Mapping T cell epitopes by flow cytometry

Abstract

Epitope mapping by flow cytometry is a very modern approach that not only identifies T-cell epitopes but simultaneously allows for detailed analysis of the responding T-cell subsets including lineage, activation marker expression, and other markers of interest. The most frequently used approach is based on the identification of intracellular cytokines in secretion-inhibited activated T cells following stimulation with peptides or peptide pools. A more recently developed assay analyzes T-cell proliferation by measuring the decrease in carboxyfluorescein diacetate succinimidyl ester staining in proliferated cells. This article includes information on peptide configuration, a section on the design and efficient application of peptide pools, and working laboratory protocols for both assays.

Introduction

Epitopes are the portions of antigens that bind specifically with the binding site of an antibody or a receptor on a T lymphocyte. While antibodies (B-cell receptors) bind free spatial epitopes, T-cell receptors require their epitopes to be presented on major histocompatibility complex (MHC) molecules in a linear fashion. The T-cell receptor makes contact with portions of the peptide and portions of the MHC molecule so that, strictly speaking, a T-cell epitope comprises both the peptide and determinants of the presenting MHC molecule. A description of the details of antigen presentation by MHC molecules is clearly beyond the scope of this article but can be found elsewhere [1]. Although a very large number of protein sequences are known and stored in databases, for most proteins it is not known whether they are immunogenic (i.e., whether they bind one or more MHC alleles within a population and are recognizable by the T-cell receptor repertoire), and where the antigenic determinants are localized within the primary amino acid sequence.

The identification of immunogenic proteins and determination of the T-cell epitopes contained in them often stand at the very beginning of vaccine development. While many vaccines use “bulk approaches” like whole recombinant proteins (hepatitis B), lysates of infectious agents (Bacillus Calmette–Guerin), complete toxoids (tetanus), there is increasing interest in more individualized peptide-based vaccinations. Possible applications include (but are not limited to) infectious diseases, tumor vaccines, and autoimmunity [2], [3], [4], [5]. Whether or not a vaccination is peptide based in its final form, the exact knowledge of relevant epitopes is helpful in studying how a vaccine is applied most efficiently, since known epitopes may be packaged with adjuvants, applied within a longer stretch of amino acids or protein, coded by DNA using free plasmids or various vectors or even modified by amino acid substitutions to generate more immunogenic molecules. Peptide-based vaccinations may induce or modulate both CD4 and CD8 T-cell responses; however, depending on the form of the vaccine (e.g., peptide or protein plus adjuvant or DNA) CD4 or CD8 T cells may be the preferred responding population. The use of defined epitopes in vaccines allows determination of which T-cell subset (helper or cytotoxic/suppressor) is going to respond in most cases, the only exception being peptides that induce both CD4 and CD8 T-cell responses. A quite recent application of epitope identification is in the quantification of epitope-specific T-cell populations by tetrameric MHC complexes (short: “tetramers”) [6]. Such tetramers are composed of four recombinant MHC molecules, each of which is coupled to a biotin molecule and linked by binding to the four biotin binding sites of a fluorochrome-conjugated streptavidin molecule. The MHC molecules all contain the same peptide in their binding groove such that the tetrameric complexes can directly bind to T-cell receptors that recognize the peptide–MHC molecule combination. Such reagents may be used to directly examine T-cell responses at an antigen-specific level. Peptides alone, on the other hand, can be used for effectively stimulating T-cell responses in peripheral blood mononuclear cells (PBMC) or other cell suspensions. A very interesting application is the use of such peptides as “diagnostic stimulant,” i.e., to find out if infection with a certain agent has taken place (and has induced T cells) or not [7]. Whether using tetramers or peptides as a stimulant, T cell responses can be identified, analyzed, and monitored. Both approaches are widely used for monitoring T-cell responses in patients with human immunodeficiency virus (HIV), Epstein–Barr virus (EBV), or cytomegalovirus (CMV) infections.

This article is dedicated exclusively to the mapping of T-cell epitopes by flow cytometry [8]. The method described is limited to epitopes within known protein sequences, since the stimulating agents for this approach are peptides that were synthesized according to a given primary protein sequence. The attraction of using a flow cytometer for mapping epitopes lies in the great versatility of this approach with respect to the number of parameters that may be acquired in a single measurement, i.e., cell lineage markers, activation markers on the cell surface and inside the cell [8]. The use of many parameters greatly enhances the resolution of the assay, i.e., identification of the T cell subset that responds to a given epitope including lineage and activation markers [9]. The approach we originally designed is based on the detection of rapid cytokine production in activated T cells in a short-term (as short as 6 h) ex vivo assay [10]. The biggest advantage of this approach apart from its unrivaled speed is that CD4 and CD8 T-cell responses can be measured in the same assay and even the same test tube [11]. However, proliferation may also be used as a readout parameter when using flow cytometry. Among the systems for measuring cell proliferation by flow cytometry that are available on the market, we favor the use of carboxyfluorescein diacetate succinimidyl ester (CFDA-SE) staining, because it is simple and reliable. The staining intensity of this protein dye is lost in regular increments as the intracellular protein content is passed on to the next generation by dividing cells. CFDA-SE is a nontoxic, fluorescein-related dye, which is able to permeate cells with the help of its two acetate side chains. Once inside cells, the acetate groups are removed by intracellular esterases and the resultant carboxyfluorescein exits at a much slower rate, allowing covalent binding to free amines of cytoplasmatic proteins, forming very stable amino bonds. Because CFSE also reacts with molecules that are rapidly degraded or transported out of the cell, a lot of the CFDA-SE that is initially taken up by cells and transformed into CFSE is lost during the first 24 h following labeling. The remaining portion that is bound to more stable intracellular components may be used for tracking cell proliferation. The use of CFDA-SE staining for the analysis of cell proliferation by flow cytometry was described in 2000 [12]. However, we have only recently started using it for the purpose of epitope identification and we therefore describe only our preliminary protocol. As regards proliferation assays, in general one fundamental difference between them and the detection of intracellular cytokines following short-term stimulation is that the latter does not require clonal expansion of responding cells for detection. On the other hand, such clonal expansion can amplify and thus enable detection of low-frequency responses that would otherwise be below the detection threshold of the cytokine-based assay. Our experience to date suggests that the CFDA-SE proliferation assay and the rapid cytokine induction assay give similar results in most donors; however, certain epitopes were identified only by one or the other approach (unpublished results).

Because the detection of intracellular cytokines in secretion-inhibited cells is a standard method, we do not discuss the basic principles of this technology, which may be found elsewhere [13]. Instead, we focus our discussion on the design of the peptides and peptide pools that may be used for epitope mapping based on intracellular cytokine detection. There are several key issues to be discussed with respect to the peptides including length, overlap between consecutive peptides, and protective groups. These topics are more or less relevant depending on whether CD4 or CD8 epitopes, or both, are defined. Another important point is the intelligent design of peptide pools. When protein sequences are long, the number of overlapping peptides often exceeds the number of single assays that can be run due to a lack of material, which is generally a preparation of peripheral blood cells. As a result, more than one peptide has to be tested per tube. We discuss two intelligent and efficient setups for such pools that increase efficiency by saving material, work, and time. All the issues we discuss in detail apply primarily to the use of intracellular cytokine detection as a readout. To what degree they also apply to the proliferation assay using CFDA-SE has not yet been examined.