In Situ Nucleosome Assembly for Single-Molecule Correlative Force and Fluorescence Microscopy

Summary

This article presents a detailed experimental procedure for reconstituting nucleosome-containing DNA tethers for single-molecule correlative force and fluorescence microscopy. It further describes several downstream experiments that can be conducted to visualize the binding behavior of chromatin-interacting proteins and analyze changes in the physical properties of nucleosomes.

Abstract

Nucleosomes constitute the primary unit of eukaryotic chromatin and have been the focus of numerous informative single-molecule investigations regarding their biophysical properties and interactions with chromatin-binding proteins. Nucleosome reconstitution on DNA for these studies typically involves a salt dialysis procedure that provides precise control over the placement and number of nucleosomes formed along a DNA tether. However, this protocol is time-consuming and requires a substantial amount of DNA and histone octamers as inputs. To offer an alternative strategy, an in situ nucleosome reconstitution method for single-molecule force and fluorescence microscopy that utilizes the histone chaperone Nap1 is described. This method enables users to assemble nucleosomes on any DNA template without the need for strong nucleosome positioning sequences, adjust nucleosome density on demand, and use fewer reagents. In situ nucleosome formation occurs within seconds, offering a simpler experimental workflow and a convenient transition into single-molecule measurements. Examples of two downstream assays for probing nucleosome mechanics and visualizing the behavior of individual proteins on chromatin are further described.

Introduction

The primary packaging unit of eukaryotic chromatin is the nucleosome, in which ~147 base pairs (bps) of DNA are wrapped around an octamer of core histones proteins^1,2. In addition to genome packaging, the nucleosome architecture serves as another rich layer of biophysical regulation that can be harnessed by chromatin-binding proteins when performing their various functions^3,4. Experimentally accessing and measuring the physical characteristics of nucleosomes has been technically challenging since these units perform at minuscule scales (e.g., nanometer lengths, piconewton forces), and thus, their probing requires sufficient sensitivity and precision to meaningfully inform function. Moreover, chromatin-binding proteins often engage with their substrates transiently, and most ensemble approaches lack appropriate temporal resolution to inform the kinetics of these interactions⁵. Fortunately, the advent of single-molecule techniques has made it possible to visualize and manipulate individual proteins and their interactions in real-time, revealing mechanistic information about these molecular events occurring at the nanoscale⁶. In particular, single-molecule correlative force and fluorescence microscopy (smCFFM) harnesses both high-resolution fluorescence detection and force manipulation tools to simultaneously resolve the mechanics, composition, and coordination of chromatin and chromatin-protein complexes^7,8.

In smCFFM that specifically employs "optical tweezers" as the force manipulation method, a tightly focused laser is used to trap a dielectric object, typically a micron-sized plastic bead, in three dimensions⁹. A biopolymer such as a piece of DNA can be conjugated to the bead (via, e.g., streptavidin-biotin, digoxigenin-anti-digoxigenin antibody linkage), and the user can both apply a calibrated force and measure the force/displacement generated by the system¹⁰. The added fluorescence module enables simultaneous multicolor fluorescence imaging to track protein composition and behavior.

The mainstream method for reconstituting nucleosome templates for smCFFM involves assembling nucleosomes on custom DNA templates by salt dialysis^11,12,13. In this procedure, purified histone octamers are incubated with DNA templates in a high-salt buffer (~1 M sodium chloride), and as the salt is slowly dialyzed away, nucleosomes spontaneously assemble on the DNA in a sequence-dependent positional manner^14,15. As such, it is customary that strong nucleosome positioning sequences (e.g., Widom 601¹⁶, X. laevis 5S rDNA¹⁷) are incorporated within the DNA templates to generate custom nucleosome arrays. Although this well-established method offers precise control over the placement and number of nucleosomes across DNA, it suffers from several disadvantages: (1) incorporation of strong nucleosome positioning DNA sequences may generate artificially stable nucleosomes that bias the activity of chromatin-binding proteins, (2) the process of salt dialysis for nucleosome formation requires high amounts of DNA and octamers, and (3) the procedure takes one to several days of work as well as a considerable effort to titrate the correct DNA to octamer ratio for optimal nucleosome array density.

In this manuscript, we describe an alternative method to form nucleosomes along DNA in situ within a single-molecule correlative force and fluorescence microscope using recombinant histone chaperone Nap1, which resembles the physiological pathway of nucleosome formation in vivo^{18,19,20,21,22}. This method allows users to assemble nucleosomes on non-specific DNA sequences, easily adjust nucleosome density, and utilize significantly fewer amounts of reagents, all within minutes. Additionally, the formation of nucleosomal DNA tethers in situ enables simpler experimental workflow and the convenience of built-in single-molecule visualization and manipulation. Thus, when uniform and specific nucleosome positioning is not a necessity, this protocol offers a useful option for the investigation of nucleosome mechanics or protein behavior on chromatin, for which we also describe example assays.

Protocol

The details of the reagents and the equipment used in the study are listed in the Table of Materials.

1. Preparation of biotinylated DNA

1. Prepare a 120 µL volume reaction containing 20 µg of λ DNA, 33 µM of biotin-dCTP, 33 µM of biotin-dUTP, 33 µM of biotin-dATP, 33 µM of dGTP, 10 units of Klenow enzyme, and 1x NEB Buffer 2.
   NOTE: This procedure specifically utilizes linear double-stranded (ds) methylation-free bacteriophage λ genomic DNA (48,502 bps), which contains 12-nucleotide (nt) 5' single-stranded (ss) DNA overhang²³. However, the protocol can be adapted to make use of any length or sequence of linear dsDNA provided it contains at least several nts of ss 5' overhangs, the length and nt sequence dictating the number of biotinylated nts installed on each end for downstream tethering between optical traps. Additionally, the reaction can be scaled down. For example, 5 µg of DNA can be used in a 30 µL volume reaction. Additionally, the DNA substrates generated in this way are not torsionally constrained. For protocols to create torsionally constrained DNA, please refer to the previously published reports^24,25.
2. Incubate the reaction at room temperature for 15 min.
3. Add 10 mM of EDTA and incubate at 75 °C for 20 min.
4. Perform ethanol precipitation to obtain the biotinylated DNA product.
   1. Add 0.1x volume of 3 M sodium acetate and 3x volume of ice cold 100% ethanol. Keep at -20 °C for at least 1 h up to overnight. Centrifuge at 13,500 x g for 30 min at 4 °C.
   2. Carefully remove the supernatant without disturbing the pellet and add 1 mL of 75% ethanol. Centrifuge at 13,500 x g for 1 min at 4 °C.
   3. Repeat the supernatant removal, ethanol addition, and centrifugation steps.
   4. Carefully remove all the supernatant and let the pellet air dry for 10 min. Dissolve the pellet in 100-200 µL of TE buffer or ddH₂O (can incubate at 37 °C to facilitate dissolving). Measure the concentration using a Nanodrop spectrophotometer.

2. Preparation of channels in a flow cell

NOTE: This procedure is based on a commercially available microfluidic chip (see Table of Materials) but can be adapted to any smCFFM instrument with a multi-channel flow cell.

1. Passivate channels (ch.) 4 and 5 (or any ch. that will contain proteins) (see schematic of channels in Figure 1A). Add 500 µL of 0.1% BSA to each channel and flow at 0.8 bar (~0.45 µL/s for tubing with a 0.010 inch inner diameter) for 8 min. Remove solution.
2. Add 500 µL of 0.05% Pluronic-F127 to each channel and flow at 0.8 bar for 8 min. Remove the solution and rinse out the syringes with 1x PBS. Add 2 mL of 1x PBS to each channel and flow at 0.8 barfor 27 min.
3. Vortex the stock solution of streptavidin-coated beads (this procedure uses 3.13 µm size beads) and add 2.15 µL of the bead solution to 1 mL of 1x PBS. Add to ch. 1.
   NOTE: The size and concentration of the streptavidin-coated beads can be adjusted according to user preference.
4. Add 30-80 ng of biotinylated λ DNA to 1 mL of 1x PBS. Add to ch. 2.
   NOTE: The concentration of biotinylated DNA can be adjusted to user preference. Greater amounts of DNA will increase the frequency of forming DNA tethers but also raise the frequency of tethering multiple molecules of DNA at once, which can add to the difficulty of the experiment.

3. Nap1-mediated in situ nucleosome formation

1. Add 500 µL of protein-free image buffer to ch. 3.
   NOTE: The buffer composition in ch. 3 is flexible and should be appropriate for both nucleosomes and the specific protein(s) of interest. Optionally, users may choose to add oxygen scavenging systems for reduced photobleaching of fluorophores and/or competitor DNA to remove histone octamers that are not properly wrapped into nucleosomes.
2. Add 2 nM S. cerevisiae Nap1 and 2-5 nM LD655-labeled H4 histone octamer (HO) to 500 µL of 1x HR buffer¹⁸. Add to ch. 4.
   NOTE: Histone octamers and Nap1 can be commercially purchased or prepared in-house (see Table of Materials). The number of nucleosomes to be formed along each tethered DNA depends on the octamer concentration as well as the time spent in this channel. The user's desired and approximate nucleosome density along each DNA molecule can be modulated by adjusting either or both of these parameters. Additionally, the user may choose to label histone octamers with their preferred fluorophore.
3. Optional: Add chromatin-binding protein or other protein of interest to ch. 5 in an appropriate buffer.
4. Flow all channels at 1.0 bar (~0.55 µL/s in the current experimental setup) for 30 s to flush the flow cell channels with samples.
5. Set the flow to 0.2 bar (~0.16 µL/s in the current experimental setup). Move the optical traps (OTs) to ch. 1 and catch a suitable pair of streptavidin-coated beads.
6. Move the OTs to ch. 3, conduct a force calibration for the bead pair, turn on the red confocal laser (or another laser line of choice), and calibrate the imaging area.
7. Set the flow to 0.2 bar and move OTs to ch. 2. Catch biotinylated DNA.
8. Move the OTs to ch. 3 and stop the flow. Increase the distance between the OTs to stretch the DNA and monitor the associated force-distance (FD) curve to confirm the presence of a single DNA tether (as opposed to multiple tethers). The curve should follow the worm-like-chain model for dsDNA at the given contour length²⁶.
9. Adjust the distance between the OTs so that the DNA tension reads approximately 1 pN.
   NOTE: 1 pN of tension is low enough to allow nucleosomes to remain wrapped by DNA²⁷ but is high enough to place the entire DNA molecule within the line scanning imaging axis.
10. Turn on line scanning to start collecting a kymograph with the red laser on.
    NOTE: Kymographs or 2D scans of tethered DNA can be obtained, as per user preference.
11. Move the OTs to ch. 4 with the flow off. Facilitated by Nap1, HOs are loaded onto DNA and wrapped to form nucleosomes in real-time. Nucleosome loading is visualized as fluorescent trajectories appearing on the DNA within kymographs. Concurrently, nucleosome formation on the DNA can be monitored by the force increase as more HOs are loaded when the positions of OTs are held constant.
    NOTE: Fluorescence excitation laser can be turned off during the nucleosome formation steps to avoid photobleaching of fluorophores. Additionally, non-specific binding of histone octamers to DNA wherein the octamers are not wrapped into nucleosomes could occur using this protocol. These binding events will not generate a "rip" when the DNA is stretched, as described in the Unwrapping Nucleosomes (step 4) protocol below, and can sometimes be washed off by a gentle buffer flow, particularly if the buffer contains free competitor DNA.
12. When the approximate desired number of nucleosomes is formed, start collecting data.

4. Unwrapping nucleosomes

1. Complete step 2 and step 3.
2. After forming a DNA tether containing nucleosomes, move the OTs to ch. 3.
3. Prepare for nucleosome unwrapping experiments by reducing the distance between OTs until the force is approximately zero. Zero the force.
4. Begin moving the OT 1 away from the OT 2 at a constant speed of 0.1 µm/s. On a kymograph, the bead in OT 1 will visually move farther from the bead in OT 2. On the FD curve, the force will rise as the distance increases. As nucleosomes unwrap, there will appear "rips" or transitions in the force (~8-37 pN) that signify the unraveling of the inner wrap of individual nucleosomes^27,28,29.
   NOTE: The force at which unwrapping occurs depends on the pulling rate, which is expected for non-equilibrium processes. Sometimes, multiple nucleosomes unwrap at once, producing a larger distance change in the transition on the FD curve. Additionally, the unraveling of the outer wrap of the nucleosome is not readily visible in this experimental setting (see the Discussion section).

5. Visualizing protein binding to chromatin

1. Complete step 2 and step 3.
2. Add 10 nM of Cy3-labeled linker histone H1.4 to 500 µL of image buffer in ch. 5.
   NOTE: Any protein of interest can be added to ch. 5 at a concentration of choice (typically below 100 nM for single-molecule visualization).
3. After forming a DNA tether containing nucleosomes, turn on the green laser in addition to the red laser and move OTs to ch. 5 for real-time imaging of H1 binding to chromatin.
4. Continue collecting kymographs or scans until sufficient statistics have been achieved (at least ~10-20 kymographs depending on the frequency of events and experimental difficulty).

Results

Using the setup described in step 3 (Figure 1A), nucleosome formation along the DNA tether was visualized as the appearance of red fluorescent foci on a 2D scan (Figures 1B, left) or trajectories over time on a kymograph (Figure 1B, right). Properly wrapped nucleosomes yielded fluorescence trajectories that were stationary over time within the diffraction limit of confocal detection (~300 nm). Notably, multiple nucleosomes formed ne...

Discussion

The described protocol provides several advantages for nucleosome reconstitution including minimizing reagents and time as well as enabling chaperone-dependent nucleosome formation along any (potentially native) DNA sequence. Moreover, the in situ method allows simpler experimental workflow and a convenient transition into single-molecule assays of nucleosome mechanics and protein-chromatin interaction. On the other hand, limitations of this approach include the inability to direct nucleosome positioning along t...

Materials

Name	Company	Catalog Number	Comments
1x HR buffer	N/A	N/A	30 mM tris acetate pH 7.5, 20 mM magnesium acetate, 50 mM potassium chloride, 0.1 mg/mL BSA
1x PBS (Phosphate-buffered saline)	N/A	N/A	137 mM sodium chloride, 2.7 mM potassium chloride, 10 mM sodium phosphate dibasic, 1.8 mM potassium phophate monobasic
Acetic acid, glacial	Millipore Sigma	AX0074-6	Use to make tris acetate
Biotin-11-dUTP	Jena Bioscience	NU-803-BIOX-S
Biotin-14-dATP	Jena Bioscience	NU-835-BIO14-S
Biotin-14-dCTP	Jena Bioscience	NU-956-BIO14-S
Bovine Serum Albumin	Millipore Sigma	A9418-50G	Dissolve in H2O and run through 0.22 um filter
Cy3 Maleimide Mono-Reactive Dye	Cytiva	PA23031	Maleimide functionalized Cy3 fluorophore
dGTP	New England Biolabs	N0442S
Eppendorf Centrifuge 5425 R	Fisher Scientific	05-414-051	Benchtop centrifuge with cooling
Ethyl alcohol, Pure	Millipore Sigma	459844
Ethylenediaminetetraacetic acid (EDTA)	Millipore Sigma	E9884	Dissolve in H2O to 0.5 M
Human histone octamer (H4, L50C; H2A, K119C)	N/A	N/A	Recombinant histone proteins and those harboring labeling mutations were purified in-house as described previously (see refs. 33, 34, 35, 36). Briefly, recombinant histones were expressed in BL21 (De3) pLySS cells (Promega). Inclusion bodies were isolated after sonication, and histones were extracted under denaturing conditions. Histones were dialyzed into buffer A (7 M urea, 10 mM tris hydrochloride pH 8.0, 100 mM sodium chloride, 1 mM EDTA, and 5 mM 2-mercaptoethanol), and the solution was added to a gravity column loaded with Q Sepharose Fast Flow (Cytiva). The flow through was then added to a gravity column loaded SP Sepharose Fast Flow (Cytiva), and the histones were eluted from the column by adding buffer A supplemented with 600 mM sodium chloride. Histones harboring labeling mutations (H4, L50C; H2A, K119C) were purified and then conjugated to the desired fluorophore via maleimide-functionalized dyes (Cytiva, Lumidyne) using a 20:1 dye-to-protein molar ratio (see refs. 33, 34). Histone octamers were assembled by adding an equal molar ratio of each wild-type or fluorophore-labeled histone under denaturing conditions, dialyzed into a high-salt buffer containing 2 M sodium chloride, and then purified by size exclusion chromatography as described previously (see refs. 36, 37). Alternatively, individual histone proteins and ready-made histone octamers can be purchased commercially (e.g., Epicypher).
Image buffer	N/A	N/A	20 mM tris hydrochloride pH 8.0, 100 mM sodium chloride
Klenow Fragment (3' to 5' exo-)	New England Biolabs	M0212S
Lambda DNA (dam-, dcm-)	Thermo Fisher Scientific	SD0021	Methylation-free λ DNA
LD655-MAL	Lumidyne Technologies	9	Maleimide functionalized LD655 fluorophore
Linker histone H1.4 (A4C)	N/A	N/A	Purified recombinant protein made in-house (see ref. 38)
LUMICKS C-Trap Dymo	LUMICKS	N/A	Dual-trap configuration; standard materials for instrument provided by manufacturer
Magnesium acetate solution	Millipore Sigma	63052-100ML	Use to make HR buffer
NEBuffer 2	New England Biolabs	B7002S	Included with Klenow Fragment kit
Pluronic F-127	Millipore Sigma	P2443-250G	Dissolve in H2O and run through 0.22 um filter
Potassium chloride	Millipore Sigma	P3911	Use to make PBS and HR buffer
Potassium phosphate monobasic	Millipore Sigma	P0662	Use to make PBS
S. cerevisiae Nap1	N/A	N/A	Nap1 was purified in-house as previously described (see refs. 33, 34). Alternatively, Nap1 can be purchased commercially (e.g., Active Motif).
Sodium acetate	Millipore Sigma	241245	Dissolve in H2O to 3 M
Sodium chloride	Millipore Sigma	S9888	Use to make PBS and image buffer
Sodium phosphate dibasic	Millipore Sigma	S9763	Use to make PBS
SPHERO Biotin Coated Particles (3.0-3.4 µm)	Spherotech	TP-30-5
Thermo Scientific NanoDrop 2000/2000c Spectrophotometer	Fisher Scientific	ND2000	NanoDrop Spectrophotometer
Tris Base	Fisher Scientific	BP152-500	Dissolve in H2O and adjust to appropriate pH; use to make image buffer and tris acetate