Chronic administration of psilocybin to cafeteria-diet-induced obese rats produced modest but significant reductions in weight gain, high-calorie food intake and central adiposity at the low and high doses. An intermediate psilocybin dose required premature termination, and the positive control metformin produced larger reductions in weight gain than psilocybin.
Background
There are currently relatively few effective pharmacological treatments for obesity, and existing ones may be associated with limiting side-effects. In the search for novel anti-obesity agents, drugs that modify central serotonergic systems have historically proven to be effective in promoting weight loss. Psilocin, which is rapidly metabolized from psilocybin, is an agonist at multiple serotonin receptors. In the present study we assessed the effects of psilocybin and a positive control (metformin) on changes in body weight in a rat model of obesity.
Methods
Five groups of adult male rats were pre-conditioned with a cafeteria diet until obese (>600 g) and then treated with either psilocybin (0.1, 1, or 5 mg/kg, i.p.), metformin (300 mg/kg, p.o.) or vehicle control. Treatments were for 27 consecutive weekdays, and body weights and high calorie food intake were recorded daily. Fasting glucose levels were recorded after 11 days of treatment. At the end of treatment rats completed a glucose tolerance test, and multiple fat pads were dissected out to assess adiposity.
Results
The medium dose psilocybin group had to be terminated from the study prematurely. Both the low and high dose psilocybin groups caused a significant decrease in changes in body weight compared to controls. The metformin group produced a greater decrease in change in body weight than either psilocybin groups or controls. Both high dose psilocybin and metformin decreased consumption of the high calorie diet, and exhibited decreased central adiposity.
Conclusion
Psilocybin demonstrated modest but significant effects on weight gain. Further study is recommended.
Papers cited by this study that are also in Blossom
Nichols, D. E. · Journal of Antibiotics (2020)
Simonsson, O., Sexton, J. D., Hendricks, P. S. · Journal of Psychopharmacology (2021)
Hendricks, P. S., Johnson, M. W., Griffiths, R. R. · Journal of Psychopharmacology (2015)
Meinhardt, M. W., Güngör, C., Skorodumov, I. et al. · Neuropsychopharmacology (2020)
Hutson and colleagues frame the study within the longstanding interest in serotonergic agents as anti-obesity treatments. They note that drugs which raise synaptic serotonin or act at specific serotonin receptors (notably 5-HT1B, 5-HT2C and 5-HT6) have historically reduced feeding and body weight, but many such agents were withdrawn because of adverse effects. Psilocybin, metabolised to the active compound psilocin, is an agonist at multiple serotonin receptors and has become of interest because of its safety profile in human research and associations in epidemiological data with lower odds of overweight/obesity and improved cardiometabolic self-reports. The authors therefore identify a gap: whether chronic administration of psilocybin can modulate weight gain in a controlled preclinical model of obesity. The present study aimed to determine whether repeated administration of psilocybin alters weight gain in a validated rodent model of diet-induced obesity. Using a cafeteria diet to induce obesity in Sprague-Dawley rats, the investigators compared three doses of psilocybin (0.1, 1 and 5 mg/kg) against an oral metformin positive control (300 mg/kg) and a vehicle control over 27 weekday treatments. Outcome measures included relative change in body weight, consumption of the high-calorie diet, fasting glucose, glucose tolerance and fat-pad measures of adiposity. The authors justify the approach by noting the translational predictive validity of rodent obesity paradigms for human weight outcomes and include a “microdose” condition to reflect emerging interest in sub-perceptual dosing.
Ninety adult male Sprague-Dawley rats (275–300 g) were acclimated and then given ad libitum access to both standard rat chow and a high-calorie cafeteria diet until reaching approximately 600 g. Animals were pair-housed on a reverse light cycle and stratified by body weight into treatment groups. Treatments were administered daily Monday–Friday for a total of 27 drug treatment days; weekend access and feeding were handled by animal-care staff because of COVID-19 restrictions, so chow intake on weekends was not measured by the research team. Five treatment conditions were planned: psilocybin at 0.1, 1.0 and 5.0 mg/kg dissolved in sterile saline and given by intraperitoneal (IP) injection, an oral gavage of metformin at 300 mg/kg, and vehicle control. Solutions were prepared fresh daily and administered in a volume of 1 ml/kg; IP injection sites were rotated. Psilocybin was given IP because of limited data on oral dosing in rats. The medium-dose psilocybin (1 mg/kg) group was terminated early after an accidental incorrect treatment at midpoint; their mid-study data are reported for informational purposes only. The cafeteria diet was a palatable paste composed chiefly of crackers, chocolate hazelnut spread and peanut butter, supplemented with mini-marshmallows and crushed chips on weekdays; nutritional values were calculated from commercial sources. Primary outcomes were relative change in body weight (percent change from baseline) and relative food consumption (food consumed as percent of body weight), measured daily on weekdays. Secondary physiological measures were fasting blood glucose measured pre-treatment, after 10–11 days, and at study end; an intraperitoneal glucose tolerance test (IGTT) after an overnight fast at study end with measurements at 60 and 120 minutes; and dissection and weighing of fat pads (subcutaneous, left retroperitoneal and left perirenal) normalized to body weight. Longitudinal outcomes were analysed by repeated-measures ANOVA, non-longitudinal continuous data by t-test or one-way ANOVA, and associations by Pearson correlation. Fisher's exact test was used for categorical comparisons with small cell counts. Significance was set at p < 0.05 and Fisher's LSD used for post-hoc comparisons.
Rats were fed the cafeteria diet for 51 days prior to treatment and, at baseline, cafeteria-fed groups weighed more than the standard chow control (mean 609.7 g vs 548.6 g; overall group difference F(5,87)=3.11, p < 0.05). The medium-dose psilocybin group was removed after an incorrect administration; their available midpoint data are reported only for information. Analysis of relative weight gain across the 27 treatment days showed significant effects of drug treatment [F(4,68)=6.92, p < 0.001], time [F(26,1768)=261.359, p < 0.001], and a drug-by-time interaction [F(104,1768)=3.391, p < 0.001]. Post-hoc tests indicated that the high-calorie diet control group exhibited the greatest relative weight gain. The low-dose psilocybin group (0.1 mg/kg) had significantly less relative weight gain than the high-calorie control (p < 0.05), and the high-dose psilocybin group (5 mg/kg) also showed reduced relative weight gain (p < 0.01). Metformin produced the largest reduction in relative weight gain compared with all other high-calorie diet groups (p < 0.05). Exploratory day-by-day analyses suggested rapid onset of effect, with reduced relative weight gain evident by day 2 for the medium- and high-dose psilocybin groups and for metformin. When relative food consumption (percent of body weight) was analysed over the full treatment period, there was no significant main effect of drug treatment [F(3,55)=1.990, p=0.126], although time was significant [F(24,1320)=4.782, p < 0.001]. Follow-up comparisons based on a priori predictions showed that both the metformin and high-dose psilocybin groups consumed relatively less of the high-calorie diet than controls across the treatment period (p < 0.05). Analysing consumption in shorter bins revealed a significant group effect for days 1–6 [F(4,73)=3.12, p < 0.01] and a trend for days 7–13 [F(4,73)=2.32, p = 0.06], with differences disappearing later in treatment. From days 1–6, mean consumption (percent of body weight) was: control 3.22 ± 0.14%, low psilocybin 2.79 ± 0.13%, high psilocybin 2.81 ± 0.13%, and metformin 2.64 ± 0.14%. Total caloric intake over the treatment period followed the same pattern (control 4430.9 ± 221.5 kcal; low psilocybin 4145.9 ± 211.5 kcal; high psilocybin 3955.5 ± 233.2 kcal; metformin 3705.7 ± 130.9 kcal). The ANOVA for total calories did not reach significance [F(3,55)=2.04, p=0.120], but metformin consumed significantly fewer calories than vehicle control (p < 0.05). Only the metformin group left mini-marshmallows uneaten (115 total over treatment; p < 0.001 vs others). Fasting glucose at the midpoint did not differ between groups [F(5,87)=1.736, p=0.136]. In the IGTT at study end, the normal diet (no cafeteria access) group showed significantly lower glucose levels than all high-calorie groups, particularly at the 1 h timepoint; no differences were observed among the high-calorie diet groups. Fat-pad analyses normalised to body weight showed no group difference for perirenal fat [F(4,72)=0.995, ns], but significant group effects for subcutaneous [F(4,72)=3.089, p < 0.05] and retroperitoneal fat [F(4,72)=10.282, p < 0.001]. Post-hoc comparisons indicated that for subcutaneous fat the normal diet group differed from the high-calorie groups except metformin, while for retroperitoneal fat the normal diet, high-dose psilocybin and metformin groups had lower relative adiposity than the high-calorie control. Correlational analyses found that subcutaneous (r = 0.528, p < 0.001) and retroperitoneal fat (r = 0.544, p < 0.001) were associated with glucose intolerance at 1 h, whereas perirenal fat was not. Total relative weight gain correlated with subcutaneous and retroperitoneal adiposity and correlated strongly with total high-calorie diet consumed (r = 0.666, p < 0.001).
Hutson and colleagues interpret their principal finding as evidence that chronic psilocybin treatment produced modest but significant reductions in relative weight gain in a rat model of diet-induced obesity. Both the low (0.1 mg/kg) and high (5 mg/kg) psilocybin doses slowed weight gain relative to high-calorie diet controls, and metformin produced a larger effect than either psilocybin dose. The high-dose psilocybin and metformin groups also showed reduced intake of the high-calorie diet, particularly during the first two weeks, and both had lower relative central adiposity than the high-calorie controls. No drug treatment altered fasting glucose or glucose tolerance among the high-calorie groups, with the IGTT differences largely reflecting the lean normal-diet control group. The authors caution that “slowed weight gain” rather than absolute weight loss is a common outcome in rodent obesity models because laboratory rats generally gain weight across their lifespan; therefore relative reductions in weight gain in drug-treated animals are how preclinical efficacy is commonly manifested and interpreted for translational purposes. They note uncertainties about dose equivalence across species: allometric scaling yields human-equivalent doses of approximately 0.016, 0.16 and 0.81 mg/kg for the rat doses used, and IP administration may provide greater bioavailability than oral dosing. The medium 1 mg/kg dose used in some behavioural studies was accidentally discontinued here, and the authors acknowledge that their highest dose may already exceed typical clinical human exposures when adjusted, underscoring the need for pharmacokinetic data in future comparisons. Mechanistically, the authors highlight that psilocin acts at 5-HT2C as well as 5-HT2A and 5-HT1A receptors, and that 5-HT2C agonism is a plausible pathway to reduce feeding and influence glucose homeostasis. They point out that the low “microdose” (0.1 mg/kg) decreased weight gain without a detectable reduction in measured cafeteria food intake, which could reflect effects on metabolism, energy expenditure or selective macronutrient intake—though the study could not resolve intake of standard chow on weekends or selectivity among macronutrients because the cafeteria diet was homogenised into a paste. Several limitations are acknowledged. The study did not investigate molecular or neural mechanisms; only male rats were used, limiting generalisability across sexes; the highest psilocybin dose may not map straightforwardly onto human-relevant doses; and chronic weekday dosing may induce tolerance (including 5-HT2A downregulation) unlike the intermittent dosing strategies used in many human psychedelic trials. Practical constraints arising from COVID-19 staffing meant weekend chow intake was not measured, complicating full dietary intake assessment. The authors recommend further work using pharmacokinetic measures, higher or varied dosing schedules, intermittent dosing to reduce tolerance, inclusion of female animals, and testing in alternative metabolic models (for example the Zucker rat) to probe diabetes-related endpoints and mechanisms. Overall, the study presents preliminary preclinical evidence that psilocybin can modestly slow weight gain and reduce central adiposity in this model. The authors emphasise that effects were smaller than those produced by metformin in the present protocol and that safety, dosing equivalence and mechanistic questions remain to be resolved before translational inferences can be made. They conclude that further investigation is warranted.
Pharmacological treatments for obesity exert their therapeutic effects on a diverse range of physiological targets. Historically, one of the most successful substrates for anti-obesity drug development has been the central serotonergic system. The monoamine neurotransmitter serotonin plays a key role in multiple processes related to feeding, and mediates central effects on appetite and satiety via both hypothalamic and extrahypothalamic brain structures. Thus, drugs which elevate synaptic levels of serotonin by either direct release (amphetamines, fenfluramine) or by reuptake inhibition (sibutramine, tesofensine) have been proven to exhibit efficacy in promoting weight loss. However, this class of drugs has been associated with a number of serious side-effects, which has resulted in their withdrawal from most markets. To reduce the risk of off-target effects caused by non-specific elevations of serotonin levels, likely mediated through the 14 known serotonin receptors, there has been a focus on a more targeted approach using receptor-specific ligands. The comprehensive preclinical literature suggests that at least the 5-HT1B, 5-HT2C, and 5-HT6 receptors play an important role in satiety, and hold promise for therapeutic development of weightloss drugs. For example, the 5-HT2C agonist lorcaserin proved to be an effective treatment for obesity before its recent withdrawal in the US due to concerns about an increased risk of cancer. The serotonergic system remains a promising substrate for developing new anti-obesity medications, despite the list of previously-withdrawn compounds-especially if products can be identified that have minimal affinity for receptors leading to serotonergic side-effects, such as the 5-HT2B receptor which has been associated with valvulopathies. One potential compound that mediates its downstream effects via serotonin receptors, and which has been used extensively by humans, is the prodrug psilocybin. Found naturally in numerous species of fungi, psilocybin has been used for centuries by many cultures for religious and recreational purposes. Interestingly, recent analysis of adults who have used serotonergic psychedelics (including psilocybin) showed significantly lower odds of being overweight/obese or having cardiometabolic diseases, with higher odds of self-reported overall health, although the mechanisms underlying these observations remain to be understood. Importantly, psilocybin has seen a surge of interest in the field of psychiatry due to evidence of a rapid and sustained antidepressant effect. Evidence indicates a bidirectional association between depression and obesity, and several of the pharmacological treatments used for the former show benefits in the latter, in preclinical and clinical studies. It remains to be seen whether serotonergic psychedelics may have weight modulating properties which warrant further study as a potential therapeutic modality. In particular, psilocybin is well-suited as a candidate to study as it is considered to be amongst the safest of all the psychedelic drugs with regards to somatotoxic and psychological side-effects. The purpose of the present study was therefore to determine whether chronic treatment with psilocybin would exert effects on weight gain in an animal model of obesity. Rodent paradigms of obesity have strong predictive validity with regards to drug development, in which most compounds that affect weight gain in rodents also affect weight in humans, and vice versa. We used an established model of obesity in which rodents are voluntarily fed a high calorie "cafeteria" diet and rapidly gain weight. For dosing psilocybin, we included two common doses of the drug (1 and 5 mg/kg) as well as a "microdose" (0.1 mg/kg). The rationale for the latter dose is the emerging interest in ingesting lower doses of psychedelics that have sub-perceptible effects (typically about 1/10th of their recreational dose).
Ninety male, adult Sprague-Dawley rats (275-300 g) were obtained from Charles River (Montreal, QC). The animals were allowed to habituate in the UBC animal colony before being exposed to the high calorie diet. Rats were randomly housed in pairs in large polycarbonate cages on ventilated racks in a reverse-light cycle, temperature-controlled (22 ± 1 • C) room and maintained on a 12-h light-dark cycle (lights on at 07:00 h). All rats, except one control group, were provided regular rat-chow and were given ad libitum access to both the rat-chow and the cafeteria diet to reach a desired weight (≈600 g) prior to experimental procedures. Cafeteria diets were prepared freshly and leftover diet food was removed and weighed daily (weekdays). Animals were familiarized to handlers through being weighed every Monday, Wednesday, and Friday for at least 30 days prior to drug treatment. The animals were all experimentally naïve, and stratified according to their body weight to obtain body weight-matched treatment groups. Animals were treated daily Monday-Friday, to be consistent with a recent study which used psilocybin to reduce alcohol seeking behavior in rats, and had a total of 27 drug treatment days; access to the colony was restricted on weekends due to safety measures related to the global COVID-19 pandemic. This latter point meant that regular rat chow was required to be changed on a daily basis by animal care staff, and so could not be measured by those running the present experiments. Animal were also monitored for signs of general health on a daily basis, as required in the protocol by the REB. This included any changes in neurological function (e.g., seizure) or general behavior, respiration, hydration/elimination, and any signs of stress and/or pain. All experimental procedures were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and approved by The University of British Columbia's Animal Care and Use Committee.
Three doses of psilocybin dissolved in sterile saline (0.9%) were used: low (0.1 mg/kg), medium (1.0 mg/kg), and high (5.0 mg/kg). Psilocybin (99.7% purity; Psygen Labs Inc., Calgary, AB, Canada) was prepared under sterile conditions by the Katz Group at the Rexall Centre for Pharmacy & Health Research at the University of Alberta. Metformin (300 mg/kg; Toronto Research Chemicals) was slightly warmed during mixing to aid in dissolution. All solutions were prepared fresh daily, and vortexed immediately prior to loading the syringe to ensure administration of a homogenous solution. Each animal received either a 1 ml/kg oral gavage of metformin, or an intraperitoneal (IP) injection of saline or the dose of psilocybin (0.1, 1, and 5 mg/kg) dissolved in a volume of 1 ml/kg of sterile saline solution. IP injection sites were rotated through the week to prevent infection and irritation. Psilocybin was administered by IP injection rather than via oral gavage (like metformin) due to the dearth of published studies on oral administration of the drug, resulting in challenges determining an optimal dose to exert behavioral effects.
The highly palatable high-fat and high-sugar diet was composed of, by weight, 10.9% crushed cheddar crackers (Goldfish: 25% fat, 65% carbohydrate, 10% protein), 53% chocolate hazelnut spread (Nutella: 31% fat, 58% carbohydrate, 6% protein) and 36% smooth peanut butter (Kraft: 47% fat, 20% carbohydrate, 27% protein). An additional 4 mini-marshmallows (100% carbohydrate) and 6g crushed chips (Ruffles regular: 38% fat, 54% carbohydrate, 7% protein) were added Monday-Thursday, and 6 mini-marshmallows and 8 g crushed chips were added on Friday. The food components were mashed into a coarse paste as per previously reported for high calorie diets, and thus the different food components were consumed in proportion to their contribution of the above makeup. The exception to this was the mini-marshmallows, which did not dissolve into the paste and so could be counted the following day if left unconsumed. Nutritional and calorie content were obtained from the commercial websites for the different food products.
Glucose measurements were collected from each rat prior to the start of drug treatment, after 10 or 11 days of drug treatment, and at the end of the study. All animals were fasted overnight prior to blood collection. Blood was collected from the lateral saphenous vein, as previously. Glucose measurements were obtained using a handheld glucometer (One Touch Ultra). At the end of the study, an intraperitoneal glucose tolerance test (IGTT) was performed, as previously described. All animals were fasted overnight, then given a glucose challenge of 1 g/ml/kg glucose administered by IP injection. Blood glucose levels were then measured at 60 and 120 min following glucose injection.
At the end of the study, the rats were euthanized by inhalant anesthetic (isoflurane), followed by decapitation to collect the brains for future studies. Fat deposits (subcutaneous [defined area], left retroperitoneal and left perirenal) were excised and weighed, as previously reported.
Longitudinal data, including daily weight gain and food consumption, were analyzed for group and time differences using repeated-measures analysis of variance (ANOVA) (SPSS Version 24.0. Armonk, NY: IBM Corp). Non-longitudinal data were compared using independent-samples t-test or one-way ANOVA, as previously. Associations between variables were determined using the Pearson correlation coefficient. Comparison of categorical variables between groups where cells had less than five counts used Fisher's exact test. Significance for all statistical tests was set at p < 0.05. Post-hoc tests were analyzed using the Fisher's LSD test.
Treatments were well-handled by the animals, who habituated quickly to the daily procedures. Rats were fed the cafeteria diet for a total of 51 days to get animals to the desired weight. At baseline, immediately prior to drug treatment, all groups previously exposed to the cafeteria diet exhibited significantly greater weight than the control group given only standard ratchow [F (5,87) = 3.11, p < 0.05], with a mean weight of 609.7 g for cafeteria-fed and 548.6 g for standard chow-only animals. At the approximate midpoint of the study, one treatment group (the medium dose of psilocybin) was accidentally treated with an incorrect drug treatment, and so treatment was discontinued for this group; their data are included at the midpoint analysis, for informative reasons only. Change in body weight over the 27 days of treatment was analyzed in relative terms (measured by change in body weight relative to baseline body weight, expressed as a percentage). Similarly, food consumption was measured in relative terms (food consumed in proportion to body weight, expressed as a percentage). When relative weight gain was analyzed, the repeated measures ANOVA indicated a significant main effect of drug treatment [F (4,68) = 6.92, p < 0.001], a main effect of time [F (26,1768) = 261.359, p < 0.001] and a drug by time interaction [F (104,1768) = 3.391, p < 0.001]. The post-hoc analysis revealed that when relative weight gain was compared across all 27 treatment days, the greatest relative weight gain was exhibited by the control group, fed the high calorie diet (Figure). The low dose psilocybin group demonstrated less relative weight gain; this was a significant effect compared to the control group (p < 0.05). The high dose psilocybin group exhibited less relative weight gain too, and this effect was also statistically significant (p < 0.01). The metformin group displayed the least relative weight gain compared to all of the other groups fed with the high calorie diet (p < 0.05). Exploratory post-hoc analyses on day-by-day changes in body weight indicated a rapid effect of psilocybin treatment, whereby relative weight gain was less in the medium and high dose psilocybin groups-as well as the metformin group-by only the second day of treatment. The low dose of psilocybin group With regards to food consumption, when relative amount of food consumed was analyzed the repeated measures ANOVA indicated a non-significant main effect of drug treatment [F (3,55) = 1.990, p = 0.126], a main effect of time [F (24,1320) = 4.782, p < 0.001] and a non-significant drug by time interaction. Follow up exploratory analyses based on a priori predictions indicated that both the metformin and high dose psilocybin groups ate relatively less of the high calorie food than the control animals over the entire treatment period (p < 0.05) (Figure). Because the analysis of weight change indicated that drug treatment had relatively rapid effects, we conducted a follow-up analysis of relative food consumption in shorter time bins (days 1-6, 7-13, 14-19, and 20-27). Interestingly, relative amount of food consumed differed significantly between groups from days 1-6 [F (4,73) = 3.12, p < 0.01] and was a non-significant trend for days 7-13 [F (4,73) = 2.32, p = 0.06], but was not significant in the latter half of the study. Post-hoc tests indicated that the high calorie diet control group consumed more food (3.22 ± 0.14% of body weight [mean ± SEM]) than all other groups given the cafeteria diet (low dose psilocybin: 2.79 ± 0.13%; high dose psilocybin: 2.81 ± 0.13%; metformin 2.64 ± 0.14% of body weight) from days 1-6, and significantly more than the high dose psilocybin and metformin groups from days 7-13 (controls: 3.19 ± 0.14%; low dose psilocybin: 2.91 ± 0.13%; high dose psilocybin: 2.63 ± 0.13%; metformin 2.72 ± 0.14% of body weight). Consumption was also analyzed based on caloric content, rather than weight, for the total amount of food consumed over the duration of the high calorie diet following drug treatment. The control group treated with saline solution consumed the most calories (4430.9 ± 221.5 kcal), followed by the low dose psilocybin (4145.9 ± 211.5 kcal), then the high dose psilocybin (3955.5 ± 233.2 kcal) and then least of all the metformin group (3705.7 ± 130.9 kcal). Similar to the relative amount of food consumed, the ANOVA did not find an overall group effect of drug treatment [F (3,55) = 2.04, p = 0.120], but based on a priori predictions from the amount of food consumed, it was indicated that the metformin group consumed significantly fewer calories than the vehicle control group (p < 0.05). Because the high calorie diet was a relatively homogeneous paste, it was not possible to determine whether the drug treatments could have caused a selective reduction in specific macronutrient (fat, carbohydrate or protein) intake, which might have been possible had the food products remained distinct. Of interest, however, only the metformin treated group had leftover minimarshmallows the following day (115 total marshmallows over the course of drug treatment, vs. none for all of the other groups; p < 0.001). Analysis of fasting glucose levels at the approximate midpoint of treatment did not indicate a significant main effect of group [F (5,87) = 1.736, p = 0.136] indicating the groups did not differ in their fasting glucose levels (Table). However, when all of the high calorie diet animals were analyzed as one group and compared to the normal diet group, there was a non-significant trend [t (86) = 1.720, p = 0.089], indicating a modest increase in fasting glucose levels for high calorie diet animals. The analysis of glucose sensitivity with the intraperitoneal glucose tolerance test at the end of treatment at the one and 2 h time points by repeated measures ANOVA revealed a group). Post-hoc analysis indicated that the significance was largely driven by the normal diet group, which exhibited significantly lower glucose levels than all of the other groups, particularly at the 1 h time point. None of the groups in the high calorie diet group differed from each other. The normal diet group exhibited the typical pattern of glucose clearance observed in non-obese Sprague-Dawley rats [e.g., as observed in our laboratory], with mildly elevated glucose levels at 1 h after glucose challenge, and a return to baseline levels by 2 h. When the weights of different fad pads at the end of treatment were analyzed, normalized to body weight (to determine relative adiposity), two of the three fat regions exhibited significant group differences (Table). While there was no group difference for perirenal fat [F (4,72) = 0.995, ns], the subcutaneous [F (4,72) = 3.089, p < 0.05] and retroperitoneal fat [F (4,72) = 10.282, p < 0.001] differed by group. Post-hoc analysis indicated that for the subcutaneous fat, the normal diet differed from all of the high calorie fed groups with the exception of the metformin group. For the retroperitoneal fat, the normal diet, high dose psilocybin and metformin groups had lower relative adiposity than the high calorie control group. Exploratory associations between weight, feeding and metabolic indices indicated that there was a significant correlation between both subcutaneous (r = 0.528, p < 0.001) and retroperitoneal fat weights (r = 0.544, p < 0.001) and glucose intolerance at the 1 h timepoint, but not for the perirenal fat (r = 0.091, NS). Total relative weight gain correlated significantly with subcutaneous and retroperitoneal adiposity, but not perirenal fat. Unsurprisingly, total relative weight gain correlated strongly (r = 0.666, p < 0.001) with total amount of high calorie diet consumed.
In the present study, we evaluated the effects of chronic treatment with three different doses of psilocybin (including a "microdose"), as well as a single dose of the oral antidiabetic drug metformin, on change in body weight in a rat model of obesity. Animals received 27 continuous weekday treatments with the drug or its vehicle. An accidental error of drug administration with the medium dose group (1 mg/kg) resulted in their early termination from the study. Psilocybin was well-tolerated by the other two treatment groups, with no obvious evidence of harmful effects, based on standard daily laboratory monitoring procedures. The principal finding of the study was that both doses of psilocybin resulted in significantly decreased relative weight gain compared to high calorie diet-fed rats treated with the control vehicle; metformin also caused decreased relative weight gain, with a greater magnitude than either of the psilocybin doses. Both the high dose psilocybin and metformin groups ate less of the high calorie diet-these effects were most pronounced in the first 2 weeks of treatment. There were no effects of any drug treatment on fasting glucose levels or glucose sensitivity in the IGTT. However, both the high dose psilocybin and metformin groups had lower relative central adiposity than the high calorie control group, and this correlated with glucose levels in the IGTT. These overall findings suggest that psilocybin may have modest weight-loss properties, which could feasibly extend to humans. While none of the groups in this study exhibited an actual reduction in absolute weight over the period of treatment compared to their baseline weight, this may reflect a limitation of the current model species. Laboratory rats will typically continue to gain weight throughout their lifetime, until at least 24-32 months of age, when they can weigh over 1 kg. In models of diet-induced obesity, high calorie fed control rats will usually continue to gain weight throughout the study, and so decreased weight gain compared to controls will often be interpreted as a "weight-loss" effect, even if there is an absolute increase in mass over time [e.g., Hansen et al.]. In the context of obese humans who are maintaining a stable weight, the slowed weight gain in psilocybin-treated rats could therefore be anticipated to correspond to a weightloss in individuals treated with the drug, rather than just a delay of any further weight gain. In support of this, both the high calorie fed group that received metformin and the control group fed normal rat chow (with no access to the high calorie diet) also continued to gain weight, with the latter group showing equivalent weight gain to the two psilocybin groups. Notably, metformin has well-established weight loss properties in obese humans, and so the manner in which the drug was only able to slow weight gain in the present study, rather than cause absolute weight-loss, again suggests that potential weight-loss effects in humans should be inferred from relative changes in the weight of drug-treated animals compared to control rats. The significantly greater effect of metformin than the two doses of psilocybin on change in body weight may indicate that the weight-altering effects of the latter drug are less effective, although uncertainties exist around equivalent dosing between the two drugs and their relative weight-altering effects in humans. The dose of metformin used in the present study would be considered at the higher end of appropriate dosing for an animal study, while the maximal physiologically relevant dose for psilocybin remains less well-established. Our dosing schedule was based on that used in a recent behavioral rat study of the effects of psilocybin on alcohol-seeking behavior, in which their dosing was determined in consultation with experts in the field, who opined that the minimum dose needed to see behavioral effects in rats is 1 mg/kg. The study byused repeated doses of psilocybin as high as 10 mg/kg. Thus, our "medium" dose may actually reflect a minimally sufficient dose to alter behavior, while our "high" dose may correspond more closely to a mid-range value. But it is also worth considering how the present doses compare to those used in human clinical trials, based on allometric comparisons between species. The typical dose of psilocybin used in landmark clinical trials for conditions such as depression is 25 mg, corresponding to a dose of approximately 0.3 mg/kg in humans. Using validated allometric guidelines, the current psilocybin doses of 0.1, 1, and 5 mg/kg equate to 0.016, 0.16, and 0.81 mg/kg Human Equivalent Doses. Additionally, the administration of psilocybin via IP injection might be expected to result in greater bioavailability than by oral consumption. Thus, it is also possible that the 5 mg/dose of psilocybin may represent a higher dose than is typically used in humans. Future studies will therefore be needed with higher doses of psilocybin to determine whether larger doses are able to affect weight gain equivalently to a high dose of metformin before final comparisons can be made between drug efficacy for weight-loss, possibly combined with pharmacokinetic indices to determine the validity of such doses to human trials. Of interest, we did not observe any effect of drug treatmentincluding metformin-on glucose levels either during a fast or following a glucose challenge in the high calorie diet animals. This may be partly because the magnitude of the effects of the diet on glucose dysregulation were modest: differences between the high calorie and regular diet groups were only significant when all animals from the high calorie diet were included as one large group. This small effect size therefore restricted the statistical power of drug treatment to produce a significant reduction in glucose values. It is important, thus, to emphasize that the current animal model is one of obesity, rather than Type 2 diabetes (T2DM). Rodent models of diet-induced obesity typically do not result in the development of diabetesalthough they can exhibit other types of metabolic dysregulation. The present model was chosen because it utilized a cafeteria diet model that included widely available supplies, is easy to dispense and measure, is economical, and does not change from day-to-day. The model was also effective previously in detecting the weight loss properties of both sibutramine and liraglutidewith chronic treatment-two drugs that promote weight loss in humans. Like the study by, our animals did not develop frank diabetes. However, in their study the cafeteria diet fed animals did demonstrate more pronounced glucose dysregulation in the GTT. As their rats were approximately 130 g heavier at the start of the study, it is possible that the current starting weight of animals in our study was not yet sufficiently large enough to maximize metabolic dysregulation. Nevertheless, it will be important to test the effects of psilocybin using more complex paradigms of obesity and metabolic syndrome/diabetes, using models such as the Zucker ratto assess the entirety of metabolic changes that might be improved. The mechanism(s) by which psilocybin slowed weight gain in the high calorie diet rats will require further study, as they were beyond the scope of the present series of experiments. It is possible that an important part of the effect is due to reduced intake of the high calorie diet-this was statistically significant for both the high dose psilocybin and metformin groups, but not the low dose psilocybin group, indicating that additional factors may be involved. Due to restrictions around weekend access to the colony, animal care staff were in charge of providing regular rat chow on a daily basis (which was coprovided with the high calorie diet, ensure that animals' nutritional needs were met) and thus consumption of regular chow could not be measured by research staff, nor could the high calorie cafeteria diet be provided to animals on these days. It is possible that the low dose psilocybin group selectively reduced their consumption of the less palatable regular chow rather than the high calorie diet; while this option seems counterintuitive, it will need to be dispelled by future experiments that measure the consumption of all food sources. The pharmacology of psilocybin is certainly consistent with effects on feeding behavior. Psilocybin is rapidly dephosphorylated in the body by alkaline phosphatase to the active metabolite psilocin. Both in vitro and in vivo studies have confirmed that psilocin is an agonist at 5-HT1A, 5-HT2A, and 5-HT2C receptors with moderate-tohigh affinity. Of these, the 5-HT2C receptor has been demonstrated repeatedly to play a key role in satiety (52), and treatment with 5-HT2C agonists results in weight-loss while 5-HT2C antagonists cause weight gain. 5-HT2C agonists may also reduce feeding through other mechanisms, including suppressing conditioned responding and impulsivity (56), which will require more complex animal feeding models to evaluate for psilocybin. In addition to effects on feeding, there is also a substantial body of evidence which indicates that 5-HT2C agonists can produce effects on blood glucose levels and insulin sensitivity independent of changes in weight and feeding. Interestingly, these drugs (such as meta-chlorophenylpiperazine and lorcaserin) were able to improve glucose homeostasis at doses well below those needed to reduce food intake. This may be relevant to the present study, where we observed that treatment with the low "microdose" (0.1 mg/kg) of psilocybinat a dose that is considered only 1/10th that needed to alter behavior in rats ()-was able to significantly decrease weight gain in the absence of any change in consumption of the high calorie diet. It is also possible that psilocybin may have altered weight by affecting metabolism and resting energy expenditure. Ongoing studies are presently addressing these possibilities. While the present findings represent a novel study of the potential effects of psilocybin on obesity, there are a number of important limitations. Firstly, we are not able to provide any insight into the molecular mechanisms that are involved in decreased weight gain. The current study was designed as an exploratory examination of the effects of psilocybin on obesity, with further study contingent on an observable effect. This phenomenon has now been produced twice in our laboratory (a pilot study showed similar effects) and so studies to determine the neural/metabolic substrates involved are proceeding. Secondly, we only studied the effects of psilocybin in male rats. Rat models of obesity typically use male rather than female animals to avoid the complicating effects of the estrous cycle, but female rodents exhibit meaningful differences in energy homeostasis and adiposity, and so it will be critical to determine whether similar benefits are observed with female rats too. Thirdly, as noted above, we may have potentially under-dosed with psilocybin at the higher end of our dosing regimen. If effects are dose-dependent, it will be important to assess where the maximal effects of psilocybin peak on weight gain at physiologically relevant doses. Use of psilocybin is complicated by regulatory requirements, access to high purity product, and (typically) cost, but chronic studies with higher doses of the drug may be necessary to make further advances in this area. However, as noted above, allometric comparison of the current doses in rats indicates that the 5 mg/kg dose may already be exceeding the Human Equivalent Dose typically used in clinical trials. Therefore, pharmacokinetic measures (such as plasma levels of psilocybin/psilocin) may be helpful in cross-species comparison to determine if higher doses are justified, as well as use of alternate behavioral assays in rodents of serotonergic activity [such as the head twitch response/wet dog shakes] in parallel. Finally, while not necessarily a limitation, the present study used repeated, chronic treatment with psilocybin, whereas most human trials typically use a much small number of doses, often in combination with psychotherapy. Repeated exposure to psychedelic drugs can result in behavioral toleranceand downregulation of the 5-HT2A receptor. This may have even occurred in the present study, as decreased food intake diminished with time more rapidly with the psilocybin than metformin groups. Thus, intermittent treatment with psilocybin may represent a future direction of study to maximize potential weight loss by preventing behavioral tolerance. In summary, the present study provides preliminary findings from a rodent model of obesity that psilocybin has potential weight reducing properties, which are relatively fast onset. Obesity represents a huge unmet medical need in many countries, with approximately 40% of the US population being affected. A recent study noted that drugs used to treat T2DM were prescribed 15 times more frequently than those used specifically for obesity. This reflects in part the small number of drugs indicated for obesity (six currently approved by the US Food and Drug Administration: phentermine/topiramate, naltrexone/bupropion, liraglutide, semaglutide, orlistat, and setmelanotide)and the issue of side-effects as well as cost. The potential of psilocybin as a weight-loss tool, whereby it could be consumed orally-possibly at a non-psychoactive dose, and therefore on a more frequent basis-may have value, even if weight loss effects were only modest; it is important to reinforce that none of the doses of psilocybin used in our study exerted effects as strong as those observed with metformin. Issues remain about the drug's safety, as we are not aware of any large-scale studies of chronic treatment with the drug, but concerns about the cardiac valvular problems associated with earlier serotonergic drugs may be less warranted, given psilocin's lesser affinity for the 5-HT2B receptor. Psychological side-effects of psilocybin treatment may also be lessened if "microdoses" could be used to obtain weight-loss benefits. These possibilities suggest that further study on this topic should proceed.
Create a free account to open full-text PDFs.
Johnstad, P. G. · Nordic Studies on Alcohol and Drugs (2018)
Ona, G., Bouso, J. C. · Neuroscience and Biobehavioral Reviews (2020)
Kuypers, K. P. C., Erritzoe, D., Knudsen, G. M. et al. · Journal of Psychopharmacology (2019)
Jefsen, O., Højgaard, K., Christiansen, S. L. et al. · Acta Neuropsychiatrica (2019)
Carhart-Harris, R. L., Giribaldi, B., Watts, R. et al. · New England Journal of Medicine (2021)
Carhart-Harris, R. L., Bolstridge, M., Rucker, J. et al. · Lancet Psychiatry (2016)
Dinis-Oliveira, R. J. · Drug Metabolism Reviews (2017)
Nichols, D. E. · Pharmacology and Therapeutics (2004)
Papers in Blossom that reference this study
Baraghithy, S., Gammal, A., Permyakova, A. et al. · ACS Pharmacology and Translational Science (2024)
Tagen, M., Mantuani, D., Van Heerden, L. et al. · Journal of Psychopharmacology (2023)