Selective biotransformation of substituted alicyclic nitriles by Rhodococcus equi A4

Article abstract of DOI:10.1139/v01-205

Nitrile hydratase from Rhodococcus equi A4 discriminated between geometric isomers of substituted alicyclic nitriles. The enzyme transformed trans-4-benzoyloxycyclohexanecarbonitrile (trans-1a), cis-3-benzoyloxy-cyclohexanecarbonitrile (cis-2a), trans-2-h

Full text of DOI:10.1139/v01-205

724
Selective biotransformation of substituted
alicyclic nitriles by Rhodococcus equi A4
Ludmila Martínková, Norbert Klempier, Margit Preiml, Mária Ovesná,
Marek Kuzma, Veronika Mylerová, and Vladimír KÍen
Abstract: Nitrile hydratase from Rhodococcus equi A4 discriminated between geometric isomers of substituted
alicyclic nitriles. The enzyme transformed trans-4-benzoyloxycyclohexanecarbonitrile (trans-1a), cis-3-benzoyloxy-
cyclohexanecarbonitrile (cis-2a), trans-2-hydroxycyclohexanecarbonitrile (trans-3a), and trans-2-hydroxycyclo-
pentanecarbonitrile (trans-4a) into the corresponding amides. On the contrary, cis-2-hydroxycyclohexanecarbonitrile
(cis-3a) and cis-2-hydroxycyclopentanecarbonitrile (cis-4a) were not converted to a significant extent. cis-4-Benzoyl-
oxycyclohexanecarbonitrile (cis-1a) was also a substrate of the enzyme but reacted slowly. Diequatorial arrangement
of the substituents in trans-1a, cis-2a, and trans-3a appears to positively influence the activity of the nitrile hydratase.
Key words: nitrile hydratase, substituted cyclohexanecarbonitriles, substituted cyclopentanecarbonitriles.
Résumé : L’hydratase des nitriles obtenue à partir de Rhodococcus equi A4 donne lieu à de la discrimination entre
des isomères géométriques de nitriles alicyliques substitués. L’enzyme a transformé les trans-4-benzoyloxycyclo-
hexanecarbonitrile (trans-1a), cis-3-benzoyloxycyclohexanecarbonitrile (cis-2a), trans-2-hydroxycyclohexanecarbonitrile
(trans-3a) et trans-2-hydroxycyclopentanecarbonitrile (trans-4a) en amides correspondants. Au contraire, le degré de
conversion des cis-2-hydroxycyclohexanecarbonitrile (cis-3a) et cis-2-hydroxycyclopentanecarbonitrile (cis-4a) est
faible. Comme substrat de cet enzyme, le cis-4-benzoyloxycyclohexanecarbonitrile (cis-1a) ne réagit que lentement.
L’arrangement diéquatorial des substituants dans les isomères trans-1a, cis-2a et trans-3a semble influencer positive-
ment l’activité de l’hydratase des nitriles.
Mots clés : hydratase des nitriles, cyclohexanecarbonitriles substitués, cyclopentanecarbonitriles substitués.
[Traduit par la Rédaction] Martínková et al. 727
Introduction
portant trans-4-aminomethylcyclohexanecarboxylic acid (tran-
examic acid) (3).
The hydrolysis of alicyclic mono- and dinitriles and
amides of cis configuration was reported to be catalyzed by
Rhodococcus rhodochrous IFO 15564 with enantioselectiv-
ity in some cases (4), but the biotransformation of the corre-
sponding trans isomers was not studied.
Whole-cell biocatalyst SP 409 (Novo Industri, not available
commercially) hydrolyzed both cis- and trans-2-hydroxycy-
clohexanecarbonitrile and cis-2-hydroxycyclopentanecarbo-
nitrile (5).
Here we report the discrimination of trans and cis isomers
of the 3- and 4-benzoyloxy- and 2-hydroxy derivatives of
cyclohexanecarbonitriles as well as 2-hydroxycyclopen-
tanecarbonitriles by the nitrile hydratase from Rhodococcus
equi A4.
The use of bacterial nitrile-converting enzymes for enantio-,
regio-, and chemoselective hydration and hydrolysis of nitriles
has been described in numerous works (1). This work re-
ports the resolution of geometric isomers using a nitrile
hydratase.
Different affinities of this enzyme for geometric isomers
of trans-1,4-dicyanocyclohexane (DCC) were shown using
Corynebacterium sp. C5 cells containing both nitrile
hydratase and amidase. The relative specific activity of the
biocatalyst for the cis isomer of DCC was only 6% of that
for the trans isomer (2). The conversion of trans-DCC to
trans-4-cyanocyclohexane-1-carboxylic acid was involved in
the chemoenzymatic synthesis of the pharmaceutically im-
Received 4 September 2001. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 4 June 2002.
Results and discussion
Dedicated to Professor J. Bryan Jones on the occasion of his
65th birthday.
Hydrolysis of alicyclic nitriles, 2-hydroxycyclohexanecar-
bonitrile, 2-hydroxycyclopentanecarbonitrile (5), cyclopenta-
necarbonitrile (6), and 2-arylcyclopropanecarbonitriles (7),
or the regioselective hydrolysis of DCC (2, 3, 8), trans-cyc-
lohexane-1,4-diacetonitrile (9), and other saturated and un-
saturated dinitriles (4) was the subject of few reports
regarding enzymatic nitrile conversion. The present work
demonstrated that the nitrile hydratase from R. equi A4,
which showed a broad substrate specificity for aromatic,
L. Martínková,1 M. Ovesná, M. Kuzma, V. Mylerová, and
V. KÍen. Institute of Microbiology, Academy of Sciences of
the Czech Republic, Laboratory of Biotransformation,
VídeÁská 1083, CZ-142 20 Prague 4, Czech Republic.
N. Klempier and M. Preiml. Institute of Organic Chemistry,
Technical University Graz, Stremayrgasse 16, A-8010 Graz,
Austria.
1Corresponding author (e-mail: martinko@biomed.cas.cz).
Can. J. Chem. 80: 724–727 (2002)
DOI: 10.1139/V01-205
© 2002 NRC Canada

Martínková et al.
725
Scheme 1.
Table 1. Biotransformations of 2-hydroxycyclohexanecarbonitriles,
2-hydroxycyclopentanecarbonitriles, and 3-, 4-benzoyloxycyclohex-
anecarbonitriles by the nitrile hydratase of Rhodococcus equi A4.
R
R
R
Reaction
Isolated
OBz
Substratea time (h) Product
Conversion (%)b yield (%)
OBz
OBz
trans-1a
cis-1a
cis-2a
trans-3a
cis-3a
trans-4a
cis-4a
2
48
5
3.5
24
3
trans-1b
cis-1b
cis-2b
trans-3b
cis-3b
trans-4b
no reaction
100
85
86
100
<5c
91
74
79
48
—
51
—
cis-2
trans-1
cis-1
R
R
R
R
OH
OH
OH
OH
trans-4
100
—
trans-3
cis-3
cis-4
24
a
R = CN
R = CONH
aThe structures (see Scheme 1) represent single enantiomers of trans-1a,
cis-1a, cis-2a, trans-3a, cis-3a, trans-4a, and cis-4a, whereas the
biotransformations were carried out with racemates.
bDetermined by TLC (for compound 4a) or HPLC (for other
compounds).
b
2
c Traces of amide detected by NMR.
arylaliphatic, and aliphatic nitriles (10), is also useful for the
hydration of alicylic nitriles. The use of the purified enzyme
afforded 2-, 3-, and 4-substituted (hydroxy-, benzoyloxy-)
alicyclic amides. On the contrary, whole-cell biocatalyst SP 409
exhibiting nitrile hydratase and amidase activities gave
2-hydroxycyclohexane- and 2-hydroxycyclopentanecarboxylic
acids as the only products (5).
The syntheses of cis- and trans-2-hydroxycyclohexanecar-
boxamide (11), 2-hydroxycyclopentanecarboxamide (11), as
well as cis- (12, 13) and trans-4-hydroxycyclohexanecar-
boxamide (13) have been published previously. Hydroxy-
cyclohexanecarboxamides, both the cis- and trans-3-
isomers, were reported as the respective O-acetyl derivatives
only (12); however, no NMR data were given for these com-
pounds.
All examined trans isomers of 2- and 4-substituted
alicyclic nitriles (Scheme 1) were readily converted into the
corresponding amides by the nitrile hydratase from R. equi
A4 (Table 1). trans-4-Benzoyloxycyclohexanecarbonitrile
(trans-1a) was transformed at a substantially higher rate than
cis-1a. The reaction rate of cis-3-benzoyloxycyclohexane-
carbonitrile (cis-2a) was somewhat lower than that of
trans-1a (Table 1). The trans isomer of 2a was not available.
An even more striking difference than between trans-1a
and cis-1a was observed when comparing the activities of
the nitrile hydratase towards trans and cis isomers of
2-hydroxycyclohexanecarbonitriles (trans-3a and cis-3a) or
2-hydroxycyclopentanecarbonitriles (trans-4a and cis-4a)
(Table 1). In both pairs of nitriles, the compounds bearing
trans-substituents were totally converted within 3 h to afford
corresponding amides trans-3b and trans-4b, respectively.
On the contrary, only traces of cis-3b were formed from
cis-3a, and nitrile cis-4a was not transformed by the enzyme
at all.
It is apparent that the enzyme prefers an equatorial posi-
tion for both substituents, which can be adopted by the
trans-1,2- and trans-1,4-isomers of the investigated substi-
tuted cyclohexanecarbonitriles. Nevertheless, this diequator-
ial arrangement of the substituents is also in accordance with
cis-3-benzoyloxycyclohexanecarbonitrile (cis-2a), which is
readily converted to the amide. The two large coupling con-
stants Jaa = 9.2 and 9.7 Hz and Jaa = 8.7 and 8.8 Hz for H-1
1
and H-3, respectively, and the H coupling pattern clearly
reveal a diequatorial substituent position in cis-2a.
Impairment of the activity of the nitrile hydratase from R.
equi A4 by steric effects, for example by nitriles with bulky
groups such as 2-(3-benzoylphenyl)propionitrile, 2,6-dichloro-
benzonitrile, and some 2-substituted benzonitriles, was pre-
viously observed (10). Its inefficient hydration of 2- and
4-substituted cyclohexane(pentane)carbonitriles with a
cis-configuration suggested that this enzyme is more sensi-
tive to steric hindrances than biocatalyst SP 409.
Experimental
NMR spectra of substrates were recorded on a Varian
1
Gemini 200 spectrometer (200 and 50.3 MHz for H and
13C, respectively) in CDCl3. NMR spectra of products of the
biotransformation were recorded on a Unity Inova 400 MHz
1
spectrometer (399.90 MHz for H, 100.57 MHz for 13C) in
DMSO-d6 at 30°C. The assignment was based on COSY,
HMQC, and HMBC experiments performed using the manu-
facturer’s software. Residual solvent signals (DMSO: dH
2.50 ppm, dC 39.60 ppm; CDCl3: dH 7.27 ppm, dC
77.27 ppm) were used as internal standards.
Synthesis of substrates
Nitriles trans-1a (14), cis-1a (14), cis-2a (15), trans-3a
(16), cis-3a (17), trans-4a (18), and cis-4a (17) were pre-
pared according to literature reported procedures. 13C and 1H
NMR data were previously reported for trans-1a (19), cis-1a
(19), trans-3a (16, 20), cis-3a (20–22), trans-4a (18), and
cis-4a (20–22) but no 1H–1H coupling constants were given.
trans-1a
1H NMR d: 1.70–1.84 (m, 4H, H-2, H-3, H-5, H-6), 2.06–
2.16 (m, 4H, H-2, H-3, H-5, H-6), 2.71 (tt, J = 7.8, 4.4 Hz,
1H, H-1), 5.11 (tt, J = 7.6, 3.7 Hz, 1H, H-4), 7.44 (m, 2H,
H-meta), 7.56 (m, 1H, H-para), 8.02 (m, 2H, H-ortho).
cis-1a
1H NMR d: 1.83–2.11 (m, 8H, H-2, H-3, H-5, H-6), 2.77
(tt, J = 7.4, 3.7 Hz, 1H, H-1), 5.10 (tt, J = 7.4, 3.7 Hz, 1H,
© 2002 NRC Canada

726
Can. J. Chem. Vol. 80, 2002
H-4), 7.46 (m, 2H, H-meta), 7.58 (m, 1H, H-para), 8.05 (m,
2H, H-ortho).
plus 0.1 % (v/v) H3PO4 were employed at a flow rate of
1.0 mL min–1. Compounds trans-1a, trans-1b, cis-1a,
cis-1b, cis-2a, and cis-2b (retention times 5.6, 2.2, 5.0, 1.8,
5.5, and 2.1 min, respectively) were detected at 230 nm.
Compounds cis-3a, trans-4a, and cis-4a (retention times 5.4,
3.5, and 2.9 min, respectively) were detected by the refrac-
tive-index detector.
cis-2a
1H NMR d: 1.40–2.10 (m, 7H), 2.31 (m, 1H), 2.72 (tt, J =
3.5, 5.9, 9.2, 9.7 Hz, 1H, H-1), 5.04 (tt, J = 3.5, 5.0, 8.7,
8.8 Hz, 1H, H-3), 7.44 (m, 2H, H-meta), 7.53 (m, 1H,
H-para), 8.06 (m, 2H, H-ortho). 13C NMR d: 21.68 (C-5),
26.22 (C-6), 28.95 (C-4), 30.56 (C-2), 33.98 (C-1), 70.36
(C-3), 121.88 (CN), 128.62 (C-meta), 129.95 (C-ortho),
130.32 (C-ipso), 133.33 (C-para), 165.98 (C=O).
Isolation and identification of products
Amides trans-1b, cis-1b, and cis-2b were extracted from
supernatants of the reaction mixtures with ethyl acetate at
pH 8.5–8.8 (NaOH). The reaction mixtures from the
biotransformations of substrates trans-3a and trans-4a were
lyophilized and extracted with methanol to give the respec-
tive amides trans-3b and trans-4b.
trans-3a
1H NMR d: 1.10–1.44 (m, 4H, H-3, H-4, H-5, H-6), 1.51–
1.84 (m, 3H, H-4, H-5, H-6), 2.06 (m, 1H, H-3), 2.42 (ddd, J
= 11.5, 9.3, 3.8 Hz, 1H, H-1), 3.03 (br s, 1H, OH), 3.69 (dt,
J = 9.4, 4.2 Hz, 1H, H-2).
trans-1b
1H NMR d: 1.41–1.55 (m, 4H, H-2a, H-3a, H-5a, H-6a),
1.85 (m, 2H, H-2b, H-6b), 2.05 (m, 2H, H-3b, H-5b), 2.14
(m, S J = 29.7 Hz, 1H, H-1), 4.82 (m, S J = 29.7 Hz, 1H,
H-4), 6.71 (br s, 1H, NH), 7.25 (br s, 1H, NH), 7.52 (m, 2H,
H-meta), 7.64 (m, 1H, H-para), 7.94 (m, 2H, H-ortho). 13C
NMR d: 27.17 (t, C-2 and C-6), 30.60 (t, C-3 and C-5),
42.44 (d, C-1), 73.01 (d, C-4), 128.82 (d, C-meta), 129.19
(d, C-ortho), 130.25 (s, C-ipso), 133.33 (d, C-para), 165.29
(s, OCOPh), 176.80 (s, CONH2).
cis-3a
1H NMR d: 1.13–2.20 (m, 8H, H-3/3, H-4/4, H-5/5,
H-6/6), 3.01 (q, J = 3.1 Hz, 1H, H-1), 3.10 (br s, 1H, OH),
3.75 (dt, J = 8.4, 4.1, Hz, 1H, H-2).
trans-4a
1H NMR d: 1.48–2.30 (m, 6H), 2.65 (m, S J = 13.2 Hz,
1H, H-1), 3.68 (d, J = 3.5 Hz, 1H, OH), 4.37 (m, S J =
11.9 Hz, 1H, H-2).
cis-1b
cis-4a
1H NMR d: 1.60–1.80 (m, 6H, H-2, H-6, H-3a, H-5a),
1.93 (m, 2H, H-3b, H-5b), 2.22 (m, S J = 28.6 Hz, 1H, H-1),
5.14 (m, S J = 13.9 Hz, 1H, H-4), 6.71 (br s, 1H, NH), 7.23
(br s, 1H, NH), 7.55 (m, 2H, H-meta), 7.66 (m, 1H, H-para),
7.99 (m, 2H, H-ortho). 13C NMR d: 23.96 (t, C-2 and C-6),
28.77 (t, C-3 and C-5), 42.10 (d, C-1), 69.64 (d, C-4),
128.76 (d, C-meta), 129.06 (d, C-ortho), 130.38 (s, C-ipso),
133.20 (d, C-para), 165.01 (s, OCOPh), 176.78 (s, CONH2).
1H NMR d: 1.46–2.10 (m, 6H), 2.71 (dt, J = 8.3, 4.8 Hz,
1H, H-1), 3.65 (br s, 1H, OH), 4.35 (dt, J = 5.3, 3.5 Hz, 1H,
H-2).
General procedure of the biotransformation
The nitrile hydratase from Rhodococcus equi A4 (10) was
diluted with KH2PO4–Na2HPO4 buffer (54 mM, pH 7.5) to a
concentration of 22 g of protein mL–1 (42 g of pro-
tein mL–1 for cis-2a) and the substrates were added from
stock solutions in methanol to make up final concentrations
of 1.25 mM of trans-1a, cis-1a, and cis-2a and 5 mM of
trans-3a, cis-3a, trans-4a, and cis-4a. Methanol did not ex-
ceed 5% (v/v) of the reaction mixture. The reactions were
carried out in shaken vessels (850 rpm, Thermomixer Com-
pact Eppendorf) at 30°C. At intervals, samples were with-
drawn, the reaction was quenched with 1 M HCl (0.1 mL
per 1 mL of sample) and the precipitated protein was re-
moved by centrifugation. The supernatants were analyzed by
HPLC as described below or by TLC on silica gel plates
(Merck) developed with chloroform–ethyl acetate (10:1) (for
substrates trans-3a, cis-3a, trans-4a, and cis-4a); the spots
were visualized by charring with 5% sulphuric acid in etha-
nol.
cis-2b
1H NMR d: 1.27 (m, 1H, H-6a), 1.37 (m, 1H, H-4a), 1.38
(m, 1H, H-5a), 1.52 (d, J = 12.2 Hz, 1H, H-2a), 1.74 (m, 1H,
H-6b), 1.86 (m, 1H, H-5b), 2.01 (m, 1H, H-4b), 2.08 (m, S J =
23.6 Hz, 1H, H-2b), 2.30 (dd, J = 3.5, 11.9 Hz, 1H, H-1),
4.88 (dd, J = 4.4, 11.0 Hz, 1H, H-3), 6.73 (br s, 1H, NH),
7.25 (br s, 1H, NH), 7.52 (m, 2H, H-meta), 7.65 (m, 1H,
H-para), 7.95 (m, 2H, H-ortho). 13C NMR d: 23.07 (t, C-5),
28.19 (t, C-6), 31.01 (t, C-4), 34.33 (t, C-2), 41.97 (d, C-1),
72.92 (d, C-3), 128.71 (d, C-meta), 129.10 (d, C-ortho),
130.13 (s, C-ipso), 133.22 (d, C-para), 165.11 (s, OCOPh),
175.75 (s, CONH2).
trans-3b
1H NMR d: 1.01–1.24 (m, 3H, H-3a, H-4a, H-5a), 1.27
(m, 1H, H-6a), 1.54–1.72 (m, 3H, H-4b, H-5b, H-6b), 1.81
(m, 1H, H-3b), 1.94 (ddd, J = 3.7, 9.8, 12.2 Hz, 1H, H-1),
3.46 (m, 1H, H-2), 4.57 (d, J = 5.1 Hz, 1H, OH), 6.61 (br s,
1H, NH), 7.13 (br s, 1H, NH). 13C NMR d: 24.39 (t, C-4),
24.84 (t, C-5), 28.73 (t, C-6), 35.16 (t, C-3), 52.10 (d, C-1),
69.80 (d, C-2), 176.67 (s, CONH2).
Analytical HPLC
The concentrations of substrates and products were deter-
mined using an HPLC system consisting of a sol-
vent-delivery system 600 (Waters), a photo-diode array
detector 996 (Waters), and a refractive-index detector 2410
(Waters) and a Nova-Pak C18 column (5 m, 3.9 × 150 mm,
Waters). As mobile phases, 50% (v/v) acetonitrile (for com-
pounds trans-1a, trans-1b, cis-1a, cis-1b, cis-2a, and cis-2b)
or 10% (v/v) acetonitrile (for other compounds) in water
trans-4b
1H NMR d: 1.42 (m, 1H, H-3a), 1.50–1.64 (m, 3H, H-4a,
H-4b, H-5a), 1.68–1.83 (m, 2H, H-3b, H-5b), 2.39 (m, 1H,
© 2002 NRC Canada

Martínková et al.
727
6. J.C.T. Dias, R.P. Rezende, C.A. Rosa, M.-A. Lachance, and
V.R. Linardi. Can. J. Microbiol. 46, 525 (2000).
7. M.X. Wang and G.C. Feng. Tetrahedron Lett. 41, 6501 (2000).
8. Y. Tani, M. Kurihara, H. Nishise, and K. Yamamoto. Agric.
Biol. Chem. 53, 3143 (1989).
H-1), 4.08 (m, 1H, H-2), 6.67 (br s, 1H, NH), 7.20 (br s, 1H,
NH). 13C NMR d: 22.50 (t, C-4), 27.90 (t, C-5), 34.96 (t,
C-3), 53.01 (d, C-1), 75.42 (d, C-2), 176.54 (s, CONH2).
9. O. Meth-Cohn and M.-X. Wang. Chem. Commun. 1041 (1997).
10. I. PÍepechalová, L. Martínková, A. Stolz, M. Ovesná, K.
Bezouška, J. Kopecký, and V. KÍen. Appl. Microbiol.
Biotechnol. 55, 150 (2001).
11. G. Bernáth, K.L. Láng, G. Göndös, P. Márai, and K. Kovács.
Acta Chim. Hung. 74, 479 (1972).
12. D.S. Noyce, B.N. Bastian, P.T.S. Lau, R.S. Monson, and B.
Weinstein. J. Org. Chem. 34, 1247 (1969).
13. R. Krieg, M. Schierhorn, H. Altmann, and H.J. Deutscher. J.
Prakt. Chem. 329, 1123 (1987).
Acknowledgments
Financial support of this work from the Grant Agency of
the Academy of Sciences of the Czech Republic (project
A4020802/1998), by the Grant Agency of the Czech Repub-
lic (project 524/00/1275), by the grant AKTION 17p12 from
the Austrian and Czech Ministries of Education and by the
Österreichische Nationalbank (project No. 7241) is grate-
fully acknowledged. Authors wish to thank Dr. Petr Sedmera
for critically reading the manuscript and Ms. Magdalena
Balvínová for her excellent technical assistance.
14. J.P. Aycard, R. Lafrance, and B. Boyer. Can. J. Chem. 57,
2823 (1979).
15. M.P. Mertes, A.A. Ramsey, P.E. Hanna, and D.D. Miller. J.
Med. Chem. 13, 789 (1970).
16. H. Hönig, P. Seufer-Wasserthal, and F. Fülöp. J. Chem. Soc.
Perkin Trans. 1, 2341 (1989).
17. P. Kozikowski and M. Adamczyk. J. Org. Chem. 48, 366 (1983).
18. F. Fülöp, I. Huber, G. Bernáth, H. Hönig, and P.
Seufer-Wasserthal. Synthesis, 43 (1991).
19. K. Praefcke and D. Schmidt. Z. Naturforsch. 35b, 1451 (1980).
20. de Raadt, N. Klempier, K. Faber, and H. Griengl. J. Chem.
Soc. Perkin Trans. 1, 137 (1992).
21. P.A. Wade and J.F. Bereznak. J. Org. Chem. 52, 2973 (1987).
22. E. Forró, K. Lundell, F. Fülöp, and L.T. Kanerva. Tetrahedron:
Asymmetry, 8, 3095 (1997).
References
1. T. Sugai, T. Yamazaki, M. Yokoyama, and H. Ohta. Biosci.
Biotech. Biochem. 61, 1419 (1997).
2. Y. Tani, M. Kurihara, and H. Nishise. Agric. Biol. Chem. 53,
3151 (1989).
3. K. Yamamoto, Y. Ueno, K. Otsubo, H. Yamane, K.-I.
Komatsu, and Y. Tani. J. Ferment. Bioeng. 73, 125 (1992).
4. K. Matoishi, A. Sano, N. Imai, T. Yamazaki, M. Yokoyama, T.
Sugai, and H. Ohta. Tetrahedron: Asymmetry, 9, 1097 (1998).
5. N. Klempier, A. de Raadt, K. Faber, and H. Griengl. Tetrahe-
dron Lett. 32, 341 (1991).
© 2002 NRC Canada