Lieve Naesens Leen De Bolle Erik De Clercq

Faculty of Pharmaceutical Sciences, Ghent University, B-9000 Gent, Belgium


In contrast to other human herpesviruses such as herpes simplex virus (HSV) or cytomegalovirus (CMV), HHV-6 has not been the subject of extensive antiviral screening, the main reason being the uncertainty about the large need for specific anti-HHV-6 therapies. Transplant recipients commonly show laboratory signs of HHV-6 reactivation, but the frequency by which this is associated with serious disease is still ill defined (Yoshikawa, 2004). Even more controversial is the role of HHV-6 in chronic neurological disorders, such as multiple sclerosis (MS) or chronic fatigue syndrome (CFS). In the absence of HHV-6-specific therapies, treatment of HHV-6 infections currently relies on the relatively broad-spectrum antiherpetic agents (val)ganciclovir and foscarnet. Although these drugs offer an indisputable benefit in the therapy or prophylaxis of CMV, their clinical efficacy against HHV-6 can only be estimated from a number of heterogeneous case reports. Long-term administration of these antivirals, as would be required in chronic diseases, is expected to cause dose-limiting side effects, and hence anti-HHV-6 drugs with a better safety and a new mode of action are urgently needed.

Antiviral assays for HHV-6

Development of new antiviral drugs requires a relevant cell culture system for virus replication. In case of HHV-6, some investigators have used the continuous T-lymphoblasts HSB-2 and Molt-3, which are the preferred host cell lines for propagation of HHV-6A or HHV-6B strains, respectively. Other T-lymphoblast lines such as SupT-1 and MT4 are less efficient in supporting HHV-6 replication (De Bolle et al., 2005b). The continuous nature of these T-cell lines ensures that the antiviral data are consistent. On the other hand, the high metabolic rate of these rapidly dividing cell lines increases their sensitivity to the cytotoxic effects exerted by the antiviral test compounds. In addition, these tumor cells may carry mutations that affect basic pathways such as nucleoside metabolism, signal transduction or cell cycle regulation. For antiviral compounds that depend on cellular factors for their activation or antiviral target, these mutations may result in cell-type-dependent antiviral effects. Therefore, primary human lymphocytes isolated from peripheral blood or cord blood provide a more relevant test system to confirm the anti-HHV-6 activity and selectivity of new compounds. Antiviral evaluation in primary or continuous cells from neuroglial origin (i.e. oligodendrocytes, microglia or astrocytes) is not commonly performed, since these cell types show a relatively inefficient replication of laboratory strains of HHV-6 (De Bolle et al., 2005b). Clinical HHV-6 isolates may show a higher neurotropism and be better suited to study the effect of antiviral compounds in these clinically relevant cell types.

Different methods can be used to estimate the inhibitory effect of antiviral compounds on HHV-6 replication. Though relatively easy and inexpensive, microscopic examination of the characteristic cytopathic effect (CPE) of HHV-6 (consisting of large ballooning cells or syncytia) may be subjective to certain extent. A definite proof of anti-HHV-6 activity requires an assay to directly measure the compound's inhibitory effect on HHV-6 replication. In antigen-based assays, HHV-6 proteins are detected with HHV-6-specific monoclonal or polyclonal antibodies. After indirect immunofluorescence staining, the fluorescence intensity in the HHV-6-infected cells is quantified by fluorescence microscopy or fluorescence-activated cell sorter (FACS) analysis (Agut et al., 1989, 1991; Reymen et al., 1995; Manichanh et al., 2000). Alternatively, HHV-6 antigen detection may be performed in an enzyme-linked immunoassay (EIA) (Takahashi et al., 1997) or by immuno-blotting ('dot-blot') assay (Yoshida et al., 1998). Unfortunately, the choice of commercial antibodies for HHV-6 is limited. DNA-based assays with custom-made primers or probes may therefore be preferred. We have good experience with a non-radioactive DNA hybridization assay to quantify viral DNA in HHV-6-infected cells (De Clercq et al., 2001). This technique can now be replaced by realtime PCR.

Care should be taken when comparing antiviral data for HHV-6 from different laboratories. The antiviral activity and selectivity may depend on the HHV-6 strain and cell system used. No standardization exists as to what should be the viral load ('multiplicity of infection'). Some compounds loose their anti-HHV-6 activity at higher viral loads. Also, the activity may depend on the assay used to follow HHV-6 replication. For instance, compounds that act at a late stage in the viral replication cycle (such as inhibitors of virus maturation) may show good activity in a CPE assay, but have a minor effect on viral DNA content.

Classical viral DNA polymerase inhibitors Mechanism of action

Since the discovery of acyclovir, inhibitors of the viral DNA polymerase have been the mainstay in anti-herpesvirus therapy. For the acyclic nucleoside analogs acyclovir and ganciclovir, inhibition of viral DNA synthesis is carried out by their triphosphate metabolite, which is formed in three consecutive phosphorylations (De Bolle et al., 2005a). The first is catalyzed by a virus-encoded kinase such as the HSV thymidine kinase, or the protein kinase encoded by the CMV UL97 gene or the HHV-6 U69 gene. The second and third phosphorylation steps are carried out by cellular enzymes. The resulting triphosphate metabolite inhibits the herpesvirus DNA polymerase by competition with the natural substrate dGTP, by incorporation in the growing DNA (effecting chain termination in case of absence of a hydroxyl group in the acyclic side chain) and/or by dead-end complex (Reardon and Spector, 1989). The selectivity of the acyclic nucleoside analogs relies on a greater affinity of their triphosphates for the viral DNA polymerase compared to cellular polymerases and, more importantly, their selective phosphorylation by a viral kinase. In comparison, the acyclic nucleoside phosphonate analog cidofovir does not require activation by a viral kinase, explaining its broad-spectrum anti-DNA virus activity. Due to the presence of the phosphonate moiety, two subsequent phosphorylations by cellular enzymes are sufficient for cidofovir to reach its active state. Cidofovir diphosphate acts as a competitive inhibitor for dCTP and an alternative substrate for the herpesvirus DNA polymerase. Its selectivity is based on its higher affinity for the viral DNA polymerase compared to cellular DNA polymerases. This is also the case for the pyrophosphate analog foscarnet, for which the inhibition of the herpesvirus DNA polymerase is based on its reversible binding to the pyrophosphate-binding site of the enzyme. Consequently, cleavage of pyrophosphate during incorporation of dNTPs is prevented and viral DNA synthesis is terminated.

In vitro data

As summarized in Table 1, several groups have determined the anti-HHV-6 activity of acyclovir, ganciclovir, cidofovir, and foscarnet in cell culture. In general, acyclovir appeared to have poor activity and selectivity, while foscarnet was consistently shown to have a pronounced effect on HHV-6 replication with relatively high selectivity. In contrast, marked differences can be seen in the data reported for ganciclovir and cidofovir, reflecting variations in the experimental conditions (Table 1). Both compounds are considerably more active and less cytotoxic in primary human lymphocytes than in continuous T-cell lines.

In case of ganciclovir, the anti-HHV-6 activity is diminished by two factors: inefficient phosphorylation of ganciclovir by the HHV-6 pU69 protein kinase and relatively low affinity of ganciclovir triphosphate for the HHV-6 DNA polymerase.

bo 4

Table 1

Activity of viral DNA polymerase inhibitors against HHV-6 in cell culture

Table 1

Activity of viral DNA polymerase inhibitors against HHV-6 in cell culture

Cell system3


Virus strain (variant)

Antiviral EC50c (mM)

Was this article helpful?

0 0

Post a comment