Short sequence-paper

Isolation of cDNA clones encoding human histone macroH2A1 subtypes

Younghee Lee a, Mihui Hong a, Jae Wha Kim a, Yun Mi Hong a, Yong-Kyung Choe a,Sung Yeoul Chang b, Kwang Soo Lee c, In Seong Choe a;*

a Molecular and Cellular Biology Research Group, Korea Research Institute of Bioscience and Biotechnology, P.O. Box 115, Yusong,Taejon 305-600, South Korea

b Institute of Rheumatology, Hanyang University Medical School, Seoul 133-792, South Koreac Department of Surgery, Hanyang University Medical School, Seoul 133-792, South Korea

Received 17 March 1998; revised 25 May 1998; accepted 5 June 1998

Abstract

cDNA clones encoding two different subtypes of histone macroH2A1, macroH2A1.1 and macroH2A1.2, have beenisolated from human liver tissue. The open reading frames in the isolated clones predicted proteins comprising 368 and 371amino acids respectively. Estimated molecular masses of the two proteins were 39.0 kDa and 39.4 kDa. Human histonemacroH2A1.1 and macroH2A1.2 showed about 98% identity with their counterparts isolated from rat. The features of thenucleotide sequences of the two macroH2A1 subtypes in human were the same as in the rat system. Northern blot analysisshowed that the macroH2A1.1 and macroH2A1.2 subtypes were expressed as mRNA species with a size of 1.5 and 4.4 kb,respectively. They were expressed in all human tissues examined. ß 1998 Elsevier Science B.V. All rights reserved.

Keywords: MacroH2A1 subtype; cDNA sequence; Tissue distribution; (Human)

The nucleosome contains about 200 bp of DNAassociated with a histone octamer that consists oftwo copies each of H2A, H2B, H3, and H4. Theseare known as the core histones. Another type ofhistone, histone H1, is called a linker histone whichcan be dissociated from the nucleosome by salt treat-ment. The amino acid sequences of four histones(H2A, H2B, H3, and H4) from a wide variety oforganisms are remarkably similar among distantlyrelated species. Transcriptional activation is sup-posed to require that transcription factors success-fully compete with histones for binding to promoters,and the conserved features of histone sequences are

likely to be important in this function [1]. A fewyears ago, another class of core histone (molecularmass 42 kDa) containing a large non-histone regionwas isolated from rat liver nucleosome and namedmacroH2A1 (mH2A1) [2]. The N-terminal region ofmH2A1 showed similarity to the entire sequence ofH2A, and the following non-histone region con-tained a basic motif and a leucine zipper-like motif[3]. Furthermore, mH2A1 appeared to associate withH2B like H2A. These features suggested that theprotein could participate in the regulation of thecore protein assembly as a competitor of H2A.mH2A1 consisted of two distinct subtype proteinsthat could be resolved by electrophoresis on SDS-polyacrylamide gels with low cross-linking. cDNAclones encoding the mH2A1 subtypes were isolatedfrom rat and named mH2A1.1 and mH2A1.2, re-

0167-4781 / 98 / $19.00 ß 1998 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 7 - 4 7 8 1 ( 9 8 ) 0 0 0 9 8 - 0

* Corresponding author. Fax: +82 (42) 8604593;E-mail : choemcbg@kribb4680.kribb.re.kr

BBAEXP 91157 24-7-98

Biochimica et Biophysica Acta 1399 (1998) 73^77

Fig. 1. (A) Nucleotide sequence of the human mH2A1.2 cDNA and its deduced amino acid sequence. The numbers at the right ofeach line denote the positions of the last nucleotide and amino acid. In-frame stop codons de¢ning the 5P and 3P ends of the readingframe are indicated by an underline and an asterisk, respectively. The original partial cDNA clone, 5F05, covered the 3P terminal se-quences, 1156^1255 bp. The region of non-identity between the two subtypes is designated in bold letters. (B) The partial sequencesof human mH2A1.1, the StyI fragment of mH2A1.1 cDNA and the corresponding amino acids are shown. The cDNA fragment in-cluding the region of non-identity was used as a mH2A1.1-directed probe in Northern analysis (Fig. 3B). StyI restriction enzyme sitesare denoted in italics and underlined.

BBAEXP 91157 24-7-98

Y. Lee et al. / Biochimica et Biophysica Acta 1399 (1998) 73^7774

spectively [4]. These two subtype proteins showed thesame amino acid sequences except the internal regionof non-identity around the leucine zipper-like region.It suggested the presence of an alternative splicingmechanism in the generation of mRNA species.

To study the function of mH2A1 subtypes in thehuman system, it is important to obtain humancDNA clones harboring the full open reading frame(ORF) of the proteins. Here, we report the cloningand sequencing of the cDNA clones containing the

full ORF of mH2A1.1 and mH2A1.2 from humanliver tissues. The expression of mH2A1 messages invarious human tissues was also studied.

To isolate noble genes up-regulated in hepatoade-nocarcinoma, we have constructed a cDNA librarywith a PCR-select cDNA subtraction system (Clon-tech) using the cDNAs from liver carcinoma andnormal liver tissue as tester and driver, respectively(manuscript in preparation). Among the cDNAclones analyzed, clone 5F05 (124 bp long) showed

Fig. 2. Sequence alignment of the predicted mH2A1 proteins (hmH2A1.1 and hmH2A1.2) with rat proteins (rmH2A1.1 andrmH2A1.2) using the CLUSTAL program [11] of PCGENE. Perfectly conserved residues are denoted with an asterisk and highly con-served ones are denoted with a dot. The region of non-identity between the two subtypes is designated in bold letters. The basic re-gion and a leucine zipper-like motif are indicated.

BBAEXP 91157 24-7-98

Y. Lee et al. / Biochimica et Biophysica Acta 1399 (1998) 73^77 75

91% homology with the nucleotide sequence of ratmH2A1 mRNA. On the basis of the informationfrom the partial cDNA sequence of clone 5F05,a primer covering the stop codon and the 3P endregion of human mH2A1 gene (primer A: 5P-GCCTAGTTGGCGTCCAGCTTG-3P) was synthe-sized and used for 5P-rapid cDNA end ampli¢cation(RACE). The Marathon cDNA library constructedfrom human fetal liver was used as a template forRACE [5]. A PCR product with an approximate sizeof 1.2 kb was isolated and directly cloned into pCRIIvector using a TA cloning kit (Invitrogen), and des-ignated as clone 5F05-4.

The nucleotide sequence of clone 5F05-4 was de-termined in both directions by the chain terminationmethod [6]. As shown in Fig. 1, the cDNA is 1255 bplong and contains an ORF encoding 371 amino acidswith a deduced molecular mass of 39.4 kDa. ThisORF is presumed to start from the ¢rst AUG codonat position 108 found in the crucial context for e¤-cient translation of eukaryotic mRNA (CC(A/G)CCAUGG) [7]. The cDNA clone contains an in-frame UGA stop codon 24 nucleotides upstreamfrom the AUG codon. BLAST search of databases

revealed that this clone has about 89% and 99% ho-mology in nucleotide and deduced amino acid se-quences with rat mH2A1.2 [8]. We named thiscDNA clone human mH2A1.2, GenBank accessionNo. AF041483. As an attempt to clone mH2A1.1cDNA sequences from human liver, a primer encod-ing the 5P end region of human mH2A1.2 ORF in-cluding the start codon (primer B: 5P-CCATGTC-GAGCCGCGGTGGGAAG-3P) was synthesizedand used as an amplimer in combination with primerA described above. The RT-PCR product with anapproximate size of 1.1 kb was directly cloned intopCRII vector, and the resulting plasmid DNAs wereanalyzed by restriction mapping and DNA sequenc-ing [9]. Four out of eight independent plasmids hadthe nucleotide sequence of mH2A1.1 and the remain-ing four had the mH2A1.2 sequence. The sequence ofhuman mH2A1.1 cDNA, GenBank accession No.AF044286, was 1111 bp long and contained anORF encoding 368 amino acids with a deduced mo-lecular mass of 39.0 kDa. The partial sequences oftwo cDNA clones including full ORF are shown inFig. 1B. The sequences of the two mH2A1 subtypesdi¡ered only in the region indicated in bold letters,

Fig. 3. Distribution of human mH2A1 mRNA in di¡erent tissues. A multiple tissue blot membrane (Clontech) was used. Lanes 1^8contain, in order, 2 Wg of poly(A)�RNA from human heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas. Sizemarkers are indicated on the left. The blot was hybridized with the entire sequence of mH2A1.2 (A), or the StyI fragment ofmH2A1.1 (B), or glyceraldehyde-3-phosphate dehydrogenase (GAPDH; C) at 65³C for 1 h in an express hybridization solution (Clon-tech). It had been washed twice to a ¢nal stringency of 0.1USSC, 0.1% SDS at 50³C for 20 min [12]. cDNA probes were 32P-labeledwith random priming method [13].

BBAEXP 91157 24-7-98

Y. Lee et al. / Biochimica et Biophysica Acta 1399 (1998) 73^7776

and matched perfectly in other regions. The featuresof the subtypes of human mH2A1 gene were thesame as in the rat system.

The alignment of amino acid sequences in Fig. 2shows that the amino acid sequences of the humanmH2A1.1 and mH2A1.2 exhibited homology of99.9% with those of rat mH2A1.1 (98.0% identity,1.9% favored substitutions) and mH2A1.2 (98.3%identity, 1.6% favored substitutions). This high de-gree of conservation of mH2A1 proteins in di¡erentspecies indicated that these proteins might have cer-tain functional signi¢cance such as in the regulationof transcription as predicted previously [4].

Northern blot analysis was performed to study theexpression of mH2A1 messages in various humantissues. The presence of mH2A1 messages in totalRNA preparations isolated from the liver tissues offetus, normal adult, and cancer patient, and fromhepatoma cell line HepG2 [10], was tested using theentire cDNA sequence of human mH2A1.2 as aprobe. However, no prominent di¡erences amongthe tissues were detected (data not shown). Thepoly(A)�RNA fractions prepared from heart, brain,placenta, lung, liver, skeletal muscle, kidney, andpancreas were subjected to Northern blot analysisusing the same probe. The transcripts of 1.5 and4.4 kb were found in all tissues examined as shownin Fig. 3A, with relatively higher abundance in heart,liver, skeletal muscle, kidney, and pancreas than inbrain, placenta, and lung. It was in agreement withthe previous report that the content of mH2A1 pro-teins was higher in liver and kidney than other tis-sues such as brain, testis, and thymus in mouse, rat,and rabbit [4]. To determine the identity of the twodi¡erent transcripts, the multiple tissue blot was re-probed with a 167 bp long StyI fragment ofmH2A1.1 cDNA which is shown in Fig. 1B. ThismH2A1.1-directed probe included the 90 bp long re-gion of non-identity and the common sequence of 77bp shared by the two subtypes. As shown in Fig. 3B,

this cDNA probe recognized mainly the transcript of1.5 kb and barely that of 4.4 kb. From the aboveresults, it could be concluded that the mRNA speciesof 1.5 kb and 4.4 kb were the messages of mH2A1.1and mH2A1.2, respectively. These di¡erent tran-scripts of mH2A1 subtypes could be produced byalternative splicing or by transcription from di¡erentloci. The distribution of human mH2A1 transcriptswas not consistent in di¡erent tissues and might sug-gest that there are some discrepancies in di¡erenttissues in the expression of human mH2A1 genes.

This study was supported by a grant from theMinistry of Science and Technology of Korea(HAN Project HS2070M).

References

[1] G. Felsenfeld, Nature 355 (1992) 219^224.[2] J.R. Pehrson, V.A. Fried, Science 257 (1992) 1398^1400.[3] E.K. O'shea, J.D. Klemm, P.S. Kim, T. Alber, Science 254

(1991) 539^544.[4] J.R. Pehrson, C. Costanzi, C. Dharia, J. Cell. Biochem. 65

(1997) 107^113.[5] Y. Lee, J.W. Kim, I.A. Lee, H.B. Kang, Y.-K. Choe, H.G.

Lee, J.S. Lim, H.J. Kim, C. Park, I.S. Choe, Biochem. Mol.Biol. Int. 39 (1996) 833^842.

[6] F. Sanger, S. Nicklen, A.R. Coulson, Proc. Natl. Acad. Sci.USA 74 (1977) 5463^5467.

[7] M. Kozak, J. Cell Biol. 108 (1989) 229^241.[8] S.F. Altschul, W. Gish, W. Miller, W.M. Eugene, D.J. Lip-

man, J. Mol. Biol. 215 (1990) 403^410.[9] E.S. Kawasaki, in: M.A. Innis, D.M. Gwelfand, J.J. Snin-

sky, T.J. White (Eds.), PCR Protocols: a Guide to Methodsand Applications, Academic Press, San Diego, CA, 1989, pp.21^27.

[10] P. Chomczynski, N. Sacchi, Anal. Biochem. 162 (1987) 156^159.

[11] D.G. Higgins, P.M. Sharp, Gene 73 (1988) 237^244.[12] J.W. Kim, Y. Lee, I.A. Lee, H.B. Kang, Y.K. Choe, I.S.

Choe, Biochim. Biophys. Acta 1350 (1997) 133^135.[13] A.P. Feinberg, B. Vogelstein, Aanl. Biochem. 132 (1983) 6^

13.

BBAEXP 91157 24-7-98

Y. Lee et al. / Biochimica et Biophysica Acta 1399 (1998) 73^77 77