View
217
Download
0
Category
Preview:
Citation preview
1
Identification of a Novel Antiviral Inhibitor of the Flavivirus Guanylyltransferase Enzyme 1
2
Hillary J. Stahla-Beek1, Daniel G. April2, Bejan J. Saeedi2, Amanda M. Hannah1, Susan M. Keenan1* and 3
Brian J. Geiss2,3* 4
1School of Biological Sciences, University of Northern Colorado, Greeley, CO, USA 5
2Department of Microbiology, Immunology, Pathology, Colorado State University, Fort Collins, CO USA 6
3Department of Biochemistry and Molecular Biology, Colorado State University, Fort Collins, CO USA 7
8
*Correspondence can be sent to either: 9
Susan Keenan Brian Geiss 10
School of Biological Sciences Department of Microbiology, Immunology, and Pathology 11
University of Northern Colorado Colorado State University 12
Greeley, CO 80639 Fort Collins, CO 80523 13
Phone: (970) 351-2510 Phone: (970) 491-6330 14
Fax: (970) 351-2335 Fax: (970) 491-2221 15
e-mail:Susan.Keenan@unco.edu e-mail: Brian.Geiss@colostate.edu 16
17
Running Title: Thioxothiazolidins as Anti-Flaviviral Inhibitors 18
Keywords: Flavivirus; RNA capping; Antiviral; Guanylyltransferase 19
20
21
22
23
24
Copyright © 2012, American Society for Microbiology. All Rights Reserved.J. Virol. doi:10.1128/JVI.00384-12 JVI Accepts, published online ahead of print on 6 June 2012
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
2
Abstract 25
Arthropod-borne flavivirus infection causes serious morbidity and mortality worldwide, but there are 26
currently no effective antiviral chemotherapeutics available for human use. Therefore, it is critical that 27
new therapeutics to virus-specific targets be developed. To identify new compounds that may be used 28
as broadly active flavivirus therapeutics, we have performed a high-throughput screen of 235,456 29
commercially available compounds for small molecule inhibitors of the dengue virus NS5 RNA capping 30
enzyme. We identified a family of compounds, the 2-thioxothiazolidin-4-ones, that show potent 31
biochemical inhibition of GTP binding and guanylyltransferase function of the capping enzyme. During 32
the course of structure-activity relationship analysis, a molecule within this family (E)-(3-(5-(4-tert-33
butylbenzylidene)-4-oxo-2-thioxo-1,3-thiazolidin-3-yl)propanoic acid (BG-323)) was found to possess 34
significant antiviral activity in a dengue virus subgenomic replicon assay. Further testing of BG-323 35
demonstrated that this molecule is able to reduce the replication of infectious West Nile and yellow 36
fever viruses in cell culture with low toxicity. The results of this study describe the first inhibitor that 37
targets the GTP-binding / guanylyltransferase activity of the flavivirus RNA capping enzyme. 38
39
Introduction 40
Arthropod-borne virus infections remain a major cause of morbidity and mortality worldwide. More 41
than two billion people are at risk of infection with dengue virus (DEN) and 600 million people at risk of 42
infection with yellow fever virus (YF) (20). Globally, an estimated 50-100 million cases of DEN and 43
200,000 cases of YF reported each year, which result in respectively ~20,000 to ~30,000 deaths annually 44
(11). There are currently no clinically useable chemotherapeutic options for the treatment of any 45
flavivirus infection, making it essential that new strategies and targets for the treatment of flavivirus 46
infections be identified. 47
48
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
3
Flaviviruses are small enveloped, single-stranded positive sense RNA viruses with genomes consisting of 49
approximately 11,000 kb RNA with a 5’ type 1 RNA cap (23). The viral genome is translated as a single 50
open reading frame (ORF) encoding a polyprotein precursor that is processed into three structural 51
proteins (capsid, premembrane, and the envelope) and eight nonstructural proteins (NS1, NS2A, NS2B, 52
NS3, NS4A, 2K, NS4B and NS5) by viral and cellular proteases (16). Currently, four viral enzymes are 53
being studied as targets for antiviral drug discovery, including the NS3 helicase and protease enzymes 54
and the NS5 RNA dependent RNA polymerase and capping enzymes (reviewed in (8)). 55
56
In particular, the capping enzyme has received a good deal of attention as a novel antiviral drug target in 57
recent years. The flavivirus capping enzyme has three distinct functions that can be targeted for 58
therapeutic intervention: the N7/2’-O methyltransferase reactions (2, 3, 6, 9) and the recently 59
discovered guanylyltransferase reaction (10, 13, 14). The formation of the 5’ cap structure is critical to 60
the survival of the virus for several reasons, including directing viral polyprotein translation and 61
protecting the 5’ end of the genome from cellular exonucleases. It has also been recently shown that a 62
fully mature type 1 cap is a mechanism that cells use to discriminate self from non-self RNAs, and 63
interference with the formation of a mature type 1 cap on the flavivirus genome limits viral replication 64
(5, 28). The flavivirus NS5 N-terminal capping enzyme is highly conserved across the flavivirus genus, 65
and the guanosine triphosphate (GTP) and S-adenosyl methionine (SAM) binding sites, as well as the 66
overall structure of the enzyme, are well conserved (4, 7, 9, 18, 27). The critical nature of the capping 67
enzyme in viral replication and immune evasion, as well as its conservation across the flavivirus genus, 68
position the capping enzyme as an important target for antiviral development efforts. The 69
methyltransferase activity has been the primary capping enzyme target for drug development (15, 21), 70
and Ribavirin triphosphate has been observed to bind to and displace GTP from the enzyme (1). 71
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
4
In this manuscript we describe the identification and characterization of the 2-thioxothiazolidin-4-one 72
family of compounds as novel inhibitors of the flavivirus RNA capping enzyme guanylyltransferase. We 73
previously reported an analysis of the capping enzyme GTP binding site (9) and a pilot high throughput 74
screen (HTS) of 46,323 small molecule compounds using a fluorescence polarization assay (10). A screen 75
for compounds capable of displacing GTP from the dengue capping enzyme was performed against 76
235,456 compounds from the NSRB library located at the National Screening Laboratory (Harvard 77
Medical School (Longwood Campus)). We detail the process of developing structure activity 78
relationships (SAR) with analogs within this family, which have lead to a solid understanding of the 79
binding parameters of these compounds to the capping enzyme. During this process we identified a 80
lead compound (E)-(3-(5-(4-tert-butylbenzylidene)-4-oxo-2-thioxo-1,3-thiazolidin-3-yl)propanoic acid) 81
(referred to as BG-323 hereafter), that competes with GTP binding to the capping enzyme, inhibits 82
capping enzyme protein guanylation activity in vitro, and displays antiviral activity against multiple 83
different flaviviruses in cell culture. These findings demonstrate that the 2-thioxothiazolidin-4-one 84
family is a novel scaffold for developing effective and specific antiviral molecules targeting the GTP 85
binding pocket of the flavivirus RNA capping enzyme. 86
87
Materials and Methods 88
Expression and Purification of flavivirus NS5 Capping Enzymes 89
Recombinant NS5 capping enzyme domains from yellow fever virus (strain 17D, AA 1-268) and dengue 90
virus type 2 (strain 16681, AA 1-267) were previously described (9). Briefly, yellow fever virus capping 91
enzyme was produced in BL21 (DE3) Codon Plus E. coli cells (Novagen). Dengue virus capping enzyme 92
was produced in BL21 (DE3) pLysS E. coli cells (Novagen). Yellow fever and dengue virus proteins were 93
induced and purified using the same protocol. Cultures (750 ml) were induced with 400 μM IPTG 94
overnight at 22oC, and the bacterial pellets were collected and stored at -80oC in low imidizole lysis 95
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
5
buffer. Frozen pellets were thawed and lysed with a microfluidizer, and the lysate was clarified by 96
centrifugation at 18K RPM in a SS-24 rotor. The histidine-tagged proteins are purified from clarified 97
lysates using a nickel-sepharose column on an AKTA Purifier FPLC system. The eluted protein was 98
concentrated using 10K Amicon Ultra concentrators (Millipore) and buffer exchanged into 400 mM NaCl, 99
20 mM Tris pH 7.5, 0.02% sodium azide, 20% glycerol, and 5mM Tris (2-carboxyethyl) phosphine 100
hydrochloride (TCEP-HCl) on a Superdex 200 gel filtration column (Amersham). Purified proteins were 101
concentrated using 10K Amicon Ultra concentrators to 100 µM and the concentrations were determined 102
by the absorbance at 280 nm using extinction coefficients obtained from the ExPASy web site. Isolated 103
proteins were >99% pure as estimated from SDS-PAGE and Coomassie blue staining. Purified protein 104
was stored at -80 °C in single-use aliquots. 105
106
High-Throughput Screening 107
High throughput screening was performed at the National Screening Lab (NRSB) located at the Harvard 108
Medical School Longwood campus (ICCB-Longwood Screening Facility). To perform the screen, 500 nM 109
of purified dengue virus capping enzyme was complexed with 10 nM of GTP-Bodipy γ-phosphate labeled 110
analog (Invitrogen, Catalog # G22183) in binding buffer (50 mM Tris-Base (pH 7.5), 0.01% NP40, 2 mM 111
dithiothreitol). 30 µl was dispensed into low-binding opaque black 384 well plates (Catalog # 3654, 112
Corning, Corning NY) using a Matrix WellMate liquid handler (Thermo Fisher Scientific, Waltham, MA). 113
One column of 10 µM GTP (final concentration) was used as a positive control on each plate, and one 114
column was treated with dimethyl sulfoxide (DMSO) as a negative control. Screening compounds were 115
added to each plate with an Epson compound transfer robot fitted with a 100 nl 384 pin transfer array. 116
Plates treated with 100 nl of compound (5 mg/ml stock concentrations) were allowed to incubate for 1 117
hr at 23oC, then total fluorescence and fluorescence polarization signals were detected on an Envision 118
2103 Multimode platereader with plate stacker attachment (Perkin Elmer, Waltham, MA). Each 119
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
6
compound was tested in duplicate. The overall Z’ score of the screen was >0.7. Compounds that 120
reduced both total fluorescence and fluorescence polarization signals by greater than 50% were cherry-121
picked and re-tested on a Victor 3V platereader. 122
123
Determination of apparent Ki values 124
Compounds were obtained from ChemDiv and Hit2Lead. All small molecules were diluted in DMSO to 125
10 mM and were stored in a -20°C freezer. All compounds were stored in small aliquots in a desiccator 126
to prevent freeze/thaw cycles. Ki values for each compound were determined based on the equation 127
detailed in (17) using a fluorescence polarization assay as previously described (9, 10). Compounds were 128
tested at least 3 times each, and standard deviations for each Ki value are reported. 129
130
Guanylation Inhibition Assay 131
Capping enzyme protein guanylation as performed as described previously (10). Briefly, 3 µM dengue 132
capping enzyme was incubated with 1 µM GTP-ATTO-680 (Catalog # NU-830-680, Jena Bioscience, Jena, 133
Germany), 500 nM MgCl2, 0.1% NP-40, and 1 µM TCEP. Reactions were treated with compounds at final 134
concentrations of 100 µM, 50 µM, 25 µM, 10 µM, and 2.5 µM or mock DMSO controls for 4 hours at 135
37oC. At the end of the incubation, samples were quenched with 1 µl of 1 M EDTA and 6X Laemmli 136
buffer was added. Samples were boiled for 15 minutes and resolved on 12% SDS-PAGE gels. The gels 137
were imaged for ATTO-680 signal on a Licor Odyssey UV scanner (Licor, Lincoln, NE), then the gels were 138
stained with Coomassie blue to verify protein equivalence. Coomassie-stained gels were analyzed using 139
the NIH Image J software package. ATTO-680 signals for experimental samples were normalized for 140
protein concentration as compared to on-gel control samples. Inhibition values were determined using 141
non-linear regression analysis in the Prism Software package (Graphpad Software Inc, La Jolla, CA). 142
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
7
Average EC50 values and standard error of the mean values are reported. Each experiment was 143
performed three times. 144
145
Antiviral (Replicon) Assay 146
BHK cells harboring a stable dengue type 2 virus subgenomic replicon (BHK-pD2hRucPac) expressing 147
Renilla luciferase have been previously described (26). BHK-pD2hRucPac cells were grown in DMEM 148
media supplemented with 10% fetal bovine serum and 3 µg/ml puromycin to maintain stability of the 149
replicon. Working stocks of BHK-pD2hRucPac cells were used for two months and then replaced with 150
fresh stocks to maintain the sensitivity of the replicon to antiviral drugs. Antiviral assays were 151
performed by seeding white opaque 96-well cell culture plates (Catalog # T-3021-13, Bioexpress, 152
Kaysville, UT) with 2000 cells/well in 100 µl DMEM supplemented with 10% FBS but without puromycin. 153
Cells were allowed to attach overnight at 37oC. The next day, test compounds were diluted in DMSO 154
and 1 µl of each dilution was added to appropriate wells (200 µM to 78 nM final concentrations). Each 155
plate included a control column of DMSO with no drug and several rows of Ribavirin (200 µM to 78 nM 156
final concentrations) to verify the sensitivity of the replicons to drug. Cells were incubated at 37oC for 157
72 hours, when cells were tested for Renilla Luciferase activity and cell viability. Media was replaced 158
with 25 µl of DMEM containing 1:1000 dilution of Viviren Live Cell Renilla Luciferase Reagent (Catalog # 159
E6492, Promega, Madison, WI), and the plates were incubated for 10 minutes at 37oC. Renilla Luciferase 160
signal was detected on a Victor Multi-Mode plate reader (Perkin Elmer). After reading the Renilla 161
Luciferase signal, 25 µl of CellTiter-Glo cell viability reagent (Catalog # G7571, Promega, Madison, WI) 162
was added to each well, and the cells were incubated for 10 minutes at 37oC. CellTiter-Glo signal was 163
detected in the same manner as the Viviren signal. Renilla and CellTiter-Glo experimental samples were 164
converted to percent of the averaged on-plate DMSO controls, and non-linear regression curves 165
(variable slope) were generated with the Prism software. EC50 values were calculated for Renilla 166
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
8
luciferase and CC50 values calculated for CellTiter-Glo curves, and averaged values and standard 167
deviations over at least three experiments for each are reported. Therapeutic index (TI) is calculated as 168
CC50/EC50. 169
170
Viral assays 171
West Nile virus (Kunjin subtype) and yellow fever (17D) virus growth curves were performed in BHK 172
cells. Cells were plated in 6-well plates at 100,000 cells/well and allowed to attach overnight. The next 173
day cells were treated with BG-323 or DMSO at the indicated concentrations and concurrently infected 174
with Kunjin or yellow fever virus at a multiplicity of infection (MOI) of 0.01. 250 µl media samples were 175
taken at 4, 12, 24, 36, 48, 72, 96, and 120 hours post infection and stored at -80oC. Samples were 176
titered for virus concentration by plaque assay on BHK cells as previously described (19). Viral growth 177
curves were generated with Prism (Graphpad Inc., La Jolla, CA). Renilla luciferase expressing Sindbis 178
virus pBG451 was previously described (24). Renilla signal was detected in infected BHK cells with 179
Viviren live cell reagent (Promega), and cell viability was detected with CellTiter Glo reagent (Promega). 180
Each experiment was performed three times and the average plaque forming unit (PFU)/ml and 181
standard error of the mean is reported. 182
183
Quantitative Reverse Transcriptase Real-Time PCR Analysis 184
West Nile virus (Kunjin) RNA was extracted from cell culture media with Trizol LS (Invitrogen, La Jolla, 185
CA) following the manufacturer’s protocol. Quantitative reverse transcriptase real-time PCR reactions 186
for Kunjin genomic RNA were performed using the Brilliant III Ultra-Fast SYBR QRT-PCR Master Mix 187
(Catalog # 600886, Agilent, Santa Clara, CA) with primers Fwd 6170 (5’-188
TGGACGGGGAATACCGACTTAGAGG) and Rev 6278 (5’-ACCCCAGCTGCTGCCACCTT). To set up qRT-PCR 189
reactions, 2 μl of extracted RNA was added to 5 μl of 2X master mix, 1 μl of 5 μM Fwd 6170 primer, 1 μl 190
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
9
of 5 μM Rev 6278 primer, and 1 μl of 100 mM DTT in 96-well PCR plates with optically clear sealing films. 191
No RNA and no primer controls were included in each experiment. qRT-PCR reactions were performed 192
on a BioRad CFX384 real-time PCR thermal cycler using the following cycling conditions: Reverse 193
transcriptase step = 50oC, 10 minutes; Denature step = 95oC, 3 minutes; PCR (40 cycles) = 95oC, 5 194
seconds / 60oC, 10 seconds. Melt curves were performed at the end of each run, to verify the specificity 195
of the detected SYBR signal. Cq values were determined by setting the threshold to 40 for all 196
experiments. Standard curves were generated from diluted media containing Kunjin virus of known titer 197
(PFU/ml), and the Cq values from experimental samples were compared to the standard curve to 198
establish PFU equivalent/ml values for each sample (y=-3.475x + 27.577 , y=Cq , X=Log10(PFU)). The limit 199
of detection in these experiments was determined to be 10 PFU equivalents/ml. All experiments were 200
performed three times, and the average and standard error of the mean reported. 201
202
Western blot analysis 203
BHK cells were infected with Kunjin virus at MOI = 0.1 and treated with increasing concentrations of BG-204
323 or DMSO. At 72 hrs post-infection, cells were collected and lysates were prepared by boiling in 1X 205
laemelli buffer. Lysates were resolved on 12% PAGE gel and protein was transferred to nitrocellulose 206
membranes. Western blot analysis was performed with anti-NS5 antibody 5D4 (12) and anti-β-actin 207
(Abcam #6276) on a separate membrane. Bands were detected with an IR-DYE-800 anti-mouse 208
secondary antibody (Rockland Scientific) on an Odyssey UV Imaging system. 209
210
Modeling Analysis 211
Each molecular structure was drawn using Maestro and minimized using Macromodel (Schrodinger, NY). 212
For minimization of all small molecules, the dielectric constant was set to 4.0 (aqueous), the maximum 213
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
10
iterations was set to 1000 and the convergence threshold was set to 0.05 kJ/Å-mol. The OPLS_2005 214
forcefield was used and conjugate gradient methodology applied. 215
216
Docking and scoring 217
All small molecules were evaluated with GOLD (25) to determine potential orientations of association 218
with dengue virus capping enzyme (PDB Code: 3EVG) and yellow fever virus capping enzyme (PDB Code: 219
3EVD). Unless discussed below, default parameters were applied. The centroid of the docking sphere is: 220
x=16.37, y=-52.73 and z=17.934, with an active site radius of 10 Å. Each run consisted of 50 iterations 221
per small molecule. The fitness function and search settings had the annealing parameters such that 222
van der waals radii = 4 Å and hydrogen bonding distance = 2.5 Å. Fifty docked conformations were 223
obtained for each compound and DSViewerPro 5.0 (Accelrys, CA) was used to visualize properties 224
between the small molecule and the capping enzyme protein. For the majority of compounds one 225
predominant conformation was obtained and was chosen as the physiologically relevant orientation for 226
further analysis. For compounds with multiple orientations (generally observed for compounds with 227
lower binding affinity), the orientation most similar (via visual inspection) to the conserved 228
physiologically relevant orientation was chosen for further analysis. The docking protocol used was 229
previously validated for ligand binding to the dengue capping enzyme (10) and can recapitulate the 230
association of GTP, GDP, and GMP within the binding site (data not shown). 231
232
233
Results 234
High-throughput screening 235
Based on our previous screen of 46,323 compounds against the yellow fever virus capping enzyme (10), 236
we performed a second screen of the remaining 235,456 compounds in the NRSB library for molecules 237
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
11
that were able to displace GTP from the dengue virus capping enzyme. This screen identified 633 238
compounds that showed > 50% inhibition in fluorescence polarization and had a corresponding 239
reduction in total fluorescence intensity, similar to what we have previously reported with our GTP-240
Bodipy binding assay (9, 10). These 633, which were determined to be tractable to medicinal chemistry, 241
were cherry picked from the NSRB libraries and were re-checked for activity using cherry-picked 242
material from the NRSB library on a different platereader. 222 of the 633 molecules repeated, a repeat 243
rate of 35.1%. Within this 222 compounds we observed a small cluster of compounds with a 244
thioxothiazolidin core and an acid moiety that appeared to displace GTP-Bodipy from the capping 245
enzyme with varying strengths based on relative displacement in the HTS assay (Table 1), indicating that 246
structure activity relationships may be explored with this family. The thioxothiazolidins were not the 247
strongest “hits” in the 222 compounds that repeated, but because the thioxothiazolidin core is found in 248
known FDA approved drugs, such as Epalrestat, and a number of analogs were commercially available 249
we chose to pursue this family of inhibitors further. 250
251
SAR Analysis of the Thioxothiazolidins 252
The four compounds from Table 1 were ordered in 5 mg quantities from ChemDiv and Hit2Lead. 21 253
additional commercially available compounds with the same core structure as BG-5 (2-thioxothiazolidin-254
4-one) (Figure 1A) were also ordered, and apparent Ki values were determined for 24 compounds 255
against dengue and yellow fever virus capping enzyme using 24-point titration curves (9, 10). 256
Compound 2044-5240 caused the protein in the Ki assays to precipitate, and no usable data was 257
obtainable. Otherwise, the thioxothiazolidin-based series of compounds exhibited apparent Ki values 258
ranging from >100 μM to 1.5 μM (Table 2). 259
260
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
12
We visually examined the structures of the small molecules in order to understand how changes in 261
compound structure relate to compound binding affinity. Interestingly, a number of trends were 262
observed. First, substitution on the aromatic ring (positions R2-R6) appears to be necessary for a small 263
molecule to be able to compete with GTP for the capping enzyme binding site. The aromatic ring of 264
compound BG-327 is not substituted (R2-R6) and this compound shows no ability to displace GTP (Ki = 265
>100 μM). The addition of an isopropyl group at R4 (BG-322) results in improved affinity (Ki = 12 μM) and 266
replacing the R4 isopropyl group of BG-322 with a larger t-butyl group (BG-323) further boosts affinity (Ki 267
= 7.5 μM). Both BG-115 (Ki = 4.0 μM) and BG-170 (Ki = 2.9 μM) provide additional evidence of this trend 268
as both contain bulky groups at position R4. Interestingly, a fluoride atom at R4 is not sufficient for 269
activity (BG-321; Ki > 100 μM). Second, while the compounds examined to date do not allow for an 270
exhaustive exploration of how the relationship between the position of an aromatic ring substituent and 271
binding affinity, the data do provide some clues. BG-317, BG-318 and BG-319 each have two chlorides 272
on the aromatic ring in varying positions. The presence of chlorides at the R3 and R4 positions (BG-318 273
Ki = 11 μM) results in increased binding when compared to molecules with chlorides in the R2 and R3 or 274
R2 and R4 positions, BG-317 (Ki = 27 μM) or BG-319 (Ki = 24 μM) respectively. Additional compounds 275
would need to be examined in order to determine whether electronegativity or simply hydrophobic 276
contacts are essential for improved affinity. 277
278
A final trend suggested by the data is that affinity is affected by the distance between the acid moiety at 279
the R1 position and the thiazolidin core. For example, BG-317 (1 carbon) and BG-324 (2 carbons) have 280
observed Ki s of 27 μM and 9.9 μM respectively. Similar effects were observed with compound BG-319 281
(1 carbon; Ki = 24 μM) and compound BG-330 (2 carbons; Ki = 9.8 μM). There are, however, data within 282
this set that do not support this trend. BG-318 (1 carbon; Ki = 11.3 μM) and BG-328 (2 carbon; Ki = 13.6 283
μM) have similar affinities for the enzyme. Elongating R1 to 3 carbons (eg. butyric acid) has a negative 284
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
13
effect on compound affinity (compare BG-330 (2 carbons) to BG-331 (3 carbons)) and modifying the 285
propionic acid to methylpropionate ablates activity. Collectively, these data suggest that there is likely a 286
structural preference for the distance of the acidic group from the thiazolidin core and that the acid 287
moiety plays a role in compound association with the binding site. Interestingly there is no observed 288
improvement in affinity without substitutions on the aromatic ring (compare BG-325 and BG-327) 289
confirming the necessity of the aromatic substitutions for activity. 290
291
Overall, our results suggest that for capping enzyme binding compounds from the thioxothiazolidin-292
based series require substitution of the aromatic ring (R2-R6) and that an acid group is necessary for 293
inhibition. Furthermore, we have preliminary information suggesting that affinity is sensitive to the 294
distance between the aromatic group and the acid moiety. These data, which is summarized in Figure 295
1B, may aid in the future optimization of this family of compounds as capping enzyme inhibitors. 296
297
Inhibition of guanylyltransferase activity 298
To determine if the thioxothiazolidin molecules interfere with guanylyltransferase activity, we tested the 299
ability of analogs with GTP Ki values less than 10 μM to interfere with the formation of the guanylated 300
intermediate of the dengue virus capping enzyme (Table 3). We observed that the IC50 for guanylation 301
inhibition for each compound was roughly equivalent to the GTP displacement Ki value, except in the 302
cases of BG-324 and BG-330, where guanylation inhibition was weakened or strengthened, respectively. 303
Chlorines in the R2 and R3 positions appear to weaken the enzymatic inhibition capacity of BG-324 (IC50 304
= 155 μM), whereas chlorines in the R2 and R4 position appear to increase enzymatic inhibition (IC50 = 305
2.6 μM), although the Ki values for BG-324 and BG-330 are identical. How the capping enzyme forms the 306
guanylated intermediate is unknown, as there is no recognizable guanylyltransferase active site motif 307
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
14
present in any flavivirus protein (14). Therefore, without understanding how the reaction occurs it is 308
unclear how the presence of chlorines at R2/R3/R4 may differentially affect enzymatic activity. Overall, 309
this data indicates that of the majority of compounds within this group that displace GTP with 310
reasonable Ki values interfere with the enzymatic activity of the capping enzyme to roughly similar 311
extents. 312
313
Preliminary antiviral testing of the Thioxothiazolidins 314
Concurrently with our biochemical SAR analysis of the thioxothiazolidin analogs, we tested the ability of 315
the six compounds that displayed lower than 10 μM Ki values to reduce Renilla luciferase expression 316
from a persistent dengue replicon in BHK cells (26). Renilla luciferase expression in replicon cells treated 317
with decreasing concentrations of the indicated compound was assayed 72 hours after treatment and 318
used to develop EC50 curves. We also calculated the CC50 of each compound and determined the 319
therapeutic index (TI) of each compound (Table 3). Compound BG-5 showed weak antiviral effect with a 320
TI value of 2.8 (EC50 = 97 μM; CC50 = 270 μM). BG-323 demonstrated a TI of 6, which was higher than the 321
Ribavirin positive control used in the replicon assays (Ribavirin TI = 4.1). The EC50 value for BG-323 is 322
30.8 μM and the CC50 value is 184 μM, indicating that while BG-323 is effective at a higher concentration 323
than Ribavirin (EC50 = 12 μM), it is also significantly less toxic than Ribavirin (CC50 = 49 μM). 324
325
Effects of BG-323 against infectious virus 326
To test if BG-323 could interfere with the replication of infectious viruses, we performed multi-step 327
growth curves with West Nile (Kunjin) and yellow fever virus in BHK cells with various concentrations of 328
BG-323 added at time of infection (Figure 2A). We observed that BG-323 reduced virus titers by close to 329
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
15
3 logs at several times during Kunjin growth, with a slightly less robust effect also seen with yellow fever 330
virus. The antiviral effect was most pronounced with the highest concentration of BG-323 (100 μM), 331
which is significantly less than its CC50 value for BG-323 (184 μM). To determine if BG-323 had activity 332
against non-flaviviruses, we treated BHK cells with BG-323 and infected the cells with a Sindbis virus that 333
encodes Renilla luciferase (24) for 24 hrs. Only a very minor reduction in Renilla signal was observed at 334
the 100 μM concentration of BG-323 with a corresponding minor reduction in cell viability (Figure 2B). 335
Therefore, BG-323 does not appear to interfere with the replication of a non-flavivirus and suggests that 336
BG-323 has specificity in inhibiting flavivirus infection. 337
338
To monitor if BG-323 was affecting viral protein accumulation in infected cells, we treated Kunjin 339
infected cells (MOI = 0.01) with increasing concentrations of BG-323 and assayed the cells for the 340
presence of the viral NS5 protein at 48 hrs post-infection by western blot analysis (Figure 3A). We 341
observed a significant decrease in NS5 with increasing BG-323 concentrations, and noted that a base 342
level of NS5 is present even in high concentrations of BG-323, likely due to translation of the input viral 343
RNA. The significant decrease in NS5 levels indicates that BG-323 may interfere with translation of viral 344
proteins from de novo synthesized genomic RNAs or alters the stability of the translated NS5 protein in 345
infected cells. Next we sought to determine if BG-323 was also reducing the amount of viral RNA being 346
released into culture media. We infected BHK cells with Kunjin virus at MOI = 0.01 and treated the cells 347
with DMSO, 50 or 100 μM BG-323, or mock treated the cells. 72 hours after infection we collected 348
media and extracted viral RNA, which was used in quantitative real-time reverse-transcriptase PCR 349
reactions. We observed that DMSO treatment had minimal effects on viral RNA in the media, whereas 350
50 μM or 100 μM BG-323 significantly reduced the amount of viral RNA detectable in the media (Figure 351
3B). 352
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
16
353
We observed that viral titers of BG-323 treated cultures reached the same level as mock treated 354
controls at later times during infection, indicating that BG-323 may not be stable throughout the entire 355
time course. To assess the activity of BG-323 over time, we incubated BG-323 in media at 37oC for 3, 2, 356
1, or 0 days. Then the pre-incubated media were added to BHK cells, Kunjin virus was added at MOI = 357
0.01, and media samples were taken 72 hours later. Viral RNA content was determined in media 358
collected from the infected cells by qRT-PCR (Figure 4A). The amount of viral RNA present in samples 359
with pre-treated media was significantly less than media treated with BG-323 at the time of infection, 360
indicating that BG-323 becomes less active over longer incubation periods. To determine if replacement 361
of BG-323 during infection could overcome this effect, we replaced BG-323 containing media on Kunjin 362
infected cells every 24 hrs for 72 hrs and monitored viral RNA in the culture media at each time point. 363
We observed that adding fresh BG-323 to the cells significantly reduced viral RNA in media over time 364
(Figure 4B), indicating that the virus remains sensitive to BG-323 and was not becoming resistant to BG-365
323 over the course of the experiment. 366
367
368
Discussion 369
This study describes the discovery of a group of novel inhibitors of flavivirus replication that may provide 370
pharmacological benefits for the medical treatment of flaviviruses. We have identified a series of 371
compounds able to inhibit an enzyme essential for viral replication and that show cell culture activity 372
similar to that observed by the known antiviral Ribavirin. 373
374
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
17
Crystal structures of the capping enzyme with GTP bound have been previously solved (9) and based on 375
mutational studies, GTP association is shown to require π/π stacking interactions with Phe 24 and the 376
formation of hydrogen bonds with Lys 13, Leu 16, Asn 17, Leu 19, Lys 28, Ser 150, Arg 213 and Ser 215. 377
A water bridge is also forms between the nitrogenous base and the backbone oxygen of Leu 19 that is 378
suggested to play a role in affinity. 379
380
Based on the computational docking of BG-5 (Figure 1A), the compound in this series with the highest 381
affinity for the enzyme, we suggest that the thioxothiazolidin family of compounds interact with the 382
enzyme in ways that mimic the association of GTP. Specifically, this compound appears to hydrogen 383
bond with Lys 28 and Ser 150 and π/π stack with the aromatic side chain of Phe 24. In addition, BG-5, 384
GTP and other compounds with improved binding affinity appear to interact with two sub-pockets 385
within the capping enzyme GTP binding site. Sub-pocket 1 allows for interactions between the small 386
molecules and Phe 24 and Lys 21 while sub-pocket 2 allows for interactions between the small 387
molecules and Asn 17 Leu 18, and Leu 19. Interaction with the two sub-pockets provides a rationale for 388
the structure-activity relationships observed In Table 2. First, compounds with increased bulk in the R2-R6 389
positions are predicted to have increased interactions within one or both of the sub-pockets (for 390
example, BG-322, BG-323). Second, the addition of a carbon-carbon bond linking the acid moiety in the 391
R1 position suggests that these compounds could protrude into these sub-pockets to a greater extent 392
while still maintaining the ability of compounds to form hydrogen bond with Ser 150 and Lys 28. 393
Interestingly, the two sub-pockets also provide a structural explanation for the similar inhibitory effects 394
of BG-318 and BG-328: based on docking studies, both compounds appear to interact with sub-pocket 2 395
to a similar (non-optimal) extent and for this reason affinity is not increased by altering the distance 396
between the thiazolidin core and the acid moiety. Currently BG-5 has the highest affinity for the capping 397
enzyme and its proposed orientation within the binding site (Figure 1A) supports our hypothesis that 398
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
18
compounds able to interact with either or both sub-pockets results in increases of compound affinity for 399
the binding site. Additional analogs would need to be tested to further clarify how the 400
thioxothiazolidins bind to the capping enzyme, although our preliminary pharmacophore presented in 401
Figure 1B provides a sound basis for further SAR development. 402
403
Two compounds within this study, BG-5 and BG-323, displayed antiviral activity in the dengue virus 404
replicon assay (Table 3). BG-323 displayed a therapeutic index of 6, which is more efficacious than our 405
Ribavirin positive controls. Further testing of BG-323 showed that it was able to suppress the replication 406
of West Nile (Kunjin) and yellow fever viruses by plaque assay, qRT-PCR analysis of media samples, and 407
western blot analysis. These data suggest that BG-323 is able to reduce the amount of infectious virus 408
that is released from infected cells, and indicates that BG-323 has significant antiviral activity in culture. 409
This finding is exciting as BG-323 (shown docked to the yellow fever capping enzyme in Figure 5) has a 410
similar structure to a known FDA approved drugs, such as the anti-diabetic drug Epalrestat (22), 411
indicating that an optimized thioxothiazolidin may be developed into a clinically usable drug for the 412
treatment of flavivirus infection. One concern is that thioxothiazolidins are considered to be somewhat 413
promiscuous in high-throughput screening assays. However, the observation that BG-323 demonstrates 414
antiviral activity against flaviviruses but not alphaviruses (Figure 2B) suggests that BG-323 is not merely 415
a “sticky” molecule but has some specificity against flavivirus replication, making it an interesting 416
candidate for further development. Additionally, we have previously identified other core structures 417
that may be substituted for the thioxothiazolidin core to mitigate any liabilities that the moiety may 418
present (10). 419
420
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
19
The biochemical data we present demonstrate that BG-323 has activity against the capping enzyme 421
guanylyltransferase activity in vitro, although it is possible that part of the antiviral effect observed in 422
cells may be due to BG-323 interfering with the methyltransferase activity of the capping enzyme or 423
interfering with a cellular protein such as aldose reductase (the target of Epalrestat). Further testing is 424
needed to definitively show the mechanism of action for this series of compounds in cell culture. We 425
have also tested if BG-323 interferes with other enzymatic assays, such as PCR, but have not observed 426
any significant effect (data not shown), indicating that BG-323 is not broadly reactive towards enzymes. 427
428
While BG-323 does show promise as a novel anti-flaviviral therapeutic, there are a few issues that need 429
to be addressed to increase its efficacy. The therapeutic index of 6 for BG-323 is somewhat narrow, and 430
increasing this value will be critical to further development. BG-323 binds to the capping enzyme with a 431
Ki ~10μM and has an EC50 value of ~30 μM, indicating that the molecule is able to pass cellular 432
membranes relatively effectively and interfere with viral replication at a concentration not much higher 433
than is inhibitory in the biochemical assays. Therefore, increased binding affinity while maintaining 434
cellular permeability may help lower the effective EC50 value of the compound series to increase the 435
therapeutic index for the molecule and bring its inhibitory activity into a more therapeutically useful 436
range. In addition, BG-323 possesses a Michael acceptor group that may increase reactivity and lead to 437
the weak toxicity we observe in cell culture. Several approaches to mitigate these issues and potentially 438
increase the therapeutic index of BG-323 are currently being investigated, such as replacing the 439
propanoic acid group with the bioisostere tetrazol, converting the propanoic acid to an ethyl ester that 440
may act as a prodrug, and/or reducing the Michael acceptor to avoid undesired reactivity. Additionally, 441
BG-323 does appear to lose activity during the course of longer experiments, suggesting that the 442
compound may degrade over time or may bind albumin. Medicinal chemistry and pharmacokinetics will 443
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
20
be necessary to identify analogs or formulations that may increase the stability of the compound over 444
time in vitro and in vivo. Further SAR development, including mutagenesis analysis of the capping 445
enzyme and co-crystallization of BG-323 and the capping enzyme, will help clarify critical interactions 446
and increase binding affinity. 447
448
BG-323 represents a valuable platform for further development of antiviral inhibitors of flavivirus RNA 449
replication. Future studies will focus on decreasing the EC50 of BG-323, improving bioavailability, and 450
testing this family of molecules for in vitro and in vivo efficacy against a number of additional 451
flaviviruses. 452
453
Acknowledgements 454
High-throughput screening capability was provided by the National Screening Laboratory (NSRB) within 455
the New England Regional Center for Excellence (NIAID U54 AI057159). We thank Dr. Su Chiang and the 456
staff at the NSRB for assistance in performing the screening and post-screen analysis. We thank Mr. 457
Hamid Gari and Mr. Kevin Haworth for their assistance in the high-throughput screen and Drs. Brian 458
McNaughton and Ritwik Burai for re-synthesizing BG-323 for validation testing. We would also like to 459
thank Ms. Stephanie Moon for help with the Kunjin qRT-PCR assay, Dr. Alexander Khromykh for 460
providing the Kunjin virus used in this study, Dr. Roy Hall for providing the NS5 antibody 5D4, and Mr. 461
Jordan Steel for performing Sindbis virus assays. This work was supported by a grant from the Rocky 462
Mountain Regional Center for Excellence (RMRCE) (NIH/NIAID U54 AI065357) and a grant from 463
NIH/NIAID (4R01AI046435-11). 464
465
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
21
References 466
1. Benarroch, D., M. P. Egloff, L. Mulard, C. Guerreiro, J. L. Romette, and B. Canard. 2004. A 467 structural basis for the inhibition of the NS5 dengue virus mRNA 2'-O-methyltransferase domain 468 by ribavirin 5'-triphosphate. J Biol Chem 279:35638-43. 469
2. Bhattacharya, D., I. Ansari, and R. Striker. 2009. The flaviviral methyltransferase is a substrate 470 of Caein Kinase 1. Virus Research 141:101-104. 471
3. Bhattacharya, D., S. Hoover, S. Falk, B. Weisblum, M. Vestling, and R. Striker. 2008. 472 Phosphorylation of yellow fever virus NS5 alters methyltransferase activity. Virology 380:276-473 284. 474
4. Bollati, M., M. Milani, E. Mastrangelo, S. Ricagno, G. Tedeschi, S. Nonnis, E. Decroly, B. Selisko, 475 X. de Lamballerie, B. Coutard, B. Canard, and M. Bolognesi. 2009. Recognition of RNA cap in 476 the Wesselsbron virus NS5 methyltransferase domain: implications for RNA-capping 477 mechanisms in Flavivirus. J Mol Biol 385:140-52. 478
5. Daffis, S., K. J. Szretter, J. Schriewer, J. Li, S. Youn, J. Errett, T. Y. Lin, S. Schneller, R. Zust, H. 479 Dong, V. Thiel, G. C. Sen, V. Fensterl, W. B. Klimstra, T. C. Pierson, R. M. Buller, M. Gale, Jr., P. 480 Y. Shi, and M. S. Diamond. 2010. 2'-O methylation of the viral mRNA cap evades host restriction 481 by IFIT family members. Nature 468:452-6. 482
6. Egloff, M., E. Decroly, H. Malet, B. Selisko, D. Benarroch, F. Ferron, and B. Canard. 2007. 483 Structural and functional analysis of methylation and 5'-RNA sequence requirements of short 484 capped RNAs by the methyltrasferase domain of dengue virus NS5. J. of Molecular Biology 485 372:723-736. 486
7. Egloff, M. P., D. Benarroch, B. Selisko, J. L. Romette, and B. Canard. 2002. An RNA cap 487 (nucleoside-2'-O-)-methyltransferase in the flavivirus RNA polymerase NS5: crystal structure and 488 functional characterization. EMBO J 21:2757-68. 489
8. Geiss, B., H. Stahla, A. Hannah, H. Gari, and S. Keenan. 2009. Focus on flaviviruses: current and 490 future drug targets. Future Medicinal Chemistry 1:327-344. 491
9. Geiss, B., A. Thompson, A. Andrews, R. Sons, H. Gari, S. Keenan, and O. Peersen. 2009. Analysis 492 of Flavivirus NS5 methyltransferase cap binding. Journal of Molecular Biology 385:1643-1654. 493
10. Geiss, B. J., H. J. Stahla-Beek, A. M. Hannah, H. H. Gari, B. R. Henderson, B. J. Saeedi, and S. M. 494 Keenan. 2011. A high-throughput screening assay for the identification of flavivirus NS5 capping 495 enzyme GTP-binding inhibitors: implications for antiviral drug development. J Biomol Screen 496 16:852-61. 497
11. Guzman, A., and R. E. Isturiz. 2010. Update on the global spread of dengue. Int J Antimicrob 498 Agents 36 Suppl 1:S40-2. 499
12. Hall, R. A., S. E. Tan, B. Selisko, R. Slade, J. Hobson-Peters, B. Canard, M. Hughes, J. Y. Leung, E. 500 Balmori-Melian, S. Hall-Mendelin, K. B. Pham, D. C. Clark, N. A. Prow, and A. A. Khromykh. 501 2009. Monoclonal antibodies to the West Nile virus NS5 protein map to linear and 502 conformational epitopes in the methyltransferase and polymerase domains. J Gen Virol 503 90:2912-22. 504
13. Henderson, B. R., B. J. Saeedi, G. Campagnola, and B. J. Geiss. 2011. Analysis of RNA binding by 505 the dengue virus NS5 RNA capping enzyme. PLoS One 6:e25795. 506
14. Issur, M., B. Geiss, I. Bougie, F. Picard-Jean, S. Despins, J. Mayette, S. Hobdey, and M. Bisaillon. 507 2009. The flavivirus NS5 protein is a true RNA guanylyltransferase that catalyzes a two-step 508 reaction to form the RNA cap structure. RNA 15:2340-2350. 509
15. Lim, S. P., L. S. Sonntag, C. Noble, S. H. Nilar, R. H. Ng, G. Zou, P. Monaghan, K. Y. Chung, H. 510 Dong, B. Liu, C. Bodenreider, G. Lee, M. Ding, W. L. Chan, G. Wang, Y. L. Jian, A. T. Chao, J. 511
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
22
Lescar, Z. Yin, T. R. Vedananda, T. H. Keller, and P. Y. Shi. 2011. Small molecule inhibitors that 512 selectively block dengue virus methyltransferase. J Biol Chem 286:6233-40. 513
16. Lindenbach, B., H. Thiel, and C. Rice. 2007. Flaviviridae: the viruses and their replication, p. 514 1101-1152. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and 515 S. E. Straus (ed.), Fields virology, 5th ed, vol. 1. Lippinoctt-Raven Publishers, Philadelphia, PA. 516
17. Martin, R. L., F. Renosto, and I. H. Segel. 1991. A simple method for calculating the dissociation 517 constant of a receptor (or enzyme).unlabeled ligand complex from radioligand displacement 518 measurements. Arch Biochem Biophys 284:26-9. 519
18. Mastrangelo, E., M. Bollati, M. Milani, B. Selisko, F. Peyrane, B. Canard, G. Grard, X. de 520 Lamballerie, and M. Bolognesi. 2007. Structural bases for substrate recognition and activity in 521 Meaban virus nucleoside-2'-O-methyltransferase. Protein Sci 16:1133-45. 522
19. Paredes, A., M. Alzuru, J. Mendez, and M. Rodriguez-Ortega. 2003. Anti-Sindbis activity of 523 flavanones hesperetin and naringenin. Biol Pharm Bull 26:108-9. 524
20. Prevention, C. f. D. C. a. 2009. Centers for Disease Control and Prevention. 525 21. Puig-Basagoiti, F., M. Qing, H. Dong, B. Zhang, G. Zou, Z. Yuan, and P. Y. Shi. 2009. 526
Identification and characterization of inhibitors of West Nile virus. Antiviral Res 83:71-9. 527 22. Ramirez, M. A., and N. L. Borja. 2008. Epalrestat: an aldose reductase inhibitor for the 528
treatment of diabetic neuropathy. Pharmacotherapy 28:646-55. 529 23. Rice, C., E. Lenches, S. Eddy, S. Shin, R. Sheets, and J. Strauss. 1985. Nucleotide sequence of 530
yellow fever virus: implications for flavivirus gene expression and evolution. Science 229:726-531 733. 532
24. Steel, J. J., B. R. Henderson, S. B. Lama, K. E. Olson, and B. J. Geiss. 2011. Infectious alphavirus 533 production from a simple plasmid transfection+. Virol J 8:356. 534
25. Verdonk, M. L., J. C. Cole, M. J. Hartshorn, C. W. Murray, and R. D. Taylor. 2003. Improved 535 protein-ligand docking using GOLD. Proteins 52:609-23. 536
26. Whitby, K., T. C. Pierson, B. Geiss, K. Lane, M. Engle, Y. Zhou, R. W. Doms, and M. S. Diamond. 537 2005. Castanospermine, a potent inhibitor of dengue virus infection in vitro and in vivo. J Virol 538 79:8698-706. 539
27. Zhou, Y., D. Ray, Y. Zhao, H. Dong, S. Ren, Z. Li, Y. Guo, K. A. Bernard, P. Y. Shi, and H. Li. 2007. 540 Structure and function of flavivirus NS5 methyltransferase. J Virol 81:3891-903. 541
28. Zust, R., L. Cervantes-Barragan, M. Habjan, R. Maier, B. W. Neuman, J. Ziebuhr, K. J. Szretter, 542 S. C. Baker, W. Barchet, M. S. Diamond, S. G. Siddell, B. Ludewig, and V. Thiel. 2011. Ribose 2'-543 O-methylation provides a molecular signature for the distinction of self and non-self mRNA 544 dependent on the RNA sensor Mda5. Nat Immunol 12:137-43. 545
546 547 548
Figure Legends 549
550
Figure 1. SAR of Thioxothiazolidin binding. A) 3D rendering of the predicted orientation of BG-5 within 551
the yellow fever capping enzyme. The yellow fever protein colors: light blue represents the entire 552
protein, bright green represents the location of the hydrophobic sub-pocket 1 and the dark green 553
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
23
represents the location of the hydrophobic sub-pocket 2. BG-5 is depicted in stick form with the 554
following color scheme: sea green are carbon atoms, purple are Iodine atoms, white are hydrogen atoms, 555
red are oxygen atoms, blue are nitrogen atoms and yellow are sulfur atoms. B) Summary pharmacophore 556
model based on Ki values for the dengue and yellow fever virus capping enzymes in Table 2. The dash 557
and dotted red line represents π/π stacking and the magenta dashed lines represent predicted hydrogen 558
bonds. 559
560
Figure 2. A) Growth Curve Analysis of Kunjin and Yellow Fever viruses with increasing concentrations 561
of BG-323. Baby hamster kidney cells were infected with Kunjin or yellow fever viruses at MOI = 0.01, 562
and the indicated concentration of BG-323 was added at time = 0. Media samples were collected at the 563
indicated times and stored at -80oC. Samples were assayed for virus titer by plaque assay. Curves were 564
generated in Prism. N=3 B) Sindbis virus is insensitive to BG-323 treatment. BHK cells were infected 565
with a Renilla luciferase expressing Sindbis virus (pBG451; (24)) at MOI = 0.01, and BG-323 was added at 566
the indicated concentrations. Relative light units (RLU) from Renilla luciferase (white bars) and CellTiter 567
Glo (black bars) signals are shown. N=3 568
569
Figure 3. A) Western blot analysis of Kunjin virus infected BHK cells treated with BG-323. BHK cells 570
were infected at MOI = 0.01 with Kunjin and treated with the indicated concentration of BG-323. At 48 571
hrs post-infection cells were collected and NS5 and β-actin proteins were detected by Western blot. B) 572
Detection of viral RNA in BG-323 treated media. BHK cells were infected with Kunjin at MOI = 0.01, and 573
cells were mock treated, treated with DMSO, 50 μM BG-323, or 100 μM BG-323 at the time of infection. 574
Cells were incubated for 48 hours and media samples were collected and stored at -80oC. Samples were 575
thawed, RNA extracted, and RNA quantified by qRT-PCR. qRT-PCR Cq values were compared to a Kunjin 576
virus RNA standard curve and converted to Log10 PFU equivalents/ml. N=3. 577
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
24
578
Figure 4. A) Stability of BG-323 at 37oC. DMSO or 50 μM BG-323 (final concentration) was added to 579
DMEM culture media, and the media samples were incubated in a CO2 incubator at 37oC for 48 or 24 580
hours prior to infection. At time=0 pre-incubated media or fresh media and BG-323 was added to BHK 581
cells, and the cells were subsequently infected with Kunjin at MOI = 0.01. 72 hours post-infection media 582
samples were collected and stored at -80oC. RNA was prepared and quantified by qRT-PCR as described 583
above. N=3 B) Replacement of BG-323 reduces Kunjin virus replication. BHK cells were infected with 584
Kunjin virus at MOI = 0.01 and treated with 100 μM BG-323 or DMSO. At 24, 48, and 72 hrs post-585
infection media samples were collected and media on infected cells was replaced with fresh media with 586
100 μM BG-323 or DMSO as indicated. Viral RNA in the 24, 48, and 72 hr post infection (PI) was 587
quantified by qRT-PCR as PFU equivalents/ml. N=3 588
589
Figure 5. Molecular docking of BG-323 with the capping enzyme. On the left a 3D rendering of the 590
predicted orientation of BG-323 within the yellow fever capping enzyme is presented. The yellow fever 591
protein colors: light blue represents the entire protein, bright green represents the location of the 592
hydrophobic sub-pocket 1 and the dark green represents the location of the hydrophobic sub-pocket 2. 593
BG-323 is depicted in stick form with the following color scheme: cyan are carbon atoms, white are 594
hydrogen atoms, yellow are sulfur atoms, red are oxygen atoms and blue are nitrogen atoms. On the 595
right is a flatland model for predicted interactions between of BG-323 and the yellow fever virus capping 596
enzyme is presented. BG-323 is depicted in line form surrounded by capping enzyme residues. The light 597
green represents the location of the hydrophobic sub-pocket 1 and the dark green represents the 598
location of the hydrophobic sub-pocket 2. The dash and dotted red line represents the predicted π/π 599
stacking interactions and the pink dashed lines represent predicted hydrogen bonds. 600
601
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
Table 2. Structure activity relationship analysis of thioxathiazolidins. Substitutions are each position are noted. Apparent Ki values for dengue and yellow fever capping enzymes were calculated as previously described (10).
N
S
O
S
R1
R2
R3
R4
R5
R6
Name R1 R2 R3 R4 R5 R6 DengueKi (µM)
DengueSD (µM)
Yellow FeverKi (µM)
Yellow FeverSD (µM)
BG-5 (1013-0252) CH2COOH OCH3 I H I H 1.5 0.55 4.6 1.6
BG-115 (5587305) CH2COOH H H OC6H5H H 4 1.8 7.7 0.15
BG-170 (0806-0194) CH2COOH H H OCH2C6H5H H 2.9 1.2 4.4 1.5
BG-309 (5629185) CH2COOH OCH3 H H OCH3 H >100 >100
BG-310 (4670-0432) CH2CH2COOH OCH3 H H H H 40 2.4 44 4.2
BG-312 (5576225) CH2COOH H H CH(CH3)2H H 18 5.5 17 3.7
BG-313 (5875208) CH2COOH OH Cl H Br H >100 >100
BG-314 (6964247) CH2COOH H Br F H H 27 2.5 27 3.5
BG-315 (5646834) CH2COOH OH Cl H Cl H >100 >100
BG-317 (5580978) CH2COOH Cl Cl H H H 27 4.3 34 9.5
BG-318 (5567899) CH2COOH H Cl Cl H H 11 2.2 16 6.1
BG-319(5221260) CH2COOH Cl H Cl H H 24 8.1 25 6
BG-320 (67975950) CH2COOH F H H Br H 17 2.6 23 6.5
BG-321 (51403100) CH2CH2COOH H H F H H >100 >100
BG-322 (6105937) CH2CH2COOH H H CH(CH3)2H H 12 1.3 13 2.2
BG-323 (7174672) CH2CH2COOH H H C(CH3)3H H 7.5 1.7 9.5 3.2
BG-324 (7426520) CH2CH2COOH Cl Cl H H H 9.9 8 16 3.5
BG-325 (5265280) CH2COOH H H H H H >100 >100
BG-327 (5140309) CH2CH2COOH H H H H H >100 >100
BG-328(5865139) CH2CH2COOH H Cl Cl H H 14 3.3 12 1.4
BG-329 (5537317) CH2CH2COOH F H H H H >100 >100
BG-330 (5558800) CH2CH2COOH Cl H Cl H H 9.8 5.6 9 0.7
BG-331 (5558233) CH2CH2CH2COOH Cl H Cl H H 15 1.3 24 9
BG-332 (6137648) CH2CH2COOCH3Cl H Cl H H >100 >100
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
Table 3. Guanylation inhibition and dengue replicon antiviral assays. Guanylation inhibition assays and replicon assays were performed as described in the Materials and Methods section. Therapeutic index was calculated as CC50/EC50.
Guanylation IC50 (µM)
Guanylation IC50 SEM
(µM)Replicon
EC50 (µM)
Replicon EC50 SEM
(µM)Replicon
CC50 (µM)
Replicon CC50 SEM
(µM)
Replicon Therapeutic Index (TI) ClogP
BG5 6.9 3.1 97.0 17.0 270.0 94.0 2.8 4.14BG115 8.8 4.9 58.0 1.2 74.0 4.7 1.3 3.94BG170 7.8 2.4 71.0 22.5 80.0 32.3 1.1 4.01BG323 7.3 2.9 30.8 1.3 184.0 41.7 6.0 4.22BG324 155.7 56.9 8.5 2.1 12.0 2.5 1.4 3.89BG330 2.6 0.1 8.7 1.2 12.0 2.0 1.4 3.89
on July 25, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
Recommended