Synthesis and characterisation of novel 3′-O- and 5′-O- modified azobenzene-thymidine phosphoramidites and their oligonucleotide conjugates as colorimeter DNA probes and FRET quenchers

Article abstract of DOI:10.1016/j.tetlet.2003.09.115

The synthesis of two new 'first generation' azobenzene based thymidine phosphoramidites 1 and 2 having the chromophore (DABCYL) covalently incorporated as an ester in the 3′-O- and 5′-O-positions of the deoxyribose, and the incorporation of these molecules into 16-mer Chronic Myeloid Leukaemia (CML) antisense oligonucleotides, giving 7 and 8, respectively, is described. These compounds were designed as highly coloured probes, and to participate in a Fluorescence Resonance Energy Transfer (FRET) mechanism in the design of novel molecular beacons.

Full text of DOI:10.1016/j.tetlet.2003.09.115

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 8571–8575
Synthesis and characterisation of novel 3%-O- and 5%-O- modified
azobenzene-thymidine phosphoramidites and their oligonucleotide
conjugates as colorimeter DNA probes and FRET quenchers
Thorfinnur Gunnlaugsson,a,* John M. Kelly,a,* Mark Nieuwenhuyzenb and Aoife M. K. O’Briena
aDepartment of Chemistry, Trinity College Dublin, Dublin 2, Ireland
bSchool of Chemistry, Queen’s University of Belfast, Belfast BT9 5AG, Northern Ireland, UK
Received 20 August 2003; revised 1 September 2003; accepted 16 September 2003
Abstract—The synthesis of two new ‘first generation’ azobenzene based thymidine phosphoramidites 1 and 2 having the
chromophore (DABCYL) covalently incorporated as an ester in the 3%-O- and 5%-O-positions of the deoxyribose, and the
incorporation of these molecules into 16-mer Chronic Myeloid Leukaemia (CML) antisense oligonucleotides, giving 7 and 8,
respectively, is described. These compounds were designed as highly coloured probes, and to participate in a Fluorescence
Resonance Energy Transfer (FRET) mechanism in the design of novel molecular beacons.
© 2003 Elsevier Ltd. All rights reserved.
The design, synthesis, and biological evaluation of
modified nucleotides and oligonucleotides has attracted
substantial interest in recent years.1 Such systems are
promising candidates for the development of novel
nucleic acid targeting therapeutic drugs for the anti-
sense approach,2 in homogeneous DNA sequence tech-
niques,3 as tools for gene-sequencing,4 as molecular
probes,5 and in particular as molecular beacons.6
Recently, Komiyama et al. conjugated simple azobenz-
ene units into short oligonucleotides as side chains.
Further they have incorporated these within the strands
using prochiral diol7 or D- or L-threoninol8 based
spacers, which were used to gain ‘external control’
(cis–trans isomerisation) over duplex and triplex DNA
formation [demonstrated by melting temperature (Tm)
experiments].8 Even though these compounds are highly
desirable, such modifications alter the local oligonucleo-
tide backbone structure as they lack both the deoxy-
ribose unit and the nucleic acid bases.9 Furthermore,
these azobenzene dyes only absorb strongly in the UV,
a drawback for their use in molecular probes or as
fluorescent quenchers.10 We are particularly interested
in the development of metal based nucleic acid target-
ing molecules. To this end we have developed several
ruthenium-based antisense oligonucleotide conjugates
capable of achieving site specific photosensitised cleav-
age of the bcr-abl signature sequence of the Philadel-
phia chromosome associated with Chronic Myeloid
Leukaemia (CML),11 as well as lanthanide based
ribonuclease mimics for phosphodiester cleavage of
mRNA.12 In addition, we have recently made several
highly coloured azobenzene based chemosensors for
Na+ and K+, where distinct colour changes occurred
upon ion recognition (which takes place through modu-
lation of the Internal Charge Transfer, ICT, mechanism
of the molecule).13
Figure 1. The structures of the two phosphoramidites 1 and
2.
* Corresponding authors. Tel.: +00 353 1 608 3459; fax: +00 353 1
671 2826; e-mail: gunnlaut@tcd.ie
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.09.115

8572
T. Gunnlaugsson et al. / Tetrahedron Letters 44 (2003) 8571–8575
Herein, we describe the synthesis of two nucleoside
phosphoramidites 1 and 2 (Fig. 1) functionalised with
the well known DABCYL (4-[4-dimethylaminophenyl-
azo] benzoic acid, Fig. 2)14 chromophore covalently
incorporated as an ester at the 3%-O- and 5%-O-posi-
tions. We also describe the incorporation of 1 and 2
into a 16-mer CML antisense oligonucleotide at the
5%-terminus (i.e. giving rise to ‘back to front’ derivative
as in the case of 1, or the more conventional ‘front to
front’ analogue 2) as potential molecular probes for
nucleic acids, and some preliminary analysis of these
conjugates using melting temperature analysis and CD
spectroscopy.
The reason for this particular design employed for 1
and 2 is threefold. Firstly, the DABCYL chromophore
(which has very low fluorescence quantum yield)13 is
able to participate in Fluorescence Resonance Energy
Transfer (FRET) as an excited state quencher, which is
important in developing novel molecular beacons. Sec-
ondly, we predict that even though the two molecules
might exert some slight influence on the structure of the
oligonucleotides (‘back to front’ versus ‘front to front’)
they should not perturb double helix formation and
therefore not alter greatly their physical properties such
as the melting temperatures (Tm).7,8 Thirdly, these com-
pounds, being strongly absorbing in the visible region
can be used for visual detection or colorimetric probes
for oligonucleotides.14 To the best of our knowledge
these are the first examples of strongly visually
absorbing azobenzene based oligonucleotide conjugates
directly incorporated into the ribose unit of a thymidine
nucleoside.
Scheme 1. Synthesis of 1 and 2. (i) Triphenylmethyl (trityl)
chloride, pyridine, 30 min; (ii) DABCYL, EDCl, DMAP,
THF, rt, 48 h; (iii) 80% AcOH, reflux 30 min; (vi) 2-cyano-
ethyldiisopropyl-chlorophosphoramidite, DIPEA, CH2Cl2, 30
min; (v) DABCYL, EDCl, DMAP, THF; (vi) 2-cyanoethyldi-
isopropyl-chlorophosphoramidite, DIPEA, CH2Cl2, 30 min.
The synthesis of 1 and 2 is shown in Scheme 1. DAB-
CYL was synthesised in accordance with published
procedure.15 Our initial synthetic strategy was to use
the N-succinimide activated ester of DABCYL 3 and to
incorporate it into either a 3%-O- and 5%-O-protected
form of thymidine 4. However, this initial attempt was
unsuccessful. Therefore, a new approach was adopted
for the synthesis of 1. Thymidine 4 was first protected
as its 5%-O-tritylthymidine, followed by coupling to
DABCYL (in its acid form), using EDCl and DMAP in
anhydrous THF at room temperature for 48 h. The
intermediate was purified using silica flash column
chromatography (CH2Cl2:EtOAc, 70:30) and isolated in
63% yield as a red coloured residue which crystallised
upon standing. This was followed by the deprotection
of the 5%-O-trityl group using 80% acetic acid under
reflux for 30 min. The resulting thymidine–DABCYL
conjugate 5 was purified by flash column chromatogra-
phy (CH2Cl2:EtOAc, 70:30) and isolated in 55% yield
as an orange coloured powder which when recrys-
tallised from CH3CN provided red crystals suitable for
X-ray crystallographic analysis.16,† The structure of 5
(Fig. 3) clearly shows the azobenzene in its expected
Figure 3. The X-ray crystal structure of 5 showing the
azobenzene chromophore in its trans-configuration. A water
molecule can also be seen hydrogen bonding to the 5% hydroxy
group. Hydrogen atoms have been omitted for clarity.
† Data were collected a Bruker-AXS SMART diffractometer using
the SAINT-NT16a software with graphite monochromated Mo-Ka
radiation. A crystal was mounted on to the diffractometer at low
temperature under nitrogen at ca. 120 K. The structures were solved
using direct methods and refined with the SHELXTL version 516b
and the non-hydrogen atoms were refined with anisotropic thermal
parameters. Additional material available from the Cambridge
Crystallographic Data Centre (CCDC No. 220683) comprises rele-
vant tables of atomic coordinates, bond lengths and angles, and
thermal parameters.
Figure 2. DABCYL (R=H) and the corresponding succin-
imide ester 3.

T. Gunnlaugsson et al. / Tetrahedron Letters 44 (2003) 8571–8575
8573
trans configuration with the angle between the aromatic
rings measured as 21°. The crystals contained a single
water molecule in the unit cell, which were hydrogen
bonded to themselves and the 5%-hydroxyl group of 5
forming an infinite hydrogen-bonded chain along the
b-axis. These chains are packed together such that they
form an interdigitated array that is held together by a
combination of edge-to-face and p–p interactions in the
a–c plane forming intricate three-dimensional hydrogen
bonded structure.
man 1000M). The chosen strand is one we have previ-
ously investigated in our antisense programme for
targeting CML.11 This yielded two antisense strands
where our modified thymidine oligonucleotides [marked
as either (1) or (2) in 7 and 8, respectively] would give
rise to two different conformational environments
within these strands.
5%-(1)-TCTTCCTTATTGATGG-3%
5%-(2)-TCTTCCTTATTGATGG-3%
7
8
Phosphoramidite 1 was formed using a standard
method by stirring 5 at 0°C with 2-cyanoethyldiiso-
propylchloro-phosphoramidite in CH2Cl2 in the pres-
ence of DIPEA, for 30 min.17 The resulting mixture was
then allowed to warm to room temperature and stirred
for a further 30 min. Subsequently, 1 was purified by
flash column chromatography (CH2Cl2:EtOAc, 50:50)
and isolated in 73% yield. The purity of this compound
was verified by reverse phase HPLC (C18), in CH3CN,
which showed two peaks after 3.1 and 3.8 min. respec-
tively in 40:60 ratio (ESI) which were assigned to the
two diastereoisomers of 1. The existence of such iso-
mers in phosphoramidite chemistry is well known,
caused by the introduction of the phosphorus moiety.18
The preparation of 7 and 8 involved dissolving the
precursors 1 and 2 in a 50:50 mixture of dry
CH3CN:pyridine and incorporating these in a large
excess onto the 16-mer to ensure efficient coupling.
These were removed from the solid support using con-
centrated ammonia (with no noticeable ester hydrolysis
of the DABCYL esters under these conditions), and
purified by semi-preparative reverse phase HPLC
(C18), using gradient elution (20%B“60%B over 15
min where B=CH3CN, and A=0.1M TEAA, pH 6.9),
which enabled the isolation of 7 or 8 after 10 min. The
UV–vis spectra of the peak eluting at 10.8 min showed
absorption bands at 260 nm, characteristic of DNA,
and a second absorption at 459 nm corresponding to
the DABCYL chromophore. The integration of the
absorption bands were in the correct ratio expected for
the additive UV–vis spectra of the 17-mer and DAB-
CYL, indicating that 7 and 8 had been successfully
synthesised (Fig. 4).
1
This was also confirmed by H NMR (ESI) and 31P
NMR which showed two peaks at 149.8 and 149.4
ppm. The ESMS of 1 showed the presence of a peak at
694.3093 (calcd for C34H45N7O7P: 694.3118) for M+H+
and a peak at 716.3 for the M+Na+.
As in the case of 1, we initially attempted to make 2 by
protecting the 3%-O hydroxy group. Several different
methods were employed, including the use of acetyl,
benzyl or tertbutylsilyl ether, followed by coupling to
DABCYL. Unfortunately, the removal of the protect-
ing group proved troublesome and often led to the
decomposition of the product. Because of these draw-
backs, we attempted to add the DABCYL moiety
directly to 4, without any protection, on the basis that
the 5%-hydroxy group should be more reactive than the
3%-hydroxy group, and this should give the desired
product in higher yield. This was indeed found to be
the case, and the 5%-O- functionalised azobenzene 6 was
obtained in 30% yield after purification by flash column
chromatography (EtOAc:DIPEA, 199:1). The structure
of 6 was also determined by X-ray crystallography, and
showed the trans isomer as previously seen for 5.
In order to investigate the effect of 7 and 8 after
incorporation into a double stranded DNA, these
modified 17-mer oligonucleotides were annealed to
a
34-mer (5%-TGACCATCAATAAGGAAGAAGC-
CTTCAGCGGCC-%3)11 target CML oligonucleotide at
80°C over 30 min. The Tm of these double stranded
oligonucleotides were then measured and determined
using a derivative plot, to be 52.0°C. Importantly, when
the Tm was measured experimentally for the same 17-
mer-34-mer duplex lacking the azobenzene moiety, it
was determined to be 51.0°C. This result strongly sug-
In an analogous fashion to 1, compound 2 was formed
by reacting 6 with 2-cyanoethyldiisopropylchlorophos-
phoramidite in DCM in the presence of DIPEA, yield-
ing 2 in 80% yield as a red-coloured residue, which was
1
characterised using conventional methods. Both the H
and 31P NMR showed the presence of two diastereoiso-
mers and the ESMS showed two peaks for M+H+
(found: 694.3062) and M+Na+, respectively, indicating
that 2 was successfully formed.
The two phosphoramidites 1 and 2 were incorporated
into a 16–mer oligonucleotide (shown below) using
standard solid phase phosphoramidite chemistry (200
nmol scale) on an automated DNA synthesiser (Beck-
Figure 4. The UV–vis spectra of 7 showing the absorption of
the DABCYL chromophore between 350 and 580 nm, and
the absorption of the DNA oligonucleotides component at ca.
260 nm.

8574
T. Gunnlaugsson et al. / Tetrahedron Letters 44 (2003) 8571–8575
gests that the incorporation of the azobenzene chro-
mophore does not adversely affect the stability of the
double helix, indicating that this does not impose sig-
nificant physical changes on the hybridised strand. This
is important for the application of 7 or 8 as probes or
quenchers in molecular beacons. Furthermore, even
though the CD spectra of 7 or 8 were somewhat
different prior to duplex formation, only minor changes
were observed to that of the non-functionalised 17-mer,
after annealing with the 34-mer target, again indicating
only small conformational effect.
(CH3)2)2), 2.53 (2H, m, C(2)H2), 1.69 (3H, s, -N-
CHꢀC(CH3)-), 1.24 (12H, m, -N(CH(CH3)2)2); lC (100
MHz, d6-acetone) 165.9, 163.8, 156.6, 153.9, 150.8,
143.9, 136.1, 131.1, 130.3, 125.9, 122.5, 118.7, 112.0,
110.8, 85.4, 84.1, 83.7, 74.1, 64.7, 60.2, 59.0, 43.6, 39.2,
24.5, 20.5, 14.1; lP (162 MHz, d6-acetone) 149.5, 149.3;
m/z 694 (M+H); (m/z found: 694.3062, [M+H]+ calcu-
lated for C34H45N7O7P: 694.3118).
Acknowledgements
We also investigated the possibilities of 5 and 6 under-
going trans–cis photo isomerisation. However, we were
unable to record this process at room temperature in
solution. This is not unexpected, as it is known that the
trans–cis photo isomerisation for Methyl Red is very
rapid in solution. However, the trans-cis isomerisation
can for be observed in polymeric matrixes or in the
presence of cyclodextrins.19 Currently we are evaluating
the ability of 7 and 8 to undergo such isomerisation
after hybridisation to the above 34-mer CML strand.
This work was supported by the Kinerton Ltd., Trinity
College Dublin, and the Higher Education Authority in
Ireland through the HEA 1999 Research Programme
on Advanced Materials. AO’B would like to thank
Enterprise Ireland for a PhD studentship. We especially
thank Professor R. Jeremy H. Davies and Clarke S.
Stevenson, Department of Biology and Biochemistry,
Queen’s University of Belfast and Drs. John O’Brien,
Frederic Pfeffer and Susan Quinn (TCD) for their help
and support, and Dr. Hazel M. Moncrieff for helpful
discussion.
In summary, we have developed two novel azobenzene
conjugated oligonucleotides by coupling DABCYL into
thymidine at the 3%-O- and 5%-O-hydroxy moieties. To
the best of our knowledge these are the first examples
of such modified and highly coloured azobenzene–
thymidine conjugates, where the dye is directly attached
to the sugar unit. Currently we are evaluating the
properties of 7 and 8 as novel DNA probes for genetic
analysis as well components of molecular beacons. We
are also undertaking the incorporation of the DAB-
CYL into thymidine via the amide linkage. This will be
the subject of future publications.
References
1. (a) Willner, I. Acc. Chem. Res. 1997, 30, 347; (b) Kool, E.
T. Chem. Rev. 1997, 97, 1437; (c) Wnuk, S. F. Tetra-
hedron 1993, 49, 9877.
2. De Mesmaeker, A.; Ha¨ner, R.; Martin, P.; Moser, H. E.
Acc. Chem. Res. 1995, 28, 366.
3. Whitcombe, D.; Theaker, J.; Guy, S. P.; Brown, T.;
Little, S. Nature Biotech. 1999, 17, 804.
4. Nurmi, J.; Ylikoski, A.; Soukka, T.; Karp, M.; Lo¨vgren,
Characterisation of 1 and 2
T. Nuclec. Acid Res. 2000, 28, e28.
5. (a) Dobson, N.; McDowell, D. G.; French, D. J.; Brown,
L. J.; Mellor, J. M.; Brown, T. Chem. Commun. 2003,
1235; (b) Tan, W.; Fang, X.; Li, J.; Liu, X. Chem. Eur. J.
2000, 6, 1107.
6. (a) Tyagi, S.; Bratu, D. P.; Kramer, F. R. Nature Biotech.
1998, 16, 49; (b) Tayagi, S.; Kramer, F. R. Nature
Biotech. 1995, 14, 303; (c) Niemeyer, C. M.; Adler, M.
Angew. Chem., Int. Ed. 2002, 41, 3779; (d) Li, J. J.; Fang,
X.; Schuster, S. M.; Tan, W. Angew. Chem., Int. Ed.
2000, 39, 1049; (e) Fang, X.; Li, J. J.; Perlette, J.; Tan,
W.; Wang, K. Anal. Chem. 2000, 747A; (f) Brown, L. J.;
Cummins, J.; Hamilton, A.; Brown, T. Chem. Commun.
2000, 621.
7. (a) Yamazawa, A.; Liang, X.; Asanuma, H.; Komiyama,
M. Angew. Chem., Int. Ed. 2000, 39, 2356; (b) Asanuma,
H.; Liang, X.; Yoshida, T.; Yamazawa, A.; Komiyama,
M. Angew. Chem., Int. Ed. 2000, 39, 1316; (c) Asanuma,
H.; Ito, T.; Yoshida, T.; Liang, X.; Komiyama, M.
Angew. Chem., Int. Ed. 1999, 38, 2393.
8. (a) Asanuma, H.; Liang, X.; Yoshida, T.; Komiyama, M.
CHEMBIOCHEM 2001, 2, 39; (b) Takarada, T.;
Tamaru, D.; Liang, X.; Asanuma, H.; Komiyama, M.
Chem. Lett. 2001, 732.
1: Obtained in yield of 73% (0.086 g, 0.12 mmol) as a
red coloured solid; mp decomposition; lH (400 MHz,
CDCl3) 8.16 (2H, d, J=8.04 Hz, Ar), 7.91 (4H, m, Ar),
7.67 (1H, s, -N-CHꢀC(CH3)), 6.77 (2H, d, J=7.52 Hz,
Ar), 6.50 (1H, m, C(1)H), 5.60 (1H, d, J=6.0, C(3)H),
4.38 (1H, m, C(4)H), 4.12 ( 2H, m, -P-O-CH2-CH2),
3.95 (3H, m, -P-O-CH2-CH2), 3.65 (2H, m, C(5)H2),
3.12 (6H, s, -N(CH3)2), 2.68 (2H, m, -N(CH(CH3)2)2),
2.35 (2H, m, C(2)H2), 1.98 (3H, s, -N-CHꢀC(CH3)),
1.25 (12H, m, -N(CH(CH3)2)2); lC (100 MHz, d6-
DMSO) 165.8, 163.6, 156.3, 152.9, 150.4, 143.6, 135.5,
130.6, 128.9, 125.5, 122.0, 117.4, 111.4, 85.0, 84.6, 75.9,
64.2, 63.4, 58.7, 43.3, 38.0, 37.7, 24.7, 20.5, 14.1; lP (162
MHz, d6-DMSO) 149.8, 149.4; m/z 694.1 (M+H); m/z
found: 694.3093 ([M+H]+ calculated for C34H45N7O7P:
694.3118).
2: Obtained in a yield of 80% (0.132 g, 0.19 mmol) as a
red coloured solid; mp decomposition; lH (400 MHz,
d6-acetone) 8.22 (2H, d, J=8.04 Hz, Ar), 7.91 (4H, m,
Ar), 7.48 (1H, s, -N-CHꢀC(CH3)), 6.89 (2H, d, J=7.52
Hz, Ar), 6.37 (1H, m, C(1)H), 4.87 (1H, d, J=6.0 Hz,
C(3)H), 4.65 (2H, m, C(5)H2), 4.42 (1H, m, C(4)H),
3.89 (2H, m, -P-O-CH2-CH2), 3.71 (2H, m, -P-O-CH2-
CH2), 3.15 (6H, s, -N(CH3)2), 2.80 (2H, m, -N(CH-
9. (a) Asanuma, H.; Ito, T.; Komiyama, M. Tetrahedron
Lett. 1999, 40, 2671; (b) Asanuma, H.; Ito, T.;
Komiyama, M. Tetrahedron Lett. 1998, 39, 9015; (c)

T. Gunnlaugsson et al. / Tetrahedron Letters 44 (2003) 8571–8575
8575
Yamana, K.; Yoshikawa, A.; Nakano, H. Tetrahedron
Lett. 1996, 38, 637.
10. (a) May, J. P.; Brown, L. J.; Rudloff, I.; Brown, T. Chem.
Commun. 2003, 970; (b) Marras, S. A. E.; Kramer, F. R.;
Tyagi, S. Nucleic Acids Res. 2002, 30, e122.
Soc., Perkin Trans. 2 2002, 141; (c) Gunnlaugsson, T.;
Nieuwenhuyzen, M.; Richard, L.; Thoss, V. Tetrahedron
Lett. 2001, 42, 4725.
14. (a) Reynolds, R. A., III; Mirkin, C. A.; Letsinger, R. L.
J. Am. Chem. Soc. 2000, 122, 3795; (b) Mirkin, C. A.;
Storhoff, J. J. Chem. Rev. 1999, 99, 1849.
15. Zimmerman, G.; Chow, J.-Y.; Paik, U.-J. J. Chem. Phys.
1958, 80, 3528.
16. (a) SAINT-NT, program for data collection and data
reduction, Bruker-AXS, Madison, WI, 1998; (b) G. M.
Sheldrick, SHELXTL Version 5.0, A System for Struc-
ture Solution and Refinement, Bruker-AXS, Madison,
WI, 1998.
17. (a) Adams, S. P.; Kavka, K. S.; Wykes, E. J.; Holder, S.
B.; Galluppi, G. R. J. Am. Chem. Soc. 1983, 105, 661; (b)
Sinha, N. D.; Biernat, J.; Koster, H. Tetrahedron Lett.
1983, 24, 5843.
18. McBride, L. J.; Caruthers, M. H. Tetrahedron Lett. 1983,
24, 245.
11. (a) Crean, C. W.; Kavanagh, Y. T.; O’Keeffe, C. M.;
Lawler, M. P.; Stevenson, C.; Davies, R. J. H.; Boyle, P.
H.; Kelly, J. M. Photochem. Photobiol. Sci. 2002, 1, 1024;
(b) Ujj, L.; Coates, C. G.; Kelly, J. M.; Kruger, P. E.;
McGarvey, J. J.; Atkinson, G. H. J. Phys. Chem. B. 2002,
106, 4854; (c) O’Reilly, F. M.; Kelly, J. M. J. Phys.
Chem. B. 2000, 104, 7206.
12. (a) Gunnlaugsson, T.; Davies, R. J. H.; Nieuwenhuyzen,
M.; Stevenson, C. S.; O’Brein, J. E.; Mulready, S. Poly-
hedron 2003, 23, 711; (b) Gunnlaugsson, T.; Davies, R. J.
H.; Nieuwenhuyzen, M.; Stevenson, C. S.; Viguier, R.;
Mulready, S. Chem. Commun. 2002, 2136; (c)
Gunnlaugsson, T.; O’Brein, J. E.; Mulready, S. Tetra-
hedron Lett. 2002, 43, 8493.
13. (a) Gunnlaugsson, T.; Leonard, J. P. J. Chem. Soc.,
Perkin Trans. 2 2002, 1980; (b) Gunnlaugsson, T.;
Nieuwenhuyzen, M.; Richard, L.; Thoss, V. J. Chem.
19. Sanchez, A. M.; de Rossi, R. H. J. Org. Chem. 1996, 61,
3446.