Animal experimentation

Animal experiments were performed according to local animal welfare laws and approved by authorities (Landesamt für Gesundheit und Soziales), covered by LaGeSo licenses ZH120, G0284/18, G021/19 and G0243/18-SGr1_G. Mice (7–12 weeks old) were housed with enrichment material in ventilated cages (humidity 45–65%, temperature 20–24 °C) on a 12 h light/dark cycle and fed ad libitum.

Cell lines and culture conditions

Wild-type, Tet1/2 DKO and Tet1Flag/Tet2V5/Tet3HA ES cells26 (all E14 background) were cultured without feeders on gelatin-coated plates (0.1%, Sigma-Aldrich G1393) at 37 °C in a humidified 5% CO2 incubator with medium containing Dulbecco’s modified Eagle medium high glucose with GlutaMAX (Thermo, 31966047), 15% fetal bovine serum (Thermo, 2206648RP), 1× nonessential amino acids, 1× penicillin–streptomycin, 1× β-mercaptoethanol and 1,000 U ml−1 LIF (homemade). Wild-type KH2, Dnmt3a/b DKO, Dnmt TKO (from Alexander Meissner Lab) and Tet1/2/3 TKO ES cells (from Jacob Hanna Lab) were cultured on mitomycin-treated feeders produced in house. Wild-type E14s were acquired from Sarah Kinkley Lab, Tet1/2 DKO cells were generated by us, and Tet1Flag/Tet2V5/Tet3HA ES cells were gifted by Ian Chambers. Cell lines were not authenticated, but genotypes were verified by PCR. Cells tested negative for mycoplasma.

In vitro diapause

ES cells or embryos were treated with the mTOR inhibitor INK128 or RapaLink-1 at 200 nM final concentration for the durations specified in individual experiments. To obtain embryos, 10- to 12-week-old b6d2F1 mice were superovulated via intraperitoneal injection with pregnant mare serum gonadotrophin (5 IU per 100 µl) on day 0, with human chorionic gonadotrophin 5 IU per 100 µl on day 2, and killed on day 3. Oocytes were collected and incubated with 10 µl of motile sperm in CARD medium (CosmoBio, KYD-003-EX) for in vitro fertilization. After overnight culture, two-cell-stage embryos were transferred to a fresh drop of K+ simplex optimised medium (KSOM) (Merck, MR-107-D) and cultured until the blastocyst stage.

Embryo transfer and in vivo diapause

Blastocysts were transferred into pseudopregnant females that have been previously mated with vasectomized males at E2.5. To induce diapause in vivo, ovariectomy was performed after embryo transfer as described previously70. Females were afterward injected every other day with 3 mg medroxyprogesterone 17-acetate subcutaneously. Diapaused blastocysts were flushed from uteri in M2 medium after 4 days of diapause at equivalent day of gestation (EDG) 7.5.

Flow cytometry

Cells were dissociated from plates using TrypLE (Thermo Fisher 12604-021) and washed in Dulbecco’s phosphate-buffered saline (Dulbecco). Cells were labeled with Alexa488-SSEA1 (BioLegend, 125610, 1:1,000) or Alexa647-AnnexinV (Invitrogen, A23204, 1:250) together with a live/dead cell stain excitable at 405 nm wavelength (Invitrogen, L34955), respectively, for 20 min in the dark on ice and subsequently washed in phosphate-buffered saline (PBS) containing 2% bovine serum albumin (BSA). After washing, fluorescence was measured on a BD FACSDIVA or BD FACSAriaII flow cytometer. Data analysis and visualization was performed using FlowJo (v10.8.2).

Pharmacological treatment of embryos

Embryos were treated with Bobcat339 (100 µM final, Sigma SML2611), vitamin C (50 µM final, Sigma A4403), Rapalink (200 nM final Hölzel HY-111373) or α-ketoglutarate (200 µM final, Merck K1128) in KSOM medium (Merck, MR-107-D) in four-well dishes (Nunc IVF multidish, Thermo Scientific, 144444) in a volume of 500 µl. Recently, it has been shown that Bobcat339 effectivity is influenced by copper, which was not separated from the inhibitor stock used here50. Survival plots and statistical tests (log-rank test) were produced using RStudio (version 1.3.1093 with R version 3.6.3) with the survminer (version 0.4.9) package.

Proliferation curves

Cells were seeded in six-well plates (Corning, 3516) at a density of 105 cells per well (on feeder cells when required, 2.5M per plate). INK128 (MedChemExpress, MCE-HY-13328) treatment was started the next day, and cells were counted every day using Cell Countess 3 (Invitrogen).

Overexpression of wild-type or catalytically dead Tet1 and/or Tet2

Wild-type Tet1 or Tet2 coding sequence was amplified from pcDNA3-Tet1 (Addgene 60938) and FH-Tet2-pEF (Addgene 41710) and cloned into a pCAGGS vector. To mutate catalytic activity, H1652Y and D1654A (Tet1) and H1304Y and D1306A (Tet2) were altered using the Q5 Site Directed Mutagenesis Kit (NEB, E0554S)71. Primer sequences:

(Tet1 F: 5′ GGCGATTCACAACATGCACAAC,

R: 3′ TTGTAAGAATGGGCACAAAAATC,

Tet2 F: 5′ AGCGCAGCAGAACATGCCAAATG,

R: 3′ CTGTAGGAATGAGCAGAGAAGTC).

Generation of Tet1/2 KO mouse ES cells

Tet1 and Tet2 genes were knocked out using gRNAs targeting Tet1 exons 4 (GATTAATCACATCAACGCCG) and 13 (GCTTTGCGCTCCCCCAAACGA) and Tet2 exons 3 (GAGTGCTTCATGCAAATTCG) and 12 (GCTACACGGCAGCAGCTTCG). gRNA sequences were cloned into the pX330 plasmid with mCherry fluorescence. Wild-type E14 cells were nucleofected with the plasmids using the Lonza 4D Nucleofector. After 48 h, cells were single-cell sorted into 96-well plates using the BD FACSAria Fusion (Software v8.0.1). After 8 days, clones were screened using two primer pairs for each Tet gene.

Primer pair 1 (Tet1 F: 5′ AGCCATAGAAGCCCTGACTC,

R: 3′ CGGAGTTGAAATGGGCGAAA,

Tet2 F: 5′ CCGAAGCAACCGAACTCTTT,

R: 3′ ACAAGTGAGATCCTGGTGGG) binds outside the guide targeted region and only produces a PCR product after successful KO.

Primer pair 2 (Tet1 F: 5′ CGCCTGTACAAAGAGCTCAC,

R: 3′ AGGCTAGTCTCAGTTGGCAG;

Tet2 F: 5′ TTCTAATGCCTGTGTTCTCTCA,

R: 3′ CAACCTCTTTTGGCTCAGCT) binds at exon 6. KOs were confirmed via western blot using the TET1 (NBP2-19290, Novus Biologicals, 1:1,000) and TET2 (Cell Signaling Technology 45010S, 1:1,000) antibodies.

IF imaging and quantifications

Cells and embryos fixed with 4% paraformaldehyde in Dulbecco’s phosphate-buffered saline for 10 min. After fixation, cells were washed, permeabilized with 0.2% Triton X-100 in PBS for 10 min at room temperature, and blocked in 0.2% Triton X-100 containing 2% BSA and 5% goat serum for 1 h at room temperature. For 5mC and 5hmC IF, cells were depurinated after permeabilization using 2 N HCl for 1 h followed by a neutralization of 30 min in 0.1 M sodium borate. Cells were incubated with the following primary antibodies: anti-5mC (Diagenode, C15200003; 1:100), anti-5hmC (ActifMotif, 39769; 1:200) and anti-TFE3 (Merck, HPA023881; 1:50) overnight at 4 °C. Cells were washed and incubated with the following secondary antibodies for 1 h at room temperature: donkey anti-rabbit AF647 (Thermo Fisher, A32795, 1:1,000) and donkey anti-mouse AF488 (Thermo Fisher, A21202, 1:1,000).

Cells were mounted in Vectashield containing 4′,6-diamidino-2-phenylindole (DAPI; Vectashield, Cat: H-1200). Images were acquired on a ZEISS LSM880 microscope at 20× magnification, with Zen black and Zen blue software (version 2.3) and processed using Fiji (version 2.3.0) and CellProfiler72 (version 4.2.1).

Western blotting

Samples were mixed with 4× ROTI loading buffer (Carl Roth, K929.2), boiled at 98 °C for 5 min and loaded on 4–15% Mini-PROTEAN®TGX precast protein gels (Bio-Rad, 4561083). Proteins were separated by electrophoresis at 70 V for 15 min followed by 100 V for 1 h using 10× Tris/glycine/sodium dodecyl sulfate running buffer (Bio-Rad, 1610772). Proteins were transferred to a polyvinylidene fluoride membrane (Thermo Fisher Scientific, IB24001) using the iBlot 2 dry blotting system (Thermo Fisher Scientific, IB21001) and run at 20 V for 7 min. Membranes were blocked with 5% milk in TBS-T buffer (Thermo Fisher Scientific, 28360) for 1 h at room temperature, and incubated with primary antibody (indicated in each method section) in 5% milk in TBS-T buffer overnight, followed by secondary antibody at room temperature for 1 h. For detection, membranes were incubated with ECL Western Blotting Substrate (Thermo Fisher Scientific, 32106) for 1 min before imaging with the ChemiDoc system (Bio-Rad).

RNA-seq

Cells were trypsinized and sorted on a BD FACSAria Fusion (Software v8.0.1 configuration 2B-5YG-3R-2UV-6V). Total RNA was extracted from 200,000 cells using the Qiagen RNeasy kit (Qiagen, 74004). External RNA Controls Consortium (ERCC)68 RNA Spike-In Mix (Thermo, 4456740) was used. Libraries were prepared from 500 ng total RNA using KAPA RNA HyperPrep Kit with RiboErase (Roche, 8098131702) following the manufacturer’s instructions, and sequenced on a NovaSeq 600, S4 flow cell, paired-end mode. Raw reads were subjected to adapter and quality trimming with cutadapt73 (version 2.4; parameters: –quality-cutoff 20–overlap 5–minimum-length 25–interleaved–adapter AGATCGGAAGAGC -A AGATCGGAAGAGC), followed by poly(A) trimming (parameters: –interleaved–overlap 20–minimum-length–adapter “A[100]”–adapter “T[100]”). Reads were aligned to the mouse reference genome (mm10) using STAR (version 2.7.5a; parameters: –runMode alignReads–chimSegmentMin 20–outSAMstrandField intronMotif–quantMode GeneCounts)74, and transcripts were quantified using stringtie (version 2.0.6; parameters: -e)75 with GENCODE annotation (release VM19). For the repeat expression quantification, reads were realigned with additional parameters ‘–outFilterMultimapNmax 50’. Differential gene expression analysis was performed on stringtie output using DEseq2 (ref. 76) (version 1.38.2).

ATAC-seq

A total of 50,000 cells per sample were collected as described above. The ATAC-seq protocol from Corces et al.77 was followed. Illumina transposase was used (20034198). Samples were purified using the Zymo DNA Clean and Concentrator-5 Kit (D4014). Libraries were amplified with eight PCR cycles with i5 and i7 primers from ref. 78. The final number of cycles was determined following ref. 78. Libraries were sequenced as above. Raw reads were subjected to adapter and quality trimming with cutadapt as above and aligned to the mouse genome (mm10) using BWA with the ‘mem’ command (version 0.7.17, default parameters)79. A sorted binary alignment map (BAM) file was obtained and indexed using SAMtools with the ‘sort’ and ‘index’ commands (version 1.10)80. Duplicate reads were identified and removed using GATK (version 4.1.4.1) ‘MarkDuplicates’ and default parameters. Replicates were merged using SAMtools ‘merge’. Peaks were called using the MACS2 (ref. 81) peakcall (2.1.2_dev) function with default parameters.

Whole genome bisulfite sequencing

A total of 50,000 cells per sample were collected as described above. Genomic DNA was isolated using the Purelink Genomic DNA mini kit (Invitrogen). Libraries were prepared using the Accel-NGS Methyl-Seq DNA Library Kit. Libraries were sequenced as above. Raw reads were subjected to adapter and quality trimming using cutadapt as above (Illumina TruSeq adapter clipped from both reads), followed by trimming of 10 and 5 nucleotides from the 5′ and 3′ end of the first read and 15 and 5 nucleotides from the 5′ and 3′ end of the second read73. Trimmed reads were aligned to the mouse genome (mm10) using BSMAP (version 2.90; parameters: -v 0.1 -s 16 -q 20 -w 100 -S 1 -u -R)82. A sorted BAM file was obtained and indexed using samtools with the ‘sort’ and ‘index’ commands (version 1.10)80. Duplicates were removed using the ‘MarkDuplicates’ command from GATK (version 4.1.4.1) and default parameters83. Methylation rates were called using mcall from the MOABS package (version 1.3.2; default parameters)84. All analyses were restricted to autosomes, and only CpGs covered by at least 10 and at most 150 reads were considered for downstream analyses.

CUT&Tag

CUT&Tag was performed as described previously in ref. 85. A total of 105 nuclei were incubated with the following primary antibodies overnight at 4 °C: TET1, NBP2-19290, Novus Biologicals, 1:100; TET2, Cell Signaling Technology 45010S, 1:50; TFE3, Sigma HPA023881, 1:50; IgG, Abcam ab46540, 1:100, H3K27ac, 9733S, CST, 1:100; H3K4me3, 9751S, CST, 1:100; H3K4me1, 5326S,CST, 1:100. Guinea pig α-rabbit secondary antibody (ABIN101961, Antibodies Online, 1:100) was used. For tagmentation, homemade 3xFLAG-pA-Tn5 preloaded with Mosaic-end adapters was used. DNA was purified using Chimmun DNA Clean & Concentrator (D5205, Zymo Research).

Libraries were amplified using the NEBNext HiFi 2× PCR Master Mix (New England BioLabs) with i5- and i7-barcoded primers78 and cleaned up using Ampure XP beads (Beckman Coulter). Library quality control was done using the Agilent High Sensitivity D5000 ScreenTape System and Qubit dsDNA HS Assay (Invitrogen). Libraries were sequenced as above. Raw reads were trimmed using cutadapt (version 2.4; parameters: –quality-cutoff 20–overlap 5–minimum-length 25–adapter AGATCGGAAGAGC -A AGATCGGAAGAGC) and aligned to the mouse genome (mm10) using BWA with the ‘mem’ command (version 0.7.17, default parameters)79. A sorted BAM file was obtained and indexed using samtools with the ‘sort’ and ‘index’ commands (version 1.10)80. Duplicate reads were identified and removed using GATK (version 4.1.4.1) ‘MarkDuplicates’ and default parameters. Relicates were merged using SAMtools ‘merge’. Peaks were called using the MACS2 (ref. 81) peakcall (2.1.2_dev) with default parameters.

Pathway expression analysis

Pathway expression value was defined as the mean expression (transcripts per million, TPM) of genes in a given pathway at the indicated time points. Kyoto Encyclopedia of Genes and Genomes86 pathways containing at least ten genes were included in the analysis.

Motif enrichment analysis

Motif enrichment was performed using Homer (v4.7, 8-25-2014). TET dormancy targets were compared against the mouse genome (mm10) using the ‘-size given’ setting.

TF footprinting analysis

ATAC-seq peaks were called by MACS2 (2.2.7.1) with 75-bp shift and 150-bp extension. Differential TF footprints inside these peak regions were identified by TOBIAS36 (0.12.11) using Homer motifs (v4.7, 8-25-2014).

Definition of mouse ES cell enhancer sets

Active and primed enhancer were retrieved from ref. 87 and are defined as follows: active enhancers, genomic regions with p300 enrichment, located within 1 kb of regions enriched in H3K27ac and not enriched in H3K27me3 (within 1 kb); primed enhancers, genomic regions with H3K4me1 enrichment and not enriched in H3K27me3 or H3K27ac (within 1 kb).

Generation of Tet KO embryos via Cas9-assisted gene editing

In vitro-fertilized zygotes were electroporated to generate KOs, as previously described88. In brief, oocytes from superovulated B6D2F1 female mice (7–9 weeks old; Envigo) and sperm from F1B6xCAST was incubated for in vitro fertilization, as previously described89. Pronuclei stage 3 zygotes were rinsed with M2 (Sigma) and OptimMEM I (Gibco, 31985062) medium before electroporation. Three gRNAs per gene were designed targeting the first few exons. gRNAs were assembled with CAS9 into ribonucleoproteins88. Embryos were electroporated on a NEPA21 (Nepagene) in a chamber with 5 mm electrode gap and the following settings: four poring pulses with a voltage of 225 V, pulse length of 2 ms, pulse interval of 50 ms, decay rate of 10% and uniform polarity, followed by five transfer pulses with a voltage of 20 V, pulse length of 50 ms, pulse interval of 50 ms, decay rate of 40% and alternating polarity. Electroporated zygotes were rinsed in KSOM drops (Merck, MR-106-D) and cultured until blastocyst stage. In the case of embryo transfer, 15 blastocysts were transferred into each uterine horn of pseudopregnant female CD-1 (21–25 g, Envigo, age 7–12 weeks) mice 2.5 days post-coitum. E8.5-stage embryos were isolated from the uteri of foster mice. The embryos were dissected in 1× Hanks’ Balanced Salt Solution (Gibco) on ice after the decicuda were removed. Embryos were washed in 1× PBS (Gibco) with 0.4% BSA and imaged on an Axiozoom (ZEISS) microscope. Images were processed with Fiji.

For KO validation qPCRs, RNA was isolated using Arcturus Pico Pure RNA isolation kit (Biosystems) and reverse transcribed using High-Capacity cDNA synthesis kit (KAPA biosystems). RT–qPCR was done using Kapa SYBR 2× master mix. β-Actin was used for normalization. qPCR results were visualized using GraphPad Prism v10.

gRNA sequences:

Tet1:

1.

TCGATCCCGATTCATTCGGG

2.

TTGGCGGCGTAGAATTACAT

3.

GATTAATCACATCAACGCCG

Tet2:

1.

AAGATTGTGCTAATGCCTAA

2.

GAGTGCTTCATGCAAATTCG

3.

GCTCCTAGATGGGTATAATA

Tet3:

1.

GAGCGCGCTGAGCATTGCCA

2.

TTCTATCCGGGAACTCATGG

3.

TCGGATTGTCTCCCGTGAGG

Control guide against green fluorescent protein:

GAAGTTCGAGGGCGACACCC

qPCR primers

Tet1:

F: 5′ ACCACAATAAATCAGGTTCACAC, R: 5′ TCTCCACTGCACAATGCCTT

Tet2:

F: 5′ GCAATCACCACCCAGTAGAA, R: 5′ TCCACGTGCTGCCTATGTAC

Tet3:

F: 5′ GCCTCAATGATGACCGGACC, R: 5′ ATGAGTTTGGCAGCGAGGAA

β-Actin:

F: 5′ TGGGTGTATTCCAGGGAGAG, R: 5′ AAGGCCAACCGTGAAAAGAT

FLASH

Tet1 and Tet2 genes were endogenously tagged with the biotinylation signal necessary to perform the FLASH method using a modified plasmid from the CRISpaint toolkit90, which contains a biotinylation signal followed by FLAG tag and a puromycin resistance gene as the donor plasmid. gRNAs targeting the C-terminus of Tet1 (GTTGCGGGACCCTACAATCGT) and Tet2 (GACAACACATTTGTATGACGC), respectively, were expressed from the px330 plasmid together with Cas9. pX330 expressing the appropriate gRNA together with the donor plasmid and the appropriate frame selector plasmid were nucleofected to generate each cell line. Colonies were delected using Puromycin for 10 days. The following primers were used for genotyping:

Tet1:

F: 5′ GGGAGTGTCCTGATGTATCCCCCG

R: 3′ CTCAGCTCATCACTCCGTGTGTTGA

Tet2:

F: 5′ CCAGTCTCTTGCTGAGAACACAGGG

R: 3′ CAGATGCTGTGACCTGTCCCTACG.

Successful knock-in of endogenous Tet1 and Tet2 was confirmed by a streptavidin pulldown followed by western blotting. Membranes were incubated with horseradish peroxidase-conjugated streptavidin (Thermo, N100) to visualize biotinylated TET1 and TET2 proteins.

FLASH was performed following the protocol in ref. 46. Cells were crosslinked with 0.15 mJ cm−2 ultraviolet (UV)-C irradiation. Isolated RNA was reverse-transcribed and RNase H-treated. cDNA was column-purified and circularized with CircLigase for 2–16 h. Circularized cDNA was directly PCR amplified, quantified and sequenced on Illumina NextSeq 500 in paired-end mode.

IP–MS

Cells were fractionated to collect nuclei. For this, cells were lysed using cold buffer A (10 mM HEPES pH 7.9, 5 mM MgCl2, 0.25 M sucrose and 0.1% NP-40), incubated for 10 min on ice and passed four times through an 18G1 needle (BD Microlance 3, 304622), and centrifuged. The pelleted nuclei was lysed in cold buffer B (10 mM HEPES pH 7.9, 1 mM MgCl2, 0.1 mM EDTA, 25% glycerol and 0,5 M NaCl), incubated for 30 min on ice, passed four times through an 18G1 needle and sonicated using Bioruptor 300 (Diagenode) with settings 30 s on, 30 s off for 5 min at 4 °C. Protein concentration was quantified using the BCA Protein Assay Kit (Pierce, 23225). One milligram of protein was used per pulldown in a minimum volume of 500 µl in IP buffer (20 mM HEPES pH 7.9, 25% glycerol, 0.15 M NaCl, 1.5 mM MgCl2, 0.2 mM EDTA and 0.02% NP-40). FLAG antibody (Merck, F3165) and Dynabeads Protein A (Invitrogen, 10001D) were added (1 mg of beads and ~8 µg of antibody) and the mixture incubated for 2 h at 4 °C. After pulldown, beads were washed 3× in IP buffer and immediately digested for MS or boiled at 96 °C in 30 µl of Leammli buffer (Roth, K930.1) for western blotting.

MS

A total of 135 µl of 100 mM ammonium bicarbonate was added to the washed magnetic beads. This was followed by a tryptic digest including reduction and alkylation of the cysteines. The reduction was performed by adding tris(2-carboxyethyl)phosphine with a final concentration of 5.5 mM at 37 °C on a rocking platform (700 rpm) for 30 min. For alkylation, chloroacetamide was added with a final concentration of 24 mM at room temperature on a rocking platform (700 rpm) for 30 min. Then, proteins were digested with 200 ng trypsin (Roche) per sample, shaking at 1,000 rpm at 37 °C for 18 h. Samples were acidified by adding 6 µl 100% formic acid (2% final), centrifuged shortly and placed on the magnetic rack. The supernatants, containing the digested peptides, were transferred to a new low-protein-binding tube. Peptide desalting was performed on C18 columns (Pierce). Eluates were lyophilized and reconstituted in 11 µl of 5% acetonitrile and 2% formic acid in water, briefly vortexed and sonicated in a water bath for 30 s before injection to nano-liquid chromatography (LC)–tandem MS.

Run parameters

LC–tandem MS was carried out by nanoflow reverse-phase LC (Dionex Ultimate 3000, Thermo Scientific) coupled online to a Q-Exactive HF Orbitrap mass spectrometer (Thermo Scientific), as reported previously. Briefly, the LC separation was performed using a PicoFrit analytical column (75 μm inner diameter × 50 cm long, 15 µm Tip inner diameter; New Objectives) in-house packed with 3-µm C18 resin (Reprosil-AQ Pur, Dr. Maisch).

Peptide analysis

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.

Enhancer and L1Md gene contact analysis

To determine the genes looping with the identified active enhancers and potentially by L1Md elements, we used the publicly available predictions from the Activity-By-Contact model49. For gene looping with enhancers, we used the recommended cut off (ABC score 0.02). For potential L1Md–promoter contacts, the HiC contact probability (>15) provided in the same file for mouse ES cells was used (mESC.AllPredictions.txt).

Gene knockdowns

Three short hairpin RNAs (shRNAs) against TFE3 were cloned into a pLKO.1 plasmid containing a puromycin resistance gene and allowing for doxycycline-inducible expression, respectively, resulting in pLKO.1.shTFE3:

shTFE3_1: ATCCGGGATTGTTGCTGATAT, shTFE3_2: GTGGATTACATCCGCAAATTA, shTFE3_3: AGCTATCACCGTCAGCAATTC.

Two micrograms of pLKO.1.shTFE3 was co-transfected with 2 µg of equal parts pVSV-G, pMDL and pRSV (packaging vectors) into HEK 293T cells grown to 80% confluency on a 10-cm uncoated culture dish. After 24 h, cell culture supernatant was collected for 3 consecutive days to enrich for produced viruses. The virus supernatant was concentrated and used to transduce E14 wild-type cells. For transduction, 100,000 E14 cells were mixed with 50 µl of concentrated virus suspension and 10 µg ml−1 polybrene in a 1.5 ml tube and rotated at 37 °C for 1 h. Subsequently, cells were plated on six-well culture dishes and grown for 48 h after which puromycin selection was applied for 6 days. Efficient knockdown of TFE3 was confirmed by IF followed by confocal microscopy.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.