Protocol for the genomic analysis of salt-fractionated chromatin from frozen murine liver

Summary

Salt fractionation is a classical approach for separating chromatin based on its differential salt solubility and physical properties. Here, we present a protocol to apply salt fractionation for genome-scale profiling of chromatin isolated from livers at different stages of aging in mice. We elaborate on the steps to isolate nuclei, digest with micrococcal nuclease, sequentially salt fractionate, purify DNA, and construct libraries for genome profiling. We also include information on a computational pipeline for data analysis.

For complete details on the use and execution of this protocol, please refer to Yang et al.¹ This protocol is an adaptation of the salt fractionation method of Teves and Henikoff.²

Graphical abstract

Highlights

• •Protocol for genomic analysis of fractionated chromatin from liver nuclei
• •Steps for nuclei isolation, nuclease digestion, salt fractionation, and DNA purification
• •Instructions for library construction, sequencing, and computational analysis

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Salt fractionation is a classical approach for separating chromatin based on its differential salt solubility and physical properties. Here, we present a protocol to apply salt fractionation for genome-scale profiling of chromatin isolated from livers at different stages of aging in mice. We elaborate on the steps to isolate nuclei, micrococcal nuclease digest, sequentially salt fractionate, purify DNA, and construct libraries for genome profiling. We also include information on a computational pipeline for data analysis.

Before you begin

The protocol below describes the specific steps to fractionate chromatin from frozen mouse livers based on their salt solubility and to generate sequencing libraries (Figure 1A). However, this protocol may be extended to other organs provided the nuclei isolation steps are optimized for that organ.

Figure 1.

Expected DNA profiles in a salt fractionation experiment

(A) Schematic of a salt fractionation experiment followed by sequencing.

(B) Representative Bioanalyzer traces of salt fractionated DNA from a representative young liver sample.

(C) Representative Bioanalyzer traces of purified DNA libraries generated from 250 mM salt extracted chromatin from a young and old mouse liver. For (B-C), FU represents fluorescence units.

(D) Genome browser snapshots of salt fraction enrichments on chr5. Inset 1 is euchromatinized and inset 2 is heterochromatinized with age and expanded on the right.

(E) Same as (D) except from chr18. This figure is adapted from Yang et al.¹

Institutional permissions

Animals used in this protocol were maintained in the NIA vivarium under Animal Study Protocol 481-LGG-2025. All procedures conform to the regulatory standards established and approved by the NIA Animal Care and Use Committee.

Prepare tissue

Timing: day 1, 30 min

• 1.Transport mice to necropsy area.
• 2.Sacrifice mice by carbon dioxide asphyxiation followed by cervical dislocation.
• 3.Rapidly dissect mice and harvest organ of choice (in this case, the liver).
• 4.Wash the dissected organ thoroughly in phosphate buffered saline (PBS) in a Petri dish and cut into 20–50 mg pieces.
• 5.Use forceps to dab each washed piece of tissue on a Kim wipe and transfer to a microcentrifuge tube.
• 6.Freeze the tube containing tissue on isopentane over liquid nitrogen.
• 7.Store the tissue long-term in a −80°C freezer.
• 8.In this protocol, take an equal amount of tissue (∼20 mg) from both animal groups including 3-month-old young mice and 18-month-old old mice.

Order reagents/kits and prepare equipment

Timing: day 1, 1 h

• 9.Order all reagents and kits (see key resources table for list) and store at appropriate temperatures upon receipt. Ensure kits/enzymes have not expired before use and that there are adequate amounts based on the number of samples. We advise that the micrococcal nuclease (MNase) be bought from the vendor listed for reproducible results.
• 10.Wear appropriate personal protective equipment such as a lab coat, gloves, mask, and eyewear.
• 11.
  Gather the following equipment/consumables and place them at the appropriate location/temperature. See the materials and equipment section and key resources table for a list of equipment used in this protocol.

  • a.
    Place a tube rotator in the cold room.
  • b.
    Pre-chill microcentrifuge to 4°C.
  • c.
    Fill two buckets with ice and set on a clean lab bench.
  • d.
    Chill Dounce homogenizer in one of the ice buckets.
  • e.
    Gather low-adhesion 1.5 mL microcentrifuge tubes, 15 mL tubes, pipettes, tips, and Kim wipes and set on lab bench.
  • f.
    Have a bench-top vortexer in the vicinity.
  • g.
    Ensure thermocycler is programmed for each step of the library construction process (as outlined in the NEBNext Ultra II kit manual).
  • h.
    Right before start, transport tubes containing tissue on dry ice to lab bench in an appropriate carrier.

Prepare solutions

Timing: day 1, 2 h

• 12.
  Prepare the buffers listed below. See materials and equipment section for buffer recipes.

  • a.
    TM2 buffer.
  • b.
    0 mM Triton buffer.
  • c.
    67.5 mM Triton buffer.
  • d.
    150 mM Triton buffer.
  • e.
    250 mM Triton buffer.
  • f.
    350 mM Triton buffer.
  • g.
    TNE buffer.

CRITICAL: Prepare all buffers fresh on the day of the salt fractionation. 1 M MgCl₂ stock solutions should also be freshly prepared while other stock solutions may be prepared up to a week in advance. All stock solutions should be filtered using a 0.2 μm filter. Alternatively, stock solutions (except 1 M MgCl₂ and CaCl₂) can be bought from commercial sources. Protease and phosphatase inhibitors should be added just before use.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Experimental models: Organisms/strains
C57BL/6JN mice (Mus musculus) Age: young (11 week), old (81 week) Sex: female	NIA Rodent Colony
Biological samples
Mouse liver	C57BL/6JN mice
Chemicals, peptides, and recombinant proteins
Ultrapure distilled water, DNase, RNase-free	Thermo Fisher Scientific	10977-015
IGEPAL CA-630	MilliporeSigma	I8896
1 M Tris-HCl pH 7.5	Quality Biological	351-006-101
Magnesium chloride hexahydrate	MilliporeSigma	M2670
(100×) Halt protease and phosphatase inhibitor cocktail	Thermo Fisher Scientific	78446
PMSF	MilliporeSigma	329-98-6
5 M sodium chloride	Quality Biological	351-036-101
EGTA	MilliporeSigma	03777
0.5 M EDTA	Thermo Fisher Scientific	15575-038
Triton X-100 solution	MilliporeSigma	93443
Micrococcal nuclease (2,000 U/μL)	New England Biolabs	M0247S
Calcium chloride	MilliporeSigma	C1016
10% SDS, RNase-free	Thermo Fisher Scientific	AM9823
RNase A, DNase, and protease-free (10 mg/mL)	Thermo Fisher Scientific	EN0531
Proteinase K solution (20 mg/mL), RNA grade	Thermo Fisher Scientific	100005393
SPRI beads	Beckman Coulter	B23318
Critical commercial assays
DNA Clean & Concentrator-5	Zymo Research	D4003
NEBNext Ultra II DNA Library Prep Kit for Illumina	New England Biolabs	E7645L
NEBNext Multiplex oligos for Illumina (96 unique dual index primer pairs)	New England Biolabs	E6440L
NEBNext Library Quant Kit for Illumina	New England Biolabs	E7630L
DNA HS kit	Agilent	5067-4626
DNA 1000 kit	Agilent	5067-1504
Qubit 1× dsDNA, High Sensitivity Assay Kit	Thermo Fisher Scientific	Q33231
Deposited data
Salt fraction DNA-seq	This paper	GEO: GSE185707
Software and algorithms
bcl2fastq2		https://support.illumina.com/sequencing/sequencing_software/bcl2fastq-conversion-software.html
BCL convert		https://sapac.support.illumina.com/sequencing/sequencing_software/bcl-convert.html
Cutadapt		https://cutadapt.readthedocs.io/en/stable/
FastQC		https://www.bioinformatics.babraham.ac.uk/projects/fastqc/
bowtie/2–2.4.2	Langmead and Salzberg3	http://bowtie-bio.sourceforge.net/index.shtml
samtools/1.10	Li et al.4	http://www.htslib.org/doc/samtools.html
picard/2.20.8		https://broadinstitute.github.io/picard/
Sambamba/0.7.1	Tarasov et al.5	https://github.com/biod/sambamba/releases
deepTools/3.5.0	Ramirez et al.6	https://deeptools.readthedocs.io/en/develop/
DANPOS	Chen et al.7
Other
Dounce homogenizer set (5 mL)	Kimble Chase	885302-0002
Tube rotator (Labquake shaker)	Thermo Fisher Scientific	13-687-12Q
Refrigerated microcentrifuge (Eppendorf model 5430R)	Fisher Scientific	13-690-005
Vortexer (BenchMixer)	Southern Labware	BV1000
Thermomixer (Eppendorf Thermomixer C)	Fisher Scientific	05-412-503PM
Thermocycler (ProFlex PCR system)	Thermo Fisher Scientific	4484073
Bioanalyzer (Agilent 2100 Bioanalyzer)	Agilent	G2939BA
Qubit 4 fluorometer and 1× dsDNA HS assay reagent	Thermo Fisher Scientific	Q33238
Magnetic rack	Diagenode	B04000001

Materials and equipment

Buffer recipes

TM2 buffer

Stock	Final concentration	Amount
1 M Tris–HCl, pH 7.5	10 mM	0.5 mL
1 M MgCl2	2 mM	0.1 mL
0.1 M PMSF∗ in isopropanol	0.5 mM	0.25 mL
100× Halt protease and phosphatase inhibitor	1×	0.5 mL
nuclease-free water	-	48.65 mL
Total	-	50 mL

0 mM Triton buffer

Stock	Final concentration	Amount
1 M Tris–HCl, pH 7.5	10 mM	10 μL
1 M MgCl2	2 mM	2 μL
0.5 M EGTA	2 mM	4 μL
10% Triton X-100∗	0.1%	10 μL
0.1 M PMSF∗ in isopropanol	0.5 mM	5 μL
100× Halt protease and phosphatase inhibitor	1×	10 μL
nuclease-free water	-	959 μL
Total	-	1 mL

67.5 mM Triton buffer

Stock	Final concentration	Amount
1 M Tris–HCl, pH 7.5	10 mM	10 μL
5 M NaCl	67.5 mM	13.5 μL
1 M MgCl2	2 mM	2 μL
0.5 M EGTA	2 mMs	4 μL
10% Triton X-100∗	0.1%	10 μL
0.1 M PMSF∗ in isopropanol	0.5 mM	5 μL
100× Halt protease and phosphatase inhibitor	1×	10 μL
nuclease-free water	-	945.5 μL
Total	-	1 mL

150 mM Triton buffer

Stock	Final concentration	Amount
1 M Tris–HCl, pH 7.5	10 mM	10 μL
5 M NaCl	150 mM	27 μL
1 M MgCl2	2 mM	2 μL
0.5 M EGTA	2 mM	4 μL
10% Triton X-100∗	0.1%	10 μL
0.1 M PMSF∗ in isopropanol	0.5 mM	5 μL
100× Halt protease and phosphatase inhibitor	1×	10 μL
nuclease-free water	-	932 μL
Total	-	1 mL

250 mM Triton buffer

Stock	Final concentration	Amount
1 M Tris–HCl, pH 7.5	10 mM	10 μL
5 M NaCl	250 mM	50 μL
1 M MgCl2	2 mM	2 μL
0.5 M EGTA	2 mM	4 μL
10% Triton X-100∗	0.1%	10 μL
0.1 M PMSF∗ in isopropanol	0.5 mM	5 μL
100× Halt protease and phosphatase inhibitor	1×	10 μL
nuclease-free water	-	909 μL
Total	-	1 mL

350 mM Triton buffer

Stock	Final concentration	Amount
1 M Tris–HCl, pH 7.5	10 mM	10 μL
5 M NaCl	350 mM	70 μL
1 M MgCl2	2 mM	2 μL
0.5 M EGTA	2 mM	4 μL
10% Triton X-100∗	0.1%	10 μL
0.1 M PMSF∗ in isopropanol	0.5 mM	5 μL
100× Halt protease and phosphatase inhibitor	1×	10 μL
nuclease-free water	-	889 μL
Total	-	1 mL

TNE buffer

Stock	Final concentration	Amount
1 M Tris–HCl, pH 7.5	10 mM	10 μL
5 M NaCl	200 mM	40 μL
0.5 M EDTA, pH 8.0∗	1 mM	2 μL
nuclease-free water	-	948 μL
Total	-	1 mL

• •0.2 M CaCl₂: add 0.022 g CaCl₂ in 1 mL nuclease-free water, filter through 0.2 μm filter.
• •1 M MgCl₂: add 0.203 g MgCl₂ in 1 mL nuclease-free water, filter through 0.2 μm filter.
• •10% NP-40: add 10 mL 100% Nonidet P-40 (NP-40) substitute (IGEPAL) in 100 mL nuclease-free water, filter through 0.2 μm filter.

Note: Store stock solutions at 24°C for up to a week. Store 0.1 M PMSF and Halt protease and phosphatase inhibitor at 4°C. Warm PMSF in 37°C water bath making sure all crystals disappear before adding to solutions. All working solutions are best prepared fresh.

CRITICAL: Toxic chemicals are marked with an asterisk. PMSF is corrosive, destructive to tissues, and toxic if swallowed. PMSF hydrolyzes upon exposure to water/moisture, liberating hydrogen fluoride that in contact with metal surfaces can generate flammable and/or explosive hydrogen gas. EDTA may cause acute inhalation toxicity. Triton X-100 causes acute oral toxicity and serious eye damage. PPE must be worn while making solutions. Avoid contact with eyes, skin, or clothing.

Alternatives: Thermomixer, thermocycler, and refrigerated microcentrifuge from other vendors can be used provided the same temperature, ramp speed, and shaking conditions are used. Library size distribution can also be assessed with a TapeStation (Agilent). DNA can be purified using any commercially available column-based DNA purification kit. Illumina sequencing libraries may also be prepared using alternative kits provided the minimum input requirements are met. Library concentration estimations can be performed with library quantification kits from other vendors such as Kapa Biosystems. This protocol requires next-generation sequencing. Any short-read sequencing platform with corresponding kits can be used (such as Illumina or Ion Torrent) provided a total of 50 million paired-end reads per sample is achieved. Here we used the NextSeq 2000 platform with the 100-cycle kit.

Step-by-step method details

Nuclei isolation and MNase digestion

Timing: day 2, 1 h

This section describes the procedures for isolating intact nuclei from frozen liver tissue and treatment with MNase to digest chromatin to mononucleosomes for extraction. The use of frozen tissue helps to ensure this protocol can be applied to archived material such as biopsies. Correct freezing techniques also keep chromatin structure in its most native form. The isolation of nuclei from frozen tissue also helps to enrich for chromatin components.

• 1.
  Nuclei isolation.

  • a.
    Add 1 mL ice-cold TM2 buffer with 60 μL 10% NP-40 (final NP-40 concentration 0.56%) to ∼20 mg of frozen liver and transfer to ice-cold 5 mL dounce homogenizer.
  • b.
    Gently dounce (5 times with pestle A and 5 times with pestle B) to release nuclei.
  • c.
    Incubate the homogenizer on ice for 3 min with 5 s gentle vortexing every min.
  • d.
    Separate the nuclei from cellular debris by gentle centrifugation (100 g) for 10 min in a refrigerated microcentrifuge set to 4°C.

    Note: The nuclei pellet appears as a white, loose pellet that is easily disturbed.

  • e.
    Carefully remove the supernatant to prevent disrupting the nuclei pellet.
  • f.
    Wash the nuclei pellet once with 1 mL of ice-cold TM2 buffer without NP-40 by gently pipetting the buffer into the tube. The pipetting action easily disrupts most of the nuclei pellet and the rest can be fully resuspended by gentle flicking of the tube.
  • g.
    Pellet the nuclei for 10 min at 100 g at 4°C and remove the supernatant without disrupting the nuclei pellet.
• 2.
  MNase digestion.

  • a.
    Set thermomixer to 23°C.
  • b.
    Add 100 μL of room temperature TM2 and 0.5 μL of 0.2 M CaCl₂ (final CaCl₂ concentration 1 mM) to the nuclei pellet and resuspend gently.
  • c.
    Add 6 μL of MNase (NEB, 2000 U/μL) and incubate at 23°C for 15 min exactly (with shaking at 1250 rpm) on a thermomixer.
  • d.
    Add 0.4 μL of 0.5 mM EGTA (final EGTA concentration 2 mM) to stop the MNase digestion.
• 3.
  Saving total MNase digest (“0”) and supernatant (“supn”) fractions.

  • a.
    Remove 10 μL (10% of reaction) and label as “0” for total MNase-digested chromatin. This sample can be frozen in a −80°C freezer.
  • b.
    Remove another 10 μL for checking digestion profile on a Bioanalyzer to ensure most of the chromatin has been digested to mononucleosome length (∼150 bp).

    Note: The DNA can be purified using a standard DNA extraction kit following manufacturer’s instructions (example, Zymo Clean and Concentrator).

    • i.
      Elute the purified DNA in ∼20 μL of supplied elution buffer.
    • ii.
      Load ∼1 μL of undiluted, purified DNA on a high sensitivity (HS) chip for analysis on a Bioanalyzer (Figure 1A, young “0”).

      Note: See troubleshooting section (problem 1) if digestion profile is suboptimal.

  • c.
    An additional 10 μL can be removed for protein analysis by western blot. This aliquot can be frozen in a −80°C freezer for future protein extraction.
  • d.
    To remove the MNase, pellet the nuclei for 10 min at 100 g and carefully aspirate the supernatant. Re-pellet the supernatant to remove any traces of nuclei and save the supernatant as “supn” fraction (∼70 μL).
  • e.
    Wash the nuclei pellet carefully with 1 mL of ice-cold TM2 buffer without NP-40, centrifuge, and remove the supernatant as above. Proceed with the nuclei pellet to the salt fractionation step.

    CRITICAL: MNase amount and incubation times should be optimized. The optimal MNase will result in an ideal digestion profile comprising mostly of mononucleosomes. An example digestion profile/Bioanalyzer trace after DNA purification is shown in Figure 1A. An idea of fragment sizes produced by suboptimal digestion can be visualized in Figure 4E in Yang et al.¹ The nuclei pellet should be held at 4°C/on ice always (except where specified to use room temperature buffer during the MNase digestion step). Nuclei integrity is critical to this protocol.

Salt fractionation

Timing: day 2, 9 h

The following section outlines the steps to sequentially extract chromatin fractions from nuclei with progressively increasing salt concentrations. Salt fractionation is founded on the differential physical and electrostatic properties of the two chromatin compartments: euchromatin and heterochromatin. Euchromatin generally has lower charge-based protein-protein and protein-DNA interactions compared to heterochromatin and is thus easier to disrupt at low salt concentrations. Note that salt here refers to sodium salts. The fractionation is performed with physiological concentrations of magnesium salts.

• 4.
  Extracting and saving sequential salt fractions (“input”, “67.5”, “150”, “250” and “350” mM).

  • a.
    Resuspend the nuclei pellet from step 3e above in 70 μL of ice-cold 0 mM Triton buffer which has no salt. Take out 35 μL as “input”. This sample can be frozen in a −80°C freezer.
  • b.
    Add 1 μL 2.5 M NaCl to the remainder 35 μL to make it a final concentration of ∼67.5 mM and incubate on a tube rotator at 4°C for 2 h.
  • c.
    Pellet the nuclei at 100 g for 10 min at 4°C and save the supernatant labeled as “67.5 mM” fraction (∼ 35 μL). This sample can be frozen in a −80°C freezer.
  • a.
    Resuspend the nuclei pellet in 35 μL of 150 mM Triton buffer and incubate on a tube rotator at 4°C for 2 h.
  • b.
    Pellet the nuclei at 100 g for 10 min at 4°C and save the supernatant labeled as “150 mM” fraction (35 μL). This sample can be frozen in a −80°C freezer.
  • c.
    Resuspend the nuclei pellet in 35 μL of 250 mM Triton buffer and incubate on a tube rotator at 4°C for 2 h.
  • d.
    Pellet the nuclei at 100 g for 10 min at 4°C and save the supernatant labeled as “250 mM” fraction (35 μL). This sample can be frozen in a −80°C freezer.
  • e.
    Resuspend the nuclei pellet in 35 μL of 350 mM Triton buffer and incubate on a tube rotator at 4°C for 2 h.
  • f.
    Pellet the nuclei at 100 g for 10 min at 4°C and save the supernatant labeled as “350 mM” fraction (35 μL). This sample can be frozen in a −80°C freezer.
  • g.
    The remaining pellet fraction corresponds to ∼5–10% of chromatin. Resuspend the pellet in 35 μL of TNE buffer and label as “pel” fraction (35 μL). This sample can be frozen in a −80°C freezer. See troubleshooting section (problem 2) if nuclei pellet is not visible in any of the salt extraction steps.

Pause point: Samples may be kept frozen at −80°C freezer for a week.

DNA purification and library construction

Timing: day 3, 10 h

This section includes two steps: DNA purification from each fraction and sequencing library construction for genome profiling. This step ensures that DNA fragments devoid of protein and RNA contaminants are carried forth for sequencing library construction. It also validates that the MNase digestion was optimal for all samples.

• 5.
  DNA purification.

  • a.
    Take out the frozen fractions labeled “input”, “supn”, “0”, “67.5”, “150”, “250”, “350” and “pel” from the −80°C freezer and thaw on ice.
  • b.
    Take 5 μL aliquot of each of “input”, “supn”, “0”, “67.5”, “150”, “250”, “350” and “pel” and put in a new tube.
  • c.
    Set thermomixer to 37°C.
  • d.
    Add 45 μL ice-cold TNE buffer (with freshly added protease and phosphatase inhibitors) to each tube to make up the volume to 50 μL.
  • e.
    Add 2.5 μL 10% SDS to lyse nuclei (final SDS concentration 0.5%).
  • f.
    Add 12 μL RNase A (10 mg/mL) and incubate at 37°C for 30 min on a thermomixer without shaking to remove RNA.
  • g.
    Add 5 μL Proteinase K (20 mg/mL) and incubate at 37°C for 30 min on a thermomixer without shaking to remove protein.
  • h.
    Purify DNA using the Zymo Clean and Concentrator kit following manufacturer’s instructions (Zymo Clean and Concentrator) Elute DNA in 10 μL of supplied DNA elution buffer.
  • i.
    Load 1 μL on a DNA HS chip and run on a Bioanalyzer. Figure 1B shows representative traces of all salt fractions from a young sample.
  • j.
    Mix 1 μL of purified DNA and mix with 199 μL of Qubit 1× dsDNA High Sensitivity assay buffer, incubate for 2 min and measure the DNA concentration on a Qubit fluorometer. See troubleshooting section (problem 3) if DNA recovery is low.

Pause point: Purified DNA may be kept frozen at −80°C freezer for weeks to months.

CRITICAL: We do not recommend phenol-chloroform extractions for DNA purification from salt extracted fractions due to the possibility of cross-sample variability in DNA recovery and impurity of the extracted DNA. Additionally, carryover phenol can interfere with downstream enzymatic reactions. We also suggest that all DNA estimations be made using a dye-based measurement such as Qubit fluorometers or qPCR for accuracy.

• 6.
  Library construction.

  • a.
    Use ∼10 ng purified DNA to make libraries using the NEBNext Ultra II kit and dual indices.
  • b.
    For every sample, make up the volume of the DNA to 50 μL with DNA elution buffer in sterile nuclease-free 200 μL PCR tubes. Add 3 μL NEBNext Ultra II End Prep Enzyme Mix and 7 μL NEBNext Ultra II End Prep Reaction Buffer for a total volume of 60 μL. Mix thoroughly using a pipette and spin quickly to collect the liquid at the bottom of the tubes.
  • c.
    Place in a thermocycler, with the heated lid set to ≥ 75°C, and run the following program: 30 min @ 20°C; 30 min @ 65°C; hold at 4°C.
  • d.
    Dilute the NEBNext adaptor for Illumina in 10 mM Tris-HCl, pH 7.5 with 10 mM NaCl. Note that this dilution is required when using 5–100 ng input DNA.
  • e.
    Following the end prep step, add in order to each tube, 2.5 μL of diluted adaptor, 30 μL NEBNext Ultra II Ligation Master Mix and 1 μL NEBNext Ligation Enhancer for a total volume of 93.5 μL. Mix thoroughly using a pipette and spin quickly to collect the liquid at the bottom of the tubes.
  • f.
    Incubate at 20°C for 15 min in a thermocycler with the heated lid off.
  • g.
    Add to each tube, 3 μL of USER enzyme.
  • h.
    Mix well and continue to incubate at 37°C for 15 min in the thermocycler with the heated lid set to ≥ 47°C.
  • i.
    Add 87 μL (0.9×) of resuspended SPRI beads to the Adaptor Ligation reaction. Mix well by pipetting up and down at least 10 times. Be careful to expel all of the liquid out of the tip during the last mix.
  • j.
    Incubate samples for 5 min at 24°C.
  • k.
    Place the tubes on an appropriate magnetic stand. Wait for 5 min to separate the beads from the supernatant and then carefully remove and discard the supernatant without removing any beads.
  • l.
    Wash the beads with 200 μL of 80% freshly prepared ethanol while on the magnetic stand. Incubate at 24°C for 30 s, and then carefully remove and discard the supernatant. Repeat the wash for a total of 2 washes.
  • m.
    Air dry the beads for 3 min. Do not overdry the beads as this may lead to poor DNA yields.
  • n.
    Remove the tube from the magnetic stand and elute the DNA target from the beads by adding 17 μL of 10 mM Tris-HCl and mixing with a pipette 10 times.
  • o.
    Incubate at 24°C for 2 min and return to the magnetic stand.
  • p.
    After 5 min when the solution is clear, transfer 15 μL to a new PCR tube.
  • q.
    Add to the tube containing eluted DNA, 10 μL Unique Dual Index Primer Mix and 25 μL NEBNext Ultra II Q5 Master Mix for a total volume of 50 μL. Mix thoroughly using a pipette and spin quickly to collect the liquid at the bottom of the tubes.
  • r.
    Place the tubes in a thermocycler, with the heated lid set to ≥ 75°C, and run the following program: 30 s @ 98°C for initial denaturation; 11 cycles of the following 2 steps: 10 s @ 98°C and 75 s @ 65°C for denaturation/annealing/extension; 5 min @ 65°C for final extension; hold at 4°C.
  • s.
    Perform cleanup of amplified library as in steps 6i-n except with 45 μL of resuspended SPRI beads and elute with 33 μL 10 mM Tris-HCl, transferring 30 μL to a new tube.
  • t.
    Load 1 μL of purified library on a DNA 1000 chip and run on a Bioanalyzer to check library size and concentration. Example libraries from the 250 mM salt fractions from young and old samples are shown in Figure 1C. See troubleshooting section (problem 4) if library yield is low.
  • u.
    Pool libraries from multiple samples in equimolar amounts (20 nM each based on Bioanalyzer concentration in the 100–1000 bp range). We recommend diluting libraries in 10 mM Tris-Cl, pH 8.5 which is compatible with downstream Illumina sequencing.
  • v.
    Measure the concentration of the pooled library with NEBNext Library Quant kit (NEB Library Quant Kit) and Qubit (optional; Qubit dsDNA HS assay) following manufacturer’s instructions. If doing both, the two estimations should closely match each other.

CRITICAL: While DNA quantities <10 ng can be used to prepare libraries using the NEBNext Ultra II kit, we recommend using at least 10 ng to achieve optimal library complexity and the advised number of PCR cycles to avoid excessive PCR duplicates. The pooling of multiple libraries in equimolar amounts also requires special attention to prevent uneven sequencing outputs and unnecessary normalization-induced differences. Ensure SPRI beads are completely resuspended before use as otherwise, DNA recovery and size distribution may be variable across samples.

Sequencing and computational analysis

Timing: day 4–10

Sequence the pooled library on an Illumina NextSeq 2000 platform using a 100-cycle kit and a 50 paired-end format. Follow Illumina instructions to load the library on the instrument. A minimum of 50 million paired-end reads per sample is optimal for downstream analysis.

The following section outlines a pipeline for computational analysis and visualization of the salt fraction libraries. It is recommended that the processes be run on a high-performance cluster or cloud computing service. The output BAM and bigWig files can be used for further downstream analysis such as peak calling, nucleosome occupancy etc.

• 7.
  Load the packages, assign samples, and set working directory.

  • a.
    Load required packages in bash.

    module load cutadapt/3.0

    module load fastqc/0.11.9

    module load bowtie/2

    module load samtools/1.13

    module load picard/2.23.7

    module load sambamba/0.7.1

    module load gcc/7.4.0

    module load bedtools/2.30.0

    module load deeptools/3.5.1
  • b.
    Set sample names. The samples referred to here are 2 biological replicates of young (Y1, Y2) and old (O1, O2) livers and their corresponding salt fractions.

    sample='Y1_Input Y1_0 Y1_67 Y1_150 Y1_250 Y1_350 Y1_Pel Y2_Input Y2_0 Y2_67 Y2_150 Y2_250 Y2_350 Y2_Pel O1_Input O1_0 O1_67 O1_150 O1_250 O1_350 O1_Pel O2_Input O2_0 O2_67 O2_150 O2_250 O2_350 O2_Pel'
  • c.
    Set project working directory.

    path="/path to your working directory/"
• 8.
  Pre-process data and perform quality control.

  • a.
    Trim Illumina adapter reads with Cutadapt.

    #Adapter trimming (NEBNext adapter sequences)

    for i in $sample

    do

    cutadapt -a AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC \

    -A AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT \

    -o ${path}/${i}_R1_val_1.fq -p ${path}/${i}_R2_val_2.fq \

    ${path}/${i}_R1.fastq ${path}/${i}_R2.fastq

    done
  • b.
    Perform FastQC to check the sequence quality before processing the data.

    for i in $sample

    do

    fastqc -f fastq -o ${path}/ ${path}/${i}_R?_val_?.fq

    done
• 9.
  Align fragments to genome.

  • a.
    Align the fragments to mouse genome (mm10) with bowtie2 to generate SAM files.
  • b.
    Generate filtered BAM files from SAM files to only keep mapped reads with “-F” setting to 4.

#Directing Bowtie2 to the genome build (mm10). Bowtie2 indices are available as part of the igenomes package.

export BOWTIE2_INDEXES=/path to Bowtie2Index/

for i in $sample

do

#Performing alignment

bowtie2 -p 24 --end-to-end --very-sensitive --no-mixed --no-discordant \

--phred33 \

-I 10 -X 700 -x genome -1 ${path}/${i}_R1_val_1.fq -2 \ ${path}/${i}_R2_val_2.fq \

-S ${path}/${i}.sam &> ${path}/${i}_bowtie2.txt

#Generating filtered BAM files

samtools view -@ 24 -h -F 4 -q 10 -bS ${path}/${i}.sam > \ ${path}/${i}_filtered.bam

done

• 10.
  Filter for uniquely mapped reads.

  • a.
    Use picard to sort the BAM files before filtering.
  • b.
    Use sambamba to only keep the uniquely mapped reads on the sorted BAM files.
  • c.
    Use picard to additionally remove the duplicate reads.

for i in $sample

do

#Sorting

java -jar $PICARDJARPATH/picard.jar SortSam -I $i -O ${path}/${i}_\

sorted.bam -SORT_ORDER coordinate

#Retaining uniquely mapped and not duplicate reads

sambamba view -h -t 2 -f bam -F "[XS] == null and not unmapped and not \ duplicate” \

${path}/${i}_sorted.bam > ${path}/${i}_sorted_unique.bam

#Removing additional duplicates

java -jar $PICARDJARPATH/picard.jar MarkDuplicates \ I=${path}/${i}_sorted_unique.bam \

O=${path}/${i}_noDUP_unique.bam REMOVE_DUPLICATES=true \ METRICS_FILE=${path}/${i}_noDUP.txt

done

• 11.
  Remove ENCODE Blacklisted regions.

  • a.
    Download the mm10 blacklist bed file from ENCODE website.
  • b.
    Remove the blacklist regions from each sample bed file with the “intersect” function in bedtools with the blacklist bed file from last step by setting the “-v” parameter to only keep the nonoverlapped regions in sample bed files.

#Downloading ENCODE mm10 blacklist bed file

wget -O ${path}/ENCFF547MET.bed.gz \

https://www.encodeproject.org/files/ENCFF547MET/@@download\

/ENCFF547MET.bed.gz \

gunzip ${path}/ENCFF547MET.bed.gz

#Removing regions that intersect with ENCODE blacklisted regions

for i in $sample

do

bedtools intersect -a ${path}/${i}_noDUP_unique.bam \

-b ${path}/ENCFF547MET.bed -v > \

${path}/${i}_noB_unique.bam

done

• 12.
  Generate bigWig files for browser tracks.

  • a.
    Index the BAM files before bigWig file generation with the “index” function in samtools.
  • b.
    Use the “bamCoverage” function in deeptools to generate bigWig files from BAM files with “RPKM” as normalization method.

    #Indexing BAM files

    for i in $sample

    do

    samtools index ${path}/${i}_noB_unique.bam

    done

    #Generating bigWig files

    for i in $sample

    do

    bamCoverage -b ${path}/${i}.bam \

    -o ${path}/${i}.bw -of bigwig --normalizeUsing RPKM

    done
  • c.
    Subtract “Input” bigWig from “Sample” bigWig to generate input subtracted browser tracks with the “bigwigCompare” function in deeptools.

    #subtracting Input from all other Y1 samples for Y1

    Sample_Y1='Y1_0 Y1_67 Y1_150 Y1_250 Y1_350 Y1_Pel’

    for i in $Sample_Y1

    do

    bigwigCompare -b1 ${path}/${i}.bw -b2 ${path}/Y1_Input.bw \

    --operation subtract \

    -o ${path}/${i}-Input.bw -of bigwig

    done

    #subtracting Input from all other Y2 samples for Y2

    Sample_ Y2='Y2_0 Y2_67 Y2_150 Y2_250 Y2_350 Y2_Pel’

    for i in $Sample_Y2

    do

    bigwigCompare -b1 ${path}/${i}.bw -b2 ${path}/Y2_Input.bw \

    --operation subtract \

    -o ${path}/${i}-Input.bw -of bigwig

    done

    #subtracting Input from all other O1 samples for O1

    Sample_ O1='O1_0 O1_67 O1_150 O1_250 O1_350 O1_Pel’

    for i in $Sample_O1

    do

    bigwigCompare -b1 ${path}/${i}.bw -b2 ${path}/O1_Input.bw \

    --operation subtract \

    -o ${path}/${i}-Input.bw -of bigwig

    Done

    #subtracting Input from all other O2 samples for O2

    Sample_ O2='O2_0 O2_67 O2_150 O2_250 O2_350 O2_Pel’

    for i in $Sample_O2

    do

    bigwigCompare -b1 ${path}/${i}.bw -b2 ${path}/O2_Input.bw \

    --operation subtract \

    -o ${path}/${i}-Input.bw -of bigwig

    done

Expected outcomes

A successful execution of the protocol will produce measurable amounts of DNA (in ng/μL) in the Qubit fluorometer with the HS assay kit. The highest amount of DNA is recovered in the 250 and 350 mM fractions while the “supn” and euchromatic (“67.5” and “150”) fractions yield relatively lower amounts of DNA compared to the other fractions. The expected average size of the DNA libraries is ∼250–270 bp, being the combined size of a mononucleosome (∼150 bp) and ligated adapter (∼126 bp). DNA and library yields may vary based on the tissue/animal used.

After sequencing and mapping, the mapped fragments are expected to cover both euchromatin and facultative heterochromatin regions of the genome. From the genome browser view in Figure 1C, inset 1 is enriched in the low salt fraction, which is open euchromatin, while the region in inset 2 is enriched in high salt and pellet fraction which is closed facultative chromatin. Interestingly, this separation is better in old samples and the region enriched in old high salt fraction also shows high H3K27me3 signal. In aged livers, due to formation of new facultative heterochromatin (marked by H3K27me3) at lamina-associated domains (LADs), a sizeable portion of the mapped fragments aligns to these regions. In contrast, in young samples LAD-bound chromatin is organized as constitutive heterochromatin and this salt fractionation protocol cannot extract this fraction (see Limitations section below). Consequently, young samples do not have any signal at LADs.

Quantification and statistical analysis

Illumina sequencing reads are first de-multiplexed generating compressed FASTQ files by typical demultiplexing software such as bcl2fastq2⁸ or BCL convert.⁹ The FASTQ files are then trimmed to remove adapter sequences with cutadapt¹⁰ and the qualities of the FASTQs assessed using FastQC.¹¹ The reads are aligned to the appropriate reference genome (GRCm38/mm10 genome in our case) using bowtie³ and the end-to-end parameter. SAM output files are filtered to keep alignments with a minimum mapping quality of 10 using samtools⁴ and PCR duplicates removed with picard.¹² Sambamba⁵ is used to retain only uniquely aligned reads. Reads mapping to the Encyclopedia of DNA Elements (ENCODE) blacklisted regions¹³ are also removed from the analysis. Optimally, there should be no statistically significant differences in sequencing depth, alignment rate, or alignable fragments per million across different groups. The bamCoverage function in deepTools¹⁴ is used to generate RPKM (reads per kilobase per million mapped reads) normalized bigWig files. Input reads are subtracted from the different salt fractions (“0”, “67.5”, “150”, “250” and “350” mM) and pellet with bigwigCompare. These input-subtracted bigWig files are then viewed in a genome browser such as UCSC genome browser or Integrative Genomics Viewer.

The input samples can be further used to infer nucleosome positions using DANPOS.⁷ The samples are read-normalized using the fold change normalization method and parameters such as occupancy, signal periodicity, fuzziness profiles are plotted using the profile or stat function.

Limitations

This protocol cannot extract constitutive heterochromatin which is typically tightly attached to the nuclear lamina. In this protocol, constitutive heterochromatin is present in the final “pel” fraction but is not solubilized in 0.5% SDS or even after RNase A and proteinase K treatment. The resultant insoluble material is therefore a protein-DNA complex which typically has different properties than free DNA molecules. They do not interact with the silica membrane in DNA binding columns present in DNA purification kits such as Zymo Clean and Concentrator, QIAGEN PCR purification or gel extraction. Consequently, the constitutive heterochromatin fraction is lost.

Troubleshooting

Problem 1

MNase digestion profile does not show mononucleosome enrichment (i.e., peak at around 150 bp) and is predominated by fragment sizes >1 kb or < 100 bp (related to step 3b).

Potential solution

• •This could indicate suboptimal MNase digestion. We recommend performing a pilot experiment with the tissue of choice, a titration with different amounts of MNase and a fixed amount (say 20 mg) of tissue (see Figure 4E in Yang et al.¹ for an example titration). The digested DNA should be purified as outlined in step 3b and 1 μL run on a Bioanalyzer HS chip. Alternatively, the digestion can also be verified on an agarose gel. Optimal MNase may depend on tissue type and amount.

Problem 2

Nuclei pellet not visible (related to step 4).

Potential solution

• •After the first few rounds of sequential salt extraction, we have noted that sometimes the pellet become invisible. This may depend on the tissue type, fat content, whether young or old, as well as the amount. While it is important to be careful during pipetting the supernatant, an invisible pellet does not necessarily imply nuclei loss. We recommend continuing to the subsequent steps. However, it is good practice to orient the microcentrifuge tubes in the rotor the same way during the centrifugation steps to pellet the nuclei consistently on the same side of the tubes.

Problem 3

Poor DNA recovery after salt fractionation (related to step 5j).

Potential solution

• •MNase digestion may not be optimal. Ensure optimal digestion profiles as shown in Figure 1A. Refer to the solution presented for Problem 1 for troubleshooting.
• •It is possible that the starting amount of tissue is not enough. Start with larger amounts of tissue. The liver typically yields high amounts of material which may not be true for other organs. If using larger amounts of tissue, ensure MNase treatment is also optimal. Additionally, it is recommended to scale up extraction buffers while keeping the elution volume the same in steps 4-5.
• •It is also possible that the nuclei recovery is poor, yielding low DNA. Optimize nuclei isolation in step 1 including the number of dounce strokes.

Problem 4

Low library yield despite optimal DNA recovery after salt fractionation (related to step 6).

Potential solution

• •Several steps in the library construction process may have gone wrong. Refer to the manufacturer’s manual (NEBNext Ultra II) and carefully assess if the proper time, temperature, and enzyme amounts have been applied each step. When dealing with a large number of samples, it is possible to miss adding an important component to one tube.
• •Check if the optimal number of PCR cycles has been used.
• •During the size selection and DNA purification steps, it is important to vortex the beads thoroughly before dispensing to ensure proper ratio of beads to sample is used and to not over dry the beads after ethanol washes. Both steps can severely affect DNA recovery.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Payel Sen (payel.sen@nih.gov).

Technical contact

Technical questions about the execution of the protocol should be directed to and will be fulfilled by the technical contact, Na Yang (na.yang2@nih.gov).

Materials availability

This study did not generate new unique reagents.

Data and code availability

All genome-wide datasets are publicly available at the Gene Expression Omnibus portal (GEO superseries: GSE185708, subseries GSE185707). All code used to analyze the salt fractionation data are available at Zenodo (https://doi.org/10.5281/zenodo.7789081) under Salt Fractionation and DANPOS.