An alternative cytoplasmic SFPQ isoform with reduced phase separation potential is upregulated in ALS
Data files
Aug 13, 2025 version files 26.17 MB
-
Doc_1_Neeves_et_al_2025_raw_data_final.xlsx
176.28 KB
-
Doc_2_BaseScope_SFPQ_HEK293T_SMG1i.xlsx
17.17 KB
-
Doc_3_BaseScope_SFPQ_D6MN_OX.xlsx
16.14 KB
-
Doc_4_Neeves_et_al_SA_Western_blots_full.pdf
7.84 MB
-
Doc_5_Neeves_et_al_sedimentation_gels.pdf
18 MB
-
Doc_6_SFPQ_cotransfection_imaging.xlsx
111.66 KB
-
README.md
15.99 KB
Abstract
Splicing factor proline- and glutamine-rich (SFPQ) is an RNA-binding protein that broadly regulates RNA metabolism. Although its nuclear roles are well-studied, evidence of SFPQ’s cytoplasmic functionality is emerging. Altered expression and nuclear-to-cytoplasmic redistribution of SFPQ have been recognised in amyotrophic lateral sclerosis (ALS) pathology, yet the mechanistic basis of this remains undetermined. We identified altered SFPQ splicing in ALS, increasing the expression of an alternative mRNA isoform lacking a nuclear localisation sequence, which we termed ‘altSFPQ’. We find that altSFPQ mRNA contributes to SFPQ autoregulation and is highly unstable yet exhibits context-specific translation with cytoplasm-predominant localisation. Notably, reduced canonical SFPQ coincides with increased altSFPQ transcript expression in familial and sporadic ALS models, providing a mechanistic basis for SFPQ nuclear-to-cytoplasmic redistribution in ALS patients. Finally, we observe that the altSFPQ protein has reduced phase separation potential and differential protein binding compared to its canonical counterpart, providing insight into its mechanistic relevance to physiology and ALS pathogenesis.
Dataset DOI: 10.5061/dryad.gb5mkkx1z
Description of the data and file structure
The attached documents contain raw data relating to the experiments undertaken within the associated manuscript (Neeves et al., 2025).
File: Doc_1_Neeves_et_al_2025_raw_data_final.xlsx
Description: tabulated Excel spreadsheet. Experimental types - qPCR, image analyses
- qPCR - raw Ct values
- High throughput immunofluorescence imaging (Opera phenix) - object data (number, area, immunofluorescence intensities)
- Image analysis of LLPS droplet assays - object data (number, area, immunofluorescence intensities)
The attached documents contain raw data relating to the experiments undertaken within the associated manuscript (Neeves et al., 2025).
Document 1: tabulated Excel spreadsheet. Experimental types - qPCR, image analyses
- qPCR - raw Ct values
- High throughput immunofluorescence imaging (Opera phenix) - object data (number, area, immunofluorescence intensities)
- Image analysis of LLPS droplet assays - object data (number, area, immunofluorescence intensities)
Document 1 individual tab notes:
Tab 1: 1B_DRB
Description: assessing levels of target RNA(s) in HEK293T cells treated with 5,6-Dichloro-1-β-D-ribofuranosylbenzimidazole (DRB) transcription inhibitor; time (hours) indicates time after treatment at which cells were harvested
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH acts as a housekeeping gene to which other target RNAs are normalised.
Comments: GADD45A was assessed on a separate plate, hence housekeeping gene GAPDH was rerun alongside it and used to normalise GADD45A values
Tab 2: 1C_S2E_SMG1i
Description: assessing levels of target RNA(s) in HEK293T cells treated with SMG1 small molecule inhibitor (or untreated, or DMSO vehicle treated)
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH and POLR2B act as housekeeping genes to which other target RNAs are normalised (using their geometric mean).
Tab 3: 1F_S2G_SFPQi
Description: assessing levels of target RNA(s) in human induced pluripotent stem cell (hiPSC)-derived motor neurons treated with SFPQ siRNA (or a control siRNA, or mock treated with vehicle/transfection reagents only).
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH acts as a housekeeping gene to which other target RNAs are normalised. Alternatively, fig1F displays the ratio of altSFPQ to wtSFPQ RNA levels (i.e., altSFPQ / wtSFPQ).
Comments: replicates reflect 3 separate experiments using the same iPSC line (control 3). SFPQ pre-mRNA (specifically primers target a constitutively spliced intronic region) was assessed on a separate plate; hence, housekeeping gene GAPDH was rerun alongside it and used to normalise SFPQ pre-mRNA values
Tab 4: 1G_S2I_S4A_SFPQ-OE
Description: assessing levels of target RNA(s) in HeLa cells transfected with recombinant SFPQ proteins (or a control empty vector plasmid, or mock treated with vehicle/transfection reagents only)
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH and POLR2B act as housekeeping genes to which other target RNAs are normalised (using their geometric mean).
Tab 5: 1H_S4B_SFPQ-OE
Description: assessing levels of target RNA(s) in NGN2-i3Neurons transduced with lentiviruses to express recombinant SFPQ proteins (or a control lentivirus, or untreated condition)
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH acts as a housekeeping gene to which other target RNAs are normalised.
Tab 6: S2A_Cq_panel
Description: assessing levels of target RNA(s) in (i) human induced pluripotent stem cells (hiPSC), (ii) hiPSC-derived DIV14 neural precursor cells (NPC), and (iii) hiPSC-derived DIV35 motor neurons (MN) biochemically separated into nuclear and cytoplasmic fractions
Data type: qPCR, mean Ct (i.e., threshold cycle) values
Comment: iPSC and NPC datasets constitute 4 different biological lines, one experimental replicate; MN dataset constitutes 4 different biological lines, two experimental replicates
Tab 7: S2B_a-amanitin
Description: assessing levels of target RNA(s) in HEK293T cells treated with alpha-amanitin transcription inhibitor; time (hours) indicates time after treatment at which cells were harvested
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH acts as a housekeeping gene to which other target RNAs are normalised.
Comment: The dataset consists of just one single experimental replicate
Tab 8: S2C_CHX
Description: assessing levels of target RNA(s) in HEK293T cells treated with cycloheximide (CHX) translation inhibitor (or untreated, or DMSO vehicle treated)
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH and POLR2B act as housekeeping genes to which other target RNAs are normalised (using their geometric mean).
Tab 9: S2F_SMG1i
Description: assessing levels of target RNA(s) in hiPSC-derived motor neurons treated with SMG1 small molecule inhibitor (or untreated, or DMSO vehicle treated)
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH and POLR2B act as housekeeping genes to which other target RNAs are normalised (using their geometric mean).
Comment: dataset consists of 3 different biological hiPSC lines (control lines 1, 3, and 4)
Tab 10: 2B_polysome
Description: assessing levels of target RNA(s) in polysome sucrose gradient fractions extracted from hiPSC-derived DIV7 neural precursors
Data type: qPCR, mean Ct (i.e., threshold cycle) values. Data inthe paper is presented as a percentage per fraction relative to the total (i.e., the sum of the target RNA amount across all fractions) for each target
Comment: dataset consists of 3 different biological hiPSC lines (control 3, ALS-VCP patient line 3, ALS-VCP patient line 4)
Tab 11: 2C_S3B_S3C_polysome
Description: assessing levels of target RNA(s) in polysome sucrose gradient fractions extracted from hiPSC-derived DIV14 neural precursors; lysates were then either treated with EDTA to disrupt polysomes/ribosomes or untreated
Data type: qPCR, mean Ct (i.e., threshold cycle) values. Data inthe paper is presented as a percentage per fraction relative to the total (i.e., the sum of the target RNA amount across all fractions) for each target
Comment: dataset consists of 3 different biological hiPSC lines (control 1, control 3, ALS-VCP patient line 4)
Tab 12: 2F_S3H_ASO20
Description: assessing levels of target RNA(s) in HEK293T cells treated with SFPQ -targeting antisense oligonucleotide 20 (ASO20) (or non-targeting control ASO)
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH acts as a housekeeping gene to which other target RNAs are normalised.
Tab 13: S3A_polysome
Description: assessing levels of target RNA(s) in polysome sucrose gradient fractions extracted from HEK293T cells; lysates were then either treated with EDTA to disrupt polysomes/ribosomes or untreated
Data type: qPCR, mean Ct (i.e., threshold cycle) values. Data in the paper is presented as a percentage per fraction relative to the total (i.e., the sum of the target RNA amount across all fractions) for each target. NA = non-applicable cells since no target RNA was identified in these fractions
Comment: dataset consists of 1 biological hiPSC line
Tab 14: S3G_ASO_screen
Description: assessing levels of target RNA(s) in HEK293T cells treated with 20 SFPQ -targeting antisense oligonucleotides (or mock, i.e., transfection reagents only)
Data type: qPCR, mean Ct (i.e., threshold cycle) values. wtSFPQ and altSFPQ are normalised to* SFPQ constitutive *(i.e. region common to splice variants) to give ‘percentage splicing’.
Tab 15: S4D_NONO_HTP_IF
Description: high throughput image analysis of NONO (antibody) fluorescent intensity in HEK293T cells transfected with recombinant SFPQ expression plasmids (or an empty vector plasmid control condition); NONO intensity was measured in cell nuclei or perinuclear regions (defined as nuclear masks expanded by 10 um to generate a perinuclear ring region) to determine the nuclear:perinuclear distribution of the protein
Data type: high throughput analysis of immunofluorescence performed on Opera Phenix; data represents summary values per well from 36 fields of view imaged per well
- Replicate_bio: biological replicates (i.e., separate transfection / experimental setup)
- Replicate_tech: technical replicates (i.e. duplicate wells per biological replicates)
- Cell – Number of objects: total number of objects identified as cells per well
- Cell - Intensity Nuclear region Alexa 568 (NONO) Mean - Mean per Well: average across well of mean intensity of AlexaFluor568-labelled NONO intensity (a.u.) in cell nuclei
- Cell - Intensity Cytoplasmic Region Alexa 568 (NONO) Mean - Mean per Well: average across well of mean intensity of AlexaFluor568-labelled NONO intensity (a.u.) in cell perinuclear ring region
- Cells positive for Alexa 488 (HA-tag) - Number of Objects: number of cell objects positive for AlexaFluor488 (HA-tag antibody; note that recombinant SFPQ proteins have an HA-tag) above a set threshold
- % of Cells 488+ (i.e., transfected cells): percentage of cell objects positive for AlexaFluor488 (HA-tag antibody; note that recombinant SFPQ proteins have an HA-tag) above a set threshold, relative to the total number of cell objects identified per well
- Ratio Alexa 568 (NONO) - Nucleus/Cytoplasm: value in ‘Cell - Intensity Nuclear region Alexa 568 (NONO) Mean - Mean per Well’ divided by value in ‘Cell - Intensity Cytoplasmic Region Alexa 568 (NONO) Mean - Mean per Well’
- Number of Analyzed Fields: number of analysed fields per well
Comments: Yellow-highlighted cells associated with one empty vector condition were not included in downstream analysis/data presentation, as was quantifiably determined as an outlier (based on 1.5 IQR rule).
Tab 16: S5F_S5G_WC_MN_stress
Description: assessing levels of target RNA(s) in hiPSC-derived motor neurons treated with sodium arsenite (oxidative stressor), sorbitol (osmotic stressor), or untreated
Data type: qPCR, mean Ct (i.e., threshold cycle) values. GAPDH acts as a housekeeping gene to which other target RNAs are normalised.
Comment: dataset consists of 3 different biological hiPSC lines (controls 2, 3, and 5)
Tab 17: 4E_4F_S5D_llps_droplet
Description: quantification of liquid-liquid phase separation (LLPS) droplet formation from homotypic eGFP-tagged wtSFPQ and altSFPQ proteins in vitro
Data type: image analysis of LLPS droplet formation. Data represents summary values per image (one image per protein condition and per replicate)
- Sum_GFP_Intensity_SolubleFraction: total GFP intensity in regions outside of droplets (a.u.)
- Sum_GFP_Intensity_droplets: total GFP intensity within droplets (a.u.)
- Median_Area: median area of droplets
Tab 18: 4IJ_S5HIJ_frac_MN_stress
Description: assessing levels of target RNA(s) in hiPSC-derived motor neurons treated with sodium arsenite (oxidative stressor), sorbitol (osmotic stressor), or untreated, and biochemically separated into nuclear and cytoplasmic fractions
Data type: qPCR, mean Ct (i.e., threshold cycle) values. NIT1 and NFX1 act as housekeeping genes (geometric mean) to which other target RNAs are normalised.
Comment: dataset consists of 3 different biological hiPSC lines (controls 2, 3, and 4)
Tab 19: 5A_S7A_SFPQ_RIP
Description: RNA immunoprecipitation performed on N2A neuroblastoma cells expressing HA-tagged SFPQ variants (or transfected with empty vector plasmid).
Data type: qPCR, mean Ct (i.e., threshold cycle) values. Each individual condition has an input and an RNA-immunoprecipitation sample for each replicate. Levels of target RNAs in RIP samples are normalised to levels in the associated input sample and expressed as ‘percentage of total input’.
Comment: The dataset consists of 3 different biological hiPSC lines (controls 2, 3, and 4). Note that different dilution factors need to be applied to RIP samples relative to input samples:
- Replicates 1 & 2: dilution factor = 40
- Replicate 3: dilution factor = 0.4
Tab 20: 6B_DIV14_SFPQ
Description: assessing levels of target RNA(s) in hiPSC-derived DIV14 neural precursor cells (NPC) biochemically separated into nuclear and cytoplasmic fractions
Data type: qPCR, mean Ct (i.e., threshold cycle) values. NIT1 and NFX1 act as housekeeping genes (geometric mean) to which other target RNAs are normalised (i.e., ‘transcript abundance’). ‘Splicing ratio’ is determined by dividing SFPQ splice variants (wtSFPQ and altSFPQ) by the constitutive SFPQ target region.
Comment: dataset constitutes 8 different biological lines (controls 1, 3, 4, 5; VCP-ALS patient-derived lines 1, 2, 3, 4), and two experimental replicates
File: Doc_2_BaseScope_SFPQ_HEK293T_SMG1i.xlsx
Description: tabulated Excel spreadsheet.BaseScope RNA-FISH analysis of day 6 terminally differentiated hiPSC-derived motor neurons treated with sodium arsenite - summary data (object number and area)
Document 2 notes:
Target:
- TV1 = wtSFPQ
- TV2 = altSFPQ
RNA refers to BaseScope RNA puncta
Conditions:
- OX = sodium arsenite oxidative stress
- UT = untreated
Replicates reflect different biological hiPSC-derived day 6 (DIV24) motor neuron cell lines (control lines 1, 2, and 3)
File: Doc_3_BaseScope_SFPQ_D6MN_OX.xlsx
Description: tabulated Excel spreadsheet. BaseScope RNA-FISH analysis of HEK293T cells treated with SMG1 inhibitor - summary data (object number and area)
Document 3 notes:
Target:
- TV1 = wtSFPQ
- TV2 = altSFPQ
RNA refers to BaseScope RNA puncta
Conditions:
- SMG1i = SMG1 small molecule inhibitor
- UT = untreated
File: Doc_4_Neeves_et_al_SA_Western_blots_full.pdf
Description: All full Western blots used within the study.
File: Doc_5_Neeves_et_al_sedimentation_gels.pdf
Description: All full gel images used for quantification of sedimentation assays within the study.
File: Doc_6_SFPQ_cotransfection_imaging.xlsx
Description: tabulated Excel spreadsheet. Image analysis of SFPQ cotransfection immunofluorescence assay- nuclear and perinuclear (generated via expansion of nuclear margins per cell) per object data (object number, area, integrated intensity, median intensity)
Document 6 notes:
Replicate: N1 – 3 (note upper case N)
Field: n1 – 7 (note lower case N)
Condition:
- wt = eGFP-wtSFPQ expressing cells
- alt = eGFP-altSFPQ expressing cells
- cotf = eGFP-wtSFPQ + mAPPLE-altSFPQ cotransfected cells
Each row reflects an individual object
Each tab represents an analysis of each of the 3 conditions for each replicate. Data contained within represents the following for each individual object (i.e., cell), in both nuclear and perinuclear cell regions:
- Area
- Integrated intensity of GFP
- Integrated intensity of mAPPLE
- Median intensity of GFP
- Median intensity of mAPPLE
For more details, please refer to the methods sections within the associated article.
Code/software
All qPCR data were analysed in Excel and Prism.
Image analysis was performed through a combination of ImageJ and CellProfiler processing, except Opera Phenix high-content imaging (Harmony high-content imaging and analysis software used).
Western blot & gel densitometry analyses were performed in ImageJ or ImageStudioLite.
For more details, please refer to the methods sections within the associated article.
Access information
NA
Human subjects data
Informed consent was obtained from healthy control donors and patients in this study. Experimental protocols were enacted according to regulations and guidelines outlined by UCLH’s National Hospital for Neurology and Neurosurgery and UCL’s Institute of Neurology joint research ethics committee (09/0272).
