The effects of maribavir on the autophosphorylation of ganciclovir resistant mutants of the cytomegalovirus UL97 protein

Abstract

Background: The UL97 protein kinase of human cytomegalovirus phosphorylates the antiviral drug ganciclovir and is the target of maribavir action. A detailed enzyme kinetic analysis of maribavir on the various enzymatic functions of wild type and ganciclovir resistant forms of UL97 is required.

Methods: Wild type and site directed mutant forms of the human cytomegalovirus UL97 gene product were expressed using recombinant baculoviruses and the purified products used to assess the effects of maribavir on the ganciclovir (GCV) kinase and protein kinase (PK) activities.

Results: Maribavir was a potent inhibitor of the autophosporylation of the wild type and all the major GCV resistant UL97 mutants analysed (M460I, H520Q, A594V and L595F) with a mean IC50 of 35 nM. The M460I mutation resulted in hypersensitivity to maribavir with an IC50 of 4.8 nM. A maribavir resistant mutant of UL97 (L397R) was functionally compromised as both a GCV kinase and a protein kinase (~ 10% of wild type levels). Enzyme kinetic experiments demonstrated that maribavir was a competitive inhibitor of ATP with a Ki of 10 nM.

Discussion: Maribavir is a potent competitive inhibitor of the UL97 protein kinase function and shows increased activity against the M460I GCV-resistant mutant which may impact on the management of GCV drug resistance in patients.

Figure 1

Coomassie Blue stained SDS-PAGE gel showing the time course of expression of the wild type UL97 protein using either the honey-bee mellitin (BVBM-UL97; lanes 1-3) or the BlueBacHis (BVBBH-UL97; lanes 4-6) recombinant baculovirus system illustrating the higher expression levels observed in the honey-bee mellitin system. For comparison, a recombinant baculovirus expressing the LacZ gene (BV-LacZ; lanes 7-10) and mock infected insect cells (mock; lanes 10-12) are also shown. Arrows indicate the UL97 and LacZ proteins and molecular weight markers.

Figure 2

A) SDS-PAGE analysis of purified wild type and a selection of mutant UL97 proteins used in the study. B) Western blot analysis showing the comparable expression levels of wild type and mutant UL97 proteins.

Figure 3

Ganciclovir (GCV) phosphorylation catalysed by the wild type and UL97 mutant proteins (designated by the amino acid change present). Insect cells were infected with the wild type or mutant UL97 expressing baculoviruses at an MOI 10. The medium was supplemented with tritiated GCV at 48 hours. The nucleotides were harvested at 72 hours and bound to DE81 filter paper. The phosphorylated GCV is plotted as percentage phosphorylation compared to the wild type UL97 (set at 100%). Data are the mean +/- one standard deviation of three experiments.

Figure 4

The kinetics of autophosphorylation of UL97 (panel A). The density of the bands was analysed by the BioRad Multianalyst software and then plotted against the corresponding ATP concentration (B). Data shown are for the wild type UL97. Data are the mean +/- one standard deviation of three experiments.

Figure 5

Effects of maribavir on the autophosphorylation of wild type UL97 and the various mutant proteins. Autoradiographs showing the Inhibition of autophosphorylation of wild type and mutant UL97 proteins by maribavir. The presence or absence of maribavir (0.5 μM) is indicated by + or-above each lane. UI = uninfected control insect cells.

Figure 6

IC₅₀ of maribavir for the wild type and mutant UL97 Proteins. The wild type and mutant UL97 proteins were subjected to protein kinase assays with varying concentrations of maribavir (0.01-1.0 μM). The autoradiographs were analysed using the BioRad Multianalyst software and UL97 phosphorylation plotted as a percentage of the total phosphorylation in the absence of maribavir. An exponential decay curve was fitted and the IC₅₀ of maribavir was determined for each of the proteins. The alteration of x-axis scale of the M460I graph should be noted. The IC₅₀ for each species is shown.

Figure 7

Competitive inhibition of ATP binding by maribavir. Protein kinase assays were performed containing increasing concentrations of ATP (2 μM-20 μM) in the absence of maribavir (0.0 μM on the graph) and repeated for increasing maribavir concentrations (0.01 μM, 0.015 μM, 0.2 μM or 0.25 μM). The inverse of the velocity of autophosphorylation was plotted against the corresponding inverse of the [ATP] and the Ki value for maribavir determined using standard enzyme kinetic equations for competitive inhibition.