Studies of Class I Genes in the Major Histocompatibility Complex of the BALB/c Mouse Citation Sher, Beverly Taylor (1985) Studies of Class I Genes in the Major Histocompatibility Complex of the BALB/c Mouse. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/4pj0-z147. https://resolver.caltech.edu/CaltechTHESIS:01312019-165301208 Abstract This thesis contains the results of investigations into the structure and organization of Class I genes in the major histocompatibility complex of the BALB/c mouse. In the body of the thesis, the sequence of the BALB/c H-2D d transplantation antigen gene is presented. This is the first complete sequence of an H-2D d gene and is the only genomic sequence to be in full agreement with the available protein sequence. The H-2D d gene sequence has been used to predict the protein sequence of the H-2D d molecule, which has been compared to the protein sequences of other Class I molecules. The H-2D d protein sequence is no more related to that of its closely linked partner, H-2L d , than it is to the sequence of its presumptive allele, H-2D b , or to the sequence of the H-2K b molecule, which is from not only another H-2 haplotype but another genetic subregion. The sequence differences between these transplantation antigens are spread throughout the molecules in a mosaic pattern that may have arisen,in part, from small gene conversion events. No obvious evidence of any recent gene conversion event affecting the H-2D d gene was observed, however. Three Class I genes have been cloned and mapped to the H-2D subregion in BALB/c. These include gene 16.1, whose product has not been identified; the H-2D d gene; and the H-2L d gene. There is serological evidence for the existence of additional H-2D-encoded transplantation antigen molecules in BALB/c, but no genes encoding these products have been identified. The sequence of the H-2D d gene contains potential alternative splice sites in and around the exon encoding the first external domain. use of these splice sites could generate a transplantation antigen molecule with different serological determinants, and might help to resolve the discrepancy between the number of H-2D-subregion Class I genes and the number of serologically defined H-2D-subregion transplantation antigens. The appendices contain the results of a number of studies related to Class I gene organization and function. Appendix A contains the sequence of gene 27.1. This gene, also known as the Q8 gene, was identified as a Qa pseudogene based on the presence of termination codons in inappropriate locations in its sequence. Appendix B contains the sequence of the H-2L d gene, which was the first transplantation antigen gene to be sequenced. Appendix C contains the results of DNA-mediated gene transfer experiments that Identified genomic clones containing the H-2Kd, H-2Ld, and H-2Dd transplantation antigen genes as well as Class I genes encoding the Qa2,3 molecule and two different TL differentiation antigen genes. Appendix D contains the results of calculations of protein sequence homology between different Class I molecules. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: Biology Degree Grantor: California Institute of Technology Division: Biology Major Option: Biology Thesis Availability: Public (worldwide access) Research Advisor(s): Hood, Leroy E. Thesis Committee: Davidson, Norman R. (chair) Lewis, Edward B. Owen, Ray David Meyerowitz, Elliot M. Hood, Leroy E. Defense Date: 27 September 1984 Funders: Funding Agency Grant Number NSF UNSPECIFIED NIH UNSPECIFIED Caltech UNSPECIFIED Record Number: CaltechTHESIS:01312019-165301208 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:01312019-165301208 DOI: 10.7907/4pj0-z147 Related URLs: URL URL Type Description https://doi.org/10.1016/0092-8674(81)90175-6 DOI Article adapted for Appendix A https://doi.org/10.1126/science.7058332 DOI Article adapted for Appendix B https://doi.org/10.1038/300231a0 DOI Article adapted for Appendix C Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 11374 Collection: CaltechTHESIS Deposited By: Mel Ray Deposited On: 01 Feb 2019 19:36 Last Modified: 16 Apr 2021 22:12 Thesis Files Preview PDF - Final Version See Usage Policy. 25MB Repository Staff Only: item control page

Stud1es of Class I Genes in the Major Histocompatibility Complex of the BALB/c Mouse Thesis by Beverly Taylor Sher In Partial Fulfilbnent of the Requirements For the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1984 (submitted September 27, 1984) Th1s thesis 1s dedicated to Ray D. Owen in gratitude for his advice and encouragement. iii Acknowledgements Many people helped to make my Cal tech expenence both intellectually and personally rewarding. Lee Hood, my I would like to thank: thesis adviser, for creating a stimulating environment in which to work; Norman Davidson, Ed Lewis, Elliot Meyerowitz, and Ray OWen, for serving on my thesis committee; Tommy Douglas, carol Sibley, and Ellen Rothenberg, for instructing me in the gentle art of teaching; Martha Bond, Kathryn calame, Richard Douglas, Kurt Eakle, John Fre 1 inger, Bob Goodenow, Steve Hunt, Donna Li vant, Minnie McMillan, Kevin Moore, Sue Schiffer, J1m Schilling, Iwona Stroynowski, and Hoodlums too numerous to mention, for their help and friendship; Edw1n H. McConkey, for introducing me to biological research; Bertha Jones, Maria De Bruyn, Jessie Smith, Connie Katz, Patty Bateman, Isabella Lubonnrski, and Bernita Lewis, for keeping things working; the NSF, NIH, and caltech, for financial support; and finally, my parents, Phyllis and Richard Taylor; my inlaws, Ralph and Judy Sher; and my husband, Marc Sher, for making it possible. iv Abstract This thesis contains the results of investigations int~ the structure and organization of Class I genes in the major histocompatibility complex of the BALB/c mouse. In the body of the thesis, the sequence of the BALB/c H-2Dd transplantation antigen gene is presented. This is the first complete sequence of an H-2Dd gene and is the only genomic sequence to be 1n full agreanent with the available protein sequence. The H-20d gene sequence has been used to predict the protein sequence of the H-20d molecule, which has been compared to the protein sequences of other Class I molecules. The H-20d protein sequence is no more related to that of its closely linked partner, H-2Ld, than it is to the sequence of its presumptive allele, H-20b, or to the sequence of the H-2Kb molecule, which is from not only another H-2 haplotype but another genetic subregion. The sequence differences between these transplantation antigens are spread throughout the molecules in a mosaic pattern that may have arisen,in part, from small gene conversion events. No obvious evidence of any recent gene conversion event affecting the H-20d gene was observed, ho~ver. Three Class I genes have been cloned and mapped to the H-20 subregion in BALB/c. These include gene 16.1, whose product has not been identified; the H-2Dd gene; and the H-2Ld gene. There is serological evidence for the existence of additional H-20encoded transplantation antigen molecules in BALB/c, but no genes encoding these products have been identified. The sequence of v the H-2Dd gene contains potential alternative splice sites in and around the exon encoding the first external domain. use of these splice sites could generate a transplantation antigen molecule with different serological determinants, and might help to resolve the discrepancy between the number of H-20-subregion Class I genes and the number of serologically defined H-20-subregion transplantation antigens. The appendices contain the results of a number of studies related to Class I gene organization and function. contains the sequence of gene 27.1. Appendix A This gene, also known as the QB gene, was identified as a Qa pseudogene based on the presence of termination codons in inappropriate locations in its sequence. Appendix B contains the sequence of the H-2Ld gene, which was the first transplantation antigen gene to be sequenced. Appendix C contains the results of DNA-mediated gene transfer experiments that Identified genomic clones containing the H-2Kd, H-2Ld, and H-2Dd transplantation antigen genes as well as Class I genes encoding the Qa2,3 molecule and two different TL differentiation antigen genes. Appendix D contains the results of calculations of protein sequence homology between different Class I molecules. vi Table of Contents Acknowledgements iii Abstract iv List of Figures and Figure Legends vii Chapter I. Introduction. 1 Chapter II. DNA Sequence of the Mouse H-2Dd Transplantation Antigen Gene. 11 Appendix A. DNA Sequence of a Class I Pseudogene. 35 Appendix B. DNA Sequence of the H-2Ld Gene. 47 Appendix C. Identification of Class I Genes. 58 Appendix D. Hanologies Between Class I Prote1n Sequences. 66 vii Figures and Figure Legends Figure Legends, Chapter I 7 The Structure of the Mouse MHC 8 Structure of a Class I Molecule 9 Figure Legends, Chapter II 31 Sequencing Strategy for the H-2Dd Gene 32 Sequence of the H-2Dd Gene 33 Campar1sons of Class I Protein Sequences 34 Corrected Sequence of the H-2Ld Gene 56 Corrections Made in the H-2Ld Gene Sequence 57 1 Chapter I Introduction 2 The major histocompatibility complex (MHC) is made up of a group of tightly linked loci whose prooucts are involved i_n the immune response (1-4). Such genetic regions have been identified in all vertebrates examined so far. It has even been suggested that the colonial tunicate, Botryllus, a protochordate, may possess a genetic region in which loci with MHC-like functions are linked (3). Thus, the association of loci whose products are involved in imnune processes is evolutionarily ancient. The major histocompatibility complex of the mouse was the first MHC to be discovered and has been extensively characterized (1,2,4). It has been divided into three major genetic regions: the H-2 canplex, the Qa region, and the Tla canplex, as diagrammed in Figure 1. The H-2 complex itself has four subre- gions: H-2k, H-2I, H-2S, and H-20. Each genetic region contains several loc1, many of wh1ch are polymorphic. Inbred strains of mice possess different sets of MHC alleles. These constellations of alleles are referred to as haplotypes. Three major classes of proteins are encoded by genes in the MHC: Class I, Class II, and Class III (1,2,4). Class I proteins can be separated into two categories: the Qa and TL differentiation antigens are encoded by genes in the Qa and Tla regions, while the transplantation antigens are encoded by genes in the H2K and H-20 subregions. The Ia roolecules, also known as the immune response-associated antigens, are Class II roolecules encoded by genes in the H-2I subregion. Class III proteins include several canplerent canponents as well as other serum proteins and 3 are encoded by genes in the H-2S subregion. As this thesis is concerned pnmarily with transplantation antigen genes, the Class II and Class III rrolecules will not be further discussed. Class I molecules share similar structures but differ in tissue distribution and extent of serological polymorphism (1,2,4,5). These rrolecules are "'45,000 dalton integral membrane glycoproteins found noncovalently associated with{3-2 micro- globulin, a 12,000 dalton polypeptide encoded by a gene outside the MHC. The structure of a typical Class I roolecule is shown in Figure 2. They are divided 1nto three external domains, each of "'90 amino acids; a hydrophobic transmembrane segment; and a carboxy-terminal cytoplasmic segment. resembles both the The third external danain p-2 microglobulin molecule and irrrnunoglobulin danains, leading to the suggestion that Class I rrolecules, p-2 microglobulin, and irrmunoglobulins are evolutionarily related (l-4). The Tla and Qa different1ation antigens are only moder- ately polymorphic and have a limited tissue distribution, being found on the surfaces of particular hematopoietic cells. The transplantation antigens, in contrast, are found on the surfaces of practically all sanatic cells and are among the most polymorphic molecules known. There are over 40 serologically de- tectable alleles at the H-2K locus and as many alleles at the H20 locus (4). The functions of the different types of Class I roolecules also differ. function. The Qa and Tla differentiation antigens have no known The functions of the transplantation antigens better understood. are They are strongly antigenic and serve as 4 prlinary targets in the rejection of grafts between mice of different H-2 haplotypes. In addition, their presence is required for cytotoxic T cell recognition of viral or tumor antigens on the surfaces of vually infected or transformed cells. Cytotoxic T cells of the proper specificity from mice of a particular H-2 haplotype can kill virally infected cells only i f these cells express H-2K or H-2D-encoded transplantation antigens of the same haplotype (6). Thus, cytotoxic T cells must have a dual speci- ficity: they must be able to recognize not only a particular foreign antigen, but also self H-2K or H-20 molecules. non is known as H-2 restriction. This phenome- It is unclear whether the cyto- toxic T cell has two receptors, one for self H-2 rrolecules and one for antigen, or whether cytotoxic T cells use a single receptor to recognize a combination of the two. Other molecules have been implicated in the interaction between cytotoxic T cells and their targets. For example, Lyt-2, a T cell-specific molecule, may help to stabilize the interaction between the T cell receptor(s), the transplantation antigen, and the foreign antigen (7). Mice of different H-2 haplotypes generally express at least one H-2K-encoded transplantation ant1gen, known as H-2Kx, and at least one H-2D-encoded transplantation, known as H-2Dx, where x is the H-2 haplotype designation. For example, C57Bl/10 m1ce (H-2b hap- lotype) express two transplantation antigens, H-2Kb and H-2Db (5). BALB/c m1ce (H-2d haplotype) express at least three trans- plantation antigens: H-2~, H-2Dd, and H-2Ld. H-2~ is encoded by the H-2K subregion, and H-2Dd and H-2Ld are encoded by genes in the H-2D subregion. All three of these transplantation antigens 5 have been purified and partially sequenced (5). In addition to these well-defined transplantation antigens, there may be more transplantation antigens encoded by H-2d haplotype genes. - In particular, there is serological evidence that the H-2D subregion in BALB/c mice may encode several other transplantation antigens, including H-2~ and H-2Rd (8,9). Class I eDNA probes have been used to determine the number of Class I genes in different inbred strains of mice and to isolate genomic clones containing Class I genes from recombinant libraries. Although the number of Class I genes appears to vary among different inbred strains, there are 25-40 Class I genes per inbred strain. Approximately 30 distinct Class I genes have been isolated from BALB/c genom1c libranes (10,11). These genes have been characterized by two different approaches: DNA-mediated gene transfer and restriction polymorphism mapping (10-13). In the first approach, a Class I gene is introduced into rrouse L cells (H-2k haplotype) by the calcium phosphate technique. The resulting transfectants are then screened by radioimmunoassay w1th a panel of monoclonal antibodies spec1fic for different Class I molecules, as described in Appendix C. In the second approach, low-copy DNA probes are obtained from the sequences flanking each distinct Class I gene. These probes are then used 1n Southern blots of genam1c DNA from various inbred, congenic, and recombinant congenic strains of mice. By correlating the observed bands with the known recombination points in these strains, it is possible to map a particular probe to a locus within the MHC. This presumably also locates the Class I gene 6 associated with the probe sequence. In BALB/c, using these approaches, two Class I genes have been mapped to the H-2K subregion; three genes have been mapped to the H-2D subregion, as discussed in Chapter 2; and the rest of the cloned Class I genes have been mapped to the Qa aoo Tla regions (13). In contrast to the situation in the H-2d haplotype, only one H2Db haplotype Class I gene mapping to the H-20 subregion has been cloned (14). This gene encodes the H-2Db protein. Canpa.risons of the protein sequences of the H-20- encoded transplantation anti- gens from the H-2d and H-2b haplotypes have brought out some interesting points. Most transplantation antigens are only 80%- 85% h:xnologous to each other at the protein level (5). This is true even for the products of allelic genes, such as H-2~ and H2Kb. This was also true in comparisons between the 183 residues of partial H-2Dd protein sequence and the H-2Ld and H-2Db protein sequences (15). Surprisingly, the protein sequences of H-2Ld and H-2Db are 96% horrologous to each other (15). This unprecedented level of hanology has led to the suggestion that the H-2Ld and H2Db genes must be allelic, and that the H-2Dd gene must be derived from a different locu~ To elucidate the relationship between the H-2Dd protein and the other H-20-encoded transplan- tation ant1gens, it is necessary to know the sequence of the H2Dd gene. This sequence and its relationship to the other H-20- region transplantation antigen gene sequences are presented in Chapter II. 7 Figure 1. The major histocanpatibility canplex of the mouse is diagrammed in this figure. The first line indicates that this canplex maps to chromosome 17. The next line indicates protein- coding loci; the class of protein produced by each locus is indicated on the third line; the fourth line indicates the boundaries of the H-2 and Tla complexes; the fifth line shows the regions into which these canplexes are divided; and the last line gives the genetic distances between these regions. Figure 2. This figure shows the domain structure of a Class I molecule. rK 1, 0<.2, and o<3 are the three external domains of the protein. o<l contains the amino terminus. o(3 associates with? -2 rn1croglobul in, which is designated ,B-2rn in the figure. The cytoplasmic tail of the rrolecule contains the carboxyl terminus. RECOMBINATION FREQUENCIES (eM) REGION 1K1 I H-2 n n nn 1 CLASS COMPLEX Ea 'lf AaE/J K LOCI S 11 Figure 1. 11 mm n Qo l Qo-2,3 11 Tlo l Tlo Tlo l Qo-1 1 ·------1.5---- 0 I I I (C4, Sip) (O,L,R) CHROMOSO~I7·~ MOUSE 00 9 CLASS I Figure 2. 10 References 1. Snell, G.D. (1981) Science 213, 172-178. 2. Hood, L., Steinmetz, M., & Mal issen, B. (1983) Ann. Rev. Lmmunol. 1, 529-568. 3. Scofield, v. L., Schlllllp::>erger, J. M., & Weissman, I. L. (1982) Amer. Zool. 22, 783-794. 4. Klein, J. (1979) Science 283, 516-521. 5. Nathenson, s. G., Uehara, H., Ewenstein, B. M., Kindt, T. J., & Coligan, J. E. (1981) Ann. Rev. Biochem. 59, 1025-1052. 6. Z inkernagel, R. M., & Doherty, P. C. (1980) Adv. Immunol. 27, 51-177. 7. Swain, S. L. (1983) Immun. Rev. 74, 129-142. 8. Ivanyi, D., & Dernant, P. (1979) Immunogenetics 8,539-550. 9. Hansen, T. H., Ozato, K., Mel ino, M. R., Col igan, J. E., Kindt, T. J., Jandinski, J. J., & Sachs, D. H. (1981) J. Immunol. 126, 1723-1716. 10. Steinmetz, M., Winoto, A., Minard, K., & Hood, L. (1982) Cell 28, 489-498. 11. Hunt, s., F1sher, D., personal communication. 12. Goodenow, R. S., McMillan, M., Nicolson, M., Sher, B. T., Eakle, K. A., Davidson, N., & Hood, L. (1982) Nature 399, 231-237. 13. Winoto, A., Steinmetz, M., & Hood, L. (1983) Proc. Natl. Acad. Sci. USA 89, 3425-3429. 14. Weiss, E. H., Golden, L., Fahrner, K., Mellor, A., Devlin, J. J., Bullman, H., Tiddens, H., Bud, H., Nature, in press. & Flavell, R. A. ll Chapter II. DNA Sequence of the Mouse H-2Dd Transplantation Antigen Gene This manuscript will be submitted to The Proceedings of the National Academy of Sciences. 12 Classification: Genetics DNA sequence of the mouse H-2Dd transplantation antigen gene (transplantation antigens/Class I genes/H-2Dd protein sequence) BEVERLY TAYLOR SHER, RODERICK NAIRN*, JOHN E. COLIGANt, and LEROY E. HOOD Division of Biology, California Institute of Technology, Pasadena, CA 91125 *Department of Microbiology and Immunology, University of Michigan ' p Medical School, 6643 Medical Science Building II, Ann Arbor, Michigan 48109. • tlaboratory of Immunogenetics, Building 5, Room Bl-04, NIAID, NIH, Bethesda, Maryland 20205. 13 ABSTRACT The inbred BALB/c mouse has three transplantation antigens, H2-Kd, H2-L d, and H2-Dd. We present the complete nu<;leotide sequence of the H2-Dd gene as well as 77 residues of previously unpublished H-2Dd protein sequence. These data complete the sequences of all the BALB/c transplantation antigen genes and permit detailed comparison with each other and with their counterparts from the inbred C57Bl/ 10 mouse. Transplantation antigens may differ from one another by as much as 5-15% of their amino acid sequence for the external domains. These extensive differences may arise by gene conversion. The H-20 region of the BALB/c mouse encodes the H2-Dd and the H2-L d genes. Serologic data suggests that at least two additional transplantation antigen molecules, H2-Rd and H2-Md, are encoded in the H-20 region of the major compatibility complex. Paradoxically, gene cloning studies have only ident ified the H2-Dd and the H2-L d genes in the H-20 region. A complete DNA sequence of the H2-Dd gene shows that a variety of alternative splice sites exist throughout the gene which may lead to additional gene products and may explain the multiplicity of H-20encoded polypeptides. 14 The major histocompatibility complex of the mouse (MHC), located on chromosome 17, has four regions which contain class I genes: _ H-20, H-2K, Qa, and the Tla complex (1-4). Class I molecules are heterodimers consisting of a 45,000 dalton integral membrane glycoprotein noncovalently associated with aTmicroglobulin, a 12,000 dalton polypeptide encoded by a gene on chromosome 2. These molecules can be divided into two categories based on differences in their expression, the extent of their serologically detectable polymorphism, and their functions. The class I heavy chains encoded by the Tla complex are differentiation antigens of unknown function which are expressed on the surfaces of certain hematopoietic cells, and are only moderately polymorphic. The transplantation antigens, encoded by genes in the H-20 and H-2K regions, are among the most polymorphic molecules known, as over 50 different alleles have been described for each of the H-20 and H-2K loci so far (3). Different strains of mice possess different sets of alleles at MHC loci. These constellations of alleles are known as haplotypes. Transplantation antigens are ~fgund on the surfaces of virtually all somatic cells. Cytotoxic T cells recognize cell-surface viral or tumor antigens only in the context of a self transplantation antigen, a phenomenon known as H-2 restriction (5). Transplantation antigens are therefore also known as restriction elements. Recently, a number of transplantation antigen genes from several different haplotypes have been isolated and characterized (6-14). Two different transplantation antigen genes, the H-2Kb and H-2Db genes, have been isolated from C57Bl/10 mice (H-2b haplotype), and three 15 transplantation antigen genes, the H-2Kd, H-2L d, and H-20d genes, have been isolated from BALB/c mice (H-2d haplotype). The H-2K genes_rhap to the H-2K region of the MHC, while the H-20 and H-2L genes map to the H-20 region ( 15). In BALB/c mice, there is serological evidence that the H-20 region may encode other transplantation antigens, including H-2Md and H-2R d ( 16, 17). Genes encoding these proteins have not yet been isolated. The H-20-region transplantation antigens H-20d, H-2L d, and H-20b are particularly interesting because of their structural interrelationships. Most transplantation antigens are only 80-85% homologous to each other at the protein sequence level(l-4). This is true for presumptive allelic gene products such as H-2Kb and H-2Kd. This is also true in comparisons between the 183 residues of partial H-20d protein sequence and the translated coding sequences of the H-20b and H-2Ld genes (9,11,18,19). Surprisingly, the protein sequences of H-2L d and H-20b are 95% .. homologous to each other. This unprecedented level of homology has led to the suggestion that the H-2L d and H-2Db genes are allelic (14,19), and that H-20d gene may be derived from a different locus. However, only one H-2D subregion H-2b haplotype Class I gene has been cloned, namely, the H-2Db gene (8). Thus, the number of Class I genes in the H-20 subregion of different haptotypes is variable, and the allelic relationships between the H-20 subregion Class I genes of different haplotypes is unclear. In order to shed light on the relationship between H-20d and the other H-20-encoded transplantation antigens, it is necessary to know the structure of the H-20d 16 gene. In this publication, we present substantial new H-2Dd protein sequence and the complete DNA sequence of the H-2Dd gene. MATERIALS AND METHODS Cosmid clone c49.2 was isolated from a BALB/c Cum sperm DNA cosmid library as previously described (12,20). The construction of Ml3 mp8 subclones and dideoxy sequencing were performed as described (9,21 ,22). Protein sequencing was performed as described (23,24). RESULTS The Class I Gene In Cosmid Clone c49.2 Encodes H-2Dd. Gene 49.2, whose structure and sequence are shown in Figs. 1 and 2, has been shown to encode H-2Dd on the basis of several criteria. Previously, it has been shown that mouse L cells (H-2k haplotype) transfected with c49.2 express a new protein which reacts with the H-20d-specific monoclonal antibodies 34-5-8 and 34-2-12 (25). Comparison of the translated coding sequence of gene c49.2 with the available protein sequence also supports the cbrklusion that this is an H-2Dd gene. The partial protein sequence of the H-2Dd molecule is present~ in Fig. 3. This sequence includes 77 previously unpublished residues, eight of which are still tentatively identified. In addition, 10 residues which had previously been published as tentative are now confirmed. The translated gene c49.2 sequence is in perfect agreement with all of the available H-2Dd protein data (254 of 254 residues). 17 The intron/exon structure of the c49.2 H-2Dd gene presented in Figs. 1 and 2 was proposed on the basis of comparison with the available H-2Dd protein sequence and with the DNA sequences of two cDNAs, pH-2I (26) and pH-2d-l (27). Each eDNA is >99% homologous to the coding sequence of the H-2Dd gene. Therefore they were probably derived from H-2d gene transcripts. The H-2Dd gene has the same general exon/intron structure as the other murine class I genes which have been characterized (6-14,28). The first exon encodes a hydrophobic signal peptide which is not present in the mature protein. The second, third and fourth exons encode the three external domains of the H-2Dd protein, a 1, a2 and a3. The third and fourth exons are separated by a large intron. The fifth exon encodes the transmembrane segment of the H-2Dd protein, and exons 6, 7 and 8 encode the cytoplasmic portion of the molecule. The polyadenylation signal sequence AA TAAA appears 463 nucleotides 3' to the termination codon in exon 8 (29). Thus, as has been observed for other transplantation antigen genes, the intron/exon structure of the H-2Dd gene corresponds precisely with the domain structure of the H-2Dd protein. The H-2Dd Gene Contains Potential Alternative Splicing Signals. Recently, an alternative intron/exon organization affecting the sec"'ond exon has been proposed for the H-2Kd gene (JO). In a eDNA clone apparently derived from a processed H-2Kd transcript, an alternative splice acceptor site in the first intron 50 nucleotides 5' to the usual acceptor site is used. In addition, an extra intron is created by the splicing out of a region inside exon 2 containing amino acids 6-38. In the H-2Dd gene 18 sequence, there is no potential alternative splice acceptor site in the position corresponding to that of the alternative acceptor site in tbe first intron of the H-2Kd gene. There is, however, a potential alternative splice acceptor site in the first intron in frame 51 nucleotides 5' to the usual splice acceptor site for exon 2. There are also potential alternative splicing signals at amino acids 6 and 38 in the H-2Dd gene sequence. It is worth noting that use of the alternative splicing signals would remove the only protein sequence that is conserved between all Class I and Class II B chain sequences, namely the sequence VRFDSD at amino acid positions 34-39 (31). It has been suggested that this sequence could be involved in interactions with T cell molecules. The potential consequences of its removal are, however, unclear. A second region of the H-2Dd gene, the seventh intron, also contains potential alternative splice sites. The H-2Dd eDNA clones pH-2d-l and pH-21 both contain eighth exon sequences that are only five nucleotides is a long and correspond to the exon 8 sequence indicated in Fig. 2. There t • potential alternative splice acceptor sequence located 27 nucleotides 5' to the splice acceptor site used by the H-2Dd cDNAs, as indicated in Fig. 2• • This potential splice acceptor site corresponds to the splice acceptor site used by the H-2Kk, H-2Kb, and H-2Kd messenger RNAs (6,7 ,32,33). The presence of potential alternative splice signals in the H-2Dd gene raises the obvious question of whether they might be used. They are not used by any H-2Dd cDNAs so far described. It is interesting to note that all of the potential alternative splice acceptor sites noted in the 19 H-2Dd gene share the following structure: AG (C/T). All but one of the splice acceptor sites known to be used by H-2Dd transcripts have the structure AG (A/G). The only exception is the acceptor site for exon 8, which has the sequence AG T. Perhaps the AG (C/T) structure is less efficiently used as a splice site by the splicing machinery of the cell than the AG (A/G) structure. From comparisons between the fourth exons of the H-2Dd, H-2L d, H-2Kd, and H-2Kb genes, it is apparent that the frequency of use of a particular splice signal may depend not only on its sequence but also on its surroundings. Exon 4 of the H-2Dd gene encodes 95 amino acids, while the fourth exons of the H-2L d, H-2Kb, and H-2Kd genes encode only 92 amino acids. This results from the absence of a splice donor site in the H-2Dd gene at a position corresponding to the position of the splice donor sites at the ends of the fourth exons of these other genes. The H-2L d, H-2Kd, and H-2Kb genes all have a potential splice donor site which corresponds to the site at the end of the fourth exon of the H-2Dd gene, but appare~ly these potential alternative splice sites are not used. It is worth noting that H-2Kd messages use the rather unusual splice donor site A GT, located • after the 92nd codon in exon 4, rather than splicing at the more common splice donor site G GT which is found nine nucleotides downstream. Thus, in this case, the position of the splice site, rather than its sequence, seems to dictate whether it is used. There is a third eDNA whose DNA sequence is >99% homologous to the coding sequence of the H-2Dd gene. This is pAG64, a eDNA derived 20 from SV40-transformed fibroblasts (34). According to the report, this eDNA represents a transcript that is present only in SV40-transformed cells. Originally, it was not identified as an H-2Dd transcript because its sequence was not identical to that of gene Ch4a-Dd (14). Gene Ch4a-Dd, which has been partially sequenced, was identified as an H-2Dd gene, but disagrees with the H-2Dd protein sequence at 11 positions (see Discussion). The pAG64 sequence differs from the coding sequence of gene c49.2 in only 1/1412 nucleotides. Thus, it is almost certainly derived from an H-2Dd gene transcript. This eDNA is unusual in two respects. It completely lacks exon 7 sequences and it fails to terminate at the polyadenylation site used by pH-2d-1, continuing instead for an extra 150 nucleotides. Other cDNAs derived from the same eDNA library as pAG64, pAG85 and pAG86, are identical in sequence to pAG64 except for the fact that they contain exon 7 sequences (34). Thus, it appears that exon 7 can also be involved in alternative splicing of H-2Dd transcripts. . The H-2Dd Gene Is Homologous to the Other H-Tf HaplotY{>E: Trans. plantation Antigen Genes. Comparisons between the sequence of the H-2Dd gene and the sequences of the other H-2d haplotype genes, the H-2Kd and H-2L d genes, reveal several interesting points. First, the introns are at least as related to each other as are the exons. The introns of H-2Dd gene are 87-9796 homologous to the corresponding intron of the H-2L d and H-2Kd genes, whereas the exons are 85-10096 homologous at the DNA level. This di : fers from the situation in other multigene families, such as the globins, where the introns are much less homologous to each 21 other than are the exons (35, 36). Perhaps the conservation of the introns arises from gene conversion events among these genes, as will be .further discussed. The only significant region of nonhomology between the DNA sequences of the H-2Dd, H-2L d, and H-2Kd introns is associated with sequences in the third intron which are homologous to the consensus sequence of the B 1 highly repetitive sequence family of the mouse (37). The areas of nonhomology result from differences in sequence and in length of the T-rich region which is located next to the Bl sequences. The Bl sequences themselves are located in homologous positions in all three genes, approximately 600 nucleotides 3' to the end of exon 3. From the DNA sequence comparisons, it is also apparent that, for most of the length of the H-2Dd gene, it is not significantly more related to the H-2L d gene than to the H-2Kd gene, despite the fact that the H-2Dd and H-2L d genes are tightly linked and separated from the H-2Kd gene by approximately 0.3 centimorgans(l-3). The only region of the H-2Dd gene to display significantly higher homology to the H-2L d gene than to t~e. H-2Kd gene is the 3' untranslated region. Up to a point 309 nucleotides 3' to the termination codon, the 3' untranslated region of the H-2Dd gene is, if anything, more homologous to the H-2Kd gene than to the H-2L d gene (94% versus 88% homology). After this point, however, the H-2Dd and H-2L d genes, which remain homologous to each other and to the H-2Db eDNA (18) diverge completely from the sequence of the H-2Kd, H-2Kb, QlO (38), and Q8 (28) genes. Ql 0 is a nonpolymorphic Qa gene; Q8 is also known as the 27.1 pseudogene, and is the BALB/c equivalent of the H-2b haplotype Q8 22 gene (8). As was noted in the discussion of the eDNA clone pAG64 (34), the divergence between these two groups of sequences results frqrh the insertion of a sequence which is 84% homologous to the consensus sequence of the 82 repeated sequence family (39). This sequence is present in the 3' untranslated region of the H-2D subregion genes, but not in the other Class I 3' flanking sequences. It is flanked by nine base pair direct repeats, as indicated in Fig. 2. As direct repeats are associated with the insertion of a transposable element, it has been suggested that this sequence represents the result of a transposition of a member of the B2 family (34). This possibility is particularly attractive because the homology to the H-2K and Qa genes resumes after the second direct repeat flanking the 82 sequence. The H-2Dd Protein Sequence Is Homologous to the Protein Sequences of Other Class I Molecules. Comparisons between the translated H-2Dd gene coding sequence and the protein sequences of other class I molecules are presented in Fig. 3. Overall, the H-2Dd protein sequence is ! • 85% homologous to the H-2L d protein, 84% homologous to the H-2Db protein, 80% homologous to the H-2Kd protein, and 86% homologous to the H-2Kb protein. In general, the H-2Dd sequence is more homologous to ·the transplantation antigen sequences than it is to the sequences of the Qaregion molecules. It has been suggested that the observed polymorphism of the transplantation antigens is generated, at least in part, by multiple small gene conversion events (40). Comparison of the H-2Dd gene sequence with 23 the sequences of the other transplantation antigen genes does not reveal any obvious signs of clearcut gene conversion events. Indeed, if the pattern of homologies observed in these comparisons was partially generated by gene conversion, it would be expected that it required many small gene conversion events that would have erased traces of previous events. In addition, even given a region in which two genes share sequence, it is difficult to determine which gene was the donor and which was the recipient in a possible gene conversion event. DISCUSSION The sequence of the H-2Dd gene in c49.2 is the first complete sequence of a H-2Dd gene and the only genomic sequence to be in full agreement with the available H-2Dd protein sequence data. The partial sequence of the class I gene in clone Ch4A-Dd, which was identified as a H-2Dd gene based on the reactivity of its transfected product with H-2Dd-specific monoclonal antibodies, disagrees with the H-2Dd protein sequence a1> 11/142 residues (14). These sequence differences are puzzling in light of the serological data, particularly as five of the changes result in ch'!rge changes. These differences could have arisen by somatic mutation, as clone Ch4A-Dd was derived from a genomic library made from MOPC41 BALB/c tumor cells. It is also possible, of course, that the Class I gene in clone Ch4a-Dd represents a second H-2Dd-like gene. If so, it is a very unusual class I gene, because class I gene introns are as homologous to each 24 other as class I gene exons, and introns 1 and 2 of gene Ch4a-Dd are almost impossible to align with introns 1 and 2 of our H-2Dd gene. It is also possible that the differences in the Ch4A-Dd sequence arose as a result of sequencing errors. The H-2Dd protein sequence predicted from the c49.2 gene sequence is between 80-85% homologous to the protein sequences of other transplantation antigens. In particular, it is 84% homologous to the H-2Db sequence, a level of homology comparable to that found between the only previously sequenced alleles, H-2Kb and H-2Kd (83%). The restriction maps of the H-2Db and H-2Dd genes and their flanking sequences are quite similar, as would be expected for allelic genes (8). Although it has been argued that, based on their unprecedented level of sequence homology (95%), the H-2L d and H-2Db genes are allelic (14,19), it could equally well be argued that the H-2Db and H-2Dd genes are alleles, as they are as homologous to each other as are the H-2Kb and H-2Kd genes, and that the H-2L d gene arose from some unequal crossover event that inserted a copy of a H-2Db-like gene next to the resident H-2Dd gene. In a situation such as this, where there is considerable polymorphism and where there are different numbers of genes in corresponding genetic subregions, it is difficult to determine which pairs of genes are allelic. Three distinct class I genes have been cloned and mapped to the H-2Dd subregion in BALB/c mice. Several clones containing the H-2L d gene have been isolated (9-12). A second H-2D-subregion gene, located on the cosmid cl6.1, has an unknown function. L cells transfected with this 25 gene do not produce any new cell-surface molecules that react with any of the available anti-BALB/c H-20 subregion antisera (13). Thus, this Class I gene could be a pseudogene, could encode a cell-surface protein not detectable in this assay, or could encode a secreted or cytoplasmic class Ilike protein. Several cosmid clones containing H-2Dd genes have been isolated (12,19,41). Their maps fall into two categories, c49.2-like and c18.1-like. The restriction maps of the two types of clones differ only in the 3' sequence flanking the H-2Dd gene. After their maps diverge, there is no DNA homology between the two types of clones. While several c49.2like clones have been isolated, cl8.1 is the only representative clone of its type (41). Based on this fact, as well as on restriction polymorphism mapping and preliminary sequence analysis, it appears that the differences in the maps are the result of a cloning artifact affecting cl8.1 (42). Thus, both types of clone probably contain the same H-2Dd gene. The presence in the H-2Dd gene of alternative splicing signals homologous to those used by other class I genes is intriguing. If U?t;.d, they could generate H-2Dd-1ike proteins that might be useful for novel functions. Use of the potential alternative signals in and around exon 2 of . the H-2Dd gene would generate cell-surface molecules that share some antigenic specificities with H-2Dd, lack other H-2Dd specificities, and have yet others of their own. This could resolve the discrepancy between the serological evidence for the existence of other H-20-encoded molecules in the H-2d haplotype and the lack of any additional cloned class I genes mapping to this region. 26 ACKNOWLEDGEMENTS We thank Donna Livant for helping to check the H-2Dd gene sequence. We thank Marc Sher and the U.C. Irvine Proton Decay Group for computing assistance. We thank Tim Hunkapiller for useful discussions. We thank Connie Katz for typing the manuscript. We also thank Mary Spengler, Robert Valas and Michael Raum for technical assistance. The protein sequence work was supported by Grant IIAI18556 to R.N. The DNA sequence work was supported by Grant #A 19624 to L.H. B. T.S. was supported by N.I.H. Training Grant #GM07616. 27 1. Snell, G.D. (l981)Science 213, 172-178. 2. Hood, L., Steinmetz, M. & Malissen, B. (1983) Ann. Rev. ImmWtol. 1, 529-568. 3. Klein, J. (1979) Science 203, 516-521. 4. Nathenson, S.G., Vehara, H., Ewenstein, B.M., Kindt, T.J., Coligan, J.E. (1981) Ann. Rev. Biochem. 50:1025-1052. 5. Zinkernagel, R.M. & Doherty, P.C. (1980) Adv. Immunol. 27,51-177. 6. Kvist, S., Roberts, L. & Dobberstein, B. (1983) EMBO J. 2, 245-254. 7. Weiss, E., Golden, L., Zakut, R., Mellor, A., Fahrner, K., Kvist, S. & Flavell, R.A. (1983) EMBO J. 12, 453-462. 8. Weiss, E.H., Golden, L., Fahrner, K., Mellor, A., Devlin, J.J., Bullman, H., Tiddens, H., Bud, H. & Flavell, R.A. Nature, in press. 9. Moore, K. W., Sher, B. T., Sun, Y.H., Eakle, K. & Hood, L. (1982) Science 215, 679-682. 10. Evans, G., Margulies, D.H., Carner ini-Otero, R.D., Oza to, K. & Seidman, J.G. (1982) Proc. Natl. Acad. Sci. USA 79, 1994-1998. 11. Sun, Y. H., Sher, B. T ., Hood, L., in preparation. 12. Steinmetz, M., Winoto, A., Minard, K. & Hood, L. (1982) Cell 28, 489498. 13. Goodenow, R.S., McMillan, M., Nicolson, M., Sher, B.T., Eakle, K., Davidson, N. & Hood, L. (1982) Nature 300, 231-237. 14. Margulies, D.M., Evans, G.A., Ozato, K., Camerini-Otero, R.D., Tanaka, K., Appella, E. & Seidman, J.G. (1983) J. Immunol. 130, 463470. 28 15. Winoto, A., Steinmetz, M. &. Hood, L. (1983) Proc. Natl. Acad. Sci. USA 80, 3425-3429. 16. Ivanyi, D. &. Demant, P. (1979) Immunogenetics 8, 539-550. 17. Hansen, T.H., Oza to, K., Meline, M.R., Coligan, J.E., Kindt, T.J., Jandinski, J.J. &. Sachs, D.H. (198l)J. Immunol. 126, 1713-1716. 18. Reyes, A. A., Schold, M. & Wallace, R. B. (1982) Immunogenetics 16, 1-9. 19. Reyes, A.A. &. Wallace, R.B. (1982) Immunogenetics 16, 265-268. 20. M. Steinmetz, unpublished data. 21. Norrander, J., Kempe, T. & Messing, J. (1983) Gene 26, 101-106. 22. Sanger, F., Coulson, A.R., Barrell, B. G., Smith, A.J.H. &. Roe, B. A. (1980) J. Mol. Biol. 143, 161-178. 23. Nairn, R., Nathenson, S.G. &. Coligan, J.E. (1980) Eur. J. Immunol. 10, 495-503. 24. Nairn, R., Nathenson, S.G. & Coligan, J.E. (1981) Biochemistry 20, 4739-4745. 25. Goodenow, R.S., unpublished data. 26. Steinmetz, M., Frelinger, J.G., Fisher, D., Hunkapiller, T., Pereira, D. Weissman, S.M., Uehara, H., Nathenson, S.G. &. Hood, L. (1981) Cell 24, 125-134. 27. Bregegere, F., Abastado, J.P., Kvist, S., Rask, L., Lalanne, J., Garoff, H., Cami, B., Wiman, K., Larhammar, D., Peterson, P.A., Gachelin, G., Kourilsky, P. & Dobberstein, B. (1981) Nature 292, 78-81. 29 28. Steinmetz, M., Moore, K.W., Frelinger, J.G., Sher, B.T., Shen, F.-W., Boyse, E.A. & Hood, L. (1981) Cell 25, 683-692. 29. Proudfoot, N. (1982) Nature 298, 516-517. 30. Lalanne, J.-L., Cochet, M., Kummer, A.-M., Gachelin, G. & Kourilsky, P. (1984) Proc. Natl. Acad. Sci. USA, in press. 31. Malissen, M., Hunkapiller, T. & Hood, L. (1983) Science 221,750-754. 32. Uehara, H., Coligan, J. & Nathenson, S.G. (1981) Biochemistry 20, 5940-5945. 33. Kress, M., Glaros, D., Khoury, G. & Jay,G. (1983) Nature 306, 602604. 34. Brickell, P. M., Latchman, D. S., Murphy, D., Willison, K., Rigby, P. W. J. (1983) Nature 306, 756-759. 35. van den Berg, J., van Ooyen, A., Mantei, N., Schambock, A., Grosveld, G., Flavell, R.A. & Weissman, C. (1978) Nature 275, 37-44. 36. Efstratiadis, A., Posakony, J., Maniatis, T., Lawn, R.M., O'Connell, C., Spritz, R.A., De Riel, J.K., Forget, B., Weissman, S.M., ~l!,ghtom, J.L., Blechl, A.E., Smithies, 0., Baralle, F.E., Shoulders, C.C. & Proudfoot, N.J. (1980) Cell 21, 653-668. 37. Krayev, A.S., Kramerov, D.A., Skryabin, K.G., Ryskov, A.P., Baev, A.A. & Georgiev, G.P. (1980) Nucleic Acids Res. 8, 1201-1215. 38. Mellor, A.L., Weiss, E. H., Kress, M., Jay, G. & Flavell, R.A. ( 1984) Cell 36, 139-144. 39. Krayev, A. S., Markusheva, T. V., Kramerov, D. A., Ryskov, A. P., Skryabin, K. G., Bayev, A. A. & Georgiev, G. P. (1982) Nucleic Acids Res. 10, 7461-7475. 39 40. Bregegere, F. (1983) Biochimie 65, 229-237. 41. Sun, Y.H., unpublished data. 42. Sher, B. T., unpublished data. 31 FIGURE LEGENDS Figure 1. This figure shows the structure of the H-2Dd gene and sequencing strategy used to determine its sequence. A battery of restriction enzymes (Hhal, Sau3A, Sau96A, Rsal, and Alul) was used to generate shotgun mp8 subclones from the 2.2 kb Bam fragment and the 4.2 kb Kpnl fragment containing the 5' and 3' ends of the gene. These subclones were sequenced by the dideoxy method. The locations of Bl and B2 sequences are indicated. The numbers given are distances in nucieotides. Figure 2. This figure contains the sequence of the H-2Dd gene. The protein trans Ia tions of the exons are given above the DNA sequence. Potential alternative splice sites are indicated by arrows. Figure 3. The protein sequences of several class I molecules are compared. The line denoted "H-2Dd pro" gives the partial H-2Dd protein sequences. Sequence identity is indicated by a dot; sequence differences are indicated by the one-letter amino acid code. The symbols on the H-ind pro line are as follows: *:newly published, tentative identification; =: previously tentative, now identified; +:previously unpublished, now identified; published. @:previously published, still tentative; and -:previously 32 J: - .-- E ~ 0 c. ID ::.::: 11 II - D- ""'q,. • N dQ' 2'0 3111 TCT GAT ATG TCT CTC CCA GAT TGT AU G GTG ACACTCTACGGTCTCATTTGCGACGOGCUTCTCCACATGATTCCCTTTCAGCGACTCCCAGUTCTCCTGTGAGTGA 3703 l"-l al Tr• , 3'1 CTCCTCGCTTCTTCCAATGTTCACTTCACAt:'l'CATCCTTCATCACTCTCATTCTCTAG TG TCA ACACACCTCCCTACTCTCCACTTGCTGACAGACAATGTCTTCACACATCTCCTQ 3820 TCACATCCAGAGACCTCACTTCTCTTTAGTCAACTCTCTCATCTTCCCTCTCACTCTGCGCCCTCUACTCAAGAACTGTGCACCCCACTCCACCCCTCCACACCACCACCCTATCCCTC 39'0 CACTCCCCTGTGTTCCCTTCCACACCCAACCTTGCTCCTCCAGCCAAACATTGCTCCACATCTCCACCCTCTCACCTCCATGCTACCCTCACCTTCAACTCCTCACTTCCACACTGAGAA -060 TAATAATTTCAATCTCCCTGCCTCG AGACATCGCTC AGCCCTGAC TCCTCTTCCA AACG TCCTGAGTTC AAATCCCAGCA ACC ACATGGTCCCTC ACA ACCATCTCTA ATCCCATC TAAC ,, 80 ACCCTCTTCTCCAGTGTCTGAACACAGCTACAGTGTACTTACATATAAT~TAAGTCTTTAAAAAATTAATTTGAAAGTGACCTTGATTGTTAACATCTTGAGCT '289 Ser Asp Met Ser Leu Pro Asp Cys Lya Y CT OCT GAC ATC ACC CTC ACC TOG CAC TTG UT COG GAG GAG CTC ACC CAG GU ATC GAG CTT CTG GAG ACC AGO CCT CCA CGO GAT GGA A 2851 hr Pho Gin Lyo Trp Ala Sor Val Val Val Pro Lou Oly Lys Glu Gin Lyo Tyr Thr Cya His Val Clu Hlo Clu Qly Lou Pro Clu Pro L 270 CC TTC CAG UC TCC GCA TCT GTC CTG CTG CCT CTT GOG AAC GAG CAG UG TAC ACA TGC CAT GTQ GAA CAT GAG GOG CTG CCT GAG CCC C 29'1 eu Thr Lou Ars Trp Gly Lya Glu G 278 TC ACC CTC ACA TOG GGC UC GAO C CTGAGCCTGCACACCTCCCCTCAGGGAUGCTOGACCCTTCTGCACACCCTOCCCTCCTCAGGCCTGAOGGCTGOCGTCATCACCCT 3052 lu Pro Pro Sor Sor Thr Lya Thr Aon Thr Val Aon Aon Ala Val Pro Val Val Leu Cly Ala Val V 300 CACCTTCATTTCCTQTACCTGTCCTTCCCAC AG CCT CCT TCA TCC ACC AAG ACT AAC ACA GTA ATC ATT GCT GTT CCG GTT GTC CTT GOA GCT GTG G 31'9 al Aon Lou Gly Ala Val Hot Ala Pho Val Hot Lya Ars Arl Arl Aon Thr G 317 TC ATC CTT GGA GCT GTG ATG GCT TTT GTG ATO HG ACG AGG AGA AAC ACA G GTATGAUGOGCAGGGTCTGAGTTCTCTCTCAGCCTCOTTTAGUGTGTAC 3251 ACTGCTCATTAATGGGGAACACACCCACACCCCACATTGCTACTGTTTCTAACTCCCTCTCCTCTCAGTTCTCGAAACTTCCAGTGTCAACATCTTCCTTCAACTCTCACACCTTTTCTT 3371 ly Cly Lyo Cly Cly Aop Tyr Ala Lou Ala Pro C 328 CTCACAC CT CCA AAA GOA CGC CAC TAT OCT CTC OCT CCA G GTTAOTATCCCCACACCATTGTCCTAGAGACATTOOAOTCAACTTCAAGATGATGCGACCTCTCGGA 3'78 ly Ser Cln Ser 331 ATCCATGATACCTCCTCCAGAGAAATCTTCTAGTTGCCTCACTTCTCCCATGAAATGAATACATTCATGTACATATGCATACACATTTCTTTTCTTTTACCCTAG GC TCC CAG AGC 359' ro Ala Asp lle Thr Leu Thr Trp Gln Leu Asn Gly Glu Glu Leu Thr Gln Glu Met Glu Leu Val Glu Thr Ara Pro Ala GlJ' Asp Gly T GAG CCC CGG GCG CGG TGG ATA GAG CAG GAG GGG CCG GAG TAT TGG GAG CGG GAG ACA CGG AGA GCC AAG GGC AAT GAG CAG ACT TTC CGA '98 Val Asp Lou Ars Thr Ala Lou Arg Tyr Tyr Aon Gin Sor Ala Cly G 91 GTG GAC CTG AGG ACC GCG CTG CGC TAC TAC AAC CAG AGC GCG GGC G CTGAGTGACCCCGGGTCGGAGGTCACGACCCCTACACTTCCCGACACAGGCACGCTGA 602 CGTCCCGCGTCCCAAGTCCGAGATTCGGCAACAGAACGGACCCGGACCCCGAGCCCGTTTCCCTTTCAGTTTCGAGGAGGTCGCCGGCGGGCGGGGCCCGGGCGGGTGAGCGGGGCTCAC 722 ly Sor His Thr Lou Cln Trp Hot Ala Gly Cys Aop Val Glu Sor Aop Gly Ars Lou Lou Ars Cly Tyr Trp Gin Pho Ala 117 CCGCGTCCCGCAG CC TCT CAC ACA CTC CAG TOG ATG OCT GGC TOT GAC GTG GAG TCG GAC CGG CGC CTC CTC CCC GGG TAC TGG CAG TTC CCC 815 Tyr Aop Gly Cyo Asp Tyr llo Ala Lou Asn Glu Asp Lou Lys Thr Trp Thr Ala Ala Aop Hot Ala Ala Gin llo Thr Ars Ars Lys Trp 1'7 TAC GAC GGC TGC GAT TAC ATC CCC CTG HC GU GAC CTO AU ACG TGG ACC GCC GCC GAC ATG OCG GCG CAG ATC ACC CCA CCC HC TGC 905 Glu Gin Ala GIJ Ala Ala Glu Arg Up Arg Ala Tyr Leu Glu Gly Glu Cyo Val Clu Trp Lou Arg Arl Tyr Lou Lye Asn Gly Aon Ala 117 GAG CAG GCT GGT GCT GCA GAG AGA GAC CCC GCC TAC CTG GAG GGC GAG TGC GTG GAG TGG CTC CGC AGA TAC CTG AAG AAC COG AAT GCT 995 Thr Lou Lou Arg Thr A 183 ACG CTG CTG CCC ACA G CTCCAGCGGCCGCCGGCAGCTCCTCCCTCTGCCCTCGGGCTGGGQCTCAGTCCTGGGGAAGAAGAAACCCTCAGCTCGGGTGATGCCCTAGTCTCAGA 1109 GGGGAGAGAGTGTCCCTGGGCCTCCTGATCC CTCATCTCAGGGACTGCACTGACTCTCCCAGGCCTCAGCTTCTCCCTGGACAGTGCCCACCCTCTCTCACCAGGGAAGGAGAGAATTTC 1229 CCTGAGGTAACAACAGCTGCTCCCTTCAGTTCCCCTGCAGCCTCTGTCAGCCATGGCCTCTCCCAGGCTGGGTTCTCTCCCCACCCCCACTCTCTGTAGACACTGACTCCTCTCCTCCTG 13'9 AGTCTCTCACCCCTTACACCTCACC ACCCCAAGTCCCCTTACCTGATAAGACACATGGACTCCCCTACACTAGCCTCCTTCCCCCACCTTCTAGAATTTTCCAGAGAATACATTCTCCCA 1'69 GATCCCTCCCCCTCCCTCTCTGTGGGCTTTGCACCCCTTCCACAACCCAATTCTCTCTATTCCTACAGTGCTCAATCCTCACATCACCCCTTATGGGTACCCTGGAGCAATATAAATCCA 1589 CAATTTTCCTTTTTCTTTTCTCTCTCTTTTTTTTTTCCGATTTTTTTTGAGACACGCTTTCTCTCTATAGCCCTGGCTGTCCTGCAACTCACTTTCTACCCCAGGTTCTCCTCAAATTCA 1709 CATCCCTCTGCCTCCTGAGTGCTGGGATTATCCTGCCCAGCCTTTCTTTTTTCTTTACTTTTTTTTTTTTTTTTTTTGAGGCCTTATTTTGTTTCTAGTCACTTTTTGTCTGCACTGGAG 1829 TCATCCTCTTTCTCCATGC CCTTGTATTATCATTTCTATCACTCTCCACAGGTGCCAGAGGAGACTGCTAACCTGGATAATTATGCHGTTAAATCAGGCTTCTCATTAUGAAGAGATT 19'9 CTTGTGAACTAAGACTGTTTCCTGTCAGAACTAAACATCCAGAAGCCCCCTGCTCTCCCTCTGCCCCACAACTTACAGTGCCCACCCCCCCCAGTGAATCAGGACTTGGACTCTGAGAGA 2069 CAGGGTCTTCTGCAATCCACGGGCTATAGTGAGAGGGAAGACCACACACCCTGTGAGCTCACTGTCTTCCAGTGAGTGCTGCACTAGGGTCCACAGCTCACTCCAGGGATCCTGTGTGAC 2189 ATACCTGTACCTTGTCCTCCAGAGTCAGGGACAGGGAGTCATTTTCTCTGGCTACAGACTTTGTGATGGCTGTTCACTCGGACTGACAGTTAACGTTGGTCAGCAAGATGACCACAGTGG 2309 TTGAGTCTCAGTGGTGGGACCCTTCCAGTAGCATATGCCCCTAATTTTGATATGAACTCAAACACATATTAAATTACTTATTTTCCATTCCCTATTCCATTCTGTGACTATCTCTCTCAT 2q29 GCTATTGAACATCACATAAGGATGGCCATGTTCACCCACTGGCTCATGTGGATTCCCTCTTAGCTTCTTTGTCCCAAAAGAAAATGTGCAGTCCTGTGCTGAGGGGACCAGCTCTGCTTT 25'9 TGGTCACTAGTGCAATGACAGTTCAAGCGTCAAACAGACACAGAGTTCACTCTCATCATTCATTTAACTGAGTCTTGTCTACATTTCACTTTCTCTTGTTAATTGTGGAATTTCTTAAAT 2669 op Pro Pro Lyo Ala His Val Thr Hlo His Arg Ars Pro Glu Gly Asp Val Thr Lou Ars Cys Trp Ala Lou Cly Pho Tyr P 210 CTTCCACACAC AT CCC CCA UG CCC CAT GTG ACC CAT CAC CCC AGA CCT CH OCT CAT CTC ACC CTC AGG TGC TCC CCC CTG CGC TTC TAC C 2761 75 -5 '08 GGC TTC GOG GAG CCC CGG TAC ATG CU GTC GGC TAC GTG GAC HC ACC GAG TTC CTG CGC TTC GAC AGC GAC CCG GAG AAT CCC AGA TAT Gly Phe Gly Glu Pro Arg Tyr Met Glu Val Gly Tyr Val Asp Aan Thr Glu Phe Val Arg Phe Asp Ser Asp Ala Glu Asn Pro Arc Tyr Glu Pro Ars Ala Arg Trp Ile Clu Gln Glu Gly Pro Glu Tyr Trp Glu Ars Glu Thr Arg Ars Ala Lya Gly Aan Glu Gln Ser Phe Ars 9' 21' 15 318 Het Cly Ala Hot Ala Pro Ars Thr Lou Lou Lou Lou Lou Ala Ala Ala Lou Cly Pro Thr Gin Thr Ars Ala G CCATCCCAC ATC CCC CCC ATG GCT CCC CCC ACC CTG CTC CTG CTC CTC CCC CCC CCC CTG GGT CCC ACT CAG ACC CGC GCT G GTCACTCCGTGC TCGGGAGCGAAACGGCCTCTGCGCCGAGCCGGCGGCACGCGGGAAGCCGGTCCCCGCGTCGCCCACCGGACCCTCCGCTCCTTCTCCACCCCACCCCCCGCGCCCAGATCCCCTCCCCCC (~) ly Sor HIs Sor Leu Arg Tyr Pho Val Thr Ala Val Sor Arg Pro CCTGCCCAGCCGCCCGCGGTCCCGGGAGGAGATCGGGGTCTCACCCCGCCCCGCCCCCAG GC TCA CAC TCC CTG AG~AT TTC GTC ACC GCC GTC TCC CGG CCC w w 34 EXIIN II H-:zod ~:,Oro H-2ld H-zol> H-2Kd H-2K'> Qe Q10 CSHSL RYrVT AVSI!Pa"G£ PRYI£VCYVONTIT VRrDSOA[NPRYEPRARW I EII:CI'tYIIE R£ TRRAKCM:QSIR Yilt RTAL~YVNQSAG ••••••••••••• Q•••••••••• D••••••• s ••••• I •• Q••••••• P ••••••••••••••••••• Q••• s .•.•• ••.•••••••• .P •• M••• E•••• • •• l. •.•• IS.,,,, ,k , ............... Q.P .M, ......... I.QI ... Q•• w., ,N ... L.C ..... .. ,P .. M•• ,[ •• • • •• ,L[ •••• IS, •• , .. k,,,., ........ , •• , .P ,M,,, . ,, ... , •• QI(,. ,Q, .W ... S •• NL .C.,,, •• , .P, , ,.,,,., •• ,, .L ••• ,r lA •• ,, .D.Q... ,,,, .D., .r •••• P.M ••• ,, ••• ,[Q,Q, •• SO. ,W ••• S.,, ,Q,, .... k. .P .............. L ............ D.....................M... . ........ QI( . . . . . . . . . . . . . . L.C ..... k • • Q., ,Q, .H ....... L., ,Wr!S ..... D.Q ............ M... ... M.......... , ,QM ••• H..... CS.,. ,QS ••• , .k. , , • ,M,, ,[,S,,, .. L.,, ,rll., ... D.Q •• , •• ••,, T.. M•••• P.M.,, •• ,,, •••• Q......... H.S,, .L.CH.,, .ES EXIIN Ill CSHT LOI"AGCDVESDGRLLRCVWQF AYJlC(DY IALNEDL KTWT MDMAAQ I TRRICWEQACMERDRA YL£G£CVElllRRYLICNCNA TLLRT ·=------------+---:-ltz:+=+-+++++ ---- ---+-+-+--~--~:--.,....._._ ...... ••••••••..... c.c ..•.••.•••••••••••••••• P•••••••••••• • 1•• •••• Y•••• c••••••• , .E •••• , .R •••••••••••• ••• ••••••••••• ••• •••• VY •••••••••••• H••••••••••••• • • • • • • Q. S., .LC •• W••• , •• L •••• E.R •• , ••• •••••••••••••••••••••• S •••• HVk ••• ••• ••••• H•••••• , •••••• •••• r .R.r •••• C•• w•••••• Q•••••• R••• •••••••••• ••• 1 •• L •••••••••• D•• VY •••••••••••••••• EL • •E..... •• .. I. VIS •• E.G ••••••••• Q. Y.... ••• •••• •••. •••••••••L •• kH •••••• E. •• L •••••• 1 ••••••••••••••• , ••• • • • • • • • • Y••• I(; ••••••••• L•••• E.R •••••••••• •••. V••••••••••••••• 1•• k.Q ••••• 1.MQS ••••• QL.kE •••• , .... 1 ••• Y•• k.C •••• r •••• L. Y•••• R. ·••••••••••••••V •• I • ••• ••••••••• VY ••••• A•••••• L ••• [l.k[, •••, EXIIN IV H-:zod DPPICAHVTHHRRPECDVTLRCWALCF YPAD I TL TWQL HCEEL TII:NEL VE 1RPACDCTFQICWASVVVPLGK£QICYTCHVEI£CLPEPL TLRWCICE H-2L H-20b .s ......•• P.Sk.E. ••••••• , ••• ••• •••••••••••••D•. ••••••••••••••••••••••••••"· • • R. Y••••••••• ••• H-20~ pro H-Zt<d H-2Kb Qe Q10 H-20d H-:zod pro H-2Ld H-zol> H-2Kd H-:zK'> Qe Q10 • +++++++ • ++++ • • ... ...... • • ~-----+--~=--------._- • .S •••••••• P.Sk.E. ••••••••••••••••••••••••••• D•••••••••••••••••••••••••••• N••• R. Y•• •••• •••••• .S ••• ••• Y,P.SQV,.,. ••• •••, ••• ••••••••••• D••• D. •••••• •• ••••••••••••••• ••• .N ••••• H.k ••• ••••••. .S •••••••• S ••• OK •• •••••••••••••••••••••• , .I.D •••••••••••••• ••. •• ••• •• •••• Y••••• Y.Q •••••••••• •••••••••• P. SY .A ••••••••••••••••••• • •••••••• DT •••• • •••• V•• • •• •• •• •• ••• • •• N••••• N•••••••••• • • • 1111 •••• 1 ••••• PCS ••••••••• P•••••••• •••• ••••••••• D•••• Q••••••••••••••••••••••• N. •••• Y•••••••••••• UIIN V (XIIN Yl EPPSS1KTNTVIIAVPYVlCAVVILCAYIW"- RRRNT CCICCGDYALAP • • • P .. DSYM •• V .. LC •••• MA.I ••• V• • ••• ••• •• •• • P•• DSYH •• V • • LC •••• MA.I ••• V ••• • • • •• •. kl.P •• VS.T ••••• L..... AIVT ... v••••• M. ••• • • • P .. VS.IIA TV •• L.,., .AI VT ••• V•••• • M••••• ••• PY.VS.IIAT ••• V.D .... A.I ••• v•••• N •• X•• ••• P, .DSIMSH •• DLLWPSlkLWYLX (XIJN Vll EXIIN YIII CSQSSDMSlPDCK YX ............ .......... ••••• E••• R•••c. ........... ..x•.• c. A• A• YNVIIlPHSlA. YMI'IllPHSLA • • ••• YN ••••• •• Q••• C.P •• Figure ]. ••••• E••• R••• ••• 1 •• L •••• ••• 1 •• L •••••• R•••• A. 35 Appendix A DNA Sequence of a Class I Pseudogene 36 The gene whose sequence 1s described in this publication was the first Class I gene to be sequenced. appeared in Cell. This publication 37 A Pseudogene Homologous to Mouse Transplantation Antigens: Transplantation Antigens Are Encoded by Eight Exons That Correlate with Protein Domains c.a. Vol. 25 , M3-e&2, September 1881. Copyright C 1181 11r MrT MlehMI Steinmetz, KeYin W. Moore, John G. Frelinger, Beveny Taytor SMr, Fung-Win Shen,• Edward A. BoyM• and Leroy Hood Division of Biology California Institute of Technology Pasadena , California 91125 •Memorial Sloan-Kettering Cancer Center New Yorll, New York 10021 Summary We have Isolated about 30 to 40 different BALB/c mouae aperm DNA genomic clones that hybridize to eDNA clones encoding proteins homologous to transplantation antigens. One of theM clones (27 .1) was aelected for aequence analysis beeauM It was polymorphic In Southern blot analyMI of the DNAs from BALB / c and CBA mice. A fragment of 5.7 kllobaMs of this clone was completely aequenced and found to contain a paeudogene whoM ... QUence Ia highly homologous to the aequencn of known transplantation antigens. Paeudogene 27.1 Ia apllt Into eight exons that correlate with the .UUcturally defined protein domains of transplantation antigens . Using Southern blot hybridization on the DNAs of different Inbred mouN strains, we mapped the pseudogene to the 0•2.3 ~ion, a of the Tla complex on chromosome 17 that Ia edjacent to the major histocompatibility complex. The 0•2,3 region encodes lymphoid differentiation antigens homologous to the transplantation antigens In size, In peptide map proflln and In their naoclation with ,g2-mlcroglobulln. TheM mapping ~udles suggest that gene 27.1 may ba a pseuc». gene for either a Oa antigen or an " yet undefined transplantation antigen. Accordingly, wa may have Isolated genes encoding lymphoid differentiation antigens of the Tla complex as well as thoM enc~ lng transplantation antigens among the 30 to 40 different genomic clones Isolated from our aperm library. .,_rt Introduction The major histocompatibility complex plays a fundamental role in the regulation of the vertebrate immune response . It encodes three families or classes of proteins-class 1. the transplantation antigens; crus II, Immune-response-associated or Ia antigens; and class Ill, oenes encoding complement components (Klein, 1975; Ploegh et al., 1981). Adjacent to the mouse major histocompatibility complex or H-2 complex on chromosome 1 7 is the Tla complex (Figure 1). wtlich contains several genes encoding lymphoid differentiation antigens. the Oa and TL antigens (Flaharty. 1 (iJ80). The class I or tranaplantation antigens (K, 0, L and R) are integral membrane proteins with molecular weights of 45,000 and are associated 1/2-microglobulin . The Oa and TL antigens appear to be similar In size to the class I molecules and also are associated with ,g2-microglobulin; this raises the possibility that the class I molecules are evolutionarily related to the Oa and TL antigens (Michaelson et al., 1977; Stanton and Hood, 1980). The transplantation antigens are expressed on most somatic cells of the mouse, whereas the TL antigens are normally expressed only on thymocyte& (Boyse et al.. 1964) and the Oa anti· gens only on hematopoietic cells (Flaherty, 1976; Stanton and Boyse, 1976; Hiimmerling et al., 1979). Structural analyses of the transplantation antigens of man and mouse. including proteolytic digestions and amino acid seQuence analyses. suggest that they are divided into a Mries of discrete domains-three extracellular or external domains. each of about 90 residues. and a transmembrane domain and a cytoplasmic domain, each of about 30 residues (Lopez de Castro et al .. 1979; Coligan et al.. 1981). Both the aecond and third external domains have a centrally placed disulfide bridge spanning about 60 residues . Thus they are similar in general structure to immunoglobulin domains, wtlich also have a centrally located disulfide bridge . Moreover. modest amino acid sequence homology (-35%) has been noted between the third external domain of transplantation antigens and immunoglobulin domains (Orr at al., 1979; Stromlnger at al., 1980). These observations suggest that transplantation antigens are evolutionarily related to immunoglobulins. A number of investigators have neported the iaolation and characterization of eDNA and genomic clones encoding proteins wf"tose seQuences are highly homologous to those of known human transplantation antigens (Jordan at al., 1 SJ81; Ploegh at al., 1980; Sood et al., 1981) and mouse transplantation antigens O<vist et al., 1981; Steinmetz et al .• 1SJ81). By DNA MQuence analysis of the eDNA clones. we could show that the third external domain of the transplantation antigen shows atriking homology to the conatant region domains of immunoglobulins (Steinmetz at al., 1981 ). This observation suggests that immunoglobulins and transplantation antigens diverged from a common ancestral gene . Moreover. Southern blot analysis with these eDNA probes against germline DNA ,.._ vealed that there are at least 1 5 crou-hybridizin; apecies of genomic DNA (Cami et al .• 1(iJ81 ; Steinmetz et al ., 1981 ). Thus the genes encoding transplantation antigens constiMe a multigene family. In this paper, we report the DNA seQuence of a 5.7 kllobase (kb) fragment of a mouse genomic clone (27 .1) that hybridizes strongly to our eDNA probes. The DNA seQuence of gene 27.1 includes eight exons. aeven of wtlich are homologous to our transplantation antigen eDNA seQuences and correlate wtth the atruc- 38 .... • a 0. 1.--ct I .. L. el 8 I I . .,....,, . ...., I ~------•1------~~-----T~-----0. I,Oc,.,.----- , .... , . Genet>c: .,., of 1t1e H-2 ...c~ n. ~··of cnr-,. 111111he MOUM TN order of loc:i - ~II not e.-. c.nn~o~orvana o.c.nc:.- In tural domains of transplantation antigens. This gene is a pseudogene that maps in the Qa-2 ,3 gene cluster of the Tla complex . Hence it may be a pseudogene either for a Qa differentiation antigen or for a transplantation antigen mapping to the right of the [).. marker locus of the major histocompatibility complex . Thus. using our eDNA probes. we may be able to iaolate a variety of closely related genes encoding lymphoid differentiation antigens (Qa and TL) u well as those encoding the claSSical transplantation antigens. Rnulta and D+acuuion Isolation of Genomic Clones Hybridizing to eDNA Pro~ for TranaplanutJon Antigens About 1.5 x 10• phage clones of a BALB/c aperm DNA library were screened with a mixture of our eDNA clones pH-2111 , which encodes portions of the NHr terminal half, and pH-21. which encodes portions of the C-terminal half of transplantation antigens (Figure 2A). Approximately 30 to 40 different clones were isolated (K. Moore, B . T. Sher and H. Sun, unpublished data) One of these genomic clones, denoted 27 .1. was chosen for DNA seQuence analysis because this gene exhibited polymorphic restriction sites when the DNAs from different inbred strains of mice were analyzed by Southern blot analysis (... below). The Eight Exons of Gene 27.1 CorTeepond to the Structurally Defined Protein Domains of the TranaplanutJon Antigen Clone 27.1 was mapped with several restriction enzymes . and the coding seQuences were identffied by hybridization with our eDNA clones. This analysis showed that the gene present In clone 27 .1 was located on a 5.7 kb Bgl 11-Bam HI fragment in the middle of the 1 9 kb mouse DNA inaert (F'tgure 28). The seQuence of the 5 .7 kb fragment was completely determined by ute of both the M 1 3 dideoxy MQuencing techniQue (Sanger et at., 1 980; Mesairtg .t al., 1981) and the chemical degradation method (Maxam and Gilbert, 1 980) (Figure 2C). The DNA MQuenca Ia given in Figure 3. A comparison of the genomic seQuence with the three eDNA seQuences of Clonea pH21, pH-211 and pH-2111. as well as a comparison of the translated genomic DNA seQuenc&a with the complete amino acid seQuence of the K" molecule (Coligan et aJ .. 1981) clearly delineates exons 2-7, which corre- late atrikingty with the atructural domains of the transplantation antigen (Figure 4). The DNA seQuences of these exons In gene 27.1 are 81-95% homologous to the corresponding seQuences in the three eDNA ctones (Table 1). Furthermore, the translated seQuences of these exons of gene.27.1 are almost as homologous to the Kb polypeptide u are the tranalated eDNA seQuences (Table 2). Exon 1 could not be identified by seQuence comparisons with the eDNA clones and transplantation antigens. It was identified on the basis that it starts with a methionine codon, encodes a lt1etch of 21 amino acids. 19 of which are hydrophobic or uncharged, and contains at its 3' end an upstream RNA aplicing signal that has its downstream counterpart in the first codon of ex on 2 (Figure 3). Ita hydrophobicity, location and size indicate that the peptide encoded by exon 1 probably represents the signal or leader sequence identified as a component of the initial translation product of the mANA for transplantation antigens (Dobberstein et al .• 1979; Ploegh et al.. 1979). Exons 2, 3 and 4 correspond to the three external domains of the transplantation antigen (Figures 3 and and have as their boundaries those amino acid residues predicted from the protein MQuence ~ ogy comparisons (Lopez de Castro et al .• 1979; Coligan et al ., 1981). The translated amino acid seQuences of the extracellular domains of gene 27.1 are 73-85% homologous to the Kb molecule at the protein level (Table 2). Exon 5 corresponds to the transmembrane domain. The transmembrane domain appears to be among the least highly conMrved domains (Table 2). in keeping with the functional reQuirement for conservation of the hydrophobic character of the transmembrane belt, but not Ita apeclfic amino acid seQuence . Two cytoplasmic exons, 6 and 7, can readMy be identtfied by their homology to the eDNA aequences and the H-2Kb amino acid seQuence (Figures 3 and 4). Exon 8 was identified both in Ita codirtg and 3' untranalated region by comparison wittl the eDNA clone pH-21 (aee below for a comparison of this rwgion with Clone pH-211). We were surprised to note that thrN exons, I, 7 and 8, encode the cytoplasmic portion of the molecule. There ia no evidence at the protein leYel to suggest that the cytoplasmic region Ia comprised of more than one functional domain. However, the cytoplasmic exona may reflect diltlnct functional entities in that thera is a possibility that the 3' end of the gene may use alternative RNA eplicing patterns that lead to distinct cytoplasmic domains. The exon encoding the third extracellular domain ia highly conserved and shows a I\81TOW8I' range of variability than any of the other ftve exona (Table 1). Since only the third domain In the transplantation antigen shows significant homology to the c:on.tant rwgion domains of Immunoglobulins, SUCh restricted dl ~nce may reflect the constraints impoeed by Ita •> Interaction with yet another ~· do- 39 COOCJit~TI()Jrlll Ol (Al r·2 1 .... · 111 r · 2Do ,..z m .. I . I l lo ·· ~ s .. !C i --------O> Figure 2 I C '·' ~ liQ'T l ·- ' o' .,.,'" &Q L• X 1' ------- 20 ) .0 WaDs of Th<M cON A ~ and the Genomic Clone 27 .1 •o .' (4> A companaon of the eDNA proDe$ UMd in !hi& paper and the coding and 3' untranslllt~ regiOn& of mANA& lot tranaplantation antigens (Me ee e Steinmetz at a! . 1 1). CBl Partial reatric:tiO<\ map of~ 1 kb inM<1 found in clone 27 .1. The Eco Rl aile& a1the ends of the insert - e generated Clunng conatruchon of the BALB / c _.-m DNA library from penial Hae IN-Aiu I ~ic fregment& 8nd Eco Rl linker fregments C0aY1s et at. . 1e80) The 3 8 kb Bg! II fragment and the 2 . 7 kb Bam HI fragment - • .ofa1ed lot DNA MQ~ analy8is Ex"'""*'tat Procedures) . The loc:ahon of~ ••ons of gene 27 .1 are indicated by blael< bo•es A llatelled bo• clenotes ~ 3' untransJat~ reg•on . (8) Bam HI . (8g) Bg! fl . (H) Hond Ill. (R) Eeo R!. (C) Map Of the 5 .7 kb Bg! 11-Bam HI fragment of clone 27 .1 ahowing the position of indMduaJ MQutneeS obtained . The &traoght arrows g ive~ 5'-3' direction and~ lenglh of~ MQuenees obtained from inetrvidual M13 ~ . Wavy arTOWI indocate ~ direchon and length of MQii&f>CH Obtained by the chemiCII degrac:lalion method (Me EJ.penmentaJ Procedures) At the Xba I aita and !he Hae Ill lite • bp UPSiflam of ~ Pst I aile marll~ by a star . no oYertawng MQuencn - · obtained . {8) Bam HI : (Bg) 8gl M: (P) Pat 1: (S) Sail : Sat 1: (U) Pvu 1: (X) Xba I. <- m main, P2-microglobulin . Thus this domain may have rigid selective constraints placed on its divergence in different ~enes for transplantation ant i ~ens . Additional seQuence data on the class I ~enes or proteins should reveal whether diversity is clustered in certain regions of the exons codin~ for extracellular domains. The 3' untranslated region of these genes is as highly conaerved as the coding regions. Gene 27.1 Is a PHudogene by Several Crtteril! The seQuence presented in Figure 3 shows that two stop codons are found in the reading frame at positions 312 in the transmembrane exon and 328 in the second cy1oplasmic exon . The part of the transmembrane ex on containing the termination codon has been determined by dideoxy seQuencing of both strands as well as by the chemical degradation method (Figure 2Cl. The two stop codons would not allow the trantr Ia !ion of an intact class I polypeptide of gene 27 .1. Moreover, in the ex on encoding the transmembrane domain , codon 292 codes for the char~ed residue, aspartic acid . The presence of an aspartic acid in this position would prevent an energetically stable integration of the transmembrane domain into the lipid b+layer . In addition. the DNA seQuence at one exonintron junction is at variance with the consensus seQuences that are believed to be necessary for RNA splicing (Lewin. 1980). The acceptor site of the second cy1oplasmic exon (exon 7 in Figure 3) Shows the deletion of a G in the last position of the intervening seQuence when compared with the consensus seQuence . A shift of the splice point by one nucleotide into the exon might generate a functional splice site but would alter the reading frame for this exon . One other feature of gene 27.1 distinguishes it from the clone pH-211. At least 165 bp at the end of the 3' untranslated region of gene 27.1 appear to have been Inverted and translocated into the intervening sequence between exons 3 and 4 (Figure 5). Since this part of the untranslated region in pH-211 is a repet itive element in the mouse genome (Steinmetz et al., 1981 ). It also is possible that the seQuence between exons 3 and 4 In gene 27.1 represents another copy of this repetitive sequence and that the corresponding homologous region at the 3' end of gene 27 .1 was deleted. Although gene 27.1 Is a pseudogene, It shows atriking homolOOY to the ~ polypeptide (Table 2) as well as to the three eDNA clones (Table 1). Moreover. the intron-exon boundaries of the first extemal domain are identical to those of one other genomic clone that appears to be expressed as a transplantation antigen in cells transformed with this clone (R. Goodenow and J . A. Frelinger, unpublished data) and is now being characterized (K. Moore and B . T. Sher, unpublished). Thus pseudooene 27.1 retains most of the fundamental characteristics of the genes encoding transplantation antigena. Gene 27.1 Maps to the Tla Region of MouN Chr~17 We have previously shown that a restriction enzyme polymorphism is observed when Bam HI-digested DNAs from BALB / c (d haplotype) and CBA (k haplotype) mice are probed with clone pH-2111 in a Soutnem blot analysis (Steinmetz et al ., 1 as 1). A strongly hy- 40 a• ... TtTTTCTTTf't''''C•M•<•••K••t~~ F icz:w::a:oti&tiWMAYW:TTTM~~~~TCT tTAT~TATt"Tt"f•••••••••••••t • •••••tACrt"'Ct"T•t"(,••"'(;•t*C'ttC••.C••"'('••+••!"C:C f!OC.AGt'llfC.JI:JCTCTT'CAIICT~~TCT;:~fCCTW G C CO'T•vm~nMCTG.t~1'CtltEACT'C.AialiiC~T;"T';.AGCTG:CC.AfT"GC.AGC;~TCCTt'TCTG~Tct.t.n:TCilieiGCTTTGC~•CCtGCGCt:Ttf'Q:TTT~AT CTTtACTCIIC.Ara:r:t.MGTliCTTl;~tt"l"CUTTC'TA~T"Ci~CCo'tTTCTTCl~TtTMCTTCTt·•c·c·•c··~~•'%••''('~TAGGGC.Cf&*GKTGT'CiGCC I WJLICIU - .. - 1"1CT'C1"C:~TtT&I~~TtTtT'QCC'(4G;CT~TCnM~T'tMGCii:T~TCiiiCT~TTC~TCTl'CCCCA ,_ T1'1"tAC1"t111t1"Crt~T'tiCTT'UTTt~Tc.t1'tioolaTCTA.IriCACTC'Ttr.ACCX:TATTCa:T'C.ACiCTGAn~TUCGCTCRZVSGF.t.GC;;;G~iTm~~Tt.A - 01111 I ~mrrrecm··r;r;, . . Ale ~ 'nor . . . L-. L-. .._ "'- '1•1 AI• AJe Ale l.-. ,..,.. ~ Jl• GJ .. n.,. .... AI• C A'f; lET ClA li£A A'f; m CTt m CR; C~ CiiCG act Get CTG: .tra: C1'G ATt W ACt C1:iC cz:; C ~TCCCIC*CGC•ur.qtrrTGCTGC I, C!• "'•• .... .._ 1'CG LTC ~TtCI"'CC*CCCt 48 U:ttcm:ux:ITCTtt.m:t;.crz:ra;'«T%~~~ CC tAo\ tleC U11"7 I I'll z liD ca .. r,.. "- ..,. 'nor a a. "•' ._ .., ....._ CJr .._ Clr CJ .. "- r.,.. "'- 1•• s- w.a c1, r.,.. ••I A-. .._, n. ca .. ,.._ ••• ..., ...._ ..., ._ ....., •t• cJ .. ._. w~=~c~=~==~=~~=~=-m=~~=~~~~====~~~=-,.,... Arof ..,. CJ .. """'- ..... Ale .,... T.,. ... ""' CJ .. CJ., Glr " - Cl.,. t,.. t..,. Cl ........ GJ .. 'nor CJ,. . . Ala L,. IIOJr llfta Gl .. Cl ...... " - ..... GJr s- t.- .... tee MilO ATC ,.... C::Z:: CZ lOa aiC ""An;"" C-" ""GGG t:CG '-" TAT TaG CoiC CCE COIIG Col CAG "" GICC ~ CGt CAT ,.... c.tC I/I;T TTt (;GI. GGG 11/Z. CTG MiiC n.. "'• ct .. ._ r,.. r,.. A.- , , .. . _ L,. ca, c IICTGC.ACM;AGC 01111 • ~ ~ L-. = = TI&.TIICMt~AGCI!JI&.CIZCC~rt:z:l'oiWiC~T~~ ~TTI:IXTTTCAGTntl.......-TtCOCirnl""'"""""'""'_;a:c__....:'TIOICL<Cna;T~ :t ~ ~ ~ ~ ~ ~ :: ~:; ~ ;rr ~~~ ~ ~ l... MA iPt CJp Tpr ~ CJ• ..._ Ala Tpr Gl .. GJ, ......... T,.. Jla AI• ~ . _ Cl. . . . . c.- L,. Tt. T. . n. AI• Wal . . . . _ AI• Ala CJ,.. Jla n.- ....... =rn==~~=~m~~--=~~-~=~-~=-~~~-=-m=~~-c~ ....., L,.. '""' Cl ... CJ .. •1• Cly J I• Ala ca.,. L,. ..... CJ., Ua r,.. c.- CJ.,. Gly n. t,.. ._ CJ .. ._ L.- .......... T,_ L.- CJ .. L.a. Cly Lpa Cl .. l'P L.a. c.- ...... =~~-~~~~~---~~~=~-~~m~===-~=~=~-~~=== ~ ~~TttTCCC~~TC'Tt'Tt•17VV·•rr"·c··n~r~TiiiCTCCCTTt.tGTTtU:t~~TQiiCCTC T ~nr:r a 1a - t~TATttTGt'~TCAGCtt'TTIIC.ACrf'tA~T'CT:rTTTAICCii.A~T'CiliotltTAftrT/IiUtCTAGCiiCTGT~IIG.UCTT"''"':':CMAGTACACTCTATt.AGATt. ZAI c:tft.CGT~;TtTr: T CililiilffTTGCA Tc:tTT~ TTt"T ATtT A T"1'a:TliiCM ~ TCiGTt.AUT~ TT AT'CIIiCTT.fiCXT ~ TACTT'TTCTT;TG"TffTTCCCtT'CTCGTTTTtTTTTlA TtTTTAr.fTTTTT ,_. TAAGiiC T ATTATGTTCiCTTAfAJora;rTTTTTtT~na;.uTC.Arr:TTCI:Tt~Tvmt'vrcectCMATATUTTT;'TATt.AC'TAGICCCt'GI::r: T~TCACTt~T ~TCIZ:t'TTC 81:1 A/II&Tt.AGM TtAGC:TGCCTCTiiiiXTtTGiiXTCTQXTC I CLt I C I CiX I C EW::X: I CI QX I L ICiX I L IQ:t I L~ IC IGiX I C I"" ICTGCCTtTQ:CTCnz:rrtTGCCTCT'C'ii:.A TT .t.MC«TCGGCCAO:ta: ACT'C'oiiCiGAa 178 ~TTCt'TG~TT/INI,#.T'CTTTTt~TTAI!CU.rtr.AGAtt.tcTCCTCiATArct:rTG~.tC.M;tttrc:c::tTGGTGMrc.r.c.u.cT~T'CC.AGGCCT 811 c.AG T ~~.-uct&CACtCCXTr:T~TCTt;T~1"CC:..IC~~T'CCTCTC1"r:otCAC.t.ft1'CT~~TTTT'C'Tt'TGiitTG.AC"T _. Cft~T~TT TCTCCT~'ft:CCTG,Af'GCCTCTTTACTT'GiC.AC~TTAATCTT;G~~TGCT TT~TCiG~T'tM:TCTl"Ct.tGTAGI:ATAT'GGTtrf:AfntTAATTT/JIIi,AT .. ACCAACTCAAACAC.A TG,A.U TntTT ATTTT'Ct.A ~ TtTT'C':.t TT AT AT ACC1"GCrT ATCTCC"TtiCT ATT""""'Tt..t&:A T&.fiG,A T'CACU T;'TTT~TCICCT't.t f';fCiCA TTCCtTCTT AGr;T'fCT'G,AG~ - •TC~TC.CTGTCit~Ttra:tT'C'It.AGtf'C,ICT ~nz::tA.~T~T~Tat;T'T'CATTC'TNtTTI&.T'GAm~T;~TTTtACT'T'TCCATTTATTTATTlATTlAT .. TTATfTATTTATTTATTTATTTAATCC,.~T~T;'TtCT'GolT'CiiC'lr;T"TTCiiX'f'fTCTCT'IiiC~nc..ATTT'TTTTTT~AT'CG - ~TCTGCT'tACT'tACTCCCTCCTTC~TTTATTTATTATTTAnAT.tlt.ATAN.T~~rr+a¥· _. ' ic:.t51i'l't1UnCMAfCIIn'TC~f;TGCT'f"CCT • ~~r~naa:t~Tt~T1'C'm1'CT1'un.ara.aT'T'1rf"'''AU~ .aT ttr.. ,..... L-,. AI• • u 'tleJ flw. act a:z: c:u Me Citol ur m 01111 • ••• ••• .._ ,., ' - T,_.. Clr AI• 'ti.J ft. r..... .... c,. T"'P Ala c.- Clr .._ T.., ..... Ala ..... Jla 1'h- L.- ft. t,... IJ., L.- .a- Clp CJ.., Cl.., L.- Tar. CJoo ,._ ~~=-m~~~=c=-~~~=-m~~~--c=c~~m~-~~=c~fa. GJ.., L.. 'tl•l GJ .. 1¥ ..... ,.,._ Ala Clp ••I CJ, 'nr. ..._ Cloo \..-,. t.,. Ala .... "-l "-1 ._, ..... L.- Clp L,. Clw g.,. .... r,.. ,_. c,. .... 'fel . _ Clv "*• ~~m~~c-~=~~~~m~-~~~~~~~~---~~~~~w~~~~ _., ~ t..., Cl,. L- .._ Caw ~... L.-. hr " - iPt Cly .... T~ C " ' CT' ttT '-" C« CTT ACt ' " AGA TGG eGG """ TGC C Gf~~TGGGTGC.AoCMa:fC*~TQC.IICICAT~TGIIGrTiiiCTCIIGiiZiCT'tACoATCTGC:GAT'GAtttT'L T.., "•1 .aaD 01111 5 J.., _.,... ,.._ _._ hr s- ._ ._ Ala 'hr II• Ala Wei 'fal . .1 ........ Glp Ala W.l AI• IJe IJ• Glp Al• "-I ... 1 Ala ACCUCATTt.a:Tr:TACC"TCtTc.tUC; AC CCl CCT teA TIIC ACT CTt fCC ...C " " ca; .act ATT GC1 GTT C:~ CTT CotC CTT CiCA liitT CTC Ct &Tt AU "" C:CT C~ r:TG C:CT ..... ..... "•l ....... ~ ...... ,_ .._ """ c TTl CTC Aft AAT ~ -"' T'-i A.t( IICA C CT~TTtTtftT"CAGTCTCtmAGMCTCT'CICTtlMTtATTM~TCTAC.ACO:CACATT'CClACtTTtT:CI..:tr.I:ICTC::T ·- 01111 • 1,. CJ,. Cl" Clr Glr ..., Cpa Ala " - Ala " - C CTtTCAGTlC~TfctMG.ATCTTCCT~TTTUTTCTCACAG CT ~ CM CJC,A C11C "" TCT GICT C1:.A ItT etA C CTT.AGTC~TTCTCTr:TC:itol£.t.TTCiCACrc.uccTCCOAGATCT~ •"'2'1 01111 7 lp . _ ,_ .... . _ .... . , .... ~Tt f~TttATAGTAA('TCT~TCTTC:C.AGI;iGC[Tc.AGTTCToiiCLUTATCo\ATI!I;I.TATATCTIICATAT'CCATATI/U.TTTTTTACLT~ ~ ct TCt TAC AGe T'LT GAT ACoA TCT _, , ... a., ,.,. ",.. A Clt tt.t GAT C,CT A.Uo .. ... r: r:Tc;.t.UCTCTr.GCGCCTCOAT~Tr:TQ;.ATaf;.fftiiGGTTTt.aGI:G.KTt::CAC.r:AAfttCCTtTli.AC""'TGCTtilii:TTGTT'CiiOAATCTT;:TtTI'CACIIGfCoi.TCiiGCTtTTCTC::C .. 01111 • -: TC.tTTCTCTAC. CA TGA loCACACCTCOCC:Tct.ACTc.t.ACT!;A.C~GATr:t,TTtATC:TCfCTCCTtTCACA.Ttt~C..CTtACTTCTtTTT~T'CfCTtAfGTTCCtfC"tACCCTA'"""fC.t.flft T CAQ;a tiW ACft T c.G.r.Gt C CAC.ttTCC.C.: T !;CACACCT""-Attc"TC.C:: t CCKTtoCC.CTCfTC.C.C l TC:.C~TTC.C:lC T lC.C.t.GC't..v.ct~tcc.Titl;,J.TtlAU.TCCT;TtAIXTtcATQ: T ~ T CC4CCTCC• T ACTTT Ia t•C.• C :.t.AtAtT.tTc;TCAC CTT t.AtJc;Tt.tTt.ATCTT""'"T~TTc;..TT"T"'TT ..... TTlAATCCfl"""""T.ttfhGtATTTTTCTCfTTTTTAA.t.AA.UU.UTu.tTCGC ... ~TTtCAGI;.AT.:TCTCltAC t Sdl ""'''TCTCTCTCTCTT~!;CT T '-""CTAT~~TGTCT~TCTCCACJrGtrccTtTt.AGTt.tTTl.tTTTTtTTTMTCC.TTCAICn;tr:Titlltt.AtGailiieiCTCTIICTCTCfCtAftACTATGoUGCAC.fl.fll( .. ~lCTCT(;AftT~l~TC.toGc:tT"TCCTr:~ TT A"-"Tcc:c.uGilCiiCTAAAC.&~TC.Attf'lliiiG,a'IU .,_ Figure 3. The DNA ~e of the 5.1 kb Bgiii-Bam HI Frag"*'l of Clone 27 .1 Amono acid MQ~ encoded by the exona of~ 27 . 1 ..-e ~hewn above tile DNA MQUenCM. The 3' untranslated regiOn ia under1ined. bridizing 6 .2 kb band seen in BALB/c DNA is replaced by a 5.-4 kb band in CBA DNA. In a similar experiment in which pH-211a is used as a probe (Figure 2A). a strongly hybridizing 2. 7 kb Bam HI fragment (lower band of the doublet) is missing in the k haplotype DNA (Figure 6). Clone 27.1 was initially selected from among the 30 to -40 distinct genomic clones from the BALB / c library because it contained a 6.2 kb Bam HI fragment hybridizing to clone pH-2111 and a 2.7 kb Bam HI fragment hybridizing to clone pH-211a. This clone therefore was a candidate for the BALB/c DNA seQuence showing the polymorphic restriction patterns described above. To test this possibility, we also have analyzed clone 27.1 by Sst I and Xba I digestions and compared it with similarly digested BALB/c and CBA DNAs probed with clone pH-211a. For both enzymes, a strongly hybridizing band in BALB/c DNA corresponded in size to that restriction fragment of 41 0 5.7kb 2 First Second Gene 27.1 Figure • Organize !ton ot Exona and lntrons in Gene 27 . 1 Gene 27 1 is compared to 1M H-2K'tranaolantation antigen (Coligan e1 et ., 1881 l 1o IIIOW 1M localiOn olln!..-.lng ~ wt!ll .-.-ct to the amino ecld 1>0SI1tons E•ons 1 toe labeled accorclong to ll>etr c:orrHI)Onding protein domains arallhown u blacl< bO. .a and the 3' untranalated regoon as a 1\atched bOx . Compared to the H-2K' molecule . 1M third exlttmal Ck>main o1 gene 27 .1 il longer by ltlree codons et ita 3' end . Three eodltoonal codons at lf\ls poailoon ,.... a1ao been found in 1M eDNA Clone I)H-2"·1 ~· e1 at .. 1881 ). Codon 30i Cmelhoonone on H-21<") os not found in gene 27 .1 . II is e1ao IICMnt In 1M eDNA~ pH-211 (Steinmetz e1 el .. 1881) and I)H-2"·1 (t(vislal at .. 1881) Tile ln!et\'flnong MQuenco between !he MCond and the lhord cytoplaam>e ••on ISI)Oalulated from a c:ompanaon be,_~ 27 .1 end pH-2t. For a companson ollhiS pen olthe 27 1 MQuenc:e a~onat pH-211 see Fogure II . Tile homology b e , _ !he tranUifl<l aequenc:e ol gene 27 .1 end !he H-2K' proteon MQuence encts preciaely at 1M begonnong of this 1311 bO ln!rOtl . A DNA a.QUet>Ce ltlet mogll1 code lor a ~ MQUenCe homologous to emono acids lC0-346 ol H-2K' has not been ielenlifiecl on gene 27.1 . Teble 1 A Companson otllle DNA SeQuer>ees CPercenl Homology) olthe Three cON A Clones and 1M Excona in Gene 27 .1• pH-2t pH-211 Fnt external CIOmein <••con 2l Second enema! domain (eaon 3l BC'Ib t5'1b TI'Wl~ domain (aaon 5) Second cytCJC)Iaamtc aaon Cuon 7) Third cyl~. .mic axon end 3'· untranalated regoon (uon 8) ceor.. IIH-2t 81'1b Third axlemal Clonwin (a aon ") Fir&! cyt~aatNC eaon Caacon 8) pH-2111 Tlltlle 2. ~ o1 the H-21<' f'ofypeptlde CPercent Homology) with the Tranalat..:l ~ 01 I)H-21, pH-211 end pH-2Ut end the Exona o1 Gene 2 7. 1• 87'1b 87'1b 88'1b to'lb 85'1b 88'1b 88'1b' 8C'Ib' • SeQuences - • ~pared lor 1M indMclual CIOmM>a (F;gurw 3 and C) where owrlaps - · praaent . • For the compan.on of the 3' end o1 gene 27 .1 wt!llthe c:ONA Clone pH-211. a different tll)licing pallem waa-"'*' u ~ 1'1 Figure II. 'Ttw ~pana()l'l of the 3' untranalatad region'-- gene 27 .1 end Clone pH-211 wu aomewtlat comt>ie• in ltlet a portion o1 lila regoon in gene 27 .1 was tranalocated adtaeent 10 the INrd eacon (Figure 5) Homology ol 87'1b .... found lor the regiotl 3' eo 1M Meond cy!CJC)Iaamtc e•on . lind homology 01 80'Ib was found tor lila translocated 3' ~ · Ttw ,..n iS~ in the II!Cie. 27.1 carrying the coding seQuence. In each ease . lttis band was missing in C8A DNA . Thus we are confident lttat clone 27 .1 represents the identified polymorphic DNA seQuence . To map gene 27.1, we used a series of congenic and recombinant lines of mice. Congenic lines are strains of mice genetically identical to a second strain except for a limited segment of one chromosome lttat carries a different allele at one or more maril.er loci (Kietn, 1975). Recombinant congenic lines in the Tla pH-211 First uwn.l domain pH-2Ht 27 .1 le'lb 82'1b lecond axternel Clonwin a..'lb 'Tl*tl en.rnel dDIMin 18'1b 85'1b tl7'1b BC'Ib T~111• Clonwin 80'Ib 73'1b Fir'll cy1QCJiumiC Pon 100'1b 100'1b 73'1b s.eoncs cytopumlc ucon IS'Ib et'lb tl2'1b • Tile ~ - • compared wtlere ~ regtOftS - e preMnt For the MCond cytoPIUtnie ol the K' moleCule. only ••on lfl\lnO ac:ida 326-331! - · c;omparad beca.-e the C-wminel rweiduea ara 1101 llomoiogoua eo any Cll lila COIIMiiOiidiiiQ DNA eeQUenCel (text). complex have been developed from recombinational events between two congenic lines (Flaherty, 1&76). Any differences observed between ...combinant congenic strains in the expression of a given trait augoest that the gene controlling lttat trait is Identical or Closely liMed to the maril.er locus. When we examined by Southern blot analysis the congenic recombinant atTains 86 .K1 and 86 .K2, which arose from recombinations between the congenic lines 86-Tla" and 86H-2" (Flaherty, 1&76), we found that the 2.7 kb Bam HI fragment was present In the 86 .K2 atrain and absent in the 86 .K1 atrain (Figure 6). When the appearance of this trait Is compared with the genetic maps of these strains (Figure 7), this trait maps to the right of the reeombinat•onal site of the 86.1<1 stratn and to the lett of the recombinational site of the 86.K2 I- I ~.1 ~ l'f.1~1114) 2 3 42 4 5 )-(- • :nzo 391e 6 I 1 ..--------.; 4923 53~4 --<~--------~-------------~·-<------1---------~·--~--------~· ----------------~~-.--<-------~··~<--~~~------------·----~------ ,...,Jl "· 1 ~ c----c--"---<~-- •------<---->-<·~___...,,---t_.. .,,.......,.....,mo,.,...,. __ "·' a1 ...,.. ~.I -+,..,---- n~4liC 1 ......,_,......._.. C81 ..1711D _____..,.,_________<---------------------------ct------~-~MT...T.........,.,.,. Figure 5 . A Companaon otthe DNA ~uences otthe 3' Untranalated Re;ion of the eDNA Clone pH-211 ancJ the HomoloQous SeQuences Found on Gene 27.1 The exon-1ntron atructu111 of~ 27 .1 u Wived from the compeliaon wi1tl pH-211 (... Figute Ill ia lhOwn In the~ pert otthe figure wrtt>the arrows •ncl1catmg the extent and 5' to 3' ~ections of 1t1a NQuencaa 11\at are ClOfTlP&'ed to 1t1a eDNA MQUanee In the lowe< pan o1 the figure. Nucleo!tcle numbers 111far to Figure 3 The stOQ COdOn at the be<;jtnn;ng of the 3' untranalated NQuance In pH-211 ia boxed. Sectuenc:a MT4M. wtltch ia lt\ought to be a recogn11ton aignal fO< pely(A) addition "' most aucaryotic mRNAa <Proudfoot ancJ aro-tea. 11178). ia alae indicated In gene 27 . t The ONA seQuence of clone e>t1-211 was praYIOUaly I>Ut>bheO (Steinmetz at al .. 11181 ). except tor a ltfalc:tl o1 172 be (inelicateO by arrows) tn 11\e mtOdle of the 3' untranalated rag1on . This sequence was CletarminaO by clidaox.Y MQuance anatr- of a Ml3mp2 Elhage containing !tie 717 be Psi fragment covenng thiS rag tOn (aH Figura 1 in St.,nmatt et al .. 1118 t ). The cloning of lt\11 fragment into the Eco Rt lila ot M13mp2 and 1t1a DNA sequence anatysos were perforrn.d u clncribed in Experimental Ptocedur" tor fr.;menta Wived from Clone 27 .1. strain-that is. in the Oa-2.3 region. The exact location of the recombination point in strain B6.K 1 between the H-2D and Oa-2.3 regions is not known. Thus gene 27 .1 may be either a Oa pseudogene or a pseudogene for a transplantation antigen mapping within the D region, but to the right of the D-mai'Xer locus. Soloski et al. ( 1981) have shown there are peptide homologies between the Oa-2 antigens and transplantation antigens. These protein homologies together with the mapping of gene 27.1 therefore raise the possibility that our eDNA probes may allow us to isolate not only genes encoding transplantation antigens but also genes encoding Oa. and perttaps other genes for lymphoid differentiation antigens encoded with the Tla complex (Figure 1). Thus the 30 to 40 genomic clones that we have isolated from the sperm library may include genes for Tla differentiation antigens as well as those tor transplantation antigens. The Intervening Sequence between the S.venth and Eighth Exona of Clone 27.1 Ia Pr.-.nt In the eDNA Clone pH-211 A comparison of the eDNA clones pH-21 and pH-211 shows that these seQuences. though encoded by distinct genes. are highly homologous except tor a 1 39 bp insertion found in clone pH-211 at a position corresponding to the last codon in clone pH-21 (Figure 8). Because of this insertion. the coding sequence of clone pH-211 is 1 3 codons longer than that of clone pH-21 (Figure 8). When these eDNA seQuences are compared with that of gene 27.1. the 139 bp insertion is found in the same alignment as tor clone pH-211 and corresponds to the last intervening sequence in clone 27.1 (Figure 9). This intervening seQuence in 27.1 ia also 139 bp in length and 90% homologous to the corresponding pH-211 eDNA sequence. We isolated a 1 38 bp Hph 1-Pvu II fragment containing only sequences from the 139 bp insert of clone pH-211 except tor 1 1 bp. This fragment does not contain the repetitive elements identified on the 3' untransJated region of clone pH-211 (Steinmetz et al.. 1981 ). When this fragment was used as a probe in Southem blot analysis of five different genomic clones. hybridization was noted tor each of the five clones. auggesting that this seQuence ia contained in all of these genomic clones . Moreover, this intervening sequence has ~ stream and downstream RNA splicing signals at its 5' and 3' ends. Several explanations for the synthesis of pH-21-like and pH-211-like mRNAs are possible. It is conceivable that the pH-211 eDNA clone wu derived from a partially spliced mRNA with the last intervening seQuence still present. Clone pH-211 would then represent an intermediary splicing Pfoduct and not a functional message. Alternatively, the retention of the last intervening seQuence could also be of physiological significance. A differential RNA splicing mechanism might operate at the 3' end of class I genes allowing the aynthesis of proteins with ITI8litedly dif- 43 ....N ~ _. II) ..... I <X ~ I la:: \0 ·- . -.... -... ~ II) II) u \0 II) .. (+I •-rtttc=::::::===:=::~-- ::cc::i:as:===== .. •- • ......-t,c: • •.• :: "---9.6 •. !!-= =;: • ••• • ·. . \.. ..· •· . .,., .•. - ... -1:....- -- - - -7.1 1(1 c::======== •A ca..,.. ......_... 016 GAI(II tiiCiflt(lrl, .. lf'lltf '-'160< A Figure 7. ~~~lapping of Gene 27., by Soulflem Blot Alwlyai& of R• c:ombinlnt Congenic Straina of Mice , •.· A llmhed map of mouN c:hr~ H lalhown. The origin of !he or atipe>ling The ~or 8bMnCe of a g'-' antigen (phenOtype) • indicated by ( +) or C-). 116-Tla• il a congenic .rra;n wi!tlltle Tla region of A on a M bllckground . 116 I( 1 and 86 K2 are rKOmbinant atraina danYed from 116-Tia• 8ftd ~2". The ew-e or -..nee olltle 2 .7 kb Bam lregmant In the Sou1hem blot ·~("Figure 6) IS inelocated by C+ l or (- ) lrldteta tor l!le and A inelocate that !he Southern blot ~~~ (not lhown In Figure 8) . . . per1ormed 1M IUbllrain 81 0 , and A/ J, ..nic:tl il of 1M ..,.. l\llpiOtype aa 116 and AI A. ~y The~ map 1M gene 27., to the right of ltle ~liOn - t In M .K1 8ftd to the left of 1M recombinatiOn - . t . , l!le K2 . F'og.n edapled from fw.arty <'MOl . ~ Ia ~ed by the ~ 8hading ~ 2.6-.- -2 .6 2.2-- ---2.2 • ~~~ (+I ... ~ ,;, ·.;_ - ... 7.1- ~ ~ H,0r ldllOhOt'l tO .,.zo ec.-Z.Oo· !IIT•. Oo;l 2.'1.. e.... . . _ r-s,o.,,...,n TL' o• 0.·2' II) • 9. 6-~ ·-~ _ • r et al., 1981) of atriking homology (up to 50%) between Figure IS Analyaia of 1t1e ON..._ from Six Dlf*errt Inbred Strains of Mtee by 5outNm Blot Hyb<idoutiOn tne DNA seQuences of the third external domain of transplantation antigens and tne constant region dcr Five micre>Qrama uc~ of lrWf ON" iaolated from 1t1e lix mouu ~ .rr11n5 - • completely Clige&ted wi!tl !Sam HI . .eparated on a 0.5'!b egaroae gel. tranaferred to 1 nltrocefluiOM filter and hybridiZed wi!tl 1M a..bclone pH-211a u a prot. ("F~gure 2Al. Thia clone wu Cienved from pH-211 by a..bcloning 1M c:odonQ regiOn and 1M pot1lon of the 3' untran&Jated reQIOn a.IIOid of ltle r-tili\18 NQuencea Into o8R322 CSte1nmetz at al .. 11181 ). HybndizatiOtl condiliOna. WUhong and the hybridization mart.er were u a&aclibad ~ (SteinmetZ e1 al .• 1881 ). The tan of 1M mari<ar DNA fragmenta are ~ In kllotluea. mains of immunoglobulin molecules. The availability of tne 27.1 genomic seQuence allowed us to extend the sequence comparison into tne Intervening aeQuences flanking the third external domain of gene 27 .1 . When compared with tne fourth constant region domain of tne I' gene with Ita flanking seQuences. moderate homology of 35% was found for the first 75 bp of tne 5' flanking seQuence of the third external domain without placing any seQuence gaps into the two seQuences compared (not shown). No significant homology was found for the 3' flanking seQuence. In · contrast to tne third external domain, the exons for the first and second external domains of tne transplantation antigens do not at10w significant homology to immunoglobulins and show only marginal homology to one another (3<&% homology between the exons coding for the first and second external domains, 30% homology between those coding for tne first and third external domains and 29% homology between those coding for the second and third external domains). The genes for transplantation antigens and antibodies both show a atriking correlation between atrvctural domains of the proteins and the exona for the correaponding genes. The class I genes of tne mouN constitute a multigene family with several interesting features. One auch feature ia that they appear to be diatributed ferent cytoplasmic domains (aee Early et al., 1880). In this respect, it is interesting to note that the Cterminal portion of the Kb molecule is atrikingly homologous to the translated seQuence of gene 27.1 except for the last seven amino acids in this polypep.. tide chain (Figure The breakpoint between tne homologous C-terminal seQuence and the last Mven amino acids occurs precisely at the boundaries of tne teventh and eighth exons. where the hypothetical aplicing event would occur. •>- GenM for Transplantation Antig ens and Antibody Molecules Are Homologous In Thelr General Structural FeaturM A computer comparison of the exons of gene 27.1 with exons of the immunoglobulin gene confirmed our previous finding for the eDNA clones (Steinmetz c. 44 r---------~~3~9~b~p~·"~'e~r~'·~o~"--------~ 313 PH-2 U 313 PH-2! TERM 40 w0rl01t()tli tf'l 272 be 85 •te hOmQIOQOUS 3 •O' ·:l' ·Ot'\ - ~ e~cp 96 •t• 1'\QI'f'IC 10QOuS Fogure 8 Scl\emahc Comparo.o<~ of !he DNA ~uencea for IN 3' R"!!'on& of !he eDNA ~ I)H-21 and I)M-211 The seQuences oreMf't on clone IIH-21 and pH-211are homolc>goua exceQ1 for a 138 b!) inMmon in clone !IH-211 (Steinmetz et aJ .. tall! and Figure 5> The lemu"atoon coelons are boaed Clone IIM-211 has rwo seQuence gapa . one of three nucleotide& and t i l e - of one nucleOtide . and clone pH-21 has a Single gap of rwo "ucleohde& lflat - • introduced 10 ach....,. maximum IIOmology . Nucleotldes on !IH-21 lflat diffe< from !he I)M-211 MQuence are 1nclocated by ~ t>ars. pH-2 n 27.1 pH-2I Figure 9 Two Mypo-I>Cal RNA Splicing Pettema at the 3' End of ~ne 27 1 That May Generate I)H-211-Uke and I)H-21-Uke Mesaenee< RNA Molecule& For gene 27 .1, tour eaons encoding the ''"'"external OO!naon . IN transmembrane domaon and tile cyto~&miC OOmaln (black boaH) U welt as the hrst pan of !he 3' untranslatea regoon (hatched box) are at>own The pH-21-!ike mRNAs would be -ated by !If\ RNA IQitce lflaf would remo .. 139 bp "'tro" ...Cicatea by !he inMned OPen boa . 1n pH-21~-~H~e mRNAs. IIlli intron .ould not b e . . -. -•aa across as much as 1.3 centiMorgans of DNA (several million nucleotides> (Boyse at al., 1965). if the Tla genes are homologous to those encoding transplantation antigens (Figure 1 ). If the Tla genes are homologous to the class I genes. perhaps the boundaries of the H-2 complex should be enlarged to include the Tla complex. Another feature is that the class I gene family is interrupted by the class II and class Ill gene families . It will be interesting to determine whether two or more of these three gene families share a common ancestor. Since the class I gene family is contiguous in other mammals, perhaps the organization of the H2 complex in the mouse reflects a high level of av~ lulionary gene duplication, deletion and rearrangement in this gene family . Finally, the class I and antibody gene families appear to be members of a supergene family as reflected in their homology and their similar organizational features . These gene families use distinct regulatory strategies. since DNA rearrangements and somatic mutation occur in the antibody but probably not in the class I gene family . It will be interesting , however. to determine whether there are still other types of regulatory mechanisms shared by these two evolutionarily related gene families. ....... Aeatrochon ~ - . purcha.Md from New Enogland BioLabs. lett>esaa ReMarch L.aboratones and Boellnn&er Mannheim . The 1 2 nucleotide pnmer UM<l for DNA MQuenconQ wu obtained from Col-.oo<an.. Research ana the C)c!NTPs from Collaboran.. R...are, and P~ Boocl\emicall . E. COli DNA polymerue I tatve rr.Qmen1 wu Obta1ned from Boehnnoger Mannherm. n DNA polymerue from P~ Biochemocala. E. COli llfail1 JM103 and phaQe M13mp7 from S. llleeda Rnearcll L.Aboratonea. T ~ DNA li;aaa from Hew England 8ooL.abs and . . .P-c!CTP (10 mCi/mf. >400 Ci/mmole) from""*sham The E . .eoli stra.n JM10t wu obtained from J . Mesalnll. and the Phage M I 3nip2 wu CICIIainea from J. o.-. DNA s.q...,_ AftalyM A 3 I! kD BQI ~ and a 2 7 kb Bam HI hgment including the COCiing negoons - • ~ted from done 27.1 by digeshon of ~ I'll P"aQe DNA wolll the ~te enzyme. gel electroclhorail on a O.S'IIo agaroaa gel and electroekllion of the desired rr.;- lollowec:l by 80-celluloM purtlleation For Cloning ana MQUM>CinQ by the M 13 dideoay eaQUenCing techniQue (Sangef el al .. 11180: MeeeinQ el al .• 1 all! l. we d!Qftted the ilolated rr.Qmenta wot11 a number of I'Mtriction enrymn lflaf - kiiOWIIIO cut DNA "*"*'tty The .._.mng rr.Q.ccorOing ""xtufft -• c:1on.a into M 1SanQef 3 wec:tcn and MQUenc«S 10 the proc..,...... CIHcr1bea el .. . ( 11180) and In the following by protocol . ~uencn - • aligned by a computer MWell program wrinen by T. Hu~ . OunnQ the courwe of INa wort< . . . louncllt . . - ana 1u1e< 10 clea .. the 1erve DNA ~ta 1n oncy a .._ plecM. using lnfr-..nt cutting restriction ~ . and 10 done the rnu!bnQ . relatt..eiy lOng DNA fragmenta to gel ex~ ar.!CtiM of in-- ~a . If cert- fregmenta - - not found c:ollecliona. they - - indiYidually ilolatea by gel electr~ and !hen c:IONd anc1 MQUeneed . For the 2 .7 kb Sal" fragment. . . ...a a " - ! euftar'l !he enzvm- Alu I, HM •. Hint I, Raa I ana S.U 3A and a ~ cutten IN ~ PYu I and Sat I. The 3 .1! kb lot I ~ wu ctu¥9d with A1u I, HM •. Hint I. Raal ana S.U HIM ~ c:un.,.. ana tor the ;en«ation of 10ng tragmenta . . , lllillturMoflamHI +~I+ XbelanaBemHI +,...,I. Aa 1 CloninQ ...:tor. - 11rwt UM<l M1 3mc17 with the E. COl . . , JM103 u '-1 tM lwltCile<l later 10 M13mcl2 wilt\ JMI01 a ' - l. About ~'llo of the mc>7 Clot-s !'hal- pocked gaw fWn! MQuendno Pllf!ems lflat - • not readable . Thilwu not~- the 11102 ~ . For donlng , the Eco Rl-cler..a M13mcl2 DNA and the tragmenta to be c:IONd-. rnact1 fluel'>.4nded by lncuCabcon wilt\ T~ DNA potyme<ua ana !her. - . blunt.-d ligated (Wartell anal'laznll<off, t MOl Phage Clot-s corrt»->tnQ genomic DNA_. octenlltled by ~ hybrldlu~~on (8enton ana O.Yia. 111n ..., <*IN 21 .1 UM<l .. a probe. I" or MQuenc:ing. tortleth of !he DNA ......., fi'Cifll a 1.15 11'11 ~ cs.ng. et al .. 11180) .m&aad ..., 2.5 ng ola 12 nucteolidl 45 prtrner DNA ~t . 2 .5 .,C. ol•~(~el el .. 1N0) ~ 0 .25 unit o! E . COlo DNA I)Oiymerue I~ hgment in a final ..a1ume o! 5 fJ1 In 7 111M T,._HCl (pM 7.5). 7 111M lolgCI,, 5 rn1o1 Clltuot!>tenol. and 50 lftM NaCI. Individual reacnona were aupp~e­ _,ted wtll'lllle following c~anona ot unlabeled nucleotidea: A raection. 11 ~ cSATP and 100 ~ cSCIATP; C I'NCtoon, 50,.M cSCICTP; G raactoon . 1.1 ~ CIGTP and 1DO ~ cSCIGTP; T reaction. 1.1 ~ cJTTP and CCXI~ cScJTTP . Eaell r. .ction alSo contained aach of the remainong lhi"M cSNTPa at the ~ation of25-33 ,M . The reactions were incubated f"' 15 mtn at 30•c . then cllaled by 111e acSCiitoon of 1 IJl of a ao4ution comaoning all lour dNTPa. eac:ll at a conc:entration of 300 ,M Allar another I 5 min inc:ubation at 30•c . 10 ,.t of a eolutoon contaonmg 115'1b lorrnamtcSe . 10 mM EOTA CPM 8 .0). 10 mlol NaOM . o .o3q,: •yiene cyanol and 0 . 03~ twomop~>enol blue - e acSCied ana the umples . .re atored troz.n. FOf ele<:trophornia. the umpfes - e treated'"' 3 mon atiiCJ•C and Quicl<-cooled on a-water ~ 2-4 ,.J were loaded on co em !l'lb and !10 em 4'1b or 5'1b acrylamiOe a.Quencong gets (Smrth and Calvo . 111!10) E..q)osure waa usually _.,.ght wtll'l irrtenaitying acr.en . To fMd Individual MQUenCH beyond POsition 200 . H was uaualty - r y to c»creaaellle cSCINTP eonc:entratoon on the reactoona by a fac10f of two to atght. As indocated on Figure 2C , lour neg iOns of the 2 .7 kt> Bam fragmen1 ~ a pan of the 3 .8 kt> Bgt II fragment also - • MQuenc.d by the Cflemocal c»gradatoon method (loluam and Gilbert. 11180). Fnegmenta -re labeled at theor 3' ends wtll'l E. COlo DNA potymeraae I large tragment (Smrth and Calvo. 111!10) and then divided by aecondary restriction enzyme c:IN..-.Qe For !he G -+ A reaction , !he procedure dHcnbed by Gray et al. (11178) wu IMCI. 1ao18t1ot1 o1 E YC¥YOtiC DNA 8fld DNA lllol Hytwldlutlon DNA was isolated !rom froze" rnouae ~ according to !he proc• Clure OHcnbed by Bltn and StaffD"' (I 1176). except !hal the liNuM - • pUI~ed in a mortar in the prewnce o1 liQuid nitrogen. and 1Wo cf'llorolorm-isoamy1alcoh01 (24 : I) axtraetionl - • acSCietd after !he phenol ••tractoons DNA blots (Sou!hem . 11175) were pr-red and hybrodiZed accorcSong to !he conditione ~ OMcnCied CStaimnett at at.. 11181 ) . .,.__ £ . A.. 0111 . L. J. anc:r Seoc:tlart. f . (1115) The Tl nt.ymus r.ukemoal antigen . A,..... _ In lm~!l>ok>gy . P Graber anc:r P . A. Wieacher. ecll. (New Yort< : Gruna and StranonJ. IICI- 23-40. ~- F., Abaatado. J . P., kvlal. 6 ., Aul< . L.. Llllanna. J . L .. Oerot!. H., Cami, B .. Wiman . K .. L.amammar. 0. , Petaraon. P. A .. Gac:Miin . G .. kourllel<y . P ~ Oobberwtein. B . (11181) Structure of C-lemlinal hall 01 two H-2 antigens !rom Cloned mANA . Natura 2112. 7&-81. c-. B. ,~- F .. ~. J. P. and kour1111<y, P. (11181) t.lultiPie aaQUenCM related to CIUaical 1\ist~ltbtloty ant.gena in ..,. mouae genome . Nature 2111 . 873-175 . Collgan, J. £ .. kincll . T. J., Uatlara. H., Wal'finl<o . J. and Nathenson. S . G . (11181 J. Primary lltTUCture o1 a murine trar\al)lantatoon antogen . Nature 2111 . 35-311. Davia. W W., c.lama. k .. Early , P. W .. Uwanl. 0 . L .. Jollo . R .. WaiUtnan. I. L. and Hood . L. (111!10) . An immunoglobulin llaary<llaon gene Ia fonned by at INSt two racombinabOnal events. Nature 283. 733-7311 Dobt>aollleilo, B .. Garoff. H , w.....,, G . and Robinaon. P J . (1117gl ~ and membrane lnaertoon of mouse H-20' hiat~ compallbiffly antigen and ,62-<nictoglobulin . Call I 7, 7511-7611 . c.»-trae Early , P .. Rogers. J., Olivia. W., Calame . K ., lionel . W., Walt. R . and Hood. L. (111!10) . Two rnRNAs can be produced !rom a aongle omm ..... noglobulin I' gena by allamallve RNA ~ palhwaya. Call 20. 313-3111 . Flaharty, l . C1871l. The Tla ~of the mouae: ldanllf>cation o1 a new .-ologically "-ffned locus, 0..2. ~ICS 3, 533-5311 . Flaherty, L. (111!10). The T1a negion antigens. In The Role of !he 1o111or Hlatocompallbiljty Compia•ln lmlnunoiogy, W. E. Oorl, acl . (New Yor!< : Ganancl STPW "'-l.IICI- 33-57. Gray , C. P.. Sommer . R .. Polka . C .. BIIC1< . E and Sctlaller. H . (187!1) Slructure of the Origin of DNA ~lion of beclanopllage !d . Proc . Hat. Aced. Sci USA 75. 50-53 . Hlmmer1ing. G. J .. Hammer11nQ. U. ~Flaharty, L. (111711) Oat~ anc:r Oat·5, ,_murine T<*! anbgens ~by !he Tla negoon and identified by monoc:lonalll'lllbodies. J. Up . lolled . '50. 1 06-116. Jordan. 8 . R., ~··F . and Kour1111<y, P. (1881). Human HLA We !hank Tim Hunllapiller tor COft1C)Uier programs. Wiellaaf Oouglaa tor compUtllf gral)hocs . Bemrta La,., lor praparat>on of the manuacripl ~ Karyt lolonara fOf ••~t tachntCAI -lance . We alae !hank Dr . J Oroum for his advice reeartling cSiGeo•y ~ in !he ... , 3 ayat&m lol S. Ia the racil>tent of a lellowsNQ !rom !he Davtac:ha Forschungsgemeinachall and K W . lol and J . G F . are IUPP0'1ed by Natoonal lnatrtutes of Hearth ~cSoc:fotal ~ B. T. S. Ia a Natoonal Science Foundatoon '-~~ow . E . A. 8 . Ia AIMnean Cancer Soc:oety Reaearcll Pro!euor of c.l1 Surface lmmunoganetlc:a. Tllia wort< was supported t>y granta !rom the Nationallnatrtut• o1 HNI!h. The costa of pUblication o1 !h,. artiCle were Cletraye<l in pan by !he payment of page cllarges Till& article lheratora be '*-Ciy marl<ed ·.avenrsement "' in accordance wtll'l 18 U.S.C . Sec:bon 1734 ao1e1y to indiCate lhoa tact. Aec:eN.cl Jvty I, 1 1181 ,..,..aew oane -.grnent iKlllted by llybridlz&tion wt111 moo.. H-2 c:OHA proOes. Nat\n 2110. 521-523. KJam. J. (1875) Biology o1.,. ~ Hiatoc:ornQallbill!r Complex (Bertin . Springer Yatlacll- kvlal. S. ~- F., Raaok . L .. c-. B .. Garoll. H .. o.n.t. F .. Wiman, K , t..amammar. 0 .. Abutaclo, J. P., Gecllalin. G .. Peter.on. P. A., Docbe< lleifl, 8 . and Kour1111< y . I'. (111111 ) c:0HA Clone COCiong tor part of I IIIOUM H-2" lnljO( Nat~tiblll!y anbgan . l'roc:. Hat . Acacl Sci. USA 78. 2772-2778. Lawin .• (11180) All~ tor aplicing : rKCIO'oizinG . . . . . . of lnlrona. Cel22. 32C-321. Lcc>az de c:.tro. J. A.. Orr. H . T., flotlt>. R. J., koalyt<. T .• IINnn. 0 L. anc1 SlrOIIW>gar, J . L. (111711). Complete amino of a ac:.c- ~acl""""" 1\ist~libiltty antigen . ~7 . 1.111>­ lafion and amino add ~ o! frag....ms and ol tryptic and Chymotryplic pap!idea llioc:hemtatry 18, 5704-5711. Waxarn. A. W. 8fld a..ert. W. (11180). Sequancjng and lllbeled DNA with baM~ c:harNcal ciMvagaa. IHtl\ . Enzymol es. c--.seo. Anc»r-aon . S .. Gall . W. J .. lola yo! . l . ~Young , I. G . (11180) A aflor1 ~ - J., CrN . R . and SMOur;. P . H . CIIM!Il A~ tor prime< lor ~ong DNA Cloned in !he alngi.-atranclecl phage ...c:tor lol13mcl2 . Nucl. Acoela Rea. 8. 1731-1743. lftc4;un DNA aaouendng Nucl . Acide Ra. 8. 3C»-321. flllcllaalaon. J .• Flaherty . L .. Vltana . E and Poult. . w. (1 ern. ftolatK. ...., aimllarltin b e - the 0. 2 aloanliQen and otlle< gene II'OCiuc:la olllla 17tll ~ ol .... mo..a. J . E.xp. Mac! 145, 10116- Benton. W. 0 . and Olivia . R . W (11177). ScrMnong ~gt ~ aor- by 1\ybridiz.atlon to aongle " ' - - in all~ Scoence I H . 1110182. 1070. Blin. N. and Stafford. 0 . (11171) A oana<al melhocS tor ~ o1 high moleCular -...ght DNA !rom -.rrot•- Nucl. Ac:iQa Aea. 3. On. H . T .. ~ - 0 ., Aoet>. R. J ., ~de Caelro. J A. and SlrC>n'llnger. J . l. (111711). ,.,.. l"tNVy cnairl o! human histocompa~ 2303-2308 . • ~ HV47 c:omaona an~ ,.goon. Nature 212. Boyae . E A . Old . l . J . and Luell . S ( 11114 l Genetic del.,.._ o1 :le&-270. ..,. TL (tllymus lauk.,..l anllgen in !he II'IOUM . Nature 201 , 778. l"'oeggl. H. L.. Cennon. l . E. ~ ShbiiW..,_ , J. L. (1818). Cal-lrM 46 ..,.,I()<\ of the lftR"'A& tor the heavy and logh! Chains of HLA-A ...., l'roc . Hal Aced Sci . USA 76. 227~2277. ~ 8llligen& ~h . H L .. On . H . T end S!Tomrnger . J L (1180) Molecu18r Clonon; o1 a ....,.." hiat~tibitrty enhgerl c:ONA ~ - Proc . ,_, Aced Sci USA 77, 1081-eo&~ . ~h H L . On. H T. end S!Tomrnger . J L (1NI) Mato< lliste>_,pallbolrty anhgens · t h e , _ (Hl.A· A. -e . -<;) end.....,.,. (H-2K. H-20> claaa 1molecules c.n 24. 2117-Ne ~~ . N J end BrowniM . G G (1876> 3' ~ NQtOI'I .-QuencH in eullaryotoc ~RNA Nature 253. 211-214 . S.nger F., Coulaon . A R . larrell. B G . Smrm . A J H . end Roe . B A ( 1180) CIOnmg ;., smg•strancled ~cten~ u an 8icl to ...0 DNA eeQuenetnQ J Mol BIOI 143. 161-178 &mitt>. 0 R. end Calvo. J M (1880) H u e - -..nee o f - E. COl• ~ eoc1onQ tor dih~otolate ..cluCIUe Nucl Aacll Aft e. 22~5-2274 loioal<o. Ill . J .. Uhr . J W .. Flatleny . L. end Viletla. E. S (IN I). Qe2. K-2K . end K-20 ..oanhgerls ~ from 1 ~ ~~ ..... J . ~ Med 153. 1010-1083 Sood . A K .• Pererra . 0 end w.......n. S 111 !1N1l lloia!l01'18nd per1aa t nvcleot.oe NQuence oil c:ONA Clone for human i'toSIOC:Ofnl)el· lbiliTy ant;;.n HLA-e by ..ae olen ~x)'ft..cleOIICie Hal Aced Sco USA 78. 616-620 . · l"'roc . Soutt.n. E Ill ( 1175) O.tectiotl ot ~ aeouence& among DNA lragments ael)&rated by gel eleciTOPIIO<~ J . .. ol a.ot. N. 503517. Slanton . T H. end Bo)'M . E. A. (11715) A,... -~c:alty ct.flned IDcua in the Til region of the II'IOUM . De- 1 . lrnmunogenellea 3 . !>25531 . llanlon . T . H end Hood. L (liMO) ~ dlntlflclllor> of the Qe-1 llloln!lgen llnmunogene!Jc:s II . 305-314 . Sl..nmetz . Ill . Fre!ang., , J . G .. Fllher . 0 .. Hur'lklpiller . T .. ~1 . D .. w..uman . S Ill .. Uehara . H .. Nalf>*naon . S . and Hood . L . (1881) . ThrH eDNA Clones encodang mouse trana~>~amatoOtl an~~gens · 110IIIOiogr to ommUMOQiobulrn genes Call 24 . 125-134 . Slrominger. J L .. On . H. T., P'artlam. P., l'loegtl , H . L .......... D. L .. -..ofsl<r . H .. S.rotl . H A . Wu. T . T . end Kabel . E . A. (1180) An evlluahOtl of 1tte a.gnafocance of aminO acid ~ homologoes .., lluma~ hllloc:ompallbohf't ln10Qenl (HLA·A and Hl...'-el Willi ommunot'Obu~MII and otllet' protem& . .-.ng reilhwet~ anon~ - Sc8nd . J Immune! II. 573-512 War1ell . R Ill and Razn•otl . W S . (1180) C1on1nQ DNA~ endonucl. .M fragmerota Willi protruding -...g-..oed ano.. Gene 1. 307-311 47 Appendix B DNA Sequence of the H-2Ld Gene. 48 The H-2Ld gene was the first transplantation antigen gene to be sequenced. The publication contained in this appendix originally appeared in Science. 49 Reprint Series 5 February 1982. Volume 215, pp. 679-682 SCIENCE DNA Sequence of a Gene Encoding a BALB/c Mouse Ld Transplantation Antigen Kevin W. Moore, Beverly Tayor Sher, Y. Henry Sun, Kurt Andrew Eakle, and Leroy Hood Copyright © 1982 by the American Association for the Advancement of Science 50 DNA Sequence of a Gene Encoding a BALB/c Mouse Ld Transplantation Antigen Abstract. The sequence of a gene, denoted 27.5. encoding a transplantation antigen for the BALBic mouse has been determined. Gene transfer studies and comparison of the translated sequence with the partial ami11o acid sequence of the Ld transplantation antigen establish that gene 27.5 encodes an L" polvpeptide. A comparison of the gene 27.5 sequence with several complementary DNA sequences suggests that the BALB/c mouse may contain a number of closely related L-like genes. Gene 27.5 has eight exons that correlate with the structural domains of the · transplantation antigen . The transplantation or class I antigens of the major histocompatibility complex (MHC) are present on the surface of all mammalian somatic cells and play a key role in T cell immunosurveillance (/, 2) . In the inbred BALB/c mouse, three genes for transplantation antigens. D . L. and R, are closely linked and these are separated from a fourth class I gene. K. by approximately 0.3 centimorgan (3) . We cloned 30 to 40 genes of the BALB/c mouse (H-2d haplotype) that are homologous to complementary DNA (eDNA) probes for transplantation antigens (4) and have demonstrated by gene transfer experiments that one of these clones (27 .5) encodes an L d transplantation antigen (5). We have determined the SCIENCE . VOL 215. 5 FEBRUARY 1982 nucleotide sequence of the L d gene present in clone 27 .5. The genomic clone 27 .5 was isolated as previously described t4l from an amplified library of BALBic sperm DNA cloned in the A vector Charon 4A . The nucleotide sequence of gene 27 .5 was determined by the dideoxy sequencing technique with M 13mp~ as cloning vector (4. 6, 7) . The sequencing strategy is shown in Fig . I. The exons of gene '17 .5 were defined on the basis of homology to available amino acid sequences for transplantation antigens (8. 9), to several DNA sequences of eDNA's encoding transplantation antigens 110 . II) . and to a genomic class I clone ~7 . I. which bears a pseudogene mapping to the Qa-:!.3 re- 0036-8075182:0:!05-0679SOI .0010 Copyright: 1982 AAAS 679 51 gion (4). The DNA sequence of gene 27.5 is given in Fig. 2. Gene 27.5 has eight exons whose boundaries correspond precisely to those determined earlier for the class I pseudogene 27.1 (Fig. Ic) (4) . Apan from introns I and 3, the lengths of the introns in genes 27.5 and 27 . I are similar. Ex on I encodes the signal peptide : exons 2 to 4 , the three external domains : exon 5. the transmembrane segment: and exons 6 to 8, the cytoplasmic domain. There is a striking correlation between the discrete exon boundaries and the structural domains of the transplantation antigen (4). All of the exon-intron boundaries have the consensus upstream or downstream RNA splicing signals (12). Gene 27.5 appears to be a functional gene by sequence analysis in that it lacks any obvious elements that would render it a pseudogene (for example . termination codons or inappropriate reading frame shifts) . This conclusion is supponed by our gene transfer studies !5). The translated sequence of gene 27.5 is identical to the amino acid sequence of the Ld molecule at 77 of 77 positions that can be compared !Table I) (13). These comparisons include residues in the first. second, and third external domains. Although the paucity of amino acid sequence data permits us to compare only 21 percent (77 of 358 positions) of the gene and protein sequences. these comparisons suppon the conclusion reached in gene transfer studies that clone 27.5 contains an L d gene (5). There is one striking observation that can be made in comparing the sequences of gene 27 .5 and of two eDNA clones, pH-211 (10) and pH-2d-3 (I/), whose translated sequences are identical with the available panial Ld amino acid sequence . Gene 27.5 is identical to eDNA clone pH-2d-3 at 467 of 470 positions compared. Two of the sequence differences are in the first 60 nucleotides of the founh exon and lead to one codon substitution. The other difference, which also leads to a codon substitution. is in the fifth exon . Likewise, the exons of gene 27 .5 are identical to clone pH-211 at 514 of 520 nucleotide positions. The differences are clustered in the first 45 nucleotides of the founh exon and lead to five codon substitutions . There a1'e also six nucleotide differences between Table I. Homology of the exons of the translated L d gene (27.5) to the corresponding regions of the translated pseudogene 27.1 (4) and to the K" (/])and Ld (JJ) molecules . The homology of the translated portion of each ex on to the corresponding portion of the K b and L d sequences . as well as to the translated 27.1 pseudogene exons 14) . is shown. Homology(%) Ex on Leader (e xon I l First domain <exon 2) Second domain fexon 3) Third domain (exon 4) · Transmembrane domain (exon 5) Cytoplasmic exon First (exon 6) Second (exon 7) Third fexon 8) Kb Ld• 27.1 84 100 100 100 76 71 76 89 69 80 88 73 73 69 100 100 64 0 •Thirt y-four of 90 posit ions were compared in exon 2. 28 of 92 in exon 3. and 15 of 92 in exon 4. a p B 8 SDHSD K SP 8 ox 0 B HA~ARA Bg ARA R p R B - b 27.51 c 27.1 2 II 4 3 2 3 (2070) 4 II II II 7 II 7 Fig . I. The organization and sequencing strategy for gene 27.5. (a) A restriction map of clone 27 .5. This map was generated by Southern blot analyses of clone 27 .5 cleaved with various restriction enzymes and probed with selected M 13 clo nes into which fragments of gene 27 .5 had been inserted (see below). The restriction sites are designated as follows : B. Bam HI; P. Pstl : S. Sau 3A: D . Dde 1: K . Kpn 1: Bg . Bglll : R . Rsa l; H . Hinf I : X. Xba 1: and A. Alu I. (bl Sequence strategy for gene 27.5 . Each arrow represents the sequence of an M 13 clone. The M 13 clones indicated by asterisks were used as probes in Southern blot analyses to generate the restriction map . At positions where clones do not overlap . alignment was determined by restriction mapping or by homology with eDNA clone pH-211 . (C) Orga ni zation of genes 27.5 and 27 . 1. E xons are represented by boxes and introns by lines. Both exon s (above boxes) and introns <below lines) are numbered . 680 gene 27.5 and clone pH-211 of the 3' untranslated region of 587 nucleotides compared . It is interesting to note that the differences in coding sequence are clustered at the beginning of the founh exon. because this exon encodes the third external domain of the Ld polypeptide, and this domain is the most conserved region of the genes encoding transplantation antigens ( /4). The genomic clone 27.5 was derived from BALB/c (H-2d haplotype) sperm DNA . The eDNA clone pH-211 was derived from a BALB/c tumor cell line ( /0), and the eDNA clone pH-2d-3 was derived from a DBA/2 CH-2d haplot ype) lymphoma line (II). It is unlikel y that all of the differences in these class I sequences arose from cloning or sequencing anifacts . Accordingly , mice of the H-2d haplotype may have three closely related class I genes that encode L d-like polypeptides . The three L-like genes might arise from genetic polymorphism at a single locus within mice of the H-2d haplotype or from duplicated genes. The possibility that these different clones represent duplicated genes is attractive in view of historical precedent and of recent evidence. By serological analysis. the D end of the H-2 complex initially appeared to have a single D gene (2). Subsequently , more refined serological analyses have suggested that this region encodes three closely linked genes for transplantation antigens-D. L, and R (3). Cosmid clones that have been isolated from BALB/c DNA contain three L-like genes according to restriction map analyses (14) . Gene transfer studies have demonstrated that at least one of these cosmid genes encodes an Ld polypeptide (15). It will be interesting to determine whether the remaining two of these putative L-like genes also are expressed as class I polypeptides reacting with the monoclonal antibodies to the L d antigen . A comparison of the translated sequences of pseudogene 27. I and the Ld gene 27 .5 with the amino acid seque nce of the mouse Kb molecule (Table I) shows that pseudogene 27 . I appears to have diverged significantly more from the translated Ld gene than the Kb molecule has (Table I). The difference between 27 . I and the other class I genes might arise because the genes encoding Qa antigens diverged from those encoding the classical transplantation antigens before the divergence of the individual class I genes . Alternatively. this divergence may reflect changes accumulating in the pseudogene. which has presumably been released from selective pressure . In addition. the exons of pseudoSCIENCE . VOL. 215 52 gene 27 .I are homologous to those of the Ld gene 27 .5. This homology suggests a common evolutionary origin . If gene 27 .I is a Qa-2.3 pseudogene [see (4) for discussion), then this homology suggests that the Qa antigens are class I molecules and that the Tla complex should be considered a part of the H-2 complex. This conclusion is supported by observations that the TL and Qa antigens resemble the classical transplantation antigens in size, peptide map profiles, and their associa- tion with ~microglobulin (/~/9). We have used a computer-generated dot matrix to analyze the homology relationships of the DNA sequences from genes 27.1 and 27.5 (Fig. 3). This analysis compares every hexamer of gene 27 .I (vertical axis) against every hexamer of gene 27.5 (horizontal axis) and places a dot in the two-dimensional matrix at positions where at least five of six nucleotides are identical. Homologies are displayed as diagonal lines in the matrix; nucleotide divergences are represented as gaps in the diagonal lines; and sequence insertions or deletions offset the diagonal lines (20). Four important points emerge from this analysis . (i) Genes 27 . I and 27.5 are quite homologous to one another, as can be seen by the strong -diagonal lines . (ii) Five of seven introns (2, 4, 5, 6, and 7) are almost as highly conserved in both genes as the exons are; the exons exhibit 87 percent homology, and these introns ex[lCON I •t 418 Pro ,..g Thr UY Leu Leu Leu Leu CCA9GGCGG.ATTGAGA8GG.UGACCACACACCCTGT&.A&CTCACT&T5TCCA;T;A&TAGCTiiiCACT~TCCACAGCTCACTCC.t.Ga&.ATCACTCCCAQ.I.T~S ATG QCT CCG C.C M:G CTS CTC CTG CTG CTG t.O All All Al8 Trp Pr-o . . p Ser AlP Pro Ar-g I 8CG act QCC T&6 C'CC 8AC TCA GAC CCG aN i IT-...T8$66CiT&C8&Hi4N""4CCVCTCT'iCSQ9G•G•W6&&C&QC•GG&GG*ASC'K' ... 8ATCCCCTcaccTC&C.A5CCGCCCGQ(;GTTT&&T.&&GAQGTC'88 Z7l [lCON 2 •t h Pr-o HU Ser Arg Tyr PM &lu TIV' Yal lef' Ar'g .,.g ;1y Leu Gly 8lu Pro Arg Tyr Ill Ser Val &ly Tyf" YIJ Aao Aen Lye llu PIN &GTC1'C.ct'GCGCGCCCGCCCCA6 &C CCA CAt TC:S AT6 CS5 TAT nc aAa ACt 8T& TCC C86 C&C ;6C CTC 88& aAG CCC CG6 TAC ATC TCT &TC iGC TAT &TG '-AC AAC AA$ $A8 TTC •t IW7 Yal Al"g PM AeQ Ser Up Ala Glu Un Pro Al"g Tyr 8lu ,...o Arg Ala PrQ Tro &lu Ciln Glu Gh Pro &lu Tyr Tr-o &lu AI"CI Ill Thr lln Ilt All Lyl IUy 8ln &lu ITS CGC TTC IliAC AIC 8AC 8C5 io\5 .UT CC& A4A TAT &Ali ct:5 Ml5 a:& CC8 TQI; ATI; &A& CA& 1M& M5 CC8 " " TAT T81'i &A& C&G .t.TC .t.C5 CA; ATC &CC .US Q8C CAl a.t.Q •u 8ln Trp Phe AI"'Q Yll .Un ~tu AI"'Q Tl'w- !Au LIU fUy Tyr Tyr Mn 8ln ler Ah 81y 8 CA6 TIJG TTC CQA GT8 .UC CTG A$6 ACt CT& CTC 8&C TAC TAC AAC CA.& Alit ac:5 &QC 8 ...... ST&.A6T&.ACCCCSC&Tt:SiA8&TCAC$&CCCCTCCCTTTCCCS•C•C•~ACGCT&.t.CTTCiiTACCC.UGT l y Thr HU Tl'W" LIU Sln Tro •t Tyr CC'Iiir.ASaTTCGGB.UCA6.UCG&ACCC&GAACCAiiTTTCCCTTTCMiTTT~TC'9CSSCC'99V99C&QGQC66GT~T~ACCSCA& 8C ACT CAC ACA CTC c.A& T$G AT& TAC [lCON '7110 J 8ly Cyt .UP Val Sly . Ser Up &ly Arg !Au Leu Arg l h Tyr 8lu lln PM All Tyr Me» &ly Cya MD Tyr Ilt All l.e'u Aan &lu Aao !Au Lyl Tlw' Trp Thr" Prtt All AaP -=~~-=~-===~-~-~=~~~-~~~~~=~-~=~~~~=~~ ~ Ser Ser Gln Ilt Thr Arg Arg Lya Tr-p llu &ln All &ly Ala All llu Tyr Tyr ArQ All Tyr LIN &lu fUy &lu Cya Val 8lu Trp Leu HU AI"'O Tyr Llu LYI Un 81y ATG TCG TCI CA.& ATC: 4CC C&A C&C AA& TQI; aA& CA5 ICT liT ICT .:A aAI TAT TAC Aa& ;cc TAC C:T& '""- 88C &-" TliiiC &T& aA& r&; CTC: CAC A.&A TAC CT& .U& .UC: $lila tOOl •t Aan All Thr' Ltu L.u Arg Thr A .UT atT ACG CTG CTG C$C ACA 6 8T&e•8&QlliCC9""99C999"'.-:TCCTCCCTCTC8CCTaa.cl&l8&86CTCAITCCT9989••S•+&•••c;ccTu.&eT888&T8.t.TiiCCCT8TCTCA;.t.GGGGAGA&AGT&.t.CCCT8 ldO &GTTCCl(M TGCCTCACiCACAGTG.ACTGCACT;.t..CTCT~C.WTTtTCCC~&TCTCAGG •&&G• 4 fi9ABAI ++TTTCCCTSA86T AAC.U.C.A&CTICTCCCTTCASTTCCCCT&T AGCCTCTSTCAI tJ00 CCA TliiGCCTClCCCAG.c6GGTTCTCT&CCCA.CaCCACT&TCiill8TiiiACACT1iACTCCT&TCCT88CTi.+ T&T&TiiTCA&CGCCTT ACACCTCA85ACCGQAAaTCSCCTT ACCT&.t TT&54.UCA T.....CTCCT AT ACACT AliiCCI TSTT t.e &JCCCCABCT"TCT AG.UCTTTCCA&J.GAA TACA TTCTCCCAGA TCCCTCttTSTCTSTG&&;TTT~T AJ.TT"CTCTCT ATTCCT AC.I.GTSGT;.u TSGTCACA TQ.A.;CTCTT AT&GGGT ACSTGG.l&au TAT U. ..0 ATC&.t.aAA T"TTTCTTTTTTST"TTTCTCTCTCTCTT"TCTCTCTCTCTCTCTCTCTtTtTC:TTTTCCQ.A TTTTTTT~TTTCTCTCT AT A&CCCTSTSTCCT;.t.CTCA TTTT A&CCCTTTTCT AA TTCACA TCTTCTCT ... Tl t?a ao Ser llro Lys Alt t_, T&G"lCACT o+;TQC.U TGACAGTGT ASTGTCA.U TA&ACACA TA&TTCACTCTCA TCA TT&ATTT AACTCAGTC"!"!IiT&T A&A TTTC.AaTTT&TCTT"&TT .U TTiiiiTifGAA TTTCTT A.U TCTTCCACACA.Iiii A"T TCC CCA U.G GCA [lCON • HU Val Thr H11 H11 Pro ArCI Ser L..ya Sly Gilu Val Thr Leu A1'11 Cyt Tro Ale Leu Sly Phi T'(f" Pr'O Ala &ap tlt TN" Leu Thr Tr-p &ln Leu Aln Sly Blu &lu !Au 1lV" ~~~~~=-=~~-=~=-=-~=-=-=~-~~=~=~~--~~~~~ Qln AlP •t 8lu Leu Vel Slu TN" Arg Pro All Gily Up fih Thr JIM lln Lya Tro All kr Y1l Yll Yel ,...o Leu SlY LYI &lu 8ln 411'1 Tyr TIV' Cyl Al"l Yel Tyr Hie ~-~~~~~a-=~-~-~=~--~=~~~==--~~-~~==~~~~ &lu Sly Leu Pro &lu Pro !Au Thr Ltu Al'1l Tt"1J 8 8A6 666 CT& CCT laG CCC CTC ACt CTS A$A Till& 8 IT~T&TGi~M;T~T&&;C;T~T~TTCT8CAQACCCT&.Aac:T&5TCAGGG.lT~TGGGSTCAT.UCCCTCACCT"TC •t [lCON 5 •t lu Pro Pro Pro ler 11'w- MD $tf" Tyr Yll Ilt Vll All Val Ltu lh Val !Au &h All All Ilt Ilt Gly All Vel Val All PM Vel ATTTCCTGTACCTSTCCTTCCCAG A6 CCT CCT CCG TCC ACT 5AC Ttl TAC AT8 &TS ATC &iTT &CT an CTS e&T STC CTT S... 8CT ATG &CC ATC ATT G&A &CT fiT& &Tiiii &CT TTT STI •t 1:1!58 D7'l L'tl ArO ArCI Vg .. n Thf' a ATG U.S AQA AWG A.fi.A .U.C ACA 6 liTAA&•+•&IiNN•~TCTIASTTTTCTCTCAQCCTCCTTTAa.U.&T&TCTCT&CTCATTU.TGGGfi.A.ACACAQCCACACCCCACATT&CTACT&TCTCTAACTIMNTCT~T&TCA&TT 1520 £XON e CTGGG.u.RTTCCA&T&T~TCTTCCTTiMCTCTCAC.AaCTTTXCTTTTCACM )y l h LYI l]y lly MQ Tyr- All leu Ala Prto 8 IT ..._ u..+ 18A CMJ8 8AC TAT 8CT CT& 8CT CCA 8 QTTANT&TGGGGACA&GATM&TTCTGGGGGACATTG&AGT&U&TTI _, £lON 1 1y Ser San Ser Ser Slu 6AGATG.ATGGG.AQCTCT&G&AATCCATAATA8CTCCTCC~TCTTCT~TaA&TT8TactAT.._..T8AATII:.ATTCAT8TAC.tTAT&CATATACATTT&TTTT8TTTTACCCTAI &C TCC CAS A6C TCT 5AA ROO ATG TCT CTC CGA QAT T&T UA 8 8T&.ACACTCTA581iTCT8ATT&G&Goi&G'fte 4 ATSTG&ACAT-...TTa&liiTTTCAG&GACTCCCA&AATCTCCTGAGAGTG.\5TGGTG&GTT&CT&G.UT6TTSTCTTCACt.;TG.ATATT a.&2 •t Str Leu Are 4e0 Cya Lya A £lON I lt Tr• CAT&.ACTCTCATTCTCTAS C& TSA A&.l.CA&CTQCCTS&ACTSTACTIAQT6ACAQAC5.AT&;T8TTCAa&TCTC"TCCT&;TQ.\CATCCA6A&CCCTCA&TTCTCTTTACACAZCATTSTCTSATGTTCCCTGT~GCTT6GGTTCA&"TW Maee TGAAG.uC"T&"TGGAGCCCAGCCT&CCCT&C.t.CACCAaw.cctT ATCCCT&CACTIK:CCT8T5TTCCCTTCCA TA8CCAACCTT6CT&CTCCA6TCA.U.CACTGGGG6ACA TCT&CATCCT&T .U&CTCCA T&CT ACCC TGAGCTGCA&CT line CCTCACTTCCACACTGAGAA T.U. TU TTTGAA T;TaGGT&&CT~A TG&CTCAaCeCTilACT&CTCTTCCA..UQfiTCCT_,.&TTCAAA TCCCA5C.UCACA Ta&T;ac:TCA.C.UCCA TCT;T .U TG& TATCT AACACCCTCTTCTS Del CAGTSTCTG.u.&ACAGCTAC.AGTGT 1414 Fig. 2. Nucleotide sequence of gene 27.5 . The amino acid translation of each exon is given above the nucleotide sequence. Certain amino acid residues are encoded by two exons and are so indicated by splits in the three-letter amino acid code. Triple dots indicate gaps in intron sequences. Gaps in introns 1 and 3 are each approximately 200 nucleotides in length. The ambiguous base code is as follows: P, either A or G: R . either A or T; Z, A. T. or C; Y, either Cor T; S. either G or C; X, A, T, or G. 5 FEBRUARY 1982 681 53 hibit 80 percent homology . Ciii ) Two of the introns. I and 3. exhibit extensive regions of nonho mology in these two genes . (iv) The homology in the 3' end of these genes terminate s almost precisely at the end of the 3' untranslated region as defined by comparison with the L d-like eDNA clone pH-211 (10) . Thus the exon s and many of the introns of these class I genes are conserved . However, extensive divergences both in siz-e (fig. I c) and sequence are seen in two introns and beyond the 3' untranslated region of these gene s. To what extent these divergences reflect a lack of selective pressure operating in these regions or the rapid changes that may occur in a pseudogene cannot be determined at thi s time . Two highly repetitive sequence elements homologous to the human Alu sequence (21) have been found in the third intron ofpseudogene 27.1 (4). Since Alu-like repetiti ve sequences appear to be ubiquitous in mammals and are transcribed in man y cells . it has been postulated that the y ma y play some undefined role in gene expression (22) . The Alu-like .·.. ; ... i .,, ~ . . . .... sequences are not present in the third intron of gene 27 .5 . Hence the presence of these Alu-like sequences in intron 3 does not play a role in regulating the expression of gene 27 .5. Funhermore . the Alu-like sequences occur in the third intron of gene 27 . 1 at precisel y the boundaries of a large region of non homology (1000 nucleotides ) in the comparison with the third intron in gene 27.5 (Fig. 3) . Thus it is tempting to spec ulate that the Alu-like sequences , which have the characteristics of transposons 122), may be responsible for the insertion of a large region of foreign DNA in intron 3 of gene 27 . I. This is consistent with the observation that the third intron of 27 . I is approximately 1000 nucleotides larger than the third intron of 27.5 . It will be interesting to determine whether Alulike sequences are present in other class I genes. In summary , we have determined the coding sequence for an L d gene of the mouse major histocompatibility complex. Thus. an H-2 gene corresponding to a serologicall y defined protein product has been identified and characterized . .r;' . .:.> l . ::J, . . . _,_.__ . • ; . '.' ••,' t• • -~ ,.-[. _-'- '. ·- I ··: .• _ . -. ~ ...:.. < . -. ~ ...:1 Fig. 3. Dot matrix ho- :. :· ;~ -.-")~::. : ~ ·- .~: -~-::;,· ~< --~ ~:/~ . : ,..' - ~-- ' ~; · · ~ ::. . ~ ~ · -~1" ......;·. :: •" .:: .. ·. t . ,; :""'. - · :- .. ... •! ·:.-.~: : ~(' . ~~::~~::· . ! .: <( :11f11 II I ll I ji:ill llii; u f 1'( II ) lllil.l ! l ·_liU:UJ Jll ) ':.· ~ - ..- . ~-. ~ -. . _ molog y analysi s of gene 27 .5 and pseudogene 27 . I. The exon and intron structures of the two genes are shown along the sides of the figure . The positions of intron-cxon boundaries are indicated by horizontal lines for 27 .I and vertical lines for 27.5 along the line of homology . lntrons are numbered below the line of homology and exons are numbered above the line of homology. The positions of the sequence gaps in gene 27.5 are indicated by arrows. The positions of the Alu-like repetitive elements in gene 27 . I are indicated by horizontal lines along the line of homology and in intron 3 of pseudogene 27 . I. Cloning and DNA sequence studies sugsest that there may be three lor more) Ld-like gene s. It will be interesting to determine whether these putative L-like genes are codominantly expressed on all cells. or whether these genes are expressed in a tissue-specific manner such as is seen with the Qa and TL antigen s. It appears that, in the future . other D-end functional genes will be defined by the strategies of gene cloning. gene tra nsfer. and sequence anal ysis . KEVIN W. MooRE DNAX Research Institute , Palo Alto, California 94304 BEVERLY TAYLOR SHER Y . HENRY S U N KURT A N DREW EAKLE LEROY Hooo• Division of Biology , California Institute of Technology, Pasadena 91125 Rd'enaces and Notes I. J. Kle in. Scitnce 203 . 5 16 11979 1. 2. G . D. Snell . J. Da usse1. S . N athenson. Histocompatibility !Academ ic Pre ss. New York. 19761 . 3. T . H . Hansen . K . Ozato. M. R . Mel ino . J . E . C oligan . T . J . Kind t. J . J . Ja nd mski. D. H . Sachs. J . Immun ol. 126. 17131 1981 1. 4 . M . Steinmetz . K . W . Moore . J . G . Frel inger. B. T . Sher . F .-W . Shen . E . A . Boyse . L. Hood . Ct ll 25 . 683 !1981 I. 5. R. S. Goodenow . M . M c Millan . A. Om. M . Nicolson . N . David so n. J .'\ . Frelinger. L. Hood. Scitnce 215 . 677 1191121 . 6 . F . Sanger . A. R. Couls on . B. G . Barrell. A. J . H . Smith . B. A. Roe . J. Mol. Bioi. 143. 161 (19801: 7. J . Messing . R . Crea . P . H . Seeburg. N ucltic Acids Rts . 9 . 309 !198 11 . 8. J. A. Lopez de Castro . H . T . Orr . R. J. Robb . T . Kostyk . D. L. Man n . J . L. Strominger. Bicr cht mistn · 18. 5704 119791. 9 . J . E . Cohgan . T . J . Kind l. H . Uehara . J . Martinko. S. G . Nathenson . Naturt ILondonr 291. 35 !198 11 . 10. M . Steinmetz .r a/ .. Cr/124. 125 !198 1I. II. F . Bregegere .rat .. Naturt !London/ 292. 78 (1981 1. 12. B. Lewin . Cr/122. 324 11 9801 . 13 . J. E . Coligan . T . J . Kind!. R . Nairn . S . G . Nathenson . D . H . Sachs . T . H . Hansen. Proc . Nat/. A.cad. Sci. U .S .A. . 77 . 1134 119801. t4 . M. Steinmetz . A . Winoto . K. Minard . L. Hood . Ct /1 . in press . t5 . R. Goode now . unpubl is hed n:su lts . 16 . J . Mic ha e lson . L . Flahert y. E . Vitetta . M. D . Poulik . J. Exp . Mtd 145 . 1066!1977 1. 17. L. Flahe rt y. Tht Rolt of the Major HistocompatibilitY Compln in Immunology !Garland . New York. 19801. 18. T . H . Stanton and L. Hood . lmmunogtnttics II . 309 !1 9801 . 19. M. J . So los ki . J . W . t..:hr. L. F laherty. E . S. Vitetta . J . Exp . Mt d . 153. 1080 I 198 11. 20. T . Hunk a piller and L. Hood . in pre pa rat ion . 21. A. S. Kra vev. D . A. Kra me ro v. K . G . Skryab in . A . P. Ry.skov. A . A . Bae v. G . P. Georgie v . Nucltic Acids Rt s. 8 . 1201 tl9801 . 22 . S . R. Hay nes . T . P. Toome y. L. Leinwand . W. R. Jeline k. Mol. Ct /1 . Bioi. I. 573 11 98 1I. 23 . Su pported b y N IH gra nt G M 06965 . K .W. M. was init iall y suppo rted by a postdoctoral fellowship from the Na tional Insti tute s of Healt h. B.T .S. was supported by a pn:doc toral fello wship fro m the Na uo na l Scie nce Founda tio n. Y .H .S. and K .A .E .were suppo rted by a nat ional n:searc h service award I I T32 GM 076 161 from the National Inst itute of Ge ne ral Med ical Sc iences . We thank M. Douglas for computer graph ics. T . Hu nkapiller for co mputer programs. and B . ursh for ma nu scri pt preparation . We thank W. R. Je linek fo r point ing out the pr e~e nce of several Alu -like sequences in gene 27.1. • To whom correspondence should be addn:ssed . 2 682 3 5 6 7 B UT S No vember 198 1: n: vised 16 December 198 1 SCIENCE . VOL. 21 5 54 The following is the corrected sequence of the H-2Ld gene. As originally published, the sequence was in error due to a cloning artifact. In addition, a number of editorial mistakes were made. The cloning artifact affected the first 90 nucleotides of the sequence. It resulted fran a Bam/Rsa-filled in ends 1 igation, and the sequence at the joint reflects the presence of some Ml3 sequence as well. This artifact also affected the restriction map, as the affected clone was used as a probe for exon I sequences. The cloning artifact and most of the editorial mistakes were corrected by extensive re-reading of the original sequencing gels and by resequencing of the clone indicated in the diagram below. In addition, Henry Sun has checked the sequence of the second intron and third exon by making new Ml3 subclones and sequencing th~ The nucleotides affected by the mistakes and subsequently corrected are underlined 1n the copy of the original sequence diagram that follows the corrected sequence. There were seven mistakes that resulted in codon changes and 32 other changes in addition to the cloning artifact. The restriction map that follows the sequence figures shows the sites that were affected by the cloning artifact. An unrelated mistake lay in the misidentification of the location of the mouse Bl and B2 repetitive sequences(called Alu1 ike sequences in the paper). Improved canputer software has allowed the precise identification of their locations. The Bl sequence in the 27.1 pseudogene occupies nucleotides 2580-2660; 55 the B2 sequence is located at nucleotides 3727-3893, just before the beginning of exon IV. The H-2Ld sequence also contains a Bl sequence which begins at nucleotide 1587 and continues up_to the gap in the large intron. This Bl sequence is in a postion analo- gous to that of the Bl sequence in the third intron of the 27.1 pseudogene. The H-2Ld sequence lacks a B2 sequence, however. The region of nonhomology between the H-2Ld and 27.1 sequences that shows up in the third intron corresponds to the T-rich area that precedes the Bl sequence: the length and sequence of this region is variable in the two sequences. 56 . . , Gly &la . . , Ala ,., lrl ft• .... ...., .... ...., .... Ala lla l1a ,.,., Pro laJ - laJ Pro lrl I :!.ttt.tce~c~~~.~oc~~~~~to~,~Till~oc~~f~~:C~~ic~tt~cm~~~ =:; h ,.. •1o ..... , 1r1 T7r Glu ,.. au 9al 1r1 ,.. Olr Leo 11r 01• ,.. 1r1 T7P ua kl 11r T7r 9a1 ' " L 11 K~~~m=~=~~~~~~=-~-~=-~m~~-~~~~·~ . . Clo - 9al &PI ,._ &aJ - laJ &la 01¥ &ae - &PI T7r lh Pro ala Ua - ,.,.. . . , Olo Ola ala 11J - l h T7P ,.,.. 0 II K~=~=m~~-~~~=~~~-~~=-m-~~~=-~-·~ h &•I lla Glo lla ua L,. 01r ala G1u Ola ,.,.. a,.. 9&1 ,.. "'"" a,.. .... &..u 01r T7r T7r ,.. Olo Ua 11r I 11 "=~~~~=~~~~~~=w~~~-~~=-~~~~~~~~~ MtT ru =~ :!: ~ :",i m~ g; ::~ :;~ ~ t .m lmaCTCACCCCc;(;CTCCG&CGTCAIOUCctTI:taTTTCttC&tACAGGIW:OCTGACC=ucTCtCACCTTC~latac&atiiG&ttCIIGa&ttc=CTTTC&GTTTCIG IOC&GTCCCCCIGCCIIGOCOOCIIGOGeCOOOOCCGnGAOCOOOOC1'Uta;ooGTCCtciC&I ~~ ~~~~==~~=~~~~=~~~~=~~~=m=~m=~=~~ a Ala &op . . , &lo Ala Olo Uo ftr lrl &rl LJO ,.,.. Olo Ola Ua 11J Ua Ala lh T7r 'f7P 1r1 &lo T7P - Olo OlJ llo Cya k IM •~~m~=~m~~-~-~~~~~~~~~~~~~--~~~~ 1 Ch1 T'rt L-e .. 811 arc Tr a..u Lr• &.ea Glr &afl AJ• "''" Leu Lev &PI ftr I tiJ I Clt ~ CTC tiC AC& TIC CTC UC UC GOG UC GCC ace CTC ~ CGC &tl C ~&OOOOCCOC:OOOC&oc=CTCCCTI:TCCCCTCOGOCT'OGGGOC _ , tt &CTC CTCCCG& &C aA"II .lCCC TC &GC'TGC.CCTCI TGC CCTG Tt tC ACiA CiCCG&CIC&CTG&C CCTGVCTCTtCTGI TGC CTC &CCAC ACTG&C T~&CTtt&CtcTCC &Gc;c.en:&CC 1 \0 J TTC TC CC TCC &CAC'rCC.CCC &CCC TG TC TC I CC&CCC a I CGJ C&CU TTTC CCTG I C.CT A&C a &C &GC TCCTC CCT'Tt ICTTtt CCtCT &CCC TCTGTC IC.CC aTGGCCTtTtCC &GGotC 111) 1«. n (TC TCC CCa CCC t aCT C. TC VG1'CAC a CTC&C1C CT C'T CCTC.C.CTClCTCT CTC t.CC ten AC IC CTC &CCIC CGCliCTC CC CTT ICCTG& TTCG.l 1 .t.C I T""IC TtCT &T&Cit 1 Jl) t.tr,t CC TG TTCC CCC& C.C TTC'T &CAlC TTTCC I CICII TAC I TTCTC CC IC &'tt CC TCCCTCTCTCTC~TTTGC &CCCCTTCCC &C II CCT U TTCtc'tCT &TTtCT At ICTCCTttU t .. ) =~~~~~i;iillmii~m~~;g~;~!~~~~~!!ii~ii!~'ieTTriT~~c!iU~~~~i!~~i!i~;~;~~~~:E:i; ~aCTCTCITI:I~TTTI&CTC&GTCTTGT'GTKITTTCAm'TCTCTTGTT&.lTTOTOGU~U&TI:TTCC-K :f t~:: ... Ia llo "'" .,., ... Lra Olr Gh hl ft• .... Ira tro ,.,., &Ja .... Olr - T7r "'" Ala aa, Ut - :t:l m~:~;~=:I~ ,.. I Ill lr Ght Gha W't~ Ttlr Gle Up . . , Glw Leu lal Glt~ ftlr arc ,.o lla GJr u, Glr ftr PM Gh LJI ,.,., Ua . .,. tal Wal hl ,... &. 1St n~=~= a.u ,.,. ,.,.. lla - ~~=~=-=~~~~=~=~~m~=~-~~~~~ &~~~~~~~~~~-~~=~~~-~=~~-~~~~~~c~ = eu Gh Lr• Glt~ Cl• &all Tyr !lar Cra &rl Wal Trr IU Glt~ Glr W• ,... Glt~ "'• &.eu ftr Leu Arl !P' 0 ns TT GGC &.lC CAC C&C au T&t &tl CGT CTG TIC CAT ~ OOG ~ CCT ~ CTC ~ ~ &G& G ~&.lOG&IOCTOTOCIOTCIC&O&C toto = .............. 1!Tt ~·~UIOCTOG&GCCTTCTOCIC&tCCTC&ICTOCTCIGOG&TGKIOCTIIGOGTC&T&&CCCTC&CCT1UTTTI:CTGTAUIGICCII~O K CCT ~ CCC 1195 . . ,. TV Up . .,. fr'r .. , lal Ua lal Ala Wel Le• Glr lal Lew Glr &la . ., &la 11• Ue Olr &la tal Wal lla P'IM tal . . , LJI JOI ~~~=~m~m~~~=~~~~~m-m~~~~~~~~m~~ ........ . . . . . . ,.. 0 ... AGI ace &GI &.&C &to G ~&.lCUIIIOO...GGCT'CTIO&GTTTTCTC&GCtTCnTT&O&AOTGTCTCTQCTCITT&&'IGOGOUC&taoet&C&CCCC&C&~&CTOTCT OH lf llr Lra GIJ Clr . ., = T7r Ala .... Ala r CTU~TI:TOCTCTCIGTTCTilllGUIITTCt&GTOTCI&foiTCTTCCTI'C~&C&C ~ OGI U& DOl ~ ~ fAT ItT 111 ICT C - ~ ~ ~UTCTGGOGICIIIGa~ITTDCl11T'GUCTT110&CITC&~.lOCTCTOOGI.ITCCITUT&OCtcc'TI:e&G&G.U&~IGOGGCC~ am 1J Ia• CIA k P kP Olv . . , ....... lrC laJ c,. L,. & Itt C.ttGUCTtUT&CITTCI'fCT.ICAT&TG-CITITACa'tTTCTTTTCTTTTICCCTIC GC fCC tiC. &U TC'T G&A lTC tt'T tTt CCI G.l'T TCT AU. G OTO&C&CTC lf)1 TIG~CTCT~ATTOGC<OlCC=uTGTCCICIT~ITTINGTT'ItiGGCICTCCCIGUTCTCCTCIGICTClCTGCTCCGTTc;(;~UTCTTCT,TTtaC&CTCITGG=ITGAC~ la TN ll'll J)t C& TTCTCT IC CCT GI &C&CI GC TCC C'TCCICTCT ICTCICTCIC IGICCI TCTCm &GCTI:TCTCCTCTCIC &TCC ICICCCCTI:&emTCTTTICIC I !Cl TTCTCTOI TOTTI: IHI tC TWTCiA CC nc.c.c; T'TC I CTC TCI&C.U.C 'TCTCGa CC CC &C.CC'CC CCTCC I C&C C&CC&C CC'T a tt CC TU IC TC.C CCTC. T CTTCCC TTCC A'f &GCC UCC TTCC T(;CTCC&Cte&llC SOli ACTC.CGCC&C &'TC TC.t ATCCTCT A.l C. CTCCITCCT IC CCTGACCTG< aCC'TCCTCAtncc at ACTG&CUTUTUTTTCU TCTCt<i1(iCCTCC&C&CITCrCCTC&GCYCTG&.tT'GC't )10& ITTCC&a.tC.CTCCTC.&CTTC a U'tC CC AGCUC it& TVCTC«Tt &C U.C Cl Tt tGT && '""'I tcT .I.&C atttTtTTCT~ IGTCT'CTCUCACAGCT A.t&GTGT 111 t 57 ~TTUI:tll'l•t••rt"•rte•arTtTMIC~~T~~TC.tC'T'CCCA&i~ All All Ale fi""' ..... 0 t.ao l r No ~o .,.. I Ia &t.C TC.l &t.C C. 1 ~T,f=8a'-""'i:f'C.C:~'8'._,..,..,,MC8f. ... ea: ea:- = cca r •t ,.,.. -· -· - A l e .... . . , - 1AU 1AU IAU ._.,._., &R IC'1 ca C8C IIC8 c:ta CTC CTI CTI CTa I&T=.,.-:crcr.. 9 -~T~ h ,._o Mil lefTYf",.,. llu nr~~~ lee" are .,... lly Leu IJy llu..,.. M"t Tyt" Ill Sr V.l l h Tyr Val Aeo Aen L.,. llu,.... C:U CAt Tl:S AT& C. TAT nt AIX 815 TCC Qjf CW' 88C CTt = C. TAC ATC TAT &t.C nt ~ K .. 1 .., - TCT lTC 88C 8TI &AC .... - """ ..,. - Alo llw ....., .., TY" llw ....., .., Alo .,.. T,.. • • llu 11• llu lh .,.. llu TYf' '"" llu .., Uo II• Uo Uo L" l h 11• llu 8Ticrnta.c-a.c~a-~cca-~-=~~acca-~-~--cca-~---~~~~ea:-.:~- •t" ,,.. filM 11r't ftl Mn Le\1 Art 'nr LMtl Leu l h Tyr Tyr .fieri lln ler Ala lly I ~- TTC- Ill &AC CTI- AIX CT& CTt 88C TAC TAC &AC CloiO- ICa 88C I Ill IT&.Oa~TCAOilf'"'I!Xi'III"*AC'C'-'CM'fiACt'ca-T~ ty TPr Mll TIW' LA\! 11ft T,.. •t TYf' CCII-I'TTlC8811oU.c.&I...._ACCX:811AAC:C,ITTTCC:ct1rtc<loiTTTI...._ITI1j~~-8CC:8-&8C:9NIC::e--ca--IIC::a-IIIIT~~ .:; CAt ACA CTt CloiO AT& TAC -· - ACT - lly Cyo Yol lh ..,. lly ""t ...., IAU . . , lh Tyr llu II• A1o YYf' """ lly c,. """ TY" Uo A1o IAU llu """ IAU L" T,.. Uo -~-~-=--===cr-~-~=ea:~a.c-~~~~ea:~&AC-a.c~~--~=~a&t.C- •t ..,. ler lln Ua l1'w' llrl Art Lye ,,.. llw 11ft 111 lly All &la llu ,,.. Trr .,... Ala '"' Leu llw lh llu C,.e Val llu Tf'O L,.-, Hla ,.,.. Tf1" Leu Ln Aan lh ~~P-~--cr---~~-~~-~~-ea:~~----8ll--crtc..c-~~---Alo- LeuiAU " " t - A AAT K'T ACS CTG CTS CIC ACA I IT.K::J''Ii"ti$15t:l'.ICIN180--C...... IFT11a::cT11CCCca::1"1CC'iltCM'CXI"'IrtW-ICTI-~TCA&T~TICCCTITCTCt9'1f?$G•s;•etSTs.a.c:ccT8 tiiO ~CCTIAT8CCT~TAACTICACTIACT~TIIACA8~~TAACAAC.IoiCTICT=TTCiolitt=T&TA8CCTCTI~ ..0 C:UT-=et~TTCTCT~TIACACTIACtccnTCC~TITUCA&CSCtTT~TCM'CTTAIXTIATTIIoUACAT&&t.CTCCTATACACT~TITT t- tc=AICTTCTAI&AC~TACATTC=-T=~TI-TTTICACaXTTC8CAAAntTCTCTATTa:TACAITIIIITIAATIITCACA~TCTTAT-T~-TATAA - ATc::aAaoU I I I i C I I I I I IITTTTCTCTCTCTCTTTC ICIt I C IClCTC IC IC ICI C ICTTn'CC:aA ~TCTCT AT Altttl'ITSTCCTIACTCATTTT IIKCtT'TT1"CT1t.A T'T'CACATCTTCTCT • • . n t7a -· .. - L " Alo TWTCACTAIT.:AAT&ACAITITA&~TA&t.CACATAI~TCATTIATTTAACT'" .... TCTTVTITAIATTTCioiTTTITCTTITTAATTI,_TTTCTT~'TCTTCCACACA& AT TCC C:U AA5 9C.l ltto ... - ""t .... L" lh llu . . 1 - HU Hlo - 1AU . . , C,. r-. Alo 1AU l h - lY" ...... A1o - Uo - 1AU - T... II• ~ - - IIY 11u 11u 1AU - ~~-~CAt=-=~~-~-=---~~-=~=~-~-~---~~---~-~ Ill'\ Me •t ll1o1 L.e\1 Val lllol nw- ~ flir'o All l h Mp Illy ""' "" lln l,_ fr1t All ..,. Val Yll ¥11 ...... L8l.l lh L" llw llPI ._, Tyr ""' Cye M"t Yal Tyr M11 ~-~-~~-~-=9C.l-~-ACCnt~--~=~~~=~---~~~--~~~~~ llu l h Leu ....., llu ....., LOW Tlr Leu ""t r,.. I CT& CCT 5&5 = CTt ~ CTI AlA I IT.lAIIIIA555~.-T~Tii8AIICCTTCT9CATIA«TIIII~T~-TCAT.u.ca:tCACCTTC - -· -· -t - lw ,._0 _,.0 (lllra 1M' .,.. Mo lef" T'lf" W.l Jla ¥11 All Val Leu lly Val LMI lly All Mt All Ill Ill lh All Yll Yll All ""- Val ATTTCCTI'ACCTITCCTTCCCA& A& CCT CCT TCC &t.C TAC ATa IT& ATC ITT KT ITT CT& .T lTC ~ KT AT& Itt ATC ATT IliA IICT IT& IICT TTT cca ACT TCT Mt LYI 111"1 ~~ M"l Mn Ttw' I ATI AA5 W ~ A&A AAC AU ' I5A Ill IT•aa••·-.e-T'CT'aA.I~M4AaTITCTCTICTCATTAATPGPS*•t•r.rr•r+fCCCtr•TTICT&CTITCTCTWTe;,;TtTICTSTCAan 810 1Y l h L" lh lh TY" Alo 1AU Alo I AAA &t.C TAT KT CTI ~ C:U I ITTAOITIT----TTCT~ATTii5AST&Aio5TTI ~ CT855AAIOTTCCA8TITUAIATCTTCCTTI&ACTCT~~TTTTCACAI IT - IA8ATIAT-.atTCT191AATCUTAAT-TCCTCCMAIAAATCTTCT~TTITKCATIAAIIT-TACATTC.lTITACATAT11CATATAC.lTTTITTTTITTTTAIXCT&a lOOt .... Leu Ill D71i -· 1Y ........... - ••• 9C TCC CAl A8C TCT - - " " t - C,. L " A -· AT& TC'T CTt CSA . . T TIT W I ITaACACTCTA&eaTC'Ta.tn$85'•8G188C••tcT~TSATTCIKTTTCAaGQ.t.CTc:ccA5AATCTCCT(;A.8.A;TWT;;TQ81TTICTI&U.TITTITCTTCACA&TiolT.TT ..., le TN CATAACTCTC.lTTCTCTAI Cl TIA AIACAQCT~TIIIIACTITACT. . . TIACAIACIAT8TITTCAIIIITCTCTCCTIT-&TCC....e=~TTCTCTTTACACAZCATTITCTIATitt=TITIAIICTT881~TI - T~n~~ASCctAIICCTICCCTiiCAC.ACC&Cii$ACCCT ATCttTIICACTIICCCT&:T&TTCCCTTCCA T..::CUCCTTQCTICTCCA&TCMACACTIG&G&rACA TCTICJ. T'CCTIT .uatTCCA fiCT ACCCT~TIC.t.5CT .. •·---•••tt•,.•_.Tal"&&T&T .... 1--4 8 8 )( )l)t HSD 'i IP I • 4----1 D a ~. ---I I • - - b 17~t: c 17.1 1 • ox • 2070) 4 • •' =I • I 4 a 7 a 7 58 Appendix C Identification of Class I Genes. This article was published in Nature. 59 Identification of the class I genes of the mouse major histocompatibility complex by DNA-mediated gene transfer RobertS. Goodenow·, Minnie McMiUan+, Margery Nicolson*, Beverly Taylor Sher·, Kurt Eakle·, Norman Davidson§ & Leroy Hood· • Division of Biology, California Inst it ute of Technology, Pasadena. California 91125 . USA ' Department of Microbiology. USC Medical School . Los Angeles . California 90033. USA ~ A\1Gen. Newburv Park . California 91320. USA ~Division of Chemistry and Chemical Engineering, Caiifornia Institute of Technology, Pasadena . Califorma 91125, USA DNA-medi~ted gene transfer was used to identifr cloned class I genes from the major histocompatibility complex of the BALB/ c mouse. Three genes encoding the transplantation antigens H-2 Kd, Dd and L d were identified as well as genes encoding the Qa-2.3 and two TL differentiation antigens. As many as 10 putative not·el class I genes were detected by the association of their gene products with f3rmicroglobulin. Alloantiserum prepared to one of the norel antigens was used to demonstrate the expression of the previously undetected antigen on spleen cells of various inbred, congeneic, and recombinant congeneic strains of mice. THE major histocompatibility complex of the mouse encodes a family of cell-surface antigens known as the class I molecules . These gene products, which fall into two categories, are encoded by distinct regions of the major histocompatibility complex. The H-2 region encodes the transplantation antigens which were initially defined by allograft rejection ' '. These antigens are expressed on most somatic cells of the mouse and are involved in the T-cell recognition of cells altered by viral or ne oplast ic transformation I for a review see refs 3 , 41 . The second category of class I antigens are the haematopoietic differentiation antigens encoded by genes in the! Tla region . The Tla region encodes three distinct types of differentiation antigens. Qa-1. Oa·2 .3 and TL ' . The TL antigens are expressed only on certain subpopulations of thymocytes and their neoplastic counterparts in certain T-cellleukaemias' . The Qa-1 antigen is expressed by thym ocy tes , peripheral T and B cells, as well as lymphoblasts'. The Qa-2,3 antigens are found on a subset of bone marrow-derived cells". The biological functions of the TL or Qa antigens are not known . Class I antigens ha' e been characterized using specific aliaantisera and monoclonal antibodies prepared by the appropriate! cross-immunizations with cells from inbred, congeneic and recombinant strains of mice . Serological analysis has demonstrated that the transplantation antigens are extremel y polymorph ic and that different strains of mice have distinct combinations or haplot ypes of class I alleles. Particular class I molecules are denoted b! appropriate letters with a superscript small letter for the hapl o type . For example. the serologically defined trans- 11 :- -c.. 'J I' plantation antigens of the BALB / c mouse which is of the H-:!" haplotype are denoted K" , D". L d or Rd Class I molecules of both categories appear to share a common structure . The class I polypeptide is a 45.000-molecular weight integral membrane prote in which is noncovalently associated with a 1:!,000-molecular weight polypeptide . (3 , microglobulin . Class I molecules consist of three external domains. each of -90 amino acids. a transmembrane region of -40 residues , and one cytoplasmic domain of 30 residues•·"'. Class I cON A clones have been characterized and used as probes to obtain genomic clones containingclass I genes''-". Two genes isolated from a BALB / c Crgl sperm DNA library constructed with the lambda bacteriophage lA 1 vector have been extensively characterized . The class I gene from clone A27. 1 is a pseudogene which has been mapped to the Qa-2.3 subregion of the Tla region by restriction enzyme site polymorphisms" . DI'A sequence analysis of gene 27 .1 shows that this class I gene is divided into eight distinct exons encoding each of the individual domains of a class I molecule. Recently, DNA-med iated gene transfer and radioimmunoassays were used to demonstrate that a gene from clone A:!7 .5 encode~ the H-2L d transplantation molecule'". The Ld molecules produced by mouse L cells transformed with A27.5 o:--;A were shown to be virtually indistinguishable from the L d molecules expressed on spleen cells from BALB / c mice by biochemical , immunological and functional criteria'"· ". The D~A sequence of gene 27.5 is identical to that of the Ld antigen at all 77 positions that can be compared " . Other laboratories have also used gene Fie. 1 Radioimmunoassay of A1.3, A:! I.! and A26 .1 transfor mants . Transformants which reacted in the panel analysis •Table II with antibodies 34-1-2 to Kd molecules were tested by the cell binding radioimmunoassay with : '), 34-1-2 antibodies to Kd: ::::;, 202 8-4 antibodies to Kb • which react with Kd molecules : and 0 , 34-2-12 antibodies to D" . Transformants derived from transfection with DNA from clones A26.1 ta 1, A21.1 tb) and Al.3 1c ). Spleen and Ltkcontrols are shown in Fig. 4 . Ill .f. < ;:; () Ill IU' Ill 60" Till* 1 Summary of tne codin& assi&nment5 for the clau I tenomic clones Reaction of_transformants with antibodies specific for Ld Qa-2,3 TL' Dd Clone Assignment Al.3 + Kd A26.1 A2!.1 cl7 .1 + Kd cl8.1 Kd Kd + + Dd + A27 .5 c59 .2 c50.2 Al7 .3 Ld + + + + c6.3 + A24.8 66.1 + + Ld Qa-2,3 TL TL TL TL Mouse L tk- cells were co-transformed with -I 1-<S of DNA from each of the A and cosmid clones !see textl and the HSV tlr. gene in pBR322 lptk51 as previously described's. For those cosmid clones containing more than one gene, the clone DNA was digested with an appropriate restriction enzyme to inactivate all but a single gene for transformation !see ref. 24). Uncloned tk. transformants selected in hypoxanthine / aminopterin / thymidine medium 3s were panel analysed by radioimmunoassa y as previously described •s using monoclonal hybridoma antibodies 34-1-2 to Kd !ref. 39), 34-2-12 toDd !ref. 39), 30-5-7 toLd (ref. 40), D3 .262 to Qa-2 ,3 (ref. 41) and TLmli J to TLB . Positives are indicated for those transformants which yielded at least 5,000 c.p.m. of iodinated 1mJ) protein A bound lsee Figs 2, 4\ with a 10 " 2 or 10- 3 dilution of the ascites fluid in duplicate transformation experiments. Background for the various antibodies and transformants ranged over 100 to 300 c.p.m., except for the lgM Qa antibodies which utilized a facilitating reagent (see Fig. 4l. The active genes on cosmid clone 17.1 , 18.1, 59.2, 50.2, 6.3 and 66 .2 are the first gene on cluster II, the first gene on cluster 13, the second gene on cluster 2, the second gene on cluster 6, the second gene on cluster 4, and the first gene on cluster 5, respectively (see Fig. 5). transfer to identify the cloned class I genes from libraries constructed from tissues or tumour cell lines from other substrains 17 and strains' 3 of mice. In an attempt to assess the total number of class I genes contained in the BALB/c mouse , genomic clones containing class I genes have been isolated from a cosmid library constructed from BALB / c sperm DNA"". Fifty-four cosmid clones containing 36 unique class I genes have been assigned to 13 distinct groups or clusters encompassing 837 kilobases of DNA. These include most of the class I genes in the BALB/c mouse. Here, we have used gene transfer and serological analyses to e\amine individual class I genomic clones from the A and cosmid libraries. Cloned class I genes encoding the serologically defined H-2 Kd, Dd, Ld, Qa-2,3 and TL antigens have been identified. Evidence is also presented for as many as 10 additional novel class I genes encoding products not previously defined serologically. Analysis of 96 A and 36 cosmid class I clones by gene transfer In an attempt to examine most class I genes from the BALB / c mouse, 96 genomic clones isolated from the A library and 36 genomic clones previously isolated from the cosmid library (designated c) were analysed. DNA from each of these genomic clones, together with the herpes viral thymidine kinase gene, was used to co-transform thymidine kinase-negative itk -) mouse L cells as previously described's. The class I H-2d products of the transformed L cells were detected by radioimmunoassay and analysed by two-dimensional gel electrophoresis. Since mouse L cells are fibroblasts derived from C3H mice of the H-2' haplotype, the endogenous K" and D' transplantation antigens can readily be distinguished from any of the BALB / c class I molecules of the H-2d haplotype by appropriate monoclonal antibodies. The mouse L-cell fibroblasts do not express the differentiation antigens Qa-1, Qa-2,3 Basic a b -Acidic e c . •·. ~ - d e - Fl&. 2 Fluorographs of two-dimensional gels of H-2Kd molecules immunoprecipitated from cell lysates of BALB / cJ spleen lymphocytes Ia 1. tk- cells transformed only with the tk gene 1b1. Al.3iCI. A26 .1 (dl and A2l.liel transformants . Cells were biosynthetically radiolabelled , immunoprecipitated with 20-8-4 monoclonal antibodies !supernatant fluids I, and analysed by two-dimensional gel electrophoreSIS as previously described'" . The gels were prepared for ftuorography, dried and exposed for -7 days. 61 or TL. These results are summarized in Table 1. In almost all cases. the binding of specific antibodies to the transformants was comparable to that of the binding of the antibodies to the corresponding antigens and not cross-reacting antigens !see Fig. 41 . Identification of genes encoding the H-2 K, D and L transplantation antigens The L-cell transformants derived from transfections with the A1.3. A26 . 1 and A21 .1 clones express Kd molecules which are virtually indistinguishable in their patterns of serological reactivity with two different monoclonal antibodies to the Kd molecule tFig. I I, both of which react as strongly with these transformants as they do with spleen cells tsee Fig. 41. No reactivity is obsen·<!d with these four transformants using monoclonal antibodies specific for D" or L" antigens tsee Fig . 1 legend I. Antibodies sp<!cific for the Kd molecule do not react with mouse L cells transformed only with the thymidine kinase gene 'see Fig. 41 . These are designated tk- transformants. To determine whether these three class I genes encode distinct gene products, the antigens were isolated by immunoprecipitation and analysed by two-dimensional gel electrophoresis tFig . :! I. As the class I antigens of different haplotypes each have distinct constellations of spots", we believe this approach is very discriminating. The products of class I genes 1.3. 26.1 and 21 . 1 appear ext remely similar to one another and to the Kd molecules isolated from normal BALB / c spleen lymphocytes . Howner. minor electrophoretic differences were observed in the faint. lower molecular weight species !Fig. 2a and 2e ). As the electrophoretic patterns between products obtained from duplicate transformations appear identical, the variation in the lower molecular weight products may reflect differences in post-translational modifications of these molecules or differential alternati,·e splicing, as has been previously suggested for class I genes 1 '. Restrict ion map analyses of clones A 1.3 , A26 .1 and A21.1 tFig. 3 1. however , suggest that the Kd genes they contain are identical to one another and also to the first gene on cluster II or clone c17.1 tsee Fig. 5! which contains a Kd gene by transformation tTable I, Fig. 4a ). Thus, all of th<!se clones appear to encode very similar or identical genes on an independently derived series of what we believe are overlapping clones. although we cannot exclude the possibility that certain of these genes may actually be distinct and yet appear identical by the limited restriction mapping we have carried out to date. These genes are currently being sequenced in order to resolve this issue . Cosmid clone 18.1 contains an H-2Dd gene as demonstrated by the radioimmunoassay of the ciS.! transformants with monoclonal antibodies to theDd antigen (Table 1, Fig. 4b 1. The 18. 1 transformants reacted with antibodies specific for Dd products and not reagents which detect Kd. 1-•. TL or Qa antigens. The Dd molecules expressed by the transformants are currently being compared with the molecules isolated from spleen cells by two-dimensional gels and tryptic peptide map analysis to confirm this identification. :-;o lambda clones containing an H-2 Dd gene were identified 1see below) . An H-2L • gene has been identified on cosmid clone 59.2 by transformation tTable I J and comparison of restriction maps between the c59.2 and A27.5 genes". The Ld molecules expressed on the corresponding transformants appear indistinguishable in radioimmunoassays from the products of the previousl y identified gene contained in clone A27.5 IFig. 4c ). The analyses based on 132 independent transformations resulted in the identification of at least one H-2Kd. a single H-2Dd and one H-2L"-gene. Serological evidence suggests that in some mice there may be multiple K and D gene products 2 <>- 31 , and additional transplantation antigens such as Rd (ref. 32J. The failure to find multiple class I Kd and Dd genes may be explained in several ways. First, the serological multiplicity of K or D antigens may arise from post-translational modifications 5"Prt't": • •3 ).26 I •z ; I " 8 ~ oM 2m 3' p, ...... .. .... P!""'2t;: o K8 "."./' 8 H K K 8 R 6 < K 8 R .. .....-... . · ,/\/\.· . ~..r;AI".r./V".tVVV'/\1 .-/V'."/.'.V/.".I'V' F'&&. J Restriction maps of the lambda clones containing the Kd genes. ON A from each of the A clones which yielded Kd -positive transformants was restriction-mapped with the indicated enzymes 24 by single and double digestion as previously described . The 5' probe pH-2111 and the 3' probe pH-2IIa 11 were used to detect fragments containing coding sequences homologous to the class I genes !dashed lines ·,. The pH -21Il probe contains the sequences encoding amino acids 63-160 of a transplantation antigen and pH-211a amino acids 167-352. The coding sequences were orientated using sequencing subclones as previously described" . The solid line indicates the insert DNA. and the A vector DNA is designated by the wavy lines . R, EcoRl; B. BamHl; H . H1ndlll ; K. Kpnl. The 3' EcoRI sites mark the linkers used in constructing the library. or alternative patterns of RNA splicing. Second. different inbred strains may have different numbers of K and D genes. Third. multiple class I genes might appear identical by restric· tion mapping and still differ in their DNA sequences. In this case, multiple distinct genes might be confused with a single gene present on overlapping clones. Finally, these genes may not have been detected for various technical reasons, for example , failure to clone the gene or express the gene at levels detectable in the radioimmunoassay. For example , the radioimmunoassay is less effective with heterogeneous alloantisera and specific monoclonal reagents for the R• or the Qa-1 gene products are lacking. Fibroblast expression of gene encoding the mouse haematopoietic differentiation antigen Qa-2,3 Transformants from c50.2 DNA seem to express low levels of the Qa-2,3 antigen as detected by radioimmunoassay with monoclonal antibodies specific for these products !Fig. 4d ). The BALB / c substrain from which the library was derived is Qa2,3-pos.itive il. Flaherty, personal communication). It is interesting that Qa-2,3, a gene normally expressed only in haematopoietic cells, appears to be expressed in transformed fibroblasts . The fact that the Qa-2,3 differentiation antigen is expressed at low levels on haematopoietic cells" may explain why limited numbers of Qa-2 ,3 molecules are expressed in the transformed mouse L cells. It is difficult to ascertain the basis for the low level of expression of this antigen by the transformants. Several possibilities are now being investigated. For example. only rare transformants may have incorporated sufficient copies of the gene to ove.rride differentiation signals. In addition, glycosylation of the polypeptide in fibroblasts might, in this particular instance, alter the antigenicity of the molecule causing a less avid reaction with the antibodies. The fact that the c50 .2 transformants exhibit elevated levels of 13 2 -microglobulin !see below) does suggest that significant amounts of the antigen may be expressed in an altered form associated with 13 2 -microglobulin. Accordingly, it has been difficult to obtain sufficient Qa-2,3 antigen from spleen cells or transformed L cells for two-dimensional gel analysis. The identification of a Qa-2,3 gene should readily permit its characterization at the DNA level. Two different class I genes encode two distinct TL T -cell differentiation antigens The TL antigens are expressed only on thymocytes at a certain 62 fla. 4 Radioimmunoassay of the transformants expressing Kd. Dd. Ld. Qa-2.3 and TL antigens. a, Transformants derived from trans· 10 fection of L tk - cells with l>N A from c17 .I 10 I were tested by radioimmunoassay for the c.. expression of Kd molecules using 34-1-2 antibodies to Kd. b, Trans· 0 formants from transfection with DNA from c18.1 101 and shown d to express Dd molecules as detec15 ted in the panel analysis were anaJy~d using 34-2-12 antibodies to D . c. Transformants derived from transfection with DNA from A27.5 10 1, which contains the Ld 22 gene '"· , and transformants from DNA from c59 .2 1e 1 were tested 0 against 30-5-7 antibodies toLd d, Transformants from transfection with c50 .2 DNA 10 1 were tested Dliulwn of "n11bod1es in a cell-binding radioimmunoassay using D3.262 monoclonal hybridoma JgM antibodies to Qa-2. Rabbit antiserum to mouse Ig\1 •Bioneticsl 12 was used as a second step reagent before incubation of the cells with '1-protein A. e, Transformants derived from transfecti.o n with DSA from Al7.3 were tested against antibodies to TL 1e 1 TLm1i 1 and I J I Tlmliii) lsee textl . f, Transformants from transfection with A24 .8 DNA were tested against the same antibodies to TL indicated in e. BALB / c spleen cells, panels a-d 1::::!1 or thymocytes. panels e-f 1=1 and Ltk ' cells. panels a-f lwl were run as controls. Thymocytes and Ltk " cells in panel e were tested against Tlmlil, panel f agamst Tlm1iii1. -- stage of differentiation . In some inbred strains of mice, distinct TL molecules are expressed on certain leukaemic cells ofT -cell origin•. Two monoclonal antibodies can distinguish these TL antigens. Antibodies denoted TLm ! i) recognize TL determinants present on both normal and leukaemic cells, whereas antibodies designated TLm1iiil react only with the TL determinants on normal thymocytes". Figure 4e, f shows that transformants from A clone 17.3 reacted with both monoclonal antibodies at levels comparable to those observed with BALB / c thymocytes 1Fig. 4e, [1, whereas transformants from A clone 24.8 reacted only with monoclonal antibodies TLm1i ). These observations suggest that there are at least two distinct genes, A17.3 and A24.8, which encode two TL antigens. From the serological analyses described above , gene A17.3 may be expressed on normal BALB / c thymocytes and leukaemic cells whereas gene A24 .8 would be expressed only on leukaemic cells. The A17.3 and A24.8 gene products are currently being characterized and compared with the molecules present on normal and leukaemic thymocytes by two-dimensional gels to clarify this. point further. The cosmid clones containing the corresponding TL genes have been identified by transformation IT able 1, Fig. 51. The 17.3 TL gene appears to be the first gene on cosmid cluster 5 or clone 6.3, and the 24.8 gene the second gene on cosmid cluster 4 or clone 66.1. Since each n gene maps to a separate cosmid cluster, they must be distinct genes. These results illustrate two important points regarding gene transfer into mouse L cells. First. as was noted for the Qa-2.3 gene, the TL genes which are de velopmentall y regulated and expressed only on thymocytes can be expressed in fibroblasts as a result of transformation. Second. genes not expressed in the mouse except on leukaemias, such as gene A24 .8. can be expressed by transformed fibroblasts . It seems unlikely that the activation of endogenous L-cell genes encoding simila r TL molecules would account for these results as restriction diges- Fig. S Radioimmunoassay of cosmid transformants for 13 2 -microglobulin . The mixed populat ion of tk • trans- -. -•o-o•H......... : .._.... formants der ived from ::: • • •• transfection with a E single gene from each of the various cosmid C::. clones, obtained by 0 - - - _____,.., - -- - - - ... . . - - -- - - · - - -r r - - - - ·-· restriction digestion , i) L Q ') , . 2. 3 was tested for the expression of /3 2 -micro._, -· 2 globulin in a cell· < binding radioimmunoassay using rabbit anti · serum to rat /3 2 -microglobulin . Each trans.t formant was assayed in .. 2 I I duplicate against a 10 8 · ----- ! ___..._. ........... _ 1 - - - - - ~--+- ( - - - .: ----dilution of the anti,., serum. The numbers 125 shown at the left indicate the c. p.m. of 1-protein A bound per 50,000 cells. Less than 10% variation in the number of c. p.m . was observed between duplicate transformations or assays. The dark portion of each bar represents the basal level of L-cell /3 2 -microglobulin measured for tk • transformants which is the average of duplicate samples tested I right I. The open portion of each bar represents the total number of c.p.m. of 1 "!-protein A bound for each transformant which exceeds the level on tk • cells. Cosmid clusters are represente~ schematically with the genes indicated as the enclosed boxes. Each cluster is comprised of a group of noncontiguous overlapping clones · •. The cosmid cluster number is indicated to the left o f the horizontal black line and the genes encoding the serologically defined products are indicated below the appropriate boxes. a, 13 2 -microglobulin on the surface of transformants derived from transfection with DNA from cosmid clones linked with mapped genes or genes encoding serologically defined antigens ; b, undefined clusters only . I I II II • •• •• II II I . _____ III II ., 63 7 "'Q 6 X E 0. ~ Stro •n BALB t cJ BtO D2 8<0 8 10 A C3H t ~e J s C3H . OL Or1•:g',~~~~n~nd .IS. .ll. d d d d ~ ~ ~ c c b •• •• a d ~ ~ d "0 c ::;) 0 D <l c ii) 0 0. :r ...... 2- "''::' Odu •ton of C3H ont o-KB-1 Fill:- 6 Radioimmunoassa" of spleen cells with C3H antibodies to KB-1 transformants . Four C3H mice were immunized by injecting 5 x I 0' UV -irradiated KB-1 cells intraperitoneally every 2 weeks for 8 weeks . Mice were bled 7 days after the last injection and the antiserum used in the cell-binding radioimmunoassay against spleen cells as previously described 18 • Reaction of spleen cells from : e. BALB / cJ : .:::;. 810.02 : • . C3H / HeJ : ::. C3H .OL; •- BIO : 0. BIO .A . The origins of the H-2 and Tla regions for each of the mice are indicated in the panel. tion of Al7 .3 or A24 .8 Dl':A destroys the ability of these genes to transform L cells to the TL-positive phenotype. Ten class I genes encode serologically undefined gene products Of the 36 distinct class I clones, six contain genes encoding serologically defined gene products !Fig. Sa 1. The remaining class I genes are either pseudogenes or encode as yet unidentified cell-surface products. For example, they could encode nonpolymorphic molecules expressed on the surface of most cells which therefore would not have been detected by classical serological approaches. Alternatively, they might be differentiation antigens expressed at defined stages on tissues not yet tested for their ability to generate alloantisera. Cytoplasmic or secreted forms of class I -molecules might represent another category of uncharacterized products. We have used a simple but effective approach to identify class I gene products for which there are no specific serological reagents. All class I molecules studied associate with /3 2 -microglobulin before expression on the cell surface ' 0 '. Therefore, the expression of foreign class I molecules on the surface of the mouse L cell should result in the elevation of the amount of 1' 2 -microglobulin expressed on the cell surface. To test this . hypothesis, a radioimmunoassay for cell-surface /3 2 -microglobulin was developed and used to quantitate the amount of /3 2 -microglobulin on several transformants expressing foreign serologically defined class I molecules as well as several which failed to react with any of the class !-specific antibodies. The tk" transformants express a basal level of /3 2 -microglobulin as do the mixed population of tk " transformants derived from transfection with the L d gene cleaved into small coding fragments with BamHI enzyme . In contrast, mouse L cells transformed with the Ld, Dd, Kd and TL genes all expressed levels of cell-surface 13 2 -microglobulin well above the background level as measured by radioimmunoassay !Fig. Sa I. This increase is reproducible in that the measured levels of 13 2 -microglobulin obtained for duplicate assays or transformations vary by less than 10%. In addition, the elevation in 13 2 -microglobulin expression appears to be a stable phenotype, observed after SO passages of some of the cell lines in culture. If elevated 13 2 microglobulin expression results from the expression of additional class I molecules by the transformants , this phenotype would be stably expressed because it has been demonstrated previously that the expression of the transferred genes is a stable phenomenon ' 8 • The ability to measure an elevation in 13 2 -microglobulin expression by the pooled transformants is related to the fact that in the conditions used for transformation, nearly 100% of the co-transformants express significant and comparable levels Basic a - Acidic e b • • • • Fie. 7 Fluorographs of two-dimensional gels of the products of the c2 . 1 gene immunoprecipitated from KB-1 cells Ia land tk • transformants lbi using C3H antibodies to K8-1 transformants. The antibodies prepared in C3H mice against c2.1 transformants tKS-1 cells• were used 18 to immunoprecipitate radiolabelled antigens for two-dimensional gel electrophoresis as previously described • The insets contain longer exposures and diagrammatic representations of the 45 .000-molecular weight region of each gel. Black spots represent the specific immunoprecipitated molecules from 8-1 cells. 64 of the foreign class I molecules". It is important to stress that the twofold increase in the magnitude of the radioimmunoassay signal does not necessarily correspond with a twofold increase in the amount of 13 : -microglobulin expressed tsee ref. 36) . These results suggest that the L cells possess the capacity to express additional class I molecules. It has recently been shown that the expression of foreign class I molecules does not occur at the expense of the endogenous H-2' products of the transformants'•. Furthermore, we have observed significant levels of expression of as many as three different products by the triple K" . D" and L" cloned transformant line, which demonstrates even greater than a twofold increase in the 13 : -microglobulin rad ioimmunoassay signal. At least I 0 of the transformants which exhibited increased levels of 13:-microglobulin expression failed to react with any of the known serological reagents to class I molecules tFig. 5a,b /, suggesting that these cells produced novel products associated with 13~-microglobulin . Unfortunately, the heterologous antibodies to 13:-microglobulin proved relatively ineffective in precipitating class I molecules for two-dimensional gel analysis. Thus, we decided to prepare antisera directly against several of these gene products in order to demonstrate formally the presence of novel class I genes. Mouse L cells transformed with the H-2L 0 gene ha ve been used to immunize C3H mice, the strain from which the L cell was derived. to generate alloantisera specific for the L 0 molecule 2 ". Similarly, C3H mice were immunized with L-cell transformants tK8-1l obtained by gene transfer with cosmid clone 2.1, the first gene on cosmid cluster I I see Fig. Sal. By radioimmunoassay. the antibodies specific for K8-J cells reacted weakly with BALB / c spleen cells. whereas significantly lowc:r reactivity was observed against BALB / c liver or thymus cells. and C3H spleen . liver or thymus cells !Fig. 61. Accordingly, these antibodies constitute an alloantiserum that was then used to screen spleen cells from a panel of different congeneic and recombinant congeneic strains of mice so that the location of gene c2 .1 in the major histocompatibility complex could be assigned by preliminary mapping. These serological analyses suggest that gene c2 .1 maps to the right of the complement region . More extensive mapping is required to pinpoint the location of this gene by serological analysis. Such analysis may be complicated if only certain strains of mice express the 2.1 gene products in a manner analogous to that of Q a or TL antigens . However, the data are consistent with the mapp ing of this gene by restriction enzyme polymorphisms. in that class I gene c2.1 is the first gene in the cos mid cluster containing gene 27 .1 , which had previously bec:n mapped to the Tla region" !Fig. Sa). The antiserum to the K8-1 cells immunoprecipitated molecules from K8-1 cells but not from tk- transformants or BALB / c spleen cells !Fig. 7). The products of the c2.1 gene appear to be -45 ,00 0 molecular weight and to consist of a number of spots of the appropriate pi in a sequential array somewhat characteristic of class I molecules analysed on twodimensional gels 1see Fig. 2 1. 13,-Microglobulin could not be readily detected in 8- 1 immunoprecipitates. This may be due to a low affinit y between 13 , -microglobulin and the 2. 1 gene products. as has been described for certain H-2° molecules " . The origin of the additional products precipitated is unknown . The fact that these products cannot be isolated from spleen cells may be due to the fact that this antigen is expressed at low levels on lymphocytes. However, the ability to generate antibodies to the putative novel gene products expressed by the K8-J transformants supports the existence of novel genes. Moreover. these genetic, serological and chemical data suggest that gene c2 . 1 encodes a class I molecule which ma y, because of its location within the major histocompatibility complex, be a d ifferentiation antigen and not a ubiquitously expressed transplantation antigen . Mapping of six serologically defined genes The serologically defined class I genes have been localized within the major histocompatibility complex through the serological analysis of the highly polymorphic congeneic mouse strains. Thus the identification of these class I genes by DSAmediated gene transfer allows us to map precisely the corresponding cosmid clusters into the major histocompatibility complex . It is interesting that only five class I genes shown in Fig. 5 map to the H-2 region !clusters 2;11 and 13!. Three class I genes encode defined transplantation antigens whereas the remaining two apparently do not encode detectable class I gene products. In contrast, 17 class I genes are contained in cosmid clusters which map to the Tla region . Four class I genes encode serologically defined differentiation antigens and eight encode novel gene products. One of the eight novel genes 1c2 . 1 l ma y encode a differentiation antigen based on the pattern of tissue distribution determined with the alloantibodies. Therefore. the Tla region contains far more class I genes than the H-2 region . Cosmid clusters 3, 7, 8, 9 , 10 and 12 do not express serol ogically defined gene products and. accordingly , cannot be mapped within the major histocompatibility complex 1Fig. 51> 1. These six cosmid clusters contain 15 class I genes, six of which appear to express novel gene products. These clusters are now being mapped through the analysis of restriction .enzyme site polymorphisms. Conclusion We have identified at least one H-2 K", D". L". possibly a Qa-2,3. and at least two TL genes from the BALB / c mouse . These genes ha ve been mapped into 6 cosmid gene clusters containing 14 class I genes iFig. 61. of which 9 lie in the Tla region and 5 in the H-2 region . As many as 10 putati'e no,e l class I gene products, detected on the basis of their association with 13 0 -microglobulin . may be encoded by genes dispersed between the mapped and unmapped cosmid gene clusters. In one case, antibodies to a novel class I antigen apparently detect the corresponding antigen expressed at low levels on BALB ;c spleen cells. We thank Drs Shen and Boyse for the gift of the monoclonal antibodies to TL, Drs Hansen and Sachs for the antibodies to the class I antigens, Lorraine Flaherty for the monoclonal antibodies to Qa-2 and Dr Michael Hunkapiller for rabbit antiserum to rat 13 0 -microglobulin . Dr Michael Steinmetz and Mr Henry Sun are acknowledged for supplying clone D!'\A. Ms Marie Krempin for assisting with tissue culture. Dr Jeffre y Frelinger for supplying the congeneic mice, Dr Anders Orn for technical assistance , Drs I won a Stroynowski, Ellen Kraig, \1itch Kronenberg, Ellen Rothenberg; Jane Parnes, Jim Forman and Jeffrey Frelinger for their helpful discussions, and Bernita Larsh for the preparation of the manuscript . This research was supported by NIH grants GM06965, CA22662. CA22199 and CA 25911. R~c~ived I~ Jul~ . accepled ~ 0 September 198::! I. Gorer. P. A . 1. parlt . Bar r 47. ::! ~ 1-:! ~:! · 1938 •. Go rer . P. A .. lyman . S & Snell . G 0 . Pr()( R. Sor . Btl!. -lQQ-~ 0 ~ . tO.at< • _, Zmkernaae l. R . \.1 _& Do her1\ . P C Adt (rnmun . 21 , ~\-\ i":' 19Ril 4 Shearer . G . M & ~chm111 - \"e;hu m . A \1 . Adr. lm.mufl ZS. ~ ~ -Q I 19 .... Flahert\', L. In n, R(llr Of ti!r .\ fa,(lr H ISII'I(Ompa uttc /11\ \olmp lt' l .... /mm. u t~ ll hll•lf1 (; \ ·ed Dorf. \i . E . • J3-S 8 ·Garl and STP~ Pren . 'e"" York . IQHl l• t> Bo ~ s.e . E . A .. Old. L. J. & Lucll. S .V arurr 261. "79 · l9t.·h '7 St anton . T . H . & Bo \'SC. E. A . /mm u ri Pii!t'"rtrn J. ~:!~-SJ1•197f'H 8 Flahert} . L. lm'"u" o grrtrf!(J 3. SB - ~~9 197tll 9. Lopez de Ca~tro , J. A. n a l . B•odunmrn 11. S704 - ~ 7 11 , 1979 1 10 Cohean . J. E .. K1nd1. T . J., \ 'ehara . H .. Martmko. J. & 'arhenson . S G ' "•wm 291. 1 35-39 <1981 • II Stemmetz.\.1 na /. lri/14 , 12S-1~4 · 1981 • 1: Sood . A . K .. Pere tra . D &. We•ssman . S M Proc . "atn . .-"cad. SCL L'.S.A 11. t-ln -t~:n • 1981 •. I~ - Ploe)!h . H . l. _Orr . H T & Stro mmaer . J. l. Proc ,wrn A cad. Sc 1- U .S.A . '17. N18 l-6ft8!' • JQ81l o K\I SI. S r t a { Pr()( •uJtrt . A caJ. S(l U S.A 78. ~"7 > :! 1 "!fol • 198 1•. Stemmetz . M . S " cJ I Crl/ lS , h8 3--h9 2 ol9Rl • 16 Ma lisse n. M .. \-1a linen . 8 6. Jordan . 8 . R PrH< nat" - --"<'tld. Sn L'.S.A. . 19. S•H-89 7 I~ 1~ . ' 19 8 ~ · 17 E\an~ . G. A .. \brguhes . 0 . H .. Ca menm -Otern. R. 0 . Ozaro. K A. Sc1dman . J G. p,,". t~ a r" A ca d S(/ . {. S A. '79. \ 99J - I 9Q~ · ~ ~~ ~~ I 8 ~oodeno"" . R S. rr a l Sc!t' r'l ct 215. t> 7"! -6 N , 198 : ' 65 19. Woodward. J . G . rl a/. f'lroc .atR . ~M. Sec. U.S.A . 'JII, l613-l617ll9i2J. 20 Woodward . J. G . rf al. UQ..A Svl'flt#' """~( . crll. IJiol. lA 11982 1 21 Orn. A. ,., al. ."''aturr 197, 41 !--'17 t 1982•. 22. Moore . K. W., Teylor-Sher , 8 .. Eakk . K. A .. Sun , H E . & Hood . L. SCitrtcr liS. 679-682 d982 L ~3 . Mellor. A. L. r1 at. ."''atutt HI. S29-133 11982 1. 2o& . S1emmcu. M S . W.no1o . A .. M tnard . K. & Hooc:l . L Cr U Zl . 481,1--498 • 198 2 t. 25. Krakaucr . T ., Hansen , T. H .. Camenn• ·Otero, R . D . &: » chs . D . H . J. lmrnu11 . 114, 21 .. 9- 2156 d980 J. 26 han )t. 0 & DCmant . P. /mrnuMCf'tlt' ncs I , ~39 - 5~ 0 • 1919 •. 2'7 ('o·an )• . 0 & DCman t, P lm'" u'WJf''u•uo 11. )9 7--408 , 198 11. 28. oemant , P. & I van ~;, D .1\'orurt- 290. 14b-149 '1 98 t I 29 Deman\ . P. &:: Roos. ~ - H /mmut~o g rt~trln 15. 46 1--4M .J 982 t. 30. lvan vt. 0 &. Demint. P. /mm&mocrrtf'tln U, •67-476 •19R2 • ll Ivan, , D .. Cherry. M. A DC:mant. tt . f"'""'~IWfles IS. C77~S.C 119121. 32 . Hanten, T . H . rf al. 1. /"'"'""" 126, 1713-1716 11981 1. · 33 Shen . f .. w .. Chorney. M &t. Bo yv . E. A . /lftmWNQttMtin lin the: preu 1. 34. Vitelli . E . S.. Poulik, N. 0 .. Klein. J. &: Uht , J. W. 1. up. Mtd. 14W, 179- 192 I 1976 1. 35 . Kran,c . M. S .. On. H. T. & Stromin~er . J. L Ctllll. 979-991 t l97Q , 36. Goodenow . R. S. tt a t. Na ru.rt · ~ubmitted l 37. Schmidt , W .. FC"'len"ern , H., Ward . P. J . &t. S.ndenon. A R./mmu.nortMfl ~J ll, 48 3-491 1!9811. . 38 Szybalska . E. H. & Szybalsh W Proc. nat"" A cad. Se t. U S . A. .U, 20 :! 6-20J.i ' 19b:! l 39 Ozato. K .. Ma ver. !'10 . M. & Sachs . 0 H Tran splo ntatiOrt 1m the pre ss •. CO. Onto . K .. Hansen . T . H. & Sachs . D . H. 1. /mmu.rt . 11!. 1473 - 2477 t 1980 • 41 Forman . J.. Tnal. J., Tonk o no n . S. & Flahert) , L 1 C':r.p. Mtd. 155. 749- 767 •198::! 1. 42 Ozato. K. &:. Sachs , D . H. J. lmmu11 . 116. 317-311 •1981 •. 66 Appendix D Homologies Between Class I Protein Sequences 67 The tables in this appendix contain the results of calculations of protein sequence homology between different Class I molecules. Because of the alternative splicing possibilities, only shared sequences were compared. For example, the third external domain of the H-2Dd molecule contains 95 amino acids, but only the 92 amino acids corresponding to the 92 amino acids in the third domain of the H-2Ld protein were compared in the Hzod /H-2Ld comparison. Homologies are given as percents. than usual homology values are indicated in boldface. Higher The protein sequences compared are those given in Figure 3 of Chapter II. 68 Whole Protein (339 residues compared) H-2Ld H-2Db H-2~ H-2Kb Q8(27.1) H-2Dd 85 H-2Ld 84 80 86 80 95 82 84 80 81 82 80 83 79 H-2Db H-2~ H-2Kb 79 <-< 1) First External Domain (90 residues compared) H-2Ld H-~ H-2~ H-2Kb Q8(27.1) Ql0 H-2Dd 80 H-2Ld 80 78 99 78 77 93 78 84 77 79 78 83 77 80 82 82 78 82 80 H-2Db H-2~ H-2Kb Q8(27.1) 79 Second External Domain (~2) (92 residues compared) H-2Ld H-2Db H-2Kd H-2Kb Q8(27.1) Ql0 H-2Dd H-2Ld H-2ob H-2Kd H-2Kb Q8(27.1) 91 86 84 85 80 80 89 86 82 80 85 83 78 79 82 80 78 85 82 80 82 69 Third External Danain (el3) (92 residues compared) H-2Ld H-2Db H-2~ H-2Kb Q8(27.1) Q1~ H-2Dd 90 H-2Ld 90 87 90 88 90 HKJ 90 90 92 91 90 90 92 91 87 88 87 87 86 H-2oh H-2~ H-2Kb Q8 (27 .1) 90 Transmembrane Segment (Exon 5) (39 residues compared) H-2Ld H-2Db H-~ H-2Kb Q8(27.1) H-2Dd 69 H-2Ld 69 67 64 64 HJ9 67 72 69 67 72 69 82 64 H-2Db H-2~ H-2Kb 74 First Cytoplasmic Segment (Exon 6) (11 residues compared) H-2Ld H-2Db H-2~ H-2Kb Q8(27.1) H-2Dd H-2Ld H-2Db H-2~ H-2Kb 100 100 82 100 73 100 82 100 73 82 100 73 82 62 77 70 Second Cytoplasmic Segment (Exon 7) (13 residues compared) H-2Ld H-2Db H-2~ H-2Kb Q8(27.1) H-2Dd H-2Ld H-2Db H-2~ H-2Kb 85 85 77 85 77 100 62 69 61 62 69 61 92 64 64