Discovery of in vitro antitubercular agents through in silico ligand-based approaches

Article abstract of DOI:10.1016/j.ejmech.2016.05.032

The development of new anti-tubercular agents represents a constant challenge mostly due to the insurgency of resistance to the currently available drugs. In this study, a set of 60 molecules were selected by screening the Asinex and the ZINC collections and an in house library by means of in silico ligand-based approaches. Biological assays in Mycobacterium tuberculosis H37Ra ATCC 25177 strain highlighted (±)-1-(4-chlorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(3,4-dichlorophenyl)piperazine-1-carboxylate (5i) and 3-(4-chlorophenyl)-5-(2,4-dimethylpyrimidin-5-yl)-2-methylpyrazolo[1.5-a]pyrimidin-7(4H)-one (42) as the most potent compounds, having a Minimum Inhibitory Concentration (MIC) of 4 and 2 μg/4g/mL respectively. These molecules represent a good starting point for further optimization of effective anti-TB agents.

Full text of DOI:10.1016/j.ejmech.2016.05.032

Accepted Manuscript
Discovery of in vitro antitubercular agents through in silico ligand-based approaches
Daniela De Vita, Fabiana Pandolfi, Roberto Cirilli, Luigi Scipione, Roberto Di Santo,
Laura Friggeri, Mattia Mori, Diego Fiorucci, Giorgio Maccari, Robert Selwyne Arul
Christopher, Claudio Zamperini, Valentina Pau, Alessandro De Logu, Silvano
Tortorella, Maurizio Botta
PII:
S0223-5234(16)30428-7
10.1016/j.ejmech.2016.05.032
EJMECH 8624
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 23 March 2016
Revised Date: 12 May 2016
Accepted Date: 16 May 2016
Please cite this article as: D. De Vita, F. Pandolfi, R. Cirilli, L. Scipione, R. Di Santo, L. Friggeri, M. Mori,
D. Fiorucci, G. Maccari, R.S.A. Christopher, C. Zamperini, V. Pau, A. De Logu, S. Tortorella, M. Botta,
Discovery of in vitro antitubercular agents through in silico ligand-based approaches, European Journal
of Medicinal Chemistry (2016), doi: 10.1016/j.ejmech.2016.05.032.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Discovery of in Vitro Antitubercular Agents Through in Silico Ligand-Based
Approaches
[
a]
[a]
[b]
[a]
Daniela De Vita, Fabiana Pandolfi, Roberto Cirilli, Luigi Scipione, Roberto Di
[
a,c]
[a]
[d,e]
[d]
[d]
Santo,
Laura Friggeri, Mattia Mori,
Diego Fiorucci, Giorgio Maccari, Robert
[
d]
[d]
[g]
[g]
Selwyne Arul Christopher, Claudio Zamperini, Valentina Pau, Alessandro De Logu,
[
a]
[d,f]
Silvano Tortorella,* and Maurizio Botta*
[
a]
Dipartimento di Chimica e Tecnologie del Farmaco, Sapienza Università di Roma,
Piazzale Aldo Moro 5, 00185 Roma (Italy)
[
b]
Dipartimento del Farmaco, Istituto Superiore di Sanità, Viale Regina Elena 299, 00161
Roma (Italy)
[
c]
Istituto Pasteur-Fondazione Cenci Bolognetti, Dipartimento di Chimica e Tecnologie del
Farmaco, Sapienza Università di Roma, Piazzale Aldo Moro 5, 00185 Roma (Italy)
[
d]
Dipartimento di Biotecnologie, Chimica e Farmacia, Università degli Studi di Siena, Via
Aldo Moro 2, 53019 Siena (Italy)
[
e]
Center for Life Nano Science@Sapienza, Istituto Italiano di Tecnologia, viale Regina
Elena 291, 00161 Roma (Italy)
[
f]
Sbarro Institute for Cancer Research & Molecular Medicine, Center for Biotechnology,
College of Science & Technology, Temple University, BioLife Science Building, Suite 333,
1
900 N 12th Street, Philadelphia, PA 19122 (USA)
[
g]
Dipartimento di Scienze della Vita e dell'Ambiente, Università degli Studi di Cagliari, Via
Porcell 4, 09124 Cagliari, Italy
Corresponding authors: maurizio.botta@unisi.it; silvano.tortorella@uniroma1.it
*

ACCEPTED MANUSCRIPT
Abstract
The development of new anti-tubercular agents represents a constant challenge mostly
due to the insurgency of resistance to the currently available drugs. In this study, a set of
6
0 molecules were selected by screening the Asinex and the ZINC collections and an in
house library by means of in silico ligand-based approaches. Biological assays in M.
tuberculosis H37Ra ATCC 25177 strain highlighted (±)-1-(4-chlorophenyl)-2-(1H-imidazol-
1
-yl)ethyl-4-(3,4-dichlorophenyl)piperazine-1-carboxylate (5i) and 3-(4-chlorophenyl)-5-
(2,4-dimethylpyrimidin-5-yl)-2-methylpyrazolo[1.5-a]pyrimidin-7(4H)-one (42) as the most
potent compounds, having a Minimum Inhibitory Concentration (MIC) of 4 and 2 µg/mL
respectively. These molecules represent a good starting point for further optimization of
effective anti-TB agents.
Keywords
Virtual screening; anti-tubercular agents; azoles; phenyl-pyrazolopyrimidinones.
Highlights
•
•
•
•
A refined anti-tubercular pharmacophore model was described;
60 molecules were selected by virtual screening of commercial and in house library;
Anti-tubercular activity was evaluated on hit compounds;
Two new lead compounds were identified with 2 and 4 µg/mL MIC.

ACCEPTED MANUSCRIPT
1
. Introduction
Tuberculosis (TB) is a widespread infectious disease caused by the bacillus
Mycobacterium tuberculosis. It typically affects the lungs (pulmonary TB) but can also
affect other parts of the body (extrapulmonary TB). M. tuberculosis is spread from person
to person through the air when people with pulmonary TB expel bacteria, by coughing and
sneezing.[1] To date, TB remains a major global health problem and it is responsible for a
large number of deaths. According to the latest projections worldwide, 9.6 million people
are estimated to have fallen ill with TB in 2014: 5.4 million men, 3.2 million women and 1.0
million children. Globally, 12% of the 9.6 million new TB cases in 2014 were HIV-
positive.[1] Generally, a relatively small proportion of people infected with M. tuberculosis
will develop the disease. It is estimated that one-third of the global population is infected
without evidence of clinically manifested active TB (latent TB infection).[2] In this form, M.
tuberculosis lives in a non-replicating stage in lungs’ cavities with caseous necrosis and
poor access to oxygen[3] and infected people do not have TB disease and symptoms.[4]
Therefore, people with latent form are often treated to prevent the development of active
TB.[5] The current treatment of active and latent TB consists in a drugs combination
(isoniazid, rifampicin, rifapentine, ethambutol and pyrazinamide) administered over six
months. Long treatment period and poor patients' adherence to the current treatment lead
to the emergence of M. tuberculosis strains resistant to some or all available anti-TB
drugs, called MDR (multidrug-resistant), XDR (extensively drug resistant) and TDR (totally
drug resistant) could arise.[5, 6]
In 2012, the diarylquinoline bedaquiline has been approved by US Food and Drug
Administration as part of combination therapy to treat MDR pulmonary TB.[7]
Nevertheless, based on the overall limitations described above, new antibiotics to treat TB
infection are urgently needed.

ACCEPTED MANUSCRIPT
Recently, many successful examples of anti-TB hit and lead compounds have been
reported by many research groups,[8-11] including our own. In particular, the identification
of new anti-TB compounds with innovative structures has been pursued by ligand-based
molecular modeling tools such as pharmacophore models, which have been used for
screening virtual libraries of small molecules and to drive the rational optimization of hit
and lead candidates. In brief detail, a 5-hydroxy-pyrazole compound,[12] characterized by
a MIC of 25 ꢀg/mL, first emerged as a hit candidate from virtual screening. Subsequent
optimization of the hit and synthesis of a library of derivatives led us to obtain
pyrazoles,[13] pyrazolone,[14] and azole analogues with polycyclic structures[15] as
interesting compounds with enhanced anti-TB activity and endowed with MIC values
ranging between 4 and 12 ꢀg/mL. In addition, a series of pyrrole derivatives endowed with
strong anti-TB activity and MIC values lower than 0.125 ꢀg/mL have been also developed
by pharmacophore modeling.[16, 17] However, as implicit in most common ligand-based
approaches, the macromolecular target(s) for these classes of hit and lead compounds is
still unknown and, for the moment, only a few hypotheses on the possible mechanism of
action have been formulated.[16, 17] Nevertheless, these pharmacophores proved to be
highly efficient in identifying hit compounds endowed with anti-TB activity and having
different chemical composition, as well as in driving the optimization of these hits up to
valuable lead compounds.[16, 17]
By following our research interest in developing anti-TB lead compounds and our research
strategies relying on pharmacophore modeling and screening, here we describe a new
pharmacophore model that was obtained by refining the previous ligand-base
pharmacophores based on literature data. Accordingly, this work represents the logical
prosecution of the previous ligand-based studies performed by our own research
group.[12-17] First, two chemical libraries were screened through the above
pharmacophore, namely i) the Asinex database of commercially available compounds, and

ACCEPTED MANUSCRIPT
ii) our in house library including azole derivatives previously synthesized as antifungal and
antiprotozoal agents. The first set of molecules was further optimized in silico by screening
the ZINC database in a ligand-based strategy. The second set of molecules has been
included because recently some azole antifungal agents have been described to possess
also anti-TB activity, thus highlighting the potentiality for exploiting the azole scaffold in
anti-TB medicinal chemistry.[18] Virtual hits have been tested in vitro in a model of TB
infection and rough structure-activity relationships (SAR) have been afforded for two
different series of molecules, most probably endowed with two different mechanisms of
action but providing remarkable anti-TB effects in vitro.
2
2
. Results and Discussion
.1. Generation of a common feature pharmacophore model
A total of 71 small molecules, which have been synthesized in our group and already
described in previous publications as inhibitors in vitro of the growth of M.tuberculosis
H37Rv ATCC 27294 strain were collected and used as training set.[12-17] Notably, these
molecules were tested in vitro by the same research group in the same experimental
condition, thus allowing the generation of a reliable and unbiased ligand-based
pharmacophore model. Out of these 71 small molecules, those having a MIC < 12 ꢀg/mL
were arbitrarily classified as “active”, those having a MIC comprised between 12 and 32
ꢀ
ꢀ
g/mL were arbitrarily classified as moderately active, whereas those having a MIC > 32
g/mL were considered as inactive during the generation of the common feature
pharmacophore model. No further descriptors were considered in this classification, thus
allowing the generation of a pharmacophore model composed of conventional features as
well as exclusion volumes, which are placed according to the biological activity of
compounds in the training set (Figure 1).

ACCEPTED MANUSCRIPT
Our pharmacophoric analysis was carried out using the CATALYST-HipHop procedure,
which evaluates the common features and the hypothetical ideal geometry of compounds
included in the training set ligands. This was performed in a three step procedure that led
to the generation of a number of common feature pharmacophore models. The highest
ranked pharmacophore hypothesis, which was composed of 5 features (1 Ring Aromatic
(RA), 1 H-Bond Acceptor (HBA), 2 Hydrophobic Aromatic (HydArom1 and HydArom2) and
1
Hydrophobic aliphatic (HydAli)) and 23 excluded volumes (Figure 1), proved to map
perfectly the most potent compounds of the training set (Figure 1), and was therefore
selected as query in virtual screening.
[
figure 1]
2
.2. Virtual screening
The whole ASINEX database collection (757900 compounds) and an in house library of
azole compounds were screened towards the selected pharmacophore model, with the
aim to identify new chemotypes of M. tuberculosis inhibitors. Pharmacophore screening
was performed with Discovery Studio from Accelrys[19] and filtered ligands were sorted
according to the FitValue score.
By screening the ASINEX database, 28298 compounds having FitValue above 2.50 were
filtered by the pharmacophore (3.7% of the ASINEX database). Since the cell wall of M.
tuberculosis is known to be noticeably hydrophobic, highly aqueous soluble molecules
were discarded by setting logP filter greater than 4.50, thus retaining 10486 compounds
(1.43% of the ASINEX database). Further, the number of compounds was reduced
through shape similarity searching performed by Discovery Studio. In particular, the shape
of the most potent compound included in the training set (MIC = 0.125 ꢀg/mL, Supporting
Information) was used to filter the remaining 10486 small molecules. Out of them, 575
compounds having at least 50% shape similarity (Tanimoto score higher than 0.5) with the

ACCEPTED MANUSCRIPT
most active compound were selected. The final shortlist of compounds was achieved by
visual inspection of the pharmacophoric alignment of these molecules as well as by
considering structural diversity and custom filters. In detail, based on previous works
performed by our own group that highlighted the para-chlorophenyl system as crucial for
activity,[13] scaffolds having the para-chlorophenyl moiety mapping to the RA feature
(Figure 1) were deemed top priority. In addition, scaffolds having a Sulphur atom mapping
to the HBA feature were discarded, because of the weak tendency to accept H-bond of
Sulphur,[20, 21] while compounds having a chiral center with unspecified geometry (most
likely they are sold as a racemic mixture) were selected only if both enantiomers were able
to map satisfactorily the pharmacophore model. This selection led to a final list of 27 small
molecules, namely 10-36 that were purchased from Asinex and tested in vitro (see
Supporting Information).
The in house library of 96 azole compounds was screened by using the same procedure
described above and detailed in the Experimental Section. In this particular case, based
on the limited dimension of the screening library, the filter based on the molecular shape of
the most potent anti-TB compound included within the training set was not applied. Indeed,
the compounds endowed with the highest FitValue after pharmacophore screening,
namely 1-4, 5a-f,l and 7 were selected and submitted to in vitro tests. Considering the
reduced structural differences in the phenyl moiety of the 2-(1H-imidazol-1-yl)-1-
phenylethanol derivatives analyzed by the pharmacophoric model, we decided to enlarge
the set of studied compounds exploring more substituents, such as a bulky lipophilic ring
i.e. naphthyl and biphenyl (5n and 6) or a small and polar nitro group (5m).
2
.3. Biological activity
Minimum Inhibitory Concentration (MIC) of each tested compound against M. tuberculosis
H37Ra ATCC 25177 was determined by the agar dilution method, as previously reported

ACCEPTED MANUSCRIPT
with slight modifications.[22] The biological assays results are reported in Table 1. Overall,
the most promising molecule selected by virtual screening of the ASINEX database of
commercially available compounds was 24, which inhibited the growth of M. tuberculosis
with a MIC of 8 ꢀg/mL, and was therefore selected as hit compound for further deepening
of the phenyl-pyrazolopyrimidinone scaffold (see below). In contrast screening the in
house library of azole compounds provided 5i as the most potent compound, having a MIC
of 4 ꢀg/mL. This molecule was further investigated, in order to depict the possible
mechanism of action of this series (see below).
2
.4. Synthesis of azole derivatives
The azole derivatives were prepared as racemates or as (S)-enantiomers, according to
suggestions coming from pharmacophore screening. Generally the azoles were prepared
following the same procedure, however for synthesizing 2-(1H-imidazol-1-yl)-1-
phenylethanols as racemic mixture the ketone group was reduced to alcohol using NaBH4,
while RuCl(p-cymene)[(R,R)-Ts-DPEN] was used for preparing (S)-2-(1H-imidazol-1-
yl)phenylethanols differently substituted.[23, 24] The final azoles (with the exception of (S)-
5
n) were synthesized using two different procedures, Method A or B. (Scheme 1). In
particular the activation of the alcohol group was performed using NaH, (Method A) or
triphosgene (Method B).[25, 26] More details are described in experimental section.
[
Scheme 1]
Compounds (S)-5n, was prepared by activation of the amino function of 1-(3,4-
dichlorophenyl)piperazine using triphosgene and pyridine; then, the corresponding 4-(3,4-
dichlorophenyl)piperazine-1-carbonyl chloride was condensed with 1-([1,1'-biphenyl]-4-yl)-
2
-(1H-imidazol-1-yl)ethanol (51) (Scheme 2). Finally, 8 and 9 were synthesized by
condensation of alcohols 53 and 54 respectively with 4-(3,4-dichlorophenyl)piperazine-1-
carbonyl chloride (Scheme 2). The 1-(4-chlorophenyl)-2-phenylethanol 53 was obtained by

ACCEPTED MANUSCRIPT
reduction with NaBH of commercial 1-(4-chlorophenyl)-2-phenylethanone. Otherwise, the
4
1
-(4-chlorophenyl)-2-(1H-pyrrol-1-yl)ethanol
54
was
prepared
starting
from
hexamethylenetetramine and 2-bromo-1-(4-chlorophenyl)ethanone and subsequent
treatment with HCl 37% in EtOH. The corresponding oxoethanaminium was reacted with
2
,5-dimethoxytetrahydrofuran according to the Clauson-Kaas procedure,[27] then 1-(4-
chlorophenyl)-2-(1H-pyrrol-1-yl)ethanone was reduced to 54 with NaBH4.
Scheme 2]
.5. SAR analysis for the azole derivatives
[
2
All tested azole derivatives share the common 2-(1H-imidazol-1-yl)-1-phenylethanolic
moiety; the main structural differences regarding: i) the length and polarity of the side
chain, comprising esters (1,4-7) and carbamates (5a-m,6); ii) substitution of the aromatic
ring of the 1-phenylethanolic moiety.
The activity data show that the compounds with the same 2-(1H-imidazol-1-yl)-1-
phenylethanolic moiety and different side chain have activities ranging from 4 to 64 ꢀg/mL
suggesting that the modification of the side chain is an extremely important factor for
obtaining variations of the in vitro biological activity.
Furthermore, it was generally observed that the presence of a para-chloro substituent
produces an improvement of activity compared to a para-fluoro group (5c vs 5d, 5i vs 5h)
in compound series with an identical side chain, and this was in agreement with our
previous findings.[13] Whereas, the presence of a second chlorine atom in the ortho-
position such as in (S)-5l reduces the anti-TB activity with respect to 5i. Moreover, the
replacement of the para-chloro of 5i with a para-nitro group [(S)-5m], a phenyl group [(S)-
5
n] or the substitution of the phenyl ring with naphtyl [(S)-6] invariably reduced the anti-TB
potency.

ACCEPTED MANUSCRIPT
Modifications of the 2-(1H-imidazol-1-yl)-1-phenylethanolic moiety led to a major changes
in the biological activity. Indeed, the replacement of imidazole ring of 5i with phenyl
(compound 8) or a pyrrole (compound 9) resulted in a loss of activity against M.
tuberculosis.
2
.6. Metabolic stability assay of 5i
In vitro metabolic stability was experimentally evaluated for the most interesting compound
5i) in order to have a reliable indication of its plasma level after oral administration.
(
The metabolic stability reported in Table 2 is expressed as percentage of the unmodified
parent drug remaining in the mixture while the metabolites (M1) were separated in HPLC
and analysed via mass spectrometry. The information on the metabolite mass obtained
with the latter spectroscopic studies were then integrated with the in silico metabolites
prediction obtained with the software Metasite[28-30] thus allowing us to confirm the
structure of each metabolite (metabolites structures are reported in the Figure S5,
Supporting Information).
[
Table 2]
This compound shows a high metabolic stability in this experimental condition. Only one
metabolite was observed, that is the product of the CYP-dependent mono-oxygenation
reaction at the α carbon position. This metabolite was found to be the 1% of the
unmodified parent drug. Moreover, experimental metabolite M1 was correctly predicted by
Metasite and ranked in the top six positions.
2
.7. Optimization of the 3-phenylpyrazolo[1-5a]pyrimidin-7(4H)-one scaffold and
SAR analysis

ACCEPTED MANUSCRIPT
The 3-phenylpyrazolo[1-5a]pyrimidin-7(4H)-one derivative 24 showed a good anti-TB
activity when compared to other compounds resulting from the pharmacophoric approach.
Accordingly, with the aim to improve its anti-TB activity, a number of chemical
modifications of this scaffold were planned. To this end, we used the shape matching
method implemented in the ROCS program.[31, 32] Starting from our hit compound 24,
we performed a survey of the available literature and found that three additional molecules
belonging to the same scaffold and endowed with anti-TB activity have been disclosed by
Mao et al. and Ananthan et al. (Figure 2A).[33, 34] These molecules and 24 were thus
used to create a shape query by means of the vROCS program (Figure 2A, B). The four
molecules were aligned using an in house developed python script which superimposes
the heavy atoms keeping the best combination in term of physicochemical features. This
alignment output was used for the query generation. The selection of the biological active
conformation is not possible for these molecules because the target is unknown. However,
since they have less than four rotatable bonds, the possible conformations are limited and
very similar between themselves. Thus, a conformation that could be adopted by all the
active molecules was trusted as reliable.
[
Figure 2]
The query was validated as described in the Supporting Information, and used to virtually
screening the “All now” subset of the ZINC database (approx. 9 million compounds).[35]
This database was pre-filtered using Filter program,[36, 37] part of the OpenEye Suite,
with a modified version of the drug-like ruleset to retain molecules with a higher probability
of becoming an antitubercular agent. In particular, we reduced the minimum molecular
weight of the compounds that could pass the filter, increased ClogP value and filtered out
molecules with reactive or unwanted features. The rationale behind the increase of the
ClogP value was the need to select molecules able to cross the thick and lipophilic barrier

ACCEPTED MANUSCRIPT
of the M. tuberculosis. After filtration, the remaining 5250000 molecules were submitted to
conformational analysis by Omega2 (OpenEye suite),[36, 37] and were screened towards
the ROCS query. The first 5000 molecules sorted by the TanimotoCombo score were
selected and visually inspected. Among them, the13 most promising compounds, namely
3
7–49 (Figure 2C and Supporting Information), were selected by taking into account the
decoration of the scaffold, with particular attention to the variable part of the four molecules
used to build our query (Figure 2A). Indeed, 37-49 bear aromatic substituents directly
linked to the scaffold, or aromatic substituents linked through linker of varying length, or
substituents with aliphatic chains or aliphatic rings. Bioassays showed that some of these
molecules have activity in the 8-16 µg/mL range, comparable with the starting hit 24 (Table
3
). Interestingly, molecule 42 showed activity of 2 µg/ml representing a substantial
improvement of the starting hit 24 (Table 3 and Figure 2C).
After hit optimization, we produced a simple structure activity relationship the 3-
phenylpyrazolo[1-5a]pyrimidin-7(4H)-one scaffold (Figure 3). The aliphatic substituents in -
R (Figure 3) not improved the activity, while the use of an etheric function provided anti-TB
activity comparable to the hit compound 24. In the same position -R, some aliphatic linkers
connecting an aromatic substituents were introduced, providing anti-TB activity
comparable to the starting hit compound, whereas the introduction of a dimethyl pyrimidine
moiety led to the best anti-TB activity of the series. The replacement of the methyl group
1
in -R with hydrogen, substituted phenyl rings and a methoxymethyl chain, proved to be
2
uneffective. In the R position the electron withdrawing chlorine atom was substituted with
one or two electron donor O-methyl groups or with hydrogen. All these modifications did
not improve the anti-TB activity of 24. Overall, the most relevant modification was the
insertion of a dimethyl pyrimidine substituent on the R position, which led to the compound
4
2 that is endowed with a MIC of 2 µg/mL.

ACCEPTED MANUSCRIPT
[
Figure 3]
3
. CONCLUSIONS
In this work, the Asinex database of commercially available compounds and an in house
library of azoles were screened in silico through a pharmacophore model. A set of 47
molecules were first selected and investigated in vitro as candidate inhibitors of the growth
of M. tuberculosis H37Ra ATCC 25177.
The screening of the azole library provided 5i as the most promising compound, having a
MIC of 4 ꢀg/mL. All the screened azoles possess a 2-(1H-imidazol-1-yl)-1-phenylethanolic
skeleton and the main structural diversity concerns the side chain linked to the alcohol or
the 1-phenylethyl moiety. Side chain modification can produce pronounced differences in
anti-TB activity; compounds with the same 2-(1H-imidazol-1-yl)-1-phenylethanolic moiety
and different side chain have activities ranging from 4 to 64 ꢀg/mL
Moreover, it was generally observed that, in compounds with the same side chain, the
presence of a para-chloro substituent produces an improvement of activity compared to
para-fluoro group (5c vs 5d, 5i vs 5h). Furthermore, the introduction of a second chlorine
atom in the ortho-position such as (S)-5l reduces the anti-TB activity with respect to 5i.
The substitution of the para-chloro group of 5i with para-nitro or phenyl group such as in
(
S)-5m and (S)-5n reduced the anti-TB potency as well as the substitution of the phenyl
ring of 5i with naphtyl moiety [(S)-6)]. The replacement of imidazole ring of 5i with phenyl
compound 8) or a pyrrole (compound 9) resulted in a loss of activity against M.
(
tuberculosis.
Compound 24 was found as the most active compound among the ASINEX database with
a MIC of 8 ꢀg/mL and was therefore selected as hit for further deepening of the 3-
phenylpyrazolo[1-5a]pyrimidin-7(4H)-one scaffold. Starting from 24 and three analogues

ACCEPTED MANUSCRIPT
endowed with anti-TB activity retrieved from the literature, a shape- and chemistry-based
query was built with ROCS. Screening the ZINC database with this query provided 13
compounds (37-49), which were selected taking into account the variations in the
decoration of the scaffold. Some of these molecules showed an in vitro antitubercular
activity in the 8-16 µg/mL range, comparable with the starting hit 24. Interestingly, 42
represented a significant improvement of the starting hit 24 with an activity of 2 µg/mL.
In conclusion, we have identified two new hits showing a good anti-mycobacterial activity
in vitro, which are attractive candidates for further development of new anti-tubercular
agents.
4
4
. Experimental Section
.1. Chemistry
General: All reagents and solvents were of high analytical grade and were purchased from
Sigma-Aldrich (Milano, Italy). The 2-(1H-imidazol-1-yl)-1-arylethanones and racemic 2-
(1H-imidazol-1-yl)-1-phenylethanols were prepared according to literature procedure.[23]
Melting points were determined on Tottoli (Buchi) or Kofler apparatus (where specified)
and are uncorrected. Infrared spectra were recorded on a Spectrum One ATR Perkin
Elmer FT-IR spectrometer. NMR spectra were acquired on a Bruker AVANCE-400 or
AVANCE-200 spectrometer in DMSO-d , CD OD or CDCl at 27 °C ; chemical shift values
6
3
3
are given in δ (ppm) relatively to TMS as internal reference. Coupling constants are given
in Hz. Abbreviations: s=singlet, d=doublet, t=triplet, q=quartet, m=multiplet, dd=doublet of
doublet. Mass spectrometric experiments were carried out with a 2000 Q TRAP instrument
(Applied Biosys-tems); a commercial hybrid triple-quadrupole linear ion-trap mass
spectrometer (Q1q2QLIT), equipped with an ESI source and a syringe pump have been
-5
used. The examined compounds were previously dissolved in methanol (10 M) and

ACCEPTED MANUSCRIPT
aqueous HCl was added just before the injection. The molecular peaks (m/z) have been
+
observed as [M+H] ; for chlorine containing compounds, the highest intensity peak was
reported. Elemental analyses were obtained by a PE 2400 (Perkin-Elmer) analyzer and
the analytical results were within ±0.4 % of the theoretical values for all compounds. The
analytical HPLC apparatus consisted of a PerkinElmer (Norwalk, CT, USA) 200 LC pump
equipped with a Rheodyne (Cotati, CA, USA) injector, a 20 µL sample loop, an HPLC
Dionex CC-100 oven (Sunnyvale, CA, USA) and a Jasco (Jasco, Tokyo, Japan) Model CD
2
095 Plus UV/CD detector. Chiral HPLC analysis were performed using the commercially
available column Chiralpak IC (250 mm x 4.6 mm i.d.) and Chiralpak IB-3 (250 mm x 4.6
mm i.d.).
4
.1.1. (±)-1-(4-chlorophenyl)-2-(1H-imidazol-1-yl)ethyl-2-phenoxyacetate (1):
Sodium hydride (13 mg, 0.54 mmol) was added to a suspension of (±)-1-(4-chlorophenyl)-
-(1H-imidazol-1-yl)ethanol (120 mg, 0.54 mmol) in dry CH CN (5 mL). After 2 h at RT, the
2
3
phenoxyacetyl chloride (220 µL, 1.59 mmol) was dropwise added and the reaction mixture
was stirred overnight at RT. The solvent was removed under reduced pressure and the
obtained residue was dissolved in CH Cl (10 mL) and washed with saturated Na CO
2
2
2
3
solution (2 x 10mL). The separated organic layer was dried over Na SO and evaporated
2
4
under reduced pressure. The residue was crystallized from EtOAc to give pure 1 as a
-1
1
white solid (183 mg, 95%); mp: 103-105 °C ; IR (neat , ʋ cm ): 1751(C=O); H NMR (400
MHz, DMSO-d ) δ (ppm): 4.42 (d, J = 5.5 Hz, 2H), 4.83 (d, J = 16.7 Hz, 1H), 4.91 (d, J =
6
1
1
6.7 Hz, 1H), 6.11 (t, J = 5.5 Hz, 1H), 6.82-6.87 (m, 3H), 6.97 (t, J = 7.4 Hz, 1H), 7.13 (s,
H), 7.27 (dd, J = 8.5, 7.4 Hz, 2H), 7.35 (d, J = 8.5 Hz, 2H), 7.43 (d, J = 8.5 Hz, 2H), 7.54
+
(
s, 1H); MS (ESI): m/z calcd for C H ClN O : 357.10 [M+H] , found 356.90; Anal. calcd
19
17
2
3
for C H ClN O : C 63.96, H 4.80, N 7.85, found: C 63.86, H 4.81, N 7.83.
19
17
2
3

ACCEPTED MANUSCRIPT
4
.1.2. (S)-1-(4-fluorophenyl)-2-(1H-imidazol-1-yl)ethyl-(3-(1H-imidazol-1-
yl)propyl)carbamate [(S)-2]:
S)-1-(4-fluorophenyl)-2-(1H-imidazol-1-yl)ethanol (100 mg, 0.48 mmol) was suspended in
mL of dry CH CN, then triphosgene (71 mg, 0.24 mmol) was added. The mixture was
(
5
3
stirred overnight at RT. The reaction mixture was treated with Et O, until the complete
2
precipitation of a white solid. The solvent was removed by decantation; the white solid was
washed with Et O (2 × 5 mL) and dissolved in anhydrous CH CN. Then, TEA (222 µL, 1.6
2
3
mmol) and 3-(1H-imidazol-1-yl)propan-1-amine (95 µL, 0.8 mmol) were added. The
reaction was stirred overnight at RT. The crude mixture was diluted with H O (5 mL), and
2
the aqueous layer was extracted with CHCl (3 × 10 mL). The combined organic layers,
3
dried under Na SO , were evaporated under reduced pressure. The crude product was
2
4
purified by silica gel column chromatography (CH Cl / MeOH 80:20) to give (S)-2 (138
2
2
-1
1
mg, 80%, e.e.>99%); mp: 101-103 °C ; IR (neat, ʋ cm ): 1710 (C=O); H NMR (400 MHz,
CD OD) δ (ppm): 1.93 (m, J = 6.8 Hz, 2H), 3.05 (m, 2H), 4.00 (t, J = 6.8 Hz, 2H), 4.40 (m,
3
2
H), 5.92 (m, 1H), 6.94 (s, 1H), 6.96 (s, 1H), 7.11 (m, 4H), 7.36 (m, 2H), 7.56 (s, 1H), 7.62
13
(
s, 1H); C NMR (100 MHz,CDCl ) δ (ppm): 31.2, 38.1, 44.4, 52.0, 74.1, 115.8 (J = 22.0
3
Hz), 118.8, 119.8, 128.0 (J = 8.0 Hz), 128.9, 129.3, 133.0 (J = 2.0 Hz), 137.2, 137.9,
+
1
3
1
55.2, 162.8 (J = 247.0 Hz); MS (ESI): m/z calcd for C H FN O : 358.17 [M+H] , found:
18
20
5
2
58.13; Anal. calcd for C H FN O : C 60.49, H 5.64, N 19.60, found: C 60.49, H 5.65, N
18
20
5
2
9.63.
.1.3. (S)-1-(4-chlorophenyl)-2-(1H-imidazol-1-yl)ethyl-3-(trifluoromethyl) benzoate
4
[(S)-3]:
Sodium hydride (13 mg, 0.54 mmol) was added to a stirred suspension of (S)-1-(4-
chlorophenyl)-2-(1H-imidazol-1-yl)ethanol (120 mg, 0.54 mmol) in dry CH CN(5 mL). After
3
2
h at RT the sodium alkoxide was treated with 3-(trifluoromethyl)benzoyl chloride (111 µL,

ACCEPTED MANUSCRIPT
0
.75 mmol). The reaction mixture was stirred for 24 h at RT. The solvent was removed
under reduced pressure and the residue was treated with CH Cl (20 mL) and washed with
2
2
saturated Na CO solution (2x10mL). The organic layer, dried over Na SO , was
2
3
2
4
evaporated under reduced pressure. The crude yellow oil was purified by silica gel column
chromatography using CH Cl / MeOH 90:10 as eluent to obtain (S)-3 as colorless oil (114
2
2
-1
1
mg, 56%, e.e.>99%); IR (neat, ʋ cm ): 1729 (C=O); H NMR (400 MHz, CDCl ) δ (ppm):
3
4
7
1
.48-4.41 (m, 2H), 6.20 (t, J = 5.3 Hz, 1H), 6.86 (s, 1H), 7.04 (s,1H), 7.21-7.28 (m, 2H),
.35-7.39 (m, 3H), 7.64 (t, J = 7.8 Hz, 1H), 7.88 (d, 1H, J = 7.8 Hz), 8.22 (d, J = 7.8 Hz,
+
H), 8.29 (s, 1H); MS (ESI): m/z calcd for C H ClF N O : 395.08 [M+H] , found: 395.00;
19
14
3
2
2
Anal. calcd for C H ClF N O : C 57.81, H 3.57, N 7.10, found: C 57.89, H 3.56, N 7.12.
19
14
3
2
2
4
.1.4. (S)-1-(4-fluorophenyl)-2-(1H-imidazol-1-yl)ethyl (4-isopropylphenyl)carbamate
(S)-4]:
Sodium hydride (13 mg, 0.53 mmol) was added to a suspension of (S)-1-(4-fluorophenyl)-
-(1H-imidazol-1-yl)ethanol (110 mg, 0.53 mmol) in dry CH CN (5 mL). After 2 h at RT, 4-
[
2
3
isopropylphenyl isocyanate (254 µL, 1.59 mmol) was added and the suspension was
stirred for 24 h at RT. The solvent was evaporated under reduced pressure, and the solid
residue was washed with MeOH (3×3 mL). The mother liquor was dried and purified by
silica gel column chromatography (CH Cl / MeOH 90:10) to give (S)-4 as a white solid (88
2
2
-1
1
mg, 45%, e.e. > 99%); mp: 200-202 °C ; IR (neat, ʋ cm ): 1715 (C=O); H NMR (400 MHz,
CD OD) δ (ppm): 1.22 (d, J = 6.9 Hz, 6H), 2.85 (m, 1H), 4.44 (m, 2H), 6.00 (t, J = 5.4 Hz,
3
1
H), 6.94 (s, 1H), 7.08-7.15 (m, 5H), 7.30 (d, J = 8.5 Hz, 2H), 7.38 (m, 2H), 7.57 (s, 1H);
13
C NMR (100 MHz, CD OD) δ (ppm): 23.1, 33.4, 51.2, 74.2, 115.1 (J = 21.0 Hz), 118.7,
3
1
1
;
20.2, 126.3, 127.4, 128.0 (J = 9.0 Hz), 133.6 (J = 3.0 Hz), 136.0, 137.8, 143.8, 153.0,
+
62.7 (J = 245.0 Hz); MS (ESI): m/z calcd for C H FN O : 368.18 [M+H] , found: 368.40
21
22
3
2
Anal. calcd for C H FN O : C 68.65, H 6.04, N 11.44, found: C 68.39, H 6.05, N 11.42.
21
22
3
2

ACCEPTED MANUSCRIPT
4
.1.5. (S)-1-(4-Fluorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(furan-2-
carbonyl)piperazine-1-carboxylate [(S)-5a]:
S)-1-(4-fluorophenyl)-2-(1H-imidazol-1-yl)ethanol (206 mg, 1 mmol) was suspended in 5
(
mL of dry CH CN, then triphosgene (148 mg, 0.5 mmol) was added. The mixture was
3
stirred overnight at RT. After the obtained mixture was treated with Et O until the complete
2
precipitation of a white precipitate. Subsequently the solvent was decanted off, the
precipitate was suspended in dry CH CN and added with TEA (278 µL, 2.0 mmol), then
3
with 1-(2-furoyl)piperazine (144 mg, 0.8 mmol). After 12 h at RT, the solvent was removed
under reduced pressure, the obtained residue was treated with CH Cl (10 mL) and
2
2
extracted with saturated aqueous Na CO . The organic layer dried over Na SO and
2
3
2
4
evaporated under vacuum give a crude residue which was purified by silica gel column
chromatography (CH Cl / MeOH 80:20) resulting in (S)-5a as a waxy solid (257 mg, 78%,
2
2
-1
1
e.e. > 99%); IR (neat, ʋ cm ): 1699, 1621 (C=O); H NMR (400 MHz, CD OD) δ (ppm):
3
3
7
5
1
.50-3.79 (m, 8H), 4.42-4.49 (m,2H), 5.99 (t, J = 6.0 Hz, 1H), 6.61 (s, 1H), 6.97 (s, 1H),
13
.08-7.14 (m,5H), 7.38-7.41 (m, 2H), 7.71 (s, 1H); C NMR (100 MHz, CD OD) δ (ppm):
3
2.5, 55.3, 73.5, 76.8, 112.5, 116.5 (J = 22.0 Hz), 118.0, 121.5, 129.0, 129.6 (J = 8.0 Hz),
34.7 (J = 3.0 Hz), 139.2, 146.1, 148.2, 155.3, 161.2, 164.2 (J = 244.0 Hz); MS (ESI): m/z
+
calcd for C H FN O 413.16 [M+H] , found: 412.99; Anal. calcd for C H FN O : C
21
21
4
4
21 21
4
4
6
1.16, H 5.13, N 13.59, found: C 61.39, H 5.14, N 13.42.
4
.1.6. (S)-1-(4-Chlorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(furan-2-
carbonyl)piperazine-1-carboxylate [(S)-5b]:
According to the procedure described for (S)-5a, (S)-1-(4-chlorophenyl)-2-(1H-imidazol-1-
yl)ethanol (111 mg, 0.5 mmol) was treated overnight with triphosgene (74 mg, 0.25 mmol),
then TEA (139 µL, 1.0 mmol) and 1-(2-furoyl)piperazine (72 mg, 0.4 mmol) were added.
The residue was purified by silica gel column chromatography using CH Cl / MeOH 80:20
2
2

ACCEPTED MANUSCRIPT
-1
as eluent to obtain (S)-5b as a waxy solid (103 mg, 60 %, e.e. > 99%); IR (neat, ʋ cm ):
1
1
701, 1622 (C=O); H NMR (400 MHz, CDCl ) δ (ppm): 3.09 (m, 4H), 3.64 (m, 4H), 4.30
3
(m, 2H), 5.92 (t, J = 6.0 Hz, 1H), 6.75 (s, 1H), 6.82 (d, J = 8.3 Hz, 2H), 7.03 (m, 2H), 7.15-
1
3
7
1
.22 (m, 4H), 7.31 (s, 1H); C NMR (100 MHz, DMSO-d ) δ (ppm): 43.9, 43.9, 52.6, 74.3,
6
11.8, 116.4, 121.4, 122.9, 128.6, 129.1, 133.6, 136.4, 136.7, 145.4, 147.2, 153.3, 158.9;
+
MS (ESI): m/z calcd for C H ClN O 429.13 [M+H] , found: 428.98; Anal. calcd for
21
21
4
4
C H ClN O : C 58.81, H 4.94, N 13.06, found: C 58.49, H 4.96, N 13.24.
21
21
4
4
4
.1.7. (S)-1-(4-fluorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(4-nitrophenyl)piperazine-1-
carboxylate [(S)-5c]:
According to the procedure described for (S)-5a, (S)-1-(4-fluorophenyl)-2-(1H-imidazol-1-
yl)ethanol (103 mg, 0.5 mmol) was treated overnight with triphosgene (74 mg, 0.25 mmol),
then TEA (139 µL, 1.0 mmol) and 1-(4-nitrophenyl)piperazine (83 mg, 0.4 mmol) were
added. The residue was purified by silica gel column chromatography using CH Cl /
2
2
MeOH 80:20 to obtain (S)-5c as a yellow solid (123 mg, 70 %, e.e. > 99%); mp: 168-169
-1
1
°
C; IR (neat, ʋ cm ): 1704 (C=O); H NMR (400 MHz, CDCl ) δ (ppm): 3.45 (m, 4H), 3.70
3
(m, 4H), 4.42 (m, 2H), 6.00 (t, J = 5.2 Hz, 1H,), 6.83 (m, 3H), 7.09 (m, 3H), 7.24 (m, 2H),
1
3
7
7
3
.87 (s, 1H), 8.16 (d, J = 8.9 Hz, 2H); C NMR (100 MHz, CDCl ) δ (ppm): 46.8, 52.1,
3
4.9, 77.2, 113.2, 116.1 (J = 22.0 Hz), 119.8, 125.9, 128.0 (J = 8.0 Hz), 128.2, 132.4 (J =
.0 Hz), 137.5, 139.3, 153.5, 154.4, 162.9 (J = 247.0 Hz); MS (ESI): m/z calcd for
+
C H FN O 440.17 [M+H] , found: 439.93; Anal. calcd for C H ClN O : C 60.13, H
22
22
5
4
21 21
4
4
5
.05, N 15.94, found: C 60.31, H 5.06, N 15.99.
.1.8. (±)-1-(4-chlorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(4-nitrophenyl)piperazine-1-
4
carboxylate (5d):
According to the procedure described for (S)-5a, (±)-1-(4-chlorophenyl)-2-(1H-imidazol-1-
yl)ethanol (111 mg, 0.5 mmol) was treated with triphosgene (111 mg, 0.5 mmol); then TEA

ACCEPTED MANUSCRIPT
(139 µL, 1.0 mmol) and 1-(4-nitrophenyl)piperazine (83 mg, 0.4 mmol) were added. The
residue was purified by silica gel column chromatography using CH Cl / MeOH 80:20 as
2
2
-1
eluent to obtain 5d as a yellow solid (146 mg, 80 %); mp: 188-190 °C ; IR (neat, ʋ cm ):
1
1
5
7
5
1
704 (C=O); H NMR (400 MHz, CDCl ) δ (ppm): 3.42 (m, 4H), 3.68 (m, 4H), 4.33 (m, 2H),
3
,94 (t, J = 5.5 Hz, 1H), 6.81 (m,3H), 7.04 (s, 1H), 7.14 (d, J = 10.5 Hz, 2H), 7.34 (m, 2H),
13
.45 (s, 1H), 8.14 (d, J = 9.3 Hz, 2H); C NMR (100 MHz, CDCl ) δ (ppm): 43.3, 46.8,
3
1.7, 75.0, 113.1, 119.7, 125.9, 127.5, 129.1, 129.2, 135.0, 135.2, 137.6, 139.2, 153.5,
+
54.4; MS (ESI): m/z calcd for C H ClN O 456.14 [M+H] , found: 455.97; Anal. calcd for
22
22
5
4
C H ClN O : C 57.96, H 4.86, N 15.36, found: C 58.18, H 4.85, N 15.41.
22
22
5
4
4
1
.1.9. (S)-1-(4-Fluorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(4-chlorophenyl)piperazine-
-carboxylate [(S)-5e]:
According to the procedure described for (S)-5a, (S)-1-(4-fluorophenyl)-2-(1H-imidazol-1-
yl)ethanol (103 mg, 0.5 mmol) was treated with triphosgene (74 mg, 0.25 mmol); then TEA
(139 µL, 1.0 mmol) and 1-(4-chlorophenyl)piperazine (78 mg, 0.4 mmol) were added. The
residue was purified by silica gel column chromatography using CH Cl / MeOH 80:20 as
2
2
-1
eluent to obtain (S)-5e as a waxy solid (158 mg, 92 %, e.e. > 99%); IR (neat, ʋ cm ): 1703
1
(
C=O); H NMR (400 MHz, DMSO-d ) δ (ppm): 3.04 (m, 4H), 3.38 (m, 4H), 4.36 (m, 2H),
6
5
5
1
1
.83 (m, 1H), 6.84 (s, 1H), 6.95 (d, J = 8.9 Hz, 2H), 7.13-7.24 (m, 5H), 7.38 (dd, J = 7.8,
13
.5 Hz, 2H), 7.56 (s, 1H); C NMR (100 MHz, DMSO-d ) δ (ppm): 43.2, 48.1, 50.5, 74.7,
6
15.3 (J = 22 Hz), 117.5, 120.1, 123.0, 128.1, 128.4 (J = 9 Hz), 128.7, 133.9 (J = 2 Hz),
37.8, 149.5, 153.2, 161.8 (J = 243 Hz); MS (ESI): m/z calcd for C H ClFN O 429.15
22
22
4
2
+
[
M+H] , found: 428.89; Anal. calcd for C H ClFN O : C 61.61, H 5.17, N 13.06, found: C
22 22 4 2
6
1.41, H 5.16, N 13.12.
4
1
.1.10. (S)-1-(4-Chlorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(4-chlorophenyl)piperazine-
-carboxylate [(S)-5f]:

ACCEPTED MANUSCRIPT
According to the procedure described for (S)-5a, (S)-1-(4-chlorophenyl)-2-(1H-imidazol-1-
yl)ethanol (113 mg, 0.5 mmol) was treated with triphosgene (74 mg, 0.25 mmol); then TEA
(13.9 µL, 1.0 mmol) and 1-(4-chlorophenyl)piperazine (78 mg, 0.4 mmol) were added. The
residue was purified by silica gel column chromatography using CH Cl / MeOH 80:20 as
2
2
eluent to obtain (S)-5f as a white solid (157 mg, 88 %, e.e. > 99%); mp: 113-116 °C ; IR
-1
1
(
neat, ʋ cm ): 1702 (C=O); H NMR (400 MHz, DMSO-d ) δ (ppm): 3.09 (m, 4H), 3.65 (m,
6
4
H), 4.38 (m, 2H), 5.88 (m, 1H), 6.84 (s, 1H), 6.97 (d, J = 8.9 Hz, 2H), 7.13 (s, 1H), 7.25
13
(d, J = 8.9 Hz, 2H), 7.37 (d, J = 8.3 Hz, 2H), 7.45 (d, J = 8.3 Hz, 2H), 7.53 (s, 1H);
C
NMR (100 MHz, CD OD) δ (ppm): 43.4, 48.9, 51.1, 75.1, 117.7, 120.1, 124.7, 127.4,
3
1
27.7, 128.2, 128.7, 134.1, 136.2, 137.8, 149.9, 153.8; MS (ESI): m/z calcd for
+
C H Cl N O 445.12 [M+H] , found: 444.87; Anal. calcd for C H Cl N O : C 59.33, H
22
22
2
4
2
22 22
2
4
2
4
.98, N 12.58, found: C 59.09, H 4.99, N 12.52.
.1.11. (S)-2-(1H-imidazol-1-yl)-1-phenylethyl-4-(3,4-dichlorophenyl)piperazine-1-
4
carboxylate [(S)-5g]:
According to the procedure described for (S)-5a, (S)-2-(1H-imidazol-1-yl)-1-phenylethanol
(
300 mg, 1.6 mmol) in 15 mL of dry CH CN was treated with triphosgene (236 mg, 0.8
3
mmol); then TEA (442 µL, 3.2 mmol) and 1-(3,4-dichlorophenyl)piperazine (293 mg, 1.3
mmol) were added. The residue was purified by silica gel column chromatography using
EtOAc / MeOH 90:10 as eluent; the obtained waxy residue was triturated with Et O to give
2
-1
a white solid (48 mg, 27 %, e.e. > 99%); mp: 118-120 °C ; IR (neat, ʋ cm ): 1700 (C=O);
1
H NMR (400 MHz, DMSO-d ) δ (ppm): 3.15 (m, 4H), 3.64 (m, 4H), 4.38 (m, 2H), 5.88 (m,
6
1
7
7
H), 6.85 (s, 1H), 6.96 (dd, J = 8.8, 2.8 Hz, 1H), 7.14 (s, 1H), 7.16 (d, J = 2.8 Hz, 2H),
13
.33-7.43 (m ,5H), 7.55 (s, 1H); C NMR (50 MHz, Acetone-d ) δ (ppm): 43.3, 48.1, 51.2,
6
5.6, 116.0, 117.2, 119.7, 121.2, 126.3 128.2, 128.5, 130.5, 132.1, 137.9, 138.2, 151.1,

ACCEPTED MANUSCRIPT
+
1
53.5; MS (ESI): m/z calcd for C H Cl N O 445.12 [M+H] , found: 445.00; Anal. calcd
22 22 2 4 2
for C H Cl N O : C 59.33, H 4.98, N 12.58, found: C 59.55, H 4.99, N 12.61.
22
22
2
4
2
4
.1.12. (S)-1-(4-Fluorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(3,4-
dichlorophenyl)piperazine-1-carboxylate [(S)-5h]:
According to the procedure described for (S)-5a, (S)-1-(4-fluorophenyl)-2-(1H-imidazol-1-
yl)ethanol (103 mg, 0.5 mmol) was treated with triphosgene (72 mg, 0.25 mmol); then TEA
(139 µL, 1.0 mmol) and 1-(3,4-dichlorophenyl)piperazine (92 mg, 0.4 mmol) were added.
The residue was purified by silica gel column chromatography using CH Cl / MeOH 90:10
2
2
as eluent; the obtained waxy residue was triturated with Et O to give a white solid (111
2
-1
1
mg, 60 %, e.e. > 99%); mp: 134-135 °C ; IR (neat, ʋ cm ): 1700 (C=O); H NMR (400 MHz,
CDCl ) δ (ppm): 3.15 (m, 4H), 3.66 (m, 4H), 4.32 (m, 2H), 5.95 (t, J = 5.2 Hz, 1H), 6.75 (m,
3
13
2
H), 6.98 (d, J = 2.8Hz, 1H), 7.05 (m, 3H), 7.18 (m ,2H), 7.32 (m, 2H); C NMR (100 MHz,
CDCl ) δ (ppm): 43.6, 48.8, 51.8, 74.9, 115.9 (J = 21 Hz), 116.0, 118.0, 119.6, 123.2,
3
1
2
27.9 (J = 8 Hz), 129.5, 130.6, 132.7 (J = 2 Hz), 132.9, 137.7, 150.3, 153.6, 162.8 (J =
+
47 Hz); MS (ESI): m/z calcd for C H Cl FN O 463.11 [M+H] , found: 463.00; Anal.
22
21
2
4
2
calcd for C H Cl FN O : C 57.03, H 4.57, N 12.09, found: C 57.23, H 4.57, N 12.12.
22
21
2
4
2
4
.1.13. (±)-1-(4-Chlorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(3,4-
dichlorophenyl)piperazine-1-carboxylate (5i):
According to the procedure described for (S)-5a, (±)-1-(4-chlorophenyl)-2-(1H-imidazol-1-
yl)ethanol (100 mg, 0.45 mmol) was treated with triphosgene (74 mg, 0.25 mmol); then
TEA (125 µL, 0.9 mmol) and 1-(3,4-dichlorophenyl)piperazine (83 mg, 0.36 mmol) were
added. The residue was purified by silica gel column chromatography using EtOAc /
MeOH 95:5 as eluent; the obtained waxy residue was triturated with Et O to give a white
2
-1
1
solid (52 mg, 30 %); mp: 143-145 °C ; IR (neat, ʋ cm ): 1702 (C=O); H NMR (400 MHz,
CD OD) δ (ppm): 3.16 (m, 4H), 3.57-3.74 (m, 4H), 4.46 (d, J = 5.8 Hz, 2H), 5.98 (t, J = 5.8
3

ACCEPTED MANUSCRIPT
Hz, 1H), 6.92 (dd, J = 8.9, 2.8 Hz, 1H), 6.96 (s, 1H), 7.11 (d, J = 2.8 Hz, 1H), 7.13 (s, 1H),
1
3
7
7
1
.34-7.41 (m, 5H), 7.58 (s, 1H); C NMR (100 MHz, Acetone-d ) δ (ppm): 43.3, 48.1, 51.0,
6
5.1, 116.1, 117.2, 120.2, 121.2, 128.0, 128.2, 128.6, 130.6, 132.1, 133.6, 136.8, 138.1,
+
51.0, 153.7; MS (ESI): m/z calcd for C H Cl N O 479.08 [M+H] , found: 479.08; Anal.
22
21
3
4
2
calcd for C H Cl N O : C 55.07, H 4.41, N 11.68, found: C 55.27, H 4.40, N 11.70.
22
21
3
4
2
4
.1.14. (S)-1-(2,4-Dichlorophenyl)-2-(1H-imidazol-1-yl)ethyl-4-(3,4-
dichlorophenyl)piperazine-1-carboxylate [(S)-5l]:
According to the procedure described for (S)-5a, (S)-1-(2,4-dichlorophenyl)-2-(1H-
imidazol-1-yl)ethanol (100 mg, 0.4 mmol) was treated with triphosgene (59 mg, 0.2 mmol);
then TEA (111 µL, 0.8 mmol) and 1-(3,4-dichlorophenyl)piperazine (74 mg, 0.32 mmol)
were added. The residue was purified by silica gel column chromatography using CH Cl /
2
2
-
MeOH 95:5 as eluent to obtain a white solid (113 mg, 61 %); mp: 61-63 °C ; IR (neat, ʋ cm
1
1
)
: 1707 (C=O); H NMR (400 MHz, CDCl ) δ (ppm): 3.15 (m, 4H), 3.65 (m, 4H), 4.29 (dd,
3
J = 14.9, 6.1 Hz, 1H), 4.43 (dd, J = 14.9, 3.5 Hz, 1H), 6.29 (dd, J = 6.0, 2.2 Hz, 1H), 6.72-
6
2
1
1
.76 (m, 2H), 6.94-6.98 (m, 2H), 7.02 (s, 1H), 7.21 (dd, J = 8.3, 2.0 Hz, 1H), 7.30-7.34 (m,
13
H), 7.45 (d, J = 2.0 Hz, 1H); C NMR (100 MHz, CDCl ) δ (ppm): 43.7, 48.8, 49.7, 72.4,
3
16.0, 118.0, 119.7, 123.3, 127.8, 127.9, 129.5, 129.8, 130.6, 132.4, 132.9, 133.2, 135.2,
+
37.7, 150.2, 153.1; MS (ESI): m/z calcd for C H Cl N O 513.04 [M+H] , found: 514.73;
22
20
4
4
2
Anal. calcd for C H Cl N O : C 51.38, H 3.92, N 10.90, found: C 51.39, H 3.91, N 10.95.
22
20
4
4
2
4
.1.15. (S)-2-(1H-imidazol-1-yl)-1-(4-nitrophenyl)ethyl-4-(3,4-
dichlorophenyl)piperazine-1-carboxylate [(S)-5m]:
According to the procedure described for (S)-5a, (S)-1-(4-nitrophenyl)-2-(1H-imidazol-1-
yl)ethanol (50) (130 mg, 0.56 mmol) was triphosgene (83 mg, 0.28 mmol); then TEA (156
µL, 1.12 mmol) and 1-(3,4-dichlorophenyl)piperazine (104 mg, 0.45 mmol) were added.
The residue was purified by silica gel column chromatography using CH Cl / MeOH / n-
2
2

ACCEPTED MANUSCRIPT
hexane 90:10:10 as eluent to obtain a yellow solid (55 mg, 25 %); mp:163-165 °C (Kofler
-1
1
apparatus); IR (neat, ʋ cm ): 1707 (C=O); H NMR (400 MHz, CDCl ) δ (ppm): 3.16 (m,
3
4
6
H), 3.63-3.71 (m, 4H), 4.39 (dd, J = 14.6, 6.1 Hz, 1H), 4.42 (dd, J = 14.6, 4.6 Hz, 1H),
.06 (m, 1H), 6.77 (m, 2H), 6.98 (d, J = 2.7 Hz, 1H), 7.05 (s, 1H), 7.28-7.38 (m, 4H), 8.24
13
(
d, J = 6.85 Hz, 2H); C NMR (100 MHz, CDCl ) δ (ppm): 43.6, 48.9, 51.4, 74.6, 116.1,
3
1
18.1, 119.5, 123.5, 124.2, 127.0, 129.8, 130.7, 133.0, 137.7, 144.0, 148.1, 150.2, 153.2;
+
MS (ESI): m/z calcd for C H Cl N O 490.10 [M+H] , found: 489.53; Anal. calcd for
22
21
2
5
4
C H Cl N O : C 53.89, H 4.32, N 14.28, found: C 53.77, H 4.33, N 14.23.
22
21
2
5
4
4
.1.16 (S)-1-([1,1'-biphenyl]-4-yl)-2-(1H-imidazol-1-yl)ethyl-4-(3,4-
dichlorophenyl)piperazine-1-carboxylate [(S)-5n]:
-(3,4-Dichlorophenyl)piperazine (173 mg, 0.75 mmol) in CH Cl (1 mL) was added
1
2
2
dropwise to a solution of triphosgene (83mg, 0.28 mmol) e pyridine (130µL, 1.61 mmol) in
CH Cl (2 mL). After 3 h at RT, the mixture was diluted with CH Cl (5 mL), then the
2
2
2
2
corresponding solution was washed with H O (2x5mL). The organic layer dried over
2
Na SO and evaporated under vacuum to give a yellow oil used without further purification,
2
4
dissolved in CH Cl (5 mL) and added to a solution of (S)-1-(biphenyl-4-yl)-2-(1H-imidazol-
2
2
1
-yl)ethanol (51) (100 mg, 0.38 mmol), TEA (47 µL, 0.41 mmol) and DMAP (20 % mmol).
From the suspension resulting after 12 h at RT the solid was removed by filtration; the
obtained filtrate was diluted with CH Cl and washed with saturated aqueous NaHCO .
2
2
3
The organic layer, dried under Na SO , was evaporated under reduced pressure and was
2
4
purified by silica gel column chromatography (CH Cl / MeOH 90:10) to give solid yellow.
2
2
Further purification was carried out by semipreparative HPLC on a Lichrosorb RP-18(5µm)
column, 250-10 mm. λ=254nm (elution profile: MeOH:H O 90:10, 2.0 mL/min) to give a
2
-1
1
white solid (25 mg, 16%); mp: 74-75 °C ; IR (neat, ʋ cm ): 1702 (C=O); H NMR (400 MHz,
DMSO-d ) δ (ppm): 3.16 (m, 4H), 3.66 (m, 4H), 4.38-4.49 (m, 2H), 5.92 (m, 1H), 6.88 (s,
6

ACCEPTED MANUSCRIPT
1
H), 6.96 (dd, J = 6.4, 2.8 Hz, 1H), 7.17 (d, J = 2.8 Hz, 1H), 7.20 (s, 1H), 7.41-7.48 (m,
13
6
H), 7.67 (s, 1H), 7.69 (m, 4H); C NMR (100 MHz, CDCl ) δ (ppm): 43.6, 48.9, 51.9,
3
7
5.4, 116.0, 118.0, 122.8, 123.2, 123.6, 126.6, 127.1, 127.7, 128.9, 130.6, 133.0, 135.8,
+
1
40.3, 141.9, 150.3, 153.7; MS (ESI): m/z calcd for C H Cl N O 521.15 [M+H] , found:
28
26
2
4
2
5
20.80; Anal. calcd for C H Cl N O : C 64.49, H 5.03, N 10.74, found: C 64.91, H 5.02,
28 26 2 4 2
N 10.70.
4
.1.17. (S)-2-(1H-imidazol-1-yl)-1-(naphthalen-1-yl)ethyl-4-(3,4-
dichlorophenyl)piperazine-1-carboxylate [(S)-6]:
According to the procedure described for (S)-5a, (S)-1-(naphtalen-1-yl)-2-(1H-imidazol-1-
yl)ethanol (52) (100 mg, 0.42 mmol) was activated with triphosgene (62 mg, 0.28 mmol);
then TEA (117 µL, 0.84 mmol) and 1-(3,4-dichlorophenyl)piperazine (76 mg, 0.34 mmol)
were added. After silica gel column chromatography (CH Cl / MeOH 90:10) the fractions
2
2
with R = 0.65 were collected and purified by a further silica gel column chromatography on
f
silica gel using CH CN / CH Cl 50:50 as eluent; the fractions with R = 0.43 were collected
3
2
2
f
to furnish a white solid, subsequently crystallized from EtOAc (34 mg, 20 %); mp: 152-
-1
1
1
54 °C (Kofler apparatus); IR (neat, ʋ cm ): 1707 (C=O); H NMR (400 MHz, CDCl ) δ
3
(ppm): 3.15 (m, 4H), 3.62-3.73 (m, 4H), 4.43 (m, 2H), 6.13 (t, J = 5.6 Hz, 1H), 6.75 (dd, J =
9
1
.2, 2.8 Hz, 1H), 6.80 (s, 1H), 6.98 (d, J = 2.8 Hz, 1H), 7.03 (s, 1H), 7.31 (s, 1H), 7.33 (s,
13
H), 7.36 (s, 1H), 7.52-7.54 (m, 2H), 7.69 (s, 1H), 7-81-7.88 (m, 3H); C NMR (100 MHz,
CDCl ) δ (ppm): 43.5, 48.9, 51.8, 75.8, 116.0, 118.0, 119.7, 123.3, 125.8, 126.7, 127.8,
3
1
28,1 128.9, 129.5, 130.6, 133.0, 133.1, 133.4, 134.2, 136.8, 137.8, 150.3, 153.7, 156.0;
+
MS (ESI): m/z calcd for C H Cl N O 495.13 [M+H] , found: 494.47; Anal. calcd for
26
24
2
4
2
C H Cl N O : C 63.04, H 4.88, N 11.31, found: C 63.14, H 4.89, N 11.35.
26
24
2
4
2
4
.1.18. (S)-1-(4-chlorophenyl)-2-(1H-imidazol-1-yl)ethyl-(4-((4-
nitrophenyl)thio)phenyl)carbamate [(S)-7]:

ACCEPTED MANUSCRIPT
According to the procedure described for (S)-5a, (S)-1-(4-chlorophenyl)-2-(1H-imidazol-1-
yl)ethanol 222 mg, 1.0 mmol) was treated with triphosgene (148 mg, 0.5 mmol) in dry
CH CN (5 mL); then TEA (417 µL, 3.0 mmol) and 4-((4-nitrophenyl)thio)aniline (197 mg,
3
0
.8 mmol) were added. The residue was purified by silica gel column chromatography
using CHCl / MeOH 90:10 as eluent to obtain a yellow solid (336 mg, 85 %, e.e. > 98% );
3
-1
1
mp: 100-102 °C ; IR (neat, ʋ cm ): 1727 (C=O); H NMR (400 MHz, CD OD) δ (ppm): 4.49
3
(m, 2H), 6.04 (t, J = 6.2Hz, 1H), 6.96 (s, 1H), 7.18 (m, 3H), 7.38 (m, 4H), 7.49 (d, J = 8.8
13
Hz, 2H), 7.57 (m, 3H), 8.08 (d, J = 9.05 Hz, 2H); C NMR (100 MHz, DMSO-d ) δ (ppm):
6
5
1
0.7, 74.6, 120.2, 120.6, 122.2, 124.7, 126.5, 128.6, 128.7, 129.0, 133.4, 136.5, 137.1,
+
38.4, 141.1, 145.3, 149.2, 152.6; MS (ESI): m/z calcd for C H ClN O S 495.09 [M+H] ,
24
19
4
4
found: 495.07; Anal. calcd for C H ClN O S: C 58.24, H 3.87, N 11.32, found: C 58.34,
24
19
4
4
H 3.88, N 11.33.
.1.19. (±)-1-(4-chlorophenyl)-2-phenylethyl-4-(3,4-dichlorophenyl)piperazine-1-
carboxylate (8):
-(3,4-Dichlorophenyl)piperazine (173 mg, 0.75 mmol) in CH Cl (1 mL) was added
4
1
2
2
dropwise to a solution of triphosgene (83mg, 0.28 mmol) e pyridine (130 µL, 1.61 mmol) in
CH Cl (2 mL). After 3 h at RT, the mixture was diluted with CH Cl (5 mL), then the
2
2
2
2
corresponding solution was washed with H O (2x5mL). The organic layer dried over
2
Na SO and evaporated under vacuum to give a yellow oil used without further purification,
2
4
dissolved in CH Cl (5 mL) and added to a solution of 1-(4-chlorophenyl)-2-phenylethanol
2
2
(53) (120 mg, 0.5 mmol), TEA (83 µL, 0.6 mmol) and DMAP (20 % mmol). The reaction
mixture was stirred overnight at RT. The corresponding orange suspension was diluted
with CH Cl (5 mL) and washed with saturated aqueous NaHCO . The organic layer, dried
2
2
3
under Na SO , was evaporated under reduced pressure. The crude product was purified
2
4
by silica gel column chromatography (CH Cl / MeOH 95:5) to give 8 as a waxy solid (35
2
2

ACCEPTED MANUSCRIPT
-1
1
mg, 15%); IR (neat, ʋ cm ): 1698 (C=O); H NMR (400 MHz, CD Cl) δ (ppm): 3.07-3.25
3
(m, 6H), 3.53-3.71 (m, 4H), 5.92 (m, 1H), 6.96 (dd, J = 8.9, 2.8 Hz, 1H), 7.12 (d, J = 2.8
13
Hz, 1H), 7.19-7.30 (m, 5H), 7.38-7.41 (m, 5H); C NMR (100 MHz, Acetone-d ) δ (ppm):
6
4
1
2.2, 42.9, 47.7, 76.2, 115.5, 116.7, 120.7, 125.9, 127.6, 127.7, 129.1, 130.0, 131.6,
32.3, 136.9, 139.7, 150.6, 153.3, 165.2; MS (ESI): m/z calcd for C H Cl N O 489.09
25
23
3
2
2
+
[
M+H] , found: 490.60; Anal. calcd for C H Cl N O : C 61.30, H 4.73, N 5.72, found: C
25 23 3 2 2
6
1.34, H 4.72, N 5.78.
.1.20. (±)-1-(4-chlorophenyl)-2-(1H-pyrrol-1-yl)ethyl-4-(3,4-
4
dichlorophenyl)piperazine-1-carboxylate (9):
±)-1-(4-Chlorophenyl)-2-(1H-pyrrol-1-yl)ethanol (54) (60 mg, 0.27 mmol) was used to
(
prepare 9 according to the procedure described for 8. The residue was purified by silica
gel column chromatography using CH Cl / MeOH / n-hexane 95:5:5 as eluent to obtain a
2
2
-1
1
waxy solid (45 mg, 35 %); IR (neat, ʋ cm ): 1698 (C=O); H NMR (400 MHz, CD OD) δ
3
(ppm): 3.16 (m, 4H), 3.56-3.74 (m, 4H), 4.29 (m, 2H), 5.99 (m, 1H), 6.02 (m, 2H), 6.64 (m,
13
2
H), 6.92 (dd, J = 8.9, 2.9 Hz, 1H), 7.11 (d, J = 2.9 Hz, 1H), 7.28-7.37 (m, 5H); C NMR
(
100 MHz, Acetone-d ) δ (ppm): 43.5, 48.2, 53.8, 75.7, 107.9, 116.0, 117.2, 121.2, 121.3,
6
1
28.1, 128.4, 130.5, 132.2, 133.3, 137.7, 151.1, 153.5; MS (ESI): m/z calcd for
+
C H Cl N O 478.08 [M+H] , found: 479.67; Anal. calcd for C H Cl N O : C 57.70, H
23
22
3
3
2
23 22
3
3
2
4
.63, N 8.78, found: C 57.80, H 4.64, N 8.75.
4
.1.21. (S)-2-(1H-imidazol-1-yl)-1-(4-nitrophenyl)ethanol (50):
A Schlenk tube was charged with CH Cl (6 mL), TEA (225 ꢀL, 1.62 mmol), 2-(1H-
2
2
imidazol-1-yl)-1-(4-nitrophenyl)ethanone (75 mg, 0.32 mmol) and [(R,R)-TsDPEN Ru-
Cymene)Cl (0.1% mol) under N flow and then HCOOH (526 µL, 12.8 mmol) was added
(
2
over a period of an hour via GC-syringe. During the addition, the temperature rose slowly
to 30 °C . The mixture was heated at 40 °C for 26 h. T he solvent volume was reduced

ACCEPTED MANUSCRIPT
approximately to 3 mL, the solid was recovered, washed with CH Cl (2x2 mL) and
2
2
1
crystallized with EtOH to give a yellow solid (45 mg, 60 %, e.e. > 99%). H NMR (400
MHz, DMSO-d ) δ (ppm): 4.11 (m, 2H), 5.01 (m, 1H), 6.07 (s br, 1H, D O exchange), 6.83
6
2
(
s, 1H), 7.10 (s, 1H), 7.48 (s, 1H), 7.61 (d, J = 7.7 Hz, 2H), 8.19 (d, J = 7.7 Hz, 2H).
.1.22. (S)-1-(biphenyl-4-yl)-2-(1H-imidazol-1-yl)ethanol (51):
A Schlenk tube was charged with CH Cl (7.0 mL), TEA (530 µL, 3.8 mmol), 1-(biphenyl-4-
4
2
2
yl)-2-(1H-imidazol-1-yl)ethanone (200 mg, 0.76 mmol) and [(R,R)-TsDPEN Ru-
Cymene)Cl (0.1 % mol) under N flow and then HCOOH(157µL, 3.8 mmol) was added
(
2
over a period of an hour via GC-syringe. During the addition, the temperature rose slowly
to 30 °C . The mixture was heated at 40 °C for 26 h, t o give a suspension. The solid was
recovered and purified by silica gel column chromatography using CHCl / MeOH / n-
3
hexane 90:10:5 as eluent to obtain a white solid subsequently crystallized from EtOH (83
1
mg, 41 %, e.e. > 99% ). H NMR (400 MHz, DMSO-d ) δ (ppm): 4.07 (dd, J = 14.0, 7.9 Hz,
6
1
1
4
H), 4.18 (dd, J = 14.0, 4.0 Hz, 1H), 4.86 (m, 1H), 5.78 (s br, 1H, D O exchange), 6.85 (s,
2
H), 7.16 (s, 1H), 7.36 (t, J = 7.3 Hz, 1H), 7.43-7.48 (m, 4H), 7.55 (s, 1H), 7.63-7.67 (m,
H).
4
.1.23. (S)-2-(1H-imidazol-1-yl)-1-(naphthalen-1-yl)ethanol (52):
(S)-2-(1H-imidazol-1-yl)-1-(naphthalen-1-yl)ethanol (52) was prepared in according to the
procedure described for 50, to give a white solid crystallized from EtOAc (55 %, e.e.>
1
9
1
7
9%). H NMR (400 MHz, DMSO-d ) δ (ppm): 4.14 (dd, J = 13.9, 7.8 Hz, 1H), 4.25 (dd, J =
6
3.9, 4.0 Hz, 1H), 4.99 (m, 1H), 5.87 (s br, 1H, D O exchange), 6.82 (s, 1H), 7.14 (s, 1H),
2
.49-7.54 (m, 4H), 7.84-7.91 (m, 4H).
4
.1.24. (±)-1-(4-Chlorophenyl)-2-phenylethanol (53):