Effects of 8-chlorodeoxyadenosine on DNA synthesis by the Klenow fragment of DNA polymerase I

Article abstract of DOI:10.1016/S0960-894X(03)00204-X

8-Chloro-2′-deoxyadenosine (8-Cl-dAdo) was incorporated into synthetic DNA oligonucleotides to determine its effects on DNA synthesis by the 3′-5′ exonuclease-free Klenow fragment of Escherichia coli DNA Polymerase I (KF^-). Single nucleotide insertion experiments were used to determine the coding potential of 8-Cl-dAdo in a DNA template. KF^- inserted TTP opposite 8-Cl-dAdo in the template, but with decreased efficiency relative to natural deoxyadenosine. Running-start primer extensions with KF^- resulted in polymerase pausing at 8-Cl-dAdo template sites during DNA synthesis. The 2′-deoxyribonucleoside triphosphate analogue, 8-Cl-dATP, was incorporated opposite thymidine (T) approximately two-fold less efficiently than dATP.

Full text of DOI:10.1016/S0960-894X(03)00204-X

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1509–1512
Effects of 8-Chlorodeoxyadenosine on DNA Synthesis by the
Klenow Fragment of DNA Polymerase I
Lisa S. Chen, Michael H. Bahr and Terry L. Sheppard*
Department of Chemistry and The Robert H. Lurie Comprehensive Cancer Center, Northwestern University,
2145 Sheridan Road, Evanston, IL 60208-3113, USA
Received 18 January 2003;revised 21 February 2003;accepted 27 February 2003
Abstract—8-Chloro-20-deoxyadenosine (8-Cl-dAdo) was incorporated into synthetic DNA oligonucleotides to determine its effects
on DNA synthesis by the 30–50 exonuclease-free Klenow fragment of Escherichia coli DNA Polymerase I (KFꢀ). Single nucleotide
insertion experiments were used to determine the coding potential of 8-Cl-dAdo in a DNA template. KFꢀ inserted TTP opposite
8-Cl-dAdo in the template, but with decreased efficiency relative to natural deoxyadenosine. Running-start primer extensions with
KFꢀ resulted in polymerase pausing at 8-Cl-dAdo template sites during DNA synthesis. The 20-deoxyribonucleoside triphosphate
analogue, 8-Cl-dATP, was incorporated opposite thymidine (T) approximately two-fold less efficiently than dATP.
# 2003 Elsevier Science Ltd. All rights reserved.
Nucleoside analogues have been used extensively as
medicinal agents, particularly in anti-viral and cancer
treatments.1,2 Several 8-modified purine analogues have
demonstrated antineoplastic activity and show sig-
nificant promise as chemotherapeutic agents.3,4
8-Chloroadenosine (8-Cl-Ado) is currently under investi-
gation as a potential treatment for multiple myeloma
and leukemia. 8-Cl-Ado induces apoptosis in multiple
myeloma cell lines in culture,4 but the mechanism of
action remains unclear. Apoptosis may result from 8-Cl-
Ado-mediated inhibition of transcription or RNA pro-
cessing, or by incorporation of the analogue into RNA,
which may alter cellular RNA structure or function. In
support of the latter hypothesis, our laboratory has
shown that direct incorporation of 8-Cl-Ado into RNA
oligonucleotides leads to RNA duplex destabilization.5
However, apoptosis induction may involve 8-chloro-20-
deoxyadenosine (8-Cl-dAdo) and its metabolites
through an indirect mechanism. After cellular uptake,
8-Cl-Ado is converted rapidly to 50-phosphorylated
8-Cl-Ado derivatives.3 These 8-Cl-Ado nucleotides then
may enter the deoxyribonucleotide pool as 8-Cl-dATP
by the combined action of ribonucleotide reductase and
a kinase. As a potential substrate for cellular DNA
polymerases, 8-Cl-dATP may be incorporated into
newly synthesized DNA.
Structural modifications of deoxynucleoside triphos-
phates (dNTPs) may affect the fidelity of nucleotide
insertion during enzymatic DNA synthesis. Purine
nucleoside analogues modified at the C-8 position fre-
quently adopt syn glycosidic torsion angles instead of the
preferred anti conformation found in unmodified
nucleosides.6ꢀ8 For example, substitution of adenine H-8
with larger groups, in analogues such as 8-methoxy-20-
deoxyadenosine9 or 8-bromoadenosine,10 biases nucleo-
side torsional preferences toward the non-standard syn
glycosidic bond conformation. NMR studies on 8-Cl-
Ado nucleotides suggest that chlorine substitution at the
C-8 position of adenine favors the syn base conformer.11
By analogy to DNA damage lesions such as 8-oxo-
dG,12 the syn preference for 8-chloroadenine (8-Cl-A)
bases may lead to base mispairing during enzymatic
DNA synthesis or in transcription. To test this hypoth-
esis experimentally, we synthesized DNA oligonucleo-
tides containing 8-Cl-dAdo and used them to assess the
effects of 8-Cl-dAdo derivatives on DNA synthesis. In
this work, we address the coding potential of 8-Cl-dAdo
in DNA templates and the insertion fidelity of 8-Cl-
dATP using exonuclease free Klenow fragment (KFꢀ)
from Escherichia coli DNA polymerase I. 8-Cl-dAdo
derivatives pair with thymidine, but exhibit decreased
nucleotide incorporation efficiencies and induce KFꢀ
polymerase pausing during enzymatic DNA synthesis.
To incorporate 8-Cl-dAdo into synthetic DNA oligo-
nucleotides, an 8-Cl-dAdo phosphoramidite derivative
*Corresponding author. Tel.: +1-847-467-7636;fax: +1-847-491-
7713;e-mail: sheppard@chem.northwestern.edu
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00204-X

1510
L. S. Chen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1509–1512
Scheme 1. (a) Chemical synthesis of 8-Cl-dAdo phosphoramidite 6, and (b) sequences of standard and 8-Cl-dAdo modified DNA oligonucleotides.
was synthesized from deoxyadenosine (dAdo) in a six-
step synthesis outlined in Scheme 1A. Deoxyadenosine
(1) was disilylated with TBDMSCl (98%)13 and sub-
jected to chlorination using tosyl chloride and LDA to
yield derivative 2 (60%).14 The silyl ethers were cleaved
with tetra-n-butylammonium fluoride (TBAF) in THF
to give 8-Cl-dAdo (3, 58%). Transient protection of the
nucleoside hydroxyls with TMSCl, followed by benzoyl-
ation15 and aqueous work up afforded compound 4
(90%). Benzoylated nucleoside 4 was converted to the
4,40-dimethoxytrityl (DMT) derivative 5 (44%) using
DMT-Cl in the presence of DMAP in pyridine. 8-Cl-
dAdo phosphoramidite 6 was synthesized from com-
pound 5 using standard phosphoramidite chemistry.16,17
Phosphoramidite analogue 6 was incorporated into
DNA oligonucleotides using standard solid-phase
synthesis methods. Oligonucleotide 9, which contained
a single 8-Cl-dAdo substitution (X=8-Cl-dAdo), was
synthesized and deprotected using standard methods
(Scheme 1B).18 Following oligonucleotide charac-
terization by RP-HPLC and MALDI-TOF mass spectro-
metry,19 9 was utilized as a template to explore the
effects of 8-Cl-dAdo substitution on DNA synthesis.20
nucleotide extension opposite dAdo. Interestingly, sin-
gle nucleotide insertion experiments with modified tem-
plate 9 (X=8-Cl-dAdo, Fig. 1, bottom panel) yielded
similar results. TTP was the nucleotide most efficiently
inserted opposite 8-Cl-dAdo, which suggests that the
coding potential of the adenine base was not altered
significantly by chlorine substitution at C-8. Although
both dAdo and 8-Cl-dAdo coded for TTP insertion,
analysis of the extended products indicated that KFꢀ
nucleotide insertion was more efficient when the tem-
plate nucleotide was dAdo compared with 8-Cl-dAdo:
template 8 (X=dAdo) showed 40% primer extension
with TTP after 30 min, whereas modified template 9
(X=8-Cl-dAdo) yielded only 26% extension product
opposite 8-Cl-dAdo over the same time course. These
results indicate that 8-Cl-dAdo shows the same nucleo-
tide insertion preferences as dAdo, but exhibits lower
incorporation efficiencies.
Single nucleotide insertion experiments with KFꢀ DNA
polymerase were conducted to determine the nucleotide
insertion preference opposite 8-Cl-dAdo compared with
the natural dAdo nucleotide. DNA primer–template
.
.
complexes (7 8 or 7 9, Scheme 1B) were incubated with
KFꢀ in the presence of a single deoxynucleoside tri-
phosphate (dNTP) under standard conditions optimized
for single insertion.21 Aliquots were removed from the
extension reactions at specific times, and the products
were analyzed by denaturing polyacrylamide gel electro-
phoresis (dPAGE).22 Results for experiments that tested
the pairing preferences of dA and 8-Cl-dA bases in
DNA templates are shown in Figure 1. The control
DNA template 8 (X=dAdo, Fig. 1, top panel) was
shown to insert exclusively TTP, the normal pairing
partner of dAdo. The three natural triphosphates
(dATP, dCTP and dGTP) were not substrates for single
Figure 1. Gel electrophoretic analysis of single nucleotide insertion by
KFꢀ. Time points at 0, 15, and 30 min for reaction of 50–32P radio-
labeled DNA primer 7 with control template 8 (top) or 8-Cl-dAdo
template 9 (bottom) are shown for each dNTP.

L. S. Chen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1509–1512
1511
Single nucleotide insertion experiments with template 9
showed that 8-Cl-dAdo is recognized by KFꢀ as deoxy-
adenosine. However, as demonstrated by Kool and
co-workers,21 the coding potential for an unnatural base
analogue in a DNA template may not be the same as
the pairing preference for the modified base as a dNTP
analogue. Thus, while 8-Cl-dAdo codes for TTP, mis-
incorporation of 8-Cl-dATP opposite bases other than
thymidine may provide an alternate mechanism for
mutagenesis. To determine the insertion fidelity of 8-Cl-
dATP, single nucleotide insertion experiments were
conducted using KFꢀ and DNA templates containing
natural bases. Either dATP or 8-Cl-dATP (TriLink
Biotechnologies, Inc.) was incubated with a primer/
template complex comprised of primer 7 and a template
containing one of the natural bases (Fig. 1B, X=dAdo
(8), dC (10), dG (11) or T (12)) in the presence of KFꢀ.
Aliquots from the reactions were removed at specific
times, and the products were analyzed by dPAGE.22
Results with triphosphate derivatives were consistent with
those observed with oligonucleotide templates 8 and 9.
8-Cl-dATP was inserted only when T was the coding base
(template 12, X=T);no extension was observed when the
template bases were dAdo, dC, or dG (data not shown).
As illustrated in Figure 2 for the T-containing template
(12), 8-Cl-dATP showed decreased insertion efficiency
relative to dATP. Enzymatic incorporation of 8-Cl-dATP
yielded 22% of extended product after 2 h of incubation
(Fig. 2, lanes 5–8), compared with 41% extension for
the natural triphosphate, dATP (lanes 1–4).
Figure 3. Gel electrophoretic analysis of running-start primer exten-
sion by KFꢀ. Primer extension time points at 0, 5, 10, 15, 30 and 60
min of incubation of 50–32P radiolabeled DNA primer 13 with control
template 8 (lanes 1–6) or 8-Cl-dAdo template 9 (lanes 8–13) are pro-
vided. Radiolabeled DNA standards were included in lane 7 (40 nt)
and lane 14 (26 nt).
completion over one hour (Fig. 3, lanes 1–6) leading to
full-length product, as shown by a 40-nucleotide DNA
marker (lane 7). Incubation of KFꢀ with modified tem-
plate 9 (X=8-Cl-Ado) gave rise to a major termination
product over the same time course (Fig. 3, lanes 8–13).
The termination product co-migrates with the 26
nucleotide DNA, 7, (Scheme 1B;Fig. 3 , lane 14), indi-
cating that termination occurs at the 8-Cl-dAdo site in
the template. This truncated product suggests that KF
pauses directly opposite the 8-Cl-dAdo site, resulting in
termination of DNA synthesis.
Single nucleotide insertion data suggest that 8-Cl-dAdo
derivatives exhibit decreased incorporation efficiencies,
both as a template nucleotide and as a nucleoside tri-
phosphate. The impaired insertion efficiency opposite
8-Cl-dAdo nucleotides may stall DNA polymerase and
impede further enzymatic DNA synthesis. To assess the
ability of KFꢀ to efficiently synthesize DNA past 8-Cl-
dAdo sites, a series of ‘running-start’ primer extension
studies were performed. These experiments utilized a
12-nucleotide primer (13, Scheme 1B), with either tem-
plate 8 or 9, and all four natural dNTPs. Template 8 or
9 places the dAdo/8-Cl-dAdo site 14 nucleotides down-
stream from the primer terminus, which requires that
the polymerase advance on the template prior to inser-
tion opposite dAdo/8-Cl-dAdo. The results for running-
start extension with KFꢀ in the presence of the four
natural dNTPs are given in Figure 3.23 Primer extension
on the control template 8 (X=dAdo) proceeded to
These studies demonstrate that 8-Cl-dAdo preferentially
pairs with thymidine bases both as a template base and
as an incoming nucleoside triphosphate derivative.
Thus, chlorine substitution at C-8 does not significantly
alter the coding ability of deoxyadenosine. The results
obtained with 8-Cl-dAdo show similarities to other
modified adenosine derivatives. The halogenated
nucleoside analogue, 2-chloro-20-deoxyadenosine (2-Cl-
dAdo or cladribine), currently is used for hairy cell
leukemia treatment.24 In contrast with 8-Cl-dAdo, 2-Cl-
dAdo is chlorinated at C-2, on the non-Watson–Crick
pairing face of adenine. In vitro experiments have
shown that 2-Cl-dATP is inserted with fidelity opposite
thymidine by bacterial polymerases.25 Other studies
have shown that DNA templates containing 2-Cl-dAdo
are transcribed with high fidelity.26 Thus, chlorine-sub-
stitution at the adenine C-2 position does not appear to
affect the coding potential of dAdo. In addition, other
studies have demonstrated that 8-oxo-deoxyadenosine
(8-oxo-dAdo) does not mispair during enzymatic DNA
synthesis.27,28 The similarity in the pairing preferences
of 8-Cl-dAdo and 8-oxo-dAdo offer a striking contrast
to the oxidative DNA damage lesion, 8-oxo-dG, which
shows both standard dC incorporation and mutagenic
dAdo insertion.12
Figure 2. Gel electrophoretic analysis of nucleotide insertion by KFꢀ
opposite thymidine. Products of extension after 0, 30, 60 and 120 min
of incubation of 50–32P radiolabeled DNA primer 7 and template 12
with dATP (lanes 1–4) or 8-Cl-dATP (lanes 5–8) are shown.

1512
L. S. Chen et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1509–1512
This work also has illustrated that KFꢀ mediated DNA
synthesis undergoes premature chain termination at
sites of 8-Cl-dAdo in the template strand. Thus,
although 8-Cl-dAdo pairs normally with thymidine,
chain termination at 8-Cl-dAdo sites may have bio-
chemical consequences. For example, decreased DNA
replication efficiency may affect cell division and has
been shown to contribute to the cytotoxic effects of
nucleoside antimetabolites.29 Additional studies with
8-Cl-Ado and 8-Cl-dAdo will be conducted to assess the
effects of 8-modified derivatives on RNA biochemistry
and will explore the potential of 8-modified nucleotides
for the treatment of human disease.
(400 MHz, CD3CN) d: 9.28 (s, 1H, NH), 8.52 (s, 1H, H2), 7.98
(d, 1H, J=7.6 Hz, Bza), 7.63 (dd, 2H, J=7.2 Hz, J=7.2 Hz,
Bzb), 7.53 (t, 2H, J=7.6 Hz, Bzc), 7.17 (m, 9H, DMT, ꢀOPh),
6.74 (m, 4H, DMT), 6.42 (m, 1H, H10), 5.19 (m, 1H, H20), 4.19
(q, 1H, J=5.2 Hz, 30OH), 3.72 (d, 6H, J=4.0 Hz, –OCH3),
3.70-3.58 (m, 1H, H30), 3.34 (m, 1H, H40), 3.19 (m, 2H, H50),
2.54 (m, 1H, ꢀOCH2CH2CN), 2.32 (m, 1H, ꢀOCH2CH2CN),
2.17 (s, 1H,-NH(CH3)2), 1.19 (t, 12H, J=6.4 Hz, –NH(CH3)2).
31P NMR (CD3CN) d: 148.6.
17. Caruthers, M. H.;Barone, A. D.;Beaucage, S. L.;Dodds,
D. R.;Fisher, E. F.;McBride, L. J.;Matteucci, M.;Stabinsky,
Z.;Tang, J. Y. Methods Enzymol. 1987, 154, 287.
18. Dey, S.;Sheppard, T. L. Org. Lett. 2001, 3, 3983.
Unmodified oligonucleotides 7, 8, 10–13 were synthesized by
Integrated DNA Technologies, purified using standard meth-
ods, and characterized by MALDI-TOF MS.
19. Characterization of oligonucleotide 9. MALDI-TOF MS
(m/z): [M]ꢀ calculated, 12261.4;found, 12261.5.
Acknowledgements
20. Stability of 8-Cl-dAdo DNA oligonucleotide. Chlorine
substitution at C-8 enhances the rate of acid-catalyzed depur-
ination of 8-Cl-dAdo (J. Am. Chem. Soc. 1978, 100, 7620.)
However, after DNA synthesis and purification, no abasic site
formation was observed either by MALDI-TOF MS or by
radiolabeling and PAGE analysis of oligonucleotides contain-
ing 8-Cl-dAdo.
We thank our collaborators Dr. Steven Rosen, Dr.
Nancy Krett, and Dr. Varsha Gandhi for insightful
discussions. We acknowledge the Keck Biophysics
Facility and the Analytical Services Laboratory of
Northwestern University, and the University of Illinois,
Urbana-Champaign Mass Spectrometry Laboratory.
This work was supported by the National Institutes of
Health (RO1 CA85915-02) and the Leukemia and
Lymphoma Society (6505-00).
21. Morales, J. C.;Kool, E. T. J. Am. Chem. Soc. 2000, 122,
1001.
22. Single nucleotide insertion with KFꢀ. Radiolabeled
(50–32P) primer 7 was annealed to the appropriate template in
a solution containing 100 mM Tris–HCl (pH 7.5), 20 mM
MgCl2, 2 mM DTT and 0.1 mg/mL BSA. A solution contain-
ing a single dNTP in the same reaction buffer was added to the
annealed primer, and primer extension was initiated by adding
KFꢀ. Aliquots were removed from the reaction at specific
times, and the products were analyzed by 20% dPAGE. Final
extension conditions were 4 mM primer/template duplex, 50
nM KFꢀ, 10 mM dNTP, 150 mM Tris–HCl (pH 7.5), 20 mM
MgCl2, 3 mM 2-mercaptoethanol, 1 mM DTT and 0.05 mg/
mL BSA.
References and Notes
1. Mitsuya, H.;Yarchoan, R.;Broder, S. Science 1990, 249, 1533.
2. Plunkett, W.;Gandhi, V. Hematol. Cell Ther. 1996, 38,
S67.
3. Gandhi, V.;Ayres, M.;Halgren, R. G.;Krett, N. L.;
Newman, R. A.;Rosen, S. T. Cancer Res. 2001, 61, 5474.
4. Halgren, R. G.;Traynor, A. E.;Pillay, S.;Zell, J. L.;Heller,
K. F.;Krett, N. L.;Rosen, S. T. Blood 1998, 92, 2893.
5. Chen, L. S.;Sheppard, T. L. Nucleosides, Nucleotides
Nucleic Acids 2002, 21, 599.
23. Running start primer extensions with KFꢀ. Radiolabeled
(50–32P) primer 13 was annealed to the appropriate template in
a solution containing 100 mM Tris–HCl (pH 7.5), 20 mM
MgCl2, 2 mM DTT and 0.1 mg/mL BSA. Primer extension
was initiated by adding KFꢀ to the annealed duplex, followed
by a dNTP solution in the same buffer. Aliquots were removed
from the reaction at 30, 60, and 120 mins, and the products
were analyzed by 20% dPAGE. Final extension conditions
were 200 nM primer/template duplex, 1 nM KFꢀ, 20 mM each
dNTP (dATP, dCTP, dGTP, TTP), 150 mM Tris–HCl (pH
7.5), 20 mM MgCl2, 3 mM 2-mercaptoethanol, 1 mM DTT
and 0.05 mg/mL BSA.
6. Tavale, S. S.;Sobell, H. M. J. Mol. Biol. 1970, 48, 109.
7. Uesugi, S.;Nagura, T.;Ohtsuka, E.;Ikehara, M.
Pharm. Bull. 1976, 24, 1884.
Chem.
8. Uesugi, S.;Shida, T.;Ikehara, M. Biochemistry 1982, 21,
3400.
9. Eason, R. G.;Burkhardt, D. M.;Phillips, S. J.;Smith,
D. P.;David, S. S. Nucleic Acids Res. 1996, 24, 890.
10. Ikehara, M.;Uesugi, S.;Yoshida, K. Biochemistry 1972,
11, 830.
24. Piro, L. D.;Carrera, C. J.;Carson, D. A.;Beutler, E.
N. Engl. J. Med. 1998, 322, 1117.
11. Chen, L. S.;Sheppard, T. L., unpublished observation.
12. Shibutani, S.;Takeshita, M.;Grollman, A. P.
1991, 349, 431.
13. Ogilvie, K. K. Can. J. Chem. 1973, 51, 3799.
Nature
25. Hentosh, P.;McCastlain, J. C.;Blakley, R. L. Biochem-
istry 1991, 30, 547.
26. Hentosh, P.;Tibudan, M. Mol. Pharmacol. 1995, 48, 897.
27. Duarte, V.;Muller, J. G.;Burrows, C. J. Nucleic Acids
Res. 1999, 27, 496.
28. Shibutani, S.;Bodepudi, V.;Johnson, F.;Grollman, A. P.
Biochemistry 1993, 32, 4615.
29. Huang, M. C.;Ashmum, R. A.;Avery, T. L.;Kuehl, M.;
Blakley, R. L. Cancer Res. 1986, 46, 2362.
14. Hayakawa, H.;Tanaka, H.;Haraguchi, K.;Mayumi, M.;
Nucleosides
Nakajima, M.;Sakamaki, T.;Miyasaka, T.
Nucleotides 1988, 7, 121.
15. Ti, G. S.;Gaffney, B. L.;Jones, R. A. J. Am. Chem. Soc.
1982, 104, 1316.
16. Characterization of phosphoramidite 6. 1H NMR