Conformation dependence of antigenic determinants on the collagen molecule

Three different types of antigenic determinants were demonstrated in soluble collagen with the aid of rat, rabbit and chicken antisera to native collagen. Helical antigenic determinants which require an intact triple-helical structure of the molecule are mainly recognized by rat antisera. Renaturation of the serologically inactive unfolded polypeptide chains (denatured collagen) is accompanied by a significant recovery of serological activity. Central antigenic determinants which are probably located in the same regions of amino acid sequence are less accessible in the native antigen and become exposed upon denaturation. Equal titres for both types of determinants are found in chicken antisera. Immunization with denatured collagen, however, revealed a response restricted to the central type. Isolated antibodies specific for terminal non-helical antigenic determinants, as yet only known to occur in rabbit antisera, reacted equally well with native collagen, the unfolded polypeptide chains and with small cyanogen bromide peptides. Independence of conformation is therefore suggested for these antigenic structures.     

Demonstration of a unique antigenic specificity for the collagen alpha1 (II) chain from cartilaginous tissue.

The rabbit antibody response to native collagen (chain composition [alpha1(II)3) from cartilaginous tissue, has been examined by agglutination assays, gel diffusion, haemagglutination-inhibition studies, and immunoadsorption. The results show that the rabbit antibody response to the cartilage-type collagen is characterized by considerable reactivity to both helical [alpha1(II)]3 as well as alpha1(II) chains. This is in contrast to the rat antibody response to the same antigens where titres are generated to largely helical antigenic determinants. Similarly to the rat response, rabbit antibodies to [alpha(II)]3 exhibit no strong cross-reaction with the genetically distinct [alpha(I)]2ALPHA2 collagen or its component chains. Strong cross-reactions were, however, observed between bovine and chick alpha1(II) chains. One of the major antigenic sites on [alpha1(II)]3 collagen appears to reside in the sequence represented by CB-11, a peptide derived from the helical portion of the [alpha1(II)]3 molecule after cyanogen bromide cleavage. The data, however, are compatible with the presence of other antigenic determinants which are probably located in the amino- and carbocy-terminal portions of the molecule.     

Independent appearance of anti-thymocyte and anti-RNA antibodies in NZB/NZW F1 mice.

Mouse antibodies to soluble bovine skin (type I) collagen react with determinants which are located in the rigid triple-helical portion of the antigen and become destroyed upon unfolding the molecule. Helical antigenic determinants are dependent on the genuine chain assembly, e.g. alpha[1(I)]2alpha2. Artefactual triplehelical structures of the composition [alpha1(I)]3 or [alpha2]3 or a genetically distinct type II collagen from cartilage showed no or only weak cross-reactivity. Pepsin treatment of type I collagen known to remove short, non-helical sequences at both ends of the molecule had virtually no effect on antigenicity and immunogenic activity. A radioimmunoassay failed to detect antibodies in three congenic resistant mouse strains immunized with denatured type I collagen. These strains had been previously classified as high or low responders to native type I collagen. Agglutination titres vs denatured collagen culd already be demonstrated in nonimmune sera. The agglutinating activity was labile against heating at 56 degrees and could not be increased by immunization. Two out of five inbred strains showed a high response against pepsin-dissolved bovine type II collagen with the chain composition [alpha1(II)]3. Lack of correlation in the responder state to both collagen types indicated control by different immune response genes. Antibodies to type II collagen also reacted against triple-helical antigenic determinants and showed neglible cross-reaction with type I collagen.     

Characterization of conformation independent antigenic determinants in the triple-helical part of calf and rat collagen

About 20 per cent of the antibodies in rabbit antisera to native calf or rat collagen exhibited affinity for denatured rabbit collagen and could be isolated by immunoadsorption. Such antibodies reacted with the unfolded α1-chain as well as with the α2-chain of collagen. Inhibition experiments suggested that the two kinds of polypeptide chains are not completely equivalent in their antigenic determinants. These determinants were not significantly influenced by a treatment of native collagen with pronase, a procedure known to remove short, non-helical sequences at both ends of the molecule. The results suggested that the antigenic determinants are conformation independent. They are, however, mainly located in the middle region of collagen, having a rather complex conformational structure.

Cyanogen bromide cleavage of collagen did not impair the serologic activity of these determinants but with one exception none of the individual cyanogen bromide peptides possessed the full activity of the entire α-chain. However, most of the peptides could be agglutinated by the antibodies when put onto tanned red cells. Inhibition studies of these agglutination reactions clearly demonstrated that virtually all of the peptides carry unique antigenic determinants, which occasionally are shared by a few other peptides. Additional evidence for heterogeneity was obtained by further cleavage of the cyanogen bromide peptides with proteases. The minimum number of antigenic determinants thus estimated in calf collagen was nine. Evidence is provided that their structure in most cases does not correspond to sequences of the type Gly-Pro-X.

     

Repeated helical epitopes of defined amino acid sequence in human type III collagen identified by monoclonal antibodies.

Two monoclonal antibodies, designated 1F8 (IgG1) and 5B10 (IgG1), have been produced in mice against native human type III collagen. These antibodies were highly type and species specific, recognizing the triple helical domain of type III as tested by ELISA. Immunofluorescence studies using each of these antibodies resulted in a fibrous staining pattern in human skin dermis. Immunogold electron microscopy resulted in a periodic distribution of gold particulates along banded collagen fibrils. Assuming that the total contour length of pepsin digested type III collagen is 300 nm, measurements of antibody-antigen complexes visualized by rotary shadowing revealed that each antibody bound at the same two sites: one approximately at the middle of the helix (153 nm from the N-terminus), the other at a site one-quarter the triple helical length from the N-terminus (75 nm). That the one-quarter binding site was closest to the N-terminus was determined by antibody incubation following tadpole collagenase treatment, which results in a larger, N-terminus containing fragment (binding antibody) and a smaller C-terminus containing fragment (not binding antibody). Located at each antibody binding epitope is a sequence of 10 amino acids: Gly-Ala-Hyp-Gly-Leu-Arg-Gly-Gly-Ala-Gly. Renatured cyanogen bromide-cleaved(CB)-peptides, CB4 and CB8, containing these repeated sequences reacted with each antibody, whereas other renatured type III CB-peptides were unreactive as determined by Western blotting analysis and ELISA. This was further confirmed by inhibition tests using a 10 residue synthetic peptide of identical sequence, which yielded 20-30% inhibition of antibody binding to native type III collagen at 4 degrees C. However, no inhibition was noted at higher temperature. These results indicate that both monoclonal antibodies recognize a specific helical conformation of 10 or slightly fewer residues in the three identical polypeptide chains comprising type III collagen.     

The structural basis of cell-mediated immunological reactions of collagen: Reactivity of separated α-chains of calf and rat collagen in cutaneous delayed hypersensitivity reactions

Denatured calf and rat skin collagen and α1- and α2-chains from these collagens were tested for their capacity to sensitize guinea-pigs for the cutaneous delayed hypersensitivity reaction and to elicit this reaction in sensitized animals. Alpha-chains from neutral salt extracted and from urea extracted collagen were also compared. All preparations were fully active as sensitizers as well as elicitors of the skin reaction. In spite of extensive cross-reactions, significant differences between α1- and α2-chains of a given species as well as between corresponding chains of different species could be detected. Differences between chains from neutral salt extracted collagen and from urea extracted collagen were also revealed.     

Bone Type V Collagen: Chain Composition and Location of a Trypsin Cleavage Site

The component α-chains of type V collagen from bovine bone were isolated and structurally characterized by gel electrophoresis, high performance liquid chromatography (HPLC) and amino acid sequence analysis. Three distinct α-chains were identified. Two of these were the well described α1(V) and α2(V) chains; the third proved to be identical to the cartilage α1(XI) chain. In adult bone the ratio between the three chains was about 1:1:1. Native type V collagen was cleaved by trypsin at 33°C or 37°C into 3/5 fragments. Aminoterminal sequence analysis of the α1(V) and α1(XI) fragments showed they both resulted from trypsin cleavage between residue 434 and 435. Trypsin apparently cleaves the type V molecule within a relatively unstable domain of the triple helix which presumably may also be a natural site of initial cleavage by a protease in vivo.     

Dependence of the antibody response of guinea-pigs to collagen on the type of adjuvant and on the conformation of the antigen.

The immunological response to collagen of guinea-pigs is strongly dependent on the conformation of the antigen and on the type of adjuvant. Freund's complete adjuvant facilitated excellent delayed hypersensitivity skin reactions to native (triple helical conformation) as well as denatured (random coil conformation) collagen. Immunization of guinea-pigs with collagen in this adjuvant gave rise to very low levels of antibody to native collagen and failed to induce antibodies to denatured collagen. Use of Freund's incomplete adjuvant resulted in excellent antibody responses to native collagen, but it did not induce antibodies to denatured collagen. Animals injected with collagen and Freund's incomplete adjuvant were not sensitized for cell-mediated immunological reactions. The antibodies to collagen were specific with regard to collagen from various species but displayed different degrees of cross-reactivities depending on the species of collagen used for immunization. They were specific for the triple helical conformation of the collagen molecule.     

Requirement for T cells in the antibody response of mice to calf skin collagen.

The antibody response to triple-helical calf skin collagen is apparently regulated by T cells since nude mice (BALB/c strain) responded to this antigen only after transfer of T cells. Syngeneic transfer of B and T cells was required to prepare thymectomized and irradiated normal C57B1/10 recipients for an anti-collagen response. Reconstitution by B cells alone was unsuccessful. The results are compatible with previous evidence on the H-2 linked genetic control of the antibody response to and the non-repetitious nature of helical antigenic determinants in calf collagen.     

Specificities of monoclonal antibodies against the activated delta-endotoxin of Bacillus thuringiensis var. thuringiensis.

Eight hybrid cell lines secreting monoclonal antibodies directed against the activated delta-endotoxin of Bacillus thuringiensis var. thuringiensis were grown in BALB/c mice. Ascites fluids were collected, and the antibodies were purified by antigen-affinity chromatography. The specificity of each monoclonal antibody for the toxin and protoxin was established by the indirect enzyme-linked immunosorbent assay. All the antibodies consisted of gamma 1 heavy and kappa light chains. They were reactive with both the native toxin and the protoxin. In contrast to specific goat antiserum, they failed, however, to bind to heat and sodium dodecyl sulfate denatured antigen. These eight cloned cell lines gave rise to five kinds of antibodies distinguished by isoelectric focusing. Competitive antibody binding studies revealed that these five antibodies recognize at least four distinct antigenic determinants of the native toxin and the protoxin. Two of the epitopes are unrelated, whereas three antibodies compete for binding to their antigenic determinants. In the bioassay with larvae of Pieris brassicae, one antibody was found to block the toxin and protoxin activity completely. A second inhibited it partially, whereas the other three antibodies did not affect it at all.     

Methods in laboratory investigation. Monoclonal antibodies to type IV collagen: probes for the study of structure and function of basement membranes

Type IV collagen is one of the main constituents of basement membranes, yet it is unknown whether the structural framework at different sites is assembled from one unique type of molecule or whether different type IV collagen molecules exist. To study the composition, chemical identity, and organization of this protein in different organs we have prepared monoclonal antibodies to a type IV collagen preparation from human placenta. Swiss Webster mice were hyperimmunized, and splenic cells were fused with the three different myeloma cell lines SP2/0, NS1, and U1. Type IV collagen-specific hybrids were selected and cloned by limiting dilution and on hard agar. Monoclonal antibodies secreted by two clones were extensively characterized by ELISA-inhibition assay, immunoprecipitation, rotary shadowing, and immunofluorescence techniques. Unlike conventionally raised antibodies in rabbits, both monoclonal antibody reagents show species-specific binding exclusively to native type IV collagen from human placenta but not to a similar preparation from calf lung or to other types of collagen. After heat denaturation of the antigen binding was no longer observed. The M3F7 antibody-binding site is located within the triple helical domain of the type IV molecule, approximately 900 A removed from the amino terminal end as visualized by a metal shadow casting technique. The monoclonal antibody M3F7 precipitates material from pepsin-derived and radiolabeled type IV collagen, and analysis of the polypeptide chains in the immunoprecipitate by sodium dodecyl sulfate polyacrylamide gel electrophoresis suggests that two major fragments are contained in the precipitate, which yield polypeptides of about 100 and 50 kilodaltons. After rotary shadowing of antigen-antibody mixtures native collagen fragments of two different size classes that bind antibody are visualized. One fragment is approximately 1500 A in length, and the other measures about 2700 to 3000 A. The localization of the antigenic site on these fragments suggests that both are generated by pepsin cleavage at a site about 900 A removed from the amino terminal end. In immunofluorescence experiments the monoclonal antibodies stained all basement membranes in kidney, lung, placenta, or skin, suggesting that at least the type IV collagen molecule recognized by these monoclonal antibodies is shared by a variety of vascular and epithelial basement membranes.     

Detection of homocytotropic antibody associated with a unique immunoglobulin class in the bovine species

Homocytotropic antibodies capable of sensitizing calf skin were demonstrated in serum of cows as early as 7 to 9 days following antigenic stimulation with rabbit serum albumin. In addition, the antibody was present in colostrum of the immunized dam and was taken up by the intestine of the newborn calf as an intact molecule. Skin-sensitizing activity was also detected in calf urine after uptake of colostrum. Evidence is presented for an anamnestic response, suggesting the existence of an immunological memory for the formation of skin-sensitizing antibodies. The homocytotropic antibody was shown to be heat labile and sensitive to treatment with 2-mercaptoethanol. Time course studies revealed that homocytotropic antibodies persist in calf skin for at least 8 weeks. The immunoglobulin class associated with homocytotropic activity was enriched from appropriate antisera by salt precipitation, DEAE-cellulose chromatography and gel filtration on Sephadex G-200 columns. The final preparation was still capable of sensitizing calf skin for positive PCA reactions using a minimum dose of 0.009 μg N. The homocytotropic activity of the vinal fraction could not be removed by antibodies specific for the IgG class, but was completely blocked by antibodies against the fraction enriched in homocytotropic activity. The immunoglobulin class associated with skin-sensitizing activity possesses antigenic determinants which are lacking in μ-, α-, and γ-chains. The results indicate that the homocytotropic activity in the bovine species is associated with a unique immunoglobulin class analogous to human IgE.     

Localization and structural features of an antigenic determinant in sea anemone collagen

A major antigenic site of sea anemone collagen was localized in the CNBr peptide, CB6, which is derived from a central, helical portion of the molecule. Further tryptic digestion of CB6 yielded a serologically active peptide containing 58 amino acid residues and including the only two tyrosine residues of the α-chain. The localization of antigenic determinants in terminal, nonhelical sequences previously found for several mammalian collagen antigens is apparently less important for invertebrate collagen.     

Immunochemical study on basement membrane (type IV) collagens.

Basement membrane (type IV) collagens were extracted from a mouse tumour with acetic acid and from human placenta after limited enzymatic digestion. Antisera were produced against both collagens in rabbits and guinea-pigs and examined by various assays. These antisera were found to be specific for basement membrane collagen and showed little or no cross-reactions with the interstitial collagens, types I, II and III or with human placenta collagen consisting of alpha A and alpha B chains. Varying degrees of cross-reaction were observed between antisera to human and mouse type IV collagen. Immunochemical analyses demonstrated the presence of three distinct determinants in the tumour type IV collagen. Rabbit antisera against this antigen reacted with either collagenase-resistant segments or with a collagenous, disulphide-bonded segment (P3). Guinea-pig antisera recognized primarily antigenic determinants in the P3 segment. Antisera to placenta type IV collagen reacted with another collagenous, pepsin fragment (P1) which lacks disulphide bonds. These antisera showed complete cross-reaction with collagenous alpha 1 (IV) chains prepared from pepsin-digests of human placenta and bovine lens capsule.     

Identification of a major antigenic epitope on CNBr-fragment 11 of type II collagen recognized by murine autoreactive B cells.

Immunization of certain strains of mice with native type II collagen (CII) induces both development of arthritis and an antibody response to autologous CII. The autoantibody response in a high-responder strain, the DBA/1 mouse, has been described earlier, and a number of monoclonal antibodies have been characterized for arthritogenicity and autoreactive binding to cartilage in vivo and in vitro. Here we map the antigenic epitope of one of these arthritogenic monoclonal antibodies (CII-C1). It belongs to a group of antibodies recognizing the CNBr fragment alpha 1(II)-CB11 of CII. Using the enzyme-linked immunosorbent assay technique, we show that the antibody reacts only with native, triplehelical CII, but not with other collagens. The antibody is able to stain specifically the CB11 fragment by immunoblotting, suggesting some partial renaturation of the CNBr fragment into triple-helical structures after blotting. The binding site of CII-C1 on CB11 was further focused by rotary shadowing of antibody-labeled CII to a site 89 +/- 8 nm from the amino end of CII, corresponding to the middle of CB11. This location was confirmed by cleavage of CB11 with trypsin, separation of the tryptic peptides by high-performance liquid chromatography and dot-blot analysis of the antigenic peptides with the CII-C1 antibody. Sequencing of the single positive peptide located the antigenic epitope within the sequence GFAGQAGPAGATGAPGRP (residues 316-333). Assuming 0.29 nm per residue, this corresponds to a position within 92-96.5 nm from NH2 terminal end of CII. Apart from glycine residues, which are not exposed on the triple-helical structure, only two amino acid residues (F-x-y-Q) are conserved in CII from different species but are not found in the triple-helix of other collagens except type IV collagen. Therefore, this structure is likely to be of critical importance for the binding of the CII-C1 antibody. Of potential importance is that this structure is also found in certain other arthritogenic proteins such as 65-kDa mycobacterial protein, in CMV and EBV.     

Anti-collagen antibodies in sera from rheumatoid arthritis patients.

Anti-cartilage antibodies, demonstrable by immunofluorescence, were found in 3.3% of rheumatoid arthritis patients. In most of these patients antibodies to type II collagen were detected. In specificity studies on these anti-collagen antibodies, they appeared to be type specific, showing no reaction with collagen types I and III. Denatured type II collagen reacted much less well than native type II, but isolated peptides from different regions of the collagen molecule were differentiated by individual sera. Removal of the glycoside side chains from native type II collagen had no effect on its antigenicity. The findings suggest that these patients produce highly specific antibodies which react with the triple helix of type II collagen.     

The immune response to human type III and type V (AB2) collagen: antigenic determinants and genetic control in mice

The genetic influences on the specific IgG antibody response to helical antigenic determinants of native soluble type III and V (AB2) human collagen have been studied in mice. H-2-linked Ir genes have been detected for each of these collagens, as have background effects. The alleles controlling high antibody responses can be distinguished from those previously described for bovine collagen type I and type I propeptide. Apparent two-gene complementation in the response to type V (AB2) may indicate that this antigen will be a useful probe in further studies of the complex nature of antigen presentation for T cell recognition. The present studies may facilitate the production of specific, high-affinity monoclonal antibodies for the human collagens which, in turn, will facilitate the study of connective tissue biology and pathology.     

Antigenicity of the M proteins of group A hemolytic streptococci: further evidence for shared determinants among serotypes.

Shared antigenic determinants between M proteins of group A streptococci appear to be widespread among serotypes. This is demonstrated by the ability of purified M proteins to absorb opsonic antibody from a variety of heterologous antisera prepared against whole cells or purified M proteins. This absorption procedure has the capacity to separate passive mouse protecting and passive hemagglutinating antibodies from opsonic antibodies measured in vitro. When whole cells or M proteins are used as heterologous absorbents, immunoglobulin may be recovered from the cells or precipitates. The recovered antibody has most of the opsonic and some of the precipitating qualities of the original unabsorbed serum, but hemagglutinating titers are significantly lower. These data provide additional evidence that shared antigenicity among M proteins can be the result of common antigenic determinants. Arguments are presented that these cross-reactions are not the result of a nonspecific protein fraction associated with purified M proteins.     

Isolation and characterization of secretory IgA (sIgA) and free secretory component (FSC) from rat bile

Secretory IgA (sIgA) and free secretory component (SC) have been purified from rat bile and compared to sIgA and free SC from rat and human milk. In rat bile and milk, sIgA exists in a series of polymers with S rates of 11,13 and 15 and molecular weights of 405,000, 540,000 and 690,000, respectively. These three forms have the same basic chain composition (light chains, -chains and SC) when reduced and analysed by polyacrylamide gel electrophoresis in SDS. Rat -chains and SC were slightly smaller than their human counterparts. SC was covalently linked to IgA in purified biliary sIgA; the latter apparently contained molecules of SC in different configurations, as revealed by antigenic analyses with anti-free-SC and anti-bound-SC antisera. There was antigenic identity between bile and milk for both FSC and sIgA. The existence of configurational determinants of SC in both sIgA and FSC was serologically demonstrated. The amino acid composition of rat bile free SC compared well to that of bovine, canine, rabbit and human free SC.     

Monoclonal anti-VH antibodies recognize a common VH determinant expressed on immunoglobulin heavy chains from various species

Our previous work using rabbit antibodies to the variable region of MOPC315 myeloma heavy chain (VH) has indicated the existence of framework determinant(s) common to many murine heavy chains. Here we report the characterization of anti-VH monoclonal antibodies (mAb) prepared in an attempt to elucidate the nature of the common VH determinant. We immunized AKR/J mice with a purified VH315 fragment and generated somatic cell hybrids by the fusion of the immune AKR/J splenocytes with the NS1 myeloma cells. Thirty-seven common anti-VH and 57 subgroup VHI-specific hybridomas have been established and characterized. Whereas the anti-subgroup mAb seemed to react with a determinant unique to the MOPC315 (mouse VHI) subgroup, all the anti-VH mAb reacted with myeloma heavy chains of different VH subgroups, class and allotypes. Antibody competition studies revealed that the VH subgroup determinants are distinct from the common VH determinants and that both were also recognized by the rabbit polyclonal antibodies. The common VH determinants were found to be "hidden" determinants on intact immunoglobulin molecules being exposed only on isolated heavy chains. Furthermore, they are sequential determinants since they are preserved on fully denatured heavy chains. The common VH determinants are shared by immunoglobulins of a wide range of vertebrates from amphibia to man and thus represent antigenic structures which were highly conserved throughout evolution.