Development of alkyl glycerone phosphate synthase inhibitors: Structure-activity relationship and effects on ether lipids and epithelial-mesenchymal transition in cancer cells

Article abstract of DOI:10.1016/j.ejmech.2018.11.050

In aggressive tumors, alkylglyceronephosphate synthase (AGPS) controls cellular ether phospholipid utilization and metabolism to promote cancer cell proliferation and motility. SAR studies on the first-in-class AGPS inhibitor 1, discovered by our group, led to the 2,6-difluoro analog 2i which showed higher binding affinity than 1 in vitro. In 231MFP cancer cells, 2i reduced ether lipids levels and cell migration rate. When tested in PC-3 and MDA-MB-231 cancer cells, 2i specifically impaired epithelial to mesenchymal transition (EMT) by modulating E-cadherin, Snail and MMP2 expression levels. Moreover, the combination of siRNAs against AGPS and 2i provided no additive effect, confirming that the modulation of 2i on EMT specifically relies on AGPS inhibition. Finally, this compound also affected cancer cell proliferation especially in MDA-MB-231 cells expressing higher AGPS level, whereas it provided negligible effects on MeT5A, a non-tumorigenic cell line, thus showing cancer specificity.

Full text of DOI:10.1016/j.ejmech.2018.11.050

Accepted Manuscript
Development of alkyl glycerone phosphate synthase inhibitors: Structure-activity
relationship and effects on ether lipids and epithelial-mesenchymal transition in
cancer cells
Giulia Stazi, Cecilia Battistelli, Valentina Piano, Roberta Mazzone, Biagina Marrocco,
Sara Marchese, Sharon M. Louie, Clemens Zwergel, Lorenzo Antonini, Alexandros
Patsilinakos, Rino Ragno, Monica Viviano, Gianluca Sbardella, Alessia Ciogli,
Giancarlo Fabrizi, Roberto Cirilli, Raffaele Strippoli, Alessandra Marchetti, Marco
Tripodi, Daniel K. Nomura, Andrea Mattevi, Antonello Mai, Sergio Valente
PII:
S0223-5234(18)31009-2
DOI:
https://doi.org/10.1016/j.ejmech.2018.11.050
EJMECH 10908
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 24 July 2018
Revised Date: 6 November 2018
Accepted Date: 20 November 2018
Please cite this article as: G. Stazi, C. Battistelli, V. Piano, R. Mazzone, B. Marrocco, S. Marchese, S.M.
Louie, C. Zwergel, L. Antonini, A. Patsilinakos, R. Ragno, M. Viviano, G. Sbardella, A. Ciogli, G. Fabrizi,
R. Cirilli, R. Strippoli, A. Marchetti, M. Tripodi, D.K. Nomura, A. Mattevi, A. Mai, S. Valente, Development
of alkyl glycerone phosphate synthase inhibitors: Structure-activity relationship and effects on ether
lipids and epithelial-mesenchymal transition in cancer cells, European Journal of Medicinal Chemistry
(2018), doi: https://doi.org/10.1016/j.ejmech.2018.11.050.
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
Cancer specificity: no
toxic effect in non-
malignant human
mesothelial MeT5A cells
EMT modulation:
↑ E-cadherin
↓ Snail and MMP2
↓ 50-70% ether lipids
↓ 78% cell migration in
aggressive 231MFP
breast cancer cells
mRNA levels in PC-3
prostate and MDA-MB-231
breast cancer cells
AGPS binding (X ray
crystallography) and
inhibition (ThermoFAD
and radioactivity assays)

ACCEPTED MANUSCRIPT
Development of Alkyl Glycerone Phosphate Synthase Inhibitors: Structure-Activity Relation-
ship and Effects on Ether Lipids and Epithelial-Mesenchymal Transition in Cancer Cells
Giulia Stazia,1, Cecilia Battistellib,1, Valentina Pianoc,d,1, Roberta Mazzonea,e, Biagina Marroccoc,
Sara Marchesec, Sharon M. Louief, Clemens Zwergela, Lorenzo Antoninig, Alexandros Patsilinakosa,
Rino Ragnoa, Monica Vivianoh, Gianluca Sbardellah, Alessia Cioglia, Giancarlo Fabrizia, Roberto
Cirillii, Raffaele Strippolib,j, Alessandra Marchettib, Marco Tripodib,j,k, Daniel K. Nomuraf, Andrea
Mattevic,*, Antonello Maia,k,*, and Sergio Valentea,*
aDepartment of Chemistry and Technologies of Drugs, Sapienza University of Rome, P. le A. Moro
5, 00185 Rome, Italy
bDepartment of Molecular Medicine, Sapienza University of Rome, Viale Regina Elena 324, 00161
Rome, Italy
cDepartment of Biology and Biotechnology, University of Pavia, Via Ferrata 9, 27100 Pavia, Italy
dDepartment of Mechanistic Cell Biology, Max-Planck Institute of Molecular Physiology, Otto-
Hahn-Str. 11, 44227 Dortmund, Germany
eCenter for Life Nano Science@Sapienza, Italian Institute of Technology, Via Regina Elena 291,
00161 Rome, Italy
fDepartments of Chemistry, Molecular and Cell Biology, and Nutritional Sciences and Toxicology,
University of California, Berkeley, Berkeley, CA 94720 USA
gAlchemical Dynamics s.r.l. 00125 Rome, Italy
hDepartment of Pharmacy, Epigenetic Med Chem Lab, University of Salerno, via Ponte Don Melil-
lo, 84084 Fisciano (SA), Italy
iCentro Nazionale per il Controllo e la Valutazione dei Farmaci, Istituto Superiore di Sanità, Viale
Regina Elena 299, I-00161 Rome , Italy
jGene Expression Laboratory, National Institute for Infectious Diseases "Lazzaro Spallanzani"
I.R.C.C.S., Rome, Italy
kPasteur Institute - Cenci Bolognetti Foundation, Sapienza Università di Roma, Viale Regina Elena
324, 00161 Rome, Italy

ACCEPTED MANUSCRIPT
KEYWORDS (AGPS inhibitors, ether lipids, cancer, E-cadherin, Snail)
ABSTRACT: In aggressive tumors, alkylglyceronephosphate synthase (AGPS) controls cellular
ether phospholipid utilization and metabolism to promote cancer cell proliferation and motility. SAR
studies on the first-in-class AGPS inhibitor 1, discovered by our group, led to the 2,6-difluoro ana-
log 2i which showed higher binding affinity than 1 in vitro. In 231MFP cancer cells, 2i reduced
ether lipids levels and cell migration rate. When tested in PC-3 and MDA-MB-231 cancer cells, 2i
specifically impaired epithelial to mesenchymal transition (EMT) by modulating E-cadherin, Snail
and MMP2 expression levels. Moreover, the combination of siRNAs against AGPS and 2i provided
no additive effect, confirming that the modulation of 2i on EMT specifically relies on AGPS inhibi-
tion. Finally, this compound also affected cancer cell proliferation especially in MDA-MB-231 cells
expressing higher AGPS level, whereas it provided negligible effects on MeT5A, a non-tumorigenic
cell
line,
thus
showing
cancer
specificity.
INTRODUCTION
Aggressive metastatic tumors exhibit an altered cell metabolism favoring increased cell proliferation
rate and invasiveness. Among the metabolic processes involved in cancer development, aberrant li-
pid metabolism and enhanced de novo lipogenesis play an essential role in promoting cell division
and motility.1-3 Although the physiological role of ether lipids is not fully understood, many findings
suggest their implication in driving cancer pathogenicity by contributing to free radical scavenging,
cell motility, cell membrane plasticity and vesicle formation, and generation of oncogenic lipid sig-
naling molecules.4, 5 The critical step in ether phospholipid synthesis is catalyzed by the peroxisomal
flavoenzyme alkylglyceronephosphate synthase (AGPS).6 AGPS enables the acyl/alkyl exchange of
the substrate acyl-dihydroxyacetone phosphate (acyl-DHAP), using the flavin cofactor (FAD)
through an unusual non-redox catalytic mechanism. Indeed, the FAD covalently binds the DHAP
moiety of acyl-DHAP, acting as a trap to immobilize the substrate and allows the exchange of the
fatty acid chain of acyl-DHAP with a fatty alcohol substrate, forming the precursor of all ether lipid
species.3, 6-8 It has been shown that AGPS is up-regulated across different types of aggressive and
metastatic tumors, such as human breast, melanoma and prostate cancers.3, 7-10 To elucidate the con-
tribution of AGPS to cancer pathogenicity, the gene encoding for the enzyme was genetically si-
lenced in breast 231MFP and melanoma C8161 cancer cell lines. Its inactivation led to an overall
reduced level of multiple structural ether lipids, as well as oncogenic signaling molecules, such as

ACCEPTED MANUSCRIPT
lysophosphatidic acid ethers (LPAe), platelet activating factor ethers (PAFe) and eicosanoids, which
contribute to the inflammatory microenvironment.11
H3C
F
CH3
O
CH3
O
O
CH3
H
N
O
O
CH3
N
H
N
H
O
HO
O
N
H
ZINC-69435460
H3C
HN
OHC
R
O
O
F
CH3
O
Antimycin A: A1, R= C6H13
A2, R= C5H11
H
N
N
H
O
A3, R= C4H9
N
H
A4, R= C3H7
1
Chart 1. First disclosed inhibitors of AGPS
As AGPS knockdown showed remarkable effects on tumor growth in a xenograft mouse model,3 we
implemented a small molecule screening to identify the first compounds able to reproduce the phe-
notype induced by AGPS silencing. We identified Zinc-69435460 (3-(2-fluorophenyl)-N-(1-(2-oxo-
2,3-dihydro-1H-benzo[d]imidazol-5-yl)ethyl)butanamide) and antimycin A (Chart 1) as first AGPS
inhibitors.12, 13 High resolution X-ray structure of the AGPS/Zinc-69435460 complex highlighted a
detailed view of the inhibitor binding mode, with all the three inhibitor moieties (fluorophenyl, pro-
panamide, and dihydrobenzoimidazole) involved in specific interactions with the surrounding pro-
tein/amino acid residues and with the FAD cofactor. Compound 1, the analog of Zinc-69435460
lacking one of the two racemic methyl groups (Chart 1), showed higher binding affinity and the
same inhibitory effect as the prototype, thus it was chosen as lead compound for further biological
evaluation. In C8161 melanoma, 231MFP breast, and SKOV3 ovarian cancer cell lines 1 was able to
selectively lower ether lipid levels as well as to induce a 50% reduction of cell survival and migra-
tion rate.13
Here, we present some SAR investigation on the 1 scaffold aimed to pinpoint the main structural re-
quirements for AGPS inhibition, and to improve the potency in both enzyme and cellular assays. A
3

ACCEPTED MANUSCRIPT
variant of the ThermoFluor assay, ThermoFAD, measuring the unfolding temperature of the protein
by monitoring the increase in FAD fluorescence upon release from the protein, performed on the 1
analogs highlighted the 2,6-difluoro compound 2i as the most effective AGPS binder. This ability
has been confirmed by the X-ray structure of the complex AGPS/2i. The 2i inhibitory effect was
validated by an in vitro radioactivity-based enzymatic assay and in 231MFP cancer cells where it
was lowering ether lipids and cell migration rate. Finally, the effects of genetic (siRNAs) and/or
pharmacological (1 and 2i) AGPS inhibition on epithelial-mesenchymal transition (EMT) in prostate
PC-3 and breast MDA-MB-231 cancer cells have been assessed.
Figure 1. Development of novel AGPS inhibitors 2a-l, q-u. aRef. 13.
4

ACCEPTED MANUSCRIPT
RESULTS AND DISCUSSION
Chemistry
We performed chemical manipulations on the 1 structure (Figure 1) including: i) removal or shift of
the 2-fluoro substituent to the 3- or 4-position of the phenyl ring (2a-c); ii) replacement of the 2-
fluoro with other substituents such as different halogens (2d,e), methyl or methoxy (2f,g), or double-
fluoro substitution (2h,i); iii) introduction of groups bulkier than methyl at the propionamide β-
carbon (2j,k); iv) shift of the methyl group from the propionamide β- to α-position (2l). The effect
of removal of the propionamide β-methyl group and/or unsaturation of the propionamide chain and
eventual insertion of a methyl group at the α-unsaturated position (2m-p) has been already reported
as at least in part detrimental for the AGPS binding activity.13
From this first round of synthesis and screening (ThermoFAD analysis), the 2,6-difluoro analog
2i clearly emerged as the most potent compound. Therefore, we further modified its structure by v)
shortening the propanamide or the aminomethyl dihydrobenzoimidazole chain (2q,r); vi) replacing
the 2-oxo-2,3-dihydro-1H-benzo[d]imidazole with 1H-benzo[d]imidazole (2s), 1,3-dimethyl-2-oxo-
2,3-dihydro-1H-benzo[d]imidazole (2t), or 3-oxo-3,4-dihydroquinoxaline (2u) (Figure 1).
The synthesis of derivatives 2a-l,r-u started from Wittig-Horner condensation between different-
ly substituted ketones and trimethyl phosphonoacetate by using 60% sodium hydride in tetrahydro-
furan at 0 °C, or alternatively between 2-fluorobenzaldehyde and triethyl 2-phosphonopropionate
with anhydrous potassium carbonate in ethanol at 80 °C, to afford the unsaturated esters 3a-d, 3f-l.
Such esters (3a-d, 3f-l) were then saturated through palladium catalyzed hydrogenation in methanol
by using a ThalesNano H-Cube Reactor to afford the corresponding propionic methyl esters 4a-l,
wherein compound 4e was commercially available. The esters 4a-l were converted to the corre-
sponding carboxylic acids 5a-l through basic hydrolysis (2N potassium hydroxide in ethanol). The
2-(2,6-difluorophenyl)propanoic acid 5m was prepared according to the literature.14 The acids 5a-m
5

ACCEPTED MANUSCRIPT
were finally coupled with the opportune amines 6-10 [see Supporting Information (SI) for their
preparation] by using N,N-carbonyldiimidazole and triethylamine in anhydrous N,N-
dimethylformamide to obtain the final amides 2a-l,q-u (Scheme 1).
Scheme 1. Synthesis of compounds 2a-l, q-u. Reagents and conditions: a) trimethyl phosphonoace-
tate, 60% NaH, dry THF, 0 °C to rt, N2 atmosphere, overnight (41-79%); or triethyl 2-
phosphonopropionate, dry K2CO3, dry EtOH, 80 °C, 2 h; b) MeOH, 10% Pd/C, PH2=10 bar, v=1
mL/min, 30 °C (79-83%); c) 2N KOH, EtOH, rt, overnight (94-97%); d) appropriate amine among
6-10, N,N-carbonyldiimidazole, dry DMF, rt or 70 °C, overnight (73-85%). The exact structures of
compounds 3-5 are reported in SI.
ThermoFAD screening of compounds 2a-l, q-u
We recently showed that ThermoFAD was a reliable assay to screen AGPS-binding molecules.13 Da-
ta determined with 2a-u (Table 1) clearly showed the crucial role of the fluoro (compare the thermal
shift (Tm) of 1, 2b-c with that of 2a) or other halogens or methyl group (see 2d-f) to have great
binding affinity. Indeed, the deletion of the fluorine at C2 position (2a) resulted in a decrease of Tm
(+ 1.5 °C) when compared with the reference 1 (+ 3.5 °C), whereas its replacement with other halo-
gens, such as chlorine (2d), bromine (2e) or with a methyl group (2f) was well tolerated, showing
Tm values for these compounds similar to the reference 1.
6

ACCEPTED MANUSCRIPT
Contrarily, the introduction of the methoxy group (2g) abrogated the AGPS binding of the com-
pound. When the fluoro atom was shifted to meta or para position on the phenyl ring (2b and 2c, re-
spectively) the meta-substitution (2b) gave a weak decrease of Tm value (+ 3.0 °C), whereas the pa-
ra-substitution (2c) provided a slight (weak) increase in Tm value (+ 4.0 °C). Hence, we also per-
formed the double insertion of 2,4- and 2,6-difluoro substitution (2h and 2i) at the phenyl ring, that
resulted in a stronger enzyme binding, with the 2,6-difluoro derivative 2i giving the highest increase
of Tm (+ 5 °C). An increase of the steric hindrance at the propionamide β–position with an ethyl
(2j) or phenyl (2k) substituent caused a decrease or complete loss in AGPS binding, respectively,
whereas the shift of the methyl group from the propionamide β- to the α-carbon atom (2l) was fully
tolerated. As already reported,13 the removal of the β-methyl group (2m) or the unsaturation of the
propionamide chain (2n,o) led to a loss of binding activity, in part restored by insertion of an α-
methyl group (2p). Based on these results, we further explored the 2i structure to acquire more SAR
data. The additional attempts to shorten the distance between the phenyl (2q) or the dihydrobenzo-
imidazole (2r) portion and the amide moiety provided compounds with very weak or no AGPS
binding, respectively, thus showing that the length of the molecule and, in particular, the distance
between the two aromatic moieties is a crucial point. Furthermore, the double methylation of the di-
hydrobenzoimidazole ureidic nitrogens (2s), or its replacement with a benzoimidazole heterocycle
(2t) or with a 3-oxo-3,4-dihydroquinoxaline ring (2u) were not tolerated, showing no protein-
inhibitor interaction.
Table 1
TermoFAD data of compounds 2a-ua
Cpd
1
Molecular structure
Tm, °C)
Cpd
2k
Molecular structure
Tm, °C)
+3.5
0
7

ACCEPTED MANUSCRIPT
+1.5
+3.0
0
2a
2b
2l
2mb
+3.0
2nb
2ob
2pb
+4.0
+3.5
+3.5
0
0
2c
2d
2e
+2.0
+3.0
+1
2f
2q
0
+3.5
+5
0
0
0
0
2g
2h
2i
2r
2s
2t
+1
2j
2u
aAll experiments were performed in triplicate; SD, standard deviations are within ±0.5°C; bRef. 13
Crystallographic studies of AGPS/2i complex
The three-dimensional structure of AGPS in complex with 2i was determined by X-ray crystallog-
raphy (2.7 Å) (Table 2). The binding mode of 2i is identical to that previously reported for 1 (the (S)
configuration has been tentatively assigned in the binding with the enzyme for both compounds), in-
cluding a pattern of small rearrangements in the side chains of the active site residues (Figure 2 and
Figure S1 in SI). The dihydrobenzoimidazole moiety is placed into the hydrophilic vertex of the V-
shaped enzyme cavity for the substrate, where it engages multiple H bonds with catalytic residues,
moreover it binds in front of the FAD cofactor, suggesting π interactions with the isoalloxazine ring.
8

ACCEPTED MANUSCRIPT
The presence of the second fluorine on the phenyl ring most likely enhances the hydrophobic inter-
actions of the 2i amide chain with the aliphatic residues surrounding the tunnel devoted to host the
acyl chain of the enzyme substrate.
Table 2
Crystal data collection and refinement statistics.
PDB code
Space group
a, b, c (Å)a
ɑ, β, ɣ (°)
6GOU
I2
147.96 120.64 148.16
90, 95.27 90
2.90
Resolution (Å)
Rsymb,c (%)
18.0 (84.2)
CC1/2c,d
0.99 (0.83)
Completenessc (%)
Redundancy
100 (100)
7.8 (8.1)
57552
11.2 (2.9)
17058
287
Unique reflections
I/σc
Protein atoms (N)
Ligands atoms (N)e
Ramachandran outliers (%)
Average B value (Å2)
Rcrystf (%)
0.4
49.0
21.8
Rfreef (%)
25.8
Rms bond length (Å)
Rms bond angles (°)
0.011
1.8
aThe unit cell of AGPS crystal contains two enzyme dimers.
bRsym=∑|Ii-<I>|/∑Ii, where Ii is the intensity of ith observation and <I> is the mean intensity of the
reflection.
cValues in parentheses are for reflections in the highest resolution shell.
9

ACCEPTED MANUSCRIPT
dA cut-off criterion for resolution limits was applied on the basis of the mean intensity correlation
coefficient of half-subsets of each dataset (CC1/2).
eThe subunit C does not contain the inhibitor.
fRcryst=∑|Fobs-Fcalc|/∑|Fobs| where Fobs and Fcalc are the observed and calculated structure fac-
tor amplitudes, respectively. Rcryst and Rfree were calculated using the working and test sets, re-
spectively.
Figure 2. Crystal structure of the AGPS/2i complex. Weighted 2Fo − Fc final electron density map
contoured at the 1.2σ level for 2i bound to subunit B of AGPS. The binding is mediated by specific
H bonds (black dashed lines) with the active site residues Asp303, Ser527, His616, and His617.
Molecular modeling studies
Molecular modeling and docking studies have been undertaken to check if different configurations
at the chiral centers of Zinc-69435460 (PDB entry code 5AE1, reported in its (S,R) configuration), 1
(PDB entry code 5ADZ) and 2i (herein described), both reported as (S) enantiomers, could have
significant differences in binding energies and thus in predicted inhibition potencies. AGPS com-
plexes were modeled with (S,S)-Zinc-69435460, (R,S)-Zinc-69435460, (R,R)-Zinc-69435460, (R)-1
and (R)-2i configurations to evaluate their likely formation (see binding mode description and meth-
ods in SI). Inspection of the minimized complexes revealed that (S,S)-Zinc-69435460, (R,S)-Zinc-
69435460, (R,R)-Zinc-69435460, (R)-1 and (R)-2i configurations adopt binding modes closely relat-
ed to the experimental conformations of (S,R)-Zinc-69435460, (S)-1 and (S)-2i with no great protein
reassembly (Figure 3 and Figure S1, S2 in SI). Analysis of protein-ligand binding free energies re-
10

ACCEPTED MANUSCRIPT
ported in Supporting Information Table S5 indicated that (S,S)-Zinc-69435460, (R,S)-Zinc-
69435460, (R,R)-Zinc-69435460, (R)-1 and (R)-2i can bind to AGPS with inhibition potencies com-
parable to those of their experimental binding configurations.
Figure 3. Experimental and modeled AGPS complexes with Zinc-69435460, 1 and 2i. (A): (S,R)-
Zinc-69435460 (light gray), (S,S)-Zinc-69435460 (blue), (R,S)-Zinc-69435460 (yellow) and (R,R)-
Zinc-69435460 (magenta); (B): (S)-1 (light green) and (R)-1 (salmon); (C): (S)-2i (gold) and (R)-2i
(light blue).
Separation, Chiroptical Characterization and AGPS binding affinity of compound 2i single
enantiomers
Although the molecular modelling and docking studies suggested that the two enantiomers of 2i
should not have significant differences in binding energies, we decided, however, to solve 2i enanti-
omers by stereoselective HPLC. Semipreparative separations of the enantiomers of 2i were carried
out by HPLC on the 1-cm i.d. Chiralpak IA column under normal phase conditions. The loading ef-
fect on the IA column was examined by keeping the concentration of the feed solution constant (i.e.
4.0 mg mL−1) and varying the injection volume from 0.25 to 1.0 mL. As can be noted in the chroma-
tograms depicted in Figure S3, for a sample loading of 4.0 mg the enantiomeric peaks were again
well separated and an easy UV signal-guided fractionation of two enantiomers was possible. Both
enantiomeric forms were obtained with high ee (> 99%) and in about 90% recovery. Polarimetric
11

ACCEPTED MANUSCRIPT
analysis indicated that the less retained enantiomer on the Chiralpak IA column was levorotatory in
ethanol solution and the pertinent approximate ORD curve calculated at six wavelengths was mono-
tonic. As shown in Figure S4, the ORD curves of the enantiomers collected on a semipreparative
scale were perfectly specular.
The pure collected single enantiomers were screened against AGPS enzyme to assess the influence
of chirality in their binding affinity through TermoFAD assay. The TermoFAD assay revealed that,
differently from the predicted result, the first eluted enantiomer (-)-2i provided +3 °C of Tm in-
crease, while the second one (+)-2i gave an higher increase of Tm (+ 6 °C), with the racemate dis-
playing + 5 °C in the same assay. This result will prompt us to investigate on the biological effects
of the single eutomer in a next work.
Effects of compounds 2i, 2q and 2s on AGPS enzymatic activity: radioactivity assay
To assess the effect of selected 2 compounds on AGPS catalytic activity, we performed a radioac-
tivity-based enzymatic assay using palmitoyl-DHAP and [1-14C]hexadecanol as substrates and de-
tecting the formation of [1-14C]hexadecyl DHAP as a function of time in the presence of com-
pounds. The catalytically inactive AGPS mutant T578F was used as a control. For the sake of clari-
ty, Figure 4 shows only the effects of 1, 2i, 2q and 2s.
12

ACCEPTED MANUSCRIPT
Figure 4. Inhibition of AGPS enzymatic activity was detected by a radioactivity assay using 100
ꢀM palmitoyl-DHAP, 96 ꢀM [1-14C]hexadecanol, and 180 ꢀM inhibitor to detect the formation of
[1-14C]hexadecyl-DHAP as a function of time. Controls were performed using AGPS alone and the
catalytically inactive AGPS mutant T578F8 as a positive control. Measurements were performed at
least in triplicate.
The effects of 2b-f,h,j-l,t are reported in Figure S5 in SI. Among the tested compounds, 2k, 2q, 2s
and 2t, which showed very weak or no binding in the ThermoFAD assay, did not display any AGPS
inhibition. The 2,6-difluoro-substituted compound 2i, for which was recorded the highest Tm shift
(+ 5 °C), provided the best inhibition profile in the progress curve being as effective as (or slightly
more potent than) the reference compound 1 (Figure 4). Enzyme kinetic assays to measure the Ki or
IC50 values for the identified AGPS inhibitors could not be performed due to the high hydrophobi-
city of the substrates which, even in presence of surfactants, form micelles and tend to precipitate in
the required experimental conditions, impairing accurate determination of the Km and rate values.
Effect of 2i on ether lipid level, on cell migration and survival in aggressive 231MFP breast
cancer cells
Compound 2i was then tested in breast 231MFP cancer cells at 500 µM and resulted able to de-
crease the levels of ether phospholipids by 50-70% after 24 h of treatment (Figure 5A), as verified
by lipidomic analyses through LC-MS/MS. At the same dose, after 6 h 2i induced 78% decrease of
the cell migration rate, higher than that observed with the prototype 1 (500 µM, 48%) after 8 h
treatment13 (Figure 5B). These findings suggest higher potency of 2i with respect to 1 in a cellular
context. Moreover, the ability of 2i to affect cell survival was also evaluated in the same cell line.
After 24 and 48 h of 2i treatment, the decrease of cell survival provided was 44% and 57%, respec-
tively, in comparison with their corresponding controls (DMSO) (Figure 5C), highlighting that the
observed effect on migration is not due to non-specific toxicity.
13

ACCEPTED MANUSCRIPT
C
140
120
100
80
60
***
40
***
20
0
DMSO
2i
DMSO
24h
2i
DMSO
48h
2i
0
Figure 5. (A) The AGPS inhibitor 2i lowered ether lipid levels in 231MFP breast cancer cells. The
lipidomic profiling data from 231MFP cells treated with vehicle (DMSO) or 2i (500 ꢀM) for 24 h
are shown. Data in bar graphs are the average ± SD. Lipid abbreviations are as follows: MAG, mon-
14

ACCEPTED MANUSCRIPT
oalkylglycerol; PA, phosphatidic acid; PI, phosphatidyl inositol; PC, phosphatidylcholine; PS, phos-
phatidylserine; LPC, lysophosphatidylcholine; LPE, lysophosphatidylethanolamine; LPA, lysophos-
phatidic acid. Lower case “e” indicates ether phospholipids; (B) effect of 2i on cell migration rate of
breast 231MFP cancer cells after 6 h. Data shown as average ± SD, n = 3−4/group. Significance is
represented as *p < 0.05 and as **p < 0.01 related to control groups; (C) Effects of 2i (500 µM) up-
on cell survival in 231MFP breast cancer cells after 24 and 48 h treatment. Significance is represent-
ed as ***p <0.001 in relation to control groups. Vehicle (DMSO) and 2i were preincubated with
cancer cells before seeding cells into serum-free migration or survival assays (Figure 5B and 5C).
Effects of 2i on proliferation and epithelial-mesenchymal transition (EMT) in PC-3 and MDA-
MB-231 cancer cells. Effect of 2i in non-malignant human mesothelial MeT5A cell prolifera-
tion
AGPS genetic knockdown in human glioma U87 and hepatic carcinoma HepG2 cells was shown to
decrease the expression of oncogenic signaling molecules and to increase the levels of adhesion
molecules such as E-cadherin, involved in the control of tumor invasiveness and differentiation.15
Thus, we investigated whether pharmacological inhibition of AGPS by 2i could phenocopy its ge-
netic knockdown by modulating EMT. The EMT process is characterized by loss of epithelial and
acquisition of mesenchymal features and plays a main role in the acquisition of invasive abilities
and chemo resistance in tumors.16 Prostate PC-3, breast MDA-MB-231 cancer cell lines and
MeT5A, a non-tumorigenic cell line used as a control, express detectable levels of AGPS (Figure
6A). A proliferation assay performed with the same cell lines showed negligible effects using 2i at
100 µM concentration, while a significant reduction of proliferation at 250 µM in PC-3 and mainly
in MDA-MB-231 (expressing higher levels of AGPS mRNA, Figure 6A) cells was recorded with 2i
(Figure 6B). In contrast, MeT5A cells were unaffected by the pharmacological treatment even at the
highest dose (Figure 6B), thus highlighting cancer specific effects by 2i treatment.
15

ACCEPTED MANUSCRIPT
B
A
Figure 6. (A) mRNA expression of AGPS in not-treated PC-3, MDA-MB-231 and MeT5A cells to
evaluate the basal levels. Quantitative RT–PCR was performed on total RNA. L32 levels were used
for normalization. Bars represent ± SEM of seven independent experiments.; (B) Dose−response
curves for antiproliferative activity of 2i, in a dose range (50, 100, 250 ꢀM) on PC-3 and MDA-MB-
231 cancer cells and on MeT5A, a non-tumorigenic cell line, after 24 h of treatment. The results rep-
resent the average ± SEM of three independent experiments (for a total of 8 measurements).
Significance is represented as *p<0.05 and **p < 0.01 related to control groups.
Thus, we treated PC-3 and MDA-MB-231 cells with 2i at 50 and 100 µM for 24 h, and we analyzed
by RT-PCR the levels of mRNA transcripts of E-cadherin, of its transcriptional regulator Snail, and
of matrix metalloproteinase-2 (MMP2), involved in the degradation of basement membrane compo-
nents and thus implicated in tumor invasion. Compound 1 was used for comparison, whereas 2s was
employed as negative control.
16

ACCEPTED MANUSCRIPT
17

ACCEPTED MANUSCRIPT
Figure 7. Effects of 2i (red), 1 (blue), and 2s (grey, negative control) at 50 and 100 ꢀM on PC-3 and
MDA-MB-231 cancer cells after 24 h treatment. Vehicle (DMSO, black) treated cells are used as
control. (A-B) E-cadherin transcript quantification via RT-PCR in PC3 and MDA-MB-231 cancer
cells; (C-D) Snail transcript quantification via RT-PCR in PC-3 and MDA-MB-231 cancer cells; (E-
F) MMP2 transcript quantification via RT-PCR in PC3 and MDA-MB-231 cancer cells. Each data is
the mean of at least four independent experiments. Significance is represented as *p < 0.05 and as
**p < 0.01 related to control groups.
The results depicted in Figure 7 show that 2i is able to increase the level of E-cadherin (Figure 7A-
B) as well as to reduce the levels of Snail (Figure 7C-D) and MMP2 (Figure 7E-F) in both the tested
cell lines. In the same assays, 2i displayed similar or higher effects than compound 1 on EMT path-
way, whereas 2s was totally inactive.
AGPS silencing in combination or not with 2i in MDA-MB-231 cells
To exclude any possible off target-dependent EMT modulation by 2i treatment, we determined the
effects on fold expression for AGPS, E-cadherin, Snail, and MMP2, in MDA-MB-231 cells treated
with siRNAs against AGPS alone or with a combination of siRNAs against AGPS and 2i (100 µM).
As shown in Figure 8, in all cases very similar results have been obtained, confirming that the ob-
served modulation of 2i on EMT is specific for AGPS inhibition.
18

ACCEPTED MANUSCRIPT
AGPS
ECAD
1.5
1.0
0.5
0.0
2.5
**
ns
*
***
2.0
***
1.5
1.0
0.5
0.0
2i
-
+
-
+
2i
-
+
-
+
siCTR
siAGPS
siCTR
siAGPS
SNAIL
MMP2
1.5
1.0
0.5
0.0
1.5
1.0
0.5
0.0
*
*
ns
ns
*
*
2i
-
+
-
+
2i
-
+
-
+
siCTR
siAGPS
siCTR
siAGPS
Figure 8. Effects of AGPS silencing either alone or in combination with 2i treatment (100 µM) in
MDA-MB-231 cancer cells. Color code: white, untreated; red, 2i treatment; grey, siAGPS; brown,
2i treatment plus siAGPS. Cells interfered for AGPS were treated after 24 h with 2i or DMSO (con-
trol) for 24 additional hours. AGPS, E-cadherin, Snail and MMP2 transcripts quantification has been
performed by RT-PCR. Each data is the mean of five independent experiments. Significance is rep-
resented as *p < 0.05, **p < 0.01 and ***p<0.001 related to control groups. ns: not significant.
CONCLUSION
Recently the FAD-dependent enzyme AGPS has been shown to be up-regulated across different
types of aggressive cancers such as human breast 231MFP, melanoma C8161, and prostate PC-3
cancers.3 AGPS knockdown led to an overall reduction of ether lipid levels, as well as oncogenic
signaling molecules, such as LPAe, PAFe and eicosanoids, and it showed lower tumor growth rates
in xenograft mice.3 The newly established small-molecule screening allowed us to identify and
19

ACCEPTED MANUSCRIPT
characterize the first AGPS inhibitor, compound 1, which recapitulates features of AGPS knock-
down. In this study, we described the design and development of some 1 analogs, compounds 2a-
l,q-u, designed to investigate structure-activity relationship and to define the pharmacophore ele-
ments of the series.
Among structural manipulations applied to the phenylpropionamide and/or the dihydrobenzimid-
azole moiety of 1 (Figure 1), only the 2,6-difluoro substitution at the phenylpropionamide chain
yielded a compound (2i) with improved AGPS binding and slightly better inhibitory activity. The
crystal structure of the AGPS/2i complex revealed that 2i retains the binding mode observed for 1,
with the second fluorine atom likely enhancing the hydrophobic interactions with the aliphatic resi-
dues surrounding the substrate tunnel. In 231MFP breast cancer cells, 2i reduced the ether phospho-
lipid levels by 50-75% and was more effective than 1 in decreasing cell migration. When tested in
PC-3 prostate and MDA-MB-231 breast cancer cells, to check putative modulation of EMT, 2i in-
creased the levels of E-cadherin and reduced those of Snail, and of MMP2, involved in invasiveness
and cancer aggressiveness. Interestingly, these effects were not dependent on reduced proliferation,
which was observed only at higher concentration (250 µM) in MDA-MB-231 and to a lesser extend
in PC-3 cells. In addition, when tested in non-cancer MeT5A cells compound 2i displayed negligible
effects on cell proliferation, thus revealing cancer specific activity. In the same assays, 1 was often
less effective and 2s, unable to bind and inhibit AGPS, was totally inactive. Genetic knockdown of
AGPS by siRNAs in MDA-MB-231 cells produced the same effects as 2i treatment, and also their
combination treatment furnished comparable increase of E-cadherin and reduction of Snail and
MMP2 levels, ruling out the possibility of off-target effects by 2i. Further studies will be conducted
to explore the effects of 2i in a panel of cancer cell lines, with particular emphasis on its effects on
migration and invasiveness.
20

ACCEPTED MANUSCRIPT
EXPERIMENTAL SECTION
Chemistry. Melting points were determined on a Buchi 530 melting point apparatus and are uncor-
1
13
rected. H NMR and C NMR spectra were recorded at 400 and 100 MHz, respectively, on a
Bruker AC 400 spectrometer; chemical shifts are reported in δ (ppm) units relative to the internal
reference tetramethylsilane (Me4Si). EIMS spectra of all compounds were recorded with a Fisons
Trio 1000 spectrometer; only molecular ions (M+) and base peaks are given. HR-MS spectra of final
compounds were recorded with an ESI Orbitrap spectrometer (capillary temperature:275°C, spray
voltage 4,0 kV (ESI +) and 3,8 kV (ESI +), sheath gas: 5, tube lens voltage +/- 90 V). The mass
spectrometry samples were solubilized in MeOH + 0,01% HCOOH (10-5M). All compounds were
routinely checked by TLC and 1H NMR. TLC was performed on aluminum-backed silica gel plates
(Merck DC, Alufolien Kieselgel 60 F254) with spots visualized by UV light. All solvents were rea-
gent grade and, when necessary, were purified and dried by standard methods. Concentration of so-
lutions after reactions and extractions involved the use of a rotary evaporator operating at reduced
pressure of ca. 20 Torr. Organic solutions were dried over anhydrous sodium sulfate. Elemental
analysis has been used to determine the purity of the described compounds, that is >95%. Analytical
results are within ±0.40% of the theoretical values. All chemicals were purchased from Sigma-
Aldrich, Milan (Italy), or from Alfa Aesar, Karlsruhe (Germany), and were of the highest purity. All
the synthetic procedures for the syntheses of 3-12 as well as the chemical and physical data for 2 are
reported in SI.
Procedure for the Synthesis of Compounds 3a-d, 3f-l, 4a-d, 4f-l, 5a-l, 6-12
Synthesis of Ethyl 3-(2-Fluorophenyl)-2-methylacrylate (3l).
Triethyl-2-phosphonopropionate (4.83 mmol, 1.2 eq, 1.04 mL) and 2-fluoro benzaldehyde (4.03
mmol, 1.0 eq, 0.42 mL) were added in sequence to a suspension of anhydrous K2CO3 (8.06 mmol, 2
eq, 1.11 g) in dry EtOH (10 mL). The reaction mixture was kept stirring at 80 °C for 7 h. After this
time the reaction was quenched with water (10 mL), concentrated in vacuo and the residue was ex-
21

ACCEPTED MANUSCRIPT
tracted with ethyl acetate (3 × 10 mL). The organic layer was dried over Na2SO4, filtrated and con-
centrated in vacuo. The crude has been purified by column chromatography on silica gel eluting
with n-hexane: EtOAc 50:1 to get the compound 3l as a light yellow oil as a mixture of the two iso-
1
mers cis:trans=1:2; H-NMR (CDCl3, 400 MHz, δ; ppm): δH 1.11 (3H, t, J= 7.2 Hz, COOCH2CH3
CIS), 1.38 (6H, J= 7.2 Hz, 2 × (3H), t, COOCH2CH3 TRANS), 2.06 (6H, 2 × (3H), s, C(2)CH3
TRANS), 2.15 (3H, s, C(2)CH3 CIS), 4.12 (2H, q, J= 6.8 Hz, COOCH2CH3 CIS), 4.30 (4H, 2 ×
(2H), q, J= 7.2 Hz, COOCH2CH3 TRANS), 6.75 (1H, s, C(3)H CIS), 7.04-7.39 (12 H, m, aromatic
protons CIS e TRANS), 7.72 (2H, s, C(3)H TRANS) ppm. 13C NMR (CDCl3, 100 MHz, δ; ppm) δC:
13.4, 14.7, 60.5, 115.8, 121.9, 124.9, 130.6, 131.0, 132.0, 132.9, 158.6, 169.2 ppm. MS (EI) m/z
[M]+: 208.08.
General Procedure for the Synthesis of Ester Intermediates 3a-d, 3f-l.
Trimethyl phosphonacetate (5.21 mmol) was added dropwise to a suspension of NaH (5.21 mmol)
in dry THF (8 mL) at 0 °C under inert atmosphere (N2). After 30 minutes, the appropriate ketone
(4.17 mmol) was added to the reaction mixture. The reaction mixture was let to warm slowly to
room temperature. The system was kept stirring at room temperature overnight. The next morning
the reaction was quenched by addition 1M NH4Cl (10 mL) and the product was extracted with Et2O
(3 × 10 mL). The combined organic layers were dried over Na2SO4, filtered, and concentrated in
vacuo. The crude was purified by silica gel chromatography eluting with n-hexane: EtOAc, 50 :1 to
give the pure 3a-d, 3f-l.
Example: Methyl-3-(2-Chlorophenyl)but-2-enoate (3d).
1H NMR (CDCl3, 400 MHz, δ, ppm): δH 2.19 (3H, d, J= 1.2 Hz, C(4)H3 TRANS), 2.52 (3H (2), d,
J= 1.2 Hz, C(4)H3 CIS), 3.57 (3H, s, COOCH3 TRANS ), 3.78 (3H (2), s, COOCH3 CIS), 5.86 (1H
(0.6), d, J= 1.6 Hz, C(2)H CIS) 6.04 (1H, d, J= 1.6 Hz, C(2)H TRANS), 7.09 (1H, dd J= 2.4 Hz, 6.1
Hz, aromatic proton), 7.18-7.20 (1H, m, aromatic proton), 7.24-7.31 (4H, m, aromatic protons
CIS+TRANS), 7.39-7.43 (2H, m, aromatic protons CIS+TRANS) ppm. 13C NMR (CDCl3, 100
22

ACCEPTED MANUSCRIPT
MHz, δ; ppm) δC: 20.7, 51.9, 121.8, 125.3, 129.2, 129.9, 130.7, 133.2, 139.5, 148.6, 166.7 ppm. MS
(EI) m/z [M]+ 210.05.
General Procedure for the Synthesis of Arylpropionic Esters Intermediates 4a-d, 4f-l.
A 0.05 M solution of esters 3a-d, 3f-l (1.31 mmol) in methanol has been reduced under flow condi-
tions (1mL/min) on Pd/C10% catalyst in presence of an H2 pressure of 10 bar by using H-Cube®
(Thales Nano) to give the products in MeOH. The solvent has been evacuated to give the pure prod-
ucts 4a-d, 4f-l.
Example: Methyl 3-(2,4-Difluorophenyl)butanoate (4h).
1H-NMR (CDCl3, 400 MHz, δ; ppm): δH 1.30 (3H, d, J= 6.8 Hz, C(3)CH3), 2.54-2.60 (2H, dd, J=
15.4 Hz, 7.6 Hz, C(2)H), 2.63-2.69 (2H, dd, J= 15.4 Hz, 7.6 Hz, C(2)H’), 3.47-3.56 (1H, m, C(2)H),
3.63 (1H, s, COOCH3), 6.74-6.84 (2H, m, C(6)H, C(8)H), 7.14-7.20 (1H, m, C(7)H) ppm. 13C NMR
(CDCl3, 100 MHz, δ; ppm) δC: 20.2, 33.0, 38.9, 51.6, 103.6, 111.6, 129.3, 131.1, 153.9, 163.3,
172.1 ppm. MS (EI) m/z [M]+: 214.09.
General Procedure for the Synthesis of Arylpropionic Acids Intermediates 5a-l.
The esters 4a-l (1.04 mmol) was solubilized in EtOH (2 mL). The system was cooled to 0 °C and a
2M aqueous solution of KOH (4.16 mmol) was added dropwise. The reaction mixture was allowed
to warm to room temperature and kept stirring for one hour. After this time the ethanol was removed
in vacuo. The residual aqueous solution was cooled to 0 °C and titrated with 1M HCl till pH = 1.
The product was extracted with ethyl acetate (3 × 10 mL). The combined organic layers were
washed with brine (10 mL) and then dried over Na2SO4, filtered and concentrated in vacuo to give
the pure compounds 5a-l.
Example: 3-(2-Fluorophenyl)pentanoic Acid (5j).
1H NMR (d6-DMSO, 400 MHz, δ, ppm): δH 0.73 (3H, t, J= 7.6 Hz, C(5)H3), 1.48-1.59 (1H, m,
C(4)H), 1.62-1.72 (1H, m, C(4)H’), 2.51-2.55 (1H, dd, J= 6.8 Hz, 15.8 Hz, C(2)H), 2.57-2.66 (1H,
dd, J= 6.8 Hz, 15.8 Hz, C(2)H’), 3.13-3.28 (1H, m, C(3)H), 7.09-7.16 (2H, m, C(8)H, C(10)H),
23

ACCEPTED MANUSCRIPT
7.21-7.27 (1H, m, C(9)H), 7.30-7.34 (1H, m, C(11)H), 12.11 (1H, bs, COOH) ppm. MS (EI) m/z
[M]+: 196.08.
Synthesis of 1,3-Dimethyl-2-oxo-2,3-dihydro-1H-benzo[d]imidazo-5-carbonitrile (11).
2-Oxo-2,3-dihydro-1H-benzo[d]imidazole-5-carbonitrile (1.88 mmol, 1.0 eq, 0.3 g) was solubilized
in dry THF (10 mL). The solution was cooled to 0 °C and NaH (4.52 mmol, 2.4 eq, 0.18 g) was
added. The reaction mixture was stirred at 0 °C for 20 min. After this time methyl iodide (4.52
mmol, 2.4 eq, 0.64 g) was added dropwise and the system was allowed to warm to room tempera-
ture and stirred overnight at room temperature. The next morning the reaction was quenched with
water (8 mL). The product was extracted with ethyl acetate (3 × 10 mL). The combined organic lay-
ers were washed with a saturated aqueous solution of NaCl (2 × 5 mL), dried over Na2SO4 and fil-
tered. The extract was then purified by silica gel chromatography eluting with hexane: ethyl acetate
1
1: 5 to give the pure compound 11; Yield: 85%, mp: 204-206 °C (ethanol); H-NMR (d6-DMSO,
400 MHz, δ; ppm) δH: 3.36 (3H, s, -N(1)CH3), 3.37 (3H, s, -N(2)CH3), 7.33 (1H, d, J= 8.4 Hz
C(3)H), 7.55 (1H, dd, J= 1.2 Hz, 8 Hz, C(4)H), 7.69 (1H, d, J= 1.2 Hz, C(6)H) ppm. 13C NMR (d6-
DMSO, 100 MHz, δ; ppm) δC: 26.8, 27.5, 107.7, 111.3, 113.1, 118.3, 128.2, 132.7, 133.5, 155.5
ppm. MS (EI) m/z [M]+: 187.09.
Synthesis of 3-Oxo-1,2,3,4-tetrahydroquinoxaline-6-carbonitrile (12).
Ethyl (4-cyano-2-nitrophenyl)glycinate (1.00 mmol, 1.0 eq, 0.25 g) was dissolved in EtOH (3 mL)
and the solution was cooled to 0 °C and was treated with SnCl2 x 2 H2O (5.01 mmol, 5.0 eq, 1.13 g)
and 37 % HCl (0.83 mL). The reaction mixture was stirred at 0 °C for 10 min, then let to warm to
room temperature and left under stirring overnight. The next morning the solvent was removed in
vacuo. The residue was suspended in 20 mL of ethyl acetate and a saturated aqueous solution of
Na2CO3 was added dropwise till pH=8. The product was extracted with ethyl acetate (3 × 10 mL).
The combined organic layers were washed with a saturated aqueous solution of NaCl (2 × 5 mL),
dried over Na2SO4, filtered and concentrated in vacuo. The obtained solid was triturated in petrole-
24

ACCEPTED MANUSCRIPT
um ether, filtered and dried to give 12 as pure compound; Yield: 61 %, mp: 190-192°C (acetoni-
trile); 1H-NMR (d6-DMSO, 400 MHz, δ; ppm): δH 3.91 (2H, d, J= 1.2 Hz, C(2)H2), 6.69 (1H, d, J=
8 Hz, C(9)H), 6.94 (1H, bs, -NHCH2CONH-), 6.95 (1H, d, J= 1.6 Hz, C(6)H), 7.17 (1H, dd, J= 1.6
Hz, 8 Hz, C(8)H), 10.51 (1H, s, -NHCH2CONH-) ppm. 13C NMR (d6-DMSO, 100 MHz, δ; ppm) δC:
42.6, 105.8, 115.9, 117.9, 118.4, 125.8, 130.5, 138.5, 163.3 ppm. MS (EI) m/z [M]+: 173.05.
Synthesis of 5-Amino-1,3-dihydro-2H-benzo[d]imidazol-2-one (7).
5-Nitro-1,3-dihydro-2H-benzo[d]imidazol-2-one (0.837 mmol, 1 eq, 0.15 g) was dissolved in etha-
nol (5 mL). SnCl2 × 2 H2O (4.187 mmol, 5 eq, 0.94 g) was added to the solution and the system was
cooled to 0 °C. Afterwards, 37 % HCl (0.5 mL) was added dropwise. The reaction mixture was
stirred at 0 °C for 10 min, then it was let warm to room temperature and then was heated to 75 °C.
The reaction mixture was stirred at this temperature for 3 h and then left overnight (10 h) at room
temperature. The next morning the reaction was cooled to 0 °C and quenched by dropwise addition
of a saturated aqueous solution of NaHCO3 till pH 7-8. The product was extracted with ethyl acetate
(3 × 10 mL). The combined organic layers were washed with a saturated aqueous solution of NaCl
(2 × 5 mL) and then dried over Na2SO4, filtered and concentrated in vacuo. The extract was purified
by silica gel chromatography eluting with chloroform: methanol 8:1, to afford the amine 7; -
1
benzo[d]imidazol-2-one (0.365 mmol, 54.5 mg, Y 44 %). m.p. 253-256 °C (methanol). H-NMR
(d6-DMSO, 400 MHz, δ; ppm) δH: 6.21-6.24 (1H, dd, J= 2 Hz, 8.2 Hz, C(4)H), 6.29 (1H, d, J= 2
13
Hz, C(2)H), 6.63 (1H, d, J= 8 Hz, C(5)H), 10.08 (1H, bs, N(1)H), 10.23 (1H, bs, N(2)H) ppm. C
NMR (d6-DMSO, 100 MHz, δ; ppm) δC: 95.4, 106.7, 108.6, 119.9, 130.2, 143.1, 156.4 ppm. MS
(EI) m/z [M]+:149.07.
General Procedure for the Synthesis of Amines 6, 8-10.
A 0.05 M solution of carbonitrile derivatives (3.49 mmol) in glacial acetic acid has been let flow
(1.5 mL/min) on Pd/C 10% catalyst in presence of an H2 pressure of 50 bar by using H-Cube® (Tha-
les Nano). The acetic acid was removed under reduced pressure to give the corresponding acetate
25

ACCEPTED MANUSCRIPT
salt of the compound. The crude residue was suspended in dry THF and was treated with 10 eq. of
HCl (2N in THF). The reaction mixture was stirred for two hours and then the suspension was fil-
trated giving a solid which was rinsed with petroleum ether to give the pure compound as a colorless
solid.
Example: (1H-benzo[d]imidazol-5-yl)methanamine hydrochloride (8).
1H-NMR (d6-DMSO, 400 MHz, δ; ppm) δH: 4.21 (2H, d, J= 5.6 Hz, C(10)H2), 7.65 (1H, d, J= 8.4
Hz, C(4)H), 7.86 (1H, d, J= 8.4 Hz, C(3)H), 8.02 (1H, s, C(6)H), 8.56-8.60 (3H, m, CH2NH2 and
13
C(8)H), 9.42 (1H, s, benzimidazole -NH-) ppm. C NMR (d6-DMSO, 100 MHz, δ; ppm) δC: 45.9,
114.2, 117.7, 124.8, 136.8, 137.1, 138.0, 143.8 ppm. MS (EI) m/z [M]+: 183.08.
General Procedure for the Amides 2a-l,q-u. The appropriate carboxylic acid 5 (0.4 mmol) was
dissolved in dry DMF (3 mL), then 1,1′-carbonyldiimidazole (0.44 mmol) was added and the mix-
ture was stirred at room temperature for 40 min. After this time triethylamine (1.2 mmol) and the
proper amine among 6-10 (0.6 mmol) was added, and the system was left under stirring at room
temperature overnight. The next morning the reaction was judged to be complete by TLC analysis.
The reaction was quenched by addition of 10 mL of brine. The product was then extracted with
ethyl acetate (3 × 10 mL). The combined organic layers were washed with a saturated aqueous solu-
tion of Na2CO3 (2 × 5 mL), a 2N HCl aqueous solution (2 × 5 mL) and saturated aqueous solution of
NaCl (2 × 5 mL) and then dried over Na2SO4, filtered and concentrated in vacuo. The obtained
white solid was triturated in diethyl ether, filtered and dried to give the pure 2.
Chemical and Physical Data, 1H NMR, 13C NMR, MS (ESI), and HR-MS (ESI) Data for Com-
pounds 2a-l,q-u.
N-((2-Oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)methyl)-3-phenylbutanamide (2a).
1H NMR (d6-DMSO, 400 MHz, δ; ppm) δH: 1.18 (3H, d, J= 6.8 Hz, C(3)CH3), 3.01-3.04 (2H, m,
C(2)H2), 3.17-3.22 (1H, m, C(3)H), 4.18 (2H, d, J= 5.6 Hz, C(10)H2), 6.67 (1H, d, J= 8.8 Hz,
C(18)H), 6.79 (2H, d, J= 8 Hz, C(5)H, C(9)H), 7.16-7.30 (5H, m, C(6)H, C(7)H, C(8)H, C(12)H,
26

ACCEPTED MANUSCRIPT
C(16)H), 8.27 (1H, t, J= 5.6 Hz, -NHCH2), 10.54 (1H, s, -NHCONH-), 10.56 (1H, s, -NHCONH-)
ppm. 13C NMR (d6-DMSO, 100 MHz, δ; ppm) δC: 22.1, 36.6, 42.6, 44.4, 108.2, 108.5, 120.7, 126.5,
127.2 (2C), 128.8 (2C), 128.9, 130.2, 132.4, 146.9, 155.9, 171.1 ppm. MS (EI) m/z [M]+: 309.16.
HR-MS (ESI) m/z [M+H]+ calculated: 310.1550, found 310.1553.
3-(3-Fluorophenyl)-N-((2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)methyl)butanamide (2b).
1H NMR (d6-DMSO, 400 MHz, δ; ppm) δH: 1.19 (3H, d, J= 7.2 Hz, C(3)CH3), 2.33-2.45 (2H, m,
C(2)H2), 3.19-3.24 (1H, m, C(3)H), 4.17 (2H, d, J= 6 Hz, C(10)H2), 6.65 (1H, d, J= 8.8 Hz,
C(18)H), 6.79 (2H, d, J= 8 Hz, C(5)H, C(9)H), 6.98-7.08 (3H, m, C(8)H, C(12)H, C(19)H), 7.29-
7.35 (1H, m, C(7)H), 8.24 (1H, t, J= 6 Hz, -NHCH2), 10.51 (1H, s, -NHCONH-), 10.53 (1H, s, -
NHCONH-) ppm. 13C NMR (d6-DMSO, 100 MHz, δ; ppm) δC: 22.0, 36.5, 42.6, 44.0, 108.2, 108.4,
113.2, 113.9, 120.0, 123.4, 129.0, 130.2, 130.5, 132.3, 149.8, 155.9, 162.7, 170.8 ppm. MS (EI) m/z
[M]+: 327.12. HR-MS (ESI) m/z [M+Na]+ calculated: 350.1275, found: 350.1281.
3-(4-Fluorophenyl)-N-((2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)methyl)butanamide (2c).
1H NMR (d6-DMSO, 400 MHz, δ; ppm) δH: 1.18 (3H, d, J= 6.8 Hz, C(3)CH3), 2.32-2.41 (2H, m,
C(2)H2), 3.18-3.24 (1H, m, C(3)H), 4.17 (2H, m, C(10)H2), 6.63 (1H, d, J= 9 Hz, C(18)H), 6.78 (2H
d, J= 7.6 Hz, C(5)H, C(9)H), 7.09 (2H, t, J= 8.8 Hz, C(12)H, C(19)H), 7.23-7.27 (2H, m, C(6)H,
C(8)H), 8.23 (1H, t, J= 6 Hz, -NHCH2), 10.51 (1H, s, -NHCONH-), 10.54 (1H, s, -NHCONH-)
ppm. 13C NMR (d6-DMSO, 100 MHz, δ; ppm) δC: 22.2, 36.0, 42.6, 44.5, 108.2, 108.4, 115.2, 115.4,
120.0, 128.9 (2C), 129.0, 130.2, 132.3, 142.8, 155.9, 161.1, 170.9 ppm. MS (EI) m/z [M]+: 327.16.
HR-MS (ESI) m/z [M+H]+ calculated: 328.1456, found: 328.1467.
3-(2-Chlorophenyl)-N-((2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)methyl)butanamide (2d).
1H NMR (d6-DMSO, 400 MHz, δ; ppm) δH: 1.16 (3H, d, J= 4 Hz, C(3)CH3), 2.43 (2H, m, C(2)H2),
3.62 (1H, m, C(3)H), 4.20 (2H, bs, C(10)H2), 6.74 (1H, bs, C(18)H), 6.80 (2H, bs, C(9)H, C(19)H),
7.14 (1H, bs, C(6)H), 7.35 (2H, bs, C(7)H, C(8)H), 7.58 (1H, d, J= 8 Hz, C(12)H), 8.33 (1H, bs, -
13
NHCH2-), 10.55 (2H, bs, -NHCONH-) ppm. C NMR (d6-DMSO, 100 MHz, δ; ppm) δC: 21.1,
27

ACCEPTED MANUSCRIPT
35.5, 42.4, 42.7, 108.2, 108.5, 120.1, 124.1, 127.9, 128.5, 128.5, 129.0, 130.2, 132.3, 133.1, 145.4,
155.9, 170.7 ppm. MS (EI) m/z [M]+: 343.10. HR-MS (ESI) m/z [M+H]+ calculated: 344.1160,
found: 344.1175.
3-(2-Bromophenyl)-N-((2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)methyl)butanamide (2e).
1H NMR (MeOD, 400 MHz, δ; ppm) δH: 1.23 (3H, d, J= 4 Hz, C(3)CH3), 2.42-2.48 (1H, dd, J= 8
Hz, 16 Hz, C(2)H2), 2.56-2.61 (1H, dd, J= 8 Hz, 16 Hz, C(2)H2), 3.80 (1H, m, C(3)H), 4.29 (2H, m,
C(10)H2), 6.84 (1H, d, J= 8 Hz, C(18)H), 6.93 (2H, d, J= 8 Hz, C(9)H, C(19)H), 7.05 (1H, t, J= 8
13
Hz, C(6)H), 7.22-7.32 (2H, m, C(7)H, C(12)H), 7.52 (1H, d, J= 8 Hz, C(8)H) ppm. C NMR (d6-
DMSO, 100 MHz, δ; ppm) δC: 21.2, 35.5, 42.4, 42.7, 108.2, 108.5, 120.1, 124.1, 127.9, 128.5,
128.5, 129.0, 130.2, 132.3, 133.1, 145.4, 155.9, 170.7 ppm. MS (EI) m/z [M]+: 387.04. HR-MS
(ESI) m/z [M+H]+ calculated: 388.0665, 390.0637, found:388.0664, 390.0644.
N-((2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)methyl)-3-(o-tolyl)butanamide (2f).
1H NMR (d6-DMSO, 400 MHz, δ; ppm) δH: 1.14 (3H, d, J= 6.8 Hz, C(3)CH3), 2.31 (3H, s,
C(5)CH3), 2.34-2.39 (2H, m, C(2)H2), 3.40-3.47 (1H, m, C(3)H), 4.18 (2H, m, C(10)H2), 6.66 (1H,
d, J= 8.4 Hz, C(18)H), 6.79 (2H, d, J= 7.2 Hz, C(9)H, C(19)H), 7.05-7.21 (4H, m, C(6)H, C(7)H,
C(8)H, C(12)H), 8.27 (1H, t, J= 5.2 Hz, -NHCH2-), 10.52 (1H, s, -NHCONH-), 10.55 (1H, s, -
13
NHCONH-) ppm. C NMR (d6-DMSO, 100 MHz, δ; ppm) δC: 19.5, 21.6, 31.7, 42.7, 43.6, 108.2,
108.5, 120.0, 125.6, 126.1, 126.6, 128.9, 130.2, 130.5, 132.4, 135.3, 145.0, 155.9, 171.2 ppm. MS
(EI) m/z [M]+: 323.15. HR-MS (ESI) m/z [M+H]+ calculated: 324.1707, found: 324.1718.
3-(2-Methoxyphenyl)-N-((2-oxo-2,3-dihydro-1H-benzo[d]imidazol-5-yl)methyl)butanamide
(2g).
1H NMR (d6-DMSO, 400 MHz, δ; ppm) δH: 1.11 (3H, d, J= 6.4 Hz, C(3)CH3), 2.34 (1H, dd, J= 5.2
Hz, 13.6 Hz, C(2)H), 2.40 (1H, dd, J= 5.2 Hz, 13.6 Hz, C(2)H’), 3.54-3.57 (1H, m, C(3)H), 3.78
(3H, s, -OCH3), 4.15-4.26 (2H, m, C(10)H2), 6.75 (1H, d, J= 8 Hz, C(18)H), 6.79-6.82 (2H, m,
C(9)H, C(19)H), 6.89 (1H, t, J= 7.6 Hz, C(7)H), 6.94 (1H, d, J= 7.6 Hz, C(6)H), 7.14-7.19 (2H, m,
28