original article Chao-Lien Liu 1, Peiqing Ye 1, Benjamin C Yen 1 and Carol H Miao 1,2

Transcription

1 The Amerian Soiety of Gene & Cell Therapy original artile In Vivo Expansion of Regulatory T ells With IL-2/IL-2 mab Complexes Prevents Anti-fator VIII Immune Responses in Hemophilia A Mie Treated With Fator VIII Plasmid-mediated Gene Therapy Chao-Lien Liu 1, Peiqing Ye 1, Benjamin C Yen 1 and Carol H Miao 1,2 1 Center for Immunity and Immunotherapies, Seattle Children s Researh Institute, Seattle, Washington, USA; 2 Department of Pediatris, University of Washington, Seattle, Washington, USA Generation of transgene-speifi immune responses an onstitute a major ompliation following gene therapy treatment. An in vivo approah to induing seletive expansion of Regulatory T (Treg) ells by injeting interleukin-2 (IL-2) mixed with a speifi IL-2 monolonal antibody (JES6-1) was adopted to modulate anti-fator VIII (anti-fviii) immune responses. Three onseutive IL-2 omplexes treatments ombined with injetion prevented anti-fviii formation and ahieved persistent, therapeuti-level of FVIII expression in hemophilia A (HemA) mie. The IL-2 omplexes treatment expanded CD4 + CD + Foxp3 + Treg ells five- to sevenfold on peak day, and they gradually returned to normal levels within 7 14 days without hanging other lymphoyte populations. The transiently expanded Treg ells are highly ativated and display suppressive funtion in vitro. Adoptive transfer of the expanded Treg ells proteted reipient mie from generation of high-titer antibodies following hallenge. Repeated plasmid transfer is appliable in tolerized mie without eliiting immune responses. Mie treated with IL-2 omplexes mounted immune responses against both T-dependent and T-independent neoantigens, indiating that IL-2 omplexes did not hamper the immune system for long. These results demonstrate the important role of Treg ells in suppressing anti-fviii immune responses and the potential of developing Treg ell expansion therapies that indue long-term tolerane to FVIII. Reeived 23 Deember 21; aepted 5 Marh 211; published online 5 April 211. doi:1.138/mt INTRODUCTION Hemophilia A (HemA) is a ongenital bleeding disorder aused by a defiieny of the blood lotting protein Fator VIII (FVIII); 1 in its severe form, hemophilia is a life-threatening, rippling hemorrhagi disease. Currently, hemophilia patients are treated with protein-replaement therapy. However, a major ongoing problem is a very high frequeny in the formation of inhibitory antibodies against FVIII following treatment. 2 Reent studies in animal models suggest that ontinuous expression of FVIII following gene therapy would provide signifiant therapeuti benefits. Nevertheless, both viral and nonviral gene transfer of FVIII 3 6 often result in transient gene expression in the absene of immunosuppression. Only in a few ases, sustained gene expression of FVIII was ahieved in animals treated with gene therapy. 7,8 It has been indiated in reent reviews 9 11 that immune responses against transgeni produts or vetors an beome major obstales to suessful appliation of gene therapy. Therefore, it is essential to develop safe and effetive methods to modulate immune responses against FVIII to ensure the suess of hemophilia treatment. Regulatory T (Treg) ells omprise a ruial T-ell ompartment that maintains peripheral tolerane by suppressing autoreative T-ell responses 12 and orhestrating a balaned immune response to foreign antigens It has an important role in the prevention of autoimmune 12,16 and inflammatory 17 diseases, regulation of immunity to viral and parasite infetions, 18,19 maintenane of maternal tolerane to fetus 2 and inhibition of antitumor immunity. 21 These findings have engendered enthusiasm for developing strategies utilizing Treg ells to regulate immune responses in linially important settings inluding transplantation, autoimmunity, allergy, and aner. Furthermore, many suessful protools for modulating FVIII-speifi immune responses involve inreases in either or both of the perentages and total number of CD4 + Foxp3 + Treg ells in gene therapy settings. 22 Reently, it was demonstrated that adoptively transferred Treg ells isolated from transgeni HemA/Foxp3 mie signifiantly redued anti-fviii inhibitory antibody titers following -mediated gene transfer into HemA mie 23, indiating that Treg ells are important regulators for anti-fviii immune responses. Interleukin-2 (IL-2) is an important growth fator that drives T ells to proliferate and differentiate into funtional effetor ells. IL-2 is also essential in the development of Treg ells; in the absene of IL-2, Treg ells annot survive or expand in the thymus or in the periphery. 24, Reent studies demonstrated that IL-2 bound to a partiular anti-il-2 monolonal antibody (mab; JES6-1A12) Correspondene: Carol H Miao, Center for Immunity and Immunotherapies, Seattle Children s Researh Institute and University of Washington, 19 Ninth Avenue, C9S-7, Seattle, WA 9811, USA. miao@u.washington.edu Moleular Therapy vol. 19 no. 8, aug

2 IL-2 Complexes Modulate Anti-FVIII Responses The Amerian Soiety of Gene & Cell Therapy an seletively and signifiantly expand CD4 + CD + Treg ells. 26 Treatment with these partiular IL-2/anti-IL-2 mab omplexes an protet mie against experimental autoimmune enephalomyelitis and panreati islet allografts, 27 and the Ab-mediated disease, experimental autoimmune myasthenia gravis. 28 In this study, we investigated whether treatment with immune omplexes onsisting of IL-2 and anti-il-2 mab (JES6-1A12) (referred to as IL-2 omplexes hereafter) to expand Treg ells an modulate anti-fviii immune responses following gene therapy. Consistent with earlier reports in other model systems, 27,29 31 we found that treatment with IL-2 omplexes indued a five- to sevenfold expansion of Treg ells in the peripheral blood, lymph nodes (LNs) and spleen of the treated mie. Most signifiantly, this expansion of Treg ells prevented the formation of inhibitory antibodies against FVIII following plasmid-mediated gene therapy in HemA mie. RESULTS IL-2 omplexes indued tolerane to FVIII after -mediated gene therapy Seletive enrihment of Treg ells by the injetion of IL-2 omplexes (IL-2/ JES6-1mAb) has the potential to modulate transgenespeifi immune responses following -mediated gene therapy. We treated HemA mie with IL-2 omplexes at two different shedules: three daily intraperitoneal injetions of IL-2 omplexes at days 5, 4, and 3 in shedule 1 mie (n = 12/ group) and at days 1,, and 1 in shedule 2 mie (n = 9/group), along with hydrodynami injetion of 5 μg at day. Plasma samples were olleted from treated mie at sheduled time points and FVIII ativities and inhibitory antibody titers were assessed. Groups of naive mie and mie treated with IL-2 omplexes only, and were inluded as ontrols. In a ontrol experiment, injetion of alone produed short-term (1 2 weeks) high levels of FVIII ativity in HemA mie, followed by a gradual derease to undetetable levels in 2 4 weeks due to the development of anti-fviii inhibitory antibodies (Figure 1). In ontrast, immune-modulation with IL-2 omplexes suessfully prevented anti-fviii immune responses. Eleven out of the 12 shedule 1 mie produed persistent therapeuti levels of FVIII ativities (1 1% of FVIII levels in normal human plasma) for 17 weeks (Figure 1a) and none of the treated mie developed anti-fviii inhibitory antibodies (Figure 1b). Of the nine mie treated with IL-2 omplexes using shedule 2, three mie produed persistent therapeuti FVIII levels without the generation of anti-fviii antibodies (Figure 1). For the remaining six mie, FVIII ativity persisted at therapeuti levels for 4 weeks before dropping to undetetable levels (Figure 1). Among these, three mie did not develop anti- FVIII inhibitory antibodies, whereas the other three developed antibodies starting at the 4th week post-treatment (Figure 1d). a FVIII ativities (% of normal) 1, , b Anti-FVIII antibodies (Bethesda units) d E94 E95 E96 E97 E98 E19 E191 E193 E626 E627 E628 E629 (n = 4) (n = 5) FVIII ativities (% of normal) Anti-FVIII antibodies (Bethesda units) E115 E116 E117 E118 E66 E661 E662 E663 E664 (n = 4) (n = 5) Figure 1 Long-term fator VIII (FVIII) expression in hemophilia A mie after -mediated gene therapy and immunomodulation with interleukin-2 (IL-2) omplexes. Two groups of hemophilia A mie were treated separately with three daily intraperitoneal (i.p.) injetions of IL-2 omplexes at days 5, 4, and 3 followed by intravenous (i.v.) injetion of 5 μg of at day (n = 12, shedule 1), or with three daily i.p. injetions of IL-2 omplexes at days 1,, and 1 onomitant by i.v. injetion of 5 μg of at day (n = 9, shedule 2). FVIII ativities were assessed by a modified ativated partial thromboplastin time assay and the anti-fviii antibody titers by Bethesda assay over time. For shedule 1 mie, (a) FVIII ativity, (b) anti-fviii antibody titers. For shedule 2 mie, () FVIII ativity, (d) anti-fviii antibody titers. Eah symbol represents data obtained from an individual mouse treated with and IL-2 omplexes using either shedule 1 or 2. Mie treated with (n = 4) and (n = 5) were used as ontrols vol. 19 no. 8 aug. 211

3 The Amerian Soiety of Gene & Cell Therapy IL-2 Complexes Modulate Anti-FVIII Responses These data indiate that shedule 1 treatment is highly effetive at preventing the formation of anti-fviii antibodies, whereas shedule 2 treatment is only partially effetive. IL-2 omplexes treatment seletively inreased the perentages and ativities of CD4 + CD + Foxp3 + Treg ells in peripheral blood, LNs and Spleen Next, we investigated if hanges in ell populations of the T-ell ompartments of HemA mie espeially Treg ells orrelated with tolerane indution by IL-2 omplexes treatment. We analyzed peripheral blood in seleted mie treated with IL-2 omplexes + (shedule 1 mie, n = 4). Groups of naive mie and mie treated with and were used as ontrols. A detailed flow ytometry analysis showed that the perentage of CD4 + T ells did not hange over time (Figure 2a; P <.5); however, the perentage of CD4 + CD + Foxp3 + T ells within the CD4 + T ell ompartment transiently and signifiantly inreased in the treated mie ompared to the naive mie (Figure 2b). The expansion of Treg ells after three injetions of IL-2 omplexes was rapid but brief, reahing a peak on day 1 (peak day; Figure 2b, left panel), and then delining gradually to bakground levels by day 21 (Figure 2b, right panel). In addition, we found that the IL-2 omplexes expanded Treg ells at 1 day after the injetion showed onsiderable higher expression levels of moleules ruial for the suppressive funtion of Treg ells, inluding CD, gluoortioid-indued tumor nerosis fator reeptor (GITR), and ytotoxi T-lymphoyte antigen 4 (CTLA-4) (Figure 2); in ontrast, injetion of alone had little or no effet on Treg ell phenotype. This highly ativated state of in vivo expanded Treg ells was brief, with the majority of these moleules returning to near-normal levels of expression by day 7. Similar results were obtained in shedule 2 mie treated with FVIII plasmid + IL-2 omplexes inluding both tolerized and untolerized mie (Supplementary Figure S1). On average, % (fivefold inrease) of CD4 + CD + Foxp3 + T ells in the CD4 + T ell ompartment was deteted on peak day in shedule 1 mie treated with IL-2 omplexes + (Figure 2b), whereas 2% (fourfold inrease) was found in shedule 2 mie (Supplementary Figure S1b), ompared to untreated mie. The substantial inrease in Treg ells following injetion of IL-2 omplexes was widespread; it ourred not only in blood, but also appeared as a fourfold inrease in LNs (Figure 3b, left panel), fivefold in spleen (Figure 3e, left panel), and threefold in thymus (data not shown) on day 1 post plasmid injetion (peak day), a b Foxp3 SSC 1, CD4 CD Peak day % of CD4 T ells % of CD4 CD Foxp3 T ells Day 1 Day 3 Day 7 Day 28 Untreated MFI 5, Untreated CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 Figure 2 Effet of interleukin-2 (IL-2) omplexes treatment on CD4 + T ells and CD4 + CD + Foxp3 + regulatory T (Treg) ells in periphery blood using shedule 1 treatment. Blood ells were isolated from untreated mie and mie treated with, fator VIII (FVIII) plasmid only and + IL-2 omplexes (n = 4/group, shedule 1) at serial time points. Cells were stained and analyzed for CD4 + and CD4 + CD + Foxp3 + T ells by flow ytometry. For CD4 + T ells, (a) left panel, representative dot plot from naive and mie treated with + IL-2 omplexes; right panel, summary plot over time. For CD4 + CD + Foxp3 + T ells gated on CD4 + T ells, (b) left panel, representative dot plot from mie treated with + IL-2 omplexes; right panel, summary plot over time, and () the expression levels of their ativation markers (shown by median fluoresene intensity (MFI) values): CD, CD62L, gluoortioid-indued tumor nerosis fator reeptor (GITR) and ytotoxi T-lymphoyte antigen 4 (CTLA-4) at days 1, 3, 7, and 28 after transfer. Data shown are MFI value of the four ativation markers of eah group. Moleular Therapy vol. 19 no. 8 aug

4 IL-2 Complexes Modulate Anti-FVIII Responses The Amerian Soiety of Gene & Cell Therapy a % of CD4 T ells , b Foxp3 CD Peak day % of CD4 CD Foxp3 T ells Day 1 Day 3 Day 7 Day 28 MFI 2, CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 d % of CD4 T ells f , e Foxp3 CD Peak day % of CD4 CD Foxp3 T ells Day 1 Day 3 Day 7 Day 28 MFI 4, CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 CD CD62L GITR CTLA4 Figure 3 Effet of interleukin-2 (IL-2) omplexes treatment on CD4 + T ells and CD4 + CD + Foxp3 + regulatory T (Treg) ells in lymph nodes (LNs) and spleen of mie using shedule 1 treatment. LNs and spleen ells were isolated from naive mie and mie treated with IL-2 omplexes only, fator VIII (FVIII) plasmid only and IL-2 omplexes + (n = 2/group, eah time point) at serial time points. Cells were stained and analyzed for CD4 + and CD4 + CD + Foxp3 + T ells by flow ytometry. For (a ) LNs and (d f) spleen ells: (a,d) Summary plot for CD4 + T ells over time. (b,e) left panel, representative dot plot from naive and mie treated with + IL-2 omplexes; right panel, summary plot for CD4 + CD + Foxp3 + T ells (gated on CD4 + T ells) over time, and (,f) the expression levels of ativation markers of CD4 + CD + Foxp3 + T ells (shown by median fluoresene intensity (MFI) values): CD, CD62L, gluoortioid-indued tumor nerosis fator reeptor (GITR) and ytotoxi T-lymphoyte antigen 4 (CTLA-4) at day 1, 3, 7, and 28 after transfer. respetively; these levels delined rapidly to normal by day14 (Figure 3b,e, right panels). In ontrast, no signifiant hange in the CD4 + T-ell population was observed in both LNs (Figure 3a) and spleen (Figure 3d). The expression levels of the ativation markers of the Treg ells inluding CD, GITR and CTLA-4 after IL-2 omplexes treatment also reahed their highest levels on the peak day in both LNs (Figure 3) and spleen (Figure 3f), whereas neither Treg ell numbers nor expression of ativation markers inreased in mie treated only with. Similarly to what was observed in peripheral blood, the highly expanded and ativated Treg ells were only transiently present; the majority of the ativation markers returned to near-normal levels of expression in both LNs (Figure 3) and spleen (Figur 3f) by day 7 following injetion. In addition, treatment with isotype ontrol antibodies + produed the same results as the plasmid only treatment in mie (data not shown). Furthermore, we tested if the speifi IL-2 omplexes indued only seletive expansion of Treg ells without affeting the other ell populations in our HemA inhibitor mouse model. The perentage and total ell numbers were examined for several ell populations. As shown in Figure 4, there were no signifiant hanges in the B ells (B22 + ells) (Figure 4a,d; P <.5), myeloid ells (CD11b + ells) (Figure 4b,e; P <.5) and CD8 + T ells (Figure 4,f) in both LNs (Figure 4a ; P <.5) and spleen (Figure 4d f; P <.5) over 35 days following IL-2 omplexes treatment. Sine transforming growth fator- 1 (TGF- 1) is ritial for Treg development, we also investigated the TGF- 1 levels in the plasma of naive mie and mie treated with + IL-2 omplexes. Three + IL-2 omplexes tolerized mie (n = 3/group, Table 1) had inreased TGF- 1 levels at day 1, 3, and 7 after the injetion, ompared to those vol. 19 no. 8 aug. 211

5 The Amerian Soiety of Gene & Cell Therapy IL-2 Complexes Modulate Anti-FVIII Responses a 1 b 5 % of B22 ells % of CD11b ells d 1 e 5 f % of CD8a T ells % of B22 ells 75 5 % of CD11b ells % of CD8a T ells IL-2 omplexes + Figure 4 Influene of interleukin-2 (IL-2) omplexes on total B ells, myeloid ells and CD8 + T ells in lymph nodes (LNs) and spleen of mie using shedule 1 treatment. (a ) LNs and (d f) spleens ells were olleted and isolated at serial time points from naive mie and mie treated with, fator VIII (FVIII) plasmid only and IL-2 omplexes + (n = 2/group, eah time point). Cells were stained and analyzed for total (a,d) B ells (B22 + ells), (b,e) myeloid ells (CD11b + ells) and (,f) CD8 + T ells (CD8a + ells). Table 1 IL-2 omplexes treatment inreases systemi TGF- 1 Systemi TGF- 1 (ng/ml) mie Mie treated with + IL-2 omplexes Day Day Day Day Abbreviations: FVIII, fator VIII; IL-2, interleukin-2; TGF- 1, transforming growth fator- 1. Systemi TGF- 1 in the plasma of mie (n = 3/group) was quantified by enzymelinked immunosorbent assay with dupliates. Day 1, 3, 7, and 14 indiate days after the injetion. Data of mean SD of eah mie group are shown. of naive mie (n = 3/group, Table 1). These data are onsistent with the results reported by Liu et al., whih showed that IL-2 omplexes treatment indued a signifiant inrease in TGF- levels in an experimental mouse model for autoimmune myasthenia gravis. 28 Funtional examination of effetor T ells and CD4 + CD + Foxp3 + Treg ells in tolerane indution by IL-2 omplexes treatment in vitro and in vivo To assess the FVIII-speifi proliferative ativity of effetor T ells post IL-2 omplexes treatment, we isolated CD4 + T ells from the spleens of four groups of mie inluding naive mie, and mie treated with, and IL-2 omplexes + 14 and 35 days after injetion. These responder ells were oultured with irradiated spleni CD4 ells serving as antigen presenting ells. When stimulated with 1 U/ml of FVIII protein whih was the optimal dose response from our previous experiment, 32 CD4 + T ells isolated from mie treated with plasmid only (with high-titer anti- FVIII inhibitory antibodies) proliferated robustly on both day 14 (Figure 5a) and day 35 (Figure 5b) after injetion. The ells isolated at day 35 had higher FVIII-speifi proliferative ativity than the ells isolated at day 14. This is as expeted, sine the mie had higher inhibitory antibody titers at day 35 than at day 14. In ontrast, CD4 + T ells isolated from mie treated with IL-2 omplexes + on day 14 (Figure 5a) and day 35 (Figure 5b) after injetion showed no FVIIIspeifi proliferation; omparable levels of nonspeifi proliferation were observed from these ells with and without FVIII stimulation. In addition, no inrease in proliferative responses to FVIII was observed from CD4 + T ells isolated from ontrol naive and treated mie. Next, we tested the suppressive funtion of Treg ells in tolerized mie treated with + IL-2 omplexes. The suppressive ativity of CD4 + CD + Treg ells isolated from tolerized mie at 3 weeks following injetion were evaluated in a FVIII-speifi suppression assay using CD4 + T ells from mie treated with as responder T ells. As antiipated, we observed signifiant FVIII-speifi suppression by CD4 + CD + Treg ells isolated from + IL-2 omplexes tolerized mie (Figure 5). Additional experiments of evaluating the suppressive funtions of expanded Treg ells by treatment with IL-2 omplexes ombined with either FVIII or nonspeifi antigens will be performed to onfirm the indution of antigen-speifi tolerane to FVIII. In addition, the in vivo funtion of expanded Treg ells were examined by adoptive transfer experiments. Treg ells were isolated from either shedule 1 or 2 mie 5 days after injetion and adoptively transferred (1 1 6 ells/mouse) Moleular Therapy vol. 19 no. 8 aug

6 IL-2 Complexes Modulate Anti-FVIII Responses The Amerian Soiety of Gene & Cell Therapy a 15, b 15, [ 3 H]thymidine inorporation (.p.m.) 1, 5, Day 14 No stimulation Day 35 FVIII stimulation [ 3 H]thymidine inorporation (.p.m.) 1, 5, No stimulation FVIII stimulation Suppression (%) Day 21 suppression IL-2 omplexes 1:2 1:4 1:8 Control Ratio of Treg to Tresp d 1 Day 21 1 Day 42 Anti-FVIII antibodies (Bethesda units) Anti-FVIII antibodies (Bethesda units) Shedule 1 Shedule 2 Figure 5 Funtional examination of CD4 + T ells and CD4 + CD + Foxp3 + regulatory T (Treg) ells from mie treated with interleukin-2 (IL-2) omplexes in vitro and in vivo. (a,b) Proliferation assay. CD4 + T ells were isolated by magneti ativated ells sorting from spleens of naive mie and mie treated with, fator VIII (FVIII) plasmid only and IL-2 omplexes + (n = 2/group, eah time point) (a) 14 and (b) 35 days after plasmid injetion CD4 + T ells were oultured with irradiated CD4 ells (antigen presenting ell) with or without the presene of FVIII at 1 U/ml for 72 hours, followed by adding 1 μci[ 3 H] thymidine for 18 hours. () FVIII suppressive assay. T-responder ells were isolated from the spleen of an inhibitor mouse treated with. CD4 + CD + T ells from pooled spleni ells of naive mie or mie at day 21 following + IL-2 omplexes treatment were used as suppressive ells. The final oulture system onsisted of CD4 + Tresp ells, irradiated CD4 + ells, and CD4 + CD + Treg ells at the indiated Treg:Tresp ratio. Data shown are mean ± SD of ounts per minute (.p.m.) of [ 3 H] thymidine inorporation in tripliate wells. (d) Adoptive transfer experiments. CD4 + CD + Treg ells were isolated from spleen of hemophilia A (HemA) mie using shedule 1 and 2 treatments 5 days post injetion, and adoptively transferred into naive HemA mie. Shedule 1 reipient mie (n = 4) and shedule 2 reipient mie (n = 4) as well as a ontrol group of naive untreated HemA mie (n = 4) were subsequently hallenged with injetion one day following adoptive transfer. Anti-FVIII antibody titers on day 21 (left panel) and day 42 (right panel) were evaluated. Eah symbol represents an individual mouse. into naive HemA mie. The two groups of reipient mie (n = 4) and a group of ontrol untreated HemA mie (n = 4) were hallenged with 5 g of 1 day after the ell transfer. On day 21 post plasmid hallenge, ontrol mie produed antihfviii antibodies, whereas both groups reeiving Treg ells from shedule 1 and shedule 2 mie generated no inhibitory antibodies (Figure 5d; left panel). Importantly, mie reeiving Treg ells from shedule 1 mie were better proteted from inhibitory antibodies formation than mie reeiving Treg ells from shedule 2 and the ontrol untreated mie on day 42 (Figure 5d; right panel). Together, these data indiate that FVIII-speifi Treg ells from tolerized mie treated with IL-2 omplexes are funtionally suppressive to anti-fviii immune responses both in vitro and in vivo. Long-term tolerane to FVIII by seond hallenge of To further investigate whether immunomodulation by IL-2 omplexes indued short-term unresponsiveness or long-term tolerane to FVIII, we hallenged the + IL-2 omplexes tolerized mie with FVIIII plasmid a seond time at 18 weeks after the 1st injetion. In this experiment, we hose 4 shedule 1 mie whih did not produe antibody responses. The 2nd hallenge with indued a short burst of high-level FVIII ativities in all four mie (Figure 6a). Three mie with persistent FVIII expression after the 1st hallenge ontinued to generate therapeuti levels of FVIII expression after the 2nd hallenge (Figure 6a). Most signifiantly, none of the mie treated with IL-2 omplexes developed detetable inhibitory antibodies against FVIII after the 2nd hallenge (Figure 6b), indiating the establishment of long-term tolerane to FVIII. Unrelated T-dependent and T-independent antigen hallenge Sine IL-2 omplexes treatment transiently inreased a large number of Treg ells, the long-term effet on the immune system was investigated. Two + IL-2 omplexes tolerized mie were inoulated with the T-dependent antigen, TNP- KLH, and another two tolerized mie with the T-independent antigen, TNP-Fioll 18 weeks after plasmid delivery. Untreated naive HemA mie (n = 2/group) were used as ontrols in both vol. 19 no. 8 aug. 211

7 The Amerian Soiety of Gene & Cell Therapy IL-2 Complexes Modulate Anti-FVIII Responses a 1st injetion b 1st injetion FVIII ativities (% of normal) 1, 1 1 2nd injetion Anti-FVIII antibodies (Bethesda units) nd injetion E626 E627 E628 E Figure 6 Maintenane of immune tolerane to fator VIII (FVIII) after a seond hallenge with. Mie treated with + interleukin-2 (IL-2) omplexes were given a seond plasmid hallenge at 18 weeks after the first plasmid injetion. Three mie whih had persistent FVIII ativity and one mouse whih had short-term FVIII ativity after first treatment were hosen for this experiment. (a) FVIII ativity, (b) anti-fviii antibody titers were examined after seond hallenge as desribed in Figure 1. Eah line represents an individual mouse. a Anti-TNP IgM (ng/ml) Anti-TNP IgM (ng/ml) 5, 4, 3, 2, 1, 12, 1, 8, 6, 4, 2, Days after immunization Days after immunization Days after immunization experiments. Peak antibody titers were observed at 2 weeks post primary immunization and 1 week post seondary TNP-KLH and TNP-Fioll immunization. The levels of speifi immunoglobulin (Ig) M (Figure 7a) and total IgG (Figure 7b) were equivalent between tolerized and untreated naive mie following hallenges with TNP-KLH. Similarly, no signifiant differenes in serum levels of anti-tnp IgM (Figure 7), and IgG3 (Figure 7d) were found between tolerized and untreated naive mie following hallenges with TNP-Fioll. Together, these data indiated that transient immunomodulation with IL-2 omplexes Anti-TNP IgG ( g/ml) Anti-TNP IgG3 ( g/ml) ,8 1, Days after immunization Figure 7 Challenge of tolerized mie with unrelated T-dependent and T-independent antigens. Tolerized mie (mie treated with interleukin-2 (IL-2) omplexes + fator VIII (FVIII) plasmid, n = 2/group) and naive mie (n = 2/group) were hallenged with the T-dependent antigen (TNP-KLH, 1 μg/mouse) and T-independent antigen (TNP-Fioll, μg/mouse) starting at week 18 after plasmid injetion, respetively. (a) Anti-TNP immunoglobulin (Ig) M and (b) anti-tnp IgG levels were assessed over time following primary (day ) and seondary (day 21) TNP-KLH immunization. () Anti-TNP IgM and (d) anti-tnp IgG3 levels were assessed over time following primary and seondary TNP-Fioll immunization. No signifiant differenes were found between the groups for anti-tnp IgM, anti-tnp IgG, and anti-tnp IgG3 levels. b d did not adversely affet immune responses to T-dependent or T-independent neoantigens in the long term. DISCUSSION Immune response against FVIII is a major obstale for suessful gene therapy for HemA. Many approahes developed toward indution of tolerane to FVIII 22 and other antigens 33 involve suppressive funtion of Treg ells. Our lab has reently demonstrated that adoptive transfer of transgeni CD4 + Foxp3 + Treg ells proteted reipient mie from high-titer anti-fviii inhibitory immune responses for prolonged periods of time following plasmid-mediated gene therapy. 23 However, some potential hallenges and risks may be involved in adoptive Treg ell therapy, inluding ineffiient isolation and expansion of Treg ells with or without defined allospeifiities and the possibility of expanded Treg ells reverting to proinflammatory effetor ells. Here, we report a promising strategy to indue seletive expansion of funtional Treg ells in vivo by injetion of a partiular IL-2 omplex to modulate anti-fviii responses in HemA mie. Three onseutive injetions of the optimal 1:2 ratio of IL-2 omplexes delivered with speifi shedule 1 together with one injetion of plasmid DNA enoding the FVIII gene prevent the formation of inhibitory antibodies, resulting in long-term, stable FVIII expression in treated HemA mie. Among mie with speifi shedule 2 treatment, only small portion of the mie developed the anti-fviii inhibitory antibody starting on day 3. We tested two different injetion shedules of IL-2 omplexes in an effort to indue higher number of FVIII-speifi Treg ells to modulate anti-fviii immune responses. For treatment using shedule 1, IL-2 omplexes were administered at days 5, 4, and 3, to first indue a robust expansion of Treg ells and was injeted at the peak of Treg expansion (day ), whereas for treatment using shedule 2, IL-2 omplexes treatment was performed at days 1,, and 1, with injetion at day so that antigen was present during the initial phase of Treg expansion. As shown in the results, shedule 1 treatment appeared to be muh more effetive in modulating FVIII-speifi responses than shedule 2 treatment, although both treatments indued signifiant expansion of Treg ells in mie inluding those mie that developed inhibitory antibodies. This evidene was also supported by the following adoptive transfer data: the mie reeiving Moleular Therapy vol. 19 no. 8 aug

8 IL-2 Complexes Modulate Anti-FVIII Responses The Amerian Soiety of Gene & Cell Therapy the FVIII-speifi Treg ells from the mie treated by shedule 1 showed more protetion form inhibitory antibody formation than those reeiving ells from the shedule 2 mie. The differene in the effetiveness of the two protools may be due to a requirement that highly expanded Treg ell populations are present upon first antigen exposure. At the time of injetion, Treg ell expansion is at its peak levels in the shedule 1 mie, whereas in shedule 2 mie, Treg expansion was just initiated with Treg ells at low levels. Thus, shedule 1 treatment an not only more effetively suppress the initial responses, but also indue antigenspeifi Treg ells upon antigen exposure at a later phase. In omparison, FVIII expression in mie treated with delined to bakground levels in 2 4 weeks and FVIII inhibitors ould be deteted as early as 3 4 weeks. 32,34 Spleni T ells isolated from mie treated with + IL-2 omplexes showed no indiation of reall proliferation to FVIII stimulation in vitro. Furthermore, following a 2nd plasmid hallenge at week 18 when mie had long reovered from the immune suppression of IL-2 omplexes, the tolerized mie failed to eliit an immune response. Moreover, hallenges with either T-dependent antigen (TNP- KLH) under a stringent ondition (emulsified in Freund s omplete adjuvant) or T-independent antigen (TNP-Fioll) promoted similar levels of immune responses in + IL-2 omplexes tolerized mie and untreated naive mie, indiating that IL-2 omplexes treatment did not affet the host s ability to mount immune responses to other T-dependent and T-independent neoantigens. Taken together, these data indiate that immunomodulation with IL-2 omplexes following -mediated gene therapy indued long-term protetion against FVIII-speifi immune responses in HemA mie. In addition to treating autoimmune disease 28 and graft rejetion in transplantation 27 with seletive and effiient expansion of Treg ells, our study demonstrated a new novel appliation of IL-2 omplexes, whih was to modulate antigen-speifi immune responses following gene therapy. In our study, we found that IL-2 omplexes treatment resulted in expansion of approximately fiveto sevenfold Treg ells in periphery blood, LNs and spleen at day 5 (peak day) after IL-2 omplexes treatment, and these levels gradually returned to normal within the next 7 14 days. Similarly to naturally ourring Treg ells, the expanded Treg ells ahieved stable Foxp3 expression. The inrease of Treg ell numbers was also aompanied by a signifiant inrease in the expression levels of the ativation markers CD, CTLA-4 and GITR on these ell populations. These short-lived expanded Treg ells are highly ativated and display in vitro suppressive funtion. In addition, adoptive transfer of the expanded Treg ells proteted reipient mie from generation of high-titer antibodies following hallenge. However, very little hanges were observed for other lymphoyte populations, inluding CD4 + T ells, CD8 + T ells, myeloid ells, and total B ells, onfirming that IL-2 omplexes indued seletive expansion of funtional Treg ells. The ommon -hain ytokine IL-2 predominantly ativates ells expressing high-affinity reeptors omposed of three hains [IL-2R (CD), IL-2R (CD122), and (CD132)], suh as ativated CD4 + and CD8 + T ells It also plays a ritial role in regulating the homeostasis of CD4 + Foxp3 + Treg ells Studies in mie engineered to express a Foxp3 reporter demonstrated that Foxp3 expression is onfined to a subset of T-ells and orrelates with their suppressive ativity. 41,42 Biologial funtion of most T-ells is to respond to diverse foreign antigens presented as peptides on the major histoompatibility omplex moleules of antigen presenting ell, and they make a major ontribution to host defense. As desribed earlier, biologial ativity of IL-2 in vivo an be greatly enhaned by assoiation with partiular anti-il-2 mabs. To indue expansion of Treg ells, we took advantage of IL-2/IL-2mAb immune omplexes that were initially desribed by Boyman et al. 26 and Kamimura et al. 31 IL-2/JES6-1 omplexes interat almost exlusively with T ells expressing high levels of CD, thus induing seletive expansion of CD4 + CD + Treg ells. This is beause JES6-1 mab binds to an IL-2 site that is ruial for interation with CD122 but less ruial for binding to CD. The other anti-il-2 mabs, S4B6 (speifi to murine IL-2) and MAB21 (speifi to human IL-2) formed IL-2/IL-2 mab omplexes preferentially interat with CD122, leading to the expansion of CD8 + T ells and natural killer ells. The mehanism by whih IL-2 omplexes exhibit superior apaity to treatment ompared with IL-2 or anti-il-2 mab alone in different appliations remains elusive. It was suggested that immune omplexes ontaining anti-il-2 mab may inrease the half-life of IL-2 in vivo 26,43 or their funtion may depend on the presene of neonatal FR. 44 The preise mehanisms remain to be investigated. It has been shown that IL-2 mrna expression was signifiantly indued by TGF- 1, whih led to the indution of IL-2 seretion. Furthermore, TGF- 1 supports the maintenane of Foxp3 expression, regulatory funtion, and homeostasis in peripheral CD4 + CD + regulatory T ells, 45 indiating the importane of TGF- 1 in maintaining tolerane. In our study, mie treated with IL-2 omplexes exhibited inreased TGF- 1 levels in blood at days 1, 3, and 7 post plasmid injetion, whih orrelated with the expansion and ativation of Treg ells. We hypothesize that FVIII plasmid + IL-2 omplexes treatment indued TGF- 1 dependent maturation of CD4 + CD ells into CD4 + CD + Foxp3 + Treg ells, whih maintain long-term FVIII-speifi tolerane. It was demonstrated in our lab that several immunosuppressive regimens inluding CTLA4-Ig/anti-CD4L ombination 32 and anti-icos 34 targeting B- and T-ell ostimulatory pathways, or anti-cd3 46 targeting T-helper ells suessfully prevented indution of anti-fviii immune responses in HemA mie treated with -mediated gene therapy. These regimens indued inreases in either or both of the perentages and total numbers of CD4 + Foxp3 + Treg ells, indiating the involvement of Treg ells in modulation of anti-fviii immune responses. Furthermore, the urrent study of seletive in vivo expansion of Treg ells by IL-2 omplexes treatment provides diret evidene for the important role of Treg ells in suppressing anti-fviii immune responses. Most importantly, short-term treatment that promotes transient expansion of Treg ells without depletion or alteration of other ell populations may be benefiial in linial appliations ompared to other immunomodulation therapies. This was demonstrated in the results that the host immune responses to other T-dependent or T-independent neoantigens were not perturbed in mie treated with IL-2 omplexes several months after treatment. IL-2 omplexes immunomodulation has the potential to be a safe and effetive strategy for modulating FVIII-speifi immune responses following gene therapy vol. 19 no. 8 aug. 211

9 The Amerian Soiety of Gene & Cell Therapy IL-2 Complexes Modulate Anti-FVIII Responses MATERIALS AND METHODS Mie. All mie were kept aording to the National Institutes of Health guidelines for animal are and the guidelines of Seattle Children s Researh Institute, and maintained at a speifi pathogen-free faility. HemA mie in a 129/SV C57BL/6 mixed geneti bakground were generated by targeted disruption of exon 16 of FVIII gene 47 and were used at the age of 1 16 weeks. Gene transfer of FVIII into HemA mie with immunomodulation by IL-2 omplexes injetion. HemA mie were intravenously injeted with 5 g of (pbs-hcrhpi-fviiia 48 ) in 2 ml phosphate-buffered saline via tail vein in 8 1 seonds. IL-2/anti-IL-2 mab (JES6-1A12) omplexes were prepared as previously desribed. 27 One mirogram reombinant mouse IL-2 (PeproTeh, Roky Hill, NJ) was mixed with 5 g anti-il-2 mab (JES6-1A12) (ebiosiene, San Diego, CA), inubated at 37 C for 3 minutes, and injeted intraperitoneally into mie aording to shedules speified in Results. Groups of mie treated with, and naive mie were inluded as ontrols. Seleted immunomodulated mie reeived a seond plasmid hallenge at 18 weeks after the first injetion. Blood samples were taken from the retro-orbital plexus at serial time points and assessed for FVIII ativity and anti-fviii antibody levels. Flow ytometry and antibodies. Cell suspensions of peripheral blood, LNs (from superfiial ervial) and spleens of eah treated mouse group were prepared aording to standard protools. Cell suspensions were stained for fluoresene-ativated ell sorting analysis using the following antibodies (obtained from ebiosiene unless otherwise stated): PE-Cy5- anti-mouse CD; FITC-anti-mouse CD62L (L-seletin); Alexa Fluor647- anti-mouse/ rat Foxp3; PE-anti-mouse CD152 (CTLA-4); Alexa Fluor 7- anti-mouse CD4 (BD Pharmingen, San Jose, CA), and PE-Cy7- anti-mouse GITR (BD Pharmingen). Cells were first stained for surfae markers CD4, CD, CD62L, and GITR, and subsequently stained intraellularly for Foxp3 and CTLA-4 following the ompany protool (ebiosiene). Samples were analyzed on an LSRII flow ytometer (Beton Dikinson, Palo Alto, CA) and data were analyzed using FlowJo software (Tree Star, Ashland, OR). FVIII ativities and inhibitor titer assays. Peripheral blood samples were taken from the experimental mie and olleted in a 3.8% sodium itrate solution. FVIII ativities were measured by a modified lotting assay using FVIII defiient plasma and reagents to measure ativated partial thromboplastin time and FVIII defiient plasma. 5,48 FVIII ativities were alulated from a standard urve generated by using serially diluted normal human pooled plasma. Anti-FVIII antibodies were measured by Bethesda assay as previously desribed. 49 Proliferation and suppressive assay. CD4 + T ells were isolated from spleens of mie by magneti ativated ells sorting (Miltenyi Biote, Auburn, CA). The CD4 + CD, CD4 + CD + subsets were further purified from the CD4 + T ells using a CD + Treg MACS isolation kit (Miltenyi Biote). For proliferation assay, CD4 + T ells were oultured with CD4 ells (irradiated, used as antigen presenting ells) per well in 96-well round-bottom plate with the presene of FVIII at 1 U/ml (1 U = 1 ng FVIII protein) 32 for 72 hours, followed by adding 1 μci [ 3 H]thymidine per well for the final 18 hours. [ 3 H]thymidine inorporation was measured as ounts per minute (.p.m.) in a Betaplate sintillation ounter (Perkin-Elmer, Shelton, CT). For suppressive assay, CD4 + T ells from spleens of mie treated with were used as responders and CD4 + CD + T ells from spleens of mie treated with + IL-2 omplexes at different time points after plasmid injetion or from spleens of age mathed naive mie were added as suppressor ells. To the oulture of CD4 + T ells and antigen presenting ells, we added CD4 + CD + T ells at indiated ratios. The ultures were stimulated with 1 U/ml of FVIII for 72 hours, followed by adding 1 μci [ 3 H] thymidine per well for the final 18 hours inubation. All ultures were done in tripliates. Suppression was alulated as: 5 % suppression = [1 (.p.m. (CD4 + CD T ells + CD4 + CD + T ells)/.p.m. CD4 + CD T ells alone)] 1% Adoptive transfer of hfviii-speifi Treg ells. Spleens were removed aseptially from shedule 1 or 2 or naïve untreated ontrol HemA mie at 5 days post injetion. CD4 + CD + spleni T ells were isolated and injeted into the tail vein of HemA reipients (1 1 6 ells/ mouse). Twenty-four hours post-transfer, reipient mie were hallenged with hydrodynami tail-vien injetion of 5 g of pbs-hcrhpi-hfviiia plasmid in 2 ml phosphate-buffered saline. Plasma samples olleted at various time points were analyzed for FVIII ontent and anti-fviii antibody titers by ativated partial thromboplastin time and Bethesda assays, respetively. Immunization with the T-independent antigen TNP-Fioll and T-dependent antigen TNP-KLH. Groups of tolerized mie (n = 2) and naive mie (n = 2) were immunized with μg TNP-Fioll and 1 μg TNP-KLH emulsified in Freund s omplete adjuvant intraperitoneally eah 24 weeks after plasmid injetion. A seond boost was administered intraperitoneally 3 weeks later with the same amount of TNP-Fioll and TNP-KLH. Mie were bled every 7 days and serum samples were analyzed for anti-tnp Ab by enzyme-linked immunosorbent assay in the plate oated with trinitrophenyl-bovine serum albumin (TNP-BSA). The TNP-speifi IgM, IgG, and/or IgG3 Abs were deteted by inubating with biotinylated anti-mouse Ab, followed by the addition of horseradish peroxidase-labeled streptavidin, and finally the wells were developed with substrate. Quantitation of TGF- levels. CD4 + T ells were prepared from the spleens of tolerized mie, mie treated with, hfviii plasmids only, or untreated ontrol mie as desribed in proliferation assays. TGF- onentrations were quantified in plasma pretreated with 1 N HCL, then assayed using the ELISA kit (BD Biosienes, San Jose, CA) aording to the manufaturer s reommendation, and the data were interpolated against the linear range on the standard urves. Statistial analysis. Results are presented as means SD The statistial signifiane of the differene between means was determined using the twotailed Student s t test. Differenes were onsidered signifiant at P <.5. SUPPLEMENTARY MATERIAL Figure S1. Treg ells and ativation markers in periphery blood using shedule 2 treatment. ACKNOWLEDGMENTS This work was supported by R1 grants (R1 HL6949 and R1 HL826) from NIH-NHLBI. We would like to thank Dr Peng for his help. REFERENCES 1. Antonarakis, SE, Youssoufian, H and Kazazian, HH Jr (1987). Moleular genetis of hemophilia A in man (fator VIII defiieny). Mol Biol Med 4: Zhang, AH, Skupsky, J and Sott, DW (29). Fator VIII inhibitors: risk fators and methods for prevention and immune modulation. Clin Rev Allergy Immunol 37: VandenDriesshe, T, Vanslembrouk, V, Goovaerts, I, Zwinnen, H, Vanderhaeghen, ML, Collen, D et al. (1999). Long-term expression of human oagulation fator VIII and orretion of hemophilia A after in vivo retroviral gene transfer in fator VIIIdefiient mie. Pro Natl Aad Si USA 96: Powell, JS, Ragni, MV, White, GC 2 nd, Lusher, JM, Hillman-Wiseman, C, Moon, TE et al. (23). Phase 1 trial of FVIII gene transfer for severe hemophilia A using a retroviral onstrut administered by peripheral intravenous infusion. Blood 12: Ye, P, Thompson, AR, Sarkar, R, Shen, Z, Lillirap, DP, Kaufman, RJ et al. (24). Naked DNA transfer of Fator VIII indued transgene-speifi, speies-independent immune response in hemophilia A mie. Mol Ther 1: Moayeri, M, Hawley, TS and Hawley, RG (). Corretion of murine hemophilia A by hematopoieti stem ell gene therapy. Mol Ther 12: Jiang, H, Lillirap, D, Patarroyo-White, S, Liu, T, Qian, X, Sallan, CD et al. (26). Multiyear therapeuti benefit of AAV serotypes 2, 6, and 8 delivering fator VIII to hemophilia A mie and dogs. Blood 18: MCormak, WM Jr, Seiler, MP, Bertin, TK, Ubhayakar, K, Palmer, DJ, Ng, P et al. (26). Helper-dependent adenoviral gene therapy mediates long-term orretion of the lotting defet in the anine hemophilia A model. J Thromb Haemost 4: Moleular Therapy vol. 19 no. 8 aug

10 IL-2 Complexes Modulate Anti-FVIII Responses The Amerian Soiety of Gene & Cell Therapy 9. Waters, B and Lillirap, D (29). The moleular mehanisms of immunomodulation and tolerane indution to fator VIII. J Thromb Haemost 7: Miao, CH (27). Reent advanes in immune modulation. Curr Gene Ther 7: Herzog, RW and Dobrzynski, E (24). Immune impliations of gene therapy for hemophilia. Semin Thromb Hemost 3: Sakaguhi, S, Ono, M, Setoguhi, R, Yagi, H, Hori, S, Fehervari, Z et al. (26). Foxp3+ CD+ CD4+ natural regulatory T ells in dominant self-tolerane and autoimmune disease. Immunol Rev 212: Belkaid, Y (27). Regulatory T ells and infetion: a dangerous neessity. Nat Rev Immunol 7: Hawrylowiz, CM and O Garra, A (). Potential role of interleukin-1-sereting regulatory T ells in allergy and asthma. Nat Rev Immunol 5: Wu, K, Bi, Y, Sun, K and Wang, C (27). IL-1-produing type 1 regulatory T ells and allergy. Cell Mol Immunol 4: Shevah, EM, DiPaolo, RA, Andersson, J, Zhao, DM, Stephens, GL and Thornton, AM (26). The lifestyle of naturally ourring CD4+ CD+ Foxp3+ regulatory T ells. Immunol Rev 212: Izue, A, Coombes, JL and Powrie, F (26). Regulatory T ells suppress systemi and muosal immune ativation to ontrol intestinal inflammation. Immunol Rev 212: Belkaid, Y, Blank, RB and Suffia, I (26). Natural regulatory T ells and parasites: a ommon quest for host homeostasis. Immunol Rev 212: Rouse, BT, Sarangi, PP and Suvas, S (26). Regulatory T ells in virus infetions. Immunol Rev 212: Aluvihare, VR, Kallikourdis, M and Betz, AG (24). Regulatory T ells mediate maternal tolerane to the fetus. Nat Immunol 5: Beyer, M and Shultze, JL (26). Regulatory T ells in aner. Blood 18: Miao, CH (21). Immunomodulation for inhibitors in hemophilia A: the important role of Treg ells. Expert Rev Hematol 3: Miao, CH, Harmeling, BR, Ziegler, SF, Yen, BC, Torgerson, T, Chen, L et al. (29). CD4+FOXP3+ regulatory T ells onfer long-term regulation of fator VIII-speifi immune responses in plasmid-mediated gene therapy-treated hemophilia mie. Blood 114: Almeida, AR, Borghans, JA and Freitas, AA (21). T ell homeostasis: thymus regeneration and peripheral T ell restoration in mie with a redued fration of ompetent preursors. J Exp Med 194: Curotto de Lafaille, MA, Lino, AC, Kuthukhidze, N and Lafaille, JJ (24). CD- T ells generate CD+Foxp3+ regulatory T ells by peripheral expansion. J Immunol 173: Boyman, O, Kovar, M, Rubinstein, MP, Surh, CD and Sprent, J (26). Seletive stimulation of T ell subsets with antibody-ytokine immune omplexes. Siene 311: Webster, KE, Walters, S, Kohler, RE, Mrkvan, T, Boyman, O, Surh, CD et al. (29). In vivo expansion of T reg ells with IL-2-mAb omplexes: indution of resistane to EAE and long-term aeptane of islet allografts without immunosuppression. J Exp Med 26: Liu, R, Zhou, Q, La Cava, A, Campagnolo, DI, Van Kaer, L and Shi, FD (21). Expansion of regulatory T ells via IL-2/anti-IL-2 mab omplexes suppresses experimental myasthenia. Eur J Immunol 4: Boyman, O, Surh, CD and Sprent, J (26). Potential use of IL-2/anti-IL-2 antibody immune omplexes for the treatment of aner and autoimmune disease. Expert Opin Biol Ther 6: Tang, Q, Adams, JY, Penaranda, C, Melli, K, Piaggio, E, Sgouroudis, E et al. (28). Central role of defetive interleukin-2 prodution in the triggering of islet autoimmune destrution. Immunity 28: Kamimura, D, Sawa, Y, Sato, M, Agung, E, Hirano, T and Murakami, M (26). IL-2 in vivo ativities and antitumor effiay enhaned by an anti-il-2 mab. J Immunol 177: Miao, CH, Ye, P, Thompson, AR, Rawlings, DJ and Ohs, HD (26). Immunomodulation of transgene responses following naked DNA transfer of human fator VIII into hemophilia A mie. Blood 18: Cao, O, Lodua, PA and Herzog, RW (29). Role of regulatory T ells in tolerane to oagulation fators. J Thromb Haemost 7 Suppl 1: Peng, B, Ye, P, Blazar, BR, Freeman, GJ, Rawlings, DJ, Ohs, HD et al. (28). Transient blokade of the induible ostimulator pathway generates long-term tolerane to fator VIII after nonviral gene transfer into hemophilia A mie. Blood 112: Cantrell, DA and Smith, KA (1983). Transient expression of interleukin 2 reeptors. Consequenes for T ell growth. J Exp Med 158: Cantrell, DA and Smith, KA (1984). The interleukin-2 T-ell system: a new ell growth model. Siene 224: Taniguhi, T and Minami, Y (1993). The IL-2/IL-2 reeptor system: a urrent overview. Cell 73: Setoguhi, R, Hori, S, Takahashi, T and Sakaguhi, S (). Homeostati maintenane of natural Foxp3(+) CD(+) CD4(+) regulatory T ells by interleukin (IL)-2 and indution of autoimmune disease by IL-2 neutralization. J Exp Med 21: D Cruz, LM and Klein, L (). Development and funtion of agonist-indued CD+Foxp3+ regulatory T ells in the absene of interleukin 2 signaling. Nat Immunol 6: Fontenot, JD, Rasmussen, JP, Gavin, MA and Rudensky, AY (). A funtion for interleukin 2 in Foxp3-expressing regulatory T ells. Nat Immunol 6: Ponpuak, M, Davis, AS, Roberts, EA, Delgado, MA, Dinkins, C, Zhao, Z et al. (21). Delivery of ytosoli omponents by autophagi adaptor protein p62 endows autophagosomes with unique antimirobial properties. Immunity 32: Wan, YY and Flavell, RA (). Identifying Foxp3-expressing suppressor T ells with a biistroni reporter. Pro Natl Aad Si USA 12: Finkelman, FD, Madden, KB, Morris, SC, Holmes, JM, Boiani, N, Katona, IM et al. (1993). Anti-ytokine antibodies as arrier proteins. Prolongation of in vivo effets of exogenous ytokines by injetion of ytokine-anti-ytokine antibody omplexes. J Immunol 151: Létourneau, S, van Leeuwen, EM, Krieg, C, Martin, C, Pantaleo, G, Sprent, J et al. (21). IL-2/anti-IL-2 antibody omplexes show strong biologial ativity by avoiding interation with IL-2 reeptor alpha subunit CD. Pro Natl Aad Si USA 17: Marie, JC, Letterio, JJ, Gavin, M and Rudensky, AY (). TGF-beta1 maintains suppressor funtion and Foxp3 expression in CD4+CD+ regulatory T ells. J Exp Med 21: Peng, B, Ye, P, Rawlings, DJ, Ohs, HD and Miao, CH (29). Anti-CD3 antibodies modulate anti-fator VIII immune responses in hemophilia A mie after fator VIII plasmid-mediated gene therapy. Blood 114: Bi, L, Sarkar, R, Naas, T, Lawler, AM, Pain, J, Shumaker, SL et al. (1996). Further haraterization of fator VIII-defiient mie reated by gene targeting: RNA and protein studies. Blood 88: Miao, CH, Ye, X and Thompson, AR (23). High-level fator VIII gene expression in vivo ahieved by nonviral liver-speifi gene therapy vetors. Hum Gene Ther 14: Kasper, CK, Aledort, L, Aronson, D, Counts, R, Edson, JR, van Eys, J et al. (1975). Proeedings: A more uniform measurement of fator VIII inhibitors. Thromb Diath Haemorrh 34: Belghith, M, Bluestone, JA, Barriot, S, Mégret, J, Bah, JF and Chatenoud, L (23). TGF-beta-dependent mehanisms mediate restoration of self-tolerane indued by antibodies to CD3 in overt autoimmune diabetes. Nat Med 9: vol. 19 no. 8 aug. 211