Nucleoside analogues as highly potent and selective inhibitors of herpes simplex virus thymidine kinase

Article abstract of DOI:10.1016/S0960-894X(01)00256-6

A series of carboxamide derivatives of 5′-amino-2′,5′-dideoxy-5-ethyluridine has been prepared as inhibitors of HSV-TK (herpes simplex virus thymidine kinase). The most potent compounds were derived from xanthene, thioxanthene and dihydroanthracene carboxylic acids. The lead compounds show subnanomolar IC₅₀ values against HSV TKs.

Full text of DOI:10.1016/S0960-894X(01)00256-6

Bioorganic & Medicinal Chemistry Letters 11 (2001) 1655–1658
Nucleoside Analogues as Highly Potent and Selective Inhibitors of
Herpes Simplex Virus Thymidine Kinase
Joseph A. Martin,a,* Robert W. Lambert,a John H. Merrett,a Kevin E. B. Parkes,a
Gareth J. Thomas,a Stewart J. Baker,a David J. Bushnell,a Julie E. Cansfield,a
a
Stephen J. Dunsdon,a AndrewC. Freeman, Richard A. Hopkins,a Ian R. Johns,a
Elizabeth Keech,a Heather Simmonite,a Andrea Walmsley,a Philippe Wong Kai-Inb
and Mark Hollandb
aDepartment of Medicinal Chemistry, Roche Discovery Welwyn, Broadwater Road, Welwyn Garden City, Herts AL7 3AY, UK
bDepartment of Antiviral Biology, Roche Discovery Welwyn, Broadwater Road, Welwyn Garden City, Herts AL7 3AY, UK
Received 4 January 2001; revised 9 January 2001; accepted 9 April 2001
Abstract—A series of carboxamide derivatives of 50-amino-20,50-dideoxy-5-ethyluridine has been prepared as inhibitors of HSV-TK
(herpes simplex virus thymidine kinase). The most potent compounds were derived from xanthene, thioxanthene and dihydroan-
thracene carboxylic acids. The lead compounds showsubnanomolar IC values against HSV TKs. # 2001 Elsevier Science Ltd.
50
All rights reserved.
Introduction
TK is not essential for virus replication in rapidly
dividing cells where sufficient thymidine triphosphate
for viral DNA synthesis is provided by cellular meta-
bolism, whereas it is required in non-dividing nerve
cells4 where little or no cellular DNA synthesis occurs.
Herpes simplex virus (HSV) is one of the most common
viral infections in man.1 Type 1 (HSV-1) primarily cau-
ses orofacial lesions, whereas type 2 (HSV-2) is mainly
but not exclusively the causative agent of genital infec-
tions, which continue to increase worldwide.2 Following
the primary infection, lesions resolve in around 10 days
but unfortunately by this stage the virus has migrated to
neuronal tissue, where it resides in a latent state. Upon
reactivation, the virus causes a recurrence of the initial
infection, which in many patients can be particularly
severe resulting in a profound effect on their quality of
life. Current therapy for the control of recurrent herpes
infections involves the use of antiviral drugs such as
Acyclovir, which prevents replication of virus after
reactivation has occurred. An alternative strategy would
be to prevent the reactivation of latent virus. Although
the precise mechanisms involved in reactivation of
latent virus are not fully understood, the virally encoded
enzyme thymidine kinase (TK) has been implicated.3 TK
catalyses the phosphorylation of thymidine to the mono-
phosphate, which is a precursor for DNA synthesis. HSV
TK deficient mutants of HSV do growin cell culture but
are defective in primary neuronal cultures. Further-
more, TK deficient mutants can cause primary infection
and establish latency in animal models but do not reac-
tivate from the latent state.5 These findings have led to
considerable interest in the search for potent and selec-
tive inhibitors of HSV TK as a novel approach to pre-
vent the reactivation of virus from latency.
We and others have previously published6ꢀ17 on the
design and biological properties of nucleoside analogues
as inhibitors of HSV TK. In our initial studies,6 we
based our design of HSV TK inhibitors on isosteric and
isoelectronic analogues of thymidine monophosphate,
*Corresponding author. Tel.: +44-1707-366452; fax: +44-1707-
373504; e-mail: joseph.martin@roche.com
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00256-6

1656
J. A. Martin et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1655–1658
the product of the biochemical process catalysed by TK.
Therefore, we synthesised phosphate, phosphonate and
sulphonate esters of thymidine as well as a number of
amide and sulphonamide derivatives of 50-amino-50-
deoxythymidine. The most potent of these was 50-phenyl-
acetamido-50-deoxythymidine. We then made the corre-
sponding 5-ethyl analogue 2 on the grounds that 20-deoxy-
5-ethyluridine is known to have a much higher affinity for
HSV TK than the cellular enzyme.18 As expected, inhibi-
tors derived from this nucleoside did showhigh selectivity
for the viral enzymes. Furthermore, 20-deoxy-5-ethylur-
idines were an order of magnitude more potent than the
corresponding thymidine analogues. More recently, we
reported19,20 a preliminary account of further studies
which culminated in the synthesis, biochemical properties
and activity in animal models of the phenoxyacetamide
1, a highly potent and very selective inhibitor of HSV
TK. We nowwish to report additional studies on
another series of highly potent inhibitors of HSV TK.
methyl 2-iodobenzoate 36 gave the diphenyl ether 37.
Hydrolysis of the ester and cyclisation to the xanthone
38 was followed by reduction to the xanthene 39, which
was then lithiated and carboxylated. Analogous strate-
gies were applied to the synthesis of 4,5-dichloro-
xanthene-9-carboxylic acid and 4,5-dimethylxanthene-9-
carboxylic acid. The 10,10-bis(trifluoromethyl)-9-10-
dihydro-9-anthracene carboxylic acid was prepared
from the lactone25 41 through reaction with phenyl
magnesium bromide and cyclisation of the product 42
to the anthrone 43.
In this case, reduction of the anthrone 43 with diborane
proceeded only as far as the anthrol, catalytic hydro-
genation of which gave the desired dihydroanthracene
44 (Scheme 3).
Chemistry
The carboxamides 2–17, 19, 20, 22–29, 31 and 32 were
obtained by acylation of the previously described19 50-
amino-20,50-dideoxyuridines 33 and 34 by reaction with
the appropriate acid chloride under Schotten–Baumann
conditions (method A) or through coupling of the acid
with the amine in the presence of 1,3-dicyclohexyl car-
bodiimide (method B) or N-ethyl-N-(3-dimethylamino-
propyl)carbodiimide (method C) as shown in Scheme 1.
Scheme 3. Reagents: (a) PhMgBr, Et2O; (b) PPA; (c) BH3, THF; (d)
H2, Pd–C.
The sulphones 18, 21 and 30 were obtained by oxidation
of the corresponding thioxanthene derivatives with m-
chloroperbenzoic acid as shown in Scheme 4.
Scheme 1. Reagents: (a) RCOCl, NaOH, H2O (method A); (b)
RCO2H, DCCI, HOBT, DMF (method B); (c) RCO2H, EDC, HOBT,
DMF (method C).
Newcarboxylic acids were synthesised through lithia-
tion and carboxylation of suitably substituted xan-
thenes, thioxanthenes or 9,10-dihydroanthracenes.
Scheme 4. Reagents: (a) MCPBA, DCM.
Results and Discussion
The compounds described herein were tested in assays
following previously described procedures.19
In the phenylacetamide series, the effect of introducing a
substituent, particularly in the ortho-position, resulted
in increased potency (Table 1). Whilst the 2-fluoro-
phenylacetamide 3 was only slightly more potent than
2, a diverse range of other 2-substituted phenylacet-
amides 4–8 showed higher and rather similar potencies
suggesting that the principle influence of the 2-sub-
stituent was steric in nature. We reasoned that a 2,6-
disubstitution pattern would restrict the rotational free-
dom of the aryl ring and could enhance potency further.
.
Scheme 2. Reagents: (a) CuO, DMA; (b) NaOH; (c) H2SO4; (d) BH3
THF; (e) LDA, THF; (f) CO2.
If these were not commercially available, they were
prepared by published procedures21ꢀ23 or by a general
route illustrated in the synthesis of the 4-trifluoro-
methylxanthene carboxylic acid24 40 (Scheme 2). Thus,
Ullmann coupling of the trifluoromethylphenol 35 with

J. A. Martin et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1655–1658
Table 1. IC50 valuesa of phenylacetamides 3–11
1657
derivative 13, which has a greater degree of conforma-
tional mobility was considerably less potent. Interest-
ingly, the planar anthracene carboxamide 14 was a
relatively weak inhibitor, whilst the heterocycles 15–18
were markedly more potent. In our earlier studies on
inhibitors of HSV TK we noted that the introduction of a
20-fluorine substituent in the arabino configuration in 50-
amino-20,30-dideoxy-5-ethyluridine carboxamides resulted
in a modest increase in potency and good oral bioavail-
ability. Therefore, we studied 20-fluoro analogues of the
most potent tricyclic carboxamides 15, 17 and 18. As
expected, these inhibitors 19–21 showed at least
equivalent potency to the parent compounds (Table 3).
Compound
R1
R2
IC50 (nM)
HSV-1 TK
HSV-2 TK
2
3
4
5
6
7
8
H
F
Cl
Me
CF3
MeO
Ph
H
H
H
H
H
1000
700
100
200
200
300
240
300
140
30
60
90
Table 3. IC50 valuesa of 20-fluoroarabinofuranosides
H
H
70
90
9
10
Cl
Me
Cl
Me
1.0
40
3.3
8.4
aValues are means of three determinations.
This proved to be the case when the 2,6-dichloro and
2,6-dimethyl analogues 9 and 10 were found to be an
order of magnitude more potent than monosubstituted
compounds. However, these compounds were still sig-
nificantly less potent than the phenoxyacetamide 1. In
an attempt to further define the lipophilic sites respon-
sible for binding of the aryl residue of our inhibitors to
the enzymes we were intrigued by the possibility of
designing conformationally restricted analogues of the
2-phenyl-phenylacetamide 8. Thus, introduction of a
direct bond between the two phenyl substituents in
structure 8 and the a-carbon atom resulted in the fluor-
enyl structure 11 (Table 2).
Compound
R1
R2
X
IC50 (nM)
HSV-1 TK
HSV-2 TK
19
20
21
22
23
24
25
26
27
28
29
30
31
32
H
H
H
Cl
Me
CF3
MeO
Ph
Cl
Me
Cl
Cl
H
H
H
H
H
H
H
H
H
Cl
Me
H
H
H
H
O
S
SO2
O
O
O
O
O
O
O
2.1
8.5
0.93
0.85
1.6
1.0
1.7
0.66
0.30
2.2
0.11
0.70
0.11
0.14
0.19
0.14
0.21
0.14
0.10
0.43
0.21
0.09
0.28
0.11
S
1.5
0.27
1.5
SO2
C(Me)2
C(CF3)2
Table 2. IC50 valuesa of compounds 12–19
H
0.19
aValues are means of three determinations.
We then examined the effect of substituents in the het-
erocyclic tricyclic residues. A variety of substituents,
particularly in the 4-position resulted in the highly
potent inhibitors 22–26, 29 and 30. By introducing a
single substituent into the tricyclic moiety, these com-
pounds suffered the disadvantage of being asymmetric.
For synthetic convenience these carboxamides were
prepared using racemic carboxylic acids and the result-
ing products were evaluated as 1:1 mixtures of diaster-
eoisomers. The stereochemical issue was overcome
through the use of symmetrically disubstituted car-
boxylic acids. This is illustrated with the 4,5-dichloro-
and 4,5-dimethylxanthene carboxamides 27 and 28.
Compounds in Table 3 are amongst the most potent
HSV TK inhibitor reported thus far, especially 21, 22,
26, 27, 30 and the symmetrical 10,10-disubstituted-
dihydroanthracene carboxamide 32, all of which are
subnanomolar against both enzymes.
Compound
X
Ring system
IC50 (nM)
HSV-1 TK HSV-2 TK
11
12
13
14
15
16
17
18
Direct bond
CH2
(CH2)2
CH
Fluorene
Dihydroxanthene
Dibenzosuberene
Anthracene
28.0
7.0
5.2
1.2
260.0
1650.0
2.9
23.0
11.5
44.0
420.0
0.13
2.6
1.4
0.13
O
Xanthene
NMe
S
SO2
NMe-acridine
Thioxanthene
Thioxanthene dioxide
0.95
aValues are means of three determinations.
This compound proved to be significantly more potent
than 8 and therefore additional tricyclic derivatives were
examined. The dihydroanthracene 12 was somewhat
more potent than 11, whereas the dibenzosuberene
Having shown that these compounds were potent inhi-
bitors of HSV TK, we wished to demonstrate their

1658
J. A. Martin et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1655–1658
selectivity for the viral enzymes. Therefore, a repre-
sentative selection of the most potent inhibitors was
evaluated against cytosolic TK from both HeLa and
Vero cells. As had been found with our previously pub-
lished inhibitors, no inhibition of cellular TK was mea-
sured at a concentration of 10 mM.
10. Kim, C. U.; Misco, P. F.; Luh, B. Y.; Terry, B.; Bisacchi,
G.; Mansuri, M. M. Bioorg. Med. Chem. Lett. 1993, 3, 1571.
11. Focher, F.; Hildebrand, C.; Freese, S.; Carrocchi, G.;
Noonan, T.; Sangalli, S.; Brown, N.; Spadari, S.; Wright, G. E.
J. Med. Chem. 1988, 31, 1496.
12. Ashton, W. T.; Meurer, R. L.; Tolman, R. L.; Karkas,
J. D.; Liou, R.; Perry, H. C.; Czelusniak, S.; Klein, R. J.
Nucleosides Nucleotides 1989, 8, 1157.
In conclusion, the xanthene, thioxanthene and dihy-
droanthracene carboxamides described herein represent
a newclass of highly potent and selective inhibitors of
HSV TK. A number of these compounds have shown
activity in murine models of HSV infection when dosed
intraperitoneally but were significantly less effective
when given by the oral route. The reason for these dif-
ferences has yet to be resolved.
13. Gambino, J.; Focher, F.; Hilderbrand, C.; Maga, G.;
Noonan, T.; Spadari, S.; Wright, G. E. J. Med. Chem. 1992,
35, 2979.
14. Xu, H.; Maga, G.; Focher, F.; Smith, E. R.; Spadari, S.;
Gambino, J.; Wright, G. E. J. Med. Chem. 1995, 38, 49.
15. Bourne, N.; Bravo, F. J.; Ashton, W. T.; Meurer, L. C.;
Tolman, R. L.; Karkas, J. D.; Stanberry, L. R. Antimicrob.
Agents Chemother. 1992, 36, 2020.
16. Kaufman, H. E.; Varnell, E. D.; Cheng, Y.-C.; Bobek, M.;
Thompson, H. W.; Dutschman, G. E. Antiviral Res. 1991, 16,
227.
17. Lieb, D. A.; Ruffner, K. L.; Hildebrand, C.; Schaffer,
P. A.; Wright, G. E. Antimicrob. Agents Chemother. 1990, 34,
1885.
18. Cheng, Y.-C. Ann. N.Y. Acad. Sci. 1977, 284, 594.
19. Martin, J. A.; Thomas, G. J.; Merrett, J. H.; Lambert,
R. W.; Bushnell, D. J.; Dunsdon, S. J.; Freeman, A. C.; Hop-
kins, R. A.; Johns, I. R.; Keech, E.; Simmonite, H.; Wong
Kai-In, P.; Holland, M. J. Antiviral Chem. Chemother. 1998, 9,
1.
References and Notes
1. Spruance, S. L. In Clinical Management of Herpes Viruses;
Sachs, S. L., Strauss, S. E., Whitley, R. J., Griffiths, P. D.,
Eds.; IOS: Washington, DC, 1995; pp 3–42.
2. Corey, L.; Wald, A. In Clinical Management of Herpes
Viruses; Sachs, S. L., Strauss, S. E., Whitley, R. J., Griffiths, P.
D., Eds.; IOS: Washington, DC, 1995; pp 43–53.
3. Tenser, R. B. Intervirol. 1991, 32, 76.
4. Jamieson, A. T.; Gentry, G. A.; Subak-Sharp, J. H. J. Gen.
Virol. 1974, 24, 465.
20. Watkins, A. M.; Dunford, P. J.; Moffatt, A. M.; Wong
Kai-In, P.; Holland, M. J.; Pole, D. S.; Thomas, G. J.; Martin,
J. A.; Roberts, N. A.; Mulqueen, M. J. Antiviral Chem. Che-
mother. 1998, 9, 9.
21. Jojima, T.; Takeshiba, H.; Konotsune, T. Chem. Pharm.
Bull. 1972, 20, 2191.
5. Coen, D. M.; Kosz-Vnenchak, M.; Jackobsen, J. G.; Lieb,
D. A.; Bogard, C. L.; Schaffer, P. A.; Tyler, K. L.; Knipe,
D. M. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 4736.
6. Martin, J. A.; Duncan, I. D.; Hall, M. J.; Wong Kai-In, P.;
Lambert, R. W.; Thomas, G. J. Nucleosides Nucleotides 1989,
8, 753.
22. Arnett, E. M.; Amarnath, K.; Harvey, N. G.; Cheng, T.-C.
J. Am. Chem. Soc. 1990, 112, 344.
7. Markham, A. E.; Newton, C. R.; Porter, R. A.; Sim, I. S.
Antiviral Res. 1982, 2, 319.
8. Nutter, L. M.; Grill, S. P.; Dutschman, G. E.; Sharma,
R. A.; Bobek, M.; Cheng, Y.-C. Antimicrob. Agents Chemo-
ther. 1987, 31, 368.
23. Leardini, R.; Nanni, D.; Pedulli, G. F.; Tundo, A.;
Zanardi, G.; Forest, E.; Palmieri, P. J. Am. Chem. Soc. 1989,
111, 7723.
24. Tamano, M.; Koketsu, J. Bull. Chem. Soc. Jpn. 1985, 58,
2577.
9. Sim, I. S.; Picton, C.; Cosstick, R.; Jones, S. A.; Walker,
R. T.; Chamiec, A. J. Nucleosides Nucleotides 1988, 7, 129.
25. Reinecke, M. G.; Chen, L. J. Acta. Chem. Scand. 1993, 47,
318.