Physicochemical Properties, Solubility, and Protein Binding of Δ⁹ -Tetrahydrocannabinol

Abstract

The rate and extent of glass binding of Δ⁹ -tetrahydrocannabinol in aqueous solution depend on the surface area and pretreatment of glass and the concentration of the drug. A total of 20 and 40% at 0.1 and 0.05 μg/ml, respectively, was bound to 50-ml volumetric flasks but could be minimized by silyl pretreatment of the glass. The drug rapidly diffused into plastics, and 70–97% was taken up by the rubber closures used for plasma vials. These bindings precluded classical methods of solubility determination, so spectral and particle-size counting determinations, which observed those concentrations at which true solution was terminated, were used. The aqueous solubility was a linear function of both the ethanol concentration (increasing) at constant ionic strength and the square root of the ionic strength (decreasing) at constant ethanol concentration. The salting-out coefficient was of high magnitude and typical solubilities were 2.8 mg/liter in water and 0.77 mg/liter in 0.15 M NaCl at 23°. The bindings also precluded the use of the classical methods of equilibrium dialysis and ultrafiltration to determine the protein binding of tetrahydrocannabinol. A method of variable plasma concentrations was devised, so protein binding was determined from the pseudoplasma concentrations of the drug after the separation of the pseudoplasma from the red blood cells added to form pseudoblood with known concentrations of Δ⁹ -tetrahydrocannabinol. This use of the competition between the high partitioning of drug between red blood cells with plasma water (D = 12.5) and the binding to plasma protein permitted an estimate of 97% binding which was not drug concentration dependent. The spectrophotometric pKa' of Δ⁹ -tetrahydrocannabinol was 10.6. Δ⁹ -Tetrahydrocannabinol degraded readily in acid solutions. Subsequent to a rapid loss, the kinetics appeared to be first-order and specific hydrogen-ion catalyzed. Concomitantly, small amounts of Δ⁸ -tetrahydrocannabinol were produced, as were two GLC observable products, P₂ and P₃, and the rate of their appearance appeared to parallel the rate of Δ⁹ -tetrahydrocannabinol degradation. A peak, P₁, also appeared almost instantaneously but did not parallel drug degradation.