Wild-type C9orf72 drives proteasomal dysfunction and mutant aggregates via a Stat1-Isg15 axis in Huntington’s disease

Abstract

Mutant C9orf72 has been extensively studied as a major genetic cause of amyotrophic lateral sclerosis and frontotemporal dementia, and is also known to generate Huntington’s disease (HD)-like phenocopies. However, despite this strong disease association, the role of wild-type C9orf72 (wt-C9orf72) in neurodegeneration remains largely unexplored. HD is a neurodegenerative disease, and characterized by the accumulation of misfolded mutant Huntingtin (mHTT) and impaired proteostasis, yet the upstream mechanisms driving ubiquitin-proteasome system (UPS) dysfunction are not fully understood. Here, we identify a previously unrecognized modulatory role of wt-C9orf72 in regulating mHTT aggregation in experimental HD models. Analysis of public transcriptomic datasets reveal context-dependent C9ORF72 expression changes across HD-related human datasets, while C9orf72 levels are increased in R6/2 mouse brain lysates. Functional analyses reveal that overexpression of wt-C9orf72 increases mHTT aggregation and is accompanied by increased apoptotic signaling and reduced cell viability. Unbiased proteomic profiling identifies Stat1 as a key downstream effector. Mechanistically, wt-C9orf72 promotes Stat1 activation and nuclear translocation, leading to transcriptional upregulation of Isg15, a ubiquitin-like modifier. Elevated Isg15 disrupts UPS function, resulting in accumulation of polyubiquitinated proteins and impaired proteasomal degradation. Importantly, genetic suppression of Stat1 or Isg15 significantly attenuates C9orf72-associated mHTT aggregation, supporting a functional C9orf72-Stat1-Isg15 axis. Consistent with these cell-based findings, Stat1, phosphorylated Stat1 and Isg15 levels are elevated in the cortex and striatum of R6/2 HD mouse brains. Collectively, our findings identify a novel wt-C9orf72-Stat1-Isg15 axis that promotes proteasomal dysfunction and mHTT aggregation, providing new insights into wt-C9orf72-associated protein homeostasis.

Graphical abstract

Introduction

Huntington’s disease (HD) is an autosomal dominant neurodegenerative disorder caused by CAG repeat expansion in the huntingtin (HTT) gene, resulting in a mutant huntingtin protein (mHTT) with an expanded polyglutamine tract. The mutant protein undergoes aberrant conformational changes that promote misfolding, oligomerization, and intracellular aggregation. Accumulation of aggregation-prone mHTT may disrupt neuronal homeostasis and potentially associate with progressive neurodegeneration in different models [[1], [2], [3], [4], [5]]. A central pathogenic feature of HD is therefore proteostasis failure, leading to the inability of neurons to maintain the balance between production, folding, and clearance of misfolded proteins [6,7].

Efficient clearance of misfolded and short-lived proteins relies predominantly on the ubiquitin-proteasome system (UPS) [8]. In this pathway, substrate proteins are tagged with polyubiquitin chains and subsequently degraded by the 26S proteasome [9]. While mHTT is a known target for ubiquitination and proteasomal turnover, growing evidence indicates that proteasomal capacity is compromised in HD. Impaired UPS activity is thought to exacerbate mHTT aggregation and proteotoxic stress, creating a feed-forward cycle of protein accumulation and neuronal dysfunction [4,8,10]. Despite extensive documentation of proteasome impairment in HD, the upstream molecular mechanisms that initiate or amplify this UPS dysfunction remain incompletely defined.

Chromosome 9 open reading frame 72 (C9orf72) encodes two major protein isoforms generated by alternative splicing. The longer isoform contains a differentially expressed in normal and neoplastic cells (DENN) domain and forms a complex with SMCR8 and WDR41 that regulates small GTPase signaling in a substrate-dependent manner, thereby participating in vesicular trafficking and autophagy [[11], [12], [13]]. Recent genetic and clinical insights suggest that the C9orf72 may intersect with HD pathology. Mutations in C9orf72, specifically hexanucleotide repeat expansions (HREs), are the most common genetic cause of familial amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) [14]. Notably, C9orf72 mutations have also been identified in a subset of patients who exhibit HD-like symptoms but lack HTT repeat expansions, a condition referred to as an HD phenocopy [15,16]. These findings strongly imply that C9orf72 may affect molecular pathways shared across neurodegenerative disorders. However, although the physiological roles of wild-type (wt) C9orf72 in the central nervous system have been increasingly characterized, its potential contribution to mHTT proteostasis in HD remains unknown.

Signal transducer and activator of transcription 1 (Stat1) is a stress-responsive transcription factor that regulates gene expression downstream of diverse signaling inputs. Among its transcriptional targets is interferon-stimulated gene 15 (Isg15), a ubiquitin-like modifier capable of conjugating to substrate proteins through ISGylation [17]. Owing to its structural similarity to ubiquitin, Isg15 has been implicated in the regulation of proteasomal degradation. However, its precise role in the UPS remains context-dependent and incompletely resolved. ISG15 has been reported to interfere with ubiquitination and proteasomal processing under certain conditions, whereas in other contexts it may facilitate protein turnover [17]. Notably, ISG15 expression is markedly elevated in several neurodegenerative diseases, including ataxia-telangiectasia [18], ALS [19], spinocerebellar ataxia type 1 [20], and Alzheimer’s disease [21]. Whether ISG15 is similarly dysregulated in HD and how it affects proteostasis in this context remain to be determined.

Here, we investigated whether wild-type C9orf72 contributes to mHTT aggregation by modulating proteasomal function. Through unbiased proteomic profiling, we identified Stat1 as a major downstream effector of C9orf72 signaling. We demonstrate that C9orf72 overexpression enhances Stat1 activation and nuclear localization, leading to transcriptional upregulation of Isg15. Elevated Isg15 in turn impairs UPS-mediated substrate clearance and promotes accumulation of mHTT aggregates. Furthermore, components of this signaling cascade are upregulated in the R6/2 mouse model of HD. Collectively, our findings identify a C9orf72-Stat1-Isg15 axis that modulates UPS function and mHTT aggregation in HD.

Materials & Methods

Plasmids

The following expression and reporter constructs were used: mHTT, GFP-mHTT, C9orf72-HA, shScramble, shStat1, Isg15, Isg15 promoter-luciferase, β-galactosidase (β-gal), sgVec, sgIsg15, Ubi-HA, and UbG76V-GFP. The mHTT (exon 1 containing 84 CAG repeats) and GFP-mHTT (GFP-84Q) constructs, driven by a human ubiquitin promoter, were previously described [8,10]. Human C9orf72 was C-terminally HA-tagged and subcloned under a CMV promoter. Mouse Isg15 (NM_015783.5) was synthesized (Mission Biotech) and cloned into the FUW lentiviral vector (Addgene #614882). For RNA interference, shStat1 (TRCN0000054923 targets the sequence 5′-GCTCACTCAGAACACTCTGAT-3′ and TRCN0000054927 targets the sequence 5′-GCCGAGAACATACCAGAGAAT-3), shC9orf72 (TRCN0000267375 targets the sequence 5′-AGCACTTATGGACTATCAATT-3′) and a non-targeting shScramble construct under a U6 promoter were obtained from the National RNAi Core Facility (Academia Sinica, Taiwan). The Isg15 promoter–luciferase reporter consists of the mouse Isg15 promoter upstream of firefly luciferase. For CRISPR/Cas9 editing, an sgRNA targeting Isg15 (5′-TGGAGTTAGTCAGCCACACCAGG-3′) was cloned into the PX459 vector (pSpCas9(BB)-2A-Puro; Addgene #62988) under a human ubiquitin promoter (sgIsg15); the empty PX459 vector served as a control (sgVec). The Ubi-HA construct encodes HA-tagged ubiquitin under a CMV promoter. The Ub^G76V-GFP reporter, encoding GFP fused to a mutant ubiquitin (G76V), was expressed under a human ubiquitin promoter.

Cell culture, transfection, and differentiation

Mouse neuroblastoma 2a (N2a) cells (ATCC CCL-131) were maintained in Eagle’s Minimum Essential Medium (MEM) supplemented with 1 mM sodium pyruvate and 10% fetal bovine serum (FBS) at 37 °C in a humidified 5% CO2 atmosphere. Transfections were performed using Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions. For neuronal differentiation, culture medium was replaced 6–8 h post-transfection with MEM containing 2% FBS and 10 μM all-trans retinoic acid, and cells were incubated for an additional 48 h prior to analysis.

R6/2 HD transgenic mice

Transgenic R6/2 mice, expressing exon 1 of the human HTT gene with 120–160 CAG repeats, were used in this study [3,4,22]. Clinical symptoms, including tremors and motor decline, manifest at 5–7 weeks of age, leading to significant deterioration and mortality by 13–16 weeks. Experimental protocols were approved by the Institutional Animal Care and Use Committee at National Cheng Kung University (NCKU), Taiwan. The laboratory animal center at NCKU is fully accredited by Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC International). The genotyping of transgenic mice was conducted via PCR using primers 5′-GGCGACCCTGGAAAAGCTGA-3′ and 5′-TGAGGAAGCTGAGGAGGCGG-3′ to detect a 490 bp amplicon. Animals were maintained at 21 ± 1 °C on a 14:10 h light-dark cycle with food and water ad libitum.

Lentivirus preparation

HEK293FT cells (Invitrogen) were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen) supplemented with 2 mM l-glutamine, 1.5 g/L sodium bicarbonate, and 10% fetal bovine serum (FBS) at 37 °C in a humidified atmosphere containing 5% CO₂. Lentiviral particles were generated by transient calcium phosphate transfection of HEK293FT cells with lentiviral transfer plasmids encoding C9orf72, RFP-Control, or GFP-mHTT, together with the VSV-G envelope plasmid and the pΔ8.9 packaging plasmid. The culture medium was replaced the following day. Forty-eight hours after medium replacement, lentivirus-containing supernatants were collected, filtered through a 0.45 μm membrane filter, and concentrated by ultracentrifugation at 17,500 rpm for 7 h at 4 °C. Viral pellets were resuspended in sterile 1 × PBS to achieve a final titer of approximately 10⁷ transducing units/mL. Viral stocks were aliquoted and stored at −80 °C until use [7,23,24].

Primary cortical neuron culture

Primary cortical neuron culture was performed with approval from the Institutional Animal Care and Use Committee of National Cheng Kung University, Taiwan. Primary cortical neurons were isolated from embryonic day 17.5 (E17.5) FVB wild-type mice. Briefly, cortices were dissected in cold Hank’s Balanced Salt Solution (HBSS; Corning) supplemented with penicillin/streptomycin, and meninges were carefully removed under a dissecting microscope. Cortical tissues were collected in Neurobasal Medium (NBM; Gibco) supplemented with 2 mM l-glutamine and B-27 Supplement (Gibco), and tissues were gently triturated to obtain a single-cell suspension. The cell suspension was filtered through a 40 μm cell strainer and centrifuged at 1000 rpm for 5 min. The resulting cell pellet was resuspended in NBM. Cells were seeded onto 12-well plates pre-coated with 0.0025% poly-d-lysine at a density of 1.5 × 10⁵ cells per well. The day of plating was defined as DIV0, and half of the culture medium was replaced every two days [25]. At DIV3, primary cortical neurons were co-infected with lentiviruses expressing GFP-mHTT together with either C9orf72 or RFP-Control. Neurons were maintained until DIV10 and then harvested for Western blot analysis.

Protein extraction and immunoblotting

Cells or mouse brain tissues were lysed in RIPA buffer (50 mM Tris-HCl, pH 8.0; 150 mM NaCl; 1 mM EDTA; 1 mM EGTA; 0.1% SDS; 0.5% sodium deoxycholate; 1% Triton X-100) supplemented with protease inhibitors (Santa Cruz Biotechnology). Lysates were sonicated, clarified by centrifugation (13,200 rpm, 10 min, 4 °C), and quantified using a Bradford protein assay (Thermo Fisher Scientific). Equal amounts of protein were resolved by SDS-PAGE and transferred to 0.22 μm PVDF membranes (Pall Life Sciences). Membranes were blocked in 5% non-fat milk, incubated overnight at 4 °C with primary antibodies, and subsequently incubated with HRP-conjugated secondary antibodies (1:10,000) for 1 h at room temperature. Primary antibodies included: mHTT (Sigma, MAB5374, 1:1000), C9orf72 (GeneTex, GTX632041, 1:1000), GFP (GeneTex, GTX113617, 1:2000), cleaved caspase-3 (Cell Signaling, 9661, 1:1000), caspase-3 (Cell Signaling, 9662, 1:1000), p-STAT1 (Tyr701) (Affinity Biosciences, AF3300, 1:1000), STAT1 (Affinity, AF6300, 1:1000), Lamin A/C (Cell Signaling, 4777, 1:1000), Isg15 (Santa Cruz, sc-166755, 1:1000), HA (Roche, 11867423001, 1:1000), Ubiquitin (Dako, Z0458, 1:2000), α-tubulin (GeneTex, GTX628802, 1:10,000), and γ-tubulin (Sigma, T6557, 1:10,000). Protein bands were visualized using chemiluminescence and quantified with ImageJ (NIH).

Quantitative real-time PCR (qPCR)

qPCR was performed using the StepOnePlus Real-Time PCR System (Applied Biosystems, USA). Gene expression analysis was conducted for Isg15, Isg20, Ccl5, Irf1, Mx1, and Oas2, with 18S rRNA serving as the endogenous housekeeping control. Reactions were carried out using SYBR Green Master Mix (Thermo Fisher Scientific) in accordance with the manufacturer’s instructions. The thermal cycling conditions were as follows: an initial denaturation step at 95 °C for 10 min, followed by 40 cycles of denaturation at 95 °C for 15 s and annealing/extension at 60 °C for 1 min. A melting curve analysis was subsequently performed with one cycle at 95 °C for 15 s, 60 °C for 1 min, and 95 °C for 15 s to verify amplification specificity. Fluorescence signals were acquired at the annealing/extension step of each cycle. Relative gene expression levels were calculated using the 2^−ΔΔCt method, with normalization to 18S rRNA.

Immunofluorescence

N2a cells grown on 12-mm glass coverslips were fixed with 4% PFA for 10 min, permeabilized, and blocked for 1 h in PBS containing 5% donkey serum, 0.2% Triton X-100, 0.1% saponin, and 2% BSA. Cells were incubated overnight at 4 °C with primary antibodies against HA (1:100), p-Stat1 (1:100), and Isg15 (1:100), followed by 1-h incubation with Alexa Fluor 488- or 594-conjugated secondary antibodies (1:1000; Invitrogen) and Hoechst 33342 (1 μg/mL). Coverslips were mounted using Fluoromount-G. Images were acquired on a DMi8 fluorescence microscope (Leica) and analyzed using ImageJ.

Cell viability assay

N2a cells were incubated with propidium iodide (PI, 1 μg/mL) and Hoechst 33342 (1 μg/mL) for 15 min at 37 °C. Cell survival was calculated as the percentage of PI-negative cells relative to the total number of Hoechst-stained nuclei. Fluorescence images were captured by fluorescence microscopy, and PI-positive (non-viable) and Hoechst-positive (total) nuclei were quantified using ImageJ. Cell survival (%) was calculated as 100 × [1 - (PI-positive cells/total cells)].

Luciferase reporter assay

Cells were co-transfected with Isg15 promoter-luciferase reporter and β-galactosidase control plasmid. Forty-eight hours post-transfection, cells were lysed, and firefly luciferase activity was measured using a luminometer. β-galactosidase activity was measured colorimetrically. Luciferase activity was normalized to β-galactosidase activity.

Ub^G76V-GFP proteasome reporter assay

Proteasome function was monitored using the Ub^G76V-GFP reporter [26]. N2a cells were co-transfected with 84Q and either C9orf72-HA or control plasmid together with Ub^G76V-GFP. After 48 h, nuclei were counterstained with Hoechst 33342 and imaged by fluorescence microscopy. Ub^G76V-GFP expression and accumulation were further evaluated by Western blotting.

Proteomic analysis

For proteomic profiling, N2a cells (8 × 10⁵ per 6-cm dish) were transfected with C9orf72-HA or control plasmid in the presence or absence of GFP-84Q. Forty-eight hours after transfection, cells were lysed in RIPA buffer with protease inhibitors, and proteins were precipitated with cold acetone. Pellets were washed with 20% acetone, air-dried, and resuspended in 6 M urea in 25 mM ammonium bicarbonate. Protein concentrations were determined by Bradford assay, and 50–100 μg protein per sample was submitted to the NTU Consortia of Key Technologies and NTU Instrumentation Center for LC-MS/MS analysis on an Orbitrap Elite mass spectrometer (Thermo Scientific).

Bioinformatic analysis

To compare C9orf72 expression between HD and control samples, public transcriptomic data were retrieved from the Gene Expression Omnibus (GEO) database (GSE1767; Borovecki et al., 2005). Expression values were analyzed using GEO2R and visualized with GraphPad Prism (GraphPad Software).

Statistical analysis

Data are presented as mean ± SD. Statistical analyses were performed using GraphPad Prism. Two-group comparisons were analyzed using unpaired two-tailed Student’s t-test. Multiple-group comparisons were performed using one-way ANOVA followed by Tukey’s post hoc test. p < 0.05 was considered statistically significant. All experiments were independently repeated at least three times (biological replicates).

Results

C9orf72 expression changes are context-dependent in human HD datasets and increased in R6/2 mouse model

Although HREs in C9orf72 are not present in genetically confirmed HD patients, mutations in C9orf72 can produce HD-like phenotypes [15,16,27]. Since no studies have addressed on the role of WT C9orf72 in HD, whether it contributes to HD pathogenesis remains unclear. To address this, we first analyzed the GSE1767 peripheral blood microarray dataset from the Gene Expression Omnibus (GEO). Quantitative analysis revealed a significant elevation of C9ORF72 transcripts in HD patients compared with healthy controls (Fig. 1b). To corroborate this finding within the central nervous system, we examined the R6/2 transgenic mouse model, which expresses exon 1 of the human HTT gene carrying ∼150 CAG repeats and recapitulates key neuropathological features of HD [28]. As expected, abundant mHTT aggregates were detected in R6/2 brains but not in WT littermates (Fig. 1c). Western blot analysis further demonstrated significantly increased C9orf72 protein levels in whole-brain lysates from 3-month-old R6/2 mice relative to controls (Fig. 1c and d). To further evaluate whether C9orf72 expression changes are consistent across additional HD-related datasets, we analyzed publicly available transcriptomic datasets from human post-mortem brain tissues (Supp Figs. 1a and 1b), hESCs (Supp Fig. 1c), hESC-derived neural stem cells (Supp Fig. 1d), and R6/2 mouse brain regions (Supp Fig. 2). In human HD datasets, C9ORF72 expression was variable, with decreased expression in one post-mortem prefrontal cortex dataset and increased expression in hESC- and neural stem cell-based HD models (Supp Fig. 1). In contrast, R6/2 mouse transcriptomic datasets showed increased C9orf72 transcript levels in the cortex and striatum (Supp Fig. 2), consistent with the increased C9orf72 protein levels observed in our R6/2 brain lysates (Fig. 1c and d). Together, these findings indicate that C9ORF72 expression changes are tissue-, dataset-, and model-dependent in human HD-related contexts, while R6/2 mouse datasets show increased C9orf72 expression at both transcript and protein levels.

Fig. 1.

C9orf72 expression is increased in HD human and mouse models. (a, b) Analysis of C9ORF72 transcript levels in human peripheral blood using the GEO dataset GSE1767. (a) Expression values of C9ORF72 in individual samples from healthy controls and HD patients. (b) Quantification of C9ORF72 transcript levels showing significantly increased expression in HD patients compared with healthy controls. (Normal n = 14; HD n = 12). (c, d) Analysis of C9orf72 protein levels in brain lysates from 3-month-old WT and R6/2 mice. (c) Representative immunoblots showing insoluble mHTT, C9orf72 and γ-tubulin. (d) Quantification of C9orf72 protein levels demonstrating increased expression in R6/2 HD mice compared with WT controls. (WT, n = 6; R6/2, n = 7). Data are presented as mean ± SD. ∗∗p < 0.01, ∗∗∗∗p < 0.0001 by unpaired t-test.

Elevated C9orf72 induces mHTT aggregation and cellular toxicity

To determine whether increased C9orf72 expression functionally influences mHTT accumulation, we co-expressed GFP-tagged mHTT with the long isoform of WT C9orf72 in N2a cells (Fig. 2a). C9orf72 overexpression markedly increased GFP-mHTT fluorescence intensity (Fig. 2b) and promoted accumulation of insoluble aggregates, as confirmed by immunoblotting (Fig. 2c and d). Conversely, C9orf72 knockdown reduced insoluble mHTT accumulation in mHTT-expressing N2a cells (Supp Fig. 3a). Given that mHTT accumulation is closely linked to cellular toxicity, we next assessed apoptotic markers and cell viability. C9orf72 overexpression significantly increased the levels of cleaved caspase-3, an apoptotic marker (Fig. 2e and f), and reduced overall cell viability, as indicated by PI staining (Fig. 2g and h). To further examine this effect in a neuronal context, we also co-expressed GFP-tagged mHTT with the C9orf72 in mouse primary cortical neurons. Consistent with the N2a cell results, C9orf72 overexpression significantly increased mHTT aggregates (Fig. 2i and j) and cleaved caspase-3 (Fig. 2i–k) as well. These results indicate that C9orf72 overexpression promotes mHTT aggregation and further reduces cell viability, suggesting that C9orf72-induced mHTT aggregation is associated with cellular toxicity under these experimental conditions.

Fig. 2.

C9orf72 overexpression promotes aggregates and cell death in mHTT-expressing cells. (a–h) N2a cells were co-transfected with GFP-mHTT and either C9orf72 or an empty control vector, and analyzed 48 h post-transfection by fluorescence microscopy and immunoblotting. (a) Schematic diagrams of the plasmids used for transfection, including GFP-mHTT (HD exon 1 containing 84 CAG repeats) and long isoform C9orf72 (C9 plasmid) with a C-terminal HA tag. Co-transfection ratio of GFP-mHTT and C9orf72 plasmid was 1:1. (b) Representative fluorescence images of N2a cells expressing GFP-mHTT with either control vector or C9orf72. Hoechst staining marks nuclei (blue), and GFP fluorescence indicates GFP-mHTT (green). Increased GFP-positive aggregates are observed in cells expressing C9orf72. Scale bar = 200 μm. (c) Representative immunoblots showing insoluble GFP-mHTT (stacking gel), soluble GFP-mHTT, C9orf72 and γ-tubulin. (d) Quantification of insoluble GFP-mHTT levels normalized to γ-tubulin, showing increased aggregation upon C9orf72 overexpression. (n = 5). (e) Representative immunoblots showing C9orf72, caspase-3, cleaved caspase-3, and γ-tubulin. (f) Quantification of cleaved caspase-3 normalized to γ-tubulin, indicating increased apoptotic signaling in cells overexpressing C9orf72. (n = 5). (g) Representative fluorescence images of N2a cells stained with Hoechst (blue) and propidium iodide (PI, red). Increased PI-positive cells are observed in the C9orf72-overexpressing group. Scale bar = 100 μm. (h) Quantification of cell survival rate based on PI staining, showing reduced viability in C9orf72-overexpressing cells compared with control. (n = 6–7). (i–k) Primary neurons were infected with GFP-mHTT together with either control vector or C9orf72 and analyzed by immunoblotting. (i) Representative immunoblots of primary neurons expressing GFP-mHTT with either control vector or C9orf72. (j) Quantification of insoluble GFP-mHTT normalized to γ-tubulin. (k) Quantification of cleaved caspase-3 normalized to γ-tubulin. (n = 6). Data are presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01 by unpaired t-test.

Proteomic profiling identifies Stat1 as a critical mediator of C9orf72-induced mHTT accumulation

To delineate the molecular mechanism driving C9orf72-mediated mHTT aggregation, we performed label-free quantitative mass spectrometry on N2a cells to identify global proteomic shifts induced by C9orf72. We compared two conditions: C9orf72 versus vector control (basal state), and GFP-mHTT + C9orf72 versus GFP-mHTT + vector (disease state). Among 1921 high-confidence proteins identified (Fig. 3a), differential expression analysis (fold change ≥1.25, p < 0.05) revealed 62 upregulated and 26 downregulated proteins in the Ctrl comparison, and 25 upregulated and 27 downregulated proteins in the mHTT background (Fig. 3b). To isolate primary downstream effectors, we focused on proteins consistently altered across both datasets. This intersection yielded four overlapping targets, with Stat1 exhibiting the most consistent and significant upregulation (Fig. 3c).

Fig. 3.

Proteomic analysis identifies Stat1 as a candidate downstream target of C9orf72. (a) Experimental workflow for label-free quantitative proteomic analysis. N2a cells were transfected with either control vector or C9orf72, in the presence or absence of GFP-mHTT expression. Cell lysates were subjected to LC-MS/MS analysis, resulting in the identification of approximately 1921 high-confidence proteins. (b) Volcano plots showing differentially expressed proteins (DEPs) identified in two comparisons: Ctrl vs. C9orf72 (left) and GFP–mHTT + Ctrl vs. GFP–mHTT + C9orf72 (right). Each dot represents a quantified protein. Upregulated proteins are indicated in red and downregulated proteins in blue, based on fold-change and statistical significance thresholds. (c) Venn diagram showing the overlap of DEPs between the two comparisons. A total of four proteins were commonly altered in both datasets. The table lists the overlapping proteins along with their fold changes (mHTT + C9orf72/mHTT + Ctrl) and P values.

We validated this proteomic hit via Western blotting, confirming marked increases in both total and phosphorylated Stat1 (p-Stat1 Tyr701) upon C9orf72 overexpression under basal and mHTT-expressing conditions (Fig. 4a–e). Immunofluorescence analysis further supported elevated Stat1 levels in C9orf72-positive cells (Supp Fig. 4). Conversely, C9orf72 knockdown reduced both p-Stat1 and total Stat1 protein levels in mHTT-expressing N2a cells (Supp Fig. 3b). To determine whether Stat1 mediates C9orf72-driven aggregation, we silenced Stat1 in cells co-expressing mHTT and C9orf72. Stat1 knockdown significantly reduced GFP-mHTT fluorescence intensity and insoluble aggregate levels compared with control shRNA-treated cells (Fig. 4f–h). This effect was further confirmed using an independent shStat1 construct targeting a distinct Stat1 sequence (Supp Fig. 5). These data demonstrate that Stat1 contributes to C9orf72-driven mHTT aggregation.

Fig. 4.

C9orf72 promotes mHTT aggregation through activation of Stat1 signaling. (a) Representative immunoblots of N2a cells transfected with control vector or C9orf72, in the presence or absence of GFP–mHTT. Immunoblotting shows phosphorylated Stat1 (p-Stat1), total Stat1, C9orf72, and γ-tubulin. (b, c) Quantification of p-Stat1 (b) and total Stat1 (c) levels normalized to γ-tubulin in cells transfected with control or C9orf72. Both p-Stat1 and total Stat1 levels were significantly increased upon C9orf72 overexpression. (d, e) Quantification of p-Stat1 (d) and total Stat1 (e) levels normalized to γ-tubulin in cells co-expressing GFP-mHTT with either control vector or C9orf72. (n = 3). ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001 by unpaired t-test. (f–h) Stat1 knockdown attenuates C9orf72-induced mHTT aggregation. N2a cells were co-transfected with GFP-mHTT, either C9orf72 or control vector, together with shScramble or shStat1 targeting on GCTCACTCAGAACACTCTGAT, and analyzed 48 h post-transfection. (f) Representative fluorescence images showing Hoechst-stained nuclei (blue) and GFP-mHTT (green). Stat1 knockdown reduced GFP–mHTT aggregation under C9orf72 overexpression conditions. Scale bar = 200 μm. (g) Representative immunoblots showing C9orf72, p-Stat1, Stat1, insoluble GFP-mHTT (stacking gel), soluble GFP-mHTT and γ-tubulin. (h) Quantification of insoluble GFP-mHTT normalized to γ-tubulin. Stat1 knockdown significantly reduced C9orf72-induced mHTT aggregation. (n = 5). ∗p < 0.05, ∗∗p < 0.01 by one-way ANOVA with Tukey’s post hoc test. Data are presented as mean ± SD.

C9orf72 drives Stat1 nuclear translocation and subsequent Isg15 transcription

Because Stat1 activation is commonly associated with inflammatory and interferon-related signaling, we next examined whether C9orf72 overexpression was accompanied by broad induction of inflammatory mediators. Using a Mouse Inflammation Antibody Array, we did not detect substantial changes in the inflammatory mediators included in the array following C9orf72 overexpression in mHTT-expressing cells (Supp Fig. 6), suggesting that broad inflammatory mediator induction was not evident under our experimental conditions although this assay does not exclude changes in specific interferon pathways or other cell-autonomous upstream mechanisms. Because Stat1 functions as a transcription factor upon phosphorylation, we next examined whether C9orf72 affects its subcellular localization. Fractionation analyses revealed increased nuclear enrichment of phosphorylated Stat1 in C9orf72-overexpressing cells (Fig. 5a–c). To identify downstream transcriptional targets, we assessed expression of Stat1-responsive genes by quantitative PCR (qPCR). Among the genes examined, Isg15 showed the strong induction following C9orf72 overexpression (Fig. 5d). Although several Stat1-responsive genes, including Oas2, were upregulated following C9orf72 overexpression, Isg15 exhibited the robust induction. Given the established role of Isg15 in regulating protein stability, ubiquitin-dependent signaling, and protein aggregation, we selected Isg15 for subsequent mechanistic analyses.

Fig. 5.

C9orf72 enhances Stat1-dependent Isg15 transcription and promotes mHTT aggregation. (a–c) C9orf72 promotes Stat1 activation and nuclear accumulation. N2a cells were co-transfected with GFP-mHTT and either control vector or C9orf72, followed by cytoplasmic and nuclear fractionation 48 h post-transfection. (a) Representative immunoblots showing p-Stat1, Stat1, and C9orf72 in cytoplasmic and nuclear fractions. α-tubulin and Lamin A/C were used as cytoplasmic and nuclear markers, respectively. (b, c) Quantification of p-Stat1 levels in cytoplasmic (b) and nuclear (c) fractions normalized to their respective loading controls. (n = 3). ∗p < 0.05, ∗∗p < 0.01 by unpaired t-test. (d) Quantitative PCR analysis of Stat1 downstream genes. N2a cells were co-transfected with GFP-mHTT and either control vector or C9orf72. Relative mRNA levels of Isg15, Isg20, Ccl5, Irf1, Mx1, and Oas2 were determined by qPCR, showing increased expression upon C9orf72 overexpression. (n = 6). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 by unpaired t-test. (e–g) C9orf72 increases Isg15 protein expression. (e) Representative immunoblots showing Isg15, C9orf72 and γ-tubulin. (f, g) Quantification of Isg15 protein levels normalized to γ-tubulin in cells transfected with control or C9orf72, in the absence (f) or presence (g) of GFP-mHTT. (n = 4). ∗p < 0.05 by unpaired t-test. (h) Schematic illustration of the Isg15 promoter-luciferase reporter construct. (i) Luciferase reporter assay showing increased Isg15 promoter activity upon C9orf72 overexpression. (j) Stat1 knockdown abolishes C9orf72-induced Isg15 promoter activation. Luciferase activity was normalized to β-galactosidase activity. (n = 9). ∗∗∗∗p < 0.0001 by unpaired t-test. (k–m) Isg15 depletion attenuates C9orf72-induced mHTT aggregation. N2a cells were co-transfected with GFP-mHTT, either C9orf72 or control vector, together with sgIsg15 or empty vector, and analyzed 48 h post-transfection. (k) Representative fluorescence images showing Hoechst-stained nuclei (blue) and GFP-mHTT (green). Isg15 depletion reduced GFP-mHTT aggregation under C9orf72 overexpression conditions. Scale bar = 200 μm. (l) Representative immunoblots showing insoluble GFP-mHTT (stacking gel), soluble GFP-mHTT, C9orf72, Isg15 and γ-tubulin. (m) Quantification of insoluble GFP-mHTT normalized to γ-tubulin. Isg15 depletion significantly reduced C9orf72-induced mHTT aggregation. (n = 3). ∗p < 0.05, ∗∗p < 0.01 by one-way ANOVA with Tukey’s post hoc test. Data are presented as mean ± SD.

We next examined whether the increase in Isg15 transcript levels was accompanied by elevated protein expression under C9orf72 overexpression using Western blot analysis. We found that Isg15 protein levels were significantly increased in C9orf72-expressing cells compared with controls (Fig. 5e–g), consistent with the transcriptional upregulation observed in the qPCR results. Immunofluorescence analysis supported increased Isg15 signal intensity in C9orf72-positive cells (Supp Fig. 7). Conversely, C9orf72 knockdown reduced Isg15 protein levels in mHTT-expressing N2a cells (Supp Fig. 3c). To determine whether Isg15 transcription is directly regulated through Stat1, we performed a luciferase reporter assay driven by the Isg15 promoter. C9orf72 significantly increased promoter activity, whereas Stat1 knockdown markedly attenuated this effect (Fig. 5h–j). Consistent with previous studies demonstrating direct binding of Stat1 to the Isg15 promoter by ChIP-PCR analyses [29,30], Stat1 regulates Isg15 promoter activity and expression levels. To assess functional relevance, we silenced Isg15 in cells co-expressing C9orf72 and mHTT. Isg15 depletion significantly reduced GFP-mHTT fluorescence intensity (Fig. 5k) and insoluble aggregate levels (Fig. 5l and m) compared with control sgRNA. Together, these findings support a model in which C9orf72 promotes mHTT aggregation through Stat1-dependent upregulation of Isg15.

Isg15 accumulation is associated with impaired UPS-mediated substrate clearance and promotes mHTT aggregation

We next investigated whether direct Isg15 elevation is sufficient to phenocopy C9orf72-mediated toxicity. N2a cells were co-expressed with GFP-mHTT and Isg15, and fluorescence imaging showed a marked increase in GFP-mHTT signal intensity in Isg15-overexpressing cells compared with controls (Fig. 6a). Consistently, Western blot analysis confirmed higher levels of insoluble mHTT aggregates (Fig. 6b and c). This accumulation was accompanied by elevated cleaved caspase-3 and decreased cell viability (Fig. 6d–g). These data indicate that increased Isg15 expression is sufficient to promote mHTT aggregation and further reduces cell viability.

Fig. 6.

Isg15 overexpression promotes aggregation and cell death in mHTT-expressing cells. N2a cells were co-transfected with GFP-mHTT and either Isg15 or control vector, and analyzed 48 h post-transfection by fluorescence microscopy and immunoblotting. (a) Representative fluorescence images showing Hoechst-stained nuclei (blue) and GFP-mHTT (green). Increased GFP-mHTT aggregates were observed in cells overexpressing Isg15 compared with control. Scale bar = 200 μm. (b) Representative immunoblots showing insoluble GFP-mHTT (stacking gel), soluble GFP–mHTT, Isg15 and γ-tubulin. (c) Quantification of insoluble GFP-mHTT levels normalized to γ-tubulin, showing increased aggregation upon Isg15 overexpression by unpaired t-test. (n = 4). (d) Representative immunoblots showing cleaved caspase-3, Isg15 and γ-tubulin. (e) Quantification of cleaved caspase-3 normalized to γ-tubulin, indicating increased apoptotic signaling in cells overexpressing Isg15. (n = 6). (f) Representative fluorescence images of N2a cells stained with Hoechst (blue) and propidium iodide (PI, red). Increased PI-positive cells were observed in Isg15-overexpressing cells compared with control. Scale bar = 100 μm. (g) Quantification of cell survival rate based on PI staining, showing reduced viability in Isg15-overexpressing cells. (n = 5). Data are presented as mean ± SD ∗∗p < 0.01, ∗∗∗p < 0.001 by unpaired t-test.

Given the ubiquitin-like nature of Isg15, we next examined whether Isg15 elevation alters ubiquitinated protein accumulation and UPS-mediated substrate degradation, further leading to accumulation of mHTT. Co-expression of HA-tagged ubiquitin with Isg15 and GFP-mHTT resulted in a marked accumulation of high-molecular-weight polyubiquitinated proteins (Fig. 7a and b), indicative of stalled proteasomal degradation. We functionally validated this using the Ub^G76V-GFP reporter, a direct substrate of the UPS. Isg15 overexpression significantly increased the number of GFP-positive cells and total reporter protein levels (Fig. 7c–e), indicating reduced proteasomal degradation capacity. Importantly, C9orf72 overexpression similarly increased polyubiquitinated proteins and Ub^G76V-GFP reporter levels (Fig. 7f–j), and knockdown of C9orf72 oppositely decreased the expression levels Ub^G76V-GFP levels upon Western blotting results (Supp Fig. 3d). In addition to UPS activity, we also examined whether C9orf72 overexpression affects autophagy, another major intracellular protein degradation pathway [[31], [32], [33]]. Analysis of autophagy-related markers revealed that C9orf72 overexpression did not significantly alter the levels of Atg5-Atg12, LC3 or p62 compared with control cells (Supp Fig. 8). These findings suggest that C9orf72 overexpression does not produce obvious changes in basal autophagy-related markers under our experimental conditions. These findings support that UPS dysfunction occurs in the downstream of C9orf72 signaling, indicating that proteasomal degradation was compromised under these conditions.

Fig. 7.

Isg15 and C9orf72 impair proteasomal degradation and promote accumulation of ubiquitinated proteins. (a, b) N2a cells were co-transfected with GFP-mHTT and Ubi-HA plasmids, with or without Isg15 overexpression, and analyzed 48 h post-transfection by immunoblotting. (a) Representative immunoblots showing HA-tagged ubiquitinated proteins, polyubiquitinated proteins, Isg15, and γ-tubulin. (b) Quantification of HA-tagged ubiquitinated proteins normalized to γ-tubulin, showing increased ubiquitinated protein accumulation upon Isg15 overexpression. (n = 4). (c–e) Analysis of proteasomal degradation using the Ubi^G76V-GFP reporter. N2a cells were co-transfected with GFP-mHTT and Ubi^G76V-GFP, with or without Isg15 overexpression. (c) Representative fluorescence images showing Hoechst-stained nuclei (blue) and Ubi^G76V-GFP (green). Increased reporter fluorescence indicates reduced proteasomal degradation in Isg15-overexpressing cells. Scale bar = 200 μm. (d) Representative immunoblots showing Ubi^G76V-GFP, ubiquitin, Isg15, and γ-tubulin. (e) Quantification of Ubi^G76V-GFP levels normalized to γ-tubulin, showing increased reporter accumulation upon Isg15 overexpression. (n = 4). (f, g) N2a cells were co-transfected with GFP-mHTT and Ubi-HA, with or without C9orf72 overexpression, and analyzed 48 h post-transfection. (f) Representative immunoblots showing HA-tagged ubiquitinated proteins, polyubiquitinated proteins, C9orf72, and γ-tubulin. (g) Quantification of HA-tagged ubiquitinated proteins normalized to γ-tubulin, showing increased ubiquitinated protein accumulation upon C9orf72 overexpression. (n = 5). (h–j) Analysis of proteasomal degradation using the Ubi^G76V-GFP reporter under C9orf72 overexpression conditions. (h) Representative fluorescence images showing Hoechst-stained nuclei (blue) and Ubi^G76V-GFP (green). Increased reporter fluorescence indicates reduced proteasomal degradation in C9orf72-overexpressing cells. Scale bar = 200 μm. (i) Representative immunoblots showing Ubi^G76V-GFP, ubiquitin, C9orf72 and γ-tubulin. (j) Quantification of Ubi^G76V-GFP levels normalized to γ-tubulin, showing increased reporter accumulation upon C9orf72 overexpression. (n = 5). Data are presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01 by unpaired t-test.

Stat1, phosphorylated Stat1, and Isg15 are elevated in R6/2 mouse brains

To determine whether Stat1-Isg15 signaling components are altered in vivo, we analyzed the expression profiling of these target proteins in cortical (Fig. 8a) and striatal (Fig. 8b) lysates from 2.5-month-old R6/2 and WT mice. Western blot analysis revealed significant upregulation of total Stat1, functionally active p-Stat1 (Fig. 8c, d, f, g), and downstream Isg15(Fig. 8e–h) in R6/2 brains compared to WT controls (Fig. 8). These in vivo findings show that Stat1-Isg15 signaling components are elevated in HD mouse brains, supporting the relevance of the C9orf72-associated pathway identified in our cellular models. To further explore the feasibility of assessing C9orf72 effects in vivo, we attempted to generate C9orf72-overexpressing transgenic mice (Supp Fig. 9a). Genotyping PCR confirmed genomic integration of the transgene (Supp Fig. 9b), and RT-PCR detected C9orf72 mRNA expression in brain tissue (Supp Fig. 9c). However, robust C9orf72 protein expression was not detected (Supp Fig. 9d), and stable transmission of the transgene to subsequent generations was not established despite breeding attempts (Supp Fig. 9e). Therefore, this model was not suitable for evaluating the in vivo consequences of C9orf72 overexpression on HD-related phenotypes.

Fig. 8.

Stat1-Isg15 signaling is activated in the brains of R6/2 mice. (a, b) Immunoblot analysis of cortical (a) and striatal (b) tissues from 2.5-month-old WT and R6/2 mice. Representative immunoblots show insoluble mHTT (stacking gel), phosphorylated Stat1 (p-Stat1), total Stat1, Isg15, and γ-tubulin as a loading control. Increased p-Stat1, Stat1, and Isg15 levels were observed in R6/2 mice compared with WT controls. (c–e) Quantification of p-Stat1 (c), total Stat1 (d), and Isg15 (e) protein levels in the cortex, normalized to γ-tubulin. (f–h) Quantification of p-Stat1 (f), total Stat1 (g), and Isg15 (h) protein levels in the striatum, normalized to γ-tubulin. Data are presented as mean ± SD (WT n = 3, R6/2 n = 5 mice). ∗p < 0.05, ∗∗p < 0.01 by unpaired t-test.

Discussion

mHTT accumulation is a key proteostasis-related feature of HD pathogenesis, disrupting multiple cellular processes that lead to neuronal dysfunction and death [[34], [35], [36]]. In this study, we investigated whether elevated wt-C9orf72 modulates mHTT accumulation and proteostasis-related signaling in experimental HD models. Our cell-based data show that C9orf72 overexpression increases mHTT aggregation and is accompanied by increased cell death marker. Specifically, C9orf72 overexpression increases Stat1 activation, increases Isg15 expression, and disrupts UPS function, culminating in proteotoxic stress. Loss-of-function experiments demonstrate that Stat1 and Isg15 are required for C9orf72-induced aggregate accumulation, establishing this pathway as a causal mediator rather than a secondary correlate of aggregation. These findings position wt-C9orf72 as an active modulator of proteostasis in the HD context.

While C9orf72 is primarily studied in the context of HREs in ALS/FTD, our findings indicate that the wt-C9orf72 itself can modulate mHTT accumulation and proteostasis-related stress markers when overexpressed. In the present study, we observed increased C9ORF72 transcript levels in one bulk peripheral blood dataset from HD patients (Fig. 1a and b). Nevertheless, this observation should be interpreted cautiously. Whole-blood transcriptomic profiles are influenced not only by gene expression changes within individual cells but also by variations in leukocyte composition [37]. Because C9orf72 is expressed and functionally important in myeloid lineage cells [38], altered proportions of circulating myeloid cells could contribute to apparent differences in bulk C9ORF72 transcript levels. Consistent with these considerations, our analysis of additional HD patient datasets revealed heterogeneous C9ORF72 expression patterns (Supp Fig. 1), potentially reflecting differences in cohort characteristics, disease stage, and sample composition. Therefore, a definitive conclusion regarding the direction and magnitude of C9orf72 dysregulation in HD cannot yet be established. These heterogeneous findings observed across peripheral blood, postmortem cortical tissues, and caudate samples further suggest that C9orf72 dysregulation in HD may be highly dependent on tissue context and disease progression, emphasizing the need for cell type-specific analyses in future studies.

Although C9orf72 has been proposed to support neuronal homeostasis through its regulation of vesicle trafficking, autophagy-lysosome function, and immune signaling [[11], [12], [13]], the observation that its overexpression significantly increases mHTT aggregate load argues against a purely compensatory or protective role (Fig. 2). Instead, these data raise the possibility that C9orf72 dosage may influence proteostasis-related phenotypes under experimental conditions; its upregulation appears to potentially be a detrimental feature of the HD “proteome stress” environment, similar to observations in Alzheimer’s disease [39], Parkinson’s disease (PD) [40], and corticobasal degeneration (CBD) [41]. However, our data do not establish whether endogenous C9orf72 is elevated in human HD cells or whether C9orf72 changes represent a driver, consequence, or compensatory response to proteostasis stress. Given the variability observed among independent transcriptomic datasets, future studies employing larger patient cohorts, standardized experimental procedures, and cell type-resolved transcriptomic analyses will be required to clarify the precise regulation and pathological significance of C9orf72 expression in HD.

Moreover, we consistently observed elevated C9orf72 protein levels in the brains of R6/2 mice using Western blot analysis (Fig. 1c), and found C9orf72 transcript levels are generally increased in R6/2 mice compared with controls (Supp Fig. 2), supporting the possibility that transcriptional upregulation contributes, at least in part, to the elevated protein expression observed in our study. However, our current data do not allow us to distinguish whether the increased C9orf72 protein levels result from enhanced transcription, increased translational efficiency, reduced protein turnover or a combination of these mechanisms. These unresolved possibilities represent an important limitation of the current study and warrant further mechanistic investigation in future work.

An important consideration is that C9orf72 functions as part of a protein complex with SMCR8 and WDR41 that regulates small GTPase signaling in a substrate-dependent manner, and has been implicated in the regulation of autophagy, lysosomal homeostasis, cellular metabolism and innate immune signaling [42,43]. These studies have suggested that disruption of the stoichiometric balance among these binding partners may influence energy metabolism and activate immune-related pathways, including STING signaling. Therefore, it is conceivable that excessive C9orf72 expression could generate supraphysiological levels of unbound C9orf72 protein, potentially leading to cellular dysfunction through mechanisms distinct from those examined in the present study. However, we did not investigate whether C9orf72 overexpression alters the expression levels of SMCR8 or WDR41, nor did we evaluate the involvement of metabolic dysregulation or STING activation. Further studies will be required to determine whether these pathways contribute to the pathogenic effects of elevated C9orf72 in HD.

In the present study, we identified activation of the Stat1-Isg15 pathway following C9orf72 overexpression (Fig. 5), but no substantial changes were detected in the inflammatory mediators included in the Mouse Inflammation Antibody Array (Supp Fig. 6), suggesting that broad cytokine induction is unlikely to be the primary driver of Stat1 signaling under our experimental conditions. Nevertheless, we cannot exclude the possibility that specific interferon-related pathways, subtle cytokine alterations or cell-autonomous signaling mechanisms contribute to Stat1 activation. Additional studies focusing on interferon signaling and innate immune responses will be necessary to clarify the upstream mechanisms linking C9orf72 overexpression to Stat1 activation.

In our data, C9orf72 overexpression increased both total Stat1 and phosphorylated Stat1 in cellular HD model (Fig. 4), indicating the activation of Stat1 signaling is one of C9orf72 working mechanisms. The elevation of total Stat1 may reflect auto-regulatory feedback, as activated Stat1 can enhance its own transcription [44]. In addition, previous work has also suggested that Stat1 ISGylation can antagonize K48-linked ubiquitination and protect phosphorylated STAT1 from proteasomal degradation, thereby preserving Stat1 activation [45]. Given that C9orf72 lacks intrinsic kinase or transcriptional activity, Stat1 activation is likely indirect. Stat1 is canonically activated by JAK-dependent phosphorylation at Tyr701, with full transcriptional activity requiring additional Ser727 phosphorylation via kinases, such as p38 MAPK or PKCδ. [46]. Although C9orf72 deficiency has been reported to enhance JAK-Stat1 signaling through dysregulated endolysosomal trafficking and cGAS/STING activation [47], our findings demonstrate that C9orf72 overexpression similarly promotes Stat1 activation. This suggests a context-dependent regulatory role, potentially reflecting differential effects of C9orf72 dosage on vesicular or signaling homeostasis. While it remains unclear whether Stat1 activation in this context occurs through canonical JAK activity or alternative kinase pathways, another key question for future studies is how C9orf72 influences upstream signaling events or cellular processes that lead to Stat1 activation.

A potential limitation of the present study is that the mechanistic involvement of Stat1 was primarily examined using gene knockdown approaches (Fig. 4 and Supp Fig. 5). However, rescue experiments involving re-expression of shRNA-resistant Stat1, as well as complementary gain- and loss-of-function studies of Isg15, would further strengthen the causal relationship between the C9orf72-Stat1-Isg15 pathway and mHTT accumulation. Future studies incorporating these approaches, together with in vivo HD models, will be valuable for establishing the mechanistic significance of this pathway in HD progression.

Isg15 emerged as a transcriptional target of C9orf72-Stat1 signaling and functions as a key effector of proteostasis disruption. We show that Isg15 overexpression increases mHTT aggregation, elevates polyubiquitinated protein levels, and stabilizes the Ub^G76V-GFP proteasome reporter, collectively indicating impaired proteasomal degradation (Fig. 6, Fig. 7). Isg15 has been reported to interfere with ubiquitin-dependent degradation through multiple mechanisms, including competition with ubiquitin for lysine residues [48], formation of mixed ubiquitin-Isg15 chains [49], and modulation of E2/E3 ligase activities [[50], [51], [52]]. These mechanisms provide plausible explanations for the accumulation of ubiquitinated substrates observed here. However, Ub^G76V-GFP accumulation and increased polyubiquitinated proteins do not distinguish whether Isg15 affects substrate ubiquitination, proteasomal recognition, substrate delivery to the proteasome, deubiquitination, or catalytic degradation. Future studies examining ubiquitin chain architecture, proteasome catalytic activities, and substrate-proteasome engagement will be needed to define the precise biochemical step affected by Isg15.

Although our data focus on the UPS, both C9orf72 and Isg15 have been implicated in autophagy regulation. Isg15 has been reported to either inhibit autophagy initiation via conjugation of Beclin 1 [53] or facilitate selective aggrephagy under specific contexts [54]. Similarly, C9orf72 participates in autophagy-lysosome trafficking [31,32]. However, our results revealed that C9orf72 overexpression did not significantly alter the levels of Atg5-Atg12, LC3 or p62 compared with control cells (Supp Fig. 8). These findings suggest that the enhanced mHTT accumulation observed in our study is unlikely to be primarily mediated through global alterations in basal autophagy-related markers. Nevertheless, the apparent discrepancy between our findings and previous reports highlights the complex and context-dependent roles of C9orf72 and Isg15 in autophagy regulation. It remains possible that C9orf72-induced Isg15 upregulation influences specific forms of selective autophagy, such as aggrephagy, that are not adequately reflected by conventional autophagy markers. Future studies employing autophagic flux assays, lysosomal functional analyses, and selective autophagy reporters will be necessary to determine whether the C9orf72-Stat1-Isg15 axis contributes to mHTT accumulation through more specialized autophagic mechanisms in HD.

Although the present study demonstrates activation of the C9orf72-Stat1-Isg15 pathway in HD cellular models and reveals increased Stat1, phosphorylated Stat1 and Isg15 expression in the cortex and striatum of R6/2 mice (Fig. 8), a direct causal role for this pathway in vivo remains to be established. Genetic manipulation of C9orf72, Stat1, or Isg15 in HD mouse models would provide important mechanistic insights into their contributions to mHTT aggregation, proteostasis impairment, neurodegeneration, and behavioral dysfunction. Although we have previously attempted to generate C9orf72-overexpressing transgenic mice, these animals were not suitable for evaluating the in vivo effects of C9orf72 on HD-related pathology due to failure of C9orf72 protein expression and germline transmission (Supp Fig. 9). Due to the importance of in vivo validation, future studies employing genetic manipulation or stereotaxic viral delivery of C9orf72, Stat1, or Isg15 into the brains of R6/2 HD mice will provide valuable opportunities to determine the contribution of this pathway to HD-associated molecular, pathological, and behavioral phenotypes.

In conclusion, our findings suggest that elevated wt-C9orf72 modulates Stat1-Isg15 signaling, proteasomal reporter accumulation, and mHTT accumulation in experimental HD models. In R6/2 mouse brains, Stat1, phosphorylated Stat1, and Isg15 were also elevated, supporting the presence of this signaling signature in vivo. However, whether this pathway causally contributes to HD-related neuronal dysfunction, behavioral abnormalities, or disease progression under physiological conditions remains to be determined. Our findings uncover a potential role of wt-C9orf72 in triggering a Stat1-Isg15 pathway, leading to proteasomal dysfunction and increasing mHTT aggregation, which provides a potential insight of wt-C9orf72-regulatory mechanisms for neurodegenerative diseases.

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors have read the manuscript, and agree with the consent for publication.

Data availability statement

All data supporting the findings of this study are available within the paper and its Supplementary Information.

Author contributions

SCC, CWT, CYC, CCS, YCC, PMW, CYT, SFC, HYK and PHC handled the experiments and analyses; SCC, CMC and SHY designed the experiments, oversaw the progress of the study and drafted the paper. All authors read and approved the final manuscript.

Funding

This work was supported by the National Science and Technology Council (114-2320-B-006 -022 and 115-2320-B-006 -035 -MY3).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: