Quinoxaline-based inhibitors of Ebola and Marburg VP40 egress

Article abstract of DOI:10.1016/j.bmcl.2016.06.053

We prepared a series of quinoxalin-2-mercapto-acetyl-urea analogs and evaluated them for their ability to inhibit viral egress in our Marburg and Ebola VP40 VLP budding assays in HEK293T cells. We also evaluated selected compounds in our bimolecular complementation assay (BiMC) to detect and visualize a Marburg mVP40–Nedd4 interaction in live mammalian cells. Antiviral activity was assessed for selected compounds using a live recombinant vesicular stomatitis virus (VSV) (M40 virus) that expresses the EBOV VP40 PPxY L-domain. Finally selected compounds were evaluated in several ADME assays to have an early assessment of their drug properties. Our compounds had low nM potency in these assays (e.g., compounds 21, 24, 26, 39), and had good human liver microsome stability, as well as little or no inhibition of P450 3A4.

Full text of DOI:10.1016/j.bmcl.2016.06.053

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Quinoxaline-based inhibitors of Ebola and Marburg VP40 egress
a
b
b
H. Marie Loughran a, Ziying Han b, Jay E. Wrobel a, , Sarah E. Decker , Gordon Ruthel , Bruce D. Freedman ,
⇑
Ronald N. Harty b, Allen B. Reitz a
a Fox Chase Chemical Diversity Center, Inc, 3805 Old Easton Road, Doylestown, PA 18902, United States
b Department of Pathobiology, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA 19104, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
We prepared a series of quinoxalin-2-mercapto-acetyl-urea analogs and evaluated them for their ability
to inhibit viral egress in our Marburg and Ebola VP40 VLP budding assays in HEK293T cells. We also
evaluated selected compounds in our bimolecular complementation assay (BiMC) to detect and visualize
a Marburg mVP40–Nedd4 interaction in live mammalian cells. Antiviral activity was assessed for selected
compounds using a live recombinant vesicular stomatitis virus (VSV) (M40 virus) that expresses the
EBOV VP40 PPxY L-domain. Finally selected compounds were evaluated in several ADME assays to have
an early assessment of their drug properties. Our compounds had low nM potency in these assays
(e.g., compounds 21, 24, 26, 39), and had good human liver microsome stability, as well as little or no
inhibition of P450 3A4.
Received 13 May 2016
Revised 17 June 2016
Accepted 18 June 2016
Available online xxxx
Keywords:
Ebola virus
Marburg virus
RNA viruses
Virus egress
PPxY
Ó 2016 Elsevier Ltd. All rights reserved.
Nedd4
L-domain
Budding
Antiviral therapeutic
The 2014–2015 outbreak of Ebola in western Africa resulted in
over 28,000 infected individuals and over 11,000 deaths (WHO:
Ebola situation report 2015). This unprecedented epidemic has
spurred a call to action on new, cost effective therapies that
combat this deadly pathogen. Among the efforts are several vacci-
nes and antiviral candidates.1–3 However the current vaccines in
clinical trials are not a complete defense. They must be given
pre-exposure and would not be effective against other RNA viruses
such as Marburg and Lassa4 that also cause lethal hemorrhagic
fever symptoms. Furthermore, oral antiviral agents used alone or
in combination may be of value for individuals who respond
adversely to the vaccine, and could be of value as prophylactic
agents for individuals deemed to be in high risk situations such
as military or healthcare workers. Therefore, effective therapeutics
are needed to safeguard the largely immunologically naive human
population by providing immediate protection.
will greatly diminish or eliminate the occurrence of drug resistant
viral mutations. Importantly, as these virus-host interactions
represent a common mechanism in a range of RNA viruses, we
predict that they represent an Achilles’ heel in the life cycle of
RNA virus pathogens.
Late budding domains (containing PPxY and PTAP motifs) are
highly conserved in the matrix proteins of a wide array of RNA
viruses
(e.g.,
filoviruses,
arenaviruses,
rhabdoviruses,
paramyxoviruses, henipaviruses, and retroviruses) and represent
broad-spectrum targets for the development of novel antiviral
therapeutics.6–16 For example, the filovirus VP40, arenavirus Z, and
rhabdovirus M proteins play central and sufficient roles in virion
assembly and egress, due in part to the presence of a PPxY
L-domain.16–23 Efficient egress of VLPs depends on viral L-domain
mediated recruitment of host proteins required for
complete virus-cell separation or pinching-off of virus
particles.7–9,11–13,15,16,24 In this regard, the viral matrix protein
VP40 (for filoviruses Ebola and Marburg) or Z (for arenavirus Lassa)
contains a PPxY L-domain motif that recruits the mammalian
cellular protein Nedd4, which is a WW-domain containing cellular
E3 ubiquitin ligase associated with the host ESCRT1 complex
(endosomal sorting complex required for transport), and this inter-
action is critical for efficient budding of filoviruses, arenaviruses,
and rhabdoviruses.9,11,15,16,18,22,24–34
We have discovered two novel series of small molecule early
leads that inhibit RNA virus budding.5 Our approach does not rely
solely on viral targets, but instead focuses on a critical virus-host
interaction required by PPxY motif-containing viruses for efficient
egress and spread. We hypothesize that targeting a virus–host
interaction necessary for efficient virus egress and dissemination
⇑
Corresponding author.
http://dx.doi.org/10.1016/j.bmcl.2016.06.053
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Loughran, H. M.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.06.053

2
H. M. Loughran et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Herein we describe our current efforts to exploit the viral
was commercially available but in a few examples (R1 = CF3 and
Et) we prepared this thiol by converting congeners 54
(R1 = CF3 and Et) using P2S5 and pyridine.37 Compound 54 where
R1 = CF3 was prepared by reaction of o-phenylenediamine with
ethyl trifluoropyruvate38 and 54 where R1 = Et is commercially
available.
PPxY–host WW domain interaction to obtain broad-spectrum
RNA antiviral therapeutics. This paper focuses on SAR around our
lead 1 (Table 1) which we described in detail previously.5 We used
our Marburg VP40 VLP budding assay in HEK293T cells as our
primary assay for the SAR analogs compiled in Tables 1–3. This
assay measures the % inhibition of viral VP40 VLP egress from
the cell versus DMSO control. In addition we provide Western anal-
ysis for selected compounds using this assay and the Ebola VP40
VLP budding assay (Figs. 2 and 3). We also evaluated selected com-
pounds in our bimolecular complementation assay (BiMC)35 to
detect and visualize a Marburg VP40–Nedd4 interaction in live
mammalian cells in the absence or presence of the indicated inhi-
bitors in Figure 1. We then assessed antiviral activity for selected
compounds using a live recombinant vesicular stomatitis virus
(VSV-M40 virus) that expresses the EBOV VP40 PPxY L-domain
and flanking residues36 in Figure 4. Relative cell viability was vali-
Several of the target compounds in Table 1 were prepared by
alternative routes highlighted in Scheme 2. For compounds with
X = O (e.g., 9 and 10), we started with commercially available
2-chloro-3-methyl-quinoxaline (58) since reaction of compounds
54 with 57 led to N-alkylation products. Substitution of the chlo-
rine of 58 with methyl glycolate, and subsequent saponification
of the methyl ester led to acid 59. Conversion to the primary amide
60 and subsequent reaction with the requisite aryl isocyanate39,40
led to target compounds 9 and 10. Preparation of N-methylated
target compound 12 (R2 = CH3) was accomplished by alkylation
of 20 with methyl iodide. Thioether 20 was also used to prepare
sulfone analog 8 via mCPBA oxidation conditions.
dated by the MTT assay at concentration ranges of 0.01–1.0 lM on
VeroE6 and HEK293T cells (data presented in Supplementary
material section). Finally selected compounds were evaluated in
several ADME assays (Table 4) to have an early assessment of their
drug properties.
We examined the X and R1–R3 substituent changes of 1 in
Table 1. With the exception of analogs 9 and 10, the compounds
2–8, 11–13 in this table had little or no inhibition of Marburg
VLP egress at 1 lM or greater. SAR highlights are summarized
Compounds 2–13 (Table 1) and 17–52 (Tables 2 and 3) were pre-
pared according to Schemes 1 and 2. While compounds 1 and 14–16
were originally purchased from Ambinter (Orléans, France), larger
quantities of 1 were synthesized by us via the methods outlined
in Scheme 1. Experimental and analytical information for
compounds 1–13 and 17–52 are described in the Supplementary
material section.
below with the full data set listed in the Supplementary section
(Table 1S). Tables 2 and 3 focused on variation of the terminal aryl
substituent of 1.
Referring to Table 1, we have not found a suitable replacement
of methyl for the R1 substituent on the quinoxaline moiety of 1. All
replacements either smaller, H (2) or larger, CF3 (3, 4), Et (5, 6) or
CH2Ph (7) were less active or not active at the 1 lM concentration
Referring to Scheme 1, target compounds in Tables 1–3 were
generally prepared by alkylation of quinoxaline thiols 55 with
in the Marburg VLP inhibition assay. The CF3 and Et groups did
provide compounds with greater stability to mouse liver micro-
somes relative to the methyl congeners however (vide infra).
We replaced the sulfur atom of compound 1 (X = S) with a SO2
moiety (8) or an O atom (9, 10). While compound 8 did not show
a-chloro-acetamidoureas 57. The alkylating agents (57) were in
turn obtained via reaction of commercially available anilines or
heteroaromatic amines 56 with commercially available
chloroacetyl isocyanate. In most cases, the quinoxaline thiol 55
activity at 1
lM, the ether analogs 9 and 10 were approximately a
Table 1
Analogs of 1. Examination of changes in highlighted areas
O
O
O
O
N
S
Ph
N
N
X
Ar
N
H
N
H
N
N
R2 R3
R1
N
CH3
Compound 1
No.
R1
R2 R3
R1
R2
H
R3
H
X
Ar
No.
X
Ar
F
2
3
H
S
S
H
H
8
9
CH3
SO2
F
CF3
H
H
CH3
O
O
H
H
H
H
CH3
Cl
CH3
4
5
6
7
CF3
Et
S
S
S
S
H
H
H
H
H
H
H
H
10
11
12
13
CH3
CH3
CH3
CH3
F
CH3
S
S
S
H
CH3
H
CH3
Cl
Et
H
F
F
CH3
NR3Ar =
F
CH2Ph
N
Please cite this article in press as: Loughran, H. M.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.06.053

H. M. Loughran et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
3
Table 2
SAR of terminal phenyl moiety of 1 with alkyl, benzyl and substituted phenyl groups
O
O
N
N
S
R
N
H
N
H
CH3
No.
R
MW
LogD (pH 7.4)
TPSA (Å2)
mVLPa (% reduction at given concentration in
lM)
1
0.3
0.1
0.03
1
Phenyl
Ethyl
Cyclopropyl
CH2-(2)-thiophenyl
Benzyl
352.4
316.4
352.4
372.5
366.4
370.4
370.4
370.4
386.9
386.9
386.9
366.4
366.4
366.4
377.4
377.4
382.4
382.4
382.4
380.5
380.5
380.5
380.5
380.5
380.5
421.3
384.4
384.4
2.9
1.4
2.9
2.5
2.6
3.1
3.1
3.1
3.5
3.5
3.5
3.4
3.4
3.4
2.8
2.8
2.8
2.8
2.8
3.9
3.9
3.9
3.9
3.9
3.9
4.1
3.6
3.6
84
84
84
84
84
84
84
84
84
84
84
84
84
84
108
108
93
93
93
84
84
84
84
84
84
84
84
84
88b
NSc
NSd
97
60
100
41
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
NS
NS
NS
29
98
2-Fluorophenyl
3-Fluorophenyl
4-Fluorophenyl
2-Chlorophenyl
3-Chlorophenyl
4-Chlorophenyl
2-Methylphenyl
3-Methylphenyl
4-Methylphenyl
2-Cyanophenyl
3-Cyanophenyl
2-Methoxyphenyl
3-Methoxyphenyl
4-Methoxyphenyl
2,3-Dimethylphenyl
2,4-Dimethylphenyl
2,5-Dimethylphenyl
2,6-Dimethylphenyl
3,4-Dimethylphenyl
3,5-Dimethylphenyl
2,5-Dichlorophenyl
3-Fluoro-2-methylphenyl
5-Fluoro-2-methylphenyl
100
49
NS
80
99e
93e
82
100
100
100
87
72
96
98f
NS
83f
NS
93
71
100
100
89
98
60
60
99
NS
NS
25
41
62
68
NS
29
44
37
86f
100
100
98
100
100
100
97
100
100f
89
a
Marburg VP40 VLP budding assay. HEK293T cells, NS: not significant versus DMSO control, a blank entry means the compound was not evaluated at the corresponding
concentration.
b
Average of twenty independent experiments at 1.0 lM.
c
d
e
f
60% reduction at 25
lM.
60% reduction at 10
lM.
Average of two independent experiments at 0.1 and 0.03
Average of four independent experiments at 0.1 and 0.03
lM concd.
l
M concd.
half log less potent than their corresponding thioether analogs 24
and 26 respectively (Table 2) and showed good inhibitory activity
at the 100 nM level (e.g., 9 showed 58% reduction of mVLPs at
100 nM and 10 showed 70% reduction of mVLPs at 100 nM). Other
general SAR results from the compounds in Table 1 reveal that
methyl substituents in place of H on the imido (R2) and amide (R3)
nitrogen atoms resulted in greatly reduced activity (compounds
11 and 12 respectively) and that the amide substituent R3 and Ar
moiety could not be tied back in a ring (compound 13).
Analogs that varied the phenyl moiety of 1 with alkyl and ara-
lkyl groups as well as the substituents on the terminal phenyl
group of 1 are shown in Table 2 with their corresponding Marburg
VLP assay data. Replacement of the phenyl moiety of 1 with alkyl
(14, 15) or aralkyl groups (16, 17) did not lead to improved potency
consistent with previous findings.5 However substituent additions
on the phenyl group of 1 led to many compounds with improved
potency. Examination of Table 2 reveals that halo and methyl
substituents particularly at the ortho and para positions [see 21
(2-Cl), 24 (2-Me) and 26 (4-Me)] showed strong activity at the
lowest concentration tested (30 nM). Compounds with either
strong electron releasing moieties (e.g., methoxy at o, m or p posi-
tions: compounds 29–31) or strong electron withdrawing groups
(e.g., cyano at o, m positions: compounds 27, 28) did not generally
result in improved potency over 1.
Addition of a second methyl group on the terminal phenyl ring
as for compounds 32–37, did not further improve potency over
monomethyl congeners 24 and 26 and this was also true of the
dichloro analog 38 when compared to mono-chloro analog 21.
The compound with a combination of a 2-methyl and 3-fluoro
group (39) did show strong potency however.
Replacement of the terminal phenyl group or
1 with a
heteroaromatic moiety is shown in Table 3. Thus far unsubstituted
and methyl/alkyl substituted 2-pyridyl (41–44), unsubstituted and
methyl substituted 3-pyridyl (45–48) and unsubstituted and
substituted pyrazole (49–52) have been investigated. The mVLP
activity is presented in the Supplementary section as Table 2S.
Generally these heterocyclic replacement compounds have not
led to improved potency increases over 1. A few exceptions are
2-pyridyl analog 43 and 3-pyridyl analogs 47 and 48 (96%, 85%
and 87% reduction in mVLP levels at 100 nM, respectively),
however these did not achieve the same magnitude of potency
increases as seen by several of the substituted phenyl analogs of
Table 2.
Our compounds have in silico properties (MW, cLogD, TPSA in
Tables 2) that are consistent with drugs delivered orally.41 We
evaluated a few of the more potent analogs in several ADMET
assays (Table 4) for a preliminary assessment of metabolic stability
to liver microsomes and inhibition of P450 3A4. The compounds
Please cite this article in press as: Loughran, H. M.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.06.053

4
H. M. Loughran et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Table 3
Analogs of 1. Replace terminal phenyl moiety with heteroaromatic groups
O
O
N
S
Ar
N
H
N
H
N
CH3
No.
41
No.
No.
49
Ar
Ar
Ar
N
NH
NH
N
45
N
N
CH3
42
43
44
46
47
48
50
51
52
CH3
N
N
N
N
N
H3C
CH3
CH3
N
N
Et
CH3
N
N
Ph
N
120
100
80
p < 0.003
***
***
60
21
40
20
0
21
0
0.1 µM
120
100
80
p < 0.001
60
***
24
39
40
20
0
24
0
0.1 µM
120
100
80
p < 0.001
60
***
40
20
0
39
0
0.1 µM
DMSO
0.1 µM
Figure 1. BiMC analysis of HEK293T cells co-expressing NYFP–Nedd4 and CYFP–mVP40 in the presence of DMSO alone, or 0.1
Representative images are shown where the total number of cells were quantified using NucBlue, and Green (YFP) indicates cells with a positive interaction between Nedd4
and mVP40. Scale bar, 200
m. YFP-positive cells were quantified in triplicate using MetaMorph software and statistical analysis as described previously.1
lM concentration of 21, 24 or 39.
l
generally had moderate to good stability against human liver
microsomes, but the 3-methyl quinoxaline analogs 21, 24, 26, 34
and 39 had poor stability against mouse liver microsomes. Since
our early antiviral animal model is the mouse, we needed to
improve stability in the mouse. This was indeed achieved by
replacing the quinoxaline 3-methyl moiety with either a tri-
fluromethyl or ethyl group (compounds 3 and 6 respectively).
Unfortunately these compounds lost significant potency, but
none-the-less provided data to guide us in the preparation of more
potent mouse liver microsome stable compounds. Thus far, none of
Please cite this article in press as: Loughran, H. M.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.06.053

H. M. Loughran et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
5
demonstrate that several of our lead inhibitors can specifically
block the PPxY-mediated interaction between mVP40 and Nedd4
in mammalian cells.
24
26
30 100
300 1000
30
100 300 1000 nM
- mVP4
Cells
VLPs
We used our validated filovirus VLP budding assay that
recapitulates live virus budding,11–13,15,17 to determine the antivi-
ral potency of our analogs in Tables 1–3. We chose the Marburg
mVP40 VLP budding assay as our primary screen because efficient
egress of mVP40 VLPs is dependent on a single PPxY L-domain
motif to recruit host Nedd4 unlike Ebola eVP40 which contains
both a PTAP and PPxY motif. Figure 2 shows a pictorial example
of this assay. Briefly HEK293T cells were transfected with an
mVP40 expression plasmid in the presence of vehicle (DMSO)
- mVP4
1
2
3
4
5
6
7
8
9
10
Figure 2. Marburg VP40 VLP budding assay. HEK293T cells were transfected with
mVP40 plasmid in the presence of DMSO alone, or the indicated inhibitor (24 or 26)
at the indicated concentrations. mVP40 was detected by Western blot in cell
extracts and VLPs at 24 h post-transfection. mVP40 was quantified using NIH
Image-J software.
alone as a negative control, 1.0 lM 1 as a positive control, or the
indicated egress inhibitors (24 and 26) in a dose-dependent
manner. Expression levels of mVP40 were detected and quantified
in cell extracts and VLPs by Western blotting and Image-J software.
As expected, budding of mVP40 VLPs was reduced by approxi-
our inhibitors have affected P450 3A4 activity at 33 lM, although a
more stringent analysis of other P450s will have to be conducted
on future analogs to best determine drug–drug interaction profiles.
We used our well established BiMC approach35 to detect and
visualize a Marburg VP40–Nedd4 interaction in live mammalian
cells in the absence or presence of the indicated egress inhibitors
(Fig. 1). Briefly, HEK293T cells were co-transfected with plasmids
expressing NYFP–Nedd4 and CYFP–mVP40 in the presence of
vehicle (DMSO) alone, or inhibitors 21, 24, or 39 at a concentration
mately 100-fold in the presence of 1.0 lM 1 compared to DMSO
control (Fig. 2, lanes 1 and 2). Importantly, inhibition of budding
of mVP40 VLPs was >90% in the presence of 10-fold lower
(100 nM) concentrations of inhibitors 24 (Fig. 2, lane 4) and 26
(lane 8). Moreover, 30 nM concentrations of both 24 and 26
resulted in an approximately 3-fold and 15-fold decrease in
mVP40 VLP budding compared to DMSO control, respectively
(Fig. 2, compare lane 1 with lanes 3 and 7).
While our initial VLP budding assays employed mVP40, we have
also tested several select analogs for antiviral activity against Ebola
virus VP40 (eVP40) budding. While eVP40 possesses a functional
PPxY-type L-domain and is thus sensitive to our PPxY budding
inhibitors, it also expresses a PTAP-type L-domain that can interact
with other host proteins. To determine if our compounds could
block egress of both mVP40 and eVP40 VLPs, we transfected
HEK293T cells with either mVP40 or eVP40 expression plasmids
in the absence (DMSO alone) of inhibitor, or in the presence of
21 at the indicated concentrations (Fig. 3). Expression levels of
mVP40 (red) and eVP40 (blue) were detected and quantified in cell
extracts and VLPs by Western blotting and Image-J software as
described previously.5 Notably, budding of both mVP40 and
eVP40 VLPs was reduced significantly at both 100 nM and 30 nM
concentrations of 21 compared to that in the DMSO control
of 0.1 lM (Fig. 1). Total cell counts based on NucBlue staining
indicated that equal numbers of cells were present in all assay
samples. YFP fluorescent complementation (i.e., green cells) is
indicative of a PPxY-mediated interaction between mVP40 and
host Nedd4, and fluorescent cells were visualized and quantified
using an inverted Leica Sp5-II confocal microscope and MetaMorph
software as described previously.5
We observed a significant decrease in the relative number of
YFP-positive cells in samples treated with either 21, 24, or 39
compared to that in vehicle alone control samples in multiple
independent experiments. As we published previously5, a PPxY
L-domain mutant of mVP40 that does not interact with Nedd4
serves as a valid negative control (data not shown). These results
120
mVP40
0.03 0.1
100
21 (µM)
0
mVP40
80
60
40
20
- mVP40
- actin
Cells
VLPs
- mVP40
0
1
2
3
21 (µM)
0
0.03
0.1
120
100
80
eVP40
0.03 0.1
21 (µM)
0
21 (nM)
21 (nM)
0
30
30
100
100
39 (nM)
0
30
100
100
eVP40
- eVP40
- actin
0
39 (nM)
0
30
Cells
60
- VSV M
- actin
- VSV M
- actin
40
20
VLPs
- eVP40
0
1
2
3
21 (µM)
0
0.03
0.1
Figure 4. Inhibition of live VSV–M40 virus budding. HEK293T cells were infected
(in triplicate) with VSV-M40 for 8 h under the indicated conditions, and virion
containing supernatants were harvested and quantified by plaque assay performed
in triplicate. Virus titers are indicated as a percent relative to control. ***Indicates a p
values <0.001 as determined by a two-tailed Student t-test. Western blots of
infected cell extracts are shown for VSV M and actin, which demonstrate that
treatment with the indicated concentrations of 21 and 39 for 8 h did not affect viral
or cellular protein levels compared to DMSO alone controls.
Figure 3. Marburg and Ebola VP40 VLP budding assays. HEK293T cells were
transfected with either mVP40 or eVP40 plasmids in the presence of DMSO (0)
alone, or budding inhibitor 21 at the indicated concentrations. mVP40 and eVP40
were detected by Western blot in cell extracts and VLPs at 24 h post-transfection.
mVP40 and eVP40 were quantified using NIH Image-J software. Bar graphs
represent the average of three independent experiments.
Please cite this article in press as: Loughran, H. M.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.06.053

6
H. M. Loughran et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Table 4
tested with more dangerous BSL-4 pathogens (collaboration at
ADMET data for select compounds
USAMRIID). Indeed, we published recently that another egress
inhibitor from different chemical series significantly decreased
PPxY-mediated budding of both VSV-WT and VSV-M40.5 Here we
demonstrate that analogs 21 and 39 effectively block budding of live
VSV-M40 from HEK293T cells (Fig. 4). Briefly, cells were infected
with VSV-M40 at a low multiplicity of infection (MOI) of 0.1 in the
presence of DMSO alone or at the indicated concentrations of 21 or
39. Virion-containing supernatants were harvested at 8 h post-
infection; the peak time for virus budding (Fig. 4). Equivalent
amounts of the VSV M protein and cellular actin were detected by
Western blot to demonstrate that the budding inhibitors had no
detectable effects on viral protein synthesis (Fig. 4). Intriguingly,
both 21 and 39 inhibited budding of VSV-M40 by approximately
5-fold at 30 nM and by >10-fold at 100 nM compared to DMSO
controls in four independent experiments (Fig. 4). These findings
correlate well with BiMC and VLP budding assays that demonstrate
specific and potent inhibition of PPxY-mediated virus budding.
To summarize, our analogs show:
Compd
Microsome stability
P450 inhib
IC50 M)
(l
Mouse
Human
t1/2
Clint (mL/min/
t1/2
Clint (mL/min/ 3A4
(min) mg protein)
(min) mg protein)
(midazolam)
1
ND
2.0
3.0
<1.4
1.9
3.5
49.9
54.8
ND
43.9
76.1
46.6
57.7
77.2
>90
1.66
0.72
1.18
0.95
0.71
0.54
0.77
1.76
ND
21
24
26
34
39
3
34.1
22.4
>50
35.1
18.8
1.3
>33
>33
>33
>33
>33
>33
ND
71.7
31.3
6
1.2
O
P2S5
pyridine
80 °C
R1
H
N
OEt
NH2
N
SH
R1
O
O
N
R1
50-70% yield
NH2
EtOH, reflux
70-90% yield
N
53
55
54
ꢀ improved potency as RNA viral egress inhibitors by >30 fold over
initial lead 1 (e.g., compounds 21, 24, 26, 39);
ꢀ low nanomolar functional activity (VLP assay);
ꢀ low nM target engagement (inhibition of Nedd4/Marburg PPxY
binding in cellular BiMC assay);
O
O
O
O
H
N
O
Ar
Cl
C
55
N
N
N
S
Ar
H
N
N
N
H
N
R3
Cl
R3
Ar
R3
DCM, rt
O
EtOH, NaOAc
R1
80 °C, 10-100% yield
90-100%
57
56
1-7, 11, 17-52
ꢀ inhibition of live virus egress (VSVM40);
ꢀ good human microsome stability and no inhibition of P450 3A4
Scheme 1. General preparation of target compounds.
at 33 lM concentration;
ꢀ little to no cytotoxicity at effective antiviral concentrations.
O
O
1) CDI, THF, rt
2) aq NH4OH
Live virus testing with infectious BSL-4 viruses (Ebola, Marburg,
Junin, Lassa) of these and other analog series of compounds we are
developing is ongoing (USAMRIID) and will be the subject of future
reports.
HO
N
Cl
N
O
OCH3
OH
O
K2CO3
N
58
CH3
N
59
CH3
100% yield
DMF/ H2O, 60 °C
50% yield
O
O
O
O
Acknowledgments
N
O
ArNCO
NH2
N
N
Ar
N
H
N
H
toluene
reflux
N
CH3
This project was supported by National Institutes of Health
(NIH): National Institute of Allergy and Infectious Diseases (NIAID)
grant R41AI113952 and AI102104. We thank C. Berry for help with
statistical analysis, and D. Lyles, S. Urata, and S. Becker for
generously providing reagents.
CH3
60
9,10
30-40% yield
F
F
O
O
MeI
O
K2CO3
DMF, rt
N
S
N
N
S
N
H
N
H
N
CH3
N
H
N
CH3
39% yield
CH3
20
20
12
Supplementary data
F
O
O
O
mCPBA
O
DCM, 0°C
N
N
S
N
H
N
H
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.06.
053.
15% yield
CH3
8
Scheme 2. Preparation of certain specific target compounds.
References and notes
samples (Fig. 3). These results were consistent in multiple indepen-
dent budding assays (Fig. 3, bar graphs), and highlight the
enhanced antiviral potency and broad-spectrum nature of these
inhibitors.
While the use of BiMC and VLP budding assays as our initial
screening assays continues to be instrumental in directing us
toward novel and potent budding inhibitors, the ultimate test of
efficacy is their ability to inhibit infectious (live) virus budding.
Toward this end, we have tested several lead candidates for their
ability to block budding of VSV and/or our previously described
live VSV recombinant virus (M40 virus) that expresses the eVP40
PPxY L-domain and flanking residues.24
1. Li, H.; Ying, T.; Yu, F.; Lu, L.; Jiang, S. Microbes Infect. 2015, 17, 109.
2. De Clercq, E. Biochem. Pharmacol. 2015, 93, 1.
3. Agnandji, S. T.; Huttner, A.; Zinser, M. E.; Njuguna, P.; Dahlke, C.; Fernandes, J.
F.; Yerly, S.; Dayer, J. A.; Kraehling, V.; Kasonta, R.; Adegnika, A. A.; Altfeld, M.;
Auderset, F.; Bache, E. B.; Biedenkopf, N.; Borregaard, S.; Brosnahan, J. S.;
Burrow, R.; Combescure, C.; Desmeules, J.; Eickmann, M.; Fehling, S. K.; Finckh,
A.; Goncalves, A. R.; Grobusch, M. P.; Hooper, J.; Jambrecina, A.; Kabwende, A.
L.; Kaya, G.; Kimani, D.; Lell, B.; Lemaître, B.; Lohse, A. W.; Massinga-Loembe,
M.; Matthey, A.; Mordmüller, B.; Nolting, A.; Ogwang, C.; Ramharter, M.;
Schmidt-Chanasit, J.; Schmiedel, S.; Silvera, P.; Stahl, F. R.; Staines, H. M.;
Strecker, T.; Stubbe, H. C.; Tsofa, B.; Zaki, S.; Fast, P.; Moorthy, V.; Kaiser, L.;
Krishna, S.; Becker, S.; Kieny, M.-P.; Bejon, P.; Kremsner, P. G.; Addo, M. M.;
Siegrist, C.-A. N. Engl. J. Med. 2015.
4. Ogbua, O.; Ajuluchukwub, E.; Uneke, C. J. J. Vector Borne Dis. 2007, 44, 1.
5. Han, Z.; Lu, J.; Liu, Y.; Davis, B.; Lee, M. S.; Olson, M. A.; Ruthel, G.; Freedman, B.
D.; Schnell, M. J.; Wrobel, J. E.; Reitz, A. B.; Harty, R. N. J. Virol. 2014, 88, 7294.
6. Adamson, C. S.; Freed, E. O. Antiviral Res. 2010, 85, 119.
7. Bieniasz, P. D. Virology 2006, 344, 55.
8. Calistri, A.; Salata, C.; Parolin, C.; Palu, G. Rev. Med. Virol. 2009, 19, 31.
Importantly, the M40-VSV recombinant virus serves as an Ebola
surrogate that we can use safely under BSL-2 conditions to rapidly
assess the antiviral potency of our inhibitors before having them
Please cite this article in press as: Loughran, H. M.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.06.053

H. M. Loughran et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
7
9. Chen, B. J.; Lamb, R. A. Virology 2008, 372, 221.
10. Chen, H.; Liu, X.; Li, Z.; Zhan, P.; De Clercq, E. Curr. Med. Chem. 2010, 17, 750.
11. Hartlieb, B.; Weissenhorn, W. Virology 2006, 344, 64.
12. Harty, R. N. Antiviral Res. 2009, 81, 189.
28. Pasqual, G.; Rojek, J. M.; Masin, M.; Chatton, J. Y.; Kunz, S. PLoS Pathog. 2011, 7,
e1002232.
29. Perez, M.; Craven, R. C.; de la Torre, J. C. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
12978.
13. Jasenosky, L. D.; Kawaoka, Y. Virus Res. 2004, 106, 181.
14. Jiang, Y.; Liu, X.; De Clercq, E. Curr. Med. Chem. 2011, 18, 16.
15. Liu, Y.; Harty, R. N. Future Virol. 2010, 5, 481.
16. Urata, S.; de la Torre, J. C. Adv. Virol. 2011, 2011, 180326.
17. Harty, R. N.; Brown, M. E.; Wang, G.; Huibregtse, J.; Hayes, F. P. Proc. Natl. Acad.
Sci. U.S.A. 2000, 97, 13871.
30. Pornillos, O.; Alam, S. L.; Davis, D. R.; Sundquist, W. I. Nat. Struct. Biol. 2002, 9,
812.
31. Pornillos, O.; Alam, S. L.; Rich, R. L.; Myszka, D. G.; Davis, D. R.; Sundquist, W. I.
EMBO J. 2002, 21, 2397.
32. Timmins, J.; Schoehn, G.; Ricard-Blum, S.; Scianimanico, S.; Vernet, T.; Ruigrok,
R. W.; Weissenhorn, W. J. Mol. Biol. 2003, 326, 493.
18. Harty, R. N.; Paragas, J.; Sudol, M.; Palese, P. J. Virol. 1999, 73, 2921.
19. Irie, T.; Licata, J. M.; McGettigan, J. P.; Schnell, M. J.; Harty, R. N. J. Virol. 2004, 78,
2657.
20. Licata, J. M.; Simpson-Holley, M.; Wright, N. T.; Han, Z.; Paragas, J.; Harty, R. N.
J. Virol. 2003, 77, 1812.
21. Urata, S.; Yasuda, J. J. Gen. Virol. 2010, 91, 228.
22. Wirblich, C.; Tan, G. S.; Papaneri, A.; Godlewski, P. J.; Orenstein, J. M.; Harty, R.
N.; Schnell, M. J. J. Virol. 2008, 82, 9730.
33. Urata, S.; Noda, T.; Kawaoka, Y.; Morikawa, S.; Yokosawa, H.; Yasuda, J. J. Virol.
2007, 81, 4895.
34. Urata, S.; Noda, T.; Kawaoka, Y.; Yokosawa, H.; Yasuda, J. J. Virol. 2006, 80, 4191.
35. Liu, Y.; Lee, M. S.; Olson, M. A.; Harty, R. N. Adv. Virol. 2011, 2011.
36. Irie, T.; Licata, J. M.; Jayakar, H. R.; Whitt, M. A.; Bell, P.; Harty, R. N. J. Virol.
2004, 78, 7823.
37. Makhloufi, A.; Baitiche, M.; Merbah, M.; Benachour, D. Synth. Commun. 2011,
41, 3532.
23. Zhadina, M.; Bieniasz, P. D. PLoS Pathog. 2010, 6, e1001153.
24. Irie, T.; Licata, J. M.; Harty, R. N. Virology 2005, 336, 291.
25. Dolnik, O.; Kolesnikova, L.; Stevermann, L.; Becker, S. J. Virol. 2010, 84,
7847.
26. Martin-Serrano, J.; Zang, T.; Bieniasz, P. D. Nat. Med. 2001, 7, 1313.
27. May, E. R.; Armen, R. S.; Mannan, A. M.; Brooks, C. L. (3rd) Proteins 2010, 78,
2251.
38. Yuan, H.; Li, X.; Qu, X.; Sun, L.; Xu, W.; Tang, W. Med. Chem. Res. 2009, 18, 671.
39. Schweim, H. G. Arch. Pharm. 1986, 319, 814.
40. Hisamatsu, Y.; Shirai, N.; Ikeda, S.; Odashima, K. Org. Lett. 2009, 11, 4342.
41. Waring, M. J.; Arrowsmith, J.; Leach, A. R.; Leeson, P. D.; Mandrell, S.; Owen, R.
M.; Pairaudeau, G.; Pennie, W. D.; Pickett, S. D.; Wang, J.; Wallace, O.; Weir, A.
Nat. Rev. Drug Disc. 2015, 14, 475.
Please cite this article in press as: Loughran, H. M.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.06.053