The Epstein-Barr virus deubiquitinating enzyme BPLF1 regulates the activity of topoisomerase II during productive infection

Abstract

Topoisomerases are essential for the replication of herpesviruses but the mechanisms by which the viruses hijack the cellular enzymes are largely unknown. We found that topoisomerase-II (TOP2) is a substrate of the Epstein-Barr virus (EBV) ubiquitin deconjugase BPLF1. BPLF1 co-immunoprecipitated and deubiquitinated TOP2, and stabilized SUMOylated TOP2 trapped in cleavage complexes (TOP2ccs), which halted the DNA damage response to TOP2-induced double strand DNA breaks and promoted cell survival. Induction of the productive virus cycle in epithelial and lymphoid cell lines carrying recombinant EBV encoding the active enzyme was accompanied by TOP2 deubiquitination, accumulation of TOP2ccs and resistance to Etoposide toxicity. The protective effect of BPLF1 was dependent on the expression of tyrosyl-DNA phosphodiesterase 2 (TDP2) that releases DNA-trapped TOP2 and promotes error-free DNA repair. These findings highlight a previously unrecognized function of BPLF1 in supporting a non-proteolytic pathway for TOP2ccs debulking that favors cell survival and virus production.

Fig 1. BPLF1 selectively binds to TOP2 and inhibits the degradation of TOP2 in cells treated with topoisomerase poisons.

HEK-293T cell expressing inducible FLAG-BPLF1 or FLAG-BPLF1^C61A were seeded into 6 well plates and treated with 1.5 μg/ml Dox for 24 h. After treatment for 3 h with 5 μM of the TOP1 poison Camptothecin (Cpt) or 6 h with 40 μM of the TOP2 poison Etoposide (Eto) with or without the addition of 10 μM MG132, protein expression was analyzed in western blots probed with the indicated antibodies. GAPDH was used as the loading control. (A) Representative western blots illustrating the expression of TOP1 in control and Cpt treated cells. The proteasome-dependent degradation of TOP1 induced by the treatment was not affected by the expression of BPLF1 or BPLF1^C61A in Dox treated cells. (B) Representative western blots illustrating the expression of TOP2α and TOP2β in Etoposide treated cells. Expression of BPLF1 protected TOP2α and TOP2β from Etoposide-induced proteasomal degradation while BPLF1^C61A had no appreciable effect. (C) The intensity of the TOP1, TOP2α and TOP2β specific bands in 5 (TOP1) or 6 (TOP2α and TOP2β) independent experiments was quantified using the ImageJ software. The data are presented as intensity of the bands in Cpt/Eto treated samples relative to untreated control after normalization to the GAPDH loading control. Statistical analysis was performed using Student’s t-test. **P≤ 0.01; ns, not significant. (D) HEK293T cells transfected with FLAG-BPLF1, FLAG-BPLF1^C61A, or FLAG-empty vector were treated with 40 μM Etoposide for 30 min and cell lysates were either immunoprecipitated with anti-FLAG conjugated agarose beads or incubated for 3 h with anti-TOP2α or TOP2β antibodies followed by the capture of immunocomplexes with protein-G coated beads. Catalytically active and inactive BPLF1 co-immunoprecipitate with both TOP2α and TOP2β in untreated and Etoposide treated cells (upper panels). Conversely, TOP2α (middle panels) and TOP2β (lower panels) interact with both catalytically active and inactive BPLF1. Representative western blots from one of two independent experiments where all conditions were tested in parallel are shown.

Fig 2. BPLF1 deubiquitinates TOP2 and stabilizes TOP2ccs.

(A) HEK293T cells were transiently transfected with plasmids expressing FLAG-BPLF1, FLAG-BPLF1-C61A, or the FLAG empty vector, and aliquots were treated with 40 μM Etoposide for 30 min. TOP2α and TOP2β were immunoprecipitated from cell lysates prepared under denaturing conditions in the presence of DUB inhibitors and western blots were probed with antibodies to TOP2α, TOP2β and ubiquitin. The expression of catalytically active BPLF1 inhibits the ubiquitination of TOP2α and TOP2β induced by Etoposide treatment. Western blots from one representative experiment out of three are shown in the figure. (B) HEK-rtTA-BPLF1 cells were treated with 1.5 μg/ml Dox for 24 h followed by treatment with 80 μM Etoposide for the indicated time with or without the addition of 10 μM MG132. RADAR assays were performed as described in Materials and Methods and TOP2 trapped in 10 μg DNA was detected in western blots using antibodies to TOP2α or TOP2β. Trapped TOP2 appears as a major band of the expected size and a smear of higher molecular weight species. The intensity of the trapped TOP2α and TOP2β smears decreased over time in control untreated cells due to proteasomal degradation, while the decrease was significantly reduced upon expression of BPLF1 in Dox treated cells. Western blots from one representative experiment out of two are shown in the figure. (C) The intensity of the TOP2 smears was quantified using the ImageJ software. Clearance was calculated as 1- (intensity of the smears after treatment for 4 h/intensity of the smears after treatment for 30 min) x100. Treatment with MG132 reduced the clearance of TOP2ccs in BPLF1 negative cells and a similar reduction was achieved by expression of BPLF1 in Dox treated cells. The mean ± SD of two independent experiments is shown in the figure. Statistical analysis was performed using Student’s t-test. *P≤ 0.05.

Fig 3. BPLF1 selectively inhibits the detection of TOP2-induced DNA damage.

HEK-rtTA-BPLF1/BPLF1^C61A cells grown on cover-slides were treated with 1.5 μg/ml Dox for 24 h to induce the expression of BPLF1 followed by treatment for 1 h with 40 μM Etoposide or 0.5 μg/ml of the radiomimetic Neocarzinostatin (NCS) before staining with the indicated antibodies. (A) The cells were co-stained with antibodies against FLAG (red) and antibodies to γH2AX or 53BP1 (green) and the nuclei were stained with DAPI (blue). The expression of catalytically active BPLF1 was associated with a significant decrease of Etoposide induced nuclear γH2AX fluorescence and decreased formation of 53BP1 foci. BPLF1^C61A had no effect. Neither the catalytically active nor the inactive BPLF1 affected the induction of γH2AX in cells treated with NCS. Representative micrographs from one of two experiments where all conditions were tested in parallel are shown. Scale bar = 10 μm. (B) Quantification of γH2AX fluorescence intensity and number of 53BP1 foci in BPLF1/BPLF1^C61A positive and negative cells from the same images. The Mean ± SD of fluorescence intensity or number of dots in at least 50 BPLF1-positive and 50 BPLF1-negative cells recorded in each condition is shown. Statistical analysis was performed using Student’s t-test. ***P ≤0.001; ns, not significant.

Fig 4. BPLF1 promotes the accumulation of SUMOylated TOP2ccs and cell viability following Etoposide treatment.

(A) HEK-rtTA-BPLF1 cells were cultured for 24 h in the presence or absence of 1.5 μg/ml Dox and then treated with 80 μM Etoposide for 30 min followed by detection of DNA trapped TOP2 by RADAR assay. Western blots of proteins bound to 10 μg DNA were probed with antibodies to TOP2, ubiquitin and SUMO2/3. The expression of BPLF1 was associated with strongly decreased ubiquitination of the TOP2ccs while SUMOylation was only marginally affected. (B) The intensity of the ubiquitin/SUMO2/3 smears and TOP2 specific bands was quantified by densitometry using the ImageJ software. Relative intensity was calculated as intensity of the smears in Dox-treated versus untreated cells after normalization to the total amount of DNA-tapped TOP2. Mean ± SE of two independent experiments. Statistical analysis was performed using Student’s t-test. *P≤0.05. (C) HEK-rtTA-BPLF1/BPLF^C61A cells were cultured for 24 h in the presence or absence of 1.5 μg/ml Dox and then treated overnight with the indicated concentration of Etoposide before assessing cell viability by MTT assays. The expression of catalytically active BPLF1 decreased the toxic effect of Etoposide over a wide range of concentrations while BPLF1^C61A had no appreciable effect. The mean ± SD of two independent experiments is shown. Statistical analysis was performed using Student’s t-test. **P≤0.01; ns = non-significant.

Fig 5. BPLF1 regulates the activity of TOP2 in productively infected HEK293-EBV cells.

The productive virus cycle was induced by treatment with 1.5 μg/ml Dox in HEK293-EBV cells carrying recombinant EBV encoding wild type or catalytic mutant BPLF1 and a tetracycline regulated BZLF1 transactivator. The expression of TOP2α and TOP2β was assessed by western blot and the intensity of the specific bands was quantified using the ImageJ software. (A) Representative western blots illustrating the small upregulation of TOP2α and TOP2β in cells expressing catalytically active BPLF1. (B) Densitometry quantification of the specific bands was performed using the ImageJ software and the values were normalized to the GAPDH loading control. The mean ± SD of three independent experiments are shown. (C) Representative RADAR assay illustrating the accumulation of TOP2α and TOP2β cleavage complexes in cells expressing catalytically active BPLF1. Western blots from one representative experiment out of three are shown. (D) The intensity of the DNA trapped TOP2α and TOP2β species was quantified using the ImageJ software. The mean ± SD fold increase in induced versus non induced cells recorded in three independent experiments is shown. Statistical analysis was performed using Student’s t-test. *P≤0.05; **P≤0.01. (E) Representative western blot illustrating the expression of the DDR marker γH2AX in induced HEK293-EBV cells expressing catalytic active or inactive BPLF1. (F) The intensity of γH2AX bands were quantified by densitometry in three independent experiments. The fold increase in induced versus control cells was calculated after normalization to the GAPDH loading control. Statistical analysis was performed using Student’s t-test. ***P≤0.001.

Fig 6. BPLF1 regulates the expression and activity of TOP2 in productively infected LCLs.

The productive virus cycle was induced by treatment with 1.5 μg/ml Dox in LCL cells carrying recombinant EBV encoding wild type or catalytic mutant BPLF1 and a tetracycline regulated BZLF1 transactivator. Induction of the productive cycle was associated with a highly reproducible downregulation of TOP2α while TOP2β was either unchanged or slightly increased. The effect was stronger in cells expressing wild type BPLF1. (A) Representative western blots illustrating the expression of TOP2α and TOP2β in control and induced cells. (B) The intensity of the specific bands was quantified using the ImageJ in three to five independent experiments and fold change in induced versus control cells was calculated after normalization to the GAPDH loading control. (C) The formation of TOP2βccs was investigated by RADAR assays in untreated and induced LCLs. Representative western blot illustrating the significant increase of TOP2ccs upon induction of the productive virus cycle in LCL cells expressing catalytically active BPLF1. BPLF1^C61A had no appreciable effect. One representative western blot is shown. (D) Quantification of the intensity of the TOP2β smears in three independent experiments. Fold increase was calculated as the ratio between the smear intensity in control versus induced cells. *P≤0.05. (E) TOP2β was immunoprecipitated from total cell lysates of control and induced LCLs and western blots were probed with antibodies to TOP2β, ubiquitin and SUMO2/3. Western blots illustrating the increase of SUMOylated TOP2β in productively infected cells and selective decrease of ubiquitinated TOP2β in cells expressing catalytically active BPLF1. One representative experiment out of three is shown. (F) The intensity of the bands corresponding to immunoprecipitated TOP2β and ubiquitinated or SUMOylated species was quantified using the ImageJ software and the SUMO/Ub ratio was calculated after normalization to immunoprecipitated TOP2β. The mean ± SE of three independent experiments is shown. *P<0.05. (G) Representative western blot illustrating the expression of the DDR marker γH2AX in control and induced LCL-EBV-BPLF1/BPLF1^C61A cells. (H) The intensity of γH2AX bands were quantified by densitometry in eight independent experiments. The fold increase in induced versus control cells was calculated after normalization to the GAPDH loading control and to the level of induction as assessed by the intensity of the BMRF1 specific band. Statistical analysis was performed using Student’s t-test. *P≤0.05. (I) The productive cycle was induced in LCL-EBV-BPLF1/BPLF1^C61A by culture for 72 h in the presence 1.5 μg/ml Dox. After washing and counting, 5x10⁴ live cells were seeded in triplicate wells of 96 well plates and treated overnight with the indicated concentration of Etoposide before assessing cell viability by MTT assays. Catalytically active BPLF1 enhanced cell viability over a wide range of Etoposide concentration while BPLF1^C61A had no appreciable effect. The mean ± SE of cell viability in three independent experiments is shown.**P≤0.01.

Fig 7. TDP2 is required for resistance to Etoposide toxicity and virus production in induced LCLs.

The productive virus cycle was induced in TK6 and TK6-TDP2^-/- LCLs expressing a tetracycline regulated BZLF1 transactivator by treatment with 1.5 μg/ml Dox. (A) TK6/TK6-TDP2^-/- LCLs were harvested 72 h after induction. Five x10⁴ live cells were seeded in triplicate wells of 96 well plates and treated overnight with the indicated concentration Etoposide before assessing cell viability by MTT assays. The rescue of Etoposide toxicity observed in induced TK6 was abolished in cells lacking TDP2. The mean ± SE of cell viability in three independent experiments is shown.**P≤0.01. (B) The amount of EBV-DNA in cell pellets and DNAse treated culture supernatants was quantified by qPCR 72 h after induction. The mean ± SE fold induction relative to untreated controls in three (supernatants) or four (cells) independent experiments is shown. *P≤0.05, **P≤0.01. (C) Model of TOP2 regulation by BPLF1. TOP2 (violet) trapped in TOP2ccs (yellow) is targeted for proteasomal degradation via SUMOylation (light blue) and ubiquitination (red) mediated by the SUMO ligase ZNF451, the SUMO-targeting ubiquitin ligase RNF4 and other cellular ubiquitin ligases, leading to the display of partially digested 5’-phosphotyrosyl-DNA adducts. Processing by the TDP2 resolvase generates protein-free DSBs that trigger the DDR. Imprecise repair leads to apoptosis and genomic instability. Ectopically expressed BPLF1 is recruited to DNA-trapped TOP2 and inhibits proteasomal degradation, which prevents the activation of the DDR. In the absence of ubiquitination, SUMOylation may alter the conformation of the TOP2 dimer allowing direct access of TDP2 to the 5’-phosphotyrosyl-DNA bonds, which promotes error-free repair. During productive EBV infection, the concomitant expression of BPLF1 and viral miRNA mediated downregulation of RNF4 favors the accumulation of SUMOylated TOP2β and the activation of non-proteolytic pathways for TOP2ccs debulking that are dependent on the resolution of TOP2-DNA adducts by TDP2. This promotes the error free repair of TOP2-induced DSBs and enhances cells survival and virus production.