Structure and Function of Murine Immunoglobulin M from Serum and Cell Membrane Citation Kehry, Marilyn Rose (1980) Structure and Function of Murine Immunoglobulin M from Serum and Cell Membrane. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/nyc2-gv40. https://resolver.caltech.edu/CaltechTHESIS:07232025-144534720 Abstract Immunoglobulin M (IgM) molecules are secreted into the bloodstream by plasma cells as soluble pentamers and also exist as monomeric integral membrane receptor proteins on the surface of B lymphocytes. Investigation of the structure of membrane and secreted μ chains has provided an understanding of the basis for the existence of IgM molecules in two very different physical environments. The complete amino acid sequence of a μ chain secreted by the murine myeloma MOPC 104E has been determined. When the μ chains of mouse, human and dog are compared, there is a striking gradient of increasing amino acid sequence homology from the NH 2 -terminus to the COOR-terminus of the μ chain, reflecting a functional conservation of structure. There are five sites of carbohydrate attachment in the mouse μ chain constant region, one of which is present 14 residues from the carboxyterminus. The secreted μ chain (μ s ) contains no stretches of unchanged amino acids long enough to allow it to exist as an integral membrane protein. IgM molecules synthesized by the murine B lymphoma, WEHI 279, have been characterized. The cells synthesize internal precursors to secreted μ chains which contain incompletely glycosylated complex carbohydrate moieties but in all other respects are identical to secreted μ chains. WEHI 279 membrane IgM is monomeric and contains mature complex carbohydrate structures. The μ m and μ s chains differ in the structure of their COOR-terminal regions. The carbos hydrate moiety and methionine residue present in the COOR-terminal 19 amino acids of μ s chains are absent in μ m chains. In addition, four COOR-terminal amino acids which are different from the carboxyterminus of μ s chains are released by carboxypeptidase treatment of μ m chains. Based on these protein and other nucleic acid data, we believe μ m chains possess an uncharged and hydrophobic C-membrane terminal domain which allows monomeric IgM molecules to be integral receptor proteins in the B cell plasma membrane. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Biochemistry) Degree Grantor: California Institute of Technology Division: Biology Major Option: Biochemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Hood, Leroy E. Thesis Committee: Unknown, Unknown Defense Date: 30 April 1980 Funders: Funding Agency Grant Number Gordon Ross Medical Foundation UNSPECIFIED Weigle Fund UNSPECIFIED Record Number: CaltechTHESIS:07232025-144534720 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:07232025-144534720 DOI: 10.7907/nyc2-gv40 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 17545 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 25 Jul 2025 18:56 Last Modified: 25 Jul 2025 19:35 Thesis Files PDF - Final Version See Usage Policy. 70MB Repository Staff Only: item control page

STRUCTURE AND FUNCTION OF MURINE IMMUNOGLOBULIN M FROM SERUM AND CELL MEMBRANE Thesis by Marilyn Rose Kehry In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1980 (Submitted April 30, 1980) ii To my mother, Edith and my husband Larry iii Many thanks to those who worked and talked with me, Sandy Ewald, Minnie Mc Millian, and John Frelinger; My appreciation goes to my advisor, Lee Hood, for his enthusiasm about science and his humor and receptiveness in listening to me; I give thanks to Carol Sibley, who patiently explained complexities and introduced me to IgM, to Mike Hunkapiller who taught me protein sequencing and allowed me to use his sequenator, to Bertha Jones who made life run smoothly with her efficiency and thoughtfulness, and to all those other Hoodniks and Caltechers who were so helpful along the way. I would like to thank the Gordon Ross Medical Foundation for continuing support and the Weigle Fund for support in preparing this thesis. iv Abstract Immunoglobulin M (IgM) molecules are secreted into the bloodstream by plasma cells as soluble pentamers and also exist as monomeric integral membrane receptor proteins on the surf ace of B lymphocytes. Investigation of the structure of membrane and secreted µ chains has provided an understanding of the basis for the existence of IgM molecules in two very different physical environments. The complete amino acid sequence of a µ chain secreted by the murine myeloma MOPC 104E has been determined. When the µ chains of mouse, human and dog are compared, there is a striking gradient of increasing amino acid sequence homology from the NH -terminus to the COOR-terminus of the µ chain, reflecting 2 a functional conservation of st,ructure. There are five sites of carbohydrate attachment in the mouse µ chain constant region, one of which is present 14 residues from the carboxyterminus. The secreted µ chain (µs) contains no stretches of unchanged amino acids long enough to allow it to exist as an integral membrane protein. IgM molecules synthesized by the murine B lymphoma, WEHI 279, have been characterized. The cells synthesize internal precursors to secreted µ chains which contain incompletely glycosylated complex carbohydrate moieties but in all other respects are identical to secreted µ chains. WEHI 279 membrane IgM is monomeric and contains mature complex carbohydrate structures. The µ m and µ s chains differ in the structure of their COOR-terminal regions. The carbo- hydrate moiety and methionine residue present in the COOR-terminal 19 amino acids of µ s chains are absent in µ m chains. In addition, four COOR-terminal amino acids which are different from the carboxyterminus of µ s chains are released by carboxypeptidase treatment of µ m chains. Based on these protein and other nucleic acid data, we believe µ m chains possess an uncharged and hydrophobic C-membrane terminal domain which allows monomeric lgM molecules to be integral receptor proteins in the B cell plasma membrane. V TABLE OF CONTENTS Page Acknowledgements . . . . . . . ii ............. ... iv Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Abstract . . . . . . . . . . . . Chapter 1 Amino acid sequence of a mouse mu chain . . . . . . . . . Chapter 2 Complete amino acid sequence of a mouse mu chain: homology among heavy chain constant region domains 24 Characterization of multiple immunoglobulin mu chains synthesized by two clones of a B-cell lymphoma . . . . 64 The immunoglobulin µ chains of membrane-bound and secreted IgM molecules differ in their C-terminal segments. . 107 Conclusion . . . . . . · . . . . . . . . . . . . . . . . . . . . . . . 17 3 Appendix 184 Chapter 3 Chapter 4 Supplementary figures 1 INTRODUCTION 2 Antibody molecules constitute an important part of an organism's major defense system against disease and foreign substances. This system of humor al immunity exhibits remarkable specificity, diversity, memory and exquisite control. Despite its importance, knowledge of the detailed molecular events governing an individual's immune response to a foreign antigen is still sparse. One of the most fascinating and as yet unanswered questions in the intricate process of antibody production is how a specific antibody secreting cell, a B lymphocyte, is stimulated by an antigen to divide and differentiate into a clone of antibody producing plasma cells. More than ten years ago, it was found that a B lymphocyte possesses an tibody molecules on its cell surface which are available for interaction with a complementary antigen (literature review in 1). This then established the specificity of the B cell triggering process. The predominant class of antibodies present as integral membrane proteins on the cell surface of B lymphocytes is called Immunoglobulin M (IgM) (1). The IgM class of antibodies is also secreted by plasma cells as a soluble molecule and circulates in the bloodstream, where it binds and clears antigen, fixes complement, and functions in the humeral immune system. In the bloodstream IgM exists as a .covalently linked pentamer, yet on the B cell surface where it plays a crucial role in antigen induced differentiation, IgM is present as a monomer. My research has used the well-characterized immune system of the mouse to investigate the structural basis which allows IgM molecules to exist in two very different environments: as water soluble pentamers in the bloodstream and as monomers in the hydrophobic plasma membrane. Delineation of the structural differences between membrane bound and secreted IgM molecules may provide an understanding of the basis for their different effector functions and perhaps lend insight into the process of B lymphocyte triggering by antigen. I will l) review antibody structure, 2) summarize what is currently known about 3 the B cell differentiation process in the mouse and 3) review the work done on this problem by other investigators over the last nine years. Antibody Structure (2) Antibody molecules are composed of two different polypeptide chains, heavy (H) and light (L) chains which are held together by disulfide bonds and form a monomeric immunoglobulin unit, H L . Each chain is divided into a variable 2 2 (V) or antigen binding region and a constant (C) region. There are five different classes of H chain which differ in their constant region structures: y, µ, a, e: and o. The five corresponding classes of immunoglobulins are designated IgG, IgM, IgA, IgE and IgD, respectively. The V regions of the H and L chains fold together to form the antigen combining portion of the antibody. The constant region of a y heavy chain, for example, is composed of three structural domains (based on X-ray chrystallographic analyses, structural studies, and amino acid sequences) (3), Cyl, Cy2 and Cy3, which carry out the effector functions of the IgG molecule. Effector functions are carried out by the antibody following antigen binding. Cyl folds with the c 1 domain and is separated from the Cy2 and Cy3 domains (formed by the homologous association of the two y chains) by a region of great structural flexibility in the y chain, the hinge. The basic structure of an IgG molecule is illustrated in Figure 1. The structure of serum IgM differs in several important respects from that of IgG. Mu chains contain four constant region domains, one additional domain than is present in y chains. Mu chains have no hinge region but possess a 20 amino acid carboxyterminal extra-domain segment after the Cµ 4 domain (3). The molecular weight of µ chains is 72,000 and they are moderately glycosylated (13% carbohydrate by weight) (4). This is the highest molecular weight of 4 Figure 1. Structure of the IgG molecule. The heavy (H) and light (L) chains are shown as they interact to form the functional domains of an IgG molecule. Variable (V) regions are stippled and the hinge region is shaded. Disulfide bridges between the chains are indicated by heavy black lines. The portions of the molecule produced by papain cleavage in the hinge region are designated as the Fab (antibody binding) and Fe (crystallizable) fragments (adapted from reference 2). V CH1 homology units domains-<. (.11 V regions C regions CH2 I\) Fe CH3 FIGURE 1. I~ LJ I Vt l CS: ( Vtf ) I CH1 I CH2 CH3 6 any of the different heavy chain classes (e.g., y chains have a molecular weight of 55,000), and mu chains are approximately three times larger than light chains. Serum µ chains have five sites of carbohydrate attachment in the constant region (5, 6). The NH -terminal three carbohydrates are complex type and the COOH2 terminal two are high mannose. Noteworthy is the fact that the most COOHterminal carbohydrate moiety is attached to the 20 amino acid extra-domain region of the µ chain. The COCH-terminal region of the µ chain plays an important role in the existence of pentameric IgM (µ L2\ (IgG exists only as a monomer). Prior 2 to secretion, IgM monomers polymerize covalently by forming disulfide bonds between the penultimate cysteine residues in the µ chains (7). A third polypeptide chain denoted joining (J) chain, also is present in the IgM pentamer (one per pentamer), and is thought to play a role in facilitating IgM secretion by catalyzing the polymerization of monomers (8, 9). IgM is also unique in that it is the most ancestral of the five immunoglobulin classes. Antibody molecules synthesized by organisms with very primitive humoral immune systems are composed of H chains similar to the µ chain of higher mammals (10). In addition, during the development of the immune system IgM is the first class of antibodies expressed - both as a membrane bound receptor and as a soluble pentamer (11). Thus, IgM is a fascinating immunoglobulin on which to perform structural and evolutionary studies. Murine B Lymphocyte Differentiation In the fetus, lymphoid cell precursors are found in the liver (12, 13, 14). (Undifferentiated lymphoid precursors or stem cells in the adult are derived from the bone marrow and migrate to the spleen where they undergo differentiation steps similar to those of fetal liver stem cells.) The earliest B cell stage is 7 detected in the fetal liver at 11 days gestation (12). This large dividing pre-B cell contains cytoplasmic µ chains, no light chains, and has no IgM on the cell surface (12, 15, 16, 17). The large pre-B cell differentiates further into a small pre-B cell that is nondividing and again contains cytoplasmic µ chains (15). Additional steps (which include initiation of light chain synthesis) then give rise to a small nondividing B lymphocyte displaying membrane bound monomeric IgM on its cell surface (1). This, then, is the B cell which may be triggered by antigen to divide and differentiate into plasma cell clones secreting IgM, IgG, IgA, etc. Most B lymphocytes in the adult spleen also display IgD on their cell surfaces (8). Membrane IgM appears before IgD on differentiating lymphocytes (19). Recent studies show that IgD may allow paucivalent antigens to more effectively stimulate the resting B lymphocyte (20), although the exact role played by IgD in initiation of the humoral response remains unclear. Structural information on c heavy chains is currently being obtained in several laboratories. The cellular changes accompanying B lymphocyte differentiation to IgM secreting plasma cells are also unknown, but clearly initiation of J chain synthesis is required (21). At the level of antibody genes, recent findings have shown that the commitment of a pre-B cell to µ chain synthesis includes a rearrangement of DNA so that a given VH gene segment becomes situated 5' to the Cµ gene (22). In addition, changes accompanying B lymphocyte differentiation to plasma cells also frequently involve DNA rearrangement; the differentiation of a B lymphocyte synthesizing membrane IgM to a plasma cell synthesizing IgG or IgA requires that the VH gene segment adjacent to the expressed Cµ gene be moved adjacent to the Cy or Ca gene. This pattern of gene rearrangement occurs in IgA and IgG secreting myelomas (23). Clearly, B cell differentiation and triggering are complicated processes. An ideal starting point for studying antigen interaction with B lymphocytes is to 8 study the antigen receptor, monomeric membrane IgM. By investigating this receptor, we hope to understand the changes involved as the differentiating B cell begins to synthesize and secrete pentameric IgM and ceases synthesis of cell surf ace receptor IgM. History of Work Done on Membrane IgM The existence of monomeric IgM on the surface of murine splenic B lymphocytes was first documented in 1973 by Vitetta and coworkers (1). The role of membrane IgM in B cell triggering was subsequently clarified by many investigators. Although antigen binding to cell surface IgM is not the only signal which is required for B cell activation (24), antigen interaction is extremely important in determining the specificity and the memory of an antibody response. The general structure of human serum IgM has been known for several years (25), and an obvious question was how the membrane and secreted forms of IgM could have such similar overall structures, yet exist in such different physical environments and carry out different effector functions. Early studies focused on the logical hypothesis that IgM molecules were attached to the cell surface via their µ chains, leaving the antigen binding V domains available for interaction with molecules in the external environment. Therefore, membrane µ (µ m) chains were isolated and compared with secretedµ (µs) chains. Membrane µ chains could be dissociated from the cell surface only by detergents, implying that membrane IgM is an integral membrane protein (26, 1). A subtle molecular weight difference between µ m and µ s chains was also apparent, µ m being larger than µs by approximately 1500 daltons (27). Attempts at examining the carbohydrate structure of µ m chains by incorporation of radiolabeled sugars showed conflicting results due to the slow turnover rate of membrane IgM in resting splenic B lymphocytes (28, 29). Anderson and coworkers found that µ m chains were not 9 glycosylated with terminal sugars as were µs chains (29). Yet, Vitetta and Uhr showed significant incorporation of galactose, fucose and glucosamine into surface IgM (28). We now know that the carbohydrate structures of µ m chains are fully glycosylated, although µ m lacks a glycosylation site present in µ s chains (30, 31). Low specific activity radiolabeled sugars and isolation procedures for membrane IgM may have been a problem in these previous experiments (28). Serological studies also suggested that certain antigenic determinants present on secreted IgM were absent on membrane lgM bound to the B cell surface. These differences were serologically localized to the Fe region of membrane IgM, implying that a more COCH-terminal portion of the µ m chain is responsible for structural differences between membrane and secreted IgM (1, 32). Detergent binding studies also showed that membrane IgM bound small but significant amounts of detergent when compared with secreted IgM (33, 34). The µ m chains used in this early work were derived from resting B lymphocytes isolated from the spleen. There are many difficulties inherent in structural studies on such B cells and their membrane IgM molecules. These difficulties include very small quantities of protein, low protein turnover rates, a heterogeneous population of µ m molecules (different Vtt's), a heterogeneous population of spleen cells in different developmental stages, and relative to plasma cells, poor incorporation of radiolabeled amino acids and sugars. For the above reasons, many investigators now have turned to studying B cell tumors as a source of membrane IgM. In the murine system of inbred strains, experimenters have found a spectrum of B cell tumors which seem to be arrested at various stages in the differentiation pathway from pre-B cells to plasma cells (36, 37). These differentiation stages can be distinguished morphologically and serologically on the basis of characteristic cell surface proteins. The cells are grown in continuous culture, divide, and synthesize homogeneous antibody molecules. Nonetheless, one must be cautious in 10 interpreting experiments performed on tumor cells. The developmental stages may simply represent, for example, blockage of a biosynthetic step (e.g., J chain synthesis) required for IgM secretion. Extensive characterization of the cells is necessary before valid conclusions can be drawn from experiments involving tumor lines. In particular, we have found that B-cell lymphomas (the tumor equivalent of a nonsecreting small B lymphocyte) synthesize an internal form of µ chains s (pre-secreted µ) in addition to µ m chains (30, 31). Solubilized whole cells then yield a population of µ chains consisting of µ and µ s m molecules which need to be completely separated from one another for use in structural studies. In particular, this situation has caused two independent research groups who performed similar experiments on different B cell tumors to draw conflicting conclusions regarding the structural nature of the COOR-terminus of µ m chains. By digesting µ m chains with carboxypeptidase, Feinstein and coworkers found that the COORterminus ofµ m chains was identical to that of µs chains (38) (which possess a COOR-terminal Tyr residue). On the other hand, Grey and coworkers found that several hydrophobic amino acids other than Tyr were released from µ m chains by carboxypeptidase (39). Our findings indicate, though, that the two results can be reconciled by considering the fact that µ m and µ s chains are indeed different at their COOR-termini and that many B cell lines synthesize both a membrane bound and an internal presecreted form of µ chains (30, 40, 41, 42). Peptide mapping studies on µ m chains derived from a B-cell lymphoma also showed that small peptide differences exist between µ m and µ s chains. Unfortunately, these differences were situated on a high background due to the design of the experiment (multiple labeled amino acids and incomplete enzyme digestion) (35). Other investigators have more recently characterized the IgM molecules synethsized by other B cell tumor lines. Our structural findings have been complemented by these more indirect studies. A brief summary to date of published 11 and unpublished findings is as follows. B cell lines do synthesize both µ and µ m s chains in varying proportions depending upon the individual cell line (30, 40, 41, 42). µ m chains are clearly larger in molecular weight than µ s chains. The molecular weight difference between µ m and µ s chains is attributable to a difference in the primary amino acid structures of µ m and µ chains. In particular, µ s m chains synthesized in the presence of the glycosylation inhibitor tunicamycin are 1500 daltons larger than nonglycosylated µ s chains (30, 34, 43, 44). In addition, translation ofµ m andµ s mRNA in vitro results in the synthesis of a µ --- m poly- peptide which is larger in molecular weight than the µ s polypeptide (43, 44, 45). µ m possesses a different and longer C-terminal domain which contains a hydro- phobic region that may be capable of interacting with the lipid bilayer (31, 34, 46). This altered C-terminal domain explains all of the observed differences between µ m and µ s chains. The two chains differ in the number of carbohydrate attachment sites, µ m lacking a high mannose carbohydrate structure present in the C-terminal region of µ s chains (30, 31). The 20 additional amino acids in µ m chains plus lack of a high mannose carbohydrate structure is reflected in the 1500 dalton molecular .weight difference between mature µ m and µ chains s (30, 43, 44, 46). Finally, the µ s chains have a COOR-terminal Tyr residue (4, 6), while µ m chains have a different C-terminal domain containing COOR-terminal Val (31). This is consistent with the inability to label a Tyr residue in µ m chains on the inside of the plasma membrane (47). The biochemical changes which occur in µ m chains will be discussed in greater detail in the following chapters and conclusion. General Approach of Research When this project was started four years ago, virtually nothing was known about the structure of mouse µ s chains (the sequence of human µ s chains was 12 determined in 1973 by Putnam and coworkers (25)). Since large quantities of µ s chains for structural determinations are obtainable from mouse myelomas, our strategy was as follows. The structural comparisons we wished to make between µ s and µ m chains were at the level of the primary amino acid sequence. Therefore, once we had detailed structural information on µ s chains, comparative studies with µ m chains should be relatively straightforward. It was decided to then characterize the cyanogen bromide (CnBr) fragment structure of the µs chains synthesized by the mouse myeloma line MOPC 104E (a plasmacytoma induced in BALB/c mice by mineral oil injection). This particular µ s chain was interesting for several reasons: i) MOPC 104E (M104E) IgM specifically binds the simple hapten ~1,3-dextran, an antigenic constitutent of many bacterial cell surfaces; ii) a particular dextran idiotype had been defined using Ml 04E IgM; iii) IgM biosynthesis and secretion had been well studied in M104E cells (4, 48, 49) and; iv) M104E µ s chains were used by a large number of investigators as a "reference standard" µ chain for overall structural characteristics such as molecular weight and glycosylation. Two years of classical and not so classical protein chemistry (which included devising procedures for handling the relatively insoluble M104E µ s chain and its CNBr fragments) brought the structural work to 80% completion. During the next year when the primary structure was completed and carbohydrate attachment sites were identified, a rabbit antiserum to M104E µ s chains was made and work was started on characterizing membrane IgM molecules. We had originally planned on large scale isolation of membrane µ chains from mouse splenic B cells. But at this time, many new B cell tumor lines became available, and because these lines offered advantages for studying membrane IgM, we began characterizing the WEHI 279 B-cell lymphoma (37), a chemically induced tumor found in the thymus of NZC mice [an inbred strain derived from (NZB x BALB/c)F mice]. These WEHI 279 tumor cells produce the parental 1 BALB/c allele ofµ chains and were a gift from Dr. N. Warner. 13 Our approach was to 1) analyze the primary amino acid sequence of µ m chains using direct radiolabeled sequencing of CNBr fragments and comparative peptide mapping and 2) label sugars and assess the glycosylation levels and sites on µ m chains. In these studies, we naturally intended to focus on the COOHterminal region of the µ m chain. Initial experiments on WEHI 279 cells revealed the existence of a rapidly labeled internal pool of µ chains, a membrane bound µ chain synthesized by these cells and little if any ( 10% of the total) µ chain secreted into the culture fluid. Thus, the cells seemed representative of small B lymphocytes from the spleen. The majority of µ chains in the in tern al ( µ . ) pool was found to consist 1 of percursors to µ s chains. These pre-secreted µ chains lacked terminal sugars on their complex carbohydrate structures, and since secretion was minimal, apparently turned over nonproductively inside the cells. We then realized that WEHI 279 µ s chains (with a WEHI 279 VH region) were needed for making structural comparisons with WEHI 279 µ m chains at the level of peptide mapping and cyanogen bromide fragmentation. The hybridoma technology was able to supply the solution. Dr. W. Raschke at the Salk Institute kindly sent us a hybrid cell line created by the fusion of WEHI 279 and MPCll myeloma cells. The MPCll x W279.2 (MxW) cells secrete pentameric IgM composed of WEHI 279 µ s chains and were used subsequently for comparison of the polypeptide structures of µ m' µ i and µ s chains. The hybridoma was amplified as ascites in BALB/c mice, and 80% of the VH region amino acid sequence of the WEHI 279 µs chain was determined in order to quantitate the different V region ai:nino acids. At this point, we were sent a replacement clone of WEHI 279 cells which was used for all further studies on µ and µ. chains. This new clone is able to m 1 polymerize and secrete the internal pre-secreted µ chains more efficiently than the original WEHI 279 cells. The more recent clone secretes 15-fold more IgM , 14 and 46% of the secreted IgM is fully polymerized into pentamers. As suggested by M. Koshland and coworkers, the ability to secrete mature IgM pentamers probably reflects the availability of J chains in the cell (21, 50). In support of this idea, Mather and Koshland have found that original WEHI 279 cells produced little J chain; MPCll cells were found to synthesize large amounts of J chain as were MxW cells (21); recent clone WEHI 279 cells synthesized 100-fold more J chain than the original WEHI 279 cells (50). We believe that this change in the cell line in no way affects the studies on µ m isolated from both cell lines. Since the V region CNBr fragments of µ m chains from both lines are identical, the µ m chains of the two lines should be identical. The following chapters are a chronology of findings first. about µ s structure, second, how the different µ chains synthesized by WEHI 279 B lymphoma cells were localized and identified, and finally the structural characterization of the WEHI 279 µ chains (with the complement of recombinant DNA technology) to show that µ and µ chains possess different primary amino acid sequences m s at their COOH-termini. 15 REFERENCES 1. Vitetta, E. S., Baur, S., and Uhr, J. W. i[_. Exp. Med. 134:242 (1971). 2. Hood, L. E., Weissman, I. L., and Wood, W. B. Immunology. The Benjamin/ Cummings Publishing Company, Inc., Menlo Park, California (1978). 3. Beale, D., and Feinstein, A. Quart. Rev. Biophysics 9:135 (1976). 4. Robinson, E. A., Appella, E., and McIntire, K. R. i[_. Biol. Chem. 248:7112 (1972). 5. Shimizu, A., Putnam, F. W., and Paul, C. Nature 231:73 (1971). 6. Kehry, M., Sibley, C., Fuhrman, J., Schilling, J., and Hood, L. E. Proc. Natl. Acad. Sci. USA 76:2932 (1979). 7. Mestecky, J., and Schrohenloher, R. E. Nature 249:650 (1974). 8. Della Corte, E., and Parkhouse, R. M. E. Biochem. i[_. 136:597 (1973). 9. Koshland, M. E. Adv. Immunol. 20:41 (1975). 10. Marchalonis, J. J. Nature 236:84 (1972). 11. Lawton, A. R., Kincade, P. W., and Cooper, M. D. Fed. Proc. 34:33 (1975). 12. Melchers, F., von Boehmer, H., and Phillips, R. A. Transplant. Rev. 25:26 (1975). 13. Phillips, R. A., and Melchers, F. i[_. Immunol. 117:1099 (1976). 14. Owen, J. J. T., Cooper, M. D., and Raff, M. C. Nature 249:361 (1974). 15. Melchers, F., Anderson, J., and Phillips, R. A. Cold Spring Harbor~Quant. Biol. 41:147 (1976). 16. Owen, J. J. T., Wright, D. E., Habo, S., Raff, M. C., and Cooper, M. D. i[_. Immunol. 118:2067 (1977). 17. Burrows, P., LeJeune, M., and Kearney, J. F. Nature 280:838 (1979). 18. Melcher, U., Vitetta, E. S., Mc Williams, M., Lamm, M. E., Phillips-Quagliata, J.M., and Uhr, J. W. i[_. Exp. Med. 140:1427 (1974). 19. Yuan, D., and Vitetta, E. S. i[_. Immunol. 120:353 (1978). 20. Vitetta, E. S. Unpublished results. 16 21. Raschke, W. C., Mather, E. L., and Koshland, M. E. Proc. Natl. Acad. Sci. USA 76:3469 (1979). 22. Early, P. W., Huang, H. V., Davis, M. D., Calame, K., and Hood, L. Cell. In Press. 23. Davis, M. M., Calame, K., Early, P. W., Livant, D. L., Joho, R., Weissman, I. L., and Hood, L. Nature 283:733 (1980). 24. M"oller, G. Transplant. Rev. 23:126 (1975). 25. Putnam, F. W., Florent, G., Paul, C., Shinoda, T., and Shimizu, A. Science 182:287 (1973). 26. Melcher, U., Eidels, L., and Uhr, J. W. Nature 258:434 (1975). 27. Melcher, U., and Uhr, J. W. !!· Immunol. 116:409 (1976). 28. Vitetta, E. S., and Uhr, J. W. !!· Exp. Med. 139:1599 (197 4). 29. Anderson, J., Lafleur, L., and Melchers, F. Eur.!!· Immunol. 4:170 (1974). 30. Sibley, C.H., Ewald, S. J., Kehry, M. R., Douglas, R.H., Raschke, W. C., and Hood, L. E. !!· Immunol. Submitted. 31. Kehry, M., Ewald, S., Douglas, R., Sibley, C., Raschke, W., Fambrough, D., and Hood, L. Cell. In preparation. 32. Fu, S. M., and Kunkel, H. G. !!· Exp. Med. 140:895 (1974). 33. Melcher, U., and Uhr, J. W. Biochemistry 16:145 (1977). 34. Vassalli, P., Tedghi, R., Lisowska-Bernstein, B., Tartakoff, A., and Jaton, J.-C. Proc. Natl. Acad. Sci. USA 76:5515 (1979). -- - - - - -- -- 35. Yuan, D., Uhr, J. W., Khapp, M. R., Slavin, S., Strober, S., and Vitetta, E. S. In B Lymphocytes in the Immune Response (M. Cooper, D. E. Mosier, I. Scher, and E. S. Vitetta, eds.). Elsevier North-Holland, Inc.: New York, New York (1979), p. 23. 36. Anderson, J., Buxbaum, J., Citronbaum, R., Douglas, S., Forni, L., Melchers, F., Pernis, B., and Stott, D. !!· Exp. Med. 140:742 (1974). 17 37. Warner, N. L., Leary, J. F., and McLaughlin, S. In!! Lymphocytes in the Immune Response (M. Cooper, D. E. Mosier, I. Scher, and E. S. Vitetta, eds.). Elsevier North-Holland, Inc.: New York, New York (1979), p. 371. 38. Mcilhinney, R. A. J., Richardson, N. E., and Feinstein, A. Nature 272:556 (1978). 39. Williams, P. B., Kubo, R. T., and Grey, H. M. ~- Immunol. 121:2435 (1978). 40. Bergman, Y., and Haimovich, J. Eur. ~. Immunol. 8:876 (1978). 41. Perry, R. P., and Kelley, D. E. Cell 18:1333 (1979). 42. Wall, R., and Rogers, J. Unpublished results. 43. Bergman, Y. Unpublished results. 44. Williamson, A. Unpublished results. 45. Raschke, W. Personal communication. 46. Early, P. W., Rogers, J., Davis, M., Calame, K., Bond, M., Wall, R., and Hood, L. Cell. Submitted. 47. Walsh, F. S., and Crumpton, M. S. Nature 269:307 (1977). 48. Melchers, F. Biochemistry 11:2204 (1972). 49. Stott, D. I., Immunochemistry 13:157 (1976). 50. Mather, E., and Koshland, M. Unpublished results. 18 CHAPTER 1 This paper was published in ~o~. ~a tl. A cad. Sci. USA. 19 Amino acid sequence of a mouse immunoglobulin µ chain (automated seq uencer/i mmunoglobulin domain / myelo ma lumor MOPC I04E / immunoglobulin evolulion / carbohydrate atlachment siles) M. KEHRY*, C. SIBLEYt. J. FUHRMAN* , J. SCHILLING*, AND L. E. Hooo* * •Division o( Biology . Ca li fornia Institut e of Tt"'(.'hoology , Pasacit•na . California 91125. and 1 Dt-parlmt"nt o f (;t'nt>lks . Univns1t y o ( Washm~ton. S.•attle. Washin~t<>n 9 1895 Co mmunicated by Edward B. Lewis. March 26. 19i9 ABSTRACT The complete amino acid sequence of the mouseµ chain from lhe BALB /c mye loma lumor MOPC 104E is reported. The C, region contains four consecutive homology regions of approximately 110 residues and a COOH-terminal region of 19 residues. A comparison of thisµ chain from mouse with a completeµ sequence from human (Ou) and a partialµ chain sequence from dog (Moo) reveals a striking gradient of increasing homology from the NHz•lerminal lo the COOH-terminal portion of theseµ chains, with the former being lhe least and the laller the most highly conserved. Four of the five sites of carbohydrate attachment appear to be al identical resi due positions when the constant regions of the mouse and human µ chains are compared. Theµ chain of \IOPC 104E has a carbohydrate moiet y attached in the second hypervariable region. This is particularly interestin,i: in view of the fact that MOPC 104E bind s a-{1-3)-de,tran . a simple carbohydrate. Tht> structural and functional constraints imposed by these comparative sequence analyses are discussed. lmmunoglobulins are comprised of two polvpeptides. li g ht ( L) and heavv (H ) chains . and ma y be divided int o fiw major classes. lgM. lgG . lgA. lgD. and lgE. which are defint-d bv their corresponding heavy chains. µ . r . n. o. and c respedively . The lgM molecule is the most primitive immunoglobu lin in thal il is the first to appear in vertebra te evolution ( I ). Mon·over . lg\l is the first immunoglobulin to appear during the ontogrnetic development of the immune system (2). Serological. immunoehemical. and evolutionary data suggest thal lgt--1 immunoglobulins have been more highly conserved than olhrr classes of immunoglobulins ( I ). For these reaso ns. the lgM mo lec ul<' is an excellent model protein for e volutionary ana lysis. Each imnrnnoglobulin chain is divided into an '.\/Hi-tt•rminal variab lt> (V) rt•gion and a C:OO H-l er minal constanl (C) region. The imnrnnoglohulin polvpeptidt>s are dividi,d into eonSt'<'utiw homo log, regions approximatelv 110 residut>s in lt•ngt h. The L chain has two homo logy regions and the humanµ c hain has five homolog, regions in add ition to a COOH-terminal segment composed of 19 rrsidu es (3). The basic unit of the immunoglobulin molecule is two light and two hean chains (HlL 1 ). in which pairs of homology regions fold tog,·lher to form globular domains. The L c hain and the N H i -lerminal portion o f tht• H c hain gene ratt> two domains each. and lhe remaining C:OO1-1terminal regions of the twoµ chai ns fold into three domains. From the NH1 lo the COO H terminus of theµ chain. tht>se domains ar .. designatrd V. C, I. C"2. C,3. and C,-l. respectivelv . Tht> V domains recognize foreign structures (a ntigens ) and the C domains carrv out a varietv of effector functions that ultimately lead to i°he destructio~ or elimination of foreign antigens. The lgM molecule has at least three special effector functions associated with its COOi-i-terminai domains. First. Th .. ptthlication costs of this artidP wt'rP cJpfravP<l in part b, pa~<' chargt' pavmt>nl. This article must thnPforP he hPrPby markPd .. ad tx•rtisenienl •• in accordancP with 18 U. S. C:. ~ I73~ sole!,· to in<li catt' this fal't. • • the IgM molecule serves as a ce ll -su rface receptor that can trigger the progeni tors of antibody-producing B cells to differen tiatr upon interaction with complementary antigen (-l ). Thus the lg\l molecu le ca n be an integral membrane protein (5. 6 ). which is displayed on tht> m e mbrane as a iS monomer . (µ1L 1) (7. 8). Second. cr lls that have been stimulatt>d to differentiate into antibody-producing cells can secrete Ig\ l m o lecul es as I 9S pentanwrs. (µ 2 L1k composed of IO light and I 0 heavv c hains held together by a third type of polypeptid e designatt'd the J c hain (9). It is not known whether the m embrane-bound and secrett-d µ cha ins ha ve the same pol,peptide structure. Finalh . after the lgM molecule binds antigen. th e C,-l domain activalt'S the effector pathways of the complement svslem ( 10 1. • In this paper we rPpo rt the amino acid seq uence of the sec reted µ chain derivt'{I from the mouse myeloma tumor \IOPC 10.JE. The mouseµ cha in is co mpart>d with a humanµ cha in whose St'qUenc,• has llt't'n complele ly determint><I ( 11 ) and a dog µ chain wh=· sequPnC't' has bet•n partially determined ( 12. 13 ). As prr,ioush- observed when the human µ sequence and th e dogµ partial sequent·e wert• co mpared. there is a striking gra dient of sequence hom olog,. with the COO H-lerminal hom o log, regions significantly more conser ved than the \"H 2-terminal homology regions ( 12 ). St>wral structu ral features of the µ chai11 are high I, rnnse rvt•d. including the placemt>nl of disulfide bridges and the loca tions of tryptophan rt>Sidues a nd carhofwdra le moietiPs in the constan t region The structu re of tht> secn·tt>d µ chain pbct>~ sevt>ral inlerest ing constrain ts on lht> naturt> of the mt>mbrant'-bound µ cha in of the lg\l receptor m olecu lt> MA TE RIALS A:\D METHODS Isolation ofµ Chain. The lg\1- seneting tumor \IOPC 10.JE (µ . ,\ ) "as grown intraperitonealh in (BA LB ·c X DB:\.' :2 1 F 1 mice . The lg\l molecules were prt•cipitatt>d from asci t('S fluid with ammonium sulfal <' a11d purified I)\· ge l filtration on a column of AC-\ 22 (LKB ) The purifit>d IgM was completeh red ucC'd with dithiothreilol and alb·lated with iodoacrtamide in fi \I guanidirw-HC:I (1-l). and lht> l;eavy and light chains were separated b,· gel filtration on a co lumn of AC.-\ :3-l (LKB ) in 3 M guanidine and 0.2 M ammonium bicarbonate Thr complete amino acid sequenct> of tht> ,\ li ght cha in nf 1'.IOPC 10-IE has been dt>termined ( 1.5. 16) Isolation of Cyanogen Bromide Fragments. Purifi ed µ chains were cleaved with cya nogen bromide in i0% (wt / vo l) formic acid for 20-22 hr at -l ., C (17) . Cyanogen bromide peptides were separatrd by gt'I filtration on a column of AC.-\ .5-1 (LKB) in 3 M gua nidine and 0.2 t--1 ammonium bicarbonate. Abbr,.,·iations Hand L. ht'an and light l'hains of immunoglobulins: \" a11d C. , aria hit• and constant rpgions of immunoglobulin l'hains: CHO. ,·arlx,h, dratP I To whom rr.print rn p1Psts shou ld he add rt'ssed. 20 Protein and peptide purity was monitored at all stages of purification by electrophoresis in the presence of sodium dodecyl sulfate on 10% or 18% polyacrylamide gels (18). Compositional and Sequence Analysis. Amino acid compositions and carbohydrate analyses for amino sugars were determined on samples hydrolyzed in 6 M HCI (I l0 '' C for 2-l hr and 3 hr. respectively) by using a Durrum D-500 amino acid analyzer. Automated sequence analyses were performed on a modified Beckman sequenator (19. 20). Samples were loaded by using Polybrene (Aldrich). and phenylthiohydantoin derivatives were identified by high-pressure liquid chromatography (Waters Associates) as described (21 ). Peptide Fragments. Fe fragments of IgM were prepared according to the method of Shimizu et al. (22) and purified by gel filtration on ACA 22 and ACA 3-l. Cyanogen bromide peptides were succinylated (23) for trypsin digestion limited to arginine residues. Cyanogen bromide peptides also were cleaved at tryptophan residues by the procedure of Ozols et al. (2-l) Smaller peptides were produced from cyanogen bromide fragments by digestion with trypsin , chymotrypsin. and thermolysin . The resulting peptides were separated in two dimensions on paper by chromatography and high-voltage electrophoresis (25 ) and where appropriate their amino acid sequences were determined in the automated sequenator in the presence of Polybrene (see ref. 21 ). the methionine residues at positions 20 (CNl-2 ) and 568 (CN8-9) were isolated. Thus. cyanogen bromide fragments covering the entireµ chain have been obtained and their sequences have been completelv determined. The 10 methionine fragments of the mouseµ chain can be unambiguously aligned in a linear order without resorting to homology comparisons with the human µ chain. Sequence analysis of the NH2-terminal 38 residues of the intactµ chain assigned cyanogen bromide fragments 1, 2, and 3. respectively, to the NH2 terminus (Fig. I ). An Fe fragment , obtained by trypsin digestion of the intact IgM in 5 M urea (22). shows that the COOH-terminal sequence of CN5 and the NH2-terminal sequence of CN6 are contiguous (Fig. 1). In addition . a cvanogen bromide digest of the Fe fragment yielded CN6, CN7. and CN8-9. Because CN-l is not in the Fe fragment. it must reside betwet>n CN-3 and C\'5 ( Fig. 1). 8e{-ause CN8-9 does not have a COOH-terminal homoserine residue it must be the COOH-terminal fragment. The onlv remaining fragment. CN7 , is then located between CN6 and CN8-9 (Fig. 1). These assignments are supported by the previous isolation and sequence analvsis of pepsin peptide fragments overlapping C:'li-l to CN5 and c:,..;, to CN8 (27 ). Thus direct sequence overlaps are available for all methionine residues except those between CN6-CN7 and CN.3-CN-t Then' is the possibilitv that additional methionine fragments between CN3-CN-l ,ind CNo-C\'7 ma,· have het'n lost in the isolation procedures. This possibility appears unlikek bt'<'ause the mouseµ chain in these two regions can be aligned without sequence gaps against a completeh sequenced humanµ chain \ Fig. 2). Disulfide Bridges. The positions of the 1-l cysteine residues of the mouseµ chain are indicated in Fig. 1. The assignments of two disulfide bridges (residues 22 to 97; residues 15:3 to 213 ) have been established through the isolation of c,a noge n bromide peptides from µ chain with intact intrachain bridges followed bv analysis on reducing and nonreducing 18~ polyacrvlamide gels. In addition. the methionine fragments of the Fe region, C:--:6, C:--:7 . and CN8. are linked by intrachain disulfide bridges. All 1-l cvsteint> residues in tht> mouseµ chain RESULTS Sequence Strategy. The methionine residues of the \IOPC 10-lE µ chain were cleaved with cyanogen bromide and 10 peptides were isolated by gel filtration. The amino acid seciuences of these peptides were determined at their NH 2 termini on the automatic sequenator. The sequences of selected peptides produced bv trvptophan . thermolysin. trypsin . and chvmotrypsin cleavages also were determined to complete the primarv structure of each cyanogen bromide fragment. Fig. I schematically illustrates the order of the cyanogen bromide fragments. Onlv fragment CN 1 was not isolated. Two cvanogen bromide fragments that resulted from incomplete cleavage of SCALE _ __ __ _ __ _ __ __ _ __ __ _ ,._ 50 __._ CNI CN2 CN3 CNBr FRAGMENTS 20 14 47 TRYPSIN5M UREA Y/C ,oo _._ 150 .__ 200 _._ 250 _._ 350 _._ 300 __. 400 _._ 450 __. 500 _._ 550 ..__ 600 ___. CN4 CN5 CN6 CN7 CNS CN9 62 84 166 109 62 8 ~-----------------1----------------------~ Fe 357 c,. 117 455 w Mu CHAIN : w c,...2 I '-s- ---s- 0 • L Cho,n ♦ I' Cho,n l J Chain Fie;. t. Schematic drawing of theµ hea\'\' chain of lgM MOP(' I0~E showing Iii siles of dea,·age h~· CN Hr and !he respertive fra~ment s I( "NI to CN9), (ii) the point of clea\'age h~· tr~·psin in ,-, M urea gi\'ing the Fr fragment. 1iii I the point 11f di,·ision 11f the ,·aria hie t \'HI and ron,tant I( '"I regions f:l), Iit·) the intraehain disulfide hridges, le) the six hom11l11::::~· rt>::::i on:-. t \" H- C.. I tu(' "~. ( •.. terminal I. a nd lt ·t I the lrn:at ion~ of th e :-.ix 11l ig11sacc:harides (CHO 1 to CHO fH. The hil:{h -mannnse oligosaccharicfp:,,:. are cirded a nd «:omplex t~·pe ulig11sarrhari<lt-:,,; are boxed . Twc • 11li:,.:11sa<Tharides are enclosed in hroken line!- lo denotf::' that the assii<nmenl of romplex 1~· pe was made from hom11l11~:s with I he hum an µ 11l ig:o:-.Jl',·haride:,,; t~hL The linkages uf three of the intrarhain hridges art- indicate<l h\ hrnkl'n lint-s henwse I he~· ha\'t· het- n a:-.signed 11111\ h~· a:--sumi n)! homolog)· with the humanµ. se4uence Ou (II 1. \\' indicates tryptophan resicha•s. To facilitate hom11l11~~- l'11rnpari sun:-.. tht- :--calt- n,rre:-.pund.-. In t hr amino acid sequence posit inns of the human myeloma µ chain Ou t 111 . Hf'n1u:-.f' t ht-rf' arP a nurnht-r of ):!:t f>!-- and insert ion:,; hetv,.:een l httwo :,;t-4ue1H·es. the numbers indirating total amino aci d resi<lue:,,; in eal'h 111-t E µ chain fra ~mt- 1\1 du 11111 ,wct-s!'-aril~· corrt>spond to !ht· numht-r:-i 11litained h~· c.:ountin~ the amino ac id residues in the homoln;:!ous Ou st>q11t-ncr . 21 l'lv 1 Lt L HOPC 104£ Ou M"" 11 lilt ~ hv3 2 H 71 U IH IH ;;r tt 111 lN rcHOl C"t - - , - - IH I.. IH DSAVYYCARO; ]Y DWYFOVWGAGTTVTVSS!ESQAJ'Pt<V1'PLVSC!SPLSDKNLVAMGCLARDFLPSTI SFTWNYCJtl'r EV I QG I Rr -T-T--VVNSVMAGY-YY- -K--<rASA--TL---NSP-BST( }-v--z--os-T-S-K~SE( )-SST-(; -T---rEl ]GoIEIPR-GQGTI c., lt1 MOPC 104£ Ou lt8QJ ti n ,a PIOPC 104£ Ou Moo hv ti EVQl,QQSGP ELVK PCASV KMSCUSGYTPTDY nu! )t',KQSHGK SLEW I GO I NP.e:;cTs y NQK FKGK AT L TVDK SSST A YMQLNS L rs E -,!-T--.E-A--KQPL T-Tf'-PSI..STSR-RVS-1 RRPP-A-I.U-888( )'.)KFYws·rSLRTRLS I SKND-KNQVVL IM I SVNPV - K-VE-Go----G-LRL-V-P-SSNG-S( )--R-OP-ltG-Q-VA-SSS: rQ-Y-ADAV-RP S IS R-NAK N- L-L-f<ED- RV- 1to '" u, 111 c,,z 1ul ""' no uo Ht FPTLRTGGKY LATSQVLLSPKN I LeGSDEY LVCK I HYGGKNROLQVP I PAV Al!MNPNVNVfVP PRDGFSGPAPRKSK L IC EATNFTPK PI( J -SVLR--A-+Z>---PS-DVMO-T-HVCKWVQHPNGBKQK3-L-VI-ELP-K-S---3-F-a~s-RQVW M c.2 _ _c.3 111 n, ,.. ua no u, I .,, no no TV SWLKDGK LVESGFX-rDPVT I ENKGSTPQTYKV I STL TI SE I DWLNLIIVYT\.:RVDHRGL TF Ll<ilvSSTCAASPS ro I LNfT I PP Sf AO I fLS SLR~G""'1'/C, ;-v-az-ZAZA-Z-G-T--T--KZS----GESMF-----Qeb'-M-VPDQD-A-RV-A---S-T --O--DA-SQS-F-K-£---00-A-M-TS OQPVG-S 1 - - - S - N T 111 MOPC 104£ Ou Moo U L MOPC 104£ Ou Mo-:, "" MOPC 104£ Ou Moo ~!;!Qi PO )It 1H I 'c®; •OO •lt •20 lilll oO C.3_ .c.• \ lil)CI KS,3. TCLV SNLATY ETLT I SWASQ,;G£PL£TK I K MESHPia:;·rrSAKGVASVCV EOWNNRK £FVCTVTHROLPS PQK KP I SK PN ElVH KH -TK--·TO--T-BSV--TREENGAVK - H S - ---A--V-£- I-£OBOWSGER-T--T--L""'1T-R-KG-AL-K-S-TO---OSV --TR£ENGA-K-HTN-S--- - -!'!- E-T--E-ESG£Q-T---T--VL""'1T- R-KG-AV- '"" .. ,. .... '" 11' uo ,,, PPAVYLi.PPAREOLNLRESATVTC LVKGFSPAO I SVQWKQRGQLLPQEKYVTSAPMPoPGAPGfY fTHS IL rVT££EWNSGETYTCVVGH R-8---zz---I--T---Vf-£-M-EP-SP~R-A---S--T-{)---AM-S-V-S--O---LS-T-Y-P-VF-V-K-PV·PDS---S,-L-A---S--A----ACw4_ _ . _c-,e,mu•,al ~1 MOPC l04E ~~ u• 1 'C""QI ! "' . r ,,. £ALP£LVTERTVOKSTG!tPTLMSLIMSD'l'GGTCY =;::::--s t p ~z=:::;:: F1<: . ~- The amino acid sequences ofµ. chains from mouse MOPC' 104r,, hum an Ou ( 111. a nd canine Moo ( 12. !Ji. iThe one-let!er code for amino acids is given in ref. 28.1 Only the NH -, -terminal and Fr sequences of the dogµ. chain are availahle. A straight lin e indicates identitv with the MOP(' I04E seque nce. Deletions are indica ted hy I I-The sit es of carhohvdrate atUlchment are boxed. Positions of earhohydrate attachmen t haw not heen determined for the dogµ cha in. The homology unit houndaries have heen assigned by an analvsis of immunoglobulin three -d imensional structure as indicated in T ahle I (:Ji. Three hypervariable reg ions ( hv I. hv 2. and hv :n are indicated b~· arrows over appropriate\" region ,egments. Q' indicates pyrrolidone carhoxylic acid. are in positions homologous to the 14 cysteine residues of the humanµ chain. Thus we assume the same cvsteine residues will l:w involved in interchain bridges and intr~domain bridges as indicated in Fig. I. Accordingly. the cysteine residue at position 140 is probably bridged to the L chain and the cysteines at 3-'37. 414. and 575 most likely form bridges with otherµ chains or with the Jchain ( 16. 29) Oligosaccharide Assignments. The I0-IE µ chain has six sitt>S of carbohydrate attachment. five in the constant region and one in the variable region (Fig. I). All six oligosaccharides contain glucosamine and are attached to asparagine residues. The sites of carhohvdrate attachment were identified as blank cvcles in automati~ sequenator runs. Subsequent enzvmatic limit- digests of the carhohvdrate-containing cyanogen bromide fragments produced small peptides that were analyzed for the presence of glucosamine. Four of the five C region oligosaccharides of the mouse µ chain are in positions identical to those of the human µ chain (Figs. I and 2). The fifth C region oligosaccharide is attached to a nonhomologous position 31 residues "-iH1-terminal to the site in Ou (positions 395 and 3tH in human and mouse. respectively). DISCUSSION The COOH-Terminal Portion of the C~ Region Appears More Highly Conserved Than the NHz-TerminalPortion. In Fig. 2 the complete sequence of the µ chains of mouse and human are listed along with the V region and Fe sequence data from the dogµ chain. Several insertions, deletions. and sequence inversions, some of which mav be technical artifacts in analvsis (2H). are noted in comparing these sequences. Using the crit~ria of !~ale and Feinstein (3 ). we have identified in the mouse c •. as in its human counterpart. four homology regions of ap- proximately 110 resi dues and an additional COOH-terminal region 19 residues in length (Fig. 2 ). The mouseµ chain. like the humanµ chain. has no hinge region separating any of the C region domains (3). Two amino acid residues. cysteine and tryptophan. are highhconSt"rved in !he C. regions (Figs. I and 2) All 12 cysteine residues are conserved in the mouse and human C. regions. as are the /l that can be compared in the dog sequence. The mouse c. regi on has six tryptophan residues. each of which is homologous to one of the seven tryptophan residues in the human C. region (Fig. 2) . All five tryptophans in the partial sequence of the dog c. region are in the same positions as their mouse and human counterparts. In contrast. the human and mouse c. regions have six and five methionine residues. respective!\·. only two of which are conserved. Because of their conservation, it is reasonable to conclude that cysteine and tryptophan residues play an important role in the domain structure of immunoglobulins. Table I is a compilation of the homologv comparisons between the various homology regions of mouse. human. and dog. There is a striking gradient of homology from the NHz-terminal to the COOH-terminal region . The C"l homology unit of mouse shows -t8% homology with humanµ as contrasted with 78% and 89% homologies in the C.4 and the COOH-terminal regions. respectively. The existence of sequence gaps in the NH 2-terminal homology regions (two in c.1 and three in C.2) but not in the COOH-terminal homology regions of theµ chain (Fig. 2) suggests that the NH2-terminal regions have diverged more. Thus. the COOH-terminal portion of theµ chain is significantlv more conserved than the NH2-terminal portion . If individual domains do earn out separate functions. these observations imph that there are strong selective constraints on 22 Table l. Comparison Homology matrix of the constant region domains of mouse. human, and dogµ chains V C-Terminus Mouse 104E vs. Human Ou Dog Moo vs. Human Ou Mouse 104E vs. Dog Moo Theµ chains were divided into four homoloi:y units and a COOH-terminal rei:ion according Lo an analysis of the threedimensional structures of several immunoglobulins not of theµ class: CH I = residues 127-2:12: CH2 = residues 2:l:1-:1:19; CH;l = residues ~40-446; CH4 = residues 44i -.'i.','i; and C terminus = residues 5.'>8-5i6 (:l) . Sequence gaps (deletions or insertions I are not counted in homology comparisons. In Lhe lower right of the square for each region and comparison is listed the number of amino acids compared in the homology calculation divided by the total number of amino acids in that homology unit. The number of sequence identities in each unit was used I.<> calculate the percent homology, which is given in the upper left of each square. the more COOH-terminal domains. These constraints mav be related to the requirements for the polymerization of penta~ers in the secreted IgM molecule. the structural requirements imposed on an integral membrane receptor with regard to membrane attachment and the triggering of differentiation. or the activation of the complement pathway, which has been localized to theC.4 domain (10). Sites of Carbohydrate Attachment to the C,. Region Are Similar in Human and Mouseµ Chains. Theµ chain of mouse has five sites of carbohydrate attachment in the constant region (Fig. I ). Four of these five carbohydrate attachment sites :ire at identical positions in the mouse and human (26) C" regions (Fig. 2). The carbohydrate moiety on residue 171 is of the complex type and those on residues 402 and 563 are of the high-mannose type (W. J. Grimes. personal communication). in agreement with the carbohydrate types on the human µ chain (26 ). Compositional analyses on the remaining two oligosaccharides are not yet complete. Each carbohydrate moiety is attached to the asparagine residue of a three-residue recognition sequence. -Asn-X-Ser / Thr. in which X may be any amino acid and Ser / Thr indicates either serine or threonine. Substitution of a methionine for a serine at position 397 in mouseµ has eliminated the recognition sequence for the carbohydrate attachment site at position 395 in human µ. In the mouse µ chain we find a carbohydrate moiety at residue 36-1, where there exists a carbohydrate recognition sequence that is absent in the dog and human µ chains. Other recognition sequences that are apparently not glycosylated also occur in theµ chain (74-76 in human Ou, 263-265 and 347-349 in mouse MOPC 104E). Thus, the elimination and formation of an attachment site implies either that this region of the protein is on the exterior of the molecule and is therefore accessible to glycosylating enzymes or that a carbohydrate moiety is required (structurally, functionally , or both) in this region of theµ chain. The carbohydrate moieties of human and mouse µ chains are present in the C" 1, c.2. and C"3 domains as well as the COOH-terminal region (Figs. 1 and 2). A variety of functions have been suggested for the carbohydrate on heavy chains. including the solubili?.ation of the heavy chain, the facilitation of the secretory process. the promotion of the assembly of pentamers. and the contribution of structural features for a variety of specific effector functions (26. 30-34). Whatever their function. the conservation of the locations of carbohvdrate attachment during the 75 million years that mouse a~d human C" genes have diverged from one another suggests that strong selective forces are maintaining the recognition sequences. The V Domain of MOPC 104E, an a-(1-3}-Dextran Binder. Has a Carbohydrate Moiety in Its Antigen-Binding Site. A carbohydrate moiety has been identified at asparagine 57, which lies in the middle of the second hypervariable region of the MOPC I 0-IE µ chain. There are several unusual features of the V region carbohydrate site. First. the usual recognition sequence is missing: this high-mannose carbohydrate moiety is attached to the asparagine of an Asn-Gly -Gly-Thr sequence. Although the majority of asparagine-linked carbohydrate moieties are associated with a recognition sequence, the sequence is clearly not an absolute requirement for the attachment of carbohydrate to asparagine residues (35). Second, the IgM molecule secreted by the MOPC 104E tumor binds a simple carbohydrate, a-(1-3)-dextran (36). Thus, we have localized a carbohydrate moiety to the antigen-binding site of the heavy chain of an immunoglobulin that binds a defined hapten. This observation raises the possibility that carbohydrate-carbohydrate interactions may constitute a portion of the binding energy for the hapten, a-(1-3)-dextran. In view of what we know about the specificities of protein and carbohydrate interactions. however. this possibility appears unlikely (37). Third, a second myeloma protein, J558(a. ,\), which also binds a-( 1-3)-dextran, appears to have a carbohydrate moiety at precisely the same position (unpublished observations). It is likely that this carbohydrate moiety also occurs in the heavy chains of normallv induced serum antibodies to a-( 1-3)dextran (unpublished observations). The role, if any, for carbohydrate in the V domains of immunoglobulins remains an interesting, but unanswered question. 23 Structure of the µ, Chain Places Constraints on the lgM Molecule as an Integral Membrane Receptor. The monomeric lgM molecule is an integral membrane protein that serves as a cell-surface receptor for antigen on B cells that ca n tri gger the B cells to differentiate into antibody-producing cells (4). The membrane (m ) and secreted (s) IgM molecules are serologically crossreactive, implying that they share structural features or may even be identical. Conflicting data suggest both that the COOH terminal portion of the JJ-m chain is identical toµ , (38) and that these regions are different (39). If the /Jm andµ,, chains are identical in amino acid sequence, the /Jm chain cannot be transmembrane in its orientation. The cv-helix is the most stable polypeptide secondary structure within a membrane (40. 41 ). This configuration requires about 21 uncharged residues to span the membrane. Alternatively . the extended# configuration requires two paired stretches of 9 uncharged amino acids to span the membrane. We have compared the three mammalianµ chains over their COOHterminal two homology units. which presumabl y contain the site of membrane attachment. The longest sequence of uncharged amino acids in any of these molecules is 14 residues in length (positions 469-482 in the human and dogµ chains ) (Fig. 2). Howevt>r, the mouse µ chain has a charged lys ine residue at position 4i7, reducing the stretch of uncharged amino acids in this region to 8 residues. In addition, this regi on contains a cysteint> at position 4i4 that is presumably joined to cysteint>536 to form the C"4 intradomain disulfide brid ge. If one assumes that all mammals will attach theirµ chains to membranf'S in a similar fashion. the re is no uncharged sequence shared by tht> COOH-terminal regions of tht> mouse. human. or dogµ chains that could span the membrane in either the n- helical or the extended unpaired {:1 configuration (41. 42). We conclude that. if the /Jm and µ , chains are identical, the IgM molecule cannot be a transmembrane protein that meets the abovt> specifications. Moreove r. tyrosine- labeling experiments on inside-out vesicles are consistent with the supposition that the IgM molecule is not transmembrane (43 ). If the /Jm chain is not a transmembrane polypeptide. then the IgM molt>cule may associate with the membrane by virtue of a specific receptor molecule. an association analogous to the interaction of lgE with its membrane receptor on mast cells (4-l). In this regard . both IgM and IgE have an additional constant region domain when compared with lgG and lgA molecules (3). We have preliminary evidence that suggests the /Jm andµ , chains differ in their primary amino acid sequences. Any primary structural differences could be generated by mR NA splicing mechanisms, posttranslational proteolysis. or the occurrence of duplicatedµ chains genes. It is possible to determine the chemical structure of /Jm and dt>fine precisely the relationship between /Jm andµ, chains. Moreover, we have isolated a genomic µ clone. whose nucleotide sequence can be determined. Such studies should allow us to distinguish unambiguously among the three models for /Jm and µ , expression cited above. We thank V. Farnsworth and M. W Hunkapiller for patiently tt"aching techniques of peptide and amino acid St>quence analyses. This work has been supported by National Institutes of Health Grants Al10i8 1 and CA-20314 and by a Gordon Ross Medical Foundation Fellowship to M .K. Marchalonis. J. J. (19i2) Nature (London) 236, 84-86. Lawton. A. R , Kincade. P. W. & C'..opper. M. D. ( 19i5) Fed. Proc. Fed . Am. Soc. E:rp. Biol. 34, 33-39. 3. Beale. D. & Feinstein . A. (1976) Q. Rev. Biophys. 9, 135-1 80. 4. Warner . N. L. (1974 ) Adv. lmmunol. 19, 6i-216 I 2. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21 . 22. 23. 24 . :25. 26. 27. 28. 29. 30. 31. 32. J.'J. 34. 35. Jo. 37 38. 39 40. 41. 42. 43. 44. Haunstein. D.. Marchalonis. J. I. & Cru mpton. M. J. (197 4) Nature (London ) 252, 602-604 . Melcher. U.. Eidels. L. & Uhr. ). W. (19i5)Nature(London)258, 434-435. Vitella. E. S . Baur. S & Uhr. J. W. ( I 9i I)}. E:rp . Med . 134, 242-264. Kennel. S. J. & Lerner. R. A. ( 1973) ]. Mo/ . Biol. 76. 485-502. Mesteckv . J.. Zikan. J. & Butler. W. T. (1971) Science 171, 1163-1170. Hurst. M. M.. Volanakis. J. E., Stroud. R. M. & Bennett. J. C. (1975 ) ]. E:rp. Med. 142, 1322-1326. Putnam. F. W .. Florenl. G.. Paul. C.. Shinoda. T. & Shimizu. A. (1973)Science 182,287-291 . Wasserman. K. L. & Capra . J. D. (1978) Science 200, I 1591161. WaSS<'rman . R. L. & Capra . J. D. (19ii ) Biochem istry 16, 3 Ifi0-3 lfi8. Konigsberg. W. (l 9i2) Methods En:ymol. 25. 185-1 88. Appella. E. (197 1l Proc. Natl. Acad. Sci. USA 68, 590-594. Weigert. M. G.. C',esa ri. I. M.. Yonkovch. S. J. & Coh n. M. (1970) Nature (London ) 228, 1045-1047. Gross. E. (1967 ) Methods En:ymol. II , 2.'38-255. Laemmli . U K. (19i0) Nature (London) 277, 680-68.5 Wittmann-Liebold. 8. (1973) Hoppe-Seylers Z. Physiol. Chem . 354 , 1415-143 1. Wittmann-Liebod. 8.. Graff under. H. & Kohls, H. (1976 ) Anal . Biochl'm . 75, 621-6.'l.'J. Hunkapiller. M. W. & Hood . L. E. (1978) Biochemistry 17. 2124-213.'J Shimizu. A.. Watanabe. S.. Yamamura . Y. & Putnam. F. W. (19i 4) lmmunochcmistry II, il9-727. Klapper. M. H. & Klotz. I. M . ( 19721 Methods E11zymol. 25, 531-536. Ozols. J.. Gerard. C. & Stachelek . C. (1977 ) }. Biol. Chem . 252, 5986-5989 Katz. A. M.. Drever. W J & :\nfinsen. C. B. ( 1959)}. Biol. Chem . 234, 2897 -2900. Shimizu. A.. Putnam. F. W. & Paul. C ( I9i I) Nature (London ) 231. 73- 76 Milstein. C. P.. Richardson. N. E.. Deverson. E. V. & Feinstein. A. (1975 ) Biochem . J. 151. 615-624 • Da yhoff.MO ( 1976) :\tlas of Prote,n Sequence and Structure (National BiomedicaLResearch Foundation. Silver Spring, MD ). Vol. 5. pp. 189-190 Mt>St('('ky. J. & Schrohenloher. R. E. (1974 ) Nat ure (London ) 249. 6.50-652. Mekhers. F. (19i2 ) Biochemistry II. :2:204-2208. Anderson. J.. Lafleur. L. & Melchers. F. (1974) Eur I lmmunol. 4, 170-1 80. Parkhouse. R M. E. & Melchers. F. ( 19il ) Biochem . }. 125. 235-240. Kornfeld . R. & Kornfeld. S. (19i6) Annu. Rec Biochem. 45, 217-23i. Hick man . S. & Kornfeld. S ( 1978) }. lmmunol. 121. 990-996. Marshall. R. D. (19i:2 ) Annu . Rev. Bioch,m, . 41. 673-702. Leon. M. A.. Young. N. M. & McIntire. K. R (1970) Biochemistry 9, I0'23- I030 Reeke. G. N.. Jr. . Becker. J. W & Edelman. G. M. (1978 ) Proc. Natl. Acad. Sci . USA 75, 2286-:2290. Mcilhinney. R. A. J.. Richardson . N. E. & Feinstein, A. (1978 ) Nat ure (London ) 272, 555-557 Williams. P B.. Kubo. R. T. & Grev. H. M. (1978) J. lmmunol. 121. 2435-2439. Guidotti. G. (1977)}. Supramol. Struct . 7, 489-497. Tanford . C & Reynolds. J. A. (1976) Biochim . Bwphys. Acta 457, 133-170. Tomita. D. 8. & Marchesi . V. T. (1976) Proc. Natl. Acad. Sci. USA 72,2964-2968 Walsh. F. S. & Crumpton. M . J. (1977 ) Nature (London ) 269. 30i-31 I. Metzger, H. (1978 ) Transplant. Rec. 41, 186-199 24 CHAPTER 2 This paper will be submitted to Biochemistry. 25 Complete Amino Acid Sequence of a Mouse Mu Chain: Homology among Heavy Chain Constant Region Domains t Marilyn R. Kehry, Jonathan S. Fuhrman, James Schilling, John Rogers, Tim Hunkapiller, William Grimes, Carol H. Sibley, and Leroy E. Hood * t From the Division of Biology, California Institute of Technology, Pasadena, California 91125 (M.K., J.F., J.S., T.H. and L.H.), Department of Genetics, University of Washington, Seattle, Washington 98195 (C.S.), Department of Microbiology and Immunology, UCLA School of Medicine, Los Angeles, California 90024 (J.R.), and Department of Biochemistry, University of Arizona Health Sciences Center, Tucson, Arizona 85724 (W.G.). 26 Footnotes 1 Abbreviations used: L, light chain; H, heavy chain; 104E, MOPC 104E myeloma; BBS, borate buffered saline; EDTA, ethylenediamine tetraacetic acid; SDS, sodium dodecyl sulfate; TPCK, tosylphenylchloroketone; Fth, phenylthiohydantoin; VH' immunoglobulin heavy chain variable region; CH' immunoglobulin heavy chain constant region; J, joining chain. 27 ABSTRACT: The complete amino acid sequence of the mouse µ chain secreted by the MOPC 104E myeloma tumor has been determined. There are four constant region domains in the µ chain which exhibit internal amino acid sequence homology and a 20-residue C-terminal segment which plays a role in the polymerization of pentameric IgM molecules. There are six sites of carbohydrate attachment in the MOPC 104E µ chain. Three complex type and two high mannose oligosaccharides are located in the µ chain constant region. In addition, a unique small carbohydrate structure is attached to an asparagine residue in the MOPC 104E second hypervariable region. The general type and location of carbohydrate moieties in the µ chain constant region is completely conserved between mouse and human µ chains. Homology in the location of carbohydrate structures on different classes of heavy chains is discussed. The amino acid sequence homologies among individual domains in the µ, y, e: and a heavy chains reveal that some interesting genetic events have occurred during the evolution of constant region domains. 28 Immunoglobulin molecules are composed of two polypeptide chains, light 1 (L) and heavy (H) chains. The five major classes of immunoglobulins, IgM, IgG, IgA, IgD, and IgE, are distinguished by the structures of their corresponding heavy chains, µ, y, a, c, and e:, respectively. Each heavy chain may be divided into a variable (V ) or antigen binding region and a constant (c ) region which performs 8 8 the various effector functions of the immunoglobulin molecule. Heavy chain constant regions may be further subdivided into homology regions or domains approximately 110 residues in length (Beale &: Feinstein, 1976). The µ chain has four c 8 domains which are designated C 1, C 2, C 3, and C 4 from the NH - to the COOR-terminus. µ µ µ µ 2 There is an additional C-terminal segment in the µ chain which is composed of 20 amino acid residues (Calame et al., 1980). The basic unit of the immunoglobulin molecule consists of two Hand two L chains (H L ) which are covalently linked together through disulfide bridges. 2 2 In this tertiary structure, pairs of homology regions fold together to form discrete functional globular domains. IgM molecules are unique among the different classes of immunoglobulins in that they exist in two different polymeric forms which perform different effector functions in the immune system. First, the IgM molecule is a cell-surface receptor which functions to trigger the differentiation of precursor B lymphocytes to antibody-producing plasma cells (Warner, 1974). This membranebound form of the IgM molecule is a monomeric ( µ 1 ) immunoglobulin embedded 2 2 in the plasma membrane of B lymphocytes (Vitetta et al., 1971). Second, antibodyproducing plasma cells secrete IgM molecules which are assembled into pentamers ( µ L \ by the addition of a joining (J) chain (Della Corte &: Parkhouse, 1973). 2 2 These soluble IgM pentamers then circulate in the bloodstream where they function to bind and neutralize foreign antigens and to fix complement. The µ chain polypeptides which compose the membrane-bound and secreted forms of the IgM molecule are known to differ in the amino acid sequence of their C-terminal segments (Kehry 29 et al., 1980; Early et al., 1980; Rogers et al., 1980). IgM molecules are also unique in being the first immunoglobulins to appear in vertebrate evolution (Marchalonis, 1972) and the first to appear during the development of the immune system (Lawton et al., 1975). The above features make the IgM molecule an interesting protein to structurally characterize and to use for investigating the evolution of immuneglobulin heavy chain classes. We present here a complete structural analysis of a secreted µ chain from the mouse myeloma tumor MOPC 104E. DNA sequencing studies on a cloned mouse µ chain gene complement the protein sequence by delineating the exact boundaries between C µ region domains (Calame et al., 1980). We discuss the composition of the six oligosaccharide moieties attached to the MOPC 104E µ chain and the homologies in the sites and types of carbohydrate attached to the different classes of heavy chains. We also give an analysis of an amino acid sequence homology comparison among the CH domains of µ, y, a and e: chains. The C-terminal segments of secreted µ and a chains function specifically in the formation of IgM and IgA polymers. Some interesting features in the evolution of CH region domains from a primordial immunoglobulin molecule have been found. Experimental Procedures Isolation of MOPC 104E µ Chain. The IgM-secreting myeloma MOPC 104E ( µ, >,.), subsequently abbreviated 104E, was obtained from Dr. M. Potter and was passaged subcutaneously in (BALB/c x DBA/2)F 1 mice. Ascites fluid was collected from mice injected intraperitoneally with 104E cells, centrifuged to remove cells, and the supernatant fluid was used for purification of the IgM immunoglobulin. Proteins were precipitated from ascites fluid (40 ml) by two cycles of treatment with 50% saturated ammonium sulfate at 4°C. The final preci_pitate was dissolved in 0.15 M sodium borate, 0.14 M NaCl, 0.02% NaN (BBS) and the pentameric 3 30 IgM molecules were purified by gel filtration on a column (5 x 100 cm) of ACA 22 (LKB) equilibrated in BBS. IgM molecules were eluted just after the void volume and the disulfide bridges were completely reduced and alkylated as follows. Appropriate fractions were pooled, precipitated with 50% saturated ammonium sulfate at 4°C, and the precipitate was dissolved in 6 M guanidine-HCl, 0.25 M Tris-HCl pH 8.5, 0.14 M EDTA. Reduction and alkylation were carried out essentially as described (Konigsberg, 1972) using 0.01 M dithiothreitol and 0.025 M iodoacetamide (3X recrystallized). When iodo(1- 14 c)-acetamide was employed for alkylation, the reaction was allowed to proceed for 30 min followed by the addition of nonradioactive iodoacetamide to 0.025 M. After dilution of the guanidine concentration to 4 M by the addition of 0.4 M ammonium bicarbonate, heavy and light chains were separated by gel filtration on a column (5 x 100 cm) of ACA 34 (LKB) equilibrated in 3 M guanidine-HCl, 0.2 M ammonium bicarbonate, 0.02% NaN 3. Isolation of Cyanogen Bromide Fragments. Purified µ chains in 3 M guanidine HCl, 0.2 M ammonium bicarbonate were concentrated (Millipore Molecular Separator), acidified by gradual addition of 88% formic acid, and desalted on a column (2.5 x 40 cm) of Bio,..Gel P-2 in 50% formic acid at 4°C. Pooled µ chains were concentrated by partial lyophilization, the concentration of formic acid was adjusted to 70%, and the protein was cleaved by cyanogen bromide (50 mg/ml) in the dark with constant stirring at 4°C (Gross, 1967). After 20-22 h, the reaction mixture was diluted with 11 volumes of water and lyophilized. The peptides were dissolved sequentially in 8 M guanidine-HCl, 3M guanidine-HCl and 0.2 M ammonium bicarbonate, and finally 0.4 M ammonium bicarbonate. The solutions were combined, and the cyanogen bromide peptides were separated by gel filtration on a column (3.5 x 140 cm) of ACA 54 (LKB) equilibrated in 3 M guanidine-HCl, 0.2 M ammonium bicarbonate, 0.02% NaN . Fractions were monitored by reading the absorbance 3 at 280 nm, and aliquots were counted for radioactivity when the cysteine residues 31 were 14 C-alkylated. The cyanogen bromide fragments were found to be of sufficient purity for sequence analyses after the one-step gel filtration on the ACA 54 column (see Figure 1). For direct sequence determinations, pooled fragments were either neutralized and desalted on a column (2.5 x 40 cm) of Bio-Gel P-2 in 20% formic acid or dialyzed (Spectra/Par 3) against three changes of 5% formic acid at 4°C followed by lyophilization. Polyacrylamide Gel Electrophoresis. SDS-polyacrylamide gel electrophoresis as described by Laemmli (1970) was routinely used to monitor the purification of IgM and µ chains (10% polyacrylamide gels) and cyanogen bromide fragments (18% polyacrylamide gels). Guanidine was removed from the samples prior to electrophoresis by dialysis (Spectra/Par 3) against 5% formic acid followed by lyophilization. Gels were stained with 0.025% Coomassie Brilliant Blue in 25% isopropanol, 10% acetic acid and destained in 10% acetic acid. Composite gels (1 % agarose and 2.5% polyacrylamide) were run by a modification of the procedure of Peacock and Dingman (1968). Gels contained 0.5 M urea, 0.2 M sodium phosphate, pH 7.2, and 0.1 % SDS. Running buffer contained 0.5 M urea, 0.1 % SDS, 0. 75 .M sodium phosphate, pH 7.2 and sample buffer was composed of 2% SDS, 10% glycerol, 0.05 M sodium phosphate, pH 6.8. Gels were 10 cm long and were electrophoresed at 3.5 ma/gel for 16 h, stained for 1 h with 0.25% Coomassie Brilliant Blue in methanol:acetic acid:water, 5:1:5, and destained in the same solution without Coomassie Brilliant Blue, followed by equiliberation in 10% acetic acid. Tryptic, Chymotryptic and Thermolysin Digestion. Cyanogen bromide fragments were dialyzed (Spectra/Par 3) exhaustively against 0.2 M ammonium bicarbonate, lyophilized and dissolved in a small volume of 0.2 M ammonium bicarbonate. TPCK-trypsin (Worthington) or chymotrypsin (Sigma) was added in 0.001 N HCl or thermolysin (Sigma) was added in 0.2 M ammonium bicarbonate 32 (freshly prepared enzyme stock) at an enzyme:substrate ratio of 1:100. Enzyme addition was performed three times at 1-h intervals. Incubation was carried out at 37°C for a total period of 2.5 h, and the digestion was terminated by lyophilization. Peptides were separated in two dimensions by chromatography followed by high voltage electrophoresis on 3 MM paper (Whatman) as described previously (Katz et al.: 1959). Ninhydrin positive spots on analytical and preparative fingerprints were excised, eluted by two extractions into 0.5 M NH OH and lyophilized. 4 Cleavage at Arginine or Tryptophan Residues. For specific cleavage at arginine residues, cyanogen bromide fragments were dialyzed exhaustively (Spectra/Per 3) against 5% formic acid at 4°C, lyophilized and dissolved in 12 ml 6 M guanidine-HCl, 0.01 M Tris pH 9.5. Amino groups on the protein were succinylated by gradual addition of finely powdered succinic anhydride (40-fold excess by weight over protein) in a pH-stat (Klapper & Klotz, 1972). When reversible blockage of lysine residues was desired, the protein was reacted with citroconic anhydride or maleic anhydride under similar conditions. In some instances the protein was radiolabeled by reacting with 14 c-succinic anhydride. Reagents were removed by exhaustive dialysis against 0.1 M NH 4OH, followed by lyophilization. Trypsin digestion was performed as described in the previous section. In some cases, peptides were separated on a column of Sephadex G-75 or G-50 (Sigma) equilibrated in 0.2 M ammonium bicarbonate or ACA 54 (1KB) equilibrated in 3 M guanidine, 0.2 M ammonium bicarbonate, 0.02% NaN . The absorbance at 3 280 nm and, when appropriate, the absorbance at 230 nm were monitored. The maleylation or citraconylation was reversed by acidifying (pH 3.5) the protein overnight. Trypsin cleavage at lysine residues was then performed as described in the previous section. For specific cleavage at tryptophan residues (Ozols et al., 1977), cyanogen bromide fragments were dialyzed exhaustively (Spectro/Por 3) against 5% formic 33 acid at 4°C, succinylated, lyophilized and dissolved in equal volumes of 88% formic acid and anhydrous heptafluorobutyric acid (Pierce). Cyanogen bromide was added to 350 mg/ml. The reaction vessel was vented into water in a fume hood, and the tryptophan cleavage reaction performed essentially as described by Ozols et al. (1977). After 24 h, the reagents were evaporated under a stream of N 2 and the peptides lyophilized after addition of 10 ml water. In some cases, peptides were then separated on a column of Sephadex G-75 or G-50 (Sigma) equilibrated in 0.2 M ammonium bicarbonate. The tryptophan cleavage reaction also modified tyrosine residues in an unknown fashion which greatly altered the chromatographic behavior of the modified phenylthiohydantoin (Pth)-tyrosine derivative obtained by sequence analysis. Fe Digestion. IgM Fe fragments were produced by a modification of the method of Shimizu et al. (1975). Whole IgM in BBS was concentrated, desalted on a column (3 x 15 cm) of Sephadex G-25 (Sigma) equilibrated in freshly prepared 5 M urea, 0.1 M Tris pH 8, 0.15 M NaCl, 0.01 % NaN . After concentration of 3 the desalted IgM molecules, a stock solution of 2 mg/ml TPCK-trypsin (Worthington) in 0.1 M Tris-HCl, pH 8.0, 0.1 M Cac1 , 0.15 M NaCl was added to give an enzyme: 2 substrate ratio of 1:100 and incubated at 25°C for 18 h. The efficiency of pentameric Fe production was monitored by removing aliquots at specific times after addition of enzyme. Proteins were then electrophoresed without reduction on composite 1% agarose, 2.5% polyacrylamide-SDS gels (Peacock &: Dingman, 1968) or the proteins were reduced and analyzed on 10% polyacrylamide-SDS gels (Laemmli, 1970). For preparative isolation of Fe fragments, lima bean trypsin inhibitor (LBI) (LBI:trypsin, 1:3, w /w) was added to inhibit further digestion. The mixture was dialyzed against BBS at 4°C to remove urea, and the pentameric Fe fragments were purified on a column of ACA 22 (1KB) equilibrated in BBS. The µ chain Fe fragments were obtained from the pentameric Fe fragments after complete 34 reduction and alkylation of disulfide bridges followed by purification on a column of ACA 34 (LKB) equilibrated in 3 M guanidine-HCl, 0.2 M ammonium bicarbonate, 0.02% NaN . 3 Mild Acid Cleavage. For cleavage predominantly at aspartic acid-proline peptide bonds, completely reduced and alkylated 104E µ chain was dissolved in 10% acetic acid and 7 M guanidine-HCl, adjusted to pH 2.5 with pyridine and incubated at 45°C for 108 h with occasional mixing. The resulting fragments were dialyzed against 3 M guanidine-HCl, 0.2 M ammonium bicarbonate at 4°C and separated by gel filtration on a column (2.5 x 100 cm) of ACA 34 (LKB) equilibrated in 3 M guanidine-HCl, 0.2 M ammonium bicarbonate. Amino Acid Analysis. Peptide samples were flushed with N and evacuated 2 three times, sealed under vacuum and hydrolyzed at 100°C for 20 h in constant boiling 6 N HCl. When quantitation of glucosamine was required, hydrolysis was for 3 h. Amino acid analysis was performed on a Durrum D-500 analyzer. Carbohydrate Determinations. Individual cyanogen bromide fragments were rechromatographed on Bio-Gel P-6 in 0.1 N acetic acid to remove galactose and glucose contamination. Alditol derivatives were separated and quantitated by gas chromatography (Kim et al., 1967). Sequence Determination. All large fragments and peptides were sequenced by loading on a modified Beckman sequenator with automatic Pth-conversion using trifluoroacetic acid (Hunkapiller & Hood, 1978; Wittmann-Liebold, 1973; Wittmann-Liebold et al., 1976). Samples were loaded into a polybrene-coated cup (containing 20 nmole glycylglycine and pre-run for 5-7 cycles) in trifluoroacetic acid (Pierce) and 10-20% water. The CN3 fragment was loaded in dilute NH OH and small peptides generated by enzymatic cleavages were dissolved in 4 water. The first step was C(?Upled twice. Pth-amino acid derivatives were identified by high performance liquid chromatography (Waters Associates, DuPont) as described 35 (Hunkapiller & Hood, 1978; Johnson et al., 1979). DNA sequences were determined by the method of Maxam and Gilbert (1977), using the G (alternative), A> C, T + C, and C reactions as described (Rogers et al., 1980). Computer Comparison of CH Amino Acid Sequences. Homology comparisons of individual CH domains were performed for the following heavy chains: MOPC 104E µ (Kehry et al., 1979; this paper), GAL µ (Watanabe et al., 1973), OU µ (Putnam et al., 1973), MOO µ (Wasserman & Capra, 1978; Mccumber et al., 1979), ND e: (Bennich & Von Bahr-Lindstrom, 1974), MOPC 21 y (Adetugbo et al., 1977), EU 1 y (Edelman et al., 1969), NIE y (Ponstingl & Hilschmann, 1972), rabbit y (Fruchter 1 1 et al., 1970), guinea pig y (Cebra et al., 1977), MOPC 173 y a (Fougereau et al., 1 2 1976), MPCll y b (Tucker et al., 1979a,b), guinea pig y (Trischmann & Cebra, 2 2 1974; Cebra et al., 1977), VIN y (Pink et al., 1970), ZUC y (Wolfenstein-Todel 4 3 et al., 1976), HER y (Michaelsen et al., 1977), TRO a (Kratzin et al., 1978), 1 3 BUR a (Low et al., 1976), BUT a2-2 (Torano & Putnam, 1978), LAN a2-1 (Tsuzukida, 1 1979), MOPC 47 A a (Robinson & Appella, 1977), MOPC 315 a and MOC 511 a (Robinson & Appella, 1977). Domains were defined on the basis of intron locations in the µ and y chain genes (Calame et al., 1980; Tucker et al., 1979) and domain homology in the a and e: chains (Beale & Feinstein, 1976). Sequences were compared using a computer program as described (Loh et al., 1979). Results Sequence Strategy. Cyanogen bromide cleavage of the MOPC 104E (104E) µ chain produced nine peptides (CN1-CN9) which were separated by gel filtra- tion (Figure 1). Determination of the NH -terminal sequence of each peptide 2 ' showed that each peak gave a single homogeneous amino acid sequence. Therefore, the one-step gel filtration procedure illustrated in Figure 1 separates each 36 of the 104E cyanogen bromide peptides from one another. Amino terminal sequence homology of the 104E CH region cyanogen bromide peptides to the human µ chain sequence (Putnam et al., 1973) permitted determination of a presumptive alignment of peptides (top portion of Figure 1). This order was confirmed by additional sequence determinations which completed the sequence of each fragment and in all but two cases directly overlapped the fragments. Fragment CNl was not isolated, but since CNl is the NH -terminal VH region fragment, this sequence had been 2 previously determined on the intact µ chain (Barstad et al., 1978). Fragment CNl was found to migrate between fragments CNl-2 and CN2 on the ACA 54 3 column by chromatographing H-leucine and tyrosine labeled 104E µ chain cyanogen bromide fragments (not shown). Incomplete cleavage of the methionine-serine bonds at positions 20 and 568 by cyanogen bromide resulted in the isolation of two partial fragments, CNl-2 and CN8-9, respectively, as indicated in Figure 1. The complete amino acid sequence of the 104E µ chain is shown in Figure 2, along with the data used to determine the sequence. The amino acid and carbohydrate composition of the 104E µ chain and the cyanogen bromide fragments as derived from the amino &cid sequence is given in Table I. There are three amino acid ,differences from the previously published sequence (Kehry et al., 1979); position 130 (alanine to serine) was miscalled due to contaminating sequences and residues ' 233 (histidine to glutamic acid) and 545 (glutamic acid to histidine) were identified by using an improved analytical method. In addition, we have determined the type of oligosaccharide attached to each of the six carbohydrate sites in the 104E µ chain. Sequence of Cyanogen Bromide Fragments and Alignment. (a) Sequence of CNl and CN2. Amino terminal sequence analysis of the first 38 residues of the intact 104E µ chain (Barstad et al., 1978) established the order of fragments CNl, 2, and 3 (Figure 1). A sequenator run on the entire CNl-2 fragment verified 37 its identity as residues 1 to 34. The sequence of CN2 as residues 21 to 34 also was completely determined in a single sequenator run (Figure 2). (b) Sequence of CN3. The CN3 fragment extends from lysine ionine 82 34 to meth- and is 47 residues in length. [To facilitate homology comparisons, the numbering of the 104E µ chain corresponds to the amino acid sequence of the human µ chain OU (Putnam et al., 1973) and the total amino acid residues in each 104E µ chain fragment are not necessarily obtained by counting the assigned residue numbers (Kehry et al., 1979).] Quantitation of an NH -terminal sequence determination 2 on CN3 is shown in Figure 3. To complete the sequence, the CN3 fragment was digested with trypsin, and the resulting peptides were separated on paper by chromatography and high-voltage electrophoresis (Katz et al., 1959). Amino acid compositions of the isolated tryptic peptides are given in Table II. Sequence determination of the peptide CN3T7 provided the overlap data which established the sequence of CN3 (Figures 1, 2 and 3). The CNl and CN3 fragments are the only µ chain cyanogen bromide fragmerits which contain no cysteine residues. The second hypervariable region is located in CN3 and includes a carbohydrate moiety attached to asparagine 57 (see below). (c) Sequence of CN4. The CN4 fragment spans the variable-constant region junction from glutamine 83 to methionine 151 and is 62 residues in length. The sequence overlap between CN3 and CN4 has not been rigorously established, but because this variable region sequence can be aligned without sequence gaps with those of other heavy chains (Kabat et al., 1976), the CN3 and CN4 fragments are probably contiguous. The NH -terminal sequence of CN4 was determined 2 to the variable-constant region junction. The remaining sequence was obtained and confirmed by determining the sequences of the major peptide produced by tryptophan cleavage (Ozols et al., 1977), and the tryptic and thermolysin peptides CN4Th3, CN4Th7, CN4Th9, CN4Th10, and CN4T8 (Figure 2). The one inserted 38 residue at the end of this fragment (position 148) also has been confirmed independently by Milstein et al. (1975) through an analysis of the corresponding cysteine-containing peptide. (d) Sequence of CN5. The CN5 fragment extends from glycine methionine 234 152 to and constitutes the major portion of the c 1 domain. This 848 residue fragment also contains the first complex carbohydrate moiety in the constant region. Determination of the NH -terminal sequence plus the sequences of the 2 two arginine fragments (CN5RI and CN5RII) and the fragment generated by cleavage at the tryptophan residue (CN5W) completed the sequence of the CN5 fragment (Figure 2). The Fe fragment produced by trypsin digestion of the IgM molecule in the presence of 5 M urea (Shimizu et al., 1975) provided the COOH-terminal sequence of CN5 and the overlap between CN5 and CN6. The first 30 amino acids of the Fe fragment contain seven proline residues which are in the boundary region between c 1 and c 2. This proline-rich stretch in the µ chain then bears a striking 8 8 resemblance to the proline-rich hinge regions of y and a heavy chains (Aldersberg, 1976). (e) Sequence of CN6. The CN6 fragment encompasses the entire c 2 8 domain and half of the c 3 domain in its 166 residues. Extending from asparagine 235 8 to methionine 397 , CN6 contains two complex carbohydrate moieties, one in c 2 8 and one in c 3 (see below and Figure 2). The amino acid sequence of the first 8 half of CN6 was established by Edman degradations of the NH -terminus and 2 of the two arginine fragments (CN6RI and CN6RII). Cleavage of the one aspartic acid-proline peptide bond in the fully reduced and alkylated 104E µ chain by mild acid produced a fragment with an NH -terminus at proline . The sequence 286 2 of this acid-derived fragment was then overlapped with the NH -terminus of the 2 CN6RII arginine fragment. The remaining COOH-terminal sequence of CN6 was determined by isolating CN6RII, followed by demaleylating or decitraconylating 39 the lysine residues and digesting the fragment with trypsin. Four predominant amino termini resulted from this procedure: CN6RIIa, beginning at the NH 2 terminus of CN6RII, glycine 326 ; CN6RIIb, produced in low yield due to incomplete cleavage of the peptide bond between lysine CN6RIIc, beginning at serine at phenylalanine 348 362 331 and glycosylated asparagine 332 ; ; and CN6RIId, the result of a chymotryptic cleavage . The sequence was confirmed and completed by the isolation of the thermolysin and trypsin plus chymotrypsin peptides, CN6TC4, CN6Th7 and CN6TC6 (Figure 2). The overlap between CN6 and CN7 has not been determined, but the sequences of CN6 and CN7 are contiguous when aligned by homology with two human µ chains and a dog µ chain (Watanabe et al., 1973; Putnam et al., 1973; Wasserman & Capra, 1978). Therefore we believe that no additional cyanogen bromide fragment is located between CN6 and CN7. (f) Sequence of CN7. The CN7 fragment, which is 109 residues in length (glutamic acid 398 to methionine 506 ), spans the remaining portion of c 3 and 8 the first two-thirds of the c 4 domain (Figure 2). The one high mannose carbo8 hydrate moiety in CN7 is attached to asparagine 402 in the CH3 domain, making c 8 4 the only µ chain constant region domain which has no covalently attached oligosaccharide moiety. In addition to an NH -terminal sequence, the sequences 4 of the two tryptophan fragments (CN7WI and CN7WII) and the sequences of the overlapping thermolysin and tryptic peptides, CN7Th10, CN7T6, CN7T7, CN7Th13, CN7Th14, CN7Th15 and CN7T8 completed the covalent structure of the CN7 fragment (Figure 2). (g) Sequence of CNS, CN8-9 and CN9. The sequence of a peptide which overlaps CN7 and CNS has been determined by Milstein et al. (1975). NH -terminal 2 sequenator runs on CNS and CNS-9, beginning at proline through tyrosine 562 507 , established the sequence . Cleavage at the single arginine residue in CNS-9 allowed determination of the remaining sequence of the C-terminal segment (to tyrosine 576 ). 40 Sequence determination on the arginine fragment of CN8 indicated that at least two-thirds of this peak lacked residues 569 to 576 (Figure 2). Isolation and sequence analysis of the CN9 fragment from a cyanogen bromide digestion of 14 c-alkylated 104E µ chains confirmed the cyanogen bromide cleavage between methionine 568 and serine 569 . The high mannose carbohydrate attached to asparagine 563 in the C-terminal segment was identified from a chymotryptic and thermolysin peptide (CN8-9CTh4). DNA Sequences. Recently, several C µ gene clones have been obtained, both from BALB/c germline chromosomal DNA and from MOPC 104E µ chain messenger-complementary DNA (Calame et al., 1980). As is shown in Table III, DNA sequence analysis of these clones provided data covering approximately 60% of the 104E µ chain constant region (260 out of 450 residues) (Calame et al., 1980; Rogers et al., 1980; Rogers, unpublished results). These DNA sequences confirm the protein sequence at all positions except for two discrepancies (asparagine 347 to threonine and threonine 379 to asparagine). Since the cloning of this gene involved no in vitro copying of the DNA, it is unlikely that the differences are due to a cloning artifact, unless a mutation has occurred during replication in E. coli. These residues also are clearly identified in the automatic sequenator runs. Since most of these sequence determinations were performed on peptide mixtures containing two or three sequences, we would like to point out that the sequence presented here does indeed conflict with the DNA sequence data. These sequence differences may represent a low level of natural polymorphism for the C µ gene. Carbohydrate Moieties. Six carbohydrate moieties are attached to asparagine residues in the 104E µ chain. One site is located in the VH region on asparagine 57 (Figure 2). The residues which are glycosylated were initially identified as blank cycles in the automatic sequenator runs. The precise location of each carbohydrate 41 was subsequently confirmed by the recovery of glucosamine from a short-term (3 h) HCl hydrolysis on a small peptide in addition to the presence of an extra aspartic acid residue which was not identified in the sequenator runs. The constant region cyanogen bromide fragments CNS, CN7 and CNS each contain one oligosaccharide. The carbohydrate compositions of these purified peptides and of two small pronase glycopeptides from the CN6 fragment are shown in Table II. These compositions establish that the first three constant region oligosaccharides attached to residues 171, 332, and 364 are of the complex type. In addition, the CNS and CN6 fragments which contain these complex type carbohydrates are characteristic glycopeptides in that these are the only 104E µ chain cyanogen bromide fragments which contain galactose residues, as judged by extrinsic labeling with galactose 3 oxidase and NaB H (Kehry et al., 1980). 4 Carbohydrate compositions on the CN7 and CNS oligosaccharides (Table II) indicate that these two COCH-terminal carbohydrates are high mannose oligosaccharides. The carbohydrate moiety attached to fragment CN3 at asparagines 7 in the V H region contains some glucosamine by ninhydrin analysis (Table I) and small amounts of mannose, galactose, and an unidentified sugar derivative (Table II). This must be a unique carbohydrate structure, not of the high mannose or complex type, consisting of one glucosamine and one mannose residue with a small proportion of the asparagines residues having galactose and/or another sugar moiety. 7 Discussion Complete µ Chain Sequence. The complete amino acid sequence of the MOPC 104E µ chain has been determined (Figure 2). The µ chain may be divided into a v and a C region. The C region can be further subdivided into four µ µ 8 domains of approximately 110 residues whose boundaries have previously been delineated by comparisons of internal heavy chain amino acid homology with the 42 three-dimensional structure of the immunoglobulin molecule (Beale & Feinstein, 1976). In addition, a 20-residue C-terminal segment is located COCH-terminal to the C µ 4 domain in the secreted µ chains. There are five sites of carbohydrate attachment in the secreted µ chain constant region, with only the C 4 domain µ lacking carbohydrate moieties. When the amino acid sequences of mouse, human and dog µ chains are compared, there is an increasing gradient of sequence homology from the NH -terminal VH domain to the C-terminal segment (Kehry et al., 1979). 2 Immunoglobulin heavy chain domains are also known to be discrete coding elements at the level of the heavy chain gene and are separated from one another by short intervening DNA sequences (Early et al., 1979; Calame et al., 1980; Tucker et al., 1979a,b). The coding region for the C-terminal segment of the secreted µ chain, though, is contiguous with that of the C 4 domain (Calame et al., 1980). µ The precise domain junctions which have been determined by DNA sequence analysis of BALB/c µ genomic and complementary DNA clones (Calame et al., 1980) are in remarkable agreement with those determined form the µ chain amino acid sequence (Beale & Feinstein, 1976; Kehry et al., 1979). For the secreted µ chain, there are five domain boundaries which have been described (Beale & Feinstein, 1976), and four of these boundaries match those determined from the µ gene DNA sequence or are only shifted by one amino acid (Calame et al., 1980). The remaining boundary between C 1 and C 2 in the µ chain gene is shifted by just three amino µ µ acids from the boundary previously described from the protein structure. Carbohydrate Moieties in the µ Chain. Two high mannose and three complex type carbohydrate moieties are attached to asparagine residues in the µ chain constant region. Both types of oligosaccharide structures are composed of a core carbohydrate containing two N-acetylglucosamine and five mannose residues (Kornfeld & Kornfeld, 1976). The high mannose structures are synthesized by the addition of mannose residues to the core to form a branched structure. 43 Complex type carbohydrates are also highly branched and are formed by the sequential additions of glucosamine, galactose, fucose and N-acetyl neuraminic acid to the core structure (Kornfeld & Kornfeld, 1976). In addition, a unique small carbohydrate structure which is not of the high mannose or complex type is located in the 104E second hypervariable region. As noted previously, four of the five C region carbohydrates are attached µ to identical positions in the mouse and human µ chains (Kehry et al., 1979). Each oligosaccharide is attached to the asparagine residue of an Asn-X-Ser/Thr recognition sequence (Marshall, 1972), where X may be any amino acid. Since three additional recognition sequences which are not glycosylated are present in OU and 104E µ chains (Kehry et al., 1979), clearly the conservation of carbohydrate attachment sites suggests they are important structural or functional features of the IgM molecule. Interestingly, the DNA sequence discrepancy which could change position 379 to an asparagine residue would create a recognition sequence for the glycosylation of that asparagine residue. In the mouse µ chain there is an oligosaccharide which is attached to asparagine 364 (Kehry et al., 1979). This oligosaccharide is in a different location (asparagine 395 ) in the human µ chain. This change in carbo- hydrate location is the result of a change in amino acid sequence between mouse and human µ chains. Thus, if the DNA sequence of the mouse µ chain indicates that residue 379 is asparagine, the selective pressure in this region to form a recognition sequence for glycosylating enzymes is remarkably high. The general type of carbohydrate moiety, complex or high mannose, is completely conserved between mouse and human µ chains (Shimizu et al., 1971). Proceeding from the CHl to the C-terminal domain, the first three oligosaccharides attached to the µ chain are of the complex type, containing terminal galactose and N-acetyl neuraminic acid residues. The remaining two oligosaccharides, including the one in the C-terminal region, are high mannose carbohydrate structures composed 44 of glucosamine and mannose. Apparently, the glycosylating enzymes involved in the synthesis of these distinct types of oligosaccharides efficiently discriminate among multiple recognition sites on the same molecule. The complex and high mannose oligosaccharides then presumably have different structural or functional roles in the IgM molecule which require a precise localization. At first glance, though, carbohydrate structures attached to heavy chains of different classes seem to differ in type and location (Torano et al., 1977). For example, most human y chains possess one oligosaccharide N-glycosidically linked to an asparagine residue in the c 2 domain (Kornfeld et al., 1971; Beale 8 & Feinstein, 1976). Human Cl chains differ remarkably in glycosylation even between the Cll and Clz subclasses. The Cl chains have one or two (depending upon the 2 Clz allele) N-linked carbohydrate structures located in the c 8 1 and hinge domains and two in the c 2 domain, one of which is identical in the a 1 chain. Both a 8 1 and a chains have one N-linked carbohydrate in the C-terminal segment and 2 only Cll chains have five 0-linked (to serine residues) oligosaccharides in the hinge region (Torano et al., 1977; Tsuzukida et al., 1979). Human e: chains, like µ chians, contain only N-linked oligosaccharides, three of which are located in the c 1 8 domain, one in the c 2, and two in the c 3 domain (Bennich & von Bahr-Lindstrom, 8 8 1974). The one oligosaccharide in y chains (in the C 2 domain) is in a homologous "( position to an N-linked carbohydrate in the c 3 domain of µ and e: chains (Beale 8 & Feinstein, 1976), suggesting that these domains either have a common evolutionary origin or that they perform similar functions. The one common feature shared by all heavy chain classes is that the last domain (C 3, C 3, C 4 or C 4) lacks y Ct µ e: a carbohydrate moiety entirely. In the case of y and e: chains, a carbohydrate moiety may interfere with Fe receptor-immunoglobulin interactions in this COOHterminal domain. 45 The one other exception to this lack of homology in the location and type of heavy chain carbohydrate structures is the high mannose oligosaccharide situated in the C-terminal segment of both µ (asparagine 563 ) and a (asparagine 459 ) chains (Putnam, 1974). This carbohydrate is located on homologous residues in human a, human µ, and mouse µ chain C-terminal segments. In addition, the amino acid sequence homology between human a and µ chain C-terminal segments is a striking 70% (14/20 identical residues). Since IgM and IgA are the only immunoglobulin classes which form polymers with the J chain, and since only a and µ chains possess the 20-residue C-terminal segment, this region must be very important in the polymerization process and in the maintenance of the polymer structure. This high amount of homology in the C-terminal segments of a and µ chains is even more impressive when one considers that the degree of c 8 region homology between µ and a chains (40% by amino acid sequence) is such that cloned C µ and Ca gene probes will not cross-hybridize (P. Early, personal communication). The penultimate cysteine residue in a and µ chains is involved in the formation of covalent linkages to the J chain and to other heavy chains in the polymer (Della Corte &: Parkhouse, 1973; Mestecky &: Schrohenloher, 1974; Mestecky et al., 1974; Mendez et al., 1973). In addition, we envision that the highly conserved C-terminal high mannose oligosaccharide plays a structural or functional role in the IgM and IgA polymer. The potential importance of the C-terminal segment in polymer formation is strengthened by the fact that membrane IgM molecules, which exist only as monomeric immunoglobulins, lack the C-terminal segment and carbohydrate moiety present in secreted µ and a chains (Kehry et al., 1980). The composition of the C-terminal high mannose oligosaccharide is different 3 from that of the other oligosaccharides in the µ chain. Quantitation of the Hmannose incorporated into 104E µ chain cyanogen bromide fragments correlates well with the number of carbohydrate moieties attached to each fragment (Kehry 46 et al., 1980). The CN7 and CNS fragments, each with one carbohydrate, and the CN6 fragment with two carbohydrates are labeled with 3 H-mannose in a 1:1:2.5 3 ratio (CN7:CN5:CN6). But, quantitation of H-mannose labeled CNS plus CN8-9 fragments shows that only 0.5 carbohydrate moiety is attached to the C-terminal segment. There are two possible explanations for the fractional amount of mannose in the C-terminal oligosaccharide. Either this carbohydrate structure contains half as many mannose residues as a mature high mannose oligosaccharide, or only a portion of the µ chains are glycosylated at this position. We favor the latter explanation in view of our amino acid sequence data on the arginine fragment of CNS. In the automatic sequenator, the asparagine residue to which a carbohydrate moiety is attached is not extracted out of the sequenator cup and therefore shows up as a blank cycle in the analytical system. In the sequence of the CN8-9 arginine fragment, asparagine 563 was a blank cycle, as expected for a position of carbohydrate attachment. But in the arginine fragment of CNS, position 563 showed approximately 0. 7 residue of asparagine, indicating that a portion of the 104E µ chains are unglycosylated in the C-terminal region. Since the fractional mannose incorporation into .the CNS fragment also is reproducibly seen in other secreted µ chains and in internal precursors to secreted µ chains (Kehry et al., 1980), it is probably significant at the structural or functional level. Domain Homologies Among Heavy Chains. (a) Mu Chain Homologies. When the mouse, dog and human C µ sequences are compared, there is an increasing homology from the C 1 domain to the C-terminal segment (Kehry et al., 1980). µ The mouse and human µ chains are 61 % homologous in the C µ region, ranging from 48% identical residues (in addition to two sequence gaps in C µ 1 and three gaps in C 2) in the C 1 domain to 90% identical residues in the C-terminal segment µ µ (no sequence gaps in C 3, C 4 and the C-terminal segment). Similarly, the mouse µ and dog C µ µ regions are 59% homologous with a range of 50% identical residues 47 in the C 1 domain (two sequence gaps in C 1 and two gaps in C 2) to 80% identical µ µ µ residues in the C-terminal segment (again, no sequence gaps after the C 2 domain). µ The amino acid sequence differences between 104E µ chains and those of human OU and dog MOO appear to be clustered in eight regions of 12 to 20 amino acids in length (residues 161-180, 210-225, 260-279, 330-347, 383-397, 410-425, 437-449, and 484-498). In all but one of these nonhomologous regions, the dog MOO and human OU chains exhibit significantly more homology to one another than either does with the mouse 104E µ chain (Kehry et al., 1979). One of the more striking examples is residues 383-397, where mouse 104E and human OU µ chains differ in 13/15 positions while dog MOO and human OU differ in only 1/15 residues. Since ON A studies on the mouse C µ gene suggest that the C µ gene is present in only one copy per haploid genome (Early et al., 1980; Calame et al., 1980; Cory &: Adams, 1980), the differing extents of homology among the mouse, dog, and human µ chains are not due to the existence of µ chain subclasses in the mouse. The difference probably reflects a closer evolutionary relationship between dogs and humans than between mice and dogs or humans. We have previously shown that among mouse, human and dog µ chains, the C µ 4 domain is the most highly. conserved of the four µ chain constant region domains (Kehry et al., 1979). Interestingly, the binding site for the first component of the complement cascade has been localized to a region in C 4, residues 468-491 µ joined to 515-546 (Hurst et al., 1975). These two stretches of amino acids are at least 75% identical in the sequences of the mouse, human, and dog µ chains. This complement binding region contributes significantly to the high degree of C 4 sequence conservation among the three species, and presumably reflects µ the structural and/or functional constraints related to complement fixation that have been imposed upon this distinct region during species evolution. 48 (b) Homologies Among µ, y, ex, and e: Chains. An interesting question in the evolution of the different heavy chain classes is how additional domains have evolved (four Ce: and C µ domains), or how regions have been deleted (no hinge region in µ or e: chains). The intervening DNA sequences between the domains in heavy chain genes provide an excellent mechanism for rearranging, duplicating, or deleting individual domains. An examination of the homologies among all of the completely sequenced CH domains may reveal the evolutionary origins of these c 8 domains. In particular, these comparisons may be used to identify a primordial domain or pair of domains from which all heavy chain genes evolved by gene duplication. For example, in an analysis of the divergence exhibited by mouse y chain protein and nucleic acid sequences, Honjo has found that the CHl domains of yl' y a and y b chains have diverged singificantly less from one another than 2 2 any of the other C c 8 y domains (Yamaw.aki-Kataoka et al., 1980). In addition, the 2 domains of y a and y b chains also diverge very little, indicating that the 2 2 CHl and c 8 2 gene segments of yl' y a and y b chains have recombined prior 2 2 to the complete divergence of these subclasses in the mouse. 49 References Adetugbo, K., Milstein, C., & Secher, S. (1977) Nature 265, 299. Aldersberg, J. B. (1976) La Ricera Clin. Lab. §_, 191. Barstad, P., Hubert, J., Hunkapiller, M., Goetze, A., Schilling, J., Black, B., Eaton, .J B., Richards, J., Weigert, M., & Hood, L. (1978) Eur. :I_. Immunol. ~' 497. Beale, D., & Feinstein, A. (1976) S· Rev. Biophys. g_, 135. Bennich, H., & Von Bahr-Lindstrom, H. (1974) Progress in Immunol . .!., 49. Calame, K., Rogers, J., Early, P., Davis, M., Livant, D., Wall, R., &. Hood, L. (1980) Nature, in press. Cebra, J., Brunhouse, R., Cordle, C., Daiss, J., Fechheimer, M., Ricardo, M., Thunberg, A., & Walfe, P. B. (1977) Prog. Immunol. ~' 269. Cory, S., & Adams, J. M. (1980) Cell 19, 37. Dayhoff, M. O. (1976) Atlas of Protein Sequence and Structure, Vol. 5, pp. 189190, National Biomedical Research Foundation, Silver Spring, MD. Della Corte, E., &. Parkhouse, R. M. E. (1973) Biochem. :I· 136, 597. Early, P. W., Davis, M. M., Kaback, D. B., Davidson, N., &. Hood, L. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 857. Early, P., Rogers, J., Davis, M., Calame, K., Bond, M., Wall, R., & Hood, L. (1980) Cell, in press. Edelman, G. M., Cunningham, B. A., Gall, w. E., Gottlieb, P. D., Rutishauser, U., &. Waxdal, M. J. (1969) Proc. Natl. Acad. Sci. U.S.A 63, 78. Fougereau, M., Bourgois, A., De Preval, C., Rocca-Serra, J., & Schiff, C. (1976) Annales D'Immunologie 127C, 607. Fruchter, R. G., Jackson, S. A., Mole, L. E., & Porter, R. R. (1970) Biochem. :I· 116, 249. Gross, E. (1967) Methods Enzymol . .!..!., 238. Hunkapiller, M. W., & Hood, L. E. (1978) Biochemistry 17, 2124. 50 Hurst, M. M., Valanakis, J. E., Stroud, R. M., &. Bennett, J. C. (1975) !!· Exp. Med. 142, 1322. Johnson, N., Hunkapiller, M., &. Hood., L. (1979) Anal. Biochem. 100, 335. Kabat, E. A., Wu, T. T., &. Bilofsky, H. (1976) Variable Regions of Immunoglobulin Chains, Medical Computer Systems, Cambridge, MA. Katz, A. M., Dreyer, W. J., &. Anfinsen, C. B. (1959) !!· Biol. Chem. 234, 2897. Kehry, M., Sibley, c., Fuhrman, J., Schilling, J., &. Hood, L. E. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 2932. Kehry, M., Ewald, S., Douglas, R., Raschke, W., Sibley, C., Fambrough, D., &. Hood, L. (1980) Cell, submitted. Kim, J. H., Shome, B., Liao, T.-H., &. Pierce, J. G. (1967) Anal. Biochem. 20, 258. Klapper, M. H., &. Klotz, E. M. (1972). Methods Enzymol. 25, 531. Konigsberg, W. (1972) Methods Enzymol. 25, 185. Kornfeld, R., Keller, J., Baenziger, J., &. Kornfeld, S. (1971) !!· Biol. Chem. 246, 3259. Kornfeld, R., &. Kornfeld, S. (1976) Ann. Rev. Biochem. 45, 217. Kratzin, H., Aldevogt, P., Kortt, A., Ruban, E., &. Hilschmann, N. (1978) ~- Physiol. Chem. 359, 1717. Laemmli, U. (1970) Nature (London) 227, 680. Lawton, A. R., Kincade, P. W., &. Cooper, M. D. (1975) Fed. Proc. Fed. Am. Soc. Exp. Biol. 34, 33. Loh, E., Black, B., Riblet, R., Weigert, M., Hood, J. M., &. Hood, L. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 1395. Low, T. L. K., Liu, Y. S. V., &. Putnam, F. W. (1976) Science 191, 390. Marchalonis, J. J. (1972) Nature (London) 236, 84. Marshall, R. D. (1972) Ann. Rev. Biochem. 41, 673. 51 Maxam, A. M., & Gilbert, W. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 560. Mccumber, L. J., Wasserman, R., & Capra, J. D. (1979) Mol. Immunol. 16, 565. Mendez, E., Prelli, F., Frangione, B., & Franklin, E. C. (1973) Biochem. Biophys. Res. Commun. 55, 1291. Mestecky, J., & Schrohenloher, R. E. (1974) Nature 249, 650. Mestecky, J., Schrohenloher, R. E., Kulhavey, R., Wright, G. P., & Tomana, M. (1974) Proc. Natl. Acad. Sci. U.S.A. 71, 544. Michaelson, T. E., Frangione, B., & Franklin, E. C. (1977) !!· Biol. Chem. 252, 833. Milstein, C. P., Richardson, N. E., Deverson, E. F., & Feinstein, A. (1975) Biochem. !!· 151, 615. Ozols, J., Gerard, c., & Stachelek, C. (1977) !!· Biol. Chem. 252, 5986. Peacock, A. C., & Dingman, C. W. (1968) Biochemistry 1, 668. Pink, J. R. L., Buttery, S. H., de Vries, G. M., & Milstein, c. (1970) Biochem. !!· 117, 33. Ponstingl, H., & Hilschmann, N. (1972) Z. Physiol. Chem. 353, 1369. Putnam, F. w., Florent, G., Paul, C., Shinoda, T., & Shimizu, A. (1973) Science 182, 287. Putnam, F. w. (1974) Prog. Immunol. (II).!., p. 25, ed. L. Brent & J. Holborrow, North Holland, Amsterdam. Robinson, E. A., & Appella, E. (1977) Proc. Natl. Acad Sci. U.S.A. 74, 2465. Rogers, J., Early, P., Carter, c., Calame, K., Bond, M., Hood, L., & Wall, R. (1980) Cell, in press. Shimizu, A., Putnam, F. W., & Paul, C. (1971) Nature (London) 231, 73. Shimizu, A., Watanabe, S., Yamamura, Y., & Putnam, F. w. (1975) Immunochemistry !.!., 719. Torano, A., Tsuzukida, Y., Liu, Y. S. V., & Putnam, F. W. (1977) Proc. Natl. / Acad. Sci. U.S.A. 74, 2301. 52 Torano, A.,&: Putnam, F. W. (1978) Proc. Natl. Acad. Sci. U.S.A. 75, 966. Trischmann, T. M., &: Cebra, J. J. (1974) Biochemistry g, 4804. Tsuzukida, Y., Wang, C. C., &: Putnam, F. W. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 1104. Tucker, P. W., Marcu, K. B., Slightom, J. L., &: Blattner, F. R. (1979a) Science 206, 1299. Tucker, P. w., Marcu, K. B., Newell, N., Richards, J., &: Blattner, F. R. (1979b) Science 206, 1303. Vitetta, E. S., Bauer, S., &: Uhr, J. W. (1971) ~- Exp. Med. 134, 242. Warner, N. (1974) Adv. Immunol. 19, 67. Wasserman, R. L., &: Capra, J. D. (~978) Science 200, 1159. Watanabe, S., Barnikol, H. U., Horn, J., Bertram, J., &: Hilschmann, N. (1973) Z. Physiol. Chem. 354, 1505. Wittmann-Liebold, B. (1973) Hoppe-Seyler's ~- Physiol. Chem. 354, 1415. Wittmann-Liebold, B., Graffunder, H., &: Kohls, H. (1976) Anal. Biochem. 75, 621. Wolfenstein-Todel, C., Frangione, B., Prelli, F., &: Franklin, E. C. (1976) Biochem. Biophys. Res. Commun. 71, 907. Yamawaki-Kataoka, Y., Kataoka, T., Takahashi, N., Oba ta, M., &: Honjo, T. (1980) Nature 283, 786. 53 TABLE I: Amino Acid Composition of Tryptic Peptides from 104E µ CN3~ CN3Tlb CN3T2 CN3T3 CN3T4 CN3T5 CN3T6 CN3T7 Q1Cys Asx 4.9 (5) 1.0 (1) Thr 1.0 1.9 (1) (1) Ser 1.0 (1) 1.7 (2) Glu 0.9 (1) 1.8 (2) 1.1 Pro Gly 1.1 (1) 1.1 (1) 2.8 (3) (1) 2.8 (3) 0.9 Ll) 1.0 (1) 1.2 Ala 1.0 (1) 0.9 (1) Val (1) Ile 1. 3 (2) Leu 1.0 Tyr 1.0 (1) 1.0 (1) (1) 0.9 (1) 0.9 (1) Phe Glc-NH e 2 0.2 0.9 His HSer (1) C 0.6 (1) 1.1 (1) 1.1 (l) 1.0 (1) 1.1 (1) 1.1 (1) Lys 1.0 (1) Arg Trpf + Total residues 3 5 20 2 2 7 7 Yield (\)d 30 35 15 60 70 75 20 41-45 46-64 65-66 67-68 69-75 76-82 Residue Numbers 9 38-40 54 Legend to Table I !!. Values reported are amino acid residues. Amino acids present at a level of less than 0.2 residue are omitted. Values in parentheses represent the nearest integral number of residues present in the sequence. ~ The name of each peptide indicates the cyanogen bromide fragment from which it was derived, followed by the enzyme used and the contiguous order in the fragment. T =trypsin; Th =thermolysin; C =chymotrypsin . .£ Includes homoserine as well as homoserine lactone. 9. Yields are based on nanomoles of peptides isolated compared with nanomoles of fragment originally digested with enzyme. ~ The presence of glucosamine was determined by a separate hydrolysis in 6 N HCl at 110° for 3 h. -f u.v. pos1·t·1ve pep t"d 1 e. S: The numbering does not correspond to the number of amino acids due to the introduction of gaps for sequence alignment. l, ?. l l Ala - 5 l 2 Lys II I. j lt ,UIJ ~ ~ l - 111 ------------------- Wt! i,~h t. l' :·1, •,) t <·(I He:; i J11t! Nwnl,e 1·:.; 1' Ml) J1 .•c11l. 111· ;•o Xd 'l'O'l'AJ. •,11\0 1'.,-11. ' o . .• ,.-, ,,., , ' I ~ \l j I , -- 11 .l-1',I •Ji1i .fJ 1·, . ·- . ~,. Hh 1H, 10 •, { ',- \ 'l°j Jl.1h 1 1.1 1'l ~ I', Jl)'{-' ,t )t1 t', ;! 0 .1 .. .' ',I ll ) ', I,'}-','((, ts l 2 l 2 l I CN'I {1 ~ , . 'I H I '.>r(? 6 .l '., JI 10 11 22 22 1.6 24 8 4'., 1'> JO 40 19 31 '.,6 '.,4 30 23 14 1'ultd \./h(Jlt> --------------- ------ -------- I . I • : ' ~I I ·1• )/1- ', llh lll') 1.0 5.4 l. '> 0.-1 ,. ! NAUA G1 c: -Nll ,). l ~-" I.'> 1.0 o.·1 i. .1 ·1.6 0.2 2 .1 1 I 0.1 o. ;~ Fuc Gul ! . .S l ., 5.) ,,, A 5 Ii 11 ,...., O.', I "i. l. O I l l l I ·1 ., 3 9 l I 4 l l l 2 2 2 4 8 l 1 0. '> 2 l l j ~? ~ ·i 0 15 4 2 6 2 2 11 ·1 5 ~ 1 4 9 2 l l CNtl 9 J' • H ll 4 ll n ., 6 1 9 6 5 2 1 JI\ 21 1() 3 3 ,, "( CN"( CUI, Mau C CHO moiety Arg l 2 us er llis I l l Phe 2 2 •rrp '.> .. 4 ;> '> ,, ( '., 4 (, 3 2 1 4 2 3 2 5 ,, ( 2 2 Tyr Leu Ile Met 3 l 2 Pro t:ly V11l 2 3 can 3 10 6 l 2 2 2 St r 4 '., 'i 4 3 CH1 , ) rn4 ;> 4 2 CN.3 Glu Thr Asn l CN ?. AHp CNl Amino Ac: id C,,1111-,os ition of CNllr Fntgrn,•nt s fn uu 1U4f u. l I I. Cys 'l'Alll.t. II C.11 C.11 56 Legend to Table II !!. Values reported are amino acid residues from the completed sequence. ~ Number of carbohydrate attachment sites in each cyanogen bromide fragment. Locations: CN3, residue 57; CN5, residue 171; CN6: A, residue 332, B, residue 364; CN7, residue 402; CNS, residue 563. £ Carbohydrate residues determined by gas chromatography of alditol acetates. Values reported are mole carbohydrate/mole glycopeptide. The two glycopeptides in CN6 were obtained by pronase digestion of CN6 followed by chromatography on Biogel P-10. A, first peak from P-10 column; B, second peak from P-10 column. Abbreviations: Fuc, Fucose; Man, Mannose; Gal, Galactose; Glc-NH2' glucosamine; NAN A, N-acetylneuraminic acid. ~ Unidentified gas chromatographic alditol acetate, present only in CN3. ~ Molecular weight in daltons calculated from the sequence of amino acids in the fragment; does not include carbohydrate. g See legend to Table I. 57 TABLE III. Sequenced Portions of Cµ Gene Clone Ref. 127 - 153 µAl * 218 - 250 µAl * 300 - 323 µ12 Calame et al., 1980 336 - 388 µAl * 432 - 466 µAl * 432 - 470 µ12 * 531 - 556 µ6 Rogers et al., 1980 542 - 576 µ12 Calame et al. , 1980 Codons (inclusive) Numbering is as for Ou. For a description of the clones and maps showing sequencing strategies, see Calame et al. (1980) and Rogers et al. (1980). µAl= SpµAl; this is a subclone from ChSpµ7, which contains chromosomal DNA from I BALB/c sperm. ~6 = pl04Eµ6, µ12 = pl04Eµl2; these contain complementary DNA from MOPC 104E mRNA. The sequences of codons 218 - 250 and 542 - 576 were obtained from two strands; the remaining sequences were obtained from only one of the two strands. *Unpublished results. 58 FIGURE 1: Separation of MOPC 104E cyanogen bromide fragments. Completely reduced and alkylated 104E µ chains (35 mg) were cleaved with cyanogen bromide and the fragments separated by gel filtration on a column of ACA 54 (1KB) as described in Experimental Procedures. Fraction volume was 5 ml. The fragments are labeled (CNl to CN9) according to their position in the µ chain sequence beginning at the NH -terminus. Agg. is the excluded column peak which consists of aggregated 2 fragments and large uncleaved peptides. At the top of the figure is a schematic drawing of the 104E µ chain (to scale) showing the linear order and sizes of the cyanogen bromide fragments (CNl to CN9). Two incomplete cleavage products, CNl-2 and CN8-9, result from partial cleavage of the methionine residues at positions 20 and 568, respectively. CHO denotes the sites of carbohydrate attachment. The complex type carbohydrates are indicated by boxes and the high mannose carbohydrates are indicated by circles. FIGURE 1. 70 I 90 I I I I 110 I ~ I A9:. I \ O.IL 00200 CN7 CN6 • A 0.2 , c,. 130 CNB I 150 I I II JI CN3 CN4 170 cNfrTI CN7 166 109 Fraction Number I I CN5 CNBr c,Nlf~ CN3, ;N ,4:~N5 Frag men ts 2014 47 62 84 v. 190 CN2 210 CN9 I I CJl (0 60 FIGURE 2: Complete sequence of the MOPC 104E µ chain. The one-letter code for amino acids is given in Dayhoff, (1976). Tryptic (T), chymotryptic (C), and thermolysin (Th) peptides are indicated. Sequences of peptides produced by cleavage at methionine (CN), arginine (R) and tryptophan (W) residues are also indicated in the left-hand margin. All peptides are identified by the cyanogen bromide fragment number from which they were derived. Residues sequenced automatically are indicated by -,. . CHO denotes the sites of carbohydrate attachment to asparagine residues. The constant region domain boundaries are from DNA sequence determinations on a cloned BALB/c µ gene (Calame et al., 1980). 61 --. _:r_, __ I le • • .. • f:ii • I 'f O I. 0 0 I e PI L 'f I I 4 4 •VI ■ IC I a I Gt Tr T o r r • I • v IQ I IO I I I. I • IO 8 I I r ■ (loo 1' I ----------------------------------- -- an 011 ; • o • , • • 1 • ; r. , • o 1 • • • ,. A r • o 1. • • 1. T 1 ; o • a • , , c • 1 'D, 'it• r , o v • o ,. o T,. v , v • ; ~ o 'i CM•--•••-•••••••-• S9 '1 ... .,.,.,, '-=•°"l'!..u.J ~ ,....-·--~=~--~ '---==--'■'--"'"'"-'S:i.'C..t..l-'1 i:, • .,. , , .... c • • , '" , o •• "' .. -: •• c " ... o , "7. ,. , • , ,. •• , ~ ~ c:.. .. ••--••-'-'--=-=----'1":_,-,::°:'::..:~t,-,_-:t C. ~ t c-11t-. I c:.,_,. I '.:ace-..":., -----------------------------------'• .r.~ .•r-"1 IC."- I CMi°'I ·of. ,. . ., , o'" 1 • ';", • , "' • , G Q • · ;- I .... .z:.·:J ...1.r.:a!".#-u1·~ - - - - - - - - - - - - - - - - - - - - - - - - - - ~----------------------- ~ 4 T I O 'f I. I. I ; I ■ I L I G I O I';" L 'I C I l I t O G i I I O L O Y Pc..;~~ A I ■ I P ■ Y I-; P Y P P I O O P I : '-St':.,. c:.• - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ' - ·-----=?I='''---"' "----21..!!..-a.D-------------------------- a. '---A!aL...., Ulll!!W ...._ _.,..."'''"''-~ .. - .. =-·· ~----------------------------------·a.---------------- - - . . . . . . . . I. l C l & T I P T P I P I T'flWLIOGIL\l'ISQPTTOP'fT I I I I C I T P Q T T I e!!I 'ca."'• <• S-,!) l'C> -'f 1 I t' I, t r I I -1 I • I. ■ I. ■ 'f f T -C I V O I I G I. T P J;! 'i:" (ii ■ V 1 I t' C'-'-r.t'-' A art r I T Q 1 I. t' P T -1 P P I P A I 1 P I. -I / -•--■ ca... ~---------------------------------~ Jr,,. ---- - -I I A.I. fi;I f CI, 'f -I I I. & f TIT(, f -I I ■ A IO I GIP -I. I 1' I t I I ■ I I _B:;i I r(ia T PI A I Iii -VA IV CY IO ■■ -I ,,.,.,,.,,., eE S-'51 PT''I ~ t I I P 'f C T 'f T ~ I ■ I. P I P O I I ; I I I ~":Ti~_.I I i P P A Y f I. I. P r-: I I O I. ■ I. I I I-: T 'f T C I. ._ I G P-:- OIFWI sr..,,. sr, u ,..,., ?!1 ~ -P a O I I Y O ■ I -O I Cl O I. I. P Q I I -T V T I a I ■ -- I I P w• C a I 4 P T P 1' I I •I L T ." T I I I ■ ■ •I Q I T f T C V \I' G •I ~~"•••-~'Qi,;•;~•-•"•~•._...... a~•,..___________________________________ _ ~•-I ·••◄•-· ~ ' - <.•..,--:;- GI - I A I. P ■ I. 'f TI I 1' 'f DI 11' GJI PT I. trl\l' I I. 1 11 IDT G GT CT FIGURE 2. 21::SN 62 FIGURE 3: Yields of phenylthiohydantoin amino acids and sequence of MOPC 104E CN3 from an NH -terminal sequenator analysis. The CN3 fragment was purified 2 and the amino terminal sequence was determined as described in Experimental Procedures. Aliquots of each cycle were analyzed by high performance liquid chromatography; peak heights were converted to nanomole yield for each derivative by comparison with a standard phenylthiohydantoin amino acid mixture. Assigned residues are indicated by a larger filled circle and scale changes are as indicated by -\\- on the graphs. Resique number 1 corresponds to residue 35 in the intact 104E µ chain. 63 104E CN3 5 5 ASN ARG ASP 4 4 3 3 2 2 2 CYS 5 HIS >- o E 5 GLN 5 GLU 5 4 4 4 3 3 3 2 2 2 5 ILE 4 3 3 2 2 LEU X5 10 LYS 8 2 / C: X5 5 MET PHE 5 4 4 3 3 2 2 PRO SER 2 x5 THR 5 4 2 TRP 5 TYR 5 4 4 3 3 VAL 2 X5 I 5 10 15 20 25 30 35 40 I 5 10 15 20 25 30 35 40 I 5 10 15 20 25 30 35 40 I 5 10 15 20 25 30 35 40 RESIDUE NUMBER I 10 20 Lys-Trp-Vol-Lys-Gln-Ser-His-Gly-Lys-Ser-Leu- Glu-Trp- Ile- Gly-Asp-Ile -Asn-Pro-Asn21 30 40 ~Gly-Gly-Thr -Ser-Tyr-Asn-Gln -Lys-Phe-Lys-Gly -Lys-Alo-Thr-Leu -Thr-Vol -Asp-Lys-Ser FIGURE 3. 64 CHAPTER 3 65 CHARACTERIZATION OF MULTIPLE IMMUNOGLOBULIN MU CHAINS SYNTHESIZED BY TWO CLONES OF A B-CELL LYMPHOMA • ** + , MARILYN R. KEHRY , + RICHARD H. DOUGLAS\ WILLIAM C. RASCHKE\ and LEROY E. HOOD+ CAROL H. SIBLEY , SANDRA J. EWALD I *Department of Genetics, University of Washington, Seattle, Washington 98195. ** Department of Microbiology, Montana State University, Bozeman, Montana 59717. +The Salk Institute for Biological Studies, P.O. Box 85800, San Diego, California 92138 +Division of Biology, California Institute of Technology, Pasadena, California 91125. This paper will be submitted to i!· Immunology. 66 ABSTRACT We have identified three species of µ chain synthesized by mouse WEHI 279 lymphoma cells that differ in cellular location, size and charge - µ. (internal), 1 µ m (membrane) and µ (secreted). Lactoperoxidase and1 galactose oxidase labeling s experiments localize the µ m chain to the plasma membrane and the µ. chain to 1 the cytoplasm. Pulse-chase experiments demonstrate that the µ. pool contains 1 precursors to both µm and µs chains. Comparative peptide mapping studies and cell labeling in the presence of tunicamycin suggest the µ m chain is 2000 daltons larger than the µs chains with the difference due to covalent polypeptide alterations. The WEHI 279 lymphoma has differentiated while in culture to a more advanced stage in the pathway of B-cell differentiation. 67 INTRODUCTION B cells pass through an orderly series of differentiation stages. The stem cell differentiates into a pre-B cell which expresses cytoplasmic µ chains, but not light (L) chains (1, 2). The pre-B cell then undergoes additional changes and becomes a B lymphocyte which synthesizes light chains as well as µ chains. This nondividing B lymphocyte synthesizes monomeric IgM molecules ( µ 1 ) that 2 2 are placed on the plasma membrane to act as antigen receptors (3). Upon stimulation with antigen, subsequent differentiation steps occur, and the B cell acquires the ability to secrete pentameric IgM molecules held together by disulfide bridges between monomers and the J or joining chain (4) [(µ L ) JJ. The terminal stage 2 25 of B lymphocyte differentiation is the plasma cell which secretes large quantities of the IgM pentamer. Fortunately, various tumor cell lines synthesizing homogeneous immunoglobulin molecules appear to correspond to these distinct stages of differentiation (5, 6). For example, the B-cell lymphoma and the plasmacytoma or myeloma tumor correspond to the B lymphocyte and the plasma cell, respectively. The plasma cell membrane and the blood constitute very different chemical environments for IgM molecules and, accordingly, the membrane-bound and secreted IgM molecules, although similar, are apparently not identical proteins (7, 8, 9, 10, 11, 12, 13). The IgM molecule is anchored in the B-cell plasma membrane via the C-terminus of the µ chain (14, 15). Therefore, specific structural adaptations in the membrane IgM (mlgM) molecule which facilitate its interaction with the lipid bilayer should occur in this region. We have initiated an analysis of secreted ( µs) and membrane-bound ( µm) mu chains by comparing the IgM molecules synthesize·d by an IgM-secreting myeloma tumor with the membrane-bound molecules synthesized by a B-cell lymphoma. We have determined the complete amino acid sequence of the µs chain secreted by the BALB/c mouse plasmacytoma, 68 MOPC 104E (Ml04E) (16). We have compared these µ chains with the µ chains s m derived from the B-cell lymphoma WEHI 279 (W279). The W279 lymphoma was isolated and initially characterized by Dr. Noel Warner and colleagues (17). It originated in the NZC inbred mouse strain derived from a (BALB/c x NZB)F 1 mouse. The W279 IgM molecules bear the allotype of the BALB/c parent. The W279 cells originally had Fe receptors, high levels of migM molecules and secreted negligible amounts of IgM molecules. This suggests that they represented lymphocytes transformed at an early stage of B-cell development. A more recent clone (W279.1/12) of the original W279 cell line exhibits increased secretion of pentameric IgM molecules and decreased levels of mlgM molecules, representing a clone of lymphocytes transformed at a more advanced stage of B-cell development. Hybrid cells created by the fusion of W279 lymphoma cells and MPCll myeloma cells secrete large quantities of W279 µs chains assembled in the pentameric IgM form. In this paper we demonstrate three discrete pools of µ chains are produced by the W279.1/12 B-cell lymphoma, µ. (internal), µ (secreted) and µ l S m (membrane). These molecules are distinguished by their cellular location, charge, structure .. and size. The W279 cell line has differentiated while in culture to a more advanced IgM-secreting stage of B-cell differentiation. MATERIALS AND METHODS Cells. The WEHI 279 (W279) uncloned cell line was a gift from Dr. Noel Warner. The WEHI 279.1/12 clone, sent to us by Dr. Vernon Oi, was isolated by Dr. Warner's laboratory as a later clone of the W279 tumor cells with greater proliferative potential than the original WEHI 279 cell line. This clone will be denoted W279.1 hereafter. W279.1 cells were grown in Dulbecco's Modified Eagle 1s Medium (DMEM) supplemented with nonessential amino acids (GIBCO) containing 5 20% heat inactivated fetal calf serum, and 5 x 10- M S-mercaptoethanol in 69 a humidified atmosphere of 90% air, 10% CO 2. W279 cells were grown in the above medium containing only 10% fetal calf serum. The MPCllxW279 (MxW) hybridoma was generated by a fusion between the MPCll myeloma and WEHI 279 lymphoma cells (18). Hybridoma cells were grown in DMEM supplemented with 10% fetal calf serum. Labeling procedures. Cells were labeled for the indicated periods with 35 s-methionine (Amersham, 900-1200 Ci/mmole), 35 s-cysteine (New England Nuclear, 700 Ci/mmole) or 3H-tyrosine and leucine (New England Nuclear, 40 mCi/ mmole and 60 mCi/mmole, respectively) in DMEM devoid of methionine or tyro6 7 sine and leucine at cell densities between 2 x 10 /ml and 1 x 10 /ml. Intact cells and cell lysates were labeled with Na 1251 and lactoperoxidase according to Mar- • chalonis, Cone and Santer (19) or with NaB 3H4 and galactose oxidase by the method of Gahmberg and Hakomori (20). Cell lysates were prepared by washing cells once in cold medium and then suspending the cells in ice cold lysing buffer (0.5% Triton X-100, 0.14 M NaCl, 0.01 M Tris-HCl, pH 7.4, 2 mM phenylmethyl sulfonyl7 fluoride [PMSF]) at a concentration of 4 x 10 cells/ml for 20 min on ice. Supernatant fractions were prepared by centrifuging the cell lysates for 20 min at 1500 x g or 5 min at 12,000 x g to remove nuclei. When cells were labeled in the presence of tunicamycin (a gift from Dr. Robert Hamill, Eli Lilly Co.), the cells were first preincubated in 5 µg/ml, 1.5 µg/ml or 0.5 µg/ml tunicamycin for 60 min at 37°C. Cell labeling was then performed in the presence of 5 µg/ml, 1.5 µg/ml or 0.5 µg/ml of tunicamycin as described above. Immunoprecipitation. Cell lysates prepared from 2 x 10 6 cells were pre-incubated with 50 µl of a washed 10% suspension of formalin-fixed, heatkilled Staphylococcus aureus, Cowan I strain (SACI) (21) for 20 min on ice in precipitating buffer (0.5% Triton X-100, 0.5% sodium dodecyl sulfate [SDS], 0.14 M 70 NaCl, 0.01 M Tris-HCl, pH 7.4, 2 mM PMSF). The SACI was removed by centrifugation at 1500 x g for 10 min, and the cleared lysate reacted with 10 µI of specific rabbit anti mouse µ (M104E µ) antiserum for 1 hr on ice. Fifty µl of the washed SACI suspension were added and the mixture incubated for an additional 20 min on ice. Precipitates were washed three times with washing buffer (0.5% Triton X-100, 0.1 % SDS, 0.14 M NaCl, 0.01 M Tris-HCl, pH 7.4) and the precipitated molecules released by heating the SACI in boiling water for 5 min in 50 to 100 µl of the loading buffer appro!)l'iate to the gel analysis. Gel analyses. Analytical composite agarose-polyacrylamide gels were poured and run according to a modification of Dingman and Peacock (22). Gels were composed of 0.5 M urea, 0.2 M sodium phosphate, pH 7.2, 0.1 % SDS, 1% agarose, 2.5% acrylamide, and 1.2% bisacrylamide. The loading buffer contained 2% SDS and 0.05 M sodium phosphate at pH 6.8. Preparative 10% polyacrylamideSDS gels were prepared according to Laemmli (23) with a two centimeter stacking gel. The loading buffer contained 4% SDS, 0.05 M Tris-HCl, pH 6.8, 15% glycerol and 4% S-mercaptoethanol. Analytical two-dimensional gel analysis was performed according to O'Farrell (24). The loading buffer contained 1.5% pH 3.5-10 ampholines (LKB), 0.5% pH 4-6 ampholines (LKB), 9 M urea, 0.1 % SDS, 2% Triton X-100 and 5% S-mercaptoethanol. Gels were processed for fluorography by the method of Bonner and Laskey (25). , Peptide mapping. W279.1 µm chains were preparatively isolated by running immunoprecipitates of IgM molecules on 10% polyacrylamide-SDS gels. The µ chains were eluted from the appropriate 1 mm gel slices by incubation with 0.5 ml of 0.5% SDS at room temperature for 24 hr. W279.1 and MxW µs chains were isolated from culture medium by the same procedure. Carrier pig IgG (0.25 mg) was added to each eluted and pooled µ m or µ sample and the chains were reduced s and alkylated as follows. The pooled µ chains were lyophilized, redissolved in 71 9% SDS, 0.5 M Tris-HCl, pH 8.5, 0.01 M EDTA, boiled for 5 min, flushed with argon for 5 min, and dithiothreitol was added to a concentration of 20 mM. The samples were then boiled an additional 2 min, sealed under argon and incubated at 37°C for 1-1/2 hr. Alkylation with 50 mM iodoacetamide (3X recrystallized) was performed for 1 hr at room temperature in the dark. The reagents were removed by desalting on Sephadex G-25 (Pharmacia) equilibrated with 0.5% SDS and 0.05 M Tris-HCl, pH 7.5, and the proteins were precipitated with 25% trichloroacetic acid, washed twice with cold 20% trichloroacetic acid, twice with ethanol:ether (1:2) and once with ether. The precipitate was air-dried and dissolved in 200 µl 0.2 M ammonium bicarbonate, digested with TPCK-trypsin (Worthington) at room temperature (total of 150 µg for 22 hr), followed by digestion with chymotrypsin (Worthington) at 37°C (total of 200 µg for 28 hr). The samples were quick frozen and lyophilized to terminate the reaction, dissolved in 45 µl of 0.5 M phosphate buff er (pH 1.8):acetone (2:1) and the peptides separated by high performance liquid chromatography as described (26). RESULTS W279.1 cells synthesize three forms of µ chains - }l·L.1:!. . l m and µ . The S W279.1 cells synthesize three forms of µ chains which we have designated as µ. l (internal µ), µ m (membrane µ) and µ (secreted µ). These µ chains differ in their s cellular location, size, charge and level of glycosylation. W279.1 cells were radiolabeled by growth in 35 s-cysteine for 16 hr. The cells were solubilized in non-ionic detergent and the lysate was specifically immunoprecipitated with a rabbit antibody directed against mouse µ chains. When these precipitates are analyzed without reduction on composite agarose-polyacrylamide gels (Fig. lA), the largest species are about 200,000 daltons, the characteristic size of cell-surface monomeric ( µ L ) IgM molecules. The other major species 2 2 72 is about 100,000 daltons, the size appropriate for assembled ( µL) half molecules. A similar analysis of IgM molecules from the cell medium (Fig. lB) reveals that about half are assembled in the pentameric form. Thus, W279.1 cells clearly synthesize both membrane and secreted IgM molecules. In order to further characterize the µ chains, W279.1 cells were labeled 3 with H-tyrosine and leucine for 8 hr. IgM molecules from the cell lysate were immunoprecipitated, reduced, separated by two-dimensional gel electrophoresis and analyzed by fluorography. Two major species of µ chains are evident (Fig. 2). The two species have an apparent molecular weight difference of about 4,000. They also differ in isoelectric point; the smaller species encompasses a range of isoelectric points from pH 6.5 to pH 5.6 and the larger a range from pH 5.8 to pH 5.2. A similar two-dimensional gel of the W279.1 secreted IgM molecules revealed that the µs chains are similar in pl to their µm counterparts but are smaller by about 3,000 daltons (data not shown). All three µ chains exhibited a cyanogen bromide cleavage pattern characteristic of C µ region peptides derived from µ chains synthesized by the myeloma tumor M104E (27). Thus, W279.1 s cells contain three distinct types of µ chains which differ in molecular weight, charge and location. As is shown below, these correspond to the previously defined µ., µ 1 m and µ chains. s The larger µ chain is on the plasma membrane. The location of the two species of µ chains contained in the cell lysate was determined using two methods of radiolabeling. Intact cells were treated with neurarriinidase and galactose oxidase 3 followed by labeling with NaB H (20). The labeled cells were solubilized in non4 ionic detergent, and the IgM molecules were immunoprecipitated, reduced and analyzed on two-dimensional gels. Only the larger, more acidic µ chains were labeled under these conditions, showing that this species of µ chain is on the plasma membrane and contains galactose residues accessible to labeling with extrinsic 73 reagents (Fig. 3A). The unglycosylated light chain was not labeled and serves as an internal control (data not shown). The labeled cell-surface IgM was also analyzed without reduction on 5% polyacrylamide-SDS gels. All of the IgM molecules labeled under these conditions possess a molecular weight of 200,000 (Fig. 4), indicating that these µ chains are present on the plasma membrane as monomeric ( µ L ) IgM molecules. Radioiodination of intact cells with lactoperoxidase and 2 2 125 Na 1 (19) also labeled only the larger and more acidic species of µ chain, confirming their location on the cell surface (Fig. 3B). In contrast, radioiodination of a cell lysate, which destroys the protective barrier of the cell membrane, labels both µ species equally (data not shown). These observations suggest that the larger µ species is the membrane-bound µ chain ( µm) and that the smaller species is located within the W279.1 cells and represents an internal pool of µ chain ( µ.). l The W279.1 secreted IgM molecules contain µ chains. To identify the 5 type of µ chains in IgM molecules from the culture fluid of W279.1 cells, we compared the tyrosine-labeled tryptic and chymotryptic peptides of W279.1 µ chains from the medium with µs chains secreted by the MxW hybridoma cells. The cell fusion has induced a high level of pentameric IgM secretion by these hybridoma cells (18) and the IgM pentamers are composed of W279 µ chains (27). W279.1 5 3 and MxW cells were labeled with H-tyrosine for 16 hr, the µs chains isolated from the culture medium of both cells and from the lysate of W279.1 cells, and cleaved with trypsin and chymotrypsin as described in Materials and Methods. The resulting peptides were separated by high performance liquid chromatography (26). As shown in Fig. SA, the tyrosine peptides of the two secreted µ chains are identical. The IgM molecules found in the culture medium from W279.1 lymphoma cells are therefore composed of µs chains. In contrast, µm chains contain several distinct tyrosine peptides not found in µ chains (Fig. 5B). The µm chains therefore 5 differ in structure from µ chains. These differences could reflect carbohydrate s 74 differences, protein differences, or both. The important point is that the µ and s µm chains can be distinguished from one another structurally. The µi species contains precursors of µm and µs chains. In order to determine the relationship of the µ.1 chains to the µ m and µs chains, we pulselabeled W279.1 cells with 35 s-methionine for 15 minutes, chased the label by washing and resuspending the cells in unlabeled medium containing a 10-fold excess of unlabeled methionine and examined the distribution of radiolabel in the µ. and 1 µm pools for 4 hr after the pulse. Figures 6A-6F display the portion of each twodimensional gel which contains the µ chains, aligning the gels on a small invariant spot (indicated with an arrow). Figure 6A shows that during the 15 min pulse, only the µ. pool is labeled. Moreover, these gels show that there are two sizes 1 of µ chains within the µ. pool. The two species of µ. chains are equally labeled 1 1 in 15 min (Fig. 6A). Over the 4 hr chase period, the smaller species progressively disappears (Figs. 6B-6F). Some of the label in the µ. pool is chased into µ 1 m chains (Fig. 6B-6F). By this method, it is not possible to determine directly the fate of the two species of the µi pool. However, it is clear that the µi pool must contain precursors to the µm chain. During a short pulse and chase, only small amounts of IgM molecules are found in the medium. However, when the chase period is extended, labeled secreted IgM molecules are easily isolated. When cells are incubated for 1 hr in 35 s-methionine, mainly the µ. pool is labeled, but the two sizes within the pool are difficult 1 to discern (Fig. 7A). As before, a portion of the label in the µ. pool is chased 1 into the µ m chains and remains as µ m chains over long periods (Fig. 7 A-7E). At the same time, label from the µ. pool chases into µ chains in the medium 1 S (Fig. 7F), demonstrating clearly that precursors to both µm and µs chains are present in the intracellular pool. 75 Tunicamycin studies demonstrate that two species of µ polypeptide chains are synthesized by W279.1 cells. In order to further characterize the µ polypeptide chains from W279.1 cells, proteins were labeled by incubation of the 35 cells in s-methionine in the presence and absence of tunicamycin, an antibiotic which inhibits glycosylation of asparagine residues (28, 29). When IgM molecules were immunoprecipitated from lysates of tunicamycin-treated cells and analyzed on 10% polyacrylamide-SDS gels, two bands in the size appropriate for unglycosylated µ chains, 67,000 daltons and 69,000 daltons, were observed (Fig. 8, lane C). The precipitates shown are from cells incubated with 5 µg/ml of tunicamycin. Identical results were obtained with either 1.5 µg/ml or 0.5 µg/ml of tunicamycin, indicating that the two bands are not an artifact of tunicamycin toxicity to cells. Since W279.1 cells synthesize both membrane and secreted IgM molecules, this observation suggests that the two size classes represent the two species of µ chain, µ and µ . This interpretation is further supported by analysis of the s m mRNA isolated from W279.1 cells and other lymphoma cells (30, 31). Cytoplasmic poly A+ mRNA was isolated, sized on agarose gels and hybridized to a clonedµ cDN A. Two classes of µ chain mRN A were found which differed in size (2. 7 and 2.4 kilobases) and appeared to encode the µ m and µ chains, respectively (30). s When W279.1 cells were labeled for short periods in the absence of tunicamycin, two size classes of µ chains could be identified (Fig. 8, lane D). These two bands correspond to the smaller, more basic µ chains seen on two-dimensional gels (Figs. 2 and 6A). When W279.1 cells are labeled in the presence of tunicamycin, IgM molecules can also be isolated from the medium. These unglycosylated µ chains were isolated and their size analyzed. The smaller of the two bands was present in proportionally greater amounts than the larger one (Fig. 8, lane B). Since the pentameric secreted form of IgM predominates in medium from these cells, it 76 appears likely that the smaller form is the µ chain ~orresponding to secreted IgM molecules. This idea is strengthened by the fact that the µ chains from tunicamycintreated M104E cells correspond in size to the smaller of the two bands in W279.1 cells (Fig. 8, lane A). Taken together, these results suggest that the larger band represents unglycosylated µ m chains and the smaller band represents unglyco- syla ted µs chains. W279 cells have differentiated while in culture and thus exhibit different stages in the B-cell differentiation pathway. We initially began characterizing the original uncloned W279 cell line. Because these W279 cells exhibited poor growth characteristics with a relatively low degree of overall cell viability, we obtained a later subclone, W279.1, which exhibited greater proliferative potential. The original W279 cells were initially characterized with respect to the size of the IgM molecules the cells synthesized. W279 cells were radiolabeled by growth 3 in H-tyrosine and leucine for 16 hr and the IgM molecules precipitated from cell lysates and culture medium. These precipitates were analyzed without reduction on composite agarose-polyacrylamide gels (Fig. 9). The two major species of IgM molecules isolated fr.om W279 cell lysates were similar in size and quantity to those isolated from W279.1 cells (compare Figs. 9A and lA). Equivalent quantities of cell lysates and medium (in cell numbers) were analyzed in Figs. 1 and 9. In comparison to IgM molecules isolated from cell lysates, IgM molecules in the medium from W279 cells were present in negligible amounts (compare Figs. 9B and lB). At least 15-fold less IgM was secreted into the medium by W279 than by W279.1 cells. Thus, the greater level of secreted pentameric IgM molecules in the W279.1 clone suggests that this clone has differentiated in culture to become a more mature B cell than the parental W279 cells. 77 DISCUSSION W279.1 cells synthesize three species of µ chains, µp_.l!m and µs' which differ in their cellular locations. The galactose oxidase and lactoperoxidase labeling experiments on intact W279.1 cells suggest that the µ m chains are integral membrane proteins inserted into the B-cell plasma membrane and assembled as monomeric ( µ 2 L2) IgM molecules of 200,000 daltons. The µm chains appear to be fully glycosylated by virtue of the fact they have terminal galactose residues 3 that can be labeled with galactose oxidase and NaB H4. The W279.1 cells secrete IgM molecules into the medium, and 50% of these are assembled as secreted IgM pentamers. The µs chains from these secreted molecules can readily be distinguished from µ m chains by comparative peptide map analysis. The W279.1 cells have an intracellular pool of µ chains ( µ.) whose complex l carbohydrate moieties lack their terminal galactose residues. Furthermore, the µ. chains have a more basic pl than their µ and µ counterparts which probably m S reflects the lack of charged sialic acid residues on the µ. chains. Thus, the µ. l l l chains appear to be lacking the terminal sugars of their complex carbohydrates. Pulse-chase studies with precursors to both the µ 35 m s-1abeled methionine suggest that the µ. pool contains l and µ chains. Following extended labeling, the majority s of the µ. chains appear to be precursors to µ chains in that structural comparisons l S of peptides suggest that the predominant species of µ. chains are very similar l to the µs chains (27). The µ chains are glycosylated in two stages. The steps required for asparagine linked glycosylation of complex carbohydrates in proteins have been carefully documented (32, 33, 34, 35). The process occurs in two distinct phases. A mannose-rich carbohydrate core is transferred en bloc from dolicholphosphate 78 to the asparagine residue in a recognition sequence Asn-X-Ser or Thr. This mannoserich core is partly degraded and then reformed by the addition of neutral sugars (N-acetyl glucosamine, galactose and fucose) and charged sialic acid. The net effect of the second stage of glycosylation is to increase the apparent molecular weight of the glycoprotein in the presence of SDS and, because of the sialic acid, to decrease its pl value. Our analysis of W279.1 µ chain synthesis and processing suggests that the complex carbohydrates in µ chains are glycosylated according to this-scheme. First, the µ. pool is labeled when W279.1 cells are incubated in 1 3 H-mannose (27). Because µ chains have only asparagine-li~ked carbohydrate moieties (16), this observation suggests that a mannose-rich core has been attached 3 to the µi chains. Second, when a lysate of W279 cells was labeled with NaB H 4 and galactose oxidase after treatment with neuraminidase, µ chains but not the µ. m 1 chains were labeled (data not shown), indicating that the complex carbohydrates attached to µ. chains contain no galactose residues. This observation and the 1 pl difference between µ. and µ chains suggests that the terminal sugar moieties 1 S have not yet been added to the mannose-rich cores in µ. chains. Thus, µ. chains 1 1 have undergone the first step but not the second step of complex carbohydrate synthesis. Moreover, µ chains synthesized in the presence of tunicamycin, a specific inhibitor of the synthesis of oligosaccharide-dolichol-phosphate (28, 29), incorporate less than 10% of the mannose found in control µ chains (C. Sibley, in preparation), have the same range of isoelectric points as µi chains (data not shown) but are lower in apparent molecular weight by about 6000 (Fig. 8). This is exactly what one would predict if the µ. chains lacked the terminal sugar residues 1 of mature complex carbohydrate moieties but contained the mannose core sugars. Taken together with the pulse-chase and labeling experiments, these results are consistent with the idea that µ chains are glycosylated during synthesis by addition of core carbohydrate and that the progression from the µi pool to 79 µm and µs chains represents the addition of the terminal neutral and charged sugars on the complex carbohydrate structures. If this is correct, the fate of the two size classes from the µ. pool should differ. Both forms should become l ~ larger in apparent molecular weight and more acidic as glycosylation proceeds from the first to the second stage. However, the µ m precursor should remain cell-associated even after it is fully glycosylated. By analogy with other immunoglobulins, the secreted form should be externalized very rapidly after the second stage of glycosylation is completed (36). Our biochemical analyses of the larger, acidic form isolated from cells suggest that it does consist principally of the membrane-bound µ chain (27). This reasoning explains why little of the secreted form remains in the cell after its glycosylation is completed. Therefore, the largest µ chain associated with W279.1 cells after a long labeling period represents µ m chains. The µm chain is 2000 daltons larger than the µ5 chain. Two lines of evidence suggest that this molecular weight difference reflects polypeptide sequence differences and not differing levels of glycosylation. First, tunicamycin blocks the addition of asparagine-linked carbohydrate moieties to polypeptide chains. Mu chains from W279.1 cells synthesized in the presence of tunicamycin consist of two size classes that differ from one another by 2000 daltons. Second, 3 H-tyrosine-labeled µsand µm chains compared by peptide mapping after digestion with trypsin and chymo~rypsin reveal µs and µm chains have several different tyrosine peptides. These data suggest the µ m and µ chains differ in their primary s amino acid sequence. A final point is that W279.1 cells synthesize two size classes of µ chain mRN A, 2. 7 kilobases and 2.4 kilobases, which apparently synthesize µm and µs chains, respectively (30). These two types of µ chain mRN As are identical except for the portion of the mRN As that encodes the C-terminal segment of the µ chain. These coding differences are consistent with a µ 2000 daltons larger than the µ5 chain. m chain approximately 80 The W279 and W279.1 cells represent lymphocytes at different stages of B-cell differentiation. The original W279 cells clearly secrete little, if any, IgM molecules. In contrast, the W279.1 clone appears to secrete significant amounts of pentameric IgM. Dr. Warner and his colleagues have demonstrated that increased section of pentameric IgM molecules correlates with the more advanced stages of B-cell differentiation (6) (Fig. 10). This observation suggests that the W279.1 clone is composed of a more mature B cell population than the original uncloned W279 cells. This supposition is further confirmed by the observation that in comparison to the original W279 cell line, W279.1 cells possess increased concentrations of Ia antigens and PCA (plasma cell antigen) and decreased concentrations of membrane IgM (M. Daley and N. Warner, personal communication). These features also are characteristic of more mature B cells (6). A question which then arises is whether W279.1-like cells are present in the uncloned W279 population or has the differentiation of W279 cells occurred over a period of two years in culture? The W279 cells were originally isolated, characterized and later cloned in Dr. Warner's laboratory. A comparison of the original W279 cells with the later W279.l cells, which were cloned and grown under different culture conditions, reveals that the W279.1 cells secrete significantly more pentameric IgM molecules and contain fewer membrane IgM molecules than the originally isolated W279 cells. Commensurate with their ability to secrete significantly increased amounts of pentameric IgM molecules, W279.1 cells also have 100-fold greater levels of J chain than their W279 counterparts (E. Mather and M. Koshland, personal communication). Accordingly, these two isolates of W279 cells appear to have differentiated with respect to one another. We conclude that the differences in culture conditions have caused the W279.1 cells to differentiate. Similar observations have been made for myeloid tumors (37-39). 81 These observations suggest that we have analyzed W279 cells at three distinct stages of differentiation from a B lymphocyte to a mature plasma cell (Fig. 10). A second B-cell lymphoma, WEHI 231, appears to have all of the characteristics of original W279 cells, namely both represent an early stage of B-cell differentiation. The W279.1 cells have differentiated to represent a later intermediate stage, and the myeloma M104E or MxW cells represent the terminal stage of B-cell differentiation. Thus the differentiation process may be induced in a single B-cell lymphoma. Perhaps these stages of B-cell tumors will be useful in delineating the regulatory mechanisms that mediate the shift in synthesis of µm to µs chains. 82 REFERENCES 1. Warner, N. 1974. Membrane immunoglobulins and antigen receptors on Band T lymphocytes. Adv. in Immunol. 19:67. 2. Burrows, P., M. Le Jeune, and J. F. Kearney. 1979. Evidence that murine pre-B cells synthesize µ heavy chains but no light chains. Nature 280:838. 3. Vitetta, E. S., S. Baur, and J. w. Uhr. 1971. Cell surface immunoglobulin. II. Isolation and characterization of immunoglobulin from mouse splenic lymphocytes. J. Exp. Med. 134:242. 4. Della Corte, E., and R. M. E. Parkhouse. 1973. Biosynthesis of immunoglobulin A (IgA) and immunoglobulin M (IgM); Requirement for J chain and a disulphide exchanging enzyme for polymerization. Biochem. J. 136:597. 5. Anderson, J., J. Buxbaum, R. Citronbaum, S. Douglas, L. Forni, F. Melchers, B. Pernis, and D. Stott. 1974. IgM-producing tumors in the BALB/c mouse. J. Exp. Med. 140:7 42. 6. Warner, N. L., J. F. Leary, and S. McLaughlin. 1979. Analysis of murine B cell lymphomas as models of B-cell differentiation arrest. In B Lymphocytes in the Immune Response. Edited by M. Cooper, D. E. Mosier, I. Scher, and E. S. Vitetta. Elsevier-North Holland, Inc., New York. P. 371. 7. Yuan, D., J. W., Uhr, and E. S. Vitetta. 1980. A peptide difference between the µ chains from cell-associated and secreted IgM of the BCL tumor. 1 J. Immunol. In press. 8. Bergman, Y ., and J. Haimovich. 1978. B Lymphocytes contain three species of µ chains. Eur. J. Immunol. 8:876. 9. Williams, P. B., R. T. Kubo, H. M. Grey. 1978. µ-chains from a non-secretor B cell line differ from secreted µ-chains at the C-terminal end. J. Immunol. 121:2435. 83 10. Melcher, U., and J. W. Uhr. 1977. Density differences between membrane and secreted immunoglobulins of murine splenocytes. Biochemistry 16:145. 11. Vassalli, P., R. Tedghi, B. Lisowska-Bernstein, A. Tartakoff, and J.-c. Jaton. 1979. Proc. Natl. Acad. Sci. USA 76:5515. 12. Parkhouse, R. M. E., J. Riffer, and Y. S. Choi. 1980. Chemical characterization of the Fab and Fe fragments from surface immunoglobulin. Nature. In press. 13. Singer, P. A., H. H. Singer, and A. R. Williamson. 1980._ Nature. In press. 14. Fu, S. M., and H. G. Kunkel. 1974. Membrane immunoglobulins of B lymphocytes: Inability to detect certain characteristic IgM and IgD antigens. J. Exp. Med. 140:895. 15. Vitetta, E. S., and J. W. Uhr. 1974. Cell surface immunoglobulin. IX. A new method for the study of synthesis, intracellular transport, and exteriorization in murine splenocytes. J. Exp. Med. 139:1599. 16. Kehry, M., C. Sibley, J. Fuhrman, J. Schilling, and L. E. Hood. 1979. Amino acid sequence of a mouse immunoglobulin µ chain. Proc. Natl. Acad. Sci. USA 76:2932. 17. Warner, N., A. Harris, and G. Gutman. 1975. In Membrane Receptors of Lymphocytes. Edited by M. Seligman, J. L. Pred'homme, and F. M. Kourilsky. Am. Elsevier, New York. P. 203. 18. Raschke, W. c., E. L. Mather, and M. E. Koshland. 1979. Assembly and secretion of pentameric IgM in a fusion between a nonsecreting B cell lymphoma and an IgG-secreting plasmacytoma. Proc. Natl. Acad. Sci. USA 76:3469. 19. Marchalonis, J. J., R. E. Cone, and V. Santer. 1971. Enzymatic iodination: A probe for accessible surface proteins of normal and neoplastic lymphocytes. Biochem. J. 124:921. 84 20. Gahmberg, C. G., and S. Hakomori. 1973. External labeling of cell surface galactose and galactosamine in glycolipid and glycoprotein of human erythrocytes. J. Biol. Chem. 248:4311. 21. Kessler, S. W. 1976. Cell membrane antigen isolation with the staphylococcae protein A-antibody adsorbant. J. Immunol. 117:1482. 22. Dingman, C. W., and A. C. Peacock. 1968. Analytical studies on nuclear ribonucleic acid using polyacrylamide gel electrophoresis. Biochemistry 7:659. 23. Laemmli, U. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680. 24. O'Farrell, P. H. 1975. High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250:4007. 25. Bonner, W. M., and R. A. Laskey. 1974. A film detection method for tritiumlabeled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem. 45:83. 26. McMillan, M., J. M. Cecka, L. Hood, D. B. Murphy, and H. O. McDevitt. 1979. Peptide map analyses of murine Ia antigens of the I-E subregion using HPLC. Nature 277:663. 27. Kehry, M., S. Ewald, R. Douglas, C. Sibley, W. Raschke, D. Fambrough, and L. Hood. 1980. The immunoglobulin µ chains of membrane-bound and secreted IgM molecules differ in their C-terminal segments. Cell. Submitted. 28. Kuo, S.-c., and J. O. Lampen. 1974. Tunicamycin: an inhibitor of yeast glycoprotein synthesis. Biochem. Biophys. Res. Commun. 58:287. 29. Takatsuki, A., K. Kohno, and G. Tamura. 1976. Inhibition of biosynthesis of polyisoprenol sugars in chick embryo microsomes by tunicamycin. Agr. Biol. Chem. 39:2089. 85 30. Rogers, J., P. Early, C. Carter, K. Calame, M. Bond, L. Hood, and R. Wall. 1980. Two mRNAs with different 3' ends encode membrane-bound and secreted forms of immunoglobulin µ chain. Cell. In press. 31. Perry, P. P., and D. E. Kelly. 1979. Immunoglobulin messenger RNAs in murine cell lines that have characteristics of immature B lymphocytes. Cell 18:1333. 32. Sturgess, J., M. Moscarello, and H. Schacter. 1978. The structure and biosynthesis of membrane glycoproteins. Curr. Topics in Membranes and Transport 11:15. 33. Kornfeld, R., and S. Kornfeld. 1976. Comparative aspects of glycoprotein structure. Ann. Rev. Biochem. 45:217. 34. Waechter, C. J., and W. J. Leunarz. 1976. The role of polyprenol-linked sugars in glycoprotein synthesis. Ann. Rev. Biochem. 45:95. 35. Robbins, P. w., S. C. Hubbard, S. J. Turco, and D. F. Wirth. 1977. Proposal for a common oligosaceharide intermediate in the synthesis of membrane glycoproteins. Cell 12:893. 36. Parkhouse, R. M. E. 1973. Assembly and secretion of immunoglobulin M (IgM) by plasma cells and lymphocytes. Transpl. Rev. 14:131. 37. Lotem, J., and L. Sachs. 1977. Control of normal differentiation of myeloid leukemia cells. XII. Isolation of normal myeloid colony-forming cells from bone marrow and the sequence of differentiation to mature granulocytes in normal and D+ myeloid leukemic cells. J. Cell. Physiol. 92:97. 38. Sachs, L. 1977. Control of normal cell differentiation in leukemia. Israel J. Med. Sci. 13:654. 39. Maeda, S., and L. Sachs. 1978. Control of normal differentiation of myeloid leukemic cells. xm. Inducibility for some stages of differentiation by dimethylsulfoxide and its disassociation from inducibility by MGI. J. Cell. Physiol. 94:181. 86 Figure 1. Composite agarose-polyacrylamide gel electrophoresis of IgM molecules synthesized by W279.1 cells. Cells were radioactively labeled by incubation in 35 s-cysteine. IgM was isolated without reduction from medium and lysates. Its size was assessed on composite agarose-polyacrylamide gels as described in the Materials and Methods. Arrows indicate the migration. of molecular weight standards 6 run in parallel gels: 104E, MOPC 104E pentameric IgM of 1 x 10 daltons; IgG, pig y globulin of 150,000 daltons; BSA, bovine serum albumin of 65,000 daltons. A. Cell lysate from W279.1 cells. The molecular weights of the two predominant • species are 200,000 and 100,000, corresponding to ( µ 1 ) IgM monomers and ( µL) 2 2 half monomers, respectively. B. Medium from W279.1 cells. 87 A WEH1279. I Lysote anti µ. 10 I04E IgG BSA ! ~ 8 ! 6 4 ,..., b 2 X Dye r-1--, ~ Cl. 0 u I g, ,..., B 8 I04E IgG t t WEH1279.I Medium antiµ. BSA ~ 6 4 2 Dye ~ 0 10 20 30 40 50 60 70 80 90 Slice Number FIGURE 1. 88 Figure 2. Two-dimensional gel analysis of µ chains from a W279.1 cell lysate. 3 W279.1 cells were radiolabeled with H-tyrosine and leucine and analyzed on twodimensional gels as described in Materials and Methods. Light indicates light chain. The 50,000 dalton molecules can be eliminated by washing the immunoprecipitates with a buffer containing 1% deoxycholate, 1% Triton X-100 and 0.15% sos. 89 acidic basic }1-,m }1-,j 1/) 1/) Q) ~ 0 .c a. 0 ~ u Q) l • I a . .... :: . ~ light FIGURE 2. ;._ .... -~ : Q) I •· 90 Figure 3. Two-dimensional gels of extrinsically labeled W279.1 cells. 3 A. Intact W279.1 cells were labeled with NaB H and galactose oxidase (20), 4 cell lysates prepared, immunoprecipitated and analyzed on two-dimensional gels. B. Intact W279.1 cells were labeled with Na 125 I and lactoperoxidase, cell lysates prepared, immunoprecipitated and analyzed by two-dimensional gel electrophoresis as described in Materials and Methods. Only the portion of the gel containing • µ chains is shown in each case. 91 B ··"?-::,· r : · • • "',:·. \~_· bt ·'.: i If~.. FIGURE 3. 92 Figure 4. SDS-polyacrylamide gel analysis of membrane IgM molecules. 3 Intact W279.1 cells were labeled with NaB H and galactose oxidase (20). IgM 4 molecules were isolated from cell lysates without reduction and the size assessed on 5% polyacrylamide gels as described in Materials and Methods. 7S IgM marks the position of the 200,000 dalton monomeric IgM molecule. FIGURE 4. 100 200 300 3H-cpm 400 10 20 30 + 40 50 60 Slice Number 7SigM 70 80 90 ! Dye (0 t,.) 94 Figure 5. Comparison of tyrosine peptides from W279.1 µ , µ and MxW µ chains. s m s Mu chains were isolated from the culture medium of W279.l and MxW cells and from the lysates of W279.1 cells incubated for 16 hr in the presence of 3 H-tyrosine. Reduction, alkylation, and digestion with trypsin and chymotrypsin were performed as described in Materials and Methods. Peptides were separated on a DuPont ODS C-18 high performance liquid chromatography column (26) and 0.5 min (0.5 ml) fractions were collected, dried and counted for radioactivity. Peptide differences are crosshatched. ( ..... ) gradient of acetone used for elution of the peptides. A. (--) MxW µ chains; (----) W279.1 µ chains. B. (--) MxW µs chains; s s (----) W279.1 µ m chains. 95 .......... iua .... 10s ::llU06J0 0 0 0 co 0 ~ l.O 0 N % 0 ---- £-01 x V'JdJ-H£ q -:t: N l.O co N ci · N 'st d 0 0 co a:: >r- 0 ~ 0 ::t 0 ----~-- ----------- '- (l) £::! .D 0 0 z E ::::, C 0 co 0 u 0 '- 0 l.O 1/) 1/) :::t. :::t. 3 X ~ q £::! 0 en 'st I'- N 3 0 N 0 Q 0 a:i q l.O 0 0 <i- -£_01 x V'Jd0-H£ N LJ_ 96 0 0 0 ----£_QI X V'Jd0-H£ q 0 r0 0 N 0 a::: 0 CE >- f- 0 ~ ', ·• .. 0 <:!' 0 N - .... (l) .D 0 E 0::, -z C o.2 Cl) -(.) ....0 oLJ.. <.!) E "' ::1.. ::1..- 0 3 ~ <:!' X N :::E3 0 N 0 d 0 a5 0 0 q N 0 0 97 Figure 6. Two-dimensional gel analysis of a short pulse chase of the µi pool into µ m in W279.1 cells. W279.1 cells were labeled for 15 min with 35 s-methionine and chased for 4 hr by resuspending them in growth medium without radiolabel. Samples were taken at the indicated times for lysis and immunoprecipitation analysis of radioactive µ chains remaining in the cells. labeling period, t = O. A. At the end of the B. After 30 min in growth medium. D. After 1 hr. E. After 2 hr. F. After 4 hr. The µ. and µ l C. After 45 min. m chains are designated on gel A. Gels were aligned with reference to the invariant spot designated with ti. Only that region of the gel containing µ chains is shown. 98 A /1-m t=O B } .• .~~min f I ' I :ii '~ -~.. ~';,.,.• :,.~;-~ :-~: -~~ - ·-· · • : :_. . • • A t"' 2 hr • t • ~ · F t: 4hr FIGURE G. .... ,,,A 99 Figure 7. Two-dimensional gel analysis of a long pulse chase of µi into µm and µ chains in W279.1 cells. W279.1 cells were labeled for 1 hr with 35 s-methionine s and chased for 20 hr by resuspending them in growth medium without radiolabel. Samples were taken for lysis and immunoprecipitation analysis of radioactive µ chains remaining in the cells or in thl medium. A. Cell lysate at the end of labeling period, t = O. B. Cell lysate after 4 hr. C. Cell lysate after 6 hr. D. Cell lysate after 8 hr. E. Cell lysate after 20 hr. F. Radioactive µ chains found in the medium after a 20 hr chase. The µ. and µ chains are designated on gel m A. Gels were aligned with reference to the invariant spot designated with 15.. 1 Cell equivalent amounts of lysate were used for the analysis at each timepoint. Only that region of the gel containing µ chains is shown. 100 ~~,~~ i r--...;.- f ' F t =20hr FIGURE 7. JJ 1 Figure 8. SDS-polyacrylamide gel analysis of µ chains synthesized by W279.l and Ml04E cells in the presence and absence of tunicamycin. Ml04E and W279.l cells were radiolabeled with 35 s-methionine in the presence or absence of tunicamycin. Samples were immunoprecipitated and analyzed on 10% SDS polyacrylamide slab gels as described in Materials and Methods. Lane A: Cell lysate from M104E cells labeled for 4 hr in the presence of 5 µg/ml tunicamycin. Lane B: Medium from W279.l cells labeled for 4 hr with 5 µg/ml tunicamycin. Lane C: Cell lysate from W279.l cells labeled for 4 hr with 5 µg/ml tunicamycin. Lane D: Cell lysate from W279.l cells labeled for 15 min without tunicamycin. Lane E: Cell lysate from W279.l cells labeled for 4 hr without tunicamycin. Arrows indicate the position of fully glycosylated Ml04E µ chains. The unglycosylated µ chain bands s in lanes B and C are not aligned since the lanes are from different polyacrylamide gels which each contained internal control lanes. 102 w - .. i . ............ . 0 u ' t CD I L1 t w V ::l. 0 . 00 ~ 0::: ;::) c., ~ 103 Figure 9. Composite agarose-polyacrylamide gel analysis of IgM molecules synthe3 sized by original W279 cells. Cells were labeled by incubation in H-tyrosine and leucine. Isolation and gel electrophoresis are as described in the legend to Fig. 1 and in Materials and Methods. A. Cell lysate from W279. B. Medium from W279. Equivalent quantities (in cell numbers) of cell lysate and medium were analyzed here and in Fig. 1. 104 A WEHI279 5 Lysate anti µ. I04E IgG BSA + + 4 + 3 2 rt) 'o X Dye ,-J--, ~ a.. u WEH1279 Medium anti µ. B I I rt) I04E IgG BSA ! i i 1.5 1.0 0.5 Dye ,-h 0 10 20 30 40 50 60 70 80 90 100 Slice Number FIGURE 9. 105 Figure 10. Differentiation stages of IgM secreting cells. The schematic drawing illustrates the various stages that have been characterized in the development of stem cells to mature plasma cells. Internal µ chains and IgM molecules, membrane IgM monomers and secreted IgM pentamers are illustrated. The different circle diameters used to represent the pre-B cell and the plasma cell indicate that the cells of these two stages are larger than the others. The B-cell lymphomas W279 and W231 correspond to a nonsecreting B cell. The W279.l lymphoma cells represent an intermediate IgM-secreting B cell stage, and M104E and MxW cells correspond to terminally differentiated plasma cells containing little if any membrane IgM molecules. FIGURE 10. MxW MI04E ~ / ~-* W279.I Stem Cell Pre-8 Cell 8 Cell B Cel I Plasma Cel I O-0~~-B-G--~ * "' * ~ W279 W231 I--' 0 0) 107 CHAPTER 4 108 The Immunoglobulin µ Chains of Membrane-Bound and Secreted IgM Molecules Differ in their C-Terminal Segments • M. Kehry , S. Ewald ••, R. Douglas•, C. Sibley••• + ++ • W. Raschke , D. Fambrough and L. Hood *Division of Biology, California Institute of Technology, Pasadena, California 91125. ** Department of Microbiology, Montana State University, Bozeman, Montana 59717. *** Department of Genetics, University of Washington, Seattle, Washington 98195. +The Salk Institute for Biological Studies, P.O. Box 85800, San Diego, California 92138. ++The Carnegie Institution of Washington, Baltimore, Maryland 21210. 109 Summary The B lymphocyte synthesizes two forms of IgM molecules during its development from a stem cell to a mature antibody-secreting plasma cell. The monomeric receptor IgM molecule is affixed to the plasma membrane and serves to trigger the later stages of B-cell differentiation, whereas the pentameric secreted IgM molecule is an effector of humeral immunity. The structural differences between membrane-bound and secreted IgM molecules are reflected in the differences between their heavy or mu chains. We have previously determined the complete amino acid sequence of a murine secreted mu(µ s) chain (Kehry et al., 1979). In this study, we have compared the structures of the secreted and membrane-bound mu(µ m) heavy chains by peptide mapping, microsequence, and carboxypeptidase analyses. These studies demonstrate that the µ m and µ chains are very similar s throughout their VH' C 1, C 2, C 3 and C 4 domains. The µ and µ chains µ µ µ µ m s differ in the amino acid sequence of their C-terminal segments. These studies in conjunction with those carried out on the µ m and µs mRNAs (Rogers et al., 1980) and the C µ gene (Early et al., 1980) suggest that the µ m and µ chains s from a given B cell are identical but for their 41 and 20 residue C-terminal segments, respectively. The amino sequence of the 41 residue C-membrane terminal segment predicted from the correspondingµ m mRNA is in perfect agreement with all of the protein studies reported in this paper with the exception that the carboxyterminal amino acid, lysine, is posttranslationally removed from the mature µ m polypeptide. Thus, these distinct C-terminal segments are produced by RN A splicing from a single nuclear transcript. 110 Introduction In the maturation of B lymphocytes to antibody-secreting plasma cells, discrete stages of differentiation may be identified (Warner, 1974). The pre-B cell synthesizes only internal mu chains (Burrows, Le Jeune and Kearney, 1979). In addition to other changes, induction of the synthesis of light chains accompanies the formation of the small B lymphocyte containing membrane IgM molecules as integral membrane receptors (Vitetta, Baur and Uhr, 1971). Additional steps of differentiation and proliferation then occur which form a clone of IgM secreting plasma cells. Thus, the IgM molecule can exist as a membrane-bound receptor on the surf ace of B lymphocytes or is secreted by plama cells as a hydrophilic serum antibody. The secreted IgM molecule is a pentamer composed of five monomeric IgM subunits plus a joining (J) chain ((µ L \Jl (Della Corte and Parkhouse, 1973) 2 2 whereas the membrane-bound IgM molecule is a monomer(µ L ). We were interested 2 2 in characterizing the structural features which distinguish these alternative forms of a single class of antibody molecules. Since the µ heavy chain affixes the IgM molecule to the plasma membrane, the first step in the comparison of the membrane-bound (µ m ) and secreted (µ ) chains was the determination of the complete s covalent structure of a µ chain synthesized by the mouse plasmacytoma MOPC s 104E (Kehry et al., 1979). This study revealed that the µs chain has five homology units or domains, one variable (V) and four constant (C) region domains, (V H' C 1, C 2, C 3 and C 4) followed by a nonhomologous hydrophilic C-terminal µ µ µ µ segment of 20 amino acid residues. Several lines of evidence suggest that the µ m and µ s chains differ in structure. IgM molecules are solubilized from the cell surface only by detergents (Melcher, Eidels and Uhr, 1975), and therefore are by definition integral membrane proteins. Detergent binding studies have shown that membrane IgM molecules 111 and isolated µ m chains bind small but significant amounts of detergent whereas secreted IgM and µ s chains bind negligible amounts (Melcher and Uhr, 1977; Vassalli et al., 1979; Parkhouse, Lifter and Choi, 1980). Thus, µ m chains possess a struc- tural region capable of interacting with the hydrophobic plasma membrane. Molecular weight comparisons of µ m and µ s chains by SDS-polyacrylamide gel electrophoresis and by ultracentrifugation have shown that the µ m chain is larger than the µs chain by approximately 1500 daltons (Melcher and Uhr, 1973; Bergman and Haimovich, 1978). Although µ m and µ are glycoproteins, µ and µ chains s m s synthesized in the presence of tunicamycin are devoid of carbohydrate. Analyses of these nonglycosylated molecules indicate that the molecular weight difference is due to differences in polypeptide structure (Vassalli et al., 1979; A. Williamson, personal communication; J. Haimovich, personal communication; Sibley et al., 1980). Translation of µ m and µ s mRN A derived from B lymphoma cells also results in the synthesis of a µ m polypeptide which is larger in molecular weight than the µ s polypeptide (A. Williamson, personal communication; J. Haimovich, personal communication). In addition, there seem to be carbohydrate differences between µ m and µ s chains (Bergman and Haimovich, 1978; Bergman, Haimovich and Melchers, 1977). At least some of the above differences are localized to the COCH-terminal region of µ m and µ s chains. This localization has been demonstrated serologically by antibody competition experiments (Fu and Kunkel, 197 4) and structurally by peptide mapping of Fe fragments (Yuan, Uhr and Vitetta, 1980) and by treatment of µ m chains with carboxypeptidase (Williams, Kubo and Grey, 1978). Peptide mapping studies on µ m and µ s chains derived from a single cell line (Yuan, Uhr and Vitetta, 1980) and on µ m and µ s chains from a B-cell lymphoma and the IgM-secreting hybrid of the B lymphoma with a plasmacytoma (Raschke, Mather and Koshland, 1979) have indicated that structural differences between µ m and µ s chains are small. However, there has been disagreement on whether 112 the COCH-terminal amino acids of µsandµ m chains are different or identical (A. Williamson, personal communication; Bergman and Haimovich, 1978; Mcllhinney, Richardson and Feinstein, 1978). Previous studies on µ and µ mRN As and the C gene (Rogers et al., s m µ 1980; Early et al., 1980) report that the µ m and µ s chains are identical in their four C µ domains but differ in their C-terminal segments. In this paper we present a structural analysis on three types of µ chains derived from a cell line. The mouse B-cell lymphoma, WEHI 279 (W279), synthesizes two species of µ chains, µ m and µ s' in addition to a µ internal form ( µ i) which contains precursors of both µ m and µ s chains (Sibley et al., 1980). These W279 µ chains contain identical variable (V) regions. Fusion of W279 cells with the myeloma cell line MPC 11 generates MPCll x W279.2 (MxW) hybrid cells that secrete pentameric IgM molecules composed of W279 µ chains. These three species of W279 µ chains, µ ., µ and µ , s 1 m s differ in molecular weight and charge. The definition of their structural differences and similarities at the carbohydrate and protein levels is the subject of the present report. Results Localization of Three Species of W279 µ Chains Three species of µ chains can be defined in W279 lymphoma and MxW hybridoma cells based on cellular location: µ m (membrane), µ. (internal) and µ (secreted). 1 s Two different cell-surface labeling techniques have demonstrated that µ m chains are present on the plasma membrane of W279 cells as a 200,000 dalton IgM monomer (Sibley et al., 1980). In contrast, the MxW hybridoma cells have very low levels of cell-surface µ chains and secrete large quantities of IgM pen tamers (W. Raschke, unpublished observations). Both W279 and MxW cells contain an internal pool of µ chains (µ .). In MxW cells, the µ. pool consists only of precursors to µ chains, 1 1 S 113 while pulse-chase experiments indicate that the µ. pool of W279 cells contains 1 precursors for both µ m and µ chains (Sibley et al., 1980). In long-term (16 hr) s "steady state" labelings (shown below), the µ. pool in W279 cells is predominantly 1 composed of pre-µ s chains. These µ chains can be distinguished by several chemical criteria. The pl of the µ chains is similar to that of µ s m chains and both are markedly more acidic than the µ i chains (Figure 1, Table 1). In addition, µ m chains are larger in apparent molecular weight than µ s chains by 4000 daltons (Figure 1, Table 1). This molecular weight difference is not due entirely to differences in the levels of glycosylation because in the presence of the glycosylation inhibitor, tunicamycin, W279 lymphoma cells synthesize two distinct µ polypeptide chains differing in molecular weight by J'l500 daltons (Table 1). The smaller of the two µ polypeptides corresponds in molecular weight to precursor µ s chains synthesized by MxW cells in the presence of tunicamycin (Table 1). Thus, the B-cell lymphoma synthesizes two µ chains, µ m and µ s' which apparently differ in size by 15 to 20 amino acid residues. The µ i Chains Are Incompletely Glycosylated Complex carbohydrate moieites are constructed from a branched mannose and glucosamine core structure by the addition of terminal N-acetyl glucosamine, galactose, fucose, and sialic acid residues (Kornfeld and Kornfeld, 1976). The µs chains have three complex carbohydrate moieties (Kehry et al., 1979). Treatment of a detergent solubilized lysate of W279 cells with galactose oxidase and NaB 3H labels µ chains but not µ. chains (Figure 2). Therefore, we conclude 4 m 1 that the µ i chains are missing terminal galactose residues and presumably the other terminal residues such as sialic acid. In contrast, the µ m chains have terminal gaiactose residues and presumably are fully glycosyla ted. The basic pl of the µ i chains as compared with the µ m chains can then be explained by the 114 lack of terminal sialic acid residues on the complex carbohydrate structures of the µ i chains. Therefore, the final stages of glycosylation have not been completed on these precursor µ . chains. l The Sites of Carbohydrate Attachment Are Similar in µ. and µ l S Chains The cyanogen bromide fragments from the constant region of µ chains can be s individually resolved by gel filtration (Kehry et al., 1979) as shown in Figure 3 for MOPC 104E (M104E) µs chains. Fragments CNl-2, 2, 3 and 4 contain variable region sequences, whereas fragments CN 4, 5, 6, 7, 8, 8-9 and 9 contain C µ sequences with the numbering extending from the NH - to the COOH-terminus 2 (CNl to CN9, respectively). The 20 residue C-terminal segment is contained in fragments CNS, 9 and 8-9. Peptides with two numbers, such as fragment CN8-9, indicate an incomplete methionine cleavage by cyanogen bromide (due to an adjacent serine residue) with the resulting peptide containing two methionine fragments. The M104E µ s chain contains five C µ region carbohydrate moieties. Fragments CNS and CN6 contain complex carbohydrate moieties (boxed in the schematic drawing of the µ s chain in Figure 3), whereas fragments CN7 and CNS contain high mannose or simple (circled) carbohydrate moieties. The excluded column peak (Agg.) consists of aggregated fragments and large uncleaved peptides. A small peak migrating on the gel filtration column between CN6 and CN7 has been found only in M104E µ s chains and consists of a combination of VH and Cµ region partial cyanogen bromide cleavage products. We have used the M104E µ chain as the standard µ chain for assessing the glycosylation of W279 µ chains. s s 3 W279 cells were labeled with H-mannose and the labeled µ. chains were l preparatively isolated as the smaller molecular weight µ peak on 10% polyacrylamide-SDS gels. These µ. chains were cleaved with cyanogen bromide and the l fragments compared by gel filtration with µ s chains that had been secreted by 3 H-mannose-labeled M104E cells and similarly prepared (Figure 4). 115 The schematic drawing at the top of Figure 4 illustrates the order of cyanogen bromide fragments in the W279 µ chain and the sites of carbohydrate attachment in the mouse µ s chain Cµ region (see Figure 3). Several points are 3 .of interest. First, the µ i and µ s chains have very similar H-mannose-labeled cyanogen bromide profiles. Therefore the majority of W279 µ. chains consists l of precursors to µ s chains. Both chains have identical numbers of 3H-mannoselabeled C µ region peptides, and corresponding peptides in µ l. and µ S chains in3 corporate H-mannose in identical proportions. Second, the peptides with simple carbohydrate moieties, CN7 and CNS, migrate to identical positions for the µ. . l and µ s chains. Third, the peptides with complex carbohydrate moieties, CN5 and CN6, are smaller in molecular weight in the µ i (denoted CN5' and CN6') than in the µs chains. This observation is consistent with the earlier conclusions drawn from the galactose oxidase studies that the µ. chains lack the terminal sugars l of the complex carbohydrate moieties. These C µ region peptides should then be smaller in molecular weight for the µ. chains. Moreover, partial radiolabeled l sequence analysis of CN5', CN6' and CN7 fragments from µ . chains demonstrates l that they are identical at their NH -termini to their µ s chain counterparts (Kehry 2 et al., 1979) (Figure 5, Table 2). Because these three cyanogen bromide peptides are contiguous (Figures 3 and 4), the smaller size of the CN5' and CN6' peptides cannot be explained by shifts in the locations of the corresponding methionine residues unless one postulates that additional small C µ region peptides are generated by cyanogen bromide cleavage of the µ. chains. Cyanogen bromide frag1 mentation studies on 3H-amino acid-labeled µ i chains (to be discussed later) do not reveal any such additional peptides. The above studies suggest that the cyanogen bromide glycopeptides in the majority of µ. and µ chains are identical, apart l S from differences in the levels of glycosylation for complex carbohydrate moieties. 116 The µi and µ Chains Appear To Be Very Similar in Primary Structure 5 W279 µ. and MxW µ chains were labeled with 3H-tyrosine and 3H-leucine, purified, l S cleaved with cyanogen bromide and the fragments analyzed by comparative gel filtration (data not shown). The VH region cyanogen bromide peptides of these µ i and µ s chains have been identified by NH 2-terminal amino acid sequence analyses (see below) and were identical in size. Because the W279 µ chain has an unusual methionine distribution in the VH region, this size correspondence supports the hypothesis that the MxW hybridoma cells secrete µ s chains with a W279 VH region. The µ. and µ chains were labeled with either 3H-tyrosine, 3H-phenyl1 S alanine, 35 s-cysteine, or 35 s-methionine and digested with trypsin and chymotrypsin. The resulting peptides were analyzed by high performance liquid chromatography (HPLC). This method of peptide separation involves the interaction of the amino acid residues with the derivatized packing followed by elution with a continuous exponential gradient of acetone (McMillan et al., 1979). The peaks in the early fractions are somewhat variable due to the shallowness of the gradient in this region. In addition, the difficulty in achieving complete digestion with chymotrypsin produces some .smaller variable peaks probably representing partial enzymatic cleavage products. The µ. and µ chains labeled with tritiated tyrosine, l S phenylalanine, cysteine, or methionine exhibited very similar tryptic and chymotryptic patterns. Figure 6 illustrates the comparison of the peptides from 35 s- methionine-labeled µ. and µ chains. The major peaks in the µ. and µ chains l S l S elute at identical positions, and these chains are therefore very similar to one another, supporting the idea that the µ. chains represent primarily an internal 1 . pool of pre-µ s chains. 3 We have incorporated various H-labeled amino acids into the µ i and µ s chains for partial amino acid sequence analyses. In addition, we isolated sufficient quantities of W279 µ chains from mice inoculated with the MxW hybridoma tumor s 117 so that we could determine directly the sequence of most of the W279 variable region. The amino acid sequence data for each cyanogen bromide fragment of the µ i chains and the V region and COCH-terminal cyanogen bromide fragments for the µ chains are compared with the amino acid sequence of the M104E µ s s chain in Table 2. These data are summarized in a diagrammatic fashion in Figure 7. Several conclusions can be drawn. First, we have determined about 80% of the amino acid sequence of the W279 µ s VH region (Table 2). The W279 VH region has an unusual amino acid sequence that readily allows it to be distinguished from many other mouse VH regions (Kabat, Wu and Bil ofsky, 19 76). Second, the VH regions of the MxW µ and W279 µ. chains are identical at all the positions where S l they can be compared (Table 2). Because the W279 VH region is unusual in its sequence, we feel there is a high probability that the W279 µ i and µ s VH regions will be identical throughout their sequences. This conclusion also is supported by the peptide map data presented above. Fusion of the W279 lymphoma cell line with the myeloma cell line therefore induces the secretion of large quantities of µ s chains containing the W279 VH region. Third, the Cµ region cyanogen bromide peptides from the W279 µ i and MxW µ s chains have identical NH -terminal 2 sequences for all the positions where they can be compared (Table 2). In addition, the W279 C µ region is identical to that of M104E µ s chains (Table 2). These partial sequence comparisons, together with peptide map comparisons discussed above, lead to the conclusion that the majority of the µ i chains and the µ s chains are probably identical in amino acid sequence. Thus, within our limits of detection, the predominant µ. chain species represents precursors to µ l S chains in that the µ. chains differ only in the glycosylation levels of their complex carbohydrate l moieties. 118 The µ m and µ 8 Chains Have Complex Carbohydrate Moieties Located in Identical Regions After treatment of intact W279 cells with galactose oxidase and NaB 3H , the 4 labeled µ m chains were isolated, cleaved with cyanogen bromide and compared by gel filtration to M104E µ chains which were similarly treated (Figure 8). s Only the fragments CN5 and CN6 contain complex carbohydrate moieties (see Figures 3 and 4), and only fragments CN5 and CN6 contain radiolabeled galactose residues (Figure 8). These peptides migrate at identical positions for the W279 µ m chains and the M104E µs chains. We therefore conclude that complex carbohydrate moieties are located in identical regions in the µ m and µ chains. s The µ m Chain Lacks the COOR-Terminal High Mannose Carbohydrate Moiety Present on Pre-µ Chains and µ 8 8 Chains 3 W279 cells were labeled with H-mannose to label all of the carbohydrate moieties. The labeled µ chains were isolated, cleaved with cyanogen bromide, and the m fragments compared by gel filtration with µ s chains that had been similarly labeled and prepared from M104E cells (Figure 9). Of the four carbohydrate containing fragments found in µ s chains, three are present in the µ m chains, fragments CN 5, CN6, and CN7. These fragments are identical in molecular weight and in the proportions of 3H-mannose incorporated into their respective carbohydrate moieties. However, the position where an expected µ CN8-9 fragment should migrate m is lacking mannose, and therefore, the high mannose carbohydrate moiety (see 3 Figures 3 and 4). This conclusion is supported by comparing H-mannose-labeled µ. (pre-µ ) and µ 1 s m chains after digestion with trypsin and chymotrypsin (Figure 10). The high performance liquid chromatographs in Figure 10 show that the µ m chains are lacking one 3H-mannose-labeled peptide (crosshatched) that is present in µ i 3 chains. As we noted earlier, the µsand µi chains have the same number of H- 119 mannose-labeled cyanogen bromide fragments (Figure 4) and trypsin plus chymotrypsin peptides (data not shown). The missing carbohydrate residue in µ m chains probably reflects the absence of the carbohydrate recognition sequence present in the Cterminal segment of µ The µ m s chains. Chain Differs from the µ Chain in Its C-Terminal Segment s To localize any amino acid sequence differences between µ 3 R-amino acid-labeled cyanogen bromide fragments of µ m m and µ s chains, the and µ. chains were I 3 compared by gel filtration (Figure 11). In particular, R-tyrosine was included in the µ i and µ m chains because it labels the COOR-terminal cyanogen bromide fragment of µ s chains, CN9 (see Figure 4; Kehry et al., 1979). The order and identity of these W279 cyanogen bromide fragments (shown in Figure 4) were deduced from a comparison of the NR -terminal radiolabeled sequences of the 2 µ. cyanogen bromide fragments and the corresponding sequences of the myeloma I M104E µ s chain (Table 2). As expected, the CN5' and CN6' fragments from the µ. chains are smaller in molecular weight than the CN5 and CN6 fragments from I the completely glycosylated µ m chains. The COOR-terminal cyanogen bromide fragment, CN9, is absent in the µ m chains, although two-fold larger aliquots of the last 40 column fractions from the cyanogen bromide digest of µ m chains were counted for radioactivity. The CN9 fragment is present in the µ i chains, supporting the idea that µ i chains are mainly composed of the internal precursor µ s chains before secretion. Identical results also have been obtained by comparing 3 R-tyrosine-labeled cyanogen bromide fragments of µ. and µ I m chains from a second B-cell lymphoma, WERI 231 (R. Douglas, unpublished observations). Since we reproducibly achieve 50% cleavage of methionine-serine bonds by cyanogen bromide, the absence of the CN9 fragment from the µ m chains suggests that the µ m chains lack the methionine residue nine amino acids from the 120 COOR-terminus of the µ s chains. This supposition is substantiated by a comparison of W279 µ m and MxW µ s chains labeled with 35 s-methionine, digested with trypsin and chymotrypsin and analyzed by HPLC (Figure 12). The HPLC peptide maps indicate that the µ m chains lack a major methionine peptide (crosshatched) that is present in the µ s chains. A similar comparison of tryptic and chymotryptic peptides from µ m and µ s chains labeled with 3H-phenylalanine (Figure 13) also shows that µ m chains have two phenylalanine peptides not present in µ s chains. Verification of the missing high mannose carbohydrate moiety and methionine residue in the C-terminal segment of µ m as compared with µ (µ.)chains using s 1 two independent analytical techniques, suggests that the µ m and µs chains possess different COOR-terminal amino acid sequences. The µ m Chains Differ in Amino Acid Sequence from the µ Chains at Their 5 COOR-Termini The amino acids present at the COCH-termini of W279 µ m' µ i and MxW µ s chains were investigated by digestion of the radiolabeled µ chains with carboxypeptidases A and B. The µ chains were labeled with either 3H-tyrosine, leucine, valine, phenylalanine or lysine, purified, the amounts of tritium in each amino acid pool quantitated and the labeled µ chains combined into µ m' µ i and µspools. Digestion of W279 µ , µ. and MxW µ chains with carboxypeptidases A and B was followed by separation m s 1 and quantitation of the released amino acids on an amino acid analyzer. The time course of release of the amino acids can be used to determine a COOR-terminal amino acid sequence. As is shown in Figure 14, different amino acids were released from µ m and µ chains. The rapid release of tyrosine from MxW µ chains in s s Figure 14C is expected since tyrosine is the COCH-terminal amino acid in µ s chains (Kehry et al., 1979). Background release of valine and phenylalanine is also observed for MxW µ chains which correlates with the low level release of these s 121 amino acids from unlabeled M104E µ chains digested with carboxypeptidase (not s shown). In addition, tyrosine is the predominant amino acid released from W279 µ i chains (Figure 14B). This confirms the previous peptide maps and sequence determinations on µ. chains that indicate µ. chains are predominantly incompletely 1 1 glycosylated precursors to µ s chains. In contrast, carboxypeptidase digestion of W279 µ m chains identifies the COOH-terminal sequence as Lys-Val-COOH (Figure 14A). Amino acids other than tyrosine which correspond to those released from µ m chains were also released from µ i chains in lower yield. Since this same sequence can be attributed to a polypeptide present as 10-15% of theµ. chains, 1 we propose that the lysine-valine COOH-terminal sequence represents precursors to µ m chains in the W279 internal pool of µ chains. Similarly, the rapid release of tyrosine from 20% of the µ m chains indicates that µ s chains are present as a contaminant in the µ m pool which was purified by isolation of the highest molec- ular weight µ chain from 10% polyacrylamide gels. The above finding that µ m chains possess a different COOH-terminal amino acid sequence from µ s chains is consistent with our previous results which indicate that all of the amino acid and carbohydrate differences between µ m and µ chains can be localized to the s C-terminal segment. Discussion W279 Cells Synthesize Two Distinct Species of µ CN!.im, µ and µ m 5 These µ chains differ in electrophoretic mobility, localization in the cells, and covalent structure. The µ m chains-have been localized to the W279 cell plasma membrane by galactose oxidase and lactoperoxidase labeling (Sibley et al., 1980). The µs chain is secreted by MxW cells as a pentamer [(µ 2L2\J in the presence of a J (joining) chain (Raschke, Mather and Koshland, 1979; M. Koshland and E. Mather, personal communication). Pulse-labeling studies (Sibley et al., 1980) 122 and carboxypeptidase experiments indicate that the internal µ pool contains precursors for µ m and µ s chains that are lacking the terminal sugars (e.g., galactose, sialic acid and presumably fucose) in their complex carbohydrate moieties. Carboxypeptidase and carbohydrate labeling studies suggest that in B-cell lymphomas the µ s precursor appears to predominate in the µ i pool (Figure 15). Perhaps the induction of J chain synthesis is an important factor in the differentiation process which converts B cells to terminally differentiated, IgM-secreting plasma cells (Raschke, Mather and Koshland, 1979). The W279 µ m and µ 8 Chains Are Identical but for Their C-Terminal Segments Gel filtration comparisons of cyanogen bromide fragments suggest that the µ m and µ s chains are identical but for their C-terminal segments (Figure 15). The extensive similarity of the µ m and µ s chains is supported by the comparison of tryptic and chymotryptic peptides labeled with a variety of different amino acid residues. In addition, the placement of high mannose and complex carbohydrate moieties appears identical except for the high mannose carbohydrate moiety which is present in the C-terminal segment of µ chains and is absent in µ s m chains (Figure 15). This observation implies that the corresponding Asn-X-Ser/Thr carbohydrate recognition sites are preserved throughout the C µ region of both µ m and µ chains. s The 40-Residue C-Terminal Segment Sequence Predicted from Nucleic Acid Studies is in Complete Accord with our Protein and Carbohydrate Findings on the µ m Chain The amino acid sequences of the C-terminal segment of the µs chain (Kehry et al., 1979) and the predicted C-membrane terminal segment of the µ m chain (Rogers et al., 1980; Early et al., 1980) are given in Figure 16. The following points are derived from a comparison of the two amino acid sequences. i) The COOH-terminal sequence of theµ m chain is -Lys-Val by carboxypeptidase analysis. This observation is consistent with the COOH-terminal sequence predicted by the DNA sequence 123 if we postulate that the COCH-terminal-most lysine residue has been removed by posttranslational proteolysis. This assumption is reasonable because COOHterminal lysine residues also are removed by posttranslational proteolysis in yl (Honjo et al., 1979) and y2b (Tucker et al., 1979; Yamawaki-Kataoka et al., 1980) chains. ii) Tryptic and chymotryptic peptide maps of 3H-phenylalanine-labeled s and µ m chains indicate that µ m chains have two phenylalanine peptides not found in µ chains. This is precisely what is expected from the predicted µ s m µ C-terminal segment sequence, taking into account the fact that three of the five phenylalanine residues in the C-terminal segment will yield free phenylalanine with this type of enzymatic digestion (Figure 16). iii) There is no methionine residue in the predicted µ C-terminal segment. Our comparative studies on m cyanogen bromide fragments and 35 s-methionine-labeled tryptic and chymotryptic peptides also indicate that µ chains lack a COCH-terminal methionine residue. m iv) The recognition sequence for asparagine-linked carbohydrate moieties, AsnX-Ser/Thr, is missing in the predicted sequence of the µ C-terminal segment. m Accordingly, the absence of the COCH-terminal high mannose carbohydrate moiety in µ chains can be readily explained. v) The µ C-terminal segment is 20 residues m m longer than its µ s counterpart. This is in excellent agreement with the size difference predicted by tunicamycin studies. vi) The µ C-terminal segment is m missing a half cystine residue found in the µ chain one residue from the COOHs terminus. Although we have not demonstrated the absence of this penultimate cystine residue in µ m chains, this residue is essential for the formation of disulfidelinked secreted IgM pentamers (Mestecky and Schrohenloher, 1974). In this regard, membrane IgM molecules are only capable of forming covalent monomers on the B cell surf ace, as was shown by our labeling studies with galactose oxidase (Sibley et al., 1980). vii) The µ m C-terminal segment contains a stretch of 25 uncharged amino acid residues which is comparable in hydrophobicity index (Segrest and 124 Feldmann, 1974) to the membrane-associated regions in glycophorin (Tomita and Marchesi, 1975), phage M13 coat protein (Wickner, 1976) and HLA heavy chains (J. Strominger, unpublished). The µ m chain is therefore an integral membrane protein possessing hydrophobic C-terminal segment capable of binding detergents (Vassalli et al., 1979; Melcher and Uhr, 1977; Parkhouse, Lifter and Choi, 1980). Thus the predicted amino acid sequence of the µ C-terminal segment is fully m consistent with all of the chemical and biological observations on µ chains. m The Membrane-Bound IgM Molecule Represents a Structurally Well-Characterized Eukaryotic Membrane Receptor Membrane receptors transduce signals between the external and internal environments of cells and as such constitute an extremely important class of biological effector molecules. Studies on the structure of most eukaryotic membrane receptors have been limited by the very small amounts of material generally available (Devreotes, Gardner and Fambrough, 1977; Carpenter and Cohen, 1979; Gill, 1976; Cuatrecassas et al., 1975; Andres, Jeng and Bradshaw, 1977). Our studies on the IgM receptor molecule and its corresponding gene and mRNA (Rogers et al., 1980; Early et al., 1980) are important in two respects. First, we have characterized in detail the structure of the IgM receptor molecule. We have demonstrated that µ m chains are very similar to µ chains throughout the v and C domains. s µ 8 The sites of carbohydrate attachment within the four C µ domains appear identical in µ m and µ chains. These chains, however, clearly differ in the amino acid s sequence of their C-terminal segments. Indeed, sequence analyses of the µ m and µ s mRN As indicate that the C-terminal segments of the µ m and µ s chains consist of 41 and 20 residues, respectively. The protein studies of µ m chains are in complete agreement with the C-membrane-terminal segment sequence predicted from the mRNA sequences (Figure 16), and therefore this C-membrane- 125 terminal segment is actually contained in µ m chains. These studies indicate that the µ m chain is a transmembrane polypeptide which spans the plasma membrane employing a 25-residue stretch of uncharged and hydrophobic residues near its COOH-terminus (Figure 16) (see Rogers et al., 1980). Second, studies on the µ andµ s mRNAs as well as the C µ gene suggest that the distinct µ m m and µ Cs terminal segments are generated by RN A splicing from the RN A transcripts of a single Cµ gene (Rogers et al., 1980; Early et al., 1980). Thus, RNA splicing allows IgM molecules and perhaps other immunoglobulins to be expressed as integral membrane receptors, or alternatively, as soluble effector molecules of humoral immunity. Experimental Procedures Cells The WEHI 279 (W279) cells (a gift from Dr. N. Warner) were grown in stationary suspension cultures at 37°C in Dulbecco's Modified Eagle's Medium (DMEM) containing 10% heat-inactivated fetal calf serum (f cs), nonessential amino acids (Gibco), and 5 x 10- 5 M 8-mercaptoethanol. We subsequently obtained a later clone of W279 cells (W279.l/12) isolated by Dr. Warner and sent to us by Dr. V. Oi. These two cell lines correspond to slightly different stages in B lymphocyte development (Sibley et al., 1980). W279 cells were used for the experiments illustrated in Figures 4, 5 and 9, some of the sequences in Table 2, and the cell fusion with • MPCll to generate MPCllxW279 cells (described below). The W279.l/12 clone was used for all remaining experiments. Our previous characterization of these two cell lines (Sibley et al., 1980; N. Warner, personal communication) indicates that µ m and µ chains isolated from the two clones are equivalent, and we refer s only to W279 cells in all descriptions. The MPCllxW279 (MxW) hybrid cells were generated by a fusion between MPCll and WEHI 279 cells (Raschke, Mather and 126 Koshland, 1979). MPCllxW279.2 cloned cells were stable for IgM production for periods of up to 6 months and were grown in DMEM supplemented with 10% heat-inactivated fetal calf serum. MOPC 104E (Ml04E) myeloma cells were passaged subcutaneously as a solid tumor in (BALB/c x DBA/2)F mice. The 1 MPCll x W279.2 (MxW) hybrid cells were grown as solid tumors in BALB/c mice. The cells were grown in ascites form for the production of milligram quantities of M104E or W279 secreted IgM or for isolation of M104E cells to be radiolabeled. Labeling W279 and MxW cells were biosynthetically labeled by washing with fresh medium and resuspending at cell densities of 2 x 10 6/ml in labeling medium [DMEM containing 5% dialyzed heat-inactivated fcs, 0.3 mg/ml glutamine, 5 x 10- 5 M 13mercaptoethanol and nonessential amino acids (Gibco)] lacking the amino acid or amino acids to be labeled, plus 50 µCi/ml of a 3H-amino acid (New England Nuclear). Cells were incubated in a humidified atmosphere lacking CO for 16 hr 2 at 37°C. M104E ascites fluid was diluted 1:2 with Hanks Balanced Salt Solution (HBSS) (Gibco), cells purified over Ficol-Hypaque, the buffy coat washed three times with HBSS, 5% heat-inactivated fcs, and resuspended in labeling medium 6 at 2 x 10 M104E cells/ml. Labeling conditions were identical for all cell lines. Following the 16 hr labeling period, the cells were harvested by centrifugation at 150 x g for 10 min and the medium was removed, made 1 mM in phenylmethyl sulfonylfluoride (PMSF), 1 x 10 -4 M 13-mercaptoethanol and stored frozen at -70°C. Cells were washed once in cold DMEM, 20% fcs and resuspended in lysis buffer (0.01 M Tris-HCl pH 7.4, 0.14 M NaCl, 0.5% Triton X-100, 1 mM PMSF, 1 x 10- 4 M 13-mercaptoethanol) at a concentration of 2.5 x 10 7 cells/ml and incubated for 20 min at 4°C. The mixture was centrifuged at 1800 x g at 4°C for 10 min to remove nuclei. Supernatants were removed, aliquoted and stored frozen at -70°C. 127 When tunicamycin was employed, cells were preincubated in 2 µg/ml tunicamycin for 1 hr prior to the addition of 35 s-methionine. Cells were labeled in the presence of tunicamycin for 1 hr. Extrinsic labeling of galactose residues with NaB 3 H and galactose oxidase 4 was performed by the method of Gahmberg and Hakomori (1973). Purified M104E IgM molecules or W279 whole cell lysates prepared as above were labeled, desalted on Sephadex G-25 equilibrated in 0.01 M Tris-HCl pH 7.4, 0.14 M NaCl, 0.5% Triton X-100 and stored frozen at -70°C. Immunoprecipitation Lysates were preincubated with washed formalin-fixed§. aureus (Kessler, 1976) [50 µl 10% suspension of§. aureus per 100 µl cell lysate] for 20 min at 4°C. After centrifugation at 1800 x g at 4°C for 5 min to remove the§. aureus, the supernatant was made 0.5% in SDS and incubated with rabbit anti M104E µ serum (rabbit anti µ ), [10 µl rabbit anti µ per 100 µl cell lysate] for 1-1/2 hr on ice. Immune complexes were precipitated by addition of washed§. aureus [100 µl 10% suspension] for 20 min at 4°C. Precipitates were washed three times in washing buffer [0.01 M Tris-HCl pH 7.4, 0.14 M NaCl, 0.5% Triton X-100, 0.1% SDS] and the proteins eluted in 50 µl to 100 µl sample buffer [50 mM Tris-HCI pH 6.8, 2% S-mercaptoethanol, 2% SDS] per 100 µI cell lysate by incubating in a boiling water bath for 2 min. For samples to be analyzed on 10% polyacrylamide-SDS gels, glycerol plus pyronin Y were added to a final glycerol concentration of 15%. Culture medium was precipitated according to the same protocol except that SDS was omitted from all the buffers and 20 µl rabbit anti µ was used per 1.0 ml medium. Samples to be analyzed by isoel~ctric focusing were precipitated by 25% trichloroacetic acid (TCA) at 4°C for 1 hr and the precipitates washed at 4°C twice with 20% TCA, twice with EtOH:ether, 1:2, once with ether, and air dried. 128 Gel Electrophoresis The µ i' µ m and µ s chains were isolated from preparative 10% polyacrylamide-SDS gels (Laemmli, 1970) with a 2 cm stacking gel. Proteins were eluted from 1 mm slices by incubation of each 1 mm slice in 0.5 ml 0.5% SDS for 24 hr. Radioactivity was determined for an aliquot of each sample by liquid scintillation counting and the appropriate fractions were pooled. Two-dimensional gel analysis was performed according to O'Farrell (1975). Isoelectric focusing gels were 5% polyacrylamide, 0.28% bisacrylamide, 9.2 M urea, 2% Triton X-100, 2% ampholines pH 3.5-10, 0.13% ampholines pH 5-7, 0.13% ampholines pH 7-9, 0.13% ampholines pH 4-6, and 0.07% ampholines pH 3.5-5 (LKB). TCA precipitates were dissolved in 50 µl sample buff er (9.5 M urea, 2% Triton X-100, 5% 8-mercaptoethanol, 0.4% ampholines pH 3.5-10, 1.6% ampholines pH 5-7) at least 1 hr before loading. Gels were stained and fixed in 25% isopropanol, 20% sulfosalicyclic acid, 0.025% Coomassie Brilliant Blue, destained in 7.5% acetic acid, 5% methanol and fluorographed according to Bonner and Laskey (1974). Cyanogen Bromide Cleavage Unlabeled M104E and MxW µ chains were purified as previously described (Kehry s et al., 1979). In order to determine the sequence of MxW CN3b, the isolated µs chains were succinylated (Klapper and Klotz, 1972) prior to cyanogen bromide cleavage. This procedure blocked the NH -terminus of the µs chains (e.g., frag2 ment CN3a). The µ ., µ 1 m or µ chains pooled from appropriate slices of 10% polys acrylamide-SDS gels were reduced anG-alkylated as follows. Carrier pig IgG and M104E IgM (3 mg each) were added, the pooled µ chains concentrated by lyophilization and brought to a final concentration of 9% SDS, 0.5 M Tris-HCl pH 8.5, boiled for 5 min, flushed with N for 5 min, dithiothreitol added to a concentration 2 129 of 20 mM, boiled 2 min, sealed under N and incubated at 37°C for 1-1/2 hr. 2 Alkylation with 50 mM iodoacetamide (3X recrystallized) was performed for 1 hr at room temperature in the dark. Reagents were removed by desalting on Sephadex G-25 equilibrated with 0.05 M Tris-HCl pH 7.4, 0.5% SDS. Proteins were TCA precipitated as described under Immunoprecipitation, and the air dried precipitate was dissolved in 88% formic acid and combined with 30 mg Ml04E µ s chains in 70% formic acid. Cyanogen bromide was added (50 mg/ml) and cleavage was performed for 20 hr at 4°C in the dark with constant stirring (Gross, 1967). Fragments were separated by gel filtration as previously described (Kehry et al., 1979) on a column of ACA54 (1KB) (3.5 x 140 cm) equilibrated with 3 M guanidineHCl, 0.2 M ammonium bicarbonate, 0.02% NaN . 5 ml fractions were collected 3 and aliquots were counted in Aquasol (New England Nuclear) in a liquid scintillation counter (Beckman LS 9000). Peptide Mapping 0.25 mg of carrier pig IgG was added to each gel-purified µ i' µ m or µ s sample intended for peptide mapping. Chains were completely reduced and alkylated as described above under Cyanogen Bromide Cleavage, TCA precipitated and air dried. Samples were dissolved in 200 µl 0.2 M ammonium bicarbonate, digested with TPCK-trypsin (Worthington) at room temperature [total of 150 µg for 22 hr], followed by digestion with chymotrypsin (Worthington) at 37°C [total of 200 µg for 28 hr]. The peptides were frozen and lyophilized to terminate the reaction, dissolved ln 45 µl 0.5 M phosphate buffer (pH 1.8):acetone, 2:1 and the peptides separated as described by high performance liquid chromatography on a DuPont ODS C-18 column (McMillan et al., 1979). Fractions were collected at 0.5 min intervals (0.5 ml), dried in a vortex evaporator, redissolved in 0.25 ml 0.01 % SDS and assayed for radioactivity in a liquid scintillation counter. 130 Sequence Analysis Cyanogen bromide fragments from the ACA54 column were prepared for sequence determinations by pooling, dialyzing exhaustively against 5% formic acid in low molecular weight cutoff dialysis tubing (Spectra/Per 3) at 4°C, and lyophilizing. Pools contained 1 to 3 mg carrier M104E cyanogen bromide fragments. Automated sequence analyses on the milligram quantities of MxW µ s chains and CN3 fragments were performed on a modified Beckman sequenator (Hunkapiller and Hood, 1978; Wittmann-Liebold, 1973; Wittmann-Liebold, Graffunder and Kohls, 1976). Samples were loaded by using Polybrene (Aldrich), and phenylthiohydantoin (Pth) derivatives were identified by high performance liquid chromatography (HPLC) (Water Associates) as described (Hunkapiller and Hood, 1978; Johnson, Hunkapiller and Hood, 1979). Automated sequence analyses of radiolabeled cyanogen bromide fragments were performed on the Caltech sequenator (Hunkapiller and Hood, 1980). Samples were loaded in trifluoroacetic acid (Pierce) and 20% H 0 by using Polybrene (Aldrich). The Pth derivatives were separated 2 by HPLC (DuPont), the peaks corresponding to the radiolabeled Pth-amino acids were fraction collected, dried, and counted for radioactivity in Liquifluor-toluene (McMillan et al., 1977). Carboxypeptidase A and B Digestions Carboxypeptidase A (Worthington) and carboxypeptidase B (Sigma) were employed at an enzyme to substrate ratio of 1:40. Carboxypeptidase A was prepared as described (Ambler, 1972) and carboxypeptidase Bin 0.1 M NaCl was thawed and diluted 1:10 in 0.2 M N-ethylmorpholine acetate pH 8.6 just prior to use. A known number of counts of isolated radiolabeled µ chain was reduced and alkylated as for peptide maps and dissolved in 0.2 M N-ethylmorpholine acetate pH 8.6 at a carrier pig IgG concentration of 2.5 mg/ml. An aliquot was removed for a substrate 131 control, and the remainder was digested with an equimolar mixture of carboxypeptidases A and Bat 37°C. Aliquots were removed at indicated times, frozen immediately on dry ice and lyophilized twice. For substrate controls, equivalent amounts of enzymes were boiled in 0.1 N acetic acid for 15 min, added to the substrate and incubated at 37°C for the same length of time as the longest reaction time point. Released amino acids were separated on a Durrum D-500 amino acid analyzer (92% of sample loaded). The ninhydrin coil temperature was turned to the minimum setting and 0.5 min fractions were collected from the time of injection and assayed for radioactivity in a liquid scintillation counter. Radiolabeled amino acid standards were analyzed by the same procedure. Total counts in each peak were quantitated (substrate controls contained no peaks) by comparison with the starting radioactivity and the number of residues of each amino acid in the W279 µ chain (Kehry et al., 1979). s Antiserum Rabbit anti-mouse M104E µ chain was produced by repeated immunization of an NZW rabbit with 300 µg partially reduced and aklylated M104E µ chain in complete Freunds adjuvant. Blood was clotted, centrifuged at 6000 x g, and the serum aliquotted and stored frozen at -70°C. Anti µ serum was used without any further treatment. 132 References Ambler, R. P. (1972). Enzymatic hydrolysis with carboxypeptidases. Methods Enzymol. 25, 143-154. Andres, R. Y., Jeng, I., and Bradshaw, R. A. (1977). Nerve growth factor receptors: Identification of distinct classes in plasma membranes and nuclei of embryonic dorsal root neurons. Proc. Nat. Acad. Sci. USA 74, 2785-2789 . .. Bergman, Y., and Haimovich, J. (1977). Characterization of a carcinogen-induced murine B lymphocyte cell line of C3H/eB origin. Eur. J. Immunol. 7_, 413-417. Bergman, Y., Haimovich, J., and Melchers, F. (1977). An IgM-producing tumor with biochemical characteristics of a small B lymphocyte. Eur. J. Immunol. ~' 574-579. Bergman, Y., and Haimovich, J. W. (1978). B lymphocytes contain three species of µ chains. Eur. J. Immunol. ~' 876-880. Bonner, W. M., and Laskey, R. A. (1974). A film detection method for tritium-labeled proteins and nucleic acids in. polyacrylamide gels. Eur. J. Biochem. 45, 83-88. Burrows, P., LeJeune, M., and Kearney, J. F. (1979). Evidence that murine pre-B cells synthesize µ heavy chains but no light chains. Nature 280, 838-840. Carpenter, G., and Cohen, S. (1979). Epidermal growth factor. Ann. Rev. Biochem. 48, 193-216. Cuatrecassas, P., Hollenberg, M. D., Chang, K.-J., and Bennett, V. (1975). Hormone receptor complexes and their modulation of membrane function. Recent Prog. Horm. Res.~' 37-94. Dayhoff, M. O. (1976). Atlas of Protein Sequence and Structure (Silver Spring, Maryland: National Biomedical Research Foundation) 5, 189-190. 133 Della Corte, E., and Parkhouse, R. M. E. (1973). Biosynthesis of immunoglobulin A (IgA) and immunoglobulin M (IgM); requirement for J chain and a disulphide exchanging enzyme for polymerization. Biochem. J. 136, 597-606. Devreotes, P. N., Gardner, J.M., and Fambrough, D. M. (1977). Kinetics of biosynthesis of acetylcholine receptor and subsequent incorporation into plasma membrane of cultured chick skeletal muscle. Cell 10, 365-373. Early, P., Rogers, J., Davis, M., Calame, K., Bond, M., Wall, R., and Hood, L. (1980). Two mRN As can be produced from a single immunoglobulin µ gene by alternative RN A processing pathways (II). Cell, submitted. Fu, S. M., and Kunkel, H. G. (1974). Membrane immunoglobulins of B lymphocytes: Inability to detect certain characteristic IgM and IgD antigens. J. Exp. Med. 140, 895-903. Gahmberg, C. G., and Hakomori, S. (1973). External labeling of cell-surface galactose and galactosamine in glycolipid and glycoprotein of human erythrocytes. J. Biol. Chem. 248, 4311-4317. Gill, D. M. (1976). The arrangement of subunits in cholera toxin. Biochemistry ~' 1242-1248. Gross, E. (1967). The cyanogen bromide reaction. Methods Enzymol.11,, 238-255. Honjo, T., Oba ta, M., Yamawaki-Kataoka, Y ., Kataoka, T., Kawakami, T., Takahashi, N., and Mano, Y. (1979). Cloning and complete nucleotide sequence of mouse immunoglobulin yl chain gene. Cell 18, 558-568. Hunkapiller, M. W., and Hood, L. E. (1978). Direct microsequence analysis of polypeptides using an improved sequenator, a nonprotein carrier (polybrene), and high pressure liquid chromatography. Biochemistry .!1., 2124-2133. 134 Hunkapiller, M. W., and Hood, L. E. (1980). New protein sequenator with increased sensitivity. Science 207, 523-525. Johnson, N., Hunkapiller, M., and Hood, L. (1979). Analysis of phenylthiohydantoin amino acids by high-performance liquid chromatography on DuPont Zorbax cyanopropylsilane columns. Anal. Biochem. 100, 335-338. Kabat, E. A., Wu, T. T., and Bilofsky, H. (1976). Variable Regions of Immunoglobulin Chains (Cambridge: Bolt Beranek and Newman, Inc.). Kehry, M., Sibley, C., Fuhrman, J., Schilling, J., and Hood, L. E. (1979). Amino acid sequence of a mouse immunoglobulin µ chain. Proc. Nat. Acad. Sci. USA 76, 2932-2936. Kessler, S. W. (1976). Cell membrane antigen isolation with the Staphylococcae protein A-antibody adsorbant. J. Immunol. 117, 1482-1490. Klapper, M. H., and Klotz, I. M. (1972). Acylation with dicarboxylic acid anhydrides. Methods Enzymol. 25, 531-536. Kornfeld, R., and Kornfeld, S. (1976). Comparative aspects of glycoprotein structure. Ann. Rev. Biochem. 45, 217-237. Laemmli, U. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 277, 680-685. Mcilhinney, R. A. J., Richardson, N. E., and Feinstein, A. (1978). Evidence for a C-terminal tyrosine residue in human and mouse B-lymphocyte membrane µ chains. Nature 272, 555-557. McMillan, M., Cecka, J. M., Murphy, D. B., McDevitt, H. O., and Hood, L. (1977). Structure of murine Ia antigens: Partial NH -terminal amino acid sequences of 2 products of the 1-E or 1-C subregion. Proc. Nat. Acad. Sci. USA 7 4, 5135-5139. 135 McMillan, M., Cecka, J. M., Hood, L., Murphy, D. B., and McDevitt, H. O. (1979). Peptide map analyses of murine Ia antigens of the I-E subregion using HPLC. Nature 277, 663-665. Melcher, U., and Uhr, J. W. (1973). An electrophoretic difference between surface and secreted IgM of murine splenocytes. J. Exp. Med. 138, 1282-1287. Melcher, U., Eidels, L., and Uhr, J. W. (1975). Are immunoglobulins integral membrane proteins? Nature 258, 434-435. Melcher, U., and Uhr, J. W. (1977). Density differences between membrane and secreted immunoglobulins of murine splenocytes. Biochemistry 1&_, 145-152. Mestecky, J., and Schrohenloher, R. E. (197 4). Site of attachment of J chain to human immunoglobulin M. Nature 249, 650-652. O'Farrell, P. H. (1975). High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007-4021. Paige, C. J., Kincade, P. W., and Ralph, P. (1978). Murine B cell leukemia line with inducible surface immunoglobulin expression. J. Immunol. 121, 641-647. Parkhouse, R. M. E., Lifter, J., and Choi, Y. S. (1980). Chemical characterization of the Fab and Fe fragments from surface immunoglobulin. Nature, in press. Putnam, F. W., Florent, G., Paul, C., Shinoda, T., and Shimizu, A. (1973). Complete amino acid sequence of the mu heavy chain of a human IgM immunoglobulin. Science 182, 287-291. Raschke, W. C., Mather, E. L., and Koshland, M. E. (1979). Assembly and secretion of pentameric IgM in a fusion between a nonsecreting B cell lymphoma and an IgG-secreting plasmacytoma. Proc. Nat. Acad. Sci. USA 76, 3469-3473. 136 Rogers, J., Early, P., Carter, C., Calame, K., Bond, M., Hood, L., and Wall, R. (1980). Two mRNAs with different 3' ends encode membrane-bound and secreted forms of immunoglobulin µ chain (I). Cell, submitted. Segrest, J. P., and Feldmann, R. J. (1974). Membrane proteins: amino acid sequence and membrane penetration. J. Mol. Biol. 87, 853-858. Sibley, C. H., Ewald, S. J ., Kehry, M. R., Douglas, R. H., Raschke, W. C., and Hood, L. E. (1980). Characterization of multiple immunoglobulin mu chains synthesized by two clones of a B-cell lymphoma. J. Immunol., submitted. Siden, E. J., Baltimore, D., Clark, D., and Rosenberg, N. (1979). Immunoglobulin synthesis by lymphoid cells transformed in vitro by Abelson murine leukemia virus. Cell.!§., 389-396. Tomita, D. B., and Marchesi, V. T. (1975). Amino-acid sequence and oligosaccharide attachment sites of human erythrocyte glycophorin. Proc. Nat. Acad. Sci. USA 72, 2965-2968. Tucker, P. W., Marcu, K. B., Slightom, J. L., and Blattner, F. R. (1979). Structure of the constant and 3' untranslated regions of the murine y2b heavy chain messenger RNA. Science 206, 1299-1303. Vassalli, P., Tedghi, R., Lisowska-Bernstein, B., Tartakoff, A., and Jaton, J.-C. (1979). Evidence for hydrophobic region within heavy chains of mouse B lymphocyte membrane-bound IgM. Proc. Nat. Acad. Sci. USA 76, 5515-5519. Vitetta, E. S., Baur, S., and Ohr, J. W. (1971). Cell surface immunoglobulin. II. Isolation and characterization of immunoglobulin from mouse splenic lymphocytes. J. Exp. Med. 134, 242-264. 137 Warner, N. (1974). Membrane immunoglobulins and antigen receptors on Band T lymphocytes. Adv. in Immunol. .!Q_, 67-216. Wickner, W. (1976). Asymmetric orientation of phage M13 coat protein in Escherichia coli cytoplasmic membranes and in synthetic lipid vesicles. Proc. Nat. Acad. Sci. USA 73 , 1159-1163. Williams, P. B., Kubo, R. T., and Grey, H. M. (1978). µ-chains from a non-secreter B cell line differ from secreted µ-chains at the C-terminal end. J. Immunol. 121, 2435-2439. Wittmann-Liebold, B. (1973). Amino acid sequence studies on ten ribosomal proteins of Escherichia coli with an improved sequenator equipped with an automatic conversion device. Hoppe-Seyler's Z. Physiol. Chem. 354, 1415-1431. Wittmann-Liebold, B., Graffunder, H., and Kohls, H. (1976). A device coupled to a modified sequenator for the automated conversion of anilinothiazolinones into PTH amino acids. Anal. Biochem. 75, 621-633. Yamawaki-Kataoka, Y., Kataoka, T., Takahashi, N., Obata, M., and Honjo, T. (1980). Complete nucleotide sequence of immunoglobulin y2b chain gene cloned from newborn mouse DNA. Nature 283, 786-789. Yuan, D., Uhr, J. W., and Vitetta, E. S. (1980). A peptide difference between the µ ·chains from cell-associated and secreted IgM of the BCL tumor. J. Immunol., 1 in press. 138 Table 1. Molecular Weights and Isoelectric Points of W279 µ Chains Molecular Weighta Cells and µ Chain Species - Tunicamycin + Tunicamycin pl Range W279 5.69-5.17 µm 82 pre-µm 78 69 5.89-5 . 58 pre-µs 76.5 67.3 6.22&6.10 µs 78b pre-µ s 76.5 MxW 3 aMolecular weight in daltons x 10- . b . Molecular weight range of 77,000 to 80,600 daltons. 5.83-5.33 67.3 6.22&6.10 LJS LJ 8 CN)b CNl CN]ab CN)a 1,1 i 1,1 CNS CNl-2 • CH4 - V - L - 1 1-1 1,1 1 1,1 9 8 CH8-9 CN7c CN7 5 HxW 1,1 8 HxW LJ y - - y - -YY-P-L- • ., L -IL) - - - V IJI I■ ,. -v-v-Ot FP-VPPLV- u, L - M ■ - ' -- ,._ - V - - - -(P)- Al - VIA)- V - V -(Al- - - -( )-( )V - - V - P-P- CHS P-FYP P-P--P-P'YP'-- P-P--P-rYr- --- L-V- ,n - ' - -(V)(V)- •• -( )-{ )-{ XV)- - - I.It w ue N • PEPGAPCP'YP'THSILTVTEEEWNSCETYTCVVGHBALAPHLVTBR -IV>- P - ESHPNGTFSAIGVASVCVEOlrfNNRIEFVCTVT .... Wt P'LP----F CCLAROFLPSTISFTWHYQNNTEVIO 11.1: -' DVWGAGTTVTVSSESOSfPNVFPLVS 11, CNB-9 W279 LJi CHB-9d H104E W279 Hl04£ P - Gp I HP N N cc T f Y NO I r I GI A f LT v o I s s s TAY HO 1,1 i CN6' L V - GYISYSCSTYYHPSLll:SRISITROTSll:NQYYLQLG - -P-V-VFV W279 P - t!VQLQESGPSLVIPSQTLSLTCSYTGOSITIG'll•NI NPNVNVFV u 1 CNS' 8 15 ~ fraq ■ ents EVQLQOSGP!LVIPGASVINSCIASCITFTDl'INIWV HlOH: LJs CN6 W279 Hl04E µ W279 Hl04E HxW LJS CH.lb W279 ._. l Hl04£ HxW W279 LJ I I Anrdno Ter•inal Sequence Analyses of W279 Hu Chain CNBc Hl04£ u a CNl-2 5 TABLE 2. w r.o ..... 140 Footnotes to Table 2 aMOPC 104E sequence from Kehry et al., 1979. bThe sequence of CN 3a was determined on intact µ s chain; the sequence of CN3b was determined on CN3 derived from µ s chains succinylated prior to CNBr cleavage. Location of CN3b in the V region was determined by alignment and homology with invariant VHI sequences (Kabat, Wu and Bilofsky, 1976). ( ) indicates residues with low yields. A key to the single letter amino acid code is provided in Dayhoff (1976). Fragments differing in molecular weight from 104E µs chain fragments are denoted by a' (prime). cTwo exceptions to the identity of the W279 µ. and 104E µ l S C region sequences have bee noted in CN7. Both residues (Tyr at position 399 and Ala at position 417) were recovered in low yield, indicating they were not a part of the major CN7 sequence. We have not identified a possible contaminating sequence which would account for the two discrepancies. d ( ..... ) sequence not analyzed. 141 Figure 1. Three Species of µ Chains Synthesized by W279 Cells Composite tracing of two-dimensional 10% polyacrylamide-SDS gel fluorographs of 3H-tyrosine and leucine labeled W279 µ chains. The tracing represents the alignment of three different two-dimensional gels using stained marker M104E ' µ chains as an internal standard. Crosshatching of the µ and light chain spots s indicates the complete blackening of the X-ray film. µ.,µ-internal;µ m, µ-membrane; µ , µ -secreted. The apparent molecular weight of the µ chains in the s electrophoretic system is approximately 78,000 daltons; light chains 22,000 daltons. 1 142 basic I/') I/') ... Q) acidic ~ 0 .c µ.5 a. -... 0 u Q) w light FIGURE 1. 143 3 Figure 2. Galactose Residues in W279 µ m Chains Labeled with NaB H and 4 Galactose Oxidase Two-dimensional gel analysis of a W279 cell lysate labeled with NaB 3H and galactose 4 oxidase as described in Experimental Procedures. After removal of the free radioactivity, the lysate was precipitated with rabbit anti µ. The basic spot at the center left of the gel labels nonspecifically when galactose oxidase is not included in the labeling reaction and is precipitated under our conditions by rabbit anti µ (data not shown). The pl ranges of µ i and µ m chain are indicated. 144 Basic Acidic \ (/') (/') ... Q) 0 . - ..c. a.. ... 0 u Q) -w -: , - - '.-~-·- ~ , :· •• · :4~~:~~~~""" •°":- . : . FIGURE 2. 145 Figure 3. Separation of M104E µ s Chain Cyanogen Bromide Fragments Completely reduced and aklylated M104E µ s chains (35 mg) were cleaved with cyanogen bromide and separated by gel filtration on a column of ACA54 (LKB) as described in Experimental Procedures. Fraction volume was 5 ml. The fragments are labeled (CNl to CN9) in the order in which they appear in the µ chain sequence. At the top of the figure is a schematic drawing to scale of the M104E µ s chain showing linear order and sizes of the cyanogen bromide fragments (CNl to CN9) and the attachment sites of complex (boxes) and simple (circles) carbohydrate moieties. Agg. denotes aggregate peak. FIGURE 3. 70 I 90 I I I I 110 I ~ I A~. I\ 0.IL 00 2so CN7 CN6 A 0.2, c,. 130 CNS I I 150 I 11 CN3 A CN4 170 c N f f J CN7 166 109 Fraction Number I I CN5 CNBr C1Nlf~ CN3, ;r,N5 Fragments 2014 47 62 84 v. 190 CN2 62 8 210 CN9 I ~ 0) ..... 147 3 Figure 4. Comparison of H-Mannose Labeled Cyanogen Bromide Peptides of µ. and µ 1 S Chains by Gel Filtration 3 W279 or Ml04E cells were incubated with H-mannose for 16 hr or 8 hr, respec1 tively, in low glucose (1000 mg/ml) Dulbecco's Modified Eagle's Medium. The µ ., µ 1 m and µ chains were preparatively isolated as in Experimental Procedures. s 3 No H-mannose was incorporated into light chains. Cyanogen bromide cleavage in the presence of carrier Ml04E µ s chains and gel filtration on a column of ACA54 was performed as described in Experimental Proced,ures. The Rf's of the fractions from the different column runs were then normalized to the internal standard Ml04E µ s chain cyanogen bromide fragments. Normalized fractions from the internal standards were then superimposable (not shown). The peak running with the excluded volume consists of aggregated material. Free salt is eluted around fraction 285, and column runs were eluted 10 fractions past this point. Ea.ch fraction volume is 5 ml. At the top of the figure is a schematic drawing to scale of the W279 µ chain showing the sites of cleavage by cyanogen bromide, the respective frags ments (CNl-2' to CN9) and the locations of the five oligosaccharides (CHOl to CHO5). The complex oligosaccharides are boxed and the high mannose oligosaccharides are circled. Fragments were ordered by comparing their NH -terminal 2 sequences to those of Ml04E µ s chains (Kehry et al., 1979) and VHI sequences (Kabat, Wu and Bilofsky, 1976). Numbering of the fragments is based on Ml04E µs chain cyanogen bromide fragments. W279 µ i (----); Ml04E µs (-). The µi fragments differing in molecular weight from corresponding Ml04E µ cyanogen bromide fragments are denoted by a' (prime). 5 chain FIGURE 4. /J-i 2000 H-CPM 100 l0001- ------ 3 30 00 40001- I ~ I I I I I I I I I I I I l \J, ~ CN 5 84 ,_ CHO 180 200 220 FRACTION NUMBER 160 ' CNB-9 b. 8 ~I' CN? -~- ,,.CN 5 CN7 µ.i 140 \ I I 120 1 I I I I I I I I ~ II II I1 I CN 30 -50 - CN6' I\ : : 11 I I ,. µ.s-J I .... I I CN6 --n~ ~ W27 9 Mu CN }o CNS, F RAGM ENT S. 48 c~ 240 CN7 109 260 280 CN8 l cN9 62 ·a· HQ' 0 200 400 1-'-s 3 600 H-CPM 800 1000 ..... .i,.. 00 149 Figure 5. Radiolabeled NH 2-Terminal Sequence of W279 µ i CN5' W279 µ. CN5' labeled with 3H-Phe, Pro, Val, Tyr, and Leu was loaded on the l Caltech sequenator in trifluoroacetic acid-H 2o as described in Experimental Procedures. Cold carrier M104E CN5 (2 mg) was included. The Pth-amino acids at each step were separated by HPLC and counted for radioactivity. The yield of Leu at step 3 is lower than that at step 8 due to failure of the automatic conversion function for the first 5 steps in the sequenator. Steps 1 to 5 were converted manually, reducing the yield. The sequence is given in Table 2. 150 Wehi 279 µ. , CN5 1 12 20 800 600 400 200 PRO 100 ::E 200 VAL a.. u I ,..,:I: 100 200 TYR 100 200 100 8 24 RESIDUE NUMBER FIGURE 5. 28 151 Figure 6. Comparative Peptide Mapping of 35 35 s-Met Labeled µ. and µ 1 S Chains s-Met labeled W279 µ chains were combined with 0.25 mg pig IgG and cleaved with trypsin followed by exhaustive digestion with chymotrypsin as described in Experimental Procedures. Peptides were separated by HPLC (McMillan et al., 1979). Fraction volume is 0.5 ml (0.5 min/fraction). ( ...... )gradient of acetone used to elute peptides. µ s (-). 35 s-Met labeled W279 µ. (-----); 1 35 s-Met labeled MxW FIGURE 6. ~~ W279 µ, 1 MPCI I XW279.2p.5 ,., II II II II I - --- I I I 100 ~ l I ~ II ! I u 140 .· .. 1: .·. . 160 I 180 Q) C: 0 0 Ol ... 5 20 ~ 40 Cf) 0 60 ~ 80 7100 0__J0 200 400 r<l L() Cf) 600 I u Q_ ~ I I I I 800 I 1200 I 11600 ~----~J ..... .. 120 I I I.•· I I n -111 Fraction Number I ~ ... I I ,,,,I II II I o l~ , 80 0 20 40 60 10001- 2000~ 30001-- 40001-- Ul t-.:1 ,_. 153 Figure 7. Location and Order of W279 Cyanogen Bromide Fragments Schematic drawing of the W279 µ i chain showing the sites of cleavage by cyanogen bromide and the respective fragments in the VH and C µ regions as compared to Ml04E µ s chains. The location of methionine residues in the W279 VH region is very different from that in the M104E VH region. The sequence of W279 µ i CN9 has not been determined. Its identity to M104E CN9 is based on the alignment of a tyrosine-labeled cyanogen bromide fragment with the gel filtration pattern of Ml04E µ s· Therefore, CN9 has not been included in the diagram. Crosshatched regions indicate the regions where a radiolabeled sequence has been determined 3 3 for W279 µ i chains ( H-Phe, Pro, Val, Tyr and Leu for all µ i fragments, H-Lys and Ala for CN7 only). * The radiolabeled sequence of MxW µs chains was also determined for these cyanogen bromide fragments. :;: Includes the sequence of MxW CNS and CN8-9. See Table 2. FIGURE 7. FRAGMENTS µi CNBr WEHI 279 CN3a * * VH CN3b M0PC 104E CNI CN2 CN3 1-'s CNBr FRAGMENTS 2014 47 84 62 CN 1- 2' CN5 c,. CN5 CN4 CN6 166 CN6 CN7 109 CN7 1 8 CN9 CN8-9* 62 CN8 "" CJ1 ..... 155 Figure 8. Comparison of NaB 3H -Galactose Oxidase Labeled Cyanogen Bromide 4 Peptides of µ m and µ s by Gel Filtration Galactose residues of cell surface W279 µ_m chains were labeled with NaB 3H 4 as described in Experimental Procedures. Pentameric M104E IgM was similarly labeled in solution. The µ m and µ s chains were isolated, cleaved with cyanogen bromide and separated as described. The column runs were normalized to internal standard M104E µs cyanogen bromide fragments. The peak running with the excluded volume consists of aggregated material. Free salt is eluted around fraction 285. Fraction volume is 5 ml. FIGURE 8. I ~ 3000 100 1000 2000 I I I I I I 4000 .le .). --i· 3H-dpm µm ,~ I\ 120 ,I ,1 I I ,I I CN6 140 160 CN5 180 220 NUMBER 200 FRACTION 240 260 280 1000 f'-s 3H-dpm 2000 CJ) CJ1 ..... 157 3 Figure 9. Comparison of H-Mannose Labeled Cyanogen Bromide Peptides of µ m and µ s Chains by Gel Filtration See legend to Figure 4. 3 H-Mannose cyanogen bromide fragments of W279 µ (---) and M104E µs (--). m FIGURE 9. f-Lm 100 ol IOOOl 2000 3H-dpm 3000[ 4000~ Ii I I I I I ,, ·120 I ' :.tA IIII,, I I I 11 1 I I I I I I I I I I -140 160 I 200 --220 FRACTION NUMBER 180 - 240 _ __ 260 280 0 , - - , - ~ 200 CN8-9 6 8 AA 1-'--s 400-- CN5 3H-cpm 600 -f 800 -11000 CN7 • I f/LcNG ti t ,, I r..-µm U1 0:> ..... 159 Figure 10. Glycopeptides of µ 3 m and µ. Chains Are Different 1 H-Mannose labeled W279 µ chains were prepared as described, cleaved with trypsin and chymotrypsin and the glycopeptides separated by HPLC. Fraction volume is 0.5 ml (0.5 min/fraction). ( .... ) gradient of acetone used to elute peptides. 3 3 H-Mannose labeled W279 µ. (----); H-mannose labeled W279 µ 1 m (--). The major peptide difference is indicated by crosshatching of the peak. The flowthrough peak from the column which is seen only in µ. is not included as a peptide 1 difference. 160 ........ iua.-.1os ::>!U06JQ % 0 0 0 0 0 co -~dG-H § 8 0 V <.O £ 0 C\J 0 0 I{) IO 00 co ''I '' I I 0 I <.O ' i I I ~' 0 V I I _,> '""""-.....,..~~ 0 ------ . C\J ... (1) ..0 ---------- 0 E :::, QZ C -... 0 0 CD <.) a LL 0 <.O -I 0 E V ::1. ::1. O') O') N N ~ ~ r-- r-- 0 N ----•="=----==~:.---:0 0 0 (\J 0 0 I{) 0 0 0 ~d0-H£ 0 0 IO 0 0 161 Figure 11. Comparison of 3 H-Amino Acid Labeled Cyanogen Bromide Peptides of µ i and µ m Chains by Gel Filtration 3 3 W279 cells were incubated with either H-Tyr and Leu(µ m) or H-Tyr, Leu, Ala, and Val ( µ i). The µ i and µ m chains were isolated, reduced, alkylated, cyanogen bromide cleaved and the peptides separated by gel filtration as described. Aliquots of isolated µ. and µ 1 m chains used for the cyanogen bromide cleavage were rerun on two-dimensional SDS gels to check the purity. The µ m and µ i chains were always 90% free of cross-contamination. The individual column runs were aligned by normalization to internal standard M104E µ s chain cyanogen bromide fragments. The peak running with the excluded volume consists of aggregated material. Free salt is eluted around fraction 285 {just after CN9). Fraction volume is 5 ml. 3 H- amino acid cyanogen bromide fragments of W279 µ m (---)andµ i (-). 0.2 ml aliquots were counted for µ i and 0.4 ml aliquots for µ m through fraction 250; 0.7 ml aliquots were counted for fractions 251-290. Cyanogen bromide fragments differing 1n molecular weight from M104E µs cyanogen bromide fragments are denoted by a '(prime). CN8 and CN8-9 in µ i and CN8-9 in µ m are not distinct peaks due to the migration of a V region incomplete Met-Ser cleavage product (CN3b-CN1-2', detected by sequence determination) between CN5 and CN8-9. FIGURE 11. 3 1-'-i H-DPM 0 l000 2000 3000 4000 100 ~-A CN6, I I 120 l I I I II II I I I ~ I JJ-m '---1I\ \ I 140 CN5' I ( j\/l,CN5 180 200 A I I I I I I I I I I I I 220 ~CN3 I FRACTION NUMBER 160 I I I I I \ I I I I I I I I /11 JJ-i CN 7 CN6' ,, 240 260 280 0 1-'-m H-DPM 1000 3 2000 3000 ~ C1) ...... 163 Figure 12. Comparative Peptide Mapping of See legend to Figure 6. 35 35 s-Met Labeled µ s-Met labeled W279 µ m (---); 35 m and µ s Chains s-Met labeled MxW µ s (-). Peptide differences are indicated by crosshatching of the peaks. FIGURE 12. r<> lO 0 ¥\ I• ,, W279 µm ,I 1'1 I . I I 100 .. . ..·•.· .· I 1 400 800 1200 ~ Cf) C 0 20 ~ 0 0 '- cr, 0 u Cf) 0 ~ Q) -- 40 ·c 60 2l= -; 80 1600 u I I ,100 I I 2000 -12400 I ""j. ~..._:E: .... ;a b ;J 0 140 160 180 120 FRACTION NUMBER 1 ' 'fv\ J_I'r•\I ....•··lit•\ .11,.," m ' I MPC 11 X W279.2 µ 5 - - 1~-~····+71'""'1 40 60 80 20 0 10001- 20001- en 3000 I u ~ 4000 Q.. I 60001 80001- ..... ~ O"> 165 3 Figure 13. Comparative Peptide Mapping of H-Phe Labeled µ See legend to Figure 6. 3 H-Phe labeled W279 µ m (---); 3 m and µ s Chains H-Phe labeled MxW µ s (-). Peptide differences are indicated by crosshatching of the peaks. FIGURE 13. ~~ 10001-- 2000f 3000t- 4000 W279 J.Lm I 100 120 Fraction Number 80 11 I . I' . 140 I 3000 ..... 160 100 0 180 roo r<> II 0 i j 0 20 40 60 ~80 1 I7 . 4000 .···••••....••12000 .·• .·. / .... l' I 1,: :, : '1 ~~·n:~ Ilj II rn:i: • 1111 • JI II I MPCII x W279.2 µ5 - - - - 0~ O 'i (f) 0 Q.) -> J 0) 0) ...... 167 Figure 14. Release of Amino Acids from µ m , µ. and µ 1 s Chains upon Digestion with Carboxypeptidases A and B W279 µ. and µ 1 m chains and MxW µ s 3 chains labeled with R-Tyr, Val, Lys, Phe and Leu were digested with carboxypeptidases A and B for varying times as described in Experimental Procedures. The µ chains labeled individually with each amino acid were pooled for the digestion, so that the quantitation of residues released is based on radioactivity in the original undigested µ chain, the proportion of the total sample removed and loaded on the amino acid analyzer for each time point, and the number of residues of each amino acid present in the W279 µ chain [22 Tyr, 45 Val, 31 Lys, 22 Phe, and 46 Leu (Kehry et al., 1979)]. Controls incubated for 2 hr on µ m' µ i and µ s chains consisted of an undigested aliquot to which boiled carboxypeptidases A and B were added. No amino acids were released in any of the control incubations. A. Release of COOR-terminal amino acids from W279 µ m chains. B. Release of COOR-terminal amino acids from W279 µ. chains. . 1 C. Release of COOR-terminal amino acids from MxW µ s chains; these chains 3 were not labeled with R-Lys and Leu. 168 A 0.7 VAL 0.5 A - - - - - t> LYS o TYR 0--------0 0.9 TYR ;:: 0. 7 0 C: 0.. ~ 6' o.s u < 0 :z: :.: < 0. 3 "' ...J 0 :c . 0.9 0 .7 0.5 o.,....-o - - - 0. l o------o -------------0 TYR .-•----------~ • VA L PHE 20 40 60 80 TIME (M IN ) FIGURE 14. 100 120 169 Figure 15. Schematic Drawing of the Structure of µ m' µ s and µ i Chains The VH and four C µ. or precursor µ 1 s µ domains are represented for the µ m (left), µ s (center), and (right) chains. The C-terminal (Ct) segment of µ m chains is shown as being longer and different in amino acid sequence (crosshatching) from the Ct segments of µ S and µ. chains. CHO indicates the sites of carbohydrate 1 attachment in the µ chains with boxes representing complex type carbohydrate moieties (dashed lines showing complex type carbohydrates lacking terminal sugars) and ovals representing high mannose carbohydrate moieties. 170 flm V V V CHI CHI CHI CH2 CH2 c~ -.fc_A_Q~ --:t:8Q: - CH3 CH4 FIGURE 15. a CH3 CH3 HO CH4 CH4 ct ct -- I -{c8oJ 171 Figure 16. Amino Acid Sequences of the C-Terminal Regions of µ m and µs Chains The amino acid sequence of the C-terminal region of µ chains (Kehry et al., s 1979) and the predicted amino acid sequence of the C-terminal membrane segment of µ m chains (Early et al., 1980; Rogers et al., 1980) are compared. The junction between the C µ 4 domain and the C-terminal domain is delineated. Boxed residues are described in the text. CHO indicates the site of carbohydrate attachment. A key to the single letter amino acid code is provided in Dayhoff (1976). FIGURE 16. /J.m ,,_, csecr~ted terminus M segment .•.. o K s rj E G E v NA EE E G[IE N L wTT As T[E]I v L F L L s L F Y s T T v T L F !K VIK Cµ.4 CH •••• DK s TIG KP TL~~ s L rffi!js o T G GTic vi Cu4 r N) -:i r-" 173 CONCLUSION 174 IgM Structures and Functions The structure of secreted IgM is well adapted to the functions performed by this 19S IgM molecule. The tertiary structure of the pentamer seems to be largely determined by the structure of the secreted µ ( µ s) chain. As is summarized in Chapter 1, µ s chains are composed of four C region domains in addition to a 20 amino acid COOR-terminal domain (1). Primary amino acid sequence homology among µ s chains from different species reveals that the more COOR-terminal regions are strikingly conserved (2). In accordance with this high degree of structural similarity among such diverse species as mouse, human and dog, the C µ 4 and C-terminal regions perform specific effector functions of the IgM molecule. The C-terminal domain plays a clear role in polymer formation. Only one other class of heavy chains contains a 20 amino acid C-terminal region. This is the a chain, the heavy chain present in IgA (3). IgA is the only other antibody class which is polymerized (usually found as dimer or trimer secretory IgA). The COOR-terminal 20 amino acids of human µ and a chains are 65% homologous (3, 4) including the location of a high mannose carbohydrate. In both a and µ chains, the penultimate cysteine residue is involved in formation of the covalent disulfide bridges between the heavy chains of monomers and between heavy and J chains. IgM and IgA are also the only antibody molecules which are polymerized with the J chain (5). The highly conserved C µ 4 domain of IgM pentamers plays an important part in activation of the classical complement pathway after antigen binding. A region has been localized in C µ 4 which is thought to be the site involved in the binding and activation of the first component of complement (6). It is notable that antigen binding by a single IgM pentamer is sufficient to trigger the entire complement cascade (7), culminating in the lytic destruction of foreign cells (8, 9). IgG molecules may also activate complement, but only when at least two monomers 175 are aggregated by antigen binding (7). Thus, 19S IgM is an efficient "pre-aggregated" effector antibody. In addition, certain types of antigens elicit predominantly IgM responses. This so-called "T-independent" antigenic stimulation occurs primarily when the immunogen consists of large multivalent complexes, most commonly polysaccharides, for example constituents of bacterial cell walls (10). The polyvalent IgM made in response to such a multivalent antigen is therefore appropriately designed to efficiently eliminate the foreign substance from the body. The function of the five carbohydrate moieties attached to each µ chain s in the Cµ region is unclear at present. Perhaps the oligosaccharide moieties contribute to the solubility of the IgM molecule in the bloodstream. Certainly secreted µ s chains are relatively insoluble when compared with other serum proteins and y, a or light chains. Perhaps, too, the carbohydrate contributes to the overall structural conformation of the IgM molecule, even rendering it resistent to proteolytic digestion (11, 12). In addition, the carbohydrate could determine plasma clearance of IgM or its intracellular transport. The only other antibody class which contains an additional CH region domain is the IgE class (1). Again, the significance of four CH domains in IgM and IgE molecules is unclear. Since individual immunoglobulin domains probably do perform discrete functions (13), we would expect IgM and IgE molecules to have a common function. One similarity is that both IgM and IgE molecules can be found attached to the surf aces of cells. IgM is bound to the membrane of resting B lymphocytes and IgE binds to the surface of mast cells via an Fe receptor (14). In both cases, the V regions are exposed to the external mileau for the purpose of antigen binding (13, 14). Indeed, the longer CH region may allow the cell surface IgE and IgM molecules to extend outwards from the cell membrane, thereby facilitating antigen interaction. In addition, the lack of a hinge region in IgM molecules 176 may serve to enforce a rigidity which could be important in the functioning of the membrane receptor IgM molecule. This brings us to the specific structural adaptations incorporated into the membrane IgM molecule. Logically, the most efficient structure for membrane receptor molecules which have the ability to recognize foreign substances and elicit the production of specific antibodies is the antibody molecule itself. What other macromolecule could be designed to convey such exquisite specificity and capacity for memory to the B cell surface? Membrane IgM incorporates all the structural features which characterize the major antigen binding and effector functions of the IgM pentamer. Only in one aspect do membrane and secreted IgM molecules differ structurally. The region governing polymerization is absent in membrane µ (µ m) chains (15, 16, 17). Instead, µ m chains possess a C-membrane terminal region COOR-terminal to the C µ 4 domain. This extra stretch of 40 amino acids seems well suited for interaction with the hydrophobic lipid bilayer. A central region of 25 uncharged amino acids is flanked by groups of charged residues (16, 17). The uncharged amino acids in µ m chains are capable of forming an alpha helix structure sufficient in length to span the lipid bilayer (18, 19), and these amino acids may serve as two membrane "anchors" for each monomeric IgM molecule. Charged residues present on the immediate interior or exterior of the plasma membrane may further stabilize membrane attachment by interacting with charged lipid heads and prevent the newly synthesized µ m chain from passing through to the interior of the endoplasmic reticulum membrane. All but one of the sites of carbohydrate attachment are identical in µ s and µ chains (15). The fifth oligosaccharide, attached to the C-terminal region m in µ s chains, is absent in µ m chains due to the absence of an Asn-X-Ser or Thr recognition sequence in the C-membrane terminal domain (16, 17). The function 177 of the fifth carbohydrate must then in some way relate to the polymerization and/or structure of the pentamer, since polymeric IgA molecules also have an identical carbohydrate moiety in the C-terminal region of the a chains (3, 20). The remaining carbohydrates on the µ chain must therefore be required directly or indirectly for maintaining IgM structure related to antigen binding and may be required for carrying out effector functions. IgM Expression: The C µ Gene B cells, then, are remarkably efficient in their mechanism of switching from membrane to secreted IgM expression. Regulation may occur at multiple levels: transcriptional, mRNA processing, translational, and/or post translational. Transcriptional control is exerted in the initiation of µ chain synthesis (including allelic exclusion) by a pre-B cell (16, 17). This step is probably related to the complex DNA rearrangements which occur in moving the VH gene closer to the Cµ gene _(16, 17). Knowledge of the control steps involved is sparse. Since the Cµ gene is present in only one copy per haploid genome (16, 21), eachµ gene primary transcript has the option of being processed for either µs or µ m chain expression. In the germline C µ gene, the coding regions for the C µ 4 and secreted C-terminal domains are contiguous (16, 22). 3' to the secreted Cterminal coding region is an 1800 nucleotide intervening sequence followed by a region of DNA encoding the C-membrane terminus (16). Expression of µ m chains simply entails removing the intervening nucleic acid sequences from the primary RNA transcript between the end of the Cµ 4 region to the beginning of the C-membrane terminal region (17). Removal of the secreted C-terminal sequence is mediated by an RNA splicing site located in the coding region of the secreted C-terminal domain (16, 17). Therefore, µ s mRN A expression is obtained by removing RNA sequences 3' to the secreted C-terminal domain. Plasma cells have differentiated so as to process the majority of µ gene transcripts in the 178 latter fashion. Presumably, this differentiation involves expression of specific RNA splicing enzymes. Many B cell tumor lines, though, seem to synthesize equal quantities of µ m and µ mRNA (23, 24). Yet in some of these cells, µ chains are synthesized s s and turn over at a much more rapid rate than µ m chains (25). Clearly, translation of the µ s and µ m mRN A pools is not equally efficient. In addition, synthesis of J chain is partially limiting for IgM polymerization and secretion. We are now able to offer an explanation for how antigen stimulated B cells secrete IgM without initially requiring RNA transcription (26), and how they may develop into IgM secreting blast cells in the absence of DNA synthesis (27). Activation of J chain mRNA translation would allow rapid maturation and polymerization of the presecreted pool of µ chains inside these cells. In addition, the relative stabilities of µm and µs mRNA following antigenic stimulation may change to favor µs synthesis. Changes accompanying large-scale IgM secretion, though, would require transcription of additional copies of µ mRN A (26). s Finally, post translational modifications, in particular glycosylation, may also regulate the rate of IgM maturation and secretion. Carbohydrate must represent an important aspect in IgM structure and/or function, since exact locations of oligosaccharide residues have been conserved in mouse and human µ chains throughout evolution (2). Treatment of plasma cells with the glycosylation inhibitor, tunicamycin, produces an 80% inhibition of IgM secretion by these cells (28). It is not unreasonable that maturing plasma cells might in part increase the level of IgM secretion by increasing the availability of oligosaccharide precursors or glycosylating enzymes. Future Questions on B Lymphocyte IgM Many important aspects of B cell IgM and the control of IgM synthesis 179 remain to be clarified. This does not even include acquiring knowledge about the molecular mechanisms underlying B cell triggering. Additional studies can be directed towards elucidating the structure and function of the proteins and of the nucleic acids involved in carrying out the B cell differentiation process. Experiments on B lymphocytes at the protein level address the functional role of IgM in the cell. 'no membrane-bound IgM molecules actually have a transmembrane orientation in the plasma membrane? Attempts should be made to label and identify regions of the µ chains which may be exposed to the cytoplasm. m A related problem is the degree to which the IgM molecule is exposed on the cell surface. Do portions of the C 3 or C 4 domains interact with C 3 or C 4 domains µ µ µ µ in other IgM monomers as in the pentamer structure, or are there any C µ domain interactions with unidentified effector molecules? Although not simply interpreted, crosslinking experiments may be used to identify regions of the µm chain which specifically interact with the lipid bilayer, with membrane proteins on the cell surface, ·or with cytoplasmic proteins. Ideally, antigen activated cells would yield more interesting receptor IgM protein associations which are significant in the immune response. The availability of inducible cell lines will be useful in future studies on the topography of membrane IgM molecules in activated cells (29). The identity of µ. chains in early pre-B cells is also unknown. Does this l µ.I pool consist of pre-µ m molecules, pre-µ S molecules or both? The tools, including protein and nucleic acid chemistry, are currently available for investigating this point. One of the most important considerations, and I believe often overlooked, is that results obtained on B cell tumor lines should be confirmed on normal B cells from spleen, lymph nodes and blood. Tumor cell lines are known to be polyploid and they possess abortive gene rearrangements. Thus, especially in questions 180 regarding the nature of irnrnunoglobulin genes and their transcription, normal B cell RNA and DNA must also be examined. Nucleic acid chemistry is already well enough advanced in the murine system to approach several problems in B cell development. Defining the control mechanisms involved in µ vs. µ rnRNA processing would greatly advance knowls rn edge about changes occurring in developing B cells. The subject of c switching 8 (expression of a given VH region with the C µ' C O, Cy' or Ca genes) also can be examined in greater detail. What is initially responsible for specifically turning on µ gene transcription? What steps are involved in processing and removing the interdornain intervening sequences in antibody rnRNAs? Perhaps even more approachable is the problem of defining when given genes are expressed in the lineage from stern cells to plasma cells. When are J chain genes and light chain genes actively expressed and when does the expression of the C µrn and Cu.r genes cease? Are, in fact, membrane IgM and IgD molecules expressed in plasma cells at low levels? For example, MOPC 104E rnyelorna cells process 10% of their µ gene transcripts to f orrn µ mRN A (16, 17). Is this µ rn m gene expression by terminally differentiated plasma cells normal and are these messengers translated and the proteins utilized for specific purposes? Examining nuclear and cytoplasmic rnRNA populations from B cells at different developmental stages for the presence of sequences complementary to J, light, µ and o chain genes may begin to define developmental expression of irnrnunoglobulins. There are obviously many problems remaining to be explored on B cell receptor irnrnunoglobulin molecules. Triggering B Cell Differentiation What, then, has our structural information on B cell receptor IgM molecules contributed to the understanding of B cell triggering by antigen? Characterization 181 of the IgM receptor is only the first step in studying B cell triggering. Now that we know membrane IgM is indeed an integral component of the lymphocyte plasma membrane, we can look for the involvement of IgM in transmembrane signaling processes. This is an enormous step forward. If cell surface IgM had been only peripherally associated with an accessory membrane protein, that secondary protein would have required characterization in order to understand its role in antigen triggering. Therefore, the path is now clear for concentrating on the primary interactions of membrane µ chains with other molecules in the B cell membrane. The molecular mechanisms involved in antigen-B cell (and T cell) interaction and the mechanisms of triggering of B cell differentiation and proliferation remain wide open to speculation - and to experimentation. Certainly insights into the triggering process will be obtained when we know: 1) What molecules on the T and B cell surfaces and even on macrophages interact with one another? 2) What direct receptor interactions are stimulated by the polyclonal B cell activators, such as LPS, PPD, and lectins? Is this nonspecific form of B cell activation similar to antigen induced triggering? 3) What role in triggering is played by the membrane IgD molecule? Since the majority of IgD molecules are membrane bound [extremely low levels are present in the bloodstream (30)], IgD must function primarily as a cell surface receptor. Do IgD molecules interact specifically with membrane IgM or with other molecules that also interact with IgM? And so on. The next few years should offer exciting opportunities for making new discoveries about immunoglobulins and B lymphocytes. 182 REFERENCES 1. Beale, D., and Feinstein, A. Quart. Rev. Biophysics 9:135 (1976). 2. Kehry, M., Sibley, C., Fuhrman, J., Schilling, J., and Hood, L. E. Proc. Natl. Acad. Sci. USA 76:2932 (1979). 3. Liu, Y.-s., Low, T. L. K., Infante, A., and Putnam, F. W. Science 193:1017 (1976). 4. Putnam, F. W., Florent, G., Paul, C., Shinoda, T., and Shimizu, A. Science 182:287 (1973). 5. Della Corte, E., and Parkhouse, R. M. E. Biochem ~- 136:597 (1973). 6. Hurst, M. M., Volanakis, J. E., Stroud, R. M., and Bennett, J. C. _!!. Exp. Med. 142:1322 (1975). 7. Mliller-Eberhard, H.J. Ann. Rev. Biochem. 45:697 (1975). 8. Podack, E. R., Biesecker, G., and Mi.iller-Eberhard, H. J. Proc. Natl. Acad. Sci. USA 76:897 (1979). 9. Shin, M. L., Paznekas, W. A., Abramovitz, A. S., and Mayer, M. M. J. Immunol. 119:1358 (1977). 10. Coutinho, A., and Motter, G. Nature 245:12 (1973). 11. Plaut, A., and Tomasi, Jr., T. B. Proc. Natl. Acad. Sci. USA 65:318 (1970). 12. Shimizo, A., Watanabe, S., Yamamura, Y., and Putnam, F. W. Immunochem. 11:719 (1974). 13. Hood, L. E., Weissman, I. L., and Wood, W. B. Immunology. The Benjamin/ Cummings Publishing Company, Inc., Menlo Park, California (1978). 14. Metzger, H. Transplant. Rev. 41:186 (1978). 15. Kehry, M., Ewald, S., Douglas, R., Sibley, C., Raschke, W., Fambrough, D., and Hood, L. CelL In preparation. 183 16. Early, P. W., Rogers, J., Davis, M., Calame, K., Bond, M., Wall, R., and Hood, L. Cell. Submitted. 17. Rogers, J., Early, P., Calame, K., Carter, M., Bond, M., Hood, L., and Wall, R. Cell. Submitted. 18. Chou, P. Y., and Fasman, G. D. Biochemistry 13:222 (1974). 19. Lim, V. I. i· Mol. Biol. 88:857 (1974). 20. Torano, A., Tsuzukida, Y., Liu, Y.-S. V., and Putnam, F. W. Proc. Natl. Acad. Sci. USA 74:2301 (1977). 21. Gough, N. M., Kemp, D. J., Tyler, B. M., Adams, J. M., and Cory, S. Proc. Natl. Acad. Sci. USA 77:554 (1980). 22. Calame, K., Rogers, J., Early, P., Davis, M., Livant, D., Wall, R., and Hood, L. Nature. In press (1980). 23. Wall, R., and Rogers, J. Unpublished results. 24. Perry, R. P., and Kelley, D. E. Cell 18:1333 (1979). 25. Sibley, C.H., Ewald, S. J., Kehry, M. R., Douglas, R. H., Raschke, W. C., and Hood, L. E. i· Immunol. Submitted. 26. Melchers, F ., and Anderson, J. Biochemistry 13:4645 (197 4). 27. Anderson, J., and Melchers, F. Eur. i· Immunol. 4:533 (1974). 28. Hickman, S., and Kornfeld, S. i· Immunol. 121:990 (1978). 29. Paige, C., Kincade, P., and Ralph, P. i· Immunol. 121:641 (1978). 30. Finkelman, F. D., Woods, V. L., Berning, A., and Scher, I. J. Immunol. 123: 1253 (1979). 184 APPENDIX 185 SUPPLEMENTARY FIGURES Included in the appendix are my supporting and unpublished results consisting of sequence determinations, representative SDS-polyacrylamide gels, preparative and analytical gel filtration and HPLC peptide maps. These data are not presented with the intention of drawing any new conclusions, but merely for purposes of documentation. A complete description of each figure and a statement as to the relevance of the data are provided. All of the figures except for the sequence determinations are presented in chronological order which parallels the preceding chapters. 186 Figure 1. Purification of MOPC 104E 19S IgM by gel filtration. Proteins were precipitated from 40 ml MOPC 104E ascites fluid (grown in BALB/c x DBA/2 mice) at 4°C in two successive cycles by a final concentration of 50% saturated ammonium sulfate each time. The precipitate was dissolved in 13 ml borate buffered saline (BBS) and chromatographed on a column of ACA22 (LKB) (5 x 100 cm) in BBS. Fraction volume was 12 ml. Bar indicates 19S IgM fractions pooled. Excluded volume consists primarily of IgM in an aggregated form. Small peak following the HIS Igl1/ peak consists of a-2 macroglobulin (MW 800,000) anc is a major contaminant of the 19S IgM. The arrow indicates position of salt elution. The pooled 19S lgl\1 is rendered at least 95% pure by this one step procedure, as assessed by 10% polyacrylamide-SDS gel electrophoresis (1). FIGURE 1. 80 0"=:::ita:::.......l.__ 0.4 0.8 1.2 1.6 0.0.280 2.0 2.4 100 _.L__ __L_ 19S IgM 120 160 _..L_ FRACTION NUMBER 140 __J__ _L _ _ . 1 . . . . _ _ . J . . __ ____J__ 180 200 ! _ _ _ . l _ ~. . . . .--..L.......J 104E ascites -.:i f--' 00 188 Figure 2. Separation of H and L chains of MOPC 104E 19S IgM by gel filtration. Pooled 19S IgM in BBS was precipitated at 4°C by 50% saturated ammonium sulfate and the precipitate dissolved in 6 M guanidine-HCl, 0.25 M Tris-HCl pH 8.5, 0.15 M NaCl, 0.02 M EDTA. Disulfide bonds were completely reduced with 0.01 M dithiothreitol (DTT) for 1-1/2 hr at 37°C under N . Free thiol groups were 2 then alkylated with 0.025 M iodoacetamide (3X recrystallized) overnight in the dark at 4°C. Guanidine concentration was diluted to 4 M by addition of 0.4 M ammonium bicarbonate, and µ and light chains were separated on a column of ACA34 (LKB) (5 X 100 cm) in 3 M guanidine-HCl, 0.2 M ammonium bicarbonate, 0.02% NaN . 3 Fraction volume was 10 ml. Bars indicate µ and light chain fractions pooled. The arrow shows the position of salt elution. Contaminating et-2 macroglobulin subunits (MW 90,000) were removed by this procedure. The pooled µ chain is at least 95% pure as assessed on 10% polyacrylamide-SDS gels (1). FIGURE 2. 0 0.4 0 .8 1.2 O.D.200 1.6 2.0 50 60 L.:--_1__ _1_.._ 70 90 100 _ . L_ 110 120 __::::::t:::::,--..___ FRACTION NUMBER 80 _ . L_ __i__...:_.:_1-.._ fL chain ~ 130 140 _.__ _...__J 104E IgM ~ 00 ~ 190 Figure 3. Separation of 14 c-Cys CNBr fragments of MOPC 104E µ 5 chains. IgM was purified and disulfide bonds reduced as described in Figures 1 and 2. The free thiol groups were then alkylated with 50 µ Ci iodo (1- 14 c) acetamide (Amersham) for 30 min in the dark at 4°C. Iodoacetamide (3X recrystallized) was added to 0.025M and the reaction incubated overnight. Heavy and light chains were separated as described in Figure 2. Pooled 14 c-Cys alkylated µ chains were con- centrated (Millipore Molecular Separator) neutralized with 88% formic acid and desalted on a column of Bio Gel P-2 (2.5 x 40 cm) in 50% formic acid at 4°C. After concentration by partial lyophilization and adjustment of the formic acid concentration to 70%, the µ chains were cleaved with CNBr (50 mg/ml) for 20 hr in the dark at 4°C with constant stirring. After 11-fold dilution with water and lyophilization, peptides were dissolved sequentially in 8M guanidine-HCl, 3M guanidine-HCl 0.2M ammonium bicarbonate, and 0,4M ammonium bicarbonate (total volume= 12 ml) and chromatographed on a column of ACA54 (1KB) (3.5 x 140 cm) eluted in 3M guanidine-HCl, 0.2 M ammonium bicarbonate, 0.02% NaN 3. Fraction volume = 5 ml. Fifty µ 1 aliquots were counted for radioactivity after reading the 00 280 of the CNBr peptides. Total radioactivity in each peak is pro- portional to the number of alkylatable cysteine residues in each fragment. Note that CN3 contains no cysteine residues and is not labeled. CN2 and CN9 were first discovered from this CNBr digestion. The 00 280 readings were not carried any further than fraction 190 because no peptides were known to chromatograph past CNl-2. Tube numbers 198 and 209 (CN2 and CN9, respectively) were desalted on Sephadex G-10 (Sigma) in 5% formic acid, lyophilized, and the sequences determined. A third peptide, labeled CN3', was also found by this procedure which migrated between CN3 and CNl-2. CN3' was pooled from fractions 157-160 and the complete amino acid sequence determined. The sequence is listed in Figure 10 and is not present in any µ chain. There is also no homology to any immunoglobulin 191 Figure 3 - Continued chain. Alignment of this sequence which generates 50% related (not identical) residues can be made with the human J chain sequence, residues 84-116 (2). No information is available on mouse J chain, although the gene has recently been cloned (3). FIGURE 3. ~ 60 A9'j CN7 CN6 t,~ ~ CN4 I 120 IOO ~ I 14() 160 180 200 220 \ J tCNZ , ~f . < N 9 A,., "' CN5j 11 V v 0 .1 0 O.l 0 .4 0.5 06 0.1 0.8 0. I. I.I 1.2 240 roo 2000 }000 4000 14 ( - CPM 5000 -16000 tv c.o ..... 193 Production of Rabbit Antiserum Against MOPC 104E µ Chain 1. Ear bled 45 ml, clotted, retracted and centrifuged. This is normal preimmune serum, 24 ml. 2. Antigen is partially reduced and alkylated MOPC 104E µ chain, sonicated as a fine suspension, 2 mg/ml in Hanks Balanced Salt Solution (HBSS); kept frozen at -20°C. 2 weeks after pre-bleed: i.m. injection in right thigh with 1 ml emulsification of 0.5 ml CFA and 0.5 ml HBSS (0.15 ml 2 mg/mlµ + 0.35 ml HBSS). 300 µg immunogen total. 3. 1 week after first injection, injected identically in left thigh. 4. 5 weeks after second injection boosted s.c. on multiple sites on back with 1 ml emulsification identical to that used in first injection. 5. Test bled 10 days after boost. 6. Repeated cycle of boosting and bleeding 10 days later at 5 week (minimum) intervals. 194 Figure 4. Scanning electron micrographs of WEHI 279 cells. Fixed original clone WEHI 279 cells were attached to poly-L-lysine coated coverslips and prepared for scanning electron microscopy (4). Scale bar is 1 µm. A. Magnification, X 10,000. B. Magnification, X 12,000. The B cell surf ace is obviously not a smooth sphere, and the density distribution of IgM receptors on the microvilli is unknown. 195 A FIGURE 4A. 196 B FIGURE 4B. 197 Figure 5. 10% polyacrylamide gel electrophoresis of WEHI 279 IgM. 3 WEHI 279 cells were incubated in the presence of H-valine for 16 hr (5). Cells were lysed in the presence of 1 mM phenylmethyl sulfonylfluoride (PMSF), and IgM was immunoprecipitated from 100 µ 1 of lysate by incubation with 10 µ 1 rabbit anti MOPC 104E µ chain followed by 100 µl 10% suspension of formalinfixed Staphalococcus aureus (6). Proteins were eluted in a boiling water bath and electrophoresed on 10% polyacrylamide-SDS gels (1). Gels were sliced into 1 mm sections, the proteins eluted by incubation of each slice in 0.5 ml 0.01 % SDS overnight and counted for radioactivity. The bar indicates WEHI 279 light chains and the arrow shows the position of the dye pyronin Y. This is a representative one-dimensional separation of µ . and µ chains, and the valine labeled m chains prepared from this lysate were used in carboxypeptidase experi1 µ. and µ m ments (7). The peaks around fractions 37-44 are nonspecifically precipitated with 1 the rabbit antiserum, can be removed under more stringent immunoprecipitation conditions (including deoxycholate), and based on two-dimensional gel electrophoresis are partly composed of actin molecules. FIGURE 5. I rt> I t ~ X 0 ~ 0 2 I I 31 4 51-- 10 20 I,"", • I 30 ~ I t n/ µ i A I 40 60 50 SI ice Number 70 . v ~...... J,l I ll n light 80 I 90 t- dye Val-Wehi 279 I 1-4 00 (0 199 Figure 6. Two-dimensional gel electrophoresis of WEHI 279 µ chains. WEHI 279 or MPCll x W279 cells were incubated in the presence of 3Htryosine and leucine for 16 hr (5). Cells were lysed, medium collected, and IgM was immunoprecipitated by incubation with rabbit anti µ followed by fixed~aureus (6). Proteins were eluted from the~- aureus and electrophoresed on 10% polyacrylamide-SDS gels (1) which were sliced into 1 mm sections. The proteins were eluted from the gel slices by incubation of each slice in 0.5 ml 0.5% SDS. Eluted µ chains were pooled, aliquots removed, and the remaining µ chains cleaved with CNBr. To check for purity in separation of the WEHI 279 µ. and µ m species, the aliquots were electrophoresed on two-dimensional gels (8) which were processed l by fluorography (9). The isoelectric focusing gels contained 5% polyacrylamide, 0.28% bis-acrylamide, 9.2M urea, 2% Triton X-100, 2% pH 3.5-10 ampholines, 0.13% pH 5-7 ampholines, 0.07% pH 3.5-5 ampholines, 0.13% pH 7-9 ampholines, and 0.13% pH 4-6 ampholines (LKB). The focusing gels were equilibrated 30 min, quick frozen and stored at -70°C. Thirty minutes before running the second dimension, gels were thawed and equilibrated in fresh SOS-sample buffer (8). Electrophoresis on 10% polyacrylamide-SDS gels was carried out in the direction of the arrow. Gels were aligned by superimposing the stained carrier MOPC 104E µ s chains in each gel. Only the region of the two-dimensional gels surrounding the µ chains is shown; the remainder of each gel was blank. Small black dots were made with a needle and used to align the gel and the fluorograph. The regions where µ i' µ m and µ s run are indicated by the bars. Note that less than 5% cross contam- ination of µ i and µ m chains is seen. A. WEHI 279 µ i chain. 4 day exposure. B. WEHI 279 µ m chain. 4 day exposure. A band corresponding to µ i is discernible in an 8 day exposure. 200 Figure 6 - Continued C. MPCll x W279 µs chain. 3 day exposure. Note that the pl of the µs band is shifted slightly to the basic e~d relative to µ m· We attribute the shift to inefficient glycosyla tion of these µ s chains by the hybrid cell line. When these µ s chains were cleaved with CNBr (see Figure 9), the CNBr fragments containing complex carbohydrates (CNS especially) were slightly smaller in molecular weight than those of µ m or MOPC 104E µ (similar instead to WEHI 279 µ .). This confirms S l the idea that the hybridoma µ s chain is not reliably glycosylated. 201 basic acidic A l ,,.. µ1 /J:i ---I I " 8 µm 1'-m - C fLs f /J-s FIGURE 6. 202 Figure 7. Size of cell associated and secreted MPCll x W279.2 IgM molecules. MPCll x W279.2 cells were incubated in the presence of 35 s-Cys for 16 hr (5). Cells were lysed and medium was collected in the presence of 1 mM PMSF. IgM was precipitated by addition of rabbit anti µ followed by~- aureus (6). Proteins were eluted without reduction in a boiling water bath and electrophoresed on 1% agarose, 2.5% acrylamide composite SDS gels (5, 10). Gels were sliced into 1 mm sections, the proteins eluted by incubating each slice in 0.5 ml 0.01% SDS for 24 hr and counted for radioactivity. Positions of pyronin Y dye and of stained marker proteins electrophoresed on a parallel gel are indicated: 104E, 19S MOPC 104E IgM (MW 1 x 10 6); IgG, pig y globulins (MW 150,000); BSA, bovine serum albumin (MW 65,000). A. 100 µ 1 cell lysate from MPCll x W279.2 cells shows the presence of two predominant peaks with MWs of 200,000 and 100,000 corresponding to intracellular 7S monomeric IgM (H L ) and HL monomers, 2 2 respectively. Note that there is no significant polymerized pool of 19S molecules inside the cells. Polymerization must occur just prior to secretion. B. 250 µ 1 medium from MPCll x W279.2 shows the predominance of secreted 19S IgM (40%), 7S IgM (10%), and HL monomers (25%). Additional minor peaks are seen which probably represent intermediate stages in polymerization. Note that less than half of the molecules are covalent pentamers. The remaining species may have dissociated under the stringent~- aureus elution conditions (2% SDS) or may have been secreted incompletely polymerized due to a limiting number of J chain molecules synthesized by the hybrid cells. 203 A Hybridoma Lysate anti µ. 104E IgG BSA i t l 5 4 3 2 ~ b X Dye ,---1-, B 104E Hybridoma Medium anti µ. !gG BSA t t 2 0 I (f) l() 8~ + 6 4 2 Dye ~ 0 10 20 30 40 Slice FIGURE 7. Cl. u 80 90 100 204 Figure 8. Peptide map comparisons of MPCll x W279.2 µ 9 chains to WEHI 279 µ .- 11 -1.z...J:..m and µ s chains. 3 µ chains were labeled by incubating cells in the presence of either H-mannose, 35 s-Met, 3H-Phe, 3H-Tyr or 35 s-Cys for 16 hr (5). Cells were lysed, medium collected and µ chains were purified on 10% polyacrylamide gels, reduced, alkylated and digested exhaustively with typsin followed by chymotypsin (7). Peptides were separated by high performance liquid chromatography on a DuPont ODS C-18 column using a y = x 3 gradient for elution (11). (..... ) indicates the percent organic solvent (acetone) used for elution. 0.5 min (0.5 ml) fractions were collected directly into scintillation vials, dried, redissolved in 0.25 ml 0.01 % SDS, and counted for radioactivity. Peptide differences are crosshatched. Set A. 3 H-mannose labeled WEHI 279 µ i (blue), µ m (red) and µ s (black) chains. Five glycopeptides are expected for µ. and µ chains; mature carbohydrate struc1 S tures in, µ s ,hains are seen to change in migration relative to µ 1. chains. µ m chains have one less glycopeptide than either µ. or µ chains. Quantitation of mannose 1 S residues per complex or high mannose oligosaccharide cannot account for these relative peak areas. 3H may therefore have been incorporated into other sugars, e.g., N-acetylglucosamine, or the peptides overlap and are generating a more complicated pattern than is revealed by this one-dimensional separation. Set B. 35 s-Met labeled MPCll x W279.2 µ (black), WEHI 279 µi (blue), µm (red), 5 and µ s (green) chains. Seven methionine peptides are expected for WEHI 279 µ. and µ chains and one is glycosylated. µ chains have one less methionine 1 s m peptide than either µ. or µ chains. If relative peak heights are significant, WEHI 279 1 µ s S chains may also consist of µ -like chains or some µ m m chains which have been proteohytically cleaved from ·the cell surface and lack the C-terminal region. 205 Figure 8 - Continued Set. C. 3 H-Phe labeled MPCll x W279.2 µs (--), WEHI 279 µi (---), µm (- - -) and µs (- - -) chains. µm chains have two additional phenylalanine peptides when compared with either µ i or µ s chains, consistent with the postulated µ m chain COCH-terminal sequence. The two µ s species are seen to be virtually identical. Twenty phenylalanine peptides are expected for µ i and µ s chains and one is glycosylated. The large number of phenylalanine residues precludes any quantiation. Set D. 3H-Tyr labeled MPCll x W279.2 µ s (black), WEHI 279 µ i (blue), µ m (red), and µ s,(grEijm) chains. Sixteen tyrosine peptides are expected for µ i and µ s chains and one is glycosylated. The patterns are very similar for all the µ chain species, although µ m is somewhat different from µ s' consistent with a gain and loss of Tyr containing peptides. Set E. 35 s-cys labeled MPCll x W279.2 µ (black), WEHI 279 µ. (blue), µ s 1 m (red), and µ s (green) chains. Fourteen cysteine peptides are expected for µ i and µ s chains and one is glycosylated. The patterns are similar for all the µ chain species but the propensity of the cysteine peptides to elute very early in the gradient prevents any quantitative comparisons. 206 f.'""·-.-~•c.•·. i l h '-• ~- [_ -. - _.,.,.,.. .,....., ~- ~ .. __ --! 1~J1• ~: ~ . ... j Lff""" • I ~-- LZ - ., ... ., ~-1. [ -i a • • • • .. .-.cTICllll&..:Jt ,a •• - i I I ...L~ •• I I I...., 7\ •• ~j •• ·•- .. =11t:~\.~[i~1.:~_~i:: J ')\'"'"""""-'"'· : ' . --~._._._..._._._._._._._._._._._._._._...·_•__, ~~f}:f;-:-:::: ~-~~----~ ---~;::r- r~~-:-·- FIG URE 8., SET A ~-11 -=-=-- - - - J, 207 ... _.,...,.. WCll X 1127&.2 ..,_. . . .., ... ,.. .. ~ L-. . ... c:..::i ·-L ...J..-1. J ---l. . .J • • • La _,.,.._ J. ., -a - - - - - , ... j ~ J, 1.J ...l . ..L_ J __ J . -1-l.. .l . L • 1• 121 1• ,. 1• "IIIICT ICNJC.la!JI' .,. M:11 r:a,.a,.z..,... I 1. • ~ I ' r· ~ , 1.D-; IL i ~i. 1- 1. 91 - x ~- .. ... I i iji,r,;;,;!;.;:_ , ,,.~. ;,:--.,; ... .· ...• -~~-::'-," •} "-:.'..-;' · :,.f,"•? )-: ,. •., ~ ~ -· - ... _._.. •• .l;IUln..r.; f~,-- -·~:·:~~ ~ _·.~.__. . ._.____. . . l~--- ~ ._~ - ---. FIGURE 8., SET B ..___._,-cn ......._....... _._..::.._._.~ . ............,.__ .._.__~. i : i 1•a J~! 208 .f ' J a. ■ - io- 1,.1 t : ~ 1.- -- : J~ ~L9 ~-b·•· -', -.1)..;JJ,..;<C • FIGURE 8., SET B J. 209 P!iE ,. 17 M • W ~-S WEH! 279 ~-• 2. 5 3. 35 2. B2 2. 53 1. 53 ~ .. X X "' I ~ JI 1. 1. 71 a, :ea 7 Bil .... ~ ri 1511 Q > ~ 1i 'a . :se . 89 I i l 2ll .. I ~ J0 . S7 . S7 21!1 "' 60 1,0 1111111 121!1 8ll FRACT ! ON NUMBER !Bil lell P!iE 4. 17 M • W Mu-S ,. 17 WEHI 279 Mu-.. 3. 3:5 "' I 3.3:5 . !,,,, 2. 53 ' ·: ~ ~ X I~: " 1. 71 "'I ~ n X ; 1013 2. 53 1. 71 :::,,, :' ~ J 8ll .... z 150"' > ~ "8 !: . IHI z < "' .II !§ 2111 .. II 2111 FIG URE 8., SET C Ill 11111 1211 FRACTION tf.MIER I.CS 1111 1111 210 Pl-£ 4. 17 . "'' M x W Mu-5 4. 17 WEHI 27g Mv-S 3. 35 3. 35 2. 53 2. 53 ~ X )( ~ J'' 1. 71 I. 72 . eg .g r. . 09 . 07 2l!l FIG URE 8., SET C l1:: lllllll, "'' "' ell lilll 120 FRACTION r«.JM!IER 8111 140 1158 !Sil ~ J BIil : ...z . ~ t" 211 .... ... ~ j mt • 4.D- IC -J [,. Ll.t- I f~:, ....j ~~~~~~~~~~~-·~·~, u.• ....... J - • • • --- • aa 1a I j . ,1 __ ...t._ .l,. . .J.- ~ 1• t11 1• IWICTICII . . . . . .-Cl I .,._, ..... T'III , ..... -----\ Z.7'\- IIC ~ LN....; . . . -. - -- - FIG URE 8., SET D .. : JI 212 J ,,. a.z,- f !La I- ri ... La- t,.. : L... "'.~ ,. jr ~~.., 1 ..l. . l - . ~ .J._ _J, .. J.. _ J. _ J__...J.....__L __ • i.a -=rta- FIG URE 8., SET D 1a 1• • ~ ia l. 213 - ... ... .. .. Off ..CII X W'Z?lt.1...,... l.at- IDtl Z7'1...,.._ 1.M- -LS, ~ M -z.• ,.. J • ~ -1.1 - - ... - I •;••; .. •;·•;:~.:...1-~--L.....___.____.______.___~r • FIG URE 8., SET E •• • ~Hlll=--sa 149 J - : ·~ l--• ~ t•• - 1 , ... - j 214 ,. ::· ··-. g·. . .. .,,_ _ -~ ·. . t t•· • ·· · \. ... ; ... t>i:. .. , .:.. .... -. l < , ·. . [-~--~-- - FIG URE 8., SET E --- ll'Ctl X..,._Z ..... ; f-- · :.· ----------------------. • • • - •: -1 ·I. • 215 Figure 9. CNBr fragments of MPCll x W279.2 µ chains. ---------------~s--- µ s chains were radio labeled by incubating cells in the presence of either 3 3 H-Tyr and Leu or H-Phe, Pro and Val for 16 hr (5). Medium was collected and µ chains were purified on 10% polyacrylamide gels, reduced, alkylated, combined s with 30 mg carrier MOPC 104E µs chains in 70% formic acid and cleaved with CNBr for 20 hr in the dark at 4°C with constant stirring (7). After removal of reagent, peptides were dissolved sequentially in 8M guanidine-HCl, 3M guanidine 0.2M ammonium bicarbonate, and 0.4M ammonium bicarbonate (total volume~ 7 ml) and chromatographed on a column of ACA54 (LKB) (3.5 x 140 cm) eluted in 3M guanidine, 0.2M ammonium bicarbonate, 0.02% NaN . Fraction volume = 5 ml. 3 Aliquots were counted for radioactivity after reading the on 280 of the MOPC 104E µ s CNBr peptides. Profiles may be aligned exactly with previous and subsequent column runs by normalizing the Rr5 of the carrier l\1OPC 104E µs CNBr fragments. Milligram quantities of the MPCll x W279.2 µs chains were obtained by 7 intraperitoneal injection of 2 x 10 MPCll x W279.2 cells into BALB/c mice. Ascites fluid was produced in BALB/c or (BALB/c x DBA/2) F mice. µ chains 1 were isolated as described and illustrated in Figures 1 and 2. CNBr fragments are numbered as in reference 7. A. 3 H-Tyr, Leu labeled µ chain CNBr fragments derived from the µ s s chains shown in Fi"ure 6C. CN6 and CNS have mobilities similar to the corresponding µ. fragments, indicating their incomplete glycosyla1 tion. 0.25 ml aliquots were counted through fraction 250, 0.5 ml was aliquoted thereafter. Fractions pooled for microsequence determinations (7) (see Figure 11) were: CN3, 199-209; CNS, 177-186; CNS-9, 171-175. B. 3 H-Phe, Pro and Val labeled µs chain CNBr fragments. Note the different incorporation of these radiolabeled amino acids (as compared 216 Figure 9 - Continued with Tyr and Leu) into CN3, CNl-2' and especially into the CN3b-1-2' partial migrating between CNS and CN8-9. CN9 is not labeled with phenylalanine, proline or valine. 0.1 ml aliquots were counted through fraction 190. 0.2 ml was aliquoted thereafter. Fractions pooled for microsequence determinations (7) (see Figure 11) were: CN3, 160-168; CNS, 139-147; CN8-9, 131-136. C. Cold CNBr fragments of MPCll x W279.2 µs chains. Fractions 191202 were pooled for sequence determination of CN3a and CN3b. CN3 was also isolated from an identical CNBr cleavage of µ s chains with succinylated NH groups (12). This allowed determination of 2 only the CN3b sequence (see Figure 10). FIGURE 9A. ::r: r<) I o,oo 2 03 0.. ~ X 04 I r<) 5 ~ 7 9 II 120 CN6 140 CN7 160 180 200 220 FRACTION NUMBER t CN5' CN5 240 x2 260 280 CN9 300 -:i ~ 1:-.) FIGURE 9B. 9 7 I 0 2 3 70 j :~ 0 0.. 6 ~ X 'o s - t0 10 11 121 CN7 90 110 150 170 CN3 190 FRACTION NUMBER 130 I\I \ I \?\ 11 I CN6 210 230 250 I 00 N> ..... FIGURE 9C. 0 .05 00 200 0,10 ' 0.15 I 110 I \I A \ CN6 130 I 150 \ I 190 I \ I 210 L. ♦ CN9 230 250 270 ~ \.,,,,,.....,, ... , FRACTION NUMBER 170 '-"\...._r<~ .. ~ I CN5 A I\ CN7 CN3 290 I (0 I-' NI 220 Figure 10. Sequenator runs on MOPC 104E µ chain CNBr fragments and fragments 8 generated from CNBr fragments by cleavage at tryptophan, arginine or lysine residues. These sequenator runs establish the entire sequence of the MOPC 104E µ s chain but for the NH -terminal 20 residues (13) and the exceptions noted in refer2 ence 14. Relative peak heights on an arbitrary scale (proportional to nanomole yield) are plotted for each automatic sequenator cycle for each of the phenylthiohydantoin (Pth)-amino acids separated by high performance liquid chromatography (HPLC) (15, 16). Scale changes and residue(s) assigned at each step are indicated on the graphs. Peptides are named as in references 14 and 17. Sequenator runs labeled A, C, D, E, I, 0, Q, R, S, T, U, V, and W were performed by M. Kehry, those designated F, G, H, J, K, L,M, N, and P were performed by J. Fuhrman and B by J. Schilling. Also included are the 3 sequenator runs establishing 8096 of the WEHI 279 V H amino acid sequence (U, V, and W). I 5 5 7 0.8 2.4 4.0 5.6 0 .8 I I ILE GLN ASN 10 +14 ·I !GLN v l~ f r 9 +"" :: 3.5 t ( LEU GLU ~2 ARG 4.0 5.6 r 14 5 I 15 20 1l-,1 10 TRP PHE 4.0 5.6 0.8 5.6 I :d't::dW1 TYR PRO J " r .. ,if/\ I~ • - J ~~ f V~L x2 SER "? !\ L YS ( GLY n • ~ ·J ,2A~ ASP 1:uuu~ t 4 5 7.2 f 7 ·I 6 f • " • I\ J10 ·I +14 L.o 56 ~ • HI ~ A 18 ·I t I ~b~t_!._ I-~ ~r-~ x 21::~ 15 20 25 30 35 THR MET --4~~=1--~4-~~~::J:q..~Lq....Y+=+~-,:: • 0.8 I 56 - 2.4 • f4 .0 x2 HIS 4.0 10 1 •:r:r I t' ' 1 : . I 24 5.G ~I - I - I 0.5 6 14 tr 13 CYS f3.5 ·I 10 5 ALA 2.5 3. 14 FIGURE lOA THAT INDICATED ON THE GRAPH. - \ \ - INDICATES CHANGE IN SCALE TO RESIDUE 11 CORRESPONDS TO RESIDUES ?. 35. AND 1 IN THE Mu SEQUENCE :::; ~ / . N ~ 30 Ser Thr Glu Gly Pro Asn Leu Val Thr Arg Cys Lys Asp ~la 21 1 10 20 Asp Asp Phe Ala Gln Phe Leu Asp Thr Cys Cys Lys Ala Ala Asp Lys Asp Thr Cys Phe Amino Acid Sequence of MOPC 104E CN3' N Ni Ni 0.5 1,5 2.5 3.5 1.5 2.5 3.5 9 5 ~ 10 • I~ ~ . Ttfl i 11 15 1 t 2 6 10 I 5 J l 1 10 I\ I 11 ~ 15 I 5 t: I\ 9 + l + RE SIDUE NUMBER TRP Pt£ ., 14 MET t ·1 ., HIS l f ILE GLN ASN f CYS t 10 =:r:: 2.5 3.5 1.5 f "' TYR PRO I~ A 9 ,=l ~ 5 7 15 I 5 I 11..---.J. T Jv • ASP j 10 I VAL SER 15 l ==-t~ ~ ~ >-1~ I LY~ GLY ~~J 5 T97 -it~ LEU GLU ARG FIGURE 108 ASN. ARC. GLN. GLU. HIS. ILE. LElJ. MET. PRO. TRP. VAL CH.Y BACKGRCUll PEAKs. Tt£ FlLLOV It«. AMU«l A.cl DS EXHI Bl TED RESIWE fl ~SPCNlS TO RESIDUE f2! IN THE Mu SEQUENCE. 1B4E CN2 NI NI w ~ I I MET 2.5 3.5 0.5 1.5 TRP I Pl£ ILE I I I -,r;:-::, i "~ 5 1. 0 .5 3.5 J r • ~ 2.5 3.5 i +3.5 ·I 2.5 3.5 09 • / t ll ·I+2.1 1.5 1.5/ x~ I 5 10 15 20 25 30 35 40 I 5 TYR PRO LEU GLU ARG v~ x, VAL SER r/1\ 10 15 20 25 30 35 40 I 5 10 15 20 25 30 35 40 I 5 RESIDUE NUMBER J J 10 15 20 25 30 35 40 L,r 'I - - 1.5 2.1 ~ HI\ ~ 7 . . • •I LYS GLY _ -1 9 I t ~ x5 n ASP ·I~ 1.5 2.5 3.5 0.5 1.5 2.5 +3.5 ·I x~\J;,' ::... d[ .1t.f',,~·····w J:: 2.1 2.5 3.5 HIS 3.5 l! + ·I • 2.5 GLN ASN 2 .i ·1 3.5 0.5 0.3 CVS 1.5 0.9 3.5 2 .5 Al.A 1.5 2 .1 MET FIGURE lOC THAT ItEICATED 00 1l£ GRAPH. -\\- ItEICATES CHANGE IN SCALE TO ARC. cvs. 1l£ FCU.OWING AMit-«l ACIDS EXHIBITED CH.. y BAO<GRCX.tll PEAKs. RESICU: f l ~ TO RESICU: 135 IN 1l£ Mu SEtU:NCE 1ME oa ~ x 2 tL 7 I 3 5 / UIIM 35 I 0.5 :.·: THRf l \ 3.5 MET x2. AA • ,. ILE x2 • r 5 LEU GLU 7 !. • ARG ·I • • ASP 7 10 14 LYS GLY : l Aftt 1 ·I 5 10 15 7 TYR PRO I 10 15 20 . 25 30 35 I 5 ::: 5.6 ! - - __1,_J __ t_ _l,. _ I RESIDUE NUMBER 5 .~ : x 2 . • 25 30 35 I TRP PHE 10 r W ~ .., .,7 ' l -I ~ 15 20 25 30 35 ll VAL SER .t ·..... r·~~-'t' ...... HIS o. 5 (17 (19 o4 3 12 _K 5 GLN 7 1 CYS 2o 2 8 ASN ., ALA I 9 o 104E CNG FIGURE 10D -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUNO PEAKS. HIS. ILE. L YS. MET. PRO RESIDUE #1 CORRESPONDS TO RESIDUES 83. 507. 21. 235 IN THE Mu CHAIN o >< v 104E CN4 104E CNB 104E CN2 C.11 NI NI 2 .0 2.8 20 TlR MET HIS 2.5 3.5 0.8 2.4 4.0 I j ·1 •15.6 CYS 3 5 7 ·1 ALA 5 ,. 10 X2 f 4.0 5.6 .5 1.5 2.5 3.5 5.6 I I I I 10 15 20 5 RESIDUE NUMBER TRP Pl£ ILE Gl.N I 15 TYR PRO LEU GLU I 20 I t I I '<2 .I VAL SER LYS X2 GLY -~ I 1 ·I I __j I 1 7 ·1' j ~ ~I ~ ~ ~ I t~~~ j 5 10 15 20 f ~ 4 .0 5.6 I 0.4 1.2 2.0 2.8 14 --·-r~T ·--~ - , --.T ~r···r ' W-1 0 W-11 V FIGURE lOE -\\- lt-lJICATES CHANGE IN SCALE TO THAT INDICATED ON TI£ GRAPH. ARG. ASP. CYS. HIS. ILE. LYS. MET. TRP. TYR ONLY BAO<GROlHl PEAKS. TI£ FCU.OWING AMINO ACIDS EXHIBITED RESICl.E f l ~ TO RESIDUES 117 AND 112 IN TI-E Mu SEa.JENCE 104E OU 104E CN4 NI NI O') 5 It 15 ~ 10 20 I 5 10 15 I 25 I 0.5 5 1 I ~~---1--- 10 15 20 TYR I 4-~ PRO LEU GLlJ 5 7 25 I 'I, 3 5 7 9 ! 9 ~3 . 5 ,, 10 15 20 VAL SER LYS GLY -1-r--y--y---.• ASP ARC -1 -- - . - . RESIDUE NUMBER 20 ~:.~. 3,5 1.5 TRP I' I 3 1: 3 25 . Pt£ 2.5 3.5 5 ·I 7 9 . +1.5 II 3 3.5 2.5 ILE ~ , •· 9 5 5 7 GLN A 2.5 3.5 - ~ . • .E.... 2.5 1lfi 9 ·1 I ASH 5 7 9 tET HIS I 3 5 J: 9 7 t3 3 I I CVS t5 +1 5 II 7 • Al.A ---.-I 25 -1 ~ •I FIGURE lOF THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. HIS, LYS, MET RESIDUE #1 CORRESPONDS TO RESIDUE #152 IN THE Mu SECUENCE 104E CNS N) N) -;J 2,5 3.5 I I o., I I I I ~ • ft f THR MET 2.0 2 .8 i 0 • _r6 7.2 • ?i~"" 4.0 • 2.4 1.5 56 • 2.5 35 • o.5 HIS 5 6 • 2.4 CYS 0.3 0.9 1.5 f2.1 ·I 1.5 X2 ALA 4,0 II I 2.5 35 - 3 5 7 9 V I ~ • X2 TRP PHE 1I vI I '.) ILE GLN -V'--.- ASN Jr 6 ~ 4.0 5.6 7.2 2. , 4.< i5.6 ..... ·---. • 14 0.8 2.4 4.0 5 6 • 7.2 0.1 0.3 0.5 0.7 TYR PRO _ LEU GLU ARG ~ • 2. 4. 5 7.2 I 14 4.0 5.6 t0.7 . ... . • • H ·t +1 _, ~ ~'> VAL SER LYS GLY ASP 35 • • i R-I o R-II • FIGURE 100 -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. MET THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. RESIDUE -1 CORRESPONDS TO RESIDUES 157 ANO 186 IN THE Mu SECUENCE 104E CNS 104E CNS 00 ~ 10 15 20 25 30 35 5 I XX X4~ Ttfl ~ r I i 3 5 -1 7 9 I 3 5 -1 7 HIS 7 -1 9 -Ir CYS ./11..A 1.2 2 .0 2.8 I 3 5 7 9 5 10 TRP 5 7 9 7 9 I I 5 3.5 f3 -I ·I 5 . PRO LEU GLU ARC 14 1. 4 ·I ·I 10 15 TYR L 9 --t- 0.2 0.6 1.0 ..~ ~: RE SIDUE NUMBER 15 20 25 30 35 " ! Pt£ ILE GLN ASH x4 25 30 35 VAL SER . LYS ·I CLY ASP FIGURE lOH -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE'FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS, ASP. CYS. HIS. MET. TRP RESIDUE #1 CORRESPONDS TO RESIDUE #168 IN THE Mu SEQUENCE 104E CNS W NI NI CD I 5 10 o.5 x2 .5 1. 5 2.5 3.5 0.5 1.5 2.5 3.5 5 7 15 20 25 30 TI-fl MET HIS ><2 CYS ALA 3 5 7 9 0 .5 1.5 2.5 3.5 L 5 10 X2.5. I 3.5 7 9 5 7 5 10 RE SIDUE NUMBER .,.u._J 15 20 25 30 TRP Pt£ ILE GLN ASN TYR PRO LEU GLU ARG I 5 7 9 5 I\ 10 d - , J · {\xrJ SER LYS GLY X2 VAL +--t-t-+-~ : : x25 3.5 n ASP ~~~7~ 4 .5 7 7 9 : 7 FIGURE 101 -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS, GLN. HIS. MET. TRP. TYR RESIDUE #1 CORRESPONDS TO RESIDUE #235 IN THE Mu SEQUENCE 104E CN6 N) w 0 f r ~ 15 rt I 5 10 20 25 30 I 4 2.0 2.8 •· I I I I I PHE ILE TRP I-+ I I GLN ~ X4 ASN I X 4 GL~ : l 4 0 6 • l Tl _x 4 i : GLYN l ,,.... ASP A I 10 I 15 ):' I RESIDUE NUMBER I 5 r 2.s 3.5 I 20 I I I 25 30 TYR I PRO 5 10 15 ~ ~ VAL i I 20 25 30 f~)J\J:1 2.0 t 2.8 i:: SER t{'i 'r=-'.'J' 1"iJ I 1{~1~1 I" I ~I LEU - l LYS • • i k I 40 • 2.4 l~ 56 • •I -1 4.0 ~ t . ARG t . ,~~I 10 15 20 25 30 I 5 t-1 I 0. I. 2.~ •1 2 .8 MET ti I •-. I : HIS I l CYS ~6 t . ALA M~\!U~ tR :) 2.8 40 • 2.4ft 5.6 • ~ 2 R- I o R-II v FIGURE lOJ -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BAO<GROUNO PEAl<S. ARG. GLN. HIS. MET. TYR RESIDUE TO RESIDUES 255 AND 11 326CORRESPONDS IN THE Mu SECUENCE 104E CN6 104E CN6 I-' c:.., t-.') I 5 5.6 0.5 1.5 2.5 3.5 4.0 5.6 Ttfl I I MET 5 4.0 . 5.6 2 I TRP I A PHE ILE GLN 10 1r GLY 14 GLU j 15 20 25 30 35 40 RESIDUE NUMBER I 5 I I 4.0 4 .0 ! 5.6 14 10 PRO 5.6 :I 7 9 5.6 5 10 15 20 25 30 35 40 'l SER :~ ~~f\~}~~x~ l:, 10 15 20 25 30 35 40 '1~ 6 10 14 4. 5. l:~ 10 15 20 25 30 35 40 I x2 HIS CYS ASP 7.2 ARG FIGURE lOK -\\- INDICATES CHANGE IN SCALE TO THAT I ND ICA TED ON THE GRAPH. MET ONL.Y BAO<GRCUIJ PEAKS. THE FCl..LOWING AMINO ACIDS EXHIBITED RESIDl.E 11 COORESPCHlS TO RESIDUES 1 AND 286 IN THE Mu SEWENCE 1B4E CN6 ASP-PRO ACID O.EAVED o 1B4E Mu AMINO TERMINUS v N) c,., N) t . _ 35 5 , ' 1.5 2.1 0.1 Tlfi tET . j 3.5 Q5 1.5 2.5 10 I I 15 20 . ._ . j 2. 1 I 5 I 10 RESIDUE NUMBER I I II I 25 30 35 I TRP Pt£ ILE ~5 1.5 5 5 o., 0.6 1.0 1.4 0. 5 I I I I I I 115 20 25 30 35 I 5 ~ ~--n PRO LEU GLU 0.3 0.5 GLN 0.9 1.5 2.1 1.5 ARG 25 ASH '- 5 HIS t . 3.5 0.3 CVS ~ ALA 2.5 ff o.5 03 3.5 X xs r x2 VAL SER LYS GL Y -.. ~ -j •I -I . J -I FIGURE lOL -\\- UlllCATES DW-a IN SCALE TO THAT INDICATED 00 TI£ GRAPH. ARG. GLN. GLU. HIS. MET. TRP. TYR Cit.Y BAD<GROLN) PEAKS. TIE FCl..LOWlt«. AMINO ACIDS EXHIBITED RESICl.E 11 COORESP(N)S TO RESICl.E 1326 IN TIE Mu SEClJENCE 104E 0-1:i Ru ts:> w w ,,,..- I 3.6 1. 5 , ~ 0.5 2 .5 , 3.5 9 J( • X2 • MET I HIS X3 CYS ALA I ,. ILE GLN 7 7 I 3 5 LEU GLU 2 1 : 1 ·2 .0 LYS GLY ASP j 9 tr¾-~,~ ,·~tt¥ ,V\J(iY~~~~~ I 3 i: 0.5 3 7 1,5 ARG 5 ASN 2 .5 •13.5 R-IIo R-Ilb R-Ila 0 () V >< FIGURE lOM -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. ARG. HIS. MET RESIDUE 11 CORRESPONDS TO RESIDUES 362. 332. 326 IN THE Mu SECUENCE UNIDENTIFIED 104E CN6 104E CN6 111J4E CN6 ""' NI w t5 17 3.0 4 .2 06 THR MET ·I • TRP LEU • 5 7 4 •2 PHE 1 >< V\_ J -1 -l ~-I ASP TYR PRO 5 7 •7 VAL SER I -I '-JI I - -I - r W 'J~ Lll H ...,5 +s N~~ f7 ~v ""- - >< 2 84 fG .0 f ~-I ARG f7 lr4,1 I'AIi • 2.5 3. 5 f ~ ~~ ASN :'1- 1i· 10 • x2 ~- +;, 5 3.0 I • I\ +3•5 ~~ ALA 4 .2 5 10 _5 R-II~ R-IIb R-IIa o O x v o FIGURE lON -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. ARG, HIS. MET RESIDUE jfl CORRESPONDS TO RESIDUES 362. 332. 326 IN THE Mu SECUENCE 104E CN6 R-1 Id UNIDENTIFIED 104E CN6 104E CN6 104E CN6 N) '" ~ 10 15 I 20 Tifio MET HIS I CYS /~\LA ~~ 4 5.1 4,5 10.5 • 7.5 13 5 : 3 5 7 9 [ I 3 5 ·I 7 • . 3 5 o r·· ·I • . 1 10 I ~ ~I 15 I I 14 _,. GLU ARG 5 RESIDUE NUMBER I o > -0~ 10 15 o 20 I t1I 5 7 '------ti I PRO 3 o 3 _1 • • 3 ~6 GLY ASP 5 ,f / 10 I 15 I ~ 20 SER LYS v 398 ANO 1 IN THE Mu SECUENCE RESIDUE 11 CORRESPONDS TO RESIDUES UJ4E CN7 o 104E AMINO TERMINUS FIGURElOQ m. ILE. MET. TYR THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. l 7 .. i9 5 LEU I 3 5 7 .: ~ A • • . 1 5fl _ 1r • r'-1'-/i'.'-,=' I · t~ . • . j J, 20 PHE ILE I GLN ASN NI o, t.., 2.1 2.5 3.5 - THR MET ALA I t 0.4 1.2 2.0 r 5 •12.8 ix.a ARG 2.0 f 2.8 ASP 10 15 20 25 30 0.6 I 0.2 ::: 1.0 1.4 • 1. 2 5 TYR - PRO 0. 1.0 1.4 2 .0 1 2 .8 5.4 ~ 10 VAL SER ~~i"""' J - 15 20 25 30 35 ~f../L X2 '\_) ~i ,~,'t1:1~ 2.0 f 2.8 RESIDUE NUMBER 10 15 20 25 30 35 TRP PHE ASN V W-1 0 FIGURE lOP -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. MET. TRP RESIDUE fl CORRESPONDS TO RESIDUES 419 AND 398 IN THE Mu SECUENCE 104E CN7 104E CN7 -:i v:, N) A HIS ALA .5 3.5 5 10 2.1 15 'f<. 20 I : t THR ' 1. 2 2.0 . MET 2.8 t . :~. ': 21 2.1 5 10 I½-( LEU ~ ARC ; 1~ 2. 1 ~ 20 I • 1.0 1.4 I I TYR 5 10 15 Al\,, 'tfVT {\ .. 5 f -.A ~ 15 VAL I LYS ASP i • ~ 1· 20 I 10 20 .1~.1v,w ~j _ 2.0 • 2.8 r:v,= I ~ ri.,.}\__ ' /\ ,o 42 t 1.4 RE SIDUE NUMBER 15 TRP I ILE ASN W-I o W-II v FIGURE lOQ -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. ILE. MET. TRP Tl£ FOLLOWING AMINO ACIDS EXHIBITED ONLY BAO<GRWNO PEAl<S. RESIOOE fl CORRESPCNlS TO RESIOOE illS 419 AND 489 IN Tl£ Mu SEQUENCE 11114E 0(7 11114E CN7 ~ 5 7 9 Tifl MET 5 7 1.5 30 50 I 3 5 7 9 L, . 10 20 RESIDUE NUMBER 40 TRP TYR 7 9 7 5 7 3 PRO GLU 5 7 9 4.5 7 9 18 ARG rrrrr-rr-rTT~rr · 5 PHE GLN ASH 1 7.5 10.5 3 1.5 7 9 5 7 5 HIS CYS ALA 9 2.5 3.5 21 , , I I -.I . VAL SER GLY ASP j I .., -I l T-:7 v FIGURE lOR -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. ARG. CYS. GLN. MET RESIDUE fl CORRESPONDS TO RESIDUE 1507 IN THE Mu SEOUENCE 104E CNS-9 o 104E AMINO TERMINAL CD ~ t,.) • 5 •· : 9 ~ I 3 5 7 9 Tifl MET • !• . J 5 7 10 15 18 n f I I 5 10 ~ 15 t 5 .7 .,..,, "' 20 TYR PRO :l 7 !18 . •9 t 5 10 n "----fN % I LEU A 20 VAL SER 1 LYS GLY ASP 1 ~ 25 l ~ . I 25 I 15 ,t~~HW~J 25 f: • r9 7 t9 lAN 'lii1 10 14 RESIDUE NUMBER 20 I TRP P't£ ILE 14 I 6 9 18 ·I 141\I ., GLU ARC 0.3 ·1 0.1 GLN ASN 10 HIS ·1 0 .5 0 .7 CYS AUi. FIGURE 10S -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. ALA. ASN. ARG. GLN. GLU. HIS. PHE. TRP 1551 IN THE Mu SEQUENCE RESIDUE 11 CORRESPONDS TO RESIDUE 104E CNB-9 R 0 1,-:1 ~ 7 j ILE 5 2.5 J,: ·; t3.5 r I 3 . LEU ~ T I +9 fi5 .: LYS GLY ~ -- • I I _J ..J 7 ...; _J I ·1 • t~ 2.1 5 0.8 + i ·I . i 5 TRP 10 2.8 .il.2 +2 .0 -! l 5 I T5 ·I ·! -+7 A ., TYR .6 1 - i ::. 5 I ., -' I ·I I -< I 7 VAL 10 ·1 7•i ·; l ~~ SER I RESIDUE NUMBER I 10 LJ~____.J~ ~ t ~ 10 _l _ _ _ __ : _ _ _ THR •I 8 ~~- fL2 PROJ + +...______ - .- - l f + 4 ·• ~- 4.0 1 PHE1 ~6 ~ ---,---i --. ~ 2.-7 ~ MET H----+----+~I ...- - - - - - - - .!.~ + ~ - - - I- - - - - - - . l T r . f .I 1 J_ _ f-~t3 I 3 T21 ~-+-i ·! 7 I -, 27 · I -1 -1 ASP -- - - - T---·- - 7 . GLU 11 I. . _____ I -1 ir~--~ 1· -t 1~ + HIS CYS ·i 1 12,5 5 7 THR t3.5 -AL:1r-----r---- - AS:r11 ----- r- ----~RG- : FIGURE lOT THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. ARG, GLN, HIS, I LE. LYS, MET. PRO. TRP RESIDUE -1 CORRESPONDS TO RESIDUE NOS. 569 AND 21 IN THE Mu SECUENCE 104E CN9 o 104E CN2 v ~ ,i,. Is) I I I I I CYS I ALA I ASN I • Tl 1:T 7 T9 t -i :,.._I 9 + Tl TRP Pt£ 7 ~7 t9 ·1 +3 l5 . 1 x2 9 T5 ·j r ·, I r5 I .I +1 ·! X2 1 7 f • J5 TYR I • 9 RESIDUE NUMBER X4 h f3 t5 1 t7 ·I • o. 5 ~t PRO LEU I ASP . ..J ~ -i I 15 20 I i ~ VAL SER LYS -1- ...J ~ 1- - 1-' Gl.Y - -¼- -1-- + ~ ...: 11.5 ~ ~ 2 -1 ► 1 ~~ Gl.U '---~......._ __ , • • f 2.5 .: 3.5 • 11111111~ 5 10 15 20 25 30 35 I 5 10 15 20 25 30 35 I · X4 -+-t-+·f ~-+ • -t - -1 3 X2 •t 1.5 lt .,I i X2 X4 ILE GLN ·I i- 06 ·I ARG r ·......--r....-, ..I ----r-,-i. I t I -- .. I -....-.--1-•__,---... I I I I I I I - -.-I It 15 ~ I - -!--+ -+-+ - +·+ l o.5 x 2 .I f 1.5 I ~t HIS I 1.0 X4 X2 I -l-2 .5 I . 1.4 I .I 3. 5 • I J -·~ - t- -I ·-1 --1- · I- I I 2.5 3.5 2.5 3.5 11 rt I -,t - ~ - . - = 7-.. -..---,-••r •-..- -· ,--T--tl FIGURE lOU -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS, ALA, HIS. MET, PHE, TRP RESIDUE 11 CORRESPONDS TD -RESIDUE fl IN THE WEHI 279 Mu SEQUENCE MPCll X W279.2 Mu ~ NI NI I I 5 2.0 2. 8 I 0.8 2.4 4.0 5.6 10 I 15 20 25 30 35 10 15 TRP THR - ASN PHE I 5 t 3 1: 9 ¥= 0 .8 2.4 4.0 5.6 ·1 MET HIS • ""f' t--,,-, X2 x2 . CYS ALA 5.6 5.6 1.0 J 1.4 TYR PRO ARG ·t 14 5.6 4.0 5.6 • I I VAL SER ASP t t FIGURE lOV -\\- INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS. ALA. HIS. MET. PHE. TRP RESIDUE -1 CORRESPONDS TO RESIDUES 1 AND 49 IN THE WEHI 279 Mu CHAIN MPCll X W279.2 CN3a o MPCl 1 X W279. 2 CN3b v """ N) c.., ' < I bCC 2.5 3. 5 I ·~ I " I ~~ 1. 5 2.5 ·I 3. 5 I I ~< •~ I 11 I ' " I 5 Lz 10 I I l I 9 I I I I I I I 5 7 9 I ~ 9 • ' • ~3 t 5 •7 ~ X2 -+ ~ X4 l r-r-1-----r ~ PRO 5 7 •9 f 7 5 I 10 J l I VAL SER LYS GLY ASP I I I j ~ • j ~ j, 15 20 25 30 35 I I i _ \ . ~ ~--+-~ LEU I GLlJ ARG I I f_ l.. __L ~ - _J _ _ L ___l_ _ .,L . .l LJ.. ...J 3 5 ·I97 • ltl I 15 20 25 30 35 I 5 10 15 20 25 30 35 I RE SIDU E NU MBER TRP pt£ ·I MET Tm ILE r ' J I HIS I • CU.:J Q2 t o .6 11 3 • 1.4 111 +1. 0 ASN 5 ·I 7 ~ I CYS ALA FIGURE lOW INDICATES CHANGE IN SCALE TO THAT INDICATED ON THE GRAPH. - \\- THE FOLLOWING AMINO ACIDS EXHIBITED ONLY BACKGROUND PEAKS, ALA. CYS. GLU. HIS. MET. PHE. TRP. VAL RESIDUE #1 CORRESPONDS TO RESIDUE #49 IN THE WEHI 279 Mu SECIUENCE MPCll X W279.2 CN3b N> .i:. .i:. 245 Figure 11. Seguenator runs on radiolabeled MPCll x W279.2 µsand WEHI 279 J:!.i chain CNBr fragments. These sequenator runs establish the identity of the WEHI 279 C 104E C µ µ and MOPC CNBr fragments (7). Fractions containing CNBr fragments from the ACA54 column were pooled, dialyzed 3X against 5% formic acid at 4°C, lyophilized and loaded on the Caltech sequenator in trifluoroacetic acid (TFA), 15% water. For each step, Pth-amino acids were separated by HPLC, the labeled residues fraction collected and counted for radioactivity (18). Scale changes and residue(s) identified at each step are indicated. Peptides are named as in reference 7. See reference 7 and Chapter 4 for the corresponding sequences. 246 Wehi 279 1-'-i CN3 800 700 600 500 400 300 200 100 ::: Cl. VAL 0 TYR u I ::c: 1200 ~ 1000 800 600 400 200 o· 800 700 600 500 400 300 200 100 0 LEU 4 8 12 16 RESIDUE NUMBER FIGURE 11 20 24 247 MPCII X W279.2 µ. 5 CN3 200 -~ •2 100 0 1200 1000 800 600 400 •2 1400 •2 -r--,__;------ 200 0 VAL 1200 1000 800 600 400 ~ a.. uI 200 0 :r:: 1000 r<") TYR 800 600 400 200 0 1800 x ~ "' LEU 1600 1400 1200 1000 800 600 x4~ 400 200 0 ~ 4 8 12 16 20 24 RESIDUE NUMBER FIGURE 11 28 32 248 Wehi 279 µ.i CN 1-2' 200 PHE 100 0 PRO 100 50 ~ 0 a.. 200 u VAL I J: rt') 100 0 800 TYR 600 400 200 0 100 50 LEl:J 4 FIGURE 11 12 16 20 8 RESIDUE NUMBER 24 249 Wehi 279 1 µ. - CN5 800 600 400 200 PRO 100 ~ Q.. 200 VAL u I ,.,:r::: 100 200 TYR 100 200 LEU 100 8 12 20 24 RESIDUE NUMBER FIGURE 11 28 250 Wehi 279 CN6 f1-i 1 400 300 200 100 400 PRO :E a. 300 (.) I :I: rt> 200 100 400 VAL 300 200 100 2 4 6 RESIDUE NUMBER FIGURE 11. 8 251 Wehi 279 µ.i + µ.m CN7 50 PHE 0 PRO 50 0 600 500 400 300 200 ~ 100 0.. uI 0 I r() 50 xz M ALA 0 50 0 LYS TYR --- x2 """ / 4 8 12 16 20 RESIDUE NUMBER FIGURE 11 24 28 32 252 Wehi 279 µi CN8-9 5 0 0 . - - - - - - - - _ _ : __ _ _ _ _ 400 PHE 1 300 200 100 0 ~:::r:::~j____l::'.~t::=...---1......-...J....___._, PRO -----,,--~---,-x2 x2 ~ ~ uI 100 CL VAL I r<) 400 300 200 100 TYR 0 ~~~-_r:::=:::::r:=-.J.___,1__..____7 100 LEU 0 L=::::::~=::;==:::;:=--,-.:.::..:...-=2~8-7.32:----:3Jc;8~4 FIGURE 11 8. 12 16 , RESIDUE NUMBER 253 MPCII x W279.2 1-'-, CNS PHE 1600 1200 800 400 0 PRO :E 0.. uI ,.,::r: VAL 100 0 300 TYR 200 100 0 LEU 200 100 0 ....... -,,,.,,.,,., _.....__ 4 • • I 6 8 IO RESIDUE NUMBER FIGURE 11 12 254 MPCII X W279.2/J-s CNS-9 800 PHE 700 600 500 400 300 200 100 0 PRO ~ a. u I ,.,J: VAL 100 0 TYR 200 100 0 LEU 100 0 --------------2 4 6 8 10 RESIDUE NUMBER FIGURE 11 12 255 REFERENCES 1. Laemmli, U. K. Nature 277:680 (1970). 2. Mole, J. E., Bhown, A. S., and Bennett, J. C. Biochemistry 16:3507 (1977). 3. Koshland, M., Mather, E., and Baltimore, D. Unpublished results. 4. Brown, S. S., and Revel, J.-P. In Advanced Techniques in Biological Electron Microscopy (J. K. Koehler, ed.). Springer-Verlag: Berlin-Heidelberg (1978). 5. Sibley, C. H., Ewald, S. J ., Kehry, M. R., Douglas, R. H., Raschke, W. C., and Hood, L. E. ~- Immunol. Submitted. 6. Kessler, S. ~- Immunol. 115:1617 (1975). 7. Kehry, M., Ewald, S., Douglas, R., Sibley, C., Raschke, W., Fambrough, D., and Hood, L. Cell. In preparation. 8. O'Farrell, P. H. ~- Biol. Chem. 250:4007 (1977). 9. Bonner, W. M., and Laskey, R. A. Eur.~- Biochem. 46:83 (1974). 10. Dingman, C. W., and Peacock, A. C. Biochemistry 7:659 (1968). 11. McMillian, M., et al. Nature 277:663 (1979). 12. Klapper, M. H., and Klotz, I. M. Methods Enzymol. 25:531 (1972). 13. Barstad, P., Hubert, J., Hunkapiller, M., Goetze, A., Schilling, J., Black, B., Eaton, B., Richards, J., Weigert, M., and Hood, L. Eur.~- Immunol. 8:497 (1978). 14. Kehry, M., Fuhrman,J., Schilling, J., Grimes, W., Rogers, J., Hunkapiller, T., Sibley, C., and Hood, L. E. Biochemistry .. In preparation. 15. Hunkapiller, M. W., and Hood, L. E. Biochemistry 17:2124 (1978). 16. Johnson, N., Hunkapille~, M., and Hood, L. Anal. Biochem. 100:335 (1979). 17. Kehry, M., Sibley, C., Fuhrman, J., Schilling, J., and Hood, L. E. Proc. Natl. Acad. Sci. USA 76:2932 (1979). 18. McMillan, M., Crecka, J. M., Murphy, D. B., McDevitt, H. O., and Hood, L. E., Proc. Natl. Acad. Sci. USA 74:5135 (1977).