Activity of Mannich bases of 7-hydroxycoumarin against Flaviviridae

Article abstract of DOI:10.1016/j.bmc.2007.11.045

Some Mannich bases of 7-hydroxycoumarin (2) and their simple derivatives (3 and 4) were prepared and tested against viruses containing single-stranded, positive-sense RNA genomes (ssRNA⁺). This study was directed toward Flaviviridae and, in particular, HCV surrogate viruses (BVDV, YFV). The 7-hydroxy derivatives 2 were generally devoid of activity, but when position 7 was propylated, the resulting 7-propyloxy derivatives 3 were in some cases endowed with an interesting activity against BVDV. The formation of 7-benzoyl derivatives 4 gave compounds generally lacking in activity against Flaviviridae, whereas the appearance of activity against RSV has been observed. Also some unsymmetrical methylene derivatives 5-7 (namely coumarins bridged to chromones or indoles) were found moderately active in antiviral tests. Derivatives 3 were submitted to a molecular modeling study using DNA polymerase of HCV as a target. The good correlation between calculated molecular modeling IC₅₀ and experimental EC₅₀ indicates that DNA polymerase is potentially involved in the inhibition of surrogate HCV viruses.

Full text of DOI:10.1016/j.bmc.2007.11.045

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 2591–2605
Activity of Mannich bases of 7-hydroxycoumarin
against Flaviviridae
Mauro Mazzei,a,* Erika Nieddu,a Mariangela Miele,a Alessandro Balbi,a Marco Ferrone,b
Maurizio Fermeglia,b Marco T. Mazzei,b, Sabrina Pricl,b Paolo La Colla,c
Fabio Marongiu,c Cristina Ibbac and Roberta Loddoc
aDepartment of Pharmaceutical Sciences, Viale Benedetto XV, 3, 16132 Genova, Italy
bMolecular Simulation Engineering (MOSE) Laboratory, Department of Chemical Engineering,
Piazzale Europa 1, 34127 Trieste, Italy
cDepartment of Biomedical Sciences and Technologies, University of Cagliari, Cittadella Universitaria,
09042 Monserrato, Cagliari, Italy
Received 12 September 2007; revised 12 November 2007; accepted 16 November 2007
Abstract—Some Mannich bases of 7-hydroxycoumarin (2) and their simple derivatives (3 and 4) were prepared and tested against
viruses containing single-stranded, positive-sense RNA genomes (ssRNA+). This study was directed toward Flaviviridae and, in par-
ticular, HCV surrogate viruses (BVDV, YFV). The 7-hydroxy derivatives 2 were generally devoid of activity, but when position 7
was propylated, the resulting 7-propyloxy derivatives 3 were in some cases endowed with an interesting activity against BVDV. The
formation of 7-benzoyl derivatives 4 gave compounds generally lacking in activity against Flaviviridae, whereas the appearance of
activity against RSV has been observed. Also some unsymmetrical methylene derivatives 5–7 (namely coumarins bridged to chro-
mones or indoles) were found moderately active in antiviral tests. Derivatives 3 were submitted to a molecular modeling study using
DNA polymerase of HCV as a target. The good correlation between calculated molecular modeling IC50 and experimental EC50
indicates that DNA polymerase is potentially involved in the inhibition of surrogate HCV viruses.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
As these drugs have little efficacy in a large percentage of
patients, the search for more active drugs is of extreme
interest and relevance. Up to now, this search has been
hampered by the lack of an efficient cell culture system.
This impediment has recently been overcome by the
development of subgenomic HCV RNA molecules that
replicate autonomously in transfected cells.7 Actually,
some research groups have reported the development
of a robust HCV infection system, which could facilitate
future in vitro experiments.8,9 Otherwise, as in the pres-
ent study, the biological tests must be performed on
viruses (namely BVDV, YFV, Dengue) similar to HCV.
Hepatitis C is a widespread viral disease caused by hep-
atitis C virus (HCV), a virus belonging to the Flaviviri-
dae family. The pathologic state may remain dormant
for many years and when the disease becomes evident,
the liver has already suffered severe damage, potentially
leading to a chronic infection. Such chronic hepatitis
may finally turn into hepatocarcinoma.1–4
Up to now, the only management of the disease was
treatment with ribavirin and IFN-a.5 The use of pegilat-
ed IFN-a has led to a drug giving weekly protection.6
After the unveiling of the HCV genoma, several sites of
the viral genoma have been identified as targets for HCV
drugs.10–13 The major targets are internal ribosome en-
try site (IRES), non-structural proteins NS3 (with prote-
ase-elicase activity) and NS5B (with RNA-dependent
RNA polymerase activity). With regards to this, many
nucleoside analogs were found to bind specifically to vir-
al targets.14,15 Also, non-nucleoside inhibitors are the
Keywords: Antiviral agents; Flaviviridae; HCV; Mannich bases; Cou-
marin derivatives.
*
Corresponding author. Tel.: +39 010 3538371; fax: +39 010
3538358; e-mail: mauro.mazzei@unige.it
Present address: DIMES—Viale Benedetto XV, 1-16132 Genova,
Italy.
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.11.045

2592
M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
NR2
CH2
objective of vigorous research for anti-HCV drugs. Such
inhibitors also include, besides drug-like molecules,16–19
peptidomimetics20,21 and oligonucleotides.22
HO
O
O
HO
O
O
HNR2 HCHO
EtOH
Following our interest in benzopyrans,23–25 this paper is
devoted to exploring the activity of some Mannich bases
of 7-hydroxycoumarin (2) and their simple derivatives in
position 7 as n-propyl ethers (3) and benzoyl esters (4)
against Flaviviridae. It is well known that natural and
synthetic coumarins have multiple biological activities.26
For instance, they have been claimed to be useful as
anticoagulant, antibacterial, antiinflammatory, antican-
cer, and anti-HIV agents.27–30 Very few are reported to
be able to counteract HCV. As far as we know, only
some coumarin derivatives present in patents and Ost-
hole (7-methoxy-8-prenylcoumarin) are capable of
inhibiting HCV replication and/or proliferation.31,32 In
particular, Osthole is able to counteract the progression
of hepatitis C into hepatocarcinoma.33 The action of
Osthole causes a strong reduction of plasma alanine
aminotransferase and also inhibits caspase-3 activa-
tion.34 Osthole derivatives exhibiting more solubility in
water in respect to the parent compound were recently
proposed as hepatoprotective drugs.35 For the biological
action of Osthole, the presence of methoxy group is
essential: in fact, the 7-hydroxy derivative, Osthenol, is
devoid of activity.33
1a,b
R'
R'
2a-q
1a: R' = H
2b: R' = CH3
C6H5COCl
Pyridine
CH3CH2CH2Br
DMF
an. K2CO3
NR2
CH2
NR2
CH2
CH3CH2CH2O
O
O
C6H5COO
O
O
3a-g
4a-g
R'
R'
Compound
R’
NR
2
2a
2b, 3a
2c
H
N
CH
H
N
N
N
3
O
O
2d, 3b, 4a
2e
CH
H
3
N
N
N
N
CH
CH
3
Since the great variability of pharmacological response
of coumarins is connected to large modifications of their
substituents, we found it interesting to synthesize some
coumarin Mannich bases in order to examine their abil-
ity to inhibit HCV replication. This idea proved to be a
winning strategy: in fact, a number of synthesized com-
pounds showed moderate but significant activity against
Flaviviridae family, paving the way for new coumarins
to be used in hepatitis C virus infections.
2f, 4b
CH
H
3
3
2g
2h
N(CH )CH COOH
3
2
N(CH )CH COOH
3
2
CH
H
3
2i, 3c, 4c
2j, 3d, 4d
N
CH
H
3
N
2k, 3e
2l, 3f, 4e
2m, 4f
N
N
N
N
CH CH CH
2
Although this work was specifically targeted toward
finding non-nucleoside drugs active against surrogate
HCV viruses, the Mannich bases 2–4 were also tested
against CVB-2, Sb-1, RSV, VSV, VV, Reo, HIV viruses.
CH
H
CH CH CH
2
3
N
N
N
CH
2
2
O
O
O
O
Some unsymmetrical methylene derivatives (5–7), re-
lated to the above Mannich bases and already synthe-
sized in previous works,25,36 were also tested to verify
their possible activity against HCV-related viruses as
well as a panel of other viruses.
2n
CH
H
N
CH
3
2o
N
N
COCH3
COCH3
2p, 3g
2q, 4g
CH
CH
3
N
N
2. Chemistry
3
N
N
CO OEt
The Mannich bases 2a–q, starting from 7-hydrox-
ycoumarins 1 with 40% formaldehyde and suitable sec-
ondary amines in 95% ethanol, were prepared to form
the title compounds. The hydroxy group in position 7
was in turn propylated or benzoylated (see Scheme 1).
The formation of ethers 3a–g seems to be significant
for two reasons: firstly, it is likely that, as mentioned
above, to possess antiviral activity the hydroxy group
should be alkylated; secondly, the Mannich bases result
stabilized in acidic media and thus soluble hydrochlo-
rides could be easily produced. To generate the alkyl
Scheme 1.
derivatives, we utilized n-propyl bromide in N,N-
dimethylformamide, in the presence of anhydrous potas-
sium carbonate, since alkylation with methyl iodide,
dimethylsulfate, and diethylsulfate also yielded the qua-
ternary ammonium salts. On the other hand, obtaining
the ether derivatives before performing the Mannich
reaction has no use for our purposes since, to insert
the dialkylaminomethyl group in position 8, the hydro-

M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
2593
Table 1. Antiviral activity of compounds 2a–q, 3a–g, 4a–g against the
RNA viruses indicated and their cytotoxicity against MT-4 cells
ORIII
RIII
RIV
O
N
O
MT-4a
BVDVb
YFVc
RSVd
RII
RICOO
Compound
O
O
NR2
O
CH2
O
NR2
2a
2b
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
5
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>24
>18
>100
75
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>59
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
P100
>100
>100
>25
CH2
CH2
O
R
AcO
O
ROII
O O
2c
2d
RI
RI
7
2e
6
5
2f
2g
Figure 1. Unsymmetrical methylene derivatives.
2h
2i
xy group must be free to activate that position. So, as a
first approach in this pharmacological survey, the con-
strained choice of propylation seems to be a reasonable
compromise in collecting useful data, leaving more in-
depth examination of other ether derivatives for the
future.
2j
2k
2l
28
13
>53
2m
2n
>100
>100
>100
>100
>100
>100
>100
>19
72
18
2o
2p
>100
>100
>100
38
59
2q
3a
>100
>100
>100
11
The same philosophy was followed in the acylation reac-
tion. It is interesting to note that it is not possible to ob-
tain the acyl derivatives on the hydroxy group in
position 7 simply by using anhydrides. In fact, the treat-
ment of such Mannich bases with acetic or propionic
anhydride leads to methylene-bis-derivatives as already
described.24,25 The formation of ester derivatives 4a–g
was easily accomplished by treatment of 2 with benzoyl
chloride in pyridine (Scheme 1). In this case too, other
investigations can be the subject of future research.
3b
3c
>100
>11
20
3d
3e
>100
16
>100
>48
>53
>93
15
27
>100
>100
>100
>100
>100
1
3f
3g
>100
25
8
>100
>100
>100
>100
>100
>100
>100
>100
1
4a
4b
>100
>100
>100
>100
45
>100
>100
>100
>100
19
4c
4d
18
>80
4e
The treatment of the Mannich bases of the 7-hydroxy-
coumarin with acetic or propionic anhydride in the pres-
ence of compounds possessing a suitable nucleophilic
center, as position 3 of 2-(dialkylamino)chromones or
position 3 of indoles, leads to unsymmetrical methylene
derivatives 5–7 (Fig. 1). The mechanism of this reaction
has already been reported.25 Moreover, to verify the
general application of anhydrides in this reaction, the
glutaric anhydride was also used. In this case, com-
pounds 7g [R = H, RI = (CH2)3COOH, RII = CH3,
4f
4g
NM108e
NM 299e
74
77
85
90
>100
>100
0.2
>100
4
>25
>100
1
25
a Compd concn (lM) required to reduce the viability of mock-infected
MT-4 (CD4+ Human T-cells containing an integrated HTLV-1
genome) cells by 50%, as determined by the colorimetric MTT
method.
b Compd concn (lM) required to achieve 50% protection of MDBK
cells from the BVDV (bovine viral diarrhea virus)-induced cyto-
pathogenicity, as determined by the MTT method.
c Compd concn (lM) required to achieve 50% protection of BHK
(kidney fibroblast) cells from the YFV (yellow fever virus)-induced
cytopathogenicity, as determined by the MTT method.
d Compd concn (lM) required to reduce the plaque number of RSV
(respiratory syncytial virus) by 50% in VERO-76 monolayers.
e Reference compounds: NM108 20-C-Me-guanosine NM299 6-Aza-
Uridine.
R
III = RIV = H] and 7h [R = CH3, RI = (CH2)3COOH,
RII = CH3, RIII = RIV = H], bearing a carboxylic group
in the side chain, were easily obtained.
All synthesized compounds are white crystals and their
structures are in agreement with elemental analyses
and spectral data.
3. Results
study ‘in silico’ to evaluate the efficacy of such com-
pounds to bind the surface of the most relevant HCV
enzymes.
3.1. Pharmacology
In the following tables the activity of compounds 2–4
(Table 1), 5 (Table 2), 6 (Table 3) and 7 (Table 4) against
Flaviviridae and RSV is shown. All compounds were
also tested against CVB-2, Sb-1, VSV, VV, Reo and
HIV-1. Against this panel of viruses all the compounds
resulted totally inactive (data not shown).
In general, all compounds 3 presented similar behavior
in this molecular modeling study: indeed, such molecules
resulted well fitted only on the surface of NS5B of HCV.
With regards to this, for better clarity, we will give some
details of compound 3d, as an example.
Figure 2 shows the optimized location of compound 3d
within the surface binding site of HCV RdRp. The
inhibitor is well buried within the lining cleft. The n-pro-
pyl-ether group is deeply inserted into the cavity, thus
maximizing its interactions with the side chain of Arg
3.2. Molecular modeling
Since compounds 3 showed encouraging activity against
BVDV, they were submitted to a molecular modeling

2594
M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
Table 2. Antiviral activity of compounds 5a–k against the RNA viruses indicated and their cytotoxicity against MT-4 cells
OR"
O
O
NR2
O
CH2
R'O
O
R
5
Compound
NR2
R
R0
R00
MT-4a
BVDVb
YFVc
RSVd
5a
5b
N(CH3)2
N(CH3)2
H
CH3CO
CH3CO
CH3CO
H
CH3
CH3
CH3
CH3
26
42
>51
>28
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
1
>25
25
CH3
H
5c
5d
N(C2H5)2
N(C2H5)2
83
>100
44
>100
>100
>100
>100
>100
>100
>100
>100
61
20
>100
11
CH3
CH3
CH3
H
5e
N(C2H5)2
N(C2H5)2
CH3CO
CH3CO
H
CH3CO
CH3(CH2)7
CH3
5f
5g
>100
>100
>100
>100
>100
>100
>100
0.2
>100
>100
>100
>100
7
Pirrolidin-1-yl
Pirrolidin-1-yl
Piperidin-1-yl
N(CH2CH2OCH3)2
N(CH2CH2OCH3)2
5h
5i
CH3
H
H
CH3
CH3
CH3CO
CH3CO
CH3CO
5j
5k
NM108e
NM 299e
H
CH3
CH3
CH3
5
4
>25
>100
1
25
a Compd concn (lM) required to reduce the viability of mock-infected MT-4 (CD4+ Human T-cells containing an integrated HTLV-1 genome) cells
by 50%, as determined by the colorimetric MTT method.
b Compd concn (lM) required to achieve 50% protection of MDBK cells from the BVDV (bovine viral diarrhea virus)-induced cytopathogenicity, as
determined by the MTT method.
c Compd concn (lM) required to achieve 50% protection of BHK (kidney fibroblast) cells from the YFV (yellow fever virus)-induced cytopath-
ogenicity, as determined by the MTT method.
d Compd concn (lM) required to reduce the plaque number of RSV (respiratory syncytial virus) by 50% in VERO-76 monolayers.
e Reference compounds: NM108 20-C-Me-guanosine NM299 6-Aza-Uridine.
Table 3. Antiviral activity of compounds 6a–f against the RNA viruses indicated and their cytotoxicity against MT-4 cells
O
O
NR2
O
CH2
AcO
O
R
6
Compound
NR2
R
MT-4a
BVDVb
YFVc
RSVd
6a
6b
N(CH3)2
N(C2H5)2
N(C2H5)2
H
80
>48
>100
>100
59
>52
>100
>100
77
>30
>100
13
H
>100
>100
>100
>100
>100
>100
0.2
6c
6d
CH3
H
Piperidin-1-yl
N(CH2CH2OCH3)2
N(CH2CH2OCH3)2
40
5
6e
6f
H
CH3
78
100
>100
1
>100
4
>25
P100
>100
1
NM108e
NM 299e
25
a Compd concn (lM) required to reduce the viability of mock-infected MT-4 (CD4+ Human T-cells containing an integrated HTLV-1 genome) cells
by 50%, as determined by the colorimetric MTT method.
b Compd concn (lM) required to achieve 50% protection of MDBK cells from the BVDV (bovine viral diarrhea virus)-induced cytopathogenicity, as
determined by the MTT method.
c Compd concn (lM) required to achieve 50% protection of BHK (kidney fibroblast) cells from the YFV (yellow fever virus)-induced cytopath-
ogenicity, as determined by the MTT method.
d Compd concn (lM) required to reduce the plaque number of RSV (respiratory syncytial virus) by 50% in VERO-76 monolayers.
e Reference compounds: NM108 20-C-Me-guanosine NM299 6-Aza-Uridine.

M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
2595
Table 4. Antiviral activity of compounds 7a–q against the RNA viruses indicated and their cytotoxicity against MT-4 cells
III
R
IV
R
N
II
R
CH2
I
R COO
O
R
O
7
Compound
R
RI
RII
RIII
RIV
MT-4a
BVDVb
YFVc
RSVd
7a
7b
H
CH3CH2
CH3
CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
H
65
72
P55
>100
>86
>100
>51
>87
>67
>69
>57
>100
>100
7
>52
>100
>73
>100
>100
>100
>100
>64
>100
>100
>82
>27
>72
>81
>58
1
>54
>100
>100
25
CH3
H
H
OCH3
H
7c
7d
CH3
CH3
CH3
CH3
CH3
CH3
C6H5
C6H5
C6H5
C6H5
62
41
CH3
H
CH3
H
7e
CH3CH2
CH3CH2
HOOC(CH2)3
HOOC(CH2)3
CH3
H
80
31
>100
45
7f
7g
CH3
H
H
H
27
29
>92
>100
>100
>100
>100
>100
60
7h
7i
CH3
H
H
H
>100
>100
>100
38
7j
7k
CH3
H
CH3
H
CH3CH2
CH3CH2
CH3
H
7l
7m
CH3
H
H
OCOC2H5
H
CH3
H
67
>64
P100
>40
4
7n
7o
NM108e
NM 299e
CH3
CH3
CH3
CH3
H
100
24
>100
26
H
>100
0.2
>100
1
>25
25
a Compd concn (lM) required to reduce the viability of mock-infected MT-4 (CD4+ Human T-cells containing an integrated HTLV-1 genome) cells
by 50%, as determined by the colorimetric MTT method.
b Compd concn (lM) required to achieve 50% protection of MDBK cells from the BVDV (bovine viral diarrhea virus)-induced cytopathogenicity, as
determined by the MTT method.
c Compd concn (lM) required to achieve 50% protection of BHK (Kidney fibroblast) cells from the YFV (yellow fever virus)-induced cytopath-
ogenicity, as determined by the MTT method.
d Compd concn (lM) required to reduce the plaque number of RSV (respiratory syncytial virus) by 50% in VERO-76 monolayers.
e Reference compounds: NM108 20-C-Me-guanosine NM299 6-Aza-Uridine.
Figure 2. Optimized binding mode for compound 3d into the allosteric binding site of NS5B of HCV (left). Details of the binding site (right). In this
figure, hydrogen atoms and water molecules are omitted for clarity.
422, His 475, Tyr 477, and Trp 528. The tetrahydroiso-
quinoline moiety is surrounded by a series of apolar
main and side chain atoms of Leu 419, Met 423, Ile
482, and Arg 501.

2596
M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
Table 5. Energy terms and total free energy of binding (kcal/mol) between HCV RdRp and compounds 3a–g
3a
3b
3c
3d
3e
3f
3g
DEVDW
DEEL
DEMM
DGPB
ꢀ44.66
ꢀ14.83
ꢀ59.49
42.42
ꢀ4.21
15.05
ꢀ6.23
27
ꢀ45.29
ꢀ47.90
ꢀ12.55
ꢀ60.45
47.29
ꢀ4.53
11.11
ꢀ6.58
15
ꢀ45.88
ꢀ11.21
ꢀ57.09
44.91
ꢀ4.85
10.88
ꢀ6.15
31
ꢀ44.77
ꢀ15.54
ꢀ60.31
46.07
ꢀ4.96
12.76
ꢀ6.44
19
ꢀ44.82
ꢀ13.74
ꢀ58.56
42.89
ꢀ4.09
13.79
ꢀ5.97
42
ꢀ46.65
ꢀ14.33
ꢀ60.98
45.57
ꢀ4.99
12.63
ꢀ7.77
2
ꢀ12.67
ꢀ57.96
46.78
ꢀ5.05
10.70
ꢀ5.53
88
DGNP
ꢀTDS
DGbind
Calc. IC50
Exp. EC50
38
>100
>11
20
15
27
8
Calculated IC50 and experimental EC50 values are in lM.
Furthermore, the analysis of the dynamic trajectory re-
veals a partial p–p stacking between this group and
the phenyl moiety of Tyr 477.
In conclusion, the most interesting compounds against
HCV-related viruses appear to be the propyloxy deriva-
tives 3. In this regard, it is possible to draw a correspon-
dence with Osthole in which the presence of a 7-methoxy
group is mandatory.
Further insights into the forces involved in substrate
binding can be obtained by analyzing the free energy of
binding and its components, which are listed in Table 5
for all type 3 compounds.
Benzoyl compounds 4 present a moderate interest
against HCV-related virus. Also compounds with a
methylene bridge 5–7 do not have a significant interest
against HCV-related viruses, with the only exception
of 7l. Occasionally, activity may emerge against other
viruses (in particular RSV). Some of these (4c, 5j, 5k,
6e) deserve deeper insight.
4. Discussion
It is possible to draw the following considerations from
the above results.
Since derivatives 3 are, in general, endowed with
remarkable activity against BVDV, we were interested
in testing their possible action on HCV studying the
docking of our molecules into the pocket of RNA-
dependent RNA polymerase (RdRp) of HCV by com-
puter assisted methods.
The Mannich bases derived from 7-hydroxycoumarins,
namely compounds 2, are generally inactive regarding
the tested viruses. In particular, cinnamyl derivatives
2k and 2l as well as benzodioxol derivative 2m and acet-
ylphenyl derivative 2o are cytotoxic.
The series of propyloxy derivatives 3 appears, on the
whole, to be endowed with some activity against BVDV,
whereas they are inactive against YFV and RSV. In par-
ticular, the tetrahydroisoquinoline derivative 3d, the cin-
namyl derivative 3e, and the acetylphenyl derivative 3g
resulted active against BVDV, but compounds 3e and
3g showed a certain degree of toxicity. It is interesting
to note that, in this series, the presence of the methyl
substituent in position 4 does seem to favor antiviral
activity. In this regard, it is valuable to remember that
position 4 of Osthole does not support any substituent.
As we may see from Table 5, both the intermolecular
van der Waals and the electrostatics are important con-
tributions to the binding. However, comparing the van
der Waals/non-polar (DEVDW + DGNP) with the electro-
static (DEEL + DGPB) contributions for all molecules, we
can see that in all cases the association between inhibi-
tors and the target protein is mainly driven by more
favorable non-polar interactions in the complex rather
than in solution. When examining the role of the electro-
statics in the inhibitor–enzyme complex formation, it is
however fundamental to consider the electrostatic com-
ponent of the molecular mechanical energy DEEL to-
gether with the electrostatic contribution to solvation,
DGPB. Indeed, electrostatics generally disfavor the dock-
ing of ligand and receptor molecules. The unfavorable
change in the electrostatics of solvation is mostly, but
not fully, compensated by the favorable electrostatics
within the resulting ligand–receptor complex. Indeed,
the total electrostatic energy contributions for all pro-
tein/drug complex formations are all unfavorable; the
3a, 3e, and 3g/protein complex formations being less
unfavorable than the remaining compound complex for-
mations because of a less positive total electrostatic
term, in which the penalty paid by the electrostatics of
solvation is better compensated by favorable electro-
static interactions within the complex. Thus, even
though electrostatics destabilize inhibitor/protein com-
plex formation, it is the optimized balance of opposing
The benzoyl derivatives 4 seem, generally, inactive or
moderately active against BVDV. It is important to note
the remarkable action of tetrahydroisoquinoline deriva-
tives 4c and 4d against RSV without sign of toxicity. In
particular, compound 4c shows a high level of activity
and this type of action needs to be validated by further
data.
Benzopyran and naphthopyran derivatives 5 and 6 do
not show appreciable activity against BVDV and
YFV. It is possible to note some level of activity against
RSV.
The series of indole derivatives 7 shows, on the whole,
remarkable toxicity; 7l being the most active compound
against BVDV.

M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
2597
The corresponding esters (compounds 4) have poor effi-
cacy against HCV-related viruses, but some of them de-
serve deeper insight for their activity against RSV.
Unsymmetrical methylene derivatives (compounds 5–7)
are, generally, lacking in activity on HCV-related viruses
but present a certain interest against RSV.
100
90
80
70
60
50
40
30
20
10
0
Since propyloxy derivatives 3 could be a candidate as a
lead structure for drugs against HCV, the use of molec-
ular modeling indicates that an allosteric receptor of
NS5B protein of HCV could be involved in the activity
of our compounds. The good correlation between found
and calculated IC50, as depicted in Table 5, corroborates
that proposal.
0
10
20
30
40
50
60
70
80
90
100
Finally, this study has highlighted a new type of activ-
ity of the 7-hydroxycoumarin derivatives and, follow-
ing this promising route, many other Mannich bases
of 7-hydroxycoumarin are under investigation in the
search for antiviral drugs. The second part of this
study will be devoted to the synthesis of Mannich
bases between 7-hydroxycoumarins and secondary
diamines.
Figure 3. Correlation between calculated IC50 (x-axis) and experimen-
tal EC50 (y-axis). Values are in lM. Trendline with R2 = 0.896.
electrostatic contributions that leads to tighter binding
in the series of compounds 3a, 3c, 3e, and 3g.
Interestingly, the values of IC50, reported in Table 5 and
calculated on the basis of Eq. 6 (see Experimental) for
binding to HCV RdRp, compare well with the corre-
sponding experimental data, expressed in terms of
EC50 on cells infected with BVDV. In Figure 3 we report
a correlation graph between calculated IC50 and experi-
mental IC50 with trendline calculated (R2 = 0.896). This
can be taken both as a clue that the RdRp is the effective
protein target for these compounds, and as a positive
indication of activity of these molecules against HCV.
6. Experimental
6.1. General
Melting points were determined using an Electrother-
mal apparatus and are uncorrected. Microanalyses
were carried out on a Carlo Erba 1106 elemental ana-
1
lyzer. H NMR and 13C NMR spectra were performed
on a Varian Gemini 200 (200 MHz) spectrometer using
TMS as internal standard (d = 0). IR spectra were
recorded on a Perkin-Elmer 398 Spectrophotometer.
The synthesis of compounds 5–7 are reported in Refs.
23–25.
Molecular dynamics simulations have shown to be able
to rank binding affinities of the most active molecules, to
provide insight into the interactions occurring in the ac-
tive site of the most probable target (the HCV RdPp),
and to rationalize the origins of variations in the corre-
sponding binding free energy. Accordingly, the compu-
tational strategy used in this paper can provide a
blueprint for new inhibitors in structure-based drug de-
sign or in predicting binding affinity of a ligand prior to
organic synthesis.
6.2. Mannich bases 2
6.2.1. General method. To 20.0 mmol of 7-hydroxy-
coumarin (or 4-methyl-7-hydroxycoumarin) dissolved
in 20 ml of 95% ethanol, 20.0 mmol of the appropriate
amine (A) and 2.0 ml of 40% formaldehyde were added.
The resulting mixture was refluxed for 4–6 h. After cool-
ing, the solvent was evaporated under reduced pressure.
The pale yellow oil obtained was treated with a little
amount of cool acetone, leaving a white solid crystal-
lized from suitable solvent (B) obtaining:
5. Conclusions
In the present study, we have synthesized some Mannich
bases of 7-hydroxycoumarins (2) and related com-
pounds (3–7) in order to investigate their ability in coun-
teracting the replication of Flaviviridae.
6.2.1.1. 8-Piperidinomethyl-7-hydroxycoumarin (2a).
(A) piperidine; (B) acetone. The compound is already
described.25
Our results indicate that the Mannich bases of 7-hydrox-
ycoumarins (2), having a free hydroxy group in position
7, do not convey any antiviral activity. On the other
hand, the formation of ethers in that position (com-
pounds 3) leads to derivatives endowed with remarkable
antiviral activity on HCV-related viruses. Interestingly,
the above-mentioned behavior of the hydroxy/alkoxy
groups in position 7 is analogous to that of Osthenol/
Osthole, which present some structural similarity to
our compounds.
6.2.1.2. 8-Piperidinomethyl-4-methyl-7-hydroxycoum-
arin (2b). (A) piperidine; (B) acetone. The compound is
already described.25
6.2.1.3. 8-Morpholinomethyl-7-hydroxycoumarin (2c).
(A) morpholine; (B) acetone. The compound is already
described.37

2598
M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
6.2.1.4. 7-Hydroxy-4-methyl-8-morpholinomethyl-
coumarin (2d). (A) morpholine; (B) acetone. The com-
pound is already described.37
4), 152.5 (C-10), 160.6 (C-7), 162.1 (C-2). Anal. Calcd
for C19H17NO3: C, 74.25; H, 5.58; N, 4.56. Found: C,
74.48; H, 5.65; N, 4.50.
6.2.1.5. 7-Hydroxy-8-[(4-methyl)piperazin-1-yl]meth-
ylcoumarin (2e). (A) N-methylpiperazine; (B) acetone.
Mp 168–169 ꢁC; yield 57%. IR (KBr), m(cmꢀ1): 3416
(broad), 2950, 2831, 1714, 1607, 1238. 1H NMR
(CDCl3), d: 3.22 (s, 3H, –N–CH3), 4.00–4.37 (m, 8H,
–N–CH2), 4.99 (s, 2H, CH2 bridge), 6.68 (d, J = 9, 1H,
H-3), 7.26 (d, J = 10, 1H, H-6), 7.83 (d, J = 10, 1H, H-
5), 8.13 (d, J = 9, 1H, H-4). 13C NMR (CDCl3), d:
45.0 (N–CH3), 51.7 (a N–CH2), 52.7 (CH2 bridge),
53.9 (b N–CH2), 106.7 (C-6), 110.5 (C-9), 110.7 (C-3),
113.0 (C-8), 127.3 (C-5), 143.5 (C-4), 152.3 (C-10),
160.4 (C-7), 161.9 (C-2). Anal. Calcd for C15H18N2O3:
C, 65.68; H, 6.61; N 10.21. Found: C, 65.79; H, 6.59;
N, 10.15.
6.2.1.10. 7-Hydroxy-4-methyl-8-(tetrahydroisoquino-
lin-1-yl)methylcoumarin (2j). (A) tetrahydroisoquinoline;
(B) ethanol. Mp 166–168 ꢁC; yield 67%. IR (KBr),
m(cmꢀ1): 2952, 2839, 1702, 1603, 1290. 1H NMR
(CDCl3), d: 2.40 (s, 3H, 4-CH3), 2.85–3.10 (m, 4H, a b
–N–CH2–CH2–Ar tetrahydroisoquinoline), 3.85 (s, 2H,
–N–CH2–Ar), 4.25 (s, 2H, CH2 bridge), 6.15 (s, 1H,
H-3), 6.80 (d, J = 9, 1H, H-6), 7.05–7.35 (m, 4H, H
arom. tetrahydroisoquinoline), 7.50 (d, J = 9, 1H, H-
5), 10.49 (br s, 1H, OH). Anal. Calcd for C20H19NO3:
C, 74.75; H, 5.96; N, 4.36. Found: C, 74.84; H, 6.06;
N, 4.37.
6.2.1.11. 8-(trans-4-Cinnamylpiperazin-1-yl)methyl-7-
hydroxycoumarin (2k). (A) trans-1-cinnamylpiperazine;
(B) ethanol. Mp 128–130 ꢁC; yield 62%. IR (KBr),
m(cmꢀ1): 3412 (broad), 2938, 2825, 1713, 1606, 1233.
1H NMR (CDCl3), d: 2.56–2.86 (m, 8H, –N–CH2),
3.22 (d, 2H, –N–CH2–CH@), 4.09 (s, 2H, CH2 bridge),
6.22 (d, J = 8, 1H, H-3), 6.35–6.54 (m, 2H, Ar–
CH@CH–), 6.66–7.79 (m, 8H, H-4, H-5, H-6, H arom.).
Anal. Calcd for C23H24N2O3: C, 73.38; H, 6.43; N, 7.44.
Found: C, 73.23; H, 6.46; N, 7.38.
6.2.1.6. 7-Hydroxy-4-methyl-8-[(4-methyl)piperazin-1-
yl]methylcoumarin (2f). (A) N-methylpiperazine; (B) ace-
tone. Mp 132–134 ꢁC; yield 55%. IR (KBr), m(cmꢀ1):
3502 (broad), 2953, 2823, 1727, 1279. 1H NMR
(CF3COOD), d: 2.65 (s, 3H, 4-CH3), 3.25 (s, 3H, –N–
CH3), 4.00–4.37 (m, 8H, –N–CH2), 4.99 (s, 2H, CH2
bridge), 6.68 (s, 1H, H-3), 7.30 (d, J = 9, 1H, H-6),
8.02 (d, J = 9, 1H, H-5). Anal. Calcd for C16H20N2O3:
C, 66.65; H, 6.99; N, 9.72. Found: C, 66.87; H, 7.08;
N, 9.68.
6.2.1.12. 8-(trans-4-Cinnamylpiperazin-1-yl)methyl-7-
hydroxy-4-methylcoumarin (2l). (A) trans-4-cinnamyl-
piperazine; (B) ethyl acetate. Mp 156–158 ꢁC; yield
63%. IR (KBr), m(cmꢀ1): 2938, 2807, 1721, 1601, 1285.
1H NMR(CDCl3), d: 2.34 (s, 3H, 4-CH3), 2.51–2.86
(m, 8H, N–CH2), 3.18 (d, J = 6, 2H, –N–CH2–CH@),
4.07 (s, 2H, CH2 bridge), 6.08 (s, 1H, H-3), 6.30–6.51
(m, 2H, Ar–CH@CH–), 6.65–7.55 (m, 7H, H-5, H-6,
H arom.), 10.17 (s, 1H, OH). Anal. Calcd for
C24H26N2O3: C, 73.82; H, 6.71; N 7.17. Found: C,
74.08; H, 6.80; N 7.08.
6.2.1.7. 7-Hydroxy-8-(sarcosin-N-yl)methylcoumarin
(2g). (A) sarcosine; (B) ethanol. Mp 209–211 ꢁC; yield
56%. IR (KBr), m(cmꢀ1): 3411, 2958, 2823, 1717, 1630,
1
1247. H NMR (CF3COOD), d: 3.30 (s, 3H, N–CH3),
4.41 (s, 2H, N–CH2–COOH), 5.02 (s, 2H, CH2 bridge),
6.63 (d, J = 10, 1H, H-3), 7.25 (d, J = 9, 1H, H-6), 7.85
(d, J = 9, 1H, H-5), 8.13 (d, J = 10, 1H, H-4). Anal.
Calcd for C13H13NO5: C, 59.31; H, 4.98; N, 5.32.
Found: C, 59.06; H, 5.05; N, 5.28.
6.2.1.8. 7-Hydroxy-4-methyl-8-(sarcosin-N-yl)methyl-
coumarin (2h). (A) sarcosine; (B) acetone. Mp 219–
221 ꢁC; yield 58%. IR (KBr), m(cmꢀ1): 3407, 2964,
6.2.1.13. 8-{4-[(1,3-Benzodioxol-5-yl)methyl]piperazin-
1-yl}methyl-7-hydroxycoumarin (2m). (A) 1-piperonyl-
piperazine; (B) ethyl acetate. Mp 182–183 ꢁC; yield
42%. IR (KBr), m(cmꢀ1): 3060, 2958, 2837, 1737, 1598,
1
2822, 1725, 1631, 1225. H NMR (CF3COOD), d: 2.65
(s, 3H, 4-CH3), 3.35 (s, 3H, N–CH3), 4.41 (s, 2H, N–
CH2–COOH), 5.02 (s, 2H, CH2 bridge), 6.65 (s, 1H,
H-3), 7.30 (d, J = 10, 1H, H-6), 8.02 (d, J = 10, 1H, H-
5). Anal. Calcd for C14H15NO5: C, 60.64; H, 5.45; N,
5.05. Found: C, 60.79; H, 5.37; N, 5.14.
1
1244. H NMR (CDCl3+DMSO-d6), d: 2.68–2.90 (m,
4H, b N–CH2), 3.25–3.70 (m, 4H, a N–CH2), 3.91 (s,
2H, N–CH2–Ar), 3.98 (s, 2H, CH2 bridge), 6.06 (s,
2H, O–CH2O), 6.46 (d, J = 10, 1H, H-3), 6.75–8.32
(m, 6H, H-4, H-5, H-6, H arom.), 10.21 (br s, 1H,
OH). Anal. Calcd for C22H22N2O5: C, 66.99; H, 5.62;
N, 7.10. Found: C, 67.14; H, 5.55; N, 7.05.
6.2.1.9. 7-Hydroxy-8-(tetrahydroisoquinolin-1-yl)meth-
ylcoumarin (2i). (A) tetrahydroisoquinoline; (B) acetone.
Mp 182–184 ꢁC; yield 65%. IR (KBr), m(cmꢀ1): 3024,
2917, 2834, 1713, 1606, 1238. 1H NMR (CDCl3), d:
2.90–3.20 (m, 4H, a + b –N–CH2–CH2-Ar tetrahydro-
isoquinoline), 3.85 (s, 2H, N–CH2–Ar), 4.25 (s, 2H,
CH2 bridge), 6.25 (d, J = 9, 1H, H-3), 6.80 (d, J = 7,
1H, H-6), 7.05–7.50 (m, 5H, H-5, H arom. tetrahydro-
isoquinoline), 7.70 (d, J = 9, 1H, H-4). 13C NMR
(CDCl3), d: 27.7 (C-70), 49.4 (C-20), 52.5 (CH2 bridge),
54.5 (C-80), 106.9 (C-6), 110.6 (C-9), 110.8 (C-3), 113.2
(C-8), 125.4 (C-40), 125.9 (C-50), 126.1 (C-60), 127.5 (C-
5), 128.0 (C-30), 131.9 (C-90), 132.3 (C-100), 143.6 (C-
6.2.1.14. 8-{4-[(1,3-Benzodioxol-5-yl)methyl]piperazin-
1-yl}methyl-7-hydroxy-4-methylcoumarin (2n). (A) 1-pip-
eronylpiperazine; (B) ethyl acetate. Mp 151–152 ꢁC;
yield 47%. IR (KBr), m(cmꢀ1): 3413, 2949, 2811, 1719,
1
1605, 1235. H NMR(CDCl3), d: 2.42 (s, 3H, 4-CH3),
2.52–2.90 (m, 4H, b N–CH2), 3.45–3.60 (m, 4H, a N–
CH2), 4.08–4.25 (m, 4H, –N–CH2–Ar, CH2 bridge),
6.02 (s, 2H, O–CH2O), 6.18–7.65 (m, 5H, H-5, H-6, H
arom.). Anal. Calcd for C23H24N2O5: C, 67.63; H,
5.92; N 6.86. Found: C, 67.53; H, 5.99; N 6.78.

M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
2599
6.2.1.15. 8-[4-(4-Acetlyphen-1-yl)piperazin-1-yl]methyl-
7-hydroxycoumarin (2o). (A) 40-piperazinoacetophe-
none; (B) ethyl acetate. Mp 162–164 ꢁC; yield 65%. IR
(KBr), m(cmꢀ1): 3444, 2947, 2818, 1717, 1664, 1601,
1235. 1H NMR (CDCl3), d: 2.46 (s, 3H, CH3–CO),
2.65–2.92 (m, 4H, b N–CH2), 3.25–3.50 (m, 4H, a N–
CH2), 4.06 (s, 2H, CH2 bridge), 6.14 (d, J = 9.4, 1H,
H-3), 6.72 (d, J = 8.6, 1H, H-6), 6.82 (d, J = 7.6, 2H, b
arom.), 7.24 (d, J = 8,6, 1H, H-5), 7.57 (d, J = 9.4, 1H,
H-4), 7.82 (d, J = 7.6, 2H, a arom.). 13C NMR (CDCl3),
d: 25.4 (CH3), 46.5 (b N–CH2), 51.5 (a N–CH2), 52.7
(CH2 bridge), 106.5 (C-6), 110.8 (C-9), 111.0 (C-3),
113.1 (C-8, C-20,60), 127.6 (C-5, C-40), 129.6 (C-30,50),
143.5 (C-4), 152.4 (C-10), 152.9 (C-10), 160.4 (C-7),
161.5 (C-2), 195.8 (CO). Anal. Calcd for C22H22N2O4:
C, 69.83; H, 5.86; N 7.40. Found: C, 70.05; H, 5.88; N
7.49.
6.3.1.1. 4-Methyl-8-piperidinomethyl-7-propyloxy-
coumarin (3a). Crystallized from ethyl acetate, mp
147–148 ꢁC; yield 65%. IR (KBr), m(cmꢀ1): 3425, 2933,
2850, 1720, 1605, 1569, 1292. 1H NMR
(CDCl3+DMSO-d6), d: 1.10 (t, J = 5, 3H, CH3–CH2),
1.35–1.62 (m, 8H, CH2 b + c piperidine, CH3–CH2),
2.32–2.71 (m, 7H, CH2a piperidine, 4-CH3), 3.82 (s,
2H, CH2 bridge), 4.08 (t, J = 5, 2H, CH2–O), 6.15 (s,
1H, H-3), 6.94 (d, J = 9, 1H, H-6), 7.58 (d, J = 9, 1H,
H-5). Anal. Calcd for C19H25NO3: C, 72.35; H, 7.99;
N, 4.44. Found: C, 72.14; H, 7.94; N, 4.52.
6.3.1.2. 4-Methyl-8-morpholinomethyl-7-propyloxy-
coumarin (3b). Crystallized from ethyl acetate, mp
146–147 ꢁC; yield 52%. IR (KBr), m(cmꢀ1): 3444, 2929,
1
2846, 1725, 1602, 1284. H NMR (CDCl3), d: 1.09 (t,
J = 5.5, 3H, CH3–CH2), 1.68–2.09 (m, 2H, CH3–CH2–
), 2.39 (s, 3H, 4-CH3), 2.48–2.73 (m, 4H, CH2 a morpho-
line), 3.55–3.79 (m, 4H, CH2 b morpholine), 3.89 (s, 2H,
CH2 bridge), 4.05 (t, J = 5.5, 2H, CH2–O), 6.16 (s, 1H,
H-3), 6.90 (d, J = 9.5, 1H, H-6), 7.56 (d, J = 9.5, 1H,
H-5). Anal. Calcd for C18H23NO4: C, 68.12; H, 7.30;
N, 4.41. Found: C, 67.88; H, 7.40; N, 4.47.
6.2.1.16. 8-[4-(4-Acetlyphen-1-yl)piperazin-1-yl]methyl-
7-hydroxy-4-methylcoumarin (2p). (A) 40-piperazinoace-
tophenone; (B) ethyl acetate. Mp 217–218 ꢁC; yield
63%. IR (KBr), m(cmꢀ1): 3413, 2947, 2843, 1716, 1702,
1
1662, 1596, 1237. H NMR (CDCl3), d: 2.39 (s, 3H,
CH3–CO), 2.51 (s, 3H, 4-CH3), 2.68–2.93 (m, 4H, b
N–CH2), 3.31–3.55 (m, 4H, a N–CH2), 4.14 (s, 2H,
CH2 bridge), 6.12 (s, 1H, H-3), 6.78 (d, J = 8.4, 1H,
H-6), 6.92 (d, J = 7.6, 2H, b arom.), 7.42 (d, J = 8.4,
1H, H-5), 7.94 (d, J = 7.6, 2H, a arom.). Anal. Calcd
for C23H24N2O4: C, 70.39; H, 6.16; N, 7.14. Found: C,
70.49; H, 6.11; N, 7.05.
6.3.1.3. 7-Propyloxy-8-(tetrahydroisoquinolin-1-yl)-
methylcoumarin (3c). Crystallized from ethyl ether, mp
113–114 ꢁC; yield 60%. IR (KBr), m(cmꢀ1): 2920, 2851,
1
1720, 1605, 1282. H NMR (CDCl3), d: 0.90 (m, 3H,
CH3–CH2), 1.50–1.85 (m, 2H, CH3–CH2–), 2.81–3.09
(m, 4H, –N–CH2–CH2–Ar), 3.81 (t, 2H, CH2–O),
3.90–4.27 (m, 4H, CH2 bridge, –N–CH2–Ar), 6.28 (d,
J = 9, 1H, H-3), 6.78–7.82 (m, 7H, H-4, H-5, H-6, H
arom.). 13C NMR (CDCl3), d: 13.4 (CH3), 25.4 (CH3–
CH2), 31.2 (C-70), 47.8 (C-20), 50.3 (CH2 bridge), 54.9
(C-80), 68.2 (CH2–O), 107.6 (C-6), 111.9 (C-8), 112.3
(C-3), 113.6 (C-9), 124.6 (C-40), 125.2 (C-50), 125.8
(C-60), 127.3 (C-5), 127.9 (C-30), 133.5 (C-90), 134.4
(C-100), 143.0 (C-4), 153.3 (C-10), 160.2 (C-7), 160.5
(C-2). Anal. Calcd for C22H23NO3: C, 75.62; H, 6.63;
N, 4.01. Found: C, 75.82; H, 6.69; N, 3.95.
6.2.1.17. 8-[4-(Ethoxycarbonyl)piperazin-1-yl]methyl-
7-hydroxy-4-methylcoumarin (2q). (A) ethyl 1-piperazine-
carboxylate; (B) ethyl acetate. Mp 137–138 ꢁC; yield 48%.
IR (KBr), m(cmꢀ1): 3440, 2950, 2867, 1731, 1709, 1603,
1
1287. H NMR (CDCl3), d: 1,26 (t, J = 6, 3H, CH3–
CH2), 2.39 (s, 3H, 4-CH3), 2.49–2.78 (m, 4H, b N–
CH2), 3.42–3.78 (m, 4H, a N–CH2), 4.10 (s, 2H, CH2
bridge), 4.18 (q, J = 6, 2H, CH3–CH2), 6.11 (s, 1H, H-
3), 6.79 (d, J = 8, 1H, H-6), 7.47 (d, J = 8, 1H, H-5),
9.20 (s, 1H, OH). Anal. Calcd for C18H22N2O5: C,
62.42; H, 6.40; N, 8.09. Found: C, 62.19; H, 6.39; N, 8.16.
6.3.1.4. 4-Methyl-7-propyloxy-8-(tetrahydroisoquino-
lin-1-yl)methylcoumarin (3d). Crystallized from ethyl
ether, mp 114–115 ꢁC; yield 64%. IR (KBr), m(cmꢀ1):
2936, 2833, 1718, 1605, 1289. 1H NMR (CDCl3), d:
1.16 (t, J = 5, 3H, CH3–CH2), 1.72–2.05 (m, 2H,
–CH2–CH2–O–), 2.41 (s, 3H, 4-CH3), 2.85–3.06 (m,
4H, –N–CH2–CH2–Ar), 3.81 (t, J = 5, 2H, –CH2–O),
3.92–4.24 (m, 4H, CH2 bridge, –N–CH2–Ar tetrahydro-
isoquinoline), 6.20 (s, 1H, H-3), 6.90–7.69 (m, 6H, H-5,
H-6, H arom.). Anal. Calcd for C23H25NO3: C, 76.01;
H, 6.93; N, 3.85. Found: C, 75.87; H, 7.01; N, 3.87.
6.3. Propyl derivatives 3
6.3.1. General method. In a round-bottom flask protected
from moisture with a calcium chloride dryness tube, a
quantity of 2.0 mmol of Mannich base 2 was heated in
10 ml of N,N-dimethylformamide until dissolution was
completed, then 0.3 g of anhydrous potassium carbonate
and 2.0 mmol (0.23 g) of 1-bromopropane were added.
The mixture was heated at 100 ꢁC for 5 h under a slight
nitrogen stream. At the end, 15 ml of water and 20 ml
of chloroform were added and the mixture was allowed
to stir for 30 min. Then, the organic phase was separated
out and the aqueous one was extracted again with chlo-
roform. The pooled organic phases were dried on anhy-
drous sodium sulfate, filtered, and evaporated under
reduced pressure. The oily residue was cooled and pre-
cipitated from ethyl ether. The precipitate was filtered
and crystallized from suitable solvent. The following
compounds were obtained:
6.3.1.5. 8-(trans-4-Cinnamylpiperazin-1-yl)methyl-7-
propyloxycoumarin (3e). Crystallized from ethyl acetate,
mp 148–149 ꢁC; yield 54%. IR (KBr), m(cmꢀ1): 2941,
1
2811, 1717, 1605, 1238. H NMR (CDCl3+DMSO-d6),
d: 1.08 (t, J = 5, 3H, CH3–CH2), 1.70–2.00 (m, 2H,
CH2–CH2–O), 2.50–2.82 (m, 8H, –N–CH2–CH2–N–),
3.15 (d, J = 5, 2H, –N–CH2–CH@), 3.95 (s, 2H, CH2
bridge), 4.05 (t, J = 5, 2H, –CH2–O–), 6.05–7.80 (m,
10H, –CH@CH, H-4, H-5, H-6, H arom.). Anal. Calcd

2600
M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
for C26H30N2O3: C, 74.61; H, 7.22; N, 6.69. Found: C,
74.72; H, 7.12; N, 6.77.
6.4.1.2. 7-Benzoyloxy-4-methyl-8-(4-methylpiperazin-
1-yl)methylcoumarin (4b). Mp 141–142 ꢁC; yield 40%.
IR (KBr), m(cmꢀ1): 2936, 2837, 1743, 1726, 1600, 1229.
1H NMR (CDCl3) d: 1.95–2.26 (m, 7H, CH3–N–
CH2CH2), 2.31–2.59 (m, 7H, 4-CH3, CH3–N–CH2CH2),
3.82 (s, 2H, CH2 bridge), 6.33 (s, 1H, H-3), 7.22 (d,
J = 8, 1H, H-6), 7.49–8.40 (m, 6H, H-5, H arom.). Anal.
Calcd for C23H24N2O4: C, 70.39; H, 6.16; N, 7.14.
Found: C, 70.18; H, 6.19; N, 7.21.
6.3.1.6. 8-(trans-4-Cinnamylpiperazin-1-yl)methyl-4-
methyl-7-propyloxycoumarin (3f). Crystallized from
ethyl acetate, mp 123–124 ꢁC; yield 46%. IR (KBr),
m(cmꢀ1): 2940, 2812, 1724, 1606, 1236. 1H NMR
(CDCl3), d: 1.06 (t, J = 5, 3H, CH3–CH2), 1.75–1.98
(m, 2H, CH2–CH2–O–), 2.38 (s, 3H, 4-CH3), 2.45–2.80
(m, 8H, –N–CH2–CH2–N–), 3.11 (d, J = 5, 2H, –N–
CH2–CH@), 3.96 (s, 2H, CH2 bridge), 4.02 (t, J = 5,
2H, –CH2–O), 6.15 (s, 1H, H-3), 6.30–7.63 (m, 9H,
–CH@CH, H-5, H-6, H arom.). Anal. Calcd for
C27H32N2O3: C, 74.97; H, 7.46; N, 6.48. Found: C,
75.23; H, 7.55; N, 6.38.
6.4.1.3. 7-Benzoyloxy-8-(tetrahydroisoquinolin-1-yl)-
methylcoumarin (4c). Mp 152–154 ꢁC; yield 65%. IR
1
(KBr), m(cmꢀ1): 2906, 2854, 1733, 1625, 1603, 1244. H
NMR (CDCl3), d: 2.55–2.78 (m, 4H, –N–CH2–CH2–
Ar), 3.55 (s, 2H, –N–CH2–Ar), 3.95 (s, 2H, CH2 bridge),
6.45 (d, 1H, J = 10, H-3), 6.80–8.25 (m, 12H, H-4, H-5,
H-6, H arom.). Anal. Calcd for C26H21NO4: C, 75.90;
H, 5.14; N, 3.40. Found: C, 76.12; H, 5.06; N, 3.38.
6.3.1.7. 8-[4-(4-Acetylphen-1-yl)piperazin-1-yl]methyl-
4-methyl-7-propyloxycoumarin (3g). Crystallized from
ethanol, mp 147–148 ꢁC; yield 46%. IR (KBr),
1
m(cmꢀ1): 2941, 2811, 1728, 1661, 1600, 1287. H NMR
6.4.1.4. 7-Benzoyloxy-4methyl-8-(tetrahydroisoquino-
lin-1-yl)methylcoumarin (4d). Mp 180–182 ꢁC; yield
40%. IR (KBr), m(cmꢀ1): 2922, 2840, 1738, 1598, 1223.
1H NMR (CDCl3) d: 2.49 (s, 3H, 4-CH3), 2.58–2.79
(m, 4H, N–CH2CH2), 3.57 (s, 2H, N–CH2–Ar), 3.99
(s, 2H, CH2 bridge), 6.36 (s, 1H, H-3), 6.90–8.26 (m,
(CDCl3), d: 1.02 (t, J = 6.4, 3H, CH3–CH2), 1.70–1.89
(m, 2H, CH2–CH2–O–), 2.41 (s, 3H, 4-CH3), 2.43 (s,
3H, CH3–CO), 2.67 (t, J = 5.2, 4H, b CH2–N), 3.26 (t,
4H, a N–CH2), 3.88 (s, 2H, CH2 bridge), 3.97 (t,
J = 6.4, 2H, CH2–O), 6.18 (d, J = 9.6, 1H, H-3), 6.74
(d, J = 9.2, 2H, b arom.), 6.79 (d, J = 9.4, 1H, H-6),
7.30 (d, J = 9.6, 1H, H-4), 7.60 (d, J = 9.4 1H, H-5),
7.82 (d, J = 9.2, 2H, a arom.). 13C NMR (CDCl3), d:
10.0 (CH3–CH2), 18.1 (4-CH3), 21.9 (CH3–CO), 25.4
(CH3–CH2), 46.5 (a N–CH2), 47.7 (CH2 bridge), 51.6
(b N–CH2), 69.7 (CH2–O), 107.6 (C-6), 111.9 (C-3),
112.3 (C-8), 112.4 (C-9), 112.5 (C-20,60), 126.6 (C-40),
127.6 (C-5), 129.6 (C-30,50), 143.1 (C-4), 153.3 (C-10),
153.4 (C-10), 160.2 (C-7), 160.3 (C-2), 195.7 (CO). Anal.
Calcd for C26H30N2O4: C, 71.87; H, 6.96; N, 6.45.
Found: C, 71.95; H, 6.91; N, 6.52.
11H, H-5, H-6,
H
arom.). Anal. Calcd for
C27H23NO4: C, 76.22; H, 5.45; N, 3.29. Found: C,
76.51; H, 5.54; N, 3.19.
6.4.1.5. 7-Benzoyloxy-8-(trans-4-cinnamylpiperazin-1-
yl)-4-methylcoumarin (4e). Mp 221–223 ꢁC; yield 68%.
IR (KBr), m(cmꢀ1): 2948, 2824, 1732, 1630, 1599, 1235.
1H NMR (CDCl3), d: 2.49 (s, 3H, 4-CH3), 2.72–2.95
(m, 8H, –N–CH2–CH2–N–), 3.63 (d, J = 5, 2H, –N–
CH2–CH@), 3.97 (s, 2H, CH2 bridge), 6.25–6.68 (m,
3H, H-3, –CH@CH–), 7.25–8.35 (m, 12H, H-5, H-6, H
arom.). 13C NMR (CDCl3), d: 18.2 (4-CH3), 48.0 (a
N–CH2), 48.4 (CH2 bridge), 48.7 (b N–CH2), 58.0
(CH2–CH@), 113.8 (C-3), 115.3 (@CH–Ar), 116.8 (C-
6), 117.4 (C-8), 118.7 (C-9), 124.2 (C-5), 126.3 (C-
200,600), 127.9 (C-400), 128.1 (C-300,500), 128.2 (C-30,50),
128.3 (C-20,60), 129.2 (C-10), 133.5 (C-40), 134.0 (C-100),
139.4 (@CH–CH2), 151.5 (C-10), 152.1 (C-7), 152.2
(C-4), 159.0 (C-2), 164.3 (CO). Anal. Calcd for
C31H30N2O4: C, 75.28; H, 6.11; N, 5.66. Found: C,
75.51; H, 6.04; N, 5.74.
6.4. Benzoyl derivatives 4
6.4.1. General method. In a round-bottom flask pro-
tected from moisture with a calcium chloride dryness
tube, a quantity of 2.0 mmol of Mannich base 2 was
heated in 10 ml of pyridine until dissolution was com-
pleted. Then 4.0 mmol (0.56 g) of benzoyl chloride was
added. The mixture was heated at 100 ꢁC for 3 h. At
the end, after cooling, ice was added and a precipitate
was formed. The solid was filtered and crystallized from
ethyl ether. The following compounds were obtained:
6.4.1.6. 8-{4-[(1,3-Benzodioxol-5-yl)methyl]piperazin-
1-yl}methyl-7-benzoyloxycoumarin (4f). Mp 227–
228 ꢁC; yield 56%. IR (KBr), m(cmꢀ1): 2958, 2837,
1737, 1598,1253. 1H NMR (CDCl3+DMSO-d6) d:
2.65–2.88 (m, 8H, N–CH2CH2–N), 3.91 (s, 2H, N–
CH2 benzodioxole), 3.98 (s, 2H, CH2 bridge), 6.06 (s,
2H, O–CH2–O), 6.46 (d, J = 9, 1H, H-3), 6.85–8.32
(m, 11H, H-4, H-5, H-6, H arom.). Anal. Calcd for
C29H26N2O6: C, 69.87; H, 5.26; N, 5.62. Found: C,
70.10; H, 5.33; N, 5.54.
6.4.1.1. 7-Benzoyloxy-4-methyl-8-morpholinomethyl-
coumarin (4a). Mp 161–162 ꢁC; yield 50%. IR (KBr),
m(cmꢀ1): 2965, 2852, 1749, 1728, 1716, 1629, 1600,
1227. 1H NMR (CDCl3) d: 2.21–2.59 (m, 7H, N–
CH2CH2–O, 4-CH3), 3.20–3.52 (m, 4H, N–CH2CH2–
O), 3.81 (s, 2H, CH2 bridge), 6.34 (s, 1H, H-3), 7.22 (d,
J = 7, 1H, H-6), 7.45–8.43 (m, 6H, H-5+H arom.). 13C
NMR (CDCl3), d: 18.1 (4-CH3), 52.0 (N–CH2CH2–O),
53.1 (CH2 bridge), 65.7 (N–CH2CH2–O), 106.3 (C-6),
110.0 (C-3), 111.7 (C-9), 112.8 (C-8), 127.6 (C-30,50),
128.0 (C-5), 129.3 (C-20,60), 129.4 (C-10), 132.6 (C-40),
151.8 (C-10), 152.7 (C-7), 160.5 (C-4), 161.3 (C-2),
170.3 (CO). Anal. Calcd for C22H21NO5: C, 69.65; H,
5.58; N, 3.69. Found: C, 69.52; H, 5.68; N, 3.62.
6.4.1.7. 7-Benzoyloxy-8-[4-(ethoxycarbonyl)piperazin-
1-yl]-4-methylcoumarin (4g). Mp 177–178 ꢁC; yield 65%.
IR (KBr), m(cmꢀ1): 2931, 2846, 1742, 1728, 1704, 1600,
1
1232. H NMR (CDCl3) d: 1.20 (t, J = 7.5, 3H, CH3–
CH2), 2.12–2.58 (m, 7H, 4-CH3, CH2–N–CO), 2.98–

M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
2601
3.29 (m, 4H, CH2–N–CH2), 3.86 (s, 2H, CH2 bridge),
4.11 (q, J = 7.5, 2H, O–CH2), 6.36 (s, 1H, H-3), 7.13–
8.42 (m, 7H, H-5, H-6, H arom.). Anal. Calcd for
C25H26N2O6: C, 66.66; H, 5.82; N, 6.22. Found: C,
66.41; H, 5.73; N, 6.28.
For cytotoxicity evaluations, exponentially growing cells
derived from human hematological tumors [CD4+ hu-
man T-cells containing an integrated HTLV-1 genome
(MT-4)] were seeded at an initial density of 1 · 105
cells/ml in 96-well plates in RPMI-1640 medium supple-
mented with 10% fetal calf serum (FCS), 100 U/ml pen-
icillin G, and 100 lg/ml streptomycin. Cell cultures were
then incubated at 37 ꢁC in a humidified, 5% CO2 atmo-
sphere in the absence or presence of serial dilutions of
test compounds. Cell viability was determined after
96 h at 37 ꢁC by the 3-(4,5-dimethylthiazol-2-yl)-2,5-di-
phenyl-tetrazolium bromide (MTT) method.38
6.5. Unsymmetrical derivatives 7
To 5 g of glutaric anhydride, heated until fusion occurred
(60 ꢁC), 2.0 mmol (0.26 g) of 2-methylindole was added,
then 2.0 mmol of 7-hydroxy-8-piperidinomethylcouma-
rin (or 7-hydroxy-4-methyl-8-piperidinomethylcouma-
rin) was added, and the solution was heated at 95 ꢁC
for 1.5 h. After cooling, the solution was poured onto
crushed ice and the mixture was stirred for 1–2 h. The pre-
cipitate was filtered and the solid was crystallized from
ethyl acetate. The following compounds were obtained:
6.6.4. Antiviral assays. Activity of compounds against
yellow fever virus (YFV), 17D vaccine strain, was based
on inhibition of virus-induced cytopathogenicity in
acutely infected BHK-21 cells. Activities against bovine
viral diarrhoea virus 1 (BVDV, strain NADL, ATCC
Number VR-534), in infected MDBK cells, and against
respiratory syncytial virus (RSV), A-2 strain, in infected
Vero 76 cells, were also based on inhibition of virus-in-
duced cytopathogenicity.
6.5.1. Glutaric acid mono 8-[(20-methylindol-30-yl)-
methyl]coumarin-7-yl ester (7g). Mp 202–203 ꢁC; yield
56%. IR (KBr), m(cmꢀ1): 3341, 2907 (broad), 1760,
1
1743, 1722, 1635, 1600. H NMR (DMSO-d6) d: 2.10–
2.81 (m, 9H, CH3, CH2), 4.07 (s, 2H, CH2 bridge),
6.46 (d, J = 9, 1H, H-3), 6.80–8.21 (m, 7H, H-4, H-5,
H-6, H arom.), 10.88 (s, 1H, NH). Anal. Calcd for
C24H21NO6: C, 68.73; H, 5.05; N, 3.34. Found: C,
68.77; H, 4.96; N, 3.27.
BHK, MDBK, and Vero 76 cells were seeded in 96-well
plates at a density of 5 · 104, 3 · 104, and 2.5 · 104 cells/
well, respectively, and were allowed to form confluent
monolayers by incubating overnight in growth medium
at 37 ꢁC in a humidified CO2 (5%) atmosphere. Cell
monolayers were then infected with 50 ll of a proper
virus dilution (in serum-free medium) to give an
m.o.i = 0.01 (0.1 in the case of RSV). One hour later,
50 ll of MEM Earle’s medium (Dulbecco’s modified Ea-
gle’s medium for Vero/RSV), supplemented with inacti-
vated fetal calf serum (FCS), 1% final concentration,
without or with serial dilutions of test compounds,
was added. After a 3-day incubation (5 days for Vero/
RSV) at 37 ꢁC, cell viability was determined by the
MTT method. In the case of Vero/RSV, cell viability
was determined with crystal violet staining of the mono-
layer (see below), followed by OD determination in
spectrophotometer at 570 nm of the dye recovered from
the monolayer solubilized with a solution containing 1%
sarkosyl and Hepes 10%. 20-C-Me-guanosine and 6-
Aza-Uridine were used as reference compounds for Fla-
viviridae and RSV, respectively.
6.5.2. Glutaric acid mono 8-[(20-methylindol-30-yl)-
methyl]-4-methylcoumarin-7-yl ester (7h). Mp 201–
202 ꢁC; yield 62%. IR (KBr), m(cmꢀ1): 3335, 2903
(broad), 1754, 1734, 1716, 1630, 1599. 1H NMR
(DMSO-d6) d: 2.12–2.68 (m, 12H, CH3, CH2), 4.13 (s,
2H, CH2 bridge), 6.25 (s, 1H, H-3), 6.76–7.72 (m, 6H,
H-5, H-6, H arom.), 10.02 (s, 1H, NH). Anal. Calcd
for C25H23NO6: C, 69.27; H, 5.35; N, 3.23. Found: C,
69.54; H, 5.40; N, 3.14.
6.6. Biological assays
6.6.1. Compounds. Compounds were dissolved in DMSO
at 100 mM and then diluted in culture medium.
6.6.2. Cells and viruses. Cell lines were purchased from
American Type Culture Collection (ATCC). The ab-
sence of mycoplasma contamination was checked peri-
odically by the Hoechst staining method. Cell lines
supporting the multiplication of RNA viruses were the
following: Madin Darby Bovine Kidney (MDBK); Baby
Hamster Kidney (BHK-21); CD4+ human T-cells con-
taining an integrated HTLV-1 genome (MT-4); Monkey
kidney (Vero 76) cells.
6.7. Molecular modeling methods
The starting 3-D model of the HCV RNA-dependent
RNA polymerase (RdRp) was based on its X-ray crys-
tallographic structure39 (chain B, PBD Code: 1CSJ).
Missing hydrogen atoms were added to the protein
backbone and side chains with the parse module of Am-
ber 7.0.40,41 All ionizable residues were considered in the
standard ionization state at neutral pH. The geometry of
added hydrogen was refined for 200 steps (steepest des-
cent) in vacuum, using the well-validated, all-atom force
field (FF) parameter set (parm94) by Cornell et al.42
Further protein geometry refinement was carried out
using the sander module of Amber 7.0 via a combined
steepest descent—conjugate gradient algorithm, using
the root-mean-square of the Cartesian elements of the
6.6.3. Cytotoxicity assays. For cytotoxicity tests, run in
parallel with antiviral assays, MDBK, BHK, and Vero
76 cells were resuspended in 96 multiwell plates at an ini-
tial density of 6 · 105, 1 · 106, and 5 · 105 cells/ml,
respectively, in maintenance medium, without or with
serial dilutions of test compounds. Cell viability was
determined after 48–120 h at 37 ꢁC in a humidified
CO2 (5%) atmosphere by the MTT method. The cell
number of Vero 76 monolayers was determined by stain-
ing with the crystal violet dye.
˚
gradient equal to 0.01 kcal/(mol A) as a convergence cri-

2602
M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
terion for the energy gradient. The generalize born/sur-
face area (GB/SA) continuum solvation model43,44 was
used to mimic a water environment. As expected, no rel-
evant structural changes were observed between RdRp
relaxed model and the original 3-D structure.
GB/SA implicit water model was again used to mimic
the solvated environment. The rotation of the angles /
and u, and the angles of the side chains was set free dur-
ing the calculations. All other parameters of the MC/SA
algorithm were kept as default. Following the docking
procedure, the structure of all compounds was subjected
˚
The model structures of the selected inhibitors were gen-
erated using Biopolymer module of Insight II (v. 2000.1,
Accelrys Inc., San Diego, CA, USA). All molecules were
subjected to an initial energy minimization, again using
the sander module of the Amber 7.0 suite of programs,
with the parm94 version of the Amber force field.42 In
this case, the convergence criterion was set at
to cluster analysis with a tolerance of 1 A for an all-
atom root-mean-square (RMS) deviation from a low-
er-energy structure representing each cluster family. In
the absence of any relevant crystallographic informa-
tion, the structure of each resulting complex character-
ized by the lowest interaction energy in the prevailing
cluster was selected for further evaluation.
10ꢀ4 kcal/(mol A). A conformational search was carried
˚
out using a well-validated, ad hoc developed combined
molecular mechanics/molecular dynamics simulated
annealing (MDSA) protocol.45–48 Accordingly, the re-
laxed structures were subjected to 5 repeated tempera-
ture cycles (from 310 to 1000 K and back) using
constant volume/constant temperature (NVT) MD con-
ditions. At the end of each annealing cycle, the struc-
tures were again energy minimized to converge below
Each best substrate/RdRp complex, resulting from the
automated docking procedure, was further refined in
the Amber 7.0 suite using the quenched molecular
dynamics (QMD) method.60 In this case, 100 ps MD
simulation at 310 K was employed to sample the confor-
mational space of the substrate–enzyme complex in the
GB/SA continuum solvation environment. The integra-
tion step was equal to 1 fs. After each ps, the system
was cooled to 0 K, the structure was extensively mini-
mized, and stored. To prevent global conformational
changes of the enzyme, the backbone of the protein
binding site was constrained by a harmonic force con-
10ꢀ4 kcal/(mol A), and only the structures correspond-
˚
ing to the minimum energy were used for further model-
ing. The atomic partial charges for the geometrically
optimized compounds were obtained using the RESP
procedure,49 and the electrostatic potentials were pro-
duced by single-point quantum mechanical calculations
at the Hartree–Fock level with a 6-31G* basis set, using
the Merz–Singh–Kollman van der Waals parame-
ters.50,51 Eventual missing force field parameters for
the molecules were generated as follows: AM152 geome-
try optimization of the structure was followed by RHF/
6-31G* single point calculation to obtain the electro-
static potentials. Next, the RESP method40 was used
for charge fitting. The missing bond, angle torsion, or
van der Waals parameters not included in the parm94
were transferred53 from the newly developed parm99
parameter set,54 or from a general AMBER force field
(GAFF).55
˚
stant of 100 kcal/A, whereas the amino acid side chains
and the ligands were allowed to move without any
constraint.
The best energy configuration of each complex resulting
˚
from the previous step was allowed to relax in a 55-A ra-
dius sphere of TIP3P water molecules.61 The resulting
system was minimized with a gradual decrease in the po-
sition restraints of the protein atoms. At the end of the
relaxation process, all water molecules beyond the first
˚
hydration shell (i.e., at a distance >3.5 A from any pro-
tein atom) were removed. Finally, to achieve electroneu-
trality, a suitable number of counterions were added, in
the positions of the largest electrostatic potential, as
determined by the cion module within Amber 7.0. To re-
duce computational time to reasonable limits, all protein
The optimized structures of the test compounds were
docked into the HCV polymerase allosteric binding site
by applying a consolidated procedure46,56,57; accord-
ingly, it will be described here only briefly. The software
AutoDock 3.058 was employed to estimate the possible
binding orientations of all compounds in the receptor.
In order to encase a reasonable region of the protein sur-
face and interior volume, centered on the crystallo-
˚
residues with any atom closer than 20 A from the center
of the mass of each bounded ligand were chosen to be
flexible in the dynamic simulations. Subsequently, a
˚
spherical TIP3P water cap of radius equal to 20 A was
centered on each inhibitor in the corresponding RdRp
complex, including the hydrating water molecules within
the sphere resulting from the previous step. After energy
minimization of the new water cap for 1500 steps, keep-
ing the protein, the ligand, and the pre-existing waters
rigid, followed by a MD equilibration of the entire water
sphere with fixed solute for 20 ps, further unfavorable
interactions within the structures were relieved by pro-
gressively smaller positional restraints on the solute
˚
graphic identified binding site, the grids were 60 A on
˚
each side. Grid spacing (0.375 A), and 120 grid points
were applied in each Cartesian direction so as to calcu-
late mass-centered grid maps. Amber 12-6 and 12-10
Lennard–Jones parameters were used in modeling van
der Waals interactions and hydrogen bonding (N–H,
O–H, and S–H), respectively. In the generation of the
electrostatic grid maps, the distance-dependent relative
permittivity of Mehler and Solmajer59 was applied.
2
˚
(from 25 to 0 kcal/(mol A )) for a total of 4000 steps.
Each system was gradually heated to 310 K at three
intervals, allowing a 5 ps interval per each 100 K, and
then equilibrated for 50 ps at 310 K, followed by
400 ps of data collection runs, necessary for the estima-
tion of the free energy of binding (vide infra). The MD
simulations were performed at constant T = 310 K using
the Berendsen et al. coupling algorithm62 with separate
For the docking of each compound to the protein, three
hundred Monte Carlo/Simulated Annealing (MC/SA)
runs were performed, with 100 constant temperature cy-
cles for simulated annealing. For these calculations, the

M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
2603
coupling of the solute and solvent to heat, an integration
time step of 2 fs, and the applications of the shake algo-
rithm63 to constrain all bonds to their equilibrium val-
ues, thus removing high frequency vibrations. Long-
range non-bonded interactions were truncated by using
in Eq. 4 is equal to zero. The Poisson–Boltzmann (PB)
calculations of DGPB were done with the Delphi pack-
age,67 with interior and exterior dielectric constants
equal to 1 and 80, respectively. A grid spacing of
˚
2/A, extending 20% beyond the dimensions of the sol-
ute, was employed. The non-polar component DGNP
˚
a dual cutoff of 9 and 13 A, respectively, where energies
˚
and forces due to interactions between 9 and 13 A were
was obtained using the following relationship68:
2
˚
updated every 20 time step. The same frequency of up-
date was employed for the non-bonded list. For the cal-
culation of the binding free energy between the RdRp
and each compound in water, a total of 400 snapshots
were saved during the MD data collection period de-
scribed above, one snapshot per each 1 ps of MD
simulation.
DGNP = cSA + b, in which c = 0.00542 kcal/(mol A ),
b = 0.92 kcal/mol, and the surface area was estimated
by means of the MSMS software.69 The last parameter
in Eq. 1, that is, the change in solute entropy upon
association—TDSsolute, was calculated through nor-
mal-mode analysis.70 In the first step of this calcula-
˚
tion, an 8 A sphere around the ligand was cut out
from an MD snapshot for each ligand–protein com-
plex. This value was shown to be large enough to
yield converged mean changes in solute entropy. On
the basis of the size-reduced snapshots of the complex,
we generated structures of the uncomplexed reactants
by removing the atoms of the protein and ligand,
respectively. Each of these structures was minimized,
using a distance-dependent dielectric constant e = 4r,
to account for solvent screening, and its entropy was
calculated using classical statistical formulas and nor-
mal-mode-analysis. To minimize the effects due to dif-
ferent conformations adopted by individual snapshots,
we averaged the estimation of entropy over 10
snapshots.
The binding free energy DGbind of each RdRp/drug
complex in water was calculated according to the pro-
cedure termed Molecular Mechanic/Poisson–Boltz-
mann Surface Area (MM/PBSA), and proposed by
Srinivasan et al.64 Since the theoretical background
of this methodology is described in detail in the origi-
nal papers by Peter Kollman and his group,65 it will
be only briefly described below. Basically, an MD sim-
ulation (typically in explicit solvent) is first carried out
which yields a representative ensemble of structures.
The average total free energy of the system, G, is then
evaluated as:
G ¼ Gsolv þ EMM ꢀ TSsolute
¼ GPB þ GNP þ EMM ꢀ TSsolute
Finally, the IC50 values were calculated from the corre-
sponding binding free energies using the following
ð1Þ
relationship71,72
:
in which GPB is the polar solvation energy component,
which is calculated in a continuum solvent, usually a fi-
nite-difference Poisson–Boltzmann (PB) model, and
GNP is the non-polar contribution to the solvation en-
ergy, which can be obtained from the solvent accessible
surface area (SA). EMM denotes the sum of molecular
mechanics (MM) energies of the molecule, and can be
further split into contributions from electrostatic (EEL),
van der Waals (EvdW), and internal (EINT) energies:
DGbind ¼ RT ln Kdiss ¼ RT lnðIC50 þ 0:50Cenz
ffi RT ln IC50
Þ
ð6Þ
The overall quality of the entire procedure described
above—that is, protein and inhibitors modeling, con-
formational analysis, docking and energetic calcula-
tions—has been previously tested by carrying out the
same calculations on three known RdRp inhibi-
tors,73,74 for which both the crystallographic structures
of the relevant HCV RdRp complexes and the corre-
sponding IC50 values were available, and for a new
series of nucleoside analogs inhibitors of the HCV
RdRp.75
EMM ¼ EEL þ EVDW þ EINT
ð2Þ
The last term in Eq. 1, TSsolute, represents the solute en-
tropy, and is usually estimated by a combination of clas-
sical statistical formulas and normal-mode analysis.
Using Eqs. 1 and 2, the binding free energy DGbind of a
given non-covalent association can be obtained as:
Acknowledgment
DGbind ¼ Gcomplex ꢀ ðGprotein þ Gligand
Þ
ð3Þ
Accordingly, the variation of each contribution to
This work was supported by grants from FIRB 2001
(Italian Foundation for Basic Research).
DGbind is given by:
DEMM ¼ DEEL þ DEVDW þ DEINT
DGsolv ¼ DGPB þ DGNP
The ensemble of structures for the uncomplexed reac-
tants is generated either by running separate MD sim-
ulations, or by using the trajectory of the complex,
simply by removing the protein or ligand atoms. As
reported below, in this work we followed the success-
ful approach proposed by Kuhn and Kollman66 and
applied the latter variant. Accordingly, the term DEINT
ð4Þ
ð5Þ
References and notes
1. Fang, J. W.; Chow, V.; Lau, J. Y. Clin. Liver Dis. 1997, 1,
493–514.
2. Thomson, B. J.; Finch, R. G. Clin. Microbiol. Infect. 2005,
11, 86–94.
3. Liang, T. J.; Heller, T. Gastroenterology 2004, 127(Suppl.
1), S62–S71.
4. Shan, Y.; Chen, X. G.; Huang, B.; Hu, A. B.; Xiao, D.;
Guo, Z. M. Liver Int. 2005, 25, 141–147.

2604
M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
5. Scott, L. J.; Perry, C. M. Drugs 2002, 62, 507–556.
25. Mazzei, M.; Dondero, R.; Sottofattori, E.; Melloni, E.;
Minafra, R. Eur. J. Med. Chem. 2001, 36, 851–861.
26. Egan, D.; O’Kennedy, E.; Moran, E.; Cox, D.; Prosser, E.;
Thornes, R. D. Drug. Metab. Rev. 1990, 22, 503–529.
27. Kontogiorgis, C. A.; Hadjipavlou-Litina, D. J. J. Med.
Chem. 2005, 48, 6400–6408.
28. Nofal, Z. M.; el-Zahar, M. I.; Abd El-Karim, S. S.
Molecules 2000, 5, 99–113.
29. Galabov, A. S.; Iosifova, T.; Vassileva, E.; Kostova, I. Z.
Naturforsch., C 1996, 51, 558–562.
30. Al-Mawsawi, L. Q.; Fikkert, V.; Dayam, R.; Witvrouw,
M., ; Burke, T. R., Jr.; Borchers, C. H.; Neamati, N. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 10080–10085.
31. Wu, J.; Yang, Z.; Fathi, R.; Zhu, Q. US Patent
2004180950, 2004, January 06.
6. Raptopoulou, M.; Tsantoulas, D.; Vafiadi, I.; Ketikoglou,
I.; Paraskevas, E.; Vassiliadis, T.; Kanatakis, S.; Hatzis,
G.; Sidiropoulos, L.; Akriviadis, E. J. Viral Hepat. 2005,
12, 91–95.
7. Pietschmann, T.; Bartenschlager, R. Curr. Opin. Drug
Discov. Dev. 2001, 4, 657–664.
8. Zhong, J.; Gastaminza, P.; Cheng, G.; Kapadia, S.; Kato,
T.; Burton, D. R.; Wieland, S. F.; Uprichard, S. L.;
Wakita, T.; Chisari, F. V. Proc. Natl. Acad. Sci. U.S.A.
2005, 9294–9299.
9. Lindenbach, B. D.; Evans, M. J.; Syder, A. J.; Wolk, B.;
Tellinghuisen, T. L.; Liu, C. C.; Maruyama, T.; Hynes, R.
O.; Burton, D. R.; McKeating, J. A.; Rice, C. M. Science
2005, 623–626.
10. Dymock, B. W.; Jones, P. S.; Wilson, F. X. Antiviral
Chem. Chemother. 2000, 11, 79–96.
11. Beaulieu, P. L.; Tsantrizos, Y. S. Curr. Opin. Invest. Drugs
2004, 5, 838–850.
32. Xu, B.; Zhu, Q.; Cho, H. J.; Fathi, R.; Yang, Z.;
Sandresagra; A., Liu, Y. US Patent 2006223783, 2005,
March 29.
33. Okamoto, T.; Kajino, K.; Hino, O. Jpn. J. Pharmacol.
2001, 87, 177–180.
34. Okamoto, T.; Kobayashi, T.; Yoshida, S. Curr. Med.
Chem. Anticancer Agents 2005, 5, 47–51.
35. Okamoto, T.; Kobayashi, T.; Yoshida, S. Med. Chem.
2007, 3, 35–44.
36. Mazzei, M.; Miele, M.; Nieddu, E.; Barbieri, F.; Bruzzo,
C.; Alama, A. Eur. J. Med. Chem. 2001, 36, 915–923.
37. Gupta, V. N.; Sharima, B. R.; Avora, R. B. J. Sci. Ind.
Res. 1961, 20B, 300.
38. Pauwels, R.; Balzarini, J.; Baba, M.; Snoeck, R.; Schols,
D.; Herdewijn, P.; Desmyster, J.; De Clercq, E. J. Virol.
Methods 1988, 20, 309–321.
39. Bressanelli, S.; Tomei, L.; Rey, F. A.; De Francesco, R.
J. Virol. 2002, 76, 3482–3492.
40. Pearlman, D. A.; Case, D. A.; Caldwell, J. W.; Ross, W.
S., ; Cheatham, T. E., III; DeBolt, S.; Ferguson, D.;
Seibel, G.; Kollman, P. A. Comput. Phys. Commun. 1995,
91, 1–41.
41. Case, D. A.; Pearlman, D. A.; Caldwell, J. W., Cheatham,
T. E.; Wang, J.; Ross, W. S.; Simmerling, C.; Darden, T.
A.; Merz, K. M.; Stanton, R. V.; Cheng, A. L.; Vincent, J.
J.; Crowley, M.; Tsui, V.; Gohlke, H.; Radmer, R. J.;
Duan, Y.; Pitera, J.; Massova, I.; Seibel, G. L.; Singh, U.
C.; Weiner, P. K.; Kollman, P. A. AMBER 7, University
of California, San Francisco, CA USA, 2000.
42. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.;
Merz, K. M.; Ferguson, D. M.; Spellmeyer, D. C.; Fox,
T.; Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc.
1995, 117, 5179–5197.
43. Onufriev, A.; Bashford, D.; Case, D. A. J. Phys. Chem. B
2000, 104, 3712–3720.
44. Feig, M.; Onufriev, A.; Lee, M.; Im, W.; Case, D. A.;
Brooks, C. L., III J. Comput. Chem. 2004, 25, 265–284.
45. Fermeglia, M.; Ferrone, M.; Pricl, S. Bioorg. Med. Chem.
2002, 10, 2471–2478.
12. Sulkowski, M. S. Curr. Gastroenterol. Rep. 2007, 9, 5–13.
13. Toniutto, P.; Fabris, C.; Bitetto, D.; Fornasiere, E.;
Rapetti, R.; Pirisi, M. Discov. Med. 2007, 7, 27–32.
14. Olsen, D. B.; Eldrup, A. B.; Bartholomew, L.; Bhat, B.;
Bosserman, M. R.; Ceccacci, A.; Colwell, L. F.; Fay, J. F.;
Flores, O. A.; Getty, K. L.; Grobler, J. A.; LaFemina, R.
L.; Markel, E. J.; Migliaccio, G.; Prhavc, M.; Stahlhut, M.
W.; Tomassini, J. E.; MacCoss, M.; Hazuda, D. J.;
Carroll, S. S. Antimicrob. Agents Chemother. 2004, 48,
3944–3953.
15. Sperandio, D.; Gangloff, A. R.; Litvak, J.; Goldsmith, R.;
Hataye, J. M.; Wang, V. R.; Shelton, E. J.; Elrod, K.;
Janc, J. W.; Clark, J. M.; Rice, K.; Weinheimer, S.;
Yeung, K. S.; Meanwell, N. A.; Hernandez, D.; Staab, A.
J.; Venables, B. L.; Spencer, J. R. Bioorg. Med. Chem.
Lett. 2002, 12, 3129–3133.
16. Dhanak, D.; Duffy, K. J.; Johnston, V. K.; Lin-Goerke, J.;
Darcy, M.; Shaw, A. N.; Gu, B.; Silverman, C.; Gates, A.
T.; Nonnemacher, M. R.; Earnshaw, D. L.; Casper, D. J.;
Kaura, A.; Baker, A.; Greenwood, C.; Gutshall, L. L.;
Maley, D.; Del Vecchio, A.; Macarron, R.; Hofmann, G.
A.; Alnoah, Z.; Cheng, H.-Y.; Chan, G.; Khandekar, S.;
Keenan, R. M.; Sarisky, R. T. J. Biol. Chem. 2002, 277,
38322–38327.
17. Gu, B.; Johnston, V. K.; Gutshall, L. L.; Nguyen, T. T.;
Gontarek, R. R.; Darcy, M. G.; Tedesco, R.; Dhanak, D.;
Duffy, K. J.; Kao, C. C.; Sarisky, R. T. J. Biol. Chem.
2003, 278, 16602–16607.
18. Be aulieu, P. L.; Bos, M.; Bousquet, Y.; DeRoy, P.; Fazal,
G.; Gauthier, J.; Gillard, J.; Goulet, S.; McKercher, G.;
Poupart, M. A.; Valois, S.; Kukolj, G. Bioorg. Med.
Chem. Lett. 2004, 14, 967–971.
19. McKercher, G.; Beaulieu, P. L.; Lamarre, D.; La Plante,
S.; Lefebvre, S.; Pellerin, C.; Thauvette, L.; Kukolj, G.
Nucleic Acids Res. 2004, 32, 422–431.
20. Llinas-Brunet, M.; Bailey, M. D.; Bolger, G.; Brochu, C.;
Faucher, A. M.; Ferland, J. M.; Garneau, M.; Ghiro, E.;
Gorys, V.; Grand-Maitre, C.; Halmos, T.; Lapeyre-
Paquette, N.; Liard, F.; Poirier, M.; Rheaume, M.;
Tsantrizos, Y. S.; Lamarre, D. J. Med. Chem. 2004, 47,
1605–1608.
46. Felluga, F.; Pitacco, G.; Valentin, E.; Coslanich, A.;
Fermeglia, M.; Ferrone, M.; Pricl, S. Tetrahedron Asym-
metry 2003, 14, 3385–3399.
47. Manfredini, M.; Solaroli, N.; Angusti, A.; Nalin, F.;
Durini, E.; Vertuani, S.; Pricl, S.; Ferrone, M.; Spadari, S.;
Focher, F.; Verri, A.; De Clercq, E.; Balzarini, J. Antiviral
Chem. Chemother. 2003, 14, 183–194.
48. Metullio, L.; Coslanich, A.; Fermeglia, M.; Ferrone, M.;
Fuchs, S.; Paneni, M. S.; Pricl, S. Biomacromolecules 2004,
5, 1371–1378.
49. Bayly, C. I.; Cieplak, P.; Cornell, W. D.; Kollman, P. A.
J. Phys. Chem. 1993, 97, 10269–10280.
21. Zhang, R.; Durkin, J. P.; Windsor, W. T. Bioorg. Med.
Chem. Lett. 2002, 12, 1005–1008.
22. Soler, M.; McHutchison, J. G.; Kwoh, T. J.; Dorr, F. A.;
Pawlotsky, J. M. Antiviral Ther. 2004, 9, 953–968.
23. Mazzei, M.; Sottofattori, E.; Dondero, R.; Ibrahim, M.;
Melloni, E.; Michetti, M. Il Farmaco 1999, 54, 452–460.
24. Mazzei, M.; Miele, M.; Nieddu, E.; Bruzzo, C.; Barbieri,
F.; Alama, A. Eur. J. Med. Chem. 2001, 36, 915–923.
50. Singh, U. C.; Kollman, P. A. J. Comput. Chem. 1984, 5,
129–145.

M. Mazzei et al. / Bioorg. Med. Chem. 16 (2008) 2591–2605
2605
51. Besler, B. H.; Merz, K. M.; Kollman, P. A. J. Comput.
Chem. 1990, 11, 431–439.
52. Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J.
J. P. J. Am. Chem. Soc. 1985, 107, 3902–3909.
53. Huo, S.; Wang, J.; Cieplak, P.; Kollman, P. A.; Kuntz, I.
D. J. Med. Chem. 2002, 45, 1412–1419.
Donini, O., ; Cieplak, P.; Srinivasan, J.; Case, D. A.;
Cheatham, T. E., III Acc. Chem. Res. 2000, 33, 889–897.
66. Kuhn, B.; Kollman, P. A. J. Med. Chem. 2000, 43, 3786–
3791.
67. Gilson, M. K.; Sharp, K. A.; Honig, B. H. J. Comput.
Chem. 1987, 9, 327–335.
54. Wang, J.; Cieplak, P.; Kollman, P. A. J. Comput. Chem.
2000, 21, 1049–1074.
68. Sitkoff, D.; Sharp, K. A.; Honig, B. H. J. Phys. Chem.
1994, 98, 1978–1988.
55. Wang, J.; Wang, W.; Kollman, P. A.; Case, D. A.
J. Comput. Chem. 2005, 25, 1157–1174.
69. Sanner, M. F.; Olson, A. J.; Spehner, J. C. Biopolymers
1996, 38, 305–320.
56. Felluga, F.; Fermeglia, M.; Ferrone, M.; Pitacco, G.;
Pricl, S.; Valentin, E. Tetrahedron Asymmetry 2002, 13,
475–489.
57. Mamolo, M. G.; Zampieri, D.; Vio, L.; Fermeglia, M.;
Ferrone, M.; Pricl, S.; Scialino, G.; Banfi, E. Bioorg. Med.
Chem. 2005, 13, 3797–3809.
70. Case, D. A. Curr. Opin. Struct. Biol. 1994, 4, 285–290.
71. Kroeger Smith, M. B.; Lamb, M. L.; Tirado-Rives, J.;
Jorgensen, W. L.; Michejda, C. J.; Ruby, S. K.; Smith, R.
H., III Protein Eng. 2000, 13, 413–421.
72. Wang, J.; Morin, P.; Wang, W.; Kollman, P. A. J. Am.
Chem. Soc. 2001, 123, 5221–5230.
58. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.;
Hart, W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem.
1998, 19, 1639–1662.
73. Wang, M.; Ng, K. K.; Cherney, M. M.; Chan, L.;
Yannopoulos, C. G.; Bedard, J.; Morin, N.; Nguyen-Ba,
N.; Alaoui-Ismaili, M. H.; Bethell, R. C.; James, M. N.
J. Biol. Chem. 2003, 278, 9489–9495.
74. Love, R. A.; Parge, H. E.; Yu, X.; Hickey, M. J.; Diehl,
W.; Gao, J.; Wriggers, H.; Ekker, A.; Wang, L.; Thomson,
J. A.; Dragovich, P. S.; Fuhrman, S. A. J. Virol. 2003, 77,
7575–7581.
75. Pricl, S.; Manfredini, S.; Angusti, A.; Ciliberti, N.; Durini,
E.; Vertuani, S.; Buzzoni, L.; Coslanich, A.; Ferrone, M.;
Fermeglia, M.; Paneni, M. S.; La Colla, P.; Sanna, G.;
Cadeddu, A.; Mura, M.; Loddo, R. Hindered Nucleoside
Analogs as Inhibitors of HCV RNA-Dependent RNA-
Polymerase: Evolving Vistas. In Framing the Knowledge of
Viral Hepatitis; Schinazi, R. F., Schiff, E. R., Eds.; IHL
Press: Cambridge MA, USA, 2005; pp 279–318, ISBN:
978-0-9754188-1-9.
59. Mehler, E. L.; Solmajer, T. Protein Eng. 1991, 4, 903–910.
60. Frecer, V.; Kabelac, M.; De Nardi, P.; Pricl, S.; Miertus,
S. J. Mol. Graphics Modell. 2004, 22, 209–220.
61. Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey,
R. W.; Klein, M. L. J. Chem. Phys. 1983, 79, 926–935.
62. Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W.
F.; Di Nola, A.; Haak, J. R. J. Comput. Phys. 1984, 81,
3684–3690.
63. Ryckaert, J. P.; Ciccotti, G.; Berendsen, H. J. C.
J. Comput. Phys. 1977, 23, 327–341.
64. Srinivasan, J.; Cheatham, T. E., III; Cieplak, P.; Kollman,
P. A.; Case, D. A. J. Am. Chem. Soc. 1998, 120, 9401–9409.
65. Kollman, P. A.; Massova, I.; Reyes, C.; Kuhn, B.; Huo,
S.; Chong, L.; Lee, M.; Lee, T.; Duan, Y.; Wang, W.;