Nitrous oxide induces hypothermia and TrkB activation: Maintenance of body temperature abolishes antidepressant-like effects in mice

Recent studies indicate that nitrous oxide (N 2 O), a gaseous anesthetic and an NMDA ( N -methyl-D-aspartate) receptor antagonist, produces rapid antidepressant effect in patients suffering from treatment-resistant depression. Our recent work implies that hypothermia and reduced energy expenditure are connected with antidepressant-induced activation of TrkB neurotrophin receptors — a key regulator of synaptic plasticity. In this study, we demonstrate that a brief exposure to N 2 O leads to a drop in body temperature following the treatment, which is linked to decreased locomotor activity; enhanced slow-wave electroencephalographic activity; reduced brain glucose utilization; and increased phosphorylation of TrkB, GSK3 β (glycogen synthase kinase 3 β ), and p70S6K (a kinase downstream of mTor (mammalian target of rapamycin)) in the medial prefrontal cortex of adult male mice. Moreover, preventing the hypothermic response in a chronic corticosterone stress model of depression attenuated the antidepressant-like behavioral effects of N 2 O in the saccharin preference test. These findings indicate that N 2 O treatment modulates TrkB signaling and related neurotrophic signaling pathways in a temperature-dependent manner, suggesting that the phenomenon driving TrkB activation — altered thermo-regulation and energy expenditure — is linked to antidepressant-like behavioral responses.


Introduction
Nitrous oxide (N 2 O) is a gaseous anesthetic which has recently shown promise in the treatment of depression (Liu et al., 2023;Nagele et al., 2015Nagele et al., , 2021)).Like the rapid-acting antidepressant ketamine, N 2 O also acts as an antagonist of the N-methyl-D-aspartate receptors (NMDARs), which has been suggested to be a key component in triggering subsequent cellular signaling mechanisms leading to antidepressant effects.The therapeutic effects of ketamine may involve the blockade of NMDARs on inhibitory GABAergic interneurons in brain areas implicated in antidepressant actions (e.g.medial prefrontal cortex and hippocampus).This increases excitatory neurotransmission and synaptic potentiation, possibly by triggering subsequent brain-derived neurotrophic factor (BDNF) release and increased neurotrophic signaling via tropomyosin-related kinase B (TrkB) receptors (Kohtala, 2021).Notably, recent studies suggest that both ketamine and N 2 O enhance excitatory neurotransmission in rat hippocampal slices, and that these effects are blocked by antagonists of TrkB and mammalian target of rapamycin (mTOR) (Izumi et al., 2022).Indeed, pharmacologically diverse antidepressants, including tricyclic antidepressants (e. g., amitriptyline), selective serotonin reuptake inhibitors (e.g., fluoxetine), and rapid-acting antidepressants (e.g., ketamine), converge in activation of TrkB and its downstream signaling cascades, which include activation of mTOR and inhibition of glycogen synthase kinase 3β (GSK3β) in the adult rodent medial prefrontal cortex and hippocampus (Alitalo et al., 2023;Aonurm-Helm et al., 2015;Autry et al., 2011;Li et al., 2010;Rantamäki et al., 2007Rantamäki et al., , 2011;;Saarelainen et al., 2003).
Attempts to explain the effects of antidepressants on TrkB signaling involve drug induced increases in BDNF expression and release, or direct receptor interactions (reviewed in Kohtala and Rantamäki, 2021;Rantamäki, 2019) Notably, some recent studies have suggested that structurally and pharmacologically distinct antidepressant drugs bind directly to the TrkB receptor (Casarotto et al., 2021;Jang et al., 2009Jang et al., , 2010;;Moliner et al., 2023).Rodent studies have also indicated that antidepressant-like effects and their associated molecular and structural changes are coupled to TrkB and its downstream signaling pathwaysmost notably the activation of mTOR and its effector p70S6K, and the inhibition of GSK3β (Beurel et al., 2016;Duman et al., 2016;Li et al., 2010).We have previously shown that N 2 O has no acute effect on TrkB activation in the mouse medial frontal cortex (Kohtala et al., 2019).Instead, TrkB phosphorylation is gradually (within 15 min) upregulated after the cessation of the gas treatment.Recently, we discovered that various classical antidepressants, as well as sedative-anesthetic drugs lacking established clinical value in the treatment of depression, induce the activation of neurotrophic signaling in the medial prefrontal cortex and hippocampus by producing hypothermia and decreasing the metabolic rate in mice (Alitalo et al., 2023).This temperature-mediated TrkB signaling is not compromised in Val66Met mice with impaired activity-dependent BDNF release and is readily prevented by manipulating the ambient temperatureconceivably through its effect on body temperature, which is determined by metabolic activity (Caro et al., 2013;Mrozek et al., 2012).It is tempting to speculate that N 2 O treatment also shares these effects on body temperature and metabolic activity, which would explain the observed delayed emergence of TrkB signaling following N 2 O exposure.
In this study, we provide evidence supporting the hypothesis that the effects of N 2 O on neurotrophic signaling pathways in the medial prefrontal cortex are connected to disrupted bioenergetics and accompanying hypothermia.Furthermore, our findings suggest that the phenomenon underlying TrkB activationchanges in energy expenditure and thermoregulationis associated with antidepressant-like behavioral responses in mice.

Animals
10-16-week-old C57BL/6JRccHsd male mice (Envigo, Venray, Netherlands) were used in the experiments.Unless otherwise described, animals were group-housed (approx.4-5 per cage) and maintained under standard laboratory conditions (22 ± 1 • C, 12 h light− dark cycle, lights on at 6 a.m. or 9 a.m.) with access to food and water available ad libitum.Treatments were performed during the light phase (Zeitgeber Time, ZT 3-7).The animal experiments were carried out in compliance with the European Communities Council Directive of September 22, 2010 (Directive, 2010/63/EU), according to the guidelines of the Society for Neuroscience, and approved by the County Administrative Board of Southern Finland (Licenses ESAVI/9793/04.July 10, 2016 and ESAVI/5844/2019).

Drug administrations
Medetomidine-HCl (0.3 mg/kg; i.p.; Domitor; Orion Pharma) was administered intraperitoneally (i.p.) in a volume of 10 ml/kg.Nitrous oxide (N 2 O) was administered as previously described (Kohtala et al., 2019).Briefly, medical grade N 2 O (Woikoski) was mixed with oxygen to achieve a gas mixture consisting of 50 or 65% N 2 O.The composition of the mixture was monitored using an oxygen sensor (73-4960; Hugo Sachs Elektronik -Harvard Apparatus GmbH; March-Hugstetten, Germany).The gas was administered into airtight acrylic chambers (for biochemical analyses (width × length × height): 14 × 25 × 9 cm; for biochemical and EEG analyses: 11.5 × 11.5 × 6.5 cm) with a flow rate of 4-8 l/min.Control animals received a sham-treatment of pressurized room air or sterile saline solution for inhaled and injected treatments, respectively.

EEG recordings
Electroencephalographic (EEG) recordings were performed as previously described (Kohtala et al., 2019).Briefly, electrodes were implanted under isoflurane anesthesia.Lidocaine (10 mg/ml) was used as local anesthetic, and buprenorphine (0.1 mg/kg, s.c.) and carprofen (5 mg/kg, s.c.) used for postoperative analgesia.From now on the animals were single housed.Two epidural screw EEG electrodes were placed above the fronto-parietal cortex.An additional screw served as mounting support.Two silver wire electrodes were implanted in the nuchal muscles to monitor the electromyogram (EMG).After a recovery period (5-7 days), animals were connected to flexible counterbalanced cables for EEG/EMG recording and habituated to recording cables for three days.The EEG and EMG signals were amplified (gain 5/10 K) and filtered (high pass: 0.3 Hz; low pass 100 Hz; notch filter) with a 16-channel AC amplifier (A-M System, model 3500), sampled at 254 Hz or 70 Hz with 1401 unit (CED), and recorded and processed using Spike2 (version 8.07, Cambridge Electronic Devices).EEG power spectra were calculated within the 1-50 Hz frequency range by fast Fourier transform (FFT = 256, Hanning window, 1.0 Hz resolution).Oscillation power in each bandwidth (delta = 1-4 Hz; theta = 4-7 Hz; alpha = 7-12 Hz; beta = 12-25 Hz; gamma low = 25-40 Hz; gamma high = 60-100 Hz) was computed in 30-300-s epochs from spectrograms (FFT size: 1024 points) for each animal.Representative sonograms were computed using a Hanning window with a block size of 512.

Corticosterone model of depression and saccharine preference test
Depressive phenotype was modeled in animals using chronic corticosterone administration (Moda-Sava et al., 2019).Corticosterone (#46148, Sigma-Aldrich) was dissolved in 99.5% ethanol and diluted with tap water to a final concentration of 100 μg/ml of corticosterone and 1% ethanol (CORT).Single-housed mice were exposed to CORT in place of drinking water for 16 days, followed by a weaning phase during which CORT mice received a solution at 50%, and then 25% of the full dose for 3 days, respectively.Drinking solutions were replaced every three days.
Saccharin preference test (SPT) examines the animal's capacity to experience hedonic pleasure evoked by sweet solutions.Saccharin was chosen instead of sucrose due to its negligible caloric value, thus controlling for energy intake.Mice were habituated to the presence of two drinking bottles prior to the experiments.On the test day, drinking was restricted for 8 h (ZT 5:30-12:30) before the start of the experiment.Saccharin solution was prepared immediately before testing by dissolving saccharin sodium salt hydrate (S1002-500G, Sigma, USA) in tap water to make 0.1% (w/v) solution.The mice were presented saccharin solution in one bottle and water in the other bottle over 4 h (from ZT 12:30-16:30).Next, the bottles were taken away and corticosterone and control solutions were returned to the cages.Water and saccharin solution consumption was calculated by weighing the bottles before and after the 4-h testing period.Saccharin preference was determined by calculating the fraction (%) of saccharin solution consumption divided by total solution consumption.

High-resolution autoradiographic analysis of functional brain activity
Autoradiographic procedure for the analysis of [ 3 H]-2-deoxy-Dglucose (2-DG; 1.5 μCi (55.5 kBq) per gram of body weight, #NET549A001MC, PerkinElmer) distribution was adapted from the method by Sokoloff et al. (1977) as previously described (Alitalo et al., 2020).Distribution and accumulation of tritiated glucose analog was detected from cryosectioned 20 μm coronal brain sections using a direct particle counting system (BeaQuant; Atlantic Instruments for Research, Nantes, France) with 21-h acquisition time.Regions of interest (ROI) were chosen for qualitative and quantitative analysis based on our previous experiments and the literature on the effects of NMDAR antagonists in rodents (Miyamoto et al., 2000).ROI were outlined in a proprietary software, Beamage (Atlantic Instruments for Research, Nantes, France), using Paxinos' and Allen's mouse brain atlas as a reference (Lein et al., 2007).For semiquantitative analysis of regional glucose utilization, the activity of a given ROI (cpm/min/mm 2 ) was averaged from multiple sections from each animal and normalized to the specific activity of corpus callosum of the same subject (Duncan and Stumpf, 1991).As the regional glucose utilization in white matter is relatively unaffected by fluctuations in metabolic demand, variation resulting from subjects' glucose homeostasis and amount of injected radiolabel is thereby accounted for (Duncan and Stumpf, 1991).Terminal blood was collected for quantifying circulating glucose and remaining radiotracer concentration using a commercial glucose monitor (FreeStyle Freedom Lite; Abbott, Lake Bluff, IL, USA) and a liquid scintillation counter (Tri-Carb 2910 TR; PerkinElmer, Waltham, MA, USA), respectively.

Temperature measurements and thermoregulation
Core body temperature was measured using small-animal rectal probe thermometers (FHC Frederick Haer & Co, Bowdoin, ME, USA; 7001H).The measurements were performed as quickly as possible in order to minimize the increase in body temperature resulting from handling stress.For longitudinal analysis of body temperature, wholebody thermal imaging was used.Thermal images were acquired using a forward-looking infrared (FLIR) P640 thermal camera (FLIR Systems, Inc., Wilsonville, OR, United States; rented from Infradex, Vantaa, Finland), which has a reported temperature range from − 40 to +500 • C and a reading accuracy of ±2%.The camera was equipped with a 19mm focal length lens with an optical angle of 45 • and was positioned perpendicular to the imaged plane above the recording chambers.Thermal data was acquired at a rate of one image every 30 s before (baseline; 3 min) and after the gas exposure (recovery; 14 min).Images were analyzed using either FLIR Tools software (version 6.4.18039.1003),where highest radiometric pixel temperature was manually acquired for each rodent cage/image using the rectangle tool, or ThermaCAM Researcher Pro (version 2.10), where the rectangle tool was used to automatically acquire the highest pixel temperatures for each rodent cage/image.
In the experiments involving controlled ambient temperature, animals were kept in a standard vented incubator (Vet-Tech Solutions Ltd, Congleton, United Kingdom) set to maintain the temperature of the chamber at 36 ± 1 • C.Besides the internal thermometer of the machine, the incubator temperature was monitored with a separate thermometer.Core temperature measurements were acquired using small-animal portable rectal thermometers (Microtherma 2, Agntho's AB, Lidingö, Sweden or DC Temperature Controller, FHC, Bowdoin, ME, USA).

Dissection and processing of the brain samples
Animals were euthanized at indicated times by rapid cervical dislocation followed by decapitation.Bilateral medial prefrontal cortex (mPFC, including prelimbic and infralimbic cortices) was immediately dissected on a cooled Petri dish and stored at − 80 • C until further processing.For the analysis of crude brain homogenates, the samples were homogenized in lysis buffer (137 mM NaCl, 20 mM Tris, 1% NP-40, 10% glycerol, 48 mM NaF, H 2 O, Pierce Protease Inhibitor Mini Tablet (Thermo Scientific; Waltham, MA, USA), Pierce Phosphatase Inhibitor Mini Tablet (Thermo Scientific; Waltham, MA, USA)).After ~15-min incubation on ice, samples were centrifuged (16,000×g, 15 min, 4 • C) and the resulting supernatant was collected for Western blot analysis.

Statistical analysis
The data were analyzed with unpaired t-tests, and one-way analysis of variance (ANOVA) with Tukey's or Dunn's post hoc tests multiple comparisons test when comparing multiple groups or multiple variables, respectively.Statistical testing was done using Graphpad Prism (v9.51;La Jolla, CA, USA).All tests were two-sided and a P ≤ 0.05 was considered statistically significant.Details of statistical tests and n numbers for each experiment are shown in Table S1.

N 2 O-induced TrkB signaling is associated with reduced body temperature
Previous studies demonstrate that N 2 O evokes TrkB signaling in the medial prefrontal cortex after its acute pharmacological effects have dissipated, which co-occurs with emergence of slow-wave EEG activity (SWA; 1-4 Hz, delta band) (Kohtala et al., 2019), commonly observed during physiological deep sleep.Consequently, we sought to investigate whether N 2 O also regulates other physiological changes associated with deep sleep: behavioral immobility (Szymusiak, 2018), attenuated brain energy expenditure (DiNuzzo and Nedergaard, 2017), and reduced body temperature (Landolt et al., 1995).The α 2 -adrenergic sedative drug medetomidine was used as a positive control due to its ability to produce a sleep-like state characterized by increased SWA and TrkB signaling (Alitalo et al., 2023;Kohtala et al., 2019;Purdon et al., 2015).
First, we replicated our earlier observations that acute treatment with medetomidine (0,3 mg/kg, i.p.) and recovery from N 2 O (65% N 2 O/ 35% O 2 , 20 min) increase slow EEG activity in the cortex (Fig. 1A).Brain samples from medial prefrontal cortex were collected from mice treated with N 2 O either at the end of a continuous 25-min exposure, or after 5 min or 15 min of recovery following 20 min N 2 O exposure.Phosphorylation of TrkB, GSK3β, and p70S6K remained unchanged in samples collected at the end of a 25 min exposure, while statistically significant increases were observed after 15 min in pTrkB, GSK3β, and pp70S6K (Fig. 1B).Statistically non-significant trend of increased phosphorylation was observed 5 min into recovery, although the subjects displayed considerable individual variation.N 2 O also increased the phosphorylation of ERK1/2 in samples collected during gas exposure, but decreased during recovery, suggesting a non-canonical form of TrkB activation (Fig. S1) (Alitalo et al., 2023;Kohtala and Rantamäki, 2019).In the open field test, animals under the influence of N 2 O were slightly more active than control animals, whereas the locomotor activity decreased drastically after cessation of the gas exposure (Fig. 1C).During this recovery phase, the animals were visually observed to be noticeably sedated, remaining still in a crouched, sleep-like posture throughout the monitoring period.
Next, we assessed the effects of N 2 O on body temperature using an infrared camera, which enables non-invasive temperature monitoring in freely moving animals.Thermographic imaging revealed a progressive decrease in the cutaneous body temperature of animals recovering from a 20-min 65% N 2 O exposure, continuing throughout the 15-min monitoring period (Fig. 1D).Overall, these findings demonstrate that a transient exposure to N 2 O activates TrkB signaling during a state characterized by increased slow EEG activity and hypothermia.

Brain energy metabolism is decreased during acute exposure to medetomidine or recovery from N 2 O
To investigate the potential bioenergetic parallels between the sedative states induced by N 2 O and medetomidine, we employed β-emission camera-based digital [ 3 H]-2-deoxy-D-glucose ( 3 H-2-DG) autoradiography (Alitalo et al., 2020;Sokoloff et al., 1977).The technique, analogous to methods used to evaluate metabolic demand during distinct sleep stages and acute drug effects (Miyamoto et al., 2000;Vyazovskiy et al., 2008), facilitates high-resolution imaging without significant interference from anesthesia, a common requirement in nuclear imaging of small animals.Cellular uptake of 3 H-2-DG mirrors that of glucose, but 2-DG cannot undergo glycolytic metabolism, thereby trapping the radiolabel inside glucose consuming cells.Consequently, the resulting autoradiograph of accumulation and distribution of the radiolabel in brain sections reflects the regional glucose utilization (i.e., functional activity) over the radiolabel uptake period.
To evaluate cerebral metabolism during recovery from N 2 O exposure, mice were administered 3 H-2-DG immediately following a 20-min 65% N 2 O exposure or a sham treatment and were subsequently terminated after a 20-min radiolabel uptake period.Animals in a separate cohort were given medetomidine (0.3 mg/kg) or saline, injected with 3 H-2-DG 10 min later when the sedative effects plateaued, and terminated after a 20-min radiolabel uptake period.Terminal blood samples were analyzed for glucose and activity concentration (see Supplementary method).
Both N 2 O and medetomidine resulted in a general decrease in cerebral glucose uptake, indicating lower energy expenditure (Fig. 2A, S4, S5).Specific brain area quantification revealed that both treatments significantly decreased glucose utilization in the retrosplenial cortex and laterodorsal thalamus (Fig. 2B and C), with medetomidine also reducing glucose utilization in the cingulate and primary motor cortices (Fig. 2B-S3A).The reduction in glucose utilization evoked by N 2 O was  S1).Abbreviations: GSK3β, glycogen synthase kinase 3β; p70S6k, ribosomal protein S6 kinase; SWA, slow-wave activity; TrkB, Tropomyosinrelated tyrosine kinase B. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)more subtle compared to that induced by medetomidine (Fig. S3B), mirroring the intensity of the electrophysiological (Fig. 1A) and molecular signaling responses elicited by the treatments (Alitalo et al., 2023;Kohtala et al., 2019).The animals treated with medetomidineunlike those exposed to N 2 Oexhibited significant hyperglycemia, demonstrating systemic bioenergetic effects of the anesthetic drug (Fig. S2A).

The antidepressant-like effects of N 2 O are prevented by maintaining body temperature post-treatment
To investigate the functional implications of TrkB activation during hypothermia and reduced energy expenditure, we tested whether antidepressant-like behavioral effects can be modulated by altering the metabolic-thermal responses evoked by nitrous oxide.We used 60-min exposure of 50% N 2 O (in O 2 ), mirroring previous studies using N 2 O in the treatment of clinical depression (Nagele et al., 2015(Nagele et al., , 2021)).Our prior research has demonstrated that this N 2 O dosing regimen also transiently augments SWA and TrkB signaling (Kohtala et al., 2019).We evaluated the behavioral effects of N 2 O using the chronic corticosterone (CORT) model of depression (Fig. 3A), a well-established model for investigating the antidepressant-like effects of various drugs, including ketamine (Moda-Sava et al., 2019).Animals subjected to CORT showed reduced preference for saccharine, an indicator for anhedonia, which was significantly alleviated by N 2 O when tested 32 h after exposure (Fig. 3B).Intriguingly, this behavioral effect was abolished when the animals were placed in a warm incubator (36 ± 1 • C) for an hour post-intervention, which prevented the N 2 O-induced decrease in body temperature.Warming alone also alleviated the CORT-induced decrease in saccharine preference.After a week from the treatment, neither N 2 O nor warming had any effect on the saccharine preference test, open field test, or tail suspension test (Fig. S6).

Discussion
The effects of antidepressant drugs are commonly explained using Fig. 2. Qualitative and semiquantitative analysis of the radiolabeled glucose analogue 2-deoxy-D-glucose (2-DG) distribution.A. Animals sedated using medetomidine (MED; 0.3 mg/kg, i.p.) or recovering from nitrous oxide (N 2 O; 65%, 20 min) exposure exhibit a general decrease in 2-DG accumulation in cortical and thalamic areas, indicating decreased functional activity.The regions of interest (ROI) used for quantification are marked in the representative brain atlas images.ROI activity was averaged from multiple sections of each animal and normalized to the activity of the corpus callosum.Medetomidine-treated animals showed significant regional decreases in the cingulate and retrosplenial cortices (CTX) and laterodorsal thalamus compared to vehicle-treated animals.During recovery from 65% N 2 O exposure, glucose uptake in the retrosplenial CTX and laterodorsal thalamus was lower than in animals exposed to compressed room air (SHAM), with a smaller, nonsignificant trend in the cingulate CTX.Data are presented as mean ± standard error of mean.p = *<0.05,**<0.01 (for statistical analyses and n numbers, see Table S1).S1).
the principles of conventional receptor pharmacology, which may sometimes overlook the complex underlying physiological processes.The findings presented in this report suggest that the activation of TrkB in the medial prefrontal cortex and associated signaling pathways-key mechanisms involved in synaptic plasticity and antidepressant-like behavioral responses-are triggered as an evoked physiological response during recovery from transient exposure to N 2 O.The recovery period was characterized by increased cortical slow-wave EEG activity, reduced body temperature, and decreased locomotor activity.Moreover, 2-DG experiments demonstrate an overall trend of decreased glucose utilization for most brain areas, including the cortex, suggesting a reduction in brain temperature.Notably, N 2 O is eliminated within minutes, essentially without metabolism, suggesting that it triggers TrkB activation at a stage when it is no longer binding to receptor targets.It remains unclear by which adaptive pharmacological mechanisms N 2 O elicits its effects on TrkB signaling and physiological changes characteristic with deep sleep.Transient NMDAR blockade is a plausible candidate, although also other mechanisms such as nitric oxide (NO) (Biojone et al., 2023;Emmanouil and Quock, 2007;Liu et al., 2020) may be involved.Notably, we have also observed "rebound" slow-wave activity after a single subanesthetic dose of ketamine (Kohtala et al., 2019).
To study the behavioral implications of this phenomenon, we utilized the chronic corticosterone model and the saccharine preference test, a measure of depressive-like behavior.The saccharine preference test was chosen because, unlike sucrose, saccharine has no caloric content or nutritional value that would directly contribute to metabolic processes.Chronic stress and corticosterone are known to dysregulate TrkB signaling, reduce the density of postsynaptic dendritic spines, and alter ensemble dynamics in the mPFC, contributing to depressive-like behavior in rodents (Li et al., 2010;Moda-Sava et al., 2019).We demonstrated that the antidepressant-like behavioral effects of N 2 O are blunted by preventing the post-treatment hypothermic response.Since we have previously shown that the hypothermia-induced TrkB signaling can be abolished by maintaining body temperature (Alitalo et al., 2023), the findings of the present study suggest that temperature-mediated TrkB signaling also has a functional role.This notion is supported by recent findings demonstrating that the activation of TrkB signaling during hypothermia mediates structural plasticity and neuroprotective effects (Peretti et al., 2021).The antidepressant-like effect of N 2 O was not evident 7 days after the delivery, although some reports have noted sustained analgesic effects a single exposure to N 2 O (Bessière et al., 2010).
Interestingly, mice subjected to warming alone also showed increased saccharine preference, although the effects were not as consistent as with N 2 O treatment.It is possible that the warming condition used in this study was sufficient to elicit therapeutic effects on its own, as whole-body hyperthermia (WBH) has been shown to have antidepressant potential in clinical and preclinical studies (Hale et al., 2017;Janssen et al., 2016).Notably, the antidepressant effects of treatments like WBH and electroconvulsive therapy have been associated with sustained thermoregulatory cooling following treatment (Hanusch et al., 2013;Szuba et al., 1997).Thus, it is tempting to speculate that reductions in body temperature following these treatments may also facilitate TrkB signaling.
Altogether, the present study strengthens previous findings of temperature being a causal driver for facilitating TrkB signaling in the mouse brain (Alitalo et al., 2023;Peretti et al., 2021) by demonstrating a similar effect following N 2 O treatment.Further characterization of the mechanisms behind this phenomenon remains an important task, as most studies on pharmacological interventions omit disclosing body or brain temperature as experimental variables, introducing a significant confounding factor for molecular and behavioral studies of antidepressants.

Conclusions
The results of this study reveal an unexpected association between N 2 O-induced TrkB phosphorylation, antidepressant-like effects, thermoregulation, and brain energy metabolism.Notably, maintaining normothermia during recovery from N 2 O exposure appears to reduce the antidepressant-like behavioral response in the saccharine preference test.These findings align with prior research indicating that TrkB activation is temperature-dependent, underscoring the significance of bioenergetics and thermoregulation in modulating antidepressant responses.Consequently, the results suggest that body temperature should be considered an important experimental variable in studies of neurotrophic signaling and antidepressant-like effects.
reported in this paper.

Fig. 1 .
Fig. 1.N 2 O evokes a sleep-like state, during which TrkB signaling becomes regulated.A. Medetomidine (MED; 0.3 mg/kg, i.p.) produces a robust and rapid increase in SWA, whereas transient exposure to nitrous oxide (N 2 O; 65%, 20 min) triggers gradual increase in SWA post-treatment.Spectrograms show the difference in spectral power normalized to baseline.Control (CTRL) animals received i.p. injection of vehicle solution.B. Phosphorylation of TrkB Y816 , GSK3β S9 , and p70SK6 T421/S424 increases gradually after N 2 O (65%, 20 min) administration is terminated, reaching significant levels by 15 min.In contrast, animals terminated after a 25-min continuous N 2 O exposure showed negligible effects on the signaling.C. Open field test during and after N 2 O exposure (65%, 20 min) displays slightly elevated locomotor activity during the gas exposure and reduced activity afterwards.D. Continuous non-invasive thermal recording reveals a progressive decrease in body temperature after N 2 O exposure.Qualitatively, the temperature of distal body parts decreased more markedly during recovery, with localized clusters of heat observed near brown adipose tissue.Phosphoproteins were normalized against corresponding total protein or Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and compared to the average signal of the control group.Data are presented as mean ± standard error of mean.p = *<0.05,**<0.01 (for statistical analyses and n numbers see TableS1).Abbreviations: GSK3β, glycogen synthase kinase 3β; p70S6k, ribosomal protein S6 kinase; SWA, slow-wave activity; TrkB, Tropomyosinrelated tyrosine kinase B. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3 .
Fig. 3. Modulation of antidepressant-like behavioral effects of N 2 O through manipulation of ambient temperature.A) Schematic overview of the experiment.B) A single exposure to nitrous oxide (N 2 O; 50%, 60 min) increased saccharine preference in chronic corticosterone (CORT)-based model of depression in mice 32 h after treatment (SPT).The effect was abolished if the animals were placed in warm ambient temperature (36 ± 1 • C) for 60 min post-intervention.Data are presented as mean ± standard error of mean.p = *<0.05,****<0.0001(for statistical analyses and n numbers, see TableS1).