Whole-organism screening in a zebrafish model of CLN2 disease identifies pregnenolone as a modulator of lysosomal functions with anti-epileptic properties

Lysosomal storage disease, neurodegeneration, epileptic seizures, autophagy, steroids, pregnenolone .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Corresponding author: Claire Russell, Comparative Biomedical Sciences, Royal Veterinary College, 4 Royal College Street, London, UK; [email protected] .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint

Lysosomal storage disorders (LSDs), a group of inherited genetic diseases, are often associated with early-onset neurodegeneration and refractory epileptic seizures. In CLN2 disease, an LSD caused by recessively inherited dysfunction of lysosomal serine protease Tripeptidyl Peptidase 1 (TPP1), lysosomes are functionally impaired through a characteristic accumulation of subcellular materials. Here, we develop and apply a whole -organism screening workflow in tpp1-/- zebrafish to identify small molecules that suppress epileptic seizures — a hallmark of the human disease — in this model. Among 640 US Food and Drug Administration-approved drugs, pregnenolone, an endogenous precursor for steroid biosynthesis, efficiently suppresses seizures and cell death in tpp1-/- zebrafish. Using a semi-automated high-content workflow, we further show that pregnenolone normalizes lysosomal architecture in tpp1-/- zebrafish. Pregnenolone stimulates steroid hormone biosynthesis and related gene expression, which is dysregulated in tpp1-/- zebrafish. Taken together, tpp1-/- zebrafish are a suitable model to study CLN2 disease , in which we have identified pregnenolone as a candidate with therapeutic properties. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint

The Neuronal Ceroid Lipofuscinoses (NCL) are a group of lysosomal storage disorders (LSDs), also known as Batten disease, which collectively constitute the most common cause of inherited paediatric neurodegenerative disorders .1,2 There are currently 13 known forms of NCL, each caused by mutation of individual genes. The majority of affected proteins in NCLs localise to the lysosomal lumen or membrane, however others are localised to other cellular compartments, including the endoplasmic reticulum (ER), Golgi, cytosol and plasma membrane.3 Despite diversity in the roles of the affected proteins, the 13 currently classified NCLs share a hallmark accumulation of auto -fluorescent lysosomal storage material and common characteristics including progressive neurodegeneration resulting in motor and cognitive decline, retinopathy leading to blindness, myoclonic epilepsy and severely premature death.2 The lysosomal storage material in the NCLs predominantly consists of ceroid - lipopigments, as well as subunit c of mitochondrial ATP synthase (SCMAS) or sphingolipid activator proteins A and D.3,4 CLN2 disease, otherwise known as the classic Late Infantile form of NCL (cLINCL), is caused by autosomal recessive inherited mutation in the CLN2 gene, leading to dysfunction of the encoded lysosomal serine protease enzyme Tripeptidyl Peptidase 1 (TPP1) .5,6 The disease causes cell death in the central nervous system (CNS) and neural retina.7 Children with CLN2 disease will typically present at 2–4 years old. Early symptoms can include seizures, language delay, motor dysfunction, behavioural problems and dementia. Progression of the disease leads to rapid motor and cognitive deterioration, progressive visual impairment, the development of refractory seizures and death normally within early adolescence.8 Current treatments for NCLs are limited and focus on symptom management.9,10 CLN2 disease is currently the only NCL in which there is a clinically approved treatment that has been shown to effectively attenuate disease progression, in the form of enzyme replacement therapy for the brain,11 but peripheral disease is likely to emerge in patients receiving enzyme replacement therap y,12 which has already been noted in the retina .13 Hence, further treatments are still needed. Development of further therapies is reliant on a deeper understanding of the disease pathophysiology, alongside effective animal models to identify and evaluate potential treatment strategies. In this study we use a zebrafish model of CLN2 disease to perform a drug screen of 640 FDA- approved compounds. The zebrafish model harbours a premature stop codon mutation in exon 3 of the tpp1 gene, owing to a single T > A point mutation in the tpp1sa0011 allele (tpp1sa0011-/-, .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint herein referred to as tpp1-/-). The mutation results in deficiency of the Tpp1 enzyme, confirmed by Western blot and an enzyme activity assay .14 Previous characterisation of the tpp1-/- zebrafish demonstrated their value as a tool for drug discovery ; various aspects of the human CLN2 disease are replicated in the tpp1-/- zebrafish, including storage accumulation and enlarged lysosomes, motor defects and retinal and neuronal degeneration. These disease phenotypes can be observed from 2 days post-fertilisation (dpf) and the zebrafish do not survive past 7 dpf.14 Another notable phenotype of tpp1-/- zebrafish is a period of increased locomotion, indicative of seizures. By using a combination of qualitative and quantitative phenotypes, we performed a relatively high-throughput drug screen. Our drug screen identified the neurosteroid pregnenolone as a potential therapeutic for CLN2 disease. Here, w e show that pregnenolone reduces duration of movement , distance and velocity of seizure-like locomotion activity in tpp1-/- zebrafish, demonstrating a rescue of the phenotype. The anti -epileptic effect of pregnenolone was confirmed by electroencephalogram (EEG) measurement of epileptiform activity in the tpp1-/- zebrafish. Using live imaging of a novel transgenic zebrafish line expressing a Lamp1 lysosomal reporter, we go on to show that pregnenolone alleviates the lysosomal phenotype in tpp1-/- zebrafish. Furthermore, we show that pregnenolone treatment leads to a reduction in cell death in the CNS and enhanced clearance of autophagic material. Finally, given that pregnenolone is a direct metabolite of cholesterol and the precursor of all steroids, we go on to investigate the cholesterol -hormone pathway in tpp1-/- zebrafish. Measurement of steroid hormone levels and whole genome RNA sequencing reveal dysregulation of hormone biosynthesis and related genes in tpp1-/- zebrafish compared with their healthy (phenotypically wild -type) ‘ WT’ siblings. This observation indicates a possible perturbation in the cholesterol pathway in CLN2 disease pathogenesis , highlighting an important area for further research.

Generation and maintenance of zebrafish This study is reported in line with the ARRIVE guidelines. Adult zebrafish were maintained in a multi-rack aquarium system and kept on a constant 14/10 hr light/dark cycle at 27–29 C and pH 6.0 –7.5.15 Adult z ebrafish were fed 2 -3 times per day a diet of Hikari, krill and live brineshrimp. Experiments conducted at the Royal Veterinary College were done so with approval from the college, the UK Home Office (PPL: PEB686695, PIL: I5DAE98B4) and local Animal Welfare Ethical Review Board (AWERB) under the Animal (Scientific .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Procedures) Act 1986 Amendment Regulations 2012. Experiments conducted at Nestlé Research (Lausanne, Switzerland) were done so according to Swiss and European Union ethical guidelines, with approval by the animal experimentation ethical committee of Canton of Vaud ( permits VD -H13 and VD3177 ). Where required, embryos were mechanically dechorionated with forceps. For Schedule 1 culling, adult and embryonic/larval zebrafish were exposed to 0.15 % 2-phenoxyethanol for 10 minutes or 24 hr respectively. Adult zebrafish were then decapitated and embryos/larvae were crushed. Generation of tpp1sa0011 homozygotes and normal healthy siblings (WT) Zebrafish carrying a T/A point mutation in the tpp1sa0011 allele were generated by N-ethyl-N- nitrosourea mutagenesis in the Tübingen strain and originally obtained from the Sanger Institute (Cambridge, UK: www.sanger.ac.uk/Projects/D_rerio/zmp/) as an outcross to WT Tupfel long fin fish14 and subsequently maintained for 12 years by backcrossing to WT Tupfel long fin fish. Adult heterozygous carriers of the tpp1sa0011 mutation (herein referred to as tpp1+/- ) were identified by KASP genotyping. Briefly, this method involves allelic discrimination determined by qPCR reaction ratio of fluorescent cassettes, FAM and HEX, which correspond to T/A allele-specific primers developed by LGC Biosearch Technologies (https://biosearch - cdn.azureedge.net/assetsv6/KASP-genotyping-chemistry-User-guide.pdf). Embryos for experiments were generated by in-cross of tpp1+/- zebrafish and raised at 28 C under standard husbandry conditions. Homozygous embryonic genotypes were assigned owing to the presence of a retinal phenotype on or after 2 dpf.14 Normal (WT) siblings were used as controls. Prior to any experiment, zebrafish embryos/larvae were viewed under a dissecting microscope to remove unfertilised embryos and to ensure they were healthy, or in the case of tpp1-/-, that they did not have any abnormal phenotypes beyond the expected ones (i.e. reduced retina size, small head, small and curved body, heart oedema, seizure-like locomotion). If any other phenotypes or injuries were observed in either WT or tpp1-/- zebrafish, they were excluded from the study. Generation of transgenic zebrafish and confocal imaging Generation of lines Transgenic zebrafish Tg(actc1b:lamp1-ZsGreen)nei08, Tg(actc1b:lc3-ZsGreen)nei08, Tg(actc1b:tfeb-ZsGreen)nei08 and Tg(actc1b:nls-mCherry)nei09 16 were independently generated using I-SCEI meganuclease-mediated transgenic insertion into 1-cell stage .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint embryos with tpp1+/- background as previously described.17 Individual founders were selected as ZsGreen-positive or mCherry-positive for propagation of each transgenic line. For ZsGreen-LC3 and ZsGreen-Lamp1 experiments, subsequent generations were crossed with non-transgenic tpp1+/- and the progeny of the second generation were selected for fluorescence and used for imaging experiments. For ZsGreen-TFEB experiments, subsequent generations of Tg(actc1b:tfeb-ZsGreen)nei08 and Tg(actc1b:nls-mCherry)nei09 were crossed to each other and the progeny were selected for both ZsGreen and mCherry fluorescence and used for imaging experiments. Image acquisition and analysis Dechorionated embryos, anaesthetised with 0.016 % tricaine (made in aquarium water, buffered with 4 % 1 M Tris pH 9 and adjusted to pH 7 with NaOH), were mounted laterally on 1.5mm glass coverslips with 1 % low-melting point agarose gel in aquarium water. Live imaging was conducted using a 40x water objective on either a LeicaTM SP5 or SP8 confocal microscope. ZsGreen-LC3 and ZsGreen-Lamp1 Confocal images were analysed as maximum projections of 8 –10 z-stack images using FIJI software. Binary images were generated by applying a threshold to highlight fluorescent signal and eliminate background; the threshold value was the mean fluorescent intensity of all fish in one experiment multiplied by 1.5 for ZsGreen -LC3 and 2 for ZsGreen -Lamp1. Region of interest was defined manually using the freehand drawing tool or by applying a threshold to outline the total area of fish muscle. Measurements of ZsGreen-Lamp1 were derived using the ‘Analyse Particles’ feature on FIJI. ZsGreen-LC3 was measured as a total area of that identified by threshold. Area and puncta count measurements were normalised to the total area o f fish muscle per image. ZsGreen-TFEB and mCherry-nls Confocal images consisting of 6 z -stack images were acquired and analysed individually. Nuclei were identified by a binary image generated by applying a threshold to highlight fluorescent signal of mCherry -nls and eliminate background; the threshold value was kept consistent per experiment. Region of interest was identified by applying a threshold using the ZsGreen-TFEB channel to find the total area of fish muscle. ZsGreen-TFEB was measured as .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint total raw fluorescent units over raw fluorescent units within the identified nuclei. Area and puncta count measurements were normalised to the total area of fish muscle per image. Drug screen At 48 hours post-fertilisation (hpf), dechorionated tpp1-/- and WT siblings were placed one embryo per well of a flat bottomed 96 round well plate in 49.5 µl aquarium water as shown (Fig 1). Each plate contained one set of WT + DMSO, one set of tpp1-/- + Dimethyl sulfoxide (DMSO) and the rest of the plates were sets of tpp1-/- + each compound. 0.5 µl of DMSO or compound (Enzo Life Sciences FDA-approved library of known bioactives ; 400 µg/ml in DMSO) was added to each well to a final compound concentration of 4 µg/ml. In Screen 1, two tpp1-/- zebrafish were used per compound and plates were incubated until the fish were 120 hpf and then placed in the DanioVision chamber (Noldus) for a 20 minute recording. Outputs from Etho Vision tracking software (Noldus) for each fish were distanced moved, mean velocity, max velocity, time spent moving and number of movement bouts (start velocity 0.4 mm/s, stop velocity 0.2 mm/s). Fish were then qualitatively scored for morphology, and quantitatively survival and touch response before culling. In Screen 2, the number of tpp1-/- zebrafish per compound was increased to three to improve reliability, and the assays were expanded to gain richer data and to include seizure-like locomotion which can only be detected at 72 hpf as loss of locomotion at later stages means that seizures can no longer be detected through assaying movement. We also improved welfare by reducing the age of zebrafish at the last assay time point. Plates were incubated at 28 C until scored qualitatively for retinal size and quantitatively for survival at 54 hpf, 73 hpf and 79 hpf and then incubated at 22 C until scoring for retinal size and survival at 97 hpf, and then culled. They were also assessed quantitatively for seizure -like locomotion using Danio Vision for 1 hour (with temperature controlled to 25 C and the light on) at 7 2 hpf (total distance moved, time spent moving and number of movement bouts; start velocity 4.5 mm/s, stop velocity 2 mm/s) and for loss of locomotion for 20 minutes at 28 C at 96 hpf (total distance moved, time spent moving and number of movement bouts; start velocity 0.4 mm/s, stop velocity 0.2 mm/s). Compounds that appeared to improve at le ast one phenotype were re -tested on 10 animals per group (tpp1-/- + DMSO, tpp1-/- + compound, WT + DMSO) using the same screen protocol but compound purchased from a new source. The compound that still had beneficial effects after re -screen (pregnenolone) was retested blind in two replicate experiments following the screen protocol .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint except using approx. Ten zebrafish per group (tpp1-/- + DMSO, tpp1-/- + pregnenolone, WT + DMSO, WT + pregnenolone). Chemical treatments Embryos were dechorionated mechanically with forceps prior to any treatment. Pregnenolone ( Preg, Sigma, P9129) treatment was performed at 2 dpf for 24 –48 hr, as indicated. 0.4 mg/mL pregnenolone dissolved in DMSO was added to aquarium water for a final concentration of 4 g/mL in 6 well plates with 4.5 mL total volume. 1 % DMSO was used as a control. Dechorionated embryos were incubated in treatment for 24 hr at 28 C. For 48 hr treatment, treatments were refreshed at 24 hr. 3-Methyladenine (3 -MA, Sigma, M9281) treatment 18 was performed at 2 dpf for 24 hr, alongside or in combination with pregnenolone as above. 3-MA was dissolved in ddH2O with gentle heating for a stock concentration of 0.33 M, then added to aquarium water for a final concentration of 10 mM in 6 well plates with 4.5 mL total volume. 1 % DMSO + ddH2O was used as a control. For LC3 -ZsGreen flux assay, 500 L 1 M NH 4Cl was added directly into 6 well plate containing 4.5 mL of 4 g/mL pregnenolone (or DMSO control), for a final concentration of 100 mM. The NH4Cl was added in the last 4 hr of pregnenolone treatment prior to imaging. 10 % ddH2O was used as a control. Electroencephalogram (EEG) measurements The method was adapted from19. Following pregnenolone treatment at 2 dpf, 3 dpf larvae were placed in 5mM of D-tubocurarine (Fluka) for 10 minutes, rinsed in aquarium water, and then stabilised in 1.8 % low-melting point agarose gel in aquarium water before placing an electrode onto the optic tectum of the midbrain. Electroencephalogram (EEG) measurements were then recorded for an average of 45 minutes using Axioscope software. Recordings were generated using Clampfit 10 and analysed using Origin 7 software to obtain the mean amplitude. Immunoblotting Dechorionated 4 dpf larvae maintained on ice, were deyolked by repeat pipetting in Phosphate Buffered Saline (PBS) with 0.1 mM EDTA through a p200 pipette. Samples were then homogenized in RIPA buffer (Sigma) containing 1x protease and phosphatase inhibitors .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint (ThermoScientific), then centrifuged at 5,000 g for 15 minutes at 4 °C to collect the soluble fraction. Protein concentrations were quantified by BCA protein assay (ThermoScientific) and equal concentrations of lysates were mixed with NuPAGE LDS Sample Buffer (4x) (Invitrogen) and NuPAGE Sample Reducing Agent (10x) (Invitr ogen), then heated at 70 °C for 10 minutes. Samples were resolved on 4 –12 % NuPAGE BiS -Tris gels (Invitrogen) and transferred to nitrocellulose membranes (Invitrogen), which were blocked w ith 5 % milk powder in TBS -Tween. The membrane was probed with primary antibodies (anti-HSC70 [SantaCruz, sc -7298]; anti -Tom20 [SantaCruz, sc -11415; anti -Subunit C of ATP Synthase (SCMAS) [Abcam, ab110273], anti -P62 [Cell Signalling, 5114S]) overnight at 4 °C and secondary antibodies (anti-mouse/rabbit IgG, HRP -labelled [Perkin Elmer, NEF822001EA/NEF812001EA]) for 1 hr at room temperature, with washing in TBS -Tween in between. Following washing, blots were visualised by ECL kit (ThermoScientific) as directed by manufacturer’s instructions. TUNEL assay Apoptosis was detected by the DeadEnd TM Fluorometric TUNEL system (Promega) , performed based on principles described by Gavr ieli et al.20 using a protocol developed from Promega (2009). Larvae were maintained in 0.003 % 1-phenyl 2-thiourea (PTU) after 24 hpf. Following pregnenolone treatment larvae at 4 8 hpf, 72 hpf zebrafish were fixed in 4 % paraformaldehyde in PBS (PFA), and stored overnight at 4 °C, before storage in methanol at - 20 °C. Larvae were rehydrated in PBT (50:50 % methanol/PBS solution containing 0.1 % TritonTM X-100 (Sigma -Aldrich)) prior to permeabilization with 10 g/mL Proteinase K (Promega) in PBT for 60 minutes. Larvae were then washed in PBT, fixed for 20 minutes in 4 % PFA, then chilled for 10 minutes at -20 °C in a 2:1 solution of ethanol and acetone. The larvae were then washed again in PBT before incubation in 100 l Equilibration Buffer (Promega) on a rocker for 30 minutes, before incubation in TUNEL stain buffer (35 L Equilibration Buffer, 10 L Nucleotide Mix, 2 L Terminal Deoxynucleotidyl Transferase, Recombinant enzyme (Promega) at 37 °C for 60 minutes inside a humified chamber, protected from light. The reaction was terminated by the addition of 500 l of 2 x SSC (Promega) for 15 minutes, followed by washes in PBT, after which the larvae were cleared in 70 % glycerol/PBS solution. Larvae were mounted laterally on 1.5mm glass coverslips with 2 % low-melting point agarose gel in aquarium water. TUNEL staining was imaged using a Leica SP5 confocal microscope at 10x magnification. Fluorescent intensity was quantified using Volocity software .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint version 6.3 (PerkinElmer). Mean fluorescent intensity was measured within a cuboid region of interest, located within the mesencephalon, identified as the region within the head of the larvae between the visually identifiable structures of the eye and otic capsule. Seizure-like locomotion assay Following dechorionation and 24 hr treatment with pregnenolone and /or 3MA, 3 dpf larvae were transferred to individual wells of a 96 square well plate in 200 L. The plate was mounted in a DanioVision (Noldus) tracking system at 25 °C with the chamber light on and movement of the larvae was tracked for 20 minutes. Movement parameters of individual larvae were quantified using Etho Vision XT (Noldus) software. Zebrafish were assayed for seizure -like activity as in the drug screen. RNA Sequencing Following dechorionation and 24 hr pregnenolone treatment from 2 dpf, 3 dpf larvae were anaesthetised with 0.016 % tricaine (made in aquarium water, buffered with 4 % 1M Tris pH 9 and adjusted to pH 7 with NaOH) and transferred to a 50 mL petri dish in a drop of water for dissection to separate heads and tails , and remove the yolk . The larvae were dissected with a 0.13 - 0.17 mm thick square glass coverslip. Two transverse cuts were made: 1, caudal to the hindbrain and rostral to the heart; 2, caudal to the yolk sac. Heads and tails were separated and all water was removed before flash freezing in liquid nitrogen. Six replicates of each condition were treated and processed in batches on the same day within a two-hour period. Samples were disrupted and lysed using the FastPrep-24 speed 6 2 x 60” in 500 L of kit lysis buffer. RNA was extracted from 400 L of lysate and eluted in 50 L. The RNA was then quantified using Quant-it RiboGreen assay (Life Technologies) and underwent quality control on a Fragment Analyzer. Libraries were prepared with 150 ng of RNA input using the QuantSeq 3’ mRNA-Seq Library Prep Kit (FWD) HT for Illumina (Lexogen) and then quantified with Quant-it Picogreen (Life Technologies). Library sizes were controlled with the High Sensitivity NHS Fragment Analysis kit on a Fragment Analyzer (Agilent). RNA sequencing was performed on HiSeq 2500 with Rapid V2 chemistry SR 65 cycles (Illumina) loaded with 3 % Phix, using two Flow cells. Raw count table was produced by HTseq -count21 and differential expression analysis was performed using Bioconductor R package edgeR (version 3.36).22 Gene set enrichment analysis .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint was performed using CAMERA (limma Bioconductor R package, version 3.50.3) with custom gene sets (Supplementary Tables 2 and 3).23 Hormone analysis in zebrafish larvae Dechorionated 48 hpf zebrafish were treated in groups of 25 with DMSO or pregnenolone for 24 hr. At 3 dpf, following treatment, two groups of samples were pooled so that 50 larvae were transferred to 1.5 mL cryotubes and all water was removed before flash freezing in liquid nitrogen. Samples were shipped on dry ice and further sample processing and hormone analysis using a Sciex QTRAP 6500+ mass spectrometry system was performed by The Metabolomics Innovation Centre, Canada. Hormone measurements were normalised to an average weight of 3 dpf WT or tpp1-/- mutant larvae. Statistical analysis Data are expressed as box -and-whisker plot (minimum to maximum) or mean ± SD, as indicated. Statistical analysis for all data were performed using unpaired t-test or linear mixed model in GraphPad Prism 9 and SPSS respectively. Differences between groups were considered statistically significant when P < 0.05. Statistical significance is indicated in figures.

Drug screen in tpp1-/- zebrafish identifies pregnenolone as a hit compound reducing seizure-like locomotion and cell death As locomotion of zebrafish is easily and rapidly monitored, locomotion in tpp1-/- zebrafish, was used as the primary quantitative measure to screen 640 FDA -approved compounds . Locomotion assays were supplemented with other quantitative and qualitative assays (Fig 1A, Supplemental Table 1 ). Among the 640 screened compounds, 34 of these were identified as potential hits due to improvement in at least one assay and re-tested using a new batch of compound. Among these, pregnenolone (from screen 2) was identified as the only hit due to apparent improvement in retinal size, seizure-like locomotion at 72 hpf, locomotion at 96 hpf and survival in 96 hpf. No other compounds showed a reduction in any phenotypes when re - tested. However, after testing pregnenolone under blind conditions, only a reduction in duration of movement in tpp1-/- zebrafish in the seizure -like locomotion assay remained statistically significant. Blind testing therefore validated th e effect of pregnenolone on seizure -like .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint locomotion, as demonstrated by a significant reduction in movement duration, distance and velocity in tpp1-/- zebrafish (Fig 1 B,C). The anti -seizure effect of pregnenolone on tpp1-/- zebrafish was verified by electroencephalogram (EEG) analysis at 3 dpf after a 24hr treatment with 4 µg/ml pregnenolone (Fig 1 D,E). Typical seizure-like brain activity in tpp1-/- zebrafish can be seen in Fig 1D, showing single representative EEG recordings from tpp1-/- zebrafish treated with DMSO or pregnenolone, alongside healthy WT siblings. Mean EEG amplitude from all measured tpp1-/- zebrafish (n=6 DMSO treated, n=8 pregnenolone treated) show a significant reduction in EEG amplitude in those treated with pregnenolone compared with the DMSO control. These results support the locomotion assay results described above, showing an anti - seizure effect of pregnenolone in tpp1-/- zebrafish. Although we did not detect a statistically significant improvement of other phenotypes in our re-screen and blind testing of pregnenolone, we wondered whether cell death in tpp1-/- zebrafish brains was altered by pregnenolone treatment. A TUNEL cell death assay revealed a significant reduction of cell death in the mesencephalon of tpp1-/- zebrafish treated with pregnenolone for 3 hours from 52 hpf, compared with DMSO (Fig 1 F,G). This finding suggests pregnenolone

in an improvement in pathology as well as seizures. Pregnenolone ameliorates lysosomal structure and function in tpp1-/- zebrafish Pregnenolone is an endogenous neurosteroid , a direct metabolite of cholesterol and the precursor of all steroid hormones.24,25 Pregnenolone is also an agonist of the sigma-1 receptor, which binds and regulates cholesterol levels within subcellular compartments, especially in lipid rafts.26 Pregnenolone was recently identified in another screen for compounds improving lysosomal storage in C LN3 disease, also known as juvenile Neuronal Ceroid Lipofuscinosis: in ARPE-19 CLN3 knockout cells pregnenolone was shown to reduce storage accumulation in the lysosomes.27 We went on to investigate whether pregnenolone has a therapeutic effect on the cell pathology of tpp1-/- zebrafish. To this end, we generated transgenic zebrafish with ZsGreen-tagged lysosomal membrane protein Lamp1 under the muscle -specific actc1b promoter in the background of tpp1-/- zebrafish. The muscle reporter provides a large area of homogenous tissue allowing for accurate high -throughput measurement of the number and .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint average size of lysosomes in live zebrafish, which would be more complicated in brain tissue owing to its smaller size and vast differences in lysosomal function between different neuronal cell populations. Quantification of confocal images show a strong lysosomal phenotype in the muscle of tpp1-/- zebrafish; the average size of lysosomes is significantly increased and the number of lysosomes is significantly decreased compared to healthy WT siblings (Fig 2 A,B). Both these parameters were significantly improved in tpp1-/- zebrafish treated with pregnenolone for 24 hr, compared to the DMSO control. Interestingly, pregnenolone had the same effect on the lysosomes of WT zebrafish , namely a significant decrease in average lysosome size and significant increase in lysosome number (Fig 2 A,B). Autophagy is a key function of the lysosome and a process that has been shown to be impaired in various LSDs, including CLN2 disease .28-32 The neuroprotective effect of autophagy induction and the inverse detriment caused by defects in autophagy have been widely demonstrated in the literature among a multitude of neurodegenerative diseases .30,33,34 Given that we see a decrease in lysosomal size with pregnenolone treatment, we measured P62, an autophagic adaptor protein, to test whether pregnenolone treatment enhances clearance of autophagic material in the lysosome, leading to the observed reduction in lysosomal size. Western blot analysis of de-yolked zebrafish tissue revealed a significant accumulation of P62 in tpp1-/- zebrafish compared to WT siblings at 4 dpf (treated with DMSO), which was significantly reduced by 48 hr pregnenolone treatment (Fig 2C). Western blot analysis showed pregnenolone had no effect on the accumulation of SCMAS, the major component of storage

in CLN2 disease (Fig 5A). RNA sequencing analysis of p62 shows that pregnenolone does not alter mRNA quantity at 3 dpf following 24 hr treatment, suggesting the reduction in P62 protein accumulation is indicative of enhanced clearance in tpp1-/- zebrafish (Supplementary Fig 1A). Similarly, mRNA quantity of three different SCMAS loci, atp5mc1, atp5mc3a and atp5mc3b, is also unaffected by pregnenolone treatment (Supplementary Fig 1B), indicating that the lack of change in SCMAS protein represents no change in clearance. Reduction in the accumulation of P62 protein in tpp1-/- zebrafish treated with pregnenolone compared with DMSO suggests enhanced autophagic clearance. To test this hypothesis further, we generated transgenic zebrafish with ZsGreen -tagged autophagosome marker LC3, again under the muscle-specific actc1b promoter in the background of tpp1-/- zebrafish. Autophagic flux was measured in zebrafish by live imaging of ZsGreen -LC3 with and without a 4 hr treatment of NH4Cl to block lysosomal degradation. At 3 dpf, tpp1-/- zebrafish did not yet show altered ZsGreen-LC3 flux compared with WT. However, preliminary results from one .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint experiment showed that both WT and tpp1-/- zebrafish revealed a significant accumulation of ZsGreen-LC3 following 24 hr pregnenolone treatment in the presence of NH4Cl, compared to the DMSO + NH 4Cl control ( Supplementary Fig 2 A,B). These results, in combination with the reduction of P62 protein accumulation, suggest pregnenolone treatment induces autophagy in both WT and tpp1-/- zebrafish. To test whether the induction of autophagy by pregnenolone is responsible for the alleviation of seizures in tpp1-/- zebrafish, we performed a locomotion assay on the zebrafish with pregnenolone treatment in combination with autophagy induction inhibitor, 3-MA.35 Results from 3 dpf zebrafish following 24 hr concurrent treatment of pregnenolone and 3 -MA show duration of movement is comparable to pregnenolone treatment only (Fig 2 D,E). This finding therefore suggests that induction of autophagy is not the mechanism of action of pregnenolone’s anti-seizure effect. Pregnenolone acts via an mTORC1/TFEB-independent mechanism Lysosomal biogenesis and function are largely regulated by mammalian target of rapamycin 1 (mTORC1) and transcription factor EB (TFEB ), which respond to nutrient queues at the surface of the lysosome to regulate lysosomal gene expression and autophagy .36,37 Given the effect on lysosomes in both WT and tpp1-/- zebrafish, we set out to test whether pregnenolone treatment acts via an mTORC1/TFEB -dependent mechanism. Western blot analysis of mTORC1 metabolite S6, revealed significantly increased phosphorylation of S6 in tpp1-/- zebrafish compared with WT, however pregnenolone had no effect on this result (Fig 3A). Furthermore, measurement of nuclear translocation of ZsGreen -TFEB using 3 dpf ZsGreen- TFEB;mCherry-nls transgenic zebrafish , as has been published previously ,16 suggested suppression of TFEB nuclear translocation (Fig 3 B,C). 24 hr treatment with pregnenolone showed no effect in either WT or tpp1-/- zebrafish. These results highlight mTORC1/TFEB signalling abnormalities in tpp1-/- zebrafish, however pregnenolone functions independently of this pathway. Enrichment analysis of a lysosomal gene set (Supplementary Table 2) supported these results, showing upregulation of lysosomal genes in tpp1-/- zebrafish compared with WT, which was not further altered by pregnenolone treatment (Fig 3 D,E). Perturbance in cholesterol and hormone metabolism in tpp1-/- zebrafish Pregnenolone is a direct metabolite of cholesterol ; the conversion of cholesterol to pregnenolone precedes the production of all steroid hormones and is a rate -limiting step in steroidogenesis.38 In this regard, we analysed the endogenous concentrations of pregnenolone .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint and a number of downstream steroid hormones in WT and tpp1-/- zebrafish, and assessed the effect of pregnenolone treatment on these. Hormone analysis by mass spectrometry was conducted in 3 dpf zebrafish following 24 hr treatment with pregnenolone or DMSO control.

showed dysregulation of steroid hormones in tpp1-/- zebrafish compared with WT siblings in control conditions (Fig 4A and supplementary Fig 3), including significantly higher levels of pregnenolone (Fig 4B). Four out of the twenty -two measured hormones – 18-OH- corticosterone, cortisol, cortisone and 20α-DH-cortisol – were decreased in tpp1-/- zebrafish compared with WT siblings under DMSO conditions and were rescued by pregnenolone treatment (Fig 4B and supplementary Fig 4). Seven steroids were significantly increased in tpp1-/- zebrafish compared with WT siblings: and eleven were unchanged (Supplementary Fig 3). The changes in steroid hormone levels are summarised in Fig 5. To further investigate the cholesterol and steroid pathway we performed enrichment analysis of a custom set of genes related to this pathway (Supplementary Table 3). The RNA Seq data from the zebrafish revealed significant enrichment of the cholesterol and steroid pathway in tpp1-/- zebrafish compared with WT siblings in both head and tail tissue (Fig 6A). Pregnenolone treatment stimulated this pathway in WT zebrafish (Fig 6B) and lead to further enrichment of the pathway in tpp1-/- zebrafish (Fig 6C). Looking at individual genes involved in the various steps of cholesterol metabolism and hormone biosynthesis,39 tpp1-/- zebrafish show significant changes in several major genes with diverse roles (Supplementary Fig 4, Supplementary Tables 4 and 5). These changes are summarised in Fig 7, which highlights the diverse functions within the cholesterol metabolism pathway of the proteins encoded by the mRNAs with altered expression in tpp1-/- zebrafish and tpp1-/- zebrafish treated with pregnenolone. Three genes are of particular interest because their mRNA levels are increased in tpp1-/- but brought nearer to normal levels by treatment with pregnenolone. These are lcat, stard3 and npc2, encoding Lecithin-cholesterol acyltransferase (LCAT), StAR-Related Lipid Transfer Domain Containing 3 (StARD3) and Niemann-Pick C2 (NPC2). StARD3 resides on the late endosome/lysosome membrane and NPC2 is an internal late endosome/lysosome protein, but both are both important for the transfer of cholesterol from late endosomes/lysosomes to other subcellular membrane compartments such as the plasma membrane, ER and mitochondria.40,41 LCAT, however, is found in the ER where it converts free cholesterol to cholesterol esters for storage in lipid droplets.42

.CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Seizures are a major symptom of CLN2 disease that worsen with disease progression and present a challenge for management of the condition as they often become resistant to anti - epileptic drugs.9,10 The present study demonstrates the use of the seizure-like locomotion assay in the tpp1-/- zebrafish model of CLN2 disease to identify potential candidate compounds to alleviate seizures. Our screen identified pregnenolone as a hit compound, shown here to elicit an anti-seizure effect in tpp1-/- zebrafish. Furthermore, reduced cell death in the CNS of tpp1-/- zebrafish treated with pregnenolone reveal a neuroprotective effect. The results here show a partial rescue of lysosomal impairment in tpp1-/- zebrafish treated with pregnenolone. This is evident from improved morphology of the lysosome in muscle cells of the zebrafish and a decrease in P62 accumulation following treatment of pregnenolone. These results present pregnenolone as a powerful candidate for further testing as a novel therapeutic for CLN2 disease. Pregnenolone is an FDA -approved compound and has been shown to be safely administered in adults for other conditions e.g.,43,44 and is now being trialled in children 14 years or older ,45 thus pregnenolone could be considered for orphan designation for CLN2 disease.46 In line with our results, pregnenolone was recently identified in another screen, in ARPE -19 CLN3 knockout cells, where it was shown to reduce storage accumulation in the lysosomes.27 Our preliminary LC3 flux results suggest storage reduction is associated with induction of autophagy by pregnenolone. Although we do not see a reduction in SCMAS protein with pregnenolone treatment in the tpp1-/- zebrafish, the decrease in lysosomal size and reduction of P62 protein accumulation suggest an overall reduction in storage accumulation. SCMAS is believed to be a direct substrate of TPP1/Tpp14,47-49 and Tpp1 is almost entirely absent in tpp1- /- zebrafish14; thus, the persistent accumulation of SCMAS despite improvements in lysosomal function is unsurprising. This observation raises the question of whether autophagy induction, by pregnenolone or other means, would provide a long -term therapeutic effect, or if the lysosomes would quickly be overwhelmed with SCMAS accumulation, despite a general improved clearance of other storage material. Autophagy induction has been shown to have significant therapeutic effects in other models of LSDs caused by specific enzyme deficiency. In a Drosophila model of Gaucher disease lacking the glucocerebrosidase enzyme, for example, induction of autophagy by means of mTORC1 inhibition by rapamycin treatment was shown to significantly increase lifespan and improve locomotor and oxidative stress phenotypes.50 Extended lifespan and phenotypic improvement by autophagy induction in the Gaucher disease model indicates a sustained therapeutic effect .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint that can lead to significant improvements in disease outcome. Despite these benefits associated with autophagy induction, inhibiting autophagy with 3 -MA in tpp1-/- zebrafish here did not affect pregnenolone’s anti -seizure effect, indicating that autophagy is unlikely to be the mechanism of action responsible for this particular therapeutic effect. Further confirmation is needed to ensure that 3 -MA blocks pregnenolone-induced autophagy, although the treatment used clearly blocks autophagy induction in zebrafish.18 Pregnenolone might act via other mechanisms to induce the anti -seizure effect. Potential mechanisms include modulation of GABA A receptors by downstream metabolites of pregnenolone51,52 and/or agonistic effects on the sigma -1 receptor, which has been shown to elicit neuroprotective effects .53-55 Neurosteroids are known for their role in modulation of neuronal excitability which has important implications in seizure susceptibility and control. Allopregnanolone, progesterone, androsterone and 11 -deoxycorticosterone are all known to enhance GABAA receptor function, eliciting anti-convulsant actions.52,56 Several studies have demonstrated the anti -seizure effect of allopregnanolone, and Ganaxolone, an analogue of allopregnanolone, has shown promising results in clinical trials for a number of epilepsies.51,52 These findings present an alternative mechanism for the seizure rescue observed in tpp1-/- zebrafish treated with pregnenolone, as raised pregnenolone levels might lead to increases in anti-convulsant hormones. Consistent with this hypothesis, we see an increase in the concentration of a number of hormones following pregnenolone treatment, including anti - convulsant steroids, progesterone and 11 -deoxycorticosterone. GABAA activation has also been shown to have neuroprotective effects,57-59 thus, action through GABAA signaling could also be responsible for the attenuation of cell death in tpp1-/- zebrafish treated with pregnenolone. Pregnenolone and its downstream metabolite DHEA, as well as their sulfated forms, act as sigma-1 receptor agonists .54,55 The sigma -1 receptor is an anti -epileptic target. Positive modulators of the receptor, fenfluramine and E1R, have been shown to elicit anti -convulsant effects.60,61 Fenfluramine has demonstrated substantial efficacy in reducing seizures in a phase 3 clinical trial for patients with Dravet syndrome, a refractory paediatric epilepsy .62 This finding suggests that actions through the sigma -1 receptor could therefore underlie the anti - seizure effect in tpp1-/- zebrafish treated with pregnenolone. The sigma-1 receptor is widely expressed in different tissues, including brain and muscle . It binds cholesterol and regulates a variety of protein binding partners in a variety of .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint membranes.26,63 Sigma-1 receptor localisation to mitochondrial-associated membrane (MAM) microdomains of the ER is most studied. Here it regulates calcium flux into the mitochondria via the inositol trisphosphate receptor 3 (InsP 3R3) calcium channel.26,64 Calcium regulation is involved in various cell death pathways and sigma -1 agonists have been shown to attenuate neuronal death in a number of models of neurodegenerative diseases .26,65 In a cell model of Huntington’s disease for example, the sigma -1 receptor agonist PRE-084 (2 -(4- morpholinoethyl)-1-phenylcyclohexane1-carboxylate hydrochloride) increased antioxidants and reduced reactive oxygen species (ROS).66 In line with a role in regulating oxidative stress, knockdown of the sigma-1 receptor has been shown to increase ROS production, likely through dysregulation of calcium homeostasis.67-69 As increased levels of ROS have been documented in CLN2 disease patient cells ,32,70 and oxidative stress can lead to cell death ,71 pregnenolone might ameliorate cell death in tpp1-/- zebrafish by alleviating oxidative stress through its action on the sigma -1 receptor. At the plasma membrane, the sigma-1 receptor decreases voltage - gated sodium and calcium channel activity whilst increasing activity of some voltage-gated potassium channels, thereby limiting excitotoxicity.26 This relationship provides an additional possible mechanism of action of pregnenolone in limiting both seizures and cell death in tpp1- /- zebrafish. Identification of pregnenolone as a hit compound led us to investigate the cholesterol and steroid hormone pathway in our zebrafish model because 1) pregnenolone is a steroid that is synthesised from cholesterol in the mitochondria, and 2) cholesterol is bound by the sigma-1 receptor, for which pregnenolone is an agonist . This investigation led to findings of dysregulation throughout the cholesterol and steroid hormone pathways that ha ve not been previously noted in CLN2 disease. We found increased endogenous pregnenolone in tpp1-/- zebrafish. As pregnenolone reduces seizures in those zebrafish, pregnenolone is the precursor of many steroids, and pregnenolone and those steroids can have anti -epileptic effects (see earlier), this increase in endogenous pregnenolone could be a compensatory mechanism. On the other hand, g iven that pregnenolone is a metabolite of cholesterol, elevated levels of endogenous pregnenolone in tpp1-/- zebrafish could result in depleted cholesterol stores , and pregnenolone treatment may reduce cholesterol depletion in addition to negating the need for the proposed compensatory increase in pregnenolone . In tpp1-/- zebrafish, we also see upregulation of npc2, stard3 and lcat mRNAs and their restoration to normal mRNA levels by pregnenolone treatment. The effect on cholesterol storage and sub -cellular levels of free cholesterol resulting from these changes is difficult to predict, so further experiments will be .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint needed. However, it is very likely that cholesterol handling is affected. Cholesterol is crucial for membrane organisation, neuronal differentiation and myelination .72,73 Furthermore, cholesterol is important for the integrity of the lysosomal membrane and modulates ion permeability.74 Disruption to the cholesterol pathway could contribute to CLN2 disease pathology via disturbance of its function in any one of these roles , highlighting cholesterol abnormality as an important area for further research in tpp1-/- zebrafish and other models of CLN2 disease. In summary, the present study employed a zebrafish model of CLN2 disease to perform a medium-throughput drug screen which identified pregnenolone as a potential therapeutic for the disease. Pregnenolone was shown to reduce seizures in our model and improve cell pathology, including improvement in lysosome morphology and function and reduced cell death. The safety profile of pregnenolone make it an appealing candidate for further testing into the use of it as a therapeutic for CLN2 disease and perhaps other NCLs.43,44 The novel Lamp1-ZsGreen transgenic zebrafish presented here revealed a striking phenotype in tpp1-/- zebrafish compared to healthy WT siblings. This transgenic line provide s a powerful tool to investigate potential therapeutics that can improve lysosomal biology in vivo, as demonstrated by the effect of pregnenolone. The use of the act1b muscle promoter provides a substantial area of tissue that can be automatically detected and imaged using a high-content confocal imaging system. These lines can therefore be used as the basis of a screen similar to the one presented in the current study, through automa ted imaging in a 96 well format. There are a number of examples of similar high -content screening studies using other transgenic zebrafish lines as effective platforms for drug screening.75-77 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Data availability statement: RNAseq and metabolomic datasets will be uploaded to. Other data will be provided by the corresponding author on request.

We thank Andrew Hibbert at the Royal Veterinary College for confocal imaging assistance and Yu Mei (Ruby) Chang for advice on statistical analysis. Funding: LK was supported by the Royal Veterinary College and the London Interdisciplinary Doctoral Training Programme (LIDo) with funding from Nestlé Research and the BBSRC (Grant BB/M009513/1). FM was supported by the Royal Veterinary College with funding from the BBSRC. The Royal Veterinary College supported EY and HY through undergraduate project consumable provision. The Royal Veterinary College supported LM and PE through hosting their postgraduate projects with consumable provision from King’s College London. VB and DN were supported by a grant from the Batten Disease Family Association, who also funded the purchase of the DanioVision equipment. A grant from SPARKS to CR and MC (13RVC01) funded some consumables. The activities in MC labs were sustained by the European Research Council Consolidator Grant COG2018- 819600_FIRM, Biotechnology and Biological Sciences Research Council grants, the Italian Association for Cancer Research grant MFAG21903 and the Fondation ARC Scheme for International Leaders. CR, AZ and MC were HEFCE-funded. GC, GL and PG were employed by Nestlé Research. Competing interests: Authors have no competing interests.

1. Nita DA, Mole SE, Minassian BA. Neuronal ceroid lipofuscinoses. Epileptic Disord . Sep 1 2016;18(S2):73-88. doi:10.1684/epd.2016.0844 2. Gardner E, Mole SE. The Genetic Basis of Phenotypic Heterogeneity in the Neuronal Ceroid Lipofuscinoses. Front Neurol. 2021;12:754045. doi:10.3389/fneur.2021.754045 3. Carcel-Trullols J, Kovacs AD, Pearce DA. Cell biology of the NCL proteins: What they do and don't do. Biochim Biophys Acta. Oct 2015;1852(10 Pt B):2242-55. doi:10.1016/j.bbadis.2015.04.027 4. Palmer DN, Fearnley IM, Walker JE, et al. Mitochondrial ATP synthase subunit c storage in the ceroid-lipofuscinoses (Batten disease). Am J Med Genet . Feb 15 1992;42(4):561 -7. doi:10.1002/ajmg.1320420428 5. Sleat DE, Donnelly RJ, Lackland H, et al. Association of mutations in a lysosomal protein with classical late -infantile neuronal ceroid lipofuscinosis. Science. Sep 19 1997;277(5333):1802 -5. doi:10.1126/science.277.5333.1802 6. Liu CG, Sleat DE, Donnelly RJ, Lobel P. Structural organization and sequence of CLN2, the defective gene in classical late infantile neuronal ceroid lipofuscinosis. Genomics. Jun 1 1998;50(2):206-12. doi:10.1006/geno.1998.5328 7. Dyke JP, Sondhi D, Voss HU, et al. Brain Region-Specific Degeneration with Disease Progression in Late Infantile Neuronal Ceroid Lipofuscinosis (CLN2 Disease). AJNR Am J Neuroradiol . Jun 2016;37(6):1160-9. doi:10.3174/ajnr.A4669 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint 8. Specchio N, Pietrafusa N, Trivisano M. Changing Times for CLN2 Disease: The Era of Enzyme Replacement Therapy. Ther Clin Risk Manag. 2020;16:213-222. doi:10.2147/TCRM.S241048 9. Williams RE, Adams HR, Blohm M, et al. Management Strategies for CLN2 Disease. Pediatr Neurol. Apr 2017;69:102-112. doi:10.1016/j.pediatrneurol.2017.01.034 10. Mole SE, Schulz A, Badoe E, et al. Guidelines on the diagnosis, clinical assessments, treatment and management for CLN2 disease patients. Orphanet J Rare Dis . Apr 21 2021;16(1):185. doi:10.1186/s13023-021-01813-5 11. Kohlschutter A, Schulz A, Bartsch U, Storch S. Current and Emerging Treatment Strategies for Neuronal Ceroid Lipofuscinoses. CNS Drugs. Apr 2019;33(4):315-325. doi:10.1007/s40263-019-00620- 8 12. Nelvagal HR, Lange J, Takahashi K, Tarczyluk -Wells MA, Cooper JD. Pathomechanisms in the neuronal ceroid lipofuscinoses. Biochim Biophys Acta Mol Basis Dis . Sep 1 2020;1866(9):165570. doi:10.1016/j.bbadis.2019.165570 13. Dulz S, Schwering C, Wildner J, et al. Ongoing retinal degeneration despite intraventricular enzyme replacement therapy with cerliponase alfa in late-infantile neuronal ceroid lipofuscinosis type 2 (CLN2 disease). Br J Ophthalmol. Jun 30 2022;doi:10.1136/bjo-2022-321260 14. Mahmood F, Fu S, Cooke J, Wilson SW, Cooper JD, Russell C. A zebrafish model of CLN2 disease is deficient in tripeptidyl peptidase 1 and displays progressive neurodegeneration accompanied by a reduction in proliferation. Brain. May 2013;136(Pt 5):1488-507. doi:10.1093/brain/awt043 15. Westerfield M. The Zebrafish Book. University of Oregon Press; 2007. 16. Collodet C, Foretz M, Deak M, et al. AMPK promotes induction of the tumor suppressor FLCN through activation of TFEB independently of mTOR. FASEB J . Nov 2019;33(11):12374 -12391. doi:10.1096/fj.201900841R 17. Thermes V, Grabher C, Ristoratore F, et al. I -SceI meganuclease mediates highly efficient transgenesis in fish. Mech Dev. Oct 2002;118(1-2):91-8. doi:10.1016/s0925-4773(02)00218-6 18. He C, Bartholomew CR, Zhou W, Klionsky DJ. Assaying autophagic activity in transgenic GFP - Lc3 and GFP -Gabarap zebrafish embryos. Autophagy. May 2009;5(4):520 -6. doi:10.4161/auto.5.4.7768 19. Zdebik AA, Mahmood F, Stanescu HC, Kleta R, Bockenhauer D, Russell C. Epilepsy in kcnj10 morphant zebrafish assessed with a novel method for long -term EEG recordings. PLoS One . 2013;8(11):e79765. doi:10.1371/journal.pone.0079765 20. Gavrieli Y, Sherman Y, Ben -Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol . Nov 1992;119(3):493 -501. doi:10.1083/jcb.119.3.493 21. Anders S, Pyl PT, Huber W. HTSeq --a Python framework to work with high -throughput sequencing data. Bioinformatics. Jan 15 2015;31(2):166-9. doi:10.1093/bioinformatics/btu638 22. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. Jan 1 2010;26(1):139 -40. doi:10.1093/bioinformatics/btp616 23. Wu D, Smyth GK. Camera: a competitive gene set test accounting for inter -gene correlation. Nucleic Acids Res. Sep 1 2012;40(17):e133. doi:10.1093/nar/gks461 24. Guennoun R, Labombarda F, Gonzalez Deniselle MC, Liere P, De Nicola AF, Schumacher M. Progesterone and allopregnanolone in the central nervous system: response to injury and implication for neuroprotection. J Steroid Biochem Mol Biol . Feb 2015;146:48 -61. doi:10.1016/j.jsbmb.2014.09.001 25. Hu J, Zhang Z, Shen WJ, Azhar S. Cellular cholesterol delivery, intracellular processing and utilization for biosynthesis of steroid hormones. Nutr Metab (Lond) . Jun 1 2010;7:47. doi:10.1186/1743-7075-7-47 26. Ryskamp DA, Korban S, Zhemkov V, Kraskovskaya N, Bezprozvanny I. Neuronal Sigma -1 Receptors: Signaling Functions and Protective Roles in Neurodegenerative Diseases. Front Neurosci. 2019;13:862. doi:10.3389/fnins.2019.00862 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint 27. Soldati C, Lopez-Fabuel I, Wanderlingh LG, et al. Repurposing of tamoxifen ameliorates CLN3 and CLN7 disease phenotype. EMBO Mol Med . Oct 7 2021;13(10):e13742. doi:10.15252/emmm.202013742 28. Kim WD, Wilson-Smillie M, Thanabalasingam A, Lefrancois S, Cotman SL, Huber RJ. Autophagy in the Neuronal Ceroid Lipofuscinoses (Batten Disease). Front Cell Dev Biol . 2022;10:812728. doi:10.3389/fcell.2022.812728 29. Lieberman AP, Puertollano R, Raben N, Slaugenhaupt S, Walkley SU, Ballabio A. Autophagy in lysosomal storage disorders. Autophagy. May 1 2012;8(5):719-30. doi:10.4161/auto.19469 30. Seranova E, Connolly KJ, Zatyka M, et al. Dysregulation of autophagy as a common mechanism in lysosomal storage diseases. Essays Biochem . Dec 12 2017;61(6):733 -749. doi:10.1042/EBC20170055 31. Settembre C, Fraldi A, Rubinsztein DC, Ballabio A. Lysosomal storage diseases as disorders of autophagy. Autophagy. Jan 2008;4(1):113-4. doi:10.4161/auto.5227 32. Vidal-Donet JM, Carcel-Trullols J, Casanova B, Aguado C, Knecht E. Alterations in ROS activity and lysosomal pH account for distinct patterns of macroautophagy in LINCL and JNCL fibroblasts. PLoS One. 2013;8(2):e55526. doi:10.1371/journal.pone.0055526 33. Djajadikerta A, Keshri S, Pavel M, Prestil R, Ryan L, Rubinsztein DC. Autophagy Induction as a Therapeutic Strategy for Neurodegenerative Diseases. J Mol Biol . Apr 3 2020;432(8):2799 -2821. doi:10.1016/j.jmb.2019.12.035 34. Sarkar S. Regulation of autophagy by mTOR -dependent and mTOR -independent pathways: autophagy dysfunction in neurodegenerative diseases and therapeutic application of autophagy enhancers. Biochem Soc Trans. Oct 2013;41(5):1103-30. doi:10.1042/BST20130134 35. Wu YT, Tan HL, Shui G, et al. Dual role of 3 -methyladenine in modulation of autophagy via different temporal patterns of inhibition on class I and III phosphoinositide 3-kinase. J Biol Chem. Apr 2 2010;285(14):10850-61. doi:10.1074/jbc.M109.080796 36. Sancak Y, Bar -Peled L, Zoncu R, Markhard AL, Nada S, Sabatini DM. Ragulator -Rag complex targets mTORC1 to the lysosomal surface and is necessary for its activation by amino acids. Cell. Apr 16 2010;141(2):290-303. doi:10.1016/j.cell.2010.02.024 37. Sancak Y, Peterson TR, Shaul YD, et al. The Rag GTPases bind raptor and mediate amino acid signaling to mTORC1. Science. Jun 13 2008;320(5882):1496-501. doi:10.1126/science.1157535 38. Miller WL, Auchus RJ. The molecular biology, biochemistry, and physiology of human steroidogenesis and its disorders. Endocr Rev. Feb 2011;32(1):81-151. doi:10.1210/er.2010-0013 39. Luo J, Yang H, Song BL. Mechanisms and regulation of cholesterol homeostasis. Nat Rev Mol Cell Biol. Apr 2020;21(4):225-245. doi:10.1038/s41580-019-0190-7 40. Meng Y, Heybrock S, Neculai D, Saftig P. Cholesterol Handling in Lysosomes and Beyond. Trends Cell Biol. Jun 2020;30(6):452-466. doi:10.1016/j.tcb.2020.02.007 41. Arenas F, Garcia-Ruiz C, Fernandez-Checa JC. Intracellular Cholesterol Trafficking and Impact in Neurodegeneration. Front Mol Neurosci. 2017;10:382. doi:10.3389/fnmol.2017.00382 42. Rousset X, Vaisman B, Amar M, Sethi AA, Remaley AT. Lecithin: cholesterol acyltransferase -- from biochemistry to role in cardiovascular disease. Curr Opin Endocrinol Diabetes Obes . Apr 2009;16(2):163-71. doi:10.1097/med.0b013e328329233b 43. Brown ES, Park J, Marx CE, et al. A randomized, double -blind, placebo -controlled trial of pregnenolone for bipolar depression. Neuropsychopharmacology. Nov 2014;39(12):2867 -73. doi:10.1038/npp.2014.138 44. Fung LK, Libove RA, Phillips J, Haddad F, Hardan AY. Brief report: an open -label study of the neurosteroid pregnenolone in adults with autism spectrum disorder. J Autism Dev Disord . Nov 2014;44(11):2971-7. doi:10.1007/s10803-014-2144-4 45. Hardan AY. Pilot Trial of Pregnenolone in Autism. NCT02627508. https://clinicaltrials.gov/study/NCT02627508 46. EMA. Orphan designation: Overview. Accessed 13/11/2021, https://www.ema.europa.eu/en/human-regulatory/overview/orphan-designation-overview .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint 47. Tian Y, Sohar I, Taylor JW, Lobel P. Determination of the substrate specificity of tripeptidyl - peptidase I using combinatorial peptide libraries and development of improved fluorogenic substrates. J Biol Chem. Mar 10 2006;281(10):6559-72. doi:10.1074/jbc.M507336200 48. Ezaki J, Takeda-Ezaki M, Kominami E. Tripeptidyl peptidase I, the late infantile neuronal ceroid lipofuscinosis gene product, initiates the lysosomal degradation of subunit c of ATP synthase. J Biochem. Sep 2000;128(3):509-16. doi:10.1093/oxfordjournals.jbchem.a022781 49. Ezaki J, Takeda -Ezaki M, Oda K, Kominami E. Characterization of endopeptidase activity of tripeptidyl peptidase -I/CLN2 protein which is deficient in classical late infantile neuronal ceroid lipofuscinosis. Biochem Biophys Res Commun. Feb 24 2000;268(3):904-8. doi:10.1006/bbrc.2000.2207 50. Kinghorn KJ, Gronke S, Castillo-Quan JI, et al. A Drosophila Model of Neuronopathic Gaucher Disease Demonstrates Lysosomal-Autophagic Defects and Altered mTOR Signalling and Is Functionally Rescued by Rapamycin. J Neurosci. Nov 16 2016;36(46):11654-11670. doi:10.1523/JNEUROSCI.4527- 15.2016 51. Lattanzi S, Riva A, Striano P. Ganaxolone treatment for epilepsy patients: from pharmacology to place in therapy. Expert Rev Neurother . Nov 2021;21(11):1317 -1332. doi:10.1080/14737175.2021.1904895 52. Miziak B, Chroscinska -Krawczyk M, Czuczwar SJ. Neurosteroids and Seizure Activity. Front Endocrinol (Lausanne). 2020;11:541802. doi:10.3389/fendo.2020.541802 53. Ruscher K, Wieloch T. The involvement of the sigma -1 receptor in neurodegeneration and neurorestoration. J Pharmacol Sci. Jan 2015;127(1):30-5. doi:10.1016/j.jphs.2014.11.011 54. Maurice T, Su TP, Privat A. Sigma1 (sigma 1) receptor agonists and neurosteroids attenuate B25-35-amyloid peptide-induced amnesia in mice through a common mechanism. Neuroscience. Mar 1998;83(2):413-28. doi:10.1016/s0306-4522(97)00405-3 55. Maurice T, Gregoire C, Espallergues J. Neuro(active)steroids actions at the neuromodulatory sigma1 (sigma1) receptor: biochemical and physiological evidences, consequences in neuroprotection. Pharmacol Biochem Behav. Aug 2006;84(4):581-97. doi:10.1016/j.pbb.2006.07.009 56. Biagini G, Panuccio G, Avoli M. Neurosteroids and epilepsy. Curr Opin Neurol . Apr 2010;23(2):170-6. doi:10.1097/WCO.0b013e32833735cf 57. Wei XW, Yan H, Xu B, Wu YP, Li C, Zhang GY. Neuroprotection of co -activation of GABA receptors by preventing caspase -3 denitrosylation in KA -induced seizures. Brain Res Bull . Sep 1 2012;88(6):617-23. doi:10.1016/j.brainresbull.2012.05.008 58. Li C, Xu B, Wang WW, et al. Coactivation of GABA receptors inhibits the JNK3 apoptotic pathway via disassembly of GluR6 -PSD-95-MLK3 signaling module in KA -induced seizure. Epilepsia. Mar 2010;51(3):391-403. doi:10.1111/j.1528-1167.2009.02270.x 59. Zhang F, Li C, Wang R, et al. Activation of GABA receptors attenuates neuronal apoptosis through inhibiting the tyrosine phosphorylation of NR2A by Src after cerebral ischemia and reperfusion. Neuroscience. Dec 19 2007;150(4):938-49. doi:10.1016/j.neuroscience.2007.09.070 60. Martin P, de Witte PAM, Maurice T, Gammaitoni A, Farfel G, Galer B. Fenfluramine acts as a positive modulator of sigma -1 receptors. Epilepsy Behav . Apr 2020;105:106989. doi:10.1016/j.yebeh.2020.106989 61. Vavers E, Svalbe B, Lauberte L, et al. The activity of selective sigma-1 receptor ligands in seizure models in vivo. Behav Brain Res. Jun 15 2017;328:13-18. doi:10.1016/j.bbr.2017.04.008 62. Bishop KI, Isquith PK, Gioia GA, et al. Improved everyday executive functioning following profound reduction in seizure frequency with fenfluramine: Analysis from a phase 3 long -term extension study in children/young adults with Dravet syndrome. Epilepsy Behav. Aug 2021;121(Pt A):108024. doi:10.1016/j.yebeh.2021.108024 63. Zhemkov V, Geva M, Hayden MR, Bezprozvanny I. Sigma -1 Receptor (S1R) Interaction with Cholesterol: Mechanisms of S1R Activation and Its Role in Neurodegenerative Diseases. Int J Mol Sci. Apr 15 2021;22(8)doi:10.3390/ijms22084082 64. Hayashi T, Su TP. Sigma -1 receptor chaperones at the ER -mitochondrion interface regulate Ca(2+) signaling and cell survival. Cell. Nov 2 2007;131(3):596-610. doi:10.1016/j.cell.2007.08.036 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint 65. Orrenius S, Gogvadze V, Zhivotovsky B. Calcium and mitochondria in the regulation of cell death. Biochem Biophys Res Commun. Apr 24 2015;460(1):72-81. doi:10.1016/j.bbrc.2015.01.137 66. Hyrskyluoto A, Pulli I, Tornqvist K, Ho TH, Korhonen L, Lindholm D. Sigma -1 receptor agonist PRE084 is protective against mutant huntingtin-induced cell degeneration: involvement of calpastatin and the NF-kappaB pathway. Cell Death Dis. May 23 2013;4:e646. doi:10.1038/cddis.2013.170 67. Tsai SY, Rothman RK, Su TP. Insights into the Sigma-1 receptor chaperone's cellular functions: a microarray report. Synapse. Jan 2012;66(1):42-51. doi:10.1002/syn.20984 68. Wang J, Shanmugam A, Markand S, Zorrilla E, Ganapathy V, Smith SB. Sigma 1 receptor regulates the oxidative stress response in primary retinal Muller glial cells via NRF2 signaling and system xc( -), the Na(+) -independent glutamate -cystine exchanger. Free Radic Biol Med . Sep 2015;86:25-36. doi:10.1016/j.freeradbiomed.2015.04.009 69. Ermak G, Davies KJ. Calcium and oxidative stress: from cell signaling to cell death. Mol Immunol. Feb 2002;38(10):713-21. doi:10.1016/s0161-5890(01)00108-0 70. Micsenyi MC, Sikora J, Stephney G, Dobrenis K, Walkley SU. Lysosomal membrane permeability stimulates protein aggregate formation in neurons of a lysosomal disease. J Neurosci. Jun 26 2013;33(26):10815-27. doi:10.1523/JNEUROSCI.0987-13.2013 71. Ryter SW, Kim HP, Hoetzel A, et al. Mechanisms of cell death in oxidative stress. Antioxid Redox Signal. Jan 2007;9(1):49-89. doi:10.1089/ars.2007.9.49 72. Ghaddar B, Lubke L, Couret D, Rastegar S, Diotel N. Cellular Mechanisms Participating in Brain Repair of Adult Zebrafish and Mammals after Injury. Cells. Feb 14 2021;10(2):391. doi:10.3390/cells10020391 73. Lu F, Ferriero DM, Jiang X. Cholesterol in Brain Development and Perinatal Brain Injury: more than a Building Block. Curr Neuropharmacol. Nov 11 2021:10.2174/1570159X19666211111122311. doi:10.2174/1570159X19666211111122311 74. Deng D, Jiang N, Hao SJ, Sun H, Zhang GJ. Loss of membrane cholesterol influences lysosomal permeability to potassium ions and protons. Biochim Biophys Acta . Feb 2009;1788(2):470 -6. doi:10.1016/j.bbamem.2008.11.018 75. Kim GJ, Mo H, Liu H, et al. A zebrafish screen reveals Renin -angiotensin system inhibitors as neuroprotective via mitochondrial restoration in dopamine neurons. Elife. Sep 22 2021;10:e69795. doi:10.7554/eLife.69795 76. Liu H, Chen S, Huang K, et al. A High-Content Larval Zebrafish Brain Imaging Method for Small Molecule Drug Discovery. PLoS One. 2016;11(10):e0164645. doi:10.1371/journal.pone.0164645 77. Westhoff JH, Steenbergen PJ, Thomas LSV, et al. In vivo High -Content Screening in Zebrafish for Developmental Nephrotoxicity of Approved Drugs. Front Cell Dev Biol . 2020;8:583. doi:10.3389/fcell.2020.00583 78. Genome.jp. Steroid hormone biosynthesis - Danio rerio (zebrafish) (PATHWAY: dre00140). Accessed 03/10/21. https://www.genome.jp/entry/pathway+dre00140 79. Miller WL, Bose HS. Early steps in steroidogenesis: intracellular cholesterol trafficking. J Lipid Res. Dec 2011;52(12):2111-35. doi:10.1194/jlr.R016675 80. Ridgway ND, Zhao K. Cholesterol transfer at endosomal -organelle membrane contact sites. Curr Opin Lipidol. Jun 2018;29(3):212-217. doi:10.1097/MOL.0000000000000506 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Figure 1: Locomotion assay drug screen in tpp1-/- zebrafish identifies pregnenolone as hit compound. (A) Drug screen design for FDA approved drug library compounds. (B) Representative locomotion assay plate. 96 -well plate containing one fish per well in 200µL aquarium water with treatment. Locomotion tracks are shown in red. White tracks are predicted movements that were not measured. (C) Locomotion assay plots from blind testing of re - sourced pregnenolone in tpp1-/- zebrafish compared to WT siblings. Pregnenolone (preg) treatment for 24 h from 48 hpf significantly improved seizure -like locomotion phenotype in tpp1-/- zebrafish, demonstrated by reduced movement duration (s), distance (mm) and mean velocity (mm/s) [Log Ln] compared to DMSO treated tpp1-/- zebrafish. Results from two replicate experiments. (D) Single representative EEG recordings of tpp1-/- zebrafish at 3 dpf treated with preg or DMSO control for 24 hr. (E) Plot showing significant reduction in mean EEG amplitude in preg treated tpp1-/- zebrafish compared to DMSO control . Results are combined from three days of experiments. (C and E) Data are shown by box and whisker plot (minimum to maximum) with individual points each representing an individual animal. (C) Linear mixed model (fixed: genotype, treatment / random: date of experiment). (E) Unpaired t-test. *P ≤ 0.033, **P ≤ 0.002, ***P ≤ 0.001. (F) Representative TUNEL stained sagittal sections from WT and tpp1-/- zebrafish treated with preg or DMSO control from 52 hpf for 3 hours, with the mesencephalon indic ated as the region within the dotted white line. ( G) Quantification of mean fluorescent intensity of TUNEL staining in the mesencephalon. Results from two replicate experiments. Data are shown by box and whisker plot (min imum to maximum) with individual points each representing an individual fish. Two-way ANOVA. *p ≤ 0.033, **p ≤ 0.002, ***p ≤ 0.001. Figure 2: Pregnenolone ameliorates l ysosomal abnormalities and dysfunction in tpp1-/- zebrafish. (A-B) ZsGreen-Lamp1 WT and tpp1-/- zebrafish at 3 dpf treated with DMSO or pregnenolone (preg) for 24 hr. (A) Representative 40x max -projection confocal images, enlarged area indicated by white box. (B) Quantification of lysosomal number (left) and size (right) from ZsGreen-Lamp1 confocal images. Results from three replicate experiments. Data are shown by box and whisker plot (min imum to max imum) with individual points each representing an individual fish. (C) Representative image of Western blot analysis from whole lysates of de-yolked non-transgenic WT and tpp1-/- zebrafish at 4 dpf treated with DMSO or preg for 48 hr , stained with antibodies against P62 normalized to HSC70 and SCMAS normalized to Tom20. Data are shown by box and whisker plot (minimum to maximum) with .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint individual points each representing an individual sample. (D,E) Seizure-like locomotion assay from 3 dpf WT and tpp1-/- mutant zebrafish treated with pregnenolone (Preg) + 3 -MA for 24 hr, with DMSO, pregnenolone, or 3 -MA alone as controls. (D) Representative locomotion assay plate. 96-well plate containing one fish per well in 200 µL aquarium water with treatment. Locomotion tracks are shown in red. White tracks are predicted movements that were not measured. (E) Plot showing duration of movement (s). No si gnificant difference between pregnenolone with and without 3 -MA, in tpp1-/- zebrafish. Results from three replicate experiments. Data are shown by box and whisker plot (min to max) with individual points each representing an individual fish. (B, C) Linear mixed model (fixed: genotype, treatment / random: date of experiment/plate). (E) Linear mixed model on log transformed data (fixed: genotype, treatment / random: date of experiment). P* <0.05, P**<0.01, P*** <0.001. Figure 3: Lysosomal abnormalities and dysfunction in tpp1-/- zebrafish are ameliorated by pregnenolone. (A) Representative image of Western blot analysis from whole lysates of de-yolked non -transgenic WT and tpp1-/- zebrafish at 3 dpf treated with DMSO or pregnenolone (preg) for 24 hr , stained with antibodies against S6 and phosphorylated(p) -S6 normalized to HSC70. Data are shown as mean ± SD with individual points each representing an individual sample. (B,C) ZsGreen-TFEB;mCherry-nls WT and tpp1-/- zebrafish at 3 dpf treated with DMSO or preg for 24 hr. (B) Representative cropped 40x single -plane images showing distribution of ZsGreen -TFEB, alongside mCherry -nls nuclear marker. ZsGreen - TFEB/mCherry-nls merge shown with corresponding ZsGreen -TFEB-only image below. Examples of nucleus indicated by yellow arrow. (C) Quantification of ZsGreen-TFEB intensity within nucleus over intensity of total area of muscle measur ed. (A,C) Linear mixed model (fixed: genotype, treatment / random: date of experiment/plate). P* <0.05, P**<0.01, P*** <0.001. (D,E) RNA Sequencing data of a custom lysosomal gene set analysed in head (left) and tail (right) tissue of 3 dpf (D) tpp1-/- vs WT zebrafish (24 hr DMSO treatment) and (E) tpp1-/- zebrafish with preg vs DMSO treatment (treated for 24 hr). Results from 6 replicate samples, n=10 fish per sample. Barcode plots from CAMERA gene set enrichment analysis. Each bar represents a single gene; red and blue sections are representations of the portion of genes that most contribute to up - and down -regulated enrichment respectively. The curves show the enrichment score of the bars in the barcode plot; parts above and below the dashed line signify up - and down -regulated enrichment respectively. CAMERA p value result and highest enriched direction indicated on each plot. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Figure 4: Steroid hormone biosynthesis is dysregulated in tpp1-/- zebrafish and boosted by pregnenolone treatment. Steroid hormone measurements from 3 dpf WT and tpp1-/- zebrafish treated with pregnenolone (preg) or DMSO for 24 hr. Results from 6 replicate samples, n=50 fish per sample. (A) Heatmap representing hormone measurements. Each block represents a single hormone from an individual sample. (B) Individual hormone measurement plots show picograms (pg) of hormones per gram of fish tissue. Data are show n by box and whisker plot (minimum to maximum) with individual points each representing an individual sample. Linear mixed model (fixed: genotype, treatment / random: date of experiment). p* <0.05, p**<0.01, p*** <0.001. Figure 5: Overview of changes in steroid hormone levels in tpp1-/- zebrafish compared with WT siblings. An overview of the steroid hormone synthesis pathway with blue and red circles representing hormones that were significantly increased or decreased respectively in tpp1-/- zebrafish compared with WT siblings (treated with DMSO for 24 hr). Grey circles represent hormones that were unchanged. Hormones that are not circled were not measured. Hormones were measured by mass spectrometry in 3 dpf zebrafish (adapted from 78). Figure 6: Cholesterol and hormone related genes are enriched in tpp1-/- zebrafish and further boosted by pregnenolone treatment. RNA Sequencing data from head (left) and tail (right) tissue of 3 dpf zebrafish treated with DMSO or pregnenolone (Preg) for 24 hr. Results from 6 replicate samples, n=10 fish per sample. (A-C) Barcode plots from CAMERA gene set enrichment analysis. Each bar represents a single gene; red and blue sections are representations of the portion of genes that most contribute to up - and down -regulated enrichment respectively. The curves show the enrichment score of the bars in the barcode plot; parts above and below the dashed line signify up- and down-regulated enrichment respectively. CAMERA p value result and highest enriched direction indicated on each plot. (A) tpp1-/- vs WT zebrafish (B) pregnenolone vs DMSO treated WT zebrafish (C) pregnenolone vs DMSO tpp1-/- zebrafish. Figure 7: Overview of major pathways of cholesterol metabolism in the cell. Cholesterol availability in the cell depends predominantly on uptake of circulating LDL by the LDL - Receptor (LDLR) and de novo biosynthesis from acetyl -coA, mediated by the rate -limiting enzymes 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) and s qualene mono- oxygenase (SM). Cholesterol biosynthesis is regulated by the transcription factor SREBF2. Excess cholesterol is exported to the blood through ABCA1 or ABCG1. Imported or stored .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint cholesterol esters are unesterified in the lysosome and free cholesterol can be distributed to other organelles in the cell via membrane contact sites, mediated by NPC1 and NPC2, and StARD3. Free cholesterol can be delivered to the inner mitochondrial membrane, mediated by StARD1, for use in hormone biosynthesis. Excess cholesterol can be esterified by ACAT or LCAT for storage in lipid droplets or excretion as HDL (adapted from39,40,79,80). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint A WT - DMSO tpp1-/- - DMSO WT - Preg tpp1-/- - Preg WT - DMSO tpp1-/- - DMSO WT - Preg tpp1-/- - Preg B DMSO Preg DMSO Preg 0 50 100 150 700 800 900 1000Movement duration (s) WT tpp1-/- ** **ns ns DMSO Preg DMSO Preg 0 500 1000 1500 2000 6000 7000 8000Distance (mm) WT tpp1-/- *** ***ns ns DMSO Preg DMSO Preg e-6 e-5 e-4 e-3 e-2 e-1 e0 e1 Velocity (mm/s) WT tpp1-/- *** ***ns ns Duration of movement Distance moved Mean velocityC ` tpp1-/- + Preg mV 0.01 mV 5 s WT tpp1-/- + DMSO D E DMSO Preg DMSO Preg 0 20 40 60 80 100Mean Intensity (AU) WT tpp1-/- ** ** TUNEL stain mean intensity F WT tpp1-/- DMSO Preg G Figure 1 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint WT DMSO Preg DMSO Preg DMSO Preg 0.00 0.02 0.04 0.06 0.08 0.10No. of lysosomes / area (AU) *** *** *** WT tpp1-/- Lysosome number DMSO Preg DMSO Preg 0 1 2 3 4 5Lysosome size (mm2) * *** *** WT tpp1-/- Lysosome size A B tpp1-/- DMSO Preg Lamp1_ZsGreen P62 SCMAS C Enlarged P62 HSC70 SCMAS TOM20 WT tpp1-/- DMSO Preg DMSO Preg Figure 2 D E .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint DMSO Preg DMSO HSC70 S6 WT tpp1-/- P-S6 Preg DMSO Preg DMSO Preg 0.0 0.5 1.0 1.5p-S6/S6 protein (AU) P-S6/S6 WT tpp1-/- **ns ns A B C Head Tail Head Tailtpp1-/- preg vs DMSO D E No. of genes: 107 Direction: Up p = 2.20E-05 No. of genes: 107 Direction: Up P = 0.01447 No. of genes: 107 Direction: Up p = 0.72 No. of genes: 107 Direction: Up p = 0.79 tpp1-/- vs WT (DMSO) Figure 3 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint DMSO Preg DMSO Preg 0 2000 4000 20000 40000 60000 pg / g Cortisol ****** ** ns WT tpp1-/- DMSO Preg DMSO Preg 0 200 400 600 2000 4000 6000 8000 10000 pg / g Cortisone ****** *** ns WT tpp1-/- DMSO Preg DMSO Preg 0 2000 4000 6000 8000pg / g 18-OH-corticosterone ** *** ns ns WT tpp1-/- DMSO Preg DMSO Preg 0.0 5.0×107 1.0×108 1.5×108 2.0×108 pg / g Pregnenolone WT tpp1-/- **** * ns 0 Pregnenolone Progesterone X17...hydroxy.pregnenolone X17.hydroxy..Progesterone X21.Deoxycortisol X11.Deoxycorticosterone X11.Deoxycortisol Cortisol Tetrahydroxy.11.deoxycortisol X18.Hydroxycortisol X20...Dihydrocortisol Cortisone Urocortisone Corticosterone X18.hydroxy.corticosterone Tetrahydroxy.aldosterone DHEA DHEAS Androstenedione Testosterone DHT.5...Dihydrotestosterone. Estrone 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 18 Hormone quantity (AU) tpp1-/- DMSO Preg WT DMSO Preg A B Figure 4 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Urocortisone 20α-DH-cortisolCortisol 18-OH- corticosterone Pregnenolone Progesterone 17α-OH- progesterone 11-Deoxy- corticosterone Corticosterone TH-aldosterone Cortisone Cholesterol 5α-DH- progesterone Allopregnanolone DHEA17α-OH- Pregnenolone Androstenedione 5α- Androstanedione Androsterone Testosterone 5α-DH- testosterone Estrone 17β-Estradiol 16α-OH-estrone Estriol 11β,21-DH-5β- pregnane-3,20-dione TH-corticosterone Aldosterone 21-Deoxycortisol 11-Deoxycortisol 18-OH-cortisol OH = Hydroxy DH = Dihydro TH = Tetrahydroxy DHEA = Dehydroepiandrosterone S = Sulfate Pregnenolone-S DHEA-S CYP11A1 HSD3B1/2 STS SULT2B1 CYP17A1 CYP17A1 STS SULT2B1 AKR1C3 HSD17B2/3/6/8 HSD3B1/2 CYP19A1 CYP1A1 HSD17B1/2/6/7/8/12 AKR1C3 CYP3A4/5 HSD17B1/2/6/7/8/12 SRD5A1/2/3 21-OH- pregnenolone CYP21A2 HSD3B1/2 CYP21A2 SRD5A1/2/3 AKR1C2/3 SRD5A1/2/3 AKR1C4 CYP21B1/2 CYP11B2 CYP11B2 AKR1D1 AKR1C4AKR1C4 HSD11B1/1LHSD11B2 17α,21-DH-5β-pregnane- 3,11,20-trione AKR1C4 AKR1D1 TH-11- deoxycortisol CYP21A2 CYP11B1/2 CYP17A1 CYP11B2 CYP11B1 HSD3B1/2 Gene encoding enzyme Hormone Bold = measured here in zebrafish KEY Increased Decreased in tpp1-/- zebrafish compared with WT siblings Unchanged = Rescued by pregnenolone treatment Figure 5 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Head Tail WT Preg vs DMSO tpp1-/- Preg vs DMSO A tpp1-/- vs WT (DMSO) B C No. of genes: 103 Direction: Up p = 0.005 No. of genes: 103 Direction: Up p = 0.0083 No. of genes: 103 Direction: Up p = 0.00053 No. of genes: 103 Direction: Up p = 0.01 No. of genes: 103 Direction: Up p = 0.0064 No. of genes: 103 Direction: Up p = 0.00014 Head Tail Head Tail Figure 6 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint Up-regulated gene in tpp1-/- mutant zebrafish vs WT siblings (DMSO) Up-regulated gene in tpp1-/- mutant zebrafish vs WT siblings (DMSO), alleviated with pregnenolone treatment KEY LD = Lipid droplet Chol = Cholesterol CE = Cholesterol ester SM = Squalene mono-oxygenase LDLR = LDL-Receptor HMGCS = HMG CoA Synthase HMGCR = HMG CoA Reductase Blood Transport Transport via membrane contact sites Conversion Down-regulated gene in tpp1-/- mutant zebrafish vs WT siblings (DMSO), alleviated with pregnenolone treatment Figure 7 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 27, 2025. ; https://doi.org/10.1101/2025.08.26.670480doi: bioRxiv preprint