The monoethyl ester of meconic acid is an active site inhibitor of HCV NS5B RNA-dependent RNA polymerase

Article abstract of DOI:10.1016/j.bmcl.2004.03.087

Screening of the in-house sample collection for compounds with HCV NS5B RNA dependent RNA polymerase inhibition led to the identification of a new lead. Afterwards, we discovered that the screening lead, rather than containing the expected structure 1, was comprised of roughly a 1:1 mixture of meconic acid 2 and its monoethyl ester 3, with all inhibitory potency residing with 3. We propose that this compound shares critical common features for activity with α,γ-diketoacids inhibitors previously discovered by our group. SAR around this molecule will be presented to provide an improved basis for structure-based ligand design.

Full text of DOI:10.1016/j.bmcl.2004.03.087

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3257–3261
The monoethyl ester of meconic acid is an active site inhibitor
of HCV NS5B RNA-dependent RNA polymerase
Paola Pace,* Emanuela Nizi, Barbara Pacini, Silvia Pesci, Victor Matassa,
Raffaele De Francesco, Sergio Altamura and Vincenzo Summa
Department of Medicinal Chemistry and Biochemistry, IRBM-MRL Rome, Via Pontina, Km 30.600, 00040 Pomezia (Rome), Italy
Received 21 January 2004; revised 24 March 2004; accepted 29 March 2004
Abstract—Screening of the in-house sample collection for compounds with HCV NS5B RNA dependent RNA polymerase inhi-
bition led to the identification of a new lead. Afterwards, we discovered that the screening lead, rather than containing the expected
structure 1, was comprised of roughly a 1:1 mixture of meconic acid 2 and its monoethyl ester 3, with all inhibitory potency residing
with 3. We propose that this compound shares critical common features for activity with a,c-diketoacids inhibitors previously
discovered by our group. SAR around this molecule will be presented to provide an improved basis for structure-based ligand
design.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
inconsistent with that obtained from the original sam-
ple. 1H, 13C NMR and mass analyses showed that, while
the second batch had the expected structure 1, the ori-
ginal screening compound was in fact a roughly 1:1
mixture of meconic acid (poppy acid or 3-hydroxy-4-
oxo-4H-pyran-2,6-dicarboxylic acid)4 2 and its mono-
ethyl ester 3. Compound 3 was therefore prepared and
shown to provide an NMR spectrum, which when
combined with the spectrum of commercial meconic
acid, accounted for the observed NMR spectrum of the
first batch of screening compound. The meconic ester
derivative was therefore identified as the active compo-
nent of this mixture, having an IC50 ¼ 2.25 lM against
HCV NS5B polymerase (Fig. 1).
The lack of a highly effective and safe treatment option
for the hepatitis C virus (HCV) highlights the necessity
of developing more efficient means of combating and,
ultimately, curing this viral disease. A primary focus is
currently on finding inhibitors of the NS5B polymerase,
an enzyme absolutely required for replication of the
virus and exhibiting important differences with cellular
polymerases.1;2
Recent discoveries from our laboratories of the biolog-
ical activity of a,c-diketoacids3 as low nanomolar
inhibitors of NS5b HCV polymerase further stimulated
our interest in this area. A new screening of our in-house
sample collection identified another low micromolar
lead as a specific and reversible inhibitor of NS5B HCV
polymerase, competitive with the diketoacid. The com-
pound was inactive at 200 lM against HIV-RT and
poliovirus RNA-dependent RNA polymerase. Surpris-
ingly, a second batch of the screening hit was inactive in
the HCV polymerase assay and provided spectral data
Our preliminary structure–activity studies are summa-
rized in Table 1 and demonstrate the absolute require-
ment of the 2-carboxylic acid functionality.
Decarboxylation or esterification abolish the activity
Keywords: Meconic acid; HCV polymerase inhibitors.
* Corresponding author. Tel.: +39-0691093313; fax: +39-0691093654;
e-mail: paola_pace@merck.com
Figure 1. Screening lead for HCV NS5b Polymerase inhibition (see
text).
Present address: Graffinity Pharmaceutical AG Im, Neuenheiner
Feld 518, 69120 Heidelberg, Germany.
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.03.087

3258
P. Pace et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3257–3261
Table 1. Inhibitory potency of meconic acids derivatives
Compd
R1
R2
IC50 (lM)a
1
CO2Et
CO2H
H
>100
>100
2.25
>100
>100
51
2
CO2H
CO2H
CO2Et
CO2Me
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
3
CO2Et
CO2Et
CO2H
4
5
6
H
7
CONHEt
CONEt2
CH2OEt
CH2OBn
CO2tBu
2
4
8
>100
16.2
>100
36
9
10
11
a See Ref. 5 for polymerase assay protocol. Data are the mean of two
to four independent experiments.
Figure 3. Displacement of 3H-DKA from NS5b HCV polymerase.
(compounds 1 and 4). Altering the position of the
carboxylate with respect to the ester gave inactive 5. Our
previous experience with a,c-diketoacids suggests a
common paradigm for HCV NS5B inhibition based on
the shared presence of a carbonyl, an enolic OH and a
carboxylate functionality. This similarity allows a very
plausible overlay of the two structures with the three
oxygen atoms in a similar spatial arrangement (Fig. 2).
potency is dependent on the type of metal used in the
assay (Mg2þ or Mn2þ, see Table 2).
Although Mg2þ is the actual cofactor in vivo, the highest
inhibitory potency measured in the presence of Mn2þ
allowed us to readily calculate an IC50 for compounds,
like meconic acid 2, whose inhibitory potency was vir-
tually undetectable under standard assays conditions.
Studies by ourselves,6 and other groups,7 showed that in
the NS5B polymerase active site there are two strictly
conserved aspartic and a glutamic acid residues that
chelate Mg2þ ions required for nucleotide incorporation.
Recently, we reported that diketoacids are active site
inhibitors and they are capable of interacting directly
with the metal ions.8 Therefore, some of the binding
energy of the meconic acid derivative too could be de-
rived from metal ion interaction in the active site. This
hypothesis is supported by a mutually exclusive binding
of the compounds to the enzyme, demonstrated by
competition experiments using tritiated diketoacid
(DKA) as probe (Fig. 3). We measured the ability of our
lead to compete for 50% of the bound [3H]-DKA
(DC50). For 3 we obtained a DC50 ¼ 1.2and 0.3 lM for
the DKA.
Further SAR studies were undertaken to evaluate the
effects of the 6-ethylcarboxylate portion. Meconic acid
(Table 1, compound 2) itself, having the 6-carboxylic
acid, is inactive in the standard assay conditions (Mg2þ).
Moreover (Table 1, compounds 6–10), decarboxylation,
replacement with an ethyl- or diethyl-amide, with an
ethylether or a benzylether was accompanied by sub-
stantial or complete loss of inhibitory potency, despite
the apparent conservation of the metal binding pattern.
The more bulky tertbutylester 11 proved 10-fold less
active than the ethyl ester 3.
It seems reasonable that all these modifications had a
secondary effect in the ionization and tautomerization
behaviour of these analogues, which are likely to show
As for the diketoacids and consistent with the direct
interaction with the active-site metal ions, the inhibitory
Table 2. Inhibition of NS5b HCV pol by meconic acid derivatives
Compound
Metal ion
IC50 (lM)a
2
Mg2þ
Mn2þ
Mg2þ
Mn2þ
Mg2þ
Mn2þ
Mg2þ
Mn2þ
Mg2þ
Mn2þ
>100
18
2
3
2.25
0.165
>100
4.0
3
4
4
9
16.2
0.8
9
13
13
23
1.9
Figure 2. Overlay of a diketoacid (green) and meconic acid monoethyl
ester 2(blue).
a See Ref. 6 for polymerase assay protocol. Data are the mean of two
to four independent experiments.

P. Pace et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3257–3261
3259
different pKa values for the enolic OH and the carboxylic
acid. This hypothesis is consistent with results from
Choux and Benoit,9 who investigated the transmission
of the substituent effect on a series of substituted 3-hy-
droxy-4-pyrones by measuring the pKa values and
establishing correlation with other properties.
Structural suggestion derived from our previous SAR
studies on diketoacids inhibitors3 (see also Fig. 2), which
had revealed the presence of an aromatic moiety to be
essential to improve binding affinity, prompted us to
place a phenyl ring in the 5-position (12) or fused in a
chromene system (13). However, instead of an increase
in activity, the chromene compound 13 (IC50 ¼ 23 lM)
lost 10-fold in potency and the 5-phenyl derivative 12
showed a somewhat larger decrease in activity.
Scheme 1. Synthesis of meconic derivatives. Reagents and conditions:
(a) CH3COCl (10 equiv), MeOH or EtOH, 0 ꢁC to room temp, 24 h or
4 days; (b) i-PrN@C(O-t-Bu)NH-i-Pr (2equiv), CH 2Cl2, room temp,
24 h; (c) EtNH2 or Et2NH, HOBT (1 equiv), EDCI (1 equiv), iPr2EtN
(2equiv), DMF, 0 ꢁC to room temp, 12h; (d) 1 N NaOH (10 equiv),
MeOH, 45 min.
Unfortunately, none of the compounds active in the
HCV NS5B polymerase assays displayed either signifi-
cant inhibition of HCV RNA replication in a cell-based
assay (EC50 > 50 lM) or toxicity at the same concen-
tration.
An issue when working with meconic acid is its stabil-
ity.10 Verkade11 has described the partial decarboxyl-
ation of meconic acid with hydrochloric acid and has
proven that the carboxyl lost during the reaction is from
the position adjacent to the hydroxyl group. The solu-
tion to this will be the subject of further publications.
Scheme 2. Synthesis of compounds 6 and 12. Reagents and conditions
(see text): (a) SeO2 (3 equiv), PhBr, 160 ꢁC, 12h; (b) NaClO 2
(1.5 equiv), H2O2 30%, NaH2PO4 (0.2equiv), MeCN/H 2O (1/1 v/v),
room temp, 1 h; (c) 18 fi 6: TMSCH2N2, MeOH, room temp; 3 N HCl,
reflux, 6 h; 1 N NaOH, MeOH, room temp; (d) 19 fi 12: TMSI
(1.3 equiv), MeOH, 1 h.
2. Synthesis
In order to identify the chemical structure of the
screening sample and further study its biological prop-
erties, we prepared compound 3 and the related 4, 5, 7, 8
and 11 as in Scheme 1. Reactivity of the two carboxylate
groups of meconic acid is dependent on both electronic
as well as steric factors. In particular, esterification with
the appropriate alcohol or O-tert-butyl-N,N0-diisopro-
pylisourea, as well as the amidification, occurs primarily
at the 6-carboxylic acid.12 Diester 4 can be achieved by
prolonged reactions times, while upon saponification, it
reverts to the free 6-carboxylic acid 5. Similarly, com-
pounds 7 and 8 were obtained by exposing meconic acid
to the appropriate amine in the presence of coupling
reagents.
bility reasons, we converted the carboxylate into its
methyl ester before removal of the O-benzyl group. A
similar protocol was followed also for the synthesis of
12. The starting material 15 was prepared from maltol
by a three step literature procedure, involving its 5-
bromination,14 O-benzylation and Suzuki coupling with
phenylboronic acid.15 Then, the two-step oxidation fol-
lowed by TMSI cleavage of the benzylether afforded
12.16
Compounds 9 and 10 were prepared form Kojic acid 20,
as depicted in Scheme 3.17 Synthesis according to liter-
ature procedures allowed preparation of compounds 23
and 24. Subsequent two step oxidation of the hydroxy-
methyl group required protection of the enolic function
as a benzyl or p-methoxybenzyl ether.
Compound 6 was prepared from O-benzylmaltol 14 as
in Scheme 2. Following literature precedents and after
trying several conditions, we used SeO2 for the oxidation
of the 2-methyl to the aldehyde 16, and NaClO2 for
further oxidation to the carboxylic acid 18.13 For sta-

3260
P. Pace et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3257–3261
more potent and stable analogues will aid future studies,
interpretation of our data in light of molecular modeling
analysis and our previous experience with diketoacids
inhibitors, led to important findings concerning the
molecular determinants of HCV NS5B inhibitory
activity.
Further efforts in modifying these lead structures are in
progress and the results will be reported in due course.
Acknowledgements
The authors thank Mauro Cerretani, Nadia Gennari
and Sergio Serafini for determination of the IC50 values
for NS5b, Fabio Bonelli for mass analysis of the
screening leads and of all the compounds, Uwe Koch for
the modeling and Michael Rowley for valuable discus-
sions. This work was supported in part by a grant from
the MIUR.
References and notes
Scheme 3. Synthesis of compounds 8 and 9. Reagents and conditions:
(a) BnBr (1.1 equiv), Cs2CO3 (1 equiv) DMF, 60 ꢁC, 3 h; EtI
(1.5 equiv), NaH (1.1 equiv), DMF, room temp, 1 h; 4 N HCl, reflux,
0.5 h; (b) PMBCl (1.1 equiv), Cs2CO3 (1 equiv), DMF; BnBr
(1.1 equiv), NaH (1.1 equiv), DMF; CH2Cl2/TFA (9/1 v/v); (c) HCHO
35% sol (1.5 equiv), 1 N NaOH, room temp, 6 h; (d) BnBr or PMBCl
(1 equiv), Cs2CO3 (1 equiv), DMF, 60 ꢁC, 3 h; (e) MnO2 (3 equiv),
CH2Cl2, room temp, 24 h; (f) Ag2O (10 equiv), NaCN (5 equiv),
MeOH, room temp, 1 h; (g) 4 N HCl, reflux, 0.5 h; (h) CH2Cl2/TFA
(9/1 v/v), room temp, 1 h.
1. For a review on HCV RNA dependent RNA polymerase
as a target for antiviral development, see: Walker, M. P.;
Hong, Z. Curr. Opin. Pharm. 2002, 2, 1.
2. Dymock, B. W. Emerging Drugs 2001, 6, 13.
3. Summa, V.; Petrocchi, A.; Pace, P.; Matassa, V. G.; De
Francesco, R.; Altamura, S.; Tomei, L.; Neuner, P. J.
Med. Chem. 2004, 47, 14.
4. Lovell, S.; Subramony, P.; Kahr, B. J. Am. Chem. Soc.
1999, 121, 7020.
5. The NS5B polymerase was expressed as a C55 truncation
in E. coli and purified to homogeneity. Reactions were
carried out in 50 lL buffer containing 20 mM Tris/HCl
(pH 7.5), 0.05% Triton X-100, 2% glycerol, 50mM NaCl,
1 mM DTT, 5 mM MgCl2 and 5 nM purified NS5B_D C55
enzyme. Poly(rA)/oligo(rU) template/primer was present
at a final concentration of 20 lg/mL and 3H-UTP (Amer-
sham) at 10 lM concentration. The NS5B_D C55 protein
was preincubated with template RNA template for 20 min
at 23 ꢁC in reaction buffer. The reaction was started by the
addition of UTP and incubated at 23 ꢁC for 30 min.
The activity was measured as the radioactivity present in
the acid-insoluble material. Compounds were dissolved in
100% DMSO and serial dilutions made in DMSO. IC50
values were calculated using three parameters logistic
equation and inhibition data were fitted by Kaleidagraph
software.
Scheme 4. Synthesis of compound 13. Reagents and conditions: (a)
mCPBA (1 equiv), benzene, reflux, 3 h; (b) 1 N HCl, reflux, 1 h; (c)
HCHO 35% sol (1. 5 equiv), 1 N NaOH, room temp, 6 h; (d) PMBCl
(1.1 equiv), Cs2CO3 (1 equiv), DMF; (e) MnO2 (30 equiv), CH2Cl2,
room temp, 2h; (f) Ag 2O (2equiv), NaOH 10%, MeOH, room temp,
1 h.
6. (a) Bressanelli, S.; Tomei, L.; Roussel, A.; Incitti, I.;
Vitale, R. L. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 13034;
(b) Bressanelli, S.; Tomei, L.; Rey, F. A.; De Francesco,
R. J. Virol. 2002, 76, 3482.
7. (a) Ago, H.; Adashi, T.; Yoshida, A.; Yamamoto, M.;
Habuka, N.; Yatsunami, K.; Miyano, M. Structure 1999,
7, 1417; (b) Lesburg, C. A.; Cable, M. B.; Ferrari, E.;
Hong, Z.; Mannarino, A. F.; Weber, P. C. Nat. Struct.
Biol. 1999, 6, 937; (c) Steitz, T. A.; Smerdon, S. J.; Jager,
J.; Joyce, C. M. Science 1994, 266, 2022.
8. Tomei, L.; Altamura, S.; De Francesco, R., in preparation.
9. Choux, G.; Benoit, R. L. J. Am. Chem. Soc. 1967, 32,
3974.
For the synthesis of compound 13, commercially avail-
able 4H-chromen-4-one 29 was selected as starting
material (Scheme 4). Epoxidation and subsequent acid
opening of the epoxide allowed for introduction of the
hydroxyl group of 30.18 Synthesis of 11 from 30 utilized
the same synthetic procedure previously reported.
3. Conclusion
10. Woods, L. L. J. Org. Chem. 1957, 339.
11. Verkade, P. E. Rec. Trav. Chim. 1924, 43, 879.
12. For the structure elucidation of 2, HMBC-NMR spec-
troscopy was performed, which unambiguously showed
The monoethyl ester of meconic acid was identified as
new inhibitor of NS5B HCV polymerase. Although

P. Pace et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3257–3261
3261
correlation between the (5)CH and the (6)COOEt but not
with the (2)COOH.
14. Looker, J. H.; Prokop, R. J.; Serbousek, W. E.; Clifton,
M. D. J. Org. Chem. 1979, 19, 3408.
15. Takao, H.; Endo, Y.; Horie, T. Heterocycles 1993, 36,
1803.
O
4
5
H
O
3
2
OH
OH
16. Jung, M.; Lister, M. A. J. Org. Chem. 1977, 23, 3761.
17. For the chemistry of kojic acid, see: (a) Ellis, B. L.;
Duhme, A. K.; Hider, R. C.; Hossain, M. B.; Rizvi, S.;
Van der Helm, D. J. Med. Chem. 1996, 39, 3659;
(b) Lovell, S.; Subramony, P.; Kahr, B. J. Am. Chem.
Soc. 1999, 121, 7020; (c) Ozturk, G.; Erol, D. D.;
Aytemir, M. D.; Uzbay, T. Eur. J. Med. Chem. 2002, 37,
829.
O
6
O
O
Similar experiments corroborated also structural assigne-
ment of the other ester and amide derivatives.
13. Tobe, Y.; Yamashita, D.; Takahashi, T.; Inata, M.; Sato,
J.; Kakiuchi, K.; Kobiro, K.; Odaira, Y. J. Am. Chem.
Soc. 1990, 112, 775.
18. Rao, C. P.; Srimannarayana, G. Synth. Commun. 1987, 17,
1507.