|Home | About | Journals | Submit | Contact Us | Français|
Recently we characterized two inhibitors targeting the human cytomegalovirus (HCMV) terminase, 2-bromo-4,5,6-trichloro-1-(2,3,5-tri-O-acetyl-β-d-ribofuranosyl) benzimidazole (BTCRB) and 2,4,5,6-tetrachloro-1-(2,3,5-tri-O-acetyl-β-d-ribofuranosyl) benzimidazole (Cl4RB). The terminase consists of the ATP-hydrolyzing subunit pUL56 and the subunit pUL89 required for duplex nicking. Because mammalian cell DNA replication does not involve cleavage of concatemeric DNA by a terminase, these compounds represent attractive alternative HCMV antivirals. We now have tested these previously identified benzimidazole ribonucleosides in order to determine if they are active against HCMV clinical isolates as well as those of herpes simplex virus type 1, mouse cytomegalovirus, rat cytomegalovirus (RCMV), and varicella-zoster virus (VZV). Antiviral activity was quantified by measurement of viral plaque formation (plaque reduction) and by viral growth kinetics. Interestingly, both BTCRB and Cl4RB had an inhibitory effect in ganciclovir (GCV)-sensitive and GCV-resistant clinical isolates, with the best effect produced by Cl4RB. Electron microscopy revealed that in cells infected with GCV-sensitive or GCV-resistant isolates, B capsids and dense bodies were formed mainly. Furthermore, pulsed-field gel electrophoresis showed that cleavage of concatenated DNA was inhibited in clinical isolates. In addition, the antiviral effect on other herpesviruses was determined. Interestingly, in plaque reduction assays, BTCRB was active against all tested herpesviruses. The best effects were observed on VZV- and RCMV-infected cells. These results demonstrate that the new compounds are highly active against GCV-resistant and GCV-sensitive clinical isolates and slightly active against other herpesviruses.
Human cytomegalovirus (HCMV) is one of eight human herpesviruses and is a serious, life-threatening, opportunistic pathogen in immunocompromised patients (organ recipients or AIDS patients) (8, 19). HCMV is widespread, with a seroprevalence throughout the world of up to 100% in adults. To date, nearly all anti-HCMV drugs for systemic treatment are inhibitors of the viral DNA polymerase (2, 13, 21, 22). Due to the low bioavailability, number of side effects, dose-dependent toxicity, and appearance of resistances caused by the current available drugs, development of new antiviral compounds which have a different mode of action is needed. Consequently, to broaden therapy of HCMV infections and to circumvent current mechanisms of drug resistance, an inhibitor of HCMV terminase would be of great value, because it would act subsequently to DNA synthesis and block the first steps in viral maturation.
Viral replication includes cleavage of newly synthesized, concatemeric DNA into unit-length genomes and packaging into preformed procaspids. These processes occur in or in close proximity to replication centers in the nucleus. Enzymes involved in the packaging process are responsible for duplex nicking and insertion of the DNA into the procapsids (1, 4, 9, 10), the so-called terminases. The HCMV terminase consists of two subunits, one subunit encoding pUL56 and the other pUL89 (5-7); each protein has a different function. Whereas subunit pUL56 mediates sequence-specific DNA binding and ATP hydrolysis, pUL89 is required for duplex nicking and enhancement of the UL56-associated ATPase activity (15, 23, 24). The hydrolysis of ATP has multiple functions during the packaging process. It is also involved in the formation of the packaging complex. The anti-HCMV benzimidazole d-nucleosides, 2-bromo-5,6-dichloro-1-(β-d-ribofuranosyl) benzimidazole (BDCRB) and 2,5,6-trichloro-1-(β-d-ribofuranosyl) benzimidazole (TCRB), developed in the laboratories of Townsend, Drach, and coworkers (12, 26), target the HCMV terminase (18, 27). However, even though these are excellent inhibitors of HCMV infection in cell culture, BDCRB is not metabolically stable in vivo. To overcome this problem, various analogs have been synthesized by Townsend and coworkers. Recently, by using a new bioluminometric assay, we found the two most active new compounds were 2-bromo-4,5,6-trichloro-1-(2,3,5-tri-O-acetyl-β-d-ribofuranosyl) benzimidazole (BTCRB) and 2,4,5,6-tetrachloro-1-(2,3,5-tri-O-acetyl-β-d-ribofuranosyl) benzimidazole (Cl4RB) (16). Since DNA packaging is ATP-dependent and the portal protein has no enzymatic activity, the interaction of the terminase subunit pUL56 with the portal protein pUL104 is essential. Therefore, analyses were undertaken to investigate the effects of the compounds on this interaction. By coimmunoprecipitation, we identified a direct interaction between pUL56 and pUL104 that was specifically inhibited by Cl4RB but not by the virologically inactive control CDMRB or by the other tetrahalogenated benzimidazole BTCRB (11). Furthermore, electron microscopy (EM) demonstrated that the formation of infectious particles was inhibited (16). Since the compounds represent promising antivirals, we performed analysis with ganciclovir (GCV)-sensitive and GCV-resistant clinical isolates and different herpesviruses.
Human foreskin fibroblasts (HFF) were grown in Dulbecco's minimal essential medium (DMEM) supplemented with 10% fetal calf serum (FCS), 2 mM glutamine, penicillin (5 U/ml), and streptomycin (50 μg/ml). HFF cells at passages 10 to 15 were used for infections, and experiments were carried out with confluent cell monolayers (1.5 × 107 cells). NIH 3T3 cells were used for experiments with mouse cytomegalovirus (MCMV), while Rat2 cells (American Type Culture Collection CRL-1764) were used for analysis with rat cytomegalovirus (RCMV).
Preparation of HCMV AD169 was performed after infection of HFF cells at a multiplicity of infection (MOI) of 0.1. The MOI was determined by the end point dilution method, with staining against HCMV IE1. A GCV-sensitive isolate was obtained from K. Korn (Institute of Clinical and Molecular Virology, Erlangen, Germany) at passage 0, which refers to direct passage from a urine sample of a patient. The GCV-resistant isolate carried the mutation M460V in HCMV UL97 and was received from K. Hamprecht (Institute of Medical Virology, Tübingen, Germany) at passage 4. In our analysis GCV-sensitive and GCV-resistant clinical isolates were used at passages 4 and 7, respectively. MCMV Smith (ATCC VR-194) was obtained from W. Brune (RKI, Berlin, Germany). We further used RCMV strain RA-70 and herpes simplex virus type 1 (HSV-1) strain KOS. For experiments with varicella-zoster virus (VZV), the vaccine strain V-Oka (Vailrix; Glaxo-SmithKline) was used.
BDCRB (26), TCRB (26), 2-chloro-5,6-dimethyl-1-(β-d-ribofuranosyl) benzimidazole (CDMRB), Cl4RB, BTCRB, and the deacetylated homologs of Cl4RB and BTCRB were synthesized in the laboratory of Townsend. Stock solutions of 50 mg/ml were obtained by dissolving the compounds in dimethylesulfoxid.
HFF cells (1 × 106) were seeded in 25-cm2 flasks, mock infected, infected with AD169 or GCV-sensitive or GCV-resistant isolates at an MOI of 1, treated with 0.001 to 30 μM BTCRB or left untreated, and prepared for pulsed-field gel electrophoresis (PFGE) as described previously (16). Samples and BAC-GFP-ΔUL89 (25) were loaded into a 1% low-melting-point agarose gel for electrophoresis (15 to 110 s, 150 V, 27 h). Upon completion of electrophoresis, DNA was stained with 1 μg/ml ethidium bromide and photographed.
HFF cells (1 × 105) were seeded in 24-well plates and infected with HCMV GCV-sensitive or GCV-resistant clinical isolates with an MOI of 0.01 in DMEM with 10% FCS. After 1 h postinfection (p.i.) the inoculum was replaced by 2 ml methyl cellulose (Methocel MC; Fluka) containing DMEM, 3% FCS, and inhibitors. All inhibitors were used in dilutions in duplicate. After incubation for 8 days at 37°C, the cells were stained with 0.1% crystal violet (in 20% EtOH) for 2 min and air dried prior to rinsing in Aqua Dest. Plaques were counted by the use of a microscope. Drug effects were calculated by comparing drug-treated and untreated cells.
HFF cells (1 × 105 cells per well) were seeded in 24-well culture plates. Confluent cells were infected with GCV-sensitive or GCV-resistant HCMV isolates at an MOI of 1 in the absence or presence of 10 μM BTCRB or Cl4RB, respectively. At 24, 48, 72, and 96 h p.i., the supernatants were harvested and frozen at −80°C. After collection at all time points, the supernatants were thawed, transferred to a 12-well plate, and overlaid with methyl cellulose, and titers were determined by crystal violet staining as described above.
Cytotoxicity (50% cytotoxic concentration [CC50]) of BTCRB, Cl4RB, and BDCRB was determined for HFF, Rat2 cells, and NIH 3T3 cells using the cell proliferation kit II (XTT; Roche Diagnostics, Mannheim, Germany) according to the manufacturer's recommendation. Briefly, HFF (5 × 104/well) were seeded in 96-well plates and incubated with different drug concentrations in a final volume of 100 μl for 8 days at 37°C. After the incubation period, 50 μl XTT [2,3-bis(2-methoxy-4-nitro-5-sulfo-phenyl)-2H-tetrazolium-5-carboxanilide] labeling mixture (XTT labeling reagent and electron coupling reagent; 50:1) was added to the cells. Four hours after incubation at 37°C, the absorbance of 492 nm with a reference wavelength of 650 nm was measured using an enzyme-linked immunosorbent assay reader. The assay is based on the cleavage of the yellow tetrazolium salt XTT and the formation of the orange formazan dye by metabolically active cells. An increase in the number of living cells directly correlates to the amount of orange formazan formed, as monitored by the absorbance.
HFF cells (1 × 106) were seeded in 25-cm2 flasks and infected with GCV-sensitive or GCV-resistant isolates at an MOI of 1 in the presence or absence of 10 μM BTCRB or Cl4RB. Cells were fixed at 72 h p.i. and embedded as described previously (14, 16). Sections were analyzed using a Tecnai G2 electron microscope (FEI Company, Eindhoven, The Netherlands) operated at 120 kV.
In order to determine the efficiency of the compounds in terms of viral yield, statistical analyses were performed. The results obtained from a paired Student t test were used to calculate significance. A P value of ≤0.05 was considered significant.
The antiviral efficacy of BTCRB and Cl4RB against GCV-sensitive or GCV-resistant clinical isolates was determined by measurement of viral plaque formation (plaque reduction assay) and measurement of growth kinetics. The 50% effective concentrations (EC50) against the clinical isolates obtained by plaque reduction ranged from 0.35 ± 0.19 (GCV-sensitive isolates) to 0.47 ± 0.95 (GCV-resistant isolates) for BDCRB, from 0.34 ± 0.20 (GCV-sensitive isolates) to 0.15 ± 0.07 (GCV-resistant isolates) for Cl4RB and from 0.33 ± 0.11 (GCV-sensitive isolates) to 0.50 ± 0.70 (GCV-resistant isolates) for BTCRB (Table (Table1).1). These results demonstrated that all compounds were active against HCMV clinical isolates and that Cl4RB appeared to be most active against the GCV-resistant isolates. In order to determine the effects of the compounds on the growth characteristics of GCV-sensitive or -resistant clinical isolates, cells were infected in the presence or absence of 10 μM BDCRB, BTCRB, or Cl4RB and their growths were monitored over 4 days (Fig. (Fig.1).1). At that concentration, all compounds effectively prevented viral growth to approximately the same extent.
To analyze the structure of viral DNA accumulated in the presence of either 30 μM BTCRB or 10 μM Cl4RB, DNA of cells infected with GCV-sensitive or GCV-resistant isolates was isolated 72 h p.i., separated by PFGE, and stained with ethidium bromide. PFGE analysis revealed that in untreated infected cells, cleavage of concatemers into unit-length DNA occurred as expected (Fig. 2B and C, lanes 1 and 3). Infection in the presence of BTCRB or Cl4RB with both clinical isolates resulted in loss of unit-length genomes (Fig. 2B and C, lanes 2 and 4). As a control, mock-infected as well as AD169-infected cells were used (Fig. (Fig.2A).2A). These data show that both compounds inhibit DNA cleavage of clinical isolates in a manner similar to that previously found for BDCRB (27).
In order to determine the formation of particles, ultra-thin sections of HFF cells infected with GCV-sensitive or GCV-resistant isolates in the absence or presence of 10 μM BTCRB, Cl4RB, or BDCRB were examined by EM. In infected cells grown without the inhibitors, all types of capsids (B, C, and A capsids) were formed (Table (Table2).2). In the presence of the compounds, the amount of B capsids was increased, while C capsids were reduced only twofold, but in the cytoplasm only noninfectious particles, mainly dense bodies, were released (Fig. (Fig.3;3; Table Table2).2). Taken together, these results indicate that the effects of the compounds on clinical isolates occur during packaging.
Plaque reduction assays were carried out to analyze the activity of the compounds against different herpesviruses. The concentration of all compounds (5 μM) in the assay was at least threefold lower than the CC50 (Table (Table3).3). Interestingly, BTCRB gave a remarkable reduction in VZV- and RCMV-infected cells, while it was less effective in HSV-1- and MCMV-infected cells (Table (Table4).4). In contrast, Cl4RB decreased the levels only in VZV and RCMV (Table (Table4).4). As a control, the effect of the compounds on HCMV was used (Table (Table4).4). These results showed that both compounds have the ability to reduce viral replication in VZV and RCMV, while BTCRB was able to reduce the plaque formation in all tested herpesviruses.
HCMV antiviral therapy to date includes drugs with less than ideal bioavailability and toxicity (2). Therefore, new antiviral therapeutics ought to be considered. This, in turn, would necessitate the search for novel compounds with modes of action different than inhibition of DNA polymerase. One of those new classes of compounds against HCMV represents benzimidazole ribonucleosides (18, 19, 21-26). Recently we have characterized the mode of action of two new benzimidazole derivates, BTCRB and Cl4RB (16). It was demonstrated that those compounds had a good antiviral activity against the laboratory strain AD169. Here we describe the antiviral activity against two clinical isolates, GCV-sensitive and GCV-resistant isolates. In addition the efficacy of these compounds against other herpesviruses was analyzed.
Plaque reduction assays determined EC50s against GCV-sensitive clinical isolates of approximately 0.35 μM for BDCRB, BTCRB, and Cl4RB, indicating that these clinical isolates have the same sensitivity as HCMV AD169 (EC50s of ~0.5 μM for BDCRB and BTCRB, and 0.25 μM for Cl4RB) (16). The same observation was made with GCV-resistant isolates (mean EC50 of 0.5 μM for BTCRB and BDCRB). However, Cl4RB with an EC50 of 0.15 μM was slightly more efficient. These results indicate that all compounds are effective against viral maturation in clinical isolates regardless of the acquired resistance against GCV. The effects of viral replication were further analyzed by growth curve kinetics. All compounds also prevented viral yield in cells infected with GCV-sensitive or GCV-resistant isolates.
One mode of action of these compounds is the inhibition of cleavage of concatemeric DNA into unit-length genomes (16). To analyze the structure of viral DNA which accumulated in cells infected with clinical isolates in the presence of BTCRB or Cl4RB, intracellular DNA was separated by PFGE. BTCRB, as well as Cl4RB, inhibited the formation of unit-length genomes in both clinical isolates. Another important point for characterization of antiviral activity is the formation of viral particles. EM demonstrated that in all infected cells, packaging was somehow affected, because the formation of infectious particles in the cytoplasm was completely blocked. Interestingly, DNA-containing C capsids were formed in only approximately two-fold-reduced amounts in GCV-sensitive isolates, but the reduction of C capsids in the presence of BTCRB or Cl4RB was significant. In cells infected with GCV-resistant isolates, both compounds led also to reduced amounts of C capsids and prevented formation of virions in the cytoplasm. Only noninfectious particles that assemble in the cytoplasm, dense bodies, were detected. A similar observation was reported by Nixon and McVoy (20), showing that BDCRB affected the packaging of DNA, leading to packaging of truncated genomes and blocking egress of these immature capsids. By nuclease treatment they found that the underlying mechanisms were premature cleavage events, thus leading to the assumption that BTCRB and Cl4RB may lead to inhibition of correct DNA packaging and prevent nuclear egress in HCMV laboratory and clinical strains.
In order to investigate whether these compounds might be promising compounds for antiviral therapy, analyses with different herpesviruses were performed. The two new compounds, as well as the parental one, were active against VZV and RCMV, even though to a much lesser extent compared to HCMV (Table (Table4).4). A slight inhibitory effect was observed against HSV-1 and MCMV with BTCRB. This spectrum of antiherpesvirus activity makes these compounds interesting, suggesting a similar mechanism of action in the alphaherpesviruses. It has been demonstrated that an l-analog of these compounds, maribavir, is active against Epstein-Barr virus but not against other human or animal herpesviruses, except for HCMV (28, 29). These differently restricted spectrums of antiviral activity could be due to the different modes of action of maribavir and BTCRB/Cl4RB. Maribavir inhibits viral replication by targeting the HCMV protein kinase pUL97 and prevents viral egress (3, 17), while BTCRB and Cl4RB prevent cleavage and packaging (16). All the benzimidazole ribonucleosides were active against GCV-resistant HCMV.
In summary, the tetrahalogenated benzimidazoles have demonstrated potent activity against HCMV clinical isolates, isolates that were resistant to GCV, as well as to other herpesviruses. In addition, these compounds target a DNA replication step that does not occur in eukaryotic cells. Taken together, these data support the conclusion that Cl4RB, BTCRB, and related analogs, such as BDCRB, are excellent lead compounds for the discovery and development of a benzimidazole d-nucleoside as a potent and highly specific anti-HCMV agent.
This work was supported by the Wilhelm Sander Foundation (no. 2004.031.2) and by the DFG (Deutsche Forschungsgemeinschaft, BO 1214/15-1).
We thank K. Korn (Institute of Clinical and Molecular Virology, Erlangen, Germany) for providing a GCV-sensitive HCMV isolate, K. Hamprecht (Institute of Medical Virology, Tübingen, Germany) for a GCV-resistant HCMV isolate, S. Voigt (Robert Koch Institute, Berlin, Germany) for the MCMV strain Smith, and M. Messerle (Institute of Virology, MHH) for BAC-GFP-ΔUL89. We are grateful to N. Bannert for unlimited access to his EM laboratory and G. Holland (Robert Koch-Institut, Berlin, Germany) for performing the sectioning. We thank C. Priemer for technical assistance concerning cell culture. E.B. thanks D. Krüger for kind support.
Published ahead of print on 28 September 2009.