NEMO is essential for Kaposi's sarcoma-associated herpesvirus-encoded vFLIP K13-induced gene expression and protection against death receptor-induced cell death, and its N-terminal 251 residues are sufficient for this process

Abstract

Kaposi's sarcoma-associated herpesvirus-encoded viral FLICE inhibitory protein (vFLIP) K13 was originally believed to protect virally infected cells against death receptor-induced apoptosis by interfering with caspase 8/FLICE activation. Subsequent studies revealed that K13 also activates the NF-κB pathway by binding to the NEMO/inhibitor of NF-κB (IκB) kinase gamma (IKKγ) subunit of an IKK complex and uses this pathway to modulate the expression of genes involved in cellular survival, proliferation, and the inflammatory response. However, it is not clear if K13 can also induce gene expression independently of NEMO/IKKγ. The minimum region of NEMO that is sufficient for supporting K13-induced NF-κB has not been delineated. Furthermore, the contribution of NEMO and NF-κB to the protective effect of K13 against death receptor-induced apoptosis remains to be determined. In this study, we used microarray analysis on K13-expressing wild-type and NEMO-deficient cells to demonstrate that NEMO is required for modulation of K13-induced genes. Reconstitution of NEMO-null cells revealed that the N-terminal 251 amino acid residues of NEMO are sufficient for supporting K13-induced NF-κB but fail to support tumor necrosis factor alpha (TNF-α)-induced NF-κB. K13 failed to protect NEMO-null cells against TNF-α-induced cell death but protected those reconstituted with the NEMO mutant truncated to include only the N-terminal 251 amino acid residues [the NEMO(1-251) mutant]. Taken collectively, our results demonstrate that NEMO is required for modulation of K13-induced genes and the N-terminal 251 amino acids of NEMO are sufficient for supporting K13-induced NF-κB. Finally, the ability of K13 to protect against TNF-α-induced cell death is critically dependent on its ability to interact with NEMO and activate NF-κB.

Importance: Kaposi's sarcoma-associated herpesvirus-encoded vFLIP K13 is believed to protect virally infected cells against death receptor-induced apoptosis and to activate the NF-κB pathway by binding to adaptor protein NEMO/IKKγ. However, whether K13 can also induce gene expression independently of NEMO and the minimum region of NEMO that is sufficient for supporting K13-induced NF-κB remain to be delineated. Furthermore, the contribution of NEMO and NF-κB to the protective effect of K13 against death receptor-induced apoptosis is not clear. We demonstrate that NEMO is required for modulation of K13-induced genes and its N-terminal 251 amino acids are sufficient for supporting K13-induced NF-κB. The ability of K13 to protect against TNF-α-induced cell death is critically dependent on its ability to interact with NEMO and activate NF-κB. Our results suggest that K13-based gene therapy approaches may have utility for the treatment of patients with NEMO mutations and immunodeficiency.

FIG 1

NEMO is essential for induction of K13-induced genes in Jurkat cells. (A) Immunoblot showing expression of Flag-tagged K13-ER^TAM (K13-ER) in WT and NEMO-deficient (NEMO^Def) Jurkat cells. (B) Scatter plot analyses of genes induced by 4OHT treatment in the control vector and K13-ER^TAM-expressing WT and NEMO-deficient Jurkat cells. Each point on the scatter plots represents the expression of an individual mRNA message, as determined by units of fluorescence intensity, in untreated cells (x axis) plotted against its expression after 4OHT treatment (y axis). The lines on the scatter plot indicate the 2-fold boundaries used for selecting genes with differential expression. (C) To validate the gene array data, 3 genes, TNFSF10, BIRC3, and NFKB1A, were randomly selected, and their relative mRNA levels in mock- and 4OHT-treated vector- and K13-ER^TAM-expressing WT and NEMO-deficient Jurkat cells were examined after 24 h of treatment using qRT-PCR. Real-time PCRs were performed in triplicate, and the data are presented as the fold change (mean ± SE) in target gene expression (***, P < 0.001, Student's t test). (D) Immunoblot showing expression of Flag-tagged vFLIP E8-ER^TAM and vFLIP HS-ER^TAM in WT Jurkat cells. (E) The relative mRNA levels of TNFSF10, BIRC3, and NFKB1A in mock- and 4OHT-treated vector- and vFLIPs-ER^TAM-expressing Jurkat cells were examined after 16 h of 4OHT treatment, as described for panel C.

FIG 2

hGLuc protein complementation assay for detecting K13-NEMO interaction. (A) Schematic representation of GLuc protein complementation assay for detecting K13-NEMO interaction. The NH₂- and COOH-terminal fragments of humanized GLuc (blue and red, respectively) are fused individually to the COOH termini of NEMO and K13 via a small linker (shown in yellow). Coexpression of the two fragments in a host cell results in interaction between NEMO (N) and K13 (K), which leads to folding of the hGLuc fragment into its native structure and the reconstitution of hGLuc activity. (B) Coexpression of an empty vector, NEMO-hGLuc[1], K13-hGLuc[2], or the LZ-hGLuc[1] control in 293T cells results in the reconstitution of hGLuc activity when full-length NEMO and K13 are present, while expression of the fusion proteins or the LZ-hGluc[1] control individually fails to do so. (C, D) hGLuc PCA assay showing that NEMO-hGLuc[2] strongly interacts with IKK1-hGLuc[1] and IKK2-hGLuc[1], while K13-hGLuc[2] fails to do so. (E) K13 interacts with NEMO(1-246) but does not effectively interact with NEMO(247-419), as determined by hGLuc PCA. 293T cells were transfected with the indicated constructs along with a respiratory syncytial virus–β-galactosidase construct. hGLuc activity was measured in cell lysates at approximately 48 h posttransfection and normalized on the basis of the β-galactosidase activity. Values shown are the mean ± SD from one representative experiment out of three performed in duplicate (*, P < 0.05, Student's t test).

FIG 3

NEMO(1-251) supports K13-induced NF-κB pathway activation. (A) (Top) Schematic representation of NEMO/IKKγ domains showing its two helices (HLX1 and HLX2), two coiled-coil domains (CC1 and CC2), NEMO ubiquitin-binding (NUB) domain, a leucine zipper (LZ) domain, and a zinc finger (ZF) domain. The regions of NEMO involved in binding to IKKα/β and K13 and cytokine/LPS signaling are also shown. (Bottom) The NEMO(1-251) mutant with a truncated C terminus. (B) Coimmunoprecipitation assay showing that K13 interacts with NEMO FL and NEMO(1-251). Flag-tagged K13-ER^TAM was immunoprecipitated from NEMO-deficient MEFs reconstituted with an empty vector, NEMO FL, or NEMO(1-251) using control (lanes C) or Flag (lanes F) antibody beads, and the presence of coimmunoprecipitated NEMO was detected by immunoblotting. All samples were treated with 20 nM 4OHT for 48 h prior to lysate preparation. Results are representative of those from two independent experiments. CL, cell lysates; IP, immunoprecipitation. (C) An ELISA-based NF-κB binding assay showing increased p65/RelA DNA-binding activity in the nuclear extracts of K13-ER^TAM-expressing NEMO^−/Y MEFs reconstituted with NEMO FL and NEMO(1-251) upon 72 h of treatment with 20 nM 4OHT. Values shown are the mean ± SD from one representative experiment out of three performed in triplicate (**, P < 0.01, Student's t test). (D) Whole-cell lysates from MEFs coexpressing K13-ER^TAM and either the vector alone, NEMO FL, or NEMO(1-251) were examined for NF-κB activation by Western blotting using an antibody against phospho-IκBα and tubulin as a loading control. MEFs were treated with 20 nM 4OHT for the indicated time intervals prior to lysate preparation. Results are representative of those from two independent experiments.

FIG 4

Truncation mutant NEMO(1-251) fails to support TNF-α-induced NF-κB activation. (A) Whole-cell lysates from MEFs coexpressing K13 and either an empty vector, NEMO FL, or NEMO(1-251) were examined for NF-κB activation upon TNF-α treatment (10 ng/ml) at various time points (0, 5, 30, 240 min) by Western blotting using antibodies against phospho-IκBα and total IκBα and antibody against GAPDH as a loading control. (B) NEMO^−/Y MEFs reconstituted with an empty vector, NEMO FL, or NEMO(1-251) and coexpressing the vector or K13-Flag were left untreated or treated with 10 ng/ml TNF-α for 12 h, and cell viability was measured using an MTS assay. WT MEFs were used as a control. Values shown are the mean ± SD from one representative experiment out of three performed in duplicate (**, P < 0.01, Student's t test). (C) NEMO^−/Y MEFs coexpressing the empty vector, NEMO FL, or NEMO(1-251) and the vector or K13-Flag were treated with TNF-α as described for panel A and imaged using a phase-contrast microscope. Magnification, ×10. Results are representative of those from three independent experiments.