Báo cáo Y học: Novel regulatory regions found downstream of the rat B29/Ig-b gene docx

Novel regulatory regions found downstream of the rat B29/Ig- b gene Ayano Komatsu, Akira Otsuka and Masao Ono Life Science Course, Department of Chemistry, College of Science, Rikkyo University, Toshima-ku, Tokyo, Japan To search for novel regulatory regions, we examined the features of chromatin structure in the rat B29/Ig- b gene and its flanking regions by determining DNase I hypersensitive sites (DHS) in plasmacytoma-derived Y3 cells. Six Y3 cell- specific DHS were detected at )8.6, promoter, +0.7, +4.4, +6.0, a nd +8.7 kb. The DHS a t + 4.4, +6.0, and +8.7 kb were present in the intergenic region between B29/Ig- b and growth hormone (GH) genes and were mapped inside con- served sequences in rat and humans. In transient transfection into Y3 cells, 2.9-kb DNA containing the +4.4 and +6.0-kb DHS demonstrated six times more enhancing activity than B29/Ig-b promoter alone. Three intergenic DHS each possessed enhancing activity that was highest in the +4.4-kb region. In the electrophoretic mobility shift assay, a major band shift was demonstrated with Y3 nuclear extract and 0.3-kb DNA containing the +4.4-kb region with a con- served 0.22-kb sequence. By footprint a nalysis, 20 bases in the middle of the 0.3-kb DNA were protected by Y3 nuclear extract in which the consensus binding site for the OCT family was present. Deletion of the footprinted region reduced enhancing activity to t hat of the B29/Ig-b promoter alone. The sequence responsible for t he major band shift and transcriptional enhancing activity in the conserved +4.4-kb region thus coincided with the 20-bp footprinted region. Keywords: B29/ Ig-b gene; cell type-specific gene expression; chromatin structure; conserved regions; D Nase I h yper- sensitive s ite(s). Along with membrane immunoglobulin (mIg) and Ig-a/ mb1 (a B-cell-spec ific membrane protein), B29/Ig-b is a component of the mIg receptor complex and belongs to the immunoglobulin superfamily [1,2]. Also called CD79b in humans, B 29/Ig-b is made of 228 amino acids in rat [3] and 229 amino acids in humans [4]. The B29/Ig-b gene is composed of a leader peptide and four domains: immuno- globulin, membrane proximal, transmembrane, and cytoplasmic. The cytoplasmic region of the mIg heavy chain is quite short and incapable o f signal transduction; instead, Ig-a/mb1 and the B29/Ig-b heterodimer play the main roles at the beginning of B-cell receptor (BCR)- mediated signal transduction [1,2]. A significant motif in Ig-a/Ig-b found by Reth [5], named the immunoreceptor tyrosine-based activation motif (ITAM), is located close to the N -terminal r egion of the cytoplasmic domain. K inases in the Src family, such as Lyn phosphorylate tyrosine residues in ITAM and phosphorylated ITAM, recruit Syk via the SH2 domain to initiate signal transduction. The B29/Ig-b genes, 3.1 kb in rat [3] and 3.6 kb in humans [6], each have six exons. In the 88-kb region of the rat B29/growth hormone (GH) locus, six genes were present in the following order [3,7,8]: skeletal muscle ( SkM) sodium channel (5¢ to 3¢), B29/Ig- b (5¢ to 3¢), GH (5¢ to 3¢), testicular cell adhesion molecule 1 (TCAM-1,3¢to 5¢), BAF60b (component of the chromatin remodeling factor, 5¢ to 3¢), SUG/p45 (transcription factor/proteasome regu latory sub- unit, 3¢ to 5¢). In humans, the SkM sodium channel i s upstream of the CD79b (B29/Ig-b) gene [3] and the GH genes are downstream [9]. A mong genes p resent at this locus, BAF60b and SUG/p45 are house-keeping genes expressed in all tissues and c ells so far examined, while SkM sodium channel, B29/Ig-b, GH,andTCAM-1 genes are expressed cell type-specifically in cells ontogenetically unrelatedtoeachother. The B29/Ig- b gene is expressed early in B-cell develop- ment when Ig genes are still in the g ermline configuration [10–12]. Mice lacking B 29/Ig-b have completely blocked B-cell development at the immature B-cell stage [13]. Mouse, human, and rat B29/Ig-b genes lack the TATA box and have multiple transcriptional initiation sites [14,15]. In the region starting from the representative initiation site to 0.17-kb upstream, SP1, ETS, OCT, and Ikaros motifs essential for B-cell-specific expression of the mouse B29/Ig-b gene were found [16]. In the 0.35-kb upstream region of the mouse gen e, two silencer elements each composed of 30 bp, have been reported [17]. In the 0.5-kb upstream region of the human gene, a 30-bp positive transcription control element is present [16]. An early B-cell factor (EBF) essential for early B lymphocyte development has been shown to be involved in B29/Ig-b gene expression from the promoter region to 0.17-kb upstream [18]. Thus, the me chanism of B -cell-specific Correspondence to M. Ono, Life Science Course, Department of Chemistry, College of Science, Rikkyo University, 3-34-1 Nishi- ikebukuro, Toshima-ku, Tokyo 171-8501 Japan. Fax/Tel.: + 81 339852387, E-mail: mono@rikkyo.ac.jp Abbreviations:BCR,B-cellreceptor;DHS,DNaseIhypersensitive sites; D-MEM, Dulbecco’s modified minimal essential medium; EBF, early B-cell factor; GH, growth hormone; GraP-DH, glyceraldehyde- 3-phosphate dehydrogenase; ITAM, immunoreceptor tyrosine-based activation motif; LCR, locus control region; Luc, luciferase; mIg, membrane immunoglobulin; SkM, skeletal muscle; TCAM-1, testi- cular cell adhesion molecule 1. Enzymes: d eoxyribonuclease I (EC 3.1.21.1); glyceraldehyde-3-phos- phate dehydrogenase (phosphorylating) (EC 1.2.1.12); firefly luciferase (EC 1.13.12.7). Note: the nucleotide sequences reported in this paper will appear in the DDBJ/EMBL/GenBank nucleotide sequence databases with accession numbers AB062673 and AB062674. (Received 2 0 September 20 01, revised 21 December 2001, accepted 2 January 2002) Eur. J. Biochem. 269, 1227–1236 (2002) Ó FEBS 2002 expression of the B29/Ig-b gene has been sought in the proximal promoter region, and s everal ci s-elements and transacting factors have been found. The chromatin of an actively transcribed locus is sensitive to DNase I digestion and cell type-specific DNase I hyper- sensitive s ites (DHS) are repeatedly found in and around the gene [19,20]. Cell type-specific DHS often correspond to promoters or enhancers and some may be locus control regions (LCR) [21,22] that confer cell type-specific expres- sion of th e introduced gene in a position-independent manner in transgenic mice. They may be considered to be landmarks in the search for novel regulatory regions of transcription. Cell type-specific DHS are present not only in promoter regions but also in regions situated far upstream or downstream along the gene [19,20]; detailed examination is essential for finding these sites. In this study, Y3 cell- specific DHS in the B29/Ig-b gene and its flanking regions were examined in rat Y3 cells expressing B29/Ig-b mRNA. In the region between th e B29/Ig-b and GH genes, three Y3 cell-specific DHS were found and each possessed transcrip- tional enhancing activity for transie nt transfection into Y3 cells. The nucleotide sequenc es of these DHS are conserved in rat and humans. Of these DHS, the 0 .3-kb region with the highest enhancing activity was analyzed exclusively. MATERIALS AND METHODS Cells, clones, Northern hybridization, and sequencing Rat plasmacytoma-derived Y3-Ag1.2.3 cells were obtained from the Japanese Cancer Research Resources Bank (Tokyo, Japan). Buffalo rat liver-derived BRL cells were obtained from RIKEN Cell Bank (Tsuku ba, Japan). Rat pituitary-derived GC cells were obtained from M. Karin (University of California, San Diego, CA, USA). The Y3 and BRL cells were propagated in Dulbecco’s modified minimal essential medium (D-MEM)/10% fetal bovine serum ( JRH Bioscience s, US A) and the GC cells were Fig. 1. Expression of B29/Ig-b mRNA (A) and DNase I hypersensitivity by SkM sodium channel and B29/Ig-b gene chromatins of Y3 cells (B). (A) Poly ( A)-rich RNA (3.0 lg) electrophoresed on a 2.2- M HCHO/1% agarose gel and hybridized with 32 P-labeled rat cDNA (nucleotides 74 to 880) [3]. The RNA size (given in kb) was determined using a commercial RNA molecular size marker (Roche Diagnostics). At the bottom, reprobing with GraP-DH is shown. (B) Isolated Y3 nuclei t reated with DNase I for 3 min at 20 °C. Concentration of DNase I for treatment of nuclei from left to right: 200, 100, 50, 25, 13, 0 UÆmL )1 . T he D NA was purifi ed from nuclei and digested with NheI. The digests (2.0 lg) were separated by electrophoresis on a 0.75% agarose gel and blotted onto n itrocellulose filters. The blot was hybridized with a 1.8-kb probe ( )12.5 to )10.7 kb; SalI/SpeI). Restriction fragment si ze (given in kb) w as determined by kHindIII. An alyzed region, DHS, and position of the probe are shown at the bottom. 1228 A. Komatsu et al. (Eur. J. Biochem. 269) Ó FEBS 2002 grown in D-MEM/F12/5%fetal bovine s erum/5% horse serum (HS; JRH Biosciences). Total RNA a nd poly(A)-rich RNA from Y3 cells and male Wistar rat spleen were prepared as previously reported [ 23]. The RNA was denatured, frac tionated i n a 1% agarose gel containing formaldehyde, transferred to a nitrocellulose filter, and hybridized with probes as in a previous study [23]. Probe DNA was labeled by the random-priming method using [a- 32 P] dCTP. A fter hybridization, the fi lter was washed in 0.3 · NaCl/Cit (1 · NaCl/Cit: 150 m M NaCl and 15 m M sodium citrate)/0.1% SDS at 65 °C. The filter was then probed again with a probe of human glyceraldehyde 3-phosphate dehydrogenase cDNA (GraP-DH Clontech, Palo Alto, CA, USA) according to the manufacturer’s instructions. Cosmid clones possessing the rat B29/GH locus have been described previously [7]. Using a dye termination c ycle sequencing kit (Applied Biosystems, Foster City, CA, USA), nucleotide sequences in both DNA strands were determined with a commercial DNA sequencer (model 377, Applied Biosystems). Isolation of nuclei, DNase I digestion, and Southern hybridization The Y3 nuclei were prepared in N1 buffer (15 m M Tris/HCl, pH 7.5, 60 m M KCl, 15 m M NaCl, 5 m M MgCl 2 ,0.5m M dithiothreitol, 0.1 m M EGTA, 0.3 M sucrose) containing 0.2% NP-40 as described previously [24]. For DNase I (TaKaRa Shuzo, Kyoto, Japan) digestion, 60 lLofenzyme in N1 buffer was mixed i n 750 lLwith3·10 7 nuclei and incubated for 3 min at 20 °C. Reactions were terminated by the addition of 75 lLstopsolutioncontaining5%SDSand 100 m M EDTA. The DNA preparation, restriction enzyme digestion, and Southern hybridization were carried out as previously reported [24]. After hybridization at 65 °C overnight, the filter was washed in 0.3 · NaCl/Cit/0.1% SDS at 65 °C. Transfection The Pica Gene basic vector 2 PGV-B2 (4818 bp; Wako Chemicals, Osaka, Japan), promoter vector SV-P (5010 bp; PGV-B2 plus SV-40 promoter) and control vector SV-P+E (5256 bp; SV-P plus SV -40 enhancer), possessing the firefly lucifer ase (Lu c) gene, se rv ed as reporter genes. The sea pansy TK control v ector pRL-TK (4045 bp; Wako Chemicals), possessing both t he herpes simplex virus thymidine kinase promoter and the sea pansy Luc gene, was used as an internal control. Transfection was conducted with a commercial kit (Effectene Transfection Kit; Quiagen, Germany) for 20 h according to the m anufacturer’s i nstruction s. In 1.6 mL of medium, 1 · 10 6 Y3 cells were transfected with 0.2 lg reporter plasmid and 5 ng pRL-TK,then cultured in a 6-well cell culture plate. Luciferase activity was measured with a comme rcial kit (Pica Ge ne Dual Sea Pansy Luminescence Kit; Wako Chemicals) and a lumiphotometer (Berthold Lumat model LB 9507). To normalize luciferase activity, activity from the firefly Luc gene was divided by activity from the sea pansy Luc gene and mean values were obtaine d for triplicate samples (± SE). With cosmid DNA (clone rcGH 10) [7] as a template, the B29/Ig-b promoter reporter B29-P was m ade by Pyrobest polymerase (TaKaRa, Kyoto, Japan) amplifica- tion of th e B29/Ig-b promoter (nucleotides )201 to +30) Fig. 2. DNase I hypersensitivity of SkM sodium channel and B29/Ig-b gene chromatins. DNase I treatment and Southern hybridization were carried out as previously described. The DNA was digested with ApaI(A)orHindIII (B) and hybridized with a 0.6-kb probe (nucleotides 1571 to 2158). Left, Y3 ; right, BRL. Ó FEBS 2002 Rat B29/Ig-b gene regulatory regions (Eur. J. Biochem. 269) 1229 [3] followed by cloning into the HindIII/BglII sites of PGV-B2. The B29/Ig-b promoter sequence was selected for its similarity of those of mouse and humans [15,16]. The DHS were amplified with a cosmid template and cloned into the MluI/NheI sites of B29-P or SV-P. For the )8.6-kb unsequenced DHS region, 0.8-kb DNA was obtained by generating XbaI()8.9 k b)/HindIII ()8.1 k b) digestion of c osmid DNA. Activity in t he following regions (nucleotide numbers from transcriptional start site of B29/Ig-b gene) was examined by transfection: +0.7 kb (540–809), +4.4a (4193–4483), +4.6b (4451–4727), +6.0c (5902–6227), +6.5d (6352–6631), +4.4a to +4.6b (4193– 4727), + 6.0c to +6.5d (5902–6631), +8.7 kb ( 8630–9238), and +3.8 kb to + 6.7 k b (3831–6652). The activities of at least two i ndependent clones were determined to b e the same, so the results of a r epresentative clone are shown. A reporter plasmid lacking a 30-bp sequence (4313– 4342) including the 20-bp footprinted region was produced as follows: two 21-bp primers (5¢ 4343–4363 3¢;5¢4312– 4292 3¢)withaSpeI site and three more nucleotides a t the 5¢ end were used for amplification by Pyrobest polymerase with the B29-P reporter t hat h ad the +4.4a DHS. Amplified DNA was digested with SpeI, ligated, and transformed. A reporter lacking basepairs 4193–4222 was prepared as a control. Electrophoretic mobility shift assay After being r insed with phosphate-buffered saline (NaCl/P i ), 1 · 10 8 Y3 cells were incubated with 5 vo l. low-salt buffer [50 m M Hepes/KOH pH 7.8, 50 m M KCl, 0.5 m M EDTA, 1m M dithiothreitol, 1% p rotease inhibitor c ocktail (Sigma, Irvine, UK)] for 10 min at 0 °C. After centrifugation at 250 g for 5 min at 4 °C, the cell pellet was suspended in 3 vol. low salt buffer and homogenized for 10 s trokes in a Dounce homo genizer. The nuclear fraction was c ollected by centrifugation at 550 g for5minat4°C, suspended in 500 lL h igh salt buffer ( 50 m M Hepes/KOH pH 7.8, 420 m M KCl, 0.1 m M EDTA, 1.5 m M MgCl 2 ,1m M dithiothreitol, 1% protease inhibitor cocktail, 20% g ly- cerol), then centrifuged again at 25 000 g for 30 min at 4 °C. The supernatant was used for electrophoretic mobility shift and footprint assays [25]. A BCA protein assay kit (Pierce, USA) was used for p rotein determination. For end- labeling, +4.4a DNA (4193–4483) was digested with MluI and then labeled with Klenow enzyme (TaKaRa), dGTP, and [a- 32 P] dCTP. The binding reaction was first carried out without labeled DNA in 17 lL binding buffer (10 m M Hepes/KOH pH 7.8, 40 m M KCl, 1 m M EDTA, 5 m M MgCl 2 ,0.12mgÆmL )1 poly(dI–dC) (Amersham Pharmacia Biotech, UK), 5 m M dithiothreitol, 1% protease inhibitor cocktail, 10% glycerol) for 10 min at room temperature; Fig. 3. Dot matrix comparison of nucleotide sequences from intergenic r egions between B29/Ig-b and GH g enes in ra t and hu mans. Each short line represents a regio n with at least 80% identity in 15 contiguous nucleotides between th e two sequences. Lo cations of conserved regions (A to G) are indicated in m atrices and at the botto m. Positions of DHS are s hown by reverse closed triangles. 1230 A. Komatsu et al. (Eur. J. Biochem. 269) Ó FEBS 2002 4000 c.p.m labeled DNA was added and the reaction continued for 30 min a t room temperature. Electrophoresis was performed on a polyacrylamide gel (4.5% acrylamide, 0.15% Bis-acrylamide in 6.7 m M Tris/HCl pH 7.5, 3.3 m M sodium acetate, 1 m M EDTA). Footprint analysis Nuclear e xtract (26 lg protein in 10 lL) was i ncubated with 5 · 10 4 c.p.m labeled DNA in 60 lL binding buffer for 30 min at 25 °C. To this was added 5 lL DNase I for 1-min incubation at 25 °C. The reaction was t erminated with 100 lLstopsolution(100m M Tris/HCl pH 7.5, 100 m M NaCl, 0.5% SDS, 10 m M EDTA, 50 lgÆmL )1 salmon sperm DNA). The DNA was extracted with phenol, precipitated with ethan ol, and analyze d on a 5% polyacryl- amide sequencing gel. RESULTS Y3 cell-specific DNase I hypersensitive sites B29/Ig-b mRNA was detected in rat plasmacytoma-derived Y3 cells by Northern hybridization with a rat B29/Ig-b cDNA probe (Fig. 1A). The features of the chromatin structure in the B29/Ig-b gene and its flanking 15-kb upstream and 31-kb downstream regions were then exam- ined by locating the DHS. In the upstream region, DHS at the promoter and at )11.2 kb and )8.6 kb inside the sodium channel g ene were fo und (Fig. 1B). Another site at +0.7 kb was located between exons 1 and 2 in the B29/Ig-b gene, four sites (+4.4, +6.0, +8.7, +11.2) were present between the B29/Ig-b and GH genes (Fig. 2), and two sites were found at +15.1 kb and +20.8 kb, between the GH and TCAM-1 genes (data not shown). I n t he B29/Ig- b nonproducing B RL and the GH-pro ducing GC cells, DHS at )11.2 kb and +11.2 kb were observed [7]. Sites at +15.1 kb and +20.8 kb are also present in GC cells [7]. Six DHS ()8.6, promoter, +0.7, +4.4, +6.0, +8.7) were found to be Y3 cell-specific. Relation of Y3 cell-specific DHS to conserved regions Regions including cell type-specific DHS have been found to correspond not only to promoters and enhancers, but also to the locus control regions (LCR) [21]. These regulatory regions are conserved in mammals such as mouse and humans [21,26–28]. To investigate this possibil- ity, intergenic nucleotide sequences between the B29/Ig-b and GH genes were compared in r at and humans (Fig. 3). The following seven c onserved regions were detected: +4.4a, +4.6b, +6 .0c, +6.5d, +8.7e, +9.7f, and +12.3 g. The +12.3 g region is the GH promoter. Conserved region size and nucleotide sequence identity were clarified as follows: +4.4a 80% 0 .22 kb, +4.6b 70% 0.20 kb, +6.0c Fig. 4. Activity of transiently introduced DHS or conserved regions in Y3 cells. Values are means from triplicate determinations (± SE) . B29-P, B29/Ig-b promoter. ., Positions of DHS. Fig. 5. Orientation dependency ( A) and promoter specificity (B) of the regions containing enhancing activity. (A) Enhancing activity of nor- mally oriented and r everse-orientation 2.9-kb and +4.4a DNA. (B) Promoter-specificity of 2.9-kb DNA and + 4.4a plus +4.6b DNA. SV-P, SV-40 promoter; SV-P+E, SV-40 promoter plus enhancer; B29-P, B29/Ig-b promoter. Luciferase activity indicated by arbitrary units. Ó FEBS 2002 Rat B29/Ig-b gene regulatory regions (Eur. J. Biochem. 269) 1231 83% 0.23 kb, +6.5d 77% 0.18 kb. Identity in +8.7e and +9.7f regions was less than in +4.4a to +6.5d. Three Y3 cell-specific DHS (+4.4, +6.0, and +8.7 kb) in the B29/ GH intergenic region were mapped inside the corresponding conserved regions +4.4a, +6.0c, and +8.7e (Fig. 3); in the +4.4a to +12.3 g region, the identity of +6.0c was highest and that of +4.4a was second highest. Transcriptional enhancing activity of the DHS Regions including cell type-specific DHS often s how transcriptional enhancing activity [19,20] and t his activity was sought for the present s tudy (Fig. 4). The Y3 cell- specific DHS in and around the B29/Ig-b gene were inserted upstream in the B29/Ig-b promoter plasmid (B29-P)with the firefly luciferase reporter gene. Recombinant plasmids were transiently transfected into Y3 cells and the luciferase activity was measured. The reporter plasmid with the B29/ Ig-b promoter ()201 to +30) possessed 21 times more luciferase activity than the pro moterless control. No signifi- cant activity was found in the 0.8-kb DNA prepared from the Y3 cell-specific )8.6-kb site or the 0.27-kb fragment from the +0.7-kb region. The 2.9-kb DNA from +3.8 kb to +6.7 kb containing the +4.4a to +6.5d regions showed six t imes more enhancing activity than the B29/Ig-b promoter alone. In the +4.4a to +6.5d regions, the highly conserved +4.4a region with one of the Y3 cell-specific DHS s howed the greatest luciferase a ctivity; th e +6.0c region showed the second greatest. Two less conserved but non-DNase I hypersensitive regions, +4.6b and + 6.5d, had n o a ctivity. A 1.1-kb fragment from the +8.7-kb region was two times more active than the B29/Ig-b promoter, although its enhancing activity was less than that for the +6.0c region alone. Experiments were c on ducted to determine whether regions containing transcriptional enhancing activity are orientation- or promoter-dependent (Fig. 5). Reverse- orientation 2.9-kb DNA containing the +4.4a to +6.5d regions displayed even more enhancing activity than normally oriented DNA. Reverse-orientation +4.4a DNA was 2.5 time s more active than the B29/Ig-b pro moter alone although its activity was less than that of normally oriented DNA. Thus, the enhancing activities of the 2.9-kb fragment and +4.4a region were orientation-independent, indicating that these regions are enhancers. When the 2.9-kb region wasinsertedupstreamwiththeSV-40 promoter reporter and t ransfected into Y3 cells, t hree times greater enhancing activity resulted compared with the activity of the SV-40 promoter alone. The region containing +4.4a to +4.6b showed twice the activity of the SV-P control and essentially the same as that of the SV-40 enhancer SV-P+E. The 2.9-kb fragment and +4.4a region thus express enhancing activity toward heterologous promoters such as the SV-40 promoter, a lthough t hey express more toward the B29/Ig-b promoter. Electrophoresis mobility shift assay, footprinting, and transfection analysis of the +4.4a region The +4.4a region containing one of the Y3 cell-specific DHS (Fig. 2B) was highly conserved in rat and humans (Figs 3 and 6) and had the highest enhancing activity o f the conserved +4.4a to +8.7e regions (Fig. 4), so it was examined in greater detail. With both the Y3 nuclear e xtract and the 0.3-kb DN A with t he +4.4a region, a major band shift was noted in the electrophoretic mobility shift assay (Fig. 7); this shift disappeared upon competition with unlabeled 0.3-kb DNA, indicating that this region has binding sites for Y3 nuclear protein. To determine these binding sites, 0.3-kb DNA was split into 0.2- and 0.1-kb fragments by HinfI digestion and the se fragments were used for competition. The 0.2-kb DNA outcompeted the major band (Fig. 7), indicating that th e b inding site for the major shift is present in this region. Fig. 6. Nucleotide sequence comparison of the conserved + 4.4a region in rat and humans. Upper, rat; lo wer, humans. Nucleotide numbers from transcriptional start site o f the B29/Ig-b gene are indicated. Identical nucleotides are shown by asterisks (*). The conserved region i s enclosed with lines. Potential binding sites o f the transcription factors are boxed. 1232 A. Komatsu et al. (Eur. J. Biochem. 269) Ó FEBS 2002 To further deline ate the major binding site, f ootprint analysis was carried out. Each 3¢ end of the 0.3-kb DNA with the + 4.4a region was labeled with 32 P for footprinting (Fig. 8). Nucleotides 4320 to 4339 and 4337 to 4320 from the 5¢ end of the B29/Ig- b gene were protected in forward and reverse strands (Fig. 8). In the protected regions, a sequence was found corresponding to the binding site for the OCT family transcription factor [29]. Presumed S RY binding sites GATA and NF-kB were present in the conserved +4.4a region (Fig. 6) but they were not pro- tected. To determine whether the footprinted region is responsible for the major electrophoretic mobility shift activity, a 30-bp fragment from nucleotides 4313 t o 4342 that include the p rotected region was u sed for competition (Fig. 7). This fragment outcompeted the major band, whereas the fragment from 4403 to 4432 did not, suggesting that the footprinted region is r esponsible for the major mobility shift activity. Deletion clones were produced to further examine the relation between the region r esponsible for transcriptional enhancing activity and the footprinted region o f the 0.3-kb DNA (Fig. 9). Although Pyrobest-DNA polymerase has high fidelity, amplification by the polymerase chain reaction method over the 5-kb region was performed to obtain t he deletion clones, and the activity of each of three constructs was separately determined. Two different activity levels were determined. Insertion of the 0.3-kb fragment upstream from B29-P resulted in 4.5 times more luciferase activity than that of the B29-P promoter alone. The 30-bp deletion (4313 to 4 342) constructs reduced luciferase activity to that of the B29-P promoter alone. The 30-bp deletions (4193 to 4222) unrelated to the footprinted region had activity similar to that of the control. The sequence responsible for the major mobility shift and the transc riptional enhancing activity in the 0 .3-kb DNA with the conserved +4.4a region thus coincided with the 20-bp footprinted region in which the s equence c orresponding to the consensus binding site for the OCT family [29] was present. DISCUSSION To understand t he mechanism for B-cell differentiation, the mechanism of gene expression specific for B lymphocytes must be understood. A component of the mIg receptor complex, B29/Ig-b is an essential molecule for BCR- mediated signal transduction [1,2] and is expressed from early in B-cell development to the p lasma cell stage [10–12]. The mechanism for B-cell-specific expression of the B29/Ig-b gene has b een studied by analyzing cis -elements located as far as the 1.2-kb upstream region and their interac- ting transcription f actors. During B -cell development, B-cell-specific gene expression seems to require not only Fig. 7. Binding of Y3 nuclear proteins to +4.4a DNA. 32 P-Labeled 0.3-kb DNA containing the +4.4a region was incubated for 3 0 m in at room temperature with Y3 nuclear extract (0.7 lg). Competition reactions were performed using full 0.3-kb DNA, a HinfI fragment (nucleotides 4193–4397), footprinted fragment A (nucleotides 4313– 4397), and control fragment B (nucleotides 4403–4432). The amount of unlabeled competitor is indicated as fold mo lar excess. Nucleotide numbers from transcriptional start site of the B29/Ig-b gene are shown. Fig. 8. DNase I footprint analysis of the +4.4a region. The 0.3-kb DNA (nucleotides 4193–4483) containing the +4.4a region labeled with 32 P was u sed. DNA binding reactions were incubated for 30 min at 25 °C with Y3 nuclear extract. F, coding strand; R, noncoding strand. Footprinted regions are indicated in margins. At bottom, features of the 0.3-kb DNA and nucleotide sequence of the footprinted region are s hown. N ucleotides 4245–4461 are the conserved region. Nucleotide numbers from transcriptional start s ite of the B29/Ig-b gene are indicated. Ó FEBS 2002 Rat B29/Ig-b gene regulatory regions (Eur. J. Biochem. 269) 1233 recruitment of transcription factors and adaptors pre- requisite for gene expression, but also change in the chromatin structure from an inactive st ate t o an active state. Cis-elements for transcription factors or the regions required f or the structural change of the chromatin are often found far upstream or downstream in the gene [21,22] as well as in pro moter regions. In this study, cell type-specific DHS in and around the B29/Ig-b gene were used as landmarks to search for novel regulatory regions. Three regions having transcriptional enhancing a ctivity w ere found in the intergenic region between the B29/Ig-b and GH genes. From the results of footprinting and deletion mutant analysis, a member of the OCT family [29] appeared to be involved in the transactivation of the + 4.4a region, which had the highest enhancing activity of the three conserved DHS. The binding site for the OCT family is present in the promoter region of mouse, human, and rat B29/Ig-b genes [3,6,14] and is essential for B-cell-specific mouse g ene e xpression [14]. The OCT family binding site is conserved in the human sequence corresponding t o the rat +4.4a region (Fig. 6), s uggesting that t he OCT family may also be important for human CD79b (B29/Ig-b)gene expression. Two OCT proteins, Oct-1 and Oct-2, are known to be expressed in B cells. In c ontrast to ubiquitously expressed Oct-1, Oc t-2 is specifically expressed in B lineage cells during all stages of development [30,31]. Y3 cells are plasmacyto- mas and thus Oct-2 with or without Oct-1 is likely to be involved in transactivation. No significant change in the expression of th e B29/Ig-b gene in pre-B cells derived from the O ct-2 null mice from that in corresponding wild type cells has been reported, indicating that Oct-2 is not required for B29/Ig-b gene expression in these cells [32]. Thus, t he actual participation of Oct-2 in B29/Ig-b gene expression during B-cell development remains obscure. Three intergenic DHS corresponding +4.4a, +6.0c, and +8.7e were each found to possess enhancing a ctivity when inserted into the B29 promoter reporter and transfected into Y3 cells (Fig. 4). Because 2.9-kb DNA containing +4.4a to +6.5d regions exhibited higher enhancing activity than e ach of the three regions individually, the +4.4a to +6.5d regions along with +8.7e s hould be involved in B29/Ig-b gene expression. Neither the 2.9-kb DNA nor any of the three regions individually had any enhancing/silencing activity when combined with the GH promoter reporter and transfected into GH-producing G C cells (data n ot shown) and thus are regulatory regions for B29/Ig-b gene expression. The LCR often coincides with a cluster of cell type-specific DHS and some DHS in the L CR po ssess transcriptional enhancing a ctivity [21,22]. Typical LCRs such as human b-glo bin, human CD2, and mouse k5/ VpreB1 loci have these features [21]. Four cell type-specific DHS that comprise the human b-globin LCR are present in the i ntergenic region between an odorant receptor gene cluster and the b-globin locus [33] and their nucleotide sequences are conserved in mouse and humans [27]. The features of three intergenic DHS demonstrated in the present study correspond well to those of the LCR, so these DHS likely form the LCR of the B29/Ig-b gene, although further examination using transgenic or knockout mice should be conducted to confirm this point. Mammalian g enes are often separated by long intergenic regions in which regulatory regions for transcription are scattered, thus making it difficult to find them. As shown this and previous studies [26,28], novel transcriptional r egulatory regions can be sought by investigating cell type-specific DHS in the intergenic region and mapping within conserved regions in appropriate species. With progress in genome projects, human and mouse g enome sequences are now easy to use, and the above method should p rove useful for finding regulatory regions in long intergen ic regions and introns. Well studied in chicken b-globin [34] and lysozyme [35] loci, changes in the chromatin s tructure of vertebrate s occur from condensed to relaxed forms in and a round many genes whose expression is regulated developmentally and cell type- specifically. Changes in chromatin structure occur not only in the gen e itself but also far upstream and downstream. Chromatin remodeling may be followed by acetylation of core histones [36,37]. In contrast, before cell type-specific gene expression, no change in chromatin structure has been reported; the human a-globin locus that evolved together with the b-globin locus from a common ancestral gene is always present in constitutively relaxed form [38]. To elucidate the mechanism for B-cell-specific gene expression of the B29/Ig-b gene, whether changes occur in chromatin structure b efore gene expression should be determined. The rat B29/GH locus is r ich in gene s with s hort intergenic regions and thus the c hromatin structure is not relaxed i n and around the expressed gene; it is condensed in a repressed gene during the expression of cell type-specific genes. Yet all regions of this locus are reportedly always relaxed in the human a-globin locus [38]. Thus, before cell type-specific gene expression in this locus, the chromatin structure o f a certain region including the particular gene with some flanking genes may change. Recently, enhance- ment of H3 and H4 histone acetylation was observed from the promoter region of human GH gene to the upstream region present in the sodium channel gene in the chromatin of GH-producing cells [39]. In B29/Ig-b-producing cells, whether enhancement of core h istone acetylation along with Fig. 9. Enhancing activity o f deletion constructs of the f ootprinted region. The 30-b p sequ en ce includ in g t he footprin ted r egion (nucleo- tides 4313–4342) deleted f rom the reporter construct having both the +4.4a region (nucleotides 4193–4474) and B29-P. Nucleotide numbers from transcriptional start site of the B29/Ig-b gene are shown. Deletion from nucleotides 4193–4222 was used for comparison. 1234 A. Komatsu et al. (Eur. J. Biochem. 269) Ó FEBS 2002 the general sensitivity to DNase I is present not only in th e B29/Ig-b gene itself but also in flanking intergenic regions and the sodium chan nel and GH genes is a point of interest. ACKNOWLEDGEMENTS We th ank M. Karin, the Japanese Cancer Research Resources Bank, and RIK EN Cell Bank for p roviding the cells. This w ork w as supported by a grant from the Foundation of Growth Sciences and b y Rikkyo University f or the Promotion of Research. REFERENCES 1. Reth, M. & Wienands, J . (1997) I nitiation and processing of sig- nals from th e B cell antigen receptor. Annu. Rev. Immunol. 15, 453–479. 2. Wienands, J. (2000) The B-cell antigen receptor: formation of signaling complexes and the function of adaptor proteins. Curr. Top. Microbiol. Immunol. 245, 53–76. 3. Nakazato, S., Nomoto, K., Kazahari, K. & Ono, M. (1998) Physical linkage of the B29/Ig-beta (CD79B) g ene to t he skeletal muscle, sodium-channel, and growth h ormone genes in rat and human. Genomics 48, 363–368. 4. Hashimoto, S., Gregersen, P.K. & Chiorazzi, N. (1993) The human Ig-b eta cDN A sequ ence, a homologue of murine B29, is identical in B cell and plasma cell lines producing all the hum an Ig isotypes. J. Immunol. 150, 491–498. 5. Reth, M. (1989) Antigen receptor tail clue. Nature 338, 3 83–384. 6.Hashimoto,S.,Chiorazzi,N.&Gregersen,P.K.(1994)The complete sequence of the human CD79b (Ig beta/B29) gene: identification of a conserved e xo n/intron o rganization, i mmuno- globulin-like r egulatory regions, a nd allelic polymorphism. Immunogenetics 40, 145–149. 7. Kazahari, K., Sano, E., Matsuura, N. & Ono, M. (1997) Chro- matin structure of the rat somatotropin gen e locus and physical linkage of the rat somatotropin gene and skeletal-muscle sodium- channel gene. Eur. J. Biochem. 244, 494 –500. 8. Ono, M., Nomoto, K. & Nakazato, S. (1999) Gene structure of rat testicular cell adhesion molecule 1 (TCAM-1) and its physical linkage to genes c oding for the growth hormone and B AF60b, a component of SWI/SNF complexes. Gene 22 6, 95–102. 9. Bennani-Baiti, I.M., Cooke, N.E. & Liebhaber, S.A. (1998) Phy- sical linkage of the h uman growth ho rmone g ene cluster and the CD79b (Ig beta/B29) gene. Genomics 48, 258–264. 10. Bain, G., Maandag, E.C., Izon, D.J., Amsen, D., Kruisbeek, A.M., Weintraub, B.C., Krop, I., Schlissel, M.S., Feeney, A.J., van Roon, M., van der Valk, M., te Riele, H.P.J., Berns, A. & Murre, C. (1994) E2A proteins are required for proper B cell development and initiation of immunoglobulin gene rearrange- ments. Cell 79, 885–892. 11. Lin, H. & Grosschedl, R . (1995) Failure of B-cell differentiation in mice lacking the transcription factor EBF. Nature 376, 263–267. 12. Urbanek, P., Wang, Z.Q., Fetka, I., Wagner, E.F. & Busslinger, M. (1994) Complete block of early B cell differentiation and altered patterning of the posterior midbrain in mice lacking Pax5/ BSAP. Cell 79 , 901–912. 13. Gong, S. & Nussenzweig, M.C. (1996) Regulation o f an early developmental checkpoint in the B cell pathway by Ig beta. Science 272, 411–414. 14. Hermanson, G.G., Briskin, M., Sigman, D. & W all, R. (1989) Immunoglobulin enhancer and promoter motifs 5¢ of the B29 B-cell-specific gene. Proc. Natl Acad. Sci. USA 86, 7341–7345. 15. Thompson, A.A., W ood, W.J. Jr, Gilly, M.J., Damore, M.A., Omori, S.A. & Wall, R. (1996) The promoter and 5¢ flanking sequences controlling human B29 gene expression. Blood 87, 666–673. 16. Omori, S.A. & Wall, R. (1993) Multiple motifs regulate the B-cell- specific promoter of the B29 gene. Proc. Natl Acad. Sci. USA 90, 11723–11727. 17. Malone, C.S., Omori, S.A. & Wall, R. (1997) Silencer elements controlling t he B29 (Ig beta) promoter are neither pro- moter- nor c ell-type-specific. Proc. Natl Acad. Sc i. USA 94 , 12314–12319. 18. Akerblad, P., Rosberg, M., Leanderson, T. & Sigvardsson, M. (1999) The B29 (immunoglobulin beta-chain) gene is a genetic target for e arly B-cell factor. Mol. Cell. Biol. 19, 392 –401. 19. Elgin, S .C. (1988) The formation a nd function of DNase I hypersensitive s ites in the process of gene activation. J. Biol. Chem. 263, 19259–19262. 20. Gross, D.S. & Garrard, W.T. (1988) Nuclease hypersensitive sites in chromatin. An nu. Rev. Biochem. 57, 159–197. 21. Dillon, N. & Sabbattini, P. (2000) Functional gene expression domains: defining the functional unit of eukaryotic gene regula- tion. Bioessays 22 , 657–665. 22. Fraser, P. & Grosveld, F. (1998) Locus control regions, chromatin activation and tra nscription. Curr. Opin. Cell Biol. 10 , 361–365. 23. Ono, M., Mochizuki, E., Mori, Y., Aizawa, A. & Harigai, T. (1995) The regulatory region and transcription factor required for the expression of rat and salmon pituitary hormone-encoding genes show cell-type an d species specificity. Gene 153, 267–271. 24. Aizawa, A., Yoneyama, T., Kazahari, K. & Ono, M. (1995) DNase I -hypersensitive sites in the chromatin of rat growth hor- mone gene locus and enhancer activity of regions with these sites. Nucleic Acids R es. 23, 2236–2244. 25. Dignam, J.D., Lebovitz, R.M. & Roeder, R.G. (1983) Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11, 1475– 1489. 26. Loots, G.G., Locksley, R.M., Blankespoor, C.M., Wang, Z.E., Miller, W ., Rubin, E.M. & Frazer, K.A. (2000) Identification of a coordinate regulator of interleukins 4 , 13, and 5 by c ross-species sequence comparisons. Science 288, 136–140. 27. Hardison, R.C. , Oeltjen, J. & Mi ller, W. (1997) Long human-mouse sequence alignments reveal no vel regulatory elements: a reason to sequence the mouse genome. Genome Res. 7, 959–966. 28. Hardison, R.C. (2000) Conserved noncoding seque nces are reli- able guides to regulatory elements. Trends Genet. 16, 369–372. 29. Matthias, P. (1998) Lymphoid-specific tran scription mediated b y the c onserved octamer site: who is d oin g what? Semin. Immunol. 10, 155–163. 30. Miller, C.L., Feldhaus, A.L., Rooney, J.W., Rhodes, L.D., Sibley, C.H. & Singh, H. (1991) R egulation a nd a possible stage-specific function of Oct- 2 during pre-B-cell differentiation. Mol. Cell. Biol. 11, 4885–4894. 31. Staudt, L.M., Clerc, R.G., Singh, H., Lebowitz, J.H., Sharp, P.A. & Baltimore, D. (1988) Cloning of a lymphoid-specific cDNA encoding a protein binding the regulatory octamer DNA motif. Science 241, 577–580. 32. Corcoran, L.M., Karvelas, M., Nossal, G.J., Ye, Z.S., Jacks, T. & Baltimore, D. (1993) Oct-2, although not required for early B-cell development, is critical for later B-cell maturation and for post- natal survival. Genes Dev. 7, 570–582. 33. Bulger, M., van Doorninck, J.H., Saitoh, N ., Telling, A., Farrell, C., Bender, M.A., Felsenfeld, G ., Axel, R ., Groudine, M . & von Doorninck, J.H. (1999) Conservation of sequence and structure flanking the mouse and human beta-globin loci: the beta-globin genes are embedded within an array of odorant receptor genes. Proc. Natl Acad. Sci. USA 96, 5129–5134. 34. Hebbes, T.R., Clayton, A.L., Thorne, A.W. & Crane-Robinson, C. (1994) Core histone hyperacetylation co-maps with generalized Ó FEBS 2002 Rat B29/Ig-b gene regulatory regions (Eur. J. Biochem. 269) 1235 DNase I sensitivity in the chicken beta-globin chromosomal domain. EMBO J. 13 , 1823–1830. 35. Bonifer, C., Jagle, U. & Huber, M.C. (1997) The chicken lysozyme locus as a paradigm for the complex developmental regulation of eukaryotic gene loci. J. Biol. Chem. 272, 26075–26078. 36. Krebs, J. & Peterson, C. (2000) Understanding ÔactiveÕ chromatin: a h istorical perspective of chromatin r emod eling. Crit. R ev. Euk- aryo. Gene Express. 10, 1–12. 37. Turner, B.M. (2000) Histone acetylation and a n epigenetic c ode. Bioessays 22, 836–845. 38. Higgs, D., Sharpe, J. & Wood, W. (1998) Understanding a globin gene expression: a step towards e ffective gene therapy. Sem. Hematol. 35, 93–104. 39. Elefant, F., Cooke, N.E. & Liebhaber, S.A. (2000) Targeted recruitment o f h istone acetyltransferase activity to a locus control region. J. Biol. C hem. 275, 138 27–13834. 1236 A. Komatsu et al. (Eur. J. Biochem. 269) Ó FEBS 2002 . [27]. The features of three intergenic DHS demonstrated in the present study correspond well to those of the LCR, so these DHS likely form the LCR of the B29/Ig-b. plasmacytoma-derived Y3 cells by Northern hybridization with a rat B29/Ig-b cDNA probe (Fig. 1A). The features of the chromatin structure in the B29/Ig-b gene and