All About Multiple Sclerosis

More MS news articles for December 2002

Hunting (auto)immune T cells in neuroimmunological diseases

Brain, Vol. 126, No. 1, 2-4, January 2003
Reinhard Hohlfeld
Department of Neuroimmunology Max–Planck Institute for Neurobiology Martinsried, Germany

In this issue of Brain, a paper by Muraro et al. (2003) takes an important step toward improving our ability to detect antigen-specific T cells. To understand better the implications of this advance, it may be helpful to recapitulate some principles of the role of antigen-specific cells in the immune system, focusing in particular on T cell biology. There are two types of antigen-specific immune responses: one is mediated by soluble antibodies secreted by B cells and the other by T lymphocytes. Various established methods of measuring antibodies [e.g. enzyme-linked immunosorbent assays (ELISA), radioimmunoprecipitation assays, western blotting, immunocytochemistry and functional assays] are readily available and of proven value, especially for confirming the diagnosis of autoantibody-mediated neurological diseases like myasthenia gravis, the Lambert–Eaton myasthenic syndrome, stiff-person syndrome and paraneoplastic neuroimmunological syndromes (Vincent and Martino, 2002).

In contrast, antigen-specific T cells continue to elude us. Most of the older techniques not only lacked sensitivity and specificity, but they were also plagued by other problems. Unlike B cells, T cells do not secrete their antigen-specific receptors; they retain them on their surface. Furthermore, whereas antibodies can bind to any type of antigen, T cells only recognize cell-associated (usually peptide) antigens, which are bound to human leukocyte antigen (HLA) molecules and are displayed on the surface of antigen-presenting cells (APC). Recognition takes place in a highly specialized zone, where the T cell and APC have direct cell-to-cell contact: the so-called immunological synapse (Krummel and Davis, 2002).

During T cell development in the thymus, billions of different receptors are generated by somatic rearrangement of germline-encoded genes in T cell precursor cells. The progeny of a single precursor cell is called a T cell clone, and all members of the clone express the same antigen-specific T cell receptor (TCR). There are several thousand copies of these unique antigen-specific TCR dimers on a T cell. Each dimer is composed of an  and a ß chain (Garcia et al., 1999). These receptors are said to be ‘clonally distributed’, because each T cell expresses only one type of (chemically identical) TCR protein. The size of the individual clone may vary with time, for example, during the course of an infection when certain pathogen-specific T cell clones are stimulated to divide (‘clonal expansion’).

Muraro and his colleagues in Roland Martin’s group at the National Institutes of Health (Bethesda, MD) used two exquisitely specific and sensitive molecular techniques to track and quantify single clones of (auto-)antigen-specific T cells in three patients with different immunological or infectious neurological diseases—multiple sclerosis, HTLV-I-associated myelopathy (HAM), and chronic Lyme neuroborreliosis. The first technique was a real-time (quantitative) RT–PCR (Walker, 2002), which allowed the authors to amplify mRNA transcripts of the TCR chains of the investigated T cell clones. The unique antigen-binding region of these chains is called the complementarity-determining region 3 (CDR3). By designing CDR3-specific (and hence clone-specific) oligonucleotide primers suitable for RT–PCR, the authors were able to track a T cell clone that recognized myelin basic protein (MBP) in the patient with multiple sclerosis, a second one that recognized an HTLV-I antigen in the patient with HAM/TSP, and a third one that recognized Borrelia burgdorferi in the patient with chronic Lyme disease.

The second technique detects the clonotypic TCR protein on the surface of individual T cells (Klenerman et al., 2002). The known antigen (in this case a peptide of the HTLV-I Tax antigen) was coupled to the same HLA molecule that normally presents this antigen on the surface of an APC. Such soluble ‘HLA–peptide multimer’ complexes can bind directly to the TCR (Klenerman et al., 2002). If the multimeric HLA–peptide complexes are labelled, e.g. with a fluorescent dye, the T cells to which they bind will also be labelled, and thus can be analysed by flow cytometry (FACS). This combination of quantitative PCR and TCR staining with HLA–peptide multimeric complexes provides an extremely powerful tool for assessing the clonal dynamics of antigen-specific T cells in neuroimmunological diseases.

One of the most intriguing findings of Muraro et al. is that the frequency of the investigated T cell clones increased during periods of clinical exacerbation. The course of the multiple sclerosis patient strikingly illustrates this. Five weeks after beginning treatment with an altered peptide ligand (APL) related to MBP, he had a severe clinical relapse. One week before the relapse, the frequency of the investigated MBP-specific T cell clone, which cross-reacted with the therapeutic APL peptide, increased 5-fold. After clinical remission, the frequency of the clone returned to baseline. This observation supports both the validity of the techniques and the pathogenic relevance of this particular T cell clone.

This elegant approach has an obvious limitation: it requires prior knowledge of the antigen specificity and the TCR sequence of the T cells being tracked (Table 1). Additional techniques are necessary to obtain this information, for example, in this case, the classic cloning of antigen-specific T cell lines and subsequent sequencing of their antigen-specific receptors. These time-consuming procedures are unsuitable for routine diagnostic purposes. Furthermore, they are geared to specific antigens (e.g. MBP, HTLV-I and B. burgdorferi) and will miss T cells specific for other (perhaps even unknown) potentially relevant antigens.
Method Material Sensitivity Comments

Functional tests, e.g. proliferation and cytotoxicity assays, Elispot, limited dilution assays Living cells Variable (e.g. antigen-induced proliferation: low sensitivity; Elispot: potentially high sensitivity) Classical techniques for detection of antigen-specific T cell responses
MHC–peptide multimers TCR surface protein Theoretically unlimited (detects single cells) T cells with down-regulated TCRs might escape detection; antigen and MHC restriction must be known
Clone-specific quantitative PCR mRNA/cDNA ˜1 cell in 105–106 TCR sequence of target T cell clone must be known
CDR3 spectratyping mRNA/cDNA ˜1–2 cells in 104 Unbiased screening technique; can detect clonal expansions of T cells with unknown TCR sequence and antigen specificity
Single-cell PCR Single DNA molecule of a single cell, e.g. in a tissue section Unlimited (detects single cells) Can relate morphological information (pathogenic relevance!) to clonal TCR sequence
Microarrays (expression profiles) cDNA Variable Can provide a global overview of gene expression in a defined population of T cells or even a single cell

cDNA, complementary DNA; CDR3, complementarity-determining region 3 (of the T cell receptor); Elispot, enzyme-linked immunospot assay; MHC, major histocompatibility complex; mRNA, messenger RNA; PCR, polymerase chain reaction; TCR, T cell receptor (for antigen).

Table 1 Different techniques for detecting and measuring T cell responses
Pathogenically relevant T cell clones should ideally be identified by unbiased means. Subsequently these cells could be tracked with the clone-specific methods described by Muraro et al. (2003). An excellent screening technique for such unbiased identification of clonal expansions is called ‘CDR3 spectratyping’ (Table 1) (Pannetier et al., 1995). This PCR-based method relies on the natural variation in length of the CDR3 region. When mixed populations of T cells are analysed with this method, their CDR3 regions show a normal (Gaussian) length distribution in each TCR V (variable) gene family. Activated T cell clones that have proliferated and are therefore expanded in number can be detected as distinct peaks. Because such peaks (clonal expansions) can occur in physiological situations, it is important to combine CDR3 spectratyping with a technique that allows the direct demonstration of the same T cell clone in sections of pathological tissue. For this purpose single infiltrating T cells can be ‘microdissected’ from tissue, and their TCR sequences can be determined by single-cell PCR. This pioneering approach has recently been applied in multiple sclerosis and polymyositis (Babbe et al., 2000; Goebels et al., 2001).
In conclusion, the paper by Muraro et al. convincingly demonstrates how clone-specific RT–PCR and staining with HLA–peptide complexes can be used to track potentially pathogenic T cell clones in the blood and CSF of patients with neuroimmunological diseases. This approach requires that the TCR sequence and/or antigen specificity of the target T cells be known in advance, but such knowledge can be obtained by combining an unbiased screening technique (like CDR3 spectratyping) with single-cell PCR (Fig. 1). The combination of these powerful techniques could begin a new era of T cell analysis in human autoimmune diseases.

Fig 1 Strategy for identifying and tracking autoimmune T cells. The first three steps (upper boxes) allow the identification of pathogenically relevant T cells in affected tissues, using a combination of different techniques. Once identified in tissue, the T cells can be tracked in blood and CSF by CDR3 spectratyping or clone-specific PCR.


I thank Ms Judy Benson for her help in editing the manuscript.


Babbe H, Roers A, Waisman A, Lassman H, Goebels N, Hohfeld R, et al. Clonal expansions of CD8(+) T cells dominate the T cell infiltrate in active multiple sclerosis lesions as shown by micromanipulation and single cell polymerase chain reaction. J Exp Med 2000; 192: 393–404.[Abstract/Free Full Text]

Garcia KC, Teyton L, Wilson IA. Structural basis of T cell recognition. [Review]. Annu Rev Immunol 1999; 17: 369–97.[Abstract/Free Full Text]

Goebels N, Hofbauer M, Wiesener S, Roers A, Pongratz D, Wekerle H, et al. Analysis of autoinvasive CD8+ T cells in polymyositis by laser-assisted microdissection, single cell PCR, and CDR3 spectratyping [abstract]. Neurology 2001; 56 (8 suppl. 3): A328.

Klenerman P, Cerundolo V, Dunbar PR. Tracking T cells with tetramers: new tales from new tools. [Review]. Nat Rev Immunol 2002; 2: 263–72.[Medline]

Krummel MF, Davis MM. Dynamics of the immunological synapse: finding, establishing and solidifying a connection. [Review]. Curr Opin Immunol 2002; 14: 66–74.[Medline]

Muraro PA, Wandinger KP, Bielekova B, Gran B, Marques A, Utz U, et al. Molecular tracking of antigen-specific T cell clones in neurological immune-mediated disorders. Brain 2003; 126: 20–31.

Pannetier C, Even J, Kourilsky P. T-cell repertoire diversity and clonal expansions in normal and clinical samples. [Review]. Immunol Today 1995; 16: 176–81.[Medline]

Vincent A, Martino G, editors. Autoantibodies in Neurological Diseases. Milano: Springer; 2002.

Walker NJ. Tech.Sight. A technique whose time has come. Science 2002; 296: 557–9.[Free Full Text]

© 2003 Guarantors of Brain