Synthesis of an ethidium nucleoside and its acyclic analog

Article abstract of DOI:10.1016/S0040-4039(03)00025-X

The protected ethidium nucleosides 8-(3′,5′-di-O-benzoyl-2′-deoxy-D-ribofuranosyl)-3- acetamido-5-ethyl-6-phenyl-phenanthridinium (5), 8-(3′,5′-di-O-acetyl-2′-deoxy-D-ribofuranosyl)-3- acetamido-5-ethyl-6-phenyl-phenanthridinium (6), and the acyclic analog 8-[(3R)-1,3-dihydroxy-4-yl]-acetamido-3-amino-5-ethyl-6-phenyl- phenanthridinium (3) were prepared. Based on to their different stability, only the acyclic derivative 3 seems to be suitable for oligonucleotide synthesis. Furthermore, the acyclic ethidium nucleoside analog 3 exhibits comparable absorption and emission properties of the underivatized ethidium (1).

Full text of DOI:10.1016/S0040-4039(03)00025-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1685–1690
Synthesis of an ethidium nucleoside and its acyclic analog
Nicole Amann and Hans-Achim Wagenknecht*
Institute for Organic Chemistry and Biochemistry, Technical University of Munich, Lichtenbergstr. 4, D-85747 Garching,
Germany
Received 24 October 2002; accepted 19 December 2002
Abstract—The protected ethidium nucleosides 8-(3%,5%-di-O-benzoyl-2%-deoxy-
D-ribofuranosyl)-3-acetamido-5-ethyl-6-phenyl-
phenanthridinium (5), 8-(3%,5%-di-O-acetyl-2%-deoxy- -ribofuranosyl)-3-acetamido-5-ethyl-6-phenyl-phenanthridinium (6), and the
D
acyclic analog 8-[(3R)-1,3-dihydroxy-4-yl]-acetamido-3-amino-5-ethyl-6-phenyl-phenanthridinium (3) were prepared. Based on to
their different stability, only the acyclic derivative 3 seems to be suitable for oligonucleotide synthesis. Furthermore, the acyclic
ethidium nucleoside analog 3 exhibits comparable absorption and emission properties of the underivatized ethidium (1). © 2003
Elsevier Science Ltd. All rights reserved.
1. Introduction
tion of the nucleic acids (>300 nm).6 Ethidium (1) also
plays an important role with respect to the studies of
photoinduced processes in DNA.7 In order to investi-
gate the distance dependence of charge transfer in
DNA, modified oligonucleotides have been used con-
taining ethidium covalently attached to the 5%-end via
an alkyl linker.8 Herein, we want to present the synthe-
sis and properties of the ethidium nucleoside 2 and its
acyclic analog 3 (Scheme 1) towards potential building
blocks for oligonucleotide synthesis.
One of the interesting properties of planar polycyclic
aromatic molecules is their ability to intercalate
between two adjacent base pairs in duplex DNA which
was first proposed by Lerman.1 This type of DNA
binding is mediated by non-covalent stacking interac-
tions with nucleobases and often combined with hydro-
gen bonding or other types of interactions. Among
these intercalators, the highly polar or even charged
systems are the most potent ones with respect to drug
positioning and sequence preferences. Thus, it is
believed that the electrostatic energy plays a major role
in the intercalation process.2
2. Results and discussion
Early steps in the synthesis of ethidium (1) and its
derivatives were made by Walls et al.,9 Watkins et al.,10
and Berg et al.11 More recently, work by Schacht et al.
showed clearly that the amino group in position C-3 of
ethidium (1) is deactivated in comparison to the C-8
amino group due to the mesomeric stabilization of the
positive charge (Scheme 1).12 Thus in general, the C-8
amino group exhibits a higher reactivity towards any
modification. Hence, we expect that a regioselective
reaction of the C-8 amino group should be possible in
order to prepare the ethidium derivatives 2 and 3.
3,8-Diamino-5-ethyl-6-phenylphenanthridium (‘ethid-
ium’, 1) is one of the well known positively charged
intercalators for duplex DNA and has been widely used
as a fluorescent staining agent due to the significant
fluorescence enhancement which it exhibits upon inter-
calation.3 Ethidium (1) and its derivatives are also well
known as potent trypanocidal drugs.4 In order to evalu-
ate the specific features of ethidium (1) such as binding
affinity, base sequence specifity and fluorescence prop-
erties, different experimental approaches were tried in
the past. Among these are the synthesis and application
of bifunctional DNA intercalators bearing ethidium
derivatives,5 theoretical calculations,2 and the applica-
tion of the photochemical reactivity beyond the absorp-
8-(2%-Deoxy-D-ribofuranosyl)-3-acetamido-5-ethyl-6-
phenyl-phenanthridinium (2) contains the ethidium
moiety glycosidically linked to the furanose form
of 2%-deoxyribose. Glycosides with ethidium as an agly-
con, e.g. 3- and 8-N-glucuronosylethidium, have been
previously identified as the major metabolites of ethid-
ium.13 Although the structure of 2 represents an N,O-
acetal, or a glycosylamine, respectively, the stability of
Keywords: ethidium; nucleoside; stacking; intercalation.
* Corresponding
author.
Fax:
+49-(89)-289-13210;
e-mail:
wagenknecht@ch.tum.de
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(03)00025-X

1686
N. Amann, H.-A. Wagenknecht / Tetrahedron Letters 44 (2003) 1685–1690
Scheme 2. Synthesis of the protected ethidium nucleosides 5
and 6: (a) 1 (1.3 equiv.), pTsOH (0.1 equiv.), MeOH, 60°C, 24
h; (b) Ac2O (3.3 equiv.), pyridine, rt, 6 h, 10–15% (a+b).
Scheme 1. Deactivation of the amino group at C-3 of ethid-
ium (1) due to mesomeric stabilization of the positive charge.
Ethidium nucleoside 2 and its acyclic analog 3.
preparation. The ethidium glycoside 5 was character-
ized by UV/vis spectroscopy and ESI mass spectrome-
try.22 The structure of 5 was confirmed by NMR
spectroscopy, including the 2D techniques HMQC and
DQF-COSY. These NMR experiments clearly revealed
that the protected 2%-deoxyribofuranoside moiety is
covalently attached to the amino group in the C-8
position of the ethidium part. The a-/b-anomers of the
glycoside 5 showed a slightly different Rf value on TLC.
Additionally, the NH-HSQC spectrum of 5 (Fig. 1)
showed two cross-peaks for the NH-protons in the
C-8-position of the ethidium group representing the
two anomers, whereas the NH-protons in the C-3 posi-
tion which is acetylated showed only one cross-peak.
2 towards acid and base hydrolysis should be high
enough to guarantee the later incorporation into
oligonucleotides. This expectation is mainly based on
the assumption that a significant mesomeric stabiliza-
tion exists within the ethidium aromatic system. We
started the preparation of the 2%-deoxyribofuranoside 2
from the precursors 1,3,5-tri-O-acetyl-2-deoxy-
D-ribo-
D-ribo-
D-ribo-
furanose,14 1-chloro-3,5-di-O-benzoyl-2-deoxy-
furanose,15
and
3,5-di-O-benzoyl-2-deoxy-
furanose.16 A whole variety of different reactions condi-
tions were tried to perform this preparation including
the Vorbrueggen conditions for the preparation of
nucleosides,17 the Koenigs–Knorr conditions18 or Lewis-
acid catalyzed conditions19 for the preparation of gly-
cosides, and, finally, special reaction conditions for the
preparation of aromatic glycosylamines.20 In all of
these experiments none or only traces of the product
could be detected by different analytical techniques,
mainly HPLC-MS.
It turned out that a complete interpretation of the
aromatic ethidium resonances in the 1H NMR spectrum
is impossible due to the overlay with the resonances of
the protons of the two benzoyl groups in the positions
3% and 5% of the 2%-deoxyribose moiety of 5. Hence, we
synthesized additionally the corresponding ethidium
glycoside 6 from 3,5-di-O-acetyl-2%-D-deoxyribofuran-
ose (7) again by treatment with NH4Cl in MeOH and
subsequent acetylation. The yield of 6 was comparable
to the synthesis of 5. Using this compound, the substi-
tution pattern of the ethidium group could be investi-
gated through 2D NMR experiments (DQF-COSY)
and was confirmed.23
In conclusion, only the reaction of 3,5-di-O-benzoyl-2-
deoxy- -ribofuranose (4) with ethidium (1) by treat-
D
ment with the weak acid NH4Cl in MeOH21 gave the
desired product in a yield of 10–15% (Scheme 2) which
was high enough in order to characterize and test the
stability of this ethidium nucleoside. Detailed investiga-
tions of the conditions of this reaction mainly by HPLC
showed that the maximum amount of the desired
product 4 could be found after 24 h reaction time under
reflux. The yield could not be improved since degrada-
tion of the product was detected after 24 h by HPLC.
In order to facilitate the chromatographic purification,
the coupling product was acetylated directly after the
We checked the stability of the protected ethidium
nucleosides 5 and 6 towards the typical reaction condi-
tions during automated DNA synthesis and workup.
Among these, the most critical ones for the glycosides 5
and 6 were expected to be the treatment with acids and
strong bases. As expected due to the mesomeric reso-
nance stabilization, both compounds, 5 and 6, stayed
stable towards the presence of trifluoroacetic acid in

N. Amann, H.-A. Wagenknecht / Tetrahedron Letters 44 (2003) 1685–1690
1687
Figure 1. NH–HSQC of the ethidium nucleoside 5.
MeCN over at least 3 h which was investigated by
HPLC analysis. In contrast, treatment of 5 and 6 with
aq. NH4OH solution completely degraded the products
within 3 h by cleaving the glycosidic bond between the
ethidium and the 2%-deoxyribofuranose part. This frag-
ments could be identified by HPLC-MS. This observed
instability of the glycosylamines 5 and 6 is in agreement
with results by Carell et al.21 and clearly indicates that
they are not suitable as building blocks for the prepara-
tion of oligonucleotides with ethidium as an artificial
DNA base.
The coupling of the carboxylic acid 1028 with ethidium
(1) was performed by three different ways. First, the
carboxylic acid of 10 was activated with iso butyl
chloroformiat.29 Subsequent treatment of the mixed
anhydride 1130 with ethidium (1) in DMF at elevated
temperature gave the coupling product 1231 in good
yield (70%). Using this synthetic pathway, only 3% of
3,8-disubstituted ethidium side product were detected
by HPLC-MS. Alternatively, the direct coupling of the
carboxylic acid 10 with ethidium (1) via typical and
common peptide coupling procedures gave 12 in com-
parable yields (60–80%). We tried this coupling using
HATU/HOAt,32 and in the presence of DCC/DMAP12.
Using the latter method, the 3,8-disubstituted ethidium
side product was formed only in traces whereas the
HATU/HOAt-catalyzed coupling gave this side
product in higher amounts (14%). Finally, the benzyli-
dene group of 12 was removed in aq. HCl yielding the
ethidium derivative 3.
As an alternative, we designed 8-[(3%R)-1%,3%-dihydroxy-
4%-yl]-acetamido-3-amino-5-ethyl-6-phenyl-phenanthri-
dinium (3) which represents an acyclic analog of the
ethidium nucleoside 2. The (3R),5-dihydroxy pentanoic
acid part mimics the 2%-deoxyribose moiety of common
nucleosides. Ethidium derivatives with an amide bond
to the exocyclic amino function at C-8 have proven to
be stable during DNA workup conditions as shown by
Barton et al.24 We started the synthesis (Scheme 3) with
the enantiomerically pure (3R),5-dihydroxypentanoic
acid methyl ester 8 which can be prepared in a multi-
step process according to literature procedures.25 The
two hydroxy functions of 8 were protected by a benzyl-
idene group and the ester function of 926 was
hydrolyzed in aq. LiOH.27
We confirmed the structure of the ethidium derivative
333 by 2D NMR experiments (DQF-COSY and
HMQC) and by mass spectrometry. Most importantly,
in contrast to the glycosides 5 and 6, the acyclic deriva-
tive 3 stays stable for hours upon treatment with acids
and bases presenting the conditions of the DNA syn-

1688
N. Amann, H.-A. Wagenknecht / Tetrahedron Letters 44 (2003) 1685–1690
Figure 2. UV/vis absorption and steady-state emission spectra
of the ethidium derivative 3 in MeCN (—) and MeOH (---).
fluorescence in MeOH is significantly lower due to
quenching. In addition to the NMR characterization,
the absorption maximum of 3 in the area of 480–490
nm clearly indicates that only one exocyclic amino
group of the ethidium moiety has been derivatized. The
additional derivatization of the exocyclic amino group
in position C-3 of the ethidium part in 3 would result in
a hypsochromic shift of the absorption spectrum to the
area of 430–440 nm.34
3. Conclusion
In conclusion, we presented here the preparation of the
ethidium nucleosides 8-(3%,5%-di-O-benzoyl-2%-deoxy-
ribofuranosyl)-3-acetamido-5-ethyl-6-phenyl-phenan-
thridinium (5), 8-(3%,5%-di-O-acetyl-2%-deoxy- -ribofura-
D-
D
nosyl)-3-acetamido-5-ethyl-6-phenyl-phenanthridinium
(6), and the acyclic analog 8-[(3%R)-1%,3%-dihydroxy-4%-
yl]-acetamido-3-amino-5-ethyl-6-phenyl-phenanthri-
dinium (3). Due to their different stability, only the
acyclic derivative 3 is suitable as a potential building
block for oligonucleotide synthesis. Furthermore, the
acyclic ethidium nucleoside 3 exhibits the comparable
absorption and emission properties of the underivatized
ethidium (1).
Scheme 3. Synthesis of the acyclic analog 3: (a) Ph-
CH(OCH3)2 (1.0 equiv.), pTsOH (0.1 equiv.), CH2Cl2, 40°C,
2 h, 63%; (b) LiOH, THF:H2O=1:1, rt, 12 h, 64%; (c)
4-ethylmorpholine (1.1 equiv.), isobutyl chloroformiate (1.1
equiv.), THF, rt, 10 min, quant; (d) 1 (1.0 equiv.), DMF,
70°C, 48 h, 70%; (e) 1 (0.9 equiv.), HATU (1.0 equiv.), HOAt
(1.0 equiv.), DMF, rt, 12 h, 80%; (f) 1 (1.0 equiv.), DCC (2.4
mmol), DMAP (0.5 equiv.), DMF, 70°C, 48 h, 60%; (g)
pTsOH (cat.), MeOH, rt, 30 min, 89%.
Acknowledgements
thesizer cycle and workup. With respect to this stability,
the ethidium derivative 3 is suitable in terms of a
potential building block for oligonucleotide synthesis.
This work was supported by the Deutsche Forschungs-
gemeinschaft and the Volkswagen-Stiftung. N.A. and
H.A.W. are grateful to Professor Horst Kessler, Tech-
nical University of Munich, for the generous support.
We thank Melina Haupt for measuring the NH-HSQC
of 5.
In order to evaluate the absorption and emission prop-
erties of the ethidium derivative 3, we finally measured
the UV/vis absorption and the steady-state fluorescence
of 3 in MeCN and MeOH, representing two typical
organic solvents without or with hydrogen-bonding
capabilities, respectively (Fig. 2). Whereas the UV/vis
absorption is very similar in both solvents, the fluores-
cence intensity is significantly different. In MeCN, the
integrated emission of 3 is 3.7 fold higher than in
MeOH, although the emission maxima reflect only a
small difference. This result stands in agreement with
the well-known electronic properties of ethidium (1)
meaning that the quantum yield of the ethidium
References
1. Lerman, L. S. J. Mol. Biol. 1961, 3, 18–30.
2. (a) Misra, V. K.; Honig, B. Proc. Natl. Acad. Sci. USA
1995, 92, 4691–4695; (b) Patterson, S. E.; Coxon, J. M.;
Stekowski, L. Bioorg. Med. Chem. 1997, 5, 277–281; (c)
Medhi, C.; Mitchell, J. B. O.; Price, S. L.; Tabor, A. B.

N. Amann, H.-A. Wagenknecht / Tetrahedron Letters 44 (2003) 1685–1690
1689
Biopolymers 1999, 52, 84–93; (d) Reha, D.; Kabela´c, M.;
(c) Bhat, C. C. In Synthetic Procedures in Nucleic Acids
Chemistry: Vol. 1; Zorbach, W. W.; Tipson, R. S., Eds.;
John Wiley and Sons: New York, 1968; pp. 521–522; (d)
Gold, A.; Sangaiah, R. Nucleosides & Nucleotides 1990, 9,
907–912.
&
&
Ryja´cek, F.; Sponer, J.; Sponer, J. E.; Elstner, M.; Suhai,
S.; Hobza, P. J. Am. Chem. Soc. 2002, 124, 3366–3376.
3. (a) Waring, M. J. J. Mol. Biol. 1965, 13, 269–282; (b)
LePecq, J.-B.; Paleotti, C. J. Mol. Biol. 1967, 27, 87–106;
(c) Olmsted, J.; Kearns, D. Biochemistry 1977, 16, 3647–
3654.
15. Hoffer, M. Chem. Ber. 1960, 93, 2777–2781.
16. Marriott, D. P.; Bantick, J. R. Tetrahedron Lett. 1981,
22, 3657–3658.
17. (a) Vorbru¨ggen, H.; Krolikiewicz, K.; Bennua, B. Chem.
Ber. 1981, 114, 1234–1255; (b) Vorbru¨ggen, H.; Ho¨fle, G.
Chem. Ber. 1981, 114, 1256–1268; (c) Vorbru¨ggen, H.;
Bennua, B. Chem. Ber. 1981, 114, 1279–1286.
4. (a) Tettey, J. N. A.; Skellern, G. G.; Grant, M. H.;
Midgley, J. M. J. Pharm. Biomed. Anal. 1999, 21, 1–7; (b)
Tettey, J. N. A.; Skellern, G. G.; Midgley, J. M.; Grant,
M. H.; Pitt, A. R. Chemico-Biol. Interact. 1999, 123,
105–115.
5. (a) Gaugain, B.; Barbet, J.; Oberlin, R.; Roques, B. P.;
Le Pecq, J.-B. Biochemistry 1978, 17, 5071–5078; (b)
Gaugain, B.; Barbet, J.; Capelle, N.; Roques, B. P.; Le
Pecq, J.-B. Biochemistry 1978, 17, 5078–5088; (c)
Kuhlmann, K. F.; Mosher, C. W. J. Med. Chem. 1981,
24, 1333–1337; (d) Lopp, I. G.; Wilson, W. D.; Boykin,
D. W. J. Heterocycl. Chem. 1982, 19, 695–696.
6. (a) Benimetskaya, L.; Bulychev, N. V.; Kozionov, A. L.;
Koshkin, A. A.; Lebedev, A. V. M.; Novozhilov, S. Y.;
Stockman, M. I. Biopolymers 1989, 28, 1129–1147; (b)
Krishnamurthy, G.; Polte, T.; Rooney, T.; Hogan, M. E.
Biochemistry 1990, 29, 981–988; (c) Gritsan, N. P.;
Koshkin, A. A.; Denisov, A. Y.; Markushin, Y. Y.;
Cherepanova, E. V.; Lebedev, A. L. J. Photochem. Pho-
tobiol. B: Biol. 1997, 37, 40–51.
7. (a) Fromherz, P.; Rieger, B. J. Am. Chem. Soc. 1986, 108,
5361–5362; (b) Atherton, S. J.; Beaumont, P. C. J. Phys.
Chem. 1987, 91, 3993–3997; (c) Pasternack, R. F.; Cac-
cam, M.; Keogh, B.; Stephenson, T. A.; Williams, A. P.;
Gibbs, E. J. J. Am. Chem. Soc. 1991, 113, 6835–6840; (d)
Dunn, D. A.; Lin, V. H.; Kochevar, I. E. Biochemistry
1992, 31, 11620–11625; (e) Brun, A. M.; Harriman, A. J.
Am. Chem. Soc. 1992, 114, 3656–3660; (f) Atherton, S. J.;
Beaumont, P. C. J. Phys. Chem. 1995, 99, 12025–12029;
(g) Jin, W. J.; Wei, Y. S.; Liu, C. S.; Shen, G. L.; Yu, R.
18. Whistler, R. L.; Wolfrom, M. L., Eds. Methods in Carbo-
hydrate Chemistry: Vol. 2; Academic Press: New York,
1963; pp. 221–222, 332–335.
19. For reference example see: Rokos, H.; Pfleiderer, W.
Chem. Ber. 1971, 104, 748–769. Other Lewis acids, such
as BF3, SnCl4, and TMSOTf, were also tried.
20. Takeuchi, Y.; Chang, M.-r.; Hashigaki, K.; Yamato, M.
Chem. Pharm. Bull. 1991, 39, 1629–1631.
21. Schwo¨gler, A.; Gramlich, V.; Carell, T. Helv. Chim. Acta
2000, 83, 2452–2463.
22. Data of 5: Rf=0.75 (silica gel, EtOAc:MeOH:H2O=
6:2:1). UV/vis (MeCN): umax=219, 292, 472 nm. 1H
signals were assigned by 2D experiments (DQF-COSY).
1H NMR (500 MHz, [d4]-MeOH): d=1.55 (t, 3 H,
CH2CH3), 1.90 (s, 3 H, NHCOCH3), 2.35–2.10 (m, 2 H,
H-2/2%), 4.20–4.05, 3.80–3.70, 3.65–3.50 (m, 3 H, H-4/5/
5%), 4.65 (q, 2 H, CH2CH3), 4.85–4.25 (m, 2 H, H-3%),
6.62, 6.54 (2d, 1 H, H-1%), 8.00–7.20 (m, 15 H, aromatic
H), 7.70 (d, 1 H, H-7), 7.90 (s, 1 H, H-2), 8.15 (d, 1 H,
H-9), 8.65 (d, 1 H, H-10), 8.75 (d, 1 H, H-1), 9.00 ppm (s,
1 H, H-4). NH-HSQC (500 MHz, [d3]-MeOH): d=7.64/
108.4 (8-NH), 7.51/90.7 (3-NH), 6.76/108.4 ppm (8-NH).
ESI-MS: m/z=680.5 (100%) [M]+.
23. Data of 6: Rf=0.70 (silica gel, EtOAc:MeOH:H2O=
6:2:1). UV/vis (MeCN): umax=217, 291, 474 nm. 1H
signals were assigned by 2D experiments (DQF-COSY).
1H-NMR (500 MHz, [d4]-MeOH): d=1.55 (t, 3 H,
CH2CH3), 2.05, 2.00, 1.85, 1.80 (2s, 6 H, COCH3), 2.20
(s, 3 H, NHCOCH3), 3.30–3.15 (m, 2 H, H-2/2%), 4.15–
3.80, 3.65–3.45 (m, 3 H, H-4/5/5%), 4.65 (q, 2 H,
CH2CH3), 5.20–4.95 (m, 2 H, H-3%), 6.56, 6.48 (2d, 1 H,
H-1%), 7.80–7.55 (m, 5 H, aromatic H), 7.65 (d, 1 H, H-7),
7.90 (s, 1 H, H-2), 8.10 (d, 1 H, H-9), 8.65 (d, 1 H, H-10),
8.75 (d, 1 H, H-1), 9.00 ppm (s, 1 H, H-4). ESI-MS:
m/z=557.5 (100%) [M]+.
24. Kelley, S. O.; Holmlin, R. E.; Stemp, E. D. A.; Barton, J.
K. J. Am. Chem. Soc. 1997, 119, 9861–9870.
25. Loubinoux, B.; Sinnes, J.-L.; O’Sullivan, A. C.; Winkler,
T. Tetrahedron 1995, 51, 3549–3558.
26. Preparation of the starting material 9 was performed
according to the procedure for the protection of the
starting material 8 by an isopropylidene group: Blechert,
S.; Dollt, H. Liebigs Ann. 1996, 2135–2140. Data of 9:
Rf=0.95 (silica gel, MeOH). 1H NMR (250 MHz,
CDCl3): d=1.55–1.61, 1.76–1.90 (m, 2 H, H-5), 2.45–2.69
(m, 2 H, H-2), 3.65 (s, 3 H, OCH3), 3.93–3.98 (m, 2 H,
H-4), 4.17–4.29 (m, 1 H, H-3), 5.49 (s, 1 H, Ph-CH),
7.27–7.58 ppm (m, 5 H, aromatic H).
Q. Spectrochim. Acta
A 1997, 53, 2701–2707; (h)
Kononov, A. I.; Moroshkina, E. B.; Tkachenko, N. V.;
Lemmetyinen, H. J. Phys. Chem. B 2001, 105, 535–541;
(i) Henderson, P. T.; Boone, E.; Schuster, G. B. Helv.
Chim. Acta 2002, 85, 135–151.
8. (a) Kelley, S. O.; Holmlin, R. E.; Stemp, E. D. A.;
Barton, J. K. J. Am. Chem. Soc. 1997, 119, 9861–9870;
(b) Hall, D. B.; Kelley, S. O.; Barton, J. K. Biochemistry
1998, 37, 15933–15940; (c) Kelley, S. O.; Barton, J. K.
Chem. Biol. 1998, 5, 413–425; (d) Wan, C.; Fiebig, T.;
Kelley, S. O.; Treadway, C. R.; Barton, J. K.; Zewail, A.
H. Proc. Natl. Acad. Sci. USA 1999, 96, 6014–6019.
9. (a) Morgan, G.; Walls, L. P. J. Chem. Soc. 1938, 389–
397; (b) Walls, L. P. J. Chem. Soc. 1945, 294–300.
10. Watkins, T. I. J. Chem. Soc. 1952, 3, 3059–3064.
11. Berg, S. S. J. Chem. Soc. 1963, 14, 3635–3640.
12. Loccufier, J.; Schacht, E. Tetrahedron 1989, 45, 3385–
3396.
13. Tettey, J. N. A.; Skellern, G. G.; Midgley, J. M.; Grant,
M. H.; Wilkinson, R.; Pitt, A. R. Xenobiotica 1999, 29,
349–360.
14. 1,3,5-Tri-O-acetyl-2-deoxy-D-ribofuranose was prepared
via a multi-step synthesis according to the following
procedures: (a) Recondo, E. F.; Rinderknecht, G. L.
Helv. Chim. Acta 1959, 42, 1171–1173; (b) Kline, P. C.;
Serianni, A. S. J. Am. Chem. Soc. 1990, 112, 7373–7381;
27. Tester, R. W.; West, F. G. Tetrahedron Lett. 1998, 39,
4631–4634.

1690
N. Amann, H.-A. Wagenknecht / Tetrahedron Letters 44 (2003) 1685–1690
1
28. Data of 10: Rf=0.95 (silica gel, MeOH). H NMR (250
MHz, CDCl3): d=1.58–1.63, 1.79–1.81 (m, 2 H, H-5),
2.50–2.77 (m, 2 H, H-2), 3.90–3.94 (m, 1 H, H-4), 3.95–
3.99 (m, 1 H, H-4), 4.20–4.30 (m, 1 H, H-3), 5.50 (s, 1 H,
Ph-CH), 7.27–7.56 ppm (m, 5 H, aromatic H).
131.7, 203.8 (C-1%). ESI-MS m/z (%): 518.4 (100) [M]+.
32. Alberico, F.; Bofill, J. M.; El-Faham, A.; Kates, S. A. J.
Org. Chem. 1998, 63, 9678–9683.
33. Data of 3: Rf=0.20 (silica gel, EtOAc:MeOH:H2O=
1
6:2:1). UV/vis (MeOH): umax=215, 295, 487. H and 13C
29. Diederichsen, U.; Weicherding, D. Synlett 1999, 917–920.
30. The pale yellow solid of 11 was used directly without
characterization.
NMR signals were assigned according to 2D experiments
(DQF-COSY and HMQC). 1H NMR (500 MHz, [d4]-
MeOH): d=1.56 (t, 3 H, CH2CH3), 1.72, 1.83 (m, 2 H,
H-4%), 2.51, 2.71 (m, 2 H, H-2%), 3.71, 3.77 (m, 2 H, H-5%),
4.21, 4.39 (m, 1 H, H-3%), 4.77 (q, 2 H, CH2CH3), 7.47 (s,
1 H, H-2), 7.49 (s, 1 H, H-4), 7.66–7.69, 7.81–7.85 (m, 5
H, aromatic H), 7.95 (s, 1 H, H-7), 8.31 (d, 1 H, H-9),
8.77 (d, 1 H, H-10), 8.79 ppm (d, 1 H, H-1). 13C NMR
(125.8 MHz, [d4]-MeOH): l=13.2 (CH2CH3), 39.1 (C-4%),
44.8 (C-2%), 49.7 (CH2CH3), 58.1 (C-5%), 65.8 (C-3%), 98.5,
108.2, 110.0, 119.0, 119.9, 122.1, 122.5, 123.3, 123.9,
125.4, 127.4, 128.0, 128.1, 129.2, 129.3, 130.0, 130.9,
131.0, 131.7, 198.4 ppm (C-1%). ESI-MS m/z (%): 430.4
(100) [M]+.
31. Data of 12: Rf=0.60 (silica gel, EtOAc:MeOH:H2O=
1
6:2:1). UV/vis (MeOH): umax=214, 291, 479 nm. H and
13C NMR signals were assigned according to 2D experi-
1
ments (DQF-COSY and HMQC). H NMR (500 MHz,
[d4]-MeOH): d=1.53 (t, 3 H, CH2CH3), 1.64, 1.81 (m, 2
H, H-4%), 2.56, 2.75 (m, 2 H, H-2%), 4.01, 4.21 (m, 2 H,
H-5%), 4.30 (m, 1 H, H-3%), 4.71 (q, 2 H, CH2CH3), 5.57
(s, 1 H, Ph-CH), 7.37 (s, 1 H, H-2), 7.36–7.38, 7.45–7.47,
7.50 (m, 5 H, Ph-H), 7.52 (s, 1 H, H-4), 7.70–7.73,
7.85–7.89 (m, 5 H, ethidium aromatic H), 7.94 (s, 1 H,
H-7), 8.30 (d, 1 H, H-9), 8.81 (d, 1 H, H-10), 8.83 ppm
(d, 1 H, H-1). 13C NMR (125.8 MHz, [d4]-MeOH):
l=13.2 (CH2CH3), 30.8 (C-4%), 43.3 (C-2%), 49.7
(CH2CH3), 66.6 (C-5%), 74.1 (C-3%), 98.5, 101.0 and 101.1
(Ph-C), 108.1, 110.2, 119.2, 120.0, 122.2, 123.3, 124.1,
125.4, 125.7, 125.8, 127.4, 127.5, 127.6, 128.0, 128.1,
128.2, 128.3, 128.4, 129.2, 129.3, 129.4, 129.9, 131.0,
34. According to the comparison of the UV/vis spectra
(MeCN)
of
8-acetamido-3-amino-5-ethyl-6-phenyl-
phenanthridium (umax=216, 291, 477 nm) and 3,8-diac-
etamido-5-ethyl-6-phenyl-phenanthridium
284, 431 nm).
(umax=213,