Tài liệu Báo cáo khoa học: Effect of deletion of the DNase I hypersensitive sites on the transcription of chicken Ig-b gene and on the maintenance of active chromatin state in the Ig-b locus docx

the transcription of chicken Ig-b gene and on themaintenance of active chromatin state in the Ig-b locus Hiroki Matsudo1, Kyoichi Osano1, Hiroshi Arakawa2and Masao Ono1 1 Department of L

the transcription of chicken Ig-b gene and on the

maintenance of active chromatin state in the Ig-b locus Hiroki Matsudo1, Kyoichi Osano1, Hiroshi Arakawa2and Masao Ono1

1 Department of Life Science, and Frontier Project ‘Life’s Adaptation Strategies to Environmental Changes’, Rikkyo University,

College of Science, Toshima-ku, Tokyo, Japan

2 GSF, Institute for Molecular Radiobiology, Neuherberg-Munich, Germany

In vertebrate cells, chromatin of active or potentially

active genes and flanking regions are characterized by

(a) sensitivity to DNase I [1–6]; (b) the presence of cell

type-specific DNase I hypersensitive sites (DHSs) [7,8];

and (c) core histone modifications such as acetylation

and methylation specific for active chromatin [9–11]

These characteristics have been reported specifically

for loci such as b-globin [1,3,12,13], and Ig-b⁄ growth

hormone (GH) [14–16] Thus, it is possible to

differen-tiate between active, or potentially active, and inactive

chromatin states by comparing differences in

chroma-tin sensitivity to nuclease and histone modifications

However, the mechanism by which the chromatin structure is modified in order to initiate cell

type-speci-fic gene expression as well as the maintenance of this active state in differentiated cells remains to be elucida-ted [17–19] Along with membrane immunoglobulin and Ig-a⁄ mb1, Ig-b is a component of the antigen receptor complex and belongs to the immunoglobulin superfamily [20,21] The Ig-b gene is expressed early in

B cell development [22–24] The mechanism of B cell-specific expression of mouse and human Ig-b genes has been studied mainly in the proximal promoter region, and several cis-elements and transacting factors have

Keywords

chicken Ig-b gene; DNase I hypersensitive

sites; DT40; histone acetylation;

transcription

Correspondence

M Ono, Department of Life Science,

College of Science, Rikkyo University,

Toshima-ku, Tokyo 171-8501, Japan

Fax ⁄ Tel: +81 339852387

E-mail: mono@rikkyo.ac.jp

(Received 11 September 2004, revised 3

November 2004, accepted 15 November

2004)

doi:10.1111/j.1742-4658.2004.04482.x

The role of DNase I hypersensitive sites (DHSs) in transcription of the B cell-specific Ig-b gene and in maintenance of active chromatin state in the Ig-b locus were examined A total of 10 DHSs were divided into four regions, and each region was deleted separately in chicken B lymphocyte-derived DT40 cells Deletion of three DHSs located between the Ig-b pro-moter and its upstream Na channel gene, resulted in the absence of Ig-b mRNA Three regions except the region in the Na channel gene were involved in the transcription of Ig-b gene The enhancing activity of DHSs

as determined by transient transfection assays did not always correlate with the effect of DHS deletion on the expression level of Ig-b mRNA In each deletion, cells contained the same DHSs as observed in the predeletion cells, indicating that deleted DHSs did not participate in the maintenance

of DT40-specific DHSs Enhanced acetylation of H3 and H4 histones at the Ig-b promoter and at DT40-specific DHSs was observed in cells in which DHSs between the Na channel gene and Ig-b promoter were deleted; therefore, these DHSs are prerequisite for transcription of the Ig-b gene but not required for the maintenance of active chromatin state in the Ig-b locus Thus, epigenetic factors required for the maintenance of the active chromatin state are suggested to reside in other regions than those deleted

in this study

Abbreviations

ChIP, chromatin immunoprecipitation; DHS, DNase I hypersensitive site; GH, growth hormone; LCR, locus control region; R-PCR, real-time polymerase chain reaction.

been identified [25–27] However, these studies have

not taken the state of chromatin into consideration

Cell type-specific DHSs not only correspond to

pro-moters and enhancers but also are likely to participate

in the establishment and maintenance of an active

chromatin state [18,28,29] In addition to their

pres-ence in and adjacent to active or potentially active

genes, cell type-specific DHSs are found in regions

situated far upstream or downstream of a gene [1,7,8]

To identify novel regions participating in cell

type-spe-cific transcription of the rat Ig-b gene, wide range

examination was carried out to find DHSs specific

for Ig-b producing cells The transcriptional enhancing

activities of the DHSs were examined by transient

transfection [30] Three regions having transcriptional

enhancing activity were found in the intergenic region

between the Ig-b and GH genes A member of the

OCT family transcription factors appeared to be

involved in the transactivation of the region which had

the highest enhancing activity of the three DHSs The

state of acetylation in H3 and H4 histones and

dime-thylation in the H3 histone Lys4 residue were

exam-ined by chromatin immunoprecipitation (ChIP) [31]

The active promoter and cell type-specific DHS with

enhancing activity showed active histone modifications

The sensitivity of chromatin to DNase I was measured

by real-time PCR (R-PCR) [16] The regions with

act-ive histone modifications were again sensitact-ive to

DNase I To determine the in vivo role of DHSs in B

cell-specific transcription of the Ig-b gene, genetic

stud-ies such as transgenics and gene targeting are required

Chicken B lymphocyte-derived DT40 cells are

partic-ularly useful for gene targeting because of the high rate

of homologous recombination [32,33] Chicken Ig-b

gene is expressed in DT40 cells [34] Previously, the

location of DHSs was surveyed in a 40 kb region

between 19 kb upstream and 21 kb downstream from

the transcriptional start site of the Ig-b gene [15]

Twelve DT40-specific DHSs were found: three in the

upstream Na channel gene, two between the Na channel

and Ig-b genes, one at the transcriptional start site,

one in the first intron of the Ig-b gene, four between

Ig-b and its downstream growth hormone (GH) genes,

and one in the downstream region of the GH gene

Transcriptional enhancing activity, as determined by

transient transfection, was associated with DHSs in the

Na channel gene and in the first intron of the Ig-b

gene Furthermore, the acetylation status of H3 and

H4 histones was examined [15] In the present study,

10 DT40-specific DHSs found in the Ig-b locus were

divided into four groups based on location For each

group, a deletion construct was introduced into DT40

cells Ig-b gene expression, acetylation of H3 and H4

histones, and the remaining DHSs in the deleted cells were examined to determine the roles of these DHSs

on the transcription of Ig-b gene and on the mainten-ance of the active chromatin state

Results

Generation of DT40/Cre cells with a long deletion

in one allele of the Ig-b locus

To genetically examine the mechanism of B cell-specific transcription of the Ig-b gene, we generated DT40⁄ Cre cells with a 16 kb deletion (16 kb Del) in one allele of the Ig-b locus (Fig 1) Introduction of the 16 kb Del construct into DT40⁄ Cre cells resulted in two deletion clones In one of these two clones, the Bsr gene between the loxP sequences was excised by Cre recombinase Following digestion of the genomic DNA with HindIII, homologous recombination and the excision of Bsr gene were confirmed by Southern hybridization using 0.9 kb DNA ()10.1 kb to )9.2 kb) as the probe (Fig 2) The sizes of the resulting bands for each clone (Fig 2B) were the same as expected from the restriction map (Fig 2A), and thus homologous recombination followed by excision of the Bsr gene was achieved

Establishment of DT40/Cre cells expressing small Ig-b mRNA

Initially, Ig-b protein appeared essential for the prolif-eration of DT40 cells To compensate for the absence

of Ig-b mRNA, a chicken b-actin promoter-driven Ig-b gene whose transcript (1.5 kb) is shorter than wild type (1.7 kb) due to deletion of the 3¢ untranslated region, was introduced into 16 kb Del cells From several puromycin resistant clones, one clone containing wild type and mutant Ig-b mRNA at a 2 : 1 ratio was selec-ted and used for subsequent studies (Fig 2C)

Generation of DT40/Cre clones deleted

in DT40-specific DHSs Ten DT40-specific DHSs are present in the 16 kb region

of the Ig-b locus (Fig 1) [15] These DHSs were divided into the following four groups (Fig 1): region I, having transcriptional enhancing activity [15] in the Na channel gene ()7.8 kb to )6.0 kb; DHSs, )7.3 kb and )6.5 kb); region II, promoter and upstream region of Ig-b gene ()2.7 kb to +0.05 kb; DHSs, )2.1 kb, )1.0 kb, and

0 kb); region III, in the first intron of the Ig-b gene (+0.6 kb to +1.1 kb; DHS, +0.9 kb); region IV, between Ig-b and GH genes (+4.5 kb to +6.7 kb; DHSs, +4.8 kb, +5.2 kb, +5.8 kb, and +6.1 kb)

Dele-tion constructs for regions I, II, and III were introduced

into the 16 kb Del cells with the extra Ig-b gene, and

between two and eight of the resulting clones were

found to have deletions for each of these regions as

judged by PCR To remove the Bsr gene from these

clones by Cre recombinase, tentative recombinants

were treated with tamoxifen and then blasticidin

sensi-tive clones were obtained In the course of the study,

Ig-b protein was found nonessential for the

prolifer-ation of DT40 cells (data not shown), and therefore

deletion of region IV was carried out by the

introduct-ion of the regintroduct-ion IV deletintroduct-ion construct into 16 kb Del

cells

Genomic Southern hybridization was performed to

confirm deletions (Fig 3) For region I (Fig 3A),

genomic DNA was digested with HindIII, and

hybrid-ized with a 0.9 kb probe for the Na channel gene; for

regions II (Fig 3B) and III (Fig 3C), DNA was cut

with EcoRI, and hybridized with Ig-b cDNA; for

region IV (Fig 3D), SacI-digested DNA was

hybrid-ized with a probe for the GH gene In each clone, the

sizes of the resulting bands that hybridized with the

probe coincided with that estimated from the

restric-tion map, and thus establishment of clones lacking region I to IV was confirmed A 1.3 kb band in Fig 3A, and a 3.1 kb band in Fig 3D are derived from the 16 kb Del allele, while the 6.6 kb band in Fig 3B,C is from the extra Ig-b gene

Expression of Ig-b mRNA in DT40/Cre clones deleted in DT40-specific DHSs

The level of Ig-b mRNA in cells with a deletion in region I was the same as that in 16 kb Del cells (Fig 4A), and thus this region was shown to be nones-sential for the transcription of Ig-b gene in DT40 cells Because region I enhances transcriptional activity when linked to the Ig-b promoter and introduced into DT

40 cells [15], activity of this region in vivo was demon-strated not to correlate with that observed in transient transfection assays Absence of Ig-b mRNA in cells with a deletion in region II (Fig 4A) demonstrated the essential role of this region in the transcription of the Ig-b gene Deletion of region III decreased the expres-sion of Ig-b mRNA to 40% of the expresexpres-sion in prede-letion cells (Fig 4A) Region III was found to enhance

Fig 1 Organization of the chicken Ig-b locus and deleted regions Exons are indicated by rectangles, and intron and intergenic regions by solid lines Horizontal arrows represent transcriptional orientations Exon numbers are indicated The hypothesized organization of a portion

of the Na channel gene for which the nucleotide sequence is unknown is represented by a dotted rectangle Sizes of the intergenic regions are shown in kb Upward arrows indicate DNase I hypersensitive sites (DHSs), distances from the transcriptional start site of the Ig-b gene

in kb are indicated underneath Black arrow, DT40-specific; gray arrow, LMH-specific; white arrow, common in both DT40 and LMH DHS arrows with enhancing activity are enclosed by rectangles Positions of the eight targets for real-time PCR (R-PCR) are shown in the middle [15]: 1, DHS )7.3 kb; 2, DHS )6.5 kb; 3, Na channel exon 24; 4, Ig-b promoter; 5, DHS +0.9 kb; 6, DHS +5.2 kb; 7, DHS +6.1 kb; 8, GH exon 4 At the bottom, positions of the arm sequences used for the targeting constructs are shown by black bars and regions deleted by dotted lines Numbers above the 5¢- and 3¢- ends of the region deleted are the distances from the transcriptional start site of the Ig-b gene.

transcriptional activity in transient transfection

four-fold in DT40 cells [15], and thus, in contrast to region

I, in vivo activity of this region correlated with the

enhancing activity in transient transfection assays

Region IV is located between the Ig-b and GH genes

and contains four DHSs within a 1.3 kb region The level of Ig-b mRNA in deletion cells was reduced to 56% of that in 16 kb Del cells (Fig 4B) The proximal part of this region (+4.6 kb to +5.5 kb containing +4.8 kb and +5.2 kb DHSs) has no transcriptional enhancing activity in transient transfection assays, while the distal part (+5.6 kb to +6.4 kb including +5.8 kb and +6.1 kb DHSs) decreases the transcrip-tional activity of the Ig-b promoter to 50% of the activity observed with the promoter alone Thus, like region I, in vivo activity of region IV did not correlate with the activity observed in transient transfection assays These observations demonstrate that regions II, III, and IV are involved in the transcription of the Ig-b gene in DT40 cells, while region I is not Further-more, the enhancing activity of DHSs as determined

by transient transfection assays does not always corre-late with the effect of DHS deletion on the expression level of Ig-b mRNA

Presence of DHSs in the Ig-b locus in cells deleted in regions I–IV

DHSs in the Ig-b locus were examined by genomic Southern hybridization in cells with region II deletion (Fig 5) DHSs located at )13 kb, )7.3 kb, and )6.5 kb in the Na channel gene were present in the region II deletion cells, similar to the parent cells (Fig 5A) DHSs located between the Ig-b and GH genes, and DHSs at +10.3 kb and +13 kb positioned downstream of the GH gene were both present in region II deletion cells as well as in the predeletion cells (Fig 5C,D) Because the sequence from )2.7 kb

to +0.05 kb was lost in region II deletion cells, no band corresponding to DHSs at)2.1 kb, )1.0 kb, and

0 kb was observed, while a DHS at +0.9 kb was pre-sent (Fig 5B) In summary, deletion of DHSs in region II did not affect the presence of DHSs in the Ig-b locus

DHSs in region I, III, and IV deletion cells were also examined DHSs detected between )13 kb and +13 kb

in the Ig-b locus [15] were again present in all three types of deletions although DHSs located at the dele-tion region disappeared (data not shown), and thus regions I, III, and IV were shown not to participate in the maintenance of DT40-specific DHSs in the Ig-b locus

Acetylation of H3 and H4 histones in cells deleted

in region II Acetylation of H3 and H4 histones is enhanced in and around potentially active and transcribed genes

A

Fig 2 Construction of DT40 ⁄ Cre cells lacking the 16 kb region in

one Ig-b allele and expressing extra Ig-b mRNA shorter than the

wild type (A) Organization of the Ig-b locus in DT40 ⁄ Cre cells

lack-ing the 16 kb region between )9.0 kb and +6.7 kb 1, wild type

allele; 2, 16 kb deletion (Del) allele with blasticidin resistance (Bsr)

gene; 3, 16 kb Del allele without Bsr gene Position of the probe

(0.9 kb; )10.1 kb HindIII to )9.2 kb AccI fragment) for hybridization

is indicated by a black rectangle Size and position of the HindIII

fragment hybridizable with the probe is shown below the map.

Exons and introns are indicated as in Fig 1 (B) Genomic Southern

hybridization HindIII digested DNA (2.5 lg) was separated by

0.75% agarose gel-electrophoresis and hybridized with the probe

shown in (A) 1, DNA from wild type DT40 ⁄ Cre cells; 2, 16 kb Del

allele with Bsr gene; 3, 16 kb Del allele without Bsr gene

Restric-tion fragment size (given in kb) was determined by kHindIII

Posi-tions of the 1.3 kb, 2.7 kb, and 6.2 kb bands are shown in the right

margin (C) Detection of the extra Ig-b mRNA by Northern

hybridiza-tion Total RNA (3.0 lg) was prepared from cells containing the

expression vector for extra Ig-b mRNA and Northern hybridized

with Ig-b cDNA probe (nucleotides 312–1542; PmaCI ⁄ EcoRI) [34].

1, 16 kb Del allele without Bsr gene; 2, 16 kb Del allele without Bsr

gene and with extra Ig-b construct Positions of 1.7 kb endogenous

and 1.5 kb extra Ig-b mRNA are shown in the right margin.

B

C

D

Fig 3 Confirmation of DHS deletions by Southern hybridization Homologous recombination and the excision of Bsr gene by tamoxifen treatment was first examined by PCR and then confirmed by Southern hybridization Deleted regions: (A), I ( )7.8 kb to )6.0 kb); (B),

II ( )2.7 kb to +0.05 kb); (C), III (+0.6 kb to +1.1 kb); (D), IV (+4.5 kb to +6.7 kb) Left, gene organization; right, autoradiogram of the South-ern hybridization Position of the probes, the size and position of restriction fragments hybridizable with the probe, and exon ⁄ introns are shown as in Fig 2A Positions of the DT40 specific-DHSs and the exon number are indicated as in Fig 1 1, wild type DT40 ⁄ Cre allele; 2, each DHS deletion allele with Bsr gene; 3, each DHS deletion allele without Bsr gene a, DNA from wild type DT40 ⁄ Cre cells; b, DNA from cells with 16 kb Del allele (with extra Ig-b gene); c, DNA from cells with 16 kb Del allele (with extra Ig-b gene) and each DHS deletion allele with Bsr gene; d, DNA from cells with 16 kb Del allele (with extra Ig-b gene) and each DHS deletion allele without Bsr gene Probes: (A), fragment from Na channel gene ( )10.1 kb to )9.2 kb; HindIII ⁄ AccI); (B) and (C), Ig-b cDNA (nucleotides 312–1542; PmaCI ⁄ EcoRI) [34]; (D),

GH gene (+6.6 kb to +7.4 kb; NcoI ⁄ NcoI) [46] Restriction enzyme digestion: A, HindIII; B and C, EcoRI; D, SacI Restriction fragment size (given in kb) was determined by kHindIII Size of the bands observed in each autoradiogram is indicated at the right side The 6.6 kb band detected in lanes b, c, and d in autoradiogram (B), and lanes c and d in (C) is derived from the extra Ig-b gene The 3.1 kb band found in lanes b, c, and d in autoradiogram (D) is derived from 16 kb Del allele.

[1,31,35–37] Enhanced acetylation is mainly observed

at DHSs and their surroundings, although low level of

acetylation has been reported at DHSs such as HS-85

and 3¢ HS1 in the mouse b-globin locus [1] Using

ChIP followed by R-PCR, the acetylation status of H3

and H4 histones in the chicken Ig-b locus has been

reported [15] In DT40 cells, acetylation of H3 and H4

histones is enhanced at the DHSs ()7.3 kb, )6.5 kb,

0 kb, +0.9 kb, +5.2 kb, and +6.1 kb) deleted in this

study [15]

The acetylation status in region II deletion cells was

compared with that of predeletion cells (Fig 6) A

prominent acetylation of both histones detected close to

the transcriptional initiation site (target 4) of the Ig-b

gene in wild type DT40 cells [15] was observed at the

same position both in predeletion cells and region II

deletion cells Similar to wild type DT40 cells [15],

enhanced acetylation was demonstrated at the targets

closely positioned to DT40-specific DHSs in the Na

channel gene (close to DHSs at )7.3 kb and )6.5 kb;

targets 1 and 2), in the Ig-b first intron (DHS at

+0.9 kb; target 5), and between the Ig-b and GH genes

(DHSs at +5.2 kb and +6.1 kb; targets 6 and 7) At

these targets (targets 1, 2, 5, 6, and 7) in the region II

deletion cells, acetylation levels of both histones were

the same as that of the predeletion cells In chicken

liver-derived LMH cells where no Ig-b mRNA is

detec-ted [15], and thus used as the negative control, no

enhanced acetylation of either histone was demonstrated

at the targets located close to the Ig-b promoter and DT40-specific DHSs Acetylation levels of both histones

at the targets in the last exon of the Na channel gene (target 3) and in the fourth exon of the GH gene (target 8) was shown to be the same as in the cells before and after region II deletion In region II deletion cells, no Ig-b mRNA was detected but the acetylation status of both histones before and after deletion was demonstra-ted to be the same, indicating that region II is essential for B cell-specific transcription of the Ig-b gene but unnecessary for the maintenance of the active chromatin state in the Ig-b locus

In transient transfection assays, the DNA fragment containing a DHS at )2.1 kb shows no enhancing activity of the Ig-b promoter in DT40 cells, while the DHS at )1.0 kb decreases promoter activity to 50% [15] It would be interesting to learn whether the dele-tion of DHSs at )2.1 kb and )1.0 kb has any effect

on the level of Ig-b mRNA; however, deletion of these DHSs has been unsuccessful so far (data not shown)

Discussion

Ten Ig-b-producing cell-specific DHSs were grouped into four regions, and each of these regions was dele-ted in DT40 cells The roles of the DHSs on Ig-b gene expression and the maintenance of active chromatin state in the Ig-b locus were examined in the dele-tion cells Establishment and maintenance of active

Fig 4 Level of Ig-b mRNA in DT40 ⁄ Cre cells lacking DT40-specific DHSs Using total RNA (3.0 lg) prepared from the deletion clones, the level of Ig-b mRNA was examined by Northern hybridization with the Ig-b cDNA probe (nucleotides 312–1542; PmaCI ⁄ EcoRI) [34] Intensity

of the bands was compared using TYPHOON 9210 and IMAGE QUANT (A) Positions of 1.7 kb endogenous and 1.5 kb extra Ig-b mRNA are shown at the left and right, respectively Lanes: RNA prepared from: 1, wild type DT40 ⁄ Cre cells (Wild type); 2, 16 kb Del allele (with extra Ig-b gene) (Control 1); 3, 16 kb Del allele (with extra Ig-b gene) and the region I deletion allele (I Del); 4 and 5, 16 kb Del allele (with extra Ig-b gene) and the region II deletion allele (II Del); 6, 16 kb Del allele (with extra Ig-b gene) and the region III deletion allele (III Del) (B) Lanes: 1 and 2, 16 kb Del allele (without extra Ig-b gene) (Control 2); 3 and 4,16 kb Del allele (without extra Ig-b gene) and the region IV deletion allele (IV Del) Probes: 1 and 3, human glyceraldehyde-3-phosphate dehydrogenase (G3PDH) cDNA; 2 and 4, Ig-b cDNA.

A B

Fig 5 DNase I hypersensitive sites in the DT40 ⁄ Cre cells lacking region II Isolated nuclei were treated with DNase I for 3 min at 20
C Concentration of DNase I for treatment of nuclei from left to right: 75, 50, 25, 0 UÆmL)1 The DNA was purified from nuclei and digested with EcoRI (A), ScaI (B), EcoRV (C) or EcoT22I (D) The digests were Southern hybridized with 0.9 kb ScaI ⁄ KpnI DNA ( )4.6 kb to )3.7 kb; nucleotides 849–1751 in accession number AB066568) (A,B) [34], 1.2 kb Ig-b cDNA (nucleotides 312–1542; PmaCI ⁄ EcoRI) [34] (C) or 0.2 kb

GH exon 5 (nucleotides 3515–3744 in D10484) (D) [46] Restriction fragment size (given in kb) was determined by kHindIII Analyzed region, DHS, and position of the probe are shown at the bottom The 2.5 kb band observed in (C) is derived from the extra Ig-b gene Autoradio-grams: left, wild type DT40 ⁄ Cre cells having 16 kb Del allele (with extra Ig-b gene); right, wild type DT40 ⁄ Cre cells having 16 kb Del allele (with extra Ig-b gene) plus the region II deletion allele (mutant).

chromatin is essential for transcription of cell

type-spe-cific genes Active chromatin is characterized by an

open chromatin structure that is generally sensitive to

DNaseI, presence of cell type-specific DHSs [7,8], and

active-type modification of core histone tails [9–11]

From the earliest stage of the study [6], DNase I

gen-eral sensitivity has been demonstrated to be a good

parameter to determine chromatin state [1–5] Distinct

differences are often obtained between positive and

negative controls in DHS and histone modification

analyses; in contrast, these differences are often much

smaller in DNase I general sensitivity analysis

Although DNase I general sensitivity analysis of the

chicken Ig-b locus has not been finished because of the technical difficulty of this type of assay, there may be

no difference in the range and extent of general sensi-tivity to DNase I with or without DHSs deleted in this study because the positions of DHSs remain the same

In the chicken b-globin locus, the DNase I sensitive region coincides with the region in which the core histones are hyperacetylated [3,38] In the mouse b-glo-bin locus, hyperacetylated H3 and H4 histones are located in a much wider DNase I sensitive region [1,12] Thus, the DNase I sensitive chromatin state is not always associated with active-type modification of core histones Furthermore, a cell type-specific DHS, HS-85, in the mouse b-globin⁄ odorant receptor locus is located in a region with a low acetylation level of the core histones and a chromatin state in the flanking region resistant to DNase I [1] In this case, the forma-tion of cell type-specific DHS again does not occur in the context of a DNase I sensitive chromatin region,

or a region with active-type histone modifications Transgenic mice with a 99 bp deletion (containing two Pit-1 binding sites) of the DHS I region which is located within the locus control region (LCR) in the human GH locus, show loss of H3 and H4 acetyla-tion ranging from 32 kb upstream of the transcrip-tional start site of the GH gene to the GH gene [18] Thus, the deleted sequence is essential for the estab-lishment and maintenance of acetylation in the GH locus Because deletion of this region has no effect on the formation of cell type-specific DHSs, the region required for H3 and H4 acetylation does not neces-sarily coincide with that involved in the DHS forma-tion Deletion of DHS1–6, the entire mouse b-globin LCR, results in extraordinarily low levels of b-globin mRNA However, both the acetylation state of the promoter region of the active b-globin gene [19] and the DNase I general sensitivity in the b-globin locus are the same as the wild type control [17] Thus, there again has been no description either of the region involved in DNase I general sensitivity or

of the region for the formation of DHSs, although chromatin opening activity of the LCR has been des-cribed in several loci using transgenic mouse models [29,39–42]

Because the chromatin state of the cells with any deletion in the regions I–IV in the Ig-b locus, judged

by DHSs as the indicator, was the same as that in pre-deletion cells, the region required for the maintenance

of the active chromatin state is presumed to be present

at the DHSs located further upstream or downstream

of the DHSs examined in this study Two

DT40-speci-fic DHSs, one at )13 kb in the Na channel gene and the other at +13 kb downstream of the GH gene, have

Fig 6 H3 and H4 acetylation of the Ig-b locus in the region II

dele-tion cells The acetyladele-tion of wild type DT40 ⁄ Cre cells with 16 kb

Del allele (with extra Ig-b gene), wild type DT40⁄ Cre cells with

16 kb Del allele (with extra Ig-b gene) plus the region II deletion

allele, and LMH cells is indicated by the black, gray and white bars,

respectively (Top) H3 acetylation (Middle) H4 acetylation

Acetyla-tion values are shown by fold enrichment compared with untreated

control DNA as reported previously [15,31] Three amplifications

were performed for each target The values of the ng control DNA

equivalents were estimated using standard calibration, and fold

enrichment was determined from these values [31] Values are

mean ± SE An over-scaled bar is shown by double wavy lines, and

its mean value is shown at the left Bottom; the positions of eight

targets are shown Exon ⁄ introns are shown as in Fig 1 Arrows,

DHSs; +, transcriptional start site.

already been described [15], and deletion of these

DHSs should be performed in the future

Transcription of the adult-type mouse b-globin gene

is suggested to require the assembly of DHSs into one

complex named the active chromatin hub at around

the transcriptional start site This complex includes six

DHSs in the LCR located 40–60 kb upstream from the

active gene, DHSs present in the odorant receptor

genes found 40 kb further upstream from the LCR,

and those present in 20 kb downstream from the

b-glo-bin gene [43] Combined deletion of the separate DHS

groups would be necessary in future studies because

the deletion of a single region showed no remarkable

effect on the chromatin state of the Ig-b locus

The activities of DHSs in the Ig-b locus examined

by transient transfection [15] did not correlate with

those determined by deletion in regions I and IV

Sim-ilar situations have been reported at DHSs within the

human b-globin LCR [44] Although the reason for the

inconsistency of the DHS activity between transient

transfection and in vivo deletion is unknown, the

phe-notype observed in deletion cells is likely to reflect the

in vivorole of the DHS because genomic DNA is

dele-ted from its native context Another reason may be the

redundancy between different elements in the Ig-b

locus

Materials and methods

Cells

Liver-derived LMH cells were obtained from the Japanese

Cancer Research Resources Bank (Tokyo, Japan)

DT40⁄ Cre cells [45] that produce MerCreMer protein, a Cre

recombinase with hormone binding domains of the mutant

mouse estrogen receptor (Mer) on both ends, which localizes

in the cytoplasm in the absence of estrogen derivative,

4-hydroxy tamoxifen Cells were propagated in RPMI1640

(Nissui, Tokyo, Japan)⁄ 10% fetal bovine serum (JRH

Bio-sciences, Lenexa, KS, USA)⁄ 1% chicken serum (GibcoBRL,

Grand Island, NY, USA) at 39.5
C for DT40 ⁄ Cre and

37
C for LMH DT40 ⁄ Cre cells were grown with

2 mgÆmL)1 of geneticin (GibcoBRL) which is required for

establishment of DT40⁄ Cre cells For selection, blasticidin

(Funakoshi, Tokyo, Japan) and puromycin (Sigma-Aldrich,

St Louis, MO, USA) were used at 30 lgÆmL)1 and

0.5 lgÆmL)1, respectively

Preparation of targeting constructs and extra Ig-b gene

expression construct pLoxBsr [45] containing the chicken

b-actin promoter-driven blasticidin resistance (Bsr) gene

with mutant loxP sequences at both ends was used as a

tar-geting vector For the tartar-geting construct, genomic DNA of

about 2 kb in size was linked on both sides with the actin

promoter-driven Bsr gene Positions of left and right arms

of the five constructs (16 kb Del, I to IV; Fig 1) are as fol-lows (the transcriptional start site of Ig-b gene is shown as base number one): 16 kb Del-L ()11.4 kb to )9.0 kb, SacI⁄ MluI; 2.4 kb), 16 kb Del-R (+6703 to +8768;

2066 bp), I-L ()10.1 kb to )7.8 kb, BamHI ⁄ XhoI; 2.3 kb), I-R ()6009 to )3942; 2068 bp), II-L ()4613 to )2692;

1922 bp), II-R (+50 to +1959; 1910 bp), III-L ()1369 to +578; 1948 bp), III-R (+1079–3130; 2052 bp), IV-L (+2721 to +4524, 1804 bp), IV-R (same as 16 kb Del-R) Based on the reported nucleotide sequences [15,34,46], arm DNAs were amplified with Pyrobest polymerase (Takara Bio, Kyoto, Japan), and ligated into the vector Restriction fragments were used for the arms located in the unsequ-enced region The Bsr gene was inserted in the same orien-tation as the flanking gene DT40⁄ Cre cells were electroporated with linearized DNA and clones were obtained as described [47]

Using cloned chicken Ig-b cDNA as the template, 0.9 kb DNA (nucleotide number 20–930) [34] was amplified by Pyrobest polymerase and cloned into HindIII⁄ NheI sites of the pExpress vector [45] The chicken b-actin promoter-driven puromycin resistance (Puro) gene was inserted into the XhoI site of the construct

Selection of homologous recombinant clones

by PCR and tamoxifen treatment

To determine homologous recombination, PCR was carried out as described [45] using the following primers: 5¢-CGATTGAAGAACTCATTCCACTCAAATATACCC-3¢ (in Bsr gene) [45] Primers (30-mer) in the genome were set

up at 10–100 bases downstream from the right arm: 5¢-TAGTTTCTCAAACACTCTGTCTGAGGTGCC-3¢ (16 k Del; +8778 to +8807); 5¢-ATGGGTTCATAGGAGACC TTTGAGGGGTTG-3¢ (I: )3782 to )3811); 5¢-TAGATG CCGTTGTCCTCGTAGCTGATCCTG-3¢ (II: +2085 to +2114); 5¢-AGTGATGTCCTCGTAGGTGGCAATCTGC TC-3¢ (III: +3438 to +3467) (IV: same as 16 k Del) Clones whose DNA contained about 3 kb predictable sequences by PCR were selected as tentative clones with homologous recombination

DT40⁄ Cre clones with homologous recombination were treated with 0.1 mm 4-OH-tamoxifen (Sigma, St Louis,

MO, USA) for 24 h and then cloned in 96 well microtiter plates Deletion of the Bsr gene was confirmed by genomic Southern hybridization

DNase I digestion, hybridization, and histone acetylation analysis

For DNase I hypersensitivity analysis, nuclei were prepared and treated with DNase I (Takara Bio) and DNA was pre-pared as described [30] Restriction enzyme digestion and

Southern hybridization were carried out as reported

previ-ously [48] Probe DNA was labeled by the random-priming

methods using [32P]dCTP[aP] The following probes were

used: Ig-b cDNA (nucleotides 312–1542, 1231 bp) [34], Na

channel gene ()10.1 kb to )9.2 kb, HindIII ⁄ AccI; 0.9 kb),

GHgene (+6560 to +7380, NcoI⁄ NcoI; 821 bp) [34]

Pre-paration of total RNA followed by Northern hybridization

was performed as described previously [31] A human

glyc-eraldehyde-3-phosphate dehydrogenase cDNA probe was

obtained from Clontech (Palo Alto, CA, USA) The

inten-sity of bands on the Northern hybridization was compared

by typhoon 9210 and image quant (Amersham

Bioscienc-es, Piscataway, NJ, USA) Acetylation status of H3 and H4

histones was examined by chromatin immunoprecipitation

(ChIP) and real-time PCR (R-PCR) as described previously

[15]

Acknowledgements

We thank the Japanese Cancer Research Resources

Bank for providing cells This work was supported by

Rikkyo University for the Promotion of Research

References

1 Bulger M, Schubeler D, Bender MA, Hamilton J,

Farrell CM, Hardison RC & Groudine M (2003) A

complex chromatin landscape revealed by patterns of

nuclease sensitivity and histone modification within

the mouse b-globin locus Mol Cell Biol 23,

5234–5244

2 Durrin LK, Weber JL & Gorski J (1984) Chromatin

structure, transcription, and methylation of the

prolac-tin gene domain in pituitary tumors of Fischer 344 rats

J Biol Chem 259, 7086–7093

3 Hebbes TR, Clayton AL, Thorne AW &

Crane-Robinson C (1994) Core histone hyperacetylation

co-maps with generalized DNase I sensitivity in the

chicken b-globin chromosomal domain EMBO J 13,

1823–1830

4 Jantzen K, Fritton HP & Igo-Kemenes T (1986) The

DNase I sensitive domain of the chicken lysozyme gene

spans 24 kb Nucleic Acids Res 14, 6085–6099

5 Lawson GM, Knoll BJ, March CJ, Woo SL, Tsai MJ

& O’Malley BW (1982) Definition of 5¢ and 3¢

struc-tural boundaries of the chromatin domain containing

the ovalbumin multigene family J Biol Chem 257,

1501–1507

6 Weintraub H & Groudine M (1976) Chromosomal

sub-units in active genes have an altered conformation

Science 193, 848–856

7 Elgin SC (1988) The formation and function of DNase

I hypersensitive sites in the process of gene activation

J Biol Chem 263, 19259–19262

8 Gross DS & Garrard WT (1988) Nuclease hypersensi-tive sites in chromatin Annu Rev Biochem 57, 159–197

9 Roth SY, Denu JM & Allis CD (2001) Histone acetyl-transferases Annu Rev Biochem 70, 81–120

10 Turner BM (2000) Histone acetylation and an epigenetic code Bioessays 22, 836–845

11 Wu J & Grunstein M (2000) 25 years after the nucleo-some model: chromatin modifications Trends Biochem Sci 25, 619–623

12 Forsberg EC, Downs KM, Christensen HM, Im H, Nuzzi

PA & Bresnick EH (2000) Developmentally dynamic his-tone acetylation pattern of a tissue-specific chromatin domain Proc Natl Acad Sci USA 97, 14494–14499

13 Litt MD, Simpson M, Gaszner M, Allis CD & Felsen-feld G (2001) Correlation between histone lysine methy-lation and developmental changes at the chicken b-globin locus Science 293, 2453–2455

14 Elefant F, Cooke NE & Liebhaber SA (2000) Targeted recruitment of histone acetyltransferase activity to a locus control region J Biol Chem 275, 13827–13834

15 Murakami R, Osano K & Ono M (2004) DNase I hypersensitive sites and histone acetylation status in the chicken Ig-b locus Gene 337, 121–129

16 Osano K, Otsuka A & Ono M (2004) State of chroma-tin sensitivity to DNase I in rat Ig-b⁄ growth hormone locus determined by real-time PCR Biol Pharm Bull 27, 222–225

17 Bender MA, Bulger M, Close J & Groudine M (2000) b-globin gene switching and DNase I sensitivity of the endogenous b-globin locus in mice do not require the locus control region Mol Cell 5, 387–393

18 Ho Y, Elefant F, Cooke N & Liebhaber S (2002) A defined locus control region determinant links chroma-tin domain acetylation with long-range gene activation Mol Cell 9, 291–302

19 Schubeler D, Groudine M & Bender MA (2001) The murine b-globin locus control region regulates the rate

of transcription but not the hyperacetylation of histones

at the active genes Proc Natl Acad Sci USA 98, 11432– 11437

20 Reth M (1992) Antigen receptors on B lymphocytes Annu Rev Immunol 10, 97–121

21 Wienands J (2000) The B-cell antigen receptor: forma-tion of signaling complexes and the funcforma-tion of adaptor proteins Curr Top Microbiol Immunol 245, 53–76

22 Bain G, Maandag EC, Izon DJ, Amsen D, Kruisbeek AM, Weintraub BC, Krop I, Schlissel MS, Feeney AJ, van Roon M, et al (1994) E2A proteins are required for proper B cell development and initi-ation of immunoglobulin gene rearrangements Cell

79, 885–892

23 Lin H & Grosschedl R (1995) Failure of B-cell differen-tiation in mice lacking the transcription factor EBF Nature 376, 263–267