Optimal T-cell receptor affinity for inducing

Sabrina Koehlia, Dieter Naehera, Virginie Galati-Fourniera, Dietmar Zehnb,c, and Ed Palmera,1

aDepartments of Biomedicine and Nephrology, University Hospital Basel and University of Basel, CH-4031 Basel, Switzerland; bSwiss Vaccine Research Institute, CH-1066 Epalinges, Switzerland; and cDivision of and , Department of Medicine, Lausanne University Hospital, 1011 Lausanne, Switzerland

Edited* by Philippa Marrack, Howard Hughes Medical Institute, National Jewish Health, Denver, CO, and approved October 14, 2014 (received for review February 17, 2014) T-cell receptor affinity for self- has an important role in them to differentiate into short-lived effector T cells (20). Below- establishing self-tolerance. Three transgenic mouse strains express- threshold T cells expand less efficiently but can still threaten the ing of variable affinity for the OVA transgenic-I T-cell host upon recognition of below-threshold antigen in an in- receptor were generated to address how TCR affinity affects the fectious environment (21). In fact, below-threshold T cells are efficiency of negative selection, the ability to prime an autoimmune indeed able to mediate autoimmunity after immunization with response, and the elimination of the relevant target cell. Mice recombinant Listeria monocytogenes (Lm) expressing a below- expressing antigens with an affinity just above the negative threshold antigen (22). selection threshold exhibited the highest risk of developing ex- Antigen affinity affects important parameters related to the perimental autoimmune diabetes. The data demonstrate that development of autoimmunity: (i) the efficiency of central toler- close to the affinity threshold for negative selection, sufficient ance, (ii) the efficiency of T-cell priming, and (iii) the efficiency of numbers of self-reactive T cells escape deletion and create an destroying a target cell expressing the self-antigen. In this study, we increased risk for the development of autoimmunity. examined these parameters individually and in combination. We show a striking threshold effect on central deletion, T-cell priming, Tcell| TCR | affinity | tolerance | autoimmunity and cytolysis of antigen-expressing cells in the target tissue. These experiments demonstrate that T cells expressing TCRs just above protective and self-tolerant T-cell repertoire is generated in the affinity threshold have the highest potential to induce an Athe (1–3), where negative selection reduces the autoimmune disease. number of self-reactive with autoimmune potential (4– Results 6). However, negative selection is not a perfect process, and a small β number of self-reactive T cells are found in the periphery (7). RIP-Variant Mice Express Variant OVA Proteins in Pancreatic Cells. Previous work from our laboratory defined the affinity threshold In RIP-OVA mice, the transgenic RIP drives OVA expression in β where negative selection is initiated (1, 8). Thymocytes expressing pancreatic Langerhans islet cells, proximal tubular epithelial MHC I restricted T-cell receptors (TCRs) undergoing negative cells in the kidney, mTECs in the thymus, and testes of male mice. For these studies, two different RIP-OVA lines were used: RIP- selection when binding a self-antigen with a Kd ≤ 6 μM and positive selection when binding a self-antigen with lower affinity (8, 9). sOVA and RIP-mOVA express the soluble and membrane forms The transcription factor ensures that of ovalbumin, respectively. We also generated three new strains tissue-restricted antigens are also expressed in the thymus within of RIP-variant transgenic mice expressing the mOVA variants: resident medullary thymic epithelial cells (mTECs) (10); how- Q4H7 (RIP-mQ4H7, below-threshold), T4 (RIP-mT4, thresh- ever, individual self-antigens are expressed on only a small per- old), and Q4R7 (RIP-mQ4R7, above-threshold). To determine whether OVA-variant antigens are expressed in pancreatic β cells, centage of mTECs, and the presentation of a single antigen is OT-I mice were crossed to the various RIP-OVA–variant mice to limited on the mTEC or surface (11–14). There- fore, self-reactive T cells might stochastically escape negative selection (15). A high frequency of self-reactive T cells in the Significance peripheral repertoire correlates with the susceptibility to develop autoimmune disease. Studies showed that mice with high fre- The adaptive has the potential to generate a quencies of myelin-specific T cells are more susceptible for the self-reactive response, which can eventually lead to an auto- induction of experimental autoimmune encephalomyelitis com- immune disease. To avoid this outcome, T with pared with mice with lower frequencies of these cells (16). In high-affinity, self-reactive antigen receptors are blocked from single transgenic mice expressing lymphocytic choriomeningitis vi- entering the mature T-cell pool (negative selection). Given this rus (LCMV) glycoprotein (gp33) or nucleoprotein with a polyclonal mechanism for removing dangerous high-affinity T cells, we T-cell repertoire, cytotoxic T lymphocytes (CTLs) generated during wondered whether autoimmunity is more likely to be caused by the immune response are of lower affinity compared with non- chronic stimulation of low-affinity T cells or by stimulation of transgenic mice, suggesting that the highest affinity T cells for the a few high-affinity T cells that escaped negative selection. In this neoantigen were removed from the T-cell repertoire by central paper, we show that T cells with an affinity just above the se- lection threshold can bypass negative selection and have the tolerance (16–18). Nevertheless, the remaining lower affinity T cells highest potential to cause an experimental autoimmune disease. were capable of inducing diabetes upon LCMV infection. Similar – findings have been reported for the rat insulin promoter (RIP) Author contributions: S.K., D.N., V.G.-F., and E.P. designed research; S.K., D.N., and V.G.-F. membrane-bound form of ovalbumin (mOVA) mouse line coex- performed research; D.N. and D.Z. contributed new reagents/analytic tools; S.K. and E.P. pressing the OT-I TCR (OVA transgenic-I T-cell receptor) β-chain analyzed data; and S.K. and E.P. wrote the paper. (17). Self-reactive T cells also play a role in the spontaneous de- The authors declare no conflict of interest. velopment of diabetes in nonobese diabetic (NOD) mice (19). *This Direct Submission article had a prearranged editor. Recently, we reported a peripheral correlate of the thymic 1To whom correspondence should be addressed. Email: [email protected]. negative selection affinity threshold such that above-threshold This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. ligands drive T cells into asymmetrical T-cell division, allowing 1073/pnas.1402724111/-/DCSupplemental.

17248–17253 | PNAS | December 2, 2014 | vol. 111 | no. 48 www.pnas.org/cgi/doi/10.1073/pnas.1402724111 Downloaded by guest on September 29, 2021 generate double-transgenic F1 animals (i.e., OT-I TCR × RIP- OVA F1, OT-I TCR × RIP-mQ4H7 F1, OT-I TCR × RIP-mT4 F1, and OT-I TCR × RIP-mQ4R7 F1). Surprisingly, double- transgenic F1 mice of all crosses showed high mortality within the first 2 wk after birth (89% in OT-I × RIP-mQ4H7, 100% in OT-I × RIP-mT4, 97% in OT-I × RIP-mQ4R7, 72% in OT-I × RIP-mOVA, and 52% in OT-I × RIP-sOVA) (Fig. 1). Blood glu- cose levels of newborn mice were elevated in all crosses, indicating severe hyperglycemia. From these results, we conclude that all RIP- OVA–variant transgenic mice expressed the OVA protein in the β cells of the corresponding RIP-OVA strains and that central tolerance mechanisms were overwhelmed in these F1 mice.

Below-Threshold T Cells Comprise a Low Risk of Autoimmunity. We wondered how target cell antigen affinity influences the de- velopment of diabetes following infection with a pathogen ex- pressing a high-affinity (i.e., molecular mimicry). Accordingly, we transferred OT-I T cells into RIP-variant mice, which were sub- sequently infected with Lm expressing the highest affinity OVA epitope (Lm-OVA). RIP-transgenic mice expressing threshold (RIP- mT4) and above-threshold (RIP-mQ4R7, RIP-OVA) OVA variants developed diabetes when as few as 3 × 103 OT-I T cells were transferred (Fig. 2A). The transgenic strain RIP-mQ4H7 expressing the below-threshold variant Q4H7 did not develop diabetes; adop- × 5 tive transfer of as many as 3 10 OT-I T cells did not result Fig. 2. Below-threshold T cells comprise a low risk for autoimmunity. (A) in glucosuria. Urine glucose was monitored in RIP-OVA– and RIP-OVA–variant mice injec- Autoimmune diseases can also be initiated by priming auto- ted with the indicated number of OT-I T cells, followed by infection with Lm- reactive T cells with self-antigens that are identical to the target OVA. (B) Immunization with self-antigen expressed in Lm. Mice injected with antigen. Therefore, OT-I T cells were transferred into various RIP- the indicated number of OT-I T cells were infected with Lm expressing the transgenic strains, followed by immunization with Lm expressing same self-antigen 1 d later. Mice were considered diabetic if urine glucose levels were sustained at >1,000 mg/dL for >2 d. The asterisk indicates pooled the same OVA variant as is expressed in the host pancreas (Fig. results of RIP-sOVA (100%, n = 2–5 mice per group) and RIP-mOVA (100%, 2B). Because diabetes induction was similar in RIP-sOVA and n = 2–5 mice per group). RIP-mOVA mice, we combined them into one group (RIP-OVA). A breakdown of the two strains is provided in the legend for Fig. 2. We observed efficient diabetes induction with only 3 × 103 trans- when the immunizing and target antigens are identical, only above- ferred OT-I T cells in mice expressing above-threshold self-anti- threshold antigens efficiently induce autoimmune diabetes. gens (RIP-OVA: 100%, nine of nine mice; RIP-mQ4R7: 73%, eight of 11 mice). Strikingly, mice expressing the threshold variant, Negative Selection Affinity Threshold Observed in Vivo. To study RIP-mT4, required 100-fold as many OT-I T cells (3 × 105)for negative selection in these transgenic strains, we generated mixed (BM) chimeras, in which a mixture of diabetes induction. In contrast, there was no induction of diabetes + + + CD45.1 B6 BM (70%) and CD45.1 /.2 OT-I Rag KO BM in RIP-mQ4H7 mice expressing the below-threshold OVA variant. + All transgenic lines receiving Lm, but no OT-I T cells, were free (30%) was injected into previously irradiated CD45.2 trans- of diabetes (SI Appendix,FigS6A). These data demonstrate that genic RIP-variant hosts (Fig. 3A). All chimeric mice were simi- larly reconstituted with OT-I BM (21–33%) (SI Appendix, Fig. S2). We assessed the frequency (Fig. 3B) and number (Fig. + 3C)ofsinglepositiveCD8αβ OT-I T cells in the lymph nodes (LNs) of the reconstituted chimeras 12 wk after BM transfer. The extent of was calculated in each chimeric strain (legend for Fig. 3C). Clonal deletion of OT-I T cells was not observed in RIP-mQ4H7 (below-threshold) or RIP-mT4 (thresh- old) hosts (Fig. 3 B and C). In contrast, the efficiency of clonal deletion was 91% in RIP-mQ4R7 (above-threshold) hosts and >99% in RIP-OVA (highest affinity) hosts (Fig. 3D). Similar results were obtained when analyzing thymocytes from the various chimeric mice (SI Appendix,Fig.S3). Taken together, these data demonstrate an affinity threshold that dictates the efficiency of negative selection in vivo (Fig. 3D). These results are similar to

what we observed in fetal thymic organ culture (8). IMMUNOLOGY AND Phenotype of T Cells Escaping Negative Selection. We wondered whether OT-I T cells escaping negative selection in mice expressing above-threshold self-antigens (RIP-mQ4R7 and RIP- Fig. 1. Incidence of overt autoimmune diabetes in RIP-mOVA–variant × OT-I – OVA) are phenotypically different from conventional OT-I cells. double-transgenic mice. Rag-2 deficient OT-I mice were crossed to RIP-sOVA, → RIP-mOVA, RIP-mQ4R7, RIP-mT4, or RIP-mQ4H7 mice. Mortality was assessed Analysis of LN cells revealed that only OT-I/B6 BM RIP- OVA (high-affinity) chimeras contained elevated percentages of by comparing the number of pups born with the number surviving at 2 wk. + + The asterisk indicates the pooled result of RIP-sOVA (52% mortality, 16 of 31 CD44 (Fig. 4A) and CD122 OT-I cells (Fig. 4B). Additionally mice) and RIP-mOVA (72% mortality, 36 of 50 mice). these hosts contained increased frequencies of OT-I cells with

Koehli et al. PNAS | December 2, 2014 | vol. 111 | no. 48 | 17249 Downloaded by guest on September 29, 2021 A OT-I (CD45.1+/.2+) / B6 (CD45.1+) Bone Marrow

Host (CD45.2+): B6 RIP-mQ4H7 RIP-mT4 RIP-mQ4R7 RIP-OVA Fig. 3. Efficiency of OT-I deletion in OT-I/B6 mixed Threshold BM chimeras expressing OVA-variant proteins is de- + + + Central Tolerance CD3 tetramer CD8 – 5 pendent on antigen affinity. (A D) Flow cytometric 105 6.3 45.8 10 12.4 29.8 105 2.9 46.9 105 10.4 1.8 105 10.2 0.2

44.9 4 49.9 46.4 87.4 analysis of lymphocytes in OT-I/B6 mixed BM chi- 104 10 104 104 81.9 104 meras was performed 12–15 wk after reconstitution. 103 103 103 103 103 (A) Representative flow cytometry plots of LN cells in 2 2 2 2 2 CD45.2 10 10 10 10 10 0 0 0 0 0 OT-I/B6 mixed BM chimeras are shown. Percentage

0 102 103 104 105 0 102 103 104 105 0 102 103 104 105 0 102 103 104 105 0 102 103 104 105 (B) and absolute numbers (C) of CD8αβ OT-I T cells CD45.1 within LNs were assessed by flow cytometry. Hori- B C D zontal bars in B and C represent the geometric mean value of the individual data points. The asterisk Host: = B6 indicates pooled results of RIP-sOVA (0.3%, 9,537, n

LN T cells RIP-mQ4H7 4) and RIP-mOVA (0.1%, 5,156, n = 3) mice. Results RIP-mT4 RIP-mQ4R7 shown were pooled from six independent experi- RIP-OVA* % Efficiency of % OT-I T cells % OT-I ments. (D) Efficiency of negative selection for each Negative Selection

nr OT-I T cells in LNs nr OT-I BM chimeric strain is plotted vs. antigen affinity (8). among CD8 1000 B6 B6 A* Efficiency of negative selection (%) = 100 − [(mean V Tetramer K (nM) Host: Host: D number of OT-I cells in RIP-OVA–variant host/mean RIP-mT4 RIP-OVA* RIP-mT4 RIP-O Affinity RIP-mQ4H7 RIP-mQ4R7 RIP-mQ4H7 RIP-mQ4R7 number of OT-I cells in B6 host) × 100].

memory phenotypes (Fig. 4). The increase of these two memory increased frequencies of CD8αα OT-I LN T cells (RIP-mQ4R7 populations was not correlated with negative selection per se, vs. B6 host: P = 0.06, RIP-OVA vs. B6 host: P = 0.03) (SI Ap- + because they were not increased in negatively selecting RIP- pendix, Fig. S5). The development of CD8αα MHC I-restricted mQ4R7 hosts. Recently, Tsai et al. (23) demonstrated that low- T cells is driven by high-affinity self-antigen (agonist selection) + + + – affinity, antigen-specific CD8 CD44 CD122 T cells from NOD (24 27). These cells carry a suboptimal coreceptor and have mice have suppressive properties. This population might develop likely bypassed negative selection. in response to strong recognition of self-ligand in the thymus. + + + Highest Potential for Autoimmunity Occurs Just Above the Affinity However, because CD8 CD44 CD122 OT-I (suppressor phe- Threshold. All chimeric mice were tolerant, because there was no notype) T cells were present in all chimeric strains (Fig. 4 A and sign of spontaneous autoimmune diabetes. To assess their risk of B, Bottom Right), there was no clear correlation of their presence developing autoimmune diabetes, we immunized each chimeric with negative selection in this experimental model. Chimeric strain with the corresponding self-peptide + LPS and plotted the mice expressing above-threshold OVA variants also contained incidence of diabetes vs. tetramer affinity (Fig. 5). Responsiveness

OT-I (CD45.1+/.2+) / B6 (CD45.1+) Bone Marrow A Host: B6 RIP-mQ4H7 RIP-mT4 RIP-mQ4R7 RIP-mOVA*

0.6 6.9 2.5 6.7 0.7 6.6 2.1 9.1 7.8 37 CD44

4.6 88 5.8 85.2 6.0 86.7 8.4 80.4 7.4 47.9

B CD62L TCM TEM naive suppressor phenotype OT-I OT-I EM CM % sup. OT-I % T % T % naive OT-I

Fig. 4. Phenotypic characterization of OT-I T cells in mixed BM chimeras. (A) LN cells from BM chimeras were stained with mAbs specific for CD44, CD62L, and CD122. Representative flow cytometry plots are shown. Numbers in the plots depict frequency (%) of cells in each quadrant. (B) Analysis of OT-I LN T cells with + + + − memory phenotypes in RIP-OVA–variant chimeras. Fractions of central memory T cells (TCM)(CD44 CD62L ), effector memory T cells (TEM) (CD44 CD62L ), naive (CD44−CD62L+), suppressor phenotype cells (CD44+CD122+), and OT-I CD8 αβ+ cells are shown. The asterisk indicates pooled results of RIP-sOVA (n = 4) and RIP-mOVA (n = 3) mice.

17250 | www.pnas.org/cgi/doi/10.1073/pnas.1402724111 Koehli et al. Downloaded by guest on September 29, 2021 A B compared with Q4H7 or containing lower numbers of peripheral Host OT-I T cells would also carry a very low risk (∼0%) of developing OVA/LPS B6 Q4H7/LPS RIP-mQ4H7 diabetes. Mice expressing a self-antigen at the affinity threshold T4/LPS RIP-mT4 ∼ Q4R7/LPS RIP-mQ4R7 have an intermediate ( 50%) risk of developing lethal autoim- OVA/LPS RIP-OVA* munity, but only if they contain large numbers (∼2 × 106) of OT-I cells (sectors 2 and 3) (Figs. 2B and 5). In contrast, the risk of

Glucosuria (%) ∼

Overt diabetes (%) developing diabetes is very high ( 90%) in RIP-mQ4R7 chi- meras. One can assume that chimeric mice expressing an even Tetramer KD (nM) Tetramer KD (nM) 5 Affinity Affinity higher affinity self-antigen and containing ≥10 OT-I cells would also have a very high (≥90%) risk of developing this form of Fig. 5. Highest risk for autoimmunity occurs slightly above the affinity autoimmunity. Finally, RIP-OVA chimeras containing 6 × 103 threshold for central tolerance. The percentage of immunized chimeric mice OT-I T cells remain tolerant to a challenge with OVA + LPS. with glucosuria (A) and sustained diabetes (B) is shown. RIP-OVA, RIP- × 3 mQ4H7, and C57BL/6 chimeras were free of diabetes for 2 wk following We assumed that chimeric mice containing as few as 6 10 OT-I immunization. The asterisk indicates pooled results of RIP-sOVA (n = 5) and cells and expressing a lower affinity self-antigen would also RIP-mOVA (n = 4) mice. have a similarly low risk of developing autoimmune diabetes (sector 5). In Fig. 6C, the hypothetical plot in Fig. 6B has been corrected of OT-I T cells to these peptides is shown in SI Appendix,Fig.S4. for the effects of central tolerance. Because above-threshold self- OT-I/B6 BM → RIP-mQ4H7 chimeras immunized with Q4H7 antigens prevent most but not all self-reactive thymocytes from peptide + LPS developed neither glucosuria nor lethal diabetes. entering the periphery, the risk of developing autoimmunity is On the other hand, 100% (eight of eight) OT-I/B6 BM → RIP- significantly reduced. As illustrated in Fig. 6C, we propose that mT4 chimeras immunized with the threshold peptide, T4 + LPS, the “dangerous” fraction of the peripheral T-cell repertoire is developed glucosuria (Fig. 5A) that resolved in 50% of these concentrated among T cells with a TCR affinity just above the animals (compare gray symbols in Fig. 5 A and B). In this model, negative selection threshold (i.e., Kd of ∼6 μM). At this affinity, the threshold ligand induces autoimmune symptoms but not al- negative selection is incomplete and the escaping T cells have ways irreversible disease (50%). Considering above-threshold sufficient affinity to induce autoimmune diabetes in this model self-antigens, 86% of OT-I/B6 BM → RIP-mQ4R7 chimeras system (20). At higher TCR affinities, negative selection is in- immunized with Q4R7 + LPS developed irreversible diabetes creasingly complete such that the few escaping T cells are in- (Fig. 5 A and B). In contrast, not a single (none of 10) OT-I/B6 capable of inducing a lethal autoimmune disease. BM → RIP-OVA chimeras immunized with OVA peptide + LPS developed glucosuria or irreversible diabetes (Fig. 5 A and B). Discussion Unlike the RIP-mQ4R7 chimeras, the tolerant state in these To address how self-antigen affinity influences several different highest affinity RIP-OVA chimeras is robust and stable. steps in the establishment of T-cell tolerance, we generated new All chimeric strains, regardless of the affinity of their OVA- transgenic mouse strains expressing antigens with variable af- variant self-antigen, contained a similar frequency (SI Appendix, finity for the OT-I TCR. All RIP-OVA–variant strains were + + Fig. S6B)ofCD4 FoxP3 regulatory T cells (Tregs; derived from crossed to OT-I cells to generate double-transgenic mice. All B6 donor BM). This finding was expected because the only dif- double-transgenic strains develop diabetes, indicating expression ference between the RIP strains is an MHC class I-presented of the neo–self-antigen in the pancreas. The incidence of di- epitope. Therefore, Treg frequency does not explain the dramatic abetes in RIP-mQ4H7, RIP-mT4, and RIP-mQ4R7 double- differences between the RIP-mQ4R7 and RIP-OVA chimeras in transgenic mice is 89–100%, suggesting that tolerance mecha- terms of susceptibility to developing autoimmune diabetes. nisms were overwhelmed. Lethal diabetes appears in only 52% of RIP-sOVA and 72% of RIP-mOVA double-transgenic mice, Self-Antigen Affinity vs. Risk of Developing Autoimmunity. Based on indicating that very high-affinity self-antigens more efficiently these results, we plotted self-antigen affinity vs. the number of deplete self-reactive thymocytes. These results are consistent peripheral OT-I cells and color-coded the risk to develop auto- with observations of Hubert et al. (28), who observed that 88% immune diabetes upon self-antigen challenge (Fig. 6). In Fig. 6A, OT-I × RIP-mOVA double-transgenic mice develop severe di- the number of OT-I cells from each of the four strains of RIP- abetes. Similarly, McGargill et al. (29) found that 80% of K14- variant chimeric mice (from Figs. 3 and 5) is plotted vs. antigen OVAp/OT-I double-transgenic mice, where OVA is expressed affinity. In Fig. 6B, the risk of developing lethal diabetes was under control of the keratin promoter 14, succumb to a lethal extrapolated to cover a wide range of self-antigen affinities and autoimmune disease due to insufficient deletion of developing numbers of self-reactive peripheral T cells. We assumed that a OT-I cells. Nevertheless, 20% of these animals survived. Surpris- chimeric mouse expressing a self-antigen of lower affinity ingly, double-transgenic animals expressing the below-threshold

A B C Fig. 6. Self-antigen affinity vs. risk of developing auto- Threshold Threshold Threshold . (A) Number (Nr.) of OT-I T cells in each chimeric 107 107 107 strain (Fig. 3) is plotted vs. antigen affinity. The negative

6 T4 6 2 6 selection affinity threshold is indicated by an arrow. The 10 Q4H7 10 10 INFLAMMATION 6 risk of developing diabetes is indicated by color (blue, ∼0% IMMUNOLOGY AND 105 105 105 ∼ ∼ Q4R7 1 3 4 risk; yellow, 50% risk; red, 90% risk). (B)DatainA are extrapolated for an extended range of self-antigen affin- 104 104 104 ities and numbers of self-reactive OT-I T cells. A description OVA 3 103 103 1 5 103 Nr. peripheral T cells T peripheral Nr. cells T peripheral Nr. Nr. peripheral T cells T peripheral Nr. of different regions of the plot is provided in Results.(C) Self-antigen affinity Self-antigen affinity Self-antigen affinity Extrapolated plot in B is corrected for the effects of neg- Experimental Data Extrapolation Effect of Central Tolerance ative selection; central deletion removes most, but not all, OVA-reactive T cells with TCR affinities above the negative selection threshold (sector 6). In principle, this model could Risk Lethal disease ~ 0% ~ 50 % ~ 90 % be applied to any self-reactive CD8 .

Koehli et al. PNAS | December 2, 2014 | vol. 111 | no. 48 | 17251 Downloaded by guest on September 29, 2021 antigen Q4H7 develop diabetes (8, 20, 22). A possible explanation diabetes). In contrast, below-threshold affinity chimeras (OT-I/ is that just after birth, an extremely large number of OT-I T cells B6 BM → RIP-mQ4H7) and the highest affinity chimeras (OT-I/ likely enter the periphery before significant numbers of Tregs B6 BM → RIP-mOVA and OT-I/B6 BM → RIP-sOVA) failed to accumulate. In this relatively lymphopenic environment, a suffi- develop diabetes. Below-threshold RIP-mQ4H7 chimeras are cient number of OT-I T cells might expand by homeostatic division likely protected due to inefficient T-cell priming and poor within the first week of life, generating diabetogenic CTLs. target cell lysis. High-affinity RIP-mOVA and RIP-sOVA In a mimicry model (30–32), where transferred OT-I cells were chimeras are likely protected due to extremely efficient neg- fully activated following Lm-OVA (high-affinity) infection (33), ative selection. Interestingly, the highest risk of developing RIP-mQ4H7 mice did not develop diabetes, indicating that host autoimmune diabetes in this model is just above the negative tissues expressing a below-threshold self-antigen are difficult to selection affinity threshold. In this narrow range of TCR af- damage. All other RIP-OVA–variant mice efficiently developed finity, negative selection is leaky and the escaping peripheral lethal diabetes. Therefore, under conditions of molecular mimicry, T cells can be sufficiently activated by their self-antigen to host tissues expressing threshold or above-threshold self-antigens induce autoimmune pathology. are highly susceptible to autoimmune attack. In a different setting, Under certain conditions, low-affinity T cells cause diabetes in where the host experiences an infection of a peripheral organ, an RIP-mOVA model. Enouz et al. (22) have convincingly shown inflammation-induced tissue damage might result in release and that below-threshold antigens induce peripheral CD8 T cells to presentation of self-tissue antigens, which subsequently induce differentiate into CTLs; furthermore, these below-threshold stim- a pathogenic response from self-antigen specific T cells that have ulated T cells cause hyperglycemia in mice expressing a threshold survived negative selection. To address the role of antigen affinity affinity antigen in the pancreas. The experiments described here in this context, RIP-OVA– and RIP-OVA–variant mice were were slightly different; although T cells can be stimulated by a adoptively transferred with OT-I cells and infected with Lm ex- below-threshold antigen, such T cells do not induce diabetes in pressing the same OVA variant expressed by the self-antigen mice expressing the same below-threshold antigen on pancreatic β transgenic mouse. Mice expressing self-antigen above the affinity cells. Therefore, our results are not inconsistent with the results threshold (RIP-mQ4R7 and RIP-mOVA) are at high risk to develop reported by Enouz et al. (22). Our data make the important point diabetes, even when low numbers of OT-I cells were transferred. In that it is quantitatively more difficult to induce autoimmune pa- contrast, mice expressing the threshold antigen T4 required 100-fold thology using a below-threshold antigen compared with an above- more transferred OT-I T cells to develop diabetes. threshold antigen. Many early studies examining tolerance to very high-affinity The frequency of lethal diabetes is high in this model (50% neo–self-antigens did not address tolerance to intermediate- for chimeras expressing a threshold antigen, 86% for chimeras and low-affinity antigens (4, 16, 18, 34, 35). More recent studies expressing an above threshold antigen). This observation is indicate that tolerance to intermediate- and low-affinity anti- likely due to the fact that the chimeras were reconstituted with gens is inefficient (17, 22). To address this issue in a more a large fraction (30%) of stem cells derived from OT-I mice. In quantitative manner, we generated mixed OT-I/B6 BM chi- contrast, the frequency of self-reactive thymocytes for a single – meric mice using RIP-OVA variant hosts. Chimeras expressing epitope is much smaller in polyclonal individuals, which likely below-threshold or at-threshold affinity antigens do not induce improves the efficiency of thymic deletion. This idea is con- negativeselectioninvivo,whereas hosts expressing above- sistent with the relatively low incidence of autoimmunity in the threshold self-antigens are highly efficient in preventing OT-I general population; 3% of the human population suffers from = T cells from entering the periphery (RIP-mQ4R7 91% effi- some form of autoimmunity (48). Nevertheless, an individual > > cient, RIP-mOVA 99.5% efficient, RIP-sOVA 99.4% ef- can probably cope with a small number of high-affinity pe- ficient). We also observed deletion of self-reactive OT-I cells ripheral T cells and still be considered functionally tolerant, within the thymus, indicating that in chimeric mice, a large allowing negative selection to be less than perfect, which seems component of their tolerance is attributable to thymic selection to be the case (49, 50). Recent computational studies (51) processes (SI Appendix,FigS3). These results are similar to propose that an immune or autoimmune response requires what we observed in fetal thymic organ culture (8). It is worth a minimum (quorum) number of antigen-specific T cells. If an noting that just above the affinity threshold, negative selection individual carries fewer than the quorum number of above- is incomplete. threshold T cells for a particular epitope, it may be extremely Because agonist-selected thymocytes enter diverse T-cell line- difficult to induce an autoimmune disease. Our studies suggest ages (36–41), we examined the phenotype of peripheral OT-I T cells – αα that T cells with an antigen affinity just above the affinity threshold in chimeric RIP-mOVA variant mice. The frequency of CD8 are more likely to exceed the quorum number in the periphery. For OT-I cells increased in chimeras expressing above-threshold anti- – αα this reason, T cells with self-reactivity just above the negative se- gens (42 45). Given their reduced antigen binding (46, 47), CD8 lection threshold have the highest autoimmune potential. thymocytes have an increased chance to evade negative selection. hi hi The proportion of OT-I T cells with a central (CD44 CD62L )or Materials and Methods hi lo effector (CD44 CD62L ) memory phenotype was clearly in- Mice. RIP-sOVA and RIP-mOVA mice (20, 52, 53), OT-I TCR transgenic mice b creased, but only in chimeras expressing the highest affinity ligand, recognizing K /Ova – , and CD45-1 congenic C57BL/6 mice were all + hi hi 257 264 OVA. CD8 CD44 CD122 cells have been reported to pre- obtained from the Jackson Laboratory. All animal work was done in accor- vent diabetes in NOD mice (23); however, the frequency of dance with the federal and cantonal laws of Switzerland. Animal research + CD8 CD44hiCD122hi OT-I cells did not correlate with the state protocols were approved by the Animal Research Commission of the Canton of tolerance in our chimeric strains. Both OT-I/B6 BM → RIP- of Baselstadt, Switzerland. mT4 (nontolerant) and OT-I/B6 BM → RIP-mOVA (tolerant) Generation of OVA-variant transgenic mice. Generation of OVA-variant trans- + mice contained a similar percentage of CD8 CD44hiCD122hi genic mice is described in SI Appendix, Fig S1. OT-I cells. On the other hand, the total OT-I number varies Generation of BM chimeric mice. Generation of BM chimeric mice is described in SI Appendix, Fig S2. considerably between these two strains. For this reason, it was + hi hi difficult to evaluate the role of CD8 CD44 CD122 OT-I cells Adoptive Cell Transfer and Infections. Mice were injected i.v. with single-cell in maintaining tolerance in this particular diabetes model. suspensions of OT-I cells; on the following day, 5,000 cfu of Lm-OVA or Lm- The highest potential for autoimmunity occurs near the affinity OVA variant was injected i.v.. Recombinant Lm expressing the full-length threshold for negative selection (RIP-mT4 chimeras: 50% irre- OVA protein containing the CD8 epitope SIINFEKL (OVA) or altered ligands versible diabetes, RIP-mQ4R7 chimeras: 86% irreversible Q4R7, T4, or Q4H7 was previously described (21).

17252 | www.pnas.org/cgi/doi/10.1073/pnas.1402724111 Koehli et al. Downloaded by guest on September 29, 2021 Blood and Urine Glucose Measurements. Blood samples from the tail vain were Data Analyses. Flow cytometry measurements were performed on a FACS read on a Contour blood glucose reader (Bayer). Mice with blood glucose Canto II (Becton Dickinson), and data were analyzed with FlowJo software levels >15 mmol/L were considered to be hyperglycemic. Urine glucose was (TreeStar). Graphs were generated with Prism (GraphPad). Unpaired t tests assessed with test strips (DIABUR-TEST 5000; ACCU-CHEK). Mice with sus- and one-way-ANOVA were performed (Prism) where indicated. tained urine glucose levels >1,000 mg/dL were considered diabetic. ACKNOWLEDGMENTS. We thank C. G. King, S. Keck, O. Stepanek, T. Rolink, Tetramer, Surface Staining, and Flow Cytometry. Lymphocytes were and D. Finke for critical discussions; M. Schmaler for advice on culturing Lm; stained with the following and/or tetramers: CD45.1-FITC, CD45.1- S.Keck,L.Wyss,R.Lang,andB.Hausmann for excellent experimental assistance; K. Thienel and E. Dalmas for advice on pancreatic islet isolation; phycoerythrin (PE), CD45.2-allophycocyanin (APC)-Cy7, CD8α-biotin, CD4- e β and the team of U. Schneider for animal husbandry. The work was Alexa700, and CD3 -APC (all from Becton Dickinson); CD8 -peridinin supported by Research Grants 310030B_133131/1, Synergia (Swiss National chlorophyll Cy5.5, CD44-APC, CD122-FITC, and CD62L-PE-Cy7 (all from Science Foundation), Sybilla (European Union Seventh Framework b BioLegend); CD8α-PE-Cy7 (eBioscience); and K -OVA (SIINFEKL) tetramers Programme), and TerraIncognita (European Research Council) (all to E.P.) (from V.G.-F. and E.P.). and by Grant 310030_130512 (Swiss National Science Foundation) (to D.Z.).

1. Palmer E (2003) Negative selection—Clearing out the bad apples from the T-cell 29. McGargill MA, et al. (2002) A spontaneous CD8 T cell-dependent autoimmune disease repertoire. Nat Rev Immunol 3(5):383–391. to an antigen expressed under the human keratin 14 promoter. J Immunol 169(4): 2. Starr TK, Jameson SC, Hogquist KA (2003) Positive and negative selection of T cells. 2141–2147. Annu Rev Immunol 21:139–176. 30. Gronski MA, et al. (2004) TCR affinity and negative regulation limit autoimmunity. 3. Stritesky GL, Jameson SC, Hogquist KA (2012) Selection of self-reactive T cells in the Nat Med 10(11):1234–1239. thymus. Annu Rev Immunol 30:95–114. 31. Hudrisier D, et al. (2001) Structural and functional identification of major histocom- 4. Kisielow P, Blüthmann H, Staerz UD, Steinmetz M, von Boehmer H (1988) Tolerance in patibility complex class I-restricted self-peptides as naturally occurring molecular T-cell-receptor transgenic mice involves deletion of nonmature CD4+8+ thymocytes. mimics of viral antigens. Possible role in CD8+ T cell-mediated, virus-induced auto- Nature 333(6175):742–746. immune disease. J Biol Chem 276(22):19396–19403. 5. von Boehmer H, Teh HS, Kisielow P (1989) The thymus selects the useful, neglects the 32. Nugent CT, et al. (2000) Characterization of CD8+ T lymphocytes that persist after useless and destroys the harmful. Immunol Today 10(2):57–61. to a self antigen expressed in the pancreas. J Immunol 164(1): 6. Goldrath AW, Bevan MJ (1999) Selecting and maintaining a diverse T-cell repertoire. 191–200. Nature 402(6759):255–262. 33. Pope C, et al. (2001) Organ-specific regulation of the CD8 T cell response to Listeria 7. Taniguchi RT, et al. (2012) Detection of an autoreactive T-cell population within the monocytogenes infection. J Immunol 166(5):3402–3409. polyclonal repertoire that undergoes distinct autoimmune regulator (Aire)-mediated 34. Kurts C, Heath WR, Carbone FR, Kosaka H, Miller JF (1998) Cross-presentation of self selection. Proc Natl Acad Sci USA 109(20):7847–7852. antigens to CD8+ T cells: The balance between tolerance and autoimmunity. Novartis 8. Daniels MA, et al. (2006) Thymic selection threshold defined by compartmentalization Found Symp 215:172–181, discussion 181–190. of Ras/MAPK signalling. Nature 444(7120):724–729. 35. Kirberg J, et al. (1994) Thymic selection of CD8+ single positive cells with a class II 9. Naeher D, et al. (2007) A constant affinity threshold for T cell tolerance. J Exp Med major histocompatibility complex-restricted receptor. J Exp Med 180(1):25–34. 204(11):2553–2559. 36. Hogquist KA (2001) Signal strength in thymic selection and lineage commitment. Curr 10. Metzger TC, Anderson MS (2011) Control of central and peripheral tolerance by Aire. Opin Immunol 13(2):225–231. Immunol Rev 241(1):89–103. 37. Singer A (2002) New perspectives on a developmental dilemma: The kinetic signaling 11. Anderson MS, et al. (2002) Projection of an immunological self shadow within the model and the importance of signal duration for the CD4/CD8 lineage decision. Curr thymus by the aire protein. Science 298(5597):1395–1401. Opin Immunol 14(2):207–215. 12. Mathis D, Benoist C (2009) Aire. Annu Rev Immunol 27:287–312. 38. Izon DJ, et al. (2001) Notch1 regulates maturation of CD4+ and CD8+ thymocytes by 13. Klein L, Hinterberger M, Wirnsberger G, Kyewski B (2009) in the modulating TCR signal strength. Immunity 14(3):253–264. thymus for positive selection and central tolerance induction. Nat Rev Immunol 9(12): 39. Hayes SM, Love PE (2006) Strength of signal: A fundamental mechanism for cell fate 833–844. specification. Immunol Rev 209:170–175. 14. Gallegos AM, Bevan MJ (2004) Central tolerance to tissue-specific antigens mediated 40. Smith-Garvin JE, et al. (2010) T-cell receptor signals direct the composition and by direct and indirect antigen presentation. J Exp Med 200(8):1039–1049. function of the memory CD8+ T-cell pool. Blood 116(25):5548–5559. 15. Gotter J, Brors B, Hergenhahn M, Kyewski B (2004) Medullary epithelial cells of the 41. Chang JT, et al. (2007) Asymmetric T division in the initiation of adaptive human thymus express a highly diverse selection of tissue-specific genes colocalized in immune responses. Science 315(5819):1687–1691. chromosomal clusters. J Exp Med 199(2):155–166. 42. Leishman AJ, et al. (2002) Precursors of functional MHC class I- or class II-restricted 16. Ohashi PS, et al. (1991) Ablation of “tolerance” and induction of diabetes by virus CD8alphaalpha(+) T cells are positively selected in the thymus by agonist self-pep- infection in viral antigen transgenic mice. Cell 65(2):305–317. tides. Immunity 16(3):355–364. 17. Zehn D, Bevan MJ (2006) T cells with low avidity for a tissue-restricted antigen rou- 43. Baldwin TA, Hogquist KA, Jameson SC (2004) The fourth way? Harnessing aggressive tinely evade central and peripheral tolerance and cause autoimmunity. Immunity tendencies in the thymus. J Immunol 173(11):6515–6520. 25(2):261–270. 44. Jordan MS, et al. (2001) Thymic selection of CD4+CD25+ regulatory T cells induced by 18. von Herrath MG, Dockter J, Oldstone MB (1994) How virus induces a rapid or slow onset an agonist self-peptide. Nat Immunol 2(4):301–306. insulin-dependent diabetes mellitus in a transgenic model. Immunity 1(3):231–242. 45. Zhou D, et al. (2004) Lysosomal glycosphingolipid recognition by NKT cells. Science 19. Clemente-Casares X, Tsai S, Huang C, Santamaria P (2012) Antigen-specific thera- 306(5702):1786–1789. peutic approaches in Type 1 diabetes. Cold Spring Harb Perspect Med 2(2):a007773. 46. Cawthon AG, Lu H, Alexander-Miller MA (2001) Peptide requirement for CTL acti- 20. King CG, et al. (2012) T cell affinity regulates asymmetric division, effector cell dif- vation reflects the sensitivity to CD3 engagement: Correlation with CD8alphabeta ferentiation, and tissue pathology. Immunity 37(4):709–720. versus CD8alphaalpha expression. J Immunol 167(5):2577–2584. 21. Zehn D, Lee SY, Bevan MJ (2009) Complete but curtailed T-cell response to very low- 47. Arcaro A, et al. (2001) CD8beta endows CD8 with efficient coreceptor function by affinity antigen. Nature 458(7235):211–214. coupling T cell receptor/CD3 to raft-associated CD8/p56(lck) complexes. J Exp Med 22. Enouz S, Carrié L, Merkler D, Bevan MJ, Zehn D (2012) Autoreactive T cells bypass 194(10):1485–1495. negative selection and respond to self-antigen stimulation during infection. J Exp 48. Cooper GS, Bynum ML, Somers EC (2009) Recent insights in the epidemiology of Med 209(10):1769–1779. autoimmune diseases: Improved prevalence estimates and understanding of cluster- 23. Tsai S, et al. (2010) Reversal of autoimmunity by boosting memory-like autor- ing of diseases. J Autoimmun 33(3-4):197–207. egulatory T cells. Immunity 32(4):568–580. 49. Moon JJ, et al. (2011) Quantitative impact of thymic selection on Foxp3+ and Foxp3- 24. Cruz D, et al. (1998) An opposite pattern of selection of a single T cell antigen receptor in subsets of self-peptide/MHC class II-specific CD4+ T cells. Proc Natl Acad Sci USA the thymus and among intraepithelial lymphocytes. J Exp Med 188(2):255–265. 108(35):14602–14607. 25. Poussier P, Ning T, Banerjee D, Julius M (2002) A unique subset of self-specific in- 50. Trudeau JD, et al. (2003) Prediction of spontaneous autoimmune diabetes in NOD traintestinal T cells maintains gut integrity. J Exp Med 195(11):1491–1497. mice by quantification of autoreactive T cells in peripheral blood. J Clin Invest 111(2): 26. Levelt CN, et al. (1999) High- and low-affinity single-peptide/MHC ligands have dis- 217–223.

tinct effects on the development of mucosal CD8alphaalpha and CD8alphabeta T 51. Butler TC, Kardar M, Chakraborty AK (2013) Quorum sensing allows T cells to dis- INFLAMMATION

lymphocytes. Proc Natl Acad Sci USA 96(10):5628–5633. criminate between self and nonself. Proc Natl Acad Sci USA 110(29):11833–11838. IMMUNOLOGY AND 27. Denning TL, et al. (2007) Mouse TCRalphabeta+CD8alphaalpha intraepithelial lym- 52. Kurts C, Kosaka H, Carbone FR, Miller JF, Heath WR (1997) Class I-restricted cross- phocytes express genes that down-regulate their antigen reactivity and suppress presentation of exogenous self-antigens leads to deletion of autoreactive CD8(+)T immune responses. J Immunol 178(7):4230–4239. cells. J Exp Med 186(2):239–245. 28. Hubert FX, et al. (2011) Aire regulates the transfer of antigen from mTECs to dendritic 53. Behrens GM, et al. (2004) Helper requirements for generation of effector CTL to islet cells for induction of thymic tolerance. Blood 118(9):2462–2472. beta cell antigens. J Immunol 172(9):5420–5426.

Koehli et al. PNAS | December 2, 2014 | vol. 111 | no. 48 | 17253 Downloaded by guest on September 29, 2021