PLK1 inhibition dampens NLRP3 inflammasome–elicited response in inflammatory disease models

The Journal of Clinical Investigation RESEARCH ARTICLE 1 Introduction NLR family pyrin domain–containing 3 (NLRP3) inflammasome activation is tightly coordinated by collaborative action of pattern recognition receptors (PRRs), and its excessive activation is associated with various diseases including cryopyrin-associated periodic syndromes, gout, atherosclerosis, diabetes, and Alzheimer’s disease (1–5). Activation of TLRs primes the NLRP3 pathway to enhance basal transcriptional expression levels of NLRP3 and IL-1β (6). Once primed, the cells are then ready to respond to various stimuli that activate the NLRP3 inflammasome to form a multiprotein complex (6), which will result in the production of active IL-1β and IL-18 and pyroptosis (7). The multiprotein complex is formed with NLRP3 connected to caspase 1 through an adaptor protein named apoptosis-associated specklike protein with a caspase recruitment domain (ASC), which is encoded by Pycard. Activation of the NLRP3 inflammasome pathway triggers caspase 1–dependent processing of immature IL-1β (pro–IL-1β) and IL-18 (pro–IL-18) into their bioactive counterparts. Previous studies showed that NLRP3 recruitment to a dispersed trans-Golgi network (TGN) is required for NLRP3 aggregation upon activation (8), that the association of cytosolic oxidized mitochondrial DNA with the NLRP3 complex is required for its activation (9), and that NLRP3 positioning at the microtubule-organizing center (MTOC) is required for speck formation (10, 11). Those data demonstrate the importance of correct subcellular localization of NLRP3 for optimal activation of the pathway. Nevertheless, how the tightly controlled NLRP3 inflammasome output is exquisitely coordinated in the different subcellular locations is still largely unknown. Polo-like kinases (PLKs) are a family of serine/threonine protein kinases that are broadly expressed (12). PLK1 is widely studied for its role in mitosis (13, 14). In recent years, it has become apparent that mitotic proteins can play versatile nonmitotic roles (15–17). The major MTOC in the eukaryotic cell is the centrosome, which consists of a pair of centrioles surrounded by a pericentriolar matrix (PCM) that controls microtubule organization by mediating microtubule nucleation and anchoring (18). PCM undergoes a drastic increase in size during centrosome maturation by accumulating γ-tubulin and other PCM proteins (18, 19), and PLK1 is involved in the recruitment of γ-tubulin and other proteins to PCM, thus regulating centrosome maturation (20–22). However, Unabated activation of the NLR family pyrin domain–containing 3 (NLRP3) inflammasome is linked with the pathogenesis of various inflammatory disorders. Polo-like kinase 1 (PLK1) has been widely studied for its role in mitosis. Here, using both pharmacological and genetic approaches, we demonstrate that PLK1 promoted NLRP3 inflammasome activation at cell interphase. Using an unbiased proximity-dependent biotin identification (Bio-ID) screen for the PLK1 interactome in macrophages, we show an enhanced proximal association of NLRP3 with PLK1 upon NLRP3 inflammasome activation. We further confirmed the interaction between PLK1 and NLRP3 and identified the interacting domains. Mechanistically, we show that PLK1 orchestrated the microtubule-organizing center (MTOC) structure and NLRP3 subcellular positioning upon inflammasome activation. Treatment with a selective PLK1 kinase inhibitor suppressed IL-1β production in in vivo inflammatory models, including LPS-induced endotoxemia and monosodium urate–induced peritonitis in mice. Our results uncover a role of PLK1 in regulating NLRP3 inflammasome activation during interphase and identify pharmacological inhibition of PLK1 as a potential therapeutic strategy for inflammatory diseases with excessive NLRP3 inflammasome activation. PLK1 inhibition dampens NLRP3 inflammasome–elicited response in inflammatory disease models Marta Baldrighi,1 Christian Doreth,1 Yang Li,2 Xiaohui Zhao,1 Emily Warner,1 Hannah Chenoweth,1 Kamal Kishore,3 Yagnesh Umrania,4 David-Paul Minde,4 Sarah Thome,1 Xian Yu,1 Yuning Lu,1 Alice Knapton,1 James Harrison,1 Murray Clarke,1 Eicke Latz,5 Guillermo de Cárcer,6,7 Marcos Malumbres,6 Bernhard Ryffel,8 Clare Bryant,9 Jinping Liu,2 Kathryn S. Lilley,4 Ziad Mallat,1,10 and Xuan Li1 1The Victor Phillip Dahdaleh Heart and Lung Research Institute, Department of Medicine, University of Cambridge, Cambridge, United Kingdom. 2Department of Cardiovascular Surgery, Zhongnan Hospital of Wuhan University, Wuhan, China. 3Cancer Research UK Cambridge Centre and 4Department of Biochemistry, Cambridge Centre for Proteomics, University of Cambridge, Cambridge, United Kingdom. 5Institute of Innate Immunity, University Hospital, University of Bonn, Bonn, Germany. 6Cell Division and Cancer Group, Spanish National Cancer Research Centre (CNIO), Madrid, Spain. 7Cell Cycle and Cancer Biomarkers Group, “Alberto Sols” Biomedical Research Institute (IIBM-CSIC), Madrid, Spain. 8UMR7355 INEM, Experimental and Molecular Immunology and Neurogenetics CNRS and Université d’Orleans, Orleans, France. 9Department of Veterinary Medicine, University of Cambridge, Cambridge, United Kingdom. 10Université Paris Cité, PARCC, INSERM, Paris, France. Authorship note: MB, CD, and YL are co–first authors. Conflict of interest: Aspects of this work are included in the United Kingdom (GB) Patent Application (No: 2313514.8) titled “Treatment of Inflammatory Diseases” with XL listed as inventor. Copyright: © 2023, Baldrighi et al. This is an open access article published under the terms of the Creative Commons Attribution 4.0 International License. Submitted: May 24, 2022; Accepted: September 6, 2023; Published: November 1, 2023. Reference information: J Clin Invest. 2023;133(21):e162129. https://doi.org/10.1172/JCI162129. The Journal of Clinical Investigation RESEARCH ARTICLE J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 2 the availability of these specific PLK1 inhibitors (cyclapolin 9 [ref. 32]; SBE13 [ref. 33]; Ro3280 [ref. 34]; BI6727 [ref. 35]), we tested the effect of PLK1 inhibition on IL-1β output upon NLRP3 inflammasome activation stimulated by ATP in WT BMDMs (Figure 1, F–H). PLK1 inhibitors were used at the activation stage (after priming) during the very short time window of ATP stimulation (30 minutes), with caspase 1 inhibitor (Ac-YVAD-FMK) applied as a positive control. Evidently, PLK1 inhibition by all of these compounds substantially reduced IL-1β production (Figure 1F) and cell death (Figure 1G) without affecting TNF-α levels (Figure 1H). We did not observe these inhibitory effects on IL-1β and cell death in NLRP3-deficient cells (Supplemental Figure 1B) or PLK1-deficient cells (Supplemental Figure 1C), indicating that the effect of PLK1 inhibition observed here depended on NLRP3 and PLK1. We detected no effect of PLK1 inhibition on Il1b mRNA levels, even when BI6727 was applied during the priming stage (Supplemental Figure 1D). Collectively, these data indicate that PLK1 inhibition during interphase affected NLRP3 inflammasome activation. PLK1 promotes NLRP3 inflammasome activation. To confirm the role of PLK1 in regulating the NLRP3 inflammasome pathway, we first used the immortalized BMDMs (iBMDMs) stably expressing ASC-mCerulean (36, 37), in which there was evident speck structure formation after inflammasome activation, as confirmed by flow cytometry (Supplemental Figure 2A). Upon NLRP3 inflammasome activation by nigericin, treatment with a PLK1 kinase inhibitor induced a significant reduction of speck-containing cells, together with reduced IL-1β levels when compared with samples without PLK1 inhibitor treatment (Supplemental Figure 2, B–D). We confirmed and extended these data using BMDMs with a wide set of NLRP3 stimuli, including nigericin and sterile crystals such as MSU crystals, cholesterol crystals (CCs), and aluminium hydroxide (15). PLK1 inhibitor treatment significantly reduced IL-1β production in response to all of those stimuli (Figure 2, A, B, and D). However, the PLK1 inhibitor had no effect on IL-1β levels in response to NLRC4 or AIM2 inflammasome stimuli (e.g., flagellin, dA:dT) (Figure 2, C and D), indicating a specific role of PLK1 in NLRP3 inflammasome activation. We further confirmed that PLK1 inhibition substantially reduced caspase 1 and IL-1β cleavage following NLRP3 inflammasome activation using Western blotting (Figure 2D). These data demonstrate that PLK1 inhibitors are effective in suppressing NLRP3 inflammasome activation. To delineate the specificity of PLK1 pharmacological inhibitors, we knocked down PLK1 in the myeloid cell lineage by using the LysM Cre recombinase system (38) and confirmed the efficiency of this system in suppressing PLK1 expression in peritoneal macrophages (Supplemental Figure 3, A and B). To account for the possibility that Plk1–/– peritoneum-resident macrophages derived from the LysM Cre system may undergo impaired differentiation and proliferation during development, we tested the CD11b+F4/80+ macrophages purified from this procedure (Supplemental Figure 3C) and plated an equal number of cells from all biological replicates. We show that knockdown of PLK1 (indicated as LysMCre/WT Plk1fl/fl) in peritoneal macrophages significantly reduced IL-1β production upon ATP-induced NLRP3 inflammasome activation in comparison with control cells (indicated as it is unclear whether PLK1 also plays a similar role in nonmitotic cells. To date, only few studies have indicated a potential role for PLKs in the regulation of inflammation (23–27), and the relevant cellular and molecular mechanisms have remained elusive. For example, Yang et al. showed that PLK4 suppresses NLRP3 inflammasome activation by specifically phosphorylating NEK7 during the LPS-mediated priming step, which in turn inhibits the interaction of NEK7 with NLRP3 and inflammasome activation (27). It is known that PLK1 and PLK4 work together to coordinate mitosis (28, 29). However, it remains unclear whether other PLK proteins are also involved in regulating NLRP3 inflammasome output and, in particular, whether PLKs play a role in the NLRP3 inflammasome activation step at cell interphase. Here, we revealed the role of interphase PLK1 upon NLRP3 inflammasome activation. We demonstrated that PLK1 was expressed throughout the cell cycle and was detectable during interphase in murine macrophages. Using both pharmacological and genetic approaches, we provided substantial evidence that PLK1 inhibition markedly reduced NLRP3 inflammasome output. Using an unbiased proteomics approach based on proximity-dependent biotin identification (Bio-ID), we revealed the PLK1 interactome in the context of NLRP3 inflammasome activation and show an increased association of NLRP3 with PLK1. We further confirmed that PLK1 interacted with NLRP3 at the endogenous level and identified the interacting domains between these 2 proteins. Mechanistically, we showed that PLK1 reinforced the MTOC structure and regulated NLRP3 subcellular positioning. Most important, low doses of PLK1 kinase inhibitor markedly reduced IL-1β level in both LPS-induced endotoxemia and monosodium urate–induced (MSU-induced) peritonitis models in vivo. This study pinpoints a readily available therapeutic strategy for limiting excessive activation of the NLRP3 pathway in inflammatory diseases. Results Interphase PLK1 inhibition reduces IL-1β levels upon NLRP3 inflammasome activation. We used differentiated murine bone marrow– derived macrophages (BMDMs) from WT C57BL/6 mice and found that mitotic cells (identified as phosphohistone H3–positive [PHH3+]) represented only a very low percentage (average, 0.56%), which was further ablated after priming with LPS (Figure 1, A and B). Therefore, we confirmed that the primed BMDMs we used for studying inflammasome activation were at the cell interphase. As expected, inflammasome activation significantly enhanced the cell population with a high ASC fluorescence (30) (ASChi) in live cells of WT BMDMs (Figure 1, A and C). To precisely interrogate PLK1 expression in the different cell subpopulations before and after priming NLRP3 activation, we assessed the intracellular expression level of PLK1 by flow cytometry as validated (Supplemental Figure 1A; supplemental material available online with this article; https:// doi.org/10.1172/JCI162129DS1). As expected, PLK1 was highly expressed in the mitotic cells before priming, and its level was significantly reduced in nonmitotic cells before and after priming (Figure 1, D and E). PLK1 was expressed throughout the cell cycle and was detectable during interphase in murine macrophages. The importance of PLK1 in regulating mitosis has led to the successful development of a number of specific PLK1 kinase inhibitors aimed at retarding tumor growth (31). Facilitated by The Journal of Clinical Investigation RESEARCH ARTICLE 3 J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 of IL-1β (Figure 2F) in Plk1-depleted BMDMs, with no significant effect on TNF-α production (Supplemental Figure 3J). The effect of Plk1 depletion on inhibition of NLRP3 inflammasome activation was further confirmed by decreased caspase 1 and IL-1β cleavage on Western blots (Figure 2G) and reduced levels of ASC speck formation (Figure 2, H and I, and Supplemental Figure 3K). Thus, we were able to recapitulate the effect of PLK1 inhibition on NLRP3 inflammasome activation by knocking out PLK1 in 2 independent genetic depletion models. Overall, our results using both pharmacological and genetic inhibition of PLK1 highlight the important role of PLK1 in NLRP3 inflammasome activation. LysMCre/WT Plk1WT/WT) (Figure 2E). To further verify our findings, we also applied a tamoxifen-inducible Cre recombination system using Rosa-CreErt2 (39). We confirmed the efficiency of this system in depleting PLK1 expression (Supplemental Figure 3, D–F). Fully-differentiated BMDMs from the RosaCreErt2/WT Plk1fl/fl genetic background were treated with tamoxifen to knock out PLK1 and were then primed with LPS. As shown, Plk1 depletion did not affect the levels of Il1b and Tnfa RNA expression (Supplemental Figure 3, H and I). Upon activation with ATP, the Plk1fl/fl genetic background itself had no effect on IL-1β production (Supplemental Figure 3G). However, we observed a significantly reduced level Figure 1. PLK1 inhibition reduces IL-1β output upon NLRP3 inflammasome activation during interphase. (A–C) After 7 days of differentiation, murine BMDMs were primed (100 ng/mL LPS, 5 hours) and activated (5 μM nigericin, for up to 2 hours). Phosphohistone H3 (PHH3) was used as a mitotic marker (A and B), and cells with high ASC fluorescence (ASChi) were identified as an activated speck-forming subpopulation (A and C). n = 3/group. PE, phycoerythrin. (D and E) Geometric (geom) MFI of PLK1 was measured in mitotic cells (dark green) and in nonmitotic cells under the untreated condition (green) and in nonmitotic cells under the primed condition (orange). An isotype antibody control (Isotype Ab. Ctrl) and a secondary antibody control (Sec. Ab. Ctrl) were used. n = 3/group. (F–H) Primed (100 ng/mL LPS, 5 hours) BMDMs were activated (5 mM ATP, 30 minutes), with or without selective PLK1 inhibitors (3 μM cyclapolin 9 [C9]; 10 nM SBE13; 50 nM Ro3280; 0.8 nM BI6727) at the activation stage. The supernatants were collected to measure the IL-1β concentration (F), cell death (G), and the TNF-α concentration (H). The caspase 1 inhibitor Ac-YVAD-FMK or no inhibitor (n.i.) treatment was used as a control. Results are representative of 4 (A–E) or 3 (F–H) independent experiments. n = 9, 7, 5, 5, 5, and 5 (in order from the left to the right bars in F and H). n = 4/group (G). One-way ANOVA with Tukey’s post hoc test was used for statistical analysis. All data are the mean ± SEM. The Journal of Clinical Investigation RESEARCH ARTICLE J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 4 Figure 2. PLK1 inhibition reduces pro–caspase 1 and pro–IL-1β cleavage and ASC speck formation in response to various NLRP3 inflammasome stimuli. (A–C) Primed (100 ng/mL LPS, 5 hours) WT or NLRP3-KO BMDMs, treated with NLRP3 inflammasome–activating stimuli as indicated (5 μM nigericin for 2 hours; 250 μg/mL MSU for 3 hours; 250 μg/mL cholesterol crystals [CC] for 3 hours; 250 μg/mL Alum crystals for 6 hours; 1 μg/mL flagellin for 3 hours; 1 μg/mL poly(dA:dT) for 3 hours), were subjected to PLK1 inhibition by cyclapolin 9 (3 μM) at the activation stage. Supernatants were collected for IL-1β quantification by ELISA. n = 6/group (A and B); n = 3–4/group (C). (D) Cell lysates and supernatants from WT BMDMs treated as in A–C as indicated were analyzed by Western blotting. (E) Peritoneal macrophages were elicited upon i.p. treatment with 1 mL 3% thioglycolate for 4 days and then isolated. Isolated macrophages were primed (100 ng/mL LPS, 5 hours) and activated (5 mM ATP, 30 minutes). Supernatants were collected for IL-1β quantification by ELISA. Control: LysMCre/WT Plk1WT/WT (n = 3); deletion: LysMCre/WT Plk1fl/fl (n = 4). (F–I) RosaCreErt2/WT Plk1fl/fl BMDMs were treated with 4-OH tamoxifen (0.002 mg/mL 4-OH Tam, 24 hours) before and during priming (100 ng/mL LPS, 5 hours) to deplete PLK1, and then cells were activated (5 mM ATP, 30 minutes). (F) Supernatants were collected for IL-1β quantification by ELISA (n = 8). (G) Cell lysates and supernatants were analyzed by Western blotting. Cells stained for ASC were used to quantify the percentage of speck-containing BMDMs across treatments. Scale bars: 20 μm (H) and quantification (No Tam, n = 10; 4-OH Tam, n = 12) (I). Results are representative of 3 (A and C) or 2 (D–G) independent experiments. Two-way ANOVA with Šidák’s post hoc test was used for statistical analysis (A–C). An unpaired t test was used for statistical analysis (E, F, and I). All data are the mean ± SEM. CASP1, caspase 1; MW, molecular weight. The Journal of Clinical Investigation RESEARCH ARTICLE 5 J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 samples were clearly distinguished from the primed samples (Supplemental Figure 5B and Supplemental Table 1.4). Most important, the mass intensity of the identified interactors with PLK1 (normalized by PLK1 value in each batch) was evidently elevated after NLRP3 inflammasome activation (Supplemental Figure 5C and Supplemental Table 1.5), suggesting expanded protein complex formation with PLK1 caused by NLRP3 inflammasome activation. After removing the candidates that responded to BASU tagging alone, the remaining candidates were split on the volcano plot to show the fold change of their proximal association with PLK1 after NLRP3 inflammasome activation (Figure 3E and Supplemental Table 1.3). A significant group of candidates involved in the immune response, including the PRR signaling pathway, showed an increased association with PLK1 after inflammasome activation (Figure 3F and Supplemental Table 2). Among these candidates, several previously known NLRP3 interactors, including DDX3X (46), BTK (47), SYK (48, 49), PKR (50), GBP5 (51), and NLRP3 itself, were identified as proximal interactors with PLK1 upon NLRP3 activation (Figure 3E and Supplemental Table 1.3). PLK1 is a central player involved in regulating cytoskeletal architecture during mitosis (52). In the context of inflammasome activation, we found that candidates involved in cytoskeleton organization constituted a major group of proteins interacting with PLK1 (Figure 3F). To delineate these interactions in further detail, we analyzed subcellular localization of the interacting proteins (Supplemental Table 1.6). We found that, of all the cytoskeletal players, proteins involved in microtubule and centrosome organization showed overall significantly increased interaction with PLK1 (Supplemental Figure 5, D and E, and Supplemental Table 1.7). These results strongly indicate a role of PLK1 in organizing MTOC upon NLRP3 inflammasome activation. PLK1 interacts with NLRP3. PLK1 consists of N-terminal Ser/ Thr kinase domain (KD) and conserved C-terminal polo-box domains (PBDs) (53, 54). Whereas KD is important in phosphorylating its substrates, phospho-dependent ligand recognition by PBD is necessary for PLK1 to target the specific substrates and subcellular locations (54). To further validate and delineate the PLK1-NLRP3 proximal association identified in the PLK1 interactome, we induced co-overexpression of HA-tagged PLK1 together with Flag-tagged NLRP3 in HEK293T cells, in which the endogenous inflammasome proteins are lacking (55). We confirmed that PLK1 coimmunoprecipitated with NLRP3 (Figure 4A). To understand the molecular interaction between PLK1 and NLRP3, we mapped this interaction using various truncations of these 2 proteins and found that full-length PLK1 interacted with pyrin-NACHT domains of NLRP3 (Figure 4B). We also found that full-length NLRP3 interacted with both the KD and PBD of PLK1 (Figure 4C), suggesting that both the kinase activity and ligand-binding function of PLK1 were required for its association with NLRP3. Furthermore, we demonstrated that purified recombinant PLK1 protein (full-length) was able to bind to immobilized recombinant NLRP3 protein (full-length) using a Bio-Layer interferometry assay (Figure 4D), suggesting a direct binding between these 2 proteins is plausible. An in situ proximity ligation assay (PLA) is suitable for quantitative studies of endogenous protein expression, protein modifications, and close protein interactions (56, 57). Using PLA to Proximity proteomics reveals the PLK1 interactome upon NLRP3 inflammasome activation. To investigate how PLK1 promotes NLRP3 inflammasome activation, we used an unbiased screen based on proximity-dependent Bio-ID (40, 41), which relies on the fusion of a promiscuous biotin ligase to the bait of interest. Here, we used BASU (41), an engineered Bacillus subtilis–derived biotin ligase, which we fused to PLK1 (Figure 3A). In our experiment, proteins that came into proximity of the bait protein PLK1 were biotinylated by the biotin ligase BASU, and then the biotinylated proteins were enriched with streptavidin bead pulldown to study the PLK1 interactome (Figure 3, B–D). The BASU enzyme covalently labels lysine residues at its proximity at a rapid speed (41), so this experiment allows the study of protein interaction and even the weak and/or transient interactions in their native states (40). Furthermore, covalent biotinylation enables harsh lysis conditions without losing this modification. Therefore, Bio-ID also allows the identification of protein interactions at the organelles, such as the cytoskeleton, nuclear envelope (40), and centrosome (42), which are refractory to conventional methods for studying protein interactions. To validate whether ectopically expressed BASU-tagged PLK1 could retain native PLK1 function, we reconstituted chemical/genetic PLK1-KO telomerase-expressing human retinal pigment epithelial (tRPE) PLK1 analog-sensitive (PLK1AS) cells (43, 44) with BASU-tagged PLK1 and then induced cell-cycle arrest introduced through PLK1 deletion by 3MB-PP1 treatment as described previously (43, 44). PLK1AS refers to a mutant tRPE cell line, which express mutant PLK1AS with an altered catalytic pocket that can bind bulky purine analogs. The addition of these purine analogs (3MP-PP1) will make a genetic/chemical inhibition that completely interrupts PLK1 function in these cells in a highly efficient manner with practically no off-target effects, as the analogs are too large to fit into the catalytic pockets of other endogenous kinases (43). Our data showed that BASU-tagged PLK1 performed as well as the HA-tagged PLK1 or endogenous PLK1 in rescuing cell arrest (Supplemental Figure 4). To identify the PLK1 interactome in the context of NLRP3 inflammasome activation, we transduced iBMDMs and established a macrophage cell line that stably expressed BASU-tagged PLK1. LPS priming (100 ng/mL for 5 hours) was applied to these cells before biotin addition. Given the presence of biotin trace in the culture medium (45), cells expressing BASU-tagged PLK1 had basal level of biotinylation even without adding more biotin, however, we observed increased intensity of biotinylated proteins after biotin addition (Figure 3C). Prior to nigericin activation of the NLRP3 inflammasome, cells were preincubated with the caspase 1 inhibitor Ac-YVAD-cmk to reduce cell death. The samples from cells with nigericin treatment were used to identify the altered PLK1 interactome after NLRP3 inflammasome activation (activated), and the samples from cells with no nigericin treatment were used as controls (primed). After purification of the biotinylated proteins, equal amounts of the total peptides (5 µg) were subjected to tandem mass tag (TMT) labeling and liquid chromatography tandem mass spectrometry (LC-MS/MS) (Figure 3D). In total, 1,252 candidates passed quality control (Supplemental Figure 5A and Supplemental Table 1.1–1.3). After batch normalization, principal component analysis (PCA) revealed that the experimental groups were correctly clustered, and the activated The Journal of Clinical Investigation RESEARCH ARTICLE J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 6 detect protein-protein interaction in situ, we confirmed that PLK1 was associated with NLRP3 at the endogenous level in BMDMs (Figure 4, E and F). Furthermore, this PLK1-NLRP3 interaction was enhanced after NLRP3 inflammasome activation, and this increased interaction was susceptible to PLK1 kinase inhibition (Figure 4, E and F). Hence, our results confirm that PLK1 interacts with NLRP3 and that NLRP3 inflammasome activation can increase PLK1 kinase activity, which further promotes PLK1-NLRP3 interaction. In the activated samples with PLK1 kinase inhibition, PLK1-NLRP3 interaction did not drop to the same level as in primed-only samples, consistent with the data that PBD was also required for its optimal interaction with NLRP3. Figure 3. An unbiased Bio-ID screen of PLK1 interactome upon NLRP3 inflammasome activation reveals a proximal association of PLK1 with NLRP3. (A) Schematic representation of the bioengineered plasmid expressing the biotin ligase BASU connected to murine PLK1 with a (GGGS)3 linker. (B) Schematic representation of the biotinylated proteins associated with PLK1 and BASU in this assay. (C) Transduced iBMDMs were treated with biotin (50 μM, 2 hours), and cell lysates were run on a Western blot to show biotinylated proteins stained by streptavidin-HRP. The size of the fusion protein BASU-PLK1 is marked by an arrow. (D) iBMDMs transduced with BASU-GS3-PLK1 were treated for inflammasome activation and biotin labeling as indicated. Cells were lysed, and biotinylated proteins were purified using magnetic beads. Trypsinization was followed by peptide quantification, and 5 μg peptides per sample were submitted for TMT labeling and mass spectrometric analysis. (E) Volcano plot for the interactome with PLK1 after NLRP3 inflammasome activation compared with the interactome with PLK1 under the primed condition. Red dots represent the enhanced protein interaction in the activated group, blue dots represent the enhanced protein interaction in the primed group, and gray dots are nonsignificant relative to the selected cutoff threshold (the cutoff threshold for the log2 fold change [FC] is 0.4, equal to a complete 1.3-fold change; the significance-adjusted P value is less than 0.05, by Benjamini-Hochberg correction). (F) GO analysis shows the upregulated (red) and downregulated (blue) proteins interacting with PLK1 after NLRP3 inflammasome activation in protein subgroups with corresponding numbers. The Journal of Clinical Investigation RESEARCH ARTICLE 7 J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 PLK1 reinforces MTOC organization and regulates NLRP3 subcellular positioning. Since PLK1 influences the recruitment of γ-tubulin complexes to the mitotic centrosomes (21), we asked if PLK1 would regulate MTOC structural composition upon inflammasome activation during interphase. In the primed and activated cells with Plk1 depletion (tamoxifen-induced PLK1 depletion in RosaCreErt2/WT Plk1fl/fl BMDMs), we analyzed γ-tubulin content at the centrosomal MTOC and measured microtubule growth by tracking microtubule plus end tip end binding 1 (EB1) growing at the microtubule minus end. In the absence of PLK1, we observed a significant reduction of centrosomal γ-tubulin accumulation (Figure 5, A and B), along with a reduction of microtubule growth from the centrosome as indicated by EB1 (Figure 5, C and D). Microtubule-associated transport is the key intracellular machinery involved in regulating protein subcellular localization (58). To further test whether PLK1 could influence NLRP3 subcellular localization, we used a biochemical fractionation assay. As shown, there were comparable levels of Nlrp3 expression in these cells (Figure 5E). However, in comparison with cells with no PLK1 Figure 4. PLK1 interacts with NLRP3. (A–C) Reconstituted HEK 293T cells were used to perform co-IPs to determine the association between PLK1 and NLRP3 using full-length proteins (A), NLRP3 domains (PYD, pyrin domain; ΔPYD, pyrin domain deletion; LRR, leucine-rich repeat; ΔLRR, LRR deletion) with full-length PLK1 (B), or PLK1 domains (ΔKD, KD deletion; ΔPBD2, PBD2 deletion) with full-length NLRP3 (C). Domain structures of NLRP3 and PLK1 are shown in B and C. Whole-cell lysates were analyzed as an indication of transfection. (D) Bio-Layer interferometric analysis with immobilized, purified NLRP3 protein as the ligand and purified PLK1 protein as the analyte of different concentrations. (E and F) BMDMs were primed (100 ng/mL LPS, 5 hours) and then activated (5 mM ATP, 30 minutes). PLK1 inhibition with 3 μM cyclapolin 9 was used at the activation stage. Interaction between PLK1 and NLRP3 was detected by PLA. Scale bars: 10 μm (E), and quantification of PLA signals per cells (n = 185, 38, 150, and 88, in order from the left bar to the right bar) (F). Two-way ANOVA with Šidák’s post hoc test was used for statistical analysis. All data are the mean ± SEM. The Journal of Clinical Investigation RESEARCH ARTICLE J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 8 kg i.v. in a MSU-induced peritonitis model. The doses were lower than the doses at which BI6727 is normally used in cancer studies (35). In these two models we have used, IL-1β production was largely or totally dependent on NLRP3 (1, 17, 60). Upon LPS challenge in the peritoneal cavity, we detected a substantial reduction of IL-1β levels in the mice treated with BI6727 when compared with mice treated with vehicle control (Figure 6, B and C), in both peritoneal fluids and serum. As shown, IL-1β levels were largely dependent on NLRP3 activation in the LPS-induced endotoxemia model, as IL-1β was substantially reduced in NLRP3KO control mice, regardless of pharmacological treatment (Figure 6, B and C). Under BI6727 treatment, there was only a limited effect of BI6727 on reducing IL-1β levels in the peritoneal fluids of NLRP3-KO mice (Figure 6, B and C), indicating that the effect of BI6727 on IL-1β was largely mediated by NLRP3. However, the effect of BI6727 on reducing the LPS-induced inflammatory depletion upon NLRP3 inflammasome activation, we detected a reduced presence of NLRP3 in both membrane and insoluble cytoskeletal fractions of the PLK1-depleted cells, despite no apparent changes in the cytosolic fractions (Figure 5F). These results suggest altered NLRP3 trafficking between different cellular organelles. Overall, our data highlight the role of PLK1 in organizing MTOC and the microtubule network in the NLRP3 inflammasome pathway and the role of PLK1 in influencing the distribution of its binding partner NLRP3 to specific subcellular locations that are likely associated with optimal NLRP3 inflammasome activation. PLK1 kinase inhibition suppresses IL-1β levels in inflammatory models in vivo. PLK1 is essential for cell proliferation and survival during development, and generic Plk1-KO mice display an embryonic lethality phenotype (59). Therefore, we applied a widely used pharmacological PLK1 kinase inhibitor, BI6727, at doses of 5 mg/ kg i.p. in a LPS-induced endotoxemia model (Figure 6A), or 1 mg/ Figure 5. PLK1 regulates microtubule nucleation and affects NLRP3 inflammasome positioning. (A and B) RosaCreErt2/WT Plk1fl/ fl BMDMs were treated with 4-OH Tamoxifen (0.002 mg/mL 4-OH Tam, 24 hours) before and during priming (100 ng/mL LPS, 5 hours) and were then activated (5 mM ATP, 30 minutes). γ-Tubulin fluorescence was quantified within a 3 × 3 μm ROI centered around the main γ-tubulin focus. Scale bars: 10 μm (lower magnification) and 3 μm (enlarged insets) (A). Quantification by fluorescence intensity (n = 20) (B). (C and D) RosaCreErt2/WT Plk1fl/fl BMDMs were treated as in A. EB1 fluorescence was quantified in a circular ROI of 3 μM diameter. Scale bars: 3 μm (C). Quantification by mean EB1 fluorescence (No Tam, n = 54; 4-OH Tam, n = 77) (D). (E) RosaCreErt2/WT Plk1fl/fl BMDMs were treated with 4-OH tamoxifen (0.002 mg/mL 4-OH Tam, 24 hours) before and during priming (100 ng/mL LPS, 5 hours). Nlrp3 relative expression was quantified by qPCR. n = 3/group. (F) RosaCreErt2/WT Plk1fl/fl BMDMs were treated as in A and processed for fractionation Western blotting. cytop, cytoplasmic; memb, membrane; ins, insoluble fractions. GAPDH, GRASP65, and vimentin, respectively, were used as markers for each fraction. (F) Results are representative of 2 independent experiments. (A–F) Treatments without 4-OH tamoxifen were used as controls. Two-way ANOVA with Šidák’s post hoc test (B) and an unpaired t test (D and E) were used for statistical analysis. All data are the mean ± SEM. The Journal of Clinical Investigation RESEARCH ARTICLE 9 J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 LPS administration leads to systematic inflammatory responses including inflammation in the lung (61) and liver (62), and the degree of the inflammatory response could largely depend on NLRP3 (11, 63). We showed that BI6727 significantly alleviated LPS-induced inflammatory responses in the lung (Figure 6, D and E) and the liver (Figure 6, F and G), as evidenced by reduced levels response was not solely through its inhibition of IL-1β. For example, we also detected that BI6727 treatment in this inflammatory model led to the reduction of TNF-α in both the peritoneal fluids and serum independently of NLRP3 (Supplemental Figure 6, A and B), consistent with the previously described role of PLK1 in TLR-mediated TNF-α production in monocytes (25). Figure 6. PLK1 inhibition suppresses the inflammatory response in a LPS-induced endotoxemia model. C57BL/6 WT and NLRP3-KO mice were treated with BI6727 (5 mg/kg, i.p.) or control (Ctrl) vehicle, followed by LPS administration (20 mg/kg, i.p.). (A) Experimental scheme. Samples for cytokine measurement and tissue assessment were collected at the indicated time points. (B and C) IL-1β levels were measured in peritoneal fluids (B) (WT: Ctrl vehicle, n = 8; WT: BI6727, n = 8; NLRP3-KO: Ctrl vehicle, n = 7; NLRP3-KO: BI6727, n = 8), and in serum (C). WT: Ctrl vehicle, n = 13; WT: BI6727, n = 11; NLRP3-KO: Ctrl vehicle, n = 5; NLRP3-KO: BI6727, n = 7. (D and E) Representative histopathological images from lung tissues and quantification of lung parenchymal area (D), and representative immunofluorescence images of Gr1+ cell staining in lung tissue and quantification (E). WT: Ctrl vehicle, n = 7; WT: BI6727, n = 7; NLRP3-KO: Ctrl vehicle, n = 5; NLRP3-KO: BI6727 group, n = 5. (F and G) Representative histopathological images of the liver and quantitative results of immune cell infiltration (F), and representative immunofluorescence images of Gr1+ cells in liver and quantitative results (G). WT: Ctrl vehicle, n = 7; WT: BI6727, n = 7; NLRP3-KO: Ctrl vehicle, n = 5; KO: BI6727, n = 5. Scale bars: 50 μm. Two-way ANOVA with Šidák’s post hoc test was used for statistical analysis. All data are the mean ± SEM. The Journal of Clinical Investigation RESEARCH ARTICLE J Clin Invest. 2023;133(21):e162129 https://doi.org/10.1172/JCI162129 16 Biotechnol Bioeng. 2019;116(6):1537–1555. 46. Samir P, et al. DDX3X acts as a live-or-die checkpoint in stressed cells by regulating NLRP3 inflammasome. Nature. 2019;573(7775):590–594. 47. Ito M, et al. Bruton’s tyrosine kinase is essential for NLRP3 inflammasome activation and contributes to ischaemic brain injury. Nat Commun. 2015;6:7360. 48. Gross O, et al. Syk kinase signalling couples to the Nlrp3 inflammasome for anti-fungal host defence. Nature. 2009;459(7245):433–436. 49. Lin YC, et al. Syk is involved in NLRP3 inflammasome-mediated caspase-1 activation through adaptor ASC phosphorylation and enhanced oligomerization. J Leukoc Biol. 2015;97(5):825–835. 50. Lu B, et al. Novel role of PKR in inflammasome activation and HMGB1 release. Nature. 2012;488(7413):670–674. 51. Shenoy AR, et al. GBP5 promotes NLRP3 inflammasome assembly and immunity in mammals. Science. 2012;336(6080):481–485. 52. Colicino EG, Hehnly H. Regulating a key mitotic regulator, polo-like kinase 1 (PLK1). Cytoskeleton (Hoboken). 2018;75(11):481–494. 53. Archambault V, Glover DM. Polo-like kinases: conservation and divergence in their functions and regulation. Nat Rev Mol Cell Biol. 2009;10(4):265–275. 54. Archambault V, et al. Understanding the polo kinase machine. Oncogene. 2015;34(37):4799–4807. 55. Khare S, et al. An NLRP7-containing inflammasome mediates recognition of microbial lipopeptides in human macrophages. Immunity. 2012;36(3):464–476. 56. Söderberg O, et al. Characterizing proteins and their interactions in cells and tissues using the in situ proximity ligation assay. Methods. 2008;45(3):227–232. 57. Weibrecht I, et al. Proximity ligation assays: a recent addition to the proteomics toolbox. Expert Rev Proteomics. 2010;7(3):401–409. 58. Bauer NC, et al. Mechanisms regulating protein localization. Traffic. 2015;16(10):1039–1061. 59. Wachowicz P, et al. Genetic depletion of Polo-like kinase 1 leads to embryonic lethality due to mitotic aberrancies. Bioessays. 2016;38 Suppl 1:S96–S106. 60. Sutterwala FS, et al. Critical role for NALP3/ CIAS1/Cryopyrin in innate and adaptive immunity through its regulation of caspase-1. Immunity. 2006;24(3):317–327. 61. Okamoto H, et al. Recombinant antithrombin attenuates acute respiratory distress syndrome in experimental endotoxemia. Am J Pathol. 2021;191(9):1526–1536. 62. Maehara T, et al. Prostaglandin F2α receptor antagonist attenuates LPS-induced systemic inflammatory response in mice. FASEB J. 2020;34(11):15197–15207. 63. Grailer JJ, et al. Critical role for the NLRP3 inflammasome during acute lung injury. J Immunol. 2014;192(12):5974–5983. 64. Liu X, et al. The Delta SARS-CoV-2 variant of concern induces distinct pathogenic patterns of respiratory disease in K18-hACE2 transgenic mice compared to the ancestral strain from Wuhan. mBio. 2022;13(3):e0068322. 65. Barretta ML, et al. Aurora-A recruitment and centrosomal maturation are regulated by a Golgiactivated pool of Src during G2. Nat Commun. 2016;7:11727. 66. Ozaki Y, et al. Poly-ADP ribosylation of Miki by tankyrase-1 promotes centrosome maturation. Mol Cell. 2012;47(5):694–706. 67. Ramani A, et al. Plk1/Polo phosphorylates Sas-4 at the onset of mitosis for an efficient recruitment of pericentriolar material to centrosomes. Cell Rep. 2018;25(13):3618–3630. 68. Kong D, et al. Centriole maturation requires regulated Plk1 activity during two consecutive cell cycles. J Cell Biol. 2014;206(7):855–865. 69. Conduit PT, et al. The centrosome-specific phosphorylation of Cnn by Polo/Plk1 drives Cnn scaffold assembly and centrosome maturation. Dev Cell. 2014;28(6):659–669. 70. Feldman AT, Wolfe D. Tissue processing and hematoxylin and eosin staining. Methods Mol Biol. 2014;1180:31–43. 71. Zhou X, Moore BB. Lung section staining and microscopy. Bio Protoc. 2017;7(10):e2286. 72. Gavilan MP, et al. The dual role of the centrosome in organizing the microtubule network in interphase. EMBO Rep. 2018;19(11):e45942. 73. Lu Y, et al. Interleukin-33 signaling controls the development of iron-recycling macrophages. Immunity. 2020;52(5):782–793. 74. Clément M, et al. Deletion of IRF8 (interferon regulatory factor 8)-dependent dendritic cells abrogates proatherogenic adaptive immunity. Circ Res. 2018;122(6):813–820. 75. Guarda G, et al. T cells dampen innate immune responses through inhibition of NLRP1 and NLRP3 inflammasomes. Nature. 2009;460(7252):269–273. 76. Zhang X, et al. The isolation and characterization of murine macrophages. Curr Protoc Immunol. 2008;Chapter 14:14.1.1–14.1.14. 77. Rajamäki K, et al. Cholesterol crystals activate the NLRP3 inflammasome in human macrophages: a novel link between cholesterol metabolism and inflammation. PLoS One. 2010;5(7):e11765. 78. Murakami Y, et al. Induction of triggering receptor expressed on myeloid cells 1 in murine resident peritoneal macrophages by monosodium urate monohydrate crystals. Arthritis Rheum. 2006;54(2):455–462. 79. Li X, et al. FRS2-dependent SRC activation is required for fibroblast growth factor receptor-induced phosphorylation of Sprouty and suppression of ERK activity. J Cell Sci. 2004;117(pt 25):6007–6017. 80. Schopp IM, et al. Split-BioID a conditional proteomics approach to monitor the composition of spatiotemporally defined protein complexes. Nat Commun. 2017;8:15690. 81. Seidi A, et al. Elucidating the mitochondrial proteome of Toxoplasma gondii reveals the presence of a divergent cytochrome c oxidase. Elife. 2018;7:e38131. 82. Shi H, et al. NLRP3 activation and mitosis are mutually exclusive events coordinated by NEK7, a new inflammasome component. Nat Immunol. 2016;17(3):250–258.