Methylene blue inhibits NLRP3, NLRC4, AIM2, and non-canonical inflammasome activation

Methylene blue (MB), which has antioxidant, anti-inflammatory, neuroprotective, and mitochondria protective effects, has been widely used as a dye and medication. However, the effect of MB on inflammasome activation has not yet been studied. Inflammasomes are multi-protein complexes that induce maturation of interleukins (ILs)-1β and -18 as well as caspase-1-mediated cell death, known as pyroptosis. Dysregulation of inflammasomes causes several diseases such as type 2 diabetes, Alzheimer’s disease, and gout. In this study, we assess the effect of MB on inflammasome activation in macrophages. As the result, MB attenuated activation of canonical inflammasomes such as NLRP3, NLRC4, and AIM2 as well as non-canonical inflammasome activation. In addition, MB inhibited upstream signals such as inflammasome assembly, phagocytosis, and gene expression of inflammasome components via inhibition of NF-κB signaling. Furthermore, MB reduced the activity of caspase-1. The anti-inflammasome properties of MB were further confirmed in mice models. Thus, we suggest that MB is a broad-spectrum anti-inflammasome candidate molecule.

Methylene blue (MB, 3,7-bis(dimethylamino)-phenothiazin-5-ium chloride) is a heterocyclic aromatic chemical compound with the chemical formula C 16 H 18 N 3 SCl 1 . It has many uses in biology and chemistry, such as a dye for the textile industry, and has potent antibiotic and antioxidant properties. Since its discovery as the first synthetic anti-malarial agent by Ehrlich in 1891, MB has been used in several clinical fields for the treatment of acute and chronic methemoglobinemia, carbon monoxide poisoning, urinary tract infection, septic shock, and cardiopulmonary bypass 1 . MB suppresses production of superoxide radicals by acting as an alternative receptor of xanthine oxide electrons. Recently, MB has received increased attention in view of studies suggesting its usefulness in treating mitochondrial dysfunction 1 . It has also been studied as an agent for the treatment of Alzheimer's disease 2 . MB may also provide neuroprotective functions based on its anti-inflammatory properties 3 . Further, expression of inflammatory genes was reduced in microglia treated with lipopolysaccharide (LPS) in the presence of MB in the culture media 3 .
Inflammation is a protective immune response mediated by the innate immune system in response to harmful stimuli such as pathogens, damaged cells, and irritants and is tightly controlled by the host 4 . Insufficient inflammation can cause continuous infection of pathogens, whereas excessive inflammation can lead to chronic or systemic inflammatory diseases. Innate immune function depends on germline-encoded pattern-recognition receptors (PRRs) recognizing pathogen-associated molecular patterns (PAMPs) derived from infectious pathogens as well as danger-associated molecular patterns (DAMPs) induced from endogenous stress. Inflammasomes, which are multi-protein complexes, consist of cytosolic PRRs and sensing cytosolic PAMPs or DAMPs in myeloid cells as well as non-myeloid cells such as keratinocytes, hepatocytes, and cardiomyocytes [5][6][7][8][9][10] . To assemble the inflammasome complex, sensing proteins and caspase-1 are linked by an adaptor protein known as apoptosis-associated speck-like protein containing a carboxy-terminal caspase recruitment domain (Asc or pycard). The sensing proteins are nucleotide-binding oligomerization domain (NOD), leucine-rich repeat (LRR)-containing protein (NLR) family members such as NLRP1, NLRP3, and NLRC4, or absent in melanoma 2 (AIM2) 5,6 . Upon detecting certain stimuli, NLR or AIM2 can oligomerize into a caspase-1-activating scaffold. Active caspase-1 subsequently functions to cleave the proinflammatory IL-1 family of cytokines into their bioactive forms, IL-1β and IL-18, as well as induce pyroptosis, a type of inflammatory cell death 5 . In addition, it has been suggested that the non-canonical inflammasome activates caspases-4, -5, and/or -11 in response to intracellular LPS, resulting in IL-1β/-18 secretion, pyroptosis, and endotoxemic death 11 .
Although MB has been used in human and veterinary medicine for over a century, there has been no study on the role of MB on inflammasome activation. In this study, we assessed the effect of MB on several well-characterized inflammasomes such as NLRP3, NLRC4, and AIM2 and non-canonical inflammasomes in murine macrophages. In addition, we demonstrated the upstream and molecular mechanisms of MB in the context of inflammasome activation. The regulatory effect of MB were further confirmed with animal models. Thus, we conclude that MB regulates inflammasome activation and inflammatory responses.

Methylene blue inhibits NLRP3 inflammasome activation.
To assess the effect of methylene blue (MB, Fig. 1A) on IL-1β maturation resulting from inflammasome activation, LPS-primed bone marrow-derived macrophages (BMDMs) were treated with MB or ATP, a NLRP3 inflammasome trigger, as a positive control. Although ATP treatment induced IL-1β secretion resulting from NLRP3 inflammasome activation, MB alone did not (Fig. 1B). This result implies that MB alone did not activate inflammasomes. Next, we tested whether or not MB inhibits inflammasome activation. LPS-primed BMDMs were subjected to NLRP3 inflammasome activation by nigericin (NG) or ATP in the presence of increasing dosages of MB, and several readouts for inflammasome activation were observed such as secretion of maturated IL-1β and caspase-1, as well as formation of Asc pyroptosome ( Fig. 1C and Supplementary Fig. 1). As the result, MB dose-dependently attenuated secretion of IL-1β (p17) and caspase-1 (p20) as well as aggregation of Asc. In addition, we confirmed the anti-inflammasome properties of MB on NG-induced IL-1β (Fig. 1D) and IL-18 ( Fig. 1E) secretion by ELISA. These data suggest that MB inhibited assembly of the NLRP3 inflammasome, which is upstream of caspase-1, IL-1β, and IL-18 maturation. Furthermore, MSU crystals, another NLRP3 inflammasome trigger, induced caspase-1 and IL-1β secretion, which were blocked by MB treatment (Fig. 1F). The current concentration of MB did not present any cytotoxicity in BMDMs (Fig. 1G). Taken together, we suggest that MB is a putative anti-NLRP3 inflammasome agent.
Methylene blue interrupts gene expression of NLRP3 and cytokines. NLRP3 inflammasome activation requires a priming step in which toll-like receptor (TLR) ligands such as LPS induce production of the pro-forms of IL-1β and NLRP3 12,13 . To elucidate the effect of MB on the priming step, BMDMs were treated with MB with/without LPS ( Fig. 2A). MB alone did not have any effect on gene expression, although it interrupted LPS-mediated pro-IL-1β and NLRP3 production. This result suggests that MB inhibits NLRP3 inflammasome activation as well as the priming step. In addition, we elucidated the effect of MB on the mRNA expression of other cytokines such as IL-1α, IL-6, IL-10, IL-12b, and TNFα in BMDMs (Fig. 2B). In the results, MB interrupted up-regulation of cytokines in response to LPS treatment, implying that MB attenuates the LPS-TLR4 signaling pathway. We next investigated whether or not MB inhibits both the priming and activation of inflammasomes. BMDMs were separately treated with MB at the 1 st or 2 nd steps (Fig. 2C). MB treatment at the 1 st step blocked secretion of IL-1β, IL-18, and caspase-1, implying that MB interrupts priming of inflammasome activation. In addition, co-treatment of MB with NG, as a 2 nd signal trigger, did not result in maturation of IL-1β, IL-18, and caspase-1. Taken together, MB interrupts both the priming and activation of inflammasomes as well as cytokine expression.
Methylene blue attenuates mitochondrial ROS production, phagocytosis, caspase 1 activity, and NLRP3 promoter activity. Next, we investigated the molecular pathway responsible for the anti-NLRP3 inflammasome effect of MB. We first tested the effect of MB on mitochondrial reactive oxygen species (ROS) production, a well-characterized molecular mediator that triggers NLRP3 inflammasome assembly 14 . As shown in Fig. 3A, mitochondrial ROS levels in LPS-primed BMDMs were induced by rotenone treatment, which interrupts electron transport in mitochondria, and rotenone-mediated ROS production was attenuated by MB co-treatment in a dose-dependent manner. Rotenone treatment in LPS-primed BMDMs induced IL-1β secretion as expected (Fig. 3B). Rotenone-mediated IL-1β secretion was inhibited by MB similar to the effect of diphenyleneiodonium (DPI), which blocks cellular and mitochondrial ROS production. This result suggests that MB blocks mitochondrial ROS generation, resulting in NLRP3 inflammasome inhibition. Indeed, MB has been reported to ameliorate mitochondrial function as well as act as an antioxidant 15 .
We further assessed the effect of MB on phagocytosis since MB also inhibited MSU crystal-mediated NLRP3 inflammasome activation (Fig. 1F). In general, crystals must be eaten by macrophages to induce phagosome rupture, resulting in cytosolic cathepsin B release and NLRP3 inflammasome activation 16 . We treated with fluorescence-conjugated latex beads (30 nm or 1 μm) to elucidate the effect of MB on phagocytosis in LPS-primed BMDMs (Fig. 3C). In our results, MB attenuated phagocytosis, implying that MB directly blocks phagocytosis before attenuating NLRP3 inflammasome activation.
Furthermore, we determined the effect of MB on caspase-1 activity, which induces maturation of IL-1β/18. MB inhibited upstream events of inflammasome activation, such as Asc pyroptosome formation, phagocytosis, and the priming step. As shown in Fig. 3D, human recombinant caspase-1 was incubated with its substrates in the presence of MB or Z-VAD-FMK, a pan caspase inhibitor, in a tube. As the result, MB inhibited casaspase-1 (C) Secretion of IL-1β and caspase-1 (Casp1) and formation of Asc pyroptosome were analyzed by immunoblotting using Sup, Lys, and cross-linked pellets (Pellet) from whole cell lysates. The below schematic graph displays the chemical treatment process for inflammasome activation. IL-1β (D) and IL-18 (E) secretions were measured by ELISA. (F) LPS-primed BMDMs were treated with monosodium urate crystals (MSU, 800 μg/mL). Secretion of caspase-1 was analyzed by immunoblotting, and IL-1β secretion was measured by ELISA. (G) For cytotoxicity, BMDMs were treated the indicated dosages of MB, and cell number was measured by an automated cell counter. Triton x-100 (1%, Triton) treatment led to cell death. All immunoblot data shown are representative of at least three independent experiments. Bar graph presents the mean ± SD. and TNFα mRNAs were quantitated by real-time PCR. C, BMDMs were treated with MB and/or LPS as the 1 st signal, after which cells were replaced by media containing nigericin (NG, 2 nd signal) with/without MB as the 2 nd signal. IL-1β and IL-18 secretion levels were measured by ELISA, and Casp1 secretion and pro-IL1β expression were analyzed by immunoblotting. All immunoblot data shown are representative of at least three independent experiments. Bar graph presents the mean ± SD.
Scientific RepoRts | 7: 12409 | DOI:10.1038/s41598-017-12635-6 activity in a dose-dependent manner. Based on a previous report 17 , we speculate that MB inhibits the activity of caspase-1 by oxidizing the catalytic cysteine.
To elucidate the molecular mechanism of MB-mediated inhibition on the priming step, we constructed two promoter activity assaying plasmids containing luciferase under the control of the mouse NLRP3 promoter (−1,327 nucleotides (nt) to + 166 nt or −1,216 nt to + 166 nt) based on a previous study 18 . As expected, the construct possessing two NF-κB-binding sites between −1,327 nt to −1,261 nt induced relative luciferase activity (RLA) in response to LPS treatment, whereas the other construct (-1,216 nt to + 166 nt) without NF-κB-binding sites did not exhibit altered RLA upon LPS treatment (Fig. 3E). We further compared RLA of the construct (−1,327 nt to + 166 nt) containing NF-κB-binding sites in the presence of MB (Fig. 3F). In the results, elevation of RLA by LPS treatment was attenuated by co-treatment with MB. This result implies that MB inhibits the priming step of inflammasome activation and cytokine expression via attenuation of NF-κB signaling.
Methylene blue inhibits NLRC4 and AIM2 inflammasomes. We further elucidated the effect of MB on other inflammasomes such as NLRC4 and AIM2. To trigger NLRC4 inflammasome activation, LPS-primed  BMDMs were transfected with flagellin or inoculated with Salmonella typhimurium (Fig. 4A). Similar to its effect on the NLRP3 inflammasome, MB dose-dependently inhibited flagellin-or Salmonella-mediated IL-1β and Caps1 secretion. In addition, dsDNA transfection and Listeria monocytogenes infection were utilized to activate AIM2 inflammasome in LPS-primed macrophages (Fig. 4B). dsDNA-and Listeria-mediated IL-1β and caspase-1 secretions were significantly blocked by MB co-treatment. We also investigated the potential bactericidal effect of MB. As shown in Fig. 4C, MB did not interrupt growth of Salmonella or Listeria. Thus, MB inhibits activation of the NLRC4 and AIM2 inflammasomes.

Methylene blue inhibits non-canonical inflammasomes.
We assessed whether or not MB attenuates activation of the non-canonical inflammasome, which acts as an upstream signal of NLRP3 inflammasome activation and is tightly involved in LPS-induced lethality 11,19 . For non-canonical inflammasome activation, LPS-primed BMDMs were transfected with LPS ( Fig. 5D) or inoculated with E.coli (Fig. 5E). As the result, MB attenuated caspase-1 and IL-1β secretion resulting from LPS or E.coli-mediated non-canonical inflammasome activation. This result implies that MB not only inhibits the canonical inflammasome but also blocks the non-canonical inflammasome.

Methylene blue inhibits inflammasomes and cytokine expression in a human cell line, THP-1.
Although MB presented anti-inflammasome properties in murine macrophages, we further investigated whether or not MB acts similarly in human cells. We adopted a human monocyte-like cell line, THP-1, and elucidated the effect of MB on inflammasome activation. THP-1 was differentiated by phorbol 12-myristate 13-acetate (PMA) treatment and subjected to priming with LPS. LPS-primed THP-1 cells showed activation of NLRP3, NLRC4, AIM2, and non-canonical inflammasomes after NG treatment, flagellin transfection, dsDNA transfection, and LPS transfection with/without MB (Fig. 6A). In the results, MB dose-dependently attenuated IL-1β secretion resulting from NLRP3, NLRC4, AIM2, and non-canonical inflammasome activation, similar to mouse BMDMs. In addition, THP-1 cells were treated with MB with/without LPS, and gene expression of cytokines was analyzed by RT-PCR (Supplementary Fig. 2B) and real-time PCR (Fig. 6B). Up-regulation of IL-1β, IL-1α, IL-6, TNFα, and NLRP3 mRNAs in response to LPS treatment was attenuated by MB co-treatment. Thus, MB attenuated the priming and activation steps of inflammasome activation in a human cell line, THP-1, similar to murine macrophages.

Methylene blue ameliorates inflammasome-mediated diseases in mice.
We adopted two inflammasome-mediated disease models, LPS-induced lethality and Listeria peritonitis, to assess the anti-inflammasome properties of MB in mice. LPS-induced lethality, also called endotoxemic shock, is a well-defined animal model for the NLRP3 inflammasome 20 and/or non-canonical inflammasome 11 . As shown in Fig. 7A, mice injected with LPS alone died within 4 h, but additional injection of MB to LPS-treated mice resulted in an increased survival rate (P = 0.0066, Log-rank test) in a dose-dependent manner. In addition, mice were injected with LPS and/or MB, after which peritoneal fluids were collected to measure IL-1β and IL-6 secretion after 6 h of injection. In the results, LPS injection induced both IL-1β and IL-6 secretion while MB co-treatment only attenuated LPS-mediated IL-1β secretion but not IL-6 secretion ( Fig. 7B and C). This result implies that MB selectively inhibits inflammasome activation in animals, although MB blocked both cytokine production and maturation in cells. Next, mice were treated with Listeria, an AIM2 inflammasome trigger, with/without MB, after which the number of peritoneal exudate cells (PECs) and secretion of peritoneal IL-1β were determined. Although we investigated Listeria-induced lethality, our Listeria did not induce any lethality within 80 h. As shown in Fig. 7D, induction of PECs by Listeria injection was not altered by MB, implying that MB could not alter the chemotactic state. Listeria also induced peritoneal IL-1β secretion, which was attenuated by MB co-treatment (Fig. 7E). Thus, in vivo treatment with MB ameliorates LPS lethality through inhibition of NLRP3 and/or non-canonical inflammasome activation as well as Listeria-induced IL-1β production via inhibition of AIM2 inflammasome activation.

Discussion
In this study, we assessed the effect of methylene blue (MB) on canonical (NLRP3, NLRC4, and AIM2) and non-canonical inflammasome activation. We demonstrated that MB acts as an anti-inflammasome agent. Specifically, MB attenuated specific inflammasome trigger-mediated IL-1β/18 and caspase-1 secretion as well as Asc pyroptosome formation. MB also blocked mitochondrial ROS production, which triggers NLRP3 inflammasome activation, as well as NLRP3 and pro-IL-1β expression, which are essential components for inflammasome activation. In addition, MB attenuated activity of casaspe-1, which directly induces maturation of IL-1β/18. The anti-inflammasome properties of MB were further confirmed in an animal model. MB treatment reduced LPS-induced lethality and Listeria-mediated IL-1β secretion. Taken together, we suggest that MB can inhibit both the beginning and end of canonical and non-canonical inflammasome activation.
MB has long been used for its therapeutic properties, and it is considered to be an effective agent to cure septic shock 21 . Sepsis models using rat, rabbit, and dog have suggested that MB infusion during endotoxic shock consistently increases mean arterial pressure (MAP) and peripheral vascular resistance while reducing catecholamine requirements in patients 22,23 . Vasodilation is mediated by cyclic guanosine monophosphate (cGMP) produced by soluble guanylate cyclases (sGC), and nitric oxide (NO) directly activates sGC 22 . As a mechanism of MB in septic shock, it has been suggested that MB down-regulates inducible nitric oxide synthase (iNOS) since LPS induces NO production, resulting in hypotension during septic shock 22 . In the present study, MB treatment increased the survival rates of LPS-treated mice (Fig. 4A). Although we demonstrated the inhibitory effect of MB on NLRP3 and/or non-canonical inflammasome activation for reduction of LPS lethality, blockage of NO production by MB might support increased survival rates in septic mice. On the other hand, we hypothesized that NO increases IL-1β secretion via inflammasome activation. For this, we treated L-arginine, an endogenous NO precursor, to LPS-primed BMDMs and assessed IL-1β secretion. As the result, L-arginine did not induce any IL-1β secretion in LPS-primed BMDMs (Supplementary Fig. 3). Thus, we conclude that MB attenuates inflammasome activation independent of NO production.
Although MB is a well-known antioxidant agent, the effect of MB on cytokine production in macrophages has been poorly studied. Based on the literature, MB treatment was shown to reduce IL-1β and increase IL-10 gene expression in a LPS-treated microglia cell line, and MB infusion after traumatic brain injury in mice reduced expression of inflammatory genes in the hippocampus 3 . In addition, MB has been shown to attenuate expression of iNOS in response to LPS by inhibiting the binding affinity of transcription factors (NF-κB and STAT1) on the promoter region of iNOS gene 21 . In the present study (Fig. 2D), we demonstrated that MB treatment blocked the priming step of inflammasome activation as well as up-regulation of pro-IL-1β and NLRP3 proteins 12,13 . Both genes are tightly regulated by NF-κB signaling 12,13,18 , and MB might interrupt the binding of NF-κB to the promoter regions of pro-IL-1β and NLRP3 genes, as shown in previous literature 21 .
MB is an oxidation-reduction (redox) agent previously used safely in humans as an antidote for certain metabolic poisons 1 . In addition, MB prevents the formation of superoxide and nitric oxide in mitochondria and is able to improve brain oxidative metabolism by enhancing mitochondrial oxygen consumption 1 . In animal studies, MB counteracts the damaging effect of rotenone, an inhibitor of the mitochondrial electron transfer complex I, on retinal neurons 24 . Thus, MB is suggested as a potential therapeutic target for mitochondrial dysfunction. Mitochondrial dysfunction plays a determinant role in a number of acute and chronic inflammatory diseases 25 . Mitochondrial dysfunction acts upstream of NLRP3 activation by providing ROS to trigger NLRP3 oligomerization or by inducing α-tubulin acetylation to relocate mitochondria in proximity to NLRP3 14,26 . Based on our results and previous reports, we conclude that MB attenuates inflammasome activation by improving mitochondrial function.
MB is used in several diagnostic procedures as a staining agent, including bacteria staining and intraoperative tissue staining 27 . For example, MB and epinephrine-containing saline are injected into the submucosal layer around a polyp during endoscopic polypectomy. MB helps to identify submucosal tissue after the polyp is removed, which is useful for determining whether more tissue should be removed or if there is a high risk for perforation 28 . MB is also used as a dye in chromoendoscopy and is sprayed onto the mucosa of the gastrointestinal tract to identify dysplasia or pre-cancerous lesions 29 . In surgeries such as sentinel lymph node dissections, MB can  (PECs, B) was calculated, and IL-1β (C) secretion levels of peritoneal lavage fluids were measured. Bar graph presents the mean ± SD. be used to visually track lymphatic drainage of involved tissues 30 . Likewise, MB is added to bone cement during orthopedic surgery to easily distinguish cement from native bone 31 . Thus, MB is used conservatively during the above procedures as a staining and/or coloring agent. However, we suggest that MB has an additional effect than just a staining molecule based on its ability to attenuate inflammasome activation and cytokine expression. Thus, the anti-inflammatory properties of MB may prevent unwanted complications after surgical procedures.
Inflammasome dysregulation has been implicated in neurologic disorders and metabolic diseases, neither of which are traditionally considered to be inflammatory diseases but which are increasingly recognized as having an inflammatory component that significantly contributes to the disease process and drives many forms of cancer in humans 5 . Therefore, researchers have become interested in the regulation of inflammasome activation. So far, several reagents such as recombinant IL-1 receptor antagonist (anakinra), neutralizing IL-1β antibody (canakinumab), soluble decoy IL-1 receptor (rilonacept), IL-18-binding protein, soluble IL-18 receptors, and anti-IL-18 receptor monoclonal antibodies have been developed and applied to control inflammasome-mediated diseases 5 . These reagents only control events downstream of inflammasome activation such as blockage of IL-1β/-18 signaling. However, we have attempted to screen natural compounds that selectively control events upstream of inflammasome activation [32][33][34][35][36][37][38] . Based on our finding, MB has the most wide range of anti-inflammasome agents and controls several events upstream of inflammasome activation. Specifically, MB blocks the NLRP3, NLRC4, and AIM2 inflammasomes as well as non-canonical inflammasome. In addition, MB attenuates crystal phagocytosis, the priming step of inflammasome activation, Asc speck formation, and caspse-1 activation.

Materials and Methods
Cell culture. Bone marrow-derived macrophages (BMDMs) were cultured as described in detail elsewhere 38,39 .
In brief, femur and tibia bones from C57BL/6 mice (6-12-weeks-old; Narabio Co., Seoul, Republic of Korea) were collected, and bone marrow cells from all bones were flushed out. Cells were than cultured in Dulbecco's Modified Eagle Medium (DMEM; WELGENE Inc. Daegu, Republic of Korea or Capricorn Scientific GmbH, Ebsdorfergrund, Germany) supplemented with 10% fetal bovine serum (FBS; Corning cellgro, Manassas, VA, USA or Capricorn Scientific GmbH) and 1% penicillin and streptomycin solution (P/S; Corning cellgro, 10,000 I.U. of penicillin and 10,000 µg/mL of streptomycin) in L-929 cell-conditioned medium containing granulocyte/macrophage colony-stimulating factor. Cells were plated in non-tissue culture-treated Petri dishes (SPL Life Science Co., Phcheon-si, Gyeonggi-do, Republic of Korea) and incubated at 37 °C in 5% CO 2   To determine the effect of MB on the priming step, BMDMs or THP-1 were treated with MB (0, 20, 100, or 200 μM) with/without LPS (10 ng/mL) for 30 min, after which cells were given fresh media with/without LPS (10 ng/mL) for 2.5 h. Details of the treatment are presented in the bottom of Fig. 2A. Western blotting sample preparation. After inflammasome activation, cellular supernatant (Sup; 350 μL of RPMI 1640) was transferred into a new tube, and remaining BMDMs were lysed with 100 μL of mild lysis buffer (150 mM NaCl, 1% Triton X-100, 50 mM Tri-base, pH 8.0) containing proteinase inhibitor cocktail (#M250-1, AMRESCO LLC, Solon, OH, USA) 36,37 . The lysate (Lys) was transferred into a new tube and collected by centrifugation at 15,000 rcf for 5 min. The remaining pellet was washed two times with PBS and then re-suspended and cross-linked with 2 mM suberic acid bis (Sigma-Aldrich Co.) for 1 h, followed by centrifugation at 15,000 rcf for 5 min. The cross-linked pellets (Pellet) were re-suspended in 50 μL of 2 X loading dye buffer (116 mM Tris, 3.4% SDS, 12% glycerol, 200 mM DTT, 0.003% bromo phenol blue) 35 . Sup, Lys, and Pellet were subjected to Western blot assay.