Human cytomegalovirus (CMV) causes significant disease in immunocompromised patients and serious birth defects if acquired in utero. Available CMV antivirals target the viral DNA polymerase, have signif- icant toxicities, and suffer from resistance. New drugs targeting different pathways would be beneficial. The anthraquinone emodin is proposed to inhibit herpes simplex virus by blocking the viral nuclease. Emodin and related anthraquinones are also reported to inhibit CMV. In the present study, emodin reduced CMV infectious yield with an EC50 of 4.9 lM but was cytotoxic at concentrations only twofold higher. Related anthraquinones acid blue 40 and alizarin violet R inhibited CMV at only high concentra- tions (238–265 lM) that were also cytotoxic. However, atanyl blue PRL inhibited infectious yield of CMV with an EC50 of 6.3 lM, significantly below its 50% cytotoxic concentration of 216 lM. Atanyl blue PRL reduced CMV infectivity and inhibited spread. When added up to 1 h after infection, it dramatically reduced CMV immediate early protein expression and blocked viral DNA synthesis. However, it had no antiviral activity when added 24 h after infection. Interestingly, atanyl blue PRL inhibited nuclease activ- ities of purified CMV UL98 protein with IC50 of 4.5 and 9.3 lM. These results indicate that atanyl blue PRL targets very early post-entry events in CMV replication and suggest it may act through inhibition of UL98, making it a novel CMV inhibitor. This compound may provide valuable insights into molecular events that occur at the earliest times post-infection and serve as a lead structure for antiviral development.1. Introduction
Human cytomegalovirus (CMV) causes a spectrum of diseases in immune compromised patients, including retinitis in HIV patients, pneumonitis in transplant patients, and serious birth defects characterized by sensorineural hearing loss and severe mental retardation when acquired during pregnancy. Drugs cur- rently licensed for the treatment of systemic CMV infections have limited therapeutic effectiveness due to dose-limiting toxicities, and prolonged therapy often results in resistance (Biron, 2006). None are approved for treatment or prevention of congenital infec- tions. Thus, there is a pressing need for more potent, less toxic therapeutics to combat CMV infections.
One possible target for development of such novel antivirals is the CMV UL98 alkaline nuclease encoded by the UL98 gene.Homologs of UL98 are encoded by all known herpesviruses. The alkaline nuclease encoded by herpes simplex virus type 1 (HSV-1), UL12, is necessary for efficient replication in cell culture (Martinez et al., 1996; Shao et al., 1993). Impaired replication has been linked to defects in DNA processing, capsid stability, and cap- sid nuclear egress (Martinez et al., 1996; Porter and Stow, 2004; Shao et al., 1993; Weller et al., 1990). Recently our group showed that a CMV UL98 null mutant is severely compromised for replica- tion (Kuchta et al., 2012). At least one function of the CMV UL98 is likely to be similar to HSV-1 UL12 since the CMV UL98 gene can functionally complement the replication defect of HSV-1 UL12 null viruses (Gao et al., 1998). A recent study indicated that the nucle- ase activity of UL12 is important for HSV-1 neurovirulence in mice yet is largely dispensable for replication in cell culture (Fujii et al., 2013). In contrast, earlier studies indicated that the nuclease activity is important for in vitro replication since only alleles of HSV-1 UL12 that encoded a functional nuclease were able to com- plement replication of UL12 null mutant viruses (Goldstein and Weller, 1998; Henderson et al., 1998).Consistent with UL12 nuclease activity playing important roles both in vitro and in vivo, recent studies found that the anthraqui- none emodin (Fig. 1) inhibits DNase activity of the HSV-1 UL12 in vitro, blocks replication of HSV-1 and herpes simplex virus type 2 (HSV-2) in cell culture, and reduces viral pathogeneses in a mouse model (Hsiang and Ho, 2008; Xiong et al., 2011). Earlier reports indicated that emodin and other anthraquinone derivatives also have CMV inhibitory activities (Barnard et al., 1992, 1995), although the mechanism of CMV inhibition has not been further studied. In the present study the anti-CMV activities of emodin and three related anthraquinones, atanyl blue PRL (also known as acid blue 129), acid blue 40, and alizarin violet R (Fig. 1) were eval- uated. Atanyl blue PRL had anti-CMV activity and acted at an early post-entry stage of replication. Atanyl blue PRL also inhibited the nuclease activity of UL98, suggesting a potential mechanism of action in which UL98 activity is important early in the CMV repli- cation cycle.
2. Materials and methods
2.1. Viruses and cell culture
Human MRC-5 fibroblasts (ATCC CCL-171) were propagated in modified Eagle medium (Gibco-BRL) supplemented with 10% fetal calf serum (HyClone Laboratories), 10,000IU/L penicillin, and 10mg/L streptomycin (Gibco-BRL) (MEM). CMV BADrUL131-Y4 (a gift from Thomas Shenk and Dai Wang) is a variant of CMV strain AD169 that contains a marker cassette encoding green fluorescent protein (GFP) under control of a simian virus 40 promoter inserted into UL21.5 (Wang and Shenk, 2005). CMV RC2626 is a variant of CMV strain Towne containing a luciferase expression cassette under control of a synthetic P1125 promoter (composed of seven tetracycline operator elements, a 23-bp TAATA-containing element from the adenovirus major late promoter, and a 17-bp initiator from the mouse TdT gene promoter) inserted into the US2-US6 region (McVoy and Mocarski, 1999). Expression of the relevant marker proteins (GFP or luciferase) encoded by these viruses can be detected as early as 24 h post infection (hpi). Viruses were prop- agated in MRC-5 cells and titered as described (Cui et al., 2012, 2008; Saccoccio et al., 2011b).
2.2. Compounds
Ganciclovir was purchased from InvivoGen. BAY 38-4766 was provided by Bayer? Pharmaceuticals (Tubingen, Germany).
Emodin, atanyl blue PRL, and acid blue 40 were purchased from Sigma–Aldrich Co. Alizarin violet R was purchased from MP Bio- medicals. Ganciclovir was dissolved in water at a concentration of 100mM. The remaining compounds were solubilized in dimethyl sulfoxide (DMSO, Sigma–Aldrich) at a stock concentra- tion of 100 mM. Compounds were stored at 20 °C.
2.3. Luciferase-based yield reduction assay
CMV yield was determined using a luciferase-based assay as described previously (Bhave et al., 2013). Briefly, 96-well plates containing confluent monolayers of MRC-5 fibroblasts were infected at low input multiplicity with luciferase-tagged CMV RC2626 (McVoy and Mocarski, 1999) and incubated for 1 hour. Various amounts of test compounds in MEM were then added. No-virus controls and no-drug controls consisting of medium con- taining 0.25% DMSO (equal to the maximum amount of DMSO present in test compound assays) were assayed on each plate and all conditions were assayed in triplicate. Cultures were main- tained with no additional media changes until 5 days post infection (dpi), when 50 ll of supernatant from each well was transferred to black-walled, clear/flat-bottomed 96-well plates containing con- fluent MRC-5 monolayers and luciferase activity was measured 24 h later using the Steady-Glo luciferase assay reagent (Promega). Luciferase activity was measured in relative light units (RLU) using a Biotek Synergy HT Multimode Microplate Reader. Data were nor- malized by converting RLU to ‘‘percent maximum luminescence’’ for each experiment and fitted to four-parameter curves using Prism 5 (GraphPad Software, Inc.) to determine the half-maximal effective concentrations (EC50) of test compounds.
For time of addition studies cultures were infected as described above and maintained in medium until compounds were added. At different times post infection compounds diluted in culture med- ium were added. Cultures were maintained with no additional media changes until 5 dpi. Two preincubation experiments were conducted in conjunction with time of addition studies. In the first, wells containing uninfected cells were incubated for 1 h with med- ium containing 28 lM atanyl blue PRL, then washed 3 times with medium to remove the atanyl blue PRL. The cultures were then infected as described above and maintained for 5 days in medium without atanyl blue PRL. In the second, a 10-fold higher concentra- tion of virus inoculum was incubated for 1 h with medium contain- ing 28 lM atanyl blue PRL, then diluted 10-fold with culture medium and added to wells containing cells.
2.4. Cytotoxicity assay
Black-walled, clear bottom 96-well plates containing confluent monolayers of MRC-5 cells were incubated in triplicate with test compounds as described above. After 5 days, 100 ll of each super- natant was removed and 100 ll CellTiter-Glo assay reagent (Pro- mega) was added to each well. Luminescence was measured in RLUs as described above and the data were normalized and fitted to four-parameter curves to determine the 50% toxic doses (TD50) of test compounds.
2.5. GFP-based assays of gene expression, virus spread, and infectivity
Clear-walled, clear-bottom 96-well plates containing confluent monolayers of MRC-5 cells were infected at 37°C with GFP-expressing CMV BADrUL131-Y4. At the time of virus infection, atanyl blue PRL in MEM was added to achieve various final concen- trations. No-virus controls and no-drug controls consisting of med- ium containing 0.25% DMSO (equal to the maximum amount of DMSO present in test compound assays) were assayed on each plate and all conditions were assayed in triplicate. On specified