Treatment with sodium butyrate induces autophagy resulting in therapeutic benefits for spinocerebellar ataxia type 3

Spinocerebellar ataxia type 3 (SCA3, also known as Machado Joseph disease) is a fatal neurodegenerative disease caused by the expansion of the trinucleotide repeat region within the ATXN3/MJD gene. Mutation of ATXN3 causes formation of ataxin‐3 protein aggregates, neurodegeneration, and motor deficits. Here we investigated the therapeutic potential and mechanistic activity of sodium butyrate (SB), the sodium salt of butyric acid, a metabolite naturally produced by gut microbiota, on cultured SH‐SY5Y cells and transgenic zebrafish expressing human ataxin‐3 containing 84 glutamine (Q) residues to model SCA3. SCA3 SH‐SY5Y cells were found to contain high molecular weight ataxin‐3 species and detergent‐insoluble protein aggregates. Treatment with SB increased the activity of the autophagy protein quality control pathway in the SCA3 cells, decreased the presence of ataxin‐3 aggregates and presence of high molecular weight ataxin‐3 in an autophagy‐dependent manner. Treatment with SB was also beneficial in vivo, improving swimming performance, increasing activity of the autophagy pathway, and decreasing the presence of insoluble ataxin‐3 protein species in the transgenic SCA3 zebrafish. Co‐treating the SCA3 zebrafish with SB and chloroquine, an autophagy inhibitor, prevented the beneficial effects of SB on zebrafish swimming, indicating that the improved swimming performance was autophagy‐dependent. To understand the mechanism by which SB induces autophagy we performed proteomic analysis of protein lysates from the SB‐treated and untreated SCA3 SH‐SY5Y cells. We found that SB treatment had increased activity of Protein Kinase A and AMPK signaling, with immunoblot analysis confirming that SB treatment had increased levels of AMPK protein and its substrates. Together our findings indicate that treatment with SB can increase activity of the autophagy pathway process and that this has beneficial effects in vitro and in vivo. While our results suggested that this activity may involve activity of a PKA/AMPK‐dependent process, this requires further confirmation. We propose that treatment with sodium butyrate warrants further investigation as a potential treatment for neurodegenerative diseases underpinned by mechanisms relating to protein aggregation including SCA3.

Spinocerebellar ataxia type 3 (SCA3), also known as Machado-Joseph disease (MJD), is a neurodegenerative disease characterized by a progressive loss of muscle control and movement, leading to wheelchair dependence and decreased lifespan. 12][3] SCA3 is the most common of the hereditary ataxias found throughout the world (21%-28% of autosomal-dominant ataxia), [4][5][6] with a high prevalence within the Azores of Portugal 7 and Indigenous communities of north-east Arnhem Land in Australia. 80][11] While the ATXN3 gene of healthy subjects contains a short CAG trinucleotide repeat region (12-40 CAG repeats), this region contains over 40, and as high as 87, CAG repeats in SCA3 patients. 7,9,12,13owever, 44-54 CAG repeats have been considered to be an intermediate length and may not lead to SCA3. 14he ATXN3 gene encodes the ataxin-3 protein, with the CAG repeat region encoding a polyglutamine (polyQ) tract toward the C-terminus of the protein. 12,13europathological staining of patient brain samples often reveals the presence of neuronal intranuclear inclusions (NII) containing the ataxin-3 protein 2,15,16 and extraction of these proteins has revealed the presence of full-length ataxin-3 protein, as well as smaller ataxin-3 protein fragments. 179][20] It has been previously reported that ataxin-3 is a histone-binding protein and that polyQ expansion within the protein increases the extent of that binding, in turn affecting histone acetylation. 185][26] However, this histone hypoacetylation has not been reported to be present in all models of the disease. 21ne therapeutic strategy that has been explored for SCA3 is treatment with drugs that may increase levels of histone acetylation, called histone deacetylase (HDAC) inhibitors.HDACs are a class of enzymes that remove acetyl groups from ε-N-acetyl lysine located on histones, causing histones to bind DNA more tightly and prevent transcription. 27,28[31][32][33] Compounds with this HDAC inhibitor capacity include suberoylanilide hydroxamic acid (SAHA), trichostatin A, resveratrol, valproic acid and sodium butyrate.27,[34][35][36][37][38] Moreover, a phase I/II clinical trial was conducted for sodium valproate in SCA3 patients with minimal adverse events.39 Sodium butyrate (SB) is a sodium salt of butyric acid, a short-chain fatty acid that is produced within the gut by intestinal microbiota during the fermentation of undigested dietary carbohydrates and fiber.28,40 While butyric acid is critical for intestinal homeostasis, it also exerts a wider range of effects including inhibition of cell proliferation, downregulation of pro-inflammatory effectors, repression of gene expression, and HDAC inhibition, resulting in increased acetylation of histones 3 and 4. 25,40 Recently, a combined treatment of taurursodiol and phenylbutyrate, another derivative of butyric acid with HDAC inhibitor effects, has been found to improve survival in a Together our findings indicate that treatment with SB can increase activity of the autophagy pathway process and that this has beneficial effects in vitro and in vivo.While our results suggested that this activity may involve activity of a PKA/AMPK-dependent process, this requires further confirmation. We prpose that treatment with sodium butyrate warrants further investigation as a potential treatment for neurodegenerative diseases underpinned by mechanisms relating to protein aggregation including SCA3.

K E Y W O R D S
autophagy, Machado-Joseph disease, neurodegeneration, polyQ, sodium butyrate, spinocerebellar ataxia type 3, trinucleotide repeat disease, zebrafish double-blind, placebo-controlled clinical trial of amyotrophic lateral sclerosis (ALS) patients. 41Since then, this combination therapy has received FDA approval for the treatment of ALS.As butyrate can elicit such a wide spectrum of positive effects, it is likely butyrate has multiple distinct mechanisms of action, 40,42 many of which are yet to be fully elucidated.
In this study, we aimed to determine the efficacy of SB in alleviating SCA3 pathology in cell culture and zebrafish models of SCA3.Furthermore, we aimed to elucidate the underlying mechanism by which SB rescues SCA3 disease phenotypes.Enhancing our understanding of the mechanisms of action underlying the neuroprotective effects induced by SB could provide therapeutic implications for a wide range of neurodegenerative and proteinopathy diseases.We hypothesized that, in addition to improving transcriptional dysregulation, SB can ameliorate SCA3 phenotypes via induction of the autophagy protein quality control pathway.Increased activation of the autophagy pathway may act to degrade or remove pathological ataxin-3 oligomeric species or aggregates in cell cultures, alleviating neurotoxicity.Importantly, here we demonstrate that administration of SB can induce autophagy and decrease SCA3 disease phenotypes both in vitro and in vivo, highlighting the therapeutic potential of autophagy inducers for the treatment of SCA3 and other related neurodegenerative diseases.

| Transgenic SCA3 zebrafish
All animal experiments were performed in accordance with the Animal Ethics Committee of Macquarie University, NSW, Australia (ARA: 2016/004 and 2017/019).Adult zebrafish (Danio rerio) were housed in a standard recirculating aquarium system at 28.5°C on a 14:10 light: dark cycle with twice daily feeding of artemia and standard pellet. 43These experiments used our previously described transgenic zebrafish model of SCA3. 44Briefly, the driver lines express a HuC (elavl3) neuronal promoter driving Gal4 VP16 [Tg(elavl3:GAL4-VP16-mCherry)mq15], and the responder lines express EGFP-ATXN-3 (containing 23 or 84 CAG repeats) and dsRED under the control of the UAS promoter that is activated only in tissues that are expressing Gal4 VP16 [Tg(UAS:Hsa.ATXN3_23xCAG-EGFP,DsRed)mq16 and Tg(UAS:Hsa.ATXN3_84xCAG-EGFP,DsRed)mq17].For the studies described here, driver and responder lines were mated to generate F1 HuC-EGFP-Ataxin-3 23Q or 84Q lines, which were in-crossed to generate F2 embryos for use in this study.

| SCA3 cell culture model
SH-SY5Y cells were grown in Dulbecco's Modified Eagle's Medium (DMEM)/Nutrient Mixture F12 Ham supplemented with 10% fetal bovine serum (FBS) and maintained at 37°C and 5% CO 2 .pcDNA3-myc-Ataxin3Q28 and pcDNA3-myc-Ataxin3Q84 were a gift from Henry Paulson (Addgene plasmids # 22124 and #22125, unmodified in house). 45These vectors, containing full-length ataxin-3 (28Q or 84Q) and a neomycin-resistant gene, were used for the stable selection of transfected cells.500 μg/mL of neomycin was used to select cells stably expressing ataxin-3.Cells were treated with 250 μg/mL of neomycin (Sigma) to maintain expression of the selective expression of the ataxin-3 protein or vector control.

| Drug treatments in SCA3 cell
cultures and transgenic SCA3 zebrafish SH-SY5Y cells stably expressing human ataxin-3 were seeded into a 24-well plate at a density of 40 000 cells/cm 2 and incubated at 37°C supplemented with 5% CO 2 .24 h after seeding, cells were treated with SB (3 mM, Cayman Chemicals [Cat #13121]) or vehicle control (autoclaved water) diluted in growth media for a total of 72 h, with SBcontaining media replenished every 24 h to ensure drug efficacy.For the co-treatment study with an autophagy inhibitor, cells were pre-treated with 3MA (5 mM, Cayman Chemicals) for 1 h before SB was added and left on for the duration of the SB treatment or bafilomycin A1 (100 nM dissolved in DMSO, Roche) was added for 4 h of prior to protein extraction.For the AMPK inhibitor experiments, SH-SY5Y cells stably expressing human ataxin-3 (-28Q or -84Q) were seeded into a 6-well plate at a denisty of 150,000 per well and incubated at 37 o C supplemented with 5% CO 2 .At 24 hours post-seeding on the plate, cells were co-treated with SB (3 mM dissolved in autoclaved water) and GSK690693 (10μM dissolved in DMSO, Sigma, #SML0428) for the co-treatment study with AMPK inhibitor.Cells were cultured for a total of 72 hours before an autophagy inhibitor (bafilomycin A1, 400 nM dissolved in DMSO, Roche) was added and left on for 4 hours prior to protein extraction.
For the zebrafish drug treatment studies, zebrafish embryos (1 day post fertilization; dpf) were screened for fluorescence (EGFP and dsRED) indicating that they were positive for the EGFP-ataxin-3 transgenes.In the drug treatment groups, positive embryos were treated with a single administration for five days with SB (250 μM, 500 μM and 1 mM, solubilized in ultrapure water).Appropriate volumes of SB were diluted in zebrafish E3 medium, with the control group containing only E3 media. 46Co-treatment with SB and the autophagy inhibitor chloroquine (1.5 mM, Sigma-Aldrich [Cat# C6628], solubilized in ultrapure water) was performed in the same manner but with chloroquine stock solution added to the E3 media containing SB so that the solution also contained 1.5 mM chloroquine.Vehicle-treated groups were left in the E3 medium.Approximately 25-30 embryos were treated per group per experiment.

| Zebrafish motor behavior testing
Motor function assays were performed as described in Watchon et al. 44 All behavioral tracking was performed using a ZebraLab Tracking System (ZebraBox; Viewpoint).Tracking of 6 dpf larvae was conducted by randomly assorting them in 24 multi-well plates within the ZebraBox housed with an enhanced light source, under conditions of 6 min light, 4 min dark, and 4 min light.The total distance traveled by each larva within the dark phase was calculated.

| Western blotting
SH-SY5Y cells stably expressing human ataxin-3 (28Q and 84Q), were washed with ice-cold PBS and incubated in 100 μL of RIPA buffer containing protease inhibitors (Complete ULTRA tablets, Roche).Cells were gently agitated on an orbital shaker for 5 min before being spun down at 18 000 g for 15 min.Protein lysates were prepared from zebrafish larvae (6 dpf) in RIPA buffer containing protease inhibitors (Roche), using a manual dounce homogenizer.Homogenates were centrifuged for 20 min at 15 000 g. Supernatants were collected and measured for protein concentration using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific).
Equal amounts of protein were separated using SDS-PAGE and transferred to the PVDF membrane for immunoblot probing.Antibodies used included anti-rabbit ataxin-3 (kind gift from H. Paulson) and anti-mouse GAPDH (Proteintech).To test the effect of drug treatments, immunoblots were probed for anti-rabbit acetylated histone 3 at lysine 9 and histone 4 at lysine 5 (ac-H3K9 and ac-H4K5, respectively, Cell Signaling), anti-rabbit LC3B (Abcam), anti-rabbit AMPK (Cell Signaling), anti-rabbit phosphorylated-ULK1 (Ser777; Thermo Fisher), antirabbit ULK1 (Cell Signaling), and anti-mouse beta-actin (Sigma).The immunoblots were probed with appropriate secondary antibodies (Promega and Li-Cor) and visualized by chemiluminescence (Supersignal detection kit, Pierce) using a BioRad GelDoc System or fluorescent detection using the LiCor Odyssey CLX.The intensity of bands within the immunoblot was quantified by Image Studio Lite and the target protein expression level was determined by normalizing against the loading control protein.

| Preparation of cultured SH-SY5Y cells for flow cytometry
For flow cytometry experiments, SH-SY5Y cells were transiently transfected with human ATXN3 plasmids containing a fluorescent tag.Human ATXN3 cDNA was subcloned into a pCS2+ vector to generate ATXN3 constructs containing a short polyQ (28Q) and an expanded polyQ (84Q) fluorescently tagged with EGFP.
To determine transfection efficiency, live cells were imaged within 6-well plates at 20× magnification using an EVOS FL monochrome microscope (Invitrogen, catalog #AMF4300) running Invitrogen image acquisition software.Cells were harvested using 0.5% trypsin/EDTA and pelleted (1500 rounds per minute, 5 min at room temperature [RT]).Cells were re-suspended in 500 μL of lysis buffer (PBS containing 1% Triton-X 100 and complete protease inhibitors [Roche]).DAPI was added to the lysis buffer (final concentration 5 μM) and incubated at RT for 5 min, protected from light.

| Dissociation of zebrafish for flow cytometry
Whole zebrafish larvae (2 or 6 dpf) were euthanized and larvae were digested and dissociated into a single-cell solution, as previously described. 47In brief, following euthanasia, whole larvae were dissected into smaller pieces using a scalpel and forceps and enzymatically digested using 0.5% Trypsin/EDTA.Dissociated cells were then lysed using PBS containing 0.5% Triton-X 100 and a marker of nuclei, either DAPI (final concentration 5 μM) or 1 × Red Dot (Gene Target Solutions, catalog # 40060).

| Flow cytometric analysis of insoluble ataxin-3
Flow cytometry was performed using a Becton Dickinson Biosciences LSR Fortessa analytical flow cytometer and calibrated using CST beads (Becton Dickinson).The fluorescence of transfected cells was compared to an untransfected control sample.Lysed samples were analyzed by plotting forward scatter (area) against relative DAPI (379-28-A) or EGFP (530-30-A/535) fluorescence.The forward scatter threshold was set at 200 to minimize the exclusion of small insoluble particles. 48A total of 20 000 events were acquired for cell culture experiments and 50 000 events for dissociated zebrafish experiments.All axes were set to log 10 .Nuclei were identified and gated based on the intensity of UV fluorescence and relative size (forward scatter).The number of insoluble GFP particles, indicating insoluble ataxin-3 particles, was analyzed based on GFP fluorescent intensity and forward scatter. 47ating of Triton-X insoluble GFP + positive particles was performed by comparing populations to an un-transfected control sample.For analysis comparing the effect of different drug treatments on SCA3 cells and zebrafish expressing ataxin-3-84Q, data were analyzed as the number of insoluble GFP + particles divided by the number of detected nuclei. 47Drug-induced effects were presented as the fold change differences when compared to vehicletreated controls. 47All flow cytometric analysis and gating were performed using FlowJo (version 10.6.2,Becton Dickinson Biosciences).

| In-gel trypsin digestion and liquid chromatography mass spectrometry (LC-MS)
SH-SY5Y cell lysates expressing ATXN3-84Q were separated in 4%-12% NuPAGE Bis-Tris polyacrylamide gels (Invitrogen) in duplicate gels.One gel was Coomassie stained and de-stained in 25% methanol while the second gel was transferred to nitrocellulose membrane and western blot analysis for ATXN3 was performed.Fractions 1-7 in the Coomassie-stained gel were identified by comparison with the molecular weight of the ataxin-3-84Q band from the western blot results.The bands in the gel were excised and cut into 1-2 mm pieces and de-stained in 50 mM ammonium bicarbonate/50% methanol pH 8, followed by 50 mM ammonium bicarbonate/50% acetonitrile pH 8.The de-stained gel pieces were dehydrated in 100% acetonitrile and the solution was removed to allow the gel pieces to air dry.The proteins were reduced and alkylated with 10 mM dithiothreitol (DTT) and 20 mM iodoacetamide (IAA), respectively, and digested with trypsin: protein [1:50 (w/w)] overnight at 37°C as described. 49ollowing overnight digestion, the supernatant was transferred to a fresh tube and the tryptic peptides from the gel pieces were extracted twice with 50% acetonitrile/2% formic acid and combined with the supernatants.
The tryptic peptides were vacuum centrifuged to remove acetonitrile and desalted on a pre-equilibrated C 18 Omix tip.The eluted peptides were further dried under vacuum centrifugation.The peptide pellet was resuspended in 12 μL of 0.1% formic acid and 10 μL was loaded for LC-MS/MS analysis.
Tryptic peptides were separated using an UHPLC Dionex Ultimate 3000 RSLC nano (ThermoFisher, USA) equipped with a Thermo Acclaim™ PepMap™ 100 C 18 column (75 μm diameter, 3 μm particle size, 150 mm length) employing a 60 min gradient (2%-26% v/v acetonitrile, 0.1% v/v formic acid for 40 min followed by 50% v/v acetonitrile, 0.1% v/v formic acid for 10 min and 80% v/v acetonitrile, 0.1% v/v formic acid for 8 min) with a flow rate of 300 nL/min.The peptides were eluted and ionized into Q-Exactive Plus mass spectrometer (ThermoFisher, USA).The electrospray source was fitted with an emitter tip 10 μm (ThermoFisher, USA) and maintained at 1.6 kV electrospray voltage.FTMS analysis was carried out with a 70 000 resolution and an AGC target of 1 × 10 6 ions in full MS (m/z range 400-2000), and MS/MS scans were carried out at 17 500 resolution with an AGC target of 2 × 10 4 ions.Maximum injection times were set to 30 and 50 ms, respectively.A top-10 method was employed for MS/MS selection, ion selection threshold for triggering MS/MS fragmentation was set to 1 × 10 4 counts, isolation width of 2.0 Da, and dynamic exclusion for 20 s was used to perform HCD fragmentation with normalized collision energy of 27.
Raw spectra files were processed using the Proteome Discoverer 2.4 software (Thermo) against the Swissprot database (organism Homo sapiens, version 25/10/2017 with 42 252 sequences) incorporating the Sequest search algorithm.Peptide identifications were determined using a 20-ppm precursor ion tolerance and a 0.1 Da MS/MS fragment ion tolerance for FTMS and HCD fragmentation.Carbamidomethylation modification of cysteines was considered a static modification while oxidation of methionine, and acetyl modification on N-terminal residues were set as variable modifications allowing for a maximum of two missed cleavages.The data were processed through Percolator for estimation of false discovery rates.Protein identifications were validated employing a q-value of 0.01.For label-free quantitation, normalized spectral abundance factors (NSAF) were calculated for each sample according to Zybailov et al. 50The relative fold-change of proteins from cells expressing ATXN-84Q was calculated by the ratio of SB treatment/DMSO vehicle control.The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE 51 partner repository with the dataset identifier PXD024626 and 10.6019/PXD024626.reviewer_pxd024626@ebi.ac.uk

| Data analysis
Data analysis was performed using GraphPad Prism software (Version 9).Group comparisons were made using one-way ANOVA tests, followed by a Tukey post hoc to identify differences.Where data did not meet assumptions for one-way ANOVA, a non-parametric test, Kruskal Wallis, was used.Co-treatment studies were analyzed using a two-way ANOVA (Factor 1: treatment with sodium butyrate, Factor 2: treatment with autophagy inhibitor, bafilomycin A1/3MA, or chloroquine).Flow cytometry experiments were analyzed using student t-tests.Co-treatment data were analyzed for main effects and interaction effects.Post hoc multiple comparisons were used to determine statistical significance between treatment groups.Comparisons involving just two groups (e.g., vehicle and treated group) were analyzed using a student t-test, or Wilcoxon test in cases of non-parametric data.All results presented are mean ± standard error mean (SEM) with statistical significance which is defined as *p ≤ .05.

| A SCA3 SH-SY5Y model recapitulates phenotypes found in human SCA3 patients, including the presence of protein aggregates and histone hypoacetylation
Immunoblot analysis of neuron-like (neuroblastoma, SH-SY5Y) cells stably expressing human ataxin-3 or an empty vector control revealed the expression of endogenous ataxin-3, ataxin-3-28Q, and ataxin-3-84Q at appropriate sizes (45, 50, and 65 kDa, respectively, Figure 1A) (Uncropped images found in Figure S1).Expression of ataxin-3 monomers did not differ between ataxin-3-28Q and ataxin-3-84Q cells, but as expected, was greater than in the empty vector control cells (Figure 1B).Cells expressing ataxin-3-84Q were found to contain an additional high molecular weight (HMW) ataxin-3 band, at approximately 140 kDa, which was not present in cells expressing ataxin-3-28Q (Figure 1C).We also found that these HMW bands were not present if we instead transfected cells with alternative constructs to express human ataxin-3 (data not shown).We confirmed that this HMW band contained ataxin-3 by comparing the location of the immunoreactive band of HMW ataxin-3 on the immunoblot with a Coomassie-stained 1D SDS PAGE gel that was run in parallel to the immunoblot.This region was excised, in-gel trypsin digestion was performed, and tryptic peptides were analyzed by LC-MS/ MS.We confirmed that this immunoreactive band indeed contained ataxin-3 (Figure S2, Table S1).
Immunostaining of the SH-SY5Y cells stably expressing human ataxin-3 or an empty vector control, showed expression of ataxin-3 throughout the cytoplasm and nucleus of the cell (Figure 1D).Previously, we demonstrated that SH-SY5Y cells transiently transfected with EGFPfused human ataxin-3-84Q carry Triton-X insoluble protein aggregates that can be detected by a flow cytometry protocol called FloIT. 47Upon immunostaining the stably expressing ataxin-3 cells for ataxin-3, the presence of cytoplasmic ataxin-3-positive puncta was revealed in cells expressing ataxin-3-28Q and -84Q (Figure 1D; white arrows).Automated quantification of the number of potential ataxin-3 protein aggregates (puncta with a diameter of 2.25-6 μm) within the images revealed that cells stably expressing ataxin-3-84Q had approximately double the number of puncta than cells expressing ataxin-3-28Q or an empty vector control (Figure 1E).
We have previously demonstrated that our transgenic SCA3 zebrafish exhibited alterations in acetylation of histone 3 at lysine 9 (ac-H3K9) and histone 4 at lysine 5 (ac-H4K5) compared to a non-transgenic control zebrafish. 26In this study, we sought to confirm whether this finding is consistent in the SH-SY5Y cells expressing polyQ-expanded human ataxin-3.Interestingly, we found that SH-SY5Y cells stably expressing ataxin-3-84Q had decreased ac-H3K9 compared to the ataxin-3 28Q expressing cells, while similar levels of ac-H4K5 were detected across examined genotypes (Figure S3).

| Treatment with sodium butyrate increases histone acetylation and decreases ataxin-3 aggregates in SCA3 cell culture models
To investigate the therapeutic potential of sodium butyrate (SB), we treated SH-SY5Y cells stably expressing human ataxin-3-84Q with SB (3 mM) for 72 h and analyzed whether ataxin-3-84Q expressing cells displayed increased histone acetylation following SB treatment.Immunoblotting for the ac-H3K9 and ac-H4K5 revealed that SB treatment produced a 50-and 7-fold increase in both ac-H3K9 and ac-H4K5, respectively (Figure 2A-C; Uncropped images found in Figure S4).Treatment with SB for 3 days was also found to decrease the number of ataxin-3 aggregates detected by ataxin-3 immunostaining by approximately 30% when compared to vehicletreated ataxin-3-84Q expressing cells (Figure 2D,E).Immunoblot analysis revealed that SB treatment also decreased the presence of the HMW ataxin-3 species by approximately 25% (Figure 2F,G) and produced a 1.5-fold increase in the amount of monomeric ataxin-3 species present (Figure 2H), suggesting the induction of a protein quality control pathway that is targeting HMW ataxin-3 (Uncropped images found in Figure S4).Interestingly, flow cytometric analysis of aggregates (FloIT) revealed an increase in the number of detergent insoluble GFP + particles with only 24 h of treatment (Figure S5A,B).

| Treatment with sodium butyrate increases histone acetylation and improves swimming of SCA3 zebrafish
Previously, we developed a transgenic zebrafish model of SCA3 with the intent to screen drug candidates for the treatment of SCA3. 44From this model, we found that our SCA3 zebrafish larvae expressing EGFP-tagged human ataxin-3-84Q had altered levels of histone acetylation, namely ac-H3K9 and ac-H4K5, compared to non-transgenic zebrafish. 26Since SB is a known HDAC inhibitor, EGFP-ataxin-3-84Q zebrafish were treated with SB (250 μM, 500 μM, and 1 mM) from 24 h post fertilization (hpf) to 6 days post fertilization (dpf) to determine whether levels of histone acetylation could be increased.Immunoblot analysis confirmed that 6 dpf transgenic zebrafish expressed human ataxin-3 protein of appropriate sizes (72 kDa and 84 kDa for EGFP-ataxin-3-23Q and -84Q, respectively), as well as endogenous zebrafish ataxin-3 (34 kDa; Figure 3A; Uncropped images found in Figure S6).Lower molecular weight ataxin-3-positive cleavage fragments were also present, consistent with our previous findings. 44Further, immunoblotting for ac-H3K9 and ac-H4K5 revealed a dose-dependent increase in levels of acetylated histone 3 and histone 4 with SB treatment (Figure 3A).This was confirmed when quantifying the amount of ac-H3K9 following treatment with 500 μM and 1 mM SB (Figure 3B) and ac-H4K5 following 1 mM SB (Figure 3C), compared to the vehicle-treated control.Quantification of levels of ataxin-3 revealed that there was also a 3-fold and 1.6-fold increase in the presence of fulllength (monomeric) and cleaved human ataxin-3 after treatment with both 500 μM and 1 mM SB (Figure 3D,E).
Next, we examined whether SB treatment could prevent motor impairment displayed in mutant ataxin-3 zebrafish at 6 dpf.We first examined whether treatment with these concentrations of SB would affect the survival and behavior of non-transgenic zebrafish at 6 dpf.SB treatment, irrespective of concentration, did not affect the survival or behavior of the non-transgenic zebrafish (Figure S7).Therefore, upon F I G U R E 3 Sodium butyrate treatment increases histone acetylation and rescues the motor phenotype in SCA3 zebrafish.Zebrafish expressing ataxin-3 84Q were treated with 250 μM, 500 μM, or 1 mM sodium butyrate (SB), or vehicle control, between 1 and 6 days post fertilization (dpf).(A) Protein lysates of the 6 dpf zebrafish treated either SB or the vehicle control underwent western blotting and checked for ataxin-3, acetylated histone 3 at lysine 9 (ac-H3K9), and acetylated histone 4 at lysine 5 (ac-H4K5).Histone 3 was probed as a control.(B) Quantification of ac-H3K9 revealed a dose-dependent increase following SB treatment (500 μM, p = .0481; 1 mM, p = .0014,n = 3-9), while (C) quantification of ac-H4K5 revealed 1 mM SB was able to increase histone acetylation compared to the vehicle control (p = .0018,n = 3-10).(D) Quantification of full-length (FL) ataxin-3 revealed an increase with 500 μM and 1 mM SB (p = .039and p = .0452,respectively, n = 4-5) while (E) levels of the cleavage fragment of ataxin-3 (CF) also demonstrated an increase with 500 μM and 1 mM SB treatment (p = .0452and p = .0109,respectively, n = 4-5).(F) Swimming trajectories of 6 dpf zebrafish after treatment with vehicle versus 1 mM SB treatment (Green, slow movement; red, fast movement).(G) Motor behavior analysis showed vehicle-treated ataxin-3-84Q zebrafish swam shorter distances compared to the non-transgenic controls (p = .0085)while 6 dpf ataxin-3-84Q larvae treated with 1 mM SB ameliorated the motor dysfunction (p = .0393,n = 64-158).(H) Flow cytometric analysis of the insoluble GFP + particles revealed a decrease with SB treatment compared with vehicle (p = .03,n = 4 group replicates).Data represent mean ± SEM.Statistical analysis was performed by a non-parametric one-way ANOVA (Kruskal Wallis) test and comparison between vehicle versus SB was analyzed using an unpaired student t-test.*p < 0.05, **p < 0.01.examination of the movement of the SCA3 zebrafish during the escape response to darkness motor assay revealed, as reported previously, 44 that zebrafish expressing ataxin-3 with a polyglutamine expansion swam shorter distances than non-transgenic controls (Figure 3F,G).Treatment with SB produced a dose-dependent increase in swimming distance, with 1 mM SB producing a significant rescue of the swimming of the EGFP-ataxin-3-84Q zebrafish (Figure 3G).
Flow cytometric analysis was also used to confirm whether SB treatment affected the number of insoluble EGFP-ataxin-3 aggregates.Previously, we have been able to show that zebrafish larvae expressing EGFP-ataxin-3-84Q develop protein aggregates detectable via whole-mount confocal imaging and a modified FloIT protocol, as early as 2 dpf. 47SB treatment from 1 to 6 dpf resulted in a 17% decrease in the number of insoluble EGFP-ataxin-3 particles compared to vehicle-treated controls (Figure 3H).In contrast, no decrease in insoluble EGFP-ataxin-3 particles was found after short-term SB treatment (from 1 to 2 dpf) (Figure S5C,D).

| Autophagy pathway is impaired in SCA3 SH-SY5Y cells and sodium butyrate treatment induces autophagy and aids removal of high molecular weight ataxin-3
As we detected ataxin-3 aggregates within the cells expressing ataxin-3-84Q, and SB treatment was able to reduce the number of these ataxin-3 aggregates and high molecular weight (HMW) ataxin-3 bands, we next explored whether the protein quality control pathway, macroautophagy (further known as autophagy) was activated in these cell cultures.3][54][55] We reported previously that the SH-SY5Y cells stably expressing polyQ expanded ataxin-3 have impaired autophagy dynamics. 56To examine whether the reduction of HMW ataxin-3 species was mediated by SB-induced autophagy, SH-SY5Y ataxin-3-84Q cells were co-treated with SB (3 mM) and an autophagy inhibitor, 3-Methyladenine (3MA, 5 mM), for 72 h (Figure 4A; Uncropped images found in Figure S8).While SB treatment decreased the amount of HMW ataxin-3 present compared to vehicle control by approximately 25%, co-treatment with SB and 3MA (SB+3MA) together prevented the decrease in the amount of HMW ataxin-3 compared to vehicle-treated cells (Figure 4A,B).Levels of monomeric ataxin-3 increased with 3MA treatment alone, as well as SB+3MA co-treatment by ≥1.7 fold (Figure 4C).Together, these findings indicate that 3MA prevented the capacity of SB to decrease HMW ataxin-3, and that the removal of HMW ataxin-3 was dependent on SB-mediated autophagic activity.
To examine whether SB treatment was increasing autophagic flux, immunoblots containing protein lysates of the SH-SY5Y cells expressing ataxin-3-84Q treated with either vehicle or SB and/or 3MA were probed with a marker of autophagosomes, LC3B (Figure 4A).Cells treated with SB displayed a 1.7-fold increase in the LC3II/LC3I ratio compared to vehicle controls (Figure 4D).In comparison, co-treatment with SB+3MA prevented this increase, indicating that it was dependent on the autophagy pathway (Figure 4D).In a similar manner, we found that co-treatment with SB and another autophagy inhibitor, bafilomycin (Baf A1), which acts to prevent the degradation of autophagosomes, resulted in an increase in LC3II/ LC3I ratio compared to following Baf A1 treatment alone, confirming that SB treatment increases autophagosome formation (Figure S9).

| Sodium butyrate treatment induces the autophagy pathway in the transgenic SCA3 zebrafish and increases zebrafish swimming in an autophagy-dependent manner
To confirm that autophagy was also being induced by SB treatment in vivo, the transgenic SCA3 zebrafish were treated from 1 to 6 dpf with either vehicle, SB (1 mM), an autophagy inhibitor (chloroquine, 1.5 mM) or co-treated with SB and chloroquine.Protein lysates of groups of larvae from these treatment groups were then subjected to western blot analysis and probed for LC3B (Figure 5A; Uncropped images found in Figure S10).Densitometric analysis revealed that SB treatment alone produced similar levels of LC3II to vehicle-treated larvae (Figure 5B).Levels of LC3II increased in co-treatment of SB and chloroquine compared to the non-transgenic control, SB treatment alone, and chloroquine treatment alone.The finding of increased LC3II levels following co-treatment with SB and chloroquine, when compared to chloroquine alone, indicates that autophagy is being induced upstream of the autophagy blockade produced by chloroquine.Taken together, these findings confirm that autophagy was induced in the SCA3 zebrafish by the SB treatment.
To confirm that the improved swimming phenotype of the SCA3 zebrafish produced by SB treatment was due to the induction of autophagy, we performed a co-treatment experiment followed by motor behavior analysis.The SCA3 zebrafish larvae were treated with SB (1 mM), chloroquine (1.5 mM), or both, from 1 to 6 dpf, prior to motor tracking.Vehicle-treated Ataxin-3-84Q zebrafish a shorter swimming distance compared to the nontransgenic control and treatment with SB alone again resulted in an increase in the distances swum by the SCA3 zebrafish, compared to vehicle treatment.The addition of 1.5 mM chloroquine with SB treatment prevented this improvement in motor function (Figure 5C,D).This indicates that chloroquine treatment was able to suppress the beneficial effect of SB, and therefore that the beneficial effect of SB treatment was indeed autophagy-dependent.

| Sodium butyrate-mediated autophagy induction is predicted to occur through the activation of protein kinase A and AMPK pathways
To investigate the potential mechanisms by which SB was inducing autophagy, we performed proteomic analysis to compare lysates extracted from SH-SY5Y cells expressing ataxin-3-84Q, treated for 72 h with either SB (3 mM) or vehicle control.From triplicate analysis, we identified 5143 and 5110 proteins from vehicle and treated cells, respectively, with 4966 (93.9%) proteins identified in common.We therefore identified 177 (3.3%) and 144 (2.7%) unique proteins between the vehicle and SB-treated cell lysates.There did not appear to be obvious differences in the protein classes that were identified between the vehicle and treated samples.Therefore, we carried out labelfree quantitative proteomics (LFQ) employing normalized spectral abundance factors (NSAF) to determine the relative abundance of each protein to determine a ratio between SB treatment/DMSO vehicle. 50,57e identified 176 differentially upregulated and 137 differentially downregulated proteins (p < .05) in the SB-treated cells compared to the DMSO vehicle control (Table S2).To determine cellular pathway changes upon 4 Removal of high molecular weight (HMW) ataxin-3 by sodium butyrate is mediated by autophagy.(A) Western blots were performed on protein lysates of SH-SY5Y cells stably expressing ataxin-3-84Q treated with sodium butyrate (SB, 3 mM) and/or 3MA (5 mM) for 72 h and probed for ataxin-3 and LC3B.(B) Quantification of HMW ataxin-3 indicated that SB treatment reduced the amount of HMW ataxin-3, compared to vehicle treatment (p = .0070),and addition of 3MA treatment alone or as a co-treatment prevented the removal of the HMW ataxin-3 (p = .0086and p = .0079,respectively, n = 6).(C) Quantification of the ataxin-3 monomeric species increased with 3MA treatment compared to the vehicle control (p = .0036).SB+3MA co-treatment further increased ataxin-3 levels compared to vehicle, SB treatment alone, and 3MA treatment alone (p < .0001,p < .0001,and p = .0214,respectively, n = 6).(D) Quantification of the LC3II/LC3I ratio revealed an increased ratio with SB treatment (p = 0002) and the addition of 3MA alone or SB+3MA co-treatment prevented this decrease in LC3II/LC3I (p = .0176and p = .0098,respectively; n = 6-7).Data represent mean ± SEM.Statistical analysis performed was a two-way ANOVA followed by a Tukey post hoc analysis.*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.SB treatment, Ingenuity Pathway Analysis was used to predict activation and/or inhibition of canonical pathways and processes using experimental values of proteins from LFQ. IPA predicted the activation of protein kinase A signaling (Z-score = 0.832, p-value = 3.81 × 10 −4 ) and AMPK signaling (Z-score = 0.816, p-value = 1.46 × 10 −3 , Figure 6A).
To confirm the upregulation of the AMPK pathway, AMPK protein levels were analyzed.SCA3 zebrafish treated with butyrate displayed a 2.5-fold increase in AMPK levels compared to the vehicle control (Figure 6B,C; Uncropped images found in Figure S11).Similarly, levels of ULK1 were examined as it is a protein known to initiate autophagosome formation.Specifically, phosphorylation of ULK1 at position Ser777 by AMPK has been previously described to induce autophagy. 58Western blotting detected phosphorylated ULK1 at Ser777 (pULK1) in the SB-treated Ataxin-3-84Q zebrafish and densitometric analysis revealed approximately a 5×-fold increase compared to the vehicle-treated control (Figure 6B,D).Upon immunoblotting for ULK1, we noted two bands at 110 and 120 kDa, whereby the 120 kDa band of ULK1 that was more marked in the SB-treated Ataxin-3-84Q larvae.We F I G U R E 5 Treatment with sodium butyrate induces autophagy in the SCA3 zebrafish model producing beneficial effects on zebrafish swimming.(A) Western blots were performed on protein lysates from non-transgenic controls and ataxin-3-84Q zebrafish larvae treated with either 1 mM sodium butyrate (SB) alone, 1.5 mM chloroquine alone or co-treatment of SB and chloroquine, between 1 and 6 days post fertilization (dpf).The immunoblots were probed for LC3B.(B) Quantification of LC3II levels after treatment with 1 mM SB and/or 1.5 mM chloroquine found an increase with SB and chloroquine treatment compared to chloroquine treatment alone, SB treatment alone, and the non-transgenic controls (p = .0475,p = .0395and p = .0303,respectively; n = 6-8).(C) Swimming trajectories of non-transgenic and SCA3 zebrafish treated with either vehicle control, 1 mM SB, 1.5 mM chloroquine, or SB/1.5 mM chloroquine and show that SB increases the amount of fast swimming and SB/chloroquine co-treatment decreases this (green, slow movement; red, fast movement).(D) Quantification of the total distance swum by each animal demonstrates that vehicle-treated SCA3 zebrafish has decreased swimming distance (p = .0009),SB treatment rescues the motor impairment seen in vehicle-treated ataxin-3-84Q zebrafish (p < .0001),and those receiving 1.5 mM chlor or SB/chlor co-treatment (p < .0001and p < .0001,respectively, n = 81-260).Data represent mean ± SEM.Statistical analysis was performed using a one-way ANOVA followed by a Tukey post hoc test.ns, non-significant.*p < 0.05, ****p < 0.0001.
previously demonstrated that the pULK1 band overthe 120 kDa ULK1 band, suggesting that the 120 kDa ULK1 band is likely pULK1. 56Quantification of these two ULK1 bands (relative to total ULK1) revealed greater amounts of the 120 kDa (1.3×), and decreased levels of the 110 kDa band (approximately 0.8×), in the presence of SB compared to the vehicle treatment, suggestive of autophagy induction (Figure 6D).
To test whether the induction of autophagy by SB was dependent on AMPK activity we performed a cotreatment study using SB and a reported AMPK inhibitor GSK690693.We treated SH-SY5Y cells expressing ataxin-3-84Q with bafilomycin, SB, and GSK690693 and compared to those treated with bafilomycin and SB alone.Western blot analysis revealed an increase in the amount of LC3II/I present in the triple-treated ataxin-3 28Q cells (p = .0156)and no difference in the triple-treated ataxin-3-84Q cells (p = .0547)(see Figure S9).These results suggest that although the proteomic data predicted that SB treatment induces autophagy through the activation of AMPK, genetic silencing approaches would be required to demonstrate this conclusively.

| Sodium butyrate increases histone acetylation and decreased movement deficits in SCA3 zebrafish
We recently reported that our SCA3 zebrafish expressing EGFP-ataxin-3-84Q have higher levels of acetylation histones 3 and 4 compared to non-transgenic control zebrafish at 6 days post-fertilization. 265][26] Here we found that treatment with sodium butyrate prevented this effect on acetylation, increasing acetylation of both histones 3 and 4.This increase in the presence of acetylated histones 3 and 4 is likely caused by histone deacetylase (HDAC) inhibition effects induced by SB treatment.Along with the increase in histone acetylation, we also found that SB treatment improved the swimming of SCA3 zebrafish when compared to control zebrafish.
9,31,[35][36][37][38]42 Other studies relating to the treatment of mouse models of SCA3 with SB are in alignment with our findings. 22,59The transgenic SCA3 mice developed by Chou and colleagues show a motor phenotype, neuropathology, and hypoacetylation of histone 3 and histone 4. Chronic treatment of SCA3 mice with SB restored movement performance, to similar levels to wild-type controls, with notable improvements in latency to fall off the rotarod, locomotor activity, foot dragging, and dendritic arborisation.In this study, we first identified that neuronal-like (neuroblastoma) SH-SY5Y cells expressing ataxin-3 containing a polyQ expansion (84 polyQ repeats) carried ataxin-3 protein aggregates.Interestingly, our study did not detect predominately nuclear ataxin-3 protein aggregates, which differs from some previous reports of polyQ expanded ataxin-3 resulting in nuclear protein aggregates. 60n our study, we found that 3-day treatment of SCA3 SH-SY5Y cells and 5-day treatment of transgenic SCA3 zebrafish with SB both resulted in the removal of the protein aggregates (detected by confocal imaging), decreased presence of uncharacterized HMW ataxin-3 species and decreased the number of insoluble ataxin-3 particles detected by flow cytometric (FloIT) analysis.In contrast with these findings, we did find that shorter periods of SB treatment (24 h in SH-SY5Y cells and transgenic SCA3 zebrafish) resulted in either an increase in the number of detergent-insoluble ataxin-3 particles or no significant effect.We hypothesize that the initial increase in insoluble particles observed may be a consequence of HDAC inhibition, causing increased transcription and translation of mutant ataxin-3 protein.In contrast, more prolonged treatment with SB (3-5 days) produced a robust decrease in ataxin-3 proteinopathy, demonstrating the removal and degradation of ataxin-3 protein species.In view of this, prolonged administration of SB may be required to increase autophagic activity and yield optimal therapeutic benefit.
We also identified that SB treatment was inducing autophagy, indicated by an effect of co-treatment with autophagy inhibitors, 3MA, bafilomycin A1, or chloroquine, on LC3II abundance, both in vitro and in vivo.We likewise found that the decrease in HMW ataxin-3 bands produced by SB treatment was autophagy-dependent.These findings align with a previous report by Qiao et al. 61 that SB treatment induced autophagy in enteroendocrine cells expressing αsynuclein.Combination treatment of SB with trehalose, a disaccharide, within a rodent model of Parkinson's disease revealed increased activation of autophagy alongside the reduction of pre-fibrillar form of phosphorylated αsynuclein. 62Nevertheless, our findings are the first evidence of SB removing protein aggregates in a neuronal-like cell line and in vivo, presumably through the autophagy pathway.
Interestingly, within this study, we have found positive effects of SB treatment, on zebrafish swimming, despite the increased abundance of monomeric ataxin-3 protein resulting from the treatment.This finding is in line with other studies using the same zebrafish model, wherein increased swimming has been present despite increased human ataxin-3 protein, likely due to other protective effects. 26Multiple protective mechanisms have been proposed to result from treatment with compounds with HDAC inhibitor activity.3][64] Our findings demonstrate that induction of autophagy is another therapeutic benefit of treatment with the HDAC inhibitor compound SB.][67][68][69] To understand the mechanism by which SB was inducing autophagy we performed proteomic analysis on lysates from SH-SY5Y cells expressing human ataxin-3-84Q treated with SB or vehicle control.Label-free proteomics analysis and Ingenuity Pathway Analysis predicted that SB treatment was activating protein kinase A and AMPK signaling.We found through validation by immunoblot analysis that indeed, SB treatment in the SCA3 zebrafish triggered increased levels of autophagy-related AMPK and a downstream substrate phosphor-ULK1 (Ser777).While increased AMPK activity following SB exposure has been previously reported in colorectal and bladder cancer treatment studies, [70][71][72] it has not been reported in neurodegenerative disease treatment studies.To test whether the induction of autophagy by SB was dependent on AMPK activity we performed a cotreatment study using SB and a reported AMPK inhibitor GSK690693.We did not detect a decrease in the autophagy induction produced by SB (indicated by increased LC3II in bafilomycin-treated cells), but instead found no difference or an increase in LC3II/I in treated cells, likely due to an effect of this compound on the activity of a range of other kinases.
Protein kinase A and AMPK signaling commonly affect similar proteins such as FOXO1.Increased activation of FOXO1 is known to induce autophagy following treatment with HDAC inhibitors by increasing transcription of autophagy-related genes (Atg4b, Atg12, Pik3c3, Becn1, and Map1lc3b). 73It has been reported that AMPK activates FOXO1 via phosphorylation of site Thr694. 74nterestingly, the FOXO family of transcription factors has been reported to have pro-longevity effects 75,76 and are reported to be downregulated with aging, 77 suggesting a potential benefit of FOXO1-related therapeutics in age-linked neurodegenerative diseases. 78Most recently, it has been demonstrated that treatment with ketone body, D-βhydroxybutyrate, can activate the autophagy pathway through the AMPK pathway and activate transcription factors FOXO1 and FOXO3a in healthy rat cortical neurons. 79Previously, overexpression of DAF-16, the equivalent gene of FOXO in C. elegans, prevented motor impairment in a C. elegans model of SCA3. 80From these studies, alongside our results, the relationship between AMPK-mediated autophagy induction and downstream activation of FOXO transcription factors warrants further investigation as a potential target for SCA3 treatment.
[55]81,82 Our team recently found impaired autophagy dynamics within SH-SY5Y cellular and transgenic zebrafish models of SCA3. 56Collectively, these factors suggest that deficits in autophagy function may contribute to ataxin-3 proteinopathy.Several studies have also demonstrated that induction of the autophagy pathway, via genetic modification 11,52,81,83 or exposure to small compounds, 44,[84][85][86][87][88][89][90] can have beneficial effects as a treatment of SCA3 models.For example, trehalose has been tested across in vitro and in vivo models of SCA3 and has shown to decrease ataxin-3 positive aggregates, 91 increase cerebellum layer thickness, 91 and improve motor function. 56,91Moreover, these pre-clinical studies led to a clinical trial for safety and efficacy in SCA3 patients with minor adverse events and stabilization of disease progression. 92he findings within this study have relevance to a broad range of neurodegenerative diseases as there is growing evidence to suggest that changes to the gut microbiome, and therefore changes to levels of metabolites released by the gut, such as butyric acid, may impact the function of the nervous system. 28,61,93,94This is because, while these metabolites are naturally produced in the gut, they may elicit effects in the CNS through either the gutbrain axis or through crossing the blood-brain barrier from the circulation. 28,61Our team has recently investigated whether the gut microbiome is altered in a SCA3 mouse model and found that indeed SCA3 mice have an altered gut microbiota structure and composition, including a decreased abundance of some populations of butyrate-producing bacteria. 95Therefore, butyrate-related treatments, or exploiting the gut microbiome to increase butyrate production, may have therapeutic potential for SCA3.While not explored within a SCA3 model, a recent study found that placing healthy mice on a ketogenic (low carbohydrate, high fat) diet resulted in increased levels of D-βhydroxybutyrate in the blood and activation of the autophagy pathway. 79Further, combined treatment of taurursodiol, a soluble form of bile acid, and phenylbutyrate, a derivative of butyric acid, has been found to improve survival in clinical cohorts of ALS patients 41 and received FDA approval for the treatment of ALS.
In contrast with our findings, previous studies investigating the therapeutic potential of SB in neurodegenerative models, including a mouse model of Huntington's disease 42 and an in vitro model of Parkinson's disease, 61 found no effect of SB treatment on the number of protein aggregates present within those models.However, one report of butyrate treatment of a SOD1G93A mouse model of amyotrophic lateral sclerosis did report a reduction in the number of SOD1-positive aggregates in the colon, compared to vehicle-treated animals. 96Despite this, the authors did not report whether SOD1 protein aggregates were reduced within neurons, or whether the autophagy pathway was induced by SB treatment.One explanation for why this finding has not been reported previously is that the optimal dose (and dosing frequency) required for SB to induce autophagy may fall within a relatively narrow range, or it may vary depending on specific protein quality control impairments induced by the individual gene mutations.Doses reported within this study varied from previous doses reported to ameliorate neuropathology in mouse model of SBMA, 23 or in a mouse model of Huntington's disease. 42Further studies are required to confirm the optimal dosage at which SB can induce degradation of ataxin-3 protein aggregates in neurons, as the dosage required may vary between different experimental animal models of SCA3.

| Concluding remarks
Sodium butyrate treatment was found to affect the aggregation status of polyQ-expanded ataxin-3 in vitro, decreasing the presence of high molecular weight (HMW) ataxin-3 bands and enhancing the degradation of ataxin-3 positive aggregates.This study also provides the first experimental evidence of sodium butyrate treatment inducing increased autophagic activity within a model of neurodegenerative disease.Induction of the autophagy protein quality control pathway may be a therapeutic option for neurodegenerative diseases due to the potential to increase the clearance and degradation of toxic protein species such as mutant ataxin-3 protein.It must also be considered that there may also be limitations to the benefit of autophagy-inducing candidates for patient treatment.A previous trial of treatment with lithium carbonate, known to induce autophagy, did not report positive outcome measures. 97e propose that our finding that treatment with SB can induce autophagy, both in vitro and in vivo, resulting in protective effects such as preventing the development of impaired movement is important for research into treatments for SCA3.These results indicate that sodium butyrate and butyric acid both warrant further investigation as autophagy-inducing agents with therapeutic potential for a wide range of neurodegenerative and protein aggregation diseases.

F I G U R E 6
Proteomic analysis reveals upregulation of the Protein Kinase A and AMPK signaling pathway with SB treatment.(A) Ingenuity pathway analysis (IPA) predicted activation of the Protein Kinase A and AMPK signaling pathway upon SB treatment on ataxin-3-84Q expressing SH-SY5Y cells.Blue indicates IPA predicted an inhibition versus orange refers to an activation.(B) Immunoblots of ataxin-3-84Q zebrafish protein samples treated with SB or a vehicle control.Blots were probed with AMPK, phosphorylated ULK1 (Ser777) and ULK1.(C) Quantification revealed increased AMPK levels with SB treatment compared to vehicle-treated (p = .0209,n = 8).(D)Densitometric analysis revealed an increased pULK1/ULK1 ratio in the SB treatment compared to the vehicle-treated SCA3 zebrafish (p = .0469,n = 7).Quantification of ULK1 revealed increased ULK1 protein at 120 kDa relative to total ULK1 and decreased 110 kDa relative to total ULK1 in SB-treated SCA3 zebrafish compared to the vehicle control (p = .0078and p = .0156,respectively; n = 8).Data represent mean ± SEM.Statistical analysis performed was a paired non-parametric t-test (Wilcoxon test).*p < 0.05, **p < 0.01.

4. 2 |
Sodium butyrate increases the activity of the autophagy protein quality control pathway