Anti-HBV nucleotide prodrug analogs: Synthesis, bioreversibility, and cytotoxicity studies

Article abstract of DOI:10.1016/j.bmcl.2005.12.058

Several pronucleotide analogs of the model anti-HBV dinucleotide 3′-dA-U2′OMe have been synthesized and evaluated for stability, bioreversibility and cytotoxicity. These studies have helped identify potential candidates for further evaluation.

Full text of DOI:10.1016/j.bmcl.2005.12.058

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1491–1494
Anti-HBV nucleotide prodrug analogs: Synthesis,
bioreversibility, and cytotoxicity studies
Seetharamaiyer Padmanabhan, John E. Coughlin, Guangrong Zhang,
Cassandra J. Kirk and Radhakrishnan P. Iyer*
Spring Bank Technologies, Inc., 113, Cedar Street, Suite S-7, Milford, MA 01757, USA
Received 24 October 2005; revised 5 December 2005; accepted 12 December 2005
Available online 4 January 2006
Abstract—Several pronucleotide analogs of the model anti-HBV dinucleotide 30-dA-U2 OMe have been synthesized and evaluated for
stability, bioreversibility and cytotoxicity. These studies have helped identify potential candidates for further evaluation.
0
ꢀ 2005 Elsevier Ltd. All rights reserved.
The concept of using chemically modified drugs as pro-
drug analogs is an established paradigm in the pharma-
ceutical development of a number of different drugs.1
The prodrug strategies permit transient modification of
the physicochemical properties of the drug in order to:
(a) improve chemical stability, (b) alter aqueous solubil-
ity, (c) improve bioavailability, (d) target specific tissues,
(e) facilitate synergistic drug combinations, (f) overcome
first-pass metabolic effects, (g) serve as lipophilic carrier
for hydrophilic drugs, and (h) serve as a chemical depot
for sustained drug delivery.2
patient-compliant delivery systems, coupled with ineffi-
cient cellular permeability, represents a significant hur-
dle in the therapeutic advancement of these molecules.
Shorter chain oligonucleotides (less than 8-mers) with
lesser number of charges and smaller molecular weight
compared to 20-mer oligonucleotides represent a prom-
ising class of novel molecules with potential therapeutic
and diagnostic properties.6 Indeed, recent reports7–10
suggest that mono-, di-, tri-, and short chain oligonucle-
otides possess significant biological activity that can be
exploited for therapeutic applications. In this context,
we have recently reported that certain di-, and tri-nucle-
oside phosphorothioate (PS) and phosphoramidate ana-
logs exhibit potent anti-HBV activity in vitro and
in vivo.9 Although dimer and trimer PS analogs are neg-
atively charged small molecules, studies of 35S-labeled
compounds in rats have revealed that these compounds
are not orally bioavailable (unpublished observations).
The lack of oral bioavailability may be due to a number
of factors including: (a) the acidic environment in the
stomach that causes substantial degradation of the
nucleotide, (b) the negative charge on the backbone that
suppresses permeation of the nucleotide through the
intestinal mucosal barrier, and (c) the presence of vari-
ous digestive enzymes in the GI tract that degrade the
compound. Given that both longer and shorter chain
oligonucleotides are not orally bioavailable, it appeared
to us that in the case of smaller nucleotide class of com-
pounds, charge rather than the size of the compound
may be a more important factor in determining bioavail-
ability and that masking the negative charge on the
backbone may potentially provide orally bioavailable
nucleotide compounds.
Prodrug strategies are also being applied in the case of
oligonucleotides (18–30 mers), which are being devel-
oped as potentially novel class of therapeutic agents
using technologies such as aptamers,3 antisense,4a,b ribo-
zymes,4c RNA interference,4d,e,f and immunostimula-
tion.4g Being highly charged, large molecular weight
compounds, oligonucleotides have unfavorable physico-
chemical attributes for cell permeation by passive diffu-
sion. Consequently, the design of prodrug analogs of
oligonucleotides has mainly focused on the partial
masking of their negatively charged backbone by biore-
versible, lipophilic groups. Several such analogs have
been synthesized and bioreversibility has been demon-
strated in vitro.5 However, there are no reports of in vivo
oral bioavailability studies of oligonucleotide prodrugs.
The lack of oral, transdermal, and other non-invasive,
Keywords: Anti-HBV nucleotide; Prodrug; Bioreversibility; Cytotox-
icity; Stability; Dinucleotide.
*
Corresponding author. Tel.: +1 508 473 5993; fax: +1 508 473
6375; e-mail: kiyer@springbanktech.com
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.12.058

1492
S. Padmanabhan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1491–1494
To develop orally bioavailable analogs of di-, and tri-
nucleotides, we carried out the synthesis and evaluation
of a number of S-functionalized, uncharged pronucleo-
tide derivatives of a model dinucleotide. Our design of
pronucleotide derivatives was based upon the ability of
a target enzyme to unmask a latent functionality to re-
veal the parent nucleotide in vivo. This paper summariz-
es our results on the design, synthesis, stability,
bioreversibility, and cytotoxicity studies of various
analogs.
anti-HBV agent adefovir dipivoxil11 are clinically used,
orally bioavailable ester prodrug analogs. In vivo, the
conversion of the prodrug to the parent molecule is
believed to occur via esterase-mediated hydrolysis in
plasma and/or liver, with concomitant liberation of
formaldehyde and carboxylic acid. Guided by this clini-
cal precedence and our previous studies,5b we designed
the pivaloyl ester pronucleotide analog 6a of the anti-
HBV PS-dimer, 30-dA-U2 OMe 5 (Fig. 2). (b) The S-(acyl-
0
oxyaryl)thiophosphate analogs (2). The acyloxyaryl
analogs of daunorubicin, doxorubicin, phosphorodia-
mide mustard, acivicin, and PEG-daunorubicin conju-
gate are well known2b and have been extensively
evaluated in vitro and in vivo. Although a reactive meth-
ylene quinone intermediate is transiently released upon
hydrolysis of these prodrugs, rapid capture of a water
molecule by the semi-quinone intermediate results in
The general structure of dinucleotide derivatives 1–3 and
the expected mechanism for their esterase-mediated con-
version to the parent dimer 4 are depicted in Figure 1
and include: (a) S-(acyloxyalkyl)thiophosphate analogs
(1). The acyloxyalkyl analogs,2b exemplified by the anti-
biotics pivampicillin, bacampicillin, as well as, the
Figure 1. General structure of dinucleotide prodrugs and the proposed mechanism of bioreversibility.
Figure 2. Synthesis of dinucleotide prodrugs and their biological evaluation.

S. Padmanabhan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1491–1494
1493
its conversion to the innocuous benzyl alcohol species
(Fig. 1) thereby minimizing any cellular injury. Using
this rationale, the analogs 6b–d (Fig. 2) were designed
to evaluate their potential as pronucleotide analogs of
5 including the ester analogs 6b, and 6c (which have a
long chain alkoxy group that imparts greater lipophilic-
ity to the molecule), as well as, the amide analog 6d, and
(c) S-alkyl derivatives with a terminal functional group
(3) are designed such that during enzyme-mediated
hydrolytic process, a latent nucleophilic group is uncov-
ered, which is juxtapositioned to attack the electrophilic
carbon a to the thiophosphate moiety resulting in the re-
lease of the parent dinucleotide. This design is similar to
the extensively used ‘cyclization-based’ and ‘trimethyl
lock’ prodrug concepts,2,12 and was conceptualized for
the pronucleotide analogs 6e–j of 5 (Fig. 2).
the analogs may have significant half-life in the Gl tract
that could facilitate oral absorption of the intact pronu-
cleotide. These observations are in contrast to the
behavior of the corresponding pronucleotides of Rp,
Sp TT-PS dimer where significant stereodifferentiation
was noted along with much slower rates of hydrolysis
in serum and PLE.17 It is possible that due to different
0
0
sugar puckering in 2 -OMe-uridine ðC3 -endoÞ com-
0
pared to a thymidine ðC2 -endoÞ, the global conforma-
tion of 6a and 6b may be significantly different from
that corresponding to TT dimer pronucleotides. Conse-
quently, the ester groups in 6a and 6b may be more
favorably poised for attack by the nucleophilic site of
the esterases.
Although the ester analogs 6e–g were hydrolyzed to the
corresponding alcohols, and 6h to the corresponding
carboxylic acid, their subsequent conversion to 5 by
cyclization did not occur. Also, the amide analogs 6d,
6i, and 6j remained unchanged even after incubation
in serum for 24 h.18 Furthermore, all analogs were stable
indefinitely when stored at ꢁ20 ꢂC as a lyophilized pow-
der. We next examined the cytotoxicity profile19 of the
pronucleotide derivatives in different cell lines such as
MDBK, Vero, and HFF. As shown in Figure 2, most
analogs, except for 6c had CC50 > 1000 lM in these cell
lines demonstrating good safety profile for these
compounds.
For our studies, we employed the Rp, Sp mixture of the
phosphorothioate analog 5, which was synthesized in
large scale (6 mmol of nucleoside-loaded controlled-
pore glass (CPG) support) using solid-phase phosphor-
amidite chemistry,13 in cꢁonjunction with a specially
fabricated LOTUS reactor .14 The dA-linked CPG sup-
port was prepared using our recently discovered ultra-
fast functionalization and loading process for solid
supports.14 For the sulfurization of the internucleotidic
dinucleoside phosphite coupled product, a solution of
3H-1,2-benzodithiole-3-one-1,1,-dioxide (0.4 M in dry
CH3CN)15 was employed. Following processing, chro-
matographic purification, and lyophilization, the sodi-
um salt of Rp, Sp 5 (ꢀ60:40 mixture) was obtained
>96% pure, which was characterized by 31P and 1H
NMR. The pronucleotide derivatives 6a–j were synthe-
sized in yields of 50–70% by chemoselective S-alkylation
of Rp, Sp-5 with the corresponding iodo-, or bromo-de-
rivatives 7a–j in aqueous acetone or methanol, followed
by work-up and chromatographic purification. The req-
uisite intermediates 7a–j were synthesized directly from
the corresponding hydroxy compounds16 or by halogen
exchange reaction from the corresponding chloro deriv-
atives (see Fig. 2).5b 31P NMR of each pronucleotide
analog 6a–j showed two peaks in the range of d 28–
34 ppm (characteristic of the thiophosphate triester
moiety) corresponding to a ꢀ55:45 ratio of the Rp, Sp
isomers.
In conclusion, we have synthesized and evaluated a
number of pronucleotides of 5 and the analogs 6a and
6b appear to be promising candidates for further evalu-
ation. Further studies are ongoing and will be reported
in due course.
Acknowledgments
Support of this research from the National Institutes of
Health, under a Research Project Cooperative Agree-
ment Grant Award 5 UO1 AI058270-03, is gratefully
acknowledged. We thank Geoffrey Dorenzo and Chris-
topher Johnson for their excellent technical assistance
in this project.
Evaluation of bioreversibility of the pronucleotides was
carried out in rabbit serum in phosphate buffer at
37 ꢂC. In order to monitor the hydrolytic conversion
of the pronucleotides to the dinucleotide 5, aliquots
of incubate were removed at different time points, pro-
cessed, and analyzed using reversed-phase HPLC. It
was found that the analogs 6a and 6b were readily con-
verted to the parent 5 with half-lives (t1/2) of 60 and
30 min, respectively. Also, complete conversion of 6a
and 6b to the parent 5 occurred in ꢀ3 h. The analogs
6a and 6b were stable for up to 24 h in phosphate buffer
(0.1 M, pH 7.2). Furthermore, there was no evidence of
any significant stereodifferentiation or desulfurization
during the hydrolysis of the Rp, Sp isomers in the mix-
ture. Interestingly, both 6a and 6b were resistant to the
hydrolytic action of pig liver esterase (PLE) and bovine
chymotrypsin (data not shown), thereby suggesting that
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2005.12.058.
References and notes
1. Bundgaard, H. In bio-reversible carriers in drug design. In
Theory and Application; Roche, E. B., Ed.; Pergamon
Press: New York, 1987; p 13.
2. For excellent reviews see (a) Oliyai, R.; Stella, V. J.
Annu. Rev. Pharmacol. Toxicol. 1993, 32, 521; (b)
Papot, S.; Tranoy, I.; Tillequin, F.; Florent, J.-C.;
Gesson, J.-P. Curr. Med. Chem. -Anticancer Agents
2002, 2, 155.

1494
S. Padmanabhan et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1491–1494
3. (a) Tuerk, C.; Gold, L. Science 1990, 249, 505; For a
review: see (b) Jayasena, S. D. Clin. Chem. 1999, 45, 1628.
15. Iyer, R. P.; Regan, J. B.; Egan, W.; Beaucage, S. L. J. Am.
Chem. Soc. 1990, 112, 1253.
4. For reviews see (a) Szymkowski, D. E. Drug Disc Today
1996, 1, 415; (b) Uhlmann, E.; Peyman, A. Chem. Rev.
1990, 90, 543; (c) Uhlenbeck, O. C. Nature 1987, 328, 596;
(d) Zamore, P. D. Science 2002, 296, 1265; (e) Manoharan,
M. Curr. Opin. Chem. Biol. 2004, 8, 570; (f) Iyer, R. P.;
Kuchimanchi, S.; Pandey, R. K. Drugs Future 2003, 28,
51; (g) Uhlmann, E.; Vollmer, J. Curr. Opin. Drug Discov.
Dev. 2003, 6, 204.
5. (a) Polushin, N. N.; Cohen, J. S. Nucleic Acids Res. 1994,
22, 5492; (b) Barber, I.; Rayner, B.; Imbach, J.-L. Bioorg.
Med. Chem. Lett. 1995, 5, 563; (c) Iyer, R. P.; Ho, N.-H.;
Yu, D.; Agrawal, S. Bioorg. Med. Chem. Lett. 1997, 7,
871.
16. (a) Hayat, S.; Rahman, A.-U.; Khan, K. M.; Choudhary,
M. I.; Maharvi, G. M.; Ullah, Z.; Bayer, E. Synth.
`
Commun. 2003, 33, 2531; (b) Fernandez, I.; Garcia, B.;
Munoz, S.; Pedro, R.; de la Salud, R. Synlett 1993, 489.
ˇ
17. Iyer, R. P.; Yu, D.; Agrawal, S. Bioorg. Med. Chem. Lett.
1995, 4, 2471.
18. Interestingly 4-[N-methyl-N-(2,2,2-trifluroacetyr)amino]-
butyl phosphotriester undergoes base-catalyzed conversion
to phosphodiesters via a novel cyclo-desterification reac-
tion and has been employed in oligonucleotide synthesis.
See: Wilk, A.; Grajkowski, A.; Phillips, L. R.; Beaucage, S.
L. J. Org. Chem. 1999, 64, 7515.
19. For bioreversibity studies, 10 lL aliquots of each analog
(2 mg in 100 lL of DMSO) were diluted with 90 lL of
phosphate buffer (0.1 M, pH 7.0) and 100 lL of rabbit
serum. The mixture was incubated at 37 ꢂC in a water
bath. Aliquots were removed at different time points and
diluted with 200 lL of methanol to stop the reaction. The
incubate was then centrifuged, supernatant concentrated
in a speed vac, and diluted with 200 lL of 0.1 M NH4OAc,
and analyzed by reversed-phase HPLC (Waters Instru-
ment equipped with a 600E gradient controller and a 996
photodiode array detector with Millennium software).
X-terra MS C18 2.5 mm, 2.1 · 20 mm column and a
gradient of 100% A to 80% B over 30 min of buffer A
(0.1 M NH4OAc) and buffer B (80:20, CH3CN:NH4OAc)
was employed. Rt for prodrugs ranged from 16 to 18 min,
whereas that of the Rp, Sp dinucleotide 5 was 13.5,
13.8 min. Cytotoxicity assays. Standard MTT assays were
performed in 96-well plates using the Promega CellTiter96
Non-radioactive Cell Proliferation Assay Kit in conjunc-
tion with a 96-well Plate Reader (ThermoMax, Molecular
devices), and MDBK, Vero, and HFF cell lines (obtained
from ATCC). Several controls were employed including
the nucleoside analogs 3TC, AZT, and ddC, as well as,
media without drugs. SDS was used as a positive cytotoxic
control. All pronucleotides were tested in triplicate at
concentrations of 100, 300, and 1000 lM. Following a
24-h incubation of cells with the test substance, the MTT
assay was carried out.
6. Wagner, R. W.; Matteucci, M. D.; Grant, D.; Huang, T.;
Froehler, B. C. Nat. Biotechnol. 1996, 14, 840.
7. Reviewed in Iyer, R. P.; Padmanabahan, S.; Zhang, G.;
Morrey, J. D.; Korba, B. E. Curr. Opin. Pharmacol. 2005,
5, 520.
8. Hannoush, R. N.; Min, K.-L.; Damha, M. J. Nucleic Acids
Res. 2004, 32, 6164.
9. (a) Iyer, R. P.; Jin, Y.; Roland, A.; Morrey, J. D.; Mounir,
S.; Korba, B. Antimicrob. Agents Chemother. 2004, 48,
2199; (b) Iyer, R. P.; Roland, A.; Jin, Y.; Mounir, S.;
Korba, B.; Julander, J. G.; Morrey, J. D. Antimicrob.
Agents Chemother. 2004, 48, 2318.
10. Nair, V. Curr. Pharm. Des. 2003, 31, 2553.
11. Rivkin, A. M. Ann. Pharmacother. 2004, 38, 625.
12. For a review see: (a) Shan, D.; Nicolaou, M. G.;
Borchardt, R. T.; Wang, B. J. Pharm. Sci. 1997, 86, 765;
(b) Borchardt, R. A.; Cohen, L. A. J. Am. Chem. Soc.
1972, 94, 9175; (c) Amsberry, K. L.; Borchardt, R. T.
Pharm. Res. 1991, 8, 323; (d) Greenwald, R. B.; Choe, Y.
H.; Conover, C. D.; Shum, K.; Wu, D.; Royzen, M. J.
Med. Chem. 2000, 43, 475.
13. For a review see Beaucage, S. L.; Iyer, R. P. Tetrahedron
1993, 49, 1925.
14. (a) Padmanabhan, S.; Coughlin, J. E.; Iyer, R. P.
Tetrahedron Lett. 2005, 46, 343; (b) Iyer, R. P.; Coughlin,
J. E.; Padmanabhan, S. Org. Prep. Proc. Intl. 2005, 37,
205.