Subjects and human samples
The generation and use of human iPSCs was approved by the Institutional Review Board (Nr. 4120: Generierung von humanen neuronalen Modellen bei neurodegenerativen Erkrankungen). Formal informed consent was obtained from all subjects. Four patients from three different families and age matched controls without history of neurological disorders were recruited. CAG repeats of fibroblasts and iPSCs were measured by the center of Human Genetics at the University Hospital Regensburg (Ute Hehr, MD).
Fibroblast culture
Fibroblasts were resuspended in fibroblast growth medium (FGM, 75% DMEM, 15% FCS, 2mM L-Glutamine, 100µg/ml Penicillin/Streptomycin, 2ng/ml fibroblast growth factor 2) and plated on polystyrene cell culture flasks. Medium was changed twice a week. Fibroblasts were split by removing FGM, adding Trypsin supplemented with 0.05% ethylenediaminetetraacetic acid (EDTA) and incubating at 37°C until cells detach. FGM was added to the detached cells, the cell suspension was transferred to a centrifugation tube and processed for 5 minutes at 300g RT. Supernatant was removed, cells were resuspended in FGM and plated on a new polystyrene cell culture flask.
iPSC generation and culture
For iPSC generation, skin biopsies of study participants were obtained. iPSCs were generated from fibroblasts using the CytoTune iPS 2.0 Sendai Reprogramming Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. Therefore, cell lines were transduced with Sendai virus containing four reprogramming factors c-MYC, KLF4, OCT3/4, and SOX2.
After generation, iPSCs were cultured in human stem cell media StemMACS iPS-Brew XF (Miltenyi Biotec) supplemented with 100 U/mL penicillin/streptomycin on 4mg/ml Geltrex (GibcoTM) coated polystyrene cell culture plates. Medium was changed every other day. When cell cultures reached 70-80% confluency, cells were passaged. Afterwards, iPSCs were washed once with DMEM/F12 (GibcoTM) and incubated with Gentle Cell Dissociation Reagent (Stemcell technologies) for 5 minutes at room temperature (RT). Gentle Cell Dissociation Reagent was aspirated and StemMACS iPS-Brew XF supplemented with 100 U/mL penicillin/streptomycin was added. Corning® Cell Lifter was used to detach hiPSCs from the cell culture plate. iPSCs were transferred to a new Geltrex coated plate.
Cortical differentiation
iPSCs were differentiated into cortical neurons using a previously reported protocol 9. In brief, iPSCs were maintained as described above. iPSCs were dissociated into a single cell suspension upon 70-80% confluency. Cells were washed once with PBS w/o Mg2+/Ca2+ and were incubated with Accutase for 5 min at 37°C. Cells were washed with DMEM/F12, centrifuged for 3 min at 300 g at RT, and resuspended in StemMACS iPS-Brew XF supplemented with 10 μM ROCK inhibitor. Cells were seeded on Geltrex coated plates with the desired density of 300’000 cells per cm2 and incubated for 24h at 37°C, 5% CO2. After cells reached confluency the next day, the medium was changed to neural maintenance medium (NMM: DMEM/F-12, neurobasal/B-27/N2, 100µM GlutaMAX, 100μM non-essential amino acids, 50 μM 2-mercaptoethanol, 1x penicillin-streptomycin) supplemented with dual SMAD inhibitors (NIM: 10μM SB431542, 100nM LDN193189) to promote neural induction. On day 12, cells differentiated into a neuroepithelial sheet and were further passaged. Cell sheet was gently washed with DMEM/F-12 and incubated for 5min with Collagenase V (2mg/ml) at 37°C for 5 min. The cell sheet was gently washed twice with DMEM/F-12 and finally detached with a 5ml serological pipette in NIM and gently resuspended into smaller pieces. Cells were passaged in a 1:2 ratio on Geltrex coated plates. Medium was changed to NMM the next day. Upon appearance of neural rosettes, medium was changed for 2 days in NMM supplemented with 20ng/ml FGF2 to promote neural stem cell proliferation. On day 19, cells were further passaged and maintained in NMM with medium changes every second day. On day 30, cells were finallly single-cell passaged with Accutase with the desired density of 50’000 cells per cm2. Cells were maintained in NMM for neuronal differentiation with medium changes twice a week till day 35 (Figure 1) or day 50 (Figures 2 – 6).
Branaplam treatment
Branaplam was reconstituted in DMSO with a concentration of 5M. Branaplam was supplemented to the cell culture media (FGM, StemMACS iPS-Brew XF or NMM) with a final concentration of 0.46nM - 1000nM and 0.002% DMSO. Supplemented medium was changed every 24h for a total of 72h.
Immunofluorescent staining
Cells were fixed in 4% paraformaldehyde (PFA) for 20 mins at RT and subsequently washed 3x with PBS each. The cells were permeabilized using 0.1% Triton X-100 and in PBS for 20 mins at RT. Then, cells were blocked in 0.3% Triton-X100 and 3% donkey serum in PBS for 1h at RT. Afterwards, cells were incubated with primary antibodies (rat anti CTIP2: ab18465, Abcam, 1:500; mouse anti beta-III-Tubulin: G7121, Promega, 1:1000; rabbit anti PAX6, 901301, BioLegend, 1:200; mouse anti Nestin, MAB5326, Millipore, 1:500) at 4°C overnight. After washing, incubation with secondary antibodies and nuclei staining using 1µg/ml DAPI was performed. The slides were mounted using ProLong(r) Antifade (Invitrogen) solution. Imaging was performed with a Zeiss Laser scanning 780 inverted confocal microscope.
FACS analysis
For flow cytometry, cells were dissociated using Accutase for 30 mins at 37°C and resuspended in FC buffer (2% FCS, 0.01% sodium azide in PBS). Cells were dispensed into 5 ml tubes (Sarstedt) at 500,000 cells per well. For intracellular antigens, cells were fixed and permeabilized using 100ul BD Fixation/Permeabilization Solution (BD Bioscience) for 10 mins, then 1ml of BD Perm/Wash Buffer was added, cells were incubated for 5 mins and subsequently centrifuged at 1,500 rpm for 3 mins. For intracellular staining of cortical progenitors anti-PAX6-APC (130-123-267, Miltenyi Biotech, 1:100) and anti-NESTIN-PerCp-Cy5.5 (561231, BD Bioscience, 1:100) for an additional 30 mins. After a wash step, cells were resuspended in 350µl FACS buffer containing DAPI (1μg/ml). For intracellular staining of neurons, cells were stained using anti-bIII-Tubulin-AF405 (NB600-1018AF405, NovusBio, 1:100) or anti-CTIP2-FITC (ab123449, Abcam, 1:100) for 30 mins. Additional controls included applying an antibody solution without one antibody in the full cocktail (“minus 1 control”) and were used to determine potential bleed-through of the fluorophores. The flow cytometry experiments were performed with a Cytoflex S machine (laser 405nm, 488nm, 561nm and 638nm; Beckman Coulter) and analyzed with the CytExpert 2.4 software.
To determine cell death via FACS, we used a commercially available kit that uses a fluorescent 660-DEVD-FMK caspase-3/7 inhibitor reagent (ab270785, abcam) and a fixable cell permeability dye (Live-or-Dye, 32008-T, Biotium). The caspase assay and Live-or-Dye assay reagents were dissolved in 50ul DMSO, respectively and aliquoted and stored at -20°C. For the assay, cortical neurons were grown in 24-well plates. At the day of analysis, media was aspirated from the plate and 150µl DMEM/F12+Glutamax containing 0.48µl 660-DEVD-FMK caspase-3/7 inhibitor reagent and 0.15μl Live-or-Dye assay were applied. After incubation for 45 mins at 37°C. Cells were dissociated, fixed, and stained as stated above. To precisely assess bleed through, single incubation controls (either with 660-DEVD-FMK caspase-3/7 inhibitor reagent or Live-or-Dye assay) were used. The number of Casp3/7+Live-or-Dye- cells vs. Casp3/7-Live-or-Dye- were determined in CTIP2+ and betaIII-Tubulin+ cells.
Protein extraction
Lysis buffer (150 mM NaCl, 20 mM Tris-HCl (pH 7.5), 1 mM EDTA, 1 mM EGTA, 1 % Triton-X-100, 10 mM NaF, 1 mM PMSF, 1x Phosphatase Inhibitor, and 1x Protease Inhibitor in dH2O) was added to the cells for 30 mins at 4°C. Cell lysate was transferred into a tube and centrifuged for 15 mins at 2000g at 4°C. Supernatant was transferred to a new centrifugation tube and stored at -80°C. Protein concentration was determined using BCA assay kit (Thermo Fisher Scientific) measuring with CLARIOStar Plus (BMG Labtech).
HTT assay
Meso-Scale-Discovery (MSD) assays to measure total and mutant Huntingtin protein levels were performed by Evotec SE, Hamburg 10. The MSD assay plate was coated with 5ug/ml of the N-terminally binding HTT antibody 2B7 (#CH03023, Coriell) in coating buffer (15mM Na2CO3, 35mM NaHCO3) overnight. The next day, the plate was washed 3x in wash buffer (0.2% (v/v) Tween 20 in DBPS), blocked for 1h at RT shaking at 350rpm (2% (w/v) Probumin in wash buffer) and subsequently washed 3x again. The MSD plate was then incubated with the protein sample derived from the various cells (10ul sample/well) for 1h at RT shaking at 350rpm. In parallel, a standard of defined concentrations of recombinant human HTT with 23Q or 73Q was applied. After incubation, the plate was washed 3x in wash buffer. Next, 10ul of the detection antibodies were added to the MSD plates: 0.5ug/ml D7F7 antibody (#5656, Cell Signaling) for tHTT detection, or 5ug/ml MW1 antibody (#MABN2427, Sigma-Aldrich) binding the polyQ region in exon 1 for mutant HTT detection. MW1 was used directly labeled with a SULFO-Tag and incubated for 1h at RT shaking at 350rpm and subsequently washed 3x. For D7F7, after 3 washes, a SULFO-Tag-labelled anti-rabbit secondary antibody (MSD) was incubated for 1h at RT and the plate subsequently washed 3x. MSD read buffer was added to the plate. If the detection antibody binds to the sample in close proximity to the MSD plate an electrochemiluminescent signal is emitted and detected at 620nm. The total and mutant HTT levels were calculated according to the generated standard curves and normalized to protein input.
Toxi light assay
Cytotoxicity was measured during Branaplam treatment using the ToxiLight Bio assay kit (Lonza) according to the manufacturer's instructions. Therefore, supernatant was collected after 72h of Branaplam treatment. The positive control was supernatant of untreated cells incubated with 10% Triton-X100 for 20 minutes at 37°C. Triplicates 20µl/sample were transferred to a 96-Well. 100µl of adenylate kinase detection reagent (ToxiLight Bio assay kit, Lonza) was added and incubated for 5 mins at RT. The resulting luminescence was measured by the CLARIOStar Plus (BMG Labtech).
RNA extraction and HTT novel exon PCR
RNA was extracted using the RNeasy kit (Qiagen) according to the manufacturer's instructions. RNA concentrations were measured using a NanoDrop. The GoScript Reverse Transcriptase cDNA Synthesis kit (Promega) was used to generate cDNA from fibroblasts and cortical neurons using random primers. RNA was mixed with random primers and incubated for 5 mins at 70°C and placed on ice for 5 mins. The remaining reaction mix was added and incubated for 5 mins at 25°C, followed by 1h 42°C extension period and a 15 mins 70°C inactivation. The GoTaq 2x Mastermix (Promega) was used to amplify novel exon inclusion in HTT amplifying 0.5ul of template with 1ul of fwd primer (100uM stock, GTCATTTGCACCTTCCTCCT) and 1ul rev primer (100uM stock, TGGATCAAATGCCAGGACAG), 5ul Mastermix and 2.5ul DNase/RNase-free water. Primer sequences were obtained from the Novartis patent (WO2021084495A1). The mix was amplified with the following conditions: 95°C for 3min, and 34 cycles of 95°C for 30s, 60°C for 20s and 72°C for 60s. A final extension of 72°C for 5min was added at the end. The products were run on a 2% Agarose gel with RotiGel stain. (Carl Roth GmbH) at 125V. A random selection of 88bp and ~200bp bands in Ctrl and HD was cut out and purified to verify their correct identity by Sanger sequencing.
RNA sequencing
A total of 500ng per sample were sent for RNA sequencing to Azenta Life Sciences (Genewiz Leipzig, Germany) for 150bp paired-end sequencing with Poly-A selection. For fibroblasts 4 Ctrl samples and 4 HD samples with DMSO or Branaplam treatment were sent and sequenced at a depth of >20 million reads in each sample. For iPSC-derived cortical neurons, 3 Ctrl samples and 3 HD samples with DMSO or Branaplam treatment were sent and sequenced at a depth of >37 million reads in each sample. After obtaining the fastq files, adapters were trimmed using Trimmomatic 16 and aligned to the human genome (GRCh38) using STAR 17. In every sample >90% of reads mapped uniquely to the human genome. Reads were assigned to genes in the gencode annotation (version 26) using the featureCounts module within the Subread package 18. Reads Per Kilobase of transcript, per Million mapped reads (RPKM) were calculated from the obtained counts to normalize for gene expression.
Alternative splicing analysis
For differential splicing rMATS (version 4.2.0) 19 was used with the --novelSS flag to identify non-annotated exons. The gencode annotation (version 26) was used to define known exons. The output files considering only the junction counts were used for further analysis. A negative value of the InclusionLevelDifference reflects an inclusion of a given exon in the samples of the target condition and a positive value of the InclusionLevelDifference reflects an exclusion of a given exon in the samples of the target condition. Subsequently, the files from the different splice types (cassette exon, A5SS, A3SS, RI and MXE) were combined into one file.
All downstream analysis were performed in Python 3. Only exon junctions that were covered with at least 10 counts in each sample of a given dataset were considered. A unique index was generated, referring to a specific AS event with the aim to identify the identical exon junction in separate rMATS analysis. An exon was called as differentially alternatively spliced in each dataset if the FDR was below 0.05 and the absolute value of the InclusionLevelDifference was more than 0.1. The overlap of differentially spliced events was visualized with the Venn function in matplotlib library.
For k-means clustering, the k-means method from the sklearn.cluster module from sciki-lean was used (specifications: init=’ranodm’, n_clusters=10, n_init=10, max_iter=300, random_state=42). AS events significant in any of the four comparisons (fibroblasts-Ctrl DMSO vs Branaplam, fibroblasts-HD DMSO vs Branaplam, cortical neurons-Ctrl DMSO vs Branaplam, cortical neurons-HD DMSO vs Branaplam) datasets were clustered into 10 clusters according to the inclusion value differences in the respective dataset. Exon junctions that were not detected with a sufficient number of reads were masked and visualized in black. The inclusion level difference of each cluster was additionally visualized with violin plots.
To determine the effect of aberrant AS reversal upon Branaplam treatment, the individual inclusion values were used from HD-DMSO, HD-Branaplam, and Ctrl-DMSO samples in fibroblasts and cortical neurons, respectively. Only significant HD alternative splicing events (Ctrl DMSO vs. HD DMSO) in a respective cell type that were detected in all samples analyzed in a cell type (>10 reads in every single sample) were used. The reversal of aberrant AS was investigated in a quantitative and qualitative manner. For quantitative measurement, the absolute inclusion value difference was calculated by subtracting HD-DMSO or HD-Branaplam inclusion values from Ctrl-DMSO inclusion values and taking the absolute value. The statistical significance of the absolute inclusion level difference was determined using scipy.stats.ranksums. For qualitative measurement, the mean inclusion values of Ctrl-DMSO and HD-DMSO and HD-Branaplam samples in each cell type were also visualized in a scatter plot. A reversal of aberrant AS was determined if the inclusion level differences in HD-Branaplam samples dropped below an absolute value of 0.1.
RBP enrichment
In order to determine RNA-binding proteins that are enriched in alternatively spliced events in HD, we made use of the ENCODE database and their eCLIP-seq datasets. We downloaded eCLIP seq peak files aligned to GRCh38 with the Irreproducible Discovery Rate (IDR) peaks (released by November 2021). A peak was considered significant if negative log10(P value) ≥ 3 and the log2(fold change) ≥ 3. To determine if an eCLIP-seq peak was present in an exon junction in HD, the rMATS output (fibroblasts Ctrl-DMSO vs HD-DMSO or cortical neurons Ctrl-DMSO vs HD-DMSO) of interest was converted into a bed format encompassing the region starting from the upstream exon start to the downstream exon end. The rMATS bed was intersected with the significant eCLIP-seq peak file using pybedtools (-u True). The statistical significance of the enrichment was computed using hypergeometric test with all events that passed the coverage threshold as the background.
Kmer enrichment
To determine the sequence preferences of Branaplam-induced AS sites at the 3' and 5' splice site, we calculated kmer enrichments (4mer, 6mer, and 8mer) 5b upstream and downstream of the 3' and 5' splice site, respectively. As a background, we also calculated the 5b upstream and downstream of the 3’ splice site of the respective upstream exon and the 5' splice site of the respective downstream exon. Kmers were counted with Kvector (https://github.com/olgabot/kvector) and significance was determined with Fisher's exact test using scipy.stats.
Statistical analysis
GraphPad Prism 9 was used to visualize data and calculate statistics for pair-wise and grouped analyses (HTT protein measurements, toxilight assay, FACS quantification, densitometric quantification of HTT PCR, HTT RPKM values). DMSO samples and their respective Branaplam samples were considered as paired. Normal distribution was assessed with Shapiro-Wilk test. When comparing two conditions, Welch's test was used if normal distribution was confirmed and Mann-Whitney test was used for non-normally distributed data. When comparing multiple groups (e.g. different Branaplam concentrations), one-way ANOVA with Geisser-Greenhouse correction was used if normal distribution was confirmed and Friedman test was used for non-normally distributed data with Dunnett's or Dunn's post hoc test respectively to identify differences between individual groups. For grouped analyses (e.g. DMSO vs. Branaplam in Ctrl vs. HD), two-way ANOVA was used. The statistical test used for calculating significance of each graph is indicated in the figure legend. A p value ≤ 0.05 was considered as significant.