Article

ISSN 0006-2979, Biochemistry (Moscow), 2024, Vol. 89, No. 4, pp. 626-636 © Pleiades Publishing, Ltd., 2024.

626

Role of Mod(mdg4)-67.2 Protein in Interactions

between Su(Hw)-Dependent Complexes

and Their Recruitment to Chromatin

Larisa S. Melnikova

1,a

*, Varvara V. Molodina

Pavel G. Georgiev

, and Anton K. Golovnin

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

e-mail: lsm73@mail.ru

Received September 25, 2023

Revised October 23, 2023

Accepted October 24, 2023

Abstract— Su(Hw) belongs to the class of proteins that organize chromosome architecture, determine promoter

activity, and participate in formation of the boundaries/insulators between the regulatory domains. This protein

contains a cluster of 12 zinc fingers of the C2H2 type, some of which are responsible for binding to the consensus

site. The Su(Hw) protein forms complex with the Mod(mdg4)-67.2 and the CP190 proteins, where the last one binds

to all known Drosophila insulators. To further study functioning of the Su(Hw)-dependent complexes, we used the

previously described su(Hw)

mutation with inactive seventh zinc finger, which produces mutant protein that

cannot bind to the consensus site. The present work shows that the Su(Hw)

protein continues to directly interact

with the CP190 and Mod(mdg4)-67.2 proteins. Through interaction with Mod(mdg4)-67.2, the Su(Hw)

protein can

be recruited into the Su(Hw)-dependent complexes formed on chromatin and enhance their insulator activity. Our

results demonstrate that the Su(Hw) dependent complexes without bound DNA can be recruited to the Su(Hw)

binding sites through the specific protein–protein interactions that are stabilized by Mod(mdg4)-67.2.

DOI: 10.1134/S0006297924040035

Keywords: transcription regulation, Su(Hw), Mod(mdg4), protein–protein interactions, chromatin insulators

Abbreviations: BTB, bric-a-brac, tramtrack, and broad complex; CP190, centrosomal protein 190 kD; Mod(mdg4), modifi-

er ofmdg4; SBS,Su(Hw) binding sites; Su(Hw),suppressor of Hairy-wing; TTK, Tramtrack group; Y2H, yeast two hybrid;

ZF,zinc fingers.

* To whom correspondence should be addressed.

INTRODUCTION

In higher eukaryotes, regulation of gene expres-

sion becomes more complex due to the cell differen-

tiation during embryonic development. Cell special-

ization is determined by the expression of various

combinations of transcription factors, which are en-

coded by a large group of developmental genes that

control differentiation [1]. Another large group of genes

encodes housekeeping proteins that are essential for

functioning of all cells. In Drosophila, housekeeping

genes are grouped in clusters, and all their regulatory

elements are usually found in close proximity to the

promoters they regulate. Unlike housekeeping genes,

developmental genes usually have complex, extensive

regulatory systems consisting of a large number of en-

hancers, each of which determines gene expression in

a specific group of cells and over a certain time period

[2-4]. Enhancers can stimulate promoters while being

located at distances from them, in some cases more

than hundreds of kb. Regulation of interactions be-

tween the enhancers and promoters is controlled by a

special group of regulatory elements called insulators

[5-7]. Certain combinations of insulators can interact

with each other to form chromatin domains that en-

hance/block long-distance interactions between the en-

hancers and promoters [5, 8, 9].

In Drosophila, architectural proteins bind to the

gene promoters and insulators. Characteristic feature

of these proteins are clusters consisting of five or more

Mod(mdg4)-67.2 PROTEIN 627

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

zinc fingers(ZF) of the C2H2 type [10,11]. It has been

shown that the architectural proteins specifically bind

to the extended motifs of 12-15bp by its 4-5 C2H2 do-

mains. In Drosophila, the best described architectural

protein is Su(Hw) (suppressor of Hairy-wing), which

has about 2000 binding sites in the genome [12-15].

Su(Hw) was originally discovered as an insulator pro-

tein that blocks interactions between the enhancers

and gene promoters by binding to 12 sites within the

gypsy retrotransposon [12-14]. Subsequently, it was

shown that Su(Hw) is involved in formation of pro-

moters and is capable of repressing transcription from

some of them [15,16]. In the center of the Su(Hw) pro-

tein there is a cluster consisting of 12 C2H2 ZFs, which

bind to the consensus sequence [17]. At the N-terminus

of the Su(Hw) protein there are two conserved regions

that interact with the BTB (bric-a-brac, tramtrack, and

broad complex) domain of the CP190 protein (centro-

somal protein 190 kD), which is involved in formation

of active insulators and promoters of housekeeping

genes [18, 19]. At the C-terminus of the protein, a do-

main is mapped that interacts with one of 30 isoforms

of the Mod(mdg4) protein (modifier of mdg4) [20-22].

All Mod(mdg4) isoforms at the N-terminus have a

TTK (Tramtrack group)-like BTB domain, which is re-

sponsible for formation of hexamers [23, 24]. The

Mod(mdg4)-67.2 isoform has a unique C-terminus that

interacts exclusively with the C-terminal domain of

the Su(Hw) protein. In addition, it was found that the

Su(Hw)-dependent complexes could also include other

proteins: HIPP1 (HP1 and insulator partner protein1),

which interacts with the same C-terminal region of

Su(Hw) as Mod(mdg4)-67.2 [25-27], ENY2 (enhancer of

yellow 2), which interacts with the zinc fingers 11 and

12 of the Su(Hw) protein [28], and RNA-binding pro-

teins [29, 30]. All isoforms of the Mod(mdg4) protein,

like the CP190 protein, are SUMOlated and, as a result

of multiple protein–protein interactions, form speckles

in the nucleus [31-33]. It is assumed that multimeriza-

tion of the BTB domains and interaction between the

SUMO (small ubiquitin-like modifier) proteins form

the speckle core, to which “passenger” proteins such as

Su(Hw) and other architectural proteins are attached.

According to the proposed model, speckles function as

reservoirs of architectural proteins that bind to new

DNA during its replication [32,33].

In the present work, we investigated the ability

of the mutant protein Su(Hw)

, which does not by

itself bind DNA, to be recruited to the Su(Hw) depen-

dent chromatin sites. A point replacement of histidine

at position 459 with tyrosine in the Su(Hw)

mutant

[34] leads to the destruction of the seventh C2H2 do-

main, which is necessary for binding of the protein to

chromatin [17]. As a result, Su(Hw)

cannot bind to

Su(Hw)-binding sites invitro and is not detected on the

polytene chromosomes [17]. However, we have demon-

strated that in the presence of the chromatin-binding

mutant of Su(Hw) with N-terminus deletion (Su(Hw)

ΔN), the Su(Hw)

the protein is recruited to the

Su(Hw)-dependent insulator sites. Efficient binding of

Su(Hw)

is mediated by the Mod(mdg4)-67.2 protein.

MATERIALS AND METHODS

Generation of recombinant genetic constructs.

All constructs for Y2H (yeast two hybrid) assay were

created on the basis of the pGBT9 vector containing

the DNA-binding domain of the yeast GAL4 protein

(Clontech, USA).

To create the pGBTSu(Hw)

construct, the PCR

product 5′-ggaacagcacaagtcacgtg 3′/5′ caccaatgcagaaaa

cttcttgtc-3′ was treated with BglII endonuclease, and

the PCR product 5′-gcccttaaaaagTatcgacgct-3′/5′-aatccgt

gcgttccataat-3′ with endonuclease EagI. Su(Hw) cDNA

was used as a template for PCR. The resulting DNA frag-

ments were co-cloned into the plasmid pGBTSu(Hw)

digested with BglII and EagI.

To create the pGBTSu(Hw)

Δ114 construct, the

XhoI-AflII fragment containing deletion of 114 a.a.

from pGBTSu(Hw)Δ114, was cloned into the plasmid

pGBTSu(Hw)

digested with XhoI and AflII.

To create the pGBTSu(Hw)

Δ283 construct, the

EagI-SalI fragment containing a 17-aa deletion from

pGBTSu(Hw)Δ283 was cloned into the pGBTSu(Hw)E8

plasmid treated with EagI and SalI.

Plasmids pGBTSu(Hw), pGBTSu(Hw)Δ283,

pGADMod(mdg4)-67.2, and pGADCP190 were obtained

and described previously [18,20].

Yeast two-hybrid system. Analysis of protein in-

teractions in Y2H was performed using plasmids and

protocols from Clontech. Plasmids were transformed

into a yeast strain pJ69-4A by lithium acetate meth-

od as described by the manufacturer and plated on a

medium without tryptophan and leucine. After 3 days

of growth at 30°C, the cells were subcultured onto a

selective medium without tryptophan, leucine, histi-

dine, and adenine, and growth of yeast colonies was

compared after 2-3 days. As a negative control, inter-

action of Su(Hw) protein derivatives expressed in the

pGBT9 vector with the pGAD24 vector was tested. In-

teractions of the full-length Su(Hw) protein with the

Mod(mdg4)-67.2 or CP190 proteins, described previ-

ously, served as a positive control [18,20]. Each exper-

iment was repeated three times.

Analysis of Drosophila transgenic lines pheno-

type. All flies were kept at 25°C on a standard yeast me-

dium (Bloomington Drosophila Stock Center). Effects

of different combinations of mutations were assessed

independently by two investigators. Level of expres-

sion of the yellow and cut phenotypes was assessed in

males aged 3-5 days, developing at 25°C. Changes in

MELNIKOVA et al.628

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

yellow gene expression (in the body and wings) were

assessed on a five-point scale, where 5 corresponds to

wild-type pigmentation; 2 corresponds to the level of

pigmentation associated with the y

mutation; 3 and

4– partial activation of basal transcription; 1– no ex-

pression. Flies in which expression of the yellow allele

was characterized previously were used as a standard.

Changes in the cut gene expression were assessed by

counting the number of gaps occurring along the edge

of the wing plate. Wild-type flies and flies carrying

the ct

allele were used as a reference. Representative

wing shapes shown in Fig. 2b were selected as “aver-

age” from the series of wings arranged in increasing

order of severity of their mutant phenotype. In each

transgenic line, the phenotype of at least 50 flies was

assessed.

Transgenic lines Su(Hw)+, Su(Hw)Δ62, Su(Hw)Δ52,

Su(Hw)Δ114, and Su(Hw)J, targeted with the 3xFLAG

epitope, were obtained and described previously

[18, 35]. Combination of the mod(mdg4)

mutation

or combinations of the su(Hw)

/su(Hw)

and su(Hw)

su(Hw)

mutations with transgenic lines was carried

out in accordance with the scheme described previ-

ously [36]. All details of genetic crosses can be provid-

ed upon request.

Immunostaining of polytene chromosomes.

Drosophila 3rd instar larvae were grown at 18°C un-

der standard conditions. Staining of polytene chro-

mosomes was performed according to the previously

described method [37]. The following primary antibod-

ies were used: anti-Su(Hw) (rabbit) 1 : 300 and anti-

FLAG 1 : 50 (mouse). The secondary antibodies used

were FITC-AffiniPure Donkey Anti-Rabbit 1 : 200 and

Cy5- AffiniPure Donkey Anti-Mouse 1 : 200 (Jackson

Immuno Research, USA). Analysis was performed us-

ing a Zeiss fluorescence microscope (Axio Observer.Z1,

Germany) equipped with an OptiGrid structured illu-

mination microscopy system (Qioptiq, Luxembourg).

TheFiji program was used for image processing.

Chromatin immunoprecipitation. Isolation of

chromatin from the pupal stage of Drosophila develop-

ment and subsequent immunoprecipitation procedure

were carried out in accordance with the method de-

scribed previously [35]. The following antibodies were

used for immunoprecipitation: anti-Su(Hw) (rabbit)

1 : 200 and anti-FLAG 1 : 300 (mouse). Amount of im-

munoprecipitated DNA was determined by quanti-

tative PCR using SYBR green (Bio-Rad, USA, Cat# 170-

8882). Sequences of primers used in PCR:

62D fw – 5′ TTTGGGCTTGGTGAGAACAG 3′

62D rev – 5′ TGATACCAGGCGAACAGAAATC 3′

50A fw – 5′ ATACAAAGTGGTTTCAGCCAAGAAG 3′

50A rev – 5′ TTGATAAATAGTCCAGCACGCATAC 3′

87E fw – 5′ GGATGTTACA TTGAGAGTGCTTAGG 3′

87E rev – 5′ TTTGCGTTTCGGCTGCTGTC 3′

1A2 fw – 5′ ACCACACATCAGTCATCGTGT 3′

1A2 rev – 5′ CTTCGTCTACCGTTGTGC 3′

gypsy fw – 5′ TTCTCTAAAAAGTATGCAGCACTT 3′

gypsy rev – 5′ CACGTAATAAGTGTGCGTTGA 3′

ras fw – 5′ GAGGGATTCCTGCTCGTCTTCG 3′

ras rev – 5′ GTCGCACTTGTTACCCACCATC 3′

Each experiment was performed in triplicate bio-

logical replicates.

Antibodies. We used polyclonal antibodies

against the N-terminal domain of the Su(Hw) protein

described previously [32,33], and monoclonal antibod-

ies against the FLAG tag (Sigma, USA, Cat# F 1804).

Statistical analysis was performed using Stu-

dent’s t-test.

RESULTS

The Su(Hw)

protein is able to directly in-

teract with the Mod(mdg4)-67.2 and CP190 pro-

teins. We have previously shown that the mutant

Su(Hw)

protein, in which the tenth ZF is inactivat-

ed, loses its ability to interact with the CP190 protein

in vitro [35]. To test how mutation in the seventh ZF

of the Su(Hw)

protein affects interaction with two

other components of the Su(Hw)-dependent complex,

CP190 and Mod(mdg4)-67.2 proteins, we used a yeast

two-hybrid system. Based on the pGBT9 vector, three

constructs were created. The first construct expressed

in yeast the full-length protein Su(Hw)

, the second

expressed its derivative Su(Hw)

Δ114 with deletion of

the region 88-202 aa responsible for interaction with

the CP190 protein, and the third expressed the deriv-

ative Su(Hw)

Δ283, in which the region of interaction

with the Mod(mdg4)-67.2 protein (760 to 778 aa) was

deleted (Fig.1a). We then tested direct interactions of

the Su(Hw)

variants with the full length CP190 and

Mod(mdg4)-67.2, which were expressed in the pGAD24

vector. The Su(Hw)

protein interacted with all insula-

tor proteins in the same way as the wild type protein.

The Su(Hw)

Δ114 protein lost its ability to interact

with CP190, and the Su(Hw)

Δ283 protein lost its abil-

ity to interact with Mod(mdg4)-67.2 (Fig.1a). Thus, the

mutation in the seventh ZF does not affect interaction

of the Su(Hw)

protein with other components of the

insulator complex. Therefore, the Su(Hw)

protein can

be recruited to chromatin through interaction with the

CP190 and Mod(mdg4)-67.2 proteins.

Su(Hw)

protein restores insulator function of

the mutant Su(Hw) proteins, interaction of which

with CP190 is impaired. To study the Su(Hw)-depen-

dent insulation, two model systems are usually used,

which are generated by integrating the transpos-

able element gypsy into the yellow(y

) and cut(ct

)

loci. The yellow gene is responsible for pigmentation

of the Drosophila cuticle [38]. In the wild type, the

body, wings, and bristles of flies are dark colored.

Mod(mdg4)-67.2 PROTEIN 629

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Fig. 1. Su(Hw) protein derivatives. a)Full length Su(Hw) protein is shown schematically. Domain designations: CID,domain in-

teracting with CP190; ZF,zinc fingers; LZ,leucine zipper. The diagram shows the regions of interaction with the Mod(mdg4)-67.2

(Mod-67.2) and CP190 proteins, which are depicted as ovals. Vertical arrow indicates the su(Hw)

mutation. The bracket below

the diagram indicates the region used for generation of antibodies. The numbers indicate amino acid residues that limit domains

and derived forms. The names of the derivatives are indicated on the left, size of the derivatives is indicated by the segments,

the dotted lines indicate internal deletions, and the asterisk indicates the su(Hw)

mutation. To the right of the diagrams are the

results obtained in the Y2H. “+”,presence of interaction and “–”,absence of interaction. b)Deletion derivatives used ingenetic

experiments and immunostaining of polytene chromosomes.

In the y

allele, the gypsy retrotransposon was inserted

between the body and wing enhancers and the yellow

gene promoter (Fig.2a). In this case, the Su(Hw) insu-

lator completely blocks activation of the yellow expres-

sion in the body and wings, that results in yellow pig-

mentation of the body and wing blades of the mutant

flies (Fig.2b). However, bristles of the flies remain pig-

mented, because the bristle enhancer is located in the

intron of the gene [39].

In the ct

allele (Fig.2a), gypsy is located between

the wing margin enhancer and the cut promoter, sep-

arated from each other by a distance of more than

70kb. Wing margin enhancer is responsible for the de-

velopment of the wing edge. In this case, the insulator

completely blocks the wing margin enhancer, resulting

in almost completely cut off the wing edge and the

wing bristles are absent (Fig.2b) [22,40].

The CP190 protein binds to two regions, 88-150aa

and 150-202aa, at the N-terminus of the Su(Hw) pro-

tein (Fig.1a) [18]. Using genetic crosses different trans-

genes were integrated into the second chromosome of

the y

; su(Hw)

/su(Hw)

line: expressing either the

full length Su(Hw)+ protein or it derivatives – Su(Hw)

Δ62 (deletion of the region 88-150aa), Su(Hw)Δ52 ( de-

letion of region 150-202aa), and Su(Hw)Δ114 (deletion

of both regions interacting with CP190 (Fig. 1b). Com-

bination of the mutations su(Hw)

/su(Hw)

inactivates

the native Su(Hw) protein, which allows us to analyze

the effect of mutant proteins on the y

phenotype

(Fig.2b) [18].

Inactivation of the Su(Hw) protein in the y

;

su(Hw)

/su(Hw)

line restored yellow expression in the

allele and cut expression in the ct

allele, demonstrat-

ing that the Su(Hw) protein binding is critical for insu-

lation. Introduction of the Su(Hw)+ transgene leads to

complete restoration of insulation (Fig. 2b). We previ-

ously showed that the CP190 protein is also required

for the Su(Hw) dependent insulation [18]. In the lines

with expression of the Su(Hw)Δ62 and Su(Hw)Δ52 pro-

teins exhibiting weakened binding of the CP190 pro-

tein to the Su(Hw) dependent complex, the insulator

completely blocked the body and wing enhancers of

the yellow gene (y

phenotype). However, the CP190 de-

ficiency produced much stronger effect on the cut gene

phenotype. In the case of Su(Hw)Δ52 protein expres-

sion, insulation in the ct

allele weakened: numerous,

but separate, gaps were present along the edge of the

wing and some of the bristles developed. The Su(Hw)

Δ62 protein demonstrated only weak insulator activity:

1-2 gaps appeared along the edge of the wing. In the

Su(Hw)Δ114 line, the CP190 protein did not bind to the

insulator complex. As a result, there was no insulator

MELNIKOVA et al.630

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Fig. 2. Effect of Su(Hw)

protein on gypsy-dependent insu-

lation. a) Schematic representation of the y

and ct

alleles.

Exons of the yellow and cut genes are shown as rectangles.

Gene transcription initiation sites are indicated by arrows.

The gypsy retrotransposon is depicted as a triangle. Rectangles

at its ends represent LTRs (long terminal repeats), orientation

of which is indicated by arrows. Designations: Su(Hw), in-

sulator Su(Hw); En-W, wing enhancer, En-B, body enhancer;

En-Br, enhancer of bristles; En-Wm is a wing margin en-

hancer. b)Effect of Su(Hw) derivatives on activity of the gyp-

sy insulator in the y

and ct

alleles analyzed in the su(Hw)

su(Hw)

(v/2) and su(Hw)

/su(Hw)

(v/E8) mutant background.

The lines used for phenotypic analysis are indicated in the

right column: wt, y

, Su(Hw)+, the transgene expressed as

the full length protein. Schemes and names of other deriv-

atives are shown in Fig. 1. The numbers in column y

show

the level of expression of the yellow gene in the cuticle of the

body and wings. The photographs show changes in the wing

phenotype of the cut gene in different mutant backgrounds.

activity: the flies had the wild-type wings, and color of

the cuticle was also close to normal (Fig.2b).

Next we studied insulation in the y

;su(Hw)

su(Hw)

line (Fig. 2b). In this case, expression of the

Su(Hw)

protein did not restore insulation. Unex-

pectedly, insulation has been partially restored when

the transgenes expressing Su(Hw) derivatives were

introduced to the y

;su(Hw)

/su(Hw)

line. In the

Su(Hw)Δ62 line, the number of wing gaps increased

significantly, and in the Su(Hw)Δ52 line, flies had

a wing phenotype close to ct

. Even in the Su(Hw)

Δ114 line, the ct

allele showed weak insulation: in-

dividual wing gaps appeared. At the same time, insu-

lation in the y

allele was restored almost complete-

ly (Fig. 2b). Analysis of the results suggests that the

derivatives of Su(Hw) recruit the Su(Hw)

protein

capable of binding with the CP190 protein, to the

sites of the gypsy insulator, that leads to restoration

of insulation.

Since the Su(Hw)Δ114 protein does not inter-

act with CP190, recruitment of the Su(Hw)

to the

Su(Hw)- dependent chromatin sites could be carried out

through the Mod(mdg4)-67.2 protein, which binds to the

C-terminal region of Su(Hw) (716-892 aa) responsible

for insulation and transcriptional repression [20, 21].

To test the role of Mod(mdg4)-67.2 in the recruitment

of the Su(Hw)

protein, we used the Su(Hw)J transgene

expressing a mutant protein with deletion of 144 aa

at the C-terminal [16]. In the mutant protein Su(Hw)J

(aa 1-801), the region of interaction with Mod(mdg4)- 67.2

is deleted, so Su(Hw)J binds only to CP190 (Fig. 1b).

In the y

;su(Hw)

/su(Hw)

mutant background, phe-

notypes of the flies expressing the Su(Hw)J and Su(Hw)

Δ114 proteins are similar: expression of yellow and cut

is restored (Fig. 2b). However, in the line expressing

Su(Hw)J in the su(Hw)

/su(Hw)

background, resto-

ration of insulation did not occur. The obtained data are

consistent with the putative role of the Mod(mdg4)-67.2

protein in recruitment of the Su(Hw)

protein to the

Su(Hw)-dependent chromatin sites.

Mod(mdg4)-67.2 mediates recruitment of the

Su(Hw)

to the binding sites of the Su(Hw)ΔN pro-

tein with polytene chromosomes. To confirm recruit-

ment of the Su(Hw)

protein to the chromatin sites

through the Mod(mdg4)-67.2 protein, we used trans-

genic lines expressing a Su(Hw)ΔN derivative with de-

letion of the N-terminal domain of the Su(Hw) (from

1 to 238 aa) tagged with the 3xFLAG (Fig. 1b) [20].

The Su(Hw)ΔN derivative, similar to the Su(Hw)Δ114,

interacts only with the Mod(mdg4)-67.2 protein. On the

polytene chromosomes of Drosophila larvae from the

;Su(Hw)ΔN-FLAG/ Su(Hw)ΔN-FLAG;su(Hw)

/su(Hw)

line, the Su(Hw)ΔN protein can be identified using an-

tibodies against FLAG, but not with antibodies against

the N terminal domain (1-150 aa) of the Su(Hw) pro-

tein (Figs.1 and3). In the y

line, the insertion sites

of the gypsy retrotransposon are located at the end

of the X chromosome distal to the chromocenter. Im-

munostaining with antibodies against FLAG showed

that the Su(Hw)ΔN protein binds to chromatin less ef-

Mod(mdg4)-67.2 PROTEIN 631

BIOCHEMISTRY (Moscow) Vol. 89 No. 4 2024

Fig. 3. Binding of the Su(Hw)

protein to polytene chromosomes. Immunostaining of polytene chromosomes of the salivary glands

of third instar larvae from the y

(wt) lines, y

;su(Hw)

/su(Hw)

(v/2), y

;su(Hw)

/su(Hw)

(v/E8), y

;su(Hw)

mod(mdg4)

su(Hw)

mod(mdg4)

(v-m/E8-m), and from the same lines expressing the Su(Hw)ΔN-FLAG or Su(Hw)+-FLAG proteins. Antibodies

against the FLAG epitope (αFLAG) and against the N-terminal domain of the Su(Hw) protein (αSu(Hw)-N) were used in the experi-

ments. Arrows indicate gypsy insertion at the end of X chromosome.

ficiently in comparison with the full length Su(Hw)+

protein, since stability of the Su(Hw) binding is medi-

ated by the CP190 protein (Fig.3) [18].

In the y

;su(Hw)

/su(Hw)

line, as in the

;su(Hw)

/su(Hw)

line, antibodies against the N-ter-

minal domain of Su(Hw) did not stain the Su(Hw)-

binding sites. However, in the su(Hw)

/su(Hw)

mutant

background, when the Su(Hw)ΔN derivative was ex-

pressed, these antibodies effectively stained numer-

ous Su(Hw) binding sites, including the gypsy sites at

the end of the chromosome X (Fig. 3). Consequently, an-

tibodies against the N-terminal domain identified the

full-length Su(Hw)

protein, which was recruited to chro-

matin through interaction with the Su(Hw)ΔN protein.

Introduction of the mod(mdg4)

mutation, which

completely inactivates the Mod(mdg4)-67.2 protein,

did not change anti-FLAG staining in the lines ex-

pressing Su(Hw)ΔN. However, in the y

;Su(Hw)ΔN-

FLAG/Su(Hw)ΔN-FLAG;su(Hw)

mod(mdg4)

/su(Hw)

mod(mdg4)

line, staining with antibodies against the

N terminal domain of Su(Hw) completely disappeared

(Fig. 3). Thus, in the absence of the Mod(mdg4)-67.2

protein, the Su(Hw)ΔN protein lost its ability to interact

with Su(Hw)

and recruit it to its binding sites.

Mod(mdg4)-67.2 mediates association of the

Su(Hw)

with the binding sites of the Su(Hw)ΔN

protein. To further confirm our results using chro-

matin immunoprecipitation, we tested binding level

of the Su(Hw)

protein in the y

;su(Hw)

/su(Hw)

line with five most studied Su(Hw)-dependent insula-

tors [41,42]. To detect the Su(Hw)

protein, antibodies

against the N-terminal domain of Su(Hw) were used.

As expected, the Su(Hw) protein was not detected at the

tested sites in either the su(Hw)

/su(Hw)

or su(Hw)

su(Hw)

mutant background (Fig.4a).

We next demonstrated, using anti-FLAG antibod-

ies, that in the lines expressing the Su(Hw)ΔN-FLAG

derivative, the Su(Hw)ΔN protein binds to all SBSs

(Su(Hw) binding sites) both in the presence and ab-

sence of the Mod(mdg4)-67.2 protein (Fig. 4b). Intro-

duction of the mod(mdg4)

mutation slightly reduced

the level of Su(Hw) binding, since Mod(mdg4)-67.2,

similarly to CP190, stabilizes association of the insula-

tor complex with SBS [20].