Copyright, Jack Brown, Department of Microbiology, University of Kansas, 1996

Back to Lecture 5

Lecture No. 6 - Antigens - Immunogenicity - Epitopes

We continued our discussion of, 1. Antigens, and:
2. What makes something immunogenic? - able to induce an immune response?

A. "Foreignness" - Relative Degree of Difference from Self Structures

B. Physical Size

C. Chemical Composition - Overall Structure

D. Degradability - Processability

E. Genetics of the Recipient

F. Dosage - Route of Immunization

E. Genetics of the Animal
Again, we can apply some common-sense reasoning here. We know that the B-cell antigen receptor (antibody/immunoglobulin) and the T-cell antigen receptor (TCR of both T-helper and T-cytotoxic cells) repetoire for an individual is encoded within inherited genes. Too, although the available MHC ClassI/II gene repetoire within any one individual is much smaller than the antigen-receptor repetoire, both MHC Class I and II molecules are inherited as well. Consequently, both the ability to specifically bind certain antigenic structures, as well as the ability to present these structures among the lymphocyte population within any given individual, is essentially unique to each individual. Therefore, different individuals among an outbred population of individuals can yield different humoral and/or cell-mediated immune responses to the same antigen. And sometimes, there will be "holes" in these repetoire, e.g., an inability to respond against certain antigenic structures, because of an inability to present a particular peptide, or perhaps because of an absence of available cells bearing receptors which bind the presented structure. Many of these observations have been made using in-bred strains of mice. Rodents apparently can withstand a pretty high amount of in-breeding. Therefore, it is possible to generate strains of mice which are genetically identical at the MHC level - the locus where MHC Class I and MHC Class II molecules are located. These mice produce progeny which are genetically identical to one another and which therefore give identical immune responses to a given immunogen. Through examination of these different in-bred strains, it has been shown that certain strains of mice give a response to certain structures, while a different in-bred strain yields no response whatsoever to the same structure.

F. Dosage and Route of Immunization
As it turns out, there is really no way yet to accurately predict whether or not a substance will be immunogenic, or to accurately predict either the level or predominant path (humoral versus cell-mediated) of the immune response if a substance is in fact immunogenic. Most of the information we have gathered has been the result of empirical (experimental) observation. There are some basic things, however that are known. First, if one immunizes an animal with very, very low amounts of material or very, very high amounts of material (for certain, relative terminology here), the immune response will be very poor. In the first case of low-dosage, common sense again tells us that if the concentration of material is too low to trigger endocytosis or phagocytosis - or the amount internalized is very small, then statistically there will not be a very high probability for subsequent interaction with receptors. However, at both the low-dosage and high-dosage ends of the spectrum - we have an interesting situation which may occur - in "old" terminology - low-zone and high-zone tolerance, respectively. When lymphocytes receive many, many very tiny dosages of something which is normally immunogenic, or receive too much (whatever "too much" may mean to the lymphocyte) antigen, the cells can become refractory to the stimulus - can in fact become non-responsive. This non-responsiveness can remain for a time, even if the immunogen is subsequently made available at an "appropriate" (dependent upon the particular substance) amount.

The route of immunization is also important and depending upon the substance, different routes, e.g., intramuscular, intravenous, subcutaneous, etc., may lead to different levels of immune response to the same immunogen. These factors must somehow relate to the half-life, processing time etc., which is best for an immune response for that particular substance. If the substance is in particulate form (either a whole cell, or tiny particles in suspension, but large molecular complexes) these forms of the immunogen tend to be more immunogenic - probably because of better trapping and phagocytosis potential. Further, there are substances known to increase the immunogenicity of other substances. These materials are called adjuvants. Some adjuvants increase the probability of lymphokine production (inflammatory signals), while others increase the longevity of the substance within the tissues, and still others increase the amount of particulate form of the imunogen. The recently- generated vaccine made for the Hemophilus influenzae bacterium is such a case. This vaccine is made by covalently complexing a H. influenzae bacterial polysaccharide to an immunogenic protein (a "hapten-carrier conjugate") - and - the preparation is made to be rather particulate in form. Let's look at the nature of the antigen/immunogen characteristics which are actually "seen" by lymphocyte antigen receptors.

3. Antigen Structures Reactive with B- and T-cell antigen Receptors
Both B- and T-cell receptors for antigen, e.g., antibody molecules and TCR's respectively, react with structures called epitopes or antigenic determinants. An epitope is a discrete region on or within an antigen molecule - a particular molecular shape expressed at a certain region of the molecule. The antibody molecules within the B-cell's plasma membrane can react with epitopes on soluble, independent antigen molecules, and therefore a circulating antibody and/or a B-cell expressing antibody can react with soluble epitopes, epitopes in suspension or - epitopes on whole cells. However, the TCR of both T-helper and T-cytotoxic cells can react with antigen only if the epitope is presented by either an MHC Class II or MHC Class I molecule (on an antigen-presenting cell), respectively. In this way then, T-cells are said to be restricted in their ability to recognize and bind to antigenic epitopes - either Class II or Class I restricted. In addition, within the same animal, a given T-cell is also restricted to recognition of antigenic epitopes presented by a particular Class II or a particular Class I molecule - depending upon the genetic repetoire of these MHC molecules.

3A. Summary Table of B and T cell-reactive Epitopes

Process                    B-Cell                     T-Cell

Form of Complex            Binary- Ab:Ag         Ternary- TCR:Ag:MHC-I/II

Bind Soluble Ag?               Yes                        No

MHC Involved?                  No                         Yes

Types of Antigen           Protein, Lipid             Protein Only
Recognized                 Polysaccharide

Epitopes Recognized        Accessible, Mobile         Denatured
                           Hydrophilic                Linear (sequential) 
                           Often Conformational       Amphipathic
                           Not Often Sequential       Bound to MHC I/II
     ______________________________________________________________________

The binding sites of antibody embedded within the plasma membrane of the B-cell can non-covalently associate with an epitope expressed on an antigen, and will form a binary complex. This reaction will occur even if the antigen is completely soluble and independent of all other molecules, or in fact whether or not the epitope itself is independently available. Epitopes which are present on proteins, carbohydrates and lipids are candidates for binding, even if the epitope is a small molecule such as a nitrophenyl ring. The types of epitopes usually recognized are accessible - that is, these epitopes are not usually hidden within the folds of the protein, but are available at the surface of the antigen. Consequently, in the natural aqueous environment such epitopes tend to be more hydrophilic than hydrophobic. In addition, the epitopes tend to be mobile - that is, the "better" epitopes are discrete antigen regions which exhibit a fair amount of molecular movement (the statistical probability of an epitope to assume an appropriate site-fitting shape increases if the region is moving rather than static). Further, individual epitopes tend to be conformational rather than sequential - although, sequential epitopes are also reactive. A sequential epitope is one which represents a linear peptide segment of amino-acid or carbohydrate sequence. A conformational epitope is one which represents a discrete folded region of polypeptide or polysaccharide which forms due to the secondary and/or tertiary folding characteristics of the protein or polysaccharide. There is another kind of epitope called a combinatorial epitope which reflects protein subunit:sububnit interaction - epitopes can form on the individual subunits and at the site of subunit interaction which are unique to the complex. We should examine some experimental results which demonstrate the importance of epitope conformational structure to antibody binding site interaction.

Experimentors have utilized information available from studies of hen egg-white lysozyme (HEL) and antibody reactivity with this enzyme. Basically, investigators wished to determine whether antibody generated against a known conformational region (conformational epitope) of this enzyme, could still react with this region if the region was somehow converted to a sequential epitope. To examine this issue, the region itself was used. The enzyme has a region called the HEL loop. This loop consists of 17 amino-acids and is formed by disulfide bonds between cysteine residues 64 and 80 of the HEL sequence. Antibody generated against this loop region were subsequently tested in inhibition assays for reactivity with HEL, the loop region isolated from HEL, synthetically-prepared closed loop, and open loop (reduced and blocked cysteine residues to linearize the structure). The results of the inhibition assays showed that the natural loop, HEL, and the closed synthetic loop were all good inhibitors of antibody binding to the natural loop; however, even when present in high concentration, the open synthetic loop failed to inhibit antibody reactivity. Consequently, these results demonstrate that if antibody is generated against a conformational epitope, any alteration of the structure of this epitope can significantly affect antibody binding-site interaction. Thus, if the conformation of a determinant/epitope is different than that against which the antibody responded, the antibody will no longer bind the altered epitope structure. This issue is not a small issue - because of the mutational alterations of virus capsid or envelope proteins, virus may escape immune activity by relatively few amino-acid substitutions within an immunodominant region.

Now let's take a look at the T-cell and Ag interactions:

Unlike antibody, the T Cell Receptor (TCR) binds extremely poorly to soluble antigen, but reacts very efficiently with peptides (protein) presented in the context of MHC Class I or II molecules on an antigen-presenting cell. The epitopes most commonly reactive with TCR's are denatured, sequential, and are usually derived from within the antigen molecule rather than the surface. By examination of what is known as a relative protrusion index for various TCR-reactive epitopes - one finds the following: if the sequence of a protein is examined, and regions of the protein are described as relatively more likely to protrude into the aqueous solvent (hydrophilic regions) - one can generate a relative protrusion index for the entire sequence of the protein. If one subsequently examines those sequential epitopes reactive with T-cells, one finds that it is always the minimally protruding epitopes which react. Too, TCR-reactive epitopes are normally amphipathic (have both hydrophobic and hydrophilic characteristics). This last characteristic is most easily explained by the requirements for epitope association with MHC molecules. The MHC Class I and MHC Class II proteins have very similar structures - each has anti-parallel ß-pleated sheets running back and forth which form a hydrophobic "floor", and helical regions on each side which form the "walls" of the site. The hydrophilic contact regions of the peptide associate predominantly with the walls, and the hydrophobic regions associate predominantly with the floor of the MHC molecule.

Consequently, without these particular contacts (amphipathic characteristics), other antigen structures such as lipids or carbohydrates, or certain relatively more hydrophilic and/or conformational epitopes just simply do not associate with these MHC molecules. Thus, the TCR is restricted to recognition of peptides - and only if the peptide is bound by one of these MHC molecules. It is also possible to now better understand why MHC molecules encoded by different individuals would or would not show specificity for certain peptides - and - why certain TCR sites would or would not interact with a given MHC molecule presenting peptide. In the next lecture we will briefly examine some experimental data upon which these statements are based.