HTLV-1 tax-induced rapid senescence is driven by the transcriptional activity of NF-κB and depends on chronically activated IKKα and p65/RelA

Abstract

The HTLV-1 oncoprotein Tax is a potent activator of classical and alternative NF-κB pathways and is thought to promote cell proliferation and transformation via NF-κB activation. We showed recently that hyperactivation of NF-κB by Tax triggers a cellular senescence response (H. Zhi et al., PLoS Pathog. 7:e1002025, 2011). Inhibition of NF-κB activation by expression of I-κBα superrepressor or by small hairpin RNA (shRNA)-mediated knockdown of p65/RelA rescues cells from Tax-induced rapid senescence (Tax-IRS). Here we demonstrate that Tax-IRS is driven by the transcriptional activity of NF-κB. Knockdown of IKKγ, the primary Tax target, by shRNAs abrogated Tax-mediated activation of both classical and alternative NF-κB pathways and rendered knockdown cells resistant to Tax-IRS. Consistent with a critical role of IKKα in the transcriptional activity of NF-κB, IKKα deficiency drastically decreased NF-κB trans-activation by Tax, although it only modestly reduced Tax-mediated I-κBα degradation and NF-κB nuclear localization. In contrast, although IKKβ knockdown attenuated Tax-induced NF-κB transcriptional activation, the residual NF-κB activation in IKKβ-deficient cells was sufficient to trigger Tax-IRS. Importantly, the phenotypes of NIK and TAK1 knockdown were similar to those of IKKα and IKKβ knockdown, respectively. Finally, double knockdown of RelB and p100 had a minor effect on senescence induction by Tax. These data suggest that Tax, through its interaction with IKKγ, helps recruit NIK and TAK1 for IKKα and IKKβ activation, respectively. In the presence of Tax, the delineation between the classical and alternative NF-κB pathways becomes obscured. The senescence checkpoint triggered by Tax is driven by the transcriptional activity of NF-κB, which depends on activated IKKα and p65/RelA.

Fig 1

IKKα and IKKγ are essential for Tax-mediated NF-κB activation as well as Tax-IRS. (A) Isolation and characterization of IKKα-, IKKβ-, and IKKγ-KD cell lines. IKKα-, IKKβ-, and IKKγ-KD clones were established in a HeLa-G reporter cell line by individually knocking down the α, β, and γ subunits of the IKK complex using their respective shRNA-expressing lentiviral vectors as described in Materials and Methods. The cells were cloned via limiting dilutions, expanded, and validated by immunoblotting. (B) Tax-mediated NF-κB activation in IKKα-, IKKβ-, and IKKγ-KD cells. IKKα-, IKKβ-, and IKKγ-KD cells were cotransfected with E-selectin-Luc or LTR-Luc and pRL-TK with or without Tax. Luciferase assays were performed at 48 h posttransfection. Activation was calculated as the ratio of reporter activities in cells with Tax over those in cells without Tax. (C) Tax-induced senescence requires IKKα and IKKγ. The three IKK knockdown cell clones, ΔN-IκBα cells, and the wild-type HeLa-G cells were plated at low density and transduced with Ad-Tax at an MOI of 1. After 5 days, cells were examined and photographed using an Olympus IX81 inverted fluorescence microscope equipped with a charge-coupled device (CCD) camera as previously reported (41). (D) Deficiency in IKKα or IKKγ confers resistance to Tax-IRS. The growth curve of Ad-Tax-transduced cells of each KD cell line was determined by counting the total number of EGFP⁺ cells at 2 and 5 days after transduction using an automated fluorescence-based cell counter as described in Materials and Methods. The data were then plotted on a line graph using Prism software.

Fig 2

Distinct impairments of I-κBα degradation, p100 processing, and p21 and p27 upregulation in IKKα-, IKK-β-, and IKKγ-deficient cells. The IKKα, IKK-β, and IKKγ knockdown cell clones (lanes 3 to 8) and the control wild-type HeLa-G cells (lanes 1 and 2) were transduced with Ad-Tax or Ad-tTa (a control adenoviral vector for the tet trans-activator) at an MOI of 10 for 48 h. The cells were then harvested and immunoblotted for the indicated proteins.

Fig 3

Nuclear p65/RelA and activated IKKα are both needed for Tax-induced NF-κB trans-activation and for triggering senescence response. (A) HeLa-G, IKKα-KD, IKKβ-KD, and IKKγ-KD cell clones were plated on chamber slides and transduced with Ad-Tax at an MOI of 1 for 48 h. The cells were then fixed, permeabilized, and immunostained for p65/RelA (red). Nuclei were stained with DAPI (blue). EGPF fluorescence indicates Tax expression. (B) The indicated cell clones were transduced with either Ad-tTa or Ad-Tax at an MOI of 10 for 48 h. Cells were harvested and fractionated into cytoplasmic and nuclear fractions and immunoblotted for the indicated proteins. GAPDH and HDAC1 were used as indicators for cytoplasmic and nuclear fractions, respectively.

Fig 4

Tax induces nuclear localization of IKK complex. (A) HeLa-G cells were transduced with either Ad-tTa or Ad-Tax at an MOI of 10 for 48 h. Cells were then harvested and fractionated into nuclear and cytoplasmic fractions and immunoblotted for the indicated proteins. HDAC1 and GAPDH were used as markers for nuclear and cytoplasmic compartments, respectively. (B) HeLa-G cells were plated on chamber slides and transduced with Ad-Tax at an MOI of 1 for 48 h. The cells were then fixed, permeabilized, and immunostained for IKKα (red). Nuclei were stained with DAPI (blue). EGPF fluorescence indicates Tax expression. The immunostaining was examined and photographed using the 63× objective of a Pascal confocal microscope equipped with a CCD camera.

Fig 5

The alternative NF-κB pathway is not required for Tax-induced cellular senescence. The RelB and p100 double KD cells were derived by methods similar to Fig. 1 except that a pool of high-titer shRNA vectors targeting both RelB and p100 was used. A clone that displayed the most significant knockdown of both RelB and p100 was then used for the study. (A) Cells of parental HeLa-G and RelB/p100-KD clone were transduced with Ad-Tax (+) or Ad-tTa (−) at an MOI of 10 for 48 h. The cells were then harvested and immunoblotted for the indicated proteins. (B) (Left) HeLa-G cells (top) and the RelB/p100-KD clone (bottom) were transduced with Ad-Tax as for Fig. 1C and photographed at day 5. (Right) The growth curve of Ad-Tax-transduced RelB/p100-KD cells was determined as in Fig. 1D.

Fig 6

Phenotypes of NIK and TAK1 knockdown parallel those of IKKα and IKKβ knockdown, respectively. (A) Derivation and characterization of TAK1-KD and NIK-KD cell lines. TAK1-KD and NIK-KD clones were derived as in Fig. 1. To verify NIK knockdown, wild-type HeLa-G and NIK-KD cells were treated with 50 μM proteasome inhibitor MG132 for 0, 2, 4, and 8 h to inhibit the constitutive proteasomal degradation of NIK. Cells were subsequently harvested and immunoblotted for NIK. (The asterisk indicates a degradation product of NIK protein.) (B) Deficiency in TAK1 compromised but deficiency in NIK abrogated Tax-mediated NF-κB activation. TAK1-KD and NIK-KD cells were cotransfected with E-selectin-Luc and pRL-TK with or without Tax. Luciferase reporter assay was performed at 48 h posttransfection. Activation was calculated as in Fig. 1B. (C) NIK-KD but not TAK1-KD prevented Tax-IRS. TAK1-KD (top left) and NIK-KD (bottom) cells were plated and transduced with Ad-Tax as for Fig. 1C. Growth curves were produced as described in Fig. 1D. (D) I-κBα, p21, and p27 levels in untreated (mock), Ad-tTa, and Ad-Tax-transduced HeLa-G, TAK1-KD, and NIK-KD cells. Cells were left untreated or were transduced with Ad-Tax or Ad-tTa at an MOI of 10 for 48 h. The cells were then harvested and immunoblotted for the indicated proteins.

Fig 7

Phenotypes of cells expressing dominant negative NIK mutant (NIK-DN) parallel those of NIK-KD cells. (A) Derivation and characterization of NIK-DN cells. An HA-tagged dominant negative NIK mutant is stably expressed in wild-type HeLa-G cells via lentiviral vector and puromycin selection. Cells were subsequently harvested and immunoblotted for HA. (B) Expression of dominant negative NIK mutant compromised Tax-mediated NF-κB activation. HeLa-G, NIK-DN, and NIK-KD cells were cotransfected with E-selectin-Luc and pRL-TK with or without Tax. Luciferase reporter assay was performed at 48 h posttransfection. Activation was calculated as for Fig. 1B. (C) NIK-DN prevented Tax-IRS. Wild-type HeLa-G (left, top) and NIK-DN (bottom) cells were plated and transduced with Ad-Tax as in Fig. 1C. Growth curves were calculated as described in Fig. 1D. The growth curve of NIK-KD cells was included as a positive control.

Fig 8

Schematic representation of the possible mechanisms by which Tax interacts and chronically activates the NF-κB pathway. Tax interacts with NEMO and recruits TAK1 and NIK to IKK complexes, which initiate the activation of predominantly the classical pathway and, to a lesser extent, the alternative NF-κB pathways. The Tax-mediated recruitment of TAK1 to IKK likely involves K63 polyubiquitination, as suggested by recent results (13, 27, 28). Whether NIK recruitment requires K63 polyubiquitin is not clear. Nuclear localization of p65/RelA and IKKα as mediated by Tax induces chromatin remodeling, chronic NF-κB transcriptional activation, and, subsequently, cellular senescence.