Zn(2+) inhibits coronavirus and arterivirus RNA polymerase activity in vitro and zinc ionophores block the replication of these viruses in cell culture

Abstract

Increasing the intracellular Zn(2+) concentration with zinc-ionophores like pyrithione (PT) can efficiently impair the replication of a variety of RNA viruses, including poliovirus and influenza virus. For some viruses this effect has been attributed to interference with viral polyprotein processing. In this study we demonstrate that the combination of Zn(2+) and PT at low concentrations (2 µM Zn(2+) and 2 µM PT) inhibits the replication of SARS-coronavirus (SARS-CoV) and equine arteritis virus (EAV) in cell culture. The RNA synthesis of these two distantly related nidoviruses is catalyzed by an RNA-dependent RNA polymerase (RdRp), which is the core enzyme of their multiprotein replication and transcription complex (RTC). Using an activity assay for RTCs isolated from cells infected with SARS-CoV or EAV--thus eliminating the need for PT to transport Zn(2+) across the plasma membrane--we show that Zn(2+) efficiently inhibits the RNA-synthesizing activity of the RTCs of both viruses. Enzymatic studies using recombinant RdRps (SARS-CoV nsp12 and EAV nsp9) purified from E. coli subsequently revealed that Zn(2+) directly inhibited the in vitro activity of both nidovirus polymerases. More specifically, Zn(2+) was found to block the initiation step of EAV RNA synthesis, whereas in the case of the SARS-CoV RdRp elongation was inhibited and template binding reduced. By chelating Zn(2+) with MgEDTA, the inhibitory effect of the divalent cation could be reversed, which provides a novel experimental tool for in vitro studies of the molecular details of nidovirus replication and transcription.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Figure 1. The zinc ionophore pyrithione inhibits nidovirus replication in cell culture.

(A) Cytotoxicity of PT in Vero-E6 cells in the absence (blue circles) or presence of 2 (black squares), 4 (red triangles), or 8 µM (gray diamonds) ZnOAc2 as determined by the MTS assay after 18 hours of exposure. (B) Dose-response curves showing the effect of PT and Zn2+ on the GFP fluorescence in Vero-E6 cells infected with a GFP-expressing EAV reporter strain at 17 h p.i. Data were normalized to GFP expression in infected, untreated control cultures (100%). The different Zn2+ concentrations added to the medium were 0 (blue circles), 1 (green triangles), or 2 µM ZnOAc2 (black squares). (C) Effect of PT and Zn2+ on the GFP fluorescence in Vero-E6 cells infected with a GFP-expressing SARS-CoV reporter strain at 17 h p.i. Data were normalized to GFP expression in infected untreated control cells (100%). Colors for different Zn2+ concentrations as in Fig. 1B. Error bars indicate the standard deviation (n = 4).

Figure 2. Inhibition of the in vitro RNA-synthesizing activity of isolated RTCs by Zn2+.

Incorporation of [α-32P]CMP into viral RNA by EAV (A) and SARS-CoV (B) in RTC assays in the presence of various Zn2+ concentrations, as indicated above each lane.

Figure 3. Inhibition of nidovirus RTC activity by Zn2+ can be reversed by chelation.

(A) Schematic representation of the in vitro assays with isolated RTCs, which were initiated with [α-32P]CTP, either in the absence (sample 1 and 2) or presence of 500 µM Zn2+. After a 30-min incubation at 30°C, both the untreated and Zn2+-treated samples were split into two aliquots, and 1 mM of the Zn2+ chelator MgEDTA was added to samples 2 and 4. All reactions were subsequently incubated for another 70 min before termination. (B) Analysis of RNA products synthesized in assays with EAV RTCs. Numbers above the lanes refer to the sample numbers described under (A). (C) In vitro activity assay with SARS-CoV RTCs.

Figure 4. EAV and SARS-CoV RdRp assays with wild-type enzyme and active-site mutants.

(A) The EAV polymerase was incapable of primer extension and required a free 3′ end and poly(U) residues to initiate. Nucleotide incorporating activity of the wild-type enzyme and D445A mutant of nsp9 on an 18-mer poly(U) template confirmed the specificity of our assay. (B) SARS-CoV nsp12 RdRp assays were performed with an RNA duplex with a 5′ U10 overhang as template. The bar graph shows the nucleotide incorporating activities of wild-type and D618A nsp12. Error bars represent standard error of the mean (n = 3).

Figure 5. The activity of the RdRps of EAV and SARS-CoV is reversibly inhibited by Zn2+.

RdRp activity of purified EAV nsp9 (A) and SARS-CoV nsp12 (B) in the presence of various Zn2+ concentrations, as indicated above the lanes. (C) Schematic representation of the experiment to test whether Zn2+-mediated inhibition of RdRp activity could be reversed with MgEDTA. RdRp reactions, either untreated controls (sample 1 and 2) or reactions containing 6 mM Zn2+ (samples 3 and 4) were incubated for 30 min. Both Zn2+-containing and control samples were split into two aliquots and 6 mM MgEDTA was added to sample 2 and 4. All reactions were incubated for an additional 30 min and then terminated. Reaction products of the RdRp assays with EAV nsp9 and SARS-CoV nsp12 are shown in (D) and (E), respectively. Numbers above the lanes refer to the sample numbers described under (C).

Figure 6. Effect of Zn2+ on initiation and elongation in in vitro assays with isolated EAV and SARS-CoV RTCs.

(A) An in vitro RTC assay with isolated EAV RTCs was allowed to initiate with unlabeled NTPs (initiation). After 30 min, [α-32P]CTP was added (pulse), the reaction was split into two equal volumes, and Zn2+ was added to a final concentration of 0.5 mM to one of the tubes. At the time points indicated, samples were taken and incorporation of [α-32P]CMP into viral RNA was analyzed. (B) Radiolabeled EAV RNA synthesized at the time points indicated above the lanes in the presence and absence of Zn2+. (C) Radiolabeled RNA synthesized by isolated SARS-CoV RTCs in reactions terminated after 100 (lane 1) and 40 (lane 2) min. Reaction products of a reaction to which 500 µM Zn2+ was added after 40 min, and that was terminated at t = 100 are shown in lane 3.

Figure 7. The effect of Zn2+ on initiation and elongation activity of purified EAV and SARS-CoV RdRps.

(A) An EAV RdRp reaction was initiated in the presence of [α-32P]ATP under conditions that do not allow elongation, i.e., low ATP concentration. After 20 min, the reaction was split into two equal volumes, and Zn2+ was added to one of the tubes. A chase with 50 µM unlabeled ATP, which allows elongation, was performed on both reactions and samples were taken after 5 and 30 min. (B) EAV RdRp reaction products that accumulated in the presence and absence of Zn2+ (indicated above the lanes) after a 5- and 30-min chase with unlabeled ATP. The length of the reaction products in nt is indicated next to the gel. (C) A SARS-CoV RdRp reaction was initiated in the presence of 0.17 µM [α-32P]ATP, which limits elongation. After 10 min, the reaction was split into two equal volumes, and Zn2+ was added to one of the tubes. A chase with 50 µM unlabeled ATP was performed on both reactions and samples were taken after 5, 10, 15, and 30 min. (D) SARS-CoV RdRp reaction products formed at the chase times indicated above the lanes in the presence and absence of Zn2+. The length of the reaction products in nt is indicated next to the gel (p is the primer length).

Figure 8. The effect of Zn2+ on SARS-CoV nsp12 template binding.

(A) Electrophoretic mobility shift assay with radiolabeled dsRNA and nsp12 in the presence and absence of Zn2+ (indicated above the lanes). The position of unbound and nsp12-bound RNA in the gel is marked on the left of the panel. (B) Determination of the nsp12 affinity for RNA in the presence and absence of Zn2+. A fixed amount of RNA was incubated with an increasing amount of nsp12. This revealed a 3–4 fold reduction in the percentage of bound RNA in the presence of zinc ions (grey) relative to the percentage of bound RNA in the absence of zinc ions (black). Error bars represent standard error of the mean (n = 3).