IVD Technology Magazine | IVDT Article Index

Originally published March, 1997

Antiidiotypic antibodies as diagnostic antigens

Eileen Skaletsky

While their potential use in vaccines has dominated research attention, antiidiotypic antibodies also promise considerable benefits for diagnostics developers.

An antibody is usually defined in terms of the antigen it recognizes. An antibody's specificity for a particular antigen is determined by its antigen-binding site, the distinct region of the antibody molecule that makes contact with an antigen. This site is found within the variable (V) regions of immunoglobulin heavy (V^_H) and light (V^_L) chains.

However, an antibody may also be defined by its idiotype (Id), an ensemble of idiotopes, or surface markers, associated with the unique V^_H and V^_L regions of a monospecific population of antibody molecules. Idiotypes are useful markers because they enable researchers to follow the appearance and persistence of particular antibodies in immune responses and inherited immunoglobulin genes.¹ Ids are also unique determinants that can stimulate production of antiidiotype (anti-Id) antibodies. Of special interest to developers of diagnostic immunoassays is the ability of some anti-Ids to mimic antigenic determinants recognized by an original antibody and, thus, to serve as surrogate antigens.

History of Antiidiotypic Antibodies

The first studies to characterize idiotypic determinants on antibody molecules were performed by Oudin and Michel, and by Kunkel et al. in 1963.^2,3 These studies examined antisera generated against homogeneous immunoglobulins (myelomas and Bence Jones proteins) and revealed the antigenic individuality of immunoglobulin molecules. In 1974, Neils Jerne proposed a biological significance for idiotypic determinants in his network theory of immune regulation.⁴ Jerne theorized that an immune response to a given antigen is controlled by a series of Id­anti-Id reactions that may either enhance or suppress an immune response. The antibody made in response to the original antigen becomes itself an antigen and elicits the synthesis of a second, antiidiotypic antibody. Thus, a given Id (Ab1) is under control of an anti-Id (Ab2), and the anti-Id may be regulated by another set of antibodies referred to as anti­anti-Id (Ab3) (see Figure 1). According to Jerne, this complex set of interactions operates via a feedback mechanism to control immune responses. Jerne's network theory successfully explained the appearance of "antiantibodies" previously noted by many laboratories. As a result, new emphasis was placed on self-recognition as an essential component of immune regulation.

Figure 1. Idiotype-antiidiotype network.

Classes of Antiidiotypic Antibodies

A heterogeneous population of anti-Id is obtained when syngeneic inbred animals are immunized with Id. The parameters used to define each anti-Id focus primarily on whether it binds within the antigen-binding site or to some other region of the Id. If the relevant antigen or hapten inhibits the binding of anti-Id to Id, the target idiotope is believed to be in the antigen-binding site, and the anti-Id is referred to as Ab2ß. Anti-Ids that are not inhibited by antigen are designated Ab2* and presumably bind to sites in the framework region. In 1984 Bona and Köhler proposed a third type of anti-Id, Ab2*, which are inhibited by antigen because of steric interference (see Figure 2).⁵

Figure 2. Interactions of different Ab2 with Ab1.

Ab2ß that can displace an antigen from the antigen-binding site of an antibody may "look" like the antigen and are referred to as internal image anti-Id. It is therefore important to define the criteria for internal imagery. The first criterion is the ability to mimic antigen. An Ab2ß must be capable of inducing an antibody response of the same specificity as the antigen it mimics or to act like a natural ligand for a cellular receptor. This functional criterion also suggests that internal images can induce T cell­mediated functions via interactions with T cell receptors. The second criterion is the ability to function as antigen in a variety of animal species or inbred strains. This criterion is a means of distinguishing Ab2ß from Ab2* or Ab2*.

The crystallographic structure of an idiotype-antiidiotype complex. Idiotype Fab Ab1 is shown in pink (heavy chain) and red (light chain); antiidiotype Fab Ab2 is shown in light blue (heavy chain) and dark blue (light chain). The two Fabs interact by juxtaposition of their complementarity determining regions (CDRs): the heavy chain of one molecule interacts almost entirely with the heavy chain of the other; the light chains similarly interact with one another, but to a much lesser extent. Photo © 1994, Proc Natl Acad Sci USA.¹⁰

Chemistry of Antigen Mimicry

Internal imagery of anti-Ids was first reported when anti-Ids raised against antibodies to insulin reproduced certain physiologic actions of the hormone itself when bound to insulin receptors.⁶ Subsequently, anti-Id mimicry of bacterial, viral, parasitic, and tumor antigens proved successful in generating experimental anti-Id vaccines.^7,8 All evidence to date indicates that anti-Id mimicry of antigens is due to similarities in three-dimensional structural conformations and, in some cases, to regions of DNA-sequence homology between antigen and anti-Id. Crystallographic analysis of a lysozyme-Ab1 Fab complex and an Ab1-Ab2 Fab-Fab complex showed that the anti-Id antibody and the antigen completely overlapped in their binding to the Ab1.⁹ In another study, our colleagues at the University of California at Riverside studying the crystal structure of an Id­anti-Id Fab-Fab complex found that the interacting molecular surfaces of the two Fab fragments greatly resembled the structural features of a typi-cal antigen-antibody complex.¹⁰ In fact, the V_^H and V_^L complementarity determining region 1s (CDR1s) of the anti-Id showed sequence homology with regions of the antigen (the peplomer protein of feline infectious peritonitis virus), suggesting that these are the sequences and structures of the external antigen mimicked by the anti-Id. Sequence homology between anti-Id and antigen has been clearly demonstrated for reovirus type 3 hemagglutinin as well.¹¹

Whether an anti-Id can accurately mimic an antigen depends greatly on the latter's chemical nature. In the most common antigens--proteins, peptides, small haptens, and carbohydrates--the groups forming the major epitopes are chemically dissimilar and present different problems for mimicry by anti-Id.¹² Of these groups, protein antigens and peptides should in principle be the best suited for this purpose, because most of their possible conformations can be easily reproduced by one or more of the CDRs. Conversely, the chemical characteristics of carbohydrate antigens and some small haptens may be more difficult to duplicate with the CDRs of antibodies.

Benefits of Antiidiotypic Antibodies as Surrogate Antigens

Even though the notion of antigen mimicry by anti-Id has been confirmed by several laboratories for numerous diverse antigens, the utility of anti-Ids as surrogate antigens in diagnostic immunoassays has received little attention. Nevertheless, they are extremely well suited to this application for both competitive immunoassay formats and direct serologic assays for specific antibody. Anti-Ids are promising alternatives to the many antigens that are not appropriate for the large-scale production and/or purification required for components of commercial immunoassays (Table I).

Table I. Disadvantages associated with large-scale production and/or purification of conventional antigens.

To overcome the limitations of conventional antigens, internal image anti-Ids can be used in place of antigen in competitive immunoassays. Anti-Ids, like any antibody, can be safely produced and purified in large quantities, can have multiple sites for the attachment of label, can be labeled without substantially diminishing their stability or ability to compete with antigen, and can be bound to a solid support without appreciable loss of immunoreactivity. For many of the same reasons, anti-Ids are also well suited for use in noncompetitive serologic assays as direct replacements for native antigen in tests for the presence of specific antibody.

It is not unreasonable to ask whether autologous anti-Id that may be present in blood or serum samples would interfere in anti-Id-based immunoassays. Data from early experiments on autoantiidiotypic antibodies indicate that internal-image anti-Ids that represent the configuration of the original antigen are rarely found.^13,14 However, anti-Ids that participate in the regulation of immune responses do arise. These types of anti-Id are termed regulatory anti-Ids, and in general they are not internal images. Thus, it appears that internal-image anti-Ids do not represent a major portion of the immunoregulatory network. They are therefore not expected to be present in levels high enough to interfere with assay performance.

Producing Antiidiotypic Antibodies

To produce anti-Ids, it is important to know the characteristics of the anti-body that will be designated the Ab1. The epitope specificity of the Ab1, and the ultimate use of the anti-Id, will guide the later characterization of Ab2. Either monoclonal or polyclonal antibodies may be used as Ab1 preparations to induce anti-Id. If polyclonal Ab1 is used, it is preferable that it be purified on an antigen-affinity column and that antigen-binding activity be confirmed following elution from the column. This procedure will increase the chances of presenting the appropriate Ab1 and generating the desired anti-Id response in immunized animals.

For generating Ab2, a variety of immunization protocols have been described. Much depends on the species from which the Ab1 was derived and the species to be immunized. To produce Ab2 in a syngeneic animal (e.g., the Ab1 is a murine monoclonal antibody and mice of the same strain are to be immunized), we use Ab1 conjugated to keyhole limpet hemocyanin (KLH) plus adjuvant.¹⁵ Following a primary intraperitoneal immunization with 50 µg of Ab1-KLH, mice are allowed to rest for several weeks and are then given a series of biweekly booster injections. Sera are analyzed for Ab2 one to two weeks following each booster injection (see "Identifying Antiidiotypic Antibodies," below). If a monoclonal Ab1 is injected into an animal of a species other than that in which it was generated, conjugation to KLH is usually not required. The injection schedule remains the same.

Surface representations of the interacting complementarity determining regions (CDRs) of an idiotype Fab (A) and an antiidiotype Fab (B), showing the surfaces of atoms involved in van der Waals contacts (purple), hydrogen bond donors (blue), and hydrogen bond acceptors (red). To facilitate identification of hydrogen bonding contact points, seven groups are labeled on the idiotype (1­7) and the antiidiotype (1'­7'). Photos courtesy Alexander McPherson, University of California, Riverside.

The disadvantage of an interspecies immunization protocol is that antibodies are produced to isotypic and allotypic as well as idiotypic determinants. If a monoclonal Ab2 is to be developed, these problems can be addressed in the initial screen of hybridoma cultures. Antiisotype antibodies will react with an irrelevant immunoglobulin (Ig) preparation from the same species as well as with the Ab1, while an anti-Id will recognize only the Ab1. Antiallotype antibodies will react with the Ab1 and with Ig in a preimmune sample obtained from the Ab1 host animal before immunization; anti-Ids will react only with the Ab1.

If a polyclonal anti-Id is to be generated, and the Ab1 is from a different species, the process of identifying the desired anti-Id is complicated by the range of antiisotype and antiallotype specificities that will be generated. However, polyclonal antiidiotypic antiserum can be adsorbed repeatedly on normal immunoglobulin to remove these unwanted specificities, leaving only anti-Ids in the antiserum.

Overall, whenever possible, the simplest approach to producing anti-Ids is to inject monoclonal Ab1 into a host animal of the same species for the generation of monoclonal anti-Id. In addition to being simpler, monoclonal anti-Id reagents represent singular epitope specificities and, therefore, lend themselves to more controllable and reproducible assay systems. In addition, monoclonal anti-Ids can be produced consistently in unlimited quantities, are easily purified from cell culture media or ascitic fluid, and, in many cases, are less expensive to produce than purified antigens.

Identifying Antiidiotypic Antibodies

Anti-Ids can be detected with a direct-binding sandwich ELISA. Wells of an ELISA plate are coated with 100­500 ng of Ab1. After serum or hybridoma culture supernatant is incubated in the wells for 30­60 minutes at room temperature, the plate is washed thoroughly. Labeled Ab1 is then added for an additional 30­60 minutes, the plate is washed again, and substrate and chromogen of choice are added. This method is likely to select Ab2*, ß, and *, but it is a useful preliminary screen for eliminating Ab2-negative cultures from a large number of hybridomas undergoing their first fusion screen.

A competitive-binding ELISA can more precisely identify anti-Id, and either of two assay formats may be used. First, wells of an ELISA plate are coated with Ab1 as described above. Anti-Id-containing serum or hybridoma supernatant is mixed with the biotinylated antigen recognized by Ab1, and this mixture is then incubated in the coated wells for 30­60 minutes at room temperature. Anti-Id in the sample will compete with the antigen for binding to Ab1 on the wells. After the plate is thoroughly washed, avidin-HRP is added, followed by substrate and chromogen. If antigen-inhibitable anti-Id is present in the sample, the reaction signal will be less than that of a control reaction in which a mixture of biotinylated antigen and irrelevant antibody is added to Ab1-coated wells. In an alternative format, wells of an ELISA plate are coated with antigen, samples containing anti-Id are mixed with labeled Ab1, and this mixture is then added to antigen-coated wells. Then the plates are washed, and an appropriate substrate and chromogen are added.

Once antigen-inhibitable Ab2 are identified, it is essential to confirm internal imagery of the anti-Id. This confirmation is most easily accomplished by use of antisera to the antigen obtained from different species. An Ab2ß, if it represents the internal image of the antigen and exhibits serological mimicry, should bind to the Id of antibodies from other species. Conversely, the ability of anti-Id to induce an Ab1 response in multiple species may be assessed. The anti-Ids can be further characterized (for instance, for relative low- or high-affinity binding to Ab1) in subsequent immunoassays. Little has been published on Id­anti-Id binding affinities, but in one study dissociation constants in the range of 10^­9 to 10^­11 mol/L were reported.¹⁶

Glossary

allotype: A distinct antigenic form of a serum protein that results from allelic variations present on the immunoglobulin heavy chain constant region.

autologous: Refers to derivation from self.

complementarity determining region (CDR): The hypervariable regions of an antibody molecule that form a three-dimensional cavity where an epitope binds to the antibody.

epitope: An antigenic determinant; the smallest structural area on a complex antigen molecule that can combine with an antibody.

framework regions: Amino acid sequences in variable regions of heavy or light immunoglobulin chains other than hypervariable sequences. Framework regions contribute to the secondary and tertiary structure of the variable region domain.

idiotype: The segment of an antibody molecule that determines its specificity for antigen. The idiotype is located in the Fab region, and its expression usually requires participation of the variable regions of both heavy and light chains that form the antigen-combining site.

isotype: Antigens that determine the class or subclass of heavy chains or the type and subtype of light chains of immunoglobulin molecules. For example, the four isotypes of IgG are designated IgG1, IgG2, IgG3, and IgG4.

syngeneic: Describes genetic identity between identical twins in humans or among members of an inbred strain of a species.

Antiidiotypic Antibodies and Diagnosis of Human Disease

In principle, anti-Id-based competitive immunoassays can be used to detect virtually any antigen present in a complex mixture of molecules. Such assays may be particularly useful for detection of antigens that are minor components of biological fluids or cell membranes. Clinical analytes for which this concept has already been demonstrated include human IgE and alpha-fetoprotein (AFP).¹⁷ By use of anti-Id-based competitive immuno-assays, IgE was determined in a range of 10­1000 IU/ml and AFP was determined in a range of 3­300 IU/ml. Coefficients of variation were 2.0­4.2% and 3.9­6.7%, respectively. In another case, adenosine deaminase­binding protein (ABP), a prognostic indicator of kidney damage, was successfully measured in urine with an anti-Id-based competitive immunoassay (see box, p. 30).¹⁸ This assay was capable of detecting a minimum of 0.07 units of ABP with a correlation coefficient of 0.92.

Perhaps one of the most promising applications of anti-Ids is in the diagnosis of antibody-mediated autoimmune disease. The Ids that are carried by autoantibodies and their possible association with disease are under active investigation. These idiotypes are ideal targets for anti-Id-based detection. Elevated expression of specific Ids has been demonstrated in myasthenia gravis, Hashimoto's thyroiditis, rheumatoid arthritis, and systemic lupus erythematosus (Table II).^19­22

Table II. Idiotypes frequently reported in the major autoimmune diseases.VirusesBacteriaProtozoa

An anti-Id-based assay has been described for autoantibody to myelin-associated glycoprotein (MAG).²³ In this study, IgM antibodies to MAG were first identified in a population of patients with a demyelinating neuropathy. Clinical observations suggested a direct pathogenic connection between this IgM antibody and associated peripheral neuropathy. This IgM antibody was ultimately used to generate a corresponding anti-Id, and this anti-Id reacted only with IgM specific for MAG (Figure 3).

Figure 3. Inhibition of the binding of ¹²⁵I-IgM from patient RC to mouse monoclonal antiidiotype antibody on ELISA plates. ¹²⁵I-IgM was incubated in various dilutions of serum from patient RC (+) or from eight controls (*). The points shown represent the mean value for the eight controls; the vertical bars indicate standard deviations.

Another potential application for anti-Ids is as surrogate antigens in the diagnosis of human infectious diseases. Anti-Id mimicry of native antigens has been successfully demonstrated for numerous infectious disease agents (Table III), and these types of anti-Id may be used as direct replacements for antigen in serologic assays or in competitive immunoassay formats. Despite the apparent utility of anti-Id antibodies as surrogate antigens, at present there are no commercial anti-Id-based diagnostic products. A likely explanation for this is that the companies skilled in developing anti-Id antibodies have focused their attention on the therapeutic applications of anti-Ids, for which commercial opportunities are substantially greater. For example, IDEC Pharmaceuticals Corp. (San Diego) has successfully developed a murine monoclonal anti-Id for treatment of B cell lymphoma, and is in the early stages of developing an anti-Id treatment for malignant melanoma.

Table III. Infectious disease agents for which anti-Id mimicry has been demonstrated.

Of course, many successful immunoassays for infectious diseases have been available for years, and there is no reason to change these existing formats to anti-Id-based assays. However, anti-Id surrogate antigens may be considered for infectious disease immunoassays currently in development, such as those for Borrelia burgdorferi, human parvovirus, or Hanta virus, or for new targets as they arise.

Id­anti-Id competitive immunoassay for adenosine deaminase­binding protein (ABP) in urine

In 1985, Thompson, Hewitt, Piper, et al. published the results of a study to determine whether an anti-ID-based competitive immunoassay could be used to detect adenosine deaminase-binding protein, a prognostic indicator of kidney damage.¹⁸ Following is the recipe used by the authors to create their anti-Id assay, together with a summary of the study results. The method used by the researchers is instructive for diagnostics manufacturers that might consider creating anti-Id-based assays for other clinical analytes or disease indicators.

1.Mix 200 µl of biotinylated anti-Id with 200 µl of each patient's urine sample, assay standards, and assay controls.

2.Add 100 µl of this mixture to Id-coated ELISA plates in triplicate.

3.Incubate for 3 hours or overnight at 25°C.

4.Wash the plate three times with Tween that contains phosphate-buffered saline.

5.Add 100 µl of avidin-HRP to each well and incubate for 1 hour at 25°C.

6.Aspirate and wash three times with Tween that contains phosphate-buffered saline.

7.Add 100 µl of chromogen solution: per L, 2 g OPD and 500 µl of 300-ml/LH^₂O^₂ solution, in citrate-phosphate buffer.

8.After 1 hour at 25°C, stop the reaction by adding 100 µl of 4.5-mol/L sulfuric acid.

9.Measure absorbance at 490 nm. Determine ABP in samples of patients' urine by comparison with standard curve (see table below).

Competitive Id­anti-Id standard curve. Data are expressed as a percentage of B/B^₀ where B represents the amount of binding of the anti-Id to the Id in the presence of different concentrations of ABP, and B^₀ represents the maximum binding of the anti-Id to the Id in the absence of ABP.

Urine samples obtained from persons with and without evidence of renal disease were tested in the competitive assay. Values shown are the mean levels of ABP obtained for each group. Source: Langone JJ (ed).⁷

Conclusion

Immunoassays that use anti-Ids as surrogate antigens can have the sensitivity and reproducibility we have come to expect from traditional assay formats. Because the methods for producing and purifying monoclonal antibodies are readily available in most laboratories, the generation of exquisitely specific monoclonal anti-Ids is, in many cases, easier, safer, and more economical than the production of bacterial or viral antigens. In addition, labeling chemistry or immobilization on solid supports is considerably more straightforward for antibody reagents than for many small haptenic antigens. Overall, the technical and commercial advantages of anti-Ids make them excellent alternatives to many conventional antigens for diagnostic immunoassays.

References

1. Greene MI, and Nisonoff A (eds), The Biology of Idiotypes, New York, Plenum, 1984.

2. Oudin J, and Michel M, "Une nouvelle forme d'allotypie des globulins * du serum de lapin, apparement liée a la fonction et a la spécificité anticorps," CR Acad Sci, 257:805­808, 1963.

3. Kunkel HG, Mannik M, and William RC, "Individual Antigenic Specificities of Isolated Antibodies," Science, 140:1218­1219, 1963.

4. Jerne NK, "Toward a Network Theory of the Immune System," Ann Immunol (Inst Pasteur), 125C:373­389, 1974.

5. Bona C, and Köhler G, "Anti-Idiotype Antibodies and Internal Images," in Monoclonal and Anti-Idiotypic Antibodies: Probes for Receptor Structure and Function, Venter JC, Fraser CM, and Lindstrom J (eds), New York, Alan R. Liss, pp 141­149, 1984.

6. Sege K, and Peterson PA, "Use of Anti-Idiotypic Antibodies as Cell Surface Receptor Probes," Proc Natl Acad Sci USA, 75: 2443­2447, 1978.

7. Langone JJ, Methods in Enzymology, Volume 178, Antibodies, Antigens, and Molecular Mimicry, San Diego, Academic Press, 1989.

8. Kennedy RC, and Dreesman GR, "Anti-Idiotypic Antibodies as Idiotope Vaccines that Induce Immunity against Infectious Agents," Intern Rev Immunol, 1:67­78, 1986.

9. Bentley GA, Boulot G, Tiottot MM, et al., "Three-Dimensional Structure of an Idiotope-Anti-Idiotope Complex," Nature, 348: 254­257, 1990.

10. Ban N, Escobar C, Garcia R, et al., "Crystal Structure of an Idiotype-Anti-Idiotype Fab Complex," Proc Natl Acad Sci USA, 91:1604­1608, 1994.

11. Williams WV, Guy HR, Weiner DB, et al., "Three-Dimensional Structure of a Functional Internal Image," Viral Immunol, 2:239­246, 1989.

12. Amzel LM, Garcia KC, and Desiderio S, "Do Anti-Idiotypic Antibodies Mimic Antigen?" Research Immunol, 145:53­55, 1994.

13. Kluskens L, and Köhler H, "Regulation of Immune Response by Autogenous Antibody against Receptor," Proc Natl Acad Sci USA, 71:5083, 1974.

14. Huang JY, Ward RE, and Köhler H, "Biological Mimicry of Antigenic Stimulation: Analysis of the in Vivo Antibody Responses Induced by Monoclonal Anti-Idiotypic Antibodies," Immunol, 63: 1­8, 1988.

15. Coligan JE, Kruisbeck AM, Margulies DH, et al. (eds), Current Protocols in Immunology, 2:9.4.1­9.4.5, New York, John Wiley, 1991.

16. Potocnjak P, Zavala F, Nussenzweig R, et al., "Inhibition of Idiotype-Anti-Idiotype Interaction for Detection of a Parasite Antigen: A New Immunoassay," Science, 215:1637­1639, 1982.

17. Walter G, Friesen H-J, Harthus H-P, "Anti-Idiotypic Antibodies: Powerful Tools in Diagnosis and Therapy," Behring Inst Mitt, 82:182­ 192, 1988.

18. Thompson RE, Hewitt CR, Piper DJ, et al., "Competitive Idiotype-Anti-Idiotype Immunoassay for Adenosine Deaminase Binding Protein in Urine," Clin Chem, 31:1833­1837, 1985.

19. Lang B, Roberts AJ, Vincent A, et al., "Anti-Acetylcholine Receptor Idiotypes in Myasthenia Gravis Analysed by Rabbit Antisera," Clin Exp Immunol, 60:637­644, 1985.

20. Delves PJ, and Roitt IM, "Idiotypic Determinants on Human Thyroglobulin Autoantibodies Derived from the Serum of Hashimoto's Patients and EB Virus Transformed Cell Lines," Clin Exp Immunol, 57:33­40, 1984.

21. Agnello V, and Barnes JL, "Human Rheumatoid Factor Cross-Idiotypes IWA and BLA are Heat-Labile Conformational Antigens Requiring both Heavy and Light Chains," J Exp Med, 164:1809­ 1814, 1986.

22. Isenberg DA, Williams W, Axford J, et al., "Comparison of Anti-DNA Antibody Idiotypes in Human Sera," J Autoimmunity, 3:393­414, 1990.

23. Page N, Murray N, Perruisseau G, et al., "A Monoclonal Anti-Idiotypic Antibody against a Human Monoclonal IgM with Specificity for Myelin-Associated Glycoprotein," J Immunol, 134:3094­3098, 1985.

24. Williams WM, and Isenberg DA, "Idiotypes and Autologous Anti-idiotypes in Human Autoimmune Disease--Some Theoretical and Practical Observations, Autoimmunity, 17:343­352, 1994.

Eileen Skaletsky, PhD, is vice president and general manager at QED Bioscience, Inc. (San Diego).