Physical interaction between the herpes simplex virus type 1 exonuclease, UL12, and the DNA double-strand break-sensing MRN complex

Abstract

The herpes simplex virus type 1 (HSV-1) alkaline nuclease, encoded by the UL12 gene, plays an important role in HSV-1 replication, as a UL12 null mutant displays a severe growth defect. The HSV-1 alkaline exonuclease UL12 interacts with the viral single-stranded DNA binding protein ICP8 and promotes strand exchange in vitro in conjunction with ICP8. We proposed that UL12 and ICP8 form a two-subunit recombinase reminiscent of the phage lambda Red α/β recombination system and that the viral and cellular recombinases contribute to viral genome replication through a homologous recombination-dependent DNA replication mechanism. To test this hypothesis, we identified cellular interaction partners of UL12 by using coimmunoprecipitation. We report for the first time a specific interaction between UL12 and components of the cellular MRN complex, an important factor in the ATM-mediated homologous recombination repair (HRR) pathway. This interaction is detected early during infection and does not require viral DNA or other viral or cellular proteins. The region of UL12 responsible for the interaction has been mapped to the first 125 residues, and coimmunoprecipitation can be abolished by deletion of residues 100 to 126. These observations support the hypothesis that cellular and viral recombination factors work together to promote efficient HSV-1 growth.

FIG. 1.

A subpopulation of UL12 is detergent resistant and localizes to replication compartments. (Top panels) KOS-infected Vero cells were fixed and permeabilized as described in Materials and Methods. (Bottom panels) KOS-infected Vero cells were extracted with CSK buffer containing Triton X-100 prior to fixation to remove nucleosolic and cytosolic fractions, followed by fixation and permeabilization. The cells were labeled with the polyclonal anti-UL12 antibody BWpUL12 (green), and the monoclonal anti-ICP8 antibody 39S (red) was used as the marker for viral replication compartments. Merged pixels are shown in yellow. In this experiment the majority of cells (34 out of 38) showed that UL12 localized in replication compartments in preextracted cells. Arrows indicate replication compartments within a single nucleus.

FIG. 2.

UL12 interacts with components of the cellular MRN complex in HSV-1-infected cell lysates. (A) Vero cells were either mock infected (lane 2) or infected with UL12.5 virus ANF-1 (lane 1), UL12 null virus AN-1 (lane 3), wild-type (WT) KOS virus (lanes 4 and 5), or ICP8 null virus HD-2 (lane 6). Anti-UL12 antibody was used to immunoprecipitate cell lysates of infected cells collected 6 hpi as described in Materials and Methods, except for lane 5, in which normal rabbit IgG was used as a control (indicated by the *). Input samples contained 3% of the precleared cell lysate collected prior to immunoprecipitation. UL12 fragments that were immunoprecipitated with anti-UL12 antibody are shown in the panel labeled IP. Proteins that coimmunoprecipitated with UL12 are shown in the panels labeled Co-IP. The arrow indicates the position of the band corresponding to full-length Mre11. (B) Vero cells were infected with KOS and collected at the indicated time points postinfection. Immunoprecipitation was performed, and results are displayed as described for panel A. (C) Vero cells were mock infected (lane 1) or infected with WT KOS virus (lane 2), UL12 null virus AN-1 (lane 3), or ICP8 null virus HD-2 (lane 4). Infected cells were collected 6 hpi, and cell lysates were prepared for immunoblotting as described in Materials and Methods. For the experiment shown in this panel, the proteins were resolved for a longer period of time to highlight the mobility shift of Nbs1 upon HSV-1 infection. Tubulin was used as the loading control.

FIG. 3.

The UL12-MRN interaction does not require DNA or other viral proteins. (A) Vero cells mock infected or infected with UL12 null virus AN-1 or the wild-type (WT) KOS virus, as indicated, were collected 6 hpi. Cell lysates were immunoprecipitated with anti-UL12 antibody. In lane 3 (+), the sample was subjected to DNase treatment during immunoprecipitation, while in lanes 1, 2, and 4 (−), immunoprecipitation was carried out in the absence of DNase. (B) Vero cells were transfected with either pSAK UL12/12.5 expressing the wild-type UL12 under the cytomegalovirus promoter (lane1) or with pSAK empty vector (lane 2). Cells were collected 20 h posttransfection, and cell lysates were immunoprecipitated using anti-UL12 antibody and subjected to DNase treatment as described in Materials and Methods.

FIG. 4.

UL12 directly binds to the MRN complex. SPR sensorgrams of the UL12 interaction with the MRN complex at increasing concentrations of 6.25, 12.5, 25, 50, and 100 nM were performed using a Biacore T100 at room temperature. The black lines represent the binding curves of the analytes at specific concentrations, and the gray lines represent the 1:1 binding model curves which were used to fit the data and calculate the kinetics of the interaction.

FIG. 5.

UL12 interacts with components of the MRN complex through the N-terminal region of UL12. (A) Schematic representation of the UL12 constructs used in panels B and C. (B) Vero cells were transfected with plasmids expressing C-terminal truncations of UL12, wild-type (WT) UL12, and UL12.5 as indicated. Transfected cells were fixed 20 h posttransfection and tested for expression and localization of the UL12 fragments by immunofluorescence. (C) Whole-cell lysates of Vero cells transfected with full-length UL12 (lane 1), C-terminal truncations (lanes 2 to 6), or empty vector (lane 7) were immunoblotted and probed with UL12 antibody. Full-length UL12 and truncations were able to express UL12 fragments of the expected sizes. (D) Vero cells transfected with plasmids expressing C-terminal truncations of UL12 as indicated (lanes 1 to 6), empty vector (lane 7), wild-type UL12 (lane 8), or UL12.5 (lane 9) were collected at 20 h posttransfection. Cell lysates were immunoprecipitated using anti-UL12 antibody and probed for the indicated proteins to detect interactions. The immunoprecipitation reaction was subjected to DNase treatment as described in Materials and Methods. The asterisks mark nonspecific bands arising due to the cross-reaction between the secondary antibody and IgG.

FIG. 6.

The N-terminal 50 amino acids of UL12 are not required for the UL12-MRN interaction. (A) Schematic representation of the N-terminal deletion constructs used for the experiment shown in panel B. (B) Vero cells were transfected with plasmids expressing N-terminal truncations of UL12 as indicated (lanes 1 to 5), empty vector (lane 6), full-length UL12 (lane 7), or UL12.5 (lane 8). Cells were collected at 20 h posttransfection, and cell lysates were immunoprecipitated using anti-UL12 antibody as described in the legend for Fig. 5. The asterisk marks a nonspecific band arising due to the cross-reaction between the secondary antibody and IgG.

FIG. 7.

The N-terminal amino acids 100 to 126 of UL12 are essential for the UL12 interaction with the components of the MRN complex. (A) Schematic representation of UL12 constructs used in the experiments shown in panels B and C. (B) Vero cells were transfected with plasmids expressing the N-terminal internal deletions of UL12 as indicated. Cells were fixed at 20 h posttransfection and tested for UL12 expression and localization by immunofluorescence. (C) Vero cells were transfected with plasmids expressing the indicated constructs described for B (lanes 1 to 4) or full-length UL12 (lane 5). Cells were collected at 20 h posttransfection and immunoprecipitated as described in the legend for Fig. 5.