Post-traumatic stress disorder (PTSD) affects around 354 million adults worldwide [1]. Patients with PTSD may experience intrusive memories of the trauma, persistent negative thoughts, hypervigilance, and trigger avoidance [2]. Trauma-focused cognitive behavioral therapy is the most effective treatment [3,4], however, more than one third of patients do not achieve remission [5,6]. Adjunct treatment with selective serotonin reuptake inhibitors (SSRIs) can reduce reactivity and avoidance of trauma reminders [7,8], but takes between 3 and 6 weeks to have a clinically significant effect [9]. These therapeutic limitations have spurred interest in identifying compounds that can rapidly reduce threat perception and enhance fear extinction.

Classical psychedelics including psilocybin and lysergic acid diethylamide (LSD) act through the 5-HT_2A receptor to alter cognitive and sensory processes, which can induce profound subjective experiences and lasting changes in mood [10,11,12]. Both psilocybin and 3,4-methyl enedioxy methamphetamine (MDMA) show great promise as medications for mood and trauma related disorders. A single dose of psilocybin confers a reduction in depression and anxiety symptoms for weeks to months [13,14] with comparable efficacy to the selective-serotonin reuptake inhibitor escitalopram [11]. Further, MDMA-assisted psychotherapy induced significant reductions in the affective symptoms of PTSD [15]. The subjective experiences of psychedelics may underlie a crucial component of their therapeutic effects [16]. However, the intensity of the psychedelic experience requires close clinical oversight for an extended period of time [17]. Psilocybin belongs to the tryptamine class of psychedelic drugs, and has a duration of 4–8 h [18], which limits practical clinical supervision [17]. By reducing clinical contact time, shorter acting psychedelics are poised to be less costly and more scalable in a clinical setting [19]. A suitable candidate would be orally available and have brief duration while retaining the ability to produce the subjective experiences of classical psychedelics.

The psilocybin derivative 4-hydroxy-diisopropyltryptamine (4-OH-DiPT) produces intense psychedelic effects in humans [20]. While anecdotal reports of use are sparse, they suggest that the subjective effects of 4-OH-DiPT are stronger and last considerably shorter (1–3 h) [20,21] than psilocybin (4–8 h) [18]. 4-OH-DiPT induces the head-twitch response (HTR) in mice [22], and substitutes for the psychedelic 2,5-dimethoxy-4-methylamphetamine (DOM) in drug discrimination in rats [23], suggesting that it shares key similarities with other classical psychedelics, such as 5-HT_2A mediated calcium influx [22]. Activation of the 5-HT_2A receptor is a critical mediator of the head twitch response in rodents [24] and psychedelic state in humans [25]. However, the compound-specific polypharmacology of psychedelics at serotonin GPCRs produces variation in subjective effects [26] and brain region activation [27]. Thus, it is essential to interrogate the functional interactions of 4-OH-DiPT with each serotonin receptor, which are largely unknown. We first used G protein dissociation and β-arrestin2 recruitment bioluminescence resonance energy transfer (BRET) assays to determine the functional efficacy of 4-OH-DiPT at 12 human serotonin GPCRs. Systemic administration of classical psychedelics reduces conditioned fear expression in rodents; this effect is blocked by 5-HT_2A receptor antagonists [28,29,30]. Therefore, we determined whether systemic administration of 4-OH-DiPT prior to fear extinction training altered fear extinction learning and recall. Systemic administration of psychedelics likely reduces conditioned fear responses by acting in the basolateral amygdala (BLA) as local activation of 5-HT_2A receptors in the BLA, but not other regions such as the prefrontal cortex, hippocampus, or the bed of the nucleus stria terminalis recapitulates the effect of systemic administration [3,4]. Excitatory principal neurons and inhibitory interneurons in the BLA have opposite roles in freezing behavior during fear conditioning and extinction [31,32,33]. Next, we used ex vivo slice electrophysiology to identify the acute effect of 4-OH-DiPT on synaptic transmission within the BLA. Immunohistochemistry studies show conflicting results whether 5-HT_2A receptor expression in the BLA is restricted to inhibitory neurons [34], or expressed in both inhibitory and excitatory neurons [35,36]. Finally, we used RNAscope, which can detect single mRNA transcripts with high specificity and sensitivity [37], to determine the co-expression ofHtr2a mRNA with cell-type specific mRNA markers for glutamate and GABA neurons in the BLA.

Materials and methods are described in detail in the Supplementary Materials

Bioluminescence resonance energy transfer (BRET) assays

HEK293T cell culture and transfections, as well as BRET² G protein dissociation and β-arrestin 2 recruitment assays were performed as previously described [38]. Briefly, cells were transfected using a reverse transfection method and plated in 1% dialyzed FBS (dFBS) at an approximate density of 15,000 cells per well into poly-L-lysine-coated 384-well white assay plates (Greiner Bio-One).

For G protein-dissociation assays, cells were transfected in a 1:1:1:1 ratio of receptor/Gα-Rluc8/Beta/GFP2-γ constructs, except for 5-HT_2C (1:2:2:2). For β-Arrestin2 recruitment assays, cells were transfected in a 1:15 ratio of 5-HTR-Rluc8/GFP2-fused human β-Arrestin2. All transfections were prepared in Opti-MEM (Invitrogen) and used a 3:1 ratio of TransIT-2020 (Mirus) µL:µg total DNA. On the assay day, plates were decanted and 20 µl of drug buffer (1× HBSS, 20 mM HEPES, pH 7.4) per well was added using a Multidrop (ThermoFisher Scientific), and plates were allowed to equilibrate at 37 °C in a humidified incubator before receiving drug administration. Drug dilutions of all compounds were made in McCorvy buffer (1× HBSS, 20 mM HEPES, pH 7.4, supplemented with 0.3% fatty acid-free BSA and 0.03% ascorbic acid) and dispensed using a FLIPR Tetra (Molecular Devices, San Jose, CA). Plates were incubated at 37 °C in a humidified incubator for 60 min. Before reading, addition of coelenterazine 400a (5 µM final concentration; Nanolight Technology; Norman, OK) was performed by the FLIPR Tetra and plates were read at 400 nm and 510 nm at 0.8 s per well using a PheraStarFSX (BMG Labtech; Cary, NC). BRET ratios of 510/400 luminescence were calculated per well and were plotted as a function of drug concentration using Graphpad Prism 5 or 9 (Graphpad Software Inc., San Diego, CA). Data were normalized to % positive control (5-HT) stimulation and analyzed using nonlinear regression “log(agonist) vs. response” to yield Emax and EC50 parameter estimates. The relative activity (log(E_MAX/EC₅₀)) for 4-OH-DiPT was normalized to the reference ligand, 5-HT, and was used to calculate Δlog(E_MAX/EC₅₀)= log(E_MAX/EC₅₀)_4-OH-DiPT - log(E_MAX/EC₅₀)_5-HT. The strength of signaling for the G protein and β-arrestin pathways were compared at each 5-HT receptor to yield ΔΔlog(E_MAX/EC₅₀)= Δlog(E_MAX/EC₅₀)_{G protein} - Δlog(E_MAX/EC₅₀)_β-arrestin and bias factors[39] (10^ΔΔlog(E_MAX/EC₅₀)) with respect to G protein activity (see Supplementary Fig. 1).

Animals

Adult C57BL/6 J male and female mice were obtained from Jackson Laboratories (stock#: 000664) and housed in a temperature and humidity-controlled room with unlimited access to food and water, unless stated otherwise. All protocols were approved by The Institutional Animal Care and Use Committee at the Medical College of Wisconsin.

Stereotaxic surgery

Mice were anesthetized isoflurane inhalation and placed in a robot stereotaxic system (Neurostar GmbH; Tübingen, Germany). AAV9-mDlx-NLS-mRuby2 (Addgene #99130, a gift from Viviana Gradinaru) was injected at: AP −1.22, ML + /−2.83, DV −4.75, −4.70, and −4.65 (150 nL). Following surgery, mice were given subcutaneous injections of buprenorphine-SR 1 mg/kg and allowed to recover for at least 7 days prior to electrophysiological recordings.

Behavior

Fear conditioning, extinction training, and retrieval tests were performed in light and sound attenuating boxes (Actimetrics, Wilmette, IL, USA). FreezeFrame 4.0 Software was used to control contingencies and measure freezing behavior (Actimetrics). For FC, mice were exposed to 5 auditory tones (75–85 dB, 20 s) that co-terminated with a foot shock (0.5 mA, 1 s) with 60–80 s intervals between cues. One day later 4-OH-DiPT (1 or 3 mg/kg, i.p.) or vehicle was administered prior to extinction training. Mice were then placed in a novel environment and exposed to 20 tones with either 5 sec intervals or 60–80 s pseudorandom intervals without foot shocks. On day 3, the cued extinction recall test was performed under the same parameters and 10 tones were delivered. The open field test, light dark box, elevated plus maze, and novelty-suppressed feeding test were performed as previously described [40] and analyzed with ANYmaze.

Slice preparation and electrophysiology

BLA brain slices were prepared as previously described [40,41,42]. Briefly, brains were extracted following isoflurane-induced anesthesia, embedded in 4% low melting point agarose, and cut into 250 µm thick sections with a VT 1200 S (Leica Biosystems; Nussloch, Germany) in an NMDG based solution. After slice cutting, Artificial Cerebrospinal Fluid (ACSF) was progressively spiked into the NMDG solution every 5 min for 20 min at room temperature to gradually reintroduce Na⁺ as previously described [43]. Slices were left to recover for at least an additional 30 min in ACSF. All solutions were saturated with 95% O₂ and 5% CO₂.

Patch clamp recordings were performed and analyzed as previously described [40,41,42]. Data was acquired using DigiData 1550B digitizers and analyzed with pClamp 10.7 software (Molecular Devices) or MiniAnalysis (Bluecell Co., Seoul, Korea). Low resistance glass pipettes (~4 MΩ) were filled with internal solution, for principal neuron recordings the solution contained (in mM): 90 K-gluconate, 50 KCl, 10 HEPES, 0.2 EGTA, 2 MgCl₂, 4 Mg-ATP, 0.3Na₂GTP and 10 Na₂-phosphocreatine; for interneuron recordings, the internal solution contained (mM): 140 K-gluconate, 5 KCl, 10 HEPES, 0.2 EGTA, 2 MgCl₂, 4 Mg-ATP, 0.3 Na₂GTP, and 10 Na₂-phosphocreatine (pH 7.2 with KOH). sIPSCs were recorded in the presence of AMPA receptor antagonist CNQX (10 µM). sIPSCs were measured and analyzed similar to our previous studies [43,44]. For 5-HT receptor blockade experiments, slices were incubated in ketanserin (10 µM) or volinanserin (150 nM). Interneurons were identified by mRuby expression and a high firing rate in current clamp after a brief depolarization.

RNAscope

Histological preparation and RNAscope in situ hybridization were performed as previously described [40]. For RNAscope, fluorescent probes targetingMus musculus mRNA forVglut1,Vgat,Htr2a, Htr2b, andHtr2c (Advanced Cell Diagnostics Inc; Hayward, CA) were incubated according to the manufacturer’s directions and imaged with a TCS SP8 confocal microscope (Leica) and quantified using Imaris (Bitplane, Zürich, Switzerland).

Statistics

All data are presented as the mean ± SEM. Tone blocks during fear conditioning were analyzed using two-way RM ANOVA with Holm-Sidakpost-hoc analysis. All other tests were analyzed with a one-way ANOVA, unpaired Student’st test, Welch’st test (unequal group variance), Mann-Whitney test or pairedt tests for before and after drug effects. Results were considered significant atp < 0.05.

Characterization of 4-OH-DiPT at human serotoninergic GPCRome

To determine the pharmacological profile for 4-OH-DiPT (Fig.1a), we used a BRET-based method to interrogate both G protein dissociation and β-arrestin2 recruitment in vitro at twelve human serotonin GPCRs, as described recently [38] (Fig. 1b, c; Supplementary Fig. 1). 4-OH-DiPT had the greatest G protein dissociation agonist activity at the 5-HT₂ receptor family with nearly full agonist activity at 5-HT_2A (EC₅₀ = 68.5 nM; 93% of 5-HT), but also activated 5-HT_2B and 5-HT_2C receptors to similar degrees (Fig.1d). 4-OH-DiPT showed similar activity (potency and efficacy) at both human and mouse 5-HT₂ receptors (Supplementary Fig. 2). Additionally, DiPT, which lacks the 4-hydroxy group of 4-OH-DiPT, did not show substantial differences on potency and activation at human or mouse 5-HT₂ receptors compared to 4-OH-DiPT (Supplementary Fig. 3). At 5-HT₁ G_i/o-coupled receptor subtypes, 4-OH-DiPT demonstrated weaker potency (EC₅₀: 496–1147 nM) and partial agonism, except at 5-HT_1B (E_MAX = 95%). At another G_i/o-coupled serotonin receptor, 5-HT_5A, 4-OH-DiPT was slightly more potent (EC₅₀ = 243 nM), but still demonstrated partial activation (E_MAX = 56%). At the G_s-coupled serotonin GPCRs, 4-OH-DiPT demonstrated little agonist activity at 5-HT_4/7a, with relatively weak partial agonism at 5-HT₆ (EC₅₀ = 697 nM, E_MAX = 55%). A summary of activity across the human serotonin GPCRome is shown as a heat map compared to 5-HT (Fig. 1e). GPCRs also recruit β-arrestin2, leading to receptor desensitization and internalization [45]. Biased signaling has been postulated as important for facets of psychedelic effects [38] especially for LSD [46]. 4-OH-DiPT induced β-arrestin2 recruitment to the 5-HT₂ receptors but indicated a lack of strong bias between Gq and β-arrestin2 activity relative to 5-HT (Supplementary Fig. 1a–f). Overall, 4-OH-DiPT is a potent balanced agonist at 5-HT_2A and 5-HT_2B with lower potency at 5-HT_2C and has strong selectivity for 5-HT₂ receptors over the other serotonin subtypes.

a Chemical structure of 5-HT and 4-OH-DiPT.b,c Activation of the 12 human 5-HT GPCRs by 5-HT (b) and 4-OH-DiPT (c) as assessed by G protein dissociation BRET assays. Data represent mean ± s.e.m. from at least three independent experiments performed in triplicate, all normalized to 5-HT.d Graphs of G protein dissociation for the top 3 targets of 4-OH-DiPT.e Heatmap showing relative agonist activity (log E_MAX/EC₅₀) comparing 5-HT with 4-OH-DiPT at the 12 5-HT GPCRs measuring G protein dissociation at 37 °C at 60 min (Supplementary Table 1).

Full size image

4-OH-DiPT suppresses freezing during fear extinction training

5-HT_2A receptor blockade enhances the expression of conditioned fear, while 5-HT_2A receptor agonism reduces conditioned fear expression [47,48,49,50,51,52]. As 4-OH-DiPT is a highly potent agonist of the 5-HT_2A receptor, we tested how it would alter fear extinction learning. Mice were first subjected to fear conditioning (FC) and were placed in a conditioning chamber and presented with five tones (20 s, 85 dB, 4 kHz) each co-terminating with a mild foot shock (0.5 mA, 1 s). Mice rapidly developed freezing behavior during the cue presentations (Fig. 2b, Supplementary Figs. 4,5b) and exhibited higher rates of spontaneous freezing after FC compared to baseline. One day after FC we tested whether injection of 4-OH-DiPT (3 mg/kg i.p.) altered freezing responses to fear cues during extinction training in a novel context (“Context B”). A dose of 3 mg/kg 4-OH-DiPT was chosen as it produced a higher head twitch response in mice compared to higher and lower doses, and the head twitch response waned by ~30 min [22]. We confirmed these findings in the present study (Supplementary Fig. 4a, b). Next, we examined how cue-induced freezing responses would be altered by exposing mice to fear extinction training during the peak (5 minutes) or the tail (30 minutes) of the head twitch response induced by i.p. injection of 3 mg/kg 4-OH-DiPT. We chose to use female mice in the initial experiments as they exhibit stronger responses to negative stimuli [53]. We found that injection of 3 mg/kg 4-OH-DiPT 30 min before extinction training produced a marginal effect (p = 0.06) on freezing responses to tones (20 tones, 60–80 s interval presented in a random manner) (Supplementary Fig. 4c) but produced a nearly complete suppression (p < 0.001) of freezing when given 5 min before extinction training (Supplementary Fig. 4d). These results indicate that 4-OH-DiPT has a rapid onset, but its fear suppressing effects are short-lasting. In subsequent studies, the effects of 4-OH-DiPT on fear extinction training were examined 5 minutes after i.p. injections. In addition, the interval between tones was shortened from 60–80 s to 5 s to capture the short-lasting peak effect of 4-OH-DiPT on extinction training.

a Timeline of behavioral and electrophysiological experiments; the arrow denotes administration of 4-OH-DiPT or vehicle.b Freezing responses to five 20 s tones co-terminating with a 0.5 mA foot shock were not significantly different between treatment groups (two-way RM ANOVA:F_2,88 = 2.017,p = 0.157).c Freezing responses to 20 tones with 5 sec intervals in a new context 24 h later. Female mice that received 4-OH-DiPT froze significantly less than vehicle controls (two-way RM ANOVA: 4-OH-DiPT dose,F_2,198 = 11.446,p < 0.001; tone block,F_9,198 = 7.549,p < 0.001; dose x tone,F_18,198 = 2.784,p < 0.001; vehicle vs. 1 mg/kg, red*; vehicle vs. 3 mg/kg, grey*).d Freezing responses to tone 24 h after extinction training. 4-OH-DiPT treatment during extinction training did not affect the overall level of freezing during the extinction test. 1 and 3 mg/kg 4-OH-DiPT treated mice had significant reduction in freezing over the entire session while vehicle treated mice did not. vehicle treated mice had no reduction in freezing over time (two-way RM ANOVA: 4-OH-DiPT dose,F_2,88 = 0.568,p = 0.575; Tone block,F_4,88 = 13.302,p < 0.001; dose x tone,F_8,88 = 2.121,p < 0.042; Tone block within 1 mg/kg (1 vs. 5,p < 0.001; 1 vs. 4,p < 0.001; 1 vs. 3,p < 0.001, red bar); Tone block within 3 mg/kg (1 vs. 3,p = 0.039, black bar).N = 8-9 mice per group.e 4-OH-DiPT treated mice showed no difference in the time spent in the center zone of an open field (one-way ANOVA on ranks:p = 0.704) or in the total distance traveled (one-way ANOVA:F_2,22 = 3.247,p > 0.05.f Mice treated with 3 mg/kg 4-OH-DiPT spent more time in the light side of the light dark box (one-way ANOVA:F_2,22 = 5.890,p = 0.009; Vehicle vs. 3 mg/kg,p = 0.007). Mice treated with 1 mg/kg 4-OH-DiPT had more entries into the light side of the light-dark box compared to mice treated with vehicle or 3 mg/kg 4-OH-DiPT (one-way ANOVA:F_2,22 = 5.466,p = 0.012; vehicle vs. 1 mg/kg,p = 0.022; 3 mg/kg vs. 1 mg/kg,p = 0.021).g Mice treated with 1 or 3 mg/kg 4-OH-DiPT spent more time in the center and open arms of the elevated plus maze compared to controls (one-way ANOVA:F_2,22 = 23.454,p < 0.001; vehicle vs. 1 mg/kg,p < 0.001, vehicle vs. 3 mg/kg,p < 0.05, 1 mg/kg vs. 3 mg/kg,p < 0.001) and mice treated with 1 mg/kg 4-OH-DiPT had more entries than controls or mice receiving 3 mg/kg 4-OH-DiPT (Kruskal Wallis ANOVA on ranks:p < 0.01; vehicle vs. 1 mg/kg,p < 0.01, 1 mg/kg vs 3 mg/kg,p < 0.05).h Mice treated with 3 mg/kg 4-OH-DiPT had a reduced latency to feed in a novel environment compared to mice receiving vehicle or 1 mg/kg 4-OH-DiPT (one-way ANOVA:F_2,22 = 5.357,p = 0.013; vehicle vs. 3 mg/kg,p = 0.021; 3 mg/kg vs. 1 mg/kg,p = 0.029). However, mice treated with 1 mg/kg 4-OH-DiPT had a longer latency to feed in the home cage compared to mice treated with 3 mg/kg 4-OH-DiPT (one-way ANOVA on ranks:p = 0.027; 3 mg/kg vs. 1 mg/kg,p = 0.023).N = 8-9 mice per group. Data are presented as mean ± s.e.m.

Full size image

Next, we tested if 4-OH-DiPT produced a dose-dependent effect on cue-associated freezing in male and female mice. Twenty-four hours after FC, mice were i.p. injected with either 1 mg/kg or 3 mg/kg 4-OH-DiPT or vehicle and placed in the home cage for 5 min. The 4-OH-DiPT-treated mice demonstrated a significant overall reduction in cue-associated freezing compared to the vehicle control group in a dose-dependent manner. Both male and female mice that received high dose (3 mg/kg) 4-OH-DiPT showed a significant reduction in cue-induced freezing (Fig. 2c and Supplementary Fig. 5c). Injection of low dose (1 mg/kg) 4-OH-DiPT significantly reduced freezing during extinction cues in female mice (Fig. 2c), but not male mice (Supplementary Fig. 5c). Thus, there was a sex and dose-dependent effect of 4-OH-DiPT on cue-induced freezing during extinction training.

The fear-suppressive effects of 4-OH-DiPT during extinction training might also affect how mice respond to fear cues after the acute effects had worn off. Therefore, we tested if 4-OH-DiPT induced altered fear responses 24 h later by presenting the mice with 10 tones in context B (“Test” session). We found that neither males nor females had a significant change in the overall level of freezing between 4-OH-DiPT and vehicle treatment. However, there was a significant interaction effect between dose and freezing as female mice that received either high or low dose 4-OH-DiPT showed a significant reduction in freezing between the first and last tone block, while control mice did not (Fig. 2d, Supplementary Fig. 5d). Together these data suggest 4-OH-DiPT acutely reduces fear expression in both male and female mice, but only enhances post-training extinction learning in female mice.

Fear conditioning precipitates avoidance behavior that might be interpreted as “anxiety-like”, and unlike freezing behavior, the avoidance behavior is often resistant to extinction [54,55,56]. Mice avoid areas with high visibility including brightly lit areas [57] and open spaces [58]. Chronic or acute treatments with anxiolytics or antidepressants reduces avoidance behavior [59]. Since 4-OH-DiPT suppressed freezing responses to learned fear, we tested if 4-OH-DiPT-treated mice would also exhibit reduced avoidance behavior in the days following extinction recall. In the open-field test, 4-OH-DiPT treated mice showed no differences in the total distance traveled or time spent in the center zone compared to vehicle treated controls (Fig. 2e). Female mice treated with high dose 4-OH-DiPT spent greater time in light side of the light dark box and the open arm and center areas of the elevated plus maze compared to controls (Fig. 2f, g). Female mice treated with low dose 4-OH-DiPT had more entries to the light side of the light dark box (Fig. 2f) and spent more time in the open and center regions of the elevated plus maze than mice treated with high dose 4-OH-DiPT or vehicle (Fig. 2g).

In the novelty-suppressed feeding (NSF) test, an increased latency to feed in the novel environment is indicative of anxiety-like behavior, as it is sensitive to anxiolytic or chronic antidepressant treatment [60,61]. Mice were food restricted for 24 h prior to being placed in an open field with food pellets in the center. Mice face the conflict between engaging in feeding and avoiding a novel environment. We found that female mice treated with high dose 4-OH-DiPT had a marked reduction in the latency to feed compared to vehicle-treated mice. The reduced latency to feed could not be due to differences in the level of hunger as there was no difference in the home cage latency to feed between the two groups (Fig. 2h). In contrast, female mice treated with low dose 4-OH-DiPT showed no difference between control groups in either measure (Fig. 2h). In summary, compared to controls female mice that received high dose 4-OH-DiPT prior to extinction training had less avoidance compared to controls in the novelty suppressed feeding test, light dark box, and elevated plus maze while female mice treated with low dose 4-OH-DiPT prior to extinction training only had reduced avoidance in the elevated plus maze. Surprisingly, male mice treated with either high or low dose 4-OH-DiPT showed no significant differences in any measures of avoidance behavior compared to controls (Supplementary Fig. 5e–h).

4-OH-DiPT modulates BLA inhibition via 5-HT_2A activation

The BLA is critically involved in associative-fear learning and fear expression [62]. Mice that received 4-OH-DiPT before extinction training spent less time freezing during cue presentations, which could be due to enhanced inhibition of BLA principal neurons. We next tested whether and how 4-OH-DiPT affected inhibitory synaptic transmission in the BLA by recording spontaneous inhibitory postsynaptic currents (sIPSCs) in BLA principal neurons. Brain slices were prepared from adult naïve mice. BLA principal neurons were identified by the pyramidal shape and spike frequency adaptation in response to the injection of depolarizing current steps (Fig. 3a). sIPSCs were recorded at a holding potential of −70 mV, in the presence in the presence of the AMPA receptor antagonist CNQX. Bath perfusion of 4-OH-DiPT (20 µM) produced a marked increase in the amplitude (Fig. 3b, c, e) and frequency (Fig. 3b, d, f) of sIPSCs in BLA principal neurons. The 4-OH-DiPT-induced potentiation of sIPSCs is likely mediated by the 5-HT_2A receptor because it was blocked by 5-HT_2A receptor antagonists ketanserin (10 µM; Fig. 3g, h) and volinanserin (150 nM; Fig. 3i, j).

a Example trace of depolarization-induced spike adaptation in a BLA principal neuron.b Example trace of sIPSCs recorded before and after bath perfusion of 20 µM 4-OH-DiPT in a BLA principal neuron. Neurons were voltage clamped at −70 mV and incubated in 5 µM CNQX for at least 15 min prior to recording.c,d Cumulative fraction of IPSC amplitude and frequency from the above neuron.e,f Bath perfusion of 4-OH-DiPT increased the amplitude (t₆ = −8.618,p = 0.000134) and frequency (Shapiro-Wilk paired t-test: Z-statistic = 2.366,p = 0.016) of sIPSCs in BLA principal neurons,n = 7 neurons.g Example traces BLA principal neuron incubated in ketanserin (top) followed by bath perfusion of 4-OH-DiPT.h Bath perfusion of 4-OH-DiPT following ketanserin incubation does not increase amplitude (left,p = 0.180) or frequency (right,p = 0.156) of sIPSCs in BLA principal neurons,n = 7 neurons.i Example trace of a BLA principal neuron incubated in volinanserin followed by bath perfusion of 4-OH-DiPT.j Bath perfusion of 4-OH-DiPT following volinanserin incubation does not enhance amplitude (p = 0.938) or frequency (p = 0.820) of sIPSCs in BLA principal neurons,n = 7 neurons. Data are presented as mean ± s.e.m.

Full size image

Most sIPSCs are driven by action potential firing in GABA interneurons. One possibility is that 4-OH-DiPT enhanced sIPSCs by activating GABA interneurons in the BLA through agonism of 5-HT_2A. To test this hypothesis, we determined whether 4-OH-DiPT affected action potential firing in BLA interneurons. As the number of principal neurons (~80%) is far more than the number of GABA interneurons (~20%) in the BLA [63], we labeled GABA interneurons by stereotaxic injections of AAV9-mDlx-mRuby into the BLA (Fig. 4a). mDlx selectively labels GABA neurons in forebrain regions [64]. After 7–14 days of viral expression, BLA slices were prepared and whole-cell current clamp recordings were made from mRuby-expressing neurons. A brief (2 s) puff of 20 µM of 4-OH-DiPT with a pressure ejector induced membrane depolarization and action potentials from GABA interneurons, which were blocked by the 5-HT_2A receptor antagonist ketanserin (Fig. 4b, c). Together, these results suggest that 4-OH-DiPT depolarizes BLA interneurons via the 5-HT_2A receptor to enhance action potential firing, which leads to enhanced sIPSCs in BLA principal neurons.

a Injection site of AAV9-mDlx-mRuby in the BLA.b Example trace in current clamp of an mRuby⁺ BLA neuron following a brief puff (2 s) of 4-OH-DiPT. Incubation in ketanserin prevented 4-OH-DiPT induced AP firing (n = 9 neurons, bottom).c Left: AP frequency twenty seconds after 4-OH-DiPT administration was compared to the 20 preceding seconds. 4-OH-DiPT administration increased AP firing frequency of mRuby+ BLA neurons (t₈ = −4.015,p = 0. 00387),n = 8 neurons. Right: 4-OH-DiPT induced depolarization was blocked by ketanserin (t_8.139 = −5.666,p = 0.000444). Data are presented as mean ± s.e.m.

Full size image

Our electrophysiological studies indicate that the 5-HT_2A receptors are expressed on GABA interneurons in the BLA. We performed RNAscope to identify the presence ofHtr2a, Htr2b, and Htr2c mRNA within excitatory and inhibitory BLA neurons (Fig. 5a). We used common isoforms of vesicular transporters to serve as cell-type specific markers of excitatory principal neurons (vesicular glutamate transporter 1,Vglut1) and inhibitory (vesicular GABA transporter,Vgat) neurons.Vglut1 was present in approximately 82% of BLA neurons, whileVgat was present in approximately 18% of neurons (Fig. 5b).Htr2a mRNA puncta were predominately expressed in GABA interneurons as determined by colocalization ofHtr2a mRNA withVgat mRNA. In contrast,Htr2a mRNA puncta were rarely expressed inVglut1-positive excitatory principal neurons in the BLA (Fig. 5a). In addition, BLA GABA interneurons expressed a significantly higherHtr2a mRNA puncta density than BLA principal neurons (Fig. 5c).Htr2c puncta density was similar inVglut1-positive neurons compared to BLA GABA interneurons (Supplementary Fig. 6a, b). However,Htr2b was not detected in the BLA (Supplementary Fig. 7a) but could be detected in the hippocampus (Supplementary Fig. 7b). These results lend further support that 4-OH-DiPT activates the 5-HT_2A receptor expressed on BLA interneurons to enhance action potential firing.

aVglut1 (yellow),VgatI (green), andHtr2a (red) are expressed in the BLA. White arrows denote areas with highHtr2a expression that overlap withVgat.b Total % of BLA neurons as indicated by expression ofVgat orVglut1 colocalized with DAPI (blue).cHtr2a puncta density is significantly higher inVgat neurons compared toVglut1 neurons (p = 0.00055,n = 4 BLA sections from 2 mice). Data are presented as mean ± s.e.m.

Full size image

In this study, we first used BRET to determine the interaction of 4-OH-DiPT with all 12 human serotonin GPCRs and found that 4-OH-DiPT is a nearly a full agonist for 5-HT_2A, 5-HT_2B, and 5-HT_2C receptor subtypes. Notably, we found that 4-OH-DiPT had similar activity at the human and mouse 5-HT₂ receptor subtypes. As it acts as a psychedelic, we next investigated if 4-OH-DiPT altered fear extinction behavior. The affective symptoms of PTSD are thought to manifest from a failure to extinguish responses to a traumatic memory [28,29,30], which can be treated by extinction-based exposure therapy [3,4]. We found that 4-OH-DiPT reduced the fear responses of mice during extinction training in a dose- and sex-dependent manner. In an investigation to dissect the circuit mechanism responsible, we found that 4-OH-DiPT activates interneurons to enhance GABAergic inhibition onto principal neurons in the BLA. Together, our results provide a comprehensive pharmacological profile of 4-OH-DiPT and reveal a potential mechanism for its effects on fear suppression.

4-OH-DiPT has been shown to activate 5-HT_2A in a calcium flux assay [22] but polypharmacological activity across the rest of the 5-HT receptors have not been fully investigated. Ligand binding studies have shown that 4-OH-DiPT has a high affinity for 5-HT₂ receptors and a low affinity for the 5-HT_1A receptor [65]. However, radioligand binding cannot determine receptor activation efficacy or signaling bias [66]. Direct G protein dissociation via BRET permits interrogation of protein-protein interactions [67], including G protein signaling and β-arrestin2 recruitment at GPCRs [38], which can be used to determine receptor signaling efficacy and pathway-specific bias. We found that 4-OH-DiPT had partial agonist activity at most 5-HT GPCRs, but none showed a bias for G protein or β-arrestin2 recruitment. However, nearly full agonism (E_max 90–100% of serotonin activity) was observed for 5-HT_2A, 5-HT_2B, 5-HT_2C, and 5-HT_1B. In humans, 5-HT_2A receptor occupancy by psilocin is positively correlated with self-reported subjective effects, such as mystical experiences [25]. The degree of mystical experiences induced by moderate to high doses of psilocybin predict efficacy for reducing anxiety and depressive behavior in terminal cancer patients and nicotine craving in smokers [16]. 4-OH-DiPT activates the 5-HT_2A receptor to a stronger degree than psilocin and produces a higher maximal head twitch response in mice [22]. Together these data indicate that 4-OH-DiPT is a strong 5-HT_2A receptor agonist that would represent an exciting alternative to psilocybin for the treatment of mental disorders.

Administration of 4-OH-DiPT prior to fear extinction produced a significant reduction in freezing during extinction training in both male and female mice. Systemic injections of either psilocybin, TCB-2, or 25CN-NBOH in mice reduced fear expression when injected shortly before fear recall; these effects were blocked by pretreatment with volinanserin, thus providing strong evidence that fear expression is reduced by 5-HT_2A receptor activation [50]. Consistent with this idea, systemic administration of DOI immediately prior to fear extinction reduces freezing during extinction and recent extinction recall, but not remote recall [48]. In the same study, microinfusion of volinanserin into the BLA reduced the effect of systemic administration of DOI on fear expression, while microinfusion of DOI into the BLA recapitulated the effects of systemic injection. Thus, 5-HT_2A receptor agonists, such as 4-OH-DiPT, are likely to act within the BLA to reduce conditioned fear expression.

Fear conditioning leads to a reduction of GABAergic inhibition in the BLA, as shown by a reduction of extracellular GABA levels [68], GABA receptor expression [69] and transcription of the GABA synthesis enzyme glutamic acid decarboxylase 65 (GAD65) [70], while intra-BLA infusion of a GABA_A agonist reduces fear expression [71]. We show that 4-OH-DiPT significantly enhanced the frequency and amplitude of sIPSCs in BLA principal neurons ex vivo, and these effects were blocked by the 5-HT_2A receptor antagonists ketanserin and volinanserin. 4-OH-DiPT and other classical psychedelics share structural similarities to serotonin which requires 5-HT_2A receptor activation to enhance sIPSCs in the BLA [72,73]. Activation of BLA interneurons drives IPSCs on BLA principal neurons [74]. Application of 4-OH-DiPT induced robust depolarization and high frequency AP firing in BLA interneurons and could be blocked by preincubation with a 5-HT_2A antagonist. RNAscope showed thatHtr2a mRNA was expressed predominately in BLA GABA interneurons in high density as it colocalized with GABA neuron markerVgat mRNA. In addition,Htr2c mRNA is expressed in both GABA interneurons and principal neurons, whileHtr2b is absent in the BLA. Given that the effects 4-OH-DiPT on BLA interneuron firing was blocked by a selective 5-HT_2A receptor antagonist, it is likely that 4-OH-DiPT reduces freezing to fear cues by activating the 5-HT_2A receptor on GABAergic interneurons, resulting in enhanced BLA inhibition and reduced fear expression.

While several studies have demonstrated that psychedelic treatment may reduce the expression of learned fear [48,49,51,52], to our knowledge, no previous studies have directly addressed whether and how extinction-paired treatment with psychedelics can influence affective state. The increased latency to feed in a novel environment is sensitive to chronic treatment with anxiolytic or antidepressant drugs and could be interpreted as “anxiety-like” behavior [60,61]. On the other hand, the open field test, light-dark box test, and elevated plus maze measure avoidance when free exploration conflicts with high visibility. Female mice that received 4-OH-DiPT prior to extinction training showed reduced avoidance to high visibility in the light-dark box, the novelty suppressed feeding and the elevated plus maze, which could be interpreted as an “anxiolytic-like” effect. In contrast, male mice given this treatment did not show significant differences in these behavioral tests. Taken together, these results suggest that administration of 4-OH-DiPT prior to extinction training produces sex-dependent effects on anxiety-like behavior. However, it is important to note that we did not determine if avoidance behavior returned to pre-fear conditioned levels. 4-OH-DiPT showed greater potency for reducing freezing responses to conditioned cues in female mice than that in male mice, which could contribute to sex differences in avoidance behavior. Future studies could address the circuit mechanism underlying these sex differences.

A main goal of exposure therapy for PTSD is to reduce affective symptoms by repeatedly presenting patients with trauma reminders in a safe setting [75]. After the continued presentation of trauma reminders, the patient learns that they are safe and fear responses dissipate. Pharmacological agents that enhance fear extinction could improve the effectiveness of behavioral therapies, ultimately reducing the economic burden on patients and healthcare systems. Overall, our study has provided a detailed characterization of the molecular profile, circuit action, and behavioral effects of 4-OH-DiPT. It is our hope that future studies will assess the safety profile of 4-OH-DiPT in humans, as it may possess therapeutic properties without requiring prolonged supervision.

This work was supported by National Institutes of Health Grants R01MH121454, DA047269 and R01DA035217 (to QSL), R35GM133421 (to JDM), and F31DA054759 (to VF). TJK is a member of the Medical Scientist Training Program at MCW, which is partially supported by a training grant from NIGMS T32-GM080202. It was also partially funded through the Research and Education Initiative Fund, a component of the Advancing a Healthier Wisconsin endowment at MCW.

Authors and Affiliations

1. Department of Pharmacology and Toxicology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA

   Thomas J. Kelly, Lianwei Mu, Xiaojie Liu, Ying Hu, Vladislav Friedman, Hao Yu, Wantang Su & Qing-song Liu

2. Department of Cell Biology, Neurobiology and Anatomy, Medical College of Wisconsin, Milwaukee, WI, 53226, USA

   Emma M. Bonniwell & John D. McCorvy

Contributions

TJK, QSL, and JDM designed the experiments. TJK, EB, LM, HY, XJ, YH, WS, and VF performed the experiments. TJK wrote the initial draft. TJK, QSL, JDM, and EB wrote the revised manuscript. VF edited the manuscript. JDM supervised the 5-HT receptor BRET experiments and were conducted by EMB.

Corresponding authors

Correspondence toJohn D. McCorvy orQing-song Liu.

Competing interests

The authors declare no competing interests.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions