Exploring the readthrough of nonsense mutations by non-acidic Ataluren analogues selected by ligand-based virtual screening

Article abstract of DOI:10.1016/j.ejmech.2016.06.048

Ataluren, also known as PTC124, is a 5-(fluorophenyl)-1,2,4-oxadiazolyl-benzoic acid suggested to suppress nonsense mutations by readthrough of premature stop codons in the mRNA. Potential interaction of PTC124 with mRNA has been recently studied by molec

Full text of DOI:10.1016/j.ejmech.2016.06.048

European Journal of Medicinal Chemistry 122 (2016) 429e435
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Short communication
Exploring the readthrough of nonsense mutations by non-acidic
Ataluren analogues selected by ligand-based virtual screening
,
, b, **
Ivana Pibiri a, Laura Lentini a *, Marco Tutone a, Raffaella Melfi a, Andrea Pace a
,
,
c
Aldo Di Leonardo a
a
ꢀ
Dipartimento di Scienze e Tecnologie Biologiche, Chimiche e Farmaceutiche (STEBICEF), Universita degli Studi di Palermo, Viale delle Scienze Ed. 16-17,
90128 Palermo, Italy
b Istituto EuroMediterraneo di Scienza e Tecnologia (IEMEST), Via Emerico Amari 123, 90139 Palermo, Italy
c Centro di OncoBiologia Sperimentale (COBS), via San Lorenzo Colli, 90145 Palermo, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Ataluren, also known as PTC124, is a 5-(fluorophenyl)-1,2,4-oxadiazolyl-benzoic acid suggested to sup-
press nonsense mutations by readthrough of premature stop codons in the mRNA. Potential interaction
of PTC124 with mRNA has been recently studied by molecular dynamics simulations highlighting the
Received 23 March 2016
Received in revised form
9 June 2016
Accepted 27 June 2016
Available online 29 June 2016
importance of H-bonding and stacking
pꢀp interactions. A series of non-acidic analogues of PTC124
were selected from a large database via a ligand-based virtual screening approach. Eight of them were
synthesized and tested for their readthrough activity using the Fluc reporter harboring the UGA pre-
mature stop codon. The most active compound was further tested for suppression of the UGA nonsense
mutation in the bronchial epithelial IB3.1 cell line carrying the W1282X mutation in the CFTR gene.
© 2016 Elsevier Masson SAS. All rights reserved.
Keywords:
PTCs readthrough
CFTR gene
Nonsense mutation
Oxadiazoles
Cystic fibrosis
1. Introduction
required for normal function of the lung, pancreas, liver, and other
organs [5].
When a premature stop codon is present in the coding region of
the mRNA, protein translation is interrupted thus producing trun-
cated polypeptides. This situation is promptly detected by the
nonsense mediated mRNA decay (NMD) pathway. The NMD is a
surveillance mechanism that targets the cytoplasmic PTC-bearing
transcript for rapid degradation [1]. Nonsense mutations are the
cause of a significant percentage of most inherited diseases,
including cystic fibrosis (CF), Duchenne muscular dystrophy
(DMD), Usher’s Syndrome, and a variety of other genetic disorders
[2e4].
Gene therapy, that is potentially a method of choice to correct
the mutated gene, is far from clinical routine. Alternative phar-
macological approaches aim at modifying gene expression and
have been applied to diseases caused by a premature stop codon.
Indeed, the translational readthrough of a nonsense mutation
might allow the synthesis of a full-length functional protein [6,7]. A
well known category of drugs, the antibiotic aminoglycosides (e.g.
gentamicin, tobramycin, paromomycin, amikacin, etc.) possess the
ability to readthrough stop codons by disturbing the translation
machinery and leading to the insertion of a near-cognate amino
acid at a PTC [8e11]. However, aminoglycoside readthrough action
lacks specificity thus resulting in readthrough of many correctly
positioned stop codons, and originating toxic aggregates that after
long-term treatments can cause nephrotoxicity and ototoxicity
[3,12e14].
Due to the presence of premature stop codons in their CF trans-
membrane regulator gene, approximately 10% of CF patients lack
adequate levels of the CFTR protein, a chloride channel that is
3-[5-(2-Fuorophenyl)- [1,2,4]oxadiazol-3-yl]-benzoic acid (also
named PTC124 or Ataluren), was developed as a drug capable to
promote selectively the readthrough of premature termination
codons [6]. Recently, also the antiinflammatory drug amlexanox
has been claimed to promote readthrough of nonsense mutations
* Corresponding author.
** Corresponding author. Dipartimento di Scienze e Tecnologie Biologiche, Chi-
miche e Farmaceutiche (STEBICEF), Universita degli Studi di Palermo, Viale delle
Scienze Ed. 16-17, 90128 Palermo, Italy.
E-mail addresses: laura.lentini@unipa.it (L. Lentini), andrea.pace@unipa.it
(A. Pace).
ꢀ
http://dx.doi.org/10.1016/j.ejmech.2016.06.048
0223-5234/© 2016 Elsevier Masson SAS. All rights reserved.

430
I. Pibiri et al. / European Journal of Medicinal Chemistry 122 (2016) 429e435
[15,16]. Overall, the general concept underlying therapeutic
nonsense suppression is that a single drug, focused on a specific
genetic defect, may be beneficial to different diseases whose
common denominator is a nonsense mutation. PTC124 is not
structurally similar to aminoglycosides and its activity was first
assessed in HEK293 cells transfected with a luciferase gene
(LUC190) harboring a premature stop codon at Thr190, replacing
the normal ACA with UAA, UAG and UGA. Such ability was also
assessed using MDX mice (mice with Duchenne Muscular Dis-
trophy caused by nonsense mutation) showing 20e25% recovery of
full-length dystrophin [6].
In these past years PTC124 has been the subject of an intense
debate about its mechanism of action, because some studies evi-
denced that it has an effective readthrough activity [4] while a few
studies did not find enough evidence to prove the readthrough
activity of PTC124 [17e19]. Our previous results on readthrough
activity of PTC124 tested with a novel reporter vector harboring a
premature stop codon (TGA) in the H2B-GFP fused gene (H2B-GFP-
opal) and a molecular dynamics simulation on the hypothetical
interaction between PTC124 and a mRNA fragment supported the
hypothesis that PTC124 is able to promote the specific readthrough
of internal UGA premature stop codons [20].
Recently, we also reported the synthesis of a set of variously
fluorinated PTC124 analogues and their biological screening in the
lower airway cell line (IB3.1) revealed three analogues showing
comparable or higher activity than PTC124 as readthrough pro-
moters [21].
However, despite recent progress on the topic, the precise bio-
logical site targeted by PTC124 is still unknown. This makes
impossible to perform docking studies to suggest convenient mod-
ifications of the drug in order to improve its activity even against
those stop codons for which PTC124 has shown a reduced effect.
In this context, based on biological data already available on
some PTC124 analogues we decided to perform a Ligand Based
Virtual Screening in order to identify promising PTC124-like can-
didates eventually providing optimized alternatives for read-
through drug development. Selected candidates were synthesized
and tested for their readthrough activity of premature termination
codons by using FLuc assay and IB3 cell lines.
2.2. Chemistry
All solvents and reagents were obtained from commercial
sources. All synthesized compounds were purified by chromatog-
raphy and analyzed by IR, HRMS, and NMR. Purity of synthesized
compounds was verified prior to biological tests by chromato-
graphic analyses and NMR (see supplementary material) and in all
the cases purity was higher than 95%. IR spectra have been regis-
tered (in Nujol) with a Shimadzu FTIR-8300 spectrophotometer;
melting points have been determined on a Reichart-Thermovar
hotstage Kofler and are uncorrected. NMR spectra have been
registered on a Bruker AVANCE DMX 300 using CDCl3 and DMSO as
solvent. HRMS spectra were recorded by analyzing a 50 ppm so-
lution of the compound in a 6540 UHD Accurate-Mass Q-TOF LC/MS
(Agilent Technologies) equipped with a Dual AJS ESI source. GC-MS
spectra have been registered by using either an Agilent 7890B/
7000C GC-MS-TQ or a GC-MS Shimadzu QP-2010 Instrument. Flash
chromatography was performed by using silica gel (Merck,
0.040e0.063 mm) and mixtures of ethyl acetate and petroleum
ether (fraction boiling in the range of 40e60 ꢁC) in various ratios. 3-
Methyl-benzamidoxime
[31],
2-picolin-amidoxime
[32],
isonicotin-amidoxime [33], nicotin-amidoxime [33], and benza-
midoxime [34] were synthesized as reported. Generally, an acqu-
eous solution of hydroxylamine was prepared by mixing
NH2OH*HCl (36 mmol) and NaOH (36 mmol) in water (20 mL). The
hydroxylamine solution was then added to an alcoholic solution of
the corresponding nitrile (30 mmol) dissolved in ethanol (100 mL)
in a 250 mL round bottomed flask. The mixture was refluxed for 8 h.
The solvent was then removed under vacuum and 100 mL water
were added to the residue. The amidoxime was filtered as a white
solid and re-crystallized from ethanol.
2.2.1. General procedure for the synthesis of 1,2,4-oxadiazoles
The synthesis of 1,2,4-oxadiazoles has been performed by the
amidoxime route [35]. The appropriate amidoxime (0.3 g) was
dissolved in 50 mL of toluene in a 250 mL round bottomed flask.
Then,1.2 eq. of the appropriate aroyl chloride and 1.2 eq. of pyridine
were added and the reaction mixture was refluxed for 6e8 h
monitoring the reaction by TLC until consumption of starting ma-
terial. The solvent was removed under vacuum and water was
added to the residue. Extraction with ethyl acetate and chromato-
graphic separation on silica gel using mixtures of petroleum ether
and ethyl acetate as eluent allowed to obtain the desired oxadia-
zole, further purified by crystallization.
2. Materials and methods
2.1. Ligand based virtual screening
Twenty PTC-124 analogs [22,23] were chosen to create the
starting dataset for the modeling analysis. The dataset structures
were processed with the LigPrep [24] software package in order to
assign the appropriate protonation states at physiological
3-(2′-pyridyl)-5-(3′-cyanophenyl)-1,2,4-oxadiazole (NV1859).
(0.49 g; 89% Yield). White solid, m.p. 148e149 ꢁC from petroleum
ether (lit. [36].148e149 ꢁC). 1H NMR (300 MHz, CDCl3):
d (ppm)
8.86 (dd, 1H, J1 ¼ 10.0 Hz, J2 ¼ 4.8 Hz), 8.60 (brs, 1H), 8.52 (1H, d,
J ¼ 8.0 Hz), 8.23 (1H, d, J ¼ 8.0 Hz), 7.90 (2H, m), 7.73 (1H, t,
J ¼ 7.8 Hz), 7.49 (1H, dd, J1 ¼ 8.0 Hz, J2 ¼ 4.8 Hz); 13C NMR (75 MHz,
pH(7.2
0.2), employing the Ionizer option. Conformers were
generated through Macro-Model torsional sampling using the
OPLS_2005 force field as reported in a previous paper [25]. The
pharmacophore modeling study was performed using the Phase
software. Phase is a versatile product for pharmacophore percep-
tion, structural alignment, activity prediction, and 3D database
creation and searching [26]. After the ligands preparation, the
pharmacophore model was developed by using a set of pharma-
cophore features to generate sites for all of the compounds. A
standard set of six pharmacophore features were used: hydrogen-
bond acceptor (A), hydrogen-bond donor (D), hydrophobic group
(H), negatively ionizable (N), positively ionizable (P), and aromatic
ring (R); Hypotheses were generated by using a previously vali-
dated protocol [27,28]. Virtual high-throughput screening was
performed on ZINC “drug-like” database [29], consisting of about
2*106 compounds and filtered according to the Lipinski’s rule of
Five [30].
CDCl3): d (ppm) 174.38,169.01,150.55, 145.93, 137.19,135.88, 132.07,
131.75, 130.22, 125.82, 125.30, 123.36, 117.34, 113.88; HRMS for
C
14H8N4O found 249.0782 [MþH]þ (Calcd. 249.0771).
3-(4′-pyridyl)-5-(3′-toluyl)-1,2,4-oxadiazole(NV1861). (0.32 g;
62% Yield). White solid, m.p. 109e110 ꢁC from petroleum ether (lit.
[37]. 111e112 ꢁC). 1H NMR (300 MHz, CDCl3):
J ¼ 5.5 Hz), 8.00e7.92 (m, 4H), 7.42 (d, 2H, J ¼ 5.5 Hz), 2.45 (s, 3H);
13C NMR (75 MHz, CDCl3):
(ppm) 177.40, 168.00, 151.09, 139.87,
d (ppm) 8.78 (d, 2H,
d
135.38, 134.65, 129.80, 129.38, 126.06, 124.37, 122.09, 21.99; HRMS
for C14H11N3O found 238.0989 [MþH]þ (Calcd. 238.0975).
3-(3′-pyridyl)-5-(3′-toluyl)-1,2,4-oxadiazole
(NV1879).
(0.46 g; 88% Yield)White solid, m.p. 101e102 ꢁC from petroleum
ether. 1H NMR (300 MHz, CDCl3):
8.78 (dd, 1H, J1 ¼ 5.0 Hz, J2 ¼ 1.8 Hz), 8.46 (dd, 1H, J1 ¼ 7.9 Hz,
d
(ppm) 9.42 (d, 1H, J ¼ 1.8 Hz),
J2 ¼ 1.8 Hz), 8.05 (m, 2H), 7.47 (m, 3H), 2.49 (s, 3H); 13C NMR

I. Pibiri et al. / European Journal of Medicinal Chemistry 122 (2016) 429e435
431
Ar
Ar'COCl
toluene/pyridine
NH
N
Ar
N
Ar'
O
NHOH
II
NV1859: Ar = 2-pyridyl; Ar' = 3-cyano-phenyl
NV1861: Ar = 4-pyridyl; Ar' = 3-toluyl
NV1879: Ar = 3-pyridyl; Ar' = 3-toluyl
NH2OH
Ar
NV1883: Ar = phenyl;
NV1894: Ar = 3-toluyl; Ar' = 2-toluyl
NV1898: Ar = 2-pyridyl; Ar' = 3-toluyl
NV1940: Ar = 2-pyridyl; Ar' = 3-methoxy-phenyl
NV1919: Ar = 4-pyridyl; Ar' = 3-methoxy-phenyl
Ar' = 3-toluyl
C
N
I
Scheme 1. Amidoxime route toward PTC124 analogues.
3-(2-pyridyl)-5-(3′-toluyl)-1,2,4-oxadiazole (NV1898). (0.47 g;
91% Yield) White solid, m.p. 127e129 ꢁC from petroleum ether. 1H
NMR (300 MHz, CDCl3):
J ¼ 8.1 Hz), 8.22 (s, 1H), 8.12 (m, 1H), 7.89 (m, 1H), 7.59 (m, 3H), 2.47
(s, 3H); 13C NMR (75 MHz, CDCl3):
(ppm) 176.65, 168.71, 150.39,
d
(ppm) 8.86 (d, 1H, J ¼ 4.4 Hz), 8.24 (d, 1H,
Fig. 1. Best pharmacophore hypothesis AARR.5 aligned with PTC-124.
d
146.53, 139.01, 137.00, 133.71, 128.96, 128.87, 125.40, 125.45, 123.89,
123.21,21.27; HRMS for C14H11N3O found 238.0952 [MþH]þ (Calcd.
238.0975).
(75 MHz, CDCl3):
d (ppm) 176.37, 166.96, 151.89, 148.66, 139.13,
134.79, 133.82, 129.07, 128.70, 125.38, 123.86, 123.63, 123.39, 21.37;
HRMS for C14H11N3O found 238.0986 [MþH]þ (Calcd. 238.0975).
3-(3′-phenyl)-5-(3′-toluyl)-1,2,4-oxadiazole(NV1883). (0.44 g;
84% Yield)White solid, m.p. 70e71 ꢁC from petroleum ether (lit.
3-(2-pyridyl)-5-(3′-anisoyl)-1,2,4-oxadiazole
(0.47 g; 84% Yield)White solid, m.p. 100e102 ꢁC from petroleum
ether. 1H NMR (300 MHz, CDCl3):
(NV1940).
[38]. 74-5 ꢁC). 1H NMR (300 MHz, CDCl3):
d
(ppm) 8.21e8.17, (m,
2H), 8.06e8.00 (m, 2H), 7.54e7.49 (m, 3H), 7.43e7.38 (m, 2H), 2.49
(s, 3H); 13C NMR (75 MHz, CDCl3):
(ppm) 175.92, 168.94, 150.72,
d
(ppm) 8.86 (d, 1H, J ¼ 4.4 Hz),
8.24 (d, 1H, J ¼ 7.9 Hz), 7.89 (m, 2H), 7.80 (brs, 1H), 7.47 (t, 2H,
J ¼ 7.9 Hz), 7.17 (dd, 1H, J1 ¼ 7.9 Hz, J2 ¼ 2.5 Hz), 3.92 (s, 3H); 13C
d
NMR (75 MHz, CDCl3): d (ppm) 177.17, 169.47, 160.73, 151.15, 147.20,
149.92, 139.02, 133.53, 131.13, 129.00, 128.83, 128.67, 127.54, 127.09,
125.33, 124.25, 21,39; HRMS for C15H14N2O found 237.1037 [MþH]þ
(Calcd. 237.1022).
137.77, 130.90, 126.21, 124.00, 121.49, 120.50, 113.32, 56.34, 30.36;
HRMS for C14H11N3O2 found 254.0921 [MþH]þ (Calcd. 254.0924).
3-(4′-pyridyl)-5-(3′-anisoyl)-1,2,4-oxadiazole
(NV1919).
3-(3-toluyl), 5-(2-toluyl)-1,2,4-oxadiazole (NV1894). (0.46 g;
92% Yield) White solid, m.p. 43e45 ꢁC from petroleum ether. 1H
(0.51 g; 91% Yield) White solid, m.p. 123e125 ꢁC from petroleum
ether. 1H NMR (300 MHz, CDCl3):
8.14 (d, 2H, J ¼ 7.5 Hz), 7.83 (brd, 1H, J ¼ 8.0 Hz), 7.73 (brs, 1H), 7.50
(1H, t, J ¼ 8.0 Hz), 7.19 (dd, 1H, J1 ¼ 8.0 Hz, J2 ¼ 2.5 Hz), 3.94 (s, 3H);
d
(ppm) 8.84 (d, 2H, J ¼ 7.5 Hz),
NMR (300 MHz, CDCl3):
d
(ppm) 8,18 (d, 1H, J ¼ 7.3 Hz), 7.99 (d, 2H,
J ¼ 7.9 Hz), 7.45 (m, 5H), 2.80 (s, 3H), 2.47 (s, 3H); 13C NMR (75 MHz,
CDCl3):
d (ppm) 176.98, 169.40, 139.80, 139.29,132.80, 132.50,
13C NMR (75 MHz, CDCl3):
d (ppm) 176.45, 167.40, 160.06, 150.50,
132.55, 130.90, 129.44, 128.76, 127.74, 126.93, 125.39, 124.27, 22.61,
22.00; HRMS for
251.1179).
C
16H14N2O found 251.1195 [MþH]þ (Calcd.
134.56, 130.35, 124.90, 121.38, 120.63, 119.57, 112.79, 55.56; HRMS
for C14H11N3O2 found 254.0907 [MþH]þ (Calcd. 254.0924).
COOH
CH3
N
N
N
N
O
N
N
O
O
F
CH3
NV1894
PTC-124
NV1883
H3C
N
N
N
N
N
N
N
O
N
N
O
O
NV1898
NV1861
NV1879
H3C
H3C
H3C
N
N
N
N
N
N
N
N
O
N
O
O
NV1919
NV1859
NV1940
NC
H3CO
H3CO
Fig. 2. PTC124 and selected analogues from the top 5% of virtual screening hits.

432
I. Pibiri et al. / European Journal of Medicinal Chemistry 122 (2016) 429e435
Fig. 3. Luciferase activity of HeLa FLuc190UGA transfected cells after treatment with PTC124 and its analogues compared to that of untreated (opal untr) and untransfected (Fluc WT)
cells.
and IB3.1 cells were cultured in DMEM supplemented with FBS 10%
(GIBCO) in a humidified atmosphere of 5% CO2 in air at 37 ꢁC.
2.3.2. Measurement of luciferase activity by luminescence
HeLa cells plated in a 6 well plate at a density of 2 ꢂ 105/ml were
transfected with WT (Fluc) and mutant (Fluc-opal) plasmids, by
using lipofectamine 2000 (Invitrogen). Cells were incubated for
24 h and PTC124 and the others compounds (12 mM) were added
for additional 24 h. Next, cells were washed with PBS, incubated
with the detection mix Steady-Glo luciferase reagent (Promega)
and 200 ml of cell suspension were plated in triplicate in a 96 well
plate. Luciferase activity was measured on
(Promega).
a luminometer
2.3.3. Cell viability assay
Fig. 4. Cell viability assay of HeLa cells after 24, 48, and 72 h treatment with G418,
PTC124, and PTC124 analogues (Control: opal untreated cells).
2 ꢂ 105/ml HeLa cells were plated and incubated with PTC124
(12 mM), G418 (300 mg/mL) and PTC124 analogues (12 mM). Cells
were counted every 24 h during the treatments (for a total of 72 h)
and the results reported in a graph.
2.3. Biology
2.3.1. Cell culture conditions and transfection of reporter plasmid
HeLa, CFBE41o- (expressing ectopic CFTR-WT, were kindly
provided by Prof. Louis Galietta, Ospedale Gaslini-Genova, Italy)
2.3.4. Immunofluorescence microscopy
To visualize the CFTR protein, cells were grown on rounded glass
coverslips and fixed with methanol for 2 min. The cell membrane
and Golgi apparatus were stained by the Wheat Germ Agglutinin
(WGA) Alexa 594 (Life Technologies). Coverslips were incubated
with a mouse monoclonal antibody (CF3) that recognizes the first
extracellular loop of human CFTR (Abcam, 1:500) overnight at 4 ꢁC,
followed by a goat polyclonal to mouse Alexa-Fluor-488 (Abcam,
1:1000) secondary antibody for 1 h at 37 ꢁC. Nuclei were visualized
with DAPI. Cells were examined under a Zeiss Axioskop microscope
equipped for fluorescence.
2.3.5. Western blotting
Proteins (50 mg) were separated by 10% SDS-PAGE and 4e12%
SDS-PAGE containing 0.1% SDS and transferred to Hybond-C
nitrocellulose membranes (Amersham Life Science) by electro-
blotting. The membrane was incubated with a goat polyclonal
antibody antieCFTR (C-19, Santa Cruz 1:500) raised against a
peptide mapping near the C-terminus of CFTR of human origin, and
HRP-conjugated anti-goat (Abcam, 1:5000). The target protein was
detected by ECL reagents (Pierce). We used
(mouse; Sigma-Aldrich 1:10.000) to confirm equal proteins
loading. Gel bands were quantified by Image Lab software (BioRad).
b-tubulin antibody
Fig. 5. Immunofluorescence assay showing the CFTR protein in CFBE41o-cells. Nuclei
(Top left), CFTR localization (Top right), Membrane and Golgi apparatus (Bottom left)
and merged images (Bottom right).

I. Pibiri et al. / European Journal of Medicinal Chemistry 122 (2016) 429e435
433
Fig. 6. Immunofluorescence assay showing Nuclei (blue), CFTR (green), cell membrane and Golgi apparatus (red) and the corresponding merged images to detect readthrough of
the CFTR’s UGA premature stop codon in IB3.1 human cells untreated (NT) or treated for 24 h with G418 (positive control), PTC124 and NV1898. (For interpretation of the references
to colour in this figure legend, the reader is referred to the web version of this article.)
3. Results and discussion
point pharmacophore variants were retrieved: seven AHRR and
six AARR variants, where R indicates aromatic rings, while A and H
are H-bond acceptor and donor sites, respectively. Six pharmaco-
phore hypothesis survived the scoring protocol and the best one
was AARR.5 as shown in Fig. 1.
Such hypothesis is represented by two aromatic rings (R7 and
R8) distant 7.67 Å from each other and separated by two hydrogen
bond acceptors (A1 and A2) at a distance of 2.28 Å from each other.
A1 is distant 4.98 Å and 3.79 Å from R7 and R8, respectively, while
A2 is distant 3.87 Å and 3.08 Å from R7 and R8, respectively. It is
quite surprising to note that no particular role is assigned to the
fluorine and carboxylic moiety by the scoring protocol, proposing
the diaryl azole as the main scaffold and suggesting only an ancil-
lary role of the substituents on the aryl rings.
The original report on the activity of PTC124 highlighted the
selectivity of such a small molecule toward the readthrough of
premature UGA codon, also pointing out the dependence of the
activity on the identity of the nucleobases surrounding the PTC.
More recently, the importance of H-bonding and
pꢀp stacking
interactions between the PTC124 and its hypothetical target have
been remarked. The inactivity of a very similar set of PTC124 ana-
logues and the finding of analogues with improved activity
prompted us to explore a ligand-based virtual screening approach
in attempting to optimize the structure of PTC124-like compounds.
The used dataset for the pharmacophore modeling study con-
sisted of PTC124 and twenty PTC124 analogs. Among the latter, the
ten compounds showing no readthrough activity, and the other ten
known to give artifacts in luciferase activity assays, were identified
as inactive compounds, while PTC-124 was identified as the active
compound. The scoring protocol allowed the different hypotheses
to be ranked so that the most appropriate can be chosen for further
investigation. Inactive molecules were scored, in order to observe
the alignment of these molecules with the pharmacophore hy-
potheses and to select the best ones. The larger the difference be-
tween the scores of the active and inactive molecules, the better the
hypothesis is at discriminating among candidates. Thirteen four-
A
virtual high-throughput screening was carried out by
matching the AARR.5 pharmacophore model with compounds
contained in the drug-like subset of ZINC database. By optimizing
the diaryloxadiazole scaffold, a set of 250 compounds was filtered
by the virtual screening of the initial database. From the top 5% of
retrieved hits in term of fitness with the hypothesis, eight com-
pounds (Fig. 2) were selected to be synthesized and tested for
readthrough activity. Interestingly, although still belonging to the
class of 3,5-diaryl-1,2,4-oxadiazoles, none of the selected hits
contained the two main features of PTC124 lead, i.e. the fluorine

434
I. Pibiri et al. / European Journal of Medicinal Chemistry 122 (2016) 429e435
Detection of increased level of the CFTR protein (Fig. 6) by
immunofluorescence microscopy confirmed that the NV1898
compound was able to readthrough the UGA premature stop codon
inducing the re-expression of CFTR and its localization on the
membrane of human IB3.1 cells.
In addition, western blot analysis indicated increased CFTR
expression after 24e72 h of treatment with NV1898 (Fig. 7; top).
Quantitation of the band intensity (Fig. 7; bottom) demonstrates
similar CFTR protein expression levels between PTC124 and
NV1898 treated cells at 72 h.
4. Conclusions
The so called “small molecule” strategy to promote the read-
through of premature termination codons is currently focused on
two types of molecules, i.e. aminoglycosides and diaryloxadiazoles,
although recent findings have pointed out at the potential role in
nonsense mutation suppressing by the antiinflammatory drug
amlexanox. However, PTC124 is both structurally and dimension-
ally different from aminoglycosides, for which a mechanistic hy-
pothesis has been proved. Therefore, there is still an urgent need for
data concerning the structure-activity relationship (SAR) of the
PTC124 scaffold, in order to provide a likely hypothesis for PTC124
mode of action thus opening the way to further drug development.
Indeed, despite the pioneering original study nature has exten-
sively tested several hundred thousands of compounds finally
leading to select PTC124 on the basis of FLuc assays, the possibility
to optimize such compound is not precluded. Recently, the SAR of
diaryl-1,2,4-oxadiazoles has been investigated mainly by focusing
on the number and position of fluorine substituents on the C(5)-
linked aromatic ring. The present work represents the first sys-
tematic approach of computer driven drug design applied through
the ligand based virtual screening of a series of PTC124 analogues.
Although not presenting an original scaffold, our study demon-
strates the possibility of designing and obtaining alternative
PTC124 analogues with similar activity, even in the absence of the
fluorine and carboxylic moieties. The used approach is particularly
valid in the absence of a known biological target, and is iteratively
improvable by implementing bioactivity data of the synthesized
compounds into the pharmacophoric dataset.
Fig. 7. Western blot showing CFTR protein levels in IB3.1 cells untreated (untr.) or
treated with PTC124 or NV1898. A primary antibody targeting the C-terminus of CFTR
was used, b-tubulin was used as a loading control (Top). Graph shows the quantitation
of the bands by densitometry using the Image Lab software (BioRad) (Bottom).
substituent and the carboxylic moiety.
Selected hits were then synthesized in good to excellent yields
(62e92%) by cyclization of arylamidoximes II, prepared from the
corresponding nitriles
I and hydroxylamine, with variously
substituted benzoyl chloride, appropriately chosen according to the
required substitution pattern (Scheme 1).
The readthrough activity of synthesized compounds was pre-
liminarily assessed by using the FLuc cell-based assay (Fig. 3) [8]. To
this aim HeLa cells were transiently transfected with the plasmids
pFLuc-WT and pFLuc-opal plasmids, and Luciferase activity was
measured by luminescence. The dose used (12 mM) was chosen on
the basis of previous results on similar compounds. Detection of
high levels of luciferase activity showed by HeLa cells transfected
with the pFLuc-WT plasmid indicated the correct functioning of
this assay.
Conflict of interest
The authors declare that they have no conflict of interest.
Cell viability (Fig. 4) experiments conducted for a total of 72 h
showed that all the PTC124 analogues were significantly less
cytotoxic, at the active dose, than G418 chosen as control.
Authors contribution
All authors have contributed to this article and particularly I.
Pibiri and L. Lentini have contributed equally.
The
3-(2-pyridyl)-5-(3-toluyl)-1,2,4-oxadiazole
(NV1898)
showed readthrough activity comparable to that of PTC124.
Considering the basic nature of the pyridyl moiety, this result re-
inforces the hypothesis that the 3-aryl moiety on the 1,2,4-
oxadiazole ring is involved as H-bond acceptor in the interaction
with the biological target, as observed in our previous molecular
dynamic study on PTC124.
The activity of NV1898 was then tested for suppression of
nonsense mutations in the CF bronchial epithelial IB3.1 cell line
(CFTR genotype W1282X/F508del). As a positive control for the
CFTR protein expression we used CFBE41o-cells that express
ectopically a wild type CFTR gene. The CFTR protein was revealed
(green) by a specific antibody targeting its first external loop
(Alexa-488). Nuclei were DAPI stained (blue) while the cell mem-
brane and Golgi apparatus were stained with WGA-Alexa-594 (red)
(Fig. 5).
Acknowledgments
This work was funded by the Italian Cystic Fibrosis Research
Foundation, grant FFC#02/2011 to Prof. Aldo Di Leonardo with the
contribution of Delegazione di Lecce and Delegazione di Vittoria
(Ragusa); FFC#1/2014 to Dr. Laura Lentini, with the contribution of
Delegazione di Palermo, Catania 2 and Vittoria (Ragusa). We thank
Prof. Jhon P. Clancy (Cincinnati Children’s Hospital Medical Center,
OHIO-USA) and Prof. Louis Galietta (Ospedale Gaslini Genova), for
respectively producing and providing us with CFBE41o-cells. We
are also grateful to Prof. J. Inglese (NIH Chemical Genomics Center,
National Institutes of Health, Bethesda) for kindly providing us with
the FLuc190UGA plasmid.

I. Pibiri et al. / European Journal of Medicinal Chemistry 122 (2016) 429e435
435
S.J. Steinberg, J.G. Hacia, Nonsense suppressor therapies rescue peroxisome
lipid metabolism and assembly in cells from patients with specific PEX gene
mutations, J. Cell Biochem. 112 (2011) 1250e1258.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2016.06.048.
[18] S.C. Harmer, J.S. Mohal, D. Kemp, A. Tinker, Readthrough of long-QT syndrome
type 1 nonsense mutations rescues function but alters the biophysical prop-
erties of the channel, Biochem. J. 443 (2012) 635e642.
[19] S.P. McElroy, T. Nomura, L.S. Torrie, E. Warbrick, U. Gartner, G. Wood,
W.H. McLean, A lack of premature termination codon read-through efficacy of
PTC124 (Ataluren) in a diverse array of reporter assays, PLoS Biol. 11 (2013)
1001593.
[20] L. Lentini, R. Melfi, A. Di Leonardo, A. Spinello, G. Barone, A. Pace, A. Palumbo
Piccionello, I. Pibiri, Towards a rationale for the PTC124 (Ataluren) promoted
read-through of premature stop codons: a computational approach and GFP-
reporter cell-based assay, Mol. Pharm. 11 (2014) 653e664.
References
[1] S. Kervestin, A. Jacobson, NMD: a multifaceted response to premature trans-
lational termination, Nat. Rev. Mol. Cell Biol. 13 (2012) 700e712.
[2] J.T. Mendell, H.C. Diez, When the message goes awry: disease-producing
mutations that influence mRNA content and performance, Cell 107 (2001)
411e414.
[3] L. Bidou, V. Allamand, J.P. Rousset, O. Namy, Sense from nonsense: therapies
for premature stop codon diseases, Trends Mol. Med. 18 (2012) 679e688.
[4] T. Goldmann, N. Overlack, F. Moller, V. Belakhov, M. van Wyk, T. Baasov,
U. Wolfrum, K. Nagel-Wolfrum, A comparative evaluation of NB30, NB54 and
PTC124 in translational read-through efficacy for treatment of an USH1C
nonsense mutation, EMBO Mol. Med. 4 (2012) 1186e1199.
[5] I. Sermet-Gaudelus, K.D. Boeck, G.J. Casimir, F. Vermeulen, T. Leal, A. Mogenet,
D. Roussel, J. Fritsch, L. Hanssens, S. Hirawat, N.L. Miller, S. Constantine,
A. Reha, T. Ajayi, G.L. Elfring, L.L. Miller, Ataluren (PTC124) induces cystic
fibrosis transmembrane conductance regulator protein expression and activ-
ity in children with nonsense mutation cystic fibrosis, Am. J. Respir. Crit. Care
Med. 182 (2010) 1262e1272.
[6] E.M. Welch, E.R. Barton, J. Zhuo, Y. Tomizawa, W.J. Friesen, P. Trifillis,
S. Paushkin, M. Patel, C.R. Trotta, S. Hwang, R.G. Wilde, G. Karp, J. Takasugi,
G. Chen, S. Jones, H. Ren, Y.C. Moon, D. Corson, A.A. Turpoff, J.A. Campbell,
M.M. Conn, A. Khan, N.G. Almstead, J. Hedrick, A. Mollin, N. Risher, M. Weetall,
S. Yeh, A.A. Branstrom, J.M. Colacino, J. Babiak, W.D. Ju, S. Hirawat,
V.J. Northcutt, L.L. Miller, P. Spatrick, F. He, M. Kawana, H. Feng, A. Jacobson,
S.W. Peltz, H.L. Sweeney, PTC124 targets genetic disorders caused by nonsense
mutations, Nature 447 (2007) 87e91.
[21] I. Pibiri, L. Lentini, R. Melfi, G. Gallucci, A. Pace, A. Spinello, G. Barone, A. Di
Leonardo, Enhancement of premature stop codon readthrough in the CFTR
gene by Ataluren (PTC124) derivatives, Eur. J. Med. Chem. 101 (2015)
236e244.
[22] D.S. Auld, N.T. Southall, A. Jadhav, R.L. Johnson, D.J. Diller, A. Simeonov,
C.P. Austin, J. Inglese, Characterization of chemical libraries for luciferase
inhibitory activity, J. Med. Chem. 51 (2008) 2372e2386.
[23] D.S. Auld, N. Thorne, W.F. Maguire, J. Inglese, Mechanism of PTC124 activity in
cell-based luciferase assays of nonsense codon suppression, Proc. Natl. Acad.
Sci. U. S. A. 106 (2009) 3585e3590.
[24] Ligprep V. 2.3, Schrodinger LLC, New York, 2009.
[25] A.M. Almerico, M. Tutone, A. Lauria, Receptor-guided 3D-QSAR approach for
the discovery of c-kit tyrosine kinase inhibitors, J. Mol. Model. 18 (2012)
2885e2895.
[26] Phase V. 3.1, Schrodinger LLC, New York, 2009.
[27] A.M. Almerico, M. Tutone, L. Pantano, A. Lauria, A3 adenosine receptor: ho-
mology modeling and 3D-QSAR studies, J. Mol. Graph. Model 42 (2013)
60e72.
[28] A. Lauria, M. Ippolito, M. Fazzari, M. Tutone, F. Di Blasi, F. Mingoia,
A.M. Almerico, IKK-beta inhibitors: an analysis of drug-receptor interaction by
using molecular docking and pharmacophore 3D-QSAR approaches, J. Mol.
Graph. Model 29 (2010) 72e81.
[7] S.M. Rowe, J.P. Clancy, Pharmaceuticals targeting nonsense mutations in ge-
netic diseases, Biodrugs 23 (2009) 165e174.
[8] D. Diop, C. Chauvin, O. Jean-Jean, Aminoglycosides and other factors pro-
moting stop codon readthrough in human cells, C. R. Biol. 330 (2007) 71e79.
[9] T. Hermann, Aminoglycoside antibiotics: old drugs and new therapeutic ap-
proaches, Cell. Mol. Life. Sci. 64 (2007) 1841e1852.
[10] L.S. Mc Coy, Y. Xie, Y. Tor, Antibiotics that target protein synthesis, Wiley
Interdiscip. Rev. RNA 2 (2011) 209e232.
[11] V. Malik, L.R. Rodino-Klapac, L. Viollet, J.R. Mendell, Aminoglycoside-induced
mutation suppression (stop codon read- through) as a therapeutic strategy for
Duchenne muscular dystrophy, Ther. Adv. Neurol. Disord. 3 (2010) 379e389.
[12] N.T. Chandrika, S. Garneau-Tsodikova, A review of patents (2011e2015) to-
wards combating resistance to and toxicity of aminoglycosides, Med. Chem.
Commun. 7 (2016) 50e68.
[13] J. Karijolich, Y.T. Yu, Therapeutic suppression of premature termination co-
dons: mechanisms and clinical considerations (Review), Int. J. Mol. Med. 34
(2014) 355e362.
[29] J.J. Irwin, B.K. Sholchet, Zinc ꢀ a free database of commercially available
compounds for virtual screening, J. Chem. Inf. Model 45 (2005) 177e182.
[30] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug
discovery and development settings, Adv. Drug Deliv. Rev. 46 (2001) 3e26.
[31] R.M. Srivastava, M.N. Ramos, L.M. Mendes e Silva, Synthesis, spectroscopic
studies and molecular orbital calculations of O-acylbenzamidoximes, Gazz.
Chim. Ital. 113 (1983) 845e850.
[32] K. Gobis, H. Foks, A. Kedzia, M. Wierzbowska, Z. Zwolska, Synthesis and
antibacterial activity of novel pyridine and pyrazine derivatives obtained from
amidoximes, J. Heterocycl. Chem. 46 (2009) 1271e1279.
[33] R. Delaby, P. Reynaud, T. Tupin, Pyridine amidines and amidoximes, Bull. Soc.
Chim. Fr. (1957) 714e717.
€
[34] P. Krüger, Ueber Abkommlinge des Benzenylamidoxims, Ber. Dtsch. Chem.
Ges. 18 (1885) 1055e1060.
[14] A. Prayle, A.R. Smyth, Aminoglycoside use in cystic fibrosis: therapeutic
strategies and toxicity, Curr. Opin. Pulm. Med. 16 (2010) 604e610.
[15] J.A. Loudon, Repurposing amlexanox as a ’run the red light cure-all’ with
readthrough-a ’no-nonsense’ approach to personalised medicine, J. Bioanal.
Biomed. 5 (2013) 79e96.
[35] A. Pace, P. Pierro, The new era of 1,2,4-oxadiazoles, Org. Biomol. Chem. 7
(2009) 4337e4348.
[36] D. Grant, R. Dahl, N.D. Cosford, Rapid multistep synthesis of 1,2,4-Oxadiazoles
in
a single continuous microreactor sequence, J. Org. Chem. 73 (2008)
7219e7223.
[16] J.A. Loudon, Preventing and correcting communicable and non-communicable
chronic disease via amlexanox-Dual ’no-nonsense’ and inflammatory axis
targeting, J. Bioanal. Biomed. 5 (2013) 138e179.
[37] M.F. Brana, J.M. Castellano, M.J.R. Yunta, Synthesis of 5-aryl-3-(4-pyridyl)-
1,2,4-oxadiazoles, J. Heterocycl. Chem. 20 (1983) 1403e1405.
[38] G. Leandri, M. Pallotti, Diazofurans. Ann. Chim. 47 (1957) 376e384.
[17] P.K. Dranchak, E. Di Pietro, A. Snowden, N. Oesch, N.E. Braverman,