Synthesis and biological activity of a bivalent nucleotide inhibitor of ribonucleotide reductase

Article abstract of DOI:10.1016/S0960-894X(00)00481-9

A novel nucleotide inhibitor (ADP-S-HBES-S-dGTP) of mouse ribonucleotide reductase was designed to span the active site and the allosteric specificity site of the enzyme. The inhibitor contains ADP and dGTP moieties which are linked by 1,6-hexane-(bis-ethylenesulfone), and has a K(i) value of 12 μM. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00481-9

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2387±2389
Synthesis and Biological Activity of a Bivalent Nucleotide
Inhibitor of Ribonucleotide Reductase
Xu Wu and Barry S. Cooperman*
Department of Chemistry, University of Pennsylvania, Philadelphia, PA 19104-6323, USA
Received 21 June 2000; accepted 17 August 2000
AbstractÐA novel nucleotide inhibitor (ADP-S-HBES-S-dGTP) of mouse ribonucleotide reductase was designed to span the active
site and the allosteric speci®city site of the enzyme. The inhibitor contains ADP and dGTP moieties which are linked by 1,6-hexane-
(bis-ethylenesulfone), and has a Ki value of 12 mM. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
receptors via a chelating e€ect.6,7 The proximity of the
active site and allosteric speci®city site in RR led us to
design a potential bivalent nucleotide inhibitor for this
enzyme, 7, which contains ADP and dGTP moieties
linked via their 8-positions by 1,6-hexane-(bis-ethylene-
sulfone) (HBES). The length of this spacer is 14.7 A.
Ribonucleotide reductase (RR) catalyzes the reduction
of ribonucleotides to 20-deoxyribonucleotides, the rate
determining step in de novo DNA synthesis.1,2 It is thus
an important target for antiviral and cancer chemo-
therapeutic agents.3
Mammalian RR, a Class I enzyme, has two subunits,
mR1 and mR2. The active site and allosteric regulation
sites are located in the mR1 subunit, while a tyrosyl
radical, required for ecient substrate turnover, is loca-
ted in the mR2 subunit. The minimal functional form of
the enzyme is the heterodimer R12R22. In Escherichia
coli R1, which is homologous to mR1 (Pender et al.,
submitted for publication), the active site and allosteric
speci®city site lie near the R12 dimer interface, such that
the straight-line distance between GDP in the active site
in one R1 monomer and dTTP in the speci®city site in
the other, measured from the centers of the bases, is 14
A4 (Fig. 1). This proximity may account for the inter-
esting functional interplay between the two sites that is
related to the maintenance of the four cellular dNTP
pools. Thus, for example, binding of dGTP in the spe-
ci®city site induces ADP reductase, exclusively, while
binding of dTTP induces GDP reductase.5
Synthesis
The synthesis of 7 was accomplished by Michael addi-
tion reaction of the vinyl sulfone thioether 3 with 8-
mercapto-dGTP 6 (Scheme 2). Compound 3 was pre-
pared in three steps from ADP sodium salt (Scheme 1),
using published protocols8,9 for the synthesis of 8-
bromo-ADP 1 and 8-mercapto-ADP 2. Compound 2
(1.3 mg, 3 mmol, in 500 mL water) was added to HBVS
(1,6-hexane-(bis-vinylsulfone), 11 mg, 45 mmol, Pierce
Chemical Company) in 50 mL of DMSO. The pH value
of the reaction mixture was adjusted to pH 7.5±8 with
1 M ammonia water. The reaction was maintained at
37 ꢀC for 24 h, yielding 3 (85%) and small amounts of
the symmetrical diadduct 4 (5%).
8-Bromo-dGTP 5 was prepared from dGTP sodium salt
according to the published protocol.10 Preparation of 6
from 5 was complicated by hydrolysis of the triphos-
phate moieties of both 5 and 6 under the conditions
normally used for replacement of the 8-Br group by an
8-SH group on a guanosine nucleus.11 Nevertheless,
careful optimization of Na2S2O3 concentration, tem-
perature, and reaction time a€orded an acceptable yield
of 6 in one step from 5, avoiding the necessity of a longer
synthetic route, with anticipated lower overall yield, via
conversion of 8-Br-dGMP to 8-mercapto-dGMP and
Bivalent ligands have recently attracted attention in the
design of chemotherapeutic agents because of their
potential for increased binding anity to biological
*Corresponding author. Tel.: +1-215-898-6330; fax: +1-215-898-
2037; e-mail: cooprman@pobox.upenn.edu
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00481-9

2388
X. Wu, B. S. Cooperman / Bioorg. Med. Chem. Lett. 10 (2000) 2387±2389
two phosphorylation reactions. One mmole of 8-Br-
dGTP was dissolved into 700 mL of butanol:ethanol:
water (4:2:1), and 200 mmol of sodium thiosulfate was
added.12 The reaction mixture was kept at 100 ꢀC for
25 min to give 6 in 40% yield, determined following
HPLC puri®cation. Compounds 3 (0.6 mmol) and 6
(0.06 mmol) were combined in 10 mL of ammonia water
(pH 8). Incubation at 37 ꢀC for 36 h a€orded 7 in 60%
yield. Excess 3 was reisolated for repeat use in syn-
thesizing additional 7.
All new compounds were puri®ed to homogeneity by
RP-HPLC (Synchropak C18, Micra Scienti®c Inc.) using
a 0 to 5% acetonitrile in 20min gradient containing
0.01M triethylammonium acetate (pH 4.6). The ¯ow rate
was 1 mL/min. Under these conditions, 6 (elution time,
16 min) was baseline resolved from 8-mercapto-dGDP
(13 min), 8-Br-dGDP (18 min) and 5 (20 min). Each
compound was characterized by UV spectrum,13,14 total
phosphate analysis15 and electrospray ionization mass
spectrometry (ESI-MS). The structure of 7 was con-
®rmed by MALDI-TOF mass spectrometry.
Figure 1. Structure of Escherichia coli R1 dimer with dTTP at the
speci®city site and GDP at the active site.4 The straight line shows the
distance between the base-ring centers of nucleotides in these two sites.
Biological Activity Evaluation
Compounds were assayed for as inhibitors using a
standard RR activity assay.16 ADP derivatives were
tested for inhibition of dGTP-activated ADP reduction,
and dGTP derivatives and 7 were tested for the inhibi-
tion of dTTP-activated GDP reduction. Km values and
Ki values were calculated using Lineweaver±Burke and
Dixon plots, respectively, giving the results summarized
in Table 1. Given the low turnover number of RR (0.18
and 0.28 s 1 for ADP and GDP, respectively),17 it is not
unreasonable to assume that Km and Kd values for sub-
strate are essentially equivalent. With this assumption,
the results show that 8-alkylthio substitution at the
ADP position is well tolerated in the active site, since
both 3 and 4, which is not expected to bind bivalently,
have slightly higher anity than ADP itself. On the
other hand, 1 and 2 bind more weakly, perhaps re¯ect-
ing unfavorable steric or electronic e€ects. The allosteric
speci®city site is somewhat more sensitive to 8-alkylthio
substitution. The 8-mercapto-dGTP mono adduct with
HBVS, 8, synthesized from 6 in a manner analogous to
Scheme 1. Synthesis of 8-substituted ADP derivatives.
Table 1. Ki values
Nucleotide
Ki (mM)
ADP
34Æ4a
77Æ3
8-Bromo-ADP 1
8-Mercapto-ADP 2
ADP-8-S-HESVS 3
ADP-8-S-HBES-S-8-ADP 4
dGTP
8-Bromo-dGTP 5
8-Mercapto-dGTP 6
ADP-8-S-HEVS-S-8-dGTP 7
dGTP-8-S-HESVS 8
158Æ8
22Æ3
19Æ2
1.8Æ0.1b
56Æ4
13Æ2
12Æ2
9Æ1
aKm of ADP as substrate.
bKd of dGTP binding to speci®city site (ref 17).
Scheme 2.

X. Wu, B. S. Cooperman / Bioorg. Med. Chem. Lett. 10 (2000) 2387±2389
2389
Figure 2. Lineweaver±Burke plots demonstrating the e€ect of added 7 on GDP reductase, measured as a function of either dTTP concentration
(speci®city activator, part a), or GDP concentration (substrate, part b). No inhibitor (&); +20 mM of 7 (^).
2 to 3 conversion but with longer reaction time (36 h),
has about a ®ve-fold lower anity than dGTP itself. 8-
Mercapto-dGTP 6, binds with approximately the same
lower anity, while the 8-Br derivative 5 has markedly
lower anity.
References
1. Ericsson, S.; Sjoberg, B. M. In Allosteric Enzymes; Herve,
G., Ed.; CRC: Boca Raton, 1989; pp 189±215.
2. Thelander, L.; Reichard, P. Annu. Rev. Biochem. 1979, 48,
133.
3. Breitler, J. J.; Smith, R. V.; Haynes, H.; Silver, C. E.;
Quish, A.; Kotz, T.; Serrano, M.; Brook, A.; Wadler, S.
Invest. New Drugs 1998, 16, 161.
4. Eriksson, M.; Uhlin, U.; Ramaswamy, S.; Ekberg, M.;
Regnstrom, K.; Sjoberg, B. M.; Eklund, H. Structure 1997, 5,
1077.
5. Jordan, A.; Reichard, P. Annu. Rev. Biochem. 1998, 67, 71.
6. Mammen, M.; Choi, S.-K.; Whitesides, G. M. Angew.
Chem., Int. Ed. 1998, 37, 2754.
Disappointingly, the designed bivalent inhibitor 7 has
approximately the same anity as 6 and 8, suggesting
that this compound binds only to the allosteric site and
not to the active site. This suggestion was con®rmed by
Lineweaver±Burke plots (Fig. 2) showing that 7 is com-
petitive with dTTP for the speci®city site binding, while
it is a noncompetitive inhibitor toward GDP.
7. Kramer, R. H.; Karpen, J. W. Nature 1998, 395, 710.
8. Ikehara, M.; Uesugi, S. Chem. Pharm. Bull. 1969, 17, 348.
9. DeCamp, D. L.; Lim, S.; Colman, R. F. Biochemistry 1988,
27, 7651.
10. Ikehara, M.; Tazawa, I.; Fukui, T. Chem. Pharm. Bull.
1969, 17, 1091.
11. Brown, R. L.; Bert, R. J.; Evans, F. E.; Karpen, J. W.
Biochemistry 1993, 32, 10089.
12. Jankowski, A. L.; Wise, D. S.; Townsend, L. B. Nucleo-
sides Nucleotides 1989, 8, 339.
Nevertheless, the present results provide grounds for
optimism that high anity bivalent inhibitors of RR are
attainable. We believe that the failure of 7 to bind biva-
lently re¯ects a nonoptimal tether, since the results in
Table 1 show that 8-alkylthio substitution is reasonably
well-tolerated at both targeted sites. Future experiments
will focus on tether optimization, via straightforward
modi®cations of the synthetic route for 7.
13. Holmes, R. E.; Robins, R. K. J. Am. Chem. Soc. 1964, 86,
1242.
14. Lin, T.-S.; Cheng, J.-C.; Ishiguro, K.; Sartorelli, A. C. J.
Med. Chem. 1985, 28, 1194.
Acknowledgements
15. Murphy, J. H.; Trapane, T. L. Anal. Biochem. 1996, 240, 273.
16. Moore, E. C.; Peterson, D.; Yang, L. Y.; Yeung, C. Y.;
Ne€, N. F. Biochemistry 1974, 13, 2904.
17. Scott, C. P. PhD Thesis, University of Pennsylvania,
Philadelphia, 1997, p 158.
We would like to thank Ossama Kashlan and Bari
Pender for providing mR1 and mR2 for enzyme assay.
Financial support was provided by the NIH grant
CA58567.