester 17 (3.40 g, 18.6 mmol) were added methanol
(75 mL) and lithium hydroxide monohydrate (LiOH-H2O)
(1.55 g, 37.1 mmol). The mixture was heated to reflux under
nitrogen for 48 h. The solvent was removed by evaporation
under reduced pressure, and the residue was dissolved in
water (50 mL). With ice bath cooling, concentrated HCl (∼2.5
mL) was added until the pH was adjusted to 6. A white
precipitate was isolated by filtration and dried under vacuum
to yield 1.22 g (39%) of 1-aminomethylspiro[2.4]heptane-1-
carboxylic acid 18 as an off-white solid: mp 231-235 °C (dec);
1H NMR (300 MHz, D2O) δ 3.43 (d, J ) 13.2 Hz, 1H), 2.86 (d,
J ) 13.2 Hz, 1H), 1.54-1.77 (m, 7 H), 1.40 (m, 1H), 1.32 (d, J
) 4.7 Hz, 1H), 0.78 (d, J ) 4.7 Hz, 1H). MS (APCI) m/z 170
[M + H]+. Anal. (C9H15NO2‚H2O) C, H, N.
Cyanocyclohexylidene-acetic Acid Methyl Ester 22. To
a solution of cyclohexanone 19 (84.1 g, 1.0 mol) in dry benzene
(100 mL) were added methyl cyanoacetate (99.1 g, 1.0 mol),
ammonium acetate (10 g) and glacial acetic acid (20 mL). The
reaction mixture was heated to reflux using Dean-Stark
apparatus for 12 h and allowed to cool to room temperature.
Excess solvent was removed in vacuo and the residue dissolved
in EtOAc (400 mL). The organic phase was washed with water,
dried over sodium sulfate, and evaporated to give cyanocyclo-
hexylideneacetic acid methyl ester 22 as pale yellow oil.
Further purification of the pale yellow oil by distillation (10
mmHg, 150-155 °C) gave 110 g (61%) of 22 as a colorless oil:
1H NMR (300 MHz, CDCl3) δ 3.80 (s, 3H), 3.00 (t, 2H), 2.70 (t,
2H), 1.70-1.85 (m, 6H). 13C NMR (75 MHz, CDCl3) δ 180.8,
162.7, 115.8, 101.9, 52.8, 37.2, 31.9, 28.9, 28.6, 25.9.
2-Cyano-4-methylpent-2-enoic Acid Methyl Ester 23.
To a solution of isobutyraldehyde 20 (18.2 mL, 200 mmol) in
dry benzene (20 mL) were added methyl cyanoacetate (18 mL,
200 mmol), ammonium acetate (2 g), and glacial acetic acid (4
mL). The reaction mixture was stirred at 60 °C for 30 min
and allowed to cool to room temperature. Excess solvent was
removed in vacuo and the residue dissolved in EtOAc (200 mL).
The organic phase was washed with water, dried (Na2SO4),
and concentrated. Further purification of the pale yellow oil
by distillation (7 mmHg) gave 27.4 g (56%) of 2-cyano-4-
methylpent-2-enoic acid methyl ester 23 as a colorless oil: 1H
NMR (300 MHz, CDCl3) δ 7.47 (d, J ) 10.5 Hz, 1H), 3.87 (s,
3H), 2.96-3.04 (m, 1H), 1.16 (d, J ) 6.6 Hz, 6H).
NMR (CDCl3) δ 166.4, 135.2, 128.8, 128.7, 128.5, 118.6, 68.2,
41.3, 36.5, 30.3, 29.8, 24.9, 19.7, 19.7, 14.2, 14.1. IR (neat) 2240,
1734 cm-1. Anal. (C18H23NO2) C, H, N.
1-Aminomethyl-2-(1-ethylpropyl)cyclopropanecarbox-
ylic Acid 12. To a solution of 1-cyano-2-(1-ethylpropyl)-
cyclopropanecarboxylic acid benzyl ester 10 (1.76 g, 6.5 mmol)
in 25 mL of THF was added 20% Pd/C (0.2 g). The mixture
was hydrogenated in a Parr shaker at 48 psi for 3 h. 1H NMR
analysis indicated the benzyl ester had been completely
cleaved. The mixture was filtered, concentrated hydrochloric
acid (2 g) was added along with PtO2 (0.42 g), and the mixture
was hydrogenated for an additional 16 h until the nitrile was
reduced. The mixture was filtered and concentrated. Flash
chromatography of the residue on silica gel (0.25:1.25:3.5
concentrated NH4OH (aq)/MeOH/CH2Cl2), followed by ion-
exchange chromatography on a DOWEX-50WX8-100 resin,
provided 0.13 g (11%) of 1-aminomethyl-2-(1-ethylpropyl)-
cyclopropanecarboxylic acid 12 as a colorless solid, mp 255-
257 °C: 1H NMR (D2O) δ 3.53 (d, J ) 13.2 Hz, 1H), 2.50 (d, J
) 13.2 Hz, 1H), 1.30 (m, 3H), 1.08-1.23 (m, 4H), 0.74 (t, J )
7.6 Hz, 3H), 0.65 (t, J ) 7.6 Hz, 3H), 0.54 (m, 1H). LRMS:
m/z 184.1 (M-1). Anal. (C10H19NO2) C, H, N.
1-Aminomethyl-2,2-dipropylcyclopropanecarboxylic
Acid 13. To a solution of 1-cyano-2,2-dipropylcyclopropane-
carboxylic acid benzyl ester 11 (0.35 g, 1.79 mmol) in 15 mL
of MeOH were added concentrated HCl (0.2 g) and platinum
dioxide (PtO2, 0.035 g). The mixture was hydrogenated in a
Parr shaker at 48 psi for 21 h. The mixture was filtered and
concentrated. Flash chromatography of the residue on silica
gel (0.25:1.25:3.5 concentrated NH4OH (aq)/MeOH/CH2Cl2),
followed by ion-exchange chromatography on a DOWEX-
50WX8-100 resin provided 0.09 g (18%) of 1-aminomethyl-
2,2-dipropylcyclopropanecarboxylic acid 13 as a colorless solid,
mp 255-257 °C: 1H NMR (CD3OD) δ 3.26 (d, J ) 12.9 Hz,
1H), 3.03 (d, J ) 12.9 Hz, 1H), 1.69 (m, 1H), 1.56 (m, 2H),
1.28-1.50 (m, 4H), 1.26 (d, J ) 4.4 Hz, 1H), 1.17 (m, 1H), 0.93
(t, J ) 7.3 Hz, 3H), 0.85 (t, J ) 7.3 Hz, 3H), 0.54 (d, J ) 4.6
Hz, 1H). LRMS: m/z 198.1 (M - 1). Anal. (C11H21NO2) C, H,
N.
Cyanocyclopentylideneacetic Acid Methyl Ester 15. To
a solution of cyclopentanone 14 (84.1 g, 1.0 mol) in dry benzene
(100 mL) were added methyl cyanoacetate (99.1 g, 1.0 mol),
ammonium acetate (10 g), and glacial acetic acid (20 mL). The
reaction mixture was heated to reflux using Dean-Stark
apparatus for 12 h and allowed to cool to room temperature.
Excess solvent was removed in vacuo and the residue dissolved
in EtOAc (400 mL). The organic phase was washed with water,
dried (Na2SO4), and evaporated to give the product as pale
yellow oil. Further purification of the pale yellow oil by
distillation (10 mmHg, 140-145 °C) provided cyanocyclopen-
tylideneacetic acid methyl ester 15 (130 g, 79%) as a colorless
oil: 1H NMR (300 MHz, CDCl3) δ 3.80 (s, 3H), 3.0 (t, 2H), 2.80
(t, 2H), 1.80 (m, 4H). 13C NMR (75 MHz, CDCl3) δ 188.3, 162.6,
115.8, 100.7, 52.5, 38.1, 35.8, 26.9, 25.4.
1-Cyanospiro[2.4]heptane-1-carboxylic Acid Methyl
Ester 16. To a solution of cyanocyclopentylideneacetic acid
methyl ester 15 (20.8 g, 126 mmol) in acetonitrile (500 mL)
was added nitromethane (34 mL, 630 mmol) followed by
dropwise addition of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)
(18.8 mL, 126 mmol). The reaction solution turned orange upon
the addition of DBU. The reaction mixture was stirred at room
temperature for 16 h. Another portion of DBU (1.0 mL) was
added and the mixture stirred for 1 h. The reaction mixture
was partitioned between ether (1 L) and 1 N HCl (400 mL),
and the phases were separated. The organic phase was washed
with 1 N HCl (2 × 300 mL) and brine (2 × 200 mL), dried
over sodium sulfate, filtered, and evaporated. The residue was
chromatographed on a wet-packed silica gel column eluting
with 4f6% EtOAc/hexanes to furnish 14.0 g (62%) of
1-cyanospiro[2.4]heptane-1-carboxylic acid methyl ester 16 as
2-Cyano-5-methylhex-2-enoic Acid Methyl Ester 24. To
a solution of isovaleraldehyde 21 (86.1 g, 1 mol) in dry benzene
(100 mL) were added methyl cyanoacetate (99.1 g, 1 mol),
ammonium acetate (10 g), and glacial acetic acid (20 mL). The
reaction mixture was stirred at 0 °C for 1 h and allowed to
cool to room temperature. Excess solvent was removed in vacuo
and the residue dissolved in EtOAc (400 mL). The organic

Cyclopropyl â-Amino Acid Analogues of Pregabalin
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8 3031
phase was washed with water, dried over sodium sulfate and
evaporated to give the product as pale yellow oil. Further
purification of the pale yellow oil by distillation (10 mmHg)
gave 100 g (60%) of 2-cyano-5-methylhex-2-enoic acid methyl
ester 24 as a colorless oil: 1H NMR (300 MHz, CDCl3) δ 7.69
(t, J ) 8.0 Hz, 1H), 3.89 (s, 3H), 2.47 (dd, J ) 7.9, 6.8 Hz, 2H),
1.91 (m, 1H), 1.00 (d, J ) 6.7 Hz, 6H).
1-Cyanospiro[2.5]octane-1-carboxylic Acid Methyl Es-
ter 25. To a solution of cyanocyclohexylideneacetic acid methyl
ester 22 (20.0 g, 112 mmol) in acetonitrile (400 mL) was added
nitromethane (30 mL, 558 mmol) followed by dropwise addition
of DBU (16.9 mL, 113 mmol) over 5 min. The solution went
from clear to orange and was stirred at room temperature for
5.5 h. The solution was diluted with ether (1 L), washed with
1 N HCl (2 × 250 mL) and then brine (2 × 250 mL). The
organic layer was dried over sodium sulfate, filtered, and
evaporated. The residue was chromatographed on a wet-
packed silica gel column (6.5 × 40 cm) eluting with 5% EtOAc/
hexanes to furnish 18.6 g (86%) of 1-cyanospiro[2.5]octane-1-
carboxylic acid methyl ester 25 as a clear oil: 1H NMR (300
MHz, CDCl3) δ 3.83 (s, 3H), 1.82 (d, J ) 5.0 Hz, 1H), 1.5-1.75
(m, 9H), 1.49 (d, J ) 5.0 Hz, 1H), 1.35 (m, 1H).
1-Cyano-2-isopropylcyclopropanecarboxylic Acid Meth-
yl Ester 26. To a solution of 2-cyano-4-methylpent-2-enoic acid
methyl ester 23 (10.5 g, 68.3 mmol) in dry acetonitrile (60 mL)
were added nitromethane (5.5 mL, 103 mmol) and a portion-
wise addition of alumina-supported potassium fluoride (40 wt
%, 22 g). The reaction mixture was stirred under reflux for 2
h and cooled to room temperature. The solid was removed by
filtration through a short pad of Celite and washed with
acetonitrile. Excess solvent was removed in vacuo and the
residue dissolved in ether. The ether phase was washed with
water and brine, dried (Na2SO4), and concentrated. The crude
product was purified by flash column chromatography (20%
EtOAc/hexanes) to afford 7.7 g (68%) of 1-cyano-2-isopropyl-
cyclopropanecarboxylic acid methyl ester 26 as a colorless oil:
1H NMR (300 MHz, CDCl3) δ 3.81 (s, 3H), 1.81 (dd, J ) 8.9,
4.4 Hz, 1H), 1.65-1.74 (m, 1H), 1.39-1.45 (m, 2H), 1.13 (t, J
) 7.2 Hz, 6H). 13C NMR (75 MHz, CDCl3) δ 168.7, 117.6, 53.7,
39.3, 31.2, 25.3, 21.9, 21.8, 19.6.
1-Cyano-2-isobutylcyclopropanecarboxylic Acid Meth-
yl Ester 27. To a solution of 2-cyano-5-methylhex-2-enoic acid
methyl ester 24 (20.2 g, 121 mmol) in dry acetonitrile (110
mL) were added nitromethane (9.8 mL, 181 mmol) and a
portionwise addition of alumina-supported potassium flouride
(40 wt %, 39.3 g). The reaction mixture was stirred under
reflux for 2 h and cooled to room temperature. The solid was
removed by filtration through a short pad of Celite and washed
with acetonitrile. Excess solvent was removed in vacuo and
the residue dissolved in ether. The ether phase was washed
with water and brine, dried (Na2SO4), and concentrated. The
crude product was purified by flash column chromatography
(20% EtOAc/hexanes) to give 15 g (69%) of 1-cyano-2-isobu-
tylcyclopropanecarboxylic acid methyl ester 27 as colorless
oil: 1H NMR (300 MHz, CDCl3) δ 3.81 (s, 3H), 1.79-1.94 (m,
3H), 1.65 (m, 1H), 1.35-1.42 (m, 2H), 1.00 (d, J ) 3.0 Hz, 3H),
0.98 (d, J ) 2.9 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 168.5,
117.3, 53.4, 39.0, 30.1, 27.9, 25.5, 22.4, 22.3, 19.3.
ester 28 as an off-white solid. The material was used without
further purification in the next step.
1-(tert-Butoxycarbonylaminomethyl)-2-isopropylcy-
clopropanecarboxylic Acid Methyl Ester 29. To a solution
of 1-cyano-2-isopropylcyclopropanecarboxylic acid methyl ester
26 (7.7 g, 46.1 mmol) and cobaltous chloride hexahydrate (21.9
g, 92.2 mmol) in MeOH (560 mL) was added sodium borohy-
dride (17.4 g, 461 mmol) in portions. The black precipitate
formed was stirred for 1 h at room temperature and was
quenched with 0.5N HCl (aq, 200 mL). After the black
precipitate dissolved, excess solvent was removed and the
aqueous phase made alkaline by addition of 2 N NaOH (aq).
The alkaline solution was added slowly to a solution of di-tert-
butyl dicarbonate (20.1 g, 92.2 mmol) in dichloromethane (400
mL). The reaction mixture was stirred at room-temperature
overnight, and the organic phase was separated, washed with
brine, and dried (Na2SO4). Evaporation of solvent followed by
purification of the residue by column chromatography (10%
EtOAc/hexanes) furnished 10.9 g (88%) of 1-(tert-butoxycar-
bonylaminomethyl)-2-isopropylcyclopropanecarboxylic acid
methyl ester 29 as a colorless oil: 1H NMR (300 MHz, CD3-
OD) δ 3.69 (s, 3H), 3.57 (d, J ) 14.3 Hz, 1H), 3.23 (d, J ) 14.3
Hz, 1H), 1.43 (s, 9H), 1.26-1.31 (m, 3H), 1.05 (d, J ) 6.3 Hz,
3H), 0.97 (d, J ) 6.3 Hz, 3H), 0.76 (m, 1H). 13C NMR (75 MHz,
CD3OD) δ 177.1, 158.5, 80.5, 53.0, 41.1, 38.1, 29.2, 23.6, 23.4,
20.5. MS (APCI) m/z 172 [M + H-100(Boc)]+.
1-(tert-Butoxycarbonylaminomethyl)-2-isobutylcyclo-
propanecarboxylic Acid Methyl Ester 30. Prepared as
above for compound 29 from 1-cyano-2-isobutylcyclopropane-
carboxylic acid methyl ester 27 (13.0 g, 71.8 mmol) to afford
16 g (78%) of 1-(tert-butoxycarbonylaminomethyl)-2-isobutyl-
cyclopropanecarboxylic acid methyl ester 30 as a colorless oil:
1H NMR (300 MHz, CDCl3) δ 3.67 (s, 3H), 5.27 (m, 1H), 3.33-
3.46 (m, 2H), 1.64-1.70 (m, 1H), 1.51-1.58 (m, 2H), 1.44 (s,
9H), 1.37 (dd, J ) 8.5, 3.8 Hz, 1H), 1.20-1.28 (m, 1H), 0.93 (t,
J ) 6.5 Hz, 6H), 0.75 (m, 1H). 13C NMR (75 MHz, CDCl3) δ
175.8, 156.1, 79.1, 52.1, 40.2, 37.8, 28.9, 28.7, 28.6, 26.7, 22.9,
22.4, 21.1. MS (APCI) m/z 186 [M + H - 100(Boc)]+.
1-(tert-Butoxycarbonylaminomethyl)spiro[2.5]octane-
1-carboxylic acid 31. To a solution of crude 1-(tert-butoxy-
carbonylaminomethyl)spiro[2.5]octane-1-carboxylic acid meth-
yl ester 28 (35.0 g, ∼96.2 mmol) in MeOH (400 mL) was added
lithium hydroxide monohydrate (12.3 g, 294 mmol) followed
by the addition of water (75 mL). The mixture was heated to
reflux. After 48 h, an additional portion of LiOH-H2O (6.15
g, 146 mmol) was added and refluxed for an additional 24 h.
The solvent was removed under reduced vacuum and the
residue partitioned between water and ether. The aqueous
phase was acidified to pH 2-3 with 2 N HCl (aq) with ice bath
cooling with a layer of CH2Cl2 present. The phases were
separated, and the aqueous phase was extracted 3x with CH2-
Cl2. The combined organic phases were dried (Na2SO4), filtered,
and evaporated to give 17.0 g (63%, two steps) of 1-(tert-
butoxycarbonylaminomethyl)spiro[2.5]octane-1-carboxylic acid
31 an off-white solid.
1-(tert-Butoxycarbonylaminomethyl)-2-isopropylcy-
clopropanecarboxylic acid 32. To a solution of 1-(tert-
butoxycarbonylaminomethyl)-2-isopropylcyclopropanecarbox-
ylic acid methyl ester 29 (10.9 g, 40.2 mmol) in MeOH (320
mL) was added a solution of lithium hydroxide monohydrate
(4.2 g, 100 mmol) in water (98 mL). The reaction mixture was
heated to reflux for 3 h and cooled to room temperature. Excess
solvent was removed, and the residue was dissolved in water
(100 mL). The aqueous solution was washed with ether,
acidified to pH ) 3 with 2 N HCl (aq), and extracted with
EtOAc (2 × 150 mL). The combined organic phases were
washed with brine, dried (Na2SO4), and concentrated to
provide 9.0 g (87%) of 1-(tert-butoxycarbonylaminomethyl)-2-
isopropylcyclopropanecarboxylic acid 32 as a colorless oil: 1H
NMR (300 MHz, CD3OD) δ 3.54 (d, J ) 14.3 Hz, 1H), 3.23 (d,
J ) 14.3 Hz, 1H), 1.42 (s, 9H), 1.28-1.36 (m, 3H), 1.06 (d, J )
6.2 Hz, 3H), 0.99 (d, J ) 6.2 Hz, 3H), 0.72 (m, 1H). 13C NMR
(75 MHz, CD3OD) δ 178.5, 158.1, 80.2, 40.8, 37.6, 30.2, 28.8,
23.3, 23.0, 20.0. MS (APCI) m/z 258 [M + H]+.
1-(tert-Butoxycarbonylaminomethyl)-spiro[2.5]octane-
1-carboxylic Acid Methyl Ester 28. To a solution of
1-cyanospiro[2.5]octane-1-carboxylic acid methyl ester 25 (18.6
g, 96.2 mmol) in methanol (480 mL) was added CoCl2-H2O
(22.8 g, 96.2 mmol) to give a deep purple color. Sodium
borohydride (14.5 g, 385 mmol) was added portionwise over
10 min with ice bath cooling to give a black colored solution.
After 40 min, the reaction was quenched carefully with 2 N
HCl (aq). The pH was adjusted to 9 with 2 N NaOH (aq) and
the solution was extracted 4x with CH2Cl2. To the combined
organic phases were added saturated K2CO3 (aq) and di-tert-
butyl dicarbonate (26.2 g, 120 mmol). After 1 h, the phases
were separated. The organic phase was dried (Na2SO4), and
concentrated to yield 35.7 g (>100%) of 1-(tert-butoxycar-
bonylaminomethyl)spiro[2.5]octane-1-carboxylic acid methyl

3032 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8
Schwarz et al.
1-(tert-Butoxycarbonylaminomethyl)-2-isobutylcyclo-
propanecarboxylic Acid 33. To a solution of 1-(tert-butoxy-
carbonylaminomethyl)-2-isobutylcyclopropanecarboxylic acid
methyl ester 30 (9.1 g, 31.9 mmol) in MeOH (260 mL) was
added a solution of lithium hydroxide monohydrate (3.34 g,
79.7 mmol) in water (80 mL). The reaction mixture was heated
to reflux for 3 h and cooled to room temperature. Excess
solvent was removed and the residue dissolved in water (100
mL). The aqueous solution was washed with ether, acidified
to pH ) 3 with 2 N HCl (aq), and extracted with EtOAc (2 ×
150 mL). The combined organic phases were washed with
brine, dried (Na2SO4), and concentrated to give 8.4 g (97%) of
1-(tert-butoxycarbonylaminomethyl)-2-isobutylcyclopropane-
carboxylic acid 33 as a colorless oil: 1H NMR (300 MHz, CD3-
OD) δ 3.40 (d, J ) 14.2 Hz, 1H), 3.23 (d, J ) 14.2 Hz, 1H),
1.64-1.69 (m, 2H), 1.49-1.55 (m, 1H), 1.43 (s, 9H), 1.35 (m,
1H), 0.95 (t, J ) 6.4 Hz, 6H), 0.71 (dd, J ) 6.4, 4.0 Hz, 1H).
162.0, 114.8, 106.0, 62.4, 16.3, 14.4, 11.7. LRMS: m/z 166.0
(M + 1). Anal. (C9H11NO2) C, H, N.
2-Cyano-bicyclopropyl-2-carboxylic Acid Ethyl Ester
39. To a solution of 2-cyano-3-cyclopropylacrylic acid ethyl
ester 38 (2.0 g, 12.1 mmol) in 40 mL of acetonitrile was added
nitromethane (3.3 mL, 60.9 mmol) followed by DBU (1.8 mL,
12.0 mmol). The mixture was stirred for 2 d, then at 50 °C for
12 h. The mixture was cooled and partitioned between Et2O
and 1 M HCl (aq), and the organic phase was separated,
washed with brine, dried (Na2SO4), and concentrated. Flash
chromatography of the residue (1f5% EtOAc/hexanes) fur-
nished 0.30 g (14%) of 2-cyano-bicyclopropyl-2-carboxylic acid
1
ethyl ester 39 as a colorless oil. H NMR (CDCl3) δ 4.19 (m,
2H), 1.74 (m, 2H), 1.39 (m, 1H), 1.28 (t, J ) 7.1 Hz, 3H), 0.89
(m, 1H), 0.70 (m, 1H), 0.62 (m, 1H), 0.34 (m, 2H). 13C NMR
(CDCl3) δ 168.0, 117.6, 62.9, 35.3, 23.7, 20.0, 14.3, 10.1, 4.8,
4.0. LRMS: m/z 180.0 (M + 1). Anal. (C10H13NO2) C, H, N.
1-Aminomethylspiro[2.5]octane-1-carboxylic Acid Hy-
drochloride 34. To a solution of 1-(tert-butoxycarbonylamino-
methyl)spiro[2.5]octane-1-carboxylic acid 31 (200 mg, 0.706
mmol) in 1,4-dioxane (3 mL) was added 4 N HCl in 1,4-dioxane
(3 mL). The reaction was stirred for 18 h at room temperature.
Ether (20 mL) was added, and the precipitate was isolated by
vacuum filtration. The solid was dried in a vacuum oven at
48 °C to give 133 mg (86%) of 1-aminomethylspiro[2.5]octane-
1-carboxylic acid hydrochloride 34 as a white solid: mp 240-
242 °C; 1H NMR (300 MHz, D2O) δ 3.61 (d, J ) 14.0 Hz, 1H),
3.12 (d, J ) 14.0 Hz, 1H), 1.4-1.6 (m, 10H), 1.40 (d, J ) 5.2
Hz, 1H), 0.92 (d, J ) 5.2 Hz, 1H). MS (APCI) m/z 184 [M +
H]+. Anal. (C10H17NO2‚HCl) C, H, N, Cl.
1-Aminomethyl-2-isopropylcyclopropanecarboxylic
Acid Hydrochloride 35. To a solution of 1-(tert-butoxycar-
bonylaminomethyl)-2-isopropylcyclopropanecarboxylic acid 32
(2.05 g, 8.0 mmol) in dry 1,4-dioxane (40 mL) was added 4 N
HCl (40 mL, in 1,4-dioxane). The reaction mixture was stirred
at room-temperature overnight and ether was added (100 mL).
The white solid was collected and dried to afford 1.35 g (88%)
of 1-aminomethyl-2-isopropylcyclopropanecarboxylic acid hy-
drochloride 35, mp 245-246 °C: 1H NMR (300 MHz, D2O) δ
3.83 (d, J ) 13.7 Hz, 1H), 2.85 (d, J ) 13.7 Hz, 1H), 1.49-
1.59 (m, 2H), 1.23-1.31 (m, 1H), 1.05 (d, J ) 6.4 Hz, 3H), 0.96
(m, 1H), 0.94 (d, J ) 6.5 Hz, 3H). 13C NMR (75 MHz, D2O) δ
179.6, 41.7, 39.6, 30.8, 27.8, 23.9, 22.3. MS (APCI) m/z 158 [M
+ H]+. Anal. (C8H15NO2‚HCl) C, H, N, Cl.
1-Aminomethyl-2-isobutylcyclopropanecarboxylic Acid
Hydrochloride 36. To a solution of 1-(tert-butoxycarbonyl-
aminomethyl)-2-isobutylcyclopropanecarboxylic acid 33 (2.8 g,
10.3 mmol) in dry 1,4-dioxane (60 mL) was added 4 N HCl
(40 mL, in 1,4-dioxane). The reaction mixture was stirred at
room-temperature overnight, and then ether was added (100
mL). The white solid was collected and dried to furnish 1.6 g
(76%) of 1-aminomethyl-2-isobutylcyclopropanecarboxylic acid
hydrochloride 36, mp 254-255 °C: 1H NMR (300 MHz, D2O)
δ 3.58 (d, J ) 13.8 Hz, 1H), 3.06 (d, J ) 13.8 Hz, 1H), 1.51-
1.80 (m, 4H), 1.21-1.28 (m, 1H), 0.96 (m, 1H), 0.95 (t, J ) 6.2
Hz, 6H). 13C NMR (75 MHz, D2O) δ 178.1, 40.2, 37.1, 28.6,
28.1, 25.3, 22.2, 21.9, 21.7. MS (APCI) m/z 172 [M + H]+. Anal.
(C9H17NO2‚HCl) C, H, N.
2-Cyano-3-cyclopropylacrylic Acid Ethyl Ester 38. To
a mixture of cyclopropanecarboxaldehyde 37 (2.4 mL, 32.1
mmol) and ethyl cyanoacetate (3.8 mL, 35.3 mmol) at 0 °C were
added acetic acid (0.2 mL) and then piperidine (0.3 mL)
dropwise. The ice bath was removed, and stirring was contin-
ued for 10 min. Additional acetic acid (0.2 mL) and piperidine
(0.3 mL) were added, followed by the addition of oven-dried 4
Å molecular sieves such that stirring was not impeded. The
mixture was stirred for 12 h, and then partitioned between
Et2O and sat. NaHCO3 (aq). The phases were separated, and
the organic phase washed with brine, dried (Na2SO4), and
concentrated. Flash chromatography of the residue (3f5f9%
EtOAc/hexanes) furnished 3.71 g (70%) of 2-cyano-3-cyclopro-
pylacrylic acid ethyl ester 38 as a colorless oil. 1H NMR (CDCl3)
δ 6.97 (d, J ) 11.2 Hz, 1H), 4.27 (q, J ) 7.2 Hz, 2H), 2.09 (m,
1H), 1.32 (m, 5H), 0.95 (m, 2H). 13C NMR (CDCl3) δ 169.0,
2-Aminomethylbicyclopropyl-2-carboxylic Acid 40. To
a solution of 2-cyanobicyclopropyl-2-carboxylic acid ethyl ester
39 (0.14 g, 0.78 mmol) in 48 mL EtOH was added RaNi (0.18
g). The mixture was hydrogenated in a Parr shaker at 48 psi
for 16 h, filtered, and concentrated. To the crude aminoester
were added 5 mL of MeOH and then 5 mL of 10% NaOH (aq),
and the mixture was stirred 12 h. The mixture was acidified
to pH 2 by addition of 6 M HCl (aq) and then concentrated to
remove MeOH. The residue was diluted with water and loaded
onto DOWEX-50WX8-100 resin. Elution with 50 mL of water
and then 50 mL 5% NH4OH (aq) and concentration of the
alkaline fractions provided a solid that was recrystallized from
MeOH/EtOAc to afford 0.036 g (30%) of 2-aminomethylbicy-
clopropyl-2-carboxylic acid 40 as a colorless solid. 1H NMR
(D2O) δ 3.22 (d, J ) 13.4 Hz, 1H), 2.95 (d, J ) 13.4 Hz, 1H),
1.24 (m, 1H), 0.99 (dd, J ) 8.8, 4.6 Hz, 1H), 0.54 (m, 1H), 0.49
(dd, J ) 6.6, 4.9 Hz, 1H), 0.33 (m, 2H), 0.05 (m, 2H). LRMS:
m/z 156.0 (M + 1). Anal. Calcd for C8H13NO2: C, 61.91; H,
8.44; N, 9.03. Found: 59.21; H, 8.26; N, 8.32.
1-Cyano-2-isobutylcyclopropanecarboxylic Acid Meth-
yl Ester 43. To a solution of 2-cyano-5-methylhex-2-enoic acid
methyl ester 4118 (6.7 g, 40 mmol) in 40 mL acetonitrile were
added nitromethane (2.2 mL, 40 mmol) and then DBU (6.0
mL, 40 mmol) resulting in a red-orange solution. The mixture
was heated to reflux 6 h, cooled, and concentrated. The residue
was partitioned between Et2O and 1 N HCl (aq). The aqueous
phase was extracted with Et2O, and the combined organics
were washed with brine, dried (MgSO4), and concentrated.
Flash chromatography of the residue (10f15f20% EtOAc/
hexanes) afforded 4.38 g (60%) of 1-cyano-2-isobutylcyclopro-
panecarboxylic acid methyl ester 43 as a yellow oil: 1H NMR
(CDCl3) δ 3.80 (s, 3H), 1.85 (m, 3H), 1.64 (m, 1H), 1.44 (m,
2H), 0.97 (d, J ) 6.6 Hz, 3H), 0.96 (d, J ) 6.6 Hz, 3H). 13C
NMR (CDCl3) δ 168.8, 117.7, 53.8, 39.2, 30.5, 28.1, 25.9, 22.7,
22.5, 19.5.
1-Cyano-2-(1-ethylpropyl)cyclopropanecarboxylic Acid
Methyl Ester 44. To a solution of 2-cyano-4-ethyl-hex-2-enoic
acid methyl ester 4230 (6.3 g, 35 mmol) in 35 mL of acetonitrile
were added nitromethane (3.0 mL, 5 mmol) and then DBU
(5.2 mL, 35 mmol), resulting in a red-orange solution. The
mixture was heated to reflux 3 h, cooled, and concentrated.
The residue was partitioned between Et2O and 1 N HCl (aq).
The aqueous phase was extracted with Et2O, and the combined
organics were washed with brine, dried (MgSO4), and concen-
trated. Flash chromatography of the residue (10f15% EtOAc/
hexanes) afforded 5.1 g (75%) of 1-cyano-2-(1-ethyl-propyl)-
cyclopropanecarboxylic acid methyl ester 44 as a yellow oil.
1H NMR (CDCl3) δ 3.80 (s, 3H), 1.84 (dd, J ) 9.0, 4.6 Hz, 1H),
1.71 (m, 1H), 1.54 (m, 3H), 1.46 (m, 1H), 1.41 (dd, J ) 8.1, 4.4
Hz, 1H), 1.12 (m, 1H), 0.93 (m, 6H). 13C NMR (CDCl3) δ 168.9,
118.0, 53.8, 42.6, 36.7, 26.2, 26.0, 25.2, 19.1, 11.2, 11.0.
1-Hydroxymethyl-2-isobutylcyclopropanecarboni-
trile 45. To a solution of 1-cyano-2-isobutylcyclopropanecar-
boxylic acid methyl ester 43 (3.6 g, 20 mmol) in 40 mL of THF
at ambient temperature was added a mixture of sodium
borohydride in THF:H2O (5:1, 30 mL). Additional NaBH4 (1.5
g) and 1.5 mL of water was added and stirred overnight and

Cyclopropyl â-Amino Acid Analogues of Pregabalin
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8 3033
quenched cautiously with 1 N HCl (aq), resulting in vigorous
gas evolution and exotherm. The mixture was concentrated
to remove THF and then partitioned between Et2O/1 N HCl
(aq). The organic phase was washed with sat. NaHCO3 (aq)
and brine, dried (MgSO4), and concentrated to provide 2.66 g
(86%) of 1-hydroxymethyl-2-isobutylcyclopropanecarbonitrile
trile 50. To a solution of 1-chloromethyl-2-(1-ethyl-propyl)-
cyclopropanecarbonitrile 48 (0.21 g, 1.1 mmol) in 5 mL of DMF
was added sodium azide (0.18 g, 2.8 mmol), and the mixture
heated to 100 °C for 1.5 h. The mixture was cooled and
partitioned between Et2O and water. The organic phase was
separated, washed with brine, dried (MgSO4), and concen-
trated to furnish 0.22 g (100%) of 1-azidomethyl-2-(1-ethyl-
propyl)cyclopropanecarbonitrile 50 as a pale yellow oil. 1H
NMR (CDCl3) δ 3.45 (d, J ) 12.9 Hz, 1H), 3.28 (d, J ) 13.2
Hz, 1H), 1.55 (m, 4H), 1.45 (m, 1H), 1.16 (m, 1H), 1.10 (m,
1H), 1.04 (m, 1H), 0.98 (t, J ) 7.6 Hz, 3H), 0.92 (t, J ) 7.3 Hz,
3H).
1
45 as a yellow oil. H NMR (CDCl3) δ 3.65 (d, J ) 11.7 Hz,
1H), 3.57 (d, J ) 11.7 Hz, 1H), 1.98 (br s, 1H), 1.77 (m, 1H),
1.54 (m, 1H), 1.35 (m, 1H), 1.20 (m, 1H), 1.12 (m, 1H), 0.97 (d,
J ) 6.6 Hz, 3H), 0.96 (d, J ) 6.6 Hz, 3H), 0.92 (m, 1H). 13C
NMR (CDCl3) δ 121.5, 66.5, 39.7, 28.4, 22.8, 22.7, 22.5, 19.1,
18.7.
2-(1-Ethyl-propyl)-1-hydroxymethylcyclopropanecar-
bonitrile 46. To a solution of 1-cyano-2-(1-ethyl-propyl)-
cyclopropanecarboxylic acid methyl ester 44 (4.5 g, 23 mmol)
in 40 mL of THF at ambient temperature was added a mixture
of sodium borohydride (4.0 g, 115 mmol) in THF/H2O (5:1, 40
mL). Additional THF and H2O was used to completely transfer
the borohydride. The mixture was stirred 3 d and then
quenched cautiously with 1 N HCl (aq). The mixture was
concentrated to remove THF and then partitioned between
Et2O/1 N HCl (aq). The organic phase was washed with brine,
dried (MgSO4), and concentrated to afford 3.5 g (91%) of 2-(1-
ethyl-propyl)-1-hydroxymethylcyclopropanecarbonitrile 46 as
a yellow oil. 1H NMR (CDCl3) δ 3.65 (d, J ) 11.7 Hz, 1H), 3.56
(d, J ) 11.7 Hz, 1H), 1.52 (m, 5H), 1.43 (m, 1H), 1.13 (m, 1H),
1.02 (m, 1H), 0.97 (t, J ) 7.6 Hz, 3H), 0.91 (t, J ) 7.6 Hz, 3H).
1-Aminomethyl-2-isobutylcyclopropanecarbonitrile 51.
To a solution of 1-azidomethyl-2-isobutylcyclopropanecarbo-
nitrile 49 (0.68 g, 3.8 mmol) in 50 mL of methanol was added
10% palladium on carbon (0.15 g), and the mixture was
hydrogenated at 48 psi in a Parr shaker for 8 h, filtered, and
concentrated to provide 0.56 g (96%) of 1-aminomethyl-2-
isobutylcyclopropanecarbonitrile 51 as a gray oil. 1H NMR
(CDCl3) δ 2.71 (d, J ) 13.9 Hz, 1H), 2.62 (d, J ) 13.7 Hz, 1H),
1.75 (m, 1H), 1.47 (m, 1H), 1.26 (m, 1H), 1.20 (m, 1H), 1.10
(m, 1H), 0.95 (d, J ) 7.3 Hz, 3H), 0.95 (d, J ) 6.6 Hz, 3H),
0.86 (m, 1H).
1-Aminomethyl-2-(1-ethylpropyl)cyclopropanecarbo-
nitrile 52. To a solution of 1-azidomethyl-2-(1-ethyl-propyl)-
cyclopropanecarbonitrile 50 (0.22 g, 1.1 mmol) in 4 mL of
methanol was added 10% palladium on carbon (0.05 g). The
mixture was hydrogenated at 48 psi in a Parr shaker for 2 h,
filtered, and concentrated to provide 0.19 g (100%) of 1-ami-
nomethyl-2-(1-ethyl-propyl)cyclopropanecarbonitrile 52 as a
1-Chloromethyl-2-isobutylcyclopropanecarbonitrile 47.
To a solution of 1-hydroxymethyl-2-isobutylcyclopropanecar-
bonitrile 45 (2.65 g, 17.3 mmol) in 50 mL of CH2Cl2 at ambient
temperature were added pyridine (5.0 mL, 62 mmol) and then
methanesulfonyl chloride (5.0 mL, 64 mmol) quickly dropwise.
The mixture was stirred 2 h and then poured into brine and
shaken. The organic phase was separated, dried (MgSO4), and
concentrated. To the residue was added 20 mL DMF and
sodium azide (2.0 g, 35 mmol) in one portion. On heating to
100 °C the mixture turned brown. It was then heated 90 min,
cooled to room temperature, and partitioned between Et2O and
water. The organic phase was washed with water and brine,
dried MgSO4, and concentrated. Flash chromatography of the
residue (5f10% EtOAc/hexanes) yielded 1.43 g (48%) of
1-chloromethyl-2-isobutylcyclopropanecarbonitrile 47 as a col-
1
pale yellow oil. H NMR (CDCl3) δ 2.76 (d, J ) 13.9 Hz, 1H),
2.60 (d, J ) 13.9 Hz, 1H), 1.52 (m, 5H), 1.14 (m, 1H), 0.99 (m,
8H).
1-Aminomethyl-2-isobutylcyclopropanecarboxylic Acid
53 (isobutyl and carboxylate groups cis). To 1-aminomethyl-
2-isobutylcyclopropanecarbonitrile 51 (0.56 g, 3.7 mmol) was
added 10 mL of 6 M HCl (aq) and the mixture refluxed for 5
d, cooled, and concentrated. The solid residue was dissolved
in water and washed with Et2O. The aqueous phase was
separated and loaded onto DOWEX-50WX8-100 ion-exchange
resin. The column was eluted with 100 mL of water and then
100 mL of 5% NH4OH (aq). The alkaline fractions were
concentrated to an off-white solid. 1HNMR of the residue
showed a trace amount of aminonitrile present. To the solid
was added hot methanol and water until it was mostly
dissolved. The liquid was withdrawn by pipet, filtered through
a glass wool plug, and concentrated to give 0.13 g (20%) of
1-aminomethyl-2-isobutylcyclopropanecarboxylic acid 53 as a
colorless solid. 1H NMR (D2O) δ 2.98 (d, J ) 13.2 Hz, 1H),
2.70 (d, J ) 13.2 Hz, 1H), 1.40 (m, 1H), 1.23 (m, 1H), 1.11 (m,
1H), 0.99 (m, 1H), 0.89 (m, 1H), 0.74 (m, 1H), 0.71 (d, J ) 6.6
Hz, 3H), 0.67 (d, J ) 6.6 Hz, 3H). Anal. Calcd. for C9H17NO2:
C, 63.13; H, 10.01; N, 8.18. Found: C, 62.52; H, 9.97; N, 8.15.
1
orless oil. H NMR (CDCl3) δ 3.60 (d, J ) 12.0 Hz, 1H), 3.45
(d, J ) 11.7 Hz, 1H), 1.79 (m, 1H), 1.53 (m, 1H), 1.38 (m, 1H),
1.31 (m, 1H), 1.24 (m, 1H), 1.13 (m, 1H), 0.97 (d, J ) 6.8 Hz,
3H), 0.96 (d, J ) 6.6 Hz, 3H). 13C NMR (CDCl3) δ 120.1, 48.7,
39.7, 28.2, 26.0, 22.8, 22.3, 21.9, 19.3.
1-Chloromethyl-2-(1-ethylpropyl)cyclopropanecarbo-
nitrile 48. To a solution of 2-(1-ethyl-propyl)-1-hydroxymeth-
ylcyclopropanecarbonitrile 46 (3.5 g, 21 mmol) in 85 mL of
CH2Cl2 at ambient temperature were added pyridine (5.0 mL,
62 mmol) and then methanesulfonyl chloride (5.0 mL, 64
mmol) quickly dropwise. The mixture was stirred 2 h and then
poured into brine and shaken. The organic phase was sepa-
rated, dried (MgSO4), and concentrated. Flash chromatography
of the residue (10f15% EtOAc/hexanes) gave 0.22 g (6%) of
1-chloromethyl-2-(1-ethyl-propyl)cyclopropanecarbonitrile 48
as a colorless oil. 1H NMR (CDCl3) δ 3.63 (d, J ) 11.7 Hz, 1H),
3.40 (d, J ) 11.7 Hz, 1H), 1.55 (m, 3H), 1.45 (m, 1H), 1.23 (m,
2H), 1.11 (m, 2H), 0.96 (t, J ) 7.6 Hz, 3H), 0.91 (t, J ) 7.3 Hz,
3H).
1-Azidomethyl-2-isobutylcyclopropanecarbonitrile 49.
To a solution of 1-chloromethyl-2-isobutylcyclopropanecarbo-
nitrile 47 (0.67 g, 3.9 mmol) in 10 mL of DMF was added
sodium azide (0.54 g, 8.3 mmol), and the mixture was heated
to 100 °C for 1 h. The mixture was cooled and partitioned
between Et2O and water. The organic phase was separated
and washed with brine, dried (MgSO4), and concentrated to
provide 0.68 g (99%) of 1-azidomethyl-2-isobutylcyclopropane-
carbonitrile 49 as a yellow oil. 1H NMR (CDCl3) δ 3.42 (d, J )
13.2 Hz, 1H), 3.28 (d, J ) 13.2 Hz, 1H), 1.78(m, 1H), 1.55 (m,
1H), 1.37 (m, 1H), 1.22 (m, 1H), 1.16 (m, 1H), 1.03 (m, 1H),
0.97 (d, J ) 6.8 Hz, 3H), 0.96 (d, J ) 6.6 Hz, 3H).
1-Aminomethyl-2-(1-ethylpropyl)cyclopropanecarbox-
ylic Acid 54 [(1-ethylpropyl) and carboxylate groups cis]. To
1-aminomethyl-2-isobutylcyclopropanecarbonitrile 52 (0.56 g,
3.7 mmol) was added 10 mL of 6 M HCl (aq) and the mixture
refluxed for 4 d, cooled, and concentrated. The solid residue
was dissolved in water and washed with Et2O. The aqueous
phase was separated and loaded onto DOWEX-50WX8-100
ion-exchange resin. The column was eluted with 100 mL of
water and then 100 mL of 5% NH4OH (aq). The alkaline
fractions were concentrated to an off-white solid. The solid was
dissolved in the minimum amount of hot MeOH and water
and filtered while hot through a glass wool plug. Concentration
afforded 56 mg (26%) of 1-aminomethyl-2-(1-ethyl-propyl)-
1
cyclopropanecarboxylic acid 54 as a colorless solid. H NMR
(D2O) δ 2.97 (d, J ) 12.9 Hz, 1H), 2.68 (d, J ) 13.2 Hz, 1H),
1.27 (m, 1H), 1.14 (m, 3H), 1.07 (m, 3H), 0.78 (m, 1H), 0.72 (t,
J ) 7.6 Hz, 3H), 0.61 (t, J ) 7.6 Hz, 3H). Anal. Calcd for C10H19-
NO2: C, 64.83; H, 10.34; N, 7.56. Found: C, 63.47; H, 10.32;
N, 7.88.
[3H]-Gabapentin Binding Assay Using Pig Brain Mem-
branes. Pig brains were purchased from Pel-Freez Biologicals
1-Azidomethyl-2-(1-ethylpropyl)cyclopropanecarboni-

3034 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8
Schwarz et al.
(Rogers, AZ). The cerebral cortex was stripped, deveined, and
stored at -80 °C before use. Tissue samples were thawed and
then homogenized on ice using a glass/Teflon homogenizer in
10 volumes (weight/volume) of Buffer A (10 mM HEPES, 1
mM EDTA, 1 mM EGTA, 320 mM sucrose, 100 µM phenyl-
methylsulfonyl fluoride (PMSF), pH 7.4 using KOH). The cell
debris was pelleted by centrifugation (1000g for 10 min) using
a Beckman centrifuge (JA17 rotor). The supernatants were
collected and centrifuged at 40 000g for 20 min at 4 °C. The
membrane pellet was resuspended in 10 volumes of Buffer B
(10 mM HEPES, 1 mM EDTA, 1 mM EGTA, 100 µM PMSF,
pH 7.4 using KOH), stirred on ice for 30 min, and then
centrifuged at 40 000g for 20 min at 4 °C. The pellet was
washed twice in 10 volumes of Buffer B before being resus-
pended in 10 mM HEPES, pH 7.4 and stored at -80 °C. The
membrane protein concentration was determined by the Pierce
bicinchonic acid method using bovine serum albumin (BSA)
as the standard.
The SPA binding assay was performed in Costar 3632 96-
well, clear bottom assay plates using wheat germ agglutinin
coated polyvinyl toluene scintillation proximity assay (SPA)
beads (Amersham Biosciences) using [3H]gabapentin. Pig
cortical membranes (10-20 µg protein per well) and SPA beads
(0.5 mg per well) were mixed with 30 nM [3H]gabapentin
(specific activity ) 44.15 Ci/mmol) in 10 mM HEPES/10 mM
MgSO4 assay buffer, pH 7.4 using KOH. The final well volume
was 200 µL, and nonspecific binding was determined in the
presence of 10 µM unlabeled (cold) pregabalin. The mixture
containing membrane protein with SPA beads and [3H]gaba-
pentin was incubated at room temperature overnight (15-24
h), and plates were then counted on a Wallace Trilux 1450
Microbeta scintillation counter. Curve fitting and IC50 values
were calculated using a four-parameter, nonlinear regression
equation from GraphPad Prism 4.0 software, while Ki values
were determined using previously determined KD values for
pig (KD ) 20 nM) and the equation of Cheng and Prusoff.31
System L Transport Assay. CHO cells were used in this
study since System L transport has been well characterized
in CHO cells.32 CHO-K1 cells were maintained in minimum
essential medium alpha medium (Gibco #32571-036) supple-
mented with 5% fetal bovine serum (Gibco #10082-139) and
1% penicillin/streptomycin (Gibco #15140-122). Cells were
typsinized, diluted and plated into 96-well tissue culture
microplates (Perkin-Elmer, Isoplate TC) the day prior to
running the transport assay. One hundred microliters of a cell
suspension at a density of 3 × 105 cells/mL was added to each
well. This resulted in 90-100% confluent cell monolayers 24
h later. Protein analysis (∼10µg/well) consistently showed e5%
difference between all wells on a plate (data not shown). The
System L transporter assays were run in sodium-free buffer
(PBC) containing (in mM): choline chloride (137), KCl (2.7),
choline phosphate (10.6), KH2PO4 (1.5), d-glucose (5.6), MgCl2
(0.49), and CaCl2 (0.9). A [3H]L-leucine uptake solution consist-
ing of PBC with 2 µCi/mL L-[4,5-3H]leucine (Amersham #TRK
510, 1 mCi/mL, 120-190Ci/mmol) was used. L-Leucine is a
protypical substrate of the System L transporter. Test com-
pound dilutions ranging from 10 mM to 300 nM in half log
increments were made in duplicate for generation of IC50
values. A 96-well drug addition plate was prepared with these
10 drug dilutions in the tritiated L-leucine uptake solution.
The drug plate also had two control wells for each test
compound dilution set, one with no drug added and one with
saturating L-leucine (10 mM) added. The osmolarities of all
the uptake mixtures were held constant by varying the
amounts of choline choride. To start the uptake assay, culture
plates were washed with Na+-free buffer two times for 20 min
each at 37 °C. This step depleted endogenous amino acids and
washed out the residual sodium. The following steps of the
assay were done on an automated Beckman Multimek 96-well
pipettor. One hundred microliters from each of the wells on
the test compound plate were added to corresponding wells
on the culture plate simultaneously. After 2 min at room
temperature, the uptake of [3H]L-leucine was terminated by
washing three times with cold (4 °C) PBC. Preliminary
experiments had indicated that [3H]L-leucine uptake was
linearly dependent on incubation time up to at least 5 min at
22 °C (data not shown). Plate wells were aspirated and then
shaken out completely dry. One hundred and fifty microliters
of scintillation cocktail (Perkin-Elmer-Trisafe) was then added
to each well. This scintillation cocktail lysed the cells, and then
the plates could be counted directly on a liquid scintillation
plate counter (Wallac-Trilux). IC50s were calculated by non-
linear least squares regression analysis (Sigmaplot 2001;
SPSS, Inc, Chicago, Illinois) using a three-parameter Logistic
fit formula: where f ) a/(1 + abs(x/x0)b), a ) max y (dependent
variable), b ) width of transition, (x0 ) x (independent
variable) value at 50% of functions amplitude (IC50), and abs
is the absolute value.
DBA/2 Anticonvulsant Model. All procedures were car-
ried out in compliance with the NIH Guide for the Care and
Use of Laboratory Animals under a protocol approved by the
Pfizer Animal Use Committee. Male DBA/2 mice 3 weeks old
(7-12 g at time of testing) were obtained from Jackson
Laboratories, Bar Harbor, ME. Mice were placed into an
enclosed acrylic plastic chamber (21 cm height, approximately
30 cm diameter) with a high-frequency speaker (4 cm diam-
eter) in the center of the top lid. An audio signal generator
(Protek Model B-810) was used to produce a continuous
sinusoidal tone that was swept linearly in frequency between
8 and 16 kHz once each 10 ms. The average sound pressure
level (SPL) during stimulation was approximately 100 dB at
the floor of the chamber. Mice were placed within the chamber
and allowed to acclimatize for 1 min. DBA/2 mice in the
vehicle-treated group responded to the sound stimulus (applied
until tonic extension occurred, or for a maximum of 60 s) with
a characteristic seizure sequence consisting of wild running
followed by clonic seizures, and later by tonic extension, and
finally by respiratory arrest and death in 80% or more of the
mice. In vehicle-treated mice, the entire sequence of seizures
to respiratory arrest lasted approximately 15 to 20 s. The
incidence of all the seizure phases in the drug-treated and
vehicle-treated mice was recorded, and the occurrence of tonic
seizures was used for calculating anticonvulsant activity. Mice
were used only once for testing, and therefore different groups
were used for each time point and for each dose. Groups of
DBA/2 mice (N ) 10) were tested for sound-induced seizure
responses at previously determined time of peak effect after
oral (PO) or intracerebroventricular (icv) drug administration.
All compounds in the present study were dissolved in distilled
water and given orally by gavage, or by icv injection and given
in a volume of 10 mL/kg of body weight PO, or 5 µL total
volume icv. Vehicle treatment consisted of water, given by oral
gavage, or 0.09% sodium chloride given icv. Immediately before
intracerebroventricular injection DBA/2 mice were restrained
by the investigator before injection of 5 µL volume to the right
lateral ventricle of the animal using a 10 µL Hamilton syringe.
Each animal received only one icv injection before placing in
chamber for anticonvulsant testing.
Acknowledgment. The authors thank Annise Good-
man for the preparation of 1-aminomethylcyclopropane
carboxylic acid 55.
Supporting Information Available: Elemental analyses.
This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) Lauria-Horner, B. A.; Pohl, R. B. Pregabalin: a new anxiolytic.
Expert Opin. Investig. Drugs 2003, 12 (4), 663.
(2) Wheeler, K. Gabapentin (Pfizer). Curr. Opin. Investig. Drugs
2002, 3 (3), 470.
(3) Field, M. J.; Oles, R. J.; Singh, L. Pregabalin may represent a
novel class of anxiolytic agents with a broad spectrum of activity.
Br. J. Pharmacol. 2001, 132, 1-4.
(4) Gee, N. S.; Brown, J. P.; Dissanayake, V. U. K.; Offord, J.;
Thurlow, R.; Woodruff, G. N. The novel anticonvulsant drug,
gabapentin (Neurontin), binds to the R2δ subunit of a calcium
channel. J. Biol. Chem. 1996, 271 (10), 5768.

Cyclopropyl â-Amino Acid Analogues of Pregabalin
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 8 3035
(5) Fink, K.; Dooley, D. J.; Meder, W. P.; Suman-Chauhan, N.; Duffy,
S.; Clusmann, H.; Gothert, M. Inhibition of neuronal Ca2+ influx
by gabapentin and pregabalin in the human neocortex. Neuro-
pharmacology 2002, 42 (2), 229.
(19) Me´lot, J.-M.; Texier-Boullet, F.; Foucaud, A. Cyclopropanation
of electrophilic alkenes with nitroalkanes in the presence of
alumina-supported potassium fluoride. Synthesis 1987, 364.
(20) For other reports describing constrained gabapentin and pre-
gabalin analogues, see: Receveur, J.-M.; Bryans, J. S.; Field,
M. J.; Singh, L.; Horwell, D. C. Synthesis and biological
evaluation of conformationally restricted gabapentin analogues.
Bioorg. Med. Chem. Lett. 1999, 9, 2329, and references therein.
See also ref 20.
(21) Ling, R.; Ekhato, I. V.; Rubin, J. R.; Wustrow, D. J. Synthesis
of 4-alkyl-pyrrolidine-3-carboxylic acid stereoisomers. Tetrahe-
dron 2001, 57 (30), 6579.
(22) Satoh, T.; Suzuki, S.; Suzuki, Y.; Miyaji, Y.; Imai, Z. Reduction
of organic compounds with sodium borohydride-transition metal
salt systems (1). Reduction of organic nitrile, nitro, and amide
compounds to primary amines. Tetrahedron Lett. 1969, (52),
4555.
(23) For the preparation of related cyclopropyl amino acids see:
Brandl, M.; Kozhushkov, S. I.; Yufit, D. S.; Howard, J. A. K.; de
Meijere, A. Cyclopropyl building blocks for organic synthesis,
49. convenient preparation of R-amino acids with bicyclopropyl-
idene and other methylenecyclopropane moieties. Eur. J. Org.
Chem. 1998, (12), 2785, and references therein.
(24) Attempts to fashion the tris-cyclopropyl amino acid failed at the
Knoevenagel condensation stage with the sterically encumbered
dicyclopropyl ketone.
(6) Dooley, D. J.; Mieske, C. A.; Borosky, S. A. Inhibition of K+-
evoked glutamate release from rat neocortical and hippocampal
slices by gabapentin. Neurosci. Lett. 2000, 280 (2), 107.
(7) Dooley, D. J.; Donovan, C. M.; Pugsley, T. A. Stimulus-dependent
modulation of [3H]norepinephrine release from rat neocortical
slices by gabapentin and pregabalin. J. Pharmacol. Exp. Ther.
2000, 295 (3), 1086.
(8) Maneuf, Y. P.; Hughes, J.; McKnight, A. T. Gabapentin inhibits
the substance P-facilitated K+-evoked release of [3H]glutamate
from rat caudal trigeminal nucleus slices. Pain 2001, 93 (2), 191.
(9) McClelland, D.; Evans, R. M.; Barkworth, L.; Martin, D. J.; Scott,
R. H. A study comparing the actions of gabapentin and pre-
gabalin on the electrophysiological properties of cultured DRG
neurones from neonatal rats. BMC Pharmacology 2004, 4 (14).
(10) Bryans, J. S.; Davies, N.; Gee, N. S.; Dissanayake, V. U. K.;
Ratcliffe, G. S.; Horwell, D. C.; Kneen, C. O.; Morrell, A. I.; Oles,
R. J.; O’Toole, J. C.; Perkins, G. M.; Singh, L.; Suman-Chauhan,
N.; O’Neill, J. A. Identification of novel ligands for the gaba-
pentin binding site on the R2δ subunit of a calcium channel and
their evaluation as anticonvulsant agents. J. Med. Chem. 1998,
41, 1838.
(11) Bryans, J. S.; Wustrow, D. J. 3-Substituted GABA analogues
with central nervous system activity: a review. Med. Res. Rev.
1999, 19 (2), 149.
(12) Belliotti, T. R.; Capiris, T.; Ekhato, I. V.; Kinsora, J. J.; Field,
M. J.; Heffner, T. G.; Meltzer, L. T.; Schwarz, J. B.; Taylor, C.
P.; Thorpe, A. J.; Vartanian, M. G.; Wise, L. D.; Su, T.-Z.; Weber,
M. L.; Wustrow, D. J. Structure-activity relationships of pre-
gabalin and analogues that target the R2-δ protein. J. Med.
Chem., in press.
(13) For other efforts to identify R2-δ ligands, see: Stearns, B. A.;
Anker, N.; Arruda, J. M.; Campbell, B. T.; Chen, C.; Cramer,
M.; Hu, T.; Jiang, X.; Park, K.; Ren, K. K.; Sablad, M.; Santini,
A.; Schaffhauser, H.; Urban, M. O.; Munoz, B. Synthesis and
biological evaluation of 6-aryl-6H-pyrrolo[3,4-d]pyridazine de-
rivatives: high affinity ligands to the R2δ subunit of voltage-
gated calcium channels. Bioorg. Med. Chem. Lett. 2004, 14, 1295,
and also refs 13-15.
(14) Lim, J.; Stock, N.; Pracitto, R.; Boueres, J. K.; Munoz, B.;
Chaudhary, A.; Santini, A. M.; Orr, K.; Schaffhauser, H.;
Bezverkov, R. E.; Aiyar, J.; Venkatraman, S. N-Acridin-9-yl-
butane-1,4-diamine derivatives: high affinity ligands to the R2δ
subunit of voltage-gated calcium channels. Bioorg. Med. Chem.
Lett. 2004, 14, 1913.
(15) Hu, T.; Stearns, B. A.; Campbell, B. T.; Arruda, J. M.; Chen, C.;
Aiyar, J.; Bezverkov, R. E.; Santini, A.; Schaffhauser, H.; Liu,
W.; Venkatraman, S.; Munoz, B. Synthesis and biological
evaluation of 6-aryl-6H-pyrrolo[3,4-d]pyridazine derivatives as
high affinity ligands of the R2δ subunit of voltage-gated calcium
channels. Bioorg. Med. Chem. Lett. 2004, 14, 2031.
(16) Lebsack, A. D.; Gunzner, J.; Wang, B.; Pracitto, R.; Schaffhauser,
H.; Santini, A.; Aiyar, J.; Bezverkov, R.; Munoz, B.; Liu, W.;
Venkatraman, S. Identification and synthesis of [1,2,4]triazolo-
[3,4-a]phthalazine derivatives as high-affinity ligands to the
R2δ-1 subunit of voltage gated calcium channel. Bioorg. Med.
Chem. Lett. 2004, 14, 2463.
(17) Babler, J. H.; Spina, K. P. A tandem Michael reaction-cycloalky-
lation utilizing 2-nitropropane: A facile route to the acid
component of insecticidal pyrethroids. Tetrahedron Lett. 1985,
26, 1923.
(18) Ono, N.; Tetsuya, Y.; Isami, H.; Kamimura, A.; Kaji, A. Nitro
compounds as nucleophilic alkylidene transfer reagents. J. Org.
Chem. 1985, 50 (15), 2806. The methyl ester of compound 5 is
disclosed in this paper.
(25) Badorrey, R.; Cativiela, C.; D´ıaz-de-Villegas, M. D.; Ga´lvez, J.
A.; Gil, A. Efficient enantioconvergent synthesis of (S)-R-benzyl-
R-methyl-â-alanine from (R)- and (S)-2-cyano-2-methyl-3-phen-
ylpropanoic acid. Tetrahedron: Asymmetry 2003, 14, 2209-2214.
(26) The direct chlorination of R-methylbenzyl alcohols with meth-
anesulfonyl chloride/pyridine was found to found to proceed in
a solvent-dependent fashion: Tanaka, K.; Katsurada, M.; Ohno,
F.; Shiga, Y.; Oda, M.; Miyagi, M.; Takehara, J.; Okano, K.
Practical Asymmetric Synthesis of (S)-MA20565,
a Wide-
Spectrum Agricultural Fungicide. J. Org. Chem. 2000, 65, 432.
(27) Ohno, M.; Tanaka, H.; Komatsu, M.; Ohshiro, Y. Novel synthesis
of 1-cyanocyclopropane-1-carboxylic acid and its application to
the synthesis of amino acids containing cyclopropane rings.
Synlett 1991, 919.
(28) Su, T.-Z.; Lunney, E.; Campbell, G.; Oxender, D. Transport of
gabapentin, a γ-amino acid drug, by system L R-amino acid
transporters: a comparative study in astrocytes, synaptosomes,
and CHO cells. J. Neurochemistry 1995, 64, 2125-2131.
(29) Schlesinger, K.; Greik, B. J. The genetics and biochemistry of
audiogenic seizures in: Contributions to Behavior-Genetic Analy-
sis: The Mouse as a Prototype Lindzey, G., Theissen, D. D., Eds.
Appleton-Century-Crofts, New York 1970, 219-257; Seyford, T.
N. Audiogenic Seizures in Mice. Fed. Proc. 1979, 38, 2399.
(30) Stevens, C. V.; Van Heecke, G.; Barbero, C.; Patora, K.; De
Kimpe, N.; Verhe, R. Synthesis of substituted cyclopropylphos-
phonates by Michael induced ring closure (MIRC) reactions.
Synlett. 2002, (7), 1089.
(31) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes
50% inhibition (IC50) of an enzymatic reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(32) Shotwell, M. A.; Jayme, D. W.; Kilberg, M. S.; Oxender, D. L.
Neutral amino acid transport systems in Chinese hamster ovary
cells. J. Biol. Chem. 1981, 256 (11), 5422-7.
JM0491086