A Kaposi's sarcoma-associated herpesvirus microRNA and its variants target the transforming growth factor β pathway to promote cell survival

Abstract

Transforming growth factor β (TGF-β) signaling regulates cell growth and survival. Dysregulation of the TGF-β pathway is common in viral infection and cancer. Latent infection by Kaposi's sarcoma-associated herpesvirus (KSHV) is required for the development of several AIDS-related malignancies, including Kaposi's sarcoma and primary effusion lymphoma (PEL). KSHV encodes more than two dozen microRNAs (miRs) derived from 12 pre-miRs with largely unknown functions. In this study, we show that miR variants processed from pre-miR-K10 are expressed in KSHV-infected PEL cells and endothelial cells, while cellular miR-142-3p and its variant miR-142-3p_-1_5, which share the same seed sequence with miR-K10a_ +1_5, are expressed only in PEL cells and not in uninfected and KSHV-infected TIME cells. KSHV miR-K10 variants inhibit TGF-β signaling by targeting TGF-β type II receptor (TβRII). Computational and reporter mutagenesis analyses identified three functional target sites in the TβRII 3' untranslated region (3'UTR). Expression of miR-K10 variants is sufficient to inhibit TGF-β-induced cell apoptosis. A suppressor of the miRs sensitizes latent KSHV-infected PEL cells to TGF-β and induces apoptosis. These results indicate that miR-K10 variants manipulate the TGF-β pathway to confer cells with resistance to the growth-inhibitory effect of TGF-β. Thus, KSHV miRs might target the tumor-suppressive TGF-β pathway to promote viral latency and contribute to malignant cellular transformation.

Fig 1

TGF-β signaling is inhibited by miRs processed from KSHV pre-miR-K10. (A) KSHV infection suppresses TGF-β signaling, shown in a reporter assay. Mock- or KSHV-infected TIME cells were transiently transfected with pSBE4-Luc and β-galactosidase constructs for 24 h, treated with 1.0 ng/ml TGF-β1 for an additional 24 h, and measured for relative luciferase activities. (B) Inhibition of TGF-β signaling by miRs derived from KSHV pre-miR-K10. TIME cells were transiently cotransfected with pSBE4-Luc and β-galactosidase constructs together with pre-miR-K10 or a vector control for 24 h, treated with 1.0 ng/ml TGF-β1 for an additional 24 h, and measured for relative luciferase activities.

Fig 2

Detection of miR-K10 and miR-142-3p variants by RT-qPCR in KSHV-infected cells and TIME, Bjab, and 293T stable cells expressing pre-miR-K10 and a vector control. (A and B) Detection of miR-K10 variants (A) or miR-142-3p variants (B) in TIME cells and KSHV-infected BC-3, BCP-1, and TIME-KSHV cells. (C and D). Detection of miR-K10 variants (C) or miR-142-3p variants (D) in stable TIME, Bjab, and 293T cultures of pre-miR-K10 or a vector control.

Fig 3

KSHV miR-K10 variants inhibit TGF-β signaling. (A to D) miR-K10 variants suppress TGF-β signaling, shown in a reporter assay. Stable TIME cells (A), Bjab cells (B), or 293T cells (C and D) expressing pre-miR-K10 or a vector control were transfected with pSBE4-Luc (A, B, and C) or pMBE-luc (D) together with the β-galactosidase construct for 24 h, treated with 1.0 ng/ml TGF-β1 for an additional 24 h, and measured for relative luciferase activities. (E) KSHV miR-K10a mimic suppresses TGF-β signaling, shown in a reporter assay. MiR-K10a mimic or a scrambled control was cotransfected into 293T cells together with pSBE4-Luc and β-galactosidase constructs for 24 h, treated with TGF-β1 for an additional 24 h, and measured for relative luciferase activities.

Fig 4

KSHV miR-K10 variants inhibit Smad3 phosphorylation. (A) Inhibition of Smad3 phosphorylation in TIME cells stably expressing pre-miR-K10. (B) Inhibition of Smad3 phosphorylation in Bjab cells stably expressing pre-miR-K10. Cells grown at the exponential phase were treated with 1.0 ng/ml of TGF-β1 for 20 min, collected, and measured for p-Smad3 by Western blotting. Numbers shown on the lanes are the relative intensities after calibration with β-tubulin.

Fig 5

KSHV miR-K10 variants downregulate the level of the TβRII protein. (A and B) Expression of pre-miR-K10 reduces the protein level of TβRII but not other components of the TGF-β pathway in 293T cells (A) or TIME cells (B). Protein levels of Smad2/3, Smad4, Smad7, TβRI, and TβRII in stable cells expressing pre-miR-K10 or a vector control were detected by Western blotting. Numbers shown on the lanes are the relative intensities after calibration with β-tubulin. (C and D) Expression of pre-miR-K10 reduces the protein level of TβRII in 293T cells (C) or TIME cells (D), shown by immunofluorescence staining. The TβRII protein in stable cells expressing pre-miR-K10 or a vector control was examined by immunofluorescence assay. (E) The level of the TβRII protein was reduced in Bjab cells expressing pre-miR-K10. The TβRII protein was detected by Western blotting. Numbers shown on the lanes are the relative intensities after calibration with β-tubulin. (F to H) KSHV miR-K10 variants don't affect the level of TβRII mRNA. TβRII mRNA levels in stable TIME cells (F), 293T cells (G), or Bjab cells (H) expressing pre-miR-K10 or a vector control were measured by RT-qPCR.

Fig 6

KSHV infection downregulates the level of the TβRII protein. (A) KSHV infection reduces the protein level of TβRII in TIME cells. Protein levels of TβRII in mock and latently KSHV-infected TIME cells were detected by Western blotting. Numbers shown on the lanes are the relative intensities after calibration with β-tubulin. (B) KSHV infection reduces the protein level of TβRII in TIME cells, shown by immunofluorescence staining. TβRII protein in mock- or KSHV-infected TIME cells was examined by immunofluorescence assay. (C) Expression of TβRII in the KSHV-infected PEL cell lines BC-3, BCP-1, BCBL-1, and BC-1 and the KSHV-negative cell line Bjab, analyzed by flow cytometry.

Fig 7

KSHV miR-K10 and miR-142-3p variants target TβRII 3′UTR. (A) KSHV miR-K10 variants suppress full-length TβRII 3′UTR reporter activity. Stable 293 cells expressing pre-miR-K10 or a vector control were transfected with the full-length TβRII 3′UTR reporter construct (TβRII 3′UTR WT) and a β-galactosidase expression construct for 48 h and measured for relative luciferase activities. (B) Sequence alignment of miR-K10 and miR-142-3p variants with the TβRII 3′UTR WT reporter and mutant reporters TβRII 3′UTR Mut1, TβRII 3′UTR Mut2, or TβRII 3′UTR Mut3, with mutation in the respective putative targeting site. (C) Mutation of any of the three predicted sites abolishes inhibition of TβRII 3′UTR reporter activity by mimics of miR-K10 and miR-142-3p variants. Cells were cotransfected with mimics or a scrambled mimic control together with TβRII 3′UTR luciferase reporters and the pRL-TK construct for 48 h and measured for relative luciferase activities using pRL-TK as a calibration control.

Fig 8

KSHV miR-K10 variants inhibit TGF-β-induced apoptosis. (A) A mimic of miR-K10a protects cells from TGF-β-induced apoptosis. Bjab cells were transfected with either a mimic of miR-K10a or a scrambled control and treated with TGF-β1 at 5 ng/ml. At 4 days posttreatment, cells were collected and apoptotic cells were examined. x and y axes indicate annexin V and PI staining intensities, respectively. (B) Summary of percentages of early, late, and total apoptotic cells in panel A. (C) KSHV miR-K10 variants inhibit TGF-β-induced PARP cleavage. Stable pre-miR-K10 Bjab cells treated with TGF-β1 at 1 and 5 ng/ml for 24 h or 48 h were harvested and examined for PARP cleavage by Western blotting.

Fig 9

Expression of TβRII without the native 3′UTR abolishes the inhibition of TGF-β-induced apoptosis by a mimic of KSHV miR-K10a. (A) Bjab cells with stable expression of TβRII without the native 3′UTR or a vector control were transfected with a mimic of miR-K10a or a scrambled control and treated with TGF-β1 at 5 ng/ml. At 4 days posttreatment, cells were collected, and apoptotic cells were examined. x and y axes indicate annexin V and PI staining intensities, respectively. (B) Summary of percentages of early, late, and total apoptotic cells in panel A. (C) Examination of TβRII expression in Bjab cells by flow cytometry. Bjab cells with stable expression of TβRII without the native 3′UTR or a vector control were transfected with a mimic of miR-K10a or a scrambled control for 4 days and examined for expression of TβRII.

Fig 10

A suppressor of miR-K10a inhibits the function of all miR-K10 and miR-142-3p variants. (A) A suppressor of miR-K10a relieved the inhibition of the miR-K10a sensor reporter by all miR-K10 variants. Cells were cotransfected with a miR-K10a sensor luciferase reporter, the pRL-TK construct, mimics, or a scrambled mimic control, together with a suppressor of miR-K10a or a scrambled suppressor control for 48 h, and measured for relative luciferase activities using pRL-TK as a calibration control. (B) A suppressor of either miR-K10a or miR-142-3p relieved the inhibition of the TβRII3′UTR WT reporter by all miR-K10 and miR-142-3p variants. Cells were cotransfected with the TβRII3′UTR WT luciferase reporter, the pRL-TK construct, mimics, or a scrambled mimic control, together with a suppressor of miR-K10a or miR-142-3p, or a scrambled suppressor control for 48 h, and measured for relative luciferase activities using pRL-TK as a calibration control.

Fig 11

A suppressor of miR-K10a increases TβRII expression, enhances TGF-β signaling, sensitizes cells to apoptosis, and decreases cell growth in KSHV-infected BCP-1 cells. (A) A suppressor of miR-K10a restores the expression of the TβRII protein in latent KSHV-infected BCP-1 PEL cells. BCP-1 cells transfected with 10 nM miR-10a suppressor or a scrambled oligonucleotide for 4 days were examined for the TβRII protein by Western blotting. Numbers shown on the lanes are the relative intensities after calibration with β-tubulin. (B) A suppressor of miR-K10a restores the responsiveness of latent BCP-1 cells to TGF-β. BCP-1 cells transfected with 10 nM miR-10a suppressor or a scrambled oligonucleotide for 4 days were treated with 1.0 ng/ml of TGF-β1 for 20 min and collected for measurement of pSmad3 by Western blotting. Numbers shown on the lanes are the relative intensities after calibration with GAPDH. (C) A suppressor of miR-K10a sensitizes BCP-1 cells to TGF-β-induced cell growth arrest. BCP-1 cells transfected with 10 nM miR-K10a suppressor or a scrambled oligonucleotide for 24 h were treated with 10 ng/ml of TGF-β1 for another 5 days. The cell number in each well was counted daily. Data are shown as averages ± SEM from 3 repeats. (D to F) A suppressor of miR-K10a sensitizes BCP-1 cells to apoptosis. BCP-1 cells transfected with 10 nM miR-K10a suppressor or a scrambled oligonucleotide for 24 h were treated with 1.0 ng/ml of TGF-β1 for another 48 h. Cells were collected and examined for apoptotic cells by dual staining of annexin V (x axis) and PI (y axis), followed by flow cytometry analysis (D), and results are summarized as the averages ± SEM from 3 repeats (E) or PARP cleavage by Western blotting (F).