Article

ISSN 0006-2979, Biochemistry (Moscow), 2024, Vol. 89, No. 4, pp. 674-687 © Pleiades Publishing, Ltd., 2024.

674

REVIEW

Studying Structure and Functions of Nucleosomes

with Atomic Force Microscopy

Alexander A. Ukraintsev

, Mikhail M. Kutuzov

, and Olga I. Lavrik

1,2,a

Institute of Chemical Biology and Fundamental Medicine, Siberian Branch of the Russian Academy of Sciences,

630090 Novosibirsk, Russia

Novosibirsk State University, 630090 Novosibirsk, Russia

e-mail: lavrik@niboch.nsc.ru

Received October 14, 2023

Revised February 19, 2024

Accepted February 22, 2024

Abstract— Chromatin is an epigenetic platform for implementation of DNA-dependent processes. Nucleosome,

as a basic level of chromatin compaction, largely determines its properties and structure. In the study of nucleo-

somes structure and functions physicochemical tools are actively used, such as magnetic and optical “tweezers”,

“DNA curtains”, nuclear magnetic resonance, X-ray crystallography, and cryogenic electron microscopy, as well as

optical methods based on Förster resonance energy transfer. Despite the fact that these approaches make it pos-

sible to determine a wide range of structural and functional characteristics of chromatin and nucleosomes with

high spatial and time resolution, atomic force microscopy (AFM) complements the capabilities of these methods.

Theresults of structural studies of nucleosome focusing on the AFM method development are presented in this

review. The possibilities of AFM are considered in the context of application of other physicochemical approaches.

DOI: 10.1134/S0006297924040072

Keywords: nucleosome, AFM, chromatin, single-molecule methods for studying biomolecules

Abbreviations: AFM, atomic force microscopy; cryoEM,cryogenic electron microscopy; DREEM,dual resonance frequency

enhanced electrostatic microscopy; FRET,Förster resonance energy transfer; HS-AFM,high-speed atomic force microscopy;

PARP,poly (ADP-ribose) polymerase; ssNCP,nucleosome with a single-stranded DNA.

* To whom correspondence should be addressed.

# These authors contributed equally to this work.

INTRODUCTION

DNA molecules are used by cells for storage and

implementation genetic information. In eukaryotic

cells, DNA is predominantly present in a chromatin

composition, and both the level of chromatin compac-

tion and chromatin spatial organization are important

[1]. The basic level of DNA compaction is nucleosome,

a complex consisting of eight histones and 147-bp DNA

with its structure defining properties of chromatin to

a large extent. The first ideas about the nucleosome

structure were formulated in the middle of the 1970s

[2-4]. Schematic representation of the nucleosome struc-

ture is shown in Fig.1.

To date, main characteristics of the composition

and structure of nucleosome, as well as its molecular

dynamics, have been studied in sufficient detail. This

knowledge was obtained largely due to the use of mod-

ern physicochemical research methods, including sin-

gle-molecule techniques such as optical and magnetic

tweezers, DNA curtains, an optical approach based on

Förster resonance energy transfer (FRET), etc. [5-7].

These methods enable to measure a wide range of be-

havior parameters of biomolecules and their complex-

es, including chromatin and nucleosomes, with high

spatial and time resolution.

Along with the listed methods, the technique of

atomic force microscopy (AFM) is actively used. AFM

is based on monitoring of the probe interaction with

the surface. A probe in the form of a needle at the end

of the cantilever scans the sample. To detect change in

the vertical coordinate of the probe, a laser directed

STRUCTURE AND FUNCTIONS OF NUCLEOSOMES BY AFM 675

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Fig. 1. Nucleosome core particle structure (PDB ID: 1KX5). Spatial structure of nucleosome is shown as a ribbon diagram in two

projections.

at the cantilever and a photodetector recording the

reflected beam are used. The beam reflected from the

cantilever shifts according to the vertical displacement

of the probe. Force of the cantilever interaction with

the surface can be assessed based on the signal from

the photodetector. Then, depending on the scanning

mode, these data enable to obtain information on ver-

tical coordinate of the sample surface directly or in-

directly through the feedback mechanism. At present

there is a wide variety of scanning modes and variants

of scanning protocols developed.

CONTACT MODE AFM

The first nucleosome images were obtained using

contact mode AFM, during which the probe-surface

repulsive force regulated by the feedback loop [8, 9]

is recorded (Fig.2a). The authors of these works were

able to show polynucleosomal structure in the form of

“beads on a string” [8, 9]. In addition, main geometric

parameters of the nucleosome disk were determined:

its height and diameter, which were ~3 and ~32 nm, re-

spectively, not including radius of the probe curvature,

[9], as well as the number of DNA supercoil turns in

the nucleosome [9].

The contact mode is characterized by a relatively

large mechanical effect of the cantilever on the sur-

face, which leads to severe deformation and damage

to nucleosome samples during scanning. In addition,

this “hard” scanning mode requires the use of probes

with a curvature radius of at least 10nm, which, as a

whole, reduces sensitivity and accuracy of the mode.

As a result, the first nucleosome images obtained by

AFM contact mode are of a low quality, which largely

limited the use of this method in studying nucleosome

structure.

TAPPING MODE AFM

Introduction of the tapping mode AFM provided a

way to reduce significantly deformation of the sample

during scanning, which made it possible to use sharp-

er probes. In this mode, the cantilever oscillates at its

resonant frequency. The feedback loop records chang-

es in the cantilever oscillation parameters and corrects

the probe position relative to the surface, providing

an equilibrium position around which the cantilever

performs harmonic oscillations (Fig.2b). Such chang-

es dramatically increased resolution of AFM in exam-

ining biological molecules. In particular, Martin et al.

[10] demonstrated advantages of this scanning mode

over the contact mode in studying chromatin struc-

ture. In their work, the authors managed to determine

the size of nucleosome more accurately than in the

case of using contact mode by employing the tapping

mode. By using the tapping mode of AFM, height and

diameter of the nucleosome disk were determined

as 4.3 ± 1.2 nm and 23.2 ± 4.5 nm, respectively, which,

considering radius of the probe curvature, are close

to the values obtained by X-ray crystallography [11].

UKRAINTSEV et al.676

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Fig. 2. Schematic representation of AFM scanning modes exemplified with a device with position of a piezo scanner at the

bottom. a)Contact mode; b)tapping mode; c)non-contact mode; d)electrostatic AFM. Designations: FB,feedback; C,computer;

AC,alternating current; DC,direct current.

Moreover, use of this approach enabled visualization

of DNA in the linker region of chromatin (which was

impossible with the contact mode scanning) as well as

to confirm experimentally the nucleosome structure

calculated by means of mathematical modeling [12].

The use of tapping mode AFM by Qian et al. [13]

in their study of chromatin from the chicken red blood

cells made it possible to visualize different levels of

DNA compaction. Both individual nucleosome “beads”

and their clusters on DNA strands were demonstrat-

ed on the obtained images. Presence of the sites with

different DNA packing densities allowed the authors

to suggest the role of compaction in the regulation of

gene expression. In this work, geometric character-

istics of fibrils that have diameter of 15-60 nm were

determined, confirming the solenoid model of nucleo-

STRUCTURE AND FUNCTIONS OF NUCLEOSOMES BY AFM 677

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

some packaging. In addition, it has been shown for the

first time that the fibrils of lower compaction degree

(~30 nm) are further folded to form chromatin struc-

tures of higher order (~60 nm), which, in turn, fold

into even more highly organized chromatin structures

(~90-110nm or more).

Design of nucleosome-positioning DNA sequences

[14], characterized by the unambiguous location of the

nucleosome core on the DNA molecule and high stabil-

ity of the formed complexes, enabled to create mod-

el structures with predicted location of nucleosomes.

Use of such nucleosome-positioning DNA sequences

made it possible to show stochasticity of nucleosome

location in the natural chromatin using telomeric sites.

These results confirmed the corresponding theoretical

model of nucleosome positioning [15].

AFM is actively used to study effects of various

histone variants, their post-translational modifications,

and epigenetic DNA modifications on the nucleosome

structure. Histone tail modifications are the key factor

in regulation of both chromatin and cellular homeo-

stasis dynamics. A hallmark of an active chromatin is

hyperacetylation of histones, which likely results in a

more open chromatin structure. It was shown in the

work of Hizume et al. [16] by AFM that histone hyper-

acetylation causes decrease in the thickness of chro-

matin fibrils from HeLa cells. This result allowed to

confirm a model of chromatin decondensation based

on electrostatic interactions. Biotinylation, on the con-

trary, promotes chromatin condensation. In the works

by Filenko et al. [17] and Singh et al. [18] the use of

AFM made it possible to demonstrate that biotinyla-

tion of H4 histone at lysines 12 or 16 leads to compac-

tion of the nucleosome structure, which is manifested

by the increase in the number of DNA turns around

the histone octamer.

Histone replacement with variants, as well as

the nucleosome movement along the DNA molecule,

are the basic mechanisms for regulating chromatin

compaction, which, as a rule, are due to the work of

remodeler proteins. RSC (“remodeling the structure

of chromatin”) is an ATP-dependent protein complex

of chromatin structure remodelers that is homolo-

gous to the human SWI/SNF (“switch/sucrose non-fer-

mentable” family of chromatin remodeling complex-

es) and mediates the nucleosome shift. Some details

of its functioning were shown using the tapping mode

AFM. In particular, it was found out that RSC moves

nucleosome along the DNA until another nucleosome

is reached or the break in the DNA chain is encoun-

tered[19].

Comparative analysis of the results of studies on

the structure of nucleosome complex with the H3 his-

tone variant (CENP-A) performed using AFM allowed

to propose an alternative model of DNA folding in the

centromeric region of chromosomes. This model in-

volves DNA folding in the centromeric region as an ar-

ray of parallel sites in the form of a “snake” instead of

the characteristic canonical nucleosome zigzag folding

[20]. The study of nucleosome structural features by

AFM involving replacement of H2A core histone with

its recently discovered H2AL2 version of the mouse

histone, expression of which is associated with sper-

matogenesis, showed, that the DNA length in the nu-

cleosome decreases from 147 to 130 bp [21]. Shorten-

ing of the DNA in the nucleosome composition is due

to relaxation of its structure by partial unwrapping.

In the same work it was also shown using cryogenic

electron microscopy (cryoEM) that nucleosomes with

H2AL2 have a more relaxed structure with a larger an-

gle value between the DNA linker regions, which is in

agreement with the data obtained by the AFM method.

It is worth paying attention to the fact that the nucle-

osome becomes resistant to the action of such remod-

eler proteins as RSC and SWI/SNF in the process. Bio-

logical significance of this modulation of nucleosome

properties remains the subject of research.

Nucleosome structure may be modulated by bind-

ing of a protein or a protein complex. One of the most

illustrative examples is the study investigating the ef-

fect of the H1 linker histone on a nucleosome compac-

tion. Using the capabilities of AFM, it has been shown

Fig. 3. Images obtained by using the tapping mode AFM. a)Nucleosome; b)nucleosomes with PARP3 protein.

UKRAINTSEV et al.678

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

that H1 does not affect the degree of nucleosome com-

paction, but significantly stabilizes it, reducing struc-

ture fluctuations [22]. In our work on the influence of

nuclear proteins poly(ADP-ribose) polymerases 1, 2,

and 3 (PARP1, PARP2, and PARP3) on the degree of nu-

cleosome compaction, it was found by AFM that PARP3

contributes to compaction of the structure of mono-

nucleosomes [23] (Fig. 3,a andb). PARP1, PARP2, and

PARP3 take part in the most DNA-dependent processes

including regulation of the DNA transcription and DNA

repair. At the same time, they catalyze the reaction of

ADP-ribose moieties transfer to the target proteins us-

ing NAD

as a substrate in response to genomic DNA

damage. PARP3 catalyzes the mono(ADP-ribosyl)ation

reaction, while PARP1 and PARP2 catalyze synthesis

of the poly(ADP-ribose). In addition, unlike PARP1 and

PARP2, PARP3 does not interact with the HPF1 his-

tone ADP-ribosylation factor [24]. Using AFM, we were

able to establish a previously unknown feature of the

PARP3 functioning, in particular, its influence on the

nucleosome structure, which may indicate the role of

this protein in regulating chromatin structure. In this

work, catalytic reaction of the PARP enzymes proteins

on nucleosomes has not been investigated, but other

researches have observed synthesis of poly(ADP-ri-

bose) catalyzed by the PARP1 and PARP2 proteins on

the extended DNA substrates by the AFM method [25,

26], which indicated the possibility of similar studies

in the case of nucleosomes.

NON-CONTACT MODE AFM

Another modification of the AFM is the non-con-

tact mode AFM. It is based on the detection of changes

in attractive forces caused by van der Waals interac-

tions between the probe and the surface, which allows

scanning without the need for direct contact of the

cantilever with the sample. While realizing this mode,

the cantilever oscillates with low amplitude at a fre-

quency exceeding its resonant frequency, which leads

to the increase in sensitivity compared to the tapping

mode AFM. Scanning nucleosomes using this mode en-

abled to obtain an image with a resolution that even

exceeded capabilities of cryoEM at the time of this

work publication [27]. However, despite the impres-

sive advantage, this scanning mode has not been wide-

ly adopted due to technical limitations. For example,

scanning in the above-mentioned work of Davies etal.

[27] was performed under a deep vacuum. It should

be noted, that analysis of biomolecules and their com-

plexes is preferable to be carried out in a solution to

facilitate interpretation of the results, which is incom-

patible with operations under vacuum. In addition,

vacuum conditions limit the study of dynamics of the

investigated complexes.

Taking into account the limitations arising during

implementation of the non-contact mode and low

quality of the images that can be obtained by the con-

tact mode, today the tapping mode AFM has become

the most widespread in the study of biomolecules.

Moreover, such AFM variants as MAC mode, PeakForce

mode, high-speed mode, as well as electrostatic AFM

are also based on the tapping mode AFM.

MAC MODE AFM

An alternative to the classic tapping mode is the

MAC mode (magnetic alternating current mode AFM),

the principles of which are based on excitation of the

cantilever oscillations by magnetic field of the coil

located under the sample [28]. The use of MAC mode

provides lower force impact on the sample than the

classic tapping mode of AFM, which contributes to

maintaining sharpness of the cantilever tip and im-

proves quality of the resulting image. For example,

methylation of the nitrogenous base C in the CpG se-

quences has been shown to result in a tighter chroma-

tin packaging in the presence of the H1 linker histone

precisely using the MAC mode [29]. These results sug-

gested a mechanism for a transcription regulation in

large chromatin domains through methylation.

Further development of this scanning mode in-

volved modification of the cantilever with an affinity

agent; for example, an antibody, which allowed to com-

bine a topographic study of the surface with molecular

recognition of the objects in real time (PicoTREC AFM)

[30,31]. When the probe contacts the sample, the sur-

face relief is recorded, and presence of the antigen at

the scanning point could be detected from the change

of the amplitude of cantilever oscillations. The use of

this approach made it possible [32, 33] to demonstrate

the mechanism of action of the SWI/SNF chromatin

remodeler. In particular, SWI/SNF has been shown

to promote removal of the H2A histone from nucleo-

some, resulting in relaxation of the nucleosome struc-

ture and accompanied by the release of ~80bp DNA.

These results enabled characterization of the changes

in geometric parameters of the nucleosome in the case

of H2A histone deletion, although, there are data ob-

tained earlier by immunoprecipitation, quantitative

PCR, and gel electrophoresis, indicating that the H2A/

H2B histones are deleted from nucleosome under the

action of SWI/SNF chromatin remodeler [34, 35].

PEAKFORCE TAPPING MODE AFM

A non-resonant scanning mode developed by

Bruker (USA), PeakForce Tapping mode AFM, is one of

alternative modifications among the scanning probe

STRUCTURE AND FUNCTIONS OF NUCLEOSOMES BY AFM 679

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Fig. 4. A cluster of nucleosome particles [37]. Three clusters of nucleosome particles are shown in the AFM image. Two lines(1,2)

are marked. The inset in the upper left corner shows height profiles for each line(1 and2) along the measured distance. The first

profile (black line) crosses DNA and shows height and width of the double-stranded DNA. The second profile (red line) crosses

the identified nucleosome.

AFM technologies, which is promising for studying bi-

ological molecules [36]. Implementation of this mode

includes contact of the cantilever with the sample

surface at a frequency significantly lower than the

resonant one, which enables to consider such kind of

contacts as a quasi-static process, and the signal is re-

corded as a fixed maximum contact force of the sam-

ple with the probe. This scanning mode additionally

reduces the force of cantilever impact on the sample

and allows topographic studies of “soft” objects to be

carried out with great accuracy. Using this mode of

AFM for the first time allowed to obtain nucleosome

images in solution without using vacuum with resolu-

tion approaching the resolution demonstrated by the

X-ray crystallography (Fig. 4) [37].

McCauley et al. [37] in their study uniquely iden-

tify nucleosomes from the height ~5.5 nm and diame-

ter ~1nm, which correlates with the data obtained by

X-ray crystallography (height was 5.7 nm; diameter

was 11 nm) [11]. Such high resolution allowed estab-

lishing that the non-histone proteins Nhp6A and Hmo1

of chromatin from the HMGB family destabilize the

nucleosome structure with release of the part of DNA,

and perform this in different ways. Hmo1 was shown

UKRAINTSEV et al.680

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

to relax the nucleosome structure more than Nhp6A,

which is highly consistent with peculiarities of local-

ization of these proteins on chromatin. Hmo1 is con-

centrated in a highly transcribed histone-depleted

region enriched with the genes of translational ap-

paratus, where high DNA compaction is not needed,

while Nhp6A is localized in the rest of euchromatin.

ELECTROSTATIC AFM

Special attention should be paid to the scanning

probe microscopy techniques that allow to determine

electrostatic potential of a surface. They proved to be

significantly more informative for studying biomole-

cules after the single-pass scanning options were op-

timized by using the resonant cantilever frequencies

at two harmonics. Implementation of this approach

made it possible to increase dramatically sensitivity

of the method [38]. Depending on the implementation

features, the Kelvin probe force microscopy (KPFM)

[38] and the dual resonance frequency enhanced elec-

trostatic microscopy (DREEM) [39] are distinguished.

Fig. 2d shows principle of signal acquisition with

KPFM and DREEM.

DREEM allows minute changes in the electrostatic

force gradient to be detected at high resolution, mak-

ing it a powerful tool for characterizing the structure

of protein-nucleic complexes at the single-molecule

level. DREEM enables visualization of DNA located

both on the surface and inside the protein-nucleic

complex, as well as DNA in the composition of large

heterogeneous complexes with several proteins. This

approach has a great potential in studying the struc-

ture of both single nucleosomes and nucleosomes in

more complex systems.

The research of Wu et al. [39] is devoted to demon-

strating capabilities of the DREEM method to study

location of a DNA molecule in large multi-protein

complexes. A nucleosome was chosen as the example.

On the one hand, the authors showed that the over-

all topographic images of nucleosomes obtained by

DREEM and other AFM methods are the same. On the

other hand, the DREEM images demonstrate regions

with reduced intensity in the nucleosome core region

location of which correlates well with the trajectory

of DNA location in the nucleosome according to the

X-ray crystallography data (Fig. 5). This fact confirms

the validity of this approach and of the results inter-

pretation.

Fig. 5. Topographic images of nucleosomes obtained by AFM and DREEM [39]. Topographic (a, 1; b, 1) DREEM phase images

(a, 2; b, 2) and DREEM amplitude (a, 3; b, 3) of nucleosomes showing how a DNA molecule is wrapped around histones once.

c,topographic(1) and DREEM phase images of the nucleosome(2) showing that DNA is wrapped around the nucleosome twice.

Theinsets show height profiles along the line drawn through the nucleosome in topographic (c, 1) and DREEM images (c, 2).

Twodots in the graphs correspond to positions of the two dots shown on the line that mark position of the peaks correspond-

ing to DNA in the DREEM image. Schematic models of histone-wrapping DNA are shown in each DREEM image (not-to-scale

models). Crystal structure of the nucleosome superimposed on the DREEM-phase image is shown in the inset to the phase image

on thepanel(c,2)

STRUCTURE AND FUNCTIONS OF NUCLEOSOMES BY AFM 681

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Fig. 6. Topographic AFM and DREEM images of repair complexes of mismatch repair nucleotides on 2-kb long DNA containing

mismatch repair GT nucleotides [39]. a)Spatial model of the Taq MutS crystal structure (created based on PDB: 1EWQ). Subunits

A, B, and DNA are colored blue, gold, and light blue, respectively. MutS bends DNA by about 60° as it passes through the DNA

binding channel. b)Topographic AFM (b, 1), as well as DREEM-phase(b, 2) and amplitude (b, 3) images of the Taq MutS-DNA

complex. c)Superimposing the of complex model on AFM images (c, 1) and to phase images (c, 2). d)Topographic AFM (d, 1) and

DREEM phase (d, 2 and 3) images of a large complex of MutSα-MutLα-DNA containing ~10 proteins. The DNA pathway is identi-

fied as the regions with the greatest reduction in the magnitude of DREEM signals, compared to the one protein, and marked on

the inset (d, 3) in blue color; MM, expected position of mismatched nucleotides.

As the second model, Wu et al. [39] used a multi-

protein complex of non-complementary base pair of

human DNA repair system consisting of MutSa and

MutLa. Application of DREEM allowed visualization of

location of the DNA containing a non-canonical pair of

nucleotides in the composition of this complex (Fig.6).

This model is designed to demonstrate applicabil-

ity of the method for determining location of the DNA

inside a protein complex. The obtained results were

included in the work devoted to studying the MutSa

and MutLa complex functioning [40].

The same approach was used by Adkins et al. [41]

to confirm the structure of nucleosome-like particles

reconstructed using single-stranded DNA (ssNCP).

Using DREEM, it has been shown that ssNCPs have a

structure and a size similar to the parameters of ca-

nonical nucleosomes, in particular, have a disk-like

shape and contain the same number of turns of the

single-stranded DNA on the histone octamer [41].

The authors of this research suggested that the his-

tones released from nucleosome composition during

the initial stages of the DNA double-strand break re-

pair process can bind to the single-strand DNA re-

gion. The possibility of reconstitution of ssNCP invitro

has been shown. ssNCPs proved to be more dynamic

than the canonical nucleosomes and capable of ATP-

independent transfer to other regions of one- or double -

-stranded DNA. Moreover, such ATP-dependent nucleo-

some remodelers as RSC and Fun30 bind effectively to

ssNCP and become activated. Based on these results,

the authors have suggested that ssNCP could be a

marker of various cellular processes accompanied by

formation of single-stranded DNA regions, and emer-

gence of these structures could be used to regulate one

or another specific cellular process.

Despite the impressive capabilities of this approach,

it has a number of technical limitations primarily re-

lated to high requirements for the quality of cantile-

vers, which must simultaneously have high sharpness

and sufficient conductivity to obtain high-resolution

topographic and DREEM images [39].

DREEM makes it possible to visualize DNA con-

formation both on the surface and within individual

protein–DNA complexes, as well as within the compo-

sition of large heterogeneous complexes with several

proteins. Ability of DREEMs to detect minute changes

in the electrostatic force gradient in combination with

high resolution makes it a powerful tool for character-

UKRAINTSEV et al.682

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

izing structure of protein-DNA complexes at the single-

molecule level. This approach has a great potential in

studying structures of both single nucleosomes and

nucleosomes in more complex systems.

HIGH-SPEED AFM

One of the key disadvantages of the above-men-

tioned AFM modes is long time– required to obtain in-

dividual images. This feature of the method limits its

applicability for investigating dynamics of the process.

The first attempts to study dynamics of the process-

es involved the use of sequential samples scanning.

In this way, the data on the RNA polymerase moving

along DNA were obtained [42]. Using the same mode, it

was possible to demonstrate the process of nucleosome

“unwinding” [43]. Shlyakhtenko etal. [43] showed that

multiple continuous scans of the same nucleosome

field allowed eight images of the same nucleosome to

be obtained, demonstrating its transformations over

time. As a result, gradual “unwinding” of DNA from

the histone octamer culminating in the complete de-

struction of the complex was shown.

Further improvement of the method due to ap-

plication of a more productive device allowed the

Lyubchenko group to obtain more detailed data on

the dynamics of nucleosomes “unwinding” with the

H3 histone and its CENP-A centromeric version using

high-speed atomic force microscopy (HS-AFM) [44, 45].

The data obtained in these studies revealed such re-

laxation mechanisms of the nucleosome structure as

DNA bleaching, as well as the possibility of transloca-

tion of nucleosomes along DNA. The authors note that

dynamic rearrangements of centromeric chromatin

can occur in the absence of remodeling factors. At the

same time, CENP-A stabilizes nucleosome particles,

preventing complete histone dissociation during the

DNA loosening or unwrapping and ensuring rapid nu-

cleosome re-assembly during their core transfer [45].

Canonical nucleosome particles with H3 histone do

not demonstrate such plasticity, and rearrange only in

the presence of such remodeling factors as SNF2h [46].

Later some dynamics details of nucleosomes and chro-

matosomes were shown in the work by Onoa etal. [47]

using HS-AFM that clarified the mechanism of their

“unwrapping”. In particular, asymmetry of the nucle-

osome “unwrapping”, step-wise nature of the process

of histone heterodimers deletion from the nucleo-

some core, as well as lability of position of the histone

tetramer (H3-H4) × 2 in the dyad region have been

demonstrated in real time.

HS-AFM scanning made it possible to analyze dy-

namics of nucleoproteins and other biomolecule com-

plexes. An even greater increase in the scanning speed

is required to estimate kinetic characteristics of the

studied proteins. In the future, machine learning ca-

pabilities are likely be involved in achieving this goal.

Inthe study by Kato et al. [48] an approach based on

machine learning has been proposed to analyze inte-

grated simulation dynamics data from asynchronous

HS-AFM scanning. Peculiarity of the described ap-

proach is that in this case asynchronous nature of the

single passes during scanning is considered. Thus, this

method with appropriate resampling frequency has

been shown to be a powerful tool for assessing dynam-

ic behavior of the object based on the low spatial and

time resolution HS-AFM data.

COMPARISON OF AFM CAPABILITIES

WITH SOME PHYSICOCHEMICAL APPROACHES

IN NUCLEOSOME STUDIES

During the research of nucleosome structure and

functioning, including dynamics studies, a number of

physicochemical research tools are used, such as: mag-

netic and optical “tweezers”, “DNA curtains”, nuclear

magnetic resonance, X-ray crystallography analysis,

and cryoEM, as well as approaches based on FRET op-

tical methods. These methods provide a wide range of

detectable parameters with high spatial and time res-

olution.

One of these approaches that have become widely

used is based on optical tweezers that allow to apply

mechanical force directly to the molecule or complex

under study and simultaneously measure elongation

of the object. Strength of the nucleosome complex [49]

was first determined by this method. Magnetic twee-

zers which in addition to stretching can generate rota-

tional movement of an object, allowed creating model

DNA structures with supercoiling. In the work of Gupta

et al. [50] it was shown that intramolecular stress oc-

curring in the structure of DNA with positive super-

coiling alters its conformation, compared to the DNA

without supercoiling making nucleosome formation

less effective.

Nevertheless, applicability of both approaches is

limited to measuring changes of the object linear char-

acteristics and amount of the applied force. Combin-

ing these methods with fluorescence microscopy en-

hances greatly the possibilities for the researchers, but

still prevents investigation of conformational changes

within the complex. Employment of fluorescence mi-

croscopy in the FRET mode, a non-radiative mode of

transferring energy between fluorophores, partial-

ly release this limitation, however, conformational

changes recorded by this method could not be always

interpreted clearly. Despite this fact, use of FRET tech-

nique is a powerful tool for studying the structure and

conformational dynamics of protein-nucleic complexes.

For example, use of FRET allowed Andreeva et al. [51]

STRUCTURE AND FUNCTIONS OF NUCLEOSOMES BY AFM 683

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

to record the effect of buffer components on nucleo-

some structure. In particular, the stabilizing effect of

potassium ions has been shown, and this stabilization

of the nucleosome structure has also been demonstrat-

ed to create a barrier for such ATP-independent pro-

teins/complexes as FACT and PARP1.

Another approach facilitating examination of the

protein complexes with nucleic acids is based on for-

mation of a nanobarrier on the substrate surface,

which ensures anchoring of DNA molecule with for-

mation of DNA curtains. Nucleosome DNA curtains are

a unique system allowing to track behavior of thou-

sands of individual nucleosome particles containing a

fluorescent label in real time. The use of this approach

made it possible to monitor dynamics of DNA compac-

tion by condensinsI and II, as well as to clarify mech-

anistic model of chromatin loop formation by yeast

condensin and human cohesin [52-56]. Visnapuu and

Greene [57] were able to determine energy landscape

of the nucleosome localization on DNA and confirm

theoretical predictions based on the model designed

using this approach. In addition, in the same work, po-

sitions of the nucleosome assembly sites were shown

to correlate with the regulatory transcription regions.

Unfortunately, this approach does not allow investi-

gating explicitly structural changes within the nucleo-

somes themselves induced by various factors.

One of the most powerful tools in studying the

structure of biopolymers is the X-ray crystallography.

Atomic structure of nucleosome has been determined

using X-ray crystallography a long time ago [58]. De-

spite the high resolution, applicability of this method

is severely limited. Limitations are largely due to the

need to grow a crystal, which does not enable study-

ing dynamics of structural changes. Moreover, obtain-

ing a crystal is on itself not a trivial task, while the

image quality directly depends on the crystal quality.

Thetask is further complicated in the case of investiga-

tion of the structure of protein complexes with nucle-

ic acids, especially of such complexity as nucleosome,

and even more so chromatin. Obtaining a X-ray-suit-

able crystal is also difficult if the studied polymer

contains unordered regions in its composition. In ad-

dition, possible difference between the protein struc-

ture in the crystal and that in the solution should be

considered while interpreting the X-ray crystallogra-

phy data. The difficulties of decoding the obtained pri-

mary data should be also mentioned, it introduces ad-

ditional limitations on the size of the object selected

for research.

At present, cryoEM allows obtaining structures

with resolution approaching the X-ray crystallography

data. For example, in the study by Markert et al. [59]

an image of nucleosome associated with complete his-

tone deacetylation complex Rpd3S was obtained using

cryoEM with final resolution of 2.8 Å. This method is

deprived of some limitations specific to X-ray crys-

tallography. For example, cryoEM allows studying

proteins with unordered structures, while they are

difficult objects to investigate with X-ray crystallog-

raphy. Nevertheless, the use of this method has some

limitations. CryoEM works best for studying large ob-

jects over 200 kDa. Like X-ray crystallography, cryoEM

is poorly adapted to study conformational dynamics.

The achievements of recent years have provided a

possibility of time-resolution analysis in the range of

milliseconds to seconds, which allows studying con-

formational dynamics of the structure, but is hard-

ly applicable to studying such large-scale changes in

the structure as nucleosome translocation or its core

transfer.

Analysis of the biomolecule conformational dy-

namics including of their complexes such as a nu-

cleosome could be performed by NMR spectroscopy.

Advances in solution NMR spectroscopy have allowed

monitoring conformational dynamics of both the

bounded core of histones and their free N- and C-tails

[60]. Solid-state NMR techniques represent additional

tools of nucleosome research [61]. For example, second-

ary structure of the H2B histone in nucleosome was

obtained using solid-state NMR, which was consistent

with the structure obtained by X-ray crystallography.

The authors were able to show dynamic interactions

of H2B histone with H4 histone, as well as with DNA

[62]. Applicability of this method is mainly limited by

the need to introduce isotopes of different elements

into the sample for unambiguous interpretation of the

spectra. Thus, in the work of Shi etal. [62] the histone

containing

S and

N isotopes was used.

The use of AFM largely complements the data ob-

tained using the above-described approaches. In par-

ticular, AFM facilitates determination of geometric

parameters of nucleosomes, as well as their changes

due to nucleosome interactions with various protein

factors or histone replacements with their variants,

as has been shown in a case of H3 replacement with

CENP-A. The possibility of scanning in a liquid medi-

um allows simulating conditions closer to physiologi-

cal ones, and development of the high-speed scanning

made it possible to observe relaxation dynamics of the

nucleosome structure. Using the capabilities of electro-

static force microscopy facilitated observation of the

trajectory of DNA folding in the composition of nucle-

osome.

One of the considered AFM advantages is the

possibility of using this technique for investigation of

complex macromolecular structures, such as chroma-

tin fragments or model polynucleosomal fragments

[15, 63]. Investigation of such objects using FRET or

tweezers could be limited by the need to add fluo-

rescent labels or microspheres into the tested sample

structure.

UKRAINTSEV et al.684

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

CONCLUSION

Rapid development of AFM as a method of study-

ing complex protein-nucleic systems allowed to deepen

significantly understanding of the nucleosome struc-

ture. Importance of the scanning environment during

the AFM application should be emphasized. Itis well

known that implementation of AFM under vacuum

conditions has a positive effect on image quality. How-

ever, in order to simplify interpretation of the results,

it is preferable to carry out scanning in an aqueous en-

vironment when studying biomolecules. Drying of bio-

molecules like DNA, proteins, or their complexes leads

to disruption of their spatial structure [64]. In addi-

tion, scanning in an aqueous environment is necessary

to study dynamics of the processes. At the same time,

implementation of AFM in a liquid is accompanied by

a number of technical difficulties, in particular it lim-

its of the range of used substrates and methods for im-

mobilization of scanned objects, as well as it increases

requirements for cantilevers preparation.

AFM capabilities are likely to be enhanced by the

development and implementation of machine learn-

ing for primary data processing. This approach would

reduce the level of “noise” when scanning at the soft-

ware level and increase accuracy of the measurements

provided AFM. In addition, performance improvement

at the software level would increase the scan speed.

Some approaches based on the AFM technique

were not widely used so far for studying nucleosomes

should be noted. For example, combination of fluores-

cence microscopy and AFM could be used for studying

interactions of protein molecules or their complexes

with the substrates in the form of nucleosomes or

chromatin. At the same time, the AFM-derived data

on the sample topology could allow to localize inter-

action area in the protein complex with the substrate,

and the fluorescence data could facilitate identifica-

tion of the labeled molecule. A similar approach was

used by the Wyman group to investigate homologous

recombination [65].

Wide use of the combination of AFM and fluo-

rescence techniques is primarily limited by the need

to use a substrate which would be atomically smooth

while being optically transparent. At present, prepa-

ration of such substrates is a non-trivial task; one of

these solutions has been suggested by Rahman et al.

[66]. This problem is likely to be solved in the future,

opening up opportunities to an arsenal of fluorescence

detection-based approaches, including FRET and fluo-

rescence polarization combined with AFM.

Implementation of electrostatic force microscopy

in a liquid medium could significantly contribute to

studying dynamics of conformational DNA changes in

the protein-nucleic complexes including nucleosome

or chromatin. At present, measuring electrostatic po-

tential of biological molecules in liquids is limited by

high dielectric constant of the medium, by the possi-

bility of a solvate shell forming on the sample surface,

as well as by shielding the probe and sample surfaces

with solution ions. Nevertheless, some studies show

the possibility of using electrostatic force microscopy

of solid materials at low ionic strength [67, 68]. Scan-

ning electrostatic potential in liquid increases de-

mands primarily on cantilever preparation. It should

not only meet all the cantilever requirements for elec-

trostatic force microscopy in DREEM mode, but also

have a dielectric coating on the probe surface, exclud-

ing the area that is in direct contact with the sample.

Apparently, further development of the method asso-

ciated with increase in the device sensitivity would fa-

cilitate measurements of electrostatic potential of bio-

molecules in liquid.

In conclusion, it should be added that studies of

the last decade have dramatically changed the views

on the structure of chromatin and its basic unit, nu-

cleosome. The static image is now complemented by

the dynamic one, and methods providing the ability to

study single molecules have played an important role

in characterizing dynamic properties of nucleosomes.

The use of AFM offers an opportunity to characterize

a complex supramolecular system at the nanoscale,

allowing direct visualization of the structural features

of nucleosomes and changes within it such as, for ex-

ample, the process of their unwrapping. In addition,

the possibility of using small amounts of biopolymer

preparations is a favorable feature of AFM measure-

ments in comparison with some other methods.

Acknowledgments. The authors express their

gratitude to A. A. Lomzov and E. A. Belousova for a con-

sultation and productive discussions during the manu-

script preparation.

Contributions. A.A.U. and M.M.K. developed the

concept of the review and prepared the manuscript;

O.I.L. supervised the study and edited the manuscript.

Funding. This research was funded by the Rus-

sian Science Foundation (project no. 22-74-10059)

and State Budget Project of Institute of Chemical Bi-

ology and Fundamental Medicine No. 121031300041-4

(forO. I. Lavrik).

Ethics declaration. This work does not describe

any studies involving humans or animals as objects

performed by any of the authors. The authors of this

work declare that they have no conflicts of interest

infinancial or any other sphere.

REFERENCES

1. Pombo, A., and Dillon, N. (2015) Three-dimensional

genome architecture: players and mechanisms,

STRUCTURE AND FUNCTIONS OF NUCLEOSOMES BY AFM 685

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Nat. Rev. Mol. Cell Biol., 16, 245-257, doi: 10.1038/

nrm3965.

2. Kornberg, R. D., and Thomas, J. O. (1974) Chroma-

tin structure; oligomers of the histones, Science, 184,

865-868, doi:10.1126/science.184.4139.865.

3. Olins, A.L., and Olins, D.E. (1974) Spheroid chroma-

tin units (vbodies), Science, 183, 330-332, doi:10.1126/

science.183.4122.330.

4. Woodcock, C.L., Safer, J.P., and Stanchfield, J.E. (1976)

Structural repeating units in chromatin. I. Evidence

for their general occurrence, Exp. Cell Res., 97, 101-

110, doi:10.1016/0014-4827(76)90659-5.

5. Robison, A. D., and Finkelstein, I. J. (2014)

High-throughput single-molecule studies of pro-

tein-DNA interactions, FEBS Lett., 588, 3539-3546,

doi:10.1016/j.febslet.2014.05.021.

6. Teif, V.B., and Cherstvy, A. G. (2016) Chromatin and

epigenetics: current biophysical views, AIMS Biophys.,

3, 88-98, doi:10.3934/biophy.2016.1.88.

7. Fierz,B., and Poirier, M.G. (2019) Biophysics of chro-

matin dynamics, Annu. Rev. Biophys., 48, 321-345,

doi:10.1146/annurev-biophys-070317-032847.

8. Vesenka, J., Hansma, H. G., Siegerist, C., Siligardi, G.,

Schabtach, E., and Bustamante, C. (1992) Scanning

force microscopy of circular DNA and chromatin

in air and propanol, Scan. Probe Microscop., 1639,

127-137, doi:10.1117/12.58189.

9. Allen, M.J., Dong, X.F., O’Neill, T.E., Yau,P., Kowalczy-

kowski, S.C., Gatewood,J., Balhorn,R., and Bradbury,

E. M. (1993) Atomic force microscope measurements

of nucleosome cores assembled along defined DNA

sequences, Biochemistry, 32, 8390-8396, doi: 10.1021/

bi00084a002.

10. Martin, L.D., Vesenka, J.P., Henderson,E., and Dobbs,

D.L. (1995) Visualization of nucleosomal substructure

in native chromatin by atomic force microscopy, Bio-

chemistry, 34, 4610-4616, doi:10.1021/bi00014a014.

11. Richmond, T. J., Finch, J. T., Rushton, B., Rhodes, D.,

and Klug, A. (1984) Structure of the nucleosome

core particle at 7 Å resolution, Nature, 311, 532-537,

doi:10.1038/311532a0.

12. Arents,G., Burlingame, R.W., Wang, B.C., Love, W.E.,

and Moudrianakis, E.N. (1991) The nucleosomal core

histone octamer at 3.1 Å resolution: a tripartite pro-

tein assembly and a left-handed superhelix, Proc.

Natl. Acad. Sci. USA, 88, 10148-10152, doi: 10.1073/

pnas.88.22.10148.

13. Qian, R.L., Liu, Z.X., Zhou, M.Y., Xie, H.Y., Jiang,C.,

Yan, Z.J., Li, M.Q., Zhang,Y., and Hu,J. (1997) Visual-

ization of chromatin folding patterns in chicken eryth-

rocytes by atomic force microscopy (AFM), Cell Res., 7,

143-150, doi:10.1038/cr.1997.15.

14. Lowary, P.T., and Widom,J. (1998) New DNA sequence

rules for high affinity binding to histone octamer and

sequence-directed nucleosome positioning, J. Mol.

Biol., 276

, 19-42, doi:10.1006/jmbi.1997.1494.

15. Pisano, S., Pascucci, E., Cacchione, S., De Santis, P.,

and Savino, M. (2006) AFM imaging and theoretical

modeling studies of sequence-dependent nucleosome

positioning, Biophys. Chem., 124, 81-89, doi: 10.1016/

j.bpc.2006.05.012.

16. Hizume,K., Araki,S., Hata,K., Prieto,E., Kundu, T.K.,

Yoshikawa, K., and Takeyasu, K. (2010) Nano-scale

analyses of the chromatin decompaction induced by

histone acetylation, Arch. Histol. Cytol., 73, 149-163,

doi:10.1679/aohc.73.149.

17. Filenko, N. A., Kolar,C., West, J. T., Smith, S. A., Has-

san, Y.I., Borgstahl, G.E., Zempleni,J., and Lyubchen-

ko, Y.L. (2011) Therole of histoneH4 biotinylation in

the structure of nucleosomes, PLoS One, 6, e16299,

doi:10.1371/journal.pone.0016299.

18. Singh, M. P., Wijeratne, S. S., and Zempleni, J. (2013)

Biotinylation of lysine 16 in histone H4 contributes

toward nucleosome condensation, Arch. Biochem. Bio-

phys., 529, 105-111, doi:10.1016/j.abb.2012.11.005.

19. Montel, F., Castelnovo, M., Menoni, H., Angelov, D.,

Dimitrov, S., and Faivre-Moskalenko, C. (2011) RSC

remodeling of oligo-nucleosomes: an atomic force

microscopy study, Nucleic Acids Res., 39, 2571-2579,

doi:10.1093/nar/gkq1254.

20. Lyubchenko, Y. L. (2014) Centromere chromatin: a

loose grip on the nucleosome? Nat. Struct. Mol. Biol.,

21, 8, doi:10.1038/nsmb.2745.

21. Syed, S. H., Boulard, M., Shukla, M. S., Gautier, T.,

Travers, A., Bednar, J., Faivre-Moskalenko, C., Dim-

itrov, S., and Angelov, D. (2009) The incorporation

of the novel histone variant H2AL2 confers unusual

structural and functional properties of the nucleo-

some, Nucleic Acids Res., 37, 4684-4695, doi: 10.1093/

nar/gkp473.

22. Würtz, M., Aumiller, D., Gundelwein, L., Jung, P.,

Schütz,C., Lehmann,K., Tóth,K., and Rohr,K. (2019)

DNA accessibility of chromatosomes quantified by

automated image analysis of AFM data, Sci. Rep., 9,

12788, doi:10.1038/s41598-019-49163-4.

23. Ukraintsev,A., Kutuzov,M., Belousova,E., Joyeau,M.,

Golyshev,V., Lomzov,A., and Lavrik,O. (2023) PARP3

affects nucleosome compaction regulation, Int.J. Mol.

Sci., 24, 9042, doi:10.3390/ijms24109042.

24. Suskiewicz, M. J., Prokhorova, E., Rack, J. G. M., and

Ahel,I. (2023) ADP-ribosylation from molecular mech-

anisms to therapeutic implications, Cell, 186, 4475-

4495, doi:10.1016/j.cell.2023.08.030.

25. Sukhanova, M.V., Abrakhi,S., Joshi,V., Pastre,D., Ku-

tuzov, M.M., Anarbaev, R.O., Curmi, P.A., Hamon,L.,

and Lavrik, O. I. (2016) Single molecule detection of

PARP1 and PARP2 interaction with DNA strand breaks

and their poly(ADP-ribosyl)ation using high-resolution

AFM imaging, Nucleic Acids Res., 44, e60, doi:10.1093/

nar/gkv1476.

26. Sukhanova, M.V., Hamon,L., Kutuzov, M.M., Joshi,V.,

Abrakhi, S., Dobra, I., Curmi, P. A., Pastre, D., and

UKRAINTSEV et al.686

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Lavrik, O. I. (2019) A single-molecule atomic force

microscopy study of PARP1 and PARP2 recognition of

base excision repair DNA intermediates, J.Mol. Biol.,

431, 2655-2673, doi:10.1016/j.jmb.2019.05.028.

27. Davies, E., Teng, K. S., Conlan, R. S., and Wilks, S. P.

(2005) Ultra-high resolution imaging of DNA and nu-

cleosomes using non-contact atomic force microsco-

py, FEBS Lett., 579, 1702-1706, doi: 10.1016/j.febslet.

2005.02.028.

28. Han, W., Lindsay, S., and Jing, T. (1996) A magnet-

ically driven oscillating probe microscope for op-

eration in liquids, Appl. Phys. Lett., 69, 4111-4113,

doi:10.1063/1.117835.

29. Karymov, M.A., Tomschik,M., Leuba, S.H., Caiafa,P.,

and Zlatanova, J. (2001) DNA methylation-dependent

chromatin fiber compaction in vivo and in vitro: re-

quirement for linker histone, FASEBJ., 15, 2631-2641,

doi:10.1096/fj.01-0345com.

30. Stroh, C., Wang, H., Bash, R., Ashcroft, B., Nelson, J.,

Gruber, H., Lohr, D., Lindsay, S. M., and Hinterdorf-

er, P. (2004) Single-molecule recognition imaging mi-

croscopy, Proc. Natl. Acad. Sci. USA, 101, 12503-12507,

doi:10.1073/pnas.0403538101.

31. Zhang, M., Chen, G., Kumar, R., and Xu, B. (2013)

Mapping out the structural changes of natural and

pretreated plant cell wall surfaces by atomic force

microscopy single molecular recognition imag-

ing, Biotechnol. Biofuels, 6, 147, doi: 10.1186/1754-

6834-6-147.

32. Wang,H., Bash,R., Lindsay, S.M., and Lohr,D. (2005)

Solution AFM studies of human Swi-Snf and its inter-

actions with MMTV DNA and chromatin, Biophys. J.,

89, 3386-3398, doi:10.1529/biophysj.105.065391.

33. Bash, R., Wang, H., Anderson, C., Yodh, J., Hager, G.,

Lindsay, S. M., and Lohr, D. (2006) AFM imaging of

protein movements: histone H2A-H2B release during

nucleosome remodeling, FEBS Lett., 580, 4757-4761,

doi:10.1016/j.febslet.2006.06.101.

34. Bruno,M., Flaus,A., Stockdale,C., Rencurel,C., Ferrei-

ra,H., and Owen-Hughes,T. (2003) Histone H2A/H2B

dimer exchange by ATP-dependent chromatin remod-

eling activities, Mol. Cell, 12, 1599-1606, doi: 10.1016/

s1097-2765(03)00499-4.

35. Vicent, G.P., Nacht, A.S., Smith, C.L., Peterson, C.L.,

Dimitrov, S., and Beato, M. (2004) DNA instructed

displacement of histones H2A and H2B at an induc-

ible promoter, Mol. Cell, 16, 439-452, doi: 10.1016/

j.molcel.2004.10.025.

36. Xu,K., Sun,W., Shao,Y., Wei,F., Zhang,X., Wang,W.,

and Li, P. (2018) Recent development of PeakForce

Tapping mode atomic force microscopy and its appli-

cations on nanoscience, Nanotechnol. Rev., 7, 605-621,

doi:10.1515/ntrev-2018-0086.

37. McCauley, M. J., Huo, R., Becker, N., Holte, M. N.,

Muthurajan, U. M., Rouzina, I., Luger, K., Maher,

L.J.,3rd, Israeloff, N.E., and Williams, M.C. (2019) Sin-

gle and double box HMGB proteins differentially de-

stabilize nucleosomes, Nucleic Acids Res., 47, 666-678,

doi:10.1093/nar/gky1119.

38. Leung,C., Maradan,D., Kramer,A., Howorka,S., Mes-

quida, P., and Hoogenboom, B. W. (2010) Improved

Kelvin probe force microscopy for imaging individu-

al DNA molecules on insulating surfaces, Appl. Phys.

Lett., 97, 203703, doi:10.1063/1.3512867.

39. Wu,D., Kaur,P., Li, Z.M., Bradford, K.C., Wang,H., and

Erie, D.A. (2016) Visualizing the path of DNA through

proteins using DREEM imaging, Mol. Cell, 61, 315-323,

doi:10.1016/j.molcel.2015.12.012.

40. Bradford, K.C., Wilkins,H., Hao,P., Li, Z.M., Wang,B.,

Burke,D., Wu,D., Smith, A.E., Spaller,L., Du,C., Gauer,

J.W., Chan,E., Hsieh,P., Weninger, K.R., and Erie, D.A.

(2020) Dynamic human MutSα-MutLα complexes com-

pact mismatched DNA, Proc. Natl. Acad. Sci. USA, 117,

16302-16312, doi:10.1073/pnas.1918519117.

41. Adkins, N.L., Swygert, S.G., Kaur,P., Niu,H., Grigor-

yev, S.A., Sung,P., Wang,H., and Peterson, C.L. (2017)

Nucleosome-like, single-stranded DNA (ssDNA)-his-

tone octamer complexes and the implication for

DNA double strand break repair, J. Biol. Chem., 292,

5271-5281, doi:10.1074/jbc.M117.776369.

42. Guthold, M., Zhu, X., Rivetti, C., Yang, G., Thomson,

N. H., Kasas, S., Hansma, H. G., Smith, B., Hansma,

P.K., and Bustamante,C. (1999) Direct observation of

one-dimensional diffusion and transcription by Esche-

richia coli RNA polymerase, Biophys.J., 77, 2284-2294,

doi:10.1016/S0006-3495(99)77067-0.

43. Shlyakhtenko, L. S., Lushnikov, A. Y., and Lyubchen-

ko, Y. L. (2009) Dynamics of nucleosomes revealed

by time-lapse atomic force microscopy, Biochemistry,

48, 7842-7848, doi:10.1021/bi900977t.

44. Miyagi, A., Ando, T., and Lyubchenko, Y. L. (2011)

Dynamics of nucleosomes assessed with time-lapse

high-speed atomic force microscopy, Biochemistry,

50, 7901-7908, doi:10.1021/bi200946z.

45. Stumme-Diers, M.P., Banerjee,S., Hashemi,M., Sun,Z.,

and Lyubchenko, Y. L. (2018) Nanoscale dynamics

of centromere nucleosomes and the critical roles of

CENP-A, Nucleic Acids Res., 46, 94-103, doi: 10.1093/

nar/gkx933.

46. Sinha, K.K., Gross, J.D., and Narlikar, G.J. (2017) Dis-

tortion of histone octamer core promotes nucleosome

mobilization by a chromatin remodeler, Science, 355,

eaaa3761, doi:10.1126/science.aaa3761.

47. Onoa,B., Díaz-Celis,C., Cañari-Chumpitaz,C., Lee,A.,

and Bustamante,C. (2023) Real-time multistep asym-

metrical disassembly of nucleosomes and chroma-

tosomes visualized by high-speed atomic force mi-

croscopy, ACS Cent. Sci., 10, 122-137, doi: 10.1021/

acscentsci.3c00735.

48. Kato, S., Takada, S., and Fuchigami, S. (2023) Parti-

cle smoother to assimilate asynchronous movie data

of high-speed AFM with MD simulations, J. Chem.

STRUCTURE AND FUNCTIONS OF NUCLEOSOMES BY AFM 687

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Theory Comput., 19, 4678-4688, doi: 10.1021/acs.

jctc.2c01268.

49. Bennink, M.L., Leuba, S.H., Leno, G.H., Zlatanova,J.,

de Grooth, B.G., and Greve, J. (2001) Unfolding indi-

vidual nucleosomes by stretching single chromatin

fibers with optical tweezers, Nat. Struct. Biol., 8, 606-

610, doi:10.1038/89646.

50. Gupta, P., Zlatanova,J., and Tomschik,M. (2009) Nu-

cleosome assembly depends on the torsion in the DNA

molecule: a magnetic tweezers study, Biophys. J., 97,

3150-3157, doi:10.1016/j.bpj.2009.09.032.

51. Andreeva, T. V., Maluchenko, N. V., Sivkina, A. L.,

Chertkov, O. V., Valieva, M. E., Kotova, E. Y., Kirpich-

nikov, M. P., Studitsky, V. M., and Feofanov, A. V.

(2022) Na

and K

ions differently affect nucleo-

some structure, stability, and interactions with pro-

teins, Microsc. Microanal., 28, 243-253, doi: 10.1017/

S1431927621013751.

52. Ganji,M., Shaltiel, I.A., Bisht,S., Kim,E., Kalichava,A.,

Haering, C.H., and Dekker, C. (2018) Real-time imag-

ing of DNA loop extrusion by condensing, Science,

360, 102-105, doi:10.1126/science.aar7831.

53. Davidson, I.F., Bauer,B., Goetz,D., Tang,W., Wutz,G.,

and Peters, J.M. (2019) DNA loop extrusion by human

cohesin, Science, 366, 1338-1345, doi:10.1126/science.

aaz3418.

54. Kim,Y., Shi,Z., Zhang,H., Finkelstein, I.J., and Yu,H.

(2019) Human cohesin compacts DNA by loop extru-

sion, Science, 366, 1345-1349, doi: 10.1126/science.

aaz4475.

55. Kong, M., Cutts, E. E., Pan, D., Beuron, F., Kaliyap-

pan,T., Xue,C., Morris, E.P., Musacchio,A., Vannini,A.,

and Greene, E. C. (2020) Human condensin I and II

drive extensive ATP-dependent compaction of nucle-

osome-bound DNA, Mol. Cell, 79, 99-114, doi:10.1016/

j.molcel.2020.04.026.

56. Kim,E., Kerssemakers,J., Shaltiel, I.A., Haering, C.H.,

and Dekker, C. (2020) DNA-loop extruding conden-

sin complexes can traverse one another, Nature, 579,

438-442, doi:10.1038/s41586-020-2067-5.

57. Visnapuu, M.L., and Greene, E. C. (2009) Single-mol-

ecule imaging of DNA curtains reveals intrinsic ener-

gy landscapes for nucleosome deposition, Nat. Struct.

Mol. Biol., 16, 1056-1062, doi:10.1038/nsmb.1655.

58. Luger,K., Mäder, A.W., Richmond, R.K., Sargent, D.F.,

and Richmond, T.J. (1997) Crystal structure of the nu-

cleosome core particle at 2.8Å resolution, Nature, 389,

251-260, doi:10.1038/38444.

59. Markert, J.W., Vos, S.M., and Farnung,L. (2023) Struc-

ture of the complete S. cerevisiae Rpd3S- nucleosome

complex, bioRxiv, doi:10.1101/2023.08.03.551877.

60. Van Emmerik, C.L., and van Ingen,H. (2019) Unspin-

ning chromatin: revealing the dynamic nucleosome

landscape by NMR, Prog. Nucl. Magn. Reson. Spec-

trosc., 110, 1-19, doi:10.1016/j.pnmrs.2019.01.002.

61. Ackermann, B. E., and Debelouchina, G. T. (2021)

Emerging contributions of solid-state NMR spectrosco-

py to chromatin structural biology, Front. Mol. Biosci.,

8, 741581, doi:10.3389/fmolb.2021.741581.

62. Shi,X., Kannaian,B., Prasanna,C., Soman,A., and Nor-

denskiöld, L. (2023) Structural and dynamical inves-

tigation of histone H2B in well-hydrated nucleosome

core particles by solid-state NMR, Commun. Biol., 6,

672, doi:10.1038/s42003-023-05050-3.

63. Melters, D. P., Neuman, K. C., Bentahar, R. S., Rak-

shit,T., and Dalal,Y. (2023) Single molecule analysis of

CENP-A chromatin by high-speed atomic force micros-

copy, Elife, 12, e86709, doi:10.7554/eLife.86709.

64. Heenan, P. R., and Perkins, T. T. (2019) Imaging DNA

equilibrated onto mica in liquid using biochemically

relevant deposition conditions, ACS Nano, 13, 4220-

4229, doi:10.1021/acsnano.8b09234.

65. Sanchez,H., Kertokalio,A., van Rossum-Fikkert,S., Ka-

naar,R., and Wyman,C. (2013) Combined optical and

topographic imaging reveals different arrangements

of human RAD54 with presynaptic and postsynaptic

RAD51-DNA filaments, Proc. Natl. Acad. Sci. USA, 110,

11385-11390, doi:10.1073/pnas.1306467110.

66. Rahman,M., Boggs,Z., Neff,D., and Norton,M. (2018)

The Sapphire (0001) Surface: a transparent and ul-

traflat substrate for DNA nanostructure imaging,

Langmuir, 34, 15014-15020, doi:10.1021/acs.langmuir.

8b01851.

67. Johnson, A.S., Nehl, C. L., Mason, M.G., and Hafner,

J.H. (2003) Fluid electric force microscopy for charge

density mapping in biological systems, Langmuir, 19,

10007-10010, doi:10.1021/la035255f.

68. Gramse,G., Gomila,G., and Fumagalli,L. (2012) Quan-

tifying the dielectric constant of thick insulators by

electrostatic force microscopy: effects of the micro-

scopic parts of the probe, Nanotechnology, 23, 205703,

doi:10.1088/0957-4484/23/20/205703.

Publisher’s Note. Pleiades Publishing remains

neutral with regard to jurisdictional claims in pub-

lished maps and institutional affiliations.