The mechanistic target of rapamycin (mTOR) signalling pathway plays a crucial role in regulating cellular growth and proliferation. While extensively studied in mammals, the phosphorylation dynamics of this pathway in non-mammalian model organisms remain largely unexplored, often due to the scarcity of suitable antibodies to measure (phosphorylated) proteins of interest. To address this gap, we developed an antibody-independent targeted phosphoproteomics method applying liquid chromatography-tandem mass spectrometry (LC-MS/MS) to quantify the abundance and phosphorylation levels of mTOR pathway-related proteins in zebrafish (Danio rerio), using the permanent cell line PAC2 as a model system. With optimized sample processing and data analysis strategies, we could successfully quantify 10 endogenous phosphosites and 15 endogenous proteins at different cell culture growth phases, revealing complex phosphorylation dynamics for both the upstream regulators (e.g., AKT, AMPK) and downstream effectors (e.g., eIF4EBP1, RPS6) of the mTOR pathway, which reflected transition from exponential growth to stationary subsistence. Our findings confirm the overall similarity of the mTOR pathway structure and functionality between zebrafish and mammals. Furthermore, this work demonstrates the high potential of the LC-MS/MS-based analytical approaches for studying phosphorylation-governed signalling in diverse organisms of interest, thus paving the way for further investigations in comparative physiology and toxicology across species.

2.2 PAC2 Cell Culture Growth Experiments and Sampling for Proteomics Analysis The starting cell number for all experiments was 39,000 cells (seeded with 13,000 cells/mL suspension) in each well of the 12-well plate. Cells were cultured as described above and the medium was changed every seven days. The growth curve was determined by quantifying the cell numbers at 15 time points, i.e., from day 1 to day 35 after seeding, using the CASY system. For proteomics analysis, five sampling points were selected throughout the different phases of PAC2 cell culture growth: day 4 (early exponential growth phase), day 7 (mid exponential growth phase), day 11 (late exponential growth phase), day 18 (early stationary phase) and day 28 (late stationary phase/early decline phase). These sampling points were selected such that changes of the culture medium were performed at least 4 days prior to sample collection, in order to avoid any bias introduced by immediate effects of nutrient introduction on phosphorylation or protein abundance levels. One plate was sampled at each time point, where three wells were used for cell counting with CASY, while cells in all other wells were lysed, pooled into one reaction tube, and processed for proteomics analysis following the procedures described in the next section. The described cell culture growth experiments and sample collection for proteomics analyses were performed in biological triplicates. 2.3 Cell lysis and S-Trap™ Protein Digestion Digestion of proteins was done using the S-Trap™ mini spin columns (Protifi) according to the manufacturer’s instructions, with minor adjustments [34]. Prior to lysis, each well of the 12-well plates was washed three times with 500 µL of phosphate-buffered saline (PBS) without magnesium and calcium (Cytiva). Cells were lysed by adding up to 100 µL (depending on the confluence) of 5% sodium dodecyl sulfate (SDS) in 0.1 M Tris-buffer (pH 7.55) for 1 min. The pooled lysates were then heated at 90°C for 10 min and stored at -70°C until further processing. The processing of samples began with thawing on ice, followed by bath sonication for 5 min. To reduce the viscosity of the lysates caused by high DNA amounts, the DNA was sheared by passing the lysate through a syringe needle (Sterican Needle 27G, 0.40x20mm; B. Braun). Total protein concentrations were determined with the bicinchoninic acid (BCA) protein assay kit (Thermo Scientific Pierce) and 200 μg of protein from each sample were processed for digestion. Protein disulfide bonds were reduced with 5 mM Tris(2-carboxyethyl)phosphine (TCEP; Sigma-Aldrich), followed by blocking of the free cysteine residues using 25 mM iodoacetamide (IAA; Sigma-Aldrich). Both reactions were carried out in the dark at room temperature for 30 min each. The samples were then acidified with formic acid (Sigma-Aldrich) at 16.7% final concentration, following recommendation for optimized suspension trapping [30], and six sample volumes of the binding/wash buffer (100 mM triethylammonium bicarbonate (TEAB; Sigma-Aldrich), pH 7.5 in 90% methanol) were added. The sample solution was then transferred to the S-TrapTM column, centrifuged at 4000 x g for 30 s and the flow-through was reloaded once. The S-TrapTM column was washed three times with 400 μL TEAB buffer, followed by an additional centrifugation step at 4000 x g for 1 min to remove any residual washing solution. On-column digestion was performed with 120 μL 50 mM TEAB (pH 8.5) containing trypsin (1:50 w/w; Roche) for 16 h at 37°C. Peptide elution was achieved sequentially with 100 μL (each) of 50 mM TEAB (pH 8.5), 0.2% formic acid, and 50% acetonitrile. The elution fractions were pooled, evaporated to dryness using a vacuum centrifuge at 30°C, and stored at −70 °C until further processing for phosphopeptide enrichment. 2.4 Phosphopeptide Enrichment and Collection of Non-Bound Fraction Mixtures of synthetic (phospho)peptides (selected as described in the next section and custom synthesized by JPT, Germany), hereafter referred to as either heavy stable isotope labelled standards (SIS-H; n = 54) or light standards (SIS-L; n = 54), were used to assess the recovery attained by two different phosphopeptide enrichment methods in solvent blank. The two methods were based on TiO2 (ProteoExtract® Phosphopeptide TiO2 Enrichment Kit, Merck) and NTA-Fe3+ (PureCube; Cube Biotech). For the enrichment with TiO2 beads, the protocol provided by the manufacturer based on previously described principles [35] was followed. For the NTA-Fe3+ approach, a previously described procedure [36] was applied with minor modifications. A total of 4.67 pmol of the SIS-H and 3.0 pmol SIS-L mixes was spiked. In case of the NTA-Fe3+ enrichment, the non-bound fraction (i.e., combined supernatants collected after sample incubation and washing steps) was additionally retained for pooling and later analysis (in case of the TiO2 procedure this is not possible). Enrichment from sample matrix was conducted with the NTA-Fe3+ method only. The dried peptide samples (~200 μg) were re-suspended in 150 μL of 80% acetonitrile with 0.2% formic acid containing 4.67 pmol of the SIS-H mix and then sonicated for 10 min. 50 μL of a 5% NTA-Fe3+ bead suspension was added to the peptide mixture and incubated for 10 min on a thermoshaker (1100 rpm, 25°C). The tubes were then placed on a magnetic rack (Cytiva) until all beads were captured (10-30 s) and the supernatant (i.e., non-bound fraction) was removed and collected. The beads were washed three times with 150 μL 80% acetonitrile with 0.2% formic acid and incubated on a thermoshaker (1100 rpm, 25°C) for 2 min each. The supernatant from the wash steps was pooled with the previously collected non-bound fraction. For elution of the phosphopeptides, 50 µL of 3% NH4OH was added, and the samples were incubated for 10 min on a thermoshaker (1100 rpm, 25°C). The eluate (i.e., bound fraction) was transferred to tubes containing 20 µl of 10% formic acid for neutralisation. All samples (i.e., both bound and non-bound fractions, corresponding to enriched phosphopeptides and non-phosphorylated peptide populations, respectively) were dried and stored at -20°C until mass spectrometry analysis. Before analysis, dried samples were re-suspended in solvent A (1% methanol in water, 0.2% formic acid) containing four heavy labelled peptide standards for quality control (two of them phosphorylated; see the Table SI1 for details) and analysed using a multiple reaction monitoring (MRM) method developed as described below. 2.5 Candidate Selection and Development of Targeted (Phospho)proteomics Assays A set of commonly known upstream regulators and downstream substrates of the mTOR pathway was selected based on mammalian literature, reviewed with a focus on the potential involvement of these proteins in mediation of chemical toxicity-induced effects. Zebrafish gene counterparts were then identified using Ensembl (https://www.ensembl.org). To identify putative zebrafish phosphosites potentially involved in growth regulation, we searched the PhosphoSitePlus database [37] and compared the known mammalian sequences with those found in zebrafish. We then used the Skyline program [38] to perform in silico digestion of zebrafish protein sequences and selected proteotypic tryptic peptides to cover, wherever possible, the identified zebrafish phosphosites of interest. Additionally, peptides without known phosphorylation sites were selected on the same proteins, where possible, to be used for quantification of protein abundance levels. Lastly, housekeeping proteins such as actin-beta1 (ACTB1) and 40S ribosomal protein S18 (RPS18) were targeted as well. All target zebrafish proteins and corresponding peptides are listed in the Table SI 1. Selected reaction monitoring (SRM) assays were then developed for the candidate peptide targets using synthetic (phospho)peptides produced by JPT (Germany), following previously published strategies [39, 40]. Assay development and sample analyses were carried out by liquid chromatography (Agilent 1290 Infinity II Bio-inert HPLC system) coupled online to the triple quadrupole mass spectrometer Agilent 6495C. Separation was done on a Poroshell 120 EC-C18 column (2.7 mm, 2.1 i.d. x 100 mm, Agilent) with a flow of 0.15 mL·min-1, using a 38 min linear gradient from 100% solvent A to 100% solvent B (98.8% methanol, 0.2% formic acid), followed by a washing step (4 min with 100% solvent B) and a re-equilibration step (8 min with 100% solvent A). Agilent 6495C was operated with a capillary voltage of 3250 V in positive mode, drying gas flow of 11 L/min, nebulizer pressure of 30 psig, sheath gas heater and flow of 290°C and 12 L/min, and peak filter width of 0.07 Da. For each (phospho)peptide, retention times were initially determined through iterative analyses of synthetic (phospho)peptide mixes where we included extensive transition (precursor-fragment pair) lists generated for each target using the Skyline program. In the next step, a set of 2-5 best transitions were selected, i.e., those with the highest signal intensities, without interferences and with a consistent peak shape and fragmentation pattern. Afterwards, collision energy (CE) optimization was performed to identify the best fragmentation conditions for the target analytes and selected transitions [41]. Information on the target peptides and LC-MS/MS parameters used in the corresponding targeted phosphoproteomics assays is provided in the Table SI 2. For the final MRM methods, the retention time window was set to 2 min, resulting in 54 and 61 maximum concurrent transitions from the scheduled SRM assays included in the methods for analysis of bound and non-bound fractions, respectively. A desired cycle time was set to 1.2 s, resulting in a minimum dwell time of 20 ms for all SRMs. The parameters of the developed SRM assays, including linearity, detection limits, and intra-/inter-assay precision are given in the Table SI 3. 2.6 Data Analysis Raw mass spectrometry data was analysed using Skyline [38, 42] and can be accessed in the Skyilne’s Panorama repository [43] under https://panoramaweb.org/Panorama%20Public/2025/EAWAG%20Environmental%20Toxicology%20-%20mTOR_zf_PAC2_growth_curve/project-begin.view or through the accession PXD059986 shared on ProteomeXchange at https://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD059986. All automatic peak integrations were additionally inspected manually. Criteria for evaluating endogenous peptide signals included retention time within a ±20 s time window from synthetic standards (heavy or light), overlapping transition profiles, matching transition intensity ratios, peak shape, and a signal-to-noise ratio greater than 3. The SRM transition data behind the quantitative proteomics assessments presented in the figures throughout this manuscript are documented in the Tables SI 4-6. Integrated raw peak areas were exported from Skyline and used to evaluate enrichment recoveries based on the spiked peptide standards in the enrichment optimisation experiments. For the analysis of endogenous phosphopeptides in the bound fraction, SIS-H mix was spiked prior to enrichment to correct for losses during sample preparation as previously applied in HeLa cells [44]; with one exception for the double-phosphorylated peptide of RPS6 protein (LpSSLRApSTSK), where no heavy-labelled counterpart was available. Consequently, its quantification in biological samples was conducted without the recovery correction applied to other phosphopeptides. While this has to be considered when interpreting the absolute quantitative values for this particular phosphopeptide, it does not affect the comparative analysis of the relative changes across experimental conditions. For peptides in the non-bound fraction, correction for differences in total protein amount across samples was achieved through normalizing to the average peak areas of two peptides from housekeeping proteins ACTB1 and RPS18, as applied previously [40, 45]. Statistical analysis of corrected/normalized peak areas of individual (phospho)peptides was subsequently performed in GraphPad. Note that several peptides in the raw Skyline files are master peptides (i.e., peptides shared between two or more isoforms of the same protein) but are labeled with the first isoform only. This specifically concerns the peptides with the following sequence identifiers: TFCGTPEYLAPEVLEDNDYGR (master peptide for AKT1, AKT2 and AKT3), IADFGLSNMMSDGEFLR (master peptide for AMPK1 and AMPK2), TGSPNYAAPEVISGR (master peptide for AMPK1 and AMPK2).

2.1 Cell Culture The permanent cell line PAC2 (CVCL_5853), derived from embryonic zebrafish [33], was cultured at 26°C in Leibovitz's L-15 medium (PAN-Biotech) supplemented with 5% fetal bovine serum (FBS; Biochrom). At 90% confluence, cells were harvested and passaged into 75 cm2 cell culture flasks (TPP). The harvesting procedure included a brief rinsing with 1 mL Versene (Invitrogen), followed by two additional washing steps of 5 min each with 1 mL Versene. For the subsequent detachment step, 1 mL trypsin solution (0.25%; Biowest) was added for 5 min with gentle agitation by tapping the flask, followed by resuspension in L-15 medium with 5% FBS. The number of cells in suspension and cell viability were determined using the electric field multichannel cell counting system CASY1 TCC (Schärfe). The cultured PAC2 cells were routinely tested for mycoplasma contamination using the MycoAlert Mycoplasma Detection Kit (Lonza). No contamination was detected throughout this study.