A carbocyclic analog of the oxidatively generated DNA lesion spiroiminodihydantoin

Article abstract of DOI:10.1002/ejoc.200600982

8-Oxo-7,8-dihydro-2′-deoxyguanosine (8-oxo-dG) is a mutagenic DNA lesion which is prone to further oxidation. One-electron oxidants like Ir ^IV oxidize 8-oxo-dG to give secondary lesions such as the spiroiminodihydantoin. This lesion blocks DNA

Full text of DOI:10.1002/ejoc.200600982

FULL PAPER
DOI: 10.1002/ejoc.200600982
A Carbocyclic Analog of the Oxidatively Generated DNA Lesion
Spiroiminodihydantoin
Heiko Müller[a] and Thomas Carell*[a]
Keywords: Oxidatively generated DNA lesion / Carbocyclic analog / 8-oxo-dG / Spiroiminodihydantoin
8-Oxo-7,8-dihydro-2Ј-deoxyguanosine (8-oxo-dG) is a muta-
genic DNA lesion which is prone to further oxidation. One-
electron oxidants like IrIV oxidize 8-oxo-dG to give secondary
lesions such as the spiroiminodihydantoin. This lesion blocks
DNA polymerases and induces G Ǟ T as well as G Ǟ C
transversions. Here, we report the synthesis of a carbocyclic
analog of the 8-oxo-dG lesion and the carbocyclic version of
the spiroiminodihydantoin lesion. Both lesion analogs were
obtained within single-stranded DNA and as their corre-
sponding nucleosides. Characterization of the lesion analogs
was achieved by HPLC, ESI-MS and ESI-MS/MS, as well as
enzymatic digestion of the oligonucleotides.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
UVA light in the presence of methylene blue.[9c] Recently,
both diastereomers of the Sp nucleobase were separated
and their absolute configurations were assigned.[10]
Introduction
The oxidation potential of 8-oxo-7,8-dihydro-2Ј-deoxy-
guanosine (8-oxo-dG), a major oxidative DNA lesion, is
only about one half of the oxidation potential of 2Ј-deoxy-
guanosine.[1] It is thus obvious that there are various sec-
ondary DNA lesions generated by the oxidation of 8-oxo-
dG.[2] Singlet oxygen reacts for example readily with 8-oxo-
7,8-dihydroguanosine (8-oxo-G) to form a variety of dif-
ferent products,[3] one of those being spiroiminodihydantoin
(Sp),[4] which was first misinterpreted as 4-OH-8-oxo-G.[5]
Later, the structure was corrected based on NMR spectro-
scopic data.[6]
Burrows et al. first described the spirocyclic structure of
the spiroiminodihydantoin lesion Sp, which they obtained
after one-electron oxidation of 8-oxo-7,8-dihydrogua-
nosine.[7] Up to now, this is the most convenient way to
produce Sp within a nucleoside or in DNA.[8] The oxidation
of 8-oxo-G to Sp was found to be strongly dependent on
both, the temperature and the pH value of the reaction
solution (Scheme 1). Whereas at lower pH and at lower
temperature the formation of guanidinohydantoin (Gh) and
of its isomer iminoallantoin (Ia) is favored, Sp is predomi-
nantly formed at higher temperatures and at higher pH val-
ues (Scheme 1).[7,8] In addition, the reaction of singlet oxy-
gen with guanosine or guanosine-5Ј-monophosphate was
shown to produce Sp as well.[9a,9b] The formation of Sp
upon reaction of cellular DNA with singlet oxygen is still a
matter of debate, although Box et al. were able to observe
Sp-containing dinucleotides after exposure of HeLa cells to
Scheme 1. Short overview of the oxidation of G and 8-oxo-G to
guanidinohydantoin Gh and spiroiminodihydantoin Sp. Iminoal-
lantoin (Ia) has not yet been identified as 1O2 oxidation product of
8-oxo-G. R = β--ribofuranosyl, 5-OH-8-oxo-G = 5-hydroxy-8-
oxo-7,8-dihydroguanosine, 8-OOH-G = 8-hydroperoxyguanosine.
[a] Department of Chemistry and Biochemistry, Ludwig-Maximil-
ians-University Munich,
Butenandtstr. 5–13, House F, 81377 Munich, Germany
Fax: +49-89-2180-77756
1438
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1438–1445

A Carbocyclic Analog of Spiroiminodihydantoin
FULL PAPER
When present in DNA, 2Ј-deoxyspiroiminodihydantoin deoxyribose unit in lesions by a stable cyclopentane unit.
(dSp) has a detrimental effect on the duplex stability. Re- This “chemical mutation” creates lesion analogs which can
gardless of the opposing base, the melting temperature of be crystallized together with active enzymes. In addition,
the duplex, containing dSp as a mixture of diastereomers, this modification does not affect the thermodynamic sta-
is greatly decreased. If at all, dSp shows a small preference bility of the corresponding base pair.[16] Herein, we report
for interacting with dG.[11a] The strong helix-disturbing ef- the synthesis and characterization of a carbocyclic analog
fect of dSp has serious consequences for the elongation of of the DNA lesion spiroiminodihydantoin, cdSp, in DNA
dSp-containing template-DNA strands by polymerases. In as well as on the nucleoside level. This should now enable
the presence of all four dNTPs, Klenow exo– is efficiently co-crystallization experiments using active repair enzymes
blocked by a template containing dSp.[11] At higher dNTP (Figure 1). Because the dSp heterocycle is rather base-labile,
concentrations and in the presence of only one correspond- it is difficult to incorporate this lesion directly into DNA
ing dNTP, Klenow exo– preferentially inserts dATP and by phosphoramidite chemistry.[15] However, Cadet et al. re-
dGTP opposite dSp.[11] Klenow exo+ in contrast is not able ported the site-specific incorporation of dSp phosphoram-
to elongate a dSp-containing template, even under single- idite, which they thought was 4-OH-8-oxo-dG, into DNA
nucleotide conditions.[11c] In vitro studies with E. coli using phosphoramidite chemistry.[17] Herein, we report the
showed that dSp is much more mutagenic than 8-oxo-dG synthesis of cdSp-containing DNA by one-electron oxi-
leading to nearly 100% G Ǟ T and G Ǟ C transversion dation of carbocyclic 7,8-dihydro-8-oxo-2Ј-deoxyguanosine
mutations.[12] Oxidatively generated DNA lesions are com- (8-oxo-cdG) in DNA. The latter was synthesized according
monly repaired by base excision repair (BER).[13] The short- to a new synthetic route and incorporated into three dif-
patch repair system requires the use of bifunctional glycos- ferent oligo-2Ј-deoxynucleotides (ODN) using phosphor-
ylases/AP lyases like the bacterial enzymes MutM/FPG or amidite chemistry.
endonuclease VIII/Nei (both of which belong to the Nei
superfamily) or its eukaryotic functional homolog OGG
(either from yeast or human, HhH superfamiliy). MutM or
FPG, which readily removes 8-oxo-dG and a variety of
other oxidized purines,[14] was also found to efficiently re-
move dSp and dGh regardless of the opposite base.[15]
In order to investigate the molecular recognition of the
lesion by repair enzymes and to study replication processes
of DNA-template strands containing the dSp lesion, it is
essential to have access to a stabilized version of this lesion.
Figure 1. 2Ј-Deoxyspiroiminodihydantoin (X = O; dSp) and its car-
We have recently introduced the concept to replace the 2- bocyclic analog (X = CH2; cdSp).
Scheme 2. a) H2, 10% Pd on charcoal in methanol; 1.05 equiv. CDI in CHCl3, room temp., 18 h, 55 °C 3 h, 29%; b) HF·pyridine/pyridine
in ethyl acetate, room temp., 18 h, 95%; c) MeNH2/NH3 (1:1), 0.5  β-mercaptoethanol, room temp., 8 h, 99%; d) 1.2 equiv. DMTCl in
pyridine, 0 °C, 2 h, 50%; e) 1.5 equiv. CEDCl in THF, 0 °C, 2 h, 70%; f) sodium phosphate 50 m, pH = 8.0, 6.8 m 6, 6 m Na2IrCl6,
60 °C, 1 h.
Eur. J. Org. Chem. 2007, 1438–1445
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1439

H. Müller, T. Carell
FULL PAPER
the carbocyclic moiety was found to proceed very slowly.
The resulting N2-acetyl-5Ј-O-(dimethoxytrityl)-8-oxo-7,8-
dihydro-2Ј-deoxycarbaguanosine (4) was finally converted
into the corresponding 3Ј-phosphoramidite 5 by reaction
with 2-cyanoethyl N,N-diisopropylchlorophosphoramidite.
The reagent was selective towards the 3Ј-position under the
conditions used here, neither the O6- nor the O8-adduct
could be detected by 31P NMR spectroscopy.
The second critical step in the synthesis required the se-
lective oxidation of the 8-oxo-dG analog to give the corre-
sponding cdSp lesion. In order to investigate this step, we
used the completely deprotected nucleoside for the oxi-
dation of 8-oxo-cdG. To this end, compound 3 was dis-
solved in a 1:1 mixture of methylamine and ammonia to
cleave the N2-acetyl group; 0.5  β-mercaptoethanol was
added to prevent aerial oxidation of 8-oxo-cdG which is a
known problem.[18] After 8 h at room temperature, no start-
ing material could be detected any more by rp-18 TLC.
The needed unprotected 8-oxo-7,8-dihydro-2Ј-deoxycarba-
guanosine (6) was finally obtained with 99% yield
(Scheme 2, Figure 2a).
Results and Discussion
The synthesis of the 8-oxo-cdG phosphoramidite was
performed using a novel strategy quite different compared
to previous studies reported by Johnson et al. starting from
aristeromycin.[18] Our convergent synthesis of 8-oxo-cdG
(Scheme 2) started with the enantiomerically pure nitropyri-
midinone 1, which was prepared in nine steps with an over-
all yield of 35% as previously reported by us.[16]
The first step towards the preparation of 7 was the re-
duction of the nitro group using hydrogen and palladium
on charcoal. The ring closure to give the 8-oxo-7,8-dihy-
droguanine heterocycle proved to be the crucial step. Both,
the 5-amino group and the 6-oxo group as well as the acet-
amide moieties are susceptible towards electrophilic attack.
Thus, the reagent needed to close the five-membered ring
substructure had to be fairly selective but also reactive
enough to attack both the 5- and the 4-amino groups of the
pyrimidinone heterocycle 1 to give the desired N2-acetyl-
3Ј,5Ј-O-bis(tert-butyldimethylsilyl)-8-oxo-7,8-dihydro-2Ј-de-
oxycarbaguanosine (2). We found that this reaction is best
performed with 1,1Ј-carbonyldiimidazole (CDI) in dry
The oxidation potential of sodium hexachloroiridate(IV)
chloroform at high dilution. This method provided com- lies between those of dG and 8-oxo-dG.[8b] This allows the
pound 2 in a respectable yield of about 30%. The following reagent to oxidize selectively 8-oxo-dG in the presence of
silyl deprotection with HF/pyridine in ethyl acetate buffered dG. All other nucleobases do not react at all. This selective
with pyridine was nearly quantitative yielding the nucleo- oxidation, however, depends critically on the pH of the
side 3 as a light yellow precipitate with both the 5Ј-and the solution as well as on the reaction temperature. Both pa-
3Ј-hydroxyl groups unprotected. The 5Ј-DMT protection of rameters have to be tightly controlled. We obtained the best
Figure 2. a) HPLC of 6 and 7 using a gradient of 0–25% buffer B in 45 min; buffer A: 2 m TEAA in water; buffer B: 2 m TEAA in
80% acetonitrile. b) Negative ESI-MS (top) and MS/MS (bottom); parent ion of 7 [M – H]–: m/z = 296.1005. c) HPLC and MALDI-
TOF of ODN1. d) HPLC and MALDI-TOF of ODN4; buffer A: 0.1  TEAA in water; buffer B: 0.1  TEAA in 80% acetonitrile (c
and d).
1440
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1438–1445

A Carbocyclic Analog of Spiroiminodihydantoin
FULL PAPER
results, when we performed the oxidation in sodium phos- Table 1. Synthesized modified ODNs.
phate buffer at pH = 8.0 and 60 °C. Under these conditions,
Name
Sequence
Mcalcd.
(amu)
Mfound
(amu)[a]
we observed that 6 cleanly reacted with Na2IrCl6 to give 7
(Figure 2a). The reaction was stopped by cooling the reac-
tion mixture to room temperature. The product was finally
purified by directly subjecting the reaction solution to ODN2
HPLC separation using 2 m triethylammonium acetate
ODN1
5Ј-CTCTTT[8-oxo-cdG]
CTCTTTG
5Ј-GCGAT[8-oxo-cdG]
TAGCG
5Ј-TGCAGT[8-oxo-cdG]
ACAGC
5Ј-CTCTTT[cdSp]CTCTTTG 4216.49
4200.49
3411.03
3684.20
4200.80
3410.14
3684.82
4216.42
ODN3
(TEAA) at pH = 7.0 in water (buffer A) and 2 m TEAA
at pH = 7.0 in 80% acetonitrile (buffer B). The peak eluting
under our conditions (gradient 0–25% B over 45 min) at
4.5 min (Figure 2a) had a detectable molecular weight of
296.1005 amu, which is in perfect agreement with the calcu-
ODN4[b]
[a] MALDI-TOF analysis. [b] As a mixture of diastereomers.
lated molecular weight of 7 (m/z = 296.1000). The fragmen- bated at 37 °C for 3 h and then set to a more basic pH by
tation pattern of this molecular ion peak can easily be ex- addition of TRIS-HCl (pH = 8.0). Alkaline phosphatase
plained by the loss of the fragments indicated in Figure 3. and snake venom phosphodiesterase I were subsequently
These data fully confirm the presence of the spiroiminodi- added and the mixture was incubated at 37 °C for another
hydantoin structure.
3 h. Following this procedure, the pH was set to approxi-
mately 7 with 0.1  HCl and the enzymes were denatured
by heating of the solution to 90 °C for 15 min prior to cen-
trifugation. 20 µL of this solution were injected directly into
the HPLC-MS system. The HPL chromatogram (Figure 4a
and b) showed three peaks, two of which could be assigned
to be dC and dG, respectively. The third peak consisted of
both, dT and 8-oxo-cdG nucleoside, as determined by ESI-
MS.
Figure 3. Negative-mode ESI-MS/MS of 7 at 35 eV. Structural as-
signments of the fragmentations are suggested as indicated in bold.
The reaction conditions used to convert ODN1 to ODN4
were found to be similar compared to those used for the
We next incorporated 8-oxo-cdG into three different synthesis of 7. ODN1 was dissolved in 14 m sodium phos-
ODNs with a sequence length of 11, 12 and 14 nucleotides phate buffer at pH = 8.0 and oxidized with 100 µ Na2IrCl6
(Table 1). The DNA synthesis was performed on controlled at 60 °C for 1 h. Preheating of the mixture before adding
pore glass (cpg) support using standard conditions with the iridate solution was necessary to avoid the formation of
benzylthiotetrazole (BTT) as the coupling reagent. Due to guanidinohydantoin side products. The mixture was then
the fact, that the coupling efficiency of 8-oxo-cdG phos- cooled to room temperature and the reaction stopped by
phoramidite 5 was significantly lower, the coupling times passing the solution through a Waters Sep-Pak® C18 car-
were set to 10 min and the coupling procedure for 5 was tridge. This removed the buffer salts as well as the remain-
repeated once. Best incorporation was finally achieved by ing iridate and therefore stopped the reaction effectively.
coupling 5 with 16 equiv. over 20 min. Cleavage from the The resulting ODN was checked for purity using analytical
solid support and removal of the protecting groups was per- HPLC. Here too, a buffer system consisting of 0.1  TEAA
formed using a 1:1 mixture of methylamine and ammonia at pH = 7.0 in water (buffer A) and 0.1  TEAA in 80%
with 0.5  β-mercaptoethanol at room temperature for acetonitrile (buffer B) was used with a gradient of 5–22%
18 h. Methylamine was used since pure ammonia gave un- B in 50 min. The chromatogram showed the presence of
satisfying deprotection results. The ODNs were purified by only one peak corresponding to pure ODN4 with a molecu-
HPLC using 0.1  TEAA at pH = 7.0 in water (buffer A) lar weight of m/z = 4216.42 as checked by MALDI-TOF
and 80% acetonitrile (buffer B) with a gradient of 5–22% analysis (Figure 2d) showing the oxidation of 8-oxo-cdG to
B in 50 min on a Macherey–Nagel Nucleodur 100-5 C18 cdSp in a clean process inside a DNA single-strand. Storage
column and checked for purity by analytical HPLC using of ODN4 at room temperature for several days was possible
the same buffer system and gradient on a Macherey–Nagel without formation of degradation products as investigated
Nucleodur 100-3 C18 column (Figure 1c). The molecular using HPLC and MALDI-TOF mass spectrometry (data
weights of the ODNs were determined using MALDI-TOF not shown).
mass spectrometry. Desalting was performed using Waters
Sep-Pak® C18 cartridges according to the manufacturer DNA, an additional enzymatic digest of ODN4 was per-
protocol. formed according to the same protocol described above. Af-
In order to proof the presence of the cdSp lesion in
Total enzymatic digest of ODN1 was performed to proof ter centrifugation of the reaction mixture, a sample of 20 µL
the correct incorporation of 8-oxo-cdG. A set of enzymes was analyzed by LC-MS. Although the structure of spiro-
was used to ensure complete degradation of the DNA. To iminodihydantoin differs significantly from the structure of
a 20 µ solution of ODN1, we added ammonium acetate 8-oxo-dG, degradation of the DNA strand by the described
buffer (pH = 5.7), nuclease P1 (Penicillium citrinum) and enzymes was successful. No partially digested fragments of
calf spleen phosphodiesterase II. The solution was incu- ODN4 could be observed. The detection of the early eluting
Eur. J. Org. Chem. 2007, 1438–1445
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1441

H. Müller, T. Carell
FULL PAPER
Figure 4. LC-MS and LC-MS/MS of enzymatic digest of: a) ODN1, HPLC on Macherey–Nagel Nucleodur 125-2 3 µ C18 pyramid, 2 m
TEAA, pH = 7.0, in water (A) and 80% acetonitrile (B), 100% A 10 min, 0–10% B 18 min, UV trace (top) and ion count at m/z =
295.5–296.5 (bottom); b) ESI-MS of peak eluting at 12 min; c) ODN4, HPLC on Uptisphere Interchrom, 2 m ammonium formate, pH
= 5.5, in water (A) and 80% acetonitrile (B), gradient equal to a), UV trace (top), ion count at m/z = 341.5–342.5 (bottom); d) ESI-MS
(top), -MS/MS (centre) and -MS/MS/MS (bottom) of peak eluting at 3 min, parent ions are [M – H + HCOOH]–: m/z = 342.1054 for
MS/MS and [M – H]–: m/z = 296.10 for MS/MS/MS; M = cdSp.
cdSp nucleoside by ESI-MS, however, appeared to be prob- complished. The oxidation of 8-oxo-cdG with the one-elec-
lematic because buffer salts were eluting with the same re- tron oxidant sodium hexachloroiridate(IV) provided the
tention time and therefore disturbed the sensitivity of the carbocyclic analog of the DNA lesion spiroiminodihydan-
ESI measurement. We found that the use of a different col- toin (cdSp) in a nearly quantitative conversion. The forma-
umn, namely an Uptisphere Interchrom 150ϫ2.1, and a tion of cdSp was also possible using 8-oxo-cdG in single-
buffer change to 2 m ammonium formate at pH = 5.5 in- stranded DNA.
stead of triethylammonium acetate at pH = 7.0 for the LC-
The presence of both, the carbocyclic lesions cdSp and
MS increased the detectability of the lesion analog signifi- 8-oxo-cdG in the corresponding ODN was confirmed by
cantly. The peak eluting at 3 min was identified to be the total enzymatic digest followed by ESI-MS and MS/MS
cdSp nucleoside (Figure 4c). The ESI-MS spectrum shows analysis. These carbocyclic lesion analogs lacking a glyco-
a molecular weight of m/z = 342.1054 in perfect agreement sidic bond between the sugar and the base moiety now pro-
with the molecular weight of the formic acid adduct of cdSp vide a powerful tool for the investigation of their recogni-
[M + HCOOH – H]– (342.1055 amu, Figure 4d). Confirm- tion by DNA repair glycosylases using protein crystallogra-
ation of the assignment is based on MS/MS measurements phy.
of this peak. The peak at 296.10 amu corresponds to the
free cdSp nucleoside with a calculated molecular weight of
m/z = 296.1000. MS/MS/MS measurements of this peak
Experimental Section
showed exactly the same fragmentation pattern compared
Materials and Methods: All solvents and reagents were obtained in
to 7, thus being the evidence for the presence of a correctly
commercially available qualities puriss., p.a., or purum. Phos-
incorporated cdSp lesion into ODN4.
phoramidites were from Glen Research or Samchully pharm, cpg
support and benzylthiotetrazole (BTT) was from Glen Research.
Dry solvents were purchased from Fluka or Acros and were used
as received. Acetonitrile for DNA synthesis was from Roth in Roti-
dry grade with a water content of 10 ppm. Acetonitrile for HPLC
and HPLC/MS was purchased from VWR in HPLC grade. All ex-
Conclusion
A new short and efficient synthesis of a carbocyclic ana-
log of 8-oxo-7,8-dihydro-2Ј-deoxyguanosine (8-oxo-cdG)
was developed and its incorporation into three different
periments involving water-sensitive compounds were conducted in
oven-dried glassware under nitrogen. Reactions were monitored by
ODNs using standard phosphoramidite chemistry was ac- analytical thin-layer chromatography (TLC) on VWR precoated al-
1442 Eur. J. Org. Chem. 2007, 1438–1445
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

A Carbocyclic Analog of Spiroiminodihydantoin
FULL PAPER
uminium plates 60 F254, visualized by UV light, anisaldehyde or
ninhydrin staining. Melting points were obtained with a Reichert-
Jung type 651501 using a Leica Galen III microscope and are not
corrected. IR spectra were obtained with a Perkin–Elmer Spectrum
BX connected with a Smiths Detection DuraSamp/IR II. 1H NMR
spectra were obtained with a Varian Mercury-200 and a Varian
INOVA-400. The chemical shifts were referenced to CHCl3 in
CDCl3 and DMSO in [D6]DMSO. If necessary, peak assignment
was carried out with the help of COSY, HMBC or HSQC experi-
ments. 13C NMR spectra were obtained with a Varian INOVA-400
and a Bruker AMX 600. 31P NMR spectra were obtained with a
Varian Mercury-200. FAB mass spectra were recorded with a Fin-
nigan MAT95Q using a 2-nitrobenzylalcohol or glycerol matrix.
ESI spectra were obtained with a Thermo Finnigan LT-FT ICR
spectrometer. MALDI-TOF spectra were obtained with a Bruker
Autoflex II spectrometer using a matrix consisting of a 1:1 mixture
of saturated 3-hydroxypicolinic acid (HPA) in water and a solution
of 50 mg of HPA, 10 mg of ammonium hydrogen citrate and 10 µL
of 15-crown-5 in 1 mL of water. HPLC analysis was performed
with a Waters 2695 and a Waters 2996 Photo Diode Array detector
using a Macherey–Nagel Nucleodur 100-3 C18 column. HPLC sep-
aration of DNA was achieved with either a Waters 1525 and a
Waters 2487 UV detector or a Merck-Hitachi L7150 and a Merck-
Hitachi L7420 UV/Vis detector. UV absorption for concentration
measurement of the ODNs was measured with an Eppendorf
Biophotometer.
(9.6) [M + Na]+ . HRMS (FAB+ ; [M + Na]+ ): calcd. for
[C25H45N5O5Si2 + Na]+ 574.2851, found 574.2795.
N2-Acetyl-8-oxo-7,8-dihydro-2Ј-deoxycarbaguanosine (3): HF·
pyridine (50 µL) and pyridine (50 µL) were added under argon to
a solution of 2 (52 mg, 0.09 mmol) in dry ethyl acetate (5 mL).
After stirring at room temperature for 18 h, a white solid precipi-
tated which was separated by centrifugation, taken up in a solution
of methoxytrimethylsilane (200 µL) in dry ethyl acetate (2 mL) and
stirred for an additional 1 h. The mixture was then centrifuged
again and the white precipitate was dried under vacuum to afford
3 in 95 % yield (29 mg, 0.09 mmol). M.p. Ͼ 230 °C. 1H NMR
(400MHz, [D6]DMSO, 25 °C): δ = 1.63–1.69 (m, 1 H, C2ЈHa),
1.77–1.85 (m, 1 H, C4ЈH), 1.87–1.97 (m, 2 H, C6ЈHa, C6ЈHb), 2.13
(s, 3 H, COCH3), 2.40–2.45 (m, 1 H, C2ЈHb), 3.31–3.53 (m, 1 H,
C5ЈHa), 3.48–3.53 (m, 1 H, C5ЈHb), 4.02–4.06 (m, 1 H, C3ЈH), 4.49
(t, 1 H, C5ЈOH), 4.60 (d, 1 H, C3ЈOH), 4.83 (m, 1 H, C1ЈH), 11.04,
11.52, 11.98 (3ϫs, 3 ϫ1 H, 3ϫNH) ppm. 13C NMR (100 MHz,
[D6]DMSO): δ = 24.42 (COCH3), 31.58 (C6Ј), 37.58 (C2Ј), 50.23,
50.62 (C1Ј, C4Ј), 63.98 (C5Ј), 72.26 (C3Ј), 103.57, 146.01, 147.50,
149.54, 152.47 (C2, C4, C5, C6, C8), 173.99 (COCH ) ppm. IR: ν
˜
3
= 3332.22 w, 3071.86 m, 2966.89 w, 2880.51 w, 2756.19 w, 2666.15
w, 1975.75 m, 1714.25 s, 1651.10 vs, 1609.80 w, 1572.22 m, 1433.21
m, 1345.42 m, 1233.84 s, 1162.05 w, 1108.04 m, 1034.23 s, 1015.82
s, 933.40 m, 882.36 m, 844.13 m, 796.07 s, 757.90 s, 742.17 s, 698.75
s cm–1. HRMS (ESI+; [M + H]+): calcd. for [C13H17N5O5 + H]+
324.1302, found 324.1302.
N2-Acetyl-3Ј,5Ј-O-bis(tert-butyldimethylsilyl)-8-oxo-7,8-dihydro-2Ј-
deoxycarbaguanosine (2): To a solution of 1 (200 mg, 0.36 mmol)
in methanol (10 mL) Pd on charcoal (50 mg) was added and the
mixture was degassed three times by freezing and evaporating. The
nitrogen was exchanged for hydrogen and the reaction mixture was
stirred at room temperature for 18 h. The solution was then filtered
under nitrogen through a 0.2 µm nylon syringe filter (Whatman
Puradisk®) and dried under vacuum. The brown product was dis-
solved in dry chloroform (20 mL). To this solution was added drop-
wise and very slowly a solution of 1,1Ј-carbonyldiimidazole
(1.05 equiv., 0.38 mmol, 61.2 mg) in dry chloroform (10 mL) at
room temperature. The reaction mixture was stirred at room tem-
perature for 20 h, followed by heating to 55 °C for another 3 h.
After cooling, additional chloroform (40 mL) was added and the
solution was washed with brine. The organic phase was dried with
sodium sulfate and concentrated in vacuo. The crude product was
purified by flash chromatography on silica gel 60 (10 g silica gel 60,
CHCl3/MeOH, 40:1) which afforded 2 as a yellow foam (58 mg,
0.11 mmol, 29 %). Rf = 0.35 (CHCl3/MeOH, 10:1). M.p. 143–
145 °C. 1H NMR (200 MHz, CDCl3, 25 °C): δ = 0.03, 0.04, 0.05
(3ϫs, 12 H, SiCH3), 0.86, 0.89 [2ϫs, 18 H, SiC(CH3)3], 1.83 (m,
1 H, C2ЈHa), 1.97 (m, 1 H, C6ЈHa), 2.01 (m, 1 H, C4ЈH), 2.07 (m,
1 H, C6ЈHb), 2.29 (s, 3 H, COCH3), 2.51 (m, 1 H, C2ЈHb), 3.64
N2-Acetyl-5Ј-O-(dimethoxytrityl)-8-oxo-7,8-dihydro-2Ј-deoxycarba-
guanosine (4): 3 (71 mg, 0.22 mmol) was coevaporated with dry pyr-
idine (2 mL) and again dissolved in dry pyridine (4 mL). Activated
molecular sieves (4 Å) was added and the mixture stirred for
30 min. The solution was then cooled in an ice bath to 0 °C and a
solution of dimethoxytrityl chloride (89.2 mg, 0.26 mmol,
1.2 equiv.) in dry pyridine (1 mL) was added dropwise. The progress
of the reaction was monitored by TLC. After a complete conver-
sion of the starting material, MeOH (4 mL) was added and the
solvent was evaporated in high vacuum. Flash chromatography of
the crude product (8 g silica gel 60, CHCl3/MeOH, 30:1, 0.5% pyri-
dine) afforded 4 as a slightly yellow solid (69 mg, 0.11 mmol, 50%).
Rf = 0.27 (CHCl3/MeOH, 10:1). M.p. 164–166 °C. 1H NMR
(200MHz, CDCl3, 25 °C): δ = 1.78–1.92 [m, 2J(C6ЈHa,C6ЈHb) =
13.8 Hz, 1 H, C6ЈHa], 2.16–2.19 (m, 3 H, C2ЈHa, C2ЈHb, C4ЈH),
2.25 (s, 3 H, NHCOCH3), 2.51–2.65 [m, 2J(C6ЈHb,C6ЈHa) =
13.4 Hz, 1 H, C6ЈHb], 3.08–3.16 [dd, 2J(C5ЈHa,C5ЈHb) = 8.8 Hz,
3 J(C5ЈHa ,C4ЈH) = 6.6 Hz, 1 H, C5ЈHa ], 3.30–3.37 [dd,
2J(C5ЈHb,C5ЈHa) = 8.8 Hz, 3J(C5ЈHb,C4ЈH) = 5.0 Hz, 1 H, C5ЈHb],
3.77, 3.78 (2ϫs, 6 H, 2ϫOCH3), 4.22–4.31 (m, 1 H, C3ЈH), 4.82–
3
5.00 (m, 1 H, C1ЈH), 6.87 [2ϫd, J(mH of pMeOPh, oH of pMe-
OPh) = 9.0 Hz, 4 H, mH of pMeOPh], 7.20–7.51 (m, 9 H, PhDMTH)
ppm; 3ϫNH, 3ЈOH not detected. 13C NMR (100 MHz, CDCl3):
δ = 23.17 (NHCOCH3), 32.27, 37.46 (C2Ј, C6Ј), 48.07, 49.70 (C1Ј,
C4Ј), 54.60 (2 ϫ CDMTOCH3), 65.21 (C5Ј), 72.91 (C3Ј), 85.72
[OCPh(pMeOPh)2], 103.11 (C5), 112.94 (4ϫmC von pMeOPh),
126.50 (pC of PhDMT), 127.61, 128.18, 130.07, 130.09 (8ϫCHDMT),
136.41 (2ϫipsoC of pMeOPh), 145.52 (ipsoC of PhDMT), 145.68,
147.07, 148.98 (3ϫCHet), 151.81 (C8), 158.64 (2 ϫCDMTOMe),
2
3
[dd, J(C5ЈHa,C5ЈHb) = 10.2 Hz, J(C5ЈHa,C4Ј) = 5.0 Hz, 1 H,
C5ЈHa], 3.69 [dd, 2J(C5ЈHb,C5ЈHa) = 10.1 Hz, 3J(C5ЈHb,C4ЈH) =
5.2 Hz, 1 H, C5ЈHb], 4.35 (m, 1 H, C3Ј), 4.89 (m, 1 H, C1ЈH), 8.78,
9.97 (2ϫs, 2 ϫ1 H, N7H, C2NH), 11.99 (s, 1 H, N1H) ppm. 13C
NMR (100 MHz, CDCl3): δ = –5.65, –5.35, –4.72, –4.53
(4 ϫ SiCH3), 18.03, 18.34 [2ϫ SiC(CH3)3], 24.31 (NHCOCH3),
25.86, 25.99 [2ϫSiC(CH3)3], 30.64 (C6Ј), 37.75 (C2Ј), 50.20, 50.35
(C1Ј, C4Ј), 63.50 (C5Ј), 72.98 (C3Ј), 103.97 (C5), 145.71, 145.95,
172.84 (NHCOCH ) ppm. IR: ν = 3200.01 s, 2931.70 vs, 1690.00
˜
3
vs, 1606.78 s, 1573.53 s, 1507.14 vs, 1439.76 m, 1246.58 s, 1174.32
m, 1031.09 m cm– 1 . HRMS (ESI– ; [M – H]– ): calcd. for
[C34H35N5O7 – H]– 624.2464, found 624.2474.
149.60, 152.26 (4 ϫ CHet), 171.30 (NHCOCH ) ppm. IR: ν =
˜
3
3230.71 vw, 2929.51 vs, 2954.06 vs, 2895.17 s, 2802.33 m, 2768.31
m, 1961.25 m, 1736.85 s, 1710.38 s, 1650.10 vs, 1606.37 w, 1560.05
vs, 1502.18 m, 1437.12 s, 1359.84 w, 1343.65 w, 1311.33 w, 1247.72 N2-Acetyl-3Ј-O-(2-cyanoethyl-N,N-diisopropyl-phosphoramidite)-5Ј-
s, 1089.34 m, 1004.49 w, 940.04 w, 832.20 s, 771.92 s, 704.36 w,
O-(dimethoxytrityl)-8-oxo-7,8-dihydro-2Ј-deoxycarbaguanosine (5):
4 (69 mg, 0.11 mmol) was coevaporated with dry THF (2 mL) and
662.91 w cm–1. MS (FAB+): m/z (%) = 552.7 (7.0) [M + H]+, 574.7
Eur. J. Org. Chem. 2007, 1438–1445
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1443

H. Müller, T. Carell
FULL PAPER
dissolved again in dry THF (6 mL) under argon. The solution was
cooled in an ice bath to 0 °C and diisopropylethylamine (110 µL)
was added through a septum. Then, 2-cyanoethyl N,N-diisopro-
pylchlorophosphoramidite (36 µL, 0.13 mmol, 1.2 equiv.) was
added slowly and the solution was stirred at 0 °C until the TLC
iminodihydantoin) (7): A solution of 6 (10 µL, 6.8 m) was added
to sodium phosphate buffer (50 µL, 100 m, pH = 8.0) and water
(6 µL) and preheated to 60 °C. A solution of sodium hexachloroiri-
date(IV) (34 µL, 6 m) was added and the mixture was shaken at
60 °C for 1 h. Purification of the oxidized nucleoside was ac-
showed no more starting material. After the reaction was com- complished by direct subjection of the oxidation mixture to analyti-
pleted, the solvent was evaporated under high vacuum and the
crude product was purified by flash chromatography on silica gel
(silica gel 60, deactivated with pyridine, CHCl3/MeOH, 40:1, 0.5%
pyridine) yielding 5 (64 mg, 70%). Rf = 0.45 (CHCl3/MeOH, 10:1).
31P NMR (200 MHz, [D6]acetone, 25 °C): δ = 147.99, 148.39 (2
diastereomers) ppm. HRMS (ESI–; [M – H]–): calcd. for
[C43H52N7O8P – H]– 824.3542, found 824.3556.
cal HPLC using a buffer system of 2 m TEAA in water (buffer
A) and 2 m TEAA in 80% acetonitrile (buffer B) and a gradient
of 0–25% buffer B in 45 min on a Macherey–Nagel Nucleodur 100-
3. LC-MS/MS analysis was done with an Interchrom Uptisphere 3
HDO 150ϫ2.1 mm column using a buffer system of 2 m ammo-
nium formate in water (buffer A) and 0.4 m ammonium formate
in 80% acetonitrile (buffer B) and a gradient of 10 min 100% buffer
A, 18 min 0–10% buffer B at a flow rate of 0.2 mL/min. The ESI
spectrometer was set to the negative mode with ion spray voltage
of 4 kV, a capillary temperature of 200 °C and a gas flow of 40.
MS/MS and MS/MS/MS experiments were done with an ionization
energy of 35 eV.
8-Oxo-7,8-dihydro-2Ј-deoxycarbaguanosine (6): 3 (29 mg,
0.09 mmol) was dissolved in a mixture of methylamine (1 mL, 40%
in water), concd. ammonia (1 mL) and β-mercaptoethanol
(200 µL). After stirring at room temperature for 8 h, the solvent
was removed with a Savant Speed Vac®. The crude product was
taken up in water/acetonitrile (20:1) and purified on RP18-silica
gel (10 g RP18-silica gel, water/acetonitrile, 20:1) which afforded 6
in 99% yield (25 mg, 0.09 mmol). Rf = 0.64 (water/MeCN, 4:1, RP-
18). M.p. Ͼ230 °C. 1H NMR (200MHz, [D6]DMSO, 25 °C): δ =
1.59–1.65 (m, 1 H, C2ЈHa), 1.81–1.93 (m, 3 H, C4ЈH, C5ЈHa,
C5ЈHb), 2.38–2.44 (m, 1 H, C2ЈHb), 3.32–3.36 (m, 1 H, C6ЈHa),
3.49–3.53 (m, 1 H, C6ЈHb), 4.02–4.04 (m, 1 H, C3ЈH), 4.70–4.79
(m, 1 H, C1ЈH), 6.36 (s, 2 H, NH2), 10.50 (br. s, 1 H, NH) ppm;
3ЈOH, NH not detected. 13C NMR (100 MHz, [D6]DMSO): δ =
31.46 (C6Ј), 37.68 (C2Ј), 50.11, 50.31 (C1Ј, C4Ј), 64.05 (C5Ј), 72.44
(C3Ј), 98.83, 148.51, 151.68, 152.62, 153.72 (C2, C4, C5, C6, C8)
Oxidation of ODN1: To a solution of ODN1 (42 µL, 0.77 m),
sodium phosphate buffer (140 µL, 100 m, pH = 8.0) and a solu-
tion of sodium chloride (28 µL, 5 ) in water (1050 µL) was added
a solution of sodium hexachloroiridate(IV) (140 µL, 1 m) after
preheating both solutions to 60 °C. The mixture was then shaken at
60 °C for 1 h and the reaction was stopped by passing the solution
through a Waters Sep-Pak® cartridge, desalting with water (6 mL)
and eluting the resulting ODN4 with water/acetonitrile (1:1, 4 mL).
The resulting solution was concentrated in vacuo with a Savant
Speed-Vac®. The sample was checked for analytical purity by sub-
jecting a sample of 5 µL to analytical HPLC using the same condi-
tions as described above for the purification of synthesized ODNs.
ODN4 was analyzed using MALDI-TOF and ESI-MS.
ppm. IR: ν = 3327.64 s, 3221.06 s, 2882.07 m, 2743.80 w, 2359.43
˜
w, 1719.19 s, 1693.35 vs, 1645.10 vs, 1597.18 s, 1523.14 w, 1438.97
m, 1346.92 m, 1216.89 w, 1048.03 s, 947.54 m, 846.24 m, 796.44 m,
756.94 m, 689.70 m cm–1. HRMS (ESI+; [M + H]+): calcd. for
[C11H15N5O4 + H]+ 282.1197, found 282.1191.
Enzymatic Digestion of ODN1 and ODN4: To a solution of either
ODN1 or ODN4 (each 100 µL, 20 µ) was added a buffer (10 µL)
containing 300 m ammonium acetate, 100 m CaCl2 and 1 m
ZnSO4 (pH = 5.7), followed by the addition of nuclease P1 (Penicil-
lium citrinum) (22 units) and calf spleen phosphodiesterase II
(0.05 units). The solution was incubated at 37 °C for 3 h. To the
resulting solution was added a buffer solution (12 µL) containing
500 m Tris-HCl, 1 m EDTA (pH = 8.0), snake venom phospho-
diesterase I (0.1 units) and alkaline phosphatase (calf intestinal
phosphatase) (10 units) were added sequentially followed by incu-
bation at 37 °C for another 3 h. To the solution thus obtained 0.1 
HCl (6 µL) was added. The solution was then heated to 90 °C for
15 min and then centrifuged at 12000 rpm for 15 min; a sample of
20 µL of the resulting solution was subjected to LC-MS/MS.
ODN1 was analyzed using 2 m TEAA at pH = 7.0 in water
(buffer A) and 2 m TEAA at pH = 7.0 in 80% acetonitrile with
a gradient of 100% A for 10 min, 0–10% B in 18 min on a Mach-
erey–Nagel Nucleodur 125-2 3µ C18 pyramid. The ESI spectrome-
ter was set to negative mode, with an ion spray voltage of 4 kV, a
capillary temperature of 200 °C and a gas flow of 40. ODN4 was
analyzed using a buffer system of 2 m ammonium formate at pH
= 5.5 in water (buffer A) and 2 m ammonium formate at pH =
5.5 in 80% acetonitrile (buffer B) with the same gradient as for
ODN1, but on a Uptisphere Interchrom 150ϫ2.1.
Oligonucleotide Synthesis: DNA synthesis was performed with an
Amersham Äkta oligopilot using standard conditions on controlled
pore glass (cpg) support at a 1 µmol scale. The concentration of
the amidite solutions was 0.1 . BTT was used as the coupling
reagent (0.25 ) and the coupling time for standard amidites was
set to 3 min. 5 was introduced by a double coupling procedure at
which two times 8 equiv. were coupled for 10 min without a cap-
ping step in between. Syntheses were performed “DMT-off”. De-
protection of the ODNs was achieved by incubation with a 1:1
mixture of concd. ammonium hydroxide and methylamine (40% in
water) at room temperature for 18 h, additional 0.5  mercapto-
ethanol was added to prevent oxidation. The solution was then
concentrated to dryness with a Savant Speed Vac® and the crude
product was resuspended in water. Purification was accomplished
with 0.1  TEAA in water at pH = 7.0 (buffer A), 0.1  TEAA at
pH = 7.0 in water/acetonitrile (20:80) (buffer B) and a gradient of
5–22% buffer B in 50 min at a flow rate of 5 mL/min on a Mach-
erey–Nagel Nucleodur 100-5 C18. The fractions were checked for
analytical purity by HPLC using the same gradient and buffer sys-
tem but a Macherey–Nagel Nucleodur 100-3 column and analyzed
by MALDI-TOF and ESI-MS. ODNs were desalted with a Waters
Sep-Pak® cartridge according to the manufacturer protocol and
quantified by UV spectroscopy. The concentration of the ODNs
was determined by measuring the UV absorption at 260 nm, the
molar extinction coefficient of the oligomer was estimated by add-
ing up the coefficients of the mononucleotides at 260 nm as pub-
lished.[19,20]
Acknowledgments
The work was supported by the Volkswagenstiftung, the Fonds der
chemischen Industrie, the Deutsche Forschungsgemeinschaft
(DFG), funds from the State of Bavaria and support from the
Gottfried Wilhelm Leibniz program of the DFG. We thank Dr.
7-Amino-1-[3-hydroxy-4-(hydroxymethyl)cyclopentyl]-1,3,6,8-tetra-
azaspiro[4.4]non-6-ene-2,4,9-trione (Carbocyclic 2Ј-Deoxyspiro-
1444
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1438–1445

A Carbocyclic Analog of Spiroiminodihydantoin
FULL PAPER
J.-L. Ravanat, P. Krajewski, J. Cadet, Chem. Res. Toxicol. 2006,
19, 1357–1365.
Werner Spahl for help with ESI measurements and the NMR and
MS division of the LMU for support.
[11] a) O. Kornyushyna, A. M. Berges, J. G. Muller, C. J. Burrows,
Biochemistry 2002, 41, 15304–15314; b) V. Duarte, J. G. Muller,
C. J. Burrows, Nucleic Acids Res. 1999, 27, 496–502; c) O. Kor-
nyushyna, C. J. Burrows, Biochemistry 2003, 42, 13008–13018.
[12] a) P. T. Henderson, J. C. Delaney, J. G. Muller, W. L. Neeley,
S. R. Tannenbaum, C. J. Burrows, J. M. Essigmann, Biochemis-
try 2003, 42, 9257–9262; b) P. T. Henderson, J. C. Delaney, F.
Gu, S. R. Tannenbaum, J. M. Essigmann, Biochemistry 2002,
41, 914–921.
[13] O. D. Schärer, Angew. Chem. Int. Ed. 2003, 42, 2946–2974.
[14] a) P. J. Berti, J. A. B. Mn, Chem. Rev. 2006, 106, 506–555; b) J.
Tchou, V. Bodepdi, S. Shibutani, I. Antoshechkin, J. Miller,
A. P. Grollman, F. Johnson, J. Biol. Chem. 1994, 269, 15318–
15324.
[15] M. Leipold, J. G. Muller, C. J. Burrows, S. S. David, Biochemis-
try 2000, 39, 14984–14992.
[16] a) M. Ober, U. Linne, J. Gierlich, T. Carell, Angew. Chem. 2003,
115, 5097–5101; b) M. Ober, H. Müller, C. Pieck, J. Gierlich,
T. Carell, J. Am. Chem. Soc. 2005, 127, 18143–18149; c) F. Co-
ste, M. Ober, T. Carell, S. Boiteux, C. Zelwer, B. Castaing, J.
Biol. Chem. 2004, 279, 44074–44083.
[1] a) S. Steenken, S. Jovanovic, J. Am. Chem. Soc. 1997, 119, 617–
618; b) H. Yanagawa, Y. Ogawa, M. Ueno, J. Biol. Chem. 1992,
267, 13320–13326.
[2] T. Gimisis, C. Cismas, Eur. J. Org. Chem. 2006, 1351–1378.
[3] a) C. Sheu, C. F. Foote, J. Am. Chem. Soc. 1995, 117, 6439–
6442; b) W. L. Neeley, J. M. Essigmann, Chem. Res. Toxicol.
2006, 19, 491–505; c) J. Cadet, J.-L. Ravanat, G. R. Martinez,
M. H. G. Medeiros, P. Di Mascio, Photochem. Photobiol. 2006,
82, 1219–1225.
[4] J. E. B. McCallum, C. Y. Kuniyoshi, C. S. Foote, J. Am. Chem.
Soc. 2004, 126, 16777–16782.
[5] C. Sheu, C. F. Foote, J. Am. Chem. Soc. 1995, 117, 474–477.
[6] a) J. C. Niles, J. S. Wishnok, S. R. Tannenbaum, Org. Lett.
2001, 3, 963–966; b) W. Adam, M. A. Arnold, M. Gruene,
W. M. Nau, U. Pischel, C. R. Saha-Möller, Org. Lett. 2002, 4,
537–540.
[7] W. Luo, J. G. Muller, E. M. Rachlin, C. J. Burrows, Org. Lett.
2000, 2, 613–616.
[8] a) W. Luo, J. G. Muller, C. J. Burrows, Org. Lett. 2001, 3, 2801–
2804; b) W. Luo, J. G. Muller, E. M. Rachlin, C. J. Burrows,
Chem. Res. Toxicol. 2001, 14, 927–938; c) B. White, M. C. Ta-
run, N. Gathergood, J. F. Rusling, M. R. Smyth, Mol. BioSyst.
2005, 1, 373–381.
[9] a) Y. Ye, J. G. Muller, W. Luo, C. L. Mayne, A. J. Shallop, R. A.
Jones, C. J. Burrows, J. Am. Chem. Soc. 2003, 125, 13926–
13927; b) L. Torun, H. Morrison, Photochem. Photobiol. 2003,
77, 370–375; c) H.-C. DeFedericis, H. B. Patrzyc, M. J. Rajecki,
E. E. Budzinski, H. Ijima, J. B. Dawidzik, M. S. Evans, K. F.
Greene, H. C. Box, Radiat. Res. 2006, 165, 445–451.
[10] a) A. Durandin, L. Jia, C. Crean, A. Kolbanovskiy, S. Ding,
V. Shafirovich, S. Broyde, N. E. Geacintov, Chem. Res. Toxicol.
2006, 19, 908–913; b) B. Karwowski, F. Dupeyrat, M. Bardet,
[17] A. Romieu, D. Gasparutto, D. Molko, J.-L. Ravanat, J. Cadet,
Eur. J. Org. Chem. 1999, 1, 49–56.
[18] F. Johnson, G. Dorman, R. A. Rieger, R. Marumoto, C. R.
Iden, R. Bonala, Chem. Res. Toxicol. 1998, 11, 19–202; R. Ma-
rumoto, T. Hamana, Y. Kashiwabara, Jpn. Kokau, Tokyo Koho
1989, JP01 75499 [8975499].
[19] G. D. Fasman, Optical properties of nucleic acids, absorption,
and circular dichroism spectra, 3 ed., CRC Press, Cleve-
land,Ohio, 1975.
[20] C. R. Cantor, M. M. Warshaw, H. Shapiro, Biopolymers 1970,
9, 1059–1077.
Received: November 10, 2006
Published Online: January 25, 2007
Eur. J. Org. Chem. 2007, 1438–1445
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1445