The Journal of Immunology
Editing Autoreactive TCR Enables Efficient Positive Selection1 Fabio R. Santori, Ivica Arsov,2 Mirjana Lili, and Stanislav Vukmanovic´3 Allelic exclusion is inefficient at the TCR␣ locus, allowing a sizeable portion of T cells to carry two functional TCRs. The potential danger of dual TCR expression is a rescue of autoreactive TCRs during selection in the thymus and subsequent development of autoimmunity. In this study, we examine the reason(s) for replacing an autoreactive TCR and for allowing the survival of cells carrying two TCRs. We compared development of TCR transgenic CD4ⴙCD8ⴚ thymocytes in the presence or absence of MHC class II autoantigen that does not induce deletion of thymocytes. Contrary to the expected negative effect of the presence of autoantigen, ⬃100% more CD4ⴙCD8ⴚ thymocytes were found in the presence of MHC class II autoantigen than in the neutral background. A further increase in the strength of autoantigenic signal via expression of a human CD4 transgene led to an additional increase in the numbers of CD4ⴙCD8ⴚ thymocytes. Thus, editing autoreactive TCR results in more efficient positive selection, and this may be both a reason and a reward for risking autoimmunity. The Journal of Immunology, 2002, 169: 1729 –1734.
R
earrangement of the TCR loci is activated by the coordinate action of recombination-activating gene (RAG)4-1 and RAG-2 molecules and is characterized by random joining of one V, (D), and J gene elements (reviewed in Refs. 1 and 2). The TCR locus is rearranged at a developmentaly earlier stage than the TCR␣ locus. Expression of functional TCR chain enables immature CD4⫺CD8⫺TCR⫺ thymocytes to pass through a developmental checkpoint and acquire CD4 and CD8 coreceptors (3, 4). TCR expression also results in temporary RAG downmodulation (5) and in allelic exclusion, that is, inaccessibility of the second germline TCR locus to recombination (6). At the subsequent stage of thymocyte differentiation (CD4⫹CD8⫹TCR⫺), the TCR␣ locus becomes accessible for recombination, and a second wave of RAG expression is induced (5). Productive TCR␣ rearrangement results in expression of low levels of cell surface TCR. The resulting CD4⫹CD8⫹TCRlow population of thymocytes is then subject to two selection processes, both based on the interaction of their TCR with self MHC/peptide complexes. Positive selection is a result of relatively weak TCR engagement that delivers survival and differentiation signals to only a fraction of total CD4⫹CD8⫹ thymocytes (7). Failure to express the TCR or expression of TCR(s) not reactive with self peptide/MHC complexes leads to CD4⫹CD8⫹ thymocyte death by default (8). Thymocytes expressing TCRs with potentially harmful reactivity with self peptide/MHC (characterized by relatively strong interactions) must be silenced or eliminated. Negative selection can be achieved by
Michael Heidelberger Division of Immunology, Department of Pathology and Kaplan Cancer Center, New York University School of Medicine, New York, NY 10016 Received for publication September 27, 2001. Accepted for publication June 6, 2002. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by the National Institutes of Health Grant AI41573 (to S.V.), and National Cancer Institute Core Support Grant 5P30 CA16087. 2 Current address: Department of Microbiology, College of Physicians and Surgeons, Columbia University, New York, NY 10032. 3
Address correspondence and reprint requests to Dr. Stanislav Vukmanovic´, Michael Heidelberger Division of Immunology, Department of Pathology, New York University School of Medicine, 550 First Avenue, New York, NY 10016. E-mail address:
[email protected] 4 Abbreviations used in this paper: RAG, recombination-activating gene; hCD4, human CD4.
Copyright © 2002 by The American Association of Immunologists, Inc.
physical deletion or by a variety of nondeletional mechanisms, including down-modulating the levels of coreceptor molecules, rendering cells anergic, or raising the activation thresholds in T cells (reviewed in Ref. 9). Unlike in the TCR locus, allelic exclusion at the TCR␣ locus is almost nonexistent (10). In fact, rearrangement at the TCR␣ locus continues until signals for positive selection terminate RAG-1 and RAG-2 expression (11, 12). Residual levels of RAG-1 and RAG-2 from the first wave of expression are sufficient to initiate recombination at the TCR␣ locus (13). However, in the absence of additional RAG expression, rearrangement is limited to the 5⬘ portion of the J␣ cluster of one locus (13). Thus, a secondary wave of RAG expression appears to be initiated with the purpose of extending recombination to the 3⬘ portion of the J␣ cluster in the rearranged TCR␣ locus and to the second TCR␣ allele. Dual TCR␣ expression at the cell surface is relatively common in immature thymocytes, but only positively selected TCR␣ is expressed in mature thymocytes (14, 15). The primary (nonselectable) TCR␣ is subject to internalization and intracellular retention (14, 15). Thus, a decision on TCR␣ cell surface expression is made based on TCR␣ engagement during positive selection. Allelic inclusion of the TCR␣ locus carries the danger of potential reassembly of an autoreactive receptor that may lead to autoimmune reactions (16 –18). A special case of allelic inclusion is termed receptor editing, in which interaction of the primary autoreactive receptor with Ag triggers secondary rearrangement (19). Receptor editing has been first described (20 –22) and subsequently extensively studied in B lymphocytes (reviewed in Ref. 23), and only later found to operate in autoreactive peripheral T cells (24) and thymocytes (18, 25). It is unclear why the risk of autoimmunity is taken by the immune system when safer mechanism(s) of elimination/silencing of autoreactive lymphocytes exists and could be used. In this study, we demonstrate that risk of autoimmunity by TCR editing may be rewarded by more efficient positive selection.
Materials and Methods Experimental animals C57BL/6 (B6), RAG2⫺/⫺, and H-Y TCR transgenic mice were purchased from Taconic Farms (Germantown, NY). B10.BR (BR), B10.A(2R) (2R), and B10.A(5R) (5R) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). hCD4 transgenic mice on mouse CD4⫺/⫺ background 0022-1767/02/$02.00
1730 were a kind gift from D. R. Littman (New York University School of Medicine). H-Y TCR transgenic B6 mice were bred for two generations with 2R mice to produce the H-Y/2R strain. Littermates were screened by immunofluorescence of peripheral blood cells using (AF6-88.5.3) anti-H2Kb mAb that does not cross-react to the H-2k haplotype. To produce H-Y transgenic, RAG2⫺/⫺, hCD4 transgenic B6 strain, we bred B10.D2/H-Y/ RAG2⫺/⫺ males with hCD4 transgenic females. hCD4-positive F1 offsprings were intercrossed, and littermates were selected by immunofluorescent staining for both human CD4 (hCD4). Anti-H-2Kb (AF6-88.5.3) and anti-H-2Kd (SF1.1.1) mAbs were used to identify the H-2d MHC haplotype, anti-hCD4 were used to screen for hCD4 or mCD4, respectively, while anti-B220 Ab was used to determine RAG2 deficiency.
Thymic epithelium grafting Neonatal B6 or 2R thymi were cultured for 5 days on sponge-supported filters in RPMI medium supplemented with 10% FCS, 50 M 2-ME, and 1.35 mM deoxyguanosine, as described (26). After 5 days of culture, thymi were placed under the left (B6) or right (2R) kidney capsule of B10.D2/ H-Y/RAG2⫺/⫺ mice. Mice were allowed to recover and after 4 wk were sacrificed for analysis.
Cell lines and proliferation Generation and maintenance of the CD4⫹ T cell line from H-Y TCR transgenic mice were described previously (27). T cells (2 ⫻ 105) were incubated with irradiated stimulator spleen cells (1 ⫻ 106) in round-bottom 96-well plates for 48 h. Each microculture was then pulsed with 0.5 Ci [3H]thymidine for 16 h, and thymidine incorporation was measured on a beta scintillation counter.
Flow cytometry Ab against the transgenic TCR␣ chain (T3.70) was used as a hybridoma supernatant. Anti-mouse CD4 (H129.19) conjugated to PE, CyChromeconjugated anti-mouse CD8␣ (53-6.7), and FITC-conjugated anti-hCD4 were purchased from BD PharMingen (Costa Mesa, CA). Thymocytes were incubated on ice for 30 min with three fluorochrome-conjugated Abs before washing in PBS (2% FCS). Cells were then fixed in 1% paraformaldehyde and analyzed using a FACScan flow cytometer (BD Biosciences, Mountain View, CA). When T3.70 Ab was used, it was followed by multiply adsorbed FITC-conjugated anti-mouse Ig (BD PharMingen).
Results H-Y TCR transgenic T cells recognize H-2Ab in vitro and in vivo Peripheral CD8⫹ T cells carrying the H-Y-specific transgenic TCR (28) recognize the peptide Smcy738 –746 presented by the H-2Db MHC class I molecule (29). The same transgenic TCR is also reactive to H-2Ab (27). Although the presence of H-2Db-associated Ag in the thymus leads to significant deletion of CD4⫹CD8⫹ thymocytes (30, 31), presence of H-2Ab has relatively mild effects (27). Thus, the H-Y TCR transgenic mouse provides a good experimental model to study escape of cells bearing autoreactive TCR. To examine how overt reactivity of the transgenic TCR to H-2Ab might affect the maturation of CD4⫹CD8⫺ thymocytes, we searched for nonstimulatory MHC class II alleles, so that we could compare CD4⫹CD8⫺ thymocyte maturation in the neutral and autoreactive backgrounds. Consistent with our previous findings (27), a CD4⫹ cell line isolated from H-Y TCR transgenic mice proliferated upon stimulation with irradiated wild-type B6 (H-2b), but not upon stimulation with MHC class II-deficient (H-2A⫺/⫺) spleen cells (Fig. 1). C3H/HeJ (H-2k) spleen cells were also unable to stimulate proliferation, indicating that the TCRs expressed by this cell line do not cross-react with H-2k class II molecules. To determine the in vivo reactivity of the transgenic TCR with H-2k or H-2b class II molecules, we assessed the relative numbers of male H-Y TCR transgenic CD4⫹CD8⫹ thymocytes in B6 or 2R backgrounds. The effect of H-2Ab is mild and is more pronounced in males in which a substantial fraction of CD4⫹CD8⫹ thymocytes is already removed by encounter with the H-Y Ag (27). H-2 recombinant 2R mice were used because they carry H-2k class II alleles and H-2Db (see Table I). The presence of H-2Db is required
TCR AUTOREACTIVITY ENHANCES POSITIVE SELECTION
FIGURE 1. Proliferative response of CD4⫹ T cells from H-Y TCR transgenic mouse to distinct MHC class II alleles. The HYCD4 cell line was tested for proliferative response to irradiated stimulator cells (1 ⫻ 106/well) of wild-type B6, MHC class II-deficient B6, or C3H/HeJ haplotypes. After 2 days of culture, proliferation was determined by tritiated thymidine incorporation. The results represent mean ⫾ SD of triplicate cultures.
for deletion induced by the male Ag. Spleen cells from 2R mice did not stimulate H-Y CD4⫹ cells (data not shown). Male H-Y TCR transgenics in the 2R background harbored twice as many CD4⫹CD8⫹ thymocytes compared with the B6 strain (Fig. 2A), suggesting that the presence of H-2Ab leads to incomplete negative selection. Taken together, the above results demonstrate that H-2Ab, but not product of any of the H-2k MHC class II loci, stimulate proliferation of peripheral H-Y TCR transgenic CD4⫹ cells in vitro and induce negative selection of H-Y CD4⫹CD8⫹ thymocytes in vivo. H-2Ab induces temporary developmental arrest of H-Y TCR transgenic thymocytes Reduced numbers of CD4⫹CD8⫹ thymocytes could result from physical deletion, or by clonal arrest (32). Apoptosis of CD4⫹CD8⫹ H-Y TCR transgenic thymocytes could be readily induced in suspension cultures by H-2Db-presented male Ag, but not by self H-2Ab autoantigen (data not shown), arguing against the physical deletion. Development of H-Y thymocytes is slower in the presence of H-2Ab relative to the MHC class II-deficient background (33), indicating that indeed, H-2Ab may induce clonal arrest. To address this issue directly, female B10.D2/H-Y/ RAG2⫺/⫺ mice were grafted with female B6 and 2R thymic epithelium. Four weeks later, there were twice as many CD4⫺CD8⫺ thymocytes in the B6 grafts than in the 2R grafts of the same hosts (Fig. 3), indicating temporal arrest in thymic development. In addition, there was 3- to 4-fold less CD4⫺CD8⫹ thymocytes in B6 grafts, most likely reflecting a cumulative effect of fewer CD4⫹CD8⫹ cells auditioning for selection due to the arrest at an earler stage. Some CD4⫹CD8⫺ thymocytes were present in both B6 and 2R grafts, but they were not fully mature as they expressed low levels of CD8 molecules (for example, mean CD8 fluorescence of CD4⫹CD8⫺ cells was 52 compared with 26 in CD4⫺CD8⫺ thymocytes). Thus, our data suggest that developmental arrest appears to be the only mechanism of H-2Ab-induced negative selection. Down-modulation of transgenic TCR␣ chains in CD4⫹CD8⫺ thymocytes One of the most immediate consequences of positive selection is up-regulation of cell surface TCR levels. When thymocytes express two TCRs, the up-regulation is limited to the TCR that promoted positive selection (14). Therefore, up-regulation of a TCR is an indicator of the TCR used to promote positive selection. In the H-Y TCR transgenic mice, CD4⫺CD8⫹ thymocytes express relatively high levels of transgenic TCR␣ chain. In contrast, few
The Journal of Immunology
1731
Table I. Efficiency of H-Y TCR transgenic CD4⫹CD8⫺ thymocyte selection in mice with 2R or B6 genetic backgrounds % Thymocytesa
MHC Alleles Breeding Strategy
Genetic Background
K
A
E
D
1
2R B6 2R B6 2R B6 F1(2R⫻D2) F1(B6⫻D2) F1(2R⫻D2) F1(B6⫻D2) 2R F1(2R⫻B6) F1(BR⫻D2) F1(5R⫻D2)
k b k b k b k/d b/d k/d b/d k k/b k/d b/d
k b k b k b k/d b/d k/d b/d k k/b k/d b/d
k ⫺ k ⫺ k ⫺ k/d ⫺/d k/d ⫺/d k k/⫺ k/d k/d
b b b b b b b/d b/d b/d b/d b b/b k/d k/d
2
3 4
L
TCRtg/ Gender
Mice/ Group
CD4⫹8⫹
CD4⫹8⫺
CD4⫹8⫺/ CD4⫹8⫹
⫺/⫺ ⫺/⫺ ⫺/⫺ ⫺/⫺ ⫺/⫺ ⫺/⫺ ⫺/d ⫺/d ⫺/d ⫺/d ⫺/⫺ ⫺/⫺ ⫺/d ⫺/d
⫹/F ⫹/F ⫺/F ⫺/F ⫹/M ⫹/M ⫹/F ⫹/F ⫹/M ⫹/M ⫹/F ⫹/F ⫺/F ⫹/F
4 3 10 3 3 3 4 4 2 1 2 3 13 7
57.5 (3.3) 58.0 (1.0) 77.7 (1.3) 83.8 (4.7) 13.4 (2.7) 6.1 (1.3) 67.7 (2.0) 66.2 (6.2) 22.1 9.4 66.3 58.8 (2.2) 41.5 (1.4) 36.7 (2.8)
6.0 (1.0) 11.0 (1.4) 8.3 (0.4) 9.5 (2.4) 7.0 (1.1) 7.3 (0.5) 6.4 (0.5) 13.1 (2.0) 6.8 6.3 8.0 11.2 (1.6) 11.4 (0.4) 15.6 (1.0)
0.1043 0.1897 0.1068 0.1134 0.5224 1.1967 0.0945 0.1978 0.3077 0.6702 0.1207 0.1905 0.2747 0.4251
Enhancement of Selection (%)b
81.9 6.2 129.1 110.3 117.8 57.8 54.8
a
Shown are means (SE), except for the groups with less than three mice, in which only means are provided. Enhancement of CD4⫹CD8⫺ thymocyte selection in H-2Ab⫹ vs H-2Ab⫺ backgrounds was calculated as H-2Ab⫹ selection ratio minus H-2Ab⫺ selection ratio, divided by H-2Ab⫺ selection ratio and multiplied by 100. b
CD4⫹CD8⫺ thymocytes express detectable cell surface transgenic TCR␣ (11). This fact, together with the presence of CD4⫺CD8⫹, and absence of CD4⫹CD8⫺ thymocytes in rearrangement-deficient backgrounds (34), indicates that positive selection of these thymocyte subsets is mediated by transgenic or endogenous TCR␣ chains, respectively. To determine which TCR␣ is used for selection of CD4⫹CD8⫺ thymocytes in 2R background, we compared the levels of transgenic TCR␣ expressed by this thymocyte subset in 2R vs B6 backgrounds. The pattern of transgenic TCR␣ expression in CD4⫹CD8⫺ vs CD4⫺CD8⫹ thymocytes in 2R and B6 backgrounds is virtually identical (Fig. 4), suggesting that endogenous TCR␣ rearrangements allow selection of CD4⫹CD8⫺ thymocytes in the 2R background as well. The expression of transgenic TCR chain was high in all thymocyte subsets in both genetic backgrounds (data not shown), indicating that allelic exclusion at the TCR locus was not affected by H-2Ab. Efficient positive selection of CD4⫹CD8⫺ H-Y TCR transgenic thymocytes by H-2Ab
genic TCR to H-2Ab present a unique opportunity to directly compare maturation of CD4⫹CD8⫺ thymocytes under conditions of allelic inclusion (2R background) with those of replacing the autoreactive TCR (B6 background). Three breeding strategies were used to compare the numbers of CD4⫹CD8⫺ thymocytes in B6 and 2R backgrounds. To our surprise, the numbers of CD4⫹CD8⫺ thymocytes were reduced in neutral 2R background, relative to the autoantigenic B6 background (Fig. 2B; Table I). This was not due to the influence of genetic backgrounds, as the numbers of CD4⫹CD8⫺ thymocytes in TCR nontransgenic B6 or 2R mice are similar (Table I). The presence of a single H-2Ab allele was sufficient to significantly increase selection of CD4⫹CD8⫺ thymocytes in the 2R background (Table I, breeding strategy 3). The difference between the 2R and B6 background was also evident when absolute numbers of CD4⫹CD8⫺ thymocytes were calculated (Fig. 5). Thus, H-2Ab induces more efficient positive selection of H-Y TCR transgenic CD4⫹CD8⫺ thymocytes than H-2Ak and H-2Ek combined.
Selection of CD4⫹CD8⫺ thymocytes by endogenous TCR␣ chains in both B6 and 2R backgrounds and reactivity of the H-Y trans-
FIGURE 2. Comparison of H-Y TCR transgenic thymocyte development in B6 or 2R backgrounds. CD4 vs CD8 plots of thymocytes isolated from male (A) or female (B) H-Y TCR transgenic mice.
FIGURE 3. Arrested development of CD4⫹CD8⫹ H-Y TCR transgenic thymocytes in B6 thymic epithelium grafts. B10.D2/RAG2⫺/⫺/H-Y TCR transgenic females were grafted with dGuo-treated B6 or 2R neonatal thymi. Same animals received both grafts under different kidney capsules. Four weeks after grafting, thymocytes were isolated from the grafts and analyzed for CD4 and CD8 expression by flow cytometry.
1732
TCR AUTOREACTIVITY ENHANCES POSITIVE SELECTION hCD4 expression enhances endogenous TCR␣-mediated selection of CD4⫹CD8⫺ H-Y TCR transgenic thymocytes In the absence of endogenous TCR␣ rearrangements, no CD4⫹CD8⫺ H-Y TCR transgenic thymocytes are selected (34), suggesting that the affinity/avidity of the transgenic TCR is either below or above the threshold required for positive selection. To address this question, we have determined the effect of hCD4 coreceptor expression (35). hCD4 interacts functionally with mouse MHC class II equally well as the mouse CD4 (36). We reasoned that bringing extra p56lck molecules into the TCR recognition cluster via additional CD4 should lead to a stronger signal. Although the expression of mouse CD4 is somewhat reduced in hCD4 transgenic mice, the overall CD4 levels (mouse plus human) are functionally higher than the levels of mouse CD4 in wild-type mice. This is suggested by more pronounced reduction of H-Y TCR transgenic CD4⫹CD8⫹ thymocytes in male mice expressing the hCD4 (Fig. 6A). Identical results were obtained with mice bred to the RAG2⫺/⫺ background (Fig. 6B). If the affinity/avidity of transgenic TCR is below the threshold for positive selection, then we would predict that an increase in the signal strength delivered by hCD4 might enable promotion of female CD4⫹CD8⫺ H-Y TCR transgenic thymocytes in the absence
FIGURE 4. Cell surface expression of transgenic TCR␣ on thymocyte subsets from female B6 or 2R H-Y TCR transgenic mice. Thymocytes isolated from female H-Y TCR transgenic mice of either B6 or 2R background were labeled with anti-CD4, anti-CD8, and T3.70 mAbs and analyzed by flow cytometry. The histograms represent comparison of transgenic TCR␣ expression on gated CD4⫹CD8⫹, CD4⫺CD8⫹, or CD4⫹CD8⫺ thymocyte populations from the two backgrounds.
A fourth breeding strategy was used to determine how the additional requirement for endogenous TCR␣ rearrangement for selection of the CD4⫺CD8⫹ lineage might affect the influence of H-2Ab on selection of CD4⫹CD8⫺ thymocytes. H-Y B10.D2 mice were bred to either B10.A(5R) or B10.BR strain (Table I). Neither of the resulting F1 offsprings expresses H-2Db, a selecting element for the transgenic TCR␣. Under these conditions, enhancement of CD4⫹CD8⫺ thymocytes by H-2Ab was still evident, but was ⬃50% less pronounced than in the presence of H-2Db. This is probably due to competition of H-2Db for engaging endogenous TCR␣ chains.
FIGURE 5. Absolute number of CD4⫺CD8⫹ or CD4⫹CD8⫺ H-Y TCR transgenic thymocytes in the absence or presence of H-2Ab. Percentage of CD4⫺CD8⫹ or CD4⫹CD8⫺ thymocytes from female F1(2R ⫻ D2) or F1(B6 ⫻ D2) offsprings (see Table I) was multiplied by individual total thymus cell numbers. Shown are mean and SEs.
FIGURE 6. The effects of hCD4 transgene expression on development of H-Y TCR transgenic thymocytes in the B6 background. Flow cytometry analysis of thymocytes isolated from male (A and B) or female (C and D) RAG2⫹/⫹ (A and C) or RAG2⫺/⫺ (B and D) H-Y TCR transgenic mice. Shown are CD4 vs CD8 plots of wild-type vs hCD4 backgrounds.
The Journal of Immunology
1733
of endogenous TCR␣ rearrangements. hCD4 exerted such an effect in RAG2⫹/⫹ background, suggesting that it can rescue selection of wide number of TCRs (Fig. 6C). However, this effect was exclusive for endogenously rearranged TCR␣ chains, because hCD4 did not alter the selection of H-Y TCR transgenic thymocytes in RAG2⫺/⫺ background (Fig. 6D). Expression of hCD4 also induced increased accumulation of CD4⫺CD8⫺ thymocytes in both RAG2⫹/⫹ and RAG2⫺/⫺ females, making the developmental thymocyte arrest by H-2Ab evident even at steady state levels. Taken together, hCD4-mediated increase in the strength of signals delivered by H-2Ab was effective to enable selection of additional endogenously rearranged TCR␣(s), but not of the transgenic TCR␣. This, together with the agonist activity of the H-2Ab for the H-Y TCR (Fig. 1), suggests that the affinity/avidity of this TCR for H-2Ab is above the threshold for positive selection.
Discussion The H-Y transgenic TCR reacts with H-2Ab sufficiently strongly to induce proliferation and/or cytokine secretion of peripheral CD4⫹ T cells (27). The autoantigen is present in the thymus, and thymocytes interact with it before positive selection. This interaction, however, does not result in positive selection, as evidenced by absence of CD4⫹CD8⫺ thymocytes in RAG-2-deficient background. Therefore, the signal delivered by H-2Ab is either too weak, or too strong for positive selection. Given that signal strength delivered by the same Ag is sufficient to induce proliferation of peripheral T cells, and that increasing the selecting signal strength via hCD4 expression failed to rescue CD4⫹CD8⫹ thymocytes when endogenous rearrangements were prevented, we believe that the latter possibility (signal too strong) is a more likely explanation for absence of positive selection by the H-2Ab. One of the characteristics of receptor editing is RAG-1/2 re-expression in the cell population interacting with the autoantigen. However, CD4⫹CD8⫹ thymocytes that encounter H-2Ab already express high levels of RAG-1/2 as part of the regular developmental program (5). Thus, direct demonstration of this aspect of receptor editing in this model is not possible. However, requirement for endogenous TCR␣ for CD4⫹CD8⫺ thymocyte selection in B6 background, reactivity of the transgenic TCR for H-2Ab, and absence of CD4⫹CD8⫺ thymocyte selection in RAG-2⫺/⫺/hCD4 transgenic background collectively argue that endogenous TCR␣ rearrangement is due to receptor editing. Replacement of the transgenic TCR by endogenously rearranged TCR␣ chains occurred in both B6 and 2R CD4⫹CD8⫺ thymocytes. However, in the B6 background, in which CD4⫹CD8⫹ thymocytes in addition encounter the autoantigen, CD4⫹CD8⫺ thymocytes are more abundant than in a neutral (2R) environment (Table I), suggesting that receptor editing results in relatively more efficient positive selection. What is the possible mechanism of efficient positive selection? The most likely explanation is that TCR itself contributes substantial avidity for H-2Ab, reaching almost the threshold required for positive selection (Fig. 7). Consequently, a wide range of endogenous TCR␣ chains can raise the total avidity of the TCR to fit between the thresholds required for positive and negative selection. The transgenic TCR␣ and occasional endogenous TCR␣ chains can raise the avidity of the TCR over the threshold for negative selection. In the case of H-2Ak/H-2Ek class II molecules, only a narrow range of endogenous TCR␣ chains can contribute to the low avidity of transgenic TCR to rescue selection. Presence of the hCD4 lowers thresholds for both positive and negative selection. In this case, the number of TCRs transferred from nonselectable to the selectable range of avidities is greater than the number of selectable TCRs eliminated because they entered the range of avidities sufficient for
FIGURE 7. Model that explains relative efficiencies of TCR repertoire selection by receptor editing vs allelic inclusion. See text for detailed explanation.
negative selection. Hence, the net result of lowering both thresholds is a gain in the number of selected TCRs. Interestingly, numbers of CD4⫺CD8⫹ thymocytes in RAG2⫺/⫺ background are lower than in the presence of RAG2 (Fig. 6). This could be explained by two possibilities. First, the peripheral lymphoid system of H-Y TCR transgenic RAG2⫺/⫺ mice is lymphopenic (data not shown). This may or may not be due to inability of H-Y TCR transgenic cells to undergo homeostatic expansion (37). A lymphopenic peripheral environment might in turn signal the thymus to more rapidly export mature T cells (38). An alternative explanation for the lower numbers might be suggested by recent study that demonstrated rescue of transgenic TCR selection by endogenous TCR␣ chains (39). Thus, endogenous TCR␣ chains could enhance selection of transgenic TCR by H-2Db. At the moment, neither possibility can be excluded with certainty, but given the up-regulation of the transgenic TCR␣ in CD4⫺CD8⫹ thymocytes (Fig. 4), we consider the first possibility more likely. In 1959, Burnet (40) proposed a clonal selection theory that attempted to explain the specificity of induction of immune responses, and mechanisms of nonresponsiveness to self. A central tenet of this widely accepted theory was that lymphocytes may express only one Ag receptor to assure the specificity of the immune response strictly to Ags. Expression of two receptors was postulated to be especially dangerous if one of the receptors was autoreactive, because activation through nonautoreactive receptor could lead to the expansion of autoreactive cells. The phenomenon of repressing expression of Ag receptor encoded by the second allele was named allelic exclusion. Although the clonal selection theory was shown to be correct in many aspects, several exceptions to the concept of “one lymphocyte-one receptor” have been described, obliging modifications of the original hypothesis to account for these observations. Even though allelic exclusion was shown to operate at the Ig H chain and the TCR loci, thanks to the inclusion of the Ig L chains and the TCR␣ loci, a portion of both B and T lymphocytes expresses two Ag receptors (reviewed in Ref. 23). That observation prompted testing Burnett’s hypothesis that lymphocytes carrying two receptors would be capable of triggering autoimmunity. Indeed, autoreactive TCRs could escape negative selection because of poor autoantigen presentation (16, 18), or because of inadequate signal strength due to low density of the autoreactive receptor (17). Thus, the immune system appears to
1734 risk development of autoimmunity for the benefit of efficient positive selection. T cells responsive to autoantigens implicated in autoimmune disorders can be isolated from healthy individuals (41). It is therefore important to distinguish autoreactivity (reactivity of immune receptors with components of self) from autoimmunity (destructive immune response directed at cells expressing components of self). Different checkpoints need to be passed to develop autoimmunity (42). The mechanisms that normally prevent autoreactive lymphocytes to become autodestructive include, but are not limited to: regulatory cells, tissue-specific expression and balance between expression of activating and inhibitory costimulatory molecules, secretion of immunomodulatory cytokines (such as TGF or IL10), and Fas-Fas ligand interaction. Because of these multiple checkpoints that prevent the development of autoimmunity, and a significant benefit of efficient TCR repertoire selection, editing autoreactive TCRs must be considered safe relative to the advantages it confers to the host.
Acknowledgments We thank John Hirst for FACS analysis, Dan Littman for providing hCD4 transgenic mouse strain, Nigel Kileen for helpful discussions, and Alan Frey for reading the manuscript.
References 1. Sadofsky, M. J. 2001. The RAG proteins in V(D)J recombination: more than just a nuclease. Nucleic Acids Res. 29:1399. 2. Kim, D. R., S. J. Park, and M. A. Oettinger. 2000. V(D)J recombination: sitespecific cleavage and repair. Mol. Cell 10:367. 3. Mombaerts, P., A. R. Clarke, M. A. Rudnicki, J. Iacomini, S. Itohara, J. J. Lafaille, L. Wang, Y. Ichikawa, R. Jaenisch, M. L. Hooper, and S. Tonegawa. 1992. Mutations in T-cell antigen receptor genes ␣ and  block thymocyte development at different stages. Nature 360:225. 4. Shinkai, Y., S. Koyasu, K.-i. Nakayama, K. M. Murphy, D. Y. Loh, E. L. Reinherz, and F. W. Alt. 1993. Restoration of T cell development in RAG2-deficient mice by functional TCR transgenes. Science 259:822. 5. Wilson, A., W. Held, and H. R. MacDonald. 1994. Two waves of recombinase gene expression in developing thymocytes. J. Exp. Med. 179:1355. 6. Uematsu, Y., S. Ryser, Z. Dembic, P. Borgulya, P. Krimpenfort, A. Berns, H. von Boehmer, and M. Steinmetz. 1988. In transgenic mice the introduced functional T cell receptor  gene prevents expression of endogenous  genes. Cell 52:831. 7. Jameson, S. C., K. A. Hogquist, and M. J. Bevan. 1995. Positive selection of thymocytes. Annu. Rev. Immunol. 13:93. 8. Surh, C. D., and J. Sprent. 1994. T-cell apoptosis detected in situ during positive and negative selection in the thymus. Nature 372:100. 9. Vukmanovic, S. 1996. The molecular jury: deciding whether immature thymocytes should live or die. J. Exp. Med. 184:305. 10. Malissen, M., J. Trucy, E. Jouvin-Marche, P.-A. Cazenave, R. Scollay, and B. Malissen. 1992. Regulation of TCR a and b gene allelic exclusion during T-cell development. Immunol. Today 13:315. 11. Borgulya, P., H. Kishi, Y. Uematsu, and H. von Boehmer. 1992. Exclusion and inclusion of ␣ and  T cell receptor alleles. Cell 69:529. 12. Brandle, D., C. Muller, T. Rulicke, H. Hengartner, and H. Pircher. 1992. Engagement of the T cell receptor during positive selection in the thymus downregulates RAG-1 expression. Proc. Natl. Acad. Sci. USA 89:9529. 13. Yannoutsos, N., P. Wilson, W. Yu, H. T. Chen, A. Nussenzweig, H. Petrie, and M. C. Nussenzweig. 2001. The role of RAG re-induction in thymocyte development in vivo. J. Exp. Med. 194:471. 14. Alam, S. M., and N. R. Gascoigne. 1998. Posttranslational regulation of TCR V␣ allelic exclusion during T cell differentiation. J. Immunol. 160:3883. 15. Boyd, R., I. Kozieradzki, A. Chidgey, H. W. Mittrucker, D. Bouchard, E. Timms, K. Kishihara, C. J. Ong, D. Chui, J. D. Marth, et al. 1998. Receptor-specific allelic exclusion of TCRV ␣-chains during development. J. Immunol. 161:1718.
TCR AUTOREACTIVITY ENHANCES POSITIVE SELECTION 16. Zal, T., S. Weiss, A. Mellor, and B. Stockinger. 1996. Expression of a second receptor rescues self-specific T cells from thymic deletion and allows activation of autoreactive effector function. Proc. Natl. Acad. Sci. USA 93:9102. 17. Sarukhan, A., C. Garcia, A. Lanoue, and H. von Boehmer. 1998. Allelic inclusion of T cell receptor ␣ genes poses an autoimmune hazard due to low-level expression of autospecific receptors. Immunity 8:563. 18. McGargill, M. A., J. M. Derbinski, and K. A. Hogquist. 2000. Receptor editing in developing T cells. Nat. Immun. 1:336. 19. Fink, P. J. 2000. If at first you don’t succeed. Nat. Immun. 1:271. 20. Tiegs, S. L., D. M. Russell, and D. Nemazee. 1993. Receptor editing in selfreactive bone marrow B cells. J. Exp. Med. 177:1009. 21. Gay, D., T. Saunders, S. Camper, and M. Weigert. 1993. Receptor editing: an approach by autoreactive B cells to escape tolerance. J. Exp. Med. 177:999. 22. Radic, M. Z., J. Erikson, S. Litwin, and M. Weigert. 1993. B lymphocytes may escape tolerance by revising their antigen receptors. J. Exp. Med. 177:1165. 23. Nussenzweig, M. C. 1998. Immune receptor editing: revise and select. Cell 95: 875. 24. McMahan, C. J., and P. J. Fink. 1998. RAG reexpression and DNA recombination at T cell receptor loci in peripheral CD4⫹ T cells. Immunity 9:637. 25. Wang, F., C. Y. Huang, and O. Kanagawa. 1998. Rapid deletion of rearranged T cell antigen receptor (TCR) V␣-J␣ segment by secondary rearrangement in the thymus: role of continuous rearrangement of TCR ␣ chain gene and positive selection in the T cell repertoire formation. Proc. Natl. Acad. Sci. USA 95:11834. 26. Jenkinson, E. J., L. L. Franchi, R. Kingston, and J. J. T. Owen. 1982. Effect of deoxyguanosine on lymphopoiesis in the developing thymus rudiment in vitro: application of the production of chimeric thymus rudiments. Eur. J. Immunol. 12:583. 27. Arsov, I., and S. Vukmanovic. 1998. Dual MHC class I and class II restriction of the single TCR: distinct modes of tolerance induction by two classes of autoantigens. J. Immunol. 162:2008. 28. Von Boehmer, H. 1990. Developmental biology of T cells in T cell receptor transgenic mice. Annu. Rev. Immunol. 8:531. 29. Markiewicz, M. A., C. Girao, J. T. Opferman, J. Sun, Q. Hu, A. A. Agulnik, C. E. Bishop, C. B. Thompson, and P. G. Ashton-Rickardt. 1998. Long-term T cell memory requires the surface expression of self-peptide/major histocompatibility complex molecules. Proc. Natl. Acad. Sci. USA 95:3065. 30. Teh, H. S., P. Kisielow, B. Scott, H. Kishi, Y. Uematsu, H. Bluthmann, and H. von Boehmer. 1988. Thymic major histocompatibility complex antigens and the ␣ T-cell receptor determine the CD4/CD8 phenotype of T cells. Nature 335:229. 31. Kisielow, P., H. Bluthmann, U. D. Stearz, M. Steimetz, and H. von Boehmer. 1988. Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4⫹8⫹ thymocytes. Nature 333:742. 32. Swat, W., H. von Boehmer, and P. Kisielow. 1994. Central tolerance: clonal deletion or clonal arrest? Eur. J. Immunol. 24:485. 33. Lacorazza, H. D., C. Tucek-Szabo, L. V. Vasovic, K. Remus, and J. Nikolich-Zugich. 2001. Premature TCR ␣ expression and signaling in early thymocytes impair thymocyte expansion and partially block their development. J. Immunol. 166:3184. 34. Scott, B., H. Bluthmann, H.-S. Teh, and H. von Boehmer. 1989. The generation of mature T cells requires interaction of the ␣ T-cell receptor with major histocompatibility antigens. Nature 338:591. 35. Killeen, N., S. Sawada, and D. R. Littman. 1993. Regulated expression of human CD4 rescues helper T cell development in mice lacking expression of endogenous CD4. EMBO J. 12:1547. 36. Konig, R., L. Y. Huang, and R. N. Germain. 1992. MHC class II interaction with CD4 mediated by a region analogous to the MHC class I binding site for CD8. Nature 356:796. 37. Ernst, B., D.-S. Lee, J. M. Chang, J. Sprent, and C. D. Surh. 1999. The peptide ligands mediating positive selection in the thymus control T cell survival and homeostatic proliferation in the periphery. Immunity 11:173. 38. Scollay, R., and D. I. Godfrey. 1995. Thymic emigration: conveyor belts or lucky dips? Immunol. Today 16:268. 39. He, X., J. C. A. Janeway, M. Levine, E. Robinson, P. Preston-Hurlburt, C. Viret, and K. Bottomly. 2002. Dual receptor T cells extend the immune repertoire for foreign antigens. Nat. Immun. 3:127. 40. Burnet, F. M. 1959. The Clonal Selection Theory of Acquired Immunity. The University Press, Cambridge. 41. Parry, S. L., F. C. Hall, J. Olson, T. Kamradt, and G. Sonderstrup. 1998. Autoreactivity versus autoaggression: a different perspective on human autoantigens. Curr. Opin. Immunol. 10:663. 42. Andre, I., A. Gonzalez, B. Wang, J. Katz, C. Benoist, and D. Mathis. 1996. Checkpoints in the progression of autoimmune disease: lessons from diabetes models. Proc. Natl. Acad. Sci. USA 93:2260.
