nucleic acid sequencing protocol
Fragments were sequenced using the Illumina HiSeq 2000
Reads were extended post-alignment in silico in their direction, to a final length equal to the fragment size estimation based on MACS (Zhang et al., 2008). Enriched regions were identified using SICER v1.03 (Zang et al., 2009) in 100bp windows, allowing a 200 bp gap for H3K4me3 and 400 bp for H3K36me3 and RNA Pol II. To eliminate PCR amplification clonal artifacts only unique reads were used to call enrichments. To control for systematic biases, a dataset derived by sequencing unenriched input DNA was used to define a background model. Only peaks at FDR<10?3 over background were retained. A post-processing step was applied in order to consolidate all H3K4me3 and H3K36me3 sites of enrichment, whereby we retained only those sites that were called in the sample with strongest enrichment values and furthermore overlapped a minimum of one base with a site called in at least one of the replicates.
All the reads overlapping a given gene were counted, and then counts were normalized by the length of the reads and by the total amount of unically aligned reads in the experiment, and multiplied by 10^9
Reads were aligned allowing only 1 mismatch per read and no multi-mapping (default parameters plus -g 1 -r 250 --coverage-search --butterfly-search), using --library-type fr-firststrand flag for strand determination when needed. The rRNA-depleted reads were mapped using Bowtie (Langmead et al., 2009), allowing for up to 3 mismatches and no multi-mapping (default parameters and -n 3 --nomaqround, with --norc and --nofw flags for determining the strand).
Illumina GAIIx flow cell preparation: GD-300-2001: TruSeq SR Cluster Kit v2 - cBot - GA; FC-104-5001: TruSeq SBS Kit v5 - GA (36-cycle).. Sequencing Control Studio 2.8
Reads were aligned allowing 1 mismatch per read and no multi-mapping
nucleic acid library construction protocol
RNA was extracted using Trizol (Invitrogen). The RNA integrity number (RIN) was assessed with a 2100 Agilent Bioanalyzer to verify RNA quality for all tested samples. RNA was treated with DNAseI (Sigma) for an extended period of 20 minutes at room temperature to prevent genomic DNA contamination following manufacturers instructions. All samples are sequenced with the Illumina Paired-end mRNA-Seq Sample Prep Kit. Samples sequenced using HiSeq200 will have one adaptation to this protocol; instead of the Illumina polyA selection, NEB (New England Biolabs) polyA selection has been used. Human PolyA+ RNA samples were used to construct random hexamer-primed non-directional cDNA libraries, whereas mouse islet PolyA+ RNA was used for unidirectional cDNA libraries, and were subsequently used for paired-end sequencing in GAIIx or HiSeq2000 Genome Analyzers. SOLiD Total RNA-Seq Kit (P/N 4445374): RNA was ribosomal depleted using the Invitrogen RiboMinus Eukaryote Kit for RNA-Seq (A1083708). RNA fragmentation by RNAse III or chemical digestion followed protocol. Libraries were barcoded using the SOLiD RNA barcoding kit (P/N 4427046) and pooled in equimolar fashion
SOLiD analysis performed using BioScope with whole transcriptome pipeline.
Emulsion for the pooled library prepared using the SOLiD EZ Bead system components as following the protocol described for E80 scale preparation. SOLiD EZ Bead E80 Systems Consumables, 4 Pack (P/N 4453095); SOLiD EZ Bead Oil Pack Kit (P/N 4457185). Templated beads deposited onto a single slide for full array sequencing. Barcode paired end sequencing performed with the SOLiD ToP Paired end sequencing kit (P/N 4459182). SOLiD 4 system software
All experiments were performed according to protocols approved by the Institutional ethical committees of the Hospital Clinic de Barcelona, Geneva University Hospitals, Istituto Scientifico Ospedale San Raffaele, Oxford University, UK Human Tissue Authority and University of Lille. Samples were isolated from multiorgan donors without a history of glucose intolerance after informed consent from family members. Donor variables and cold ischemia times of the samples used in this study are provided separately. Pancreatic islets were isolated and purified according to established isolation procedures. After isolation, islets were incubated at 37 degrees C in a humidified chamber with 5% CO2 in CMRL 1066 medium with 10% fetal calf serum for various periods prior to shipment at room temperature in the same culture medium. Upon arrival, islets were re-cultured at 37 degrees C in a humidified chamber with 5% CO2 in RPMI 1640 medium supplemented with 10% fetal calf serum, 100 U/ml penicillin, and 100 U/ml streptomycin for three days before extraction of RNA or chromatin. Islet purity was assessed by dithizone staining using an aliquot of islets immediately prior to harvest. Further assessment of islet purity was carried out by qPCR analysis of duct, acinar, and beta-cell specific markers (SOX9, CPA1, MIST1, NKX6.1, and INS). Only samples showing marginal exocrine contaminant mRNAs were further processed. Acinar-enriched fractions were obtained after gradient purification of islets.
nucleic acid library construction protocol
Human islets were fixed and sonicated as described previously (Nammo et al, 2012). Briefly, frozen crosslinked cell pellets of ~4,000 islets were thawed on ice in 1 mL lysis buffer and disrupted with five 1-min cycles with 0.5 mm glass beads (BioSpec). Samples were sonicated for 10-20 cycles of 30 pulses each (1s on and 0.5 s off) using a Brandson Sonifier 450D at 15 % amplitude. The size for the bulk of chromatin obtained with this procedure was checked to be in the range from 200-1000 bp. For ChIPs, around 300-400 human islet equivalents (IEQs) were pre-cleared with A/G sepharose beads (Ge Healthcare) for 1 h, rotating at 4 degrees C, and then incubated with each antibody. Afterwards, samples were rotated for 2 hr at 4 degrees C with protein A/G sepharose beads and then sequentially washed with low salt, high salt, LiCl and TE buffers. Next, chromatin was eluted with SDS 1% buffer and sequentially treated with RNAse (greater than 5 hr at 65 degrees C) and Proteinase K (overnight at 45 degrees C). Finally, DNA was extracted with phenol-chloroform followed by ethanol precipitation. The immunoprecipitated DNA was modified for sequencing following the Illumina sequencing following the manufacturer protocol (Preparing Sample for ChIP Sequencing of DNA (11257047 Rev A)).
All animal experiments were conducted as approved by the Ethical Committee of Animal Experimentation of our institutions and in accordance with national regulations. Islets were obtained from 10-14 week-old male C57BL/6J mice. For sequencing we obtained RNA from islets after culture at 37 degrees C in a humidified chamber with 5% CO2 for 72 hr in RPMI 1640 medium with 11 mM Glucose supplemented with 10% fetal calf serum, 100 U/ml penicillin, and 100 U/ml streptomycin.
Human beta-cells were further purified from isolated islets by Fluorescence Activated Cell Sorting using the zinc dye Newport Green and exclusion of duct and dead cells. beta-cell purity was ascertained by immunofluorescence, using polyclonal guinea pig anti-insulin antiserum (1:500), followed by texas red-conjugated goat anti-guinea pig antiserum (1:50, Vector Laboratories, Peterborough, UK) in conjunction with nuclear staining with DAPI (Vector Laboratories). Images were captured using a Zeiss LSM510 confocal laser scanning microscope (Carl Zeiss, Jena, Germany). Preparations with greater than 90% insulin-positive cells were used for sequencing.