Discovery and structural diversity of the hepatitis C virus NS3/4A serine protease inhibitor series leading to clinical candidate IDX320

Article abstract of DOI:10.1016/j.bmcl.2015.09.009

Exploration of the P2 region by mimicking the proline motif found in BILN2061 resulted in the discovery of two series of potent HCV NS3/4A protease inhibitors. X-ray crystal structure of the ligand in contact with the NS3/4A protein and modulation of the

Full text of DOI:10.1016/j.bmcl.2015.09.009

Accepted Manuscript
Discovery and structural diversity of the hepatitis C virus NS3/4A serine pro-
tease inhibitor series leading to clinical candidate IDX320
Christophe C. Parsy, François-René Alexandre, Valérie Bidau, Florence
Bonnaterre, Guillaume Brandt, Catherine Caillet, Sylvie Cappelle, Dominique
Chaves, Thierry Convard, Michel Derock, Damien Gloux, Yann Griffon, Lisa
B. Lallos, Frederic Leroy, Michel Liuzzi, Anna-Giulia Loi, Laure Moulat,
Musiu Chiara, Virginie Roques, Elodie Rosinovsky, Simon Savin, Maria Seifer,
David Standring, Dominique Surleraux
PII:
S0960-894X(15)30041-X
http://dx.doi.org/10.1016/j.bmcl.2015.09.009
BMCL 23090
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
1 June 2015
2 September 2015
4 September 2015
Please cite this article as: Parsy, C.C., Alexandre, o-R., Bidau, V., Bonnaterre, F., Brandt, G., Caillet, C., Cappelle,
S., Chaves, D., Convard, T., Derock, M., Gloux, D., Griffon, Y., Lallos, L.B., Leroy, F., Liuzzi, M., Loi, A-G.,
Moulat, L., Chiara, M., Roques, V., Rosinovsky, E., Savin, S., Seifer, M., Standring, D., Surleraux, D., Discovery
and structural diversity of the hepatitis C virus NS3/4A serine protease inhibitor series leading to clinical candidate
IDX320, Bioorganic & Medicinal Chemistry Letters (2015), doi: http://dx.doi.org/10.1016/j.bmcl.2015.09.009
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.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Discovery and structural diversity of the
hepatitis C virus NS3/4A serine protease
inhibitor series leading to clinical candidate
IDX320
Leave this area blank for abstract info.
a, 
a
a#
a#
a
Christophe C. Parsy , François-René Alexandre , Valérie Bidau , Florence Bonnaterre , Guillaume Brandt ,
a#
a
a#
a
a#
Catherine Caillet , Sylvie Cappelle , Dominique Chaves , Thierry Convard , Michel Derock , Damien
a#
a#
b#
a#
b#
b#
a#
Gloux , Yann Griffon , Lisa B. Lallos , Frederic Leroy , Michel Liuzzi , Anna-Giulia Loi , Laure Moulat ,
b#
a#
a#
a#
b
b#
Musiu Chiara , Virginie Roques , Elodie Rosinovsky , Simon Savin , Maria Seifer , David Standring and
Dominique Surleraux
a#
3
5
6
: R = CH3
3
7
8
: R = Cl
(IDX320)

Bioorganic & Medicinal Chemistry Letters
Discovery and structural diversity of the hepatitis C virus NS3/4A serine
protease inhibitor series leading to clinical candidate IDX320
a, 
a
a#
a#
Christophe C. Parsy , François-René Alexandre , Valérie Bidau , Florence Bonnaterre , Guillaume
a
a#
a
a#
a
a#
Brandt , Catherine Caillet , Sylvie Cappelle , Dominique Chaves , Thierry Convard , Michel Derock ,
Damien Gloux , Yann Griffon , Lisa B. Lallos , Frederic Leroy , Michel Liuzzi , Anna-Giulia Loi ,
Laure Moulat , Musiu Chiara , Virginie Roques , Elodie Rosinovsky , Simon Savin , Maria Seifer ,
David Standring and Dominique Surleraux
a#
a#
b#
a#
b#
b#
a#
b#
a#
a#
a#
b
b#
a#
a
IDENIX, a wholly-owned subsidiary of Merck & Co, 1682 rue de la Valsière, Cap Gamma, BP 50001, 34189 MONTPELLIER Cedex 4, FRANCE
IDENIX, a wholly-owned subsidiary of Merck & Co, 320 Bent Street - 4th Floor, Cambridge, MA 02139, USA
Former Idenix employees
b
#
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
Exploration of the P2 region by mimicking the proline motif found in BILN2061 resulted in the
discovery of two series of potent HCV NS3/4a protease inhibitors. X-Ray crystal structure of the
ligand in contact with the NS3/4A protein and modulation of the quinoline heterocyclic region
by structure based design and modeling allowed for the optimization of enzyme potency and
cellular activity. This research led to the selection of clinical candidate IDX320 having good
genotype coverage and pharmacokinetic properties in various species.
Keywords:
Hepatitis C
HCV NS3/4A protease inhibitors
Macrocycle
2015 Elsevier Ltd. All rights reserved.
Current reports indicate that 3% of the world’s population is
infected with Hepatitis C virus (HCV), which leads to chronic
hepatitis. In some cases, this progress to liver cirrhosis and
hepatocellular carcinoma; HCV is the main cause of liver
The understanding of the HCV genomic organization, life
cycle of the virus, as well as the development of HCV
replicons has enhanced rational drug design efforts and led to
the discovery of direct acting antiviral inhibitors of HCV
1
4–7
transplantation in developed countries. The previous standard
protease. Currently, two classes of protease inhibitors have
of care introduced in 2001 covering all HCV genotypes,
combining pegylated interferon and ribavirin, induces a
sustained viral response (SVR) in about 50% of patients
affected by HCV genotype 1, highlighting the need for more
emerged (Chart 1). The serine trap inhibitors, such as
8
9
telaprevir (VX-950) (1) and boceprevir (SCH-503034) (2),
include the first class and are constructed around an α-
ketoamide moiety that covalently and reversibly binds S139,
part of the catalytic triad of the HCV protease. Due to the
recent progress in HCV treatment option, (1) and (2) are no
2
effective chemotherapeutic agents to treat HCV patients.
Recent advances in treatment options with the approval of
sofosbuvir (nucleoside polymerase inhibitor) and the
combination of sofosbuvir and ledipasvir (NS5A replication
inhibitor) has seen dramatic cure rate up to 99% in some
1
0
longer recommended for regimen therapy and discontinued.
The second class resulted from the discovery that the N-
terminal cleavage product from the NS5A/B cleavage site is a
competitive inhibitor of the NS3 protease. Truncation of this
hexapeptide to a tetrapeptide on a rigid macrocyclic scaffold
yielded the non-covalent macrocyclic protease inhibitor, BILN
2061 (3). This was the first HCV protease inhibitor to enter
clinical trials, and in only 2 days of monotherapy reduced viral
RNA by 2-3 log10 IU/mL in patients, thereby establishing
3
specific genotype population patients. It is likely that the next
evolution in HCV therapy will be the identification of a pan-
genotypic direct acting antiviral combination therapy including
an NS3/4A protease inhibitor that could shorten the treatment
duration below the current 12 weeks across all HCV patient
population.
—
——

Corresponding author. Tel.: +33(0)499522252; fax: +33(0)499522250; e-mail: christophe.parsy@merck.com

2
1
proof of concept for HCV protease as a viable antiviral
developed in our laboratories. This synthetic strategy enabled
us to introduce diversity on the macrocycle using different
quinoline rings (13a-w) with a straightforward Mitsunobu
coupling. After coupling of the quinoline and hydrolysis of the
ester, the resulting acid (15) was activated by CDI to form an
intermediate oxazolidinedione that is open in the presence of
the P1 sulfonamide to yield the final target inhibitors (16a-x)
and (7).
11–14
target.
Unfortunately, BILN-2061 had to be discontinued
15
due to cardiac toxicities in animal studies. Despite the fact
that the mechanism of cardiac toxicity has not been reported, it
is believed to be compound specific and several groups,
inspired by this pioneer drug, have pursued macrocyclic
16
inhibitors leading to the approval of simeprevir (4).
a
b
c
9
10
11
12
1
1a: R7=TBDMS
11b: R =4-NO2Bz
7
1
2
Telaprevir (VX-950)
Boceprevir (SCH 503034)
d
e
f
P2
P1
14a-x
15a-x
16a-x, 7
Scheme 1. Reagents and conditions: (a) TBTU, diisopropylethylamine, N-
methylhex-5-en-1-amine tosylate salt, DMF, 0°C to RT, 16h, 95%; (b) (1)
TFA, DCM, RT, 3h, quant. (2) ethyl (1R,2S)-1-[(3-methylimidazol-3-ium-
P4
P3
1
-carbonyl)amino]-2-vinyl-cyclopropanecarboxylate, Et
16h, 70%, (3i) TBDMSCl, Et N, DCM, RT, 16h, 70% or (3ii) 4-
nitrobenzoyl chloride, Et N, DCM, RT, 16h, 46-70%; (c) (1) Hoveyda-
Grubbs 2 generation cat. or Zhan II cat., DCE reflux, 3h, 43-64%, (2i)
TBAF, THF, RT, 2h, 94% or (2ii) LiOH, H O/THF, 0°C, 45min, 71-81%;
d) quinolines 13a-w, PPh , DIAD, THF, 0°C to RT, 16h, 15-89%; (e)
LiOH, H O/THF, RT, 16h, 10-90%; (f) (1) CDI, THF or EDCI, DCM,
0°C,
methylcyclopropylsulfonamide
3
N, DCM, RT,
3
4
3
Ciluprevir (BILN 2061)
Simeprevir (TMC 435)
3
nd
Chart 1. First generation of NS3/4A protein inhibitors
2
(
3
2
8
microwaves
irradiations,
or
50min,
(2)
DBU,
1-
In this letter, we present our continuous efforts originated from
cyclopropylsulfonamide,
80°C,
the 2,4-disubstituted azetidine motif (5) and (6) in the P2
1
7
microwaves irradiations, 1h30, 9-51%.
region. We designed two novel series of macrocyclic HCV
protease inhibitors by introducing structure diversity at the P2
region (Chart 2). The first series differs from BILN2061 by
switching position 1 and 2 of the proline ring therefore creating
a proline urea moiety (7). The second series represents a ring
expansion of P2 region by using pipecolic acid instead of
pyrrolidine acid as a starting building block. These efforts led
to the clinical drug candidate IDX320 (8).
R2=
R2=
3
4
5
6
4
3
5
6
R =Cl, R =OMe, R =R =H : 13j
R =OMe, R =R =R =H : 13a
R3=Me, R4=OMe, R5=R6=H : 13b
3
4
5
6
R =F, R =OMe, R =R =H : 13c
R2=
3
4
5
6
R =Cl, R =OMe, R =R =H : 13d
3
4
5
6
R =Cl, R =OMe, R =R =H : 13k
3
6
4
5
R =R =H, R =Cl,R =OMe : 13e
3
4
5
6
R =Br, R =OMe, R =R =H : 13f
R2=
3
4
5
6
R =Me, R =OMe, R =R =H : 13l
R2=
3
4
5
6
R =Cl, R =OMe, R =R =H : 13g
R2=
5
6
: R3= CH3
= Cl
3
4
5
6
: R
3
7
IDX320
8
R =Cl, R =OMe, R =R =H : 13t
R2=
3
4
5
6
R =Me, R =OMe, R =R =H : 13u
3
4
5
6
R =Cl, R =OMe, R =R =H : 13h
Chart 2. From a lead series to the drug candidate IDX320
R2=
3
4
5
6
R =Cl, R =OMe, R =R =H : 13v
R2=
The synthesis of the proline urea series was based on a 4-
hydroxy proline scaffold and started from the commercially
available N-Boc protected 4-hydroxy-2-pyrrolidine carboxylic
acid (9) (Scheme 1). Peptide coupling with the N-methyl
hexenyl mediated by TBTU, Boc-deprotection, coupling with
the 1,1’-carbonyldiimidazole (CDI) activated vinylcyclopropyl
amino acid ester followed by protection of the pyrrolidine
hydroxyl by tert-butyldimethylsilyl chloride (TBDMSCl) or 4-
nitrobenzoyl chloride resulted in the diene (11). This diene
3
4
5
6
R =Cl, R =OMe, R =R =H : 13i
R2=
3
4
5
6
R =Cl, R =OMe, R =R =H : 13w
Chart 3. Quinolines (13a-l), (13t-w).
2
-Thiazoles, 4-thiazoles, 3-oxazoles and 2-oxazoles
substituted quinolines (13a-l) and (13t-w) (Chart 3) were
synthesised using described methods, starting from ortho-acyl
anilines and corresponding thiazole or oxazole acids.
should be noted that for (13t-w), functionalizations
Sonogashira and Heck reactions) at the position 4 of the 4-
bromo-thiazole were realized before the cyclisation step,
16,22
(11) was then engaged in a Ruthenium catalysed ring closing
It
metathesis (RCM) followed by deprotection of the hydroxyl
18
group to yield the 14-membered macrocycle (12). The RCM
(
19
st
nd
can be carried out with Hoveyda-Grubs 1 or 2 generation
2
0
catalyst or Zhan’s catalyst along with DarPhin catalyst

unless no reaction was observed. Quinolines (13m-s) were
prepared according to scheme 2 using a concise synthesis in
four or five steps from anilines (17a-b). First step consisted in
a cyclisation reaction of malonic acid in the presence of
phosphorus oxychloride to yield the 2,4-dichloroquinolines
a
b
c
21
22
23
24
(18a-b). Then the 4 position was protected in 50% yield with
the reaction of para-methoxybenzyl alcohol (PMB) giving a
mixture of major 4-OPMB and minor 2-OPMB adducts, which
were easily separable by chromatography on silica gel. N-
branched pyrazoles were then introduced via a SNAr reaction
in moderate yields, followed by PMB deprotection using TFA
to afford (13m-n). (13p-r) were obtained via Negishi, Suzuki-
Miyaura, or Stille reaction at position 2, followed by TFA
treatment. Triazole (13o) was synthesised from 2-azido
quinoline (20b) followed by cyclisation with ethynyl
derivatives. Oxadiazole (13s) was obtained from 2-cyano-
quinoline (20a) then reaction with hydroxylamine and
followed by cyclisation with isobutyryl chloride.
d
e
f
25
26
27
Scheme 3. Reagents and conditions: (a) (1) H
quant., (2) Et N, N-hex-5-enyl-N,3-dimethyl-imidazol-3-ium-1-
carboxamide iodide, THF, 120°C, microwaves irradiations, 1h, 66%; (b)
1) LiOH, H O/THF, RT, quant., 16h, (2) TBDMSCl, Imidazole, DMF,
RT, 20h, 76%, (3) [(1R,2S)-1-methoxy-carbonyl-2-vinyl-
cyclopropyl]ammonium tosylate salt, DIPEA, TBTU, DCM, RT, 16h,
48%; (c) Hoveyda-Grubbs 1 generation cat., DCE reflux, 4h, 54%; (d)
13u, PS-PPh
RT, 16h, 46%; (f) (1) EDCI, DCM, RT, 3h, (2) DBU, 1-
methylcyclopropylsulfonamide, RT, 3h, 50%.
2
, Pd/C, EtOH, RT, 16h,
3
(
2
st
3
, DIAD, THF, 0°C to RT, 16h, 42%; (e) LiOH, H
2
O/THF,
g
2
0b: R3=Cl
a
b
f
a
b
c
21
28
29
1
1
7a: R3=Me
7b: R3=Cl
18a: R3=Me
18b: R3=Cl
19a: R3=Me
19b: R3=Cl
13m-s
d
c
d
e
2
0a: R3=Me
3
3
0a: R1=H
0b: R1=Me
31a: R1=H
8: R1=Me
R2=
3
Scheme 4. Reagents and conditions: (a) 13l, PPh , DIAD, THF, 0°C to RT,
2
h, 52%; (b) (1) TFA, 60°C, 5h, 97%, (2) Et N, N-hex-5-enyl-N,3-
3
R3= Me
R3= Cl
13m
-
-
13p
13q
-
-
dimethyl-imidazol-3-ium-1-carboxamide iodide, THF, 120°C, microwaves
irradiations, 1h, 78%; (c) (1) LiOH, H O/THF, 40°C, 16h, 90%, (2)
-
13n
13o
-
-
13r
13s
2
(
1R,2S)-1-amino-N-cyclopropyl-sulfonyl-2-vinyl-cyclopropane-
carboxamide or (1R,2S)-1-amino-N-(1-methylcyclopropyl)sulfonyl-2-
vinyl-cyclopropanecarboxamide, DIPEA, TBTU, DMF, RT, 16h, 30-93%;
Scheme 2. Reagents and conditions: (a) malonic acid, POCl
3-75% ; (b) p-methoxybenzylalcohol, NaH, 15-crown-5, DMF, 0°C to
RT, 16h, 50-72%; (c) (i) pyrazole, NMP, 200°C, microwaves irradiations,
3
, reflux, 16h,
6
(d) Zhan IB cat., DCE reflux, 1h, 20-36%.
3
5
5
0min, 33%, for 13m; or (ii) (1) pyrazole, NaH, DMF, 0°C to 90°C, 6h,
0%; (2) CeCl .7H O, NaI, CH CN, 85°C, microwaves irradiations, 1h,
8%; for 13n; or (iii) (1) (3-(2-trimethylsilylethynyl) phenyl) boronic acid,
, PPh , Na CO , H O, dioxane, 100°C, microwaves irradiations,
h, 52%; (2) TFA, RT, 10min, quant.; for 13q; or (iv) (1) 4-methyl-2-
tributylstannyl)thiazole, PdCl (PPh , K CO , DMF, 90°C, 16h, 65%; (2)
TFA, RT, 10min, quant.; for 13r; (d) Zn, Zn(CN) , Pd dba , dppf, DMA,
20°C, microwaves irradiations, 70min, 67%; (e) (i) TFA, RT, 10min,
2% for 13p; or (ii) (1) NH OH.HCl, TEA, EtOH, reflux, 3h30; (2)
Our previous research efforts on the discovery of HCV
3
2
3
protease inhibitors led to a novel series of -strand mimic
17
having 2,4-disubstituted azetidine motif in the P2 region. The
best compounds from this series (5) and (6) (Chart 2)
demonstrated excellent biochemical activity IC50 6.6 and 2.1
nM and good inhibition of HCV replicon replication in Huh-7
cells containing Con1 subgenomic replicon (GS4.1 cells)
monitored for expression of the NS5A protein with EC50 = 45
and 60 nM, respectively, and good selectivity indices (SI)
Pd(OAc)
1
(
2
3
2
3
2
2
3
)
2
2
3
2
2
3
1
9
2
3
isobutyryl chloride, isobutyric acid, 100°C, 3h, 87%; for 13s; (f) NaN ,
DMF, 80°C, 48h, 12%; (g) (1) 3-methylbutyne, CuI, DIPEA, DMF,
1
9
40°C, microwaves irradiations, 45min, 84%; (2) TFA, DCM, RT, 10min,
4%; for 13o.
>
1500. Further in vitro evaluation of these two compounds
demonstrated high mean clearance (>150 l/min/mg prot) in
rat, cynomolgus monkey and human liver microsomes, along
with a short half life (<10 minutes) confirmed by in vivo
studies in the rat (Table 3). Anticipating a possible sensitivity
of the azetidine ring toward metabolic degradation, this series
was no longer pursued.
The pipecolic series was based on a 4-hydroxy pipecolic
scaffold. Two different synthetic routes were used for this
scaffold. The first strategy was similar to the previously
described in Scheme 1 for the 4-hydroxy proline scaffold and
was used to synthesize compound (27) (Scheme 3). The second
route involved coupling the quinoline motif as the first step of
the synthesis then building-up the macrocycle stepwise to
finish using the ring closing metathesis reaction to yield the
final inhibitor. This route was less convergent but represented
a possible strategy that could be used in a scale-up process.
This synthetic route has been successfully used for the
synthesis of compound (31a) and (8) (Scheme 4).
We investigated the 3-hydroxy-pyrrolidine as a possible
mimic keeping the urea motif within the macrocycle.
Molecular modeling of this series indicated that canonical
hydrogen bonds between the amide moieties of the inhibitor
and the NS3 main chain would be maintained. The first
compound synthesized bore an 8-fluorine on the quinoline
moiety, compound (16c). The compound exhibited excellent
biochemical activity (IC50=3.5 nM) but moderate inhibitory

activity in a luciferase-replicon cell based assay (EC50=538
nM) (Table 1).
sulfonamide oxygen atoms is involved in the hydrogen
bonding interaction with G137 and S139 in the oxyanion hole
of the serine protease. In this example, the adjacent carbonyl is
also involved in hydrogen bonding with G137 and S139 main
chain thereby forming a strong hydrogen bond network. A
remarkable difference between our modeling and the crystal
structure is the orientation of K136; the alkyl side chain is
overlapping the macrocycle with the terminal amine making a
hydrogen bond interaction with the urea carbonyl. K136 is
acting like a claw forming a well defined pocket with the
oxyanion hole and the catalytic triad in which the macrocyclic
moiety of the inhibitor is making strong main chain and side
chain interactions. Having such an intricate network of
interactions on the macrocyclic moiety, we decided to turn our
attention to the quinoline heterocyclic region. The quinoline
motif is parallel and above the axis formed by the δ-NH of
Compound (16c) was selected for co-crystallization with the
NS3/4A protein and rapidly yielded suitable crystals. The
structure was solved and refined to a final resolution of 2.8 Å
in which the stereochemistry of all chiral centers was
2
3
confirmed. Examination of the crysta1 structure revealed that
the protease active site is extremely hydrophobic and flat
whereas the catalytic center is relatively exposed to solvent.
Nevertheless, ligand (16c) forms six specific hydrogen bonds
with the protein (Figures 1 and 2): four main chain interactions
(G137, S139, A157 and R155) and two side chain interactions
(K136 and S139). It is also possible to observe a sigma-π
interaction between H57 and one of the proline C5-hydrogen
24,25
on the ligand.
-
H57 and the COO of D81 shielding the ionic interaction from
solvent exposition by its hydrophobic nature. It is involved in a
cation-π interaction between R155 and its left-hand side
aromatic ring and in a π-π interaction with its right hand side
aromatic with H57. The thiazole ring is involved in a π-π
interaction with H57.
Figure 1. Representation of the amino acids of NS3/4A protein directly
involved in main chain or side chain interactions with compound (16c) in
the catalytic centre of the HCV protease. (PDB code: 4U01)
Figure 3. Overlay of compound (7) on the X-ray crystal structure of
NS3/4A protein complex with compound (16c) (in green).

Modulating the substitution on the quinoline ring (Table 1)
and changing the nature of the heteroaromatic ring (Table 2) at
position 2 could lead to more potent compounds. Compound
(16d) and (16b) with a chlorine or methyl group at position 8,
while maintaining a methoxy at position 7, demonstrated good
potency in the biochemical assay with 2.3 nM and 1.8 nM,
respectively, and significantly improved the inhibitory activity
of the replicon compared to the fluorine analogue (16c) by
seven to over nine fold. The increasing activity observed with
Br>Cl>F in both assays can be explained by a possible halogen
bonding interaction with the oxygen carbonyl of V78 as




θ=65°
γ=20°
d=4.03Å
H57
θ=31°
d=4.45Å

R155

A157
2.71Å
2
7,28
studied previously.
The oxygen of the methoxy group at
S139
position 7 can be directly involved in hydrogen bonding with
the arginine side chain of R155 but can also participate
indirectly to the parallel displaced cation-π interaction of R155
3
.05Å
2
9
G137
with the quinoline by inductive effect. Due to the excessive
payload of the bromine atom as compared to the gain in
activity observed, only the chlorine atom and methyl group
were selected for optimization at position 2 of the quinoline
K136
Figure 2. 2D-cartography of the interactions observed between
compound (16c) and the NS3/4A protein.
Green doted arrow: Main chain hydrogen bond interactions; Blue doted
arrow: Side chain hydrogen bond interactions; Magenta doted arrow:
Charge interactions; Orange Line: π interactions
(Table 2). The X-ray of compound (16c) revealed a close
proximity of the 2-iso-propyl-thiazole and tyrosine (Y56).
Compounds (16p-q) and (16t-x) were designed to increase
interaction with the tyrosine via σ-π or π-π interaction, the
vinyl (16w) and ethynyl (16u) and (16v) groups are the most
active substituents with subnanomolar activity in the
biochemical assay and low nanomolar activity in the cell-based
26
Rönn et al emphasized the importance of both sulfonyl
oxygens for excellent enzymatic and cell based potencies. It
can be seen from this crystal structure that one of the

assay. The carbonitrile group in compound (16x), being less
electron rich than its ethynyl analogue in compound (16t),
probably has a reduced interaction with the Y56 resulting in a
lower binding affinity with an IC50 value of 10 nM and an EC50
value of 89 nM. Changing the nature of the heteroatom to a
pyrazole (16m), a 1,2,3-triazole (16o), isoxazole (16i) or
oxadiazole group (16s) and therefore lowering the electron
density on the ring was detrimental to the activity. The 1,2,3-
triazole (16o), isoxazole (16i) or oxadiazole (16s) groups could
also affect cell permeability as seen with the fold change
increase between the biochemical assay and cell-based assay
The crystal structure of compound (16c) was the basis for
parallel research on the pipecolic series where the P2
pyrrolidine moiety was replaced with a 4-hydroxy pipecolic
acid moiety (Figure 3). The pipecolic motif has been seen
previously in peptidomimetic structures such as the HIV
32
protease inhibitor, palinavir , which demonstrated good
33
pharmacokinetic in the rat. Modelling this substitution with
the trifluoromethyl moiety on the thiazole ring demonstrated a
good overall fit with the X-ray structure of compound (16c).
The main chain and side chain hydrogen bond interactions
were maintained along with the σ-π interaction between H57
and one of the pipecolic C5-hydrogen, plus the cation-π
interactions of thiazole and quinoline ring with R155.
Encouraged by these findings, a few analogues with the best
substitution identified in the 3-hydroxy-pyrrolidine series were
synthesised to validate the binding hypothesis and evaluate the
potential of this new series. The pipecolic motif series yielded
extremely potent compounds (27), (31a) and (8) with sub
nanomolar activity both in the enzymatic and the cellular
inhibition assays (Table 4). These results validated the
modelling hypothesis and binding energies calculated of
-93kcal/mol for compounds (16c) as compared to the more
favourable binding energies calculated of -108.9kcal/mol for
(>80). It is interesting to observe that in the co-crystal structure
of ligand (16c) with the NS3/4A protein and in other co-crystal
structures generated (data not shown), the thiazole ring was
always orientated in such that the sulfur atom of the thiazole
0
ring could be engaged in a stabilizing intramolecular n →σ*
non-bonding interaction with the quinoline nitrogen. This type
of interaction between sulfur and heteroatom has been
previously observed and is believed to be important in
stabilizing the orientation of the two ring system to one
another, therefore lowering the entropic energy of the ligand
30
and favouring the overall binding energy. However in this
case, the thiazole regioisomer (16j) is slightly more potent than
34
(16d) in the enzymatic assay and equipotent in the cellular
compounds (8).
assay; this interaction does not play a significant role in the
The best compounds in each series, compounds (7) and (8),
were selected for rat PK analysis. The plasma kinetics, oral
bioavailability and liver tissue distribution in male Sprague-
Dawley rats were determined after a single oral (po)
administration of 10 mg/kg of the compounds using the
sodium salt of (7) in a VitamineE-TPGS-PEG400-buffer pH
9.25 (2.5/17.5/80%) solution or the parent compound (8) in
VitamineE-TPGS-PEG400-buffer pH 7.4 (2.5/17.5/80%)
solution as vehicle. These data were analyzed and compared to
those obtained after a single intravenous (iv) administration of
2 mg/kg of either compound prepared under the same
conditions as above. The plasma levels were determined up to
8h post-administration iv and up to 24h post-administration po.
The mean maximum concentration was obtained for both
compounds after 1 hour, indicating a rapid absorption (Table
5). The clearance (Cl) was lower for (7) compared to (8) and
was associated with a longer mean half-life (t1/2), yet
compound (8) yielded better oral absorption and higher area
under the curve (AUC) concentrations that translated to a
higher oral bioavailability (%F).
overall binding energy of the protease inhibitor. The steric
3
1
isostere
of the isopropyl thiazole ring bearing
a
trifluroromethyl group (7) was the best compound from this
series with an IC50 value of 0.5 nM, an EC50 value of 13 nM
and a CC50 value of >75 µM.
A few compounds from this series were selected to assess
their stability in different microsome species. Replacing the
2,4-disubstituted azetidine motif in the P2 region by the 3-
hydroxy-pyrrolidine as a possible structural mimic was
beneficial to microsomal stability with compounds (16d) and
(16b) having significant lower clearance in rat, cynomolgus
monkey and human liver microsomes along with a longer half
life in rat liver microsomes compared to compounds (5) and
(6). The clearance obtained in rat liver microsomes and rat
cryopreserved hepatocytes indicated a primary oxidative
metabolism process for this series as the main route of
elimination rather than a conjugative process.
The clearance in human liver microsomes remained
moderate for this series, but demonstrated more sensitivity in
monkey liver microsomes (Table 3).
Table 1. Optimization of the substitution on the quinoline ring
6
5
4
3
a
b
c
Compound
R
R
R
R
IC50 (µM)
IC50 SD
±0.0008
±0.0009
±0.0004
±0.0008
±0.0031
EC50 (µM)
EC50 SD
±0.027
±0.008
±0.073
±0.003
±0.009
CC50 (µM)
CC50 SD
±7.042
-
1
1
1
1
1
6a
6b
6c
6d
6e
H
H
H
H
H
H
OCH
OCH
OCH
OCH
Cl
3
3
3
3
H
0.0017
0.0018
0.0035
0.0023
0.0086
0.508
0.072
0.538
0.056
0.191
42.70
>75
H
CH
F
3
H
33.52
48.48
>75
±3.048
±7.068
-
H
Cl
H
OCH
3

1
6f
H
H
OCH
3
Br
0.0005
±0.0003
0.029
±0.003
51.18
±8.016
a
Biochemical FRET assay for inhibition capacity on the full length 1b Con1 HCV NS3/4A protease in triplicate.
b
Inhibition of HCV replication Huh-7 cells containing HCV Con1 subgenomic replicon (GS4.1 cells) with a luciferase read-out in triplicate.
EC50 values were determined from the 50% inhibition versus concentration data.
c
CC50’s were calculated as the concentration that caused a 50% of cells death versus concentration data.
TM
Cleavage of a synthetic peptide by purified, recombinant HCV NS3/4A protease from genotype 1b was measured with the SensoLyte 620 HCV Protease Assay
kit from AnaSpec, Inc. (San Jose, CA). Luciferase-replicon cells (Con1, genotype 1b) were seeded onto 96-well plates, cultured for 3 days in the presence of
compound and subjected to a luciferase assay. Compound cytotoxicity was measured in parallel using a colorimetric proliferation assay. IC50, EC50 and CC50
values were determined from % inhibition (IC50 and EC50) or % cytotoxicity (CC50) versus concentration data using a sigmoidal non-linear regression analysis
based on four parameters.
Table 2. Changing the nature of the heteroaromatic ring at position 2 of the quinoline
1
2
3
a
b
c
Compound
R
R
R
IC50 (µM)
IC50 SD
±0.0002
EC50 (µM)
EC50 SD
±0.0029
CC50 (µM)
CC50 SD
-
7
CH
H
3
CH
Cl
3
0.0005
0.0010
0.013
0.021
>75
1
1
1
6g
6h
6i
±0.0001
±0.0001
±0.0021
±0.0069
±0.0010
±0.0741
31.25
±2.934
CH
CH
3
3
Cl
Cl
0.0013
0.0037
0.029
0.669
>75
>75
-
-
1
6j
CH
3
3
Cl
0.0008
±0.0001
0.032
±0.0059
30.21
±2.865
1
1
1
6k
6l
CH
H
Cl
0.0010
0.0009
0.0009
±0.0001
±0.0003
±0.0003
0.019
0.008
0.065
±0.0058
±0.0024
±0.0051
59.00
>75
±4.963
-
CH
CH
3
3
6m
CH
3
9.61
±1.365
1
1
6n
6o
CH
CH
3
3
Cl
Cl
0.0004
0.0073
±0.0001
±0.0016
0.027
0.593
±0.0028
±0.1279
21.92
28.54
±1.927
±0.537
1
1
6p
6q
CH
CH
3
3
CH
CH
3
3
0.0064
0.0070
±0.0012
±0.0010
0.499
0.085
±0.0577
±0.0115
55.80
>75
±1.151
-
1
1
6r
6s
H
H
Cl
0.0010
0.0257
±0.0001
±0.0063
0.022
±0.0088
-
>75
>75
-
-
CH
3
>0.750
1
1
1
6t
CH
CH
H
3
3
Cl
0.0005
0.0009
0.0009
±0.0001
±0.0001
±0.0001
0.004
±0.0005
±0.0013
±0.0010
17.04
>75
±6.644
-
6u
6v
CH
CH
3
3
0.0027
0.0028
14.41
±0.815
1
1
a
6w
6x
CH
CH
3
3
Cl
Cl
0.0008
0.010
±0.0001
±0.0006
0.004
0.089
±0.0008
±0.0114
57.12
>75
±2.475
-
Biochemical FRET assay for inhibition capacity on the full length (1b) Con1 HCV NS3/4A protease in triplicate.
b
Inhibition of HCV replication Huh-7 cells containing HCV Con1 subgenomic replicon (GS4.1 cells) with a luciferase read-out in triplicate
EC50 values were determined from the 50% inhibition versus concentration data
c
CC50’s were calculated as the concentration that caused a 50% of cells death versus concentration data
Table 3. Mean clearance (n=2) observed in different species liver microsomes and cryopreserved hepatocytes.
Compound
Rat Liver
Microsomes Clint
Rat Liver
Microsomes t1/2
Rat Cryopreserved
Hepatocytes Clint
Human Liver
Microsomes Clint
Monkey Liver
Microsomes Clint

6
(
µl/min/mg)
(min)
3.8
8
(µl/min/10 cells)
(µl/min/mg)
299.1
471.9
24
(µl/min/mg)
500
5
6
7
1
1
1
364.1
182.5
11.4
12.6
9.7
N/D
N/D
17.4
12.3
11.2
30
500
122
112
143
59
46.7
6b
6d
6l
9.8
37.8
18.5
66.8
23.7
27.1
39.9
N/D: not determined
Table 4. Structure activity of the pipecolic scaffold
1
2
a
b
c
Compound
R
R
IC50 (µM)
IC50 SD
±0.0001
EC50 (µM)
EC50 SD
±0.0001
CC50 (µM)
CC50 SD
-
2
3
7
CH
3
3
0.0006
0.0002
>75
1a
H
0.0004
0.0005
±0.0001
±0.0001
0.0009
0.0007
±0.0001
±0.0002
>75
-
8
CH
52.78
±0.299
a
Biochemical FRET assay for inhibition capacity on the full length (1b) Con1 HCV NS3/4A protease in triplicate.
b
Inhibition of HCV replication Huh-7 cells containing HCV Con1 subgenomic replicon (GS4.1 cells) with a luciferase read-out in triplicate.
EC50 values were determined from the 50% inhibition versus concentration data.
c
CC50’s were calculated as the concentration that caused a 50% of cells death versus concentration data.
Table 5. male Sprague-Dawley rats mean (n=3) plasma levels and pharmacokinetic of compounds (7) and (8) analysis
Compound Route
Dose
C
max
T
max
AUC
t
AUCinf
(ng.h/mL)
2572
t
1/2
Cl
F
Liver to plasma ratio
concentration at 24h
(mg/kg)
(ng/mL)
2674
(h)
(ng.h/mL)
2537
(h)
1.5
5.2
(L/h/kg)
(%)
/
Iv
2
0.08
1.0
0.79
/
7
po
10
406
2227
2377
/
18%
N/C
iv
2
2064
934
0.08
1.0
1751
3915
1773
3966
1.5
3.6
1.13
/
/
8
po
10
/
45%
61.7
N/C: not calculable; Oral bioavailability (F%)
a
Table 6. Phamacokinetic parameters of compound (8) in mice and monkeys administered a single 2 mg/kg dose
b
Species
Route
C
max
C
24h
T
max
AUC
Vd
t
1/2
Cl
F
(
ng/mL)
(ng/mL)
7.9±1.1
1.5±0.3
(h)
N/A
2
(ng.h/mL)
4060±356
1410±86
(L/kg)
4.4±0.3
N/A
(h)
(L/h/kg)
0.48±0.04
N/A
(%)
n/a
34.7
iv
N/A
174
6.3±1.0
N/C
CD-1 Mouse
po
iv
N/A
66±26
81±47
N/A
1.0
9580±1420
8920±2190
3.5±0.5
N/A
10.3±1.0
7.6±2.0
0.23±0.03
N/A
N/A
Cynomolgus
Monkey
po
1030 ±270
107±40
a
vehicle: 70% PEG300/30%D5W for IV dose and PEG400 for oral dose
b
AUC 0-24h for mouse and AUCinf for monkey
N/A: not applicable; N/C: not calculable; Oral bioavailability (F%)

Due to the specific replication of the virus in the liver, the
liver to plasma ratio was measured. While the liver level of
compound (7) was undetectable at 24h, the liver level of
compound (8) was detectable and the liver to plasma ratio
concentration was 61.7. Based on this promising in vivo data,
compound (8) was further characterised both in vitro and in
vivo. The macrocycle (8) inhibited NS3/4A proteases from
genotypes 1a, 1b, 2a and 4a (IC50 values from 0.8 to 1.9 nM),
as well as from genotype 3a (IC50 value of 23 nM). Also,
compound (8) did not inhibit nine tested cellular proteases
2.
Hoofnagle, J. H.; Seeff, L. B. N. Engl. J. Med. 2006, 355, 2444.
Azvolinsky, A. Nat. Biotechnol. 2015, 33, 575.
Chary, A.; Holodniy, M. Rev. Recent Clin. Trials 2010, 5, 158.
Morikawa, K.; Lange, C. M.; Gouttenoire, J.; Meylan, E.;
Brass, V.; Penin, F.; Moradpour, D. J. Viral Hepat. 2011, 18,
305.
3
4
5
.
.
.
6.
Reiser, M.; Timm, J. Expert Rev. Anti Infect. Ther. 2009, 7,
5
37.
7
8
.
.
Chen, K. X.; Njoroge, F. G. Curr. Opin. Investig. Drugs Lond.
Engl. 2000 2009, 10, 821.
Forestier, N.; Zeuzem, S. Expert Opin. Pharmacother. 2012,
13, 593.
Rizza, S. A.; Talwani, R.; Nehra, V.; Temesgen, Z. Drugs
Today 2011, 47, 743.
(IC50 > 10 µM; data not shown), indicating high selectivity.
9.
Favourable oral bioavailability of (8) was observed in mouse
and monkey species, with substantial plasma concentrations
observed 24 h after a single 2 mg/kg oral dose (Table 6). Low
clearance (less than 9% of hepatic blood flow) and relatively
long plasma half-lives in both the mouse and monkey along
1
1
0. For the latest update on the best practise for HCV treatment,
please refer to: http://www.hcvguidelines.org.
1. Llinàs-Brunet, M.; Bailey, M. D.; Ghiro, E.; Gorys, V.;
Halmos, T.; Poirier, M.; Rancourt, J.; Goudreau, N. J. Med.
Chem. 2004, 47, 6584.
-6
1
2. Llinàs-Brunet, M.; Bailey, M.; Fazal, G.; Goulet, S.; Halmos,
T.; Laplante, S.; Maurice, R.; Poirier, M.; Poupart, M.-A.;
Thibeault, D.; Wernic, D.; Lamarre, D. Bioorg. Med. Chem.
Lett. 1998, 8, 1713.
with a high permeability (Papp=1.8x10 cm/s) and a low
efflux (ER=2.7) in Caco-2 cell monolayers. The selected
compound was not cytotoxic to fresh mouse, rat, monkey and
human hepatocytes, with CC50 values > 10 µM. Compound (8)
showed no significant inhibition of human CYP450 1A2, 2B6,
13. Rancourt, J.; Cameron, D. R.; Gorys, V.; Lamarre, D.; Poirier,
M.; Thibeault, D.; Llinàs-Brunet, M. J. Med. Chem. 2004, 47,
2
511.
2C9, 2D6, 3A4 or human UGT1A1 (IC50 ≥ 10 µM).
1
4. Tsantrizos, Y. S.; Bolger, G.; Bonneau, P.; Cameron, D. R.;
Goudreau, N.; Kukolj, G.; LaPlante, S. R.; Llinàs-Brunet, M.;
Nar, H.; Lamarre, D. Angew. Chem. Int. Ed. 2003, 42, 1356.
Additionally, the cardiac hERG potassium channel current in
HEK293 cells was not affect by (8). At single oral doses up to
250 mg/kg, compound (8) had no effect on the cardiovascular
15. Hinrichsen, H.; Benhamou, Y.; Wedemeyer, H.; Reiser, M.;
Sentjens, R. E.; Calleja, J. L.; Forns, X.; Erhardt, A.; Crönlein,
J.; Chaves, R. L.; Yong, C.-L.; Nehmiz, G.; Steinmann, G. G.
Gastroenterology 2004, 127, 1347.
and respiratory systems of monkeys or on the central nervous
system, renal function or gastrointestinal motility of mice. (8)
demonstrated no genotoxicity in the bacterial mutation,
lymphocyte chromosomal aberration and mouse micronucleus
tests. In 4-week GLP toxicology studies in mice and monkeys,
the compound was well tolerated and all in-life and post-
mortem parameters were generally unremarkable at oral doses
up to 250 mg/kg/day. Those data supported the potential for
once-daily dosing in patients.
1
6. Raboisson, P.; de Kock, H.; Rosenquist, Å.; Nilsson, M.;
Salvador-Oden, L.; Lin, T.-I.; Roue, N.; Ivanov, V.; Wähling,
H.; Wickström, K.; Hamelink, E.; Edlund, M.; Vrang, L.;
Vendeville, S.; Van de Vreken, W.; McGowan, D.; Tahri, A.;
Hu, L.; Boutton, C.; Lenz, O.; Delouvroy, F.; Pille, G.;
Surleraux, D.; Wigerinck, P.; Samuelsson, B.; Simmen, K.
Bioorg. Med. Chem. Lett. 2008, 18, 4853.
1
7. Parsy, C.; Alexandre, F.-R.; Brandt, G.; Caillet, C.; Cappelle,
S.; Chaves, D.; Convard, T.; Derock, M.; Gloux, D.; Griffon,
Y.; Lallos, L.; Leroy, F.; Liuzzi, M.; Loi, A.-G.; Moulat, L.;
Musiu, C.; Rahali, H.; Roques, V.; Seifer, M.; Standring, D.;
Surleraux, D. Bioorg. Med. Chem. Lett. 2014, 24, 4444.
Investigation of two series of P2 analogues that mimic the
β-strand using 4-hydroxy-2-pyrrolidine acid and 4-hydroxy
pipecolic acid moieties led to the discovery of potent inhibitors
with improved metabolic stability compared to the 2,4-
disubsituted azetidine series. X-ray co-crystal structures of the
protein helped us to model the key interactions of the inhibitor
and to rationalize the design of 4-hydroxy pipecolic acid series.
This series with excellent potency, multi-genotypic activity in
addition to good pharmacokinetic profiles in rodent and non
human primate led to the selection of (8) (IDX320) as an HCV
18. Arumugasamy, J.; Arunachalam, K.; Bauer, D.; Becker, A.;
Caillet, C. A.; Glynn, R.; Latham, G. M.; Lim, J.; Liu, J.;
Mayes, B. A.; Moussa, A.; Rosinovsky, E.; Salanson, A. E.;
Soret, A. F.; Stewart, A.; Wang, J.; Wu, X. Org. Process Res.
Dev. 2013, 17, 811.
19. Hoveyda, A. H.; Gillingham, D. G.; Veldhuizen, J. J. V.;
Kataoka, O.; Garber, S. B.; Kingsbury, J. S.; Harrity, J. P. A.
Org. Biomol. Chem. 2004, 2, 8.
3
5
20. Zhao, Z.; Zhan, Z.-Y. J. Chim. Oggi 2006, 24, 8.
drug candidate for clinical development.
2
2
1. Parsy, C.; Alexandre, F.-R.; Bonnaterre, F.; Surleraux, D.
WO2010090976A1 2010.
Acknowledgments
2. Alexandre, F.-R.; Brandt, G.; Caillet, C.; Chaves, D.; Convard,
T.; Derock, M.; Gloux, D.; Griffon, Y.; Lallos, L.; Leroy, F.;
Liuzzi, M.; Loi, A.-G.; Moulat, L.; Musiu, C.; Parsy, C.;
Rahali, H.; Roques, V.; Seifer, M.; Standring, D.; Surleraux, D.
Bioorg. Med. Chem. Lett., DOI information:
We thank Proteros Biostructure GmbH, Dr. Christine
Wenzkowski and Dr. Stefan Steinbacher for X-Ray structure
determination of HCV NS3/4A protease in complex with the
ligand (16c).
1
0.1016/j.bmcl.2015.07.020.
2
3. Co-Crystallization: The NS3/4a protein was recombinantly
expressed in an E. coli expression system. An apparent
homogeneity of more than 95% as judged by Coomassie-
stained SDS-PAGE was achieved after a purification procedure
comprising affinity, ion exchange and gel filtration
Supplementary data
Supplementary data associated (detailed characterization of
compound (8) (IDX320)) with this article can be found in the
online version.
chromatography steps. Crystals of the HCV NS3/4A ligand
complex were prepared by the method of co-crystallisation.
Diffraction data of the HCV NS3/4A ligand complex
containing ligand (16c) were collected at the SWISS LIGHT
SOURCE (SLS, Villigen, Switzerland). The structure was
solved and refined to a final resolution of 2.8 Å (PDB code:
References and notes:
4
U01). The crystal of space group C 2221 contains nine
1
.
Marubashi, S.; Dono, K.; Miyamoto, A.; Takeda, Y.; Nagano,
H.; Umeshita, K.; Monden, M. J. Hepatobiliary. Pancreat.
Surg. 2006, 13, 382.
monomers of HCV NS3/4A in the asymmetric unit. The
resulting electron density shows an unambiguous binding mode
for ligand 16c, including the orientation and conformation of

the ligand. The ligand shows high occupancy as judged by
temperature factors of the atomic model.
2
4. Marsman, A. W.; Havenith, R. W. A.; Bethke, S.; Jenneskens,
L. W.; Gleiter, R.; van Lenthe, J. H.; Lutz, M.; Spek, A. L. J.
Org. Chem. 2000, 65, 4584.
2
2
5. Singh, R. K.; Tiwari, M. K.; Kim, I.-W.; Chen, Z.; Lee, J.-K.
Appl. Environ. Microbiol. 2012, 78, 8817.
6. Rönn, R.; Sabnis, Y. A.; Gossas, T.; Åkerblom, E.; Helena
Danielson, U.; Hallberg, A.; Johansson, A. Bioorg. Med. Chem.
2
006, 14, 544.
2
2
7. Lu, Y.; Shi, T.; Wang, Y.; Yang, H.; Yan, X.; Luo, X.; Jiang,
H.; Zhu, W. J. Med. Chem. 2009, 52, 2854.
8. Lemke, C. T.; Goudreau, N.; Zhao, S.; Hucke, O.; Thibeault,
D.; Llinàs-Brunet, M.; White, P. W. J. Biol. Chem. 2011, 286,
1
1434.
9. Imai, Y. N.; Inoue, Y.; Yamamoto, Y. J. Med. Chem. 2007, 50,
189.
0. Beno, B. R.; Yeung, K.-S.; Bartberger, M. D.; Pennington, L.
D.; Meanwell, N. A. J. Med. Chem. 2015, 58, 4383.
2
3
1
3
3
1. Smart, B. E. J. Fluor. Chem. 2001, 109, 3.
2. Beaulieu, P. L.; Wernic, D.; Abraham; Anderson, P. C.; Bogri,
T.; Bousquet, Y.; Croteau, G.; Guse, I.; Lamarre, D.; Liard, F.;
Paris, W.; Thibeault, D.; Pav, S.; Tong, L. J. Med. Chem. 1997,
4
0, 2164.
3. Liard, F.; Jaramillo, J.; Paris, W. L.; Yoakim, C. J. Pharm. Sci.
998, 87, 782.
4. The binding energies were calculated using the Discovery
Studio 4.0 suite from BIOVIA.
3
3
3
1
5. During a drug-drug interaction study of the combination of
IDX184 (nucleoside polymerase inhibitor) and IDX320
(protease inhibitor) in healthy volunteers, three serious adverse
events occurred. These observed serious adverse events were
elevated liver function tests detected in three subjects during
post exposure safety visits. The cause of these serious adverse
events in the combination study of IDX184 and IDX320 was
further attributed to IDX320 and the clinical development of
this drug was halted.