D N A Sequence-].DNA Sequencing and Mapping, Vol. 1, pp. 409-41 3 Reprints available directly from the publisher Photocoping permitted by license only
0 1991 Harwood Academic Publishers GrnbH Printed in the United Kingdom
SHORT COMMUNICATION
Mitochondrial DNA Downloaded from informahealthcare.com by Nanyang Technological University on 08/20/15 For personal use only.
A novel divergently transcribed human histone H2A/H2B gene pair THOMAS DOBNER,* INGRID WOLF, BERNARD MA1 and MARTIN LlPP lnstitut fur Biochernie, Ludwig-Maximilians-Universifat, 0-8000 Munchen 2, Germany
EMBL Data Library Accession No. X57138
A genomic clone containing a novel closely linked human histone H2A/H2B gene pair has been isolated and sequenced along with extensive 5’ and 3’ flanking regions. Both genes are devoid of introns and code for core histone proteins. The nucleotide sequences are 84% and 87% homologous to the coding regions of a human genomic H2A and H2B gene, respectively. A comparison of the nucleotide-derived amino acid sequences shows that the histone H2A protein corresponds to the human H2A.1 subtype, whereas the H2B histone gene predicts an H2B protein sequence which i s almost identical to the histone H2B.2 variant from human and bovine obtained by direct protein sequencing. The 3’ flanking regions contain previously identified conserved sequence elements thought to be involved in transcription termination and processing of replication-dependent histone gene poly(A)- mRNAs. Primer extension analyses of the histone mRNAs encoded within this clone demonstrate that both genes are divergently transcribed from a 313 bp intergene promoter region. The spatial arrangement and orientation of two TATA-boxes, four CAAT-boxes, and one HZB-box within this region suggests that the linked genes share common promoter elements for transcriptional regulation.
an apparently random gene order (reviewed by Stein eta/., 1984). However, a common motif i s a tight linkage of H2A/H2B and H3/H4 genes (Hentschel and Birnstiel, 19811. Functional studies of H2A/H2B gene pairs mainly from chicken have shown that these genes are often divergently transcribed (Harvey et a/., 1982). The detailed analysis of the corresponding intergenic promoter regions has revealed several highly conserved regulatory elements required for the co-ordinated expression for both genes (Sturm et a/., 1988). Despite extensive knowledge of the promoter organization and expression of H2A/H2B gene pairs from many species, the particulars of the fine structure and transcriptional regulation of the homologous genes in humans still remain to be elucidated. This lack of information encouraged us to isolate the corresponding genes from a human genomic library. These experiments should serve as a basis for detailed studies of the transcriptional regulation of divergent transcribed human H2A/H2B histone genes. In this communication we report the complete nucleotide sequence of a novel tightly linked human histone H2A/H2B gene pair. Both genes are separated by a 313 bp intergene region and code for replication-dependent core histone proteins which are identical or closely related to the human H2A.1 and H2B.2 variants. In addition, we have mapped the 5‘ ends of both transcripts encoded by these genes. These data as well as the arrangement of conserved promoter elements within the intergene region suggests that transcription of the H2A/H2B gene pair i s regulated by an overlapping intergenic promoter which shares common regulatory elements.
KEY WORDS: divergent transcription, H2A/H2B gene pair, human histone genes, intergene promoter, multigene family
The chromosomal D N A of all eukaryotes is complex by evolutionary highly conserved basic proteins known as histones. The five major types of histone proteins H1, H2A, H2B, H3, and H4, are encoded by ten to several hundred copies of genes per haploid genome in all species examined. Furthermore in higher vertebrates several non-allelic variants of each histone protein exist, which differ in their primary structures and their expression patterns. Analysis of the genomic organization of the multigene family i n higher eukaryotes has revealed
*Present address: Department of Molecular Biology, Lewis Thomas Laboratory, Princeton, New Jersey08544, USA.
409
410
A
T . DOBNER E J A L .
H lhb
B
H, X
Sf
s
S 3 H !,t
pK.3 B Xm I
[
:,",,
Sm
I
--
hi0114
H&&p H P2 X
plOll4BX
Mitochondrial DNA Downloaded from informahealthcare.com by Nanyang Technological University on 08/20/15 For personal use only.
t--+--ipBMl
pBM213 7-
pBM4
t-------cC--pBMSH1 (
Figure 1 Cloning and characterization of the H2A.1 and H2B.2 gcwe. T-he restriction maps show only the location of the restriction cnzyrncs usid for generation of subclones. Abbreviations of used restrii tiori enzymes: B BamHI. H: Hindlll. Pv2: Pvull. S: Sall. Sm: SmA. St: Stul. X : Xbal Xm: Xmnl. (A) Re.striction map of A10/14. Subciones pK3 and plO/l4BX are shown beneath the restriction map. LA and RA indicate the left and right arm of phage vector FMBl3A. Methods: 5x1 0' recombinant phages were screened with the rcindom-primed "P-labelled H2A.1 cDNA probe as described by Benton and Davis (1977). Positive phages were plaque-purified by thiree successive platings. Phage DNA was isolated from mini-lysates, characterized by restriction enzyme analysis, and Southern blotting as described by Maniatis et a/. (1982). A 0.7 kb Hindll/Stul (clone pK3) and a 1.5 kb BamHVXbal iragnient (clone plO/14BX) were isolated and subcloned in p U C l 2 according to standard protocols (Maniatis et a/., 1982). (B) Subciuning and sequencing strategy. Arrows indicate the orientation and length of the sequenced DNA regions within the suhclones. The positicins and orientations of the H2A and H2B genes are indicated by large arrows on the map. Methods: After restriction enzyme analysis of pl0/14BX restriction fragments were tiirthcr subcloncd into pUC vectors using the appropriate enzyme sites. Plasrnid DNAs from pK3, pBM1, pBM2/3, pBM4 and pHh.1SW1 were isolated from mini-preparations purified over C2iagt.n tip-20 columns, and the DNA sequences were determined twice on both strands by the dideoxy chain termination method ih'ith r: DNA Polymerase Sequencing Kit (Pharmacia). Products were analyzed on 6% polyacrylamide sequencing gels and tietc,rtd b y rxxposure to X-ray film (Kodak XAR-5).
A genomic EME!L3A library from the human Burkitt's lymphoma cell line BL64. (Hart1 and Lipp, 1987) was screened with a previously identified human ti2A.1 (EDNA probe (Dobner et a/., submitted). Homologous genomic clones were selected using high stringency conditions for hybridization and post-hybridization washes. Among .5 positivle clones one clone, designated 21 0/14, was selected for further detailed analysis. The characterization of d10/14 and the subsequent cloning of H2A and H2B-containing genomic D N A fragments in plasmid vectors for D N A sequence analysis i s summarized in Fig. 1. The nucleotide
sequence has been submitted to the EMBL Data Library (accession number X57138). The overlapping subclones, pK3 and p10/1 4BX, contain the complete 1661 bp nucleotide sequence of the H2A.1 cDNA probe and a novel H2B gene along with extensive 3' flanking regions. Both genes are arranged in a divergent orientation and are separated by a 313 bp intergene region (Fig. 2). The histone H2A and H2B sequences encode a 129 and 125 amino acid histone protein with a deduced molecular weight of 14093 and 13909, respectively. The nucleotide sequences are 82% (H2A) and 87% (H2B) homologous to the coding regions of a previously identified human genomic H2A and H2B gene (Zhong e t a / . , 1983). The predicted H2A protein corresponds to the human histone H2A.1 variant obtained by direct protein sequencing (Hayashi eta/., 1988), whereas the H2B protein is almost identical to the amino acid sequence of the human H2B.2 subtype (Ohe et a/., 1979), showing a 97% homology or 3 conservative amino acid changes as indicated in Fig. 2. Finally, the 3' flanking regions of the H2A #and H2B gene contain classical regions of dyad symmetry and a conserved downstream purine-rich element. Both elements are required for termination of transcription and 3' end processing of nonpolyadenylated replication-dependent histone premRNAs (Mowry eta/., 1989). In order to prove that the H2A/H2B gene pair is divergently transcribed from the intergene region we have mapped the 5' ends of both histone mRNAs. Figure 3 shows the results of primer extension analyses which demonstrate that the major start of the H2B.2 mRNA i s at the A residue (position 916, Fig. 2), whereas the H2A.1 mRNA starts at position 1056 (Fig. 2) at A. These results are consistent with other start sites of H2A and H2B mRNAs from chicken (Harvey et a/., 1982) and human (Zhong et a / . , 1983) published previous I y . The mapped H2A and H2B mRNA CAP sites are located approximately 30 bp downstream from the TATA-boxes (Fig. 4). Beside characteristic promoter elements of eukaryotic transcription units (TATA and CAAT-boxes) this divergent transcribed promoter has particularly interesting features. Remarkably the H2B subtype-specific element (H2B-box) which is normally found very close r6 to 30 bp) upstream of the TATA-box of almost all H2B genes (Harvey et a/., 1982; Wells and McBride, 1989) is located 112 bp upstream of the TATA-box
41 1
HUMAN HISTONE H2NH2B GENE PAIR
I I 1 I I I I I I 1 I I CCCGGGAGACGGAGCTTACAGTGAGCCGAGATCGCGGCACTGCACTCTATCCTGGGCGACAGAGCCAAGACTCCATTTAAAAAAAAAAAAAAAAAAAACACAAAAGGAAAGAGTAACCGA GGGCCCTCTGCCTCGAATGTCACTCGGCTCTAGCGCCGTGACGTGAGATAGGACCCGCTGTCTCGGTTCTGAGGTAAATTTTTTTTTTTTTTTTTTTTGTGTTTTCCTTTCTCATTGGCT
120
AAACCGAAAGCAAACTTGTAATTATAGCTTAAAGCACAACTGCAAGCCTTTTTCCCCAAGGGTATCTAAGAACCCAAGCTAAATTTGTCTGATGGAGCGTCTATTCTCACACGCAAAAAG 240 TTTGGCTTTCGTTTGAACATTAATATCGAATTTCGTGTTGACGTTCGGAAAAAGGGGTTCCCATAGATTCTTGGGTTCGATTTAAACAGACTACCTCGCAGATAAGAGTGTGCGTTTTTC CCTAGGGAGCTTAGCTTACCTTTTCCCGCCGTCTTCCCTCCTAATGGAGGCCGGCCGGAGTCCACCCAAAAGACTACAAGCAAAATGCACAGGCTCTTTCTCTGGAAGCTGGAGTGGCTC GGATCCCTCGAATCGAATGGAAAAGGGCGGCAGAAGGGAGGATTACCTCCGGCCGGCCTCAGGTGGGTTTTCTGATGTTCGTTTTACGTGTCCGAGAAAGAGACCTTCGACCTCACCGAG
TGAAAAGAGCCTTTGGATTGCTGGCAGCTGCGAGAGCTCACTTGGAGCTGGTGTACTTGGTGACGGCCTTTGTGCCCTCGGACACGGCGTGCTTGGCCAGCTCCCCGGGCAGCAGCAGGC ACTTTTCTCGGAAACCTAACGACCGTCGACGCTCTCGAGTGAACCTCGACCACATGAACCACTGCCGGAAACACGGG~GCCTGTGCCGCACGAACCGGTCGAGGGGCCCGTCGTCGTCCG
0 1991 Harwood Academic Publishers GrnbH Printed in the United Kingdom
SHORT COMMUNICATION
Mitochondrial DNA Downloaded from informahealthcare.com by Nanyang Technological University on 08/20/15 For personal use only.
A novel divergently transcribed human histone H2A/H2B gene pair THOMAS DOBNER,* INGRID WOLF, BERNARD MA1 and MARTIN LlPP lnstitut fur Biochernie, Ludwig-Maximilians-Universifat, 0-8000 Munchen 2, Germany
EMBL Data Library Accession No. X57138
A genomic clone containing a novel closely linked human histone H2A/H2B gene pair has been isolated and sequenced along with extensive 5’ and 3’ flanking regions. Both genes are devoid of introns and code for core histone proteins. The nucleotide sequences are 84% and 87% homologous to the coding regions of a human genomic H2A and H2B gene, respectively. A comparison of the nucleotide-derived amino acid sequences shows that the histone H2A protein corresponds to the human H2A.1 subtype, whereas the H2B histone gene predicts an H2B protein sequence which i s almost identical to the histone H2B.2 variant from human and bovine obtained by direct protein sequencing. The 3’ flanking regions contain previously identified conserved sequence elements thought to be involved in transcription termination and processing of replication-dependent histone gene poly(A)- mRNAs. Primer extension analyses of the histone mRNAs encoded within this clone demonstrate that both genes are divergently transcribed from a 313 bp intergene promoter region. The spatial arrangement and orientation of two TATA-boxes, four CAAT-boxes, and one HZB-box within this region suggests that the linked genes share common promoter elements for transcriptional regulation.
an apparently random gene order (reviewed by Stein eta/., 1984). However, a common motif i s a tight linkage of H2A/H2B and H3/H4 genes (Hentschel and Birnstiel, 19811. Functional studies of H2A/H2B gene pairs mainly from chicken have shown that these genes are often divergently transcribed (Harvey et a/., 1982). The detailed analysis of the corresponding intergenic promoter regions has revealed several highly conserved regulatory elements required for the co-ordinated expression for both genes (Sturm et a/., 1988). Despite extensive knowledge of the promoter organization and expression of H2A/H2B gene pairs from many species, the particulars of the fine structure and transcriptional regulation of the homologous genes in humans still remain to be elucidated. This lack of information encouraged us to isolate the corresponding genes from a human genomic library. These experiments should serve as a basis for detailed studies of the transcriptional regulation of divergent transcribed human H2A/H2B histone genes. In this communication we report the complete nucleotide sequence of a novel tightly linked human histone H2A/H2B gene pair. Both genes are separated by a 313 bp intergene region and code for replication-dependent core histone proteins which are identical or closely related to the human H2A.1 and H2B.2 variants. In addition, we have mapped the 5‘ ends of both transcripts encoded by these genes. These data as well as the arrangement of conserved promoter elements within the intergene region suggests that transcription of the H2A/H2B gene pair i s regulated by an overlapping intergenic promoter which shares common regulatory elements.
KEY WORDS: divergent transcription, H2A/H2B gene pair, human histone genes, intergene promoter, multigene family
The chromosomal D N A of all eukaryotes is complex by evolutionary highly conserved basic proteins known as histones. The five major types of histone proteins H1, H2A, H2B, H3, and H4, are encoded by ten to several hundred copies of genes per haploid genome in all species examined. Furthermore in higher vertebrates several non-allelic variants of each histone protein exist, which differ in their primary structures and their expression patterns. Analysis of the genomic organization of the multigene family i n higher eukaryotes has revealed
*Present address: Department of Molecular Biology, Lewis Thomas Laboratory, Princeton, New Jersey08544, USA.
409
410
A
T . DOBNER E J A L .
H lhb
B
H, X
Sf
s
S 3 H !,t
pK.3 B Xm I
[
:,",,
Sm
I
--
hi0114
H&&p H P2 X
plOll4BX
Mitochondrial DNA Downloaded from informahealthcare.com by Nanyang Technological University on 08/20/15 For personal use only.
t--+--ipBMl
pBM213 7-
pBM4
t-------cC--pBMSH1 (
Figure 1 Cloning and characterization of the H2A.1 and H2B.2 gcwe. T-he restriction maps show only the location of the restriction cnzyrncs usid for generation of subclones. Abbreviations of used restrii tiori enzymes: B BamHI. H: Hindlll. Pv2: Pvull. S: Sall. Sm: SmA. St: Stul. X : Xbal Xm: Xmnl. (A) Re.striction map of A10/14. Subciones pK3 and plO/l4BX are shown beneath the restriction map. LA and RA indicate the left and right arm of phage vector FMBl3A. Methods: 5x1 0' recombinant phages were screened with the rcindom-primed "P-labelled H2A.1 cDNA probe as described by Benton and Davis (1977). Positive phages were plaque-purified by thiree successive platings. Phage DNA was isolated from mini-lysates, characterized by restriction enzyme analysis, and Southern blotting as described by Maniatis et a/. (1982). A 0.7 kb Hindll/Stul (clone pK3) and a 1.5 kb BamHVXbal iragnient (clone plO/14BX) were isolated and subcloned in p U C l 2 according to standard protocols (Maniatis et a/., 1982). (B) Subciuning and sequencing strategy. Arrows indicate the orientation and length of the sequenced DNA regions within the suhclones. The positicins and orientations of the H2A and H2B genes are indicated by large arrows on the map. Methods: After restriction enzyme analysis of pl0/14BX restriction fragments were tiirthcr subcloncd into pUC vectors using the appropriate enzyme sites. Plasrnid DNAs from pK3, pBM1, pBM2/3, pBM4 and pHh.1SW1 were isolated from mini-preparations purified over C2iagt.n tip-20 columns, and the DNA sequences were determined twice on both strands by the dideoxy chain termination method ih'ith r: DNA Polymerase Sequencing Kit (Pharmacia). Products were analyzed on 6% polyacrylamide sequencing gels and tietc,rtd b y rxxposure to X-ray film (Kodak XAR-5).
A genomic EME!L3A library from the human Burkitt's lymphoma cell line BL64. (Hart1 and Lipp, 1987) was screened with a previously identified human ti2A.1 (EDNA probe (Dobner et a/., submitted). Homologous genomic clones were selected using high stringency conditions for hybridization and post-hybridization washes. Among .5 positivle clones one clone, designated 21 0/14, was selected for further detailed analysis. The characterization of d10/14 and the subsequent cloning of H2A and H2B-containing genomic D N A fragments in plasmid vectors for D N A sequence analysis i s summarized in Fig. 1. The nucleotide
sequence has been submitted to the EMBL Data Library (accession number X57138). The overlapping subclones, pK3 and p10/1 4BX, contain the complete 1661 bp nucleotide sequence of the H2A.1 cDNA probe and a novel H2B gene along with extensive 3' flanking regions. Both genes are arranged in a divergent orientation and are separated by a 313 bp intergene region (Fig. 2). The histone H2A and H2B sequences encode a 129 and 125 amino acid histone protein with a deduced molecular weight of 14093 and 13909, respectively. The nucleotide sequences are 82% (H2A) and 87% (H2B) homologous to the coding regions of a previously identified human genomic H2A and H2B gene (Zhong e t a / . , 1983). The predicted H2A protein corresponds to the human histone H2A.1 variant obtained by direct protein sequencing (Hayashi eta/., 1988), whereas the H2B protein is almost identical to the amino acid sequence of the human H2B.2 subtype (Ohe et a/., 1979), showing a 97% homology or 3 conservative amino acid changes as indicated in Fig. 2. Finally, the 3' flanking regions of the H2A #and H2B gene contain classical regions of dyad symmetry and a conserved downstream purine-rich element. Both elements are required for termination of transcription and 3' end processing of nonpolyadenylated replication-dependent histone premRNAs (Mowry eta/., 1989). In order to prove that the H2A/H2B gene pair is divergently transcribed from the intergene region we have mapped the 5' ends of both histone mRNAs. Figure 3 shows the results of primer extension analyses which demonstrate that the major start of the H2B.2 mRNA i s at the A residue (position 916, Fig. 2), whereas the H2A.1 mRNA starts at position 1056 (Fig. 2) at A. These results are consistent with other start sites of H2A and H2B mRNAs from chicken (Harvey et a/., 1982) and human (Zhong et a / . , 1983) published previous I y . The mapped H2A and H2B mRNA CAP sites are located approximately 30 bp downstream from the TATA-boxes (Fig. 4). Beside characteristic promoter elements of eukaryotic transcription units (TATA and CAAT-boxes) this divergent transcribed promoter has particularly interesting features. Remarkably the H2B subtype-specific element (H2B-box) which is normally found very close r6 to 30 bp) upstream of the TATA-box of almost all H2B genes (Harvey et a/., 1982; Wells and McBride, 1989) is located 112 bp upstream of the TATA-box
41 1
HUMAN HISTONE H2NH2B GENE PAIR
I I 1 I I I I I I 1 I I CCCGGGAGACGGAGCTTACAGTGAGCCGAGATCGCGGCACTGCACTCTATCCTGGGCGACAGAGCCAAGACTCCATTTAAAAAAAAAAAAAAAAAAAACACAAAAGGAAAGAGTAACCGA GGGCCCTCTGCCTCGAATGTCACTCGGCTCTAGCGCCGTGACGTGAGATAGGACCCGCTGTCTCGGTTCTGAGGTAAATTTTTTTTTTTTTTTTTTTTGTGTTTTCCTTTCTCATTGGCT
120
AAACCGAAAGCAAACTTGTAATTATAGCTTAAAGCACAACTGCAAGCCTTTTTCCCCAAGGGTATCTAAGAACCCAAGCTAAATTTGTCTGATGGAGCGTCTATTCTCACACGCAAAAAG 240 TTTGGCTTTCGTTTGAACATTAATATCGAATTTCGTGTTGACGTTCGGAAAAAGGGGTTCCCATAGATTCTTGGGTTCGATTTAAACAGACTACCTCGCAGATAAGAGTGTGCGTTTTTC CCTAGGGAGCTTAGCTTACCTTTTCCCGCCGTCTTCCCTCCTAATGGAGGCCGGCCGGAGTCCACCCAAAAGACTACAAGCAAAATGCACAGGCTCTTTCTCTGGAAGCTGGAGTGGCTC GGATCCCTCGAATCGAATGGAAAAGGGCGGCAGAAGGGAGGATTACCTCCGGCCGGCCTCAGGTGGGTTTTCTGATGTTCGTTTTACGTGTCCGAGAAAGAGACCTTCGACCTCACCGAG
TGAAAAGAGCCTTTGGATTGCTGGCAGCTGCGAGAGCTCACTTGGAGCTGGTGTACTTGGTGACGGCCTTTGTGCCCTCGGACACGGCGTGCTTGGCCAGCTCCCCGGGCAGCAGCAGGC ACTTTTCTCGGAAACCTAACGACCGTCGACGCTCTCGAGTGAACCTCGACCACATGAACCACTGCCGGAAACACGGG~GCCTGTGCCGCACGAACCGGTCGAGGGGCCCGTCGTCGTCCG
