Evenamide reverses schizophrenia-related dysfunction in a neurodevelopmental animal model

Abstract

Schizophrenia is characterized by positive, negative, and cognitive symptoms. However, current D2-based antipsychotic drugs only address primarily positive symptoms. Limbic hippocampus hyperexcitability is a key pathological state of schizophrenia, representing an ideal therapeutic target. Evenamide is a selective voltage-gated sodium channel blocker that reduces neuronal hyperexcitability. We examined the effect of acute evenamide treatment on the hyperdopaminergic state, hippocampal hyperexcitability, social deficits, and recognition memory in the methylazoxymethanol acetate (MAM) neurodevelopmental model. Male and female Sprague-Dawley offspring from dams treated with saline or MAM on gestational day 17 were tested as adults (postnatal day >65). Electrophysiological recordings were made in the ventral tegmental area (VTA) and ventral hippocampus (vHipp) and social approach and novel object recognition were tested. Evenamide (3 mg/kk i.p.) normalized the number of spontaneously active DA neurons in the VTA of female and male MAM rats and reduced pyramidal neuron hyperactivity in the vHipp. The hyperdopaminergic state in the VTA of female and male MAM rats was also rescued by local evenamide injection in the vHipp (1μM). Systemic evenamide also reversed the recognition memory impairment of female and male MAM rats. For social deficits, only male MAM rats exhibit a reduced social sniffing time that was normalized by evenamide. These findings suggest that evenamide’s efficacy in downregulating the hyperdopaminergic state, social deficits, and recognition memory impairment may result from its ability to attenuate vHipp hyperexcitability. Therefore, evenamide could offer a novel therapeutic strategy that is capable of addressing positive, cognitive, and negative symptoms of schizophrenia.

1. Introduction

Schizophrenia is a neurodevelopmental disorder with a prevalence of ~1% worldwide [1,2]. The disorder consists of positive symptoms (delusions and hallucinations), negative (social withdrawal, anhedonia), and cognitive symptoms (working memory and decision-making impairments) [1,3,4]. A large portion of available antipsychotic drugs are only effective in treating positive symptoms [5,6] which is associated with increased dopaminergic states [7–9]. The neurobiological basis of antipsychotic agents has historically been focused on the dopamine (DA) system and relies on its ability to block type 2 DA receptor (D2) [10–12]. However, there is a substantial population of schizophrenia patients who fail to respond to antipsychotic drug treatment, known as treatment-resistant patients [13]. Furthermore, even in patients who respond, antipsychotic drugs are ineffective in controlling other major symptoms associated with schizophrenia, primarily cognitive and negative symptoms [14–16] which are often associated with poor functional outcomes [17–19].

A better approach to treat all symptoms would be targeting the key site of pathology in schizophrenia; i.e. the limbic anterior hippocampus [20,21]. Imaging studies have pointed to hyperactivity in the anterior hippocampus of schizophrenia patients [20,22], and in postmortem studies there is a substantial loss of GABAergic parvalbumin inhibitory neurons in this area [23]. Preclinical models have successfully recapitulated the hippocampal hyperexcitability domain, such as the neurodevelopmental methylazoxymethanol acetate (MAM) model [24,25]. The MAM model consists of injecting MAM during gestational day 17 into pregnant rats at a time that approximates the human second trimester; a period of vulnerability during pregnancy during which insults can result in increased schizophrenia incidence in adult offspring [26,27]. The MAM-treated offspring as adults show multiple anatomical, behavioral, neurochemical, and physiological changes consistent with schizophrenia [26,28]. The MAM model exhibits different dysregulated behaviors that includes impaired cognitive performance, social deficits, and enhanced stress responsivity, reflecting behavioral domains associated with the cognitive and negative symptoms of schizophrenia [28]. Indeed, the MAM rats show a loss of hippocampal PV neurons, limbic ventral hippocampal hyperactivity (vHipp), and altered rhythmicity [24,25], and consequently, overdrive of the DA neurons projecting to the associative striatum [25,29]. This disruption of hippocampal activity is thought to contribute to the cognitive and affective behavioral disturbances observed in this model potentially via other efferent pathways[16,28,30]. It has been proposed that, by targeting the site of origin of the pathology within the hippocampus, a drug should be able to alleviate not only the DA pathology and psychosis but also the negative and cognitive deficits [9,16]^.

Evenamide (formerly NW-3509) is a unique compound able to target hippocampal hyperexcitability, as it selectively inhibits hyperactive neurons [31]. Evenamide is a voltage-gated sodium channel blocker that is devoid of activity at any other central nervous system target, and it normalizes excessive synaptic glutamate induced by NMDA hypofunction along with a reduction of cortical and hippocampal hyperexcitability [32,33]. This is achieved by modulating neurons that have sustained repetitive firing properties without impacting baseline activity [34,35]. It has been described that evenamide alone or in combination with clozapine normalizes the prepulse inhibition impairment induced by ketamine [36]. Indeed, the potential effects of evenamide on negative symtoms were established by reversing the asociality induced by the glutamate antagonist phencyclidine, when given in combination with aripiprazole [37]. In the clinical field, evenamide is a safe compound [32] that was proven effective as an add-on for 4-week trials in improving the Positive and Negative Syndrome Scale (PANSS) [38]. The long-term efficacy was investigated in a phase-II open-label clinical trial as an add-on in treatment-resistant patients who demonstrated progressive clinical improvement (PANSS score and Clinical Global impression) across the 1-year study. There were no significant safety and tolerability abnormalities for the tested doses (7.5, 15, and 30 mg bid), and a high retention rate was observed at up to 6-month and 1-year tests [39,40]. A phase III double-blind and placebo-controlled study with evenamide (30mg bid) in patients with chronic schizophrenia that have poor response to second-generation antipsychotic drugs demonstrated improvement of PANSS score after 4 weeks of treatment compared to the placebo group [41]. Along with efficacy in treating positive symptoms, the hypothesized downregulation of the hyperactive limbic hippocampus would also suggest a potential to impact negative symptoms and cognitive deficits. In fact, in patients with chronic schizophrenia not responding to their current 2^nd generation antipsychotic drugs, the addition of evenamide led to a significant improvement in negative symptoms compared with patients receiving placebo [41]. Indeed, this could account for the high level of compliance of patients in evenamide trials. Therefore, the study’s aim was to investigate the effect of acute evenamide treatment on hyperdopaminergic state in the VTA, hyperactive vHipp, and recognition memory and social behaviors in a preclinical neurodevelopmental MAM model of schizophrenia.

2. Material and Methods

2.1. Animals

Pregnant Sprague-Dawley rats were housed with food and water available ad libitum under standard conditions in temperature-controlled rooms (22 ± 1 °C). The animal facilities were kept in a constant 12-hour light/dark cycle with one group experiencing light from 7 am-7 pm and one group experiencing light from 7 pm-7 am. Behavioral experiments (novel object recognition and social approach) were tested during the dark cycle while the electrophysiological recordings were done during the light cycle, as anesthesia is known to suppress behavioral-state-dependent and circadian modulation of neuronal activity [42]. Both male and female rats were tested after acute evenamide treatment, with testing occurring at exactly 10 minutes following injection of the drug for behavior tests. All procedures were performed according to the guidelines outlined in The National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at the University of Pittsburgh.

2.2. MAM Treatment

The MAM treatment procedure followed methods established in our previous studies [27]. Pregnant Sprague-Dawley rats were obtained on GD15 and received treatment on GD17 of either MAM (20mg/kg i.p) or Saline (1ml/kg i.p). The dams were housed individually in well-ventilated plastic breeding containers. The date of birth usually occurred on gestational day 22 and the pups were monitored closely until they reached the weaning stage at postnatal day (PD) 21, where they were then housed in pairs with their littermates and tested at adulthood (PD > 65). For each treatment group, only two male and female rats of the same dam were used (one drug, one veh for each sex), totaling four rats from the same dam per experiment. Over the course of the study, we used 11 Sal-treated and 12 MAM-treated dams. Independent groups of rats were used for behavioral and electrophysiological experiments. For behavioral assays, a different set of rats was used for each test to rule out long-term effects of evenamide on subsequent behaviors.

2.5. Drug preparation and administration

Only acute drug treatments were utilized in this study. Acute treatment was chosen based on the hypothesis that evenamide would act at the pathological site of dysfunction, the vHipp, leading to immediate electrophysiological and behavioral responses. Indeed, the acute dose of 3mg/kg of evenamide was based on previous studies conducted in rats [36,37]. Evenamide was diluted using Saline 0.9% and was prepared before injection. Each dose was administered via intraperitoneal injection at 3mg/kg (injection volume 5ml/kg) from a solution of concentration 0.6mg/ml. All behavioral tests occurred 10 minutes following the initial injection and electrophysiological recording started right after the injection.

2.6. Novel Object Recognition

Rats were tested in a rectangular arena with measurements L70 cm × W40 cm × H30 cm. Day 1 of the test consisted of a habituation phase in which the rats were recorded exploring the empty box for 10 minutes. The following day featured a 5-minute training phase and a 5-minute testing phase. During the training phase, the rats were exposed to two identical objects placed in symmetrical locations within the box. The time spent exploring each object was recorded, which can be classified as animal-directed attention to the object, such as sniffing, licking, or touching. Sitting on or using the object for purposes other than intentional exploration was not recorded. The interval between the training and testing phase was 1 hour. The testing phase began with the injection of either evenamide (3mg/kg) or saline (5 mL/kg), the return of animals to their cage, and subsequent testing after 10 minutes. One familiar object from the training phase and one novel object were placed in symmetrical locations in the box. The discrimination index was used to quantify recognition memory and was calculated as follows: [(Novel object. exploration) − (Familiar object exploration)]/Total Exploration (total time exploring both the novel and the familiar). The arena and objects were cleaned thoroughly with Quatricide after each animal. All tests were recorded and saved to ensure accurate data collection.

2.7. Social Approach

Testing occurred in the same rectangular arena divided into three chambers – one smaller middle chamber and two larger side chambers. All rats were habituated to the testing room for at least 30 minutes, and the test rats were habituated to the arena with no stimuli for at least 5 minutes, with a minimum exploration time of 1 minute per chamber. The middle chamber was empty, and each side chamber was assigned a Social Chamber or Toy Chamber. The social chamber contained a metal cage with a younger rat (PD 30–45) of the same sex, which allowed the test rat to interact with but not directly contact the social stimulus rat which prevents aggressive or sexual behaviors. The toy chamber contained an identical metal cage with a toy rat. For the testing phase, rats were recorded in the arena for 10 minutes. Six variables were recorded: social chamber time (s), toy chamber time (s), middle chamber time (s), time sniffing social stimulus (s), # of entries to social, and # of entries to toy chamber.

2.8. Electrophysiology

Rats were anesthetized using 8% chloral hydrate (400mg/kg, i.p.), with anesthesia checked every 30 minutes using the hindlimb compression reflex and supplemented as needed. Once sedated, animals were fixed in a stereotaxic frame (Kopf), and their body temperature was maintained at 37°C via a heating pad and monitored with a thermometer. In vivo extracellular single-unit electrophysiological recordings were performed using glass microelectrodes constructed from Omegadot 2.0mm glass tubing with a Narishige P-5 vertical electrode puller (Japan). The electrode was filled with 2M NaCl and 2% Sky Blue dye solution to confirm the recording site in histological analysis by dye ejection at the end of the recording. Electrodes were advanced into the VTA using a hydraulic microdrive system (Kopf) targeting VTA coordinates of anteroposterior −5.3–5.7 mm from bregma, mediolateral ±0.6–1.0 mm from midline, and ventral depths ranging from 6.5 – 9.0mm from brain surface. For vHipp, the coordinates used were anteroposterior −5.7–5.9 mm from bregma, mediolateral ±4.6–5.0 from midline, and ventral −6.5–8 mm from brain surface.

DA neurons and pyramidal neurons were identified based on electrophysiological characteristics established in previous studies [43–45]. To assess population activity within the VTA and vHipp, neurons were counted within 6–9 vertical electrode passes per animal, with each pass spaced 0.2 μm apart (Fig 1B). Each neuron was recorded for 3 min, during which the following factors were collected: (1) population activity, or the number of spontaneously active neurons recorded per track; (2) firing rate; and (3) % of action potentials in bursts, with burst initiation defined as the presence of two spikes with an interspike interval of 80ms and termination with interspike intervals >160 ms [45,46]. Analysis was conducted by categorizing the electrode recordings based on their relative locations in the medial, central, and lateral tracks of the VTA in relation to population activity, firing rate, and the percentage of spikes in each burst. VTA and vHipp were recorded in both brain hemispheres, with the first hemisphere serving as the baseline/no drug condition and the second the hemisphere tested after evenamide treatment (systemic i.p. administration or direct infused in the vHipp). Although all measures were analyzed for both VTA and vHipp, the number of spontaneously active DA neurons per track and firing rate represent the most relevant measures for VTA and vHipp, respectively, in the MAM model [25–27,30].

Figure 1. Evenamide treatment (3 mg/kg, i.p.) normalized the increased number of spontaneously active DA neurons in both male and female MAM rats.

Single-unit extracellular electrophysiological recordings were performed in the VTA (A) following a grid pattern across the anteroposterior and mediolateral axes (B), as illustrated by a representative histological electrode placement (C). Evenamide reversed the increased number of spontaneously active DA neurons per track in both male and female MAM rats (D and G, respectively), without affecting the Saline groups. In male MAM rats, the reduction in DA neuron number was confined to the central and lateral regions of the VTA (E), while in female MAM rats, it was limited to the medial and lateral regions (H). Time-course analysis revealed that the decrease in DA neuron number in male MAM rats persisted over a 3-hour period (F), whereas in female MAM rats, the reduction was evident only at the 2–3-hour time point (I).

2.9. Histology

Following electrophysiological recording, Chicago Sky Blue dye was administered via electrophoretic ejection (−20mA for 20 min) into the electrode site. Rats were euthanized using a CO₂ chamber. Their brains were removed and stored in an 8% paraformaldehyde solution for 48 hours, followed by a 25% sucrose solution to ensure cryoprotection. Brains were sliced in coronal sections (60mm) using a Leica Frigocut 280 cryostat. Neutral red and cresyl violet stains were used for the identification of VTA electrode locations.

2.10. Statistical analysis

The data are represented as the mean ± SEM. All data sets were tested for normality using the Shapiro-Wilk test. For a normal data set, a two-way ANOVA test, with factors of treatment (evenamide, veh) vs condition (Saline, MAM) was used which was followed by Sidak’s multiple comparison test. The Kruskal-Wallis followed by Dunn’s multiple comparison tests was used for data that did not have a normal distribution. P<0.05 was set as statistically significant.

3. Results

3.1. Evenamide normalizes the increased dopaminergic activity in the VTA of MAM rats.

Electrophysiological recordings of the VTA were performed in adult rats (Figure 1A) using a grid pattern across the VTA region (Figure 1B), with subsequent histological verification of electrode placement (Figure 1C). Male and female MAM rats exhibited an increased number of spontaneously active DA neurons per electrode track in the VTA which was normalized by evenamide treatment (Male: MAM condition, F_1,26=4.32, p=0.0476; Treatment, F_1,26=55.18, p<0.0001; Interaction, F_1,26=33.41, p<0.0001, Figure 1D/Females: MAM condition, F_1,24=4.89, p=0.0367; Treatment, F_1,24=16.55, p=0.0004; Interaction, F_1,24=15.41, p=0.0006, Figure 1G; ANOVA). The analysis of DA neuron firing rate indicated no effect of evenamide treatment in males and females across Saline and MAM conditions (Male, p=0.1857; Female, p=0.2824; Kruskal-Wallis, Supplementary Figure 1A and G). The analysis of % of spikes in bursts indicated a significant effect in both male (U=7.92, p=0.0477, Kruskal-Wallis, Supplementary Figure 1B) and female (U=10.08, p=0.0179, Kruskal-Wallis, Supplementary Figure 1H) groups. In males, the multiple comparison test (Dunn’s test) did not reveal any significant change across the groups while in females a difference was found between the Saline – no drug and MAM – evenamide group (p=0.0143, Dunn’s test).

The VTA data was analyzed based on the electrode placement within the medial, central, and lateral portions of the VTA. In male rats, a significant increase in the number of spontaneously active DA neurons per track was observed in the lateral portion of the VTA of MAM – drug compared to Saline – no drug in which evenamide was able to normalize the MAM condition (U=17.86, p=0.0005, Kruskal-Wallis, Dunn’s multiple comparison test, Figure 1E). In the central portion of the VTA, evenamide decreased the number of spontaneously active DA neurons per track only in the male MAM group (U=14.11, p=0.0028, Kruskal-Wallis, Dunn’s multiple comparison test, Figure 1E). No effect was observed in the medial portion of VTA of males across conditions and treatment (MAM, p=7507; Treatment, p=0.2486; Interaction, p=0.924; ANOVA, Figure 1E). In females, a significant effect was observed in the medial segment of the VTA with MAM – no drug group having a higher number of active DA neurons per track than both Saline groups, and in which evenamide was able to normalize it (MAM, F_1,24=4.95, p=0.0358; Treatment, F_1,24=4.95, p=0.0358; Interaction, F_1,24=4.95; p=0.0358; ANOVA, Sidak’s multiple comparisons test, Figure 1H). In the central portion of the VTA of females, a treatment effect was observed regarding the number of active DA neurons per track but multiple comparison tests did not reveal any difference across all groups (Treatment, F_1,24=4.77, p=0.0389; MAM, p=0.3379; Interaction, p=0.1039; ANOVA, Sidak’s multiple comparisons test, Figure 1H). Analysis of the lateral portion of VTA in females indicated that evenamide decreased the number of spontaneously active DA neurons per track only in the MAM group (Treatment, F_1,24=7.79, p=0.0101; Interaction, F_1,24=9.35, p=0.0054; MAM, p=0.6149; ANOVA, Sidak’s multiple comparisons test, Figure 1H). Firing rate and % of spikes in bursts were not different across conditions and treatment in male rats (Kruskal-Wallis, Supplementary Figure 1C and D). In females, no changes were observed for firing rate and % of spikes in bursts in the medial and central portion of VTA in all groups (Kruskal-Wallis, Supplementary Figure 1I and J). For the lateral portion of the VTA, a difference was observed between the MAM – evenamide group and both Saline groups (no drug and evenamide) for firing rate in females (U=9.46, p=0.0237, Kruskal-Wallis, Dunn’s multiple comparisons test, Supplementary Figure I). Indeed, the difference was noted for % of spikes in bursts in the lateral portion of VTA of females between Saline – no drug and MAM – evenamide groups (U=7.99, p=0.0463, Kruskal-Wallis, Dunn’s multiple comparisons test, Supplementary Figure J).

The VTA data was also analyzed based on different times across the recording period (0–1h, 1–2h, and 2–3h). In males, a consistent and sustained decrease in the number of spontaneously active DA neurons per track was observed in all time points after evenamide treatment in the MAM group (0–1h, U=8.28, p=0.0406; 1–2h, U=11.53, p=0.0092; 2–3h, U=14.63, p=0.0022; Kruskal-Wallis, Dunn’s multiple comparisons test, Figure 1F). Females only demonstrated a decrease in the number of active DA neurons per track of MAM rats in the last time-point of recording (0–1h: MAM, p=0.072; Treatment, p=0.1154; Interaction, p=0.7044; ANOVA; 1–2h, p=0.0613, Kruskal-Wallis; 2–3h, Treatment, F_1,24=10.53, p=0.0034; Interaction, F_1,24=5.53, p=0.0272; MAM, p=0.5777; ANOVA, Sidak’s multiple comparisons test; Figure 1I). The firing rate of DA neurons did not differ across conditions and treatment in all the time points measured in both males and females (Kruskal-Wallis, Supplementary Figure 1E and K). In males, a difference in the % of spikes in bursts was noted between Saline – no drug and MAM – no drug only in the last recording time-point (0–1h, p=0.4439; 1–2h, p=0.5522; 2–3h, U=9.62, p=0.0221; Kruskal-Wallis, Supplementary Figure 1F). For females, the difference in % of spikes in bursts was observed between Saline – no drug and MAM – evenamide only during the first recording time-point (0–1h, U=9.61, p=0.0222; 1–2h, p=0.5576; 2–3h, p=0.2058; Kruskal-Wallis, Dunn’s multiple comparisons test, Supplementary Figure 1L).

3.2. Evenamide attenuates the hyperactivity of pyramidal neurons in the vHipp observed in MAM rats.

The effect of evenamide (3mg/kg) was assessed in vHipp recordings (Figure 2A) in a grid pattern across vHipp extension (Figure 2B) to determine if it could normalize the hyperexcitability of pyramidal neurons. The electrode locations were confirmed to be in the vHipp through histological analysis (Figure 2C). There was no effect of MAM treatment on the number of spontaneously active pyramidal neurons per track in males (Figure 2D) (p=0.7106, ANOVA), evenamide treatment (p=0.2191, ANOVA), or interaction (p=0.2352, ANOVA). In females, only an effect of MAM was observed (F_1,24=4.29, p=0.0492, ANOVA, Figure 2G) but not treatment (p=0.4424) or interaction (p=0.074) for the number of active pyramidal neurons. However, the multiple comparisons test did not indicate any difference across all groups. The firing rate of pyramidal neurons in the vHipp increased in both male and female MAM rats (MAM – no drug) compared to Saline groups (Male, U=12.44, p=0.006, Figure 2E; Female, U=16.09, p=0.0011, Figure 2H; Kruskal-Wallis followed by Dunn’s comparisons test). The % of spikes in bursts was not changed across treatment and condition in males (p=0.1621, Kruskal-Wallis, Figure 2F). In females, a decrease in the % of spikes in bursts was noted in the MAM – evenamide group compared to MAM – no drug group (U=9.73, p=0.021, Kruskal-Wallis, Figure 2I).

Figure 2. Evenamide treatment (3 mg/kg, i.p.) reduced the elevated firing rate of pyramidal neurons in the vHipp of male and female MAM rats.

Single-unit electrophysiological recordings of pyramidal neurons in the vHipp (A) were conducted using a grid pattern spanning the mediolateral and anteroposterior axes (6 tracks; B), as illustrated by a representative histological electrode placement (C). Evenamide did not affect the number of spontaneously active pyramidal neurons in either male or female Saline or MAM rats (D and G, respectively). However, it significantly reduced the increased firing rate of pyramidal neurons in male (E) and female (H) MAM rats. In males, evenamide did not alter the % of spikes in bursts (F), whereas, in female MAM rats, it significantly reduced this parameter (I).

The analysis of vHipp recording time did reveal an effect of evenamide treatment only in the first hour of recording in males for the number of pyramidal neurons per track (Treatment, F_1,24=4.33, p=0.0482, ANOVA, Supplementary Figure 2A); however, Sidak’s multiple comparisons test did not find any difference between the groups. In females, evenamide decreased the number of pyramidal neurons per track in MAM rats in the last hour of the recording (2–3h, U=10.16, p=0.0173, Kruskal-Wallis, Dunn’s multiple comparisons test, Supplementary Figure 2D). The analysis of the pyramidal neuron firing rate across recording time found an effect in the last hour of recording in males with the MAM – evenamide group having a lower firing rate average than the Saline – evenamide group (2–3h, U=9.91, p=0.0194; Kruskal-Wallis, Supplementary Figure 2B). The same analysis in females found a decrease in the firing rate of MAM – evenamide compared to MAM – no drug group in the first hour of recording (0–1h, U=7.82, p=0.0498, Kruskal-Wallis, Supplementary Figure 2E). For % of spikes in bursts, no effect was found in both saline and MAM groups after evenamide treatment in males (Kruskal-Wallis, Supplementary Figure 2C) across the 3-hour long recording. However, a decrease in the % of spikes in bursts was observed in the MAM-evenamide group compared to saline–evenamide in the first hour of recording (0–1h, U=8.81, p=0.0319, Kruskal-Wallis, Supplementary Figure 2F).

As the firing rate of pyramidal neurons is the most relevant data affected by evenamide in the vHipp, analysis of the distribution of neurons in different firing rate ranges was conducted to evaluate if evenamide would affect a specific range of pyramidal neurons. However, no effect was found between evenamide groups in both sexes of Saline and MAM rats (Kolmogorov-Smirnov test, Supplementary Figure 3A–F).

3.3. Local injection of evenamide in the vHipp normalizes the increased dopaminergic activity in the VTA of MAM rats.

VTA activity was evaluated following evenamide infusion (1 μM) into the vHipp (Figure 3A), with subsequent confirmation of electrode placement in the VTA and infusion site in the vHipp (Figure 3B). Both male and female MAM rats exhibited increased number of spontaneously active DA neurons per track in the VTA, which was normalized by a 1μM local evenamide infused into the vHipp (Male: MAM, F_1,22=18.50, p=0.0003; Treatment, F_1,22=39.12, p<0.0001; Interaction, F_1,22=8.17, p=0.0091, Figure 3C; Females: MAM, F_1,22=6.23, p=0.0205; Treatment, F_1,22=20.75, p=0.0002; Interaction, F_1,22=12.53, p=0.0018, Figure 3F, ANOVA). The DA neuron firing rate was unaffected by evenamide treatment in both males and females across all conditions (Male, p=0.6823; Female, p=0.0889; Kruskal-Wallis, Supplementary Figure 4A and G). The % of spikes in bursts also remained unaffected in both males and females following evenamide treatment (Male, p=0.9429; Female, p=0.7097; Kruskal-Wallis, Supplementary Figure 4B and H).

Figure 3. Evenamide Treatment (1μM vHipp local injection) normalized the increased number of active DA neurons per track in the VTA of both male and female MAM rats.

Single-unit electrophysiological recordings of cells in the VTA (A) were performed according to a grid pattern along the anteroposterior and mediolateral axes, represented by histological verification of electrode position (B). Males and Females treated with MAM exhibited an increase in active DA neurons per track, which was normalized by evenamide treatment (C and F, respectively); Saline groups were not affected. In male rats, a difference was noted in relation to Saline – evenamide groups and MAM – no drug for the medial and lateral portion of VTA (D). In female MAM rats, evenamide treatment produced a decrease in DA neurons per track in the medial and central regions (G). Further analysis showed that males exhibited a decrease in the number of DA neurons per track during the 0–1hr and 2–3hr periods (E), while females did not exhibit this decrease in any specific period (H).

VTA data was analyzed based on electrode location in the medial, central, and lateral tracks of the VTA. In male rats, an increase in the number of DA neurons per track was observed between MAM – no drug group only in relation to Saline – evenamide group in the medial and lateral portion of the VTA (Medial, U=15.06, p=0.0018; Lateral, U=11.24, p=0.0105; Kruskal-Wallis, Dunn’s multiple comparison test, Figure 3D). In female MAM rats, evenamide treatment produced a significant decrease in DA neurons per track in the medial portion of the VTA (U=8.61, p=0.0349, Kruskal-Wallis, Dunn’s multiple comparison test, Figure 3G). Female MAM rats showed an increase in DA neurons per track compared to Saline-evenamide, which was normalized by evenamide treatment in the central portion of VTA (MAM, F_1,22=16.70, p=0.0005; Treatment, p=0.0766; Interaction, p=0.1157; ANOVA, Sidak’s multiple comparison test, Figure 1H). There were no significant differences found in the lateral VTA of female rats (p=0.7551, Kruskal-Wallis, Figure 3H). For both males and females, there were no changes in firing rate or % of spikes in bursts observed in the medial, central, or lateral tracks of the VTA across all conditions (Kruskal-Wallis; Males, Supplementary Figure 4C–J; Females, Supplementary Figure 4I–L).

The VTA data was analyzed according to timed sections of recording (0–1h, 1–2h, 2–3h). In male MAM rats, there was a significant increase in active DA neurons per track when compared to Saline-Veh groups, which was normalized by evenamide treatment in the first hour of recording (0–1h: MAM, p=0.0536; Treatment, F_1,21=11.71, p=0.0026; Interaction, F_1,21=4.35, p=0.0493; ANOVA, Sidak’s multiple comparison test, Figure 3E). An increased number of DA neurons per track in MAM-no drug was observed compared to saline-evenamide in the 1–2 h time point in males (MAM, p=0.0699; Treatment, F_1,22=10.66, p=0.0035, Interaction, p=0.5748, ANOVA, Figure 3E). In the last hour of recording (2–3h), an increased number of DA neurons per track was noted in the MAM-no drug in relation to saline-evenamide and MAM-evenamide group (MAM, F_1,22=5.77, p=0.0252; Treatment, F_1,22=8.05, p=0.0096, Interaction, p=0.1758, ANOVA, Figure 3E). Evenamide had no effect on firing rate for all time points in male rats (Kruskal-Wallis Test, Supplementary Figure 4E). There was a significant decrease in % of spikes in bursts observed when male MAM rats were treated with evenamide during the 1–2h period (U=8.41, p=0.0383, Kruskal-Wallis Test, Dunn’s Multiple Comparison Test, Supplementary Figure 4F). Female MAM rats showed no changes in active DA neurons per track across all time points (Kruskal-Wallis Test, Figure 3H). There was a significant increase in firing rate found in female MAM rats when compared to Saline-evenamide during the 2–3h period (U=10.70, p=0.0134, Kruskal-Wallis Test, Dunn’s Multiple Comparison Test, Supplementary Figure 4K). No changes were detected in % of spikes in bursts in female rats across all time points (Kruskal-Wallis Test, Supplementary Figure 4L).

3.4. Evenamide restores the recognition memory of MAM rats in the novel object recognition test.

Evenamide was tested in the novel object recognition test to evaluate the recognition memory. Evenamide was injected 10 min before the testing phase (Figure 4A). In males, evenamide rescued the decreased discrimination index observed in MAM rats without affecting the Saline group (Treatment, F_1,25=8.90, p=0.0063; MAM, F_1,25=4.41, p=0.0461; Interaction, F_1,25=5.96, p=0.022; ANOVA, Figure 4B). The total exploration of both objects in the testing phase was not affected by evenamide treatment in male Saline and MAM rats (Treatment, p=0.0976; MAM, p=0.7361; Interaction, p=0.8033; ANOVA, Figure 4C). In females, evenamide also reversed the reduced discrimination index in MAM rats without affecting the Saline condition (U=12.48, p=0.0059, Kruskal-Wallis, Figure 4D). The total exploration was reduced in MAM rats treated with evenamide compared to Saline rats treated with vehicle (Treatment, F_1,28=7.0, p=0.013; MAM, F_1,28=5.3, p=0.028; Interaction, p=0.3235; ANOVA, Figure 4E).

Figure 4. Evenamide rescues cognitive deficits in male and female MAM rats.

Rats were habituated to the testing chamber for 10 minutes. On the following day, they were exposed to two identical objects during the acquisition phase (5 minutes). One hour later, one object was replaced with a novel object for the testing phase (5 minutes). Evenamide was administered 10 minutes before the testing phase (A). Evenamide normalized the reduced discrimination index in male and female MAM rats (B and D, respectively). Evenamide did not affect total exploration time in males (C), but a decrease was observed in MAM females treated with evenamide compared to saline-treated with vehicle (E).

3.5. Evenamide improves the social interaction of male MAM rats.

Evenamide was injected 10 min before the social approach test (Figure 5A). Evenamide treatment reversed the reduced sniffing time of male MAM rats without affecting the Saline group (Treatment, F_1,31=4.28, p=0.047; MAM, F_1,31=16.64, p=0.0003; Interaction, F_1,31=20.34, p=0.0001; ANOVA, Figure 5B). However, female MAM rats did not show a decrease in the sniffing time and evenamide did not change MAM and Saline responses (Treatment, p=0.2254; MAM, p=1459; Interaction, p=0.3793; ANOVA, Figure 5F). The time spent in the social chamber (male, Figure 5C; female, Figure 5G), % of entries in the social chamber (male Figure 5D; female, Figure 5H), and total crosses (male, Figure 5E; female, Figure 5I) were not changed by evenamide treatment or MAM condition in male and female rats. The time in the middle and toy chamber was also not affected by evenamide and MAM (ANOVA, Supplementary Figure 5).

Figure 5. Evenamide rescues social deficits in male MAM rats.

Evenamide was administered 10 minutes before the social approach test (A). Treatment with evenamide reversed the reduced sniffing time observed in male MAM rats (B), without affecting time spent in the social chamber (C), % of entries into the social chamber (D), or total number of crosses (E). In females, neither evenamide nor MAM treatment altered sniffing time (F), time in the social chamber (G), % of entries into the social chamber (H), or total number of crosses (I).

4. Discussion

Evenamide has emerged as a promising compound to treat schizophrenia, especially in patients who are treatment-resistant or poor responders to existing D2-based antipsychotic agents [31,40,41]. Unlike most current antipsychotic drugs that target D2 receptors, evenamide targets the site of proposed deficit in schizophrenia patients; i.e., hippocampal hyperactivity due to loss of parvalbumin interneuron inhibition [16,21,47]. However, the neurobiological mechanism underlying the efficacy of evenamide remain unclear. In this study, we investigated evenamide using the well-established MAM neurodevelopmental animal model of schizophrenia. Evenamide was effective in rescuing the hyperdopaminergic state, hippocampal hyperexcitability, and recognition memory deficits in MAM rats, with a sex-specific effect on social interaction observed only in MAM males. These findings suggest that evenamide may normalize aberrant dopaminergic activity by modulating hippocampal excitability, which could also contribute to the observed improvements in cognitive and social behaviors in a sex-specific manner.

The hyperdopaminergic state represents a hallmark of schizophrenia [5,8,29] which is associated with the occurrence of psychotic symptoms [48,49]. Current antipsychotic agents primarily exert their effects through D2 receptor antagonism and are effective in reducing positive symptoms [9,12,16,50]. Different preclinical models to study schizophrenia have exhibited this increased dopaminergic activity and are also important tools to evaluate the efficacy of antipsychotics in modulating DA activity [26,51–53]. Among them, the MAM model consistently demonstrates a hyperdopaminergic phenotype which is reversed by many antipsychotic drugs [6,9]. In the present study, both male and female MAM rats exhibited a significantly higher number of spontaneously active DA neurons per track which was normalized by evenamide administration. This domain in MAM rats is proposed to parallel the increased fluorodopa uptake observed in the associative striatum of patients with schizophrenia [29,54,55]. Interestingly, the increase in active DA neurons appeared to be region-specific and sex dependent. In males, it was predominantly observed in the lateral portion of the VTA, while in females, the increase was confined to the medial VTA with a trend in the lateral VTA. The medial-lateral subdivisions of the VTA differ in their projection targets, with the medial/central portions projecting to reward-related nucleus accumbens shell and lateral VTA to associative striatum and nucleus accumbens core [56,57]. The associative striatum is implicated in stimulus salience, whereas the nucleus accumbens shell is associated with affective regulation [1,57]. These differences in DA activity may reflect the clinical observation that schizophrenia symptoms differ between male and females. Males typically exhibit more severe negative symptoms, particularly those related to social withdrawal, whereas females often present with more pronounced mood disturbance and affective symptoms [58–61]. In fact, our behavioral data demonstrated that only male MAM rats displayed a social interaction deficit, as reflected by the reduced sniffing time towards a novel, younger, sex-matched conspecific, which is consistent with the clinical picture. Despite these sex-specific differences in dopaminergic activity and social response in the MAM rats, systemic administration of evenamide was able to reverse all observed dysregulations. Evenamide even decreased the number of DA neurons per track in the lateral portion of VTA of females and central portion in males in MAM rats which could reflect evenamide’s ability to modulate different schizophrenia symptom categories. Although the sex-specific dysregulation of VTA subregions may be associated with behavioral outcomes, these are correlational observations and should be interpreted with caution. Differences within the VTA may not directly correspond to specific behaviors, and future studies are needed to evaluate the role of distinct VTA projections using more direct behavioral approaches.

Analysis of VTA activity across time indicates that in male MAM rats, there was a persistent decrease in the number of spontaneously active DA neurons per track throughout the three-hour recording period. In contrast, a significant reduction in females was observed only during the final hour of recording. However, female MAM rats did not show a consistent pattern of increased dopamine neuron population activity across time points, which may represent a limitation in interpreting the effects of evenamide over time in females. This variability may be due to differences in sensitivity to anesthesia and/or the anterior-posterior segregation of dopamine activity within the VTA, as the electrode placement progresses posteriorly over time according to the grid-pattern used in the VTA recording experiment. In addition, the lack of an evident effect of evenamide in MAM females may reflect sex differences in either the pharmacokinetics of evenamide or plasticity changes within the circuits regulating VTA dopaminergic activity. A previous report indicates that evenamide exhibits rapid kinetics in humans, with a Tmax of 1.5–2 hours. Doses ranging from 1 to 30 mg/kg produce plasma concentrations comparable to those observed in rats 15 minutes after a 1 mg/kg injection; a dose that significantly attenuates PCP-induced social interaction deficits in rats [37]. Indeed, evenamide clinical trials often are associated with bid doses of 15 and 30 mg/kg [40,41] which supports that evenamide has a rapid pharmacokinetic profile. However, the sustained effects of evenamide on VTA dopaminergic activity in MAM male rats suggest that its impact may extend beyond its peak plasma concentration. This could indicate that evenamide may induce a circuit-level plasticity that results in long-lasting modulation of dopaminergic activity observed over the three-hour recording period.

Dopaminergic dysfunction in schizophrenia is proposed to result from dysregulation in upstream brain regions, particularly the limbic hippocampus [21,47], which we hypothesize to be the evenamide main site of action in downregulating dopaminergic activity. The limbic hippocampus is reported to be hyperactive in patients with schizophrenia [20,22], a state that arises due to a loss of parvalbumin-expressing interneurons [23,24]. The vHipp is homologous with the anterior limbic hippocampus in humans [62], and regulates VTA dopaminergic activity via nucleus accumbens and ventral pallidum pathway [29]. Consistent with previous findings [25], pyramidal neurons in the vHipp of MAM rats exhibited a higher firing rate compared to Saline rats; a pattern observed in both male and female animals. Evenamide, a voltage-gated sodium channel blocker, has been shown to attenuate the excitability of cortical neurons with a selective impact on neurons showing elevated levels of activity [32]. In fact, evenamide was able to reduce the firing rate of pyramidal neurons in the vHipp of male and female MAM rats. This suggests that evenamide selectively decreased the heightened excitability of vHipp neurons without altering the overall number of active neurons. Notably, the decrease in firing rate appeared more pronounced during the first hour of recording in MAM females, while in males, the effect was less time-dependent and likely reflected a cumulative reduction over the entire three-hour period.

The site of action of evenamide within the vHipp was confirmed by infusing evenamide locally in the vHipp and recording VTA DA neuron activity. Similar to systemic administration, local vHipp infusion reversed the increased number of spontaneously active DA neurons per track in male and female MAM rats. This aligns with the hypothesis that evenamide may normalize VTA DA activity by reducing vHipp hyperexcitability. Interestingly, the mediolateral distribution of VTA DA neuron activity after local vHipp infusion did not fully align with the pattern observed after systemic evenamide administration. In males, no evident data across mediolateral portions of VTA was observed in response to vHipp local evenamide injection in MAM rats. The absence of effect across the mediolateral VTA axis in this cohort of MAM rats may reflect variability between experimental groups or minor environmental differences during neurodevelopment. In addition, the difference between systemic and vHipp local evenamide effects may indicate that systemically administered evenamide may exert additional actions via other brain areas, e.g. the prefrontal cortex [63,64], which is also involved in regulating VTA dopaminergic activity. Despite these differences, the temporal analysis of VTA response to local vHipp infusion of evenamide in males was consistent with those observed following systemic administration, in which there is a significant reduction in the number of active DA neurons during the first and last hours of recording. In females, only a trend toward reduction was observed during the final hour of recording in MAM rats. These findings suggest that evenamide may induce its therapeutic effects by modulating vHipp activity to normalize the elevated glutamatergic vHipp outputs to the VTA in MAM rats.

The ability of evenamide to regulate vHipp excitability could also underlie its effects on improving behavioral deficits in MAM rats, as the vHipp has connectivity with crucial brain regions involved in negative and cognitive symptoms, such as amygdala and prefrontal cortex [65–67]. One of cognition aspect measured in the present study was recognition memory using the novel object recognition test, in which both male and female MAM rats exhibited reduced discrimination indexes. This is an indicative of impaired recognition memory. Systemic administration of evenamide restored the discrimination index to levels comparable to saline rats. This behavioral approach could offer some insights into working memory deficits commonly observed in patients with schizophrenia [68,69]. Evenamide may improve recognition memory by normalizing vHipp hyperexcitability and its subsequent impact on prefrontal cortex function; a region critically involved in cognitive control [70]. However, it is also possible that evenamide directly modulates prefrontal cortex activity by regulating the excitability of pyramidal neurons. The recognition memory improvement induced by evenamide in the preclinical MAM model may indicate that it may also enhance cognitive function in patients with schizophrenia and ultimately lead to a better functional outcome [18,71]. Current D2-based antipsychotic agents do not effectively address cognitive symptoms [14], which limits their overall efficacy and produces a significant functional burden on patients [72]. Therefore, evenamide would offer advantages over existing antipsychotic drugs by targeting both positive symptoms and cognitive deficits. However, future evenamide studies should extend the investigation to other cognitive domains, such as working memory and attention. It would provide a better understanding of the potential cognitive benefits of evenamide in the context of schizophrenia.

Another behavioral test used in this study was the social approach test, which assesses social interaction. Social withdrawal represents an important negative symptom that significantly impacts the functional outcomes of patients with schizophrenia and is not effectively treated by D2-based antipsychotic drugs [15]. Our data showed that only male MAM rats exhibited reduced sniffing time, indicative of impaired social interaction, and that evenamide was able to rescue this social deficit. In contrast, female MAM rats did not display a reduction in social behavior. As mentioned previously, the symptom profile in schizophrenia was found to differ between sexes, with men often expressing more social-related negative symptoms than women [58–61]. This pattern is consistent with our findings in the social approach test in MAM males. Therefore, evenamide was effective in reversing social behavior deficits in male MAM rats without affecting social response in saline and MAM females. This suggests that evenamide could effectively treat the broad and sex-specific domains of behavioral dysregulation. In addition, the improvement in social behavior in male MAM rats may indicate that evenamide could be modulating vHipp projections to the amygdala, a key area involved in emotional regulation [67,73,74]. In another preclinical model of schizophrenia, peripubertal stress also induces a hyperdopaminergic state in adult rats that is associated with amygdala hyperexcitability in males [75]. Interestingly, this amygdala involvement appears to be absent in females [75] which may explain the lack of social deficits in MAM females. However, the specific contribution of the amygdala in female MAM rats remains unclear.

While evenamide is thought to preferentially affect abnormally hyperactive glutamatergic neurons, anatomical specificity within vHipp is unlikely to be absolute. Other brain regions with high glutamatergic excitability, such as the prefrontal cortex and amygdala, may also be influenced by evenamide in the context of dysfunctional transmission. Although the present study focused on the vHipp, evenamide would likely be acting directly in these regions, which may contribute to the normalization of cognitive and social behaviors. For example, it has been reported previously that the basolateral amygdala exhibits increased firing rates in MAM rats [76], and that the prefrontal cortex displays an excitatory–inhibitory imbalance due to interneuron disruption which likely result in a hyperexcitable state [77]. In both cases, evenamide may act to normalize this dysfunctional excitability and contribute to the behavioral rescue associated with cognitive and negative symptoms of schizophrenia. Moreover, evenamide may also influence connectivity between regions and improve the top-down control from the prefrontal cortex to subcortical structures such as the thalamus and striatum [78–80]. Taken together, evenamide may exert its therapeutic effects through a combination of actions across multiple brain regions along vHipp.

These findings suggest that evenamide has high therapeutic potential for treating multiple symptom domains of schizophrenia in both males and females. Notably, evenamide rescued the hyperdopaminergic state, hippocampal hyperexcitability, and recognition memory deficits in both sexes, while the rescue of social deficits was sex-specific, occurring only in males. Evenamide arises as a promising agent capable of acting at the site of deficit in schizophrenia by reducing hippocampal hyperexcitability. This represents a significant advancement in treatment, as evenamide can downregulate the hyperdopaminergic state without producing D2 blockade-related side effects while also improving behavioral deficits that are not properly treated by D2 blocking antipsychotic agents. Therefore, this approach would potentially increase the functional abilities and overall outcomes of patients with schizophrenia, offering a novel strategy for managing the disorder.