- Email: [email protected]
Characterization of a new subgroup of human Ig Vλ cDNA clone and its expression
Molecular Immunology, Vol. 29, No. 6, pp. 723-728, 1992 Printed in Great Britain.
0161~5890/92 $5.00 + 0.00 0 1992 Pergamon Press Ltd
CHARACTERIZATION OF A NEW SUBGROUP OF HUMAN Ig V, cDNA CLONE AND ITS EXPRESSION YASUKO TSUNETSUGU-YOKOTA,*~ TAKAYUKI MINEKAWA,* KAZUHIRO SHIGEMOTO,* TAKUJI SHIRASAWA$ and TOSHITADA TAKEMORI* *Department of Cellular Immunology, National Institute of Health of Japan, Tokyo, Japan and SDepartment of Molecular Pathology, Metropolitan Institute of Senile Science, Tokyo, Japan (First received 17 October 1991; accepted in revised form 26 November 1991)
Abstract-From a human bone marrow cDNA library, we have cloned and sequenced a gene which cross-hybridized with murine pre-B cell-specific gene 8HS-20 cDNA under the low-stringent condition. Sequence analysis predicted that this gene (YM-1) encoded 240 amino acids which had the basic structure of immunoglobulin 1 light chain. The 3’ half of the YM-1 sequence was identical to the Jn2C,Z region except for four nucleotides. The 5’ part of the gene had 87.6% sequence homology with the reported V, gene called Tl. Comparison of the deduced amino acid sequences with representative members of the seven other VI subgroups showed considerable structural homology, but the maximum homology with these chains was 44%. Therefore, we conclude that YM-1 belongs to a new V, subgroup. Interestingly YM-1 showed higher homology with VpreBl (56%) than with any of the other V1 subgroups. By Southern blot analysis four to six cross-hybridizing Vn bands were detected at high stringency. Expression of the V, gene was observed in immature as well as mature B cell lines without accompanying V-JC gene joining, suggesting that V, of the YM-1 locus is activated at the early stage of maturation.
INTRODUCTION The development of B lymphocytes follows a wellorganized program. The first step is DJ and then VDJ joining of the immunoglobulin heavy (IgH) chain gene, resulting in the expression of p chain in the cytoplasm (pre-B cell). At the next step, the immunoglobulin light (IgL) chain (K, A) gene rearranges and expresses mature IgL chain followed by H,LI immunoglobulin (Ig) tetramer formation, which is an essential process for B cell maturation. In the pre-B cell, stage-specific, p-associated I-like proteins called VpreBl and 15 have been identified and their genes molecularly cloned (Kudo and Melchers, 1987; Sakaguchi and Melchers, 1986). They are thought to play an important role in B cell development, inducing IgL chain rearrangement. Recently we obtained from a mouse an immature B cell-specific cDNA clone, 8HS-20, by the subtraction method (T. Shirasawa et al., manuscript in preparation). This gene product also is a I-like protein and is partially associated with p chain. Since the results of genomic Southern blotting at low stringency suggested the existence of an 8HS-20 homologue in humans, we have tried to obtain this gene. The clone (YM-1) that cross-hybridized with murine 8HS-20 was obtained from a human bone marrow TAuthor to whom correspondence should be addressed: Yasuko Tsunetsugu-Yokota, Department of Cellular Immunology, National Institute of Health of Japan, 2-10-35, Kami-osaki, Shinagawa-ku, Tokyo 141, Japan. The nucleotide sequence data reported in this paper will appear in the DDBJ, EMBL and GenBank Nucleotide Sequence Databases under the accession number D01059.
cDNA library, but turned out to be the AIgL chain gene. The IgL chain is either K or 1 type and in humans 40% of Igs contain 1 chains (Fu et al., 1976) whereas in mice only 5% do (Cotner and Eisen, 1978). The study of the human 3, IgL chain locus revealed six tandemly repeated non-allelic C1 genes on chromosome 22, each of which is associated with its own Ja segment (Heiter et al., 1981). Recently a seventh CA gene and its J, segment have been identified (Vasicek and Leder, 1990; Bauer and Blomberg, 1991) and Ckd,,C,, and Ci, genes are now know to be pseudogenes. On the other hand, the organization of human V, genes is not well known. The number of human VA genes is thought to be much larger than the limited number of mouse V, genes. Seven subgroups have been identified according to the amino acid sequences of the VA chain (Anderson et al., 1984), and each subgroup appears to contain about 10 members or so. Here we report that the YM-1 clone consists of a V, gene, which belongs to a novel V, subgroup joined to the JA,Ci2 gene. We characterized this gene and its expression.
MATERIALS Screening
AND METHODS
of a library
A cDNA library prepared from human bone marrow in bacteriophage lgtl0 was purchased (Clontech Lab. Inc., Palo Alto, CA) and screened with 8HS-20, which was isolated from a cDNA library of a murine pre-B cell line (496138) by subtraction and differential hybridization (T. Shirasawa, manuscript in preparation). This gene encodes a 14 kDa protein associated with p chains in immature B cells. The nucleotide sequence of 8HS-20 723
Y. TSUNETSUGU-YOKOTAet al.
124
has 59% homology with human V, and 5460% homology with Vrc. On the other hand, the immature B-cell specific gene, VpreBl, shares 36% amino acid and less nucleotide sequence homology with 8HS20. Plaque hybridization was carried out under the low-stringent condition: 5 x SSC, 0.5% SDS at 65°C for hyb~dization followed by washing in 2 x SSC, 0.5% SDS at 50°C. The solution for prehybridization was 4 x SSC, 1 x Denhardt, 50 mM Tris-HCl at pH 7.5, I mM EDTA, 0.5% SDS, 100 pug of denatured herring sperm DNA per ml, and for hybridization 10% dextran sulphate was added to the prehybridization solution. Nine clones obtained after the third screening were analyzed further. Sequencing DNA restriction fragments were subcloned into the appropriate site of either the M 13 cloning vector or BSK plasmid. Sequencing was carried out with a Sequenase Version 2.0 DNA sequencing kit (US Biochemical Corp., Cleveland, OH) and [a-35S]-dCTP. Cell lines ARH and Wa cell lines were mature B cell lines kindly provided by Dr A. Yano (Nagasaki University, Nagasaki, Japan) (Yano et al., 1989). The cytoplasmic p-positive pre-B cell lines, HPB-null and Nalm-6, and a null cell line, Nalm-16, were obtained by the courtesy of Dr J. Minowada (Fujisaki Cell Center, Okayama, Japan). Ball-l, CEM and U937 were B, T and monocytic cell lines, respectively. RNA isolation and analysis Total RNA from cultured cells was extracted by the method described previously (Chomczynski and Sacchi, 1987). RNA samples were denatured at 65°C with 50% formamide and 2.2 M fo~aldehyde and electrophoresed on 1.2% agarose gels containing 2.2 mol of formaldehyde, Ten microgrammes of RNA was transferred to a nylon membrane (Hybond N, Amersham), and the conditions of hybridization and washing were essentially the same as for the genomic Southern blotting described below. The VpreBl probe was synthesized by the polymerase chain reaction using Taq polymerase (Cetus Corp., Norwalk, CT).
analyzed cDNA inserts from nine clones, There were two groups of clones; the size of one group (6 clones) was 1 kbp and that of the other group (3 clones) was 2.5 kbp. One clone from each group was subcloned into plasmid BSK and they were designated as YM-1 and YM-6. Restriction mapping analysis revealed that they were completely different genes (data not shown). The genomic Southern blots of human and mouse DNA were analyzed first by hybridizing with the 8HS-20 probe at low stringency, then by rehybridizing with the YM-6 probe followed by the YM-1 probe at high stringency. The size of the main band in the human genome hybridized with each clone was not the same as that expected from the results with the 8HS-20 probe. However, a weak signal corresponding to the band hybridized with 8HS-20 was detectable with the YM-1 probe, which suggests that at least YM-1 has weak homology with the human homologue of 8HS-20 (data not shown). Characterization of the YM-1 sequence We determined the nucleotide sequence of the YM-I clone. The sequencing strategy is outlined in Fig. 1, and the complete nucleotide sequence and the predicted amino acid sequence of YM-1 are shown in Fig. 2. The translation initiation codon starts at nucleotide (nt.) 23 from the EcoRI cloning site and continues to the termination codon at nt. 742, This open reading frame is able to encode 240 amino acids and the predicted molecular weight is about 26 kd. There are two ploy(A) signals, at nt. positions 848 and 860. The overall nucleotide sequence homology of YM- 1 and 8HS20 was turned out to be less than 40%. However, from nt. 187 through nt. 225 of YM-I,33 nucleotides were completely matched with the sequence of 8HS20 (87% homology), which suggests the possibility of the cross-hybridization of two genes. Homology search by the computer revealed a maximum homology of the 3’ part of the YM-1 sequence with human J,CAZ gene of the immunoglobulin lambda light chain reported by Yamasaki et al. (Yamasaki et al., 1987). There were four single base changes, at nt. 397 in the J region and at 637 in the C region, which did not result in amino acid change, and at nt. 555 and 737 in the C region, which resulted in an amino acid changes
Genomic Southern blotting Five microgrammes of high mol. wt DNA was digested with EcoRI restriction enzyme, and agarose gel electrophoresis was carried out. DNA was transferred to nylon filters and hybridized with probes under the high-stringent condition; 5 x SSC, 0.5% SDS at 65°C for hybridization, 0.1 x SSC, 0.5% SDS at 65°C for washing.
L
J
V
I EcoRl
BamHl
C
Ilu~ll Smal
i 1 BarnHI
I EcoRl
4
c
I 100 bp
RESULTS AND DISCUSSION Isolation of human 8US-20 homologue We screened a human bone marrow cDNA library using mouse 8HS-20 cDNA as a probe under the low-stringent condition. After the third screening we
Fig. 1. Restriction map of YM-1. L, leader; V, variable; J, joining; C, constant region. The open bar indicates exon and the closed bar shows the untranslated region. Fragments shown below it were subcloned into either the Ml3 phage cloning vector or Bluescript plasmid and the arrows indicate the direction of sequencing.
725
A new subgroup of human Ig V,
from glycine to alanine and from cysteine to arginine, respectiveIy. The 5’ part of the YM-1 nucleotide sequence showed 87.6% homology with the V1 gene of Tl (Bernstein et aE_,1989). The first 19 amino acids were leucine rich, having an overall hydrophobic character, which is a typical structure of the signal peptides. V, genes have been divided into seven subgroups according to amino acid sequences of the V, region (Anderson et a/., 1984). Compa~son of the deduced amino acid sequence of the V region of YM-I with those of representative members of the other subgroups of VAis shown in Fig. 3, in which the sequences were aligned for maxima homology. YM-1 showed considerable st~ctural homology with those Vi,poly~ptides, but had a unique feature of amino acid compositiou. Interestingly, the highest homology was observed with human pre-B all-s~cific gene VpreBl (56%), while the other V, subgroups had 35 to 44% homology. These results indicate the the V region of YM-1 (V~/YM-I) along with Tl belongs to a new subgroup of V, , and that VpreB 1 possibly diverged from this V, subgroup after the diversification of a V, family.
To characterize the V,,WM-1 locus, DNAs from various cell lines were digested with EcoRI, and Southern blot analysis was carried out by probing with Vn/YM-I (EcoRI-BamHI fragment in Fig. 2) under the high-stringent condition. Then the same filter was rehybridized with the Cu gene probe (Smal-Pstl fra~ent in Fig. 2). In EcoRI digests of DNA from various cell lines, four cross-hybridizing bands of V,/YM-1 of 7.6 kb, 4.4 kb, 4.2 kb and 3.9 kb were observed commonly in four of the seven cell lines, as indicated by arrowheads in Fig. 4A (lanes 1,3,4 and 7). Less intense bands of 7.0 kb, 2.8 kb and 2.4 kb also appeared in all cell lines, which suggests that these bands represent the less homologous genes of the YM-1 family. In the monocytic cell
Fig, 2. ~ucleotide
sequence of YM-1, The amino acid sequences deduced are shown above the nucleotide sequence starting at ATG and ending at TAG (double underline). Figures below the nucleotide sequence refer to the nucleotide number starting at the sequence recognized by EcoRXrestriction enzyme. Poly(A) signals are shown by dotted underling. CDR denotes a complementa~ty~e~~ining region. Boxed nucleotides indicate base changes compared to those of the C region of p3C4;1 S (12). Restriction sites used to prepare probes are underlined.
CO_
60
1x1 hIi1
B”R en.
I
SO
70 G G G
homology
100
TSAS-UTAS-TTVT--
44
44 NSRDSS
i:
SSYVDN
40 36 38 56
Fig. 3. Comparison of the predicted amino acid members of seven known subgroups and VpreBl. are shaded and the per cent homology with the calculated. Figures above the amino acid
sequence of Amino acid total number sequence are
YM-1 with those of representative sequences homologous with YM-1 of amino acids in this figure was based on the V,, sequence.
726
et (11.
Y.Tsmmsucu-YofcoT~
1234567
1234567 kb
kb
23
23
9.4
6.6
6.6
4.4
2.3
2.3
Fig. 4. Southern blot analysis of different cell lines under the stringent condition. A, The filter was hybridized with a V,/YM-1 probe. B, The same filter was washed and rehybridized with a C,, probe. Lane 1, CEM; lane 2, U937; lane 3, ARH; lane 4, Wa; lane 5, HPB-null; lane 6, Nalm-6; lane 7, Nalm-16. The size marker is lamda DNA digested with Hind III. Arrowheads indicate the main bands observed in common five of the seven cell lines.
line U937 (lane 2) and the pre-B cell line HPB-null (lane 5), a 7.3 kb instead of a 7.6 kb band was observed and an additional 3.6 kb band was detected in the U937 cell line. In another pre-B cell line, Nalm-6 (lane 6) both 7.6 and 7.3 kb bands were observed in addition to a 3.6 kb band. Meanwhile the pattern of Southern blot hybridization of CI genes showed the germ-line configuration, which is 16 kb, 14 kb and 8 kb bands in EcoRI-digested DNA (Taub et al., 1983), in most of the cell lines. Two pre-B cell lines, HPB-null and Nalm-6, showed rearrangement of the 8 kb band containing Cnz (Fig. 4B, lanes 5 and 6). However, none of the rearranged C, bands hybridized with VJYM-1 (Fig. 4A). From these results taken together, we assume either that an abnormal rearrangement has occurred in U937 and the two pre-B cell lines or, more likely, that the 7.6 kb, 7.3 kb and
3.6 kb bands reflect the polymorphism of the YM-1 family. The genomic structure of VJYM-1 must be determined to know the exact nature of these patterns of Southern blot hybridization. Both C, and V,/ YM- 1 are expressed independently as a sterile message in immature and mature B cell lines 1. A sterile C, message. The expression of YM-1 in the cell lines used for Southern blot analysis and an additional mature B cell line, Ball- 1, was investigated by Northern blot analysis. As shown in Fig. 5A, the C,, probe revealed abundant expression of C, in Wa, HPBnull, Nalm-6 and Nalm-16 cell lines (lanes 5,6, 7 and 8) and weak expression in the U937 cell line (lane 2). The major C, transcript (0.95 kb) was smaller than mature 1. mRNA (1.2 kb) (Sun et al., 1985). Taking into account
C 12345678
12345678
12345678
-
Fig. 5. Northern blot analysis. The filter was prepared as described in Materials and Methods, and hybridized with probe C,, (A) (Zday exposure), V,/YM-1 (B) (8-day exposure) or VpreBl (C) (l-day exposure). Lane 1, CEM; lane 2, U937; lane 3, Ball-l; lane 4, ARH; lane 5, Wa; lane 6, HPB-null; lane 7, Nalm-6; lane 8, Nalm-16.
727
A new subgroup of human Ig V,
the results of Southern blot analysis of the C1 genes, we consider that these C1 transcripts are sterile messages. While very little is known about the sterile C1 message, the sterile message is characterized as non-functional transcriptional activation of unrearranged Cp (Kemp et al., 1980; Alt et al., 1982) or C, (Perry et al., 1980) gene and is observed even in T and myeloid cells. We observed the sterile C, message in the CEM (T) cell line and additional heterogenous messages in Wa, HPB-null, Nalm-6 and Nalm- 16 cell lines after long exposure of the film (data not shown), only one of which is visible in Fig. 5A, lane 7. Thus, the activation of the C, gene appears to be initiated early in a progenitor cell of the T, B, and myeloid lineages as reported in the case of Cp sterile messages (Kemp et al., 1980). However, the origin of transcription and enhancer/promoter region of the sterile Cn message is not known. 2. A sterile Vn message. When the same filter was rehybridized with a V1/YM-1 probe, a low but detectable level of a l.O-kb transcript was observed in Wa (lane S), HPB-null (lane 6), Nalm-6 (lane 7), and Nalm-16 (lane 8) (Fig. 5B). The mature B cell (Wa) showed a higher level of expression than the immature cells. These V, messages were not the result of the gross-hybridization with the transcript of VpreBl, because the Vpre Bl message was smaller (0.8 kb) than that of V,/YM-I and observed only in immature B cells (lanes 68), not in Wa (lane 5) (Fig. 5C). Since these transcripts do not appear to be derived from the rearranged V, gene, we conclude that they are germ-line V, messages (sterile V, messages). Expression of the unrearranged V region of the heavy chain (vu) is limited to the early stage of the B-lymphocyte differentiation pathway (Yancopoulos and Alt, 1985) and is temporarily detected in murine fetal liver (Lennon and Perry, 1990). In this case, the lack of unrearranged V, expression in mature cells is thought to be due to the inactivation of a locus which was previously active (Yancopoulos and Alt, 1985). In contrast, the unrearranged V, locus appears to be activated along with B cell maturation. Although the germ-line ?J1 message in humans is not known, studies by Picard and Shaffner (Picard and Shaffner, 1984) on various murine cell lines showed that the unrearranged Vi region was transcribed at a low but steady-state level of l-10 RNA molecules per cell, but only in myelomas and hybridomas, not in cell lines representing the pre-B cell and immature B cell stage. In the present study on human cell lines we were able to detect germ-line V, messages even in the null cell as well as immature B cells albeit at much lower level than in the mature B cell. This difference may reflect the fact that mouse Ig contains only 5% 1 IgL chain while human Ig is 40% ;1.type. Our novel finding in humans indicates that the chromatin structure of V, locus is already open before the synthesis of p chain in a cell of the B lineage, which is also suggested by Leclercq et al. (1989) in their study of a germ-line J, transcript. Whether a germ-line V, transcript similar to YM-1 is also expressed in other VA subgroups is unknown.
Further study is required to know the regulatory mechanism for the activation of the V, gene during B cell maturation. Acknowledgement-This work was supported by a Research Grant for Aging and Health from the Ministry of Health and Welfare.
REFERENCES Ah F. W., Rosenberg N., Enea V., Siden E. and Baltimore D. (1982) Multiple immuuoglobulin heavy-chain gene transcripts in Abelson murine leukemia ~rus-t~nsfo~ed lymphoid cell lines. Molec. cell. Biol. 2, 386-400, Anderson M. L. M., Szajnert M. F., Kaplan J. C., McCall L. and Young B. D, (1984) Cloning and sequence analysis of an Ig 2 light chain mRNA expressed in the Burkitt’s lymphoma cell line EB4. Nucl. Acids Res. 12, 6647-6661. Bauer T. R., Jr and Blomberg B. (1991) The human A L chain Ig locus: Recharacterization of JC16 and identification of a functional JC17. J. Zmmun. 146, 2813-2820. Berinstein N., Levy S. and Levy R. (1989) Activation of an excluded immunoglobulin allele in a human B lymphocyte cell line. Science 244, 337-339. Chomczynski P. and Sacchi N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanate-phenolchloroform extraction. Anafyt. Biochem. 162, 156-I 59. Cotner T. and Eisen H. V. (1978) The natural abundance of y2-fight chains in inbred mice. J. exp. Med. 148, 1388-1399. Fu S. M., Chiorarri N. and Kunkel H. G. (1976) Differentiation capacity and other properties of the leukemic cells of chronic lymphocytic leukemia. Zmmun. Rev. 48, 23-44. Heiter P. A., Hollis G. F., Korsmeyer S. J., Waldmann T. A. and Leder P. (1981) Clustered arrangement of immunoglobulin t constant region genes in man. Nature 294, 536-540. Kemp D. J., Harris A. W. and Adams J. M. (1980) Transcripts of the immunogIobulin Cp gene vary in structure and splicing during lymphoid de~lopment. Proc. natn. Acad. Sci. U.S.A. 77, 7~7~. Kudo A. and Melchers F. (1987) A second gene, Vpre B in the 15 locus of the mouse, which appears to be selectively expressed in me-B lymphocytes. EMBO J. 6, 2267-2272. Leclercq L., Butkeraitis P. and Reth M. (1989) A novel germ-line J, transcript starting immediately upstream of J, 1. Nuci. Acids Res. 17, 6809-6819. Lennon G. G. and Perry R. P. (1990) The temporal order of appearance of transcripts from unrearranged and rearranged Ig genes in murine fetal liver. J. Zmmun. 144, 1983-1987. Perry R. P., Kelley D. E., Coleclough C., Seidman J. G., Leder P., Tonegawa S., Matthyssens 6. and Weigert M. (1980) Transcription of mouse ICchain genes: Implications for allelic exclusion. Proc. rtafn. Acad. Sci. U.S.A. 77, 1937-1941. Picard D. and Shaffner W. (1984) Unrearranged immunoglobulin R variable region is transcribed in K-producing myelomas. EMBO J. 3, 3031-3035. Sakaguchi N. and Melchers F. (1986) 15, a new light chainrelated locus selectively expressed in pre-B lymphocytes. Nature 324, 579-582. Sun L.-H. K., Croce C. M. and Showe L. C. (1985) Cloning and sequencing of a rearranged V, gene from a Burkitt’s
72x
Y.
TSUNETSUGU-Y0K0T.4
lymphoma cell line expressing kappa light chains. Nucl. Acids Res. 13, 49214934. Taub R. A., Hollis G. F., Heiter P. A., Korsmeyer S., Waldmann T. A. and Leder P. (1983) Variable amplification of immunoglobulin i light-chain genes in human populations. Nature 304, 1722174. Vasicek T. J. and Leder P. (1990) Structure and expression of the human immunoglobulin i genes. J. exp. Med. 172, 6099620. Yamasaki N., Komori S. and Watanabe T. (1987) Comp-
et al.
lementary DNA for a human subgroup IV immunoglobulin I-chain. Molec. Immun. 24, 981-985. Yancopoulos G. D. and Alt F. W. (1985) Developmentally controlled and tissue-specific expression of unrearranged V, gene segments. Cell 40, 271-281. Yano A., Aosai F., Ohta M., Hasekura H., Sugane K. and Hayashi S. (1989) Antigen presentation by Toxoplasma Gondii-infected cells to CD4 + proliferative T cells and CD8 + cytotoxic cells. J. Purasitol. 75, 411-416.
0161~5890/92 $5.00 + 0.00 0 1992 Pergamon Press Ltd
CHARACTERIZATION OF A NEW SUBGROUP OF HUMAN Ig V, cDNA CLONE AND ITS EXPRESSION YASUKO TSUNETSUGU-YOKOTA,*~ TAKAYUKI MINEKAWA,* KAZUHIRO SHIGEMOTO,* TAKUJI SHIRASAWA$ and TOSHITADA TAKEMORI* *Department of Cellular Immunology, National Institute of Health of Japan, Tokyo, Japan and SDepartment of Molecular Pathology, Metropolitan Institute of Senile Science, Tokyo, Japan (First received 17 October 1991; accepted in revised form 26 November 1991)
Abstract-From a human bone marrow cDNA library, we have cloned and sequenced a gene which cross-hybridized with murine pre-B cell-specific gene 8HS-20 cDNA under the low-stringent condition. Sequence analysis predicted that this gene (YM-1) encoded 240 amino acids which had the basic structure of immunoglobulin 1 light chain. The 3’ half of the YM-1 sequence was identical to the Jn2C,Z region except for four nucleotides. The 5’ part of the gene had 87.6% sequence homology with the reported V, gene called Tl. Comparison of the deduced amino acid sequences with representative members of the seven other VI subgroups showed considerable structural homology, but the maximum homology with these chains was 44%. Therefore, we conclude that YM-1 belongs to a new V, subgroup. Interestingly YM-1 showed higher homology with VpreBl (56%) than with any of the other V1 subgroups. By Southern blot analysis four to six cross-hybridizing Vn bands were detected at high stringency. Expression of the V, gene was observed in immature as well as mature B cell lines without accompanying V-JC gene joining, suggesting that V, of the YM-1 locus is activated at the early stage of maturation.
INTRODUCTION The development of B lymphocytes follows a wellorganized program. The first step is DJ and then VDJ joining of the immunoglobulin heavy (IgH) chain gene, resulting in the expression of p chain in the cytoplasm (pre-B cell). At the next step, the immunoglobulin light (IgL) chain (K, A) gene rearranges and expresses mature IgL chain followed by H,LI immunoglobulin (Ig) tetramer formation, which is an essential process for B cell maturation. In the pre-B cell, stage-specific, p-associated I-like proteins called VpreBl and 15 have been identified and their genes molecularly cloned (Kudo and Melchers, 1987; Sakaguchi and Melchers, 1986). They are thought to play an important role in B cell development, inducing IgL chain rearrangement. Recently we obtained from a mouse an immature B cell-specific cDNA clone, 8HS-20, by the subtraction method (T. Shirasawa et al., manuscript in preparation). This gene product also is a I-like protein and is partially associated with p chain. Since the results of genomic Southern blotting at low stringency suggested the existence of an 8HS-20 homologue in humans, we have tried to obtain this gene. The clone (YM-1) that cross-hybridized with murine 8HS-20 was obtained from a human bone marrow TAuthor to whom correspondence should be addressed: Yasuko Tsunetsugu-Yokota, Department of Cellular Immunology, National Institute of Health of Japan, 2-10-35, Kami-osaki, Shinagawa-ku, Tokyo 141, Japan. The nucleotide sequence data reported in this paper will appear in the DDBJ, EMBL and GenBank Nucleotide Sequence Databases under the accession number D01059.
cDNA library, but turned out to be the AIgL chain gene. The IgL chain is either K or 1 type and in humans 40% of Igs contain 1 chains (Fu et al., 1976) whereas in mice only 5% do (Cotner and Eisen, 1978). The study of the human 3, IgL chain locus revealed six tandemly repeated non-allelic C1 genes on chromosome 22, each of which is associated with its own Ja segment (Heiter et al., 1981). Recently a seventh CA gene and its J, segment have been identified (Vasicek and Leder, 1990; Bauer and Blomberg, 1991) and Ckd,,C,, and Ci, genes are now know to be pseudogenes. On the other hand, the organization of human V, genes is not well known. The number of human VA genes is thought to be much larger than the limited number of mouse V, genes. Seven subgroups have been identified according to the amino acid sequences of the VA chain (Anderson et al., 1984), and each subgroup appears to contain about 10 members or so. Here we report that the YM-1 clone consists of a V, gene, which belongs to a novel V, subgroup joined to the JA,Ci2 gene. We characterized this gene and its expression.
MATERIALS Screening
AND METHODS
of a library
A cDNA library prepared from human bone marrow in bacteriophage lgtl0 was purchased (Clontech Lab. Inc., Palo Alto, CA) and screened with 8HS-20, which was isolated from a cDNA library of a murine pre-B cell line (496138) by subtraction and differential hybridization (T. Shirasawa, manuscript in preparation). This gene encodes a 14 kDa protein associated with p chains in immature B cells. The nucleotide sequence of 8HS-20 723
Y. TSUNETSUGU-YOKOTAet al.
124
has 59% homology with human V, and 5460% homology with Vrc. On the other hand, the immature B-cell specific gene, VpreBl, shares 36% amino acid and less nucleotide sequence homology with 8HS20. Plaque hybridization was carried out under the low-stringent condition: 5 x SSC, 0.5% SDS at 65°C for hyb~dization followed by washing in 2 x SSC, 0.5% SDS at 50°C. The solution for prehybridization was 4 x SSC, 1 x Denhardt, 50 mM Tris-HCl at pH 7.5, I mM EDTA, 0.5% SDS, 100 pug of denatured herring sperm DNA per ml, and for hybridization 10% dextran sulphate was added to the prehybridization solution. Nine clones obtained after the third screening were analyzed further. Sequencing DNA restriction fragments were subcloned into the appropriate site of either the M 13 cloning vector or BSK plasmid. Sequencing was carried out with a Sequenase Version 2.0 DNA sequencing kit (US Biochemical Corp., Cleveland, OH) and [a-35S]-dCTP. Cell lines ARH and Wa cell lines were mature B cell lines kindly provided by Dr A. Yano (Nagasaki University, Nagasaki, Japan) (Yano et al., 1989). The cytoplasmic p-positive pre-B cell lines, HPB-null and Nalm-6, and a null cell line, Nalm-16, were obtained by the courtesy of Dr J. Minowada (Fujisaki Cell Center, Okayama, Japan). Ball-l, CEM and U937 were B, T and monocytic cell lines, respectively. RNA isolation and analysis Total RNA from cultured cells was extracted by the method described previously (Chomczynski and Sacchi, 1987). RNA samples were denatured at 65°C with 50% formamide and 2.2 M fo~aldehyde and electrophoresed on 1.2% agarose gels containing 2.2 mol of formaldehyde, Ten microgrammes of RNA was transferred to a nylon membrane (Hybond N, Amersham), and the conditions of hybridization and washing were essentially the same as for the genomic Southern blotting described below. The VpreBl probe was synthesized by the polymerase chain reaction using Taq polymerase (Cetus Corp., Norwalk, CT).
analyzed cDNA inserts from nine clones, There were two groups of clones; the size of one group (6 clones) was 1 kbp and that of the other group (3 clones) was 2.5 kbp. One clone from each group was subcloned into plasmid BSK and they were designated as YM-1 and YM-6. Restriction mapping analysis revealed that they were completely different genes (data not shown). The genomic Southern blots of human and mouse DNA were analyzed first by hybridizing with the 8HS-20 probe at low stringency, then by rehybridizing with the YM-6 probe followed by the YM-1 probe at high stringency. The size of the main band in the human genome hybridized with each clone was not the same as that expected from the results with the 8HS-20 probe. However, a weak signal corresponding to the band hybridized with 8HS-20 was detectable with the YM-1 probe, which suggests that at least YM-1 has weak homology with the human homologue of 8HS-20 (data not shown). Characterization of the YM-1 sequence We determined the nucleotide sequence of the YM-I clone. The sequencing strategy is outlined in Fig. 1, and the complete nucleotide sequence and the predicted amino acid sequence of YM-1 are shown in Fig. 2. The translation initiation codon starts at nucleotide (nt.) 23 from the EcoRI cloning site and continues to the termination codon at nt. 742, This open reading frame is able to encode 240 amino acids and the predicted molecular weight is about 26 kd. There are two ploy(A) signals, at nt. positions 848 and 860. The overall nucleotide sequence homology of YM- 1 and 8HS20 was turned out to be less than 40%. However, from nt. 187 through nt. 225 of YM-I,33 nucleotides were completely matched with the sequence of 8HS20 (87% homology), which suggests the possibility of the cross-hybridization of two genes. Homology search by the computer revealed a maximum homology of the 3’ part of the YM-1 sequence with human J,CAZ gene of the immunoglobulin lambda light chain reported by Yamasaki et al. (Yamasaki et al., 1987). There were four single base changes, at nt. 397 in the J region and at 637 in the C region, which did not result in amino acid change, and at nt. 555 and 737 in the C region, which resulted in an amino acid changes
Genomic Southern blotting Five microgrammes of high mol. wt DNA was digested with EcoRI restriction enzyme, and agarose gel electrophoresis was carried out. DNA was transferred to nylon filters and hybridized with probes under the high-stringent condition; 5 x SSC, 0.5% SDS at 65°C for hybridization, 0.1 x SSC, 0.5% SDS at 65°C for washing.
L
J
V
I EcoRl
BamHl
C
Ilu~ll Smal
i 1 BarnHI
I EcoRl
4
c
I 100 bp
RESULTS AND DISCUSSION Isolation of human 8US-20 homologue We screened a human bone marrow cDNA library using mouse 8HS-20 cDNA as a probe under the low-stringent condition. After the third screening we
Fig. 1. Restriction map of YM-1. L, leader; V, variable; J, joining; C, constant region. The open bar indicates exon and the closed bar shows the untranslated region. Fragments shown below it were subcloned into either the Ml3 phage cloning vector or Bluescript plasmid and the arrows indicate the direction of sequencing.
725
A new subgroup of human Ig V,
from glycine to alanine and from cysteine to arginine, respectiveIy. The 5’ part of the YM-1 nucleotide sequence showed 87.6% homology with the V1 gene of Tl (Bernstein et aE_,1989). The first 19 amino acids were leucine rich, having an overall hydrophobic character, which is a typical structure of the signal peptides. V, genes have been divided into seven subgroups according to amino acid sequences of the V, region (Anderson et a/., 1984). Compa~son of the deduced amino acid sequence of the V region of YM-I with those of representative members of the other subgroups of VAis shown in Fig. 3, in which the sequences were aligned for maxima homology. YM-1 showed considerable st~ctural homology with those Vi,poly~ptides, but had a unique feature of amino acid compositiou. Interestingly, the highest homology was observed with human pre-B all-s~cific gene VpreBl (56%), while the other V, subgroups had 35 to 44% homology. These results indicate the the V region of YM-1 (V~/YM-I) along with Tl belongs to a new subgroup of V, , and that VpreB 1 possibly diverged from this V, subgroup after the diversification of a V, family.
To characterize the V,,WM-1 locus, DNAs from various cell lines were digested with EcoRI, and Southern blot analysis was carried out by probing with Vn/YM-I (EcoRI-BamHI fragment in Fig. 2) under the high-stringent condition. Then the same filter was rehybridized with the Cu gene probe (Smal-Pstl fra~ent in Fig. 2). In EcoRI digests of DNA from various cell lines, four cross-hybridizing bands of V,/YM-1 of 7.6 kb, 4.4 kb, 4.2 kb and 3.9 kb were observed commonly in four of the seven cell lines, as indicated by arrowheads in Fig. 4A (lanes 1,3,4 and 7). Less intense bands of 7.0 kb, 2.8 kb and 2.4 kb also appeared in all cell lines, which suggests that these bands represent the less homologous genes of the YM-1 family. In the monocytic cell
Fig, 2. ~ucleotide
sequence of YM-1, The amino acid sequences deduced are shown above the nucleotide sequence starting at ATG and ending at TAG (double underline). Figures below the nucleotide sequence refer to the nucleotide number starting at the sequence recognized by EcoRXrestriction enzyme. Poly(A) signals are shown by dotted underling. CDR denotes a complementa~ty~e~~ining region. Boxed nucleotides indicate base changes compared to those of the C region of p3C4;1 S (12). Restriction sites used to prepare probes are underlined.
CO_
60
1x1 hIi1
B”R en.
I
SO
70 G G G
homology
100
TSAS-UTAS-TTVT--
44
44 NSRDSS
i:
SSYVDN
40 36 38 56
Fig. 3. Comparison of the predicted amino acid members of seven known subgroups and VpreBl. are shaded and the per cent homology with the calculated. Figures above the amino acid
sequence of Amino acid total number sequence are
YM-1 with those of representative sequences homologous with YM-1 of amino acids in this figure was based on the V,, sequence.
726
et (11.
Y.Tsmmsucu-YofcoT~
1234567
1234567 kb
kb
23
23
9.4
6.6
6.6
4.4
2.3
2.3
Fig. 4. Southern blot analysis of different cell lines under the stringent condition. A, The filter was hybridized with a V,/YM-1 probe. B, The same filter was washed and rehybridized with a C,, probe. Lane 1, CEM; lane 2, U937; lane 3, ARH; lane 4, Wa; lane 5, HPB-null; lane 6, Nalm-6; lane 7, Nalm-16. The size marker is lamda DNA digested with Hind III. Arrowheads indicate the main bands observed in common five of the seven cell lines.
line U937 (lane 2) and the pre-B cell line HPB-null (lane 5), a 7.3 kb instead of a 7.6 kb band was observed and an additional 3.6 kb band was detected in the U937 cell line. In another pre-B cell line, Nalm-6 (lane 6) both 7.6 and 7.3 kb bands were observed in addition to a 3.6 kb band. Meanwhile the pattern of Southern blot hybridization of CI genes showed the germ-line configuration, which is 16 kb, 14 kb and 8 kb bands in EcoRI-digested DNA (Taub et al., 1983), in most of the cell lines. Two pre-B cell lines, HPB-null and Nalm-6, showed rearrangement of the 8 kb band containing Cnz (Fig. 4B, lanes 5 and 6). However, none of the rearranged C, bands hybridized with VJYM-1 (Fig. 4A). From these results taken together, we assume either that an abnormal rearrangement has occurred in U937 and the two pre-B cell lines or, more likely, that the 7.6 kb, 7.3 kb and
3.6 kb bands reflect the polymorphism of the YM-1 family. The genomic structure of VJYM-1 must be determined to know the exact nature of these patterns of Southern blot hybridization. Both C, and V,/ YM- 1 are expressed independently as a sterile message in immature and mature B cell lines 1. A sterile C, message. The expression of YM-1 in the cell lines used for Southern blot analysis and an additional mature B cell line, Ball- 1, was investigated by Northern blot analysis. As shown in Fig. 5A, the C,, probe revealed abundant expression of C, in Wa, HPBnull, Nalm-6 and Nalm-16 cell lines (lanes 5,6, 7 and 8) and weak expression in the U937 cell line (lane 2). The major C, transcript (0.95 kb) was smaller than mature 1. mRNA (1.2 kb) (Sun et al., 1985). Taking into account
C 12345678
12345678
12345678
-
Fig. 5. Northern blot analysis. The filter was prepared as described in Materials and Methods, and hybridized with probe C,, (A) (Zday exposure), V,/YM-1 (B) (8-day exposure) or VpreBl (C) (l-day exposure). Lane 1, CEM; lane 2, U937; lane 3, Ball-l; lane 4, ARH; lane 5, Wa; lane 6, HPB-null; lane 7, Nalm-6; lane 8, Nalm-16.
727
A new subgroup of human Ig V,
the results of Southern blot analysis of the C1 genes, we consider that these C1 transcripts are sterile messages. While very little is known about the sterile C1 message, the sterile message is characterized as non-functional transcriptional activation of unrearranged Cp (Kemp et al., 1980; Alt et al., 1982) or C, (Perry et al., 1980) gene and is observed even in T and myeloid cells. We observed the sterile C, message in the CEM (T) cell line and additional heterogenous messages in Wa, HPB-null, Nalm-6 and Nalm- 16 cell lines after long exposure of the film (data not shown), only one of which is visible in Fig. 5A, lane 7. Thus, the activation of the C, gene appears to be initiated early in a progenitor cell of the T, B, and myeloid lineages as reported in the case of Cp sterile messages (Kemp et al., 1980). However, the origin of transcription and enhancer/promoter region of the sterile Cn message is not known. 2. A sterile Vn message. When the same filter was rehybridized with a V1/YM-1 probe, a low but detectable level of a l.O-kb transcript was observed in Wa (lane S), HPB-null (lane 6), Nalm-6 (lane 7), and Nalm-16 (lane 8) (Fig. 5B). The mature B cell (Wa) showed a higher level of expression than the immature cells. These V, messages were not the result of the gross-hybridization with the transcript of VpreBl, because the Vpre Bl message was smaller (0.8 kb) than that of V,/YM-I and observed only in immature B cells (lanes 68), not in Wa (lane 5) (Fig. 5C). Since these transcripts do not appear to be derived from the rearranged V, gene, we conclude that they are germ-line V, messages (sterile V, messages). Expression of the unrearranged V region of the heavy chain (vu) is limited to the early stage of the B-lymphocyte differentiation pathway (Yancopoulos and Alt, 1985) and is temporarily detected in murine fetal liver (Lennon and Perry, 1990). In this case, the lack of unrearranged V, expression in mature cells is thought to be due to the inactivation of a locus which was previously active (Yancopoulos and Alt, 1985). In contrast, the unrearranged V, locus appears to be activated along with B cell maturation. Although the germ-line ?J1 message in humans is not known, studies by Picard and Shaffner (Picard and Shaffner, 1984) on various murine cell lines showed that the unrearranged Vi region was transcribed at a low but steady-state level of l-10 RNA molecules per cell, but only in myelomas and hybridomas, not in cell lines representing the pre-B cell and immature B cell stage. In the present study on human cell lines we were able to detect germ-line V, messages even in the null cell as well as immature B cells albeit at much lower level than in the mature B cell. This difference may reflect the fact that mouse Ig contains only 5% 1 IgL chain while human Ig is 40% ;1.type. Our novel finding in humans indicates that the chromatin structure of V, locus is already open before the synthesis of p chain in a cell of the B lineage, which is also suggested by Leclercq et al. (1989) in their study of a germ-line J, transcript. Whether a germ-line V, transcript similar to YM-1 is also expressed in other VA subgroups is unknown.
Further study is required to know the regulatory mechanism for the activation of the V, gene during B cell maturation. Acknowledgement-This work was supported by a Research Grant for Aging and Health from the Ministry of Health and Welfare.
REFERENCES Ah F. W., Rosenberg N., Enea V., Siden E. and Baltimore D. (1982) Multiple immuuoglobulin heavy-chain gene transcripts in Abelson murine leukemia ~rus-t~nsfo~ed lymphoid cell lines. Molec. cell. Biol. 2, 386-400, Anderson M. L. M., Szajnert M. F., Kaplan J. C., McCall L. and Young B. D, (1984) Cloning and sequence analysis of an Ig 2 light chain mRNA expressed in the Burkitt’s lymphoma cell line EB4. Nucl. Acids Res. 12, 6647-6661. Bauer T. R., Jr and Blomberg B. (1991) The human A L chain Ig locus: Recharacterization of JC16 and identification of a functional JC17. J. Zmmun. 146, 2813-2820. Berinstein N., Levy S. and Levy R. (1989) Activation of an excluded immunoglobulin allele in a human B lymphocyte cell line. Science 244, 337-339. Chomczynski P. and Sacchi N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanate-phenolchloroform extraction. Anafyt. Biochem. 162, 156-I 59. Cotner T. and Eisen H. V. (1978) The natural abundance of y2-fight chains in inbred mice. J. exp. Med. 148, 1388-1399. Fu S. M., Chiorarri N. and Kunkel H. G. (1976) Differentiation capacity and other properties of the leukemic cells of chronic lymphocytic leukemia. Zmmun. Rev. 48, 23-44. Heiter P. A., Hollis G. F., Korsmeyer S. J., Waldmann T. A. and Leder P. (1981) Clustered arrangement of immunoglobulin t constant region genes in man. Nature 294, 536-540. Kemp D. J., Harris A. W. and Adams J. M. (1980) Transcripts of the immunogIobulin Cp gene vary in structure and splicing during lymphoid de~lopment. Proc. natn. Acad. Sci. U.S.A. 77, 7~7~. Kudo A. and Melchers F. (1987) A second gene, Vpre B in the 15 locus of the mouse, which appears to be selectively expressed in me-B lymphocytes. EMBO J. 6, 2267-2272. Leclercq L., Butkeraitis P. and Reth M. (1989) A novel germ-line J, transcript starting immediately upstream of J, 1. Nuci. Acids Res. 17, 6809-6819. Lennon G. G. and Perry R. P. (1990) The temporal order of appearance of transcripts from unrearranged and rearranged Ig genes in murine fetal liver. J. Zmmun. 144, 1983-1987. Perry R. P., Kelley D. E., Coleclough C., Seidman J. G., Leder P., Tonegawa S., Matthyssens 6. and Weigert M. (1980) Transcription of mouse ICchain genes: Implications for allelic exclusion. Proc. rtafn. Acad. Sci. U.S.A. 77, 1937-1941. Picard D. and Shaffner W. (1984) Unrearranged immunoglobulin R variable region is transcribed in K-producing myelomas. EMBO J. 3, 3031-3035. Sakaguchi N. and Melchers F. (1986) 15, a new light chainrelated locus selectively expressed in pre-B lymphocytes. Nature 324, 579-582. Sun L.-H. K., Croce C. M. and Showe L. C. (1985) Cloning and sequencing of a rearranged V, gene from a Burkitt’s
72x
Y.
TSUNETSUGU-Y0K0T.4
lymphoma cell line expressing kappa light chains. Nucl. Acids Res. 13, 49214934. Taub R. A., Hollis G. F., Heiter P. A., Korsmeyer S., Waldmann T. A. and Leder P. (1983) Variable amplification of immunoglobulin i light-chain genes in human populations. Nature 304, 1722174. Vasicek T. J. and Leder P. (1990) Structure and expression of the human immunoglobulin i genes. J. exp. Med. 172, 6099620. Yamasaki N., Komori S. and Watanabe T. (1987) Comp-
et al.
lementary DNA for a human subgroup IV immunoglobulin I-chain. Molec. Immun. 24, 981-985. Yancopoulos G. D. and Alt F. W. (1985) Developmentally controlled and tissue-specific expression of unrearranged V, gene segments. Cell 40, 271-281. Yano A., Aosai F., Ohta M., Hasekura H., Sugane K. and Hayashi S. (1989) Antigen presentation by Toxoplasma Gondii-infected cells to CD4 + proliferative T cells and CD8 + cytotoxic cells. J. Purasitol. 75, 411-416.