Nitrous Oxide, a Rapid Antidepressant, Has Ketamine-like Effects on Excitatory Transmission in Adult Hippocampus

Abstract

BACKGROUND:

Nitrous oxide (N2O) is a non-competitive inhibitor of NMDA receptors (NMDARs) that appears to have ketamine-like rapid antidepressant effects in patients with treatment resistant major depression. In preclinical studies, ketamine enhances glutamate-mediated synaptic transmission in hippocampus and prefrontal cortex. The present study examined effects of N₂O on glutamate transmission in hippocampus, and compared its effects to ketamine.

METHODS:

Glutamate-mediated synaptic transmission was studied in the CA1 region of hippocampal slices from adult albino rats using standard extracellular recording methods. Effects of N₂O and ketamine at sub-anesthetic concentrations were evaluated by acute administration.

RESULTS:

Akin to 1 μM ketamine, 30% N₂O administered for 15-20 min resulted in persistent enhancement of synaptic responses mediated by both AMPA receptors (AMPARs) and NMDARs. Synaptic enhancement by both N₂O and ketamine was blocked by co-administration of a competitive NMDAR antagonist at saturating concentration, but only ketamine was blocked by an AMPAR antagonist. Synaptic enhancement by both agents involved tropomyosin receptor kinase B (TrkB), mechanistic target of rapamycin (mTOR), and nitric oxide synthase (NOS), with some differences between N₂O and ketamine. N₂O potentiation occluded enhancement by ketamine, and in vivo N₂O exposure occluded further potentiation by both N₂O and ketamine.

CONCLUSIONS:

These results indicate that N₂O has ketamine-like effects on hippocampal synaptic function at a sub-anesthetic, but therapeutically relevant concentration. These two rapid antidepressants have similar, but not identical mechanisms that result in persisting synaptic enhancement, possibly contributing to psychotropic actions.

INTRODUCTION

Major Depressive disorder (MDD) is a serious psychiatric illness causing disability and death (1). While current antidepressants are effective, they have significant limitations including slow onset and often incomplete remission of symptoms. Approximately 30% of patients fail to respond to antidepressant treatments and have “treatment resistant major depression” (TRMD) (2). The emergence of ketamine as a rapid antidepressant (3) with FDA approval of esketamine for TRMD and acute suicidal ideation, has been a major breakthrough (4). Yet, ketamine has limitations, including side effects and short duration of effect, prompting need for agents with similar rapid onset but fewer side effects and longer-lasting benefits (1,5).

While mechanisms underlying antidepressant effects remain incompletely understood, ketamine has several properties that likely contribute. These include partial inhibition of N-methyl-D-aspartate receptors (NMDARs) at sub-anesthetic concentrations (1,5). NMDAR block, likely on interneurons and possibly via receptors expressing GluN2B subunits, results in disinhibition and enhancement of excitatory transmission on principal neurons in prefrontal cortex and hippocampus (6-11). Potentiating effects involve brain-derived neurotrophic factor (BDNF) and tropomyosin receptor kinase B (TrkB) (12-16), along with mechanistic target of rapamycin (mTOR) (17). These kinases trigger downstream effects that persistingly enhance excitatory transmission. Even this simplified description has controversies, including the role of NMDARs (5,18,19).

Insights from ketamine prompt efforts to identify other rapid antidepressants. An example is nitrous oxide (N₂O), an anesthetic used clinically for over 150 years (1). N₂O is a non-competitive NMDAR antagonist acting by mechanisms distinct from ketamine (20,21). Akin to ketamine, N₂O has effects other than NMDAR antagonism, including modulation of the opiate system and other receptors and channels (1,5). Based on success of ketamine as an antidepressant, studies in humans examined N₂O in TRMD. A proof-of-concept study in highly refractory TRMD patients found that one-hour inhalation of 50% N₂O had rapid antidepressant effects (22). Subsequently, it was shown that both 25% and 50% N₂O are effective in TRMD and effects were sustained for at least two weeks (23).

Despite its history as an anesthetic, less is known about N₂O in terms of mechanisms and whether N₂O shares effects of ketamine on synaptic function. In this study, we examined effects of N₂O on excitatory synaptic function in hippocampus and compared effects to ketamine under the hypothesis that the two agents would have overlapping actions.

METHODS AND MATERIALS

Hippocampal slice preparation & physiology

Protocols for animal experiments were approved by the Washington University IACUC. Hippocampal slices were prepared for most experiments from postnatal day (P) 58-63 (adult) Harlan Sprague-Dawley male albino rats (Indianapolis IN) (24). In some experiments, slices from P28-32 (juvenile) rats were used. Male mice were studied to avoid variability from differences in estrus cycle. For slice preparation, rats were anesthetized with isoflurane (4-5 min duration), and dissected hippocampi were pinned on an agar base in artificial cerebrospinal fluid (ACSF) containing (in mM): 124 NaCl, 5 KCl, 2 MgSO₄, 2 CaCl₂, 1.25 NaH₂PO₄, 22 NaHCO₃, 10 glucose, gassed with 95% O₂-5% CO₂ at 4-6°C. The dorsal two-thirds of the hippocampus was cut into 500 μm slices with a rotary slicer and maintained in 30°C ACSF for at least 1 hour before transfer to a submersion-recording chamber perfused with 30°C ACSF at 2 ml/min. We did not include GABA-A receptor inhibitors, which would interfere with network changes from disinhibition that may play a key role in effects of ketamine and nitrous oxide (25).

Extracellular recordings were obtained from the apical dendrites of CA1 to monitor field excitatory postsynaptic potentials (EPSPs), which were evoked once per minute using 0.1 ms constant current pulses to the Schaffer collateral pathway via a bipolar stimulating electrode. Stimulus intensity was half-maximal based on baseline input-output (IO) curves. IO curves were repeated 60 min following drug administration and were the primary measure of synaptic change in comparison to baseline. Tetanus-induced LTP was evoked using a single 100 Hz x 1 sec high frequency stimulus (HFS). For display in figures, responses are shown at 5 min intervals.

Isolated NMDAR EPSPs were recorded in stratum radiatum once per minute in ACSF containing 0.1 mM Mg²⁺and 2.5 mM Ca²⁺ to promote NMDAR activation and 30 μM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) to block AMPARs (26). Under these conditions, field EPSPs are completely inhibited by the NMDAR-antagonist, 2-amino-5-phosphonovalerate (APV) (26,27). We quantified NMDAR EPSPs as the rising slope of the field potential.

Chemicals

Cyclotraxin B and 6,7-dinitroquinoxaline-2,3-dione (DNQX) were purchased from Tocris (Minneapolis MN). Rapamycin was from Cayman Chemical (Ann Arbor MI). Ketamine, CNQX other agents were obtained from Millipore Sigma (St. Louis MO). N₂O with 5.04% carbon dioxide was purchased from Puritan Medical Products (Overland KS) and was bubbled into a reservoir containing ACSF at the time of experiments (25). We measured N₂O in ACSF as a way to monitor our solutions (Supplemental Methods and Results). In some experiments, we administered 30% N₂O in vivo prior to slice preparation. Because of use of different gas conditions and recording paradigms, experimenters were not blind to recording conditions. Experiments were done with 30% nitrous oxide, a maximal concentration for use in ex vivo slices (25).

Experimental Design & Analysis

EPSPs were collected using pClamp software (Molecular Devices, Union City CA), and are expressed as mean ± SEM 60 min following drug exposure normalized to baseline EPSPs (100%). A two-tailed unpaired Student’s t-test was used for most comparisons between groups. When appropriate, paired t-tests were used. For non-normally distributed data, the Mann-Whitney U-test was used for independent samples and the Wilcoxon signed-rank test was used for matched/dependent samples. Statistical comparisons were based on IO curves at baseline and 60 minutes following drug application to determine changes in EPSP slope at 50% of maximum on IO curves, with p < 0.05 considered significant (28). Numbers in the text are the number (N) of animals in a condition. Some analyses in Figures 1 and 2 compared baseline responses to responses 25-60 min after drug. Statistics were performed using SigmaStat (Systat Software, Richmond City, CA). Data in figures display continuous monitoring of responses at low frequency and thus may differ from numerical results described in the text, which are based on IO curve analysis.

Figure 1.

N₂O, like ketamine, potentiates CA1 hippocampal synaptic responses. A. The graph shows the time course of change in EPSPs in response to 1 μM ketamine. White circles are with low frequency synaptic stimulation during ketamine administration while black circles are without stimulation. B. Perfusion of 30% N₂O (white bar) for 15 min resulted in an initial depression of responses followed by persisting enhancement (white circles are with stimulation during drug administration and black circles are without stimulation). C. Perfusion of 30% nitrogen caused transient synaptic depression during perfusion, but had no persistent effect on EPSPs. D. N₂O failed to enhance EPSPs in slices from P30 rats, but showed initial depression during administration. Traces depict EPSPs before (dashed lines) and 60 min after drug exposure (solid lines). Scale bar: 1 mV, 5 ms.

Figure 2.

Ketamine and N₂O enhance NMDAR EPSPs and NMDAR activation during drug perfusion is required for persisting effects on AMPAR-mediated transmission. A. Perfusion of 1 μM ketamine caused a small depression of NMDA EPSPs during drug administration that was followed by persisting enhancement. B. N₂O depressed NMDAR EPSPs during drug administration and this was followed by persisting enhancement after drug washout. C. Administration of the competitive NMDAR antagonist, APV, just prior to and during ketamine inhibited potentiation of AMPAR-mediated EPSPs. D. A saturating concentration of ketamine (100 μM) that blocks NMDARs completely failed to potentiate EPSPs. E. Akin to ketamine, APV blocked persisting enhancement by N₂O. Traces show field EPSPs as in Figure 1. Scale bar: 1 mV, 5 ms.

RESULTS

Previous studies provide strong support for the ability of ketamine to enhance excitatory synaptic transmission in the CA1 region of hippocampal slices from adult rodents (13,29). In initial studies, we compared effects of sub-anesthetic concentrations of ketamine and nitrous oxide on dendritic EPSPs evoked by low frequency stimulation of the Schaffer collateral pathway in slices prepared from adult (~P60) rats. At 1 μM, a concentration readily achieved in humans during treatment for MDD (30), ketamine produced clear enhancement of excitatory synaptic responses using either of two drug application methods (Figure 1A). Based on prior work indicating that synaptic enhancement by ketamine results from effects on spontaneous transmission (13,29), we administered ketamine for 30 min in the absence of Schaffer collateral stimulation. This protocol resulted in EPSPs that were 146.0 ± 13.2% (N = 6) of baseline 60 min following ketamine; p = 0.0136 vs. baseline, Figure 1A, solid circles; Supplemental Figure 1). Similar results were obtained when low frequency stimulation was employed during ketamine exposure (130.6 ± 10.5% 60 minutes following ketamine, N = 6, p = 0.0369 vs. baseline and p = 0.3827 vs. ketamine with no stimulation; Figure 1A, open circles), indicating that activation dependence of ketamine does not influence results under these conditions. We also found that persistent enhancement was observed with a higher sub-saturating concentration of ketamine (20 μM) (124.7 ± 8.3% of baseline, N = 6; p = 0.2023 vs. 1 μM ketamine; not shown). For subsequent studies, we administered ketamine in the absence of evoked transmission during ketamine perfusion.

For studies of N₂O, we administered 30% N₂O for 15-20 min, based on our prior studies in slices from juvenile (P30) rodents (25) and effects observed clinically in humans (23). In adult (P60) slices, we observed an initial acute depression of EPSPs during N₂O perfusion that differed from 1 μM ketamine (Figure 1A,B). This synaptic depression reversed rapidly following N₂O washout and was followed by enhancement of EPSPs that persisted for at least 60 min (143.3 ± 6.4% of baseline 60 min after N₂O, N = 11, p = 0.0130 vs baseline; p = 0.8374 vs. ketamine, Figure 1B). Similar to ketamine, we also administered N₂O in the absence of synaptic stimulation and still observed potentiation (123.4 ± 4.7%, N = 7; p = 0.0074 vs. baseline; p = 0.0594 vs. N₂O with synaptic stimulation, Figure 1B). To control for changes in ACSF gas composition and possible mild hypoxia, we examined 30% nitrogen. While nitrogen depressed transmission acutely, it did not produce lasting synaptic enhancement after washout (100.8 ± 4.9%, N = 5, p = 0.0009 vs. nitrous oxide, Figure 1C). For subsequent studies, we administered N₂O with synaptic stimulation to monitor effects during and after drug.

These initial studies indicate that N₂O produces ketamine-like enhancement of excitatory transmission. Prior studies (27,31), indicate that ketamine produces no persisting enhancement in slices from juvenile rodents. Thus, we also examined N₂O in slices from P30 rats. Unlike P60 but similar to ketamine, we found no persisting effect of N₂O on evoked EPSPs at P30 (92.6 ± 3.6% of baseline 60 min following N₂O, N = 5; Figure 1D), although N₂O transiently depressed EPSPs during perfusion.

Ketamine and N₂O are non-competitive NMDAR antagonists that act by distinct mechanisms (1,5,21). This prompted us to examine effects on isolated NMDAR-mediated EPSPs in adult slices. At 1 μM, ketamine produced a small depression of NMDA EPSPs (~25% decrease at the end of 30 min perfusion). Following drug washout, we observed enhancement of NMDA EPSPs that, like effects on AMPA EPSPs, persisted for at least an hour (138.3 ± 11.2% of baseline, p = 0.0398 vs. baseline, N = 8, Figure 2A; Supplemental Figure 2). We also observed acute depression of NMDA EPSPs during perfusion with N₂O (~50% depression at the end of 15 min perfusion) followed by persisting enhancement that lasted for at least 60 min after washout (131.0 ± 9.2% of baseline 60 min after N₂O, N = 6, p = 0.0364 vs. baseline; p = 0.4828 vs. ketamine; Figure 2B).

In prior studies, we found that persisting metaplastic effects of ketamine in slices from P30 rats were eliminated by co-administration of a saturating concentration of the competitive NMDAR antagonist, APV (27), indicating that opening of unblocked NMDA channels is important in ketamine’s effects. Similarly, we found that co-administration of 50 μM APV with ketamine eliminated ketamine’s ability to enhance AMPA EPSPs (ketamine alone: 151.7 ± 10.1% of baseline 60 min after ketamine, N = 5 vs. 96.3 ± 10.7% with ketamine + APV, N = 5, p = 0.0055; Figure 2C). While this result strongly suggests that access to NMDA channels is important in synaptic enhancement, it leaves open the question whether activation of unblocked NMDA channels in the presence of low ketamine drives synaptic change. Thus, we also examined whether a saturating concentration of ketamine that completely blocks NMDARs exhibits synaptic enhancement. In contrast to 1 μM ketamine, 100 μM ketamine had no persisting effect on EPSPs (107.5 ± 10.7% of baseline 60 min following ketamine, N = 6, p = 0.4818 vs. baseline; p = 0.016 vs. 1 μM ketamine, Figure 2D). These results are consistent with our previous observations at P30 (27) and effects of ketamine in prefrontal cortex (17).

Akin to ketamine, we found that co-administration of APV with N₂O, eliminated persisting synaptic enhancement, although acute depression during N₂O was still observed (93.1 ± 9.6% of baseline with N₂O + APV, N = 5, p = 0.0006 vs. N₂O alone; Figure 2E). This result strongly suggests that activation of unblocked NMDARs during N₂O exposure plays a role in driving persisting enhancement. Prior studies indicate that effects of ketamine require activation of AMPARs (13,17). This prompted us to examine ketamine and N₂O in the presence of an AMPAR antagonist. For these experiments, we used DNQX (32), a competitive antagonist that washes out of slices faster than other AMPAR antagonists. We found that 10 μM DNQX administered for 10 min before and during ketamine eliminated synaptic enhancement (91.5 ± 2.4%, N = 5, p = 0.0004 vs. ketamine alone; Figure 3A). In contrast, DNQX had no effect on synaptic potentiation by N₂O (131.3 ± 10.2% change, N = 8, p = 0.3095 vs. nitrous alone; Figure 3B). Thus, a triggering role for AMPARs appears to separate the actions of ketamine from those of N₂O.

Figure 3.

Inhibition of AMPARs blocked potentiation by ketamine but not N₂O. A. At 10 μM, the competitive AMPAR antagonist, DNQX, administered before and during ketamine prevented persisting enhancement of EPSPs. The gap in the graph reflects time during which a repeat IO curve was obtained 60 min following ketamine, prior to responses returning completely to baseline. B. In contrast to ketamine, DNQX had no effect on synaptic enhancement following N₂O. Traces show EPSPs as in Figure 1. Calibration bar: 1 mV, 5 ms.

Prior studies indicate that ketamine activates intracellular signaling systems (4,33), including TrkB receptors via BDNF and perhaps direct drug binding (13,14), along with mTOR (17) and nitric oxide synthase (NOS) (27,34). In the following studies, we examined roles of these effectors in the actions of ketamine and N₂O. To examine TrkB receptors, we used the selective antagonist, ANA-12 (35), an agent that inhibits effects of ketamine on these receptors (14). In the presence of 1 μM ANA-12, ketamine failed to enhance EPSPs (95.8 ± 5.8% of baseline 60 min after ketamine, N = 5; p = 0.0014 vs. ketamine alone; Figure 4A). In contrast, we found no effect of 1 μM ANA-12 on persisting enhancement by N₂O (151.7 ± 12.7% of baseline, N = 5, p = 0.3228 vs. N₂O alone; Figure 4B). Because ANA-12 has biphasic effects on TrkB, we also examined a higher concentration, and found that 20 μM ANA-12 eliminated N₂O-induced synaptic enhancement (105.8 ± 12.2% of baseline, N = 5, p = 0.0317 vs. N₂O + 1 μM ANA-12; Figure 4B). We also examined a structurally and mechanistically distinct TrkB antagonist, cyclotraxin B (36). At 0.1 μM, cyclotraxin B completely blocked synaptic enhancement by both ketamine (92.0 ± 6.6%, N = 5, p = 0.0011 vs. ketamine alone; Figure 4C) and N₂O (101.6 ± 3.9%, N = 5, p = 0.0009 vs. N₂O alone; Figure 4D).

Figure 4.

Synaptic potentiation by ketamine and N₂O involve TrkB receptors. A. At 1 μM, the TrkB antagonist ANA-12 completely blocked synaptic enhancement by ketamine. B. ANA-12 also blocked synaptic enhancement by N₂O, but was ineffective at 1 μM (gray circles). At 20 μM, ANA-12 completely inhibited enhancement by N₂O (black circles). C,D. A second TrkB antagonist, cyclotraxin B, blocked synaptic enhancement by both ketamine (C) and N₂O (D) at a submicromolar concentration. Traces show EPSPs as in prior figures. Scale bar: 1 mV, 5 ms.

We examined mTOR using the inhibitor, rapamycin (37), an agent that blocks ketamine-induced effects in prefrontal cortex (17). At 1 μM, a concentration that inhibits both mTORC1 and mTORC2 (38), rapamycin eliminated synaptic enhancement by both ketamine and N₂O (ketamine + rapamycin: 94.7 ± 5.1% of baseline, N = 5, p = 0.001 vs. ketamine alone; N₂O + rapamycin: 99.3 ± 8.0%, N = 5, p = 0.0026 vs. N₂O alone, not shown). At 20 nM, rapamycin has selective effects on mTORC1 (37,38). We found that 20 nM rapamycin also blocked the enhancing effects of both ketamine (112.0 ± 10.6%, N = 5, p = 0.0266 vs. ketamine alone, Figure 5A) and N₂O (104.6 ± 8.7%, N = 5; p = 0.0084 vs. N₂O alone, Figure 5B), indicating a predominant role for mTORC1 in these effects.

Figure 5.

Persisting synaptic enhancement by both ketamine and N₂O involve mTOR and NOS. A,B. A nanomolar concentration of rapamycin that inhibits mTORC1 blocked synaptic enhancement by both ketamine (A) and N₂O (B). C,D. Synaptic enhancement by both ketamine (C) and N₂O (D) is also inhibited by a selective neuronal NOS inhibitor. In the case of N₂O, synaptic enhancement was blocked completely in 5 of 6 slices, with one slice showing potentiation. Traces depict representative EPSPs. Scale bar: 1 mV, 5 ms.

In prior studies, we found that metaplastic effects of ketamine are blocked by inhibitors of NOS (27). Other studies indicate that certain behavioral effects of N₂O also involve NO (39,40). Thus, we examined 3-bromo-7-nitroindazole (7-NIA), a specific inhibitor of neuronal NOS (41). At 5 μM, 7-NIA inhibited the effects of ketamine on synaptic responses (102.9 + 8.5%, N = 5, p = 0.0061 vs. ketamine alone, Figure 5C). The NOS inhibitor also markedly dampened the effects of N₂O, eliminating synaptic enhancement in five of six slices with one slice still showing significant synaptic potentiation (110.6 ± 18.2%, N = 6; p = 0.0208 vs. N₂O alone, Figure 5D). In the presence of the NOS inhibitor, ketamine did not differ from N₂O (p = 0.7289).

Similarities in N₂O and ketamine led us to examine whether N₂O-induced potentiation alters further potentiation by either N₂O or ketamine. In slices treated with N₂O, we observed no further potentiation with a second administration of N₂O (N₂O:147.4 ± 12.5% of baseline vs. 121.1 ± 5.8% following a second application of N₂O, N=5; p = 0.1536; Figure 6A). Similarly administration of ketamine following N₂O-induced potentiation resulted in no further enhancement (N₂O: 147.5 ± 9.6% vs. 146.8 ± 10.4% after ketamine, N = 5, p = 0.7047; Figure 6B). However, N₂O did not occlude LTP induced by tetanic stimulation (N₂O: 148.4 ± 7.9% of baseline vs. HFS: 190.4 ± 17.9%, N=5; p = 0.0353, Figure 6C). We also examined whether drug-induced potentiation was altered following tetanus-induced LTP. Sixty minutes following HFS, neither N₂O nor ketamine showed further enhancement (Supplemental Figure 3).

Figure 6.

N₂O-induced potentiation occludes further enhancement by either N₂O or ketamine. A,B. The graphs show enhancement of synaptic responses by N₂O and lack of further potentiation following a second administration of N₂O (A) or ketamine (B). C. In contrast, N₂O potentiation did not occlude further enhancement by tetanic stimulation. Traces show representative EPSPs at the times depicted in the graphs with solid filled symbols. Scale: 1 mV, 5 ms.

The ex vivo experiments in Figure 6A indicate that synaptic enhancement by N₂O is saturated following a single administration. Based on this observation, we examined whether administration of N₂O in vivo altered subsequent potentiation. Consistent with ex vivo studies, in vivo N₂O (30% for 15 min) occluded potentiation by both N₂O (87.7 ± 3.9%, N=5; p = 0.0001 vs. N₂O with no pretreatment; Figure 7A) and ketamine (93.2 ± 5.5%, N=5; p= 0.0076 vs. ketamine alone; Figure 7B), but not HFS (141.6 ± 17.9%, N=5; p = 0.9707 vs. HFS alone; Figure 7C) in slices prepared from N₂O-treated rats.

Figure 7.

In vivo N₂O occludes further potentiation by N₂O and ketamine but not HFS. For these experiments, rats received 30% N₂O by inhalation in vivo for 15 min prior to slice preparation. A,B. Neither N₂O (A) nor ketamine (B) enhanced synaptic responses in slices from rats that received in vivo N₂O. C. In contrast, tetanus (HFS)-induced LTP remained intact in slices from these rats. Traces depict representative EPSPs; scale: 1 mV, 5 ms.

DISCUSSION

The emergence of ketamine as a psychiatric treatment is a major, even revolutionary, advance. Because it differs from traditional antidepressants, ketamine offers unique opportunities to identify other novel treatments and mechanisms contributing to antidepressant actions. While mechanisms of ketamine are not completely understood, this agent has clear effects on neural circuitry in prefrontal cortex and hippocampus, regions thought to be important for depression-like behaviors. At sub-anesthetic concentrations, ketamine promotes neuroplasticity resulting in enhanced excitatory (glutamate-mediated) transmission in both regions (13,17). Contributors to the effects of ketamine include local disinhibition, activation of TrkB receptors and activation of mTOR (1,4,5,33).

Based on recent clinical trials indicating that N₂O has antidepressant effects in TRMD (22,23), we compared N₂O oxide and ketamine on synaptic transmission in the CA1 hippocampal region. Our results indicate that, like ketamine, N₂O persistently enhances CA1 transmission through mechanisms shared, in part, with ketamine. Shared mechanisms include activation of unblocked NMDARs during drug administration. This conclusion is based on two sets of experiments: 1) at effective concentrations, block of NMDARs by both ketamine and N₂O is submaximal, possibly involving NMDARs expressing GluN2B subunits (27,51); 2) complete block of NMDARs with APV (or saturating ketamine) prevents synaptic enhancement. In prior studies, sub-anesthetic, but not anesthetic ketamine produces antidepressant-like effects in rodents (17). Effects of ketamine and N₂O on CA1 transmission are also blocked by an mTOR antagonist, at a concentration implicating mTORC1 (42). There is controversy about the role of mTORC in effects of ketamine in human clinical trials with rapamycin pretreatment (43). It is also uncertain whether effects in the CA1 region relate to antidepressant actions of these agents (though see ref 11) or to other effects on memory and pain, and it is likely that changes in other brain regions such as medial prefrontal cortex and subcortical areas are important.

Effects of ketamine and N₂O are blocked by TrkB antagonists, although there are differences in how TrkB receptors appear to be engaged. Synaptic enhancement by ketamine is blocked completely by low micromolar ANA-12, while block of N₂O-mediated enhancement requires higher antagonist concentration. ANA-12 is a non-competitive TrkB antagonist that has biphasic effects with partial inhibition at an IC50 of ~45 nM and more complete inhibition above 10 μM (35). Another TrkB antagonist, cyclotraxin B, inhibited both ketamine and N₂O at a submicromolar concentration. Cyclotraxin B is a non-competitive, partial TrkB inhibitor that blocks both BDNF-dependent and BDNF-independent TrkB activation at similar concentrations (36). The effectiveness of this latter agent suggests a possible role of BDNF-independent effects of N₂O, although ANA-12 does not appear to inhibit BDNF-independent TrkB activation even at high concentrations (35,36). Another possibility is that 30% N₂O represents a higher effective concentration for triggering enhanced neurotransmission than 1 μM ketamine, thus requiring stronger TrkB inhibition. We also note that the anesthetic we used for slice preparation, isoflurane, may alter TrkB signaling (44,45). However, we found that our slices exhibited robust tetanus-induced LTP and responses to N₂O and ketamine (Figure 6); thus effects of isoflurane were minimal by the time of our experiments. Consistent with shared mechanisms, we found that N₂O-induced potentiation occludes effects of ketamine, and effects of both N₂O and ketamine are occluded by tetanus-induced LTP. However, N₂O does not occlude tetanus-induced LTP, indicating that HFS triggers additional mechanisms. We also found that in vivo exposure to N₂O occluded further potentiation by N₂O and ketamine, but not HFS.

Based on our prior study in which metaplastic effects of ketamine in juvenile hippocampus required NOS (27), we also examined a specific inhibitor of neuronal NOS. The NOS inhibitor completely blocked synaptic enhancement by ketamine, but had variable effects against N₂O, completely blocking potentiation in five of six slices. Prior studies indicate that NOS activation is stimulated by NMDARs and participates in effects of N₂O on anxiety and depression-like behaviors (39,46,47), and on tetanus-induced plasticity (48,56).

An unexplained difference between ketamine and N₂O is that N₂O transiently depresses AMPAR-mediated transmission during acute perfusion. This depression recovers quickly following drug washout and is followed by synaptic enhancement. No depression of AMPAR EPSPs was observed with ketamine. Synaptic depression was also observed with nitrogen, but nitrogen did not mimic synaptic potentiation by N₂O. This result suggests that changes in gas content likely contributed to initial depression, although N₂O has several actions (1) that could also depress transmission including presynaptic effects (5) and weak inhibition of AMPARs (21).

Ketamine and N₂O are non-competitive NMDAR antagonists that act by different mechanisms (1,5). While ketamine’s effects are use-dependent and involve open channel block, effects of N₂O show no use dependence, limited voltage dependence, and are unlikely to involve open channel block (21). Prior work indicates that activation of AMPARs is important for ketamine-induced behavioral and synaptic changes (12,13,17,49). Consistent with this, we found that AMPAR antagonism during drug administration blocked effects of ketamine, but not N₂O. The role of AMPARs in the effects of ketamine but not N₂O could reflect differences in mechanisms of the two drugs on NMDARs (1,5). In studies examining isolated NMDAR-mediated EPSPs, we found that both ketamine and N₂O caused acute depression of responses that was followed by persisting enhancement following drug washout, indicating that both aspects of glutamate transmission are augmented.

A poorly understood aspect of ketamine is that the drug has differential effects in juveniles and adults. Antidepressant-like effects and synaptic enhancement are observed in adult but not juvenile rodents (27,31), although some data suggest antidepressant effects in human adolescents (50,51). Similarly, the ability of ketamine to induce psychotic symptoms is observed in adults but not children (52,53). Ketamine also causes vacuolar changes in neurons in adult but not juvenile rodents (54). The latter effects, like synaptic enhancement, appear to involve network disinhibition (54). N₂O also causes neuronal vacuoles in adult rodents (20). In our prior studies, N₂O disinhibited the CA1 region of juvenile rats (25), but, as shown in the present work, did not cause synaptic enhancement in juveniles.

In summary, our results indicate that N₂O shares ketamine’s ability to enhance hippocampal excitatory transmission. These effects result from partial inhibition of NMDARs during drug exposure, but persisting effects result from activation of unblocked NMDARs during drug administration. In turn, these network effects involve similar, but not identical cellular mechanisms that result in persisting excitatory synaptic enhancement. Whether and how persisting enhancement drives antidepressant actions remains uncertain. While our experiments focused on NMDARs and excitatory transmission, both agents have effects on other transmitter and signaling systems that could contribute to clinical effects (1,5). It is intriguing that N₂O has antidepressant effects in humans that can last for a few weeks (23). In contrast to ketamine, which persists in the body for several hours and has active metabolites that may contribute, N₂O is rapidly cleared following inhalation and has no metabolites (1,5,57). Thus, sustained effects of N₂O derive from changes initiated during exposure to the gas.

KEY RESOURCES TABLE

Resource Type	Specific Reagent or Resource	Source or Reference	Identifiers	Additional Information
Add additional rows as needed for each resource type	Include species and sex when applicable.	Include name of manufacturer, company, repository, individual, or research lab. Include PMID or DOI for references; use “this paper” if new.	Include catalog numbers, stock numbers, database IDs or accession numbers, and/or RRIDs. RRIDs are highly encouraged; search for RRIDs at https://scicrunch.org/resources.	Include any additional information or notes if necessary.
Chemical Compound or Drug	Compressed gas/nitrous oxide	Praxair (Danbury , CN)	UN1070, CAS 10024-97-2
Chemical Compound or Drug	Compressed gas/ 95% O2 5% CO2	Airgas (Radnor, PA)	UN 3156
Chemical Compound or Drug	Compressed gas/95% N2O, 5.04% CO2	Puritan Medical Products (Overland,KS)/Airgas	UN 3156
Chemical Compound or Drug	D-APV	Millipore-Sigma	CAS: 79055-68-8, Cat# A8054
Chemical Compound or Drug	CNQX	Millipore-Sigma	CAS: 115066-14-3, Cat# C239
Chemical Compound or Drug	3-Bromo-7-nitroindazole	Millipore-Sigma	CAS:74209-34-0, Cat# B2050
Chemical Compound or Drug	ANA-12	Millipore-Sigma	CAS:219766-25-3, Cat# SML0209
Chemical Compound or Drug	Cyclotraxin B	Millipore-Sigma	CAS 1203586-72-4, Cat# 1769
Chemical Compound or Drug	Ketamine	Millipore-Sigma	CAS:61763-23-3, Cat# K113
Chemical Compound or Drug	DNQX	Tocris Bioscience/R&D system	CAS:2379-57-9, Cat# 0189
Chemical Compound or Drug	Rapamycin	Cayman Chemical	CAS: 53123-88-9, Cat# cay 13346
Organism/Strain	Rat: Sprague Dawley, male	Charles River Laboratories (Raleigh, NC)	RRID:SCR_003792
Software; Algorithm	pClamp 5,51 software	Molecular Devices	RRID:SCR_011323