Nuclear targeting of the cauliflower mosaic virus coat protein

Abstract

The entry of the viral genomic DNA of cauliflower mosaic virus into the nucleus is a critical step of viral infection. We have shown by transient expression in plant protoplasts that the viral coat protein (CP), which is processed from the product of open reading frame IV, contains an N-terminal nuclear localization signal (NLS). The NLS is exposed on the surface of the virion and is thus available for interaction with a putative NLS receptor. Phosphorylation of the matured CP did not influence the nuclear localization of the protein but improved protein stability. Mutation of the NLS completely abolished viral infectivity, thus indicating its importance in the virus life cycle. The NLS seems to be regulated by the N terminus of the precapsid, which inhibits its nuclear targeting. This regulation could be important in allowing virus assembly in the cytoplasm.

FIG. 1

CaMV CP constructs. (A) Schematic representation of CaMV CP. The yellow regions are rich in acidic amino acids. The black region is defined as the jelly roll (53) and is believed to be involved in protein-protein interaction between CP subunits in viral assembly (9). The blue region is the lysine-rich domain which is important for interaction with nucleic acids (9). The red region is the zinc finger believed to be involved in RNA binding. The amino acid sequence of the N terminus of p44 is shown below the scheme. The phosphorylation targets are marked in violet, and the NLS is in green. (B) Different forms of the CaMV CP used in the experiments. The small violet box represents the position of the phosphorylation target. The number 3 appears in the box when three of the serines are mutated to alanines. The letter A appears in the green box when the wt NLS sequence RKRK is mutated to AAAA. Each construct is named based on the numbering of the full-length ORF IV construct. The flag HA11 is fused to the N termini of all of the constructs to facilitate the immunodetection of these proteins.

FIG. 2

Analysis by indirect immunofluorescence of the intracellular localizations of different forms of CaMV CP after transient expression in N. plumbaginifolia protoplasts. (A, D, and G) p(77-480), p(77-411), and p(77-332) were detected in the nuclei of transfected cells. (B, E, and H) The same field of view was stained with DAPI to visualize the nuclei. (C, F, and I) Confocal pictures of different cells from the same transfections.

FIG. 3

Mapping of the NLS of CP. (A and B) Confocal pictures showing indirect immunofluorescence of the intracellular localizations of different forms of CaMV CP after transient expression in N. plumbaginifolia protoplasts. (A) p(77-332), containing the wt sequence. (B) pNL⁻(77-332), with the four positively charged amino acids of RKRK replaced by AAAA. (C) Alignment of CPs of four caulimoviruses in the NLS region. CERV, carnation etched ring virus; FMV, figwort mosaic virus; SoyCMV, soybean chlorotic mottle virus. The SV40 T-antigen NLS is also shown for comparison. The basic clusters are highlighted in the black box. The acidic amino acids are shown in boldface italics.

FIG. 4

The NLS of CaMV CP is exposed outside the virion and is present in the three processed forms of CP in the purified virion. (A and B) Gold labeling of partially purified CaMV particles with preimmune serum (A) or NLS-IgG (B). Bar, 50 nm. (C) Western blot of proteins from the purified virus probed with NLS-IgG or CP-IgG. (D) Western blot of p(126-411), pNL⁻(77-411), and p(77-332) expressed in E. coli and probed with NLS-IgG. (E) Same as panel D but probed with CP-IgG.

FIG. 5

Influence of phosphorylation and the N-terminal acidic region of CP on nuclear targeting of CaMV CP. (A) In vitro phosphorylation of CaMV CP. Serines 82, 86, and 88 were mutated to alanines in pS³⁻(77-265). (B to D) Analysis by indirect immunofluorescence of the intracellular localizations of different forms of CaMV CP after transient expression in N. plumbaginifolia protoplasts. (B) pS³⁻(77-332), in which serines 82, 86, and 88 were mutated to alanine. (C) pS³⁻NL⁻(77-332), in which serines 82, 86, and 88 and all of the basic amino acids of the NLS were mutated to alanine. (D) p(1-362), with the first 76 amino acids of the ORF IV product. (E) Western blot of transfected protoplasts expressing wt p(77-332), pS³⁻(77-332), or pS³⁻NL⁻(77-332) probed with tag antibody.

FIG. 6

Model describing the involvement of CaMV CP in the nuclear import of the CaMV genomic DNA to the nucleus.