Abstract
The bovine papillomavirus type 1 E2 transactivator protein is required for viral transcriptional regulation and DNA replication and may be important for long-term episomal maintenance of viral genomes within replicating cells (M. Piirsoo, E. Ustav, T. Mandel, A. Stenlund, and M. Ustav, EMBO J. 15:1–11, 1996). We have evidence that, in contrast to most other transcriptional transactivators, the E2 transactivator protein is associated with mitotic chromosomes in dividing cells. The shorter E2-TR and E8/E2 repressor proteins do not bind to mitotic chromatin, and the N-terminal transactivation domain of the E2 protein is necessary for the association. However, the DNA binding function of E2 is not required. We have found that bovine papillomavirus type 1 genomes are also associated with mitotic chromosomes, and we propose a model in which E2-bound viral genomes are transiently associated with cellular chromosomes during mitosis to ensure that viral genomes are segregated to daughter cells in approximately equal numbers.
Certain DNA viruses, such as papillomavirus or Epstein-Barr virus (EBV), are able to maintain their genomes as stable extrachromosomal elements in the nuclei of infected cells. Papillomaviruses infect and replicate in stratified epithelium and give rise to benign lesions called warts or papillomas (reviewed in reference 17). There appear to be three stages of DNA replication that take place in the papillomavirus life cycle. Initially, the virus infects basal epithelial cells, and after uptake of the virus, the viral genome is transported to the nucleus of the basal cell, where it is presumed to be amplified to a low copy number. Most experimental studies have examined transient DNA replication in cultured cells, a system that is most analogous to this initial amplification stage and which requires the E1 and E2 proteins and the viral replication origin (44, 45). Infected basal cells of a papilloma proliferate and are thought to maintain low levels of extrachromosomal viral DNA. The genomes of papillomaviruses can also be stably maintained as high-copy-number extrachromosomal elements in certain cell lines (9, 24), and the viral genomes replicate in synchrony with cellular DNA. Overall, the viral genome copy number remains constant, but the genomes are replicated by a random choice mechanism (11, 36). The third stage of viral replication is vegetative DNA synthesis and is required to generate progeny virus. Vegetative DNA replication occurs only as the basal cells of a papilloma migrate upwards and differentiate in the stratified epithelium. However, very little is known about vegetative viral DNA replication because of the requirement for terminally differentiating keratinocytes and difficulties in reproducing these conditions in a culture system.
Papillomavirus DNA replication requires the full-length E2 transactivator protein, the viral E1 protein, and the replication origin (44, 45). The minimal origin of replication consists of an E1 binding site, an E2 binding site, and an AT-rich region that may facilitate origin unwinding. The E1 protein has several replication-associated activities such as origin-specific binding and helicase activities and forms a complex with the E2 transactivator (19, 20, 48). The E2 protein is the major transcriptional transactivator of the virus, but it is also required for viral DNA replication. The E2 protein plays an auxiliary role in replication by enhancing and regulating the functions of the E1 protein. E2 has been shown to cooperatively bind to the origin with the E1 protein (4, 33, 38, 39, 42), to alleviate repression of replication by nucleosomes (26), and to interact with cellular replication proteins (RPA) (25). The bovine papillomavirus type 1 (BPV-1) E2 open reading frame also encodes two shorter polypeptides that repress E2-mediated transactivation (8, 23) (Fig. 1B). These proteins, E2-TR and E8/E2, contain the DNA binding-dimerization domain, but their role in replication is not clear.
FIG. 1.
(A) Diagram of the BPV-1 genome. The open reading frames E1 to E8 and L1 and L2 are shown. Promoters are represented by arrows, and E2-specific DNA binding sites are represented by small black circles. The LCR origin of replication (ori), and MME are also indicated. (B) Map of the E2 transactivator and repressor proteins. The full-length E2 protein is a transcriptional transactivator that can be expressed from the P2443 promoter. The E2-TR repressor protein is expressed from the P3080 promoter and initiated at an internal initiation codon. The E8/E2 repressor protein is encoded by a spliced message that links 11 amino acids of the E8 open reading frame to the C-terminal half of the E2 open reading frame.
Plasmids containing the minimal replication origin can replicate transiently in cells expressing the E1 and E2 proteins, but the replicated DNA is lost with time. Long-term, stable maintenance of such plasmids requires expression of the E1 and E2 proteins, the replication origin, and a region from the long control region (LCR) that has been designated a minichromosome maintenance element (MME) (34). This element contains multiple high-affinity E2 binding sites and can be replaced with a sequence of 10 tandem E2 sites (34). This suggests that the E2 protein may play a role in plasmid copy number control and viral genome segregation. To gain insight into the mechanism by which papillomavirus genomes are stably replicated, we have examined the intracellular localization of the viral genome and E2 proteins in mitotic cells.
MATERIALS AND METHODS
Cell culture.
COS-7, CMT4 (10), ID13 (24), CV-1, and C127-derived lines were cultured in Dulbecco’s minimal essential medium supplemented with 10% fetal calf serum. CHO-derived lines were cultured in F-12 medium supplemented with 10% fetal calf serum. Recombinant simian virus 40 (SV40) PAVA E2 virus was produced in CMT4 cells, as described previously (40). C127 cells expressing the E2 proteins under the control of a tetracycline-regulated promoter were generated by cotransfecting pSV2neo, a plasmid expressing a tetracycline-regulated transcriptional repressor, pTET-tTAK (GIBCO BRL) with pTET-splice plasmids (GIBCO BRL) expressing the E2-TA and E2-TR proteins. G418-resistant colonies were isolated and screened for expression of the E2 proteins by immunofluorescence. CHO cells expressing the E2 proteins were generated by transfecting a CHO line (AA8) that expresses a tetracycline-regulated transcriptional repressor (Clontech) with pTK-Hyg (Clontech) and pTET-splice plasmids (GIBCO BRL) expressing the E2-TA, E2-TR, and E8/E2 proteins from a tetracycline-regulated cytomegalovirus promoter. Hygromycin B-resistant colonies were isolated and screened for expression of the E2 proteins by immunofluorescence. 137 cells were established from a clone of C127 cells transformed with a BPV-1 genome containing mutations in the major phosphorylation sites of the E2 proteins (serine to alanine at positions 290, 298, and 301) (31). 209 cells were derived from C127 cells transformed with a cloned BPV-1 genome, p1472, that is unable to express the E2-TR protein (22); in both cases, the viral genome was cleaved from the prokaryotic vector sequences and religated before being transfected into cells.
Plasmids and viruses.
The recombinant PAVA virus expressing the E2 protein, pPAVAkzE2, has been described previously (pSB-E2kz [29]). pPAVAkzE2-TA and pPAVAkzE2-TA K344 were designed to express the E2-TA protein only. The initiating methionine of E2-TR (amino acid 162 of E2-TA) in these constructs was changed to an isoleucine by mutating nucleotide 3093 from G to C (codon change of ATG to ATC) (41). An E2 fragment (Asp718 to BstXI) containing the K344 mutation was subcloned from pTZE2R344K (47) into pPAVAkzE2-TA to generate pPAVAkzE2-TA K344. pPAVAkzE2-TAΔ41-120 and pPAVAkzE2-TAΔ51-120 have been described previously (41). BPV-1 genomes containing a serine-to-alanine mutation at position 301 (142-6 A301) and a mutation changing the initiating methionine of E2-TR to a threonine (p1472-1) have been described previously (22, 31).
Transient expression and immunofluorescence.
Cells were plated onto glass slides 16 h before infection or induction of the tetracycline-regulated promoter. For PAVA virus E2 expression, CV-1 cells were infected with virus at a high multiplicity of infection and were analyzed for E2 expression after 40 to 44 h. Where indicated, cultures were treated with 30 ng of colchicine per ml for 30 min before fixation to block cells in metaphase. Cells were fixed for 30 min in 3.7% formaldehyde solution in phosphate-buffered saline (PBS) and permeabilized with 0.1% Triton X-100 in PBS. Mouse monoclonal anti-E2 antibodies, B201 and B202 (provided by Elliot Androphy), were added at dilutions of 1:10 and 1:100, respectively, in PBS block solution. Antiserum against SV40 T antigen was obtained from Oncogene Sciences and used at a 1:30 dilution. Slides were incubated with the primary antibody, washed with PBS, and incubated with goat anti-mouse or anti-rabbit immunoglobulin G conjugated to fluorescein isothiocyanate (FITC; 1:100 dilution; Jackson Immunochemicals). Following washes in PBS, slides were mounted in Vectashield mounting fluid (Vector Laboratories) containing 0.2 μg of propidium iodide per ml. Immunofluorescence was detected and photographed with a Bio-Rad MRC600 confocal laser scanning imaging system.
Fluorescent in situ hybridization (FISH).
Cells were grown on glass slides and treated with colchicine as described above. Cells were fixed for 20 min in methanol-acetic acid (3:1). Where indicated, cells were swollen for 20 min in a hypotonic solution to separate the metaphase chromosomes before fixation. Slides were treated with 0.1 mg of RNase A per ml–200 U of Aspergillus oryzae RNase per ml in 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate) for 1 h at 37°C, rinsed in 2× SSC, and dehydrated in a graded series of ethanol. Cellular and viral DNA on the slides was denatured for 15 min at 75°C in 50% formamide–5× SSC, and the slides were dehydrated in a graded series of chilled ethanol. BPV-1 and SV40 probe DNAs were prepared by labeling with fluor-12-dUTP with a Prime-It Fluor fluorescence labeling kit (Stratagene) and 50 ng (per slide) coprecipitated with 6 μg of sheared competitor DNA. The probe was added to 30 μl of hybridization solution (50% formamide, 10% dextran sulfate, 4× SSC), denatured for 15 min, and incubated on the slides at 37°C overnight. Slides were washed three times in 50% formamide–2× SSC at 50°C and three times in 0.01× SSC at 65°C. Slides were rinsed briefly in 4× SSC–0.01% Tween 20 and were mounted in Vectashield mounting fluid (Vector Laboratories) containing 0.2 μg of propidium iodide per ml. Fluorescence was detected and photographed with a Bio-Rad MRC600 confocal laser scanning imaging system.
DNA binding assay.
Plasmid pTZE2290–410, which encodes the DNA binding domain of E2, has been described elsewhere (30). The R344K substitution was generated in this background as described previously (32). E2 proteins were generated by in vitro transcription and translation (30) and tested for DNA binding by an electrophoretic mobility shift assay as described previously (30), except that no poly(dI-dC) or nonspecific DNA was included in the reaction mixture. Varying amounts of double-stranded end-labeled oligonucleotide probe (5′-TCGAACCGAAAACGGTGTCGA-3′) were incubated with either 0.1 μl of control reticulocyte lysate, E2290–410 lysate, or E2290–410 K344 lysate. The amount of bound probe was measured with a radioanalytic imaging system (AMBIS Systems, San Diego, Calif.).
RESULTS
BPV-1 viral genomes and E2 proteins are closely associated with mitotic chromatin.
To investigate the mechanism of papillomavirus genome segregation and to analyze the role of E2 in this process, the localization of BPV genomes and E2 proteins was determined in dividing cells that stably maintain the viral genomes as extrachromosomal elements. Initial studies used a cell line (137) that contains a BPV-1 genome with mutations in the major phosphorylation sites of the E2 proteins (serine to alanine at positions 290, 298, and 301) (31). The E2 proteins are normally present at very low levels in cells transformed by wild-type BPV-1 (18), but the A301 mutation results in very high levels of episomal viral DNA and detectable levels of E2 protein (28, 31). Cellular DNA was stained with propidium iodide to allow identification of cells undergoing mitosis, and the localization of viral genomes was determined by hybridization with a viral DNA probe. As shown in Fig. 2, most viral DNA is closely associated with mitotic cellular chromosomes. The viral DNA signal forms a random speckled pattern over the mitotic chromosomes and is not specific for any particular chromosome or chromosomal domain. In most experiments, 137 cells were treated with colchicine for 30 min prior to fixation to block cells in metaphase, but similar results were obtained without colchicine (data not shown). Cellular chromosomes were also spread on slides, and viral genomes were detected by FISH (Fig. 2g and h). In these spreads, the majority of the signal was also closely associated with the condensed chromosomes in a random pattern.
FIG. 2.
BPV DNA was detected by FISH in C127 cells (a and b) and 137 cells (c to h). In panels a, c, e, and g, cellular DNA was detected by the propidium iodide (PI) signal. In panels b, d, f, and h, the same fields of cells are shown with the FITC-labeled BPV DNA signal. In panels a to f, cells were grown on slides and treated with colchicine for 30 min before fixation. In panels g and h, the cells were treated with colchicine for 30 min and the chromosomes were spread on slides as described in Materials and Methods. Mitotic cells are indicated with white arrowheads in the propidium iodide-stained images.
The intracellular localization of the E2 proteins was determined by immunofluorescence. Monoclonal antibody B201 binds the E2-TA and E2-TR proteins, and B202 interacts with all three E2 species (Fig. 1B). Experiments with both antibodies showed that the E2 proteins in 137 cells were also localized to the mitotic chromatin (Fig. 3 and data not shown). The E2 proteins were localized in a random speckled pattern over the condensed chromosomes, as was seen for the viral genomes.
FIG. 3.
E2 proteins were detected in C127 cells (a, b, g, and h) and 137 cells (c to f, i, and j) by indirect immunofluorescence with the E2-specific B201 antibody. In panels a, c, e, g, and i, cellular DNA was detected by the propidium iodide (PI) signal. In panels b, d, f, h, and j, E2 protein was detected with an FITC-labeled secondary antibody in the same field of cells. Mitotic cells are indicated with white arrowheads in the propidium iodide-stained images.
Wild-type BPV-1 genomes are associated with mitotic chromatin.
To ensure that association of viral DNA with mitotic cellular chromosomes was not an artifact of the high-copy-number BPV-1 137 viral genome, FISH was performed on ID13 cells that contain episomal wild-type BPV genomes (24). As shown in Fig. 4a to d, BPV DNA was found to be associated with mitotic chromosomes in ID13 cells in a pattern similar to that of 137 cells. As stated above, the low level of expression prevents detection of the E2 proteins in these cells.
FIG. 4.
BPV DNA was detected by FISH in ID13 cells (a to d) and 209 cells (e to h). In panels a, c, e, and g, cellular DNA was detected by propidium iodide (PI) staining. In panels b, d, f, and h, FITC-labeled BPV DNA was detected in the same fields of cells. Cells were grown on slides and treated with colchicine for 30 min before fixation. Mitotic cells are indicated with white arrowheads in the propidium iodide-stained images.
The E2-TR protein is not required for the association of BPV-1 DNA with mitotic chromosomes.
C127 cells transformed with a BPV-1 genome that is unable to express the E2-TR protein also stably maintain a high copy number of extrachromosomal viral genomes (22). A cell line (209) was established by transformation of C127 cells with a cloned genome that is unable to express the E2-TR protein (p1472 [22]). The location of viral genomes in mitotic 209 cells was determined by FISH and was found to be very similar to that of the wild-type and E2 A301 viral DNA (Fig. 4e to h), indicating that the E2-TR protein is not required for the association of viral DNA with mitotic chromosomes.
SV40-derived DNA and SV40 T antigen are not associated with mitotic chromatin.
To determine whether the association of episomal DNA with mitotic chromatin is a common feature of any extrachromosomal DNA, we performed FISH analysis on COS-7 cells replicating either SV40 viral DNA or SV40-BPV E2 recombinant viral DNA (PAVAkzE2-TA). The recombinant virus consists of SV40 with the early region replaced with the E2-E5 region of BPV-1 (29, 40, 41). It can replicate, can express the E2-TA and E5 proteins, and is packaged in cells expressing large T antigen. COS-7 cells were infected with either SV40 or PAVAkzE2-TA, and the intracellular location of viral genomes was determined with a fluor-labeled SV40 DNA probe. Although the population of SV40-infected mitotic cells was low, in these cells SV40 DNA was found to be dispersed throughout the cell and did not concentrate on cellular chromosomes (Fig. 5, subpanels c and d). Mitotic cells infected with pPAVAkzE2-TA also contained SV40-derived DNA dispersed throughout the cell, despite the fact that this virus also expresses the BPV-1 E2-TA protein (Fig. 5, subpanels e and f). Therefore, it appears that the colocalization of BPV-1 DNA with mitotic chromatin is a specific phenomenon and does not occur with any extrachromosomal DNA, even in the presence of the E2 protein.
FIG. 5.
(a) SV40 and pPAVAkzE2-TA DNA was detected in COS-7 cells by FISH. Subpanels a and b show uninfected cells. Cells in subpanels c and d were infected with SV40, and those in subpanels e and f were infected with pPAVAkzE2-TA recombinant virus. In subpanels a, c, and e, cellular DNA was detected by propidium iodide (PI) staining. In subpanels b, d, and f, FITC-labeled BPV DNA was detected in the same fields of cells. Cells were treated with colchicine for 30 min, and the chromosomes were spread on slides as described in Materials and Methods. Mitotic cells are indicated with white arrowheads in the propidium iodide-stained images. (b) SV40 T antigen was detected in uninfected COS-7 cells by indirect immunofluorescence. In subpanels a, c, e, and g, cellular DNA was detected by propidium iodide (PI) staining. In subpanels d, f, and h, FITC-labeled secondary antibody was used to detect T antigen in the same fields of cells. As a negative control, cells in subpanel b were stained with an anti-E2 antibody. Cells were grown on slides and treated with colchicine for 30 min before fixation. Mitotic cells are indicated with white arrowheads in the propidium iodide-stained images.
The localization of SV40 T antigen was also examined by immunofluorescence in mitotic COS-7 cells. As shown in Fig. 5, subpanels c to h, T antigen was completely excluded from mitotic chromatin in dividing cells. In fact, most transcription factors are displaced from mitotic chromatin in dividing cells (27). Therefore, the specific association of E2 with mitotic chromatin, while not unique, is uncommon for transcription factors.
The E2-TA protein associates with mitotic chromatin in the absence of viral DNA.
To determine which of the three BPV-1 E2 proteins were associated with mitotic chromatin and to determine whether the association requires the viral genome, the E2-TA protein was overexpressed in two different cell systems. C127-derived cell lines that stably express the E2-TA protein under the control of a tetracycline-regulated promoter were established. As shown in Fig. 6c and d, the E2-TA transactivator protein was associated with mitotic chromatin in these cells. In addition, in CV-1 cells infected with the SV40-BPV E2 recombinant virus, PAVAkzE2-TA, the E2-TA transactivator protein was also observed as random speckles associated with the chromatin of mitotic cells (see Fig. 7). Therefore, the E2-TA transactivator is associated with mitotic chromatin, and this association is not mediated through and does not require the BPV-1 viral genome.
FIG. 6.
E2 proteins were detected in C127 and CHO-derived cell lines expressing the E2-TA, E2-TR, and E8/E2 proteins by indirect immunofluorescence with the B201 (a to f) and B202 (g to j) E2-specific antibodies. The cell lines shown in each panel are as follows: C127 (a and b), C127/E2-TA (c and d), C127/E2-TR (e and f), CHO (g and h), and CHO/E8/E2 (i and j). In panels a, c, e, g, and i, cellular DNA was detected by propidium iodide (PI) staining. In panels b, d, f, h, and j, the E2 proteins were detected by the FITC signal in the same fields of cells. Mitotic cells are indicated with white arrowheads in the propidium iodide-stained images.
FIG. 7.
E2 proteins were detected in COS-7 cells infected with pPAVAkzE2-TA viruses by indirect immunofluorescence with the B201 E2-specific antibody. (a and b) SV40-infected COS-7 cells used as a control; (c and d) cells infected with pPAVAkzE2-TA; (e and f) cells infected with pPAVAkzE2-TAΔ41–120; (g and h) cells infected with pPAVAkzE2-TAΔ51–120. In panels a, c, e, and g, cellular DNA was detected by propidium iodide (PI) staining. In panels b, d, f, and h, FITC-labeled E2 protein is detected in the same fields of cells. Mitotic cells are indicated with white arrowheads in the propidium iodide-stained images.
The E2-TR and E8/E2 proteins are not associated with mitotic chromatin.
To determine whether the shorter E2 repressor proteins, E2-TR and E8/E2, were associated with mitotic chromatin, these proteins were overexpressed in cells that do not contain BPV-1 genomes. C127 cells and CHO cell lines that express the E2-TR and E8/E2 repressor proteins, respectively, under the control of a tetracycline-regulated promoter were established. In each case, the viral repressor proteins, as detected by immunofluorescence, were excluded from mitotic chromatin and were dispersed throughout the cytoplasm of the cell (Fig. 6e, f, i, and j). Therefore, the E2-TR and E8/E2 proteins do not associate with mitotic chromatin, and it is unlikely that they are involved in the interaction of viral genomes with mitotic cellular chromosomes.
Deletions in the E2 transactivation domain abrogate the association with mitotic chromatin.
The fact that the full-length E2 protein interacts with mitotic chromatin and the shorter repressor species do not suggests that the N-terminal transactivation domain might be important for this interaction. To analyze this further, two proteins with in-frame deletions in the N-terminal domain (Δ41 to 120 and Δ51 to 120) were expressed in COS-7 cells from recombinant PAVA viruses. These E2 proteins have previously been shown to be localized in the nucleus (41). As shown in Fig. 7, wild-type E2-TA protein was closely associated with mitotic chromosomes in PAVA-infected COS-7 cells. In contrast, the two proteins with deletions in the transactivation domain were excluded from chromosomes. This suggests either that the deleted region contains important determinants for the interaction with mitotic chromosomes or that the deletions have disrupted the structure of the N-terminal domain, thereby indirectly destroying a region(s) important for chromosomal association.
The DNA binding property of the E2-TA protein is not required for association with mitotic chromatin.
It is possible that the E2-TA protein interacts with mitotic cellular chromosomes by binding to specific DNA binding sites in the cellular genome (in addition to the requirement for the N-terminal domain). To see if this was the case, the location of an E2 protein defective in DNA binding was determined. This protein has an arginine-to-lysine substitution at amino acid 344, which is one of the DNA contact residues in the recognition helix of the DNA binding domain (15). This E2 protein is unable to bind DNA (Fig. 8) but retains other properties such as dimerization (7). COS-7 cells and CV-1 cells were infected with a recombinant PAVA virus expressing the E2 K344 protein, and the protein was detected by immunofluorescence. As shown in Fig. 9, the K344 protein appeared to be associated with mitotic chromatin in a pattern indistinguishable from that of the wild-type protein. Therefore, the association with mitotic chromatin does not require the DNA binding property of the E2 transactivation protein.
FIG. 8.
E2 K344 is defective in DNA binding. The DNA binding domains (E2 residues 290 to 410) of wild-type and K344 E2 proteins were synthesized in vitro and tested for DNA binding in an electrophoretic mobility shift assay. The amount of oligonucleotide probe bound to the E2 proteins is plotted against the concentration of probe in the reaction mixtures. Background amounts of probe bound by the control lysate have been subtracted from the values shown. Values for wild-type E2 are represented by circles, and those for E2 K344 are represented by squares.
FIG. 9.
E2 proteins were detected in CV-1 cells infected with pPAVAkzE2-TA and pPAVAkzE2-TA K344 by immunofluorescence with the B201 E2-specific antibody. (a and b) Uninfected CV-1 cells; (c to f) pPAVAkzE2-TA-infected CV-1 cells; (g to j) pPAVAkzE2-TA K344-infected cells. In panels a, c, e, g, and i, cellular DNA was detected by propidium iodide (PI) staining. In panels b, d, f, h, and j, FITC-labeled E2 protein was detected in the same fields of cells. Mitotic cells are indicated with white arrowheads in the propidium iodide-stained images.
DISCUSSION
In this study, we have shown that both the BPV-1 E2 transactivator protein and the BPV-1 viral genomes are closely associated with mitotic chromatin in dividing cells. It is tempting to speculate that this association is important for segregation of papillomavirus genomes in dividing cells. Rodent cells transformed by BPV-1 maintain approximately 50 to 200 copies of the viral genome indefinitely as extrachromosomal nuclear plasmids (24). Cell lines derived from cervical carcinomas can also maintain human papillomavirus genomes as extrachromosomal elements (3). The E1 and E2 proteins are required for transient replication of plasmids containing the viral origin of replication; however, stable maintenance of origin-containing plasmids also requires regions from the LCR that contain multiple high-affinity E2 DNA binding sites (34). This region has been designated the MME and can be replaced by 10 tandem copies of E2 DNA binding sites, suggesting that the E2 protein and the E2 DNA binding sites are important for genome segregation. The findings presented in this study suggest that the E2 protein may facilitate genome segregation by interacting with condensed mitotic chromatin and support a model in which viral genomes are attached to mitotic chromatin indirectly via the E2 protein and E2 DNA binding sites. This interaction would ensure that approximately equal numbers of viral genomes are segregated to daughter cells. Viral genomes that replicate as extrachromosomal plasmids may also require a mechanism to ensure that they are not lost from the nucleus during cell division. Association with cellular chromosomes would ensure that viral genomes are enclosed in the nuclear membrane during telophase. The genomes may also interact with some cellular component that ensures that they are in a transcriptionally active region of the nucleus as the cells move into the G1 stage of the cell cycle.
Although the overall viral copy number in a population of BPV-1-transformed cells remains relatively constant, several studies have shown that individual cells contain a wide range of copy numbers (35–37). Also noted in this study was that BPV-transformed cell lines hybridized with FITC-labeled viral DNA showed varying fluorescence in individual cells. Stewart et al. also demonstrated that there was significant randomization in replication and/or partitioning (43). This suggests that segregation does not occur by a very precise mechanism and is consistent with the model in which the E2 proteins and viral genomes randomly associate with mitotic chromatin as passenger molecules. This model would also predict that the viral copy number depends on the levels of the E2-TA protein. Notably, the levels of E2 proteins in 137 cells (Fig. 3) also vary greatly, and it would be informative to determine whether the amount of E2 protein correlated with the viral copy number in individual cells.
A similar phenomenon has been observed for EBV. EBV infects and immortalizes B lymphocytes, and the viral genome is maintained indefinitely as an extrachromosomal element. The EBNA-1 protein of EBV is both a transcriptional transactivator and a replication protein, and it is the only viral protein required for replication and maintenance of plasmids containing the oriP origin of replication (which contains a number of EBNA DNA binding sites) (49). The EBNA-1 protein and EBV genomes have also been shown to be randomly associated with mitotic chromatin (12, 14), and it has been suggested that these properties might be important for the genome segregation and nuclear retention function of EBNA-1. The EBNA-1 protein also promotes prolonged nuclear retention of plasmids containing EBNA-1 DNA binding sites but no origin of replication (21), and this function has been exploited in the design of extrachromosomal vectors for gene therapy (6). It seems that the EBNA-1 and E2 proteins have some common roles in the life cycles of their respective viruses (13). Notably, both proteins have dimeric DNA binding domains with almost identical structures despite no amino acid homology (5). Studies are in progress to determine whether the E2 protein has a similar nuclear retention function that is separate from its role in DNA replication. If so, this system might be more suitable for inclusion in extrachromosomal gene therapy vectors, as it has been shown that the EBNA-1 protein can cause lymphomas in transgenic mice expressing this protein in B cells (46).
In this study, it was shown that the E2-TA protein could interact with mitotic chromatin in the absence of viral genomes. Conversely, the E2-TR and E8/E2 proteins were found to be dispersed throughout the cell during mitosis and were excluded from mitotic chromatin. This indicates that the DNA binding domain of the E2 protein is not sufficient for the interaction with mitotic chromosomes and suggests that the interaction is not mediated by binding to cellular DNA sequences. This is also supported by the finding that a DNA-binding-defective E2-TA protein retains the ability to interact with mitotic chromatin. Furthermore, deletions within the N-terminal domain abrogate the ability of E2 to interact with mitotic chromosomes. These findings indicate that the N-terminal transactivation domain of E2-TA is necessary for the interaction, and studies are in progress to determine whether this domain is sufficient.
It has been argued that cells containing high copy numbers of extrachromosomal elements do not require a specific mechanism for plasmid segregation because approximately equal numbers of genomes should be passively segregated to daughter cells during mitosis. Although many cell lines do contain high copy numbers of papillomaviral genomes, this is probably not the case in infected epithelial lesions. Papillomaviruses infect and stimulate proliferation of basal keratinocytes, which provide a reservoir of infected cells that can differentiate and amplify viral DNA and produce virion particles. The viral genome copy number in the basal cells of a papilloma appears to be quite low, and therefore, a specific mechanism for genome segregation may be important to ensure that viral genomes are not lost in the proliferating basal cells of a papilloma. BPV-1 causes fibropapillomas, which have a large dermal fibroma in addition to the epithelial lesion. Efficient segregation of the viral genomes may also be important in ensuring that all cells in the fibroma contain BPV-1 DNA.
As yet, it is not known what component of mitotic chromatin is important for interaction of the E2 protein with mitotic chromatin. One possibility is that E2 is interacting with some constituent of the chromosomal scaffold or chromosomal periphery. The chromosomal periphery is a region around the condensed chromatids that contain many proteins, some of which form a network of fibrils and granules (16). Several components of the nuclear matrix are found in the perichromosomal region, as well as a number of passenger proteins from the nucleus and nucleoli. The E2-TA protein (but not the E2-TR or E8/E2 protein) has been shown to be associated with the nuclear matrix (18), and it will be interesting to determine whether the same interactions are important for the association with mitotic chromosomes. Nuclear matrix attachment sites have also been identified in BPV-1 (1, 2), and it is possible that these sites are also important for interaction of the genomes with mitotic chromatin instead of, or in addition to, E2 DNA binding sites. Future studies will determine whether E2 binding sites and nuclear matrix attachment sites are required for the association with mitotic chromosomes.
ACKNOWLEDGMENTS
We thank Bernard Moss and Thomas Kristie for their comments on the manuscript, Elliot Androphy for the E2 monoclonal antibodies, and Maritza Blanco for technical assistance in generating the CHO E2 cell lines.
REFERENCES
- 1.Adom J N, Gouilleux F, Richard-Foy H. Interaction with the nuclear matrix of a chimeric construct containing a replication origin and a transcription unit. Biochim Biophys Acta. 1992;1171:187–197. doi: 10.1016/0167-4781(92)90119-k. [DOI] [PubMed] [Google Scholar]
- 2.Adom J N, Richard-Foy H. A region immediately adjacent to the origin of replication of bovine papilloma virus type 1 interacts in vitro with the nuclear matrix. Biochem Biophys Res Commun. 1991;176:479–485. doi: 10.1016/0006-291x(91)90949-8. [DOI] [PubMed] [Google Scholar]
- 3.Bedell M A, Hudson J B, Golub T R, Turyk M E, Hosken M, Wilbanks G D, Laimins L A. Amplification of human papillomavirus genomes in vitro is dependent on epithelial differentiation. J Virol. 1991;65:2254–2260. doi: 10.1128/jvi.65.5.2254-2260.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Blitz I L, Laimins L A. The 68-kilodalton E1 protein of bovine papillomavirus is a DNA binding phosphoprotein which associates with the E2 transcriptional activator in vitro. J Virol. 1991;65:649–656. doi: 10.1128/jvi.65.2.649-656.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bochkarev A, Barwell J A, Pfuetzner R A, Furey W J, Edwards A M, Frappier L. Crystal structure of the DNA-binding domain of the Epstein-Barr virus origin-binding protein EBNA 1. Cell. 1995;83:39–46. doi: 10.1016/0092-8674(95)90232-5. [DOI] [PubMed] [Google Scholar]
- 6.Calos M P. The potential of extrachromosomal replicating vectors for gene therapy. Trends Genet. 1996;12:463–466. doi: 10.1016/0168-9525(96)40049-x. [DOI] [PubMed] [Google Scholar]
- 7.Carruth, M., and A. A. McBride. 1997. Unpublished data.
- 8.Choe J, Vaillancourt P, Stenlund A, Botchan M. Bovine papillomavirus type 1 encodes two forms of a transcriptional repressor: structural and functional analysis of new viral cDNAs. J Virol. 1989;63:1743–1755. doi: 10.1128/jvi.63.4.1743-1755.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Dvoretzky I, Shober R, Chattopadhyay S K, Lowy D R. A quantitative in vitro focus assay for bovine papilloma virus. Virology. 1980;103:369–375. doi: 10.1016/0042-6822(80)90195-6. [DOI] [PubMed] [Google Scholar]
- 10.Gerard R D, Gluzman Y. New host cell system for regulated simian virus 40 DNA replication. Mol Cell Biol. 1985;5:3231–3240. doi: 10.1128/mcb.5.11.3231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gilbert D M, Cohen S N. Bovine papilloma virus plasmids replicate randomly in mouse fibroblasts throughout S phase of the cell cycle. Cell. 1987;50:59–68. doi: 10.1016/0092-8674(87)90662-3. [DOI] [PubMed] [Google Scholar]
- 12.Grogan E A, Summers W P, Dowling S, Shedd D, Gradoville L, Miller G. Two Epstein-Barr viral nuclear neoantigens distinguished by gene transfer, serology, and chromosome binding. Proc Natl Acad Sci USA. 1983;80:7650–7653. doi: 10.1073/pnas.80.24.7650. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Grossman S R, Laimins L A. EBNA1 and E2: a new paradigm for origin-binding proteins? Trends Microbiol. 1996;4:87–89. doi: 10.1016/0966-842X(96)81520-4. [DOI] [PubMed] [Google Scholar]
- 14.Harris A, Young B D, Griffin B E. Random association of Epstein-Barr virus genomes with host cell metaphase chromosomes in Burkitt’s lymphoma-derived cell lines. J Virol. 1985;56:328–332. doi: 10.1128/jvi.56.1.328-332.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hegde R S, Grossman S R, Laimins L A, Sigler P B. Crystal structure at 1.7 A of the bovine papillomavirus-1 E2 DNA-binding domain bound to its DNA target. Nature. 1992;359:505–512. doi: 10.1038/359505a0. [DOI] [PubMed] [Google Scholar]
- 16.Hernandez-Verdun D, Gautier T. The chromosome periphery during mitosis. Bioessays. 1994;16:179–185. doi: 10.1002/bies.950160308. [DOI] [PubMed] [Google Scholar]
- 17.Howley P M. Papillomavirinae: the viruses and their replication. In: Fields B N, Knipe D M, Howley P M, editors. Virology. Philadelphia, Pa: Lippincott-Raven; 1995. pp. 2045–2076. [Google Scholar]
- 18.Hubbert N L, Schiller J T, Lowy D R, Androphy E J. Bovine papilloma virus-transformed cells contain multiple E2 proteins. Proc Natl Acad Sci USA. 1988;85:5864–5868. doi: 10.1073/pnas.85.16.5864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hughes F J, Romanos M A. E1 protein of human papillomavirus is a DNA helicase/ATPase. Nucleic Acids Res. 1993;21:5817–5823. doi: 10.1093/nar/21.25.5817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Jenkins O, Earnshaw D, Sarginson G, Del Vecchio A, Tsai J, Kallender H, Amegadzie B, Browne M. Characterization of the helicase and ATPase activity of human papillomavirus type 6b E1 protein. J Gen Virol. 1996;77:1805–1809. doi: 10.1099/0022-1317-77-8-1805. [DOI] [PubMed] [Google Scholar]
- 21.Krysan P J, Haase S B, Calos M P. Isolation of human sequences that replicate autonomously in human cells. Mol Cell Biol. 1989;9:1026–1033. doi: 10.1128/mcb.9.3.1026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lambert P F, Monk B C, Howley P M. Phenotypic analysis of bovine papillomavirus type 1 E2 repressor mutants. J Virol. 1990;64:950–956. doi: 10.1128/jvi.64.2.950-956.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lambert P F, Spalholz B A, Howley P M. A transcriptional repressor encoded by BPV-1 shares a common carboxy-terminal domain with the E2 transactivator. Cell. 1987;50:69–78. doi: 10.1016/0092-8674(87)90663-5. [DOI] [PubMed] [Google Scholar]
- 24.Law M F, Lowy D R, Dvoretzky I, Howley P M. Mouse cells transformed by bovine papillomavirus contain only extrachromosomal viral DNA sequences. Proc Natl Acad Sci USA. 1981;78:2727–2731. doi: 10.1073/pnas.78.5.2727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Li R, Botchan M R. The acidic transcriptional activation domains of VP16 and p53 bind the cellular replication protein A and stimulate in vitro BPV-1 DNA replication. Cell. 1993;73:1207–1221. doi: 10.1016/0092-8674(93)90649-b. [DOI] [PubMed] [Google Scholar]
- 26.Li R, Botchan M R. Acidic transcription factors alleviate nucleosome-mediated repression of DNA replication of bovine papillomavirus type 1. Proc Natl Acad Sci USA. 1994;91:7051–7055. doi: 10.1073/pnas.91.15.7051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Martinez-Balbas M A, Dey A, Rabindran S K, Ozato K, Wu C. Displacement of sequence-specific transcription factors from mitotic chromatin. Cell. 1995;83:29–38. doi: 10.1016/0092-8674(95)90231-7. [DOI] [PubMed] [Google Scholar]
- 28.McBride, A. A. 1997. Unpublished observations.
- 29.McBride A A, Bolen J B, Howley P M. Phosphorylation sites of the E2 transcriptional regulatory proteins of bovine papillomavirus type 1. J Virol. 1989;63:5076–5085. doi: 10.1128/jvi.63.12.5076-5085.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.McBride A A, Byrne J C, Howley P M. E2 polypeptides encoded by bovine papillomavirus type 1 form dimers through the common carboxyl-terminal domain: transactivation is mediated by the conserved amino-terminal domain. Proc Natl Acad Sci USA. 1989;86:510–514. doi: 10.1073/pnas.86.2.510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.McBride A A, Howley P M. Bovine papillomavirus with a mutation in the E2 serine 301 phosphorylation site replicates at a high copy number. J Virol. 1991;65:6528–6534. doi: 10.1128/jvi.65.12.6528-6534.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.McBride A A, Klausner R D, Howley P M. Conserved cysteine residue in the DNA-binding domain of the bovine papillomavirus type 1 E2 protein confers redox regulation of the DNA-binding activity in vitro. Proc Natl Acad Sci USA. 1992;89:7531–7535. doi: 10.1073/pnas.89.16.7531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Mohr I J, Clark R, Sun S, Androphy E J, MacPherson P, Botchan M R. Targeting the E1 replication protein to the papillomavirus origin of replication by complex formation with the E2 transactivator. Science. 1990;250:1694–1699. doi: 10.1126/science.2176744. [DOI] [PubMed] [Google Scholar]
- 34.Piirsoo M, Ustav E, Mandel T, Stenlund A, Ustav M. Cis and trans requirements for stable episomal maintenance of the BPV-1 replicator. EMBO J. 1996;15:1–11. [PMC free article] [PubMed] [Google Scholar]
- 35.Ravnan J-B, Cohen S N. Transformed mouse cell lines that consist predominantly of cells maintaining bovine papillomavirus at high copy number. Virology. 1997;213:526–534. doi: 10.1006/viro.1995.0025. [DOI] [PubMed] [Google Scholar]
- 36.Ravnan J-B, Gilbert G M, Ten Hagen K G, Cohen S N. Random-choice replication of extrachromosomal bovine papillomavirus (BPV) molecules in heterogeneous clonally derived BPV-infected cell lines. J Virol. 1992;66:6946–6952. doi: 10.1128/jvi.66.12.6946-6952.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Roberts J M, Weintraub H. Cis-acting negative control of DNA replication in eukaryotic cells. Cell. 1988;52:397–404. doi: 10.1016/s0092-8674(88)80032-1. [DOI] [PubMed] [Google Scholar]
- 38.Sedman J, Stenlund A. Co-operative interaction between the initiator E1 and the transcriptional activator E2 is required for replicator specific DNA replication of bovine papillomavirus in vivo and in vitro. EMBO J. 1995;14:6218–6228. doi: 10.1002/j.1460-2075.1995.tb00312.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Seo Y-S, Muller F, Lusky M, Gibbs E, Kim H-Y, Phillips B, Hurwitz J. Bovine papilloma virus (BPV)-encoded E2 protein enhances binding of E1 protein to the BPV replication origin. Proc Natl Acad Sci USA. 1993;90:2865–2869. doi: 10.1073/pnas.90.7.2865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Settleman J, DiMaio D. Efficient transactivation and morphologic transformation by bovine papillomavirus genes expressed from a bovine papillomavirus/simian virus 40 recombinant virus. Proc Natl Acad Sci USA. 1988;85:9007–9011. doi: 10.1073/pnas.85.23.9007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Skiadopoulos M H, McBride A A. The bovine papillomavirus type 1 repressor proteins use different nuclear localization signals. J Virol. 1996;70:1117–1124. doi: 10.1128/jvi.70.2.1117-1124.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Spalholz B A, McBride A A, Sarafi T, Quintero J. Binding of bovine papillomavirus E1 to the origin is not sufficient for DNA replication. Virology. 1993;193:201–212. doi: 10.1006/viro.1993.1116. [DOI] [PubMed] [Google Scholar]
- 43.Stewart A-C, Jareborg N, Simonsson M, Alderborn A, Burnett S. Segregation properties of bovine papillomaviral plasmid DNA. J Mol Biol. 1994;236:480–490. doi: 10.1006/jmbi.1994.1159. [DOI] [PubMed] [Google Scholar]
- 44.Ustav M, Stenlund A. Transient replication of BPV-1 requires two viral polypeptides encoded by the E1 and E2 open reading frames. EMBO J. 1991;10:449–457. doi: 10.1002/j.1460-2075.1991.tb07967.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Ustav M, Ustav E, Szymanski P, Stenlund A. Identification of the origin of replication of bovine papillomavirus and characterization of the viral origin recognition factor E1. EMBO J. 1991;10:4321–4329. doi: 10.1002/j.1460-2075.1991.tb05010.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Wilson J B, Bell J L, Levine A J. Expression of Epstein-Barr virus nuclear antigen-1 induces B cell neoplasia in transgenic mice. EMBO J. 1996;15:3117–3126. [PMC free article] [PubMed] [Google Scholar]
- 47.Winokur P L, McBride A A. The transactivation and DNA binding domains of the BPV-1 E2 protein have different roles in co-operative origin binding with the E1 protein. Virology. 1996;221:44–53. doi: 10.1006/viro.1996.0351. [DOI] [PubMed] [Google Scholar]
- 48.Yang L, Mohr I, Fouts E, Lim D A, Nohaile M, Botchan M. The E1 protein of bovine papillomavirus 1 is an ATP-dependent DNA helicase. Proc Natl Acad Sci USA. 1993;90:5086–5090. doi: 10.1073/pnas.90.11.5086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Yates J L, Warren N, Sugden B. Stable replication of plasmids derived from Epstein-Barr virus in various mammalian cells. Nature. 1985;313:812–815. doi: 10.1038/313812a0. [DOI] [PubMed] [Google Scholar]