Diastereoselective, large-scale synthesis of β-amino acids via asymmetric aza-Michael addition as α2δ ligands for the treatment of generalized anxiety disorder and insomnia

Article abstract of DOI:10.1016/j.tetlet.2009.08.119

Scalable synthetic routes to β-amino acids 1 and 2 are presented. These two compounds, which bind to the α2δ subunit of calcium channels and have important medical applications, have been prepared on kilogram scale in our pilot plant through an improved synthesis that avoids the use of highly toxic reagents and hazardous chemistry present in the original Medicinal Chemistry route. The two chiral centers are introduced through asymmetric Michael and aza-Michael reactions with excellent diastereoselectivity.

Full text of DOI:10.1016/j.tetlet.2009.08.119

Tetrahedron Letters 50 (2009) 6325–6328
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Diastereoselective, large-scale synthesis of b-amino acids via asymmetric
aza-Michael addition as a2d ligands for the treatment of generalized
anxiety disorder and insomnia
a
a
b
c
Javier Magano a, , Daniel Bowles , Brian Conway , Thomas N. Nanninga , Derick D. Winkle
*
a Pfizer Global Research & Development, Eastern Point Road, Groton, CT 06340, USA
b Bridge Organics, 311 W Washington, Vicksburg, MI 49097, USA
c Pleotint LLC, 7705 West Olive Road, West Olive, MI 49460, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 July 2009
Revised 27 August 2009
Accepted 28 August 2009
Available online 2 September 2009
Scalable synthetic routes to b-amino acids 1 and 2 are presented. These two compounds, which bind to
the 2d subunit of calcium channels and have important medical applications, have been prepared on
kilogram scale in our pilot plant through an improved synthesis that avoids the use of highly toxic
reagents and hazardous chemistry present in the original Medicinal Chemistry route. The two chiral cen-
ters are introduced through asymmetric Michael and aza-Michael reactions with excellent
diastereoselectivity.
a
Keywords:
b-Amino acid
Ó 2009 Elsevier Ltd. All rights reserved.
a2d Ligand
Asymmetric Michael addition
Asymmetric aza-Michael addition
Generalized anxiety disorder
Insomnia
a
2d-Ligands are compounds that selectively displace 3H-gaba-
presence of LiCl and CuCl provided alkenes 5 and 6, respectively.4
The double bond was then oxidized with Jones reagent to give
acids 7 and 8.5 The rest of the synthesis is illustrated through the
use of acid 7 as a representative example. Thus, 7 underwent reac-
tion with pivaloyl chloride (9) in the presence of Et3N to give mixed
anhydride 10, which was treated with oxazolidinone 11 to gener-
ate intermediate 12.6 The deprotonation of 12 with NaHMDS fol-
lowed by the addition of tert-butyl bromoacetate afforded tert-
butyl ester 13 in good yield and excellent diastereoselectivity.7
The cleavage of the oxazolidinone with LiOH and H2O2 gave acid
14 as a yellow viscous oil,8 which was used directly in the next
step. At this point, the chiral auxiliary could be recovered in nearly
quantitative yield. The key Curtius rearrangement was accom-
plished by heating acid 14 with diphenylphosphoryl azide (DPPA)9
and triethylamine in tert-butyl methyl ether at 45 °C to give crude
isocyanate 15, which was immediately treated with 6 N HCl at re-
flux to afford b-amino acid 1 as its HCl salt. Homologue 2 was pre-
pared in a similar fashion from carboxylic acid 8.
pentin from brain membranes, indicating a high affinity interaction
with the a
2d subunit of voltage-gated calcium channels.1 This type
of compound can be used to treat a number of conditions such as
generalized anxiety disorder (GAD), insomnia, fibromyalgia, epi-
lepsy, neuropatic pain, anxiety, depression, and attention hyperac-
tivity disorder, among others.2 Recently, we have been interested
in the synthesis of compounds 1 and 2 as candidates for the treat-
ment of generalized anxiety disorder and insomnia, respectively.
With the goal of preparing large quantities of these materials for
clinical trials, we have developed two different and efficient ap-
proaches: the ‘Michael addition route’, described in this Letter,
and the ‘Enamide hydrogenation route’ described in the following
article.
The medicinal chemistry route for the preparation of these
compounds is shown in Scheme 1.3 The synthesis started with
(R)-b-citronellol (3), which already possesses one of the chiral cen-
ters of the target molecule in the desired configuration. Mesylation
under standard conditions followed by displacement of the leaving
group in 4 with either methyl or ethylmagnesium bromide in the
This route proved satisfactory to provide small amounts of
material for preliminary studies but several issues were identified
that did not make it amenable for scale up, such as the cost of (R)-
b-citronellol, the highly toxic waste generated by the Jones oxida-
tion, and the use of hydrogen peroxide and DPPA, both of which are
* Corresponding author. Tel.: +1 860 6869021.
E-mail addresses: Javier.Magano@Pfizer.com, jmaganop@yahoo.com (J. Magano).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.08.119

6326
J. Magano et al. / Tetrahedron Letters 50 (2009) 6325–6328
hazardous reagents. As a result of these limitations, we developed
an alternate route toward 1 and 2 that circumvented these prob-
lems and enabled a more process-friendly synthesis. Even though
this new route is essentially the same for both 1 and 2, it employs
different chiral auxiliaries for each substrate as well as alternative
protocols for the removal of the auxiliary to produce an intermedi-
ate b-ketoester.
The preparation of b-ketoesters 22 and 24 is described in
Scheme 2. Oxazolidinones 11 and 16, derived from (1S,2R)-(+)-nor-
ephedrine and L-(+)-a-phenylglycine, respectively, were deproto-
(17) to afford intermediates 18 and 19.10 This direct method was
preferred to the mixed anhydride methodology described in the
original synthesis (Scheme 1), since it is operationally simpler
and generates a much smaller amount of waste. Without isolation,
N-acyl oxazolidinones 18 and 19 were treated with the cuprates
derived from n-propyl and n-butylmagnesium bromide, respec-
tively, and CuBrꢀMe2S to give Michael addition products 20 and
21 in excellent diastereoselectivity after recrystallization from
hexane.11 Although the level of conversion was not affected, the le-
vel of enantioselectivity was improved by utilizing multiple equiv-
alents of the cuprate reagent (97% ee using 3.1 equiv vs 92% ee
nated with n-BuLi followed by acylation with crotonyl chloride
RMgBr, LiCl, CuCl
MsCl, Et3N
OH
R
CH2Cl2, 0 °C
94%
OMs
THF, 0 °C
R = Me, Et
4
3
CrO3, H2SO4
Acetone, 20 °C
59% (2 steps)
R
HO2C
R = Me (7)
R = Et (8)
R = Me (5)
6
R = Et ( )
O
HN
Me
O
O
11
O
O
Cl
9
Ph
Et3N
O
O
N
+
7
THF, 0 oC
LiCl, THF, 0 °C
79%
O
O
Me
Ph
12
10
O
O
O
1) NaHMDS, THF,
-50 °C
LiOH, H2O2
O
OH
CO2t-Bu
N
THF/H2O, 20 °C
Ph
93%
2) BrCH2CO2t-Bu,
-50 °C to 20 °C
t-BuO2C
Me
14
13
74%, 95% de
N=C=O
(PhO)2PON3
6 N HCl
CO2H
HCl
Et3N, MTBE, 45 °C
100%
60 °C
71%
NH2
CO2t-Bu
15
1
Scheme 1. Medicinal chemistry route for the preparation of
a2d ligands 1 and 2.
O
O
O
O
O
1. n-BuLi, THF, -55 °C
2.
R3MgBr, CuBr•DMS
TMEDA, THF, -55 °C
O
NH
R2
O
N
O
N
R3
O
17
R1
R1
R2
20
: 41% (2 steps)
R1
R2
Cl
21: 70% (2 steps)
11
16
(R1 = Ph, R2 = Me)
(R1 = H, R2 = Ph)
20
(R1 = Ph, R2 = Me, R3 = n-propyl)
18
(R1 = Ph, R2 = Me)
21 (R1 = H, R2 = Ph, R3 = n-butyl)
19 (R1 = H, R2 = Ph)
d.r: 99.8/0.2
1. KO2CCH2CO2Et, MgCl2,
MeCN, Et3N, 80 °C
O
O
20
+
11
2. Aq. HCl
EtO
50% recovered
83% (2 steps)
22
1. CDI, THF
2. KO2CCH2CO2Et, MgCl2
1. 30% H2O2, LiOH,
H2O/THF
O
O
O
21
HO
EtO
2. NaHSO3
3. Aq. HCl
23
+
24
80% (2 steps)
16
85% recovered
Scheme 2. Scalable syntheses of b-ketoesters 22 and 24 en route to a2d ligands 1 and 2.

J. Magano et al. / Tetrahedron Letters 50 (2009) 6325–6328
6327
using 1.6 equiv). Running the reaction at higher concentration also
improved the selectivity but extremely thick slurries were
obtained.
90
80
70
60
50
40
30
20
10
0
3.5
3
Different conditions were employed for the removal of the chi-
ral auxiliaries. Thus, the oxazolidinone moiety in 21 was cleaved
via the original oxidative protocol with H2O2 and LiOH12 to give
acid 23 which, after activation with 1,10-carbonyldiimidazole,
underwent reaction with potassium ethyl malonate (KEM) in the
presence of MgCl2 to afford b-keto ester 24. Alternatively, the di-
rect displacement of the chiral auxiliary in 20 was carried out in
one step with KEM and MgCl2 to provide b-keto ester 22 in similar
yield. This methodology, which was developed in our laborato-
ries,13 considerably simplifies this transformation with respect to
the traditional approach. At this point, the chiral auxiliaries 11
and 16 could be recovered in fair and good yield, respectively, even
though this aspect of the synthesis was not optimized. Prior to the
hydrogenation step, the purity of keto esters 22 and 24 was up-
graded via treatment with Darco G-60 in ethanol at 50 °C for 3 h.
With b-keto esters 22 and 24 in hand, the reduction of the keto
group was then investigated (Scheme 3). Several hydrogenation
screens were completed in an Argonaut Endeavor Parallel Reactor
using (Ph3P)3RuCl2 as a catalyst14 to determine the influence of
pressure, temperature, catalyst loading, concentration, and the
choice of acid on the process. The data showed that catalyst load-
ing and temperature had the biggest impact on the reaction rate,
while pressure had a lesser influence on the outcome. Also,
H2SO4 was shown to be a viable alternative to HCl since this trans-
formation had to be carried out in stainless steel equipment. The
most effective conditions that emerged from the screen were
0.13 mol % of both Ru catalyst and H2SO4 under 50 psi of hydrogen
in ethanol at 60 °C, which gave alcohols 25 and 26 in excellent
yield. The NaBH4 reduction in methanol was also investigated,
but the product was contaminated with boron by-products, which
made the purification difficult.
2.5
2
R
R
1.5
1
Yield, %
(5R, 3R), %
0.5
0
0
2
4
6
8
10
12
14
Temperature C
Figure 1. Diastereomeric purity and yield of 2 as a function of the crystallization
temperature.
palladium hydroxide on carbon and acetic acid as co-catalyst to
accelerate the rate of reaction. Unfortunately, a heavy loading of
the Pd catalyst (1 g/2.5 g of substrate) was found to be necessary
to push the reaction to completion and to prevent the isolation
of partially debenzylated intermediates. Finally, the ester group
was hydrolyzed in hot, concentrated HCl to afford crudes 1 and
2. After an extensive solvent screen, it was found that crystalliza-
tion from the acid reaction mixture was most effective in affording
the desired b-amino acids 1 and 2 in 66% and 40% yield, respec-
tively, and de P99%, with residual Pd and Rh contents below our
specification of 620 ppm. In addition, it was found that slow cool-
ing helped reduce the amount of the major impurity identified as
the (3R,5R)-diastereoisomer. The influence of the crystallization
temperature in the diastereomeric purity and the yield of 2 is
shown in Figure 1. Compounds 1 and 2 can be further recrystal-
lized from 2-propanol to upgrade the optical purity to P99.8% de.
Subsequent mesylate formation and elimination in the presence
of Et3N gave a,b-unsaturated esters 27 and 28. This transformation
was originally carried out in acetonitrile, but ethyl acetate proved a
greener alternative. In addition, it allowed for a more straightfor-
ward work-up after aqueous quench since, when acetonitrile was
employed, it was necessary to distill most of the solvent to obtain
good phase separation.
One of the key steps in this route is the introduction of the sec-
ond chiral center. Based on the work of Davies et al.,15 chiral amine
29 was first deprotonated with n-BuLi at ꢁ65 °C and the resulting
anion was added to acrylates 27 and 28 to afford b-amino esters 30
and 31, respectively, in excellent yield and diastereoselectivity.
Good conversion was obtained using 1.3 equiv of amine 29. In
addition, quenching the reaction with dilute HCl gave a cleaner
product profile in comparison with concentrated HCl or acetic acid.
The debenzylation of intermediates 30 and 31 was carried out with
In conclusion, a scalable route for the preparation of a2d ligands
1 and 216 has been described that avoids the limitations encoun-
tered in the original synthesis. The feasibility of this technology
has been demonstrated in our pilot plant on multi-kilogram scale,
making possible the timely advance of associated drug develop-
ment programs.
Acknowledgments
The authors thank Tom Mulhern, Jim Zeller, Dan Belmont, Denis
Sobieray, Margaret Evans and Sandra Jennings for helpful discus-
sions during the implementation of this project and Mark Maloney
and Phil Nixon for reviewing this Letter.
O
OH
O
H2, (Ph3P)3RuCl2
1. MsCl, EtOAc, -25 °C
2. Et3N, -20 to 60 °C
22 24
or
EtO
25
R3
Ph
EtO
R3
EtOH, H2SO4,
60 °C, 50 psi
(R3 = n-propyl)
26 (R3 = n-butyl)
27 (R3 = n-propyl): 77% (3 steps)
28 (R3 = n-butyl): 69% (3 steps)
Ph
1. H2, Pd(OH)2/C, EtOH,
HOAc, 50 psi, 45 °C
1. n-BuLi, hexanes,
O
NH2
Ph
Ph
THF, -65 °C
O
N
HO
R3
NH
29
27
28
or , -75 °C
2.
2. 12 M HCl, 85 °C
HCl
EtO
R3
1
2
(R3 = n-propyl): 66% (2 steps); de >99%
(R3 = n-butyl): 40% (2 steps); de >99%
30
31
(R3 = n-propyl): 92%; de: 93%
(R3 = n-butyl): 98%; de: 96%
Scheme 3. Scalable synthesis for the preparation of a2d ligands 1 and 2. Completion of the synthesis.

6328
J. Magano et al. / Tetrahedron Letters 50 (2009) 6325–6328
L. Org. Process Res. Dev. 1997, 1, 26; (b) Shioiri, T.; Hayashi, K.; Hamada, Y.
References and notes
Tetrahedron 1993, 49, 1913; (c) Fadel, A.; Salaun, J. Tetrahedron Lett. 1988, 29,
6257.
13. Magano, J.; Nanninga, T. N.; Winkle, D. D. Tetrahedron Lett. 2008, 49, 2956.
14. (a) Sánchez-Delgado, R. A.; de Ochoa, O. L. J. Organometal. Chem. 1980, 202, 427;
(b) Zsigmond, Á.; Balatoni, I.; Bogár, K.; Notheisz, F.; Joó, F. J. Catal. 2004, 227,
428.
1. Ohashi, K.; Kawai, M.; Ninomiya, N.; Taylor, C.; Kurebayashi, Y. Pharmacology
2008, 81, 144.
2. Rawson, D. J.; Schwarz, J. B. WO 2007/052134 A1 20070510. CAN 146:482239.
3. Barta, N. S.; Schwarz, J. B.; Thorpe, A. J.; Wustrow, D. J. WO 2003082807 A2
20031009. CAN 139:308005.
4. Nunomoto, S.; Kawakami, Y.; Yanashita, Y. J. Org. Chem. 1983, 48, 1912.
5. Davis, M. A.; Hickinbottom, W. J. J. Chem. Soc. 1958, 2205.
6. Ho, G.-J.; Mathre, D. J. J. Org. Chem. 1995, 60, 2271.
7. (a) Evans, D. A.; Wu, L. D.; Wiener, J. J. M.; Johnson, J. S.; Ripin, D. H. B.; Tedrow,
J. S. J. Org. Chem. 1999, 64, 6411; (b) Sibi, M. P.; Deshpande, P. K. J. Chem. Soc.,
Perkin Trans. 1 2000, 1461.
15. (a) Davies, S. G.; Ichihara, O. Tetrahedron: Asymmetry 1991, 2, 183; (b) Bull, S.
D.; Davies, S. G.; Smith, A. D. J. Chem. Soc., Perkin Trans. 1 2001, 2931; (c) Davies,
S. G.; Smith, A. D.; Price, P. D. Tetrahedron: Asymmetry 2005, 16, 2833.
16. Analytical data for 1: 1H NMR (400 MHz, methanol-d4) d ppm 0.93–1.00 (m,
6H), 1.17–1.26 (m, 1H), 1.32–1.52 (m, 4H), 1.61–1.77 (m, 2H), 2.63–2.78 (m,
2H), 3.61–3.68 (m, 1H), 4.99 (br s, 3H). 13C NMR (100 MHz, methanol-d4) d ppm
13.47, 18.42, 19.67, 28.69, 36.55, 39.08, 40.14, 46.65, 172.44. MS (APCI+): m/z
174 (M+H). Anal. Calcd for C9H19NO2ꢀHCl: C, 51.54; H, 9.61; N, 6.68. Found: C,
8. (a) Evans, D. A.; Britton, T. C.; Ellman, J. A. Tetrahedron Lett. 1987, 28, 6141; (b)
Yuen, P.-W.; Kanter, G.; Taylor, C. P.; Vartanian, M. G. Bioorg. Med. Chem. Lett.
1994, 4, 823.
51.91; H, 9.41; N, 6.80. ½a D22
ꢂ
(14.35, c 0.64, MeOH).
Analytical data for 2: 1H NMR (methanol-d4) d ppm 0.90–1.03 (m, 6H), 1.17–
1.47 (m, 7H), 1.56–1.72 (m, 2H), 2.23–2.42 (m, 1H), 2.42–2.59 (m, 1H), 3.38–
3.58 (m, 1H), 5.13 (bs, 3H). 13C NMR (methanol-d4) d ppm 13.33, 18.76, 22.84,
28.91, 28.94, 36.42, 38.78, 40.58, 47.68, 176.83. MS (APCI+): m/z 188 (M+H).
Anal. Calcd for C10H21NO2ꢀHCl: C, 53.68; H, 9.91; N, 6.26. Found: C, 53.30; H,
9. Arvanitis, E.; Ernst, H.; Ludwig, A. A.; Robinson, A. J.; Wyatt, P. B. J. Chem. Soc.,
Perkin Trans. 1 1998, 521.
10. (a) Evans, D. A.; Chapman, K. T.; Bisaha, J. J. Am. Chem. Soc. 1984, 106, 4261; (b)
Evans, D. A.; Chapman, K. T.; Bisaha, J. J. Am. Chem. Soc. 1988, 110, 1238.
11. Williams, D. R.; Kissel, W. S.; Li, J. J. Tetrahedron Lett. 1998, 39, 8593.
12. (a) Hoekstra, M. S.; Sobieray, D. M.; Schwindt, M. A.; Mulhern, T. A.; Grote, T.
M.; Huckabee, B. K.; Hendrickson, V. S.; Franklin, L. C.; Granger, E. J.; Karrick, G.
9.69; N, 6.23. ½a D22
ꢂ
(30.73, c 1.0, MeOH).