Synthesis and biological evaluation of novel cYY analogues targeting Mycobacterium tuberculosis CYP121A1

Article abstract of DOI:10.1016/j.bmc.2019.02.051

The rise in multidrug resistant (MDR) cases of tuberculosis (TB) has led to the need for the development of TB drugs with different mechanisms of action. The genome sequence of Mycobacterium tuberculosis (Mtb) revealed twenty different genes coding for cy

Full text of DOI:10.1016/j.bmc.2019.02.051

Accepted Manuscript
Synthesis and Biological Evaluation of Novel cYY Analogues targeting Myco-
bacterium tuberculosis CYP121A1
Safaa M. Kishk, Kirsty J. McLean, Sakshi Sood, Mohamed A. Helal, Mohamed
S. Gomaa, Ismail Salama, Samia M. Mostafa, Luiz Pedro S. de Carvalho,
Andrew W. Munro, Claire Simons
PII:
S0968-0896(19)30024-0
https://doi.org/10.1016/j.bmc.2019.02.051
BMC 14787
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
4 January 2019
20 February 2019
25 February 2019
Please cite this article as: Kishk, S.M., McLean, K.J., Sood, S., Helal, M.A., Gomaa, M.S., Salama, I., Mostafa,
S.M., de Carvalho, L.P.S., Munro, A.W., Simons, C., Synthesis and Biological Evaluation of Novel cYY Analogues
targeting Mycobacterium tuberculosis CYP121A1, Bioorganic & Medicinal Chemistry (2019), doi: https://doi.org/
1
0.1016/j.bmc.2019.02.051
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.

Synthesis and Biological Evaluation of Novel cYY Analogues
targeting Mycobacterium tuberculosis CYP121A1
a,b
c
d
b,e
Safaa M. Kishk, Kirsty J. McLean, Sakshi Sood, Mohamed A. Helal, Mohamed
b,f
b
b
d
S. Gomaa, Ismail Salama, Samia M. Mostafa, Luiz Pedro S. de Carvalho, Andrew
a*
c
W. Munro, and Claire Simons
a
School of Pharmacy & Pharmaceutical Sciences, Cardiff University, King Edward VII Avenue,
Cardiff, CF10 3NB, U.K.
b
Medicinal Chemistry Department, Faculty of Pharmacy, Suez Canal University, Ismailia,
Egypt
c
Manchester Institute of Biotechnology, School of Chemistry, The University of
Manchester, 131 Princess Street, Manchester M1 7DN, UK
d
Mycobacterial Metabolism and Antibiotic Research Laboratory, The Francis Crick Institute, 1
Midland Road, London NW1 1AT, UK
e
Biomedical Sciences Program, University of Science and Technology, Zewail City of Science and
Technology, Giza 12588, Egypt
f
Department of Chemistry, College of Clinical Pharmacy, Imam Abdulrahman Bin Faisal University,
Dammam, Kingdom of Saudi Arabia
*
Address for correspondence: Claire Simons, School of Pharmacy and
Pharmaceutical Sciences, Cardiff University, Redwood Building, King Edward VII
Avenue, Cardiff CF10 3NB, Wales, UK. Tel: 02920-876307. E-mail:
simonsc@cardiff.ac.uk
1

Abstract
The rise in multidrug resistant (MDR) cases of tuberculosis (TB) has led to the need
for the development of TB drugs with different mechanisms of action. The genome
sequence of Mycobacterium tuberculosis (Mtb) revealed twenty different genes
coding for cytochrome P450s. CYP121A1 catalyzes a C-C crosslinking reaction of
dicyclotyrosine (cYY) producing mycocyclosin and current research suggests that
either mycocyclosin is essential or the overproduction of cYY is toxic to Mtb. A
series of 1,4-dibenzyl-2-imidazol-1-yl-methylpiperazine derivatives were designed
and synthesised as cYY mimics. The derivatives substituted in the 4-position of the
phenyl rings with halides or alkyl group showed promising antimycobacterial activity
(MIC 6.25 g/mL), with the more lipophilic branched alkyl derivatives displaying
i
t
optimal binding affinity with CYP121A1 ( Pr K = 1.6 M; Bu K = 1.2 M).
D
D
Computational studies revealed two possible binding modes within the CYP121A1
active site both of which would effectively block cYY from binding.
Keywords
CYP121A1;
Mycobacterium
tuberculosis;
1,4-dibenzyl-2-imidazol-1-yl-
methylpiperazine derivatives; binding affinity assays; molecular modelling
2

1
. Introduction
Tuberculosis (TB) is the ninth leading cause of death worldwide, ranking above
Human Immunodeficiency Virus (HIV/AIDS). In 2017, the estimated TB deaths were
1
.3 million among HIV-negative people and 300,000 among HIV-positive people, and
1
the estimated incident TB cases were 10.0 million worldwide. The standard
therapeutic regimens involve the combination of four first-line drugs, often isoniazid,
rifampicin, pyrazinamide and either streptomycin or ethambutol, depending upon
2
whether there is a latent or active type of infection. Although the current medications
for treating drug-sensitive TB are effective when there is optimum adherence of
patients, a common problem in real-life conditions appears with TB-infected patients
3
,4
who generally do not adhere strictly to dosages.
The duration for treating drug-sensitive TB is usually 6 months with the four
most effective agents from the first-line oral drugs administered in a single
prescription for the first two months of treatment, and two of the four taken for a
subsequent four months in the continuation phase, resulting in patient adherence
5
issues. When administered in sub-optimum conditions, chronic cases of infectious
drug-resistant TB appear. Long treatment durations are usually required because
Mycobacterium tuberculosis (Mtb), the causative pathogen, can develop a dormancy
phenotype under nutrient depletion and anaerobiosis conditions which is tolerant to
6
,7
several anti-TB drugs. The development of novel anti-TB drugs has become a
priority in view of the increasing global incidence of strains resistant to at least
rifampicin and isoniazid (MDR-TB), or to rifampicin, isoniazid and one of the
injectable second-line anti-TB drugs as well as to any of the fluoroquinolone drug
1
series (XDR-TB), in addition to the interactions between anti-TB drugs and
antiretroviral medications, and the need to treat latent TB-infected patients before the
3

8
bacteria transform into their active form.
Studies following the unraveling of the genome sequence of the virulent Mtb H37Rv
9
strain revealed numerous genes with unknown functions. Among these were twenty
1
0
different cytochrome P450 (CYP) genes coding for P450s. This large number of
CYP genes was uncommon for a bacterium and, as a result, studies over the past
twenty years have focused on characterizing these Mtb P450s, several of which play
crucial roles in the survival of Mtb. Subsequent genome studies with M. smegmatis
revealed several P450s, confirming the multiple important functions for the P450s in
1
1
the actinobacteria.
Figure 1. Cyclodityrosine synthase and CYP121A1 reactions to produce the cyclic dipeptide,
cyclo-L-Tyr-L-Tyr (cYY), and the novel secondary metabolite mycocyclosin.
CYP121A1 (mycocyclosin synthase) was shown to be essential for bacterial growth
1
2-14
by in vitro gene knockout studies.
Also, CYP121A1, along with CYP128A1 and
CYP141A1, are the only Mtb P450s that appear to be conserved within members of
the Mtb complex because their homologues do not appear in the genomes of the other
4

members of the actinobacteria family. The first evidence of CYP121A1 function in
Mtb was derived from its gene position, which is located in an operon harboring two
enzymes involved in the formation of cyclo-di-L-tyrosine (cYY). The first enzyme is
a cyclodipeptide synthase (encoded by Rv2275), which uses aminoacyl-tRNA
Tyr
synthetases (L-tyrosyl-tRNA ) to catalyse the ATP-independent formation of cYY
1
5,16
(Figure 1).
Then, CYP121A1 (encoded by Rv2276) catalyzes a C-C crosslinking
reaction between the respective carbons in the ortho position of the phenolic hydroxyl
of cYY, producing mycocyclosin (Figure 1). Rv2276 was found to be an essential Mtb
gene, and it was suggested that either mycocyclosin was essential or the
14,17
overproduction of cYY is toxic.
Assays conducted on CYP121A1 utilising a
ferredoxin and ferredoxin reductase system demonstrated that CYP121A1 catalyzes
multiple turnovers of cYY in a complicated multistep process to form mycocyclosin
1
7
as the single major product in the presence of NADPH.
A
B
Figure 2. (A) Flexible alignment of cYY (yellow) and lead compound (cyan) showing
overlap (B) Binding affinity and MIC data for lead compound.
In the development of CYP121A1 inhibitors, we have described several series of
1
8,19
compounds with chemical scaffolds that mimic the natural substrate cYY.
From
these series 1,4-dibenzyl-2-imidazol-1-yl-methylpiperazine derivatives were
1
8
identified as potential leads for further development. Flexible alignment of these
5

derivatives with cYY showed good overlap (Figure 2) suggesting they would occupy
a similar area as cYY within the CYP121A1 active site, while the addition of an azole
haem-binding group (imidazole) may allow direct interaction between the compound
and the haem iron.
The lead compound showed weak binding affinity, as determined by UV-vis optical
titration, and modest inhibitory activity against Mtb (Figure 2). Therefore, the aim of
this study was to develop this series of compounds, through varying the substituent on
the benzene rings, and to explore the structure-activity relationship to improve both
binding affinity with CYP121A1 and MIC against Mtb.
Scheme 1. Reagents and conditions: (i) EtOH, r.t., o/n (ii) NaBH , MeOH, r.t., 4 h (iii) Et N,
4
3
2
,3-dibromopropionic acid ethyl ester, toluene, 80 °C, o/n (iv) LiAlH , THF, r.t., o/n (v)
4
SOCl , CH Cl , r.t., 48 h (vi) (a) Imidazole, K CO , CH CN, 45 °C, 1 h (b) chloride 7, 70 °C,
2
2
2
2
3
3
48 h.
6

2
2
. Results and Discussion
.1 Chemistry
The imidazole products (8) were obtained from a six-step synthetic route (Scheme 1).
Substituted benzaldehyde derivatives (1) were reacted with ethylenediamine (2) in
ethanol for 6 h at room temperature to give the di-Schiff base (di-imine) (3), which
precipitated from the reaction as pure crystalline solids in high yields of 81-99%. The
imine bond was reduced by NaBH followed by careful aqueous work-up while
4
cooling in an ice bath to give the diamines (4) in good yields (75-97%) with the
exception of the pyridinyl derivatives (4i and 4j), which were obtained in yields of 29
and 31% respectively owing to their water solubility. The synthesised diamines (4) on
2
0
reaction with ethyl 2,3-dibromopropionate resulted in the formation of ethyl 1,4-
bis(substituted) piperazine-2-carboxylate (5) with yields in the range of 65-85%.
Piperazine, as a six-membered cyclic system, can exist in both chair and boat
conformations. The lowest energy and most stable conformation for the synthesised
21
22
esters, using the conformational import algorithm in MOE followed by
2
3
optimisation using the PM3 model Hamiltonian is shown in Figure 3.
From the proposed 3D structure, the axial conformation for the ester moiety at C-
2
4
1
2
was preferred, this conformation of the ester intermediates was analysed by H
NMR and 2D COSY experiments. The H-2 proton had vicinal three-bond couplings
across single bonds with both axial and equatorial H-3 protons, and appeared as
1
doublet of doublets (dd) in the H NMR spectra in derivatives 5a, 5c, 5e, 5h and 5i
with coupling constants in the range of 3.4-3.6 Hz for the vicinal equatorial-equatorial
couplings, and larger values in the range of 5.5-6.0 Hz for the vicinal equatorial-axial
couplings. In some derivatives, the H-2 proton couplings appeared as a multiplet (m)
as seen in 5b and 5d, or interfered with other protons as seen in 5f, 5g and 5j.
7

Geminal spin-spin couplings were conserved between the H-7 protons and the H-8
1
protons, and in the H NMR spectra, each proton appeared as a doublet (d) with a
geminal coupling constant in the range of 13.0-14.0 Hz.
Figure 3. Ethyl 1,4-bis(4-fluorobenzyl)piperazine-2-carboxylate (5a) chair conformation with
vicinal and geminal couplings of H-2, H-7 and H-8 protons.
The synthesised esters (5) were reduced to the corresponding primary alcohols
(
6) by LiAlH after overnight stirring at room temperature with yields in the range of
4
5
6-91% obtained. Conversion of the alcohols (6) to the alkyl chlorides proceeded
through a nucleophilic substitution S 2 mechanism using thionyl chloride (10
N
equivalents) at room temperature for 48 h. A trial to accelerate the chlorination
reaction rate was done by refluxing the synthesised alcohol with thionyl chloride (2
equivalents) at 80 °C, but complex mixtures were obtained, and the products could
not be separated or identified upon refluxing, especially for the halide derivatives (7a-
7
d). The yields of the chloride products (7) from reaction at room temperature for 48
h were in the range of 50-88%. The imidazolate ion, which acts as a nucleophile, was
8

formed in situ by reacting imidazole with a base, for 1 h at 45 °C in acetonitrile,
followed by a S 2 reaction between the imidazolate and the synthesised alkyl chloride
N
(7). When sodium hydride (60% dispersion in mineral oil) was used as a base
18
according to the reported method, a complex mixture was obtained, and the product
could only be separated in a very low yield (< 10%). Potassium carbonate was
selected as the optimum base, and the reaction was complete after refluxing at 70 °C
for 48 h with yields in the range of 51-69%.
Figure 4. 2-((1H-Imidazol-1-yl)methyl)-
1,4-bis(4-(tert-butyl)benzyl)piperazine (8h)
chair conformation after energy
minimisations to explain vicinal and
geminal couplings between H-9, H-2, H-7,
H-8 protons.
A characteristic feature for the final imidazoles was the geminal and vicinal
proton-proton coupling,. The doublet of doublets (dd) at 4.28-4.29 ppm was assigned
to the geminal (J = 13.8-14.0 Hz) and vicinal (J = 4.3-4.4 Hz) couplings of the H-9a
proton with H-9 and H-2 protons, respectively. The doublet of doublets at 4.20-4.21
b
ppm was assigned to the geminal (J = 13.8-14.0 Hz) and vicinal (J = 7.6-8.0 Hz)
coupling of the H-9 proton with H-9 and H-2 protons, respectively. The vicinal spin-
b
a
spin coupling constants illustrated the difference in the chemical shifts between H-9a
and H-9 protons. Vicinal spin-spin coupling constants were generally in the range of
b
6
-8 Hz and, as the proton was closer to the imidazole group (electron-deficient
group), smaller vicinal J values were noticed. From the proposed 3D-structure after
9

energy minimisations (Figure 4), the H-9 proton appeared to be closer to the
a
imidazole group than the H-9 proton. Consequently, the H-9 proton was less
b
a
shielded and its vicinal coupling constant was less than 6.0 Hz while the H-9 proton
b
was more shielded and its vicinal coupling constant was within the range of 6-8 Hz.
2
.2 CYP121A1 Ligand Binding Affinity
The CYP121A1 binding affinity (K ) of the various compounds was determined by
D
2
5
UV-vis optical titration. Binding titrations were carried out for the synthesised
imidazoles (8), which showed a type II (inhibitor-like) red shift in the haem Soret
peak to a longer wavelength (Figure 5, for example 8h) indicating that, in the solution
state, most of these compounds coordinate either (i) directly to the CYP121A1 haem
iron or (ii) indirectly to the haem iron through an interstitial water molecule.
Figure 5. UV-Vis optical binding titration for compound 8h binding to CYP121A1. The
left hand panel shows data from a compound 8h titration with CYP121A1 (~4.7 M) with the
ligand-free spectrum as a thick black line, spectra following progressive additions of 8h as
thin solid lines, and the final near-saturated protein spectrum shown as a thick red line. The
inset shows overlaid difference spectra generated by the subtraction of the starting spectrum
from each consecutive ligand-bound spectrum collected in the titration. The right hand panel
shows a plot of compound 8h-induced absorbance change – calculated as the difference
between the peak and trough in the difference spectra in the left hand panel, using the same
1
0

wavelength pair (429 and 392 nm, respectively) throughout. Data were fitted using the Hill
equation to give a 8h K value of 1.2 ± 0.1 M.
D
Halogenated derivatives (8b-8d) displayed moderate binding affinity with the
best K value observed for the 3,5-diCl compound 8d. The 4-F compound 8a was
D
noted to undergo photo- and thermal-decomposition in this assay so the data obtained
was not reliable.
Table 1. Binding affinity (K ) values for compounds against M. tuberculosis H37Rv
D
a
Soret peak shift
nm)
Compound
R
K (M)
D
(
Lead
H
1968 ± 137 416.5 to 417.5
26.8 ± 7.9 416.5 to 419
11.9 ± 2.1 416.5 to 421
8
8
8
8
8
8
8
8
8
b
c
d
e
f
4-Cl
4-Br
3,5-diCl
8.1 ± 0.4
416.5 to 419
4-OCH
4-CH CH
4-CH(CH
4-C(CH
42.0 ± 14.3 416.5 to 419.5
3
3.7 ± 0.1
1.6 ± 0.1
1.2 ± 0.1
416.5 to 423.5
416.5 to 422.5
416.5 to 419.5
2
3
g
h
i
)
3 2
)
3 3
Pyridin-4-yl 39.2 ± 7.9 416.5 to 419.5
j
Pyridin-3-yl 489 ± 27
8.6 ± 0.2
416.5 to 421.5
Fluconazole
Clotrimazole
0.07 ± 0.01
cYY
5.8 ± 0.2
a
compound 8a (R = 4-F) noted to undergo photo- and thermal-decomposition during this assay.
1
1

The lipophilic derivatives had the best binding of the imidazole nitrogen to the
haem iron with K values of 3.7 ± 0.1, 1.6 ± 0.1 and 1.2 ± 0.1 M for 8f, 8g and 8h,
D
respectively, while the natural substrate cYY had a K value of 5.8 ± 0.2 M. The
D
moderately hydrophilic 4-OMe (8e) and pyridines (8i and 8j) had weak binding
affinity to CYP121A1 (Table 1). Compounds 8f and 8g induce the most extensive
Soret absorbance shifts (416.5 to 422.5/423.5 nm), suggesting that the CYP121A1
haem iron is predominantly coordinated by a direct imidazole nitrogen bond in these
cases. Compounds 8b-8e and 8h-8j show less extensive Soret red shifts on binding
the CYP121A1 haem, suggesting that ligation modes in these cases may involve
imidazole nitrogen interactions with haem iron mediated through an interstitial water
ligand.
2
.3 MIC determination against Mycobacterium tuberculosis
The derivatives were screened against M. tuberculosis H37Rv by the REMA
2
6
(
Resazurin Microtiter Assay) method. With the exception of the pyridyl compounds
i and 8j, which were inactive (MIC >100 g/mL), all the compounds were more
active than the lead parent compound (MIC 50 g/mL). In the halogenated series, the
-monosubstituted benzene ring derivatives 8a-8c showed good inhibitory activity
MIC 6.25 g/mL) compared with the 3,5-dichloro substituted derivative 8d (MIC 25
g/mL). The alkyl substituted series all showed promising inhibitory activity (MIC
.25 g/mL), while the more hydrophilic 4-methoxy derivative (8e) was less effective
MIC 25 g/mL).
8
4
(

6
(
1
2

Table 2. MIC values for compounds against M. tuberculosis H37Rv
a
MIC
cLogP
Compound
R
Mol.Wt.
(
g/mL)
Lead
H
346.469
382.450
415.359
504.261
484.249
406.521
402.575
430.628
458.681
50
6.25
6.25
6.25
25
3.56
3.87
4.67
5.21
5.79
3.3
8
8
8
8
8
8
8
8
8
8
a
b
c
d
e
f
4-F
4-Cl
4-Br
3,5-diCl
4-OCH
4-CH CH
4-CH(CH
4-C(CH
25
3
6.25
6.25
6.25
>100
>100
>100
20
5.36
6.02
6.97
0.88
0.88
0.87
5.97
1.26
2
3
g
h
i
)
3 2
)
3 3
Pyridin-4-yl 348.445
Pyridin-3-yl 348.445
306.271
j
Fluconazole
Clotrimazole
344.837
cYY
326.347
-
a
cLogP was determined using Crippen’s fragmentation
1
3

2
.4 Molecular Modelling
Molecular modelling of the compounds was performed using Molecular Operating
Environment (MOE) sofware and the crystal structure of cYY co-crystallised with
1
7
CYP121A1 (PDB 3G5H). Most of the compounds displayed two binding modes
when docked within the CYP121A1 active type, both of which show indirect binding
with the haem via an interstitial water molecule.
A
B
Figure 6. Docking of representative piperazines (cyan) (A) 8b in binding mode 1 and (B) 8g
in binding mode 2 in the CYP121A1 active site compared with cYY (magenta). [Haem
(
orange), water molecules (red spheres)]
The first docking mode (mode 1) mimicked the cYY conformation as illustrated by
the chloro derivative 8b (Figure 6A). The most consistent binding interactions were
observed between the imidazole ring and Arg386 via a water molecule and the
protonated piperazine NH+ and Met86 effectively bridging the active site above the
haem (Table 2). In the second binding mode (mode 2) a more open/flexible
conformation was observed as illustrated by the isopropyl derivative 8g (Figure 6B).
This increased flexibility is possible owing to the piperazine ring which can exist in
both chair and boat conformations. In mode 2, binding interactions were observed
1
4

between the imidazole and a number of different amino acids (e.g. Asn74, Gly283,
Gln385 and Arg386) via a water molecule with additional binding between the
protonated piperazine NH+ and Met86. The pyridine derivatives 8i and 8j formed
additional H-bonding interactions through the two pyridine rings. Multiple
hydrophobic interactions were observed between the piperazines (8) for both binding
modes 1 and 2 including residues Met62, Thr77, Val78, Val82, Val83, Asn85,
Leu164, Ala167, Phe168, Thr229, Ala233, Gly232, Phe280 and Leu284. For the
unsubstituted lead compound and the fluoro (8a), chloro (8b) and pyridine (8i and 8j)
derivatives, binding mode 1 was energetically prefered (Table 2).
Table 2. Computational binding energies of piperazines 8 and CYP121A1
Mode 2
S-value (kcal/mol)
Compound
R
Mode 1
S-value (kcal/mol)
Lead
H
-4.69
-6.29
-2.97
-4.61
8
a
4-F
8
b
4-Cl
4-Br
-4.53
-4.23
-2.78
-3.64
-4.79
-4.29
-4.21
0.04
8
c
-
-
8
d
3,5-diCl
8
e
4-OCH
4-CH CH
4-CH(CH
4-C(CH
3
-4.73
-3.34
-3.18
-
8
f
2
3
8
g
3
)
2
8
h
)
3 3
8
i
Pyridin-4-yl
Pyridin-3-yl
-5.00
-5.32/-6.58
-2.35
-4.26
8
j
S-value = lowest binding energy
The methoxy derivative (8e) showed comparable S-values for binding modes 1 and 2,
while the ethyl (8f) and isopropyl (8g) derivatives had improved S-values for binding
1
5

mode 2 (Table 2). For the more bulky bromo (8c), dichloro (8d) and tert-butyl (8h)
derivatives only binding mode 2 was observed with the tert-butyl derivative the least
energetically favourable. Two possible mode 1 orientations were observed for the 3-
pyridyl derivative (8j) with indirect interaction with the haem via a water molecule
either through imidiazole (S-value -5.32) or through one of the pyridine rings (S-value
-6.58.).
To investigate the binding modes further molecular dynamics simulations were run
for 50ns for the CYP121A1 and isopropyl derivative (8g) and tert-butyl derivative
2
7,28
(
8h) complexes using the Desmond programme of Maestro.
8g)-CYP121A1 complex, the RMSD changed from 1.2 Å at zero time to 2.1 Å at ~
0 ns; then the protein was equilibrated with no evident RMSD fluctuations observed
For the isopropyl
(
1
and the ligand aligned. For the tert-butyl (8h)-CYP121A1 complex, equilibrium and
ligand alignment required a longer run time with the RMSD changing from 0.8 Å at
zero time to 2.0 Å at ~ 30 ns. (Figure 7A). Protein interactions with the ligand were
checked throughout the simulation and interactions ordered by their type; specifically
hydrogen bonds, hydrophobic, ionic, and water bridges (Figure 7B). The isopropyl
derivative (8g) formed more interactions over the course of the 50ns simulation
compared with the tert-butyl derivative (8h). As illustrated in Figure 7C, water bridge
interactions between the imidazole and Asp282 and Gln385, and the protonated
piperazine and Val83 were significant for 8g with additional hydrophobic interactions
between the 4-iPr-phenyl rings and Phe168 and Pro285. For 8h the major contributor
was a water bridge interaction between the imidazole ring and Arg386, with
additional binding interactions between one of the tBu-phenyl rings and His343 (-
stacking) and Phe168 (hydrophobic) (Figure 7C).
1
6

iPr (8g)
tBu (8h)
A
B
C
Figure 7. (A) Protein (blue) and ligand (red) root mean square of the CYP121A1-8g and 8h complexes
B) Illustration of protein interactions with the ligand. The red column means ionic bond, green column
(
a hydrogen bond, blue column water bridges, and a purple column a hydrophobic bond. The stacked
bar charts are normalised over the course of the trajectory: for example, a value of 0.5 suggests that for
5
0% of the simulation time, the specific interaction is maintained (C) A schematic of detailed ligand
atom interactions with the protein residues. Interactions that occur more than 30.0% of the simulation
time in the selected trajectory (0.00 through 50.05 nsec), are shown (note: it is possible to have
1
7

interactions with >100% as some residues may have multiple interactions of a single type with the
same ligand atom e.g. the ARG side chain has four H-bond donors that can all hydrogen-bond to a
single H-bond acceptor).
Sterically, the transition from an isopropyl (8g) to a tert-butyl (8h) would appear
sufficient to hinder closer binding interaction. As shown in Figure 8, the isopropyl
derivative (8g) is able to take a position within the active site that is comparable
(mode 1 binding) with the natural substrate cYY although one of the iPr-phenyl rings
deviates from the cYY position through hydrophobic interaction with Pro285.
However the tert-butyl derivative (8h) stabilises in the more open/flexible mode 2
conformation, with one of the tBu-phenyl rings deviating substantially, most likely
owing to the reduction in conformational space within the binding pocket composed
of Met62, Val82, Val83, Asn85 and Met86 (Figure 6).
Figure 8. Positioning of the isopropyl derivative (8g, cyan) and tert-butyl derivative (8h,
magenta) compared with cYY (yellow) after molecular dynamics simulation.
Both indirect binding modes 1 and 2 are possible and both would be effective in
blocking the natural substrate cYY from binding at the active site. Crystallographic
studies are needed to explore binding modes further.
1
8

3
. Conclusions
With the exception of the pyridyl derivatives (8i and 8j), all the new piperazine
derivatives showed improved binding affinity to CYP121A1 and inhibitory activity
against Mtb compared with the lead compound. Preliminary SAR for the described
1
,4-dibenzyl-2-imidazol-1-yl-methylpiperazine derivatives would suggest that 4-
substitution of the benzyl rings with either halides (8a-8c) or alkyl substitutents (8f-
h) is beneficial for antimycobacterial activity with improved activity (MIC = 6.25
g/mL) compared with clotrimazole (MIC = 20 g/mL) (Table 1). Only one
8

compound did not have substitution in the 4-position, namely the 3,5-dichloro
derivative (8d), which had a reduced inhibitory activity compared with clotrimazole,
however mono-substitutions in the 2- and 3-positions have not been investigated thus
far and will be the focus of further studies to determine effect on binding and
inhibitory activity. Two binding modes were observed from molecular modeling
studies with the larger substituents (Br, 3,5-diCl, alkyl) showing a preference for the
more open binding mode 2, while smaller substituents (F, Cl) showed a preference for
the more compact binding mode 1 and more closely resemble the conformation of
cYY (Table 2, Figure 6). The alkyl derivatives (8f-8h) were optimal with respect to
binding affinity (K = 3.7, 1.6 and 1.2 M, respectively), with the ethyl (8f) and
D
isopropyl derivatives (8g) inducing the most extensive Soret absorbance shifts (416.5
to 422.5/423.5 nm) (Table 1). However, binding affinity did not clearly correlate with
MIC, for example, the 4-chloro derivative (8b) and the tert-butyl derivative (8g)
showed similar MIC values (6.25 g/mL) but had completely different binding
affinity to CYP121A1 (26.8 ± 7.9 compared with 1.6 ± 0.1M), suggesting different
ligation modes with either indirect or direct haem coordination.
1
9

These piperazine derivatives are useful compounds for further development with
future studies focusing on replacement of one of the phenyl rings to optimise
positioning within the binding pocket composed of Met62, Val82, Val83, Asn85 and
Met86 identified from the computational studies as restricting optimal fit within the
CYP121A1 active site.
4
. Experimental Section
4
.1 General Experimental
All chemicals, reagents and solvents were purchased from Sigma-Aldrich, Fisher
Scientific, Alfa-Aesar, Fluka and Acros Chemicals. Whenever required, solvents were
dried prior to use as described by the handbook Purification of Laboratory Chemicals
and stored over 4 Å molecular sieves under nitrogen. Flash column chromatography
was performed with silica gel (230-400 mesh) (Merck) and TLC was performed on
pre-coated silica gel plates (Merck Kiesel gel 60F254, BDH). Melting points were
determined on an electrothermal instrument (Gallenkamp), and are uncorrected.
Compounds were visualised by irradiation with UV light at 254 nm and 365 nm or by
using KMnO stain or vanillin stain followed by heating. NMR spectra were recorded
4
on a Bruker AVANCE DPX500 spectrometer operating at 500,125 and 470 MHz for
1
13
19
H, C and F NMR, respectively, and auto calibrated to the deuterated solvent
reference peak. The assignments were made using one-dimensional (1D) and two-
dimensional (2D) HSQC and COSY spectra. Chemical shifts are given in δ relative to
tetramethylsilane (TMS); the coupling constants (J) are given in Hertz. TMS was used
1
as an internal standard (δ = 0 ppm) for H NMR and CDCl served as an internal
3
1
3
standard (δ = 77.0 ppm) for C NMR. Multiplicity is denoted as s (singlet), br s
(broad singlet), d (doublet), dd (doublet of doublet), t (triplet), q (quartet), m
2
0

(multiplet) or combinations thereof. High resolution mass spectra (HRMS) were
determined at the EPSRC National Mass Spectrometry Facility at Swansea University
and Medac Ltd., Brunel Science Centre, Surrey using ESI (Electrospray Ionisation).
The following compounds were prepared as previously described: diimines 3a-3c, 3e
2
9-34
29,30,34,35
and 3g-3j;
diamines 4a-4c, 4e-4f and 4i;
ethyl 1,4-bis(4-
methoxybenzyl)piperazine-2-carboxylate (5e), (1,4-bis(4-methoxybenzyl)piperazin-2-
18
yl)methanol (6e) and 2-(chloromethyl)-1,4-bis(4-methoxybenzyl)piperazine (7e).
All compounds were more than 95% pure.
4
.2 Chemistry
.2.1. General method for the preparation of diimines (3). To a stirred solution of
4
benzaldehyde (1) (2 m.eq.) in ethanol (1mL/mmol of benzaldehyde), was added
ethylenediamine (2) (1 m.eq.) dropwise. The reaction mixture was stirred for 6 h at
room temperature, and the precipitated white coloured crystals were collected by
filtration, washed with diethyl ether and dried to give the desired diimine.
4
.2.1.1. N,N´-(Ethane-1,2-diyl)bis(1-(3,5-dichlorophenyl)methanimine) (3d). Prepared
from 3,5-dichlorobenzaldehyde (1d) (5 g, 28.6 mmol). Product obtained as a white
o
solid, yield 5.10 g (96%). M.p. 130-132 C. TLC (3:1 petroleum ether/EtOAc), Rf =
1
0
.84. H NMR (CDCl ): δ 8.34 (s, 2H, 2 x CH=N), 7.74 (d, J = 1.9 Hz, 4H, Ar), 7.71
3
1
3
(
t, J = 1.9 Hz, 2H, para-Ar), 3.91 (s, 4H, 2 x CH ). C NMR (CDCl ): δ 161.3 (2 x
2 3
CH=N), 135.7 (2 x C, Ar), 135.3 (4 x C, Ar), 129.9 (2 x CH, para-Ar), 129.3 (4 x CH,
+
Ar), 61.2 (2 x CH ). [ESI-HRMS] calculated for C H Cl N : 372.9833 [M+H] .
2
16 13
4
2
+
Found: 372.9840 [M+H] .
4
.2.1.2. N,N´-(Ethane-1,2-diyl)bis(1-(4-ethylphenyl)methanimine) (3f). Prepared from
-ethylbenzaldehyde (1f) (10 mL, 72.96 mmol). Product obtained as a white solid,
4
o
1
yield 9.89 g (93%). M.p. 68-70 C. TLC (4:1 petroleum ether/EtOAc), Rf = 0.84. H
2
1

NMR (CDCl ): δ 8.29 (s, 2H, 2 x CH=N), 7.62 (d, J = 8.1 Hz, 4H, Ar), 7.26 (d, J =
3
8
.0 Hz, 4H, Ar), 3.84 (s, 4H, 2 x CH ), 2.62 (q, J = 7.6 Hz, 4H, 2 x CH CH ), 1.17 (t,
2 2 3
1
3
J = 7.6 Hz, 6H, 2 x CH CH ). C NMR (DMSO-d ): δ 162.2 (2 x CH=N), 147.1 (2 x
2
3
6
C, Ar), 134.2 (2 x C, Ar), 128.5 (4 x CH, Ar), 128.0 (4 x CH, Ar), 61.5 (2 x CH2),
8.5 (2 x CH ), 15.8 (2 x CH ). [ESI-HRMS] calculated for C H N : 293.2018
2
2
3
20 25
2
+
+
[
M+H] . Found: 293.2033 [M+H] .
4
.2.2. General method for the preparation of diamines (4). To an ice-cooled stirred
suspension of diimine (3) (1 m.eq.) in methanol (10 mL/mmol) was added sodium
borohydride (4.6 m.eq.) in portions. The reaction mixture was stirred for 4 h at room
temperature. After the reaction was complete, methanol was evaporated under
vacuum and cold water (10 mL/mmol) was added to the remaining residue until
cessation of effervescence. The mixture was extracted with EtOAc (10 mL/mmol),
and the organic layer was washed with water (2 x 10 mL/mmol), brine (2 x 10
mL/mmol), dried (MgSO ) and evaporated under vacuum.
4
1
2
4
.2.2.1. N ,N -Bis(3,5-dichlorobenzyl)ethane-1,2-diamine (4d). Prepared from N,N'-
(ethane-1,2-diyl)bis(1-(3,5-dichlorophenyl)methanimine) (3d) (5.0 g, 13.36 mmol).
o
Product obtained as a white solid, yield 4.41 g (87%). M.p. 106-108 C. TLC (2:1
1
petroleum ether/EtOAc), Rf = 0.12. H NMR (CDCl ): δ 7.28 (m, 2H, para-Ar), 7.26
3
13
(
m, 4H, Ar), 3.77 (s, 4H, 2 x CH ), 2.76 (s, 4H, 2 x CH ), 1.79 (br s, 2H, 2 x NH). C
2 2
NMR (CDCl ): δ 143.6 (2 x C, Ar), 135.0 (4 x C, Ar), 127.2 (2 x CH, Ar), 126.5 (4 x
3
CH, Ar), 52.8 (2 x CH ), 48.4 (2 x CH ). [ESI-HRMS] calculated for C H Cl N :
2
2
16 17
4
2
+
+
3
77.0146 [M+H] . Found: 377.0159 [M+H] .
1
2
4
.2.2.2. N ,N -Bis(4-isopropylbenzyl)ethane-1,2-diamine (4g). Prepared from N,N'-
(ethane-1,2-diyl)bis(1-(4-isopropylphenyl)methanimine) (3g) (10.4 g, 32.76 mmol).
o
Product obtained as a white solid, yield 9.89 g (93%). M.p. 58-60 C. TLC (2:1
2
2

1
petroleum ether/EtOAc), Rf = 0.51. H NMR (CDCl ): δ 7.22 (d, J = 8.2 Hz, 4H, Ar),
3
7
.16 (d, J = 7.8 Hz, 4H, Ar), 3.62 (s, 4H, 2 x CH ), 2.85 (m, 2H, 2 x CH(CH ) ), 2.57
2 3 2
13
(
s, 4H, 2 x CH ), 1.49 (br s, 2H, 2 x NH), 1.18 (d, J = 6.8 Hz, 12H, 4 x CH ). C
2
3
NMR (DMSO-d ): δ 146.9 (2 x C, Ar), 138.9 (2 x C, Ar), 128.3 (4 x CH, Ar), 126.4
6
(4 x CH, Ar), 53.3 (2 x CH ), 48.9 (2 x CH ), 33.6 (2 x CH(CH ) ), 24.43 (4 x CH ).
2 2 3 2 3
+
+
[
ESI-HRMS] calculated for C H N : 325.2638 [M+H] . Found: 325.2643 [M+H] .
2
2
33
2
1
2
4
.2.2.3. N ,N -Bis(4-(tert-butyl)benzyl)ethane-1,2-diamine (4h). Prepared from N,N'-
(ethane-1,2-diyl)bis(1-(4-tert-butylylphenyl)methanimine) (3h) (7.0 g, 20.08 mmol).
Product obtained as a colourless oil, yield 6.5 g (92%). TLC (2:1 petroleum
1
ether/EtOAc), Rf = 0.53. H NMR (DMSO-d ): δ 7.31 (d, J = 10.0 Hz, 4H, Ar), 7.21
6
(
d, J = 10.0 Hz, 4H, Ar), 3.61 (s, 4H, 2 x CH ), 2.56 (s, 4H, 2 x CH ), 1.99 (br s, 2H,
2 2
1
3
2
x NH), 1.26 (s, 18H, 6 x CH ). C NMR (DMSO-d ): δ 149.2 (2 x C, Ar), 138.3 (2
3 6
x C, Ar), 128.1 (4 x CH, Ar), 125.2 (4 x CH, Ar), 53.1 (2 x CH ), 48.7 (2 x CH ), 34.6
2
2
+
(
2 x C), 31.7 (6 x CH ). [ESI-HRMS] calculated for C H N : 353.2951 [M+H] .
3 22 37 2
+
Found: 353.2948 [M+H] .
1
2
4
.2.2.4. N ,N -Bis(pyridin-3-yl methyl)ethane-1,2-diamine (4j). Prepared from N,N'-
(ethane-1,2-diyl)bis(1-(pyridine-3-yl)methanimine) (3j) (11.57 g, 48.57 mmol).
Product obtained as a yellow oil, yield 3.7 g (31%). TLC (2:1 petroleum ether/EtOAc),
1
Rf = 0.12. H NMR (DMSO-d ): δ 8.84 (s, 2H, pyridine), 8.60 (m, 2H, pyridine), 8.15
6
(t, J = 5.8 Hz, 2H, pyridine), 7.41 (d, J = 5.7 Hz, 2H, pyridine), 3.85 (s, 4H, 2 x CH2),
1
3
3
.71 (s, 4H, 2 x CH ), 2.46 (br s, 2H, 2 x NH). C NMR (DMSO-d ): δ 150.3 (2 x
2 6
CH, pyridine), 148.8 (CH, pyridine), 148.7 (CH, pyridine), 136.7 (2 x CH, pyridine),
1
4
2
34.7 (C, pyridine), 133.9 (C, pyridine), 123.7 (2 x CH, pyridine), 53.6 (2 x CH2),
+
8.6 (2 x CH ). [ESI-HRMS] calculated for C H N : 242.1531 [M+H] . Found:
2
14 19
4
+
42.1536 [M+H] .
2
3

4
.2.3. General method for the preparation of ethyl 1,4-bis(aryl)piperazine-2-
carboxylates (5). To a stirred solution of diamine (4) (1 m.eq., 2.84 g, 10.27 mmol) in
anhydrous toluene at 80 °C (10 mL/mmol), was added triethylamine (2.5 m.eq.) and
2
,3-dibromopropionic acid ethyl ester (1.05 m.eq.) dropwise, and the reaction mixture
was heated at 80 °C overnight. The solvent was evaporated and the residue extracted
with CH Cl (10 mL/mmol), washed with saturated aqueous NaHCO (5 x 5
2
2
3
mL/mmol) and brine (3 x 5 mL/mmol). The organic layer was dried (MgSO4),
concentrated under reduced pressure, and crude product was purified by gradient flash
column chromatography.
4
.2.3.1. Ethyl 1,4-bis(4-fluorobenzyl)piperazine-2-carboxylate (5a). Prepared from
1
2
N ,N -bis(4-fluorobenzyl)ethane-1,2-diamine (4a) (2.84 g, 10.27 mmol). The product
was eluted with petroleum ether – EtOAc 87.5:12.5 v/v to give the product as a white
o
solid, yield 2.49 g (65%). M.p. 46-48 C. TLC (4:1 petroleum ether/EtOAc), Rf =
1
0
.82. H NMR (CDCl ): δ 7.29 (m, 4H, Ar), 6.99 (m, 4H, Ar), 4.19 (q, J = 7.0 Hz, 2H,
3
CH CH ), 3.89 (d, J = 13.3 Hz, 1H, H-7 ), 3.58 (d, J = 13.0 Hz, 1H, H-7 ), 3.53 (d, J
2
3
a
b
=
13.2 Hz, 1H, H-8 ), 3.40 (d, J = 13.2 Hz, 1H, H-8 ), 3.30 (dd, J = 3.6, 5.5 Hz, 1H,
a b
H-2), 3.07 (m, 1H, H-3 ), 2.76 (m, 1H, H-3 ), 2.59-2.39 (m, 4H, H-6 , H-5 ), 1.26 (t,
a
b
a,b
a,b
1
3
J = 7.1 Hz, 3H, CH CH ). C NMR (CDCl ): δ 172.0 (C=O), 163.0 (C, Ar), 162.9 (C,
2
3
3
Ar), 161.1 (C, Ar), 161.0 (C, Ar), 130.5 (2 x CH, Ar), 130.4 (2 x CH, Ar), 130.3 (2 x
CH, Ar), 130.2 (2 x CH, Ar), 62.6 (CH, piperazine), 61.8 (CH CH ), 60.4 (CH ), 58.8
2
3
2
(CH ), 55.4 (CH , piperazine), 53.0 (CH , piperazine), 48.5 (CH , piperazine), 14.3
2 2 2 2
1
9
(
CH CH ). F NMR (470 MHz, DMSO-d ): δ -115.99 (2 x C H F). [ESI-HRMS]
2
3
6
6
5
+
+
calculated for C H F N O : 375.1885 [M+H] . Found: 375.1867 [M+H] .
2
1
25
2
2
2
4
.2.3.2. Ethyl 1,4-bis(4-chlorobenzyl)piperazine-2-carboxylate (5b). Prepared from
1
2
N ,N -bis(4-chlorobenzyl)ethane-1,2-diamine (4b) (4.00 g, 12.93 mmol). The product
2
4

was eluted with petroleum ether – EtOAc 87.5:12.5 v/v to give the product as a white
o
solid, yield 4.48 g (85%). M.p. 58-60 C. TLC (4:1 petroleum ether/EtOAc), Rf =
1
0
.60. H NMR (CDCl ): δ 7.26 (m, 8H, Ar), 4.19 (q, J = 7.1 Hz, 2H, CH CH ), 3.90
3
2
3
(
d, J = 13.5 Hz, 1H, H-7 ), 3.59 (d, J = 11.6 Hz, 1H, H-7 ), 3.54 (d, J = 13.4 Hz, 1H,
a
b
H-8 ), 3.41 (d, J = 13.4 Hz, 1H, H-8 ), 3.32 (m, 1H, H-2), 3.08 (m, 1H, H-3 ), 2.77 (m,
a
b
a
1
7
H, H-3 ), 2.59 (m, 1H, H-6 ), 2.53 (m, 1H, H-6 ), 2.40 (m, 2H, H-5 ), 1.26 (t, J =
b a b a,b
1
3
.1 Hz, 3H, CH CH ). C NMR (CDCl ): δ 172.1 (C=O), 163.0 (2 x C, Ar), 161.1 (2
2
3
3
x C, Ar), 130.3 (4 x CH, Ar), 128.4 (4 x CH, Ar), 62.7 (CH, piperazine), 61.8
(
CH CH ), 60.4 (CH ), 58.8 (CH ), 55.4 (CH , piperazine), 53.1 (CH , piperazine),
2 3 2 2 2 2
4
6
8.4 (CH , piperazine), 14.0 (CH CH ). Anal. Calcd for C H Cl N O (407.3335): C,
2 2 3 21 24 2 2 2
1.92; H, 5.94; N, 6.87. Found: C, 61.66; H, 5.91; N, 6.84.
.2.3.3. Ethyl 1,4-bis(4-bromobenzyl)piperazine-2-carboxylate (5c). Prepared from
4
1
2
N ,N -bis(4-bromobenzyl)ethane-1,2-diamine (4c) (9.35 g, 23.48 mmol). The product
was eluted with petroleum ether – EtOAc 87.5:12.5 v/v to give the product as a white
o
solid, yield 9.68 g (83%). M.p. 64-66 C. TLC (4:1 petroleum ether/EtOAc), Rf =
1
0
.55. H NMR (CDCl ): δ 7.44 (m, 4H, Ar), 7.20 (m, 4H, Ar), 4.19 (q, J = 7.1 Hz, 2H,
3
CH CH ), 3.89 (d, J = 13.6 Hz, 1H, H-7 ), 3.59 (d, J = 13.4 Hz, 1H, H-7 ), 3.52 (d, J
2
3
a
b
=
13.4 Hz, 1H, H-8 ), 3.39 (d, J = 13.4 Hz, 1H, H-8 ), 3.32 (dd, J = 3.4, 5.7 Hz, 1H,
a b
H-2), 3.08 (m, 1H, H-3 ), 2.79 (m, 1H, H-3 ), 2.55-2.38 (m, 4H, H-6 , H-5 ), 1.26 (t,
a
b
a,b
a,b
1
3
J = 7.1 Hz, 3H, CH CH ). C NMR (CDCl ): δ 171.9 (C=O), 137.5 (2 x C, Ar),
2
3
3
1
37.1 (2 x C, Ar), 131.4 (2 x CH, Ar), 131.3 (2 x CH, Ar), 130.7 (2 x CH, Ar), 130.6
(
(
2 x CH, Ar), 62.5 (CH, piperazine), 61.8 (CH CH ), 60.5 (CH ), 58.9 (CH ), 55.4
2
3
2
2
CH , piperazine), 53.1 (CH , piperazine), 48.4 (CH , piperazine), 14.3 (CH CH ).
2
2
2
2
3
Anal. Calcd for C H Br N O (496.2355): C, 50.83; H, 4.84; N, 5.64. Found: C,
21
24
2
2
2
5
0.51; H, 4.80; N, 5.57.
2
5

4
.2.3.4. Ethyl 1,4-bis(3,5-dichlorobenzyl)piperazine-2-carboxylate (5d). Prepared
1
2
from N ,N -bis(3,5-dichlorobenzyl)ethane-1,2-diamine (4d) (6.0 g, 16.87 mmol). The
product was eluted with petroleum ether – EtOAc 9:1 v/v to give the product as a buff
o
solid, yield 4.46 g (71%). M.p. 80-84 C. TLC (4:1 petroleum ether/EtOAc), Rf =
1
0
7
3
3
2
1
.85. H NMR (DMSO-d ): δ 7.79 (m, 2H, para-Ar), 7.36 (m, 4H, Ar), 4.09 (q, J =
6
.1 Hz, 2H, CH CH ), 3.87 (d, J = 13.7 Hz, 1H, H-7 ), 3.72 (d, J = 14.0 Hz, 1H, H-7 ),
2
3
a
b
.61 (d, J = 14.0 Hz, 1H, H-8 ), 3.47 (d, J = 14.0 Hz, 1H, H-8 ), 3.38 (m, 1H, H-2),
a
b
.36 (m, 1H, H-3 ), 3.03 (m, 1H, H-3 ), 2.78 (m, 1H, H-6 ), 2.62 (m, 1H, H-6 ), 2.44-
a
b
a
b
13
.30 (m, 2H, H-5 ), 1.15 (t, J = 7.1 Hz, 3H, CH CH ). C NMR (DMSO-d ): δ
a,b
2
3
6
71.7 (C=O), 144.4 (2 x C, Ar), 143.4 (2 x C, Ar), 134.4 (C, Ar), 134.3 (C, Ar), 127.5
2 x CH, Ar), 127.3 (2 x CH, Ar), 127.1 (CH, Ar), 126.9 (CH, Ar), 62.5 (CH,
piperazine), 60.4 (CH CH ), 60.4 (CH ), 57.8 (CH ), 55.2 (CH , piperazine), 53.3
(
2
3
2
2
2
(
CH , piperazine), 48.5 (CH , piperazine), 14.5 (CH CH ). Anal. Calcd for
2 2 2 3
C H Cl N O (476.2236): C, 52.96; H, 4.66; N, 5.88. Found: C, 52.69; H, 4.57; N,
21
22
4
2
2
5
.81.
.2.3.5. Ethyl 1,4-bis(4-ethylbenzyl)piperazine-2-carboxylate (5f). Prepared from
4
1
2
N ,N -bis(4-ethylbenzyl)ethane-1,2-diamine (4f) (9.0 g, 30.35 mmol). The product
was eluted with petroleum ether – EtOAc 9:1 v/v to give the product as a colourless
1
oil, yield 8.8 g (74%). TLC (95:5 CH Cl /MeOH), Rf = 0.97. H NMR (DMSO-d ): δ
2
2
6
7
3
.15 (m, 8H, Ar), 4.06 (q, J = 7.7 Hz, 2H, CH CH ), 3.79 (d, J = 13.4 Hz, 1H, H-7 ),
2 3 a
.57 (d, J = 13.3 Hz, 1H, H-7 ), 3.48 (d, J = 13.2 Hz, 1H, H-8 ), 3.29 (m, 2H, H-8 ,
b
a
b
H-2), 2.98 (m, 1H, H-3 ), 2.69 (m, 1H, H-3 ), 2.56 (q, J = 7.5 Hz, 4H, 2 x CH CH ),
a
b
2
3
1
3
2
1
.40 (m, 4H, H-6 , H-5 ), 1.15 (m, 9H, 3 x CH CH ). C NMR (DMSO-d ): δ
a,b a,b 2 3 6
71.7 (C=O), 142.8 (2 x C, Ar), 135.6 (2 x C, Ar), 129.0 (4 x CH, Ar), 128.0 (4 x CH,
Ar), 64.7 (CH, piperazine), 63.0 (CH CH ), 61.9 (CH ), 60.7 (CH ), 58.8 (CH ,
2
3
2
2
2
2
6

piperazine), 55.4 (CH , piperazine), 53.3 (CH , piperazine), 28.3 (2 x CH CH ), 16.1
2
2
2
3
(2 x CH CH ), 14.5 (CH CH ). [ESI-HRMS] calculated for C H N O : 395.2699
2 3 2 3 25 35 2 2
+
+
[
M+H] . Found: 395.2692 [M+H] .
4
.2.3.6. Ethyl 1,4-bis(4-isopropylbenzyl)piperazine-2-carboxylate (5g). Prepared from
1
2
N ,N -bis(4-isopropylbenzyl)ethane-1,2-diamine (4f) (9.0 g, 27.73 mmol). The
product was eluted with petroleum ether – EtOAc 9:1 v/v to give the product as a
1
colourless oil, yield 10.0 g (85%). TLC (95:5 CH Cl /MeOH), Rf = 0.98. H NMR
2
2
(
DMSO-d ): δ 7.17 (m, 8H, Ar), 4.08 (q, J = 7.8 Hz, 2H, CH CH ), 3.79 (d, J = 13.3
6
2
3
Hz, 1H, H-7 ), 3.57 (d, J = 13.2 Hz, 1H, H-7 ), 3.48 (d, J = 13.1 Hz, 1H, H-8 ), 3.27
a
b
a
(
m, 2H, H-8 , H-2), 2.99 (m, 1H, H-3 ), 2.84 (sept, 2H, J = 6.8 Hz, 2 x CH(CH ) ),
b a 3 2
2
.70 (m, 1H, H-3 ), 2.47-2.30 (m, 4H, H-6 , H-5 ), 1.17 (d, J = 7.0 Hz, 12H, 4 x
b a,b a,b
1
3
CH ), 1.11 (t, J = 7.2 Hz, 3H, CH CH ). C NMR (DMSO-d ): δ 171.7 (C=O), 147.4
3
2
3
6
(2 x C, Ar), 136.3 (C, Ar), 135.8 (C, Ar), 129.0 (2 x CH, Ar), 128.9 (2 x CH, Ar),
1
26.5 (2 x CH, Ar), 126.4 (2 x CH, Ar), 65.4 (CH, piperazine), 61.9 (CH CH ), 60.2
2
3
(
CH ), 60.1 (CH ), 58.8 (CH , piperazine), 55.4 (CH , piperazine), 53.4 (CH ,
2 2 2 2 2
piperazine), 33.6 (2 x CH(CH ) ), 24.4 (4 x CH ), 14.5 (CH CH ). [ESI-HRMS]
3
2
3
2
3
+
+
calculated for C H N O : 423.3012 [M+H] . Found: 423.3016 [M+H] .
2
7
39
2
2
4
.2.3.7. Ethyl 1,4-bis(4-(tert-butyl)benzyl)piperazine-2-carboxylate (5h). Prepared
1
2
from N ,N -bis(4-tert-butylbenzyl)ethane-1,2-diamine (4h) (6.0 g, 17.01 mmol). The
product was eluted with petroleum ether – EtOAc 92.5:7.5 v/v to give the product as a
o
white solid, yield 5.0 g (65%). M.p. 82-84 C. TLC (5:1 petroleum ether/EtOAc), Rf
1
=
0.83. H NMR (CDCl ): δ 7.31 (m, 4H, Ar), 7.18 (m, 4H, Ar), 4.06 (q, J = 7.1 Hz,
3
2
H, CH CH ), 3.80 (d, J = 13.1 Hz, 1H, H-7 ), 3.54 (d, J = 13.2 Hz, 1H, H-7 ), 3.39
2 3 a b
(
m, 2H, H-8 ), 3.30 (dd, J = 3.6, 5.6 Hz, 1H, H-2), 2.99 (m, 1H, H-3 ), 2.72 (m, 1H,
a
,
b
a
H-3 ), 2.41 (m, 4H, H-6 , H-5 ), 1.25 (s, 18H, 6 x CH ), 1.11 (t, J = 6.7 Hz, 3H,
b
a,b
a,b
3
2
7

1
3
CH CH ). C NMR (CDCl ): δ 171.5 (C=O), 135.8 (2 x C, Ar), 132.2 (2 x C, Ar),
2
3
3
1
28.8 (4 x CH, Ar), 125.4 (4 x CH, Ar), 64.1 (CH, piperazine), 63.0 (CH CH ), 61.8
2 3
(
CH ), 60.3 (CH ), 58.6 (CH , piperazine), 53.2 (CH , piperazine), 47.9 (CH ,
2 2 2 2 2
piperazine), 34.6 (2 x C(CH ) ), 31.6 (6 x CH ), 14.5 (CH CH ). [ESI-HRMS]
3
3
3
2
3
+
+
calculated for C H N O : 451.3324 [M+H] . Found: 451.3327 [M+H] .
2
9
43
2
2
4
.2.3.8. Ethyl 1,4-bis(pyridin-4-ylmethyl)piperazine-2-carboxylate (5i). Prepared from
1
2
N ,N -bis(pyridin-4-yl)ethane-1,2-diamine (4i) (3.66 g, 15.10 mmol). The product
was eluted with CH Cl -MeOH-Et N 97.5:1.5:1 v/v/v to give the product as a brown
2
2
3
1
oil, yield 3.36 g (65%). TLC (95:5 CH Cl /MeOH), Rf = 0.32. H NMR (DMSO-d ):
2
2
6
δ 8.50 (m, 4H, Ar), 7.30 (m, 4H, Ar), 4.10 (q, J = 7.6 Hz, 2H, CH CH ), 3.90 (d, J =
2
3
1
3
3.0 Hz, 1H, H-7 ), 3.74 (d, J = 13.7 Hz, 1H, H-7 ), 3.61 (d, J = 13.4 Hz, 1H, H-8 ),
a b a
.46 (d, J = 13.4 Hz, 1H, H-8 ), 3.40 (dd, J = 3.6, 5.9 Hz, 1H, H-2), 3.04 (m, 1H, H-
b
3a), 2.81 (m, 1H, H-3 ), 2.58 (m, 1H, H-6 ), 2.45 (m, 1H, H-6 ), 2.36 (m, 2H, H-5 ),
b
a
b
a,b
13
1
.15 (t, J = 7.0 Hz, 3H, CH CH ). C NMR (DMSO-d ): δ 171.6 (C=O), 150.0 (4 x
2 3 6
CH, Ar), 148.7 (C, Ar), 147.7 (C, Ar), 124.0 (2 x CH, Ar), 123.8 (2 x CH, Ar), 62.0
(
CH, piperazine), 60.7 (CH CH ), 60.4 (CH ), 57.8 (CH ), 55.4 (CH , piperazine),
2 3 2 2 2
5
3.4 (CH , piperazine), 52.5 (CH , piperazine), 14.6 (CH CH ). [ESI-HRMS]
2 2 2 3
+
+
calculated for C H N O : 341.1899 [M+H] . Found: 341.1910 [M+H] .
1
9
25
4
2
4
.2.3.9. Ethyl 1,4-bis(pyridin-3-ylmethyl)piperazine-2-carboxylate (5j). Prepared from
1
2
N ,N -bis(pyridin-3-yl)ethane-1,2-diamine (4j) (3.50 g, 14.44 mmol). The product
was eluted with CH Cl -MeOH-Et N 96.5:2.5:1 v/v/v to give the product as a brown
2
2
3
1
oil, yield 2.30 g (47%). TLC (95:5 CH Cl /MeOH), Rf = 0.31. H NMR (DMSO-d ):
2
2
6
δ 8.47 (s, 2H, pyridine), 8.46 (m, 2H, pyridine), 7.67 (t, J = 7.7 Hz, 2H, pyridine),
.35 (d, J = 5.6 Hz, 2H, pyridine), 4.08 (q, J = 7.8 Hz, 2H, CH CH ), 3.88 (d, J = 14.0
7
2
3
Hz, 1H, H-7 ), 3.70 (d, J = 13.8 Hz, 1H, H-7 ), 3.60 (d, J = 13.5 Hz, 1H, H-8 ), 3.38
a
b
a
2
8

(
(
(
m, 2H, H-8 , H-2), 3.00 (m, 1H, H-3 ), 2.75 (m, 1H, H-3 ), 2.54 (m, 1H, H-6 ), 2.42
b a b a
13
m, 1H, H-6 ), 2.34 (m, 2H, H-5 ), 1.13 (t, J = 7.7 Hz, 3H, CH CH ). C NMR
b
a,b
2
3
DMSO-d ): δ 171.6 (C=O), 150.4 (CH, pyridine), 150.2 (CH, pyridine), 148.8 (CH,
6
pyridine), 148.7 (CH, pyridine), 136.9 (CH, pyridine), 136.6 (CH, pyridine), 134.7 (C,
pyridine), 133.9 (C, pyridine), 123.9 (CH, pyridine), 123.8 (CH, pyridine), 62.3 (CH,
piperazine), 60.4 (CH CH ), 59.1 (CH ), 56.3 (CH ), 56.0 (CH , piperazine), 55.2
2
3
2
2
2
(
CH , piperazine), 53.2 (CH , piperazine), 14.5 (CH CH ). [ESI-HRMS] calculated
2 2 2 3
+
+
for C H N O : 341.1899 [M+H] . Found: 341.1903 [M+H] .
1
9
25
4
2
4
.2.4. General method for the preparation of alcohols (6). To an ice-cooled solution
of ethyl carboxylate (5) (1 m.eq.) in dry THF (3 mL/mmol) was added LiAlH (1M in
4
THF, 1.5 m.eq.) dropwise over 25 min. The reaction was then stirred at room
temperature overnight, then cooled in an ice-bath and carefully quenched with H O
2
until cessation of effervescence. The reaction mixture was extracted with EtOAc (2 x
1
0 mL/mmol), then the combined organic layers washed with H O (3 x 10 mL/mmol),
2
dried (MgSO ) and concentrated under reduced pressure. The crude product was
4
purified by gradient flash column chromatography.
4
.2.4.1. (1,4-Bis(4-fluorobenzyl)piperazin-2-yl)methanol (6a). Prepared from ethyl
,4-bis(4-fluorobenzyl)piperazine-2-carboxylate (5a) (1.2 g, 3.20 mmol). The product
1
was eluted with CH Cl -MeOH 95:5 v/v to give the product as a yellow oil, yield 0.86
2
2
1
g (81%). TLC (95:5 CH Cl /MeOH), Rf = 0.55. H NMR (CDCl ): δ 7.30 (m, 8H, Ar),
2
2
3
3
.61 (m, 2H, H-9 ), 3.48 (m, 2H, H-7 ), 2.96 (m, 2H, H-8 ), 2.67 (m, 3H, H-2, H-
a,b a,b a,b
1
3
3a,b), 2.59 (m, 1H, H-6 ), 2.53-2.45 (m, 2H, H-6 , H-5 ), 2.37 (m, 1H, H-5 ). C NMR
a
b
a
b
(
CDCl ): δ 163.1 (C, Ar), 161.1 (C, Ar), 138.4 (C, Ar), 137.5 (C, Ar), 128.9 (4 x CH,
3
Ar), 128.4 (4 x CH, Ar), 62.3 (CH ), 58.7 (CH, piperazine), 58.0 (CH ), 57.2 (CH ),
2
2
2
1
9
5
5.9 (CH , piperazine), 52.4 (CH , piperazine), 49.9 (CH , piperazine). F NMR
2 2 2
2
9