Research report: In vivo blood and ex vivo brain solid-phase microextraction coupled with LC-MS/MS for mapping unbound cannabidiol, phase I metabolites, and endocannabinoid tone

For the first time, researchers have tracked the pharmacokinetics of cannabidiol directly in the circulating blood of living animals, using a hair-thin extraction probe that acts like a minimally invasive chemical biopsy. A team from the Medical University of Gdańsk, Wrocław Medical University, St. Anne's University Hospital Brno and the University of Waterloo reports in Frontiers in Pharmacology that a solid-phase microextraction (SPME) platform coupled to liquid chromatography-tandem mass spectrometry (LC-MS/MS) can simultaneously monitor unbound cannabidiol (CBD), its phase I metabolites, and the endocannabinoid 2-arachidonoylglycerol (2-AG) in living rat blood and in four discrete regions of the intact brain. The proof-of-concept study reveals a striking mismatch between where the drug's total mass accumulates and where its freely available, pharmacologically relevant fraction actually resides.

The motivation lies in a long-standing problem of neuropharmacology: conventional analysis grinds brain tissue into a homogenate and extracts everything, which conflates the inert, protein-bound reservoir of a lipophilic drug with the small unbound fraction that can diffuse to and modulate its molecular targets. According to the free drug hypothesis, only this unbound fraction drives pharmacological effect. For CBD, a highly lipophilic molecule that binds plasma lipoproteins and tissue proteins to an extent exceeding 95 percent, total tissue concentrations measured by solid-liquid extraction (SLE) therefore say little about active exposure.

SPME sidesteps this by using a biocompatible probe coated with a C18 polymeric phase that adsorbs small molecules directly from the biological matrix without removing fluid or tissue, preserving the pre-existing binding equilibrium and immediately immobilizing analytes before local enzymes or post-mortem degradation can alter them. In the experiment, male Wistar Han rats received a single oral dose of CBD at 100 mg/kg body weight, delivered voluntarily as a medicinal-cannabis jelly prepared from decarboxylated, GMP-certified CBD-dominant plant material containing 7.6 percent CBD. At baseline and at 1, 2, 4 and 6 hours after dosing, anesthetized animals had a C18-coated SPME probe inserted into the lateral tail vein for a 10-minute extraction window operating under pre-equilibrium conditions.

Immediately after blood sampling, the animals were euthanized and 5 mm probes were placed into the hippocampus, amygdala and cerebellum, with 10 mm probes in the brainstem, for 30-minute static ex vivo extraction. The same brain regions then underwent classical SLE to quantify total tissue concentrations, allowing a direct free-versus-total comparison. Quantification relied on a validated LC-MS/MS method with calibration in surrogate matrices—agarose gel mimicking the semi-solid brain and artificial serum—yielding correlation coefficients above 0.97 and detection limits of 1 to 2 ng/mL or ng/g for most analytes.

The in vivo blood data captured a rapid oral absorption phase: free circulating CBD rose steeply within the first hour and reached a maximum concentration of roughly 160 μg/mL at 2 hours post-dose, followed by a marked decline at 4 and 6 hours. The authors attribute this sharp clearance to intense hepatic first-pass metabolism by cytochrome P450 enzymes and rapid distribution into highly perfused, lipid-rich tissues including the brain. Among the phase I metabolites, only 6-hydroxy-CBD was recovered from circulating blood, and only at trace levels at the 2-hour interval.

The other major biotransformation products, 7-hydroxy-CBD and 7-carboxy-CBD, remained entirely below detection limits in their unbound forms throughout the 6-hour time course, suggesting that systemic generation of these metabolites is strictly rate-limited by parent drug availability, or that they distribute rapidly into organs. The same 10-minute blood protocol also tracked endocannabinoid tone in living animals. Mean circulating 2-AG concentrations were numerically higher than the vehicle-control baseline of 34.45 ng/mL at every post-dose time point, rising by approximately 178, 172, 201 and 424 percent at 1, 2, 4 and 6 hours respectively, with the highest mean value at the terminal 6-hour mark.

Because CBD does not directly activate CB1 or CB2 receptors but is thought to modulate endocannabinoid signaling indirectly, including through negative allosteric modulation of CB1, this delayed systemic pattern is consistent with an indirect upregulation of the endocannabinoid system. The authors stress, however, that these preliminary findings rest on a very small exploratory sample of two to three rats per time point and must be interpreted as proof of concept rather than definitive pharmacology. Ex vivo brain microextraction uncovered pronounced regional heterogeneity in the freely accessible CBD fraction.

Parent CBD was detected in the amygdala, cerebellum and hippocampus at every time point from 1 to 6 hours, confirming passage across the blood-brain barrier. The amygdala and cerebellum both peaked at 2 hours, reaching 140.83 ± 11.87 ng/g and 157.80 ± 10.68 ng/g respectively, before declining by 44 and 59 percent at 6 hours. The hippocampus told a different story: its peak arrived late, at 4 hours, at 411.03 ± 90.66 ng/g—about 2.9-fold the amygdala peak and 2.6-fold the cerebellar peak—before falling 89 percent by 6 hours.

Kruskal-Wallis tests found significant temporal effects in the cerebellum (p = 0.0018) and hippocampus (p = 0.0014), with Dunn's post hoc tests identifying significant differences between the 2- and 6-hour points in the cerebellum and the 4- and 6-hour points in the hippocampus, while the amygdala showed no statistically significant overall change. Descriptive pharmacokinetic parameters derived from the composite mean profiles reinforced this regional picture. The hippocampal area under the concentration-time curve over 6 hours was 1,280.95 ng/g·h, roughly 2.1-fold that of the amygdala (601.70 ng/g·h) and 2.2-fold that of the cerebellum (595.09 ng/g·h).

The brain-to-serum AUC ratio was likewise numerically highest for the hippocampus at 0.00492 mL/g, versus nearly identical values of 0.00231 and 0.00228 mL/g for the amygdala and cerebellum. Because brain tissue collection was terminal, individual concentration-time profiles could not be obtained from the same animals, and these parameters were explicitly treated as exploratory and descriptive rather than grounds for formal statistical comparison between regions. The endocannabinoid data from brain mirrored this region-dependence.

Only 2-AG was detectable by SPME in all sampled structures. In the amygdala, mean 2-AG concentrations sat at or below control values across the entire time course without statistical significance. The cerebellum showed an early numerical dip—about 36 percent below control at 2 hours—followed by recovery, but the overall test did not reach significance (p = 0.0794).

The hippocampus, by contrast, displayed numerically higher mean 2-AG concentrations at 4 and 6 hours, roughly 88 and 123 percent above control, with the overall Kruskal-Wallis test significant at p = 0.0227; notably, however, Dunn's post hoc comparisons identified no significant pairwise differences between individual time points, so the authors caution that this pattern should be read as descriptive rather than as a confirmed CBD-induced increase. Intriguingly, the hippocampal 2-AG rise began as CBD peaked at 4 hours and continued as CBD declined, hinting at a delayed endocannabinoid system response, though the study did not measure the enzymes or receptors that would be needed to establish any mechanism. The starkest result emerged from the cross-correlation of free and total measurements.

SLE of the same tissues detected abundant phase I metabolites in every brain region, with total 7-OH-CBD accumulating at levels 3 to 8 times higher than parent CBD and peaking at 4 hours, while 6-OH-CBD persisted detectably in all four structures at 6 hours. Yet SPME recovered essentially no free metabolites from any brain region—the sole exception being 7-OH-CBD in the hippocampus and amygdala of one rat at 4 hours, an animal that showed extreme CBD and metabolite levels across all tissues, underscoring inter-individual variability in metabolism and blood-brain barrier penetration. Free CBD captured by SPME accounted for only 2 to 40 percent of the total tissue burden measured by SLE.

The authors note that this discrepancy cannot be attributed to tissue sequestration alone: it may equally reflect free concentrations falling below the SPME-LC-MS/MS detection limit of 2.0 ng/g, or region-specific differences in diffusion and extraction kinetics that were not characterized for individual structures. The 30-minute static extraction was optimized in agarose gel, but tissue-specific diffusion coefficients and tortuosity remain undetermined, and future in vivo brain applications will require thinner coatings, shorter extractions or diffusion-based calibration to guarantee non-depleting conditions. As a methodology-driven proof of concept with small cohorts and no statistical confirmation of region-specific endocannabinoid modulation, the study's value lies chiefly in its analytical framework rather than in clinical implications.

It demonstrates that coupling living intravascular extraction with localized regional brain microextraction can resolve the active biophase kinetics of lipophilic neuropharmaceuticals—something bulk homogenization fundamentally cannot—and provides a more accurate bioanalytical baseline for evaluating novel endocannabinoid system modulators and other lipophilic central nervous system drug candidates in future, larger preclinical studies.

Subject of Research: Mapping unbound cannabidiol and endocannabinoid dynamics in rat blood and brain using solid-phase microextraction coupled with LC-MS/MS

Article Title: In vivo blood and ex vivo brain solid-phase microextraction coupled with LC-MS/MS for mapping unbound cannabidiol, phase I metabolites, and endocannabinoid tone

Article References: Mosińska, A., Kaszewska, M., Sztormowska-Achranowicz, K., Smarzewska, K., Owczarek, K., Woźniczka, K., Trojan, V., Miś, M., Pawliszyn, J., Bączek, T., & Roszkowska, A. (2026). In vivo blood and ex vivo brain solid-phase microextraction coupled with LC-MS/MS for mapping unbound cannabidiol, phase I metabolites, and endocannabinoid tone. Frontiers in Pharmacology, 17, Article 1980760. https://doi.org/10.3389/fphar.2026.1980760

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1980760

Keywords: cannabidiol, solid-phase microextraction, LC-MS/MS, endocannabinoid system, 2-arachidonoylglycerol, neuropharmacokinetics, blood-brain barrier, free drug hypothesis, hippocampus, phase I metabolites, pharmacokinetics, rats