Immunomodulation Followed by Antigen-Specific TregInfusion Controls Islet AutoimmunityCecilia Cabello-Kindelan,1 Shane Mackey,1 Alexander Sands,1 Jennifer Rodriguez,1 Claudia Vazquez,1

Alberto Pugliese,1,2,3 and Allison L. Bayer1,2

Diabetes 2020;69:215–227 | https://doi.org/10.2337/db19-0061

Optimal immune-based therapies for type 1 diabetes(T1D) should restore self-tolerance without inducingchronic immunosuppression. CD41Foxp31 regulatoryT cells (Tregs) are a key cell population capable of facil-itating durable immune tolerance. However, clinical trialswith expanded Tregs in T1D and solid-organ transplantrecipients are limited by poor Treg engraftment withouthost manipulation. We showed that Treg engraftment andtherapeutic benefit in nonautoimmune models requiredablative host conditioning. Here, we evaluated Treg en-graftment and therapeutic efficacy in the nonobese di-abetic (NOD) mouse model of autoimmune diabetesusing nonablative, combinatorial regimens involving theanti-CD3 (aCD3), cyclophosphamide (CyP), and IAC(IL-2/JES6–1) antibody complex. We demonstrate thataCD3 alone induced substantial T-cell depletion, impact-ing both conventional T cells (Tconv) and Tregs, subse-quently followed by more rapid rebound of Tregs. Despiterobust depletion of host Tconv and host Tregs, donor Tregs

failed to engraft even with interleukin-2 (IL-2) support. Asingle dose of CyP after aCD3 depleted rebounding hostTregs and resulted in a 43-fold increase in donor Treg

engraftment, yet polyclonal donor Tregs failed to reversediabetes. However, infusion of autoantigen-specific Tregs

after aCD3 alone resulted in robust Treg engraftmentwithin the islets and induced remission in all mice. Thisnovel combinatorial therapy promotes engraftment ofautoantigen-specific donor Tregs and controls islet auto-immunity without long-term immunosuppression.

The key role played by regulatory T cells (Tregs) in self-tolerance (1,2) and suppression of rejection (3–6) makes

them attractive for tolerogenic cell-based therapies. Mucheffort is being devoted to developing Treg therapy forrecent-onset type 1 diabetes (T1D) and in transplantsettings (7–10). Several groups have established in vitroTreg expansion protocols (11–15); clinical trials with au-tologous, expanded Tregs are ongoing in T1D (9,10) usingunselected, polyclonal Tregs (14,16). A phase 1 studyrevealed that in vitro expanded, autologous Tregs weresafe and tolerable in children with recent-onset T1D,with evidence of improved fasting C-peptide and reducedinsulin requirement 4 months posttreatment. Therapeuticeffects correlated with increased Tregs post-infusion butonly persisted for a short time. A subsequent trial con-firmed the limited persistence of expanded Tregs even aftera second infusion (9,10,17).

Data emerging from these trials highlight the limitationsof protocols that rely solely on Treg infusion without re-cipient manipulation, including immunomodulation andhomeostatic support. In fact, our previous work identifiedcritical requirements for infused Treg engraftment andfunction: 1) generation of peripheral space for engraftment,2) overcoming competition from host Tregs, 3) sufficientinterleukin 2 (IL-2), and 4) antigen availability to selectdisease-relevant Tregs (18–20). Our observations are sup-ported by other studies (for review see Cabello-Kindelanet al. [21]) and should be considered when designing Treg-based clinical trials, whether these Tregs are expandedin vitro or infused after isolation for in vivo expansion.Our previous protocols involved ablative (radiation) condi-tioning of the host for Treg engraftment and therapeuticefficacy. Although effective, ablative conditioning bears

1Diabetes Research Institute, University of Miami Miller School of Medicine, Miami,FL2Department of Microbiology and Immunology, University of Miami Miller School ofMedicine, Miami, FL3Division of Diabetes, Endocrinology, and Metabolism, Department of Medicine,University of Miami Miller School of Medicine, Miami, FL

Corresponding author: Allison L. Bayer, abayer@med.miami.edu

Received 29 January 2019 and accepted 6 November 2019

This article contains Supplementary Data online at http://diabetes.diabetesjournals.org/lookup/suppl/doi:10.2337/db19-0061/-/DC1.

© 2019 by the American Diabetes Association. Readers may use this article aslong as the work is properly cited, the use is educational and not for profit, and thework is not altered. More information is available at http://www.diabetesjournals.org/content/license.

Diabetes Volume 69, February 2020 215

IMMUNOLOGY

AND

TRANSPLANTATIO

ND

ownloaded from

http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 Decem

ber 2021

translational concerns for clinical application that in T1Dwould also involve children. Hence, we sought to developeffective but safer methods to create immunological spaceand enable engraftment of autologous Tregs. To this end, weexplored novel, nonablating, clinically applicable immuno-modulatory regimens that included combinations of a shortcourse of intact aCD3, a single injection of cyclophospha-mide (CyP), and the addition of IL-2 complex (IAC, IL-2/clone JES6-1), which progressively improved engraftment.However, the reversal of recent-onset diabetes in nonobesediabetic (NOD) mice was only observed when donor Tregswere autoantigen specific rather than polyclonal, and thissetting immunomodulation with aCD3 alone resulted in100% durable diabetes remission (.6 months), with robustTreg engraftment within the islets. Thus, our results dem-onstrate that robust engraftment and disease-relevant,antigen-specific Tregs are key requirements for successfulimmunotherapy in a preclinical model of autoimmunediabetes.

RESEARCH DESIGN AND METHODS

MiceNODmice were from The Jackson Laboratory (Bar Harbor,ME). Breeder pairs of NOD.NON-Thy1a/1LtJ andNOD.Cg-Tg(TcraBDC2.5,TcrbBDC2.5)1Doi/DoiJ werepurchased from The Jackson Laboratory and colonies weremaintained at the University of Miami. Animal studieswere performed in accordance with protocol approval bythe University of Miami Institutional Animal Care and UseCommittee.

Cell PurificationsCD41CD251 Tregs were purified from spleen (SP) andlymph nodes (LNs) as previously described (22) fromNOD or Thy1.11 NOD mice. Tregs on average were.92% pure. For remission studies, islet-specificCD41TCRVb41CD251 Tregs were purified from SP andLNs of NOD.Cg-Tg(TcraBDC2.5,TcrbBDC2.5)1Doi/DoiJmice by cell sorting and were on average .98.5% pure.In brief, CD4 T cells were enriched by anti-CD4 MACSmicrobeads (Miltenyi Biotec, San Diego, CA), stained withAlexa Flour 700–conjugated CD4, phycoerythrin (PE)-conjugated Vb4 antibody, and PE-CF594–conjugatedCD25 (BD Biosciences, San Diego, CA) and cell sortedon Beckman Coulter Astrios. Purities were assessed byflow cytometry.

Adoptive Treg TransferFreshly isolated Tregs (0.53 106), either polyclonal or islet-specific Tregs, were adoptively transferred by i.v. injectionsthrough tail vein to NOD mice that received immunomo-dulation in various combinations using intact aCD3(50 mg, clone 2C11) (Leinco Technologies, St. Louis,MO) one time per day for five consecutive days, a singleinjection of CyP (200 mg/kg) (Sigma-Aldrich, St. Louis,MO), and IL-2 complex (IAC) given every other day for1 week starting at the time of Treg infusion, or mice were

left untreated (Fig. 3A). IAC was prepared using recombi-nant murine IL-2 with anti–IL-2 monoclonal antibody(mAb; clone JES6-1A12) (eBioscience, San Diego, CA)and incubated at a molar ratio of 2 to 1 (1 mg IL-2 and5 mg Ab) for 15 min at room temperature. The resultingIAC was intraperitoneally injected in 200 mL PBS as in-dicated in each illustration.

Remission StudiesRecent-onset diabetic NOD mice (.250 mg/dL for twoconsecutive blood readings) received a single insulin pelletsubcutaneously to initially control blood sugars but noinsulin thereafter. These mice received immunomodula-tion in various combinations using intact aCD3, CyP, IAC,and/or Treg infusion (autologous, polyclonal NOD, or islet-specific NOD Tregs), or mice were left untreated as in-dicated in each illustration. Diabetes reoccurrence wasmonitored for 6 months. Mice were tested two to threetimes per week for weight and glycosuria. The absence ofglycosuria and hyperglycemia indicates lack of diabetesrelapse.

Skin TransplantationSkin grafting was performed as a modification of theBillingham and Medawar technique (23). Anesthetizedmice received full-thickness donor skin (prepared fromdorsal tissue from ear) on separate graft beds on the backof NOD mice 6 months after aCD31BDC2.5 Treg in-fusion. Each recipient received two grafts, syngeneic NODand allogeneic B6 grafts. A bandage was placed over thesetwo grafts for 7 days. Grafts were monitored every otherday and scored rejected when .75% of the original graftwas lost or became necrotic as assessed by visual exam-ination.

Antibodies and Flow Cytometry AnalysisSP and LN were made into single cell suspensions bymechanic disruption. Pancreatic LN (pLN) and pancreaswere digested with collagenase D (2 mg/mL) (Roche) at37°C for 30 min. Equal volumes of collected blood wereused for peripheral blood mononuclear cells and purifiedfrom whole blood on a Ficoll-Paque PLUS gradient (GE LifeSciences). Red blood cells from tissues were lysed withACK lysing buffer. Total cell counts were performed ontissues and blood using a hemocytometer.

Lymphocytes were stained with LIVE/DEAD FixableNear-IR according to the manufacturer’s instructions(Invitrogen/Thermo Fisher Scientific, Eugene, OR) andwashed twice with PBS. Cells were then incubated withrat anti-CD16/32 (clone 24G2) to block nonspecific Abbinding followed by cell surface staining with fluorescence-conjugated Abs against mouse CD45, CD19, CD4, CD8,Thy1.2, CD25 (BD Biosciences), and Thy1.1 (BioLegend,San Diego, CA). Intracellular Foxp3 (clone FJK-16s) stain-ing was performed according to the manufacturer’sinstructions (eBioscience) along with Ki67 staining (cloneB56; BD Biosciences).

216 Regulatory T-cell Immunotherapy Diabetes Volume 69, February 2020

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

Tetramer staining was performed on islet lympho-cytes. Murine islets were isolated as described previously(24) with minor modifications. Animals were killed undergeneral anesthesia, and pancreas was exposed andinjected with Hanks’ balanced salt solution containing0.5 mg/mL collagenase P (Sigma-Aldrich) via the mainbile duct until distension was achieved. Digestion wasperformed at 37°C for 10–15 min with gentle agitationand terminated by the addition of cold buffer (RPMIcontaining 10% FCS and 2 mmol/L L-glutamine). Thetissue was filtered through 70-mm mesh, placed on Euro-Ficoll (Mediatech) gradients by centrifugation at 2,000rpm for 15 min, and washed twice with PBS. Single cellsuspensions were incubated with rat anti-CD16/32 (clone24G2) followed by staining with PE-conjugated BDC2.5tetramer Ab [I-A(g7) BDC2.5 mimetope RTRPLWVRME;National Institutes of Health Tetramer Core] for 3 h at

37°C, and fluorescence-conjugated Abs against cell sur-face mouse CD45, Thy1.2, CD4, CD8, and CD25 wereadded to the last 30 min of incubation. Cells were thenstained with LIVE/DEAD Fixable Near-IR followed byintracellular Foxp3 staining.

FACS analysis was performed using a BD BiosciencesLSRII and analyzed with Diva or Kaluza software. The totalnumber of events collected was between 50,000 and200,000 cells, except for pancreas samples in which entiresamples were collected, and CD451 cells were set as thestorage gate.

Statistical AnalysisOne-way ANOVA was followed by Dunnett multiple com-parison test. Two-way ANOVA was followed by Sidak mul-tiple comparison test. Unpaired Student t test was performedin which aCD3 was compared with aCD31CyP at each time

Figure 1—T-cell depletion and rebound after immunomodulation with aCD3, CyP, and IAC in the circulation. A: Experimental scheme.Percentage of Thy1.2 (B), CD4 (C), CD8 (D), and CD41Foxp31 (E) in the gated CD451 population and percentage of CD41Foxp31 in thegated CD4 T cells (F) (shaded area 1 and 2 indicate increase in the percentage of CD41Foxp31 in the CD4 T cells) in prediabetic female NODmice receiving aCD3 (solid lines) or aCD31CyP (dashed lines). G: Percentage of Ki671 in the gated CD41Foxp31, CD41Foxp32, and CD8T cells in the peripheral blood. Time points examined are 2, 4, 7, 11, 15, 18, 22, 25, 32, and 39 days. n 5 5–6 mice per group. In E, the grayarrow indicates recovery in the Treg compartment, and the shaded area indicates the decrease in CD41Foxp31 host Tregs following single CyPinjection after aCD3 treatment. In F, shaded area 1 indicates the increase in CD41Foxp31 cells in the gated CD4 T cells after CD3 alone, andshaded area 2 indicates the increase in CD41Foxp31 cells in total CD4 T cells after CD31CyP. *P, 0.0001;1P, 0.001; @P, 0.01; ^P,0.05. One-way ANOVA followed by Dunnett multiple comparison test compared with day 0. Two-way ANOVA followed by Sidak multiplecomparison test; aCD3 compared with aCD31CyP at each time point. #P , 0.05. d, day.

diabetes.diabetesjournals.org Cabello-Kindelan and Associates 217

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

point. Comparisons yielding P , 0.05 were considered statis-tically significant. Survival curves were subjected to Mantel-Cox log-rank test. The P value is indicated on the graph.

Data and Resource AvailabilityThe data sets generated during and/or analyzed during thecurrent study are available from the corresponding authorupon reasonable request. All resources, including animalmodels and reagents, are commercially available.

RESULTS

A Combinatorial Regimen of aCD3 and CyP CreatesSpace in the Host Treg Compartment in FemaleNOD MiceIn previous studies, intact aCD3 was more effective at de-pleting T cells than the F(ab9)2 form (25–27); while regimensvaried in timing and duration of administration, depletion wasdose dependent. Yang et al. (27) reported that a single injec-tion of intact aCD3 led to transient T-cell depletion in both

C57BL/6 and NOD mice, but NOD required a higher dose(50 mg) than C57BL/6 mice; depletion was age dependent inNOD mice. Moreover, in NOD mice, 50 mg of intact aCD3more efficiently depleted CD4 T cells than 100 mg of F(ab9)2-aCD3 (27). Hence, we used intactaCD3 antibody (50-mg dose)as a depleting agent in our study.We investigated the effects ofaCD3 therapy on conventional T-cell and Treg compartments inthe circulation of 5- to 6-week-old prediabetic NOD femalemice. We used a 5-day course of aCD3 and followed the effectsfor 39 days (Fig. 1A). aCD3 readily decreased conventionalT cells, including CD41Foxp32 and CD8 T cells (Fig. 1B–D,solid lines). CD41Foxp32 T cells returned to normal levels ataround 39 days after the first aCD3 dose; CD8 T cells did notfully recover. Our assessments at earlier time points (as early as2days) revealed that Treg depletion occurs very early post-aCD3(Fig. 1E); in fact, Tregs returned to normal levels within 7 days ofthe first dose (Fig. 1E, solid lines, gray arrow). Thus, wedemonstrated a previously unknown early Treg depletionpost-aCD3. We also showed that Tregs rebound faster than

Figure 2—Treg depletion and rebound in the tissues after aCD3 and CyP immunomodulation. A: Experimental scheme. Number of CD81 andCD41Foxp32 T cells (per 13 106 lymphocytes) in peripheral blood (B), pLNs (C), and pancreas (D). Number of CD41Foxp31 cells (per 13 106

CD451 lymphocytes) and percentage of CD41Foxp31 in the gated CD4 T cells in peripheral blood (E), pLNs (F ), and pancreas (G) in thepresence or absence of CyP of young, prediabetic NOD mice (aged 5–6 weeks). n5 3–6 mice per group. In E and F, gray arrows indicate anearly decrease in the number of CD41Foxp31 cells after the start of aCD3 in the blood and pLNs. In G, hatched arrow indicates an initialincrease in the number of CD41Foxp31 cells after the start of aCD3 in the pancreas. *P, 0.0001;1P, 0.001; @P, 0.01; ^P, 0.05. One-way ANOVA followed by Dunnett multiple comparison test compared with day 0. Unpaired Student t test, aCD3 compared with aCD31CyPat each time point. n 5 3–6 mice per group. d, day.

218 Regulatory T-cell Immunotherapy Diabetes Volume 69, February 2020

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

conventional T cells, which recover around day 32–39, leadingto an increased proportion of CD41Foxp31 cells in the CD41

T-cell compartment (Fig. 1F, solid line, shaded area 1).Penaranda et al. (28) noted that treatment with F(ab9)2-aCD3induced a transient increase in the proportion of Foxp31 cellsin the CD4 T-cell compartment that was accompanied bya slight reduction in the total count of Tregs, but that theincreased Foxp31 proportion was due to greater depletion of

the CD41Foxp32 conventional T cells. In our study, mostT cells were initially proliferating after depletion with intactaCD3 (Fig. 1G); however, only Tregs maintained robust pro-liferation, including a much higher Ki67 mean fluorescenceintensity, over the 39-day follow-up (Supplementary Fig.1), which may contribute to the increased proportion of Foxp31

cells in the CD41 T-cell compartment after aCD3 treatmenttogether with the slow recovery of the CD41Foxp32 T cells.

weeks )

Figure 3—Adoptive transfer of Tregs leads to engraftment after immunomodulation in young, prediabetic NODmice.A: Experimental scheme.B: Percentage of Thy1.11 donor Treg engraftment among the total gated CD41Foxp31 T cells in peripheral blood. C: Number of donor Tregs/1 3 106 CD451 lymphocytes in peripheral blood. D: Percentage of Thy1.11 donor Treg engraftment in the total gated CD41Foxp31 T cells4 weeks post–Treg infusion in SP, LN, pLN, and pancreas (PAN) of young, female prediabetic NOD mice receiving aCD31Treg,aCD31Treg1IAC, aCD31CyP1Treg, or aCD31CyP1Treg1IAC. IAC 5 anti–IL-2 (JES6)1rmIAC. E: Percentage of Thy1.11 donor Tregs inthe gated CD41Foxp31 T cells in mice receiving the aCD31CyP1Treg1IAC regimen and CD41Foxp31 in the gated total CD4 T cells in micereceiving the aCD3 regimen in the peripheral blood. F: Percentage of CD41Foxp31 in the gated CD451 population. G: CD41Foxp31 in thegated CD4 T cells in the peripheral blood of young, prediabetic female NODmice (aged 5–6weeks) in NODmice receiving aCD3, aCD31IAC,aCD31CyP, aCD31Treg, aCD31Treg1IAC, aCD31CyP1Treg, or aCD31CyP1Treg1IAC. n5 5–6mice per group. *P, 0.0001;1P, 0.001;@P, 0.01; ^P, 0.05. B and C: Multiple Student t test compared with aCD31Treg. F and G: Two-way ANOVA followed by Dunnett multiplecomparison test compared with aCD3. d, day; rmIAC, recombinant murine IAC.

diabetes.diabetesjournals.org Cabello-Kindelan and Associates 219

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

We next examined T-cell compartments after aCD3treatment up to 15 days in pancreas, SP, LN, pLN, andPeyer’s patches (PP), lamina propria lymphocytes (LPLs),and intraepithelial lymphocytes (IELs) in the small in-testine (Fig. 2A). Similar to blood, aCD3 therapy rapidlydepleted CD41Foxp32 and CD81 T cells in LN and pLN,with,30% recovery from baseline at the end of the 15-dayfollow-up (Fig. 2B and C and Supplementary Fig. 2). Incontrast, there was an initial CD81 T-cell increase in SP,PP, IEL, and LPL followed by a decrease beginning on day4, while CD41Foxp32 T cells were rapidly decreased afteraCD3 treatment (Supplementary Figs. 2 and 3). There wasa slow recovery in the small intestine, with ,35% ofCD41Foxp32 T cells recovering from baseline, excludingT cells in SP with;36–58% recovery (Supplementary Figs.2 and 3). In pancreas, there was an early increase followedby decrease of CD41Foxp32 T cells beginning on day 4,with no recovery by the end of the follow-up period. CD81

T cells had a more persistent increase before returning tobaseline, leading to an inversion of the CD4:CD8 T-cellratio after aCD3 treatment (Fig. 2D).

Resembling effects in blood (Fig. 2E, left panel, grayarrow), aCD3 treatment was immediately followed bya decrease of CD41Foxp31 Tregs in LN and pLN, with44–60% recovery by day 15 (Fig. 2E and F and Supple-mentary Fig. 2, gray arrows). The slower recovery ofCD41Foxp32 T cells in these tissues compared with Tregsled to increased proportions of CD41Foxp31 in the CD41

compartment (Fig. 2E and F and Supplementary Fig. 2).There was delayed Treg depletion in SP (Supplementary Fig.2, white arrow) compared with blood and LNs, but re-covery was still around 50% at the end of follow-up (Fig. 2and Supplementary Fig. 2). In contrast, aCD3 treatmentresulted in an initial Treg increase in the pancreas (Fig. 2G,hatched arrow) followed by delayed depletion starting atday 11, with no recovery of Tregs accompanied by anincrease in the proportion of CD41Foxp31 cells in theCD41 T-cell compartment. A similar pattern was seen inLPL, IEL, and PP, with an initial Treg increase, but Tregs didbegin to recover (Supplementary Fig. 3). Treg changes post-aCD3 did not appear to be solely due to redistribution ofcirculating Tregs to tissues. By linear regression, we founda link between Treg rebound and proliferation in pLN(Supplementary Fig. 4A), pancreas, LPL, and peripheralblood (not shown). However, there was no correlationobserved in LN (Supplementary Fig. 4B), SP, IEL, andPP (not shown), suggesting that Treg rebound may onlyoccur in certain compartments, and may be partlyexplained by Treg redistribution. Moreover, consistentwith earlier reports (25,29), lymphocytes isolated onday 15 from the SP or pancreas of treated, prediabeticmice exhibited inflammatory cytokine production with18.7 6 1.9% IFNg- and 23.4 6 7.5% TNFa-producingcells from the pancreas and 19.1 6 2.5% IFNg- producingcells from the SP (Supplementary Fig. 5). Circulating cellsof aCD3-treated mice showed little production of thesecytokines. Lymphocytes from untreated mice showed very

little production of these cytokines from any of the tissuesexamined. Overall, this extended longitudinal evaluationin tissues reveals the complex effects of aCD3 treatmenton Tregs, which vary in different tissues and include de-pletion, rebound, and redistribution. Moreover, aCD3treatment does not appear to open the Treg compartmentfor extensive periods of time in the pancreas or lymphoidtissues, suggesting that this treatment alone may notpromote robust polyclonal Treg engraftment.

A Single Dose of CyP to aCD3 Therapy DepletesRebounding Host Tregs

We have previously shown that robust donor Treg engraft-ment requires not only creating peripheral space but alsolessening competition from rebounding host Tregs (20).Because aCD3 treatment induces only an early, temporaryTreg depletion that is followed by rebound, achievingoptimal engraftment of autologous donor Tregs may re-quire further manipulation to control rebounding popu-lations. CyP is known to eliminate proliferating cells inresponse to alloantigen but also depletes Tregs because oftheir higher proliferation rates compared with naïve T cells(30,31). To determine whether CyP could effectively de-plete rapidly rebounding host Tregs post-aCD3, a singleCyP injection was given at day 10 (Fig. 1A). CyP after aCD3significantly decreased host Tregs by 62% (day 11, two-wayANOVA, P 5 0.0215) for ;5 days post-CyP in blood (Fig.1E, dashed lines), thus creating a more supportive Tregenvironment (Fig. 1E, shaded area). CyP also depletedCD41 and CD81 T cells, which remained lower comparedwith aCD3 alone 39 days after the first aCD3 dose (Fig.1B–D, dashed lines). After CyP, a substantial increase inthe proportion of CD41Foxp31 cells in the CD41 T-cellcompartment was observed during days 18–32 (Fig. 1F,dashed line, shaded area 2). Moreover, CyP after aCD3decreased rebounding host Tregs by 21–82% in SP, LN, PP,and IEL (Supplementary Figs. 2 and 3). In contrast, hostTregs showed a modest increase in pancreas and pLN afterCyP (Fig. 2F and G). Overall, these findings highlight thata single dose of CyP after aCD3 can effectively depleterebounding host Tregs, which are expected to compete withinfused donor Tregs, and can be used to synergize withaCD3 therapy to open the Treg compartment to promoterobust Treg engraftment.

A Combinatorial Regimen of aCD3 and CyP PromotesRobust Engraftment of Polyclonal, Autologous DonorTregs in NOD MiceWe tested whether these combinatorial therapies (Fig. 3A)supported engraftment of polyclonal, autologous donorTregs in 5- to 6-week-old female NOD mice. Despite ro-bust T-cell depletion with aCD3 alone, donor congenicThy1.11 NOD Tregs were minimally detected 1–4 weekspost-infusion in blood (Fig. 3B and C and SupplementaryTable 1, aCD31Treg), SP, LN, pLN, and pancreas (Fig. 3D).The addition of IL-2 support with a short course of IACtreatment after aCD3 enhanced engraftment during

220 Regulatory T-cell Immunotherapy Diabetes Volume 69, February 2020

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

treatment (aCD31Treg1IAC), but by the end of follow-up,engraftment was no better than aCD3 alone. However,a single dose of CyP after aCD3 (aCD31CyP1Treg), whichdepletes rebounding host Tregs (Fig. 1E, dashed line,shaded area), resulted in a 43-fold increase in donorTreg engraftment compared with aCD3; the additionof IAC further enhanced engraftment by several fold(aCD31CyP1Treg1IAC) (Fig. 3B and C and Supplemen-tary Table 1). The significant increase in the proportion ofhost CD41Foxp31 cells in the CD41 T-cell compartmentduring the rebound phase after aCD31CyP (Fig. 1F,shaded area 2) did not inhibit donor Treg engraftment,which was robust (Fig. 3D). Thus, key to robust engraft-ment is to create space that minimizes host Treg

competition at the time of Treg infusion. The additionof CyP and IAC resulted in the greatest increase of thetotal CD41Foxp31 compartment compared with allother regimens tested (Fig. 3F), and was accompaniedby an increased proportion of CD41Foxp31 cells in theCD41 T-cell compartment (Fig. 3G). While the total Tregcompartment returned to pretreatment levels at the endof the 39-day follow-up (Fig. 3F and G), donor Tregscomprised 8–25% of the Treg compartment in mice givenaCD3 and CyP with or without IAC (Fig. 3B and D).Importantly, congenic Thy1.11 donor Tregs maintaineda phenotype similar to host Tregs in all treatments withcomparable Foxp3/CD25 expression in treated mice(Supplementary Fig. 6A).

Figure 4—Diabetes remission after immunomodulation and infusion of polyclonal Tregs in recently diabetic NOD mice. A: Experimentalscheme. B: Percentage of diabetic mice after aCD3, CyP, and/or IAC immunomodulation and adoptive transfer of polyclonal Tregs isolatedfrom Thy1.1 congenic NODmice into recently diabetic NODmice.C: Percentage of diabetic mice after aCD3, CyP, and/or IAC. Diabetic micewere initially given a single insulin pellet at the start of aCD3 treatment to control diabetes. Urine and blood glucose levels were monitored for6 months after the start of aCD3 treatment. Median values are indicated to the right of the legends. d, day.

diabetes.diabetesjournals.org Cabello-Kindelan and Associates 221

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

We then treated late, prediabetic NOD mice aged 16–22 weeks with our various immunomodulation strategiesfollowed by infusion of congenic, autologous Thy1.11

NOD Tregs and examined Treg engraftment. Similar tothe young, prediabetic NODmice, substantial engraftmentoccurred in mice that received aCD3 and CyP with orwithout IAC in the peripheral blood and SP, LN, pLN, andpancreas (Supplementary Fig. 7A, B, and E). Again, optimalTreg engraftment was observed in the aCD31CyP1Treg1IAC group, with a 53-fold increase in engraftmentcompared with aCD3 alone (day 18 after the start ofaCD3), and led to the greatest increase in the totalCD41Foxp31 population (Supplementary Fig. 7A–C).

Similarly, we observed a decline in engraftment, but donorTregs comprised 25–30% of the Treg compartment at theend of follow-up in mice given aCD3/CyP (SupplementaryFig. 7A). Collectively, our results demonstrate a new ther-apeutic synergy between aCD3 and CyP in promotingengraftment of polyclonal, autologous Tregs.

A Combinational Regimen of CD3, CyP, and Polyclonal,Autologous Donor Tregs Fails to Induce DiabetesRemission in Female NOD MiceWe then examined whether diabetes remission could beinduced by combinational regimens themselves or in com-bination with infusion of polyclonal, autologous Tregs (Fig.

Figure 5—Donor Treg engraftment after immunomodulation in recent diabetic NOD mice. A: Percentage of Thy1.11 donor NOD Treg inCD41Foxp31 in peripheral blood (PB). B: 4–8 weeks in the SP, LN, pLN, and pancreas (Pan). C: Foxp3 and Thy1.1 staining in gated CD4T cells day 22–39 after Treg infusion in recent-onset diabetic NOD mice receiving aCD31CyP1Treg1IAC. D: Percent positive Thy1.21 totalT cells, CD41Foxp32 T cells, or CD8 T cells in the peripheral blood of recent-onset NODmice receiving aCD3 treatment. *P, 0.0001; @P,0.01, one-way ANOVA followed by Dunnett multiple comparison test compared with day 0. E: Percentage of CD41Foxp31 in gated CD451

population. F: Percentage of CD41Foxp31 in the gated CD4 T cells with aCD3 (solid) or aCD31CyP (dashed). Diabetic NOD mice weremaintainedwith insulin pellets to control blood glucose. n5 2–6mice per group. In E, the shaded area indicates the decrease in CD41Foxp31

host Tregs after single CyP injection after aCD3 treatment. In F, shaded area 1 indicates an increase in CD41Foxp31 cells among total in thegated CD4 T cells after CD3 alone, and shaded area 2 indicates an increase in CD41Foxp31 cells in the gated CD4 T cells after CD31CyP. d,day.

222 Regulatory T-cell Immunotherapy Diabetes Volume 69, February 2020

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

4A). Although our combinatorial regimen was very effec-tive at promoting engraftment in prediabetic NOD mice,long-term diabetes remission was not achieved in recentlydiagnosed NOD mice treated with immunomodulationalone or in combination with polyclonal Tregs (Fig. 4Band C). We found that in mice treated with theaCD31CyP1polyclonal Treg6IAC regimen, the time todisease relapse significantly correlated with the levels ofdonor Tregs in blood (r 5 0.72223, P 5 0.028). Therefore,we examined Treg engraftment in blood and tissues; donor,

congenic Thy1.11NOD Tregs were readily detected in bloodand SP, LN, pLN, and pancreas 4 weeks post-infusion(aCD31CyP1Treg1IAC) (Fig. 5A and B and Supplemen-tary Table 2), and donor Tregs had a similar level of Foxp3/CD25 expression as host Tregs in blood and tissues (Sup-plementary Fig. 6B). However, there was a decrease (2.0–2.6-fold) in the level of Treg engraftment observed inrecent-onset mice compared with late prediabetic NODmice at the end of follow-up (Supplementary Fig. 7E andF), but overall engraftment with aCD3/CyP treatment was

Figure 6—Diabetes remission after immunomodulation and infusion of antigen-specific Tregs in recently diabetic NODmice. A: Experimentalscheme. B: Percentage of diabetic mice after aCD3, CyP, and/or IAC immunomodulation and adoptive transfer of antigen-specific Tregsisolated from BDC2.5 TCR NOD mice into recently diabetic NOD mice. Diabetic mice were initially given a single insulin pellet at the start ofaCD3 treatment to control diabetes. Urine and blood glucose levels were monitored for 6 months after the start of aCD3 treatment. Medianvalues are indicated to the right of the legends. C: Percentage of skin graft survival of NOD and B6 skin placed 6 months after the start ofaCD3 and BDC2.5 Treg infusion. Mice were monitored for skin rejection and diabetes development for 70 days. Inset shows blood glucosevalues for the transplanted mice. Survival curves were subjected to Mantel-Cox log-rank test. The P value is indicated on the graph. d, day;UD, undefined.

diabetes.diabetesjournals.org Cabello-Kindelan and Associates 223

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

still robust compared with prediabetic mice treated withaCD3 alone (Fig. 3A–C and Supplementary Fig. 7A and B).Several studies showed that inflammatory environmentspromote unstable Tregs, termed ex-Tregs, which becomepathogenic (32–34). However, the observed lower engraft-ment was not explained by a loss of Foxp3 expression indonor Treg populations (CD41Foxp31Thy1.11 cells), asthere were very few Thy1.11Foxp32 cells in blood, SP, LN,pLN, or pancreas (Fig. 5C). Lower engraftment was not dueto a failure of aCD3 to effectively deplete T cells, as CD4,CD8, and CD41Foxp31 cells were decreased post-aCD3

(Fig. 5D and E). Engraftment was not impacted byrebounding host Tregs, as CyP was very effective at de-pleting rebounding cells post-aCD3 (Fig. 5E, shaded area,and Fig. 5F, shaded area 1). However, unlike in prediabeticmice in which host Tregs recovered after CyP (Fig. 1F,shaded area 2), such a recovery was not observed inrecent-onset mice (Fig. 5F, shaded area 2). Together, theseresults suggest that in recent-onset mice, despite initialinsulin treatment, the host may not fully support the Tregcompartment, which could represent an obstacle to suc-cessful Treg therapy with polyclonal Tregs.

Figure 7—Antigen-specific donor Treg engraftment after immunomodulation in recent diabetic NOD mice. A: BDC2.5 tetramer staining ofCD451 cells from isolated pancreatic islets of recent-onset NODmice subjected to immunomodulation and adoptive transfer of BDC2.5 Tregsof gated CD41Foxp31 andCD41Foxp32 cells 7 days after Treg infusion (day 18).B: Representative staining of CD25 and BDC2.5 tetramer ongated CD41Foxp31 cells in the pancreatic islets 7 days after Treg infusion after aCD3 treatment. C: Percentage of BDC2.5 tetramer1 cellsamong the CD41Foxp31 cells from pancreatic islets 18, 40, and 90 days after aCD3 or CyP treatment or in mice left untreated.D: Correlationanalysis between the level of BDC2.5 Treg engraftment among the gated CD41Foxp31 cells and percentage of total CD41Foxp31 in thegated CD451 population within islets on day 18 after the start of aCD3 treatment. Percentage of BDC2.5 tetramer1 cells among theCD41Foxp31 cells (E) or CD41Foxp32 cells (F ) in the pancreatic islets, pLN, or other LN 18, 40, and 90 days post–Treg infusion after aCD3treatment. P values are indicated in C and E following unpaired Student t test. d, day.

224 Regulatory T-cell Immunotherapy Diabetes Volume 69, February 2020

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

Combinational Therapies That IncludeAntigen-SpecificTregs Lead to Durable Remission in Newly DiagnosedDiabetic NOD MiceSelection of therapeutic Tregs by antigen could potentiallyovercome the less supportive environment observed inrecent-onset mice. Moreover, antigen-specific Tregs areknown to be more effective than polyclonal Tregs at reg-ulating diabetes and responses to alloantigen (35–37).Therefore, we tested whether the infusion of islet-specificTregs together with aCD3 and/or CyP immunomodulationcould lead to durable diabetes remission in recently di-abetic NOD mice (Fig. 6A). In sharp contrast to polyclonalTregs, infusion of islet-specific (chromogranin A) Tregs iso-lated from BDC2.5 TCR transgenic NOD mice after immu-nomodulation induced durable diabetes remission (Fig.6B). This remission was observed after immunomodula-tion with aCD3 alone or in combination with CyP with orwithout a short course of IAC. All the NOD mice receivingan islet-specific Treg infusion after aCD3 alone went intodurable remission (6 months). aCD3 prior to BDC2.5 Treginfusion is critical for achieving durable remission as in-fusion of BDC2.5 Tregs after CyP injection resulted in only38% of mice experiencing long-term remission, and in-fusion of equal numbers of BDC2.5 Tregs without any priorimmunomodulation failed to reverse diabetes (Fig. 6B).These data support that manipulation of the host immunesystem is necessary to achieve a therapeutic benefit withadoptive Treg therapy. Antigen specificity of infused Tregs iscritical, since equal numbers of polyclonal Tregs failed toachieve durable remission in most mice (Fig. 4B). Impor-tantly, the lack of diabetes relapse with aCD31BDC2.5Treg therapy was not due to loss of effective immuneresponses. Fully allogeneic B6 skin grafts placed 6 monthsafter the start of therapy on mice given aCD31BDC2.5Treg therapy were all rejected, whereas syngeneic NODgrafts were maintained (Fig. 6C), and mice continued to bediabetes free at end of the 70-day follow-up (Fig. 6C, inset).Therefore, the destructive autoimmune response had beenreset without long-term immunosuppression and thesemice had a fully competent immune system.

Combinational Regimen of aCD3 and Antigen-SpecificTregs Results in Robust, Long-Term EngraftmentWithin IsletsWe determined the level of islet-specific Treg engraftmentafter immunomodulation. BDC2.5 Treg engraftment wasassessed from isolated islets 7 days post–Treg infusion (day18 after start of aCD3) and up to 90 days post-aCD3 orCyP treatment or in unmanipulated recent-onset NODmice. We detected the highest engraftment in pancreaticislets with aCD3 treatment and persisted up to 90 dayscompared with CyP treatment; mice that did not receiveimmunomodulation had the lowest engraftment (Fig. 7Aand C). These BDC2.5 tetramer-positive cells expressedhigh levels of CD25 (Fig. 7B). Importantly, this level ofBDC2.5 Treg engraftment in the pancreas of mice 40 dayspost–aCD3 treatment was similar to the level of polyclonal

Treg engraftment in the pancreas when aCD3, CyP, andIAC were given (Fig. 5B). There was a significant correla-tion between the level of BDC2.5 Treg engraftment andpercentage of total CD41Foxp31 T cells within islets (Fig.7D). BDC2.5 Tregs were detected in pLN or nondrainingLNs, but much less compared with islets (Fig. 7E), and thiswas similar to the level of engraftment observed withpolyclonal, autologous Tregs after aCD3 treatment alone(Fig. 3D). Moreover, tetramer-positive CD41Foxp32

T cells were minimally detected, suggesting that BDC2.5Tregs are stable within these tissues (Fig. 7E and F).Collectively, our data suggest that antigen-specific Tregsmay be capable of better engraftment specifically in thepancreatic islets where their target antigen is present, andthis may explain the therapeutic efficacy observed.

DISCUSSION

Our previous work focused on understanding critical factorsrequired for the generation/homeostasis of thymic Tregsand demonstrated the importance of host environmentas a key determinant of long-term Treg engraftment andtherapeutic efficacy in therapies involving Treg infusion(18–20,22,38,39). Our data from this study demonstratehow these requirements also apply for a successful outcomeof a Treg-based immunomodulatory regimen in a preclinicalT1D model. Although therapeutic success in NOD mice doesnot guarantee a translation to patient benefit, it isnonetheless important to conduct preclinical testing,when feasible, before starting clinical experimentation.The NOD mouse presents critical similarities at geneticand immunological levels with the human diseases and,despite its limitations, is the most widely used and ac-cepted experimental model. Our findings provide therationale for the design of future clinical trials to testthe efficacy of Treg-based therapies and highlight theimportance of incorporating immunomodulation priorto Treg infusion and the key role of antigen-specific Tregsto treat islet autoimmunity.

Our experimental studies used the intact aCD3e anti-body for T-cell debulking because it is more efficient in thissetting than the F(ab9)2 form (25–27). Studies with a hu-manized CD3 NOD mouse model (NOD-huaCD3) (29)demonstrated that a single injection of intact aCD3 an-tibody (clone 2C11, 5mg) led to a higher percentage of CD4and CD8 cell apoptosis accompanied by greater depletionof CD4 T cells and a higher rebound of CD8 T cellscompared with nonmitogenic F(ab9)2 fragments (50 mg).Similar results on T-cell apoptosis, depletion, and reboundwere observed when a single dose of mitogenic (YTH12.5,2 mg) and nonmitogenic (otelixizumab, 100 mg) humanCD3e antibodies were used. In the context of our studiesexamining the in vivo environment to support persistentTreg engraftment, a 5-day course of a higher dose of intactaCD3 (50 mg) led to significant depletion of CD4 and CD8T cells, which was accompanied by slow recovery, while theTregs were depleted early after aCD3 treatment with a rapidrecovery. However, intact aCD3 treatment allowed only

diabetes.diabetesjournals.org Cabello-Kindelan and Associates 225

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

limited Treg engraftment after T-cell debulking and waslikely due to competition with rapidly rebounding hostTregs. In fact, when the recovering host Treg population wasdecreased again with CyP just before Treg infusion, robustengraftment of polyclonal Tregs occurred. This level ofdonor Treg engraftment 4 weeks post–Treg infusion inprediabetic NOD is similar to what was observed withharsher ablative conditioning in nonautoimmune C57BL/6mice (20), yet relying on short courses of clinically relevantagents. Improving Treg engraftment could have increasedfrequencies of disease-relevant Tregs and improve thera-peutic benefit, but durable remission was not achieveddespite robust polyclonal Treg engraftment. However, weobserved a strong correlation with the level of polyclonalTreg engraftment after immunomodulation and clinicaloutcome in recent-onset NOD mice, suggesting that in-creased numbers of adoptively transferred polyclonal,autologous Tregs, perhaps through ex vivo expansion, couldlead to improved benefit. This will require further inves-tigation but could be more easily implemented in a clinicalsetting given that it is more challenging to use antigen-specific Tregs. Additionally, the positive effects of aCD3 ina recent prevention clinical trial (40), and our new data onengraftment in late prediabetic mice, support the hypoth-esis that this regimen could be tested in this preclinicalmodel with polyclonal and antigen-specific Tregs.

Antigen-based therapies have been used in experimen-tal models and clinical trials with mixed results. However,antigens could be helpful to drive the selection/expansionof infused antigen-specific Tregs. Indeed, diabetes wasprevented in adoptive transfer studies with islet-specificNOD Tregs into NOD mice (NOD.Rag2/2, NOD TCRa2/2,or NOD.CD282/2) that lack either T cells or Tregs, re-spectively (35,37,41,42). Critically, these models resembleIL-2Rb2/2mice, in that there is natural space with limitedhost Treg competition. A major benefit of using antigen-specific Tregs is that these Tregs will likely act only whereantigen is present, thereby providing local immunotherapywithout impacting the remainder of the immune system.Here, we found that our combinational regimens allowengraftment of polyclonal Tregs throughout the immunecompartments and in tissues, but the presence of thetarget antigen in the pancreas led to islet-specific Tregsbeing found largely within islets after infusion. Both aCD3or aCD31CyP treatments prior to islet-specific Treg in-fusion led to durable remission; however, aCD3 alone withislet-specific Tregs was sufficient. The level of engraftmentin these mice was similar to what was observed in micetreated with aCD31CyP1polyclonal Tregs1IAC, but nowislets also include disease-relevant Tregs in the pancreaticmicroenvironment. These NOD mice were still fully capa-ble of rejecting allogeneic skin transplants without break-ing the islet autoimmune tolerance established by our Tregimmunotherapy with islet-specific Tregs. Importantly, by cre-ating this supportive environment for donor Tregs, durablediabetes remission was achieved using 20-fold less numbers ofantigen-specific Tregs in our studies compared with a previous

study that used in vitro expanded BDC2.5 Tregs in unmanip-ulated NOD mice (35). Like others (25,29), we found thatintact aCD3 antibody led to IFNg and TNFa production, butthis did not hinder Treg engraftment or the induction ofdurable diabetes remission with antigen-specific Treg infusion.

Challenges with antigen-specific Treg-based therapiesstill center on the availability of a sufficient number ofdisease-relevant Tregs and whether pools of antigen spe-cificities, and what specificities, will be needed to achievetherapeutic benefit, given that autoimmunity is likely to bemore heterogeneous in patients. The discovery of hybridinsulin peptides and other modified peptides highlightsthe importance of further investigation on the antigenspecificities needed for therapeutic gains (43,44). Althoughcurrent methodologies are limited in terms of the ability toisolate and expand a sufficient number of autologousantigen-specific Tregs, new strategies aimed at generatinglarge numbers of antigen-specific Tregs are being explored.These include lentiviral TCR gene transfer into expandedpolyclonal Tregs, Foxp3 gene editing into antigen-specificCD41 T cells, and conversion of effector T cells into Tregulatory–like cells by CRISPR/Cas9-mediated integrationof a Foxp3 transgene, and open the possibility of over-coming this limitation (45–47). Our work sets the con-ceptual framework and a key innovative approach that wasaimed at addressing a critical and unmet need in theclinics, the need for new therapeutic protocols for T1Dtreatment that will enhance the efficacy and durability ofTreg-based cell therapies.

Acknowledgments. The authors thank Oliver Umland of the DiabetesResearch Institution Flow Cytometry Core.Funding. This work was supported by the American Diabetes Association(Junior Faculty grant 07-09-JF-06 to A.L.B.), the National Institute of Allergy andInfectious Diseases (High Priority Short-term grant 1R56-AI-101278-01A1 toA.L.B.), the University of Miami Miller School of Medicine (Dean’s NationalInstitutes of Health Bridge Program Award to A.L.B.), and the Diabetes ResearchInstitute Foundation (to A.L.B.).Duality of Interest. No potential conflicts of interest relevant to this articlewere reported.Author Contributions. C.C.-K. performed experiments and providedintellectual feedback. S.M., A.S., J.R., and C.V. performed experiments. A.P.provided intellectual feedback and reviewed the manuscript. A.L.B. designed andconceived the studies, performed experiments, analyzed data, and wrote themanuscript. A.L.B. is the guarantor of this work and, as such, had full access to allthe data in the study and takes responsibility for the integrity of the data and theaccuracy of the data analysis.Prior Presentation. This study was presented at the Immunology of DiabetesSociety Congress 2018, London, U.K., 25–29 October 2018; the Keystone Symposia:Immune Regulation in Autoimmunity and Cancer, Whistler, British Columbia, Canada,26–30 March 2017; and the Keystone Symposia: Inflammatory Diseases: RecentAdvances in Basic and Translational Research and Therapeutic Treatments, Vancouver,British Columbia, Canada, 17–22 January 2014.

References1. Sakaguchi S, Sakaguchi N, Asano M, Itoh M, Toda M. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains(CD25). Breakdown of a single mechanism of self-tolerance causes variousautoimmune diseases. J Immunol 1995;155:1151–1164

226 Regulatory T-cell Immunotherapy Diabetes Volume 69, February 2020

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021

2. Shevach EM, McHugh RS, Thornton AM, Piccirillo C, Natarajan K, Margulies DH.Control of autoimmunity by regulatory T cells. Adv Exp Med Biol 2001;490:21–323. Graca L, Cobbold SP, Waldmann H. Identification of regulatory T cells intolerated allografts. J Exp Med 2002;195:1641–16464. Graca L, Thompson S, Lin CY, Adams E, Cobbold SP, Waldmann H. BothCD4(1)CD25(1) and CD4(1)CD25(-) regulatory cells mediate dominant trans-plantation tolerance. J Immunol 2002;168:5558–55655. Hara M, Kingsley CI, Niimi M, et al. IL-10 is required for regulatory T cells tomediate tolerance to alloantigens in vivo. J Immunol 2001;166:3789–37966. Kingsley CI, Karim M, Bushell AR, Wood KJ. CD251CD41 regulatory T cellsprevent graft rejection: CTLA-4- and IL-10-dependent immunoregulation of al-loresponses. J Immunol 2002;168:1080–10867. Juvet SC,Whatcott AG, Bushell AR, Wood KJ. Harnessing regulatory T cells for clinicaluse in transplantation: the end of the beginning. Am J Transplant 2014;14:750–7638. Marek-Trzonkowska N, Myśliwec M, Siebert J, Trzonkowski P. Clinical appli-cation of regulatory T cells in type 1 diabetes. Pediatr Diabetes 2013;14:322–3329. Marek-Trzonkowska N, Mysliwiec M, Dobyszuk A, et al. Administration ofCD41CD25highCD127- regulatory T cells preserves b-cell function in type1 diabetes in children. Diabetes Care 2012;35:1817–182010. Marek-Trzonkowska N, Mysliwiec M, Dobyszuk A, et al. Repetitive ad-ministration of CD41CD25highCD1272 T regulatory cells prolongs survival ofpancreatic islets in type 1 diabetes in children. Oral session (IT-OR01, HumanImmunology in Clinical Trials) at the 73rd Scientific Sessions of the AmericanDiabetes Association, 21–25 June 2013, at the McCormick Place ConventionCenter, Chicago, IL11. Brunstein CG, Miller JS, Cao Q, et al. Infusion of ex vivo expanded T regulatorycells in adults transplanted with umbilical cord blood: safety profile and detectionkinetics. Blood 2011;117:1061–107012. Earle KE, Tang Q, Zhou X, et al. In vitro expanded human CD41CD251

regulatory T cells suppress effector T cell proliferation. Clin Immunol 2005;115:3–913. Hippen KL, Merkel SC, Schirm DK, et al. Massive ex vivo expansion of humannatural regulatory T cells (T(regs)) with minimal loss of in vivo functional activity.Sci Transl Med 2011;3:83ra4114. Putnam AL, Brusko TM, Lee MR, et al. Expansion of human regulatory T-cellsfrom patients with type 1 diabetes. Diabetes 2009;58:652–66215. Trzonkowski P, Szary�nska M, Myśliwska J, Myśliwski A. Ex vivo expansion ofCD4(1)CD25(1) T regulatory cells for immunosuppressive therapy. Cytometry A2009;75:175–18816. Trzonkowski P, Bieniaszewska M, Juści�nska J, et al. First-in-man clinical resultsof the treatment of patients with graft versus host disease with human ex vivoexpanded CD41CD251CD127- T regulatory cells. Clin Immunol 2009;133:22–2617. Marek-Trzonkowska N, Myśliwiec M, Dobyszuk A, et al. Therapy of type1 diabetes with CD4(1)CD25(high)CD127- regulatory T cells prolongs survival ofpancreatic islets - results of one year follow-up. Clin Immunol 2014;153:23–3018. Adeegbe D, Bayer AL, Levy RB, Malek TR. Cutting edge: allogeneicCD41CD251Foxp31 T regulatory cells suppress autoimmunity while establishingtransplantation tolerance. J Immunol 2006;176:7149–715319. Bayer AL, Yu A, Malek TR. Function of the IL-2R for thymic and peripheralCD41CD251 Foxp31 T regulatory cells. J Immunol 2007;178:4062–407120. Cabello-Kindelan C, de la Barrera A, Malek TR, Bayer AL. In vivo environmentnecessary to support transplanted donor mouse T regulatory cells. Am JTransplant 2014;14:1032–104521. Cabello-Kindelan C, Mackey S, Bayer AL. Adoptive T regulatory cell therapyfor tolerance induction. Curr Transplant Rep 2015;2:191–20122. Bayer AL, Yu A, Adeegbe D, Malek TR. Essential role for interleukin-2 forCD4(1)CD25(1) T regulatory cell development during the neonatal period. J ExpMed 2005;201:769–77723. Billingham RE, Medawar PB. The technique of free skin grafting in mammals.J Exp Biol 1951;28:385–40224. Berney T, Molano RD, Cattan P, et al. Endotoxin-mediated delayed islet graftfunction is associated with increased intra-islet cytokine production and islet cellapoptosis. Transplantation 2001;71:125–132

25. Chatenoud L, Primo J, Bach JF. CD3 antibody-induced dominant self tol-erance in overtly diabetic NOD mice. J Immunol 1997;158:2947–295426. Herold KC, Bluestone JA, Montag AG, et al. Prevention of autoimmune diabeteswith nonactivating anti-CD3 monoclonal antibody. Diabetes 1992;41:385–39127. Yang W, Hussain S, Mi QS, Santamaria P, Delovitch TL. Perturbed ho-meostasis of peripheral T cells elicits decreased susceptibility to anti-CD3-inducedapoptosis in prediabetic nonobese diabetic mice. J Immunol 2004;173:4407–441628. Penaranda C, Tang Q, Bluestone JA. Anti-CD3 therapy promotes tolerance byselectively depleting pathogenic cells while preserving regulatory T cells. J Im-munol 2011;187:2015–202229. Kuhn C, You S, Valette F, et al. Human CD3 transgenic mice: preclinicaltesting of antibodies promoting immune tolerance. Sci Transl Med 2011;3:68ra1030. Brode S, Raine T, Zaccone P, Cooke A. Cyclophosphamide-induced type-1diabetes in the NOD mouse is associated with a reduction of CD41CD251Foxp31 regulatory T cells. J Immunol 2006;177:6603–661231. Lutsiak ME, Semnani RT, De Pascalis R, Kashmiri SV, Schlom J, Sabzevari H.Inhibition of CD4(1)251 T regulatory cell function implicated in enhanced im-mune response by low-dose cyclophosphamide. Blood 2005;105:2862–286832. Bailey-Bucktrout SL, Martinez-Llordella M, Zhou X, et al. Self-antigen-drivenactivation induces instability of regulatory T cells during an inflammatory auto-immune response. Immunity 2013;39:949–96233. Komatsu N, Okamoto K, Sawa S, et al. Pathogenic conversion of Foxp31

T cells into TH17 cells in autoimmune arthritis. Nat Med 2014;20:62–6834. Zhou X, Bailey-Bucktrout SL, Jeker LT, et al. Instability of the transcriptionfactor Foxp3 leads to the generation of pathogenic memory T cells in vivo. NatImmunol 2009;10:1000–100735. Tang Q, Henriksen KJ, Bi M, et al. In vitro-expanded antigen-specific regulatoryT cells suppress autoimmune diabetes. J Exp Med 2004;199:1455–146536. Tang Q, Henriksen KJ, Boden EK, et al. Cutting edge: CD28 controls pe-ripheral homeostasis of CD41CD251 regulatory T cells. J Immunol 2003;171:3348–335237. Yamazaki S, Inaba K, Tarbell KV, Steinman RM. Dendritic cells expandantigen-specific Foxp31 CD251 CD41 regulatory T cells including suppressors ofalloreactivity. Immunol Rev 2006;212:314–32938. Bayer AL, Chirinos J, Cabello C, et al. Expansion of a restricted residual host Treg-cell repertoire is dependent on IL-2 following experimental autologous he-matopoietic stem transplantation. Eur J Immunol 2011;41:3467–347839. Bayer AL, Jones M, Chirinos J, et al. Host CD41CD251 T cells can expandand comprise a major component of the Treg compartment after experimental HCT.Blood 2009;113:733–74340. Herold KC, Bundy BN, Long SA, et al.; Type 1 Diabetes TrialNet Study Group.An anti-CD3 antibody, teplizumab, in relatives at risk for type 1 diabetes. N Engl JMed 2019;381:603–61341. Masteller EL, Warner MR, Tang Q, Tarbell KV, McDevitt H, Bluestone JA.Expansion of functional endogenous antigen-specific CD41CD251 regulatoryT cells from nonobese diabetic mice. J Immunol 2005;175:3053–305942. Tarbell KV, Yamazaki S, Olson K, Toy P, Steinman RM. CD251 CD41 T cells,expanded with dendritic cells presenting a single autoantigenic peptide, suppressautoimmune diabetes. J Exp Med 2004;199:1467–147743. Delong T, Wiles TA, Baker RL, et al. Pathogenic CD4 T cells in type 1 diabetesrecognize epitopes formed by peptide fusion. Science 2016;351:711–71444. Wiles TA, Powell R, Michel CR, et al. Identification of hybrid insulin peptides (HIPs)in mouse and human islets by mass spectrometry. J Proteome Res 2019;18:814–82545. Brusko TM, Koya RC, Zhu S, et al. Human antigen-specific regulatory T cellsgenerated by T cell receptor gene transfer. PLoS One 2010;5:e1172646. Honaker Y, Xiang Y, Fisher L, et al. Conversion of T-effector cells to im-munosuppressive T-regulatory-like cells by CRISPR/Cas9-mediated integration ofa FOXP3 transgene. Blood 2018;132:349047. Xiang Y, Sommer K, Honaker Y, et al. Gene-editing of FOXP3 in antigen-specific CD41 T cells for restoration of immune tolerance in type 1 diabetes.Diabetes 2019;68(Suppl. 1):1740

diabetes.diabetesjournals.org Cabello-Kindelan and Associates 227

Dow

nloaded from http://diabetesjournals.org/diabetes/article-pdf/69/2/215/431504/db190061.pdf by guest on 29 D

ecember 2021