Properties of cryptic epitopes and their corresponding antibodies as indicated by the study of human and ovine growth hormones

Antibodies (Ab) directed to hidden antigenic determinants (cryptotopes) are undesirable because they are not neutralizing. Additionally, we have previously demonstrated a close association between the extent of Ab to cryptic determinants and the expression of autoantibodies (autoAb) under some experimental conditions. Thus, the first objective of this work was to establish the physicochemical characteristics of Ab to cryptotopes and the second one was to examine the structural features of cryptic epitopes themselves. Using human and ovine growth hormones (hGH and oGH) as antigenic models and competition ELISA under different conditions of temperature, pH or ionic strength, we did not find any difference between the binding properties of anti-cryptic epitope antibodies (Ab) and anti-native epitope Ab. Then, using synthetic peptides and tryptic digests and direct and competition ELISAs we studied the structures of cryptic hGH and oGH epitopes. Isolated peptides either in solution or adsorbed on microplates failed to react. Partially digested hGH was recognized only when insolubilized on microplates, and anti-oGH Ab only reacted with a large fragment of the hormone either in solution or insolubilized. These results indicate that, at least in the case of hGH and oGH, cryptic epitopes are not simple linear sequences, as commonly referred without any evidence, but new exposed conformational structures different from those found in the native antigen.     

Changes in the Specificity of Antibodies in Mice Infected with Lactate Dehydrogenase-Elevating Virus

Infection with lactate dehydrogenase-elevating virus (LDV) modifies the isotypic distribution of antibodies (Ab) directed to several antigenic proteins with a preferential production of IgG2a. Because it was not known whether the virus could also affect the Ab specificity, the authors addressed this point using human growth hormone (hGH) as a model antigen. Anti-hGH monoclonal antibodies (MoAb) were used as probes to study the occurrence of Ab to three native hGH epitopes (3C11, F11 and 10D6) in sera from LDV-infected CBA/Ht and BALB/c mice immunized with hGH. Competition ELISA was used to determine the extent of Ab directed to cryptic hGH epitopes, i.e. antigenic determinants hidden in the native hormone. Results indicated that in LDV-infected CBA/Ht mice the titres of anti-hGH Ab were lower than in controls, although a consistent isotypic shift to IgG2a subclass was observed. Concurrently, the presence of Ab to epitopes 3C11, F11 and/or 10D6 were markedly reduced in infected animals and most anti-hGH Ab were directed to hGH cryptic epitopes. By contrast, LDV infection increased the amount of anti-KLH Ab elicited by CBA/Ht mice and did not affect Ab specificity, whilst control and LDV-infected BALB/c mice showed similar concentrations of anti-hGH Ab. Furthermore, the proportion of Ab to cryptic hGH epitopes did not change in infected animals even though an important shift to IgG2a was detected. Thus, data presented herein suggest that LDV infection modifies Ab specificity depending on the mice genetic background and on the antigenic characteristics of the immunogen.     

Sequence similarity and structural homologies are involved in the autoimmune response elicited by mouse hepatitis virus A59

The features of autoantibodies (autoAb) to liver fumarylacetoacetate hydrolase (FAH) elicited in mice infected with mouse hepatitis virus (MHV) were studied by ELISA and western-blot competition assays. All sera tested contained Ab to cryptic FAH epitopes according with results from western-blot tests, whereas ELISA data indicated that some of these same sera did recognize native epitopes of the autoantigen (autoAg). Such differences were detected in individual sera from various mouse strains, and were ascribed to the fact that proteins insolubilized on solid supports expose a variety of conformational and cryptic antigenic determinants. On the other hand, whereas results from both experimental protocols showed that anti-MHV Ab did not cross-react with the soluble autoAg, the opposite situation did not show analogous results. Thus, binding of autoAb to insolubilized FAH could be inhibited by MHV depending on the mouse serum or the experimental protocol used. Additionally, a set of synthetic homologous peptides from mouse FAH and various viral proteins was employed to analyze the Ab repertoire of MHV-infected mice. Results indicated that two homologous peptides were recognized by most Ab: the N-terminal sequences (1-10) from FAH and the nucleocapside, both sharing 50% of identity, and sequence 2317-2326 of the RNA polymerase, a peptide showing 30% of identity with FAH 11-20. Results indicated that MHV-infection triggers at least three distinct Ab populations: anti-MHV, anti-FAH and cross-reacting Ab. This cross-reaction implies either sequential or conformational epitopes from both the viral proteins and the autoAg and may differ between individuals.     

Effects of various adjuvants and a viral infection on the antibody specificity toward native or cryptic epitopes of a protein antigen

An immunization protocol that induces antibodies (Abs) directed to cryptic epitopes of a protein antigen (Ag) reduces the efficacy of vaccines that ideally should induce Abs against native epitopes. We have shown earlier that viral infections concomitant with immunization against a protein tend to shift the Ab specificity toward cryptic epitopes and tend to induce the production of autoantibodies (autoAbs). Here, we show the effects of three adjuvants on the Ab specificity in the absence or presence of a viral infection (lactate dehydrogenase-elevating virus or LDV), with human growth hormone (hGH) being, as before, the protein Ag. Pathogen-free CBA/Ht and BALB/c mice were immunized with hGH in the presence of complete Freund's adjuvant (CFA), monophosphoryl lipid A (MPL) or alum, with the animals being either infected with LDV or not infected with LDV. Conventional and competition enzyme-linked immunosorbent assays (ELISAs) indicated that in noninfected mice, CFA induced higher titres of anti-hGH Ab than did MPL or alum, with the Ab being almost totally directed to cryptic hGH epitopes. Strikingly, CFA plus LDV infection in CBA/Ht mice shifted the specificity of the anti-hGH Ab toward native epitopes, whereas the virus decreased the Ab titre when MPL or alum was used. Our Western blot results showed that 70% of mice immunized with hGH in the presence of any adjuvant produced autoAbs against a variety of tissue Ags. The amount of autoAb and the concentration of Ab to hGH cryptic epitopes did correlate, suggesting a relationship between both kinds of Ab. Significant differences were observed in the various effects of adjuvants and the viral infection between the two mouse strains used in this work.     

Monoclonal antibodies to human growth hormone induce an allosteric conformational change in the antigen.

We re-investigated the properties of a monoclonal antibody (mAb), 4D11, to human growth hormone (hGH) that showed a very weak affinity, recognizing hGH only when the hormone was solubilized on a solid surface. MAb4D11 did not significantly bind 125I-hGH. It was found that three mAb directed to different hGH epitopes (mAb 3C11, 10C1 and NA71) were able to induce the binding of the soluble antigen to mAb 4D11. The co-operative effect could be demonstrated by the formation of binary complexes (Ag:Ab, 1:2) detected by high-performance liquid chromatography (HPLC) and by the increase of radioactivity found when the synergistic mAb were added to 125I-hGH incubated with mAb 4D11 immobilized on polyvinyl microplates. Other possible explanations, such as the formation of cyclic complexes or the generation of a new epitope in the Fc fragment of the first antibody (Ab), were dismissed because the Fab fragment of one of the enhancing mAb (3C11) gave the same effect as the intact Ab. The data suggest that the hGH molecule undergoes a localized conformational change after binding to mAb 3C11, NA71 or 10C1 and that mAb 4D11 binds with high affinity to the modified region of the hormone. The formation or not of ternary complexes (Ag:Ab, 1:3) was used to localize the 4D11 epitope on the surface of the Ag. It is suggested that mAb 4D11 recognizes a conformational change produced in the region defined by the AE5/AC8 epitopes, which is close to the hGH antigenic domain only expressed when the protein is immobilized on plastic surfaces.     

Epitope analysis of GAD65Ab using fusion proteins and rFab

The identification of disease-specific autoantibodies to the 65-kDa isoform of glutamate decarboxylase (GAD65Ab) epitopes in type 1 diabetes has been hampered by their conformational nature. Here, we compared two methods of GAD65Ab epitope analysis: GAD65/67 fusion proteins and competition assays using GAD65-specific recombinant fraction antigen binding (rFab). Sera from newly diagnosed type 1 diabetes patients (n=61) were studied using both approaches. Competition of GAD65 binding by an rFab to a specific epitope did not correlate with binding to the fusion protein that represented this epitope. Conversely, samples that bound to specific fusion proteins were not necessarily competed with rFab specific to determinants in the same region. We conclude that epitopes of different characteristics are detected by fusion proteins and by competition with rFab. Fusion proteins allow the definition of large epitope regions; however, some conformational GAD65Ab epitopes, especially those residing in the middle region, are destroyed or distorted in the fusion proteins. Competition studies using rFab allow the identification of conformational epitopes. However, monoclonal rFab may only reflect a limited proportion of the epitopes recognized by polyclonal sera. A combined analysis using both approaches may therefore be necessary to gain best understanding of autoantibody characteristics and affinity maturation.     

A general strategy to enhance immunogenicity of low-affinity HLA-A2. 1-associated peptides: implication in the identification of cryptic tumor epitopes

Low-affinity MHC class I-associated cryptic epitopes derived from self proteins overexpressed in a wide variety of human tumors or derived from antigens of viruses exhibiting a high mutation rate, could be interesting candidates for tumor and virus immunotherapy, respectively. However, identification of low-affinity MHC-associated epitopes comes up against their poor immunogenicity. Here we describe an approach that enhances immunogenicity of nonimmunogenic low-affinity HLA-A2.1-binding peptides. It consists of modifying their sequence by introducing a tyrosine in the first position (P1Y). P1Y substitution enhances affinity of HLA-A2.1-associated peptides without altering their antigenic specificity. In fact, P1Y variants of ten nonimmunogenic low-affinity peptides exhibited a 2.3- to 55-fold higher binding affinity and/or stabilized the HLA-A2.1 for at least 2 h more than the corresponding native peptides. More importantly, P1Y variants efficiently triggered in vivo native peptide-specific CTL which also recognized the corresponding naturally processed epitope. The possibility for generating CTL against any low-affinity HLA-A2.1-associated peptide provides us with the necessary tool for the identification of cryptic tumor and virus epitopes which could be used for peptide-based immunotherapy.     

Relative distribution of various antigenic determinants on the human growth hormone surface

The relative distribution of 12 antigenic determinants on the surface of the human growth hormone (hGH) molecule has been established. The necessary information was obtained by testing the ability of paired monoclonal antibodies (MAb) to bind simultaneously or not, to 125I-hGH which leads to the formation of 1:2 or 1:1, Ag-Ab complexes, respectively. The results obtained indicate that the epitopes occupy a large percentage of the total hGH molecular surface and revealed the existence of; an antigenic region specific for hGH; at least two independent domains of immunological identity between hGH and human placental lactogen (hPL), one of them also shared by heterologous GH; and other independent areas of partial cross reactivity with hPL. MAb competition experiments in a solid-phase RIA showed the unreliability of this technique for mapping purposes. The distribution of the hGH epitopes suggested in this work is in accord with present views on protein antigenicity and also explains data existing in the literature concerning the behavior of some of the MAb tested here.     

Peptide mimotopes of oncoproteins as therapeutic agents in breast cancer

Generation of an immune response to oncoproteins can lead to a cancer specific protective immunity. Several such oncoproteins are being examined as tumor targets with mixed results. We are evaluating the clinical utility of synthetic peptides that would mimic the antigen immunologically and elicit a tumor specific immune response. HER-2/neu, an oncoprotein whose expression in breast cancer is associated with poor prognosis, lower disease free-survival and a propensity for metastases was chosen as a model. Antibodies, Ab2, Ab4 and Ab5 directed towards the extracellular domain of HER-2/neu were reacted to peptides from two synthetic phage display peptide libraries, LX-8 (12-mer peptide library containing disulfide bridge) and X-15 (linear 15-mer). The isolated peptides were sequenced and characterized for ability to produce high titer antibodies and cross-reactivity. The peptides isolated did not show any sequence homology to protein databases but did show a hierarchy of immunogenic epitopes. Antibodies generated against peptides selected against the same antibody Ab2 or Ab4 showed affinity variation. Phages selected against Ab2 were also able to compete with binding of Ab2 to HER-2/neu. These results validate our hypothesis that synthetic peptides that mimic the antigenic epitope of oncoprotein can be generated and their clinical utility rests on devising a screening mechanism to identify peptides that can elicit an immune response directed to the oncoprotein and if possible its antigenic variants.     

Non-germ-line encoded residues are critical for effective antibody recognition of a poorly immunogenic neutralization epitope on glycoprotein B of human cytomegalovirus

The capability of the antibody (Ab) repertoire to mount a response to appropriate epitopes on infectious agents will strongly affect the ability of the immune system to provide protection against disease. Certain epitopes may be poor inducers of immunity but are nevertheless able to promote biologically important protection when recognized by the immune repertoire. We have investigated the recognition by Ab of one such poorly immunogenic target, antigenic domain 2 (AD-2) on human cytomegalovirus glycoprotein B. To date, only two well-characterized human monoclonal Ab reactive with this epitope are known. To define parameters important for establishment of a human paratope recognizing this epitope, we created variants of the variable genes utilized by one of these Ab and used phage display technology to select Ab fragments with retained antigen specificity. We show here that residues in the first complementarity determining region of both the heavy and the light chain are involved in determining the AD-2 specificity and, in addition, that key mutations in the germ-line sequence are required for effective interaction with this epitope. Thus, the products of the human germ-line IGHV3-30 and IGKV3-11 genes, the only V genes that have been demonstrated to participate in an AD-2 specific Ab response, do not have the intrinsic features required for high-affinity recognition of this epitope. We propose that the inability of the human germ-line gene-encoded Ab repertoire to directly recognize this and possibly other antigenic determinants results in their poor immunogenicity in vivo. This may favor responses to other epitopes, which have a high-affinity imprint in the human germ-line Ab repertoire.     

Biochemical and biological characteristics of epitopes on the E1 glycoprotein of western equine encephalitis virus

Antigenic determinants identified by monoclonal antibodies (Mabs) on the E1 glycoprotein of western equine encephalitis (WEE) virus have been characterized by their serological activity, requirements for secondary structure, expression on the mature virion, and their role in protecting animals from WEE virus challenge. On the basis of a cross-reactivity enzyme-linked immunosorbent assay (ELISA) and hemagglutination inhibition assay, eight antigenic determinants (epitopes) on the E1 glycoprotein have been identified, ranging in reactivity from WEE-specific to alphavirus group reactive. No neutralization of virus infectivity was demonstrable with any of the Mabs. An alphavirus group-reactive hemagglutination (HA) site, a WEE complex-reactive HA site, and a WEE virus-specific HA site were identified. Spatial arrangement of these epitopes was determined by a competitive binding ELISA. Four competition groups defining three distinct antigenic domains were identified. Antibodies directed against four E1 epitopes were capable of precipitating the E1/E2 heterodimer from infected cells or purified virus disrupted with nonionic detergents. These same antibodies precipitated only E1 in the presence of 0.1% SDS. That E1 conformation was important was shown by the inability of antibodies specific for seven of the epitopes to bind to virus denatured in 0.5% SDS. As determined by equilibrium gradient analysis of virus-antibody mixtures, four epitopes were found to be fully accessible on the mature virion, three epitopes were inaccessible, and one epitope was partially accessible to antibody binding. Antibodies specific for three epitopes were able to passively protect mice from WEE virus challenge.     

Immunogenicity of peptides for B cells is not impaired by overlapping T-cell epitope topology.

The epitope specificity of T-cell help to B cells and of surface immunoglobulin-mediated B-cell-binding of antigens usually involves topographically distinct antigenic determinants. The possibility of cross-recognition of the same peptide sequence by both T cells and antibodies has been a matter of conflicting opinions. We investigated this subject by detailed mapping of T- and B-cell epitopes within four immunogenic mycobacterial peptides. The identified core sequences of T- and B-cell epitopes showed different topology within each peptide: they were partially overlapping or adjacent in two P38-derived peptides, but entirely overlapping in two P19-derived peptides. The critically important result using the two truncated peptides (P19/67-78 and P19/146-155) containing only the fully overlapping epitope cores was, that they retained full potency for inducing antibody responses. However, despite this desirable overlap of determinants, antipeptide sera failed to block the proliferation of corresponding T-cell hybridomas. We conclude, that our study, in contrast to previous findings, suggests that overlapping topology of T- and B-cell epitopes within synthetic peptides does not necessarily impair B-cell immunogenicity.     

Fine mapping of a peptide sequence containing an antigenic site conserved among arenaviruses

The lymphocytic choriomeningitis virus (LCMV) structural glycoproteins GP-1 (Mr 44K) and GP-2 (Mr 35K) are encoded on a single intracellular proteolytic cleavage precursor glycoprotein, GP-C (Mr 76K). We have used a series of synthetic peptides derived from the deduced amino acid sequence of LCMV GP-C to define an antigenic site containing two topographically overlapping epitopes. Three mouse monoclonal antibodies directed against two epitopes on GP-2 were assayed for binding in solution phase blocking and solid-phase enzyme-linked immunoadsorbant assays to a series of peptides representing the sequence of the intracellular precursor glycopeptide GP-C. Both epitopes were initially localized to a single peptide GP-C 370-382 (Cys-Asn-Tyr-Ser-Lys-Phe-Trp-Tyr-Leu-Glu-His-Ala-Lys) in the GP-2 segment of GP-C. Further analysis demonstrated that both epitopes were contained within a nine amino acid segment, GP-C 370-378, which contains five residues conserved among LCMV, Lassa, Pichinde, and Tacaribe viruses. Assays with N-terminal deletions from this sequence suggested that the minimal epitope recognized by the broadly cross-reactive monoclonal 33.6 (epitope GP-2a) consisted of five amino acids, GP-C 374-378 (Lys-Phe-Trp-Tyr-Leu). Reactivity of a second monoclonal, 9-7.9 (epitope GP-2B) but not 33.6, was abolished when substitution of tyrosine for phenylalanine was made at position 375 in the antigenic sequence corresponding to a naturally occurring sequence difference between LCM and Lassa viruses. Polyclonal sera from human cases and from animals experimentally infected with Junin, LCM, and Lassa viruses, respectively, bound to the antigenic peptide GP-C 370-382 but not to control peptides. As was the case with the monoclonals, this binding activity was abrogated by blocking with the antigenic peptide but not with control peptides in solution.     

Topological mapping of murine leukemia virus proteins by competition-binding assays with monoclonal antibodies

Monoclonal antibodies produced by hybrid cells in culture are chemically homogeneous reagents that react with constant avidity to single antigenic determinants (epitopes). As such, these antibodies are remarkably specific probes for small regions of complex antigenic proteins. We have used a panel of monoclonal antibodies in competition binding assays to investigate the arrangement of six epitopes on the envelope proteins of AKR leukemia virus. It was reasoned that if two epitopes were adjacent to each other on a single protein, the binding of antibody to one epitope would sterically hinder the binding of antibody to the other epitope. On the other hand, if the two epitopes were at distant sites on the protein, the binding of antibody to one epitope would not influence the binding of antibody to the other epitope. The results of these competition binding assays demonstrated the presence of two distinct antigen sites on both the gp70 and p15(E) envelope proteins. With the gp70 protein, one antigen site contained the gp70^b and gp70^c epitopes; the other antigen site on this protein contained the gp70^a epitope. With p15(E), one of the antigen sites contained the p15(E)^b and p15(E)^c epitopes, while the other site contained the p15(E)^a epitope. These findings demonstrate the utility of this type of serological analysis for the study of the tertiary structure of individual viral proteins.     

Inhibition of T-cell responses by feeding peptides containing major and cryptic epitopes: studies with the Der p I allergen.

H-2b mice respond to the 222 residue allergen Der p I by producing T cells sensitized to the dominant epitopes encompassed in peptides 21-49, 78-100, 110-131 and 197-212. Immunization with the synthetic peptides 120-143 and 144-169, however, revealed cryptic epitopes which could sensitize T cells for responses to the respective peptides and, providing splenic adherent cells were added to lymph node cultures, to the whole allergen. It is shown that feeding recombinant fusion peptides can markedly inhibit the ability of the whole antigen to immunize mice, as measured by the in vitro interleukin-2 (IL-2) and granulocyte-macrophage colony-stimulating factor (GM-CSF)/IL-3 release on stimulation with protein or peptides, although inhibition measured by IL-2 release was more marked. The inhibition extended to epitopes other than those in the fusion peptides used for feeding. Thus feeding peptide 101-154 inhibited responses to 110-131 and 78-100. Fusion peptides 1-14 and 188-222 did not inhibit responses, although 188-222 did contain an epitope. Inhibition was also obtained when mice were fed a fusion containing the cryptic epitope 144-169. The ability of peptides containing the cryptic epitopes to inhibit responses has significant implications for peptide-based immunotherapy.     

Antigenic structure of human chorionic somatomammotropin: four epitopes revealed by monoclonal antibodies

The characterization of epitope specificity of a panel of 15 monoclonal antibodies against human chorionic somatomammotropin (hCS), previously prepared in our laboratory, allowed us to distinguish 4 antigenic determinants on the hCS molecule. Experiments were carried out with competition and sandwich assays. The results allowed us to divide our MAbs into 4 groups, recognizing 4 different epitopes, and to select 4 MAbs each distinguishing a different epitope in order to pursue further studies on the antigenic structure of hCS.     

Epitopes functional in neutralization of varicella-zoster virus.

By competition neutralization assay using monoclonal antibodies (MAbs) to varicella-zoster virus (VZV) glycoproteins (gps), we attempted to determine the topographical relationship of epitopes which are functional in VZV neutralization. MAbs against gpI interfered moderately to strongly with neutralization of MAbs against gpIII, and one antigenic domain with two distinct epitopes was identified on gpIII. Competition neutralization assays performed with MAbs to gpI revealed at least three distinct antigenic domains: the first contained two complement-dependent neutralizing epitopes; the second contained five complement-dependent neutralizing, overlapping epitopes and one nonneutralizing, nonoverlapping epitope; and the third contained one complement-enhanced neutralizing epitope. Competition neutralization assays performed with MAbs to gpIV showed one antigenic domain with two distinct epitopes which competed with nonneutralizing gpI MAbs. gpII did not interfere with neutralization of gpI, gpIII, or gpIV. Our data suggest that neutralizing and nonneutralizing MAbs can interfere with the action of viral neutralization either by inhibition or by enhancement. This report describes the epitope mapping of VZV gps by a functional biological assay.     

The recognition by monoclonal antibodies of various portions of a major antigenic site of human growth hormone

The reactivities of five mouse monoclonal antibodies against human growth hormone (hGH) were defined by either a competitive radioimmunoassay with insolubilized antibodies or by an agglutination-inhibition method with hGH-coated polystyrene particles. The five antibodies reacted significantly but to various degrees with human placental lactogen and at least three antibodies reacted with human prolactin and three synthetic peptides extending from residues 19 to 128, 73 to 128 and 98 and 128 of hGH. Four tested monoclonal antibodies failed to react with bovine growth hormone and with hGH oxidized by performic acid. The antibodies were further distinguished by their different reactions with hGH modified by reduction and alkylation or by adsorption on a polystyrene surface. The unique specificity of each antibody was confirmed for most of them by an agglutination method in which the agglutinating activity of hGH was tested on latex particles coated with various paired combinations of the monoclonal antibodies. The lack of agglutination with certain combinations suggested that the specificities of such a pair of antibodies overlapped each other. These results suggest that the sequences corresponding to the synthetic peptides participate in the structure of a major antigenic site of which various portions are recognized by the monoclonal antibodies.     

Determination of B-cell epitopes of nef HIV-I protein: immunogenicity related to their structure.

Determination of the B-cell epitopes of the nef molecule encoded by the human immunodeficiency virus type 1 (HIV-1) was undertaken using a set of six synthetic peptides. Sequences that were both antigenic and immunogenic and stimulated the production of antibodies recognizing the full length molecule, were considered as B-cell epitopes. Two peptidic sequences were antigenic both in rodents (mice and rats) and in non-human primates (chimpanzee). They were located in the regions 45-69 and 176-206 of the nef molecule. Two additional antigenic sequences were determined, one in chimpanzees, region 79-94, and the second in rodents, region 148-161. Immunogenicity was investigated in the rodents. Only the 45-69 and 176-206 sequences were immunogenic, and specific antibodies present in the sera of the immunized animals reacted with the nef protein. Therefore, each of these two sequences could be considered as containing at least one B-cell epitope. The fine epitopic specificity was determined using subfragments of these two sequences and it was shown that the antigenic determinants were contained in the C-terminal region of each sequence overlapping with the T-cell epitopes. These results raised the importance of vicinity of B- and T-cell determinants and their immunogenicity.