Off Label Antiviral Therapeutics for Henipaviruses: New Light Through Old Windows

Reagents

2,5-di-(t-butyl)-1,4-hydroquinone (BHQ); (-)-(r,r)-N,N′-bis[11-(ethoxycarbonyl)undecyl]-n,n′-4,5-tetramethyl-3,6-dioxaoctanediamide (CA1001); ryanodine and xestospongin C were purchased from Merck Scientific (Kilsyth, VIC Australia). (-)-7-octylindolactam V, (+/-)-verapamil hydrochloride, (+/-)-propranolol hydrochloride, [arg⁸]-vasopressin acetate salt, 1,2-bis(2-aminophenoxy)ethane-N,N,N′N′—tetraacetic acid (acetoxymethyl ester) (BAPTA-AM), 25-hydroxy-cholecalciferol, 4-hydroxytamoxifen, bradykinin acetate, caffeine, calcium ionophore A23187, chemiluminescent peroxidase substrate-3 (CPS-3), cholecalciferol, clotrimazole, ethylene glycol-bis(2-aminoethylether)-N, N, N′N′-tetraacetic acid (EGTA), ionomycin from streptomyces conglobatus, nifedipine, phorbol 12-myristate 13-acetate (PMA), protein kinase C (PKC) inhibitor (meristoylated), ribavirin, thapsigargin and U73122 hydrate were purchased from Sigma Chemical Company (St. Louis, MO, USA). The CellTiter-Glo® cytotoxicity kit was purchased from Promega (Madison, USA). Indo-1 was purchased from Invitrogen (Carlsbad, CA, USA). RNeasy kits were purchased from Qiagen (Valencia, CA, USA). First strand synthesis Superscript III Reverse Transcriptase kits were purchased from Invitrogen (Carlsbad, CA, USA). Jumpstart SYBRGreen real time PCR reagent was purchased from Sigma-Aldrich (St. Louis, MO, USA).

Virus and cells

African Green Monkey Kidney (Vero) cells were grown in Minimal Essential Medium containing Earle's salts (EMEM), antibiotics (100 U penicillin, 100 μg/ml streptomycin and 500 μg/ml Fungizone) and 10% fetal calf serum (FCS), designated EMEM-10. NiV was isolated in Vero cells from the brain of a human fatally infected in the 1998-99 Malaysian outbreak and was passaged three times in Vero cells then double plaque purified and passaged a further three times in Vero cells as previously described (Shiell et al., 2003). HeV was isolated in Vero cells from the lung of a horse infected in the Brisbane outbreak in October 1994 and was passaged five times in Vero cells followed by triple plaque purification and a further five passages in Vero cells as previously described (Hyatt et al., 1996). HeV and NiV stock titer were adjusted to 1×10⁶ TCID₅₀/ml. All work with live virus was carried out under Biosafety Level 4 (BSL 4) conditions at CSIRO Australian Animal Health Laboratory (AAHL).

Henipavirus infection of Vero cells and antiviral screening

Vero cells were seeded at a density of (2×10⁴) into individual wells of 96-well microtitre plates and incubated at 37°C overnight in 100 μl EMEM-10. Prior to NiV inoculation, medium was discarded and 100 μl of 20 μM of different test compounds were added to each well in triplicate. Under BSL4 conditions, 1,000 TCID₅₀ of virus in EMEM-10 were added to each well of Vero cells in volumes of 100 μl (final compound concentration of 10 μM). After an overnight incubation at 37°C, the culture medium was discarded; plates were immersed in ice-cold absolute methanol, enclosed in heat sealed plastic bags and the bags surface sterilized with Lysol for removal from the BSL4 laboratory. Methanol-fixed plates were air dried at room temperature for a minimum of 30 minutes prior to immunolabeling. In pretreatment experiments cells were treated for 60 minutes with one of the following compounds: 10 μM BAPTA, 10 μM EGTA, 5 mM calcium chloride, 10 μM xestospongin C, 1 μM ryanidine or 10 μM verapamil. Medium was then removed and cells were washed with fresh media. Test compound was then applied at half-log concentrations, immediately followed by the addition of virus (1,000 TCID₅₀ in EMEM-10). Viral infection was quantified as described above.

HTS immunolabeling assay

Virus replication was assessed as previously described (Aljofan et al., 2008). Briefly, plates were washed 3 times with phosphate buffered saline containing 0.05% Tween-20 (PBS-T). Plates then were protein blocked with 100 μl of 2% skim milk in PBS-T and incubated at 37°C for 30 minutes. After blocking, plates were washed 3 times with PBS-T, followed by incubation with 100 μl anti-NiV-N antibody (prepared by AAHL Bioreagents Development Group) diluted 1:1,000 in PBS-T containing 2% skim milk for 30 minutes at 37°C and then washed 3 times with PBS-T. Plates were incubated with 1% H₂O₂ (Sigma) for 15 minutes at room temperature then washed with PBS-T 3 times. Aliquots of 100 μl of goat anti-rabbit HRP conjugated antibody (Sigma) diluted 1:2,000 in PBS-T containing 2% skim milk, were added to each well and plates incubated at 37°C for 30 minutes. Plates were washed 3 times with PBS-T and 100 μl aliquots of Chemiluminescent Peroxidase Substrate-3 (CPS-3, Sigma) diluted 1:10 in chemiluminescent assay buffer (20 mM Tris-HCl, 1 mM MgCl₂, pH 9.6) were added to all wells. Plates were incubated at room temperature for approximately 15 minutes, and then read using a Luminoskan Ascent luminometer (Thermo Fisher Scientific, Waltham, USA) using 100 millisecond integration per well.

Antiviral cytotoxicity assay

Compounds were evaluated for cytotoxicity using the CellTiter-Glo cytotoxicity kit (Promega) according to the manufacturer's instructions. The CellTiter-Glo Luminescent Cell Viability Assay is a homogeneous method of determining the number of viable cells in culture based on quantitation of the ATP present, which signals the presence of metabolically active cells. Briefly, Vero cell monolayers were incubated with 200 μl of half-log dilutions in EMEM-10 (20 μM-63 nM) of each compound (n=3) overnight at 37°C. Medium was removed and 100 μl of CellTiter-Glo Reagent, diluted 1:5 with chemiluminescent assay buffer, was added to each well, mixed well to lyse cells, equilibrated to room temperature for 10 minutes, and then read using a luminometer as described above. Non-linear regression analysis performed using GraphPad Prism software to determine the concentration resulting in 50% cytotoxicity (CC₅₀).

Determination of cytokines gene expression

Vero cells of approximately 1 × 10⁵ densities, were seeded in 48 well plates, and treated with compounds at various concentrations (10 μM, 5 μM, and 1.25 μM) for 24 hours. RNA was extracted using Qiagen RNeasy kits. RNA from each sample (1 μg) was treated with DNase I (Invitrogen) for 15 minutes, reverse transcribed with 0.5 μg oligo dT(12-18) primers (Invitrogen) using Superscript II RT (Invitrogen), and diluted 1:5 before 2 μl of each diluted sample was subject to real-time PCR in triplicate using Jumpstart SYBRGreen kits (Sigma) on an ABI 7900HT Realtime PCR system (Applied Biosystems). Real-time PCR primers specific for TNF-α and IL-8 were obtained from SuperArrayBiosciences (Applied Biosystems) and were used at a concentration of 200 nM for each reaction. Fold change in mRNA transcripts over mock levels were determined using the delta-delta-Ct method.

Effect of intracellular calcium releasing compounds on henipavirus replication

To examine the effect of intracellular calcium release on viral infection the application of thapsigargin, a potent releaser of inositol triphospate (IP3) dependent ER calcium stores (Treiman et al., 1998; Thomas et al., 1999) induced highly effective inhibition of henipavirus replication in vitro (Figure 1) as did the calcium ionophore, A23187, which makes permeable to calcium both the plasma membrane (allowing influx of extracellular calcium) and to the ER membrane facilitating calcium efflux into the cytosol. The addition of other calcium ionophores also resulted in the inhibition of henipavirus replication, but their efficacy varied by several orders of magnitude (A23187 ≫ ionomycin > CA1001), perhaps reflecting differences in their specificity for calcium transport (Table 1). Similarly, BHQ, another releaser of calcium from intracellular stores via IP3 stores also inhibited virus replication, but was over 100-fold less potent than thapsigargin. We also observed that the IP3 mediated calcium release antagonists, 2-APB and xestospongin C, inhibited virus replication at concentrations similar to BHQ (xestospongin C) or greater (2-APB) as did an inhibitor of ryanodine-mediated calcium release from the ER (dantrolene) in addition to ryanodine itself (Table 1).

Modulation of antiviral efficacy by calcium chelation

The calcium ionophore, A23187, mediated viral replication inhibition of HeV, which was partially reduced by either the removal of extracellular calcium through the addition of EGTA or the addition of exogenous calcium (Figure 2) suggesting at least some of their effects may be due to the flux of calcium either into or out of the cell. EGTA had no effect on thapsigargin efficacy supporting the hypothesis that its antiviral effect is mediated through release of intracellular calcium. In contrast, A23187 mediated inhibition of NiV replication was not significantly modulated by calcium chelation. For both A23187 and thapsigargin, chelation of intracellular calcium using BAPTA-AM resulted in approximately 30-50% reduction of antiviral efficacy of thapsigargin and A23187, respectively against HeV, highlighting the importance of available cytosolic calcium for antiviral efficacy with these agents. Individual addition of neither BAPTA, EGTA, nor exogenously added calcium exhibited considerable inhibition of virus replication (data not shown).

Blockade of intracellular calcium release from either IP3 dependant stores (xestospongin C) or ryanodine sensitive stores (ryanodine) partially reversed the inhibition due to thapsigargin and A23187 (Figure 2). Additionally, pretreatment with a calcium channel antagonist (verapamil) significantly reduced virus replication, further supporting the possibility that these antiviral effects are partially mediated by the flux of calcium across the plasma membrane. While thapsigargin exerts its effect by inhibiting sarco/endoplasmic reticulum Ca²⁺-ATPases (SERCA) (Paula et al., 2004), thus depleting the ER calcium stores and increasing the cytosolic calcium levels, virtually all of this, mobilised cytosolic calcium, is immediately extruded from the cell via plasma membrane calcium ATPase (PMCA) (Hong et al., 1999).

This potentially provides a common pathway with A23187 involving calcium efflux. Given the relative modulation by EGTA of the antiviral effects exerted by these compounds, it is possible that A23187 relies on calcium influx as well, while thapsigargin exerts its effects via calcium efflux.

Effect of calcium and sodium channel antagonists

To further evaluate the direct role of ion channel blockade on viral infectivity, ten calcium channel antagonists, four sodium channel antagonists and one sodium ionophore were evaluated for direct effect on viral infectivity. Of the calcium channel antagonists, only mibefradil, verapamil and praziquantel afforded considerable reduction in viral infectivity in the low micromolar range (Table 2). Interestingly, despite the direct inhibitory effects of verapamil, it significantly reversed the inhibition associated with all other calcium antagonists (data not shown). The most significant effect of sodium channel antagonists was evident with 5-N-ethyl isopropyl amiloride (EIPA) followed by benzamil. Both compounds are inhibitors of the Na/H antiport (Abrahamse et al., 1994) suggesting a possible role for pH modulation of virus replication. Phenytoin and amiloride were less effective at inhibiting either virus while the sodium ionophore, monensin, was effective in the low micromolar range (Table 2). Monensin has previously been shown to inhibit HeV F protein processing (Pager et al., 2004) and this was attributed to the inhibition of vesicular transport between the medial and trans-Golgi cisternae (Griffiths et al., 1983).

Further investigation of additional calcium modulators revealed that the calmodulin antagonist, phenoxybenzamine, also resulted in considerable inhibition of viral replication suggesting the level of free calcium intracellularly is an important factor in viral infection (Table 3). The fungal metabolite gliotoxin, reported to suppress rises in intracellular calcium concentration via store operated Ca²⁺ entry (SOCE) in mast cells (Niide et al., 2006), also produced potent inhibition of viral infectivity, as we described earlier (Aljofan et al., 2009b), as did cyclosporine, which, is also an activator of SERCA, which is reported to result in removal of cytosolic calcium (Hadri et al., 2006).

Effect of PKC activators and inhibitors on henipavirus replication

As calcium regulation is inherently linked to protein kinase C (PKC) pathways, we also examined a range of specific PKC activators and inhibitors for the potential henipavirus antiviral efficacy. Both rottlerin (PKC inhibitor) and U73122 (phospholipase C (PLC) inhibitor) produced moderate inhibition of viral replication but the effects of other PKC activators or inhibitors was quite variable. Interestingly PKC activators (PMA and 7-octyl-indolactam) inhibited virus replication at high concentrations (>10 μM), presumably through acute toxicity (Figure 3), but actually enhanced replication in the nanomolar range. Note the absence of effect of the PKC inhibitor in the nanomolar range but efficacy (without toxicity) at high micromolar concentration.

Effect of therapeutically licensed drugs on henipaviruses

The potent antiviral actions observed with both A23187 and thapsigargin, and to a lesser extent by gliotoxin and phenoxybenzamine, possibly attributable to the modulation of free calcium in the cytosol has prompted us to evaluate clinically available compounds, which are reported to act as modulators of cytosolic calcium, as potential antivirals (Table 4). Of the eight compounds randomly selected based on reports in the literature of modulation of intracellular calcium, seven compounds inhibited henipavirus replication in the low micromolar range while caffeine also inhibited virus infectivity, but only in the millimolar range (Table 5). Of these, only five showed a strong dependance on intracellular calcium for their inhibition effects, determined through a series of pretreatment experiments with various calcium chelators, channel antagonists or ER release inhibitors (Table 5). Importantly, antiviral efficacy was enhanced by pretreatment of cells with thapsigargin with four of these compounds. In general, supplementation of the media with calcium chloride 10 μM prior to and during drug treatment produced similar levels of efficacy inhibition as did the removal of calcium via EDTA. This suggests calcium flux modulation by these compounds may involve the release of calcium from intracellular stores, in addition to the influx of calcium from the extracellular space. As either process may trigger the other, the order of this sequence may vary for each compound. Interestingly, the least effective of these compounds, caffeine, is reported to release ryanodine sensitive calcium stores in skeletal muscle cells (Chen et al., 1996). As supported by the lack of modulation by thapsigargin in the current study (Table 5). The five compounds with greater efficacy in this study have been reported to release IP3 sensitive calcium (Reilly et al., 1998; Sitrin et al., 1999; Wang et al., 2001; Sergeev et al., 2005) suggesting there may be an IP3 dependent portion to the antiviral effects attributable to many of these compounds. While clotrimazole exhibited a weak dependance on intracellular calcium, the effects of 4-hydroxy tamoxifen and propranolol were insensitive to changes in calcium availability, suggesting their antiviral efficacy is likely to be via an unrelated mechanism.

Determination of cytokine gene expression following antiviral treatment

While these studies provide support for the hypothesis that the antiviral efficacy of a number of these compounds is associated with the release of intracellular calcium, this release alone may have profound effects on a broad range of cellular processes including general cellular inflammatory processes. In an effort to evaluate the possibility of changes in cytokine production as an indicator of general inflammation, we evaluated the expression of two inflammatory cytokines, IL-8 and TNF-α, in response to a number of drug treatments including A23187, thapsigargin, 4-hydroxytamoxifen (as a representative of the ER calcium releasing drugs), calcium channel antagonists and PKC modulators (Figure 4). As expected our control compound, PMA stimulated an almost 60-fold increase in production of both these cytokines, while several compounds induced greater than 20-fold increase in IL-8 expression, without accompanying increase in TNF-α production. Interestingly, verapamil and mibefradil produced a much greater stimulation of IL-8 expression than did another calcium antagonist, nifedipine, partially reflective of their relative antiviral potency. Neither 4-hydroxytamoxifen nor thapsigargin induced obvious changes in cytokine expression suggesting their antiviral effects may be distinct from general cellular anti-viral responses via inflammation.