Psilocybin Combines Rapid Synaptogenic And Anti-Inflammatory Effects In Vitro

The introduction frames the central nervous system as a generally growth-inhibitory environment maintained by factors such as the Nogo signalling system, while neurotrophins like BDNF and GDNF act to oppose that restriction and promote structural plasticity. Interest in pharmacological agents that enhance neuronal plasticity has revived with research on serotonergic psychedelic compounds, several of which have been reported to induce neurite outgrowth, spine formation and synaptogenesis. Psilocybin, a prodrug rapidly dephosphorylated to the active metabolite psilocin, engages serotonergic receptors (notably 5-HT2A) and is under clinical investigation for several neuropsychiatric conditions; however, mechanisms underlying its rapid and long-lasting effects on synaptic and immune-related processes remain poorly defined. Smedfors and colleagues set out to characterise acute and short-term cellular effects of psilocybin in vitro, focusing on synaptic markers in primary mouse hippocampal and cortical neuronal cultures and on inflammatory responses in an immortalised microglial (IMG) cell line. The study aims to map time courses of pre- and postsynaptic protein expression from minutes to days after exposure, to assess regional specificity (hippocampus versus cortex), and to test whether psilocybin modulates microglial cytokine release after an LPS challenge, thereby probing both plasticity-promoting and anti-inflammatory actions.

Primary neuronal cultures were prepared from embryonic day 18 C57BL/6 mouse hippocampus and cortex. Cells were dissociated with trypsin, plated on polyornithine-coated coverslips (hippocampal density 37,500 cells/well; cortical 75,000 cells/well), maintained in Neurobasal-based media with appropriate supplements, and analysed at day in vitro (DIV) 17. An immortalised mouse microglial (IMG) cell line was cultured in high-glucose DMEM with 10% foetal bovine serum and used up to 10 passages. Animal housing conditions are reported for the source animals but all experiments described are in vitro. Psilocybin (Merck) was used for cell treatments; the authors note that psilocybin is converted to psilocin in vitro and attribute observed effects to psilocin. The extracted text gives a final concentration for the synapse experiments as "1x10M", but the exponent is unclear in the extraction and thus the exact concentration is not clearly reported for those experiments. For the microglial inflammation assay, a concentration series of 10-5, 10-6, 10-7, 10-8, and 10-9 M psilocybin was tested, with n = 3 biological replicates per group. Vehicle controls contained the same acetonitrile concentration used in psilocybin dilution. Synaptic effects were probed over multiple timepoints: hippocampal cultures were assessed at short intervals (5, 15, 30 min; 1, 2, 3, 4, 5, 6 h) and longer intervals (24 and 72 h) depending on the marker. Presynaptic Piccolo and postsynaptic Homer1 were measured at 5 min through 24 h (with key reporting at 1, 3, 6 and 24 h), while Synapsin-1 was quantified across eleven timepoints including 72 h. Immunocytochemistry used Piccolo, Homer1 and Synapsin-1 antibodies, phalloidin to visualise dendrites, and DAPI for nuclei. Imaging for synaptic markers used Zeiss Airyscan confocal microscopes (63x oil, NA 1.4) with second-order dendrites imaged from ten cells per well. SynQuant software was employed to detect and quantify synaptic puncta (number, size, intensity); settings were optimised on a subset of images. Imaging and analysis were performed blinded to treatment. For the inflammation assay, IMG cells were preincubated with psilocybin for 1 h, then challenged with 10 ng/mL LPS for 15 h. Media were collected for ELISA measurement of mouse TNF-α and IL-6 (BioLegend kits), and cell viability was assessed by an MTS assay. Statistical analysis of synaptic puncta and cytokine data used generalised linear models (GLM) followed by post-hoc comparisons with Holm adjustment; the multicomp package was used for inflammatory assays. Analyses were performed in R/RStudio; significance thresholds and figure notation are reported by p-value ranges.

In hippocampal neuronal cultures, psilocybin produced rapid, biphasic changes in synaptic markers. Counts of presynaptic Piccolo and postsynaptic Homer1 puncta per µm were significantly increased at 1 h (Piccolo p = 0.000049; Homer1 p = 0.0053) and 3 h (Piccolo p = 0.000049; Homer1 p = 0.014) after treatment. Colocalised Piccolo–Homer1 puncta (sites where pre- and postsynaptic markers overlap) also increased markedly at 1 h (p = 0.0000015) and 3 h (p = 0.0000016). Both proteins' puncta densities fell below baseline at 6 h (Piccolo p = 0.046; Homer1 p = 0.0000058) and returned toward baseline by 24 h. Intensity measures per punctum mirrored the number changes, with significant intensity increases for Piccolo at 1 h (p = 0.00000766), 3 h (p = 0.000000092) and 6 h (p = 0.00061095), and for Homer1 at 1 h (p = 0.000008698), 3 h (p = 0.000007855) and 6 h (p = 0.000008931), before normalisation at 24 h. Using Synapsin-1 as a second presynaptic marker revealed an early rise beginning by 15 minutes and a peak at 3 h, followed by a decline by 4 h. Notably, a second significant increase in the total number of Synapsin-1 puncta occurred at 72 h. At that later timepoint the number of puncta was high but their fluorescence intensity was not increased, which the authors suggest could reflect less active or dormant synapses. In primary cortical neuronal cultures the direction of effects differed. Piccolo puncta density decreased significantly at 1 h (p = 0.00157) and 3 h (p = 0.000037), and colocalised Piccolo–Homer1 puncta were reduced at 1 h (p = 0.012) and 3 h (p = 0.0068). Homer1 puncta showed a non-significant decreasing trend overall, with no individual timepoint differing significantly from control. The size and intensity of puncta in cortical cultures were generally unchanged across treatments. In the IMG microglial model, LPS (10 ng/mL) induced a clear increase in TNF-α secretion. Pretreatment with psilocybin produced a dose-dependent trend toward reduced TNF-α, with a statistically significant reduction at 10-5 M (p ≈ 0.00035). A trend toward decreased TNF-α was also observed at 10-7 M. IL-6 levels showed a downward trend at the highest psilocybin dose but the change did not reach statistical significance. MTS viability assays indicated that the highest psilocybin dose used (10-5 M) did not reduce cell viability alone or in combination with LPS. The inflammation experiment used n = 3 biological replicates per group; synapse quantification sampled second-order dendrites from ten cells per well and analyses were blinded.

Smedfors and colleagues interpret their findings as evidence that psilocybin rapidly and differentially modulates synaptic protein expression in a region- and time-dependent manner, producing a short-lived peak in hippocampal pre- and postsynaptic markers at 1–3 h and a later presynaptic increase at 72 h. They suggest these effects reflect a transient ‘‘window of plasticity’’ that might be exploitable for timed adjunctive therapies, noting that the early intensity increases are consistent with prior associations between heightened synaptic protein fluorescence and increased functional synapses. The lack of increased Synapsin-1 intensity at 72 h led the authors to propose that newly counted presynaptic puncta at that time might be less active or dormant. The inverted response in cortical cultures—early presynaptic downregulation without strong postsynaptic changes—led the investigators to hypothesise a pruning-like effect, where psilocybin (via serotonergic signalling) could remove weaker or ‘‘erroneous’’ synapses prior to formation of new connections. The discussion links these cell-type and region-specific patterns to the complex pharmacology of psilocin, an unspecific serotonergic agonist that can activate multiple 5-HT receptor subtypes with different temporal dynamics. The authors further contextualise their synaptic results with in vivo reports of psilocybin-induced changes in synaptic markers and gene expression, and with studies showing prolonged increases in presynaptic proteins such as SV2A after single doses. Regarding inflammatory effects, the authors report that psilocybin reduced LPS-induced TNF-α secretion from IMG cells at the highest tested concentration, while IL-6 showed a non-significant downward trend. They note complementary literature showing immunomodulatory effects of psychedelics and identify candidate intracellular mediators mentioned in other studies—such as IkBα and Dusp1—that could link synaptogenic and anti-inflammatory pathways, but acknowledge that mechanistic details remain to be elucidated. The discussion also recognises variability across studies (for example, differences in timing and tissue versus cell-culture models) as a likely reason for some discrepancies with prior reports. Limitations implied in the text include the in vitro nature of the experiments, regional and species comparison caveats with in vivo studies, and incomplete mechanistic resolution; the extracted text ends mid-sentence while describing Dusp1, indicating the full discussion and any further limitations or future directions are not completely available in the provided extract.