Original Article The Type and Frequency of Immunoregulatory CD4  T-Cells Govern the Efficacy of Antigen-Specific Immunotherapy in Nonobese Diabetic Mice Shannon M. Pop, 1 Carmen P. Wong, 2 Qiuming He, 2 Yaming Wang, 2 Mark A. Wallet, 2 Kevin S. Goudy, 2 and Roland Tisch 1,2 Antigen-specific immunotherapy, an approach to selec- tively block autoimmune diabetes, generally declines in nonobese diabetic (NOD) mice as disease progresses. To define the parameters influencing the efficacy of antigen- specific immunotherapy once diabetes is established, plas- mid DNA (pDNA) vaccination was used to suppress autoimmune-mediated destruction of syngeneic islet grafts in diabetic NOD recipients. pDNAs encoding a glutamic acid decarboxylase 65 (GAD65)-Ig molecule (pGAD65), interleukin (IL)-4 (pIL4), and IL-10 (pIL10) significantly delayed the onset of recurrent diabetes compared with pGAD65pIL10-vaccinated recipients. Despite differences in efficacy, a similar frequency of GAD65-specific CD4  T-cells secreting IL-4, IL-10, or interferon- were de- tected in mice treated with pGAD65pIL4pIL10 and pGAD65pIL10. However, the frequency of FoxP3-ex- pressing CD4  CD25  CD62L hi T-cells was increased in the renal and pancreatic lymph nodes of diabetic recipients vaccinated with pGAD65pIL4pIL10. These immuno- regulatory CD4  CD25  T-cells (CD4  CD25  Treg) exhib- ited enhanced in vivo and in vitro suppressor activity that partially was transforming growth factor- dependent. Furthermore, duration of islet graft protection in pGAD65pIL4pIL10-vaccinated diabetic recipients cor- related with the persistence of CD4  CD25  Treg. These data demonstrate that the frequency and maintenance of FoxP3-expressing CD4  CD25  Treg influence antigen-in- duced suppression of ongoing -cell autoimmunity in dia- betic recipients. Diabetes 56:1395–1402, 2007 T ype 1 diabetes is characterized by the destruc- tion of the pancreatic -cells (1–3). The primary mediators of -cell destruction are CD4 + and CD8 + T-cells, which typically exhibit a proin- flammatory type 1 phenotype (1–3). Skewing toward type 1 differentiation by naive -cell–specific T-cells is believed, in part, to be because of a deficiency in immunoregulatory CD4 + T-cells. Reduced frequencies of -cell–specific “adaptive” immunoregulatory CD4 + T-cells, such as Th2 and Tr1 cells, have been reported in diabetic individuals and nonobese diabetic (NOD) mice (4 – 8). These CD4 + T effectors regulate autoimmunity primarily through the bystander effects of interleukin (IL)-4 and -10 secretion. Diminished numbers and/or suppressor function of “natu- ral” regulatory CD4 + CD25 + CD62L hi T-cells (CD4 + CD25 + Treg) also have been documented in diabetic NOD mice and patients (9 –14). CD4 + CD25 + Treg cells are character- ized by a potent suppressor capacity mediated by cell-to- cell contact and/or expression of transforming growth factor (TGF)-1 and possibly IL-10 (15–18). Whereas adap- tive immunoregulatory effectors are established once na- ı¨ve CD4 + T-cells encounter antigen in the periphery, the phenotype and suppressor function of CD4 + CD25 + Treg is induced in the thymus upon recognition of self-antigen and expression of the FoxP3 transcription factor (19 –22). In view of the functional imbalance between pathogenic type 1 and immunoregulatory -cell–specific T-cells, vari- ous strategies of immunotherapy to prevent and treat type 1 diabetes have been devised to reestablish peripheral immunoregulation (23). Previous studies (24 –27) have demonstrated that administration of -cell autoantigens, such as glutamic acid decarboxylase 65 (GAD65), or insulin by various means induces Th2- and Tr1-like cells and prevents the onset of diabetes in NOD mice. Antigen- specific immunotherapy also provides a possible strategy to rescue residual -cell mass and/or block autoimmune- mediated destruction of islet grafts in diabetic individuals. However, whether sufficient -cell–specific immunoregu- latory CD4 + T-cells can be induced to suppress autoim- munity under overt diabetic conditions, in which the number of pathogenic effectors is high and only a limited pool of naı¨ve -cell–specific CD4 + T-cells exists (28,29), remains largely untested. We previously demonstrated that administration of plas- mid DNA (pDNA) vaccines encoding a GAD65 immuno- globulin (Ig) chimeric molecule, IL-4, and/or IL-10 is an effective approach to suppress late preclinical type 1 From the 1 Curriculum in Oral Biology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; and the 2 Department of Microbiology and Immunology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina. Address correspondence and reprint requests to Roland Tisch, PhD, De- partment of Microbiology and Immunology, Mary Ellen Jones Bldg., Room 804, Campus Box no. 7290, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7290. E-mail: rmtisch@med.unc.edu. Received for publication 21 April 2006 and accepted in revised form 8 February 2007. Published ahead of print at http://diabetes.diabetesjournals.org on 22 Feb- ruary 2007. DOI: 10.2337/db06-0543. GAD65, glutamic acid decarboxylase 65; HEL, hen egg lysozyme; IL, interleukin; IFN, interferon; pDNA, plasmid DNA; PLN, pancreatic lymph node; RLN, renal lymph node; Treg, regulatory T-cell; TGF, transforming growth factor. © 2007 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. DIABETES, VOL. 56, MAY 2007 1395