Previous studies have reported that the Fas-FasL pathway and the downstream caspase-8 play important roles in islet mass and insulin secretion (Chinnaiyan et al., 1995). As a classical adaptor in Fas-FasL signaling, Fas-associated death domain-containing protein (FADD) also participates in a series of nonapoptotic processes regulated by its phosphorylation (Bonnet et al., 2011; Hua et al., 2003; Imtiyaz et al., 2009; Kabra et al., 2001; Lee et al., 2012; Matsuyoshi et al., 2006; Pyo et al., 2005; Shimada et al., 2006; Thomas et al., 2004; Welz et al., 2011; Yao et al., 2013; 2015). However, until now, its role in insulin degradation has not yet been evaluated.

It has been reported that overexpression of Fas in β cells leads to the formation of autoimmune diabetes (Moriwaki et al., 1999). In addition, Fas was also found to affect the production of insulin; insulin secretion is inhibited in Fas-deficient mice (Strandmark and Hallberg, 2007). These studies proposed that the Fas signaling pathway might play important roles in insulin production and secretion. Previous studies performed in our lab showed that FADD-D mice, which mimic the constitutive phosphorylation of FADD, displayed an obviously decreased islet area (Yao et al., 2015). FADD phosphorylation was found to be involved in glucose metabolism, which resulted in low blood glucose, elevated insulin levels, decreased mRNA expression levels of insulin, and decreased glucose-stimulated insulin secretion (GSIS) in FADD-D mice. The islet secretion function of FADD-D mice was impaired probably due to developmental defects. Impaired insulin secretion and continuously elevated insulin levels in serum thus present us with the hypothesis that FADD-D mice might be associated with insulin degradation disorders (Yao et al., 2015).

The liver is the major tissue that helps remove approximately 50% of circulating insulin, which regulates liver glucose and lipid metabolism. Hepatic insulin clearance therefore contributes to regulating insulin action by controlling insulin availability to peripheral tissues (Bojsen-Moller et al., 2018). The underlying mechanism of hepatic metabolism by insulin is vital for understanding both normal physiology and the pathogenesis of metabolic diseases, such as hyperglycemia and hypertriglyceridemia (Rubenstein et al., 1972; Titchenell et al., 2017).

Insulin-degrading enzyme (IDE) is responsible for the degradation of a number of hormones and peptides, including insulin, and clears ~50% of insulin during the first portal passage (Duckworth et al., 1998; Villa-Perez et al., 2018). IDE knockout mice displayed a diabetic phenotype, exhibiting hyperinsulinemia and glucose intolerance (Abdul-Hay et al., 2011). IDE mutant mice were found to exhibit impaired GSIS as well as increased levels of insulin in pancreatic islets (Steneberg et al., 2013). In addition, human genetic studies have suggested that IDE polymorphisms are associated with the pathogenesis of diabetes mellitus (Bonnefond et al., 2010). Although it is well-known that IDE plays an important role in modulating blood glucose by insulin degradation, relatively little is known about the mechanisms of IDE regulation.

To the best of our knowledge, our investigations on IDE in livers of FADD-D mice reveal, for the first time, that FADD, especially in its phosphorylated form, has an impact on IDE regulation. We therefore have expanded the understanding of the non-apoptotic functions of FADD and its phosphorylation.

Plasmids and reagents

The pEGFP-N1-FoxO1 plasmid and pEGFP-N1-TSS plasmid were kindly provided by Prof. D. Accili (Naomi Berrie Diabetes Center, Department of Medicine, College of Physicians & Surgeons of Columbia University, USA). The human IDE promoter (1-900) was cloned into the SAC I and Hind III restriction sites of the pGL3-basic vector (Promega, USA). The polymerase chain reaction (PCR) primer sequences used for plasmid construction were forward-CGAGCTCAAAGTAACCACTGAAGCCG and reverse-GGAAGCTTATTCCGCTTAC-CCACACAGG. Dulbecco’s modified Eagle’s medium (DMEM), Opti-MEM, and Lipofectamine 2000 reagent were obtained from Invitrogen/Thermo Fisher Scientific (USA). Fetal bovine serum (FBS) was obtained from Sigma-Aldrich (USA). Anti-IDE, anti-GAPDH, anti-SIRT1, and anti-β-actin antibodies were from Santa Cruz Biotechnology (USA); anti-FADD and anti-CK-18 were from Abcam (USA); anti-FoxO1 was from Cell Signaling Technology (USA); and HRP goat-anti-mouse IgG, HRP rabbit-anti-goat IgG, and HRP goat-anti-rabbit IgG were purchased from Jackson ImmunoResearch Europe (USA). Alexa Fluor 488 donkey anti-rabbit IgG and goat anti-rabbit IgG were obtained from Invitrogen/Thermo Fisher Scientific. The immunohistochemical and DAB substrate kits were obtained from Shenzhen Fumax Technology (China). The SIRT1 inhibitor EX527 and activator SRT1720 were purchased from APExBIO (USA). DNase I was purchased from Sangon Biotech (China). Act D, collagenase IV, and DAPI were obtained from Sigma-Aldrich. All other chemicals were purchased from Sigma-Aldrich.

Animals

All studies involving mice were conducted in accordance with the high standards of animal welfare approved by the Nanjing University Animal Care and Use Committee in China (20180413). Animals were maintained in a specific pathogen-free animal facility on a 12-h light-dark cycle at an ambient temperature of 21°C. They were given free access to food and water. Detailed protocols for generating FADD-D mice have been described previously (Hua et al., 2003). Adult FADD-D and wild-type (WT) control mice of both sexes were used for the experiments.

Cell culture

HepG2, HEK293, and HEK293T cells were obtained from the Cell Bank of the Chinese Academy of Science (China). Cells were cultured in DMEM (Gibco/Thermo Fisher Scientific) supplemented with 10% FBS, 50 μg/ml streptomycin, and 50 U/ml penicillin. The primary hepatocytes were maintained in low-glucose DMEM supplemented with 20% FBS, 50 μg/ml streptomycin, and 50 U/ml penicillin. Cells were maintained at 37°C in a humidified incubator with 5% CO₂.

Primary hepatocytes extraction and cell culture

A 1.5% pentobarbital sodium solution was warmed in a water bath (37°C); thereafter, 5 min before perfusion, 100 µl of this solution was injected. The hepatic portal vein was exposed by carefully moving the viscera to the right of the abdominal cavity. Then, an 18-gauge angiocath was inserted into the hepatic portal vein. The perfusate tubing was connected to the needle, and infusion was initiated in situ at a low flow rate (10 ml/min) with 5 ml of prewarmed (37°C) D-hank’s solution with 1,000 U heparin (Sigma-Aldrich). If performed properly, the liver would instantly begin to blanch. Once successful cannulation was confirmed, a cut was made at the inferior vena cava (IVC) to allow efflux. A further test for successful cannulation was performed by applying light pressure with a sterile swab on the IVC, leading to all lobes of the liver quickly beginning to swell. The flow rate of the infusion was increased to 25 ml/min. The liver became pale in color. The perfusion solution was replaced with 5 ml of D-hank’s solution plus collagenase IV (3 mg/ml) (Sigma-Aldrich) with no interruption in the flow for an additional 6 min. After collagenase perfusion, the liver began to appear spongy. The liver was resected and placed in a prechilled sterile beaker with 5 ml of DMEM and DNase I (1 mg/ml) (Sangon Biotech) and then moved to a tissue cell culture hood. Under the hood, a cell scraper was used to gently disperse the cells into DMEM in a sterile Petri dish. The cell dispersion was filtered through a 100 µm pore size cell strainer into a 50 ml conical tube in order to remove the connective tissues and undigested tissue fragments. The cells were then suspended in 40 ml of DMEM and centrifuged at 50g for 3 min at 4°C. The supernatant was aspirated, and the cells were gently resuspended in 40 ml of cold DMEM to wash the cells. Centrifugation was repeated, the supernatant was aspirated, and the cells were gently resuspended in 15 ml of DMEM. The cells were counted within the cell suspension using a hemocytometer, and cell viability was determined by trypan blue staining.

siRNAs and transfection

All synthetic siRNAs and negative controls (NCs) were purchased from GenePharma (China). Cells were transiently transfected with siRNAs or plasmids using Lipofectamine 2000 according to the manufacturer’s instructions. The siRNA sequence of the target was as follows: FADD-5’-CACAGAGAAGGAGAACGCA-3’.

Luciferase reporter assay

For dual-luciferase reporter assays, HEK293T cells were seeded in a 24-well plate at a density of 1 × 105 cells per well in antibiotic-free medium one day prior to transfection. HEK293T cells were transfected with siRNA/NC and then cotransfected with IDE-promoter luciferase and control pCMV-RL reporter for 24 h. Each well of cells was transiently cotransfected with 0.8 µg of pGL-basic3-IDE plasmids, 0.8 µg of pEGFP-N1-FoxO1 plasmid or pEGFP-N1-TSS plasmid, and 8 ng of Renilla reporter plasmid (pCMV-RL; Promega) as an internal control using Lipofectamine 2000. The luciferase activity was measured consecutively using the Dual-Luciferase reporter assay (Promega).

RNA extraction and quantitative real-time PCR (RT-qPCR)

Total RNA was extracted from tissues and cells with TRIzol reagent (Invitrogen/Thermo Fisher Scientific) following the manufacturer’s instructions. RT-qPCR was performed using a reverse transcription kit (Takara, Japan) and SYBR Green PCR Master Mix (Roche, Switzerland). RT-qPCR was performed on an ABI system (Applied Biosystems, USA). Primer sequences were as follows: IDE, 5’-GCCGCCGATCACAGCAGCAGGAG-3’ and 5’-CCCACAGACCGGTAGCCCAGGAT-3’; FADD, 5’-GCCGCGCCTGGGGAAGAAGAC-3’ and 5’-GCAAAGCAGCGGCCCATCAGGA-3’; β-actin, 5’-CATCGAGCACGGCATCGTCA-3’ and 5’-TAGCACAGCCTGGATAGCAAC-3’. All data are represented as the means of the fold change from triplicate analyses and normalized to β-actin.

Western blotting (WB)

Liver samples were taken from the mice and snap-frozen in liquid N₂ and then stored for subsequent protein extractions. The tissue samples and cells were lysed in ice-cold lysis buffer for 30 min. The supernatant was boiled with Laemmli sample buffer for SDS-PAGE. Antibodies were sourced as follows: anti-FADD (1:500) (Abcam); anti-IDE (1:500), anti-SIRT1 (1:1,000), anti-β-actin (1:2,000), and anti-GAPDH (1:2,000) (Santa Cruz Biotechnology); HRP goat-anti-mouse IgG (1:3,000), HRP goat-anti-rabbit IgG (1:3,000), and HRP rabbit-anti-goat IgG (1:1,000) (Jackson ImmunoResearch Europe)

Histology

Livers obtained from the euthanized mice were fixed in 10% phosphate-buffered formalin and embedded in paraffin, and then 5-µm-thick sections were stained for routine histology. For immunohistochemical staining, the liver sections and primary hepatocytes were immunostained with anti-IDE antibody (1:300) (Santa Cruz Biotechnology) and visualized with DAB (Shenzhen Fumax Technology). From immunofluorescence staining, the expression levels of CK-18 and FoxO1 were detected. An immunofluorescence assay was used to detect the localization of cells with anti-FoxO1 antibody (1:1,000) (Cell Signaling Technology) and the identification of the primary hepatocyte cells with anti-CK-18 (1:1,000). Alexa Fluor 488-conjugated anti-rabbit IgG antibody and anti-mouse IgG antibody (1:1,000) (Invitrogen/Thermo Fisher Scientific) were used as secondary antibodies. Nucleic DNA was stained with 4,6-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich). Images were then visualized with an Axiophot 2 microscope (Zeiss, Germany). Cells climbing to the carry sheet glass were stained according to the instructions of the PAS kit (Nanjing JianCheng Bioengineering Institute, China). Briefly, cell slides were washed with phosphate-buffered saline (PBS) three times. Primary hepatocytes were stained for 10 min for glycogen detection. After being rinsed under running water for 5 min, the cell slides were then stained for 10 s using the other reagent and then washed under running water again. The cell slides were sealed after drying. Periodic acid-Schiff (PAS)-positive glycogen was stained red or purple-red with blue nuclei. Images were then visualized with an Axiophot 2 microscope.

Statistical analysis

Data are presented as the mean ± SEM. Differences between two groups was analyzed by two-tailed Student’s t-test using GraphPad Prism 6.0 (GraphPad Software, USA). Values were considered statistically significant at P < 0.05.

Genotyping of FADD-D mice

The construction of the mouse model was described in detail in our previously published paper (Hua et al., 2003). As shown in Fig. 1, the correct bands for the FADD-Neo gene and TgD gene were amplified except for the FADD-Endo gene in FADD-D mice, while the correct bands for the FADD-Endo gene were amplified except for the FADD-Neo gene and the TgD gene in WT mice. The results indicated that the FADD-Endo gene had been knocked out completely, suggesting that FADD-D mice were the target mice for our subsequent study.

Identification of the primary hepatocytes

Primary hepatocytes needed to be used in the following experiments; thus, we first investigated whether the primary hepatocytes were extracted from the livers of living mice successfully. As shown in Fig. 2A, the cells displayed attachment and stretching, round, oval, polygonal, and dual-core phenotypes under the microscope. The cytoplasm was granular, and the cell boundary profile was clearly visible. PAS staining showed that the cytoplasm was stained by glycogen in the primary hepatocytes extracted from FADD-D mice and the control mice. Moreover, the amount of glycogen was found to be elevated in FADD-D mice compared with the control mice (Fig. 2B), which might be associated with the elevated serum insulin levels in FADD-D mice. Cytokeratin-18 (CK-18) is a normal component of the cytoskeleton of hepatocytes. Our immunofluorescence assay also showed that the hepatocytes were stained with green fluorescence by the CK-18 antibody (Fig. 2C).

Downregulation of IDE in the livers and primary hepatocytes from FADD-D mice

Previous studies performed in our lab showed that FADD-D mice exhibited increased insulin levels in serum and decreased insulin secretion (Yao et al., 2015). To explore the underlying mechanisms, the expression level of IDE was examined in the current study. Our data showed that both the mRNA and protein expression levels of IDE were downregulated in the liver (Figs. 3A and 3B) and primary hepatocytes (Figs. 3D and 3E) in FADD-D mice by qPCR and WB analyses. Immunohistochemistry staining also confirmed that the protein level of IDE in FADD-D liver and primary hepatocytes was downregulated (Figs. 3C and 3F). In FADD-D mice, the expression level of IDE was reduced, while the concentration of serum insulin remained elevated (Yao et al., 2015). Thus, we speculate that the increased insulin level might be caused by less insulin degradation due to the reduced expression of IDE.

Reduced IDE expression in FADD-knockdown HepG2 cells

Our previous studies indicated that FADD deficiency sometimes displays similar effects or phenotypes to FADD phosphorylation (Cheng et al., 2014; Zhuang et al., 2016). Thus, to further investigate the effect of FADD on IDE, we next determined the stability of IDE mRNA in response to FADD knockdown in HepG2 cells. As shown in Yao et al., 2015 Figs. 4A and 4C, the expression level of FADD also decreased in cells transfected with FADD siRNA. We found that both the mRNA and protein levels of IDE were downregulated in HepG2 cells transfected with FADD siRNA fragments by qPCR and WB analyses (Figs. 4B and 4C). The mRNA level of IDE may be regulated by IDE transcription or the stability of mRNA. To further address this issue, 5 μg/ml actinomycin D (Act D) was added to HepG2 cells transfected with siFADD or NC fragments on a variable time gradient. The results showed that the stability of IDE mRNA did not change markedly after transfection with siFADD compared with the control (Fig. 4D). Thus, we concluded that FADD knockdown might affect the transcription of IDE but not the stability of IDE mRNA.

Effects of FADD phosphorylation or deficiency on IDE protein stability

To further assess the effect of FADD phosphorylation on IDE, we concentrated our subsequent analysis on IDE protein levels with cycloheximide (CHX) stimulation in primary hepatocytes from FADD-D and control mice. The results showed that the degradation rate of the IDE protein increased in cells from FADD-D mice compared to those from the control (Zhuang et al., 2016 Figs. 5A and 5B). Similarly, the degradation rate of IDE protein also increased in HEK293 cells transfected with siFADD stimulated with CHX for 24 h compared to those transfected with NC (Figs. 5C and 5D). In addition, we found that the expression levels of IDE decreased progressively following insulin stimulation at different concentrations in primary hepatocytes from FADD-D mice compared to the control (Fig. 5E). Together, these data suggested that FADD might be involved in the regulation of IDE protein stability.

FoxO1 inhibited the luciferase activity of the IDE promoter in FADD-D or FADD knockdown cells

Insulin activity in maintaining balanced glycolipid metabolism partly accounts for the phosphorylation of FoxO1 (Cheng and White, 2011). Phosphorylation of FoxO1 at Thr24, Ser256, and Ser319 (TSS) can promote its nuclear to cytoplasmic translocation and inhibit FoxO1-dependent transcription. We first aimed to explore the relationship between FoxO1 phosphorylation and the IDE promoter by luciferase activity analysis. We found that the transcription activity of the IDE promoter in HEK293T cells transfected with the pEGFP-N1-FoxO1 plasmid was the lowest, while HEK293T cells transfected with the pEGFP-N1 plasmid had the highest transcription activity (Fig. 6A), indicating that FoxO1 might be critical to block the transcription of the IDE promoter. Interestingly, cells transfected with pEGFP-N1-TSS exhibited higher transcription activity than those transfected with pEGFP-N1-FoxO1 (Fig. 6A). To further investigate the relationship between IDE and FADD, we next determined the transcriptional activity of the IDE promoter in response to FADD knockdown in HEK293T cells. As shown in Fig. 6B, by interfering with the expression of FADD in HEK293T cells, reduced luciferase activity was observed, indicating that the IDE transcription activity in HEK293T cells transfected with FADD siRNA fragments was decreased. In addition, by WB analysis, we found that the protein level of IDE was downregulated in HEK293T cells transfected with FADD siRNA fragments (Fig. 6C). To further observe the effect of FADD phosphorylation on the cellular localization of FoxO1, an immunofluorescence experiment was carried out on primary hepatocytes from FADD-D and control mice. The results showed that FADD phosphorylation promoted the movement of FoxO1 from the cytoplasm to the nucleus (Fig. 6D), suggesting that FADD phosphorylation might affect FoxO1 nuclear localization and subsequently decrease IDE transcription activity.

Effects of Sirt1 on IDE expression

It is well known that Sirt1 has a prominent role in the liver, where it can deacetylate FoxO1 proteins, resulting in a pronounced effect on glucose homeostasis and insulin secretion. Sirt1 deacetylates FoxO1 and enables the activation of FoxO1 transcription in multiple systems (Cao et al., 2016; Sin et al., 2015; Xiong et al., 2011). Here, in our study, to observe the effect of Sirt1 on IDE expression, different concentrations of the Sirt1 inhibitor EX527 and activator SRT1720 were added to HepG2 cells. The results showed that Sirt1 negatively regulated the expression of IDE in HepG2 cells. We noticed that when Sirt1 was inhibited, the expression of IDE was upregulated; on the other hand, when Sirt1 was activated, the expression of IDE was downregulated (Fig. 7A). Furthermore, Sirt1 was found to be elevated in FADD-knockdown HepG2 cells (Fig. 7B). Based on these data, we therefore speculated that FADD might regulate the expression of IDE through the Sirt1-FoxO1 signaling pathway. However, the detailed regulatory mechanisms of FADD on Sirt1 still need to be investigated in our future research.

FADD is a critical adaptor in apoptosis signaling shared by several death receptors, and its phosphorylation mediates many nonapoptotic activities (Hua et al., 2003). In our previous research, we found that FADD phosphorylation plays an important role in islet morphology and function via proteomics discovery and further physiological validation. FADD-D mice displayed increased insulin levels in serum; however, the function of insulin secretion was impaired in FADD-D mice (Yao et al., 2015). Therefore, we speculate that elevated insulin levels in the serum of FADD-D mice might not be due to abnormal synthetic progress but may be related to abnormal degradation. To address this issue, we thus focused on insulin degradation in FADD-D mice. In this study, the FADD-D mouse model was used to study the effect of FADD phosphorylation on the expression of IDE for the first time, providing new insights into the function of FADD. FADD phosphorylation or downregulation might inhibit IDE stability and affect the nuclear location of FoxO1, leading to reduced IDE expression and the subsequent slowed degradation of insulin and decreased blood glucose.

IDE is an intracellular 110 kDa thiol zinc-metalloendopeptidase that is located in the cytosol, peroxisomes, endosomes, and on the cell surface. IDE catalyzes the degradation of several small proteins, such as insulin, amylin and β-amyloid (Shen et al., 2006). Genetic association studies in humans have proven that IDE plays an important role in the pathogenesis of T2DM (Pivovarova et al., 2015). Liver-specific deletion of IDE (L-IDE-KO) in mice resulted in elevated plasma glucose levels and reduced plasma membrane insulin receptor levels in the liver (Villa-Perez et al., 2018). Once the function of IDE is impaired, the biological effects and degradation of insulin will be abnormal, thus leading to hyperinsulinemia (Rezende et al., 2014). Valid and selective modulators of IDE activity can be potential drugs for T2DM treatment (Pivovarova et al., 2016). Vettorazzi et al. (2017) used taurine conjugated bile acid (TUDCA) which can stimulate the expression of IDE in HepG2 cells to ameliorate hyperinsulinemia induced by obesity.

IDE is the principal enzyme in the insulin degradation process in the liver (Pivovarova et al., 2015; Valera Mora et al., 2003). We hypothesized that FADD phosphorylation may have an effect on the expression of IDE in the liver. Our data showed that the mRNA and protein levels of IDE were significantly decreased in FADD-D mouse hepatic tissue and the same results were also obtained in the liver primary hepatocyte cells. After the primary hepatocytes cells were stimulated by varying exposure times of CHX and varying concentrations of insulin, the protein level of IDE reduced gradually in FADD-D primary hepatocytes, but did not change significantly in control cells. Similar results were also found in HEK293 cells transfected with FADD siRNA. Together, these results indicated that FADD phosphorylation or FADD downregulation could affect the protein stability of IDE.

FoxO1 was shown to play a major role in regulating the insulin response in the liver, which is one of its critical sites of action. When FoxO1 constitutively expresses in the liver, fasting blood glucose rises (Matsumoto et al., 2006). Conversely, liver specific FoxO1 knockout mice develop fasting hypoglycemia (Matsumoto et al., 2007). FoxO1 is active in the fasted state and distributed in the nucleus, which results in the transcriptional induction of two gluconeogenic enzymes and increased hepatic glucose production (Zhang et al., 2006). In the fed state, phosphorylation of three highly conserved protein kinase B sites, corresponding to TSS in human FoxO1, suppresses transactivation and promotes nuclear exclusion of FoxO1 proteins (Lu et al., 2012). In our study, we found that FoxO1 negatively regulated on IDE promoter activity. In FADD-D primary hepatocytes, more FoxO1 moved from the cytoplasm into the nucleus, indicating that FADD phosphorylation affected the nuclear localization of FoxO1. Due to the negative regulation of FoxO1, the expression of IDE was then downregulated with subsequent suppression of gluconeogenesis and hyperinsulinemia.

Previous studies have shown that FADD is an adaptor in the Fas-FasL apoptosis signaling pathway, which takes part in the development of autoimmune diabetes by β cells (Silva et al., 2003). In addition, since FADD is located on chromosome 11q13.3, a diabetes susceptibility locus, it has been suspected that changes in FADD might be a possible contributor to type 1 diabetes (Pirot et al., 2008). Subsequent studies revealed that FADD is located outside of the type 1 diabetes mellitus region on chromosome 11q13 (IDDM4); thus it is unlikely to be the gene for IDDM4 (Eckenrode et al., 2000). Zhang et al. (2013) identified an upstream promoter element that blocks the reverse transcription of the mouse IDE promoter through promoter deletion and mutation analysis. Although there are a few reports showing that FADD is related to glucose, the specific mechanism is not yet clear. In the present study, FADD phosphorylation may reduce the expression of IDE by promoting the nuclear translocation of FoxO1. In addition, Sirt1 might also take part in regulating the expression of IDE, which then affects insulin degradation, the growth of glycogen, and the decrease in blood glucose (Fig. 7C).

Our study preliminarily explored the impact of FADD phosphorylation on IDE. However, further studies are needed; for instance, determination of the binding site of FoxO1 on the IDE promoter and how Sirt1 is involved in the regulation of insulin degradation. When HepG2 cells were treated with a Sirt1 inhibitor and activator, Sirt1 negatively regulated the expression of IDE. The detailed regulatory mechanism of FADD on Sirt1 requires further experimental confirmation.

In summary, our investigations revealed for the first time that FADD, especially in its phosphorylated form, has an impact on IDE expression. FADD phosphorylation or downregulation reduces the expression of IDE by inhibiting its transcription by promoting the translocation of FoxO1 into the nucleus and decreasing IDE protein stability. We have thereby expanded the understanding of the nonapoptotic functions of FADD and its phosphorylation. These findings implied a novel role of FADD in the reduction of protein stability and expression levels of IDE.

This work was supported by grants from the National Key R&D Research Program by Ministry of Science and Technology (2017YFA0506002), the Chinese National Natural Sciences Foundation (81773099, 81630092, 81570790, 81573338), Jiangsu Provincial Department of Science and Technology (BK20171202) and Changzhou Bureau of Science and Technology (CJ20180013).