The zinc finger antiviral protein directly binds to specific viral mRNAs through the CCCH zinc finger motifs

Abstract

The zinc finger antiviral protein (ZAP) is a recently isolated host antiviral factor. It specifically inhibits the replication of Moloney murine leukemia virus (MLV) and Sindbis virus (SIN) by preventing the accumulation of viral RNA in the cytoplasm. For this report, we mapped the viral sequences that are sensitive to ZAP inhibition. The viral sequences were cloned into a luciferase reporter and analyzed for the ability to mediate ZAP-dependent destabilization of the reporter. The sensitive sequence in MLV was mapped to the 3' long terminal repeat; the sensitive sequences in SIN were mapped to multiple fragments. The fragment of SIN that displayed the highest destabilizing activity was further analyzed by deletion mutagenesis for the minimal sequence that retained the activity. This led to the identification of a fragment of 653 nucleotides. Any further deletion of this fragment resulted in significantly lower activity. We provide evidence that ZAP directly binds to the active but not the inactive fragments. The CCCH zinc finger motifs of ZAP play important roles in RNA binding and antiviral activity. Disruption of the second and fourth zinc fingers abolished ZAP's activity, whereas disruption of the first and third fingers just slightly lowered its activity.

FIG. 1.

Mapping the target sequence of ZAP in the MLV-Luc vector. The 5′-LTR or 3′-LTR of MLV-Luc was cloned into the control reporter pGL3-Luc as indicated. The resultant constructs were transfected into Rat2-HA-Zeo or Rat2-NZAP-Zeo cells. At 48 h posttransfection, the cells were lysed and the luciferase activities were measured. The expression level of pRL-TK, a plasmid expressing Renilla luciferase, was used to normalize the transfection efficiencies. The inhibition (n-fold) was calculated as the normalized luciferase activity expressed in Rat2-HA-Zeo cells divided by the normalized luciferase activity expressed in Rat2-NZAP-Zeo cells. The data are means plus standard deviations SD of at least three independent experiments.

FIG. 2.

Mapping the target sequences of ZAP in the SIN genome. The test sequences of SIN were cloned into pGL3-Luc-linker between the luciferase coding sequence and the poly(A) signal in either a (+) or (−) orientation. The numbers indicate the positions of the ends of the fragments in the SIN genomic clone. The constructs were transfected into Rat2-HA-Zeo or Rat2-NZAP-Zeo cells. At 48 h posttransfection, the cells were lysed and analyzed for inhibition as described in the legend to Fig. 1. The data are means plus SD of at least three independent experiments. (A) The infectious clone of SIN was divided by restriction digestion into fragments, designated as indicated, and each fragment was tested for inhibition. V, control vector pGL-3-Luc-linker. (B) The D fragment from panel A was truncated from the 5′ or 3′ end, as indicated, and analyzed for inhibition. M, a fragment of D retaining most of the sensitivity. (C) The M fragment from panel B was truncated from the 5′ or 3′ end, as indicated. The fragments were cloned into pGL3-Luc-linker in the antisense orientation. N, a fragment retaining most of the sensitivity of M. (D) The N fragment from panel C was further analyzed by deletion mutagenesis, as indicated. The fragments were cloned into pGL3-Luc-linker in the antisense orientation.

FIG. 3.

ZAP does not target AREs. pGL3-Luc reporters containing different types of AREs were cotransfected into 293A cells with an empty vector or a vector expressing TTP or ZAP. At 48 h posttransfection, the cells were lysed and the luciferase activities were measured. The luciferase activity in cells transfected with the pGL3-Luc reporters was arbitrarily defined as 100. The data are means plus SD of three independent experiments.

FIG. 4.

Analysis of the effect of each zinc finger mutation on the function of ZAP. Each zinc finger of NZAP-Zeo was disrupted by the mutation of a critical residue in the CCCH motif. The mutants were packaged into MLV pseudoviruses to transduce Rat2 cells and analyzed for the inhibition of viruses or luciferase reporters. NZAP, NZAP-Zeo; H86K, NZAP-Zeo-H86K; C88R, NZAP-Zeo-C88R; C168R, NZAP-Zeo-C168R; C191R, NZAP-Zeo-C191R; HA, HA-Zeo, empty vector control. (A) The cells were infected in duplicate wells with SIN for 1 h at a multiplicity of infection of 0.01 PFU per cell. At 24 h postinfection, the supernatants were collected. For each well, the mean virus titer was obtained by titration in duplicate on BHK-J cells. For each set of infections, the inhibition (n-fold) was calculated as the mean titer of the virus produced by infection of Rat2-HA-Zeo cells divided by the mean titer of the virus produced by infection of Rat2 cells expressing NZAP-Zeo or an NZAP-Zeo mutant. The data are mean fold inhibitions plus SD. (B) Cells were infected with the Eco-Luc virus (hatched bars) or transiently transfected with the pGL3-SIN-D(+)-Luc (gray bars) or pGL3-SIN-D(−)-Luc (solid bars) reporter and then analyzed for inhibition as described in the legend to Fig. 1. (C) Cells were infected with the Eco-Luc virus or transiently transfected with pGL3-SIN-D(−)-Luc. At 48 h postinfection or posttransfection, the total RNA was extracted from the cells and the reporter RNA level was analyzed by Northern blotting using a ³²P-labeled luciferase probe. The expression level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA served as a loading control. The arrows indicate the positions of the RNAs of interest.

FIG. 5.

In vitro binding of ZAP to target RNA. The lysates of ZAP-expressing or control cells were mixed with 9E10 anti-Myc antibody and protein G-agarose resin for 2 h to immobilize ZAP to the resin. The resins were washed and incubated with the indicated ³²P-labeled RNA probes for 30 min in binding buffer. Bound RNAs were eluted by boiling in RNA sample buffer, subjected to urea-polyacrylamide gel electrophoresis, and detected by autoradiography; bound ZAP proteins were eluted by boiling in protein sample buffer and detected by Western blotting with the 9E10 antibody. (A) Binding of RNAs to NZAP-Zeo-myc. The NZAP-Zeo-myc protein was expressed in Rat2 cells and analyzed for binding to the indicated RNA probes. Rat2-vector, lysate of Rat2 cells transfected with the empty vector pcDNA4/TO-myc-His; Rat2-NZAP-Zeo-myc, lysate of Rat2 cells transfected with pcDNA4/TO/myc-NZAP-Zeo. The arrow in the lower panel indicates the position of NZAP-Zeo-myc. (B) Binding of RNAs to full-length ZAP. 293TRex, lysate of 293TRex control cells treated with tetracycline; 293TRex-ZAP, lysate of 293TRex-ZAP cells treated with tetracycline. The arrow in the lower panel indicates the position of ZAP-myc. (C) A mutation in the second zinc finger of ZAP abolished protein binding to the target RNA. 293TRex-ZAP-C88R, lysate of 293TRex cells expressing the second zinc finger mutant ZAP-C88R. The arrow in the lower panel indicates the position of ZAP-C88R-myc.