Skip to main content
NCBI home page
The Journal of Pharmacology and Experimental Therapeutics logoLink to The Journal of Pharmacology and Experimental Therapeutics
. 2024 Dec 18;392(3):100061. doi: 10.1016/j.jpet.2024.100061

Agmatine inhibits NMDA receptor–mediated calcium transients in mouse spinal cord dorsal horn via intact PSD95-nNOS signaling

Tongzhen Xie 1, Rachel E Schorn 2, Kelley F Kitto 2, Stephanie K Florio 3, Cristina D Peterson 2,4, George L Wilcox 2,5,6, Lucy Vulchanova 2, Carolyn A Fairbanks 1,2,4,5,∗
PMCID: PMC11969267  PMID: 39969272

Abstract

Intrathecal administration of agmatine, an NMDA receptor (NMDAr) antagonist and nitric oxide synthase inhibitor, prevents neuropathic pain behavior in a dose-dependent manner by acting at the GluN2B subunit of the NMDAr. The present study investigated the pharmacological mechanism of agmatine’s inhibitory effect using calcium imaging and an in vivo assay of nociceptive responses induced by NMDA. The application of NMDA-evoked calcium transients in the mouse spinal cord dorsal horn slice was inhibited by the NMDAr antagonist, 2-amino-5-phosphonovalerate. Agmatine also concentration-dependently inhibited NMDA-evoked calcium responses. To evaluate the role of the GluN2B subunit of the NMDAr in the agmatine response, we conditionally knocked-down Grin2B, the gene encoding GluN2B, in spinal cord dorsal horn neurons (GluN2B knockdown [GluN2B-KD]). In control spinal cord slices, ifenprodil inhibited NMDAr-mediated calcium transients, but it was not effective in GluN2B-KD. Surprisingly, agmatine was equally effective in reducing calcium transients in control and GluN2B-KD mouse spinal cord slices. To determine whether the effect of agmatine could be attributed to an action downstream of the NMDAr (eg, neuronal nitric oxide synthase [nNOS]), we used the PSD95-nNOS tethering inhibitor, IC87201, to disrupt the link between NMDAr and nNOS. In the presence of IC87201, agmatine’s attenuation of NMDA-evoked calcium transients in ex vivo spinal cord dorsal horn was significantly reversed as was agmatine’s antihyperalgesic effect in the intrathecal NMDA-evoked thermal hyperalgesia in vivo model. These results indicated that agmatine requires an intact NMDAr-PSD95-nNOS pathway to attenuate NMDAr-mediated calcium transients and thermal hyperalgesia induced by intrathecal NMDA.

Significance Statement

Chronic pain is an urgent public health concern, and effective long-term treatments are still needed. Agmatine reduces pain in preclinical models without the side effects of motor dysfunction or addiction. Clarifying the pharmacological mechanism of agmatine’s analgesic effect in spinal neurotransmission may facilitate the development of novel pain-alleviating therapeutics.

Key words: N-methyl-D-aspartate (NMDA), nitric oxide synthase/NOS, Calcium imaging

Graphical abstract

graphic file with name ga1.jpg

Open in a new tab

1. Introduction

Spinal N-methyl D-aspartic acid (NMDA) receptors (NMDArs) play crucial roles in persistent pain development and synaptic plasticity (MacDermott et al, 1986; Dickenson and Sullivan, 1987; Svendsen et al, 1998; Ikeda et al, 2003). Calcium influx through NMDAr activates Ca2+/calmodulin-dependent kinase II (CaMKII) and neuronal nitric oxide synthase (nNOS), leading to downstream signaling amplifying the glutamate response and maladaptive plasticity (Infante et al, 2007; Li et al, 2017). NMDAr antagonists, such as ketamine, have been considered for the treatment of neuropathic pain, but their use has been limited due to dysphoria, hallucinations, and potential abuse liability (Niesters et al, 2014). Agmatine is an endogenous NMDAr antagonist (Reynolds, 1990; Gibson et al, 2002; Molderings and Haenisch, 2012) that reduces nerve injury-induced tactile hypersensitivity without adverse effects in a manner dependent on the GluN2B subunit of the NMDAr (Fairbanks et al, 2000; Piletz et al, 2013; Peterson et al, 2021). Similarly, we have observed through a whole-cell patch-clamp electrophysiology study that agmatine inhibits stimulation-evoked excitatory postsynaptic currents (EPSCs) in spinal cord dorsal horn, also with dependence on the GluN2B subunit of the NMDAr (Waataja et al, 2019). In addition to the antagonism of the NMDAr, agmatine inhibits nNOS (Demady et al, 2001; Aricioglu et al, 2004; Kim et al, 2004), which is also an important contributor to the development of spinal neuroplasticity (Meller et al, 1992; Gilad et al, 1996; Thomas et al, 1996; Yoon et al, 1998).

To evaluate the impact of agmatine inhibition on the NMDA-evoked response of a broad population of dorsal horn neurons activated simultaneously, we applied an ex vivo spinal cord slice calcium imaging technique (Iseppon et al, 2022). We developed an assay of NMDAr-mediated calcium transients in the spinal cord dorsal horn to determine whether agmatine also inhibits the NMDAr-mediated calcium transients. Our calcium imaging assay was performed by incubating ex vivo spinal cord slices with calcium indicator dye Fluo-4, and applying NMDA and other compounds through bath application to evoke or inhibit calcium responses (Heinke et al, 2004; Skorput et al, 2018).

In this report, we used this NMDAr-mediated calcium transients assay to evaluate the mechanisms by which agmatine inhibits NMDA-evoked calcium transients in the spinal cord dorsal horn. Similar to the commonly used NMDAr antagonist 2-amino-5-phosphonovalerate (APV), agmatine concentration-dependently inhibited NMDA-evoked calcium transients in ex vivo spinal cord. We also determined the effect of genetic knockdown of the GluN2B subunit of the NMDAr on agmatine’s inhibition of NMDA-evoked calcium transients. Moreover, our observations demonstrate the impact of pharmacological disruption of the PSD95-nNOS protein–protein interaction on agmatine’s inhibition of NMDA-evoked calcium transients. The present study indicates that agmatine’s inhibition of NMDA-evoked calcium transients is mediated through an intact NMDAr-PSD95-nNOS pathway.

2. Material and methods

2.1. Animals

Male and female Hsd:ICR (CD-1) mice (4–7 weeks, Envigo) or C57Bl/6 mice (as described below) were housed with continuous access to water and food. Mice were housed in a humidity- and temperature-controlled environment. Up to 4 male or 5 female mice were housed per cage with 14-hour light/10-hour dark cycles. All experiments were approved by the Institutional Animal Care and Use Committee at the University of Minnesota.

2.2. Chemicals

Agmatine sulfate, APV, ifenprodil, NMDA, and IC87201 were purchased from Sigma-Aldrich. 2,3-Dioxo-6-nitro-1,2,3,4-tetrahydrobenzo[f]quinoxaline-7-sulfonamide (NBQX) was purchased from Bio-Techne Corporation. All drugs were dissolved in regular artificial cerebral spinal fluid (aCSF). IC87201 was aliquoted and stored as a stock solution of 0.005 mg/mL in 50% DMSO/50% 0.9% saline in a −80 °C freezer. A new aliquot of the stock was freshly thawed for every experiment and diluted to target concentrations with a final DMSO concentration of 5% or less.

For preparation of compounds intended for intrathecal delivery, IC87201 was diluted from the IC87201 stock (0.005 mg/mL in 50% DMSO/50% 0.9% saline) with 0.9% sterile saline. The dilution was done to the extent that the composition of DMSO was less than 1%. An agmatine stock solution and an NMDA stock solution were each prepared in 0.9% saline, aliquoted and stored at −20 °C. For each experiment new aliquots were thawed and used fresh each day.

2.3. Spinal cord slice preparation

The spinal cord slice preparation was adapted from Skorput et al (2018). ICR(CD-1) (4–7 weeks) and GluN2B-floxed mice (7–9 weeks) were anesthetized with 10% isoflurane and perfused transcardially with 45 mL of ice-cold high-sucrose aCSF at a rate of 8 mL/min, which contained (in mM) NaCl 95; KCl 1.8; KH2PO4 1.2; CaCl2 0.5; MgSO4 7; NaHCO3 26; glucose 15; sucrose 50; kynurenic acid 1, bubbled with 95% O2/5% CO2. The spinal cord was extracted following laminectomy and transferred to a chamber holding ice-cold, bubbled, high-sucrose aCSF. Both the dura and arachnoid mater were removed, and the ventral roots were trimmed. The lumbar enlargement was mounted on a 4% agar block and sliced to 400 μm transverse sections using a vibratome (Leica VT 1200S) with 0.04 mm/s sectioning speed and 85 Hz sectioning frequency. Slices from ICR(CD-1) mice were incubated at 35 °C for 45 minutes with 150 μL of 20% Pluronic F-127 in DMSO (Thermo Scientific) suspended in 4.5 mL of regular aCSF containing (in mM): NaCl 127; KCl 1.8; KH2PO4 1.2; CaCl2 2.4; MgSO4 1.3; NaHCO3 26; glucose 15, and 5 μM Fluo-4 AM (Invitrogen) while blowing 95% O2/5% CO2 on the incubant. Slices from GluN2B-floxed mice were incubated in regular aCSF at 35 °C for 45 minutes. After incubation at 35 °C, a 30-minute recovery incubation was carried out at room temperature in regular aCSF before imaging.

2.4. Calcium imaging

The calcium imaging methods were adapted from Skorput et al (2018). Slices were placed in a room temperature stage-mounted imaging chamber on a Nikon A1R multiphoton microscope (objective: 25X, NA 1.1), and confocal images were captured digitally with Nikon Elements software (Nikon; University Imaging Centers; University of Minnesota). The imaging chamber was continuously perfused with oxygenated regular aCSF or drug-containing aCSF using a peristaltic pump (1.5 mL/min). One mM of glutamate was applied for 10 seconds at the end of acquisition in all imaging sessions to ensure cellular viability. Single-plane two-photon images were taken at 30–60 μm below the surface of the slice with 800 nm (Fluo-4 incubated slices) and 920 nm infrared excitation (GCaMP6s-expressing slices). Slices were imaged within 8 hours of tissue harvest. It is noteworthy that drug concentrations applied in intact central nervous system slices (Garthwaite, 1985; Luo et al, 2008; Doolen et al, 2012; Waataja et al, 2019) frequently differ from those used in in vitro dissociated cell culture preparations (Garthwaite, 1985; Donevan and Rogawski, 1993; Wyllie and Cull-Candy, 1994; Olmos et al, 1999; Strauss and Marini, 2002; Shirwany et al, 2009). Examples of such differences in drug concentrations between in vitro cultures and intact central nervous system slices are highlighted in the Supplemental Table 1.

2.5. Measurement of NMDAr-mediated calcium transients and analysis

To isolate the NMDAr-mediated calcium transients, slices were incubated with a mixture of AMPA receptor antagonists with NMDAr agonists for 15 seconds (in μM): NBQX 10; glycine 3; and NMDA 100. After the first NMDAr drug mixture application, a washout period (5 minutes) was performed before incubation with NMDAr antagonists. NMDAr antagonists APV (2, 5, and 10 μM) and agmatine (0.33, 1, 3.3, and 10 mM) were applied singly to each slice. Following the antagonist incubation, NMDAr agonist was applied for the second NMDA-evoked response. One mM of glutamate was perfused for 10 seconds at the end of every imaging session to ensure the cellular viability. Time-lapse of images acquired by ND Acquisition in Nikon Elements software (Nikon; University Imaging Centers; University of Minnesota) was analyzed using Fiji (Schindelin et al, 2012). After applying the fire lookup table and image stabilization (ImageJ, http://www.cs.cmu.edu/∼kangli/code/Image_Stabilizer.html), 1 mM glutamate-responsive cells were marked as the region of interest with the region of interest manager. The intensity acquired 20 seconds before each NMDAr drug mixture application was averaged for assigning resting values. The peak response intensity was determined with the maximum search function in the next 300 frames (5 minutes) after drug application. The maximum response amplitude was calculated by subtracting resting values from the peak response intensity. All statistical analyses were conducted in GraphPad Prism 9 (GraphPad), and all data were analyzed by Student’s t test or two-way ANOVA. The percentage inhibition of the maximum amplitude was calculated by % inhibition = 100 × (1 − [response post-drug]/[response pre-drug]).

2.6. Generation of GluN2B-deficient mice via intraspinal injection of adeno-associated virus 9 virus

The GluN2B-floxed mouse line was generated by the Gene-Targeted Mouse Core of the INIAstress consortium and provided by Dr E. Delpire (Vanderbilt University) (Brigman et al, 2010). With INIAstress consortium approval, a breeding colony of homozygous GluN2B-floxed mice was established at the University of Minnesota. At the time of weaning (postnatal day 21–28), intraspinal injections of adeno-associated virus (AAV) were performed in GluN2Bfl/fl mice (GluN2B knockdown [GluN2B-KD]). Cre-containing AAV9-hSyn-GCaMP6s-P2A-Cre (4.47 × 1013/mL) and control AAV9-hSyn-GCaMP6s-P2A-Δcre (3.71 × 1013 GC/mL) viruses were generated by the University of Minnesota Viral Vector and Cloning Core. Both serotypes of the viruses encoded a Ca2+ indicator, GCaMP6s, to visualize calcium transients in transfected cells. The difference between the 2 vectors was that the Cre-containing vector resulted in GluN2B-KD and expressed GCaMP6s to enable visualizing calcium, whereas the AAV9-hSyn-GCaMP6s-P2A-ΔCre contained ΔCre, which is a noneffective scrambled version of Cre recombinase (Anderson et al, 2017); those subjects were effectively the wild-type control as they did not manifest GluN2B-KD but expressed GCaMP6s to visualize calcium transients.

For spinal cord surgery, an incision was made in the skin overlying T11–T12 and T12–T13 intervertebral spaces, the overlaying muscle was bluntly dissected, and the intervertebral ligaments were separated to allow unilateral injection of 500 nL of virus (0.5 mm lateral to spinal vein, 0.3 mm depth) at both L2–L3 and L4–L5 levels of the spinal cord using a Nanoject III microinjector (Drummond).

During each injection, the mouse was elevated on a gauze cushion, and its spine was stabilized in a spinal clamp to minimize movement artifacts. The muscles and tissue were sutured using veterinary surgical sutures (4–0 monocryl with PC-5 needle) prior to closing the skin with surgical staples. All animals were anaesthetized with isoflurane (2.5%) during injections and treated with meloxicam (2 mg/kg s.c.) for postoperative analgesia for 3 consecutive days following surgery. Within GluN2B-KD mice, neurons expressing Cre recombinase were distinguished from neurons not transduced with AAV vector by GCaMP6s expression.

2.7. Nitric oxide synthase inhibition assays

The nNOS assay was performed by Eurofins Panlabs Discovery Services, Taiwan, and the inducible nitric oxide synthase (iNOS) and endothelial nitric oxide synthase (eNOS) assays were performed by Eurofins Cerep. All assessments of agmatine inhibition of nitric oxide synthase (NOS) included the following concentrations: 0.003, 0.01, 0.03, 0.1, 0.3, 1, 3, 10, 30, 100, 300, 700, 1000 μM. The methods for each assay are briefly described.

2.7.1. nNOS

The nNOS source was Wistar rat cerebellum and [3H]L-arginine (20 nM) was the substrate. The tissue was preincubated for 15 minutes at 25 °C and incubated with [3H]L-arginine for 10 minutes at 25 °C. Production of [3H]citrulline was the indicator of NOS activity, which was measured by scintillation counting. IC50 values were determined by a nonlinear, least squares regression analysis using MathIQTM (ID Business Solutions Ltd).

2.7.2. iNOS

The iNOS method was based on and adapted from a previous report (Tayeh and Marletta, 1989). Briefly, the iNOS source was mouse recombinant iNOS in Escherichia coli L-arginine (100 μM) was the substrate and was incubated for 120 minutes at 37 °C. NO2− was the indicator of iNOS activity and was detected by photometry.

2.7.3. eNOS

The eNOS method was based on and optimized from a past study (Pollock et al, 1991). Briefly, the eNOS source was human umbilical vascular endothelial cells. [3H]L-arginine (28 nM) + arginine (50 nM) were the substrate and stimulus tracers, respectively. Incubation time was 30 minutes at 37 °C. Production of [3H]citrulline was the indicator of NOS activity, which was measured by scintillation counting. A compound’s enzyme inhibition effect was calculated as a % inhibition of control enzyme activity. The IC50 by nonlinear regression analysis of the inhibition/concentration–response curves generated with mean replicate values using Hill equation. This analysis was performed using software developed at Eurofins Cerep (Hill software) and validated by comparison with data generated by the commercial software SigmaPlot 4.0 for Windows (1997 by SPSS Inc).

2.8. Intrathecal injections

All intrathecal injections were delivered to conscious mice between the L5/L6 spinous processes based on the reported method (Hylden and Wilcox, 1980), as described in detail (Fairbanks, 2003). A 30-gauge 0.5-inch needle connected to a 50-μL Luer-hub Hamilton syringe was used to deliver 5 μL of solution to the subarachnoid space. To accomplish this, each individual mouse was gently held by the hip bones and the experimenter placed the needle (bevel side up) at midline, inserted it at an approximate 70°–80° relative to the horizontal until it contacted the bone of the vertebral column. At that point the angle was lowered to about 30° as the experimenter slipped the needle in between the vertebrae. Puncture of the dura was indicated by a reflexive flick of the tail. The injectate was released in a 5 μL volume. All intrathecal doses are expressed as total dose in moles.

2.9. NMDA-evoked thermal hyperalgesia

To evaluate thermal hyperalgesia the warm water tail immersion test was used at a temperature of 49 °C. Baseline tail flick measurements were taken on all mice (baseline of tail flick latency of males was 7.9 ± 0.08 seconds; baseline of females was 7.8 ± 0.2 seconds; mean ± SEM, n = 6). Intrathecal administration of NMDA (0.3 nmol) to mice resulted in typical scratching and biting responses for the first minute after injection, which were counted and recorded (nociceptive behaviors of males was 35 ± 2.9 seconds, that of females was 46.7 ± 6.2 seconds; mean ± SEM, n = 6). Five minutes following the intrathecal injection of NMDA (0.3 nmol), a thermal hyperalgesia arose, as measured by the warm water (49 °C) tail immersion test. Data are expressed as either tail flick latencies (in seconds) or as the average percent inhibition ± SEM, calculated using the formula: Δ Tail flick latency = control value − experimental value.

3. Results

3.1. Agmatine concentration-dependently inhibited NMDA-evoked Ca2+ transients

To visualize Ca2+ transients in ex vivo spinal cord slices, we performed 2-photon microscopy in Fluo-4-incubated spinal cord slices. To determine the optimal concentration of NMDA for a submaximal and repeatable calcium response, 30, 100, and 300 μM of NMDA mixture were applied to spinal cord slices, showing that 100 μM NMDA produced a robust yet moderate response compared with 30 and 300 μM NMDA (representative trace, Fig. 1A). The NMDA mixture also contained endogenous concentration of glycine (3 μM), an obligatory coagonist for NMDAr activation, and NBQX (5 μM), an α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor antagonist, to ensure that the NMDA-evoked calcium influx is NMDAr-specific. The application of the 100 μM NMDA mixture evoked a calcium response, which was reproduced to a comparable degree following 5 minutes of drug-free aCSF washout (representative trace, Fig. 1B). There was no apparent change in the response amplitude between the first and second application of the 100 μM NMDA mixture, indicating stability and reproducibility of the response in the same spinal cord cell after a 5-minute washout period. Using Student’s t test, the response amplitudes of the first and second NMDA applications were not significantly different (Fig. 1C, P > .05), which shows the reproducibility of the NMDA-induced calcium response in the spinal cord.

Fig. 1.

Fig. 1

Open in a new tab

NMDA-evoked calcium transients in spinal cord dorsal horn. Calcium transients were measured in mouse spinal cord dorsal horn slices using two-photon microscopy. ΔF/F is the change in fluorescence intensity relative to the resting fluorescence intensity. (A) Fifteen seconds of 30, 100, and 300 μM NMDA perfusion respectively increased NMDA-evoked calcium transients in mouse spinal cord dorsal horn. Images corresponding to pre-drug application as well as NMDA- (1, 2) and glutamate- (3) induced increases of fluorescence are shown on the right. (B) Submaximal 100 μM NMDA response before and after 5 minutes aCSF washout was reproduced within the same spinal cord cell. Images corresponding to pre-drug application as well as NMDA- (1, 2) and glutamate- (3) induced increases of fluorescence are shown on the right. (C) First and second NMDA-evoked responses were measured with a 5-minute aCSF washout in between NMDA applications. There was no difference in NMDA-induced calcium transients between the 2 NMDA applications (paired Student’s t test, n = 3, P = .19). ns means not significant (P > .05).

Upon establishing the NMDAr-mediated calcium transients assay, we examined the inhibition of calcium transients by the NMDAr antagonists to ensure that our NMDA-evoked calcium transients assay is specific for NMDAr. We tested the effects of the well-established NMDAr antagonist APV. APV is an NMDAr antagonist that competes with glutamate and NMDA at the glutamate binding site of the NMDAr (Davies et al, 1981; Evans et al, 1982). In this preparation, the first NMDA agonist mixture was applied to measure baseline NMDA-evoked calcium response. After a subsequent 5-minute washout with drug-free aCSF, spinal cord slices were then incubated with an NMDAr antagonist (2 minutes), immediately followed by the second NMDA drug mixture (15 seconds). Inhibition of NMDA-evoked calcium transients was determined by measuring the amplitude of the NMDA-evoked response pre- and post-drug incubation. With increasing concentrations of APV (2, 10, 50 μM, 2 minutes each), the NMDA-evoked calcium transients was concentration-dependently inhibited in both male and female subjects (Fig. 2A). Using the same assay of NMDA-evoked calcium transients, agmatine was applied to spinal slices (0.33, 1, 3.3, and 10 mM, 2 minutes each). Agmatine also concentration-dependently attenuated NMDA-evoked calcium responses in both male and female mice (Fig. 2B). These data indicate that agmatine inhibited NMDAr-mediated calcium transients in the spinal cord dorsal horn in a manner similar to that of the established NMDAr antagonist, APV. Neither APV nor agmatine demonstrated any sex difference at any dose. Responses to each dose of APV and each dose of agmatine were compared between male and female by Student’s t test (P = .45 and P = .84, respectively).

Fig. 2.

Fig. 2

Open in a new tab

Agmatine and APV inhibited NMDA-evoked Ca2+ transients concentration-dependently. NMDAr-evoked Ca2+ transients were measured after 2-minute incubation of APV (A) and agmatine (B) in mouse spinal cord ex vivo slices. NMDA-evoked Ca2+ response was attenuated concentration-dependently. Symbols represent mean ± SEM. Each symbol is represented by at least 6 cells per slice; 3 slices per animal; 2 animals per concentration; no sex difference was observed (two-way ANOVA).

3.2. Role of the GluN2B subunit in inhibition of NMDA-evoked calcium transients

We have reported that agmatine requires GluN2B-containing NMDArs in order to fully reduce NMDAr-mediated EPSCs (Waataja et al, 2019) and to fully reverse NMDA-induced mechanical and thermal hyperalgesia (Peterson et al, 2021). In the present study, agmatine inhibited NMDAr-mediated calcium responses in spinal cord dorsal horn, but it is unknown whether agmatine’s inhibition on calcium responses is also preferentially mediated by GluN2B-containing NMDArs. To conditionally knock down the GluN2B subunit-containing NMDArs, GluN2B-floxed mice were intraspinally injected with either AAV9-hSyn-GCaMP6s-P2A-Cre or AAV9-hSyn-GCaMP6s-P2A-ΔCre (Fig. 3A). ΔCre-recombinase is an inactive form of Cre, whereas hSyn allows conditional transduction of AAV in neurons (Anderson et al, 2017). GCaMP6s expression allowed us to measure neuronal calcium transients and also indicates transduction of Cre or ΔCre recombinase in the GCaMP6s-labeled neurons. Three to 4 weeks following the injection of AAV, two-photon microscopy was conducted to measure the inhibition of NMDA-evoked calcium transients by agmatine or ifenprodil. Calcium imaging and analysis were only conducted in GCaMP6s-labeled spinal cord neurons. The NMDA-evoked calcium response in dorsal horn neurons of control and GluN2B-KD animals were compared (Student’s t test, n = 6 per concentration). The response amplitude of 30 μM, 100 μM, and 300 μM NMDAr-mediated calcium transients was not significantly different between control and GluN2B-KD animals (Fig. 3B). As expected, ifenprodil reduced NMDA-evoked calcium transients in control slices but not in GluN2B-KD slices (Fig. 3C.1). The % inhibition of ifenprodil significantly differed between control and GluN2B-KD (Fig. 3C.2, ∗∗∗P < .001). Surprisingly, agmatine reduced NMDA-evoked calcium transients in both control slices and in GluN2B-KD slices (Fig. 3D.1). Also, in contrast to ifenprodil, the % inhibition of agmatine did not differ between control and GluN2B-KD (Fig. 3D.2).

Fig. 3.

Fig. 3

Open in a new tab

GluN2B-NMDAr knockdown (GluN2B-KD) reversed the Ca2+ transients’ attenuation by ifenprodil but not agmatine. (A) Schematic of GluN2B-KD. GluN2B floxed mice were intraspinally injected with AAV containing Cre or Delta Cre. Calcium imaging was conducted 3–4 weeks following the injections. (B) The first NMDA responses (white circle, gray bar) were compared with the second responses (red circle, red bar). 30, 100, and 300 μM NMDA showed no significant difference between GluN2B-KD and control subjects. (C.1, C.2) 100 μM ifenprodil inhibited the NMDA-evoked Ca2+ transients in GluN2B-KD as compared with the control (control in white circles and gray bars; GluN2B-KD in red squares and red bars). (D.1, D.2) 3.3 mM agmatine attenuated NMDA-evoked Ca2+ transients significantly in both control and GluN2B-KD (control in white circles and gray bars; GluN2B-KD in red squares and red bars). Symbols represent single cells’ values, and lines and bars represent mean ± SEM. Each symbol is represented by at least 6 cells per slice; 3 slices per animal; 6 animals per concentration; Student’s t test, ∗∗∗∗P < .001; ∗∗∗P < .005.

3.3. Role of the PSD95-nNOS protein–protein interaction in inhibition of NMDA-evoked calcium transients

Previous reports indicated that agmatine inhibits nNOS, iNOS, and eNOS (Galea et al, 1996). Concurrent with the present study, we contracted with Eurofins to provide independent confirmation that agmatine inhibits nNOS (IC50, 112 μM; confidence interval [CI], 77–153), iNOS (IC50, 120 μM; CI, 68–220), and eNOS (IC50, 850 μM; CI, 670–1100). Based on these data, it seems plausible that the mechanism of agmatine to reduce calcium-evoked transients involves nNOS. To address this question, we developed a strategy to dissociate nNOS from the NMDAr.

nNOS activation requires CaMKII and PSD95, a postsynaptic scaffolding protein in excitatory neurons (Ishii et al, 2006; Courtney et al, 2014). PSD95-nNOS tethering allows nNOS to efficiently interact with NMDAr-mediated calcium influx (Brenman et al, 1996). We previously showed that the nNOS inhibitor 7-nitroindazole (7-NI) dose-dependently inhibited intrathecal NMDA-induced thermal hyperalgesia but not nociceptive behaviors (Roberts et al, 2005), suggesting that NMDA-induced thermal hyperalgesia involves spinal nNOS activation (Kitto et al, 1992; Fairbanks et al, 2000). Agmatine inhibits NMDA-induced thermal hyperalgesia at doses significantly lower than those needed to inhibit the nociceptive behaviors (Fairbanks et al, 2000; Roberts et al, 2005), indicating agmatine’s dual inhibition of nNOS (Galea et al, 1996) and NMDAr (Yang and Reis, 1999). However, it is unknown if agmatine’s attenuation of NMDA-evoked calcium influx is also mediated by nNOS inhibition. We reasoned that if agmatine reduces the NMDAr-mediated calcium transients via inhibition of nNOS that an intact PSD95-nNOS protein-protein interaction would be necessary for agmatine’s inhibitory effect in this assay. To address this question, we used a subeffective concentration of IC87201, which disrupts the PSD95-nNOS protein–protein interaction without disrupting nNOS function (Florio et al, 2009). Two concentrations of IC87201 (10 and 30 μM) were applied to ex vivo mouse spinal cord slices, and NMDA-evoked calcium transients were measured before and after the incubation with IC87201. We observed that 10 μM IC87201 was ineffective but that 30 μM significantly inhibited NMDA-evoked calcium transients (Fig. 4A). The attenuation of NMDA-induced calcium influx in spinal cord dorsal horn by IC87201 is consistent with our previous in vivo IC87201 report, which found dose-dependent inhibition of NMDA-induced thermal hyperalgesia following intrathecal administration of IC87201 (Florio et al, 2009). In order to determine whether the PSD95-nNOS tethering is necessary for the inhibitory effect of agmatine, we first applied the 10 μM concentration of IC87201 that did not inhibit NMDAr-mediated calcium transients. The subinhibitory concentration of IC87201 enabled evaluation of the impact of IC87201 on agmatine inhibition of NMDAr-mediated calcium transients without a confounding effect. We also tested the 30 μM concentration of IC87201 for an impact on agmatine inhibition of NMDAr-mediated calcium transients with the knowledge that it also inhibited NMDAr-mediated calcium transients (Student’s t test, ∗P < .05). Subinhibitory (10 μM) or inhibitory (30 μM) concentrations of IC87201 and approximate IC50 of agmatine (3.3 mM) were bath-perfused simultaneously to ex vivo spinal cord slices. NMDA-induced calcium transients in the spinal cord dorsal horn were quantified before and following coincubation with IC87201 and agmatine. The attenuation of NMDA responses by agmatine was significantly reversed in the 10 μM (one-way ANOVA, ∗P < .05) and by the 30 μM concentration of IC87201 (one-way ANOVA, ∗∗P < .01) (Fig. 4B). These results indicate that an intact NMDAr-PSD95-nNOS pathway is required for the inhibitory effect of agmatine on NMDAr-induced calcium transients.

Fig. 4.

Fig. 4

Open in a new tab

PSD95-nNOS inhibitor, IC87201, reversed agmatine’s inhibition of NMDA-evoked Ca2+ transients. (A) 30 μM (magenta) but not 10 μM (pink) IC87201 significantly inhibited NMDA-evoked Ca2+ transients, Student’s t test; ∗P < .05. (B) 10 (pink squares, pink bar) and 30 μM (purple downward triangles, purple bar) IC87201 significantly reversed the inhibition by 3.3 mM agmatine (white circles, gray bar). Symbols represent single cells values, and lines and bars represent mean ± SEM. Each symbol is represented by at least 17 cells per slice; 5 slices per animal; 3 animals per concentration. No sex difference was observed. One-way ANOVA; ∗P < .05; ∗∗P < .01. ns means not significant (P > .05).

Notably, the NMDA-induced nociceptive behavior is attenuated by NMDAr antagonists, but not by NOS inhibitors; therefore, the nociceptive behavior appears to be NO-independent (Roberts et al, 2005). In contrast, the accompanying thermal hyperalgesia is inhibited by NOS inhibitors (Roberts et al, 2005) as well as by intrathecal hemoglobin (Kitto et al, 1992) and thus requires NO signaling. The reversal of agmatine’s inhibition by a subinhibitory concentration of IC87201 suggests that agmatine’s attenuation of NMDA-evoked calcium responses is mediated, at least in part, by agmatine’s inhibition of nNOS.

3.4. Role of PSD95-nNOS protein–protein interactions in inhibition of NMDA-evoked thermal hyperalgesia

We have reported that intrathecally delivered agmatine inhibits NMDA-induced thermal hyperalgesia (Fairbanks et al, 2000; Roberts et al, 2005), consistent with agmatine’s inhibitory action on nNOS (Galea et al,1996; Demady et al, 2001; present study). We used an in vivo model of short-term NMDA-induced neurobehavioral plasticity whereby intrathecally delivered NMDA evokes 2 distinct responses in the same mouse (Aanonsen and Wilcox, 1987). Following intrathecal delivery of NMDA (0.3 nmol), mice scratch and bite at their hindquarters for 1 minute, after which the behavior ceases. Following cessation of the scratching and biting responses, mice develop thermal hyperalgesia of the tail. The initial nociceptive response is inhibited by MK-801 and other NMDAr antagonists but not by NOS inhibitors (eg, L-NAME or 7-NI) (Fairbanks et al, 2000; Roberts et al, 2005), and the second thermal hyperalgesia response is inhibited by both NMDAr antagonists and NOS inhibitors (Kitto et al, 1992). Therefore, the first response is dependent on NMDAr activation (but not NOS), and the second response is dependent on activation of both NMDAr and NOS. We have noted previously that agmatine inhibits both responses but with different potencies (NMDA nociceptive behaviors [53 nmol (25–112)] and thermal hyperalgesia [0.45 nmol (0.089–2.2)]), suggesting independent action at each (Fairbanks et al, 2000; Roberts et al, 2005). By selecting the lower dose of agmatine, which reduces thermal hyperalgesia but does not inhibit nociceptive behaviors, we may selectively engage the aspect of agmatine’s inhibition on NMDA-induced thermal hyperalgesia that is associated with NOS. We applied the same strategy as in Fig. 4A by using IC87201 to disrupt the PSD95-nNOS interaction. Coadministering a subanalgesic dose (0.001 pmol) of IC82701 (Florio et al, 2009) intrathecally with agmatine significantly reversed agmatine’s inhibition on NMDA-induced thermal hyperalgesia in both males and females. We observed that the inhibition of tail flick hyperalgesia by intrathecal delivery of 2 nmol agmatine in female mice was significantly reversed in the 0.001 pmol IC87201 group (one-way ANOVA, n = 6, ∗P < 0.05) (Fig. 5). In male mice, we intrathecally administered a 1 nmol dose of agmatine to achieve a similar extent of inhibitory effect on thermal hyperalgesia in females (Fig. 5). The inhibition of tail flick hyperalgesia by intrathecal agmatine (1 nmol) was also significantly reversed in the 0.001 pmol IC87201 group (one-way ANOVA, n = 6, ∗∗∗∗P < .0001).

Fig. 5.

Fig. 5

Open in a new tab

IC87201 reversed agmatine’s inhibition of NMDA-induced thermal hyperalgesia. 0.001 pmol intrathecal IC87201 significantly reversed the inhibition of NMDA-mediated thermal hyperalgesia by agmatine in female (left panel) and male (right panel) mice. Images corresponding to pre-drug application as well as traces 1, 3, and 3 are shown on the right. Symbols represent the individual animals’ values, and lines and bars represent mean ± SEM, n = 6 per group. One-way ANOVA, ∗P < .05; ∗∗∗∗P < .0001.

4. Discussion

The mechanisms underlying the development and maintenance of both chronic pain and opioid analgesic tolerance include maladaptive neuroplasticity arising in the dorsal horn of the spinal cord. Opioid analgesic tolerance and neuropathic pain are both dependent on NMDAr activation (Mayer et al, 1995). Long-term potentiation (LTP, a persistent increase in synaptic strength) in the spinal cord is known to drive the spinal neuroplasticity attributed to neurological dysfunction associated with maladaptive signaling in the dorsal horn of the spinal cord. As in classic hippocampal neuroplasticity, spinal LTP requires NMDAr activation (Liu and Sandkuhler, 1995) and NO retrograde signaling (Fenselau et al, 2011). Complementarily, intrathecal delivery of the commonly used NMDAr antagonist MK-801 and nNOS inhibitor 7-NI reduces the development of spinal opioid analgesic tolerance (Fairbanks and Wilcox, 1997) and chronic pain (Guan et al, 2007; Peterson et al, 2021).

We have previously observed that intrathecally delivered agmatine, a decarboxylated form of the amino acid L-arginine, inhibited the development of acute morphine (Fairbanks and Wilcox, 1997) and endomorphin-2 (Wade et al, 2009) analgesic tolerance as well as inflammatory (Fairbanks et al, 2000) and peripheral nerve injury-induced pain (Peterson et al, 2021). These studies used preclinical models of neurological disorders known to involve NMDAr-dependent neuroplasticity. Our past study also reported that bath-applied agmatine concentration-dependently reduced high-frequency stimulation-induced increases in the amplitude of the spinal cord dorsum potential in a rat model of spinal LTP (Peterson et al, 2023). In addition, we previously observed that intrathecal delivery of an AAV carrying the gene for arginine decarboxylase (agmatine-synthesizing enzyme) reversed nerve injury-induced hypersensitivity in mice (Peterson et al, 2023). These results suggest that agmatine application can reverse or prevent the development of spinal LTP, hence hindering the maladaptive spinal plasticity leading to pain.

In the present study, by bath-applying NMDAr agonists and antagonists to ex vivo spinal cord slices to modulate calcium responses, we showed that NMDA-evoked Ca2+ responses were concentration-dependently reduced by subsequent application of agmatine, supporting that agmatine is an established NMDAr antagonist in spinal cord dorsal horn profiles. We previously observed that agmatine’s inhibition of NMDA-evoked scratching and biting behavior and neuropathic pain required the GluN2B subunit of the NMDAr (Fairbanks et al, 2000; Peterson et al, 2021). In a spinal cord slice preparation, we similarly observed that agmatine reduced the amplitude, duration, and decay constant of NMDAr-mediated EPSCs in a manner comparable to the GluN2B subunit-selective NMDAr antagonist ifenprodil and distinct from the GluN2A subunit-selective NMDAr antagonist PEAQX. Moreover, the effects of both agmatine and ifenprodil (but not PEAQX) on NMDAr-mediated EPSCs in mice with conditional knockdown of GluN2B were attenuated (Waataja et al, 2019). To examine whether agmatine’s inhibition of NMDAr-mediated calcium transients also involves the GluN2B subunit of the NMDAr, we used a conditional knockdown model of the GluN2B-containing NMDAr in the spinal cord dorsal horn. Previously, we showed significant mRNA and protein level knockdown of the GluN2B containing NMDAr following intrathecal injection of Cre-containing AAV9 in a GluN2B floxed mouse (Waataja et al, 2019; Peterson et al, 2021). In this study, we used intraspinal injection of AAV9, which restricts the delivery of Cre-containing AAV9 to a more limited number of the dorsal horn spinal cord neurons. Past literature indicates that decreased sensitivity to ifenprodil reflects a decrease in GluN2B-containing NMDArs (Williams et al, 1993; Barth and Malenka, 2001). Consistent with those findings, we showed that ifenprodil’s inhibition is significantly attenuated in GluN2B-KD animals compared with controls. Therefore, we validated the knockdown model using the well-known GluN2B subunit-selective NMDAr antagonist ifenprodil. As expected, ifenprodil reduced NMDAr-mediated calcium transients in the spinal cords of control mice injected with AAV9-hSyn-GCaMP6s-P2A-ΔCre, but not in GluN2B-KD mice (injected with AAV9-hSyn-GCaMP6s-P2A-Cre). Unexpectedly, agmatine attenuated NMDA-evoked calcium transients with equal effectiveness in spinal cords from both control and GluN2B-KD mice. The pharmacological effect of agmatine in the calcium transient assay was not reversed in GluN2B-KD animals, an outcome that differs from our electrophysiological (Waataja et al, 2019) and in vivo pharmacological studies (Peterson et al, 2021). These data suggest that agmatine’s inhibition of NMDA-mediated calcium transients does not require the GluN2B subunit of the NMDAr. The difference between our prior studies compared with the present calcium imaging study may be attributable to agmatine’s multiple pharmacological effects and not limited to inhibition of the GluN2B subunit-containing NMDArs. Glun2A and GluN2B subunits of NMDArs differentially contribute to the NMDA-evoked calcium transients in neurons. A previous study showed that both GluN2A-selective antagonist NVP-AAM077 and GluN2B-selective antagonist ifenprodil decreased the ratio of increased intracellular calcium in live neurons to a different extent (Zhou et al, 2013). In a previous electrophysiological study, it was demonstrated that the GluN2B subunit contributes to synaptic NMDAr responses at lamina I synapses in spinal cord dorsal horn to a greater extent than does the GluN2A subunit (Hildebrand et al, 2014). Both GluN2B and GluN2A contribute to NMDA spinal responses; it is possible that GluN2A subunit-containing NMDArs are involved in agmatine-mediated inhibition of NMDA-evoked calcium transients.

Agmatine has also been shown to inhibit multiple isoforms of NOS (Galea et al, 1996; Gilad et al, 1996; Demady et al, 2001), an outcome which we have also confirmed with a contemporary study. Other studies showed that inhibition of nNOS and iNOS via selective NOS antagonists (Malmberg and Yaksh, 1993), or genetic knockout of NOS (Boettger et al, 2007), can attenuate inflammatory and neuropathic pain. Based on the inhibitory action of agmatine on NOS, we speculated that agmatine’s action in the NMDAr-mediated calcium transients could be associated with its action at NOS. Of different NOS isoforms, we focused on nNOS. nNOS is a calcium-dependent NOS, which is mainly regulated by NMDAr activation and intracellular calcium concentration (Zhu et al, 2023). Agmatine is also an inhibitor of nNOS (Gilad et al, 1996; Demady et al, 2001; Aricioglu et al, 2004; Kim et al, 2004), which participates in the generation of neuroplasticity in the spinal cord. Studies have shown that PSD95-nNOS plays a crucial role in the development of neuropathic, inflammatory, and cancer pain (Florio et al, 2009; D'Mello et al, 2011; Lee et al, 2015; Wei et al, 2021). We previously compared the impact of intrathecal delivery of MK-801, agmatine, and 7-NI on intrathecal NMDA-induced thermal hyperalgesia and nociceptive behaviors in mice (Fairbanks et al, 2000; Roberts et al, 2005). The responses to the 3 ligands are quite different. The nNOS inhibitor 7-NI dose-dependently inhibits the NMDA-induced thermal hyperalgesia but has no effect on NMDA-induced nociceptive behaviors. In contrast, NMDAr channel blocker MK-801 showed dose-dependent inhibition of both NMDA-induced thermal hyperalgesia and nociceptive behaviors with equivalent potency. Unlike MK-801, agmatine inhibits thermal hyperalgesia at lower doses than the dose range that it inhibits nociceptive behaviors. We speculate that the difference in potency at which agmatine inhibits the 2 distinct behaviors is consistent with agmatine’s actions as an NMDAr antagonist (Reynolds, 1990) and also an nNOS inhibitor (Galea et al, 1996).

To test whether agmatine’s inhibition of NMDAr-mediated calcium transients involves signaling downstream of the NMDAr, we developed a strategy to disrupt the association of nNOS with NMDAr without directly inhibiting the enzyme. To use an nNOS inhibitor or a knockout mouse would preclude assessment of the action of agmatine since it is, itself, an nNOS inhibitor. Rather, we used an inhibitor (IC87201) that disrupts the PSD95-nNOS interaction. The PSD95 domain allows close positioning of nNOS to the NMDAr and subsequent activation by CaMKII. The small molecule inhibitor IC87201 disrupts the tethering between PSD95 and nNOS, as previously demonstrated by maintaining PSD95 binding sites and reduction in cGMP production (Florio et al, 2009). We speculated that the reduction of PSD95-nNOS tethering would reduce the effectiveness of nNOS inhibition by agmatine; through this approach, we evaluated whether agmatine’s attenuation on NMDA-evoked calcium responses is dependent on an NMDAr-PSD95-nNOS pathway. The result showed that subinhibitory concentrations of IC87201 significantly reversed the attenuation of calcium transients by agmatine in both male and female ex vivo spinal cord dorsal horn, showing that the intact PSD95-nNOS tethering is required for agmatine-mediated inhibition of calcium transients. This suggests that agmatine mediates the effect via an intact NMDAr-PSD95-nNOS pathway.

To determine whether agmatine’s inhibitory effects on NMDA-stimulated thermal hyperalgesia are partially or significantly mediated by nNOS inhibition, we used an ineffective concentration of the PSD95-nNOS tethering inhibitor IC87201 to disrupt NMDAr-PSD95-nNOS signaling while applying agmatine. Similar to our ex vivo calcium imaging study, we selected the lower dose of agmatine that moderately reduces thermal hyperalgesia but does not inhibit nociceptive behaviors. This specific dose enables the assessment of agmatine’s inhibition of NMDA-induced thermal hyperalgesia, which is known to be nNOS-dependent. Coadministering a subanalgesic dose of IC82701 intrathecally with agmatine showed that NMDA-induced thermal hyperalgesia was significantly reversed in both males and females. The result from in vivo pharmacology is congruent with the calcium imaging study, where IC87201 also significantly reversed agmatine’s inhibition on NMDAr-mediated calcium transients. IC87201 inhibits the protein–protein interaction between PSD95 and nNOS, shown in an in vitro binding assay (Florio et al, 2009). IC87201 itself did not inhibit nNOS function at 40 μM in primary neuronal culture (Florio et al, 2009). The function of IC87201 is likely to limit agmatine’s access to NMDAr-PSD95-tethered nNOS, as studies suggested that the NMDAr-PSD95-nNOS complex allows efficient downstream nNOS activation and NO signaling (Kiedrowski et al, 1992).

We observed that agmatine concentration-dependently inhibited NMDA-evoked calcium responses in spinal cord dorsal horn in a manner that requires an intact PSD95-nNOS protein–protein interactions. We similarly observed that an intact PSD95-nNOS protein–protein interaction is required for agmatine’s inhibition of NMDA-induced thermal hyperalgesia. This correspondence suggests that the in vivo calcium imaging assay may offer a robust environment in which to characterize agmatine-mediated synaptic dynamics. Through complementary calcium imaging and in vivo pharmacology experiments, the present study substantially extends our understanding of agmatine’s inhibitory mechanisms in spinal cord dorsal horn.

Conflict of interest

No author has an actual or perceived conflict of interest with the contents of this article.

Acknowledgments

The authors would like to thank Dr Ezequiel Marron Fernandez de Velasco for providing AAV9 virus, generated by the University of Minnesota Viral Vector and Cloning Core (Minneapolis, Minnesota), University of Minnesota University Imaging Centers, for providing resources and assistance from staff members and Ms H. Oanh Nguyen for editing the manuscript.

Financial support

The funding for these studies was provided by National Institutes of Health National Institute on Drug Abuse [Grant DA035931] (to C.A.F.), Department of Defense-PRMRP [Grant W81XWH-19-1-0673] (to C.A.F.), and Department of Defense-CMDRP [Grant W81XWH-20-1-0509] (to L.V.) and support was received from the College of Pharmacy of the University of Minnesota.

Data availability

The data that support the findings of this study are available upon request from the corresponding author.

Authorship contributions

Participated in research design: Xie, Schorn, Peterson, Wilcox, Vulchanova, Fairbanks.

Conducted experiments: Xie, Schorn, Kitto.

Performed data analysis: Xie, Kitto.

Wrote or contributed to the writing of the manuscript: Xie, Schorn, Peterson, Florio, Wilcox, Vulchanova, Fairbanks.

Footnotes

This article has supplemental material available at jpet.aspetjournals.org.

Supplemental material

Supplemental Table 1
mmc1.pdf (82KB, pdf)

References

  1. Aanonsen L.M., Wilcox G.L. Nociceptive action of excitatory amino acids in the mouse: effects of spinally administered opioids, phencyclidine and sigma agonists. J Pharmacol Exp Ther. 1987;243:9–19. [PubMed] [Google Scholar]
  2. Anderson G.R., Maxeiner S., Sando R., Tsetsenis T., Malenka R.C., Sudhof T.C. Postsynaptic adhesion GPCR latrophilin-2 mediates target recognition in entorhinal-hippocampal synapse assembly. J Cell Biol. 2017;216:3831–3846. doi: 10.1083/jcb.201703042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Aricioglu F., Paul I.A., Regunathan S. Agmatine reduces only peripheral-related behavioral signs, not the central signs, of morphine withdrawal in nNOS deficient transgenic mice. Neurosci Lett. 2004;354:153–157. doi: 10.1016/j.neulet.2003.10.010. [DOI] [PubMed] [Google Scholar]
  4. Barth A.L., Malenka R.C. NMDAR EPSC kinetics do not regulate the critical period for LTP at thalamocortical synapses. Nat Neurosci. 2001;4:235–236. doi: 10.1038/85070. [DOI] [PubMed] [Google Scholar]
  5. Boettger M.K., Uceyler N., Zelenka M., Schmitt A., Reif A., Chen Y., Sommer C. Differences in inflammatory pain in nNOS-, iNOS- and eNOS-deficient mice. Eur J Pain. 2007;11:810–818. doi: 10.1016/j.ejpain.2006.12.008. [DOI] [PubMed] [Google Scholar]
  6. Brenman J.E., Chao D.S., Gee S.H., McGee A.W., Craven S.E., Santillano D.R., Wu Z., Huang F., Xia H., Peters M.F., et al. Interaction of nitric oxide synthase with the postsynaptic density protein PSD-95 and alpha1-syntrophin mediated by PDZ domains. Cell. 1996;84:757–767. doi: 10.1016/s0092-8674(00)81053-3. [DOI] [PubMed] [Google Scholar]
  7. Brigman J.L., Wright T., Talani G., Prasad-Mulcare S., Jinde S., Seabold G.K., Mathur P., Davis M.I., Bock R., Gustin R.M., et al. Loss of GluN2B-containing NMDA receptors in CA1 hippocampus and cortex impairs long-term depression, reduces dendritic spine density, and disrupts learning. J Neurosci. 2010;30:4590–4600. doi: 10.1523/JNEUROSCI.0640-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Courtney M.J., Li L.L., Lai Y.Y. Mechanisms of NOS1AP action on NMDA receptor-nNOS signaling. Front Cell Neurosci. 2014;8:252. doi: 10.3389/fncel.2014.00252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. D'Mello R., Marchand F., Pezet S., McMahon S.B., Dickenson A.H. Perturbing PSD-95 interactions with NR2B-subtype receptors attenuates spinal nociceptive plasticity and neuropathic pain. Mol Ther. 2011;19:1780–1792. doi: 10.1038/mt.2011.42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Davies J., Francis A.A., Jones A.W., Watkins J.C. 2-Amino-5-phosphonovalerate (2APV), a potent and selective antagonist of amino acid-induced and synaptic excitation. Neurosci Lett. 1981;21:77–81. doi: 10.1016/0304-3940(81)90061-6. [DOI] [PubMed] [Google Scholar]
  11. Demady D.R., Jianmongkol S., Vuletich J.L., Bender A.T., Osawa Y. Agmatine enhances the NADPH oxidase activity of neuronal NO synthase and leads to oxidative inactivation of the enzyme. Mol Pharmacol. 2001;59:24–29. doi: 10.1124/mol.59.1.24. [DOI] [PubMed] [Google Scholar]
  12. Dickenson A.H., Sullivan A.F. Evidence for a role of the NMDA receptor in the frequency dependent potentiation of deep rat dorsal horn nociceptive neurones following C fibre stimulation. Neuropharmacology. 1987;26:1235–1238. doi: 10.1016/0028-3908(87)90275-9. [DOI] [PubMed] [Google Scholar]
  13. Donevan S.D., Rogawski M.A. GYKI 52466, a 2,3-benzodiazepine, is a highly selective, noncompetitive antagonist of AMPA/kainate receptor responses. Neuron. 1993;10:51–59. doi: 10.1016/0896-6273(93)90241-i. [DOI] [PubMed] [Google Scholar]
  14. Doolen S., Blake C.B., Smith B.N., Taylor B.K. Peripheral nerve injury increases glutamate-evoked calcium mobilization in adult spinal cord neurons. Mol Pain. 2012;8:56. doi: 10.1186/1744-8069-8-56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Evans R.H., Francis A.A., Jones A.W., Smith D.A., Watkins J.C. The effects of a series of omega-phosphonic alpha-carboxylic amino acids on electrically evoked and excitant amino acid-induced responses in isolated spinal cord preparations. Br J Pharmacol. 1982;75:65–75. doi: 10.1111/j.1476-5381.1982.tb08758.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fairbanks C.A. Spinal delivery of analgesics in experimental models of pain and analgesia. Adv Drug Deliv Rev. 2003;55:1007–1041. doi: 10.1016/s0169-409x(03)00101-7. [DOI] [PubMed] [Google Scholar]
  17. Fairbanks C.A., Schreiber K.L., Brewer K.L., Yu C.G., Stone L.S., Kitto K.F., Nguyen H.O., Grocholski B.M., Shoeman D.W., Kehl L.J., et al. Agmatine reverses pain induced by inflammation, neuropathy, and spinal cord injury. Proc Natl Acad Sci U S A. 2000;97:10584–10589. doi: 10.1073/pnas.97.19.10584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fairbanks C.A., Wilcox G.L. Acute tolerance to spinally administered morphine compares mechanistically with chronically induced morphine tolerance. J Pharmacol Exp Ther. 1997;282:1408–1417. [PubMed] [Google Scholar]
  19. Fenselau H., Heinke B., Sandkuhler J. Heterosynaptic long-term potentiation at GABAergic synapses of spinal lamina I neurons. J Neurosci. 2011;31:17383–17391. doi: 10.1523/JNEUROSCI.3076-11.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Florio S.K., Loh C., Huang S.M., Iwamaye A.E., Kitto K.F., Fowler K.W., Treiberg J.A., Hayflick J.S., Walker J.M., Fairbanks C.A., et al. Disruption of nNOS-PSD95 protein-protein interaction inhibits acute thermal hyperalgesia and chronic mechanical allodynia in rodents. Br J Pharmacol. 2009;158:494–506. doi: 10.1111/j.1476-5381.2009.00300.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Galea E., Regunathan S., Eliopoulos V., Feinstein D.L., Reis D.J. Inhibition of mammalian nitric oxide synthases by agmatine, an endogenous polyamine formed by decarboxylation of arginine. Biochem J. 1996;316(Pt 1):247–249. doi: 10.1042/bj3160247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Garthwaite J. Cellular uptake disguises action of L-glutamate on N-methyl-D-aspartate receptors. With an appendix: diffusion of transported amino acids into brain slices. Br J Pharmacol. 1985;85:297–307. doi: 10.1111/j.1476-5381.1985.tb08860.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gibson D.A., Harris B.R., Rogers D.T., Littleton J.M. Radioligand binding studies reveal agmatine is a more selective antagonist for a polyamine-site on the NMDA receptor than arcaine or ifenprodil. Brain Res. 2002;952:71–77. doi: 10.1016/s0006-8993(02)03198-0. [DOI] [PubMed] [Google Scholar]
  24. Gilad G.M., Salame K., Rabey J.M., Gilad V.H. Agmatine treatment is neuroprotective in rodent brain injury models. Life Sci. 1996;58:PL 41–PL 46. doi: 10.1016/0024-3205(95)02274-0. [DOI] [PubMed] [Google Scholar]
  25. Guan Y., Yaster M., Raja S.N., Tao Y.X. Genetic knockout and pharmacologic inhibition of neuronal nitric oxide synthase attenuate nerve injury-induced mechanical hypersensitivity in mice. Mol Pain. 2007;3:29. doi: 10.1186/1744-8069-3-29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Heinke B., Balzer E., Sandkuhler J. Pre- and postsynaptic contributions of voltage-dependent Ca2+ channels to nociceptive transmission in rat spinal lamina I neurons. Eur J Neurosci. 2004;19:103–111. doi: 10.1046/j.1460-9568.2003.03083.x. [DOI] [PubMed] [Google Scholar]
  27. Hildebrand M.E., Pitcher G.M., Harding E.K., Li H., Beggs S., Salter M.W. GluN2B and GluN2D NMDARs dominate synaptic responses in the adult spinal cord. Sci Rep. 2014;4:1–12. doi: 10.1038/srep04094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Hylden J.L., Wilcox G.L. Intrathecal morphine in mice: a new technique. Eur J Pharmacol. 1980;67:313–316. doi: 10.1016/0014-2999(80)90515-4. [DOI] [PubMed] [Google Scholar]
  29. Ikeda H., Heinke B., Ruscheweyh R., Sandkuhler J. Synaptic plasticity in spinal lamina I projection neurons that mediate hyperalgesia. Science. 2003;299:1237–1240. doi: 10.1126/science.1080659. [DOI] [PubMed] [Google Scholar]
  30. Infante C., Diaz M., Hernandez A., Constandil L., Pelissier T. Expression of nitric oxide synthase isoforms in the dorsal horn of monoarthritic rats: effects of competitive and uncompetitive N-methyl-D-aspartate antagonists. Arthritis Res Ther. 2007;9:R53. doi: 10.1186/ar2208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Iseppon F., Linley J.E., Wood J.N. Calcium imaging for analgesic drug discovery. Neurobiol Pain. 2022;11 doi: 10.1016/j.ynpai.2021.100083. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ishii H., Shibuya K., Ohta Y., Mukai H., Uchino S., Takata N., Rose J.A., Kawato S. Enhancement of nitric oxide production by association of nitric oxide synthase with N-methyl-D-aspartate receptors via postsynaptic density 95 in genetically engineered Chinese hamster ovary cells: real-time fluorescence imaging using nitric oxide sensitive dye. J Neurochem. 2006;96:1531–1539. doi: 10.1111/j.1471-4159.2006.03656.x. [DOI] [PubMed] [Google Scholar]
  33. Kiedrowski L., Costa E., Wroblewski J.T. Glutamate receptor agonists stimulate nitric oxide synthase in primary cultures of cerebellar granule cells. J Neurochem. 1992;58:335–341. doi: 10.1111/j.1471-4159.1992.tb09315.x. [DOI] [PubMed] [Google Scholar]
  34. Kim J.H., Yenari M.A., Giffard R.G., Cho S.W., Park K.A., Lee J.E. Agmatine reduces infarct area in a mouse model of transient focal cerebral ischemia and protects cultured neurons from ischemia-like injury. Exp Neurol. 2004;189:122–130. doi: 10.1016/j.expneurol.2004.05.029. [DOI] [PubMed] [Google Scholar]
  35. Kitto K.F., Haley J.E., Wilcox G.L. Involvement of nitric oxide in spinally mediated hyperalgesia in the mouse. Neurosci Lett. 1992;148:1–5. doi: 10.1016/0304-3940(92)90790-e. [DOI] [PubMed] [Google Scholar]
  36. Lee W.H., Xu Z., Ashpole N.M., Hudmon A., Kulkarni P.M., Thakur G.A., Lai Y.Y., Hohmann A.G. Small molecule inhibitors of PSD95-nNOS protein-protein interactions as novel analgesics. Neuropharmacology. 2015;97:464–475. doi: 10.1016/j.neuropharm.2015.05.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Li D.P., Zhou J.J., Zhang J., Pan H.L. CaMKII regulates synaptic NMDA receptor activity of hypothalamic presympathetic neurons and sympathetic outflow in hypertension. J Neurosci. 2017;37:10690–10699. doi: 10.1523/JNEUROSCI.2141-17.2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Liu X.G., Sandkuhler J. Long-term potentiation of C-fiber-evoked potentials in the rat spinal dorsal horn is prevented by spinal N-methyl-D-aspartic acid receptor blockage. Neurosci Lett. 1995;191:43–46. doi: 10.1016/0304-3940(95)11553-0. [DOI] [PubMed] [Google Scholar]
  39. Luo C., Seeburg P.H., Sprengel R., Kuner R. Activity-dependent potentiation of calcium signals in spinal sensory networks in inflammatory pain states. Pain. 2008;140:358–367. doi: 10.1016/j.pain.2008.09.008. [DOI] [PubMed] [Google Scholar]
  40. MacDermott A.B., Mayer M.L., Westbrook G.L., Smith S.J., Barker J.L. NMDA-receptor activation increases cytoplasmic calcium concentration in cultured spinal cord neurones. Nature. 1986;321:519–522. doi: 10.1038/321519a0. [DOI] [PubMed] [Google Scholar]
  41. Malmberg A.B., Yaksh T.L. Spinal nitric oxide synthesis inhibition blocks NMDA-induced thermal hyperalgesia and produces antinociception in the formalin test in rats. Pain. 1993;54:291–300. doi: 10.1016/0304-3959(93)90028-N. [DOI] [PubMed] [Google Scholar]
  42. Mayer D.J., Mao J., Price D.D. The association of neuropathic pain, morphine tolerance and dependence, and the translocation of protein kinase C. NIDA Res Monogr. 1995;147 269–268. [PubMed] [Google Scholar]
  43. Meller S.T., Pechman P.S., Gebhart G.F., Maves T.J. Nitric oxide mediates the thermal hyperalgesia produced in a model of neuropathic pain in the rat. Neuroscience. 1992;50:7–10. doi: 10.1016/0306-4522(92)90377-e. [DOI] [PubMed] [Google Scholar]
  44. Molderings G.J., Haenisch B. Agmatine (decarboxylated L-arginine): physiological role and therapeutic potential. Pharmacol Ther. 2012;133:351–365. doi: 10.1016/j.pharmthera.2011.12.005. [DOI] [PubMed] [Google Scholar]
  45. Niesters M., Martini C., Dahan A. Ketamine for chronic pain: risks and benefits. Br J Clin Pharmacol. 2014;77:357–367. doi: 10.1111/bcp.12094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Olmos G., DeGregorio-Rocasolano N., Paz Regalado M., Gasull T., Assumpcio Boronat M., Trullas R., Villarroel A., Lerma J., Garcia-Sevilla J.A. Protection by imidazol(ine) drugs and agmatine of glutamate-induced neurotoxicity in cultured cerebellar granule cells through blockade of NMDA receptor. Br J Pharmacol. 1999;127:1317–1326. doi: 10.1038/sj.bjp.0702679. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Peterson C.D., Kitto K.F., Verma H., Pflepsen K., Delpire E., Wilcox G.L., Fairbanks C.A. Agmatine requires GluN2B-containing NMDA receptors to inhibit the development of neuropathic pain. Mol Pain. 2021;17 doi: 10.1177/17448069211029171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Peterson C.D., Waataja J.J., Kitto K.F., Erb S.J., Verma H., Schuster D.J., Churchill C.C., Riedl M.S., Belur L.R., Wolf D.A., et al. Long-term reversal of chronic pain behavior in rodents through elevation of spinal agmatine. Mol Ther. 2023;31:1123–1135. doi: 10.1016/j.ymthe.2023.01.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Piletz J.E., Aricioglu F., Cheng J.-T., Fairbanks C.A., Gilad V.H., Haenisch B., Halaris A., Hong S., Lee J.E., Li J., et al. Agmatine: clinical applications after 100 years in translation. Drug Discov Today. 2013;18:880–893. doi: 10.1016/j.drudis.2013.05.017. [DOI] [PubMed] [Google Scholar]
  50. Pollock J.S., Forstermann U., Mitchell J.A., Warner T.D., Schmidt H.H., Nakane M., Murad F. Purification and characterization of particulate endothelium-derived relaxing factor synthase from cultured and native bovine aortic endothelial cells. Proc Natl Acad Sci U S A. 1991;88:10480–10484. doi: 10.1073/pnas.88.23.10480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Reynolds I.J. Arcaine uncovers dual interactions of polyamines with the N-methyl-D-aspartate receptor. J Pharmacol Exp Ther. 1990;255:1001–1007. [PubMed] [Google Scholar]
  52. Roberts J.C., Grocholski B.M., Kitto K.F., Fairbanks C.A. Pharmacodynamic and pharmacokinetic studies of agmatine after spinal administration in the mouse. J Pharmacol Exp Ther. 2005;314:1226–1233. doi: 10.1124/jpet.105.086173. [DOI] [PubMed] [Google Scholar]
  53. Schindelin J., Arganda-Carreras I., Frise E., Kaynig V., Longair M., Pietzsch T., Preibisch S., Rueden C., Saalfeld S., Schmid B., et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9:676–682. doi: 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Shirwany N.A., Xie J., Guo Q. Regulation of intracellular calcium in cortical neurons transgenic for human Abeta40 and Abeta42 following nutritive challenge. Int J Clin Exp Med. 2009;2:149–158. [PMC free article] [PubMed] [Google Scholar]
  55. Skorput A.G.J., Zhang X., Waataja J.J., Peterson C.D., Riedl M.S., Kitto K.F., Truong H., Huffman C., Salton S.R., Fairbanks C.A., et al. Involvement of the VGF-derived peptide TLQP-62 in nerve injury-induced hypersensitivity and spinal neuroplasticity. Pain. 2018;159:1802–1813. doi: 10.1097/j.pain.0000000000001277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Strauss K.I., Marini A.M. Cyclooxygenase-2 inhibition protects cultured cerebellar granule neurons from glutamate-mediated cell death. J Neurotrauma. 2002;19:627–638. doi: 10.1089/089771502753754091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Svendsen F., Tjolsen A., Hole K. AMPA and NMDA receptor-dependent spinal LTP after nociceptive tetanic stimulation. Neuroreport. 1998;9:1185–1190. doi: 10.1097/00001756-199804200-00041. [DOI] [PubMed] [Google Scholar]
  58. Tayeh M.A., Marletta M.A. Macrophage oxidation of L-arginine to nitric oxide, nitrite, and nitrate. J Biol Chem. 1989;264:19654–19658. [PubMed] [Google Scholar]
  59. Thomas D.A., Ren K., Besse D., Ruda M.A., Dubner R. Application of nitric oxide synthase inhibitor, N omega-nitro-L-arginine methyl ester, on injured nerve attenuates neuropathy-induced thermal hyperalgesia in rats. Neurosci Lett. 1996;210:124–126. doi: 10.1016/0304-3940(96)12670-7. [DOI] [PubMed] [Google Scholar]
  60. Waataja J.J., Peterson C.D., Verma H., Goracke-Postle C.J., Séguéla P., Delpire E., Wilcox G.L., Fairbanks C.A. Agmatine preferentially antagonizes GluN2B-containing N-methyl-d-aspartate receptors in spinal cord. J Neurophysiol. 2019;121:662–671. doi: 10.1152/jn.00172.2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Wade C.L., Eskridge L.L., Nguyen H.O., Kitto K.F., Stone L.S., Wilcox G., Fairbanks C.A. Immunoneutralization of agmatine sensitizes mice to micro-opioid receptor tolerance. J Pharmacol Exp Ther. 2009;331:539–546. doi: 10.1124/jpet.109.155424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Wei W., Liu W., Du S., Govindarajalu G., Irungu A., Bekker A., Tao Y.X. A compound mitigates cancer pain and chemotherapy-induced neuropathic pain by dually targeting nNOS-PSD-95 interaction and GABA(A) receptor. Neurotherapeutics. 2021;18:2436–2448. doi: 10.1007/s13311-021-01158-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Williams K., Russell S.L., Shen Y.M., Molinoff P.B. Developmental switch in the expression of NMDA receptors occurs in vivo and in vitro. Neuron. 1993;10:267–278. doi: 10.1016/0896-6273(93)90317-k. [DOI] [PubMed] [Google Scholar]
  64. Wyllie D.J., Cull-Candy S.G. A comparison of non-NMDA receptor channels in type-2 astrocytes and granule cells from rat cerebellum. J Physiol. 1994;475:95–114. doi: 10.1113/jphysiol.1994.sp020052. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Yang X.-C., Reis D.J. Agmatine selectively blocks the N-methyl-aspartate subclass of glutamate receptor channels in rat hippocampal neurons. J Pharmacol Exp Ther. 1999;288:544–549. [PubMed] [Google Scholar]
  66. Yoon Y.W., Sung B., Chung J.M. Nitric oxide mediates behavioral signs of neuropathic pain in an experimental rat model. Neuroreport. 1998;9:367–372. doi: 10.1097/00001756-199802160-00002. [DOI] [PubMed] [Google Scholar]
  67. Zhou X., Ding Q., Chen Z., Yun H., Wang H. Involvement of the GluN2A and GluN2B subunits in synaptic and extrasynaptic N-methyl-D-aspartate receptor function and neuronal excitotoxicity. J Biol Chem. 2013;288:24151–24159. doi: 10.1074/jbc.M113.482000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Zhu L.J., Li F., Zhu D.Y. nNOS and neurological, neuropsychiatric disorders: a 20-Year story. Neurosci Bull. 2023;39:1439–1453. doi: 10.1007/s12264-023-01060-7. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Table 1
mmc1.pdf (82KB, pdf)

Data Availability Statement

The data that support the findings of this study are available upon request from the corresponding author.

Articles from The Journal of Pharmacology and Experimental Therapeutics are provided here courtesy of American Society for Pharmacology and Experimental Therapeutics

RESOURCES