Adult male C57BL/6 mice (postnatal day 17-25; Taconic, Germantown, NY) were housed with food and water available ad libitum in a temperature-controlled environment with light/dark cycle of 12:12 hour. All procedures were conducted in compliance with Georgetown University's Animal Care and Use regulations. All experiments were performed with the approval of the Georgetown University Animal Care and Use Committee. Effort was made to minimize the number of animals used to avoid unnecessary suffering.
Mice were anaesthetized via inhalation of 3% isoflurane and maintained with 1.5% isoflurane while the left footpad was injected. Mice recover with spontaneous activity promptly after removal of the anesthetic. Inflammation was induced by injecting 20 μL of 5% formalin (37% formaldehyde solution diluted in saline solution, 30 gauge needle) into the intraplantar surface of the mouse's hindpaw [59, 60]. Control mice received 20 μL of saline solution (0.9%) into their hindpaw. Animals were monitored regularly after injections.
Nociceptive testing: Thermal hyperalgesia
Hargreaves plantar apparatus (Ugo Basile, Camerio VA, Italy) was used to assess the thermal withdrawal latency (TWL) in the injected hindpaw . Briefly, mice were allowed to acclimate (20 minutes) within a Plexiglas enclosure on a clear glass plate maintained at 30°C. When the mice were stationary, a mobile radiant heat source (i.e., high intensity projector lamp) located under the glass table was activated with a timer and focused onto the plantar surface of the left hindpaw. Thermal paw-withdrawal latency was determined by a motion detector that halted both lamp and timer when the paw was withdrawn. A maximal cut-off of 30 s was employed to prevent tissue damage. Paw withdrawal latencies were recorded automatically. Naïve (uninjected control), saline-treated, and formalin-treated animals were tested. Trials were repeated in the same animals prior to injection (baseline) and at 1, 3, 6 and 24 hours post-injection of either formalin or saline into the left footpad. Each trial was repeated twice at 3 minutes intervals, and the TWL is the average of the two responses. Behavioral response of flinching that may or may not be accompanied by licking of the heated paw was taken as a response. The heat stimulus gave a basal TWL of 7.1 ± 0.1 seconds.
Nociceptive testing: Mechanical allodynia and threshold
For the evaluation of mechanical allodynia, mice were placed individually in clear Plexiglas enclosures on elevated mesh stand (large base 36" × 16") (IITC Life Science, Woodland hills, CA) to allow access to the surface of the hindpaw. The withdrawal threshold (WT) in grams was obtained by application of the rigid tip attached to the anesthesiometer transducer probe (2390 series Electronic von Frey Anesthiometer (EVF), Stoelting Co, Wood Dale, IL). The EVF system allows measurement and displays test readings upon reaction in grams based upon the amount of pressure applied. The EVF is zero calibrated before each use. When the mice were stationary, the stimuli were delivered from below, to the plantar surface of the hindpaw. The animals were acclimatized for 1 hour before behavioral testing and the mechanical allodynia was evaluated at several time-points (1, 6, and 24 hours). In order to determine the basal mechanical thresholds, all the groups were evaluated before the test procedures. Each trial was repeated twice at 5 minutes interval, the withdrawal threshold is an average of the two responses. Saline treated, formalin treated, and ZJ43- Formalin treated animals were tested.
ZJ43 (150 mg/kg) was administered intraperitoneally (IP) 30 minutes prior to and approximately 8 hours after formalin injection into the mouse's left hind paw. Unpublished data from our lab indicate that a single i.p. injection of 150 mg/kg of ZJ43 inhibits NAAG peptidase inhibition for no more than 6 hours (Olszewski R, Ball S, Wegorzewska M, Lee M, Janczura K, Neale JH in preparation). As a result, in this dual injection paradigm there would be no significant peptidase inhibition ongoing at the time of behavioral testing or assessment of the brain slices at 24 hours post inflammation. The ZJ43 treated animals were divided into two groups: (1) used in electrophysiology experiments and (2) used in mechanical threshold (von Frey) experiments 24 hours post formalin injection.
Coronal brain slices (300 μM) (Bregma -1.06 to -1.58 mm) containing the right central nucleus of amygdala, laterocapsular part (CeLC) were prepared using techniques that emulate those used in rats [52, 62]. Briefly, mice (i.e. postnatal day 17-25, C57BL/6, Taconic Lab.) were decapitated, and the brain rapidly removed and placed in ice cold sucrose containing artificial cerebrospinal fluid (ACSF) buffer. No anesthesia was used prior to decapitation to avoid contamination of the brain tissue. Brain slices were then prepared using a micro slicer vibratome 3000 (Vibratome Inc, St Louis MO) and incubated at 32°C for 30 minutes in a sucrose based (ACSF) (mM): (NaCl 86, KCl 3, MgCl2 4, NaH2PO4 1, sucrose 75, glucose 25, CaCl2 1, NaHCO3 25). Slices were then maintained at 32°C for 30 minutes in standard ACSF of the following composition (mM): (NaCl 124, KCL 4.5, MgCl2 1, NaH2PO4 1, glucose 10, CaCl2 2, NaHCO3 26). The slice bathing medium was bubbled with O2/CO2 at a composition of 95% and 5%. After the incubation period, slices were transferred to the recording chamber and, unless stated otherwise, were perfused at 2 ml/min with a nominally Mg2+ free extracellular solution (Mg2+ was omitted from the standard ACSF). All experiments were performed at 30 ± 2°C in the presence of picrotoxin (30 μM, Sigma Aldrich St Louis, MO) and CGP52432 (10 μM, Tocris Bioscience Ellisville, MO) to block GABAergic currents. D-serine a co-agonist at glycine site on NMDA receptor (10 μM, Sigma) was included in nominal Mg2+ free extracellular solution. Coronal brain slices were obtained from untreated control mice, saline treated mice (24 hours post injection), ZJ43 treated formalin mice, and formalin-treated mice at 1, 6 and 24 hours post footpad injection. Using video as a visual guide the stimulating electrode was positioned on the fibers dorsomedial to the CeA and ventral to but outside the caudaputamen, (see fig 8A in ). The spinoparabrachial amygdaloid (PB-CeA) afferents reaching the central nucleus of the amygdala were stimulated using a tungsten bipolar stimulating electrode at a frequency of 0.1 Hz. Stimulation of the afferents in the external capsule (EC), which sends cortical inputs that originate outside of the pain pathway (PB-CeA) to the amygdala was used as a control for input specificity. Recordings were made from neurons in the anterior part of the CeLC region.
Input-output experiments [63, 64] were used to assess the synaptic strength response to incremental increases in stimulus intensity (0.08-5 mA) in the CeLC region and to compare changes in synaptic transmission between treatment groups. The same intensities were used for each animal group; the maximum stimulus intensity was that which failed to further increase the eEPSCs size. Current clamp experiments, with increasing depolarizing current injections of 20 pA steps were performed. Rheobase current was defined as the first current step, within a series of increasing 20 pA steps that elicited an action potential.
Electrodes were pulled in two stages on a vertical pipette puller from borosilicate glass capillaries (Wiretrol II, Drummond, Broomall, PA). The patch pipette solution composition (mM) was: (Kgluconate 145, EGTA 5, MgCl2 5, HEPES 10, ATP-Na 5, GTP-Na 0.2, and pH 7.2 with KOH). For the input-output experiments, the local anesthetic QX-314 (5 mM) was included in the pipette solution to block voltage gated Na+ conductance. Typical pipette resistance was 5-7 MΩ. Whole-cell recordings at Vhold = -60 mV were performed with a patch-clamp amplifier (Axopatch 200B, Axon Instrument, Foster City, CA) under video visual control with a Nikon Eclipse E600FN microscope (Nikon Japan).
Urea based glutamate carboxypeptidase (GCP) inhibitor N-[N-[(S)]-1, 3-dicarboxypropyl] carbamoyl]-L-leucine (ZJ43) was provided by Dr Alan Kozikowski at University of Illinois Chicago. At concentrations as high as 10 μM, ZJ43 fails to act as an agonist or antagonist at group I, II and III mGluRs . The group II metabotropic glutamate receptor (mGluR) antagonist (LY341495), the GABAB antagonist (CGP 52432), the AMPA receptor antagonist [sodium-2, 3-dihydro-6-nitro-7-sulfamoyl-benzo[f]quinoxaline (NBQX)], NMDA receptor antagonist receptor antagonist [3-[(±)-2-carboxypiperazin-4-yl]-propyl-1-phosphonic acid (CPP)], and tetrodotoxin (TTX) were from Tocris Bioscience (Ellisville, MO). N-acetylaspartylglutamate (NAAG) from Tocris. We have long recognized that commercial NAAG is contaminated with 0.1-0.4% glutamate. As a result for more than a decade, we have routinely re-purified by ion exchange chromatography to remove traces of glutamate contamination (final glutamate levels less than 0.1%). The group II mGluR agonist SLx-3095-1 [see  was the generous gift of Alessandra Bartolozzi at Surface Logix, Inc. Boston, MA. SLx-3095-1 is the racemate (+/- isomer as HCl salt) of the highly selective group II mGluR agonist LY379268 (- isomer). Its synthesis is described as compound 9 in .
Stock solutions of TTX, NBQX, picrotoxin and CPP were dissolved in water. ZJ43, LY341495, and SLx-3095-1 were dissolved in saline. NAAG was dissolved in equimolar NaOH. CGP52432 was dissolved in dimethylsulfoxide (DMSO) at 0.01% final concentration.
The recording chamber was perfused continuously with nominal Mg2+ free or standard ACSF in order to minimize spread of the drug beyond the application site. Drugs were diluted in the recording ACSF to obtain optimal concentrations. All drugs solutions were locally applied adjacent to the recording site via a Y-tube  for 5 minutes and modified for optimal solution exchange in brain slices . AMPA receptor antagonist (NBQX, 5 μM) or NMDA receptor antagonist (CPP, 10 μM) were used at the end each experiment based on the type of current being measured to confirm the proper Y tube placement relative to the recording site.
Spontaneous and miniature excitatory postsynaptic currents (sEPSCs and mEPSCs) were identified with semi-automated template based analysis software in Clampfit 9.2 or using a semi-automated threshold based mini detection software (Mini Analysis, Synaptosoft Inc., http://www.synaptosoft.com, Fort Lee, NJ) and visually confirmed. Event detection threshold was set at 5 times the root mean square level of baseline noise. mEPSC and sEPSC averages were based on ≥30 events in each cell studied and the decay kinetics were determined using exponential curve fitting. Evoked excitatory postsynaptic currents (eEPSCs) inhibited or enhanced by various pharmacological agents are represented as absolute magnitude and were determined by comparison of mean baseline before and immediately after drug application. Failures were excluded from evoked current analysis. The change in the peak of the Gaussians was calculated to determine the drug effect. Cells with > 20% change in access resistance were discarded. Off-line evoked current data analysis, curve fitting, and figure preparation was performed using Clampfit 9.2 (Molecular Devices, Inc. Sunnyvale CA) and Microsoft excel.
TWL and mechanical WT values were obtained from the digital readout of the Hargreaves plantar and EVF apparatus respectively.
Statistical significance was determined using a two-tailed Student's t-test unpaired when comparing two populations of cells, and paired when comparing conditions in the same cell population or animal group. In ZJ43 response experiments one way repeated measure-analysis of variance (RM-ANOVA) was followed by pairwise Tukey t-test post hoc analysis (Sigmaplot 11 software, http://www.sigmaplot.com).
In the von Frey experiments, differences between the three groups (saline, formalin and naïve) and changes over time were analyzed for ipsilateral and contralateral hindpaw with a two way RM-ANOVA, followed by Tukey post-hoc comparisons between groups. Significant changes post-injection were established by comparison with baseline values (pre-injection) using Student's t-test for paired data. In each group of the von Frey experiments, differences between contralateral and ipsilateral sides to injection were tested over time with a two way RM-ANOVA. Two way RM-ANOVA analysis was used to analysis changes in TWL over time. The Tukey t-test post-hoc analysis was used for multiple comparisons between groups. All data are presented as mean ± SEM. The significance levels were set at *p ≤ 0.05. In all figures, *p ≤ 0.05 and **p ≤ 0.005 compared to control values, ◊ or #p ≤ 0.05 across groups. The term "control" when applied to brain slices or treatment groups refers to untreated mice.