Prenatal exposure to cannabinoids evokes long-lasting functional alterations by targeting CB1 receptors on developing cortical neurons

Abstract

The CB1 cannabinoid receptor, the main target of Δ(9)-tetrahydrocannabinol (THC), the most prominent psychoactive compound of marijuana, plays a crucial regulatory role in brain development as evidenced by the neurodevelopmental consequences of its manipulation in animal models. Likewise, recreational cannabis use during pregnancy affects brain structure and function of the progeny. However, the precise neurobiological substrates underlying the consequences of prenatal THC exposure remain unknown. As CB1 signaling is known to modulate long-range corticofugal connectivity, we analyzed the impact of THC exposure on cortical projection neuron development. THC administration to pregnant mice in a restricted time window interfered with subcerebral projection neuron generation, thereby altering corticospinal connectivity, and produced long-lasting alterations in the fine motor performance of the adult offspring. Consequences of THC exposure were reminiscent of those elicited by CB1 receptor genetic ablation, and CB1-null mice were resistant to THC-induced alterations. The identity of embryonic THC neuronal targets was determined by a Cre-mediated, lineage-specific, CB1 expression-rescue strategy in a CB1-null background. Early and selective CB1 reexpression in dorsal telencephalic glutamatergic neurons but not forebrain GABAergic neurons rescued the deficits in corticospinal motor neuron development of CB1-null mice and restored susceptibility to THC-induced motor alterations. In addition, THC administration induced an increase in seizure susceptibility that was mediated by its interference with CB1-dependent regulation of both glutamatergic and GABAergic neuron development. These findings demonstrate that prenatal exposure to THC has long-lasting deleterious consequences in the adult offspring solely mediated by its ability to disrupt the neurodevelopmental role of CB1 signaling.

Keywords: CB1 cannabinoid receptor; cannabis; corticospinal; neurodevelopment; seizures.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.

Embryonic THC exposure impairs subcerebral projection neuron development. (A) Subcerebral projection neurons in embryonically vehicle- and THC-administered mice at P15 were stained with an anti-ER81 antibody (n = 7 and 5, respectively). (B) ER81+ cell number was quantified and referred to the total cell number (DAPI) per cortical column. (C) CSMNs were labeled by injecting retrogradelly transported red fluorescent beads (RetroBeads) in the cervical spinal cord at P10. (D) Representative image showing RetroBead colocalization with the subcerebral projection neuron marker ER81. (E) Representative images of retrogradelly labeled somata in cortical layer V at P15. (F) RetroBead-labeled somata per cortical column were quantified in vehicle- and THC-exposed mice (n = 5 and 4, respectively). *P < 0.05, **P < 0.01 vs. vehicle-treated mice. [Scale bar, (A and E) 200 µm (Insets, 60 µm) and (D) 10 µm.]

Fig. 2.

Embryonic THC exposure impairs corticospinal motor function. (A and B) The skilled pellet-reaching test was evaluated in adult CB1+/− and CB1−/− littermates prenatally exposed to vehicle or THC from E12.5 to E16.5. The success ratio of pellets retrieved in the skilled paradigm (A) and the total number of trials performed (B) were quantified [n = 11 and 8 (CB1+/− vehicle- and THC-treated mice, respectively); n = 6 and 4 (CB1−/− vehicle- and THC-treated mice, respectively)]. (C and D) Mice were subjected to the staircase pellet-reaching test, and the sum of pellets retrieved from skilled steps (from four to eight) was compared between vehicle- and THC-administered mice (C). As a control, the number of pellets reached in unskilled steps 1–3 was calculated (D) [n = 8 and 12 (CB1+/− vehicle- and THC-treated mice, respectively); n = 5 and 3 (CB1−/− vehicle- and THC-treated mice, respectively)]. (E and F) THC- and vehicle-treated CB1+/− and CB1−/− mice were subjected to a PTZ administration paradigm. Latency to tonic-clonic seizure appearance after the first PTZ injection (E) and the cumulative dose of PTZ required (F) were calculated [n = 15 and 18 (CB1+/− vehicle- and THC-treated mice, respectively); n = 13 and 11 (CB1−/− vehicle- and THC-treated mice, respectively)]. *P < 0.05, **P < 0.01 vs. vehicle-treated CB1+/− mice.

Fig. 3.

Embryonic THC exposure transiently down-regulates CB1Rs. (A–D) CB1R protein levels were determined by Western blot in brain samples at E17.5 (A and C) and P2.5 (B and D) after THC or vehicle administration from E12.5 to E16.5. The optical density of the CB1R band was quantified and normalized to β-actin [n = 9 and 6 (E17.5 vehicle- and THC-treated brain samples, respectively); n = 7 and 7 (P2.5 vehicle- and THC-treated brain samples, respectively)]. (E–H) Radiolabeled CP-55,940 binding was quantified in coronal brain sections at E17.5 (E and G) and P2.5 (F and H) [n = 4 and 6 (E17.5 vehicle- and THC-treated brains, respectively); n = 8 and 8 (P2.5 vehicle- and THC-treated brains, respectively)]. **P < 0.01 vs. vehicle-treated mice. (Scale bar, 2 mm.)

Fig. 4.

Characterization of selective CB1R expression rescue in dorsal glutamatergic neurons and forebrain GABAergic neurons. (A) Immunofluorescence analysis of CB1R reexpression driven by Nex (Glu-CB1-RS) (A3), Dlx5/6 (GABA-CB1-RS) (A4), or EIIa (CB1-RS) (A2) promoters in Stop-CB1 mice (A1) was performed in P2.5 coronal brain sections. (B) Cortical CB1R expression analysis in the same mouse strains was performed at P20 (B1–B4). (C) Double immunofluorescence was performed with anti-CB1 antibody combined with anti-VGLUT1 (C1–C4) or anti-VGAT (C1’–C4’) antibodies. Arrowheads point to the glutamatergic CB1R signal, which is abundant in pyramidal neuron fibers at P2.5 (A) and to scarce immunopositive puncta corresponding to glutamatergic (VGLUT1-positive) terminals at P20 (B and C). Arrows point to the abundant CB1R signal corresponding to GABAergic (VGAT-positive) neurons within the hippocampal formation at P2.5 (A) and the mature cortex at P20 (B and C). [Scale bars, (A) 200 µm, (B) 50 µm (Inset, 10 µm), and (C) 5 µm.]

Fig. 5.

Selective CB1R expression rescue restores functional alterations and reestablishes THC susceptibility in Stop-CB1 mice. (A and B) Skilled motor activity was assessed by the skilled pellet-reaching (A) and staircase (B) tests in adult CB1R-rescued mice [n = 15 and 15 (Stop-CB1 vehicle- and THC-treated mice, respectively); n = 20 and 17 (CB1-RS vehicle- and THC-treated mice, respectively); n = 9 and 6 (Glu-CB1-RS vehicle- and THC-treated mice, respectively); n = 5 and 7 (GABA-CB1-RS vehicle- and THC-treated mice, respectively)]. (C and D) Seizure susceptibility to subconvulsive doses of PTZ was determined. Latency to seizures (C) and the cumulative dose of PTZ required (D) are shown [n = 13 and 19 (Stop-CB1 vehicle- and THC-treated mice, respectively); n = 18 and 19 (CB1-RS vehicle- and THC-treated mice, respectively); n = 9 and 5 (Glu-CB1-RS vehicle- and THC-treated mice, respectively); n = 9 and 8 (GABA-CB1-RS vehicle- and THC-treated mice, respectively)]. *P < 0.05, **P < 0.01 vs. corresponding Stop-CB1 mice; #P < 0.05, ##P < 0.01 vs. corresponding vehicle-treated group.