ASKA technology-based pull-down method reveals a suppressive effect of ASK1 on the inflammatory NOD-RIPK2 pathway in brown adipocytes﻿

Recent studies have shown that adipose tissue is an immunological organ. While inflammation in energy-storing white adipose tissues has been the focus of intense research, the regulatory mechanisms of inflammation in heat-producing brown adipose tissues remain largely unknown. We previously identified apoptosis signal-regulating kinase 1 (ASK1) as a critical regulator of brown adipocyte maturation; the PKA-ASK1-p38 axis facilitates uncoupling protein 1 (UCP1) induction cell-autonomously. Here, we show that ASK1 suppresses an innate immune pathway and contributes to maintenance of brown adipocytes. We report a novel chemical pull-down method for endogenous kinases using analog sensitive kinase allele (ASKA) technology and identify an ASK1 interactor in brown adipocytes, receptor-interacting serine/threonine-protein kinase 2 (RIPK2). ASK1 disrupts the RIPK2 signaling complex and inhibits the NOD-RIPK2 pathway to downregulate the production of inflammatory cytokines. As a potential biological significance, an in vitro model for intercellular regulation suggests that ASK1 facilitates the expression of UCP1 through the suppression of inflammatory cytokine production. In parallel to our previous report on the PKA-ASK1-p38 axis, our work raises the possibility of an auxiliary role of ASK1 in brown adipocyte maintenance through neutralizing the thermogenesis-suppressive effect of the NOD-RIPK2 pathway.


Surface plasmon resonance assay
Setting 1NA-PP1-Lx and ASK1 KD as a ligand and an analyte, respectively, the SPR signal was measured with Biacore T100 (GE Healthcare). 1NA-PP1-L1 or 1NA-PP1-L2 was immobilized to a CM5 sensor chip. GST-tagged kinase domain recombinants of mouse wild-type ASK1 and as-ASK1 (V745L/M761A/S828A) were purified as previously described 21 and injected with a series of concentrations. The obtained sensorgrams were fitted to the bivalent analyte model because the analyte ASK1 KD can be dimerized in solution 20 . The dissociation constant for the first phase (KD1) was calculated as the dissociation rate for the first phase (kd1) divided by the association rate constant for the first phase (ka1).

Gel filtration column chromatography
To fractionate the as-ASK1 signalosome in the brain, whole brains were isolated from six Ask1 ASKA knock-in mice and triturated with a Dounce homogenizer in H buffer (50 mM HEPES-KOH (pH 7.5), 10 mM KCl, 1 mM EDTA, 1 mM EGTA, 1.5 mM MgCl2) containing 0.2% Triton X-100 and 150 mM NaCl. To fractionate the as-ASK1 signalosome in primary brown adipocytes, differentiated primary brown adipocytes were collected by scraping with H buffer containing 1% digitonin (Wako Pure Chemicals Industry, #044-02121) and homogenized with a pellet mixer (TreffLab). The homogenate was clarified by centrifugation, and the supernatant was filtered through a Millex-HV syringe filter unit (0.45 µm), PVDF (33 mm), and gamma sterilized (Millipore, #SLHV033RB) twice. The clarified homogenate was processed by gel filtration column chromatography using AKTA (GE Healthcare) with a Superose 6 10/300 GL column (GE Healthcare) and H buffer.
In Fig. 1g, the sample "after purification" was processed for the following section "purification of the as-ASK1 signalosome" prior to gel filtration column chromatography. For sampling for immunoblotting analysis, each fraction was precipitated using ethanol by incubation at −20 °C overnight and subsequent sedimentation through centrifugation. After air-drying for an hour, the samples were dissolved in SDS sample buffer (80 mM Tris-HCl pH 8.8, 80 μg/mL bromophenol blue, 28.8% glycerol, 4% SDS and 10 or 20 mM dithiothreitol).
Preparation of 1NA-PP1-Lx-immobilized beads N-Hydroxysuccinimide (NHS) FG beads (TAMAGAWA SEIKI, #TAS8848 N1141) were sonicated in N,N-dimethylformamide (DMF; Sigma-Aldrich, #227056) several times and subsequently incubated with 130 µM 1NA-PP1-Lx for an hour at room temperature. The beads were shaken in 10% volume/volume ethanolamine (Sigma-Aldrich, #398136) in DMF at room temperature for 2 h, and the supernatant was replaced with 2 M Tris-HCl (pH 8.0). After sonication, the beads were shaken for another 2 h at room temperature. The medium was replaced with 10% volume/volume MeOH (Wako Pure Chemicals Industries, #137-01823) in H buffer, and the beads were sonicated. The prepared beads were stored at 4 °C. Prior to pull-down analysis, beads were sonicated again with H buffer and resuspended with H buffer.

Purification of the as-ASK1 signalosome
The fractionated samples or the clarified cell lysates of the as-ASK1 signalosome were incubated with 1NA-PP1-Lx-immoribilized beads for 16 h at 4 °C with gentle shaking, followed by addition of Washing H buffer (H buffer containing 0.2% Triton X-100). The beads were clarified by centrifugation followed by the addition of lysis buffer, and washed with Washing H buffer four times using a magnetic stand (TAMAGAWA SEIKI, #TA4899N10). The as-ASK1 complex was eluted from the beads by 1 mM 1NA-PP1 or DMSO diluted in Washing H buffer for 2 h incubation on ice with tapping every 15 min. After the elution step, the supernatants were purified using a magnetic stand.
In Fig. 1f, for sampling for immunoblotting analysis, the supernatants or beads were sampled by adding SDS sample buffer.

Coimmunoprecipitation assay
The supernatants of cell extracts were incubated with anti-Flag-tag antibody-immobilized beads (Wako Pure Chemicals Industries, clone 1E6, #016-22784) for 10-30 min at 4 °C or incubated with anti-RIPK2 (#ab75257) antibody followed by the addition of Dynabeads Protein G (Invitrogen, #DB10004). The beads were washed with lysis buffer four times, followed by the direct addition of SDS sample buffer.

TUBE pull-down assay
The K63-linked polyubiquitin conjugates were purified using K63 TUBE-coupled magnetic beads (LifeSensors, #UM404M)) following the manufacturer's instructions. In brief, cells were lysed in lysis buffer, and the supernatants of the cell extracts were incubated with TUBE for 3 h at 4 °C. The beads were washed three times with chilled TBS-T. The samples were dissolved in SDS sample buffer.

Supplementary Figure
Supplementary Figure S1. Systematic analysis of the effects of ASK1 knockdown on cytokine secretion from NOD-RIPK2 pathway-activated cells.

S-8
Supplementary Figure S2. ASK1 does not suppress the NOD-RIPK2 pathway and cytokine induction in white adipocytes.
(a) Effects of ASK1 knockdown on NOD-RIPK2 pathway activation in white adipocytes. 3T3-L1 cells were differentiated into white adipocytes and treated with C12-iE-DAP (10 µg/mL) for the indicated times. Note that cycloheximide (50 µg/ml) was treated to prevent the rapid feedback synthesis of IκBα. (b) Relative mRNA levels of inflammatory cytokines under NOD-RIPK2 pathway activation. 3T3-L1 cells were differentiated into white adipocytes and stimulated with C12-iE-DAP (10 µg/mL, 6 h). Data are represented as the mean ± SEM. n = 8-10 (pooled from 5 independent experiments). n.s.: not significant according to one-tailed Dunnett's test. S-9 Supplementary Figure S3. Inhibition of the NOD-RIPK2 pathway contributes to maintenance of thermogenic potential in brown adipocytes.
(a) Experimental model for evaluating the paracrine effect of the NOD-RIPK2 pathway on brown adipocyte marker expression. Control or ASK1 knockdown HIB 1B cells (referred to as "donor cells") were treated with C12-iE-DAP. After changing the medium to avoid contamination of remnant C12-iE-DAP in the conditioned medium, the cells were incubated for 9 h in total. Then, the conditioned medium from the donor cells was used to treat another set of HIB 1B cells ("acceptor cells") for 6 h, followed by administration of 0.1 µM of a β3-adrenergic receptor agonist CL316,243 for 12 h to induce the thermogenic genes in the acceptor cells. (b) Intercellular effects of the NOD-RIPK2 pathway on brown adipocyte marker induction. Acceptor HIB 1B cells were incubated with the conditioned medium from donor HIB 1B cells or donor cell-absent wells, with or without stimulation with C12-iE-DAP, and subsequently stimulated with CL316,243. Thermogenic marker induction in the acceptor cells was evaluated with quantitative RT-PCR. (c) Effects of ASK1 knockdown in NOD-RIPK2 pathway-activated donor cells on brown adipocyte marker induction in acceptor cells. Acceptor HIB 1B cells were incubated with the conditioned medium from ASK1-knockdown donor HIB 1B cells and subsequently stimulated with CL316,243. Thermogenic marker induction in the acceptor cells was evaluated with quantitative RT-PCR. Data are represented as the mean ± SEM, n = 9 (pooled from 3 independent experiments) in (b), n = 11-12 (pooled from 3 independent experiments) in (c). *P < 0.05, **P < 0.01, ***P < 0.001. n.s.: not significant according to two-tailed Student's t-test with the Bonferroni correction (b) or two-tailed Dunnett's test (c). b a c