We gene-edited the iPSC line previously generated from the SPG10 patient carrying the novel KIF5A variant (c.50G>A, p.R17Q, NM_004984.4) (Santangelo et al., 2023).

The web resource https://www.idtdna.com/site/order/designtool/index/CRISPR_SEQUENCE was employed to design the single-guide RNA (sgRNA) sequenceTCAAGGTGCTCTGCCGATTCCGG, targeting the region in the first exon of KIF5A gene containing the c.50G>A single nucleotide variant. The same web resource by IDT (above) and an additional software by Wellcome Sanger Institute Genome Editing (https://wge.stemcell.sanger.ac.uk/find) were used to check and exclude high probability off-target sites in coding sequences in the first 10 hits (Supplementary Table S2).

The following Alt-R single-stranded oligodeoxynucleotide (ssODN) sequence was used to introduce the WT KIF5A sequence (c.50G): AGA​CCA​ACA​ACG​AAT​GTA​GCA​TCA​AGG​TGC​TCT​GTC​GAT​TCC​GGC​CCC​TGA​ACC​AGG​C TGA​GAT​TCT​GCG​GGG​AGA​CAA​GTT​C.

For the iPSC gene-editing, we adapted the protocol provided by IDT company. Briefly, 8 × 10⁵ iPSCs were transfected with a ribonucleoprotein (RNP) complex containing synthetic sgRNA, Alt-R HiFi Cas9 nuclease V3, Electroporation Enhancer and Alt-R ssODN (all from IDT) using the Amaxa nucleofector II system. Cells were then plated on Synthemax II-SC (Corning)-coated wells in StemFlex medium (Life Technologies) supplemented with the Homology Directed Repair (HDR) enhancer (IDT), incubated at 32 °C for 48 h and then at 37 °C. On day 11, the grown colonies were dissociated, clonally diluted, and transferred to multi-well plates for expansion and subsequent Sanger sequencing analysis.

Genomic DNA was extracted from iPSC pellets using Wizard® Genomic DNA Purification Kit (Promega). Amplicons encompassing the target KIF5A variant (c.50G>A) were obtained using specific primer pairs (Fwd: CAG​AGA​CTG​AGC​ACC​TGT​CCT​CC; Rev: GGG​GAA​GAG​GAT​GAA​GGA​TGA​GC) and sequenced with BigDye Terminator kit (Applied Biosystems) on an ABI 3500 Genetic Analyzer (Applied Biosystems).

The two iPSC lines (WT and p.R17Q) were cultured in E8 medium (Thermo Fisher Scientific), splitted twice a week and cultured in 5% CO₂, at 37 °C.

For motoneurons differentiation, iPSCs were grown in suspension for 21 days to generate embryoid bodies (EBs), that were then dissociated and cultured for additional 13 days on poly-D-lysine and laminin (Thermo Fisher Scientific) pre-coated multi-well plates to obtain iPSC-MNs, as previously described (Casiraghi et al., 2025). The efficiency of iPSC-MNs differentiation was assessed by quantitative PCR and immunofluorescence.

Total RNA was isolated using the TRIzol Reagent (Thermo Fisher Scientific) according to the manufacturer’s protocol, DNAse I-treated and then retro-transcribed using SuperScript II reverse transcriptase (Thermo Fisher Scientific) along with oligo-dT primers. Q-PCR was performed in technical duplicates for each sample using SYBR Green PCR Master Mix (Applied Biosystems) with 300 nM of specific primer pairs for 45 cycles on the QuantStudio 12K Flex system (Applied Biosystems). Specific primer pairs for iPSC stemness markers (SOX2, OCT3/4, and NANOG) and for iPSC-MNs characterization (TUBB3, MAP2 and ChAT) were designed and used as previously detailed (Santangelo et al., 2023; 2024). Mean values of the threshold cycles (Ct) for each gene were normalized over mean Ct values of the housekeeping RPL10A gene (ΔCt). Each ΔCt was normalized over the ΔCt of the WT control sample (ΔΔCt) and gene expression values were expressed as fold change (2^−ΔΔCT).

Cell pellets were resuspended in lysis buffer (20 mM Tris–HCl pH 7.5,150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, protease/phosphatase inhibitors cocktail) for 15 min at 4 °C, and protein content quantified by Bicinchoninic (BCA) method as previously described (Casiraghi et al., 2025). Protein lysates (30 µg) were resolved on precast 10% NuPAGE Bis-Tris polyacrilamide gels in 3-N-morpholino propanesulfonic acid (MOPS) buffer and then transferred onto nitrocellulose membranes (all from Thermo Fisher Scientific). Membranes were blocked with 5% Bovine Serum Albumin (BSA) in Tris-buffered saline (Santa Cruz Biotechnology) with Tween-20 (Sigma-Aldrich) (TBST) and incubated with the primary antibodies for KIF5A and α-Tubulin (Supplementary Table S1) in blocking solution at 4 °C overnight. The membrane was then washed 4 times with TBST for 5 min and incubated 1 h at room temperature (RT) with the appropriate HRP-conjugated secondary antibody (Supplementary Table S1). The Clarity™ Western ECL Substrate (Biorad) was used for signal detection and densitometric analyses were performed using the Quantity One software (Biorad).

IF was performed as previously described (Santangelo et al., 2024). Cells on coverslips were fixed using 4% paraformaldehyde (Santa Cruz Biotechnology) for 20 min at RT and permeabilized with ice-cold methanol and 0.3% Triton X-100 (Sigma-Aldrich) for 5 min each. Fixed cells were then incubated with 10% Normal Goat Serum (Gibco) in phosphate buffered saline (PBS) at RT for 20 min, followed by incubation with specific primary antibodies (Supplementary Table S1) at 37 °C for 1.5 h. Coverslips were then incubated with fluorescent labelled secondary antibodies (Supplementary Table S1) for 45 min at RT. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich).

To visualize mitochondria, live cells were incubated with 10 nM Mitotracker Red CMXRos (Thermo Fisher Scientific) for 15 min in Neurobasal medium (Thermo Fisher Scientific). Cells were washed twice with PBS and then fixed for IF.

Images were acquired on a Nikon Eclipse Ti confocal microscope at 40x or 60x as Z-stacks (0.5 µm step size) at a resolution of 1,024 × 1,024 pixels.

The ImageJ software (https://imagej.net/software/fiji/) was used for different image analyses:

- Mean fluorescence intensity (MFI): the multiple stacks of each channel were combined using the sum slices criterion. The “Rectangle” or “Free Hand” functions were used to select the regions of interest (ROI), whose fluorescence signal intensity was measured with the “Measure” function. KIF5A protein MFI was measured in the soma cytoplasm (2.5 × 2.5 μm), in neurite segments (10–20 μm long) proximal to the soma and in distal neurite segments (50–60 μm long) ∼50 μm far from the soma.

The confocal microscope Nikon Eclipse Ti with a 20x and a ×40 objective was used for live imaging of mitochondria and lysosomes, respectively. The cell chamber conditions were set at 5% CO₂, 37 °C, 95% Relative Humidity (RH). At day 21 of differentiation, EB-dissociated cells were plated on 35 mm glass-bottom dishes and cultured for 13 days to obtain iPSC-MNs as described above.

To image mitochondrial transport along neurites, iPSC-MNs were incubated with 10 nM Mitotracker Red CMXRos (Invitrogen) diluted in Neurobasal medium for 15 min at 37 °C. After replacement of the medium, cells were washed twice with PBS and then iPSC-MNs culture medium was added to image the mitochondria transport with 1 frame every 2 s for 3 min with a ×20 objective.

To image lysosomal axonal transport, iPSC-MNs were labelled with 50 nM LysoTracker Red DND-99 (Invitrogen) for 15 min at 37 °C. Cells were washed with PBS and incubated with artificial cerebrospinal fluid (aCSF) pH 7.4 solution (NaCl 121 mM, KCl 2.5 mM, CaCl2 2.2 mM, MgSO4 1 mM, NaHCO3 29 mM, NaH2PO4 0.45 mM, Na2HPO4 0.5 mM, Glucose 20 mM) for 15 min at 37 °C (Feole et al., 2024). Images were recorded every 2 s for 1 min with a ×40 objective.

Given the difficulty in discriminating between axons and dendrites in iPSC-MNs, we performed MN differentiation at low density culture condition (5,000 EB-dissociated cells) and we considered the longest neurite extending from the soma as the main axon in our transport analyses. Mitochondrial and lysosomal velocities were tracked along segments approximately 50 μm in length and located within 100 μm distance from the nucleus.

Representative kymographs were generated from the time-lapse images using ImageJ software to illustrate the directional movement. The ImageJ “MTrackJ” plugin was used to analyze the mean velocity (μm/second) and the distance (μm) travelled by mitochondria and lysosomes along the selected segments (Supplementary Figure S3).

Statistical analysis was conducted using Graphpad Prism 9 software, using non-parametric Mann–Whitney U test. Results were considered statistically significant when p-value was <0.05.

The novel N-terminal KIF5A VUS (p.R17Q), which we recently identified in a SPG10 pedigree and predicted to impair ATP/ADP binding in the motor domain, was corrected in the iPSC line we recently generated and characterized from one out of the three affected members (Santangelo et al., 2023). By CRISPR/Cas9 gene-editing, we obtained the isogenic KIF5A WT iPSC line, as confirmed by Sanger sequencing (Figure 1a).

Cytogenetic analysis by Q-banding showed a normal karyotype and no gross rearrangements in the original mutant KIF5A p.R17Q iPSCs, as already reported (Santangelo et al., 2023), and in its gene-edited derivative WT cell line (data not shown). Both the KIF5A iPSC lines tested positive for the stemness markers OCT3/4, SOX2, and NANOG by Q-PCR (Figure 1b), and TRA-1–60, NANOG, AFP (Alpha-Fetoprotein), and SSEA-4 by IF (Figure 1c) confirming their pluripotency features.

We differentiated the two KIF5A iPSC lines into iPSC-MNs following a 34 days protocol and confirmed they were positive for the neuronal cytoskeletal markers SMI-312 and βIII-tubulin and the motoneuronal marker HB9 (Figure 1d). The differentiation efficiency was evaluated by Q-PCR of the neuronal markers TUBB3 (βIII-tubulin) and MAP2, as well as of the motoneuronal marker ChAT. Our results showed no differences in gene expression levels of both neuronal and motoneuronal markers between the mutant p.R17Q and the WT KIF5A iPSC-MNs in three differentiation replicates (Supplementary Figure S1). Analysis of KIF5A protein levels by Western blot (WB) revealed no difference between the mutant KIF5A p.R17Q and the isogenic WT iPSC-MNs (Figure 1e).

Since KIF5A is responsible for the axonal anterograde transport acting in dimers, we investigated whether the p.R17Q variant, while not affecting KIF5A protein levels, might influence kinesin motor activity and indirectly its distribution along neurites. For this purpose, we performed IF staining and measured the KIF5A protein mean fluorescence intensity (MFI) in three different motoneuronal cell regions: the soma cytoplasm, the neurite segments proximal to the soma, and the distal neurite segments (∼50 μm distance from the soma) (Figure 2a). To evaluate KIF5A protein redistribution from the cell soma into neurites, we first calculated the ratio between the KIF5A MFI in proximal neurite segments and in the soma. No statistically significant differences in KIF5A protein distribution in proximal neurites were observed between the two KIF5A iPSC-MN lines (Figure 2b).

To assess the ability of the mutant KIF5A protein to move and distribute from proximal to distal neurites, we determined the ratio between KIF5A MFI in neurite segments distal and proximal to the soma. We observed a significant increased ratio in the mutant KIF5A iPSC-MNs compared to the WT ones (Figure 2c), likely indicating an impaired transport and accumulation of the mutant protein in more distal neurites.

To further assess possible axonal transport defects, we analyzed the presence of axonal swellings by performing IF staining for SMI-312, an axonal marker of highly phosphorylated medium and heavy neurofilaments (Figure 2d). We observed a significant increase in the number of neurofilament-positive axonal swellings in the mutant KIF5A p.R17Q iPSC-MNs compared to the WT iPSC-MNs (Figure 2e). By a parallel labelling with Mitotracker Red CMXRos dye, we representatively showed that mitochondria were also trapped at axonal swellings in the KIF5A p.R17Q iPSC-MNs (Supplementary Figure S2).

To examine the impact of the KIF5A p.R17Q VUS on mitochondrial transport, we performed a live cell time-lapse assay after mitochondria labelling with the Mitotracker dye. The representative kymographs for both the mutant and the WT KIF5A iPSC-MN lines are shown (Figure 3a). We analyzed the mean mitochondrial anterograde velocity and the distance traveled by motile mitochondria in selected neurites regions as detailed in Materials and Methods, focusing on anterogradely moving mitochondria. The results showed a significant decrease in both mean anterograde velocity (Figure 3b) and distance travelled by the mitochondria (Figure 3c) in KIF5A p.R17Q iPSC-MNs compared to the WT iPSC-MNs.

To investigate whether the observed changes in mitochondrial anterograde transport also affected mitochondrial morphology (Figure 3d), we calculated the Aspect Ratio (AR) parameter related to the mitochondria along the analyzed neuritic processes. AR did not significantly differ between the two KIF5A iPSC-MN lines (Figure 3e).

To evaluate the transport of other cargoes, such as lysosomes, KIF5A iPSC-MNs were incubated with the LysoTracker Red DND-99 dye. Lysosomal movement was assessed by time-lapse imaging and representative kymographs were generated for both the mutant and the WT KIF5A iPSC-MN lines (Figure 4a). We analyzed the mean anterograde velocity and the distance travelled by motile lysosomes in specific neurites regions, focusing on anterogradely moving lysosomes. Our results showed a trend towards a reduced velocity in mutant p.R17Q iPSC-MNs compared to the WT cells (Figure 4b). Consistent with the reduced velocity, the lysosome travel distance was significantly reduced in the mutant KIF5A p.R17Q iPSC-MNs compared to the WT cells (Figure 4c).