London, United Kingdom
London, United Kingdom

Time filter

Source Type

No statistical methods were used to predetermine sample size. The experiments were not randomized and the investigators were not blinded to allocation during experiments and outcome assessment. All mouse experiments were approved by the Animal Research Committee, the Norwegian Food Safety Authority (NFDA), and conducted in accordance with the rules and regulations of the Federation of European Laboratory Animal Science Associations (FELASA). C57BL6/CBA mice were housed in IVC SealSafe Plus Greenline cages in an Aero IVC Greenline system at SPF status. They were maintained on a 12 h light, 12 h dark cycle with ad libitum access to food and water. Four- to eight-week-old donors were injected with 5 units of pregnant mare serum gonadotropin (for oocytes and 2-cells at 14:00; for 8-cells at 15:00, 100 μl of 50 I.U. (international units) ml−1 solution) followed by 5 units of human chorionic gonadotropin (hCG) (for oocytes and 2-cells at 11:00; for 8-cells, 15:00, 100 μl of 50 I.U. ml−1 solution) 45 h (for oocytes and 2-cells) or 48 h (for 8-cells) after injection of pregnant mare serum gonadotropin. For 2-cells and 8-cells collection, females were transferred to cages with males for breeding immediately after hCG injections. Donor mice were killed by cervical dislocation 18 h after hCG injection (no mating). Oviducts were transferred to a clean dish with M2 (Sigma) medium. The ampulla was identified under a stereomicroscope, and the oocytes released followed by removal of cumulus mass by room temperature incubation in M2 containing 0.3 mg ml−1 hyaluronidase. The oocytes were further washed in M2. Donor mice were killed by cervical dislocation 45 h after hCG injection. Oviducts were transferred to a clean dish with M2 medium. Infundibulum was identified and the 2-cells were released by placing a syringe containing M2 inside the infundibulum opening, followed by flushing the M2 through the whole oviduct. The 2-cells were further washed in M2 medium. Donor mice were killed by cervical dislocation 68 h after hCG injection/mating. Oviducts were transferred to a clean dish with M2 medium. Infundibulum was identified and the 8-cells were released by placing a syringe containing M2 inside the infundibulum opening, followed flushing the M2 through the whole oviduct. The 8-cells were further washed in M2 medium. The oocytes, 2-cells and 8-cells were transferred to a 150 μl drop of Acidic Tyrode’s solution (Sigma), and further transferred to a drop of M2 immediately after the zona had been removed. 5 steps of washing in M2 were carried out, and the oocytes, 2-cells and 8-cells were ready for fixation. Immature oocytes were isolated from 12-day-old and 15-day-old prepubertal CD-1 mice (RjOrl:SWISS) as follows. Ovaries were removed with fine scissors and carefully freed from surrounding tissues with a 25G needle. Batches of five ovaries were placed in 800 μl DPBS in a 60 mm culture dish, 400 μl of Trypsin-EDTA (0.05%) (Gibco) was added immediately before fine mincing of the ovaries with a scalpel. After mincing, 5 μl of DNase I (10 U μl−1) (Sigma, 04716728001) was added and the minced ovaries were incubated at 37 °C for 20 min. Next, 20 μl of Collagenase Type II (100 mg ml−1) (Sigma, C9407), 800 μl of DBPS and 400 μl of Trypsin-EDTA (0.05%) was added and the dish was incubated for 10 min at 37 °C. Mechanical dissociation with a pipette then resulted in denuded oocytes. To remove any possible traces of somatic contaminants, and to remove the zona, oocytes were washed four times in M2 medium, incubated in two consecutive drops of M2 containing 0.3 mg ml−1 hyaluronidase, washed two times in M2 medium, then in two drops of Acidic Thyrode’s solution (Sigma) and again washed four times in M2 medium. Batches of oocytes to be analysed for DNA methylation were washed once in WGBS lysis solution (20 mM Tris-HCl, 20 mM KCl, 2 mM EDTA), transferred in a volume of maximum 5 μl to a 1.5 ml tube, snap-frozen in liquid nitrogen and stored at −80 °C before further processing. Batches of oocytes for ChIP−seq were treated as described below. Mouse embryos were immunostained using an adapted protocol from ref. 31 in 96-well plates. Briefly, embryos were subjected to thinning of the zona pellucida using acidic DPBS (pH 2.5), and fixation in 2% paraformaldehyde for 30 min. Embryos were permeabilized in 0.3% BSA, 0.1% Triton X-100, 0.02% NaN PBS solution. Blocking was carried out in 0.3% BSA, 0.01% Tween-20, and 0.02% NaN in PBS. Embryos were incubated in blocking solution with 1:200 H3K4me3 antibody (Merck Millipore, 04-745) for 60 min at room temperature. After further blocking, embryos were finally incubated with goat anti-rabbit Alex Fluor 488 (Invitrogen, A-11008) or Alexa Fluor 568 for morpholino injected embryos (Invitrogen, A-21069) at 1:200 dilution and placed on a slide in SlowFade Gold with DAPI (Invitrogen). Quantitative measurements of H3K4me3 were obtained using a Zeiss Axio Observer epi-fluorescence microscope with a Coolsnap HQ2 camera. Confocal images were obtained with a Zeiss Axio Observer LSM 710 confocal microscope. Images were processed and quantified in Axiovision and ImageJ software. Isolated zygotes were injected at 0.5 dpc (days post coitum) with either fluorescein-tagged morpholino oligonucleotides (Gene-tools) targeted at Kdm5a (5'-TGACGGCCACCAAAGCCCTCTCA-3') and Kdm5b (5'-AGCACAGGGCAGGCTCCGCAACC-3') or five base mismatch control morpholinos for Kdm5a (5'-TGAaGGaCACaAAAcCCCTaTCA-3') and Kdm5b (5'-AcCAaAGGGaAGGaTCCGaAACC-3'). Embryos were cultured in G1 plus media (Vitrolife) until late 2-cell (35 h after hCG) and fixed in 2% PFA. Embryos were further treated as described above. Lysates from 132 two-cell embryos were prepared adding lysis buffer (20 mM Tris-HCl, pH 7.4, 20% glycerol, 0.5% NP40, 1 mM MgCl , 0.150 M NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, and 1 mM PMSF, 1× PIC, 1% SDS) to a final volume of 7 μl. Embryos were lysed on ice for 30 min with occasional vortexing and spinning. Embryos were vortexed and spun down at the end and were frozen on dry ice. Samples were subsequently thawed, sonicated on ultrasound bath for 1 min and centrifuged at 16,000g for 10 min followed by transfer of 5 μl supernatant to a new tube. The simple western immunoblots were performed on a PeggySue (ProteinSimple) using the Size Separation Master Kit with Split Buffer (12–230 kDa) according to the manufacturer’s standard instruction, using the following abtibodies: anti-Kdm5A (CellSignalling, 3876), anti-Kdm5b (Abcam, 181089) and anti-β-actin (Abcam, ab8227). The Compass software (ProteinSimple, version 2.7.1) was used to program the PeggySue-robot and for presentation (and quantification) of the western Immunoblots. Output data was displayed from the software-calculated average of seven exposures (5–480 s). Human NCCIT pluripotent embryonal carcinoma cell line was obtained from ATCC (CRL-2073), and cultured according to ATCC specifications. Mouse E14 ES cells were obtained from a stock at passage P2 equal to what was used for the mouse ENCODE project, and cultured according to that specified by the mouse ENCODE project (https://www.encodeproject.org/biosamples/ENCBS171HGC/). Cell lines were validated by ChIP–seq confirming species and a highly conserved profile. Cell lines were never passaged passed passage 15 for the work described here. Mycoplasma testing was carried out on a regular basis and both of the cell lines were free for Mycoplasma. Cross-linking of oocytes, 2-cell or 8-cell embryos. We added 50 μl M2 medium to a 0.6-ml tube. Embryos were then added and let settle to the bottom. Volume was controlled by eye by comparing to another 0.6-ml tube with 50 μl M2 medium and adjusted with mouth pipette to 50 μl. 50 μl of PBS with 2% formaldehyde was added to get a 1% final concentration and vortexed carefully, incubated at room temperature for 8 min, and vortexed once more. 12 μl of 1.25 M glycine stock (final concentration 125 mM) was added, mixed by gentle vortexing, incubated for 5 min at room temperature, and vortexed once during the incubation step. This was centrifuged at 700g for 10 min at 4 °C in a swinging-bucket rotor with soft deceleration settings and washed twice with 400 μl ice-cold PBS. A volume of 10 μl was left after the last wash, snap-frozen in liquid nitrogen and stored at −80 °C. Binding of antibodies to paramagnetic beads. The stock of paramagnetic Dynabeads Protein A was vortexed thoroughly to ensure the suspension was homogenous before pipetting. 100 μl of Dynabeads stock solution was transferred into a 1.5-ml tube, which was placed in a magnetic rack and the beads captured on the tube wall. The buffer was discarded, and the beads washed twice in 500 μl of RIPA buffer (10 mM Tris-HCl pH 8.0, 140 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 1% Triton X-100, 0.1% SDS, 0.1% Na-deoxycholate) and resuspended in RIPA buffer to a final volume of 100 μl. 96 μl of RIPA buffer was aliquoted into 200-μl PCR tubes on ice, the washed beads were vortexed thoroughly, and 2 μl of bead suspension and 2 μl of either antibody against H3K4me3 (Merck Millipore, 04-745) or to H3K27ac (Active Motif, catalogue number AM39133) was added to each of the 200 μl PCR tubes. This was then incubated at 40 r.p.m. on a ‘head-over-tail’ tube rotator for at least 4 h at 4 °C. Chromatin preparation. The desired number of cross-linked and frozen pools of embryos was removed from −80 °C storage and placed on dry ice in an insulated box (for example, four tubes with a total number of 1,000 2-cell embryos). 10 minutes of cross-linking was carried out during thawing as follows: one tube was moved at the time from dry-ice to ice for 5 s, and any frozen droplets quick pelleted by a brief spin in a mini-centrifuge. 100 μl of 1.1% formaldehyde solution was added (PBS with 1 mM EDTA, 1.1% formaldehyde, 20 mM sodium butyrate, 1 mM PMSF and protease inhibitor cocktail). The tubes were incubated for 10 min at room temperature and vortexed gently twice. 7 μl was added of 2.5 M glycine, vortexed gently and incubated for 5 min before the tube was moved to ice. Tubes were centrifuged at 750g for 10 min at 4 °C in a swinging-bucket rotor with soft deceleration settings, then washed twice with 400 μl PBS with 1 mM EDTA, 20 mM sodium butyrate, 1 mM PMSF and protease inhibitor cocktail. A volume of 10 μl was kept after the last wash. For four tubes, a total of 120 μl of 0.8% SDS lysis buffer with 20 mM sodium butyrate, 1 mM PMSF and protease inhibitor cocktail was used. First 60 μl, then 2 × 30 μl was used for two consecutive rounds of washing through the four tubes by pipetting. The same tip was used and the entire volume (160 μl) left in the last of the four tubes. The sample was sonicated for 5 × 30 s using a UP100H Ultrasonic Processor (Hielscher) fitted with a 2-mm probe. We allowed 30 s pauses on ice between each 30 s session, using pulse settings with 0.5 s cycles and 27% power. 170 μl RIPA Dilution buffer (10 mM Tris-HCl pH 8.0, 175 mM NaCl, 1 mM EDTA, 0.625 mM EGTA, 1.25% Triton X-100, 0.125% Na-deoxycholate, 20 mM sodium butyrate, 1 mM PMSF and protease inhibitor cocktail) was added. The sample was centrifuged at 12,000g in a swinging-bucket rotor for 10 min at 4 °C and the supernatant transferred to a 1.5-ml tube. 200 μl of RIPA Dilution buffer was added to the pellet and sonicated 3 × 30 s. The sample was centrifuged at 12,000g in a swinging-bucket rotor for 8 min, then the supernatant was removed and mixed well with the first supernatant, resulting in a total volume of about 530 μl of ChIP-ready chromatin. Immunoprecipitation and washes. Pre-incubated antibody–bead complexes were washed twice in 130 μl RIPA buffer by vortexing roughly. The tubes were centrifuged in a mini-centrifuge to bring down any solution trapped in the lid and antibody–bead complexes were captured in a magnetic rack cooled on ice. 250 μl of chromatin was added to each of anti H3K4me3 or H3K27ac reactions, and 25 μl kept for input control. 2 μl of cross-linked recombinant histone octameres and 1.25 μg of non-immunized rabbit IgG was immediately added to ChIP reactions, then incubated at 4 °C, 40 r.p.m. on a ‘head-over-tail’ rotator for 30 h. The chromatin–antibody–bead complexes were washed four times in 100 μl ice-cold RIPA buffer. The concentration of SDS and NaCl was titrated for each antibody to find optimal conditions for maximized signal-to-noise ratio. For H3K4me3, we washed 1× RIPA buffer with 0.2% SDS and 300 mM NaCl, 1× RIPA buffer with 0.23% SDS and 300 mM NaCl followed by 2× RIPA buffer with 0.2% SDS and 300 mM NaCl. For H3K27ac, we washed 4× RIPA buffer with 0.1% SDS and 140 mM NaCl. Each wash involved rough vortexing on full speed, repeated twice with pauses on ice in between. Next, a wash in 1 × 100 μl TE and tube shift was carried out as previously described32, 33. DNA isolation and purification. We removed TE and added 150 μl ChIP elution buffer (20 mM Tris-HCl pH 7.5, 50 mM NaCl, 5 mM EDTA. 1% SDS, 30 μg RNase A) and incubated at 37 °C, 1 h at 1,200 r.p.m. on a Thermomixer. 1 μl of Proteinase K (20 mg ml−1 stock) was added to each tube and incubated at 68 °C, 4 h at 1,250 r.p.m. Eluate was transferred to a 1.5-ml tube. A second elution with 150 μl was performed for 5 min and pooled with the first supernatant. ChIP DNA was purified by phenol-chloroform isoamylalcohol extraction, ethanol-precipitated with 10 μl acrylamide carrier as described previously32, 33 and dissolved in 10 μl EB (10 mM Tris-HCl). Library preparation and sequencing. ChIP and input library preparations were carried out according to the ThruPLEX (Rubicon Genomics) procedure with some modifications, including increased incubation times for the library purification and size selection. 12 ChIP libraries were pooled before AMPure XP purification and allowed to bind for 10 min after extensive mixing. Increased elution time, thorough mixing and the use of a strong neodymium bar magnet allowed for high recovery in elution volumes of 25 μl buffer EB. Sequencing procedures were carried out as described previously according to Illumina protocols with minor modifications (Illumina,). We sequenced all P12, P15, oocyte, 2-cell, and 8-cell ChIP–seq libraries as paired-end and all NCCIT ChIP–seq libraries as single-end. Mouse ES cell ChIP–seq libraries were sequenced as paired-end and single-end and the results were combined after mapping for analysis. Single-end and paired-end library information for all samples have been deposited in the GEO database (GSE72784). Sequence read alignment. We aligned single- and paired-end μChIP–seq reads from H3K27ac and H3K4me3 experiments to the mm10 reference genome by using BWA-mem34. For human ChIP–seq samples performed with human NCCIT cells, we aligned reads to the hg19 reference genome using BWA-mem. Unmapped and non-uniquely mapped reads were removed. We also removed PCR duplicate reads with Picard. H3K4me3 ChIP–seq data for heart, liver, and cerebellum were downloaded from the mouse ENCODE project26. H3K4me3 ChIP–seq data for sperm was downloaded from GEO database under accession number GSE42629 (ref. 23). Culture and collection of embryos. 2-cell stage embryos (2 × 25 embryos) were transferred to 350 μl of Buffer RLT (QIAgen RNaseay) (including β-Me according to manufacturer’s description) in a 1.5 ml low-binding tube and snap-frozen in liquid nitrogen and stored at −80 °C. Zygotes were cultured in M16 medium. For α-amanitin treatment the medium contained 10 μg ml−1 α-amanitin (Sigma, A2263). RNA extraction with QIAGEN RNeasy Micro Kit. RNAs were extracted by following QIAGEN RNeasy Micro handbook. Briefly, 25 embryos were disrupted by addition of buffer RLT followed by homogenization of the lysate. One volume of 70% EtOH was added to the lysate, transferred to an RNeasy MinElute spin column, and centrifuged for 15 s. Next, 350 μl of buffer RW1 from QIAGEN RNeasy Micro Kit was added to wash the RNeasy MinElute spin column by 15 s centrifugation. 10 μl DNase I mix was added to the RNeasy MinElute spin column membrane (10 μl DNase + 70 μl buffer RDD) and incubated at room temperature for 15 min. The spin column membrane was washed twice with 500 μl buffer RPE from the QIAGEN RNeasy Micro Kit followed by 500 μl of 80% EtOH. 14 μl RNase-free water was added directly to the centre of the spin column membrane and centrifuged to elute RNA. RNA amplification with NuGEN Ovation RNA-seq system V2. Extracted RNA was amplified by following the NuGEN ovation RNA-seq handbook. Briefly, step 1 was first-strand cDNA synthesis. 2 μl of First Strand Primer Mix from NuGEN Ovation RNA-seq system V2 was added to a PCR tube followed by addition of 5 μl of total RNA sample, and the thermal cycler for primer annealing was run (65 °C for 2 min, held at 4 °C). 3 μl of the First Strand Master Mix from NuGEN Ovation RNA-seq system V2 was added to each tube and thermal cycler for first strand synthesis was run (4 °C for 1 min; 25 °C for 10 min; 42 °C for 10 min, 70 °C for 15 min, held at 4 °C). Step 2 was second-strand cDNA synthesis. 10 μl of the second-strand mix from NuGEN Ovation RNA-seq system V2 was added to each first-strand reaction tube and the thermal cycler was run (4 °C for 1 min; 25 °C for 10 min; 50 °C for 30 min; 80 °C for 20 min; hold at 4 °C). Step 3 was purification of cDNA with Agencourt RNAClean XP beads. Step 4 was SPIA amplification. 40 μl of the SPIA master Mix from NuGEN Ovation RNA-seq system V2 was added to each tube containing the double-stranded cDNA bound to the Agencourt RNAClean XP beads, and the thermal cycler was run to amplify double-stranded cDNA (4 °C for 1 min; 47 °C for 60 min; 80 °C for 20 min, held at 4 °C). The tubes were transferred to the magnet and 40 μl of the supernatant containing the SPIA cDNA was transferred to a new tube, followed by SPIA cDNA purification with QIAGEN MinElute reaction cleanup kit. Library preparation and sequencing. The volume and concentration of purified SPIA cDNA to 500 ng in 100 μl and sonicated SPIA cDNA with Covaris M220 ultrasonicator was adjusted to 400 bp DNA fragment size. Library preparation was carried out according to TruSeq library preparation. Sequencing procedures were carried out as described previously according to Illumina HiSeq2500 protocols with minor modifications (Illumina). DNA methylation libraries of growing oocytes obtained from day 12 (P12) and day 15 (P15) mice were constructed with a modified library protocol from ref. 2. Briefly, 100-500 embryos were lysed in 5 μl lysis buffer (20 mM Tris, 2 mM EDTA, 20 mM KCl, 1 mg ml−1 proteinase K (QIAGEN)) for 1.5 h at 56 °C. followed by heat-inactivation for 30 min at 75 °C. 45 μl nuclease-free water and 0.5% Lamda DNA (Promega) spike-in was added into the lysate. DNA was fragmented with Covaris M220 ultrasonicator and incubated at 37 °C to reduce volume to 30 μl. The fragmented DNA was end-repaired by incubating with 5 μl end-repair enzyme mixture (3.5μl T4 DNA ligase buffer (NEB), 0.35 μl 10 mM dNTP, 1.15 μl NEBNext End Repair Enzyme Mix (NEB)) for 30 min at 20 °C, followed by heat-inactivation for 30 min at 75 °C. Then, 5 μl of dA-tailing mixture (0.5 μl T4 DNA ligase buffer, 1 μl Klenow exo- (NEB), 0.5 μl 100 mM dATP and 3 μl nuclease free water) was added and incubated for 30 min at 37 °C, followed by heat-inactivation for 30 min at 75 °C. Finally, 10 μl ligation mixture (1 μl T4 DNA ligase buffer, 0.5 μl 100 mM ATP, 1.5 μl 50 mM cytosine methylated Illumina adaptor, 2 μl T4 DNA ligase (NEB) and 5 μl nuclease-free water) was added and incubated at 16 °C overnight. 100 ng Carrier RNA (Ambion) was added into the tube. Bisulfite conversion reaction was performed with the EZ DNA methylation-Gold Kit (Zymo Research) according to the manufacturer’s instructions. The purified DNA was then amplified with 6 cycles PCR by using KAPA HiFi HotStart Uracil+ DNA polymerase (KAPA). Amplified DNA was purified with Ampure XP beads (Beckman) to discard the short fragments and adaptor-self ligations. Then, another round of 6–8 cycles of PCR was performed to obtain sufficient molecules for sequencing. Sequencing procedures were carried out as described previously according to Illumina HiSeq2500 protocols with minor modifications (Illumina). Reads were trimmed by Trimmomatic35 with default parameters to remove the reads containing adapters and showing low quality. Trimmed reads were aligned by using Bismark (V12.5)36 Bisulfite Mapper against the mouse reference genome mm10 with parameters: -N 1 –score_min L,0,-0.6. Duplicate reads were removed with Picard after splitting aligned reads into Watson and Crick strands. CpG methylation level was extracted with Samtools mpileup. Strands were merged to calculate the CpG methylation level per dinucleotide CpG site. Methylation level was calculated for each site spanned by at least 4 reads. During RNA-seq data analysis we used GENCODE gene annotation v3. We considered all level 1 and 2 genes and included level 3 protein-coding genes. To define gene expression levels, mouse oocytes, 2-cell, and 8-cell stage embryos RNA-seq data sets were downloaded from the GEO database with accession number GSE44183 (ref. 8). Mouse ES cell RNA-seq data were downloaded from GEO database with accession number GSE39619. RNA-seq reads were aligned to the mm10 reference genome using BWA-mem. Unmapped and non-uniquely mapped reads were removed. Gene expression values were obtained based on GENCODE annotation v3 and normalized to fragments per kilobase of transcript per million mapped (FPKM) values using Cufflinks37. Whole-genome bisulfite sequencing (WGBS) data from sperm was obtained from GEO database with accession number GSE56697 (ref. 2). Oocyte DNA methylation data was obtained from GEO database under accession number GSE56879 (ref. 19). We combined all data from 12 individual MII oocytes and the bulk oocyte sample. Deeply sequenced results were used for MII oocyte with number 2 and 5. WGBS data for GVO and NGO stage oocytes were obtained through personal communication with the authors22. We performed broad peak calling for H3Kme3 in oocytes based on MACS2 broad peak calling algorithm with default parameters (–format = BAM -g mm -m 5 50 -p 1e-5 –broad) followed by combining adjacent peaks within 5 kb. We determined the optimal distance to combine adjacent peaks on the basis of the number of broad H3K4me3 domains at varying distance thresholds. At 5 kb distance threshold, the number of broad domains became stable as shown in Extended Data Fig. 5b. On the basis of the location of transcription start sites (GENCODE v3), we classified broad H3K4me3 domains into two groups as TSS-containing and non-TSS-containing domains. The basic idea of RPKM values is to calculate relative ChIP signal enrichment for a given genomic region compared to the entire genome to normalize different sequencing depth between samples. This approach is reasonable when the total amount of ChIP DNA is similar between samples, and in general the fraction of genomic regions covered by each histone modification mark is similar between samples such as that H3K4me3 marks around 1–3% of the human genome in cells/tissues assessed to date. Therefore by using RPKM values, one can simply avoid a potential bias caused by different sequencing depth. However, if a sample shows an extraordinary ChIP signal distribution, the sample with much larger genomic regions covered with, for example, H3K4me3 tends to show relatively lower RPKM values owing to the large amount of total ChIP signal. In oocytes, we observed such notably broadly distributed H3K4me3 signals, resulting in lower RPKM values than other samples when we consider the top-ranked promoter regions in terms of H3K4me3 signal (Extended Data Fig. 5d). In this regard, to compare H3K4me3 signals fairly between samples, we need to adjust H3K4me3 RPKM values in each cell type. In order to adjust H3K4me3 RPKM values between samples, we used the top-5,000 ranked promoters in terms of H3K4me3 level as internal control regions during H3K4me3 normalization. We calculated H3K4me3-adjustment scaling factors on the basis of the H3K4me3 ChIP signals at the top 5,000 ranked promoters, with the assumption that the promoters with the highest H3K4me3 levels in each cell type represent fully H3K4me3-modified promoters and have similar H3K4me3 signal levels. In support of this assumption, all oocyte and embryo samples represent highly homogenous cell populations, thus it is plausible that most or all cells carry the H3K4me3 mark at the cell-type-specific top-ranked promoters. Furthermore, ChIP conditions were kept the same for all samples. As expected, we observed very similar H3K4me3 signals between samples exhibiting only canonical H3K4me3 patterns when we consider the same number of top-ranked promoters (Extended Data Fig. 5e). On the basis of this observation, we calculated H3K4me3 RPKM adjustment scaling factors for different numbers of the top-ranked promoters (Extended Data Fig. 5f). The scaling factors were calculated by dividing median H3K4me3 RPKM values at the top-ranked promoters in each sample by median H3K4me3 RPKM value at the top-ranked promoters in mES cells. Importantly, the scaling factors are very robust regardless of the number of promoters analysed, indicating that there is a systematic bias caused by different genomic coverage of H3K4me3. Indeed, the adjustment scaling factors are supported by the qPCR-quantified amount of ChIP DNA that is precipitated in each experiment (Supplementary Table 2). Therefore, in this study, we defined the scaling factors based on the top 5,000 most highly ranked promoters. We downloaded a list of maternally expressed genes from GEO database under accession number GSE45719 (ref. 7). We excluded all genes expressed in oocytes to allow us to distinguish maternally expressed genes in the early embryo. We considered genes with less than 0.3 FPKM values as not expressed. On the basis of the extracted genes, we tested whether maternally expressed genes are enriched within broad H3K4me3 domains. The number of genes expected by chance was calculated on the basis of the fraction of all genes located within broad H3K4me3 domains. The significance of enrichment of maternally expressed genes was calculated by Fisher-exact tests. P values were 1.9−8, 2.9−9, 1.6−4, 8.2−6, 2.8−4, 8.0−4 for zygote, early 2-cell-, mid 2-cell-, late 2-cell-, 4-cell-, and 8-cell-stage embryos, respectively. We used a predefined ZGA gene list obtained from a previous study5. The list of oocyte-specific genes (denoted as maternal RNA) was also obtained from the same study after excluding any genes showing less than 0.3 FPKM values in oocytes. Visualization and preceding analysis was done using EaSeq and its integrated tools38. Heat maps were generated using the ‘HeatMap’ tool, and superimposed tracks were generated using the ‘FillTrack’ tool. Data were imported using default settings and all values were normalized to FPKM and scaled as described above (see ‘An adjustment of H3K4me3 RPKM values’). Distances from and orientation of each TSS to the nearest domain centre were calculated using the ‘Colocalize’ tool, and the ‘Sort’ tool was used to order the TSS in the heat maps according to these distances or for ordering heat maps according to domain size. We only considered broad H3K4me3 domains that span more than 5 kbp DNA to avoid any overlapping information at domain boundaries between 5′ and 3′ends of domains. Clustering of domain boundaries was carried out by: (1) quantifying normalized and scaled H3K4me3 RPKM values for P12, P15, and oocyte samples at a set of regions corresponding to the most proximal 2 kbp within the boundary and average DNA methylation frequency for NGO, P12, P15, GVO and oocyte samples at a set of regions corresponding to the most proximal 2 kbp outside of the domain boundaries using the EaSeq (ref. 38) ‘Quantify’-tool (settings: ‘Start = Center, offset = -1000, Fixed width’, ‘End = Center, offset = 1000, Fixed width’, ‘Normalize to reads pr. million, checked’, ‘Normalize to signal size of, unchecked’, ‘Normalized counts to fragments, checked’, ‘Present values as Z-scores, unchecked’); then (2) clustering the boundaries based on this quantified signal using EaSeq’s ‘ClusterP’-tool (settings: ‘Log-Transform, unchecked’, ‘Normalize parameters to average signal’, ‘k-means clustering, checked’, ‘k = 10’, ‘g = 0’). The order of the clusters was changed manually. Distances from each boundary to nearest CGI were calculated using a set of CGIs downloaded from the UCSC table browser and ‘Colocalize’-tool. We predicted distal cis-regulatory elements on the basis of H3K27ac μChIP–seq results. We combined two biological replicates for each cell type and called H3K27ac peaks using MACS2 with the following parameters (–format = BAM -g mm -m 5 50 -p 1e-5). To directly compare the activity of distal cis-regulatory elements between cell types, we defined putative distal cis-regulatory elements by combining all H3K27ac peaks from oocytes, 2-cell and 8-cell embryos and ES cells after excluding chrY, chrM, and any peaks within 2.5 kb from known transcription start sites (GENCODE v3). The activity for each distal cis-regulatory element in each cell type was defined by taking the log ratio between H3K27ac ChIP–seq and input RPKM values. On the basis of the activity of distal cis-regulatory elements, we performed k-means clustering. 20 clusters were defined with Euclidian distance metric followed by reordering clusters manually. On the basis of the clustered patterns, we identified stage-restricted distal cis-regulatory elements. We defined nearby genes of each cRE when the distance between gene TSS and each distal cRE is less than 15 kb. Similarly, in order to define nearby distal cREs for ZGA genes, we combined all distal cREs within 15 kb from each ZGA gene TSS. We used HOMER to find enriched transcription factor motif sequences in distal cREs for each developmental stage. We also performed GREAT39 analysis for each class of stage-restricted distal cREs using the settings ‘single nearest gene’, ‘within 300 kb’ of the enriched H3K27ac region, and no curated regions. In order to identify genes with a certain transcription factors in nearby cREs, we carried out STORM40 motif search with –f –t 0.9 parameters for nearby cREs within 15 kb from each TSS. Each transcription factor motif position weight matrix was obtained from HOMER motif search41 results. The genes with a certain transcription factor in nearby cREs were called if any cREs within 15 kb from the TSS matched with the corresponding transcription factor motif sequence. We called downregulated genes between Kdm5a and Kdm5b MO injected and control MO injected 2-cell embryos when gene FPKM values were 1.5-fold or more reduced in both of the two biological replicates. Additionally, we called experimental stage specific ZGA genes when gene FPKM values were twofold or more reduced in α-amanitin treated embryos as compared to control MO injected embryos. The rational for identifying the experimental stage-specific ZGA genes comes from the observation that the composition of the transcriptome changes dramatically and rapidly during the 2-cell stage7. Although α-amanitin-treated embryos blocked polymerase II transcription from the early 1-cell stage onwards, de novo transcription-independent degradation of maternal RNA may still occur. Therefore, they provide a well-suited control for defining the experimental stage-specific ZGA genes when compared to the control morpholino-injected embryos. As a result, we identified 7,132 putative experimental stage-specific ZGA genes and these genes are significantly overlapped with the ZGA gene list obtained from a previous study5 (hypergeometric P value is 0). KDM5A- and KDM5B-depleted embryos showed that 1,303 ZGA genes are downregulated among 7,132 experimental stage-specific ZGA genes, whereas 980 non-ZGA genes are downregulated among 25,155 genes. We visualized ChIP–seq and RNA-seq data on the basis of raw read depth after converting aligned bam files to wig files using genomeCoverageBed and wigTobigWig utilities.


News Article | November 23, 2016
Site: www.nature.com

The experiments were not randomized and the investigators were not blinded to allocation during experiments and outcome assessment. No statistical methods were used to determine sample size. Intestinal crypts were extracted from 5–10-week-old heterozygous Lgr5–eGFP-IRES-CreERT2 mice (Jackson Laboratory), following animal experimentation protocols prescribed by EPFL and FELASA. Murine intestinal crypts were isolated following procedures described previously. In brief, the proximal part of the intestine was collected, opened longitudinally and washed with ice-cold PBS. The luminal side of the intestine was scraped using a glass slide to remove luminal content and villous structures. After washing with ice-cold PBS again, the intestine was cut into 2–4 mm pieces with scissors. The pieces were transferred to a 50 ml Falcon tube and further washed with cold PBS (5–10 times) with gentle vortexing. Intestinal fragments were then incubated in 20 mM EDTA/PBS for 20 min on ice. EDTA was removed, 10 ml of cold PBS was added and crypts were released by manual shaking of the suspension for 5 min. The supernatant was collected and passed through a 70-μm strainer (BD Biosciences). The remaining tissue fragments were resuspended in 10 ml cold PBS, triturated 5–10 times and the supernatant was passed through a 70-μm strainer. The previous step was repeated a second time. The three crypt-containing fractions were combined and centrifuged at 110g for 5 min. The pellet was resuspended in 10 ml cold Advanced DMEM/F12 (Invitrogen) and centrifuged at 84g to remove single cells and tissue debris. The resulting pellet was enriched in crypts, which were subsequently dissociated or directly embedded in PEG gels or in Matrigel (BD Biosciences; growth factor reduced, phenol red-free formulation). When needed, crypts or ISC colonies were dissociated enzymatically by incubating for 8 min at 37 °C in 1 ml TrypLE Express (Life Technologies), supplemented with DNase I (2000 U ml−1; Roche), 0.5 mM N-acetylcysteine (Sigma) and 10 μM Y27632 (Stemgent). Undigested clusters were removed by passing the suspension through a 40 μm strainer. Freshly isolated mouse crypts or single cells from dissociated mouse ISC colonies were embedded in Matrigel or PEG gels, which were cast into 20-μl droplets at the bottom of wells in 24-well plate. Following polymerization (15 min, 37 °C), the gels were overlaid with 500 μl of ISC expansion medium (Advanced DMEM/F12 containing Glutamax, HEPES, penicillin-streptomycin, B27, N2 (Invitrogen) and 1 μM N-acetylcysteine (Sigma)), supplemented with growth factors, including EGF (50 ng ml−1; R&D), Noggin (100 ng ml−1; produced in-house) and R-spondin (500 ng ml−1; produced in-house), and small molecules, including CHIR99021 (3 μM; Millipore) and valproic acid (1 mM; Sigma). For single-cell culture, thiazovivin (2.5 μM; Stemgent) was included in the medium during the first two days. To induce stem cell differentiation and organoid formation, the medium was removed, the gels were washed with PBS and fresh medium containing EGF, Noggin and R-spondin was added. Human small intestinal and colorectal cancer organoids were generated as described previously31, 32 and grown in 20-μl droplets of Matrigel or PEG gels overlaid with Advanced DMEM/F12 containing Glutamax, HEPES, penicillin-streptomycin, B27 (Life Technologies), Wnt3a (50% conditioned medium; produced in-house; only for small intestinal organoids), R-spondin 1 (20% conditioned medium; produced in-house), Noggin (10% conditioned medium; produced in-house), N-acetylcysteine (2 μM; Sigma), Nicotinamide (10 mM; Sigma), human EGF (50 ng ml−1; Peprotech), A83-01 (500 nM; Tocris), SB202190 (10 μM; Sigma), Prostaglandin E2 (10 nM; Tocris); Gastrin (10 nM; Tocris), and Y-27632 (10 μM; Abmole). In general, growth factors were replenished every two days, with full medium change taking place every four days. Where indicated, the following compounds were used at the specified concentrations: blebbistatin (Sigma, 12.5 μM), ML7 (Calbiochem, 10 μM), cytochalasin D (Merck-Millipore, 0.1 μg ml−1), echistatin (Sigma, 500 nM). Vinylsulfone-functionalized 8-arm PEG (8-arm PEG-VS or sPEG) was purchased from NOF, and acrylate-functionalized 8-arm PEG (8-arm PEG-Acr or dPEG) was purchased from Creative PEGWorks. The transglutaminase (TG) factor XIII (FXIIIa) substrate peptides Ac-FKGGGPQGIWGQ-ERCG-NH2 (TG-DG-Lys), Ac-FKGG-GDQGIAGF-ERCG-NH2 (TG-NDG-Lys) and H-NQEQVSPL-ERCGNH2 (TG-Gln) and the RGD-presenting adhesion peptide H-NQEQVSPL-RGDSPG-NH2 (TG-Gln-RGD) were purchased from GL Biochem. To couple the FXIIIa substrate peptides to the 8-arm PEG-VS or 8-arm PEG-Acr, they were mixed with the PEG powder in a 1.2 stoichiometric excess (peptide-to-VS group); the combined solids were dissolved in triethanolamine (0.3 M, pH 8.0), and allowed to react for 2 h at 37 °C. The reaction solution was dialysed (Snake Skin, MWCO 10K, PIERCE) against ultrapure water for 3 days at 4 °C, after which the five products ((PEG-VS)-DG-Lys, (PEG-VS)-NDG-Lys, (PEG-VS)-Gln, (PEG-Acr)-NDG-Lys, (PEG-Acr)-Gln) were lyophilized. The resulting solid precursors were dissolved in ultra-pure water to make 13.33% w/v stock solutions. Appropriate volumes of 13.33% w/v PEG precursor solutions were mixed in stoichiometrically balanced ratios to generate hydrogel networks of a desired final PEG content. Hydrogel formation was triggered by the addition of thrombin-activated FXIIIa (10 U ml−1; Galexis) in the presence of Tris-buffered saline (TBS; 50 mM, pH 7.6) and 50 mM CaCl . The spare reaction volume was used for the incorporation of dissociated ISCs, fragments of human small intestinal or colorectal cancer organoids, and ECM components: TG-RGD-Gln, fibronectin (0.5 mg ml−1; Invitrogen), laminin-111 (0.1 mg ml−1; Invitrogen), collagen-IV (0.25 mg ml−1; BD Bioscience), hyaluronic acid (0.5 mg ml−1; gift from D. Ossipov, Uppsala University), perlecan (0.05 mg ml−1; Sigma). Gels were allowed to crosslink by incubating at 37 °C for 15 min. Dissociation and release of colonies grown in PEG for downstream cell processing or re-embedding was accomplished by enzymatic digestion of the gels. Gels were carefully detached from the bottom of the plate using the tip of a metal spatula and transferred to a 15-ml Falcon tube containing 1 ml of TrypLE Express (Life Technologies), supplemented with DNase I (2,000 U ml−1; Roche), 0.5 mM N-acetylcysteine (Sigma) and 10 μM Y27632 (Stemgent). Following digestion (10 min, 37 °C), the cell suspension was washed with 10 ml of cold medium, passed through a 40-μm strainer (BD Biosciences) and centrifuged at 1,200 r.p.m. for 5 min. To form mechanically dynamic PEG hydrogels, which underwent varying extents of spontaneous softening, hybrid hydrogels were formed from both PEG-VS and PEG-Acr hydrogel precursors. Specifically, to form a fast-softening 100% Acr gel, stoichiometric quantities of (PEG-Acr)-NDG-Lys and (PEG-Acr)-Gln precursors were allowed to crosslink. A slow-softening 50% Acr gel was formed by crosslinking stoichiometric amounts of (PEG-VS)-NDG-Lys and (PEG-Acr)-Gln precursors. A 75% Acr gel with intermediate kinetics of softening was formed by crosslinking the (PEG-Acr)-Gln precursor with half of the stoichiometric equivalent of (PEG-VS)-NDG-Lys and half of the stoichiometric equivalent of (PEG-Acr)-NDG-Lys. Regardless of the relative proportions of (PEG-VS) and (PEG-Acr) precursors within the hydrogel, its overall PEG content was varied to tune its initial mechanical properties. It should be noted that, by providing an initially stiff and later a soft environment, the mechanically dynamic matrices support both ISC expansion and organoid formation. Hence, ISCs can be expanded and organoids can be formed in the same hydrogel. Hybrid PEG–alginate (PEG–alg) gels were employed to induce a controlled drop in stiffness at a desired time. PEG–alg gels were formed by the simultaneous presence of activated FXIII enzyme—to drive the crosslinking of the PEG macromers—and Ca2+ ions, which induce the crosslinking of the alginate polysaccharides. Hybrid gels were formed by casting a solution, containing 2% (w/v) of stoichiometrically balanced (PEG-VS)-NDG-Lys and (PEG-VS)-Gln precursors, 10 mM TG-RGD-Gln, 10 U ml−1 FXIIIa, 0.8% (w/v) alginate (Sigma) and dissociated ISCs, within a 1% agarose/2% gelatin mould, containing 20 mM CaCl . The solution was incubated at 37 °C for 15 min, carefully de-molded from the agarose substratum and transferred to a 12-well plate containing 1 ml of complete ISC expansion medium. Matrix softening was induced at the desired time by adding 1 U ml−1 alginate lyase (Sigma), and incubating for 1 h at 37 °C. The digested gels were washed and transferred to freshly prepared ISC expansion medium. The shear modulus of the PEG gels was determined by performing small-strain oscillatory shear measurements on a Bohlin CVO 120 rheometer. In brief, preformed hydrogel discs 1–1.4 mm in thickness were allowed to swell in complete cell culture medium for at least 3 h, and were subsequently sandwiched between the parallel plates of the rheometer. The mechanical response of the gels was recorded by performing frequency sweep (0.1–10 Hz) measurements in a constant strain (0.05) mode, at 37 °C. The shear modulus (G') is reported as a measure of gel mechanical properties. To quantify the colony formation efficiency of single embedded ISCs, phase contrast z-stacks spanning the entire thickness of the cell-laden Matrigel or PEG gels were collected (Zeiss Axio Observer Z1) at 5 different locations within the gels. The Cell Counter plugin in ImageJ (NIH) was used to quantify the fraction of cells which had formed colonies at day 4 after seeding. To quantify colony circularity, phase contrast images of ISC colonies grown in the condition of interest (between 5 and 38 colonies per condition) were taken, and their contours traced manually in ImageJ. The circularity of the contours was measured using the Measure algorithm in ImageJ. To characterize the cell morphology within ISC colonies grown in different conditions, phase contrast images of at least 50 colonies were taken and the numbers of colonies containing packed columnar cells versus spread cells were counted. To quantify Lgr5 expression within ISC colonies grown within different matrices, fluorescence images of at least 50 colonies per condition were recorded and the number of colonies expressing Lgr5–eGFP was counted. To identify a short sequence that supports intestinal organoid culture, we created a library of soft (G' = 200 Pa) hydrogels in which binding sequences from the laminin α1 subunit previously shown to be biofunctional33, 34 (Extended Data Table 4) were tethered to the PEG backbone. Embedding fragments of pre-formed organoids and screening the library for organoid survival and growth revealed that two laminin-derived peptides—AG73 and A55—significantly enhanced organoid viability and supported further growth (Extended Data Fig. 5a). Presenting these two sequences (A55 and AG73) alongside in the same gel did not appear to have an additive effect, likely owing to a redundant adhesion mechanism. Hence, we focused on the sequence with a stronger individual effect, that is, AG73 and the corresponding PEG gels (referred to as TG PEG-AG73). Varying the amount of AG73 peptide tethered to the PEG gel backbone revealed a dose-dependent effect on intestinal organoid viability and growth (Extended Data Fig. 5b, c). Despite the improved rate of survival and morphogenesis in TG PEG-AG73 matrices compared with plain PEG or PEG RGD, the process was significantly less efficient compared with Matrigel, and morphological differences were apparent. Keeping in mind that the effect of AG73 was concentration-dependent, we attributed these differences to a potentially sub-optimal AG73 ligand density within the synthetic system. By design, there is an upper limit to the concentration of tethered factors that can be incorporated into the PEG system used thus far in the study: exceeding this limit disrupts the structural integrity of the gels. To overcome this limitation and enhance the biofunctionality of the synthetic matrix by increasing the concentration of AG73 ligands, we turned to chemically crosslinked PEG gels. Here, vinyl sulfone (VS)-conjugated 4-arm PEG precursors are covalently linked into solid hydrogels through Michael-type addition between VS groups and the thiols of a short crosslinker containing two cysteine residues. To incorporate the AG73 ligand at a high density, we designed a crosslinker in which the AG73 sequence was flanked by two short cysteine-containing sequences. The resulting gels (hereafter referred to as MT PEG-AG73) presented the AG73 ligand at a concentration of 3.1 mM, thus significantly surpassing the highest concentration achieved in the enzymatically crosslinked matrices. Embedding intestinal organoid fragments into MT PEG-AG73 revealed that the percentage of tissues that remained viable and continued to undergo morphogenesis approached that observed in Matrigel (Extended Data Fig. 5e). To verify the maintenance of ISCs within the organoids grown in MT PEG-AG73, we embedded tissues extracted from the Lgr5-eGFP reporter mouse and monitored eGFP expression. We observed that Lgr5–eGFP was expressed in the expected pattern: localized to the crypt-like buds of the organoids (Extended Data Fig. 5f). The fraction of organoids expressing the marker was significantly higher than in those cultured in TG PEG AG73, and at least as high as in organoids cultured in Matrigel (Extended Data Fig. 5g). We also confirmed that the organoids cultured in MT PEG-AG73 were polarized and contained differentiated cells (Extended Data Fig. 5h). ISC colonies or organoids embedded in Matrigel or PEG gels were fixed with 4% paraformaldehyde (PFA) in PBS (30 min, room temperature). The fixation process typically led to complete degradation of the Matrigel. Suspended tissues were collected and centrifuged (800 r.p.m., 5 min) to remove the PFA, washed with ultra-pure water and pelleted. Following resuspension in water, the organoids were spread on glass slides and allowed to attach by drying. Attached organoids were rehydrated with PBS. Following fixation, organoids embedded in PEG or spread on glass were permeabilized with 0.2% Triton X-100 in PBS (1 h, room temperature) and blocked (10% goat serum in PBS containing 0.01% Triton X-100) for at least 3 h. Samples were subsequently incubated overnight at 4 °C with phalloidin-Alexa 546 (Invitrogen) and primary antibodies against lysozyme (1:50; Thermo Scientific PA1-29680), mucin-2 (1:50; Santa Cruz sc-15334), chromogranin-A (1:50; Santa Cruz sc-13090), L-FABP (1:50; Santa Cruz sc-50380) and YAP1 (1:50; Santa Cruz sc-101199) diluted in blocking buffer. After washing with PBS for at least 3 h, samples were incubated overnight at 4 °C with secondary antibody Alexa 647 goat-α-rabbit (1:1000 in blocking solution; Invitrogen). Following extensive washing, stained organoids were imaged in epifluorescence (Zeiss Axio Observer Z1) or confocal (Zeiss LSM 710) mode. Alternatively, ISC colonies or organoids cultured in PEG were released from the hydrogel before PFA fixation, by incubating the gels with 1 mg ml−1 Dispase (Gibco) for 7 min at 37 °C. The released colonies or organoids were fixed with PFA in suspension, and attached to glass coverslips, as described above. Human organoids were fixed in 10% neutral buffered formalin, washed with PBS, dehydrated, and embedded in paraffin. Sections were stained with H&E or Ki67 antibody (1:250; Monosan). Lentiviral particles encoding for shRNA recognizing YAP (two sequences validated for knockdown, purchased from Sigma) or the pLKO.1-puro Non-Target control shRNA (Sigma) were generated in HEK 293T cells, using third generation lentivirus packaging vectors. Transfection was carried out using the X-tremeGENE HP Transfection kit (Roche). After 48 h, the supernatant was collected, filtered and ultracentrifuged at 50,000g for 2 h at 20 °C. The resulting pellet was resuspended in PBS and stored at −80 °C. Lentiviral infection of ISCs was performed by dissociating the ISC colonies (described above), resuspending the resulting single cells in ice-cold liquid Matrigel, containing 10 μM Y27632 and the concentrated lentiviral particles at a dilution of 1:10. ISCs were incubated with viral particles in a liquid suspension for 45 min on ice. The suspension was subsequently cast into droplets and allowed to form gels, which were overlaid with ISC expansion medium. The embedded ISCs proceeded to form colonies, which carried the transgene encoded by the virus. The cells were allowed to recover and form colonies for 36 h, after which they were dissociated and encapsulated within PEG hydrogels or used for quantification of knockdown efficiency by qPCR. ISC colonies or organoids grown in Matrigel or PEG gels were dissociated as described above, and RNA was extracted using an RNeasy Micro Kit (Qiagen). cDNA was synthesized using the iScript cDNA Synthesis Kit (Bio-Rad). qPCR was carried out using the Power SYBR Green PCR Master Mix (Applied Biosystems) and the primers listed Extended Data Table 1. RNA-seq libraries were prepared using 500 ng of total RNA and the Illumina TruSeq Stranded mRNA reagents (Illumina; San Diego, California, USA) on a Sciclone liquid handling robot (PerkinElmer; Waltham, Massachusetts, USA) using a PerkinElmer-developed automated script. Cluster generation was performed with the resulting libraries using the Illumina TruSeq SR Cluster Kit v4 reagents and sequenced on the Illumina HiSeq 2500 using TruSeq SBS Kit v4 reagents. Sequencing data were processed using the Illumina Pipeline Software version 1.84. Purity-filtered reads were adapters and quality trimmed with Cutadapt and filtered for low complexity with seq_crumbs (v. 0.1.8). Reads were aligned against the Mus musculus.GRCm38.82 genome using STAR35. The number of read counts per gene locus was summarized with htseq-count36 using the Mus musculus.GRCm38.82 gene annotation. Quality of the RNA-seq data alignment was assessed using RSeQC37. Reads were also aligned to the Mus musculus.GRCm38.82 transcriptome using STAR and the estimation of the isoforms abundance was computed using RSEM38. Statistical analysis was performed for genes in the R software package. Genes with low counts were filtered out according to the rule of 1 count per million in at least 1 sample. Library sizes were scaled using TMM normalization in the EdgeR package39 and log-transformed with the limma voom function40. Differential expression was computed with limma41 by fitting paired samples data into a linear model and performing all pairwise comparisons. To control for false discovery and multiple testing, we computed an adjusted P value, using the Benjamini–Hochberg method. Gene set expression analysis was performed with the freely available GSEA software42 (Broad Institute), using differential expression values and pre-defined gene signatures as inputs. In particular, to test for upregulation of stress-related genes, the MSigDB gene set ‘BIOCARTA_STRESS_PATHWAY’ was used. We checked for upregulation of colon cancer-related genes by using the inflammatory colon cancer signature, as identified by Sadanandam et al.43. Functional annotation and gene ontology analysis of significantly enriched gene sets was conducted using the MetaCore software. The accession numbers for the gene expression profiles described here are GEO: GSE85391. Statistically significant differences between the means of two groups were assessed by using a Student’s t-test, whereas data containing more than two experimental groups were analysed with a one-way ANOVA followed by a Bonferroni’s multiple comparison test. All statistical analyses were performed in the GraphPad Prism 6.0 software. RNA sequencing data that support the findings of this paper have been deposited to the Gene Expression Omnibus (GEO) public repository (GSE85391; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? acc=GSE85391). Source Data for Figs. 1, 2, 3, 4 and Extended Data Figs 1, 2, 3, 4, 5 are provided with the paper. All additional relevant data are available upon request from the corresponding author.


The future laboratory animal technician in Europe will be provided with three different levels of education. All candidates have to start with an introductory course to reach level A0. At this level (A0) they will be able to assist in the laboratory animal facility by undertaking limited specific duties under supervision. Most A0 assistants will continue their education and training for at least one year while in full-time employment. This process will include continual assessment with the option of a final examination to become qualified at level A1. A1 represents a comprehensively educated laboratory animal technician with theoretical background knowledge and practical skills. Some of the A1 laboratory animal technicians may continue specific education for at least another year of full-time employment. They will develop knowledge and expertise as well as supervisory and basic managerial skills in order to obtain level A2.

Loading FELASA collaborators
Loading FELASA collaborators