AUT00206, a novel and selective Kv3.1/3.2 channel modulator, restores cognitive dysfunction and negative symptoms in an animal model for schizophrenia symptomatology Authors: Marianne Leger, Daniela Cadinu, Jennifer Fletcher, Agostino Marasco, Giuseppe Alvaro, Daniel Umbricht, Charles Large, Ben Grayson, Joanna C Neill, Michael K Harte Published: 2026 ABSTRACT AbstractBackground:Voltage-gated Kv3.1 and KV3.2 potassium channels are mainly located on parvalbumin (PV) containing interneurons, where they play a key role in synchronising the coordinated firing of pyramidal neurons and regulating cognitive function.Aims:We aim to explore the efficacy of acute treatment with AUT00206 (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione), a novel and selective positive modulator of Kv3.1/3.2 channels, to improve cognitive and social behaviour deficits in our validated animal model relevant to schizophrenia.Methods:Female Lister Hooded rats were treated with phencyclidine for 7 days, followed by a 7-day washout (scPCP). The efficacy of AUT00206 was tested in the novel object recognition (NOR), reversal learning (RL) and social interaction (SI) paradigms.Results:The cognitive and social behaviour deficits induced by scPCP were significantly attenuated by AUT00206 (10 and 30 mg/kg) in the three behavioural tasks. These data demonstrate, for the first time, the efficacy of a novel Kv3.1/3.2 channel modulator, AUT00206, in two cognitive domains (short-term recognition memory and an aspect of cognitive flexibility) and an aspect of negative symptoms in a validated animal model of schizophrenia symptomatology.Conclusions:Modulation of Kv3.1/3.2 channels on PV interneurons could be an important novel approach for the treatment of schizophrenia.Keywords: Kv3.1/2 channel modulator, AUT00206, cognition, novel object recognition, reversal learning, social interaction, rat, schizophrenia, subchronic phencyclidine model Introduction Schizophrenia is a psychiatric disorder affecting approximately 1% of the world’s population (Bromet and Fennig, 1999). This chronic disease is characterised by positive (hallucinations, delusions), negative (social withdrawal, lack of motivation) and cognitive symptoms (working memory, attention, executive function) (Tandon et al., 2009). While positive symptoms are reasonably well controlled by current medication, cognitive and negative symptoms remain poorly treated, demonstrating the urgent need to develop new and improved therapeutic strategies (Stępnicki et al., 2018). There is accumulating evidence that cognitive disturbances in schizophrenia are related to the dysfunction of parvalbumin (PV) containing interneurons, a subtype of gamma-aminobutyric acid (GABA) interneurons (Lewis, 2014). The microcircuits between the fast-firing PV interneurons and pyramidal cells are responsible for gamma oscillations, which are considered critical for cognitive function (for a review, see Glausier and Lewis, 2017). Disturbances in these fast-spiking networks and the associated network desynchrony are thought to contribute to the pathophysiology of cognitive dysfunction in schizophrenia (see Dienel and Lewis, 2019). One of the mechanisms thought to impair PV interneuron function is N-methyl-D-aspartate receptor (NMDAR) hypofunction (Lewis, 2014; Nakazawa et al., 2012). This hypothesis originated from the observation that NMDAR antagonists induce schizophrenia-like symptoms in healthy individuals (Greifenstein et al., 1958; Luby et al., 1959), which was later replicated after subchronic administration of these antagonists in animal models (Abdul-Monim et al., 2007; see Reynolds and Neill, 2016). Acute administration of NMDAR antagonists, like phencyclidine, increases the activity of glutamatergic neurons due to a preferential reduction in the firing of fast-spiking inhibitory interneurons (Homayoun and Moghaddam, 2007). In the subchronic phencyclidine (scPCP) model, we and others have demonstrated that scPCP administration results in reduced PV density in brain regions involved in schizophrenia pathophysiology (Abdul-Monim et al., 2007; Amitai et al., 2012; Landreth et al., 2020). The involvement of PV interneurons in NMDAR hypofunction raises the possibility that modulation of NMDAR function and PV interneuron activity could contribute to the amelioration of NMDAR hypofunction-induced cognitive deficits. The voltage-gated Kv3.1 and Kv3.2 potassium channels are key contributors to the fast-firing phenotype of the PV interneurons. These channels are highly expressed by PV-containing interneurons, where they facilitate rapid membrane repolarisation and the firing of pyramidal neurons at gamma frequencies (Caillard et al., 2000; Rudy et al., 1999; Rudy and McBain, 2001; Vreugdenhil et al., 2003). A postmortem study of people with schizophrenia showed no change in Kv3.2, and a reduction in Kv3.1 channel, messenger ribonucleic acid (mRNA) expression in the dorsolateral prefrontal cortex, a brain region critical for cognitive function (Yanagi et al., 2014). Reduced Kv3.1 mRNA was also detected in the prefrontal cortex of scPCP-treated rats, a model for schizophrenia symptomatology in rats (Pratt et al., 2008). To our knowledge, Kv3.2 changes have not been specifically characterised in NMDAR antagonism models. Here, we hypothesise that the positive modulation of the Kv3.1/3.2 channels may provide another strategy to normalise the PV interneuron deficiency and alleviate the clinical symptoms of schizophrenia. We thus investigated the efficacy of AUT00206 (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione), a novel and selective Kv3.1 and Kv3.2 channel positive modulator, to reverse the cognitive and social behaviour deficits in our established scPCP rat model (Kaar et al., 2022, 2024; Neill et al., 2010). We assessed recognition memory and cognitive flexibility, two cognitive processes known to be affected in schizophrenia and other disorders, using novel object recognition (NOR) and operant reversal learning (RL) tasks. Social interaction (SI) was used to measure negative symptomatology, as scPCP-treated rats show reduced interaction with other rats (see Neill et al., 2016 for an evaluation of the antipsychotic cariprazine using this model; see Cadinu et al., 2018; and Neill et al., 2014 for reviews on the validity of this model for assessing the efficacy of novel drug targets for cognition and negative symptoms). Materials and methods Animals In total, 200 adult female Lister Hooded rats (237 ± 2 g; Charles River, UK) were used. The rats were housed in groups of five in individually ventilated cages with two levels (GR1800 Double-Decker Cage, Techniplast, UK) under a standard 12-hour light: dark cycle (lights on 07:00 h) in regulated temperature (21°C ± 2°C) and humidity (55% ± 5%). All animals had free access to food (Special Diet Services) and water, except when undergoing behavioural testing. Animals used for RL experiments were placed under food restriction starting 1 week before the beginning of the training period. Food was restricted to maintain a minimum of 90% of free-feeding body weight (approximately 12 g rat chow/rat/day). Food restriction continued throughout training and RL testing. Behavioural testing was performed during the light cycle. Experiments were conducted in accordance with the UK Animals (Scientific Procedures) Act 1986 and approved by the University of Manchester Animal Welfare Ethical Review Body (AWERB). Our recent publication explained our use of female rats and described the three tests in more detail (Grayson et al., 2025). Drug administration Subchronic treatments Rats were treated with either phencyclidine (PCP) (Sigma-Aldrich; 2 mg/kg) or vehicle (VEH) solution (0.9% saline) via the intraperitoneal (i.p.) route twice daily for 7 days, followed by a 7-day washout as previously described (see Neill et al., 2016 as an example). Acute treatments AUT00206 was dosed via the oral route (p.o.) at doses of 0, 3, 10 and 30 mg/kg (dose volume 5 ml/kg) (n = 10 per group per behaviour). AUT00206 was prepared in 50% Gelucire 44/14, 30% PEG400 and 20% distilled water and administered 90 minutes prior to testing. Dose routes and ranges were selected from pharmacokinetic studies conducted by Autifony. Behavioural experiments Novel object recognition test The NOR test was performed as previously described (Grayson et al., 2007). Briefly, the animals (n = 50) were habituated in cage groups to the empty test arena the day before the test. The arena has black Plexiglas® walls and a white floor divided into nine squares (52 cm L; 52 cm W; 31 cm H). On the test day, rats underwent an acquisition trial for 3 minutes, during which they were placed in the testing arena containing two identical objects for free exploration. The animals were placed in a holding cage for a 1-minute inter-trial interval (ITI). Rats were returned to the test arena for the 3-minute retention trial with one duplicate familiar object and one novel object. The arena was cleaned with 70% ethanol between rats to remove any olfactory cues. Behaviour in all trials was recorded and scored by an experimenter blinded to treatment groups. The object exploration time (the time animals spent licking, sniffing, or touching the object but not including time spent standing, sitting on or leaning against the object) was recorded for the familiar and novel objects in the acquisition and retention trials. Locomotor activity (measured by the number of lines crossed in both trials) and discrimination index (DI) (calculated as the difference in time spent exploring the novel and the familiar object divided by the total time spent exploring both objects) were also collected. Any animal that failed to explore one or both objects (<0 seconds) in either trial was excluded from the final analysis. In this study, no animals were excluded from analysis. RL test Rats (n = 50) were trained to perform an operant RL task as previously described (Abdul-Monim et al., 2003; Neill et al., 2016). Following habituation to the operant chambers, rats were trained to respond to food on a Fixed Ratio 1 schedule of reinforcement with both levers active for 30 minutes. When responding stabilised (128 lever presses in 30 minutes), rats were trained to press the active lever (left or right) for food delivery in a 20-minute training session. The side of the active lever varied daily using a pseudorandom Gellerman schedule. This phase of training lasted approximately 2 weeks. Next, rats were trained to respond to food according to the presence or absence of a visual cue (Stimulus LED light above the lever). The session was terminated following 128 lever presses (approximately 30 minutes). Rats were trained once daily until they reached a criterion of ⩾90% correct responses on three consecutive days (generally achieved within 2 weeks), after which rats were trained until they reached the same criterion on the opposite contingency (light cue position relative to the active lever). Three animals failed to reach the criterion, so they did not proceed to the next stage of training. Each RL task session was preceded by a 30-minute operant training session to ensure a stable response. In the first 5 minutes of the initial phase of the RL task session, the contingency was the same as for the operant training session. This was followed by a 2-minute time-out period, signalled by the house light being turned off, that is, rats were given a signal that the rule was about to be reversed. The contingency was reversed for the subsequent 5-minute reversal phase. Animals undertook 4-6 RL task sessions before the beginning of the drug studies to ensure that they reached a stable level of performance. Once all animals had stabilised, the cohort was dosed with vehicle or PCP and given appropriate washouts. During the drug study, after the pre-treatment time, animals underwent one RL session: that is, an initial phase of 5 minutes, a time-out for 2 minutes, then a reversal phase of 5 minutes. Social interaction test The SI test was performed as previously described (Neill et al., 2016). Briefly, the test rat (n = 50) was placed in the arena with a weight-matched unfamiliar conspecific rat (n = 50) together with an unfamiliar object (e.g. an unopened drink can) and recorded for a 10-min period. The time spent sniffing the conspecific and the object was recorded and scored by an experimenter blinded to treatment groups using behavioural scoring software (Hindsight, Scientific Programming Services, Wokingham, UK). Locomotor activity was recorded by counting the total number of lines crossed, as in the NOR. Experimental design For each study, there was a subchronic vehicle (scVeh) group (pretreated with subchronic saline followed by vehicle treatment), a scPCP group (pretreated with scPCP followed by vehicle treatment) and drug dose-response groups (pretreated with scPCP followed by AUT00206 treatment) (Figure 1). Experimental procedures and behavioural testing. Female Lister Hooded rats were dosed with either scPCP (2 mg/kg, i.p.) or 0.9% saline twice daily for 7 days, followed by at least a 7-day washout. Rats were then acutely treated with AUT00206 (3, 10, and 30 mg/kg; p.o.) 90 minutes prior to behavioural testing. AUT00206: 5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione; scPCP: subchronic phencyclidine; i.p.: intra-peritoneal; p.o.: per os. Statistical analysis Data were assessed for equal variance and normality of residuals. NOR data were analysed by a two-way repeated measures analysis of variance (two-way repeated measures analysis of variance (RM ANOVA)) (where exploration is the within-subjects factor and treatment was the between-subjects factor) followed by planned multiple comparisons with Šídák’s correction. DI was analysed with a one-way ANOVA, followed by Dunnett’s multiple comparison test, where all treatment groups were compared to scPCP. Total object exploration for each trial and the number of line crossings were used to assess locomotor activity. A one-way ANOVA was used, followed by Tukey’s test. A repeated measures general linear model (RM GLM) was used for the RL data. Response type (correct or incorrect) and the phase (initial and reversal) were matched factors, and the treatment group was a between-subjects factor. In this analysis, the number of lever presses and their interaction with treatment for both phases were compared to ensure consistent engagement in the task. Next, the interaction between response type (correct or incorrect), phase and treatment was investigated to assess whether correct performance was affected by phase or treatment. The residuals from the RM GLM analysis were evaluated for normality to ensure the correct model fit. In the event of a significant RM GLM interaction, multiple comparison testing with Šídák’s correction was used. To compare between treatment groups, the percentage of correct responses in the reversal phase was compared using a one-way ANOVA, followed by Dunnett’s multiple comparison testing, where all groups were compared to the scPCP control. SI data were analysed by a one-way ANOVA or a Kruskal–Wallis test to identify the main effect of drug treatment on the measured exploratory parameters. These analyses were followed by Dunnett’s or Dunn’s multiple comparison tests as appropriate. In both comparisons, all groups were compared to the scPCP control. The number of line crossings in the SI test was analysed using the same parameters stated in the NOR task. Statistical significance was assumed when p < 0.05. Data were analysed and visualised in GraphPad Prism (Version 10) with the exception of the RM GLM, which was carried out in SPSS (Version 29). Results Investigating the efficacy of AUT00206 to reverse scPCP-induced short-term recognition memory deficits in the NOR test We first assessed AUT00206 in the NOR test for recognition memory. In the acquisition trial, a two-way RM ANOVA found no significant effect of the object (F(1, 45) = 0.000, p > 0.999), treatment (F(4, 45) = 1.441, p = 0.236) or a treatment * object interaction (F(4, 45) = 0.636, p = 0.639), showing that all groups, regardless of treatment, explored the left and right object equally. In the retention trial, there was a significant effect of the object (F(1, 45) = 24.760, p < 0.001), no effect of treatment (F(4, 45) = 0.557, p = 0.695) and a near-significant effect of the treatment*object interaction (F(4, 45) = 2.428, p = 0.062). Planned multiple comparison testing revealed a significant difference between the time scVeh-treated animals spent at the familiar and novel objects (p = 0.002), indicating intact recognition memory. The scPCP-treated animals did not prefer the novel object (p > 0.999), suggesting that the scPCP treatment induces a NOR deficit. The scPCP-induced deficit was maintained following treatment with 3 mg/kg AUT00206 with these animals unable to differentiate between objects (p = 0.849). However, in the higher AUT00206 doses, animals were able to differentiate between objects (10 mg/kg: p = 0.019; 30 mg/kg: p = 0.017) (Figure 2(a)). The effect of acute doses of AUT00206 on novel object recognition memory. Rats were treated with PCP (2 mg/kg, i.p., twice daily for 7 days, followed by 7 days washout). Exploration times for acquisition and retention phases of the NOR test (a) and DI (b) following acute treatment with AUT00206 (AUT; 3, 10 and 30 mg/kg; p.o.). For (a), data were analysed with a repeated-measures two-way ANOVA followed by planned multiple comparisons with Šídák’s correction. δ = significant difference between time spent at familiar and novel objects, δ p < 0.050, δδ p < 0.010, δδδ p < 0.001. For (b), DI was analysed by one-way ANOVA followed by Dunnett’s multiple comparison test. * = significantly different from scVeh, *p < 0.050. # = significantly different from scPCP group, #p < 0.050. Graphs show individual data points and the mean ± SEM, n = 10. ANOVA: analysis of variance; AUT00206: (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione); i.p.: intra-peritoneal; p.o.: per os; (RM) GLM: (repeated measure) general linear model ; (sc)PCP: (subchronic) phencyclidine; scVeh: subchronic vehicle; SEM: standard error of the mean. A one-way ANOVA with Welch’s correction revealed a significant effect of drug treatment on the DI (F(4.00, 22.15) = 4.499, p = 0.008). Compared to the scVeh cohort, scPCP-treated rats had a significantly reduced DI (p = 0.014). Treatment with AUT00206 at 10 mg/kg (p = 0.015) and 30 mg/kg (p = 0.033) significantly increased DI compared to scPCP-treated rats (Figure 2(b)). The total exploration times in the acquisition and retention trials were compared between treatments to ensure the drugs were not affecting object exploration. There was no effect of treatment in the acquisition (F(4, 45) = 1.144, p = 0.236) or the retention (F(4.00, 25.84) = 1.236, p = 0.325, Welch’s correction applied) trials. In addition, locomotor activity was assessed by the total number of line crossings. Here, there was no effect of treatment (F(4, 45) = 1.982, p = 0.113) (Table 1). Table 1.Object exploration and locomotor activity in the novel object recognition test.Total exploration time of objectsLocomotor activity (line crossings)Groups + TreatmentAcquisition trialRetention trialscVeh + Veh33.50 (1.34)28.70 (1.10)78.80 (2.38)scPCP + Veh31.30 (1.72)29.70 (0.56)75.90 (2.80)scPCP + AUT00206 3 mg/kg31.50 (1.43)28.00 (1.96)70.50 (5.05)scPCP + AUT00206 10 mg/kg32.40 (2.16)28.20 (0.89)74.00 (3.28)scPCP + AUT00206 30 mg/kg32.90 (1.97)27.30 (1.01)74.10 (2.95)Mean and (SEM) of total exploration time of two objects and line crossings in NOR following acute treatment with AUT00206 (3–30 mg/kg; p.o.) in scPCP-treated rats (2 mg/kg, i.p., twice daily for 7 days followed by 7 days washout). Data were analysed by one-way ANOVA followed by Tukey’s test (n = 10 per group).ANOVA: analysis of variance; AUT00206: (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione); i.p.: intra-peritoneal; NOR: Novel Object Recognition; p.o.: per os; (sc)PCP: (subchronic) phencyclidine; scVeh: subchronic vehicle; SEM: standard error of the mean. Object exploration and locomotor activity in the novel object recognition test. Mean and (SEM) of total exploration time of two objects and line crossings in NOR following acute treatment with AUT00206 (3–30 mg/kg; p.o.) in scPCP-treated rats (2 mg/kg, i.p., twice daily for 7 days followed by 7 days washout). Data were analysed by one-way ANOVA followed by Tukey’s test (n = 10 per group). ANOVA: analysis of variance; AUT00206: (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione); i.p.: intra-peritoneal; NOR: Novel Object Recognition; p.o.: per os; (sc)PCP: (subchronic) phencyclidine; scVeh: subchronic vehicle; SEM: standard error of the mean. Investigating the efficacy of AUT00206 to reverse scPCP-induced problem-solving deficits in the RL task The reversal learning paradigm comprises two phases: initial and reversal. There were no differences in the total number of lever presses in the initial phase compared to the reversal phase of the test (F(1, 42) = 0.132, p = 0.718), nor was there a significant interaction between treatment and phase lever presses (F(4, 42) = 1.336, p = 0.273), indicating maintained engagement in the task across both phases (Table 2). The number of correct lever presses in each phase was compared. Here, the phase of the task significantly predicted the number of correct lever presses (F(1, 42) = 32.514, p < 0.001), and this interacted with treatment (F(4, 42) = 3.849, p = 0.009). Multiple comparison testing was used to compare the correct and incorrect responses in each phase for every treatment. In the initial trial, all treatment groups pressed the correct lever significantly more than the incorrect lever (p < 0.001). In the reversal trial, the scVeh group continued to press the correct lever more (p < 0.001). scPCP administration reduced correct lever presses, which did not differ from the number of incorrect (p = 0.676). In rats treated with AUT00206, there was a significant difference in the number of correct lever presses in rats treated with 10 mg/kg (p = 0.017) and 30 mg/kg (p < 0.001) of AUT00206 but not at 3 mg/kg (p = 0.724) (Figure 3(a)). Table 2.Number of lever presses in the initial and reversal phase of the reversal learning test.Number of lever pressesGroups + TreatmentInitialReversalscVeh + Veh25.50 (0.45)25.10 (0.50)scPCP + Veh25.40 (0.43)25.40 (0.37)scPCP + AUT00206 3 mg/kg25.22 (0.36)25.78 (0.28)scPCP + AUT00206 10 mg/kg25.22 (0.28)25.56 (0.18)scPCP + AUT00206 30 mg/kg25.56 (0.24)24.67 (0.33)Mean and (SEM) of lever presses in the RL task following acute treatment with AUT00206 (3–30 mg/kg; p.o.) in scPCP-treated rats (2 mg/kg, i.p., twice daily for 7 days followed by 7 days washout). Data were analysed by an RM GLM followed by Šídák’s test (n = 9–10 per group).ANOVA: analysis of variance; AUT00206: (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione); i.p.: intra-peritoneal; (RM) GLM: (repeated measure) general linear model; (sc)PCP: (subchronic) phencyclidine; scVeh: subchronic vehicle; SEM: standard error of the mean. Number of lever presses in the initial and reversal phase of the reversal learning test. Mean and (SEM) of lever presses in the RL task following acute treatment with AUT00206 (3–30 mg/kg; p.o.) in scPCP-treated rats (2 mg/kg, i.p., twice daily for 7 days followed by 7 days washout). Data were analysed by an RM GLM followed by Šídák’s test (n = 9–10 per group). ANOVA: analysis of variance; AUT00206: (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione); i.p.: intra-peritoneal; (RM) GLM: (repeated measure) general linear model; (sc)PCP: (subchronic) phencyclidine; scVeh: subchronic vehicle; SEM: standard error of the mean. The effect of acute doses of AUT00206 on reversal learning. Rats were treated with scPCP (2 mg/kg, i.p., twice daily for 7 days, followed by 7 days washout). For (a), graphs show the total number of levers pressed (light bar) and correct lever presses (overlayed dark bars and individual points) in the initial and reversal phase following acute treatment with AUT00206 (AUT; 3, 10 and 30 mg/kg; p.o.). Data were analysed with an RM GLM followed by multiple comparisons with Šídák’s correction. δ = correct lever was used significantly more than incorrect, δ p < 0.050, δδδ p < 0.001. For (b), the percentage of correct responses in the reversal phase is shown. For (b), data were analysed with a one-way ANOVA followed by Dunnett’s multiple comparison test, * = significantly different from scVeh, *p < 0.050. # = significantly different from scPCP group, #p < 0.050. Graphs show individual data points and the mean ± SEM, n = 9–10. ANOVA: analysis of variance; AUT00206: (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione); i.p.: intra-peritoneal; p.o.: per os; (RM) GLM: (repeated measure) general linear model ; (sc)PCP: (subchronic) phencyclidine ; scVeh: subchronic vehicle; SEM: standard error of the mean. The percentage of correct responses in the reversal phase was compared between treatment groups. There was a significant effect of treatment on the correct responses (F(4, 42) = 6.721, p < 0.001). Multiple comparison testing comparing all groups to the scPCP-control confirmed a significant deficit when compared to the scVeh control (p < 0.001). This deficit was attenuated by the 30 mg/kg group only (p = 0.003), whereas no significant effects were observed in the other treatment groups (Figure 3(b)). Investigating the efficacy of AUT00206 to reverse scPCP-induced social behaviour deficits in the SI test In this study, scPCP treatment produced a significant impairment in social behaviour. Statistical analysis revealed a significant drug treatment effect on social sniffing behaviour (H(4) = 16.90, p = 0.002). Multiple comparison testing showed reduced time spent sniffing the conspecific in the scPCP-treated rats compared to scVeh (p = 0.008). This deficit was significantly attenuated in rats treated with AUT00206 at 10 mg/kg (p = 0.022) but not 3 mg/kg (p > 0.999) and 30 mg/kg (p = 0.070), though the latter neared significance (Figure 4(a)). There was no treatment effect on object exploration (F(4, 45) = 0.617, p = 0.653) (Figure 4(b)) or line crossing (F(4, 45) = 0.803, p = 0.503) (Table 3). The effect of acute doses of AUT00206 on the social interaction test. Rats were treated with scPCP (2 mg/kg, i.p., twice daily for 7 days, followed by 7 days washout). Graphs show the time spent exploring the conspecific (a) and the object (b) following acute treatment with AUT00206 (AUT; 3, 10 and 30 mg/kg; p.o.). For (a), data were analysed with a Kruskal–Wallis test followed by Dunn’s multiple comparisons test. For (b), data were analysed with a one-way ANOVA using Dunnett’s multiple comparison test. * = significantly different from scVeh, *p < 0.050, **p < 0.010. # = significantly different from scPCP group, #p < 0.050. Graphs show individual data points and the mean ± SEM, n = 10. ANOVA: analysis of variance; AUT00206: (5,5-dimethyl-3-[2-(7-methylspiro[2H-benzofuran-3,1'-me]-4-yl)oxypyrimidin-5-yl]imidazolidine-2,4-dione); i.p.: intra-peritoneal; p.o.: per os; (RM) GLM: (repeated measure) general linear model ; (sc)PCP: (subchronic) phencyclidine; scVeh: subchronic vehicle; SEM: standard error of the mean. Table 3.Locomotor activity in the social interaction test.Groups + TreatmentLocomotor activity (line crossings)scVeh + Veh202.2 (13.95)scPCP + Veh212.9 (14.00)scPCP + AUT00206 3 mg/kg196.3 (6.44)scPCP + AUT00206 10 mg/kg206.2 (10.10)scPCP + AUT00206 30 mg/kg187.6 (6.71)Mean and (SEM) of the number of line crossings in SI following acute treatment with AUT00206 (3–30 mg/kg; p.o.) in scPCP-treated rats (2 mg/kg, i.p. twice daily for 7 days followed by 7 days washout). Data were analysed by one-way ANOVA followed by Tukey’s multiple comparison test (n = 10 per group). Locomotor activity in the social interaction test. Mean and (SEM) of the number of line crossings in SI following acute treatment with AUT00206 (3–30 mg/kg; p.o.) in scPCP-treated rats (2 mg/kg, i.p. twice daily for 7 days followed by 7 days washout). Data were analysed by one-way ANOVA followed by Tukey’s multiple comparison test (n = 10 per group). Discussion As reported by several previous studies in our laboratory, the scPCP treatment regimen has been thoroughly validated for modelling specific aspects of cognitive and negative symptoms of schizophrenia (see Cadinu et al., 2018; Neill et al., 2016). We demonstrate that a novel molecule, AUT00206, a Kv3.1 and Kv3.2 channel positive modulator, attenuates scPCP-induced deficits in three paradigms relevant to the cognitive and negative symptoms of schizophrenia when administered acutely. AUT00206 alleviated behavioural deficits relevant to schizophrenia AUT00206, a modulator of Kv3.1 and Kv3.2 channels, ameliorated scPCP-induced recognition memory (NOR), executive function (RL) and SI deficits in adult female rats. The deficit in the NOR task was significantly attenuated by acute administration of AUT00206 at 10 and 30 mg/kg. Treatment with scPCP also induced a selective deficit in the reversal phase of the RL paradigm in rats, as we have shown previously (Idris et al., 2010; McLean et al., 2010), indicating impaired cognitive flexibility (Neill et al., 2010). As in our previous studies, scPCP treatment did not affect learning performance in the initial phase of the task but induced a robust deficit in the reversal phase. AUT00206 at 10 and 30 mg/kg significantly improved performance relative to the scPCP control groups. In the SI test, scPCP treatment typically reduces social behaviour with an untreated conspecific rat (Snigdha and Neill, 2008a, 2008b). In the present study, scPCP treatment reduced conspecific sniffing. AUT00206 attenuated these effects at 10 mg/kg and trended towards an improvement at the highest dose, 30 mg/kg. We measured non-specific outcomes in each test, which suggested the drug did not markedly induce hyperactivity or sedation. In each case, AUT00206 did not significantly alter task engagement (total object exploration time or number of lever presses) or locomotor activity; however, a small, non-significant reduction in locomotor activity was observed at the highest dose, which warrants further discussion. Kv3.1 and Kv3.2 channels are highly expressed in PV-positive interneurons, including within basal ganglia circuits, which play a key role in controlled movement (Hegeman et al., 2016; Kaczmarek and Zhang, 2017). Modulation of these circuits may influence motor pathways and, in turn, alter locomotor output. Therefore, the non-significant reduction in locomotion observed at higher doses of AUT00206 may reflect enhanced inhibitory tone in these movement circuits, although our data suggest this does not preclude the cognitive and social benefits at these doses. AUT00206 has been tested in vivo previously in female Lister hooded rats. AUT00206 was administered at 30 mg/kg in an auditory fear conditioning task before training and retrieval. When there was extended training (3 days) before retrieval, AUT00206 treatment resulted in decreased passive (freezing) and increased active (darting) fear responses (Stubbendorff et al., 2023). These data suggest that modulation of Kv3.1/3.2 channels may influence fear responses, potentially through effects on PV interneuron activity. However, further studies incorporating cell-type-specific approaches would be required to determine the contribution of PV-positive interneurons to the observed behavioural effects. A limited number of clinical imaging studies have assessed the effects of AUT00206 on frontal gamma power (Kaar et al., 2024), dopamine synthesis and capacity in striatal regions (Angelescu et al., 2022) and associated reward anticipation (Kaar et al., 2022). Despite small sample sizes, these preliminary studies demonstrated the promising capabilities of AUT00206. AUT00206 resulted in a trend towards reducing gamma power across the 28 days of the study (Kaar et al., 2024). AUT00206 improved the activity in the associative striatum during reward anticipation tasks, a trait associated with reduced Positive and Negative Syndrome Scale score at baseline (Kaar et al., 2022). The dopamine-related readouts from these studies were less clear, with no clear change in dopamine synthesis capacity (Angelescu et al., 2022) and no alterations in the ventral striatum during the reward anticipation task (Kaar et al., 2022). In the latter study, the authors suggested this may be due to antipsychotic use (Kaar et al., 2022). Targeting parvalbumin interneurons in schizophrenia Multiple studies support the perspective that inhibitory microcircuit abnormalities are a core clinical failure in schizophrenia (Gonzalez-Burgos et al., 2010; Lewis, 2014). A key player in microcircuit abnormalities is the impaired inhibitory PV-containing interneurons, as evidenced by postmortem studies (Lewis, 2014) and preclinical models (for a review, see Ruden et al., 2021). GABAergic inhibition from PV-containing cells projecting to neighbouring pyramidal neurons is essential for synchronising neural activity (Bartos et al., 2007; Mann and Paulsen, 2007). Dysfunction in their firing pattern contributes to impaired cortical gamma oscillations (Dienel and Lewis, 2019), a prominent physiological feature of schizophrenia (reviewed in Uhlhaas and Singer, 2010). It should be noted that the present study employed an acute dosing paradigm, which limits the ability to directly infer underlying molecular or cellular adaptations associated with AUT00206 treatment. Changes in interneuron function, including those involving PV levels or activity, may be more readily detectable following chronic administration. However, the extent to which such changes contribute to the behavioural effects observed following acute dosing remains unclear. As such, while modulation of Kv3.1/3.2 channels provides a plausible mechanism for influencing inhibitory circuit function, further studies incorporating chronic treatment designs, alongside electrophysiological or molecular approaches, will be required to determine the contribution of PV interneurons to the behavioural effects reported here and to establish whether these findings generalise across different preclinical models of schizophrenia. Limitations and future directions While the present study demonstrates that AUT00206 attenuates scPCP-induced behavioural deficits, several limitations should be considered. First, all experiments were conducted in female rats. Although this is consistent with previous work using this model, future studies should determine whether these findings are generalisable to males and indeed other preclinical models relevant to schizophrenia. Second, although AUT00206 improved behavioural performance following scPCP treatment, its effects were not extensively characterised in control (scVeh-treated) animals across all paradigms. As such, it remains unclear whether the observed effects are specific to the disease-like phenotype or reflect more general modulation of behaviour. It should be noted, however, that to characterise behavioural effects in scVeh animals, different behavioural tests would need to be selected, as the tests used here likely have a ceiling effect, with the scVeh/Veh animals already performing optimally. Conclusion A novel molecule for the treatment of schizophrenia, AUT00206, shows robust efficacy in our rodent model in the two cognitive domains and in ameliorating social behaviour deficits, an aspect of negative symptoms. The modulation of Kv3.1 and Kv3.2 channels thus represents a promising new approach for treating these unmet clinical needs in schizophrenia patients. Footnotes ORCID iDs: Ben Grayson https://orcid.org/0000-0001-6843-898X Joanna C. Neill https://orcid.org/0000-0002-2717-9739 Michael K. Harte https://orcid.org/0000-0002-4702-4785 Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: These studies were funded by Innovate UK (formerly Technology Strategy Board, TSB), Autifony and b-neuro at the University of Manchester. ML and DC were funded by a grant awarded to the University of Manchester from Autifony Therapeutics Ltd with Innovate UK (formerly TSB). The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. References