Identifying Disease Signatures in the Spinocerebellar Ataxia Type 1 Mouse Cortex

The neurodegenerative disease spinocerebellar ataxia type 1 (SCA1) is known to lead to the progressive degeneration of specific neuronal populations, including cerebellar Purkinje cells (PCs), brainstem cranial nerve nuclei and inferior olive nuclei, and spinocerebellar tracts. The disease-causing protein ataxin-1 is fairly ubiquitously expressed throughout the brain and spinal cord, but most studies have primarily focused on the role of ataxin-1 in the cerebellum and brainstem. Therefore, the functions of ataxin-1 and the effects of SCA1 mutations in other brain regions including the cortex are not well-known. Here, we characterized pathology in the motor cortex of a SCA1 mouse model and performed RNA sequencing in this brain region to investigate the impact of mutant ataxin-1 towards transcriptomic alterations. We identified progressive cortical pathology and significant transcriptomic changes in the motor cortex of a SCA1 mouse model. We also identified progressive, region-specific, colocalization of p62 protein with mutant ataxin-1 aggregates in broad brain regions, but not the cerebellum or brainstem. A cross-regional comparison of the SCA1 cortical and cerebellar transcriptomic changes identified both common and unique gene expression changes between the two regions, including shared synaptic dysfunction and region-specific kinase regulation. These findings suggest that the cortex is progressively impacted via both shared and region-specific mechanisms in SCA1.


Introduction
Spinocerebellar ataxia type 1 (SCA1) is an inherited neurodegenerative disorder characterized by progressive gait impairment, limb incoordination, respiratory dysfunction, cognitive impairment, and memory decline [1][2][3][4][5][6][7]. SCA1 is a monogenic disorder caused by a CAG repeat expansion in the gene ATAXIN-1 (ATXN1), resulting in a polyglutamine (polyQ) expansion in the ataxin-1 protein [8]. Interestingly, the disease-causing gene ATXN1 is fairly ubiquitously expressed throughout the central nervous system [9], but is known to lead to pronounced neuron loss and pathology in specific brain regions, including the cerebellum and brainstem [10][11][12]. However, pathology and molecular disease signatures in other brain regions have not yet been fully understood.
The pronounced cerebellar Purkinje cell (PC) loss and ataxic gait impairment have led to impactful SCA1 research within the cerebellum and brainstem. Despite large focus in these regions, analysis of SCA1 patient brains by magnetic resonance imaging (MRI) and post-mortem imaging has identified broad degeneration and pathology in both cerebellar and extra-cerebellar regions, including the thalamus, striatum, and frontal, temporal,

Cell Culture
Neuro-2a (N2A; ATCC, CCL-131) cells were plated at 5 × 10 4 cells per well in a 24 well plate with a cover slip 24 h prior to transfection, maintained in a 37 • C, 5.5% CO 2 incubator and cultured in DMEM (Gibco) with 10% FBS, 100 U/mL penicillin, and 100 U/mL streptomycin. When 70-80% confluent, media was replaced with antibiotic free DMEM with 10% FBS, and cells were transiently transfected with 0.5 ug DNA, 1 uL P3000 enhancer reagent (Invitrogen, L3000015, Waltham, MA, USA), and 1 uL Lipofectamine 3000 (Invitrogen, L3000015) in 50 uL Opti-MEM (Gibco). After 48 h, cells were washed twice with PBS and subsequently fixed with 4% PFA for 10 min. Cells were washed two more times with PBS, prior to blocking for 1 h in 10% normal goat serum and 0.3% Triton-X in PBS. Cells were then incubated at 4 • C in primary antibody in blocking buffer overnight (chicken anti-DDDDK (Abcam, ab1170, 1:400, Cambridge, UK), mouse anti-p62 (SQSTM1) (Abnova, H00008878-M01, 1:700, Taipei City, Taiwan)). The following morning, cells were washed three times in PBS prior to incubation in secondary antibody in blocking buffer for two hours at room temperature (goat anti rabbit-488 and goat anti mouse-555 (Invitrogen, 1:500)). Following incubation, cells were washed three times in PBS and mounted on slides using Vectashield mounting media and DAPI (Vector Laboratories, H-1500, Newark, NJ, USA). Fluorescent images were acquired on a Zeiss LSM880 confocal microscope at 20× magnification, with the same settings used across similar experiments. Images were acquired and quantified for 50 cells per condition.

RNA Extraction and Bulk RNA Sequencing
Whole mouse cerebella and cortex were macro-dissected, flash frozen, and stored at −80 • C until processing. RNA was extracted using the Qiagen RNeasy Mini Kit, and DNA was removed with DNase I as described in the manufacturer's protocol. Total RNA was sent to the Yale Center for Genome Analysis (YCGA) for processing. RNA integrity was measured with capillary electrophoresis (Agilent BioAnalyzer 2100, Agilent Technologies, Santa Clara, CA, USA) and RIN values were checked to ensure RNA integrity before proceeding with RNA-seq. Libraries were generated using oligo-dT purification of poly-adenylated RNA, which was then reverse transcribed into cDNA that was fragmented, blunt ended, and ligated to adaptors. Base pair size was determined with capillary electrophoresis, and the library was quantified before pooling and sequencing on a HiSeq 2500 using a 75 bp paired-end read strategy. Sequencing data can be accessed under GEO accession # (GSE211678).

RNA Sequencing Data Analysis
STAR (Spliced Transcripts Alignment to a Reference) [30] was utilized to align FASTQ reads to mouse reference genome (Mm10). FeatureCounts [31] was used to quantify read counts per gene in Python 3.10. The obtained read counts were used for subsequent quantification and differential expression analysis, completed using the DESeq2 package [32] in the R environment. Gene expression with an FDR-adjusted p-value < 0.05 was considered significant, and genes with a Log2FoldChange greater than |0.25| were used in the final analyses. The Pheatmap (RRID:SCR_016418) and BioVenn [33,34] packages were used to generate all heatmaps and Venn diagrams, respectively, in the R environment (Version 4.1.3, the r-project). Normalized read counts were used as heatmap input. Intervene [35] was used to generate Upset plots. Gene Ontology analysis was conducted using g: Profiler [36,37]. Plots were generated using GraphPad Prism 9 (Version 9.3.1, GraphPad Software, San Diego, CA, USA) and Python 3.10 (Python Software Foundation, Wilmington, DE, USA).

Fluorescent Image Quantification
ImageJ (National Institutes of Health) was used for all image processing and quantification. For all images across experiments, the maximum intensity projection was utilized, converted to 8-bit, and thresholded using identical parameters across all images of each experiment. ROI manager was used to measure equal areas across images. To quantify cortical thickness, the length of the cortical layer was measured at three locations in the image approximately 100 µm apart, from layer 1 to layer 6 based on NeuN stain, and averaged per image. To quantify neuron number and size, clumped nuclei were split using the watershed function, and only NeuN-positive nuclei greater than 30 µm 2 were counted. To quantify microglia number and intensity, only Iba1-positive cells greater than 15 µm 2 were counted, and intensity was recorded as mean gray value. To quantify p62-positive puncta per mm 2 , only puncta greater than 3 µm 2 were counted. To quantify colocalization of p62-positive puncta with ataxin-1 nuclear inclusions (ataxin-1 [11NQ] antibody-specific), the ImageJ plugin JaCoP was used [39]. Images were cropped to contain a single cell of interest and thresholded to remove background intensity. Percent colocalization was then calculated using the Manders' overlap coefficient. For all quantifications of mouse tissue, 3-6 images were quantified per mouse, and for quantifications of in vitro cells, 50 cells per condition were quantified. GraphPad Prism was used to plot and analyze all fluorescent image data.

Statistical Analysis
All statistical analyses were completed using GraphPad Prism software. All data are shown as mean ± SEM. When comparing two experimental groups, a two-tailed unpaired student's t tests was performed. When comparing more than two experimental groups, a one-way analysis of variance (ANOVA) with Tukey's post hoc analysis was performed. Statistical significance was determined using a significance cutoff of p < 0.05.

Progressive Cortical Thinning in the SCA1 KI Mouse Model
In order to identify region-specific disease signatures between the SCA1 motor cortex and cerebellum, we first aimed to broadly investigate the degree in which the cortex might be affected or altered in mouse models of SCA1. It is known that ataxin-1 is expressed in the cortex [9], and that polyQ-expanded mutant ataxin-1 aggregates in nuclear inclusions across brain regions in SCA1 mouse models, including the cortex [10]. However, the functional impacts of this expression and aberrant aggregation in extra-cerebellar regions are not fully understood. We analyzed SCA1 KI mouse brain tissue at 12 and 30 weeks, as these are known to represent mid-and late stages of SCA1 disease progression in the mouse cerebellum, and are time points at which motor and cognitive impairment is observed in SCA1 KI mice [10,40,41]. At 12 weeks, there is significant astro-and microgliosis observed in the cerebellum [40], as well as motor and cognitive behavioral deficits, including impaired spatial learning by Morris Water Maze, and impaired learning and memory by conditioned fear response [10,41]. At 30 weeks, these phenotypes are more severe, and cerebellar PC loss is observed as well [10,41].
As cortical thinning has been observed in studies of SCA1 patients [12,42], we first looked at whether this was replicated in the SCA1 KI mouse model. We confirmed progressive cortical thinning in the SCA1 KI mouse motor cortex, with trending changes at 12 weeks (early stage of disease progression), and a significant decrease at 30 weeks (late stage of disease progression), compared to WT littermate controls ( Figure 1A,B). To identify whether this may be due to neuronal loss or atrophy, we characterized the density and size of NeuN-positive neuronal nuclei in the mouse cortex. We observed no significant neuronal loss at 12 or 30 weeks ( Figure 1A,C), but did observe a progressive decrease in size of neuronal nuclei, with a significant decrease at 30 weeks in both layers 2/3 (L2/3) and layer 5 (L5) of the SCA1 KI motor cortex ( Figure 1A,D). This could suggest that the observed cortical thinning may be due to atrophy, not necessarily death, of cortical neurons. However, this does not necessarily exclude the possibility of neuronal subpopulations being differentially affected in the cortex.
Next, since gliosis is reported in the SCA1 KI cerebellum at these time points [40], we investigated gliosis in the motor cortex. We observed no significant changes in microgliosis in Iba1-positive populations at 12 or 30 weeks ( Figure 1E-K), although we did observe a trending increase in Iba1-positive microglia number at 12 weeks ( Figure 1E,G). We also did not observe any significant changes in protein levels of astrocyte marker Gfap at 12 weeks ( Figure 1J,L). Interestingly, we previously analyzed gliosis in the SCA1 KI mouse cerebellum and inferior olive, and observed significant astro-and micro-gliosis at 12 weeks in the inferior olive [27]. This suggests that there are differential patterns of gliosis across brain regions in SCA1. As we only analyzed a subset of astrocyte and microglia markers in this study, this does not necessarily exclude the possibility of glial subpopulations being differentially impacted in the SCA1 cortex. Taken together, these data suggest that the cortical thinning observed in SCA1 patients is replicated in an age-dependent manner in the SCA1 KI mouse model, and that this thinning may in part be due to neuronal shrinkage or atrophy. Interestingly, the gliosis observed in the SCA1 KI cerebellum and inferior olive at 12 and 30 weeks [27,40] was not significantly changed in the SCA1 KI cortex at these same time points.

Progressive, Region-Specific, PolyQ-Dependent Colocalization of p62 with Mutant Ataxin-1 Nuclear Inclusions
Next, we were interested in investigating the ataxin-1 nuclear inclusions in the cortex. Aggregations of disease-causing proteins are observed across diverse neurodegenerative disorders [15][16][17][18][19][20][21][22]; however, the impact of these aggregates towards disease progression is still not fully understood. In SCA1, ataxin-1-positive aggregates are observed across diverse brain regions, although less so in the cerebellum [10]. Here, we imaged ataxin-1 aggregates in the motor cortex and observed that these aggregates colocalized with the autophagy/proteaseome marker p62 (Figure 2A-C). Interestingly, p62 aggregates were progressively increased in the SCA1 cortex, with significant increase compared to WT littermate controls at both 12 and 30 weeks (Figure 2A-C). We also found that colocalization of p62-positive aggregates and ataxin-1 nuclear inclusions was observed in other brain regions of SCA1 mice, including the striatum and hippocampus, with minimal to no colocalization in the SCA1 brainstem and cerebellum at 30 weeks ( Figure S1). Given the role of p62 in autophagy and proteasomal degradation of ubiquitinated proteins [43], as well as previous literature suggesting that p62 may aid in degrading aggregated polyQexpanded ataxin-1 in vitro [44], we were interested if this colocalization occurred in a polyQ-dependent manner. To investigate this, we transfected N2A cells with 2Q, 30Q, and 82Q human ATXN1 (Flag-hATXN1) plasmids ( Figure 2D). After 48 h, 44.3% of N2A cells transfected with hATXN1 [82Q] formed mutant ataxin-1 nuclear inclusions colocalized with endogenous p62, in contrast with 9.9% and 11.3% of N2A cells transfected with hATXN1 [2Q] and −[30Q] having p62-positive ataxin-1 nuclear inclusions, respectively (n = 50 cells per condition) ( Figure 2D,E). This significant increase in hATXN1 [82Q] colocalization with p62 in nuclear inclusions, along with the observed increase in p62 aggregates in SCA1 cortex in vivo, suggests that this colocalization occurs in a polyQ-dependent and region-specific manner.

Transcriptomic Alterations in the SCA1 KI Mouse Cortex
Given these observed morphological changes in the SCA1 KI mouse cortex, we were interested in investigating potential molecular mechanisms in which mutant ataxin-1 impacts the cortical transcriptome, as the role of ataxin-1 as a transcriptional regulator, and the impact of this towards the transcriptional signature of the cerebellum and inferior olive, has been well established [8,[24][25][26][27]45]. We performed bulk RNA sequencing of the SCA1 KI mouse cortex at 12 weeks, and observed a high number of up-and downregulated genes in the SCA1 cortex compared to WT littermate controls (1853 downregulated differentially expressed genes (DEGs), 819 upregulated DEGs) ( Figure 3A,B, Table S1).

Transcriptomic Alterations in the SCA1 KI Mouse Cortex
Given these observed morphological changes in the SCA1 KI mouse cortex, we were interested in investigating potential molecular mechanisms in which mutant ataxin-1 impacts the cortical transcriptome, as the role of ataxin-1 as a transcriptional regulator, and the impact of this towards the transcriptional signature of the cerebellum and inferior olive, has been well established [8,[24][25][26][27]45]. We performed bulk RNA sequencing of the SCA1 KI mouse cortex at 12 weeks, and observed a high number of up-and downregulated genes in the SCA1 cortex compared to WT littermate controls (1853 downregulated differentially expressed genes (DEGs), 819 upregulated DEGs) ( Figure 3A,B, Table S1). Genes with an FDR-adjusted p-value < 0.05 and a log 2 foldchange (LFC) greater than |0.25| were considered significant and used for further analyses. The top Gene Ontology (GO) terms for the downregulated DEGs were largely related to phosphorylation and kinase regulation, calcium channel activity, synaptic organization and function, and dendrite development, etc. (Figure 3C,D, Table S2). In contrast, the top GO terms for the upregulated DEGs were mainly related to transcriptional regulation and chromatin binding ( Figure S2, Table S3). Together, these data support that mutant ataxin-1 expression is significantly impacting the cortical transcriptome in the SCA1 KI mice.

Cross-Tissue Comparison of the Cortex and Cerebellum Identifies Shared and Unique Transcriptomic Response to SCA1 Disease
Given the significant degree of transcriptomic alterations in the SCA1 KI mouse cortex, we were interested in seeing how this compares to that of the SCA1 KI mouse cerebellum. To do this, we reanalyzed our lab's previously published cerebellar transcriptomic dataset [27] and compared this to the cortical transcriptomic dataset (Figure 4). Surprisingly, we observed a larger degree of DEGs in the cortex at 12 weeks compared to the cerebellum, with 2672 DEGs in the cortex (819 upregulated, 1853 downregulated), and 1154 DEGs in the cerebellum (232 upregulated, 922 downregulated) at this time point ( Figure 4B, Table S1). Of the upregulated DEGs, there was a 20.26% or 5.74% (n = 47) overlap of the cerebellar DEGs (n = 47/232) with cortical DEGs (47/819), respectively ( Figure 4C,D). For the downregulated DEGs, this overlap was 30.37% or 15.11% (n = 280) of cerebellar DEGs (n = 280/922) with cortical DEGs (n = 280/1853), respectively, suggesting these genes are commonly altered across these brain regions, with the remainder being unique to each region ( Figures 4C-E and S3). Interestingly, there were a subset of DEGs that behaved in an opposing manner between these two brain regions, showing upregulation in the cortex but downregulation in the cerebellum (n = 54), and vice versa (n = 45) ( Figure 4C). Together, these data suggest that mutant ataxin-1 impacts gene expression in both the cerebellum and cortex, with a larger degree of transcriptomic changes observed in the SCA1 KI mouse cortex compared to the cerebellum at 12 weeks.
Next, we were interested in understanding what biological and molecular pathways were commonly or uniquely altered across the cerebellum and cortex, as this may suggest common or distinct disease signatures across brain regions. Comparison of the SCA1 KI mouse cerebellar and cortical transcriptomic changes at 12 weeks identified both shared and unique GO hits ( Figures 4F and S4-S6, Tables S4 and S5). The top shared GO hits that were downregulated in both the cerebellum and cortex related to synaptic transmission, cell to cell signaling, and nervous system development, etc. (Figures 4F, S4 and S5A, Table S4). This suggests that synaptic dysfunction is a shared common disease signature in SCA1 KI mice across these regions at 12 weeks. In contrast, the top shared GO hits that were upregulated in both the cerebellum and cortex included system development, neuron differentiation, and regulation of macromolecule metabolic process, etc. ( Figure S6A, Table S5).
We also identified region-specific differences in transcriptional alterations. We found that the top GO hits for downregulated DEGs that were unique to the cerebellum related to G-protein coupled receptor signaling, dopamine response, and serotonin receptor signaling, etc. (Figures 4F, S4 and S5B, Table S4). In contrast, kinase regulation and phosphorylation were consistently downregulated in the SCA1 KI mouse cortex, but unchanged in the SCA1 cerebellum ( Figures 4F, S4 and S5C, Table S4). Among the upregulated DEGs, cell morphogenesis and transcription regulation were specific to the SCA1 KI mouse cerebellum and the cortex, respectively ( Figure S6B,C, Table S5). Given previous literature investigating kinases in SCA1 [46][47][48], including region-specific targeting of kinases as a potential effective therapeutic approach [48], we further investigated this change in gene expression. We confirmed a significantly altered expression of serine-threonine kinases in the SCA1 KI mouse cortex that was not observed in the cerebellum ( Figure 5A). We validated that this downregulation was observed only in the SCA1 KI mouse cortex and not the cerebellum at 12 weeks for a subset of kinases, specifically Camkk2 and Prkcb, as these were among the top 10 differentially expressed kinases in the cortex ( Figure 5B-D). Taken together, these data suggest that there are region-specific and shared cross-regional molecular changes occurring in the SCA1 KI mouse brain (Figures 4-6, S4 and S5).

Discussion
In an effort to further understand the functional impact of mutant ataxin-1 expr in broad brain regions in SCA1, we characterized cortex-specific changes in the SC mouse model as this region has been largely understudied, and performed a cro gional comparison of the transcriptomic changes between brain regions of the corte cerebellum. We identified region-specific pathological and molecular changes in including progressive cortical thinning, protein aggregation and colocalization o with ataxin-1 nuclear inclusions, as well as common and cortex-specific transcrip alterations, including regulation of kinase activity.
Importantly, we observed progressive cortical shrinkage in the SCA1 KI mous tor cortex, which is consistent with the cortical thinning observed in human SCA1 pa [12,42]. Our data suggest that this shrinkage may be due to neuronal atrophy, rathe significant neuronal loss even at 30 weeks in SCA1 KI mice. Further research investi the progression of this phenotype at later time points would be insightful, as well vestigation into greater detail of how unique cortical neuron subtypes might be dif tially affected during SCA1 disease initiation and progression, as certain population have differential responses to mutant ataxin-1 expression compared to others. Furth did not observe significant astro-or micro-gliosis in the SCA1 motor cortex at 12 weeks, although we saw a trending increase in microglia density at 12 weeks. In ingly, gliosis is observed in the SCA1 KI mouse cerebellum and inferior olive at thes points [27,40]. Region-specific differences in gliosis [27], and the functional impact o sis towards neuronal health and pathology is an interesting area of future study worth mentioning that our data cannot exclude the possibility of glial subtypes bein ferentially affected in the cortex, as our findings are based on pan-astrocyte (Gfap microglia (Iba1) markers. Additionally, in this study we investigated the SCA1 m motor cortex, primarily layers 2/3 and 5. Thus, it is possible that other cortical area layers may have diverse or different responses to mutant ataxin-1 expression.
We observed differential region-specific aggregation and colocalization of p6 ataxin-1 nuclear inclusions. Colocalization of p62 aggregates with ataxin-1 nuclear sions was observed in the SCA1 KI mouse cortex, hippocampus, and striatum, but the cerebellum or brainstem. Interestingly, this colocalization occurred in a poly pendent manner in N2A cells in vitro. The functional impact of this colocalization interest, especially given that the regulation and role of p62 has been quite well stud

Discussion
In an effort to further understand the functional impact of mutant ataxin-1 expression in broad brain regions in SCA1, we characterized cortex-specific changes in the SCA1 KI mouse model as this region has been largely understudied, and performed a crossregional comparison of the transcriptomic changes between brain regions of the cortex and cerebellum. We identified region-specific pathological and molecular changes in SCA1, including progressive cortical thinning, protein aggregation and colocalization of p62 with ataxin-1 nuclear inclusions, as well as common and cortex-specific transcriptomic alterations, including regulation of kinase activity.
Importantly, we observed progressive cortical shrinkage in the SCA1 KI mouse motor cortex, which is consistent with the cortical thinning observed in human SCA1 patients [12,42]. Our data suggest that this shrinkage may be due to neuronal atrophy, rather than significant neuronal loss even at 30 weeks in SCA1 KI mice. Further research investigating the progression of this phenotype at later time points would be insightful, as well as investigation into greater detail of how unique cortical neuron subtypes might be differentially affected during SCA1 disease initiation and progression, as certain populations may have differential responses to mutant ataxin-1 expression compared to others. Further, we did not observe significant astro-or micro-gliosis in the SCA1 motor cortex at 12 or 30 weeks, although we saw a trending increase in microglia density at 12 weeks. Interestingly, gliosis is observed in the SCA1 KI mouse cerebellum and inferior olive at these time points [27,40]. Region-specific differences in gliosis [27], and the functional impact of gliosis towards neuronal health and pathology is an interesting area of future study. It is worth mentioning that our data cannot exclude the possibility of glial subtypes being differentially affected in the cortex, as our findings are based on pan-astrocyte (Gfap) and microglia (Iba1) markers. Additionally, in this study we investigated the SCA1 mouse motor cortex, primarily layers 2/3 and 5. Thus, it is possible that other cortical areas and layers may have diverse or different responses to mutant ataxin-1 expression.
We observed differential region-specific aggregation and colocalization of p62 with ataxin-1 nuclear inclusions. Colocalization of p62 aggregates with ataxin-1 nuclear inclusions was observed in the SCA1 KI mouse cortex, hippocampus, and striatum, but not in the cerebellum or brainstem. Interestingly, this colocalization occurred in a polyQ-dependent manner in N2A cells in vitro. The functional impact of this colocalization is of interest, especially given that the regulation and role of p62 has been quite well studied in other neurodegenerative disease, including Alzheimer's disease (AD) [49], Parkinson's disease (PD) [50,51], Huntington's disease (HD) [52], amyotrophic lateral sclerosis (ALS), spinocerebellar ataxias (SCAs) [44,53], as well as various cancers [54][55][56]. P62, encoded by SQSTM1, is a scaffold protein known to promote degradation of ubiquitinated proteins, and therefore thought to have protective functions by promoting clearance of protein aggregates [54,57]. Therapeutic compounds targeting autophagy, including rapamycin and metformin [54,[58][59][60][61][62][63][64][65], have shown some promise in neurodegenerative diseases. Rapamycin that inhibits mTOR and enhance autophagy has been reported to be beneficial in models of AD [59][60][61], HD [62][63][64], but detrimental in ALS models [65]. In SCA1 specifically, an in vitro study found that p62 played a positive role in regulating clearance of mutant ataxin-1 protein [44]. In addition to autophagy, p62 is also known to play roles in nucleo-cytoplasmic shuttling, regulation of inflammation, and oxidative stress [54,55,66,67]. Given this wide range of functions of p62, and its progressive, region-specific distribution in the SCA1 KI mouse model, further investigation into the role of p62 in SCA1 progression would be of interest.
Our bulk RNA sequencing of the whole SCA1 KI mouse cortex was performed at 12 weeks, as this is considered to be an early to-intermediate stage of disease progression in SCA1 KI mice, with motor and cognitive impairments observed, as well as gliosis and nuclear inclusions, but no major cellular loss in the cerebellum [10,27,40,41]. We selected an early to-intermediate stage rather than an end-stage of disease progression in order to avoid any potential distorted transcriptional alterations that may be due to substantial cell loss. Further, this 12-week time point aligns with previously generated bulk RNA sequencing data of the SCA1 cerebellum [27], and allowed for a cross-regional comparison between these two brain regions. Our bulk RNA sequencing analyses of the whole SCA1 KI mouse cortex identified a large number of transcriptomic changes occurring in the cortex at 12 weeks, including hits relating to synapse function, transcription regulation, and kinase activity, etc. The functional impact of these changes, whether they are pathogenic, protective, or secondary to disease progression still requires further investigation. Interestingly, in our cross-regional comparison between the cortex and cerebellum, we observed a larger degree of transcriptomic changes in the cortex compared to the cerebellum, with a subset of genes overlapping across brain regions. This suggests that there are significant changes occurring in the SCA1 KI cortex, and how these changes are involved in disease progression is an important area of further investigation. To understand how these two brain regions might be differentially or similarly impacted in SCA1, we compared the top GO hits across cortex to cerebellum, and observed shared GO hits relating to synaptic transmission and cell to cell signaling, suggesting that these may be commonly affected across brain regions in SCA1. In contrast, dysregulation, in particular downregulation, of serine-threonine kinases was observed in the SCA1 KI cortex, with no significant changes in the cerebellum. The functional impact of this remains unknown; however, it is of particular interest given previous research in SCA1, showing inhibition or reduction in several kinases to be sufficient to rescue pathology in the cerebellum and brainstem [46][47][48]. Once again, it remains unclear at this moment what the functional impact of kinase reduction in the cortex is; for example, whether this is pathogenic, compensatory, or neutral to SCA1 pathology.
Lastly, our present study describes pathological and molecular changes in the SCA1 KI mouse cortex and clearly suggests that the cortex is indeed affected in this SCA1 mouse model. Importantly, SCA1 patients also experience a wide range of symptoms related to cortex. These include cognitive impairment and memory decline, specifically executive dysfunction, impaired verbal memory, and decreased visuospatial performance, which tends to correlate with ataxia symptoms [4][5][6][7]. The functional impact of these cortex-specific and shared molecular or pathological changes, and how these may contribute towards clinical symptoms in patients, remain an important area of future studies, as this may aid in developing more effective therapeutic targets to improve patient outcomes.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/cells11172632/s1, Figure S1: Supplement to Figure 2, Colocalization of p62 with ataxin-1 nuclear inclusions across SCA1 brain regions; Figure Table S1: Supplement to Figures 3 and 4, DEG list of the SCA1 KI mouse cortex and cerebellum at 12 weeks; Table S2: Supplement to Figure 3, Gene Ontology terms of downregulated DEGs in the SCA1 KI mouse cortex at 12 weeks, generated using g:Profiler; Table S3: Supplement to Figure 3, Gene Ontology terms of upregulated DEGs in the SCA1 KI mouse cortex at 12 weeks, generated using g:Profiler; Table S4: Supplement to Figure 4, Shared and unique Gene Ontology (GO) terms of the SCA1 KI mouse cortex and cerebellum at 12 weeks for downregulated DEGs in each region, generated using g:Profiler; Table S5: Supplement to Figure 4, Shared and unique Gene Ontology (GO) terms of the SCA1 KI mouse cortex and cerebellum at 12 weeks for upregulated DEGs in each region, generated using g:Profiler.