Middle Cerebral Artery Occlusion-induced Deficits in Fine Skilled Motor Movement and Its Recovery Following Combined Treatment with Sodium Valproate and CIMT Authors: Rini PG, Yoganarasimha Doreswamy, Laxmi TR Published: 2026 ABSTRACT AbstractBackgroundPreclinical studies have shown that sodium valproic acid (VPA) has neuroprotective properties following ischaemic injury. However, its compatibility as an adjuvant therapy with physical rehabilitation in ischaemic stroke recovery, with underlying neurophysiological mechanisms, has not been explored yet.PurposeThe present study has investigated the advantage of combination therapy over monotherapy in ameliorating ischaemic stroke-induced deficits in skilled fine motor movements with associated changes in local and interhemispheric neural connectivity.MethodsCerebral ischaemic stroke was induced by 60 min of middle cerebral artery occlusion in adult male Sprague–Dawley rats. VPA (300 mg/kg, subcutaneously, twice daily) was administered within 3 h after reperfusion and continued for 7 days, followed by constraint-induced movement therapy (CIMT) for another week (6 h/day). The ischaemic stroke phenotype was assessed based on neurological deficits, gait, forelimb use, rotarod stay time, reach success and infarct volume. Functional recovery, with associated changes in neural activities, was carried out by analysing the local field potential signals from the first reach success during the reach-to-grasp task on day 8 and day 15 post-stroke. Cortical oscillations were analysed at delta (0–4 Hz), theta (4–8 and 8–12 Hz) and beta (12–20 Hz) during the first reach success.ResultsCombination therapy after ischaemic stroke significantly improved motor and neurological functions compared to monotherapy and vehicle controls. The neural oscillations revealed a significant increase in theta power in the injured motor cortex after combination therapy when compared to that with monotherapy and control groups. The significant increase in the cross-correlation between hemispheres in the combination therapy group over monotherapy indicating an increased synchronised functional connectivity between the injured (left) and uninjured motor (right) cortices. Interestingly, VPA as monotherapy did not significantly reduce infarct volume compared to the vehicle controls. Furthermore, the therapeutic efficacy of monotherapies with VPA and early-onset CIMT (initiated immediately post-stroke) showed limited efficacy in promoting motor and neurological recovery.ConclusionIn conclusion, the study demonstrates that VPA combined with physical rehabilitative training (CIMT) enhances functional recovery by enhancing inter-hemispheric functional connectivity and represents a promising therapeutic strategy to alleviate persistent post-stroke motor deficits.Keywords: Ischaemic stroke, sodium valproate, constraint-induced movement therapy, neural plasticity, local field potentials Introduction Ischaemic stroke is classified as one of the leading causes of motor disability, with limited potential for spontaneous recovery.1, 2 Post-stroke recovery heavily relies on the effectiveness of rehabilitation. However, current rehabilitative training emphasises mechanical or task-oriented training to enhance experience-dependent plasticity. However, persistent motor deficits and several unmet clinical needs in patients emphasise the requirement for combination approaches harnessing the brain’s intrinsic capacity for reorganisation.3–6 The histone deacetylase (HDAC) inhibitor drug valproic acid (VPA) could be useful for treating different conditions, such as bipolar disorder,7, 8 migraines,9, 10 mood-related disorders,11, 12 epilepsy,13, 14 cancer15, 16 and motor-related disorders. 17 It has been used as an anxiolytic agent, 18 generalised anxiety disorder, 19 bipolar depression, 20 panic disorder 21 and post-traumatic stress disorder 22 due to its gamma-aminobutyric acid (GABA)-enhancing actions. Studies using in vitro and in vivo models have shown multimodal actions of VPA attenuating glutamate excitotoxicity, 23 anti-inflammatory responses, 24 HDAC inhibition, 25 increased acetylation, 26 GABA enhancement 27 and reducing microglial activation, 28 indicating VPA as a neuroprotective agent. Rehabilitative strategies such as constraint-induced movement therapy (CIMT) are considered one of the most effective physical therapies in clinical settings to improve motor disability. 29 Previously, we have shown that CIMT after 1 week of middle cerebral artery occlusion (MCAO) was beneficial in ameliorating ischaemic stroke-induced deficits in skilled fine motor movements and associated changes in structural plasticity in the penumbral motor cortex. 29 CIMT promotes contra-lesional oriented structural and hemispheric connectivity.30, 31 However, no studies have yet explored whether valproate enhances CIMT-induced motor and neurological recovery following ischaemic stroke. We hypothesised that administration of VPA within 3 h following ischaemia, followed by a week of treatment and subsequent rehabilitative training using CIMT for 7 days, would result in enhanced behavioural recovery compared with monotherapies. Methods Animals Male Sprague–Dawley rats (360–400 g) aged 2–3 months were procured from the Central Animal Research Facility, NIMHANS, Bengaluru. They were kept in a standard polypropylene cage (22.5 × 35.5 × 15 cm) at a temperature of 25 °C under light-controlled (12 h light:12 h dark cycle) conditions. They had access to ad libitum food and water. These rats were handled regularly before subjecting them to stroke surgery. The experimental procedure described in the present study has obtained the Institutional Animal Ethics Committee’s approval (Approval ID: AEC/75/498/NP). All efforts were made to avoid stress and to minimise animal suffering. All animal experiments were conducted in accordance with the ARRIVE 2.0 guidelines. Groups Rats were divided randomly into six groups (n = 6/group): (a) sham control (SC), (b) MCAO followed by 0.9% saline injection from day 0 to day 14 (ischaemic stroke control [ISC] + vehicle control [VC]), (c) MCAO with CIMT from day 1 to day 14 post-stroke (ISC + CIMT), (d) a combination of 0.9% saline injection from day 0 to day 7 post-stroke and CIMT from day 8 to day 14 post-stroke (ISC + VC + CIMT), (e) MCAO with VPA injection from day 0 to day 14 (ISC + VPA), (f) combinational treatment group was with MCAO, followed by VPA treatment from day 0 to day 7 post-stroke and CIMT from day 8 to day 14 post-stroke (ISC + VPA + CIMT). Different cohorts of animals were used to evaluate the role of combination therapy in motor recovery and associated electrophysiological changes, and to assess the effect of valproate on infarct volume. Study Design The study design and experimental timeline are explained schematically in Figure 1. A Study Design Depicting the Timeline of the Procedures Carried Out. Middle Cerebral Artery Occlusion and Treatments The ischaemic stroke in rats was induced by occluding the middle cerebral artery (MCA) as described previously. 32 In brief, a silicone rubber-coated nylon monofilament (Doccol Corporation, USA, diameter with coating 0.41 ± 0.02 mm, length 30 mm) was inserted through an incision in the internal carotid artery and advanced towards the cranial base until slight resistance was felt (approximately 23 ± 2 mm from the bifurcation of the common carotid artery). The MCAO lasted for 60 min, after which the filament was removed for reperfusion. The SC rats were subjected to vascular surgery without occluding the MCA. Sodium valproate (300 mg/kg, s.c., twice daily) was administered as previously described (Sigma–Aldrich, USA). 33 VPA was dissolved in 0.9% normal saline, while the vehicle used was 0.9% saline (w/v) (Abaris Healthcare Pvt. Ltd, India). CIMT was administered for 6 h (16.00–22.00 h) daily. 29 Both sodium valproate and 0.9% saline were administered subcutaneously twice daily to their respective groups, starting within 3 h after MCAO. Stereotaxic Surgery and Local Field Potentials Recording During Reach-to-grasp Task The coated stainless steel electrodes (125 µm) (A-M Systems, Inc., USA) were implanted stereotaxically (Stoelting Co., USA) under anaesthetised conditions (ketamine 80 mg/kg + xylazine 10 mg/kg) 34 in layer V of the right and the left motor cortex (anterior-posterior, +3.0 mm; mediolateral, ±2.0 mm; and dorsoventral, 1.5 mm). The ground and reference screws were placed subdurally in the right cerebellum and in the left cerebellum, respectively. The electrode was fixed onto the skull surface using dental acrylic cement. The opposite ends of the electrode were soldered to an integrated circuit socket, and to ensure proper fixation, the assembly was encased in dental acrylic above the rat’s head, preventing the electrodes from being exposed. Subsequently, the MCA was exposed and occluded for 60 min to induce transient focal ischaemic stroke. A 10% povidone-iodine solution was applied along the wound’s edges to prevent infection. Finally, the incision was carefully sutured using Vicryl 3.0 absorbable sutures. Meloxicam (10 mg/kg, Intas Pharmaceuticals Ltd, India) was given subcutaneously and 0.9% saline (Pharma Impex Pvt. Ltd, India) was injected to prevent dehydration following surgical blood loss. The rats were kept under observation in a heated chamber until they showed recovery in motor functions and began to eat and drink. Rats were allowed to recover from the surgery for a period of 7 days. The entire procedure was conducted under strict aseptic precautions. Care was taken to minimise pain or distress during and after stereotaxic surgery. The local field potential (LFP) was recorded during the reach-to-grasp task on days 8 and 15 post-stroke. Data Analysis In the present study, 73 rats were subjected to electrophysiological studies, and among them, 23 of 37 rats died from surgical trauma and MCAO; 14 of 37 rats had an infection, poor body weight gain, and reduced food intake. Rats with a high level of signal-to-noise ratio (n = 6/group) were considered for the data analysis. The AXXONET BESS software was used for the entire data acquisition process, which included signal amplification, digitisation and storage for subsequent analysis. The LFP signals were amplified with a gain of 1,000 and digitised at a sampling rate of 1,024 Hz. Recordings were done at a voltage scale of 7.0 µV/mm, filtered with 50 and 100 Hz notch filters, and a band-pass filter of 0.1–100 Hz. The behaviour of the animals was recorded using a video camera with a sampling rate of 30 frames per second through the AXXONET BESS software. Events were marked offline after the recording. The first reach success was considered and marked as an event for the subsequent data analysis. After the event marking, the raw data were exported and stored in *edf format for further analysis. The data acquisition for LFP recording was conducted using BESS software (Axxonet System Technologies, India). LFP data were extracted and processed using Spike2 software version 6.0 (Cambridge Electronic Design, UK). The *edf file was imported into Spike2 and converted to *smr file. Channel processing for the *smr file includes direct current removal and smoothing with a time constant of 0.0097 s. The data analysis was carried out using Spike 2 software. The oscillatory activities in the injured (left) and uninjured (right) motor cortex in layer V were analysed offline from the first reach success event for 5 s. To evaluate if delta-theta-beta oscillations were modulated during reach success, a power spectral density (PSD) and waveform correlation analysis of LFP signals from motor cortices (injured and uninjured) were performed. Power Spectral Analysis The power spectrum for reach success was analysed using a fast Fourier transform with the Hanning window, which gave the frequency range from 0 to 512 Hz in 512 bins with a sampling rate of 1,024 Hz. The frequency band was divided into delta (0–4 Hz), low theta (4–8 Hz), high theta (8–12 Hz) and beta (12–20 Hz). 34 The relative power is then computed as the ratio of the absolute power in each frequency as a percentage, P(f), of the maximum absolute power of the total LFP data, P(fmax). Waveform Correlation The cross-correlation was measured using a 5 s epoch during reach success in the reach-to-grasp task from the motor cortices. The peak cross-correlation was calculated in the time-series analysis. Histology The electrode placement in the target brain area was confirmed on day 16 post-stroke using the Nissl staining method after transcardial perfusion of the rat brain. Constraint-induced Movement Therapy CIMT was administered as described in our previous study. 29 Briefly, the cast was applied manually around the uninjured forelimb. The cast was prepared using plaster of Paris. All animals that received CIMT therapy were subjected to reach-to-grasp task training, promoting the use of the injured forelimb. Neurological Examination A neurological examination was conducted 1 day before, on day 8, and on day 15 post-MCAO, using the Garcia scoring method. 35 Cylinder Test A cylinder test was used to assess the spontaneous forelimb paw preference in a natural environment 36 and was conducted on days 8 and 15 post-MCAO in rats. The rat was placed in a plexiglass cylinder (34 × 21 × 40 cm) and allowed to explore the cylinder wall for 5 min, during which video recordings were made for offline analysis. The limb preference was measured as the ipsilateral bias index and calculated using the following formula: Ipsilateral bias index = i–c/i + b + c Where i represents the number of independent unimpaired forelimb contacts; c is the number of dependent impaired forelimb contacts; b is the number of both forelimbs in contact simultaneously. Rotarod Test The rotarod test was carried out to assess the balance and coordination in rats by using the Rotamex-5 system (CR-1 Rotamex System, Columbus Instruments, USA). All the rats were subjected to 5 days of training before the stereotaxic surgery and continued for 5 days after MCAO, up to day 14 post-MCAO. The test was carried out on days 8 and 15 post-stroke. The maximum speed at which the animal stayed for a given duration was measured through Rotamex software. The longest duration of stay in the rotating rung was considered a measure of balance and coordination in the rotating wheel. 37 Gait Analysis Rats were allowed to walk in a narrow passage on plain paper 1 m in length. Ink was applied to the ventral side of the forepaw to get a footprint. Then, stride length was measured by calculating the distance between two consecutive footprints on the same side. Reach-to-grasp Task The reach-to-grasp task was used to assess and train skilled fine forelimb movements. The behavioural procedure was carried out as described previously. 29 Both behavioural and electrophysiological studies were subjected to 2 weeks of training, receiving 30 pellets over 30 min (1 pellet/minute). To evaluate the effect of MCAO on skilled fine movements and their electrophysiological correlates, only the rats that successfully learned the reach-to-grasp task were selected for stereotaxic electrode implantation and subsequent MCAO surgery. Following surgery, training was resumed for an additional 2 weeks. Finally, the skill tests were performed on post-stroke days 8 and 15 for a duration of 10 min (1 pellet/minute) per session. During the entire training and testing, the videos were recorded for offline analysis. The percentage success was calculated for each day by using the following formula: The 2,3,5-Triphenyltetrazolium Chloride Staining The infarct size confirmation was carried out 7 days after MCAO by placing the rat brain in a brain matrix slicer to obtain 2 mm coronal sections and immediately exposed to 2,3,5-triphenyltetrazolium chloride (TTC) staining. The brain sections were then incubated in 1% buffered TTC for 30 min in a dark room at room temperature. The infarct volume was measured as described previously. 29 Statistical Analysis All data are expressed as mean ± standard error of the mean (SEM). One-way analysis of variance (ANOVA) followed by a multiple comparison test was used to assess the valproate effect on infarct size. To check the effect of different therapeutic strategies (monotherapy vs combinational therapy) on functional recovery after stroke, a repeated measures ANOVA followed by Tukey’s multiple comparison test was performed. The data were analysed using GraphPad Prism software (version 9). The criterion for statistical significance was set at p < .05. Results Infarct Size Figure 2a shows the TTC-stained images of SC and VC (ISC + VC) and VPA-treated (ISC + VPA) rat brain sections after 7 days of MCAO. One-way ANOVA followed by Tukey’s multiple comparison test revealed an increased infarct volume in the ISC + VC (p < .0001) and ISC + VPA (p = .0001) rats when compared to SC rats (F2,15 = 33.15, p < .0001) (Figure 2b). No significant difference in infarct volume was observed in the ISC + VPA group compared with the ISC + VC group. The Ischaemic Injury in the Rat After 7 Days of Middle Cerebral Artery Occlusion (MCAO) Surgery. (a) Representative Images of Rat Brain Section Showing the Infarct Size/Area. (b) Infarct Volume (mm 3 ) on Day 8 Post-stroke from Sham Control (SC), Ischaemic Stroke Control (ISC) + Vehicle Control (VC) and ISC + Valproic Acid (VPA) Rat Groups. Note: Data are expressed as mean ± standard error of the mean (SEM). One-way analysis of variance (ANOVA) followed by a post hoc test showed significant differences between groups. ****p < .0001 and ***p < .001 versus SC group (n = 6/group). Neurological Score As shown in Figure 3a, all ISC and rats treated with VPA and/or CIMT groups showed an overall deficit in Garcia score compared to the SC group after 1 and 2 weeks of MCAO. The two-way ANOVA with repeated measures analysis revealed a statistically significant effect of groups (F5,36 = 199.8, p < .0001) and groups × days (F5,36 = 19.35, p < .0001). Tukey’s multiple comparison test further revealed a significant effect on neurological functions, showing significantly reduced Garcia scores in ISC + VC (p < .0001), ISC + CIMT (p < .0001), ISC + VC + CIMT (p < .0001), ISC + VPA (p < .0001), and ISC + VPA + CIMT (p < .0001) when compared to the SC group on both day 8 and day 15 post-stroke. Compared with the ISC + VC, the ISC + VC + CIMT (p < .01), ISC + VPA (p < .05), and ISC + VPA + CIMT (p < .05) groups showed significant improvement on day 15 post-stroke. Behavioural Assessment on Day 8 and Day 15 After Middle Cerebral Artery Occlusion (MCAO) Surgery. (a) Garcia Score. (b) Gait Analysis by Measuring Stride Length. (c) Ipsilateral Bias Index to Assess Forelimb Preference. (d) Rotarod Stay Time for Gross Motor Movement Assessment: Pre- and Post-MCAO Day 8 and Day 15. Note: Data are represented as mean ± standard error of the mean (SEM). Two-way analysis of variance (ANOVA), followed by Sidak’s multiple comparison test, was performed (n = 6/group). ****p < .0001, **p < .01 versus sham control (SC); @@@@p < .0001, @@p < .01, @p < .05 versus ischaemic stroke control (ISC) + vehicle control (VC); ####p < .0001, ##p < .01, #p < .05 between day 8 and day 15. Gait The stride length was assessed in rats subjected to ischaemic stroke for the gait analysis. The two-way ANOVA with repeated measures analysis revealed a statistically significant effect of groups (F5,36 = 53.07, p < .0001) and groups × days (F5,36 = 6.494, p < .001). Sidak’s multiple comparison test further revealed a significant effect on stride length, showing reduced stride length in ISC + VC (p < .0001), ISC + CIMT (p < .0001), ISC + VC + CIMT (p < .0001), and ISC + VPA (p < .0001) when compared to the SC group on days 8 and 15 (Figure 2b). The ISC + VPA + CIMT group, on the other hand, showed a reduced stride length on day 8 (p < .0001) but not on day 15 when compared to the SC group. Comparison of post-stroke days 8 and 15 revealed a significant improvement in stride length in both the ISC + VC + CIMT (p < .0001) and ISC + VPA + CIMT (p < .0001) groups, but not with the other groups. The ISC + VC + CIMT (p < .0001) and ISC + VPA + CIMT (p < .0001) groups exhibited a significantly greater stride length on day 15 post-stroke compared with the ISC + VC group (Figure 3b). Forelimb Use The forelimb use asymmetry in rats subjected to ischaemic stroke was assessed using a cylinder test. The two-way ANOVA with repeated measures analysis revealed a statistically significant effect of groups (F5,36 = 34.25, p < .0001) and groups × days (F5,36 = 2.568, p < .05). Sidak’s multiple comparison test has further revealed a significant effect on the ipsilateral bias index, showing increased ipsilateral bias index score in ISC + VC (p < .0001), ISC + CIMT (p < .0001), ISC + VC + CIMT (p < .0001), and ISC + VPA (p < .01) when compared to the SC group on day 8. The ISC + VPA + CIMT group, on the other hand, showed an increased ipsilateral bias index on day 8 (p < .01) but not on day 15 when compared to the SC group. When comparisons were made between post-stroke days 8 and 15, there were significant differences found in ISC + VC (p < .05), but not with the other groups (Figure 3c). Gross Motor Movements The rotarod test was performed at baseline and on post-stroke days 8 and 15. The two-way ANOVA with repeated measures of analysis revealed a statistically significant effect on groups (F5,36 = 50.34, p < .0001) and groups × days (F10,72 = 24.06, p < .001). Tukey’s multiple comparison test further revealed a significantly reduced rotarod stay time on day 8 and day 15 in ISC + VC (p < .0001), ISC + CIMT (p < .0001), ISC + VC + CIMT (p < .0001), ISC + VPA (p < .01), and ISC + VPA + CIMT (p < .01) when compared to the SC group. When comparisons were made between post-stroke days 8 and 15, we found significant differences in rotarod stay time in SC (p < .0001), ISC + VC + CIMT (p < .01), and ISC + VPA + CIMT (p < .05), but with no differences in ISC + VC, ISC + VPA and ISC + CIMT groups. Compared with the ISC + VC group, the ISC + VC + CIMT (p < .01) and ISC + VPA + CIMT (p < .0001) groups showed significant improvement on day 15 post-stroke (Figure 3d). Fine Skilled Movements The two-way ANOVA with repeated measures analysis revealed a statistically significant effect of groups (F5,36 = 197.8, p < .0001) and groups × days (F5,36 = 6.655, p < .001). Tukey’s multiple comparison tests further revealed a significantly reduced reach success rate on day 8 and on day 15 in ISC + VC (p < .0001), ISC + CIMT (p < .0001), ISC + VC + CIMT (p < .0001), ISC + VPA (p < .0001), and ISC + VPA + CIMT (p < .0001) compared to SC groups. When compared to day 8, the reach success rate on day 15 was significantly increased in ISC + CIMT (p < .01), ISC + VC + CIMT (p < .001), ISC + VPA (p < .05), and ISC + VPA + CIMT (p < .0001), but with no differences in ISC + VC and SC groups. Additionally, we observed a significant improvement in skilled reaching in the ISC + CIMT (application of CIMT immediately after stroke) from day 8 to day 15 post-stroke (p < .01). This indicates that chronic treatment for 2 weeks with VPA and/or CIMT was beneficial in restoring skilled fine motor movements in ischaemic stroke rats (Figure 4). Reach-to-grasp Task to Assess Skilled Forelimb Movements. Reach Success Rate on Day 8 and on Day 15 Post-Middle Cerebral Artery Occlusion (MCAO) Surgery. Note: Data are represented as mean ± standard error of the mean (SEM). Two-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison test (n = 6/group). ****p < .0001 versus sham control (SC); ####p < .0001, ##p < .01, #p < .05 between day 8 and day 15. Electrophysiological Correlates of Fine Skilled Movements Power Spectral Density Histological verification of electrode placement was performed before the analysis of LFPs (Figure 5). Oscillatory activity in the motor cortex was classified into delta (0–4 Hz), low theta (4–8 Hz), high theta (8–12 Hz), beta (12–20 Hz) and gamma (20–40 Hz) frequency bands. The statistical analysis was carried out using two-way ANOVA with repeated measures analysis followed by Tukey’s multiple comparison tests. Histological Verification of Electrode Placement in Rat Motor Cortex. (a) Confirmation of Electrode Placement for Local Field Potential (LFP) Recording Showing Electrodes in Layer 5 of the Motor Cortex. (b) Representative Examples of LFP Recordings and Spectrogram from the Right and Left Motor Cortex During the Skilled Forelimb Movement from (b) Sham Control (SC) and (c) Ischaemic Stroke Control + Vehicle Control (ISC + VC) Rat Groups. The Event Marker Shows the Event Selected to Analyse the LFPs During the Specific Phase of the Reach-to-grasp Task. At the delta frequency band, there were no significant differences in relative PSD between groups and days in both the injured (Figure 6a) and the uninjured motor cortex (Figure 6e). Averaged Power Spectral Density (PSD) on Day 8 and Day 15 Following Middle Cerebral Artery Occlusion (MCAO) from the Injured (Left) and the Uninjured (Right) Motor Cortex. Relative Power (%) at (a) 0–4 Hz, (b) 4–8 Hz, (c) 8–12 Hz and (d) 12–20 Hz Frequency Band in the Injured Motor Cortex; (e) 0–4 Hz, (f) 4–8 Hz, (g) 8–12 Hz and (h) 12–20 Hz in the Uninjured Motor Cortex Following 7 and 14 Days of Treatment with Valproic Acid (VPA)+/Constraint-induced Movement Therapy (CIMT). Two-way Analysis of Variance (ANOVA) Followed by Tukey’s Multiple Comparison Tests Was Performed. Note: Data are represented as mean ± standard error of the mean (SEM). *p < .01 versus sham control (SC); @p < .05 versus ischaemic stroke control (ISC) + vehicle control (VC) and #p < .05 between day 8 and day 15 post-stroke (n = 6/group). The two-way ANOVA with repeated measures on the theta frequency band (4–8 Hz) revealed a significant effect of group (F5,30 = 2.802, p < .05) and post-stroke days versus group (F5,30 = 3.9476, p < .01) on day 8 post-stroke. Tukey’s multiple comparison test further revealed significantly increased theta power at 4–8 Hz in ISC + VPA + CIMT (p < .05) compared to the ISC + VC + CIMT group on day 8 post-stroke. There was also a significant increase in theta power on day 15 post-stroke in ISC + VPA + CIMT compared to the ISC + VC group (p < .003). The significant difference was found between day 8 and day 15 post-stroke in ISC + VC and ISC + VC + CIMT rat groups (Figure 6b). However, there were no differences observed in the uninjured motor cortex (Figure 6f). Tukey’s multiple comparison test revealed a significant difference in PSD at the high theta frequency band (8–12 Hz) on day 15 between ISC + VC and ISC + VPA + CIMT in the injured motor cortex (p < .05) (Figure 6c). Similarly, between day 8 and day 15, there were significant differences in the ISC + VC group (p < .05) in the high theta band. The significant differences were not observed in the uninjured motor cortex (Figure 6g). Two-way ANOVA was carried out to identify the impact of treatment with VPA and CIMT in the injured motor cortex and revealed that at the beta frequency band (12–20 Hz), there was a significant effect on groups (F5,30 = 4.198, p < .01). Tukey’s multiple comparison test has revealed a significant increase in ISC + VPA + CIMT when compared to SC (p < .01), ISC + VC (p < .01), and ISC + VC + CIMT (p < .01) on post-stroke day 15, but not on day 8 (Figure 6d). Similar observations were observed in the uninjured motor cortex (Figure 6h), wherein the ISC + VPA + CIMT group showed significantly increased PSD at 12–20 Hz (p < .05) on day 15 but not on day 8 when compared to ISC + VC and ISC + CIMT. Next, PSD analysis was carried out based on the area under the curve (AUC) during the reach-to-grasp task for day 8 and day 15 post-stroke in rats. The two-way ANOVA revealed a statistically significant effect of group (F5,30 = 5.016, p < .001). Sidak’s multiple comparison test revealed significantly increased PSD in the injured (left) motor cortex during the reach-to-grasp task (Figure 7a) on day 15 in the ISC + VPA + CIMT when compared to SC (p < .05), ISC + VC (p < .001), ISC + VC + CIMT (p < .05), and ISC + VPA (p < .05) groups. When compared to day 8, there were significant differences (p < .05) on day 15 in the ISC + VC rat groups. In the uninjured (right) motor cortex, there were no significant differences in PSD between groups and post-stroke days (Figure 7b). Averaged Power Spectral Density (PSD) Based on Area Under the Curve (AUC) on Day 8 and Day 15 Following Middle Cerebral Artery Occlusion (MCAO) from the (a) Injured (Left) and (b) the Uninjured (Right) Motor Cortex After 7–14 Days of Treatment with Valproic Acid (VPA)+/Constraint-induced Movement Therapy (CIMT). Two-way Analysis of Variance (ANOVA) Followed by Tukey’s Multiple Comparison Test. Note: Data are represented as mean ± standard error of the mean (SEM). *p < .05 versus sham control (SC); #p < .05 between d8 and d15; @@@p < .001 versus ischaemic stroke control (ISC) + vehicle control (VC). Interhemispheric Connectivity To investigate the effects of combinational therapy on interhemispheric functional connectivity, the waveform cross-correlation analysis was carried out on day 8 and day 15 post-stroke. As compared to the SC group, there was a significant effect of group (F5,30 = 11.01, p < .0001) on day 15, but not on day 8 post-stroke period. Sidak’s multiple comparison test revealed that the peak cross-correlation was significantly reduced more in ISC + VC (p < .0001), ISC + CIMT (p < .0001), ISC + VC + CIMT (p < .01), ISC + VPA (p < .01), and less in ISC + VPA + CIMT (p < .05) when compared to the SC group of rats. The significant differences in peak correlation level were not observed on day 8 post-stroke. The peak correlation between day 8 and day 15 was significantly greater in the SC rat group (p < .05) and not in other stroke groups (Figure 8). Waveform Correlation Analysis. Peak Correlation Level Indicates the Synchronisation Level Between the Left and the Right Motor Cortex. Data Are Expressed as Mean ± Standard Error of the Mean (SEM), Two-way Analysis of Variance (ANOVA) (F5,30 = 11.01, p < .0001) Followed by Sidak’s Multiple Comparison Test (n = 6/group). Note: Data are expressed as mean ± SEM. ***p < .001, **p < .01, *p < .05 versus sham control (SC); ##p < .01 between day 8 and day 15. Discussion Despite the growing interest in understanding the role that HDAC inhibitors play in the management of stroke rehabilitation, there are no studies yet available using combined treatment to ameliorate ischaemic stroke-induced changes in skilled fine motor movements. The hypothesis was that combined treatment with VPA and CIMT could positively influence interhemispheric communication and stroke outcomes in skilled fine motor movements. In the present study, we subjected stroke rats to either monotherapy or combination therapy and examined the electrophysiological correlates of skilled fine motor movement success. In particular, VPA administration combined with physical rehabilitative training (CIMT) significantly enhanced functional recovery following sequential combination treatment over monotherapy and control groups. The electrophysiological and behavioural studies provide substantial evidence for triggered neuroplasticity in the injured motor cortex and thus re-establishing the functional connectivity between hemispheres to reduce long-term motor deficits induced by MCAO. Behavioural Recovery in Response to Combinational Treatment with Valproic Acid and Constraint-induced Movement Therapy The neuroprotective effect of HDACi is mainly due to its role in epigenetic modulation of gene expression. 38 The behavioural response to VPA is possibly mediated by inhibiting the inflammatory and oxidative pathways. 25 It was shown previously that the delayed application of CIMT for a week has ameliorated the stroke-induced deficits in fine motor movements and increased the quality of movements by increasing dendritic arborisation in the pyramidal neurons of layer III and layer V motor cortex. 29 Even though CIMT as a monotherapy is effective in the chronic stage of ischaemic stroke to overcome ‘learned non-use’, initiating it immediately after the stroke onset as a monotherapy can have a detrimental effect on true functional recovery, as shown in the previous studies. 39 The present study further substantiated these findings, showing that early treatment with VPA was more effective at reducing the neurological deficits (ipsilateral bias for forelimb use) than early implementation of CIMT within 3 h after MCAO. The delayed application of CIMT successfully improved both neurological and behavioural deficits caused by MCAO, which is in line with our previous findings. 29 The success of VPA in post-stroke recovery is possibly due to its neuroprotective properties, 40 protecting the vulnerable brain tissue from injury during the post-stroke window, while CIMT forces the reorganisation of neural connectivity in the injured motor cortex, resulting in higher functional outcomes than with monotherapy. Combination Treatment Enhances Connectivity Between the Injured and Uninjured Motor Cortex During Fine Motor Moments The present study has revealed that MCAO has severely disrupted the local and interhemispheric communication until 2 weeks of post-stroke. Locally, ischaemic injury resulted in a severe reduction in PSD within the theta band of the injured motor cortex. While monotherapies did not yield a significant effect on beta-theta-beta power, the combination treatment successfully restored theta band (PSD) to baseline physiological levels by day 15 post-stroke. A treatment with monotherapy for ischaemic stroke was not beneficial in ameliorating the interhemispheric connectivity up to 2 weeks after MCAO. On the other hand, with combination treatment, there was a significantly enhanced waveform synchronisation between hemispheres from day 8 to day 15. Thus, the functional recovery in fine forelimb movements 41 and in the neurological recovery 42 is attributable to partial restoration of interhemispheric communications as observed in the present study. The fine motor movements in rodents are primarily controlled by neural circuits involving the motor cortex. 43 The fine motor skills are typically the most delayed motor functions to recover from the stroke. 44 In several cases, even after rehabilitation therapies, patients are left with persistent fine motor deficits. 45 It is interesting to note that stroke-induced deficits in skilled fine motor movements are long-lasting, but not permanent due to direct association with the neural plasticity in the penumbral motor cortex, and are associated with the restoration of interhemispheric connectivity. This suggests that changes in network integrity may require much more time before translating into measurable, true motor recovery measures. Enhanced synchronisation at theta and beta frequency ranges during the reach-to-grasp task indicates that low-frequency oscillations such as theta rhythms are critical for long-range communication between motor circuits, compensatory network reorganisation, 46 indicative of ongoing plasticity and recovery mechanisms. 47 In stroke-induced rats, a combinational treatment resulted in increased theta oscillations in the injured motor cortex during the task, reflecting attentional processing of movement control and possibly indicating ongoing reorganisation within the injured motor cortex. 48 An increase in beta power in the injured motor cortex suggests an enhanced cortical excitability during motor output. 49 However, few studies have indicated that an increased beta frequency range may suggest maladaptive plasticity50, 51 with engagement of compensatory cognitive mechanisms during the task. 52 The reappearance of these oscillatory patterns with combined treatment, thus, is associated with gradual improvements in fine motor performance. 48 These findings support the idea that combination therapies facilitate network reorganisation in the peri-lesioned areas necessary for fine motor recovery, most likely by directing the neuroprotective qualities of VPA to specific goal-oriented synaptic remodelling and connectivity with the help of CIMT. Taken together, the results from the present study provide a strong link between the recovery of fine forelimb movements and frequency-specific cortical reorganisation within and between the hemispheres. Specifically, the study demonstrated that 2 weeks of combination treatment, but not monotherapy, partially restores interhemispheric connectivity, particularly long-range connections following ischaemic stroke. Future studies will focus on dose-dependent changes in these oscillations during execution of skilled fine motor movements. Acknowledgements The authors wish to thank the National Institute of Mental Health and Neuro Sciences (NIMHANS) for providing infrastructural facilities to the candidate for carrying out the study. Sincere thanks to Ravi S. Muddashetty (Centre for Brain Research, Bengaluru) for his insightful feedback and valuable guidance throughout the study. The authors would like to thank Mr Gnan Kumar for his invaluable assistance, which contributed to the completion of this project. RPG contributed to the research design, experiments, analysis and writing the original draft; DYN was involved in the study design and proofreading of the manuscript; TRL contributed to conceptualisation and study design, and was involved in data analysis, review and editing the manuscript. The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article. Funding: The authors received no financial support for the research, authorship and/or publication of this article. ORCID iD: Laxmi T.R. https://orcid.org/0000-0001-8147-9559 Data Availability All data analysed during this study are included in this published article. Statement of Ethics Ethical approval was obtained from the Institutional Animal Ethics Committee (Approval ID: AEC/75/498/NP), which is under the guidelines of the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), Government of India. References Associated Data Data Availability Statement All data analysed during this study are included in this published article.