Suppression of HTLV-1 replication by Tax-mediated rerouting of the p13 viral protein to nuclear speckles

Abstract

Disease development in human T-cell leukemia virus type 1 (HTLV-1)-infected individuals is positively correlated with the level of integrated viral DNA in T cells. HTLV-1 replication is positively regulated by Tax and Rex and negatively regulated by the p30 and HBZ proteins. In the present study, we demonstrate that HTLV-1 encodes another negative regulator of virus expression, the p13 protein. Expressed separately, p13 localizes to the mitochondria, whereas in the presence of Tax, part of it is ubiquitinated, stabilized, and rerouted to the nuclear speckles. The p13 protein directly binds Tax, decreases Tax binding to the CBP/p300 transcriptional coactivator, and, by reducing Tax transcriptional activity, suppresses viral expression. Because Tax stabilizes its own repressor, these findings suggest that HTLV-1 has evolved a complex mechanism to control its own replication. Further, these results highlight the importance of studying the function of the HTLV-1 viral proteins, not only in isolation, but also in the context of full viral replication.

Figure 1

p13 expression decreases viral production and Tax activity. (A) Schematic representation of the HTLV-1 genome that encodes the enzymatic genes gag-pro-pol and env and the regulatory proteins p12, p30, p13, Rex, Tax, and HBZ (arrow indicates antisense direction of the HBZ gene) at the 3′ of the genome. Box under env indicates the second exon that contains the ATG-initiating codons for the envelope p30, Tax, and Rex proteins. The singly spliced orf-II p13 mRNA is illustrated below. (B) Single-letter amino acid code of p13 from HTLV-1_LAF. (C) 293T cells transfected with the HTLV-1 molecular clone pAB (0.5 μg) without and with increasing amounts of p13-HA and the reporter construct HTLV-1 LTR-Luc (0.1 μg). DNA concentrations were equalized using the backbone vector pMH. Viral replication was measured 48 hours after transfection by extracellular p19 Gag ELISA (pg/mL) and intracellular anti-p24 Gag immunoblotting. Lysates were examined for p13 and Tax expression. Tubulin was used as a loading control. Tax-dependent LTR-Luc activity was measured and adjusted for protein concentration. p19 Gag concentration and luciferase activity were set to 100% for pAB alone, and the standard deviation represents an average of 3 independent experiments. (D) Quantitative RT-PCR was performed on cytoplasmic and total RNA isolated from 293T cells transfected with pAB (0.5 μg) and p30, p13, or vector control (pMH) (1.0 μg). RNA levels of cytoplasmic over total for gag (top panel) and tax/rex (bottom panel) are shown. The graphs correspond to gag or tax signals adjusted for gapdh level. Graphs represent values from duplicate experiments. (E) 293T cells were transfected with pcTax (50 ng) without (−) or with (+) increasing amounts of p13-HA and the reporter construct HTLV-1 LTR-Luc (0.1 μg). Luciferase activity is normalized for protein concentration and is the result presented is representative of > 3 independent experiments. Lysates were examined by immunoblotting for expression of p13, Tax, and tubulin.

Figure 2

p13 competes with p300 for Tax interaction. (A) 293T cells were transfected with (+) or without (−) Tax and with increasing amounts of p13-HA. Tax-normalized lysates were immunoprecipitated using anti-Tax antibody and immunoblotted with anti-p300 and anti-Tax antibodies. The right panel shows immunoblots of total cell lysates (input) with anti-HA and anti-p300 antibodies. (B) 293T cells were transfected with empty vector, without (−) and with (+) pcTax and increasing amounts of p13-HA and with and without p300-Flag (5 μg). Lysates were immunoprecipitated with anti-p300 antibody and both immunoprecipitate and input (right panel) were immunoblotted with anti-HA, anti-Tax, and anti-Flag. (C) 293T cells were transfected with empty vector, p13-HA without (−) and with (+) Tax, or Tax alone. Cell lysates were immunoprecipitated using anti-Tax antibody. Immunoprecipitates (left panel) and total cell lysates (input, right panel) were immunoblotted with anti-HA and anti-Tax antibodies. (D) In vitro association of Tax and p13. In vitro transcribed and translated S-methionine–labeled p13 protein was incubated with GST (top panel lane 2) or GST-Tax (top panel lane 3) immobilized on glutathione-agarose beads. Radiolabeled p13 bound to GST-Tax was visualized with a PhosphorImager. The bottom panel is a Coomassie blue staining of the same gel as in the top panel showing GST (lane 1) and GST-Tax (lane 2) protein present in each reaction. (E) 293T cells were transfected with vector control, Tax without and with increasing amounts of p300, or Tax with p13 and increasing amounts of p300. The reporter construct HTLV-1 LTR-Luc was included and luciferase activity adjusted for transfection efficiency using RSV-βgal. DNA concentrations were equalized using the appropriate backbone vectors. SD is given for the average of 3 independent experiments. Lysates were examined by immunoblotting for expression of Tax, Flag-p300, and tubulin.

Figure 3

p13 is stabilized by Tax. (A) 293T cells were transfected without (−) and with increasing amounts of p13 and lysates immunoblotted by anti-HA and anti-tubulin antibodies (top panel, −Tax). The +Tax (bottom panel) is the same experiment presented in Figure 1E and added here for comparison. (B) RT-PCR was performed on 293T cells transfected with vector control DNA, p13 alone, p13 and Tax, or Tax alone. Total RNA was collected 48 hours after transfection and subjected to real-time PCR using primers specific for p13 (top panel), Tax (middle panel), or actin (bottom panel). The graph represents p13 mRNA levels from 3 independent experiments. (C) 293T cells were transfected with the CMV-driven, Rex-dependent reporter plasmid pRXRE-CAT and Rex without and with increasing amounts of p13-HA. Whole-cell lysates were immunoblotted for expression of Rex, p13-HA, CAT, and tubulin. (D) To measure the half-life of the p13 protein, 293T cells were transfected with p13 in the absence or presence of Tax. The cells were treated with 10μM cycloheximide 48 hours after transfection. Whole-cell lysates were prepared at the indicated times and the protein levels were examined by Western blot analysis for p13 (anti-HA). The bottom panel is a Coomassie blue–stained gel shown as a loading control.

Figure 4

p13 translocates to the nucleus when coexpressed with Tax. HeLa cells were transfected with p13-HA (2μg) or Tax (0.5μg) and fixed 24 hours after transfection (A-D). DNA concentrations were adjusted using pMH plasmid. Subcellular localization of p13 was examined by immunofluorescence with anti-HA and anti-CoxIV (A) and anti-HA and anti-SC35 (B). Tax localization with anti-Tax and anti-SC35 (C) and anti-Tax and anti-CoxIV (D) antibodies is shown. (E-G) HeLa cells transfected with the p13-HA (2μg) and Tax (0.5μg) constructs and fixed 24 hours after transfection. Subcellular localization of Tax and p13 was examined by immunofluorescence with anti-HA and anti-Tax antibodies (E). Subcellular localization of p13 was analyzed with anti-HA and anti-SC35 (F) and with anti-HA and anti-sp100 and (G). An overlay of these images is shown on the right.

Figure 5

p13 is ubiquitinated in the presence of Tax. (A) 293T cells were transfected with p13 (4.0 μg; lanes 2 and 3) in the absence or presence of Tax (1.0 μg). Lysates were prepared with buffer containing NEM (10μM) and immunoblots were made with antibodies to Tax, HA, and tubulin. (B) 293T cells were transfected with Tax (1.0 μg) and Ub-WT (2.0 μg), Ub-K63 (2.0 μg), or Ub-K48 (2.0 μg) and p13-HA (4.0 μg) as indicated. Lysates were immunoblotted for the expression of Tax, p13-HA, and tubulin. Arrows indicate ubiquitinated forms of p13.

Figure 6

p13 modulates virus expression in HTLV-1–infected cells. (A) 293T cells were transfected with HTLV-1 molecular clones pAB (gray bars) or p13KO (black bars). Culture supernatants were harvested 48 and 72 hours after transfection and extracellular p19 Gag levels measured by ELISA assay. Western blot analysis for intracellular p24 Gag, Tax, and actin was performed on whole-cell lysates from 72-hour cultures. The values given are an average of 4 independent experiments. (B) Luciferase activity for the reporter construct LTR-Luc (0.5 μg or 1.0 μg) was measured for 293T cells transfected with HTLV-1 molecular clones pAB (gray bars) or p13KO (black bars). Values are an average of 2 independent experiments and are adjusted for transfection efficiency using pRLTK-Luc. (C) MT-2 cells were transfected with the reporter construct HTLV-1 LTR-Luc and without (−) and with increasing amounts of p13-HA. Lysates were immunoblotted for the expression of p24 Gag, p13-HA, and tubulin. Luciferase activity was measured 48 hours after transfection and normalized for protein concentration. Luciferase activity was set at 100% for cells transfected in the absence of p13-HA. SD is given for the average of 3 independent experiments. (D) MT-2 cells were transfected with p13-HA and, 48 hours after transfection, cells were added to fibronectin-precoated coverslips, fixed, and stained with anti-HA, anti-Tax, or anti-SC35 antibodies for analysis by confocal microscopy. DAPI staining identifies the cell nucleus. (E) Schematic representation of the predicted interplay between Tax and p13. Tax binds to the viral LTR in the nucleus through interaction with CREB. Tax activates transcription through the recruitment of basal transcription machinery and coactivators such as p300/CBP. The p13 protein expressed alone resides in the mitochondrial membrane, affects the electron transport chain (ETC), the membrane potential (Δψ), and the production of reactive oxygen species (ROS). In the presence of Tax, however, p13 is stabilized, becomes ubiquitinated, and part of it interacts with Tax, colocalizes with Tax to nuclear speckles, and reduces viral expression.