- Email: [email protected]
Limited clonality in autoimmunity: drivers and regulators
Autoimmunity Reviews 3 (2004) 524 – 529 www.elsevier.com/locate/autrev
Limited clonality in autoimmunity: drivers and regulators Peter van den Elzen, Juscilene S. Menezes, Akio Ametani, Emanual Maverakis, Loui Madakamutil, Xiao-lei Tang, Vipin Kumar, Eli E. Sercarz* Torrey Pines Institute for Molecular Studies, 3550 General Atomics Court, San Diego, CA 92121, United States Accepted 20 July 2004 Available online 17 August 2004
Abstract The available T cell repertoire directed against self is appreciable owing to the escape of many clones from negative selection, largely because many determinants on self proteins are cryptic and not presented adequately. In addition, the degeneracy of T cell receptor specificity permits each lymphocyte a broad recognitive potential. Within the available selfreactive repertoire are T cells with high affinity, and these can compete favorably with other T cells with the same specificity. We have studied a bdriver cloneQ and its two specific regulators in the B10.PL model of experimental autoimmune encephalomyelitis and found that each of these repertoires is highly limited. There is a single major clonal family comprising the aggressive driver population, which is public and of high affinity, and just one other minor public clonotype. The receptors of this Vh8.2/Jh2.7 driver are presented to a CD4 regulator and a CD8 suppressor, each of limited clonality, the latter killing the driver clone by apoptosis, completing a feedback control loop. This tightly regulated group of three cell types furnishes an excellent example of the immune homunculus. D 2004 Elsevier B.V. All rights reserved. Keywords: Driver clones; Competition; Degeneracy of T cell receptor recognition; Immune homunculus; Antigen processing
Contents 1. 2. 3. 4. 5. 6.
The breadth of the available T cell repertoire . . . . . . . . . . . Invisible determinants and the escape of self reactive T cells. . . Dominance of antigenic determinants: processing for presentation Dominance of a clonal family. . . . . . . . . . . . . . . . . . . What makes DAGGGY public and dominant? . . . . . . . . . . The self-reactive homunculus . . . . . . . . . . . . . . . . . . .
* Corresponding author. Tel.: +1 858 455 3775; fax: +1 858 455 3715. E-mail address: [email protected] (E.E. Sercarz). 1568-9972/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.autrev.2004.07.008
. . . . . . . . . . . . . . . . . . to CD4 T cells . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
525 525 526 526 527 527
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Take-home messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. The breadth of the available T cell repertoire Lymphocytes are unique in their ability to rearrange their antigen receptor genes at the DNA level in order to achieve a repertoire of maximal diversity. T lymphocytes follow a program of development where positive selection in the thymus first rescues any T cell from death that is capable of recognizing self MHC. Negative selection removes clones that recognize self peptides presented on MHC with too high an affinity [1,2]. However, there are a limited number of peptides from each self antigen that are expressed within the binding grooves of the self MHC molecules. This is a result of the unpredictability of antigenic processing: the availability of both endopeptidic cleavage sites and neighboring determinants with a reasonable affinity, as well as the competitive nature of peptide–MHC interaction [3]. Thus, the residual repertoire, a considerable population, although a minority compared to the potential repertoire, is typically weakly reactive to self, while able to respond to any theoretical foreign antigen with some high affinity clones. In the world of potential foreign peptides, however, it has been estimated that a spleen the size of a house would be necessary in order to match a single high affinity T cell to every dominant antigenic determinant. Thus, the T cell repertoire relies on degeneracy of antigen recognition to be able to respond to a maximal number of targets. Each T cell receptor seems highly flexible in its recognition—an apparent lack of specificity, but at the population level this likely translates into a broad recognition potential. Neither T, B nor APC can discern self (or harmless) from foreign (or dangerous). Such a discrimination is made by the immune system as a whole, and a variety of innate signal receptor systems help in identifying a given externally derived signal as dangerous. It must be remembered that even in an overwhelming infection, as in normal life, it will still be the dominant determinants on self antigens that are presented to the T cell repertoire, although many of its members have
525
528 528 528
been rendered non-aggressive either because they are purged, or are of very low affinity for target Ag-MHC complexes.
2. Invisible determinants and the escape of self reactive T cells Education of T cells in thymus relies on the complete exposure of the self to developing T cells. Lack of visibility at this stage results in potentially selfreactive T cells being released to the periphery. Thus, the thymus has adopted mechanisms to maximize the display of all possible self-antigens. The autoimmune regulatory element (AIRE) protein seems to be one of these mechanisms, a transcription factor which promiscuously promotes expression of genes that are otherwise unique to other tissues (pancreatic antigens, thyroid antigens, myelin antigens) [4,5]. Lack of this factor results in multi-system autoimmunity. Nevertheless, as stressed above, although many self-antigens are displayed, only some determinants on each antigen will become visible to ambient T cells. Another mechanism of escape can occur when a protein is differentially spliced in the thymus compared to the periphery. The myelin protein PLP is only expressed in the thymus as its truncated splice variant, DM20. Thus, the region of PLP not found in DM20 has never been exposed to T cells developing in the thymus, leaving a potential hole with respect to tolerance induction [6]. Indeed, the immunodominant peptide of PLP, 139–151, is within this invisible region, and is the resultant target of autoimmune attack when displayed in the central nervous system. In another model of autoimmunity, the immunodominant peptide of MBP is the acetylated N-terminus of the protein, Ac1-9, which has a repertoire of T cells capable of recognizing it despite the expression of MBP in the thymus. In this case, immune invisibility occurs due to competition of flanking determinants being able to completely outcompete Ac1-9 for
526
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529
display on MHC [7,8]. Instead of a splice variant resulting in lack of display of an epitope (as in DM20), in MBP, a splice variant which results in a higher affinity flanking peptide (golli-MBP determinant) [9], and another strongly binding determinant on the other flank conspire to prevent Ac1-9 from being properly exposed to delete self reactive T cells. When the truncated version of MBP is later exposed in the maturing animal’s CNS, Ac1-9 gains greater availability and can induce high affinity T cells despite its low affinity for the MHC. Interestingly, one such clone gains predominance and becomes a bdriverQ of the disease process (called bDAGGGYQ because of its CDR-3 sequence) [10]. In this particular case, most T cells specific for the efficiently binding 7–16 determinant should have been rendered tolerant, permitting DAGGGY to win the competition for activation [8].
3. Dominance of antigenic determinants: processing for presentation to CD4 T cells Responses to protein antigens occur following processing of that protein in professional antigenpresenting cells. Peptides are presented on MHC II to T cells, and in the case of self-peptides, will only be recognized in the unfortunate circumstance where autoreactive T cells have escaped tolerance (described above). Selective processing of a protein determines which peptides are displayed by the MHC. The unfolding and proteolytic degradation of the protein does not result in an equal likelihood of any peptide from the protein binding to the MHC molecule [11]. As the unfolding and digestion takes place in the endocytic compartment, preferential display occurs due to availability, within the context of the unfolded protein, of a peptide sequence and its affinity for the MHC. Furthermore, the proteases involved in degradation have preferential cleavage sites on the protein, destroying some potential epitopes while leaving some intact. We envision the ideal circumstance in yielding a dominant class II MHC binding determinant is an endopeptidase target site close to a determinant with relatively high affinity for the MHC. Any instance where previously cryptic determinants become dominant can result in autoimmunity [12]. This situation often arises in an inflammatory focus where Ag processing becomes greatly enhanced.
4. Dominance of a clonal family In recent studies, we have characterized the repertoire responding to the immunodominant peptide of myelin basic protein (MBP), the self antigenic target in experimental autoimmune encephalomyelitis (EAE). Due to the competition of determinants during thymic antigen presentation, ineffective central tolerance occurs, leaving a repertoire of potentially autoreactive T cells to Ac1-9 (described above). Despite original reports of a limited repertoire to Ac1-9, we have found that the repertoire is quite heterogenous when analyzed ex vivo. The selection imposed by in vitro culturing of T cells led to the mistaken view that very few clones persisted during the competitive establishment of the self-reactive response. We utilized a PCR technique, TcR spectratyping or the bimmunoscopeQ, that allowed a more direct analysis of those clones responding to Ac1-9 in vivo [13]. A diverse set of clones was found to respond to the antigen, each with a unique T cell receptor. From animal to animal, this large repertoire was comprised of mostly non-overlapping, so-called bprivateQ members. However, a few TcR seemed to be unique in being expressed dominantly in all mice, referred to as the bpublicQ repertoire. The selection that led to a particular TcR was not due to the preferential rearrangement of this TcR sequence, since we saw that the public DAGGGY clone was not actually a clone, but was a clonal family, with members of distinct origin using different DNA sequences to encode the DAGGGY CDR3 region. This implied a selection at the level of antigen recognition. What is so unique about DAGGGY to give it such an advantage over the private repertoire? Indeed, it was found to have an affinity advantage over other Ac1-9 reactive clones when assessed with I-Au Ac1-9 tetramers [14]. Probably as a result of this higher affinity, the DAGGGY population was particularly polarized to be Th1 [15]—being found in the IFNg producing population, but not the IL-4 population. In contrast, private clones seemed to be distributed into both Th1 and Th2 populations. A particularly unique feature of this clonal family was its association with disease. Analysis of the spinal-cord infiltrating population of T cells showed this expansion to always be present during active disease and lost during remission. Private Ac1-9-
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529
reactive clones from the splenic repertoire were rarely or never found in the spinal cord of the same mice.
5. What makes DAGGGY public and dominant? The emergence of high and low affinity clones is perhaps an expected outcome when injecting a high dose of a single peptide antigen such as MBP-Ac1-9 into an animal. Interestingly, the same situation occurred with respect to MBP-specific clones when we used whole myelin as the immunogen. Still, the DAGGGY public clone emerged in all animals tested. In fact, with the multitude of potential epitopes available in the antigenic mixture of crude myelin and complete Freund’s adjuvant, there still emerged just a few dominant TcR sequences. What other features of DAGGGY could have led to the establishment of its dominance? Let us consider three areas— degeneracy, homeostatic proliferation, and trafficking. (a) Degeneracy. Although it is now considered that each T cell receptor has the capacity to recognize as many as 104–106 epitope/MHC structures, some TcRs may be more degenerate than others [16]. For example, the string of three glycines in DAGGGY may provide for particularly degenerate recognition, owing to the freedom of rotation permitted by the lack of sequential R groups. Indeed, the DAGGGY clone is known to recognize many different peptides bound to I-Au [17]. We have therefore hypothesized that receptor degeneracy may be a prime factor in the preferential expansion of certain clones. (b) Homeostatic Proliferation. Circulating T cells appear to receive constant tonic TcR stimulation in order to maintain them in a quasi-primed state, ready to receive their bcognateQ signal. Thus, it may not be surprising to find that those clones that end up topping the hierarchy in a heterogenous response are those that combine a flexible degenerate recognition with a propensity to compete favorably during homeostatic proliferation. This may be occurring throughout the lifetime of the clone, and probably even during its homeostatic existence as a naRve T cell. Thus, DAGGGY may have already been pre-selected in the naRve repertoire. We will question this possibility experimentally by attempting to quantify the precursor population of DAGGGY relative to other potential Ac1-9 reactive clones.
527
In addition to a possible precursor frequency advantage in the emergence of public clones, there probably is a post-antigen selection, based presumably on the combination of high affinity for Ac1-9 and degenerate recognition of many determinants (in CFA, or in myelin) that may select this clone to be dominant. (c) Trafficking. Particularly intriguing is the consistent finding of DAGGGY in the spinal cord. What makes DAGGGY particularly infiltrative? Perhaps its high affinity and consequent state of activation allows selective (re)expansion in the spinal cord. Perhaps, Th1 polarization leads to a homing receptor profile which favors its passage through the blood–brain barrier. Once in the spinal cord, its especially high degeneracy likely allows for the recognition of a broader array of stimulatory myelin epitopes.
6. The self-reactive homunculus Cohen described the immune homunculus as a requirement for regulation in the system. It represents those motifs in the immune system that become overrepresented because they are essential for its function [18,19]. He foresaw that certain self-reactive clones would be public, because it was necessary to have a focus on specific regulation: thus, their receptors would be recognized by the regulators and since the homuncular clone would be dominant, this regulation should safeguard homeostasis. As long as the dominant driver clone can be regulated, it is not necessary to target its recruits for destruction [20]. As a further extension of the premise, if regulators were also limited to a small panel, then the whole structure of the homunculus would be highly cogent and focused. The alternative situations could be avoided, of having single regulators for each activated effector cell, creating a rather chaotic web of recognition; another solution would be acceptance of broad bystander regulation. Indeed in fact, the regulatory circuitry in the B10.PL mouse which has been studied extensively in our laboratory [21], displays all of the characteristics of a homuncular regulatory system. The T cell receptors of the major driver and perpetrator of disease, DAGGGY, are transferred to antigen
528
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529
presenting cells which then activate a Vh14 CD4 and a Vh6 CD8 duo of regulators. Both the CD4 and the CD8 regulators appear to employ limited TcR repertoires, although further confirmation is necessary for the CD8 T cells. The CD8 T cells have been shown to apoptose activated DAGGGY T cells [20], which presumably present peptide 42– 50 of Vh8.2 in the context of class I MHC. The existence of this loop of self-reactive aggressors and regulators, once primed, manages to provide the individual with protection from aggressive high affinity CD4 T cells, maintaining a status quo of disease-free life. A most noteworthy feature of this feedback regulatory system is the limited repertoire of each of its cellular components. It is tempting to speculate that this limitation is selected for in evolution, and may be a characteristic of other such systems.
Acknowledgements Research for this paper is supported by two grants from: National Institute of Health: RO1AI48077 and RO1AI42396-01; and by a grant from: National Multiple Sclerosis Society: RG263C8T.
Take-home messages ! The T cell repertoire relies on the degenerate recognition of antigen by its members, allowing a response to a maximal number of determinants. ! Cryptic determinants, not engaged in negative selection owing to their poor presentation, can result in permitting the maturation of high affinity, dangerously aggressive, self-reactive T cells. ! Previously cryptic determinants can be revealed under inflammatory conditions and thereby induce non-tolerized self-reactive lymphocytes. ! bDAGGGYQ is a public and dominant clone in the B10.PL strain which is induced by MBP1-9 and can cause experimental autoimmune encephalomyelitis, and functions as a bdriver cloneQ to enlist other clones and propagate the autoimmune response. This driver is an excellent competitor of high affinity, exhibiting degenerate recognition.
! DAGGGY induces a feedback regulation loop through the presentation of its TcR receptor Vh peptides to a CD4 and a CD8 regulator with a limited repertoire. This loop is a fine example of homuncular regulation. ! Interestingly, both the DAGGGY effector and its two regulators have a very limited repertoire, resulting in a cogent and centralized, regulatable system.
References [1] von Boehmer H, Aifantis I, Gounari F, Azogui O, Haughn L, Apostolou I, et al. Thymic selection revisited: how essential is it? Immunol Rev 2003;191:62 – 78. [2] Starr TK, Jameson SC, Hogquist KA. Positive and negative selection of T cells. Annu Rev Immunol 2003;21:139 – 76. [3] Sercarz EE. Processing creates the self. Nat Immunol 2002; 3:110 – 2. [4] Anderson MS, Venanzi ES, Klein L, Chen Z, Berzins SP, Turley SJ, et al. Projection of an immunological self shadow within the thymus by the aire protein. Science 2002;298: 1395 – 401. [5] Kyewski B, Derbinski J, Gotter J, Klein L. Promiscuous gene expression and central T-cell tolerance: more than meets the eye. Trends Immunol 2002;23:364 – 71. [6] Greer JM, Sobel RA, Sette A, Southwood S, Lees MB, Kuchroo KV. Immunogenic and encephalitogenic epitope clusters of myelin proteolipid protein. J Immunol 1996;156: 371 – 9. [7] Maverakis E, Beech J, Stevens DB, Ametani A, Brossay L, van den Elzen P, et al. Autoreactive T cells can be protected from tolerance induction through competition by flanking determinants for access to class II MHC. Proc Natl Acad Sci U S A 2003;100:5342 – 7. [8] Seamons A, Sutton J, Bai D, Baird E, Bonn N, Kafsack BF, et al. Competition between two MHC binding registers in a single peptide processed from myelin basic protein influences tolerance and susceptibility to autoimmunity. J Exp Med 2003;197:1391 – 7. [9] Pribyl TM, Campagnoni C, Kampf K, Handley VW, Campagnoni AT. The major myelin protein genes are expressed in the human thymus. J Neurosci Res 1996;45:812 – 9. [10] Urban JL, Kumar V, Kono DH, Gomez C, Horvath SJ, Clayton J, et al. Restricted use of T cell receptor V genes in murine autoimmune encephalomyelitis raises possibilities for antibody therapy. Cell 1988;54:577 – 92. [11] Sercarz EE, Maverakis E. Mhc-guided processing: binding of large antigen fragments. Nat Rev, Immunol 2003;3: 621 – 9. [12] Sercarz EE, Lehmann PV, Ametani A, Benichou G, Miller A, Moudgil K. Dominance and crypticity of T cell antigenic determinants. Annu Rev Immunol 1993;11:729 – 66.
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529 [13] Pannetier C, Cochet M, Darche S, Casrouge A, Zoller M, Kourilsky P. The sizes of the CDR3 hypervariable regions of the murine T-cell receptor beta chains vary as a function of the recombined germ-line segments. Proc Natl Acad Sci U S A 1993;90:4319 – 23. [14] Garcia KC, Radu CG, Ho J, Ober RJ, Ward ES. Kinetics and thermodynamics of T cell receptor–autoantigen interactions in murine experimental autoimmune encephalomyelitis. Proc Natl Acad Sci U S A 2001;98:6818 – 23. [15] Kumar V, Bhardwaj V, Soares L, Alexander J, Sette A, Sercarz E. Major histocompatibility complex binding affinity of an antigenic determinant is crucial for the differential secretion of interleukin 4/5 or interferon gamma by T cells. Proc Natl Acad Sci U S A 1995;92:9510 – 4. [16] Sercarz E, Cohen I (guest editors). Degeneracy of T cell recognition and its relationship to molecular mimicry. Mol. Immunol. 2004;40:983–1145.
529
[17] Bhardwaj V, Kumar V, Geysen HM, Sercarz EE. Degenerate recognition of a dissimilar antigenic peptide by myelin basic protein-reactive T cells. Implications for thymic education and autoimmunity. J Immunol 1993;151:5000 – 10. [18] Cohen RI, Young DB. Autoimmunity, microbial immunity and the immunological homunculus. Immunol Today 1991;12: 105 – 10. [19] Cohen IR. The cognitive paradigm and the immunological homunculus. Immunol Today 1992;13:490 – 4. [20] Madakamutil LT, Maricic I, Sercarz E, Kumar V. Regulatory T cells control autoimmunity in vivo by inducing apoptotic depletion of activated pathogenic lymphocytes. J Immunol 2003;170:2985 – 92. [21] Kumar V, Sercarz E. An integrative model of regulation centered on recognition of TCR peptide/MHC complexes. Immunol Rev 2001;182:113 – 21.
The World of Autoimmunity; Literature Synopsis Mast cell in early phase vasculitis in the Brown Norway rat model of vasculitis Vinen et al. (Int J Exp Pathol 2004;85:165) report about the central role of mast cells in a model of vasculitis. Administration of mercuric chloride to Brown Norway rats causes Th2-dominated autoimmunity with raised immunoglobulin E concentrations and gut vasculitis, both of which are T-cell dependent, peak at 14 days after starting mercuric chloride and then spontaneously resolve. If animals are re-challenged with mercuric chloride 6 weeks after initial exposure, they are resistant to autoimmunity, and developing only attenuated disease. Recently, a separate phase of early caecal vasculitis was described beginning 24 h after initiating mercuric chloride and prior to caecal entry of T cells. The authors performed a histological study during the first 93 hours following mercuric chloride challenge defining the precise relationship between gut mast cell degranulation and appearing caecal vasculitis. Their findings indicate a direct correlation between mast cell degranulation and early caecal vasculitis following the initial mercuric chloride challenge.
B lymphocyte signalling established by the CD19/CD22 loop and autoimmunity in the tight-skin mouse Systemic sclerosis is characterized by fibrosis and autoantibody production. Peripheral blood B cells from systemic sclerosis patients specifically over express CD19, a critical cell-surface signal transduction molecule in B cells. Asano et al. (Am J Pathol 2004;165:641) analyzed transgenic mouse lines that over express CD19. Their findings indicate that 20% increase of CD19 expression in mice spontaneously induced systemic sclerosis-specific anti-DNA topoisomerase I antibody production, which was further augmented by 200% over expression. However, no associated skin thickness increase was found. The molecular mechanisms for abnormal CD19 signaling were assessed by the authors: B-cell antigen receptor cross linking induced exaggerated calcium responses and augmented activation of extra cellular signalregulated kinase in TSK/+ B cells. In addition, CD22 function was specifically impaired in TSK/+ B cells. Consistently, CD19, a major target of CD22-negative regulation, was hyperphosphorylated in TSK/+ B cells. Therefore, reduced inhibitory signal provided by CD22 results in abnormal activation of signaling pathways including CD19 in TSK/+ mice; this disrupted B cell signaling contribute to specific autoantibody production.
Limited clonality in autoimmunity: drivers and regulators Peter van den Elzen, Juscilene S. Menezes, Akio Ametani, Emanual Maverakis, Loui Madakamutil, Xiao-lei Tang, Vipin Kumar, Eli E. Sercarz* Torrey Pines Institute for Molecular Studies, 3550 General Atomics Court, San Diego, CA 92121, United States Accepted 20 July 2004 Available online 17 August 2004
Abstract The available T cell repertoire directed against self is appreciable owing to the escape of many clones from negative selection, largely because many determinants on self proteins are cryptic and not presented adequately. In addition, the degeneracy of T cell receptor specificity permits each lymphocyte a broad recognitive potential. Within the available selfreactive repertoire are T cells with high affinity, and these can compete favorably with other T cells with the same specificity. We have studied a bdriver cloneQ and its two specific regulators in the B10.PL model of experimental autoimmune encephalomyelitis and found that each of these repertoires is highly limited. There is a single major clonal family comprising the aggressive driver population, which is public and of high affinity, and just one other minor public clonotype. The receptors of this Vh8.2/Jh2.7 driver are presented to a CD4 regulator and a CD8 suppressor, each of limited clonality, the latter killing the driver clone by apoptosis, completing a feedback control loop. This tightly regulated group of three cell types furnishes an excellent example of the immune homunculus. D 2004 Elsevier B.V. All rights reserved. Keywords: Driver clones; Competition; Degeneracy of T cell receptor recognition; Immune homunculus; Antigen processing
Contents 1. 2. 3. 4. 5. 6.
The breadth of the available T cell repertoire . . . . . . . . . . . Invisible determinants and the escape of self reactive T cells. . . Dominance of antigenic determinants: processing for presentation Dominance of a clonal family. . . . . . . . . . . . . . . . . . . What makes DAGGGY public and dominant? . . . . . . . . . . The self-reactive homunculus . . . . . . . . . . . . . . . . . . .
* Corresponding author. Tel.: +1 858 455 3775; fax: +1 858 455 3715. E-mail address: [email protected] (E.E. Sercarz). 1568-9972/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.autrev.2004.07.008
. . . . . . . . . . . . . . . . . . to CD4 T cells . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
525 525 526 526 527 527
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Take-home messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. The breadth of the available T cell repertoire Lymphocytes are unique in their ability to rearrange their antigen receptor genes at the DNA level in order to achieve a repertoire of maximal diversity. T lymphocytes follow a program of development where positive selection in the thymus first rescues any T cell from death that is capable of recognizing self MHC. Negative selection removes clones that recognize self peptides presented on MHC with too high an affinity [1,2]. However, there are a limited number of peptides from each self antigen that are expressed within the binding grooves of the self MHC molecules. This is a result of the unpredictability of antigenic processing: the availability of both endopeptidic cleavage sites and neighboring determinants with a reasonable affinity, as well as the competitive nature of peptide–MHC interaction [3]. Thus, the residual repertoire, a considerable population, although a minority compared to the potential repertoire, is typically weakly reactive to self, while able to respond to any theoretical foreign antigen with some high affinity clones. In the world of potential foreign peptides, however, it has been estimated that a spleen the size of a house would be necessary in order to match a single high affinity T cell to every dominant antigenic determinant. Thus, the T cell repertoire relies on degeneracy of antigen recognition to be able to respond to a maximal number of targets. Each T cell receptor seems highly flexible in its recognition—an apparent lack of specificity, but at the population level this likely translates into a broad recognition potential. Neither T, B nor APC can discern self (or harmless) from foreign (or dangerous). Such a discrimination is made by the immune system as a whole, and a variety of innate signal receptor systems help in identifying a given externally derived signal as dangerous. It must be remembered that even in an overwhelming infection, as in normal life, it will still be the dominant determinants on self antigens that are presented to the T cell repertoire, although many of its members have
525
528 528 528
been rendered non-aggressive either because they are purged, or are of very low affinity for target Ag-MHC complexes.
2. Invisible determinants and the escape of self reactive T cells Education of T cells in thymus relies on the complete exposure of the self to developing T cells. Lack of visibility at this stage results in potentially selfreactive T cells being released to the periphery. Thus, the thymus has adopted mechanisms to maximize the display of all possible self-antigens. The autoimmune regulatory element (AIRE) protein seems to be one of these mechanisms, a transcription factor which promiscuously promotes expression of genes that are otherwise unique to other tissues (pancreatic antigens, thyroid antigens, myelin antigens) [4,5]. Lack of this factor results in multi-system autoimmunity. Nevertheless, as stressed above, although many self-antigens are displayed, only some determinants on each antigen will become visible to ambient T cells. Another mechanism of escape can occur when a protein is differentially spliced in the thymus compared to the periphery. The myelin protein PLP is only expressed in the thymus as its truncated splice variant, DM20. Thus, the region of PLP not found in DM20 has never been exposed to T cells developing in the thymus, leaving a potential hole with respect to tolerance induction [6]. Indeed, the immunodominant peptide of PLP, 139–151, is within this invisible region, and is the resultant target of autoimmune attack when displayed in the central nervous system. In another model of autoimmunity, the immunodominant peptide of MBP is the acetylated N-terminus of the protein, Ac1-9, which has a repertoire of T cells capable of recognizing it despite the expression of MBP in the thymus. In this case, immune invisibility occurs due to competition of flanking determinants being able to completely outcompete Ac1-9 for
526
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529
display on MHC [7,8]. Instead of a splice variant resulting in lack of display of an epitope (as in DM20), in MBP, a splice variant which results in a higher affinity flanking peptide (golli-MBP determinant) [9], and another strongly binding determinant on the other flank conspire to prevent Ac1-9 from being properly exposed to delete self reactive T cells. When the truncated version of MBP is later exposed in the maturing animal’s CNS, Ac1-9 gains greater availability and can induce high affinity T cells despite its low affinity for the MHC. Interestingly, one such clone gains predominance and becomes a bdriverQ of the disease process (called bDAGGGYQ because of its CDR-3 sequence) [10]. In this particular case, most T cells specific for the efficiently binding 7–16 determinant should have been rendered tolerant, permitting DAGGGY to win the competition for activation [8].
3. Dominance of antigenic determinants: processing for presentation to CD4 T cells Responses to protein antigens occur following processing of that protein in professional antigenpresenting cells. Peptides are presented on MHC II to T cells, and in the case of self-peptides, will only be recognized in the unfortunate circumstance where autoreactive T cells have escaped tolerance (described above). Selective processing of a protein determines which peptides are displayed by the MHC. The unfolding and proteolytic degradation of the protein does not result in an equal likelihood of any peptide from the protein binding to the MHC molecule [11]. As the unfolding and digestion takes place in the endocytic compartment, preferential display occurs due to availability, within the context of the unfolded protein, of a peptide sequence and its affinity for the MHC. Furthermore, the proteases involved in degradation have preferential cleavage sites on the protein, destroying some potential epitopes while leaving some intact. We envision the ideal circumstance in yielding a dominant class II MHC binding determinant is an endopeptidase target site close to a determinant with relatively high affinity for the MHC. Any instance where previously cryptic determinants become dominant can result in autoimmunity [12]. This situation often arises in an inflammatory focus where Ag processing becomes greatly enhanced.
4. Dominance of a clonal family In recent studies, we have characterized the repertoire responding to the immunodominant peptide of myelin basic protein (MBP), the self antigenic target in experimental autoimmune encephalomyelitis (EAE). Due to the competition of determinants during thymic antigen presentation, ineffective central tolerance occurs, leaving a repertoire of potentially autoreactive T cells to Ac1-9 (described above). Despite original reports of a limited repertoire to Ac1-9, we have found that the repertoire is quite heterogenous when analyzed ex vivo. The selection imposed by in vitro culturing of T cells led to the mistaken view that very few clones persisted during the competitive establishment of the self-reactive response. We utilized a PCR technique, TcR spectratyping or the bimmunoscopeQ, that allowed a more direct analysis of those clones responding to Ac1-9 in vivo [13]. A diverse set of clones was found to respond to the antigen, each with a unique T cell receptor. From animal to animal, this large repertoire was comprised of mostly non-overlapping, so-called bprivateQ members. However, a few TcR seemed to be unique in being expressed dominantly in all mice, referred to as the bpublicQ repertoire. The selection that led to a particular TcR was not due to the preferential rearrangement of this TcR sequence, since we saw that the public DAGGGY clone was not actually a clone, but was a clonal family, with members of distinct origin using different DNA sequences to encode the DAGGGY CDR3 region. This implied a selection at the level of antigen recognition. What is so unique about DAGGGY to give it such an advantage over the private repertoire? Indeed, it was found to have an affinity advantage over other Ac1-9 reactive clones when assessed with I-Au Ac1-9 tetramers [14]. Probably as a result of this higher affinity, the DAGGGY population was particularly polarized to be Th1 [15]—being found in the IFNg producing population, but not the IL-4 population. In contrast, private clones seemed to be distributed into both Th1 and Th2 populations. A particularly unique feature of this clonal family was its association with disease. Analysis of the spinal-cord infiltrating population of T cells showed this expansion to always be present during active disease and lost during remission. Private Ac1-9-
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529
reactive clones from the splenic repertoire were rarely or never found in the spinal cord of the same mice.
5. What makes DAGGGY public and dominant? The emergence of high and low affinity clones is perhaps an expected outcome when injecting a high dose of a single peptide antigen such as MBP-Ac1-9 into an animal. Interestingly, the same situation occurred with respect to MBP-specific clones when we used whole myelin as the immunogen. Still, the DAGGGY public clone emerged in all animals tested. In fact, with the multitude of potential epitopes available in the antigenic mixture of crude myelin and complete Freund’s adjuvant, there still emerged just a few dominant TcR sequences. What other features of DAGGGY could have led to the establishment of its dominance? Let us consider three areas— degeneracy, homeostatic proliferation, and trafficking. (a) Degeneracy. Although it is now considered that each T cell receptor has the capacity to recognize as many as 104–106 epitope/MHC structures, some TcRs may be more degenerate than others [16]. For example, the string of three glycines in DAGGGY may provide for particularly degenerate recognition, owing to the freedom of rotation permitted by the lack of sequential R groups. Indeed, the DAGGGY clone is known to recognize many different peptides bound to I-Au [17]. We have therefore hypothesized that receptor degeneracy may be a prime factor in the preferential expansion of certain clones. (b) Homeostatic Proliferation. Circulating T cells appear to receive constant tonic TcR stimulation in order to maintain them in a quasi-primed state, ready to receive their bcognateQ signal. Thus, it may not be surprising to find that those clones that end up topping the hierarchy in a heterogenous response are those that combine a flexible degenerate recognition with a propensity to compete favorably during homeostatic proliferation. This may be occurring throughout the lifetime of the clone, and probably even during its homeostatic existence as a naRve T cell. Thus, DAGGGY may have already been pre-selected in the naRve repertoire. We will question this possibility experimentally by attempting to quantify the precursor population of DAGGGY relative to other potential Ac1-9 reactive clones.
527
In addition to a possible precursor frequency advantage in the emergence of public clones, there probably is a post-antigen selection, based presumably on the combination of high affinity for Ac1-9 and degenerate recognition of many determinants (in CFA, or in myelin) that may select this clone to be dominant. (c) Trafficking. Particularly intriguing is the consistent finding of DAGGGY in the spinal cord. What makes DAGGGY particularly infiltrative? Perhaps its high affinity and consequent state of activation allows selective (re)expansion in the spinal cord. Perhaps, Th1 polarization leads to a homing receptor profile which favors its passage through the blood–brain barrier. Once in the spinal cord, its especially high degeneracy likely allows for the recognition of a broader array of stimulatory myelin epitopes.
6. The self-reactive homunculus Cohen described the immune homunculus as a requirement for regulation in the system. It represents those motifs in the immune system that become overrepresented because they are essential for its function [18,19]. He foresaw that certain self-reactive clones would be public, because it was necessary to have a focus on specific regulation: thus, their receptors would be recognized by the regulators and since the homuncular clone would be dominant, this regulation should safeguard homeostasis. As long as the dominant driver clone can be regulated, it is not necessary to target its recruits for destruction [20]. As a further extension of the premise, if regulators were also limited to a small panel, then the whole structure of the homunculus would be highly cogent and focused. The alternative situations could be avoided, of having single regulators for each activated effector cell, creating a rather chaotic web of recognition; another solution would be acceptance of broad bystander regulation. Indeed in fact, the regulatory circuitry in the B10.PL mouse which has been studied extensively in our laboratory [21], displays all of the characteristics of a homuncular regulatory system. The T cell receptors of the major driver and perpetrator of disease, DAGGGY, are transferred to antigen
528
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529
presenting cells which then activate a Vh14 CD4 and a Vh6 CD8 duo of regulators. Both the CD4 and the CD8 regulators appear to employ limited TcR repertoires, although further confirmation is necessary for the CD8 T cells. The CD8 T cells have been shown to apoptose activated DAGGGY T cells [20], which presumably present peptide 42– 50 of Vh8.2 in the context of class I MHC. The existence of this loop of self-reactive aggressors and regulators, once primed, manages to provide the individual with protection from aggressive high affinity CD4 T cells, maintaining a status quo of disease-free life. A most noteworthy feature of this feedback regulatory system is the limited repertoire of each of its cellular components. It is tempting to speculate that this limitation is selected for in evolution, and may be a characteristic of other such systems.
Acknowledgements Research for this paper is supported by two grants from: National Institute of Health: RO1AI48077 and RO1AI42396-01; and by a grant from: National Multiple Sclerosis Society: RG263C8T.
Take-home messages ! The T cell repertoire relies on the degenerate recognition of antigen by its members, allowing a response to a maximal number of determinants. ! Cryptic determinants, not engaged in negative selection owing to their poor presentation, can result in permitting the maturation of high affinity, dangerously aggressive, self-reactive T cells. ! Previously cryptic determinants can be revealed under inflammatory conditions and thereby induce non-tolerized self-reactive lymphocytes. ! bDAGGGYQ is a public and dominant clone in the B10.PL strain which is induced by MBP1-9 and can cause experimental autoimmune encephalomyelitis, and functions as a bdriver cloneQ to enlist other clones and propagate the autoimmune response. This driver is an excellent competitor of high affinity, exhibiting degenerate recognition.
! DAGGGY induces a feedback regulation loop through the presentation of its TcR receptor Vh peptides to a CD4 and a CD8 regulator with a limited repertoire. This loop is a fine example of homuncular regulation. ! Interestingly, both the DAGGGY effector and its two regulators have a very limited repertoire, resulting in a cogent and centralized, regulatable system.
References [1] von Boehmer H, Aifantis I, Gounari F, Azogui O, Haughn L, Apostolou I, et al. Thymic selection revisited: how essential is it? Immunol Rev 2003;191:62 – 78. [2] Starr TK, Jameson SC, Hogquist KA. Positive and negative selection of T cells. Annu Rev Immunol 2003;21:139 – 76. [3] Sercarz EE. Processing creates the self. Nat Immunol 2002; 3:110 – 2. [4] Anderson MS, Venanzi ES, Klein L, Chen Z, Berzins SP, Turley SJ, et al. Projection of an immunological self shadow within the thymus by the aire protein. Science 2002;298: 1395 – 401. [5] Kyewski B, Derbinski J, Gotter J, Klein L. Promiscuous gene expression and central T-cell tolerance: more than meets the eye. Trends Immunol 2002;23:364 – 71. [6] Greer JM, Sobel RA, Sette A, Southwood S, Lees MB, Kuchroo KV. Immunogenic and encephalitogenic epitope clusters of myelin proteolipid protein. J Immunol 1996;156: 371 – 9. [7] Maverakis E, Beech J, Stevens DB, Ametani A, Brossay L, van den Elzen P, et al. Autoreactive T cells can be protected from tolerance induction through competition by flanking determinants for access to class II MHC. Proc Natl Acad Sci U S A 2003;100:5342 – 7. [8] Seamons A, Sutton J, Bai D, Baird E, Bonn N, Kafsack BF, et al. Competition between two MHC binding registers in a single peptide processed from myelin basic protein influences tolerance and susceptibility to autoimmunity. J Exp Med 2003;197:1391 – 7. [9] Pribyl TM, Campagnoni C, Kampf K, Handley VW, Campagnoni AT. The major myelin protein genes are expressed in the human thymus. J Neurosci Res 1996;45:812 – 9. [10] Urban JL, Kumar V, Kono DH, Gomez C, Horvath SJ, Clayton J, et al. Restricted use of T cell receptor V genes in murine autoimmune encephalomyelitis raises possibilities for antibody therapy. Cell 1988;54:577 – 92. [11] Sercarz EE, Maverakis E. Mhc-guided processing: binding of large antigen fragments. Nat Rev, Immunol 2003;3: 621 – 9. [12] Sercarz EE, Lehmann PV, Ametani A, Benichou G, Miller A, Moudgil K. Dominance and crypticity of T cell antigenic determinants. Annu Rev Immunol 1993;11:729 – 66.
P. van den Elzen et al. / Autoimmunity Reviews 3 (2004) 524–529 [13] Pannetier C, Cochet M, Darche S, Casrouge A, Zoller M, Kourilsky P. The sizes of the CDR3 hypervariable regions of the murine T-cell receptor beta chains vary as a function of the recombined germ-line segments. Proc Natl Acad Sci U S A 1993;90:4319 – 23. [14] Garcia KC, Radu CG, Ho J, Ober RJ, Ward ES. Kinetics and thermodynamics of T cell receptor–autoantigen interactions in murine experimental autoimmune encephalomyelitis. Proc Natl Acad Sci U S A 2001;98:6818 – 23. [15] Kumar V, Bhardwaj V, Soares L, Alexander J, Sette A, Sercarz E. Major histocompatibility complex binding affinity of an antigenic determinant is crucial for the differential secretion of interleukin 4/5 or interferon gamma by T cells. Proc Natl Acad Sci U S A 1995;92:9510 – 4. [16] Sercarz E, Cohen I (guest editors). Degeneracy of T cell recognition and its relationship to molecular mimicry. Mol. Immunol. 2004;40:983–1145.
529
[17] Bhardwaj V, Kumar V, Geysen HM, Sercarz EE. Degenerate recognition of a dissimilar antigenic peptide by myelin basic protein-reactive T cells. Implications for thymic education and autoimmunity. J Immunol 1993;151:5000 – 10. [18] Cohen RI, Young DB. Autoimmunity, microbial immunity and the immunological homunculus. Immunol Today 1991;12: 105 – 10. [19] Cohen IR. The cognitive paradigm and the immunological homunculus. Immunol Today 1992;13:490 – 4. [20] Madakamutil LT, Maricic I, Sercarz E, Kumar V. Regulatory T cells control autoimmunity in vivo by inducing apoptotic depletion of activated pathogenic lymphocytes. J Immunol 2003;170:2985 – 92. [21] Kumar V, Sercarz E. An integrative model of regulation centered on recognition of TCR peptide/MHC complexes. Immunol Rev 2001;182:113 – 21.
The World of Autoimmunity; Literature Synopsis Mast cell in early phase vasculitis in the Brown Norway rat model of vasculitis Vinen et al. (Int J Exp Pathol 2004;85:165) report about the central role of mast cells in a model of vasculitis. Administration of mercuric chloride to Brown Norway rats causes Th2-dominated autoimmunity with raised immunoglobulin E concentrations and gut vasculitis, both of which are T-cell dependent, peak at 14 days after starting mercuric chloride and then spontaneously resolve. If animals are re-challenged with mercuric chloride 6 weeks after initial exposure, they are resistant to autoimmunity, and developing only attenuated disease. Recently, a separate phase of early caecal vasculitis was described beginning 24 h after initiating mercuric chloride and prior to caecal entry of T cells. The authors performed a histological study during the first 93 hours following mercuric chloride challenge defining the precise relationship between gut mast cell degranulation and appearing caecal vasculitis. Their findings indicate a direct correlation between mast cell degranulation and early caecal vasculitis following the initial mercuric chloride challenge.
B lymphocyte signalling established by the CD19/CD22 loop and autoimmunity in the tight-skin mouse Systemic sclerosis is characterized by fibrosis and autoantibody production. Peripheral blood B cells from systemic sclerosis patients specifically over express CD19, a critical cell-surface signal transduction molecule in B cells. Asano et al. (Am J Pathol 2004;165:641) analyzed transgenic mouse lines that over express CD19. Their findings indicate that 20% increase of CD19 expression in mice spontaneously induced systemic sclerosis-specific anti-DNA topoisomerase I antibody production, which was further augmented by 200% over expression. However, no associated skin thickness increase was found. The molecular mechanisms for abnormal CD19 signaling were assessed by the authors: B-cell antigen receptor cross linking induced exaggerated calcium responses and augmented activation of extra cellular signalregulated kinase in TSK/+ B cells. In addition, CD22 function was specifically impaired in TSK/+ B cells. Consistently, CD19, a major target of CD22-negative regulation, was hyperphosphorylated in TSK/+ B cells. Therefore, reduced inhibitory signal provided by CD22 results in abnormal activation of signaling pathways including CD19 in TSK/+ mice; this disrupted B cell signaling contribute to specific autoantibody production.