Ribonucleotide reductase inhibitors hydroxyurea, didox, and trimidox inhibit human cytomegalovirus replication in vitro and synergize with ganciclovir

Abstract

Ganciclovir (GCV) is a deoxyguanosine analog that is effective in inhibiting human cytomegalovirus (HCMV) replication. In infected cells GCV is converted to GCV-triphosphate which competes with dGTP for incorporation into the growing DNA strand by the viral DNA polymerase. Incorporated GCV promotes chain termination as it is an inefficient substrate for elongation. Because viral DNA synthesis also relies on cellular ribonucleotide reductase (RR) to synthesize deoxynucleotides, RR inhibitors are predicted to inhibit HCMV replication. Moreover, as dGTP competes with GCV-triphosphate for incorporation, RR inhibitors may also synergize with GCV by reducing intracellular dGTP levels and there by promoting increased GCV-triphosphate utilization by DNA polymerase. To investigate potential of RR inhibitors as anti-HCMV agents both alone and in combination with GCV, HCMV-inhibitory activities of three RR inhibitors, hydroxyurea, didox, and trimidox, were determined. In both spread inhibition and yield reduction assays RR inhibitors had modest anti-HCMV activity with 50% inhibitory concentrations ranging from 36±1.7 to 221±52μM. However, all three showed significant synergy with GCV at concentrations below their 50% inhibitory and 50% toxic concentrations. These results suggest that combining GCV with relatively low doses of RR inhibitors could significantly potentiate the anti-HCMV activity of GCV in vivo and could improve clinical response to therapy.

Keywords: Antivirals; Cytomegalovirus; Didox; Ganciclovir; Hydroxyurea; Trimidox.

Figure 1

Structures of HU, DX, and TX.

Figure 2

Inhibition of HCMV yield in fibroblasts by RR inhibitors. (A) Confluent MRC-5 cultures in 96-well plates were infected with increasing infectious units (pfu/well) of RC2626 and luciferase activities in cell lysates were determined at the indicated times after infection (for simplicity, data for only four inocula are shown). (B) Luciferase data collected 24 and 48 hpi were plotted vs. pfu/well. (C and D) Confluent MRC-5 cultures in 96-well plates were infected with RC2626 (MOI = 0.03) and incubated in the presence of different concentrations of the indicated drugs for five days. 50 µl of day-five culture supernatants were transferred to fresh confluent MRC-5 cultures. After 24 h luciferase activities in cell lysates were determined. Data from three independent experiments were normalized by converting RLU to “ percent of maximum RLU” for each experiment and then averaged. Best-fit four-parameter curves were fitted to the data and used to calculate EC₅₀ values for each drug. (E) Toxicity of RR inhibitors. MRC-5 cultures in 96-well plates were incubated in the presence of different concentrations of the indicated drugs for 5 days and cell viability was measured using CellTiter-Glo. Best-fit four-parameter curves were fitted to the data and used to calculate TD₅₀ values for each drug. Each data point represents the mean of two replicate wells.

Figure 3

Synergistic inhibition of HCMV replication by combinations of GCV with HU, DX, or TX. Checkerboard arrays of GCV-HU (A), GCV-DX (B), GCV-TX (C) combinations were evaluated using the luciferase-based yield reduction assay described in figure 2. MacSynergy II software was used to calculate % inhibition above predicted additive % inhibitions for each drug combination. Positive values in the Z-axis indicate synergy for a given drug combination. Data shown represent means of data from three independent experiments.

Figure 4

Toxicity of GCV-RR inhibitor combinations. MRC-5 cultures in 96-well plates were incubated with checkerboard arrays of GCV combinations with HU, DX, or TX for 5 days, then cell viability was measured using CellTiter-Glo. Toxicity (Z-axis) for all drug combinations was calculated as described in materials and methods. Data shown represent means of data from three independent experiments.