**4. Discussion**

This study revealed that chronic treatment with CBD reduced seizure-related mortality and prolonged the latency for the appearance of generalized seizures in PTZ-kindled rats. However, the development of kindling and the maximum seizure severity were not influenced. Interestingly, CBD treatment decreased the cognitive performance of rats in the NOR test. Furthermore, at the cellular level, the CA3 and CA1 regions of the hippocampal formation of CBD-treated rats showed significant differences compared to PTZ-kindled controls, which may explain the decreased mortality and cognitive performance.

CBD has been demonstrated to affect the central nervous system dose-dependently without any psychoactive action. In low doses, it showed anxiolytic and antidepressive-like effects [4,42–47] like serotonergic drugs [48,49], whereas in higher doses (30–100 mg/kg), its anticonvulsant and antipsychotic actions were proven [1,16,18,19,37]. There are, however, many discrepancies between the previously reported results, which may partly be attributed to the unfavorable pharmacokinetic properties of CBD. It is possible that highly lipophilic drugs such as CBD can precipitate in the stomach when administered orally in the form of suspensions, resulting in prolonged absorbtion, a long time to peak plasma

concentration, and low bioavailability [50]. Moreover, the vehicle of administration was shown to significantly influence the Cmax and AUC of CBD [51]. Although this issue is clearly present in laboratory experiments as well, very few studies have provided data about the pharmacokinetics of CBD formulations administered to animals [37,52].

To achieve better bioavailability, CBD was dispersed in olive oil and incorporated into the food pellets. The obtained results showed that brain concentrations correlated well with the plasma concentrations, and the brain-to-plasma ratio was consistently higher 24 h after administration, which indicates the high affinity of CBD to brain tissue. These brain concentrations are very similar to those reported recently by Uttl et al. [37]. However, plasma concentrations obtained at 1 h after administration showed that, in some cases, absorbtion was not complete. So, this study confirmed that CBD, even when dissolved in oil, which is the most common formulation of CBD used in humans, showed significant variations in the absorption phase, making the study of peak-time effects challenging, which could underlie the controversial results published previously. Hence, future studies should relate CBD's effects to its plasma concentration, rather than the dose administered.

The most important finding of this study is that CBD decreased seizure-related mortality and increased the latency to tonic-clonic seizures in the PTZ-kindling model, although it was not tested at peak plasma/brain concentrations. From a translational point-of-view, this result may provide evidence for the long-term protective action of CBD against the progression of epileptic syndromes and the potential prevention of breakthrough seizures. As previously reported, a single administration of CBD at different doses exhibits anticonvulsant effects in several animal models of epilepsy, but it does not necessarily reduce the frequency or the severity of seizures. In contrast, CBD modified the duration of generalized seizures, the latency to seizure onset, and the mortality [1,12,14,16,18,19,53]. In pilocarpine-induced seizures, CBD administered at 100 mg/kg doses did not affect the severity of seizures [14]. However, the same study reported a reduction in mortality and the occurrence of tonic-clonic seizures in the penicillin model of partial seizure [14]. The reduction of seizure duration and the increase of the latency to first seizure by CBD at a dose of 60 mg/kg were also reported after the acute intraperitoneal administration of PTZ [19].

When administered for a longer period, CBD delayed the progression of kindling in PTZ-kindled mice, but it did not prevent generalized seizures [19]. In rats, Mao et al. reported a decrease of the mean seizure score and the kindling rate in the CBD-treated group (50 mg/kg administered intraperitoneally) [18]. However, CBD was administered before PTZ injections in both studies; thus, the observed anticonvulsant action corresponds to the peak-time effect of CBD described after acute administration. In an interesting approach, Hosseinzadeh et al. demonstrated that the intracerebroventricular injection of CBD might have long-term protective effects in the pilocarpine model of epilepsy in rats [16], and the neuroprotective effects of CBD described by in vitro and in vivo studies [1,54] point to the potential benefits of chronic adjunctive treatment with CBD. On the other hand, it is important to differentiate between epileptic seizures and epilepsy syndromes. CBD is currently approved for the treatment of seizures associated with Lennox–Gastaut syndrome or Dravet syndrome based on a favorable benefit–risk profile [10], but there is little evidence about its effectiveness in other types of epileptic seizures. Therefore, the present study proposed to study the effects of CBD on seizures by delivering the seizure-precipitating factor (i.e., PTZ injection) outside of CBD's maximal anticonvulsant protection. The only significant difference between the treatment groups was the prolonged latency to the first generalized seizure; the other parameters related to kindling were not modified by CBD treatment. Interestingly, this led to a reduced mortality rate in the CBD-treated group, which is in agreemen<sup>t</sup> with previous studies [12].

The next important finding of this study is the effects of chronic CBD treatment on the cognitive performance of PTZ-kindled rats. CBD has been proposed to exert neuroprotective effects in epilepsy [55] and possesses several mechanisms to protect against memory impairments in various diseases [23]. However, a dose-dependent nature of

these neuroprotective effects was also demonstrated, the lower and middle range doses (5–20 mg/kg) showing improvements in memory assessment tasks [56]. In this experiment, the decrease of the discrimination index in the NOR test clearly showed that CBD-treated animals could not recall the familiar object and interacted with both familiar and novel objects more equally. Studies have shown that the hippocampus plays a significant role in object-recognition memory. If this structure is damaged, there will be moderate and reliable changes in anterograde memory [57]. The histological analysis of the hippocampal formation revealed that CBD-treated animals had a reduced thickness of the stratum pyramidale in the CA3 and CA1 regions. As confirmed by others, this measurement of stratum pyramidale thickness is robust and constant, because it does not differ across a series of coronal sections collected around the anteroposterior axis of the same animal [58]. Additionally, stratum pyramidale thickness was shown to reflect the neurodegeneration caused by toxic substances such as cadmium [59] and amyloidbetapeptide 1–40 [60] in rat models with cognitive impairments similar to those observed in this study.

On the other hand, CBD treatment decreased the number of GFAP-positive astrocytes to sham levels. The results of this study confirmed again that PTZ-kindling induces astrogliosis in the CA3 and CA1 regions of the hippocampus [61] and that CBD treatment prevents the proliferation of astrocytes [18]. Astrocytes were demonstrated to play an important role in epileptogenesis by releasing inflammatory cytokines and altering the excitability of the neurons [62]. The reduction in mortality in the CBD-treated group may be due to the long-term antiproliferative effects of CBD on astrocytes, which may be part of a promising novel therapeutic strategy targeting neuroinflammation [63].

Furthermore, it is important to note that the number of microglial cells in the hippocampus was not significantly influenced by CBD; however, a decreasing trend was observed. Microgliosis was also linked with epileptogenesis, especially in the status epilepticus models of epilepsy [64]. Although it seems to play a role in the increased excitability of the neurons, its involvement in cognitive impairment was not demonstrated yet. As a limitation of the study, it should be noted that only one microglia-specific histological marker was used (i.e., IBA1), which did not allow the investigation of microglial activation. Another important cellular marker of neuronal apoptosis may be neuronal debris, as described by Lana et al. [41,65–67]. This study showed that CBD-treated rats had a decreased number of neuronal debris in the stratum radiatum, despite a lower number of astrocytes, which suggests that microglial activity had an important contribution to the clearance of neuronal debris in this group. Further studies using specific microglial activation and apoptosis markers are needed to fully understand the effects of CBD on the interplay between neurons, astrocytes, and microglia.
