The major tegument structural protein VP22 targets areas of dispersed nucleolin and marginalized chromatin during productive herpes simplex virus 1 infection

Abstract

The herpes simplex virus (HSV) major tegument structural protein VP22 resides in multiple subcellular regions during productive infection. During an analysis of the molecular determinants of these localizations, we observed that a transfected fusion of the C-terminal portion of VP22, containing its pat4 nuclear localization signal, with GFP lacked nucleolar sparing compared to GFP alone. Thus, the initial goal was to determine whether VP22 associates with nucleoli. Using an optimized indirect immunofluorescence system to visualize nucleolin and viral proteins, we observed that VP22 present in VP22-expressing Vero (V49) cells "surrounded" nucleolin. These two initial findings implied that VP22 might associate directly with nucleoli. We next analyzed HSV-infected cells and observed that at late times, anti-nucleolin immune reactivity was dispersed throughout the nuclei while it retained uniform, circular staining in mock-infected cells. Time course infection experiments indicated that nucleolin initiated its transition from uniform to dispersed structures between 2 and 4 hpi. Comparison of Hoechst stained nuclei showed bright anti-nucleolin staining localized to regions of marginalized chromatin. These effects required de novo infected cell protein synthesis. A portion of VP22 detected in nuclei at 4 and 6 hpi localized to these areas of altered nucleolin and marginalized chromatin. VP22 was excluded from viral replication compartments containing the viral regulatory protein ICP22. Finally, altered nucleolin and marginalized chromatin were detected with a VP22-null virus, indicating that VP22 was not responsible for these nuclear architecture alterations. Thus, we conclude that nuclear VP22 targets unique subnuclear structures early (<6hpi) during herpes simplex virus 1 (HSV-1) infection.

Fig. 1

Schematic representation of the recombinant VP22pat4GFP construct used (A). Line 1 - the HSV-1 (F) genome with the unique long (U_L), unique short (U_S), and terminal repeat segments (a, b, c, a′, b′, and c′) indicated. Coordinates of the BamF fragment are indicated. Line 2 - gene U_L49 encoding full length VP22 showing approximate locations of pat4 and pat7 nuclear localization signal motifs. Solid arrowhead shows the direction of transcription. Line 3 - orientation of GFP indicating that a portion of U_L49 remains at its carboxy terminus and the potential for a GFP carboxy-terminal VP22 (GFP-CTermVP22) chimera expressed by pJB177. Line 4 -sequence of the in frame amino acids of VP22pat4GFP between GFP and 15 aa of the VP22 carboxy terminus, including linker residues. The pat4 nuclear localization signal motifs is shown. Transfected VP22pat4GFP does not spare subnuclear circular structures (B). Vero cells were transfected with either plasmid pJB177 expressing VP22pat4-GFP or pJB49 expressing GFP, nuclei were stained with Hoechst dye, and cells were visualized live by fluorescence microscopy. Merged images (overlay) were prepared to emphasize nuclear staining. Arrows mark staining patterns of GFP sparing of intranuclear structures in representative cells. Panels were enlarged for better visualization. Subnuclear structures bind RNA dye (C). Live Vero cells were stained with Hoechst and SYTO RNASelect dyes and cells were visualized by fluorescence microscopy. Arrows mark consistent SYTO staining with a Hoechst void pattern. Subnuclear structures contain nucleolin (D). Vero cells were fixed/permeablized, stained with antibody specific for nucleolin, and cells were visualized by fluorescence microscopy. Arrows mark nucleolin staining patterns. Phase contrast (Phase) and merged (overlay) images are also shown. Magnification, 60×.

Fig. 2

The circular nucleolin staining pattern is independent of fixing conditions (A). Vero cells were fixed with 2.5% paraformaldehyde and then permeablized using two different conditions; either acetone (−20°C) or Triton X-100 (4°C). Phase contrast (Phase) and merged (overlay) images are shown. VP22 surrounds nucleolin in VP22-expressing V49 cells (B). Vero and V49 cells were immunostained with antibodies specific for VP22 and nucleolin. Arrows mark a portion of VP22 staining in V49 cells which is circular and resembles that of nucleolin. Nuclei were visualized following Hoechst staining (A and B). Magnification, 100×.

Fig. 3

Altered nucleolin staining observed at late infection time points. Vero cells were synchronously mock- or HSV-1(F)-infected. At 13 and 24 hpi, cells were fixed/permeablized and immunostained with antibodies specific for VP22 and nucleolin. Nuclei were visualized following Hoechst staining. Magnification, 100×.

Fig. 4

Altered nucleolin staining detected at 4 h post infection. Vero cells were synchronously mock- or HSV-1(F)-infected. At 2, 4, and 8 hpi, cells were fixed/permeablized and immunostained with antibodies specific for VP22 and nucleolin at 4 hpi. Nuclei were visualized following Hoechst staining. Overlay images of Hoechst and nucleolin at 8 hpi show colocalization of altered nucleolin with marginalized chromatin in infected cells. Arrows mark a portion of VP22 staining which is circular and surrounds altered nucleolin. Magnification, 100×.

Fig. 5

Altered nucleolin in infected VP22 expressing V49 cells. V49 cells were synchronously mock- or HSV-1(F)-infected. At 2, 4, and 8 hpi, cells were fixed/permeablized and immunostained with antibodies specific for VP22 and nucleolin. Nuclear chromatin was visualized following Hoechst staining. Arrows mark a portion of VP22 staining which surrounds nucleolin; at 4 and 8 hpi VP22 targets altered nucleolin and marginalized chromatin. Magnification, 100×.

Fig. 6

Alteration of subnuclear structures is not dependent on VP22. Vero cells were mock- or HSV-1(F), Bac-F-, and Bac-ΔVP22-infected. At 6 hpi, cells were fixed/permeablized and immunostained with antibodies specific for VP22 and nucleolin. Nuclei were visualized following Hoechst staining. Magnification, 100×.

Fig. 7

Alteration of subnuclear structures is not dependent on VP22. Vero cells were synchronously mock- or HSV-1(F), Bac-F-, Bac-ΔVP22-, and Bac-ΔVP22Repair-infected. At 6 hpi, cells were fixed/permeablized and immunostained with antibodies specific for VP22 and nucleolin. Nuclei were visualized following Hoechst staining. Arrows mark a portion of VP22 staining which targets altered nucleolin and marginalized chromatin. Magnification, 100×.

Fig. 8

Alteration of subnuclear structures is not dependent on VP22. Vero cells were synchronously mock- or HSV-1(F), Bac-F, Bac-ΔVP22-, and Bac-ΔVP22Repair-infected. At 6 hpi, cells were fixed/permeablized and immunostained with antibodies specific for ICP22 and nucleolin. Nuclei were visualized following Hoechst staining. Arrows mark ICP22 exclusion of nucleolin. Magnification, 100×.

Fig. 9

Alteration of subnuclear structures requires de novo infected cell protein synthesis. Vero cells were synchronously mock- or HSV-1(F), Bac-F-, Bac-ΔVP22-, and Bac-ΔVP22Repair-infected in the presence (plus) and absence (minus) of CHX. At 6 hpi, cells were fixed/permeablized and immunostained with antibodies specific for ICP22 and nucleolin. Nuclei were visualized following Hoechst staining. Merged images (overlay) were prepared to emphasize staining patterns. Arrows mark ICP22 exclusion of nucleolin. Magnification, 100×.