Cutting Edge: Negative Regulation of Inflammasome Activation by TRAF1 Can Limit Gout

Gout is a common form of acute arthritis that dramatically reflects the cardinal features of inflammation. Local monosodium urate (MSU) crystal deposition in the joint initiates a rapid influx of neutrophils and monocytes. This process is initiated, in part, by the secretion of the potent proinflammatory cytokine, IL-1β, and is mediated by the NLRP3 inflammasomes (1, 2). On receiving a priming signal (i.e., signal 1) from pathogen-associated molecular patterns (e.g., LPS) and an activation signal (i.e., signal 2) from danger-associated molecular patterns (e.g., MSU crystals, extracellular ATP, or the bacterial toxin nigericin), NLRP3 proteins self-oligomerize and recruit the adaptor protein ASC, which itself oligomerizes to form a platform for the recruitment of procaspase-1. This causes the activation of caspase-1, which then promotes the proteolytic cleavage of pro–IL-1β into its biologically active forms (3). Notably, ASC is required for IL-1β production after activation of other inflammasomes, including AIM2 (4) and NLRC4 (5) inflammasomes.

Dysregulated inflammasome activation, especially NLRP3, is implicated in the pathogenesis of numerous inflammatory arthritides, including gout (6). Hence tight regulation of NLRP3 inflammasome activity is critical to prevent disease. As a result, posttranslational modification of inflammasome components has emerged as a key regulatory mechanism to control inflammasome activity (7). For instance, linear ubiquitination of ASC by the linear ubiquitin chain assembly complex is required for NLRP3 inflammasome activation (8). Moreover, TFNR-associated factor 1 (TRAF1) is a key signaling adaptor protein that is involved in TNFR and TLR signaling pathways (9). Genome-wide association studies have shown that single-nucleotide polymorphisms in TRAF1 are linked to increased risk of rheumatoid arthritis (10, 11). A later study showed that TRAF1 acts as a negative regulator of TLR receptor signaling by limiting linear ubiquitination of NEMO, the regulatory subunit of the IKK complex, and subsequent NF-κB activation and proinflammatory cytokine production (12).

Based on the regulatory role of linear ubiquitin chain assembly complex on the NLRP3 inflammasome, we hypothesized that TRAF1 may be implicated in ASC-dependent inflammasome assembly by regulating linear ubiquitination of ASC. These studies may facilitate the development of novel therapeutics for inflammasome-driven inflammatory diseases.

THP-1 human monocytic leukemia cells (Sigma) were cultured, and primary bone marrow–derived macrophages (BMDMs) were cultured as previously reported (13).

MSU crystals preparation was done by dissolving 1 g uric acid (Sigma) in 200 ml boiling ddH₂O water containing 1 N NaOH (pH 7.2). Endotoxins were removed by autoclaving the solution for 6 h at 120°C, and MSU crystals were confirmed to be free of detectable endotoxin contamination (<0.02 endotoxin units/ml) by the limulus amebocyte lysate assay (Pierce). MSU crystals were assessed for appropriate crystal shape and birefringence before and after autoclaving.

C57BL/6NCrl mice (Charles River) were crossed with TRAF1^−/− mice, and F1 progeny were bred for all in vivo experiments. Six- to eight-week-old female littermate mice were injected i.p. with either PBS or 5 mg/ml MSU crystals. Cells were then collected from the peritoneal lavage (3–5 ml PBS) and evaluated by flow cytometry. Supernatants were used to measure active IL-1β by ELISA (ThermoFisher). For MSU crystal–induced arthritis, 10- to 12-wk-old male littermate mice were injected intra-articularly into the knee joint with either PBS or 0.5 mg MSU crystals into contralateral sides (14,15). Knee thickness was measured and after 4 h, mice were euthanized, serum was collected for measurement of IL-1β by ELISA, synovial tissues were dissected, and knee joints were fixed and stained for histology with hematoxylin (Sigma) and eosin (ThermoFisher). All animals were housed under specific pathogen-free conditions, and animal-related procedures were approved by York University animal care committee in accordance with Canadian Council on Animal Care regulations (protocol 2019-12).

Supernatants and whole-cell extracts from inflammasome-stimulated THP-1 and BMDMs were probed with Abs to detect human/mouse IL-1β (R&D Systems); mouse caspase-1 (BioLegend); human caspase-1 (Millipore); and TRAF1 (45D3), TRAF2 (C192), and TRAF3 (E8H3B) (Cell Signaling). For linear ubiquitin immunoprecipitations, whole-cell lysates were incubated with protein A/G beads (Pierce) and 2.5 μg linear ubiquitin Ab (clone 1F11/3F5/Y102L, obtained under Material Transfer Agreement from Genentech), as previously described (12), resolved on a 4–15% gel, and then probed with ASC PolyAb (Proteintech). For ASC oligomerization assay, cells were lysed in ASC polymerization lysis buffer, and pellets were cross-linked with disuccinimidyl suberate solution (Sigma) and probed with ASC Ab, as described previously (16).

All statistical analysis was done using GraphPad software (Prism) using two-way ANOVA for comparison of multiple groups, or unpaired t test (nonparametric Mann–Whitney U test) for two groups, with p values as indicated in the figure legends.

To determine whether TRAF1 can regulate inflammasome activation, we activated the NLRP3 inflammasome in THP-1 cell lines that have been transduced with short hairpin RNA targeting TRAF1 (shTRAF1) or a control short hairpin RNA (shCtrl) (Supplemental Fig. 1A, 1B). Partial reduction of TRAF1 expression (shTRAF1) led to a significant increase in caspase-1 cleavage (Fig. 1A) and its proteolytic activity (Fig. 1B). Moreover, shTRAF1 cells secreted higher levels of active IL-1β compared with shCtrl cells (Fig. 1C, 1D). We also checked whether TRAF1 is important for controlling other ASC-dependent inflammasomes. Indeed, shTRAF1 THP-1 cells secreted higher levels of IL-1β when following activation of AIM2 (Fig. 1D) and NLRC4 (Fig. 1E) inflammasomes. Furthermore, IL-1β secretion after activation of noncanonical inflammasome by Escherichia coli outer membrane vesicles, which relies on ASC (17, 18), was similarly increased in shTRAF1 THP-1 cells (Fig. 1F). This suggests that TRAF1 is inhibiting a common component of these inflammasomes, likely through ASC oligomerization. Remarkably, shTRAF1 THP-1 cells had a significant increase in ASC oligomer formation compared with shCtrl cells (Fig. 1G). Next, we asked whether TRAF1 negatively regulates linear ubiquitination of ASC. After activation of NLRP3 inflammasome, shTRAF1 THP-1 cells showed enhanced linear ubiquitination of ASC compared with control cells (shCtrl), as evident by an increase in high-m.w. bands (Fig. 1H). Taken together, these data show that TRAF1 restricts inflammasome activation and IL-1β production by negatively regulating linear ubiquitination of ASC and subsequent oligomer formation. Importantly, TRAF1 knockdown did not cause a significant change in intracellular inactive forms of IL-1β or caspase-1 (Supplemental Fig. 1C), or in other TRAF proteins, such as TRAF2 and TRAF3 (Supplemental Fig. 1D).

To explore the ability of TRAF1 to regulate inflammasome activation in vivo, we first tested inflammasome activation in BMDMs prepared from wild type (WT) and TRAF1 knockout (TRAF1^−/−) littermate mice. After inflammasome activation, secretion of active IL-1β and of caspase-1 were significantly higher in TRAF1^−/− BMDMs compared with WT controls (Fig. 2A, 2B). Intracellular levels of pro–IL-1β and procasapse-1 were slightly increased in the absence of TRAF1 (Supplemental Fig. 2A). Next, the in vivo model of NLRP3 inflammasome activation, MSU-induced peritonitis, was employed in WT and TRAF1^−/− littermate mice. i.p. injection of MSU crystals, absent priming (i.e., signal 1), in TRAF1^−/− mice led to an increase of active IL-1β protein in the peritoneum (Fig. 2C) and a significant enhancement in the recruitment of total leukocytes, neutrophils, monocyte-derived macrophages, and peritoneal macrophages, when compared with WT littermates (Fig. 2D, Supplemental Fig. 2B). Collectively, we show that TRAF1 can control inflammasome activation in vitro and in vivo.

TRAF1 has been linked to increased risk for inflammatory arthritis, and thus we employed MSU crystal–induced arthritis, an NLRP3-inflammasome–dependent model of gout (19), in WT (TRAF1^+/+), heterozygote (TRAF1^+/−), and TRAF1 knockout (TRAF1^−/−) mice. Intra-articular injection of MSU crystals into the knee joints caused significantly more swelling in TRAF1^−/− mice than in TRAF1^+/− and WT littermates (Fig. 3A, 3B). This was accompanied by increased infiltration of inflammatory cells into the synovium of TRAF1^−/− mouse knee joints (Fig. 3C, 3D). Similarly, induction of gene expression of proinflammatory cytokines (IL-1β, TNF-α, and IL-6) and inflammatory mediators (S100A8 and S100A9), but not inflammasome components (ASC, AIM2, caspase-1), after intra-articular injection of MSU crystals was higher in TRAF1^−/− mice (Fig. 3E, Supplemental Fig. 3D–E). Similar results were observed when mice were injected with a mixture of MSU crystals and C16:0 palmitic acid (Supplemental Fig. 3A–C). Finally, serum IL-1β was significantly more elevated in TRAF1^−/− mice compared with heterozygote and WT littermates (Fig. 3F). These results show that TRAF1-mediated control of inflammasome activation is important to prevent exacerbated gout flares.

Despite the mounting evidence from genome-wide association studies that linked TRAF1 single-nucleotide polymorphisms to increased risk for rheumatoid arthritis (10, 11), a direct involvement of TRAF1 in driving inflammatory arthritis is yet to be proved. These results demonstrate that TRAF1 knockout mice were significantly more susceptible to a model of MSU crystal–induced arthritis than WT littermates. Importantly, triggering inflammasome activation with MSU crystal, which does not require priming with LPS/PMA, also led to increased inflammasome activation in TRAF1-deficient cells. This indicated that the role of TRAF1 in controlling the inflammasome activation is independent from our previous finding regarding its role in limiting LPS-induced sepsis (12). TRAF1 was also shown to play a key role in reducing TNF-induced apoptosis (20), so it is possible that cell death in TRAF1-deficient cells might have contributed to increased inflammasome activation.

Overall, we believe that our data clearly establish a, to our knowledge, novel role for TRAF1 in regulating inflammasome activation in in vitro and in vivo models of inflammasome-driven peritonitis and gout and elucidate the mechanism of action.

The authors have no financial conflicts of interest.