Complete amino acid sequence of the 6 heavy chain of human immunoglobulin D ( antibody structure / diversity segment / joining region / lymphocyte receptor )

We have determined the amino acid sequence of the variable (V) region of the 8 heavy (H) chain of human IgD isolated from the plasma ofmyeloma patient WAH. This V region is unusual in its amino end group (arginine) and in its length (129 residues). The length is due to 10 insertions in the third complementarity-determining region (CDR3). A computer search showed that no reported CDR3-joining region (-JH) sequences are identical and that they appear to be unrelated to the constant (C) region sequences of immunoglobulins. The V region sequence together with our previous results for the C region give the complete sequence of the human 8 chain WAH, which has 512 amino acid residues and a Mr 65,000. The human 8 chain has four domains (V, C81, C82, and C83) and a long hinge region; by comparison, the mouse 8 chain lacks a continuous segment of 135 residues, including half the hinge region and the entire C82 domain. The human and mouse 8 chains also differ in the number, kind, and location ofGlcN and GalN glycans and probably in conformation and quaternary structure. These and other considerations suggest that there may be multiple forms of both secreted and membranebound IgD that differ in size, structure, and function. Although sequence data have been published for more than 50,000 amino acid residues of immunoglobulins of various classes and species, until recently (1-8) little was known about the amino acid sequence of human IgD. The reasons include: (i) the low concentration of serum IgD except in rare patients with multiple myeloma, (ii) the apparent lack of a characteristic antibody activity or effector function, (iii) the extreme sensitivity of IgD to "spontaneous" proteolytic degradation (1, 9), and (iv) the unusual structure of the hinge region, which presented difficult technical problems for its determination (5, 6). However, mounting evidence that this minor class of circulating immunoglobulins has a major function as an antigen receptor on the membrane of B lymphocytes (10, 11) led us to undertake determination of the complete primary structure of human IgD. We have reported the complete amino acid sequence of the Fc region (2) and of the C81 and hinge regions (3), which together constitute the entire constant (C) region of the 8 chain of the myeloma protein WAH. Here, we report the structure ofthe variable (V) region ofthe heavy (H) chain (the VH region), which completes the amino acid sequence of the 8 H chain; in work to be reported separately, we have determined the entire sequence ofthe A light (L) chain, which completes the structure of a human IgD molecule. Interest in the structure of IgD was stimulated by evidence that it functions as a major receptor on the surface of B lymphocytes, where it is coexpressed with IgM after antigen capture triggers cell differentiation (10, 11). Like IgM, IgD exists in two forms: sIgD, which is secreted into the serum, and mIgD, which is bound to the B-cell membrane. Cloning studies (12-16) have shown that the A& and 8 structural genes may both be expressed in a single primary transcript that can be processed to yield either IgM or IgD having the same VH region. Although the recent findings on the cellular role and biosynthesis of tL and 8 chains are similar for mice and humans (9-16), and although human and mouse IgM are similar in structure (17, 18), the mouse 8 chain is unusual in that it lacks the C82 domain present in the human 8 chain (1-7). The hinge structures of the 8 chains ofthe two species also differ, and there is unexpectedly low homology in the C81 domains (3). These results suggest that there may be several different forms of IgD in addition to the alternative forms of sIgD and mIgD that appear to differ only in their carboxyl-terminal tailpieces. EXPERIMENTAL PROCEDURES Methods described by Lin and Putnam (1) were used to purify WAH IgD and to prepare the tryptic Fab and Fc fragments of IgD [Fab(t) and Fc(t)], the papain Fab and Fc fragments ofIgD [Fab(p) and Fc(p)], and the A L and 8H chains. CNBr fragments were prepared (1-3) and were separated on Sephadex G-75 and purified by high-performance liquid chromatography (5, 6). Peptides were prepared by digestion with trypsin, chymotrypsin, or Staphylococcus aureus V8 protease (2, 3). The aminoterminal CNBr fragment of the 8 chain was resistant to these enzymes and so was digested with pepsin and also cleaved by iodosobenzoic acid or dilute acid (0.03 M HCl). Peptide digests were separated by gel filtration on Sephadex G-50 and purified by high-performance liquid chromatography (4-6). The methods for amino acid analysis and sequence determination of the peptides with the Beckman model 890C sequencer have been described (1-6). Whereas most human and animal heavy chains begin with a blocked end group (pyrrolidone carboxylic acid) or with glutamic acid (19-21), the WAH 8 chain begins with arginine, which has not been reported previously as the amino end group of any heavy chain. Although Edman degradation of the intact Abbreviations: Fab(t) and Fc(t), tryptic Fab and Fc fragments of IgD; Fab(p) and Fc(p), papain Fab and Fc fragments of IgD. Abbreviations for classes, fragments, regions, and domains ofimmunoglobulins accord with official World Health Organization recommendations for human immunoglobulins published in ref. 34: CDR, complementarity-determining region; L, light; H, heavy; V, variable; VH, variable region of H chain; JH, joining region ofH chain; D, diversity segment ofH chain; C, constant region. * Present address: U-124 INSERM, Institut de Recherches sur le Cancer, BP 311 F-59020 Lille-Cedex, France. t Present address: Indiana University School ofMedicine, Indianapolis, IN 46223. t To whom reprint requests should be addressed. 2850 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact. Proc. Natl. Acad. Sci. USA 79 (1982) 2851 8 chain was successful for only 10 steps, clear results were obtained for 42 steps with the amino-terminal CNBr'fragment. The sequence of the segment from Gln-41 through Arg-68 was difficult to determine because the peptides in this region were obtained only in low yield by enzymatic digestion and chemical cleavage. All positions in the V region were established by at least two independent methods, and all required overlaps of peptides were obtained. RESULTS AND DISCUSSION Length and Mr of the 8 Chain. The WAH 6 chain has the longest V region sequence yet reported for human immunoglobulins (i.e., 129 amino acid residues compared to 126 in the Ou ,u chain (17)-the longest previously recorded). Taken together, the sequences of the V region, the C81 domain (2); the hinge region (2), and the Fc region (3) yield the complete amino acid sequence of the 8 chain of IgD WAH, including the identification of all acids and amides and the sites and nature of the oligosaccharides (Fig. 1).' The entire 8 chain contains 512 amino acid residues; the Mr of the polypeptide portion is 56,213. The three GlcN glycans and the four (or five) GalN glycans contribute a total Mr of --9,000; thus, Mr = 65,000 for the human 8 chain, which accords with our earlier estimate of66,000 +' 1,000 (1). Previous estimates have ranged from Mr = 60,000 to Mr = 69,000, which led to suggestions thatthe 8 chain might contain four C region domains (9, 10). However, our sequence determination shows that there are only three C region domains (C81, C82, and C83) and that it is the extended length of the 8 hinge that makes the 8 chain intermediate in size between the ,u and e chains on one hand and y and a on the other. Structural Model of Human IgD. Because we also have determined the amino acid sequence of the A light chain ofWAH IgD (unpublished data), we can construct a structural model for human IgD. In this model, the light chain is linked by a disulfide bond to Cys-C15 in the C81 domain, and the two heavy chains are joined by a single disulfide bridge at Cys-C161 at the end of the hinge to form a tetrachain monomeric molecule. IgD is the only human immunoglobulin class in which the two heavy chains are joined by a single disulfide bridge. In gel filtration, WAH IgD behaves as a four-chain monomer; the Fc fragment is a dimer because the disulfide bond linking the heavy chains remains intact after limited cleavage with papain or trypsin (1). Special Characteristics of the V Region. Although the V region of the WAH 8 chain has a unique amino end group (arginine), a high concentration and clustering of aromatic amino acids, and exceptional length, it clearly belongs to the VHI1 subgroup of heavy chains, for it has 37 of the 41 residues assigned by Kabat et al (20) as "invariant" in subgroup II. The WAH 8 chain is notable for the clustering of aromatic residues in the first complementarity-determining region (CDR1), where the sequence Tyr-Tyr-Trp-Gly-Trp occurs (see Tyr-34) and in CDR2, where there are two Tyr-Tyr sequences (see Tyr54 and Tyr-60); in addition, there is one Tyr-Tyr sequence in CDR3 (see Tyr-105). There is a similar clustering of aromatic residues in CDR1 ofthe A L chain, where the sequence Tyr-ValTyr-Trp-Tyr occurs. Thus, the CDR regions of this IgD protein epitomize the "ring-of-rings" identified by Edmundson et al (22) as important determinants of combining specificity particularly for aromatic ligands. Length ofthe CDR3 and JH Regions. The greatest variability in amino acid sequence in H chains occurs in the third hypervariability region, earlier called HV3 or the hypervariable deletion region (19) and now often called CDR3 (20). Differences in length of the V region are primarily determined by the CDR3 region, which is also the major determinant ofthe size and shape of the antibody combining site and of idiotypic determinants (19, 22, 23). CDR3 and the joining segment (JH) make up the carboxyl-terminal portion of the VH region. The CDR3 region of a H chain is encoded by a separate DNA segment called the diversity (D) segment, and two recombination events (VH-D and D-JH joinings) are necessary to form a complete V gene (24, 25). The boundaries of the JH and CDR3 segments also have been predicted by comparison of a series of amino acid sequences (26). There are many JH amino acid sequences, but the prototype sequence is X-X-Trp-Gly-Gln-Gly-Thr-X-Val-ThrVal-Ser-Ser in whichX represents positions where substitutions most often occur. Kabat et at (20) set the boundaries of the CDR3 region as V95-V102 in their numbering system and made provision for possible insertions to be numbered as 100A, 100B, etc. For human heavy chains, Andrews and Capra (27) showed that the length ofCDR3 varied from 7 to 17 amino acid residues. By the same criteria, the length of CDR3 in the WAH 8 chain is 19 residues; thus, the WAH 8 chain is -unique in the unmatched length of its CDR3 region. Genetic Origin of the D Region Encoding CDR3. With one apparent exception, all known DNA exons for immunoglobulin chains code precisely for functional segments ofthe polypeptide chains (e.g., the exons for the V and J segments and for each C region domain in H and L chains and the exons for the tailpieces of the secreted and membrane forms of Au chains). The exception appears to be the DNA segment for the D (or CDR3) sequence. Potential germ-line sequences for the D segment are just beginning to be reported for mouse (24) and human (25) DNA. Their size is variable and is smaller than the usual CDR3 region, and their number in the germ-line is unknown. Recent evidence suggests that the human D segments are encoded in tandem multigene families and that they may recombine and be read in different frames (25). Because of the uncertainty about the genetic origin ofthe D region encoding CDR3, we requested Winona C. Barker of the Atlas of Protein Sequence and Structure to use its programs (21, 28) to make a computer search of all reported. amino acid -sequence data for segments homologous to the 30-residue section oftheWAH 8chain from positions 100 to 129 that represent CDR3 and JH. We expected no identical sequence, and there was none. Of almost 210,000 segments of 30-residue length compared, the first 41 having nine or more identities included the CDR3-JH area of the V region of immunoglobulin chains; most of the identities were in the J area, and the sequences all terminated at the end ofthe V region. To our surprise, the highest score (14 identities) was for two mouse H chain V regions (T601 and MOPC315) (Fig. 2). Human and mouse H chain V regions were about equally represented for scores of 13 and 12, and dog and rabbit H chains were among those scoring 11. With the exception ofone human A chain V region, all segments scoring 10 were H chain V regions ofvarious species. No significant relationship was found between the CDR3-JH sequence of the WAH 8 chain and the C region sequence of any immunoglobulin. Hence, we conclude that neither CDR3 nor JH have any evolutionary relationship to genes coding for the C region. Furthermore, the degree of homology in the CDR3-JH region is unrelated to the species, class, or subgroup ofthe H chain. Also, when the computer search was made with the genetic code matrix (21, 28), the ordering in the scores of the H chains was very similar to that shown in Fig. 2. Location, Structure, and Function of Oligosaccharides. Human IgD contains about 12% carbohydrate; this is all bound to the 8 chain and consists of three large GlcN oligosaccharides linked to the Fc region and four or five GalN trisaccharides in close proximity in the hinge region. The GlcN is always Nlinked to an asparagine in the obligatory tripeptide acceptor Biochemistry:'Takahashi et aL 2852 Biochemistry: Takahashi et al Proc. Natl. Acad. Sci. USA 79 (1982) 1 10 20 30 Arg-Leu-Gln-Leu-Gln-Glu-Ser-Gly-Pro'-Gly-Leu-Val-Lys-Pro-Ser-Glu-Thr-Leu-Ser-Leu-Thr-Cys-Ile-Val-Ser-Gly-Gly-Pro-Ile-Arg ArgVi V20 I V30 CDRI 40 50 CDR2 Thr-Gly-Tyr-Tyr-Trp-Gly Trp-Ile-Arg-Gln-Pro-Pro-Gly-Lys-Gly-Leu-Glu-Trp-Ile-Gly Gly-Val-Tyr-Tyr-Thr-Gly-Ser-Ile-Tyr-Tyr35A 35B V40. V50 I 70 80 90 Asn-Pro-Ser-Leu-Arg-Gly1Arg-VAl -ThrIl e-Ser-Val -Asp-Thr-Ser-Arn-Asn-Gl n-Phe-Ser-teu-Asn-Leu-Arg-Ser-Met-Ser-Al a-Ala-AspS V60 V70 V80 82A 82B 82C l ~ ~ ~~~~~ CDR3 I ADJ11 Thr-Al a-Met-Tyr-Tyr-Cys-Ala-Arg Gly-Asn-Pro-Pro-Pro-Tyr-Tyr-AspIle-Gly-Thr-Gly-Ser-Asp-Asp-Gl yIl e]Asp-Val Trp-Gly-Gl nV90 V100 A B C D E F G H I J K(ViOl JH Cdl 140 Light Gly-Thr-Thr-Val -Hi s-Val -Ser-Ser Al a-Pro-Thr-Lys-Al a-Pro-Asp-Val -Phe-Pro-I l eIl e-Ser-Gly-Cys-Arg-Hi s-Pro-Lys-Asp-Asn-SerV113 C1 C1o C20 160 170 180 Pro-Va l-Val -Leu-Ala-Cys-Leu-I l e-Thr-Gly-Tyr-Hi s -Pro-Thr-Ser-Va l -Thr-Va l -Thr-Trp-Tyr-tiet-Gl y-Thr-Gl n-Ser-Gl n-Pro-Gl n-ArgC30 C40 C50 190 200 210 S Thr-Phe-Pro-Gl u-Il e-Gl n-Arg-Arg-Asp-Ser-Tyr-Tyr-Met-Thr-Ser-Ser-Gl n -Leu-Ser-Thr-Pro-Leu-Gl n-Gl n-Trp-Arg-Gl n-Gly-Gl u-TyrC60 C70 C80 220 Cd1a-. Hinge Fd "-ri240 Lys-Cys-Val -Val -Gl n-Hi s-Thr-Ala-Ser-Lys-Ser-Lys-Lys-Gl u-Il e-Phe Arg-Trp-Pro-Glu-Ser-Pro-Lys-Ala-Gl n*Ala Ser-Ser-Val -ProC90 C10o Clio