Cannabidiol Modulates the Effects of Levetiracetam on Seizure Parameters and Behavioral Outcomes in Pentylenetetrazol-Kindled Rats

Abstract

Background/Objectives: The antiseizure effects of cannabidiol (CBD) were extensively studied when used as a monotherapy. However, there is conflicting evidence regarding its use in combination with levetiracetam (LEV). Methods: This study explored the effects of chronic co-administration of CBD and LEV in a pentylenetetrazole-kindling rat model to evaluate potential antiseizure and neuropsychiatric interactions. Male and female Wistar rats (n = 48) were divided into four treatment groups: one control and three treated by receiving LEV 300 mg/kg and LEV + CBD at 10 and 60 mg/kg, respectively. Seizure parameters were assessed using the Racine scale, and behavior was evaluated using the open field (OF), novel object recognition (NOR), and social interaction (SI) tests. Results: While both combinations, LEV + CBD 10 mg/kg and 60 mg/kg, significantly reduced maximal seizure intensity, the LEV + CBD 10 mg/kg attenuated LEV’s anti-kindling effect. Additionally, only LEV + CBD 60 mg/kg reduced seizure duration compared to LEV alone (p = 0.0002). In behavioral assessments, LEV + CBD 10 mg/kg showed anxiolytic effects in the OF test by increasing central activity (p = 0.0141). In contrast, the LEV + CBD 60 mg/kg impaired social behavior in both sexes (p = 0.0019). LEV improved the cognitive performance of female rats in the NOR test (p = 0.0301), but this improvement was not observed in LEV + CBD groups. Conclusions: CBD exhibited dose-dependent effects when combined with LEV: low doses might offer anxiolytic effects but promote kindling, and high doses enhance seizure control but potentially worsen social interaction. The results support the therapeutic potential of LEV-CBD co-treatment, while highlighting the need for careful dose optimization when considering CBD as an adjunctive therapy.

1. Introduction

Epilepsy, a disorder of the central nervous system, affects approximately 0.5–1% of the population worldwide and significantly impacts quality of life [1]. Available antiseizure medications (ASM) are effective in approximately 60% of patients with epilepsy; however, most of these medications have significant side effects or adverse reactions, including hypersensitivity and interactions [2]. Patients with epilepsy frequently present with associated neuropsychiatric disorders, such as anxiety, depression, or suicidal tendencies [2,3,4,5]. Additionally, epileptic patients often report other somatic and cardiovascular diseases that require medication alongside antiepileptic treatment [6]. It is important to mention here that some ASMs have an enzyme-inducing effect, so treatment is a challenge due to drug–drug interactions [7,8].

Owing to its well-established efficacy and tolerability, levetiracetam (LEV) has progressively become a first-line treatment option for epilepsy over the years [9]. LEV’s distinct mechanism of action involves a high affinity for the SV2A vesicular protein, which regulates neurotransmitter release, thereby blocking excessive glutamate release [9]. The pharmacokinetic profile of LEV is linear and nearly ideal [10]. It is rapidly and almost completely absorbed after oral administration [11]. The metabolism of LEV does not affect the cytochrome P450 system, resulting in minimal risk of drug interactions [12]. However, Joyce A. and colleagues noted that the use of LEV may cause psychiatric symptoms, with the more significant ones including depression, irritability, and aggression [13]. Given that LEV’s neuropsychiatric side effects pose significant challenges, the complementary anxiolytic and antidepressant properties of cannabidiol (CBD) present a compelling rationale for exploring their combined therapeutic potential.

CBD is a non-psychoactive component of Cannabis sativa, exhibiting highly complex mechanisms of action [14]. Its neuroprotective potential is attributed to pronounced antioxidant and anti-inflammatory effects, which have made it a focal point of scientific research. CBD demonstrated therapeutic promise in the management of various neurodegenerative disorders, including amyotrophic lateral sclerosis, Parkinson’s disease, Huntington’s disease, and Alzheimer’s disease [15]. Additionally, it is approved as an adjunctive therapy for severe epileptic syndromes, such as Lennox–Gastaut and Dravet [16].

Several studies support the antiepileptic effects of CBD when used as monotherapy. Devinsky and colleagues demonstrated that CBD reduces seizure frequency and has an adequate safety profile in children and young adults with highly treatment-resistant epilepsy [17,18,19]. CBD has also shown protective effects against pentylenetetrazol (PTZ)-induced chronic seizures [20,21]. Given CBD’s antiepileptic, antidepressant, and anxiolytic effects, it may be a promising option for adjunctive use in combination with other ASMs, especially with those having neuropsychiatric side effects [13,22]. Although LEV is not metabolized by CYP450 enzymes, a recent study demonstrated that acute administration of CBD at doses of 50–100 mg/kg body weight reduces LEV’s antiepileptic efficacy in mice, possibly through pharmacodynamic interactions that do not affect drug concentrations [23].

The main objective of the current study was to evaluate the therapeutic effects and possible interactions of chronic LEV-CBD co-administration in the PTZ-kindling rat model of epilepsy using behavioral tests. A secondary objective was to compare the outcomes between male and female rats to identify any sex-related differences in response to the treatment.

2. Materials and Methods

2.1. Experimental Animals

Adult female (n = 24, mean 295 g, SD 19 g) and male (n = 24, mean 504 g, SD 27 g) Wistar rats with approximately uniform body weights were used in groups. The rats were provided by the Biobase of the George Emil Palade University of Medicine, Pharmacy, Science, and Technology of Targu Mures. Before the experiments, all animals were subjected to an 8-day habituation period and were housed individually in plexiglass cages to monitor seizures and drug pellet consumption accurately. All animals were subjected to identical handling protocols, and the experimental conditions were standardized across all groups. Standard environmental conditions (12 h light–dark cycle, 20 ± 2 °C temperature, and 60 ± 10% humidity) were ensured by heating, ventilation, and air conditioning systems. The light–dark cycle was provided by a programmable light source. Throughout the experiment, water was available ad libitum. However, to encourage the consumption of CBD- and solvent-containing pellets, the required amount of food was provided daily. Body weight was recorded once weekly, and their health status and well-being were monitored daily. To minimize potential confounders such as the order of treatments and measurements, all interventions were sequentially performed from the first to the last cage, ensuring that animals from different groups were tested each day. The applied procedures were in accordance with the European Directive 2010/63/EU and approved by the Ethics Committee for Scientific Research of the George Emil Palade University of Medicine, Pharmacy, Science, and Technology of Targu Mures (approval no. 1950/07.12.2022) and by the Directorate of Veterinary Sanitary and Food Safety (DSVSA Mures) (approval no. 57/06.03.2023).

Following the habituation period, animals were randomly divided by sex into four equal groups (n = 6 per group for both females and males): control groups (FC for females, MC for males) received no treatment but underwent PTZ-kindling; LEV-treated groups (FL for females, ML for males) underwent PTZ-kindling and received levetiracetam (LEV) at 300 mg/kg body weight; LEV + 10 mg/kg CBD groups (FLC10 for females, MLC10 for males) underwent PTZ-kindling and received LEV (300 mg/kg) plus CBD (10 mg/kg); LEV + 60 mg/kg CBD groups (FLC60 for females, MLC60 for males) underwent PTZ-kindling and received LEV (300 mg/kg) plus CBD (60 mg/kg). Male and female rats were housed separately throughout the experiment to prevent any potential social or pheromonal influences on behavior and seizure susceptibility. Estrous cycle staging was not performed in female rats, as seizure-induced disruptions of reproductive cycling are well-documented in epilepsy models [24], making cycle determination unreliable and potentially confounded by the treatment protocol itself [25]. Sample sizes were determined through power analysis by repeatedly simulating datasets using distribution parameters—mean, standard deviation, and distribution type—obtained from prior studies for each behavioral test’s main parameter.

2.2. Pentylenetetrazol Kindling Model of Epilepsy

By repeated administration of PTZ, we established an epileptic experimental model that exhibits characteristics similar to human temporal lobe epilepsy, which is one of the most common epileptic disorders [26,27,28]. This model is based on repeated chemical stimulation, where a subconvulsive dose of a proconvulsant substance (e.g., PTZ) is used. The kindling model demonstrates the clinical symptoms of complex partial seizures, as well as structural, cellular, and molecular changes [28].

Chronic hyperexcitability was induced in all animal groups by administering subconvulsive doses of PTZ (37.5 mg/kg body weight) intraperitoneally (i.p.) every other day for 30 days, following the experimental protocol described by Davoudi and colleagues [29]. After each PTZ injection, seizures were monitored for a one-hour period and recorded using a video system (DAHUA Technology, HDCVI 2 MP camera). Seizures were classified during the monitoring period by two non-blinded experimenters according to the modified Racine scale [30], while seizure latency and duration were analyzed retrospectively based on video recordings. Seizure intensities were included in the following stages: 0 = no response; 1 = ear and facial twitching; 2 = myoclonic jerks without rearing; 3 = myoclonic jerks with rearing; 4 = turning over into a side position with tonic–clonic seizures; and 5 = turning over into a back position, generalized tonic–clonic convulsions, and loss of balance and falling. An animal was considered kindled when it had experienced stage 4 or 5 seizures on two consecutive trials [31].

A common practice in PTZ-kindling studies is to allow a washout or stabilization period after the final kindling injection before administering a “challenge dose” to confirm whether the animals remain in a persistently lowered seizure threshold. By re-challenging the animals with a subconvulsive PTZ dose, researchers can demonstrate whether the animals have indeed developed a chronic, long-term increase in seizure susceptibility— a “kindled” state—rather than experiencing merely transient seizure activity from ongoing PTZ exposure [32]. Eighteen days after the last PTZ dose, an additional PTZ dose (a challenge PTZ dose of 37.5 mg/kg body weight) was administered. During this period, treatments were continued, and behavioral tests were performed. The 18-day window is often chosen based on prior experimental evidence indicating that this time frame is long enough to clear acute drug effects, yet short enough that kindling-induced neuroplastic changes remain stable. At the end of the experiment, all animals were euthanized via an overdose of isoflurane.

2.3. Preparation and Administration of Active Substances

In this experiment, CBD was administered orally. Crystalline CBD (99.5% purity from Trigal Pharma GmbH, Wien, Austria), dissolved in coconut oil (MCT Oil, Ostrovit, Poland), was dropped onto the food pellets (INCDMM Cantacuzino, Bucharest, Romania) to allow absorption. The pharmacokinetic parameters of this formulation were tested previously [21]. The administered doses were determined based on a thorough literature review. LEV was administered in a 300 mg/kg body weight dose, corresponding to the maximum recommended human dose calculated by Sarangi and colleagues [33]. The CBD doses were chosen based on previous studies. The lower dose was intended to be from the anxiolytic and antidepressant-like dose range of 1–30 mg/kg [22,34], while a higher CBD dose was selected to study the possible anticonvulsant synergy of the combination [21,23].

Treatment administration followed different strategies depending on the PTZ administration schedule. On PTZ administration days, every two days, LEV was administered i.p. in the form of an infusion solution (Levetiracetam SUN, 100 mg/mL, Sun Pharmaceutical Industries Europe B.V., The Netherlands) 30 min before the administration of PTZ to ensure testing at peak plasma concentration [35,36]. The control group was injected i.p. with sodium chloride 0.9% solution for infusion (B. Braun Melsungen Ag, Melsungen, Germany). On non-PTZ administration days, the LEV treatment was given orally to the rats in the form of pills along with CBD pellets.

A food plate was placed in each cage, from which the animals took and consumed the pill. The consumption of all pellets was checked and registered to ensure that each rat received the entire LEV and CBD dose within one hour. The control group always received a placebo pill without the active ingredient. For the preparation of the pills, the LEV was obtained by grinding commercially available film-coated tablets (Levetiracetam Terapia 500 mg, Sun Pharmaceutical Industries Europe B.V., Hoofddorp, The Netherlands). Lactose monohydrate was used as an excipient, food-grade glucose as a flavor enhancer, and sugar syrup as a binder. All substances were food-grade.

Table 1. Administration of active substances during the experiment.

Treatment Group	On PTZ Days	On Non-PTZ Days
FC and MC	i.p.—sodium chloride 0.9% solution for infusion	p.o.—placebo tablet
FL and ML	i.p.—300 mg/kg body weight LEV solution for infusion	p.o.—300 mg/kg body weight LEV tablet
FLC10 and MLC10	i.p.—300 mg/kg body weight LEV solution for infusion + p.o. 10 mg/kg body weight CBD pellet	p.o.—300 mg/kg body weight LEV tablet + 10 mg/kg body weight CBD pellet
FLC60 and MLC60	i.p.—300 mg/kg body weight LEV solution for infusion + p.o. 60 mg/kg body weight CBD pellet	p.o.—300 mg/kg body weight LEV tablet + 60 mg/kg body weight CBD pellet

Abbreviations: p.o.—peroral administration; i.p.—intraperitoneally administration; PTZ—pentylenetetrazol; CBD—cannabidiol; LEV—levetiracetam; FC—control group, females; MC—control group, males; FL—control group treated with LEV, females; MC—control group, males; FLC10—group treated with LEV + CBD 10 mg/kg body weight, females; MLC10—group treated with LEV + CBD10 mg/kg body weight, males; FLC60—group treated with LEV + CBD 60 mg/kg body weight, females; and MLC60—group treated with LEV+ CBD 10 mg/kg body weight, males.

shows the administration schedule for the treatments, indicating the route of administration and the pharmaceutical form.

2.4. Behavioral Tests

The behavioral assessment of the animals was performed using open-field (OF), novel object recognition (NOR), and social interaction (SI) tests. The timeline and intertrial intervals are shown in Figure 1. All trials were performed before drug administration, between 10 AM and 12 PM, corresponding to the period of lowest plasma drug concentration. Behavior was recorded using a video camera (2 MP CMOS USB camera). The recordings were later analyzed using the EthoVision^® XT software (version 11.5, Noldus IT, Wageningen, The Netherlands).

2.4.1. Open–Field Test

The test was conducted in a 60 × 60 cm box with a black floor and transparent walls that were 50 cm in height to observe the rats’ locomotor and exploratory behavior, as well as potential anxiety levels. The lighting in the testing room was adjusted to provide an average illuminance of 150 ± 10 lux at the floor. The animals were placed in the center of the test zone and allowed to explore the space undisturbed, and their behavior was recorded for 5 min by a CMOS camera (1280 × 720 pixels, 30 fps) and analyzed using EthoVision^® XT software (version 11.5, Noldus IT, Wageningen, The Netherlands). The software determined the time traveled and the time spent in the center zone based on the three-point detection of the head, nose, and tail. The following outcome measures were analyzed: total distance traveled, the number of entries, and the time spent in the center zone.

2.4.2. Novel Object Recognition Test

The NOR test was used to assess cognitive functions and memory. It consisted of two phases: (1) rats were placed in a 60 × 60 cm box with transparent walls (50 cm high), containing two identical objects, and their behavior was recorded; (2) four hours later, one of the objects was replaced with a new object of the same size but a different material. To eliminate any confounding factors, the box was cleaned with 70% ethanol between animals or phases to reduce odor cues. The animals’ behavior was observed again to evaluate short-term memory. The discrimination index (DI) shows the ability of animals to differentiate the novel and familiar objects, which was calculated using the following formula: DI = (TN − TF)/(TN + TF), where TN is the exploration time for the novel object and TF is the exploration time for the familiar object. A cutoff criterion of three seconds was set for minimal total exploration time.

2.4.3. Social Interaction Test

The treatment’s impact on social interaction was assessed after the animals had been singly housed throughout the 7-week experiment, inadvertently introducing social novelty. In this test, four behavioral categories were observed after putting together two rats of the same sex and from the same treatment group in a 60 × 60 cm enclosure for a five-minute session. For a clear response, four behavioral categories were defined as follows: passive—minimal movement or active avoidance (e.g., crouching, freezing, or remaining largely immobile); active—non-aggressive exploration or interaction (e.g., sniffing, following, gentle contact, or mild vocalization); aggressive—hostile or forceful behavior (e.g., biting, wrestling, attacking, or forceful chasing); and non-interactive—neither engaging nor avoiding (e.g., ignoring the partner, self-grooming away from the other rat, or resting without a response). Experimenters performing the analysis were not blind to the treatment groups.

2.5. Statistical Analysis

Data were analyzed and graphical representations were created using GraphPad^® Prism (version 9.3.1, San Diego, CA, USA). All data were checked for normal distribution using the Kolmogorov–Smirnov test, and depending on the outcome, either parametric or non-parametric statistical tests were applied accordingly. For detecting between-treatment group and between-sex differences, two-way ANOVA followed by Sidak’s post hoc test was used for normally distributed data, while the Kruskal–Wallis test followed by Dunn’s multiple comparisons test was used for non-parametric data. Kaplan–Meier analysis was performed to analyze the effect of treatments on the kindling process. To control for Type I errors in multiple pairwise comparisons, the false discovery rate (FDR) correction was applied to adjust the significance threshold. Data are presented as mean ± standard error of the mean (SEM) for normally distributed variables and as median with range for non-parametric data. A p-value < 0.05 was considered statistically significant.

3. Results

3.1. Antiseizure and Anti-Kindling Effects

Forty-eight rats underwent the PTZ-kindling protocol, of which 47 reached the end of the experiment. Afterward, each of them was subsequently tested with a challenge dose of PTZ. One male rat from the LEV group died for unknown reasons, but this was not related to seizures.

PTZ-induced seizures showed significant sex-dependent differences throughout the experiment. Consequently, the effects of treatment were analyzed separately on males and females. Kindling progression was characterized by the number of PTZ injections to reach the kindled state at an individual level and by the percentage of fully kindled animals at a group level. A significant difference between groups was observed with the log-rank test. After applying the FDR correction, the LEV and LEV + CBD60 groups significantly differed from the control (p = 0.0305; Figure 2). While in both LEV and LEV + CBD60 groups, kindling began to develop after the 13th dose of PTZ only, it can be observed that the LEV + CBD10 group showed a similar kindling rate to the control group, and earlier development of a fully kindled state. This means that adding 10 mg/kg CBD to LEV counteracted the protective effect of LEV in both genders.

The maximum seizure severity according to the Racine scale (RSMax), the duration of seizures with a severity of Racine 3 or higher (DRS), and the latency to first RS4 or RS5 seizures (LAT) were analyzed using a two-way ANOVA test considering treatment, sex, and the interaction between these factors. We found that RSMax significantly differed between males and females (F(1, 40) = 7.236; p = 0.0104), with females showing lower values for RSMax. The sex × treatment interaction was not significant. The ANOVA analysis, followed by Sidak’s multiple comparison test, showed a significant difference in RSMax in males between the control group and all treated groups (Figure 3a). However, females exhibited a significant difference between the FLC10 and FLC60 groups only (Figure 3b).

Conversely, DRS exhibited no differences between sexes, but the treatment had a significant impact on it (p = 0.0008). In males, the Welch’s ANOVA test showed significant differences between groups (p < 0.0001). The post hoc multiple comparisons test with FDR revealed that the MLC60 group had a significantly reduced duration of RS3-RS5 seizures compared to both MC and ML groups, while MLC10 significantly differed from MC only (Figure 4a). Although the same trend was observed in females, the differences between groups did not reach the significance level (Figure 4b).

Analyzing LAT, we found no difference between males and females. All treatment groups tended to increase LAT; however, we did not find any significant differences between the groups (Figure 5).

3.2. Behavioral Effects

3.2.1. Open Field Test

In the OF test, we evaluated four parameters: the number of entries into the central zone, the time spent in the central zone (s), the total distance moved in the central zone (cm), and the total distance moved in the test arena during the 5 min testing session (cm). The number of entries into the central zone was analyzed with the Kruskal–Wallis test, followed by Dunn’s multiple comparisons test, and showed significant differences between FL and FLC10 groups (p = 0.027, Figure 6a). The time spent in the center zone showed the same modification pattern (Figure 6b), but statistically significant differences were not observed (F(3, 39) = 1.526; p = 0.2229). The distance moved in the central zone of the apparatus showed a significant treatment effect (F(3, 35) = 2.910; p = 0.048), while no differences between males and females were observed (Figure 6c). The post hoc multiple comparisons test revealed a significant difference between the control and LEV + CBD10 groups (p = 0.0422). The total distance moved was not affected by treatments (Figure 6d), as the two-way ANOVA analysis demonstrated (F(3, 39) = 1.022; p = 0.3933). In addition, no differences between sexes were observed (F(1, 39) = 1.845; p = 0.1821 and F(1, 39) = 0.9345; p = 0.3396, respectively).

3.2.2. Novel Object Recognition Test

Two animals did not fulfill the cutoff criterion of three seconds for exploration time; thus, they were excluded from further analysis. The DI values were analyzed using the Kruskal–Wallis test, followed by Dunn’s multiple comparison test, because of the non-parametric distribution of the data. In males, no significant differences were found (KW = 0.3478; p = 0.95; Figure 7a). On the other hand, in females, the DI values were significantly different (KW = 8.941; p = 0.03) and the Dunn’s multiple comparison test showed an increase in FL group compared to FC group (median (range) −0.22 (−0.44–−0.12) vs. 0.08 (−0.10–0.31)), while the LEV + CBD combination treatments tended to decrease DI (Figure 7b).

3.2.3. Social Interaction Test

The social interaction test revealed a significant treatment effect between groups (F(3, 39) = 5.487; p = 0.003) according to the two-way ANOVA test. However, no significant sex-related differences were observed in active or passive interaction. The multiple comparisons test with FDR showed that the LEV + CBD60 group exhibited significantly reduced time spent with active interactions compared to all other groups (control = 86.9 ± 9.02 s, LEV = 83.8 ± 8.27 s, LEV + CBD10 = 82.8 ± 9.82 s, LEV + CBD60 = 44.2 ± 6.12 cm). Significant difference between the two CBD-treated groups was also observed (Figure 8).

4. Discussion

This study aimed to investigate the antiseizure and anxiolytic effects of CBD in combination with LEV using the PTZ-kindling model of epilepsy and found dose- and sex-dependent differences. The higher dose of CBD potentiated LEV’s action on reducing seizure duration, while the lower dose altered its anti-kindling effect. However, the lower CBD dose exerted some anxiolytic action by increasing central zone activity in the OF test, especially in female rats. On the other hand, it is important to mention that both LEV + CBD combinations significantly reduced the maximal intensity of PTZ-induced seizures, similar to LEV alone, but this effect was observed only in males.

The PTZ-kindling model of epilepsy was previously demonstrated to mimic some important features of epileptic syndromes, such as the progressive increase in seizure duration and severity associated with behavioral disturbances and morphological alterations of the brain. The development of the kindling could serve as an experimental parameter to assess the disease-modifying effects of ASMs. LEV was previously demonstrated to possess anti-kindling effects in various chronic epilepsy models [37,38,39,40]. On the other hand, we have previously shown that CBD alone was not capable of modifying the development of PTZ-kindling, as only the latency to generalized seizure was increased [21], but adding CBD to LEV treatment has not been characterized yet. Furthermore, this study addressed an important gap in epilepsy research, where studies have historically focused predominantly on male subjects, despite evidence of sex differences in seizure patterns, severity, and treatment responses [41]. In this study, we observed a sex- and dose-dependent effect of the LEV + CBD combination. It was observed that female rats showed lower susceptibility to PTZ-induced seizures, so the treatment did not have a significant effect on seizure parameters, while highly significant differences were observed in males. However, the behavioral assessment did not show any sex-dependent differences. Conversely, in males, the addition of 60 mg/kg CBD to LEV reduced the duration of seizures without affecting the kindling process, i.e., the onset of generalized seizures and the maximal intensity remained the same as in the case of LEV alone, while the addition of a 10 mg/kg CBD dose induced a more rapid progression in Racine score over time. This observation suggests that CBD might attenuate the anti-kindling effect of LEV.

The difference between males and females can be attributed to the influence of sex hormones on seizure susceptibility. Both estrogens and progesterone were previously demonstrated to modulate neuronal excitability, thus increasing or decreasing seizure susceptibility through allosteric modulation of GABA receptors or enhancing glutamatergic transmission [42,43]. However, it is not as straightforward as females being more or less susceptible to proconvulsants depending on the estrous cycle. Conversely, the effects of reproductive hormones on seizure intensity and duration are complex, and contradictory results were reported depending on the agent used to trigger the seizures [44]. The PTZ-kindling model was also shown to affect male and female rats differently, and the results of the current study are in accordance with a previous study by Pollo et al., with males exhibiting higher seizure scores in both studies [45]. On the other hand, previous research demonstrated that CBD’s antidepressant-like effects can also differ between males and females [46]. This study provided evidence that antiseizure and anxiety-like behavior might also diverge.

Regarding dose-dependency, this study only used two different doses, selected based on existing evidence on the CBD’s antiseizure and anxiolytic-like effects. It was found that only the higher CBD dose of 60 mg/kg was able to increase the antiseizure effects of LEV. These findings are in line with those reported by Lucchi and colleagues, who reported in the kainic acid model of epilepsy that 120 mg/kg CBD combined with 300 mg/kg body weight of LEV enhanced the effects of LEV [47]. In their study, neither LEV nor CBD administered alone significantly affected seizure parameters; however, the LEV + CBD combination reduced the occurrence and duration of spontaneous seizures [47]. Similarly, in the present study, the duration of PTZ-kindling seizures was not significantly reduced by LEV treatment alone, but was reduced by the chronic administration of LEV + CBD combination. Conversely, potential interactions between LEV and CBD were presented by Socala in acute seizure models, showing that increasing doses of CBD attenuated LEV antiseizure effects [23]. Although a pharmacokinetic interaction between LEV and CBD could not be ruled out—given evidence of increased LEV serum levels in clinical trials [48]—several mechanisms may underlie these observations. As Gilmartin and colleagues previously summarized, CBD interacts with a large number of ASMs, either by influencing pharmacokinetics or altering pharmacodynamic activity. Increased pharmacodynamic activity has been reported for clobazam, gabapentin, pregabalin, tiagabine, and topiramate, while a decrease was reported only for LEV [49]. However, this decrease was observed in a single experiment that examined the interaction between LEV and CBD under a single-dose setting [23]. The present study explored the possible interactions after chronic treatment with LEV and CBD. One possible explanation for the synergic interaction between CBD and LEV is that LEV increases the levels of the anticonvulsant neurosteroid allopregnanolone in the hippocampus [50], which is a positive allosteric modulator of GABA-A. CBD also modulates GABA-A receptors, and this may synergize with the neurosteroid-mediated enhancement induced by LEV, leading to amplified inhibition of neuronal excitability and contributing to the reduction in seizure duration and severity. Additionally, both compounds influence synaptic vesicle function, with LEV acting via SV2A binding [9,51] and CBD reported to modulate presynaptic neurotransmitter release [52], potentially resulting in overlapping or additive effects on synaptic transmission regulation. However, this dual modulation might also explain the disruption of LEV’s protective effect at lower CBD doses, where subthreshold or partial receptor activation may unbalance the excitatory–inhibitory homeostasis. Future studies should focus on combining lower CBD doses with LEV to better understand the potential disruptions of LEV’s efficacy.

The behavioral features associated with seizures and ASMs might pose further challenges to the management of epileptic patients. Anxiety and depression are common comorbidities that need drug therapy, with several studies reporting the anxiety-related side effects of LEV in humans [53]. Consequently, combining CBD with LEV might not only improve seizure control but also address neuropsychiatric issues. However, in animal models, chronic LEV administration showed no anxiogenic effects in the OF test [54,55,56]. In addition, no influence of LEV treatment was reported on the OF test parameters in PTZ-kindled mice [57]. In this study, in accordance with previously published data, the OF test showed that LEV treatment did not influence behavioral parameters; however, when 10 mg/kg CBD was added to LEV treatment, some anxiolytic effects were observed, reflected by increased central zone activity. Conversely, LEV administered in combination with a high dose of 60 mg/kg CBD did not show any effect on OF parameters. Previous studies also confirmed that CBD alone in a dose of 60 mg/kg had no effects on anxiety-related parameters in the OF test [21]. In an earlier study performed using the chronic mild stress model of depression, 10 mg/kg CBD alone increased the vertical exploration of rats, another important parameter characterizing anxiety level [34]. Moreover, Campos and colleagues also described anxiolytic effects of CBD in the elevated plus maze test [58]. Nevertheless, several mechanisms were proposed to explain the inverted U-shape dose–response curve of the anxiolytic effects of CBD [59], which is similar to some other widely used antidepressants with serotonergic mechanisms [60]. Other research suggests that low CBD doses enhance GABAergic transmission, promoting anxiolytic effects, while higher doses might counteract these effects by inhibiting the GABA–A receptor modulation, thereby increasing anxiety-like behaviors [61]. Overall, the anxiolytic action of low-dose CBD in combination with LEV was observed in this study, and further investigations should focus on its clinical therapeutic benefits.

The NOR test results showed no significant impact of either dose of CBD or LEV on cognitive functions, though a slight, but non-significant, increase in the LEV group was noted. It is important to highlight this negative finding because many previous studies reported ASMs to have an undesirable impact on cognition. Although a clinical study showed that LEV was superior to topiramate regarding the cognitive effects, showing that no changes in cognitive performance were observed in the LEV group [62], the combination of LEV and CBD could have had a different influence on cognition. An early preclinical study used the Morris water maze (MWM) test to demonstrate that LEV does not impair cognitive functions, making it a safe choice for epilepsy treatment without negatively affecting learning and memory [63]. Recently, Zwierzyńska and colleagues conducted an interesting study to assess LEV’s effects on short- and long-term memory using MWM and NOR tests and found an improved performance in rats, which had been exposed to ethanol. However, LEV caused memory and locomotor disturbances after repeated doses [64,65]. On the other hand, our group previously found that chronic treatment with 60 mg/kg CBD significantly impaired the discrimination between novel and familiar objects, i.e., reduced DI compared to controls [21]. Consequently, this study also evaluated the recognition memory of rats and showed that no significant difference between controls and LEV + CBD-treated rats was found.

Conversely, the LEV + CBD60 group exhibited a significant reduction in active social interactions and locomotor activity in the SI test. This is interesting because although the reduction in social interaction was previously demonstrated to occur due to cannabinoid treatment, 20 mg/kg CBD was demonstrated to reverse the Δ9-tetrahydrocannabinol-induced reduction [66]. Recent evidence suggest that CBD’s action on social interaction might manifest in low doses up to 30 mg/kg [67,68,69]. On the other hand, LEV was also shown to ameliorate stress-induced reduction in social interaction [70]. However, our findings are in accordance with Almeida and colleagues, who performed a dose–response investigation of social interaction and found that 60 mg/kg CBD significantly reduced the total social interaction in spontaneously hypertensive rats [71]. This study adds to the existing evidence that epileptic rats might be more vulnerable to the action of high-dose CBD in terms of altering social interaction.

This study has several limitations that should be considered when interpreting the results. First, although interactions between CBD and LEV were observed, no pharmacokinetic data were collected to determine whether these effects were due to altered drug metabolism or plasma levels. Second, the behavioral testing was unblinded and limited to OF, NOR, and SI tests, potentially overlooking other relevant domains such as depression-like behavior or spatial memory. Furthermore, using a single OF test to assess anxiety might not capture all aspects of behavioral changes. Third, while sex differences were noted, the biological mechanisms underlying these effects—such as hormonal influences or receptor expression—were not explored. Finally, repeated PTZ injections and single housing may have introduced stress-related confounds that could have affected anxiety-related behavioral outcomes in the absence of stress control groups or biomarker assessments.

Overall, the study underscores the importance of CBD dose titration and confirms its anticonvulsant action, which may be beneficial even when combined with the first-line ASM levetiracetam. On the other hand, the results presented here highlight the need for caution when using CBD as an anxiolytic adjunct to minimize ASM-related side effects, as it may paradoxically increase seizure susceptibility in certain forms of epilepsy.

5. Conclusions

The LEV-CBD combination showed a more effective anticonvulsant effect than LEV in monotherapy in the PTZ-induced chronic epilepsy model using rats. LEV alone prevented kindling and reduced the maximal intensity of seizures, but seizure duration and latency to generalized seizures were not affected. Conversely, by combining LEV with CBD, a significant reduction in seizure duration was also observed in addition to the reduced maximal intensity of seizures. On the other hand, the LEV + CBD combination showed dose-dependent anti-kindling and anxiolytic effects in rats: at the lower CBD dose, it exhibited an anxiolytic effect, but accelerated the development of kindling, while at the high CBD dose, it further reduced the duration of seizures compared to LEV alone without affecting the kindling process. However, the addition of a high CBD dose negatively impacted social interactions and exploratory behavior.

Another important finding was the sex-specific differences in seizure parameters and anxiety-like behaviors. Female rats showed no significant changes in maximal seizure intensity and seizure duration after LEV + CBD combination treatment but exhibited greater sensitivity to the anxiolytic effects of a low CBD dose, as well as enhanced cognitive benefits from LEV treatment in the novel object recognition test. These suggest that hormonal factors may modulate the behavioral response to CBD-LEV co-administration. Our findings support that the combination of CBD and LEV may be potentially useful for improving seizure control, while it is important to consider the optimal dosing of CBD to minimize neuropsychiatric side effects. However, further studies and validations are needed to assess the safety of this combination.