{"id":448894,"date":"2024-01-03T19:00:00","date_gmt":"2024-01-04T00:00:00","guid":{"rendered":"https:\/\/platohealth.ai\/foxo1-mediated-lipid-metabolism-maintains-mammalian-embryos-in-dormancy-nature-cell-biology\/"},"modified":"2024-01-04T07:11:42","modified_gmt":"2024-01-04T12:11:42","slug":"foxo1-mediated-lipid-metabolism-maintains-mammalian-embryos-in-dormancy-nature-cell-biology","status":"publish","type":"post","link":"https:\/\/platohealth.ai\/foxo1-mediated-lipid-metabolism-maintains-mammalian-embryos-in-dormancy-nature-cell-biology\/","title":{"rendered":"FOXO1-mediated lipid metabolism maintains mammalian embryos in dormancy – Nature Cell Biology","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
<\/div>\n

Animal experimentation<\/h3>\n

All animal experiments were performed according to local animal welfare laws and approved by the local authority Landesamt f\u00fcr Gesundheit und Soziales (license numbers ZH120, G0284\/18, G021\/19 and G0243\/18). Mice were housed in individually ventilated cages (Techniplast) on bedding (S-SELECT-09322) on a 12\u2009h dark\/light cycle and fed ad libitum (Ssniff, V1124-300).<\/p>\n

Cell lines and culture conditions<\/h3>\n

Mouse ESCs<\/h4>\n

E14 ESCs (S. Kinkley Lab, MPIMG) were used. Cells were plated on 0.1% gelatin-coated dishes and grown in Dulbecco\u2019s modified Eagle medium (DMEM) high glucose with GlutaMAX medium (Thermo Fisher, 31966047) supplemented with 15% foetal bovine serum (FBS; Thermo Fisher, 2206648RP), 1\u00d7 non-essential amino acids (Thermo Fisher, 11140-035), 1\u00d7 \u03b2-mercaptoethanol (Thermo Fisher, 21985023), 1\u00d7 penicillin\/streptomycin (Life Technologies, 15140148) and 1,000\u2009U\u2009ml\u22121<\/sup> LIF (homemade) and grown at 37\u2009\u00b0C in 20% O2<\/sub> and 5% CO2<\/sub>.<\/p>\n

Mouse TSCs<\/h4>\n

TSCs (M. Zernicka-Goetz Lab, Caltech) were grown on mitotically inactivated mouse embryonic fibroblasts in media containing Roswell Park Memorial Institute 1640 medium with GlutaMAX (Gibco 61870010, Thermo), 20% FBS, 1\u00d7 \u03b2-mercaptoethanol (Thermo Fisher, 21985023), 1\u00d7 penicillin\/streptomycin (Life Technologies, 15140148), 1\u00d7 sodium pyruvate, 25\u2009ng\u2009\u00b5l\u22121<\/sup> FGF4 (R&D Systems, 235-F4-025) and 1\u2009\u00b5g\u2009ml\u22121<\/sup> heparin (Sigma, H3149). Cells were depleted off feeders at the time of collection for analysis. Cell lines have not been authenticated but have been sextyped. Cells tested negative for mycoplasma.<\/p>\n

Developmental pausing setup<\/h3>\n

Mouse ESC\/TSC pausing<\/h4>\n

Cells were treated with the mTOR inhibitor INK-128 (MedChem Express, HY-13328) at 200\u2009nM final concentration. Medium was replenished daily.<\/p>\n

In vitro fertilization<\/h4>\n

The 8\u201312-week-old b6d2F1 mice (Janvier Labs or produced in-house at the MPIMG) were superovulated by an intraperitional injection with pregnant mare serum gonadotrophin (5\u2009IU per 100\u2009\u00b5l) on day 0 of the superovulation protocol. On day 2, mice were intraperitonially injection with human chorionic gonadotropin (5\u2009IU per 100\u2009\u00b5l). On day 3, mice were killed with CO2<\/sub> and cervical dislocation, after which the cumulus oocyte complexes were retrieved. Oocytes were fertilized using b6casF1 (MPIMG) sperm. After overnight culture, the obtained two-cell stage embryos were transferred to a fresh drop of KSOM (Merck, MR-107-D).<\/p>\n

Mouse blastocyst pausing in vitro<\/h4>\n

Embryos as generated above were used. Starting at E3.5, embryos were cultured in reduced KSOM medium made in-house (Supplementary Table 5<\/a>). A 2\u00d7 homemade reduced KSOM medium was prepared and stored for up to 3\u2009months at \u221280\u2009\u00b0C. The medium was filtered through a 0.22\u2009\u00b5m filter (Corning, 431118). A 1\u00d7 reduced KSOM medium was freshly prepared from the 2\u00d7 frozen stock before each experiment with the addition of CaCl2<\/sub>\u20222H2<\/sub>O and bovine serum albumin (BSA) fraction V. The reduced KSOM contains a total nutrient content of 10.4\u2009mM, compared with 13.5\u2009mM in commercially available KSOM56<\/a><\/sup>. Embryos were cultured in four-well dishes (Nunc IVF multidish, Thermo Fisher, 144444) in a volume of 500\u2009\u00b5l reduced KSOM under hypoxic conditions (5% O2<\/sub> and 5% CO2<\/sub> at 37\u00b0C). Embryo survival was evaluated daily under a stereomicroscopy and collapsed embryos were removed.<\/p>\n

Embryo supplements<\/h4>\n

Supplements are listed in Supplementary Table 5<\/a>. The supplement solvent concentration did not exceed 2%.<\/p>\n

Survival analysis<\/h4>\n

Embryo survival curves were plotted in RStudio (version 1.3.1093 with R version 3.6.3) with the survminer (version 0.4.9) package. The survival (version 3.3-1) package was used to extract median and maximum survival values.<\/p>\n

In vivo diapause induction<\/h4>\n

In vivo diapause was induced as previously described57<\/a><\/sup> after natural mating of CD1 mice. Briefly, pregnant females were ovariectomized at E2.5 and afterwards injected every other day with 3\u2009mg medroxyprogesterone 17-acetate (subcutaneously). Diapaused blastocysts were flushed from uteri in M2 media after 4\u2009days of diapause at EDG7.5.<\/p>\n

Global proteomics<\/h3>\n

Sample preparation<\/h4>\n

Proteomics sample preparation was done following a published protocol with minor modifications58<\/a><\/sup>. In brief, 5\u2009\u00d7\u2009106<\/sup> cells were lysed under denaturing conditions in 500\u2009\u00b5l of buffer containing 3\u2009M guanidinium chloride (GdmCl), 10\u2009mM tris(2-carboxyethyl)phosphine, 40\u2009mM chloroacetamide and 100\u2009mM Tris\u2013HCl pH 8.5. Lysates were denatured at 95\u2009\u00b0C for 10\u2009min by shaking at 1,000\u2009rpm and sonicated for 10\u2009min. A total of 100\u2009\u00b5l lysate was diluted with dilution buffer (10% acetonitrile and 25\u2009mM Tris\u2013HCl, pH 8.0, to reach a 1\u2009M GdmCl concentration). Proteins were digested with LysC (Roche; enzyme-to-protein ratio 1:50, mass spectroscopy (MS) grade) shaking at 700\u2009rpm at 37\u2009\u00b0C for 2\u2009h. The digestion mixture was diluted again to reach 0.5\u2009M GdmCl, followed by a tryptic digestion (Roche, enzyme-to-protein ratio 1:50, MS grade) and incubation at 37\u2009\u00b0C overnight in at 700\u2009rpm. Peptide desalting was performed according to the manufacturer\u2019s instructions (Pierce C18 Tips, Thermo Fisher Scientific). Desalted peptides were reconstituted in 0.1% formic acid in water and separated into four fractions by strong cation exchange chromatography (SCX 3\u2009M purification). Eluates were dried in a SpeedVac, dissolved in 5% acetonitrile and 2% formic acid in water, briefly vortexed and sonicated for 30\u2009s before injection to nano liquid chromatography\u2013tandem mass spectometry (LC\u2013MS\/MS).<\/p>\n

Run parameters<\/h4>\n

LC\u2013MS\/MS was carried out by nano-flow reverse phase LC (Dionex Ultimate 3000, Thermo Fisher) coupled online to a Q Exactive HF Orbitrap mass spectrometer (Thermo Fisher), as reported previously59<\/a><\/sup>. LC separation was performed using a PicoFrit analytical column (75\u2009\u03bcm inner diameter (ID)\u2009\u00d7\u200950\u2009cm long, 15\u2009\u00b5m tip ID; New Objectives) packed in-house with 3-\u00b5m C18 resin (Reprosil-AQ Pur, Dr. Maisch).<\/p>\n

Peptide analysis<\/h4>\n

Raw MS data were processed with MaxQuant software (v1.6.10.43) and searched against the mouse proteome database UniProtKB with 55,153 entries, released in August 2019. The MaxQuant processed output files can be found in Supplementary Table 1<\/a>.<\/p>\n

Global proteomics analysis<\/h4>\n

The DEP package (version 1.14.0) was used for the global proteomics analysis60<\/a><\/sup>. Potential contaminants were filtered, unique gene names were generated and only proteins that were quantified in two out of three replicates were included for further analysis. Data were normalized, and missing values were imputed using random draws from a Gaussian distribution centred around a minimal value.<\/p>\n

Pseudotime analysis<\/h4>\n

Unique peptide abundances were averaged across replicates. The 500 most variable proteins were selected for downstream analysis. Z<\/i>-score was calculated and used for diffusion maps via the destiny package version 3.8.0 in R version 4.1.0 with parameter n_pcs\u2009=\u200910. Diffusion pseudotime was calculated using the DPT function of the destiny package.<\/p>\n

Pathway divergence analysis<\/h4>\n

Differentially expressed genes were identified using the DEP package. log2<\/sub> fold change (FC) and adjusted P<\/i> values during entry into pausing compared with 0\u2009h were computed. KEGG61<\/a><\/sup> pathways containing at least ten genes were retained for the divergence analysis. Pathway expression value corresponds to the mean log2<\/sub>FC of proteins in the pathway.<\/p>\n

Divergence score was calculated as:<\/p>\n

\n

$$begin{array}{l}mathrm{Divergence};mathrm{score}=m_{{mathrm{ESC}}}left(mathrm{mean};{log }_{2}mathrm{FC};mathrm{pathway};mathrm{expression}frac{mathrm{tP}}{mathrm{tN}}right)-m_{{mathrm{TSC}}}left(mathrm{mean};{log }_{2}mathrm{FC};mathrm{pathway};mathrm{expression}frac{mathrm{tP}}{mathrm{tN}}right)end{array}$$<\/span><\/p>\n<\/div>\n

where tP is the pause timepoint, tN is the normal (untreated) timepoint and m<\/i> denotes the slope of pathway expression over time, computed with linear regression.<\/p>\n

Pathways with a divergence score of >0.0075 were considered divergent. Waterfall plots were generated using ggplot2 (version 3.3.5).<\/p>\n

Time-series differential expression analysis<\/h3>\n

Potential contaminants were removed, and label-free quantitation values were log2<\/sub> transformed in Perseus version 1.6.14.0. For each cell type, rows were filtered and only proteins expressed in two out of three samples of at least one timepoint were retained. Values were extracted, and a one-way repeated measures ANOVA with MetaboAnalyst 4.0 was used for time-series analysis62<\/a><\/sup>. Proteins with an adjusted P<\/i> value of <0.05 were regarded as differentially expressed.<\/p>\n

Dynamics of differentially expressed proteins<\/h4>\n

k<\/i>-Means clustering was used to identify the dynamic behaviour of the differentially expressed proteins. The R stats package \u2018stats\u2019 (version 4.1.0) was used.<\/p>\n

GO term analysis<\/h4>\n

The clusterProfiler R package (version 4.0.5) and the Benjamini\u2013Hochberg correction was used, with P<\/i> value cutoff of 0.05 and q<\/i> value cutoff of 0.1.<\/p>\n

Metabolomics<\/h3>\n

A total of 1\u2009\u00d7\u2009108<\/sup> ESCs were collected in three biological replicates. Metabolite extraction and LC\u2013MS\/MS measurements were done as previously reported by us63<\/a>,64<\/a><\/sup>. Methyl-tert<\/i>-butyl ester (Sigma-Aldrich), methanol, ammonium acetate and water were used for metabolite extraction. Subsequent separation was performed on an LC instrument (1290 series UHPLC, Agilent), coupled to a triple quadrupole hybrid ion trap mass spectrometer QTrap 6500 (Sciex), as reported previously65<\/a><\/sup>.<\/p>\n

Data analysis<\/h4>\n

The metabolite identification was based on the correct retention time, up to three multiple reaction monitorings and a matching multiple reaction monitoring ion ratio of tuned pure metabolites as a reference. Relative quantification was performed using MultiQuantTM software v.2.1.1 (Sciex), and all peaks were reviewed manually. Only the average peak area of the transition was used for calculations. Normalization was based on cell number of the samples and subsequently by internal standards. Metabolites with a P<\/i>\u2009<\u20090.05 and absolute log2<\/sub>FC >0.75 were regarded as statistically significant.<\/p>\n

Embryo metabolomics sample prep<\/h4>\n

To distinguish the different experimental conditions, embryos were fluorescently labelled with MitoSpy Green (E4.5), MitoSpyRed (D8 mTORi) and 4\u2032,6-diamidino-2-phenylindole (DAPI; D15 mTORi\u2009+\u2009c). Embryos were washed 3\u00d7 with phosphate-buffered saline (PBS), 3\u00d7 with 100\u2009mM ammonium acetate (Merck, 1.01116.1000), deposited onto microscope slides, desiccated for 1\u2009h and stored at \u221280\u2009\u00b0C. Sample slides were desiccated for 1\u2009h in a vacuum chamber and matrix deposition was performed using a TM Sprayer (HTX Technologies) according to the protocol. Matrix for negative mode was 1,5-diaminonaphthalene (Sigma) at 7\u2009mg\u2009ml\u22121<\/sup> sprayed at a constant rate of 0.05\u2009ml\u2009min\u22121<\/sup> for seven passes. For positive mode, 2,5-dihydroxybenzoic acid (Sigma) at 15\u2009mg\u2009ml\u22121<\/sup> was sprayed at a constant rate of 0.07\u2009ml\u2009min\u22121<\/sup> for eight passes. Mass spectrometry imaging was performed with the AP-SMALDI5-Orbitrap MS. The mass range was 100\u2013500 m\/z<\/i> for negative mode and 250\u20131200 m\/z<\/i> for positive mode. After matrix coating, samples were imaged on an AP-SMALDI5 ion source (TransMIT) coupled to a Q Exactive Plus mass spectrometer (Thermo Fisher). Samples were imaged at 15\u2009\u00b5M step size with attenuator level 28. Typical acquisition time was 1\u20132\u2009h per sample, and the area covered was ~50\u2013100\u2009px2<\/sup>. Raw data were analysed by converting to imzML and IBD format using Thermo Fisher ImageQuest software with centroiding. Results were annotated and interpretated using metaspace (https:\/\/metaspace2020.eu\/<\/a>).<\/p>\n

Embryo metabolomics data analysis<\/h4>\n

A total of 26 ions with a false discovery rate <10% and embryonic localization were detected. Metabolite ion intensity was represented as a grey value in grey-scale images. To quantify each metabolite across the full embryo respectively, z<\/i>-projections of all metabolites per spot were created and the embryos masked manually in ImageJ. The resulting masks were used to quantify the average intensity for each metabolite. Integrated intensity was calculated as average intensity multiplied by number of pixels per embryo. The log2<\/sub>FC over E4.5 was computed as integrated intensity (sample) divided by mean E4.5 integrated intensity per metabolite, which was then log2<\/sub> transformed and visualized with a heatmap using ggplot2 (version 3.3.5).<\/p>\n

Seahorse analysis<\/h3>\n

A total of 35,000 cells were plated onto gelatin-coated cell culture miniplates and incubated overnight at 37\u2009\u00b0C and 5% CO2<\/sub>. Seahorse analysis was performed following the manufacturer\u2019s recommendations for Mitostress, Glycostress and Mito Fuel Flex kits. An Agilent Seahorse XFp Extracellular Flux Analyzer was used for measurements.<\/p>\n

Data analysis<\/h4>\n

Basal respiration and glycolysis rates, and glucose and fatty acid dependency values, were computed using the Seahorse 2019 software. Three technical and two biological replicates were performed. Data were plotted with ggplot2.<\/p>\n

FAO rate assay<\/h3>\n

Cells were treated with 50\u2009mM etomoxir and\/or 200\u2009nM INK128 for 48\u2009h as described in Results<\/a>. Cells were trypsinized and plated on Matrigel (Sigma-Aldrich, CLS356237) pre-coated Ibidi \u00b5-Slide 8 Well slides (Ibidi, 80821). After 30\u2009min of attachment, cells were stained for 1\u2009h with SPY650-DNA (SpiroChrome, SC501) in culture medium with inhibitors at 37\u2009\u00b0C, washed twice with HEPES-buffered saline (Sigma-Aldrich, 40010) and stained for 2\u2009h with 20\u2009\u00b5M FAOBlue (DiagnoCine, FNK-FDV-0033) in DMEM with inhibitors at 37\u2009\u00b0C in 20% O2<\/sub> and 5% CO2<\/sub>. Cells were washed once with HEPES-buffered saline and imaged with 200\u2009\u00b5l HEPES-buffered saline. Imaging was done on a Zeiss Plan-Apochromat 20\u00d7\/0.8 objective on the Zeiss LSM880 Airy microscope using longitudinal-section magnetic mode. Images were processed using Fiji ImageJ2 (version 2.3.0) and quantified using CellProfiler (version 4.2.1).<\/p>\n

ESC\/TSC derivation<\/h3>\n

ESCs were derived from E3.5 or D12\u2009+\u2009c embryos of b6d2F1 females (Janvier Labs or produced in-house at the MPIMG, 8\u201312\u2009weeks old) and b6casF1 males (produced in-house at the MPIMG, 2\u20136\u2009months old) as described previously66<\/a>,67<\/a><\/sup>. In brief, embryos were seeded onto mouse embryonic fibroblasts (MEFs) and allowed to attach for 3\u2009days in knockout DMEM (Thermo Fisher, 10829018) supplemented with 20% knockout Serum Replacement (Thermo Fisher, 10828028), GlutaMAX (1.7\u2009mM final, Thermo Fisher Scientific, 35050038), 1\u00d7 non-essential amino acids (Thermo Fisher, 11140-035), 1\u00d7 \u03b2-mercaptoethanol (Thermo Fisher, 21985023), 1\u00d7 penicillin\/streptomycin (Life Technologies, 15140148) and 1,000\u2009U\u2009ml\u22121<\/sup> LIF. Cells were grown at 37\u2009\u00b0C in 20% O2<\/sub> and 5% CO2<\/sub>. After attachment, medium was refreshed every second day, until the attached outgrowths were disaggregated. After forming colonies, cells were passaged and expanded.<\/p>\n

TSCs were derived following previous protocols68<\/a><\/sup>. In brief, embryos were seeded onto MEFs and allowed to attach for 3\u2009days in Roswell Park Memorial Institute 1640 medium with GlutaMAX (Thermo Fisher, 10438026), supplemented with 20% FBS (Thermo Fisher, 2206648RP), 1\u00d7 sodium pyruvate (Thermo Fisher, 11360070), 1\u00d7 \u03b2-mercaptoethanol (Thermo Fisher, 21985023) and 1\u00d7 penicillin\/streptomycin (Life Technologies, 15140148). Cells were grown at 37\u2009\u00b0C in a 20% O2<\/sub> and 5% CO2<\/sub> incubator. Medium was refreshed every second day, until the attached outgrowths were disaggregated. After colony formation, cells were passaged and expanded.<\/p>\n

Electron microscopy<\/h3>\n

Diapaused (EDG7.5) or E4.5 CD1 (produced from 8\u201312-week-old CD1 females bred with 2\u20136-month-old CD1 males produced in house at the MPIMG) embryos were isolated and immediately fixed in 2.5% glutaraldehyde (grade I, Sigma) in PBS for 90\u2009min at room temperature. The fixed embryos were subsequently embedded in 2% low melting point agarose (Biozym). Agarose was cut into cubes with an edge length of ~1\u2009mm, containing one embryo per cube. The cubes were collected in 3-ml glass vials and stored overnight at 4\u2009\u00b0C in PBS. Afterwards samples were post-fixed in 0.5% OsO4<\/sub> for 2.5\u2009h at room temperature on a specimen rotator followed by four washes in ddH2<\/sub>O at room temperature for 5\u2009min each. Samples were then incubated for 1\u2009h in 0.1% tannic acid (Science Services) dissolved in 100\u2009mM HEPES on a rotator, followed by three washes in ddH2<\/sub>O at room temperature for 10\u2009min each. Samples were incubated in 2% uranyl acetate (Sigma-Aldrich, Merck) for 90\u2009min at room temperature on a specimen rotator. Samples were washed in ddH2<\/sub>O and dehydrated through a series of increasing ethanol concentrations. Samples were incubated in a 1:1 mixture of propylene oxide and absolute ethanol for 5\u2009min at room temperature followed by 2\u2009\u00d7\u200910\u2009min incubation in pure propylene oxide (Sigma-Aldrich, Merck) and a 30-min incubation in a 1:1 mixture of propylene oxide and Spurr resin (Low Viscosity Spurr Kit, Ted Pella, Plano GmbH) at room temperature. Samples were transferred to pure Spurr resin mixture and infiltrated overnight at 4\u2009\u00b0C. Embryos were then polymerized at 60\u2009\u00b0C for 48\u2009h. Ultrathin sections (70\u2009nm) were cut using a Leica UC7 ultramicrotome equipped with a 3-mm diamond knife (Diatome) and placed on 3.05-mm formvar carbon-coated TEM copper slot grids (Plano GmbH). Sections were post-contrasted with UranyLess EM stain and lead citrate (both from Science Services). Sections were imaged fully automatically at 4,400\u00d7 nominal magnification using Leginon on a Tecnai Spirit transmission electron microscope (FEI) operated at 120\u2009kV, which was equipped with a 4k\u2009\u00d7\u20094k F-416 CMOS camera (TVIPS). Acquired images were then stitched to a single montage using the TrakEM2 plugin69<\/a><\/sup> implemented in Fiji70<\/a><\/sup>.<\/p>\n

Analysis<\/h4>\n

Semi-automated analysis was done using the ZEN 3.4 Blue software (Zeiss). Blastocysts and subcellular structures were manually masked. The ZEN 3.4 Blue inbuilt function zone of influence analysis was carried out on the masked images to determine the number of mitochondria surrounding each LD (radius of 80\u2009pixels). In cases where LDs occurred as clusters, binary dilation with a count of 15 was used to fuse the LDs in a cluster.<\/p>\n

IF staining<\/h3>\n

Cell lines<\/h4>\n

ESCs were directly cultured on the Nunc Lab-Tek II Chamber Slide System (Thermo Fisher Scientific, 154534). TSCs were trypsinized after culture and colonies were plated on Matrigel-coated (Sigma-Aldrich, CLS356237) Nunc Lab-Tek II Chamber Slides for 30\u2009min. Cells were fixed for 10\u2009min in 4% paraformaldehyde, then permeabilized for 15\u2009min in 0.2% Triton X-100 (Sigma, T8787) in PBS. Permeabilized cells were incubated in blocking buffer (0.2% Triton X-100 in PBS, 2% BSA (BSA Fraction V 7.5%, Gibco, 15260-037) and 5% goat serum (Jackson Immunoresearch\/Dianova, 017-000-121)) for 1\u2009h. Cells were incubated overnight at 4\u2009\u00b0C with the following primary antibodies: Lamin B1 (Abcam, 16048), H3K9me2 (Abcam, 1220), phospho-mTOR (Abcam, ab131538), pS6 (CST, 4858\u2009S), pAKT (CST,), OCT3\/4 (Santa Cruz, sc-5279) and CDX2 (Biogenex, MU392A-UC). Cells were washed and incubated with the following secondary antibodies for 1\u2009h at room temperature: donkey anti-rabbit AF647 (Thermo Fisher, A32795<\/a>) and donkey anti-mouse AF488 (Thermo Fisher, A21202<\/a>). Cells were washed, embedded and imaged with the Zeiss Plan-Apochromat 20\u00d7\/0.8 or 63\u00d7 1.4\u2009NA oil objectives on the Zeiss LSM880 Airy microscope using Airy scan. Airy processing (two-dimensional, strength of one) was done with the Zen Black software. Images were processed using Fiji ImageJ2 (version 2.3.0) and quantified using CellProfiler (version 4.2.1). Nuclei were identified as primary objects using the DAPI stain. The integrated intensities were normalized to the DAPI intensity or nuclear area. The signal intensity of H3K9me2, LAMIN B1 and DAPI were quantified per cell, using the multichannel plot profile of the BAR plugin in Fiji-2. The cell size was scaled, and H3K9me2 and LAMIN B1 intensity were plotted.<\/p>\n

Embryos<\/h4>\n

Embryos were fixed, permeabilized and incubated in blocking buffer as described above, then incubated overnight at 4\u2009\u00b0C with the following primary antibodies: Oct3\/4 (Santa Cruz, sc-5279), H4K16ac (Millipore, 7329), c-Caspase3 (CST, 9661S), FoxO1 (CST, 2880T), Lamin B1 (Abcam, cat#. 16048), H3 (dimethyl K9) antibody (Abcam, 1220), phospho-mTOR (Abcam, ab131538), pS6 (CST, 4858S), pAKT (CST, 4060T), pAMPK (Abcam, ab23875), pACC (CST, 11818T), pULK (CST, 14202S), LC3B (CST, 83506S), LAMP1 (CST, 99437S), CPT1A (Abcam, ab128568) and SOX2 (R&D System, AF2018). Embryos were washed and incubated with the same secondary antibodies as above and donkey anti-goat AF594 (Thermo Fisher, A11058<\/a>) for 1\u2009h at room temperature. Embryos were washed and mounted on a microscope slide with a Secure-Seal Spacer (eight wells, 9\u2009mm diameter and 0.12\u2009mm deep, Thermo Fisher, S24737<\/a>). LDs were stained using BODIPY, together with CellMask Deep Red (Thermo Fisher, A57245<\/a>) and Hoechst 33342 (Thermo Fisher, H3570) in PBS. Imaging, processing and quantifications were done as described above. H3K9me2, LAMIN B1 and DAPI were quantified using the multichannel plot profile of the BAR plugin in Fiji-2. The cell size was scaled, and intensities were normalized for DAPI.<\/p>\n

Alkaline phosphatase staining<\/h3>\n

ESCs were grown on MEFs and allowed to form colonies. The Vector Red Substrate Kit, with alkaline phosphatase (Vector Labs, VEC-SK-5100), was used according to the manufacturer\u2019s instructions.<\/p>\n

Single embryo proteomics experimental setup<\/h4>\n

Embryos were cultured in reduced KSOM, paused for 5\u2009days via mTOR inhibition with 200\u2009nM RapaLink-1 and supplemented with 1\u2009mM l<\/span>-carnitine, or treated with 50\u2009\u00b5M etomoxir or 25\u2009mM AS1842856 together with 1\u2009mM l<\/span>-carnitine.<\/p>\n

Sample preparation<\/h4>\n

Mouse embryos (consisting of approximately 100\u2013120 cells each) were isolated and transferred individually in low protein binding 96-well plates containing 5\u2009\u00b5l lysis buffer (PreOmics). The plate was incubated at 95\u2009\u00b0C for 2\u2009min and centrifuged at 100g<\/i> for 1\u2009min. Lysed proteins were digested with 5\u2009ng trypsin (1\u2009ng\u2009\u00b5l\u22121<\/sup> in ammonium bicarbonate, pH 8, Roche) at 37\u2009\u00b0C for 3\u2009h. Samples were acidified by adding 15\u2009\u00b5l 1% formic acid in water and then put on a rocking platform at 300\u2009rpm for 1\u2009min to mix and centrifuged at 100g<\/i> for another minute.<\/p>\n

Run parameters<\/h4>\n

The digests were loaded onto Evotip Pure (Evosep) tips according to the manufacturer\u2019s protocol. Peptide separation was carried out by nano-flow reverse phase LC (Evosep One, Evosep) using the Aurora Elite column (15\u2009cm\u2009\u00d7\u200975\u2009\u00b5m ID, C18 1.7\u2009\u00b5m beads, IonOpticks) with the 40 samples a day method (Whisper_100_40SPD). The LC system was online coupled to a timsTOF SCP mass spectrometer (Bruker Daltonics) applying the data-independent acquisition with parallel accumulation serial fragmentation method.<\/p>\n

Peptide analysis<\/h4>\n

MS data were processed with Dia-NN (v1.8.1) and searched against an in silico-predicted mouse spectra library. The \u2018match between run\u2019 feature was used, and the mass search range was set to 400\u20131,000\u2009m<\/i>\/z<\/i>, with the MS1 and MS2 accuracy set to 10 and 15\u2009ppm, respectively.<\/p>\n

The DEP package (version 1.14.0) was used for analysis60<\/a><\/sup>. Potential contaminants were filtered, unique gene names were generated and only proteins that were quantified in all replicates of at least one condition were retained for further analysis. Samples with fewer than 3,000 identified proteins were excluded from further analysis, data were normalized and missing values were imputed as described above.<\/p>\n

KEGG pathway analysis<\/h4>\n

Data were prepared and filtered as described above. log2<\/sub>FC and adjusted P<\/i> values for all conditions against E4.5 were computed. KEGG61<\/a><\/sup> pathways containing at least ten gene symbols were included in the divergence analysis. The pathway expression value was defined as the mean log2<\/sub>FC of proteins between any of the timepoints during entry and the 0\u2009h control.<\/p>\n

Adult data RNA seq<\/h3>\n

The following datasets were used: GSE126338<\/a> (GSM3596803<\/a>, GSM3596804<\/a>, GSM3596805<\/a>, GSM3596806<\/a>, GSM3596807<\/a>, GSM3596831<\/a>, GSM3596832<\/a>, GSM3596833<\/a>, GSM3596808<\/a> and GSM3596834<\/a> were omitted due to low mapping quality), PRJEB13002<\/a> (ERR1308085, ERR1308086, ERR1308087, ERR1308082, ERR1308083 and ERR1308084), GSE121589<\/a> (GSM4140537<\/a>, GSM4140538<\/a>, GSM4140539<\/a>, GSM4140540<\/a>, GSM4140541<\/a>, GSM4140542<\/a>, GSM4140556<\/a>, GSM4140557<\/a>, GSM4140558<\/a>, GSM4140559<\/a> and GSM4140560<\/a>), GSE101576<\/a> (GSM2706308<\/a>, GSM2706309<\/a>, GSM2706310<\/a>, GSM2706311<\/a>, GSM2706312<\/a>, GSM2706313<\/a> and GSM2706314<\/a>), GSE116997<\/a> (GSM3267160<\/a>, GSM3267161<\/a>, GSM3267162<\/a>, GSM3267157<\/a>, GSM3267158<\/a> and GSM3267159<\/a>), GSE138243<\/a> (GSM4103308<\/a>, GSM4103309<\/a>, GSM4103310<\/a>, GSM4103311<\/a>, GSM4103316<\/a>, GSM4103317<\/a>, GSM4103318<\/a> and GSM4103319<\/a>), GSE138884<\/a> (GSM4121228<\/a>, GSM4121229<\/a>, GSM4121230<\/a>, GSM4121234<\/a>, GSM4121235<\/a>, GSM4121236<\/a>, GSM4121237<\/a>, GSM4121231<\/a>, GSM4121232<\/a>, GSM4121233<\/a>, GSM4121238<\/a>, GSM4121239<\/a>, GSM4121240<\/a> and GSM4121241<\/a>), GSE135705<\/a> (GSM4716555<\/a>, GSM4716556<\/a>, GSM4716557<\/a> and GSM4716558<\/a>) and GSE168617<\/a> (GSM5151338<\/a>, GSM5151339<\/a>, GSM5151340<\/a> and GSM5151341<\/a>).<\/p>\n

Reads were aligned to the reference genome mm10 using STAR-2.5.3a (ref. 71<\/a><\/sup>) with default parameters except: –outFilterMultimapNmax 5, –outFilterMismatchNoverLmax 0.1, –chimSegmentMin 10 and –quantMode GeneCounts. Reads were counted using HTSeq 0.11.4 (ref. 72<\/a><\/sup>) using GRCm38.gtf file (Ensembl Release 101) and default parameters, except –nonunique all and -a 0.<\/p>\n

Differentially expressed genes were identified using DESeq2 (ref. 73<\/a><\/sup>). To measure the effect of the conditions and to simultaneously control for batch differences the design parameter \u2018~ batch\u2009+\u2009condition\u2019 was used. In addition, the six tissues were analysed separately. For tissues represented by more than one dataset, the design parameter \u2018~ batch\u2009+\u2009condition\u2019 was used, while design parameter \u2018~ condition\u2019 was used for tissues represented by one dataset.<\/p>\n

The merged DESeq2 results were filtered for adjusted P<\/i>\u2009<\u20090.05. Among them, the genes belonging to one of the following categories were selected: commonly upregulated genes (adjusted P<\/i>\u2009<\u20090.05 and positive log2<\/sub>FC values in at least four out of six tissues), commonly downregulated genes (adjusted P<\/i>\u2009<\u20090.05 and negative log2<\/sub>FC values in at least four out of six tissues) and tissue-specific genes (adjusted P<\/i>\u2009<\u20090.05 only in the given tissue and not in the remaining ones). The heatmap (Fig. 7b<\/a>), generated using R package \u2018ComplexHeatmap\u201974<\/a><\/sup>, represents log2<\/sub>FC values from all DESeq2 runs for genes belonging to one of these categories.<\/p>\n

GO term and pathway enrichment analysis<\/h4>\n

EnrichGO from the R package \u2018clusterProfiler\u201975<\/a><\/sup> were used with default parameters.<\/p>\n

Statistics and reproducibility<\/h3>\n

No statistical method was used to pre-determine sample size, but our sample sizes are similar to those reported in previous publications13<\/a>,76<\/a><\/sup>. No data were excluded from the analyses. The experiments were not randomized and the investigators were not blinded to allocation during experiments and outcome assessment. Data distribution was assumed to be normal, but this was not formally tested. The precise number of biological replicates is indicated in the figure legends. A minimum of three different embryos were used for embryo stainings and electron microscopy. For image quantifications, precise numbers of quantified cells are provided in the figures or legends. Number of embryos used for survival analyses are provided in the figures or legends. Source data<\/a> are provided with this study. Exact P<\/i> values that were not provided in the figures or legends are available in Supplementary Table 11<\/a>.<\/p>\n

Reporting summary<\/h3>\n

Further information on research design is available in the Nature Portfolio Reporting Summary<\/a> linked to this article.<\/p>\n