Article

ISSN 0006-2979, Biochemistry (Moscow), 2025, Vol. 90, No. 11, pp. 1584-1601 © Pleiades Publishing, Ltd., 2025.

Russian Text © The Author(s), 2025, published in Biokhimiya, 2025, Vol. 90, No. 11, pp. 1689-1707.

1584

REVIEW

Biomolecular Condensates in the Regulation

of Transcription and Chromatin Architecture

Arseniy V. Selivanovskiy

1,2,3

, Sergey V. Razin

1,3

, and Sergei V. Ulianov

1,3,a

Institute of Gene Biology, Russian Academy of Sciences, 119334 Moscow, Russia

Moscow Institute of Physics and Technology, 141700Dolgoprudny, Russia

Faculty of Biology, Lomonosov Moscow State University, 119234 Moscow, Russia

e-mail: sergey.v.ulyanov@gmail.com

Received August 26, 2025

Revised October 1, 2025

Accepted October 3, 2025

Abstract— Recent studies have highlighted the pivotal role of biomolecular condensates (liquid-like complexes)

in gene control. Biomolecular condensates create a specific microenvironment around enhancers and gene

promoters, which can activate transcription, repress it, or maintain at an appropriate level. They can also

influence the chromatin structure and are important participants in the enhancer–promoter communication.

Finally, biomolecular condensates represent promising therapeutic targets, as their dysregulation results in a

broad spectrum of pathologies. The review present most recent, as well as fundamental studies establishing

the role of condensates in the regulation of gene expression and enhancer–promoter communication.

DOI: 10.1134/S0006297925602746

Keywords: phase separation, transcriptional condensates, RNA polymeraseII, transcription factors, loop extru-

sion, multivalent interactions, enhancers, promoters, chromatin

* To whom correspondence should be addressed.

INTRODUCTION

The activation of eukaryotic gene transcription

requires a coordinated action of transcription factors

(TFs), co-activators, and RNA polymerase  2 (RNAP2).

These proteins bind to chromatin in the regions of

gene promoters and enhancers (regulatory DNA se-

quences) [1,  2]. Large enhancers that bind the high-

est number of transcription regulators are referred

to as super-enhancers[3,  4]. Enhancers and super-en-

hancers form spatial contacts with activated genes.

This process is facilitated by the ring-shaped ATPase

cohesin that extrudes chromatin loops [5]. Another

key participant of the enhancer–promoter (EP) com-

munication is the CTCF protein, which acts as a phys-

ical barrier preventing the movement of the cohesin

complex [6, 7].

Approximately 30years ago, RNAP2 clusters were

observed in the nucleus by electron microscopy. These

clusters were often associated with several cis-regu-

latory elements and, therefore, designated as “tran-

scription factories” [8]. The following development

of light microscopy and genome editing techniques

has made it possible to observe dynamic clusters of

protein transcription regulators with the properties

of liquid-phase condensates in live cells. Based on the

concept of phase separation, they were named tran-

scriptional condensates (TCs) [9] (Fig. 1).

Biomolecular condensates (Fig.  1) are nonstoi-

chiometric complexes formed as a result of multiva-

lent interactions between their components (proteins,

RNA, DNA). Although structured protein domains may

play an important role in these interactions [10], the

latter almost always, to a greater or lesser extent, oc-

cur between intrinsically disordered regions (IDRs).

Such interactions are highly specific and maintain a

constant composition of the condensates despite the

absence of membrane envelope  [11]. The key factors

providing specific interactions between the IDRs are

their repetitive short linear motifs (SLiMs) [11-14],

typically composed of 4 to 12 amino acid residues

(no more than 8 residues in most cases) [11, 13, 15].

Despite their small length, SLiMs are evolutionary

conserved sequences. The interactions between SLiMs

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1585

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

Fig. 1. The model of TC in the region of EP contact. The condensate concentrates transcription factors (TFs), co-activa-

tors (Mediator), and RNA polymerase 2 (RNAP2) and promotes transcription. Protein components of the condensate often

contain extended unstructured domains (shown in RNAP2 and TF).

are diverse and include hydrophobic, π–π, π–cation,

electrostatic, dipole–dipole interactions, and hydro-

gen bonds [12]. Mutations in these repeats impair

phase separation, while their post-translational mod-

ification regulates phase formation in cells [12,  13].

At the same time, usually only one amino acid residue

within a SLiM is crucial for the phase separation[15].

Hence, in contrast to the interaction between the

structured domains, which is based on the mutual

recognition of protein fragments that are tens or hun-

dreds amino acids in length, the central role in the

interaction between IDRs belongs to individual amino

acids [16, 17].

The formation of condensates is finely regulated

via numerous mechanisms. The propensity of proteins

for the phase separation varies between cell types and

cellular compartments [18]. It depends on the protein

concentration, post-translational modifications, het-

erotypic interactions, the presence of multimerization

domains, and external conditions (e.g., temperature

and pH) [9]. In addition, the clustering of TF-binding

sites on chromatin and their spatial proximity create

the regions of locally high TF concentration.

In this review, we briefly describe the main ex-

perimental approaches used in the studies of TCs, the

functions of TCs in the regulation of transcription and

chromatin architecture, and their role in the devel-

opment of various pathologies. Multiple mechanisms

have been proposed to explain the formation of TCs,

the most discussed of them being liquid–liquid phase

separation (LLPS)[19], surface condensation[20], and

phase separation coupled to percolation (PSCP) [21].

All these mechanisms describe the formation of non-

stoichiometric dynamic complexes driven by weak

multivalent interactions. In this review, as well as in

some other works [22,  23], such complexes are re-

ferred to as TCs.

METHODS FOR TC VISUALIZATION

Since the size of TCs is often at the diffraction

limit, their detection required the development of

super-resolution microscopy techniques, such as

tcPALM (time-correlated photoactivation localization

microscopy) [24]. The most common method of TC

visualization is immunofluorescence, which allows to

visualize focal clusters of transcription proteins and

observe their transition from the uniform distribu-

tion in the nucleoplasm or cytosol to clusters upon

exposure to a stimulus [25-28]. Besides its relative

simplicity, the advantages of this method include the

possibility to work with endogenous protein concen-

trations in cells. An obvious disadvantage is that this

method requires cell fixation, thus preventing inves-

tigation of the dynamic properties of such clusters.

In addition, a fixator (usually formaldehyde or para-

formaldehyde) can interfere with the process of con-

densate formation by disturbing or, on the contrary,

promoting protein–protein or DNA–protein inter-

actions [29-32].

The use of endogenously labeled proteins helps

to bypass these limitations. For example, CRISPR-me-

diated knock-ins can be employed to add a fluoro-

phore (most often, fluorescent or photoconvertible

protein or HaloTag) to the protein reading frame,

which allows to study the dynamics of formation/dis-

solution of TCs and to observe their liquid-like behav-

ior in  vivo [33-35]. Thus, FRAP (fluorescence recovery

after photobleaching) was used to demonstrate both

diffusion within the condensates and exchange of

molecules between the condensates and their envi-

ronment [36]. Although expression of a labeled pro-

tein from an endogenous promoter makes it possi-

ble to avoid its overexpression, some studies have

used exogenously expressed labeled proteins, usually,

SELIVANOVSKIY et al.1586

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

in cells not expressing such protein under normal

conditions [37, 38] or cells where this protein was

knocked out/knocked down [39]. In these cases, a

doxycycline- or tetracycline-inducible promoter is typ-

ically used to adjust protein expression levels to the

endogenous ones.

Super-resolution microscopy makes it possible

not only to visualize condensates, but also to evalu-

ate their colocalization with the active transcription

sites in live and fixed cells. DNA- or RNA-FISH (flu-

orescence in  situ hybridization) techniques are used

in fixed cells [24, 28, 33, 40, 41], while in live cells,

the transcripts can be visualized using modified nu-

cleotides (e.g., 5′-ethynyl uridine) or viral tags (usually

MS2/MCP or PP7/PCP systems) [15, 18, 24, 42].

Light microscopy can be helpful in evaluating the

size, shape, and dynamics of condensates, as well as

in determining the site and time of their emergence.

However, it provides little information on the TC com-

position, because in most studies, the condensates are

visualized using only one or two component(s), e.g.,

Mediator and RNAP2. A more detailed characteriza-

tion of the TC composition can be achieved by using

proteomics-based approaches.

METHODS FOR THE ANALYSIS

OF TC COMPOSITION

Characterization of the TC proteome is import-

ant for understanding the action mechanisms of

condensates. For example, using biotinylated inac-

tive Cas9 nuclease (dead Cas9, dCas9) fused with

the FUS protein IDR allowed to identify the key

transcription and architecture proteins in TCs [43].

The authors of [44] used inducible delivery of bio-

tin ligase (potentially, any enzyme) in a complex

with IDR of the ubiquitous transcriptional co-activa-

tor BRD4 for the local biotinylation of all TC compo-

nents [44]. In both studies [43,  44], RNAP2, Mediator,

BRD4, and other transcription proteins were found as

the most frequently occurring TC components. Using

chemical crosslinking and mass spectrometry, FUS

was identified as a key partner of the TAZ TF, re-

quired to maintain fluidity and robust transcription-

al activity of TAZ condensates  [28]. Finally, a similar

method helped to identify AMPK (AMP-dependent ki-

nase) as a negative regulator of pathological conden-

sates of the FOXM1 TF and a promising therapeutic

target [16].

Combined with modern microscopy techniques,

proteomics methods allow detailed characterization

of the composition of TCs in live cells. However, it

remains unclear whether condensates make any spe-

cific contribution to the transcription regulation in

addition to that of the soluble complexes.

METHODS FOR EVALUATION

OF CONDENSATE FUNCTIONALITY

The attention of researchers is currently focused

on the biological functions of condensates compared

to soluble protein complexes. There are several ways

to resolve this issue.

1)  Creation of artificial TCs and analysis of their

transcription activation capacity. The condensates are

often reconstructed on the studied genes through the

targeted recruitment of dCas9–IDR complexes with

fluorescent proteins, which enables visualization of

condensates in live cells and evaluation of their bio-

physical properties[43,45]. Such systems are used for

the targeted regulation of both transcription [46] and

chromatin architecture [47,  48] (see “Transcriptional

condensates as transcriptional activators”). Another

frequently used approach is the use of the LacO/LacI-

or TetO/TetR-based systems.

2)  Genetic complementation[49]. The principle of

this method consists in the identification of domains

responsible for the phase separation followed by their

substitution with functionally analogous domains of

other proteins. If such substitution restores both the

biological function of the protein and its ability to

form condensates, this is indicative of the conden-

sate functionality. However, this does not exclude the

contribution of soluble complexes[19, 33, 50-52]. It is

important that the replaced domains lack the amino

acid sequence similarity (18% on the average[49], but

can reach 0% [51]) despite their common role in the

phase separation. The substitution of domains does

not always restore the protein biological function, be-

cause the amino acid composition of IDRs may affect

the consistency of condensates or their ability to con-

centrate co-activators. In particular, the substitution

of the IDR of MYOCD (myocardin) by IDRs from FUS,

EWS, or DDX4 proteins fully restored the ability of

MYOCD to form condensates and activate transcrip-

tion, while its replacement with the IDR from CDT1 led

to the formation of nonfunctional condensates unable

to concentrate RNAP2 and Mediator. Another example

is histone deacetylase UTX, whose phase separation

underlies its chromatin-regulatory activity in tumor

suppression. The substitution of its IDR by the IDRs

from eIF4G2 and AKAP95 (8 and 18%identity, respec-

tively) restored the condensates and their tumor-sup-

pressing activity. At the same time, replacement with

the IDR from its paralog UTY (74% identity) resulted

in the formation of more solid, nonfunctional conden-

sates [52]. Interestingly, the catalytic activity of UTX is

not necessary for its tumor suppression function, in

contrast to the ability to form condensates [52].

3)  Point mutagenesis aimed at the uncoupling

of protein–protein interaction from the phase sep-

aration [53-56]. However, in some cases, these two

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1587

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

activities cannot be separated. For example, in the

yeast transcription factor Gcn4, the same amino acid

residues are involved in the protein interaction with

Med15 and condensate formation [53]. Hence, these

proteins exist in live cells as both soluble complexes

and phase condensates [53]. At the same time, in the

IDR of the chromatin remodeling factor ARID1A/B,

the residues responsible for the condensate formation

and interaction with other proteins are different, and

both activities are essential for the protein binding to

chromatin and implementation of its biological func-

tions [54].

4)  Recruitment of solubilizing proteins (e.g., fruc-

tose- or mannose-binding proteins), which dissolve

condensates but do not abolish the protein–protein

interactions [57-60], to condensates. A major advan-

tage of this approach is its high selectivity. Moreover,

it does not affect the level of synthesis of cellular pro-

teins and does not require introduction of mutations

into them. These methods allow identification of genes

specifically regulated by the condensates [57-59].

5)  Treatment of cells with small organic mole-

cules that dissolve condensates without changing the

levels of synthesis of their protein components or pro-

tein–protein interactions[27, 34,61]. Such approaches

are effectively used for the regulation of gene expres-

sion in both cultured cell and live organisms.

In addition, by using live-cell microscopy, it was

shown that condensates form on the regulated genes

before the onset of transcription [62]. For example,

one of the recent works demonstrated that the emer-

gence of large dynamic clusters of the Nanog TF

onthe actively expressed mir430 gene in Danio rerio

embryos preceded the start of its transcription [63].

Another example is clustering of nonphosphorylated

RNAP2 on various genes in mouse embryonic stem

cells (mESCs) before the transcription initiation [64].

At least for some proteins, e.g., MYOCD, the forma-

tion of condensates and activation of transcription

occurred at the same critical threshold concentration.

The condensates formed specifically on the activated

genes, presumably, due to the MYOCD interaction with

TFs [33]. Therefore, available methods demonstrate

the contribution of TCs to the regulation of gene ex-

pression, including both activation and repression

of transcription.

TRANSCRIPTIONAL CONDENSATES

AS TRANSCRIPTIONAL ACTIVATORS

The colocalization of condensates and active tran-

scription sites in live cells has been shown for both

native and synthetic TCs[18, 24, 26, 28, 33, 40-43, 45,

65,66]. Early observations demonstrated a direct cor-

relation between the levels of mRNA synthesis and

stability of transcriptional factories associated with

the β-actin gene in live mouse cells[62]. Later experi-

ments with endogenously labeled proteins showed that

the key transcriptional activators BRD4, Mediator, and

RNAP2 formed dynamic condensates on active genes

in mESCs [40]. Immunofluorescence analysis in com-

bination with RNA- and DNA-FISH revealed that these

condensates form on super-enhancers and are in a

close proximity to or overlap with the sites of mRNA

synthesis[24,  40]. Moreover, TCs colocalized (although

episodically) with the MS2-labeled transcripts of the

Esrrb gene actively expressed in mESCs [24]. Later

studies have shown a significant inverse correlation

between the levels of gene transcription and distance

to the TC [35,  36]. TCs formed in mESCs were over

300  nm in size and contained up to 400 Mediator and

RNAP2 molecules [24], suggesting their nonstoichio-

metric nature. Promoter-associated clusters contain-

ing 10 to 90 molecules of RNAP2 phosphorylated at

Ser5 at the moment of transcriptional bursting, were

identified using super-resolution microscopy [67,  68].

Beside the regulation of transcription initiation, pro-

teins forming condensates with the unstructured C-ter-

minal domain (CTD) of RNAP2 were found to regulate

transcription elongation[65, 69], splicing [70] and, in

some cases, transcription termination [50].

Optogenetic techniques were used to reveal a di-

rect correlation between the intensity of condensate

fluorescence and transcription levels of associated

genes [18]. The use of dCas9 in a complex with the

CRY2 domain allowed to create TCs at the genomic

loci of interest and to analyze their effects on the

transcription of individual genes[43,  45,  71,  72]. Such

synthetic condensates efficiently concentrated tran-

scriptional co-activators and RNAP2 phosphorylated

at Ser2  [46]. In HeLa cells, synthetic light-induced

condensates based on dCas9 and guide RNAs to the

beta-globin gene HS2 enhancer and BCL11A gene

promoter, increased transcription of the target genes

11-23 and 24-35 times, respectively[43]. The complex-

es of the viral protein VP64 (commonly used tran-

scriptional activator) with IDRs of the phase-form-

ing proteins FUS and NUP98 enhanced transcription

much more efficiently than VP64 alone [45, 72, 73].

Finally, the DroprCRISPRa system based on the FUS

IDR fused with dCas12a-VP64 activated transcription

in vivo in mice [46]. Besides providing targeted ex-

pression activation, these approaches demonstrate

the need for the optimal intensity of multivalent in-

teractions in order to ensure an efficient transcription

activation [45,  71,  73]. For example, mutations in the

IDR of FUS, causing the dissolution of condensates or

changes in their material properties, reduced tran-

scription [46].

These observations are in good agreement

with the data obtained for native TFs. For example,

SELIVANOVSKIY et al.1588

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

Fig.  2. The “kiss-and-kick” and the re-entrant phase transition models of the transcriptional bursts. In the kiss-and-kick

model [24, 35], transcriptional bursts are initiated by the periodic gene convergence with TCs and enhancers. In the

model of recurrent phase transitions [81], TCs form on weakly expressed genes, thus enhancing their transcription,

and then dissolve at a high local transcript concentration (negative feedback).

in the presence of androgens, diffusely distributed

androgen receptor (AR) translocated to the nucleus

and formed condensates on enhancers via the IDR–

IDR interactions [61]. The treatment of cells with

1,6-hexanediol, which dissolved the AR condensates

but did not affect the level of AR biosynthesis, sig-

nificantly reduced chromatin accessibility and tran-

scription on the androgen-regulated enhancers [61].

At the same time, an increased number of glutamine

residues in the IDR of AR in various pathological

states results in the formation of stable aggregates

and noticeable decrease in the enhancer activity [61].

A similar picture was observed for the excessive stim-

ulation of estrogen enhancers [74]. In general, an in-

crease in the multivalency can inhibit transcription

due to alterations in the material properties of con-

densates or their formation outside of the chromatin

body [23, 74-76].

The intensity of multivalent interactions can be

influenced by external factors. For example, at tem-

peratures above 27°C, the nucleoplasmic conden-

sates in Arabidopsis thaliana cells concentrate the

ELF3 protein (negative regulator of flowering time),

thus preventing its binding to chromatin and repres-

sion of genes involved in flower development  [77].

The biomolecular condensates that play a key role

in the regulation of flowering are typically tempera-

ture-sensitive  [78]. The transcription factor Hsf1, an

activator of heat shock response genes, uses a sim-

ilar mechanism to form condensates at elevated

temperatures  [79]. TCs can also appear in response

topH changes. Thus, acidification of the nucleoplasm

in macrophages, which is typical for inflammatory

processes, partially dissolved BRD4 and MED1 con-

densates via protonation of His residues in the IDR

of BRD4  [80]. This mechanism most strongly reduc-

es transcription of proinflammatory genes regulated

by distant super-enhancers (negative feedback-mech-

anism) [80]. Hence, TCs can act as sensors of ex-

ternal conditions.

The formation of TCs allows to explain some fea-

tures of eukaryotic transcription. For example, it is

known that transcription of eukaryotic and some pro-

karyotic genes occurs in bursts, which can be due to

the periodic formation of contacts between the genes

and TCs  [35]. For example, the intensity of the Sox2

gene transcription in mESCs was proportional to its

proximity to the condensate associated with its su-

per-enhancer[35] (Fig.  2). The alternative mechanism

is periodic emergence and dissolution of condensates

due to the electrostatic repulsion of mRNAs accumu-

lated in the active transcription sites [81] (Fig.  2).

This suggestion was confirmed by the fact that the

treatment of cells with transcription elongation inhib-

itors stabilized Mediator-containing initiatory conden-

sates[81]. At the same time, this treatment led to the

dissolution of the elongation condensates appearing

due to the interaction between proteins and the CTD

of RNAP2 phosphorylated at Ser2 [65].

Special attention has been focused on a potential

role of condensates in the regulation of TF binding to

DNA [22, 82-86]. If the properties of the nucleoplasm

were similar to those of a diluted homogeneous solu-

tion, then, according to the Smoluchowski equation,

it would take days for a TF to find its binding mo-

tifs in the volume of the nucleus[82]. However, DNA,

proteins, and RNA are intrinsically present in the nu-

cleus at the concentrations that prevent free diffu-

sion and create the effect of macromolecular crowd-

ing [87], which could significantly increase the time

required for such search. However, some transcrip-

tional responses are observed within several minutes.

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1589

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

Phase separation intensifies the interactions between

some TFs and chromatin[16,  19] (see also “Transcrip-

tional Condensates in Pathology” section), while chro-

matin-associated condensates help some TFs find their

binding sites[80,82]. The role of IDRs in the search of

binding motifs on DNA has been well demonstrated.

It supplements and, in some cases, even exceeds the

significance of DNA-binding domains in the identifica-

tion of binding sites[79,83]. Interestingly, TFs diffuse

slower inside the condensates than in the nucleop-

lasm  [76]. Analysis of diffusion trajectories of single

TF molecules revealed two types of diffusion: rapid

free and slow limited  [88]. At least in some cases,

limited diffusion required the presence of IDRs  [89].

Based on these observations, it was suggested that lo-

cal diffusion does occur inside the condensates, which

facilitates the binding of TFs to DNA and slows down

their dissociation [84, 89].

However, the situation might be more compli-

cated. For example, in the yeast Msn2 TF, extended

IDRs play a key role in the search of binding sites

independently of the phase separation, as Msn2 is dif-

fusely distributed in the nucleoplasm [90]. The lim-

ited diffusion of RNAP2 can be observed in the ab-

sence of evident signs of condensate formation [91].

To summarize, although IDRs and multivalent inter-

actions help TFs find their binding sites, the need for

condensates in this process remains debatable.

The role of TCs is not always limited to the tran-

scription activation or repression; sometimes, it is to

maintain required levels of gene expression. This

phenomenon was observed in mESCs, which stably

express differentiation genes at low levels. Recent

study has shown an association between these genes

and specific dual-activity TFs combining the func-

tions of transcriptional activators and repressors[34].

Using endogenous fluorescence labeling of inves-

tigated TFs, the authors were able to observe the

formed condensates in vivo. The condensates neither

colocalized with the markers of active and inactive

chromatin, nor associated with the bivalent chro-

matin. In contrast to classical TCs, the condensates

formed by the dual-action TFs concentrated RNAP2

very moderately and were almost entirely depleted of

Mediator. Moreover, artificial reconstruction of these

condensates on chromatin in HEK293 cells stabilized

intermediate expression levels of genes that had been

transcribed above or below these levels. Experiments

with reporter genes have shown that after reaching

a threshold concentration necessary for the phase

separation, further increase in the TF concentration

did not lead to any significant enhancement of tran-

scription. This fact distinguishes dual-action TFs from

the classical activators, which activate transcription

proportionally to their recruited amount [34]. Finally,

chimeric proteins obtained by fusing these TFs with

solubilizing proteins (e.g., mannose-binding protein)

did not form condensates and displayed no dual ac-

tion at relatively low concentrations; however, both

effects were restored at the high concentrations.

Therefore, the formation of microcompartments with

a specific protein composition characterized by the

absence of repressors and relatively low content of

transcription activators allows to maintain gene ex-

pression at a necessary level [34].

Some TCs, in particular, those including compo-

nents of facultative and constitutive heterochroma-

tin and regulators of the promoter-proximal pausing

of RNAP2, specialize in the repression of regulated

genes.

TRANSCRIPTIONAL CONDENSATES

AS TRANSCRIPTIONAL REPRESSORS

Some molecular condensates are formed by

the key components of heterochromatin. One of

these components is MeCP2, which binds methylated

DNA and histones and represses transcription by ei-

ther displacing transcription activators or recruiting

corepressors (e.g., histone deacetylases). In mESCs,

MeCP2 formed condensates that selectively concen-

trated transcription repressors [92]. MeCP2 droplets

actively concentrated HP1-α, but did not fuse with

the BRD4 or MED1 droplets even after physical con-

tact in  vitro [92]. Mutations affecting the ability of

MeCP2 to form the condensates reduced its capacity

to bind DNA and repress transcription [92].

Repressive condensates can also emerge in re-

sponse to changing environmental conditions. NELF

(negative elongation factor) is homogeneously dis-

tributed in the nucleoplasm under normal condi-

tions, but forms condensates under heat stress [25].

The substitution of its IDR by IDRs from FUS or

EWSR1 restored both cluster formation and NELF-

mediated repression [25].

Other important transcription repressors are pro-

teins from the Polycomb group (in particular, com-

ponents of the PRC1 and PRC2 complexes) that are

necessary for the facultative heterochromatin forma-

tion. Polycomb condensates were originally discov-

ered by the super-resolution light microscopy[84,93].

Recently, their ultrastructure and mechanisms of

formation were elucidated by electron tomography.

For example, a subunit of the PRC1–CBX8 com-

plex condenses with chromatin due to the multiva-

lent interactions with DNA and nucleosomes [94].

Although chromatin in the CBX8 condensates was

more static, there were pores between the nucleo-

somes  [94] which allowed the passage of complexes

up to 600 kDa (~8  nm). Hence, RNAP2 (~550  kDa)

or CBX8 (~43  kDa) could freely diffuse within such

SELIVANOVSKIY et al.1590

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

heterochromatin, and transcription repression was

achieved not due to the chromatin compaction but

rather to the inability of RNAP2 to stably associate

with chromatin [94-97].

Polycomb proteins are also required for the X

chromosome inactivation, a mechanism that provides

balanced expression of X-linked genes in males and

females. The key role in this process in placental

mammals belongs to the long noncoding RNA Xist

that recruits transcriptional repressors. However, the

mechanism of its distribution over inactive X chro-

mosome had remained unclear for a long time. Re-

cently, it was demonstrated that it occurs through the

formation of a repressive condensate covering the

entire X chromosome [98]. At the same time, it was

found that HNRNPK protein condensates concentrate

Xist and its protein partners, leading to the increase

in their adhesiveness and fluidity invitro, which pre-

sumably limits Xist diffusion and facilitates its distri-

bution in cis in vivo [98].

Another interesting example of transcription-

al repression was observed for the circadian genes

of Drosophila melanogaster [29], which are period-

ically activated and inactivated throughout the day.

The binding of the PER and CLK TFs in the promot-

er regions of these genes leads to their repression.

Endogenous labeling of PER in D.  melanogaster live

cells showed that during repression, PER forms sev-

eral large (~300-400  nm) condensates that colocalize

with the repressed genes[29] and translocate them to

the nuclear lamina by interacting with lamin B [29].

Therefore, condensates are also able to position ge-

nomic loci in the nuclear space and to determine

chromatin architecture.

The architectural function of condensates is uni-

versally observed. In most cases, it involves creation

and maintenance of loop contacts[21, 99,100]. At the

same time, the condensates often partially or com-

pletely associate with super-enhancers, which are

probable sites of their emergence [35, 40, 101].

TRANSCRIPTIONAL CONDENSATES

AS MEDIATORS IN THE EP COMMUNICATION

The concept of TCs was originally based on the

observations of super-enhancers as structures with

an abnormally high level of associated transcription-

al IDR-containing proteins [102-104] and emerging

as a result of cooperative assembly through a sin-

gle nucleation event [40, 105]. Super-enhancers form

spatial hubs that do not depend on the loop extru-

sion[106]. Taken together, these observations suggest-

ed the existence of TCs [105], which have then been

found on super-enhancers in live cells [24, 26, 40,

107]. The ability of super-enhancers to participate

in the nucleation of condensates was demonstrat-

ed in vitro [101].

The idea that condensates can regulate chromatin

architecture has originated from the two types of ev-

idence. First, it was found that mutations disrupting

phase separation result in the weakening of genome

loops, while the overexpression of phase-forming IDRs

strengthens certain EP contacts [19, 21, 99]. Second,

artificial nucleation of condensates on chromatin can

alter its 3D structure [43, 47, 48, 108]. In particular,

condensates artificially reconstructed on chromatin

were able to form loops and concentrate cohesin[43].

Moreover, many proteins impeding cohesin move-

ment (e.g., RNAP2, Mediator, MAZ, RUNX2, and other

TFs) form condensates, although the detailed relation-

ship between these activities requires further eluci-

dation [109]. At the same time, condensates can stop

at least some molecular motors on chromatin [50].

There is evidence (although contradictory) [110, 111]

that the CTCF protein, the most canonical barrier for

the cohesin-dependent loop extrusion, also forms con-

densates [112, 113]. In some cases, chromatin folding

promotes the emergence of condensates. Thus, the

chromatin framework formed by CTCF and cohesin

is necessary for the appearance of BRD4 and RNAP2

clusters in human HCT116 cells [110], presumably,

due to the convergence of actively transcribed ge-

nomic elements. Also, a recent study showed that the

high local density of piRNA (piwi-interacting RNA)

genes in C.elegans germ cells is crucial for the emer-

gence of condensates activating expression of these

genes [14].

Regardless of whether formation of condensates

is a consequence or a mechanism of chromatin loop

formation/maintenance, the condensate paradigm ex-

plains many observations about EP communication,

such as long (comparable to the condensate size) dis-

tances separating active enhancers and corresponding

promoters [114], coregulation of several genes by a

single enhancer [66,  115], co-expression of spatial-

ly convergent genes [116], coupling of transcription

on enhancers and promoters [115,  117], formation of

hubs of multiple enhancers and promoters [118-120],

as well as the increase in the local viscosity between

them [121]. All the above implies the existence of a

common compartment where the enhancer and the

promoter can exchange associated regulators [122].

Another important indicator of enhancer activity is

the production of enhancer RNAs, which, in many

cases, contribute to the condensate formation and can

play the crucial role in the EP communication [123].

Therefore, condensates play the key role in the

EP communication, chromatin architecture, and gene

regulation. Impairments in their formation, localiza-

tion, or composition can lead to the development

of various pathologies.

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1591

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

TRANSCRIPTIONAL CONDENSATES

IN PATHOLOGY

An impaired regulation of TC formation and/or

composition can result in various pathologies, such

as disorders in the development of the nervous sys-

tem [92] and limbs [55], viral infections, and many

types of cancer. Recently, tens of thousands of mu-

tations in condensate-forming proteins have been

identified, which can lead to more than 1000 types

of genetic disorders and hundreds of types of cancer

[124]. These pathologies result from the formation of

aberrant condensates and changes in their material

properties or localization [124] (Fig. 3a).

Disruptions in the condensate localization as

the result of the derepression of endogenous ret-

roviruses led to embryonic lethality in mice [125].

RNA transcribed from the endogenous retroviruses

effectively recruited RNAP2 and Mediator, resulting

in the TC emergence on retrotransposons and their

disappearance on enhancers and promoters [125]

(Fig.  3b). The authors also observed the hyperactiva-

tion of genes located closely to the derepressed ret-

roviruses [125]. Hence, in this case, the derepressed

retroviruses exerted a pathological effect not via

retrotransposition, but through the disruption of TC

localization. Similarly, the condensates of ICP22 pro-

tein of the human herpes virus (HSV-1) successfully

competed with the genome of infected cells for ac-

tive (capable of elongation) RNAP2 phosphorylated

at Ser2 [126].

Another example of pathology associated with

the disruptions in the condensate localization and

material properties is brachyphalangy, polydactyly

and tibial aplasia syndrome (BPTAS). It is caused by

the frameshift mutations in the IDR of the chroma-

tin protein HMGB1, which alter its amino acid se-

quence [127]. The role of impaired phase separation

in the development of this disorder has been demon-

strated in a recent study [127]. Under normal condi-

tions, HMGB1 is localized in the nucleoplasm, where

it forms numerous small dynamic (capable of rapid

fluorescence recovery) spherical clusters [127]. The

frameshift mutations associated with BPTAS lead to

the enrichment of the arginine residues in the HMGB1

IDR, which is typical of nucleolar proteins. As a con-

sequence, the protein is redistributed to the nucleo-

lus. The frameshift also results in the appearance of

a hydrophobic patch in the IDR, leading to solidifi-

cation of nucleolar aggregates of HMGB1. The solidi-

fied aggregates of the mutant protein are irregularly

shaped and much less dynamic. Pathological clusters

of HMGB1 in the nucleolus lead to the organelle dys-

function and defects in the limb formation [127].

Although mutant HMGB1 noticeably reduced the sur-

vival of U2OS cells, this effect was neutralized by

the exogenous expression of HMGB1 variant lacking

the hydrophobic patch [127], thus showing the cru-

cial significance of physical nature of HMGB1 clusters

for the BPTAS development.

Mutations in IDRs altering the composition of

condensates can also result in pathologies. For ex-

ample, an increased number of alanine residues in

the IDR of HOXD13 (regulator of limb development)

significantly reduces the ability of condensates con-

taining this TF to concentrate Mediator and to ac-

tivate transcription, leading to the development of

synpolydactyly [55]. A similar situation is observed

for the HOXA13, RUNX2 and TBP TFs [55].

The formation of chimeric transcription factors

via chromosomal translocations underlies many types

of cancer. In these cases, the structured DNA- or nu-

cleosome-binding domain of one protein is fused

with the IDR of another protein, which enables the

mutant protein to form condensates and affects its

protein interactome. Quite often, such aberrant con-

densates concentrate large amounts of transcriptional

co-activators and hyperactivate transcription of asso-

ciated genes [128, 129]. For example, chimeric TFs

generated by the fusion of the DNA-binding domain

of HOXA9 and IDR of nucleoporin NUP98, form leuke-

mia-causing condensates on chromatin [19, 128]. The

phenylalanine/glycine repeats in the IDR of NUP98

contribute to the concentration of cofactors and pro-

mote oncogene expression [128]. Similarly, mutations

in the ENL protein recognizing acetylated chromatin,

lead to the emergence of condensates excessively

concentrating transcription elongation factors, result-

ing in the cancer development [23]. Interestingly, the

mutant TF in this case is formed due to the point

mutations in the structured domain and IDR and not

as a result of translocation [23]. A recent study has

shown that the amino acid sequences of transacti-

vation domains in chimeric oncogenic TFs typically

recruit RNAP2 more intensively due to the increased

number of aromatic residues and amino acids inter-

acting with them, which promotes overexpression of

the target genes [129].

In addition to the hyperactivation of target genes

via a higher recruitment of RNAP2, aberrant conden-

sates can form new EP contacts. For example, the chi-

meric TF NUP98-HOXA9 forms super-enhancers and

CTCF-independent spatial contacts with oncogenes as

a result of phase separation [19, 119]. The substitu-

tion the specific Phe residues in the IDR of NUP98-

HOXA9, making it incapable of phase separation, led

to the disappearance of EP contacts anchored by

this factor and significantly reduced pathogenicity

of the chimeric protein [19]. Another example is the

Ewing sarcoma, in which aberrant EP contacts are

formed due to the chimeric EWS/FLI1 TF and chro-

matin remodeler ARID1A [39, 76, 130].

SELIVANOVSKIY et al.1592

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

Fig. 3. The role of condensates in the development of pathologies. a)Condensate involvement in the pathology development;

b)expression deregulation caused by disruptions in the condensate localization as a result of derepression of endogenous

retroviruses competing with the genomic loci for the transcription proteins.

The therapeutic approaches for dissolving aber-

rant TCs are being actively developed [16, 131]. For

example, in breast cancer cells, aberrant TCs are

formed by the FOXM1 TF. This promotes extensive

FOXM1 binding to chromatin and increases its ac-

tivating capacity [16]. AMPK, which phosphorylates

the IDR of FOXM1 at a single serine residue, acts as

an antagonist of this process. The agonists of AMPK

and synthetic peptides containing phosphorylated

serine facilitate dissolution of FOXM1 condensates in

live cells [16]. The treatment of cells with these pep-

tides significantly reduced the proliferative potential

of cells. When injected into mice as components of

nanoparticles liposomes, such peptides significantly

reduced tumor growth and metastasis and activated

the immune system [16].

Micropeptides can also produce a therapeutic

effect, although via the opposite mechanism, i.e., by

causing the solidification of condensates. Recently, a

17-amino acid micropeptide with a high propensity

for oligomerization has been obtained [132], whose

targeted delivery to condensates (including oncogen-

ic condensates formed by chimeric TFs, viral conden-

sates, and nucleolus) made them much less dynamic

and fully stopped exchange of components between

the condensates and their environment [132]. At the

same time, the interactions between the soluble com-

plexes were not affected, which emphasizes the key

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1593

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

contribution of condensates to the pathogenesis of

diseases. The treatment of murine leukemia tumor

cells isolated from the bone marrow with this pep-

tide decreased their proliferative potential to nearly

zero[132]. Similarly, this peptide significantly reduced

production of new viral particles in infected HEK293T

cells [132].

CONCLUSION

Recent studies of TCs have resulted in a consid-

erable progress in understanding molecular mecha-

nisms involved in the regulation of gene expression

and chromatin architecture. New methods of su-

per-resolution microscopy and genome editing have

demonstrated the possibility of transcriptional regula-

tion by dynamic biomolecular condensates emerging

due to the multivalent interactions, which determine

their liquid nature and nonstoichiometric composi-

tion.

Over the past seven years, significant advanc-

es have been achieved in understanding the role of

TCs in the transcription regulation. TCs are involved

in the differentiation of animal and plants cells [40,

41, 77], stress response [79, 118], signaling [27, 38],

and many other processes. Due to their selectivity,

condensates can create specific microenvironments

characterized by a unique composition that can facili-

tate transcription activation or repression[15, 34,92].

At least in some cases, the formation of condensates

increases the enzymatic activity of chromatin-associ-

ated proteins [52] and promotes the activation func-

tion of TFs [133]. Moreover, phase separation can

be accompanied by the appearance of new material

properties. For example, it has been shown recently

that in live cells, the condensates of BRD4-NUT and

BRD4S TFs behave as a viscous liquid while on chro-

matin, thus limiting diffusion of these TFs and mo-

bility of nucleosomes associated with them [134]. It

is interesting that the zones of TCs corresponded to

the A compartments [134].

However, many aspects of TC functioning remain

poorly studied. In what situations do condensates

become necessary for the transcription activation/re-

pression compared to soluble complexes? It is possible

that in most cases, both condensates and soluble com-

plexes are involved in the transcription regulation,

although there are genes whose expression strongly

depends on phase separation. The mechanisms of TC

formation still require detailed investigation. For ex-

ample, it remains unclear why DNA in some cases

facilitates [92, 101] and in other cases prevents [53]

the formation of condensates of DNA-associated pro-

teins. Other interesting issues are the relationship

between TCs and loop extrusion in the formation of

the genome 3D conformation [109] and contribution

of phase separation to the maintenance of large loops

based on Polycomb proteins.

Recent studies have emphasized the necessity of

the optimal level of multivalent interactions for the

proper regulation of gene expression and demonstrat-

ed that the phase separation can not only activate

but also inhibit transcription [75, 76]. Interestingly,

phase separation can enhance the activator function

of TFs by reducing their binding specificity[133]. The

disruption of this fine balance can lead to a wide

range of pathologies. Hence, condensates are actively

studied as therapeutic targets. Manipulations with TCs

have already helped to control gene expression in cul-

tured cells and mice. Elucidation of molecular mecha-

nisms of TC formation and regulation will become an

important step in understanding the universal prin-

ciples of cell nucleus organization and transcription

regulation.

Abbreviations

EP enhancer–promoter

IDR intrinsically disordered region

mESC mouse embryonic stem cell

RNAP2 RNA polymerase2

TC transcription condensate

TF transcription factor

Contributions

A.V.S. analyzed the data, wrote and edited the text of

the article, and prepared the figures; S.V.U. and S.V.R.

edited the manuscript.

Funding

The work was supported by the Russian Science Foun-

dation, project no.21-64-00001-P.

Ethics approval and consent to participate

This work does not contain any studies involving hu-

man or animal subjects.

Conflict of interest

The authors of this work declare that they have

noconflicts of interest.

REFERENCES

1. Andersson, R., and Sandelin, A. (2020) Determi-

nants of enhancer and promoter activities of regu-

latory elements, Nat. Rev. Genet., 21, 71-87, https://

doi.org/10.1038/s41576-019-0173-8.

2. Vernimmen, D., and Bickmore, W. A. (2015) The hi-

erarchy of transcriptional activation: from enhancer

to promoter, Trends Genet., 31, 696-708, https://

doi.org/10.1016/j.tig.2015.10.004.

SELIVANOVSKIY et al.1594

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

3. Grosveld, F., van Staalduinen, J., and Stadhouders, R.

(2021) Transcriptional regulation by (super)enhancers:

from discovery to mechanisms, Annu. Rev. Genom-

ics Hum. Genet., 22, 127-146, https://doi.org/10.1146/

annurev-genom-122220-093818.

4. Blobel, G. A., Higgs, D. R., Mitchell, J. A., Notani, D.,

and Young, R. A. (2021) Testing the super-enhancer

concept, Nat. Rev. Genet., 22, 749-755, https://

doi.org/10.1038/s41576-021-00398-w.

5. Karpinska, M.A., and Oudelaar, A.M. (2023) The role

of loop extrusion in enhancer-mediated gene acti-

vation, Curr. Opin. Genet. Dev., 79, 102022, https://

doi.org/10.1016/j.gde.2023.102022.

6. Oudelaar, A. M., and Higgs, D. R. (2021) The rela-

tionship between genome structure and function,

Nat. Rev. Genet., 22, 154-168, https://doi.org/10.1038/

s41576-020-00303-x.

7. Li,Y., Haarhuis, J.H.I., Sedeño Cacciatore,Á., Olden-

kamp,R., van Ruiten, M.S., Willems,L., Teunissen,H.,

Muir, K.W., de Wit,E., Rowland, B.D., and Panne,D.

(2020) The structural basis for cohesin-CTCF-anchored

loops, Nature, 578, 472-476, https://doi.org/10.1038/

s41586-019-1910-z.

8. Rippe,K., and Papantonis,A. (2025) RNA polymerase

II transcription compartments – from factories to

condensates, Nat. Rev. Genet., 26, 775-788, https://

doi.org/10.1038/s41576-025-00859-6.

9. Pei, G., Lyons, H., Li, P., and Sabari, B. R. (2025)

Transcription regulation by biomolecular conden-

sates, Nat. Rev. Mol. Cell Biol., 26, 213-236, https://

doi.org/10.1038/s41580-024-00789-x.

10. Boeynaems, S., Alberti, S., Fawzi, N. L., Mittag, T.,

Polymenidou, M., Rousseau, F., Schymkowitz, J.,

Shorter,J., Wolozin,B., Van Den Bosch,L., Tompa,P.,

and Fuxreiter, M. (2018) Protein phase separation: a

new phase in cell biology, Trends Cell Biol., 28, 420-

435, https://doi.org/10.1016/j.tcb.2018.02.004.

11. Holehouse, A. S., and Alberti, S. (2025) Molecu-

lar determinants of condensate composition, Mol.

Cell, 85, 290-308, https://doi.org/10.1016/j.molcel.

2024.12.021.

12. Cermakova, K., and Hodges, H. C. (2023) Interac-

tion modules that impart specificity to disordered

protein, Trends Biochem. Sci., 48, 477-490, https://

doi.org/10.1016/j.tibs.2023.01.004.

13. Van der Lee, R., Buljan, M., Lang, B., Weatheritt,

R. J., Daughdrill, G. W., Dunker, A. K., Fuxreiter, M.,

Gough,J., Gsponer,J., Jones, D.T., Kim, P.M., Kriwacki,

R. W., Oldfield, C. J., Pappu, R. V., Tompa, P., Uversky,

V.N., Wright, P.E., and Babu, M.M. (2014) Classifica-

tion of intrinsically disordered regions and proteins,

Chem. Rev., 114, 6589-6631, https://doi.org/10.1021/

cr400525m.

14. Davey, N. E., Van Roey, K., Weatheritt, R. J.,

Toedt, G., Uyar, B., Altenberg, B., Budd, A., Diella, F.,

Dinkel, H., and Gibson, T. J. (2012) Attributes of

short linear motifs, Mol. Biosyst., 8, 268-281, https://

doi.org/10.1039/c1mb05231d.

15. Xu, M., Jiang, S.-Y., Tang, S., Zhu, M., Hu, Y., Li, J.,

Yan, J., Qin, C., Tan, D., An, Y., Qu, Y., Song, B.-L.,

Ma, H., and Qi, W. (2025) Nuclear SREBP2 conden-

sates regulate the transcriptional activation of lipo-

genic genes and cholesterol homeostasis, Nat. Me-

tab., 7, 1034-1051, https://doi.org/10.1038/s42255-025-

01291-0.

16. Xie,F., Zhou,X., Ran,Y., Li,R., Zou,J., Wan,S., Su,P.,

Meng, X., Yan, H., Lu, H., Ru, H., Hu, H., Mao, Z.,

Yang, B., Zhou, F., and Zhang, L. (2025) Targeting

FOXM1 condensates reduces breast tumour growth

and metastasis, Nature, 638, 1112-1121, https://

doi.org/10.1038/s41586-024-08421-w.

17. Hess, N., and Joseph, J. A. (2025) Structured protein

domains enter the spotlight: modulators of biomo-

lecular condensate form and function, Trends Bio-

chem. Sci., 50, 206-223, https://doi.org/10.1016/j.tibs.

2024.12.008.

18. Wei, M.-T., Chang, Y.-C., Shimobayashi, S. F., Shin, Y.,

Strom, A. R., and Brangwynne, C. P. (2020) Nucleat-

ed transcriptional condensates amplify gene expres-

sion, Nat. Cell Biol., 22, 1187-1196, https://doi.org/

10.1038/s41556-020-00578-6.

19. Ahn, J. H., Davis, E. S., Daugird, T. A., Zhao, S.,

Quiroga, I. Y., Uryu, H., Li, J., Storey, A. J., Tsai,

Y.-H., Keeley, D. P., Mackintosh, S. G., Edmondson,

R. D., Byrum, S. D., Cai, L., Tackett, A. J., Zheng, D.,

Legant, W.R., Phanstiel, D.H., and Wang, G.G. (2021)

Phase separation drives aberrant chromatin loop-

ing and cancer development, Nature, 595, 591-595,

https://doi.org/10.1038/s41586-021-03662-5.

20. Morin, J. A., Wittmann, S., Choubey, S., Klosin, A.,

Golfier, S., Hyman, A. A., Jülicher, F., and Grill,

S. W. (2022) Sequence-dependent surface condensa-

tion of a pioneer transcription factor on DNA, Nat.

Phys., 18, 271-276, https://doi.org/10.1038/s41567-021-

01462-2.

21. Foran,G., Hallam, R.D., Megaly,M., Turgambayeva,A.,

Antfolk,D., Li,Y., Luca, V. C., and Necakov, A. (2024)

Notch1 Phase Separation Coupled Percolation fa-

cilitates target gene expression and enhancer loop-

ing, Sci. Rep., 14, 21912, https://doi.org/10.1038/

s41598-024-71634-6.

22. Stortz, M., Presman, D. M., and Levi, V. (2024) Tran-

scriptional condensates: a blessing or a curse for gene

regulation? Commun. Biol., 7, 187, https://doi.org/

10.1038/s42003-024-05892-5.

23. Song, L., Yao, X., Li, H., Peng, B., Boka, A. P., Liu, Y.,

Chen,G., Liu,Z., Mathias, K.M., Xia,L., Li,Q., Mir,M.,

Li, Y., Li, H., and Wan, L. (2022) Hotspot mutations

in the structured ENL YEATS domain link aberrant

transcriptional condensates and cancer, Mol. Cell,

82, 4080-4098.e12, https://doi.org/10.1016/j.molcel.

2022.09.034.

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1595

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

24. Cho, W.-K., Spille, J.-H., Hecht, M., Lee, C., Li, C.,

Grube, V., and Cisse, I. I. (2018) Mediator and RNA

polymerase II clusters associate in transcription-

dependent condensates, Science, 361, 412-415, https://

doi.org/10.1126/science.aar4199.

25. Rawat,P., Boehning,M., Hummel,B., Aprile- Garcia,F.,

Pandit, A.S., Eisenhardt,N., Khavaran,A., Niskanen,E.,

Vos, S. M., Palvimo, J. J., Pichler, A., Cramer, P., and

Sawarkar, R. (2021) Stress-induced nuclear condensa-

tion of NELF drives transcriptional downregulation,

Mol. Cell, 81, 1013-1026.e11, https://doi.org/10.1016/

j.molcel.2021.01.016.

26. Zamudio, A. V., Dall’Agnese, A., Henninger, J. E.,

Manteiga, J. C., Afeyan, L. K., Hannett, N. M., Coffey,

E.L., Li, C.H., Oksuz,O., Sabari, B.R., Boija,A., Klein,

I. A., Hawken, S. W., Spille, J.-H., Decker, T.-M., Cisse,

I.I., Abraham, B.J., Lee, T.I., Taatjes, D.J., Schuijers,J.,

and Young, R.A. (2019) Mediator condensates localize

signaling factors to key cell identity genes, Mol. Cell, 76,

753-766.e6, https://doi.org/10.1016/j.molcel.2019.08.016.

27. Xie,J., He,H., Kong,W., Li,Z., Gao,Z., Xie,D., Sun,L.,

Fan,X., Jiang,X., Zheng,Q., Li,G., Zhu,J., and Zhu,G.

(2022) Targeting androgen receptor phase sepa-

ration to overcome antiandrogen resistance, Nat.

Chem. Biol., 18, 1341-1350, https://doi.org/10.1038/

s41589-022-01151-y.

28. Shao,Y., Shu, X., Lu, Y., Zhu, W., Li, R., Fu, H., Li,C.,

Sun, W., Li, Z., Zhang, Y., Cao, X., Ye, X., Ajiboye, E.,

Zhao, B., Zhang, L., Wu, H., Feng, X.-H., Yang, B.,

and Lu, H. (2024) A chaperone-like function of FUS

ensures TAZ condensate dynamics and transcrip-

tional activation, Nat. Cell Biol., 26, 86-99, https://

doi.org/10.1038/s41556-023-01309-3.

29. Xiao, Y., Yuan, Y., Jimenez, M., Soni, N., and

Yadlapalli, S. (2021) Clock proteins regulate spa-

tiotemporal organization of clock genes to control

circadian rhythms, Proc. Natl. Acad. Sci. USA, 118,

e2019756118, https://doi.org/10.1073/pnas.2019756118.

30. Irgen-Gioro,S., Yoshida,S., Walling,V., and Chong, S.

(2022) Fixation can change the appearance of phase

separation in living cells, eLife, 11, e79903, https://

doi.org/10.7554/eLife.79903.

31. Schnell, U., Dijk, F., Sjollema, K. A., and Giepmans,

B. N. G. (2012) Immunolabeling artifacts and the

need for live-cell imaging, Nat. Methods, 9, 152-158,

https://doi.org/10.1038/nmeth.1855.

32. Teves, S.S., An,L., Hansen, A.S., Xie,L., Darzacq,X.,

and Tjian,R. (2016) A dynamic mode of mitotic book-

marking by transcription factors, eLife, 5, e22280,

https://doi.org/10.7554/eLife.22280.

33. Gan, P., Eppert, M., De La Cruz, N., Lyons, H., Shah,

A.M., Veettil, R.T., Chen,K., Pradhan,P., Bezprozvan-

naya,S., Xu,L., Liu,N., Olson, E.N., and Sabari, B.R.

(2024) Coactivator condensation drives cardiovascu-

lar cell lineage specification, Sci. Adv., 10, eadk7160,

https://doi.org/10.1126/sciadv.adk7160.

34. He, J., Huo, X., Pei, G., Jia, Z., Yan, Y., Yu, J., Qu, H.,

Xie, Y., Yuan, J., Zheng, Y., Hu, Y., Shi, M., You, K.,

Li, T., Ma, T., Zhang, M. Q., Ding, S., Li, P., and Li, Y.

(2024) Dual-role transcription factors stabilize inter-

mediate expression levels, Cell, 187, 2746-2766.e25,

https://doi.org/10.1016/j.cell.2024.03.023.

35. Du,M., Stitzinger, S.H., Spille, J.-H., Cho, W.-K., Lee,C.,

Hijaz, M., Quintana, A., and Cissé, I. I. (2024) Direct

observation of a condensate effect on super- enhancer

controlled gene bursting, Cell, 187, 331-344.e17,

https://doi.org/10.1016/j.cell.2023.12.005.

36. Lu, Y., Wu, T., Gutman, O., Lu, H., Zhou, Q., Henis,

Y. I., and Luo, K. (2020) Phase separation of TAZ

compartmentalizes the transcription machinery to

promote gene expression, Nat. Cell Biol., 22, 453-464,

https://doi.org/10.1038/s41556-020-0485-0.

37. Datta, D., Navalkar, A., Sakunthala, A., Paul, A.,

Patel, K., Masurkar, S., Gadhe, L., Manna, S.,

Bhattacharyya, A., Sengupta, S., Poudyal, M.,

Devi, J., Sawner, A. S., Kadu, P., Shaw, R., Pandey, S.,

Mukherjee,S., Gahlot,N., Sengupta,K., and Maji, S.K.

(2024) Nucleo-cytoplasmic environment modulates

spatiotemporal p53 phase separation, Sci. Adv., 10,

eads0427, https://doi.org/10.1126/sciadv.ads0427.

38. Stortz, M., Pecci, A., Presman, D. M., and Levi, V.

(2020) Unraveling the molecular interactions in-

volved in phase separation of glucocorticoid recep-

tor, BMC Biol., 18, 59, https://doi.org/10.1186/s12915-

020-00788-2.

39. Kim, Y. R., Joo, J., Lee, H. J., Kim, C., Park, J.-C., Yu,

Y.S., Kim, C.R., Lee, D.H., Cha,J., Kwon,H., Hanssen,

K. M., Grünewald, T. G. P., Choi, M., Han, I., Bae, S.,

Jung, I., Shin, Y., and Baek, S. H. (2024) Prion-like

domain mediated phase separation of ARID1A pro-

motes oncogenic potential of Ewing’s sarcoma, Nat.

Commun., 15, 6569, https://doi.org/10.1038/s41467-

024-51050-0.

40. Sabari, B. R., Dall’Agnese, A., Boija, A., Klein, I. A.,

Coffey, E. L., Shrinivas, K., Abraham, B. J., Hannett,

N. M., Zamudio, A. V., Manteiga, J. C., Li, C. H., Guo,

Y. E., Day, D. S., Schuijers, J., Vasile, E., Malik, S.,

Hnisz, D., Lee, T. I., Cisse, I. I., Roeder, R. G., Sharp,

P.A., Chakraborty, A.K., and Young, R.A. (2018) Coact-

ivator condensation at super-enhancers links phase

separation and gene control, Science, 361, eaar3958,

https://doi.org/10.1126/science.aar3958.

41. Huoh, Y.-S., Zhang, Q., Törner, R., Baca, S. C., Ar-

thanari, H., and Hur, S. (2024) Mechanism for con-

trolled assembly of transcriptional condensates by

Aire, Nat. Immunol., 25, 1580-1592, https://doi.org/

10.1038/s41590-024-01922-w.

42. Ma, L., Gao, Z., Wu, J., Zhong, B., Xie, Y., Huang, W.,

and Lin, Y. (2021) Co-condensation between tran-

scription factor and coactivator p300 modulates

transcriptional bursting kinetics, Mol. Cell, 81, 1682-

1697.e7, https://doi.org/10.1016/j.molcel.2021.01.031.

SELIVANOVSKIY et al.1596

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

43. Kim, Y. J., Lee, M., Lee, Y.-T., Jing, J., Sanders, J. T.,

Botten, G. A., He, L., Lyu, J., Zhang, Y., Mettlen, M.,

Ly,P., Zhou,Y., and Xu,J. (2023) Light-activated macro-

molecular phase separation modulates transcription

by reconfiguring chromatin interactions, Sci. Adv.,

9, eadg1123, https://doi.org/10.1126/sciadv.adg1123.

44. Lee, C., Quintana, A., Suppanz, I., Gomez-Auli, A.,

Mittler, G., and Cissé, I. I. (2024) Light-induced tar-

geting enables proteomics on endogenous conden-

sates, Cell, 187, 7079-7090.e17, https://doi.org/10.1016/

j.cell.2024.09.040.

45. Schneider, N., Wieland, F.-G., Kong, D., Fischer,

A.A.M., Hörner,M., Timmer,J., Ye,H., and Weber,W.

(2021) Liquid-liquid phase separation of light-induc-

ible transcription factors increases transcription ac-

tivation in mammalian cells and mice, Sci. Adv., 7,

eabd3568, https://doi.org/10.1126/sciadv.abd3568.

46. Ma, S., Liao, K., Li, M., Wang, X., Lv, J., Zhang, X.,

Huang, H., Li, L., Huang, T., Guo, X., Lin, Y., and

Rong, Z. (2023) Phase-separated DropCRISPRa plat-

form for efficient gene activation in mammalian cells

and mice, Nucleic Acids Res., 51, 5271-5284, https://

doi.org/10.1093/nar/gkad301.

47. Shin,Y., Chang, Y.-C., Lee, D.S. W., Berry,J., Sanders,

D.W., Ronceray,P., Wingreen, N.S., Haataja,M., and

Brangwynne, C.P. (2018) Liquid nuclear condensates

mechanically sense and restructure the genome,

Cell, 175, 1481-1491.e13, https://doi.org/10.1016/j.cell.

2018.10.057.

48. Strom, A.R., Kim,Y., Zhao,H., Chang, Y.-C., Orlovsky,

N. D., Košmrlj, A., Storm, C., and Brangwynne,

C. P. (2024) Condensate interfacial forces reposi-

tion DNA loci and probe chromatin viscoelastici-

ty, Cell, 187, 5282-5297.e20, https://doi.org/10.1016/

j.cell.2024.07.034.

49. Sabari, B. R., Hyman, A. A., and Hnisz, D. (2025)

Functional specificity in biomolecular condensates

revealed by genetic complementation, Nat. Rev.

Genet., 26, 279-290, https://doi.org/10.1038/s41576-

024-00780-4.

50. Liu, W., Deng, L., Wang, M., Liu, X., Ouyang, X.,

Wang, Y., Miao, N., Luo, X., Wu, X., Lu, X., Xv, X.,

Zhang, T., Li, Y., Ji, J., Qiao, Z., Wang, S., Guan, L.,

Li, D., Dang, Y., Liu, C., Li, W., Zhang, Y., Wang, Z.,

Chen, F. X., Chen, C., Lin, C., Goh, W. S. S., Zhou, W.,

Luo, Z., Gao, P., Li, P., and Yu, Y. (2025) Pcf11/Spt5

condensates stall RNA polymerase II to facilitate

termination and piRNA-guided heterochromatin

formation, Mol. Cell, 85, 929-947.e10, https://doi.org/

10.1016/j.molcel.2025.01.023.

51. Kuang, J., Zhai, Z., Li, P., Shi, R., Guo, W., Yao, Y.,

Guo,J., Zhao,G., He,J., Xu,S., Wu,C., Yu,S., Zhou,C.,

Wu, L., Qin, Y., Cai, B., Li, W., Wu, Z., Li, X., Chu, S.,

Yang, T., Wang, B., Cao, S., Li, D., Zhang, X., Chen, J.,

Liu, J., and Pei, D. (2021) SS18 regulates pluripo-

tent-somatic transition through phase separation,

Nat. Commun., 12, 4090, https://doi.org/10.1038/

s41467-021-24373-5.

52. Shi, B., Li, W., Song, Y., Wang, Z., Ju, R., Ulman, A.,

Hu, J., Palomba, F., Zhao, Y., Le, J. P., Jarrard, W.,

Dimoff, D., Digman, M. A., Gratton, E., Zang, C., and

Jiang, H. (2021) UTX condensation underlies its tu-

mour-suppressive activity, Nature, 597, 726-731,

https://doi.org/10.1038/s41586-021-03903-7.

53. Bremer, A., Lang, W. H., Kempen, R. P., Sweta, K.,

Taylor, A.B., Borgia, M.B., Ansari, A.Z., and Mittag,T.

(2025) Reconciling competing models on the roles

of condensates and soluble complexes in transcrip-

tion factor function, Mol. Cell, 85, 2718-2732.e7,

https://doi.org/10.1016/j.molcel.2025.06.008.

54. Patil, A., Strom, A. R., Paulo, J. A., Collings, C. K.,

Ruff, K.M., Shinn, M. K., Sankar,A., Cervantes, K.S.,

Wauer,T., St. Laurent, J.D., Xu,G., Becker, L.A., Gygi,

S.P., Pappu, R.V., Brangwynne, C. P., and Kadoch, C.

(2023) A disordered region controls cBAF activi-

ty via condensation and partner recruitment, Cell,

186, 4936-4955.e26, https://doi.org/10.1016/j.cell.2023.

08.032.

55. Basu,S., Mackowiak, S.D., Niskanen,H., Knezevic,D.,

Asimi, V., Grosswendt, S., Geertsema, H., Ali, S.,

Jerković, I., Ewers, H., Mundlos, S., Meissner, A.,

Ibrahim, D. M., and Hnisz, D. (2020) Unblending of

transcriptional condensates in human repeat expan-

sion disease, Cell, 181, 1062-1079.e30, https://doi.org/

10.1016/j.cell.2020.04.018.

56. Kosno,M., Currie, S.L., Kumar,A., Xing,C., and Rosen,

M. K. (2023) Molecular features driving condensate

formation and gene expression by the BRD4-NUT

fusion oncoprotein are overlapping but distinct,

Sci. Rep., 13, 11907, https://doi.org/10.1038/s41598-

023-39102-9.

57. Yang, J., Chung, C.-I., Koach, J., Liu, H., Navalkar, A.,

He, H., Ma, Z., Zhao, Q., Yang, X., He, L., Mittag, T.,

Shen,Y., Weiss, W. A., and Shu,X. (2024) MYC phase

separation selectively modulates the transcrip-

tome, Nat. Struct. Mol. Biol., 31, 1567-1579, https://

doi.org/10.1038/s41594-024-01322-6.

58. Chung, C.-I., Yang, J., Yang, X., Liu, H., Ma, Z.,

Szulzewsky, F., Holland, E. C., Shen, Y., and Shu, X.

(2024) Phase separation of YAP-MAML2 differen-

tially regulates the transcriptome, Proc. Natl. Acad.

Sci. USA, 121, e2310430121, https://doi.org/10.1073/

pnas.2310430121.

59. Brumbaugh-Reed, E.H., Gao,Y., Aoki,K., and Toettcher,

J.E. (2024) Rapid and reversible dissolution of biomo-

lecular condensates using light-controlled recruitment

of a solubility tag, Nat. Commun., 15, 6717, https://

doi.org/10.1038/s41467-024-50858-0.

60. Zhang, J. Z., Greenwood, N., Hernandez, J., Cuperus,

J. T., Huang, B., Ryder, B. D., Queitsch, C., Gestwicki,

J. E., and Baker, D. (2025) De novo designed Hsp70

activator dissolves intracellular condensates,

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1597

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

Cell Chem. Biol., 32, 463-473.e6, https://doi.org/10.1016/j.

chembiol.2025.01.006.

61. Chen,L., Zhang,Z., Han,Q., Maity, B.K., Rodrigues,L.,

Zboril,E., Adhikari,R., Ko, S.-H., Li,X., Yoshida, S.R.,

Xue, P., Smith, E., Xu, K., Wang, Q., Huang, T. H.-M.,

Chong, S., and Liu, Z. (2023) Hormone-induced en-

hancer assembly requires an optimal level of hor-

mone receptor multivalent interactions, Mol. Cell,

83, 3438-3456.e12, https://doi.org/10.1016/j.molcel.

2023.08.027.

62. Cho, W.-K., Jayanth, N., English, B. P., Inoue, T.,

Andrews, J.O., Conway,W., Grimm, J.B., Spille, J.-H.,

Lavis, L. D., Lionnet, T., and Cisse, I. I. (2016) RNA

Polymerase II cluster dynamics predict mRNA output

in living cells, eLife, 5, e13617, https://doi.org/10.7554/

eLife.13617.

63. Chabot, N. M., Purkanti, R., Ridolfi, A. D. P., Nogare,

D. D., Oda, H., Kimura, H., Jug, F., Co, A. D., and

Vastenhouw, N. L. (2024) Local DNA compaction

creates TF-DNA clusters that enable transcription,

bioRxiv, https://doi.org/10.1101/2024.07.25.605169.

64. Wang, H., Li, B., Zuo, L., Wang, B., Yan, Y., Tian, K.,

Zhou, R., Wang, C., Chen, X., Jiang, Y., Zheng, H.,

Qin, F., Zhang, B., Yu, Y., Liu, C.-P., Xu, Y., Gao, J.,

Qi, Z., Deng, W., and Ji, X. (2022) The transcriptional

coactivator RUVBL2 regulates Pol II clustering with

diverse transcription factors, Nat. Commun., 13, 5703,

https://doi.org/10.1038/s41467-022-33433-3.

65. Guo, Y. E., Manteiga, J. C., Henninger, J. E., Sabari,

B. R., Dall’Agnese, A., Hannett, N. M., Spille, J.-H.,

Afeyan, L.K., Zamudio, A.V., Shrinivas,K., Abraham,

B. J., Boija, A., Decker, T.-M., Rimel, J. K., Fant, C. B.,

Lee, T. I., Cisse, I. I., Sharp, P. A., Taatjes, D. J., and

Young, R.A. (2019) Pol II phosphorylation regulates a

switch between transcriptional and splicing conden-

sates, Nature, 572, 543-548, https://doi.org/10.1038/

s41586-019-1464-0.

66. Kawasaki, K., and Fukaya, T. (2023) Functional co-

ordination between transcription factor cluster-

ing and gene activity, Mol. Cell, 83, 1605-1622.e9,

https://doi.org/10.1016/j.molcel.2023.04.018.

67. Forero-Quintero, L. S., Raymond, W., Handa, T.,

Saxton, M. N., Morisaki, T., Kimura, H., Bertrand, E.,

Munsky,B., and Stasevich, T.J. (2021) Live-cell imag-

ing reveals the spatiotemporal organization of endog-

enous RNA polymerase II phosphorylation at a single

gene, Nat. Commun., 12, 3158, https://doi.org/10.1038/

s41467-021-23417-0.

68. Li,J., Dong,A., Saydaminova,K., Chang,H., Wang,G.,

Ochiai, H., Yamamoto, T., and Pertsinidis, A. (2019)

Single-molecule nanoscopy elucidates RNA poly-

merase II transcription at single genes in live cells,

Cell, 178, 491-506.e28, https://doi.org/10.1016/j.cell.

2019.05.029.

69. Zhang, Q., Kim, W., Panina, S. B., Mayfield, J. E.,

Portz,B., and Zhang, Y.J. (2024) Variation of C-termi-

nal domain governs RNA polymerase II genomic loca-

tions and alternative splicing in eukaryotic transcrip-

tion, Nat. Commun., 15, 7985, https://doi.org/10.1038/

s41467-024-52391-6.

70. Giudice, J., and Jiang, H. (2024) Splicing regulation

through biomolecular condensates and membrane-

less organelles, Nat. Rev. Mol. Cell Biol., 25, 683-700,

https://doi.org/10.1038/s41580-024-00739-7.

71. Fu,Y., Yang,X., Li,S., Ma,C., An,Y., Cheng,T., Liang,Y.,

Sun,S., Cheng,T., Zhao,Y., Wang,J., Wang,X., Xu,P.,

Yin,Y., Liang,H., Liu,N., Zou,W., and Chen,B. (2025)

Dynamic properties of transcriptional condensates

modulate CRISPRa-mediated gene activation, Nat.

Commun., 16, 1640, https://doi.org/10.1038/s41467-

025-56735-8.

72. Liu, J., Chen, Y., Nong, B., Luo, X., Cui, K., Li, Z.,

Zhang, P., Tan, W., Yang, Y., Ma, W., Liang, P., and

Songyang, Z. (2023) CRISPR-assisted transcription ac-

tivation by phase-separation proteins, Protein Cell,

14, 874-887, https://doi.org/10.1093/procel/pwad013.

73. Chen, R., Shi,X., Yao,X., Gao,T., Huang, G., Ning,D.,

Cao,Z., Xu,Y., Liang,W., Tian, S.Z., Zhu,Q., Fang,L.,

Zheng, M., Hu, Y., Cui, H., and Chen, W. (2024) Spe-

cific multivalent molecules boost CRISPR-mediated

transcriptional activation, Nat. Commun., 15, 7222,

https://doi.org/10.1038/s41467-024-51694-y.

74. Nair, S. J., Yang, L., Meluzzi, D., Oh, S., Yang, F.,

Friedman, M. J., Wang, S., Suter, T., Alshareedah, I.,

Gamliel, A., Ma, Q., Zhang, J., Hu, Y., Tan, Y., Ohgi,

K.A., Jayani, R.S., Banerjee, P.R., Aggarwal, A.K., and

Rosenfeld, M. G. (2019) Phase separation of ligand-

activated enhancers licenses cooperative chromo-

somal enhancer assembly, Nat. Struct. Mol. Biol., 26,

193-203, https://doi.org/10.1038/s41594-019-0190-5.

75. Trojanowski,J., Frank,L., Rademacher,A., Mücke,N.,

Grigaitis, P., and Rippe, K. (2022) Transcription acti-

vation is enhanced by multivalent interactions in-

dependent of phase separation, Mol. Cell, 82, 1878-

1893.e10, https://doi.org/10.1016/j.molcel.2022.04.017.

76. Chong, S., Graham, T. G. W., Dugast-Darzacq, C.,

Dailey, G. M., Darzacq, X., and Tjian, R. (2022) Tun-

ing levels of low-complexity domain interactions

to modulate endogenous oncogenic transcription,

Mol. Cell, 82, 2084-2097.e5, https://doi.org/10.1016/

j.molcel.2022.04.007.

77. Jung, J.-H., Barbosa, A. D., Hutin, S., Kumita, J. R.,

Gao, M., Derwort, D., Silva, C. S., Lai, X., Pierre, E.,

Geng, F., Kim, S.-B., Baek, S., Zubieta, C., Jaeger,

K.E., and Wigge, P.A. (2020) A prion-like domain in

ELF3 functions as a thermosensor in Arabidopsis,

Nature, 585, 256-260, https://doi.org/10.1038/s41586-

020-2644-7.

78. Huang, X., Yang, Y., and Xu, C. (2025) Biomolecular

condensation programs floral transition to orchestrate

flowering time and inflorescence architecture, New

Phytol., 245, 88-94, https://doi.org/10.1111/nph.20204.

SELIVANOVSKIY et al.1598

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

79. Ren, Q., Li, L., Liu, L., Li, J., Shi, C., Sun, Y., Yao, X.,

Hou, Z., and Xiang, S. (2025) The molecular mecha-

nism of temperature-dependent phase separation of

heat shock factor 1, Nat. Chem. Biol., 21, 831-842,

https://doi.org/10.1038/s41589-024-01806-y.

80. Wu,Z., Pope, S.D., Ahmed, N.S., Leung, D.L., Hong,Y.,

Hajjar, S., Krabak, C., Zhong, Z., Raghunathan, K.,

Yue, Q., Anand, D. M., Kopp, E. B., Okin, D., Ma, W.,

Zanoni,I., Kagan, J.C., Thiagarajah, J.R., Hargreaves,

D. C., Medzhitov, R., and Zhou, X. (2025) Regulation

of inflammatory responses by pH-dependent tran-

scriptional condensates, Cell, 188, 5632-5652.e25,

https://doi.org/10.1016/j.cell.2025.06.033.

81. Henninger, J. E., Oksuz, O., Shrinivas, K., Sagi, I.,

LeRoy, G., Zheng, M. M., Andrews, J. O., Zamudio,

A. V., Lazaris, C., Hannett, N. M., Lee, T. I., Sharp,

P. A., Cissé, I. I., Chakraborty, A. K., and Young, R. A.

(2021) RNA-mediated feedback control of transcrip-

tional condensates, Cell, 184, 207-225.e24, https://

doi.org/10.1016/j.cell.2020.11.030.

82. Mazzocca, M., Fillot, T., Loffreda, A., Gnani, D., and

Mazza,D. (2021) The needle and the haystack: single

molecule tracking to probe the transcription factor

search in eukaryotes, Biochem. Soc. Trans., 49, 1121-

1132, https://doi.org/10.1042/BST20200709.

83. Jonas, F., Navon, Y., and Barkai, N. (2025) Intrin-

sically disordered regions as facilitators of the

transcription factor target search, Nat. Rev. Gen-

et., 26, 424-435, https://doi.org/10.1038/s41576-025-

00816-3.

84. Kent,S., Brown,K., Yang, C.-H., Alsaihati,N., Tian,C.,

Wang, H., and Ren, X. (2020) Phase-separated tran-

scriptional condensates accelerate target-search

process revealed by live-cell single-molecule im-

aging, Cell Rep., 33, 108248, https://doi.org/10.1016/

j.celrep.2020.108248.

85. Wang, X., Bigman, L. S., Greenblatt, H. M., Yu, B.,

Levy, Y., and Iwahara, J. (2023) Negatively charged,

intrinsically disordered regions can accelerate tar-

get search by DNA-binding proteins, Nucleic Ac-

ids Res., 51, 4701-4712, https://doi.org/10.1093/nar/

gkad045.

86. Meeussen, J. V. W., Pomp, W., Brouwer, I., de Jonge,

W. J., Patel, H. P., and Lenstra, T. L. (2023) Tran-

scription factor clusters enable target search but

do not contribute to target gene activation, Nucle-

ic Acids Res., 51, 5449-5468, https://doi.org/10.1093/

nar/gkad227.

87. Alfano,C., Fichou,Y., Huber,K., Weiss,M., Spruijt,E.,

Ebbinghaus,S., De Luca,G., Morando, M.A., Vetri,V.,

Temussi, P. A., and Pastore, A. (2024) Molecular

crowding: the history and development of a scien-

tific paradigm, Chem. Rev., 124, 3186-3219, https://

doi.org/10.1021/acs.chemrev.3c00615.

88. Rudnizky, S., Murray, P. J., Wolfe, C. H., and Ha, T.

(2024) Single-macromolecule studies of eukaryotic

genomic maintenance, Annu. Rev. Phys. Chem., 75,

209-230, https://doi.org/10.1146/annurev-physchem-

090722-010601.

89. Garcia, D. A., Johnson, T. A., Presman, D. M., Fett-

weis, G., Wagh, K., Rinaldi, L., Stavreva, D. A., Paak-

inaho,V., Jensen, R.A.M., Mandrup,S., Upadhyaya,A.,

and Hager, G. L. (2021) An intrinsically disordered

region-mediated confinement state contributes to

the dynamics and function of transcription factors,

Mol. Cell, 81, 1484-1498.e6, https://doi.org/10.1016/

j.molcel.2021.01.013.

90. Jonas, F., Carmi, M., Krupkin, B., Steinberger, J.,

Brodsky, S., Jana, T., and Barkai, N. (2023) The mo-

lecular grammar of protein disorder guiding ge-

nome-binding locations, Nucleic Acids Res., 51, 4831-

4844, https://doi.org/10.1093/nar/gkad184.

91. Ling, Y. H., Ye, Z., Liang, C., Yu, C., Park, G., Corden,

J.L., and Wu,C. (2023) Disordered C-terminal domain

drives spatiotemporal confinement of RNAPII to en-

hance search for chromatin targets, bioRxiv, https://

doi.org/10.1101/2023.07.31.551302.

92. Li, C.H., Coffey, E.L., Dall’Agnese,A., Hannett, N.M.,

Tang, X., Henninger, J. E., Platt, J. M., Oksuz, O.,

Zamudio, A. V., Afeyan, L. K., Schuijers, J., Liu, X. S.,

Markoulaki, S., Lungjangwa, T., LeRoy, G., Svoboda,

D. S., Wogram, E., Lee, T. I., Jaenisch, R., and Young,

R. A. (2020) MeCP2 links heterochromatin conden-

sates and neurodevelopmental disease, Nature, 586,

440-444, https://doi.org/10.1038/s41586-020-2574-4.

93. Tatavosian,R., Kent,S., Brown,K., Yao,T., Duc, H.N.,

Huynh, T.N., Zhen, C.Y., Ma,B., Wang,H., and Ren,X.

(2019) Nuclear condensates of the Polycomb pro-

tein chromobox 2 (CBX2) assemble through phase

separation, J. Biol. Chem., 294, 1451-1463, https://

doi.org/10.1074/jbc.RA118.006620.

94. Uckelmann, M., Levina, V., Taveneau, C., Ng, X. H.,

Pandey, V., Martinez, J., Mendiratta, S., Houx, J.,

Boudes, M., Venugopal, H., Trépout, S., Fulcher, A. J.,

Zhang,Q., Flanigan,S., Li,M., Sierecki,E., Gambin,Y.,

Das, P. P., Bell, O., de Marco, A., and Davidovich, C.

(2025) Dynamic PRC1-CBX8 stabilizes a porous struc-

ture of chromatin condensates, Nat. Struct. Mol.

Biol., 32, 520-530, https://doi.org/10.1038/s41594-

024-01457-6.

95. Collombet,S., Rall,I., Dugast-Darzacq,C., Heckert,A.,

Halavatyi,A., Le Saux,A., Dailey,G., Darzacq,X., and

Heard, E. (2023) RNA polymerase II depletion from

the inactive X chromosome territory is not mediat-

ed by physical compartmentalization, Nat. Struct.

Mol. Biol., 30, 1216-1223, https://doi.org/10.1038/

s41594-023-01008-5.

96. Murphy, S. E., and Boettiger, A. N. (2024) Polycomb

repression of Hox genes involves spatial feedback

but not domain compaction or phase transition,

Nat. Genet., 56, 493-504, https://doi.org/10.1038/

s41588-024-01661-6.

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1599

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

97. Eeftens, J. M., Kapoor, M., Michieletto, D., and

Brangwynne, C. P. (2021) Polycomb condensates can

promote epigenetic marks but are not required for

sustained chromatin compaction, Nat. Commun., 12,

5888, https://doi.org/10.1038/s41467-021-26147-5.

98. Ding, M., Wang, D., Chen, H., Kesner, B., Grimm,

N.-B., Weissbein, U., Lappala, A., Jiang, J., Rivera, C.,

Lou, J., Li, P., and Lee, J. T. (2025) A biophysical ba-

sis for the spreading behavior and limited diffusion

of Xist, Cell, 188, 978-997.e25, https://doi.org/10.1016/

j.cell.2024.12.004.

99. Zhang, Y., Li, X.-H., Peng, P., Qiu, Z.-H., Di, C.-X.,

Chen, X.-F., Wang, N.-N., Chen,F., He, Y.-W., Liu, Z.-B.,

Zhao, F., Zhu, D.-L., Dong, S.-S., Hu, S.-Y., Yang, Z.,

Li, Y.-P., Guo, Y., and Yang, T.-L. (2025) RUNX2 phase

separation mediates long-range regulation between

osteoporosis-susceptibility variant and XCR1 to pro-

mote osteoblast differentiation, Adv. Sci. Weinh. Bad-

enWurtt. Ger., 12, e2413561, https://doi.org/10.1002/

advs.202413561.

100. Vicioso-Mantis, M., Fueyo, R., Navarro, C., Cruz-

Molina, S., van Ijcken, W. F. J., Rebollo, E., Rada-

Iglesias, Á., and Martínez-Balbás, M.A. (2022) JMJD3

intrinsically disordered region links the 3D-genome

structure to TGFβ-dependent transcription activa-

tion, Nat. Commun., 13, 3263, https://doi.org/10.1038/

s41467-022-30614-y.

101. Shrinivas, K., Sabari, B. R., Coffey, E. L., Klein, I. A.,

Boija,A., Zamudio, A.V., Schuijers,J., Hannett, N.M.,

Sharp, P. A., Young, R. A., and Chakraborty, A. K.

(2019) Enhancer features that drive formation of

transcriptional condensates, Mol. Cell, 75, 549-561.e7,

https://doi.org/10.1016/j.molcel.2019.07.009.

102. Xing, Y.-H., Dong,R., Lee,L., Rengarajan,S., Riggi,N.,

Boulay,G., and Rivera, M. N. (2024) DisP-seq reveals

the genome-wide functional organization of DNA-

associated disordered proteins, Nat. Biotechnol., 42,

52-64, https://doi.org/10.1038/s41587-023-01737-4.

103. Yu, Z., Wang, Q., Zhang, Q., Tian, Y., Yan, G., Zhu, J.,

Zhu, G., and Zhang, Y. (2024) Decoding the genom-

ic landscape of chromatin-associated biomolecu-

lar condensates, Nat. Commun., 15, 6952, https://

doi.org/10.1038/s41467-024-51426-2.

104. Wang, S., Wang, Z., and Zang, C. (2025) Genomic

clustering tendency of transcription factors reflects

phase-separated transcriptional condensates at su-

per-enhancers, Nucleic Acids Res., 53, gkaf015, https://

doi.org/10.1093/nar/gkaf015.

105. Hnisz, D., Shrinivas, K., Young, R. A., Chakraborty,

A.K., and Sharp, P.A. (2017) A phase separation mod-

el for transcriptional control, Cell, 169, 13-23, https://

doi.org/10.1016/j.cell.2017.02.007.

106. Rao, S.S.P., Huang, S.-C., Glenn St Hilaire,B., Engreitz,

J. M., Perez, E. M., Kieffer-Kwon, K.-R., Sanborn,

A. L., Johnstone, S. E., Bascom, G. D., Bochkov, I. D.,

Huang, X., Shamim, M. S., Shin, J., Turner, D., Ye, Z.,

Omer, A. D., Robinson, J. T., Schlick, T., Bernstein,

B.E., Casellas,R., Lander, E.S., and Aiden, E.L. (2017)

Cohesin loss eliminates all loop domains, Cell, 171,

305-320.e24, https://doi.org/10.1016/j.cell.2017.09.026.

107. Boija, A., Klein, I. A., Sabari, B. R., Dall’Agnese, A.,

Coffey, E. L., Zamudio, A. V., Li, C. H., Shrinivas, K.,

Manteiga, J.C., Hannett, N.M., Abraham, B.J., Afeyan,

L.K., Guo, Y.E., Rimel, J.K., Fant, C.B., Schuijers,J., Lee,

T.I., Taatjes, D.J., and Young, R.A. (2018) Transcription

factors activate genes through the phase-separation

capacity of their activation domains, Cell, 175, 1842-

1855.e16, https://doi.org/10.1016/j.cell.2018.10.042.

108. Li,J., Liu,S., Kim,S., Goell,J., Drum, Z.A., Flores, J.P.,

Ma, A. J., Mahata, B., Escobar, M., Raterink, A., Ahn,

J. H., Terán, E. R., Guerra-Resendez, R. S., Zhou, Y.,

Yu, B., Diehl, M. R., Wang, G. G., Gustavsson, A.-K.,

Phanstiel, D. H., and Hilton, I. B. (2025) Biomolec-

ular condensation of human IDRs initiates endog-

enous transcription via intrachromosomal loop-

ing or high-density promoter localization, Nucleic

Acids Res., 53, gkaf056, https://doi.org/10.1093/nar/

gkaf056.

109. Selivanovskiy, A. V., Molodova, M. N., Khrameeva,

E. E., Ulianov, S. V., and Razin, S. V. (2025) Liquid

condensates: a new barrier to loop extrusion? Cell.

Mol. Life Sci., 82, 80, https://doi.org/10.1007/s00018-

024-05559-8.

110. Lee, R., Kang, M.-K., Kim, Y.-J., Yang, B., Shim, H.,

Kim, S., Kim, K., Yang, C. M., Min, B., Jung, W.-J.,

Lee, E.-C., Joo, J.-S., Park, G., Cho, W.-K., and Kim,

H.-P. (2022) CTCF-mediated chromatin looping pro-

vides a topological framework for the formation of

phase-separated transcriptional condensates, Nu-

cleic Acids Res., 50, 207-226, https://doi.org/10.1093/

nar/gkab1242.

111. Hansen, A. S. (2020) CTCF as a boundary factor for

cohesin-mediated loop extrusion: evidence for a

multi-step mechanism, Nucleus, 11, 132-148, https://

doi.org/10.1080/19491034.2020.1782024.

112. Wei, C., Jia, L., Huang, X., Tan, J., Wang, M., Niu, J.,

Hou, Y., Sun, J., Zeng, P., Wang, J., Qing, L., Ma, L.,

Liu,X., Tang,X., Li, F., Jiang,S., Liu,J., Li,T., Fan,L.,

Sun, Y., Gao, J., Li, C., and Ding, J. (2022) CTCF or-

ganizes inter-A compartment interactions through

RYBP-dependent phase separation, Cell Res., 32,

744-760, https://doi.org/10.1038/s41422-022-00676-0.

113. Gao, C., Gao, A., Jiang, Y., Gao, R., Guo, Y., Peng, Z.,

Jiang, W., Zhang, M., Zhou, Z., Yan, C., Fang, W.,

Hu,H., Zhu,G., and Zhang,J. (2025) Hypoxia-induced

phase separation of ZHX2 alters chromatin looping to

drive cancer metastasis, Mol. Cell, 85, 1525-1542.e10,

https://doi.org/10.1016/j.molcel.2025.03.009.

114. Heist, T., Fukaya, T., and Levine, M. (2019) Large

distances separate coregulated genes in living Dro-

sophila embryos, Proc. Natl. Acad. Sci. USA, 116,

15062-15067, https://doi.org/10.1073/pnas.1908962116.

SELIVANOVSKIY et al.1600

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

115. Fukaya, T., Lim, B., and Levine, M. (2016) En-

hancer control of transcriptional bursting, Cell, 166,

358-368, https://doi.org/10.1016/j.cell.2016.05.025.

116. Bohrer, C. H., and Larson, D. R. (2023) Synthetic

analysis of chromatin tracing and live-cell imaging

indicates pervasive spatial coupling between genes,

eLife, 12, e81861, https://doi.org/10.7554/eLife.81861.

117. Mahat, D. B., Tippens, N. D., Martin-Rufino, J. D.,

Waterton, S. K., Fu, J., Blatt, S. E., and Sharp, P. A.

(2024) Single-cell nascent RNA sequencing unveils co-

ordinated global transcription, Nature, 631, 216-223,

https://doi.org/10.1038/s41586-024-07517-7.

118. Chowdhary,S., Kainth, A.S., Paracha,S., Gross, D.S.,

and Pincus, D. (2022) Inducible transcriptional con-

densates drive 3D genome reorganization in the heat

shock response, Mol. Cell, 82, 4386-4399.e7, https://

doi.org/10.1016/j.molcel.2022.10.013.

119. Oka, M., Otani, M., Miyamoto, Y., Oshima, R.,

Adachi, J., Tomonaga, T., Asally, M., Nagaoka, Y.,

Tanaka, K., Toyoda, A., Ichikawa, K., Morishita, S.,

Isono, K., Koseki, H., Nakato, R., Ohkawa, Y., and

Yoneda, Y. (2023) Phase-separated nuclear bodies of

nucleoporin fusions promote condensation of MLL1/

CRM1 and rearrangement of 3D genome structure,

Cell Rep., 42, 112884, https://doi.org/10.1016/j.celrep.

2023.112884.

120. Christou-Kent, M., Cuartero, S., Garcia-Cabau, C.,

Ruehle,J., Naderi,J., Erber,J., Neguembor, M.V., Plana-

Carmona, M., Alcoverro-Bertran, M., Andres-Aguayo,

L. D., Klonizakis, A., Julià-Vilella, E., Lynch, C.,

Serrano, M., Hnisz, D., Salvatella, X., Graf, T., and

Stik,G. (2023) CEBPA phase separation links transcrip-

tional activity and 3D chromatin hubs, Cell Rep., 42,

112897, https://doi.org/10.1016/j.celrep.2023.112897.

121. Ohishi,H., Shinkai,S., Owada,H., Fujii,T., Hosoda,K.,

Onami, S., Yamamoto, T., Ohkawa, Y., and Ochiai, H.

(2024) Transcription-coupled changes in genom-

ic region proximities during transcriptional burst-

ing, Sci. Adv., 10, eadn0020, https://doi.org/10.1126/

sciadv.adn0020.

122. Panigrahi, A. K., Lonard, D. M., and O’Malley,

B. W. (2023) Enhancer-promoter entanglement ex-

plains their transcriptional interdependence, Proc.

Natl. Acad. Sci. USA, 120, e2216436120, https://

doi.org/10.1073/pnas.2216436120.

123. Altendorfer, E., Mundlos, S., and Mayer, A. (2025) A

transcription coupling model for how enhancers

communicate with their target genes, Nat. Struct.

Mol. Biol., 32, 598-606, https://doi.org/10.1038/s41594-

025-01523-7.

124. Banani, S.F., Afeyan, L.K., Hawken, S.W., Henninger,

J.E., Dall’Agnese,A., Clark, V.E., Platt, J.M., Oksuz,O.,

Hannett, N. M., Sagi, I., Lee, T. I., and Young, R. A.

(2022) Genetic variation associated with condensate

dysregulation in disease, Dev. Cell, 57, 1776-1788.e8,

https://doi.org/10.1016/j.devcel.2022.06.010.

125. Asimi,V., Sampath Kumar, A., Niskanen, H., Riemen-

schneider, C., Hetzel, S., Naderi, J., Fasching, N., Pop-

itsch,N., Du,M., Kretzmer,H., Smith, Z.D., Weigert,R.,

Walther, M., Mamde, S., Meierhofer, D., Wittler, L.,

Buschow,R., Timmermann,B., Cisse, I.I., Ameres, S.L.,

Meissner, A., and Hnisz, D. (2022) Hijacking of tran-

scriptional condensates by endogenous retroviruses,

Nat. Genet., 54, 1238-1247, https://doi.org/10.1038/

s41588-022-01132-w.

126. Qi,H., Yin,M., Ren,X., Chen,G., Li,A., Li,Y., Cao,X.,

and Zhou, J. (2025) HSV-1 ICP22 condensates impair

host transcription by depleting promoter RNAPII

Ser-2P occupation, Front. Microbiol., 16, 1538737,

https://doi.org/10.3389/fmicb.2025.1538737.

127. Mensah, M. A., Niskanen, H., Magalhaes, A. P.,

Basu, S., Kircher, M., Sczakiel, H. L., Reiter, A. M. V.,

Elsner, J., Meinecke, P., Biskup, S., Chung, B. H. Y.,

Dombrowsky, G., Eckmann-Scholz, C., Hitz, M. P.,

Hoischen, A., Holterhus, P.-M., Hülsemann, W.,

Kahrizi,K., Kalscheuer, V.M., Kan,A., Krumbiegel,M.,

Kurth, I., Leubner, J., Longardt, A. C., Moritz, J. D.,

Najmabadi, H., Skipalova, K., Snijders Blok, L.,

Tzschach, A., Wiedersberg, E., Zenker, M., Garcia-

Cabau,C., Buschow,R., Salvatella,X., Kraushar, M.L.,

Mundlos, S., Caliebe, A., Spielmann, M., Horn, D.,

and Hnisz, D. (2023) Aberrant phase separation and

nucleolar dysfunction in rare genetic diseases, Na-

ture, 614, 564-571, https://doi.org/10.1038/s41586-022-

05682-1.

128. Ahn, J.H., Guo, Y., Lyons, H., Mackintosh, S. G., Lau,

B. K., Edmondson, R. D., Byrum, S. D., Storey, A. J.,

Tackett, A.J., Cai,L., Sabari, B.R., and Wang, G.G. (2025)

The phenylalanine-and-glycine repeats of NUP98 on-

cofusions form condensates that selectively partition

transcriptional coactivators, Mol. Cell, 85, 708-725.e9,

https://doi.org/10.1016/j.molcel.2024.12.026.

129. Lyons, H., Pradhan, P., Prakasam, G., Vashishtha, S.,

Li, X., Eppert, M., Fornero, C., Tcheuyap, V. T.,

McGlynn,K., Yu,Z., Raju, D.R., Koduru, P.R., Xing,C.,

Kapur, P., Brugarolas, J., and Sabari, B. R. (2025)

RNA polymerase II partitioning is a shared feature

of diverse oncofusion condensates, Cell, 188, 3843-

3862.e28, https://doi.org/10.1016/j.cell.2025.04.002.

130. Showpnil, I.A., Selich-Anderson,J., Taslim,C., Boone,

M. A., Crow, J. C., Theisen, E. R., and Lessnick, S. L.

(2022) EWS/FLI mediated reprogramming of 3D chro-

matin promotes an altered transcriptional state in

Ewing sarcoma, Nucleic Acids Res., 50, 9814-9837,

https://doi.org/10.1093/nar/gkac747.

131. Wang,Y., Yu,C., Pei,G., Jia,W., Li,T., and Li,P. (2023)

Dissolution of oncofusion transcription factor conden-

sates for cancer therapy, Nat. Chem. Biol., 19, 1223-

1234, https://doi.org/10.1038/s41589-023-01376-5.

132. Zhang, Y., Stöppelkamp, I., Fernandez-Pernas, P.,

Allram,M., Charman,M., Magalhaes, A.P., Piedavent-

Salomon, M., Sommer, G., Sung, Y.-C., Meyer, K.,

STRUCTURE AND FUNCTIONS OF TRANSCRIPTION CONDENSATES 1601

BIOCHEMISTRY (Moscow) Vol. 90 No. 11 2025

Grams, N., Halko, E., Dongre, S., Meierhofer, D.,

Malszycki, M., Ilik, I. A., Aktas, T., Kraushar, M. L.,

Vastenhouw, N., Weitzman, M. D., Grebien, F.,

Niskanen, H., and Hnisz, D. (2025) Probing conden-

sate microenvironments with a micropeptide kills-

witch, Nature, 643, 1107-1116, https://doi.org/10.1038/

s41586-025-09141-5.

133. Naderi,J., Magalhaes, A.P., Kibar,G., Stik,G., Zhang,Y.,

Mackowiak, S.D., Wieler, H.M., Rossi,F., Buschow,R.,

Christou-Kent, M., Alcoverro-Bertran, M., Graf, T.,

Vingron,M., and Hnisz,D. (2024) An activity-specific-

ity trade-off encoded in human transcription factors,

Nat. Cell Biol., 26, 1309-1321, https://doi.org/10.1038/

s41556-024-01411-0.

134. Semeigazin, A., Minami, K., Shimazoe, M. A.,

Khochbin, S., Panne, D., Ide, S., and Maeshima, K.

(2025) Liquid-like transcription condensates locally

constrain chromatin in living human cells, bioRxiv,

https://doi.org/10.1101/2025.07.05.663270.

Publisher’s Note. Pleiades Publishing remains

neutral with regard to jurisdictional claims in published

maps and institutional affiliations. AI tools may have

been used in the translation or editing of this article.