Antigen-Encoding Bone Marrow Terminates Islet-Directed Memory CD8+ T-Cell Responses to Alleviate Islet Transplant Rejection

1328 Diabetes Volume 65, May 2016

Miranda A. Coleman,1 Claire F. Jessup,2,3 Jennifer A. Bridge,1 Nana H. Overgaard,1 Daniella Penko,2 Stacey Walters,4 Danielle J. Borg,5 Ryan Galea,1 Josephine M. Forbes,5 Ranjeny Thomas,1 Patrick T.C. Coates,2 Shane T. Grey,4 James W. Wells,1 and Raymond J. Steptoe1

Diabetes 2016;65:1328–1340 | DOI: 10.2337/db15-1418

Islet-specific memory T cells arise early in type 1 diabetes therapies. After immunogenic antigen encounter, na- (T1D), persist for long periods, perpetuate disease, and ive T cells give rise to effectors and a small population are rapidly reactivated by islet transplantation. As memory of residual memory cells with increased longevity, “ ” T cells are poorly controlled by conventional therapies, faster response kinetics, and a reduced dependence on – memory T cell mediated attack is a substantial challenge costimulatory signals (1). While this is advantageous for in islet transplantation, and this will extend to application re-encounter with pathogens, formation of T-cell memory of personalized approaches using stem cell–derived re- can be detrimental under some circumstances. Pathogenic placement b-cells. New approaches are required to limit memory autoimmune attack of transplanted islets or re- memory T cells are prominent in type 1 diabetes (T1D), placement b-cells. Here, we show that transfer of bone where they arise during the early, preclinical phase of marrow encoding cognate antigen directed to dendritic disease and are well established by the time of diagnosis cells, under mild, immune-preserving conditions, inacti- (2). In addition to perpetuating disease, islet-specific vates established memory CD8+ T-cell populations and CD4+ and CD8+ memory T cells are persistent and may generates a long-lived, antigen-specific tolerogenic envi- remain dormant for extended periods after T1D onset but ronment. Consequently, CD8+ memory T cell–mediated are rapidly reactivated by islet antigen exposure during targeting of islet-expressed antigensispreventedandislet islet transplantation and contribute substantially to re- IMMUNOLOGY AND TRANSPLANTATION graft rejection alleviated. The immunological mechanisms jection of transplanted islets (3,4). Thus, immunother- of protection are mediated through deletion and induction apy for T1D must address memory T cells regardless of unresponsiveness in targeted memory T-cell popula- of disease stage, and this is crucial for success of islet tions. The data demonstrate that hematopoietic stem transplantation. cell–mediated gene therapy effectively terminates antigen- Conventional immunosuppressants such as calcineurin specific memory T-cell responses, and this can alleviate destruction of antigen-expressing islets. This addresses a inhibitors and rapamycin control memory much less effec- key challenge facing islet transplantation and, importantly, tively than naive T-cell responses (5). Many immunother- – the clinical application of personalized b-cell replacement apies for T cell mediated diseases attempt to reintroduce or therapies using patient-derived stem cells. expand regulatory T cells (Tregs); however, memory T cells can resist suppression by Tregs (6,7). Therefore, development of new approaches to terminate memory T-cell re- It is increasingly recognized that islet-specificmemory sponses in a controlled, antigen-specificcontextwouldbe T cells represent a major barrier to b-cell replacement highly beneficial. In the clinical context, this is particularly

1The University of Queensland Diamantina Institute, The University of Queens- Corresponding author: Raymond J. Steptoe, [email protected]. land, Translational Research Institute, Brisbane, Queensland, Australia Received 13 October 2015 and accepted 22 February 2016. 2Discipline of Medicine, University of Adelaide, Adelaide, South Australia, Australia This article contains Supplementary Data online at http://diabetes 3Department of Anatomy and Histology, Flinders University, Adelaide, South .diabetesjournals.org/lookup/suppl/doi:10.2337/db15-1418/-/DC1. Australia, Australia 4Garvan Institute of Medical Research, Sydney, New South Wales, Australia © 2016 by the American Diabetes Association. Readers may use this article as 5Mater Research Institute, The University of Queensland, Translational Research long as the work is properly cited, the use is educational and not for profit, and Institute, Brisbane, Queensland, Australia the work is not altered. diabetes.diabetesjournals.org Coleman and Associates 1329 pertinent for application of islet transplantation and person- succinimidyl ester (CFSE) labeled before transfer. Unless alized b-cell replacement therapies using patient-derived stated otherwise, 2 3 106 Tmem were transferred (intra- stem cells where memory autoimmune responses would at- venous, lateral tail vein). tack replacement cells. BM Transplantation Dendritic cells (DCs) are increasingly considered clin- Femurs and tibias were collected into mouse tonicity ically appropriate therapeutics for T1D and other auto- PBS. BM was flushed with mouse tonicity PBS/2.5% FCS, immune diseases. Indeed, memory CD4+ and CD8+ T-cell erythrocytes were lysed (NH Cl/TRIS buffer), and BM was responses are terminated by DCs or other antigen- 4 injected intravenously within 3 h of irradiation (300cGy, presenting cells (APCs) engineered to express cognate an- 137Cs source). HSPCs for transfer were prepared by high- tigen (8–10). One strategy to achieve APC-targeted antigen 2ve speed FACS sorting of lin-negative (lin )/c-kit-positive expression to prevent diabetes development therapeutically +ve (c-kit ) cells to typically .95% purity from bulk BM. is to transfer bone marrow (BM) or hematopoietic stem and +ve 2ve HSPCs were depleted from BM by sorting lin /c-kit progenitor cells (HSPCs) genetically modified to encode an- cells from BM. tigen expression that, once engrafted, continually gives rise to antigen-expressing APCs (11). Autologous hematopoietic In Vitro and In Vivo Assays stem cell transplantation (HSCT) has been tried for T1D CFSE labeling and flow cytometric analyses including with exciting outcomes but, as currently applied, uses toxic bead-based counting assays were performed as previously conditioning leading to complete immune ablation to de- described (15,17). OVA/QuilA immunization was as pre- plete pathogenic immune cells including memory T cells viously described (9). In vivo cytotoxic T lymphocyte (CTL) (12,13). assays were preformed as previously described (15). Mono- Here, we tested the hypothesis that addition of gene clonal antibodies were purchased from Biolegend, BD Bio- therapy to HSCT, to achieve tolerogenic self-antigen ex- sciences, or BioXcell (Lebanon, NH) or grown, purified, and pression in APCs, would terminate established memory conjugated in-house. ELISpot assays were as previously T-cell responses, restoring long-lasting tolerance to islet described (15) using OVA257–264 (0.5 mg/mL) or keyhole lym- antigens while maintaining protective immunity. We dem- phocyte hemocyanin (10 mg/mL) stimulation and counted onstrate that, under immune-preserving conditions, trans- with an ELISpot reader (AID GmbH, Strassburg, Germany). plantation of BM encoding DC-expressed antigen terminates Skin and Islet Transplantation + cognate CD8 memory T-cell responses and alleviates im- Skin grafting used double grafting of skin from K5.mOVA mune destruction of newly transplanted islets. and BALB/c donors as previously described (18). For islet transplantation, mice were administered a single dose of RESEARCH DESIGN AND METHODS streptozotocin (200 mg/kg i.p., pH 4.2 citrate buffer), and Mice blood glucose (BG) was monitored daily using an Accu- OT-I (14), 11c.OVA (15), and K5.mOVA (16) mice were Check Advantage (Roche). Hyperglycemic (.16.6 mmol/L) maintained under specific pathogen-free conditions in the mice were transplanted under the kidney capsule with islets 2 Princess Alexandra Hospital Biological Resource Facility prepared from OVA+ve RIP.mOVA or OVA ve littermate or Translational Research Institute Biological Resource donors at a ratio of three donors per recipient. BG was Facility, Brisbane, Australia. C57BL/6 and B6.SJLptprca monitored as indicated. Mice in which euglycemia (,16.6 mice were purchased from the Animal Resources Centre mmol/L) was not initially restored were excluded due to (Perth, Australia). B6.SJL and OT-I mice were crossed to primary graft nonfunction. Rejection was defined as two generate CD45.1+CD45.2+ OT-I mice. 11c.OVA mice were consecutive BG levels .16.6 mmol/L after initial euglycemia. backcrossed to B6.SJLptprca to make CD45.1+ 11c.OVA Mice were nephrectomized $100 days postgrafting to de- mice. Unless stated otherwise, CD45.1+ B6.SJLptprca termine graft dependence of euglycemia. Kidneys containing mice were used as recipients and CD45.2+ mice used as islet grafts were formalin fixed and embedded for histolog- BM donors. All animal procedures were approved by the ical analysis (19). University of Queensland Animal Ethics Committee. Statistical Analysis t OT-I T-Cell Transfers Student test was used for comparison of means and one- For the transfer of naive OT-I T cells, suspensions of way ANOVA with Newman-Keuls or Tukey posttest for pooled mesenteric, inguinal, axillary, and brachial lymph multiple comparisons (GraphPad Prism 5 or Prism 6). Sur- nodes (LNs) were prepared and transferred (5 3 106) vival data were analyzed using a Log-rank test (Mantel-Cox) as previously described (15). Memory OT-I T cells were (GraphPadPrism5orPrism6). generated as described previously (8). Briefly, LNs were harvested from OT-I or CD45.1+CD45.2+ OT-I mice and RESULTS cultured in complete RPMI with 1% mouse serum, 0.1 In Vitro–Generated OT-I Tmem Establish Functional In mg/mL OVA257–264, and 10 ng/mL rhIL-2. After 3 days, cells Vivo Memory Populations werewashedandreculturedwith10ng/mLrmIL-15for2 To study induction of tolerance to islet antigen, we days. Where indicated, OT-I T cells were carboxyfluorescein adapted an established system (8,20) to generate and 1330 Reversal of Autoimmune Islet Rejection by BMT Diabetes Volume 65, May 2016 monitor CD8+ T-cell memory at the individual cell level. were transferred and recipients irradiated (300cGy) 1 week LN cells from OVA-specific OT-I T-cell receptor transgenic after transfer, and then BM from CD45.2+ non-Tg or (Tg) mice were cultured in peptide/interleukin (IL)-2 washed 11c.OVA donors, which expresses OVA in DCs (15), and cultured in IL-15 over a period of 5 days (8,21), which was transferred. Donor-type (CD45.2+) leukocytes accumu- 2 gives rise to predominantly CD44high/pCD62L+/CD69 cen- lated in blood and spleen, indicating that BM effectively tral memory phenotype cells, referred to here as “Tmem.” engrafted from OVA-negative and 11c.OVA donors (Fig. These cells establish long-lived memory populations in vivo 1C and D). Donor-type DCs developed similarly in recip- that readily respond to immunogenic rechallenge (Fig. 1A), ients of 11c.OVA or non-Tg BM (Fig. 1E), indicating that which would target OVA-expressing b-cells. OVA expression was no impediment to DC development in the presence of OVA-specific Tmem. Low-Dose Irradiation Permits Effective Engraftment of Antigen-Encoding BM While Preserving CD8+ T-Cell Transplantation of Antigen-Encoding BM Inactivates Memory OT-I Memory T Cells Low-dose irradiation facilitates engraftment of BM in We next determined whether antigen-encoding BM naive and primed mice (22) and preserves immunity. For transfer (BMT) under immune-preserving conditions inacti- testing of whether immunity was preserved in mice with vated preexisting memory CD8+ T-cell responses. OT-I TCR-Tg memory CD8+ T cells, OT-I Tmem were trans- Tmem were transferred to non-Tg mice and 1 week later, ferred to CD45.1+ recipients. Recipients were left un- mice were irradiated (300cGy) and 11c.OVA or non-Tg BM treated or irradiated 1 week later (300cGy), and BM was transferred. Six weeks later, responsiveness of OT-I from CD45.2+ non-Tg mice was transferred. No signifi- cells was tested by immunogenic challenge with OVA/ cant loss of responsiveness in OT-I memory populations QuilA (8), which substantially expanded OT-I T-cell num- resulted from low-dose irradiation and transfer of non-Tg ber in spleens of no bone marrow transfer (no-BMT) con- BM (Fig. 1B). Antigen-encoding BM engraftment in immune- trols and non-Tg BM recipients relative to sham-challenged competent mice is enabled by restricting antigen expression controls (Fig. 2A–C)butnotinrecipientsof11c.OVABM to differentiated CD11c+ DCs (22). For testing of this in (Fig. 2A–C). Intracellular cytokine staining for interferon mice carrying TCR Tg memory CD8+ T cells, OT-I Tmem (IFN)-g, as a measure of effector function, showed that

Figure 1—Immune-preserving conditioning permits engraftment of 11c.OVA BM in the presence of OVA-speciﬁc memory CD8+ T cells. A: OT-I Tmem (2 3 106) were transferred to non-Tg (C57BL/6, CD45.2+) mice that were sham (PBS/QuilA) or OVA/QuilA challenged 6 weeks later. One week later, OT-I (CD45.1+CD45.2+CD8+Va2+) T cells were enumerated. Fold expansion is shown in brackets. B: OT-I Tmem were transferred to non-Tg mice and 1 week later recipients left untreated or irradiated (300cGy) and injected intravenously with non-Tg BM (107). Six weeks later, mice were sham (PBS/QuilA) or OVA/QuilA challenged. OT-I T cells were enumerated 1 week later. Fold expansion is shown in brackets. C–E: OT-I Tmem were transferred to non-Tg mice and 1 week later recipients irradiated (300cGy) and non-Tg or 11c.OVA BM (107) injected intravenously. Peripheral blood as indicated (C) or spleen at 7 weeks (pooled sham-challenged and challenged mice) was analyzed (D), and donor-type DCs (CD11chigh) were enumerated in peripheral blood (E). diabetes.diabetesjournals.org Coleman and Associates 1331

OVA/QuilA challenge elicited an ;200-fold increase in total IFN-g–producing OT-I T cells in spleens of both no-BMT controls and non-Tg BM recipients (Fig. 2D). The number of splenic IFN-g–producing OT-I T cells in 11c.OVA BM recipients remained low, however, regardless of challenge (Fig. 2D). Similarly, little CTL activity was elicited in vivo by OVA/QuilA challenge of 11c.OVA BM recipients, but CTL activity was substantial in recipients of non-Tg BM or in no-BMT controls (Fig. 2E). As we have previously reported (22), BMT using low-dose irradiation does lead to some mild, mostly nonsignificant, reduction in responsiveness to OVA challenge (Fig. 2A–D). Together, these data indicate that responsiveness of an established OT-I Tmem population was effectively inhibited by OVA- encoding BM transfer. In additional studies where OT-I Tmem recipients were immunized with KLH/QuilA prior to BM transfer, KLH responses were preserved regardless of whether OVA-encoding or non-Tg BM was transferred (Supplementary Fig. 1), indicating antigen specificity of the tolerogenic effect of OVA-encoding BM transfer. We observed no change in the proportion of donor- type DCs in either non-Tg or 11c.OVA BM recipients after OVA/QuilA challenge (Fig. 2F), indicating that the lack of OVA257–264-specific CTL activity in 11c.OVA BM recipients (Fig. 2E) is attributable to inactivation rather than redirection of OVA-specific CTL activity to cytotoxic destruction of OVA-expressing DCs. BM Transfer Is Effective at High CD8+ Memory T-Cell Frequencies For determination of whether the effectiveness of Ag- encoding BM transfer was limited if Ag-specificTmem were highly abundant, a large number (4 3 107) of OT-I Tmem were transferred and then non-Tg or 11c.OVA BM was transferred. OVA/QuilA challenge, 4 weeks after BM transfer, showed that while OT-I Tmem retained responsiveness and the capacity to proliferate and produce IFN-g in recipients of non-Tg BM (Fig. 2G and H), this was completely ablated in recipients of OVA-encoding 11c.OVA BM (Fig. 2G and H). This indicates that Ag-encoding BM transfer is an effective strategy to achieve antigen-specifictolerance, particularly for Tmem, even when the frequency of Ag- specific CD8+ Tmem is very high, as might occur in T1D.

Spleens were analyzed 1 week after OVA/QuilA challenge (C, D, and F). In vivo CTL was performed 1 week after challenge (E). G and H: OT-I Tmem (4 3 107, CD45.1+/CD45.2+) were transferred to non-Tg mice (CD45.1+), and 1 week later mice were irradiated (300cGy) and non-Tg or 11c.OVA BM (CD45.2+,107) was transferred. Four weeks Figure 2—Inactivation of memory OT-I T-cell populations by after, BMT mice were sham or OVA/QuilA challenged and, 1 week ﬂ OVA-encoding BMT. A–F: OT-I Tmem (CD45.1+/CD45.2+) were later, analyzed by ow cytometry. Total number of OT-I per spleen G + H transferred to non-Tg mice (CD45.1+), and 1 week later mice were ( ) and total IFN-g OT-I per spleen ( ) were determined. Data are 6 n – irradiated (300cGy) or not and non-Tg or 11c.OVA BM (CD45.2+, pooled from 3 experiments and represent mean SEM ( =6 11) A 6 B–F 107) was transferred. Six weeks after, BMT mice were sham or OVA ( ) or individual mice with mean SEM ( ) or pooled from two 6 challenged and, 1 week later or as indicated, analyzed by ﬂow separate experiments showing individual mice with mean SEM G H cytometry. OT-I T cells (CD45.1+CD45.2+CD8+Va2+) were serially ( and ). monitored in blood by cytometry (A) or 1 week after challenge (B). 1332 Reversal of Autoimmune Islet Rejection by BMT Diabetes Volume 65, May 2016

Transplantation of Antigen-Encoding BM Prevents induction. To define whether engraftment of HSPCs was Islet Rejection required and to negate any contribution from differenti- We sought to determine whether ablation of a cognate ated cells cotransferred in BM such as Tregs, sort-purified + 2 preexisting memory CD8 T-cell response might prevent lin ve/c-kit+ve HSPCs were transferred to mice carrying memory T-cell–mediated rejection of “replacement” tissues. OT-I Tmem under engrafting (300cGy) or nonengrafting For islet transplants, this would model the autoimmune (no irradiation) conditions and tolerance induction was component of responses mounted against islet grafts in tested by OVA/QuilA challenge. Donor-type leukocyte ac- T1D where, here, OVA substitutes for b-cell autoantigens. cumulation, indicating engraftment, was lower in HSPC + Memory CD8 T cells were established by Tmem transfer, than BM recipients (Fig. 4A), as anticipated from previous and 1 week later mice were left untreated or irradiated studies (22). However, 5 weeks after HSPC or BM trans- (300cGy) and non-Tg or 11c.OVA BM was transferred. Four fer, total spleen cell number did not differ between groups weeks later, mice were rendered diabetic by streptozotocin (Fig. 4B). Whether mice received 11c.OVA HSPCs or BM, +ve treatment, islets from syngeneic OVA RIP.mOVA mice OVA/QuilA elicited little expansion of the memory OT-I 2ve expressing OVA in pancreatic b-cells or OVA littermate population as expected (Fig. 4C). Substantial OT-I expan- controls were transplanted under the kidney capsule, and sion, similar to that in non-Tg BM or HSPC recipient glycemia was monitored to determine islet graft survival. controls, occurred, however, if mice received HSPC- 2ve OVA islets transplanted as technical controls survived depleted 11c.OVA BM under engrafting conditions or long-term, stably restoring euglycemia in a graft-dependent 11c.OVA HSPCs under nonengrafting conditions. To- manner (Fig. 3A and Supplementary Fig. 2). When grafted gether, these results indicate that engraftment of into OT-I Tmem recipients to which control non-Tg BM had antigen-encoding HSPCs is necessary and sufficient for been transferred, five of eight OVA-expressing islet grafts antigen-specific CD8+ tolerance and nonengrafting com- failed to stably restore euglycemia, and graft failure occurred ponents of BM do not contribute. A prerequisite for soon after transplantation in most cases (Fig. 3A and Sup- tolerance induction is steady-state antigen presenta- plementary Fig. 2). In contrast, in mice carrying memory tion by antigen-expressing DCs (15). Transfer of CFSE- OT-I T cells that received OVA-encoding 11c.OVA BM, eight labeled OT-I T cells revealed that 11c.OVA BM transfer, led of nine islet grafts stably restored euglycemia (Fig. 3A and to long-term presentation of OVA determinants for at least Supplementary Fig. 2). Insulin-positive cells were prominent 12 weeks in recipient mice (Fig. 4D). Furthermore, in addi- in islet grafts of mice that received 11c.OVA BM (Fig. 3B). tion to ablating responsiveness of antigen-specificmemory To further define the ability of antigen-encoding BM CD8+ T cells, OVA-encoding BM establishes a long-term transfer to prevent rejection of transplanted antigen- tolerogenic environment in recipients (Supplementary Fig. expressing tissues, we compared survival of allogeneic and 3) that would prevent reemergence of targeted antigen- OVA-expressing skin grafts. OT-I Tmem were transferred to specificT-cellpopulations. non-Tg recipients and non-Tg or 11c.OVA BM was transferred after low-dose irradiation. Six weeks later, recipients Ablation of Memory OVA-Specific CD8+ T-Cell and no-BMT controls were double grafted with skin from Responses by OVA-Encoding BM Does Not Establish allogeneic BALB/c and syngeneic K5.mOVA donors express- OVA-Specific “Regulation” ing OVA in skin. Control, no-BMT Tmem recipients Using irradiation to facilitate BM engraftment has been rejected .80% of OVA-expressing skin grafts rapidly, as associated with generation or expansion of CD4+ or CD8+ did recipients of non-Tg BM. Remaining grafts promptly Treg populations that modulate subsequent immune re- rejected in response to OVA/QuilA challenge 100 days sponses (23,24). To determine whether OVA-encoding after placement (Fig. 3C), indicating residual OVA-specific BM transfer induced a regulatory response that inhibited memory. Profoundly, almost two-thirds of OVA-expressing subsequent OVA-specific T-cell activation, we adapted a skin grafts survived long-term in the 11c.OVA BMT group widely used “regulation” assay and compared proliferation (Fig. 3C), and none rejected in response to OVA/QuilA chal- of a second adoptively transferred population of OVA- lenge, indicating robust graft acceptance in this group. In all specific CD8+ T cells in mice where OVA-specific memory mice tested, allogeneic skin rejected with a similar tempo CD8+ T cells had been inactivated or not by BM transfer. (Fig. 3D), indicating no gross immunosuppression that Mice were injected with CD45.2+ OT-I Tmem or not and might contribute to a difference between test groups. These 1 week later irradiated (300cGy), and non-Tg or 11c.OVA data demonstrate successful induction of tolerance, which BM was transferred. Six weeks later, a new cohort of alleviates Tmem attack of antigen-expressing islets and congenically distinct (CD45.1+) CFSE-labeled naive OT-I skin, addressing a key challenge facing islet transplantation cells was transferred as the “test” population, and CFSE and personalized b-cell replacement therapies. dilution was measured 3 days later. For provision of a consistent source of antigen stimulation between recipi- Tmem Inactivation Requires Engraftment of Antigen- ents of non-Tg and OVA-encoding BM, some mice were Encoding Hematopoietic Progenitors immunized subcutaneously with OVA/QuilA. In non-Tg As gene-engineered BM transfer could have clinical applica- BM recipients, CFSE dilution indicated that OT-I T cells tion, we sought to understand the mechanisms of tolerance divided only after immunization (Fig. 5A, left and center diabetes.diabetesjournals.org Coleman and Associates 1333

Figure 3—Antigen-encoding BM transfer promotes survival of antigen-expressing islet and skin grafts. A and B: OT-I Tmem (4 3 107) were transferred to non-Tg mice. One week later, mice were irradiated (300cGy) and non-Tg or 11c.OVA BM was transferred. Four weeks after BM transfer, mice were rendered diabetic with streptozotocin (200 mg/kg) and islets from RIP.mOVA mice transplanted under the kidney capsule. BG was monitored and rejection determined as 2 consecutive readings >16.6 mmol/L. Graft sites were embedded, sectioned, and stained for insulin (B). C and D: OT-I Tmem were transferred to non-Tg mice. One week later, mice were irradiated (300cGy) and non-Tg or 11c.OVA BM was transferred. Six weeks after BMT, mice were double grafted with BALB/c and K5.mOVA skin. Mice retaining skin grafts at 100 days postgrafting were immunized with OVA/QuilA. Data are pooled from 2–3 experiments per group. panels) and proliferation was most extensive in LN drain- However, after immunization, proliferation of test T cells ing the site of immunization (Fig. 5A, center panels). In was similar in draining LN of non-Tg and 11c.OVA BM nondraining LN where immunization-derived OVA was recipients (Fig. 5A, center and right upper panels), and not presented, proliferation was most extensive in 11c.OVA this was not altered by the presence of the initially trans- BM recipients (Fig. 5A,centerandrightlowerpanels). ferred, but inactivated, OT-I Tmem (Fig. 5B and C). 1334 Reversal of Autoimmune Islet Rejection by BMT Diabetes Volume 65, May 2016

Deletion Contributes to Ablation of CD8+ Memory Responses After Transfer of Antigen-Encoding BM Serial analyses of peripheral blood (e.g., Fig. 1A) indicated that after an initial expansion phase, the population of OT-I T cells may have been contracting, possibly through deletion. When we serially sampled blood over time, the frequency of OT-I increased similarly in non-Tg and 11c.OVA BM recipients for 3–4 weeks but then appeared to contract in 11c.OVA BM recipients (Fig. 6A). In spleens, where only late time points were analyzed, a similar pattern was apparent and the number of OT-I T cells significantly dropped in 11c.OVA BM recipients between 28 and 49 days after BM transfer (Fig. 6B). To- gether, these observations were consistent with a pattern of OT-I Tmem expansion followed by deletion in 11c.OVA BM recipients that would be anticipated based on previous studies (22). As host T-cell dynamics after low-dose irradiation might confound this analysis, we tested the contribution of deletion more directly. The proapoptotic bcl-2 family member bim is a crucial mediator of DC-induced T-cell deletion (25). Therefore, we transferred deletion-resistant 2 2 OT-I.bim / Tmem or deletion-competent OT-I Tmem, and 1 week later, BM from non-Tg or 11c.OVA mice was 2 2 also transferred. OT-I and OT-I.bim / T cells were enumerated in peripheral blood 1 day before and at regular intervals after BM transfer. In recipients of non-Tg BM, few OT-I T cells accumulated in blood, and the number of these diminished slowly with time (Fig. 6C). In 11c.OVA BM recipients, wild-type (WT) OT-I T cells expanded and then began to diminish in number consistent with slow contraction of the population. On the other hand, deletion- impaired bim-deficient OT-I T cells continued to expand and accumulated in many-fold greater numbers than WT equivalents in blood (Fig. 6C) and in spleen (Fig. 6D). Together these results demonstrate that bim-mediated deletion limits OT-I T accumulation after transfer of 11c.OVA Figure 4—Termination of CD8+ Tmem responses requires HSPC BM and enforces contraction of the OVA-specificTmem engraftment. A–C: OT-I Tmem were transferred to non-Tg mice that 2 population. were irradiated (300cGy) or not 1 week later and BM, lin ve/c-kit+ve HSPCs, or BM depleted of lin2ve/c-kit+ve HSPCs transferred as indicated. Four weeks after BM transfer, mice were challenged or not Residual, Unresponsive CD8+ Memory Cells Exhibit a A with OVA/QuilA and, 1 week later, analyzed. Donor engraftment ( ), CD5high Phenotype Distinct From “Exhaustion” total spleen cell number (B), and OT-I in spleen (C) were determined by flow cytometry. D: Mice were irradiated (300cGy), and non-Tg or Although OT-I Tmem undergo deletion after 11c.OVA BM 11c.OVA BM (107) was transferred. CFSE-labeled naive OT-I T cells transfer, the number of residual OT-I Tmem in spleen is were transferred as indicated and analyzed 3 days later. Prolifera- similar to or higher than that in recipients of non-Tg BM 6 n tion index is indicated (mean SD, = 4). Data are pooled from 4 (e.g., Figs. 1C and 2C), but functional studies (Figs. 1 and experiments (A–C) and show individual mice (mean 6 SEM) (n =4 [sham] or 8 [OVA/QuilA] per group) or representative of 4 mice from 2) indicate that these cells are inactivated. Long-term 2 separate experiments. expression of OVA by DC in 11c.OVA BM recipients (Fig. 5) could induce inactivation by “exhaustion” or other mechanisms. For investigation of this, OT-I Tmem were Together, these observations show that 11c.OVA BM transferred to recipient mice and then 11c.OVA or non- transfer, whether or not in the presence of OT-I Tmem, Tg BM was transferred and residual Tmem phenotyped had not established OVA-specificregulation.Inline 4 weeks later. OT-I Tmem were also transferred to with this, no significant differences were observed unirradiated 11c.OVA or non-Tg controls (Fig. 7) (no- in the proportion of CD4+CD25+FoxP3+ Treg be- BMT) for parallel analysis. “Exhaustion” is a prominent tween groups, and transfer of 11c.OVA did not lead mechanism inhibiting CD8+ T-cell effector function, to differentiation of OT-I Tmem to CD8+FoxP3+ Treg particularly in chronic viral infection and cancer, where (Fig. 5D). “inflammation” or upregulation of coinhibitory ligands diabetes.diabetesjournals.org Coleman and Associates 1335

Figure 5—Transfer of OVA-encoding BM does not establish OVA-specific regulation. A–C: OT-I Tmem were or were not transferred to non- Tg recipients. One week later, mice were irradiated (300cGy) and non-Tg (C57BL/6) or 11c.OVA BM was transferred. Six weeks after BM transfer, 5 3 106 naive CFSE-labeled OT-I cells were transferred and mice immunized with OVA/QuilA subcutaneously. CFSE dilution in the draining (inguinal) and nondraining (pooled axillary, mesenteric) LN was assessed 3 days later by flow cytometry. D: OT-I Tmem were transferred to non-Tg mice that were irradiated (300cGy) and injected intravenously with BM from non-Tg or 11c.OVA donors (BMT) 1 week later. Controls (no-BMT) were non-Tg or 11c.OVA mice injected with Tm in parallel with BM transfer test mice. Four weeks after BM transfer, spleen cells were examined by flow cytometry. Data are representative of 6 mice in 2 separate experiments (A) or pooled from two separate experiments (B–D). Lines depict mean 6 SEM (n =4–6 per group). 1336 Reversal of Autoimmune Islet Rejection by BMT Diabetes Volume 65, May 2016

Figure 6—Deletion is an important contributor to ablation of preexisting memory CD8+ T-cell responses. A and B: OT-I Tmem (CD45.1+/ CD45.2+) were transferred to non-Tg mice (B6.SJL, CD45.1+) and, 1 week later, mice were irradiated (300cGy) and injected intravenously with BM (107) from non-Tg (C57BL/6, CD45.2+) or 11c.OVA (CD45.2+) donors. At the indicated time points, blood or spleen was examined by flow cytometry. C and D: WT or bim2/2 OT-I Tmem (CD45.2+,23 106) were transferred to non-Tg (B6.SJL, CD45.1+) mice. One week later, mice were irradiated (300cGy) and injected intravenously with BM (107) from non-Tg (B6.SJL, CD45.1+) or 11c.OVA (CD45.1+) donors. OT-I T cells were monitored in blood the day before BM transfer or at the indicated points (C) or in spleen at the end of the experiment (38 days post-BMT) (D). Data are pooled from 2–3 experiments per time point with typically 2–4 mice per group, some of which were sham immunized (PBS/QuilA) at day 42 (A and B), or pooled from 2 experiments (C and D). Data show individual points or mean 6 SEM. B: *11c.OVA BM 49 days after BM transfer is significantly lower than 11c.OVA BM 28 days after BM transfer (P < 0.05). C: **OT-I.bim2/2 11c.OVA BM is significantly greater than WT OT-I 11c.OVA BM transfer at days 23 and 37 (P < 0.001). **OT-I.bim2/2 11c.OVA BM days 23 and 37 significantly greater than days 11 and 18 (P < 0.001).

may be present (26). As programmed cell death protein-1 however, typically been associated not with chronic viral (PD-1) is considered an archetypal marker of exhaustion infection–induced exhaustion but, rather, with T-cell “tun- and has been implicated in induction and maintenance of ing” in response to chronic antigen exposure. Relatively little tolerance, we examined expression of PD-1 and a range of CD5 was expressed by OT-I in non-Tg BM recipients and other markers associated with exhaustion. In non-Tg BM non-Tg controls, but a small population expressed moderate recipients and non-Tg controls, PD-1 was expressed at low levels (Fig. 7A and B). Almost all residual OT-I in 11c.OVA levels by approximately one-third of residual OT-I cells BM recipients and 11c.OVA controls expressed high levels of (Fig. 7A–C). In 11c.OVA BM recipients, PD-1 was expressed CD5, mostly coexpressed with PD-1, particularly in 11c.OVA by the majority of cells at moderate levels (Fig. 7A–C). In controls (Fig. 7A), which reﬂected overall expression levels of 11c.OVA controls, PD-1 was expressed on almost all resid- CD5 and PD-1 (Fig. 7B and C). Consistent with expression of ual OT-I Tmem and at high levels relative to the other OVA by all DC in the 11c.OVA control OT-I recipients, experimental groups. CD5 negatively regulates intracellular compared with the ;30–40% in 11c.OVA BMT mice, the signaling in lymphocytes and dampens T-cell responsive- overall expression of both PD-1 and CD5 was signiﬁcantly ness by inhibiting TCR signaling (27). CD5 expression has, higher on OT-I T cells from 11c.OVA control recipients diabetes.diabetesjournals.org Coleman and Associates 1337

Figure 7—Residual inactivated OT-I Tmem exhibit a PD-1highCD5high phenotype. A–C: OT-I Tmem were transferred to non-Tg mice that were irradiated (300cGy) and injected intravenously with BM from non-Tg or 11c.OVA donors (BMT) 1 week later. Controls (no BMT) were non-Tg or 11c.OVA mice injected with Tm in parallel with BM transfer test mice. Four weeks after BM transfer, spleen cells were examined by ﬂow cytometry. OT-I T cells were gated as CD45.1+CD8+Va2+ lymphocyte-sized cells. Data are pooled from 2 experiments and show representative dot plots (A), histograms (B), or individual mice (C). Lines denote mean 6 SEM (n = 6).

than those from mice that received 11c.OVA BM transfers detrimental memory T-cell responses are difficult to (Fig. 7B and C). Of a range of surface markers indicative of control using conventional therapies (5). Thus, current T-cell exhaustion, CD160 and to a lesser extent CD244 were therapeutic approaches are poorly effective or have sub- expressed by residual OT-I T cells only in non-Tg BM recip- stantial detrimental side effects, and new approaches are ients and non-Tg controls, whereas T cell immunoglobulin required to control pathogenic memory T-cell responses. and mucin domain protein 3 (TIM-3) and lymphocyte- Here, using transfer of BM or HSPCs genetically encoding activation gene 3 (LAG-3) were not expressed. cognate antigen expression under mild, immune-preserving pre-BMT conditioning, we demonstrate a highly effective al- DISCUSSION ternative to current therapies. This approach, which antigen- The autoimmune anti–b-cell response reawakened by islet specifically ablates pathogenic memory T-cell responses, pro- transplantation is a significant component of recipient vides an avenue by which autologous HSCT therapies could immune resistance to islet transplants and remains a sig- be modified and optimized to avoid immune suppression nificant clinical challenge. Because memory T cells are while antigen-specifically targeting and preventing the difficult to control with conventional immunosuppression reemergence of detrimental memory T-cell responses. (3) and are resistant to, for example, cytotoxic drugs and Memory T cells represent the immunological driving the inhibitory effects of Tregs (6,7,17,28), unwanted, force in many pathological states. As they are programmed 1338 Reversal of Autoimmune Islet Rejection by BMT Diabetes Volume 65, May 2016 to rapidly expand and generate large effector populations One current approach used clinically to ablate patho- upon antigen re-encounter, they pose a significant barrier to genic T-cell responses in severe autoimmune disease is tissue, cell, and protein replacement therapies in “primed” autologous HSCT. This approach, as currently practiced, individuals. Due to their terminal differentiation and while largely effective, relies on the use of cytotoxic costimulation independence, memory T cells have long been agents, typically with or without T-cell depletion (38), to considered to be resistant to tolerance induction (29,30). ablate the entire immune repertoire and achieve an im- However, pioneering studies including our own, where anti- mune “reset” in the absence of specific mechanisms lead- gen expression has been enforced in APCs, have shown that ing to protective, antigen-specific tolerance. Some studies memory CD4+ or CD8+ T-cell responses can be silenced (8– have tested the principle of antigen-encoding BM transfer 10,31–33). We now advance this to show application to the in a primed setting using “nonmyeloablative” approaches, clinical challenge of b-cell–specific memory T-cell responses but these have exclusively used T-cell depletion or other that represent a strong impediment to islet graft survival. immunodepleting strategies (39–41) so that while recipi- Previous studies showed when using gene-modified BM ents are ultimately reconstituted with a “tolerant” T-cell transfer that restricting antigen expression to differentiated repertoire, this is not immune preserving. Our findings APCs permitted BM engraftment under immune-preserving represent a significant step forward by providing an conditions in naive and primed recipients (22). However, the immune-preserving procedure and predicate a potential fate of established memory T cells was not examined. Mech- clinically applicable approach to ablating unwanted mem- anistic studies here show that with use of this approach, ory T-cell responses. Additionally, this demonstrates that memory CD8+ T-cell populations are inactivated through a induction of mixed allogenic chimerism, with its atten- combination of deletion and induction of unresponsiveness dant risks, is not required. For instance, the risk of graft- in residual undeleted cells. Here, we identified deletion by versus-host disease is negated, as the donor cells are using T-cell donors with a complete functional loss of bim. syngeneic or autologous to recipients. Similarly, for islet Alteredfunctionofsomeofthebclfamilyofapoptosis transplantation in T1D there is no risk of disease exacer- regulators and in deletional pathways has been reported in bation, as replacement islets would be transplanted after the NOD mouse (34) and may exist in humans with T1D. T-cell inactivation is complete. If this procedure were to While potentially contributing to disease, these alterations be used to control autoaggressive T-cell responses in at- aresubtleandnotcomparabletocompletegeneknockout. risk individuals, then transient inflammatory exacerba- In NOD mice, these alterations are unlikely to have a sig- tion could potentially be controlled by a short course of nificant impact on tolerance, as our previous studies (11) immunosuppression. indicate that enforced expression of the proinsulin islet an- Allogeneic islet or whole pancreas transplantation is tigen is sufficient to promote toleranceinthisstrain.Over- currently the only curative therapy for T1D, but its all, the data indicate that residual OT-I remaining after application is limited by several factors including insuffi- deletion, which are rendered unresponsive, exhibit not a cient donor pancreata and the substantial side effects of classical “exhausted” phenotype but, rather, a phenotype long-term immunosuppression. With significant advances that suggests modulation of responsiveness by mechanisms made recently in generation of stem cell–derived b-cells such as tuning—asindicatedbyPD-1andCD5expression— (42,43) and genetically engineered insulin-secreting cells of TCR responsiveness. In the BM transfer setting, ;30– (44,45), it is conceivable that generation of personalized 40% of DCs express OVA and, accordingly, OT-I expression insulin-secreting cells will be feasible in the future. How- of PD-1 and CD5 is reduced compared with OT-I T cells ever, persistent anti-islet memory T-cell responses repre- transferred to 11c.OVA mice, where all DCs express OVA. sent a threat to clinical application of such approaches. This is consistent with previous reports that T-cell tuning is Further development of gene-engineered BM transfer dependent on antigen expression levels (35,36). The data protocols may lead to regimens that can be applied with also indicate that inactivation is effective even when less risk than current organ transplant procedures. Under antigen is expressed by a minority of all DCs. The cur- these circumstances, using gene-engineered BM to ablate rent study specifically examines the response of CD8+ memory T-cell responses could reduce the require- T cells. We have reported previously that the presence ments for immunosuppression in allogeneic islet trans- of cognate CD4+ T cells alters the kinetics but does not plantation or, in the case of personalized autologous prevent tolerance induction (15) and that adoptively trans- replacement, avoid the need for immunosuppression ferred memory CD4+ T cells are inactivated (10) when completely. antigen is targeted to DCs. In studies where CD4+ Tcells The current study was designed to provide proof-of- have been adoptively transferred to recipients of gene- principle preclinical data and to define the mechanistic engineered BM, differentiation of CD4+ Tregs has been contributions to the therapeutic effects of BMT on auto- reported (37). Such induced CD4+ Tregs are likely to aggressive memory CD8+ T cells. Nevertheless, before this provide bystander suppression and provide additional approach can be translated, a number of challenges re- therapeutic value. Therefore, we propose that this ap- main. An ideal translational application of the single an- proach would also be effective for control of pathogenic tigen approach demonstrated in these preclinical studies memory CD4+ T-cell responses. would be to protect insulin-secreting cells engineered diabetes.diabetesjournals.org Coleman and Associates 1339 from non-b-cells such as hepatocytes (44,45). However, 6. Yang J, Brook MO, Carvalho-Gaspar M, et al. Allograft rejection mediated by prevention of T1D progression in at-risk or recent-onset memory T cells is resistant to regulation. Proc Natl Acad Sci U S A 2007;104: human subjects may require a more complex approach. 19954–19959 In an appropriate preclinical model, it would be impor- 7. Afzali B, Mitchell PJ, Scottà C, et al. Relative resistance of human CD4(+) tant to demonstrate simultaneous inactivation of re- memory T cells to suppression by CD4(+) CD25(+) regulatory T cells. Am J – sponses to multiple relevant antigens and/or bystander Transplant 2011;11:1734 1742 8. Kenna TJ, Thomas R, Steptoe RJ. Steady-state dendritic cells expressing tolerance where regulatory T cells induced toward one cognate antigen terminate memory CD8+ T-cell responses. Blood 2008;111: antigen regulate immunity toward additional relevant 2091–2100 local antigens. Additionally, interpretation based on 9. Kenna TJ, Waldie T, McNally A, et al. Targeting antigen to diverse APCs TCR Tg T cells may be limited, and in future studies inactivates memory CD8+ T cells without eliciting tissue-destructive effector endogenous antigen-specific cells would need to be ex- function. J Immunol 2010;184:598–606 amined using tetramers. Humanized mice grafted with 10. Nasreen M, Waldie TM, Dixon CM, Steptoe RJ. Steady-state antigen- relevant human tissue antigens may represent one pre- expressing dendritic cells terminate CD4+ memory T-cell responses. Eur J clinical model where such advances could be made. Fi- Immunol 2010;40:2016–2025 nally, mild conditioning regimens would need to be 11. Steptoe RJ, Ritchie JM, Harrison LC. Transfer of hematopoietic stem cells developed to support HPSC transfer in young individuals encoding autoantigen prevents autoimmune diabetes. J Clin Invest 2003;111: with, or at risk for, T1D. Notwithstanding these chal- 1357–1363 lenges, we demonstrate a highly effective method to 12. Voltarelli JC, Couri CE, Stracieri AB, et al. Autologous nonmyeloablative silence autoaggressive islet-destructive memory T-cell hematopoietic stem cell transplantation in newly diagnosed type 1 diabetes – responses that, with further development, has clinical mellitus. JAMA 2007;297:1568 1576 13. Couri CE, Oliveira MC, Stracieri AB, et al. C-peptide levels and insulin in- potential. dependence following autologous nonmyeloablative hematopoietic stem cell transplantation in newly diagnosed type 1 diabetes mellitus. JAMA 2009;301:1573–1579 14. Hogquist KA, Jameson SC, Heath WR, Howard JL, Bevan MJ, Carbone FR. Acknowledgments. The authors thank Professor Leonard Harrison and T cell receptor antagonist peptides induce positive selection. Cell 1994;76:17–27 Dr. Phillipe Boulliet (Walter and Eliza Hall Institute) and Professor Francis Carbone 15. Steptoe RJ, Ritchie JM, Wilson NS, Villadangos JA, Lew AM, Harrison LC. (University of Melbourne) for providing mice. Cognate CD4+ help elicited by resting dendritic cells does not impair the in- Funding. This work was supported by a Diabetes Australia Research Trust duction of peripheral tolerance in CD8+ T cells. J Immunol 2007;178:2094–2103 general research grant Y13G-JESC (to C.F.J.), JDRF project 32-2008-250 (to 16. Azukizawa H, Kosaka H, Sano S, et al. Induction of T-cell-mediated skin R.J.S.), and National Health and Medical Research Council (NHMRC) project disease specific for antigen transgenically expressed in keratinocytes. Eur J GNT1013066 (to R.J.S.). J.M.F. is supported by the NHMRC of Australia and Immunol 2003;33:1879–1888 the Mater Foundation. R.T. was supported by an Australian Research Council 17. Steptoe RJ, Stankovic S, Lopaticki S, Jones LK, Harrison LC, Morahan G. Future Fellowship and Arthritis Queensland. J.W.W. was supported in part by Persistence of recipient lymphocytes in NOD mice after irradiation and bone grants from the University of Queensland and Cancer Council Queensland and by marrow transplantation. J Autoimmun 2004;22:131–138 a Perpetual Trustees Fellowship. R.J.S. is a recipient of an Australian Research 18. Rahimpour A, Mattarollo SR, Yong M, Leggatt GR, Steptoe RJ, Frazer IH. gd Council Future Fellowship (FT110100372). T cells augment rejection of skin grafts by enhancing cross-priming of CD8 fl Duality of Interest. No potential con icts of interest relevant to this article T cells to skin-derived antigen. J Invest Dermatol 2012;132:1656–1664 were reported. 19. Penko D, Rojas-Canales D, Mohanasundaram D, et al. Endothelial pro- Author Contributions. M.A.C., C.F.J., J.M.F., R.T., P.T.C.C., S.T.G., genitor cells enhance islet engraftment, influence b-cell function, and modulate J.W.W., and R.J.S. designed experiments. M.A.C., C.F.J., J.A.B., N.H.O., D.P., islet connexin 36 expression. Cell Transplant 2015;24:37–48 S.W., D.J.B., and R.G. performed experiments. M.A.C., C.F.J., S.W., and R.J.S. 20. Weninger W, Crowley MA, Manjunath N, von Andrian UH. Migratory properties analyzed experiments. M.A.C., R.T., S.T.G., and R.J.S. wrote the manuscript. of naive, effector, and memory CD8(+) T cells. J Exp Med 2001;194:953–966 R.J.S. is the guarantor of this work and, as such, had full access to all the data in 21. Blake SJP, Ching ALH, Kenna TJ, et al. Blockade of PD-1/PD-L1 promotes the study and takes responsibility for the integrity of the data and the accuracy of adoptive T-cell immunotherapy in a tolerogenic environment. PLoS One 2015;10: the data analysis. e0119483 References 22. Coleman MA, Bridge JA, Lane SW, et al. Tolerance induction with gene- 1. Sprent J, Surh CD. T cell memory. Annu Rev Immunol 2002;20:551– modified stem cells and immune-preserving conditioning in primed mice: re- 579 stricting antigen to differentiated antigen-presenting cells permits efficacy. Blood 2. Monti P, Scirpoli M, Rigamonti A, et al. Evidence for in vivo primed and 2013;121:1049–1058 expanded autoreactive T cells as a specific feature of patients with type 1 di- 23. McFarland HI, Puig M, Grajkowska LT, et al. Regulatory T cells in g abetes. J Immunol 2007;179:5785–5792 irradiation-induced immune suppression. PLoS One 2012;7:e39092 3. Vendrame F, Pileggi A, Laughlin E, et al. Recurrence of type 1 diabetes after 24. Robb RJ, Lineburg KE, Kuns RD, et al. Identification and expansion of highly simultaneous pancreas-kidney transplantation, despite immunosuppression, is suppressive CD8(+)FoxP3(+) regulatory T cells after experimental allogeneic associated with autoantibodies and pathogenic autoreactive CD4 T-cells. Di- bone marrow transplantation. Blood 2012;119:5898–5908 abetes 2010;59:947–957 25. Davey GM, Kurts C, Miller JF, et al. Peripheral deletion of autoreactive CD8 T 4. Diz R, Garland A, Vincent BG, et al. Autoreactive effector/memory CD4+ and cells by cross presentation of self-antigen occurs by a Bcl-2-inhibitable pathway CD8+ T cells infiltrating grafted and endogenous islets in diabetic NOD mice mediated by Bim. J Exp Med 2002;196:947–955 exhibit similar T cell receptor usage. PLoS One 2012;7:e52054 26. Wherry EJ. T cell exhaustion. Nat Immunol 2011;12:492–499 5. Valujskikh A. Targeting T-cell memory: where do we stand? Curr Opin 27. Tarakhovsky A, Kanner SB, Hombach J, et al. A role for CD5 in TCR- Organ Transplant 2008;13:344–349 mediated signal transduction and thymocyte selection. Science 1995;269:535–537 1340 Reversal of Autoimmune Islet Rejection by BMT Diabetes Volume 65, May 2016

28. Casorati G, Locatelli F, Pagani S, et al. Bone marrow-resident memory 37. de Andrade Pereira B, Ackermann M, Chaudhary S, et al. Tolerance of T cells survive pretransplant chemotherapy and contribute to early immune reactivated pathogenic CD4+ T cells by transcriptional targeting of dendritic cells. constitution of patients with acute myeloid leukemia given mafosfamide-purged Gene Ther 2015;22:382–390 autologous bone marrow transplantation. Exp Hematol 2005;33:212–218 38. Coleman MA, Steptoe RJ. Induction of antigen-speciﬁc tolerance through 29. Chung Y, Chang SY, Kang CY. Kinetic analysis of oral tolerance: memory hematopoietic stem cell-mediated gene therapy: the future for therapy of auto- – lymphocytes are refractory to oral tolerance. J Immunol 1999;163:3692 3698 immune disease? Autoimmun Rev 2012;12:195–203 30. London CA, Lodge MP, Abbas AK. Functional responses and costimulator 39. Hosseini H, Oh DY, Chan ST, et al. Non-myeloablative transplantation of – dependence of memory CD4+ T cells. J Immunol 2000;164:265 272 bone marrow expressing self-antigen establishes peripheral tolerance and com- 31. Kreuwel HT, Aung S, Silao C, Sherman LA. Memory CD8(+) T cells undergo pletely prevents autoimmunity in mice. Gene Ther 2012;19:1075–1084 peripheral tolerance. Immunity 2002;17:73–81 40. Eixarch H, Espejo C, Gómez A, et al. Tolerance induction in experimental 32. Liblau RS, Tisch R, Shokat K, et al. Intravenous injection of soluble antigen autoimmune encephalomyelitis using non-myeloablative hematopoietic gene induces thymic and peripheral T-cells apoptosis. Proc Natl Acad Sci U S A 1996; therapy with autoantigen. Mol Ther 2009;17:897–905 93:3031–3036 41.KoHJ,ChungJY,NasaZ,etal.Targeting MOG expression to dendritic cells 33. Higgins AD, Mihalyo MA, Adler AJ. Effector CD4 cells are tolerized delays onset of experimental autoimmune disease. Autoimmunity 2011;44:177–187 upon exposure to parenchymal self-antigen. J Immunol 2002;169:3622– 42. Pagliuca FW, Millman JR, Gürtler M, et al. Generation of functional human 3629 pancreatic cells in vitro. Cell 2014;159:428–439 34. Liston A, Lesage S, Gray DH, et al. Generalized resistance to thymic deletion b 43. Rezania A, Bruin JE, Arora P, et al. Reversal of diabetes with insulin- in the NOD mouse; a polygenic trait characterized by defective induction of Bim. Immunity 2004;21:817–830 producing cells derived in vitro from human pluripotent stem cells. Nat Bio- – 35. Mandl JN, Monteiro JP, Vrisekoop N, Germain RN. T cell-positive selection technol 2014;32:1121 1133 uses self-ligand binding strength to optimize repertoire recognition of foreign 44. Lawandi J, Tao C, Ren B, et al. Reversal of diabetes following trans- antigens. Immunity 2013;38:263–274 plantation of an insulin-secreting human liver cell line: Melligen cells. Mol Ther 36. Stamou P, de Jersey J, Carmignac D, Mamalaki C, Kioussis D, Stockinger B. Methods Clin Dev 2015;2:15011 Chronic exposure to low levels of antigen in the periphery causes reversible 45. Gerace D, Martiniello-Wilks R, O’Brien BA, Simpson AM. The use of b-cell functional impairment correlating with changes in CD5 levels in monoclonal CD8 transcription factors in engineering artiﬁcial b cells from non-pancreatic tissue. T cells. J Immunol 2003;171:1278–1284 Gene Ther 2015;22:1–8