Salvia divinorum: from recreational hallucinogenic use to analgesic and anti-inflammatory action

Salvia divinorum is a herbal plant from southwestern Mexico long used in Mazatecan medical and religious practice. Traditional preparations have been applied to ailments that overlap with inflammatory conditions and pain such as headaches, gastrointestinal complaints and rheumatism, while modern extracts with elevated salvinorin A concentrations are widely used recreationally for their potent, short-lived hallucinogenic effects. Phytochemical work has identified a series of neo-clerodane diterpenes (salvinorins A–F), with salvinorin A (SA) the dominant active component; pharmacological screening shows SA is a high-affinity agonist at κ opioid receptors (KORs) and modulates cannabinoid type 1 (CB1) receptors, but lacks action at the classical 5-HT2A serotonin receptor targeted by psilocybin and LSD. This review aims to synthesise available evidence on S. divinorum, SA and semi-synthetic analogues with emphasis on analgesic and anti-inflammatory effects, while also describing psychoactive properties and safety signals. Coffeen and Pellicer focus mainly on preclinical pharmacology, mechanistic biochemical data, animal models of inflammation and pain, and early reports in humans and poison-control data to consider therapeutic potential and remaining translational challenges.

The extracted text presents a narrative review rather than a methods-driven systematic review: no explicit search strategy, databases, date ranges, or formal inclusion/exclusion criteria are reported in the provided sections. The investigators draw on ethnobotanical reports, biochemical receptor screening, in vitro experiments, animal models (rodent and zebrafish), selected human experimental and observational reports, and studies of semi-synthetic analogues to assemble the account. Because primary study designs vary across the sources cited, the review integrates receptor binding and signalling data, pharmacokinetic and imaging findings (for example PET in rodents and baboons), behavioural assays of reward and nociception (conditioned place preference, self-administration, tail-flick, hot-plate and formalin tests), models of intestinal inflammation and colitis, and toxicology results from in vivo and in vitro studies. The extracted text does not describe a formal risk-of-bias assessment or quantitative meta-analysis approach.

Psychoactivity: Salvinorin A produces a distinct profile of brief, intense subjective effects. Field reports of traditional aqueous leaf preparations described lightheadedness, altered proprioception and amplified imagery without frank hallucinations, but modern concentrated extracts give potent hallucinatory experiences in humans. The reported hallucinogenic dose of SA in humans is 250–500 µg and its potency is slightly lower than LSD. Experimental human reports and animal work indicate rapid onset and short duration of action: intravenous SA in baboons reached 3.3% of the injected dose in brain within 40 seconds and cleared to half the peak by 8 minutes, consistent with rapid smoked effects in humans. Neuroimaging and electrophysiology studies implicate KOR-rich regions (periaqueductal grey, bed nucleus of the stria terminalis, cerebellar vermis, hypothalamus, auditory and sensory cortices) and changes in anterior versus posterior cortical spectral power. Behavioural models of reward show strain- and species-dependent results: SA induced conditioned place preference in zebrafish and Wistar rats (with increased nucleus accumbens dopamine), but failed to support self-administration in some rat strains; the rewarding effects in zebrafish and some rodent models were blocked by the KOR antagonist nor-BNI and by the CB1 antagonist rimonabant. Chronic human use can produce withdrawal-like symptoms (anxiety, irritability, malaise) and functional impairment consistent with substance use disorder criteria in some cases. Inflammation: Ethnobotanical use for gastrointestinal disorders prompted preclinical studies showing anti-inflammatory actions of SA and extracts. In mouse models of intestinal inflammation, SA inhibited motility and hypermotility during inflammation via mechanisms involving KOR and CB1 receptors and in some assays CB2, suggesting cross-talk between opioid and endocannabinoid systems in the gut. In vitro, SA reduced inducible nitric oxide synthase (iNOS) expression following LPS challenge in macrophages, lowered pro-inflammatory TNF-α and restored anti-inflammatory IL-10 production, and prevented epithelial barrier dysfunction via NOS-2–related pathways. SA inhibited leukotriene B4 (LTB4) biosynthesis in a concentration-dependent manner in isolated macrophages and reduced LT levels in zymosan-induced peritonitis and carrageenan pleurisy, indicating potential utility against LT-mediated inflammatory states. Intraperitoneal, oral and intracolonic SA showed anti-inflammatory effects in two experimental colitis models and produced analgesia in mice with ongoing intestinal inflammation; these effects were mediated by KOR and CB1 receptors. Analgesia: A body of preclinical evidence indicates SA and S. divinorum extracts have antinociceptive properties. Peripheral administration produced dose-dependent effects in thermal (tail-flick) and chemical (acetic-acid abdominal constriction) nociception assays as well as in the hot-plate test; analgesia was reversed by nor-BNI, implicating KOR. Co-administration of a hydroponic S. divinorum extract with anandamide (a cannabinoid agonist) produced synergistic analgesia in the hot-plate test at low doses. Central administration studies showed intrathecal SA increased tail-flick latencies at 13.9–23.1 nmol, and intracerebroventricular SA or a derived ester produced analgesia in wild-type but not KOR knockout mice. Ligand-binding assays report high affinity of SA for the κ1 receptor (K i 18.7 ± 3.4 nM) and very low affinity for a κ2 subtype (K i >10,000 nM), indicating receptor subclass selectivity. Repeated SA administration over about 7 days reduced formalin-induced mechanical allodynia, decreased neuronal hyperexcitability and glial activation in models of chronic pain via spinal KOR and CB1 mechanisms. Local microinjection of SA into the insular cortex produced marked antinociception in a rat neuropathic pain model, a finding the investigators attribute to KOR affinity and CB1 modulation. Analogues: Because SA has limitations for therapeutic use—poor oral viability, very short duration of action, rapid metabolism to inactive salvinorin B and P-glycoprotein efflux—considerable medicinal chemistry work has produced roughly 600 semi-synthetic analogues, principally modifying the C2 acetate or the furan ring to extend duration and alter receptor profiles. Examples include herkinorin, a µ-opioid receptor (MOR)–selective agonist that produced antinociception in formalin tests with apparent minimal tolerance and which activated MOR and KOR without recruiting β-arrestin pathways; PR-38 (2-O-cinnamoylsalvinorin B) with K i 9.6 nM at KOR and K i 52 nM at MOR, which modulates intestinal motility via MOR and KOR, retains CB1 interaction and is orally bioavailable; and β-THP SalB, a C2-substituted derivative with preserved binding affinity (full characterisation not provided in the extracted text). Several analogues show efficacy in acute, inflammatory and neuropathic pain models, and some have improved pharmacokinetic or receptor-signalling properties. Side effects and toxicity: Human poison-control data over a 10-year period report neurologic, cardiovascular and gastrointestinal effects following intentional S. divinorum exposures, with worse outcomes when combined with other substances; the study cited is limited by small sample size and lack of laboratory confirmation. Rodent studies reported no major changes in cardiac conduction, temperature or galvanic skin response after acute or chronic SA administration, and no significant histological changes in spleen, blood, brain, liver, kidney or bone marrow at examined doses, suggesting relatively low in vivo toxicity in those assays. By contrast, in vitro studies across multiple cell lines (neuronal, hepatic, pulmonary, renal, intestinal) indicated dose- and time-dependent cytotoxicity, with concentrations such as 50 µM and, after prolonged exposure, 10 µM reducing cell viability in hepatic lines. The investigators note that extensive in vivo and clinical toxicology studies are still needed. Other findings: The investigators' own laboratory work reported that intraperitoneal S. divinorum ethyl acetate extract did not increase nucleus accumbens dopamine in Wistar rats, supporting the view that salvinorins may have lower addiction potential relative to classical drugs of abuse.

Coffeen and Pellicer interpret the assembled evidence as indicating that S. divinorum, salvinorin A and selected analogues possess credible anti-inflammatory and analgesic effects in preclinical models, mediated primarily via high-affinity KOR agonism and modulation of endocannabinoid receptors (CB1 and in some assays CB2). They highlight multiple mechanistic actions—reduction of iNOS, suppression of TNF-α, restoration of IL-10, inhibition of leukotriene biosynthesis and suppression of motility and nociceptive signalling in inflamed tissues—as supportive of potential utility in conditions such as inflammatory bowel disease, neuropathic pain and other LT-related inflammatory states. Positioning these findings against earlier research, the investigators note the pharmacological distinctness of SA from classical serotonergic hallucinogens because it lacks 5-HT2A activity, and they underscore the rapid pharmacokinetics that explain short-lived subjective effects. The discussion emphasises medicinal chemistry efforts to overcome limitations of SA—poor oral bioavailability, brief receptor binding and rapid metabolism to inactive salvinorin B—and presents several analogue strategies that have shown promise in prolonging activity or shifting receptor selectivity. Key limitations acknowledged include the predominance of preclinical data, variability across species and strains in behavioural responses, sparse and limited human safety data, and mixed signals from in vitro cytotoxicity studies. The extracted text also highlights pharmacokinetic barriers (enzymatic degradation, P-glycoprotein efflux) and the need to separate analgesic/anti-inflammatory efficacy from undesired hallucinogenic effects. As implications, the investigators propose that further development of non-hallucinogenic, longer-acting salvinorin analogues could provide alternative therapies for inflammatory and neuropathic pain and help address gaps in analgesic drug development; they also suggest that the apparent lack of increased nucleus accumbens dopamine in some models points to potentially lower addiction risk. Ultimately, Coffeen and Pellicer call for more extensive in vivo toxicology and clinical studies to define safety, dosing, efficacy and translational potential.