A Dual Role for FADD in Human Precursor T-Cell Neoplasms

Abstract

A reduction in FADD levels has been reported in precursor T-cell neoplasms and other tumor types. Such reduction would impact on the ability of tumor cells to undergo apoptosis and has been associated with poor clinical outcomes. However, FADD is also known to participate in non-apoptotic functions, but these mechanisms are not well-understood. Linking FADD expression to the severity of precursor T-cell neoplasms could indicate its use as a prognostic marker and may open new avenues for targeted therapeutic strategies. Using transcriptomic and clinical data from patients with precursor T-cell neoplasms, complemented by in vitro analysis of cellular functions and by high-throughput interactomics, our results allow us to propose a dual role for FADD in precursor T-cell neoplasms, whereby resisting cell death and chemotherapy would be a canonical consequence of FADD deficiency in these tumors, whereas deregulation of the cellular metabolism would be a relevant non-canonical function in patients expressing FADD. These results reveal that evaluation of FADD expression in precursor T-cell neoplasms may aid in the understanding of the biological processes that are affected in the tumor cells. The altered biological processes can be of different natures depending on the availability of FADD influencing its ability to exert its canonical or non-canonical functions. Accordingly, specific therapeutic interventions would be needed in each case.

Keywords: FADD; T-cell acute lymphoblastic leukemia/T-cell lymphoblastic lymphoma (T-ALL/LBL); canonical and non-canonical functions; interactomics; precursor T-cell neoplasms; transcriptomics.

Figure 1

Reduced FADD levels associated with poor prognosis in precursor T-cell neoplasms. (A) Analysis of FADD expression in cancer types. The gene summary for FADD in cancers was gained from Oncomine^TM Research Edition (http://www.oncomine.org (accessed on 18 September 2018)). (B) Kaplan–Meier disease-free survival curve analysis in precursor T-cell neoplasms with high and low FADD expression obtained from the dataset published in the Ref. [29]. HR, hazard ratio (Mantel–Haenszel); 95% CI, 95% confidence interval of ratio. (C,D) Violin plots representing FADD expression (DESeq2 normalized counts) in patients belonging to the ETP, nearETP, and notETP subtypes and in patients belonging to the Pre-Cortical, Cortical, and Post-Cortical subtypes (as defined in the Ref. [26]), respectively. Krustal–Wallis was used to test for statistical significance.

Figure 2

The FADD-negative phenotype showed a significant association with oncogenic signatures. Gene Set Enrichment Analysis was performed in 264 patients with precursor T-cell neoplasm, based on their FADD expression levels, which defined FADD-negative and FADD-positive phenotypes. These signatures were significantly enriched in the FADD-negative phenotype. (A) This signature was created ad hoc based on previous literature and is defined in Supplementary Table S1. (B,C) These signatures were selected from the Molecular Signatures Database, and their systematic names are M3323 (Doxorubicin Resistance) and M15835 (Tamoxifen Resistance). The green curve corresponds to the enrichment score (ES) curve, which is the running sum of the weighted enrichment score obtained with the GSEA software. NES, normalized enrichment score; p, nominal p value.

Figure 3

Pharmacological arrest at the G2/M stage of the cell cycle. The cell cycle of the FADD-expressing (FADD) and FADD-deficient (NEG) JURKAT cell lines was evaluated without any treatment (untreated) or after treatment with 100 nM etoposide, 50 ng/mL nocodazole, and 10 nM paclitaxel for 18 h. (A) Representative histograms of cells stained with propidium iodine showing DNA content distribution. Cell cycle phase distribution was analyzed by flow cytometry. (B) Bar chart of cell cycle distribution from six independent experiments. One-way ANOVA was used to test for statistical significance: * p ≤ 0.05; ** p ≤ 0.01. Error bars represent the standard error of the mean (SEM).

Figure 4

The FADD-positive phenotype showed a significant association with energy metabolism signatures. Gene Set Enrichment Analysis was performed for 264 patients with precursor T-cell neoplasm, based on their FADD expression levels, which defined FADD-negative and FADD-positive phenotypes. These signatures were significantly enriched in the FADD-positive phenotype and selected from the Molecular Signatures Database. Their systematic names are M5936 (Hallmark_Oxidative Phosphorylation) (A), M893 (Reactome_Respiratory Electron Transport) (B), M13293 (GOBP_Electron Transport Chain) (C) and M25889 (GOCC_Respirasome) (D). The green curve corresponds to the enrichment score (ES) curve, which is the running sum of the weighted enrichment score obtained with the GSEA software. NES, normalized enrichment score; p, nominal p-value.

Figure 5

Interactome analysis using DIA-MS confirms FADD participation in energy metabolism processes. (A) Workflow of co-Immunoprecipitation-Mass Spectrometry. Endogenous FADD protein was co-immunoprecipitated from FADD-expressing (FADD) and FADD-deficient (NEG) JURKAT cell lines using magnetic beads. Samples were trypsin-digested and injected into the mass spectrometer for data-independence acquisition. Finally, the raw data were analyzed with Spectronaut software and R. (B) STRING interaction map of significant proteins that interact with FADD in JURKAT cells compared to JURKAT-deficient FADD (FADD NEG). Proteins were clustered based on K-means into five clusters. The functional enrichment analysis and adjusted p-value (adj. p-value) are shown for the three main clusters. (C) Validation of DIABLO–FADD interaction. The input and immunoprecipitated (IP) fractions were separated by SDS-PAGE and blotted with antibodies against anti-FADD and -DIABLO. The densitometry values are shown below. (D) Schematic representation to show interactions of FADD with proteins involved in different biological processes. Under non-apoptotic conditions, FADD interacts with DIABLO/Smac, which hampers FADD function inducing apoptosis. FADD also interacts with several proteins involved in the energy metabolism.