Identification of a lymphoid cell-specific human minor histocompatibility antigen that arises from differential protein translation

5122

Minor histocompatibility antigens (mHAg) that are expressed selectively in hematopoietic cells, including leukemic cells, but not widely expressed in nonhematopoietic tissues represent potential targets for T cell therapy after MHC-matched allogeneic hematopoietic cell transplantation (HCT) to enhance graft-versus-leukemia (GVL) reactions without inducing or aggravating graft-versus-host disease. We isolated from an HLA-matched HCT recipient a CTL clone that recognized a novel HLA-A3-restricted mHAg with hematopoietic-specific expression. In vitro cytotoxicity assays revealed that the mHAg was presented at high levels on recipient EBV-transformed B cells (BLCL), and at low levels on recipient PHA-stimulated T cells, but was not presented on recipient dermal fibroblasts or any donor cells. When tested against a panel of allogeneic HLA-A3⁺ leukemic cells, the CTL lysed malignant cells of B-lymphoid origin but not of myeloid origin. To identify the epitope, HLA-A3-associated peptides were extracted from a mHAg⁺ BLCL, fractionated by reversed-phase HPLC, and tested for the ability to reconstitute CTL recognition of mHAg^- donor BLCL. After 3 rounds of RP-HPLC, candidate peptides within biologically active fractions were determined using nanospray Fourier transform mass spectrometry. The most abundant candidate peptide ion was sequenced by collision activated dissociation, and the corresponding synthetic peptide sensitized donor BLCL to CTL lysis, with half-maximal lysis observed at a peptide concentration of 2 nM. A database search revealed that an alternative transcript of the C22orf18 gene encodes this peptide. To elucidate the basis for differential CTL recognition of donor and recipient BLCL, C22orf18 alleles from each were sequenced and compared. The mHAg^- donor was homozygous for a C→T single nucleotide polymorphism (SNP) within the codon for the first residue of the epitope that created a translation stop codon, suggesting that this mHAg’s antigenicity results from differential protein translation in donor and recipient cells. PCR-RFLP analysis of this SNP in 30 additional HLA-A3⁺ individuals revealed that the frequency of the T (stop codon) allele was 0.25, and phenotyping confirmed that homozygosity for the T allele was associated with resistance to CTL lysis. RT-PCR analysis revealed that the C22orf18 alternative transcript is expressed almost exclusively in resting and activated CD19⁺ B-lymphoid cells, at low levels in CD4⁺, CD8⁺, and CD14⁺ cells, and negligibly in non-hematopoietic tissues. Thus, therapeutic strategies targeting this mHAg could selectively enhance the GVL effect in HCT recipients with B-lymphoid malignancies. These studies additionally provide the first evidence that mHAg can result from differential allelic protein translation.