A combination of Δ9-tetrahydrocannabinol and cannabidiol modulates glutamate dynamics in the hippocampus of an animal model of Alzheimer's disease

Abstract

A combination of Δ⁹-tetrahydrocannabinol (Δ⁹-THC) and cannabidiol (CBD) at non-psychoactive doses was previously demonstrated to reduce cognitive decline in APP/PS1 mice, an animal model of Alzheimer's disease (AD). However, the neurobiological substrates underlying these therapeutic properties of Δ⁹-THC and CBD are not fully understood. Considering that dysregulation of glutamatergic activity contributes to cognitive impairment in AD, the present study evaluates the hypothesis that the combination of these two natural cannabinoids might reverse the alterations in glutamate dynamics within the hippocampus of this animal model of AD. Interestingly, our findings reveal that chronic treatment with Δ⁹-THC and CBD, but not with any of them alone, reduces extracellular glutamate levels and the basal excitability of the hippocampus in APP/PS1 mice. These effects are not related to significant changes in the function and structure of glutamate synapses, as no relevant changes in synaptic plasticity, glutamate signaling or in the levels of key components of these synapses were observed in cannabinoid-treated mice. Our data instead indicate that these cannabinoid effects are associated with the control of glutamate uptake and/or to the regulation of the hippocampal network. Taken together, these results support the potential therapeutic properties of combining these natural cannabinoids against the excitotoxicity that occurs in AD brains.

Keywords: Alzheimer; Cannabidiol; Cannabinoid; Glutamate; Hippocampus; Δ(9)-tetrahydrocannabinol.

Fig. 1

Extracellular glutamate levels in the hippocampus of chronically treated WT and APP/PS1 mice measured by in vivo microdialysis. (A) Chronic treatment with Δ⁹-THC ​+ ​CBD (orange symbols) significantly decreased glutamate extracellular levels when compared to vehicle (VEH)-treated animals (white symbols) after the 20 ​min infusion with the depolarizing agent veratridine (50 ​μM) on fractions 6 and 7. (B) Area under the curve (AUC) values for the glutamate extracellular levels after veratridine infusion from fraction 5 to 8 (colored area in A) revealed a similar treatment effect. (C) Infusion during 2 ​h with the GLT-1 inhibitor DHK (5 ​mM) increased hippocampal glutamate extracellular levels, an increase significantly lower in mice chronically treated with Δ⁹-THC ​+ ​CBD (orange symbols) than in VEH-treated (white symbols) mice at fractions 6, 7 and 10. A genotype effect was also significant in fractions 9 and 10. (D) AUC values for the glutamate extracellular levels after DHK infusion from fraction 5 to 10 (colored area in C), revealing a similar treatment effect. Data are expressed as mean ​± ​SEM (WT VEH n ​= ​7 males ​+ ​4 females; WT Δ⁹-THC ​+ ​CBD n ​= ​6 males ​+ ​8 females; APP/PS1 VEH n ​= ​4 males ​+ ​5 females; APP/PS1 Δ⁹-THC ​+ ​CBD n ​= ​2 males ​+ ​5 females). T∗: Treatment effect, p ​< ​0.05; T∗∗: Treatment effect, p ​< ​0.01; G∗: Genotype effect, p ​< ​0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

(A) Representative immunoblots for EAAT3, GLT-1, GluA1, GluN1, mGlu5R and β-tubulin as loading control. Quantification by western blotting of the main glutamate transporters EAAT3 (B), GLT-1 (C) and glutamate receptors subunits GluA1 (D), GluN1 (E) and mGlu5R (F) in hippocampal samples from APP/PS1 (black bars) and WT (white bars) mice chronically treated with Δ⁹-THC ​+ ​CBD or vehicle (VEH). Densitometric quantifications of each protein are normalized respect β-tubulin and referred to vehicle-treated WT mice. (F) Individual values and mean ​± ​SEM (WT VEH n ​= ​2 males ​+ ​3 females; WT Δ⁹-THC ​+ ​CBD n ​= ​4 males ​+ ​2 females; APP/PS1 VEH n ​= ​3 males ​+ ​3 females; APP/PS1 Δ⁹-THC ​+ ​CBD n ​= ​2 males ​+ ​4 females) are presented in the bar graphs. T∗: Treatment effect, p ​< ​0.05.

Fig. 3

(A) Representative immunoblots for peEF2, eEF2, pFyn, Fyn and β-tubulin as loading control. Quantification by western blotting of the phosphorylated and total forms of the glutamate signaling mediators eEF2 (B–C) and Fyn (E–F) in hippocampal samples from APP/PS1 (black bars) and WT (white bars) mice chronically treated with Δ⁹-THC ​+ ​CBD or vehicle (VEH). Densitometric quantifications of each protein are normalized respect β-tubulin and referred to vehicle-treated WT mice. (D) and (G) show the ratio between the phosphorylated and the total form of each signaling mediator. Individual values and mean ​± ​SEM (WT VEH n ​= ​3 males ​+ ​1 females; WT Δ⁹-THC ​+ ​CBD n ​= ​5 males ​+ ​5 females; APP/PS1 VEH n ​= ​2 males ​+ ​2 females; APP/PS1 Δ⁹-THC ​+ ​CBD n ​= ​2 males ​+ ​5 females) are presented in the bar graphs. T∗: Treatment effect, p ​< ​0.05; G∗: Genotype effect, p ​< ​0.05; G∗∗: Genotype effect, p ​< ​0.01.

Fig. 4

(A) Input–output curves obtained in hippocampal slices of vehicle- (VEH) and Δ⁹-THC ​+ ​CBD-treated WT and APP/PS1 mice are displayed as the relationship between fEPSP slope (ordinates) and stimulus intensity (abscissa). Δ⁹-THC ​+ ​CBD chronic treatment induced a significant reduction in the maximum fEPSP slope in both genotypes. (B) Paired pulse ratio (PPR) showed a paired-pulse facilitation (PPR >1) with an interpulse interval of 50 ​ms with no differences between groups. (C) Time-course of the variation of the slope of fEPSPs, expressed as percentage of baseline values, in the CA1 stratum radiatum upon stimulation of afferent Schaffer collateral fibers, before and after induction of a long-term potentiation (LTP) with a train of high-frequency stimulation (HFS, 5 trains of 100 ​Hz for 1 ​s, arrow). No significant differences due to genotype or chronic treatment were observed. The inserts show representative recordings of the fEPSPs obtained for the indicated experimental groups. (D) Bar graph of LTP amplitude at 50–60 ​min after HFS. No significant differences due to genotype or chronic treatment were observed. Individual values and mean ​± ​SEM (WT VEH n ​= ​1 male ​+ ​5 females; WT Δ⁹-THC ​+ ​CBD n ​= ​5 females; APP/PS1 VEH n ​= ​3 males ​+ ​2 females; APP/PS1 Δ⁹-THC ​+ ​CBD n ​= ​2 males ​+ ​5 females) are presented in the bar graphs. ∗p ​< ​0.05, ∗∗p ​< ​0.01 Tukey's post hoc test respect WT vehicle (VEH) group.

Fig. 5

(A) Representative dendrites of CA1 pyramidal neurons from vehicle- and Δ⁹-THC ​+ ​CBD-treated male WT and APP/PS1 mice. Scale bars represent 5 ​μm. (B) Quantitative analysis showing dendritic spine density per micrometer of dendritic length in CA1 pyramidal neurons from male and female WT and APP/PS1 mice at 6 months of age. A significant reduction in the dendritic spine density in female WT and male APP/PS1 mice was observed when compared to male WT littermates. (C) Density of each morphological type of dendritic spine in male treated mice. A significant decrease in the density of mushroom spines was observed in Δ⁹-THC ​+ ​CBD-treated APP/PS1 respect treated WT mice. (D) No significant differences due to genotype or treatment were observed in the total dendritic spine density in female treated mice, in spite of a significant treatment effect in the density of stubby spines. Data are expressed as mean ​± ​SEM (n ​= ​45–55 dendrites from WT VEH n ​= ​4 males ​+ ​4 females; WT Δ⁹-THC ​+ ​CBD n ​= ​4 males ​+ ​4 females; APP/PS1 VEH n ​= ​4 males ​+ ​4 females; APP/PS1 Δ⁹-THC ​+ ​CBD n ​= ​3 males ​+ ​3 females). ∗p ​< ​0.05, ∗∗p ​< ​0.01 respect control groups.

Fig. 6

(A) Diagram showing the ventral hippocampal area where glutamate biosensor (iGluSnFr) was injected and the optic fiber cannula attached to a bipolar electrode were placed for photometry recordings (upper panel). Lower panel shows a representative slide expressing iGluSnFr sensor. (B) Representative traces of iGluSnFR signal after electrical stimulation of the hippocampus in WT (upper panel) and APP/PS1 (lower panel) mice chronically treated with vehicle (VEH, black lines) or Δ⁹-THC ​+ ​CBD (orange lines). (C) Chronic treatment with Δ⁹-THC ​+ ​CBD did not affect dF/F₀ measurements for the elicited peaks at any frequency, but reversed the increased decay time observed in APP/PS1 treated with VEH (D). Data are expressed as mean ± SEM (WT VEH n = 4 male + 3 females; WT Δ⁹-THC+CBD n = 5 male + 3 females; APP/PS1 VEH n = 2 males + 3 females; APP/PS1 Δ⁹-THC+CBD n = 5 males + 4 females). ∗p ​< ​0.05 respect APP/PS1 VEH group. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)