Mechanisms of Transplacental Transfer of (‐)‐∆ 9‐Tetrahydrocannabinol (THC) in the Human Dual Cotyledon, Dual Perfusion, Placenta Model

Abstract

Background(‐)‐∆ 9‐Tetrahydrocannabinol (THC) is the most abundant psychoactive compound in marijuana. With legalization of marijuana, its use has increased including by pregnant women. The latter raises concerns of fetal toxicity which is likely driven by fetal THC exposure. Interestingly, in the catheterized maternal‐fetal macaque model, fetal THC exposure is lower than its corresponding maternal exposure (fetal/maternal AUC0‐inf ratio of 0.37). This difference is supported by sparse human THC umbilical vein/maternal plasma concentrations at term. The most likely explanation for these observations is that THC is effluxed by P‐glycoprotein (P‐gp) which is highly expressed in the human placenta. THC oral pharmacokinetics studies in P‐gp knock‐out mice support the hypothesis that P‐gp can transport THC. Therefore, the objective of our study was to investigate if THC is effluxed in human placentae by estimating the maternal to fetal clearance (m‐f‐CL) and fetal to maternal clearance (f‐m‐CL) using the dual cotyledon, dual perfusion, placenta model.MethodsTerm placentae were utilized from uncomplicated, unlabored, cesarean deliveries with no known history of tobacco or drug use. Two cotyledons from each placenta were perfused at physiological rates (m: 10 mL/min; f: 4 mL/min) for 2‐4 hours in a single‐pass mode with oxygenated HBSS buffer containing 0.2% BSA, 2000 U/L sodium heparin and 5 mg/mL gentamicin. To determine m‐f CL, the maternal perfusate also contained 5 μM THC, a P‐gp transporter probe (1 or 10 μM saquinavir), and a passive diffusion probe (20 μg/mL antipyrine) and effluent from the fetal vein and intervillous space were collected at multiple timepoints. To determine f‐m‐CL, the fetal perfusate of the second cotyledon contained the aforementioned drugs. All perfusions were conducted with (n = 7) and without (n = 5) a pan efflux transporter inhibitor, 4 μM valspodar. THC concentrations in the effluent were quantified using LC‐MS/MS. THC CL indices, i.e. m‐f‐CL and f‐m‐CL, normalized to the corresponding antipyrine CL, were compared in the absence and presence of valspodar.ResultsTHC m‐f‐CL index, 0.40 ± 0.15, was not significantly different from the f‐m‐CL index, 0.54 ± 0.13 (p = 0.31; two‐tailed Wilcoxon signed rank test), suggesting that THC passively diffuses across the placental barrier. Surprisingly, in the presence of valspodar, THC m‐f‐CL index, 0.24 ± 0.14, was significantly lower than the corresponding f‐m‐CL index, 0.44 ± 0.14 (p = 0.016; two‐tailed Wilcoxon signed rank test). The reason for this difference is unknown.ConclusionsBecause THC passively crosses the human placenta, changes in activity of placental transporters (e.g. drugs that inhibit/induce placental transporters or pharmacogenetics) should not alter fetal THC exposure. Further, mechanisms other than P‐gp efflux (e.g. binding to placental tissue) need to be explored to explain the fetal and maternal THC exposure of less than unity detailed in the introduction.