TBK1 is a signaling hub in coordinating stress-adaptive mechanisms in head and neck cancer progression

Hyo Jeong Kim; Haeng-Jun Kim; Sun-Yong Kim; Jin Roh; Ju Hyun Yun; Chul-Ho Kim

doi:10.1080/15548627.2025.2481661

. 2025 Apr 1;21(8):1744–1766. doi: 10.1080/15548627.2025.2481661

TBK1 is a signaling hub in coordinating stress-adaptive mechanisms in head and neck cancer progression

Hyo Jeong Kim ^a, Haeng-Jun Kim ^b, Sun-Yong Kim ^c, Jin Roh ^d, Ju Hyun Yun ^e, Chul-Ho Kim ^a,^f,^✉

PMCID: PMC12282999 PMID: 40114316

ABSTRACT

Tumorigenesis is closely linked to the ability of cancer cells to activate stress-adaptive mechanisms in response to various cellular stressors. Stress granules (SGs) play a crucial role in promoting cancer cell survival, invasion, and treatment resistance, and influence tumor immune escape by protecting essential mRNAs involved in cell metabolism, signaling, and stress responses. TBK1 (TANK binding kinase 1) functions in antiviral innate immunity, cell survival, and proliferation in both the tumor microenvironment and tumor cells. Here, we report that MUL1 loss results in the hyperactivation of TBK1 in both HNC cells and tissues. Mechanistically, under proteotoxic stress induced by proteasomal inhibition, HSP90 inhibition, or Ub⁺ stress, MUL1 promotes the degradation of active TBK1 through K48-linked ubiquitination at lysine 584. Furthermore, TBK1 facilitates autophagosome-lysosome fusion and phosphorylates SQSTM1, regulating selective macroautophagic/autophagic clearance in HNC cells. TBK1 is required for SG formation and cellular protection. Moreover, we found that MAP1LC3B is partially localized within SGs. TBK1 depletion enhances the sensitivity of HNC cells to cisplatin-induced cell death. GSK8612, a novel TBK1 inhibitor, significantly inhibits HNC tumorigenesis in xenografts. In summary, our study reveals that TBK1 facilitates the rapid removal of ubiquitinated proteins within the cell through protective autophagy under stress conditions and assists SG formation through the use of the autophagy machinery. These findings highlight the potential of TBK1 as a therapeutic target in HNC treatment.

Abbreviations: ALP: autophagy-lysosomal pathway; AMBRA1: autophagy and beclin 1 regulator 1; BaF: bafilomycin A₁; CC: coiled-coil; CD274/PDL-1: CD274 molecule; CHX: cycloheximide; CQ: chloroquine; DNP: dinitrophenol; EGFR: epidermal growth factor receptor; ESCC: esophageal squamous cell carcinoma; G3BP1: G3BP stress granule assembly factor 1; HNC: head and neck cancer; HPV: human papillomavirus; IFN: interferon; IGFBP3: insulin like growth factor binding protein 3; IRF: interferon-regulatory factor 3; KO: knockout; LAMP1: lysosomal associated membrane protein 1; MAP1LC3B: microtubule associated protein 1 light chain 3 beta; NPC: nasopharyngeal carcinoma; PABP: poly(A) binding protein; PI: proteasome inhibitor; PQC: protein quality control; PROTAC: proteolysis-targeting chimera; PURA/PURα: purine rich element binding protein A; RIGI: RNA sensor RIG-I; SD: standard deviation; SG: stress granule; SQSTM1: sequestosome 1; STING1: stimulator of interferon response cGAMP interactor 1; TBK1: TANK binding kinase 1; UPS: ubiquitin-proteasome system; USP10: ubiquitin specific peptidase 10; VCP: valosin containing protein; VHL: von Hippel-Lindau tumor suppressor; WT: wild type

KEYWORDS: Autophagic flux, GSK8612, head and neck cancer, MUL1, stress granule formation, TBK1

Introduction

Cancer cells encounter various stressors, including hypoxia, nutrient deprivation, heat shock, or oxidative stress, which can lead to protein misfolding and aggregation, ultimately compromising cell survival [1]. To avoid protein accumulation, cancer cells activate stress-adaptive mechanisms, which primarily rely on the protein quality control (PQC) system. This multicompartmental system coordinates HSPA/HSP70-HSP90-assisted protein folding and disaggregation [2], the ubiquitin-proteasome system (UPS), and the autophagy-lysosomal pathway (ALP) [3], and participates in mRNA translation [4,5].

Protective autophagy is a stress-adaptive mechanism that promotes the survival of cancer cells in advanced stages by ameliorating stress in the microenvironment [6]. However, in HNC, increased expression of autophagy markers, such as SQSTM1 (sequestosome 1) and MAP1LC3B (microtubule associated protein 1 light chain 3 beta)-II, has been associated with reduced overall and disease-specific survival [7]. Targeting autophagy affects the self-renewal, malignancy, chemoresistance, and survival of cancer stem cells in HNC [8,9]. Human papillomavirus/HP infection plays a critical role in HNC progression, with HPV-positive patients generally having better prognoses and treatment responses than HPV-negative patients [10]. Autophagic activity, and AMBRA1 (autophagy and beclin 1 regulator 1) expression are low in HPV⁺ compared to HPV^– HNC cell lines, and AMBRA1-dependent autophagy contributes to cisplatin resistance in HPV^– HNC cell lines [11]. A correlation between elevated MAP1LC3B-II and poor prognosis has been observed in HPV^– HNC patients, supporting the association of protective autophagy with cisplatin resistance [11,12].

In contrast to autophagy, stress granules (SGs) are non-membrane-enclosed RNA granules that assemble in response to changes in environmental conditions, enabling cellular adaptation to stressors. SGs attenuate global mRNA translation to preserve energy, maintaining selective mRNA translation while protecting cells [13,14]. Cancer cells use SGs to support survival and metastatic capacity when exposed to radio- or chemotherapy. A recent study showed that the G3BP1 (G3BP stress granule assembly factor 1) may serve as a biomarker of proliferation, apoptosis, and prognosis in oral squamous cell carcinoma [15]. The SG-associated RNA-binding protein CAPRIN1 drives tumor progression and regulates treatment response in nasopharyngeal carcinoma (NPC) [16]. YBX1 (Y-box binding protein 1) regulates stress SG formation by translationally activating G3BP1 [17]. Clinically, YB1 expression is correlated with advanced-stage NPC, and NPC patients with positive YB1 expression have significantly lower survival rates. YB1 knockdown significantly reduced the proliferation, migration, and invasion abilities of NPC cells through impaired SG formation [18]. In esophageal squamous cell carcinoma (ESCC), PURA/PURα (purine rich element binding protein A), an SG component, represses the mRNA translation initiation of IGFBP3 (insulin like growth factor binding protein 3) by forming cytoplasmic SGs [19]. Knockdown of IGFBP3 reversed the inhibitory effects of PURA loss on ESCC cell proliferation, migration, and invasion [19]. Despite recent research into SG-associated proteins in HNC, the precise role and mechanism of SG in chemotherapy response remains unclear.

Recent research suggests that the PQC system closely monitors SGs. Molecular chaperones such as HSP90, HSPA/HSP70, and VCP (valosin containing protein) can prevent the accumulation of misfolded and ubiquitinated proteins inside SGs, maintaining their dynamics [20]. When surveillance by the PQC machinery fails, persistent aberrant SGs are targeted by the UPS or ALP, involving SQSTM1 and VCP [21–23]. The binding of deubiquitinating enzyme USP10 (ubiquitin specific peptidase 10) to G3BP1 stabilizes a soluble conformation of G3BP1 bound to 40S subunits and PABP (poly(A) binding protein), causing SG disassembly [24]. Inhibition of ALP has been shown to impair SG formation [25]. Although the exact role of these PQC components in SG dynamics is only partly understood, these findings suggest that the PQC system and SGs are interconnected.

TBK1 (TANK binding kinase 1) is a serine/threonine kinase that plays an important role in regulating many cellular processes, including innate immunity, inflammatory cytokine production, autophagy, mitochondrial metabolism, and cancer development [26]. Upon DNA or RNA virus infection, STING1 (stimulator of interferon response cGAMP interactor 1) and the RNA sensor RIGI (RNA sensor RIG-I) activate the production of type I interferon (IFN), which phosphorylates the transcription factor IRF3 (interferon regulatory factor 3) via TBK1 [27]. Therefore, TBK1 is a central kinase in innate immune sensing of nucleic acids [28]. Recent reports have expanded the role of TBK1 in cancer development. In VHL (von Hippel-Lindau tumor suppressor) null clear-cell renal cell carcinoma (ccRCC) cells, loss of VHL copy number results in hyperactivation of TBK1, which increases autophagy due to TBK1-mediated phosphorylation of SQSTM1 on Ser366, thereby contributing to ccRCC tumorigenesis [29]. A novel cereblon-based TBK1 proteolysis-targeting chimera (PROTAC) efficiently degraded TBK1 in both VHL null and WT VHL ccRCC cells and dramatically inhibited VHL null ccRCC colony growth [29] by recruiting VHL E3 ligase [30]. Thus, degradation or inhibition of TBK1 could be an effective way to decrease cancer cell viability and invasiveness [31–33].

We investigated the mechanism responsible for the negative regulation of TBK1 stability by its interaction partners. HSP90 chaperone inhibition can trigger the ubiquitination and proteasomal degradation of many oncogenic protein kinases. The mechanism by which HSP90 inhibition triggers TBK1 ubiquitination is not understood, and the involvement of E3 ligase in HNC is unknown. We previously showed that MUL1 contributes to HNC progression by negatively regulating AKT [34] or HSPA5 [35]. Our current findings indicate that upon HSP90 inhibition, MUL1 E3 ligase binds to the active form of TBK1 and catalyzes its K48-linked ubiquitination at K584. In contrast, MUL1 deficiency leads to increased TBK1 phosphorylation and defective TBK1 ubiquitination in response to proteotoxic stress, including proteasome inhibitor, HSP90 inhibitors, and Ub⁺ stress. Additionally, we investigated the autonomous function of TBK1 in HNC, which is distinct from its role in innate immune signaling. Although SG formation and autophagic clearance of harmful proteins are essential for proteostasis, the relationship between these two stress-adaptive responses remains elusive. Our results highlight that TBK1 is a central signaling hub in stress-adaptive mechanisms, positively regulating autophagy and SG formation. Our findings provide new insights into autophagy-SG crosstalk and its potential link with HNC progression.

Results

TBK1 is negatively regulated by MUL1 E3 ligase

MUL1/MULAN/GIDE/MAPL is a mitochondrial E3 ubiquitin ligase embedded in the outer mitochondrial membrane via its RING finger domain [36]. It functions as a tumor suppressor protein that inhibits tumor cell proliferation and migration and is notably suppressed in HNC patients and cell lines [34,35,37,38]. Given its downregulation in HNC, targeting MUL1 to enhance its protein stability or gene expression holds significant therapeutic potential. MUL1’s E3 ligase activity facilitates K48-linked ubiquitination of target proteins, promoting their proteasomal degradation. This activity is essential for maintaining the stability of various proteins involved in cellular stress responses and oncogenic signaling pathways. For instance, MUL1-mediated degradation of overexpressed AKT and HSPA5 has previously been demonstrated to suppress HNC progression [35,37]. This is particularly significant because the activation of AKT and HSPA5 establishes a mutually reinforcing positive feedback loop, highlighting the therapeutic importance of targeting MUL1 in HNC.

To further explore MUL1’s tumor suppressor mechanisms, we are currently mapping the MUL1-specific substrate interaction network, which is essential for developing PROTAC-based strategies to ubiquitinate and degrade oncoproteins in HNC [39]. Based on this, we aimed to identify novel substrate targets for MUL1-mediated tumor suppression in HNC. To this end, we screened oncogenic proteins that exhibit opposite expression patterns to MUL1 in MUL1 knockout (KO) HNC cells and patient tissues. Among the candidate MUL1 substrates identified, TBK1 was discovered as a novel interacting protein.

Therefore, we conducted experiments to investigate whether MUL1 directly interacts with and degrades TBK1 in HNC cells. Investigation of MUL1 and TBK1 expression in tissue samples from 14 hNC patients using western blotting revealed that the TBK1 protein expression and phosphorylation were increased, while MUL1 expression was significantly suppressed in tumor samples compared to non-tumor samples (Figure 1A,B). Accordingly, in HNC cell lines, MUL1-deficient cells (SCC15, Cal27, and SCC1483) showed higher levels of p-TBK1 protein than low MUL1-expressing HNC cell lines (FaDu, MSKQLL1, and SNU899) (Figure S1A). These findings suggested that MUL1 and p-TBK1 levels are inversely correlated in tumor and normal tissues, as well as HNC cell lines.

Figure 1. — MUL1 is novel negative regulator of TBK1. (A) TBK1 or MUL1 expression levels were different between tissues from human head and neck cancer (T) and their adjacent non-tumor head and neck (N). Proteins were isolated from tissues of 14 patients with HNC, and p-TBK1, TBK1, or MUL1 expression levels were determined by western blot. (B) Quantification of the p-TBK1:GAPDH, TBK1:GAPDH, and MUL1:GAPDH (14-pair HNC cohort, **p = 0.0058, ****p < 0.0001 by the Mann-Whitney t test). (C) WT or *MUL1* KO FaDu cells were treated with PI+Tg (10 nM bortezomib +100 nM thapsigargin) for the indicated time periods, and subject to western blot. (D) WT or *MUL1* KO FaDu cells were treated with 20 μg/ml cycloheximide (CHX) and 10 nM PI +100 nM tg for the indicated times, and subjected to western blot. (E) MUL1 induces TBK1 degradation by the UPS. MYC-His-TBK1 was transfected into *MUL1* KO FaDu cells together with or without flag-MUL1. Cells were treated with 10 μM MG132, 10 nM bortezomib, 10 μM lactacystin, or 250 nM epoxomicin for 12 h. (F) TBK1 interacts with MUL1. At 24 h after co-transfection with the indicated plasmids, *MUL1* KO FaDu cells were treated with 10 μM MG132 for 8 h before cell harvest and then subjected to Ni-nta affinity-isolation under denaturing conditions. The obtained affinity-isolated samples were subjected to western blot using the indicated antibodies. (G) WT or *MUL1* KO FaDu cells were transfected with NC or *STING1* siRNA for 24 h, followed by incubation with 10 nM PI +100 nM tg for 30 h. (H) Endogenous p-TBK1 ubiquitination assay. WT or *MUL1* KO FaDu cells were treated with 10 nM PI +100 nM tg for the indicated time, followed by ubiquitination assays for analysis with an anti-FK2 antibody. (I) MUL1 induces K48-linked ubiquitination of TBK1. *MUL1* KO FaDu cells were transfected MYC-His-TBK1 and flag-MUL1 together with HA-WT ub or ubiquitin mutants (HA-Ub-K48, HA-Ub-K63, HA-Ub^K48R, or HA-Ub^K63R). Ubiquitinated TBK1 was identified by Ni-nta affinity-isolation assays for analysis with an anti-ha antibody. (J) MUL1 preferentially induces degradation in active TBK1. Active TBK1 (MYC-His-wt TBK1) or inactive TBK1 (MYC-His-TBK1^S172A) were co-transfected with flag-MUL1 plasmids (0, 0.25, 0.5, or 1 μg) in *MUL1* KO FaDu cells. (K) Activated TBK1 is efficiently ubiquitinated by MUL1. After transfection with plasmids as indicated, *MUL1* KO FaDu cells were treated with 10 μM MG132 and then subjected to Ni-nta affinity-isolation ubiquitination assays. (L) Colocalization of EGFP-MUL1 (green), COX4l1 (red), and HSPD1/HSP60 (far red) with or without 20 nM PI +100 nM tg treatment for 12 h in *MUL1* KO FaDu cells. (M) Colocalization of EGFP-MUL1 (green), mitochondria (red), and MYC-His-TBK1 (far red) with or without 20 nM PI +100 nM tg treatment for 12 h in *MUL1* KO FaDu cells. Mitochondria were labeled with MitoTracker red CMXRos dye. Images were obtained using a nikon N-SIM confocal microscope and overlaid to assess protein localization. Scale bars: 10 μm and 1 μm (inset). (N) The percentages of TBK1 containing colocalization of mitochondria (COX4l1 and HSPD1/HSP60) per field. The data represent the mean ± SD of 25 fields, each of which contains at least 4 cells that meet statistical requirements, from three independent experiments. (O) Quantification of the colocalization of TBK1 and MUL1 signal. Merged images from (M) were analyzed for TBK1 and MUL1 colocalization using NIS Elements software and Pearson’s correlation coefficient. The data represent the mean ± SD of 40 fields, each of which contains at least 4 cells that meet statistical requirements, from three independent experiments. ****p < 0.0001 by unpaired t test.

The induction of ER stress and inhibition of proteasomal degradation lead to an extensive accumulation of aggregates [40]. Moreover, co-treatment with a proteasome inhibitor (PI) and the ER stress inducer thapsigargin (Tg) [41] resulted in synergistic activation of TBK1, suggesting that these stresses facilitate the formation of misfolded or ubiquitinated proteins. We observed that stable MUL1 KO cells had elevated p-TBK1 levels at early time points after PI+Tg treatment [35], and even higher basal levels of p-TBK1 (Figure S1B). At late time points, both p-TBK1 and TBK1 were degraded in WT MUL1 cells, but their levels remained sustained in MUL1 KO cells (Figure 1C). In addition to proteasomal inhibition and ER stress, Ub⁺ stress [42] induced by the overexpression of His6-tagged Ub, which facilitates the formation of misfolded proteins, also induced TBK1 phosphorylation (Figure S1C), and this was accelerated in MUL1 KO cells. Furthermore, the degradation of p-TBK1 and TBK1 in MUL1 KO cells was delayed compared to that in WT cells, as demonstrated by the cycloheximide (CHX)-chase assay (Figure 1D; Figure S1D). Moreover, reconstitution of MUL1 expression in MUL1 KO cells reduced the levels of TBK1. Proteasomal degradation of TBK1 by several proteasome inhibitors, including MG132, bortezomib, lactacystin, and epoxomicin, reduced the TBK1 protein levels (Figure 1E). To confirm that MUL1 targets TBK1, we examined the interaction between MUL1 and TBK1. His-tagged TBK1 (MYC-His-TBK1) was co-transfected into MUL1 KO cells, and the proteins precipitated using Ni-NTA agarose resin on subjection to western blot analysis revealed the presence of Flag-tagged MUL1 (Flag-MUL1) using an anti-Flag antibody. This suggested that TBK1 and MUL1 can form a complex (Figure 1F). MG132 increased the binding between MUL1 and TBK1, as MUL1 levels were elevated in TBK1 precipitates. PPM1A [43] and PPM1B [44] catalytically dephosphorylate TBK1. We questioned whether MUL1 may affect PPM1A or PPM1B binding with TBK1, thus contributing to TBK1 dephosphorylation. We found that TBK1-PPM1A or -PPM1B association and TBK1 dephosphorylation were undetectable with or without MUL1 expression (Figure 1F).

It has been reported that MUL1 promotes the K63-linked ubiquitination and activation of STING1, which subsequently activates TBK1 signaling [45]. To determine whether MUL1 regulates p-TBK1 and TBK1 degradation through STING1, we examined the effects of STING1 knockdown (KD) in WT MUL1 and KO cells. PI+Tg-induced p-TBK1 and TBK1 protein degradation were significantly reduced in MUL1 KO cells, regardless of STING1 expression (Figure 1G). Furthermore, TBK1 ubiquitination was significantly inhibited in MUL1 KO cells (Figure 1H). The ability to generate diverse substrate-ubiquitin structures is essential for targeting proteins to different fates [46]. To explore the type of ubiquitination involved, we performed ubiquitination assays in FaDu cells expressing WT or lysine-mutant Ub plasmids. MUL1 overexpression induced TBK1 ubiquitination in MUL1 KO cells expressing WT Ub or Ub-K48 to a similar degree (Figure 1I). However, Ub-K63 and Ub^K48R, but not WT Ub and Ub^K63R, greatly reduced MUL1-mediated TBK1 ubiquitination, indicating that MUL1 ubiquitinates TBK1 primarily through K48-linked chains.

Additionally, we assessed the K48-linked ubiquitination of TBK1 and found that STING1 KD did not affect MUL1-mediated ubiquitination of MYC-His-TBK1 (Figure S1E). These results indicate that MUL1 regulates TBK1 K48-linked ubiquitination and degradation independently of STING1. Interestingly, in HNC cells, the basal level of p-STING1 was lower in MUL1 KO cells compared to WT cells. This observation is consistent with a previous study reporting that MUL1 promotes STING1 activation through K63-linked ubiquitination at K224 [45].

MUL1 may specifically target the active form of TBK1 for degradation. To test this, we co-transfected MUL1 KO cells with MUL1, WT TBK1, or two kinase-inactive mutants (TBK1^S172A and TBK1^K38A) [47]. MUL1 failed to promote the degradation of the TBK1 mutants (Figure S1F). In contrast, overexpressed WT TBK1 is constitutively phosphorylated at S172 [48], showed a dose-dependent reduction in protein levels upon MUL1 overexpression. TBK1^S172A levels, however, remained unaffected by MUL1 (Figure 1J). Ubiquitination assays further demonstrated that the ubiquitination level of WT TBK1, but not that of the TBK1 mutants, increased in a MUL1-dependent manner (Figure 1K). Together, these results indicate that MUL1 promotes K48-linked ubiquitination and degradation of TBK1 in a catalytic activity-dependent manner.

The interaction between TBK1 and a stimulus-adaptor protein can be induced by different upstream stimuli, and the subcellular localization of TBK1 varies depending on the specific stimulus [49]. For example, TBK1-TANK is localized in the perinuclear region in a punctate pattern, inducing IRF3 activation and the production of IFNA/IFNα and IFNB/IFNβ [49]. TBK1 also binds to SQSTM1 or OPTN on phagophores, regulating autophagy [50,51] and mitophagy [52]. PLA1A facilitates the recruitment of activated TBK1 to mitochondria and interaction with MAVS (mitochondrial antiviral signaling protein) [53]. We hypothesized that the first and key step in terminating TBK1 activity is the recruitment of activated TBK1 from the cytosol to the outer mitochondrial membrane. As MUL1 is localized to the outer mitochondrial membrane [54], it was examined using GFP-tagged MUL1 and co-staining with MitoTracker, COX4l1, HSPD1/HSP60, or TOMM20, confirming its mitochondrial localization (Figure 1L,N; Figure S1G). Confocal microscopy revealed that MYC-tagged TBK1 protein was cytosolic and partially colocalized in the mitochondria. The colocalization of MUL1 and TBK1 increased upon PI+Tg treatment (Figure 1M,O). Collectively, these data suggested that MUL1 induces ubiquitination and proteasomal degradation of TBK1 on the cytosolic face of the mitochondrial outer membrane.

MUL1 induces K48-linked ubiquitination of TBK1 at lysine 584

To gain insight into the mechanism of MUL1 in regulating TBK1 ubiquitination, we generated four truncations of TBK1 (Figure 2A) and found that MUL1 could promote the degradation of the TBK1 deletion mutant containing a coiled-coil (CC) domain, but not that of mutants containing only the kinase domain or the kinase and/or ubiquitin-like domains. The CC domain displayed a significant reduction in MUL1 expression through the UPS (Figure 2B). MUL1 bound to the CC domain as well as WT TBK1 (Figure 2C). These results indicated that the CC domain of TBK1 is responsible for MUL1-mediated degradation and ubiquitination of TBK1. Examination of TBK1 using the Ubisite, UbNET2.0, dbPTM, and BDM-PUB programs identified numerous lysine residues as putative ubiquitination sites [55–57]. To evaluate the correlation between TBK1 ubiquitination and mutations on 12 lysine residues in the CC domain, all lysine (K) to alanine (R) mutants of the putative ubiquitin-conjugation residues in TBK1 were generated by site-directed mutagenesis. MUL1-mediated degradation of TBK1^K584R was blocked compared with that of WT TBK1, TBK1^K416R, TBK1^K484R, TBK1^K504R, TBK1^K567R, TBK1^K570R, TBK1^K628R, TBK1^K661R, TBK1^K670R, TBK1^K691R, TBK1^K692R, and TBK1^K694R (Figure 2D). The ubiquitination of TBK1^K584R was also inhibited by MUL1 (Figure 2E). The expression of the TBK1^K584R mutant decreased only slightly, even when MUL1 was overexpressed (Figure 2F). The protein half-life of TBK1^K584R was prolonged compared with that of WT TBK1 in a CHX-chase assay (Figure 2G). Taken together, these results suggested that K584 in TBK1 is a specific residue for MUL1-mediated TBK1 ubiquitination.

HSP90 regulates TBK1 stability and activity

HSP90 fulfills housekeeping functions by assisting the folding and maturation of a large variety of proteins, referred to as the HSP90 clients. The association of client proteins with HSP90 is a prerequisite for their maturation into an activation-competent state. As a result, inhibition of HSP90 disrupts signaling mediated by SRC kinases [58] and the PI3K-AKT cascade [59], often leading to client protein degradation, causing a decrease in their steady-state levels. HSP90 has been previously identified as a TBK1 interactor [60,61], suggesting that TBK1 stability depends on the chaperone activity of HSP90 [62]. We observed strong degradation of endogenous p-TBK1 and TBK1 in wild-type cells treated with HSP90 inhibitors (17AAG, AUY922, and HSP990) in a dose-dependent manner. However, this degradation was significantly reduced in MUL1 KO cells, indicating that p-TBK1 and TBK1 are indeed targets of MUL1-mediated degradation (Figure 3A,B; Figure S2). Similar results were obtained when depleting HSP90 using siRNA (Figure 3C). Consistent herewith, further experiments indicated that p-TBK1 and TBK1 were partly degraded by 17AAG treatment in MUL1 KO cells but were completely degraded by 17AAG upon reconstitution of MUL1 expression (Figure 3D). To test whether HSP90 inhibition promotes MYC-tagged TBK1 degradation via the proteasome or lysosomes, we co-treated TBK1-overexpressing cells with MG132 and bafilomycin A₁ (BaF; autophagosome-lysosome fusion inhibitor). MYC-His-TBK1 levels were partially recovered upon proteasome inhibition, indicating rapid proteasomal degradation when HSP90 is inhibited by 17AAG or HSP990. The lack of TBK1 accumulation in the BaF-treated group confirms that its degradation is proteasome-dependent, not autophagy-related (Figure 3E). We next confirmed that HSP90 interacts with overexpressed TBK1 by Ni-NTA affinity isolation and that this interaction was sensitive to 17AAG (Figure 3F). We found that the interaction between HSP90 and TBK1 was also sensitive to the inhibition of TBK1 kinase activity by GSK8612 [63] (Figure 3F). Likewise, upon treatment of the cells with 17AGG, the degradation of TBK1^S172A or TBK1^K38A was inhibited compared with that of WT TBK1 (Figure 3G). Together, these data suggested that TBK1 requires repeated rounds of binding and release to HSP90 to maintain its active state; preventing TBK1 reloading onto HSP90 redirects it to degradation, as has been observed for other client kinases of HSP90 [60]. Because TBK1 is regulated by ubiquitin-mediated degradation, we tested whether HSP90 inhibitors could promote TBK1 ubiquitination. 17AAG and HSP990 significantly increased TBK1 ubiquitination (Figure 3H). Moreover, MUL1 KD significantly reversed 17AAG-induced ubiquitination of TBK1 (Figure 3I), corroborating that MUL1 is required for 17AAG-stimulated TBK1 ubiquitination and degradation. Accordingly, 17AAG promoted the binding of TBK1 to MUL1 (Figure 3F). In summary, HSP90 appeared to be essential for the stability of activated TBK1. Inhibition of HSP90 function led to increased MUL1 binding to TBK1 and subsequent TBK1 ubiquitination and degradation through the UPS (Figure 3J).

Figure 3. — HSP90 inhibition lead to the degradation of active TBK1 by MUL1. (A) 17AAG treatment significantly reduces p-TBK1 and TBK1 levels in WT cells but not in *MUL1* KO FaDu cells. The cells were treated with 0, 0.25, 0.5, 1, or 2 μM 17AAG for 24 h. (B) Same as in (A), the cells were treated with 0, 0.5, 1, or 2 μM AUY922 for 24 h, and the effects on p-TBK1 and TBK1 levels were measured. (C) TBK1 levels are reduced by *HSP90* knockdown. FaDu cells were transfected with NC or *HSP90* siRNAs for 48 h, and the levels of endogenous p-TBK1 and TBK1 were examined by western blot. (D) FaDu cells were transfected with pCMV-flag or flag-MUL1 for 24 h. The cells were treated with 0, 0.5, 1, or 2 μM 17AAG for 24 h. (E) FaDu cells were transfected with MYC-His-TBK1 for 24 h. The cells were then incubated for an additional 8 h in the presence of 2 μM 17AGG or 1 μM HSP990 alone or combined with 10 μM MG132 or 100 nM BaF, and analyzed by western blot. (F) FaDu cells were transfected with MYC-His-TBK1 for 24 h. Where indicated, cells were treated with either 1 μM 17AAG or 20 μM GSK8612 in combination with 10 μM MG132 for 4 h. Cells were lysed and subjected to Ni-nta affinity-isolation under denaturing conditions. Levels of MYC-His-TBK1 and endogenous HSP90 are shown in the input and bead fractions. (G) FaDu cells were transfected with MYC-His-wt TBK1, MYC-His TBK1^S172A, or MYC-His-TBK1^K38A. at 24 h transfection, cells were treated with 1 μM 17AAG for 24 h and analyzed by western blot. (H) FaDu cells were transfected MYC-His-TBK1 together with HA-Ub. After 24 h, cells were treated with 1 μM 17AAG or 1 μM HSP990 combined with 10 μM MG132 for 12 h, and then subjected to Ni-nta affinity-isolation ubiquitination assay under denaturing conditions. The obtained affinity-isolation samples were subjected to western blot using an anti-ha antibody. (I) After transfecting FaDu cells with NC or *MUL1* siRnas, and either MYC-His-wt TBK1, MYC-His TBK1^S172A, or MYC-His-TBK1^K38A. Cells were treated for 12 h with 1 μM 17AAG alone or combined with 10 μM MG132, and then subjected to Ni-nta affinity-isolation ubiquitination assay, followed by western blot with an anti-ha antibody. (J) Schematic model showing how HSP90 could regulate TBK1 stability and activity.

TBK1 inhibition perturbs autolysosome formation and selective autophagic clearance of aggregates

To evaluate the involvement of autophagy in HNC progression, we examined autophagy marker protein levels in tissue samples from HNC patients using western blotting. Notably, the extent of autophagy, as indicated by increased p-SQSTM1 and SQSTM1 levels and MAP1LC3B-II formation, was higher in tumor samples than in non-tumor samples (Figure S3A). TBK1 has been associated with selective autophagy receptors such as OPTN [64], CALCOCO2 [52,65], and SQSTM1 [29,50]; however, other studies have suggested a more generalized role in autophagosome maturation [66] or trafficking events associated with ATG9A (autophagy related 9A) [67]. Recent studies demonstrated that TBK1 is involved in the initiation step of autophagy and is essential for ULK1 complex assembly under basal conditions [68,69]. Moreover, TBK1 phosphorylates STX17 (syntaxin 17), thereby controlling the formation of the ATG13⁺ RB1CC1/FIP200⁺ mammalian phagophore assembly sites in response to autophagy induction [70]. These findings indicate that TBK1 plays crucial roles in multiple stages of autophagy. Hence, we considered a model in which these factors could play an active role in regulating the autophagic pathway in HNC. Next, we found that TBK1 KD or GSK8612 treatment resulted in a significant accumulation of SQSTM1 and MAP1LC3B-II, indicating that TBK1 depletion may impact autophagic flux (Figure 4A,D,G; Figure S4A,B,D). Furthermore, the disrupted autophagic flux observed after TBK1 inhibition did not seem to be linked to lysosomal defects, as we observed no significant differences in lysosomal acidification following GSK8612 treatment (Figure S3B). Additionally, the transcriptional regulation of SQSTM1 and MAP1LC3B was unchanged in response to TBK1 inhibition (Figure S3C). Despite exposure to 3 MA, an inhibitor of autophagosome formation, the accumulation of SQSTM1 and MAP1LC3B-II induced by GSK8612 was not eliminated (Figure S4A).

We monitored the autophagic flux using the MAP1LC3B-II turnover assay [71] (Figure 4A,D,G) and mCherry-GFP-LC3B reporter [72] (Figure 4B,E,H). Prolonged and strong proteasome inhibition causes cells to induce mRNAs for nearly all ATG genes (e.g., ATG7 and MAP1LC3B) and autophagy receptors to degrade ubiquitinated proteins [73]. First, to determine whether TBK1 affects the proper initiation of autophagy, we induced autophagy using a proteasome inhibitor [42] in TBK1 KD cells. In the presence of the inhibitor, MAP1LC3B-II levels were significantly higher in TBK1 KD cells compared to control cells (Figure 4A). Similar observations were made when cells with TBK1 KD or inhibition were incubated with torin 1, which facilitates autophagosome formation (Figure 4D; Figure S4B), confirming that TBK1 inhibition may affect a different phase of autophagic flux, i.e., one that is independent of the upstream induction stage. TBK1 KD or GSK8612 treatment in combination with the lysosomal inhibitor BaF did not result in a further increase in the level of MAP1LC3B compared to BaF treatment alone (Figure 4G; Figure S4D). Second, during autophagy induction (in the PI- or torin1-treated group), there was a significant increase in red-only puncta, indicating the presence of autolysosomes/ALs. However, TBK1 inhibition (through TBK1 KD or GSK8612 treatment) hindered the formation of red-only puncta and instead led to a dramatic increase in yellow puncta, indicative of autophagosomes (APs) (Figure 4B,C,E,F; Figure S4C). In cells where autophagosome-lysosome fusion or lysosomal function is impaired (such as BaF-treated cells), GFP puncta will not be quenched, resulting in the observation of more yellow puncta. Interestingly, after GSK8612 treatment or TBK1 KD, we observed a similar pattern to that in BaF-treated cells, indicating that inhibiting TBK1 hindered the autophagic flux (Figure 4H,I; Figure S4E). Collectively, these results strongly suggested that chemical or genetic inactivation of TBK1 impairs autophagic degradation of MAP1LC3B in the late stages of autophagic flux, rather than upregulating overall cellular autophagy. Immunofluorescence microscopy analysis indicated that the induction of GSK8612-mediated SQSTM1 punctum formation coincided with that of MAP1LC3B-positive cytosolic puncta (Figure 4J). GSK8612-induced puncta colocalized with each other (Figure 4L), suggesting that GSK8612 stimulated the accumulation of undegraded autophagosomes and impeded autophagic flux. We further assessed autophagosome-lysosome fusion by investigating the colocalization of MAP1LC3B with LAMP1 (lysosomal associated membrane protein 1). Notably, MAP1LC3B punctum formation increased in FaDu cells upon treatment with GSK8612, but the puncta did not colocalize with LAMP1 (Figure 4K,M). These findings indicated that GSK8612 blocked autophagosome-lysosome fusion. Transmission electron microscopy further revealed that GSK8612 led to substantial autophagosome accumulation and enlargement (Figure 4N).

We investigated the role of TBK1 in the autophagic degradation of poly-Ub proteins through SQSTM1, a protein essential for autophagic clearance of ubiquitinated proteins and aggregates [29,50,74]. FaDu cells were pretreated with PI+Tg for 16 h (pre) and then incubated in fresh growth medium supplemented with GSK8612 for another 24 h (post) [75]. Endogenous SQSTM1 after immunoprecipitation with an anti-SQSTM1 antibody showed the presence of endogenous poly-Ub proteins. The relative amount of poly-Ub protein that co-immunoprecipitated was lower in the GSK8612-treated group than in the PI+Tg-treated group (Figure 4O). These results suggested that GSK8612 reduced the affinity between SQSTM1 and poly-Ub proteins by inhibiting SQSTM1 phosphorylation. We further evaluated the function of MUL1 in the autophagic degradation of poly-Ub proteins through SQSTM1. FaDu cells were treated with PI+Tg for 16 h (pre) and then transfected with a cDNA encoding Flag-MUL1 or empty vector (Mock) in fresh medium for another 24 h (post). The relative amount of poly-Ub proteins that co-immunoprecipitated was lower in the MUL1 overexpression group than in the PI+Tg-treated group (Figure S4F). MUL1-induced TBK1 degradation inhibited the binding of SQSTM1 to poly-Ub cargo by reducing SQSTM1 phosphorylation. In addition, immunostaining analysis showed that PI-induced Ub conjugates formed cytosolic puncta that colocalized with SQSTM1 (Figure 4P). GSK8612-treated or MUL1-overexpressing cells showed impaired Ub conjugate formation in SQSTM1-positive puncta upon proteasomal inhibition (Figure 4P,Q; Figure S4G), indicating that S403-phosphorylated SQSTM1 could stably store ubiquitinated proteins in the sequestosome. Next, we observed that the degradation of oxidatively carbonylated proteins was significantly accelerated in cells overexpressing Ub [42]. This could be blocked by genetic ablation of Atg5, GSK8612 treatment, or MUL1 overexpression (Figure 4R,S; Figure S4H), indicating that cellular carbonylated (damaged) proteins were degraded via the autophagy pathway. Accordingly, GSK8612 treatment to inhibit autophagy led to the stabilization of oxidized proteins. In summary, these results provided evidence that proper TBK1 activity is required for autophagosome-lysosome fusion and selective autophagic clearance of protein aggregates upon proteotoxic stress.

Elevated expression and predominant cytosolic localization of G3BP1 and PABP in HNC patients

Cancer cells use SGs to adapt to various stresses, especially those encountered within the tumor microenvironment and during chemotherapy. Remarkably, elevated SG marker expression has been reported in various human cancers and predicts a poor prognosis [76]. We investigated the potential of SGs as biomarkers of proliferation and apoptosis in HNC. Fourteen pairs of tumor and adjacent non-tumor head and neck tissue samples were obtained from HNC patients. Western blotting analysis revealed that the protein levels of two SG markers, G3BP1 and PABP, were higher in tumor (T) samples than in non-tumor (N) samples (Figure 5A-C), suggesting that elevated SG formation is a pathological feature of HNC. Through H&E staining, we distinguished tumor (T1 and T2) from non-tumor tissues (N1 and N2) in the overall HNC patient tissue images. Immunofluorescence staining of the paired HNC sections using antibodies against G3BP1 and PABP confirmed that G3BP1 and PABP expression was higher in tumor regions than in non-tumor regions, with distinct puncta in the cytosol (Figure 5D,E).

Figure 5. — High levels of G3BP1 and PABP expression were found in HNC. (A) G3BP1 and PABP expression from both HNC tissues (T) and their adjacent non-tumor head and neck tissues (N) were analyzed using western blotting. Representative western blot with anti-G3BP1 and anti-pabp antibodies, GAPDH as loading control. Quantification of the G3BP1:GAPDH (B) and PABP:GAPDH (C) (14-pair HNC cohort, ****p < 0.0001 by the mann-whitney t test). (D) A digital histologic image of a hematoxylin and eosin (H&E)-stained slide of squamous cell carcinoma (SCC) at the oral cavity (tongue). Representative immunofluorescence staining of G3BP1 (green) and PABP (red) was performed to determine the SG formation in HNC tissues and adjacent non-tumor head and neck tissues (yellow; merge/colocalization, 10-pair HNC cohort). Scale bar: 200 μm and 20 μm (inset). (E) the average percentage of cells with G3BP1 and pabp-positive SGs in tumor tissues was counted. Average percentages of SGs ± SD for each tumor is plotted based on individual values. The colocalization of G3BP1 and PABP signal in cytosol was quantified by analyzing the merged images from (D) using NIS Elements software (5 fields of view imaged at 60× in each of 10-pair HNC tissues, ****p < 0.0001 by the Mann-Whitney t-test).

TBK1 is required for SG formation in stressed HNC cells

Because of the heterogeneous composition of SGs and the crowded molecular environment, SGs may, indirectly, require PQC assistance for proper assembly and disassembly in cancer. A number of SG components have a role in PQC, including ubiquitin and E3 ubiquitin ligases, while proteasome inhibition induces SGs [77]. One essential and ubiquitous molecular chaperone required for the folding and maturation of a large variety of proteins, including SG components, is HSP90 [78]. HSP90 has been identified as a TBK1 interactor (Figure 3) [60], suggesting that HSP90 may act upstream of TBK1 to regulate SG assembly. However, whether TBK1 regulates SG dynamics remained unknown. We employed immunofluorescence microscopy to examine G3BP1 and PABP levels to establish that TBK1 is required to maintain the activity of key factors that regulate SG formation. The results revealed that the HSP90 inhibitor 17AGG decreased TBK1 activity (Figure 3) and the number of cells displaying SGs in FaDu cells experiencing proteotoxic stress induced by MG132 (Figure 6A,B). SGs were absent in the absence of stress stimuli. Furthermore, treatment with 17AGG and GSK8612 reduced the percentage of SG-positive FaDu cells by more than 91% and 60%, respectively (Figure 6B). These data suggested that HSP90 inhibition may induce an unstable conformation in TBK1, redirecting it to proteasomal degradation (Figure 3). Thus, a reduction in TBK1 levels as a result of HSP90 inhibition suppresses the expression of SGs. Upon TBK1 inhibition, SG assembly in SCCQLL1 and SCC25 hNC cells decreased by approximately 50% and 70%, respectively (Figure S5A,B). Similar observations were made after siRNA-mediated knockdown of TBK1 (Figure 6C). In FaDu cells exposed to MG132, TBK1 knockdown attenuated SG formation by approximately 32% (Figure 6D). Therefore, we investigated whether TBK1 influences the expression of SG-associated proteins. G3BP1 and PABP levels were not altered in TBK1 KD cells (Figure S5C).

Figure 6. — TBK1 inhibition impairs SG formation and sensitizes HNC cells to cisplatin. (A) FaDu cells were pre-treated with 1 μM 17AGG or 20 μM GSK8612 for 3 h. Then, 20 μM MG132 was added, and the cells were incubated for an additional 6 h. After that, the cells were fixed and representative immunofluorescence staining for G3BP1 (green) and PABP (red) was performed in FaDu cells to determine the SG formation. Scale bar: 10 μm. (B) The average percentage of cells with G3BP1 and pabp-positive SGs was counted (n = 10 fields, each field has at least 8 cells that meet statistical requirements). Average percentages of SGs and SD for three independent experiments are plotted. **p = 0.0013, ****p < 0.0001 by unpaired t test. (C) FaDu cells were transfected with NC or *TBK1* siRnas. After 24 h, 20 μM MG132 was added, and the cells were incubated for an additional 6 h. Cells were fixed and labeled with anti-G3BP1 and anti-pabp antibody. Scale bar: 10 μm. (D) The average percentage of cells with G3BP1 and pabp-positive SGs was counted (n = 10 fields). Average percentages of SGs and SD for three independent experiments are plotted. **p = 0.0066 by unpaired t test. (E) FaDu cells treated for 6 h with 20 μM MG132 alone or with chloroquine (CQ) were fixed and labeled with anti-G3BP1 and anti-pabp antibody. Scale bar: 10 μm. (F) The average percentage of cells with G3BP1 and pabp-positive SGs was counted (n = 10 fields). Average percentages of SGs and SD for three independent experiments are plotted. ****p < 0.0001 by unpaired t test. (G) FaDu cells were transfected with NC, *MAP1LC3B, ATG5, or SQSTM1* siRnas. After 24 h, 20 μM MG132 was added, and the cells were incubated for an additional 6 h. Cells were fixed and labeled with anti-G3BP1 and anti-pabp antibody. Scale bar: 10 μm. (H) The average percentage of cells with G3BP1 and pabp-positive SGs was counted (n = 10 fields). Average percentages of SGs and SD for three independent experiments are plotted. **p = 0.0019, **p = 0.0034 vs. NC by unpaired t test. (I) FaDu cells were pre-treated with 100 nM BaF for 3 h. Then, 20 μM MG132 was added, and the cells were incubated for an additional 6 h. After that, the cells were fixed and representative immunofluorescence staining for G3BP1 (red) and MAP1LC3B (green) or SQSTM1 (red) and PABP (green) was performed in FaDu cells to determine the SG formation. Scale bar: 10 μm and 1 μm (inset). (M) HNC cells transfected with NC or *TBK1* siRNAs were treated with 10 μM cisplatin for 24 h, and apoptosis was measured by ANXA5/Annexin V and PI flow cytometry (n = 3, unpaired t-test). (N) HNC cells transfected with NC or *TBK1* siRNAs were treated with 0, 1, 2.5, 5, or 10 μM cisplatin for 24 h and analyzed by western blot using anti-p-TBK1, anti-TBK1, and anti-cleaved CASP3 antibodies, with GAPDH as loading control. (O) FaDu cells transfected with NC, *TBK1*, *ATG5*, or *G3BP1* siRNAs were treated with 10 μM cisplatin for indicated time (24 h or 48 h), and cell viability was measured using the CCK-8 assay (n = 3). For 24 h, ***p = 0.0008, ****p < 0.0001, and ***p = 0.0003 vs. NC; **p = 0.0067, and ***p = 0.0008 vs. cis; ^##p = 0.0080 (siTBK1+cis vs. siATG5+cis); and ^#p = 0.0415 (siTBK1+cis vs. siG3BP1+cis). For 48 h, ****p < 0.0001, **p = 0.0014, and ***p = 0.0001 vs. NC; ***p = 0.0003, *p = 0.0299, and ***p = 0.0002 vs. cis; ^##p = 0.0031 (siTBK1+cis vs. siATG5+cis); ns (non-significant) for siTBK1+cis vs. siG3BP1+cis, determined by unpaired t-test. (P) FaDu cells transfected with NC, *TBK1*, *ATG5*, or *G3BP1* siRNAs were treated with 10 μM cisplatin for 48 h. Protein expression levels were analyzed by western blot using antibodies against p-TBK1, TBK1, and cleaved CASP3, with GAPDH serving as the loading control.

SG formation occurs as a consequence of, but is not required for, translation arrest. Translational arrest was induced by incorporating puromycin into nascent peptides. Results revealed puromycin-labeled proteins were detected in control cells, but not after treatment with MG132 alone or in combination with the TBK1 inhibitors GSK8612 or compound 1 [79] (Figure S5D). This finding indicated that TBK1 regulates SG formation via an EIF2A phosphorylation-independent mechanism.

Autophagy, lysosomes, and VCP inhibition affect SG morphology and composition, implying that proteostasis imbalances directly impact SG formation [25]. Chloroquine (CQ), a compound known to block autophagosome-lysosome fusion, suppressed SG formation by approximately 96% when co-administered with MG132 in FaDu cells (Figure 6E,F). This suggests that the abnormal accumulation of MAP1LC3B observed after GSK8612 treatment may be linked to reduced autophagic flux (Figure 4). Consistent with this, GSK8612 treatment increased the expression of autophagic marker proteins, MAP1LC3B and SQSTM1, in a time-dependent manner (Figure S5E). To identify which autophagy-related proteins contribute to SG dynamics, we knocked down MAP1LC3B, ATG5, and SQSTM1 and compared their effects on MG132-induced SG formation. Knockdown of MAP1LC3B (~41%) and ATG5 (~48%) significantly reduced SG formation, indicating their involvement in SG regulation. In contrast, siSQSTM1 had no effect, suggesting that SQSTM1 does not directly influence SG formation under these conditions (Figure 6G,H). Interestingly, MAP1LC3B was observed to partially colocalize with SGs during their formation, While SQSTM1 did not, highlighting the distinct roles of these autophagic markers in SG dynamics under stress conditions (Figure 6I). Further analysis revealed that BaF pretreatment, which increases MAP1LC3B and SQSTM1 levels by inhibiting lysosomal degradation, led to the accumulation of lipidated MAP1LC3B and SQSTM1. However, neither MAP1LC3 nor SQSTM1 localized to SGs during proteotoxic stress (Figure 6I), suggesting that MAP1LC3B plays a critical role in SG formation and that TBK1 may regulate SG dynamics indirectly through its control of autophagic flux.

We postulated that TBK1-mediated control of SG formation may enhance the sensitivity of HNC cells to cisplatin, which is a standard chemotherapeutic drug [80,81]. To investigate this, we examined the effects of cisplatin on SG formation in FaDu cells. Cisplatin treatment significantly induced SG formation (Figure 6J), while TBK1 KD reduced SG-positive cells by approximately 46% (Figure 6K). Furthermore, cisplatin treatment led to a decrease in SQSTM1 levels and an increase in MAP1LC3B levels, indicating the activation of autophagy in FaDu cells (Figure 6L). These findings suggest that TBK1 plays a critical role in modulating SG dynamics and autophagy during cisplatin treatment, potentially influencing the therapeutic response of HNC cells. To further explore this, in vitro assays were conducted using FaDu, Detroit562, SNU1041, and SCC25 cells to evaluate the combinatorial effect of GSK8612 and cisplatin on HNC cell death. While incubation of HNC cells with low-dose cisplatin alone for 24 h slightly reduced cell viability, cisplatin cytotoxicity was augmented in TBK1-depleted cells (Figure S5F). Flow cytometry revealed a 4- and 3-fold increase in ANXA5/Annexin V-FITC-positive apoptotic cells in TBK1-depleted FaDu and SCC25 cells, respectively, when treated with cisplatin (Figure 6M). CASP3 (caspase 3) cleavage was markedly increased by the combination treatment (Figure 6N), as shown by western blotting, confirming the synergistic effect of GSK8612 and cisplatin on HNC cell death. Similar results were obtained in FaDu cells treated with PI (Figure S5G-I).

To clarify whether TBK1 enhances cisplatin-induced cell death through inhibition of autophagy flux or attenuation of SG formation, we examined the effects of siTBK1, siATG5, and siG3BP1 on cell viability (Figure 6O) and apoptosis (Figure 6P). At 24 h, knockdown of G3BP1, resulted in the largest reduction in cell viability (~60%), followed by siTBK1 (~50%) and siATG5 (~30%). These results suggest that SG formation, primarily regulated by G3BP1, plays a more immediate role in cisplatin sensitivity compared to autophagic flux, which is modulated by ATG5. At 48 h, knockdown of ATG5 led to a ~ 60% reduction in cell viability, while siTBK1 and siG3BP1 both caused a similar and more pronounced reduction of ~ 80%. Consistently, cleaved CASP3 levels were higher in siTBK1- and siG3BP1-treated cells compared to siATG5-treated cells (Figure 6P). These results suggest that G3BP1-mediated SG formation has a stronger early impact on cisplatin sensitivity, while the effects of TBK1 knockdown reflect its broader role in modulating both SG dynamics and autophagy over time. Interestingly, in the absence of cisplatin, knockdown of TBK1, ATG5, or G3BP1 reduced proliferation compared to control cells, with siTBK1-treated cells showing the slowest proliferation rate (Figure S5J). This suggests that TBK1 may have additional roles beyond its regulation of autophagy and SG formation, particularly under basal conditions where stress-induced pathways are less active.

To further investigate the role of SG formation in cisplatin-induced apoptosis, we evaluated the effects of G3BP1 inhibition using the small molecule inhibitor FAZ-3532 [82] and cisplatin significantly increased ANXA5-FITC-positive apoptotic cells, with approximately a 3-fold increase compared to cisplatin alone (Figure S5K). Additionally, tumorsphere assays revealed that FAZ-3532 treatment reduced tumorsphere size and increased global cell death (Figure S5L,M). These findings emphasize the critical role of G3BP1 and SG formation in modulating cisplatin-induced apoptosis and highlight the potential therapeutic benefit of targeting SG dynamics in combination with chemotherapeutic agents.

A novel TBK1 inhibitor, GSK8612, suppresses xenograft tumor growth in vivo

To assess the role of TBK1 in tumorsphere formation, FaDu cells harboring an siRNA against TBK1, or a negative control siRNA were cultured as tumorsphere (Figure S6A). The tumorsphere size decreased in TBK1-depleted spheroids (Figure S6B). A LIVE/DEAD assay of tumorspheres revealed that TBK1 KD increased the number of global dead cells (Figure S6C), suggesting that TBK1 may promote the proliferation and survival of HNC cells.

Given that TBK1 regulated SG formation in cultured HNC cells (Figure 6), we determined the relevance of TBK1 in SG formation and whether the inhibitory effect of GSK8612 on HNC progression occurred in vivo. Tumor growth was significantly attenuated in FaDu-bearing mice treated with GSK8612 compared to control mice. Tumor growth was inhibited after the seventh injection, and the anticancer effect was sustained for 10 days (Figure 7A). Tumor weight in the GSK8612-treated group was reduced by approximately 50% compared with that in the control group (Figure 7B), whereas body weight did not significantly differ between the two groups (Figure 7C). To determine whether GSK8612 induced HNC cell death through autophagy inhibition and regulation of SG formation, we analyzed the changes in the levels of p-TBK1, TBK1, p-SQSTM1, SQSTM1, MAP1LC3B, and cleaved CASP3 by western blotting (Figure 7D,E). p-TBK1 expression in tumors was significantly reduced in GSK8612-treated mice compared to control mice, and GSK8612 strongly correlated with the reduction in p-SQSTM1 expression in the same lysates. We observed ~ 2.9- and ~ 2-fold increases in the levels of the autophagy markers SQSTM1 and MAP1LC3B-II, respectively, in GSK8612-treated tumors, indicating the accumulation of protein aggregates. Cleaved CASP3 expression was significantly increased in the treatment group (~10-fold), suggesting that GSK8612 induced apoptosis. MAP1LC3B and SQSTM1 puncta, reminiscent of autophagosome formation, were detected in GSK8612-treated tumors, whereas diffuse MAP1LC3B staining was observed in control tumors (Figure 7F). Lipidation and clustering of MAP1LC3B may be the result of both the induction and suppression of autolysosomal maturation. The cargo protein SQSTM1 is a useful marker to confirm that autophagosome accumulation is due to autophagy inhibition rather than autophagy induction. GSK8612-treated tumors showed increased SQSTM1 immunofluorescence, indicating the accumulation of MAP1LC3B-positive aggregates (Figure 7G).

Figure 7. — GSK8612 inhibition impairs autophagic flux, SG formation and tumor growth in HNC xenograft. (A) subcutaneous injection of 5 × 10⁶ FaDu cells was performed in BALB/c nu/nu mice, followed by oral administration of GSK8612 every day for 10 days. Data are mean tumor volume at endpoint ± SD and individual tumor volume for each mouse. CON n = 7, GSK8612 n = 9. **p < 0.01 by the mann-whitney t test. (B) top, the endpoint images of tumors formed by CON and GSK8612 in BALB/c mice. Bottom, data are mean tumor weight at endpoint ± SD and individual tumor weights for each mouse. CON n = 7, GSK8612 n = 9. **p = 0.0046 by the mann-whitney t test. (C) data are mean body weight at endpoint ± SD. ns; non-significant by the mann-whitney t test. (D) proteins isolated from tumor tissues from (A) were subjected to western blot analysis using the indicated antibodies, GAPDH as loading control. (E) the amounts of p-TBK1 (***p = 0.0006 by the mann-whitney t test), TBK1 (ns; non-significant), p-SQSTM1 (**p = 0.0070 by the mann-whitney t test), SQSTM1 (*p = 0.0111 by the mann-whitney t test), MAP1LC3B (*p = 0.0379 by the mann-whitney t test), and cleaved CASP3 (***p = 0.0006 by the mann-whitney t test) were measured and normalized to GAPDH. CON n = 7 and GSK8612 n = 7. (F) colocalization of SQSTM1 (green) and MAP1LC3B (red) in tumor sections. Arrowheads, autophagosomes (yellow). Scale bar: 10 μm. (G) the average percentage of cells with the SQSTM1⁺MAP1LC3B⁺ puncta per field (n = 10 fields, each field has at least 20 cells that meet statistical requirements). Average percentages of autophagosomes ± SD for each tumor is plotted based on individual values. CON n = 7, GSK8612 n = 7. ***p = 0.0006 by the mann-whitney t test. (H) Representative immunofluorescence staining for G3BP1 (green) and PABP (red) in tumors. Scale bar: 20 μm and 10 μm (inset). (I) the average percentage of cells with G3BP1 and pabp-positive SGs in tumor tissues was counted (n = 4 fields, each field has at least 100 cells that meet statistical requirements). Average percentages of SGs ± SD for each tumor is plotted based on individual values. CON n = 7, GSK8612 n = 7. **p = 0.0070 by the mann-whitney t test. (J) Representative immunofluorescence staining of MKI67 (green)- or cleaved CASP3 (red)-positive areas in tumors. Scale bar: 100 μm. (K) quantification of MKI67- or cleaved CASP3-positive areas in tumors. The values for each tumor represent the average of 10 fields of view imaged at 20× and covering ~ 50% of each section. Data are mean MKI67 or cleaved CASP3 area over tumor area ± SD of individual tumors in mice. CON n = 7, GSK8612 n = 7. ***p = 0.0006 by the mann-whitney t test.

Immunofluorescence of tumor sections revealed that SGs were significantly reduced in GSK8612-treated tumors compared to control tumors (Figure 7H). Quantification of the SG index by computing the cell area occupied by SGs as a fraction of the total cell area showed that inhibition of TBK1 attenuated SG formation by ~ 1.6-fold (Figure 7I). Further, we evaluated whether the differential impact of SG inhibition on tumor growth was due to cell death and/or proliferation. SG inhibition led to higher cell death levels in GSK8612-treated mice than in control mice (Figure 7J). Quantification of the fraction of tumor area positive (+) for cleaved CASP3 and MKI67 indicated a ~ 13-fold increase in cell death and a ~ 3-fold decrease in proliferation in GSK8612-treated tumors compared to control tumors (Figure 7K) These results indicated that SGs may contribute to both cancer cell proliferation and survival in vivo. Taken together, these findings suggested that TBK1 contributes to HNC progression by promoting SG formation and is a potential therapeutic target in HNC.

Discussion

The critical role of TBK1 in tumorigenesis and the association between aberrant TBK1 expression and prognosis in different cancers have been reported [83]. Data from The Cancer Genome Atlas and Gene Expression Omnibus databases consistently demonstrated that TBK1 is upregulated in HNC compared to normal tissues [84]. However, little was known about the effect of TBK1 on HNC progression. We investigated the role of TBK1 in two cellular processes, protective autophagy and SG formation, during HNC progression, and report several novel findings that address critical gaps in current knowledge. First, the mechanism by which TBK1 activation is terminated in HNC was not fully understood. We identified MUL1 as a novel E3 ligase targeting TBK1. MUL1 induced K48-linked ubiquitination of lysine 584 and degradation of active TBK1 through the UPS. Second, we showed that HSP90 plays an essential role in maintaining TBK1 stability. By destabilizing TBK1 and preventing its reactivation, HSP90 inhibition redirects the client kinase toward ubiquitination and degradation by MUL1, which then redirects inactive TBK1 to an alternative destination similar to that observed following direct inhibition of TBK1 by GSK8612. Further, in HNC, hyperactivated TBK1 is important for regulating autophagic flux and p-SQSTM1-dependent clearance of aggregates. We also uncovered new functions of TBK1 in regulating SG assembly through autophagic machinery and in protecting HNC cells from proteotoxic stress and cisplatin, both in vitro and in vivo. Further, we evaluated the potential antitumor activities of the selective TBK1 inhibitor GSK8612 in HNC xenografts. The mechanisms that lead to TBK1 hyperactivation in HNC, and whether this activation is related to its role in innate immunity, remain unclear. Here, we characterized the role of TBK1 as a stress-adaptive signaling hub in HNC, which is distinct from its role in innate immune signaling.

Suppression of MUL1 expression is a hallmark of HNC [35]. We have previously reported that MUL1 plays a pivotal role in HNC progression by inducing the degradation of oncoproteins such as HSPA5 and AKT [34,35,37]. When acting as a tumor suppressor, MUL1 uses TBK1 as a substrate by ubiquitinating K584 for UPS. In the present study, MUL1 deficiency resulted in higher amounts of p-TBK1 and defective TBK1 ubiquitination in response to proteotoxic stressors, including proteasome inhibitors, HSP90 inhibitors, and Ub overexpression (Figures 1–4). TBK1 is inactive until adaptor proteins recruit it to signaling complexes, where it can be auto-phosphorylated due to a high local concentration [85] or phosphorylated by other kinases [86] localized to the same molecular scaffold. Given the robust auto-phosphorylation capabilities of TBK1 and the potential oncogenic effect thereof, important regulatory mechanisms are in place to prevent TBK1 activation in the absence of pathway stimulation [48]. Although the mechanisms regulating TBK1 activation have been studied extensively, the mechanisms regulating its termination are less well understood. TBK1 does not appear to be ubiquitinated until after association with proteotoxic stress, but MUL1 may interact with TBK1 in resting cells. Moreover, 17AAG, which selectively inhibits HSP90 substrate interactions, specifically released HSP90 from multiple chaperone complexes with TBK1 [61] and stimulated the MUL1-mediated translocation of TBK1 from the cytosol to the mitochondrial outer membrane and its subsequent degradation. How MUL1 ubiquitinates TBK1 in the presence of stress is yet to be determined, although a conformational change or chaperone interaction is thought to facilitate this process. However, the precise function of these factors in mitochondrial targeting requires further investigation.

Our findings suggest a novel link between TBK1-mediated SG formation and HNC progression, as TBK1 inhibition dramatically reduced autophagic functions and selective autophagy in HNC cells. Additionally, we confirmed the presence of elevated levels of G3BP1 and PABP with excessive SG formation in HNC tissues obtained from human patients. Cancer responds to stresses, such as oxidative and metabolic stress or hypoxia, via protein aggregation by activating PQC and attenuating translation [2,3,5]. PQC involves chaperones and degradation systems [3] and is essential in the proteotoxic stress response. The findings that SGs are targeted for ubiquitin-dependent autophagy for degradation by phosphorylation of VCP by ULK1 or ULK2 [87], SG disassembly and restoration of translation activity requires chaperone (Hsp104 and HSPA/HSP70)-driven protein disaggregation [88], and MG132-induced SGs were significantly reduced in atg5^–/– or atg16l1^–/– mouse embryonic fibroblasts (MEFs) compared to wild-type MEFs [25], consistent with the idea that autophagy may assist SG dynamics. This suggests that specific components may need to be extracted from autophagosomes, where SGs are targeted for lysosomal degradation. Our data demonstrate that inhibition of TBK1 impairs SGs (Figure 6) through autophagic flux (Figure 4), supporting that inhibition of autophagy, lysosomes, and VCP function modulates SG formation [25]. A challenge for future work will be to identify how TBK1 and co-factors orchestrate the selection of components to be extracted from SGs and to understand the components of the molecular machinery for autophagy in SGs [25,87]. Most chemotherapies induce SGs, promoting the acquisition of therapy resistance in cancers. Disrupting SG formation may represent an effective strategy to enhance cancer therapeutic effectiveness [17,89]. We showed that TBK1 inhibition can enhance the sensitivity of HNC cells to undergo cell death upon treatment with a low dose of cisplatin or PI.

Over the past decade, research interest in TBK1 has expanded, and many small molecules targeting TBK1 have been identified and developed. Although most compounds are quite potent toward TBK1 and its homolog IKBKE/IκB kinase ε, not all are highly selective [90]. GSK8612, a recently developed highly selective TBK1 inhibitor [63], was used in our study and selectively killed HNC cells and xenografts. A recent report showed that depletion of TBK1 in lung cancer reduced the number of both CD274/PD-L1-expressing and myeloid-derived suppressor cells in the tumor microenvironment, which was associated with a local increase in CD8⁺ T cells [91]. IFN gamma-inducible protein 16 promotes HPV⁺ cervical cancer progression by upregulating CD274/PD-L1 via the STING1-TBK1-NFKB/NF-κB pathway [92]. Similarly, GSK8612 attenuates hepatocellular carcinoma progression by enhancing tumor immune infiltration of CD8⁺ T cells. Targeting TBK1 boosted the efficacy of immune checkpoint blockade in murine- and patient-derived organotypic tumor spheroids [79,93]. Therefore, targeting TBK1 may not only inhibit tumor cell-autonomous signaling but also potentially overcome resistance to immunotherapy in HNC.

Our study elucidated the multifaceted role of TBK1 in HNC progression. Initially, we discovered that the E3 ligase MUL1 triggers K48-linked ubiquitination of TBK1 at lysine 584, leading to its degradation. When HSP90 is inhibited, TBK1 dissociates from HSP90 and subsequently undergoes UPS-mediated degradation, a process facilitated by MUL1. Inhibiting TBK1 negatively affects SG formation by disrupting autophagic flux, thereby enhancing apoptosis in response to cisplatin or proteotoxic stress. These findings greatly enhance our understanding of how TBK1 contributes to HNC progression and highlight its potential as a therapeutic target. Further investigation is needed to unravel the mechanisms underlying TBK1 hyperactivation in HNC and its potential interactions with innate immunity pathways.

Materials and methods

Reagents

For in vitro experiments, bortezomib (Selleckchem, S1013), thapsigargin (Sigma-Aldrich, T9033), lactacystin (Sigma-Aldrich, L6785), epoxomicin (Sigma-Aldrich, E3652), MG132 (Selleckchem, S2619), cycloheximide (Cell Signaling Technology, 2112), 17AAG (Selleckchem, S1141), AUY922 (Selleckchem, S1069), HSP990 (Selleckchem, S7097), bafilomycin A₁ (Selleckchem, S1413), chloroquine (Selleckchem, S6999), GSK8612 (Selleckchem, S8872), torin 1 (Selleckchem, S2827), compound 1 (Selleckchem, S8922), and puromycin (Sigma-Aldrich, P9620) were dissolved in DMSO (Sigma-Aldrich, D2650) and further diluted to the required concentration. 3-Methyladenine (Selleckchem, S2767) was dissolved in water and then diluted in the medium to the specified concentration, while cisplatin (Selleckchem, S1166) was dissolved in normal saline and also diluted in the medium to the specified concentration. For in vivo experiments, the stock solution concentration is 20 mg/ml GSK8612 in DMSO. The working solution (5% stock solution, 40% PEG300 [Selleckchem, S6704], 5% Tween 80 [Selleckchem, S6702] and 50% water) should be freshly prepared and used on the same day.

Cell culture and transfection

Human HNC cell lines FaDu (HTB-43), SCC15 (CRL-1623), SCC25 (CRL-1628), Detroit562 (CCL-138), and Cal27 (CRL-2095) were purchased from American Type Culture Collection. SNU1041 (01041) and SNU899 (00899) cells were obtained from the Korean Cell Line Bank. MSKQLL1, SCCQLL1, and SCC1483 cells were gifted from Dr. Se-Heon Kim (Yonsei University, Korea). The atg5^−/− and the corresponding WT MEFs were provided by Dr. You-Sun Kim (Ajou University, Korea). MUL1 KO FaDu is a monoclonal cell line generated in our laboratory [35]. FaDu, Detroit562, SCCQLL1, and SCC1483 cells were cultured in MEM (Gibco 11,095,080). SNU1041 and SNU899 cells were cultured in RPMI 1640 (Gibco 22,400,089). SCC15, SCC25, and MSKQLL1 cells were cultured in DMEM/F12 (Gibco 11,320,033). Cal27 and MEFs were cultured in DMEM (Gibco 11,965,092). All growth media were supplemented with 10% FBS (Gibco, A5670801) and antibiotic-antimycotic (Gibco 15,240,096) at 37°C with 5% CO₂ under humidified conditions. All cell lines were routinely verified to be mycoplasma negative using a mycoplasma detection kit (Lonza, LT07–318). For transfection, Lipofectamine LTX with PLUS reagent (Thermo Fisher Scientific,15338100) was used for transient transfection of plasmids, and RNAiMAX (Thermo Fisher Scientific 13,778,150) was used for transfection of small interfering RNAs based on the manufacturer’s instructions. Experiments were performed 24 and 48 h after transfection of cDNAs or siRNA, respectively. The siRNAs targeting TBK1 (L-003788-00-0010), MUL1 (L-007062-00-0010), G3BP1 (L-012099-00-0010), ATG5 (L-004374-00-0010), SQSTM1 (L-010230-00-0010), MAP1LC3B (L-012846-00-0010), HSP90 (L-005186-00-0010), STING1 (L-024333-00-0010), and non-targeting control (D-001810-01-50) were purchased from Dharmacon (SMARTpool: ON-TARGETplus siRNAs). Cells were transfected at a final concentration of 50 nM into the indicated HNC cells.

Plasmids

EGFP-MUL1, His-WT Ub, HA-WT Ub, HA-Ub-K48, HA-Ub-K63, HA-Ub^K48R, and HA-Ub^K63R were used as described previously [35,37]. pDEST-CMV mCherry-GFP-LC3B WT was obtained from Addgene (123230; deposited by Robin Ketteler). His-human TBK1 (Sino Biological, HG11023-NH) and human pCMV6-human MUL1 (OriGene, SC321149) were subcloned into pEF1α-MYC vector (TaKaRa 631,991) and pCMV-Flag (DYKDDDDK)-N vector (TaKaRa 635,688), respectively. Truncation mutants of TBK1; MYC-His-TBK1 (K), MYC-His-TBK1 (KU), MYC-His-TBK1 (CC), and MYC-His-TBK1 (UCC) were generated by PCR from MYC-His-WT TBK1. The TBK1 inactive mutants (MYC-His-TBK1^S172A and TBK1^K38A) and the TBK1 lysine mutants (MYC-His-TBK1^K416R, TBK1^K484R, TBK1^K504R, TBK1^K567R, TBK1^K570R, TBK1^K584R, TBK1^K628R, TBK1^K661R, TBK1^K670R, TBK1^K691R, TBK1^K692R, and TBK1^K694R) were generated using the QuikChange II Site-Directed Mutagenesis Kit (Agilent Technologies 200,524) according to the manufacturer’s instructions. All the mentioned constructs were fully verified by Sanger sequencing. PCR primer sequences in this study are shown in Table S1.

Immunoprecipitation assay

Twenty-four hours post-transfection, cells were lysed in lysis buffer (50 mm Tris-HCl, pH 7.4, 150 mm NaCl, 1 mm EDTA, 1 mm EGTA, 10% glycerol, and 1% Triton X-100 [Sigma-Aldrich, P1379]) containing a protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific 78,440) on ice for 30 min. Cell lysates were centrifuged at 12,000 × g for 30 min at 4°C. The supernatant was then precleared with protein A/G PLUS agarose beads (Santa Cruz Biotechnology, sc-2003) at 4°C for 4 h and subsequently mixed with washed agarose beads conjugated with anti-p-TBK1 (Ser172; Cell signaling Technology, 5483; D52C2), anti-SQSTM1 (Cell Signaling Technology 88,588; D5L7G), normal rabbit IgG control (Cell Signaling Technology, 2729), or mouse mAb IgG1 isotype control (Cell signaling Technology, 5415) at 4°C overnight. Immunocomplexes were washed extensively 4 times with washing buffer (50 mm Tris-HCl, pH 7.4, 150 mm NaCl, 1 mm EDTA, 1 mm EGTA, and 0.1% Triton X-100). Both co-immunoprecipitated proteins and input fractions were resolved on SDS-PAGE and analyzed by western blot.

Ni-NTA affinity-isolation assay

The ubiquitination status of MYC-His-tagged TBK1 protein, either wild-type (WT) or lysine mutant variants was determined by Ni-NTA affinity isolation or in vivo ubiquitination assays. Briefly, cells transfected with various constructs, together with MYC-His-TBK1 or HA-Ub, were treated with MG132 for 16 h and lysed in 200 μl of denaturing lysis buffer A (50 mm Tris-HCl, pH 7.4, 2% SDS, 70 mm β-mercaptoethanol) by vortexing and boiling for 20 min at 95°C. The lysates were diluted with 800 μl lysis buffer A (50 mm NaH₂PO₄, 300 mm NaCl, and 10 mm imidazole, pH 8.0) containing protease inhibitor cocktail and MG132. The diluted lysates were incubated with 50 μl of Ni-NTA agarose beads (Qiagen 30,230) overnight at 4°C. The beads were washed five times with buffer B (50 mm NaH₂PO₄, 300 mm NaCl, and 20 mm imidazole, pH 8.0), and bound proteins were eluted by boiling in a mixture of 5× SDS-PAGE loading buffer and buffer C (50 mm NaH₂PO₄, 300 mm NaCl, and 250 mm imidazole, pH 8.0) (1:4). Thereafter, ubiquitinated TBK1 was identified with anti-His Tag (Cell Signaling Technology 12,698), anti-MYC Tag (Cell Signaling Technology, 2276; 9B11), and anti-HA Tag (Cell signaling Technology, 3724; C29F4) antibodies on a western blot.

Western blot

Cells were lysed with RIPA buffer (25 mm Tris-HCl pH 7.4, 150 mm NaCl, 1% NP-40 [Thermo Fisher Scientific 85,124], 1% sodium deoxycholate [Thermo Fisher Scientific 89,905], 0.1% SDS [Thermo Fisher Scientific 15,553,027]) containing a protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific 78,440) on ice for 30 min. Total protein levels were quantified using a Pierce BCA Protein Assay Kit (Thermo Fisher Scientific 23,227), and the lysates were diluted to approximately equal concentrations before heating in SDS sample buffer (with a final concentration of 50 mm Tris-Cl, pH 6.8, 2% SDS, 10% glycerol, 5% β-mercaptoethanol, 0.01% bromophenol blue) at 95°C for 15 min. Equal amounts of sample (typically 20 μg total protein per lane) were separated by SDS-PAGE and transferred to a polyvinylidene difluoride/PVDF membrane (Thermo Fisher Scientific 88,518). Each membrane was blocked with 5% skim milk for 1 h at room temperature (RT) and incubated overnight with primary antibody (1:1,000) at 4°C. The following primary antibodies were used: anti-p-TBK1 (Ser172; 5483; D52C2), anti-TBK1 (38066; E8I3G), anti-GAPDH (5174; D16H11), anti-MYC Tag (2276 and 2272; 9B11), anti-Flag Tag M2 (14793; D6W5B, and 8146; 9A3), anti-HA Tag (3724; C29F4, and 2367; 6E2), anti-p-STING1 (Ser366; 50907; E9A9K), anti-STING1 (13647; D2P2F), anti-HSP90 (4877; C45G5), anti-MAP1LC3B (3868; D11), anti-p-SQSTM1 (Ser403; 39786; D8D6T), anti-SQSTM1 (88588; D5L7G), anti-dinitrophenol (DNP; 14681; D1D6), anti-ATG5 (12994; D5F5U), anti-cleaved CASP3 (9664; 5A1E), anti-p-EIF2A (S51; 3398; D9G8), and anti-EIF2A (5324; D7D3) antibodies were from Cell Signaling Technology. Anti-MUL1 (ab209263; EPR20241), anti-PPM1A (ab14824; p6c7), anti-PPM1B (ab70804), anti-G3BP1 (ab56574; 2F3), and anti-PABP (ab21060) were from Abcam. Anti-HA Tag (sc-7392; F7) and anti-His Tag (sc-8036; H-3) antibodies were from Santa Cruz Biotechnology. Anti-ubiquitinated proteins (04–263; FK2) antibody was from Millipore. After washing with 0.1% Tween-20 (Sigma-Aldrich, P1379) in Tris-buffered saline (TBS; Sigma-Aldrich, T8912), the membranes were incubated with an HRP-conjugated secondary rabbit antibody (1:5,000; Cell Signaling Technology, 7074 and 93,702) and secondary mouse antibody (1:5,000; Cell Signaling Technology, 7076 and 58,802) for 2 h at RT. Proteins were visualized using ECL reagents (GE Healthcare Life Sciences, RPN2235) and detected with ImageQuant™ LAS 4000 (FujiFilm) and Amersham™ ImageQuant™ 650 imager (Cytiva). Densitometric values were determined and quantified on western blot at non-saturating exposures using the ImageJ software (NIH, Bethesda, Maryland, USA, Java 1.8.0_112) and normalized against GAPDH, which acted as internal loading controls. Using ImageJ, we recorded the target intensity for each lane and plotted the intensity against the protein load. Using spiked proteins allowed us to control the amount of total protein per lane while only changing the amount of spiked proteins. These procedures were consistently followed across all western blot studies.

Immunofluorescence

For cell immunofluorescence, HNC cells (8 × 10⁴) were grown on coverslips, fixed in 4% paraformaldehyde for 15 min. After permeabilizing with 0.1% Triton X-100 in PBS (Gibco 10,010,023) for 10 min and blocked with 10% normal goat serum (Thermo Fisher Scientific, 50062Z) in PBS for 1 h, coverslips were incubated with primary antibodies (1:200) overnight at 4°C, including anti-TOMM20 (Abcam, ab56783; 4F3), anti-COX4l1 (Abcam, ab33985; 33985), anti-HSPD1/HSP60 (Cell Signaling Technology 12,165), anti-MYC Tag (Cell Signaling Technology, 2276; 9B11), anti-SQSTM1, anti-MAP1LC3B (Cell Signaling Technology, 3868; D11), anti-LAMP1 (Cell Signaling Technology, 9091; D2D11), anti-ubiquitinated proteins, anti-G3BP1, and anti-PABP. The coverslips were washed with PBST (0.1% Tween-20, PBS) for three times and were labeled with the secondary antibodies (1:300) goat anti-Mouse IgG (H+L) Alexa Flour Plus 488 (Thermo Fisher Scientific, A32723), goat anti-Rabbit IgG (H+L) Alexa Flour Plus 488 (Thermo Fisher Scientific, A32731), goat anti-Mouse IgG (H+L) Alexa Fluor Plus 555 (Thermo Fisher Scientific, A32727), goat anti-Rabbit IgG (H+L) Alexa Fluor Plus 555 (Thermo Fisher Scientific, A32732), and goat anti-Rabbit IgG (H+L) Alexa Fluor Plus 647 (Thermo Fisher Scientific, A32733) for 2 h at RT. The slides were mounted with ProLong^TM Gold Antifade Mountant with DAPI (Thermo Fisher Scientific, P36931) and imaged by Nikon A1R Spectral Confocal Laser Dual Scanning Microscope (Nikon) and Zeiss LSM900 Confocal Microscope with Airyscan 2 (Carl Zeiss). For MitoTracker staining, cells were incubated with 250 nM MitoTracker Red CMXRos (Thermo Fisher Scientific, M7512) in complete medium at 37°C for 20 min, then were removed of excess unbound dyes by washing twice. Structured illumination microscopy (SIM) super-resolution images were taken using a Nikon A1R HD25 N-SIM system with a 100× oil immersion objective lens, 1.49 NA (Nikon). Images were captured using NIS Elements Acquisition Software (Nikon) and reconstructed using slice reconstruction in NIS elements.

For tissue immunofluorescence, paraffin sections of patient and xenograft tumors with HNC were baked at 60°C for 30 min followed by deparaffinization and rehydration using xylene and graded ethanol. Then sections were subjected to heat-mediated antigen retrieval with citrate buffer (pH 6.0; Abcam, ab93678) for 20 min. Tissue samples were blocked with Image-iT^TM FX Signal Enhancer (Thermo Fisher Scientific, I36933) for 30 min and 10% normal goat serum for 1 h at RT. Following, slides were incubated with primary antibodies (1:200) overnight at 4°C, including anti-SQSTM1, anti-MAP1LC3B, anti-G3BP1, anti-PABP, anti-MKI67 (Cell Signaling Technology, 9449; 8D5), and anti-cleaved CASP3 (Cell Signaling Technology, 9664; 5A1E) antibodies. The slides were washed in TBST (0.1% Tween-20 in TBS) and incubated with the appropriate secondary antibodies (1:300) for 1 h at RT. After washing in TBST, Vector® TrueVIEW® Autofluorescence Quenching Kit (Vector laboratories, SP-8400-15) was used to quench spontaneous fluorescence for 20 min at RT followed by washing twice. The slides were mounted with ProLong^TM Gold Antifade Mountant with DAPI (Thermo Fisher Scientific, P36931) and imaged by Nikon A1R Spectral Confocal Laser Dual Scanning Microscope (Nikon).

Imaging quantification

For colocalization analysis, Pearson’s correlation coefficient was determined for each cell using NIS Elements Acquisition Software (Nikon). An individual cell was manually selected as an ROI using the polygon selection tool to generate each data point. Quantification of cells positive for SGs was based on a manual count of a minimum of 200 cells per experimental condition in ten random fields of view. The percentage of cells with SGs was calculated by counting the number of cells displaying SGs and expressing them as 100 × [(number of cells with SGs)/(total number of cells)] [17]. For the quantification of tumor area positive for MKI67 and cleaved CASP3, protein levels in tissue sections were done by measuring the fluorescence intensity using NIS Elements Acquisition Software (Nikon). The final value for each condition was obtained by averaging the relative KIM67- or cleaved CASP3-positive area per tumor across a minimum of eight random fields of view.

Transmission electron microscopy (TEM)

Samples were fixed for 12 h in 2% glutaraldehyde-paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) and washed in 100 mm phosphate buffer. The samples were postfixed with 1% OsO₄ dissolved in 100 mm phosphate buffer for 2 h and dehydrated in an ascending ethanol series (50 to 100%) and infiltrated with propylene oxide (Sigma-Aldrich 110,205). The specimens were embedded with the Poly/Bed 812 kit (Polysciences 21,844–1). The specimens were embedded with pure fresh resin and polymerized at 65°C in an electron microscope oven (Dosaka, TD-700) for 24 h. Sections 200- to 250-nm thick were cut and stained with toluidine blue (Sigma-Aldrich, T3260) for light microscopy. The 70-nm sections were double-stained with 6% uranyl acetate (EMS 22,400) for 20 min and then with lead citrate for contrast staining. The sections were cut with a LEICA EM UC7 ultramicrotome (Leica Microsystems, Buffalo Grove, IL, USA) with a diamond knife and transferred to copper and nickel grids (MERCK, F4776). All thin sections were observed with an electron microscope (JEOL, Tokyo, Japan, JEM-1011) at an acceleration voltage of 80 kV, and images were analyzed with the Camera-Megaview III Soft imaging system. A Formvar-carbon coated EM grid (EM Sciences, FCF100-Au) was placed Formvar side down on top of the (sample) drop for approximately 1 min to prepare a negative stain. The grid was removed, blotted with filter paper, and placed onto a drop of 2% uranyl acetate for 15 s. The uranyl acetate was removed, and the EM grid was examined and photographed for TEM. Autophagosomes were identified based on their characteristic double-membrane structure enclosing cytoplasmic material or organelles. Quantification was performed by counting the number of autophagosomes per cell in randomly selected fields. A total of 40 cells per experimental group were analyzed to ensure statistical reliability.

Tumorsphere formation

FaDu cells were transfected with 50 nM of the siRNAs targeting TBK1, or negative control. After 24 h transfection, the single-cell suspension of the transfected cells (3 × 10³ cells/30 μl/drop) in a complete culture medium was placed on the inner side of a 96-well hanging drop plate (SPL Life Sciences 331,096). After 48-h incubation, the spheroids were harvested from the hanging drop plate by pipetting 100 μl of PBS. The spheroids were then washed with PBS and stained using the LIVE/DEAD^TM Viability/Cytotoxicity Kit (Thermo Fisher Scientific, L3224) by incubating with calcein AM and ethidium homodimer-1 (EthD-1) at 37°C for 30 min. A confocal microscope (Nikon A1R, Japan, x20 objective, n = 20) was used for imaging, and z-section imaging was performed.

Xenograft in vivo tumor model

Human FaDu HNC cells (5 × 10⁶) were resuspended in 100 μl of PBS and inoculated subcutaneously into the lower right flank BALB/c nu/nu mice (aged 4–6 weeks, Harlan Laboratories, UK) [35]. After 7 days, when the tumors reached approximately ~50 mm in diameter, the mice were randomly divided into control and GSK8612 groups. The TBK1 inhibitor, GSK8612, was administered orally to the experimental group at a dose of 10 mg/kg once per day for 10 days. The tumors were measured using a sliding caliper daily, and the volumes (mm³) were calculated as described previously [35]. Mice were housed in an environmentally controlled room with a 12-h/12-h light/dark cycle and free access to laboratory chow and water. All animal-related experimental procedures and animal handling were conducted in accordance with the Committee for Ethics in Animal Experiments of the Ajou University School of Medicine (AUMC, IACUC No. 2017–0020, 2021–0063).

Statistical analysis

All statistical analyses were done with the Prism software package (Prism 9.0, GraphPad Software). Statistical significance was determined with the Mann-Whitney two-tailed unpaired t test for nonparametric values and the Student two-tailed unpaired t test for parametric values. All graphs depict mean ± standard deviations (SD) unless otherwise indicated. *, **, ***, and **** denote p value of < 0.05, 0.01, 0.001, and 0.0001, respectively, ns; non-significant.

Supplementary Material

Supplementary Materials R2.docx

KAUP_A_2481661_SM5196.docx^{(2.7MB, docx)}

Funding Statement

This research was supported by grants of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant numbers: RS-2021-KH113820 and RS-2024-00438448) and Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (grant numbers: RS-2023-NR077227 to C-HK and RS-2024-00351359 to HJK).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data generated in this study are available within the article and its Supplementary data files.

Author contributions

HJK, H-JK, and S-YK performed the experiments and data analyses. JR and JHY acquired patient samples and performed the experiments and data analyses. HJK and C-HK developed the study design and interpreted data, and supervised the study. HJK and C-HK wrote and revised the manuscript. All authors read and approved the final manuscript.

Ethics statement

The Institutional Review Board of Ajou University Medical Center (AUMC) approved the study protocol, and all patients provided written informed consent (AJIRB-BMR-SMP-18-150). (AUMC). All enrolled patients provided written informed consent before participation.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15548627.2025.2481661

References

[1].Luo J, Solimini NL, Elledge SJ.. Principles of cancer therapy: oncogene and non-oncogene addiction. Cell. 2009;136(5):823–837. [DOI] [PMC free article] [PubMed] [Google Scholar]
[2].Morán Luengo T, Mayer MP, Rüdiger SGD. The Hsp70–Hsp90 chaperone cascade in protein folding. Trends Cell Biol. 2019;29(2):164–177. doi: 10.1016/j.tcb.2018.10.004 [DOI] [PubMed] [Google Scholar]
[3].Pohl C, Dikic I. Cellular quality control by the ubiquitin-proteasome system and autophagy. Science. 2019;366(6467):818–822. doi: 10.1126/science.aax3769 [DOI] [PubMed] [Google Scholar]
[4].Thedieck K, Holzwarth B, Prentzell MT, et al. Inhibition of mTORC1 by astrin and stress granules prevents apoptosis in cancer cells. Cell. 2013;154(4):859–874. doi: 10.1016/j.cell.2013.07.031 [DOI] [PubMed] [Google Scholar]
[5].Moon SL, Morisaki T, Stasevich TJ, et al. Coupling of translation quality control and mRNA targeting to stress granules. J Cell Biol. 2020;219(8):e202004120. doi: 10.1083/jcb.202004120 [DOI] [PMC free article] [PubMed] [Google Scholar]
[6].Rybstein MD, Bravo-San Pedro JM, Kroemer G, et al. The autophagic network and cancer. Nat Cell Biol. 2018;20(3):243–251. doi: 10.1038/s41556-018-0042-2 [DOI] [PubMed] [Google Scholar]
[7].Cosway B, Lovat P. The role of autophagy in squamous cell carcinoma of the head and neck. Oral Oncol. 2016;54:1–6. doi: 10.1016/j.oraloncology.2015.12.007 [DOI] [PubMed] [Google Scholar]
[8].Smith AG, Macleod KF. Autophagy, cancer stem cells and drug resistance. J Pathol. 2019;247(5):708–718. [DOI] [PMC free article] [PubMed] [Google Scholar]
[9].Chen Y, Zhao H, Liang W, et al. Autophagy regulates the cancer stem cell phenotype of head and neck squamous cell carcinoma through the noncanonical FOXO3/SOX2 axis. Oncogene. 2022;41(5):634–646. doi: 10.1038/s41388-021-02115-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
[10].Chung CH, Gillison ML. Human papillomavirus in head and neck cancer: its role in pathogenesis and clinical implications. Clin Cancer Res. 2009;15(22):6758–6762. doi: 10.1158/1078-0432.CCR-09-0784 [DOI] [PubMed] [Google Scholar]
[11].Antonioli M, Pagni B, Vescovo T, et al. HPV sensitizes OPSCC cells to cisplatin-induced apoptosis by inhibiting autophagy through E7-mediated degradation of AMBRA1. Autophagy. 2021;17(10):2842–2855. doi: 10.1080/15548627.2020.1847444 [DOI] [PMC free article] [PubMed] [Google Scholar]
[12].Lai K, Matthews S, Wilmott JS, et al. Differences in LC3B expression and prognostic implications in oropharyngeal and oral cavity squamous cell carcinoma patients. BMC Cancer. 2018;18(1):624. doi: 10.1186/s12885-018-4536-x [DOI] [PMC free article] [PubMed] [Google Scholar]
[13].Kedersha N, Ivanov P, Anderson P. Stress granules and cell signaling: more than just a passing phase? Trends Biochem Sci. 2013;38(10):494–506. doi: 10.1016/j.tibs.2013.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
[14].Khong A, Matheny T, Jain S, et al. The stress granule transcriptome reveals principles of mRNA accumulation in stress granules. Mol Cell. 2017;68(4):808–20.e5. doi: 10.1016/j.molcel.2017.10.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
[15].Hu X, Xia K, Xiong H, et al. G3BP1 may serve as a potential biomarker of proliferation, apoptosis, and prognosis in oral squamous cell carcinoma. J Oral Pathol Med. 2021;50(10):995–1004. doi: 10.1111/jop.13199 [DOI] [PubMed] [Google Scholar]
[16].Yang T, Huang L, Qin H, et al. STRESS granule-associated rna-binding protein CAPRIN1 drives cancer progression and regulates treatment response in nasopharyngeal carcinoma. Med Oncol. 2022;40(1):47. doi: 10.1007/s12032-022-01910-w [DOI] [PMC free article] [PubMed] [Google Scholar]
[17].Somasekharan SP, El-Naggar A, Leprivier G, et al. YB-1 regulates stress granule formation and tumor progression by translationally activating G3BP1. J Cell Biol. 2015;208(7):913–929. doi: 10.1083/jcb.201411047 [DOI] [PMC free article] [PubMed] [Google Scholar]
[18].Zhan Y, Chen X, Zheng H, et al. YB1 associates with oncogenetic roles and poor prognosis in nasopharyngeal carcinoma. Sci Rep. 2022;12(1):3699. doi: 10.1038/s41598-022-07636-z [DOI] [PMC free article] [PubMed] [Google Scholar]
[19].Tian L, Xie X, Das U, et al. Forming cytoplasmic stress granules PURα suppresses mRNA translation initiation of IGFBP3 to promote esophageal squamous cell carcinoma progression. Oncogene. 2022;41(38):4336–4348. doi: 10.1038/s41388-022-02426-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
[20].Mediani L, Antoniani F, Galli V, et al. Hsp90-mediated regulation of DYRK3 couples stress granule disassembly and growth via mTORC1 signaling. EMBO Rep. 2021;22(5):e51740. doi: 10.15252/embr.202051740 [DOI] [PMC free article] [PubMed] [Google Scholar]
[21].Ganassi M, Mateju D, Bigi I, et al. A surveillance function of the HSPB8-BAG3-HSP70 chaperone complex ensures stress granule integrity and dynamism. Mol Cell. 2016;63(5):796–810. doi: 10.1016/j.molcel.2016.07.021 [DOI] [PubMed] [Google Scholar]
[22].Buchan JR, Kolaitis RM, Taylor JP, et al. Eukaryotic stress granules are cleared by autophagy and Cdc48/VCP function. Cell. 2013;153(7):1461–1474. doi: 10.1016/j.cell.2013.05.037 [DOI] [PMC free article] [PubMed] [Google Scholar]
[23].Chitiprolu M, Jagow C, Tremblay V, et al. A complex of C9ORF72 and p62 uses arginine methylation to eliminate stress granules by autophagy. Nat Commun. 2018;9(1):2794. doi: 10.1038/s41467-018-05273-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
[24].Kedersha N, Panas MD, Achorn CA, et al. G3BP–Caprin1–USP10 complexes mediate stress granule condensation and associate with 40S subunits. J Cell Biol. 2016;212(7):845–860. doi: 10.1083/jcb.201508028 [DOI] [PMC free article] [PubMed] [Google Scholar]
[25].Seguin SJ, Morelli FF, Vinet J, et al. Inhibition of autophagy, lysosome and VCP function impairs stress granule assembly. Cell Death Differ. 2014;21(12):1838–1851. doi: 10.1038/cdd.2014.103 [DOI] [PMC free article] [PubMed] [Google Scholar]
[26].Runde AP, Mack R, Peter Breslin SJ, et al. The role of TBK1 in cancer pathogenesis and anticancer immunity. J Exp Clin Cancer Res. 2022;41(1):135. doi: 10.1186/s13046-022-02352-y [DOI] [PMC free article] [PubMed] [Google Scholar]
[27].Zevini A, Olagnier D, Hiscott J. Crosstalk between cytoplasmic rig-I and STING sensing pathways. Trends Immunol. 2017;38(3):194–205. doi: 10.1016/j.it.2016.12.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
[28].Yum S, Li M, Fang Y, et al. TBK1 recruitment to STING activates both IRF3 and nf-κB that mediate immune defense against tumors and viral infections. Proc Natl Acad Sci USA. 2021;118(14):e2100225118. doi: 10.1073/pnas.2100225118 [DOI] [PMC free article] [PubMed] [Google Scholar]
[29].Hu L, Xie H, Liu X, et al. TBK1 is a synthetic lethal target in cancer with VHL loss. Cancer Discov. 2020;10(3):460–475. doi: 10.1158/2159-8290.CD-19-0837 [DOI] [PMC free article] [PubMed] [Google Scholar]
[30].Crew AP, Raina K, Dong H, et al. Identification and characterization of Von hippel-lindau-recruiting proteolysis targeting chimeras (PROTACs) of tank-binding kinase 1. J Med Chem. 2018;61(2):583–598. doi: 10.1021/acs.jmedchem.7b00635 [DOI] [PubMed] [Google Scholar]
[31].Eskiocak B, McMillan EA, Mendiratta S, et al. Biomarker accessible and chemically addressable mechanistic subtypes of BRAF melanoma. Cancer Discov. 2017;7(8):832–851. doi: 10.1158/2159-8290.CD-16-0955 [DOI] [PMC free article] [PubMed] [Google Scholar]
[32].Cooper JM, Ou YH, McMillan EA, et al. TBK1 provides context-selective support of the activated AKT/mTOR pathway in lung cancer. Cancer Res. 2017;77(18):5077–5094. doi: 10.1158/0008-5472.CAN-17-0829 [DOI] [PMC free article] [PubMed] [Google Scholar]
[33].Liu S, Marneth AE, Alexe G, et al. The kinases IKBKE and TBK1 regulate myc-dependent survival pathways through YB-1 in AML and are targets for therapy. Blood Adv. 2018;2(23):3428–3442. doi: 10.1182/bloodadvances.2018016733 [DOI] [PMC free article] [PubMed] [Google Scholar]
[34].Bae S, Kim SY, Jung JH, et al. Akt is negatively regulated by the MULAN E3 ligase. Cell Res. 2012;22(5):873–885. doi: 10.1038/cr.2012.38 [DOI] [PMC free article] [PubMed] [Google Scholar]
[35].Kim SY, Kim HJ, Kim HJ, et al. HSPA5 negatively regulates lysosomal activity through ubiquitination of MUL1 in head and neck cancer. Autophagy. 2018;14(3):385–403. doi: 10.1080/15548627.2017.1414126 [DOI] [PMC free article] [PubMed] [Google Scholar]
[36].Zhang B, Huang J, Li HL, et al. GIDE is a mitochondrial E3 ubiquitin ligase that induces apoptosis and slows growth. Cell Res. 2008;18(9):900–910. doi: 10.1038/cr.2008.75 [DOI] [PMC free article] [PubMed] [Google Scholar]
[37].Kim HJ, Kim SY, Kim DH, et al. Crosstalk between HSPA5 arginylation and sequential ubiquitination leads to AKT degradation through autophagy flux. Autophagy. 2021;17(4):961–979. doi: 10.1080/15548627.2020.1740529 [DOI] [PMC free article] [PubMed] [Google Scholar]
[38].Wang H, Wang X, Zhang H, et al. The HSF1/miR-135b-5p axis induces protective autophagy to promote oxaliplatin resistance through the MUL1/ULK1 pathway in colorectal cancer. Oncogene. 2021;40(28):4695–4708. doi: 10.1038/s41388-021-01898-z [DOI] [PubMed] [Google Scholar]
[39].Bekes M, Langley DR, Crews CM. PROTAC targeted protein degraders: the past is prologue. Nat Rev Drug Discov. 2022;21(3):181–200. doi: 10.1038/s41573-021-00371-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
[40].Nunziante M, Ackermann K, Dietrich K, et al. Proteasomal dysfunction and endoplasmic reticulum stress enhance trafficking of prion protein aggregates through the secretory pathway and increase accumulation of pathologic prion protein. J Biol Chem. 2011;286(39):33942–33953. doi: 10.1074/jbc.M111.272617 [DOI] [PMC free article] [PubMed] [Google Scholar]
[41].Cha-Molstad H, Sung KS, Hwang J, et al. Amino-terminal arginylation targets endoplasmic reticulum chaperone BiP for autophagy through p62 binding. Nat Cell Biol. 2015;17(7):917–929. doi: 10.1038/ncb3177 [DOI] [PMC free article] [PubMed] [Google Scholar]
[42].Peng H, Yang J, Li G, et al. Ubiquitylation of p62/sequestosome1 activates its autophagy receptor function and controls selective autophagy upon ubiquitin stress. Cell Res. 2017;27(5):657–674. doi: 10.1038/cr.2017.40 [DOI] [PMC free article] [PubMed] [Google Scholar]
[43].Li Z, Liu G, Sun L, et al. PPM1A regulates antiviral signaling by antagonizing TBK1-mediated STING phosphorylation and aggregation. PlOS Pathog. 2015;11(3):e1004783. doi: 10.1371/journal.ppat.1004783 [DOI] [PMC free article] [PubMed] [Google Scholar]
[44].Zhao Y, Liang L, Fan Y, et al. PPM1B negatively regulates antiviral response via dephosphorylating TBK1. Cell Signal. 2012;24(11):2197–2204. doi: 10.1016/j.cellsig.2012.06.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
[45].Ni G, Konno H, Barber GN. Ubiquitination of STING at lysine 224 controls IRF3 activation. Sci Immunol. 2017;2(11). doi: 10.1126/sciimmunol.aah7119 [DOI] [PMC free article] [PubMed] [Google Scholar]
[46].Sun H, Mali SM, Singh SK, et al. Diverse fate of ubiquitin chain moieties: the proximal is degraded with the target, and the distal protects the proximal from removal and recycles. Proc Natl Acad Sci USA. 2019;116(16):7805–7812. doi: 10.1073/pnas.1822148116 [DOI] [PMC free article] [PubMed] [Google Scholar]
[47].Shu C, Sankaran B, Chaton CT, et al. Structural insights into the functions of TBK1 in innate antimicrobial immunity. Structure. 2013;21(7):1137–1148. doi: 10.1016/j.str.2013.04.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
[48].Ma X, Helgason E, Phung QT, et al. Molecular basis of tank-binding kinase 1 activation by transautophosphorylation. Proc Natl Acad Sci USA. 2012;109(24):9378–9383. doi: 10.1073/pnas.1121552109 [DOI] [PMC free article] [PubMed] [Google Scholar]
[49].Seo M, Lee SO, Kim JH, et al. MAP4-regulated dynein-dependent trafficking of BTN3A1 controls the TBK1–IRF3 signaling axis. Proc Natl Acad Sci USA. 2016;113(50):14390–14395. doi: 10.1073/pnas.1615287113 [DOI] [PMC free article] [PubMed] [Google Scholar]
[50].Pilli M, Arko-Mensah J, Ponpuak M, et al. TBK-1 promotes autophagy-mediated antimicrobial defense by controlling autophagosome maturation. Immunity. 2012;37(2):223–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
[51].Herhaus L, Bhaskara RM, Lystad AH, et al. TBK1-mediated phosphorylation of LC3C and GABARAP-L2 controls autophagosome shedding by ATG4 protease. EMBO Rep. 2020;21(1):e48317. doi: 10.15252/embr.201948317 [DOI] [PMC free article] [PubMed] [Google Scholar]
[52].Richter B, Sliter DA, Herhaus L, et al. Phosphorylation of OPTN by TBK1 enhances its binding to ub chains and promotes selective autophagy of damaged mitochondria. Proc Natl Acad Sci USA. 2016;113(15):4039–4044. doi: 10.1073/pnas.1523926113 [DOI] [PMC free article] [PubMed] [Google Scholar]
[53].Gao X, Chen D, Hu X, et al. PLA1A participates in the antiviral innate immune response by facilitating the recruitment of tank-binding kinase 1 to mitochondria. J Innate Immun. 2018;10(4):315–327. doi: 10.1159/000489832 [DOI] [PMC free article] [PubMed] [Google Scholar]
[54].Jenkins K, Khoo JJ, Sadler A, et al. Mitochondrially localised MUL1 is a novel modulator of antiviral signaling. Immunol Cell Biol. 2013;91(4):321–330. [DOI] [PubMed] [Google Scholar]
[55].Kim W, Bennett EJ, Huttlin EL, et al. Systematic and quantitative assessment of the ubiquitin-modified proteome. Mol Cell. 2011;44(2):325–340. doi: 10.1016/j.molcel.2011.08.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
[56].Beltrao P, Albanèse V, Kenner LR, et al. Systematic functional prioritization of protein posttranslational modifications. Cell. 2012;150(2):413–425. doi: 10.1016/j.cell.2012.05.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
[57].Akimov V, Barrio-Hernandez I, Hansen SVF, et al. UbiSite approach for comprehensive mapping of lysine and N-terminal ubiquitination sites. Nat Struct Mol Biol. 2018;25(7):631–640. doi: 10.1038/s41594-018-0084-y [DOI] [PubMed] [Google Scholar]
[58].Bijlmakers MJ, Marsh M, Craig E. Hsp90 is essential for the synthesis and subsequent membrane association, but not the maintenance, of the Src-kinase p56 lck. Mol Biol Cell. 2000;11(5):1585–1595. doi: 10.1091/mbc.11.5.1585 [DOI] [PMC free article] [PubMed] [Google Scholar]
[59].Basso AD, Solit DB, Chiosis G, et al. Akt forms an intracellular complex with heat shock protein 90 (Hsp90) and Cdc37 and is destabilized by inhibitors of Hsp90 function. J Biol Chem. 2002;277(42):39858–39866. doi: 10.1074/jbc.M206322200 [DOI] [PubMed] [Google Scholar]
[60].Taipale M, Krykbaeva I, Koeva M, et al. Quantitative analysis of HSP90-client interactions reveals principles of substrate recognition. Cell. 2012;150(5):987–1001. doi: 10.1016/j.cell.2012.06.047 [DOI] [PMC free article] [PubMed] [Google Scholar]
[61].Saul VV, Seibert M, Krüger M, et al. ULK1/2 restricts the formation of inducible sint-speckles, membraneless organelles controlling the threshold of TBK1 activation. iScience. 2019;19:527–544. doi: 10.1016/j.isci.2019.08.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
[62].Yang K, Shi H, Qi R, et al. Hsp90 regulates activation of interferon regulatory factor 3 and TBK-1 stabilization in Sendai virus-infected cells. Mol Biol Cell. 2006;17(3):1461–1471. doi: 10.1091/mbc.e05-09-0853 [DOI] [PMC free article] [PubMed] [Google Scholar]
[63].Thomson DW, Poeckel D, Zinn N, et al. Discovery of GSK8612, a highly selective and potent TBK1 inhibitor. ACS Med Chem Lett. 2019;10(5):780–785. doi: 10.1021/acsmedchemlett.9b00027 [DOI] [PMC free article] [PubMed] [Google Scholar]
[64].Li F, Xu D, Wang Y, et al. Structural insights into the ubiquitin recognition by OPTN (optineurin) and its regulation by TBK1-mediated phosphorylation. Autophagy. 2018;14(1):66–79. doi: 10.1080/15548627.2017.1391970 [DOI] [PMC free article] [PubMed] [Google Scholar]
[65].Thurston TL, Ryzhakov G, Bloor S, et al. The TBK1 adaptor and autophagy receptor NDP52 restricts the proliferation of ubiquitin-coated bacteria. Nat Immunol. 2009;10(11):1215–1221. doi: 10.1038/ni.1800 [DOI] [PubMed] [Google Scholar]
[66].Matsumoto G, Shimogori T, Hattori N, et al. TBK1 controls autophagosomal engulfment of polyubiquitinated mitochondria through p62/SQSTM1 phosphorylation. Hum Mol Genet. 2015;24(15):4429–4442. doi: 10.1093/hmg/ddv179 [DOI] [PubMed] [Google Scholar]
[67].Saitoh T, Fujita N, Hayashi T, et al. Atg9a controls dsDNA-driven dynamic translocation of STING and the innate immune response. Proc Natl Acad Sci USA. 2009;106(49):20842–20846. doi: 10.1073/pnas.0911267106 [DOI] [PMC free article] [PubMed] [Google Scholar]
[68].Ravenhill BJ, Boyle KB, von Muhlinen N, et al. The cargo receptor NDP52 initiates selective autophagy by recruiting the ULK complex to cytosol-invading bacteria. Mol Cell. 2019;74(2):320–9 e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
[69].Vargas JNS, Wang C, Bunker E, et al. Spatiotemporal control of ULK1 activation by NDP52 and TBK1 during selective autophagy. Mol Cell. 2019;74(2):347–62 e6. doi: 10.1016/j.molcel.2019.02.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
[70].Kumar S, Gu Y, Abudu YP, et al. Phosphorylation of syntaxin 17 by TBK1 controls autophagy initiation. Dev Cell. 2019;49(1):130–44.e6. doi: 10.1016/j.devcel.2019.01.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
[71].Mizushima N, Yoshimori T. How to interpret LC3 immunoblotting. Autophagy. 2007;3(6):542–545. doi: 10.4161/auto.4600 [DOI] [PubMed] [Google Scholar]
[72].Kimura S, Noda T, Yoshimori T. Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent-tagged LC3. Autophagy. 2007;3(5):452–460. doi: 10.4161/auto.4451 [DOI] [PubMed] [Google Scholar]
[73].Sha Z, Schnell HM, Ruoff K, et al. Rapid induction of p62 and GABARAPL1 upon proteasome inhibition promotes survival before autophagy activation. J Cell Biol. 2018;217(5):1757–1776. doi: 10.1083/jcb.201708168 [DOI] [PMC free article] [PubMed] [Google Scholar]
[74].Matsumoto G, Wada K, Okuno M, et al. Serine 403 phosphorylation of p62/SQSTM1 regulates selective autophagic clearance of ubiquitinated proteins. Mol Cell. 2011;44(2):279–289. [DOI] [PubMed] [Google Scholar]
[75].Lim J, Lachenmayer ML, Wu S, et al. Proteotoxic stress induces phosphorylation of p62/SQSTM1 by ULK1 to regulate selective autophagic clearance of protein aggregates. PlOS Genet. 2015;11(2):e1004987. doi: 10.1371/journal.pgen.1004987 [DOI] [PMC free article] [PubMed] [Google Scholar]
[76].Aulas A, Finetti P, Lyons SM, et al. Revisiting the concept of stress in the prognosis of solid tumors: a role for stress granules proteins? Cancers (Basel). 2020;12(9):2470. doi: 10.3390/cancers12092470 [DOI] [PMC free article] [PubMed] [Google Scholar]
[77].Mazroui R, Di Marco S, Kaufman RJ, et al. Inhibition of the ubiquitin-proteasome system induces stress granule formation. Mol Biol Cell. 2007;18(7):2603–2618. doi: 10.1091/mbc.e06-12-1079 [DOI] [PMC free article] [PubMed] [Google Scholar]
[78].Markmiller S, Soltanieh S, Server KL, et al. Context-dependent and disease-specific diversity in protein interactions within stress granules. Cell. 2018;172(3):590–604.e13. doi: 10.1016/j.cell.2017.12.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
[79].Jenkins RW, Aref AR, Lizotte PH, et al. Ex vivo profiling of PD-1 blockade using organotypic tumor spheroids. Cancer Discov. 2018;8(2):196–215. doi: 10.1158/2159-8290.CD-17-0833 [DOI] [PMC free article] [PubMed] [Google Scholar]
[80].Shen DW, Pouliot LM, Hall MD, et al. Cisplatin resistance: a cellular self-defense mechanism resulting from multiple epigenetic and genetic changes. Pharmacol Rev. 2012;64(3):706–721. doi: 10.1124/pr.111.005637 [DOI] [PMC free article] [PubMed] [Google Scholar]
[81].Kim HR, Lee SJ, Park S, et al. A single-arm, prospective, phase II study of cisplatin plus weekly docetaxel as first-line therapy in patients with metastatic or recurrent salivary gland cancer. Cancer Res Treat. 2022;54(3):719–727. doi: 10.4143/crt.2021.1019 [DOI] [PMC free article] [PubMed] [Google Scholar]
[82].Freibaum BD, Messing J, Nakamura H, et al. Identification of small molecule inhibitors of G3BP-driven stress granule formation. J Cell Biol. 2024;223(3):e202308083. doi: 10.1083/jcb.202308083 [DOI] [PMC free article] [PubMed] [Google Scholar]
[83].Barbie DA, Tamayo P, Boehm JS, et al. Systematic RNA interference reveals that oncogenic kras-driven cancers require TBK1. Nature. 2009;462(7269):108–112. doi: 10.1038/nature08460 [DOI] [PMC free article] [PubMed] [Google Scholar]
[84].Jiang Y, Chen S, Li Q, et al. Tank-binding kinase 1 (TBK1) serves as a potential target for hepatocellular carcinoma by enhancing tumor immune infiltration. Front Immunol. 2021;12:612139. doi: 10.3389/fimmu.2021.612139 [DOI] [PMC free article] [PubMed] [Google Scholar]
[85].Hou F, Sun L, Zheng H, et al. MAVS forms functional prion-like aggregates to activate and propagate antiviral innate immune response. Cell. 2011;146(3):448–461. doi: 10.1016/j.cell.2011.06.041 [DOI] [PMC free article] [PubMed] [Google Scholar]
[86].Clark K, Peggie M, Plater L, et al. Novel cross-talk within the IKK family controls innate immunity. Biochem J. 2011;434(1):93–104. doi: 10.1042/BJ20101701 [DOI] [PubMed] [Google Scholar]
[87].Wang B, Maxwell BA, Joo JH, et al. ULK1 and ULK2 regulate stress granule disassembly through phosphorylation and activation of VCP/p97. Mol Cell. 2019;74(4):742–57.e8. doi: 10.1016/j.molcel.2019.03.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
[88].Cherkasov V, Hofmann S, Druffel-Augustin S, et al. Coordination of translational control and protein homeostasis during severe heat stress. Curr Biol. 2013;23(24):2452–2462. doi: 10.1016/j.cub.2013.09.058 [DOI] [PubMed] [Google Scholar]
[89].Li H, Lin PH, Gupta P, et al. MG53 suppresses tumor progression and stress granule formation by modulating G3BP2 activity in non-small cell lung cancer. Mol Cancer. 2021;20(1):118. doi: 10.1186/s12943-021-01418-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
[90].Cruz VH, Brekken RA. Assessment of tank-binding kinase 1 as a therapeutic target in cancer. J Cell Commun Signal. 2018;12(1):83–90. doi: 10.1007/s12079-017-0438-y [DOI] [PMC free article] [PubMed] [Google Scholar]
[91].Zhu L, Li Y, Xie X, et al. TBKBP1 and TBK1 form a growth factor signalling axis mediating immunosuppression and tumourigenesis. Nat Cell Biol. 2019;21(12):1604–1614. doi: 10.1038/s41556-019-0429-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
[92].Cai H, Yan L, Liu N, et al. IFI16 promotes cervical cancer progression by upregulating PD-L1 in immunomicroenvironment through STING-TBK1-NF-kB pathway. Biomed Pharmacother. 2020;123:109790. doi: 10.1016/j.biopha.2019.109790 [DOI] [PubMed] [Google Scholar]
[93].Sun Y, Revach OY, Anderson S, et al. Targeting TBK1 to overcome resistance to cancer immunotherapy. Nature. 2023;615(7950):158–167. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Materials R2.docx

KAUP_A_2481661_SM5196.docx^{(2.7MB, docx)}

Data Availability Statement

The data generated in this study are available within the article and its Supplementary data files.

[cit0001] [1].Luo J, Solimini NL, Elledge SJ.. Principles of cancer therapy: oncogene and non-oncogene addiction. Cell. 2009;136(5):823–837. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0002] [2].Morán Luengo T, Mayer MP, Rüdiger SGD. The Hsp70–Hsp90 chaperone cascade in protein folding. Trends Cell Biol. 2019;29(2):164–177. doi: 10.1016/j.tcb.2018.10.004 [DOI] [PubMed] [Google Scholar]

[cit0003] [3].Pohl C, Dikic I. Cellular quality control by the ubiquitin-proteasome system and autophagy. Science. 2019;366(6467):818–822. doi: 10.1126/science.aax3769 [DOI] [PubMed] [Google Scholar]

[cit0004] [4].Thedieck K, Holzwarth B, Prentzell MT, et al. Inhibition of mTORC1 by astrin and stress granules prevents apoptosis in cancer cells. Cell. 2013;154(4):859–874. doi: 10.1016/j.cell.2013.07.031 [DOI] [PubMed] [Google Scholar]

[cit0005] [5].Moon SL, Morisaki T, Stasevich TJ, et al. Coupling of translation quality control and mRNA targeting to stress granules. J Cell Biol. 2020;219(8):e202004120. doi: 10.1083/jcb.202004120 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0006] [6].Rybstein MD, Bravo-San Pedro JM, Kroemer G, et al. The autophagic network and cancer. Nat Cell Biol. 2018;20(3):243–251. doi: 10.1038/s41556-018-0042-2 [DOI] [PubMed] [Google Scholar]

[cit0007] [7].Cosway B, Lovat P. The role of autophagy in squamous cell carcinoma of the head and neck. Oral Oncol. 2016;54:1–6. doi: 10.1016/j.oraloncology.2015.12.007 [DOI] [PubMed] [Google Scholar]

[cit0008] [8].Smith AG, Macleod KF. Autophagy, cancer stem cells and drug resistance. J Pathol. 2019;247(5):708–718. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0009] [9].Chen Y, Zhao H, Liang W, et al. Autophagy regulates the cancer stem cell phenotype of head and neck squamous cell carcinoma through the noncanonical FOXO3/SOX2 axis. Oncogene. 2022;41(5):634–646. doi: 10.1038/s41388-021-02115-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0010] [10].Chung CH, Gillison ML. Human papillomavirus in head and neck cancer: its role in pathogenesis and clinical implications. Clin Cancer Res. 2009;15(22):6758–6762. doi: 10.1158/1078-0432.CCR-09-0784 [DOI] [PubMed] [Google Scholar]

[cit0011] [11].Antonioli M, Pagni B, Vescovo T, et al. HPV sensitizes OPSCC cells to cisplatin-induced apoptosis by inhibiting autophagy through E7-mediated degradation of AMBRA1. Autophagy. 2021;17(10):2842–2855. doi: 10.1080/15548627.2020.1847444 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0012] [12].Lai K, Matthews S, Wilmott JS, et al. Differences in LC3B expression and prognostic implications in oropharyngeal and oral cavity squamous cell carcinoma patients. BMC Cancer. 2018;18(1):624. doi: 10.1186/s12885-018-4536-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0013] [13].Kedersha N, Ivanov P, Anderson P. Stress granules and cell signaling: more than just a passing phase? Trends Biochem Sci. 2013;38(10):494–506. doi: 10.1016/j.tibs.2013.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0014] [14].Khong A, Matheny T, Jain S, et al. The stress granule transcriptome reveals principles of mRNA accumulation in stress granules. Mol Cell. 2017;68(4):808–20.e5. doi: 10.1016/j.molcel.2017.10.015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0015] [15].Hu X, Xia K, Xiong H, et al. G3BP1 may serve as a potential biomarker of proliferation, apoptosis, and prognosis in oral squamous cell carcinoma. J Oral Pathol Med. 2021;50(10):995–1004. doi: 10.1111/jop.13199 [DOI] [PubMed] [Google Scholar]

[cit0016] [16].Yang T, Huang L, Qin H, et al. STRESS granule-associated rna-binding protein CAPRIN1 drives cancer progression and regulates treatment response in nasopharyngeal carcinoma. Med Oncol. 2022;40(1):47. doi: 10.1007/s12032-022-01910-w [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0017] [17].Somasekharan SP, El-Naggar A, Leprivier G, et al. YB-1 regulates stress granule formation and tumor progression by translationally activating G3BP1. J Cell Biol. 2015;208(7):913–929. doi: 10.1083/jcb.201411047 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0018] [18].Zhan Y, Chen X, Zheng H, et al. YB1 associates with oncogenetic roles and poor prognosis in nasopharyngeal carcinoma. Sci Rep. 2022;12(1):3699. doi: 10.1038/s41598-022-07636-z [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0019] [19].Tian L, Xie X, Das U, et al. Forming cytoplasmic stress granules PURα suppresses mRNA translation initiation of IGFBP3 to promote esophageal squamous cell carcinoma progression. Oncogene. 2022;41(38):4336–4348. doi: 10.1038/s41388-022-02426-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0020] [20].Mediani L, Antoniani F, Galli V, et al. Hsp90-mediated regulation of DYRK3 couples stress granule disassembly and growth via mTORC1 signaling. EMBO Rep. 2021;22(5):e51740. doi: 10.15252/embr.202051740 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0021] [21].Ganassi M, Mateju D, Bigi I, et al. A surveillance function of the HSPB8-BAG3-HSP70 chaperone complex ensures stress granule integrity and dynamism. Mol Cell. 2016;63(5):796–810. doi: 10.1016/j.molcel.2016.07.021 [DOI] [PubMed] [Google Scholar]

[cit0022] [22].Buchan JR, Kolaitis RM, Taylor JP, et al. Eukaryotic stress granules are cleared by autophagy and Cdc48/VCP function. Cell. 2013;153(7):1461–1474. doi: 10.1016/j.cell.2013.05.037 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0023] [23].Chitiprolu M, Jagow C, Tremblay V, et al. A complex of C9ORF72 and p62 uses arginine methylation to eliminate stress granules by autophagy. Nat Commun. 2018;9(1):2794. doi: 10.1038/s41467-018-05273-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0024] [24].Kedersha N, Panas MD, Achorn CA, et al. G3BP–Caprin1–USP10 complexes mediate stress granule condensation and associate with 40S subunits. J Cell Biol. 2016;212(7):845–860. doi: 10.1083/jcb.201508028 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0025] [25].Seguin SJ, Morelli FF, Vinet J, et al. Inhibition of autophagy, lysosome and VCP function impairs stress granule assembly. Cell Death Differ. 2014;21(12):1838–1851. doi: 10.1038/cdd.2014.103 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0026] [26].Runde AP, Mack R, Peter Breslin SJ, et al. The role of TBK1 in cancer pathogenesis and anticancer immunity. J Exp Clin Cancer Res. 2022;41(1):135. doi: 10.1186/s13046-022-02352-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0027] [27].Zevini A, Olagnier D, Hiscott J. Crosstalk between cytoplasmic rig-I and STING sensing pathways. Trends Immunol. 2017;38(3):194–205. doi: 10.1016/j.it.2016.12.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0028] [28].Yum S, Li M, Fang Y, et al. TBK1 recruitment to STING activates both IRF3 and nf-κB that mediate immune defense against tumors and viral infections. Proc Natl Acad Sci USA. 2021;118(14):e2100225118. doi: 10.1073/pnas.2100225118 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0029] [29].Hu L, Xie H, Liu X, et al. TBK1 is a synthetic lethal target in cancer with VHL loss. Cancer Discov. 2020;10(3):460–475. doi: 10.1158/2159-8290.CD-19-0837 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0030] [30].Crew AP, Raina K, Dong H, et al. Identification and characterization of Von hippel-lindau-recruiting proteolysis targeting chimeras (PROTACs) of tank-binding kinase 1. J Med Chem. 2018;61(2):583–598. doi: 10.1021/acs.jmedchem.7b00635 [DOI] [PubMed] [Google Scholar]

[cit0031] [31].Eskiocak B, McMillan EA, Mendiratta S, et al. Biomarker accessible and chemically addressable mechanistic subtypes of BRAF melanoma. Cancer Discov. 2017;7(8):832–851. doi: 10.1158/2159-8290.CD-16-0955 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0032] [32].Cooper JM, Ou YH, McMillan EA, et al. TBK1 provides context-selective support of the activated AKT/mTOR pathway in lung cancer. Cancer Res. 2017;77(18):5077–5094. doi: 10.1158/0008-5472.CAN-17-0829 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0033] [33].Liu S, Marneth AE, Alexe G, et al. The kinases IKBKE and TBK1 regulate myc-dependent survival pathways through YB-1 in AML and are targets for therapy. Blood Adv. 2018;2(23):3428–3442. doi: 10.1182/bloodadvances.2018016733 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0034] [34].Bae S, Kim SY, Jung JH, et al. Akt is negatively regulated by the MULAN E3 ligase. Cell Res. 2012;22(5):873–885. doi: 10.1038/cr.2012.38 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0035] [35].Kim SY, Kim HJ, Kim HJ, et al. HSPA5 negatively regulates lysosomal activity through ubiquitination of MUL1 in head and neck cancer. Autophagy. 2018;14(3):385–403. doi: 10.1080/15548627.2017.1414126 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0036] [36].Zhang B, Huang J, Li HL, et al. GIDE is a mitochondrial E3 ubiquitin ligase that induces apoptosis and slows growth. Cell Res. 2008;18(9):900–910. doi: 10.1038/cr.2008.75 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0037] [37].Kim HJ, Kim SY, Kim DH, et al. Crosstalk between HSPA5 arginylation and sequential ubiquitination leads to AKT degradation through autophagy flux. Autophagy. 2021;17(4):961–979. doi: 10.1080/15548627.2020.1740529 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0038] [38].Wang H, Wang X, Zhang H, et al. The HSF1/miR-135b-5p axis induces protective autophagy to promote oxaliplatin resistance through the MUL1/ULK1 pathway in colorectal cancer. Oncogene. 2021;40(28):4695–4708. doi: 10.1038/s41388-021-01898-z [DOI] [PubMed] [Google Scholar]

[cit0039] [39].Bekes M, Langley DR, Crews CM. PROTAC targeted protein degraders: the past is prologue. Nat Rev Drug Discov. 2022;21(3):181–200. doi: 10.1038/s41573-021-00371-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0040] [40].Nunziante M, Ackermann K, Dietrich K, et al. Proteasomal dysfunction and endoplasmic reticulum stress enhance trafficking of prion protein aggregates through the secretory pathway and increase accumulation of pathologic prion protein. J Biol Chem. 2011;286(39):33942–33953. doi: 10.1074/jbc.M111.272617 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0041] [41].Cha-Molstad H, Sung KS, Hwang J, et al. Amino-terminal arginylation targets endoplasmic reticulum chaperone BiP for autophagy through p62 binding. Nat Cell Biol. 2015;17(7):917–929. doi: 10.1038/ncb3177 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0042] [42].Peng H, Yang J, Li G, et al. Ubiquitylation of p62/sequestosome1 activates its autophagy receptor function and controls selective autophagy upon ubiquitin stress. Cell Res. 2017;27(5):657–674. doi: 10.1038/cr.2017.40 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0043] [43].Li Z, Liu G, Sun L, et al. PPM1A regulates antiviral signaling by antagonizing TBK1-mediated STING phosphorylation and aggregation. PlOS Pathog. 2015;11(3):e1004783. doi: 10.1371/journal.ppat.1004783 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0044] [44].Zhao Y, Liang L, Fan Y, et al. PPM1B negatively regulates antiviral response via dephosphorylating TBK1. Cell Signal. 2012;24(11):2197–2204. doi: 10.1016/j.cellsig.2012.06.017 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0045] [45].Ni G, Konno H, Barber GN. Ubiquitination of STING at lysine 224 controls IRF3 activation. Sci Immunol. 2017;2(11). doi: 10.1126/sciimmunol.aah7119 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0046] [46].Sun H, Mali SM, Singh SK, et al. Diverse fate of ubiquitin chain moieties: the proximal is degraded with the target, and the distal protects the proximal from removal and recycles. Proc Natl Acad Sci USA. 2019;116(16):7805–7812. doi: 10.1073/pnas.1822148116 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0047] [47].Shu C, Sankaran B, Chaton CT, et al. Structural insights into the functions of TBK1 in innate antimicrobial immunity. Structure. 2013;21(7):1137–1148. doi: 10.1016/j.str.2013.04.025 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0048] [48].Ma X, Helgason E, Phung QT, et al. Molecular basis of tank-binding kinase 1 activation by transautophosphorylation. Proc Natl Acad Sci USA. 2012;109(24):9378–9383. doi: 10.1073/pnas.1121552109 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0049] [49].Seo M, Lee SO, Kim JH, et al. MAP4-regulated dynein-dependent trafficking of BTN3A1 controls the TBK1–IRF3 signaling axis. Proc Natl Acad Sci USA. 2016;113(50):14390–14395. doi: 10.1073/pnas.1615287113 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0050] [50].Pilli M, Arko-Mensah J, Ponpuak M, et al. TBK-1 promotes autophagy-mediated antimicrobial defense by controlling autophagosome maturation. Immunity. 2012;37(2):223–234. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0051] [51].Herhaus L, Bhaskara RM, Lystad AH, et al. TBK1-mediated phosphorylation of LC3C and GABARAP-L2 controls autophagosome shedding by ATG4 protease. EMBO Rep. 2020;21(1):e48317. doi: 10.15252/embr.201948317 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0052] [52].Richter B, Sliter DA, Herhaus L, et al. Phosphorylation of OPTN by TBK1 enhances its binding to ub chains and promotes selective autophagy of damaged mitochondria. Proc Natl Acad Sci USA. 2016;113(15):4039–4044. doi: 10.1073/pnas.1523926113 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0053] [53].Gao X, Chen D, Hu X, et al. PLA1A participates in the antiviral innate immune response by facilitating the recruitment of tank-binding kinase 1 to mitochondria. J Innate Immun. 2018;10(4):315–327. doi: 10.1159/000489832 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0054] [54].Jenkins K, Khoo JJ, Sadler A, et al. Mitochondrially localised MUL1 is a novel modulator of antiviral signaling. Immunol Cell Biol. 2013;91(4):321–330. [DOI] [PubMed] [Google Scholar]

[cit0055] [55].Kim W, Bennett EJ, Huttlin EL, et al. Systematic and quantitative assessment of the ubiquitin-modified proteome. Mol Cell. 2011;44(2):325–340. doi: 10.1016/j.molcel.2011.08.025 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0056] [56].Beltrao P, Albanèse V, Kenner LR, et al. Systematic functional prioritization of protein posttranslational modifications. Cell. 2012;150(2):413–425. doi: 10.1016/j.cell.2012.05.036 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0057] [57].Akimov V, Barrio-Hernandez I, Hansen SVF, et al. UbiSite approach for comprehensive mapping of lysine and N-terminal ubiquitination sites. Nat Struct Mol Biol. 2018;25(7):631–640. doi: 10.1038/s41594-018-0084-y [DOI] [PubMed] [Google Scholar]

[cit0058] [58].Bijlmakers MJ, Marsh M, Craig E. Hsp90 is essential for the synthesis and subsequent membrane association, but not the maintenance, of the Src-kinase p56 lck. Mol Biol Cell. 2000;11(5):1585–1595. doi: 10.1091/mbc.11.5.1585 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0059] [59].Basso AD, Solit DB, Chiosis G, et al. Akt forms an intracellular complex with heat shock protein 90 (Hsp90) and Cdc37 and is destabilized by inhibitors of Hsp90 function. J Biol Chem. 2002;277(42):39858–39866. doi: 10.1074/jbc.M206322200 [DOI] [PubMed] [Google Scholar]

[cit0060] [60].Taipale M, Krykbaeva I, Koeva M, et al. Quantitative analysis of HSP90-client interactions reveals principles of substrate recognition. Cell. 2012;150(5):987–1001. doi: 10.1016/j.cell.2012.06.047 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0061] [61].Saul VV, Seibert M, Krüger M, et al. ULK1/2 restricts the formation of inducible sint-speckles, membraneless organelles controlling the threshold of TBK1 activation. iScience. 2019;19:527–544. doi: 10.1016/j.isci.2019.08.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0062] [62].Yang K, Shi H, Qi R, et al. Hsp90 regulates activation of interferon regulatory factor 3 and TBK-1 stabilization in Sendai virus-infected cells. Mol Biol Cell. 2006;17(3):1461–1471. doi: 10.1091/mbc.e05-09-0853 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0063] [63].Thomson DW, Poeckel D, Zinn N, et al. Discovery of GSK8612, a highly selective and potent TBK1 inhibitor. ACS Med Chem Lett. 2019;10(5):780–785. doi: 10.1021/acsmedchemlett.9b00027 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0064] [64].Li F, Xu D, Wang Y, et al. Structural insights into the ubiquitin recognition by OPTN (optineurin) and its regulation by TBK1-mediated phosphorylation. Autophagy. 2018;14(1):66–79. doi: 10.1080/15548627.2017.1391970 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0065] [65].Thurston TL, Ryzhakov G, Bloor S, et al. The TBK1 adaptor and autophagy receptor NDP52 restricts the proliferation of ubiquitin-coated bacteria. Nat Immunol. 2009;10(11):1215–1221. doi: 10.1038/ni.1800 [DOI] [PubMed] [Google Scholar]

[cit0066] [66].Matsumoto G, Shimogori T, Hattori N, et al. TBK1 controls autophagosomal engulfment of polyubiquitinated mitochondria through p62/SQSTM1 phosphorylation. Hum Mol Genet. 2015;24(15):4429–4442. doi: 10.1093/hmg/ddv179 [DOI] [PubMed] [Google Scholar]

[cit0067] [67].Saitoh T, Fujita N, Hayashi T, et al. Atg9a controls dsDNA-driven dynamic translocation of STING and the innate immune response. Proc Natl Acad Sci USA. 2009;106(49):20842–20846. doi: 10.1073/pnas.0911267106 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0068] [68].Ravenhill BJ, Boyle KB, von Muhlinen N, et al. The cargo receptor NDP52 initiates selective autophagy by recruiting the ULK complex to cytosol-invading bacteria. Mol Cell. 2019;74(2):320–9 e6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0069] [69].Vargas JNS, Wang C, Bunker E, et al. Spatiotemporal control of ULK1 activation by NDP52 and TBK1 during selective autophagy. Mol Cell. 2019;74(2):347–62 e6. doi: 10.1016/j.molcel.2019.02.010 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0070] [70].Kumar S, Gu Y, Abudu YP, et al. Phosphorylation of syntaxin 17 by TBK1 controls autophagy initiation. Dev Cell. 2019;49(1):130–44.e6. doi: 10.1016/j.devcel.2019.01.027 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0071] [71].Mizushima N, Yoshimori T. How to interpret LC3 immunoblotting. Autophagy. 2007;3(6):542–545. doi: 10.4161/auto.4600 [DOI] [PubMed] [Google Scholar]

[cit0072] [72].Kimura S, Noda T, Yoshimori T. Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent-tagged LC3. Autophagy. 2007;3(5):452–460. doi: 10.4161/auto.4451 [DOI] [PubMed] [Google Scholar]

[cit0073] [73].Sha Z, Schnell HM, Ruoff K, et al. Rapid induction of p62 and GABARAPL1 upon proteasome inhibition promotes survival before autophagy activation. J Cell Biol. 2018;217(5):1757–1776. doi: 10.1083/jcb.201708168 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0074] [74].Matsumoto G, Wada K, Okuno M, et al. Serine 403 phosphorylation of p62/SQSTM1 regulates selective autophagic clearance of ubiquitinated proteins. Mol Cell. 2011;44(2):279–289. [DOI] [PubMed] [Google Scholar]

[cit0075] [75].Lim J, Lachenmayer ML, Wu S, et al. Proteotoxic stress induces phosphorylation of p62/SQSTM1 by ULK1 to regulate selective autophagic clearance of protein aggregates. PlOS Genet. 2015;11(2):e1004987. doi: 10.1371/journal.pgen.1004987 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0076] [76].Aulas A, Finetti P, Lyons SM, et al. Revisiting the concept of stress in the prognosis of solid tumors: a role for stress granules proteins? Cancers (Basel). 2020;12(9):2470. doi: 10.3390/cancers12092470 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0077] [77].Mazroui R, Di Marco S, Kaufman RJ, et al. Inhibition of the ubiquitin-proteasome system induces stress granule formation. Mol Biol Cell. 2007;18(7):2603–2618. doi: 10.1091/mbc.e06-12-1079 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0078] [78].Markmiller S, Soltanieh S, Server KL, et al. Context-dependent and disease-specific diversity in protein interactions within stress granules. Cell. 2018;172(3):590–604.e13. doi: 10.1016/j.cell.2017.12.032 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0079] [79].Jenkins RW, Aref AR, Lizotte PH, et al. Ex vivo profiling of PD-1 blockade using organotypic tumor spheroids. Cancer Discov. 2018;8(2):196–215. doi: 10.1158/2159-8290.CD-17-0833 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0080] [80].Shen DW, Pouliot LM, Hall MD, et al. Cisplatin resistance: a cellular self-defense mechanism resulting from multiple epigenetic and genetic changes. Pharmacol Rev. 2012;64(3):706–721. doi: 10.1124/pr.111.005637 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0081] [81].Kim HR, Lee SJ, Park S, et al. A single-arm, prospective, phase II study of cisplatin plus weekly docetaxel as first-line therapy in patients with metastatic or recurrent salivary gland cancer. Cancer Res Treat. 2022;54(3):719–727. doi: 10.4143/crt.2021.1019 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0082] [82].Freibaum BD, Messing J, Nakamura H, et al. Identification of small molecule inhibitors of G3BP-driven stress granule formation. J Cell Biol. 2024;223(3):e202308083. doi: 10.1083/jcb.202308083 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0083] [83].Barbie DA, Tamayo P, Boehm JS, et al. Systematic RNA interference reveals that oncogenic kras-driven cancers require TBK1. Nature. 2009;462(7269):108–112. doi: 10.1038/nature08460 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0084] [84].Jiang Y, Chen S, Li Q, et al. Tank-binding kinase 1 (TBK1) serves as a potential target for hepatocellular carcinoma by enhancing tumor immune infiltration. Front Immunol. 2021;12:612139. doi: 10.3389/fimmu.2021.612139 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0085] [85].Hou F, Sun L, Zheng H, et al. MAVS forms functional prion-like aggregates to activate and propagate antiviral innate immune response. Cell. 2011;146(3):448–461. doi: 10.1016/j.cell.2011.06.041 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0086] [86].Clark K, Peggie M, Plater L, et al. Novel cross-talk within the IKK family controls innate immunity. Biochem J. 2011;434(1):93–104. doi: 10.1042/BJ20101701 [DOI] [PubMed] [Google Scholar]

[cit0087] [87].Wang B, Maxwell BA, Joo JH, et al. ULK1 and ULK2 regulate stress granule disassembly through phosphorylation and activation of VCP/p97. Mol Cell. 2019;74(4):742–57.e8. doi: 10.1016/j.molcel.2019.03.027 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0088] [88].Cherkasov V, Hofmann S, Druffel-Augustin S, et al. Coordination of translational control and protein homeostasis during severe heat stress. Curr Biol. 2013;23(24):2452–2462. doi: 10.1016/j.cub.2013.09.058 [DOI] [PubMed] [Google Scholar]

[cit0089] [89].Li H, Lin PH, Gupta P, et al. MG53 suppresses tumor progression and stress granule formation by modulating G3BP2 activity in non-small cell lung cancer. Mol Cancer. 2021;20(1):118. doi: 10.1186/s12943-021-01418-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0090] [90].Cruz VH, Brekken RA. Assessment of tank-binding kinase 1 as a therapeutic target in cancer. J Cell Commun Signal. 2018;12(1):83–90. doi: 10.1007/s12079-017-0438-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0091] [91].Zhu L, Li Y, Xie X, et al. TBKBP1 and TBK1 form a growth factor signalling axis mediating immunosuppression and tumourigenesis. Nat Cell Biol. 2019;21(12):1604–1614. doi: 10.1038/s41556-019-0429-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[cit0092] [92].Cai H, Yan L, Liu N, et al. IFI16 promotes cervical cancer progression by upregulating PD-L1 in immunomicroenvironment through STING-TBK1-NF-kB pathway. Biomed Pharmacother. 2020;123:109790. doi: 10.1016/j.biopha.2019.109790 [DOI] [PubMed] [Google Scholar]

[cit0093] [93].Sun Y, Revach OY, Anderson S, et al. Targeting TBK1 to overcome resistance to cancer immunotherapy. Nature. 2023;615(7950):158–167. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

TBK1 is a signaling hub in coordinating stress-adaptive mechanisms in head and neck cancer progression

Hyo Jeong Kim

Haeng-Jun Kim

Sun-Yong Kim

Jin Roh

Ju Hyun Yun

Chul-Ho Kim

ABSTRACT

Introduction

Results

TBK1 is negatively regulated by MUL1 E3 ligase

Figure 1.

MUL1 induces K48-linked ubiquitination of TBK1 at lysine 584

Figure 2.

HSP90 regulates TBK1 stability and activity

Figure 3.

TBK1 inhibition perturbs autolysosome formation and selective autophagic clearance of aggregates

Figure 4.

Elevated expression and predominant cytosolic localization of G3BP1 and PABP in HNC patients

Figure 5.

TBK1 is required for SG formation in stressed HNC cells

Figure 6.

A novel TBK1 inhibitor, GSK8612, suppresses xenograft tumor growth in vivo

Figure 7.

Discussion

Materials and methods

Reagents

Cell culture and transfection

Plasmids

Immunoprecipitation assay

Ni-NTA affinity-isolation assay

Western blot

Immunofluorescence

Imaging quantification

Transmission electron microscopy (TEM)

Tumorsphere formation

Xenograft in vivo tumor model

Statistical analysis

Supplementary Material

Funding Statement

Disclosure statement

Data availability statement

Author contributions

Ethics statement

Supplementary material

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases