X-ray crystal structures of Enterococcus faecalis thymidylate synthase with folate binding site inhibitors

Article abstract of DOI:10.1016/j.ejmech.2016.07.066

Infections caused by Enterococcus faecalis (Ef) represent nowadays a relevant health problem. We selected Thymidylate synthase (TS) from this organism as a potential specific target for antibacterial therapy. We have previously demonstrated that species-specific inhibition of the protein can be achieved despite the relatively high structural similarity among bacterial TSs and human TS. We had previously obtained the EfTS crystal structure of the protein in complex with the metabolite 5-formyl-tetrahydrofolate (5-FTHF) suggesting the protein role as metabolite reservoir; however, protein–inhibitors complexes were still missing. In the present work we identified some inhibitors bearing the phthalimidic core from our in-house library and we performed crystallographic screening towards EfTS. We obtained two X-ray crystallographic structures: the first with a weak phthalimidic inhibitor bound in one subunit and 5-hydroxymethylene-6-hydrofolic acid (5-HMHF) in the other subunit; a second X-ray structure complex with methotrexate. The structural information achieved confirm the role of EfTS as an enzyme involved in the folate pool system and provide a structural basis for structure-based drug design.

Full text of DOI:10.1016/j.ejmech.2016.07.066

European Journal of Medicinal Chemistry 123 (2016) 649e664
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
X-ray crystal structures of Enterococcus faecalis thymidylate synthase
with folate binding site inhibitors
,
,
Alessia Catalano a 1, Rosaria Luciani b 1, Alessia Carocci a, Debora Cortesi b, Cecilia Pozzi c,
Chiara Borsari b, Stefania Ferrari b **, Stefano Mangani c
Dipartimento di Farmacia-Scienze del Farmaco, Universita degli Studi di Bari “Aldo Moro”, Via Orabona 4, 70125 Bari, Italy
Dipartimento di Scienze della Vita, Universita degli Studi di Modena e Reggio Emilia, Via Campi 103, 41125 Modena, Italy
Dipartimento di Biotecnologie, Chimica e Farmacia, Universita degli Studi di Siena, Via Aldo Moro 2, 53100 Siena, Italy
,
, *
a
ꢀ
b
c
ꢀ
ꢀ
a r t i c l e i n f o
a b s t r a c t
Article history:
Infections caused by Enterococcus faecalis (Ef) represent nowadays a relevant health problem. We
selected Thymidylate synthase (TS) from this organism as a potential specific target for antibacterial
therapy. We have previously demonstrated that species-specific inhibition of the protein can be achieved
despite the relatively high structural similarity among bacterial TSs and human TS. We had previously
obtained the EfTS crystal structure of the protein in complex with the metabolite 5-formyl-tetrahy-
drofolate (5-FTHF) suggesting the protein role as metabolite reservoir; however, proteineinhibitors
complexes were still missing. In the present work we identified some inhibitors bearing the phthalimidic
core from our in-house library and we performed crystallographic screening towards EfTS. We obtained
two X-ray crystallographic structures: the first with a weak phthalimidic inhibitor bound in one subunit
and 5-hydroxymethylene-6-hydrofolic acid (5-HMHF) in the other subunit; a second X-ray structure
complex with methotrexate. The structural information achieved confirm the role of EfTS as an enzyme
involved in the folate pool system and provide a structural basis for structure-based drug design.
© 2016 Elsevier Masson SAS. All rights reserved.
Received 26 April 2016
Received in revised form
25 July 2016
Accepted 26 July 2016
Available online 29 July 2016
Keywords:
Enterococcus faecalis
Enzyme inhibition
Thymidylate synthase
X-ray crystallography screening
Phthalimide scaffold
Antibacterial
1. Introduction
present in more than 90% of clinical isolates. Enterococcus faecalis
accounts for 80e90% of clinical strains, while Enterococcus faecium
Enterococci can be found in soil, food and water and make up a
significant portion of the normal gut flora of humans and animals.
As for other bacteria of the gut flora, enterococci can also cause
infectious diseases. Typical enterococcal infections occur in
immunosuppressed patients and in hospitalized patients suffering
from a wide spectrum of severe illnesses. Enterococci nowadays
rank second to third in frequency among bacteria isolated from
hospitalized patients. They can be often isolated from urinary tract
infections, intra-abdominal and pelvic infections, bacteremias,
wounds, tissue infections and endocarditis as part of a poly-
microbial flora. Enterococcus faecalis (Ef) and Enterococcus faecium
are the most prevalent species cultured from humans and they are
for only 5e10%. Some strains of E. faecalis and many of E. faecium are
resistant to multiple antimicrobials [1]. The identification of new
antibiotics targeting Enterococci infections is challenging since no
specific antibacterial agents are known so far. This encourages the
starting of active drug discovery programs. Molecules targeting the
Enterococci folate pathway represent a source of potential and
specific drugs able to inhibit the bacterial folate pathway without
altering, in principle, the mammalian cells. In our antibacterial
discovery programs we have identified some lead compounds
bearing a phthalimide moiety targeting bacterial Lactobacillus casei
(Lc) and Escherichia coli (Ec) thymidylate synthase (TS) and able to
discriminate between the bacterial and the host (human) protein.
This selectivity led to the discovery of antibacterial agents with a
low toxicity [2e10]. TS (EC: 2.1.1.45) catalyzes the reductive
methylation of 20-deoxyuridine-50-monophosphate (dUMP) to 20-
deoxythymidine-50-monophosphate (dTMP), assisted by the
cofactor N5,N10-methylene tetrahydrofolate (MTHF) [11,12]. TS is
the only synthetic source of dTMP in human cells, thus it represents
a major target for the design of chemotherapeutic agents [13]. The
* Corresponding author.
** Corresponding author.
E-mail addresses: stefania.ferrari@unimore.it (S. Ferrari), stefano.mangani@
unisi.it (S. Mangani).
1
Alessia Catalano and Rosaria Luciani contribute equally to the work. The work
was written through contributions of all authors.
http://dx.doi.org/10.1016/j.ejmech.2016.07.066
0223-5234/© 2016 Elsevier Masson SAS. All rights reserved.

650
A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
well-known anticancer agents, 5-fluorouracil (the prodrug of 5-
fluoro-20-deoxyuridine monophosphate), methotrexate (MTX)
and raltitrexed (Fig. 1), are classical TS inhibitors belonging to the
the racemic mixture and the single enantiomers.
2.2. Chemistry
antimetabolite class. In our previous work, compound a156 (Fig. 1)
had been proposed as an antibacterial lead able to face vancomycin-
resistant infections caused by gram positive bacteria such as
Staphylococcus aureus. This molecule also inhibited the Entero-
coccus faecalis clinical isolates and showed a low toxicity towards
human cells [7]. In a parallel work we had identified a library of
phthalimidic derivatives, 4-methylisoindoline-1,3-dione N-de-
rivatives (I, Fig. 1) and 4-(aminomethyl)-2-methylisoindoline-1,3-
dione derivatives substituted at the amino position (II, Fig. 1),
that presented inhibitory activity against bacterial TS enzymes.
These phthalimidic derivatives inhibited also Enterococcus faecalis
TS (EfTS), with the best inhibitors showing Ki in the range of
Only synthetic procedures followed to obtain compounds not
previously characterized in literature are described in this paper.
The synthesis of the racemic mixture and of the enantiomers of
compound 8 is shown in Scheme 1. Ester (RS)-60 was obtained
following Williamson reaction of (RS)-bromopropanoate, (RS)-58
with 2,6-dimethylphenol, while (R)- and (S)-61 were obtained by
Mitsunobu reaction [34,35] of 2,6-dimethylphenol with (S)- or (R)-
methyl lactate [(S)- or (R)-59], respectively. After esters reduction,
the obtained alcohols 61 were submitted to Mitsunobu reaction
with phthalimide to give the desired compounds. Compound 13
was obtained through hydrolysis of compound 12 (Scheme 2). The
synthesis of compounds (S)-17, (S)- and (R)-33 is depicted in
Scheme 3. (S)- and (R)-33 were obtained by protecting (S)- and (R)-
2-amino-2-phenylethanol, respectively, with phthalic anhydride.
Iodination of compound (S)-33 yielded compound (S)-17. Com-
pounds 20 and 22 were synthesized as reported in Scheme 4. Cyano
derivative 18 was converted into the corresponding trime-
thyltintetrazole derivative 62 by reaction with azido-trimethyltin.
The treatment of compound 62 with gaseous HCl gave tetrazole
derivative 22, which was alkylated with methyl iodide to give
compound 20. Compound 24 was obtained by reacting 2-
(methoxycarbonyl)benzoic acid with p-toluidine in presence of
EEDQ as previously described (Scheme 5) [36]. The nitration of (R)-
63, followed by reduction of the nitro group gave (R)-7 and, suc-
cessively, (R)-43 (Scheme 6), as reported in the literature [26,37,38].
Compounds 25 and 55 were prepared by protecting 4-
2e8 mM. These compounds exhibited a competitive behavior with
respect to the folate cofactor and a significant selectivity towards
bacterial TS [8]. EfTS X-ray crystal structure of the native protein
had been obtained [14]. According to structural analysis, the protein
crystallized in a ternary complex where a folate substrate analogue,
5-formyl-tetrahydrofolate (5-FTHF), originating from the bacterial
metabolites pool, co-crystallized with the enzyme. Thus, we can
conclude that EfTS could be considered a reservoir for folate sub-
strates in the bacteria folate dependent metabolism. Although no
structural information about EfTS-inhibitors complexes are avail-
able yet, we could suppose that the target druggability is influenced
by the presence of the folate metabolites inside the bacterial cell
and that higher affinity inhibitors are necessary to displace 5-FTHF.
Aiming to obtain X-ray crystallographic structures to gain addi-
tional information on EfTS target druggability and to carry out
further structure-based drug design studies, we analyzed our in-
house library based on a N-substituted phthalimide scaffold
bearing a flexible chain linked to the nitrogen in position 2. In de-
tails, 56 compounds (Charts 1 and 2) were tested against EfTS, EcTS
and human TS (hTS). In the present work, X-ray crystallographic
screenings were carried out and one X-ray crystallography complex
with a protein inhibitor, (S)-12, was obtained. In addition, we solved
the X-ray crystal structure of EfTS in complex with methotrexate
(MTX), a known folate analogue. These studies represent an
important starting point for a structure-based drug design aimed at
identifying new antibacterial agents against Enterococcus faecalis
infections.
phenylbutan-1-amine
with
phthalimide
or
tetra-
chlorophthalimide, respectively (Scheme 7). The synthesis of
compound 52 is reported in Scheme 8. 3-aminopropanol was
protected with phthalic anhydride to give compound 64, which
was submitted to Mitsunobu reaction with 1-naphthol. Compound
65 (Scheme 9) was synthesized by the reaction of 3-aminopropanol
with tetrachlorophthalic anhydride and converted into compound
53. The synthesis of compound 54 is depicted in Scheme 10. Wil-
liamson reaction on 3-bromopropanol with thiophenol gave com-
pound 67 which was converted into the thioether 55. Finally, the
synthesis of (R)-1 was obtained via a Mitsunobu reaction starting
from (S)-68 (Scheme 11).
2. Results and discussion
2.3. Biological evaluation of the compounds library
2.1. Library selection and compound synthesis
The selected 56 compounds were tested against EfTS and hTS
enzymes in order to evaluate the inhibitory activity towards TS and
the specificity towards the bacterial enzyme (Fig. 2, Table S1, see
Supplementary Material). The percentage of inhibition (I %) at
With the aim of exploring the chemical space of the phthalimide
core, a library of 56 compounds (Charts 1 and 2) was selected from
our in-house libraries (for details see experimental section), pre-
viously synthesized as intermediates for the preparation of com-
pounds with different pharmacological activities [15e33]. These
compounds underwent in vitro screening towards EfTS, hTS and
EcTS. All the molecules of the new library show an N-substituted
phthalimide ring (III, Fig. 1). The majority of the compounds have
an aromatic ring in the side-chain linked to position 2 of the
phthalimide moiety. In our previously published library the aro-
matic moiety in the side-chain was directly linked to the nitrogen in
position 2 (I, Fig. 1), whereas in the present work almost all the
compounds have a spacer between the phthalimide core and the
aromatic system of the side-chain, thus being more flexible. Few
compounds (27e31, 34, 56) have a non-aromatic side-chain with a
lower steric hindrance. This novel phthalimide library includes
several chiral compounds (1e8, 12, 13, 17, 21, 27e34, 40e43, 50, 51,
56) and for some of them we explored the biological profile of both
100 mM are reported as heat-map plot in Fig. 2 in which different
color codes represent the inhibition range of the compounds.
Moreover, the inhibitory activity towards Escherichia coli TS (EcTS)
was assessed, aiming to explore a larger biological activity profile.
Several compounds of the library are chiral, thus they were tested
as racemic mixture and/or as homochiral. Thirteen compounds (4,
7, 11, 18, 20, 23, 35, 39, 40, 45, 47, 52 and 54) show inhibitory ac-
tivity towards EfTS (I % ¼ 20e60% at 100
40 and 54, with Ki of 25,13 and 24 M, respectively, turned out to be
the best inhibitors of the library. Moreover, they were species-
specific inhibitors of EfTS, since they were inactive at 100 M to-
wards hTS. Few compounds (RS)-1, (R)-7, (R)-8, 9, 14,16, 24, 25 were
able to decrease the EcTS activity by 35e48% at 100 M and
mM). Compounds 18, (R)-
m
m
m
represent the most potent EcTS inhibitors of this set of compounds.
In our previous paper [8], we studied the inhibitory activity to-
wards EfTS of more rigid phthalimide derivatives (6A, 8A, 12A,

A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
651
Fig. 1. Chemical structure of TS inhibitors and the prodrug 5FU.
Fig. 1). The phthalimide moiety had been directly connected to
substituted aromatic or heteroaromatic rings in position 2. Com-
pound 6A, 8A and 12A showed a significant EfTS inhibitory activity,
inhibitor bound to EfTS was complete and confidently interpreted
(PDB: 4O7U) and compared with that of EfTS-MTX complex (PDB:
5J7W). MTX has a Ki equal to that of the best phtalimidic inhibitor of
the series, compound (R)-40. The two independent EfTS dimers
present in the crystal asymmetric unit of the EfTS-(S)-12 complex
(Fig. 3A) display subunit heterogeneity: one subunit is empty and in
the ‘open’ conformation [14], while the other one hosts the exog-
enous ligand and is in a ‘closed’ conformation (Fig. 3B). The ‘open’
conformation is characterized by the loop 92e102 being open and
mostly disordered and by the short helix 103e109 and the first two
turns of helix 111e130 being completely unfolded. In the ‘closed’
conformation the secondary structure elements described above
are fully ordered with the loop closer to the active site cavity (see
Fig. 3B). Further heterogeneity is present in the ‘closed’ subunits as
they bind two different molecules into their active sites. We
with compound 12A being the most active (Ki ¼ 2.0 0.35
mM). In
the present study, we evaluated the activity of more flexible mol-
ecules. The spacer between the phthalimide moiety and the aro-
matic side chain was from 2 to 10 carbons long (Fig. 1, III). In most
cases, an ether or a thioether spacer was used. If we compare the
overall inhibitory activity of the previously published library with
that of the present library, we can state that more rigid compounds
are preferable with respect to flexible molecules, for EfTS binding.
2.4. X-ray crystal structures of EfTS in complex with (S)-12 and
with MTX
observed
a
mixture of two different complexes: 5-
X-ray crystallographic studies were carried out on structurally
different N-substituted phthalimides. Overall, the present library
showed medium to low affinity towards EfTS. All the compounds
showing a percentage of inhibition towards EfTS of 10%e60% were
selected for the crystallographic screening. From the screened
crystals, we obtained usable diffraction data only for EfTS com-
plexes with (R)-2, (S)-12 and 14. Unfortunately, the Fourier differ-
ence maps for (R)-2 and 14 complexes showed only blurred
features corresponding to the bound inhibitor into the active site
that prevented model building. In addition, we also obtained a
hydroxymethylene-6-hydrofolic acid (5-HMHF) is found in one
dimer (Fig. 4A), while the other dimer hosts the S-enantiomer of
compound (RS)-12 used for co-crystallization (Fig. 4B). Subunit
heterogeneity and half-site reactivity is commonly found in TS
enzymes [39e41]. The electron density present in the active site of
the
B subunit led to identify the folate analogue as 5-
hydroxymethylene-6-hydrofolate molecule (5-HMHF), thus
different from the 5-formyltetrahydrofolate (5-FTHF) already pub-
lished [14]. The active site cavity of the B subunit shows electron
density above noise, which is consistent with the planar structure
of the pteridine moiety and in contrast with the puckered pyrazine
dataset for EfTS bound by MTX (Ki ¼ 13
mM) that has been used for
comparison. Eventually, only the electron density for the (S)-12

652
A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
Compd
R
Compd
R
Compd
R
Cl
1
2
3
O
O
Cl
O
4
7
5
8
6
9
O
O
O
O
N
Cl
O
O
O
O
O
O
Cl
Cl
Cl
O
10
13
11
14
12
15
O
Cl
O
OH
O
Cl
O
O
Chart 1. Compounds 1e52.
ring of tetrahydrofolate found in our previous crystal structure of
EfTS [14] (Fig. 4A). The remaining parts of the molecule establish
interactions similar to those of 5-FTHF. We have previously shown
that EfTS obtained from our preparations contains bound metabo-
lites from the folate pathway, such as 5-formyl-tetrahydrofolate (5-
FTHF; PDB: 3UWl) [14]. 5-HMHF is a product of the conversion of 5-
FTHF by bacterial enzymes (e.g. 5,10-methenyl-tetrahydrofolate
synthetase enzyme). In fact, 5-HMHF was discovered for the first
time bound to 5,10-methenyl-tetrahydrofolate synthetase from
Mycoplasma pneumoniae (PDB 1U3G) [42]. Analyzing the binding
mode of (S)-12, we can suggest that this compound displaces 5-
HMHF (or other folate structural analogues), at least partially,
from EfTS active site. This finding provides structural evidence of
the competitive inhibition pattern of compound (S)-12 with respect
to MTHF. The orientation of compound (S)-12 into EfTS active site is
similar to that of 5-FTHF and 5-HMHF (Fig. 4AeB). Compound (S)-
12 establishes interactions with the hydrophobic residues sur-
rounding the active site (Ile80, Trp81, Trp84, Leu194, Cys197,
Leu223, Phe227, Tyr260 and Ile313). Only one hydrogen bond be-
tween the carbonyl group of compound (S)-12 and a water mole-
cule is formed. The phthalimide core of compound (S)-12 places
itself in the same position occupied by the pteridine moiety of the
5-HMHF and the para-substituted aromatic chain replaces the PABA
moiety of 5-HMHF. The phthalimide core is wrapped in a lipophilic
binding site and interacts with Leu194, Trp84 and Trp81, typical
residues involved in the cofactor binding. The aromatic side chain
has contact distance from the side chains of Glu59 (3.4 Å), Trp84
(3.1 Å) and Phe227 (3.4 Å), while the acetoxy substituent in para

A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
653
Cl
Cl
C
N
O
16
19
22
17
20
23
18
21
24
I
Cl
COOMe
N
O
Cl
N
N
N
N
NH
O
N
N
O
25
28
26
29
27
30
Br
Br
I
OH
31
34
32
35
33
36
OH
OH
Cl
Cl
O
O
O
37
38
39
S
S
Chart 1. (continued).
position of the phenyl ring is at 3.7 Å from Leu223. In addition, the
acetoxy group is located within 6.0e7.0 Å from several carbonyl
groups of the main chain and from the backbone chains of Lys49
and Lys310. The absence of direct H-bonds and the prevalence of
lipophilic interactions explain the low affinity of the compound for
EfTS. Compound (S)-12 was the only compound of the series
providing a crystal structure of its complex with EfTS despite (S)-12
to the active site. A structural comparison between our enzyme and
other bacterial TS that form ternary complexes with dUMP and
folate analogues (e.g. PDB: 1NCE) [41,44] highlighted that 5-HMHF
and (S)-12 are bound closer to EfTS catalytic Cys197 where usually
dUMP binds, if compared to the folate analogues. However, a slight
rearrangement of the cofactor could be expected to allow space for
dUMP binding. Probably, the sulfate anions used at high concen-
tration for crystallization compete for the same site where the
phosphate group of dUMP binds [14]. Indeed, a sulfate anion is
bound in the active site cavity at the same site in all subunits. This
site corresponds to the binding site of the dUMP phosphate group
in 1NCE as well as other Escherichia. coli TS complexes with dUMP
is a weak inhibitor (10% at 100 mM). This occurs frequently when
hits screenings are performed to identify X-ray crystal structures
needed to start a structure-based drug design [43]. The puzzling
aspect of our EfTS is that, despite the addition of the dUMP to the
crystallization medium, the substrate has never been found bound

654
A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
O
OH
O
O
40
43
41
44
42
45
O
O
O
OH
NH2
O
N
O
O
( )
O
8
Cl
O
O
Cl
O
O
Cl
( )
46
49
52
47
50
48
51
( )
( )
8
O
N
O
8
8
N
O
O
N
O
O
O
S
O
OH
Chart 1. (continued).
Compd
53
Structure
Compd
Structure
54
56
O
O
N
O
OH
55
N
O
Chart 2. Compounds 53e56.
and the inhibitors F9 (PDB: 4LRR) [10] or BGC 945 (PDB: 4ISK)
[45,46]. In our previous paper [8] the crystallographic complexes of
LcTS with phthalimidic compounds (I in Fig. 1) have been obtained.
Due to the structural similarity between LcTS and EfTS, we can
compare the complexes of LcTS-inhibitors already published [8]
with the complex EfTS-(S)-12. The compounds of the previously
published library presented a phthalimide core directly linked to
the side chain, thus they cannot assume a folded conformation and
they show a binding mode different to that of compound (S)-12.
Moreover, the presence of dUMP in LcTS complexes suggested that
dUMP induces a different orientation of the phtalimidic core intoTS
active site and contributes to stabilize the protein-ligand complex

A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
655
Scheme 1. Reagents and conditions: (i) 2,6-dimethyphenol, Na, abs EtOH, 70 ꢀC, for (RS)-59; (ii) 2,6-dimethylphenol, PPh3, DIAD, dry THF, rt, for (R)- and (S)-60; (iii) LiAlH4, dry THF,
reflux; (iv) phthalimide, PPh3, DIAD, dry THF, rt.
Scheme 2. Reagents and conditions: (i) 10% HCl, THF, 0 ꢀC then rt.
Scheme 3. Reagents and conditions: (i) phthalic anhydride, Et3N, toluene, 130 ꢀC; (ii) gaseous HI, dry toluene, ꢁ5 ꢀC then rt.
Scheme 4. Reagents and conditions: (i) (CH3)SnN3, dry toluene, 110 ꢀC; (ii) gaseous HCl, dry toluene/THF, 0 ꢀC; (iii) CH3I, K2CO3, rt. then reflux.
interaction. On the other hand, the compounds of the present paper
have spacers between the phthalimide core and the aromatic
moiety, therefore only a folded conformation is possible for binding
into EfTS active site; the obtained results suggest that this decreases
the inhibition activity and complex stability. The structure deter-
mination of EfTS complexes with rigid inhibitors (e.g. 6A, 8A and
12A) could further clarify the potential of phthalimide derivatives
Scheme 5. Reagents and conditions: (i) EEDQ, Et3N, CHCl3, reflux.

656
A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
O
O
O
O
O
O
ii
i
N
N
N
H2N
O
O2N
O
O
43
( )-
R
63
( )-
R
7
( )-
R
Scheme 6. Reagents and conditions: (i) NaNO3, TFA, rt, (ii) 10% Pd/C, abs EtOH, rt.
Scheme 7. Reagents and conditions: (i) Et3N, toluene, reflux.
Scheme 8. Reagents and conditions: (i) phthalic anhydride, Et3N, toluene, reflux; (ii) 1-naphthol, PPh3, DEAD, anhyd THF, rt.
Scheme 9. Reagents and conditions: (i) tetrachlorophthalic anhydride, Et3N, toluene, reflux; (ii) phenol, PPh3, DEAD, anhyd THF, rt.
Scheme 10. Reagents and conditions: (i) thiophenol, K2CO3, anhyd DMF, reflux; (ii) tetrachlorophthalimide, PPh3, DEAD, anhyd THF, rt.
“closed” conformation with the MTX inhibitor bound. At variance
OH
O
O
with the previous structure, no 5-HMHF is found, possibly sug-
gesting that MTX has replaced the folate derivative in both sub-
units. The pose and the electron density of MTX bound to EfTS is
reported in Fig. 5A and compared with the EfTS-(S)-12 in Fig. 5B.
(S)-12 and MTX assume essentially the same pose in the EfTS active
site and make similar interactions. The glutamic acid moiety ter-
minal portion of MTX, that is missing in (S)-12, does not interact
with the cavity, the closest approach being that of the vicinal
carboxylate group of MTX lying at 3.4 Å and 4.0 Å from the side
chains of Phe227 and Lys48, respectively. It is noteworthy that, as
for the (S)-12 EfTS structure, dUMP is not bound into the active site
of the EfTS-MTX complex and the binding site of dUMP phosphate
is occupied by a sulfate anion in both crystal structures. Comparison
O
i
N
O
(S)-68
(R)-1
Scheme 11. Reagents and conditions: (i) phthalimide, PPh3, DIAD, dry THF, rt.
as tool for the EfTS inhibition. The structure of EfTS-MTX shows the
same subunit heterogeneity observed for the EfTS-(S)-12 complex
with two subunits (A and B) in “open” conformation and two in

A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
657
Compound
Absolute configuration
EfTS % inhibition
hTS % inhibition
EcTS % inhibition
EcTS % inhibition
hTS % inhibition
EfTS % inhibition
Absolute configuration
# # # #
I
# #
I
I
# #
I
I
I
# #
I
I
# #
I
I
# # D # # D D #
I
# #
I
I
I
# #
I
I
#
I
I
I
I
# #
I
D
I
# #
# #
I
D
I
I
#
I
# # # # #
I
I
#
I
# # # #
I
I D I # D D I # #
I
I
I
I
D
D
D D D D D
D
I
D
I
#
I
I
#
#
# D D
ND #
#
I
#
I
#
I
#
NI
#
#
NI
# # NI
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I D D D D D D D D D D D D D D
NI
NI
I
I
I
I
I
# D #
I
#
I
I
#
I
I
I
# # # #
I
I
I
I
# # # I D I
I
I
I
I
I
# D # # #
I
#
# # I # # # #
# # # # # D #
Compound
Fig. 2. Inhibitory activity of compounds 1e56 against EfTS, hTS and EcTS. The compound activity is given as % of inhibition at 100
and 35), light green (values between 34 and 10), yellow (values lower than 10), red (no detectable inhibition at 100
compound. Ki EfTS ¼ 13 M. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
m
M. Color code: dark green (values between 100
a
m
M) and light blue (not tested). MTX is the reference
m
of the EfTS-(S)-12 and MTX structures with the closed form of
human TS with bound raltitrexed (PDB: 1HVY) suggests a possible
way to increase specificity and affinity of inhibitors towards the
bacterial enzyme. The only relevant difference between human TS
and EfTS cavity consists in the presence of Glu88 that is replaced by
Ala111 in the human enzyme. The side chain of Glu88 is located at
about 7.0 Å from the dimethyl benzene ring of (S)-12 and from the
PABA ring of MTX. The design and synthesis of (S)-12 derivatives
where the benzene ring bears meta substituents able to form H-
bonds with the side chain of Glu88 could lead to an increased af-
finity towards EfTS together with specificity.
3,6,41 [23,24], 5,27e32,56 [25], 9,11,14e16,26,35,44e49,65 [27], 18
[29], 19 [47], 34 [30], 36e40,51 [18], 63 [31,32] has been already
published in literature. For compound 7, only the (R)-enantiomer,
not previously reported, has been herein described, while (RS)-7 is
described in the literature [26]. Compounds 22, 24, 25, 52, 55 had
been previously described in the literature (as referenced below)
but not fully characterized and/or synthesized following different
procedures. All chemicals were purchased from Sigma-Aldrich or
Lancaster in the commercially available highest quality. Solvents
were reagent grade unless otherwise indicated. Yields refer to pu-
rified products and were not optimized (Table S4, see Supple-
mentary Material). The structures of the compounds were
confirmed by routine spectrometric analyses. Only spectra for
compounds not previously described are given. Melting points
(mp) were determined on a Gallenkamp melting point apparatus in
open glass capillary tubes and are uncorrected (Table S4, see Sup-
plementary Material). The infrared spectra were recorded on a
Perkin-Elmer SpectrumOne FT spectrophotometer and band posi-
tions are given in reciprocal centimetres (cmꢁ1). 1H NMR and 13C
NMR spectra were recorded on a Varian VX Mercury spectrometer
operating at 300 and 75 MHz for 1H and 13C, respectively, using
CDCl3 and DMSO-d6 as solvents. Chemical shifts are reported in
3. Conclusions
Aiming to obtain novel specific inhibitors of EfTS enzyme, we
combined an in vitro target-based screening approach with X-ray
crystallographic studies. The most interesting compounds (18, (R)-
40 and 54) showed Ki of 13e25 mM, with compound (R)-40 being
the most potent and selective towards EfTS with respect to the
human enzyme. X-ray crystallographic studies were carried out on
structurally different phthalimides. The X-ray crystal structure of
EfTS in complex with a low affinity inhibitor, (S)-12, was deter-
mined and refined. In this structure EfTS behaves as half-site
enzyme. According to the crystallographic structure, compound
(S)-12 binds into the active site with an orientation analogue to that
of folate metabolites, thus suggesting a competitive behavior with
respect to the folate co-substrate. One subunit binds a folate
metabolite (5-HMHF), suggesting that high affinity ligands are
needed to fully displace the substrates in both the subunits. We can
conclude that target druggability is affected by the presence of
folate analogues bound into one EfTS subunit. Moreover, we solved
the X-ray crystal structure of EfTS in complex with MTX
parts per million (ppm) relative to solvent resonance: CDCl3,
d 7.26
(1H NMR) and 77.3 (13C NMR); DMSO-d6, 2.48 (1H NMR). J values
d
d
are given in Hz (NMR Spectra for compounds 2, 3, 5, 6, 9, 11, 12, 14,
16, 18, 36e42, 44e50 are available in the Supplementary Material.).
Gas chromatography (GC)/mass spectroscopy (MS) was performed
on a Hewlett-Packard 6890e5973 MSD at low resolution. Liquid
chromatography (LC)/mass spectroscopy (MS) was performed on a
spectrometer Agilent 1100 series LC-MSD Trap System VL. The
molecular ion was designed as “Mþ”. Elemental analyses were
performed on a Eurovector Euro EA 3000 analyzer and the data for
C, H, N were within 0.4 of theoretical values (Table S3, see Sup-
plementary Material). Optical rotations were measured on a Perkin
(Ki ¼ 13
mM) in which the folate metabolites, such as 5-HMHF, are
not bound. This confirms the ability of high affinity ligands to
displace the metabolites from the binding site. The present study
discloses the EfTS druggability and represents an interesting
starting point for a target-based drug design aimed at developing
more potent EfTS inhibitors and antibacterial drugs.
Elmer (Norwalk, CT) Mod 341 spectropolarimeter; concentrations
21
are expressed in g/100 mL, and the cell length was 1 dm, thus [
a]
D
values are given in units of 10ꢁ1 deg cm2
g
ꢁ1. Chromatographic
separations were performed on silica gel columns by flash chro-
matography (Kieselgel 60, 0.040e0.063 mm, Merck, Darmstadt,
Germany) as previously described [48,49].
4. Experimental section
4.1. Chemistry
4.1.1. (R)-2-[1-Methyl-2-(1-naphthyloxy)ethyl]-1H-isoindole-
1,3(2H)-dione (R)-1
The synthesis of compounds 1,4,10,21,23 [19], 2,12,42,50 [21,22],
Prepared as reported below for (RS)-8 starting from (S)-68 and

658
A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
6H, CH3), 3.96 (dd, J ¼ 9.3, 5.5 Hz, CHHCH), 4.48 (t, J ¼ 9.3 Hz, 1H,
CHHCH), 4.80e4.95 (m, 1H, CH), 7.70e7.76 (m, 2H, ArHC-5,6),
7.82e8.00 (m, 4H, ArOHC-3,5, ArHC-4,7). Other spectroscopic data
were in agreement with those reported in the literature for the
racemate [26]. Anal. calcd for C19H18N2O5%: C, 64.40; H 5.12; N 7.91.
Found: C, 64.45; H 5.16; N 7.95.
4.1.1.2. (RS)-2-[2-(2,6-dimethylphenoxy)propyl]-1H-isoindole-
1,3(2H)-dione (RS)-8. To a stirred solution of (RS)-61 (1.97 g,
10.9 mmol), phthalimide (2.40 g, 16.3 mmol) and triphenylphos-
phine (4.28 g, 16.3 mmol) in dry THF (100 mL), under N2 atmo-
sphere, a solution of diisopropylazodicarboxylate (DIAD) (3.30 g,
16.3 mmol) in dry THF (50 mL) was added dropwise. The mixture
was stirred at room temperature for 24 h. The solvent was then
evaporated under reduced pressure, Et2O was added and the pre-
cipitate formed was filtered off. The filtrate was evaporated in
vacuum and the residue was purified by flash chromatography
(eluent EtOAc/petroleum ether 2:8) to give 3.24 g of a white solid
(96%) which was recrystallized from EtOAc: mp 90e91 ꢀC; IR (KBr):
1772, 1717 (C]O) cmꢁ1
;
1H NMR (CDCl3):
d
1.19 (d, J ¼ 6.0 Hz, 3H,
CH3CH), 2.22 (s, 6H, CH3Ar), 3.85 (dd, J ¼ 13.5, 6.3 Hz,1H, CHH), 4.07
(dd, J ¼ 13.6, 7.0 Hz, 1H, CHH), 4.52 (sextet, J ¼ 6.3 Hz, 1H, CH),
6.82e7.0 (m, 3H, Ar), 7.68e7.78 (m, 2H, Ar), 7.82e7.90 (m, 2H, Ar);
13C NMR (CDCl3):
d 17.3 (2C), 18.2 (1C), 43.4 (1C), 74.2 (1C), 123.5
(2C), 123.7 (1C), 129.2 (2C), 131.4 (2C), 132.2 (2C), 134.2 (2C), 153.1
(1C), 168.5 (2C); GC/MS (70 eV) m/z (%) 309 (Mþ, 16), 188 (100).
Anal. calcd for C19H19NO3.0.25H2O %: C, 72.71; H 6.26; N 4.46.
Found: C, 72.85; H 6.19; N 4.51.
4.1.1.3. (R)-2-[2-(2,6-Dimethylphenoxy)propyl]-1H-isoindole-
1,3(2H)-dione (R)-8. Prepared as reported above for (RS)-8 starting
from (R)-61. Yield: 87%; white crystals: mp 69e70 ꢀC (EtOAc);
20
[a
]
¼ ꢁ81 (c 2, CHCl3). Spectroscopic data were in agreement with
D
the racemate. Anal. calcd for C19H19NO3%: C, 73.77; H 6.19; N 4.53.
Found: C, 73.82; H 6.24; N 4.57.
4.1.1.4. (S)-2-[2-(2,6-Dimethylphenoxy)propyl]-1H-isoindole-
1,3(2H)-dione (S)-8. Prepared as reported above for (RS)-8 starting
Fig. 3. EfTS x-ray crystallographic structure. (A) The EfTS crystal asymmetric unit
content. The two subunits of each heterodimer differ in the conformation of the small
domain due to the presence of a ligand bound to the active site cavities. The subunit
having a ligand bound are in closed conformation, while the unbound subunits are in
open conformation. (B) comparison of closed and open conformations by super-
imposing subunit A (open; light blue ribbon) and B (closed; gold ribbon) of 4O7U. The
5-HMHF molecule is represented to show the position of the active site cavity. (For
interpretation of the references to colour in this figure legend, the reader is referred to
the web version of this article.)
from (S)-61. Yield: 80%; white crystals: mp 70e71 ꢀC (EtOAc);
20
[a
]
¼ þ79 (c 2, CHCl3). Spectroscopic data were in agreement with
D
the racemate. calcd for C19H19NO3%: C, 73.77; H 6.19; N 4.53. Found:
C, 73.59; H 6.18; N 4.57.
4.1.1.5. 2-[1-(4-Hydroxy-2,6-dimethylphenoxy)propan-2-yl]-1H-iso-
indole-1,3(2H)-dione (RS)-13. A solution of (RS)-12 (0.82 g,
2.23 mmol) in THF (6 mL) and 10% HCl solution (5 mL) was stirred in
an ice-bath for 3 h. Then the mixture was stirred at room temper-
ature for 48 h. The solvent was evaporated under vacuum and
extracted with EtOAc. The organic layer was dried (Na2SO4) to give
a slightly yellowish solid which was recrystallized from EtOAc/pe-
troleum ether to give 0.43 g (59%) of slightly yellowish crystals: mp
178e179 ꢀC; IR (KBr): 3412 (OH), 1770, 1699 (C]O); 1H NMR
20
phthalimide. Yield: 57%; mp 79e81 ꢀC (EtOAc/hexane); [
a]
¼ ꢁ38
D
(c 1, CHCl3). Spectroscopic and spectrometric data were in agree-
ment with those found in the literature for the racemate [19]. Anal.
Calcd. for C21H17NO3%: C, 76.12; H 5.17; N 4.23. Found: C, 75.85; H
5.16; N 4.31.
(CDCl3):
d
1.53 (d, J ¼ 6.9 Hz, 3H, CH3CH), 2.12 (s, 6H, CH3Ar), 3.84
4.1.1.1. (R)-2-[2-(2,6-Dimethyl-4-nitrophenoxy)-1-methylethyl]-1H-
isoindole-1,3(2H)-dione (R)-7. 0.65 g (2.1 mmol) of (R)-63 were
dissolved in 5 mL of trifluoroacetic acid at room temperature. Then,
0.22 g (2.5 mmol) of sodium nitrate were added to the solution in
small portions during the period of 60 min. The mixture was stirred
overnight, then diluted with 15 mL of diethyl ether and 3 mL of
brine. The aqueous phase was extracted with Et2O for three times.
The organic layers, washed with 20% NaOH aqueous solution, were
dried over anhydrous Na2SO4 to give 0.65 g (85%) of an orange foam
(dd, J ¼ 9.2, 5.6 Hz, 1H, CHHCH), 4.29 (t, J ¼ 9.2 Hz, 1H, CHHCH), 4.58
(br s, 1H, OH), 4.76e4.92 (m, 1H, CH), 6.41 (s, 2H, Ar), 7.65e7.74 (m,
2H, Ar), 7.78e7.90 (m, 2H, Ar); 13C NMR (CDCl3):
d15.4 (1C), 16.5
(2C), 47.4 (1C), 72.2 (1C), 115.2 (2C), 123.4 (2C), 132.1 (2C), 132.2
(2C),134.2 (2C),149.3 (1C),151.5 (1C),168.7 (2C); GC/MS (70 eV) m/z
(%) 325 (Mþ, 5), 188 (100). Anal. calcd for C19H19NO4.0.20H2O %: C,
69.37; H 5.94; N 4.26. Found: C, 69.59; H 5.84; N 4.37.
4.1.1.6. (S)-2-(2-Iodo-1-phenylethyl)-1H-isoindole-1,3(2H)-dione
(S)-17. Into a stirred ice-cooled solution of (S)-33 (1.74 g, 6.5 mmol)
in dry toluene (30 mL) under N2 atmosphere, HI was bubbled until
saturation occurred. The mixture was stirred at ꢁ5 ꢀC for 1 h and at
which was recrystallized from Et2O/petroleum ether to give orange
20
crystals: mp 105e106 ꢀC (Et2O/petroleum ether); [
a]
¼ ꢁ30 (c 2,
D
CHCl3); 1H NMR (CDCl3):
d
1.55 (d, J ¼ 7.2 Hz, 3H, CH3CH), 2.26 (s,

A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
659
Fig. 4. Detailed view of the binding ligands. (A) Stick representation of the (S)-12 inhibitor (yellow) bound to EfTS with superimposed the 2Fo-Fc electron density map at 1.3
s
(blue
wire), computed with refined phases and the omit map (green wire) contoured at 3.0
(B) Stick representation of 5-HMHF (yellow) bound to EfTS with superimposed the 2Fo-Fc electron density map at 1.3
(green wire) contoured at 3.0 . Cys197 is also shown as sticks together with other relevant residues in the active site cavity. Notice the different extent of the omit map in the PABA
s. Cys197 is also shown as sticks together with other relevant residues in the active site cavity.
s
(blue wire), computed with refined phases and the omit map
s
moiety of 5-HMHF with respect to the same region in Fig. 4A. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this
article.)
Fig. 5. The binding of MTX to EfTS. (A) Stick representation of the MTX inhibitor (yellow) bound to EfTS with superimposed the 2Fo-Fc electron density map at 1.3
s (blue wire),
computed with refined phases and the omit map (green wire) contoured at 3.0 . Cys197 is also shown as sticks together with other relevant residues in the active site cavity. (B)
s
Least-squares superimposition of MTX (green sticks) and (S)-12 (yellow sticks) bound to EfTS. Notice the similarity of the pose. (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)
room temperature for 48 h. The solvent was then evaporated under
vacuum and the residue, taken up with EtOAc, was washed with 6 N
NaOH, 2 N HCl, and brine. The organic layer was dried (Na2SO4) and
concentrated under vacuum to give 1.7 g (70%) of a slightly
Anal. calcd for C16H12NO2I %: C, 50.95; H 3.21; N 3.71. Found: C,
51.15; H 3.25; N 3.79.
4.1.1.7. 2-{[20-(1-Methyl-1H-tetrazol-5-yl)biphenyl-4-yl]methyl}-1H-
isoindole-1,3(2H)-dione 20. Potassium carbonate (65.2 g,
0.47 mmol) was added to a solution of 22 (0.15 g, 0.39 mmol) in
10 mL of acetone. The mixture was stirred at room temperature for
30 min, and then methyl iodide (0.11 g, 0.77 mmol) was added. The
reaction mixture was refluxed for 2 h. After cooling, the reaction
mixture was filtered and evaporated. The residue was extracted
with EtOAc and washed twice with H2O. After drying (Na2SO4), the
organic phase was evaporated to afford a crude solid which was
yellowish solid which was recrystallized from MeOH to give (S)-17
20
as white crystals: mp 123e124 ꢀC; [
a]
¼ ꢁ16 (c 2, CHCl3); IR (KBr):
D
1780, 1707 (C]O); 1H NMR (CDCl3):
d
3.83 (dd, 1H, J ¼ 10.5, 5.2 Hz,
CHH), 4.53 (t, 1H, J ¼ 10.8 Hz,CH), 5.56 (q, 1H, J ¼ 5.2 Hz, CHH),
7.28e7.38 (m, 3H, Ar), 7.50e7.58 (m, 2H, Ar), 7.68e7.76 (m, 2H, Ar),
7.82e7.88 (m, 2H, Ar); 13C NMR (CDCl3):
d 4.8 (1C), 57.8 (1C), 123.8
(2C), 128.1 (2C), 128.9 (2C), 129.2 (1C), 131.8 (2C), 134.5 (1C), 138.1
(2C), 168.1 (2C); GC/MS (70 eV) m/z (%) 377 (Mþ, <1), 250 (100).

660
A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
purified by crystallization from CHCl3/hexane affording 83 mg
(54%) of white crystals: mp 210e211 ꢀC; IR (KBr): 1766, 1705 (C]
R-enantiomer [53,54].
O); 1H NMR (CDCl3):
d
4.22 (s, 3H, CH3), 4.82 (s, 2H, CH2), 7.04e7.16
(m, 2H, Ar), 7.28e7.38 (m, 2H, Ar), 7.40e7.90 (m, 8H, Ar); 13C NMR
(CDCl3): 33.8 (1C), 41.3 (2C), 122.6 (1C), 123.6 (2C), 123.7 (1C),
4.1.1.13. (R)-2-[1-(4-amino-2,6-dimethylphenoxy)propan-2-yl]-1H-
isoindole-1,3(2H)-dione hydrochloride (R)-43. Catalytic hydrogena-
tion of (R)-7 (0.35 g, 1.0 mmol) in 10 mL of absolute EtOH was
conducted at room temperature for 12 h in the presence of 10%
palladium on carbon at 10 bar. The catalyst was filtered off and the
solvent was removed under reduced pressure to give 0.23 g (70%) of
the free amine (R)-43 as a yellow oil which was converted into its
hydrochloride salt by dissolving the free base in a small amount of
aqueous 1 N HCl and azeotropically removing water. The crude
d
128.2 (1C), 128.4 (2C), 129.1 (1C), 130.6 (2C), 131.7 (1C), 132.0 (1C),
134.3 (2C), 135.3 (1C), 136.6 (1C), 138.7 (1C), 141.5 (1C), 168.3 (2C);
GC/MS (70 eV) m/z (%) 395 (Mþ, 34), 394 (100). Anal. calcd for
C
23H17N5O2%: C, 69.86; H 4.33; N 17.71. Found: C, 69.47; H 4.37; N
17.43.
4.1.1.8. 2-{[20-(1H-tetrazol-5-yl)biphenyl-4-yl]methyl}-1H-isoindole-
1,3(2H)-dione 22. A suspension of compound 62 (0.60 g, 1.1 mmol)
in a mixture of toluene (8 mL) and THF (2 mL) was stirred in an ice
bath. Gaseous HCl was then added to give a clear solution followed
by precipitation of the desired product. The precipitate was filtered,
and the filtrate was washed with toluene to give a white solid
(0.40 g, 96%): mp: >250 ꢀC; IR (KBr): 3458 (NH), 1767, 1714 (C]O),
solid obtained was recrystallized from EtOH/Et2O: mp 256e258 ꢀC;
20
[a
]
¼ ꢁ42 (c 2, MeOH). Spectroscopic and spectrometric data
D
were in agreement with those found in the literature for the
racemate [26]. Anal. calcd for C19H20N2O3.HCl.0.33H2O %: C, 62.21;
H 5.95; N 7.64. Found: C, 62.16; H 6.06; N 7.24.
4.1.1.14. 2-[3-(Naphthalen-1-yloxy)propyl]-1H-isoindole-1,3(2H)-
dione 52. Prepared as reported above for (RS)-8 starting from (R)-
64 and 1-naphthol using diethyl azodicarboxylate (DEAD) instead
of DIAD. Yield: 36%; mp 150e151 ꢀC (EtOꢁA1c/hexane), lit [55]
;
1H NMR (CDCl3):
1463,1252 (NeN]N) cmꢁ1; 1H NMR (DMSO-d6):
d 4.72 (s, 2H, CH2),
7.04 (d, J ¼ 8.2 Hz, 2H, Ar), 7.23 (d, J ¼ 8.2 Hz, 2H, Ar), 7.46e7.70 (m,
4H, Ar), 7.78e7.96 (m, 4H, Ar), 16.4 (1H, NH); LC/MS m/z 380
[Mþ ꢁ 1]. Other spectroscopic data were in agreement with the
literature [50]. Anal. calcd for C22H15N5O2.0.50H2O %: C, 67.68; H
4.13; N 17.94. Found: C, 67.73; H 4.03; N 17.70.
154e154.5 ꢀC; IR (KBr): 1765, 1713 (C]O) cm
d
2.34(quintet, J ¼ 6.9 Hz, 2H, CH2CH2CH2),4.01 (t, J ¼ 6.8 Hz, 2H,
CH2N),4.22 (t, J ¼ 6.0 Hz, 2H, CH2O), 6.76 (d, J ¼ 6.6 Hz, 1H, Ar),
7.33e7.44 (m, 4H, Ar), 7.64e7.68 (m, 2H, Ar),7.75 (d, J ¼ 8.2 Hz, 1H,
ArO), 7.78e7.82 (m, 2H, Ar),8.16 (d, J ¼ 7.7,1H, Ar); 13C NMR (CDCl3):
4.1.1.9. 2-(4-Methylphenyl)-1H-isoindole-1,3(2H)-dione
24.
EEDQ (1.48 g, 6 mmol), p-toluidine (0.59 g, 5.5 mmol) and Et3N
(1.04 mL, 7.5 mmol) were successively added to a stirring solution
of 2-(methoxycarbonyl)benzoic acid (1.0 g, 5 mmol) in CHCl3
(160 mL). The reaction mixture was heated under reflux for 6 h. The
solvent was removed under reduced pressure and the residue,
taken up with EtOAc, was washed three times with 2 N HCl, twice
with 2 N NaOH, and then dried over anhydrous Na2SO4. The organic
phase was evaporated to afford a crude solid which was purified by
crystallization from CHCl3/hexane to give 0.90 g (67%) of 24 as beige
crystals: mp 205e206 ꢀC; IR (KBr): 1747, 1716 (C]O); 1H NMR
d
28.7 (1C), 35.9 (1C), 66.1 (1C), 104.7 (1C), 120.6 (1C), 122.1 (1C),
123.4 (2C), 125.3 (1C),125.7 (1C), 126.0 (1C), 126.5 (1C), 127.6 (1C),
132.4 (2C), 134.1 (2C), 134.6 (1C), 154.6 (1C), 168.6 (2C); GC/
MS(70 eV) m/z (%) 331 (Mþ, 13), 188 (100).
4.1.1.15. 4,5,6,7-Tetrachloro-2-(3-phenoxypropyl)-1H-isoindole-
1,3(2H)-dione 53. Prepared as reported above for (RS)-8 starting
from 65 and phenol using diethyl azodicarboxylate (DEAD) instead
of DIAD. Yield: 30%; mp 211e212 ꢀC (EtOAc/hexane); IR (KBr): 1777,
1712 (C]O) cmꢁ1; GC/MS (70 eV) m/z (%) 419 (Mþ, 6), 326 (100).
Anal. calcd for C17H11Cl4NO3.0.50H2O %: C, 47.70; H 2.83; N 3.27.
Found: C, 47.62; H 2.65; N 3.41.
(CDCl3):d 2.41 (s, 3H, CH3), 7.31 (s, 4H, Ar), 7.74e7.82 (m, 2H, Ar),
7.92e7.99 (m, 2H, Ar); GC/MS (70 eV) m/z (%) 237 (Mþ, 100). Other
spectroscopic data were in agreement with the literature [51]. Anal.
calcd. for C15H11NO2.0.20H2O %: C, 74.80; H 4.77; N 5.82. Found: C,
75.06; H 5.08; N 6.00.
4.1.1.16. 4,5,6,7-Tetrachloro-2-[3-(phenylsulfanyl)propyl]-1H-iso-
indole-1,3(2H)-dione 54. Prepared as reported above for (RS)-8
starting from 67 and tetrachlorophthalimide, using dieth-
ylazodicarboxylate (DEAD) instead of DIAD. Yield: 20%; mp
140e141 ꢀC (THF/petroleum ether); IR (KBr): 1775, 1713 (C]O)
4.1.1.10. 2-(4-Phenylbutyl)-1H-isoindole-1,3(2H)-dione
25.
Prepared as reported for compound 55 starting from 4-
phenylbutan-1-amine and phthalic anhydride. Yield: 62%; white
solid: mp 84e85 ꢀC. Spectroscopic data were in agreement with the
literature [52].
cmꢁ1
;
1H NMR (CDCl3):
d
1.97e2.04 (m, 2H, CH2CH2CH2), 2.93 (t,
J ¼ 7.1 Hz, 2H, CH2S), 3.83 (t, J ¼ 7.0 Hz, 2H, CH2N), 7.15e7.35 (m, 5H,
Ar); 13C NMR (CDCl3):
27.9 (1C), 31.7 (1C),38.1 (1C),126.7
d
(2C),127.8 (1C), 129.2 (2C), 129.8 (2C), 130.3 (2C), 135.7 (2C),140.3
4.1.1.11. (R)-2-(2-Hydroxy-1-phenylethyl)-1H-isoindole-1,3(2H)-
dione (R)-33. Prepared as reported above for compound 64 starting
(1C), 163.7 (2C); GC/MS (70 eV) m/z (%) 433 (Mþ, 54), 149 (100).
from (R)-phenylglycinol and phthalic anhydride. Yield: 88%; white
20
crystals: [
a
]
D
¼ þ 41 (c 2, CHCl3); IR (neat): 3464 (OH), 1772, 1710
4.1.1.17. 4,5,6,7-Tetrachloro-2-(4-phenylbutyl)-1H-isoindole-1,3(2H)-
dione 55. A mixture of 4-phenylbutan-1-amine (2.0 g, 13.4 mmol),
tetrachlorophthalic anhydride (3.83 g, 13.4 mmol), triethylamine
(0.13 g, 1.34 mmol) in toluene (20 mL) was heated under reflux in a
flask fitted with a DeaneStark tube for 7 h. During this period, the
temperature of the oil bath is maintained at about 130 ꢀC and water
separates. All volatile matter were then evaporated under vacuum
and the solid residue was taken up with EtOAc and washed with 2 N
HCl, NaHCO3, and H2O. The organic phase was dried (Na2SO4) and
concentrated under vacuum to give a white solid which was
recrystallized from EtOAc/hexane to give 3.90 g (70%) of white
crystals: mp 119e120 ꢀC. Other spectroscopic data were in agree-
ment with the literature [52]. Anal. calcd for C18H13Cl4NO2%: C,
51.83; H 3.14; N 3.36. Found: C, 51.71; H 3.13; N 3.44.
(C]O); 1H NMR (CDCl3):
d
2.64 (br s, 1H, OH), 4.21 (dd, J ¼ 11.6,
5.0 Hz, 1H, CHH), 4.65 (dd, J ¼ 11.6, 9.3 Hz, 1H, CHH), 5.41e5.53 (m,
1H, CH), 7.23e7.38 (m, 3H, Ar), 7.40e7.50 (m, 2H, Ar), 7.63e7.73 (m,
2H, Ar), 7.74e7.85 (m, 2H, Ar). Other spectroscopic data were in
agreement with those reported in the literature [53,54]. Anal. calcd.
for C16H13NO3%: C, 71.90; H 4.90; N 5.24. Found: C, 72.24; H 4.99; N
5.40.
4.1.1.12. (S)-2-(2-Hydroxy-1-phenylethyl)-1H-isoindole-1,3(2H)-
dione (S)-33. Prepared as reported above for compound 64 starting
from (S)-phenylglycinol and phthalic anhydride. Yield: 85%; slightly
yellowish crystals: [
20
a
]
¼ ꢁ32 (c 2, CHCl3). Other spectroscopic
D
data were in agreement with those reported in the literature for the

A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
661
4.1.1.18. (RS)-Ethyl 2-(2,6-dimethylphenoxy)propanoate (RS)-59.
Sodium (0.38 g, 16.5 mmol) was added in small pieces to absolute
EtOH (30 mL) and the mixture was heated at 70 ꢀC. After comple-
tion of sodium dissolution, a solution of 2,6-dimethylphenol (2.0 g,
16.5 mmol) in absolute EtOH (15 mL) was added dropwise. After
30 min, a solution of (RS)-ethyl 2-bromopropanoate [(RS)-57]
(2.98 g, 16.5 mmol) in absolute EtOH (15 mL) was added dropwise
and the mixture was heated at reflux for 5 h. The solid residue was
filtered off, the solvent was evaporated in vacuum and the residue
was taken up with diethyl ether, washed with 2 N NaOH and dried
(Na2SO4). Removal of the solvent under vacuum gave 3.11 g (85%) of
4.1.1.23. (R)-2-(2,6-Dimethylphenoxy)propan-1-ol
(R)-61.
Prepared as reported above for (RS)-61 starting from (R)-60. Yield:
20
62%; colorless oil: [
a
]
D
¼ ꢁ9.1 (c 2.1, CHCl3). Spectroscopic data
were in agreement with the (S)-isomer.
4.1.1.24. 2-{[2'-[1-(trimethylstannyl)-1H-tetrazol-5-yl]biphenyl-4-yl}
methyl]-1H-isoindole-1,3(2H)-dione 62. A solution of compound 18
(1.50 g, 4.44 mmol) in dry toluene (75 mL) and azidotrimethyltin
(1.0 g, 4.89 mmol) was kept at 110 ꢀC under nitrogen atmosphere
for 24 h. The precipitate was filtered off and washed with hot
toluene to give compound 62 as a white solid (0.90 mg, 37%):
mp > 250 ꢀC; IR (KBr): 1763,1709 (C]O), 550 (SneN) cmꢁ1;1H NMR
(RS)-59 as a yellow oil: IR (neat): 1737 (C]O) cmꢁ1
(CDCl3):
;
1H NMR
d
1.28 (t, 3H, J ¼ 7.1 Hz, CH3CH2), 1.53 (d, 3H, J ¼ 6.6 Hz,
(DMSO-d6): d 0.32 (br s, 9H, (CH3)3Sn), 4.71 (s, 2H, CH2), 6.98 (d,
CH3CH), 2.28 (s, 6H, CH3Ar), 4.23 (q, 2H, J ¼ 7.1 Hz, CH2), 4.49 (q, 1H,
J ¼ 8.2 Hz, 2H, Ar), 7.16 (d, J ¼ 9.3 Hz, 2H, Ar), 7.35e7.55 (m, 4H, Ar),
J ¼ 6.6 Hz, CH), 6.86e6.94 (m, 1H, Ar), 6.95e7.03 (m, 2H, Ar); 13C
7.75e7.95 (m, 4H, Ar); LC-MS m/z (%) 568 [Mþ þ 23].
NMR (CDCl3):
d 14.4 (1C), 17.2 (2C), 18.7 (1C), 61.3 (1C), 77.1 (1C),
124.2 (1C), 129.3 (2C), 131.1 (2C), 154.8 (1C), 172.3 (1C); GC/MS
4.1.1.25. 4,5,6,7-Tetrachloro-2-(3-hydroxypropyl)-1H-isoindole-
1,3(2H)-dione 65. Prepared as reported above for (R)-33, starting
from 3-aminopropanol and tetrachlorophthalic anhydride.
Yield:32%; mp193e194 ꢀC (EtOAc/hexane); IR (KBr): 3319 (OH),
(70 eV) m/z (%) 222 (Mþ, 91), 149 (100).
4.1.1.19. (R)-Methyl 2-(2,6-dimethylphenoxy)propanoate (R)-60.
Prepared as reported above for (RS)-8 starting from 2,6-
1774, 1707 (C]O) cmꢁ1 1H NMR (CDCl3):
; d 1.85e1.92 (m, 2H,
CH2CH2CH2),1.98 (br s,1H, OH),3.64 (t, J ¼ 5.8 Hz, 2H, CH2N), 3.86 (t,
dimethylphenol and ethyl (S)-(ꢁ)-lactate [(S)-58]. Yield: 72%;
J ¼ 6.3 Hz, 2H, CH2O); 13C NMR (CDCl3):
d 31.1 (1C), 35.7 (1C), 59.5
20
slightly yellowish oil:[
a]
D
¼ þ40 (c 2.1, CHCl3). Spectroscopic data
(1C),127.7 (2C), 129.9 (2C),140.4 (2C); 164.2 (2C); GC/MS(70 eV) m/z
(%) 343 (Mþ, 24), 298 (100). Anal. calcd for C11H17Cl4NO3.H2O %: C,
36.60; H 2.51; N 3.88. Found: C, 36.53; H 2.11; N 3.86.
were in agreement with the (S)-isomer.
4.1.1.20. (S)-Methyl 2-(2,6-dimethylphenoxy)propanoate (S)-60.
Prepared as reported above for (RS)-8 starting from 2,6-
4.1.1.26. 3-(Phenylsulfanyl)propan-1-ol
67. K2CO3
(0.32
g,
dimethylphenol and methyl (R)-(þ)-lactate [(R)-58]. Yield: 37%;
2.32 mmol) was added to a solution of 3-bromopropanol (0.29 g,
2.11 mmol) in dry DMF (30 mL) under N2 atmosphere. The reaction
mixture was heated at 130 ꢀC and then a solution of thiophenol
(0.26 g, 2.32 mmol) in 20 mL of dry DMF was added dropwise
during a period of 3 h. The mixture was stirred at this temperature
for 24 h. After evaporation of the solvent, the residue was taken up
with EtOAc, washed with 2 N NaOH, and then with brine. The
organic phase was dried (Na2SO4) and concentrated under vacuum.
The residue was purified by column chromatography on silica gel
(EtOAc/petroleum ether 2:8) to give 0.36 g (69%) of a yellow oil; IR
20
slightly yellowish oil: [
a]
¼ ꢁ39 (c 2.1, CHCl3); IR (neat): 1759 (C]
D
O) cmꢁ1; 1H NMR (CDCl3):
6H, CH3Ar), 3.78 (s, 3H, CH3O), 4.51 (q, J ¼ 6.9 Hz,1H, CH), 6.88e7.04
(m, 3H, Ar); 13C NMR (CDCl3):
17.2 (2C), 18.7 (1C), 52.3 (1C), 77.0
d
1.53 (d, J ¼ 6.9 Hz, 3H, CH3CH), 2.27 (s,
d
(1C), 124.2 (1C), 129.3 (2C), 131.1 (2C), 154.7 (1C), 172.7 (1C); GC-MS
(70 eV) m/z (%) 208 (Mþ, 1), 122 (100).
4.1.1.21. (RS)-2-(2,6-dimethylphenoxy)propan-1-ol (RS)-61. To
a
suspension of LiAlH4 (1.28 g, 33.7 mmol) in dry THF (60 mL), a
solution of (RS)-59 (3.74 g, 16.9 mmol) in dry THF (60 mL) under N2
atmosphere was added. The mixture was stirred under moderate
reflux for 1 h and then at room temperature for 24 h. The reaction
was quenched by the careful addition of cold water until the end of
gas evolution. The residue was removed by filtration and the filtrate
shaken with H2O. The aqueous phase was extracted several times
with Et2O. The combined extracts were dried over Na2SO4 and the
filtrate was concentrated under vacuum. Purification by flash
chromatography (EtOAc/petroleum ether 1:9) gave 1.85 g (61%) of
(neat): 3356 (OH) cmꢁ1 1H NMR (CDCl3):
; d1.83e1.92 (m, 3H,
CH2CH2CH2 þ OH), 3.03 (t, J ¼ 7.15 Hz, 2H, CH2SAr), 3.75 (t,
J ¼ 6.1 Hz, 2H, CH2OH), 7.14e7.36 (m, 5H, ArS); 13C NMR (CDCl3):
d
30.5 (1C), 31.9 (1C), 61.6 (1C), 126.2 (1C), 129.1 (2C), 129.4 (2C),
136.5 (1C); GC/MS (70 eV) m/z (%) 168 (Mþ, 10), 110 (100).
4.2. Enzymology
The compounds were screened for their activity and their
species-specificity against EcTS, EfTS, and hTS. EcTS, and hTS were
purified as described [8]. EfTS [14] was purified following the pro-
cedure used for EcTS. Enzyme kinetics experiments were con-
ducted by chromogenic assay under standard conditions [8,14]. The
Km values, by curve-fitting to hyperbolic curves, of EcTS, EfTS, and
(RS)-61 as a colorless oil: IR (neat): 3416 (OH) cmꢁ1
;
1H NMR
(CDCl3):
d
1.17 (d, J ¼ 6.3 Hz, 3H, CH3CH), 2.28 (s, 7H, CH3Ar þ OH),
3.68e3.84 (m, 2H, CH2), 4.18e4.30 (m, 1H, CH), 6.87e6.95 (m, 1H,
Ar), 7.01 (d, J ¼ 3.8 Hz, 2H, Ar); 13C NMR (CDCl3):
d 16.4 (1C), 17.4
(2C), 67.1 (1C), 78.1 (1C), 123.8 (1C), 129.3 (2C), 131.4 (2C), 154.0
hTS for MTHF were, respectively, 5.8
mM, 13.6 mM, and 6.9 mM.
(1C); GC/MS (70 eV) m/z (%) 180 (Mþ, 37), 122 (100).
These data represent the mean of triplicate measurements; in a
typical study, the standard deviation of the data fell within 20% of
the mean. The apparent inhibition constant (Ki app, simply defined
as Ki within the text) values were obtained from the linear least-
squares fit of the residual activity as a function of inhibitor con-
centration, assuming a competitive inhibition [8]. In the inhibition
assays, the reaction mixture was the same as in the TS standard
assay. Stock solutions of each inhibitor were freshly prepared in
dimethyl sulfoxide (DMSO) and stored at e 80 ꢀC. In the reaction
mixture, the concentration of DMSO never exceeded 5%. Each
experiment was repeated at least three times and an individual
measurement did not differ more than 20% from the mean.
4.1.1.22. (S)-2-(2,6-Dimethylphenoxy)propan-1-ol
(S)-61.
Prepared as reported above for (RS)-61 starting from (S)-60. Yield:
20
66%; colorless oil: [
a]
¼ þ9.0 (c 2.1, CHCl3); IR (neat): 3401 (OH)
D
cmꢁ1;1H NMR (CDCl3):
overlapping s at 2.28 ppm, exch. D2O, 1H, OH), 2.28 (s overlapping
br s at 2.21 ppm, 6H, CH3Ar), 3.68e3.84 (m, 2H, CH2), 4.18e4.29 (m,
1H, CH), 6.88e6.96 (m, 1H, Ar), 6.97e7.05 (m, 2H, Ar); 13C NMR
d
1.17 (d, J ¼ 6.3 Hz, 3H, CH3CH), 2.21 (br s
(CDCl3):d 16.4 (2C), 17.4 (1C), 67.1 (1C), 78.0 (1C), 123.8 (1C), 129.3
(2C), 131.4 (2C), 154.0 (1C); GC/MS (70 eV) m/z (%) 180 (Mþ, 26), 122
(100).

662
A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
4.3. Crystallization
Supplementary Material). All EfTS crystals belong to space group
P21 with four EfTS subunits in the asymmetric unit. The structures
were solved by molecular replacement technique as implemented
in the software MolRep [62] from the CCP4 Suite [61] using the
crystal structure of Lactobacillus casei TS as model (PDB: 2TDM; 73%
sequence homology). Because of the structural flexibility that
characterize some domains of TS enzymes the LcTS loop 82e144
was removed from the model. Molrep provided the correct solution
that consists of a couple of EfTS homodimers in the crystal asym-
metric unit (A-B and C-D). The EfTS molecule was manually rebuilt
in the electron density maps. The whole polypeptide chain could be
completely reconstructed only in the subunits were the inhibitor
was bound. The unbound subunits presented disorder in the re-
gions corresponding to residues 81e132 that were not clearly
visible in the electron density maps and could be only partially
built. The Fourier difference maps showed electron density inter-
pretable with the presence of the inhibitor only for EfTS-(S)-12, and
EfTS-MTX complexes. The final models were refined using the
program Refmac5 [63] from the CCP4 Suite. Water molecules have
been added using the program ArpWARP [64] and checked
manually. The diffraction pattern of the crystals presented split
reflections due to the lattice disordering as described above. This
resulted in medium quality indicators of the refinement as shown
in Table S2 (see Supplementary Material). Nevertheless, the elec-
tron density maps were quite well defined and could be confidently
interpreted, as can be judged from the reported figures. The ste-
reochemical quality of the final model was checked using the
programs PROCHECK [65] and COOT [66]. The program COOT was
also used for molecular modeling and to superimpose structures.
Structural models were rendered using CCP4MG [67]. Atomic co-
ordinates and structure factors for EfTS-(S)-12, and EfTS-MTX
complexes have been deposited in the Protein Data Bank (PDB),
www.rcsb.org, under the accession codes 4O7U and 5J7W,
respectively.
Crystallization trials for EfTS were performed using the vapor
diffusion sitting drop method [56] at 297 K. Crystal Screen, Crystal
Screen II and Greed Screen Ammonium Sulfate from Hampton
Research were used for screening crystallization conditions. Drops
consisting of 2
25 mM KH2PO4/K2HPO4 pH 6.8, 40 mM dUMP, 1 mM EDTA) were
equilibrated against 100 L reservoir solution. After one month
mL precipitant and 1 mL EfTS solution (7.8 mg/mL in
m
crystals were observed into a drop containing 3.2 M ammonium
sulfate and 0.1 M HEPES buffer, pH 7.0 as precipitant solution. These
crystals were ill-formed and provided a poor diffraction pattern
when exposed to X-ray radiation, therefore we tried to improve the
crystal ordering/quality by applying seeding techniques [56] The
crystals obtained were crushed and seeding solutions were pre-
pared by dilution with the successful precipitant solution (3.2 M
ammonium sulfate and 0.1 M HEPES buffer, pH 7.0). 4
2.5 mg/mL EfTS solution (buffer 25 mM KH2PO4/K2HPO4 pH 6.8,
40 mM dUMP, 1 mM EDTA) were mixed with 2 L of a precipitant
mL drops of a
m
solution containing 2.7 M ammonium sulfate and 0.1 M HEPES
buffer, pH 7.0. Streak-seeding has been performed at different
dilution of the seeding solution. Drops were allowed to equilibrate
at room temperature over 0.6 mL precipitant solution in the well.
Crystals suitable for diffraction appeared in 2e3 weeks into the
drops in which a seeding solution diluted 1:100 was applied. Co-
crystallization trials were performed also for the complexes EfTS-
inhibitor using the same precipitant solution identified in the
previous screen for the enzyme (3.2 M ammonium sulfate and
0.1 M HEPES buffer, pH 7.0). Trials of co-crystallization were per-
formed at 297 K using both the microbatch-under-oil crystalliza-
tion technique [57,58] with a 1:1 mixture of paraffin and silicon oil
(Hampton Research) as well as the sitting-drop vapor-diffusion
method. Because of the low solubility in water, concentrated so-
lutions of each inhibitor were prepared in DMSO. A 4 mM solution
of each inhibitor in DMSO has been added in a 1:10 vol ratio to a
6.5 mg/mL EfTS solution (buffer 25 mM KH2PO4/K2HPO4 pH 6.8,
40 mM dUMP, 1 mM EDTA). Drops have been prepared by mixing
Funding sources
2
m
L of the EfTS-inhibitor solution with 1
m
L of a precipitant solution
The project has been supported by MIUR-PRIN2009 number
200925BPZ5_004.
containing 3.2 M ammonium sulfate and 0.1 M HEPES buffer, pH
7.0. Crystals of EfTS complexes with the library molecules have been
finally obtained using the microbatch-under-oil crystallization
technique. As it can be appreciated from the above description, the
crystallization screen has been extremely difficult and was a
cumbersome process, rich of failures. Under the microscope, all the
crystals obtained presented defects like cracks and lines indicating
internal disorder. Before data collection, crystals were transferred
to a cryoprotectant solution containing 20% ethylene glycol and 80%
precipitant solution (2.7 M ammonium sulfate and 0.1 M HEPES pH
7.0) and then flash frozen in liquid nitrogen.
Acknowledgment
We acknowledge the European Synchrotron Radiation Facility
(ESRF), especially beamline ID29-1 for data collection facilities.
Abbreviations
DIAD
DMF
diisopropylazodicarboxylate
dimethylformamide
DMSO dimethyl sulfoxide
4.4. Structure solution and refinement
dTMP
DTT
20-deoxythymidine-50-monophosphate
dithiothreitol
About 70 crystals were screened for diffraction at the ESRF
(European Synchrotron Radiation Facilities) beamline ID29-1. The
data collection demonstrated that all crystals were characterized by
severe splitting and suffered from radiation damage, with the
diffraction pattern decaying after about 100 s of exposure to X-rays
(0.5ꢀ oscillation images). Finally, we were able to obtain usable
datasets referring to (R)-2, (S)-12, 14 and MTX complexes. However,
all of them were complete to a maximum of about 75e80%, but the
missing data were randomly distributed in reciprocal space
without jeopardizing the quality of the electron density maps [46].
Intensities were integrated using the program Mosflm [59] and
scaled with the program Scala [60] or Aimless from the CCP4 Suite
[61]. Data collection statistics are reported in Table S2 (see
dUMP
EcTS
EEDQ
EDTA
EfTS
20-deoxyuridine-50-monophosphate
Escherichia coli Thymidylate Synthase
2-ethoxy-1-ethoxycarbonyl-1,2-dihydroquinoline
Ethylenediaminetetraacetic acid
Enterococcus faecalis Thymidylate Synthase
Ethanol
EtOH
Et3N
triethylamine
5-FTHF 5-formyl tetrahydrofolate
5-HMHF 5-hydroxymethyl-6-hydrofolate
hTS
Ki
LcTS
Thymidylate Synthase
Inhibition Constant
Lactobacillus casei Thymidylate Synthase
MeOH Methanol

A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
663
N.A. Colabufo, N. Galeotti, F. Corbo, R. Matucci, C. Ghelardini, C. Franchini,
Chiral aryloxyalkylamines: selective 5-HT1B/1D activation and analgesic ac-
tivity, ChemMedChem. 5 (2010) 696e704.
MTHF
NBS
SD
TFA
THF
TMS
TS
N5,N10-methylene tetrahydrofolate
N-bromosuccinimide; PEG, Poly(ethylene glycol)
Small Domain
Trifluoroacetic acid
Tetrahydrofuran
[20] A. De Luca, S. Talon, M. De Bellis, J.F. Desaphy, C. Franchini, G. Lentini,
A. Catalano, F. Corbo, V. Tortorella, D. Conte Camerino, Inhibition of skeletal
muscle sodium currents by mexiletine analogues: specific hydrophobic in-
teractions rather than lipophilia per se account for drug therapeutic profile,
Naunyn-Schmied. Arch. Pharmacol. 367 (2003) 318e327.
[21] A. Catalano, A. Carocci, G. Fracchiolla, C. Franchini, G. Lentini, V. Tortorella,
A. De Luca, M. De Bellis, J.F. Desaphy, D. Conte Camerino, Stereospecific syn-
thesis of para-hydroxymexiletine and sodium canne blocking activity evalu-
ation, Chirality 16 (2004) 72e78.
Trimethylsilane
Thymidylate Synthase.
Appendix A. Supplementary data
[22] A. Catalano, A. Carocci, M.M. Cavalluzzi, A. Di Mola, G. Lentini, A. Lovece,
A. Dipalma, T. Costanza, J.F. Desaphy, D. Conte Camerino, C. Franchini, Hy-
droxylated analogs of mexiletine as tools for structural-requirements inves-
tigation of the sodium canne blocking activity, Arch. Pharm. Chem. Life Sci.
343 (2010) 325e332.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2016.07.066.
[23] A. Catalano, J.F. Desaphy, G. Lentini, A. Carocci, A. Di Mola, C. Bruno,
R. Carbonara, A. De Palma, R. Budriesi, C. Ghelardini, M.G. Perrone,
N.A. Colabufo, D. Conte Camerino, C. Franchini, Synthesis and toxico-
pharmacological evaluation of m-hydroxymexiletine, the first metabolite of
mexiletine more potentthan the parent compound on voltage-gated sodium
channels, J. Med. Chem. 55 (2012) 1418e1422.
[24] A. Catalano, R. Budriesi, C. Bruno, A. Di Mola, I. Defrenza, M.M. Cavalluzzi,
M. Micucci, A. Carocci, F. Franchini, G. Lentini, Searching for new antiar-
rhythmic agents: evaluation of meta-hydroxymexiletine enantiomers, Eur. J.
Med. Chem. 65 (2013) 511e516.
[25] A. Carocci, A. Catalano, F. Corbo, A. Duranti, R. Amoroso, C. Franchini,
G. Lentini, V. Tortorella, Stereospecific synthesis of mexiletine and related
compounds: Mitsunobu versus Williamson reaction, Tetrahedron Asymm. 11
(2000) 3619e3634.
[26] C. Bruno, M.M. Cavalluzzi, A. Carocci, A. Catalano, C. Franchini, G. Lentini,
Capillary zone electrophoresis for separation and quantitative determination
of mexiletine and its mainphase I metabolites, Drug Metab. Lett. 7 (2013)
52e57.
[27] R. Pascale, A. Carocci, A. Catalano, G. Lentini, A. Spagnoletta, M.M. Cavalluzzi,
F. De Santis, A. De Palma, V. Scalera, C. Franchini, New N-(phenoxydecyl)
phthalimide derivatives displaying potent inhibition activity towards alpha-
glucosidase, Bioorg. Med. Chem. 18 (2010) 5903e5914.
[28] M. De Bellis, A. De Luca, F. Rana, M.M. Cavalluzzi, A. Catalano, C. Franchini,
V. Tortorella, D. Conte Camerino, Evaluation of the pharmacological activity of
the major mexiletine metabolites on skeletal muscle sodium currents, Br. J.
Pharmacol. 149 (2006) 300e310.
[29] C. Lamanna, A. Catalano, A. Carocci, A. Di Mola, C. Franchini, V. Tortorella,
M.S. Sinicropi, P.M.L. Vanderheyden, K.A. Watson, S. Sciabola, AT1 receptor
ligands: virtual-screening-based design with TOPP descriptors, synthesis, and
biological evaluation of pyrrolidine derivatives, ChemMedChem. 2 (2007)
1298e1310.
[30] A. Duranti, C. Franchini, G. Lentini, F. Loiodice, V. Tortorella, A. De Luca,
S. Pierno, D. Conte Camerino, 2000 Homologation of mexiletine alkyl chain
and stereoselective blockade of skeletal muscle sodium channels, Eur. J. Med.
Chem. 35 (2000) 147e156.
References
[1] F.H. Kayser, Safety aspects of enterococci from the medical point of view, Int. J.
Food Microb. 88 (2003) 255e262.
[2] T.J. Stout, D. Tondi, M. Rinaldi, D. Barlocco, P. Pecorari, D.V. Santi, I.D. Kuntz,
R.M. Stroud, B.K. Shoichet, M.P. Costi, Structure-based design of inhibitors
specific for bacterial thymidylate synthase, Biochemistry 38 (1999)
1607e1617.
[3] M.P. Costi, M. Rinaldi, D. Tondi, P. Pecorari, D. Barlocco, S. Ghelli, R.M. Stroud,
D.V. Santi, T.J. Stout, C. Musiu, E.M. Marangiu, A. Pani, D. Congiu, G.A. Loi, P. La
Colla, Phthalein derivatives as a new tool for selectivity in thymidylate syn-
thase inhibition, J. Med. Chem. 42 (1999) 2112e2124.
[4] S. Ghelli, M. Rinaldi, D. Barlocco, A. Gelain, P. Pecorari, D. Tondi, G. Rastelli,
M.P. Costi, Ortho-Halogen naphthaleins as specific inhibitors of Lactobacillus
casei thymidylate synthase. Conformational properties and biological activity,
Bioorg. Med. Chem. 11 (2003) 951e963.
[5] S. Calo, D. Tondi, A. Venturelli, S. Ferrari, P. Pecorari, M. Rinaldi, S. Ghelli,
M.P. Costi, A step further in the discovery of phthalein derivatives as thymi-
dylate synthase inhibitors, ARKIVOC (2004) 382e396.
[6] J.S. Finer-Moore, A.C. Anderson, R.H. O'Neil, M.P. Costi, S. Ferrari, J. Krucinski,
R. Stroud, The structure of Cryptococcus neoformans thymidylate synthase
suggests strategies for using target dynamics for species-specific inhibition,
Acta Crystallogr. Sect. D. Biol. Crystallogr. D61 (2005) 1320e1334.
[7] M.P. Costi, A. Gelain, D. Barlocco, S. Ghelli, F. Soragni, F. Raniero, T. Rossi,
A. Ruberto, C. Guillou, A. Cavazzuti, C. Casolari, S. Ferrari, Anti-bacterial agent
discovery using thymidylate synthase bio-library screening, J. Med. Chem. 49
(2006) 5958e5968.
[8] S. Mangani, L. Cancian, R. Leone, C. Pozzi, S. Lazzari, R. Luciani, S. Ferrari,
M.P. Costi, Identification of the binding modes of N-phenylphthalimides
inhibiting bacterial thymidylate synthase through X-ray crystallography
screening, J. Med. Chem. 54 (2011) 5454e5467.
[9] S. Ferrari, M. Ingrami, F. Soragni, R.C. Wade, M.P. Costi, Ligand-based discovery
of N-(1,3-dioxo-1H,3H-benzo[de]isochromen-5-yl)-carboxamide and sulfon-
amide derivatives as thymidylate synthase a inhibitors, Bioorg. Med. Chem.
Lett. 23 (2013) 663e668.
[31] C. Franchini, A. Carocci, A. Catalano, M.M. Cavalluzzi, F. Corbo, G. Lentini,
A. Scilimati, P. Tortorella, D. Conte Camerino, A. De Luca, Optically active
mexiletine analogues as stereoselective blockers of voltage-gated Naþ chan-
nels, J. Med. Chem. 46 (2003) 5238e5248.
[32] M.M. Cavalluzzi, A. Catalano, C. Bruno, A. Lovece, A. Carocci, F. Corbo,
C. Franchini, G. Lentini, V. Tortorella, Synthesis of (R)-, (S)-, and (RS)-
hydroxymethylmexiletine, one of the major metabolites of mexiletine, Tet-
rahedron Asymm. 18 (2007) 2409e2417.
ꢀ
[10] S. Ferrari, S. Calo, R. Leone, R. Luciani, L. Costantino, S. Sammak, F. Di Pisa,
C. Pozzi, S. Mangani, M.P. Costi, 2-Deoxyuridine 50-monophosphate substrate
displacement in thymidylate synthase through 6-hydroxy-2H-naphtho[1,8-
bc]furan-2-one derivatives, J. Med. Chem. 56 (2013) 9356e9360.
[11] M.P. Costi, S. Ferrari, A. Venturelli, S. Calo, D. Tondi, D. Barlocco, Thymidylate
synthase structure, function and implication in drug discovery, Curr. Med.
Chem. 12 (2005) 2241e2258.
[33] M. De Bellis, A. De Luca, J.F. Desaphy, J. Heiny, A. Kennedy, A. Carocci,
M.M. Cavalluzzi, G. Lentini, C. Franchini, D. Conte Camerino, Combined
modifications of mexiletine pharmacophores for new lead blockers of Nav1.4
channels, Biophys. J. 104 (2013) 344e354.
[34] A. Catalano, A. Carocci, I. Defrenza, M. Muraglia, A. Carrieri, F. Van Bambeke,
A. Rosato, F. Corbo, F. Franchini, 2-Aminobenzothiazole derivatives: search for
new antifungal agents, Eur. J. Med. Chem. 64 (2013) 357e364.
[35] A. Carocci, A. Catalano, C. Bruno, G. Lentini, C. Franchini, M. De Bellis, A. De
Luca, D. Conte Camerino, Synthesis and in vitro sodium channel blocking
activity evaluation of novel homochiral mexiletine analogs, Chirality 22
(2010) 299e307.
[12] C.W. Carreras, D.V. Santi, The catalytic mechanism and structure of thymi-
dylate synthase, Ann. Rev. Biochem. 64 (1995) 721e762.
[13] E. Chu, M.A. Callender, M.P. Farrell, J.C. Schmitz, Thymidylate synthase in-
hibitors as anticancer agents: from bench to bedside, Cancer Chemother.
Pharmacol. 52 (2003) S80eS89.
[14] C. Pozzi, S. Ferrari, D. Cortesi, R. Luciani, R.M. Stroud, A. Catalano, M.P. Costi,
S. Mangani, The structure of Enterococcus faecalis thymidylate synthase
provides clues about folate bacterial metabolism, Acta Crystallogr. D68 (2012)
1232e1241.
[15] A. Catalano, A. Carocci, M.S. Sinicropi, Mexiletine metabolites: a review, Curr.
Med. Chem. 22 (2015) 1400e1413.
[36] A. Catalano, A. Carocci, F. Corbo, C. Franchini, M. Muraglia, A. Scilimati, M. De
Bellis, A. De Luca, D. Conte Camerino, M.S. Sinicropi, V. Tortorella, Constrained
analogues of tocainide as potent skeletal muscle sodium channel blockers
towards the development of antimyotonic agents, Eur. J. Med. Chem. 43
(2008) 2535e2540.
[16] A. Carocci, A. Catalano, C. Bruno, A. Lovece, M.G. Roselli, M.M. Cavalluzzi, F. De
Santis, A. De Palma, M.R. Rusciano, M. Illario, C. Franchini, G. Lentini, N-
(Phenoxyalkyl) amidesas MT1 and MT2ligands: antioxidantproperties and
inhibition of Ca2þ/CaM-dependentkinase II, Bioorg. Med. Chem. 21 (2013)
847e851.
[37] A. Catalano, A. Carocci, G. Lentini, I. Defrenza, C. Bruno, C. Franchini, Stereo-
specific synthesis of m-hydroxymexiletine enantiomers, Drug. Metab. Lett. 6
(2012) 182e186.
[38] A. Catalano, A. Carocci, G. Lentini, I. Defrenza, M.M. Cavalluzzi, C. Franchini, An
improved synthesis of m-hydroxymexiletine, a potent mexiletine metabolite,
Drug. Metab. Lett. 6 (2012) 124e128.
[39] F. Maley, J. Pedersen-Lane, L. Changchien, Complete restoration of activity to
inactive mutants of Escherichia coli thymidylate synthase: evidence that E.
coli thymidylate synthase is a half-the-sites activity enzyme, Biochemistry 34
[17] J.F. Desaphy, A. Dipalma, T. Costanza, R. Carbonara, M.M. Di Nardo, A. Catalano,
A. Carocci, G. Lentini, C. Franchini, D. Conte Camerino, Molecular insights into
the local anesthetic receptor within voltage-gated sodium channels using
hydroxylated analogs of mexiletine, Front. Pharmacol. 3 (2012) 1e12.
[18] A. Carocci, A. Catalano, A. Lovece, G. Lentini, A. Duranti, V. Lucini, M. Pannacci,
F. Scaglione, C. Franchini, Design, synthesis, and pharmacological effects of
structurally simple ligands for MT1 and MT2melatoninreceptors, Bioorg. Med.
Chem. 18 (2010) 6496e6511.
[19] A. Carocci, G. Lentini, A. Catalano, M.M. Cavalluzzi, C. Bruno, M. Muraglia,

664
A. Catalano et al. / European Journal of Medicinal Chemistry 123 (2016) 649e664
[52] H. Takahashi, S. Sou, R. Yamasaki, M. Sodeoka, Y. Hashimoto, a-Glucosidase
(1995) 1469e1474.
[40] A.C. Anderson, R.H. O'Neil, W.L. DeLano, R.M. Stroud, The structural mecha-
nism for half-the-sites reactivity in an enzyme, thymidylate synthase, in-
inhibitors with a phthalimide skeleton: structure activity relationship study,
Chem. Pharm. Bull. 48 (2000) 1974e1983.
ꢀ
volves
a
relay of changes between subunits, Biochemistry 38 (1999)
[53] N. Aguilar, A. Moyano, M.A. Pericas, A. Riera, A mild, selective, PyBOP medi-
13829e13836.
ated procedure for the conversion of primary amines into phthalimides,
Synthesis (1998) 313e316.
[54] A.A. Bahajaj, M.H. Moore, J.M. Vernon, Asymmetric synthesis of spiro 2-
pyrrolidin-5-ones, 2-piperidin-6-ones and 1-isoindolin-3-ones. Part 1: N-
Acyliminium ion cyclisations with an internal arene nucleophile, Tetrahedron
60 (2004) 1235e1246.
[55] R.A. Glennon, N.A. Naiman, M.E. Pierson, J.D. Smith, A.M. Ismaiel, M. Titeler,
R.A. Lyon, N-(Phthalimidoalkyl) derivatives of serotonergic agents: a common
interaction at 5-HT1A serotonin binding sites? J. Med. Chem. 32 (1989)
1921e1926.
[56] M. Benvenuti, S. Mangani, Crystallization of soluble proteins in vapor diffusion
for x-ray crystallography, Nat. Protoc. 2 (2007) 1633e1651.
[57] N.E. Chayen, A novel technique for containerless protein crystallization, Pro-
tein Eng. 9 (1996) 927e929.
[41] D.L. Birdsall, J. Finer-Moore, R.M. Stroud, The only active mutant of thymi-
dylate synthase D169, a residue far from the site of methyl transfer, dem-
onstrates the exquisite nature of enzyme specificity, Protein Eng. 16 (2003)
229e240.
[42] S. Chen, D.H. Shin, R. Pufan, R. Kim, S.H. Kim, Crystal structure of meth-
enyltetrahydrofolate synthetase from Mycoplasma pneumoniae (GI:
13508087) at 2.2 A resolution, Proteins 56 (2004) 839e843 (PDB: 1U3G).
[43] B.K. Shoichet, R.M. Stroud, D.V. Santi, I.D. Kuntz, K.M. Perry, Structure-based
discovery of inhibitors of thymidylate synthase, Science 259 (1993)
1445e1450.
[44] Z. Wang, T. Abeysinghe, J.S. Finer-Moore, R.M. Stroud, A. Kohen, A remote
mutation affects the hydride transfer by disrupting concerted protein motions
in thymidylate synthase, J. Am. Chem. Soc. 134 (2012) 17722e17730.
[45] A. Tochowicz, S. Dalziel, O. Eidam, J.D. O'Connell, S. Griner, J.S. Finer-Moore,
R.M. Stroud, Development and binding mode assessment of N-[4-[2-propyn-
1-yl[(6S)-4,6,7,8-tetrahydro-2-(hydroxymethyl)-4-oxo-3H-cyclopenta[g]qui-
[58] N.E. Chayen, The role of oil in macromolecular crystallization, Structure 5
(1997) 1269e1274.
[59] A.G. Leslie, The integration of macromolecular diffraction data, Acta Crys-
tallogr. D. Biol. Crystallogr. 62 (2006) 48e57.
nazolin-6-yl]amino]benzoyl]-l-g-glutamyl-D-glutamic acid (BGC 945), a novel
thymidylate synthase inhibitor that targets tumor cells, J. Med. Chem. 56
(2013) 5446e5455.
[60] P. Evans, Scaling and assessment of data quality, Acta Crystallogr. D. Biol.
Crystallogr. 62 (2006) 72e82.
[46] Z. Dauter, Data-collection strategies, Acta Crystallogr. D. Biol. Crystallogr. 55
(1999) 1703e1717.
[47] A. Catalano, A. Carocci, A. Di Mola, P.M.L. Vanderheyden, C. Franchini, 1-
Pentanoyl-N-{[20-(1H-tetrazol-5-yl)biphenyl-4-yl]methyl}-pyrrolidine-2-
carboxamide: investigation of structural variations, Arch. Pharm. Chem. Life
Sci. 344 (2011) 617e626.
[48] C. Bruno, A. Carocci, A. Catalano, M.M. Cavalluzzi, F. Corbo, C. Franchini,
G. Lentini, V. Tortorella, Facile, alternative route to lubeluzole, itsenantiomer
and the racemate, Chirality 18 (2006) 227e231.
[49] M. Muraglia, M. De Bellis, A. Catalano, A. Carocci, C. Franchini, A. Carrieri,
C. Fortugno, C. Bertucci, J.F. Desaphy, A. De Luca, D. Conte Camerino, F. Corbo,
N-aryl-2,6-dimethylbenzamides, a new generation of tocainide analogues as
blockers of skeletal muscle voltage-gated sodium channels, J. Med. Chem. 57
(2014) 2589e2600.
[50] W.V. Murray, M. P. Wacther, Substituted N-biphenylyl lactams, Patent WO
1993/5182288.
[51] M.V. Khedkar, S.R. Khan, K.P. Dhake, B.M. Bhanage, Carbonylative cyclization
of o-halobenzoic acids for synthesis of N-substituted phthalimides using
polymer-supported palladium-N-heterocyclic carbene as an efficient, het-
erogeneous, and reusable catalyst, Synthesis 44 (2012) 2623e2629.
[61] M.D. Winn, C.C. Ballard, K.D. Cowtan, E.J. Dodson, P. Emsley, P.R. Evans,
R.M. Keegan, E.B. Krissinel, A.G. Leslie, A. McCoy, S.J. McNicholas,
G.N. Murshudov, N.S. Pannu, E.A. Potterton, H.R. Powell, R.J. Read, A. Vagin,
K.S. Wilson, Overview of theCCP4 suite and current developments, Acta
Crystallogr. D. Biol. Crystallogr. 67 (2011) 235e242.
[62] A. Vagin, A. Teplyakov, Molecular replacement with MOLREP, Acta Crystallogr.
D. Biol. Crystallogr. 66 (2010) 22e25.
[63] G.N. Murshudov, A.A. Vagin, E.J. Dodson, Refinement of macromolecular
structures by the maximum-likelihood method, Acta Crystallogr. D53 (1997)
240e255.
[64] R.J. Morris, A. Perrakis, V.S. Lamzin, ARP/wARP and automatic interpretation of
protein electron density maps, Methods Enzymol. 374 (2003) 229e244.
[65] R.A. Laskowski, M.W. MacArthur, D.S. Moss, J.M. Thornton, PROCHECK: a
program to check the stereochemical quality of protein structures, J. Appl.
Crystallogr. 26 (1993) 283e291.
[66] P. Emsley, B. Lohkamp, W.G. Scott, K. Cowtan, Features and development of
coot, Acta Crystallogr. D. Biol. Crystallogr. 66 (2010) 486e501.
[67] S. McNicholas, E. Potterton, K.S. Wilson, M.E.M. Noble, Presenting your
structures: the CCP4mg molecular-graphics software, Acta Cryst. D67 (2011)
386e394.