Herpesvirus saimiri episomal persistence is maintained via interaction between open reading frame 73 and the cellular chromosome-associated protein MeCP2

Abstract

Herpesvirus saimiri (HVS) is the prototype gamma-2 herpesvirus, which naturally infects the squirrel monkey Saimiri sciureus, causing an asymptomatic but persistent infection. The latent phase of gamma-2 herpesviruses is characterized by their ability to persist in a dividing cell population while expressing a limited subset of latency-associated genes. In HVS only three genes, open reading frame 71 (ORF71), ORF72, and ORF73, are expressed from a polycistronic mRNA. ORF73 has been shown to be the only gene essential for HVS episomal maintenance and can therefore be functionally compared to the human gammaherpesvirus latency-associated proteins, EBNA-1 and Kaposi's sarcoma-associated herpesvirus (KSHV) latency-associated nuclear antigen (LANA). HVS ORF73 is the positional homologue of KSHV LANA and, although it shares limited sequence homology, has significant structural and functional similarities. Investigation of KSHV LANA has demonstrated that it is able to mediate KSHV episomal persistence by tethering the KSHV episome to host mitotic chromosomes via interactions with cellular chromosome-associated proteins. These include associations with core and linker histones, several bromodomain proteins, and the chromosome-associated proteins methyl CpG binding protein 2 (MeCP2) and DEK. Here we show that HVS ORF73 associates with MeCP2 via a 72-amino-acid domain within the ORF73 C terminus. Furthermore, we have assessed the functional significance of this interaction, using a variety of techniques including small hairpin RNA knockdown, and show that association between ORF73 and MeCP2 is essential for HVS chromosomal attachment and episomal persistence.

FIG. 1.

The HVS ORF73 C terminus associates with MeCP2 but not DEK during in vitro binding assays. (a) Bacterially expressed ORF73C was bound to nickel-agarose beads. (b) The ORF73C-conjugated beads were then incubated with HeLa cell extracts previously transfected with a plasmid expressing an empty FLAG vector (−ve), FLAG-tagged DEK, or FLAG-tagged MeCP2. Proteins attached to the ORF73C-conjugated beads were analyzed by Western blotting using a monoclonal anti-FLAG antibody as a primary antibody. Total-cell extracts representing 5% of total input were run as positive controls (+ve).

FIG. 2.

ORF73C is coimmunoprecipitated with MeCP2 in vivo. (a) 293T cells were transfected with a vector expressing the ORF73C-GFP fusion protein (pORF73C-GFP) and an empty EGFP vector (pEGFP). Cells were harvested, lysed, and incubated with or without a polyclonal anti-MeCP2 antibody (Ab). Immunocomplexes were captured using protein A agarose, washed, and separated on an SDS-PAGE gel. Analysis by Western blotting using an anti-GFP antibody indicates that ORF73C is immunoprecipitated in association with MeCP2. (b) To ensure that ORF73C is specifically immunoprecipitated by the anti-MeCP2 antibody, the immunoprecipitation was repeated using an alternative polyclonal antibody directed against HVS ORF57. Following Western blotting with an anti-GFP antibody, no interaction was observed, indicating that ORF73C is specifically immunoprecipitated in association with MeCP2. For both panels, 10% of the pORF73C-GFP and pEGFP total-cell extract was run as a positive control.

FIG. 3.

(a) Schematic representation of the ORF73 C-terminal deletion series. CAS 1 and CAS 2, described previously, are indicated (9). (b) Primers used during cloning of ORF73C fragments into the bacterial expression construct pET21b are listed and the sequences described.

FIG. 4.

In vitro binding assays using ORF73 C-terminal deletion proteins (a, b, and c) and DNase I-treated cellular extracts (d). pET21b-73C deletion proteins were expressed and bound to nickel-agarose beads (ai, bi, ci, and di) prior to incubation with HeLa cell extracts previously transfected with pFLAG-MeCP2 or an empty FLAG vector (pcDNA3.1-FLAG). Proteins associated with ORF73C-conjugated beads were analyzed by Western blotting using an anti-FLAG primary antibody (aii, bii, cii, and dii). Deletion analyses shown in panels aii, bii, and cii indicate that only full-length 73C and deletion proteins through 73CΔ4 are able to associate with MeCP2. Five to 10% of total pFLAG-MeCP2 and pcDNA3.1-FLAG cell lysates was run as a positive control where shown. Note that a small amount of MeCP2 lysate control has spilled over in the input lanes of cii; MeCP2 is not expressed in the empty FLAG control. PCR analysis using primers directed against GAPDH indicates that DNase I treatment has successfully digested cellular DNA in the treated cell extracts (diii). Panel dii demonstrates that DNase I treatment does not affect the ability of ORF73C and MeCP2 to associate, indicating that the interaction is not dependent on the presence of cellular DNA.

FIG. 5.

ORF73 and MeCP2 colocalize on mitotic chromosomes during the metaphase (a) and telophase (b) stages of mitosis. HeLa cells were serum starved, induced to divide by the replenishment of serum, stained with anti-FLAG and Texas red-conjugated antibodies, and visualized by laser scanning confocal microscopy. Images show bright field (i), TO-PRO-3 iodide-stained nucleic acid (ii), ORF73C-GFP (iii), and FLAG-MeCP2 protein (iv) at a magnification of ×400 (a) or ×630 (b). (av and bv) merged GFP and Texas red fluorescent staining, highlighting ORF73 and MeCP2 colocalization on the cellular DNA.

FIG. 6.

HVS cannot persist in NIH 3T3 cells when virus persistence is selected. (a) Mouse NIH 3T3 cells do not express MeCP2 at detectable levels when analyzed by RT-PCR. (b) (i) NIH 3T3 and SW480 cells infected with HVS-BAC-GFP successfully form colonies following episomal persistence analysis without virus selection. (ii) NIH 3T3 cells are unable to form colonies when virus persistence is selected with hygromycin. (c) Expression of heterologous MeCP2 in the NIH-MeCP2 stable cell line rescues the ability of NIH 3T3 cells to maintain HVS episomal persistence. (i) Immunofluorescent microscopy indicates that approximately 50 to 60% of NIH-MeCP2 cells express heterologous MeCP2. (ii) Colony formation analysis indicates that heterologous expression of MeCP2 in NIH-MeCP2 cells is able to rescue hygromycin-resistant colony formation.

FIG. 7.

Analysis of HVS episomal persistence following shRNA knockdown of MeCP2 expression in a latently infected cell line. (a) 293T-HVS-GFP latently infected cells constitutively express GFP as well as hygromycin and chloramphenicol resistance. The HVS genome can be maintained indefinitely as a nonintegrated, circularized episome in these cells. (b) Vectors expressing shRNA directed against MeCP2 or luciferase protein (control) were transfected into 293T-HVS-GFP cells. The cells were incubated for 0, 24, 48, and 72 h before harvest and analysis by Western blotting. Analysis of GAPDH expression indicates that all samples were loaded equally. Detection of MeCP2 with an anti-MeCP2 antibody indicates that the MeCP2 shRNA specifically inhibits MeCP2 expression after 48 and 72 h posttransfection. (c) (i) Analysis of colony formation in the absence of virus selection using hygromycin indicates that the shRNA treatment alone does not inhibit episomal colony formation. (ii) In the presence of virus selection, inhibition of MeCP2 expression in the 293T-HVS-GFP latently infected cells correlates with a substantial decrease in episomal colony formation in a 96-well episomal persistence assay.