Limited Effect of Chronic Valproic Acid Treatment in a Mouse Model of Machado-Joseph Disease

Machado-Joseph disease (MJD) is an inherited neurodegenerative disease, caused by a CAG repeat expansion within the coding region of ATXN3 gene, and which currently lacks effective treatment. In this work we tested the therapeutic efficacy of chronic treatment with valproic acid (VPA) (200mg/kg), a compound with known neuroprotection activity, and previously shown to be effective in cell, fly and nematode models of MJD. We show that chronic VPA treatment in the CMVMJD135 mouse model had limited effects in the motor deficits of these mice, seen mostly at late stages in the motor swimming, beam walk, rotarod and spontaneous locomotor activity tests, and did not modify the ATXN3 inclusion load and astrogliosis in affected brain regions. However, VPA chronic treatment was able to increase GRP78 protein levels at 30 weeks of age, one of its known neuroprotective effects, confirming target engagement. In spite of limited results, the use of another dosage of VPA or of VPA in a combined therapy with molecules targeting other pathways, cannot be excluded as potential strategies for MJD therapeutics.


Introduction
Polyglutamine (PolyQ) diseases are neurodegenerative disorders caused by an expansion of trinucleotide CAG repeats within the coding region of specific genes [1]. This group of disorders includes spinal bulbar muscular atrophy (SBMA), Huntington's disease (HD), Dentatorubral-Pallidoluysian atrophy (DRPLA), and six types of spinocerebellar ataxias (SCA's) [2]. Machado-Joseph disease (MJD) or Spinocerebellar Ataxia type 3 (SCA3) is the most common dominantly inherited SCA worldwide and is caused by the expansion of a polyQ tract in the Cterminus of the ATXN3 gene product [3]. Both the normal and expanded ataxin-3 (ATXN3) proteins are expressed ubiquitously, although the neurodegeneration in MJD is limited to some brain regions, mainly in cerebellum, brainstem and spinal cord [4]. The symptoms include ataxia, progressive external ophthalmoplegia, pyramidal and extrapyramidal signs, peripheral amyotrophies, intention fasciculation-like movements of facial and lingual muscles, rigidity, and bulging eyes [5][6][7]. The pathological hallmark of the disease is the presence of neuronal intranuclear inclusions (NIIs) of aggregation-prone expanded ATXN3 in the patients' brain, being the pathogenic relevance of these aggregates still unclear [8][9][10][11].
Despite the recent efforts towards the understanding of the pathogenesis of this disorder, the molecular pathways that ultimately lead to neuronal demise remain mostly unknown and no effective treatments are yet available for MJD, as for other polyQ diseases. Nevertheless, there seem to be common pathways between all polyQ diseases that were shown to be altered, which could be explored in the development of therapeutics, related to transcriptional dysregulation, mitochondrial dysfunction, oxidative stress, Ubiquitine Proteasome System (UPS) impairment, excitotoxicity, DNA damage and activation of apoptotic pathways [12].
Nevertheless, the translation of candidate therapies to clinical trials is a very long process due to uncertainty for human safety and has not improved significantly in the last years. In this context, drug re-purposing strategies, which relies on finding new uses for existing FDAapproved compounds, has been gaining attractiveness due to the faster translation to the clinic, with predictably less safety issues.
VPA is an FDA-approved compound that has been used over the years as an anticonvulsant and mood-stabilizing drug in the treatment of epilepsy, bipolar disorder and migraine [13], with a relatively safe profile in clinical use. In the last years, a growing body of evidence indicates that VPA holds promise in treating other neurodegenerative diseases due to its diverse mechanisms of action. Its pharmacological effects comprise a range of mechanisms, including inhibition of histone deacetylases, increased gamma-aminobutyric acid (GABA)-ergic transmission, reduced release and/or effects of excitatory amino acids, blockade of voltage-gated sodium channels and modulation of dopaminergic and serotoninergic transmission [14]. VPA treatment is also known to produce changes in the expression of multiple genes, involved in transcription regulation, cell survival, ion homeostasis, cytoskeletal modifications, signal transduction, endoplasmic reticulum stress and longevity [13,15] This drug has been shown to delay the disease onset, to reduce neurological deficits and/or to prolong survival in several models of neurodegenerative diseases, including HD, SBMA and Amyotrophic Lateral Sclerosis (ALS) [16][17][18]. In MJD, VPA was reported to alleviate neurodegeneration in a Drosophila model of the disease [19] and to attenuate mutant ATXN3-induced cell toxicity in a human neuronal cell model [20]. Moreover, we have previously shown a significant reduction of mutant ATXN3 aggregation and neurological dysfunction in a C. elegans model of MJD upon VPA treatment through the protective role of the transcription factor DAF-16, supporting a role in protection against proteotoxicity related to aging and cell survival [21]. However, its therapeutic efficacy is still not demonstrated in a mammalian model of MJD. The goal of this work was to test the therapeutic efficacy of chronic VPA treatment in a mouse model of MJD, CMVMJD135 [22]. Our results show that chronic VPA treatment at the dosage used in this pre-clinical trial, lead to very limited and transient phenotypic effects in the CMVMJD135 mouse model, and did not change the ATXN3 inclusion load neither astrogliosis in affected brain regions.

Results
Effect of VPA acute treatment in histone acetylation and neuroprotective molecules in the cerebellum Although CMVMJD135 mouse model do not present a general hypoacetylation of histones in specific brain regions, we measured the H3 acetylation and we observed a trend towards an increase in H3 acetylation levels upon 5 days of VPA acute treatment (Fig 1A) This result did not present statistical difference due to high variability between samples, however, a trend was observed to an hyperacetylation of histone H3 in the cerebellum upon VPA treatment, when normalized to the nuclear protein Lamin B1. Furthermore, and considering our previous results in C.elegans model of MJD upon VPA treatment, we also assessed the mRNA levels of GADD45α and BIM, two genes related to stress resistance and apoptosis regulation, respectively [23][24][25][26], in the cerebellum of CMVMJD135 mice, where a trend towards an increase upon VPA acute treatment was also observed (Fig 1B). Plasma VPA concentration was also assessed after 30 and 120 minutes post-injection. An average of 252 ± 15.8 μmol/L after 30 minutes was observed in VPA-treated animals while a 25.7 ± 8.7 μmol/L concentration was detected after 120 minutes post-injection ( Fig 1C).
VPA treatment had limited effects in neurological deficits and decreased the body weight gain in CMVMJD135 mice Chronic VPA treatment was initiated at 5 weeks of age, with a dosage of 200mg/kg, for 5 consecutive days each week, until 30 weeks of age. A battery of neurological and motor coordination tests was performed since 4 weeks of age until 30 weeks of age (Fig 2).
No differences were found between WT and CMVMJD135 mice at 4 weeks of age before the beginning of the injections. CMVMJD135 mice start showing less body weight gain at 16 weeks of age, being statistically different from age-matched WT littermates at 24 weeks of age ( Fig 3A). VPA treatment significantly reduced the already diminished body weight gain of the transgenic animals since very early in this trial ( Fig 3A) suggesting some toxicity to these animals.
Other than reduced weight gain, no apparent clinical symptoms indicative of significant health impact were observed during long-term VPA treatment of WT and transgenic animals, and no more than 20% of their total body weight was lost at any instance. The first sign of neurological disease in the CMVMJD135 mouse model is the presence of muscular grip strength abnormalities at 6 weeks of age, given by the significant decrease in the latency to fall off in the hanging wire test [22]. VPA treatment did not alter the progression of the CMVMJD135 animals in the hanging wire, demonstrating an absence of effect in muscle strength and/or fine motor coordination of the paws (Fig 3B). Spontaneous locomotor activity of transgenic animals, given by the number of squares travelled in the arena, was markedly increased upon VPA treatment at 24 weeks of age ( Fig 3C). However, spontaneous vertical exploratory activity, tremors and clasping were not improved by VPA treatment (Fig 3D, 3E and 3F).  , * represent p<0.05, ** represent p<0.01 and *** represent p<0.001, black asterisks represents the difference between WT and CMVMJD135, blue asterisks represents the difference between non-treated and VPA-treated CMVMJD135 (Repeated-measures ANOVA, Tukey correction for continuous variables, One-Way ANOVA for differences between groups in specific ages of the continuous variables and Chi-square Fisher's exact test for categorical non-continuous variables).
Gait abnormalities were assessed qualitatively by the analysis of the footprint pattern. CMVMJD135 mice model presented foot dragging already at 12 weeks progressing through age. VPA was able to decrease the severity of this phenotype only at 14 weeks (Fig 3G), while no effect was observed at more advanced stages of the trial. In addition, stride length was also measured and although not significant, a trend towards and improvement in transgenic animals upon VPA treatment was observed at 22 and 24 weeks of age ( Fig 3H).
Long-term VPA treatment led to limited improvement in balance and motor coordination at later disease stages Since chronic VPA treatment induced a decrease in body weight of transgenic animals, behavior performance in the motor and balance coordination tests was normalized to body weight. Results of behavioral tests without normalization to body weight are included in supplementary data (S1 Fig).
VPA treatment ameliorated balance and motor coordination of CMVMJD135 mice at 24 weeks of age, after 20 weeks of daily treatment, as assessed by the time taken to cross the 11 mm circle and 12 mm beams in the balance beam walk test. However, this improvement was not maintained at 30 weeks of age ( Fig 4A and 4B).
In the motor swimming test, VPA-treated CMVMJD135 mice also had a better performance later in life (22 and 30 weeks of age) when compared to vehicle-treated mice ( Fig 4C). Other motor deficits observed in CMVMJD135 mice, namely the loss of motor coordination observed in the Rotarod test, were alleviated by VPA treatment at 22 weeks of age for 8 and 20 rpm, but not maintained at 30 weeks of age. (Fig 4D and 4E).
VPA treatment did not change the ataxin-3 inclusion load and astrogliosis in specific brain regions of CMVMJD135 mice At the pathological level, CMVMJD135 mice show presence of ATXN3 NIIs in the nucleus of cells in different regions of the CNS including the pontine nuclei, reticulotegmental nucleus of the pons, spinal cord neurons, facial motor nuclei, anterior olfactory nuclei, ventral tenia tecta, inferior olive, dentate nuclei, locus coeruleus, cuneate nuclei and lateral reticular nuclei [22]. The analysis of brain tissue of VPA-treated and non-treated CMVMJD135 by immunohistochemistry of ATXN3 in facial motor nuclei (7N) and lateral reticular nuclei (LRt), did not reveal significant differences between groups (Fig 5A-5D).
Astrogliosis observed in substantia nigra of the CMVMJD135 mouse model, was also not mitigated upon VPA treatment (Fig 5E and 5F).

Chronic VPA treatment increases GRP78 protein levels in cerebellum of CMVMJD135 mice
One of the neuroprotective actions described for VPA is its ability to increase GRP78 protein levels, through HDAC inhibition [15]. GRP78, also known as binding immunoglobulin protein (BiP), is a stress chaperone protein found in the lumen of the endoplasmic reticulum (ER) which binds newly synthesized proteins as they are translocated into the ER, and keeps them in a competent state for subsequent folding and oligomerization [27]. We specifically investigated the potential role of HDAC inhibition by monitoring the GRP78 protein levels induction upon VPA treatment. We observe a striking increase in GRP78 protein levels in CMVMJD135-treated animals when compared to non-treated animals, in the cerebellum at 30 weeks of age (Fig 6).

Discussion
In the past years, the use of VPA as a treatment for neurodegenerative disease models has been shown to improve neurological phenotypes, decrease cell degeneration and toxicity, together with the increase of histone acetylation and subsequent gene transcription activation [28][29][30][31][32][33]. The dosage of VPA of 200mg/kg used in this pre-clinical trial was previously described in a pre-clinical trial ALS, in which neuroprotective and histone acetylation effects were shown [34,35]. Here, we also show that five days treatment with 200mg/kg of VPA was able to exert a trend towards an increase in histone acetylation in cerebellum and a tendency to increase GADD45α and BIM mRNA levels, which may be used as VPA target engagement [36][37][38]. The acute treatment and/or high variability between animals may account for the lack of statistical differences between groups. Acute VPA treated animals presented therapeutic levels of 252 ± 15.8 μmol/L and 25.7 ± 8.7 μmol/L in plasma, after 30 and 120 minutes post-injection, respectively, which was within the usually accepted therapeutic range of VPA. Therefore, we performed a pre-clinical trial with five consecutive day treatment starting at 5 and ending at 30 weeks of age in CMVMJD135 mouse model. Animals in the pre-clinical trial were always maintained in an appropriate and healthy environment avoiding the development of any stressful condition that may interfere with their motor performance. Although previous findings by several groups suggested a therapeutic effect for VPA in cellular and invertebrate animal models of SCA3, in the present study, chronic VPA treatment of CMVMJD135 mice lead to a late and limited improvement in the motor performance, given by the beam balance, motor swimming, rotarod and spontaneous locomotor activity tests. For other general health and neuromuscular function, VPA-treatment had only marginal or even no effects comparing to vehicle-treated transgenic animals. In addition, the dosage used in this study is not considered toxic for mice and is already described as safe in the literature [18,34,35,39]. Nevertheless, chronic VPA treatment at similar dosage was already described to reduce the body mass gain in WT animals [34], even though chronically treated human patients are known to increase their body weight upon VPA treatment [40][41][42]. Metabolic differences and biological system diversity could be the reason for these contradictory observations.
At the pathological level, we examined the presence of ATXN3 neuronal nuclear inclusions in facial nuclei (7N) and lateral reticular nuclei (LRt), two regions described to be affected in MJD human patients [43,44], as well as in CMVMJD135 mouse model [22]. No differences were observed in the amount of nuclear neuronal inclusions of ATXN3 in both VPA-treated and non-treated CMVMJD135 mice.
Additionally, astrogliosis is a consistent pathological phenotype of CMVMJD135 mouse model and human patients, which was also not mitigated by VPA chronic treatment.
The overexpression of specific chaperones has been shown to allow protection against cellular damage and/or death caused from an extensive group of agents and conditions including cytotoxic chemicals [45], oxidative stress [46] and ER stress [27]. Here we show that the induction of GRP78 protein levels upon VPA chronic treatment, whose overexpression may be neuroprotective in proteinopathies, including MJD [47,48], could be one of its cytoprotective actions in the MJD context, enhancing the folding capacity of the ER. The induction of GRP78 also indicates VPA target engagement during the chronic treatment in the pre-clinical trial, as this protein is known to be induced by VPA [15]. Although we did not observe a statistical difference in the CMVMJD135 animals when comparing to WT, enhancing the expression of this molecular chaperone in the cerebellum of these mice could be one of the neuroprotective mechanisms responsible for the late and mild improvement of the CMVMJD135 animal motor performance.
The observed beneficial effects of VPA were transient, occurred mostly later in life and thus at an advanced stage of disease (between 22 and 30 weeks of age), mainly in behavioral tests more related with motor coordination. These results are comparable to some extent with previous findings, in which we have shown a significant reduction in neurological dysfunction in a C. elegans model of MJD after VPA treatment that was more relevant later in the worm's life (day10) in spite of early treatment [21]. Previous evidence also suggested a protective role for VPA in the context of MJD, both in cell and Drosophila models, by attenuating mutant ATXN3 induced cell toxicity and alleviating polyQ-induced phenotypic abnormalities, without major impact on ATXN3 inclusion [19,20]; in C. elegans, there was some improvement of aggregation, but less prominent than that observed for other compounds, for instance Hsp90 inhibitors [21].
Although the effects observed in our mouse model were not striking, only one dosage of VPA was tested; thus, the possibility of testing other dosages, far from toxic and lethal ones [49,50], should be considered, as they could exert more pronounced effects. Moreover, the complex activity and a broad range of VPA effects also create the need for further clarification of the effects of this drug not only at the symptom level, but also molecular and pathological levels in the CMVMJD135 mouse model. In fact, and although the use of HDACi's in the context of polyQ diseases has showed promising results, the evidence for a globally decreased histone acetylation is not fully consistent [51], and this strategy still lacks some target specificity/ selectivity [52] and requires a more in depth study of the mechanisms of action of these compounds in the central nervous system. Furthermore, chronic VPA treatment in human patients can produce some side effects, such as weight gain [53,54], decreased reproductive potential [55,56] and increased susceptibility to birth defects [57][58][59]. Nevertheless, the strategy of repurposing FDA/EMA-approved molecules, as VPA, could be of benefit for MJD and other rare diseases lacking effective therapies. Additionally, there is still an open window for different VPA dosages to be tested and since in the past years, a growing number of efforts are being developed for the formulation of a new generation of more selective and specific compounds this could be useful for the treatment not only of MJD, but also of other polyglutamine diseases.

Ethics statement
All animal procedures were conducted in accordance with European regulations (European Union Directive 86/609/EEC). Animal facilities and the people directly involved in animal experiments (SE, SDS, ANC and ASF) were certified by the Portuguese regulatory entity-Direcção Geral de Alimentação e Veterinária. All of the protocols performed were approved by the Ethics Subcommittee for Life and Health Sciences of the Life and Health Sciences Research Institute, University of Minho. All experiments were designed with commitment to the principles of refinement, reduction, and replacement and performed according to the FELASA guidelines to minimize discomfort, stress, and pain to the animals, with defined humane endpoints [60]. Humane endpoints for the preclinical trial were defined as 20% reduction of the body weight, inability to reach food and water, presence of wounds in the body and dehydration), however they were not needed as the study period was conceived to include ages at which animals do not reach these endpoints.

Transgenic mice model and drug administration
We used the CMVMJD135 (background C57BL/6) mouse model, expressing an expanded version of the human MJD1-1 cDNA (the 3 UIMs-containing variant of ATXN3) under the regulation of the CMV promoter, ubiquitously and at near-endogenous levels [22]. These animals show a slowly progressive motor phenotype and CNS pathology consistent with that of MJD patients [22]. Male transgenic and non-transgenic drug-and placebo-treated animals were sequentially assigned and housed at weaning in groups of 5 animals in filter-topped polysulfone cages 267 × 207 × 140 mm (370 cm 2 floor area) (Tecniplast, Buguggiate, Italy), with corncob bedding (Scobis Due, Mucedola SRL, Settimo Milanese, Italy) in a conventional animal facility. DNA extraction, animal genotyping and CAG repeat size analyses were performed as previously described [61], with the mean repeat size (±SD) for all mice of (133±1). Male littermates wild-type (WT) animals were used as controls. All animals were maintained under standard laboratory conditions: an artificial 12 h light/dark cycle (lights on from 8:00 to 20:00 h), with an ambient temperature of 21±1°C and a relative humidity of 50-60%; the mice were given a standard diet (4RF25 during the gestation and postnatal periods, and 4RF21 after weaning, Mucedola SRL, Settimo Milanese, Italy) and water ad libitum. We administered Valproic acid sodium salt (PG-4543, Sigma) during five consecutive days per week through intraperitoneal injection (i.p), in a dosage of 200mg/kg dissolved in 0,9% saline. Control animals were given a placebo of injection buffer (0.9% NaCl) with the same frequency. Treatment was initiated at five weeks of age, i.e. one week before the onset of the first neurological symptoms, until 30 weeks of age of the pre-clinical trial. For a pilot study, WT animals were treated for 5 consecutive days with i.p. injections of VPA at 200mg/kg or saline.

Behavioral analysis
Behavioral analysis was performed during the diurnal period in groups of 5 male animals per cage including CMVMJD135 hemizygous transgenic mice and WT littermates (n = 10-15 per genotype) treated and non-treated with VPA. All behavioral tests started in a pre-symptomatic stage of the disease (4 weeks of age) and were conducted until an age at which the phenotype is fully established (30 weeks) [22]. The animals were weighed one week before the start of drug treatment (4 weeks) and then every two weeks until 30 weeks of age (Fig 2).
Neurological examination (based on SHIRPA protocol). Based on the SHIRPA protocol we established an adapted protocol for phenotypic assessment applied since 4 weeks until 24 weeks of age, in which we used the tests for which, based on our previous experience, CMVMJD135 mice usually present significant phenotypic alterations [22]. We assessed motor function through the spontaneous activity test, by counting wall-leanings during five minutes, and locomotor activity in which we counted the number of squares travelled over 30 secs, in an arena (55×33×18 cm) with 15 labeled squares. Other observational measurements included tremors and limb clasping, in which we suspended the animal by the tail and classified the extensor reflexes. In this protocol we also included the hanging wire test, as a measure of muscle strength and fine motor coordination of the paws. This protocol was adjusted in order to minimize animal handling and to generate uniformity in waiting times between the tests [62].
Footprint analysis. To evaluate the dragging of the paws, the footprint test was used since 10 weeks of age. To obtain footprints, the hind-and forepaws of the mice were coated with black and red non-toxic paints, respectively. We used a clean rectangular paper sheet placed on the floor of the runway for each run. The animals were allowed to walk along a 100-cmlong × 4.2 cm width × 10 cm height corridor in the direction of an enclosed black box. Each animal was allowed to achieve one valid trial per age. To evaluate the severity of foot-dragging through age the footprinting pattern was classified at each time point considering six consecutive steps (0 = absent dragging, up to three steps; 1 = moderate dragging, less than three steps out of six; 2 = severe dragging, all steps out of six show dragging). The stride length was also measured through the footprinting pattern by measuring the length between three consecutive steps.
Motor swimming test. To assess swimming movement coordination, the time that animals take to reach a safe platform at the end of a container (60 cm distance) with 15 cm depth of water at 24-26°C was recorded bi-weekly since 22 weeks of age. The protocol consisted of 2 days of training with three trial followed by three days of test with two trials as previously described [22,63].
Beam walk test. Balance and fine motor coordination of mice were assessed by measuring the ability of the mice to traverse, without falling, a graded series of narrow beams to reach an enclosed safety platform as previously described [22,63]. During training, mice were placed at the start of the 12 mm square beam and trained over 3 days (3 trials per day) to traverse the beam to the safe platform. On the fourth day, they were tested in the training beam (12 mm square) and 11 mm round beam (2 trials per beam).
Rotarod test. To evaluate motor skill learning and coordination with another paradigm, mice were tested in a rotarod apparatus (TSE systems, Bad Homburg, Germany). The protocol is comprised of 3 training days at a constant speed (15 rpm) for a maximum of 60 s in four trials, with a 10 min interval between each trial. On the fourth day, animals were tested for each of 6 different speeds (5 rpm, 8 rpm, 15 rpm, 20 rpm, 24 rpm and 31 rpm) for a maximum of 60s in two trials, with a 10-min-long interval between each trial, as previously described [61].

Immunohistochemistry and quantification of ataxin-3 neuronal inclusions and astrogliosis
Thirty week-old WT and CMVMJD135 littermate mice, VPA-treated and non-treated (n = 4 for each group) were deeply anesthetized-with a mixture of ketamine hydrochloride (150 mg/ kg) plus medetomidine (0.3 mg/kg) and transcardially perfused with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde (PFA) (Panreac, USA). Brains were removed and post fixed overnight in PFA 4% and embedded in paraffin. Slides with 4-μm-thick paraffin sections were subjected to antigen retrieval (Buffer Citrate, 1M) and then incubated with mouse anti-ATXN3 (1H9) (1:1000, MAB5360, Milipore) and GFAP (1:500, Z0334, Dako corporation) which were detected by incubation with a biotinylated anti-polyvalent antibody, followed by detection through biotin-streptavidin coupled to horseradish peroxidase and reaction with the DAB (3, 3'-diaminobenzidine) substrate (Lab VisionTM Ultra-VisionTM Detection kit, Thermo Scientific). The slides were counterstained with 25% hematoxylin according to standard procedures. ATXN3 positive inclusions in the facial motor nucleus (7N) and lateral reticular nucleus (LRt), and GFAP positive cells in substantia nigra (SN) of vehicle or VPA-treated animals (n = 4 for each conditions, 4 slides per animal) were quantified and normalized for total area using the Olympus BX51 stereological microscope (Olympus, Japan) and the Visiopharma integrator system software (Visopharm, Denmark) as previously described [22]. The total area of 7N,LRT and SN were chosen based on the mouse brain atlas [64].

Determination of Valproic acid Plasma Levels
The plasma valproic acid levels were measured applying the VALP assay using the Dimension Vista1 System (VALP Flex1 reagent cartridge)-SIEMENS.

Statistical analysis
The experimental unit used in this study was a single animal. Experimental design was based on power analyses for optimization of sample size [65]. Mouse sample size calculations were performed for each behavioral test and pathological analyses assuming a power of 0.8 and a significance level of p < 0.05. The effect size was calculated aiming at detecting 50% improvement. We used n = 10 to 15 per genotype/treatment for behavioral tests, and a group size of four animals per group for quantification of ATXN3 NIIs analysis. Data was analyzed through the non-parametric Mann-Whitney U-test when variables were non-continuous or when a continuous variable did not present a normal distribution (Kolmogorov-Smirnov test, p<0.05) (Rotarod). Continuous variables with normal distributions and with homogeneity of variance (Levene's test) were analyzed with Repeated-Measures ANOVA for longitudinal multiple comparisons, using Tukey test for post-hoc comparisons and One-way ANOVA for paired comparisons Non-continuous categorical variables were analyzed through Chi-Square Fisher exact test. All statistical analyses were performed using SPSS 22.0 (SPSS Inc., Chicago, IL) and G-Power 3.1.9.2 (University Kiel, Germany). A critical value for significance of P < 0.05 was used throughout the study. Values were expressed as mean ± SEM for continuous variables and as percentages for non-continuous variables.
Supporting Information S1 Checklist. NC3Rs ARRIVE Guidelines Checklist is provided as supporting information.
(DOCX) S1 Fig. Balance and motor coordination were improved at later stages upon VPA treatment given by the balance beam and motor swimming performance. (A) Amelioration of balance and motor coordination at 24 weeks of age in 11 mm circle and (B) 12 mm square beams in beam walk test; (C) Motor swimming coordination improvement; (D,E) no improvement in increasing rotations in Rotarod were observed between VPA-treated and non-treated CMVMJD135. Bars represent the mean ± SEM (WT veh, n = 10; WT VPA, n = 15, CMVMJD135 vehicle, n = 10, CMVMJD135 VPA, n = 13), Ã represent p<0.05, ÃÃ represent p<0.01 and ÃÃÃ represent p<0.001, black asterisks represents the difference between WT and CMVMJD135, blue asterisks represents the difference between non-treated and VPA-treated CMVMJD135, (Repeated-measures ANOVA, Tukey correction for continuous variables, One-Way ANOVA for differences between groups in specific ages of the continuous variables and Mann-Whitney U test for continuous variables without normal distribution (Rotarod)). (TIF)