Animals

All procedures with experimental animals were approved by the Institutional Animal Care and Use Committee and the Institutional Biosafety Committee of the Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY in accordance with the National Institutes of Health Guidelines. C57 BL/6, B6.129PF, TRPV1 knockout (KO) (B6.129X1-Trpv1tm1Jul/J), TRPA1 KO (B6;129P-Trpa1tm1Kykw/J), α7nAChR KO (B6.129S7-Chrna7tm1Bay/J), Syn-Cre (B6.Cg-Tg(Syn1-Cre) 671Jxm/J), TRPA1fl/fl (129S-Trpa1tm2Kykw/J), Gq-DREADD (B6N;129-Tg(CAG-CHRM3*,-mCitrine)1Ute/J), Vglut2-ires-Cre (Slc17a6tm2(cre)Lowl/J), and Rosa26-GCaMP3 (B6.Cg-Gt (ROSA)26Sortm38(CAG-GCaMP3)Hze/J), mice were purchased from Jackson Laboratory (Jackson Laboratory, Bar Harbor, ME, USA). TRPA1-Cre-DTR mice were a generous donation from Columbia University (C. Zuker laboratory). Syn-Cre female mice were bred with TRPA1fl/fl male mice to generate heterozygous F1 generation Syn-Cre/TRPA1fl/+ mice. F1 generation Syn-Cre/TRPA1fl/+ mice were then back crossed to TRPA1fl/fl mice to create an F2 generation of Syn-Cre/TRPA1fl/fl mice to be used for experiments (Additional file 1: Fig. S2). TRPA1-Cre-DTR mice were bred with IL1R1fl/fl mice to generate TRPA1-Cre/IL1R1fl/+ mice. F1 generation TRPA1-Cre/IL1R1fl/+ mice were then back crossed to IL1R1fl/fl mice to create an F2 generation of TRPA1-Cre/ IL1R1fl/fl mice to be used for experiments (Additional file 1: Fig. S2). TRPA1-Cre-DTR mice were bred with Gq-DREADD mice to generate TRPA1-Cre/Gq-DREADD mice. VGlut2-ires-Cre mice were bred with Rosa26-GCaMP3 mice to generate Vglut2-Cre/GCaMP3 mice that express GCaMP3 allele under control of the Vglut2 locus. The genotypes of the transgenic strains were confirmed using PCR (Transnetyx, Cordova, TN). For these experiments male mice were used as these exhibit a more reliable and severe inflammatory response to endotoxin (Cai et al. 2016). For vagus nerve recording experiments, food was withheld for the 3–4 h prior to recording; animals continued to have access to water.

Telemetry system for temperature recordings

Mice were anesthetized using isoflurane at 2.5% in 100% oxygen at a flow rate of 1 L/min and maintained in supine position at 2.0% isoflurane. A midline incision was made, and an ETA-F10 temperature implant (DSI New Brighton, MN) was placed in the peritoneal cavity tacked to the peritoneal wall. After a minimum 5-day recovery period, mice were placed onto the DSI receiver. Baseline body temperature and relative activity was recorded for 1 h prior to intraperitoneal injection of IL-1β (0.5 µg/kg, 5 µg/kg, 50 µg/kg), saline (8.0 µL/g), clozapine-N-oxide (5 mg/kg), vehicle, or TNF (0.4 mg/kg). Recordings were continued for up to 7 h post injection during their active cycle (lights off). Animals were housed and experiments conducted in a precision temperature-controlled environment assuring constant ambient temperature. A baseline was established at the time of injection (time zero) for area under the curve (AUC) analysis.

Electrophysiological recording

The vagus nerve recordings were performed as described previously (Zanos et al. 2018; Silverman et al. 2018). Briefly, mice were anesthetized using isoflurane at 2.5% in 100% oxygen at a flow rate of 1 L/min, and maintained in supine position at 1.5% isoflurane on a heating pad to maintain core body temperature around 37 °C. The cervical vagus nerve was then exposed and placed on the recording electrode. Recordings were sampled at 40 kHz with a 120 Hz filter and 1:50 gain at 1.25% isoflurane. Electrophysiological signals were recorded using a bipolar cuff electrode (CorTec, Freiburg, Germany) referenced to the animal ground placed between the right salivary gland and the skin. Following acquisition of the baseline activity (5 min), 350 ng/kg recombinant human IL-1β was administered intraperitoneally; recordings were then continued for 5 min post-injection. The electrophysiological signals were digitized from the vagus nerve using a Plexon data acquisition system (OmniPlex, Plexon Inc., Dallas, Texas) and analysed using Spike2 software (version 7, CED) as described previously (Zanos et al. 2018).

Opto-pharmacological stimulation of TRPA1 positive fibers on the vagus nerve

Mice were anesthetized and the left cervical vagus nerve was surgically exposed as described above. Optovin (2 µL of 15 mM solution) was directly applied on the nerve 2 min prior to light stimulation. Animals were subjected to light stimulation (1000 mA, 10 Hz, with a 10% duty cycle for 5 min) using Thor Labs LED driver DC4100, with a 405 nm LED model M405L3 (Newton, New Jersey). Sham stimulation group underwent similar surgical procedure and optovin application, but no light stimulation. Following light stimulation, the wound was approximated in two layers using 5-0 vicryl sutures (Ethicon, Somerville, NJ, USA) and skin clips. Mice were allowed to recover on a heat pad prior to endotoxemia.

Endotoxemia

LPS (Escherichia coli 0111:B4; Sigma; 1 mg/mL in saline) was sonicated for minimum of 30 min and administered intraperitoneally as indicated. Mice subjected to light or sham stimulation were allowed to recover for 2 h prior to LPS administration. In experiments with TRPA1 agonist administration, mice received polygodial (Tocris, Minneapolis, MN, USA) (5 mg/kg, i.p.) or vehicle (i.p.) 30 min prior to LPS administration (0.1 mg/kg). TRPA1-Cre/Gq-DREADD mice were subjected to either chemogenetic stimulation receiving either clozapine N-oxide (Tocris, Minneapolis, MN, USA) at 5 mg/kg (i.p.) or vehicle (i.p.) 1 h prior to LPS administration (0.3 mg/kg i.p.); or direct application of clozapine N-oxide (CNO, 0.25 mg/kg) or vehicle to the vagus nerve for 5 min. Following direct application of clozapine N-oxide to the vagus nerve, mice were recovered for 2 h prior to LPS (1.0 mg/kg). Animals were euthanized by CO2 asphyxiation 90 min post-LPS administration and blood was collected by cardiac puncture. Serum TNF levels were quantitated using commercial enzyme-linked immunosorbent assay (ELISA) (Invitrogen, Thermofisher Scientific, Waltham, MA, USA).

Vagotomy

A 6.0 suture was tied around the vagus nerve, and a surgical cut of the left cervical vagus nerve was completed either rostral or caudal to the optovin application site, using the brain as the point of reference. Animals were then subjected to optovin application, light stimulation and endotoxin administration as previously described.

Neuronal cultures

Nodose ganglia from B6.129PF, TRPA1 KO or Vglut2-Cre/GCaMP3 were dissected into ice-cold neurobasal medium (Gibco, Thermo Fisher Scientific), and dissociated with 1 µg/mL collagenase/dispase (Roche Life Science, Germany) for 90 min at 37 °C on a rotator-shaker. After trituration using to dissociate intact nodose ganglia, cells were filtered with a 70 µm nylon cell strainer, and centrifuged. Cells were plated on poly-l lysine (100 µg/mL, Sigma-Aldrich, St. Louis, MO) and laminin (50 µg/mL, Sigma-Aldrich) coated glass cover slips in 24 well tissue culture plate in complete neurobasal medium [Neurobasal™ medium supplemented with penicillin–streptomycin (Gibco), GlutaMax™ (Gibco), B-27® serum-free supplement (Gibco), 50 ng/mL NGF (Sigma-Aldrich)], and allowed to adhere for 24–48 h at 37 °C (with 5% CO2) prior to proximity ligation assay and intracellular calcium measurements.

Calcium imaging

Sensory nodose ganglion neurons isolated from wildtype (B6.129PF) and TRPA1 KO mice were loaded with Fluo-4 NW with pluronic acid F-127 (Molecular Probes) for 60 min at 37 °C for 45 min in neurobasal medium, washed and imaged at room temperature. For imaging the nodose ganglion neurons from VGlut2-Cre/GCaMP3 mice, cells were isolated, cultured on coverslips as described above and imaged. Confocal images were acquired continuously at a frame rate of 4 Hz with an imaging resolution of 512 × 512 pixels while chemical agonists 20 µg/mL IL-1β, 100 µM AITC (Sigma-Aldrich), 10 µM polygodial, and 10 µM capsaicin (Sigma-Aldrich) were applied via a valve-controlled perfusion system (Warner Instruments) or pipette on a Zeiss LSM-880 confocal laser microscope. 1-[[3-[2-(4-chlorophenyl)ethyl]-1,2,4-oxadiazol-5-yl]methyl]-1,7-dihydro-7-methyl-6H-purin-6-one (AM0902; Tocris, Minneapolis, MN, USA) was perfused as indicated in respective experiments. Cells were washed with HBSS before application of each new agonist. Individual neurons were identified and analyzed for fluorescence intensity changes offline.

Ex vivo intact nodose ganglion imaging

For ex vivo imaging, intact nodose ganglia along with a part of the cervical vagus nerve were excised from VGlut2-Cre/GCaMP3 mice, placed into HBSS buffer, and mounted into a custom-fabricated glass holding chamber. This chamber is designed to sit within a standard liquid perfusion system (Warner Instruments, Hamden, CT, USA) mounted onto the imaging stage of a Zeiss LSM-880 confocal laser microscope. Confocal images were acquired continuously at a frame rate of 4 Hz with an imaging resolution of 512 × 512 pixels while chemical agonists were applied via a valve-controlled perfusion system (Warner Instruments). Neurons were challenged with IL-1β (20 µg/mL), followed by polygodial (200 µM) and capsaicin (10 µM). Cells were washed with HBSS before application of each new agonist. Individual neurons were identified and analyzed for fluorescence intensity changes offline using ImageJ.

Whole-cell patch-clamp recordings

Whole-cell patch-clamp recordings were performed as described previously (Chang and Martin 2016). Nodose ganglia neurons from Vglut2-Cre/GCaMP3 mice were selected for recording by their calcium fluorescence response to batch application of IL-1β (20 µg/mL). For patch-clamp recordings, neurons were visualized on a SliceScope system (Scientifica, Thermo Fisher Scientific) with an Olympus BX51 microscope and Luigs and Neumann micromanipulators. Glass electrodes had a resistance of 2–4 MΩ and contained an intracellular solution (in mM): KCl 140, HEPES 10, EGTA 5, Mg-ATP 2, NaGTP 0.3, MgCl2 2, phosphocreatine 10, pH 7.25 adjusted with KOH. The cells were perfused at a rate of 2 mL/min at room temperature (20–22 °C) with an external bath solution containing the following (in mM): 140 NaCl, 5 KCl, 10 HEPES, 10 d-glucose, 2 CaCl2, and 1 MgCl2, pH 7.3–7.4. Whole-cell currents were acquired using an Multiclamp 700B amplifier (Molecular Devices, Union City, CA) and Windows PC running pClamp 11 software (Molecular Devices). Recordings were filtered at 2 Hz and sampled at 10 Hz. Data was analyzed offline using Clampfit 11 (Molecular Devices) and Origin 2019.

Duolink proximity ligation assay (PLA)

The interactions between TRPA1 and IL1R1 in nodose ganglia cultures were detected by PLA with Duolink in situ kit (Sigma-Aldrich). Primary nodose ganglion cells were harvested and cultured on glass coverslips as described above. The cells were fixed with 4% paraformaldehyde/PBS for 15 min, washed with PBS, permeabilized with 0.2% Triton X-100 for 15 min, blocked with Duolink blocking solution for 60 min at 37 °C, and then incubated in primary antibodies diluted in Duolink antibody diluent overnight at 4 °C. Primary antibodies were rabbit anti-TRPA1 (polyclonal, 1:500; Millipore, Temecula, CA) and goat anti-IL1R1 (polyclonal, 1:40; R&D Systems). After incubation with primary antibodies, cells were washed and then incubated for 60 min at 37 °C in Duolink anti-goat PLUS and anti-rabbit MINUS PLA probe solutions. The coverslips were then washed and incubated with the ligation solution for 30 min at 37 °C. After ligation, the cells were washed and incubated with amplification solution for 100 min at 37 °C. After washing, cells were mounted using Duolink mounting medium with DAPI. Labeled cells were visualized and imaged using a confocal microscope (Zeiss LSM 880). The resulting positive signals were recognized as discrete fluorescent spots. Each spot represents one interaction event. All specimens were imaged under identical conditions and analyzed using identical parameters.

Cecal ligation and puncture

A standardized model of cecal ligation and puncture (CLP)-induced sublethal polymicrobial sepsis was used as previously described (Yang et al. 2015). Mice were anesthetized using ketamine (100 mg/kg, i.p.) and xylazine (8 mg/kg, i.p.). The cecum was isolated and ligated below the ileocecal valve and then punctured with a 22-gauge needle. Approximately 2 mm of stool was then extruded, with the cecum returned to the abdominal cavity. The abdomen was closed with surgical clips. An antibiotic, Primaxin (Imipenem–Cilastatin, 0.5 mg/kg, subcutaneously, in a total volume of 0.5 mL/mouse) was administered immediately after CLP as part of the resuscitation fluid. Mice were monitored for survival and sepsis-associated clinical signs. Disease severity was scoring done on days 0, 3, 4, 5, and 6 post-CLP using the murine sepsis score (MSS) (Shrum et al. 2014).

Statistical analysis

All statistical tests were performed with GraphPad Prism 9 software (GraphPad, La Jolla, CA). Values are presented as individual samples and or mean ± SEM. Statistical analysis of mean differences between groups was performed using two-way ANOVA, paired t-test, Student’s t-test, Welch’s t-test, Mann–Whitney U-test, and mixed effects analysis with multiple comparisons as indicated in respective results. For all analyses, P ≤ 0.05 was considered statistically significant.