Human T-Cell Lymphotropic Virus Type 1 Transactivator Tax Exploits the XPB Subunit of TFIIH during Viral Transcription

Abstract

Human T-cell lymphotropic virus type 1 (HTLV-1) Tax oncoprotein is required for viral gene expression. Tax transactivates the viral promoter by recruiting specific transcription factors but also by interfering with general transcription factors involved in the preinitiation step, such as TFIIA and TFIID. However, data are lacking regarding Tax interplay with TFIIH, which intervenes during the last step of preinitiation. We previously reported that XPB, the TFIIH subunit responsible for promoter opening and promoter escape, is required for Tat-induced human-immunodeficiency virus promoter transactivation. Here, we investigated whether XPB may also play a role in HTLV-1 transcription. We report that Tax and XPB directly interact in vitro and that endogenous XPB produced by HTLV-1-infected T cells binds to Tax and is recruited on proviral LTRs. In contrast, XPB recruitment at the LTR is not detected in Tax-negative HTLV-1-infected T cells and is strongly reduced when Tax-induced HTLV-1 LTR transactivation is blocked. XPB overexpression does not affect basal HTLV-1 promoter activation but enhances Tax-mediated transactivation in T cells. Conversely, downregulating XPB strongly reduces Tax-mediated transactivation. Importantly, spironolactone (SP)-mediated inhibition of LTR activation can be rescued by overexpressing XPB but not XPD, another TFIIH subunit. Furthermore, an XPB mutant defective for the ATPase activity responsible for promoter opening does not show rescue of the effect of SP. Finally, XPB downregulation reduces viability of Tax-positive but not Tax-negative HTLV-1-transformed T cell lines. These findings reveal that XPB is a novel cellular cofactor hijacked by Tax to facilitate HTLV-1 transcription.IMPORTANCE HTLV-1 is considered the most potent human oncovirus and is also responsible for severe inflammatory disorders. HTLV-1 transcription is undertaken by RNA polymerase II and is controlled by the viral oncoprotein Tax. Tax transactivates the viral promoter first via the recruitment of CREB and its cofactors to the long terminal repeat (LTR). However, how Tax controls subsequent steps of the transcription process remains unclear. In this study, we explore the link between Tax and the XPB subunit of TFIIH that governs, via its ATPase activity, the promoter-opening step of transcription. We demonstrate that XPB is a novel physical and functional partner of Tax, recruited on HTLV-1 LTR, and required for viral transcription. These findings extend the mechanism of Tax transactivation to the recruitment of TFIIH and reinforce the link between XPB and transactivator-induced viral transcription.

Keywords: TFIIH; oncoprotein; promoter opening; retrovirus; transcription.

FIG 1

XPB binds to Tax. (A) GST-pulldown assay performed with GST, GST-CA, and GST-Tax (wt and M47 mutant) and in vitro transcribed/translated (TnT) XPB. Binding of XPB to the GST constructs was analyzed by immunoblot using an anti-XPB antibody. Protein inputs were determined by taking UV images of the SDS-PAGE gel before transfer. (B) Coimmunoprecipitation of Tax and endogenous XPB in 293T cells. Cells were transfected with a control or Tax plasmid and total proteins prepared 24 h posttransfection were analyzed by immunoblot before (lysates) or after anti-Tax immunoprecipitation (IP Tax). (C) Coimmunoprecipitation of Tax and overexpressed XPB in 293T cells. Cells were cotransfected with a control or XPB plasmid together or not with the Tax plasmid and proteins were analyzed as in (B). (D) Coimmunoprecipitation of endogenous Tax and XPB produced in HTLV-1-transformed C8166 T cells. Immunoprecipitated proteins (IP Tax, upper) or total proteins obtained before (input) or after (flowthrough) IP (lower) were analyzed by immunoblot. Ig, signal corresponding to immunoglobulins used for the IP. Each panel corresponds to one representative experiment out of two or three performed.

FIG 2

Endogenous XPB is recruited at the proviral LTR in a Tax-dependent manner. (A and B) ChIP experiments to detect XPB recruitment at various genomic loci. Sonicated chromatins prepared for Tax-positive C8166 and CB-CD4/HTLV-1 (A) or Tax-negative/low MT-1 and TL-Om1 HTLV-1-infected T cells (B) were amplified by PCR before (input) or after immunoprecipitation with control IgG or anti-XPB antibody. PCRs were performed using primers specific for the R or the 5′ LTR U5 region of the HTLV-1 LTR or for the IκBα promoter (positive control) and the α-satellite regions (negative control). PCR amplification of the chromatin (input) gave positive signals for both R and the 5′ LTR U5 regions in the four T cell lines (18 to 22 PCR cycles compared to 37 to 45 for water), confirming that they all carried integrated HTLV-1 genomes. Data correspond to means ± SEM of two independent experiments performed in triplicate. (C) Effect of chaetocin on XPB, Tax, and Tax cofactor expression levels. C8166 T cells were treated with 100 nM chaetocin for 24 h and the levels of endogenous p300, CBP, XPB, CREB, and Tax were analyzed by immunoblots (one representative experiment out of two). (D) Effect of chaetocin treatment on XPB recruitment to either the LTR or the IκBα promoter. ChIP analysis was performed as in (A). Data correspond to means ± SEM of one experiment performed in triplicate out of two. Statistical significance: ns, not significant; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

FIG 3

XPB is involved in Tax-mediated LTR activation. (A) Effect of XPB overexpression on HTLV-1 LTR activation in Jurkat T cells. Cells were transfected with the LTR U3R-Luc and pRL-TK reporter plasmids along with the control (basal) or Tax (transactivation) plasmids and with the XPB-GFP or XPD-GFP construct and then treated with DMSO or with EPL or SP at 10 μM for 24 h. Upper panel: relative luciferase activity calculated by normalizing firefly/renilla ratios to that of DMSO-treated Tax-transfected cells (set to 1). Data are means ± SEM of three independent experiments performed in duplicate. Lower panel: protein expression levels detected by immunoblots. Level of each band was normalized to the level of DMSO-treated control cells (lane 1). Data are from one representative experiment out of three. (B) XPB-GFP interaction with Tax. Human 293T cells were transfected with a control plasmid or with the Tax plasmid together with a plasmid encoding the untagged or GFP-fused version of XPB. Total (lysates) or anti-Tax immunoprecipitated (IP) proteins were blotted with the anti-XPB or anti-Tax antibody. (C) Effect of untagged XPB overexpression on HTLV-1 LTR activation in Jurkat T cells. Cells were transfected with the LTR U3R-Luc and pRL-TK reporter plasmids along with the Tax and XPB constructs and then treated with DMSO or with EPL or SP at 10 μM for 24 h. Left panel: relative luciferase activity calculated by normalizing firefly/renilla ratios to that of DMSO-treated Tax-transfected cells (set to 1). Data are means ± SEM of three independent experiments performed in duplicate. Right panel: protein expression levels detected by immunoblots. Level of each band was normalized to the level of DMSO-treated Tax-transfected cells (lane 1). Data are from one representative experiment out of two. (D) Effect of XPB knockdown on LTR activation by Tax. RFP positive-sorted Jurkat T cells producing the control of XPB shRNA were induced (dox+) or not (dox-) and transfected with the U3R-Luc and pRL indicator plasmids with or without the Tax plasmid. Left panel: expression of Tax and XPB analyzed by immunoblots. Data correspond to one representative experiment out of three. Right panel: luciferase production. Data are means ± SEM of three independent experiments performed in duplicate. Statistical significance: *, P < 0.05; **, P < 0.01.

FIG 4

The R-E-D domain of XPB is required for proviral LTR transactivation and endogenous Tax production. (A) Expression levels of LTR-regulating factors or TFIIH components in C8166 T cells treated with SP. (B) Capacity of XPB-GFP or XPD-GFP to rescue SP-mediated inhibition of LTR transactivation or endogenous Tax production. HTLV-1-transformed C8166 T cells were transfected with the LTR U3R-Luc and pRL-TK reporter plasmids and either the XPB-GFP or XPD-GFP construct and were then treated with DMSO or with EPL or SP at 10 μM for 48 h. Upper panel: relative luciferase activities calculated by normalizing firefly/renilla ratios to that of DMSO-treated control cells (bar 1). Data are means ± SEM of three independent experiments performed in duplicate. Lower panel: Tax, XPB-GFP, or XPD-GFP expression levels detected by immunoblotting. Data are from one representative experiment out of three. (C) Interaction of Tax with the XPB mutants. Upper panel: schematic representation of the XPB protein and position of the XPB mutations. Lower panel: interaction of Tax with wild-type XPB or XPB mutants. Human 293T cells were cotransfected with one of the XPB constructs along with a control or Tax expressor. Total proteins prepared 24 h posttransfection were analyzed before (lysates) or after the anti-Tax immunoprecipitation (IP Tax). Data correspond to one representative experiment out of two. (D) Capacity of wt or mutated XPB proteins to rescue SP-mediated inhibition of LTR transactivation or endogenous Tax production. HTLV-1-transformed C8166 T cells were transfected with the LTR U3R-Luc and pRL-TK reporter plasmids and one of the XPB constructs and were then treated with DMSO or with EPL or SP at 10 μM for 48 h. Upper panel: relative luciferase activity calculated by normalizing firefly/renilla ratios to that of DMSO-treated control cells (bar 1). Data are means ± SEM of five independent experiments performed in duplicate. Lower panel: Tax and XPB expression levels detected by immunoblotting. Data are from one representative experiment out of three. Statistical significance: ns, not significant; *, P < 0.05; **, P < 0.01; ****, P < 0.0001.

FIG 5

XPB downregulation inhibits viral transcription. (A) C8166 and MT-4 HTLV-1-transformed T cells or HTLV-1-immortalized CB-CD4/HTLV-1 T cells were treated with DMSO or with SP or EPL at 10 μM for 24 h. The level of Gag or Tax mRNA was quantified by RT-qPCR and normalized to the level of housekeeping genes (NQ). For MT-4 and CB-CD4/HTLV-1 T cells, the level of Gag protein (p19) was also analyzed by immunoblotting. Data correspond to one representative experiment out of two or three. (B) Time-course experiment to quantify the level of Gag mRNA over the course of SP treatment. C8166 T cells were either untreated (NT) or treated with DMSO or with SP or EPL at 10 μM. The level of XPB was analyzed at indicated times by immunoblotting in half of the cells. At each time point, the intensity of the XPB signal in EPL- or SP-treated cells was compared to that of DMSO-treated cells (set to 1). Data correspond to one representative experiment out of three. (C) The other half of the cells was used to quantify the level of Gag mRNA. Left panel shows the SP/DMSO ratios calculated from the level of RNA normalized to the level of housekeeping genes (NQ). Right panel shows the correlation between the level of XPB protein quantified from (B) and the SP/DMSO ratios shown in the left panel. (D) Effect of SP on the levels of housekeeping gene mRNAs. Data show the SP/DMSO ratios for EEF1G or HPRT calculated from relative quantities using the formula: Q = 2^−ΔCp. Data correspond to one representative experiment performed in triplicate out of three. Statistical significance: *, P < 0.05; ***, P < 0.001; ****, P < 0.0001.

FIG 6

XPB downregulation reduces viability of Tax-positive HTLV-1-infected T cells. C8166 (A), MT-4 (B), TL-om1 (C), MT-1 (D), Jurkat transformed T cells (E), and HTLV-1 immortalized CB-CD4/HTLV-1 cells (F) were treated every day with DMSO (0) or with EPL or SP at the indicated concentrations and cell viability was determined by the MTT method after 3 days of culture. The viability values for SP or EPL were normalized to that of DMSO-treated cells (set as 100%). For each cell line, viability results correspond to two independent experiments performed in triplicate. Ns, not significant; **, P < 0.01; ***, P < 0.001, ****, P < 0.0001.

FIG 7

Model illustrating the capacity of Tax to target essential steps of the Pol II-mediated transcription process, including the promoter-opening step controlled by the XPB subunit of TFIIH.