Male Sprague-Dawley (SD) rats (250–350 g) were obtained from the Animal Facility of University of Chile. Animal care and experimental protocols for this study were approved by the Institutional Animal Use Committee of University of Chile (N°17027-MED-UCH) and followed the guidelines for ethical protocols and animal care established by the National Institute of Health, MD, United States and the International Association for the Study of Pain in conscious animals. Every effort was made to minimize animal suffering.

Animals were bred and housed in controlled laboratory conditions, received standard rat chow diet and water ad libitum and were housed on a 12-h light/dark cycle at a constant room temperature of 23°C.

We constructed a three-dimensional (3D) molecular model of the heteropentameric GlyR (2α1:3β) based on the X ray crystal structure of α1GlyR subunits from zebrafish (Du et al., 2015a) (Protein Data Bank ID code: 3JAE; Resolution: 3.9 Å). The software MODELLER version 9.18 was used to build the homology models for all sets of receptors (Šali and Blundell, 1993). The GlyR was oriented towards the z axis and embedded in a membrane composed by POPC (1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine) lipids. The system was solvated in a water periodic box (162 × 160 × 123 Å) of TIP3 water molecules and ionized at 0.15 mM NaCl. The initial conformations of each system were subjected to cycles of energy minimization of 10,000 steps and further equilibrated by molecular dynamics of 20 ns with α carbons restrained, in order to relax the conformation of the lateral chains and avoid conformation tension generated during the construction of the model. The calculations were performed using NAMD software and the CHARMM36 united-atom force field for membrane lipids and the CHARMM27 force field with the CMAP correction for proteins (Phillips et al., 2005). All analyses were performed using the VMD software (Humphrey et al., 1996). To study the interaction of IL-1β with the GlyR, the crystal structure of the IL-1β Receptor complex bound to IL-1β (ID code: 3O4O; Resolution: 3.3 Å; RValue: 0.29) was used to build the IL-1β homology model. All models were validated using PROCHEK (Laskowski et al., 1993).

Protein-protein interactions between GlyR with IL-1β were explored. The system was named 2α1:3β-IL-1β (heteropentameric α1βGlyR conformation bound to IL-1β). In order to study the putative interactions between IL-1β and the 2α1:3β GlyR the ClusPro2.0 server was used (https://cluspro.org). ClusPro2.0 is based on three key computational stages. First, an extensive sampling of conformations on a rigid-body docking was performed. Then, resulting conformations were grouped by RMSD according to those with the lowest energy, and finally, a refinement of the structures was performed by energy minimizations. Clusters were classified using electrostatic and Van der Waals interactions. Clusters with sizes greater than the sum of their average plus twice their standard deviation were considered representative, as previously reported (Bottegoni et al., 2006).

Point mutations in the βGlyR subunit were made using a QuikChange kit from Agilent Technologies (Santa Clara, CA, United States), following the manufacturer’s instructions, and checked by sequencing (Macrogen), using the following primers GlyRY240A: Forward: tga​tat​taa​aaa​gga​gga​tat​cga​agc​tgg​caa​ctg​tac​aaa​at and Reverse: att​ttg​tac​agt​tgc​cag​ctt​cga​tat​cct​cct​ttt​taa​tat​ca.

Human embryonic kidney 293 cells (HEK293T) were cultured in Dulbecco’s modified Eagle’s medium supplemented with L-glutamine DMEM-GlutaMAX (Gibco) and 10% fetal bovine serum (Lonza Basel, Switzerland) at 37 °C and 5% CO₂. Once the cells reached 60–70% confluence, they were transfected with a pcDNA3.1 plasmid containing either the rat αGlyR subunit, rat βGlyR subunit (NovoPro Biosciences) or site mutated rat Y240AβGlyR subunits, using the Effectene transfection kit (Qiagen). GlyR subunits were co-transfected with CD-8 as reporter, and labeled using Dynabeads coupled with anti-CD8 antibody (Invitrogen). Cells were transfected with 0.3 μg total DNA at a co-transfection ratio 1:2:0.1 or 1:0.1 for heteropentameric α1GlyR:βGly:CD8 or homopentameric α1GlyR:CD8, respectively. Cell cultures were incubated for 24 h with the Effectene-DNA complexes, followed by trypsinization and seeding in poly-l coated 12 mm coverslips at a density of 300,000 cells/µl. Electrophysiology experiments were performed 48 h after transfection, in 12-mm glass coverslips coated with poly-lysine (Sigma-Aldrich).

Patch Clamp recordings in whole cell configuration where used to measure evoked current amplitudes on transiently transfected HEK293T cells at −60 mV holding potential. Patch electrodes (3.0–5.0 MΩ) were pulled from borosilicate glass (World Precision Instruments, INC 1B150F-4) in a Sutter P97 puller (Sutter Instruments, Novato, CA, United States) and filled with an internal solution buffer composed of (in mM): 140 CsCl, 10 BAPTA, 10 HEPES, 4 MgCl₂, 2 ATP, 0.5 GTP. The external solution buffer was composed of (in mM): 140 NaCl, 3 KCl, 10 HEPES, 1.3 Mg₂Cl and 2.4 CaCl₂ and 10 glucose. Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA, United States) in conjunction with an I/O board NI 6221 PCI (National Instruments) and the Strathclyde electrophysiology software WinWCP (University of Strathclyde, UK) were used for data acquisition. EC₅₀ and nHill values were obtained experimentally from a normalized dose response curve using data points from the evoked currents in response to 30, 50, 100, 300 and 1,000 µM of glycine, using a nonlinear algorithm in the GraphPad Prism 6 software. The dose-response curves were fitted to Hill’s equation: I g l y c i n e = I m a x ∗ [ g l y c i n e ] E C 50 n H i l l + [ g l y c i n e ] n H i l l

The current evoked at a determined concentration (I glycine), the maximal current obtained at saturating concentrations of glycine (I max), the concentration of glycine to reach the half maximal current response (EC₅₀) and the Hill coefficient (nHill), which indicates cooperativity in the binding of glycine to GlyR (Hussein et al., 2019), were estimated from the dose response curves. Perfusion experiments were performed by evoking currents with 50 µM glycine pulses. Then IL-1β (10 ng/ml) was perfused for 60 s, followed by intermittent 50 µM glycine pulses in 10 s intervals until sensitivity was restored. All experiments were performed at room temperature (20°–21°C). These currents were analyzed by fitting a single exponential function to adjust the decay time constant.

Four male SD rats were anesthetized with isoflurane and then decapitated. Their brains were rapidly removed and placed in a beaker with cold artificial cerebrospinal solution (ACSF) composed of (in mM): 85 NaCl, 75 sucrose, 3 KCl, 1.25 NaH₂PO₄, 25 NaHCO₃, 10 dextrose, 3.5 MgSO₄, 0.5 CaCl₂, 3 sodium pyruvate, 0.5 sodium L-ascorbate and 3 myo-inositol (305 mOsm, pH 7.4 with 95% O₂/5% CO₂). Two coronal slices of 300 µm from each hemisphere were obtained from each animal using a vibratome and left for 1 h at 36°C in the ACSF solution. Then the ACSF solution was replaced with a “recording solution,” composed of (in mM): 126 NaCl, 3.5 KCl, 1.25 NaH₂PO₄, 25 NaHCO₃, 10 dextrose, 1 MgSO₄, 2 CaCl₂, 3 sodium pyruvate, 0.5 sodium L-ascorbate and 3 myo-inositol (305 mOsm, pH 7.4 with 95% O₂/5% CO₂) at room temperature (22°C). Slices were recorded in a submerged-style chamber solution with recording solution at 30–32°C, under an upright infrared-differential interference contrast (IR-DIC) fluorescence microscope (Eclipse FNI, Nikon) equipped with a 40× water objective. Voltage-clamp whole cell recordings were performed in the soma of visually identified neurons in the CeA. We used a borosilicate glass electrode (World Precision Instruments, Sarasota, FL, United States) pulled on a P-97 Flaming/Brown Micropipette Puller (Sutter Instruments, Novato, CA, United States). The glass pipette, ranging from 3.5 to 4.2 MΩ, was filled with intracellular solution containing the following (in mM): 130 Cs-gluconate, 3.5 CsCl, 4 ATP-Mg, 0.3 GTP-Na, 10 Na-phosphocreatine, 1 EGTA, 10 HEPES and 0.4% Biocytin (286 mOsm, pH 7.4 adjusted with CsOH). After seal formation and successful transition to whole-cell configuration, access resistance usually between 4 and 15 MΩ, was continuously monitored; series resistance was monitored and compensated between 75 and 80%. Glycine-mediated currents were isolated in the presence of synaptic blockers CNQX (10 µM), APV (50 µM), and picrotoxin (50 µM) to block AMPAR, NMDAR and GABA_A receptors, respectively. The spontaneous inhibitory postsynaptic currents (sIPSCs) obtained as outward currents were recorded at a holding potential of +20 mV. The remaining sIPSCs obtained after further application of bicuculline (10 µM) was considered glycine-based sIPSC. At the end of the recordings, strychnine was applied to block glycinergic currents and thus confirm that the observed currents were glycinergic. We used PClamp detection events to choose single events of about 3–10 pA that were greater than peak to peak noise, to analyze their amplitude, area, and decay constant. Spontaneous synaptic events were detected over 100 s of continuous recording, and two different times following IL-1β application were analyzed (10 ng/ml, at 5 and 15 min). For these experiments, 16 slices out of 4 animals were recorded, and 13 cells were selected for the calculations (the criterium was the recording with stable input resistance (R_in) along the experiment). Voltage-clamp signals were acquired using a MultiClamp 700B amplifier (Axon CNS, Molecular Devices LLC), low-pass filtered at 10 kHz, digitally sampled at 30 kHz and recorded through a Digidata-1440A interface (Axon CNS, Molecular Devices) and PClamp 10.3 software.

The results are expressed as mean ± standard error of the mean (SEM). We used one-way analysis of variance (ANOVA) to determine the effects of IL-1β treatment, followed by a Bonferroni’s post hoc test. The threshold for statistical significance was set at p < 0.05, with a 95% confidence interval. Statistical analyses were performed using Prism software (GraphPad, United States).

To determine whether IL-1β can modulate GlyR activity in the CeA, we performed whole-cell voltage-clamp recordings in CeA neurons from brain slices. Representative traces of isolated spontaneous inhibitory postsynaptic glycinergic currents (sIPSCs), before (Control) and during the application of IL-1β, are presented in Figure 1A. We generated an amplitude distribution plot of the events recorded in all the slices/cells. We observed that the application of IL-1β induced a significant change in the amplitude distribution of sIPSCs, from the beginning of IL-1β perfusion (5 min), and up to 15 min afterward (Figure 1B). GlyR currents transiently shifted toward events of higher amplitude after 5 min of IL-1β perfusion, and reduced the events in all the amplitudes after 15 min (See the table with summarized values). The cumulative probability distribution of the amplitudes showed that the half-probability is transiently shifted to the right at the beginning of IL-1β perfusion. At 15 min, the half-probability is similar to the control. However, the curve is accelerated by the effects of IL-1β (Figure 1C).

The analysis of the mean current amplitude, frequency, and decay constant (tau) is summarized in Figure 1D. We compared the control average with the 5 and 15 min time points after IL-1β perfusion. The data showed that IL-1β transiently and significantly modulated spontaneous glycine current amplitudes (control: 20.4 ± 0.22 pA; IL-1β 5 min: 23.7 ± 0.18 pA; IL-1β 15 min: 19.5 ± 0.23 pA; *p < 0.05, ***p < 0.001; Figure 1D, left). Instead, IL-1β did not significantly change the frequency of spontaneous glycinergic activity (control: 0.15 ± 0.01 Hz; IL-1β 5 min: 0.18 ± 0.02 Hz; IL-1β 15 min: 0.15 ± 0.02 Hz; Figure 1D, middle). However, the presence of IL-1β did transiently increase the decay time constant of the current (control: 7.2 ± 0.72 ms; IL-1β _5min: 18.4 ± 0.5 ms; IL-1β _15min: 8.6 ± 0.9 ms; ***p < 0.001). These analyses suggest that IL-1β can dynamically modulate the amplitude and decay time constant of spontaneous isolated glycinergic currents, suggesting a modulatory effect on GlyR subunits in the CeA.

Previous studies demonstrated that auxiliary βGlyR subunit is broadly expressed in the CeA (Delaney et al., 2010). To determine if βGlyR subunit could be involved in the effects of IL-1β, we constructed a three-dimensional (3D) molecular model of the heteropentameric GlyR (2α1:3β) and IL-1β based on the most complete X-ray crystal structure of α1GlyR available, which was obtained from zebrafish (PDBid: 3JAE) and the IL-1β (PDBid:3o4o). Several three-dimensional reconstructions of αGlyR structure are available, but they are only partial (Du et al., 2015b; Huang et al., 2017; Moraga-Cid et al., 2015). We modeled the ECD and TMD of the βGlyR subunit since there is no available crystallographic structure of the ICD. The sequence identity between both proteins is nearly 50%, which is acceptable for this type of procedure (Burgos et al., 2016). We carried out protein-protein coupling experiments to predict the binding modes of IL-1β on the heteromeric GlyR. The system 2α1:3β-IL-1β showed a representative cluster (See Supplementary Appendix S1). Finally, the result was enhanced on Rosie (Rosseta Online Server, https://rosie.graylab.jhu.edu/docking2). The relevant site identified in the interaction with the IL-1β, corresponded to the back of the loop C of β-subunit in the system 2α1:3β. Molecular Docking suggested that the residues Tyrosine 240, Lysine 235 and Aspartate 237 in the back of loop C of the ECD domain of the βGlyR subunit could have a predominant role in the modulation of IL-1β, mostly interacting through hydrogen bonds (Figure 2A). These residues were identified as hot spots for interactions with IL-1β (Wang et al., 2010). Besides, three important putative interactions between β-subunit and IL-1β through hydrogen bonds were described. These interactions correspond to the hydrogen bonds that were established between residues Gln48 (IL-1β)-Tyr240 (βGlyR), Glu51 (IL-1 β)-Lys235 (βGlyR) and Lys93 (IL-1β)-Asp237 (βGlyR), and showed as red dotted lines in Figure 2B. These results suggested that IL-1β may modulate GlyRs currents by interacting with the βGlyR auxiliary subunit, however to prove it we performed site-directed mutations.

We performed an alignment of human ECD domains of αGlyR and βGlyR subunits focusing on the domain that contains the amino acids proposed to interact with IL-1β (Figure 3). The Tyrosine 240 residue that was divergent in βGlyR was selected to perform a site directed mutagenesis. Then, whole-cell patch clamp electrophysiological recordings were performed on HEK293T cells transfected with α1 alone, α1β wild type (α1βWT) or α1β mutated (α1βY240A), and perfused with glycine (50 μM) and IL-1β (10 ng/ml) as already described (Kawasaki et al., 2008) (Figure 4). The heteropentameric α1βWT exhibited greater current decay time constant when compared to the mutated α1βY240A GlyR (Figure 4A; τ = 35,420 ± 2,671 ms and τ = 3,367 ± 1,684 ms, respectively (n = 3 cells from three different cultures; one-way ANOVA, ****p < 0.0001). The plotted current decay time constants of glycine-activated currents in homo- and heteropentameric GlyRs after the application of glycine or glycine + IL-1β are shown in Figure 4B. The effect of IL-1β significantly increases the decay time of the chloride currents in the presence of a βGlyR subunit in heteropentameric GlyRs; whereas, when the βGlyR was mutated (βY240A) the effect of IL-1β was abolished. This data shows that this subunit is important for the cytokine-induced effect in the GlyRs kinetics. Remarkably, it also suggests that the Tyrosine 240 residue from the βGlyR subunit is critical for IL-1β effects on GlyRs (Figure 4B). On the other hand, the EC₅₀ for glycine was not affected by the mutated subunit (comparing EC₅₀ of α1βWT with α1βY240A; Figure 4C and table), showing that the mutation affects the modulation of GlyRs by IL-1β without altering the pharmacological properties of the channel. Altogether these results suggest that IL-1β modulates GlyR activity through an interaction with the βGlyR subunit.