Accurate Quantification of Mutant and Wild-Type polyQ Proteins Using Simple Western Capillary Immunoassays

Bas Röttgering; Janwillem Testerink; Rudie Weij; Chantal Beekman; Nicole Datson

doi:10.1007/s12035-025-05089-9

. 2025 May 31;62(10):12768–12790. doi: 10.1007/s12035-025-05089-9

Accurate Quantification of Mutant and Wild-Type polyQ Proteins Using Simple Western Capillary Immunoassays

Bas Röttgering ^1,², Janwillem Testerink ¹, Rudie Weij ¹, Chantal Beekman ¹, Nicole Datson ^1,^✉

PMCID: PMC12433360 PMID: 40450087

Abstract

Polyglutamine (polyQ) diseases are monogenic fatal neurodegenerative disorders caused by a CAG repeat expansion that is translated into a toxic polyQ tract. There are nine polyQ diseases: Huntington’s disease (HD), spinocerebellar ataxias 1, 2, 3, 6, 7 and 17 (SCA1, 2, 3, 6, 7, 17), dentatorubral-pallidoluysian atrophy (DRPLA) and spinal and bulbar muscular atrophy (SBMA). Although no disease-modifying therapies are available, lowering levels of the causative mutant polyQ protein is a promising potential treatment. Preclinically, the efficacy of polyQ protein-lowering compounds is often assessed using time-consuming Western blots (WB), which can produce variable results. Therefore, to improve throughput and accuracy of polyQ protein level quantification, Simple Western (SW) capillary immunoassays were developed. A panel of antibodies was screened for reactivity to the polyQ proteins on SW. The most promising antibodies were selected for further assay development. This resulted in optimised SW immunoassays for huntingtin (HTT), ataxin 1, 2 and 3 (ATXN1, 2, 3), atrophin 1 (ATN1) and androgen receptor (AR). Additionally, size-separation of the wild-type and polyQ-expanded mutant protein isoforms on SW was shown for ATXN1, ATXN3 and ATN1, allowing for their separate quantification. To facilitate size-separation of the larger HTT protein (≥ 348 kDa), a novel caspase 3-based assay was developed to generate N-terminal wild-type and mutant HTT fragments that could be separately quantified on SW in contrast to full-length HTT. In conclusion, SW capillary immunoassays were developed for polyQ proteins to improve preclinical research and aid the development of polyQ-lowering therapies for polyQ diseases.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12035-025-05089-9.

Keywords: Protein quantification, Polyglutamine diseases, Capillary immunoassay, Assay development, Huntington’s disease, Spinocerebellar ataxias

Introduction

Polyglutamine (polyQ) diseases are a class of rare monogenic neurodegenerative disorders caused by a pathological expansion of a CAG trinucleotide repeat in the coding region of a gene that is translated into a toxic expanded polyQ stretch in the corresponding mutant protein. There are nine well-established polyQ diseases: Huntington’s disease (HD), spinocerebellar ataxia types 1, 2, 3, 6, 7 and 17 (SCA1, 2, 3, 6, 7, 17), dentatorubral-pallidoluysian atrophy (DRPLA) and spinal and bulbar muscular atrophy (SBMA), each caused by a CAG repeat expansion in a different gene (Table 1). The polyQ diseases are autosomal dominantly inherited, except for SBMA that follows an X-linked recessive inheritance pattern [1, 2]. PolyQ disease patients develop movement disorders and cognitive impairments caused by progressive neurodegeneration, with longer polyQ expansions correlating with earlier age of onset and faster disease progression [3–11]. However, there are distinct characteristics in the clinical presentation of the nine polyQ diseases due to differences in the most affected regions of the central nervous system (CNS) [1, 12]. HD is the most common polyQ disease, with a pooled prevalence of 4.88 out of 100.000 (95% CI, 3.38–7.02) [13], followed by SCA3, with an estimated prevalence of 1–2 per 100.000 according to the polyQ database [14]. However, the prevalence of polyQ diseases varies significantly between geographical populations, making accurate estimation of global prevalence difficult [13–15].

Table 1.

Overview of repeat lengths in CAG repeat expansion diseases (adapted from [12])

Disease	Disease (full)	Gene	Gene (full)	CAG length
				Healthy	Patient
HD	Huntington’s disease	HTT	Huntingtin	6–35	> 36
SCA1	Spinocerebellar ataxia 1	ATXN1	Ataxin 1	6–34	> 39
SCA2	Spinocerebellar ataxia 2	ATXN2	Ataxin 2	14–31	> 32
SCA3	Spinocerebellar ataxia 3	ATXN3	Ataxin 3	12–40	> 55
SCA6	Spinocerebellar ataxia 6	CACNA1A	Calcium voltage-gated channel subunit alpha1 A	4–18	> 21
SCA7	Spinocerebellar ataxia 7	ATXN7	Ataxin 7	7–18	> 38
SCA17	Spinocerebellar ataxia 17	TBP	TATA-box binding protein	25–43	> 45
DRPLA	Dentatorubral-pallidoluysian atrophy	ATN1	Atrophin 1	3–38	> 49
SBMA	Spinal and bulbar muscular atrophy	AR	Androgen receptor	6–36	> 38

Open in a new tab

How the CAG repeat expansion and the resulting expanded polyQ tract in the mutant proteins lead to neurodegeneration is not fully understood. Nevertheless, the mutant polyQ proteins play a central role in polyQ disease pathogenesis. The expanded polyQ tract confers a toxic gain-of-function to the protein making it prone to aggregation [16–19]. Expression and aggregation of the mutant polyQ proteins lead to disturbance of various cellular processes like transcription, axonal transport, autophagy and mitochondrial metabolism, although the exact mechanism of toxicity differs between the polyQ proteins [1]. Several pathologic mechanisms not directly related to the mutant polyQ proteins may also contribute to disease progression, including RNA toxicity [20–23], repeat-associated non-AUG (RAN) translation [24, 25] and somatic expansion of the CAG repeat [26, 27]. Additionally, aberrant splicing or protein processing of mutant polyQ transcripts or proteins can result in short toxic polyQ protein fragments, as observed with toxic N-terminal huntingtin (HTT) exon 1 in HD [28, 29]. Besides the toxic gain-of-function, there is evidence that loss of wild-type polyQ protein function caused by the polyQ expansion may contribute to disease progression as well [30–34].

At the time of writing of this manuscript, there are no approved disease modifying treatments available for any of the polyQ diseases. Because the disease burden is high, there is a substantial unmet medical need. Lowering levels of the toxic mutant polyQ proteins is a promising potential therapy for polyQ diseases. Possible therapeutic approaches to regulate polyQ protein expression include gene therapy [35], small molecules [36] and RNA modulation modalities targeting polyQ protein transcripts, such as small interfering RNAs (siRNA) [37] and antisense oligonucleotides (ASO) [38]. These approaches can reduce polyQ protein levels either in a non-allele-preferential manner, affecting both wild-type and mutant polyQ protein levels, or in an allele-preferential manner, mainly targeting the mutant isoform. Allele-preferential reduction of the polyQ proteins may be desirable considering the potential loss-of-function contribution to disease progression mentioned before. Indeed, it has been shown that knock-out of wild-type HTT in YAC128 HD mice [39] and wild-type ATXN1 in 154Q ATXN1 knock-in SCA1 mice [40] exacerbates the disease phenotype. Furthermore, haploinsufficiency of the polyQ protein calcium voltage-gated channel subunit alpha1 A (CACNA1A) causes various neurological disorders in humans [41]. Together, this highlights the importance of maintaining a certain level of wild-type polyQ protein expression. However, the exact level of expression needed to maintain physiological cellular functions is unclear and will differ per polyQ protein.

Although many therapeutic approaches are still in early development, both non-allele-preferential and allele-preferential polyQ protein-lowering compounds have reached the clinical trial phase. Three examples of non-allele-preferential therapies that are being tested in clinical trials are PTC518, AMT-130 and Tominersen. PTC518 is a small molecule which promotes splicing of a pseudo-exon from within intron 49 into mature HTT mRNA, leading to nonsense mediated decay and lowering of wild-type HTT protein levels in healthy volunteers in a phase 1 clinical trial [42]. The efficacy of the small molecule is currently being further examined in HD patients in a phase 2a/b trial (ClinicalTrial.gov identifier: NCT05358717, registered on 27 th of April 2022/NCT06254482). AMT-130 is a gene therapy consisting of an expression cassette delivered by an adeno-associated virus (AAV) that codes for an miRNA, which binds and degrades HTT transcripts non-allele-preferentially, reducing mutant and wild-type HTT protein expression. A phase 1/2 clinical trial was initiated in 2019 with AMT-130 in HD patients (NCT04120493). Tominersen is a gapmer ASO that reduces mutant and wild-type HTT protein expression by degrading HTT mRNA in a non-allele-preferential, RNAse H-dependent manner. Although Tominersen failed to meet primary endpoints in a phase 3 clinical trial (NCT03761849) [43], a new phase 2 trial was initiated in 2023 to examine the therapeutic potential of Tominersen in a subpopulation of early manifest and prodromal HD patients (NCT05686551). Several allele-preferential compounds are also being explored in clinical trials. A phase 1b/2a clinical trial with WVE-003, an allele-preferential gapmer ASO for HD, concluded in 2024 (NCT05032196). WVE-003 targets a single nucleotide polymorphism (SNP) often located on mutant HTT alleles. After treatment with WVE-003, lower levels of mutant HTT protein were observed in the cerebrospinal fluid of HD patients, while wild-type HTT expression was preserved [44]. Another allele-preferential ASO is VO659, which directly targets the CAG repeat and has therapeutic potential for the entire class of polyQ diseases. As of 2023, VO659 is being examined in a phase 1/2a clinical trial that includes HD as well as SCA1 and SCA3 patients (NCT05822908). VO659 is a steric blocking ASO that, in contrast to gapmers, does not degrade mRNA. In HD and SCA1, VO659 preferentially reduces mutant polyQ protein expression through CAG-repeat-length-dependent steric hindrance of HTT and ATXN1 protein synthesis [45, 46]. In SCA3, VO659 induces exon-skipping of the CAG-repeat-containing exon 10 of ATXN3, effectively reducing mutant ATXN3 protein levels [46].

Studying the efficacy of potential polyQ protein-lowering compounds is a critical step in preclinical therapy development. It is important to develop methods that accurately quantify both wild-type and mutant polyQ protein levels to allow for precise preclinical efficacy screens and lead compound selection. Conventional Western blot (WB) is often used for this purpose, but is time-consuming, labour-intensive and requires a high protein input per measurement. Additionally, WB results are often variable and poorly reproducible, with minor changes in methodology, experimental design and quantitation method significantly affecting experimental outcomes [47–51]. Therefore, there is a need for more consistent and faster methods to quantify polyQ proteins with higher sensitivity and accuracy.

The Simple Western (SW) capillary immunoassay can accurately detect various proteins with higher sensitivity, a wider dynamic range and considerably shorter run times than WB, while also using a lower protein input [52–54]. In SW, proteins are size-separated using matrix-filled capillaries, followed by incubation with primary and horseradish peroxidase (HRP)-conjugated secondary antibodies for detection of the protein of interest. After incubation with luminol/peroxidase, chemiluminescent signal is measured over the length of the capillary using multiple exposure times. The measured chemiluminescent signal is then converted to a blot-like image and an electropherogram. Subsequently, protein expression is quantified by calculating the area under the curve (AUC) of the relevant peak in chemiluminescence in the electropherogram.

The aim of this study was to develop SW capillary immunoassays to allow for sensitive, reproducible and time-efficient quantification of human wild-type and mutant polyQ proteins. Here, we developed assays for human huntingtin (HTT), ataxin 1 (ATXN1), ataxin 2 (ATXN2), ataxin 3 (ATXN3), atrophin 1 (ATN1) and androgen receptor (AR) proteins. For therapies that allele-preferentially target the mutant protein, it is also important that the developed assays can distinguish between the wild-type and mutant isoforms. Therefore, we examined whether the wild-type and mutant proteins could be size-separated on SW based on polyQ length in polyQ disease patient-derived fibroblasts that express both isoforms. Furthermore, we developed a novel procedure to facilitate size-separation of the large and difficult-to-size-separate wild-type and mutant HTT protein isoforms (≥ 348 kDa depending on polyQ length) using a caspase 3 cleavage step to generate shorter N-terminal HTT fragments that contain the polyQ domain. These methods can be implemented to assess polyQ protein-lowering efficacy in vitro to facilitate screening studies for accurate selection of the most promising candidate molecules, thus contributing to the preclinical development of new polyQ-lowering therapeutic approaches.

Materials and Methods

Cell Lines and Cell Culture

Patient- and healthy control-derived fibroblast cell lines were obtained from the Coriell Institute for Medical Research. The length of the CAG repeat encoding the polyQ expansion in the relevant disease gene was examined with Sanger sequencing (Table 2). For an overview of all examined CAG repeat lengths per cell line, see Online Resource 1. For SCA2 and SBMA, no patient cell lines were available. All fibroblast cell lines were cultured using Minimum Essential medium (MEM) (Gibco, 10370070) supplemented with 15% premium foetal bovine serum (FBS) (Gibco, A5670701), 1% penicillin/streptomycin (Pen/Strep) (Gibco, 15140122) and 1% GlutaMAX supplement (Gibco, 35050038). All cells were cultured at 37 °C with 5% CO₂. For passaging and plating, cells were washed with calcium- and magnesium-free Dulbecco’s phosphate-buffered saline (dPBS) (Gibco, 14190144), released from culture dishes by incubating at 37 °C for 5 min (min) with 0.05% trypsin–EDTA (Gibco, 25300054), followed by replating in a new culture dish with fresh culture medium. To seed a specific number of cells for experiments, live cells were counted using a Luna-II automated cell counter (Logos Biosystems, L40002) by diluting an aliquot of cells 1:1 in 0.4% Trypan Blue (Invitrogen, T10282).

Table 2.

List of fibroblast cell lines

Cell line ID	Disease	Gene	CAG length
			Wild type	Mutant
GM02147	HD	HTT	15	43
GM04022	HD	HTT	18	44
GM21756	HD	HTT	15	68
GM09197	HD	HTT	18	175
GM06927	SCA1	ATXN1	29^a	52
GM06153	SCA3	ATXN3	17	68
GM03561	SCA7	ATXN7	7	56
GM13716	DRPLA	ATN1	16	68
GM02171	-	-	-	-

Open in a new tab

^aCAG length includes two CAT interruptions that code for histidines

Protein Isolation for SW Capillary Immunoassay Optimisation

For SW capillary immunoassay optimisation, a high volume of protein was needed to be able to test multiple conditions with the same sample. To this end, fibroblasts were cultured in 10-cm dishes or T-175 flasks as outlined before, depending on the amount of protein needed. To isolate protein, the fibroblasts were washed twice with dPBS followed by trypsinization with 0.05% trypsin–EDTA for 5 min at 37 °C. The fibroblasts were collected in culture medium, spun down at 300–350 g for 5 min and washed twice with dPBS. After the final wash, the dPBS was aspirated and the cell pellet lysed using 90–400 µL ice cold CelLytic M (Sigma-Aldrich, C2978) supplemented with cOmplete, Mini, EDTA-free protease inhibitor cocktail (Roche, 04693159001) depending on the volume and confluency of the culture flask. The samples were lysed for 15–20 min on ice and vortexed every 2–3 min. After lysis, the samples were spun down at 21,000 g for 10 min at 4 °C to pellet cellular debris. Then, the supernatant protein lysate was aliquoted and stored at − 80 °C. The protein concentration was measured using the Pierce™ BCA protein assay kit (Thermo Scientific, 23,227) according to manufacturer specifications.

siRNA Transfection

To generate negative control samples for SW, fibroblasts were transfected with siRNAs (Invitrogen) targeting the different polyQ proteins (Table 3). To transfect fibroblasts with siRNAs, 3000 cells were plated per well in 96-well plates 3 days prior to transfection. On the day of transfection, cells were washed twice with 150 µL dPBS followed by addition of 150 µL fresh MEM medium containing 10% FBS, 1% Pen/Strep and 1% GlutaMAX. The siRNAs were transfected using 0.17 µL Lipofectamine RNAiMAX (Invitrogen, 13,778,150) per well. The siRNAs and RNAiMAX were diluted separately in Opti-MEM I reduced serum medium (Gibco, 31985062). The siRNA and RNAiMAX dilutions were mixed 1:1 and incubated for 5 min at room temperature (RT) to allow for transfection complex formation, according to the manufacturer’s instructions. Then, 37.5 µL of the siRNA/RNAiMAX Opti-MEM mixture was added per well to the cells, resulting in a final siRNA concentration of 75 nM. For non-treated (NT) samples, 37.5 µL Opti-Mem without siRNA or RNAiMAX was added instead. A non-targeting siRNA (siCtrl) was included in every transfection experiment as control. Protein was isolated from the fibroblasts 4 days after transfection by aspirating the culture medium, washing with 150 µL PBS and adding 12 µL of ice cold CelLytic M supplemented with cOmplete, Mini, EDTA-free protease inhibitor cocktail directly into the wells. The plates were subsequently placed on a shaker (Heidolph, L1298544-11200–00) for 15 min at 900 RPM at 4 °C to lyse the cells, followed by freezing and storage at − 80 °C.

Table 3.

siRNAs and target genes

Target gene	siRNA ID (Invitrogen)
ATXN1	142073
ATXN2	12415
ATXN3	s230539
ATN1	106622
AR	s1539
HTT	s6490
No-target siControl	AM4635

Open in a new tab

SW Capillary Immunoassay Procedure

All SW capillary immunoassay experiments were performed on Wes (Bio-Techne, ProteinSimple, #004–600) or Jess (Bio-Techne, ProteinSimple, #004–650) systems. Although the systems can be used interchangeably, the Jess system has RePlex capabilities, allowing for a second staining in the same capillary to control for protein loading. The Jess system was used for siRNA-transfected samples to stain for total protein to correct for protein input using the RePlex capabilities, as no BCA was performed for the siRNA-transfected samples due to the lower protein yield of the 96-well culture plates. The Wes system was used for further optimisation of selected antibodies.

The Wes and Jess SW capillary immunoassay runs were performed according to the manufacturer’s specifications. Either 12–230 kDa (Bio-Techne, SM-W004) or 66–440 kDa (Bio-Techne, SMW-008) separation modules were used depending on protein size, in combination with a No secondary Detection Module (Bio-Techne, DM-003). In short, for Jess runs, 2 µL (1 µL for HTT assays) protein sample was diluted in 1.5 µL (2.5 µL for HTT assays) 0.1 × sample buffer and 0.9 µL Fluorescent Master Mix. For Wes runs, protein samples were diluted to specific protein concentrations (0.0125–0.8 mg/mL) in a volume of 4.4 µL using 0.1 × sample buffer and 0.9 µL Fluorescent Master Mix. Then, for both Wes and Jess runs, the samples were denatured at 95 °C for 5 min before loading 3 µL sample on the SW plate together with the ladder, antibody diluent, primary antibody diluted in antibody diluent (for the complete list of all tested antibodies see Online Resource 2), HRP-conjugated rabbit secondary antibody (Bio-Techne, 042–206), luminol/peroxide and wash buffer. For Jess runs, several additional reagents were pipetted on the plate for the RePlex total protein staining as per the manufacturer’s specifications. This included the RePlex purge solution mix (Bio-Techne, RP-001), the total protein labelling, and the total protein streptavidin-HRP detection solutions (Bio-Techne, DM-TP01).

After loading all reagents and samples, the SW plate was spun down at 2000 g for 5 min and placed in the appropriate SW system together with the matrix-filled separation capillaries (included in the separation module). The default run settings were used for both Wes protein and Jess protein-total protein RePlex assays. Chemiluminescence was detected using the standard high dynamic range (HDR) detection profile for Wes or the RePlex dynamic range (RDR) for Jess.

All SW runs were performed and analysed using the Compass software (Bio-Techne). To quantify polyQ protein signal, the electropherogram resulting from the chemiluminescence detection was examined and the automated peak and baseline detection were adjusted manually if necessary. The area under the curve (AUC) of a peak of interest was calculated by the Compass software. For total protein correction of Jess assays, the values of the AUCs of peaks of interest were corrected for the values of the AUCs of the total protein RePlex assay. To determine whether the AUC of a peak of interest in the electropherogram was sufficiently high for further assay development with a specific antibody, a signal-to-noise (S/N) threshold calculated by the Compass software of > 10 was used.

Caspase 3 HTT Cleavage Assay

To facilitate size-separation of wild-type and mutant HTT protein on SW based on polyQ length, protein lysate was treated with caspase 3 to generate smaller N-terminal HTT fragments that contain the polyQ region. Recombinant human caspase 3 (R&D systems, 707-C3/CF) was diluted to a concentration of 6–48 µg/mL in assay buffer (0.1% CHAPS, 10 mM DTT, 25 mM HEPES). Protein lysate was diluted to the desired protein concentration in assay buffer if needed and subsequently mixed 1:1 with the caspase 3 dilution resulting in a final caspase 3 concentration of 3–24 µg/mL. The final reaction volume was 4 µL for the siRNA transfection samples (2 µL undiluted protein lysate + 2 µL 12 µg/mL caspase 3) and 10 µL for optimisation experiments (5 µL protein lysate dilution + 5 µL 6–48 µg/mL caspase 3). The protein lysate/caspase 3 mixture was incubated for 2–4 h (h) at 37 °C and subsequently analysed with SW as described above using the N-terminal HTT antibody ab109115 (Abcam).

Results

To accurately determine expression levels of polyQ proteins, a large panel of 41 primary antibodies (Online Resource 2) was screened to select those that performed best for further assay development. The following stepwise approach was taken for each polyQ protein assay:

First, we examined whether using the selected antibodies on SW resulted in a peak in chemiluminescence with an S/N ratio of > 10 around the predicted molecular weight (MW) of the target protein. Second, to confirm that the detected peak was the polyQ protein of interest, we examined whether the AUC of the peak decreased in lysates of cells transfected with an siRNA targeting the relevant polyQ protein. Additionally, recombinant protein, if available, was included as positive control to verify the observed MW of the polyQ protein on SW. Third, once it had been confirmed that an antibody detected a specific polyQ protein, the assay conditions were further optimised. The optimal antibody concentration was determined by performing an antibody dilution series to establish the point of epitope saturation, after which increasing the antibody concentration does not result in higher chemiluminescent signal. The linear (e.g. quantitative) range of an assay was assessed by performing a protein loading dilution series.

This resulted in the development of SW capillary immunoassays for six out of the nine polyQ proteins. Despite testing eight antibodies for ATXN7 and six antibodies for TBP, no SW assays were optimised for these proteins as none of the tested antibodies showed promising initial results, except for the ab220788 TBP antibody (Abcam) which showed a peak around the predicted MW of TBP (38 kDa). However, further testing revealed high background and inconsistent TBP signal (Online Resource 3), so the antibody was not further optimised. For CACNA1A, only two antibodies were acquired, neither of which showed promising results. The antibodies that were not selected for further assay development either showed lower signal than the selected antibody, no signal in the range of the predicted MW of the specific polyQ protein, did not meet the S/N threshold or showed no reduction of the AUC of the peak of interest in the siRNA-transfected samples. Examples of antibodies that were not selected for further SW assay development are found in Online Resource 4. For ATXN1, ATXN2, ATXN3, ATN1, AR and HTT, promising antibodies were identified and SW capillary immunoassays developed, which will be further discussed per protein.

ATXN1

For ATXN1, 10 antibodies were screened of which the recombinant monoclonal anti-ATXN1 antibody 16H8L13 (Invitrogen, #703273) showed the most promising results. A sharp main peak in chemiluminescence was observed around 115 kDa with S/N ratios of > 540, using the 16H8L13 antibody on protein lysate of Huntington’s disease (HD) patient fibroblasts (GM21756, 29/30 CAG in ATXN1) (Fig. 1a, b). The peak ran at a higher MW than the predicted MW of ATXN1 (87 kDa). However, transfection with 75 nM ATXN1-targeted siRNA (siATXN1) confirmed that the peak represented wild-type ATXN1 (WtATXN1), as the AUC of the peak decreased by ~ 86% compared to NT, while no decrease was observed in the 75 nM siCtrl samples (Fig. 1c). Additionally, the 16H8L13 antibody was also able to detect a recombinant ATXN1 protein (Origene, TP322862), which ran at a similar height (~ 118 kDa) as endogenous WtATXN1 (Fig. 1a). Furthermore, in SCA1 patient fibroblast (GM06927, 29/52 CAG in ATXN1) lysate, a second peak at 130 kDa which was not present in HD patient fibroblast lysate was observed, which corresponded to mutant ATXN1 (MtATXN1) (Fig. 1a, d). Similar to WtATXN1, the AUC of the MtATXN1 peak decreased by nearly 100% after transfection with 75 nM siATXN1. WtATXN1 and MtATXN1 signal also decreased in siCtrl-transfected SCA1 patient fibroblasts, but to a lesser extent (~ 11% and 19%, respectively) (Fig. 1e). Regardless, these results confirmed that 16H8L13 detected both WtATXN1 and MtATXN1 proteins on SW. The relatively small difference in MW of WtATXN1 and MtATXN1 resulted in peaks that did not fully size-separate, but the individual peaks could still be quantified accurately using the AUC fitting of the Compass software when using the 66–440 kDa separation module. Strikingly, there was no size-separation of WtATXN1 and MtATXN1 when using the 12–230 kDa separation module, as only a single total ATXN1 (TotATXN1) peak was observed (Fig. 1f). Therefore, further optimisation of the ATXN1 SW capillary immunoassay was performed using the 66–440 kDa separation module.

Fig. 1 — ATXN1 SW capillary immunoassay optimisation. a Representative lane view of a 66–440 kDa separation module SW run with transfected HD (GM21756) or SCA1 (GM06927) patient fibroblasts protein lysate and 1 µg/mL 16H8L13 anti-ATXN1 antibody. In the right most lane, 1.25 ng/mL recombinant ATXN1 was loaded. b Electropherogram of the HD patient fibroblast lanes in a. The S/N ratio of the NT peak is indicated in blue. c Relative WtATXN1 protein levels in untreated (NT), 75 nM siCtrl or 75 nM siATXN1-transfected HD patient fibroblast (GM21756) lysate measured using 1 µg/mL 16H8L13. Corrected for total protein input using Jess RePlex. NT was set at 100% and data is shown as averages with standard deviations (SD) of 3 biological replicates. d Electropherogram of the SCA1 patient fibroblast capillaries in a. e Relative WtATXN1 and MtATXN1 protein levels in NT, 75 nM siCtrl or 75 nM siATXN1-transfected SCA1 patient fibroblast (GM06927) lysate measured using 1 µg/mL 16H8L13. Corrected for total protein input using Jess RePlex. NT was set at 100% and data is shown as averages with SD of 3 biological replicates. f Representative electropherogram of 0.5 mg/mL SCA1 patient fibroblast (GM06927) lysate loaded on a 12–230 kDa separation module plate detected using 1 µg/mL 16H8L13. g Antibody dilution series of 0.167–10 µg/mL 16H8L13 using 0.1, 0.3 or 0.5 mg/mL of protein lysate from NT control fibroblasts (GM02171). Data was expressed as percentage of the average AUC value of the WtATXN1 peaks per plate. Data points represent averages of 2 technical replicates with SD. Saturation curves (one site – total binding) were fitted to the data using GraphPad Prism. h Protein dilution series of 0.05–0.8 mg/mL of NT control fibroblast (GM02171) lysate using 1 µg/mL 16H8L13. Data was expressed as percentage of the average AUC value of the WtATXN1 peaks per plate. Data points represent averages of 2 measurements with SD. A sigmoidal 4 parameter curve was fit to the data using GraphPad Prism. The plotted dotted line represents the linear part of the curve

To determine the optimal conditions to measure ATXN1 protein expression using the anti-ATXN1 antibody 16H8L13, both antibody and protein loading dilution series were performed. The antibody dilution series in control fibroblast (GM02171, 32/35 CAG in ATXN1) lysate showed a steady increase in ATXN1 signal up to 1.0 µg/mL. Further increasing the antibody concentration only resulted in marginal increases in chemiluminescence, indicating epitope saturation (Fig. 1g). Using the optimal 1.0 µg/mL 16H8L13 antibody resulted in high ATXN1 signal and low background, with S/N ratios of > 600. Performing a protein loading dilution series using control fibroblast (GM02171) lysate revealed that this ATXN1 assay was linear between 0.1 and 0.6 mg/mL (Fig. 1h). Therefore, to examine efficacy of ATXN1-lowering compounds, a maximum protein loading concentration of 0.6 mg/mL is recommended using 1.0 µg/mL of the 16H8L13 anti-ATXN1 antibody.

ATXN2

Initial screening of three ATXN2 antibodies identified the recombinant monoclonal anti-ATXN2 antibody ab254362 (Abcam) as the most promising. A strong but broad peak in chemiluminescence with S/N ratios of > 1200 was observed at 173 kDa (Fig. 2a, b). The peak ran at a somewhat higher MW than the predicted MW of ATXN2 of 140 kDa. No other non-specific peaks were observed, aside from slight background signal between 50 and 140 kDa. The AUC of the 173 kDa peak decreased by ~ 86% after transfection with an ATXN2-targeted siRNA (siATXN2) compared to NT HD patient fibroblasts (GM21756, 22/23 CAG in ATXN2). Remarkably, the non-targeting siCtrl also reduced ATXN2 signal by ~ 46% (Fig. 2c). Although no SCA2 lysate or recombinant ATXN2 was available, the siATXN2 transfection still suggested that the peak in chemiluminescence detected by ab254362 on SW represented ATNX2, even with the observed decrease in AUC induced by the siCtrl. Therefore, the ATXN2 assay using the ab254362 antibody was further optimised.

Fig. 2 — ATXN2 SW capillary immunoassay optimisation. a Representative lane view of a 12–230 kDa separation module plate loaded with protein lysate from NT, 75 nM siCtrl or 75 nM siATXN2-transfected HD patient fibroblasts (GM21756) and 6.7 µg/mL ab254362 anti-ATXN2 antibody. b Electropherogram of lane view in a. c Relative WtATXN2 protein levels in NT, 75 nM siCtrl or 75 nM siATXN2-transfected HD patient fibroblast (GM21756) protein lysate measured using 6.7 µg/mL ab254362. Corrected for total protein input using Jess RePlex. NT was set at 100% and data is shown as averages with SD of 3 biological replicates. d Antibody dilution series of 0.5–50.5 µg/mL ab254362 using 0.1, 0.3 or 0.5 mg/mL of protein lysate from NT control fibroblasts (GM02171). Data was expressed as percentage of the average AUC value of the WtATXN2 peaks per plate. Data points represent averages of 2 technical replicates with SD. Saturation curves (one site – total binding) were fit to the data. e Protein dilution series of 0.05–0.8 mg/mL of NT control fibroblast (GM02171) lysate using 6.7 µg/mL ab254362. Data was expressed as percentage of the average AUC value of the WtATXN2 peaks per plate. Data points represent averages of 2 measurements with SD. A sigmoidal 4 parameter curve was fit to the data. The plotted dotted line represents the linear part of the curve

The optimal assay conditions to quantify ATXN2 using ab254362 were determined by performing both antibody and protein loading dilutions curve experiments. ATXN2 signal increased up to an antibody concentration of 3.4 µg/mL when loading up to 0.5 mg/mL control fibroblast (GM02171, 22/22 CAG in ATXN2) lysate. Further increasing the antibody concentration up to 50.5 µg/mL only minimally increased chemiluminescent signal (Fig. 2d). The protein loading dilution series using control fibroblast (GM02171) lysate revealed assay linearity between 0.1 and 0.5 mg/mL (Fig. 2e) and high S/N ratios of > 250 even with lower protein inputs. Therefore, loading up to 0.5 mg/mL protein lysate is recommended using 3.4 µg/mL of the ab254362 anti-ATXN2 antibody for the optimal assay to measure ATXN2 protein levels.

ATXN3

For ATXN3, three antibodies were screened of which the recombinant monoclonal anti-ATXN3 antibody ab221143 (Abcam) showed the clearest signal. When using the ab221143 antibody, a sharp peak in chemiluminescence was observed around 49 kDa, close to the predicted MW of wild-type ATXN3 (WtATXN3) of 43 kDa, in HD patient fibroblast lysate (GM21756, 8/8 CAG in ATXN3) (Fig. 3a, b). Background levels were low, corresponding to high S/N ratios of > 550. Additionally, a recombinant ATXN3 protein (MyBioSource, MBS203345) was recognised by the ab221143 antibody, which ran at a MW of 54 kDa, similar to endogenous WtATXN3 (Fig. 3a). In samples transfected with 75 nM ATXN3-targeted siRNA (siATXN3), the AUC of the WtATXN3 peak decreased by ~ 71% compared to NT (Fig. 3c). In SCA3 patient fibroblast (GM06153, 17/68 CAG in ATXN3) lysate, a second peak that was not observed in the HD patient fibroblasts appeared around 70 kDa, corresponding to mutant ATXN3 (MtATXN3) (Fig. 3a, d). Additionally, the WtATXN3 peak in the SCA3 patient fibroblasts (GM06153) ran at a higher MW of 52 kDa than the WtATXN3 peak in the HD patient fibroblasts (GM21756) due to the difference in repeat lengths of 17 CAG and 8 CAG, respectively, in the WtATXN3 alleles. Like WtATXN3, MtATXN3 signal decreased by ~ 84% compared to NT after transfection with 75 nM siATXN3 (Fig. 3e). Interestingly, increases in ATXN3 levels were observed after transfection with 75 nM siCtrl of ~ 30% in HD (GM21756) and ~ 14%/18% (WtATXN3/MtATXN3) in SCA3 (GM06153) patient fibroblasts. Furthermore, in SCA3 patient fibroblast (GM06153) lysate, a peak was observed at 44 kDa that was not reduced by siATXN3 (Fig. 3d). In the HD patient fibroblasts, this peak was mostly hidden by the WtATXN3 peak, only appearing as a shoulder, which decreased with lower antibody concentrations (Fig. 3f).

Optimising the ATXN3 assay showed that epitope saturation was reached at 10.4 µg/mL in HD patient fibroblast (GM21756) lysate (Fig. 3g). The protein loading dilution series revealed assay linearity between 0.05 and 0.6 mg/mL protein lysate (Fig. 3h). When loading 0.5 mg/mL HD patient fibroblast lysate and 10.4 µg/mL antibody, high S/N ratios of > 1400 were observed. Therefore, for optimal and quantitative ATXN3 signal, loading up to 0.6 mg/mL of protein lysate is recommended using 10.4 µg/mL of the ab221143 antibody.

ATN1

Of the five tested ATN1 antibodies, the polyclonal NBP1-90044 (Novus) anti-ATN1 antibody yielded the best results. In NT HD patient fibroblast (GM21756, 12/16 CAG in ATN1) lysate, a narrow peak in chemiluminescence was observed at 194 kDa using the NBP1-90044 antibody, with S/N ratios of > 425 (Fig. 4a, b). The AUC of this peak decreased by 95% after transection with 75 nM ATN1-targeted siRNA (siATN1) compared to NT (Fig. 4c). Similar to WBs in literature, the identified wild-type ATN1 (WtATN1) peak ran higher than the predicted MW of ATN1 of 125 kDa [55–57]. Additionally, two other peaks were observed at 123 and 148 kDa which decreased after siATN1 siRNA transfection, which possibly corresponded to proteolytic cleavage products of ATN1 [55–58]. However, not every lot of the polyclonal NBP1-90044 antibody detected these peaks, although the main ATN1 peak was always clearly present (data not shown). Furthermore, in DRPLA patient fibroblast (GM13716, 16/68 CAG in ATN1) lysate, an additional peak with a higher MW (225 kDa) was observed. This 225 kDa peak was not present in the HD patient fibroblasts and therefore likely corresponded to mutant ATN1 (MtATN1) (Fig. 4a, d). The MWs of WtATN1 and MtATN1 were too close to allow complete size-separation of the peaks on the 66–440 kDa separation module (Fig. 4d). However, the peaks were sufficiently distinct for separate AUC fitting and quantification of the WtATN1 and MtATN1 protein isoforms. Indeed, WtATN1 decreased by ~ 92% and MtATN1 decreased by ~ 93% after transfection with 75 nM siATN1 in the DRPLA patient fibroblasts (Fig. 4e). Comparable to previously discussed polyQ proteins, transfection with 75 nM siCtrl also reduced ATN1 protein expression in both HD and DRPLA patient fibroblast lysate, by 45% (WtATN1) and 18%/35% (WtATN1/MtATN1), respectively. Although no recombinant ATN1 protein was available as additional positive control, the siRNA results and presence of a mutant peak in DRPLA patient fibroblast lysate confirmed that the NBP1-90044 antibody detected ATN1 on SW. It should be noted that WtATN1 and MtATN1 did not size-separate when using the 12–230 kDa separation module, as only a single total ATN1 (TotATN1) peak could be observed (Fig. 4f). Therefore, usage of the 66–440 kDa separation module is recommended for ATN1.

Fig. 4 — ATN1 SW capillary immunoassay optimisation. a Representative lane view of a Jess run using a 66–440 kDa kit plate loaded with protein lysate from NT, 75 nM siCtrl or 75 nM siATN1-transfected GM21756 (HD) or GM13716 (DRPLA) patient fibroblasts and 4.8 µg/mL NBP1-90044 anti-ATN1 antibody. b Electropherogram of the HD patient fibroblast lanes in a. c Relative WtATN1 protein levels in NT, 75 nM siCtrl or 75 nM siATN1-transfected HD patient fibroblast (GM21756) lysate measured using 4.8 µg/mL NBP1-90044. Corrected for total protein input using Jess RePlex. NT was set at 100% and data is shown as averages with standard deviations (SD) of 3 biological replicates. d Electropherogram of the DRPLA patient fibroblast capillaries in a. e Relative WtATN1 (blue) and MtATN1 (green) protein levels in NT, 75 nM siCtrl or 75 nM siATN1-transfected DRPLA patient fibroblast (GM13716) lysate measured using 4.8 µg/mL NBP1-90044. Corrected for total protein input using Jess RePlex. NT was set at 100% and data is shown as averages with SD of 3 biological replicates. f Representative electropherogram of 0.5 mg/mL DRPLA patient fibroblast (GM13716) lysate loaded on a 12–230 kDa module plate detected using 2.4 µg/mL NBP1-90044. g Antibody dilution series of 0.24–12.0 µg/mL NBP1-90044 using 0.1, 0.3 or 0.5 mg/mL of protein lysate from NT control fibroblasts (GM02171). Data was expressed as percentage of the average AUC value of the WtATN1 peaks per plate. Data points represent averages of 2 technical replicates with SD. Saturation curves (one site – total binding) were fit to the data using GraphPad Prism. h Protein dilution series of 0.05–0.8 mg/mL of NT control fibroblast (GM02171) lysate using 4.8 µg/mL NBP1-90044. Data was expressed as percentage of the average AUC value of the WtATN1 peaks per plate. Data points represent averages of 2 measurements with SD. A sigmoidal 4 parameter curve was fit to the data using GraphPad Prism. The plotted dotted line represents the linear part of the curve

The optimal assay conditions were examined as previously described. The antibody dose–response curve showed increasing chemiluminescence ATN1 signal up to the highest tested concentration of 12 µg/mL, a 1:10 dilution of the antibody stock, in control fibroblast (GM02171, 8/15 CAG in ATN1) lysate. The curve levelled off, but did not reach complete epitope saturation at 12 µg/mL antibody. Since using concentrations higher than 12 µg/mL would consume an unreasonably high amount of antibody, and considering that ATN1 chemiluminescence signal was sufficiently high, a NBP1-90044 concentration between 4.8 and 12 µg/mL was considered appropriate (Fig. 4g). The protein loading dilution series revealed that the ATN1 assay was linear when loading 0.05 to 0.6 mg/mL of protein lysate (Fig. 4h) when using 4.8 µg/mL of the NBP1-90044 antibody. Therefore, to quantify ATN1 protein levels, using 4.8–12 µg/mL NBP1-90044 anti-ATN1 antibody is recommended, which yielded high S/N ratios (> 630) in 0.5 mg/mL control fibroblast (GM02171) lysate, with loading concentrations up to 0.6 mg/mL fibroblast protein lysate.

AR

A total of four AR antibodies were screened. The recombinant monoclonal anti-AR antibody mAb#5153 (Cell Signaling Technology) showed more promising results than the other antibodies. A peak in chemiluminescence was observed in HD patient fibroblast (GM21756, 18/18 CAG in AR) lysate at 112 kDa, around the predicted molecular weight of AR of 99 kDa (Fig. 5a, b). At 159 kDa, a low intensity non-specific peak was observed. At 44 kDa, a peak was observed that likely does not correspond to AR or an AR isoform, but to the smaller fluorescent standard protein that is included in every sample as part of the SW run, as this peak appears in all SW runs using the 66–440 kDa module regardless of antibody. This peak could also be observed in the previously discussed SW assays but seems more prominent here due to the relatively lower signal of the AR peak. Although background signal was higher and AUC lower than in previously discussed assays, the S/N ratios of the 112 kDa peak (> 43 S/N) were well above the set threshold. As no SBMA cell line or recombinant AR protein was available, an AR-targeted siRNA (siAR) was used to confirm that the 112 kDa peak represented wild-type AR (WtAR). Indeed, transfection with 75 nM siAR in HD patient fibroblasts (GM21756) decreased the AUC of the 112 kDa peak by ~ 95% compared to NT, while transfection with 75 nM siCtrl decreased the AUC by only ~ 26% (Fig. 5c). No clear reduction of the AUC of the 44 and 159 kDa peaks was observed after 75 nM siAR transfection. These results suggest that the 112 kDa peak detected by mAb#5153 indeed represented WtAR.

Fig. 5 — AR SW capillary immunoassay optimisation. a Representative lane view of a 12–230 kDa module plate loaded with protein lysate from NT, 75 nM siCtrl or 75 nM siAR-transfected HD (GM21756) patient fibroblasts and 8.2 µg/mL mAb#5153 anti-AR antibody. b Electropherogram of lane view in a. c Relative WtAR protein levels in NT, 75 nM siCtrl or 75 nM siAR-transfected HD patient fibroblast (GM21756) protein lysate measured using 8.2 µg/mL mAb#5153. Corrected for total protein input using Jess RePlex. NT was set at 100% and data is shown as averages with SD of 3 biological replicates. d Antibody dilution series of 0.33–16.4 µg/mL mAb#5153 using 0.1, 0.3 or 0.5 mg/mL of protein lysate from NT HD patient fibroblasts (GM21756). Data was expressed as percentage of the average AUC value of the WtAR peaks per plate. Data points represent averages of 2 technical replicates with SD. Saturation curves (one site – total binding) were fit to the data. e Protein dilution series of 0.05–0.8 mg/mL of NT HD patient fibroblast (GM21756) lysate using 8.2 µg/mL mAb#5153. Data was expressed as percentage of the average AUC value of the WtAR peaks per plate. Data points represent averages of 2 measurements with SD. A sigmoidal 4 parameter curve was fit to the data. The plotted dotted line represents the linear part of the curve

An antibody dilution series revealed the point of epitope saturation. The AUC of the 112 kDa WtAR peak increased with antibody concentrations up to 3.3 µg/mL when loading lysate from NT HD patient fibroblasts (GM21756). Further increasing the antibody concentration only resulted in marginal increases in chemiluminescence (Fig. 5d). Performing a protein loading dilution series with HD patient fibroblast (GM21756) lysate showed that this AR assay was linear when loading between 0.1 and 0.5 mg/mL (Fig. 5e). When loading 0.5 mg/mL HD patient fibroblast (GM21756) lysate and 3.3 µg/mL antibody, S/N ratios of > 240 were observed. Therefore, to measure AR protein levels using SW, it is recommended to use 3.3 µg/mL of the mAb#5153 and to maximally load 0.5 mg/mL fibroblast protein lysate.

HTT

For HTT, the monoclonal anti-N-terminal HTT antibody ab109115 (Abcam) is commonly used in literature on WB [59, 60]. This antibody showed good results when used on SW as well, so no additional antibodies were tested. In HD patient fibroblast (GM21756, 15/68 CAG in HTT) lysate, both wild-type HTT (WtHTT) (347 kDa) and mutant HTT (MtHTT) (354 kDa) proteins were present, which were difficult to size-separate due to the high molecular weight of the full-length protein and the small relative difference in MW between the wild-type and mutant protein isoforms. On the 66–440 kDa separation module, only a single clear peak in chemiluminescence with S/N ratios of > 220 was observed at 307 kDa using the ab109115 antibody (Fig. 6a, b). Transfection with a HTT-targeted siRNA (siHTT) decreased the AUC of the 307 kDa peak in lysate of HD patient fibroblasts (GM21756) by ~ 74%, while the siCtrl increased HTT signal by ~ 38% (Fig. 6c). This confirmed that the 307 kDa peak represented total HTT (TotHTT). To attempt to size-separate MtHTT from WtHTT, the SW running time was increased. The standard running time was 30 min at 475 V when using the 66–440 kDa separation modules. Increasing the running time to 35 or 40 min did not result in better allele separation (Fig. 6d, e) when loading HD patient fibroblast (GM09197, 18/175 CAG in HTT) lysate. Further elongating the running time to 45 min did not increase allele separation and caused the smallest fluorescent standard and the 66/116 kDa ladder proteins to run of the capillary (Fig. 6f). Both are necessary for accurate determination of protein sizes across the capillary and correct plotting of the electropherogram by the Compass software. As seen in Fig. 6f, inconsistencies in estimated MW between the blot-like lane view and the exported electropherogram can be observed. While the remaining 200/280/440 kDa ladder proteins are plotted normally in the lane view, the MW of the HTT peak in the electropherogram is wrongly estimated at ~ 467 kDa, resulting in unreliable results of the SW run when the running time is too long. Therefore, another approach to size-separate WtHTT from MtHTT was necessary.

Fig. 6 — TotHTT SW capillary immunoassay optimisation. a Representative lane view of a 66–440 kDa module plate loaded with protein lysate from NT, 75 nM siCtrl or 75 nM siHTT-transfected HD patient fibroblasts (GM21756) and 15.2 µg/mL ab109115 anti-HTT antibody. b Electropherogram of lane view in a. c Relative TotHTT protein levels in NT, 75 nM siCtrl or 75 nM siHTT-transfected HD patient fibroblast (GM21756) protein lysate measured using 15.2 µg/mL ab109115. Corrected for total protein input using Jess RePlex. NT was set at 100% and data is shown as averages with SD of 3 biological replicates. d–f Representative electropherograms and lane views of 0.1 mg/mL protein lysate from HD patient fibroblasts (GM09197) loaded on 66–440 kDa module plates with increasing running times (35–45 min) at 475 V. g Antibody dilution series of 1.5–152.4 µg/mL ab109115 using 0.1, 0.3 or 0.5 mg/mL of protein lysate from NT control fibroblasts (GM02171). Data was expressed as percentage of the average AUC value of the WtHTT peaks per plate. Data points represent averages of 2 technical replicates with SD. Saturation curves (one site – total binding) were fit to the data. h Protein dilution series of 0.025–0.6 mg/mL of NT control fibroblast (GM02171) lysate using 15.2 µg/mL ab109115. Data was expressed as percentage of the average AUC value of the WtHTT peaks per plate. Data points represent averages of 2 measurements with SD. A sigmoidal 4 parameter curve was fit to the data. The plotted dotted line represents the linear part of the curve

However, before proceeding with size-separation experiments, the optimal assay conditions to quantify TotHTT in HD cell lines or WtHTT in control cell lines using the ab109115 antibody were determined. The antibody dilution series showed that epitope saturation was reached at 15.2 µg/mL antibody when using NT control fibroblast (GM02171, 15/17 CAG in HTT) lysate. Further increasing the antibody concentration resulted in decreased and variable HTT signal (Fig. 6g). Performing a protein loading dilution series using control fibroblast (GM02171) lysate revealed that full-length WtHTT signal was linear between 0.025 and 0.2 mg/mL (Fig. 6h). Therefore, for quantification of TotHTT or full-length WtHTT protein levels, maximally 0.2 mg/mL fibroblast protein lysate should be loaded, using the 15.2 µg/mL dilution of the ab109115 anti-HTT antibody, for high chemiluminescence signal with high S/N ratios (> 90) even at lower protein inputs.

HTT Caspase 3 Cleavage Assay

Because WtHTT and MtHTT did not readily size-separate using SW, a novel caspase 3 cleavage assay was developed. The N-terminal part of the HTT protein contains two main caspase 3 cleavage sites after the polyQ stretch at amino acids 510 and 549 (based on a polyQ length of 22) [61, 62] (Fig. 7a). The presence of these sites was utilised to develop a HTT cleavage assay, allowing for generation of shorter N-terminal WtHTT (N-WtHTT) and MtHTT (N-MtHTT) fragments that contain the polyQ region, which can be detected by the ab109115 antibody. Whether these smaller fragments size-separated more readily than full-length HTT due to the larger relative difference in MW between N-WtHTT (~ 56 kDa) and N-MtHTT (> 59 kDa depending on Q length) was examined. Protein samples from NT, 75 nM siHTT or 75 nM siCtrl-transfected HD patient fibroblasts (GM21756, 15/68 CAG in HTT) were treated with caspase 3 and subsequently ran on SW (Fig. 7b, c). While in the no-caspase 3 control only the single TotHTT peak was observed at 307 kDa, two lower MW peaks corresponding to the N-WtHTT and N-MtHTT fragments appeared at 77 and 120 kDa, respectively, after caspase 3 treatment. In protein lysate of HD patient fibroblasts (GM21756) transfected with 75 nM siHTT, the AUC of the 77 and 120 kDa peaks decreased by ~ 73% and 77%, respectively, confirming these peaks to be the N-terminal HTT fragments (Fig. 7d). Similarly to the TotHTT assay, the AUC of the N-WtHTT and N-MtHTT peaks increased by ~ 36% and 40%, respectively, after transfection with the siCtrl.

Fig. 7 — Caspase 3 assay to size-separate N-WtHTT and N-MtHTT on SW. a Schematic of caspase 3 cleavage of full-length Wt/MtHTT and detection with an N-terminal antibody. Table shows predicted molecular weights of the full-length and the N-terminal cleavage products of Wt/MtHTT and their relative sizes compared to WtHTT. b Representative lane view using a 66–440 kDa module plate. Protein lysate from NT, 75 nM siCtrl or 75 nM siHTT-transfected HD patient fibroblasts (GM21756) treated with caspase 3 for 2 h at 37 °C, or assay buffer as negative control, was loaded on the plate and detected with 15.2 µg/mL ab109115 N-terminal HTT antibody. c Electropherogram of lane view in b. d Relative N-WtHTT (blue) and N-MtHTT (green) protein levels in caspase 3 treated NT, 75 nM siCtrl or 75 nM siHTT-transfected HD patient fibroblast (GM21756) protein lysate measured using 15.2 µg/mL ab109115. Corrected for total protein input using Jess RePlex. NT set at 100% and data is shown as averages with SD of 3 biological replicates. e HD protein lysate (GM21756) was incubated with increasing amounts of caspase 3 (0–24 µg/mL) for either 2 h (uninterrupted lines) or 4 h (dotted lines) after which 0.07 mg/mL protein lysate was loaded and detected using 15.2 µg/mL ab109115. Full-length TotHTT (red), N-WtHTT (blue) and N-MtHTT (green) were quantified. f, g Antibody dilution series of 1.5–152.4 µg/mL ab109115 using 0.1, 0.25 or 0.4 mg/mL of protein lysate from HD patient fibroblasts (GM21756). Data was expressed as percentage of the average AUC value of the f N-WtHTT or g N-MtHTT peaks per plate. Data points represent averages of 2 technical replicates with SD. Saturation curves (one site – total binding) were fitted to the data. h Protein dilution series of 0.0125–0.4 mg/mL of caspase 3 treated HD patient fibroblast (GM21756) lysate using 15.2 µg/mL ab109115. Data was expressed as percentage of the average AUC value of the N-WtHTT (blue) or N-MtHTT (green) peaks per plate. Data points represent averages of 2 measurements with SD. A sigmoidal 4 parameter curve was fitted to the data. The dotted line represents the linear part of the curve. i Peak view of capillaries from different 66–440 module kDa SW runs using 15.2 µg/mL ab109115. Top row was loaded without caspase 3 treatment. Bottom row was incubated with 6 µg/mL caspase 3 (R&D) for 1 h at 37 °C prior to loading. From left to right: GM06153 SCA3 and GM03561 SCA7 patient fibroblast lysate (both 15/17 CAG in *HTT*), GM02147 HD patient fibroblast lysate (15/43 CAG in *HTT*), GM21756 HD patient fibroblast lysate (15/68 CAG in *HTT*), GM09197 HD patient fibroblast lysate (18/175 CAG in *HTT*)

Next, we further optimised the caspase 3 reaction by incubating 0.26 mg/mL HD patient fibroblast (GM04022, 18/44 CAG in HTT) protein lysate with increasing amounts of caspase 3 (3–24 µg/mL) for 2 h or 4 h. After incubation, the caspase 3-treated protein lysates were run on SW using the 66–440 kDa separation module at a concentration of 0.07 mg/mL with 15.2 µg/mL ab109115. Full-length TotHTT, N-WtHTT and N-MtHTT were quantified to establish optimal caspase 3 incubation conditions (Fig. 7e). At a concentration of 6 µg/mL caspase 3, 2 h of incubation at 37 °C was sufficient to almost completely cleave full-length TotHTT into N-WtHTT and N-MtHTT. Increasing the incubation time to 4 h slightly decreased N-WtHTT or N-MtHTT signal, possibly indicating breakdown of the N-terminal fragments over the incubation period at 37 °C. Therefore, 6 µg/mL caspase 3 with an incubation time of 2 h at 37 °C is recommended.

Subsequently, we examined whether the epitope saturation and linear range of the assay changed due to the caspase 3 treatment. To this end, 0.5 mg/mL HD patient fibroblast (GM21756) lysate was treated with 6 µg/mL caspase 3 for 2 h at 37 °C. The treated lysate was loaded in increasing concentrations (0.0125–0.4 mg/mL) with increasing antibody concentrations (1.5–152.4 µg/mL) on 66–440 kDa separation modules and ran on SW. The antibody dilution curves showed that the AUC of the N-WtHTT (Fig. 7f) and N-MtHTT (Fig. 7g) peaks markedly increased up to an antibody concentration of 15.2 µg/mL, the same as observed with the TotHTT assay. Conversely, the linear range of the caspase 3 HTT cleavage assay was wider than the linear range of the TotHTT assay. The assay was linear over the entire range of tested protein concentrations for both N-WtHTT and N-MtHTT (Fig. 7h). The S/N ratios for the N-WtHTT and N-MtHTT peaks were both > 10 even at the 0.0125 mg/mL protein concentration, increasing to > 100 when loading 0.4 mg/mL. Therefore, after caspase 3 treatment, loading up to 0.4 mg/mL protein lysate is recommended using 15.2 µg/mL of the N-terminal HTT ab109115 antibody to quantify N-terminal HTT fragments.

Finally, we examined whether the caspase 3 cleavage assay could separate N-terminal HTT fragments in a variety of patient fibroblast cell lines with different CAG repeat lengths in HTT ranging from 15 to 175 CAGs (Fig. 7i). It was not possible to size-separate the different WtHTT proteins in SCA3 patient fibroblast lysate (GM06153, 15/17 CAG in HTT) with the caspase 3 cleavage assay due to the small difference in CAG repeat length between the two HTT alleles. In contrast, N-WtHTT could be size-separated from N-MtHTT in all tested HD cell lines. This included HD patient fibroblasts with a mutant HTT repeat of 43 CAG, which is in the range of the most prevalent CAG repeat lengths in HD [63, 64]. Moreover, longer polyQ stretches in the MtHTT allele resulted in bands with a progressively higher MW. In GM02147 (15/43 CAG in HTT), N-MtHTT ran at 97 kDa; in GM21756 (15/68 CAG in HTT), N-MtHTT ran at 118 kDa; and in GM09197 (18/175 CAG in HTT), N-MtHTT ran at 250 kDa. Note that the N-terminal HTT fragments ran at a higher MW than predicted as listed in Fig. 7a, especially for N-MtHTT with longer polyQ stretches. It should again be noted that the peak observed at 44 kDa likely corresponds to the fluorescent standard protein that is included in every sample for the SW run. This peak can be observed both with and without caspase 3 treatment. Together, this showed that caspase 3 treatment of protein samples allowed for separate quantification of WtHTT and MtHTT across a wide range of different wild-type and mutant polyQ lengths.

Optimised SW Capillary Immunoassays

In total, functional antibodies were identified and SW capillary immunoassays developed for ATXN1, ATXN2, ATXN3, ATN1, AR, total HTT and N-terminal HTT fragments. The optimal antibody and protein loading concentrations for all the immunoassays were also determined. An overview of the optimised conditions, which SW module was used for optimisation, and which controls were included for which immunoassay is seen in Table 4.

Table 4.

Optimised SW capillary immunoassay conditions per PolyQ protein

Target protein	Antibody ID	Supplier	Immunogen	Separation module	siRNA	Recombinant protein	Mutant allele separation	Optimal (antibody) (µg/mL)	Linear range (mg/mL)
ATXN1	16H8L13	Invitrogen	aa 562–693	66–440 kDa	✓	✓	✓	1.0	0.1–0.6
ATXN2	ab254362	Abcam	Unknown	12–230 kDa	✓	x	Not tested	6.7	0.1–0.5
ATXN3	ab221143	Abcam	Unknown	12–230 kDa	✓	✓	✓	10.4	0.05–0.6
ATN1	NBP1-90044	Novus biologicals	aa 528–673	66–440 kDa	✓^a	x	✓	4.8–12	0.05–0.6
AR	mAb#5153	Cell Signaling Technology	N-terminal	66–440 kDa	✓^a	x	Not tested	3.3	0.1–0.5
HTT	ab109115	Abcam	N-terminal	66–440 kDa	✓	x	x	15.2	0.05–0.2
N-HTT	ab109115	Abcam	N-terminal	66–440 kDa	✓	x	✓	15.2	0.0125–0.4

Open in a new tab

^aLarge reductions with non-targeting siCtrl were observed, but to a lesser extent than with the targeted siRNA

Discussion

SW Capillary Immunoassays for Development of polyQ Protein-Lowering Therapies

In the preclinical development phase of new polyQ-lowering therapies, accurate and fast quantification of polyQ protein levels is essential. Often, large libraries of potential polyQ-lowering therapeutics are screened for efficacy to select a lead candidate. It is essential that the most promising compound is selected for further development, as erroneous selection based on unreliable data can lead to a considerable waste of resources and time. Conventional WB is commonly used to examine efficacy of polyQ protein-lowering therapeutics. However, WB is not well suited as a screening tool given its limited quantitative properties, proneness to a high level of technical variability and issues with reproducibility [47–51]. In addition, WB is not well suited for higher throughput screening applications as it requires a high amount of protein and is labour-intensive and time-consuming. Consequently, development of more effective and reliable tools to examine the efficacy of polyQ protein-lowering therapies is important. Therefore, we developed polyQ immunoassays using the next-generation protein analysis platform, the SW capillary immunoassay.

Advantages of SW Capillary Immunoassays

The SW capillary immunoassay has several clear advantages over WB. First, SW is less labour-intensive and notably faster than WB, leading to higher sample throughput. A single WB run, which includes SDS-page electrophoresis, blotting, staining and imaging steps, can take up to multiple days. In contrast, a SW run with up to 24 samples takes 2.5–3 h (or 5.5 h for a RePlex run), excluding plate loading, allowing for multiple runs in a day. Second, the SW capillary immunoassay has a wider dynamic range and is more sensitive than WB, requiring lower protein input [52–54]. This facilitates more efficient and higher throughput screening as smaller cell culture volumes can be used. We demonstrated this here by performing siRNA transfections in 96-well plates, which required a low number of cells, low volume of medium and low quantities of the polyQ-lowering agent. Third, detection and quantification of target proteins is more reliable and reproducible compared to WB [52, 53]. The lack of a blotting step with SW bypasses potential variability caused by inconsistent protein transfer from the SDS-page gel during blotting, which is especially relevant for larger proteins that are generally more difficult to transfer [65]. Additionally, blots are delicate and small handling mistakes during protein transfer, blocking and staining may lead to technical artifacts and issues during quantification. Furthermore, in WB, quantification of protein levels involves manually identifying bands and setting thresholds, often using the open-source imaging program, ImageJ. In SW capillary immunoassays, baselines and peaks are automatically detected and set by the Compass software. This makes quantification of protein levels more consistent between experiments and operators, resulting in more reliable quantification and better reproducibility.

Potential Drawbacks of SW Capillary Immunoassays

Besides advantages, there are some drawbacks the SW approach as well. Firstly, switching from WBs to SW capillary immunoassays can be challenging. Not all primary antibodies that can be used on WB work with the SW capillary immunoassay. Similarly, the lysis buffer used for protein extraction may not be compatible with the used SW system. Potentially, high concentrations of certain salts, denaturing agents or detergents in lysis buffers can lead to failed runs [66]. Furthermore, SW capillary immunoassays are generally more expensive to run than WB, requiring specific separation modules, reagents and machinery.

Secondly, size-separation of specific proteins may be more challenging on SW than on WB, as seen with the HTT assay in which WtHTT and MtHTT did not readily size-separate. With WB, size-separation can be optimised more conveniently as gel percentages and running times can be more flexibly adjusted. When using the Wes or Jess systems, proteins are size-separated using specific modules, which are limited to 2–40 kDa, 12–230 kDa and 66–440 kDa versions. Adjusting the running time of the SW using the Wes and Jess systems also has limitations. The maximum running time depends on the ladder and the fluorescent standards which are added to every sample to assess the MW distribution over the capillary. The standards need to still be in the capillary for accurate plotting of the electropherogram together with the ladder to obtain reliable data.

Lastly, although more consistent and more reproducible than WB, SW capillary immunoassays are still semiquantitative in nature. Similarly to WB, it is important to measure protein expression within the linear range of the assay; otherwise, protein levels can be over- or underestimated [49]. To further improve the quantitative nature of the SW capillary immunoassay, a standard curve of recombinant protein can be included. However, regardless of the drawbacks, SW capillary immunoassays are likely better suited than WB for preclinical development of polyQ therapies in most situations, given the previously discussed advantages.

SW Capillary Immunoassay Systems

For this manuscript, all SW capillary immunoassays were performed on the Wes and Jess systems. The Wes system allows for detection of single protein of interest per run, but unlike WB, it does not allow for stripping and probing to detect a second target. In contrast, the Jess system has a RePlex functionality comparable to blot stripping, allowing for detection of an additional protein or a total protein loading control measurement in the same capillary. Additionally, the Jess has a near infrared (NIR) and infrared (IR) channel, possibly allowing for multiplex detection of different proteins. Note that it may be possible to multiplex to detect multiple proteins using solely the chemiluminescence channel, which would also be possible with the Wes system. However, multiplexing in a single channel requires the proteins of interest to be sufficiently different in MW and similar in expression level. Additionally, high background signal, presence of non-specific peaks and cross-reactivity between the antibodies may interfere with multiplex assays. Although not tested here, it may be possible to detect multiple polyQ proteins in the same capillary by multiplexing the selected antibodies, but this will need to be confirmed experimentally. It should also be noted that the Wes and Jess systems can be used interchangeably as shown with the developed singleplex polyQ immunoassays, indicating that these assays may be well suited for use with other similar SW systems.

Observed Versus Predicted MW on SW Capillary Immunoassays

For assay development of WB or SW capillary immunoassays, the observed MW of a protein is often used as indicator that a band or peak represents the protein of interest. However, for the six developed SW capillary immunoassays, the observed MW of a polyQ protein peak was often not consistent with the predicted MW. There are several factors which can influence the MW of a protein on SW. The predicted MW is based purely on the amino acid sequence of a specific protein. This does not take post-translational modifications such as phosphorylation and ubiquitination into account, which increase the MW weight of a protein [67, 68]. Indeed, post-translational modifications have been described for the polyQ proteins [69–72]. Several other protein characteristics that influence migration through a gel, like hydrophobicity and whether a protein is cell membrane-bound or not, have been described [73, 74], which may also be relevant for SW. Other factors that may affect the observed MW of a protein on SW include the type of size-separation module used and the presence of high concentrations of certain salts, reducing agents, denaturing agents and detergents in the sample [66, 75]. For most of the polyQ proteins, these factors may explain the difference between observed and predicted MW. Only for WtATN1 the difference between observed and predicted MW (194 vs 125 kDa) seemed too large. However, WtATN1 has been shown to run at a higher MW of > 190 kDa on WB as well [55–57], strengthening the conclusion that the observed peak represented WtATN1. Overall, this demonstrated that the observed MW of a peak by itself is not reliable enough to conclude that a peak corresponds to a protein of interest. Whether a peak represents a certain polyQ protein could be verified by testing multiple antibodies against the same protein, as the observed MW of a protein should not change based on the used antibody. Indeed, we identified some other antibodies that likely detected ATXN1, ATXN3 and ATN1 based on recombinant protein and siRNA data as indicated in Online Resource 2. These antibodies detected a peak at the same MW as the selected 16H8L13 (ATXN1), ab221143 (ATXN3) and NPB-90044 (ATN1) antibodies but had lower specific or higher non-specific signal. For ATXN2 or AR, only a single antibody was identified that clearly detected the relevant ATXN2 or AR peak, and only a single antibody was tested for HTT. Therefore, it was essential to include proper positive or negative controls, like recombinant proteins or lysates depleted for the protein of interest, especially when the MW of a protein of interest on SW cannot be verified with multiple different antibodies.

Use of siRNA Samples to Validate Antibody Specificity with SW Capillary Immunoassays

To generate negative control samples where expression of a polyQ protein of interest was markedly lower, fibroblasts were transfected with targeted siRNAs. It should be noted that we used HD patient fibroblasts (GM21756) and not the control fibroblasts (GM02171) for all the wild-type polyQ protein siRNA experiments because these fibroblasts can be efficiently transfected and have similar expression of wild-type ATXN1, ATXN2, ATXN3, ATN1 and AR protein levels compared to other fibroblast cell lines (data not shown). Indeed, the six developed SW capillary immunoassays all detected the reduction in polyQ protein expression induced by the siRNA. However, it is important to establish whether the observed reduction in signal was caused by the on-target effect of the siRNA. Performing transfections using liposome-based agents such as RNAiMAX can lead to sequence non-specific effects like cytotoxicity, changes in cell viability and changes in RNA and protein expression [76, 77]. Although we did not see any obvious cell death or cytotoxicity after transfection, it is difficult to rule out non-specific effects of transfection based solely on morphology of the cells under the microscope. This is why it was important to examine sequence non-specific effects using the siCtrl. If a targeted siRNA and non-targeting siCtrl similarly decrease the AUC of a peak detected by an antibody, the peak may not correspond to the protein of interest as the decrease in AUC may not be caused by a sequence-specific effect of the targeted siRNA. For the ATXN1, ATXN3, ATN1 and HTT assays, transfection with the siCtrl did not markedly reduce signal intensity while the targeted siRNA did, confirming that the detected peaks represented the proteins of interest. For the ATXN2 and ATN1 assays, larger reductions in signal were observed after transfection with the siCtrl. However, because the targeted siRNA was clearly more efficacious, the results still suggested that the peaks of interest represented ATXN2 and ATN1, although not as strongly as for the other assays. This does show that sequence non-specific effects of transfection can indeed affect protein expression and therefore must be kept in mind when interpreting siRNA results. As an alternative to siRNAs, proteins of interest could be knocked out using CRISPR/Cas9, which may circumvent issues with the sequence non-specific effects of transfection. In finite fibroblast cell lines, single cell cloning to generate pure knockout cultures is time-consuming and difficult to achieve, but in immortalised cell lines, the use of CRISPR/Cas9 to generate negative control samples could be a viable alternative to siRNA transfections.

Detection and Size-Separation of Wild-Type and Mutant polyQ Proteins

Whether the wild-type and mutant polyQ protein isoforms could be separately detected using the developed assays was examined for two reasons. Firstly, allowing for separate quantification of wild-type and mutant polyQ protein levels is important in the development of allele-preferential polyQ-lowering therapies which aim to mainly reduce mutant polyQ protein expression. Secondly, the presence of a mutant peak in patient lysate provides further confirmation that an antibody binds to the correct protein. We showed detection of the mutant protein and size-separation from the wild-type isoform for the ATXN1, ATXN3 and ATN1 SW assays. Additionally, for HTT, we showed detection and size-separation of N-terminal N-WtHTT and N-MtHTT fragments with the novel caspase 3 assay. In contrast, for ATXN2 and AR, we were unable to show detection of the mutant protein isoforms due to unavailability of SCA2 and SBMA patient cell lines. Therefore, the evidence that the presumed ATXN2 and AR peaks correspond to ATXN2 and AR protein is not as strong as for the other assays. Nevertheless, as both the anti-ATXN2 antibody (ab254362) and the anti-AR antibody (mAb#5153) still likely detect the wild-type protein based on the siRNA data, it is unlikely that the antibodies would not detect the mutant isoform as well. Indeed, the mAb#5153 antibody has been shown to recognise AR with an expanded polyQ tract on WB [78]. Furthermore, the observed MWs of WtATXN2 (172 kDa) and WtAR (112 kDa) on SW are similar to WtATN1 (194 kDa) and WtATXN1 (115 kDa). Because WtATN1 and WtATXN1 did size-separate from the mutant isoform, it is likely that WtATXN2 and WtAR can be size-separated from mutant using SW as well. However, whether the developed ATXN2 and AR SW immunoassays can detect mutant ATXN2 (MtATXN2) and mutant AR (MtAR) and whether MtATXN2 and MtAR can be size-separated from the wild-type isoform on SW will need to be experimentally confirmed in SCA2 and SBMA patient lysate.

Antibody Considerations

When developing SW capillary immunoassays, it is important to note that there can be lot-to-lot inconsistencies in antibody performance. Lot-to-lot variability can result in changes in intensity of the peaks of interest, background signal and detection of non-specific peaks. This is most prominently observed with polyclonal antibodies, which we also encountered during antibody screening, where in some cases immunoassay development needed to be repeated with the new antibody lot. Therefore, if available, recombinant monoclonal antibodies were selected for assay development, which should have more consistent lot-to-lot performance than conventional monoclonal and polyclonal antibodies [79, 80]. For the optimised ATXN1, ATXN2, ATXN3, AR and HTT assays, recombinant monoclonal antibodies were used. However, for the ATN1 assay, no promising monoclonal antibody was identified, so the polyclonal antibody NBP1-90044 was used instead, which showed acceptable levels of lot-to-lot variability over three different lot numbers (data not shown).

Detection of polyQ Proteins and Isoforms in Different Cell and Tissue Types

The polyQ SW capillary immunoassays were developed using protein lysates derived from healthy control- or patient-derived fibroblast cell lines. Theoretically, the assays should be well suited to measure polyQ protein levels in other commonly used cell or tissue types like induced pluripotent stem cells (iPSC)-derived neurons, HeLa cells or human post-mortem brain material. However, SW assays are likely not sensitive enough to detect polyQ proteins in blood plasma or cerebrospinal fluid, as these often require ultra-sensitive single-molecule counting (SMC) methods as seen with HTT [81] and ATXN3 [82, 83]. When using these SW capillary immunoassays in other cell types or tissues, it is important to note that polyQ protein expression may be different than in fibroblasts. According to data from proteinatlas.org [84], the majority of the polyQ proteins have similar or higher expression levels in disease relevant tissues and cell types (e.g. different brain regions and neurons) than in fibroblasts. Even if a specific polyQ protein is expressed at lower levels in a relevant tissue or cell type, use of the developed assays is not necessarily precluded because of the high sensitivity of the SW capillary immunoassay. However, it should be noted that differences in protein expression may result in an altered protein loading linear range. Background signal may also differ depending on cell and tissue type. For instance, AR signal in fibroblasts was low compared to the other assays, possibly due to lower expression levels. In cell types that highly express AR, loading a lower amount of protein may be necessary. Likewise, the reason that no acceptable ATXN7, CACNA1A or TBP antibody was identified here may be due to low protein expression levels in fibroblasts. Furthermore, different cell types may express different isoforms of specific proteins. For several of the polyQ proteins, alternative splicing and several disease-specific protein isoforms have been described [28, 29, 56, 85–87]. In fibroblast lysate, no clear polyQ protein isoforms, besides the expected wild-type and mutant proteins, were observed on SW with the HTT, ATXN1, ATXN2 and AR assays. For ATXN3, a peak at 44 kDa was observed that may correspond to an alternative isoform, as ATXN3 is extensively alternatively spliced [88]. However, transfection with the siATXN3 did not decrease signal of this peak, making it less likely to correspond to an ATXN3 isoform, unless it is an isoform where the exon that the siRNA binds to is spliced out. Confirmation of expression of a specific splicing isoform at the RNA level, followed by use of an siRNA targeting the specific splicing isoform, may be necessary to establish whether a specific peak indeed represents an alternatively spliced isoform. For ATN1, some evidence of smaller proteolytic cleavage products [55–57] was observed, but not consistently depending on the antibody lot. Whether a specific polyQ protein isoform can be detected with these assays will depend on the expression level of the isoform and on whether the epitope of the selected antibody is still present in the isoform. For example, in HD, a toxic aggregation prone N-terminal HTT exon 1 protein fragment caused by miss-splicing of mutant HTT transcripts has been described [28, 29]. Because the HTT assay uses an anti-N-terminal HTT antibody, it should be possible to study this isoform in models where the expression is high enough, although we did not observe this isoform here in HD patient fibroblasts. However, it should be noted that protein aggregates are often insoluble, potentially leading to lower levels of dissolved monomeric peptide in the lysis buffer. Similarly, mutant polyQ protein aggregates are insoluble and resistant to denaturing by SDS, requiring rigorous sonication and boiling in a stringent lysis buffer of the insoluble fraction to investigate [89, 90]. Here, we only used the fraction of proteins that was soluble in the CelLytic M lysis buffer for SW assay development. However, when screening or optimising antibodies, we mostly used lysate from fibroblasts that only expressed the less aggregation-prone wild-type isoform of the relevant polyQ protein. Nevertheless, in patient fibroblasts, the intensity of the mutant polyQ protein peak, although clearly detectable on SW, was often lower than that of the wild-type isoform. Whether this was caused by differences in protein expression of the isoforms or due to insoluble mutant protein aggregates is unclear. Furthermore, because no SCA2 or SBMA patient cells were available, we were unable to examine whether aggregation would possibly affect detection of MtATXN2 and MtAR on SW. Still, it is important to note that the detection of very aggregation prone protein isoforms may depend on the lysis protocol. Additionally, detection of alternative protein isoforms will depend on the antibody. Unfortunately, the exact epitope or immunogen used to raise an antibody is often proprietary (Table 3), which makes predicting whether an antibody can recognise alternative protein isoforms challenging.

Additionally, most of the antibodies used here are reactive to multiple species, according to the manufacturers. This suggests that the developed SW capillary immunoassays could be used for analysis of in vivo studies in both knock-in as well as humanised transgenic disease models. For example, the R6/2 mouse is a commonly used HD mouse model which expresses a human N-terminal HTT fragment [91] which may be detected by the anti-N-terminal HTT antibody ab109115. Furthermore, there are a wide variety of other commonly used transgenic and knock-in HD mouse and rat models [92]. Because the ab109115 antibody binds to mouse and rat HTT according to the manufacturer, it may be possible to use the HTT SW assay for these in vivo HD models as well. Similarly, the antibodies used for the ATXN1, ATXN2, ATXN3 and ATN1 assays are predicted to recognise the mouse and rat proteins. Conversely, the AR antibody mAb#5153 only reacts to human AR, according to the manufacturer, but has been used to detect AR in a human transgenic 97Q SBMA mouse model on WB [78].

Conclusion

Here, we developed SW capillary immunoassays to accurately quantify ATXN1, ATXN2, ATXN3, ATN1, AR and HTT protein expression. Additionally, we developed a novel caspase 3-based HTT SW assay allowing for separate quantification of mutant and wild-type HTT. We demonstrated that including the proper controls is necessary when selecting antibodies, as the observed MW of a detected peak is not sufficient to conclude that a peak corresponds to the protein of interest. We showed that the selected antibodies could detect decreases in polyQ protein levels induced by siRNAs and we optimised antibody and protein loading concentrations. Because the SW capillary immunoassay is time-efficient, very sensitive and yields reproducible data, these assays will be a valuable resource for preclinical research supporting the development of new mutant polyQ protein-lowering therapies for the progressive and fatal polyQ diseases.

Supplementary Information

Below is the link to the electronic supplementary material.

ESM1^{(258.4KB, docx)}

(DOCX 258 KB)

ESM2^{(260KB, docx)}

(DOCX 260 KB)

ESM3^{(266.9KB, docx)}

(DOCX 266 KB)

ESM4^{(7.4MB, docx)}

(DOCX 7.42 MB)

ESM5^{(7.5MB, docx)}

(DOCX 7.51 MB)

Acknowledgements

We are grateful to Prof. Willeke van Roon-Mom from the Human Genetics department of the Leiden University Medical Center (LUMC) for critical review of the manuscript. We are grateful to Anneke Janson for her assistance with the HTT SW running time experiments.

Author Contribution

BR wrote and revised the manuscript and performed experiments. JT significantly contributed to the identification of suitable antibodies and siRNA experiments. CB and RW contributed to the optimisation of the caspase 3 HTT assay. ND conceptualised and supervised the study and the writing of the manuscript. All authors critically reviewed the manuscript and have read and approved the final version.

Funding

This work was funded and supported by VICO Therapeutics B.V.

Data Availability

The datasets generated for this manuscript can be obtained from the corresponding author on reasonable request.

Declarations

Ethics Approval

Not applicable.

Consent to Participate

Not applicable.

Consent for Publication

Not applicable.

Competing Interests

Financial interest: All authors are employed and paid by VICO Therapeutics B.V. and hold employee stock options.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1.Orr HT, Zoghbi HY (2007) Trinucleotide repeat disorders. Annu Rev Neurosci 30:575–621. 10.1146/annurev.neuro.29.051605.113042 [DOI] [PubMed] [Google Scholar]
2.Torii R, Hashizume A, Yamada S et al (2023) Clinical features of female carriers and prodromal male patients with spinal and bulbar muscular atrophy. Neurology 100:e84–e93. 10.1212/WNL.0000000000201342 [DOI] [PubMed] [Google Scholar]
3.Gusella JF, MacDonald ME (2009) Huntington’s disease: the case for genetic modifiers. Genome Med 1:80. 10.1186/gm80 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Goldfarb LG, Vasconcelos O, Platonov FA et al (1996) Unstable triplet repeat and phenotypic variability of spinocerebellar ataxia type 1. Ann Neurol 39:500–506. 10.1002/ana.410390412 [DOI] [PubMed] [Google Scholar]
5.Figueroa KP, Hilary C, Nieves S et al (2017) Genetic analysis of age at onset variation in spinocerebellar ataxia type 2. Neurol Genet 3:e155. 10.1212/NXG.0000000000000155 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Schöls L, Vieira-Saecker AMM, Schöls S et al (1995) Trinucleotide expansion within the MJD1 gene presents clinically as spinocerebellar ataxia and occurs most frequently in German SCA patients. Hum Mol Genet 4:1001–1005. 10.1093/hmg/4.6.1001 [DOI] [PubMed] [Google Scholar]
7.Matsumura R, Futamura N, Fujimoto Y et al (1997) Spinocerebellar ataxia type 6. Neurology 49:1238–1243. 10.1212/WNL.49.5.1238 [DOI] [PubMed] [Google Scholar]
8.Johansson J, Forsgren L, Sandgren O et al (1998) Expanded CAG repeats in Swedish spinocerebellar ataxia type 7 (SCA7) patients: effect of CAG repeat length on the clinical manifestation. Hum Mol Genet 7:171–176. 10.1093/hmg/7.2.171 [DOI] [PubMed] [Google Scholar]
9.Toyoshima Y, Takahashi H (2018) Spinocerebellar ataxia type 17 (SCA17). Polyglutamine Disorders 1049:219–231. 10.1007/978-3-319-71779-1_10 [DOI] [PubMed] [Google Scholar]
10.Komure O, Sano A, Nishino N et al (1995) DNA analysis in hereditary dentatorubral-pallidoluysian atrophy. Neurology 45:143–149. 10.1212/WNL.45.1.143 [DOI] [PubMed] [Google Scholar]
11.Atsuta N, Watanabe H, Ito M, et al (2006) s. Natural history of spinal and bulbar muscular atrophy (SBMA): a study of 223 Japanese patient Brain 129:1446–1455. 10.1093/brain/awl096 [DOI] [PubMed]
12.Lieberman AP, Shakkottai VG, Albin RL (2019) Polyglutamine repeats in neurodegenerative diseases. Annu Rev Pathol 14:1–27. 10.1146/annurev-pathmechdis-012418-012857 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Medina A, Mahjoub Y, Shaver L, Pringsheim T (2022) Prevalence and incidence of Huntington’s disease: an updated systematic review and meta-analysis. Mov Disord 37:2327–2335. 10.1002/mds.29228 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Estevam B, Matos CA, Nóbrega C (2023) PolyQ database—an integrated database on polyglutamine diseases. Database 2023:baad060. 10.1093/database/baad060 [DOI] [PMC free article] [PubMed]
15.Ruano L, Melo C, Silva MC, Coutinho P (2014) The global epidemiology of hereditary ataxia and spastic paraplegia: a systematic review of prevalence studies. Neuroepidemiology 42:174–183. 10.1159/000358801 [DOI] [PubMed] [Google Scholar]
16.Buijsen RAM, Hu M, Sáez-González M et al (2023) Spinocerebellar ataxia type 1 characteristics in patient-derived fibroblast and iPSC-derived neuronal cultures. Mov Disord 38:1428–1442. 10.1002/mds.29446 [DOI] [PubMed] [Google Scholar]
17.Reiner A, Dragatsis I, Dietrich P et al (2011) Genetics and neuropathology of Huntington’s disease. Pathophysiology, Pharmacology, and Biochemistry of Dyskinesia 98:325–372. 10.1016/B978-0-12-381328-2.00014-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Paulson HL, Perez MK, Trottier Y et al (1997) Intranuclear inclusions of expanded polyglutamine protein in spinocerebellar ataxia type 3. Neuron 19:333–344. 10.1016/S0896-6273(00)80943-5 [DOI] [PubMed] [Google Scholar]
19.Yang W, Dunlap JR, Andrews RB, Wetzel R (2002) Aggregated polyglutamine peptides delivered to nuclei are toxic to mammalian cells. Hum Mol Genet 11:2905–2917. 10.1093/hmg/11.23.2905 [DOI] [PubMed] [Google Scholar]
20.Martí E (2016) RNA toxicity induced by expanded CAG repeats in Huntington’s disease. Brain Pathol 26:779–786. 10.1111/bpa.12427 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Nalavade R, Griesche N, Ryan DP et al (2013) Mechanisms of RNA-induced toxicity in CAG repeat disorders. Cell Death Dis 4:e752–e752. 10.1038/cddis.2013.276 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Mykowska A, Sobczak K, Wojciechowska M et al (2011) CAG repeats mimic CUG repeats in the misregulation of alternative splicing. Nucleic Acids Res Suppl 39:8938–8951. 10.1093/nar/gkr608 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Skariah G, Albin RL (2021) Repeat RNA toxicity drives ribosomal RNA processing defects in SCA2. Mov Disord 36:2464–2467. 10.1002/mds.28795 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Bañez-Coronel M, Ayhan F, Tarabochia AD et al (2015) RAN translation in Huntington disease. Neuron 88:667–677. 10.1016/j.neuron.2015.10.038 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Jazurek-Ciesiolka M, Ciesiolka A, Komur AA et al (2020) RAN translation of the expanded CAG repeats in the SCA3 disease context. J Mol Biol 432:166699. 10.1016/j.jmb.2020.10.033 [DOI] [PubMed] [Google Scholar]
26.Kacher R, Lejeune FX, Noël S, et al (2021) Propensity for somatic expansion increases over the course of life in Huntington disease. eLife 10:e64674. 10.7554/eLife.64674 [DOI] [PMC free article] [PubMed]
27.Kacher R, Lejeune F-X, David I et al (2024) CAG repeat mosaicism is gene specific in spinocerebellar ataxias. Am J Hum Genet 111:913–926. 10.1016/j.ajhg.2024.03.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Sathasivam K, Neueder A, Gipson TA et al (2013) Aberrant splicing of HTT generates the pathogenic exon 1 protein in Huntington disease. Proc Natl Acad Sci U S A 110:2366–2370. 10.1073/pnas.1221891110 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Hoschek F, Natan J, Wagner M et al (2024) Huntingtin HTT1a is generated in a CAG repeat-length-dependent manner in human tissues. Mol Med 30:36. 10.1186/s10020-024-00801-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Egorova PA, Bezprozvanny IB (2019) Molecular mechanisms and therapeutics for spinocerebellar ataxia type 2. Neurotherapeutics 16:1050–1073. 10.1007/s13311-019-00777-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Crespo-Barreto J, Fryer JD, Shaw CA et al (2010) Partial loss of ataxin-1 function contributes to transcriptional dysregulation in spinocerebellar ataxia type 1 pathogenesis. PLoS Genet 6:e1001021. 10.1371/journal.pgen.1001021 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Cattaneo E, Rigamonti D, Goffredo D et al (2001) Loss of normal huntingtin function: new developments in Huntington’s disease research. Neurotoxins 24:182–188. 10.1016/s0166-2236(00)01721-5 [DOI] [PubMed] [Google Scholar]
33.Moldovean-Cioroianu NS (2024) Reviewing the structure-function paradigm in polyglutamine disorders: a synergistic perspective on theoretical and experimental approaches. Int J Mol Sci 25:6789. 10.3390/ijms25126789 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Schulte J, Littleton JT (2011) The biological function of the huntingtin protein and its relevance to Huntington’s disease pathology. Curr Trends Neurol 5:65–78 [PMC free article] [PubMed] [Google Scholar]
35.Afonso-Reis R, Afonso IT, Nóbrega C (2021) Current status of gene therapy research in polyglutamine spinocerebellar ataxias. Int J Mol Sci 22:4249. 10.3390/ijms22084249 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Bahat A, Itzhaki E, Weiss B et al (2024) Lowering mutant huntingtin by small molecules relieves Huntington’s disease symptoms and progression. EMBO Mol Med 16:523–546. 10.1038/s44321-023-00020-y [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Fiszer A, Krzyzosiak WJ (2014) Oligonucleotide-based strategies to combat polyglutamine diseases. Nucleic Acids Res Suppl 42:6787–6810. 10.1093/nar/gku385 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Silva AC, Lobo DD, Martins IM et al (2020) Antisense oligonucleotide therapeutics in neurodegenerative diseases: the case of polyglutamine disorders. Brain 143:407–429. 10.1093/brain/awz328 [DOI] [PubMed] [Google Scholar]
39.Raamsdonk JMV, Pearson J, Rogers DA et al (2005) Loss of wild-type huntingtin influences motor dysfunction and survival in the YAC128 mouse model of Huntington disease. Hum Mol Genet 14:1379–1392. 10.1093/hmg/ddi147 [DOI] [PubMed] [Google Scholar]
40.Lim J, Crespo-Barreto J, Jafar-Nejad P et al (2008) Opposing effects of polyglutamine expansion on native protein complexes contribute to SCA1. Nature 452:713–718. 10.1038/nature06731 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Damaj L, Lupien-Meilleur A, Lortie A et al (2015) CACNA1A haploinsufficiency causes cognitive impairment, autism and epileptic encephalopathy with mild cerebellar symptoms. Eur J Hum Genet 23:1505–1512. 10.1038/ejhg.2015.21 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Gao L, Bhattacharyya A, Beers B et al (2024) Pharmacokinetics and pharmacodynamics of PTC518, an oral huntingtin lowering splicing modifier: a first-in-human study. Br J Clin Pharmacol 90:3242–3251. 10.1111/bcp.16202 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.McColgan P, Thobhani A, Boak L et al (2023) Tominersen in adults with manifest Huntington’s disease. N Engl J Med Overseas Ed 389:2203–2205. 10.1056/NEJMc2300400 [DOI] [PubMed] [Google Scholar]
44.Wave Life Sciences USA, inc (2024) Wave life sciences announces positive results from phase 1b/2a SELECT-HD trial with first clinical demonstration of allele-selective mutant huntingtin lowering in Huntington’s disease. wavelifesciences. https://ir.wavelifesciences.com/news-releases/news-release-details/wave-life-sciences-announces-positive-results-phase-1b2a-select. Accessed on 09 January 2025
45.Datson NA, González-Barriga A, Kourkouta E et al (2017) The expanded CAG repeat in the huntingtin gene as target for therapeutic RNA modulation throughout the HD mouse brain. PLoS ONE 12:e0171127. 10.1371/journal.pone.0171127 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Kourkouta E, Weij R, González-Barriga A et al (2019) Suppression of mutant protein expression in SCA3 and SCA1 mice using a CAG repeat-targeting antisense oligonucleotide. Mol Ther Nucleic Acids 17:601–614. 10.1016/j.omtn.2019.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Butler TAJ, Paul JW, Chan E-C et al (2019) Misleading Westerns: common quantification mistakes in Western blot densitometry and proposed corrective measures. Biomed Res Int 2019:1–15. 10.1155/2019/5214821 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Janes KA (2015) An analysis of critical factors for quantitative immunoblotting. Sci Signal 8:rs2. 10.1126/scisignal.2005966 [DOI] [PMC free article] [PubMed]
49.Pillai-Kastoori L, Schutz-Geschwender AR, Harford JA (2020) A systematic approach to quantitative Western blot analysis. Anal Biochem 593:113608. 10.1016/j.ab.2020.113608 [DOI] [PubMed] [Google Scholar]
50.Gassmann M, Grenacher B, Rohde B, Vogel J (2009) Quantifying Western blots: pitfalls of densitometry. Electrophoresis 30:1845–1855. 10.1002/elps.200800720 [DOI] [PubMed] [Google Scholar]
51.Rees PA, Lowy RJ (2023) Optimizing reduction of Western blotting analytical variations: use of replicate test samples, multiple normalization methods, and sample loading positions. Anal Biochem 674:115198. 10.1016/j.ab.2023.115198 [DOI] [PubMed] [Google Scholar]
52.Nguyen U, Squaglia N, Boge A, Fung PA (2011) The Simple Western™: a gel-free, blot-free, hands-free Western blotting reinvention. Nat Methods 8:v–vi. 10.1038/nmeth.f.353 [Google Scholar]
53.Sormunen A, Koivulehto E, Alitalo K et al (2023) Comparison of automated and traditional Western blotting methods. Methods Protoc 6:43. 10.3390/mps6020043 [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Chen J-Q, Wakefield LM, Goldstein DJ (2015) Capillary nano-immunoassays: advancing quantitative proteomics analysis, biomarker assessment, and molecular diagnostics. J Transl Med 13:182. 10.1186/s12967-015-0537-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Ellerby LM, Andrusiak RL, Wellington CL et al (1999) Cleavage of atrophin-1 at caspase site aspartic acid 109 modulates cytotoxicity. J Biol Chem 274:8730–8736. 10.1074/jbc.274.13.8730 [DOI] [PubMed] [Google Scholar]
56.Suzuki Y, Nakayama K, Hashimoto N, Yazawa I (2010) Proteolytic processing regulates pathological accumulation in dentatorubral-pallidoluysian atrophy. FEBS J 277:4873–4887. 10.1111/j.1742-4658.2010.07893.x [DOI] [PubMed] [Google Scholar]
57.Nucifora FC Jr, Ellerby LM, Wellington CL et al (2003) Nuclear localization of a non-caspase truncation product of atrophin-1, with an expanded polyglutamine repeat, increases cellular toxicity. J Biol Chem 278:13047–13055. 10.1074/jbc.M211224200 [DOI] [PubMed] [Google Scholar]
58.Suzuki Y, Yazawa I (2011) Pathological accumulation of atrophin-1 in dentatorubralpallidoluysian atrophy. Int J Clin Exp Pathol 4:378–384 [PMC free article] [PubMed] [Google Scholar]
59.Evers MM, Schut MH, Pepers BA et al (2015) Making (anti-) sense out of huntingtin levels in Huntington disease. Mol Neurodegener 10:21. 10.1186/s13024-015-0018-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Matthes F, Massari S, Bochicchio A et al (2018) Reducing mutant huntingtin protein expression in living cells by a newly identified RNA CAG binder. ACS Chem Neurosci 9:1399–1408. 10.1021/acschemneuro.8b00027 [DOI] [PubMed] [Google Scholar]
61.Wellington CL, Singaraja R, Ellerby L et al (2000) Inhibiting caspase cleavage of huntingtin reduces toxicity and aggregate formation in neuronal and nonneuronal cells. J Biol Chem 275:19831–19838. 10.1074/jbc.M001475200 [DOI] [PubMed] [Google Scholar]
62.Wellington CL, Ellerby LM, Hackam AS et al (1998) Caspase cleavage of gene products associated with triplet expansion disorders generates truncated fragments containing the polyglutamine tract. J Biol Chem 273:9158–9167. 10.1074/jbc.273.15.9158 [DOI] [PubMed] [Google Scholar]
63.Hećimović S, Klepac N, Vlasić J et al (2002) Genetic background of Huntington disease in Croatia: molecular analysis of CAG, CCG, and Delta2642 (E2642del) polymorphisms. Hum Mutat 20:233–233. 10.1002/humu.9055 [DOI] [PubMed] [Google Scholar]
64.Langbehn DR, Hayden MR, Paulsen JS et al (2010) CAG-repeat length and the age of onset in Huntington disease (HD): a review and validation study of statistical approaches. Am J Med Genet B Neuropsychiatr Genet 153B:397–408. 10.1002/ajmg.b.30992 [DOI] [PMC free article] [PubMed] [Google Scholar]
65.Kurien BT, Scofield RH (2003) Protein blotting: a review. J Immunol Methods 274:1–15. 10.1016/s0022-1759(02)00523-9 [DOI] [PubMed] [Google Scholar]
66.ProteinSimple (2014) Simple Western size assay development guide. Bio-Techne resources. https://resources.bio-techne.com/bio-techne-assets/docs/literature/042-889_Rev1_Size_Assay_Development_Guide.pdf. Accessed 09 January 2025
67.Xu G, Jaffrey SR (2013) Proteomic identification of protein ubiquitination events. Biotechnol Genet Eng Rev 29:73–109. 10.1080/02648725.2013.801232 [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Bagwan N, Ali HHE, Lundby A (2021) Proteome-wide profiling and mapping of post translational modifications in human hearts. Sci Rep 11:2184. 10.1038/s41598-021-81986-y [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Alegre-Martí A, Jiménez-Panizo A, Martínez-Tébar A, et al (2023) A hotspot for posttranslational modifications on the androgen receptor dimer interface drives pathology and anti-androgen resistance. Sci Adv 9:eade2175. 10.1126/sciadv.ade2175 [DOI] [PMC free article] [PubMed]
70.Wan L, Xu K, Chen Z et al (2018) Roles of post-translational modifications in spinocerebellar ataxias. Front Cell Neurosci 12:290. 10.3389/fncel.2018.00290 [DOI] [PMC free article] [PubMed] [Google Scholar]
71.Costa RG, Conceição A, Matos CA, Nóbrega C (2024) The polyglutamine protein ATXN2: from its molecular functions to its involvement in disease. Cell Death Dis 15:415. 10.1038/s41419-024-06812-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Ehrnhoefer DE, Sutton L, Hayden MR (2011) Small changes, big impact: posttranslational modifications and function of huntingtin in Huntington disease. Neuroscientist 17:475–492. 10.1177/1073858410390378 [DOI] [PMC free article] [PubMed] [Google Scholar]
73.Rath A, Glibowicka M, Nadeau VG et al (2009) Detergent binding explains anomalous SDS-PAGE migration of membrane proteins. Proc Natl Acad Sci U S A 106:1760–1765. 10.1073/pnas.0813167106 [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Shirai A, Matsuyama A, Yashiroda Y et al (2008) Global analysis of gel mobility of proteins and its use in target identification. J Biol Chem 283:10745–10752. 10.1074/jbc.m709211200 [DOI] [PubMed] [Google Scholar]
75.Wiesner R, Scheller C, Krebs F et al (2021) A comparative study of CE-SDS, SDS-PAGE, and Simple Western: Influences of sample preparation on molecular weight determination of proteins. Electrophoresis 42:206–218. 10.1002/elps.202000199 [DOI] [PubMed] [Google Scholar]
76.Wang T, Larcher LM, Ma L, Veedu RN (2018) Systematic screening of commonly used commercial transfection reagents towards efficient transfection of single-stranded oligonucleotides. Molecules 23:2564. 10.3390/molecules23102564 [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Liang X-H, Hart CE, Crooke ST (2013) Transfection of siRNAs can alter miRNA levels and trigger non-specific protein degradation in mammalian cells. Biochim Biophys Acta Gene Regul Mech 1829:455–468. 10.1016/j.bbagrm.2013.01.011 [DOI] [PubMed] [Google Scholar]
78.Iida M, Sahashi K, Kondo N et al (2019) Src inhibition attenuates polyglutamine-mediated neuromuscular degeneration in spinal and bulbar muscular atrophy. Nat Commun 10:4262. 10.1038/s41467-019-12282-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Bradbury A, Plückthun A (2015) Reproducibility: standardize antibodies used in research. Nature 518:27–29. 10.1038/518027a [DOI] [PubMed] [Google Scholar]
80.Bradbury ARM, Trinklein ND, Thie H, et al (2018) When monoclonal antibodies are not monospecific: hybridomas frequently express additional functional variable regions. mAbs 10:539–546. 10.1080/19420862.2018.1445456 [DOI] [PMC free article] [PubMed]
81.Wild EJ, Boggio R, Langbehn D et al (2015) Quantification of mutant huntingtin protein in cerebrospinal fluid from Huntington’s disease patients. J Clin Invest 125:1979–1986. 10.1172/jci80743 [DOI] [PMC free article] [PubMed] [Google Scholar]
82.Hübener-Schmid J, Kuhlbrodt K, Peladan J et al (2021) Polyglutamine-expanded ataxin-3: a target engagement marker for spinocerebellar ataxia type 3 in peripheral blood. Mov Disord 36:2675–2681. 10.1002/mds.28749 [DOI] [PubMed] [Google Scholar]
83.Gonsior K, Kaucher GA, Pelz P et al (2021) PolyQ-expanded ataxin-3 protein levels in peripheral blood mononuclear cells correlate with clinical parameters in SCA3: a pilot study. J Neurol 268:1304–1315. 10.1007/s00415-020-10274-y [DOI] [PMC free article] [PubMed] [Google Scholar]
84.Pontén F, Jirström K, Uhlen M (2008) The Human Protein Atlas—a tool for pathology. J Pathol 216:387–393. 10.1002/path.2440 [DOI] [PubMed] [Google Scholar]
85.Lastres-Becker I, Nonis D, Nowock J, Auburger G (2019) New alternative splicing variants of the ATXN2 transcript. Neurol Res Pract 1:22. 10.1186/s42466-019-0025-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
86.Lim WF, Forouhan M, Roberts TC, et al (2021) Gene therapy with AR isoform 2 rescues spinal and bulbar muscular atrophy phenotype by modulating AR transcriptional activity. Sci Adv 7:eabi6896. 10.1126/sciadv.abi6896 [DOI] [PMC free article] [PubMed]
87.Weishäupl D, Schneider J, Pinheiro BP et al (2019) Physiological and pathophysiological characteristics of ataxin-3 isoforms. J Biol Chem 294:644–661. 10.1074/jbc.ra118.005801 [DOI] [PMC free article] [PubMed] [Google Scholar]
88.Raposo M, Hübener-Schmid J, Tagett R et al (2024) Blood and cerebellar abundance of ATXN3 splice variants in spinocerebellar ataxia type 3/Machado-Joseph disease. Neurobiol Dis 193:106456. 10.1016/j.nbd.2024.106456 [DOI] [PubMed] [Google Scholar]
89.Ochaba J, Morozko EL, O’Rourke JG, Thompson LM (2018) Fractionation for resolution of soluble and insoluble huntingtin species. J Vis Exp 132:57082. 10.3791/57082 [DOI] [PMC free article] [PubMed] [Google Scholar]
90.Cohen A, Ross L, Nachman I, Bar-Nun S (2012) Aggregation of PolyQ proteins is increased upon yeast aging and affected by Sir2 and Hsf1: novel quantitative biochemical and microscopic assays. PLoS ONE 7:e44785. 10.1371/journal.pone.0044785 [DOI] [PMC free article] [PubMed] [Google Scholar]
91.Mangiarini L, Sathasivam K, Seller M et al (1996) Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell 87:493–506. 10.1016/S0092-8674(00)81369-0 [DOI] [PubMed] [Google Scholar]
92.Ramaswamy S, McBride JL, Kordower JH (2007) Animal models of Huntington’s disease. ILAR J 48:356–373. 10.1093/ilar.48.4.356 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ESM1^{(258.4KB, docx)}

(DOCX 258 KB)

ESM2^{(260KB, docx)}

(DOCX 260 KB)

ESM3^{(266.9KB, docx)}

(DOCX 266 KB)

ESM4^{(7.4MB, docx)}

(DOCX 7.42 MB)

ESM5^{(7.5MB, docx)}

(DOCX 7.51 MB)

Data Availability Statement

The datasets generated for this manuscript can be obtained from the corresponding author on reasonable request.

[CR1] 1.Orr HT, Zoghbi HY (2007) Trinucleotide repeat disorders. Annu Rev Neurosci 30:575–621. 10.1146/annurev.neuro.29.051605.113042 [DOI] [PubMed] [Google Scholar]

[CR2] 2.Torii R, Hashizume A, Yamada S et al (2023) Clinical features of female carriers and prodromal male patients with spinal and bulbar muscular atrophy. Neurology 100:e84–e93. 10.1212/WNL.0000000000201342 [DOI] [PubMed] [Google Scholar]

[CR3] 3.Gusella JF, MacDonald ME (2009) Huntington’s disease: the case for genetic modifiers. Genome Med 1:80. 10.1186/gm80 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR4] 4.Goldfarb LG, Vasconcelos O, Platonov FA et al (1996) Unstable triplet repeat and phenotypic variability of spinocerebellar ataxia type 1. Ann Neurol 39:500–506. 10.1002/ana.410390412 [DOI] [PubMed] [Google Scholar]

[CR5] 5.Figueroa KP, Hilary C, Nieves S et al (2017) Genetic analysis of age at onset variation in spinocerebellar ataxia type 2. Neurol Genet 3:e155. 10.1212/NXG.0000000000000155 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR6] 6.Schöls L, Vieira-Saecker AMM, Schöls S et al (1995) Trinucleotide expansion within the MJD1 gene presents clinically as spinocerebellar ataxia and occurs most frequently in German SCA patients. Hum Mol Genet 4:1001–1005. 10.1093/hmg/4.6.1001 [DOI] [PubMed] [Google Scholar]

[CR7] 7.Matsumura R, Futamura N, Fujimoto Y et al (1997) Spinocerebellar ataxia type 6. Neurology 49:1238–1243. 10.1212/WNL.49.5.1238 [DOI] [PubMed] [Google Scholar]

[CR8] 8.Johansson J, Forsgren L, Sandgren O et al (1998) Expanded CAG repeats in Swedish spinocerebellar ataxia type 7 (SCA7) patients: effect of CAG repeat length on the clinical manifestation. Hum Mol Genet 7:171–176. 10.1093/hmg/7.2.171 [DOI] [PubMed] [Google Scholar]

[CR9] 9.Toyoshima Y, Takahashi H (2018) Spinocerebellar ataxia type 17 (SCA17). Polyglutamine Disorders 1049:219–231. 10.1007/978-3-319-71779-1_10 [DOI] [PubMed] [Google Scholar]

[CR10] 10.Komure O, Sano A, Nishino N et al (1995) DNA analysis in hereditary dentatorubral-pallidoluysian atrophy. Neurology 45:143–149. 10.1212/WNL.45.1.143 [DOI] [PubMed] [Google Scholar]

[CR11] 11.Atsuta N, Watanabe H, Ito M, et al (2006) s. Natural history of spinal and bulbar muscular atrophy (SBMA): a study of 223 Japanese patient Brain 129:1446–1455. 10.1093/brain/awl096 [DOI] [PubMed]

[CR12] 12.Lieberman AP, Shakkottai VG, Albin RL (2019) Polyglutamine repeats in neurodegenerative diseases. Annu Rev Pathol 14:1–27. 10.1146/annurev-pathmechdis-012418-012857 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR13] 13.Medina A, Mahjoub Y, Shaver L, Pringsheim T (2022) Prevalence and incidence of Huntington’s disease: an updated systematic review and meta-analysis. Mov Disord 37:2327–2335. 10.1002/mds.29228 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Estevam B, Matos CA, Nóbrega C (2023) PolyQ database—an integrated database on polyglutamine diseases. Database 2023:baad060. 10.1093/database/baad060 [DOI] [PMC free article] [PubMed]

[CR15] 15.Ruano L, Melo C, Silva MC, Coutinho P (2014) The global epidemiology of hereditary ataxia and spastic paraplegia: a systematic review of prevalence studies. Neuroepidemiology 42:174–183. 10.1159/000358801 [DOI] [PubMed] [Google Scholar]

[CR16] 16.Buijsen RAM, Hu M, Sáez-González M et al (2023) Spinocerebellar ataxia type 1 characteristics in patient-derived fibroblast and iPSC-derived neuronal cultures. Mov Disord 38:1428–1442. 10.1002/mds.29446 [DOI] [PubMed] [Google Scholar]

[CR17] 17.Reiner A, Dragatsis I, Dietrich P et al (2011) Genetics and neuropathology of Huntington’s disease. Pathophysiology, Pharmacology, and Biochemistry of Dyskinesia 98:325–372. 10.1016/B978-0-12-381328-2.00014-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Paulson HL, Perez MK, Trottier Y et al (1997) Intranuclear inclusions of expanded polyglutamine protein in spinocerebellar ataxia type 3. Neuron 19:333–344. 10.1016/S0896-6273(00)80943-5 [DOI] [PubMed] [Google Scholar]

[CR19] 19.Yang W, Dunlap JR, Andrews RB, Wetzel R (2002) Aggregated polyglutamine peptides delivered to nuclei are toxic to mammalian cells. Hum Mol Genet 11:2905–2917. 10.1093/hmg/11.23.2905 [DOI] [PubMed] [Google Scholar]

[CR20] 20.Martí E (2016) RNA toxicity induced by expanded CAG repeats in Huntington’s disease. Brain Pathol 26:779–786. 10.1111/bpa.12427 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Nalavade R, Griesche N, Ryan DP et al (2013) Mechanisms of RNA-induced toxicity in CAG repeat disorders. Cell Death Dis 4:e752–e752. 10.1038/cddis.2013.276 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.Mykowska A, Sobczak K, Wojciechowska M et al (2011) CAG repeats mimic CUG repeats in the misregulation of alternative splicing. Nucleic Acids Res Suppl 39:8938–8951. 10.1093/nar/gkr608 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Skariah G, Albin RL (2021) Repeat RNA toxicity drives ribosomal RNA processing defects in SCA2. Mov Disord 36:2464–2467. 10.1002/mds.28795 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Bañez-Coronel M, Ayhan F, Tarabochia AD et al (2015) RAN translation in Huntington disease. Neuron 88:667–677. 10.1016/j.neuron.2015.10.038 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Jazurek-Ciesiolka M, Ciesiolka A, Komur AA et al (2020) RAN translation of the expanded CAG repeats in the SCA3 disease context. J Mol Biol 432:166699. 10.1016/j.jmb.2020.10.033 [DOI] [PubMed] [Google Scholar]

[CR26] 26.Kacher R, Lejeune FX, Noël S, et al (2021) Propensity for somatic expansion increases over the course of life in Huntington disease. eLife 10:e64674. 10.7554/eLife.64674 [DOI] [PMC free article] [PubMed]

[CR27] 27.Kacher R, Lejeune F-X, David I et al (2024) CAG repeat mosaicism is gene specific in spinocerebellar ataxias. Am J Hum Genet 111:913–926. 10.1016/j.ajhg.2024.03.015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR28] 28.Sathasivam K, Neueder A, Gipson TA et al (2013) Aberrant splicing of HTT generates the pathogenic exon 1 protein in Huntington disease. Proc Natl Acad Sci U S A 110:2366–2370. 10.1073/pnas.1221891110 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Hoschek F, Natan J, Wagner M et al (2024) Huntingtin HTT1a is generated in a CAG repeat-length-dependent manner in human tissues. Mol Med 30:36. 10.1186/s10020-024-00801-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Egorova PA, Bezprozvanny IB (2019) Molecular mechanisms and therapeutics for spinocerebellar ataxia type 2. Neurotherapeutics 16:1050–1073. 10.1007/s13311-019-00777-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Crespo-Barreto J, Fryer JD, Shaw CA et al (2010) Partial loss of ataxin-1 function contributes to transcriptional dysregulation in spinocerebellar ataxia type 1 pathogenesis. PLoS Genet 6:e1001021. 10.1371/journal.pgen.1001021 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Cattaneo E, Rigamonti D, Goffredo D et al (2001) Loss of normal huntingtin function: new developments in Huntington’s disease research. Neurotoxins 24:182–188. 10.1016/s0166-2236(00)01721-5 [DOI] [PubMed] [Google Scholar]

[CR33] 33.Moldovean-Cioroianu NS (2024) Reviewing the structure-function paradigm in polyglutamine disorders: a synergistic perspective on theoretical and experimental approaches. Int J Mol Sci 25:6789. 10.3390/ijms25126789 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR34] 34.Schulte J, Littleton JT (2011) The biological function of the huntingtin protein and its relevance to Huntington’s disease pathology. Curr Trends Neurol 5:65–78 [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.Afonso-Reis R, Afonso IT, Nóbrega C (2021) Current status of gene therapy research in polyglutamine spinocerebellar ataxias. Int J Mol Sci 22:4249. 10.3390/ijms22084249 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR36] 36.Bahat A, Itzhaki E, Weiss B et al (2024) Lowering mutant huntingtin by small molecules relieves Huntington’s disease symptoms and progression. EMBO Mol Med 16:523–546. 10.1038/s44321-023-00020-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Fiszer A, Krzyzosiak WJ (2014) Oligonucleotide-based strategies to combat polyglutamine diseases. Nucleic Acids Res Suppl 42:6787–6810. 10.1093/nar/gku385 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Silva AC, Lobo DD, Martins IM et al (2020) Antisense oligonucleotide therapeutics in neurodegenerative diseases: the case of polyglutamine disorders. Brain 143:407–429. 10.1093/brain/awz328 [DOI] [PubMed] [Google Scholar]

[CR39] 39.Raamsdonk JMV, Pearson J, Rogers DA et al (2005) Loss of wild-type huntingtin influences motor dysfunction and survival in the YAC128 mouse model of Huntington disease. Hum Mol Genet 14:1379–1392. 10.1093/hmg/ddi147 [DOI] [PubMed] [Google Scholar]

[CR40] 40.Lim J, Crespo-Barreto J, Jafar-Nejad P et al (2008) Opposing effects of polyglutamine expansion on native protein complexes contribute to SCA1. Nature 452:713–718. 10.1038/nature06731 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR41] 41.Damaj L, Lupien-Meilleur A, Lortie A et al (2015) CACNA1A haploinsufficiency causes cognitive impairment, autism and epileptic encephalopathy with mild cerebellar symptoms. Eur J Hum Genet 23:1505–1512. 10.1038/ejhg.2015.21 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Gao L, Bhattacharyya A, Beers B et al (2024) Pharmacokinetics and pharmacodynamics of PTC518, an oral huntingtin lowering splicing modifier: a first-in-human study. Br J Clin Pharmacol 90:3242–3251. 10.1111/bcp.16202 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.McColgan P, Thobhani A, Boak L et al (2023) Tominersen in adults with manifest Huntington’s disease. N Engl J Med Overseas Ed 389:2203–2205. 10.1056/NEJMc2300400 [DOI] [PubMed] [Google Scholar]

[CR44] 44.Wave Life Sciences USA, inc (2024) Wave life sciences announces positive results from phase 1b/2a SELECT-HD trial with first clinical demonstration of allele-selective mutant huntingtin lowering in Huntington’s disease. wavelifesciences. https://ir.wavelifesciences.com/news-releases/news-release-details/wave-life-sciences-announces-positive-results-phase-1b2a-select. Accessed on 09 January 2025

[CR45] 45.Datson NA, González-Barriga A, Kourkouta E et al (2017) The expanded CAG repeat in the huntingtin gene as target for therapeutic RNA modulation throughout the HD mouse brain. PLoS ONE 12:e0171127. 10.1371/journal.pone.0171127 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Kourkouta E, Weij R, González-Barriga A et al (2019) Suppression of mutant protein expression in SCA3 and SCA1 mice using a CAG repeat-targeting antisense oligonucleotide. Mol Ther Nucleic Acids 17:601–614. 10.1016/j.omtn.2019.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR47] 47.Butler TAJ, Paul JW, Chan E-C et al (2019) Misleading Westerns: common quantification mistakes in Western blot densitometry and proposed corrective measures. Biomed Res Int 2019:1–15. 10.1155/2019/5214821 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.Janes KA (2015) An analysis of critical factors for quantitative immunoblotting. Sci Signal 8:rs2. 10.1126/scisignal.2005966 [DOI] [PMC free article] [PubMed]

[CR49] 49.Pillai-Kastoori L, Schutz-Geschwender AR, Harford JA (2020) A systematic approach to quantitative Western blot analysis. Anal Biochem 593:113608. 10.1016/j.ab.2020.113608 [DOI] [PubMed] [Google Scholar]

[CR50] 50.Gassmann M, Grenacher B, Rohde B, Vogel J (2009) Quantifying Western blots: pitfalls of densitometry. Electrophoresis 30:1845–1855. 10.1002/elps.200800720 [DOI] [PubMed] [Google Scholar]

[CR51] 51.Rees PA, Lowy RJ (2023) Optimizing reduction of Western blotting analytical variations: use of replicate test samples, multiple normalization methods, and sample loading positions. Anal Biochem 674:115198. 10.1016/j.ab.2023.115198 [DOI] [PubMed] [Google Scholar]

[CR52] 52.Nguyen U, Squaglia N, Boge A, Fung PA (2011) The Simple Western™: a gel-free, blot-free, hands-free Western blotting reinvention. Nat Methods 8:v–vi. 10.1038/nmeth.f.353 [Google Scholar]

[CR53] 53.Sormunen A, Koivulehto E, Alitalo K et al (2023) Comparison of automated and traditional Western blotting methods. Methods Protoc 6:43. 10.3390/mps6020043 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Chen J-Q, Wakefield LM, Goldstein DJ (2015) Capillary nano-immunoassays: advancing quantitative proteomics analysis, biomarker assessment, and molecular diagnostics. J Transl Med 13:182. 10.1186/s12967-015-0537-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR55] 55.Ellerby LM, Andrusiak RL, Wellington CL et al (1999) Cleavage of atrophin-1 at caspase site aspartic acid 109 modulates cytotoxicity. J Biol Chem 274:8730–8736. 10.1074/jbc.274.13.8730 [DOI] [PubMed] [Google Scholar]

[CR56] 56.Suzuki Y, Nakayama K, Hashimoto N, Yazawa I (2010) Proteolytic processing regulates pathological accumulation in dentatorubral-pallidoluysian atrophy. FEBS J 277:4873–4887. 10.1111/j.1742-4658.2010.07893.x [DOI] [PubMed] [Google Scholar]

[CR57] 57.Nucifora FC Jr, Ellerby LM, Wellington CL et al (2003) Nuclear localization of a non-caspase truncation product of atrophin-1, with an expanded polyglutamine repeat, increases cellular toxicity. J Biol Chem 278:13047–13055. 10.1074/jbc.M211224200 [DOI] [PubMed] [Google Scholar]

[CR58] 58.Suzuki Y, Yazawa I (2011) Pathological accumulation of atrophin-1 in dentatorubralpallidoluysian atrophy. Int J Clin Exp Pathol 4:378–384 [PMC free article] [PubMed] [Google Scholar]

[CR59] 59.Evers MM, Schut MH, Pepers BA et al (2015) Making (anti-) sense out of huntingtin levels in Huntington disease. Mol Neurodegener 10:21. 10.1186/s13024-015-0018-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR60] 60.Matthes F, Massari S, Bochicchio A et al (2018) Reducing mutant huntingtin protein expression in living cells by a newly identified RNA CAG binder. ACS Chem Neurosci 9:1399–1408. 10.1021/acschemneuro.8b00027 [DOI] [PubMed] [Google Scholar]

[CR61] 61.Wellington CL, Singaraja R, Ellerby L et al (2000) Inhibiting caspase cleavage of huntingtin reduces toxicity and aggregate formation in neuronal and nonneuronal cells. J Biol Chem 275:19831–19838. 10.1074/jbc.M001475200 [DOI] [PubMed] [Google Scholar]

[CR62] 62.Wellington CL, Ellerby LM, Hackam AS et al (1998) Caspase cleavage of gene products associated with triplet expansion disorders generates truncated fragments containing the polyglutamine tract. J Biol Chem 273:9158–9167. 10.1074/jbc.273.15.9158 [DOI] [PubMed] [Google Scholar]

[CR63] 63.Hećimović S, Klepac N, Vlasić J et al (2002) Genetic background of Huntington disease in Croatia: molecular analysis of CAG, CCG, and Delta2642 (E2642del) polymorphisms. Hum Mutat 20:233–233. 10.1002/humu.9055 [DOI] [PubMed] [Google Scholar]

[CR64] 64.Langbehn DR, Hayden MR, Paulsen JS et al (2010) CAG-repeat length and the age of onset in Huntington disease (HD): a review and validation study of statistical approaches. Am J Med Genet B Neuropsychiatr Genet 153B:397–408. 10.1002/ajmg.b.30992 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR65] 65.Kurien BT, Scofield RH (2003) Protein blotting: a review. J Immunol Methods 274:1–15. 10.1016/s0022-1759(02)00523-9 [DOI] [PubMed] [Google Scholar]

[CR66] 66.ProteinSimple (2014) Simple Western size assay development guide. Bio-Techne resources. https://resources.bio-techne.com/bio-techne-assets/docs/literature/042-889_Rev1_Size_Assay_Development_Guide.pdf. Accessed 09 January 2025

[CR67] 67.Xu G, Jaffrey SR (2013) Proteomic identification of protein ubiquitination events. Biotechnol Genet Eng Rev 29:73–109. 10.1080/02648725.2013.801232 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR68] 68.Bagwan N, Ali HHE, Lundby A (2021) Proteome-wide profiling and mapping of post translational modifications in human hearts. Sci Rep 11:2184. 10.1038/s41598-021-81986-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR69] 69.Alegre-Martí A, Jiménez-Panizo A, Martínez-Tébar A, et al (2023) A hotspot for posttranslational modifications on the androgen receptor dimer interface drives pathology and anti-androgen resistance. Sci Adv 9:eade2175. 10.1126/sciadv.ade2175 [DOI] [PMC free article] [PubMed]

[CR70] 70.Wan L, Xu K, Chen Z et al (2018) Roles of post-translational modifications in spinocerebellar ataxias. Front Cell Neurosci 12:290. 10.3389/fncel.2018.00290 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR71] 71.Costa RG, Conceição A, Matos CA, Nóbrega C (2024) The polyglutamine protein ATXN2: from its molecular functions to its involvement in disease. Cell Death Dis 15:415. 10.1038/s41419-024-06812-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR72] 72.Ehrnhoefer DE, Sutton L, Hayden MR (2011) Small changes, big impact: posttranslational modifications and function of huntingtin in Huntington disease. Neuroscientist 17:475–492. 10.1177/1073858410390378 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR73] 73.Rath A, Glibowicka M, Nadeau VG et al (2009) Detergent binding explains anomalous SDS-PAGE migration of membrane proteins. Proc Natl Acad Sci U S A 106:1760–1765. 10.1073/pnas.0813167106 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR74] 74.Shirai A, Matsuyama A, Yashiroda Y et al (2008) Global analysis of gel mobility of proteins and its use in target identification. J Biol Chem 283:10745–10752. 10.1074/jbc.m709211200 [DOI] [PubMed] [Google Scholar]

[CR75] 75.Wiesner R, Scheller C, Krebs F et al (2021) A comparative study of CE-SDS, SDS-PAGE, and Simple Western: Influences of sample preparation on molecular weight determination of proteins. Electrophoresis 42:206–218. 10.1002/elps.202000199 [DOI] [PubMed] [Google Scholar]

[CR76] 76.Wang T, Larcher LM, Ma L, Veedu RN (2018) Systematic screening of commonly used commercial transfection reagents towards efficient transfection of single-stranded oligonucleotides. Molecules 23:2564. 10.3390/molecules23102564 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR77] 77.Liang X-H, Hart CE, Crooke ST (2013) Transfection of siRNAs can alter miRNA levels and trigger non-specific protein degradation in mammalian cells. Biochim Biophys Acta Gene Regul Mech 1829:455–468. 10.1016/j.bbagrm.2013.01.011 [DOI] [PubMed] [Google Scholar]

[CR78] 78.Iida M, Sahashi K, Kondo N et al (2019) Src inhibition attenuates polyglutamine-mediated neuromuscular degeneration in spinal and bulbar muscular atrophy. Nat Commun 10:4262. 10.1038/s41467-019-12282-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR79] 79.Bradbury A, Plückthun A (2015) Reproducibility: standardize antibodies used in research. Nature 518:27–29. 10.1038/518027a [DOI] [PubMed] [Google Scholar]

[CR80] 80.Bradbury ARM, Trinklein ND, Thie H, et al (2018) When monoclonal antibodies are not monospecific: hybridomas frequently express additional functional variable regions. mAbs 10:539–546. 10.1080/19420862.2018.1445456 [DOI] [PMC free article] [PubMed]

[CR81] 81.Wild EJ, Boggio R, Langbehn D et al (2015) Quantification of mutant huntingtin protein in cerebrospinal fluid from Huntington’s disease patients. J Clin Invest 125:1979–1986. 10.1172/jci80743 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR82] 82.Hübener-Schmid J, Kuhlbrodt K, Peladan J et al (2021) Polyglutamine-expanded ataxin-3: a target engagement marker for spinocerebellar ataxia type 3 in peripheral blood. Mov Disord 36:2675–2681. 10.1002/mds.28749 [DOI] [PubMed] [Google Scholar]

[CR83] 83.Gonsior K, Kaucher GA, Pelz P et al (2021) PolyQ-expanded ataxin-3 protein levels in peripheral blood mononuclear cells correlate with clinical parameters in SCA3: a pilot study. J Neurol 268:1304–1315. 10.1007/s00415-020-10274-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR84] 84.Pontén F, Jirström K, Uhlen M (2008) The Human Protein Atlas—a tool for pathology. J Pathol 216:387–393. 10.1002/path.2440 [DOI] [PubMed] [Google Scholar]

[CR85] 85.Lastres-Becker I, Nonis D, Nowock J, Auburger G (2019) New alternative splicing variants of the ATXN2 transcript. Neurol Res Pract 1:22. 10.1186/s42466-019-0025-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR86] 86.Lim WF, Forouhan M, Roberts TC, et al (2021) Gene therapy with AR isoform 2 rescues spinal and bulbar muscular atrophy phenotype by modulating AR transcriptional activity. Sci Adv 7:eabi6896. 10.1126/sciadv.abi6896 [DOI] [PMC free article] [PubMed]

[CR87] 87.Weishäupl D, Schneider J, Pinheiro BP et al (2019) Physiological and pathophysiological characteristics of ataxin-3 isoforms. J Biol Chem 294:644–661. 10.1074/jbc.ra118.005801 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR88] 88.Raposo M, Hübener-Schmid J, Tagett R et al (2024) Blood and cerebellar abundance of ATXN3 splice variants in spinocerebellar ataxia type 3/Machado-Joseph disease. Neurobiol Dis 193:106456. 10.1016/j.nbd.2024.106456 [DOI] [PubMed] [Google Scholar]

[CR89] 89.Ochaba J, Morozko EL, O’Rourke JG, Thompson LM (2018) Fractionation for resolution of soluble and insoluble huntingtin species. J Vis Exp 132:57082. 10.3791/57082 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR90] 90.Cohen A, Ross L, Nachman I, Bar-Nun S (2012) Aggregation of PolyQ proteins is increased upon yeast aging and affected by Sir2 and Hsf1: novel quantitative biochemical and microscopic assays. PLoS ONE 7:e44785. 10.1371/journal.pone.0044785 [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR91] 91.Mangiarini L, Sathasivam K, Seller M et al (1996) Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell 87:493–506. 10.1016/S0092-8674(00)81369-0 [DOI] [PubMed] [Google Scholar]

[CR92] 92.Ramaswamy S, McBride JL, Kordower JH (2007) Animal models of Huntington’s disease. ILAR J 48:356–373. 10.1093/ilar.48.4.356 [DOI] [PubMed] [Google Scholar]

PERMALINK

Accurate Quantification of Mutant and Wild-Type polyQ Proteins Using Simple Western Capillary Immunoassays

Bas Röttgering

Janwillem Testerink

Rudie Weij

Chantal Beekman

Nicole Datson

Abstract

Supplementary Information

Introduction

Table 1.

Materials and Methods

Cell Lines and Cell Culture

Table 2.

Protein Isolation for SW Capillary Immunoassay Optimisation

siRNA Transfection

Table 3.

SW Capillary Immunoassay Procedure

Caspase 3 HTT Cleavage Assay

Results

ATXN1

Fig. 1.

ATXN2

Fig. 2.

ATXN3

Fig. 3.

ATN1

Fig. 4.

AR

Fig. 5.

HTT

Fig. 6.

HTT Caspase 3 Cleavage Assay

Fig. 7.

Optimised SW Capillary Immunoassays

Table 4.

Discussion

SW Capillary Immunoassays for Development of polyQ Protein-Lowering Therapies

Advantages of SW Capillary Immunoassays

Potential Drawbacks of SW Capillary Immunoassays

SW Capillary Immunoassay Systems

Observed Versus Predicted MW on SW Capillary Immunoassays

Use of siRNA Samples to Validate Antibody Specificity with SW Capillary Immunoassays

Detection and Size-Separation of Wild-Type and Mutant polyQ Proteins

Antibody Considerations

Detection of polyQ Proteins and Isoforms in Different Cell and Tissue Types

Conclusion

Supplementary Information

Acknowledgements

Author Contribution

Funding

Data Availability

Declarations

Ethics Approval

Consent to Participate

Consent for Publication

Competing Interests

Footnotes

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases