Hsp90 inhibitor 17-DMAG decreases expression of conserved herpesvirus protein kinases and reduces virus production in Epstein-Barr virus-infected cells

Abstract

All eight human herpesviruses have a conserved herpesvirus protein kinase (CHPK) that is important for the lytic phase of the viral life cycle. In this study, we show that heat shock protein 90 (Hsp90) interacts directly with each of the eight CHPKs, and we demonstrate that an Hsp90 inhibitor drug, 17-dimethylaminoethylamino-17-demethoxygeldanamycin (17-DMAG), decreases expression of all eight CHPKs in transfected HeLa cells. 17-DMAG also decreases expression the of the endogenous Epstein-Barr virus protein kinase (EBV PK, encoded by the BGLF4 gene) in lytically infected EBV-positive cells and inhibits phosphorylation of several different known EBV PK target proteins. Furthermore, 17-DMAG treatment abrogates expression of the human cytomegalovirus (HCMV) kinase UL97 in HCMV-infected human fibroblasts. Importantly, 17-DMAG treatment decreased the EBV titer approximately 100-fold in lytically infected AGS-Akata cells without causing significant cellular toxicity during the same time frame. Increased EBV PK expression in 17-DMAG-treated AGS-Akata cells did not restore EBV titers, suggesting that 17-DMAG simultaneously targets multiple viral and/or cellular proteins required for efficient viral replication. These results suggest that Hsp90 inhibitors, including 17-DMAG, may be a promising group of drugs that could have profound antiviral effects on herpesviruses.

Fig 1

EBV PK associates with Hsp90 and Cdc37. (A) 293T cells were transfected with a vector expressing wild-type EBV PK fused in frame to a HaloTag or a kinase-dead mutant (PK-Dead), and the EBV PK proteins were then purified from the cell lysates (and cleaved from the tag). A Coomassie-stained gel showing the total cell lysate and purified EBV PK proteins is shown. (B) In vitro kinase assays using [³²P]ATP were performed by using GST or GST-BMRF1 (C terminus) fusion protein and purified EBV PK proteins. (Top) Autoradiography showed phosphorylation of BMRF1 and EBV PK autophosphorylation, as indicated. (Bottom) A Coomassie stain of the GST proteins used for the kinase assay is also shown. Molecular weights (in thousands) are indicated on the left of panels A and B. (C) 293T cells were transfected with the Halo control vector or Halo-EBV PK, purified as described above for panel A, and immunoblotted with antibodies against Hsp90 (left) or Cdc37 (right). (D) HeLa cells were transfected with various combinations of the vector control, HA-tagged EBV PK, HA-tagged Daxx, and Flag-tagged Hsp90 expression vectors, as indicated, and then immunoprecipitated (IP) by using a FLAG antibody. Immunoblot analyses were performed to examine the expression of transfected proteins (direct load) or immunoprecipitated proteins by using HA, Hsp90, and actin antibodies.

Fig 2

Knockdown of Cdc37 does not affect EBV PK expression. HONE-Akata cells were transfected with pSG5-EBV PK in the presence of Cdc37 siRNA or equivalent amounts of a control siRNA. Whole-cell extracts were prepared, and immunoblot analysis was performed to analyze the expression of Cdc37, EBV PK, Cdc2 (a known cellular substrate of Cdc37), and actin.

Fig 3

17-DMAG decreases the expression level of EBV PK in EBV-positive cells. (A) HONE-Akata cells were transfected with an empty vector or a BZLF1 expression vector (SG5-Z), followed by a 48-h treatment with 17-DMAG (0.17 μM) or the DMSO control beginning at 4 h posttransfection. After 2 days, whole-cell extracts were prepared, and immunoblot analysis was performed to analyze the expression of EBV PK, BMRF1 (both derived from the endogenous viral genome), transfected BZLF1, Cdc2, and actin, as indicated. (B) SG5-Z-transfected AGS-Akata cells were treated with or without 17-DMAG (0.17 μM), as indicated, and immunoblot analyses were performed to detect the level of EBV PK, transfected BZLF1, or actin. (C) Different amounts of the protein extracts from the experiment shown in panel B were loaded onto a gel, as indicated, and examined by immunoblotting using antibodies directed against EBV PK and actin. (D) Latent EBV-positive HONE-Akata cells were transfected with SG5, SG5-EBV PK (HA tagged) (top), or SG5-Z (bottom), as indicated, followed by a 48-h treatment with 17-DMAG (0.17 μM) or the DMSO control beginning at 4 h posttransfection. Whole-cell extracts were prepared after 2 days, and immunoblot analysis was performed to analyze the expression of EBV PK, Cdc2, actin, and BZLF1, as indicated. (E) HeLa cells were transfected with SG5, SG5-EBV PK (HA-tagged), or SG5-Z (HA-tagged) expression vectors, and immunoblot analysis was performed by using an anti-HA antibody to detect the expression of the transfected EBV PK and BZLF1 proteins, Cdc2, or actin. (F) HeLa cells were transfected with the SG5-EBV PK (HA-tagged) vector in the presence or absence of 17-DMAG, and different amounts of the protein extracts were loaded onto a gel, as indicated, and examined by immunoblotting using antibodies directed against EBV PK (HA antibody) and actin.

Fig 4

17-DMAG decreases phosphorylation of EBV PK substrates. (A) AGS-Akata cells were treated with 17-DMAG or the DMSO control for 48 h. Whole-cell extracts were prepared 2 days later, and immunoblot analysis was performed to analyze the expression of BMRF1, BZLF1, Cdc2, and actin, as indicated. (B) AGS-Akata cells transfected with the vector control or BZLF1 (in the presence or absence of 17-DMAG) were examined by immunoblotting using antibodies to detect phosphorylated serine 22 of lamin A, total lamin A, EBV PK, transfected BZLF1, and tubulin. (C) HeLa cells were transfected with the vector control, a myc-tagged expression vector for EBV thymidine kinase (EBV TK) (150 ng), or the EBV TK expression vector (50 ng) and the EBV PK expression vector (50 ng) in the presence or absence of 17-DMAG. Immunoblot analyses were performed to examine EBV TK, EBV PK, and actin expression. (D) Mutu I Burkitt lymphoma cells were treated with and without TGF-β to induce lytic infection (in the presence or absence of 17-DMAG), and immunoblot analyses were performed to detect the expression of EBV PK, EBV TK, and actin.

Fig 5

Hsp90 is coprecipitated with multiple different CHPKs. (A) HeLa cells were transfected with the empty vector control (expressing the HA tag only) or vectors expressing different HA-tagged CHPKs, as indicated. After 28 h, cells were harvested, and the cell lysate was immunoprecipitated with an anti-HA antibody or loaded directly onto the gel (direct-load control). The precipitates and direct-load control were immunoblotted with antibodies against Hsp90 or HA, as indicated. (B) HeLa cells were transfected with various combinations of the vector control, HA-tagged HMCV UL97 or HA-tagged Daxx expression vectors, or a FLAG-tagged Hsp90 expression vector, as indicated, and then immunoprecipitated by using a FLAG antibody. Immunoblot analyses were performed to examine the expression of transfected proteins under each condition (direct load) or immunoprecipitated proteins by using antibodies against HA, Hsp90, and actin.

Fig 6

17-DMAG decreases expression of multiple CHPKs. (A) HeLa cells were transfected with the empty vector control or vectors expressing UL97, HHV6 PK, HHV7 PK, KSHV PK, RCMV PK, MHV68 PK, or EBV PK (all tagged with HA). A 48-h treatment with 17-DMAG (0.17 μM) or the DMSO control began at 4 h posttransfection. Whole-cell extracts were prepared, and immunoblot analyses were performed to analyze the expression of PKs using antibodies against HA, Cdc2, or actin, as indicated. (B) HeLa cells were transfected with the SG5-UL97 (HA-tagged) vector in the presence or absence of 17-DMAG, and different amounts of the protein extracts were loaded onto a gel, as indicated, and examined by immunoblotting using antibodies directed against UL97 (HA antibody) and actin.

Fig 7

17-DMAG inhibits expression of HCMV PK in HCMV-infected fibroblasts. (A) Primary human fibroblasts were infected with HCMV (strain AD169) and treated with 17-DMAG (0.17 μM) or the DMSO control for 6 or 24 h. Whole-cell extracts were prepared, and immunoblot analysis was performed to analyze the expression of UL97, IE1, UL44, and GAPDH, as indicated. M, mock infection. (B) Primary human fibroblasts were infected with HCMV in the presence or absence of 17-DMAG, and extracts were harvested at 24, 48, and 72 h postinfection (hpi). Immunoblot analysis was performed to analyze the expression of UL97, IE1, and GAPDH, as indicated.

Fig 8

17-DMAG reduces the EBV titer in lytically infected cells. AGS-Akata cells were transfected with SG5 or SG5-Z expression vectors, followed by a 48-h treatment with 17-DMAG (0.17 μM) or the DMSO control beginning at 4 h posttransfection. The cell medium was replaced after 48 h with medium free of drug. At 72 h posttransfection, cells and supernatant were harvested. (A) The virus titer was quantitated by infecting Raji cells with various amounts of the supernatant at 72 h posttransfection and counting the number of GFP-positive cells (green Raji units [GRU]) by using a fluorescence microscope. The results from two independent experiments are shown. (B) The number of viable AGS-Akata cells under each condition shown in panel A was determined by counting the cells (with trypan blue) at the end of the experiment. (C) Whole-cell extracts were prepared from the AGS-Akata cells used to derive the virus titer shown in panel A, and immunoblot analysis was performed to analyze the expression of BMRF1, BZLF1, and actin.

Fig 9

17-DMAG decreases intracellular lytic EBV DNA replication. (Left) AGS-Akata cells were treated with 17-DMAG (0.17 μM), the DMSO control, and/or acyclovir (50 μg/ml) for 2 days, and the DNA was harvested. The DNA was cut with BamHI and hybridized to a probe spanning the EBV termini for Southern blotting. The forms of the EBV genome containing fused viral termini (a mixture of latent and lytically replicated genomes) as well as the cleaved linear forms of the genome (produced only by lytically replicated viral DNA) are indicated. (Right) Immunoblotting was performed to ascertain the protein levels of BMRF1 and BZLF1 in the AGS-Akata samples used in the left panel.

Fig 10

Overexpression of EBV PK does not rescue viral replication in 17-DMAG-treated cells. AGS-Akata cells were transfected with SG5-Z, SG5-EBV PK, or both, followed by a 48-h treatment with 17-DMAG (0.17 μM) or the DMSO control beginning at 4 h posttransfection. The cell medium was replaced after 48 h with medium free of drug. At 72 h posttransfection, cells and supernatant were harvested. (A) The virus titer was quantitated by infecting Raji cells with various amounts of the supernatant at 72 h posttransfection and counting the number of GFP-positive cells (GRU) by using a fluorescence microscope. The results from two independent experiments are shown. (B) Whole-cell extracts were prepared from the AGS-Akata cells used to derive the virus in the experiments in panel A, and immunoblot analysis was performed to analyze the expression of BMRF1, BZLF1, and actin.

Fig 11

EBV PK and HCMV UL97 expression is not completely rescued by proteasomal or autophagy inhibitors in the presence of 17-DMAG. (A) AGS-Akata cells were transfected with SG5 or SG5-Z, as indicated, followed by a 24-h treatment with 17-DMAG (0.17 μM) or the DMSO control beginning at 12 h posttransfection in the presence or absence of the proteasomal inhibitor lactacystin or the autophagy inhibitor 3-MA. Immunoblot analysis was performed to analyze the expression of EBV PK, Cdc2, transfected BZLF1, and actin. (B) Primary human foreskin fibroblasts were subjected to serum starvation for 48 h and then infected with HCMV (strain AD169) at an MOI of 1. Before infection, cells were pretreated with 0.17 μM 17-DMAG for 1 h and incubated with the same amount of this drug until harvesting. Under some conditions, lactacystin or 3-MA was added to the medium at 6 h postinfection. Cells were harvested at 24 h postinfection, and immunoblot analysis performed to analyze the expression of UL97, IE1, and GAPDH. M, mock infection.