TRIM21, a New Component of the TRAIL-Induced Endogenous Necrosome Complex

Abstract

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is a well-known apoptosis inducer and a potential anticancer agent. When caspases and inhibitors of apoptosis proteins (IAPs) are inhibited, TRAIL induces necroptosis. Molecular mechanisms of necroptosis rely on kinase activation, and on the formation of a necrosome complex, bringing together the receptor-interacting protein kinases 1 and 3 (RIPK1, RIPK3), and the mixed lineage kinase domain-like protein (MLKL). In this study, mass spectrometry approach allowed to identify the tripartite motif containing 21 (TRIM21), an E3 ubiquitin-protein ligase as a new partner of the endogenous TRAIL-induced necrosome. Alteration of TRIM21 expression level, obtained by transient transfection of HT29 or HaCat cells with TRIM21-targeted siRNAs or cDNA plasmids coding for TRIM21 demonstrated that TRIM21 is a positive regulator of TRAIL-induced necroptosis. Furthermore, the invalidation of TRIM21 expression in HT29 cells by CRISPR-Cas9 technology also decreased cell sensitivity to TRAIL-induced necroptosis, a shortcoming associated with a reduction in MLKL phosphorylation, the necroptosis executioner. Thus, TRIM21 emerged as a new partner of the TRAIL-induced necrosome that positively regulates the necroptosis process.

Keywords: TRAIL; TRIM21; necroptosis; necrosome; proteomics.

FIGURE 1

TRAIL/z-VAD-fmk/Birinapant (TzB) induces necroptosis and necrosome formation in HT29 cells. (A) HT29 cells were pretreated with vehicle (DMSO 0.1%), Necrostatin-1 (Nec1, 10 µM) or Necrosulfonamide (NSA, 1 µM) for 1 h. Cells were then treated with TzB (in grey) or not (NT, in white) for 24 h. Percentage of PI positive dead cells were measured by flow cytometry (mean ± SD, n = 3). (###), p < 0.001 compared treated cells to NT cells; (***), p < 0.001 compared treated conditions. (B) HT29 cells were treated or not (0 min) with TRAIL-SK (500 ng/ml), z-VAD-fmk (25 µM) and Birinapant (Bp, 1 µM) for the indicated times. Four mg of cell lysates were immunoprecipitated (IP) with RIPK1 or RIPK3 antibody or control IgG. The RIPK1 or RIPK3 immunocomplexes were analyzed by western blot. *: an upper band for RIPK3 or a smear for RIPK1. Anti-human β-actin antibody was used as protein loading control. Representative data of three independent experiments. (C) Label-free quantification of protein of interest by mass spectrometry. Weighted spectral counts were calculated (means ± SD, n = 2) from the analysis of two different pools of eight RIPK3 immunoprecipitates for RIPK3, MLKL, RIPK1, FADD, Caspase 8, HS90B, TRIM21, and PGAM5 observed in IgG control (aIgG) and at times 0 and 3 h post-stimulation. (**) p < 0.01.

FIGURE 2

RIPK1, MLKL, P-MLKL, and TRIM21 are recruited to the endogenous RIPK3-dependent necrosome. HT29 cells were treated or not (0 min) with TRAIL-SK (500 ng/ml), z-VAD-fmk (25 µM), and Birinapant (Bp, 1 µM) (TzB) for the indicated times. Four mg of cell lysates were immunoprecipitated (IP) with RIPK3 antibody or control IgG as described in materials and methods. Upper panel: RIPK3 immunocomplexes were analyzed for indicated proteins by immunoblotting. *: an upper band for RIPK3 or a smear for RIPK1. Lower panel: expression of the indicated proteins was analyzed in the total cell lysates (input) by immunoblotting. Anti-human β-actin antibody was used as protein loading control. Representative data of three independent experiments.

FIGURE 3

Downregulation or overexpression of TRIM21 expression negatively or positively regulates TRAIL/z-VAD-fmk/Birinapant (TzB)-induced necroptosis, respectively. (A) HT29 and HaCat cells were transiently transfected with siTRIM21 or siNT1 (negative control). Immunoblot analysis of TRIM21, RIPK1, RIPK3, and MLKL expressions was carried out 72 h post-transfection. Anti-human Hsc70 antibody was used as protein loading control. Representative data of three independent experiments. (B) Forty eight hours after siRNA transfections, HT29, and HaCat cells were treated (in grey) or not (in white) with TRAIL-SK (100 ng/ml), z-VAD-fmk (25 µM) and Birinapant (Bp, 1 µM) (TzB) for 24 h. Percentages of necrotic cells (propidium iodide (PI) positive) were estimated (means ± SD, n = 3). (C) HT29 and HaCat cells were transiently transfected with pTRIM21-GST or control plasmid (used as a negative control). Immunoblot analysis of TRIM21, RIPK1, RIPK3, and MLKL expressions was carried out 72 h after transfection. Anti-human Hsc70 antibody was used as protein loading control. Representative data of three independent experiments. (D) Forty eight hours after transfection, HT29, and HaCat cells were treated (in grey) or not (in white) with TRAIL-SK (100 ng/ml), z-VAD-fmk (25 µM), and Birinapant (Bp, 1 µM) (TzB) for 24 h. Percentages of necrotic cells (propidium iodide (PI) positive) were estimated (means ± SD, n = 3). (###), p < 0.001 compared treated cells to NT cells; (*), p < 0.05, and (**), p < 0.01 compared treated conditions.

FIGURE 4

TRIM21-KO HT29 cells are more resistant to TRAIL/z-VAD-fmk/Birinapant (TzB)-induced necroptosis and their sensitivity can be rescued by TRIM21 re-expression. (A) WT (black lines) or TRIM21-KO (grey lines) HT29 cells were treated (dashed lines) or not (solid lines) with TRAIL-SK (100 ng/ml), z-VAD-fmk (25 µM), and Birinapant (Bp, 1 µM) (TzB) during 8 h. Necroptosis was evaluated by measuring SYTOX™ Green (SG) intensity (RFU) as described in materials and methods. Representative data of three independent experiments. (B) WT and TRIM21-KO HT29 cells were treated with TzB during the indicated times. Immunoblot analysis of TRIM21, RIPK1, RIPK3, MLKL, and phosphorylated forms of RIPK1, RIPK3, and MLKL were evaluated. Anti-human Hsc70 antibody was used as protein loading control. Representative data of three independent experiments. (C) TRIM21-KO HT29 cells were transfected with EGFP-control plasmid or TRIM21-EGFP plasmid for 48 h and approximately 40% of cells were GFP positive for each transfection conditions. Immunoblot analysis of TRIM21, RIPK1, RIPK3 and, MLKL expressions was carried out 48 h after transfection. Anti-human Hsc70 antibody was used as protein loading control. (D) After transfection with the EGFP-control plasmid (black lines) or with the pTRIM21-EGFP (grey lines), cells were treated or not (NT, solid lines) with TRAIL-SK (100 ng/ml), z-VAD-fmk (25 µM), and Birinapant (Bp, 1 µM) (TzB, dashed lines) during 8 h. Necroptosis was evaluated by measuring propidium iodide (PI) intensity (RFU) as described in materials and methods (means ± SEM, n = 3). (*), p < 0.05 compared treated conditions.