Practical access to the proline analogs (S,S,S)- and (R,R,R)-2- methyloctahydroindole-2-carboxylic acids by HPLC enantioseparation

Article abstract of DOI:10.1002/chir.20952

An efficient methodology for the preparation of the α- tetrasubstituted proline analog (S,S,S)-2-methyloctahydroindole-2-carboxylic acid, (S,S,S)-(αMe)Oic, and its enantiomer, (R,R,R)-(αMe)Oic, has been developed. Starting from easily available substrates and through simple transformations, a racemic precursor has been synthesized in excellent yield and further subjected to HPLC resolution using a cellulose-derived chiral stationary phase. Specifically, a semipreparative (250 mm × 20 mm ID) Chiralpak IC column has allowed the efficient resolution of more than 4 g of racemate using a mixture of n-hexane/tert-butyl methyl ether/2-propanol as the eluent. Multigram quantities of the target amino acids have been isolated in enantiomerically pure form and suitably protected for incorporation into peptides.

Full text of DOI:10.1002/chir.20952

CHIRALITY 23:507–513 (2011)
Practical Access to the Proline Analogs (S,S,S)- and
(R,R,R)-2-Methyloctahydroindole-2-carboxylic
Acids by HPLC Enantioseparation
´
´
*
FRANCISCO J. SAYAGO, MARIA J. PUEYO, M. ISABEL CALAZA, ANA I. JIMENEZ, AND CARLOS CATIVIELA
Departamento de Quı´mica Orga´nica, Instituto de Ciencia de Materiales de Arago´n, Universidad de Zaragoza–CSIC, 50009 Zaragoza, Spain
ABSTRACT
An efficient methodology for the preparation of the a-tetrasubstituted proline
analog (S,S,S)-2-methyloctahydroindole-2-carboxylic acid, (S,S,S)-(aMe)Oic, and its enantiomer,
(R,R,R)-(aMe)Oic, has been developed. Starting from easily available substrates and through simple
transformations, a racemic precursor has been synthesized in excellent yield and further subjected
to HPLC resolution using a cellulose-derived chiral stationary phase. Specifically, a semipreparative
(250 mm 3 20 mm ID) Chiralpak1 IC column has allowed the efficient resolution of more than 4 g
of racemate using a mixture of n-hexane/tert-butyl methyl ether/2-propanol as the eluent. Multi-
gram quantities of the target amino acids have been isolated in enantiomerically pure form and suit-
C
VV
ably protected for incorporation into peptides. Chirality 23:507–513, 2011.
2011 Wiley-Liss, Inc.
KEY WORDS: a-tetrasubstituted amino acid; perhydroindole-2-carboxylic acid; polysaccharide-
derived chiral stationary phase; chiral HPLC; HPLC resolution; Chiralpak IC
INTRODUCTION
As the a-methylated counterpart of Oic, 2-methyloctahy-
droindole-2-carboxylic acid [(aMe)Oic, Fig. 1)] is of high
intrinsic value as a potential modulator of the conformational
constraints induced by proline. Both Oic and (aMe)Oic bear
an additional cyclohexane ring with reference to the parent
amino acid, proline. This structural feature is appropriate to
build up hydrophobic recognition interactions at the binding
site of receptors and to endow peptides with greater lipophi-
licity, which, in turn, may translate into enhanced affinity and
bioavailability. Moreover, the stereochemical diversity of Oic
and (aMe)Oic greatly increases their potential for the design
of peptides. It should be noted that, for a given configuration
at the a carbon, four different dispositions of the fused cyclo-
hexane ring are possible. The variety of shapes provided by
the different Oic and (aMe)Oic stereoisomers may be of
help to fine-tune the interaction with the hydrophobic pocket
of a receptor and to elucidate the bioactive conformations of
peptides.
In spite of its great potential value, the behavior of
(aMe)Oic when included into peptides or other biologically
relevant systems remains unexplored. This is due to the lack
of synthetic methodologies providing access to the different
stereoisomers of (aMe)Oic in enantiomerically pure form. In
fact, at variance with Oic, attention on the a-methylated coun-
terpart (aMe)Oic has been focused only recently. Within
this context, we have described the preparation of enantio-
pure (S,S,S)-(aMe)Oic13 and its epimer at the a carbon
(R,S,S)-(aMe)Oic14 by means of stereoselective alkylation
Peptides are mediators of key biological functions in our or-
ganism and, thereby, provide a great opportunity for therapeu-
tic applications. However, the use of natural peptides as drugs
is strongly limited by their rapid metabolism, low bioavailability,
and poor receptor subtype selectivity.1 Therefore, much effort
is being devoted to endow peptides with enhanced therapeutic
properties. In this context, the incorporation of constrained a-
amino acids into peptides constitutes a very useful strategy to
reduce their flexibility and retard enzymatic degradation.1,2
Moreover, restricted a-amino acids stabilize particular confor-
mational features, which may lead to the improvements in the
biological potency if the bioactive conformation is tethered.1,2
In particular, a-methylation of a-amino acids is an efficient
way to produce surrogates with defined conformational pref-
erences.3 Indeed, the incorporation of a-methyl amino acids
into peptides has shown to reduce their conformational
freedom and to prevent the hydrolytic cleavage of neighbor-
ing peptide bonds.1,2,4 Among them, a-methylproline5
[(aMe)Pro, Fig. 1)] has frequently been used as a replace-
ment for proline to control the secondary structure and,
hence, the biological behavior of peptides.6–8 The great num-
ber of patents dealing with bioactive peptides containing
(aMe)Pro provide evidence for the enormous value of this
amino acid (in general, of a-methylated amino acids) in the
development of therapeutic agents.
Octahydroindole-2-carboxylic acid (Oic, Fig. 1) is a proline
analog containing a fused cyclohexane ring that has proven
extremely useful to optimize the pharmacological profile of
bioactive peptides. Particularly notable has been the use of
(S,S,S)-Oic for the design of angiotensin-converting enzyme
(ACE) inhibitors,9 antagonists for the B2 receptor of bradyki-
nin,10,11 and prolyl oligopeptidase (POP) inhibitors.12 Such
(S,S,S)-Oic-containing compounds are useful, among other
applications, as antihypertensive drugs, anti-inflammatory
agents, for the prevention of heart failure, or the treatment of
neurodegenerative diseases and, in fact, a number of them
are already on the market or have entered clinical evaluation.
Contract grant sponsors: Ministerio de Ciencia
e Innovacio´n–FEDER,
Gobierno de Arago´n; Contract grant numbers: CTQ2007-62245, CTQ2010-
17436, PIP206/2005 and research group E40
*Correspondence to: C. Cativiela, Departamento de Qu´ımica Orga´nica,
ICMA, Universidad de Zaragoza–CSIC, 50009 Zaragoza, Spain.
E-mail: cativiela@unizar.es
Received for publication 6 December 2010; Accepted 4 February 2011
DOI: 10.1002/chir.20952
Published online 15 April 2011 in Wiley Online Library
(wileyonlinelibrary.com).
C
VV
2011 Wiley-Liss, Inc.

508
SAYAGO ET AL.
X-Ray Diffraction
Colorless single crystals of racemic (S*,S*,S*)-10 were obtained by
slow evaporation from a dichloromethane/ethyl acetate solution. The
X-ray diffraction data were collected at 100 K on an Oxford Diffraction
Xcalibur diffractometer provided with a Sapphire CCD detector, using
˚
Fig. 1. Structure of different proline analogs: a-methylproline
graphite-monochromated Mo-Ka radiation (k 5 0.71073 A). The struc-
ture was solved by direct methods using SHELXS-97,16 and refinement
was performed using SHELXL-9717 by the full-matrix least-squares tech-
nique with anisotropic thermal factors for heavy atoms. Hydrogen atoms
were located by calculation and affected by an isotropic thermal factor
fixed to 1.2 times the Ueq of the carrier atom (1.5 for the methyl pro-
tons). Crystallographic data (excluding structure factors) for the struc-
ture reported in this article have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication number
CCDC 746151. Copies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK [fax:
144(0)-1223-336033 or email: deposit@ccdc.cam.ac.uk].
[(aMe)Pro], octahydroindole-2-carboxylic acid (Oic), and 2-methyloctahy-
droindole-2-carboxylic acid [(aMe)Oic]. Chiral carbons in the octahydroin-
dole system (corresponding to positions 2, 3a, and 7a) are indicated with
asterisks. All throughout the text, a nomenclature without numerical locants
will be used for the sake of simplicity; accordingly, a (2S,3aS,7aS) stereo-
chemistry will be denoted as (S,S,S).
processes. To the best of our knowledge, these are, together
with the preparation of (S,S,S)-(aMe)Oic described in a pat-
ent,15 the only synthetic routes to (aMe)Oic reported to
date. Therefore, development of efficient methodologies that
ensure accessibility to these and other (aMe)Oic stereoiso-
mers, in optically pure form and significant quantities,
remains a challenge. We describe here a convenient route
for the multigram-scale preparation of both (S,S,S)-(aMe)Oic
and (R,R,R)-(aMe)Oic in enantiomerically pure form. The
procedure combines the synthesis of a racemic precursor
and a chromatographic resolution procedure.
Crystallographic data: orthorhombic, space group Pbca; a 5 7.5997(2)
A, b 5 17.0955(6) A, c 5 18.1963(6) A; Z 5 8; dcalcd 5 1.313 g cm–3
;
˚
˚
˚
18,156 reflections collected, 3133 unique (Rint 5 0.0401); data/parame-
ters: 3133/137; final R indices (I > 2rI): R1 5 0.036, wR2 5 0.089; final
R indices (all data): R1 5 0.062, wR2 5 0.102; highest residual electron
–3
˚
density: 0.39 e A
.
Syntheses
Synthesis of methyl N-(tert-butoxycarbonyl)indole-2-carboxylate
(9). To a solution of indole-2-carboxylic acid (8) (4.0 g, 24.82 mmol) in
methanol (135 ml), sulfuric acid (2.30 ml, 40.77 mmol) was added dropwise.
The reaction mixture was heated under reflux for 6 h. The solvent was
evaporated, and the residue was dissolved in dichloromethane (70 ml) and
washed with saturated aqueous NaHCO3. The organic layer was dried over
MgSO4 and filtered. After evaporation of the solvent, the remaining residue
was dissolved in tetrahydrofuran (100 ml) and 4-(dimethylamino)pyridine
(361 mg, 2.93 mmol) and di-tert-butyl dicarbonate (6.40 g, 29.32 mmol)
were added. The reaction mixture was kept at room temperature for 2 h,
and then the solvent was evaporated. The resulting residue was dissolved
in ethyl acetate (70 ml) and washed with saturated aqueous NaHCO3. The
organic phase was dried over MgSO4 and filtered. The crude product
obtained after evaporation of the solvent was purified by column chromatog-
raphy (eluent: hexanes/ethyl acetate 10/1) to give pure 9 as a white solid
(6.36 g, 23.10 mmol, 93% yield). Mp 68–708C. IR (KBr) m 3019, 2981, 1745,
MATERIALS AND METHODS
General
All reagents from commercial suppliers were used without further pu-
rification. Thin-layer chromatography was performed on Macherey-
Nagel Polygram syl G/UV-precoated silica gel polyester plates. The
products were visualized by exposure to UV light (254 nm), iodine
vapor, or submersion in cerium molybdate stain [aqueous solution of
phosphomolybdic acid (2%), CeSO4ꢀ4H2O (1%), and H2SO4 (6%)]. Col-
˚
umn chromatography was performed using Macherey-Nagel 60-A silica
gel. Melting points were determined on a Gallenkamp apparatus and are
uncorrected. IR spectra were registered on a Mattson Genesis or a Nico-
let Avatar 360 FTIR spectrophotometer; mmax is given for the main
absorption bands. 1H and 13C NMR spectra were recorded on a Bruker
AV-400 instrument at room temperature using the residual solvent signal
as the internal standard; chemical shifts (d) are expressed in ppm and
coupling constants (J) in Hertz. Optical rotations were measured at room
temperature using a JASCO P-1020 polarimeter. High-resolution mass
spectra were obtained on a Bruker Microtof-Q spectrometer.
1
1712, 1556 cm–1. H NMR (CDCl3, 400 MHz) d 1.78 (s, 9H), 4.08 (s, 3H),
7.26 (m, 1H), 7.42 (m, 1H), 7.57 (m, 1H), 7.76 (m, 1H), 8.25 (m, 1H). 13C
NMR (CDCl3, 100 MHz) d 27.78, 52.28, 84.55, 114.82, 122.12, 123.25,
126.77, 127.44, 130.34, 137.79, 149.20, 162.31. HRMS (ESI) C15H17NNaO4
[M1Na]1: calcd 298.1050, found 298.1039.
High-Performance Liquid Chromatography
Synthesis of methyl (2S*,3aS*,7aS*)-N-(tert-butoxycarbonyl)octa-
hydroindole-2-carboxylate [(S*,S*,S*)-4]. A solution of 9 (6.0 g,
21.79 mmol) in acetic acid (80 ml) was hydrogenated at room tempera-
ture and atmospheric pressure in the presence of PtO2 (600 mg). After
24 h, the catalyst was filtered off and washed with acetic acid. The sol-
vent was evaporated to dryness, and the residue was dissolved in dichloro-
methane (70 ml) and washed with saturated aqueous NaHCO3. The or-
ganic layer was dried over MgSO4 and filtered. The solvent was evapo-
rated, and the residue was purified by column chromatography (eluent:
hexanes/ethyl acetate 10/1) to afford pure (S*,S*,S*)-4 as a colorless oil
(5.87 g, 20.72 mmol, 95% yield). HRMS (ESI) C15H25NNaO4 [M1Na]1:
calcd 306.1676, found 306.1679. Spectroscopic data are identical to those
reported previously13 for the (S,S,S) enantiomer.
HPLC was carried out using a Waters 600 HPLC system equipped
with a 2996 photodiode array detector and a 2487 dual wavelength ab-
sorbance detector (respectively used for monitoring analytical and prepa-
rative separations). The solvents used as mobile phases were of spectro-
scopic grade. Analytical assays were performed on Chiralpak1 IA, IB,
and IC columns (Daicel Chemical Industries, Japan) of 250 mm 3
4.6 mm ID using different binary and ternary mixtures as eluents and
working at flow rates ranging from 0.7 to 1.0 ml/min. The preparative re-
solution was carried out on a 250 mm 3 20 mm ID Chiralpak1 IC col-
umn eluting with n-hexane/tert-butyl methyl ether/2-propanol 50/44/6
(see below for further details). Both the column loading capacity (i.e.,
the maximum sample mass that the column can hold) and the optimum
sample concentration had previously been determined on the analytical
250 mm 3 4.6 mm ID Chiralpak1 IC column by injecting increasing
amounts of sample at different concentrations. The capacity (k0), selectiv-
ity (a), and resolution (RS) factors are defined as follows: k0 5 (tr – t0)/
t0, a 5 k02/k01, RS 5 1.18 (t2 – t1)/(w2 1 w1), where subscripts 1 and 2
refer to the first and second eluted enantiomers, tr (r 5 1, 2) are their
retention times, and w1 and w2 denote their half-height peak widths; t0 is
the dead time.
Synthesis of methyl (2S*,3aS*,7aS*)-N-(tert-butoxycarbonyl)-2-
methyloctahydroindole-2-carboxylate [(S*,S*,S*)-5]. A 1.8 M solu-
tion of LDA in hexanes (26.20 ml, 47.16 mmol) was added dropwise to a
stirred solution of (S*,S*,S*)-4 (5.70 g, 20.11 mmol) in anhydrous tetra-
hydrofuran (105 ml) kept under an inert atmosphere, at –788C. After stir-
ring at this temperature for 30 min, methyl iodide (5.20 ml, 81.35 mmol)
was added dropwise. The resulting solution was warmed to –508C and kept
Chirality DOI 10.1002/chir

HPLC RESOLUTION OF (S*,S*,S*)-(ꢀME)OIC
509
at this temperature overnight. The reaction mixture was quenched with sat-
urated aqueous NH4Cl (20 ml) and extracted with dichloromethane several
times. The organic phase was dried over anhydrous MgSO4 and filtered.
The solvent was eliminated, and the residue was purified by column chro-
matography (eluent: hexanes/ethyl acetate 10/1) to afford pure (S*,S*,S*)-
5 as a colorless oil (4.48 g, 15.06 mmol, 75% yield). IR (neat) m 1743, 1693
cm–1. 1H NMR (DMSO-d6, 400 MHz) d 1.06 (m, 1H), 1.14–1.50 (m, 3H)
overlapped with 1.30, 1.36 (two s, 9H) and with 1.43, 1.45 (two s, 3H), 1.54–
1.68 (m, 4H), 1.94–2.10 (m, 1H), 2.11–2.25 (m, 1H), 2.40–2.48 (m, 1H),
3.60–3.72 (m, 1H) overlapped with 3.60, 3.65 (two s, 3H). 13C NMR
(DMSO-d6, 100 MHz) d (duplicate signals are observed for most carbons)
20.10, 20.21; 21.77, 23.13; 23.00, 23.23; 25.03, 25.11; 25.65, 26.46; 27.85,
28.01; 33.62, 34.54; 38.87, 39.97; 51.86, 52.04; 57.52, 57.68; 64.40; 78.41,
78.52; 151.66, 152.32; 174.37, 174.66. HRMS (ESI) C16H27NNaO4
[M1Na]1: calcd 320.1832, found 320.1835.
aqueous KHSO4 and extracted with ethyl acetate (3 3 50 ml). The com-
bined organic layers were dried over anhydrous MgSO4 and filtered.
Evaporation of the solvent gave (S,S,S)-12 as a white foam (1.94 g, 6.11
mmol, 99% yield). [a]D 111.8 (c 0.81, MeOH). IR (KBr) m 3500–2480,
3064, 2931, 1743, 1697 cm–1. 1H NMR (CDCl3, 400 MHz) d 1.03–1.35 (m,
2H), 1.44–1.78 (m, 6H) overlapped with 1.53, 1.62 (two s, 3H), 2.05, 2.26
(two m, 1H), 2.40–2.55 (m, 2H), 3.88, 3.95 (two m, 1H), 5.05–5.14 (m,
2H), 7.22–7.38 (m, 5H), 9.48 (bs, 1H). 13C NMR (CDCl3, 100 MHz) d
(duplicate signals are observed for most carbons) 20.37, 20.59; 21.98,
23.05; 23.49, 23.65; 25.53, 25.60; 26.00, 27.05; 34.46, 34.98; 39.63, 41.24;
58.61, 58.98; 64.95, 65.74; 66.89, 66.94; 127.66, 127.70; 127.77; 127.88;
128.29; 128.42; 136.41, 136.58; 153.49, 154.51; 179.24, 180.82. HRMS (ESI
neg.) C18H22NO4 [M–H]–: calcd 316.1554, found 316.1554.
Synthesis of (2R,3aR,7aR)-N-(benzyloxycarbonyl)-2-methylocta-
hydroindole-2-carboxylic acid [(R,R,R)-12]. An identical procedure
to that described above was applied to transform (R,R,R)-11 (2.01 g,
6.06 mmol) into (R,R,R)-12 (1.91 g, 6.03 mmol, 99% yield). [a]D –12.0 (c
0.85, MeOH). HRMS (ESI neg.) C18H22NO4 [M–H]–: calcd 316.1554,
found 316.1558. Spectroscopic data are the same as those described for
(S,S,S)-12.
Synthesis of methyl (2S*,3aS*,7aS*)-2-methyloctahydroindole-2-
carboxylate hydrochloride [(S*,S*,S*)-10]. A 3 N solution of hydro-
gen chloride in anhydrous ethyl acetate (20 ml) was added to
(S*,S*,S*)-5 (4.40 g, 14.80 mmol), and the reaction mixture was stirred
at room temperature for 4 h. The solvent was eliminated, and the solid
was dissolved in water and lyophilized to afford (S*,S*,S*)-10 (3.46 g,
14.80 mmol, 100% yield). Mp 167–1698C. HRMS (ESI) C11H20NO2 [M–
Cl]1: calcd 198.1489, found 198.1485. Spectroscopic data are identical to
those reported in our previous work13 for the (S,S,S) enantiomer.
RESULTS AND DISCUSSION
Synthesis of a Racemic Precursor
The first stage in the synthesis of enantiomerically pure
(S,S,S)- and (R,R,R)-(aMe)Oic involved the preparation of a
racemic common precursor to be subsequently subjected to
HPLC resolution. The synthesis of such a precursor was
addressed by the modification of the asymmetric route previ-
ously reported13 for the preparation of enantiopure (S,S,S)-
(aMe)Oic, which is presented in Figure 2. It began with the
hydrogenation of commercially available (S)-indoline-2-car-
boxylic acid, (S)-1, using PtO2 as a catalyst to afford a 90:10
mixture of the (S,S,S) and (S,R,R) octahydroindole deriva-
tives. The two epimeric compounds retain the (S) configura-
tion at the a carbon present in the starting material and differ
in the stereochemistry of the two new chiral centers formed.
The desired (S,S,S) stereoisomer was isolated pure by
recrystallization and was protected at the carboxylic acid and
amino functionalities to give (S,S,S)-4. This N-Boc amino
ester underwent a stereoselective a-methylation reaction and
the major diastereoisomer, (S,S,S)-5, was isolated by column
chromatography. Subsequent deprotection of the carboxylic
acid yielded (S,S,S)-(aMe)Oic in the N-Boc protected form,
that is, adequately protected for use in peptide chemistry.
The racemic version of this methodology would require
the use of racemic indoline-2-carboxylic acid (rac-1) as a sub-
strate (Fig. 3). Alternatively, the completely unsaturated com-
pound, indole-2-carboxylic acid (8) (Fig. 3), can be envisaged
as a less expensive starting material for the preparation of
the racemate (S*,S*,S*)-4. However, according to the litera-
ture,18,19 the catalytic hydrogenation of the indole system
requires harsh reaction conditions (high temperatures and
pressures of hydrogen gas). Yet, some ester derivatives of 8
have been reported20 to undergo hydrogenation under Rh/C
catalysis and hydrogen pressure at room temperature.
Synthesis of methyl (2S*,3aS*,7aS*)-N-(benzyloxycarbonyl)-2-
methyloctahydroindole-2-carboxylate [(S*,S*,S*)-11]. To an ice-
cooled solution of (S*,S*,S*)-10 (3.42 g, 14.63 mmol) and N,N-diisopro-
pylethylamine (10.4 ml, 59.68 mmol) in dry dichloromethane (75 ml),
benzyl chloroformate (4.40 ml, 29.28 mmol) was added dropwise. Once
the addition was completed, the reaction was kept at room temperature
overnight. The mixture was washed with saturated aqueous NaHCO3,
and the organic phase was dried over anhydrous MgSO4 and filtered.
The solvent was evaporated, and the residue was purified by column
chromatography (eluent: hexanes/ethyl acetate 10/1) to give (S*,S*,S*)-
11 as a colorless oil (4.47 g, 13.49 mmol, 92% yield). IR (neat) m 3035,
2950, 1743, 1702 cm–1 1H NMR (DMSO-d6, 400 MHz) d 1.00–1.28 (m,
.
2H), 1.34–1.73 (m, 6H) overlapped with 1.46 and 1.48 (two s, 3H), 1.94,
2.08 (two m, 1H), 2.16–2.27 (m, 1H), 2.44–2.56 (m, 1H), 3.48, 3.62 (two
s, 3H), 3.79 (m, 1H), 4.94–5.11 (m, 2H), 7.25–7.45 (m, 5H) 13C NMR
(DMSO-d6, 100 MHz) d (duplicate signals are observed for most car-
bons) 19,98, 20.12; 21.63; 22.97, 23.07; 24.89, 24.97; 25.60, 26.61; 33.61,
34.53; 38.80, 40.23; 52.11; 57.66, 58,23; 64.53, 64.96; 65.55, 65.85; 127.21;
127.42; 127.64, 127.69; 128.23; 128.30; 136.49, 136.93; 152.48, 152.80;
174.09, 174.35. HRMS (ESI) C19H25NNaO4 [M1Na]1: calcd 354.1676,
found 354.1676.
Resolution of (S*,S*,S*)-11: Isolation of methyl (2S,3aS,7aS)- and
(2R,3aR,7aR)-N-(benzyloxycarbonyl)-2-methyloctahydroindole-2-
carboxylate [(S,S,S)-11 and (R,R,R)-11]. HPLC resolution of race-
mic (S*,S*,S*)-11 (4.142 g) dissolved in chloroform (6.90 ml) was car-
ried out by successive injections (one every 11.5 min) of 600 ll on a 250
mm 3 20 mm ID Chiralpak1 IC column. A mixture of n-hexane/tert-
butyl methyl ether/2-propanol 50/44/6 was used as the eluent working
at a flow rate of 16 ml/min and with UV monitoring at 220 nm. Three
separate fractions were collected. Optically pure (S,S,S)-11 (2.043 g)
and (R,R,R)-11 (2.006 g) were, respectively, obtained by evaporation of
the first and third fractions. The second fraction contained 38 mg of a
22/78 mixture of (S,S,S)-11/(R,R,R)-11 and was discarded. (S,S,S)-11:
colorless oil. [a]D –9.6 (c 0.92, CHCl3). (R,R,R)-11: colorless oil. [a]D
19.5 (c 0.95, CHCl3). Spectroscopic data for (S,S,S)- and (R,R,R)-11 are
identical to those given above for the racemic compound.
On the basis of this precedent, we undertook the prepara-
tion of the target amino acids, (S,S,S)- and (R,R,R)-(aMe)Oic,
starting from an adequately protected derivative of indole-2-
carboxylic acid (8), namely the N-Boc protected ester 9
(Fig. 3). The latter compound was readily obtained by the
reaction of 8 with methanol in the presence of sulfuric acid
followed by treatment with di-tert-butyl dicarbonate. The hy-
drogenation of 9 using PtO2 as a catalyst was accomplished
Synthesis of (2S,3aS,7aS)-N-(benzyloxycarbonyl)-2-methyloctahy-
droindole-2-carboxylic acid [(S,S,S)-12]. A 1 M solution of potassium
hydroxide in methanol (50 ml) was added to (S,S,S)-11 (2.04 g, 6.16
mmol), and the reaction mixture was heated under reflux for 24 h. After
evaporation of the solvent, the residue was taken up in water and
washed with diethyl ether. The aqueous phase was acidified with 5%
Chirality DOI 10.1002/chir

510
SAYAGO ET AL.
tially cheaper than rac-1), global yield, hydrogenation condi-
tions (the process in Fig. 2 requires heating at 608C), and
stereoselectivity of the hydrogenation process (complete vs.
90:10). The synthetic pathway in Figure 3 is therefore highly
convenient for the production of (S*,S*,S*)-Oic derivatives
on a large scale.
Once racemic (S*,S*,S*)-4 was obtained according to the
procedure in Figure 3, the synthetic route proceeded in a
similar way to that described in the asymmetric version pre-
viously reported13 and shown in Figure 2. Thus, the treat-
ment of (S*,S*,S*)-4 with lithium diisopropylamide gener-
ated an intermediate lithium enolate that reacted with methyl
iodide to provide an 89:11 mixture of the two possible dia-
stereoisomers, (S*,S*,S*)-5 and (R*,S*,S*)-6 (Fig. 4). It
should be noted that the high steric hindrance imposed by
the cis-fused cyclohexane ring was responsible for the effi-
cient facial stereodifferentiation observed.13 The major dia-
stereoisomer, (S*,S*,S*)-5, was isolated pure from the crude
mixture by column chromatography in 75% yield.
At this stage, we undertook the exchange of the N-Boc
protecting group by the Cbz one, as the latter has the advant-
age of absorbing in the UV range, thus favoring monitoring
of the subsequent chromatographic resolution process. This
was achieved by the treatment of (S*,S*,S*)-5 with a satu-
rated solution of hydrogen chloride in ethyl acetate followed
by the reaction with benzyl chloroformate (Fig. 4). In this
way, several grams of (S*,S*,S*)-11 were prepared in 61%
overall yield from indole-2-carboxylic acid (8). The interme-
diate amino ester hydrochloride, (S*,S*,S*)-10, furnished
single crystals suitable for X-ray diffraction analysis. The
crystalline structure (Fig. 5) confirmed the expected
(S*,S*,S*) relative stereochemistry. As shown in Figure 5,
the cyclohexane ring adopts a chair conformation, which is
slightly distorted to accommodate the cis-fused five-mem-
bered cycle. The latter exhibits an envelope shape, with one
of the carbon atoms involved in the fusion (3a) occupying
the flap of the envelope. It should be noted that inversion of
the reactions sequence carried out to transform (S*,S*,S*)-4
into (S*,S*,S*)-11 was examined, but only poor results were
obtained. Thus, when the Cbz group was introduced in
(S*,S*,S*)-4 before the a-methylation reaction, the latter pro-
cess afforded only a 25% yield of (S*,S*,S*)-11 (Fig. 4).
Accordingly, this alternative pathway was discarded.
Fig. 2. Stereoselective synthesis for enantiopure (S,S,S)-(aMe)Oic previ-
ously reported (ref. 13). Abbreviations: Boc, tert-butoxycarbonyl; LDA, lith-
ium diisopropylamide.
at room temperature and atmospheric pressure of hydrogen
gas. It seems, therefore, that both the amino and carboxylic
acid functions in 8 are to be in the protected form (as are in
9) for the hydrogenation of the indole system to proceed
under such mild conditions. The choice of PtO2 as the cata-
lyst for this reaction was based on our previous observation13
that the aromatic ring in (S)-1 is efficiently hydrogenated at
atmospheric pressure in the presence of PtO2 (Fig. 2),
whereas the use of Pd/C or Rh requires high pressure of
hydrogen.21,22 Moreover, we were delighted to observe that the
hydrogenation of 9 provided racemic (S*,S*,S*)-4 as a single
product. This differential behavior with respect to that
described above for (S)-1 (Fig. 2) is due to the fact the three
chiral centers are formed at once on hydrogenation of 9 (Fig.
3). Accordingly, only the compound exhibiting a cis relative dis-
position of the two bridgehead hydrogen atoms and that at the
a position, (S*,S*,S*)-4, is obtained in this process.
Therefore, the route in Figure 3 constitutes a highly effi-
cient manner to access racemic (S*,S*,S*)-4 as a single ste-
reoisomer in high overall yield (88%) starting from inexpen-
sive indole-2-carboxylic acid (8). Certainly, this route is
much more convenient for the production of racemic Oic
derivatives exhibiting an (S*,S*,S*) relative stereochemistry
than the racemic version of the asymmetric synthesis
reported before13 (Fig. 2) that would require the use of indo-
line-2-carboxylic acid (rac-1) as a substrate. As outlined
above, the procedure in Figure 3 is more advantageous in
terms of accessibility of the starting material (8 is substan-
Fig. 3. Synthesis of racemic (S*,S*,S*)-4.
Chirality DOI 10.1002/chir

HPLC RESOLUTION OF (S*,S*,S*)-(ꢀME)OIC
511
Fig. 5. X-ray crystal structure of racemic (S*,S*,S*)-10 [only the (S,S,S)
enantiomer is shown]. Heteroatoms are drawn as thermal ellipsoids. Most
hydrogens have been omitted for clarity.
derived ones and was therefore not considered for further
assays. The addition of a third component to the initial n-hex-
ane/2-propanol mixture was then evaluated. Among those
tested, tert-butyl methyl ether provided the best results, par-
ticularly for the IC column (Table 1). Thus, when working
on this stationary phase, the presence of a high percentage
of this solvent in the mobile phase not only improved the se-
lectivity but also led to a substantial enhancement of the re-
solution factor. Moreover, this change also resulted benefi-
cial to the sample solubility, which is an extremely important
issue for the subsequent scaling-up stage. Accordingly, the
analytical conditions considered to be optimal for extension
to the preparative-scale enantioseparation were elution with a
50/44/6 (v/v/v) mixture of n-hexane/tert-butyl methyl
ether/2-propanol at a flow rate of 0.8 ml/min on Chiralpak1
IC. The profile obtained under such chromatographic condi-
tions is shown in Figure 6. Further assays carried out on the
IC analytical column working in an overload mode confirmed
that these eluting conditions provided the optimal results in
terms of both resolution and column loadability.
Fig. 4. Synthesis of the racemic precursor of the target amino acids,
(S*,S*,S*)-11. Abbreviations: Cbz, benzyloxycarbonyl.
HPLC Resolution: Isolation of Enantiopure Amino Acids
Once the racemic precursor of the target amino acids,
(S*,S*,S*)-11, had been obtained, we addressed its resolution
by HPLC. The separation of enantiomers by preparative chro-
matography using chiral stationary phases is nowadays recog-
nized as a powerful tool to produce enantiopure compounds.23–
25 Polysaccharide-based stationary phases are particularly suita-
ble for this purpose due to their excellent chiral recognition
ability and high loading capacity.23–25 Moreover, HPLC col-
umns containing immobilized cellulose and amylose deriva-
tives have recently become commercially available (Chir-
alpak1 IA, IB, and IC).26–28 In these columns, the chiral sup-
port is covalently bonded to the silica matrix, which results in
a high stability in the presence of all organic solvents.25–29 This
feature means a great advantage for resolutions on a prepara-
tive scale. Chiral stationary phases of this type, either commer-
cial or made at the laboratory (before their commercialization),
have been used in our group for the preparative HPLC resolu-
tion of a wide variety of non-natural protected amino acids.30–41
The resolution of racemic (S*,S*,S*)-11 was first tested at
the analytical level using 250 mm 3 4.6 mm Chiralpak1 IA,
IB, and IC columns that, respectively, contain tris(3,5-dime-
thylphenylcarbamate) of amylose,26 tris(3,5-dimethylphenyl-
carbamate) of cellulose,27 and tris(3,5-dichlorophenylcarba-
mate) of cellulose as the chiral selector.28 The screening
started using mixtures of n-hexane/2-propanol as the eluent.
The enantiomers of 11 could not be distinguished on the IA
column, whereas both Chiralpak1 IB and IC showed excel-
lent enantiodiscrimination abilities, with selectivity factors
around 1.6 (Table 1). Replacement of the alcohol component
in the eluting mixture by acetone was next tested. This
change had little effect for the cellulose-based phases,
whereas very much improved the chiral recognition ability of
Chiralpak1 IA. In fact, an almost complete baseline separa-
tion of peaks was achieved when eluting with n-hexane/ace-
tone 96/4 (v/v) on this stationary phase (Table 1). In spite of
this, it remained much less effective that the cellulose-
These conditions were subsequently scaled-up for the
preparative resolution of (S*,S*,S*)-11 on a 250 mm 3 20
mm ID Chiralpak1 IC column. Successive injections of 600
ll of a highly concentrated solution of the sample in chloro-
form (600 mg per ml of solvent) were performed (Fig. 7).
TABLE 1. Selected chromatographic data for the analytical
HPLC resolution of compound (S*,S*,S*)-11 on Chiralpak1
IA, IB, and IC columnsa
Eluentb
% (v/v)
k1
a
RS
0
Chiralpak1 IA
n-Hx/2-propanol
98/2
96/4
2.4
2.4
1.0
1.1
–
1.3
n-Hx/acetonec
Chiralpak1 IB
n-Hx/2-propanol
98/2
96/4
80/19/1
1.6
2.1
1.6
1.6
1.5
1.6
5.4
5.8
5.6
n-Hx/acetonec
n-Hx/t-BuOMe/2-propanol
Chiralpak1 IC
n-Hx/2-propanol
80/20
1.7
1.0
1.3
1.6
1.5
1.9
5.2
5.3
6.7
n-Hx/acetonec
90/10d
50/44/6d
n-Hx/t-BuOMe/2-propanol
aColumn size: 250 mm 3 4.6 mm ID. Flow rate: 1.0 ml/min (unless otherwise
indicated). UV detection at 210 nm (unless otherwise indicated). The chro-
0
matographic parameters k1 , a, and RS are defined in the Materials and Meth-
ods section.
bn-Hx: n-hexane; t-BuOMe: tert-butyl methyl ether.
cUV detection at 220 nm.
dFlow rate: 0.8 ml/min.
Chirality DOI 10.1002/chir

512
SAYAGO ET AL.
Fig. 8. HPLC analytical profile of the resolved enantiomers of 11. Chro-
matographic conditions as in Figure 6.
Fig. 6. HPLC resolution of (S*,S*,S*)-11 at the analytical level. Column:
Chiralpak1 IC (250 mm 3 4.6 mm ID). Eluent: n-hexane/tert-butyl methyl
ether/2-propanol 50/44/6. Flow rate: 0.8 ml/min. UV detection: 210 nm.
Finally, each enantiomer of 11 was subjected to saponifi-
cation to provide the corresponding N-Cbz amino acid in
almost quantitative yield (Fig. 9). Thus, the target com-
pounds, (S,S,S)- and (R,R,R)-(aMe)Oic, were isolated in enan-
tiomerically pure form and suitably protected for further deri-
vatization, including their direct use in peptide synthesis.
The Cbz-protecting group can be eliminated to liberate the
amino function under mild reaction conditions that do not
affect the optical integrity of the compounds.
Each run was collected into three separate fractions accord-
ing to the cut points indicated in Figure 7, with equivalent
fractions of successive injections being combined. Following
this protocol, a total of 4.14 g of racemate was submitted to
the resolution procedure, which was completed in about 3 h.
Evaporation of the first and third fractions provided 2.04 and
2.01 g of the first and second eluted enantiomers, respec-
tively, both of them in optically pure form, as assessed at the
analytical level (Fig. 8). This means that 98% of the injected
material was recovered in enantiomerically pure form after a
single passage through the column. The high productivity of
the enantioseparation process is therefore remarkable
because more than 1 g of optically pure material was isolated
per hour working on a semipreparative size column.
The absolute configuration of the isolated enantiomers
was assigned by comparison of their retention times with
that of a sample of enantiomerically pure (S,S,S)-11 prepared
separately. Thus, a small amount of (S,S,S)-5 obtained from
(S)-indoline-2-carboxylic acid in our previous work13 follow-
ing the synthetic procedure in Figure 2 was transformed into
(S,S,S)-11 by exchange of the Boc-Cbz N-protecting groups,
and its retention time was found to be coincident with that of
the first eluted enantiomer of 11 (Fig. 8). Accordingly, the
more strongly retained enantiomer of 11 was assigned an
(R,R,R) configuration.
CONCLUSIONS
We have developed an efficient and practical method for
the production of multigram quantities of the a-tetrasubsti-
tuted proline analogs (S,S,S)- and (R,R,R)-2-methyloctahy-
droindole-2-carboxylic acids in enantiomerically pure form
and suitably protected for incorporation into peptides. The
procedure relies in the use of inexpensive indole-2-carboxylic
acid as the starting material, which allows the preparation of
a racemic precursor in high overall yield through simple
transformations, and the efficiency of the HPLC resolution
process carried out on a chiral column (Chiralpak1 IC). The
procedure has been applied to the isolation of significant
Fig. 7. Preparative HPLC resolution of (S*,S*,S*)-11. Profile correspond-
ing to the injection of 600 ll of a solution of 11 in chloroform (concentration:
600 mg solute/ml of solvent). The cut points for the three fractions collected
are indicated. Column: Chiralpak1 IC (250 mm 3 20 mm ID). Eluent: n-hex-
ane/tert-butyl methyl ether/2-propanol 50/44/6. Flow rate: 16 ml/min. UV
detection: 220 nm.
Fig. 9. Preparation of N-protected enantiomerically pure (S,S,S)- and
(R,R,R)-(aMe)Oic.
Chirality DOI 10.1002/chir

HPLC RESOLUTION OF (S*,S*,S*)-(ꢀME)OIC
513
20. Blankley CJ, Kaltenbronn JS, DeJohn DE, Werner A, Bennett LR, Bobow-
ski G, Krolls U, Johnson DR, Pearlman WM, HoefIe ML, Essenburg AD,
Cohen DM, Kaplan HR. Synthesis and structure-activity relationships of
potent new angiotensin converting enzyme inhibitors containing satu-
rated bicyclic amino acids. J Med Chem 1987;30:992–998.
quantities of optically pure compounds and can be easily
scaled-up to the production of larger amounts. The amino
acids thus prepared are proline analogs of great value in the
design of peptides with pharmacological applications and
other medicinally relevant compounds.
21. Vincent M, Marchand B, Re´mond G, Jaguelin-Guinamant S, Damien G,
Portevin B, Baumal JY, Volland JP, Bouchet JP, Lambert PH, Serkiz B,
Luitjen W, Laubie M, Schiavi P. Synthesis and ACE inhibitory activity of
the stereoisomers of perindopril (S 9490) and perindoprilate (S 9780).
Drug Des Discov 1992;9:11–28.
ACKNOWLEDGMENTS
The authors thank Ana Lidia Bernad for assistance with
HPLC.
22. Vincent M, Baliarda J, Marchand B, Remond G. Process for the industrial
synthesis of perindopril. US Pat.4,914,214, 1990.
LITERATURE CITED
23. Cox GB. Preparative Enantioselective Chromatography. Oxford: Black-
well; 2005.
1. Adessi C, Soto C. Converting a peptide into a drug: strategies to improve
stability and bioavailability. Curr Med Chem 2002;9:963–978.
24. Francotte E. Isolation and production of optically pure drugs by enantio-
selective chromatography. In: Francotte E, Lindner W, editors. Chirality
in drug research. Weinheim: Wiley-VCH; 2006. p 155–187.
2. Cowell SM, Lee YS, Cain JP, Hruby VJ. Exploring Ramachandran and Chi
space: conformationally constrained amino acids and peptides in the design
of bioactive polypeptide ligands. Curr Med Chem 2004;11:2785–2798.
25. Okamoto Y, Ikai T. Chiral HPLC for efficient resolution of enantiomers.
Chem Soc Rev 2008;37:2593–2608.
3. Toniolo C, Crisma M, Formaggio F, Peggion C. Control of peptide con-
formation by the Thorpe-Ingold effect (Ca-tetrasubstitution). Biopoly-
mers (Pept Sci) 2001;60:396–419.
26. Zhang T, Kientzy C, Franco P, Ohnishi A, Kagamihara Y, Kurosawa H.
Solvent versatility of immobilized 3,5-dimethylphenylcarbamate of amy-
lose in enantiomeric separations by HPLC. J Chromatogr A 2005;1075:
65–75.
4. Sagan S, Karoyan P, Lequin O, Chassaing G, Lavielle S. N- and Ca-meth-
ylation in biologically active peptides: synthesis, structural and functional
aspects. Curr Med Chem 2004;11:2799–2822.
27. Zhang T, Nguyen D, Franco P, Murakami T, Ohnishi A, Kurosawa H.
Cellulose 3,5-dimethylphenylcarbamate immobilized on silica: a new chi-
ral stationary phase for the analysis of enantiomers. Anal Chim Acta
2006;557:221–228.
5. Calaza MI, Cativiela C. Stereoselective synthesis of quaternary proline
analogues. Eur J Org Chem 2008;3427–3448.
6. Thaisrivongs S, Pals DT, Lawson JA, Turner SR, Harris DW. a-Methyl-
proline-containing renin inhibitory peptides: in vivo evaluation in an anes-
thetized, ganglion-blocked, hog renin infused rat model. J Med Chem
1987;30:536–541.
28. Zhang T, Nguyen D, Franco P, Isobe Y, Michishita T, Murakami T. Cel-
lulose tris(3,5-dichlorophenylcarbamate) immobilised on silica: a novel
chiral stationary phase for resolution of enantiomers. J Pharm Biomed
Anal 2008;46:882–891.
7. Hinds MG, Welsh JH, Brennand DM, Fisher J, Glennie MJ, Richards
NGJ, Turner DL, Robinson JA. Synthesis, conformational properties, and
antibody recognition of peptides containing b-turn mimetics based on
a-alkylproline derivatives. J Med Chem 1991;34:1777–1789.
29. Zhang T, Franco P. Analytical and preparative potential of immobilized
polysaccharide-derived chiral stationary phases. In: Subramanian G, edi-
tor. Chiral separation techniques. Weinheim: Wiley-VCH; 2007. p 99–134.
30. Cativiela C, D´ıaz-de-Villegas MD, Jime´nez AI, Lo´pez P, Marraud M, Oli-
veros L. Efficient access to all four stereoisomers of phenylalanine cyclo-
propane analogues by chiral HPLC. Chirality 1999;11:583–590.
8. Bisang C, Weber C, Inglis J, Schiffer CA, van Gunsteren WF, Jelesarov I,
Bosshard HR, Robinson JA. Stabilization of type-I b-turn conformations in
peptides containing the NPNA-repeat motif of the Plasmodium falciparum
circumsporozoite protein by substituting proline for (S)-a-methylproline.
J Am Chem Soc 1995;117:7904–7915.
31. Al´ıas M, Cativiela C, Jime´nez AI, Lo´pez P, Oliveros L, Marraud M. Prepa-
rative HPLC resolution of the cis cyclohexane analogs of phenylalanine.
Chirality 2001;13:48–55.
9. Hurst M, Jarvis B. Perindopril: an updated review of its use in hyperten-
sion. Drugs 2001;61:867–896.
32. Jime´nez AI, Lo´pez P, Oliveros L, Cativiela C. Facile synthesis and highly
efficient resolution of a constrained cyclopropane analogue of phenylala-
nine. Tetrahedron 2001;57:6019–6026.
10. Reissmann S, Imhof D. Development of conformationally restricted ana-
logues of bradykinin and somatostatin using constrained amino acids
and different types of cyclization. Curr Med Chem 2004;11:2823–2844.
33. Royo S, Lo´pez P, Jime´nez AI, Oliveros L, Cativiela C. First synthesis of
the two enantiomers of a-methyldiphenylalanine [(aMe)Dip] by HPLC
resolution. Chirality 2002;14:39–46.
11. Stewart JM. Bradykinin antagonists: discovery and development. Pep-
tides 2004;25:527–532.
34. Gil AM, Bu˜nuel E, Lo´pez P, Cativiela C. Synthesis of enantiopure 7-azanor-
bornane proline-a-amino acid chimeras by highly efficient HPLC resolution
of a phenylalanine analogue. Tetrahedron: Asymmetry 2004;15:811–819.
12. Lawandi J, Gerber-Lemaire S, Juillerat-Jeanneret L, Moitessier N. Inhibi-
tors of prolyl oligopeptidases for the therapy of human diseases: Defining
diseases and inhibitors. J Med Chem 2010;53:3423–3438.
35. Cativiela C, Lo´pez P, Lasa M. Synthesis and preparative resolution of the
trans-cyclohexane analogues of phenylalanine. Eur J Org Chem 2004;
3898–3908.
13. Sayago FJ, Calaza MI, Jime´nez AI, Cativiela C. A straightforward route to
enantiopure a-substituted derivatives of (2S,3aS,7aS)-octahydroindole-2-
carboxylic acid. Tetrahedron 2009;65:5174–5180.
36. Al´ıas M, Lo´pez MP, Cativiela C. An efficient and stereodivergent
synthesis of threo- and erythro-b-methylphenylalanine. Resolution of each
racemic pair by semipreparative HPLC. Tetrahedron 2004;60:885–891.
14. Sayago FJ, Calaza MI, Jime´nez AI, Cativiela C. Versatile methodology for
the synthesis and a-functionalization of (2R,3aS,7aS)-octahydroindole-2-
carboxylic acid. Tetrahedron 2008;64:84–91.
37. Lasa M, Lo´pez P, Cativiela C. Synthesis of the four stereoisomers of
cyclobutane analogues of phenylalanine in enantiomerically pure form.
Tetrahedron: Asymmetry 2005;16:4022–4033.
15. Barvian NC, Connolly CJC, Guzzo PR, Hamby JM, Hicks JL, Johnson
MR, Le V-D, Mitchell LH, Roark WH. [B]-Fused bicyclic proline deriva-
tives and their use for treating arthritic conditions. PCT Int. Appl.
WO2004092132, 2004.
38. Jime´nez AI, Lo´pez P, Cativiela C. Synthesis and HPLC enantioseparation
of the cyclopropane analogue of valine (c3Val). Chirality 2005;17:22–29.
16. Sheldrick GM. SHELXS-97, Program for the Solution of Crystal Struc-
tures; Go¨ttingen: University of Go¨ttingen; 1997.
39. Cativiela C, Lasa M, Lo´pez P. Synthesis of enantiomerically pure cis- and
trans-cyclopentane analogues of phenylalanine. Tetrahedron: Asymmetry
2005;16:2613–2623.
17. Sheldrick GM. SHELXL-97. Program for the Refinement of Crystal Struc-
tures; Go¨ttingen: University of Go¨ttingen; 1997.
40. Royo S, Jime´nez AI, Cativiela C. Synthesis of enantiomerically pure b,b-
diphenylalanine (Dip) and fluorenylglycine (Flg). Tetrahedron: Asymme-
try 2006;17:2393–2400.
18. Adkins H, Coonradt HL. The selective hydrogenation of derivatives
of pyrrole, indole, carbazole and acridine. J Am Chem Soc 1941;63:
1563–1570.
41. Sayago FJ, Jime´nez AI, Cativiela C. Efficient access to N-protected deriva-
tives of (R,R,R)- and (S,S,S)-octahydroindole-2-carboxylic acid by HPLC
resolution. Tetrahedron: Asymmetry 2007;18:2358–2364.
19. Cui Y, Kwok S, Bucholtz A, Davis B, Whitney RA, Jessop PG. The effect
of substitution on the utility of piperidines and octahydroindoles for re-
versible hydrogen storage. New J Chem 2008;32:1027–1037.
Chirality DOI 10.1002/chir