 |
 |
GEO help: Mouse over screen elements for information. |
|
| Status |
Public on Nov 25, 2015 |
| Title |
MNase-seq T35 |
| Sample type |
SRA |
| |
|
| Source name |
red blood cell stage, 27-35 hours post invasion
|
| Organism |
Plasmodium falciparum |
| Characteristics |
strain: 3D7
developmental stage: medium schizont
selection: var2csa
|
| Growth protocol |
3D7 Plasmodium falciparum blood-stage parasites obtained from the Swiss Tropical and Public Health Institute (Basel, Switzerland - a kind gift from Dr. T. Voss) were cultured in standard RPMI medium supplemented with 10% human serum and 0.2% NaHCO3 in 2.5% human O+ red blood cells. 75 cm2 tissue culture flasks were incubated at 37ºC in candle jars under low oxygen conditions. Multiple rounds of panning were performed as described in (Noviyanti et al., 2001) with the following modifications, to obtain a parasite population homogenously expressing a single var gene (var2csa, PF3D7_1200600). In short, CSA coated plates were blocked using 1% casein sodium salt from bovine milk (Sigma C8654-500G) solution and washed three times with RPMI media before parasites were allowed to bind for 30 min at 37 ºC without agitation followed by 5 RPMI washes. Bound parasites were collected in RPMI media supplemented with serum and NaHCO3 and allowed to grow to ~5-10% parasitemia before the next round of panning was performed. Preferential expression of var2csa and silencing of all other var genes was verified by RT-qPCR prior to synchronization with multiple rounds of sorbitol treatments. After a final Percoll gradient centrifugation human red blood cells were added with a few hours delay as in (Bartfai et al., 2010), to prevent “premature” invasion and therefore ensure better synchronicity of the culture. The blood was pre-filtered twice over sterile PlasmoPure (EuroProxima) filters to remove human white blood cells. T0 was set as the time when the first invasions were observed, resulting in an ~8 hour synchronicity window for >95% of parasites. Cross-linked nuclei (for MNase-Seq) and RNA (for RNA-Seq) were collected every 5 hours. Medium was changed at least every 10 hours, but not within 10 hours from collection. For T25 and T30 collections culture volume was doubled at the T9, for T35 and T40 collections at the T18 media change, respectively. Cultures were divided over separate culture flasks (20ml each), but mixed upon every media change.
|
| Extracted molecule |
genomic DNA |
| Extraction protocol |
Cross-linked nuclei and RNA were collected from the pooled culture flasks of the same batch of synchronized parasites to ensure completely matched nucleosome-positioning and transcriptome. MNase-seq and pellet Control: Formaldehyde was immediately added to 37 ºC parasite cultures to a 1% final concentration, followed by 15 min incubation at 37 ºC while shaking. Cross-linking reaction was quenched by addition of 0.125M glycine (final concentration) and all subsequent steps were performed at 4 ºC unless noted otherwise. Parasites were isolated from red blood cells by 0.05% saponin lysis, followed by the isolation of parasite nuclei on a 0.25 M sucrose cushion using cell lysis buffer (10 mM Tris pH8, 3 mM MgCl2, 0.2% NP40, Roche Protease Inhibitor Cocktail). Isolated nuclei were divided over multiple aliquots, which were used to determine the optimal digestion time for each time-point. Enzymatic digestion of the chromatin was performed in digestion buffer (50 mM Tris pH7.4, 4 mM MgCl2, 1mM CaCl2, 0.075% NP40, 1mM DTT, Roche Protease Inhibitor Cocktail) with 0.5 U MNase and 100 U Exonuclease III in 150 µl aliquots for 3.5-14 min at a 37 ºC waterbath with regular agitation. Digestion reactions were stopped by addition of 1 reaction volume quenching solution (2% Triton X100, 0.6% SDS, 300 mM NaCl, 6 mM EDTA, Roche Protease Inhibitor Cocktail) and placed at 4 ºC. Nuclei were mildly sonicated for 6x 10 sec (setting low, BioruptorTM Next Gen, Diagenode) to free the cross-linked nucleosomes from the nuclear membranes and the supernatant containing soluble chromatin was collected after centrifugation for 10 min at 9800 x g (this is the X-linked chromatin used for MNase-Seq input and α-H4 ChIP reactions). Nuclear pellets were decrosslinked (see conditions below) to assess whether the material residing with the nuclear membranes is equally distributed over the genome or enriched for certain genomic regions (pellet control). DNA yield from the pellet of each time-point was quantified and DNA size distribution was confirmed to be marginally higher in size-range to MNase digested soluble chromatin. Since for the T15 time-point the highest percentage of chromatin remained with the nuclear membranes, this sample was selected for sequencing as the pellet control. MNase-Seq chromatin and nuclear pellets were decrosslinked overnight at a 45 ºC shaking heatblock in decrosslinking buffer (1% SDS, 0.1 M NaHCO3, 1 M NaCl) after which DNA was isolated via QIAquick column purification (Qiagen). Digestion efficiency was verified on 2% agarose gel. Importantly, to enable, as much as possible, differentiation of technical variation from biological variation between the different samples, a technical replicate was included for the T40 time-point, where two different aliquots from the same nuclear collection were independently digested and processed for MNase-Seq. These technical replicates were called T40A and T40B and for most analysis sequence reads from both samples were combined (T40 sample). Chromatin Immunoprecipitation: 200ng X-linked T40-stage chromatin (from sample T40A) in ChIP buffer (20 mM Tris pH8, 2 mM EDTA, 1% Triton-X100, 0.15% SDS, 150 mM NaCl, Roche Protease Inhibitor Cocktail) was incubated with 1 µg α-H4 core antibody (Abcam Ab17036, lot GR8733-1) or 1 µg normal rabbit IgG control antibody (Upstate 12-370, lot DAM1421465) O/N while rotating at 4 ºC. 10 µl ProtA Dynabeads (Life Technologies 10008D) and 10 µl ProtG Dynabeads (Life Technologies 10009D) were added to each reaction and allowed to rotate for another 2 hours at 4 ºC, after which ChIPs were washed with 400 µl wash buffer: 1x ChIP wash 1 (20 mM Tris pH8, 2 mM EDTA, 1% Triton-X100, 0.1% SDS, 150 mM NaCl), 2x ChIP wash 2 (20 mM Tris pH8, 2 mM EDTA, 1% Triton-X100, 0.1% SDS, 500 mM NaCl), 2x ChIP wash 3 (10 mM Tris pH8, 1 mM EDTA). Reactions were rotated for 5 min at 4 ºC in between each wash. Immunoprecipitated chromatin was eluted in elution buffer (1% SDS, 0.1M NaHCO3) during 20 min rotation at RT and decrosslinked as described above (“MNase-Seq and pellet control” section). ChIP reaction efficiency with α-H4 core and control antibody was verified by qPCR. Genomic DNA extraction and fragmentation: Genomic DNA was extracted from sorbitol synchronized ring stage 3D7 Plasmodium falciparum cultures of the same strain as was used for MNase-Seq. Cultures were PlasmoPure (EuroProxima) filtered before collection. Native nuclei collection, proteinase K treatment and genomic DNA isolation were performed as described in (Bartfai et al., 2010) with the following modifications. An RNaseA incubation step ensured degradation of RNA. gDNA was fragmented using a BioruptorTM Next Gen (Diagenode) to a ~100-400 bp range. RNA isolation: Total RNA isolation and oligo-dT-selection to enrich for polyA+ mRNA were performed as described elsewhere (Hoeijmakers et al., 2013a) with addition of an extra in-solution DNase treatment (TURBO DNase, Ambion) to remove contaminating gDNA. Integrity of total RNA was confirmed on 1.5% agarose gel and gDNA contamination was tested by qPCR.
MNase/gDNA Sequencing library preparation: 10 ng of MNase-Seq inputs, gDNA control, T15 pellet control and 5.7 ng of T40 α-H4 ChIP DNA were used to generate sequencing libraries. All samples were first end repaired, a 3’ A-overhang added and NextFlex barcoded adapters (Bio Scientific) were as described in (Hoeijmakers et al., 2011). Adapter ligation was followed by 2 successive Agencourt AMPure XP bead purifications (Beckman Coulter) and library amplification by an in-house Plasmodium-optimized kapa PCR amplication protocol. 2x kapa HiFi HotStart ready-mix (KAPA Biosystems) and NextFlex primers were used for 9 cycles of PCR amplification using the following conditions: 98 ºC for 2 min, (9 cycles of: 98 ºC for 20 sec, 62 ºC for 3 min), 62 ºC for 5 min. Libraries were again Agencourt AMPure XP bead purified before sequencing. Importantly, no size-selection step was applied to the MNase-Seq input and control libraries allowing assessment of the full range of DNA fragments resulting from enzymatic chromatin digestion and in silico size-selection of paired-end sequenced libraries. cDNA synthesis and preparation of strand-specific RNA-seq libraries: 2 µg of total RNA-equivalent polyA+ mRNA was fragmented by hydrolysis as described in (Hoeijmakers et al., 2013a). cDNA synthesis was modified as follows from (Hoeijmakers et al., 2013a) to allow maintenance of directional information for strand-specific RNA-Seq. During first strand synthesis 0.2 µg Actinomycin D was included in the reaction and a 15 min 70 ºC enzyme deactivation step was included after first strand synthesis followed by QIAquick MinElute purification (Qiagen) of the first strand cDNA. During second strand synthesis dTTP was replaced by dUTPs, resulting in incorporation of U-bases instead of T-bases in the second cDNA strand only. 4 ng double-stranded cDNA was used for sequencing library preparation as described above (“Sequencing library preparation” section) with the following adaptation: after adapter ligation a 15 min incubation at 37 ºC with USER enzyme (New England Biolabs) ensured degradation of the second cDNA strand specifically (by degrading the U-base containing strand) thereby retaining directional information for sequencing. Subsequently, 4 cycles of Plasmodium-optimized kapa pre-PCR (for details see “Sequencing library preparation” section) were performed, followed by size-selection of 200-300bp cDNA fragments and subsequent amplification of another 9 cycles of Plasmodium-optimized kapa PCR resulting in total 13 cycles of PCR amplification.
|
| |
|
| Library strategy |
MNase-Seq |
| Library source |
genomic |
| Library selection |
MNase |
| Instrument model |
Illumina HiSeq 2000 |
| |
|
| Data processing |
Illumina Casava 1.8.2 software for base calling. We only used the bcl to fast conversion.
The single-end RNA-seq samples were mapped with BWA (Version 0.6.2-r126, default parameters) to either the complete P. falciparum assembly for the genomic tracks or the spliced transcriptome for transcript quantification (PlasmoDB 6.1) and filtered to mapping quality ≥15. Between 13.9 and 24.9 million reads mapped to transcripts.
Paired-end reads were clipped to 72bp (FASTX Toolkit, 0.0.13.2) and mapped with BWA (0.6.2-r126) to the PlasmoDB 6.1 assembly. We decided to trim MNase-seq reads to 72bp, to be able to map sub-nucleosomal fragments shorter than 100bp, but retain decent mappability of the sequence reads. The paired-end mapping was done such that up to 1000 alternative alignment positions were reported for each read (bwa sampe –n 1000). Subsequently, read pairs aligning with mapping quality below 30 or aligning to more than a single position with alignment pair distances below 1 kb were discarded using a custom script.
Genome_build: PlasmoDB 6.1
Supplementary_files_format_and_content: .raw.bedGraphFull: Raw coverage by full length fragments for paired-end or 92 bp reads for single-end. .scale1.bedGraphFull: Normalized coverage per site (ni) calculated as ni = ri × g/b, where ri is the raw coverage (.raw.bedGraphFull) covering the sites i in the sample, b is the total number of bases in fragments, and g is the genome size. Therefore, a normalized coverage value of 1 represents the coverage expected if fragments are randomly distributed across the genome. .scaled.*.scale1.bedGraphFull: Normalized RNA-seq coverage as in .scale1.bedGraphFull multiplied by scaling constants representing the estimated total RNA content per cell. gDNArs*: As a means to control for amplification and sequencing biases we sequenced a sonicated genomic DNA sample, which unavoidably has a different distribution of fragment lengths than MNase-digested and nucleosome-protected DNA. To avoid potential biases associated with such differences in the insert-size distributions (ISD), we transformed the ISD of the gDNA control using a sampling approach. For every MNase-seq sample we created a resampled gDNA control with a matching ISD. To that aim, for every insert-size the number of fragments was calculated based on the target ISD and a desired total number of 150 million fragments. Fragments were then randomly selected from the fragments of that length out of the 274 million gDNA fragments. Samples T40 & Tall: T40 represents the pooled fragments of samples T40A-4511 and T40B-4513. Tall represents the pooled fragments of T5-4504, T10-4505, T15-4506, T20-4507, T25-4508, T30-4509, T35-4510, and T40.
|
| |
|
| Submission date |
Feb 23, 2015 |
| Last update date |
May 15, 2019 |
| Contact name |
Richard Csaba Bartfai |
| Organization name |
Radboud University
|
| Department |
Department of Molecular Biology
|
| Street address |
Geert Groteplein 26-26
|
| City |
Nijmegen |
| ZIP/Postal code |
6525 GA |
| Country |
Netherlands |
| |
|
| Platform ID |
GPL16607 |
| Series (1) |
| GSE66185 |
The Nucleosome Landscape of P. falciparum Reveals Chromatin Architecture and Dynamics of Regulatory Sequences |
|
| Relations |
| BioSample |
SAMN03366073 |
| SRA |
SRX885817 |
| Supplementary file |
Size |
Download |
File type/resource |
| GSM1616490_MNaseSeq_T35-4510.Plfal_3D7_6.1.bwa.unique.mapq30.submononuc.raw.bedGraph.gz |
70.7 Mb |
(ftp)(http) |
BEDGRAPH |
| GSM1616490_MNaseSeq_T35-4510.Plfal_3D7_6.1.bwa.unique.mapq30.submononuc.scale1.bedGraph.gz |
75.4 Mb |
(ftp)(http) |
BEDGRAPH |
SRA Run Selector |
| Processed data provided as supplementary file |
| Raw data are available in SRA |
|
|
|
|
 |
