NET-Triggered NLRP3 Activation and IL18 Release Drive Oxaliplatin-Induced Peripheral Neuropathy

Abstract

Oxaliplatin is an antineoplastic agent frequently used in the treatment of gastrointestinal tumors. However, it causes dose-limiting sensorimotor neuropathy, referred to as oxaliplatin-induced peripheral neuropathy (OIPN), for which there is no effective treatment. Here, we report that the elevation of neutrophil extracellular traps (NET) is a pathologic change common to both cancer patients treated with oxaliplatin and a murine model of OIPN. Mechanistically, we found that NETs trigger NLR family pyrin domain containing 3 (NLRP3) inflammasome activation and the subsequent release of IL18 by macrophages, resulting in mechanical hyperalgesia. In NLRP3-deficient mice, the mechanical hyperalgesia characteristic of OIPN in our model was reduced. In addition, in the murine model, treatment with the IL18 decoy receptor IL18BP prevented the development of OIPN. We further showed that eicosapentaenoic acid (EPA) reduced NET formation by suppressing the LPS-TLR4-JNK pathway and thereby abolished NLRP3 inflammasome activation and the subsequent secretion of IL18, which markedly prevented oxaliplatin-induced mechanical hyperalgesia in mice. These results identify a role for NET-triggered NLRP3 activation and IL18 release in the development of OIPN and suggest that utilizing IL18BP and EPA could be effective treatments for OIPN.

Figure 1.

Chemotherapy triggers NET formation. A and B, Adult male C57BL/6J mice were intraperitoneally injected with L-OHP (3 mg/kg) for 5 consecutive days for a total dose of 15 mg/kg to induce mechanical hyperalgesia. The mechanical pain threshold was tested for 14 days by the von Frey test (n = 8). L-OHP = oxaliplatin. C, The protein level of H3Cit in the DRG was evaluated by western blot on the 14th day after the initiation of L-OHP treatment (n = 4). D, Quantification of MPO activity was assessed in the DRG on day 14 after the initiation of L-OHP treatment (n = 8). E–G, The content of H3Cit, NE, and cfDNA in plasma after intraperitoneal injection of L-OHP was evaluated using the H3Cit ELISA kit, NETosis Assay, and Quant-iTPico green dsDNA assay, respectively, at day 14 after the initiation of L-OHP treatment (n = 8). H, Mice were intraperitoneally injected with carboplatin (CBP; 10 mg/kg) three times (days 1, 3, and 7); cisplatin (DDP; 2.3 mg/kg) for 5 consecutive days; paclitaxel (PTX; 8 mg/kg) three times (days 1, 3, and 5); and vincristine (VLB; 100 μg/kg) for 5 consecutive days to induce mechanical hyperalgesia. Then, the level of cf-NDA in plasma was evaluated by Quant-iTPico green dsDNA assay at day 7 after the initiation of chemotherapeutic drugs (n = 6). I, Representative IHC images of hind paw cross-sections stained with the NET marker H3Cit from mice treated with saline or L-OHP. The zones of the dermis and subcutaneous tissue are labeled with dotted lines, and the arteries are labeled with asterisks (n = 3). Scale bar, 50 μm. J, Representative immunofluorescence microscopy images of DRG from mice treated with saline or L-OHP stained for Ly6G, H3Cit, and DNA (n = 3). Scale bar, 50 μm.K, Representative in vivo multiphoton microscopy images of neutrophils (purple), NETs (red), and vessels (green) in the ears of mice 5 days after the initiation of L-OHP treatment. Neutrophils were localized in ear vessels and in the parenchyma. Blood vessels (green) were labeled by intravenous injection of FITC-dextran (MW = 2,000,000 Da). Neutrophils (purple) were labeled by intravenous injection of APC-conjugated monoclonal Ly6G antibody. NETs (red) were labeled by intravenous injection of Sytox Orange (n = 3). Scale bar, 50 μm. L, Mice were administrated 300 μg anti-mouse Ly6G 1 day before the initiation of L-OHP treatment and every 3 days after that until day 14. IgG2α isotype control was administered in the same way. The mechanical pain threshold was tested for 14 days by the von Frey test (n = 6). M, WT and Pad4^−/− mice were intraperitoneally injected with L-OHP (3 mg/kg) for 5 days. The mechanical pain threshold was tested for 14 days by the von Frey test. N, PAD4 inhibitor Cl-amidine (10 mg/kg, i.p.) and MPO inhibitor PF-1355 (50 mg/kg, i.p.) were administered 1 day before the initiation of L-OHP administration once a day until the end of 14 days. The mechanical pain threshold was tested for 14 days by the von Frey test (n = 8). O, The concentration of H3Cit in the plasma of mice mentioned in M and N was evaluated by ELISA on day 14 (n = 6). ***, P < 0.001. Data, mean ± SEM.

Figure 2.

Chemotherapy induces gut microbe–derived NET formation. Adult male C57BL/6J mice were intraperitoneally injected with L-OHP (3 mg/kg) for 5 days for a total dose of 15 mg/kg. A, The degree of intestinal barrier disruption and inflammatory infiltration were evaluated by H&E staining (n = 3). Scale bar, 50 μm. B and C, LPS levels in the serum and DRG of mice were evaluated by Pierce Chromogenic Endotoxin Quant Kit (n = 8). D and E, For gut microbiota eradication, mice were provided drinking water containing 0.5 g/L ampicillin, 0.5 g/L neomycin, 0.5 g/L metronidazole, and 0.25 g/L vancomycin with 3 g/L artificial sweetener Splenda for 3 weeks. Then, the mice were intraperitoneally injected with L-OHP (3 mg/kg) for 5 days. The mechanical pain threshold was tested for 14 days by the von Frey test (n = 8). F and G, The level of H3Cit and NE in plasma was evaluated at day 14 after the initiation of L-OHP treatment using the H3Cit ELISA kit and NETosis Assay, respectively (n = 8). H, NET formation was examined by scanning electron microscopy following 4 hours of stimulation of neutrophils with LPS (10 ng/mL), L-OHP (10 μmol/L), or LPS (10 ng/mL) + L-OHP (10 μmol/L). Cells treated with LPS (1 μg/mL) were used as a positive control (n = 3). Scale bar, 10 μm. I, Neutrophils derived from mouse bone marrow were pretreated with LPS (10 ng/mL), L-OHP (10 μmol/L), or LPS (10 ng/mL) + L-OHP (10 μmol/L) for 4 hours, and immunofluorescence staining of NETs was performed: H3Cit (green), MPO (red), and DAPI (blue). The white arrow showed the details of NETs (n = 3). Scale bar, 50 μm. ***, P < 0.001. Data are shown as mean ± SEM.

Figure 3.

Oxaliplatin induces NET-mediated proinflammatory cytokine production. A–F, Adult male C57BL/6J mice were intraperitoneally injected with L-OHP (3 mg/kg) for 5 days for a total dose of 15 mg/kg. The levels of IL1β, IL6, and TNFα in plasma and DRG were evaluated by ELISA on day 14 after the initiation of L-OHP treatment (n = 8). G, The level of phospho-NF-κB p65 protein in the DRG was evaluated by western blot on the 14th day after the initiation of L-OHP treatment (n = 4). H, Representative immunofluorescence microscopy images of DRGs from WT and Pad4^−/− mice stained with F4/80 (green) and H3Cit (red) on the 14th day after the initiation of L-OHP treatment. The white boxed area shows the details of NETs colocalizing with macrophages (white arrow; n = 3). Scale bar, 50 μm. I–K, The levels of Il1b, Il6, and Tnfa mRNA in the DRG were evaluated by qPCR at day 14 after the initiation of L-OHP treatment in WT and Pad4^−/− mice (n = 4). ***, P < 0.001. Data are shown as mean ± SEM.

Figure 4.

The TLR7/TLR9 signaling pathway contributes to the activation of the NLRP3 inflammasome induced by NETs. A and B, Adult male C57BL/6J mice were injected with DNase1 (150 U, i.v.) or RNase A (1 mg/kg, i.v.) from the initiation of L-OHP treatment to the 14th day. The mechanical pain threshold was tested for 14 days by the von Frey test (n = 8). C, The level of H3Cit in plasma from the mice in A–B was evaluated by H3Cit ELISA kit on day 14 after the initiation of L-OHP treatment (n = 8). D, Neutrophils were incubated in LPS (1 μg/mL) for 4 hours and stained with DNA Sytox Green, RNA Syto RNAselect Green and DAPI (blue). White arrows indicate NETs (n = 3). Scale bar, 50 μm. E–G, BMDMs were pretreated with a TLR2 inhibitor (C29, 10 μmol/L), TLR4 inhibitor (TAK242, 10 μmol/L), TLR7 inhibitor (IRS661, 1 μmol/L), TLR8 inhibitor (CU-CPT9a, 10 μmol/L), or TLR9 inhibitor (IRS869, 1 μmol/L) for 1 hour and then treated with NETs (500 ng/mL) for 3 hours. The levels of Il1b, Il6, and Tnfa mRNA were measured by qPCR (n = 3). H–J, BMDMs were pretreated with a TLR7 inhibitor (IRS661, 1 μmol/L) plus a TLR9 inhibitor (IRS869, 1 μmol/L) for 1 hour and then treated with NETs for 3 hours. The levels of Il1b, Il6, and Tnfa mRNA were measured by qPCR (n = 3). K, Representative immunofluorescence microscopy images of NETs in BMDMs. Neutrophils were incubated with LPS (1 μg/mL) for 4 hours to induce NET formation. Then, BMDMs were treated with NETs for 1 hour and stained with H3Cit (green), TLR7 or TLR9 (red), and DAPI (blue). Each panel represents a section from the stack on the z axis appropriately chosen to visualize the colocalization of H3Cits with TLR7 or TLR9. For each field, two 3D reconstruction sections perpendicular to the plane of the monolayer and parallel to the x or y axis are shown above (x–z section, green line) and to the right (y–z section, red line) of each panel (n = 3). Scale bar, 10 μm. L, Adult male C57BL/6J mice were administered with inhibitors for TLR7 and TLR9 (IRS954, 10 mg/kg, i.p.) 1 day before the initiation of L-OHP treatment until the end of 14 days. The mechanical pain threshold was tested for 14 days by the von Frey test (n = 8). M–O, BMDMs were pretreated with cytochalasin D (5 μmol/L) for 1 hour and then treated with NETs for 3 hours. The levels of Il1b, Il6, and Tnfa mRNA were measured by qPCR (n = 3). P, BMDMs transfected with siRNA targeting the genes encoding NLRP1, AIM2, NLRP3, or NLRC4 were treated with NETs for 3 hours and then treated with oxaliplatin (10 μmol/L) for 1 hour (n = 3). Western blot was used to evaluate the levels of IL1β in supernatants.  *, P < 0.05; **, P < 0.01; ***, P < 0.001. Data are shown as mean ± SEM.

Figure 5.

NLRP3 activation and subsequent IL18 release contribute to oxaliplatin-induced mechanical hyperalgesia. A, The level of ROS in the DRG at day 14 after the initiation of L-OHP treatment was assessed by DCFH-DA staining (n = 8). B and C, The level of ROS in BMDMs stimulated with L-OHP (1, 5, or 10 μmol/L) for 1 hour was assessed by calculating the ratio of DCFH-DA–positive cells to 10,000 cells through flow cytometry. BMDMs stimulated with rotenone (10 μmol/L, 6 hours) were used as a positive control (n = 4). D–K, BMDMs were stimulated with NETs for 3 hours and then stimulated with L-OHP (10 μmol/L) or left unstimulated for 1 hour. LPS (1 μg/mL)-primed BMDMs were treated with ATP (1.5 mmol/L) as a positive control (n = 3). Western blot was used to assess the activation of NLRP3 inflammasome–related proteins [NLRP3, Pro-IL1β, Pro-caspase-1, Caspase-1 (p20), IL1β] in BMDMs treated with NETs and/or L-OHP in both culture supernatants [caspase-1 (p20), IL1β] and cell lysates (NLRP3, Pro-IL1β, Pro-caspase-1; D-I). IL1β or IL18 in supernatants were analyzed by ELISA (J and K). L–Q, WT and Pad4^−/− mice were treated with L-OHP (3 mg/kg) for 5 consecutive days to induce mechanical hyperalgesia. The protein levels of NLRP3, caspase-1 (p20), IL1β, IL18, and H3Cit in the DRG were evaluated on the 14th day by western blot (n = 3). R, The levels of IL18 in WT and Pad4^−/− mice plasma were evaluated by ELISA at day 14 after the initiation of L-OHP treatment (n = 8). S, WT and Nlrp3^−/− mice were treated with L-OHP (3 mg/kg) for 5 consecutive days for a total dose of 15 mg/kg to induce mechanical hyperalgesia. The mechanical pain threshold was tested for 14 days by the von Frey test (n = 8). T, NLRP3 inhibitor MCC950 (20 mg/kg, i.p.) was administered one day before L-OHP administration until the end of 14 days (n = 8). The mechanical pain threshold was tested for 14 days by the von Frey test. U, Mice were treated with IL18BP (200 μg/kg, i.p.) or anakinra (2.5 mg/kg, i.p.) 1 day before L-OHP administration. The mechanical pain threshold was tested for 14 days by the von Frey test (n = 8). V, IL18 (20 ng per mouse, i.p.) was administered one day before L-OHP administration to Nlrp3^−/− mice until the end of 7 days. The mechanical pain threshold was measured for 7 days (n = 6). W, The phosphorylation level of Glu2B in the spinal cord was evaluated on the 14th day after the initiation of L-OHP treatment by western blot (n = 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001. Data are shown as mean ± SEM.

Figure 6.

NETs and IL18 are detected in the plasma of patients after chemotherapy. A and B, The level of H3Cit and cfDNA, as measured by H3Cit ELISA kit and Quant-iTPico green dsDNA assay, respectively, in plasma from healthy volunteers and cancer patients who had or had not received chemotherapy (n = 10). C and D, The levels of IL18 and IL1β in the same plasma samples used in A–B were measured by ELISA (n = 10). E, Correlations between IL18 and H3Cit were analyzed by linear regression analysis (R² = 0.6993, P = 0.0026). F, Correlations between IL18 and cfDNA were analyzed by linear regression analysis (R² = 0.5209, P = 0.0184). G–I, Correlations between VAS and H3Cit, cfDNA or IL18 were analyzed by linear regression analysis (R² = 0.7673, P = 0.0009; R² = 0.7107, P = 0.0022; R² = 0.5148, P = 0.0195, respectively). *, P < 0.05; ***, P < 0.001. Data are shown as mean ± SEM.

Figure 7.

EPA prevents OIPN by suppressing the formation of NETs and abolishing the activation of NLRP3 inflammasome. A, NET formation was quantitatively detected by Sytox Green staining. Bone marrow–derived neutrophils were incubated with LPS (1 μg/mL) for 4 hours. DHA (20 μmol/L) or EPA (20 μmol/L) was added to the medium 1 hour before LPS. Then, the cells were stained with 1 μmol/L Sytox Green for 15 minutes, and fluorescence intensity was captured at an emission peak of 523 nm when excited by a 488 nm argon-ion laser (n = 8). B, Immunofluorescence staining of NETs was performed: H3Cit (green), MPO (red) and DAPI (blue). The images are representative of three independent experiments. Scale bar, 50 μm. C, Bone marrow–derived neutrophils were incubated with LPS (10 μg/mL) for 1 hour, and EPA (20 μmol/L) was added to the medium for 1 hour before LPS treatment. The phosphorylation of JNK was analyzed by western blotting (n = 3). D–G, BMDMs transfected with Arrb2-targeting siRNA were pretreated with EPA (20 μmol/L) for 1 hour before NET stimulation. BMDMs were stimulated with NETs for 3 hours in presence of EPA, and then oxaliplatin (10 μmol/L) was added for 1 hour. Western blot was used to test the activation of NLRP3 inflammasomes–related proteins [Pro-IL18, caspase-1 (p20)] in BMDMs in both culture supernatants (Pro-IL18) and cell lysates [caspase-1 (p20); n = 3)]. D–F, IL18 in supernatants was analyzed by ELISA (n = 3). G. H–O, EPA (1 g/kg, i.g.) was administered by gavage 5 days before the initiation of L-OHP treatment to the 14th day. The levels of cfDNA and H3Cit in the plasma were evaluated by the Quant-iTPico green dsDNA kit and the H3Cit ELISA kit (n = 8; H–I). The mechanical pain threshold was tested for 14 days by the von Frey test (n = 8; J). The protein levels of H3Cit, NLRP3, caspase-1(p20) and IL18 in the DRG were evaluated on the 14th day after the initiation of L-OHP treatment by western blot (n = 3; K–O). P, Schematic illustration indicating that oxaliplatin-induced NETs contribute to the development of OIPN via NLRP3 activation and IL18 release. *, P < 0.05; **, P < 0.01; ***, P < 0.001. Data, mean ± SEM.