Activity of the Anti-Orthopoxvirus Compound ST-246 Against Vaccinia, Cowpox and Camelpox Viruses in Cell Monolayers and Organotypic Raft Cultures

Andrei 21/11/07 15:46 Page 1205

Antiviral Therapy 12:1205–1216 Activity of the anti-orthopoxvirus compound ST-246 against vaccinia, cowpox and camelpox viruses in cell monolayers and organotypic raft cultures

Sophie Duraffour1,2, Robert Snoeck1, Rita de Vos 3, Joost J van Den Oord3, Jean-Marc Crance 2, Daniel Garin2, Dennis E Hruby4, Robert Jordan4, Erik De Clercq1 and Graciela Andrei1*

1Rega Institute For Medical Research, KU Leuven, Leuven, Belgium 2CRSSA Emile Pardé, Virology Laboratory, La Tronche, France 3Pathology Department, UZ Leuven, Leuven, Belgium 4SIGA Technologies, Inc., Corvallis, Oregon, CA, USA

*Corresponding author: Tel: +32 16 33 73 72; Fax: +32 16 33 73 40; E-mail: [email protected]

Background: The potential use of variola virus as a biolog- Results: ST-246 inhibited preferentially the production of ical weapon has renewed efforts in the development of extracellular virus compared with intracellular virus antiviral agents against orthopoxviruses. ST-246 [4- production in HEL and PHK cells (for VV) and in PHK cells trifluoromethyl-N-(3,3a,4,4a,5,5a,6,6a-octahydro-1,3-di (for CMLV). In organotypic epithelial raft cultures, ST-246 oxo-4,6-ethenocycloprop [f]isoindol-2(1H)-yl)-benza- at 20 μg/ml inhibited extracellular VV and CMLV produc- mide] is an anti-orthopoxvirus compound active against tion by 6 logs, whereas intracellular virus yield was several orthopoxviruses including vaccinia virus (VV), reduced by 2 logs. In the case of CPV, both extracellular cowpox virus (CPV), camelpox virus (CMLV), ectromelia and intracellular virus production were completely inhib- virus (ECTV) and variola virus in cell culture. The ited by ST-246 at 20 μg/ml. Histological sections of the compound has been shown to inhibit the release of extra- infected rafts, treated with increasing amounts of drug, cellular virus by targeting the F13L VV protein and to confirmed the antiviral activity of ST-246: the epithelium protect mice from VV, CPV and ECTV orthopoxvirus- was protected and there was no evidence of viral infection. induced disease. Electron microscopic examination confirmed the absence Methods: The antiviral activity of ST-246 was assessed of intracellular enveloped virus forms in VV-, CPV- and against extracellular and intracellular VV, CPV and CMLV CMLV-infected cells treated with 10 μg/ml of ST-246. production in human embryonic lung (HEL) fibroblasts Conclusions: These data indicate that ST-246 is a potent and primary human keratinocyte (PHK) cell monolayers, anti-orthopoxvirus compound; the mode of inhibition is as well as in three-dimensional raft cultures. dependent on the virus and cell type.

Introduction

Smallpox has been eradicated by a worldwide vaccina- of cytomegalovirus (CMV) retinitis in AIDS patients, is tion campaign in the late 1970s [1]. In today’s popula- permitted for use as an emergency treatment in the case tion, the level of residual immunity is considered to be of a smallpox outbreak. CDV exhibits activity in cell low or non-existent. Currently, the potential release of culture against a wide range of DNA viruses including variola virus (VARV) by bioterrorists and the growing adenoviruses, herpesviruses, hepadnaviruses, polyoma- number of cases of human monkeypox have prompted viruses, papillomaviruses and orthopoxviruses [5]. the development of new antiviral chemotherapies CDV has been shown to block the replication of VARV effective against orthopoxvirus infections [2,3]. and monkeypox virus (MPXV) in cell culture and to A number of new treatments are currently the subject protect mice from lethal challenge of ectromelia virus of active investigations. However, antiviral drugs offer (ECTV), vaccinia virus (VV) or cowpox virus (CPV) a combination of chemical stability and simplicity of [6,7]. However, CDV has to be administered by the delivery that is especially attractive from a public health intravenous route due to its poor bioavailability. The perspective. Currently, only cidofovir (CDV; (S)-1-(3- risk of nephrotoxicity induced by CDV can be reduced hydroxy-2-phosphonylmethoxypropyl)cytosine; by administration of probenecid. CDV analogues and HPMPC; Vistide®) [4], a drug approved for the treatment esters of CDV have been shown to block the replication

S Duraffour et al.

of VARV, MPXV, CPV and ECTV in vitro [8–10], and Materials and methods when delivered orally protect mice from a lethal challenge with VV, CPV or ECTV [11–14]. CDV and other Cells known nucleotide analogues act by interfering with Human embryonic lung (HEL) fibroblasts (HEL-299; DNA polymerase activity [15]. ATCC CCL-137) were cultured in MEM Earle’s medium A need exists for active compounds with mechanisms (Gibco, Invitrogen Corporation, UK) supplemented with of action different from that of CDV [16]. ST-246 (4- 10% fetal calf serum (FCS), 1% L-glutamine, 1% non- trifluoromethyl-N-(3,3a,4,4a,5,5a,6,6a-octahydro-1,3- essential amino acids, 1% sodium pyruvate and 1%

dioxo-4,6-ethenocycloprop[f]isoindol-2(1H)-yl)-benza antibiotic/antimycotic at 37˚C in a 5% CO2 atmosphere. mide) has already been described as an orally bioavail- These HEL cells were used for antiviral assays. Primary able antiviral compound that is active against several human keratinocytes (PHKs) were isolated from orthopoxvirus species, including VV, CPV, camelpox neonatal foreskins and cultured as already described virus (CMLV), MPXV, ECTV, and two strains of VARV, [10]. These PHKs were used for both antiviral assays in in Vero, BSC-40 and human foreskin fibroblasts [17]. monolayers and for organotypic raft cultures. Importantly, ST-246 was found to be active against a CDV-resistant CPV variant. The compound was inac- Compounds tive against unrelated DNA- and RNA-containing The sources of the compounds were as follows: CDV viruses, demonstrating specificity for inhibition of (HPMPC), Gilead Sciences, Foster City, CA; and ST-246, orthopoxvirus replication. ST-246 has been shown to ViroPharma, Inc., Exton, PA. act at the level of extracellular virus production, reducing extracellular VV formation by 10-fold while Viruses having little effect on the production of intracellular The following viral strains were used: vaccinia virus virus [17]. One of the potential targets of the compound (VV), strain Copenhagen-GFP, kindly provided by has been validated as the F13L VV protein, which is one Dr R Drillien (E0345 ISERM, EFS-Alsace, France), of the proteins required for wrapping of intracellular cowpox virus (CPV), strain Brighton, and camelpox virus mature virus to form intracellular enveloped virus (CMLV), strain Iran, kindly provided by Dr H Meyer [18–22]. Orally administered ST-246 protected mice (Bundeswehr Institute of Microbiology, Germany) [24]. from lethal orthopoxvirus challenges and prevented poxvirus-induced disease. Mice acquired protective Antiviral assays immunity and were resistant to subsequent lethal chal- The antiviral activities of the compounds against each lenge [17]. ST-246 was proven to be highly effective in of the viruses listed above were evaluated in HEL and animal models of systemic orthopoxvirus disease, even PHK cells. Both cell lines were cultured in 96-well when treatment was delayed up to 72 h post-viral inoc- microtitre plates. Confluent monolayers were infected ulation and dosing was reduced to once daily [23]. with each virus at a multiplicity of infection (MOI) of Unlike CDV, which targets the viral DNA polymerase, 0.01 in 2% FCS. After 2 h of incubation at 37˚C with

ST-246 inhibits virus spread by targeting a protein 5% CO2, residual virus was removed and the infected involved in the release of the virus. cells were further incubated with their respective The present study was conducted to provide a more growth medium containing serial dilutions of the test in-depth view of the antiviral activity of ST-246 against compounds (in duplicate). After 2–3 days for VV and extracellular and intracellular virus production. All CPV or 6 days for CMLV in both HEL and PHK cells, experiments were performed against three the cells were fixed with ethanol and stained with orthopoxviruses, VV strain Copenhagen, CPV strain Giemsa solution. Viral cytopathic effect (CPE) was

Brighton, and CMLV strain Iran [24]. The inhibitory recorded, and the 50% inhibitory concentration (IC50) effects of ST-246 were determined on both extracellular was defined as the compound concentration required

and intracellular virus production in human embryonic to reduce viral CPE by 50%. The IC50 values of the lung fibroblast and primary human keratinocyte mono- compounds tested against each strain were calculated layers, as well as in a three-dimensional model of as the mean of three independent experiments. organotypic epithelial raft cultures. The raft culture system has been considered as an effective model for Cytotoxicity assays evaluating antiviral compounds against some epithe- The cell toxicities of the compounds were evaluated liotropic viruses such as papilloma, herpes and based upon the inhibition of cell growth. The cells poxviruses [10,25–27]. Finally, infected cells treated were seeded into 96-well microtitre plates at with 10 μg/ml of ST-246 were examined using electron 3.5×103 cells/well/100 μl for HEL cells and at microscopy. The mode of action of this compound is 5×103 cells/well/100 μl for PHK cells. After 1 day also discussed. (HEL cells) or 3 days (PHK cells) of incubation, 100 μl

Inhibitory effects of ST-246 on orthopoxvirus replication

of medium containing serial dilutions of the test were included for the normal differentiated epithelium compounds were added (in duplicate). After 4 days of (uninfected rafts), viral replication (infected untreated incubation for both cell lines, the cells were trypsinized rafts), and rafts infected and treated with several and the cell number per well was determined with a concentrations of CDV (20, 5, 2, 0.5 and 0.2 μg/ml). coulter counter. Cytotoxicity was expressed as the At day 5 (VV or CPV) or 7 (CMLV) post-infection, the

50% cytotoxic concentration (CC50), defined as the first set of rafts was fixed in 10% buffered formalin concentration required to reduce cell growth by 50%. and embedded in paraffin. Micrometre sections were stained with haematoxylin-eosin for histological exam- Virus yield assays ination. The second set of rafts was first washed gently The effects of several dilutions of CDV (20, 5, 2, 0.5 with 5 ml of sterile PBS to collect extracellular viruses, and 0.2 μg/ml) and ST-246 (20, 2, 0.2, 0.02, 0.002, and then immersed in 5 ml of sterile PBS to collect 0.0002, 0.00002 and 0.000002 μg/ml) were evaluated intracellular viruses. Samples were frozen at -20˚C. in both HEL and PHK cells infected with VV, CPV or Rafts were thawed, centrifuged at 1,800 rpm for CMLV. Both cell lines were grown in six-well microtitre 10 min, and the supernatants containing the released plates, and the confluent monolayers were infected virus were collected. Extracellular and intracellular with each virus at an MOI of 0.01. After 2 h incuba- fractions were used for quantification of infectious

tion at 37˚C in a 5% CO2 atmosphere, residual virus virus. Titration of infectious virus was performed in was removed, each well was washed two times with HEL cells with 10-fold dilutions of the samples, and 2 ml of sterile phosphate-buffered saline (PBS) per well, the viral titre was expressed as the number of PFU/raft. and the different concentrations of the test compounds were added. A virus control (infected untreated cells) Electron microscopy and a cell control (non-infected, non-treated cells) were Confluent 75 cm2 flasks of HEL cells were infected included at each time point. The supernatants and the with VV, CMLV or CPV in the presence or absence of infected cell monolayers were recovered separately to ST-246 (10 μg/ml) at an MOI of 0.06 and incubated at investigate the reduction of the production of extracel- 37˚C. At 30 h post-infection (hpi) for VV and CPV and lular and intracellular viruses. After 2 days (VV and 48 hpi for CMLV, the cells were first washed with CPV on HEL cells), 3 days (VV and CPV on PHKs) or trypsin and then trypsinized. The cells were collected 6 days (CMLV on both cell types) post-infection, the by centrifugation and then fixed with 2.5% glutaralde- supernatants were harvested and frozen at -20˚C. The hyde in 0.1 mol/l PBS at 4˚C overnight. After 1 h post- cell monolayers were washed two times with 2 ml of fixation in 1% osmium tetroxide, 0.1 mol/l PBS at 4˚C, sterile PBS per well and then frozen at -20˚C after the the samples were dehydrated in a graded series of addition of 3 ml of sterile PBS per well. In order to titre alcohol and embedded in epoxy resin. Ultra-thin the extracellular virus, the supernatants were thawed sections 50–60 nm were cut, stained with uranyl and the virus yield was evaluated by virus titration in acetate and lead citrate and examined at 50 kV using a 96-well microtitre plates. For the intracellular titration, Zeiss EM 900 electron microscope. Images were the microplates were thawed and the disrupted cells recorded digitally with a Jenoptik Progress C14 camera contained in 3 ml PBS were collected and centrifuged at system operated using Image-Pro express software. 1,800 rpm for 10 min. The titration was performed with serial 10-fold dilutions of the samples, and the Results viral titre was expressed in plaque-forming units (PFU)

per ml. The IC99 values were defined and calculated as Antiviral and cytotoxicity assays the compound concentrations required to reduce the The antiviral activities of the two compounds ST-246 viral titre by 99%. and CDV against the three orthopoxviruses were determined in HEL fibroblasts and in PHK monolayers Organotypic epithelial raft cultures (Table 1). In both cell lines, ST-246 exhibited a more Epidermal equivalents were prepared as already potent activity against the different orthopoxviruses described [10]. The epithelial cells were then allowed to tested than the reference compound CDV. ST-246 stratify for 5 days (CMLV) or 7 days (VV and CPV) inhibited poxvirus-induced cytopathic effect (CPE) with μ μ and, at this time, the cultures were infected with 100 l IC50 values ranging from 0.007 to 0.16 g/ml in HEL of VV, CPV or CMLV (1,000 PFU/100 μl), which were and from 0.003 μg/ml to 0.013 μg/ml in PHK cells,

placed on top of the rafts. To test the effects of ST-246 while IC50 values for CDV were in the range of 2.6–13.3 on the replication of VV, CPV and CMLV ex vivo, at μg/ml in HEL cells and 1.7–3.2 μg/ml in PHK cells. The the time of infection, medium containing different concentrations required to reduce cell growth by 50% > μ ± μ concentrations of this compound (20, 2, 0.2, 0.02 and (CC50) were 200 g/ml and 60 24 g/ml for ST-246 0.002 μg/ml) was added. In the same assay, controls and CDV, respectively, in HEL cells. The selectivity

Antiviral Therapy 12:8 1207 Andrei 21/11/07 15:46 Page 1208

S Duraffour et al.

Table 1. Antiviral activities of ST-246 and CDV against VV, CPV and CMLV in HEL and PHK cell monolayers HEL cells PHK cells † † Virus Compound IC50, µg/ml* Selectivity index IC50, µg/ml* Selectivity index

VV ST-246 0.007 ±0.003 >28,570 0.003 ±0.00006 26,970 CDV 3.7 ±0.5 16.5 2.3 ±1.0 1.0 CPV ST-246 0.16 ±0.09 >1,270 0.013 ±0.0005 6,225 CDV 13.3 ±3.0 4.5 3.2 ±0.6 0.7 CMLV ST-246 0.03 ±0.004 >6,670 0.02 ±0.01 4,495 CDV 2.6 ±1.2 23.3 1.7 ±0.8 1.4

± *Concentration required to inhibit 50% of virus-induced cytopathic effect. The IC50 values for each compound are the mean SD of the IC50 values of at least three independent experiments. The concentrations required to reduce cell growth by 50% were >200 µg/ml and 60 ±24 µg/ml, respectively, for ST-246 and CDV in HEL ± ± ± cells, and 80.9 14.8 µg/ml and 2.36 2.88 µg/ml for ST-246 and CDV in PHK cells, respectively. The CC50 values for each compound represent the mean SD of the † CC50 values from at least two independent assays. Selectivity index (ratio of CC50 to IC50). HEL; human embryonic lung fibroblasts; CMLV, camelpox virus; CPV, cowpox virus; PHK, primary human keratinocytes; VV, vaccinia virus.

indexes (SIs) for ST-246 in HEL cells were 280- to replication by 3 to 5 logs CPV, respectively (Figure 1B). 1,700-fold higher than those obtained for CDV In HEL cells, 20 μg/ml of ST-246 reduced by 3 logs both

(Table 1). The CC50 values obtained in PHKs were 80.9 intracellular and extracellular CPV production, but did ±14.8 μg/ml for ST-246 and 2.36 ±2.88 μg/ml for not abolish completely CPV replication (Figure 1B). The

CDV. Moreover, ST-246 exhibited the highest SIs in IC99 values of ST-246 for extracellular virus production monolayer cultures of PHKs, with values increasing were 1.9 ±0.1 μg/ml in HEL cells and 0.12 ±0.08 μg/ml from 2,900- to 26,970-fold higher than SIs obtained in PHK cells, compared with 1.3 ±0.7 μg/ml (HEL cells) for CDV. The antiviral activity of ST-246 was also and 0.15 ±0.007 μg/ml (PHK cells) for intracellular

determined against two CDV-resistant strains of VV virus yield (Table 2). The IC99 I/E ratios were compa-

strain Lederle, and showed an IC50 value of rable for ST-246- and CDV-treated cells. μ 0.005 g/ml for both strains; the IC50 values of CDV In the case of CMLV, we found different results were >50 μg/ml against both resistant strains (data not depending on the cell line used. As shown in shown) [28]. We also assessed ST-246 against the Figure 1C and in Table 2, both extracellular and parapoxvirus orf, but the compound showed an intracellular CMLV formation in HEL cells were > μ μ IC50 50 g/ml (data not shown). inhibited by 3 logs with 2 g/ml of ST-246. IC99 values

obtained were similar giving an IC99 I/E ratio of 1.7.

Intracellular and extracellular virus yield assays In a similar manner, the IC99 values for CDV were also The inhibitory effect of ST-246 on intracellular and comparable for both extracellular (18.9 ±0.6 μg/ml) extracellular virus production was evaluated in HEL and intracellular (17.2 ±3.0 μg/ml) virus production

cells and in PHK monolayers. As shown in Figure 1A, giving an IC99 I/E ratio of 0.9. In contrast, intra- ST-246 inhibited preferentially the production of extracellular CMLV production was inhibited by <1 log in cellular VV compared with intracellular virus produc- PHK cells at 2 μg/ml of ST-246, while extracellular tion in both cell lines. A treatment of 2 μg/ml of ST-246 virus formation was decreased by >2 logs (Figure 1C).

reduced the viral titres of extracellular viruses by 5 logs The IC99 values obtained were different with values of compared with a decrease of 2 logs for intracellular 5.1 ±1.7 μg/ml for extracellular virus and ≥20 μg/ml

viruses. Thus, in PHK monolayers IC99 values for intra- for intracellular CMLV yield, giving an IC99 I/E ratio cellular and extracellular virus yields were, respectively, >3.9 (Table 2). 0.10 ±0.11 μg/ml and 0.009 ±0.010 μg/ml, giving a

ratio of intracellular IC99 to extracellular IC99 (I/E) of 11 Organotypic epithelial raft cultures (Table 2). In a similar manner, an I/E ratio of 187 was The antiviral activity of ST-246 was also evaluated in observed in VV-infected HEL cells treated with ST-246. human organotypic epithelial raft cultures. As shown

In comparison, the IC99 values observed with CDV in in Figure 2, PHK cells from uninfected rafts were able PHK cells were 4.8 ±0.04 μg/ml for extracellular VV to generate a fully differentiated multi-stratified epithe- and 4.6 ±0.06 μg/ml for intracellular VV. After treat- lium, similar to that seen in the clinic setting. Rafts ment with CDV of VV-infected cells, the I/E ratios were infected with VV (Figure 2A), CPV (Figure 2B) or 0.8 in HEL cells and 0.9 in PHK cells (Table 2). CMLV (Figure 2C) allowed virus replication with In contrast to VV, ST-246 offered a similar level of evidence of cytopathic effects. Histopathological exam- inhibition of extracellular and intracellular CPV ination revealed cytoplasmic swelling and ballooning of production in HEL and PHK monolayers reducing the keratinocytes in the infected controls. A reduction of

Inhibitory effects of ST-246 on orthopoxvirus replication

Figure 1. The inhibitory effect of ST-246 on the production of VV, CPV and CMLV A HEL cells PHK cells

6 6

5 5

4 4

3 3

Log PFU/ml 2 Log PFU/ml 2

1 1

0 0 0 0 0.2 2 20 0.000002 0.00002 0.0002 0.002 0.02 0.2 2 0.000002 0.00002 0.0002 0.002 0.02 20 Concentration, µg/ml Concentration, µg/ml B

6 6

5 5 4 4

3 3

Log PFU/ml 2 Log PFU/ml 2

1 1

0 0 0 0 2 20 0.000002 0.00002 0.0002 0.002 0.02 0.2 2 0.000002 0.00002 0.0002 0.002 0.02 0.2 Concentration, µg/ml 20 Concentration, µg/ml C 6 6

5 5

4 4

3 3

Log PFU/ml 2 Log PFU/ml 2

1 1

0 0 0 0 2 20

0.000002 0.00002 0.0002 0.002 0.02 0.2 2 0.000002 0.00002 0.0002 0.002 0.02 0.2 Concentration, µg/ml 20 Concentration, µg/ml

Extracellular Intracellular

The inhibitory effect of ST-246 on the production of (A) vaccinia virus (VV), (B) cowpox virus (CPV) and (C) camelpox virus (CMLV). The effects of several concentrations of ST-246 on extracellular and intracellular virus production were determined after 2–6 days post-infection depending on the virus (see Materials and methods for details). Viral titres were determined by plaque assay on HEL cells. The error bars represent the standard deviation of the means of duplicate virus yield determinations. HEL, human embryonic lung fibroblasts; PHK, primary human keratinocytes.

Antiviral Therapy 12:8 1209 Andrei 21/11/07 15:46 Page 1210

S Duraffour et al.

Table 2. Inhibition of extracellular and intracellular VV, CPV and CMLV production in HEL and PHK cell monolayers

IC99, µg/ml* Virus Cell type Compound E† I‡ I/E§

VV HEL ST-246 0.03 ±0.02 5.6 ±6.3 187 CDV 12.9 ±1.0 10.6 ±7.9 0.8 PHK ST-246 0.009 ±0.01 0.10 ±0.11 11 CDV 4.8 ±0.04 4.6 ±0.06 0.9 CPV HEL ST-246 1.9 ±0.1 1.3 ±0.7 0.7 CDV 4.8 ±0.04 5.4 ±0.7 0.3 PHK ST-246 0.12 ±0.08 0.15 ±0.007 1.2 CDV 16.4 ±1.2 16.9 ±0.1 1.0 CMLV HEL ST-246 1.1 ±0.9 1.9 ±0.03 1.7 CDV 18.9 ±0.6 17.2 ±3.0 0.9 PHK ST-246 5.1 ±1.7 ≥20 ≥3.9 CDV 4.5 ±0.7 10.8 ±8.5 2.4

*Concentration of ST-246 or CDV required to inhibit 99% of extracellular and intracellular vaccinia virus (VV), cowpox virus (CPV) and camelpox virus (CMLV) production ± in human embryonic lung (HEL) fibroblasts and primary human keratinocytes (PHK). The IC99 values represent the mean SD of the IC99 values from two independent virus † ‡ § yield experiments (see Materials and methods for details). Extracellular viruses (E). Intracellular viruses (I). Ratio of intracellular IC99 to extracellular IC99 values.

Figure 2. Activities of ST-246 in organotypic epithelial raft cultures

Infected 0.002 µg/ml 0.02 µg/ml 0.2 µg/ml 2 µg/ml 20 µg/ml Uninfected

Rafts were infected after 7 days of differentiation with (A) vaccinia virus (VV) or (B) cowpox virus (CPV) and after 5 days of differentiation with (C) camelpox virus (CMLV). They were then treated with different concentrations of ST-246 (see Materials and methods for details). At 12 days of differentiation, the rafts were fixed and stained with haematoxylin-eosin. The far right panels show uninfected rafts; the far left panels were infected but untreated with ST-246. Magnification, ×40 for all panels.

the cytopathic effects was observed following the (Figure 3A,3B,3C). A complete inhibition of VV and addition of an increasing amount of drug. In some CMLV replication was observed at 20 μg/ml of CDV experiments, we noticed a cytotoxic effect at 20 μg/ml (Figure 3A,3C). As shown in Figure 3A, 20 μg/ml of of ST-246 (Figure 2B). The activity of ST-246 against ST-246 inhibited extracellular VV production by 6 VV, CPV and CMLV replication was also confirmed in logs, while intracellular virus yield was only reduced the extracellular and intracellular virus yield of the by 2 logs, which is in agreement with the results in rafts and compared with the inhibitory effect of CDV monolayers. Similarly, extracellular CMLV production

Inhibitory effects of ST-246 on orthopoxvirus replication

was reduced by 4.5 logs at 20 μg/ml of ST-246, infected and treated cells, but we could not detect any whereas intracellular virus production was only CMLV enveloped forms (Figure 4C iii,iv). decreased by 2 logs (Figure 3C). As expected, the effect of ST-246 against CMLV replication found in the three- Discussion dimensional model was similar to that seen in PHK monolayers but differed from the observations in HEL In these studies, ST-246 was shown to be a potent cells (Figure 1C). In contrast, in the case of CPV (Figure inhibitor of the orthopoxviruses VV, CPV and CMLV, 3B), both extracellular and intracellular virus production confirming that it is a very promising antiviral drug were completely inhibited by ST-246 at 20 μg/ml. At candidate. In cell monolayers as well as in three- 2 μg/ml of ST-246, the intracellular virus fraction had a dimensional epithelial raft cultures, the potency of mean virus titre of 1 log and the extracellular CPV repli- ST-246 appears to exceed that of CDV, which is today cation was abolished. A comparable inhibition profile the best-known poxvirus antiviral drug. Whereas CDV μ μ was observed with CDV at 5 and 20 g/ml. has IC50 values ranging from 2.6–13.3 g/ml against

the three orthopoxviruses assessed and a CC50 of 60 Electron microscopic analyses ±24 μg/ml in HEL cells, ST-246 inhibits poxvirus- To further investigate the defect in extracellular virus induced cytopathic effect under similar conditions μ formation due to ST-246, VV-, CPV- or CMLV-infected with IC50 values ranging from 0.007–0.16 g/ml HEL cells were either treated or untreated with without significant cytotoxicity (Table 1). These 10 μg/ml of ST-246 and used to prepare thin sections. results are correlated with those obtained by Quenelle Cultures were fixed at 30 h post-infection for VV and and coworkers [23] in human foreskin fibroblast cells CPV and at 48 h post-infection for CMLV and and by Yang et al. [17] in Vero cells. In primary human processed for electron microscopy analysis. As shown keratinocyte cells, ST-246 demonstrated a higher in Figure 4A (i) with VV there were virus particles inhibitory activity and a lower toxicity when within the cell including immature virions (IV), intra- compared to CDV (Table 1). We also demonstrated cellular mature virus (IMV) and many intracellular that ST-246 was active against two CDV-resistant enveloped viruses (IEV). At higher magnification, we strains of VV strain Lederle [28]. ST-246 has already could see that the IEV observed were surrounded by a been shown to be inactive against non- double-layer of membrane derived from the trans- orthopoxviruses [17], and we confirmed its inactivity Golgi network (Figure 4A ii). In the ST-246-treated against orf, a parapoxvirus. samples, we observed less infected cells than in In vivo toxicity of ST-246 has been investigated untreated controls. IV, as well as IMV particles were and the compound demonstrated good oral bioavail- also present in treated infected cells (Figure 4A iii). ability (rat model), safety and was well tolerated in IMV particles were able to migrate from the viral facto- mice and cynomolgus macaques with no adverse ries, but we could not find wrapped virions in the effects at levels of 2,000 mg/kg and 600 mg/kg, treated samples (Figure 4A iv). respectively [29]. Recently, a child with a life-threat- In the case of CPV, we confirmed the presence of the ening eczema vaccinatum was successfully treated full repertoire of virion intermediates, including IV with ST-246 [30]. In the case of CDV, which is poorly and IMV particles (not shown) in cultures that were bioavailable, dose-limiting nephrotoxicity requires infected and untreated. We detected many enveloped the use of probenecid [5]. forms within the cells (Figure 4B i) surrounded by a The activity of ST-246 against VV, CPV and CMLV double-layer of membrane, as shown at higher magni- was compared with that of CDV in virus yield assays in fication (Figure 4B ii). We could still see viral factories fibroblast and keratinocyte monolayers, as well as in with IV and IMV in CPV-infected cells treated with three-dimensional cultures. The raft system has already ST-246, but we did not observe any IEV particles been well described and proved to be a reliable model (Figure 4B iii,iv). for the study of the antiviral activities of several Concerning CMLV, we found many infected and compounds against epitheliotropic viruses lysed cells in the virus control. Viral factories were [9,10,25,26]. The examination of the histological found containing IV and many IMVs as compared with sections obtained from this three-dimensional model, the observations for VV and CPV (Figure 4C i). In mimicking the morphological and physiological char- general, we found few IEV particles in the untreated acteristics of the human skin, allowed us to confirm the samples. As shown in Figure 4C (ii), we observed IEV inhibitory effect of ST-246 on viral growth particles surrounded by membranes. In the samples (Figure 2A,2B,2C). As shown in Figure 2, after treated with ST-246, we found fewer infected cells and 5–7 days of ST-246 treatment, VV-, CPV- or CMLV- fewer lysed cells in comparison with the untreated infected rafts were fully protected at 20 μg/ml. Toxic control. We saw IV and IMV particles within the effects on the epithelium seen at 20 μg/ml of ST-246 in

Antiviral Therapy 12:8 1211 Andrei 21/11/07 15:46 Page 1212

S Duraffour et al.

Figure 3. The effects of CDV and ST-246 on extracellular and intracellular virus yield in organotypic raft cultures A CDV ST-246 7 7 6 6 5 5 4 4 3 3 Log PFU/raft Log PFU/raft 2 2 1 1 0 0 0 0 0.2 0.5 2 5 0.00.2 0.02 0.2 2 B Concentration, µg/ml 20 Concentration, µg/ml 20 6 6

5 5

4 4

3 3

Log PFU/raft 2 Log PFU/raft 2

1 1

0 0 0 0 0.2 0.5 2 µ 5 0.00.2 0.02 µ0.2 2 C Concentration, g/ml20 Concentration, g/ml 20 6 6

5 5

4 4

3 3

Log PFU/raft 2 Log PFU/raft 2

1 1

0 0 0 0 0.2 2 0.5 5 0.002 0.02 0.2 2 Concentration, µg/ml 20 Concentration, µg/ml 20

Extracellular Intracellular

Rafts were infected after 7 days of differentiation with (A) vaccinia virus (VV) or (B) cowpox virus (CPV) and after 5 days of differentiation with (C) camelpox virus (CMLV). They were then treated with different concentrations of CDV or ST-246 (see Materials and methods for details). At 12 days of differentiation, extracellular and intracellular viruses were collected. Intracellular and extracellular virus production were determined by plaque assay of HEL cell monolayers. The error bars represent the standard deviation of the means of duplicate virus yield determinations. HEL, human embryonic lung fibroblasts; PHK, primary human keratinocytes.

Inhibitory effects of ST-246 on orthopoxvirus replication

some experiments may have been donor-dependent. Figure 4. Electron microscopic analysis of VV-, CPV- and First evidences of cytoplasmic ballooning and swelling CMLV-infected cells treated with ST-246 due to viral infection were observed in histopatholog- A ical sections at 0.02 μg/ml for CPV- and at 0.002 μg/ml for VV- and CMLV-infected rafts treated with ST-246. (i) (ii) ST-246 has been shown to be a specific inhibitor of extracellular orthopoxvirus formation by targeting the F13L protein of VV. ST-246 resistance has been mapped to the CPV V061 gene, which encodes for a major enveloped protein homologous to the VV F13L gene product [17–19]. The single amino acid change (position 277) responsible for the ST-246-resistant (iii) (iv) phenotype was re-engineered back into wild-type CPV and VV genomes, and the resulting recombinants were found to be resistant to ST-246 [17]. The non-glycosy- lated and palmitoylated F13L protein is a component of IEV, cell-associated enveloped virus (CEV), and extracellular enveloped virus (EEV) [31–36]. This protein is present in the Golgi network as well as in B endosomes. Analyses of a VV mutant lacking the entire (i) (ii) F13L gene (vΔ13L) have shown that the F13L protein is required for efficient IMV wrapping to form IEV [18,20,21,37,38]. The vΔ13L mutants exhibited tiny plaque size, associated with a normal production of IMV, and a 10-fold decrease of EEV formation [18]. Payne and Kristenson [39] have shown that extracellular VV particle formation in cell culture was inhibited (iii) (iv) by N1-isonicotinoyl-N2-3-methyl-4-chlorobenzoylhydrazine (IMCBH) by targeting the F13L protein. As shown in Figures 1A and 3A, ST-246 at 20 μg/ml inhibits preferentially the production of extracellular VV by 6 logs compared with intracellular VV formation, which decreases by 2 logs in both monolayers and three-dimensional systems. According to these results C and the observations of Yang et al. [17], the F13L protein could be one of the potential targets of (i) (ii) ST-246, which would act by preventing the formation of extracellular virus. In the case of CMLV, we found a similar mode of action of ST-246 that acts by inhibiting preferentially extracellular virus formation compared with intracellular virus production in keratinocyte monolayers (Figure 1C right-hand panel) and in the rafts (Figure 3C). However, ST-246 (iii) (iv) had a different antiviral effect on CMLV replication in HEL cells, where it inhibited both extracellular and intracellular virus yields. Moreover, this behav- iour was also found in the experiments performed with CPV. In the case of CMLV, the differences we observed in HEL and PHK monolayers could be explained by a cell-specific event in morphogenesis. The effect of ST-246 on virus replication was evaluated in cells infected with (A) vaccinia virus (VV), (B) cowpox virus (CPV) and (C) camelpox virus (CMLV). Indeed, the extent of EEV production is known to Human embryonic lung fibroblasts (HEL) cells monolayers were infected with depend on the cell type and virus strain [40]. VV each virus at a multiplicity of infection of 0.06. After 1 h incubation, residual virus was removed and the test compound added (10 µg/ml). Treated and IHD-J, for instance, releases considerably more EEV untreated infected cells were processed for electron microscopic examination at from RK-13 cells than from Statens Seruminstitut 30 h post-infection: (i) and (ii) untreated cells; (iii) and (iv) treated cells. Scale bars are shown on the top of each panel. G, Golgi; IEV, intracellular enveloped Rabbit Cornea (SIRC) or HeLa cells [40]. virus; IMV, intracellular mature virus; IV, immature virions.

Antiviral Therapy 12:8 1213 Andrei 21/11/07 15:46 Page 1214

S Duraffour et al.

Because the F13L gene is highly conserved among Another possibility is that the inhibition of VV, CPV orthopoxviruses (V061L CPV and V048L CMLV and CMLV production cannot entirely be explained by proteins having 99% and 98% amino acid identity the targeting of a single protein, F13L, and that one or with F13L VV protein, respectively), we expected to several other proteins of CPV and CMLV are also see similar inhibition profiles for VV, CPV and CMLV targeted by ST-246. One potential target would be a in both cell lines and in the rafts with regards to the conserved protein that exhibits variations in amino production of extracellular and intracellular viruses. acid sequence among orthopoxviruses, conferring Interestingly, there is a similar effect of ST-246 on differences in conformation and, consequently, on extracellular and intracellular CPV production in both drug susceptibility. Although VV, CPV and CMLV systems (Figures 1B and 3B). Both extracellular and belong to the same genus, several lines of evidence intracellular CPV yields were reduced by 3 and 4 logs, revealed differences between them. For instance, respectively, in fibroblast and keratinocyte monolayers Pulford et al. [42] showed that among (Figure 1B) and by 5 logs in raft cultures (Figure 3B), orthopoxviruses the A13L protein exhibits conserved which differs from what was observed with VV. In sequences at the N and C termini and species-specific contrast to VV, these results in monolayers could heterogeneity in the central domain. It can be hypoth- suggest that the spread of CPV and CMLV (in HEL esized that different orthopoxvirus species have cells) may be less dependent on extracellular virus evolved distinct amino acid motifs across domains of production and is cell-type specific (CMLV), although conserved proteins, which have led to a variable the I/E ratios could also potentially reflect the stage of susceptibility of these proteins towards antiviral the viral infection at the time of harvest. compounds. Recently, studies have demonstrated Upon electron microscopy analyses we found that in differences in the activities of the CrmB proteins the presence of 10 μg/ml of ST-246, IEV particles were encoded by VARV, MPXV and CPV. These proteins not produced for each of the three viruses studied. We target tumour necrosis factor. Sequence analyses could conclude for CPV and VV that ST-246 showed a revealed a high homology of these proteins. In addi- block in the trans-Golgi network wrapping of IMV. tion, a comparative analysis of the C-terminal These observations were correlated with the results smallpox-virus-encoded chemokine receptor obtained in virus yield assays for HEL cell monolayers. (SECRET) domain sequences of CrmB proteins was These data indicated that the primary block in performed. Differences in the amino acid sequences of morphogenesis caused by ST-246 is situated at the these domains characteristic of each particular IMV–IEV transition for VV and CPV. In the case of orthopoxvirus species were detected. It was assumed CMLV, it was difficult to draw a conclusion as few IEV that the species-specific distinctions between the CrmB particles were found in the untreated control. In proteins might underlie the differences in their comparison to VV and CPV, we observed many lysed physicochemical and biological properties [43]. cells and an obvious accumulation of IMV particles Taken together, our results confirm the potency of within the cell in CMLV-infected cells from 30 h post- ST-246 against orthopoxviruses and further support infection (unpublished data). This observation might the notion that the F13L protein is one of the antiviral suggest that the spread of CMLV in HEL cells could be targets of ST-246. Our data also point to some differ- more dependent on IMV forms than on EEV particles. ences in the mode of action of ST-246 against different Our results suggest several hypotheses concerning the orthopoxviruses such as VV, CMLV and CPV. ST-246 antiviral activity of ST-246. One hypothesis is that the represents a promising drug for the treatment of F13L protein has species-specific biological properties human orthopoxvirus diseases. and plays a role in the different mechanisms of spread of VV, CPV and CMLV. EEV forms precede the systemic Acknowledgements viraemic spread to target organs playing a significant role in viral pathogenesis [41]. In a recent study, it has We would like to thank Dr H Meyer (Bundeswehr been shown that the dosing of ST-246 is necessary for Institute of Microbiology, Munich, Germany) for periods >5 days in CPV intranasally infected mice in providing the camelpox virus. We thank Steven contrast to VV intranasally infected mice [23]. These Carmans, Lies Van den Heurck, Anita Camps, results were correlated with the viral detection in target Rolande Renwart and Christiane Armee for excellent organs on days 3–5 and on days 5–7 post-inoculation technical assistance. SD is a recipient of a grant of the for VV- and CPV-infected mice, respectively [17,23]. Délégation Générale pour l’Armement (DGA). This Our in vitro experiments could be in agreement with work was supported by grant G.0267.04 from the these animal studies, showing that the spread of VV is Fund for Scientific Research Flanders (FWO, more dependent on extracellular virus production than Vlanderen) and by grant GAI062540.01 from NIH, the spread of CPV and CMLV (in HEL cells). Bethesda, MD.

Inhibitory effects of ST-246 on orthopoxvirus replication

References 21. Husain M, Moss B. Vaccinia virus F13L protein with a conserved phospholipase catalytic motif induces 1. Geddes AM. The history of smallpox. Clin Dermatol 2006; colocalization of the B5R envelope glycoprotein in 24:152–157. post-Golgi vesicles. J Virol 2001; 75:7528–7542. 2. Damon IK, Roth CE, Chowdhary V. Discovery of 22. Smith GL, Law M. The exit of vaccinia virus from infected monkeypox in Sudan. N Engl J Med 2006; 355:962–963. cells. Virus Res 2004; 106:189–197. 3. Mackelprang RW, Mackelprang RD, Thirkill AD. 23. Quenelle DC, Buller RM, Parker S et al. Efficacy of delayed Bioterrorism and smallpox: policies, practices, and treatment with ST-246 given orally against systemic implications for social work. Soc Work 2005; 50:119–127. orthopoxvirus infections in mice. Antimicrob Agents Chemother 2007; 51:689–695. 4. Lalezari JP, Stagg RJ, Kuppermann BD et al. Intravenous cidofovir for peripheral cytomegalovirus retinitis in patients 24. Ramyar H, Hessami M. Isolation, cultivation and with AIDS. A randomized, controlled trial. Ann Intern Med characterization of camel pox virus. Zentralbl 1997; 126:257–263. Veterinarmed B 1972; 19:182–189. 5. De Clercq E. Clinical potential of the acyclic nucleoside 25. Andrei G, van den OJ, Fiten P et al. Organotypic epithelial phosphonates cidofovir, adefovir, and tenofovir in raft cultures as a model for evaluating compounds against treatment of DNA virus and retrovirus infections. Clin alphaherpesviruses. Antimicrob Agents Chemother 2005; Microbiol Rev 2003; 16:569–596. 49:4671–4680. 6. Knorr CW, Allen SD, Torres AR, Smee DF. Effects of 26. Dal Pozzo F, Andrei G, Holy A et al. Activities of acyclic cidofovir treatment on cytokine induction in murine models nucleoside phosphonates against Orf virus in human and of cowpox and vaccinia virus infection. Antiviral Res 2006; ovine cell monolayers and organotypic ovine raft cultures. 72:125–133. Antimicrob Agents Chemother 2005; 49:4843–4852. 7. Robbins SJ, Jackson RJ, Fenner F et al. The efficacy of 27. Delvenne P, Hubert P, Jacobs N et al. The organotypic cidofovir treatment of mice infected with ectromelia culture of HPV-transformed keratinocytes: an effective (mousepox) virus encoding interleukin-4. Antiviral Res in vitro model for the development of new 2005; 66:1–7. immunotherapeutic approaches for mucosal (pre)neoplastic lesions. Vaccine 2001; 19:2557–2564. 8. Kern ER, Hartline C, Harden E et al. Enhanced inhibition of orthopoxvirus replication in vitro by alkoxyalkyl esters 28. Andrei G, Gammon DB, Fiten P et al. Cidofovir resistance of cidofovir and cyclic cidofovir. Antimicrob Agents in vaccinia virus is linked to diminished virulence in mice. Chemother 2002; 46:991–995. J Virol 2006; 80:9391–9401. 9. Lebeau I, Andrei G, Dal Pozzo F et al. Activities of 29. Bailey TS, Rippin SR, Opsitnick E et al. N-(3,3a,4,4a,5,5a, alkoxyalkyl esters of cidofovir (CDV), cyclic CDV, and (S)- 6,6a-Octahydro-1,3-dioxo-4,6-ethenocycloprop[f]isoindol- 9-(3-hydroxy-2-phosphonylmethoxypropyl)adenine against 2-(1H)-yl)carboxamides: identification of novel orthopoxviruses in cell monolayers and in organotypic orthopoxvirus egress inhibitors. J Med Chem 2007; cultures. Antimicrob Agents Chemother 2006; 50:1442–1444. 50:2525–2529. 30. Marris E. Dramatic rescue relieves rare case of smallpox 10. Snoeck R, Holy A, Dewolf-Peeters C et al. Antivaccinia infection. Nat Med 2007, 13:517–517. activities of acyclic nucleoside phosphonate derivatives in 31. Grosenbach DW, Ulaeto DO, Hruby DE. Palmitylation of epithelial cells and organotypic cultures. Antimicrob Agents the vaccinia virus 37-kDa major envelope antigen. Chemother 2002; 46:3356–3361. Identification of a conserved acceptor motif and biological 11. Buller RM, Owens G, Schriewer J et al. Efficacy of oral relevance. J Biol Chem 1997; 272:1956–1964. active ether lipid analogues of cidofovir in a lethal 32. Grosenbach DW, Hruby DE. Analysis of a vaccinia virus mousepox model. Virology 2004; 318:474–481. mutant expressing a nonpalmitylated form of p37, a 12. Quenelle DC, Collins DJ, Wan WB et al. Oral treatment of mediator of virion envelopment. J Virol 1998; cowpox and vaccinia virus infections in mice with ether 72:5108–5120. lipid esters of cidofovir. Antimicrob Agents Chemother 33. Grosenbach DW, Hansen SG, Hruby DE. Identification and 2004; 48:404–412. analysis of vaccinia virus palmitylproteins. Virology 2000; 13. Smee DF, Bailey KW, Sidwell RW. Treatment of lethal 275:193–206. vaccinia virus respiratory infections in mice with cidofovir. 34. Koonin EV. A duplicated catalytic motif in a new Antivir Chem Chemother 2001; 12:71–76. superfamily of phosphohydrolases and phospholipid 14. Smee DF, Sidwell RW. A review of compounds exhibiting synthases that includes poxvirus envelope proteins. Trends anti-orthopoxvirus activity in animal models. Antiviral Res Biochem Sci 1996; 21:242–243. 2003; 57:41–52. 35. Roper RL, Moss B. Envelope formation is blocked by 15. Magee WC, Hostetler KY, Evans DH. Mechanism of mutation of a sequence related to the HKD phospholipid inhibition of vaccinia virus DNA polymerase by cidofovir metabolism motif in the vaccinia virus F13L protein. diphosphate. Antimicrob Agents Chemother 2005; J Virol 1999; 73:1108–1117. 49:3153–3162. 36. Sung TC, Roper RL, Zhang Y et al. Mutagenesis of 16. LeDuc JW, Damon I, Relman DA, Huggins J, Jahrling PB. phospholipase D defines a superfamily including a trans- Smallpox research activities: U.S. interagency collaboration, Golgi viral protein required for poxvirus pathogenicity. 2001. Emerg Infect Dis 2002; 8:743–745. EMBO J 1997; 16:4519–4530. 17. Yang G, Pevear DC, Davies MH et al. An orally 37. Borrego B, Lorenzo MM, Blasco R. Complementation of bioavailable antipoxvirus compound (ST-246) inhibits P37 (F13L gene) knock-out in vaccinia virus by a cell line extracellular virus formation and protects mice from lethal expressing the gene constitutively. J Gen Virol 1999; orthopoxvirus challenge. J Virol 2005; 79:13139–13149. 80:425–432. 18. Blasco R, Moss B. Extracellular vaccinia virus formation 38. Husain M, Moss B. Intracellular trafficking of a and cell-to-cell virus transmission are prevented by deletion palmitoylated membrane-associated protein component of of the gene encoding the 37,000-Dalton outer envelope enveloped vaccinia virus. J Virol 2003; 77:9008–9019. protein. J Virol 1991; 65:5910–5920. 39. Payne LG, Kristenson K. Mechanism of vaccinia virus 19. Hiller G, Weber K. Golgi-derived membranes that contain release and its specific inhibition by N1-isonicotinoyl-N2-3- an acylated viral polypeptide are used for vaccinia virus methyl-4-chlorobenzoylhydrazine. J Virol 1979; envelopment. J Virol 1985; 55:651–659. 32:614–622. 20. Husain M, Moss B. Similarities in the induction of post- 40. Payne LG. Identification of the vaccinia hemagglutinin Golgi vesicles by the vaccinia virus F13L protein and polypeptide from a cell system yielding large amounts of phospholipase D. J Virol 2002; 76:7777–7789. extracellular enveloped virus. J Virol 1979; 31:147–155.

Antiviral Therapy 12:8 1215 Andrei 21/11/07 15:46 Page 1216

S Duraffour et al.

41. Payne LG. Significance of extracellular enveloped virus in 43. Gileva IP, Nepomnyashchikh TS, Antonets DV et al. the in vitro and in vivo dissemination of vaccinia. J Gen Properties of the recombinant TNF-binding proteins from Virol 1980; 50:89–100. variola, monkeypox, and cowpox viruses are different. 42. Pulford DJ, Meyer H, Ulaeto D. Orthologs of the vaccinia Biochim Biophys Acta 2006; 1764:1710–1718. A13L and A36R virion membrane protein genes display diversity in species of the genus Orthopoxvirus. Arch Virol 2002; 147:995–1015. Accepted for publication 1 August 2007