Gefitinib (GF), the tyrosine kinase inhibitor (TKI) targeting epidermal growth factor receptor, initiates lung inflammation through the NLR family pyrin domain containing 3 (NLRP3) inflammasome. However, the molecular targets and mechanisms underlying the inflammatory action of GF remain unknown. In this study, we identified mitochondrial Src family kinases (mSFKs) as key determinants of GF-induced NLRP3 inflammasome activation. Comprehensive analysis of the TKIs revealed that all TKIs we tested act as potent agonists for the NLRP3 inflammasome in human monocytic THP-1 cells and bone marrow–derived macrophages. Moreover, these TKIs share a common off-target activity against the mSFKs, such as c-Src, Fgr, and Fyn. Interestingly, loss of each kinase spontaneously stimulated the NLRP3 inflammasome activation in THP-1 cells. These results together suggest that NLRP3 senses hypoactivity of the mSFKs that is responsible for mitochondrial dysfunction. Thus, our findings demonstrate a mechanistic link between the NLRP3 inflammasome and mSFKs, which, to our knowledge, provides insights into a novel molecular basis and cellular function of the NLRP3 inflammasome.

In response to pathogen infection, endogenous metabolites, and various environmental chemicals, the NLR family pyrin domain containing 3 (NLRP3) is activated and then forms a multiprotein complex with apoptosis-associated speck-like protein containing CARD (ASC) and the cysteine protease caspase-1, called the NLRP3 inflammasome (1–6). On the NLRP3 inflammasome, caspase-1 is processed and activated, leading to the maturation and secretion of the inflammatory cytokines IL-1β and IL-18 (7). Although the NLRP3 inflammasome plays an essential role in innate immunity, inappropriate activation of the NLRP3 inflammasome is known to contribute to the development of diverse inflammatory diseases, such as arteriosclerosis, gout, type 2 diabetes mellitus, and Alzheimer’s disease (8–12). However, precise mechanisms of the NLRP3 inflammasome activation are not fully understood, although accumulating evidence indicates that its activation is closely associated with mitochondrial dysfunction.

A series of tyrosine kinases, such as epidermal growth factor receptor (EGFR), plays critical roles in cell proliferation and survival, and its gain-of-function mutations are often seen in cancer cells (13, 14). Therefore, the exploitation of tyrosine kinase inhibitors (TKIs) has been considered as a promising therapeutic strategy for malignant cancers (15). To date, a wide variety of TKIs, including imatinib (IM), gefitinib (GF), erlotinib (ER), sorafenib, sunitinib, and dasatinib, are approved for the treatment for patients with advanced cancers (15, 16). Although TKIs exert excellent therapeutic effects, serious inflammatory-based side effects of TKIs, such as interstitial pneumonitis, hinder appropriate use of TKIs and limit their therapeutic efficacy. Our recent study has revealed that the EGFR TKI GF causes the NLRP3 inflammasome activation through mitochondrial damage, which is responsible, at least in part, for GF-induced interstitial pneumonitis (17, 18). However, it is unknown how GF initiates mitochondrial damage and whether other TKIs share this mechanism.

Src family kinases (SFKs) are nonreceptor tyrosine kinases that regulate in cell differentiation, proliferation, survival, adhesion, morphology, and motility (19). Eleven types of SFKs are present in the human genome, eight of which (c-Src, Fyn, Lyn, Hck, Yes, Blk, Fgr, and Lck) are well characterized (20). Moreover, of the SFKs, Fgr, Fyn, Lyn, and c-Src are localized in mitochondria and play an essential role in mitochondrial functions (21–24). In this study, we found that all TKIs we tested nonspecifically inhibit the kinase activity of the mitochondrial SFKs (mSFKs), which is responsible for the activation of the NLRP3 inflammasome. Therefore, the off-target activity against mSFKs appears to trigger the inflammatory-based side effects.

THP-1 cells from Japanese Collection of Research Bioresources Cell Bank were cultured in RPMI 1640 containing 10% heat-inactivated FBS, 1% penicillin-streptomycin solution, and Plasmocin at 37°C under a 5% CO2 atmosphere. For experiments, THP-1 cells were differentiated for 3 h with 100 nM PMA on the day before stimulation. Human embryonic kidney 293A (HEK293A) and A549 cells were cultured in DMEM containing 5% heat-inactivated FBS and 1% penicillin-streptomycin solution at 37°C under a 5% CO2 atmosphere. Bone marrow–derived macrophages (BMDMs) were isolated from mouse femurs in sterile RPMI 1640 and were cultured in RPMI 1640 containing 10 ng/ml M-CSF (Sino Biological) and 10% heat-inactivated FBS and 1% penicillin-streptomycin solution (25). For experiments, BMDMs were primed for overnight with 100 ng/ml LPS.

FLAG–c-Src, FLAG–Fyn, FLAG–Fgr, FLAG–NLRP3, FLAG–ASC, FLAG–caspase-1, and FLAG–IL-1β were inserted into pcDNA3.2 plasmids. cDNAs encoding human c-Src, Fyn, Fgr, NLRP3, ASC, caspase-1, and IL-1β were obtained by performing PCR and were inserted into pcDNA3.2 with FLAG tag plasmids. Plasmid transfection was performed using Polyethylenimine “Max” (Polysciences), according to the manufacturer’s instructions.

Knockout (KO) cells were generated using the CRISPR/Cas9 system as previously described (26). Guide RNAs were designed to target a region in exon 1 of NLRP3 gene (5′-CTGCAAGCTGGCCAGGTACC-3′) and exon 4 of EGFR gene (5′- CTTTCTCAGCAACATGTCGATGG-3′) using CRISPRdirect. These KO cells have been characterized in previous studies (17, 27). Guide RNA–encoding oligonucleotide was cloned into lenti-CRISPRv2 plasmid (Addgene), and KO cells were established as previously described. To determine the mutations of each gene in cloned cells, we analyzed the genomic sequence around the target region by PCR-direct sequencing using extracted DNA from each clone as a template and the following primers: 5′-AGTGTGGACCGAAGCCTAAG-3′ and 5′-TTCTCCTCCCCATTGAAGTC-3′ for NLRP3 and 5′-GGGCTAATTGCGGGACTCTT-3′ and 5′-CCCAGTGCTGTAGAGCTGTC-3′ for EGFR.

c-Src, Fyn, Fgr, Lyn, gasdermin D (GSDMD), or control shRNA (pLKO.1) lentivirus vector (MISSON shRNA; Sigma) was cotransfected with the VSV-G envelope and psPAX2 packaging plasmids (Addgene) into HEK293A cells using Polyethylenimine “Max” (Polysciences) as previously described (5). After 48 h, lentivirus-containing supernatants were harvested and centrifuged at 1000 rpm for 5 min to discard the debris. THP-1 cells were infected with the virus for 24 h and then selected by 1 μg/ml puromycin for 72 h. Knockdown efficacy was analyzed by immunoblotting.

All reagents were obtained from commercial sources: GF, ER, H2O2, sodium orthovanadate, JC-1, SP600125, and U0126 (Wako); lapatinib (LP) and osimertinib (OS; Chemscene); pazopanib (PZ), axitinib (AX), saracatinib, and bosutinib (Selleck); afatinib (AF), regorafenib (RG), and IM (Tokyo Chemical Industry); PP2 (Adipogen); MCC950 and dasatinib (Cayman); Ultrapure LPS (Invivogen); and PMA, SB203580, and Mito-TEMPO (Santa Cruz). The Abs used were against NLRP3 (AG-20B-0014) (Adipogen); IL-1β (#D3A3Z, #12242), Fyn (#4023T), Fgr (#27555), IκBα (#9242), and p-p65 (#3033) (Cell Signaling); ASC (cat# D086-3) (MBL); FLAG (F7425) (Sigma); and caspase-1 (M-20, A-19), ASC (B-3), GSDMD (64-Y), c-Src (B-12), p-c-Src (Y419, 9A6), β-actin (C4), EGFR (A-10), p-Tyr (PY20), α-tubulin (B-7), Tom20 (F-10), p-STAT3 (B-7), STAT3 (F-2), Lyn (H-6), and p65 (C-20) (Santa Cruz).

Proteins from cell culture supernatants were extracted by methanol/chloroform precipitation. In brief, cell-free supernatant was mixed with methanol/chloroform at a 5:5:1 (cell culture supernatant/methanol/chloroform) ratio. The mixture was vortexed and centrifuged for 10 min at 15,000 rpm. The clear upper phase was discarded, and 1000 μl methanol was added to the interphase. The mixture was centrifuged for 10 min at 15,000 rpm, and the liquid phase was removed. The protein pellet was dried and resuspended with 8 M urea. Cells were lysed in ice-cold lysis buffer containing 20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 10 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, and 1% protease and phosphatase inhibitor mixtures (Nacalai Tesque). Samples were resolved by SDS-PAGE and analyzed as described previously (28).

Cells were incubated in ice-cold hypotonic buffer containing 50 mM HEPES (pH 7.0), 1 mM EDTA, 1 mM DTT, 1 mg/ml BSA, and 1% protease inhibitor cocktails (Nacalai Tesque) for 5 min. For cell lysis, cell suspensions were added to 0.03% digitonin and then incubated on ice for 5 min and vortexed every minute. Cell lysates were added to 2× isotonic buffer containing 100 mM HEPES (pH 7.0), 2 mM EDTA, 1.2 M d-sorbitol, 1 mM DTT, 1 mg/ml BSA, and 1% protease inhibitor cocktails and then centrifuged at 4°C at 5000 × g for 15 min. The supernatants containing cytoplasmic fraction were collected and removed. After washing two times with 1× isotonic buffer, the pellets were resuspended in ice-cold lysis buffer containing 20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% Triton X-100, 10% glycerol, and 1% protease inhibitor cocktails for 10 min. Cell lysates were then centrifuged at 4°C at 15,000 rpm for 15 min, and then the supernatants were collected as mitochondrial fractions.

Concentrations of IL-1β in cell culture supernatants were measured by specific ELISA kits (Thermo Fisher Scientific) according to the manufacturer’s instructions.

Mitochondrial membrane potential was measured by JC-1 MitoMP Detection Kit (MT09) (Wako) according to the manufacturer’s instructions. Cells were treated with indicated reagents and then incubated with 2 µM JC-1 dye for 30 min at 37°C. After incubation, the cells were measured for fluorescence intensity by SPECTRA max GEMINI XPS (Molecular Devices) as described previously (29). JC-1 dye is captured by mitochondria and exhibits strong red fluorescence. However, loss of mitochondrial membrane potential (MMP) blocks the capture, and then JC-1 exhibits green fluorescence. Thus, reduction of MMP can be measured by the ratio of the red/green fluorescence.

Cells were treated with indicated reagents and then incubated with 5 µM MitoSOX Red Mitochondrial Superoxide Indicator (Invitrogen) for 30 min at 37°C according to the instructions. After incubation, the cells were suspended and measured for fluorescence intensity by Cytoflex flow cytometer (Beckman Coulter) as described previously (30).

Immunocytochemistry was performed as described previously (31). Cells were fixed with 3.7% formaldehyde, permeabilized with 0.5% Triton X-100, blocked with 3% BSA-PBS, and incubated with primary Abs overnight at 4°C, followed by incubation with secondary Abs (Alexa Fluor 555; Invitrogen) for 1 h at room temperature. The immunostained samples were enclosed with Fluoro-KEEPER Antifade Reagent and Non-Hardening Type with DAPI (Nacalai Tesque) and then observed with KEYENCE BZ-X810 all-in-one fluorescence microscope.

Cell death was monitored by using Lactate Dehydrogenase (LDH)-Cytotoxic Test Kit (Wako) according to the manufacturer’s protocol. The activity level of the LDH released into the culture media was quantified as a percentage of the total activity level of LDH as described previously (32).

Stimulated cells were briefly harvested, 20% of which were saved for DNA extraction of whole cells, and suspended with ice-cold hypotonic lysis buffer. For cell lysis, cell suspensions were added to 0.03% digitonin and then incubated on ice for 5 min and vortexed every minute. Cell lysates were added to 2× isotonic buffer and then centrifuged at 4°C at 5000 × g for 15 min. The supernatants containing cytoplasmic fraction were collected. The cytoplasmic fraction and whole-cell extracts were incubated with 0.2 mg/ml Proteinase K (Wako) and 0.1 mg/ml RNase A (MACHEREY-NAGEL) for 1 h at 42°C and purified with phenol/chloroform extraction and ethanol precipitation. To quantify mitochondrial DNA (mtDNA) level, we subjected DNA extracted from both whole-cell extracts and cytosolic fractions to qPCR with Luna Universal qPCR Master Mix (New England Biolabs). The following primers were used: D-loop forward, 5′-ACCAACAAACCTACCCACCC-3′, and reverse, 5′-GTAGCACTCTTGTGCGGGAT-3′; and cytochrome c oxidase I (Cox1) forward, 5′-ATCCTACCAGGCTTCGGAAT-3′, and reverse, 5′-ACCTACGGTGAAAAGAAAGATGA-3′. The relative content of cytosolic mtDNA versus total mtDNA was calculated.

The value was expressed as the mean ± SD using Prism software (GraphPad). All experiments were repeated at least three independent times. Two groups were compared using Student t test. Data were considered significant when *p < 0.05, **p < 0.01, and ***p < 0.001.

This study was performed in accordance with the Tohoku University Center for Gene Research and Center for Laboratory Animal Research.

Our recent study has demonstrated that GF acts as an agonist for the NLRP3 inflammasome (17). However, its molecular basis is largely unclear. As for the molecular basis, we have speculated that there are two possible mechanisms by which GF initiates the NLRP3 inflammasome activation (18): one is that the chemical structure of GF triggers the NLRP3 inflammasome activation, and the other is that nonspecific inhibition of tyrosine kinases as off-target effects leads to its activation. To understand its molecular basis, we performed a comprehensive analysis of TKIs that possess different and distinct structural features (Table I). Interestingly, it turned out that all TKIs clearly stimulate mature IL-1β release in human monocytic THP-1 cells, in which PMA was primed to differentiate into macrophages (PMA-primed THP-1 cells) (Fig. 1A). Immunoblot analysis showed that the mature IL-1β release occurred in a concentration-dependent manner (Fig. 1B). Moreover, the mature IL-1β release was mostly canceled in NLRP3 KO THP-1 cells or by the NLRP3 inhibitor MCC950 (Fig. 1C, Supplemental Fig. 1). Collectively, these observations suggest that the TKIs stimulate the NLRP3 inflammasome activation via a common mechanism, and they further suggest that the agonistic activity of GF is independent of its structural properties.

Table I.

The characteristics of the TKIs

The characteristics of the TKIs
The characteristics of the TKIs
FIGURE 1.

The TKIs initiate the NLRP3 inflammasome activation. (A) PMA-differentiated THP-1 cells were treated with 20 μM GF, ER, AF, RG, AX, IM, LP, OS, or PZ for 8 h. IL-1β release was analyzed by ELISA. Data shown are the mean ± SD. Significant differences were determined by Student t test; **p < 0.01, ***p < 0.001. (B) PMA-differentiated THP-1 cells were treated with the indicated concentrations of GF, ER, AF, RG, AX, IM, LP, OS, or PZ for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (C) PMA-differentiated WT and NLRP3 KO THP-1 cells were treated with 20 μM GF, ER, AF, RG, AX, IM, LP, OS, or PZ for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs.

FIGURE 1.

The TKIs initiate the NLRP3 inflammasome activation. (A) PMA-differentiated THP-1 cells were treated with 20 μM GF, ER, AF, RG, AX, IM, LP, OS, or PZ for 8 h. IL-1β release was analyzed by ELISA. Data shown are the mean ± SD. Significant differences were determined by Student t test; **p < 0.01, ***p < 0.001. (B) PMA-differentiated THP-1 cells were treated with the indicated concentrations of GF, ER, AF, RG, AX, IM, LP, OS, or PZ for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (C) PMA-differentiated WT and NLRP3 KO THP-1 cells were treated with 20 μM GF, ER, AF, RG, AX, IM, LP, OS, or PZ for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs.

Close modal

We next addressed a more detailed characterization of TKI-induced activation of the NLRP3 inflammasome. It is well known that the NLRP3 inflammasome activation is proceeded by two steps: the priming (signal I) step and the activating (signal II) step (33). In signal I, priming stimuli transcriptionally upregulate the components of the NLRP3 inflammasome, such as NLRP3 and pro–IL-1β (precursor IL-1β), basically through the activation of the NF-κB pathways (33). However, several TKIs, such as GF, PZ, OS, and IM, failed to upregulate both NLRP3 and pro–IL-1β in mouse BMDMs, whereas LPS, a TLR4 ligand that transcriptionally upregulates NLRP3 and pro–IL-1β expression, clearly did so (Fig. 2A). Consistent with these observations, GF did not promote the degradation of the IκBα and phosphorylation of p65/RelA, representative indices of the NF-κB activation, in BMDMs (Fig. 2B). In addition, the TKIs did not stimulate the mature IL-1β release in the absence of LPS or PMA priming in BMDMs or THP-1 cells, respectively (Fig. 2C, 2D). Thus, these results suggest that the TKIs do not stimulate the priming step of the NLRP3 inflammasome activation. In contrast, on caspase-1 activation, cells usually form only one ASC punctum, termed “ASC speck” in their cytosol, which indicates the inflammasome activation (34). Interestingly, as well as a typical NLRP3 ligand nigericin, the TKIs promoted the ASC speck formation (Fig. 2E). Moreover, the TKIs stimulated the caspase-1 activation to a similar extent as nigericin (Fig. 2F). Collectively, these observations suggest that the TKIs activate the NLRP3 inflammasome without promoting transcriptional upregulation of NLRP3 and IL-1β.

FIGURE 2.

The TKIs activate caspase-1 without affecting the upregulation of NLRP3 and IL-1β. (A) BMDMs were treated with 20 μM GF, PZ, OS, or IM for 8 h or 100 ng/ml LPS for 8 h. Cell lysates were subjected to immunoblotting with the indicated Abs. (B) BMDMs were treated with 20 μM GF for the indicated periods or 100 ng/ml LPS for 0.5 h. Cell lysates were subjected to immunoblotting with the indicated Abs. (C) LPS-primed or nonprimed BMDMs were treated with 20 μM GF, PZ, OS, or IM for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (D) PMA-differentiated or undifferentiated THP-1 cells were treated with 20 μM GF, PZ, OS, or IM for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (E) PMA-differentiated THP-1 cells were treated with 5 µM nigericin for 2 h or 20 μM GF, PZ, OS, IM, or ER for 8 h, and then we performed immunofluorescence staining with ASC Ab and DAPI nuclear staining. Scale bars, 5 µm. The aggregation of ASC (red fluorescence) shows “ASC speck.” (F) LPS-primed BMDMs were treated with 20 μM GF, PZ, or IM for 8 h, 20 μM OS for 4 h, or 5 µM nigericin for 2 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. DM, DMSO; Nig, nigericin.

FIGURE 2.

The TKIs activate caspase-1 without affecting the upregulation of NLRP3 and IL-1β. (A) BMDMs were treated with 20 μM GF, PZ, OS, or IM for 8 h or 100 ng/ml LPS for 8 h. Cell lysates were subjected to immunoblotting with the indicated Abs. (B) BMDMs were treated with 20 μM GF for the indicated periods or 100 ng/ml LPS for 0.5 h. Cell lysates were subjected to immunoblotting with the indicated Abs. (C) LPS-primed or nonprimed BMDMs were treated with 20 μM GF, PZ, OS, or IM for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (D) PMA-differentiated or undifferentiated THP-1 cells were treated with 20 μM GF, PZ, OS, or IM for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (E) PMA-differentiated THP-1 cells were treated with 5 µM nigericin for 2 h or 20 μM GF, PZ, OS, IM, or ER for 8 h, and then we performed immunofluorescence staining with ASC Ab and DAPI nuclear staining. Scale bars, 5 µm. The aggregation of ASC (red fluorescence) shows “ASC speck.” (F) LPS-primed BMDMs were treated with 20 μM GF, PZ, or IM for 8 h, 20 μM OS for 4 h, or 5 µM nigericin for 2 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. DM, DMSO; Nig, nigericin.

Close modal

Accumulating evidence has shown that conventional IL-1β release to extracellular space is mediated by the pore-forming protein GSDMD (35). We therefore examined whether GSDMD is involved in TKI-induced IL-1β release. In GSDMD knockdown THP-1 cells, nigericin failed to promote IL-1β release (Fig. 3A). Similar to nigericin, TKI-induced IL-1β release was also attenuated in GSDMD knockdown THP-1 cells (Fig. 3B). In contrast, it is well known that GSDMD mediates pyroptosis, a form of proinflammatory programmed cell death of innate immune cells accompanied by release of cellular contents (35). Importantly, pyroptosis is involved in the necroinflammation, which refers to an autoamplification loop of inflammation driven by pyroptosis (36). Therefore, we examined whether pyroptosis is involved in TKI-induced inflammatory responses, using LDH assay that is commonly used to evaluate pyroptosis (37). As shown in Fig. 3C, the TKIs promoted LDH release, which is inhibited by the NLRP3 inhibitor MCC950, suggesting that the TKIs induce pyroptosis mediated by the NLRP3 inflammasome. Moreover, TKI-induced LDH release was significantly inhibited in GSDMD knockdown THP-1 cells, although not as much as nigericin (Fig. 3D, 3E). Therefore, the TKIs appear to initiate inflammatory responses through GSDMD-mediated pyroptosis.

FIGURE 3.

The TKIs initiate GSDMD-mediated pyroptosis. (A) PMA-differentiated control (Ctr) or GSDMD knockdown THP-1 cells were treated with 2.5 µM nigericin for 1.5 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (B) PMA-differentiated control (Ctr) or GSDMD knockdown THP-1 cells were treated with 20 μM GF, OS, or LP for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (C) PMA-differentiated THP-1 cells were treated with 20 μM GF, ER, AF, IM, LP, OS, or PZ for 8 h with or without 1 μM MCC950. Cell cytotoxicity was measured by LDH release assay. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05, **p < 0.01. (D) PMA-differentiated control (Ctr) or GSDMD knockdown THP-1 cells were treated with 5 μM nigericin for 2 h. Cell cytotoxicity was measured by LDH release assay. Data shown are the mean ± SD. Significant differences were determined by Student t test; ***p < 0.001. (E) PMA-differentiated control (Ctr) or GSDMD knockdown THP-1 cells were treated with 20 μM GF, AF, LP, or OS for 8 h. Cell cytotoxicity was measured by LDH release assay. Data shown are the mean ± SD. Significant differences were determined by Student t test; **p < 0.01, ***p < 0.001.

FIGURE 3.

The TKIs initiate GSDMD-mediated pyroptosis. (A) PMA-differentiated control (Ctr) or GSDMD knockdown THP-1 cells were treated with 2.5 µM nigericin for 1.5 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (B) PMA-differentiated control (Ctr) or GSDMD knockdown THP-1 cells were treated with 20 μM GF, OS, or LP for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (C) PMA-differentiated THP-1 cells were treated with 20 μM GF, ER, AF, IM, LP, OS, or PZ for 8 h with or without 1 μM MCC950. Cell cytotoxicity was measured by LDH release assay. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05, **p < 0.01. (D) PMA-differentiated control (Ctr) or GSDMD knockdown THP-1 cells were treated with 5 μM nigericin for 2 h. Cell cytotoxicity was measured by LDH release assay. Data shown are the mean ± SD. Significant differences were determined by Student t test; ***p < 0.001. (E) PMA-differentiated control (Ctr) or GSDMD knockdown THP-1 cells were treated with 20 μM GF, AF, LP, or OS for 8 h. Cell cytotoxicity was measured by LDH release assay. Data shown are the mean ± SD. Significant differences were determined by Student t test; **p < 0.01, ***p < 0.001.

Close modal

Our previous study has demonstrated that GF promotes the loss of MMP that leads to the NLRP3 inflammasome activation (17). When cells were stained with a fluorescent probe 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolyl-carbocyanine iodide (JC-1), the TKIs enhanced the fluorescence intensity of the JC-1 monomer (green), indicating that not only GF but also other TKIs promote the loss of MMP (Fig. 4A). These findings support the idea that the TKIs stimulate the NLRP3 inflammasome activation via a common mechanism, which prompted us to elucidate the mechanism. A reasonable expectation is that the TKIs share a common off-target activity to some of the tyrosine kinases, which results in mitochondrial dysfunction that causes the loss of MMP. Surprisingly, GF substantially reduced the tyrosine phosphorylation levels of a wide variety of mitochondrial proteins in PMA-primed THP-1 cells, under conditions in which tyrosine phosphatases are inhibited by a vanadate and H2O2 mixture (peroxovanadate), even though the expression of EGFR was not seen in THP-1 cells (Fig. 4B, 4C) (38). Moreover, similar results were observed in EGFR KO A549 cells, a human lung adenocarcinoma cell line (Fig. 4D). These results suggest that GF can broadly inhibit tyrosine kinase signaling even in the absence of EGFR. We next explored how the blockade of tyrosine kinase signaling leads to mitochondrial dysfunction. In this regard, the SFKs such as c-Src play an essential role in mitochondrial function through tyrosine phosphorylation of respiratory chain components (22, 24). Therefore, given that the TKIs cause mitochondrial dysfunction by nonspecifically inhibiting c-Src, it is possible to explain the mechanism by which the TKIs initiate mitochondrial dysfunction that triggers the NLRP3 inflammasome activation. As shown in Fig. 4E, phosphorylation of c-Src induced by peroxovanadate was clearly inhibited by PP2, the specific inhibitor of the SFKs. Interestingly, GF and other TKIs also inhibited the phosphorylation of c-Src, although much higher concentrations are needed to inhibit the phosphorylation (Fig. 4E, Supplemental Fig. 2). Moreover, GF inhibited the basal phosphorylation of c-Src before the mature IL-1β release (Fig. 4F). Taken together, these findings suggest that the TKIs inhibit c-Src as off-target effects and suggest the possibility that the inhibition of c-Src is responsible for TKI-induced activation of the NLRP3 inflammasome.

FIGURE 4.

The TKIs exhibit off-target activity against c-Src. (A) PMA-differentiated THP-1 cells were treated with 20 μM GF, ER, AF, AX, IM, OS, or PZ for 8 h. PMA-differentiated THP-1 cells were treated with 20 μM RG or LP for 3 and 1 h, respectively. MMP was measured using JC-1 probe. Upper panel, Representative images of JC-1 monomer (green fluorescence) and JC-1 aggregate (red fluorescence) were shown. Scale bars, 10 µm. Lower graphs depict the measured value of means. Data shown are the mean ± SD. Significant differences were determined by Student t test; **p < 0.01, ***p < 0.001. (B) PMA-differentiated THP-1 cells were pretreated with 20 μM GF for 1 h and then treated with 250 µM peroxovanadate for 0.5 h. The mitochondrial and cytoplasmic extracts were subjected to immunoblotting with the indicated Abs. (C) The cell extracts from A549 and PMA-differentiated THP-1 cells were subjected to immunoblotting with the indicated Abs. (D) WT and EGFR KO A549 cells were treated with 20 μM GF for 1 h and then 250 µM peroxovanadate for 0.5 h. The mitochondrial and cytoplasmic extracts were subjected to immunoblotting with the indicated Abs. (E) PMA-differentiated THP-1 cells were pretreated with the indicated concentrations of indicated reagents for 1 h and then 250 µM peroxovanadate for 0.5 h. Cell lysates were subjected to immunoblotting with the indicated Abs. *Nonspecific band. (F) PMA-differentiated THP-1 cells were treated with 20 μM GF for the indicated periods. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. *Nonspecific band.

FIGURE 4.

The TKIs exhibit off-target activity against c-Src. (A) PMA-differentiated THP-1 cells were treated with 20 μM GF, ER, AF, AX, IM, OS, or PZ for 8 h. PMA-differentiated THP-1 cells were treated with 20 μM RG or LP for 3 and 1 h, respectively. MMP was measured using JC-1 probe. Upper panel, Representative images of JC-1 monomer (green fluorescence) and JC-1 aggregate (red fluorescence) were shown. Scale bars, 10 µm. Lower graphs depict the measured value of means. Data shown are the mean ± SD. Significant differences were determined by Student t test; **p < 0.01, ***p < 0.001. (B) PMA-differentiated THP-1 cells were pretreated with 20 μM GF for 1 h and then treated with 250 µM peroxovanadate for 0.5 h. The mitochondrial and cytoplasmic extracts were subjected to immunoblotting with the indicated Abs. (C) The cell extracts from A549 and PMA-differentiated THP-1 cells were subjected to immunoblotting with the indicated Abs. (D) WT and EGFR KO A549 cells were treated with 20 μM GF for 1 h and then 250 µM peroxovanadate for 0.5 h. The mitochondrial and cytoplasmic extracts were subjected to immunoblotting with the indicated Abs. (E) PMA-differentiated THP-1 cells were pretreated with the indicated concentrations of indicated reagents for 1 h and then 250 µM peroxovanadate for 0.5 h. Cell lysates were subjected to immunoblotting with the indicated Abs. *Nonspecific band. (F) PMA-differentiated THP-1 cells were treated with 20 μM GF for the indicated periods. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. *Nonspecific band.

Close modal

To investigate this possibility, we examined whether the c-Src inhibitors initiate the activation of the NLRP3 inflammasome. As shown in Fig. 5A, PP2 clearly promoted the mature IL-1β release, whereas GF failed to do so at the lower concentrations (0.25–1 μM). This observation supports the idea that inhibition of c-Src is responsible for TKI-induced activation of the NLRP3 inflammasome. In addition, the mature IL-1β release induced by PP2 was almost blocked by the NLRP3 inhibitor MCC950 or in NLRP3 KO THP-1 cells (Fig. 5B, 5C). Similar to the TKIs, PP2 failed to stimulate IL-1β release in the absence of PMA priming in THP-1 cells and upregulate pro–IL-1β and NLRP3 protein in BMDMs (Fig. 5D, 5E). Moreover, PP2 promoted the ASC speck assembly (Fig. 5F). Meanwhile, other c-Src inhibitors, such as dasatinib, saracatinib, and bosutinib, also promoted the mature IL-1β release (Supplemental Fig. 3A). Moreover, the c-Src inhibitors caused the loss of MMP, as well as GF (Fig. 5G, Supplemental Fig. 3B). Therefore, we next examined whether the TKIs actually inhibit the kinase activity of c-Src. As shown in Fig. 5H, overexpressed c-Src increased the phosphorylation levels of endogenous STAT3, a well-known substrate of the SFKs (39). Interestingly, its phosphorylation was potently inhibited by AX, PZ, RG, and AF, similar to that of PP2, and moderately inhibited by GF, ER, OS, and IM. Because Fyn and Fgr are also known as mSFKs and are actually observed in mitochondrial fractions, we tested the inhibitory effect of the TKIs on Fyn and Fgr (Fig. 5I). As for Fyn, only AX, PZ, RG, and AF inhibited the phosphorylation levels of STAT3 (Fig. 5J). In contrast, all TKIs inhibited the STAT3 phosphorylation induced by the overexpression of Fgr (Fig. 5K). However, the lower concentrations of GF failed to inhibit Fgr-dependent STAT3 phosphorylation, which is correlated well with the mature IL-1β release (Fig. 5A, 5L). Taken together, these results show that the TKIs exhibit off-target activity against the mSFKs, and especially share Fgr as a common off-target kinase.

FIGURE 5.

The TKIs nonspecifically inhibit the activity of mSFKs. (A) PMA-differentiated THP-1 cells were treated with the indicated concentrations of GF or PP2 for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (B) PMA-differentiated THP-1 cells were pretreated with 1 μM MCC950 for 0.5 h and then treated with 1 µM PP2 for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (C) PMA-differentiated WT and NLRP3 KO THP-1 cells were treated with 1 μM PP2 for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (D) PMA-differentiated or undifferentiated THP-1 cells were treated with 1 µM PP2 for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (E) BMDMs were treated with 1 µM PP2 for 8 h or 100 ng/ml LPS for 8 h. Cell lysates were subjected to immunoblotting with the indicated Abs.(F) PMA-differentiated THP-1 cells were treated with 1 µM PP2 for 8 h, and then we performed immunofluorescence staining with ASC Ab and DAPI nuclear staining. Scale bars, 5 µm. The aggregation of ASC (red fluorescence) shows “ASC speck.” (G) PMA-differentiated THP-1 cells were treated with 1 μM PP2 for 8 h. MMP was measured using JC-1 probe. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05. (H, J, and K) HEK293A cells were transfected with the FLAG-c-Src (H), FLAG-Fyn (J), and FLAG-Fgr (K) plasmids for 8 h and then treated with 1 µM PP2 or 20 µM indicated TKI for 16 h. Cell extracts were subjected to immunoblotting with the indicated Abs. (I) The mitochondrial and cytoplasmic extracts from PMA-differentiated THP-1 cells were subjected to immunoblotting with the indicated Abs. *Nonspecific band. (L) HEK293A cells were transfected with the FLAG-Fgr constructs for 8 h and then treated with 1 µM PP2 or indicated concentrations of GF for 16 h. Cell lysates were subjected to immunoblotting with the indicated Abs. DM, DMSO; PP, PP2.

FIGURE 5.

The TKIs nonspecifically inhibit the activity of mSFKs. (A) PMA-differentiated THP-1 cells were treated with the indicated concentrations of GF or PP2 for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (B) PMA-differentiated THP-1 cells were pretreated with 1 μM MCC950 for 0.5 h and then treated with 1 µM PP2 for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (C) PMA-differentiated WT and NLRP3 KO THP-1 cells were treated with 1 μM PP2 for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (D) PMA-differentiated or undifferentiated THP-1 cells were treated with 1 µM PP2 for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (E) BMDMs were treated with 1 µM PP2 for 8 h or 100 ng/ml LPS for 8 h. Cell lysates were subjected to immunoblotting with the indicated Abs.(F) PMA-differentiated THP-1 cells were treated with 1 µM PP2 for 8 h, and then we performed immunofluorescence staining with ASC Ab and DAPI nuclear staining. Scale bars, 5 µm. The aggregation of ASC (red fluorescence) shows “ASC speck.” (G) PMA-differentiated THP-1 cells were treated with 1 μM PP2 for 8 h. MMP was measured using JC-1 probe. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05. (H, J, and K) HEK293A cells were transfected with the FLAG-c-Src (H), FLAG-Fyn (J), and FLAG-Fgr (K) plasmids for 8 h and then treated with 1 µM PP2 or 20 µM indicated TKI for 16 h. Cell extracts were subjected to immunoblotting with the indicated Abs. (I) The mitochondrial and cytoplasmic extracts from PMA-differentiated THP-1 cells were subjected to immunoblotting with the indicated Abs. *Nonspecific band. (L) HEK293A cells were transfected with the FLAG-Fgr constructs for 8 h and then treated with 1 µM PP2 or indicated concentrations of GF for 16 h. Cell lysates were subjected to immunoblotting with the indicated Abs. DM, DMSO; PP, PP2.

Close modal

Given that nonspecific inhibition of the mSFKs caused by the TKIs is responsible for the NLRP3 inflammasome activation, loss of these kinases spontaneously provokes the NLRP3 inflammasome activation. To examine this possibility, we established knockdown THP-1 cells of these kinases (Fig. 6A). The expression levels of the components of the NLRP3 inflammasome were not altered by knockdown of the mSFKs (Supplemental Fig. 4). Surprisingly, mature IL-1β release was spontaneously promoted in c-Src, Fyn, and Fgr, but not Lyn, KD THP-1 cells, even in the absence of agonists for the NLRP3 inflammasome (Fig. 6B, 6C). Moreover, the spontaneous mature IL-1β release was clearly inhibited by MCC950 (Fig. 6D). However, in overexpression experiments, overexpressed Fyn did not affect the NLRP3 inflammasome activation induced by their overexpression (Fig. 6E). Thus, these observations suggest that loss of the mSFKs somehow activates the NLRP3 inflammasome, although the mSFKs do not work as inhibitors of the NLRP3 inflammasome. In contrast, similar to the TKIs, MAPKs such as ERK, p38 MAPK, and JNK are protein-serine/threonine kinases that play an important role in mitochondrial homeostasis (40). Indeed, the specific inhibitors of MAPK promoted the loss of MMP, as well as the TKIs, suggesting that the MAPKs contribute to maintaining mitochondrial functions (Fig. 6F). However, the MAPK inhibitors did not stimulate the mature IL-1β release (Fig. 6G). Therefore, the NLRP3 inflammasome seems to be specifically activated by the mitochondrial damage related to dysfunction of the mSFKs, which raises the possibility that the NLRP3 inflammasome works as a sensor that specifically detects the hypoactivity of mSFKs.

FIGURE 6.

Loss of the SFKs spontaneously activates the NLRP3 inflammasome. (A) The cell extracts from PMA-differentiated control (Ctr), c-Src, Fyn, Fgr, and Lyn knockdown THP-1 cells were subjected to immunoblotting with the indicated Abs. *Nonspecific band. (B) PMA-differentiated Ctr, c-Src, Fyn, Fgr, and Lyn knockdown THP-1 cells were incubated for 8 h. IL-1β release was analyzed by ELISA. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05, **p < 0.01, ***p < 0.001. (C) PMA-differentiated Ctr, c-Src, Fyn, Fgr, and Lyn knockdown THP-1 cells were incubated for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (D) PMA-differentiated Ctr, c-Src, Fyn, and Fgr knockdown THP-1 cells were treated with 1 µM MCC950 and incubated for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (E) HEK293A cells were transfected with Flag-NLRP3, procaspase-1, ASC, pro–IL-1β, and Fyn plasmids for 24 h and then incubated for 8 h after medium change. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (F) PMA-differentiated THP-1 cells were treated with 20 μM SP600125 (JNK inhibitor), SB203580 (p38 inhibitor), U0126 (MEK inhibitor), or GF for 8 h. MMP was measured using JC-1 probe. Data shown are the mean ± SD. Significant differences were determined by Student t test; ***p < 0.001. (G) PMA-differentiated THP-1 cells were treated with 20 μM SP600125 (JNK inhibitor), SB203580 (p38 inhibitor), U0126 (MEK inhibitor), or GF for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs.

FIGURE 6.

Loss of the SFKs spontaneously activates the NLRP3 inflammasome. (A) The cell extracts from PMA-differentiated control (Ctr), c-Src, Fyn, Fgr, and Lyn knockdown THP-1 cells were subjected to immunoblotting with the indicated Abs. *Nonspecific band. (B) PMA-differentiated Ctr, c-Src, Fyn, Fgr, and Lyn knockdown THP-1 cells were incubated for 8 h. IL-1β release was analyzed by ELISA. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05, **p < 0.01, ***p < 0.001. (C) PMA-differentiated Ctr, c-Src, Fyn, Fgr, and Lyn knockdown THP-1 cells were incubated for 8 h. Cell-free supernatants (Sups) and cell lysates were subjected to immunoblotting with the indicated Abs. (D) PMA-differentiated Ctr, c-Src, Fyn, and Fgr knockdown THP-1 cells were treated with 1 µM MCC950 and incubated for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (E) HEK293A cells were transfected with Flag-NLRP3, procaspase-1, ASC, pro–IL-1β, and Fyn plasmids for 24 h and then incubated for 8 h after medium change. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs. (F) PMA-differentiated THP-1 cells were treated with 20 μM SP600125 (JNK inhibitor), SB203580 (p38 inhibitor), U0126 (MEK inhibitor), or GF for 8 h. MMP was measured using JC-1 probe. Data shown are the mean ± SD. Significant differences were determined by Student t test; ***p < 0.001. (G) PMA-differentiated THP-1 cells were treated with 20 μM SP600125 (JNK inhibitor), SB203580 (p38 inhibitor), U0126 (MEK inhibitor), or GF for 8 h. Cell-free Sups and cell lysates were subjected to immunoblotting with the indicated Abs.

Close modal

Finally, we explored the mechanistic connections between the loss of mSFKs and the NLRP3 inflammasome activation. Considering that the TKIs cause the loss of MMP as shown in Fig. 4A, loss of functions of the mSFKs may lead to mitochondrial damage or dysfunction. Notably, a recent report supports this idea. In Huntington’s disease, the mitochondrial dysfunction associated with reduced expressions of c-Src and Fyn proteins was improved by its restoration (41). Interestingly, we found that the loss of Fyn promoted leakage of mtDNA to cytosol (Fig. 7A). The leakage of mtDNA was also observed when wild-type (WT) THP-1 cells were treated with the TKIs (Fig. 7B). Recent evidence has demonstrated that the leakage of mtDNA to cytosol was promoted when mitochondrial reactive oxygen species (mtROS) cause oxidation and fragmentation of mtDNA, which triggers the NLRP3 inflammasome activation (42). Indeed, c-Src, Fyn, and Fgr, but not Lyn, knockdown THP-1 cells exhibited increased generation of mtROS even in the absence of stimulation (Fig. 7C). Moreover, the spontaneous mature IL-1β release was clearly inhibited by the mitochondria-targeted antioxidant Mito-TEMPO (Fig. 7D). Thus, we concluded that the mitochondrial damage that causes mtROS generation and subsequent leakage of mtDNA, initiated by the nonspecific inhibition of any of c-Src, Fyn, or Fgr, stimulates the NLRP3 inflammasome activation, and off-target activity of the TKIs for these SFKs is responsible for NLRP3-mediated inflammation.

FIGURE 7.

Loss of the SFKs initiates leakage of mtROS and mtDNA. (A) Control (Ctr) and Fyn knockdown THP-1 cells were incubated for 8 h. Relative cytosolic mtDNA amounts were quantified by quantitative PCR (qPCR). Relative ratios of cytosolic D-loop or Cox1 are shown. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05, **p < 0.01, ***p < 0.001. (B) THP-1 cells were treated with 20 μM GF, PZ, OS, or IM for 4 h. Relative cytosolic mtDNA amounts were quantified by qPCR. Relative ratios of cytosolic D-loop or Cox1 are shown. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05, **p < 0.01. (C) Ctr, c-Src, Fyn, Fgr, and Lyn knockdown THP-1 cells were measured for mtROS using MitoSOX. Data shown are the mean ± SD. Significant differences were determined by Student t test; ***p < 0.001. (D) PMA-differentiated Ctr, c-Src, Fyn, and Fgr knockdown THP-1 cells were treated with 20 µM Mito-TEMPO and incubated for 8 h. Cell-free supernatants (Sup) and cell lysates were subjected to immunoblotting with the indicated Abs.

FIGURE 7.

Loss of the SFKs initiates leakage of mtROS and mtDNA. (A) Control (Ctr) and Fyn knockdown THP-1 cells were incubated for 8 h. Relative cytosolic mtDNA amounts were quantified by quantitative PCR (qPCR). Relative ratios of cytosolic D-loop or Cox1 are shown. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05, **p < 0.01, ***p < 0.001. (B) THP-1 cells were treated with 20 μM GF, PZ, OS, or IM for 4 h. Relative cytosolic mtDNA amounts were quantified by qPCR. Relative ratios of cytosolic D-loop or Cox1 are shown. Data shown are the mean ± SD. Significant differences were determined by Student t test; *p < 0.05, **p < 0.01. (C) Ctr, c-Src, Fyn, Fgr, and Lyn knockdown THP-1 cells were measured for mtROS using MitoSOX. Data shown are the mean ± SD. Significant differences were determined by Student t test; ***p < 0.001. (D) PMA-differentiated Ctr, c-Src, Fyn, and Fgr knockdown THP-1 cells were treated with 20 µM Mito-TEMPO and incubated for 8 h. Cell-free supernatants (Sup) and cell lysates were subjected to immunoblotting with the indicated Abs.

Close modal

TKIs are essential anticancer drugs. However, TKIs frequently initiate inflammation-based side effects, such as interstitial pneumonitis (43). Our recent study has uncovered that GF, a representative EGFR TKI, initiates lung inflammation by promoting NLRP3-dependent IL-1β release from macrophages (17). In this study, we found that other typical TKIs also promote the activation of the NLRP3 inflammasome. Moreover, our further results suggest that the common off-target effect of the TKIs on mSFKs, such as c-Src, Fgr, and Fyn, is responsible for the NLRP3 activation. In other words, NLRP3 may be sensing the dysfunction or hypoactivity of the mSFKs. Considering that mSFKs play essential roles in cellular functions, to sense and respond to the dysfunction or hypoactivity of mSFKs appears to have physiological significance. However, further studies are needed to elucidate the mechanisms by which the dysfunction or hypoactivity of the mSFKs leads to the NLRP3 inflammasome activation. Voltage-dependent anion channels (VDACs), which are localized in the outer mitochondrial membrane, play a critical role in the NLRP3 activation (44). In contrast, VDACs are identified as potential mitochondrial targets of c-Src (45). Therefore, it is possible that the inactivation of mSFKs allows the VDAC activation, which triggers VDAC-dependent activation of the NLRP3 inflammasome.

Although TKIs are molecular-targeted drugs that specifically inhibit the target tyrosine kinase, it is not surprising that the TKIs exhibit off-target activity against other tyrosine kinases under experimental conditions in which cells are stimulated at much higher concentrations (10–20 µM TKIs) than clinical conditions (up to ∼1.4 µM). Indeed, previous studies have demonstrated that a wide variety of the kinase inhibitors, including TKIs, have a propensity to inhibit multiple targets, implying that the therapeutic dose is critical to prevent adverse reactions to TKIs (46, 47). Moreover, it has been reported that the GF is concentrated in breast tumor tissue (∼43-fold higher than in plasma) (48). Therefore, not only the overdose but also the biased distribution of TKIs may elicit the inactivation of the mSFKs that stimulates the NLRP3 inflammasome in macrophages. Taken together, our findings may explain why a wide variety of TKIs initiate inflammatory-based side effects and may help to prevent the inflammatory-based side effects of TKIs. Thus, our results provide renewed insight into both the biological and the clinical significance of the NLRP3 inflammasome and the SFKs.

The authors have no financial conflicts of interest.

We thank all members of Laboratory of Health Chemistry for helpful discussions.

This work was supported by the Japan Agency for Medical Research and Development (Grants JP19lm0203002 and JP20lm0203011j0004 to T.N.); Ministry of Education, Culture, Sports, Science and Technology, Japan Society for the Promotion of Science KAKENHI (Grant JP21H02691 to T.N.; Grant JP21J10592 to Y.S.; and Grants JP21H02620, JP21K19325, and JP21H00268 to A.M.); the Fugaku Trust for Medicinal Research (A.M.); and the Takeda Science Foundation (A.M.).

The online version of this article contains supplemental material.

AF

afatinib

ASC

apoptosis-associated speck-like protein containing CARD

AX

axitinib

BMDM

bone marrow–derived macrophage

Cox1

cytochrome c oxidase I

EGFR

epidermal growth factor receptor

ER

erlotinib

GF

gefitinib

GSDMD

gasdermin D

HEK293A

human embryonic kidney 293A

IM

imatinib

KO

knockout

LDH

lactate dehydrogenase

LP

lapatinib

MMP

mitochondrial membrane potential

mSFK

mitochondrial Src family kinase

mtDNA

mitochondrial DNA

mtROS

mitochondrial reactive oxygen species

NLRP3

NLR family pyrin domain containing 3

OS

osimertinib

PZ

pazopanib

RG

regorafenib

SFK

Src family kinase

TKI

tyrosine kinase inhibitor

VDAC

voltage-dependent anion channel

WT

wild-type

1
Allen
,
I. C.
,
M. A.
Scull
,
C. B.
Moore
,
E. K.
Holl
,
E.
McElvania-TeKippe
,
D. J.
Taxman
,
E. H.
Guthrie
,
R. J.
Pickles
,
J. P. Y.
Ting
.
2009
.
The NLRP3 inflammasome mediates in vivo innate immunity to influenza A virus through recognition of viral RNA
.
Immunity
30
:
556
565
.
2
Rodrigues
,
T. S.
,
K. S. G.
de Sá
,
A. Y.
Ishimoto
,
A.
Becerra
,
S.
Oliveira
,
L.
Almeida
,
A. V.
Gonçalves
,
D. B.
Perucello
,
W. A.
Andrade
,
R.
Castro
, et al
.
2020
.
Inflammasomes are activated in response to SARS-CoV-2 infection and are associated with COVID-19 severity in patients
.
J. Exp. Med.
218
:
e20201707
.
3
Duewell
,
P.
,
H.
Kono
,
K. J.
Rayner
,
C. M.
Sirois
,
G.
Vladimer
,
F. G.
Bauernfeind
,
G. S.
Abela
,
L.
Franchi
,
G.
Nuñez
,
M.
Schnurr
, et al
.
2010
.
NLRP3 inflammasomes are required for atherogenesis and activated by cholesterol crystals. [Published erratum appears in 2010 Nature 466: 652.]
Nature
464
:
1357
1361
.
4
Halle
,
A.
,
V.
Hornung
,
G. C.
Petzold
,
C. R.
Stewart
,
B. G.
Monks
,
T.
Reinheckel
,
K. A.
Fitzgerald
,
E.
Latz
,
K. J.
Moore
,
D. T.
Golenbock
.
2008
.
The NALP3 inflammasome is involved in the innate immune response to amyloid-beta
.
Nat. Immunol.
9
:
857
865
.
5
Kagi
,
T.
,
R.
Naganuma
,
A.
Inoue
,
T.
Noguchi
,
S.
Hamano
,
Y.
Sekiguchi
,
G. W.
Hwang
,
Y.
Hirata
,
A.
Matsuzawa
.
2022
.
The polypeptide antibiotic polymyxin B acts as a pro-inflammatory irritant by preferentially targeting macrophages
.
J. Antibiot. (Tokyo)
75
:
29
39
.
6
Allam
,
R.
,
M. N.
Darisipudi
,
K. V.
Rupanagudi
,
J.
Lichtnekert
,
J.
Tschopp
,
H. J.
Anders
.
2011
.
Cutting edge: cyclic polypeptide and aminoglycoside antibiotics trigger IL-1β secretion by activating the NLRP3 inflammasome
.
J. Immunol.
186
:
2714
2718
.
7
Schroder
,
K.
,
J.
Tschopp
.
2010
.
The inflammasomes
.
Cell
140
:
821
832
.
8
Mangan
,
M. S. J.
,
E. J.
Olhava
,
W. R.
Roush
,
H. M.
Seidel
,
G. D.
Glick
,
E.
Latz
.
2018
.
Targeting the NLRP3 inflammasome in inflammatory diseases. [Published erratum appears in 2018 Nat. Rev. Drug Discov. 17: 688.]
Nat. Rev. Drug Discov.
17
:
588
606
.
9
Heneka
,
M. T.
,
M. P.
Kummer
,
A.
Stutz
,
A.
Delekate
,
S.
Schwartz
,
A.
Vieira-Saecker
,
A.
Griep
,
D.
Axt
,
A.
Remus
,
T. C.
Tzeng
, et al
.
2013
.
NLRP3 is activated in Alzheimer’s disease and contributes to pathology in APP/PS1 mice
.
Nature
493
:
674
678
.
10
van der Heijden
,
T.
,
E.
Kritikou
,
W.
Venema
,
J.
van Duijn
,
P. J.
van Santbrink
,
B.
Slütter
,
A. C.
Foks
,
I.
Bot
,
J.
Kuiper
.
2017
.
NLRP3 inflammasome inhibition by MCC950 reduces atherosclerotic lesion development in apolipoprotein E-deficient mice: brief report
.
Arterioscler. Thromb. Vasc. Biol.
37
:
1457
1461
.
11
Mulay
,
S. R.
,
O. P.
Kulkarni
,
K. V.
Rupanagudi
,
A.
Migliorini
,
M. N.
Darisipudi
,
A.
Vilaysane
,
D.
Muruve
,
Y.
Shi
,
F.
Munro
,
H.
Liapis
,
H. J.
Anders
.
2013
.
Calcium oxalate crystals induce renal inflammation by NLRP3-mediated IL-1β secretion
.
J. Clin. Invest.
123
:
236
246
.
12
Mridha
,
A. R.
,
A.
Wree
,
A. A. B.
Robertson
,
M. M.
Yeh
,
C. D.
Johnson
,
D. M.
Van Rooyen
,
F.
Haczeyni
,
N. C.
Teoh
,
C.
Savard
,
G. N.
Ioannou
, et al
.
2017
.
NLRP3 inflammasome blockade reduces liver inflammation and fibrosis in experimental NASH in mice
.
J. Hepatol.
66
:
1037
1046
.
13
Saraon
,
P.
,
S.
Pathmanathan
,
J.
Snider
,
A.
Lyakisheva
,
V.
Wong
,
I.
Stagljar
.
2021
.
Receptor tyrosine kinases and cancer: oncogenic mechanisms and therapeutic approaches
.
Oncogene
40
:
4079
4093
.
14
Herbst
,
R. S.
2004
.
Review of epidermal growth factor receptor biology
.
Int. J. Radiat. Oncol. Biol. Phys.
59
(
Suppl. 2
):
21
26
.
15
Huang
,
L.
,
S.
Jiang
,
Y.
Shi
.
2020
.
Tyrosine kinase inhibitors for solid tumors in the past 20 years (2001-2020)
.
J. Hematol. Oncol.
13
:
143
.
16
Esteban-Villarrubia
,
J.
,
J. J.
Soto-Castillo
,
J.
Pozas
,
M.
San Román-Gil
,
I.
Orejana-Martín
,
J.
Torres-Jiménez
,
A.
Carrato
,
T.
Alonso-Gordoa
,
J.
Molina-Cerrillo
.
2020
.
Tyrosine kinase receptors in oncology
.
Int. J. Mol. Sci.
21
:
8529
.
17
Noguchi
,
T.
,
Y.
Sekiguchi
,
Y.
Kudoh
,
R.
Naganuma
,
T.
Kagi
,
A.
Nishidate
,
K.
Maeda
,
C.
Ishii
,
T.
Toyama
,
Y.
Hirata
, et al
.
2021
.
Gefitinib initiates sterile inflammation by promoting IL-1β and HMGB1 release via two distinct mechanisms
.
Cell Death Dis.
12
:
49
.
18
Kagi
,
T.
,
T.
Noguchi
,
A.
Matsuzawa
.
2021
.
Mechanisms of gefitinib-induced interstitial pneumonitis: why and how the TKI perturbs innate immune systems?
Oncotarget
12
:
1321
1322
.
19
Martin
,
G. S.
2004
.
The road to Src
.
Oncogene
23
:
7910
7917
.
20
Manning
,
G.
,
D. B.
Whyte
,
R.
Martinez
,
T.
Hunter
,
S.
Sudarsanam
.
2002
.
The protein kinase complement of the human genome
.
Science
298
:
1912
1934
.
21
Salvi
,
M.
,
A. M.
Brunati
,
L.
Bordin
,
N.
La Rocca
,
G.
Clari
,
A.
Toninello
.
2002
.
Characterization and location of Src-dependent tyrosine phosphorylation in rat brain mitochondria
.
Biochim. Biophys. Acta
1589
:
181
195
.
22
Miyazaki
,
T.
,
L.
Neff
,
S.
Tanaka
,
W. C.
Horne
,
R.
Baron
.
2003
.
Regulation of cytochrome c oxidase activity by c-Src in osteoclasts
.
J. Cell Biol.
160
:
709
718
.
23
Salvi
,
M.
,
N. A.
Morrice
,
A. M.
Brunati
,
A.
Toninello
.
2007
.
Identification of the flavoprotein of succinate dehydrogenase and aconitase as in vitro mitochondrial substrates of Fgr tyrosine kinase
.
FEBS Lett.
581
:
5579
5585
.
24
Hebert-Chatelain
,
E.
2013
.
Src kinases are important regulators of mitochondrial functions
.
Int. J. Biochem. Cell Biol.
45
:
90
98
.
25
Noguchi
,
T.
,
K.
Ishii
,
H.
Fukutomi
,
I.
Naguro
,
A.
Matsuzawa
,
K.
Takeda
,
H.
Ichijo
.
2008
.
Requirement of reactive oxygen species-dependent activation of ASK1-p38 MAPK pathway for extracellular ATP-induced apoptosis in macrophage
.
J. Biol. Chem.
283
:
7657
7665
.
26
Tsuchida
,
M.
,
T.
Yokosawa
,
T.
Noguchi
,
T.
Shimada
,
M.
Yamada
,
Y.
Sekiguchi
,
Y.
Hirata
,
A.
Matsuzawa
.
2020
.
Pro-apoptotic functions of TRAF2 in p53-mediated apoptosis induced by cisplatin
.
J. Toxicol. Sci.
45
:
219
226
.
27
Sekiguchi
,
Y.
,
M.
Yamada
,
T.
Noguchi
,
C.
Noomote
,
M.
Tsuchida
,
Y.
Kudoh
,
Y.
Hirata
,
A.
Matsuzawa
.
2019
.
The anti-cancer drug gefitinib accelerates Fas-mediated apoptosis by enhancing caspase-8 activation in cancer cells
.
J. Toxicol. Sci.
44
:
435
440
.
28
Noguchi
,
T.
,
M.
Suzuki
,
N.
Mutoh
,
Y.
Hirata
,
M.
Tsuchida
,
S.
Miyagawa
,
G. W.
Hwang
,
J.
Aoki
,
A.
Matsuzawa
.
2018
.
Nuclear-accumulated SQSTM1/p62-based ALIS act as microdomains sensing cellular stresses and triggering oxidative stress-induced parthanatos
.
Cell Death Dis.
9
:
1193
.
29
Suzuki
,
M.
,
Y.
Asai
,
T.
Kagi
,
T.
Noguchi
,
M.
Yamada
,
Y.
Hirata
,
A.
Matsuzawa
.
2020
.
TAK1 mediates ROS generation triggered by the specific cephalosporins through noncanonical mechanisms
.
Int. J. Mol. Sci.
21
:
9497
.
30
Shimada
,
T.
,
Y.
Yabuki
,
T.
Noguchi
,
M.
Tsuchida
,
R.
Komatsu
,
S.
Hamano
,
M.
Yamada
,
Y.
Ezaki
,
Y.
Hirata
,
A.
Matsuzawa
.
2022
.
The distinct roles of LKB1 and AMPK in p53-dependent apoptosis induced by cisplatin
.
Int. J. Mol. Sci.
23
:
10064
.
31
Hirata
,
Y.
,
M.
Takahashi
,
Y.
Yamada
,
R.
Matsui
,
A.
Inoue
,
R.
Ashida
,
T.
Noguchi
,
A.
Matsuzawa
.
2021
.
trans-Fatty acids promote p53-dependent apoptosis triggered by cisplatin-induced DNA interstrand crosslinks via the Nox-RIP1-ASK1-MAPK pathway
.
Sci. Rep.
11
:
10350
.
32
Yokosawa
,
T.
,
M.
Yamada
,
T.
Noguchi
,
S.
Suzuki
,
Y.
Hirata
,
A.
Matsuzawa
.
2019
.
Pro-caspase-3 protects cells from polymyxin B-induced cytotoxicity by preventing ROS accumulation
.
J. Antibiot. (Tokyo)
72
:
848
852
.
33
Bauernfeind
,
F. G.
,
G.
Horvath
,
A.
Stutz
,
E. S.
Alnemri
,
K.
MacDonald
,
D.
Speert
,
T.
Fernandes-Alnemri
,
J.
Wu
,
B. G.
Monks
,
K. A.
Fitzgerald
, et al
.
2009
.
Cutting edge: NF-kappaB activating pattern recognition and cytokine receptors license NLRP3 inflammasome activation by regulating NLRP3 expression
.
J. Immunol.
183
:
787
791
.
34
Stutz
,
A.
,
G. L.
Horvath
,
B. G.
Monks
,
E.
Latz
.
2013
.
ASC speck formation as a readout for inflammasome activation
.
Methods Mol. Biol.
1040
:
91
101
.
35
Pandeya
,
A.
,
L.
Li
,
Z.
Li
,
Y.
Wei
.
2019
.
Gasdermin D (GSDMD) as a new target for the treatment of infection
.
MedChemComm
10
:
660
667
.
36
Linkermann
,
A.
2019
.
Death and fire-the concept of necroinflammation
.
Cell Death Differ.
26
:
1
3
.
37
Rayamajhi
,
M.
,
Y.
Zhang
,
E. A.
Miao
.
2013
.
Detection of pyroptosis by measuring released lactate dehydrogenase activity
.
Methods Mol. Biol.
1040
:
85
90
.
38
Huyer
,
G.
,
S.
Liu
,
J.
Kelly
,
J.
Moffat
,
P.
Payette
,
B.
Kennedy
,
G.
Tsaprailis
,
M. J.
Gresser
,
C.
Ramachandran
.
1997
.
Mechanism of inhibition of protein-tyrosine phosphatases by vanadate and pervanadate
.
J. Biol. Chem.
272
:
843
851
.
39
Yu
,
C. L.
,
D. J.
Meyer
,
G. S.
Campbell
,
A. C.
Larner
,
C.
Carter-Su
,
J.
Schwartz
,
R.
Jove
.
1995
.
Enhanced DNA-binding activity of a Stat3-related protein in cells transformed by the Src oncoprotein
.
Science
269
:
81
83
.
40
Lim
,
S.
,
K. R.
Smith
,
S. T.
Lim
,
R.
Tian
,
J.
Lu
,
M.
Tan
.
2016
.
Regulation of mitochondrial functions by protein phosphorylation and dephosphorylation
.
Cell Biosci.
6
:
25
.
41
Fão
,
L.
,
P.
Coelho
,
L.
Duarte
,
R.
Vilaça
,
M. R.
Hayden
,
S. I.
Mota
,
A. C.
Rego
.
2022
.
Restoration of c-Src/Fyn proteins rescues mitochondrial dysfunction in Huntington’s Disease
.
Antioxid. Redox Signal.
DOI: 10.1089/ars.2022.0001.
42
Xian
,
H.
,
K.
Watari
,
E.
Sanchez-Lopez
,
J.
Offenberger
,
J.
Onyuru
,
H.
Sampath
,
W.
Ying
,
H. M.
Hoffman
,
G. S.
Shadel
,
M.
Karin
.
2022
.
Oxidized DNA fragments exit mitochondria via mPTP- and VDAC-dependent channels to activate NLRP3 inflammasome and interferon signaling
.
Immunity
55
:
1370
1385.e8
.
43
Enomoto
,
N.
2022
.
Pathological roles of pulmonary cells in acute lung injury: lessons from clinical practice
.
Int. J. Mol. Sci.
23
:
15027
.
44
Zhou
,
R.
,
A. S.
Yazdi
,
P.
Menu
,
J.
Tschopp
.
2011
.
A role for mitochondria in NLRP3 inflammasome activation. [Published erratum appears in 2011 Nature 475: 122.]
Nature
469
:
221
225
.
45
Guedouari
,
H.
,
Y.
Ould Amer
,
N.
Pichaud
,
E.
Hebert-Chatelain
.
2021
.
Characterization of the interactome of c-Src within the mitochondrial matrix by proximity-dependent biotin identification
.
Mitochondrion
57
:
257
269
.
46
Karaman
,
M. W.
,
S.
Herrgard
,
D. K.
Treiber
,
P.
Gallant
,
C. E.
Atteridge
,
B. T.
Campbell
,
K. W.
Chan
,
P.
Ciceri
,
M. I.
Davis
,
P. T.
Edeen
, et al
.
2008
.
A quantitative analysis of kinase inhibitor selectivity
.
Nat. Biotechnol.
26
:
127
132
.
47
Fabian
,
M. A.
,
W. H.
Biggs
III
,
D. K.
Treiber
,
C. E.
Atteridge
,
M. D.
Azimioara
,
M. G.
Benedetti
,
T. A.
Carter
,
P.
Ciceri
,
P. T.
Edeen
,
M.
Floyd
, et al
.
2005
.
A small molecule-kinase interaction map for clinical kinase inhibitors
.
Nat. Biotechnol.
23
:
329
336
.
48
McKillop
,
D.
,
E. A.
Partridge
,
J. V.
Kemp
,
M. P.
Spence
,
J.
Kendrew
,
S.
Barnett
,
P. G.
Wood
,
P. B.
Giles
,
A. B.
Patterson
,
F.
Bichat
, et al
.
2005
.
Tumor penetration of gefitinib (Iressa), an epidermal growth factor receptor tyrosine kinase inhibitor
.
Mol. Cancer Ther.
4
:
641
649
.

Supplementary data