One Dose of a Psychedelic Drug Restores Brain Cell Connections in Thinking Areas of Mice Exposed to Alcohol Before Birth

All procedures and use of animals were approved by the University of Michigan Institutional Animal Care and Use Committee. Sample sizes were chosen based on previous publications (Ly et al. 2018; Nardou et al. 2019, 2023; Revenga et al. 2021; Shao et al. 2021; Ekins et al. 2023, 2025; Vargas et al. 2023). Eighty‐four total neurons from 19 (seven female, 12 male) adolescent mice, postnatal days 29–49 (P29–49), were used for this study. Experimental groups undergoing the PAE protocol and subsequent saline or psychedelic injections were all on a C57BL/6 background. Control data consisted of cell type‐, region‐, layer‐, and age‐matched neurons derived from control mice used in two of our recent studies (Ekins et al. 2023, 2025). These control data included Ai14, Ai32, Scnn1a‐Cre, Pvalb‐Cre, Grp‐Cre, Kj319‐Cre, and Htr2a‐flox, all on a C57BL/6 background. Notably, Htr2a‐flox mice were used only as controls (without Cre), as in our recent publication (Ekins et al. 2025). Additional details about these CRISPR‐Cas9‐generated Htr2a floxed mice are available in our recent publication (Ekins et al. 2025). All other lines were obtained from The Jackson Laboratory and/or bred in‐house and backcrossed to C57BL/6 mice. For additional comparisons with a limited subset of control data, genotypes were matched to those used in the PAE experiments (C57BL/6 and Pvalb‐Cre).

The PAE protocol used here was based on an established protocol using chronic alcohol administration to model heavy alcohol use during pregnancy, and prior studies using similar protocols have reported blood alcohol concentrations of approximately 80–150 mg/dL per day (Tajuddin and Druse 1989; Kim et al. 1997; Sickmann et al. 2014; Skorput et al. 2015, 2019; Skorput and Yeh 2016). Throughout pregnancy, drinking water was replaced with a 6.6% ethanol solution (Tajuddin and Druse 1989; Kim et al. 1997). At birth, the ethanol solution was removed, and standard drinking water was returned to the cage. Experimentation began in adolescence (P29–49).

25CN‐NBOH, a selective 5‐HT_2AR agonist (Fantegrossi et al. 2015; Buchborn et al. 2018, 2020; Märcher‐Rørsted et al. 2021; Odland et al. 2021; Wang et al. 2025), was purchased from Tocris Bioscience and was dissolved in sterile water before mixing with sterile saline for in vivo injections. Light sonication was used to assist solubilization in water. DMSO was avoided as a solvent as it is known to alter intrinsic electrophysiological properties of neocortical and hippocampal neurons at doses far lower than those regularly used in neuroscience experiments (Tamagnini et al. 2014). 25CN‐NBOH was administered via intraperitoneal injection at 10 mg/kg, selected based on our previous study, which found a long‐lasting enhancement in reversal learning after this dose (Brouns et al. 2025). After 24–48 h, electrophysiology experiments were conducted. The control group for this study was a region‐matched and age‐matched subset of control neurons from two of our recent datasets—one published (Ekins et al. 2025) and one preprinted (Ekins et al. 2023).

Mice were deeply anesthetized with isoflurane before decapitation. Brains were dissected out in ice‐cold sucrose‐substituted ACSF, saturated with 95% O₂ and 5% CO₂, and containing the following (in mM): 3 KCl, 1.25 NaH₂PO₄, 26 NaHCO₃, 10 dextrose, 234 sucrose, 0.2 CaCl₂, and 4 MgSO₄. Coronal slices (300 µm thick) were prepared using a VT1200 vibratome (Leica) and placed in a high‐magnesium ACSF solution at 32°C for 30 min, after which the slices rested at room temperature for at least 30 more minutes prior to recording. During experiments, slices were incubated in a recording chamber with physiological temperature (32°C) ACSF (126 mM NaCl, 1.25 mM NaH₂PO₄, 26 mM NaHCO₃, 3 mM KCl, 10 mM dextrose, 1.20 mM CaCl₂, and 1 mM MgSO₄) perfused at a rate of 3 mL/min. Slices containing the anterior cingulate cortex were obtained at anteroposterior coordinates relative to the bregma from 2.1 to −0.5 mm.

Neurons were visualized on an Olympus BX51WI microscope with an Olympus 60x water immersion lens and Andor Neo sCMOS camera (Oxford Instruments, Abingdon, Oxfordshire, UK). Patch electrodes were pulled from borosilicate glass (Sutter Instruments) and had diameters of 2–4 µm and resistances of 3–5 MΩ. Internal solution was potassium gluconate based and contained (in mM): 130 K‐gluconate, 0.6 EGTA, 10 HEPES, 2 Na₂ATP, 0.3 NaGTP, 6 KCl, 3 MgCl₂, and 0.5% biocytin (calculated chloride reversal potential of −68 mV; pH of 7.25; osmolarity of 290 mOsm). Pipette capacitance compensation and bridge balance were applied; recordings were not corrected post hoc for liquid junction potential. When conducting intracellular recordings in brain slice electrophysiology experiments, not all neurons in the brain slice are healthy or viable, which requires extensive quality control (Gouwens et al. 2019, 2020; Scala et al. 2019; Vormstein‐Schneider et al. 2020). To ensure high‐quality intracellular recordings, the following standard quality control metrics were used (Tricoire et al. 2011; Brennan et al. 2020, 2021; D'Amour et al. 2020; Ekins et al. 2020, 2023, 2025; Pelkey et al. 2020; Shao et al. 2021; Jedrasiak‐Cape et al. 2025): cells were excluded if baseline resting membrane potential was more depolarized than −50 mV or if uncompensated series resistance (R_s) was >35 MΩ. All whole‐cell recordings were conducted with a MultiClamp 700B amplifier and digitized at 20 kHz with a Digidata 1550B (Molecular Devices) for acquisition on a computer equipped with pClamp 10.7 software (Molecular Devices). We included only putative intratelencephalic‐type pyramidal cells in the study, as identified by co‐expression of a regular spiking firing pattern and a physiologic index less than 5 (Baker et al. 2018; Gulledge 2024). Biocytin enabled post hoc streptavidin labeling and confocal imaging to confirm pyramidal cell morphology.

Current‐clamp experiments were conducted similarly to our previous studies (Cruikshank et al. 2012; Brennan et al. 2020, 2021; D'Amour et al. 2020; Ekins et al. 2020, 2023, 2025; Pelkey et al. 2020; Jedrasiak‐Cape et al. 2025). Membrane potentials were biased to −65 mV at the start of each sweep. Firing patterns were investigated using a series of 1‐s‐long current injections (step size 50 pA) up to 400 pA or until depolarization block was induced. Firing frequency was the number of spikes that occurred for each 1‐s‐long current step. Spikes that did not have a peak voltage reaching at least −10 mV were not counted. Frequency–current (f–I gain) was defined as the slope of a linear regression on firing frequency for each current injection from 0 pA to the injection at which maximum firing frequency was attained, or 400 pA, whichever occurred first. Total spike output was the sum of all spikes until maximum firing frequency or 400 pA. We analyzed intrinsic cell properties using Python (version 3.10.4) code with pyABF, NumPy, pandas, and Matplotlib as dependencies, as previously described (Ekins et al. 2023, 2025). For spike properties, we defined threshold as the voltage where the rise slope (dV/dt) reached 10 mV/ms. Amplitude was the difference between the threshold and the peak voltage. Half‐width was measured at the voltage corresponding to half‐maximal amplitude. Maximum rise slope was the maximum of dV/dt; maximum decay slope was the minimum of dV/dt. These properties were measured for the first spike evoked by the 1‐s current injection. Rheobase and latency to first spike were determined with a series of 2‐s‐long depolarizing current steps that were sufficient to evoke spiking (5–10 pA step size), repeated one to three times. Latency to first spike was determined by measuring the interval between the onset of the current step and the time at which the initial spike threshold was reached; rheobase was defined as the minimal current necessary to induce a spike during the 2‐s current step. Input resistance (R_in), input capacitance (C_in), and the membrane time constant (TC) were assessed with a series of small hyperpolarizing current steps (typically −10 to −20 pA, 20–40 sweeps). R_in was computed using the voltage difference between the maximum response during the current injection and the mean voltage during the 100 ms prior to the current injection, divided by the injected current. TC was computed by fitting an exponential decay curve to the voltage trace from the start of the current injection to the peak voltage deflection. Input (membrane) capacitance (C_in) was calculated using the relationship C_in = TC/R_in. Injection of hyperpolarizing current can produce a noticeable “sag,” which indicates the presence of a hyperpolarization‐activated cation current (I_h), which activates after the initial hyperpolarization peak. Sag ratio was computed as the maximum response amplitude divided by the mean response amplitude of the last 50 ms of the current injection. Response amplitudes were computed relative to the baseline voltage, defined as the mean of the 100 ms prior to the current injection. Physiology index was computed following previously described methods (Baker et al. 2018; Gulledge 2024).

Voltage‐clamp recordings were conducted at a holding potential of −70 mV, which yields approximately no driving force for chloride due to the internal solution composition (listed above). sEPSC recordings were taken in 30‐s sweeps with a brief hyperpolarizing test pulse (−5 mV, 250 ms) at the start to monitor R_s and R_in throughout the experiment. sEPSCs were detected using Easy Electrophysiology (version 2.6.1). sEPSCs were analyzed by an analyst blinded to experimental conditions in 30‐s sweeps (two to three analyzed per cell); the first second of each sweep was discarded because it contained the hyperpolarizing test pulse. Events were first identified with threshold‐based detection with the following settings: negative peak direction, 5‐ms local maximum period, 30‐ms decay search period, 8 pA threshold, 10‐ms search period, 1‐ms averaged baseline, and curved baseline and threshold. To eliminate false positives, events were manually inspected, and noise events were rejected. Frequency was determined for each sweep by dividing the total number of events per sweep by 29 s; amplitude was computed by averaging event amplitudes within each sweep; decay TC was determined by fitting an exponential to the averaged event for each sweep.

To reveal cell morphology, biocytin (0.5%; Sigma–Aldrich) was added to the intracellular solution as in previous studies (Brennan et al. 2020, 2021; D'Amour et al. 2020; Ekins et al. 2020, 2023, 2025; Pelkey et al. 2020; Jedrasiak‐Cape et al. 2025). Biocytin slowly diffused into cells during the course of the recording (∼10 min). At the end of each recording, the patch pipette was slowly retracted from the cell to allow membrane resealing, then the slice was transferred to 4% PFA for overnight fixation. Afterward, each slice was washed in PBS and incubated for 24–48 h in streptavidin‐conjugated Alexa Fluor (488 or 647; Thermo Fisher Scientific) with 0.4% Triton X‐100 (Sigma–Aldrich). The fluorescent Nissl stain, NeuroTrace (435/455; Thermo Fisher Scientific), was added to the Alexa Fluor incubation step (at 1:200 dilution). After incubation, slices were washed in PBS, mounted on slides, and cover‐slipped using Fluoromount‐G Slides (Southern Biotech).

Statistical analyses were performed in GraphPad Prism (version 10.5.0). Additional details on statistical tests are provided throughout the figure legends and Table. S1

Adolescent PAE mice (Zhang 2004; Virtanen et al. 2018; Yang et al. 2018; Chen et al. 2025) received a single intraperitoneal injection of either saline or the psychedelic drug 25CN‐NBOH (NBOH; 10 mg/kg). At 24–48 h post injection, brain slices containing the PFC were prepared, and recordings were obtained from layer 5 (L5) pyramidal cells using whole‐cell patch‐clamp electrophysiology. Intrinsic excitability (membrane and firing properties) and synaptic input (sEPSCs) were assessed (Figure 1). Neuronal morphological reconstructions were done for representative neurons of each group, confirming the pyramidal morphology of recorded neurons (Figure S1).

First, we compared intrinsic excitability and synaptic properties in L5 pyramidal cells from PAE mice (treated only with saline) with those in control neurons from age‐matched mice. Relative to control neurons, PAE markedly decreased the intrinsic excitability of PFC neurons (Figure 2A). Frequency–current curves constructed from depolarizing step pulses revealed that PAE lowered f–I gain and reduced maximum firing frequency compared to control cells (Figure 2B,C). The total spike output of these neurons was also significantly lower (Figure 2C). Observed spike kinetics were consistent with alterations to voltage‐gated channels mediating spike generation: action potential maximum rise and fall slopes were reduced, and spikes were broader (Figure 2C). No changes were observed in spike amplitudes (Figure 2C), and the passive membrane properties did not differ significantly between groups (Figure S2).

Next, we examined the effects of PAE (again, only PAE mice treated with saline) on synaptic activity in PFC L5 pyramidal cells (Figure 2D). We found that PAE decreased the frequency of sEPSCs (Figure 2E), consistent with previous findings of reduced dendritic spine density and indicative of decreased synaptic connectivity (Reyes et al. 1983; Ferrer and Galofré 1987). The sEPSC amplitudes were unchanged, but kinetics were altered, as reflected by the longer sEPSC decay TCs in neurons from PAE mice (Figure 2E).

We next asked if treatment with a psychedelic drug (25CN‐NBOH; NBOH) would affect the electrophysiological properties of PFC L5 pyramidal cells compared to saline treatment in adolescent PAE mice. We found that a single dose of NBOH enhanced the intrinsic excitability of PFC L5 pyramidal cells relative to PAE with saline treatment (Figure 3A,B), significantly increasing maximum firing rates and total spike output compared to saline treatment (Figure 3C). While the f–I gain showed a trend toward an increase, this was not significant (Figure 3C). This result is comparable to previous studies’ findings that psychedelic treatment is without lasting impact on the intrinsic excitability of neocortical neurons from healthy mice (Cameron et al. 2023; Ekins et al. 2023, 2025) but partially normalizes intrinsic excitability alterations of neurons from chronically stressed mice (Lu et al. 2021). Spike waveform changes were selective: action potential half‐width was reduced, but spike rise and fall slopes were not significantly changed (Figure 3C). Again, no changes were observed in spike amplitudes (Figure 3C) or passive membrane properties (Figure S3). NBOH treatment of PAE mice significantly increased excitatory synaptic input (Figure 3D), boosting the sEPSC frequency compared to PAE mice given only a saline injection (Figure 3E). No significant changes were observed in either sEPSC amplitude or decay TC (Figure 3E).

Finally, to assess whether NBOH treatment corrects PAE‐induced synaptic input and intrinsic excitability deficits, we compared neurons from NBOH‐treated PAE mice to control neurons (Figure 4A,B). We found that while NBOH treatment increased intrinsic excitability compared to saline treatment in neurons from PAE mice, intrinsic excitability was only partially rescued compared to control neurons (Figure 4C). Maximum firing frequency and total spike output were not different from control neurons, but f–I gain was still lower (Figure 4C). Spike kinetics also showed residual abnormalities, with slower spike rise and fall slopes and broader half‐widths (Figure 4C). No changes were observed in spike amplitudes (Figure 4C) or passive membrane properties (Figure S4). In contrast, NBOH treatment restored excitatory synaptic input, as evidenced by the lack of statistically significant differences in sEPSC frequency, amplitude, or decay TC compared to control neurons (Figure 4D,E), and this pattern was unchanged in parallel analysis (Figure S5). Thus, NBOH treatment partially normalized intrinsic excitability deficits from PAE and restored excitatory synaptic drive to baseline levels (Table S1).