5-Substituted N4-hydoxy-2'-deoxycytidines and their 5'-monophosphates: Synthesis, conformation, interaction with tumor thymidylate synthase, and in vitro antitumor activity

Article abstract of DOI:10.1021/jm000975u

Convenient procedures are described for the synthesis of 5-substituted, N⁴,hydroxy-2'-deoxycytidines 5a,b,d-h via transformation of the respective 5-substituted 3',5'-di-O-acetyl-2'-deoxyuridines 1a-c,e-h. These procedures involved site-specific triazolation or N-methylimidazolation at position C(4), followed by hydroxylamination and deblocking with MeOH-NH₃. Nucleosides 5a,b,d-h were selectively converted to the corresponding 5'-monophosphates 6a,b,d-h with the aid of the wheat shoot phosphotransferase system. Conformation of each nucleoside in D₂O solution, deduced from ¹H NMR spectra and confirmed by molecular mechanics calculations, showed the pentose ring to exist predominantly in the conformation S (C-2'-endo) and the N⁴-OH group as the cis rotamer. Cell growth inhibition was studied with two L5178Y murine leukemia cell lines, parental and 5-fluoro-2'-deoxyuridine (FdUrd)-resistant, the latter 70-fold less sensitive toward FdUrd than the former. With FdUrd-resistant L5178Y cells, 5-fluoro-N⁴-hydroxy-2'-deoxycytidine (5e) caused almost 3-fold stronger growth inhibition than FdUrd; 5e was only some 3-fold weaker growth inhibitor of the resistant cells than of the parental cells. Thymidylate synthase inhibition was studied with two forms of the enzyme differing in sensitivities toward 5-fluoro-2'-deoxyuridine 5'-monophosphate (FdUMP), isolated from parental and FdUrd-resistant L1210 cell lines. All N⁴-hydroxy-dCMP (6a,b,d-h) and dUMP analogues studied were competitive vs dUMP inhibitors of the enzyme. Analogues 6b,d-h and 5-hydroxymethyl-dUMP, similar to N⁴-hydroxy-dCMP (6a) and FdUMP, were also N⁵,N¹⁰-methylenetetrahydrofolate-dependent, hence mechanism-based, slow-binding inhibitors. 5-Chloro-dUMP, 5-bromo-dUMP, and 5-iodo-dUMP, similar to dTMP, did not cause a time-dependent inactivation of the enzyme. Instead, they behaved as classic inhibitors of tritium release from [5-³H]dUMP. 5-Bromo-dUMP and 5-iodo-dUMP showed substrate activity independent of N⁵,N¹⁰-methylenetetrahydrofolate in the thymidylate synthase-catalyzed dehalogenation reaction. The =N-OH substituent of the pyrimidine C(4) prevented the enzyme-catalyzed release from the C(5) of Br^- and I^- (the same shown previously for H⁺). While FdUMP and 6a showed a higher affinity and greater inactivation power with the parental cell than FdUrd-resistant cell enzyme, an opposite relationship could be seen with 5-hydroxymethyl-dUMP.

Full text of DOI:10.1021/jm000975u

J . Med. Chem. 2000, 43, 4647-4656
4647
5-Su bstitu ted N4-Hyd r oxy-2′-d eoxycytid in es a n d Th eir 5′-Mon op h osp h a tes:
Syn th esis, Con for m a tion , In ter a ction w ith Tu m or Th ym id yla te Syn th a se, a n d in
Vitr o An titu m or Activity
Krzysztof Felczak,† Agnieszka Miazga,† J arosław Poznan´ski,† Maria Bretner,† Tadeusz Kulikowski,*,†
J olanta M. Dzik,§ Barbara Gołos,§ Zbigniew Zielin´ski,§ J oanna Cies´la,§ and Wojciech Rode§
Institute of Biochemistry and Biophysics, Polish Academy of Sciences, 5a Pawin´skiego Street, 02-106 Warszawa, Poland, and
Nencki Institute of Experimental Biology, Polish Academy of Sciences, 3 Pasteur Street, 02-093 Warszawa, Poland
Received May 1, 2000
Convenient procedures are described for the synthesis of 5-substituted N4-hydroxy-2′-
deoxycytidines 5a ,b,d -h via transformation of the respective 5-substituted 3′,5′-di-O-acetyl-
2′-deoxyuridines 1a -c,e-h . These procedures involved site-specific triazolation or N-meth-
ylimidazolation at position C(4), followed by hydroxylamination and deblocking with MeOH-
NH3. Nucleosides 5a ,b,d -h were selectively converted to the corresponding 5′-monophosphates
6a ,b,d -h with the aid of the wheat shoot phosphotransferase system. Conformation of each
nucleoside in D2O solution, deduced from 1H NMR spectra and confirmed by molecular
mechanics calculations, showed the pentose ring to exist predominantly in the conformation S
(C-2′-endo) and the N4-OH group as the cis rotamer. Cell growth inhibition was studied with
two L5178Y murine leukemia cell lines, parental and 5-fluoro-2′-deoxyuridine (FdUrd)-resistant,
the latter 70-fold less sensitive toward FdUrd than the former. With FdUrd-resistant L5178Y
cells, 5-fluoro-N4-hydroxy-2′-deoxycytidine (5e) caused almost 3-fold stronger growth inhibition
than FdUrd; 5e was only some 3-fold weaker growth inhibitor of the resistant cells than of the
parental cells. Thymidylate synthase inhibition was studied with two forms of the enzyme
differing in sensitivities toward 5-fluoro-2′-deoxyuridine 5′-monophosphate (FdUMP), isolated
from parental and FdUrd-resistant L1210 cell lines. All N4-hydroxy-dCMP (6a ,b,d -h ) and
dUMP analogues studied were competitive vs dUMP inhibitors of the enzyme. Analogues
6b,d -h and 5-hydroxymethyl-dUMP, similar to N4-hydroxy-dCMP (6a ) and FdUMP, were also
N5,N10-methylenetetrahydrofolate-dependent, hence mechanism-based, slow-binding inhibitors.
5-Chloro-dUMP, 5-bromo-dUMP, and 5-iodo-dUMP, similar to dTMP, did not cause a time-
dependent inactivation of the enzyme. Instead, they behaved as classic inhibitors of tritium
release from [5-3H]dUMP. 5-Bromo-dUMP and 5-iodo-dUMP showed substrate activity
independent of N5,N10-methylenetetrahydrofolate in the thymidylate synthase-catalyzed de-
halogenation reaction. The dN-OH substituent of the pyrimidine C(4) prevented the enzyme-
catalyzed release from the C(5) of Br- and I- (the same shown previously for H+). While FdUMP
and 6a showed a higher affinity and greater inactivation power with the parental cell than
FdUrd-resistant cell enzyme, an opposite relationship could be seen with 5-hydroxymethyl-
dUMP.
In tr od u ction
involves a time-dependent formation of a ternary co-
valently bound complex of the enzyme with FdUMP and
N5,N10-methylenetetrahydrofolate, in a reaction similar
to that with dUMP. However, at this step the reaction
stops, as the C(5)-fluorine fails to dissociate (due to the
strength of the C-F bond). This results in slowly
reversible enzyme inactivation.
Thymidylate synthase (EC 2.1.1.45) catalyzes the
dUMP methylation reaction involving a concerted trans-
fer and reduction of the one-carbon group (at the
aldehyde oxidation level) of N5,N10-methylenetetrahy-
drofolate, with concomitant production of thymidylate
and dihydrofolate.1,2Active forms of several drugs used
in anticancer, antiviral, and antifungal chemotherapy
are thymidylate synthase inhibitors being either dUMP
or N5,N10-methylenetetrahydrofolate analogues.3-6
Among the dUMP analogues that are good inhibitors
of thymidylate synthase, most involve an electron-
withdrawing substituent at the pyrimidine C(5)-posi-
tion.7-9 The most prominent example is 5-fluoro-dUMP
(FdUMP). Inhibition of thymidylate synthase by FdUMP
A rare example of a dUMP analogue that is C(4)-
substituted and, nevertheless, a strong inhibitor of
thymidylate synthase,10 is N4-hydroxy-2′-deoxycytidine
5′-monophosphate (6a ). Similar to FdUMP, it is a slow-
binding inhibitor, covalently bound by the enzyme, and
inactivating it in the presence of N5,N10-methylenetet-
rahydrofolate.10-12 5-Fluoro substitution in 6a potenti-
ated inhibition, although in 5-fluoro-N4-hydroxy-2′-de-
oxycytidine 5′-monophosphate (6e) the N4-OH substitu-
ent probably remained the cause of inactivation.12
* To whom correspondence should be addressed. Tel/Fax: 48-39-
121623. E-mail: tk@ibbrain.ibb.waw.pl.
To pursue those studies, we decided to synthesize a
series of new 5-substituted N4-hydroxy-2′-deoxycytidine
† Institute of Biochemistry and Biophysics.
§ Nencki Institute of Experimental Biology.
10.1021/jm000975u CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/11/2000

4648 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24
Felczak et al.
Sch em e 1a
5′-phosphates, with different electronic, hydrophobic,
and steric properties of their 5-substituents and to
compare interactions with thymidylate synthase of those
analogues. The nucleoside forms of the analogues were
expected to be potential leukemia cell growth inhibitors.
Ch em istr y
After the observation that hydroxylamine in aqueous
solutions reacts selectively with cytosine residues in
DNA,13 N4-hydroxycytosine (N4-OH-Cyt) and its ana-
logues were synthesized by direct treatment of cytosine
and cytidine with aqueous or anhydrous hydroxylamine,
followed by mild acid treatment at an elevated temper-
ature.14 The reaction is dependent on the addition of
hydroxylamine to the 4,5-double bond and replacement
of the amino group by hydroxylamine. This led to 4,6-
dihydroxylamino derivatives which were converted into
the corresponding N4-hydroxycytosines by acid catalysis
at elevated temperature. This reaction was also applied
to the preparation of 6a and dCDP, as well as to
hydroxylamination of poly(C).11,15,16 However, when this
procedure was applied to 2′-deoxycytidine, the yield was
very poor (16%) due to the substantial deamination and
cleavage of glycosidic bond under acidic conditions and
elevated temperature.
Better results were obtained by the treatment of 2′-
deoxy-3′,5′-di-O-benzoyl-4-thiouridine with a methanolic
solution of hydroxylamine at reflux temperature to give
N4-hydroxy-2′-deoxycytidine (5a ) in 48% yield.17 In the
case of 5-fluoro-N4-hydroxy-2′-deoxycytidine (5e) it was
advantageous to S-methylate 2′-deoxy-5-fluoro-4-thio-
uridine with diazomethane and to treat the resulting
4-methylthio derivative with hydroxylamine at high
dilution. Under these conditions the 4,6-dihydroxyl-
amino derivative was not formed, 5e being the single
product.18
a
Reagents: (a) 1,2,4-triazole, POCl3, TEA; (b) MeIm, POCl3,
Py; (c), (d) NH2OH; (e) NH3/MeOH; (f) p-nitrophenyl phosphate/
wheat shoot phosphotransferase, 37 °C.
photransferase-4-nitrophenyl phosphate system26 which
gave exclusively 5′-phosphates in 50-70% yields. These
compounds proved to be good substrates for snake
venom phosphodiesterase (5′-nucleotidase) to give ex-
clusively their mother nucleosides 5a ,b,d -h .
Another method of preparation leading to N4-hy-
droxycytidines was based on hydroxylamination of the
corresponding O′-acetylated 4-chlorouridines. However,
a four-step procedure and the instability of 4-chloro
nucleosides led to poor yields.19 More recently it was
announced that some 2′-deoxyribofuranosyl-4-(1,2,4-
triazol-1-yl)pyrimidin-2(1H)-ones reacted smoothly with
hydroxylamine at room temperature.20 We decided to
apply this modified procedure to the preparation of O′-
acetylated 5-substituted N4-hydroxy-2′-deoxycytidines
4a -c,e,f,h .
Con for m a tion a l Asp ects
Molecular mechanics calculations of a series of C(5)-
substituted N1-methyl-N4-hydroxycytosines were made
with the aid of the Sybyl package. The tripos force field27
was used with Pullman atomic charges. In the absence
of the explicit water molecules the electrostatic interac-
tions were scaled by the distance-dependent dielectric
constant ꢀ ) 4.5r. The initial geometry was adapted
from the Sybyl database coordinates of the cytosine.
According to quantum mechanical calculations28 the
imino form of the N(4) nitrogen was assumed.
2′-Deoxy-3′,5′-di-O-acetyluridines21,22
1a -c,e-h
(Scheme 1) were prepared according to Robins’ proce-
dure23 and subjected to site-specific triazolation24 at C(4)
with triazolide formed from 1,2,4-triazole, POCl3, and
TEA. This procedure gave 1,2,4-1H-triazolated nucleo-
sides 2a -c,e,f,h in good yield (60-90%).
1H NMR spectra of the compounds in neutral aqueous
medium (D2O) were recorded on Varian Unity Plus (500
MHz) spectrometer at 25 °C with the standard program
of the solvent signal suppression via presaturation.
The conformational analysis of the deoxyribose moiety
was based on 1H NMR spectroscopic data. The seven
spin systems were analyzed by the iterative LAOCOON-
like29 program. The sugar ring puckering parameters
To improve the yield and to shorten the preparation
of 4f,g, we applied the “one-pot” procedure25 employing
N-methylimidazolide intermediates, hitherto not used
in the synthesis of N4-hydroxy-2′-deoxy nucleosides,
which increased the yield of 3f,g to ca. 80%. The
standard deprotection of acetyl groups with MeOH-
NH3 led to the free 5-substituted N4-hydroxy-2′-deoxy-
cytidines 5a ,b,d -h in ca. 90% yield.
were estimated from the vicinal H-1H coupling con-
1
stants by Remin interpretation30 of the algorithm
proposed originally by Altona.31
Con for m ation of th e N4-OH Gr ou p. Detailed quan-
tum mechanical calculations of Les´ et al.28 demonstrated
that in both N4-hydroxycytosine and its 5-fluoro ana-
logue, the imino form of the N(4) nitrogen was strongly
5′-Monophosphates of the N4-hydroxy-2′-deoxycy-
tidines 6a ,b,d -h were prepared via selective phospho-
rylation at the 5′-position with the wheat shoot phos-

Deoxycytidines and Their 5′-Monophosphates
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24 4649
Ta ble 1. Energy Terms and Structural Details Leading to Cis N4-OH Conformer Stabilization of 5-Substituted
N4-Hydroxy-2′-deoxycytidines
substituent
5-H
5-F
5-Cl
trans
Energy Terms (kcal/mol)
5-Br
5-I
5-Me
5-CH2OH
cis trans
cis
trans
cis
trans
cis
cis
trans
cis
trans
cis
trans
bond stretching
angle bending
torsional
0.29
1.06
0.83
0.02
2.51
0.32
2.18
0.71
0.01
2.74
0.28
1.04
0.84
0.01
2.25
0.33
2.47
0.66
0.00
2.45
0.27
1.05
0.84
0.00
2.21
0.42
3.82
0.73
0.00
2.21
0.27
1.06
0.83
0.00
2.18
0.45
4.24
0.74
0.00
2.18
0.27
1.06
0.82
0.00
2.07
0.46
4.41
0.77
0.00
2.22
0.33
1.28
1.64
0.01
2.42
0.67
4.69
1.66
0.01
2.37
0.38
1.44
1.48
0.02
2.92
0.75
4.93
1.54
0.01
2.91
1.88
out of plane
1-4 van der Waals
van der Waals
1-4 electrostatic
electrostatic
total
-0.48
0.33 -0.56
0.30 -0.71
0.82 -0.76
0.97 -0.83
0.90 -0.52
1.55 -0.44
-3.38 -3.45 -1.07 -1.60 -2.23 -2.63 -3.23 -3.43 -3.71 -3.79 -1.65 -2.09 -3.27 -3.60
-0.62 -0.37 -1.01 -0.40 -0.88 -0.41 -0.73 -0.41 -0.64 -0.38 -1.22 -0.72 -0.43 -0.06
0.22
2.47
1.78
4.22
0.55
4.96 -0.39
4.75 -0.96
4.58
2.29
8.13
2.09
8.35
trans population
estimated at 25 °C
2 × 10-2
2 × 10-2
6 × 10-4
2 × 10-4
9 × 10-5
6 × 10-5
3 × 10-5
Geometry (Angles, deg]
N(3)-C(4)-N(4)
C(5)-C(4)-N(4)
asymmetry
124.1 120.7 123.9 120.4 123.9 119.2 124.0 118.9 124.0 118.9 123.7 118.4 123.7 118.2
118.4 123.9 118.7 124.3 118.7 126.1 118.7 126.6 118.6 126.5 118.3 126.4 118.2 126.6
-5.7
3.2
-5.3
3.9
-5.2
6.9
-5.3
7.7
-5.4
7.6
-5.4
8.0
-5.5
8.3
Ch a r t 1. Rotamers of N4-Hydroxy-2′-deoxycytidines
5a ,b,d -h
whereas for the trans conformers the asymmetry value
was correlated with the C(5)-substituent’s dimensions.
In the case of 1-methyl-N4-hydroxycytosine and its C(5)-
fluoro derivative, asymmetry did not exceed 4°. It
increased with the increase of the radius of C(5)-
substituent up to 8.3° in the case of the hydroxymethyl
group. The increase of the asymmetry was not depend-
ent on other, including electrostatic, properties of the
C(5)-substituent. An almost linear relation between
asymmetry in the trans form of the N4-OH group and
the cis conformer stabilization energy proved that the
C(5)-substituent-induced C(4)-N(4) bond deformation
played the main role in destabilization of the trans
conformation. This led to the sterically driven mecha-
nism of the stabilization of the cis conformer of the C(5)-
substituted N4-hydroxy-2′-deoxycytidines. Finally, it
may be postulated that apart from 5a ,e the population
of the trans form could be neglected. In fact, NMR
spectra of 5a ,e proved that both compounds exhibited
small populations of the corresponding trans conformers
(data not presented).
favored with the oxygen atom lying in the plane of the
base. In the case of N4-hydroxy-2′-deoxycytidine (5a ),
the conformer with the OH group pointed toward the
N(3) proton of the base (cis) was more stable than the
other one, with the hydroxyl pointed toward C(5) (trans)
(Chart 1). This effect was even more pronounced in the
case of 5-fluoro-N4-hydroxy-2′-deoxycytidine (5e).
Deoxyr ibose Con for m a tion . The conformational
parameters of the foregoing nucleosides in neutral
1
With all compounds investigated the N4-OH cis-trans
equilibrium was analyzed by molecular mechanics. The
differences of this equilibrium were driven mainly by
steric hindrances between the N4-OH group and the
respective C(5)-substituent. The analysis of the energy
terms proved that the total stabilization effect mainly
resulted from the direct van der Waals interactions as
well as from the deformation of the N(3)-C(4)-N(4) and
C(5)-C(4)-N(4) angles (cf. van der Waals and angle
bending terms in Table 1). In the trans conformer the
increase of the value of the C(5)-C(4)-N(4) angle
accompanied by the decrease of the N(3)-C(4)-N(4)
angle resulted in more pronounced separation of the N4-
OH oxygen from the C(5)-substituent. Enlargement of
the C(5)-substituent dimension increased the deforma-
tion of the trans conformation, stabilizing the cis one.
The deformation was defined by the asymmetry and
measured as the difference between the N(3)-C(4)-N(4)
and C(5)-C(4)-N(4) angles. For the cis conformers the
asymmetry value was constant (-5.4 ( 0.2°) and
appeared not to be influenced by the C(5)-substituent,
aqueous medium (D2O), derived from the vicinal H-
1H coupling constants (see Experimental Section) with
the aid of the modified Karplus relationship,31 are listed
in Table 2. No significant conformational differences
influencing interaction with the enzyme could be found
for the sugar ring. For each compound its puckering as
well as the population of C(5′)-exocyclic group rotamers
are almost identical. The S conformer is populated at
the level of 0.62 ( 0.02, mean amplitude of pseudo-
rotation, and the population equals 32.8 ( 1.4%, whereas
the population of the g+ rotamer of the C(4′)-C(5′) bond
dominates (0.51 ( 0.02).
Biologica l Resu lts
Cell Gr ow th In h ibition . It was studied with two
L5178Y murine leukemia cell lines, parental (L5178Y-
P) and 5-fluoro-2′-deoxyuridine resistant (L5178Y-R),
the latter 70-fold less sensitive toward FdUrd than the
former. Although the resistance mechanism has not
been yet fully recognized, preliminary results (not
shown), concerning ATP-dependent phosphorylation of

4650 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24
Felczak et al.
Ta ble 2. Solution Conformation of 5-Substituted N4-Hydroxy-2′-deoxycytidines in D2O Calculated from 500-MHz 1H NMR
conformation
compd
S
g+
t
g-
Ng+
Nt
Sg+
St
Sg-
F12a
F34a
5a
5b
5d
5e
5f
0.63
0.61
0.64
0.63
0.59
0.61
0.60
0.47
0.49
0.50
0.53
0.52
0.51
0.53
0.38
0.36
0.37
0.35
0.29
0.34
0.34
0.15
0.15
0.13
0.12
0.19
0.15
0.13
0.23
0.25
0.22
0.24
0.29
0.26
0.26
0.14
0.14
0.13
0.13
0.12
0.13
0.13
0.24
0.24
0.27
0.29
0.22
0.25
0.27
0.24
0.22
0.24
0.22
0.17
0.21
0.21
0.15
0.15
0.13
0.12
0.19
0.15
0.13
1.62
1.75
1.67
1.86
2.41
1.93
1.94
1.00
1.08
1.14
1.32
1.30
1.20
1.31
5g
5h
a
Lit. ref 30.
N4-hydroxy-5-iodo-2′-deoxycytidine (5h ) and 5-iodo-2′-
deoxyuridine (IdUrd) showed the opposite relation. N4-
Hydroxy-5-hydroxymethyl-2′-deoxycytidine (5d) and 5-hy-
droxymethyl-2′-deoxyuridine (HmdUrd) were similarly
potent cell growth inhibitors with each cell line (Table
3).
Ta ble 3. Inhibition of Growth of L5178Y Parental and
FdUrd-Resistant Cells by 5-Substituted 2′-Deoxy Nucleosides
and Their N4-Hydroxy Congeners
a
IC50 (µM)
drug
growth assay
[
14C]Leu incorp
[3H]Thd incorp
L5178Y-P
0.0024
FdUrdb
CldUrd
BrdUrd
IdUrd
HmdUrd
5a
5e
5f
5g
5h
0.002
0.002
Th ym id yla te Syn th a se-Ca ta lyzed Deh a logen a -
t ion of 5-Br om o-2′-d eoxyu r id in e 5′-Mon op h os-
p h a te (Br d UMP ) a n d 5-Iod o-2′-d eoxyu r id in e 5′-
Mon op h osp h a te (Id UMP ). With the analogues 5-
substituted with chlorine, bromine, or iodine a possibility,
their dehalogenation in the thymidylate synthase reac-
tion was tested32 (see Experimental Section).
32 ( 10 (3)
26 ( 2% (2)
3.8 ( 4% (2)
6.2 ( 11% (2)
31 ( 21% (2)
16 ( 1.4 (3)
20 (1)
4.6 ( 39% (2)
33 (1)
26 ( 10% (2)
5.4 ( 28% (2)
8.6 ( 6% (2)
0.018 ( 0.002 (3) 0.018 ( 3% (2) 0.026 ( 4% (2)
1.15 ( 4% (2)
0.98 ( 26% (2)
32 ( 3% (2)
4.0 ( 5% (2)
56 ( 39% (2)
1.48 ( 6% (2)
1.24 ( 8% (2)
7.0 ( 25% (2)
With both BrdUMP and IdUMP the absorption maxi-
mum changed with time (from 277 to 271 nm for
BrdUMP and from 278 to 271 nm for IdUMP). No such
changes were visible with CldUMP, N4-OH-BrdCMP
(6g), N4-OH-IdCMP (6h ), or when the enzyme was
omitted. Therefore BrdUMP and IdUMP, but not Cl-
dUMP or either 5-halogeno-N4-hydroxy-2′-deoxycytidine
5′-monophosphate analogue, become dehalogenated by
recombinant rat hepatoma thymidylate synthase.
5d
5b
L5178Y-R
0.15 ( 4% (2)
FdUrd
CldUrd
BrdUrd
IdUrd
HmdUrd
5a
5e
5f
5g
5h
0.14 ( 0.01 (3)
12 ( 11% (2)
12 ( 19% (2)
2.40 ( 12% (2)
11 ( 39% (2)
193 ( 19% (2)
0.13 ( 5% (2)
0.052 ( 0.005 (3) 0.053 ( 8% (2)
1.66 ( 2% (2)
2.5 ( 61% (2)
34 ( 23% (2)
9.1 ( 42% (2)
163 ( 29% (2)
3.01 ( 18% (2) 1.03 ( 8% (2)
Spectrophotometric monitoring at 338 nm of the
dehalogenation reaction mixture, containing either 0.04
mM BrdUMP or 0.04 mM IdUMP, with 0.3 mM meth-
ylenetetrahydrofolate demonstrated a slow extinction
increase, pointing to dihydrofolate production. The
reaction velocity was 20.5 and 34.2 nmol/min‚mg protein
for BrdUMP and IdUMP, respectively.
5d
5b
a
IC50 is the drug concentration required for 50% reduction in
cell number, [14C]Leu incorporation, or [3H]Thd incorporation.b Lit.
ref 38. c Mean ( SEM or mean ( % difference between the mean
and each of the two results, followed by the number of experiments
(N) in parentheses.
Th ym id yla te Syn th a se In h ibition by Nu cleo-
tid es. Thymidylate synthase inhibition was studied
with two forms of the enzyme differing in sensitivities
toward FdUMP inhibition, isolated from parental and
FdUrd-resistant L1210 cell lines.12 All N4-OH-dCMP
and dUMP analogues studied were competitive vs
dUMP inhibitors of the enzyme (Table 4). The 5-sub-
stituted N4-OH-dCMP analogues and HmdUMP, similar
dThd and FdUrd by cell crude extracts, point to altered
substrate specificity of this reaction, reflecting presum-
ably thymidine kinase activity, in the resistant cells
(with FdUrd being a better substrate than dThd in
L5178Y-P but worse than dThd in L5178Y-R cells).
None of the nucleoside analogues studied was as strong
a growth inhibitor as FdUrd for L5178Y-P cells although
5-fluoro-N4-hydroxy-2′-deoxycytidine (5e) was only about
10-fold weaker (Table 3). Interestingly, with L5178Y-R
cells 5e caused almost 3-fold stronger inhibition than
FdUrd. At the same time 5e was some 3-fold weaker
growth inhibitor of the resistant cells than of the
parental cells. All other analogues (5d -h ), except for
FdUrd and to a smaller extent N4-hydroxy-2′-deoxycy-
tidine (5a ) and N4-hydroxy-5-methyl-2′-deoxycytidine
(5b), were also rather similarly potent inhibitors of both
cell lines. While each 5-chloro-N4-hydroxy-2′-deoxycy-
tidine (5f) and 5-bromo-N4-hydroxy-2′-deoxycytidine (5g)
was over 10-fold stronger inhibitor of L5178Y (both
parental and FdUrd-resistant) cell growth than the
corresponding 5-halogeno-substituted dUrd congener,
to N4-OH-dCMP (6a ) and FdUMP, were also N5,N10
-
methylenetetrahydrofolate-dependent, hence mecha-
nism-based, slow-binding inhibitors (Table 5). CldUMP,
BrdUMP, and IdUMP, similar to dTMP, did not cause
time-dependent inactivation of the enzyme. Instead,
they behaved as classic inhibitors of tritium release from
[5-3H]dUMP, in agreement with substrate activity of
BrdUMP and IdUMP in the dehalogenation reaction
(see above) and CldUMP being a classic inhibitor.
While previously studied FdUMP and N4-OH-dCMP
(6a ) showed higher affinity (Table 4) and greater
inactivation power (Table 3) with the parental cell than
a FdUrd-resistant cell enzyme, a definitely opposite
relationship could be seen with HmdUMP (Tables 4 and
5).

Deoxycytidines and Their 5′-Monophosphates
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24 4651
Ta ble 4. Parameters Describing Time-Independent
Inhibition of Thymidylate Synthase from L1210 Parental
and FdUrd-Resistant Cells by 5-Substituted
N4-Hydroxy-2′-deoxycytidine 5′-Monophosphates and
dUMP Analogues
monophosphate, in an ATP-dependent reaction inhib-
ited by thymidine and deoxyuridine, but not deoxycy-
tidine, and as the monophosphate to inhibit thymidylate
synthase.33 That is why it was interesting to compare
the effect of N4-OH-dCyd (5a ) and its analogues on two
cell lines differing in sensitivity to another thymidine
and deoxyuridine antimetabolite, FdUrd, also phospho-
rylated to the monophosphate and targeted at the same
enzyme. 2-4 Interestingly, while both N4-OH-dCyd (5a )
and its 5-fluoro congener N4-OH-FdCyd (5e) were sev-
eralfold weaker inhibitors of L5178Y-R (FdUrd-resis-
tant) than L5178Y-P (parental) cells, combination of the
C(4)dN-OH with C(5)-F substituents in N4-OH-FdCyd
(5e) resulted in stronger inhibition of the resistant line
than that caused by FdUrd (although the latter inhib-
ited the parental line 9-fold stronger than 5e). One
possible suggestion of an explanation is apparently
altered substrate specificity of nucleoside phosphoryla-
tion in the resistant cells (see above), but further studies
of the resistance mechanism are needed to clarify this
point.
L1210-P enzyme
L1210-R enzyme
compd
Ki (µM)
Ki/Km
Ki (µM)
Ki/Km
6a a
1.7
0.02
0.8
0.3
8.5
0.68
0.008
0.32
0.12
3.4
9.3
0.03
1.4
0.15
8.4
5.5
0.018
0.82
0.088
4.9
FdUMPa
6ea
CldUMP
6f
BrdUMP
6g
IdUMP
6h
HmdUMP
6d
dTMP
1.5
0.6
13
3.1
3
1.5
0.76
2.9
0.46
36.6
6.99
5.7
0.9
2.18
0.27
22
4.1
3.4
0.53
1.3
32.1
7.68
7.46
3.7
1.91
7.3
a
Lit. ref 12.
Discu ssion
Cell Gr ow th In h ibition . The parent N4-OH-dCyd
(5a ) has been previously demonstrated as a L5178Y and
Burkitt’s lymphoma cell growth inhibitor with IC50
values of 12 µM33 and 1.7-4.0 µM,34 respectively. In our
hands it was some 3-fold weaker inhibitor of L5178Y-P
cell growth (Table 3). However, the 5-fluoro substituent,
in N4-OH-FdCyd (5e), potentiated inhibition of cell
growth almost 2000-fold (Table 3), whereas with Bur-
kitt’s lymphoma cells the corresponding effect amounted
to 60-fold, IC50 value for 5e being 0.027 µM.34 Potentia-
tion of L5178Y cell growth inhibition by a 5-substituent
at N4-OH-dCyd (5a ) was apparent also with N4-OH-
CldCyd (5f), N4-OH-BrdCyd (5g), and N4-OH-HmdCyd
(5d ) but not with N4-OH-IdCyd (5h ) or N4-OH-MedCyd
(5b) (Table 3). Combinations of the C(4)dN-OH with
either C(5)-Cl in 5f or C(5)-Br in 5g resulted clearly in
a synergistic effect on inhibition of both cell lines,
reflected by lower IC50 values for each combination than
for each of the corresponding singly substituted conge-
ners (Table 3).
In ter a ction w ith En zym e. Dehalogenation of Br-
dUMP and IdUMP by bacterial thymidylate synthase
from Lactobacillus casei has been previously described.32
Lack of dehalogenation of the corresponding derivatives
of N4-OH-dCMP (6a ) is similar to the earlier demon-
strated lack of C(5)-proton release from 6a in thymidy-
late synthase-catalyzed reaction.12 The latter result has
been interpreted in terms of the C(4)dN-OH, probably
interacting with N5,N10-methylenetetrahydrofolate, be-
ing the cause of the reaction arrest at a step preceding
release of H+ from C(5). Such an explanation has been
further supported by molecular modeling studies.2,35
Considering lack of dependence of the dehalogenation
reactions (with BrdUMP and IdUMP) on N5,N10-meth-
ylenetetrahydrofolate, failure of the enzyme to release
Br- or I- from 6g or 6h in the absence of the cofactor
suggests an influence of the C(4)dN-OH on the C(5)-
Br and C(5)-I bonds, rendering the corresponding
halogen anion release impossible. Hence, the dN-OH
substituent at the pyrimidine C(4) appears to provide
both BrdUMP and IdUMP with potency to inactivate
the enzyme. This also concerns CldUMP, although it
shows no thymidylate synthase-catalyzed dehalogena-
tion in agreement with previous results,32 which ap-
N4-OH-dCyd (5a )33,34 and its 5-fluoro congener N4-
OH-FdCyd (5e)34 have been shown to be thymidine and
deoxyuridine, rather than deoxycytidine, antimetabo-
lites. An elaborated study demonstrated N4-OH-dCyd
(5a ) to be phosphorylated in L5178Y cells to the 5′-
Ta ble 5. Parametersb for Time-Dependent Inactivation, Reflecting Slow-Binding Inhibition, by FdUMP and 5-Substituted
N4-Hydroxy-2′-deoxycytidine 5′-Monophosphates of Thymidylate Synthase from L1210 Parental and FdUrd-Resistant Cells
compd
enzyme source
Ki′ (µM)
0.0018
0.0122
0.063
0.184
0.073
0.093
Ki′′ (µM)
0.02
0.014
0.226
1.46
k2′ (min-1
)
k2′′ (min-1
)
FdUMPa
L1210-P
L1210-R
L1210-P
L1210-R
L1210-P
L1210-R
L1210-P
L1210-R
L1210-P
L1210-R
L1210-P
L1210-R
L1210-P
L1210-R
L1210-P
L1210-R
0.17
0.12
FdUMPa
0.25
0.17
0.20
0.24
0.24
0.17 ( 0.01 (3)
0.19 ( 0.03 (3)
0.23 ( 0.05 (3)
0.22 ( 0.03 (5)
0.21 ( 0.01 (3)
0.30 ( 0.03 (3)
0.74 ( 0.09 (3)
0.63 ( 0.01 (3)
0.45 ( 0.09 (3)
0.72 ( 0.17 (3)
0.06
0.02
0.09
0.07
0.06
0.06 ( 0.00 (3)
0.06 ( 0.02 (3)
0.07 ( 0.01 (3)
0.05 ( 0.01 (5)
0.02 ( 0.00 (3)
0.11 ( 0.03 (3)
0.02 ( 0.00
6a a
6a a
6ea
0.056
0.071
6ea
6f
6f
6g
6g
6h
6h
20 ( 2 (3)
16 ( 2 (3)
31 ( 9 (3)
33 ( 8 (3)
15 ( 6 (3)
26 ( 3 (3)
36 ( 5 (3)
11 ( 1 (3)
4.9 ( 1.1 (3)
4.2 ( 1.9 (3)
57 ( 18 (3)
17 ( 1 (3)
14 ( 2 (3)
8.6 ( 1.2 (3)
16 ( 2 (3)
HmdUMP
HmdUMP
6d
9.9 ( 3.0 (3)
4.7 ( 1.3 (3)
1.4 ( 0.3 (3)
0.02 ( 0.01
0.23 ( 0.06 (3)
0.15 ( 0.02 (3)
3.8 ( 1.1 (3)
5.7 ( 1.1 (3)
6d
a
Lit. ref 12. b Mean ( SEM or mean ( % difference between the mean and each of the two results, followed by the number of experiments
(N) in parentheses.

4652 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24
Felczak et al.
employed for pH measurements. Liquid matrix secondary ion
mass spectra (LSIMS) were recorded for nucleosides with an
AMD-604 spectrometer. High-resolution H NMR spectra were
pears to be only a classic inhibitor of the enzyme.
However, with the dN-OH substituent at the pyrimi-
dine C(4) it becomes a slow-binding inhibitor.
1
recorded on a Varian 500 MHz in D2O with DSS as internal
standard or in CDCl3 with tetramethylsilane as internal
standard. Thin-layer chromatography (TLC) was run on Merck
silica gel F254 glass plates (DC, 20 × 20 cm, 0.25 mm; no.
1.05715) and Merck cellulose F glass plates (DC, 20 × 20 cm,
0.1 mm; no. 1.05718). Preparative-layer chromatography (PLC)
was run on Merck silica gel F254 glass plates (PLC, 20 × 20
cm, 2 mm; no. 1.05717). The following solvents (v/v) were
used: (A) CHCl3-MeOH, 80:20; (B) CHCl3-MeOH, 95:5; (C)
i-PrOH-EtOAc-toluene, 10:20:70; (D) MeOH-concentrated
aqueous NH3-H2O, 70:10:20; (E) EtOH-1 M CH3COONH4,
70:50; (F) saturated aqueous (NH4)2SO4-0.05 M phosphate
buffer pH 6-i-PrOH, 79:19:2; DEAE-Sephadex A-25 was
purchased from Pharmacia. Relation of phosphorus/extinction
(P/ꢀ) was determined using micromethod of Chen, Toribara and
Warner.37
Cell Lin es. Mouse leukemia L5178Y cells were grown as
reported earlier.38 The FdUrd-resistant cell line was developed
by growing in the presence of the drug in the cell medium.
FdUrd concentration was increased (stepwise and always after
cells became adapted to the previous concentration) up to 0.1
µM.
In Vitr o Cell Gr ow th In h ibition . The influence of each
analogue on viability of exponentially growing cells and [14C]-
leucine and [3H]thymidine incorporation was followed, and IC50
values were determined as previously described.38
Th ym id yla te Syn th a se. Highly purified preparations of
thymidylate synthase, differing in sensitivity to FdUMP slow-
binding inhibition, from parental and FdUrd-resistant L1210
cells,12,39 as well as the pure recombinant rat hepatoma
enzyme40 have been described in more detail elsewhere. The
[5-3H]dUMP tritium release activity assay was performed as
previously described.12 The dUMP and dCMP analogues were
added to the reaction mixtures as neutral aqueous solutions.
Reaction mixtures and procedures used in inhibition studies
were as earlier described.12,41
Th ym id yla te Syn th a se-Ca ta lyzed Deh a logen a tion of
d UMP An a logu es. Highly purified recombinant rat hepatoma
thymidylate synthase (specific activity 0.6 µmol/min/mg pro-
tein at 30 °C) was used.12,40 The reaction mixture contained
50 mM N-methylmorpholine-HCl pH 7.4, 6.5 mM DTT, 25 mM
MgCl2, 1 mM EDTA, one of the nucleotide analogues (0.042
mM 5-chloro-2′-deoxyuridine 5′-monophosphate (CldUMP),
0.04 mM BrdUMP, 0.038 mM IdUMP, 0.043 mM 6g or 0.043
mM 6h ) and the enzyme (0.004 mM) in a total volume of 0.42
mL. Samples were incubated at 30 °C and UV spectra were
monitored in the course of 120 min using Beckman DU-64
spectrophotometer.
In view of the results of the preliminary QSAR study
of the 5-substituted dUMP analogues,36 a large steric
effect causes weaker interaction with thymidylate syn-
thase by IdUMP, as compared with other 5-halogenated
congeners. However, comparison of enzyme inhibition
by the 5-halogenated analogues of 6a shows an interest-
ing departure from this relationship, reflected by a
stronger inhibition, particularly of the slow-binding
type, caused by N4-OH-IdCMP (6h ) than N4-OH-Brd-
CMP (6g) or N4-OH-CldCMP (6f) (Tables 4 and 5). The
latter suggests an interplay between the substituents
at C(4) and C(5) in 6h interacting with thymidylate
synthase. A similar interplay between the C(4)dN-OH
and C(5)-F substituents in N4-OH-FdCMP (6e) was
indicated by our previous results.12 To explain it, an
intramolecular hydrogen bond dN4-OH‚‚‚F-C(5) was
hypothesized, influencing an assumed cis-trans equi-
librium of rotamers, resulting in stabilization of the rare
species trans, found to be the only inhibitory form.12
However our present results, based on molecular me-
chanics calculations (Table 1), bring into question the
dN4-OH‚‚‚F-C(5) hydrogen bond formation as a main
factor of stabilization of the trans species (cf. ref 28).
The compounds exhibiting lowered population of the
rare trans conformer, caused by the unfavorable O4-
C(5)-substituent steric interaction (such as in 6b,e-h ),
show significantly lower inhibitory activity against
thymidylate synthase than either 6a or 6e.
Con clu sion s
The most convenient procedures for the synthesis of
5-substituted N4-hydroxy-2′-deoxycytidines 5a ,b,d -h
involve site-specific triazolation or N-methylimidazola-
tion of the respective 5-substituted 3′,5′-di-O-acetyl-2′-
deoxyuridines followed by hydroxylamination and de-
blocking with MeOH-NH3.
Steric interaction between the 5-substituent and the
N(4)-oxygen strongly destabilizes the trans conformation
of the N4-OH group when the van der Waals radius of
the substituent exceeds 1D.
The dN-OH substituent at the pyrimidine C(4)
provides 6a and its analogues with potency to inactivate
thymidylate synthase. Inactivation depends on the
population of the rare trans conformer (with the N4-OH
pointed toward the C(5)); hence compounds exhibiting
lowered population of the trans conformer, caused by
an unfavorable O4-C(5)-substituent steric interaction,
are weaker slow-binding inhibitors of the enzyme than
6a or its 5-fluoro congener.
Kin etic Stu d ies. To identify the type of inhibition involved,
the effects of the 6a analogues on the dependence of reaction
rate on dUMP concentration, in the form of Liveweaver-Burk
plots, were analyzed as previously reported with the use of a
program, based on nonlinear regression and designed for
estimation of kinetic constants describing competitive (both
linear and parabolic), noncompetitive, uncompetitive, or mixed-
type linear inhibition.41
Quantitative analyses of thymidylate synthase inhibition
by 5-substituted N4-OH-dCMP analogues or HmdUMP leading
to time-dependent inactivation of the enzyme were performed
by following the decrease of enzyme activity with time (usually
at 0.5, 1.0, 1.5, 4, 6, 8 and 10 min) during preincubation of
the enzyme at 37 °C in the presence of 0.8 mM (6RS,aS)-
CH2H4PteGlu, 3.3 µM dUMP (to prevent thermal inactivation),
and various concentrations of inhibitor. Activity remaining
after preincubation was determined by addition of 25 µM
[5-3H]dUMP (7 × 104 dpm/nmol) and tritium release after 4
min incubation was measured. The slopes of the semilog plots
of percent (%) remaining activity vs preincubation time,
expressing apparent inactivation rate constants (kapp) and
corresponding inhibitor concentrations [I], were then replotted
as double-reciprocal plots, according to the relationship:41
The dN-OH substituent of the pyrimidine nucleotide
C(4) prevents the enzyme-catalyzed release of Br- and
I- from the C(5) (the same shown previously for H+).
5e shows a strong cell growth inhibitory activity against
FdUrd-resistant tumor cells.
Exp er im en ta l Section
Gen er a l Met h od s. Melting points (uncorrected) were
measured on a Boetius microscopic hot stage; UV spectra were
recorded on a Cary 300 instrument, using 10-mm path length
cuvettes. Extremes of pH made use of standard solutions of
HCl and NaOH. A phosphate buffer was used (pH 7.00). A
Cole-Parmer instrument with combination electrode was

Deoxycytidines and Their 5′-Monophosphates
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24 4653
184 °C; UV λmax (pH 0) 280 nm (ꢀ 10.4 × 103), 219 nm (ꢀ 6.9 ×
103); λmax (pH 1) 279.5 nm (ꢀ 10.2 × 103), 220 nm (ꢀ 7.0 × 103);
λmax (pH 2) 278.5 nm (ꢀ 8.7 × 103), 224 nm (ꢀ 6.9 × 103); λmax
(pH 7) 270 nm (ꢀ 5.2 × 103), 234 nm (ꢀ 10.0 × 103); λmax (pH
12) 240 nm (ꢀ 8.2 × 103); TLC (silica gel) Rf (C) 0.21; 1H NMR
δ (CDCl3) 6.80 (1H, d, H5), 6.30 (1H, dd, H1′), 5.65 (1H, d H6),
5.19-5.18 (1H, m, H3′), 4.34-4.28 (2H, m, H5′, H5′′), 4.20 (1H,
m, H4′), 2.38-2.35 (1H, m, H2′′), 2.17-2.14 (1H, m, H2′), 2.11,
2.10 (6H, 2 × s, 2 × CH3CO). Anal. (C13H17N3O7‚0.5MeOH) C,
H, N.
Ki[S] Ki
1
kapp
1
[I]
1
k2
)
+
+
(
)
Kmk2 k2
where k2 is the inactivation rate constant. The values of k2
and Ki were determined from the plot intercept and slope,
respectively.38,42
Sta tistica lly Eva lu a ted Resu lts. These are presented as
means ( SEM or means ( percent difference between the
mean and each of the two results, followed by the number of
experiments (N) in parentheses.
1-(3,5-Di-O-acetyl-â-D-2-deoxyr ibofu r an osyl)-4-h ydr oxy-
a m in o-5-m eth ylp yr im id in -2(1H)-on e (4b). 200 mg (0.53
mmol) of 1-(3,5-di-O-acetyl-â-D-2-deoxyribofuranosyl)-4-(1,2,4-
triazol-1-yl)-5-methylpyrimidin-2(1H)-one (2b) was dissolved
in 10 mL of pyridine and hydroxylamine hydrochloride (150
mg, 2.16 mmol) was added. The mixture was stirred at room
temperature for 12 h, evaporated under reduced pressure and
coevaporated successively with 2 × 20 mL of toluene and
EtOH. 50 mL of CHCl3 was added and the mixture was washed
with water (2 × 30 mL). The organic layer was dried (Na2-
SO4), evaporated under reduced pressure and crystallized from
MeOH to yield white crystals: 145 mg (80%); mp 140-143
°C; UV λmax (pH 0) 285 nm (ꢀ 13.60 × 103), λmax 220 nm
(ꢀ 9.60 × 103) λmax (pH 1) 282.5 nm (ꢀ 12.40 × 103), λmax 220
nm (ꢀ 9.00 × 103), λmax (pH 2) 278 nm (ꢀ 9.90 × 103), λmax 224
nm (ꢀ 9.40 × 103), λmax (pH 7) 268 nm (ꢀ 8.60 × 103), λmax 236
nm (ꢀ 11.55 × 103) λmax (pH 12) 252 nm (ꢀ 10.80 × 103); TLC
Gen er a l P r oced u r e for th e Syn th esis of 1,2,4-Tr ia zole
5-Su bstitu ted 3′,5′-Di-O-acetyl-2′-deoxyu r idin es 2a-c,e,f,h .
1,2,4-Triazole (625 mg, 9.05 mmol) was suspended in MeCN
(5 mL) at 0 °C, phosphoryl chloride (180 µL, 1.93 mmol) was
added over 2 min and the mixture was stirred at 0 °C for 10
min. After addition of triethylamine (1.2 mL, 8.61 mmol) over
5 min the mixture was stirred at 0 °C for a further 20 min.
3′,5′-Di-O-acetyl derivative of the appropriate 2′-deoxy nucleo-
side (1a -c,e,f,h )21-24 (1 mmol) in MeCN (5 mL) was added
and the solution stirred at room temperature overnight.
Triethylamine (820 µL, 5.93 mmol) and water (215 µL, 11.94
mmol) were then added and after 10 min the solvents were
evaporated under reduced pressure. The residue was parti-
tioned between CHCl3 (35 mL) and saturated aqueous NaH-
CO3. The organic layer was dried (Na2SO4) and evaporated.
The crude products were purified as follows.
1-(3,5-Di-O-a cet yl-â-D-2-d eoxyr ib ofu r a n osyl)-4-(1,2,4-
tr ia zol-1-yl)p yr im id in -2(1H)-on e (2a ). Crude product was
crystallized from MeOH to yield white crystals: 330 mg (91%);
mp 150-153 °C; UV λmax (MeOH) 312 nm (ꢀ 8.1 × 103), 248.5
nm (ꢀ 13.5 × 103), 214.5 (ꢀ 17.35 × 103); TLC (silica gel) Rf (C)
0.18; 1H NMR δ (CDCl3) 9.27, 8.13 (2H, 2 × s, 2 × triazolo
CH), 8.28 (1H, d, H5), 7.10 (1H, d, H6), 6.27 (1H, dd, H1′),
5.25-5.23 (1H, m, H3′), 4.42-4.40 (2H, m, H5′, H5′′), 2.95 (1H,
m, H4′), 2.13, 2.09 (6H, 2 × s, 2 × CH3CO), 2.20-2.08 (2H, m,
H2′, H2′′).
1
(silica gel) Rf (C) 0.32; H NMR δ (CDCl3) 6.34 (1H, dd, H1′),
6.11 (1H, d, H6), 5.21-5.18 (1H, m, H3′), 4.36-4.30 (2H, m,
H5′, H5′′), 4.20-4.18 (1H, m, H4′), 2.35-2.30 (1H, m, H2′′),
2.19-2.13 (1H, m, H2′), 2.14, 2.11 (6H, 2 × s, 2 × CH3CO),
1.83 (3H, s, CH3. Anal. (C14H19N3O7‚1.5H2O) C, H, N.
1-(3,5-Di-O-acetyl-â-D-2-deoxyr ibofu r an osyl)-4-h ydr oxy-
a m in o-5-a cetoxym eth ylp yr im id in -2(1H)-on e (4c). 5-Hy-
droxymethyl-2′-deoxyuridine44 258 mg (1 mmol), 6 mL of Ac2O
and 7.5 mg of DMAP were stirred at room temperature for 24
h. At this point TLC (solvent C) indicated the completion of
the reaction. The reaction mixture was evaporated to dryness
in vacuo and evaporated twice with EtOH. Chromatographi-
cally homogeneous residue (385 mg) was directly used in
triazolation reaction, to give crude 2c which was used in the
preparation of 4c. 200 mg (0.46 mmol) of 1-(3,5-di-O-acetyl-
â-D-2-deoxyribofuranosyl)-4-(1,2,4-triazol-1-yl)-5-acetoxymeth-
ylpyrimidin-2(1H)-one (2c) was dissolved in 20 mL of pyridine
and hydroxylamine hydrochloride (150 mg, 2.16 mmol) was
added. The mixture was stirred at room temperature for 12
h, evaporated under reduced pressure and coevaporated suc-
cessively with 2 × 20 mL of toluene and EtOH. 50 mL of CHCl3
was added and the mixture was washed with water (2 × 30
mL). The organic layer was dried (Na2SO4) and evaporated
under reduced pressure. The residue was crystallized from a
mixture of toluene and MeOH to yield white crystals: 167 mg
(91%); mp 122-125 °C; UV λmax (pH 0) 281.5 nm (ꢀ 14.20 ×
103), λmax 219 (ꢀ 12.40 × 103); λmax (pH 1) 281.5 nm (ꢀ 9.60 ×
103), λmax 224.5 nm (ꢀ 8.80 × 103); λmax (pH 2) 275 nm
(ꢀ 7.60 × 103), λmax 230 nm (ꢀ 12.10 × 103); λmax (pH 7) 274 nm
(ꢀ 6.90 × 103), λmax 231.5 nm (ꢀ 12.75 × 103); λmax (pH 12) 251
nm (ꢀ 10.60 × 103); TLC (silica gel) Rf (C) 0.33; 1H NMR δ
(CDCl3) 7.03 (1H, s, H6), 6.30 (1H, dd, H1′), 5.21 (1H, m, H3′),
4.82-4.70 (2H, dd, CH2OH), 4.36 (1H, dd, H5′′), 4.30 (1H, dd,
H5′), 4.21 (1H, m, H4′), 2.39-2.34 (1H, m, H2′′), 2.20-2.05
(1H, m, H2′), 2.17, 2.11, 2.07 (9H, 3 × s, 3 × CH3CO). Anal.
(C16H21N3O9‚0.33MeOH) C, H, N.
1-(3,5-Di-O-a cet yl-â-D-2-d eoxyr ib ofu r a n osyl)-4-(1,2,4-
tr ia zol-1-yl)-5-m eth ylp yr im id in -2(1H)-on e (2b). The prod-
uct was crystallized from MeOH to yield white crystals: 351
mg (93%); mp 126-131 °C (lit.43 120-122 °C); UV λmax (MeOH)
326.5 nm (ꢀ 7.8 × 103), 249.5 nm (ꢀ 10.3 × 103), 215.5 nm (ꢀ
18.2 × 103); TLC (silica gel) Rf (C) 0.18.
1-(3,5-Di-O-a cet yl-â-D-2-d eoxyr ib ofu r a n osyl)-4-(1,2,4-
tr ia zol-1-yl)-5-flu or op yr im id in -2(1H)-on e (2e). Crude com-
pound 2e, homogeneous on TLC (silica gel) Rf (B) 0.75, was
used in the next step without further purification.
1-(3,5-Di-O-a cet yl-â-D-2-d eoxyr ib ofu r a n osyl)-4-(1,2,4-
tr iazol-1-yl)-5-ch lor opyr im idin -2(1H)-on e (2f). Crude prod-
uct was chromatographed on a silica gel column (1 × 30 cm)
with CHCl3 as eluent to give a foam. Crystallization from
MeOH yielded white crystals: 290 mg (73%); mp 140-145 °C;
UV λmax (MeOH) 332.5 nm, 249 nm; TLC (silica gel) Rf (C) 0.25;
1H NMR δ (CDCl3) 9.20, 8.20 (2H, 2 × s, 2 × triazolo CH),
8.45 (1H, d, H6), 6.26 (1H, dd, H1′), 5.26-5.22 (1H, m, H3′),
4.46-4.42 (3H, m, H4′, H5′, H5′′), 2.95-2.89 (1H, m, H2′′),
2.27-2.20 (1H, m, H2′), 2.18, 2.15 (6H, 2 × s, 2 × CH3CO);
MS m/z 398.08470 [(M + H)+, calcd for C15H17O6N535Cl
398.08674].
1-(3,5-Di-O-a cet yl-â-D-2-d eoxyr ib ofu r a n osyl)-4-(1,2,4-
tr ia zol-1-yl)-5-iod op yr im id in -2(1H)-on e (2h ). Crude com-
pound 2h , homogeneous on TLC (silica gel) Rf (C) 0.30, was
used in the next step without further purification.
1-(3,5-Di-O-acetyl-â-D-2-deoxyr ibofu r an osyl)-4-h ydr oxy-
a m in op yr im id in -2(1H)-on e (4a ). 200 mg (0.55 mmol) of
1-(3,5-di-O-acetyl-â-D-2-deoxyribofuranosyl)-4-(1,2,4-triazol-1-
yl)pyrimidin-2(1H)-one (2a ) was dissolved in 20 mL of pyridine
and hydroxylamine hydrochloride (150 mg, 2.16 mmol) was
added. The mixture was stirred at room temperature for 12
h, evaporated under reduced pressure and coevaporated suc-
cessively with 2 × 20 mL of toluene and EtOH. 50 mL of CHCl3
was added and the mixture was washed with water (2 × 30
mL). The organic layer was dried (Na2SO4), evaporated under
reduced pressure and crystallized from the mixture of toluene
and MeOH to yield white crystals: 154 mg (86%); mp 177-
1-(3,5-Di-O-acetyl-â-D-2-deoxyr ibofu r an osyl)-4-h ydr oxy-
a m in o-5-flu or op yr im id in -2(1H)-on e (4e). To the solution
of 332 mg (0.87 mmol) of 2′-deoxy-3′,5′-di-O-acetyl-5-fluoro-4-
triazolouridine (2e) in 10 mL of MeOH were added 530 mg
(6.6 mmol) of NH2OH × HCl dissolved in 10 mL of MeOH and
sodium methoxide prepared by dissolving of 152 mg (6.6 mmol)
of Na in 5 mL of MeOH. The solution was stirred at room
temperature and progress of the reaction was monitored by
TLC on silica gel in solvent B. After completion of the reaction
(ca. 1 h) the mixture was concentrated under reduced pressure,
dissolved in water and extracted with CHCl3. The organic layer

4654 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24
Felczak et al.
was dried (Na2SO4) and evaporated, the residue crystallized
from EtOH to yield white crystals 220 mg. The mother liquors
were purified on a preparative silica gel plate developed in
solvent B. Products were combined to yield colorless crystals:
276 mg (92%); mp 162-163 °C; UV λmax (pH 0) 286 nm
(ꢀ 14.6 × 103), 222 nm (ꢀ 10.6 × 103); λmax (pH 2) 270 nm
(ꢀ 9.7 × 103), 234 nm (ꢀ 12.2 × 103); λmax (pH 7) 270 nm
(ꢀ 10.4 × 103), 234 nm (ꢀ 13.3 × 103); λmax (pH 12) 253 nm
(ꢀ 13.5 × 103); TLC (silica gel) Rf (C) 0.37. Anal. (C13H16FN3O7)
C, H, N.
(ꢀ 7.95 × 103), 233 nm (ꢀ 11.60 × 103), λmax (pH 7) 284.5 nm (ꢀ
8.05 × 103), 233 nm (ꢀ 11.60 × 103), λmax (pH 12) 247.5 nm (ꢀ
1
11.90 × 103); TLC (silica gel) Rf (C) 0.45; H NMR δ (CDCl3)
7.30 (1H, s, H6), 6.30 (1H, dd, H1′), 5.23-5.20 (1H, m, H3′),
4.38 (1H, dd, H5′′), 4.31 (1H, dd, H5′), 4.23 (1H, m, H4′), 2.42-
2.36 (1H, m, H2′′), 2.18-2.11 (1H, m, H2′), 2.22, 2.11 (6H, 2 ×
s, 2 × CH3CO). Anal. (C13H16IN3O7‚0.33MeOH) C, H, N.
Gen er a l P r oced u r e for th e Deblock in g a n d P u r ifica -
tion of N4-OH Nu cleosid es. 0.5 mmol of appropriate com-
pound 4a -c,e-h was dissolved in 15 mL of MeOH saturated
at 0 °C with ammonia. The mixture was stirred at room
temperature for 12 h, concentrated under reduced pressure
and purified on Dowex 50W(H+) column (0.5 × 10 cm). Product
was eluted with linear gradient of water-1 N ammonia. The
major UV absorbing fractions were collected and loaded on
Chelex100 (H+ form) column (0.5 × 3 cm). The product was
eluted with water and concentrated under reduced pressure
to give crude compounds 5a ,b,d -h which were purified as
follows.
1-(3,5-Di-O-acetyl-â-D-2-deoxyr ibofu r an osyl)-4-h ydr oxy-
a m in o-5-ch lor op yr im id in -2(1H)-on e (4f). 1.06 g (3 mmol)
of 1-(3,5-di-O-acetyl-â-D-2-deoxyribofuranosyl)-5-chlorouracil
(1f) was dissolved in 15 mL of dry MeCN and added to
N-methylphosphoimidazolide prepared from 2.4 mL (30 mmol)
of N-methylimidazole and 840 µL (9 mmol) POCl3 in 60 mL of
MeCN. The mixture was stirred for 2h at room temperature
and 9.5 mmol of hydroxylamine in 10 mL of MeOH was added.
Stirring was continued for 2 h at room temperature and
reaction mixture was evaporated to dryness in vacuo. The
residue was dissolved in 20 mL of water and extracted (3 ×
50 mL) with EtOAc. The extract was washed with 30 mL
water, dried over MgSO4, evaporated to dryness in vacuo to
give ca. 1 g of crude product, which was dissolved in MeOH
and deposited on Dowex 50W(H+) column (1.5 × 20 cm) and
eluted with a gradient of TEA in MeOH (0-1 M). The fractions
containing 4f were concentrated under vacuum and crystal-
lized from EtOH to yield white crystals: 850 mg (78%); mp
178-180 °C; UV λmax (pH 0) 294 nm (ꢀ 15.3 × 103), 226 nm
(ꢀ 11.9 × 103); λmax (pH 1) 288 nm (ꢀ 10.8 × 103), 226 nm (ꢀ
12.4 × 103); λmax (pH 2) 282 nm (ꢀ 10.2 × 103), 233 nm
(ꢀ 13.2 × 103); λmax (pH 7) 281 nm (ꢀ 10.1 × 103), 234 nm
(ꢀ 13.8 × 103); λmax (pH 12) 284 nm (ꢀ 9.4 × 103), 250 nm
(ꢀ 14.2 × 103); TLC (silica gel) Rf (C) 0.41; 1H NMR δ (CDCl3)
7.13 (1H, s, H6), 6.34 (1H, dd, H1′), 5.23-5.20 (1H, m, H3′),
4.37 (1H, dd, H5′′), 4.32 (1H, dd, H5′), 4.24-4.22 (1H, m, H4′),
2.42-2.38 (1H, m, H2′′), 2.19-2.12 (1H, m, H2′), 2.18, 2.11
(6H, 2 × s, 2 × CH3CO). Anal. (C13H16ClN3O7) C, H, N.
1-(â-D-2-Deoxyr ibofu r a n osyl)-4-h yd r oxya m in op yr im i-
d in -2(1H)-on e (5a ).33 Crude 5a was crystallized from mixture
of MeOH and EtOAc to yield white crystals: 98 mg (81%); mp
147-150 °C; UV λmax (pH 0) 280 nm (ꢀ 10.4 × 103), 219 nm
(ꢀ 6.9 × 103); λmax (pH 1) 279.5 nm (ꢀ 10.2 × 103), 220 nm (ꢀ
7.0 × 103); λmax (pH 2) 278.5 nm (ꢀ 8.7 × 103), 224 nm (ꢀ 6.9 ×
103); λmax (pH 7) 270 nm (ꢀ 5.2 × 103), 234 nm (ꢀ 10.0 × 103);
λmax (pH 12) 240 nm (ꢀ 8.2 × 103); TLC (silica gel) Rf (A) 0.36;
1H NMR δ (D2O) 7.06 (1H, d, H5), 6.27 (1H, t, H1′, J 1′2′ ) 7.58
Hz, J 1′2′′ ) 6.60 Hz), 5.74 (1H, d, H6) 4.43 (1H, m, H3′, J 3′4′
)
3.84 Hz), 3.97 (1H, m, H4′, J 4′5′ ) 3.68 Hz, J 4′5′′ ) 5.19 Hz),
3.79 (1H, dd, H5′, J 5′5′′ ) -12.42 Hz), 3.72 (1H, dd, H5′′), 2.33
(1H, m, H2′′, J 2′2′′ ) -14.25 Hz, J 2′′3′ ) 3.63 Hz), 2.27 (1H, m,
H2′, J 2′3′ ) 6.98 Hz); MS m/z 244 (M + H)+.
1-(â-D-2-Deoxyr ibofu r a n osyl)-4-h yd r oxya m in o-5-m eth -
ylp yr im id in -2(1H)-on e (5b). Crude 5b was crystallized from
mixture of EtOAc and MeOH to yield white crystals: 110 mg
(85%); mp 91-95 °C (lit.17 114 °C); UV λmax (pH 0) 285 nm (ꢀ
13.60 × 103), 220 nm (ꢀ 9.60 × 103); λmax (pH 1) 282.5 nm (ꢀ
12.40 × 103), 220 nm (ꢀ 9.00 × 103); λmax (pH 2) 278 nm
(ꢀ 9.90 × 103), 224 nm (ꢀ 9.40 × 103); λmax (pH 7) 268 nm
(ꢀ 8.60 × 103), 236 nm (ꢀ 11.55 × 103); λmax (pH 12) 252 nm
1-(3,5-Di-O-acetyl-â-D-2-deoxyr ibofu r an osyl)-4-h ydr oxy-
a m in o-5-br om op yr im id in -2(1H)-on e (4g). 1.28 g (3 mmol)
of 1-(3,5-di-O-acetyl-â-D-2-deoxyribofuranosyl)-5-bromouracil
(1g) was dissolved in 15 mL of dry MeCN and added to the
N-methylphosphoimidazolide prepared from 2.4 mL (30 mmol)
of N-methylimidazole and 840 µL (9 mmol) POCl3 in 60 mL of
MeCN. The mixture was stirred for 2 h at room temperature
to give intermediate 3g and 9.5 mmol of hydroxylamine in 10
mL of MeOH was added. Stirring was continued for 2 h at
room temperature and reaction mixture was evaporated to
dryness in vacuo. The residue was dissolved in 20 mL of water
and extracted (3 × 50 mL) with EtOAc. The extract was
washed with 30 mL water, dried over MgSO4, evaporated to
dryness in vacuo to give ca. 1 g of crude product. An analytical
sample was purified by PLC in the mixture of CHCl3-MeOH
97:3 to yield after crystallization from MeOH 20 mg of 4g: mp
82-85 °C; Rf (C) 0.48; 1H NMR δ (CDCl3) 7.22 (1H, s, H6),
6.32 (1H, dd, H1′), 5.22 (1H, m, H3′), 4.38 (1H, dd, H5′′), 4.32
(1H, dd, H5′), 4.23 (1H, m, H4′), 2.43-2.38 (1H, m, H2′′), 2.20-
2.10 (1H, m, H2′), 2.19, 2.11 (6H, 2 × s, 2 × CH3CO). Anal.
(C13H16BrN3O7‚0.25MeOH) C, H, N.
1-(3,5-Di-O-acetyl-â-D-2-deoxyr ibofu r an osyl)-4-h ydr oxy-
a m in o-5-iod op yr im id in -2(1H)-on e (4h ). Crude nucleoside
2h was dissolved in 20 mL of pyridine and hydroxylamine
hydrochloride (150 mg, 2.16 mmol) was added. The mixture
was stirred at room temperature for 12 h, evaporated under
reduced pressure and coevaporated successively with 2 × 20
mL of toluene and EtOH. 50 mL of CHCl3 was added and the
mixture was washed with water (2 × 30 mL). The organic layer
was separated, dried (Na2SO4) and evaporated under reduced
pressure. Resulting crude 4h was purified on a silica gel
column (1 × 50 cm) with the use of CHCl3 to yield 385 mg
(84%) of 4h . An analytical sample was crystallized from MeOH
to yield white crystals: mp 90-92 °C; UV λmax (pH 0) 305.5
nm (ꢀ 9.9 × 103), 230 nm (ꢀ 11.4 × 103), λmax (pH 1) 293 nm
(ꢀ 8.05 × 103), 232 nm (ꢀ 11.65 × 103), λmax (pH 2) 284 nm
1
(ꢀ 10.80 × 103); TLC (silica gel) Rf (A) 0.49; H NMR δ (D2O)
6.96 (1H, s, H6), 6.28 (1H, t, H1′, J 1′2′ ) 7.54 Hz, J 1′2′′ ) 6.58
Hz), 4.44 (1H, m, H3′, J 3′4′ ) 3.99 Hz, J 3′2′ ) 6.89 Hz, J 3′2′′
)
3.70 Hz), 3.97 (1H, m, H4′, J 4′5′ ) 3.99 Hz, J 4′5′′ ) 5.02 Hz),
3.81 (1H, dd, H5′, J 5′5′′ ) -12.41 Hz), 3.74 (1H, dd, H5′′), 2.34
(1H, m, H2′′), 2.27 (1H, m, H2′, J 2′2′′ ) -14.25 Hz), 1.81 (3H,
s, 5-CH3); MS m/z 258 (M + H)+.
1-(â-D-2-Deoxyr ib ofu r a n osyl)-4-h yd r oxya m in o-5-h y-
d r oxym eth ylp yr im id in -2(1H)-on e (5d ). Crude 5d was crys-
tallized from mixture of toluene and MeOH to yield white
crystals: 127 mg (93%); mp 188-190 °C dec; UV λmax (pH 0)
281.5 nm (ꢀ 14.20 × 103), 219 (ꢀ 12.40 × 103); λmax (pH 1) 281.5
nm (ꢀ 9.60 × 103), 224.5 nm (ꢀ 8.80 × 103); λmax (pH 2) 275 nm
(ꢀ 7.60 × 103), 230 nm (ꢀ 12.10 × 103); λmax (pH 7) 274 nm (ꢀ
6.90 × 103), 231.5 nm (ꢀ 12.75 × 103); λmax (pH 12) 251 nm (ꢀ
10.60 × 103); TLC (silica gel) Rf (A) 0.25; 1H NMR δ (D2O) 7.14
(1H, s, H6), 6.30 (1H, t, H1′, J 1′2′ ) 7.44 Hz J 1′2′′ ) 6.63 Hz),
4.46 (1H, m, H3′, J 3′4′ ) 3.70 Hz, J 3′2′ ) 6.79 Hz, J 3′2′′ ) 3.80
Hz), 4.29 (2H, d, 5-CH2OH), 3.99 (1H, m, H4′, J 4′5′ ) 3.53 Hz,
J 4′5′′ ) 5.07 Hz), 3.83 (1H, dd, H5′, J 5′5′′ ) -12.43 Hz), 3.76
(1H, dd, H5′′), 2.36 (1H, m, H2′′), 2.30 (1H, m, H2′, J 2′2′′
)
-14.14 Hz); MS m/z 274.10151 [(M + H)+ calcd for C10H16N3O6
274.10391]. Anal. (C10H15N3O6) C, H, N.
1-(â-D-2-Deoxyr ibofu r a n osyl)-4-h yd r oxya m in o-5-flu o-
r op yr im id in -2(1H)-on e (5e).18 Title compound was obtained
as a glass: 115 mg (88%); mp 94-97 °C; UV λmax (pH 0) 286
nm (ꢀ 14.6 × 103), 222 nm (ꢀ 10.6 × 103), λmax (pH 7) 270 nm
(ꢀ 10.4 × 103), 234 nm (ꢀ 13.3 × 103), λmax (pH 12) 253 nm (ꢀ
13.5 × 103); TLC (silica gel) Rf (A) 0.44; 1H NMR δ (D2O) 7.34
(1H, d, H6) 6.32 (1H, td, H1′), 4.47 (1H, m, H3′, J 2′3′ ) 5.14
Hz, J 2′′3′ ) 5.14 Hz), 4.02 (1H, m, H4′, J 4′3′ ) 3.65 Hz), 3.85
(1H, dd, H5′, J 4′5′ ) 3.65 Hz), 3.78 (1H, dd, H5′′, J 4′5′′ ) 4.91

Deoxycytidines and Their 5′-Monophosphates
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24 4655
Hz) 2.33 (2H, dd, H2′, H2′′, J 1′2′+J 1′2′′/2 ) 6,96 Hz,); MS m/z
262 (M + H)+.
lyzed to their mother nucleosides. An additional control, 2′-
(3′)-GMP was unaffected, pointing to the absence of nonspecific
phosphatases.
1-(â-D-2-Deoxyr ibofu r a n osyl)-4-h yd r oxya m in o-5-ch lo-
r op yr im id in -2(1H)-on e (5f). The product 5f was obtained
as an amorphous foam: 120 mg (87%); mp 92-95 °C; UV λmax
(pH 0) 294 nm (ꢀ 15.3 × 103), 226 nm (ꢀ 11.95 × 103); λmax (pH
1) 288 nm (ꢀ 10.8 × 103), 226 nm (ꢀ 12.45 × 103); λmax (pH 2)
282 nm (ꢀ 10.2 × 103), 233 nm (ꢀ 13.2 × 103); λmax (pH 7) 281
nm (ꢀ 10.1 × 103), 234 nm (ꢀ 13.8 × 103); λmax (pH 12) 284 nm
(ꢀ 9.4 × 103), 250 nm (ꢀ 1.42 × 103); TLC (silica gel) Rf (A)
0.54; 1H NMR δ [ppm] (D2O) 7.40 (1H, s, H6), 6.25 (1H, t, H1′,
J 1′2′ ) 6.95 Hz, J 1′2′′ ) 6.74 Hz), 4.44 (1H, m, H3′, J 3′4′ ) 4.14
Hz), 3.98 (1H, m, H4′), 3.81 (1H, dd, H5′, J 4′5′ ) 4.02 Hz), 3.74
(1H, dd, H5′′, J 4′5′′ ) 4.47 Hz, J 5′5′′ ) -12.46 Hz), 2.32 (1H, m,
H2′′, J 2′2′′ ) -14.21 Hz, J 2′′3′ ) 3.42 Hz), 2.30 (1H, m, H2′,
J 2′3′ ) 7.32 Hz). Anal. (C9H12ClN3O5) C, H, N.
1-(â-D-2-Deoxyr ib ofu r a n osyl)-4-h yd r oxya m in o-5-b r o-
m op yr im id in -2(1H)-on e (5g). Crude 5g was crystallized
from MeOH to yield white crystals: 500 mg (67%); mp 187-
190 °C dec; UV λmax (pH 0) 297 nm (ꢀ 11.35 × 103), 225 nm (ꢀ
9.00 × 103); λmax (pH 1) 291 nm (ꢀ 8.75 × 103), 229 nm (ꢀ 10 ×
103); λmax (pH 2) 283 nm (ꢀ 7.80 × 103), 231 nm (ꢀ 10.80 ×
103); λmax (pH 7) 282 nm (ꢀ 7.80 × 103), 231 nm (ꢀ 11.05 ×
103); λmax (pH 12) 282 nm (ꢀ 6.70 × 103), 245 nm (ꢀ 10.30 ×
103); TLC (silica gel) Rf (A) 0.54; 1H NMR δ (D2O) 7.49 (1H, s,
H6), 6.25 (1H, dd, H1′, J 1′2′ ) 6.90 Hz, J 1′2′′ ) 6.68 Hz), 4.44
(1H, m, H3′, J 2′3′ ) 7.12 Hz), 3.98 (1H, m, H4′, J 3′4′ ) 3.96 Hz),
3.81 (1H, dd, H5′, J 4′5′ ) 3.73 Hz, J 5′5′′ ) -12.44), 3.74 (1H,
dd, H5′′J 4′5′′ ) 4.82 Hz), 2.33 (1H, m, H2′′, J 2′′3′ ) 3.73 Hz),
2.30 (1H, m, H2′, J 2′2′′ ) -14.22 Hz). Anal. (C9H12BrN3O5) C,
H, N.
1-(â-D-2-Deoxyr ibofu r a n osyl)-4-h yd r oxya m in op yr im i-
d in -2(1H)-on e 5′-m on op h osp h a te (6a ): 18 mg (70%); UV
λ
max (pH 0) 278 nm (ꢀ 10.30 × 103), 218 nm (ꢀ 6.83 × 103); λmax
(pH 7) 271 nm (ꢀ 5.15 × 103), 234 nm (ꢀ 9.99 × 103); λmax (pH
12) 241 nm (ꢀ 8.12 × 103); TLC (cellulose) Rf (D) 0.49, (E) 0.30,
(F) 0.42; P/ꢀ 1.00.
1-(â-D-2-Deoxyr ibofu r a n osyl)-4-h yd r oxya m in o-5-m eth -
ylp yr im id in -2(1H)-on e 5′-m on op h osp h a te (6b): 18.3 mg
(68%); UV λmax (pH 0) 283 nm (ꢀ 13.70 × 103), 219 nm
(ꢀ 9.50 × 103); λmax (pH 7) 269 nm (ꢀ 8.51 × 103), 236 nm
(ꢀ 11.67 × 103); λmax (pH 12) 253 nm (ꢀ 10.69 × 103); TLC
(cellulose) Rf (D) 0.46, (E) 0.32, (F) 0.42; P/ꢀ 1.01.
1-(â-D-2-Deoxyr ib ofu r a n osyl)-4-h yd r oxya m in o-5-h y-
dr oxym eth ylpyr im idin -2(1H)-on e 5′-m on oph osph ate (6d):
15.8 mg (57%); UV λmax (pH 0) 280 nm (ꢀ 14.00 × 103), 218
(ꢀ 12.30 × 103); λmax (pH 7) 275 nm (ꢀ 6.83 × 103), 231 nm
(ꢀ 12.63 × 103); λmax (pH 12) 252 nm (ꢀ 10.50 × 103); TLC
(cellulose) Rf (D) 0.41, (E) 0.27, (F) 0.27; P/ꢀ 1.02.
1-(â-D-2-Deoxyr ibofu r a n osyl)-4-h yd r oxya m in o-5-flu o-
r op yr im id in -2(1H)-on e 5′-m on op h osp h a te (6e): 16.8 mg
(62%); UV λmax (pH 0) 282 nm (ꢀ 14.5 × 103), 223 nm (ꢀ 10.5 ×
103), λmax (pH 7) 269 nm (ꢀ 10.3 × 103), 234 nm (ꢀ 13.2 × 103),
λ
max (pH 12) 254 nm (ꢀ 13.4 × 103); TLC (cellulose) Rf (D) 0.44,
(E) 0.28, (F) 0.41; P/ꢀ 1.01.
1-(â-D-2-Deoxyr ibofu r a n osyl)-4-h yd r oxya m in o-5-ch lo-
r op yr im id in -2(1H)-on e 5′-m on op h osp h a te (6f): 14 mg
(50%); UV λmax (pH 0) 293 nm (ꢀ 15.2 × 103), 225 nm
(ꢀ 11.80 × 103); λmax (pH 7) 282 nm (ꢀ 10.0 × 103), 234 nm
(ꢀ 13.7 × 103); λmax (pH 12) 285 nm (ꢀ 9.3 × 103), 250 nm
(ꢀ 1.30 × 103); TLC (cellulose) Rf (D) 0.46, (E) 0.30, (F) 0.27;
P/ꢀ 1.00.
1-(â-D-2-Deoxyr ib ofu r a n osyl)-4-h yd r oxya m in o-5-b r o-
m op yr im id in -2(1H)-on e 5′-m on op h osp h a te (6g): 17.5 mg
(58%); UV λmax (pH 0) 295 nm (ꢀ 11.20 × 103), 224 nm
(ꢀ 8.90 × 103); λmax (pH 7) 283 nm (ꢀ 7.70 × 103), 231 nm
(ꢀ 10.90 × 103); λmax (pH 12) 283 nm (ꢀ 6.60 × 103), 246 nm
(ꢀ 10.20 × 103); TLC (cellulose) Rf (D) 0.48, (E) 0.28, (F) 0.29;
P/ꢀ 1.01.
1-(â-D-2-Deoxyr ibofu r an osyl)-4-h ydr oxyam in o-5-iodopy-
r im id in -2(1H)-on e (5h ). Crude 5h after crystallization from
MeOH yield white crystals: 146 mg (79%); mp 181-184 °C
dec; UV λmax (pH 0) 305.5 nm (ꢀ 9.9 × 103) 230 nm (ꢀ 11.4 ×
103); λmax (pH 1) 293 nm (ꢀ 8.05 × 103), 232 nm (ꢀ 11.65 ×
103); λmax (pH 2) 284 nm (ꢀ 7.95 × 103), 233 nm (ꢀ 11.60 ×
103); λmax (pH 7) 284.5 nm (ꢀ 8.05 × 103), 233 nm (ꢀ 11.60 ×
103); λmax (pH 12) 247.5 nm (ꢀ 11.90 × 103); TLC (silica gel)
1
Rf (A) 0.59; H NMR (D2O) 7.56 (1H, s, H6), 6.22 (1H, t, H1′,
J 1′,2′ ) 6.87 Hz, J 1′,2′′ ) 6.71 Hz), 4.44 (1H, m, H3′, J 3′4′ ) 4.00
Hz), 3.97 (1H, m, H4′, J 4′,5′ ) 3.53 Hz, J 4′5′′ ) 4.75 Hz), 3.82
(1H, dd, H5′, J 5′5′′ ) -12.55 Hz), 3.75 (1H, m, H5′′), 2.34 (1H,
m, H2′′, J 2′2′′ ) -14.25 Hz, J 2′′3′ ) 3.98 Hz), 2.30 (1H, m, H2′,
J 2′3′ ) 6.81 Hz). Anal. (C9H12IN3O5‚0.33 MeOH) C, H, N.
1-(â-D-2-Deoxyr ibofu r an osyl)-4-h ydr oxyam in o-5-iodopy-
r im id in -2(1H)-on e 5′-m on op h osp h a te (6h ): 19.5 mg (60%);
UV λmax (pH 0) 304 nm (ꢀ 9.8 × 103) 229 nm (ꢀ 11.3 × 103);
λ
max (pH 7) 285 nm (ꢀ 7.97 × 103), 233 nm (ꢀ 11.50 × 103); λmax
Gen er a l P r oced u r e for th e Syn th esis of Nu cleosid e 5′-
P h osp h a tes. To a solution of 0.05 mmol of the appropriate
nucleoside analogue 5a ,b,d -h in 1.5 mL of 0.1 M acetate
buffer pH 4 was added 280 mg (0.75 mmol) of p-nitrophenyl
phosphate and the pH was brought to 4 by addition of
concentrated acetic acid. To this was added 1.5 mL of a crude
extract of wheat shoot nucleoside phosphotransferase. The
mixture was incubated at 37 °C for 40 h, concentrated to half-
volume and extracted 3 times with ether. After evaporation
of the aqueous layer the product was purified by column
chromatography on a DEAE-Sephadex A-25 eluted with a
linear gradient of aqueous Et3N-H2CO3 (0-0.5 M). The
homogeneous fractions containing pure monophosphate were
pooled and lyophilized to remove excess bicarbonate buffer to
yield the appropriate nucleoside 5′-monophosphates 6a ,b,d -h
as triethylammonium salts. UV spectra of nucleotides 6a,b,d-h
were almost identical to spectra of their mother nucleosides
5a ,b,d -h (see below).
(pH 12) 248 nm (ꢀ 11.80 × 103); TLC (cellulose) Rf (D) 0.46,
(E) 0.29, (F) 0.27; P/ꢀ 1.00.
Ack n ow led gm en t. This work was supported by the
Polish State Committee for Scientific Research, Grants
4P05F03011p01 and 4P05F03011p02.
Refer en ces
(1) Carreras, C. W.; Santi, D. V. The Catalytic Mechanism and
Structure of Thymidylate Synthase. Annu. Rev. Biochem. 1995,
64, 721-762.
(2) Rode, W.; Les´, A. Molecular Mechanism of Thymidylate Syn-
thase-catalyzed Reaction and Interaction of the Enzyme with
2- and/or 4-substituted Analogues of dUMP and 5-Fluoro-dUMP.
Acta Biochim. Pol. 1996, 43, 133-142.
(3) Danenberg, P. V. Thymidylate Synthetase - a Target Enzyme
in Cancer Chemotherapy. Biochim. Biophys. Acta 1977, 473, 73-
92.
(4) Heidelbereger, C.; Danenberg, P. V.; Moran, R. G. Fluorinated
Pyrimidines and Their Nucleosides. Adv. Enzymol. 1983, 54, 57-
119.
(5) J ackman, A. L.; J ones, T. R.; Calvert, A. H. Thymidylate
Synthetase Inhibitors: Experimental and Clinical Aspects. In
Experimental and Clinical Progress in Cancer Chemotherapy;
Muggia, F. M., Ed.; Martinus Nijhoff Publishers: Boston, 1985;
pp 155-210.
(6) Ealick, S. E.; Armstromg, S. R. Pharmacologically Relevant
Proteins. Curr. Opin. Struct. Biol. 1993, 3, 861-867.
(7) Santi, D. V. Perspectives on the Design and Biochemical
Pharmacology of Inhibitors of Thymidylate Synthetase. J . Med.
Chem. 1980, 23, 103-111.
Gen er a l P r oced u r e for th e En zym a tic Hyd r olysis of
Nu cleosid e 5′-P h osp h a tes. To a solution of 40 µL 0.1 M Tris/
HCl buffer pH 8.8 + 20 µL 0.1 M MgCl2 was added 0.05 µmol
of nucleoside 5′-phosphate, followed by 5 µL of a 10 mg/mL
stock solution of Crotalus adamanteus (EC 3.1.3.5.) snake
venom phosphodiesterase (5′-nucleotidase). Following 2 h
incubation at 37 °C, nucleoside 5′-monophosphates 6a ,b,d -h
underwent ∼50% hydrolysis to their mother nucleosides
5a ,b,d -h while the “natural” substrate dCMP was quantita-
tively converted to the parent nucleoside dCyd. After overnight
incubation at 37 °C all nucleotides were quantitatively hydro-

4656 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 24
Felczak et al.
(8) Lewis, C. A., J r.; Dunlap, R. B. Thymidylate Synthase and its
Interaction with 5-Fluoro-2′-deoxyuridylate. In Topics in Mo-
lecular Pharmacology; Burgen, A. S. V., Roberts, G. C. K., Eds.;
Elsevier/North-Holland Biomedical: New York, 1981; pp 170-
219.
(9) De Clercq, E.; Balzarini, J .; Torrence, P. F.; Mertes, M. P.;
Schmidt, C. L.; Shugar, D.; Barr, P. J .; J ones, A. S.; Verhelst,
G.; Walker, R. T. Thymidylate Synthetase As Target Enzyme
for the Inhibitory Activity of 5-substituted 2′-deoxyuridines on
Mouse Leukemia L1210 Cell Growth. Mol. Pharmacol. 1981, 19,
321-330.
(10) Lorenson, M. Y.; Maley, G. F.; Maley, F. The Purification and
Properties of Thymidylate Synthetase From Chick Embryo
Extracts. J . Biol. Chem. 1967, 242, 3332-3344.
(27) Clark, M.; Cramer, R. D.; Van Opdenbosh, N. Validation of the
General Purpose Tripos 5.2 Force Field. J . Comput. Chem. 1989,
10, 982-1012.
(28) Les´, A.; Adamowicz, L.; Rode, W. Structure and Conformation
of N4-hydroxycytosine and N4-hydroxy-5-fluorocytosine. A Theo-
retical Ab Initio Study. Biochim. Biophys. Acta 1993, 1173, 39-
48.
(29) Gunther, H. NMR Spectroscopy. An Introduction; J ohn Wiley
and Sons: Chichester, 1980; pp 125-140.
(30) Remin, M. Thermodynamic cycle between DNA and RNA
Constituents for Conformation of the Sugar Ring from Nuclear
Magnetic Resonance Study. J . Biomol. Struct. Dyn. 1997, 15,
251-264.
(31) De Leeuw, F.A. A. M.; Altona, C. Conformation Analysis of â-D-
Ribo, â-D-Deoxyribo, â-D-Arabino, â-D-Xylo, â-D-Lyxo-nucleo-
sides from Proton-Proton Coupling Constant. J . Chem. Soc.,
Perkin Trans. 2 1982, 375-384.
(32) Garret, C.; Wataya, Y.; Santi, D. V. Thymidylate Synthase.
Catalysis of Dehalogenation of 5-bromo- and 5-iodo-2′-deoxy-
uridylate. J . Med. Chem. 1979, 13, 2798-2804.
(33) Nelson, D. J .; Carter, C. E. Inhibition of the Biosynthesis of
Thymidylic Acid by 4-N-hydroxy-2′-deoxycytidine in L5178Y
Leukemic Cells. Mol. Pharmacol. 1966, 2, 248-258.
(34) Dollinger, M. R.; Burchenal, J .h.; Kreis, W.; Fox, J . J . Analogues
of 1-â-D-arabinofuranosylcytosine. Studies on Mechanisms of
Action in Burkitt’s Cell Culture and Mouse Leukemia, and in
vitro Deamination Studies. Biochem. Pharmacol. 1967, 16, 689-
706.
(35) Les´, A.; Adamowicz, L.; Rode, W. Modelling of Reaction Steps
Relevant to Deoxyuridylate (dUMP) Enzymatic Methylation and
Thymidylate Synthase Mechanism-based Inhibition. J . Biomol.
Struct. Dyn. 1998, 15, 703-715.
(11) Goldstein, S.; Pogolotti, A. L.; Garvey, J r.; E. P.; Santi, D. V.
Interaction of N4-hydroxy-2′-deoxycytidylic Acid with Thymidy-
late Synthetase. J . Med. Chem. 1984, 27, 1259-1262.
(12) Rode, W.; Zielin´ski, Z.; Dzik, J . M.; Kulikowski, T.; Bretner M.;
Kierdaszuk B.; Cies´la, J .; Shugar D. Mechanism of Inhibition
of Mammalian Tumor and Other Thymidylate Synthases by N4-
hydroxy-dCMP, N4-hydroxy-5-fluoro-dCMP, and Related Ana-
logues. Biochemistry 1990, 29, 10835-10842.
(13) Brown, D. M.; Shell, P. The Reaction of Hydroxylamine with
Cytosine and Related Compounds. J . Mol. Biol. 1961, 3, 709-
710.
(14) Brown, D. M.; Shell, P. Nucleosides Part XLVIII. The Reaction
of hydroxylamine with Cytosine and Related Compounds. J .
Chem. Soc. 1965, 208-215.
(15) Maley, G. F.; Maley, F. The Purification and Properties of
Deoxycytidylate Deaminase From Chick Embryo Extracts. J .
Biol. Chem. 1964, 239, 1168-1176.
(16) J anion, C.; Shugar, D. Studies on Possible Mechanisms of
Hydroxylamine Mutagenesis. Acta Biochim. Polon. 1968, 15,
107-121.
(36) Wataya, Y.; Santi, D. V. Inhibition of Loctobacillus casei
Thymidylate Synthetase by 5-Substituted 2′-Deoxyuridylates.
Preliminary Quantitative Structure-Activity Relationship. J .
Med. Chem. 1977, 20, 1469-1473.
(37) Chen, P. S., J r.; Toribara, T. Y.; Warner, H. Microdetermination
of Phosphorus. Anal. Chem. 1956, 28, 1756-1758.
(17) Fox, J . J .; Van Praag, D.; Wempen, I.; Doerr, I. L.; Cheong, L.;
Knoll, J . E.; Eidinoff, M. L.; Bendich, A.; Brown, G. B. Thiation
of Nucleosides. II. Synthesis of 5-Methyl-2′-deoxycytidine and
Related Pyrimidine Nucleosides. J . Am. Chem. Soc. 1959, 81,
178-187.
(18) Wempen, I.; Miller, N.; Falco, E. A.; Fox, J . J . Nucleosides.
XLVII. Syntheses of Some N4-substituted Derivatives of 1-â-D-
arabinofuranosylcytosine and -5-Fluorocytosine. J . Med. Chem.
1968, 11, 144-148.
(19) Mertes, M. P.; Smrt, J . Nucleic Acid Components and Their
Analogues. CXV. Synthesis of N4-Hydroxy-6-azacytidine 5′-
Phosphate and 5′-Diphosphate. Collect. Czech. Chem. Commun.
1968, 33, 3304-3312.
(20) Herdewijn, P.; Balzarini, J .; Baba, M.; Pauwels, R.; J anssen, G.;
De Clercq, E. Synthesis and Anti HIV Activity of Different Sugar
Modified Pyrimidine and Purine Nucleosides. J . Med. Chem.
1988, 31, 2040-2048.
(21) Robins, M. J .; Barr, P. J .; Giziewicz, J . Nucleic Acid Related
Compounds. 38. Smooth and High-yield Iodination and Chlori-
nation at C-5 of Uracil Bases and p-Tolyl-Protected Nucleosides.
Can. J . Chem. 1982, 60, 554-557.
(22) Asakura, J .; Robins, M. J . Cerium (IV)-Mediated Halogenation
at C-5 of Uracil Derivatives. J . Org. Chem. 1990, 55, 4928-4933.
(23) Robins, M. J .; MacCoss, M.; Naik, S. R,; Ramani, G. Nucleic Acid
Related Compounds. 21. Direct Fluorination of Uracil and
Cytosine Bases and Nucleosides Using Trifluoromethyl Hypo-
fluorite. Mechanism, Stereochemistry, and Synthetic Applica-
tions. J . Am. Chem. Soc. 1976, 98, 7381-7390.
(24) Divakar, K. J .; Reese, C. B. 4-(1,2,4-Triazol-1-yl)- and 4-(3-Nitro-
1,2,4-triazol-1-yl)-1-(â-D-2,3,5-tri-O-acetylarabinofuranosyl)py-
rimidin-2(1H)-ones. Valuable Intermediates in the Synthesis of
Derivatives of 1-(â-D-Arabinofuranosyl)cytosine (Ara-C). J . Chem.
Soc., Perkin Trans. 1 1982, 1171-1176.
(25) Matsuda, A.; Obi, K.; Miyasaka, T. Reaction of Uracil Nucleo-
sides with 1-Methylimidazole in the Presence of Phosphoryl
Chloride. A Convenient Method for the Synthesis of 4-Substi-
tuted Pyrimidin-2(1H)-one. Nucleosides. Chem. Pharm. Bull.
1985, 33, 2575-2578.
(26) Giziewicz, J .; Shugar, D. Enzymatic Phosphorylation of Nucleo-
sides to the 5′-phosphates. In Nucleic Acid Chemistry; Townsend,
L. B., Tipson, R. S., Eds.; J ohn Wiley: New York, 1985; pp 955-
961.
(38) Dzik, J . M.; Bretner, M.; Kulikowski, T.; Gołos, B.; J armuła, A.;
Poznan´ski, J .; Rode, W.; Shugar, D. Synthesis and Interactions
with Thymidylate Synthase of 2,4-dithio- Analogues of dUMP
and 5-fluoro-dUMP. Biochim. Biophys. Acta 1996, 1293, 1-8.
(39) Zielin´ski, Z.; Dzik, J . M.; Rode, W.; Kulikowski, T.; Bretner, M.;
Kierdaszuk, B.; Shugar, D. Interaction of N4-hydroxy-dCMP and
N4-hydroxy-5-FdCMP with L1210 Thymidylate Synthase Dif-
fering in Sensitivity Towards 5-FdUMP Inhibition. In Chemistry
and Biology of Pteridines 1989. Pteridines and Folic Acid
Derivatives; Curtius, H.-Ch., Ghisla, S., Blau, N., Eds.; Walter
de Gruyter: Berlin, 1990; pp 817-820.
(40) Cies´la, J .; Weiner, K. X. B.; Weiner, R. S.; Reston, J . T.; Maley,
G. F.; Maley, F. Isolation and Expression of Rat Thymidylate
Synthase cDNA: Phylogenetic Comparison with Human and
Mouse Thymidylate Synthases. Biochim. Biophys. Acta 1995,
1261, 233-242.
(41) Da¸browska, M.; Zielin´ski, Z.; Wranicz, M.; Michalski, R.; Pawełc-
zak, K.; Rode, W. Trichinella Spiralis Thymidylate Synthase:
Developmental Pattern, Isolation, Molecular Properties and
Inhibition by Substrate and Cofactor Analogues. Biochem.
Biophys. Res. Commun. 1996, 228, 440-445.
(42) Brouillette, C. B.; Chang, C. T.-C.; Mertes, M. P. 5-(R-Bro-
moacetyl)-2′-Deoxyuridine 5′-phosphate: An Affinity Label for
Thymidylate Synthetase. J . Med. Chem. 1979, 22, 1541-1544.
(43) Tanty, C. R.; Lopez-Canovas, L.; Brauet, A. L.; Paredes, M. R.;
Clarke, D. H.; Castro, H. V.; Rojas, A. M. R.; Cabrera, A. M.
Introduction of an Immunochemical Label in a Cytidine Ana-
logue. Nucleosides Nucleotides 1995, 14, 219-228.
(44) Poznan´ski, J .; Felczak, K.; Kulikowski, T.; Remin, M. 1H NMR
Conformational Study of Antiherpetic C5-Substituted 2′-Deoxy-
uridines: Insight into the Nature of Structure-Activity Rela-
tionships. Biochem. Biophys. Res. Commun. 2000, 272, 64-74.
J M000975U