Induction of dominant transplantation tolerance by an altered peptide ligand of the male antigen Dby

Abstract

T cell reactivity to minor histocompatibility (mH) antigens is responsible for rejection of HLA-matched allografts, limiting the effectiveness of transplantation for the treatment of end-stage organ failure. The deadbox gene Dby is located on the Y chromosome and encodes an mH antigen that prompts rejection of male tissues by female mice. Establishing a network of regulatory T (T(reg)) cells that is capable of coercing naive cells to adopt a tolerant phenotype offers an attractive strategy for immune intervention in such deleterious immune responses. While various approaches have successfully induced a dominant form of transplantation tolerance, they share the propensity to provoke chronic, incomplete activation of T cells. By identifying the T cell receptor (TCR) contact sites of the dominant epitope of the Dby gene product, we have designed an altered peptide ligand (APL) that delivers incomplete signals to naive T cells from A1 infinity RAG1(-/-) mice that are transgenic for a complementary TCR. Administration of this APL to female transgenic mice polarizes T cells toward a regulatory phenotype, securing a form of dominant tolerance to male skin grafts that is capable of resisting rejection by naive lymphocytes. Our results demonstrate that incomplete signaling through the TCR may establish a network of T(reg) cells that may be harnessed in the service of transplantation tolerance.

Figure 1

Identification of the H-2Ek binding motif of the dominant epitope of Dby. (A) Proliferative responses of naive female A1 ∞ RAG1–/– T cells to the 479–493 peptide from the male antigen, Dby (filled circles), compared with its female homolog, Dbx (open triangles). (B) Competition binding assay showing the enhanced capacity of Dby (filled circles) to compete with HEL_1-18 for binding to H-2Ek compared to Dbx (open triangles), as determined by the extent of inhibition of IL-2 secretion by the 2G7.1 hybridoma. (C) Stimulatory capacity of analogs of the Dby epitope, substituted with alanine at putative TCR contact residues. (D) Schematic diagram highlighting residues of the Dby epitope that determine its interaction with H-2Ek and the A1 TCR. The conventional arrows define those residues that interact directly with either the MHC or TCR, while the broad arrow denotes the putative primary TCR contact residue. All data shown are representative of at least three independent experiments.

Figure 2

Characterization of APLs bearing substitutions at critical TCR contact residues. (A) IL-2 release by 2G7.1 cultured in the presence of 0.5 ∝M HEL_1–18 and a 100 ∝M excess of APLs. All APLs competed as efficiently as the cognate peptide for binding to H-2Ek with the exception of 484H Ø R, which bound with similar affinity to the control peptide, Dbx. (B) Proliferative responses of naive female A1 ∞ RAG1–/– T cells to APLs bearing conservative substitutions at critical TCR contact sites. All experiments were repeated at least twice and yielded similar results on each occasion.

Figure 3

The 490R Ø H APL delivers partial activation signals to naive female A1 ∞ RAG1–/– T cells. (A) The A1 TCR is rapidly downregulated in response to its cognate ligand, Dby (filled circles), and the agonist peptide 487R Ø K (open squares). In contrast, neither the female homolog, Dbx (open triangles), nor the partial agonist 490R Ø H (filled squares) have any detectable impact on levels of expression of the TCR. (B) Release of IL-10 by naive female A1 ∞ RAG1–/– T cells in response to Dby (white bars) or the 490R Ø H APL (black bars). Asterisks denote samples in which IL-10 could not be detected. The results presented are representative of at least three independent experiments.

Figure 4

The 490R Ø H APL polarizes naive T cells toward a predominantly IL-10–secreting phenotype. (A–D) Intracellular cytokine staining of naive A1 ∞ RAG1–/– T cells stimulated for 5 days with 500 nM 490R Ø H (A and B) or 5 nM Dby (C and D). (E and F) Intracellular cytokine staining of A1 ∞ RAG1–/– T cells stimulated five times at 14-day intervals with the 490R Ø H APL to simulate chronic, incomplete stimulation. The results presented are typical of at least three independent experiments.

Figure 5

The 490R Ø H APL induces dominant transplantation tolerance in vivo. (A) Long-term acceptance of syngeneic male skin grafts by female A1 ∞ RAG1–/– mice treated with five doses of 20 μg of 490R Ø H (n = 16) (filled squares). All control mice receiving the cognate Dby peptide rejected their grafts with a mean survival time of 12 days (n = 7) (filled inverted triangle), as did 10/14 mice treated with PBS alone (filled triangles). The results presented are pooled from three independent experiments. (B) Mice rendered tolerant by administration of 490R Ø H (n = 5) fail to reject existing and fresh skin grafts from male donors when infused with 107 T cells from naive A1 ∞ RAG1–/– mice (filled triangle). The same number of T cells administered to control RAG1–/– mice (n = 5) rapidly mediated graft rejection (filled squares). (C) Neutralization of IL-10 activity fails to abrogate tolerance induced by 490R Ø H. Mice tolerant of male skin grafts were treated with 1B1.2, specific for the IL-10R (n = 10) (filled triangle) or an isotype-controlled mAb (n = 10) (filled inverted triangle) before being infused with 5 ∞ 106 naive A1 ∞ RAG1–/– splenocytes and a fresh male skin graft. The same number of naive splenocytes induced rapid rejection of male skin grafts when infused into control RAG1–/– recipients (n = 5) (filled squares).

Figure 6

Phenotypic analysis of T cells from mice rendered tolerant of male skin grafts after administration of 490R Ø H. (A) Flow-cytometric analysis of T cells pooled from six tolerated male grafts confirming the accumulation of CD4+CD25+ cells. (B) Real-time quantitative RT-PCR analysis of Foxp3 mRNA expression by tolerated and control skin compared with unfractionated cells from the spleens and thymi of tolerant and naive mice. Populations of CD4+CD25- and CD4+CD25+ T cells, sorted from naive CBA/Ca mice to greater than 90% purity, were included as controls. Foxp3 mRNA expression was normalized against CD3γ mRNA levels, so as to correct for the T cell content of each tissue.

Figure 7

Female A1 ∞ RAG1–/– T cells rendered tolerant in vivo by repeated administration of 490R Ø H secrete predominantly IL-10 when challenged in vitro with Dby. Cytokine release, measured by ELISA, was determined for T cells purified from naive (white bars) or tolerant mice (black bars) stimulated ex vivo with 100 nM Dby. Asterisks denote samples in which cytokines could not be detected. Results are representative of four independent experiments.