Alkylation of thymine with 1,2-dibromoethane

Article abstract of DOI:10.1016/S0040-4020(01)00277-0

Alkylation of thymine with 1,2-dibromoethane depends strongly on the reaction conditions. Various alkyl derivatives may be produced, including N-1- or N-3-monosubstituted alkylthymines and products of their cyclisation, as well higher molecular weight products resulting from intermolecular substitution of N-1- and N-3-mono- and N-1,N-3-dialkylthymines. We have identified two cyclic products 5 and 6 and the dialkylated derivative 7, for which detailed structural analyses have been performed.

Full text of DOI:10.1016/S0040-4020(01)00277-0

TETRAHEDRON
Pergamon
Tetrahedron 57 32001) 3979±3985
Alkylation of thymine with 1,2-dibromoethane
B. Nawrot,a,p O. Michalak,a S. Olejniczak,a M. W. Wieczorek,b T. Lisc and W. J. Steca
aDepartment of Bioorganic Chemistry, Laboratory for Spectroscopic Studies, Centre of Molecular and Macromolecular Studies,
Polish Academy of Sciences, Sienkiewicza 112, 90-363 Lodz, Poland
bDepartment of Food Chemistry, Technical University, 90-925 Lodz, Poland
cInstitute of Chemistry, Wroclaw University, 50-383 Wroclaw, Poland
Received 16November 2000; revised 5 February 2001; accepted 22 February 2001
AbstractÐAlkylation of thymine with 1,2-dibromoethane depends strongly on the reaction conditions. Various alkyl derivatives may be
produced, including N-1- or N-3-monosubstituted alkylthymines and products of their cyclisation, as well higher molecular weight products
resulting from intermolecular substitution of N-1- and N-3-mono- and N-1,N-3-dialkylthymines. We have identi®ed two cyclic products 5
and 6 and the dialkylated derivative 7, for which detailed structural analyses have been performed. q 2001 Elsevier Science Ltd. All rights
reserved.
1. Introduction
2. Results
Reaction of 2,4-bis3trimethylsilyloxy)-5-methylpyrimidine
with 1,2-dibromoethane as solvent, in the presence of
sodium iodide at 1008C for 24 h, provided a mixture of
DMSO soluble compounds, which were crystallised from
ethanol. Low solubility in organic solvents and dif®culties
in chromatographic separation of this crystalline material
prompted us to investigate the chemical composition of
There are numerous reports in the chemical literature
concerning the synthesis of N-1-32-aminoethyl)thymine
31).1 Direct aminoethylation of thymine with aziridine,
performed in aqueous solution provides 1, although the
predominant product of this reaction is N-1,N-3-di32-
aminoethyl)thymine accompanied by minute amounts of
N-3-32-aminoethyl)thymine 32). Isolation of pure 1, 2 or
the product of bis-alkylation from this reaction mixture is
a tedious process, since N-aminoalkylation competes with
polymerisation of aziridine.1 An alternative approach to 1
involves reaction of thymine with 1,2-dibromoethane
providing N-1- and N-3-32-bromoethyl)thymines 33 and 4)
with subsequent substitution of the terminal bromine with
azide ion followed by reduction to the desired 1 or 2.
According to a recently published procedure, compound 3
was obtained in 44% yield by the reaction of 2,4-3bis-
trimethylsilyloxy)-5-methylpyrimidine with 1,2-dibromo-
ethane as the solvent at 1008C.2 Interestingly, in another
report, the reaction of thymine with 1,2-dibromoethane
was described as an ef®cient route to 3 3yield 65%). This
reaction was performed in DMF solution at ambient
temperature, using 4 molar equivalents of 1,2-dibromo-
ethane, in the presence of K2CO3.3 Comparison of the
physical properties of compounds 3 presented in both
reports revealed essential differences in melting points,
solubilities and spectral characteristics. Due to our interest
in ®nding an effective synthesis of 1 and 2 we reinvestigated
both the above procedures. The results of our ®ndings are
presented below.
1
the products by means of mass spectrometry and H NMR
3DMSO-d6). Fig. 1 represents a mass spectrum of this
mixture recorded under conditions of chemical ionisation
3positive mode) and reveals a low content of the desired 3
3m/z 233) and the presence of polycondensation products.
1
Such conclusions are supported by the H NMR spectrum,
which contains at least six resonances, of nearly equal inten-
sity, characteristic for thymine 6-H protons. Formation of
polycondensation products, namely N-1,N-10-3alkanediyl)-
bis-thymines, by treatment of thymine with Br3CH2)nBr
3nˆ3±10) in the presence of t-BuOK in DMF, has been
described by Itahara.4 However, with 1,2-dibromoethane,
the reaction gave a cyclic derivative 5 3Fig. 2). Indeed, in
the above-mentioned mass spectrum 3Fig. 1), in addition to
the high molecular mass peaks, we observed a low-intensity
peak of m/z 153, which corresponds to molecular ion of
compound 5. Since the mixture of products obtained in
this reaction of 2,4-bis3trimethylsilyloxy)-5-methylpyrimi-
dine with 1,2-dibromoethane was not of any synthetic value,
the structures of the individual components of the mixture
were not investigated in detail.
To avoid formation of intermolecular alkylation products
we used less drastic conditions for the alkylation of thymine
using 1,2-dibromoethane 34 equiv.) as reagent in DMF as
solvent.3 The reaction was carried out in the presence of
Keywords: alkylation; pyrimidines; oxazolidines; X-ray structures.
Corresponding author. Tel.: 148-42-6816970; fax: 148-42-6815483;
e-mail: bnawrot@mail.lodz.pdi.net
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020301)00277-0

3980
B. Nawrot et al. / Tetrahedron 57 42001) 3979±3985
Figure 1. Mass spectrum of products obtained by reaction of 2,4-bis3trimethylsilyloxy)-5-methylpyrimidine with 1,2-dibromoethane in the presence of
sodium iodide at 1008C for 24 h.
resulting from intermolecular substitution. The predominant
product was isolated in ca. 10% yield and the two minor
products with yields of 1.2% and less than 1%. Only the
most hydrophobic compound, the lowest yielding
product, contained bromine. Spectroscopic analysis
revealed that this compound was the N-1,N-3-bis-alkyl-
derivative 7. The two remaining products did not contain
bromine. The lack of imino protons in the 1H NMR
spectrum, and of bromine isotope peaks in the MS 3m/z
153), indicated their cyclic structure. It was important
Figure 2. Possible cyclic structures of thymidine derivatives obtained by
alkylation of thymine with 1,2-dibromoethane.
to distinguish between the N-1 and N-3 substituted
isomers. By analysis of the H and 13C NMR data we
1
identi®ed the lower yield product as the N-1/O-2 cyclic
derivative 5, already described in the literature.4 NOE
correlations between H-8 and H-6of compound 5 con®rmed
the N-1 substitution 3Fig. 2). X-ray analysis 3details
included in Experimental) of compound 5 con®rmed
unequivocally its N-1/O-2 cyclic structure 3Fig. 3a,b,
Tables 1±3).
K2CO3 at room temperature. After 48 h more than 60% of
thymine was still present in the reaction mixture. Prolonga-
tion of the reaction time did not improve the product/
substrate ratio. Three products were isolated by silica gel
column chromatography 3Scheme 1), in addition to a small
amount 3less than 1%) of high molecular weight material
Scheme 1.

B. Nawrot et al. / Tetrahedron 57 42001) 3979±3985
3981
Figure 3. 3a) Thermal ellipsoidal view with the atom numbering scheme and 3b) crystal packing diagram for compound 5. Ellipsoids are shown with 50%
probability, 3c) thermal ellipsoidal view with the atom numbering scheme and 3d) crystal packing diagram for compound 6. Ellipsoids are shown with 50%
probability.
The main product 310% yield) could be either N-3/O-2 36) or
N-3/O-4 36a) cyclic derivative 3Fig. 2). Simple spectral
contains the experimentally determined 13C chemical shifts
of both isolated cyclic compounds and the calculated shield-
ing parameters siso of the three possible cyclic structures 5,
6 and 6a. The a data correspond to the lower yield product 5
and the b data to the higher yield product 6. Correlation of
experimental chemical shifts with calculated shielding para-
meters gave six possible solutions 3supplementary data).
Comparison of the correlation coef®cients R2 for every
possible combination 3Table 5) remains in agreement with
X-ray results.
1
analysis by means of the H and 13C NMR data did not
give us any clear answer and no appropriate NOEs were
observed. The chemical shift of C-63 d 151.0 ppm) was
considerably deshielded relative to the corresponding
carbon of the heterocyclic ring of compound 5 3d
134 ppm) indicating a change in the nature of the
conjugated system. The structure of the main reaction
product was determined by X-ray analysis 3Fig. 3c,d, Tables
1±3). It occurred that the investigated compound is the N-3/
O-2 cyclic derivative 6.
The calculation method GIAO also allowed the deter-
mination of the theoretical dipole moment for all of
the structures under investigation. Compound 5 has a
much higher theoretical dipole moment 3mˆ8.73 D)
than compound 6 3mˆ2.6 3 D). These data explain
the much higher polarity of compound 5 relative to
compound 6, which results in their very different
mobility on silica gel thin layer chromatography. Rf
values of 0.14 and 0.53 were observed for compound
5 and 6, respectively on TLC in methanol/chloroform
9/1, v/v.
To test usefulness of a GIAO 3B3PW91/6-311G**) method5
for molecular modelling of the three possible cyclic
ethylenethymine structures, N-1/O-2 35), N-3/O-2 36) and
N-3/O-4 36a) 3Fig. 2) we performed calculations using
shielding parameters of 13C NMR spectroscopy. All
calculations were run on a Silicon Graphics Challenge
computer by means of standard methods and functions
routinely implemented in the Gaussian package.5 Optimisa-
tion geometry was done on the level of HF/6-31G*. Table 4

3982
B. Nawrot et al. / Tetrahedron 57 42001) 3979±3985
Table 1. Crystal data and experimental details of X-ray analysis of 5 and 6
Table 3. Bond angles 38) for non-hydrogen atoms
Compound 5
Compound 5
Compound 6
Compound 6
Molecular formula
Formula weight
C7H8N2O2
152.15
monoclinic
P21/m
6.065737)
6.548139)
8.8844312)
90.00
106.424311)
90.00
C7H8N2O2
152.15
Orthorhombic
Pnma
11.6628311)
6.567037)
8.771038)
90.00
90.00
90.00
C2
C2
C6Nx
Nyb
N1
O10
C2
O11
O11
N3
N1
Nxa
C6118.72312)
113.39311)
C8
C8
O10
N3
Nxa
C4
N3
C5
C5
111.73314)
111.91312)
125.63310)
122.31312)
126.20313)
111.49311)
122.46310)
120.00311)
126.52313)
113.48311)
a
Crystallographic system
Space group
129.55313)
121.95313)
126.73314)
111.32313)
116.83312)
119.22313)
121.63314)
119.14312)
C2
C2
C2
N3
C4
C4
C4
C4
C7
C5
Ê
a 3A)
Ê
b 3A)
Ê
c 3A)
a 38)
b 38)
g 38)
Ê 3
V 3A )
Z
338.4838)
2
1.493
671.77311)
4
1.504
C6C5
C6C5
C4
118.94314)
122.94313)
118.12313)
118.51313)
124.03312)
Dc 3g/cm3)
C7
117.46312)
m [cm21
]
1.12
0.25£0.35£0.40
1.13
0.30£0.30£0.70
C5
C6N1
C8
C9
43131) 19.6
125.95312)
101.44313)
107.16313)
108.35313)
Crystal dimensions 3mm)
Maximum 2u 38)
Ê
Radiation, l 3A)
Scan mode
hkl ranges:
Nxa
O10
C2
C9
C8
C9
101.39310)
106.36311)
108.85310)
60.14
MoKa, 0.71073
v
61.86
MoKa, 0.71073
v
O10
a
NxˆN1 for Compound 5, NxˆN3 for compound 6
NyˆN3 for Compound 5, NyˆN1 for compound 6
b
hˆ
26 8
28 8
212 11
939
939
215 15
28 8
210 11
966
966
904
kˆ
lˆ
Table 4. Experimental 13C chemical shift parameters of cyclic products a
and b and theoretical shielding parameters calculated for a three possible
cyclic structures
No. of re¯ections: unique
with I.0s3I)
obs. with I.2s3I)
789
8686
No. of parameters re®ned
Largest diff. Peak 3eA
Largest diff. Hole 3eA
shift/esd max
Robs
wRobs
Sobs
Rint
Ê 23
Ê 23
Carbon number d [ppm]
s
iso [ppm]
)
)
0.3460.315
20.238
0.000
0.0420
0.1188
1.113
0.0168
20.267
0.000
0.0411
0.1091
1.117
a
b
N-1/O-2
N-3/O-2 N-3/O-4
2
4
160.6
171.7
115.7
158.0 28.9
161.5 22.4
116.9 61.4
34.3
170.6 173.0 .5 176
42.4
66.1
29.8
28.8
66.9
36.7
25.4
97.1
5
6134.4
7
8
9
0.0372
151.0 58.1
13.612.4
46.2
19.1
140.1
123.6
144.6
122.8
142.5
120.4
66.6
Ê
Table 2. Bond lengths 3A) for non-hydrogen atoms
Compound 5
Compound 6
1.29632)
Table 5. Correlation coef®cients R2 calculated for both isolated cyclic
products and their possible cyclic structures 3N-1/O-2, N-3/O-2 and N-3/
O-4)
N1
N1
Nxa
C2
C2
N3
C4
C4
C5
C5
C8
C9
C2
C61.37132)
C8
N3
O10
C4
O11
C5
1.35232)
1.39032)
1.45532)
1.29532)
1.33732)
1.39032)
1.24232)
1.45932)
332)
1.49532)
1.50233)
1.45432)
1.46132)
1.35932)
1.33832)
1.39332)
1.23032)
1.45032)
Compound
Possible cyclic structure
N-1/O-2
N-3/O-2
N-3/O-4
a
b
0.9918
0.9733
0.9740
0.9986
0.9155
0.9552
C61.35032)
C7
C9
1.36
1.49632)
1.52732)
1.46732)
O10
32-aminoethyl)thymine 1 3Scheme 2). An analogous
transformation of the cyclic derivative 6 gave the N-3-32-
aminoethyl)thymine 2 3Scheme 3). On the other hand, heat-
ing compound 6 with aqueous ammonium hydroxide gave
predominantly N-3-32-hydroxyethyl)thymine 10 340%) in
addition to the amine 2, obtained in 30% yield. This N-3-
hydroxyethyl-derivative 10 was not identi®ed in a direct
hydroxyalkylation of thymine, performed with an equimolar
amount of ethylene carbonate in the presence of a catalytic
amount of sodium hydroxide.8 N-1-32-Hydroxyethyl)-
thymine 370%) and a small amount of the N-1,N-3-bis-
alkylated derivative were the only identi®ed products.
a
NxˆN1 for Compound 5, NxˆN3 for compound 6
The desired N-1-32-bromoethyl)thymine 3 was ®nally
obtained by alkylation of 2,4-bis-3trimethylsilyloxy)-5-
methylpyrimidine with 1,2-dibromoethane under mild
conditions 3ambient temperature, 10 days).6 The yield of
the reaction was very modest 310%) and much unreacted
substrate remained. However we did not observe formation
of the N-3-alkylation product 4 or any other cyclic or higher
substituted products. The spectral data of compound 3 were
identical to those described earlier in the literature.2,4
Reaction of compound 3 or the cyclic oxazolopyrimidine 5
with trimethylsilylazide in the presence of ¯uoride ions7
gave a good yield 370±85%) of azide 8. Catalytic reduction
of compound 8 resulted in the formation of the desired N-1-
3. Discussion
In general, N-alkylation of pyrimidine bases gives predomi-
nantly N-1-monosubstituted and N-1,N-3-bis-substituted

B. Nawrot et al. / Tetrahedron 57 42001) 3979±3985
3983
Scheme 2.
Scheme 3.
derivatives rather than N-3 or acid-labile O-alkyl deriva-
tives.9 The regioselectivity of this reaction is determined
by the acidity of the two ionisable protons of the hetero-
cyclic ring. The acidity of the N-1 proton is much higher
3pKaˆ9.43 for uracil, and 9.86for thymine) than that of the
N-3 proton 3pKa.13 for uracil,10 13.96for thymine 11).
Despite the large difference in the pKa values of the protons
it is not possible to obtain the N-1 product selectively, since
N-1 substitution of the pyrimidine ring generates a signi®-
cant increase in the acidity of the N-3 proton 3pKaˆ9.71
for an N-1 substituted uracil12,13). The ratio of N-1-mono- to
N-1,N-3-bis-substituted products depends on the reaction
conditions, mostly on the molar ratio of the heterocyclic
base and the alkylating agent.9,14 When dibromoalkanes
[Br3CH2)nBr 3nˆ3±10)] were used as alkylating agents,
formation of polycondensation products namely N-1,N-10-
3alkanediyl)bis-thymines was observed.4 However, alkyl-
ation of thymine with 1,2-dibromoethane in the presence
of t-BuOK resulted in the formation of the cyclic derivative
5.4 We isolated such a N-1/O-2 cyclic derivative 5 as a
minor reaction product only when 1,2-dibromoethane was
used in four-fold excess over thymine, and the reaction was
carried out in the presence of potassium carbonate. In such
conditions, however, the main reaction product was N-3/O-2
cyclic derivative 6. Formation of such N-3/O-2 cyclic
derivatives most probably occurs via an alkylation of N-3
nitrogen atom with 1,2-dibromoethane followed by subse-
quent nucleophilic substitution of the second bromine by the
oxygen atom at position 2 of the heterocyclic ring. Such
nucleophilic substitution by the oxygen O-2 is rather
surprising, since it is well known that the oxygen atom
O-4 of the nucleobase ring is more electronegative than
oxygen O-215 and thus one would expect formation of an
N-3/O-4 cyclic derivative 6a rather than an N-3/O-2 isomer
6. In none of the performed reactions we were able to isolate
N-3-32-bromoethyl)thymine 4, the product of selective
substitution at N-3 position and the precursor of cyclic
derivative 6. Alkylation at ambient temperature without
catalyst resulted in monoalkylation of thymine at position
N-1 3product 3), although the yield of this reaction was
rather low. A small amount of polycondensation product
was present in all the reactions, while a considerable amount
of substrate was always left. When the alkylation reaction
was performed under drastic conditions 31008C, without
solvent) we observed, as expected, the formation of poly-
alkylated products rather than monosubstituted thymines.

3984
B. Nawrot et al. / Tetrahedron 57 42001) 3979±3985
4. Conclusions
MeOH, 9/1), mp 197±2008C, 3199±2008C,2 200±2028C,4
164±1658C3), 1H NMR 3200 MHz, DMSO-d6, ppm):
11.33 3br s, 1H, NH), 7.55 3q, Jˆ1.1 Hz, 1H, H-6), 4.02
3t, Jˆ12.6Hz, 2H, CH 2N, 3.69 3t, Jˆ12.6Hz, 2H,
CH2Br), 1.75 3d, Jˆ1.1 Hz, 3H, CH3), FAB MS 233
[M11]1, 235 [M13]1.
In summary, the outcome of the alkylation of thymine with
1,2-dibromoethane depends strongly on the reaction condi-
tions. Various alkyl derivatives may be produced, including
N-1- or N-3-monosubstituted alkylthymines and products of
their cyclisation, as well higher molecular weight products
resulting from intermolecular substitution of N-1- and N-3-
mono- and N-1,N-3-dialkylthymines. We have identi®ed
two cyclic products 5 and 6 and the dialkylated derivative
7, for which detailed structural analyses have been
performed.
5.1.1. N-1--2-azidoethyl)thymine 8. To a solution of
bromide 3 3520 mg, 2.24 mmol) in anhydrous THF
320 mL), azidotrimethylsilane 30.6mL) and tetrabutyl-
ammonium ¯uoride 34.5 mL) were added. The reaction
mixture was stirred at 658C for 3 h. After removal of the
solvent, the residue was crystallised from ethanol. The
product 8 was obtained as a white solid 3320 mg, 80%).
Mp 152±1558C, Rfˆ0.61 3ethyl acetate), 1H NMR
3200 MHz, CDCl3, ppm): 7.02 3s, 1H, H-6), 3.82 3br q,
2H, CH2), 3.67 3br q, 2H, CH2), 1.94 3s, 3H, CH3), IR
3KBr, cm21): 2122; HR MS for C7H9N5O2: [M1H]1
196.08345 3calc.), 196.08407 3exp.). Synthesis of azide 8
was performed analogously as described above starting
from cyclic compound 5 376mg, 0.5 mmol). The product
was obtained as a white solid 355 mg, 56%).
5. Experimental
5.1. Alkylation of thymine
Procedure A: 1,2-Dibromoethane 335.52 g, 189 mmol) was
added to a suspension of thymine 36g, 47.6mmol) and
anhydrous potassium carbonate 316.32 g, 124 mmol) in
anhydrous DMF 3100 mL). The mixture was kept at room
temperature for 48 h, then the solid was ®ltered off and the
solvent was evaporated under reduced pressure. The reac-
tion products were separated by ¯ash column chromato-
graphy on silica gel 60G in gradient of methanol in
chloroform.
5.1.2. N-3--2-azidoethyl)thymine 9. Synthesis of azide 9
was performed analogously as described above starting
from cyclic compound 6 3340 mg, 2.23 mmol). The product
was obtained as a white solid 3320 mg, 73%). Mp 119±
1208C, Rfˆ0.75 3ethyl acetate), 1H NMR 3200 MHz,
CD3OD, ppm): 7.24 3q, Jˆ1.1 Hz, 1H, H-6), 4.15 3t,
Jˆ6.2 Hz, 2H, CH2), 3.50 3t, Jˆ6.2 Hz, 2H, CH2), 1.88
3d, Jˆ1.1 Hz, 3H, CH3); IR 3KBr, cm21): 2110, 2088; HR
MS for C7H9N5O2: [M1H]1 196.08345 3calc.), 196.08269
3exp.).
Compound
5 32,3-dihydro-6-methyl-7H-oxazolo[3,2-a]-
pyrimidin-7-one) 30.09 g, 1.2%): Rfˆ0.14 3CHCl3/MeOH,
9/1), mp 285±2898C,4 1H NMR 3200 MHz, DMSO-d6,
ppm): 7.63 3q, Jˆ1.2 Hz, 1H, H-6), 4.64 3t, Jˆ8.45 Hz,
2H, CH2O), 4.18 3t, Jˆ8.45 Hz, 2H, CH2N), 1.77 3d,
Jˆ1.2 Hz, 3H, CH3), FAB MS 153 [M11]1, IR 3KBr,
cm21): 3049, 2922, 1667, 1611, 1550, 1504, 1377, 880.
5.1.3. N-1--2-aminoethyl)thymine 1. 10% Pd/C 320 mg)
was added to the solution of azide 8 3300 mg, 1.54 mmol)
in methanol 310 mL) and hydrogen was bubbled through the
reaction mixture for 2.5 h 3TLC control). The reaction
mixture was ®ltered and the solvent removed under reduced
pressure. The crude product was puri®ed by ¯ash column
chromatography in gradient of methanol in chloroform.
Amine 1 was obtained as a white solid 3230 mg, 90%).
Mp 180±1858C; Rfˆ0.08 32£CHCl3/MeOH, 8/2); 1H
NMR 3200 MHz, CD3OD): 7.40 3q, Jˆ1.2 Hz, 1H, H-6),
3.763t, Jˆ6.28 Hz, 2H, CH2), 2.89 3t, Jˆ6.28 Hz, 2H,
CH2), 1.87 3d, Jˆ1.2 Hz, 3H, CH3); FAB MS 170
[M11]1, 168 [M-1]2; IR 3KBr, cm21): 3342, 3289, 3054,
2960, 1705, 1660, 1481, 1358, 1046, 760.
Compound
6 32,3-dihydro-6-methyl-5H-oxazolo[2,3-a]-
pyrimidin-5-one) 30.70 g, 10.0%): Rfˆ0.53 3CHCl3/
MeOH, 9/1), Rfˆ0.15 3ethyl acetate), mp.3908C, 1H
NMR 3200 MHz, DMSO-d6, ppm): 7.57 3q, Jˆ1.1 Hz,
1H, H-6), 4.67 3t, Jˆ8.45 Hz, 2H, CH2O), 4.29 3t,
Jˆ8.45 Hz, 2H, CH2N), 1.863d, Jˆ1.1 Hz, 3H, CH3); HR
MS for C7H8O2N2: [M]1 152.05858 3calc.), 152.05731
3exp.); IR 3KBr, cm21): 2922, 1663, 1611, 1561, 1437,
1358, 1256, 1000, 766.
Compound 7 30.118 g, 0.9%): Rfˆ0.53 3CHCl3/MeOH, 9/
1
1), Rfˆ0.52 3ethyl acetate), H NMR 3200 MHz, CDCl3,
ppm): 7.05 3q, Jˆ1.2 Hz, 1H, H-6), 4.03 3t, Jˆ6.0 Hz, 2H,
CH2), 3.58 3t, Jˆ6.0 Hz, 2H, CH2), 1.84 3d, Jˆ1.2 Hz, 3H,
CH3); HR MS for C9H13O2N2Br2379): 338.93437 3calc.)
338.93372 3exp.), for C9H13O2N2Br379)Br381): 340.93233
3calc.), 340.931263exp.).
5.1.4. N-3--2-aminoethyl)thymine 2. Synthesis of amine 2
was performed analogously as described above starting
from azide 9 3340 mg, 2.23 mmol). The product was
obtained as a white foam 3320 mg, 73%). Rfˆ0.16
32£CHCl3/MeOH, 8/2), 1H NMR 3200 MHz, CD3OD,
ppm): 7.23 3q, Jˆ1.2 Hz, 1H, H-6), 4.05 3t, Jˆ6.4 Hz, 2H,
CH2), 2.93 3t, Jˆ6.4 Hz, 2H, CH2), 1.88 3d, Jˆ1.2 Hz, 3H,
CH3); HR MS for C7H11N3O2: [M1H]1 170.09295 3calc.),
170.09392 3exp.); IR 3KBr, cm21): 3343, 3281, 3058, 2895,
1703, 1637, 1343, 1000, 770.
Procedure B: A solution of 2,4-bis3trimethylsilyloxyl)-5-
methylpyrimidine16 39.0 g, 33.3 mmol) in 1,2-dibromo-
ethane 3115 mL) was stirred for 10 days at room tempera-
ture. Then water 3300 mL) was added to the reaction
mixture and the products were extracted with chloroform
33£300 mL). After removal of solvents, the residue was
poured onto the silica gel column. The product was eluted
with gradient of methanol in chloroform. Bromide 3 30.78 g,
10%) was obtained as a white solid. Rfˆ0.59 3CHCl3/
5.1.5. N-3--2-hydroxyethyl)thymine 10 and N-3--2-
aminoethyl)thymine 2. Cyclic derivative 6 3390 mg,
2.56mmol) was treated with aqueous ammonium hydroxide

B. Nawrot et al. / Tetrahedron 57 42001) 3979±3985
3985
320 mL) at 508C for 16h 3TLC). After removal of solvent
References
the mixture was separated by ¯ash column chromatography
on silica gel with a gradient of methanol in chloroform.
Hydroxy-derivative 10 3Rfˆ0.36in CHCl 3/MeOH, 9/1)
was isolated as a fast eluting product 3170 mg, 40%) and
amine 2 as a slow eluting product 3120 mg, 30%). 10: Mp
1. Markiw, R. J. Org. Chem. 1972, 37, 2165±2168.
2. Adams, D. R.; Boyd, A. S. F.; Ferguson, R.; Grieson, D. S.;
Monneret, C. Nucleosides Nucleotides 1998, 17, 1053±1075.
3. Ciapetti, P.; Taddei, M. Tetrahedron 1998, 54, 11305±11310.
4. Itahara, T. Bull. Chem. Soc. Jpn. 1997, 70, 2239±2247.
5. Frish, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E.,
Robb, M. A., Cheeseman, J. R., Zakrzewski, V. G., Montgom-
ery, J. A. J., Stratmann, R. E., Burant, J. C., Dapprich, S.,
Millam, J. M., Daniels, A. D., Kudin, K. N., Strain, M. C.,
Farkas, O., Tomasi, J., Barone, V., Cossi, M., Cammi, R.,
Mennucci, B., Pomelli, C., Adamo, C., Clifford, S., Ochterski,
J., Petersson, G. A., Ayala, P. Y., Cui, Q., Morokuma, K.,
Malick, K., Rabuck, A. D., Raghavachari, K., Foresman, J.
B., Cioslowski, J., Oritz, J. V., Stefanov, B. B., Liu, G.,
Liashenko, A., Piskorz, P., Komaromi, I., Gomperts, R.,
Martin, R. L., Fox, D. J., Keith, T., Al-Laham, M. A., Peng,
C. Y., Nanayakkara, A., Gonzalez, C., Challacombe, M., Gill,
P. M. W., Johnson, B., Chen, W., Wong, M. W., Andres, J. L.,
Head-Gordon, M., Replogle, E. S., Pople, J. A., Gaussian 98,
Revision A.6, Gaussian, Pittsburgh, PA, 1998.
1
170±1758C, H NMR 3200 MHz, DMSO-d6, ppm): 10.83
3d, Jˆ5.5 Hz, 1H, NH), 7.28 3dq, Jˆ1.2 Hz, Jˆ5.5, 1H,
H-6), 4.72 3t, Jˆ5.8 Hz, 1H, OH), 3.85 3t, Jˆ6.7 Hz, 2H,
CH2), 3.463dt, Jˆ6.7 Hz, Jˆ5.8 Hz, 2H, CH2), 1.763d,
Jˆ1.2 Hz, 3H, CH3); HR MS for C7H10N2O3: [M1H]1
171.07697 3calc.), 171.07693 3exp.); IR 3KBr, cm21):
3423, 3221, 3075, 1712, 1637, 1449, 1216, 1084, 767.
5.2. X-Ray analysis
Crystal and molecular structures of 5 and 6 were determined
using data collected at low temperature on a KM4 diffract-
ometer 3KUMA Diffraction Instruments GmbH) with
graphite monochromatized MoKa radiation. Compound 5
crystallises from ethanol in monoclinic system, in space
group P21/m with the unit cell consisting of 2 molecules.
Compound 6 crystallises from chloroform in orthorhombic
system, in space group Pnma with the unit cell consisting of
4 molecules. Crystal data and experimental details are
shown in Table 1. The lattice constants were re®ned by
least-squares ®t of 2138 re¯ections in the u range 3.648±
30.078 for 5 and 3731 re¯ections in the u range 3.888±
30.938 for 6. For 5 and 6, respectively, a total of 939 and
966 independent re¯ections with I$0 were used to solve the
structure by direct methods and to re®ne it by full matrix
least-squares using F2.17,18 Hydrogen atoms were found on a
difference Fourier map and re®ned isotropically. Aniso-
tropic thermal parameters were re®ned for all non-hydrogen
atoms. The ®nal re®nement of 5 converged to Rˆ0.0420
for 86re®ned parameters and 789 observed re¯ections with
I$2s3I). The ®nal re®nement of 6 converged to Rˆ0.0411
for 86re®ned parameters and 904 observed re¯ections with
I$2s3I). Data processing was carried out with the
KM4CCD software, structure solution shelxs17 and
structure re®nement shelxl.18
6. Browne, D. T.; Eisinger, J.; Leonard, N. J. J. Am. Chem. Soc.
1968, 90, 7302±7323.
7. Soli, E. D.; DeShong, P. J. Org. Chem., 1999, 64, 9724±9726.
8. Gi, H.-J.; Xiang, Y.; Schinazi, R. F.; Zhao, K. J. Org. Chem.
1997, 62, 88±92.
9. Shaw, G. In Comprehensive Heterocyclic Chemistry,
Katritzky, A. R., Rees, C. W., Eds.; Pergamon: Oxford,
1985; pp 499.
10. Jonas, J.; Gut, J. Collect. Czech. Chem. Commun. 1962, 27,
716.
11. Ganguly, S.; Kundu, K. K. Can. J. Chem. 1994, 72, 1120±
1126.
12. Nakanishi, K.; Suzuki, N.; Yamazaki, F. Bull. Chem. Soc. Jpn.
1961, 34, 53.
13. Wittenburg, E. Chem. Ber. 1966, 99, 2391.
14. Yamauchi, K.; Kinoshita, M. J. Chem. Soc., Perkin Trans 1
1973, 391±392.
15. Miller, K. J. J. Am. Chem. Soc. 1990, 112, 8533±8542.
16. Nishimura, T.; Iwai, I. Chem. Pharm. Bull. 1964, 12, 352.
17. G. M. Sheldrick, G. M. Kruger, R. Goddard, shelxs-86.
Program for the Solution of Crystal Structures, University of
The authors have deposited all crystallographic data for
these structures with the Cambridge Crystallographic Data
Centre19 under CCDC 156329 and 156330 deposition
numbers for compounds 5 and 6, respectively.
È
È
Gottingen: Gottingen, Germany 1986.
18. G. M. Sheldrick, shelxl-93. Program for the Re®nement of
È È
Crystal Structures, University of Gottingen: Gottingen,
Germany, 1993.
19. University Chemical Laboratory, Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, United
Kingdom.
Acknowledgements
Results presented here were obtained within the project
assisted by the State Committee for Scienti®c Research
3KBN, Grant No. 3T09A 03917 for B. N.).