Elsevier

Molecular Immunology

Volume 35, Issue 13, 1 September 1998, Pages 865-879
Molecular Immunology

Antigenic and immunological mimicry of peptide mimotopes of Lewis carbohydrate antigens

https://doi.org/10.1016/S0161-5890(98)00067-4Get rights and content

Abstract

Peptides may substitute for carbohydrates in reactions with carbohydrate-specific molecules. Recently, we found that peptides containing aromatic residues mimic mucin and histo-blood group related carbohydrate epitopes, eliciting polyclonal responses cross-reactive with bacterial and viral antigens that express these carbohydrate forms. These results demonstrate that peptides can function in in vivo and in vitro models as carbohydrate surrogate antigens. To further explore the nature of the antigenic and immunogenic properties of such mimotopes, synthetic peptides with aromatic amino acids were tested to delineate reactivity patterns with several anti-neolactoseries monoclonal antibodies (MAbs). These MAbs recognize biologically important conformations of the histo-blood group related Lewis antigens expressed on the surface of a variety of human cancers. Results by ELISA demonstrate that the MAbs can distinguish particular peptide motifs that include the sequences GGIYYPYDIYYPYDIYYPYD, GGIYWRYDIYWRYDIYWRYD and GGIYYRYDIYYRYDIYYRYD. Substitution of Arg by Pro diminished the reactivity of the anti-Lewis Y (LeY) MAb BR55-2. Binding of LeY to BR55-2 was inhibitable by the Arg containing peptides. Serum against all three peptides displayed reactivity with synthetic histo-blood group related antigen probes. Immunologic presentation of the peptides as multiple antigen peptides (MAPs) improved peptide ability to induce LeY specific immune responses. Serum bound to human tumor cells that preferentially expressed neolactoseries antigens, but not to normal tissues. Immunoprecipitation of human breast tumor cell lysates before and after treatment with tunicamycin confirmed serum carbohydrate binding. The anti-peptide sera mediated tumor cell killing by complement mediated cytotoxicity. These results indicate that mapping peptide epitopes with anti-carbohydrate antibodies can lend to defining antibody fine specificities that can go undetected by screening of carbohydrate antigens alone. In addition, these results confirm that peptides and carbohydrates can bind to the same antibody binding site and that peptides can structurally mimic salient features of carbohydrate epitopes.

Introduction

The production of specific antibodies against polysaccharides play a pivotal role in the defense against gram-negative and gram-positive bacteria, and have shown promise in vaccine approaches in the immunotherapy of cancer (Livingston et al., 1997; Livingston and Ragupathi, 1997; Ravindranath et al., 1997a, Ravindranath et al., 1997b). However, carbohydrate antigens are typically poorly immunogenic, very difficult to purify in large quantities, difficult to synthesize and usually induce mostly short-lived IgM type antibodies in a vaccinated host without long lasting immunity. Most carbohydrate antigens belong to the category of T cell-independent antigens that reflect their inability to stimulate MHC class II-dependent T cell help (Mond et al., 1995). Consequently, carbohydrates alone are not capable of induction of a sufficient anamnestic or secondary immune response and require extensive adjuvanticity. An alternative approach for augmentation of carbohydrate immunity is the use of peptide or polypeptide surrogate antigens.

The idea of using surrogate antigens as immunogens, requires that antigenic mimicry, accomplished using amino acids in place of sugars, induces a precisely reproduced immune specificity pattern for the nominal antigen (Olsson, 1987; Diakun and Matta, 1989; Hutchins et al., 1996; Tsuyuoka et al., 1996). Mimicking peptides represent a new and very promising tool to overcome T cell-independence and to increase the efficiency of the immune response to carbohydrates (Shikhman and Cunningham, 1997). Peptides with a high prevalence of tryptophan and tyrosine occur in peptides that mimic carbohydrates (Oldenburg et al., 1992; Scott et al., 1992; Hoess et al., 1993; Shikhman et al., 1994; Shikhman and Cunningham, 1994; Westerink et al., 1995; Valadon et al., 1996; Harris et al., 1997; Zhang et al., 1997b; Agadjanyan et al., 1997; Phalipon et al., 1997; Taki et al., 1997). A more precise understanding of the binding properties of carbohydrate-mimicking peptides is required to determine whether the occurrence of aromatic containing motifs is due to molecular mimicry or simply reflects an advantage provided by aromatic rings for interaction between proteins (Valadon et al., 1996).

Peptides with the aromatic motif Wā§¹YXY have been previously defined to mimic several carbohydrate subunits that include YPY as a mimic of mannose as identified from peptide phage screening with Con A (Oldenburg et al., 1992; Scott et al., 1992) WRY found to mimic a (1ā€“4)glucose as identified from analysis of protein that bind to a-amylase (Murai et al., 1985; Mirkov et al., 1995), WLY found to mimic Lewis Y (LeY) as identified from peptide phage screening with an anti-LeY antibody, B3 (Hoess et al., 1993) and YRY derived from an anti-idiotypic antibody found to mimic the major C polysaccharide a (2ā€“9) sialic acid (MCP) of Neisseria meningitidis (Westerink et al., 1995). This aromatic motif is also implicated as a mimic for a Cryptococcus neoformans epitope (Valadon et al., 1996).

The sequence similarities that define this motif suggest that antibodies to homologous peptides might cross-react with similar subunits expressed on what are otherwise dissimilar carbohydrate structures; if so the basis of cross-reactivity would be structural. Mapping peptide epitopes containing the Wā§¹YXY motif with anti-carbohydrate antibodies could lend to defining fine specificities that might go undetected by screening carbohydrate antigens alone. Molecular modeling suggests that the neolactoseries LeY and sialyl-lex (sLeX) tetrasaccharide structure is similar to the core structure of MCP, providing a structural basis for potential cross-reactivity of antibodies (Agadjanyan et al., 1997). Consequently, evaluating responses to neolactoseries antigens would further define recognition specificities for antibodies directed to these important carbohydrate forms.

Here, we delineate specificity patterns associated with the antigenic and immunological mimicry of peptide mimotopes that induce antibodies to Lewis (Le) antigens. The blood group-related neolactoseries carbohydrate structures Lewis X (LeX), sLeX, Lewis a (Lea), sialyl-Lea (sLea) and LeY are examples of terminal carbohydrate structures related to tumor prognosis (Miyake et al., 1992; Dabelsteen, 1996). These antigens (Fig. 1) constitute carbohydrate moieties on some tumor associated gangliosides, the human carcinoembryonic antigen family, the human pancreatic MUC-1 antigen and identified on glycoproteins and glycolipids on carcinomas of the skin, stomach, pancreas, lung, colon, breast and prostate. SLeX and sLea, are implicated as immunogenic antigens in human melanoma as well (Ravindranath et al., 1997a). Histo-blood group related antigens are observed only at the secretory borders on normal tissues (Zhang et al., 1997a). This location appears to be inaccessible to the immune system, inducing neither tolerance nor autoimmune responses. Consequently, some Le antigens, notably LeY, are excellent targets for passive immunotherapy or a vaccine approach in the treatment of cancer.

Contrasting the antigenic and immunological properties of peptide mimotopes forms provides information about how amino acid differences lend to specific carbohydrate mimicry. We observe that the immunological presentation of Wā§¹YXY containing peptides influences the specificity pattern for histo-blood group synthetic carbohydrate probes. We also observe that the peptide mimotopes elicit polyclonal sera that specifically bind to human tumor cells, but not to normal tissues, and can mediate complement dependent cytotoxicity (CDC) of representative human breast cell lines, albeit at low titers.

Section snippets

Preparation of peptide immunogens

Several peptides were synthesized, repeating putative centralized motifs, suggested to mimic carbohydrate forms and correspond to GGIYYPYDIYYPYDIYYPYD (K61105), GGIYWRYDIYWRYDIYWRYD (K61106), GGIYYRYDIYYRYDIYYRYD (K61107), GGGAPWLYGAPWLYGAPWLY (K61223), GGAPWLYGGAPWLYAPWLY (K61108). Other peptides were synthesized as variants that include GGAGRWVFSAPGVRSIL (K6111), GGGWPYLRFSPWVSPLG (K61110), GGARVSFWRYSSFAPTY (K61109). Peptides were synthesized with the addition of a tripeptide YGG spacer and

Antigenic mimicry of peptide motifs

To determine the extent of antigenic mimicry of the Wā§¹YXY putative motifs in shown in Table 1, a variety of peptides were synthesized, some repeating the respective putative centralized motifs. MAP forms were synthesized for detection of reactivity patterns with the anti-LeY monoclonal antibodies BR55-2 and 15.6A and the anti-sialylated LeX reactive antibody FH-6 (Fig. 2a). These antibodies are very specific for their respective antigens. It is hypothesized that MAP forms may represent

Discussion

Immunization with simple synthetic Lewis antigen-conjugates, as observed with simple LeY-conjugates, only results in sera and MAbs, reactive with the immunizing antigen (Kitamura et al., 1994). In contrast, immunization with LeY expressing cell lines or multivalent LeY forms, yield MAbs that react with both synthetic carbohydrate forms and native carbohydrate configurations (Vlasova et al., 1994). One difference between these antigen sources is the distribution of LeY epitopes on the carrier ā€”

Acknowledgements

This work was supported by the USAMRMC (DAMD17-94-J-4310) Breast Cancer Program. Computer equipment support from the Cancer Center of the University of Pennsylvania is also gratefully acknowledged. We also thank Charlotte Read Kensil of Aquilia Pharmaceuticals (Worcester, MA) for supplying the QS-21.

References (54)

  • S.V. Evans et al.

    The crystallization

    (1994)
  • R. Hoess et al.

    Gene

    (1993)
  • W. Hutchins et al.

    Molecular Immunology

    (1996)
  • J. Lescar et al.

    Journal of Biological Chemistry

    (1995)
  • G.W. McLean et al.

    Journal of Immunological Methods

    (1992)
  • M. Rajadhyaksha et al.

    Vaccine

    (1995)
  • S. Saito et al.

    Journal of Biological Chemistry

    (1994)
  • S.M. Silver et al.

    Blood

    (1987)
  • T. Taki et al.

    Febs Letters

    (1997)
  • P. Valadon et al.

    Journal of Molecular Biology

    (1996)
  • A.C. Young et al.

    Journal of Molecular Biology

    (1997)
  • M. Agadjanyan et al.

    Nature Biotechnology

    (1997)
  • E. Dabelsteen

    Journal of Pathology

    (1996)
  • P.P. Deshpande et al.

    Nature

    (1997)
  • K.R. Diakun et al.

    III. Specificity analysis of an antiā€“anti-idiotypic antibody to a carbohydrate tumor-associated antigen. Journal of Immunology

    (1989)
  • K.R. Diakun et al.

    Immunological Investigations

    (1996)
  • C.P. Garin et al.

    Journal of Histochemistry and Cytochemistry

    (1989)
  • J. Garrigues et al.

    Journal of Cell Biology

    (1994)
  • S.L. Harris et al.

    Proceedings of the National Academy of Sciences of the United States of America

    (1997)
  • D. Iliopoulos et al.

    Journal of the National Cancer Institute

    (1989)
  • S. Kanda et al.

    Cell Biophysics

    (1994)
  • T. Kieber-Emmons et al.

    Hybridoma

    (1997)
  • K. Kitamura et al.

    Proceedings of the National Academy of Sciences of the United States of America

    (1994)
  • V. Kudryashov et al.

    Cancer Immunology

    (1998)
  • P.O. Livingston et al.

    Cancer Immunology

    (1997)
  • P.O. Livingston et al.

    Cancer Immunology

    (1997)
  • G.H. Lowell et al.

    Science

    (1988)
  • Cited by (41)

    • Varicella seroepidemiology in United States air force recruits: A retrospective cohort study comparing immunogenicity of varicella vaccination and natural infection

      2017, Vaccine
      Citation Excerpt :

      Previous studies have reported higher post-vaccination seroprevalence of anti-measles, mumps, rubella, and influenza antibodies among women compared to men [25,26], suggesting sex may modify the effect of seropositivity following vaccine administration. Additionally, HBGA expression has been hypothesized to play a role in viral and vaccine immunogenicity [27,28]. Age may also confound the relationship between exposure status and serologic immunity in the setting of variable vaccine availability, evolving vaccination recommendations, and decreasing exposure to wild-type VZV since the introduction of the varicella vaccine.

    • Toward a better understanding of the basis of the molecular mimicry of polysaccharide antigens by peptides: The example of Shigella flexneri 5a

      2006, Journal of Biological Chemistry
      Citation Excerpt :

      However, available data suggest that independently of the peptide mimic under study, all mIgA-peptide interactions derived mostly from the direct contact of peptide aromatic residues and methyl groups with the mAb-binding site, suggesting that recognition was basically driven by hydrophobic and van der Waals contacts. In that matter, data provided for the S. flexneri 5a system fully support previous observations made for other models implicating peptides mimicking polysaccharide antigens in complex with specific mAbs (11). In addition, all data reported here strongly emphasize the crucial role played by the type II Ī²-turn topology in governing molecular mimicry of S. flexneri 5a O-antigen.

    View all citing articles on Scopus
    View full text