|Home | About | Journals | Submit | Contact Us | Français|
Infection with human cytomegalovirus (HCMV) continues to be a threat for pregnant women and immunocompromised hosts. Although limited anti-HCMV therapies are available, development of new agents is desired. The Wnt signaling pathway plays a critical role in embryonic and cancer stem cell development and is targeted by gammaherpesviruses, Epstein-Barr virus (EBV), and Kaposi's sarcoma-associated herpesvirus (KSHV). HCMV infects stem cells, including neural progenitor cells, during embryogenesis. To investigate the role of Wnt in HCMV replication in vitro, we tested monensin, nigericin, and salinomycin, compounds that inhibit cancer stem cell growth by modulating the Wnt pathway. These compounds inhibited the replication of HCMV Towne and a clinical isolate. Inhibition occurred prior to DNA replication but persisted throughout the full replication cycle. There was a significant decrease in expression of IE2, UL44, and pp65 proteins. HCMV infection resulted in a significant and sustained decrease in expression of phosphorylated and total lipoprotein receptor-related protein 6 (pLRP6 and LRP6, respectively), Wnt 5a/b, and β-catenin and a modest decrease in Dvl2/3, while levels of the negative regulator axin 1 were increased. Nigericin decreased the expression of pLRP6, LRP6, axin 1, and Wnt 5a/b in noninfected and HCMV-infected cells. For all three compounds, a correlation was found between expression levels of Wnt 5a/b and axin 1 and HCMV inhibition. The decrease in Wnt 5a/b and axin 1 expression was more significant in HCMV-infected cells than noninfected cells. These data illustrate the complex effects of HCMV on the Wnt pathway and the fine balance between Wnt and HCMV, resulting in abrogation of HCMV replication. Additional studies are required to elucidate how HCMV targets Wnt for its benefit.
Infection with human cytomegalovirus (HCMV) continues to be a major threat for pregnant women and the immunocompromised population, including patients with HIV/AIDS (1–3). Because of the limited agents available for HCMV therapy, the side effects associated with anti-HCMV compounds (all viral DNA polymerase inhibitors), and the emergence of resistant viral mutants during therapy (4–6), there is a pressing need to develop anti-HCMV compounds with novel mechanisms of action. Understanding the complex and evolving interaction of HCMV with the cellular machinery may lead to the development of novel anti-HCMV inhibitors.
HCMV persistently infects CD34+ hematopoietic progenitor and neural stem cells (7–10). Infection of neural stem cells reduces their capacity to differentiate into astrocytes, an observation that may explain, at least in part, the abnormalities in brain development observed in congenitally infected children (9). The Wnt signaling pathway plays an important role in embryonic development, a time in which HCMV infects multiple cells and causes injury to major organs. Recent reports from cancer chemotherapy suggest that cure of cancers depends on targeting stem cells within the tumor environment which are usually resistant to available chemotherapeutic agents (11). Compounds that inhibit cancer stem cell growth via modulation of Wnt have recently been reported (11, 12). Because components of the Wnt signaling pathway are targeted by gammaherpesviruses, Epstein-Barr virus (EBV), and Kaposi's sarcoma-associated herpesvirus (KSHV) (13), we hypothesized that HCMV may also target the Wnt pathway and that modulation of Wnt by small molecules may affect HCMV replication. In this study, the anti-HCMV activities of monensin, nigericin, and salinomycin, compounds that were reported to inhibit Wnt signaling, were tested (12). These compounds demonstrated potent inhibition of HCMV replication, an effect associated with changes in the expression of proteins in the Wnt pathway, not previously known to be affected by HCMV.
Monensin, nigericin, salinomycin, and ganciclovir (GCV) were purchased from Sigma Chemicals (St. Louis, MO). A 10 mM stock solution of all compounds was stored at −80°C.
The pp28-luciferase HCMV Towne strain was constructed as previously described (14). This virus expresses luciferase under the control of the late CMV gene promoter pp28. Luciferase expression is strongly activated 48 to 72 h postinfection (hpi). This recombinant virus provides a highly sensitive and reproducible reporter system which correlates with the classic plaque reduction assay (14). A GCV-resistant HCMV strain was obtained from a patient with CMV disease. It has a UL97 mutation (H520Q) and a 50% effective concentration (EC50) of 7.6 μM for GCV. Human herpesvirus strains were luciferase HSV1-KOS/Dlux/oriS (15) and clinical isolates of herpes simplex virus 1 (HSV-1) and HSV-2. All clinical isolates were provided by the clinical virology laboratory with no identifiers that can link to a specific subject. The Johns Hopkins Office of Human Subjects Research Institutional Review Board determined that this research qualified for an exemption.
Human foreskin fibroblast (HFFs) passages 12 to 16 (ATCC CRL-2088) were grown in Dulbecco's modified eagle medium (DMEM) containing 10% fetal bovine serum (FBS) (Gibco, Carlsbad, CA) in a 5% CO2 incubator at 37°C and used for infection with HCMV or HSV at a multiplicity of infection of 1 PFU/cell (MOI = 1) unless otherwise specified. Following 90 min of adsorption (60 min for HSV), medium was removed and cells were washed with phosphate-buffered saline (PBS). DMEM with 4% FBS containing compounds was added to each well. For HCMV, infected, treated HFFs were collected at 72 hpi and lysates were assayed for luciferase activity using a luciferase assay kit (Promega, Madison, WI) on a GloMax-Multi+ detection system (Promega) according to the manufacturer's instructions. For HSV1-KOS/Dlux/oriS, a luciferase assay was performed 24 hpi. A yield reduction assay was performed with HSV1-KOS/Dlux/oriS. HFFs were infected and treated with the compounds. At 48 hpi, the supernatants from infected HFFs were collected and used for infection of fresh HFFs. Luciferase activity was measured after 16 h. Plaque assays were performed with clinical isolates of HSV-1 and HSV-2. Vero cells were seeded at 3 × 105 cells per well in 12-well plates and were infected 24 h later with HSV-1 or HSV-2 strains at 200 PFU/well. Following 60 min of adsorption, the virus was aspirated, and DMEM containing 0.5% carboxymethyl-cellulose, 4% fetal bovine serum (FBS), and compounds at indicated concentrations was added into triplicate wells. After incubation at 37°C for 2 days, the overlay was removed and plaques were counted after crystal violet staining.
The quantitative CMV real-time PCR assay is based on detection of a 151-bp region from the highly conserved US17 gene (16). The limit of detection of the assay is 100 copies/ml (3.0 copies/reaction), and the measureable range is 2.4 to 8.0 log10 copies/ml. The PCR was performed using a total reaction volume of 50 μl, including TaqMan 2× universal PCR master mix (Applied Biosystems, Foster City, CA), primers (300 nM final concentration), 6-carboxyfluorescein (FAM)-labeled probe (200 nM final concentration), distilled water (dH2O), and template (10 μl). Amplification was performed on a 7500 real-time PCR system (Applied Biosystems, Foster City, CA). PCR conditions were as follows: 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s and 60°C for 60 s. Quantification standards were prepared by cloning the US17 amplicon in the pCR2.1-TOPO plasmid vector (Invitrogen, Carlsbad, CA). Serial 10-fold dilutions of plasmid from 7.0 to 1.0 log10 copies/reaction were included with each assay and used to establish a standard curve. The slope and R2 of the standard curve were −3.3 ± 0.1 and >0.990, respectively. Assay controls included quantified CMV AD169 DNA (Advanced Biotechnologies Inc.) and quantified Towne CMV at 3.0 and 5.0 log10 copies/ml. Quantitative CMV data were expressed as viral DNA copies per milliliter. The real-time PCR was used for quantification of HCMV replication in cell lysates at 48 hpi and virus DNA yield in supernatants at 96 hpi and for a reversibility assay (14, 16).
Cell viability was determined using a colorimetric MTT [3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide] cell proliferation assay by following the manufacturer's instructions (Sigma-Aldrich, St. Louis, MO). HFFs were treated with various concentrations of monensin, nigericin, and salinomycin and incubated at 37°C for 3 days. After the addition of 20 μl/well of MTT (5 mg/ml in PBS) and shaking at 150 rpm for 5 min, the plates were incubated at 37°C for 3 h. Conversion of yellow solution into dark blue formazan by mitochondrial dehydrogenases of living cells was quantified by measuring absorbance at 570 nM. Viable cells in culture medium containing vehicle alone (0.1% dimethyl sulfoxide [DMSO]) were referred to as 100% cell viability. For each cell type used for virus infection and drug treatment, the MTT assay was performed at the same time points as the antiviral assay.
Cell lysates were quantified for protein content using a bicinchoninic acid (BCA) protein assay kit (Pierce Chemical, Rockford, IL). An equivalent amount of proteins was mixed with an equal volume of 2× sample buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 20% glycerol, and 5% β-mercaptoethanol) and boiled at 100°C for 10 min (except for LRP6 and pLRP6 detection, for which samples were not boiled). Denatured proteins were resolved in Tris-glycine polyacrylamide gels (10 to 12%) and transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad Laboratories, Hercules, CA) by electroblotting. Membranes were incubated in blocking solution (5% nonfat dry milk and 0.1% Tween 20 in PBS [PBST]) for 1 h, washed three times with PBST, and incubated with appropriately diluted primary antibodies at 4°C overnight. Membranes were washed with PBST and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies in PBST for 1 h at room temperature. Following washing with PBST, protein bands were visualized by chemiluminescence using SuperSignal West Dura and Pico reagents (Pierce Chemical, Rockford, IL). The following antibodies were used for detection of CMV proteins: mouse anti-IE1 and -IE2, (MAb810), mouse anti-beta-actin antibody (Millipore, Billerica, MA), mouse anti-UL44 (Santa Cruz Biotechnology, Santa Cruz, CA), mouse anti-pp65 (Vector Laboratories, Burlingame, CA), HRP-conjugated anti-rabbit IgG (Cell Signaling Technology, Beverly, MA), and HRP-conjugated anti-mouse IgG (GE Healthcare, Waukesha, WI). The following antibodies were used for detection of proteins in the Wnt pathway: anti-Wnt5a/b, -LRP6, -phospho-LRP6, -Dvl2, -Dvl3, and -axin1 (Cell Signaling) and anti-β-catenin (E-5) (Santa Cruz Biotechnology).
The anti-HCMV activities of monensin, nigericin, and salinomycin were tested. At 1 μM, >95% inhibition of HCMV replication (based on relative luciferase units) was achieved with all three compounds. A dose-response curve was generated using concentrations ranging from 1 nM to 1 μM (Fig. 1A). Although monensin and nigericin had a lower EC50 (effective concentration resulting in 50% virus inhibition) than salinomycin, the selectivity indexes (SI) of the three compounds were similar and higher than 100 (Table 1). The slopes of the dose-response curve were also similar among the compounds and close to 1 (similar to the slope of GCV), suggesting an inhibitory effect via a possible shared target. The three compounds were toxic to colon carcinoma HCT116 and cervical cancer HeLa cells, but toxicity in HFFs was significantly lower (Table 1). The concentration resulting in 50% cell toxicity (CC50) in noninfected HFFs was 30 to 80 times higher than that measured in the tested cancer cells (Fig. 1B and Table 1).
The three compounds were similarly effective in inhibiting a GCV-resistant HCMV strain (Fig. 2), based on virus DNA yield measured in supernatants of infected cells, suggesting an activity that is independent of the UL97 kinase.
To determine the timing of HCMV inhibition, add-on and removal assays were performed. Compounds were added or removed at 0, 6, 12, 24, and 48 hpi, and luciferase activity was measured at 72 hpi. The three compounds were similar to each other in both the add-on and removal assays. When compounds were added at or after 24 h, they were less effective against HCMV replication, suggesting that HCMV inhibition occurred during the immediate-early (IE) and early (E) stages of HCMV replication. However, the removal assay revealed a gradual decrease of virus replication by the three compounds (Fig. 3). Approximately 75% virus inhibition was achieved when compounds were removed at the 48-h time point. This pattern of anti-HCMV activity is notably different from that of GCV and other HCMV inhibitors, such as artemisinins (17).
To test whether HCMV inhibition by monensin, nigericin, and salinomycin was reversible, infected HFFs were treated for 1, 2, or 3 days, followed by removal of the medium containing the compounds and addition of fresh medium until 6 days postinfection. HCMV DNA was quantified in supernatants of infected cells at day 6 postinfection by real-time PCR. Similar to GCV, salinomycin was fully reversible when removed after 3 days (Fig. 4A). Monensin and nigericin showed weak irreversible inhibition of HCMV replication when present in infected cells longer than 48 h.
We recently reported that HCMV inhibitors may have varied effects on DNA replication and virus yield (based on real-time PCR in supernatants of infected cells) (17). While the DNA polymerase inhibitor GCV inhibited virus yield and DNA replication at similar potencies, the inhibition of virus yield by artemisinins was approximately 10-fold higher than the inhibition of DNA replication. The effects of monensin, nigericin, and salinomycin on DNA replication, late gene expression, and virus DNA yield were evaluated. Similar to artemisinins, the three compounds inhibited virus DNA yield at least 10-fold more than the inhibition of DNA replication (Fig. 4B), suggesting that HCMV inhibition was not simply a result of direct targeting of the DNA replication machinery.
The effect of nigericin on HCMV gene expression was tested (Fig. 5A). There was no obvious inhibition of IE1 expression, but major inhibition of IE2 and UL44 was observed at different time points (24, 48, and 72 hpi). There was also a significant inhibition in the expression of the late HCMV gene pp65. Determination of infectious virus released into the medium from infected HFFs treated with monensin or nigericin at 0.1 μM or salinomycin at 1 μM (yield reduction) revealed a complete absence of plaques, suggesting that the detection of DNA by real-time PCR from supernatants of infected and treated cells represents noninfectious virus. These results further support our observations that HCMV inhibition by monensin, nigericin, and salinomycin is likely a multistage process that is potentiated from early to late stage of virus replication.
The effect of nigericin and GCV on the expression of Wnt proteins (LRP6, phospho-LRP6, Wnt5a/b, Dvl2, Dvl3, and axin 1) was tested in noninfected and HCMV-infected HFFs (Fig. 5B). At 24, 48, and 72 hpi, HCMV infection resulted in a significant decrease in the expression of Wnt 5a/b and β-catenin, with no significant change in Dvl2 and Dvl3 expression between infected and noninfected cells. Phosphorylated and total lipoprotein receptor-related protein 6 (pLRP6 and LRP6) levels were significantly reduced in HCMV-infected cells at 24 and 48 hpi. HCMV infection resulted in enhanced and sustained expression of the negative Wnt regulator axin 1. At all time points, treatment with nigericin resulted in decreased expression of pLRP6, LRP6, axin 1, and Wnt 5a/b in noninfected and HCMV-infected cells. Treatment with nigericin also resulted in a significant decrease in β-catenin levels, an effect that was increased as infection proceeded from 24 to 48 and 72 hpi. A correlation between HCMV inhibition and expression of Wnt 5a/b and axin 1 was observed with all three compounds (see Fig. S1 in the supplemental material).
The activity of monensin, nigericin, and salinomycin against HSV-1 and HSV-2 replication was tested. The compounds did not inhibit luciferase expression measured at 24 hpi with HSV1-KOS (Dlux/OriS), but plaque formation was strongly inhibited at 48 hpi. Since the luciferase in HSV1-KOS is regulated by an immediate-early kinetics, these compounds did not inhibit HSV-1 replication at an IE/E stage while the inhibition of HCMV started at an IE stage. The supernatants of HFFs infected with HSV1-KOS (Dlux/OriS) and treated with the three compounds were harvested at 48 hpi and used for infection of fresh HFFs. Luciferase activity was measured after 16 h in the newly infected HFFs and revealed that the compounds were active against HSV-1 replication (Table 2), suggesting that inhibition occurred at a later stage of HSV-1 replication. In addition, monensin, nigericin, and salinomycin inhibited plaque formation of HSV-1 and HSV-2 in Vero cells (Table 2). The effect of monensin, nigericin, and salinomycin on expression of components of Wnt in HSV-1-infected HFFs was different from their effect in HCMV-infected and treated HFFs (see Fig. S2 in the supplemental material). At 48 hpi with HSV1-KOS (Dlux/OriS) and treatment with concentrations resulting in full HCMV inhibition, monensin, nigericin, and salinomycin did not reduce the expression of β-catenin or axin 1 while inhibition of Wnt 5a/b was observed (see Fig. S2a in the supplemental material). A similar effect on Wnt protein expression was also observed at concentrations resulting in full inhibition of HSV-1 (see Fig. S2b in the supplemental material). Taken together, the mechanisms of HCMV and HSV inhibition by monensin, nigericin, and salinomycin are likely distinct and may depend on pathways that are not fully shared between these viruses.
The data presented show for the first time that HCMV infection of HFFs is associated with complex changes in the expression of multiple proteins of the Wnt pathway and that modulators of this pathway, monensin, nigericin, and salinomycin, are potent inhibitors of HCMV replication without causing toxicity to HFFs. These three compounds may share a similar mechanism of HCMV inhibition, reflected by their slope, SI, and antiviral assays. The add-on and removal assays suggest that inhibition of HCMV replication occurs at several stages. Since the slopes of the dose-response curves of the three compounds were close to 1 (Table 1), their effects may involve a viral protein that associates with a cellular target which is critical throughout the full replication cycle. Studies are ongoing to determine whether a viral target is indeed involved in the activity of these compounds. While the tested compounds did not affect IE1 expression at 24, 48, or 72 hpi, the early late gene UL44 and IE2 were significantly decreased. The inhibitory effect was noted to augment itself as replication proceeded: DNA replication was inhibited at 48 hpi, pp65 gene expression was undetectable at 72 hpi, and a yield reduction assay revealed a complete absence of plaques in infected HFFs treated with the three compounds. In contrast to the activation of Wnt/β-catenin by the oncogenic EBV and KSHV, HCMV infection resulted in inhibition of several components of the Wnt pathway, while axin 1 expression was enhanced at all tested time points. Moreover, HCMV inhibition with Wnt modulators was associated with significant and further downregulation of several components of both the canonical and noncanonical Wnt pathways. The expression of Wnt5a, which activates the β-catenin-independent pathway but negatively regulates β-catenin-dependent activity (18), was significantly decreased throughout HCMV infection of HFFs and even more so after treatment with nigericin.
Several anticancer agents (imatinib [Gleevec], roscovitine, rapamycin) were reported to inhibit HCMV replication, likely by interfering with one or more cell signaling pathways (19–21). The drawback of these agents is their toxicity to cancer cells as well as to noncancerous primary HFFs. We have shown that the HCMV inhibitors, artemisinin dimers, have strong anticancer activities but are nontoxic to HFFs at concentrations that inhibit HCMV replication (22). A similar differential effect was reported with the Wnt modulators: while strongly inhibiting cancer cells, no toxic effects were found in the surrounding healthy cells (12, 23, 24). In agreement with these reports, we show here that cancer cells are exceedingly more sensitive to monensin, nigericin, and salinomycin than are primary HFFs.
Wnt signaling plays a crucial role in embryonic development and cancer, processes that are affected by HCMV. Originally, Wnt signals were classified into canonical (β-catenin dependent) and noncanonical (β-catenin independent). However, this pathway is now known to be more complex, Wnt action is context dependent, and multiple intracellular cascades can be triggered, some a blend of canonical and noncanonical components (25, 26). The key characteristics of canonical signaling are the requirement for the LRP5/6 coreceptor to enable β-catenin accumulation and the involvement of LEF/TCF transcription factors. LRP6 phosphorylation by glycogen synthase kinase 3 (GSK-3) and casein kinase 1γ (CK1) is crucial for activation of the canonical Wnt/β-catenin signaling (27). When a Wnt ligand binds to the Fz (Frizzled) receptor and its coreceptor LRP6, this complex, together with the scaffolding protein Dishevelled (Dvl), results in LRP6 phosphorylation, activation, and recruitment of the axin complex to the receptors. These events lead to inhibition of axin-mediated β-catenin phosphorylation and stabilization of β-catenin, which accumulates and travels to the nucleus to activate Wnt target gene expression. The noncanonical Wnts avoid LRP coreceptors and β-catenin stabilization to activate intracellular kinases and regulate distinct β-catenin-independent pathways. These include the planar cell polarity (PCP) pathway and the Wnt/calcium pathway. The PCP pathway, mediated by Fz and Dvl, activates c-Jun N-terminal kinase (JNK) and Rho-associated kinase. Noncanonical Wnts binding to Fz can also stimulate an increase in intracellular Ca2+ levels, thereby activating calcium-sensitive proteins, such as Ca2+/calmodulin-dependent protein kinase II (CaMKII) and protein kinase C (PKC) (28).
Monensin, an ionophorous antibiotic isolated from Streptomyces cinnamonensis and used as an antibiotic in dairy cattle, was recently reported as a novel antineoplastic compound in prostate cancer cells (24, 29). Salinomycin, structurally similar to monensin, was identified as a breast cancer stem cell inhibitor in vitro and in vivo (30). It also reduced cancer and cancer stem cell growth in leukemias and uterine sarcoma cells (23). Nigericin and salinomycin inhibited Wnt signaling by blocking the phosphorylation of the Wnt coreceptor LRP6 and inducing its degradation (12, 31–34). Monensin was reported in the past to block protein transport from the Golgi apparatus to the cell membrane and to inhibit HSV-1, HSV-2, and HCMV (35–38). Monensin treatment inhibited transport of progeny virus to the surface of infected cells, while viral protein synthesis and DNA replication were not inhibited. The reduction of extracellular virus release was more significant for HSV-2 than HSV-1 (35). In the case of HCMV, monensin inhibited DNA replication and generation of virus progeny in HFFs and its activity was MOI dependent (37). These results correlate with our findings and suggest differences in activities of Wnt modulators between HSV- and HCMV-infected cells. Although the reports on monensin predated the discovery of Wnt and stem cell cancer, a correlation between the Wnt pathway and transport of proteins from the Golgi apparatus in HCMV-infected cells remains to be studied.
HCMV infection of HFFs resulted in decreased expression of multiple members (both canonical and noncanonical) of the Wnt pathway. Intriguingly, even further inhibition of Wnt signaling by monensin, nigericin, or salinomycin resulted in HCMV inhibition, suggesting a very fine balance between virus and Wnt components that either maintains or abrogates lytic replication. A recent study reported on dysregulation of the canonical Wnt/β-catenin signaling pathway by HCMV. Infection induced degradation of β-catenin, which resulted in a decrease in its transcriptional activity in response to Wnt ligand stimulation (39). HCMV also inhibited Wnt/β-catenin signaling in human extravillous cytotrophoblasts. Similarly, we found that HCMV infection resulted in decreased expression of total β-catenin. Another study showed that overexpression of HCMV-encoded US28 in intestinal epithelial cells inhibited GSK-3β function, promoted accumulation of β-catenin, and increased expression of Wnt target genes involved in the control of the cell proliferation (40). These studies and ours suggest that the Wnt pathway is tightly regulated by HCMV. Virus replication mostly targets this pathway for inhibition of its components (with the exception of axin 1), and further inhibition of specific components results in virus inhibition.
HCMV infection of the developing brain results in long-term neurological sequelae. How brain damage is induced by HCMV is not well understood, but neural stem cells in the fetal brain appear to be an important cell type affected by the virus (41). The Wnt pathway may play an important role in neural stem cell development, differentiation, and migration. In a murine CMV (MCMV) model, MCMV inhibited neuronal differentiation and decreased expression of Wnt-1 and neurogenin 1 (42).
We conclude that HCMV targets components of the Wnt pathway. While the gammaherpesviruses KSHV and EBV activate the canonical Wnt/β-catenin pathway, HCMV exerts different and mostly inhibitory effects on this pathway. While the described compounds may not be useful for HCMV therapy, studies with additional small molecules that modulate the Wnt pathway may assist in dissecting the complex interaction between HCMV and Wnt.
This study was supported by National Institutes of Health grant 1R01AI093701 (to R.A.-B.).
We thank David A. Leib, Dartmouth Medical School, for providing KOS/Dlux/oriS and Prashant Desai for amplifying HSV-2.
Published ahead of print 9 April 2013
Supplemental material for this article may be found at http://dx.doi.org/10.1128/AAC.00029-13.