Specific interaction between VH and VL regions of human monoclonal immunoglobulins

Preferential reassociation of immunoglobulin H and L chains was investigated using a method of competitive hybridization. A model was built in which two monoclonal light chains, one autologous, L_A, and one heterologous, L_B, were competitively reassociated with a given monoclonal heavy chain, H_A. A total of 12 human myeloma proteins were used with known isotypic, allotypic and variability subgroups: 44 distinct combinations were studied by competitive hybridizations. It was found that a preferential reassociation occurred between the complementary H and L chains that were associated in the native molecule in 80% of the cases. It was clearly established that the subgroups had no influence on the preferential reassociations that seem, therefore, to rely exclusively on individual (“idiotypic”) structural differences. It was shown that, although H and L chains had been fully reduced and denatured, the same degree of preferential reassociation was observed after the chains had been reoxidized and refolded. These experiments suggest, therefore, that the observed preferential reassociations are the consequence of an antigen-independent selection process that must have taken place during the differentiation of the antibody-forming cells.     

Relative noncovalent association constant between immunoglobulin H and L chains is unrelated to their expression or antigen-binding activity

With H and L chains derived from the murine hybridoma and myeloma proteins NQ5-89.4, 93G7, AIDA10/3, AIDA10/16, HyHEL-10, HyHEL-9, HyHEL-8, HyHEL-5, XRPC25, J539, UPC10, 6684, and C101, the relationship between the relative affinity between H-L pairs, their antigen-binding characteristics, and the primary structure of their VH and VL domains was assessed. Using competitive chain reassociation assays in which two different L chains were allowed to compete for a limiting amount of an H chain, it was observed that different pairs of L chains tended to compete to the same degree regardless of which H chain was used as the limiting reagent and regardless of whether they were the autologous or heterologous L chain. In agreement with our previous results, it was also observed that when there was limited diversity between the Vk segments of the competing L chains, the relative competitive ability of an L chain was dictated by the nature of the first residue of the Jk segment, residue 96. However, when a high degree of diversity existed between the Vk segments of the competing L chains, the relative affinity was dictated by the V segment. It was further demonstrated that junctional diversity in the L chain may not necessarily be essential for antibody activity, determined using autologous and heterologous, noncovalently reassociated immunoglobulin molecules in antigen-binding assays. Combined with the results of the competitive reassociation assays, it was evident that no correlation between the competitive ability of these L chains existed or, by inference, the relative mutual affinity between different H-L pairs and their ability to form an antigen-binding site. These results were in agreement with the random rearrangement of VH and VL domain gene segments and argue against any restrictions in the expression of the full repertoire of immunoglobulin molecules due to combinatorial (H-L pairing) mechanisms.     

Comparison of the amino acid sequences of the variable domains of two homogeneous rabbit antibodies to type III pneumococcal polysaccharide.

The amino acid sequences of the V (variable) regions of the H (heavy) and L (light) chains derived from rabbit antibody K-25, specific for type III pneumococci, were determined; this is the second homogeneous rabbit antibody besides antibody BS-5 whose complete sequence of the V domain has been established (Jaton, 1974d). The V regions of L chains BS-5 and K-25 (both of allotype b4) differ from each other by 19 amino acid residues; 11 of these 19 substitutions are located within the three hypervariable sections of the V region. On the basis of seven amino acid differences within the N-terminal 28 positions, it is suggested that L chain K-25 belongs to a different subgroup of rabbit K chains and L chain BS-5. H chain K-25 (allotype a2) differs from another H chain of the same allotype by one amino acid substitution within the N-terminal 70 positions in addition to interchanges occurring in the first two hypervariable sections. H chain K-25 was compared with H chain BS-5 (allotype a1) and with the known V-region rabbit sequences. Allotype-related differences between a1, a2 and a3 chains appear to occur within the N-terminal 16 positions and possibly in scattered positions throughout the V-region. In the hypervariable positions, variability between the two antibodies is remarkably more pronounced within the third hypervariable section of both H and L chains than within the first two.     

Cloning,expression, and crystallization of the V delta domain of a human gamma delta T-cell receptor.

T-lymphocytes recognize a wide variety of antigens through highly diverse cell-surface glycoproteins known as T-cell receptors (TCRs). These disulfide-linked heterodimers are composed of alpha and beta or gamma and delta polypeptide chains consisting of variable (V) and constant (C) domains non-covalently associated with at least four invariant chains to form the TCR-CD3 complex. It is well established that alpha beta TCRs recognize antigen in the form of peptides bound to molecules of the major histocompatibility complex (MHC); furthermore, information on the three-dimensional structure of alpha beta TCRs has recently become available through X-ray crystallography. In contrast, the antigen specificity of gamma delta TCRs is much less well understood and their three-dimensional structure is unknown. We have cloned the delta chain of a human TCR specific for the MHC class I HLA-A2 molecule and expressed the V domain as a secreted protein in the periplasmic space of Escherichia coli. Following affinity purification using a nickel chelate adsorbent, the recombinant V delta domain was crystallized in a form suitable for X-ray diffraction analysis. The crystals are orthorhombic, space group P2(1)2(1)2 with unit cell dimensions a = 69.9, b = 49.0, c = 61.6 A. and diffract to beyond 2.3 A resolution. The ability of a V delta domain produced in bacteria to form well-ordered crystals strongly suggests that the periplasmic space can provide a suitable environment for the correct in vivo folding of gamma delta TCRs.     

Demonstration of a homogeneous noncompetitive immunoassay based on bioluminescence resonance energy transfer

We describe a noncompetitive homogeneous bioluminescent immunoassay based on the antigen-dependent reassociation of antibody variable domains (open sandwich bioluminescent immunoassay, OS-BLIA). The reassociation of two chimeric proteins, an antibody heavy-chain fragment (V(H))-Renilla luciferase (Rluc) and an antibody light-chain fragment (V(L))-enhanced yellow fluorescent protein (EYFP), was monitored by a bioluminescence resonance energy transfer (BRET) between the two. Upon simple mixing of the reagents with the sample, an antigen-dependent increase in BRET was observed with a measurable concentration range of 0.1 to approximately 10 microg/ml antigen hen egg lysozyme. Compared with our comparable assays based on fluorescence resonance energy transfer (FRET), a 10-fold improvement in the sensitivity was attained, probably due to a reduction in reagent concentration.     

The crystallizable human myeloma protein Dob has a hinge-region deletion.

During experiments to prepare heavy-metal derivatives of the crystallizable human IgG1 (k) immunoglobulin Dob, it became apparent that this protein has several unusual features. (1) Instead of the four labile interchain disulfide bridges ordinarily found in IgG1, the Dob protein has only a single interchain disulfide bridge, which connects its two light chains. (2) The Dob heavy chain appears to be slightly smaller than a control gamma1 chain, as judged by polyacrylamide gel electrophoresis in sodium dodecyl sulfate and by gel filtration in guanidine. (3) The Dob heavy chain has three fewer residues of half-cystine than expected in gamma1 chains. (4) The Dob IgG is relatively resistant to digestion with papain and trypsin; however, it is readily digested with pepsin, although at an unusual site. These findings suggest that some or all of the gamma1 hinge region is missing in Dob. To localize the deletion, we prepared an F(ab')2 fragment consisting of two heavy-chain pieces (Fd') noncovalently associated with the light-chain dimer. The Fd' piece was isolated and digested with trypsin. The sequence of the C-terminal tryptic peptide was Val-Ala-Pro-Glu-Leu-Leu-Gly-Gly-Pro-Ser-Val. Positions 2-11 of this peptide are identical with residue positions 231-240 of the gamma1 chain. The N-terminal valine could be either Val-211 or Val-215 of the gamma1 sequence. A tryptic peptide, Val-Asp-Lys-Lys, was also isolated from Dob Fd'; this sequence is not found in the variable region of the Dob heavy chain [Steiner, L. A., Garcia Pardo, A., & Margolies, M. N. (1979) Biochemistry (following paper in this issue)] but corresponds to positions 211-214 of the gamma1 constant region. Therefore, the deletion cannot include these residues and must begin after Val-215; normal gamma1 sequence resumes at Ala-231. The same 15-residue deletion has been found in two other IgG1 proteins, Mcg [Fett, J. W., Deutsch, H. F., & Smithies, O. (1973) Immunochemistry 10, 115] and Lec [Rivat, C., Schiff, C., Rivat, L., Ropartz, C., & Fougereau, M. (1976) Eur. J. Immunol. 6, 545]. Possible explanations for the occurrence of identical hinge-region deletions in three different immunoglobulins are suggested by recent experiments demonstrating that the three constant domains and the hinge region of mouse gamma1 chains are each encoded by separate segments of DNA [Sakano, H., Rogers, J. H., Hüppi, K., Brack, C., Traunecker, A., Maki, R., Wall, R., & Tonegawa, S. (1979) Nature (London) 277, 627].     

Membrane immunoglobulins are stabilized by interchain disulfide bonds occurring within the extracellular membrane-proximal domain

Membrane-bound immunoglobulins have, in addition to the transmembrane and cytoplasmic portions, an extracellular membrane-proximal domain (EMPD), absent in the secretory forms. EMPDs of immunoglobulin isotypes alpha, gamma, and epsilon contain cysteines whose role has so far not been elucidated. Using a genetic strategy, we investigated the ability of these cysteines to form disulfide bridges. Shortened versions of human membrane immunoglobulins, depleted of cysteines known to form intermolecular disulfide bonds, were constructed and expressed on the surface of a B-cell line. The resulting membrane proteins contain a single chain fragment of variable regions (scFv) linked to the dimerizing domain from the immunoglobulin heavy chains (CH3 for alpha and gamma or CH4 for epsilon isotypes), followed by the corresponding EMPD and the transmembrane and cytoplasmic domains. The two functional membrane versions of the epsilon chain, containing the short and long EMPD, were analyzed. Our results show that the single cysteine within alpha1L and gamma1 EMPD and the short version of epsilon EMPD form an interchain disulfide bond. Conversely, the cysteine resident in the epsilon transmembrane domain remains unreacted. epsilon-long EMPD contains four cysteines; two are involved in interchain bonds while the remaining two are likely forming an intrachain bridge. Expression of a full-length membrane epsilon heavy chain mutant, in which Cys(121) and Cys(209) within domain CH2 (involved in interchain bridges) were mutated to alanines, confirmed that, within the complete IgE, EMPD cysteines form interchain disulfide bonds. In conclusion, we unveil evidence for additional covalent stabilization of membrane-bound immunoglobulins.     

Using a heavy chain-loss hybridoma 26.4.1LL for studying the structural basis of immunoglobulin chain association

One of the mechanisms contributing to antibody diversity is created by the association of different heavy and light chains. The combinability of heavy and light chains has been studied previously in two systems: in vitro chain recombination and hybrid hybridoma. Here, a novel in vivo chain combination assay system involving a heavy chain-loss variant, 26.4.1LL, producing two kappa light chains (L(DEX) and L(MPC)) different in size is described. In conjunction with DNA transfection, immunoprecipitation and SDS-PAGE, the structural basis of noncovalent interaction between heavy and light chains can be elucidated systematically by examining the relative association tendency of a heavy chain with two light chains. To demonstrate the usefulness of this system, three stably transfected 26.4.1LL cell lines expressing gamma2b heavy chains, designated as H(DEX), H(CHI) and H(ARS), respectively, with structural interrelated variable regions were generated: H(DEX) differs from H(CHI) only in framework regions whereas H(CHI) differs from H(ARS) in complementarity-determining regions. The relative amounts (R values) of L(DEX) and L(MPC) associated with the heavy chains H(DEX), H(CHI) and H(ARS) in the assembled immunoglobulin molecules were found to be 1.02, 0.64 and 0.05, respectively, suggesting that the complementarity-determining regions and framework regions contribute equally to the V(L)-V(H) interaction. This conclusion is consistent with previous observations based on calculation of the buried area in the V(L)-V(H) interface, thus demonstrating the usefulness of this system.     

Shared N-terminal sequences in monclonal IgMkappa and IgGkappa proteins from a patient with a complex multiple paraprotein disorder.

The N-terminal sequence analyses were performed on the heavy (H) and light (L) chains of the idiotypically identical IgM kappa and IgG kappa paraproteins isolated from the serum of patient, Cam. The N-terminal 39 residues of the kappa chains of the IgM and IgG were identical and belonged to the human V kappa III subgroup. This sequenced stretch included the first L chain hypervariable region. The N-terminal 27 residues of the variable regions (VH) of the respective mu and gamma heavy chains were also identical and belonged to the human VHIII subgroup. These identical VH sequences were unique with lysine residues at positions 13 and 19. This dual lysine substitution has not been seen in 37 other human VHIII sequences reported in the literature. This N-terminal sequence homology in the V-regions of Cam IgM kappa and IgG kappa paraproteins and the shared idiotypy expressed by Cam IgM, IgG, and IgA proteins strongly suggest the existence of complete structural homology in the variable regions of the and L chains of these Ig molecules of three separate Ig classes. At the cellular and genetic level, these results point toward a common clonal origin for the idiotypically related Ig molecules and suggest that identical V-region (VH and VL) genes were utilized by the Cam lymphoid clone in the biosynthesis of the respective IgM, IgC, and IgA proteins.     

Characterization of the gamma subunits of the 7S nerve growth factor complex.

The gamma subunits of the 7S nerve growth factor complex (7S NGF) display arginine esteropeptidase activity. By varying the conditions of electrophoresis in acrylamide gel, it has been demonstrated that the gamma-subunit fraction of 7S NGF contains five different proteins, in contrast to the three (gamma1, gamma2, and gamma3) originally described (Smith, A.P., Varon, S. and Shooter, E.M. (1968), Biochemistry 7, 3259-3268); the gamma1 and gamma2 subunits, previously thought to be single species, can each be resolved into two components. The two components of the gamma1 subunit have the same isoelectric point, as do the two components of the gamma2 subunit. The distribution of protein among the two components of each of the gamma1 and gamma2 subunits varied from preparation to preparation. Moreover, a shift in the distribution for the gamma1 subunit was accompanied by a parallel shift for the gamma2 subunit. All of the different gamma proteins have the same molecular weight. On the basis of the molecular weights of the peptide chains of the gamma subunits and of the species which are formed by cross-linking with dimethyl suberimidate, it was concluded, that both the gamma1 and gamma2 subunits contain one species with two peptide chains and another with three peptide chains, while the gamma3 subunit is a single species with three peptide chains. The results also suggest that two of the chains in the three-chain species are derived, by proteolytic cleavage, from the larger chain in the two-chain species.     

Heat-treated smooth muscle tropomyosin.

Gizzard smooth muscle tropomyosin, which is close to 100% gamma beta heterodimer in the native state, was heated to about 100 degrees C, at which temperature the chains are dissociated, followed by reassociation by rapid cooling to 0 degree C. This heat-treated tropomyosin was composed of about 58% heterodimer and 42% of the gamma gamma and beta beta homodimers and had a lower viscosity than that of the native protein, indicating a reduced end-to-end polymerization. Close to 100% heterodimer was regenerated if the heat-treated tropomyosin was subjected to slow cooling from 50 degrees C. However, the viscosity remained low and did not return to the value for untreated tropomyosin, suggesting that the 100 degrees C treatment results in irreversible chemical damage to tropomyosin which affects its end-to-end interaction. Therefore, heat treatment of tropomyosin, a procedure widely used in the preparation of smooth muscle and non-muscle tropomyosins, may result in tropomyosin with a different heterodimer/homodimer distribution and different properties from those of the native protein and should be used with caution.     

The CD3-gamma and CD3-delta subunits of the T cell antigen receptor can be expressed within distinct functional TCR/CD3 complexes.

The T cell receptor for antigen (TCR) consists of two glycoproteins containing variable regions (TCR-alpha/beta or TCR-gamma/delta) which are expressed on the cell surface in association with at least four invariant proteins (CD3-gamma, -delta, -epsilon and -zeta). CD3-gamma and CD3-delta chains are highly homologous, especially in the cytoplasmic domain. The similarity observed in their genomic organization and their proximity in the chromosome indicate that both genes arose from duplication of a single gene. Here, we provide several lines of evidence which indicate that in human and murine T cells which expressed both the CD3-gamma and CD3-delta chains on their surface, the TCR/CD3 complex consisted of a mixture of alpha beta gamma epsilon zeta and alpha beta delta epsilon zeta complexes rather than a single alpha beta gamma delta epsilon zeta complex. First, a CD3-gamma specific antibody failed to co-immunoprecipitate CD3-delta and conversely, several CD3-delta specific antibodies did not coprecipitate CD3-gamma. Secondly, analysis of a panel of human and murine T cell lines demonstrated that CD3-gamma and CD3-delta were expressed at highly variable ratios on their surface. This suggested that these chains were not expressed as a single complex. Thirdly, CD3-gamma and CD3-delta competed for binding to CD3-epsilon in transfected COS cells, suggesting that CD3-gamma and CD3-delta formed mutually exclusive complexes. The existence of these two forms of TCR/CD3 complexes could have important implications in the understanding of T cell receptor function and its role in T cell development.     

Absence of preferential homologous H/L chain association in hybrid hybridomas

Bispecific mAb contain two Ag-combining sites each composed of a different combination of H and L chains. The resulting ability to react with and cross-link two different Ag makes these molecules a novel tool for application in biology and medicine. Intact bispecific mAb can be made only by biologic means, e.g., by fusion of two established hybridomas. Appropriate assembly of bispecific mAb by these hybrid cells depends on H = L chain behavior: strong preferential homologous H-L pairing would benefit the yield of bispecific antibodies. We have analyzed the Ig species produced by eight hybrid hybridomas (quadromas). Quadroma-produced IgG was fractionated and characterized for H and L chain content. The Ag reactivities were verified by using ELISA and immunofluorescence. Preferential homologous pairing was seen only with a minority of H-L chain pairs; L chains associated on average in a random fashion with H chains. This indicates that in the B cells from which the parental hybridomas were obtained, no strong selection had occurred on H-L recombination. Our results extend recent biochemical competitive H-L reassociation experiments, where on average an at random association of L chains with H chains was found; evidently this random association occurs in our biologic system as well. For the biologic production of bispecific antibodies this means that only in a small number of cases the "ideal" producer will be met. From the viewpoint of generation of antibody diversity, our results favor a large freedom for combinatorial binding of H and L chains during B cell ontogeny.     

Analysis of the diversity of murine antibodies to dextran B1355: N-terminal amino acid sequences of heavy chains from serum antibody.

The N-terminal amino acid sequences of two gamma and two mu chains from normally induced serum antibodies to dextran in BALB/c mice are presented. These heavy chains are derived from antibodies with three distinguishable idiotypes. These variable region (VH) sequences are all identical as far as they have been analyzed (27 to 53 residues). The light chains from these antibodies are all of the lambda type and are identical by isoelectric focusing analysis. Accordingly, the diversity of dextran antibodies appears to reside primarily in the heavy chains. The implications of these observations for antibody diversity are discussed.     

Evolution of variable and constant domains and joining segments of rearranging immunoglobulins

The rearranging immunoglobulins (Igs) are a family of recognition and defense proteins found in all vertebrate classes. These proteins consist of two types of polypeptide chains; each of these contains a variable (V) domain, a joining (J) segment, and a constant (C) region, which can itself consist of one to four domains. The distinction between light and heavy chains is an ancient one phylogenetically that is reflected in the structures of V, J, and C regions. Despite the early emergence of these genetic elements, conservatism is apparent in the peptide structures encoded by V, J, and C exons. C regions of heavy chains did not evolve as single units; rather the individual domains show their own clustering patterns, which apparently are independent of heavy-chain designation or species. C-region domains of light chains and the T cell receptor beta chain are similar to one another and to the most carboxyl-terminal domain of heavy chains. Comparison of the light chains of sharks, bullfrogs, chickens, and mammals indicated that a phylogenetic distinction can be made between kappa and lambda light chains. V and J segments of the rearranging T cell receptors alpha, gamma, and delta are homologous to the corresponding segments of Igs, but their C regions form a group that is markedly distinct from those of conventional Igs and Tcr beta.     

Ligand-induced assembly and activation of the gamma interferon receptor in intact cells.

Functionally active gamma interferon (IFN-gamma) receptors consist of an alpha subunit required for ligand binding and signal transduction and a beta subunit required primarily for signaling. Although the receptor alpha chain has been well characterized, little is known about the specific role of the receptor beta chain in IFN-gamma signaling. Expression of the wild-type human IFN-gamma receptor beta chain in murine L cells that stably express the human IFN-gamma receptor alpha chain (L.hgR) produced a murine cell line (L.hgR.myc beta) that responded to human IFN-gamma. Mutagenesis of the receptor beta-chain intracellular domain revealed that only two closely spaced, membrane-proximal sequences (P263PSIP267 and I270EEYL274) are required for function. Coprecipitation studies showed that these sequences are necessary for the specific and constitutive association of the receptor beta chain with the JAK-2 tyrosine kinase. These experiments also revealed that the IFN-gamma receptor alpha and beta chains are not preassociated on the surface of unstimulated cells but rather are induced to associate in an IFN-gamma-dependent fashion. A chimeric protein in which the intracellular domain of the beta chain was replaced by JAK-2 complemented human IFN-gamma signaling and biologic responsiveness in L.hgR. In contrast, a c-src-containing beta-chain chimera did not. These results indicate that the sole obligate role of the IFN-gamma receptor beta chain in signaling is to recruit JAK-2 into the ligand-assembled receptor complex.     

Internalization of the amino terminal 1-78 sequence of the gamma chain of native fibrinogen and exposure associated with catabolism and plasmin cleavage.

Specific antibodies of the gamma 1--78 peptide of human fibrinogen were employed in binding assays and in equilibrium competitive inhibition assays to analyze the expression of gamma 1--78 antigenic determinants as an indication of the relative exposure of the gamma 1--78 sequence in the E domain of fibrinogen, high solubility fibrinogen subfractions I-8 and I-9, and plasmic cleavage fragments of fibrinogen and fibrin. A very limited exposure of gamma 1--78 sequences was found to occur concomitant with proteolytic deletions of the major carboxyterminal segment of the A alpha chains in fgI-8, fgI-9. Exposure of gamma 1--78 is not influenced by further proteolysis to fg-X which is associated with B beta 1--43 deletion. Further proteolysis to fg-Y, which is associated with deletion of beta 43--53 and of one of the D domains, is associated with additional exposure of gamma 1--78. This is not significantly influenced by further proteolysis to fg-E with deletion of the second D domain, deletion of A alpha 1--19, and proteolysis at the carboxyterminal aspects of the E domain chains.     

Simultaneous expression of mu- and gamma-chain mRNA in cloned murine B-lymphoma cell lines

The expression of immunoglobulin heavy chain genes was studied in five murine B-lymphomas known from previous studies to express either mu (38C-13), mu + delta (L10A, K46, BCL1) or gamma chains (A20). The presence of mu- and gamma-mRNAs in these tumors was determined by Northern blot analyses of the total cell poly(A)+ mRNA, using the appropriate 32P-labeled recombinant plasmid probes. In four out of the five lymphomas examined, both mu- and gamma-mRNAs were detected. The mu-mRNA appeared as multiple discrete bands of 1.9-3.0 kb. In three out of the four lymphomas, the gamma-mRNA appeared as two bands, a major one of 1.9 and a minor one of 3.9 kb. Three myelomas examined by similar methods did not contain more than one class of heavy chain mRNA. Reexamination of the Ig chains produced by the B-lymphomas which expressed both mu- and gamma-mRNAs revealed that two of them preserved their original phenotype and expressed mu (38C-13) or gamma chains only (A20). In contrast, two of the cell lines previously shown to express mu but not gamma chains (i.e., L10A and K46R) had changed during growth in culture and 'switched' to the production of gamma chains only. These results indicate that, in contrast to myelomas, B-lymphomas possess two classes of mRNA. However, the production of heavy chain mRNA in B-lymphomas is not necessarily accompanied by synthesis of the corresponding polypeptide chains. More studies are necessary to find out whether the expression of 'non-productive' heavy chain mRNA molecules in B-lymphomas is related to the phenomena of 'allelic exclusion' and/or the 'heavy chain switch' which occurs during the maturation of B-cells.     

Identification of B beta chain domains involved in human fibrinogen assembly.

Fibrinogen chains are assembled in a stepwise manner in the rough endoplasmic reticulum prior to secretion of the final six-chain dimeric molecule. Previous studies indicated that the synthesis of B beta may be a rate-limiting factor in the assembly of human fibrinogen. To determine the domains of B beta which interact with the other two component chains of fibrinogen, deletion mutants of B beta were transiently co-expressed, together with A alpha and gamma chains, in COS cells, and fibrinogen assembly and secretion were measured. Deletion of the COOH-terminal half of the B beta chain (amino acids 208-461) did not affect assembly and secretion. Assembly of A alpha, gamma, and B beta also occurred when the first NH2-terminal 72 amino acids of B beta were deleted, but not when 93 amino acids were deleted. This indicates that the B beta domain between amino acids 73 and 93 is necessary for the assembly of the three fibrinogen chains. This domain marks the start of the alpha-helical "coiled-coil" region of fibrinogen.     

Assembly and secretion of fibrinogen. Degradation of individual chains.

Hep G2 cells produce surplus A alpha and gamma fibrinogen chains. These excess chains, which are not secreted, exist primarily as free gamma chains and as an A alpha-gamma complex. We have determined the intracellular location and the degradative fate of these polypeptides by treatment with endoglycosidase-H and by inhibiting lysosomal enzyme activity, using NH4Cl, chloroquine, and leupeptin. Free gamma chain and the gamma component of A alpha-gamma are both cleaved by endoglycosidase-H, indicating that the gamma chains accumulate in a pre-Golgi compartment. Lysosomal enzyme inhibitors did not affect the disappearance of free gamma chains but inhibited A alpha-gamma by 50%, suggesting that A alpha-gamma is degraded in lysosomes. The degradative fate of individual chains was determined in transfected COS cells which express but do not secrete single chains. Leupeptin did not affect B beta chain degradation, had very little affect on gamma chain, but markedly inhibited A alpha chain degradation. Antibody to immunoglobulin heavy chain-binding protein (GRP 78) co-immunoprecipitated B beta but not A alpha or gamma chains. Preferential binding of heavy chain-binding protein to B beta was also noted in Hep G2 cells and in chicken hepatocytes. Taken together these studies indicate that B beta and gamma chains are degraded in the endoplasmic reticulum, but only B beta is bound to BiP. By contrast A alpha chains and the A alpha-gamma complex undergo lysosomal degradation.