Sex-specific increased reactivity of the PVT and prolonged PVT→CeA circuit engagement following psilocin administration

Abstract

The psychedelic psilocybin has shown therapeutic potential, yet underlying neural mechanisms remain poorly understood. We investigated the impact of psilocin—the active metabolite of psilocybin—on basal activity and reactivity within the paraventricular nucleus of the thalamus (PVT) and PVT projections to central amygdala (CeA) in rats. Psilocin administration increased PVT c-Fos expression and selectively engaged PVT→CeA neurons in females, but not males. Psilocin enhanced PVT reactivity to an aversive air-puff stimulus, with effects primarily driven by passive responders. In PVT→CeA neurons, psilocin prevented time-dependent reductions in stimulus-evoked activity and maintained reactivity across timepoints in females but not males. The sustained engagement of PVT→CeA circuitry was driven by active responders. These findings identify sex-specific modulation of thalamic-limbic circuitry and behavior by psilocin, implicating PVT→CeA circuitry in the neural and behavioral effects of psychedelic compounds, advancing our understanding of how psychedelics modulate emotional brain circuits to further inform potential therapeutic mechanisms.

Psychedelics have therapeutic potential, yet their underlying mechanisms remain unclear. Here, authors show that psilocin increased PVT and PVT◄CeA activity in female rats, revealing sexspecific effects on thalamo-limbic circuitry and behavior.

Introduction

Psychedelic compounds such as psilocybin, and it’s active metabolite psilocin^1,2, have been shown to produce rapid-acting and persistent improvements in symptomology for a variety of different psychiatric disorders^3–13. Despite observed clinical outcomes, little is known in terms of the neurobiological alterations underlying these persistent effects. Clinical imaging studies suggest that psilocybin promotes alterations in neural functional connectivity^14–22, however less is known about the circuit-specific mechanisms underlying large-scale alterations in function.

The thalamus is a small structure within the diencephalon responsible for integrating signals from various brain regions within the central nervous system. Traditionally, the thalamus has been viewed as a passive relay station for sensory information to cortical regions, but has recently emerged as an integration hub for functional brain networks²³. The thalamus has inputs from nearly all regions of the brain and outputs to regions critical for affective processes including cortex, striatum, and amygdala^24–26. Though often thought of as a single brain region, the thalamus is comprised of several functionally distinct subnuclei²⁷. The paraventricular nucleus of the thalamus (PVT) is a midline thalamic nucleus heavily involved in appetitive and aversive behaviors and is considered an integral hub within the limbic system governing emotional processing^28,29 and goal directed actions to maintain homeostatic balance³⁰. The PVT sends glutamatergic and peptidergic projections to various limbic regions^31,32, including dense projections to the nucleus accumbens^33,34, and to a lesser degree to the BNST³⁴, central nucleus of the amygdala^34,35, prefrontal cortex³⁶, and hypothalamus³². The PVT has been shown to regulate threat responding, with projections to the central amygdala (CeA)³⁷, specifically the lateral capsule of the CeA, which has been shown to regulate freezing behavior and immediate anxiogenic effects in conditioned fear assays^38–40. Additionally, lesions of the PVT result in increased stress-induced c-Fos expression within CeA⁴¹, suggesting that the PVT may act as an upstream modulator of CeA reactivity to stressful stimuli.

Administration of the serotonergic psychedelic, lysergic acid diethylamide (LSD) increases functional connectivity within the thalamus and promotes connectivity changes between the thalamus and sensory networks and striatum⁴². Previous work showed that administration of psilocin, the active metabolite of psilocybin, promoted acute increases in reactivity to an aversive air-puff in the central amygdala (CeA) in females but not males⁴³. Given what is known about projections from PVT→CeA and their involvement in fear responding, the PVT may serve as an upstream mediator of previously observed effects. Recent studies have shown that psilocybin administration promotes increases in PVT c-Fos expression in mice^44,45 and that chronic, low-dose administration of psilocybin strengthened cortical connections to the PVT⁴⁶. Together these studies highlight the PVT as a critical region implicated in acute psychedelic drug action, however how psychedelics alter in-vivo functioning of the PVT remains poorly understood.

In the present study, we utilized in-vivo fiber photometry to assess the effects of psilocin administration on functional reactivity within the PVT and PVT→CeA circuitry. We found that psilocin administration produced increases in c-Fos expression in the PVT of females, but not males. We also found that psilocin produced acute increases in PVT reactivity in response to an aversive air-puff that persisted for 2 days following injection and was driven by passive responding females. Next, we utilized retrograde tracing and found increased c-Fos expression in PVT→CeA neurons following psilocin administration in females, but not males. Finally, using an intersectional viral approach, we found that psilocin administration reduced a time-dependent dampening of reactivity in PVT→CeA projections in females but not males and that this effect was driven by females employing active threat responding at baseline.

Results

The effects of psilocin administration on basal activity within the PVT

To assess the acute effects of psilocin administration on basal PVT activation, male and female Sprague Dawley rats received either vehicle or psilocin (2 mg/kg, s.c.) injection. Animals were then perfused 2hrs post injection for histological analysis. Changes in c-Fos expression were examined between groups. In males, there were no differences between the vehicle (mean: 101.3, SD: 15.4) and psilocin (mean: 119.4, SD: 71) treated animals (Fig. 1A). However, in females there was significantly higher cFos expression in psilocin (mean: 136.4, SD: 36.12) treated animals compared to vehicle control (mean: 59.92, SD: 30.08) group (t(8) = 3.636, p = 0.006; Fig. 1A). Interestingly, there were differences seen between vehicle treated males (mean: 101.3, SD: 15.4) and females (mean: 59.92, SD: 30.08) suggesting there may be sex differences in basal activation within the PVT (t(8) = 2.740, p = 0.02; Fig. 1B).

Fig. 1. Effects of psilocin on basal activity within the PVT.

A Top: representative images showing c-Fos (green) expression and histograms displaying the mean +/- S.E.M. of c-Fos expressing cells for each group in vehicle (n = 5) and psilocin (n = 5) treated male (p = 0.5927). Analysis comparing c-Fos expression was done using a two-tailed unpaired t-tests. Bottom: representative images showing c-Fos (green) expression and histograms displaying the mean +/- S.E.M. of c-Fos expressing cells for each group in vehicle (n = 5) and psilocin treated (n = 5) females (p = 0.0066). Analysis comparing c-Fos expression was done using a two-tailed unpaired t-test. Scale bar = 100 μM. B Histograms comparing the mean +/- S.E.M. of c-Fos expressing cells at baseline between males (n = 5) and female (n = 5) vehicle treated animals. Females showed significantly higher levels of c-Fos+ cells (p = 0.0254). Analyses comparing c-Fos were done using a two-tailed unpaired t-test. Scale bar = 100 μM. C representative image showing fiber tract and jGCaMP7f-AAV9 injection site. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

The effects of psilocin administration on PVT reactivity to an aversive air-puff stimulus

To assess the effects of psilocin on in-vivo PVT activity, we utilized jGCaMP7f-AAV9 to target neurons within the PVT utilizing an aversive air-puff stimulus in conjunction with behavioral fiber photometry (FP) assay to capture acute and prolonged changes in PVT reactivity (Fig. 1C). In vehicle-treated females, there were no significant differences in PVT reactivity traces, area under the curve (AUC), or peak point (PP) between baseline and injection. Additionally, no differences were seen in average max speed or time immobile following air-puff (Fig. 2A). In psilocin-treated females, there was a significant increase in reactivity following psilocin administration (2-way ANOVA: F_interaction(2.860,14.30) = 4.106, p = 0.0284; F_injection(1.000, 5.000) = 11.97, p = 0.0180; F_time(1.606, 8.030) = 19.45, p = 0.0011; Fig. 2B) with a significant increase in area under the curve (t(5) = 3.192, p = 0.0242; Fig. 2B) and peak point (t(5) = 3.574, p = 0.0160; Fig. 2B). The increase in PVT reactivity was associated with a decrease in max speed (2-way ANOVA: F_interaction(9,117) = 5.212, p < 0.0001; F_injection(1, 13) = 5.181, p = 0.0404; F_time(9,117) = 22.17, p < 0.0001; Fig. 2B) and an increase in time immobile in the first 3 seconds following air-puff (2-way ANOVA: F_interaction(9,117) = 2.377, p = 0.0166; F_injection(1, 13) = 1.832, p = 0.1989; F_time(9,117) = 11.79, p < 0.0001; Fig. 2B).

Fig. 2. Effects of psilocin on general PVT reactivity to an aversive stimulus in females.

A–F ΔF/F trace plots (top left) show PVT activity in response to a 500 ms air-puff at baseline vs. post-injection, 2-day, or 14-day follow-up. Traces are 500 ms binned group averages ± S.E.M. (shaded area) and analyzed using 2-way ANOVA or mixed-effects model depending on group size variance. Heatmaps (bottom left) show individual responses aligned to air-puff onset (dotted line). Behavioral metrics, average maximum speed (top right) and time immobile (middle right), are presented as group means ± S.E.M. and analyzed using 2-way ANOVA. AUC and PP values (bottom right) were compared between groups via 2-tailed unpaired t-tests. Histogram at the bottom demonstrating between group averages between psilocin and vehicle groups +/- S.E.M. and were analyzed using a 2-way ANOVA. For all histograms, each point represents an individual subject. Vehicle: n = 8; Psilocin: n = 6. Statistical Results: A Acute Vehicle: No significant effects reported. B Acute Psilocin: ΔF/F: treatment × time interaction (p = 0.0284), main effect of treatment (p = 0.0180); Max speed: interaction (p < 0.0001) and main effect of injection (p = 0.0355); AUC (p = 0.0242), PP (p = 0.0160) increased. C Vehicle – 2-day Follow-up: No significant effect of session (p = 0.0927). D Psilocin – 2-day Follow-up: Significant effect of session (p = 0.0272). E Vehicle – 14-day Follow-up: No significant effects of session (p = 0.0589). F Psilocin – 14-day Follow-up: No significant effect of session (p = 0.6434); Max speed: interaction (p = 0.0002). G Histogram comparing AUC between vehicle and psilocin groups at each time point: No significant effects reported. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. AUC area under the curve, PP peak point, ΔF/F change in fluorescence relative to baseline.

Next, we tested animals at 2- and 14-days post-injection to assess prolonged effects of psilocin administration on PVT reactivity in response to air-puff. At 2-days, there were no differences between baseline and the 2-day recording in PVT reactivity, max speed, or time immobile following air-puff in vehicle animals (Fig. 2C). In the psilocin treated animals, reactivity at 2-days remained elevated compared to baseline (2-way ANOVA: F_interaction(2.325,11.62) = 2.063, p = 0.1677; F_session(1.000, 5.000) = 9.546, p = 0.0272; F_time(1.772,8.859) = 13.38, p = 0.0025; Fig. 2D). No differences were seen in max speed or time immobile following air-puff (Fig. 2D). At the 14-day follow up, there were no differences seen in vehicle treated animals for PVT reactivity, max speed, or time immobile following air-puff (Fig. 2E). In the psilocin group, signal was restored to baseline levels at the 14-day follow up, with no differences in max speed or time immobile (Fig. 2F). To determine if there were differences in PVT activity between vehicle and psilocin treated groups, a between groups comparison of calcium dynamics AUC across time was conducted. There were no significant differences between vehicle and psilocin groups across time (Fig. 2G). All together, these data suggest that psilocin administration in females produced increased basal activation of the PVT as well as acute increases in reactivity to the air-puff stimulus that persisted 2 days following administration but returned to baseline levels by 14 days.

Active vs. passive responding

Based on our previous findings demonstrating that baseline threat responding predicted drug sensitivity^43,47, we divided groups by baseline behavioral responding to the air-puff into active and passive responding groups. Active responders were defined as animals that employed evasive locomotor behavior in response to the air-puff and actively moved away from the quadrant where the air-puff occurred (Fig. 3A), while passive responders were defined as rats that did not display enhanced locomotor behavior or actively try to escape the air-puff (Fig. 3B). At baseline, animals from both vehicle and psilocin groups had a roughly equal proportion of animals that displayed the active responding phenotype. On the day of injection, active responding was reduced in psilocin treated animals (Fig. 3C), with a significant difference between vehicle and psilocin treated groups (Fisher’s exact test: p = 0.0256). On the 2-day follow up, active responding in the psilocin-treated animals was increased such that levels were restored to baseline and were higher than levels in vehicle treated animals, however there were no significant differences seen between groups at the 2- and 14-day follow up (Fig. 3C). In the vehicle active responders, the proportion of animals employing an active responding strategy decreased across recording sessions. In the psilocin treated animals, active responding behavior was completely blocked on the day of injection but recovered to baseline levels that were higher than vehicle at the 2-day follow up recording with no significant differences between groups at 2- or 14-day follow up (Fig. 3D). In the passive responders, vehicle treated animals showed an increase in active responding on the day of injection, an effect not seen in psilocin treated animals (Fig. 3D). There were no differences in baseline PVT reactivity to the air-puff between active and passive responders for both vehicle and psilocin groups (Fig. 3E).

Fig. 3. Active vs. passive responding.

A Active responders were defined as animals that exhibited darting behavior and escaped the quadrant in which the air-puff occurred at baseline. B Passive responders remained immobile and did not escape the air-puff quadrant. C Left: Active responding over time across groups; each point reflects the percentage of animals displaying active responses. A 2-tailed Fisher’s exact test revealed a significant group difference on the day of injection (p = 0.0256). Right: Distribution of active vs. passive phenotypes within vehicle and psilocin groups. D Active responding over time plotted separately for left: active responders and right: passive responders. E Baseline PVT reactivity compared between active and passive responders in left: vehicle and right: psilocin groups; no significant differences observed. **p < 0.01, ***p < 0.001, ****p < 0.0001. ΔF/F change in fluorescence relative to baseline, BL baseline, INJ day of injection, 2 d 2-day, 14 d 14-day.

Increases in PVT reactivity with psilocin are driven by passive responders

We next assessed the differential effects of psilocin treatment between active and passive responders. In the vehicle active responders, there were no differences in PVT reactivity between baseline and injection (Fig. 4A), however there was a significant interaction in average max speed post air-puff, with significantly lower speeds at 2-seconds compared to baseline (2-way ANOVA: F_interaction(9,54) = 2.292, p = 0.0293; F_session(1, 6) = 0.3380, p = 0.5821; F_time(9,54) = 21.42, p < 0.0001; Fig. 4A). In the vehicle passive responders, there were also no differences in PVT reactivity between baseline and injection. For average max speed, there was an interaction between treatment and time, with a significant increase in speed at 3 s post air-puff compared to baseline (2-way ANOVA: 2-way ANOVA: F_interaction(9,54) = 3.202, p = 0.0036; F_session(1, 6) = 3.761, p = 0.1005; F_time(9,54) = 30.45, p < 0.0001; Fig. 4B), which could be due to injection stress. In the psilocin active responders, no differences were seen in PVT reactivity between baseline and injection, however, psilocin treated animals showed a significant increase in AUC compared to baseline (p = 0.0120). There was a reduction in max speed (2-way ANOVA: F_interaction(9,63) = 7.357, p < 0.0001; F_session(1, 7) = 6.749, p = 0.0355; F_time(9,63) = 16.49, p < 0.0001; Fig. 4C) and time immobile (2-way ANOVA: F_interaction(9, 63) = 4.272, p = 0.0002; F_session(1, 7) = 4.406, p = 0.074; F_time(9, 63) = 5.909, p < 0.0001; Fig. 4C), which could also be impacted by injection stress, although in a manner different to what was observed in passive responders, where activity was increased. In the psilocin passive responders, there was a significant increase in PVT reactivity between baseline and injection (2-way ANOVA: F_interaction(1.844, 3.688) = 3.462, p = 0.1427; F_treatment(1.000, 2.000) = 20.13, p = 0.0462; F_time(1.290, 2.580) = 10.75, p = 0.0571), with a significant increase seen in AUC (t(2) = 4.487, p = 0.0462, Fig. 4D) and no changes in max speed or time immobile. To determine if there were differences in PVT activity between vehicle and psilocin treated active and passive responders, a between groups comparison of calcium dynamics as measured by AUC across time was conducted. There were no significant differences between groups in the active or passive responders (Fig. 4E, F).

Fig. 4. Increases in PVT reactivity driven by baseline passive responders.

A–F ΔF/F trace plots (top left) show PVT activity in response to a 500 ms air-puff at baseline vs. post-injection. Traces are 500 ms binned group averages ± S.E.M. (shaded area). Heatmaps (bottom left) show individual responses aligned to air-puff onset (dotted line). Behavioral responses (top right: max speed; middle right: time immobile) are presented as group averages ± S.E.M. and analyzed via 2-way ANOVA. AUC and PP values (bottom right) are compared within groups using 2-tailed paired t-tests; Histograms at the bottom demonstrating between group averages between psilocin and vehicle active (left) and passive (right) responders +/- S.E.M. and were analyzed using a 2-way ANOVA. For all histograms, each point represents an individual subject. Active Vehicle: n = 4; Active Psilocin: n = 3; Passive Vehicle: n = 4; Passive Psilocin: n = 3. Statistical Results: A Active Vehicle: Max speed: time × treatment interaction (p = 0.0293), main effect of time (p < 0.0001); No significant effects reported for ΔF/F or AUC/PP. B Passive Vehicle: Max speed: significant interaction (p = 0.0036). C Active Psilocin: ΔF/F: Peak point increased (p = 0.0120); Max speed: interaction (p < 0.0001), main effect of treatment (p = 0.0355), main effect of time (p < 0.0001); Time immobile: interaction (p = 0.0002). D Passive Psilocin: ΔF/F: main effect of treatment (p = 0.0462); AUC increased (p = 0.0462). E Histogram comparing psilocin and vehicle active responders across time: no significant effect of treatment found (p = 0.8909). F Histogram comparing psilocin and vehicle passive responders across time: no significant effect of treatment found (p = 0.4706). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. AUC area under the curve, PP peak point, ΔF/F change in fluorescence relative to baseline.

The effects of psilocin administration on basal activity within PVT→CeA circuitry

Next, to determine how psilocin alters activity within the PVT→CeA circuit, we utilized intracranial injection of retrobeads in the CeA to specifically examine the effects of psilocin on PVT neurons that project to the CeA (Fig. 5A). In males, there were no differences in the proportion of retrobead/cFos+ PVT neurons between vehicle (mean: 29.3, SD: 3.85) and psilocin (mean: 31.32, SD: 6.03) groups (Fig. 5B, C). However, in the female psilocin (mean: 48.27, SD: 17.98) group there was a greater proportion of cFos + /retrobead⁺ PVT neurons compared to vehicle (mean: 23.47, SD: 2.79) control (t(7) = 2.697, p = 0.03; Fig. 5B, C), suggesting increased activity of PVT→CeA projections in females but not males.

Fig. 5. Sex-specific effects of psilocin administration on PVT→CeA circuitry.

A Top: Experimental timeline for retrobead and drug administration. Bottom: Representative image showing retrobead injection into the CeA. B Representative images of c-Fos (green) and retrobead (red) co-expression in males (top) and females (bottom) across groups. C Histograms showing the proportion of retrobead-labeled cells co-expressing c-Fos in males (top) and females (bottom). Male vehicle: n = 4; Male psilocin: n = 3; Female vehicle: n = 4; Female psilocin: n = 5. Statistical analysis utilized 2-tailed unpaired t-tests found a significant difference between vehicle in psilocin in females (p = 0.0308) but not in males (p = 0.6084). D Top: Schematic of injection site and fiber placement for the intersectional viral approach. Bottom: Timeline for fiber photometry recordings. E Representative image showing fiber placement and Cre-dependent GCaMP expression in PVT. Scale bar = 100 μM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. BL baseline, 0 d day of injection, 2 d 2-day follow-up, 14 d 14-day follow-up.

The effects of psilocin administration on reactivity of PVT→CeA circuitry in females

Next, we utilized an intersectional viral strategy to investigate how psilocin administration alters reactivity within this specific population of PVT→CeA neurons. To do this, we performed injections into the CeA with a retrograde Cre-recombinase virus (pENN.AAV.hSyn.Cre.WPRE.hGH) followed by injection of a Cre-dependent GCaMP (AAV9-pGP-AAV-syn-FLEX-jGCaMP7s-WPRE) and fiber cannula installation in the PVT (Fig. 5D, E).

In the vehicle treated females, we found that PVT→CeA reactivity was significantly reduced on the day of injection compared to baseline (2-way ANOVA: F_interaction(29, 290) = 3.907, p < 0.0001; F_injection(1, 10) = 7.252, p = 0.0226; F_time(29, 290) = 6.657, p < 0.0001; Fig. 6A). Area under the curve (AUC) (t(10) = 2.725, p = 0.0214) and peak point (PP) (t(10) = 2.661, p = 0.0239) were both significantly reduced (Fig. 6A). Additionally, no differences were seen in max speed or time immobile following air-puff (Fig. 6A). In the psilocin group, no differences were seen in PVT reactivity between baseline and injection (Fig. 6B), suggesting that psilocin treatment prevented a reduction in reactivity seen in vehicle control animals. Interestingly, no changes were seen in average max speed or time immobile following air-puff (Fig. 6B). At the 2-day follow up, vehicle females showed a persistent reduction in PVT→CeA reactivity compared to baseline (2-way ANOVA: F_interaction(29, 290) = 2.637, p < 0.0001; F_session(1, 10) = 2.933, p = 0.1175; F_time(29, 290) = 7.084, p < 0.0001; Fig. 6C). However, no significant reductions in AUC and PP were observed, though there was a trend for PP (p = 0.0585, Fig. 6C). Additionally, there were no changes in average max speed or time immobile. In the psilocin group, no differences were seen in PVT→CeA reactivity at the 2-day follow up (Fig. 6D), suggesting a persistent prevention of reduction in signal reactivity seen in vehicle control. At the 14-day follow up, the vehicle group showed a sustained reduction in PVT→CeA reactivity compared to baseline (2-way ANOVA: F_interaction(29,290) = 5.480, p < 0.0001; F_session(1, 10) = 8.027, p = 0.0178; F_time(29,290) = 6.454, p < 0.0001; Fig. 6E). Significant reductions were also observed in AUC (t(10) = 2.837, p = 0.0178) and PP (t(10) = 2.910, p = 0.0156; Fig. 6E) with no differences in average max speed or time immobile. In the psilocin group, there was a significant decrease in reactivity compared to baseline (2-way ANOVA: F_interaction(29, 290) = 3.568, p < 0.0001; F_session(1, 10) = 6.321, p = 0.0307; F_time(29, 290) = 12.53, p < 0.0001; Fig. 6F). There were also significant reductions observed in AUC (t(10) = 3.331, p = 0.0076) and PP (p = 0.0217; Fig. 6F). However, no changes in average max speed or time immobile were observed. To determine if there were differences in PVT activity between vehicle and psilocin treated females, a between groups comparison of calcium dynamics as measured by AUC across time was conducted. There were no significant differences found between groups across time (Fig. 6G). All together, these data suggest that psilocin delays time-dependent reductions in PVT reactivity to repeated aversive air-puff.

Fig. 6. The effects of psilocin administration on reactivity of PVT→CeA circuitry in females.

A–F ΔF/F trace plots (top left) show PVT reactivity to a 500 ms air-puff across timepoints (baseline vs. post-injection, 2-day, and 14-day follow-ups). Traces represent 500 ms binned group averages ± S.E.M. (shaded area). Heatmaps (bottom left) display individual responses aligned to the air-puff (dotted line). Behavioral responses—average maximum speed (top right) and time immobile (middle right)—are presented as group means ± S.E.M. and analyzed via 2-way ANOVA. AUC and PP values (bottom right) are compared by 2-tailed unpaired t-tests. Histogram at the bottom demonstrating between group averages between psilocin and vehicle groups +/- S.E.M. and were analyzed using a 2-way ANOVA. For all histograms, each point represents an individual subject. Vehicle: n = 11; Psilocin: n = 11. Statistical Results: A Acute Vehicle: ΔF/F: time × treatment interaction (p < 0.0001), main effects of time (p < 0.0001) and treatment (p = 0.0226). AUC (p = 0.0214). PP (p = 0.0239). B Acute Psilocin: No significant effects reported. C Vehicle – 2-day Follow-up: ΔF/F: time × treatment interaction (p < 0.0001), main effect of time (p < 0.0001). D Psilocin – 2-day Follow-up: ΔF/F: no significant effects reported. Time immobile: main effect of treatment (p = 0.0014). E Vehicle – 14-day Follow-up: ΔF/F: time × treatment interaction (p < 0.0001), main effects of time (p < 0.0001) and treatment (p = 0.0178). AUC (p = 0.0176). PP (p = 0.0156). F Psilocin – 14-day Follow-up: ΔF/F: time × treatment interaction (p < 0.0001), main effects of time (p < 0.0001) and treatment (p = 0.0307). AUC (p = 0.0076). PP (p = 0.0217). G Histogram comparing AUC between vehicle and psilocin groups at each time point: No significant effects reported. # indicates significance of ranging values for multiple points. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. AUC area under the curve, PP peak point, ΔF/F change in fluorescence relative to baseline.

Psilocin-induced sustained reactivity of PVT-CeA circuitry is driven by baseline active responders in females

To assess the impact of baseline threat responding on changes in reactivity within the PVT→CeA circuit, females were split into active and passive groups for subsequent analysis. At baseline, 60% of animals in the vehicle group exhibited an active response to the air-puff. Similarly, 54.5% of the psilocin group exhibited an active response. No differences were seen between groups in the proportion of active responding across time (Fig. 7A). This relatively even split allowed for animals to be separated into active and passive responding groups for further analysis. Notably, in the passive responding group, psilocin treatment may have delayed a shift towards active responding following air-puff (Fig. 7B), however there were no significant differences between groups at any of the time points. In the active responders, vehicle treated females appeared to show a desensitization to the air-puff, exhibited by a decrease in the proportion of animals exhibiting an active response across time. Psilocin treated females did not show this same trend and appeared to maintain their responding strategy across time. No significant differences were seen between groups across time (Fig. 7B).

Fig. 7. Psilocin induced sustained reactivity of PVT→CeA circuitry driven by baseline active responder females.

A Left: Percentage of animals displaying active responses over time. Right: Proportion of active vs. passive responders at baseline. B Proportion of animals exhibiting active responding across time, separated by left: passive and right: active responder groups. C–F ΔF/F trace plots (top left) show PVT activity following a 500 ms air-puff in baseline or injection conditions. Traces represent 500 ms binned group averages ± S.E.M. (shaded area). Heatmaps (bottom left) display individual responses aligned to air-puff onset (dotted line). Behavioral responses—average maximum speed (top right) and time immobile (middle right)—are shown as group means ± S.E.M. and analyzed via 2-way ANOVA. AUC and PP values (bottom right) were compared by 2-tailed unpaired t-tests. Histograms at the bottom demonstrating within group averages between active and passive responding groups +/- S.E.M. in active (left) and passive (right) responders and were analyzed using a 2-way ANOVA. For all histograms, each point represents an individual subject. Active Vehicle: n = 7; Active Psilocin: n = 6; Passive Vehicle: n = 4; Passive Psilocin: n = 5; Statistical Results: C Active Baseline – Vehicle vs. Psilocin: No significant effects reported. D Passive Baseline – Vehicle vs. Psilocin: No significant effects reported. E Active Injection – Vehicle vs. Psilocin: ΔF/F: time × treatment interaction (p < 0.0001). F Passive Injection – Vehicle vs. Psilocin: ΔF/F: no significant effects reported. Max speed: treatment (p = 0.0238). Time immobile: treatment (p = 0.0437). G Histogram comparing AUC for ΔF/F between vehicle and psilocin groups for active responders. Time x treatment interaction (p = 0.0139). Uncorrected Fisher’s LSD showed significant reduction in vehicle between baseline and injection (p = 0.0359). H Histogram comparing AUC for ΔF/F between vehicle and psilocin groups for passive responders. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. AUC area under the curve, PP peak point, ΔF/F change in fluorescence relative to baseline.

In the active responders, at baseline there were no differences seen in PVT→CeA reactivity, average max speed, or time immobile between groups (Fig. 7C). Similarly, in the passive responders, there were no differences in reactivity, max speed, or time immobile (Fig. 7D). On the day of injection, vehicle treated active responders exhibited blunted PVT→CeA reactivity compared to psilocin treated animals (F_interaction(29,319) = 2.495, p < 0.0001; F_injection(1, 11) = 3.852, p = 0.0755; F_time(29, 319) = 14.27, p < 0.0001; Fig. 7E). Interestingly, there were no differences in max speed or time immobile (Fig. 7E). In the passive responders, no differences were seen between groups in reactivity, however there were significant reductions in max speed following air-puff (2-way ANOVA: F_interaction(9, 63) = 0.9139, p = 0.5191; F_treatment(1, 7) = 8.270, p = 0.0238; F_time(9, 63) = 6.622, p < 0.0001; Fig. 7F) and differences in time immobile (F_interaction(9, 63) = 0.9646, p = 0.4775; F_treatment(1, 7) = 6.034, p = 0.0437; F_time(9, 63) = 12.38, p < 0.0001; Fig. 7F). A within group comparison of calcium transient AUC across time found a significant difference between baseline and injection in vehicle control active responders (2-way ANOVA: F_interaction(1, 11) = 8.537, p = 0.0139; F_treatment(1, 11) = 0.7655, p = 0.4003; F_time(1, 11) = 0.1051, p = 0.7518; Fig. 7G), with no differences in the passive responders (Fig. 7H).

The effects of psilocin administration on reactivity of PVT→CeA circuitry in males

Given that sexual divergent behavior of active vs. passive responding⁴⁸, and the implication of the PVT→CeA circuit in this behavior³⁷, we sought to determine the effects of psilocin within this circuit in males. In vehicle treated males, we found that PVT→CeA reactivity was not changed between baseline and injection. While there were no differences seen in PP, males showed an interesting effect in which the signal decreased below baseline in the last 5 seconds of the trace following air-puff, reducing AUC (F_time(1, 12) = 9.106, p = 0.0107; Fig. 8A). No differences were seen in max speed or time immobile (Fig. 8A). Similarly, no differences were seen in the psilocin treated animals between baseline and injection, though the same reduction in signal was seen in the second half of the post air-puff reactivity signal and AUC (F_time(1, 10) = 11.40, p = 0.0070; Fig. 8B). No differences were seen in max speed or time immobile (Fig. 8B).

Fig. 8. The effects of psilocin administration on reactivity of PVT→CeA circuitry in males.

A–F ΔF/F trace plots (top left) show PVT→CeA reactivity to a 500 ms air-puff across timepoints (baseline vs. post-injection, 2-day, and 14-day follow-ups). Traces represent 500 ms binned group averages ± S.E.M. (shaded area). Heatmaps (bottom left) display individual responses aligned to the air-puff (dotted line). Behavioral responses—average maximum speed (top right) and time immobile (middle right)—are presented as group means ± S.E.M. and analyzed via 2-way ANOVA. AUC and PP values (bottom right) are compared by 2-tailed t-tests. Histograms at the bottom demonstrating between group averages between psilocin and vehicle groups +/- S.E.M. in early response AUC (left) and late response AUC (right) and were analyzed using a 2-way ANOVA. Vehicle: n = 7; Psilocin: n = 6; Statistical Results: A Acute Vehicle: ΔF/F: no significant effects. Significant reduction seen between early and late response AUC (time: p = 0.0107). No significant effects were observed for PP, maximum speed, or time immobile. No significant effects were observed for PP, maximum speed, or time immobile. B Acute Psilocin: ΔF/F: no significant effects. Significant reduction seen between early and late response AUC (time: p = 0.0070). No significant effects were observed for PP, maximum speed, or time immobile. C Vehicle – 2-day Follow-up: ΔF/F: no significant effects. Significant reduction seen between early and late response AUC (time: p = 0.0120). Time immobile was significantly increased at 2–3 s post air-puff (p < 0.0001). D Psilocin – 2-day Follow-up: ΔF/F reactivity was significantly blunted relative to baseline (p = 0.0223), with a significant reduction in PP (p = 0.0337). Significant reduction seen between early and late response AUC (interaction: p = 0.0132; time: p = 0.0337). Uncorrected Fisher’s LSD revealed lower AUC between baseline and 2-day follow up in the early response (p = 0.0066) and between early and late response in baseline (p = 0.0001) but not the 2-day follow up (p = 0.0864). E Vehicle – 14-day Follow-up: ΔF/F reactivity was significantly reduced relative to baseline (interaction: p = 0.0008), with significant reductions in PP (p = 0.0294) and a significant early–late AUC difference (time: p = 0.0047). No significant effects were observed for maximum speed or time immobile. F Psilocin – 14-day Follow-up: ΔF/F reactivity showed a sustained reduction relative to baseline (p < 0.0001), with significant reductions in PP (p = 0.0181). Significant reduction seen between early and late response AUC (interaction: p = 0.0094; time: p = 0.0055). Uncorrected Fisher’s LSD revealed lower AUC between baseline and 14-day follow up in the early response (p = 0.0050) and between early and late response in baseline (p = 0.0002) but not the 2-day follow up (p = 0.1084). G Histogram comparing early response AUC (0-5 s) for ΔF/F between vehicle and psilocin groups. H Histogram comparing late response AUC (5-10 s) for ΔF/F between vehicle and psilocin groups. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. AUC area under the curve, PP peak point, ΔF/F change in fluorescence relative to baseline.

At the 2-day follow up, vehicle treated males showed no significant difference in PVT→CeA reactivity compared to baseline (Fig. 8C). No differences were seen in max speed, though there was a significant difference in time immobile (2-way ANOVA: F_interaction(9, 108) = 5.734, p < 0.0001; F_treatment(1, 12) = 0.01963, p = 0.8909; F_time(9, 108) = 54.90, p < 0.0001; Fig. 8C), with significantly greater immobility at 2- and 3-seconds post air-puff compared to baseline (Fig. 8C). In the psilocin treated males, PVT→CeA reactivity was significantly blunted compared to baseline at the 2-day follow-up (2-way ANOVA: F_interaction(1.901, 9.506) = 5.928, p = 0.0223; F_session(1.000, 5.000) = 3.501, p = 0.1202; F_time(1.307, 6.535) = 9.104, p = 0.0176 Fig. 8D), with a significant reduction in PP (t(5) = 2.903, p = 0.0337). For AUC, a significant interaction (2-way ANOVA: F_interaction(1, 10) = 9.030, p = 0.0132; F_session(1, 10) = 3.329, p = 0.0980; F_time(1, 10) = 14.12, p = 0.0037) revealed that the reduction in signal between the first and second half seen at baseline (p = 0.0001), was no longer present at the 2-day follow up (p = 0.0864, Fig. 8D).

At the 14-day follow up, vehicle treated males showed a significant reduction in reactivity compared to baseline (2-way ANOVA: F_interaction(29, 174) = 2.233, p = 0.0008; F_session(1, 6) = 0.3299, p = 0.5866; F_time(29, 174) = 7.003, p < 0.0001, Fig. 8E), with significant reduction in PP (t(6) = 2.843, p = 0.0294) and a difference in signal between the first and second half of AUC (2-way ANOVA: F_time(1, 12) = 12.01, p = 0.0047, Fig. 8E). No differences were seen in max speed or time immobile following air-puff. In the psilocin treated males, there was a sustained reduction in signal compared to baseline (2-way ANOVA: F_interaction(29, 145) = 8.450, p < 0.0001; F_session(1, 5) = 3.066, p = 0.1403; F_time(29, 145) = 7.939, p < 0.0001, Fig. 8F), with significant reductions in PP (t(5) = 3.459, p = 0.0181). For AUC, a significant interaction (2-way ANOVA: F_interaction (1, 10) = 10.29, p = 0.0094; F_session (1, 10) = 3.459, p = 0.0925; F_time (1, 10) = 12.41, p = 0.0055) revealed that the reduction in signal between the first and second half seen at baseline (p = 0.0002), was no longer present at the 14-day follow up (p = 0.1084, Fig. 8F). To determine between group differences, AUC was analyzed across time. No differences were found for either the early (0-5 s) or late (5-10 s) signal response between groups (Fig. 8G, H).

Next, we assessed the impact of baseline threat responding on changes in reactivity within the PVT→CeA circuit for males. At baseline, 66.7% of psilocin treated males and 57% of vehicle treated males were active responders (Fig. 9A). Psilocin treated males showed an overall reduction in active responding over time, while vehicle males appeared to show increases in active responding to the air-puff (Fig. 9A). In passive responding males, there appeared to be a shift towards active responding between vehicle and psilocin groups. In active responders, psilocin treated animals appeared to show decreases in active responding across time, while active responses remained relatively stable in vehicle control (Fig. 9B).

Fig. 9. The effects of psilocin administration on reactivity of PVT→CeA circuitry in active and passive males.

A–J Behavioral categorization and fiber photometry analyses stratified by baseline threat response strategy (active vs. passive). A, B Proportion of active and passive responders across timepoints for vehicle- and psilocin-treated males. Psilocin-treated males showed a reduction in active responding over time, whereas vehicle-treated males showed an increase in active responding. C–F ΔF/F trace plots (top left) show PVT → CeA reactivity to a 500 ms air-puff at baseline and on the day of injection for active and passive responders. Traces represent 500 ms binned group averages ± S.E.M. (shaded area). Heatmaps (bottom left) display individual responses aligned to the air-puff (dotted line). Behavioral responses—average maximum speed (top right) and time immobile (middle right)—are presented as group means ± S.E.M. and analyzed via 2-way ANOVA. AUC values (bottom right) are compared by 2-tailed t-tests. Histograms at the bottom demonstrating within group averages between psilocin and vehicle groups +/- S.E.M. in active (left) and passive (right) responders for early (top) and late responses (bottom) and were analyzed using a 2-way ANOVA. For all histograms, each point represents an individual subject. Active Vehicle: n = 4; Active Psilocin: n = 4; Passive Vehicle: n = 3; Passive Psilocin: n = 2; Statistical Results: A Baseline Threat Strategy: At baseline, 66.7% of psilocin-treated males and 57% of vehicle-treated males were classified as active responders. B Longitudinal Threat Strategy: Psilocin-treated males showed a reduction in active responding across time, whereas vehicle-treated males showed increased active responding. C Active Responders – Baseline: No significant differences were observed between vehicle- and psilocin-treated males in ΔF/F reactivity, maximum speed, or time immobile. A significant reduction in AUC between early and late post–air-puff epochs was observed (p = 0.0278). D Passive Responders – Baseline: No significant differences were observed between vehicle- and psilocin-treated males in ΔF/F reactivity, maximum speed, or time immobile. No significant early–late AUC reduction was observed (p = 0.0714). E Active Responders – Day of Injection: No significant differences were observed between vehicle- and psilocin-treated males in ΔF/F reactivity, maximum speed, or time immobile. A significant reduction in AUC between early and late post–air-puff epochs was present (p = 0.0341). Psilocin-treated males showed a trend toward reduced late-phase signal attenuation relative to vehicle (p = 0.08). F Passive Responders – Day of Injection: No significant differences were observed between vehicle- and psilocin-treated males in ΔF/F reactivity, maximum speed, or time immobile. No significant early–late AUC reduction was observed (p = 0.0849). G–J AUC Comparisons Across Time: No significant differences were observed in early (0–5 s) or late (5–10 s) AUC responses across time between or within treatment groups for either active or passive responders.

Next, fiber photometry data were split into active and passive subgroups to assess differences in CeA→PVT reactivity to the air-puff. At baseline, there were no differences in reactivity, max speed, or time immobile between vehicle and psilocin treated active responders (Fig. 9C). A main effect of time in AUC demonstrated the previously noted reduction in signal within the late response of the 10 s following air-puff (F_time(1, 6) = 8.341, p = 0.0278, Fig. 9C). In passive responders, there were no differences in reactivity, max speed, or time immobile seen between vehicle and psilocin treated animals. Interestingly, there was no reduction in AUC between the early and late response in post-air-puff activation seen in the combined group analysis and inactive responders (F_time(1, 3) = 7.501, p = 0.0714, Fig. 9D). On the day of injection, active responders showed no differences in reactivity, max speed, or time immobile between vehicle and psilocin treated animals, though there was a trend towards reduced max speed in psilocin treated animals (p = 0.054, Fig. 9E). The reduction in signal between the early and late response of air-puff induced increases in reactivity was present on day of injection for the active responders (F_time(1, 4) = 9.993, p = 0.0341, Fig. 9E). Psilocin treated active responders showed a trend for less of a reduction in signal during the late response (multiple comparisons-late response-vehicle vs. psilocin; p = 0.08, Fig. 9E). For the passive responders on the day of injection, there were no significant differences in reactivity, max speed, or time immobile between the vehicle and psilocin treated males (Fig. 9F). Similar to baseline, passive responders did not show a decrease in signal between the early and late reactivity response (F_time(1, 5) = 4.595, p = 0.0849, Fig. 9F). There were no differences in AUC for both early (Fig. 9G, H) and late response (Fig. 9I, J) across time between or within groups in active and passive responders.

Discussion

In the current study we utilized immunohistochemistry and in-vivo fiber photometry (FP) to investigate the effects of the psychedelic psilocin on basal activity and reactivity of the PVT and PVT→CeA circuitry. Reactivity to the previously developed air-puff stimulus was used to assess changes in calcium dynamics in the presence of drug and at more prolonged time points. We found that psilocin administration produced increases in c-Fos expression and proportion of c-Fos + /retrobead+ cells within the PVT of females, but not males. Using a standard GCaMP approach in females, we found that psilocin administration increased PVT reactivity to the air-puff stimulus that persisted at a 2-day follow up recording, an effect not seen in vehicle controls. When animals were analyzed based on basal threat responding to air-puff, we found that the increased PVT reactivity was driven by females employing a passive responding strategy. Next, using an intersectional Cre-recombinase/Cre-GCaMP viral strategy, we found that psilocin prevented a decrease in reactivity seen in vehicle control on day of injection and sustained PVT→CeA activation at the 2-day follow up compared to a significantly reduced reactivity seen in vehicle control in females but not males. Notably, when females were split into groups based on baseline responding strategy, the sustained activation seen on day of injection in the psilocin treated group was driven by females employing an active baseline responding strategy.

The PVT is a significant hub within the central nervous system responding to stress^28–30,49. Dysregulation of the PVT has been implicated in behaviors relating to psychiatric disorders such as major depressive disorder (MDD)^50,51 and substance use disorder^52–54. A hallmark of these disorders is a compromised ability to make emotionally and logically sound decisions^55–59, including dysfunctional emotional regulation and responses to stress^60–62. The bi-directional connections between cortical neurons and thalamic nuclei is referred to as the thalamocortical system⁶³, which involves complex and functionally specific connections between these two regions⁶⁴, and is involved in many cognitive processes such as decision making⁶⁵. Dysregulation of thalamocortical connectivity is associated with disorders such as MDD^66,67. When exposed to a stressor, animals employ either an active or passive coping strategy. In certain circumstances, it is more advantageous to freeze (passive) or flee (active). It has been shown that cortico-thalamic circuitry governs the decision to engage in active vs. passive responding to a threat⁶⁸. From the PVT, there are divergent circuits mediating active vs. passive threat responding, with projections to the CeA orchestrating a passive response and projections to the nucleus accumbens (NAc) promoting an active response³⁷. Previous findings showed that changes in reactivity within the CeA were primarily driven by animals employing an active threat responding strategy⁴³, suggesting the possibility that threat responding circuitry may be implicated in psychedelic drug action. Positron emission tomography (PET) imaging studies in humans have illuminated changes in thalamic glucose metabolism, suggestive of increased activation, following administration of serotonergic psychedelics, including psilocybin^16,69–71. Additionally, recent work highlighted strengthened cortical connections to the PVT following repeated, low-dose psychedelic administration in rats⁴⁶. We found that global PVT reactivity to an aversive air-puff was increased with psilocin injection in females, and that this occurred in parallel with reductions in locomotor activity in response to the air-puff. This may suggest that PVT activity correlates with or is causal to reduces in threat responding. However, when we separated animals into active and passive threat responders based on baseline responding, we found that the increased PVT reactivity occurred in the passive responders, while reductions in threat responding occurred in the active responders. This could be due to limits of detecting reductions in threat responding in passive responders (i.e., a floor effect), or it could suggest that threat responding is not governed by global PVT activity, but by specific cell populations or discrete circuits. Together, our work contributes to our knowledge on the acute effects of psychedelic drug administration on reactivity within this key thalamic nucleus.

Here, we showed that psilocin induced acute increases in basal activity in PVT of females, but not males and increased reactivity to air-puff in the PVT of psilocin-treated females. Notably, vehicle injected females had lower basal activation compared to vehicle males, which could have contributed to greater increases in females, however similar sex-specific increases with psilocin were observed in PVT→CeA neurons despite similar basal activation levels. We have previously reported region-specific differences in basal reactivity to the aversive puff stimulus as well as psilocin effects in male and female rats. In a previous study we found no sex differences in basal reactivity to air-puff in the central amygdala (CeA) but females showed greater active responding to air-puff at baseline and displayed greater increases in basal activity and air-puff reactivity after acute psilocin, while males showed reductions in reactivity at later time points⁴³. In another study, we found that males showed greater basal reactivity to the air-puff in the paraventricular nucleus of the hypothalamus (PVN) compared to females, and while there were no differences in basal activity with acute psilocin, males showed increased PVN reactivity that was primarily driven by active responders⁴⁷. Previous work has shown sex-differences in threat responding behavior, with females primarily employing an active threat response^48,72. This may suggest that females have heightened activation of the PVT→NAc pathway, rather than PVT→CeA circuitry³⁷. A previous study found divergent effects of dopamine receptor2 (DRD2) PVT projections to CeA or NAc during fear conditioning with opposing results on active or passive conditioned fear responses. in contrast to PVT^DRD2→NAc neurons, activity of PVT^DRD2→CeA neurons was associated with reduced avoidance behavior and increased freezing and optogenetic inhibition of PVT^DRD2→CeA neurons increased active avoidance behavior and reduced freezing, suggesting that of PVT^DRD2→CeA circuitry drives passive threat responding behavior³⁷. Another study identified that corticotropin-releasing factor (CRF) PVT neurons and upstream projections from ventromedial hypothalamus are bidirectionally modulated during approach avoidance conflict to food or predator odor exposure⁷³. It has also been shown that female rats show compromised CRF receptor internalization, leading to heightened sensitivity to CRF signaling in females compare to males⁷⁴. These studies highlight the potential for population- or circuit-specific effects and the possibility that different local or projection populations in PVT may explain why psilocin has differential effects on PVT reactivity between males and females. Another example is the neuropeptide, pituitary adenylate cyclase-activating polypeptide (PACAP), a modulator of stress responding^75,76. Studies looking at PACAP expression within the PVT of mice showed greater expression in females than males⁷⁷. Recent work has also shown that PACAP modulates 5-hydroxytryptamine (5-HT)_2AR signaling via internalization of receptors in cell culture, with cerebrospinal administration of PACAP suppressing 5-HT_2AR agonist induced head twitch responses⁷⁸, a behavioral output thought to be involved in psychedelic drug action in rodents^79–81. Given potential sex differences in PACAP expression and involvement of PACAP in 5-HT_2AR activity, it is possible that local differences in PVT could be responsible for observed sex-specific effects of the 5-HT_2AR partial agonist psilocin. However, future studies are needed to identify the specific signaling mechanisms involved in these effects. Future work should examine the effects of 5-HT_2AR agonism on PVT reactivity through the use antagonist pretreatment or genetic knockdown.

The current study provides insight into the effects of psilocin administration on reactivity of PVT→CeA circuitry. Here, we showed that psilocin administration increased basal activation of PVT→CeA projecting neurons in females but not males. Additionally, we showed that psilocin prevented a reduction in activation of PVT→CeA neurons seen in vehicle group in response to repeated air-puffs. Previous findings in the lab demonstrated an acute increase in CeA reactivity of females but not males⁴³. While this suggests that PVT→CeA circuitry may be an upstream mediator of the observed CeA effects, it is not the only circuit involved in threat responding behavior. Future work is needed to explore psilocin effects on upstream and downstream PVT circuitry and that employs manipulation of different circuits to determine how psychedelics impact reactivity to an aversive air-puff stimulus.

All together, these data provide valuable insight into the effects of the partial 5-HT_2AR agonist psychedelic psilocin on PVT basal activity and reactivity to an aversive stimulus. Given the relevance of implicit threat and stress responding in the development of psychiatric disorders, studying PVT circuitry may provide key insights into better understanding the etiology of psychiatric disorders and potentially uncovering previously unknown mechanisms of action for possible therapeutic effects of psychedelics.

Methods

Animals

All experiments utilized male and female Sprague Dawley rats (200-400 g, ~8-weeks old at arrival, Envigo, Indianapolis, IN) that were group-housed prior to surgery followed by individual housing with nestlet enrichment in a humidity- and temperature-controlled (22 °C) vivarium on a 12 hr light/dark cycle with ad libitum access to food and water.

Stereotaxic surgery

At ~9-weeks, male and female rats were anesthetized with isoflurane/oxygen (5% induction 2-3% maintenance), placed in a (KOPF) stereotaxic frame and intracranial surgery was performed using standard sterilization procedures according to approved protocols. For standard fiber photometry cohorts, rats were injected with pGP-AAV-syn-jGCaMP7f-WPRE; Addgene: 104488-AAV9) into the PVT (AP:−3.4 mm, ML:+/−0 mm, DV:−4.9 mm). For intersectional fiber photometry cohorts, 250 nl of retrograde-Cre recombinase (pENN.AAV.hSyn.Cre.WPRE.hGH; Addgene: 105553-AAVrg) was infused in the CeA (AP: −2 mm, ML: +/−3.9 mm, DV: −8 mm) at a rate of 100 nl/min using a 2.0 µL Hamilton Syringe (2.0Neuros Syringe, Model 7002 KH, 30 gauge). Immediately following CeA injections, a single injection of 400 nl of Cre-dependent GCaMP (AAV9-pGP-AAV-syn-FLEX-jGCaMP7s-WPRE; Addgene: 104491-AAV9) was injected into the PVT (AP: −3.4 mm, ML: +/−0 mm, DV: −5.2 mm). For all microinjections syringe was left in place for 5 minutes prior to removal. Following infusion, a mono-fiber cannula (Doric: MFC_200/230-0.37_6.5mm_ZF1.25_FLT) was installed at the PVT, 0.3 mm above injection site at coordinates (AP: −3.4 mm, ML: +/−0 mm, DV: −4.9 mm). Three screws were placed into the skull followed by application of Metabond dental cement to create a headcap and stabilize fiber implants. Animals were treated with 2.5 mg/kg Flunixin on the day of surgery and for two days following. All procedures were approved by the University of North Carolina Chapel Hill’s Institutional Animal Care and Use Committee (IACUC).

Drug administration

Psilocin (Cayman Chemical) was dissolved in 2% glacial acetic acid and brought up to a 1 mg/ml dilution in 0.9% saline. On the day of injection animals received subcutaneous (s.c.) injections of either vehicle (2% glacial acetic acid in 0.9% saline) or psilocin (2 mg/kg) and were returned to their home cage for 30 min before being placed into open field box for a 10 min habituation prior to air-puff administration. For the c-Fos cohort, animals received injections 2 h prior to perfusion and collection of brain tissue for immunohistochemical analysis.

Retrograde tracing

At ~8-weeks of age, male and female Sprague Dawley rats were anesthetized with isoflurane/oxygen (5% induction 2-3% maintenance), placed in a stereotaxic frame (KOPF) and intracranial surgery was performed using standard sterilization procedures according to approved protocols. Rats received intracranial injections (300nL) of red retrograde beads (Lumafluor Inc.) into the CeA (AP: −2 mm, ML: +/−3.9 mm, DV: −8 mm). Seven days following intracranial retrobead surgery, animals were administered either vehicle or psilocin (2 mg/kg, s.c.) depending on group. Animals were then perfused 2 hr post injection and brains were collected for tissue processing.

Fiber photometry

To look for changes in PVT reactivity following psilocin administration, an air-puff stimulus was used in conjunction with in-vivo behavioral fiber photometry (FP). Baseline recordings were used to demonstrate changes in fluorescence (ΔF/F) prior to drug administration. Animals underwent FP recordings at ~12 weeks of age. All FP recordings were collected with the TDT RZ5 (Tucker-Davis Technologies, Alachua, FL) real time processor and Doric external components. Fiber implants were silica mono fiber cannula (Doric: MFC_200/230-0.37_6.5mm_ZF1.25_FLT) with a numerical aperture of 0.37 NA. Light output at the tip of the recording cable was confirmed and standardized to 50 μW on each recording day. Cohorts were split into two subgroups ran consecutively in order to control for time of day within each group and across subjects. On the day of recording, animals were habituated to the room for 1 hr after which they were placed in chambers for 10 min habituation and baseline recording. To test for acute effects, animals were given an injection of either vehicle or psilocin (2 mg/kg, s.c.) and placed in home cage for 20 min prior to 10 min habituation for a total of 30 min prior to the beginning of the air-puff session. Follow-up recordings were collected at 2- and 14-days post administration.

Air-puff stimulus

An air-puff stimulus was performed using house air supply under the control of a regulator to administer a short (500 ms) burst of 85 psi air by a Parker solenoid (Part #003-0868-900), powered by a custom 12 V electrical circuit box connected to the TDT RZ5 system to trigger solenoid opening and closing simultaneous to fiber recording timestamps through the Synapse program. Each session consisted of a 10 min habituation baseline period before the delivery of two air-puff administrations with a 5 min inter-stimulus interval (ISI). Air-puffs were delivered by positioning of silicone tubing connected to a metal rod lowered into the behavioral chamber from above and directed at the face of the animal. Tubing and rod were retracted and positioned above the behavioral chamber during baseline and between stimulus deliveries. Baseline fluorescence was calculated through a least-squares linear fitting of the 405 nm isosbestic control to the 465 nm GCaMP signal to create a resting GCaMP signal over the entire session not containing event-dependent fluctuations in GCaMP signal. Peri-event plots were then generated by averaging changes in fluorescence (ΔF/F) across two different air-puff trials to generate a mean change in reactivity. A minimum of two air-puffs were collected for each subject and averages were combined into corresponding groups to compare group means.

Locomotor behavior

Locomotor activity was assessed in a clear, plexiglass open field apparatus (50 cm × 50 cm) which was located within a sound dampened activity cabinet. Behavioral sessions were recorded and analyzed with ANY-maze video tracking system (Stoelting Co, IL, USA). Behavioral measurements were collected within two time periods: a 10-minute habituation prior to the onset of the air puff and then a 5-minute ISI immediately following exposure to the air puff. Analysis across time were calculated at 1 min resolution for the overall locomotor behaviors, and additionally at 1 s for the immediate post puff behavioral analysis to correspond with fiber photometry time points. Measures of locomotor activity included distance traveled (m), max speed (m/s), % time spent in the center zone, and time immobile (at 65% sensitivity, animal must remain immobile for at least 2 s). The center zone was defined within the tracking software ( ~ 7.6 cm × 7.6 cm; 45% of total area), and the time spent variable was calculated as (time spent in center zone/(total time spent in center + time spent in periphery)) × 100).

Perfusion

Transcardial perfusions were conducted in both retrograde tracing and fiber photometry experiments to prepare tissue for IHC processing. Animals were anesthetized with isoflurane until rendered insensitive to toe-pinch. Subsequently, an incision was made in ventricle of the heart and the circulatory system was flushed with phosphate buffered saline (PBS) followed by ~120 mL of 4% paraformaldehyde (w/v). Brains were stored for 24 h in 4% paraformaldehyde (w/v) and then transferred to a 30% sucrose solution for storage until sectioning.

Immunohistochemistry

To measure changes in PVT activation, 40μm sections (AP coordinates −2.9 mm to −3.6 mm) were collected and prepared for immunohistochemistry (IHC) to stain for the protein c-Fos. Brain slices were rinsed in PBS, treated with a 50% methanol/PBS solution for 30 min, and exposed to 3% hydrogen peroxide for 5 min. After a blocking step with 0.3% Triton X-100 and 1% BSA at room temperature for 1 hr, slices were incubated with rabbit anti-cFos (1:3000, Millipore) at 4 °C for 24hrs. Following washes in TNT buffer (0.1% Tween-20 in TBS), slices were blocked with TNB buffer for 30 min, then incubated with goat anti-rabbit HRP (1:200, Abcam #ab6721) for 2hrs. Fluorescent labeling was achieved using tyramide-conjugated fluorescein (1:50, Akoya) in amplification diluent for 10 min at room temperature. Slices were washed, mounted with DAPI-containing antifade medium, and stored at 4 °C. Imaging was performed using a Keyence BZ-X800 fluorescence microscope. Imaging and quantification were performed by experimenters that were blind to the condition and sex of each animal. Images were taken at 20× magnification and stitched together using the BZ-X800 Analyzer program. Regions of interest were outlined and cell quantification was performed by manual counting using the FIJI multipoint counter tool.

Statistics and reproducibility

Fiber photometry signal collection by TDT RZ5 processor and TDT Synapse program and subsequent data analysis were performed with 500 ms time bins using Matlab (Version 2023a, The Mathworks, Inc.) and a custom Matlab script (10.5061/dryad.3ffbg79qr)⁴³. For all comparisons of c-Fos/retrobead data, unpaired t-tests were used. For all comparisons between baseline and injection/follow-up session FP recordings, 2-way ANOVAs were used with Sidak multiple comparisons for any post-hoc analysis done following discovery of interaction. AUC and PP in control and experimental conditions were compared using paired t-tests. In all cases, p < 0.05 was the threshold for statistical significance and relative significance is reported as follows: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. All data were graphed and statistically analyzed with Prism 10.0.3 (GraphPad Software). Subjects were excluded from data analysis if site verification revealed incorrect injection location and/or fiber placement outside of the PVT. Sample sizes for all experimental groups are listed within figure legends. Experimenters were blinded during manual cell counting and manual behavioral scoring. The experimenter was not blinded to condition during experimentation or during data analysis as these were all automated processes measuring changes in physiology and, therefore, did not involve any subjective interpretation during data collection. Sample sizes were determined based on previous work⁴³.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

D.P.E. contributed to the initial conception, experimental design, apparatus design, surgical procedures, behavioral recording, histology, data analysis, custom code, data visualization, the initial draft of the manuscript, and manuscript editing. J.L.H. contributed to the behavioral recording, behavioral analysis, a section in methods, and manuscript editing. S.G.Q. contributed to histological analyses, schematic illustrations, and manuscript editing. C.S.R. contributed to histology. D.T. contributed to histology. M.E.S. contributed to histology. M.W.H. contributed to histology. C.W.H. contributed to the behavioral recording and manuscript editing. M.A.H. contributed to the initial conception, experimental design, behavioral assay, and acquisition of funds.

Peer review

Peer review information

Nature Communications thanks Mikael Palner and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

All data generated in this study, including a source data file containing raw data values for each figure, have been deposited in the Dryad database 10.5061/dryad.fbg79cp7b. Source data are provided with this paper.

Code availability

Custom-written MATLAB code and resulting MATLAB structures have been made available on the Dryad database 10.5061/dryad.fbg79cp7b.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-71481-1.