Background: The Kell blood group system contains over 25 antigens, with the K1 antigen being recognised as the most clinically important antigen after the ABO and RhD. K1 and its antithetical antigen K2 differ by a polymorphism at aa193, resulting in the substitution of Met for Thr in the K2 antigen. Our focus is on the role of helper T (Th) cells in the alloresponse to K1, which, occurs as a result of antigen mismatch during transfusion or pregnancy. In particular, the aim was to map the helper epitopes spanning the K1 polymorphism and to characterise the responding Th cells.

Methods: To map the T-helper cell epitopes responsible for driving the alloresponse K1, peripheral blood mononuclear cells (PBMC) were isolated from K1 negative donors who had anti-K1 alloantibodies and stimulated with panels of overlapping 15-mer peptides, “walking” one amino acid at a time, down the sequences containing either the K1 Met, or K2 Thr polymorphism. The peptides were screened for the ability to stimulate recall proliferative responses and Th cytokine production, and the responding cells phenotyped by flow cytometric analysis.

Results: PBMC from over 90% of alloimmune donors tested exhibited significant proliferation in response to the panel of K1 Met peptides, particularly those with the polymorphism at or near the C-terminus. Most notably, K1 peptide 1m(179–193), was found to induce significant proliferative responses from over 80% of K1 alloimmunised donors tested; proliferation was not associated with any particular HLA Class II type. There was no clear helper subset bias in the responses to the stimulatory peptides, with the balance between production of the respective Th1 and Th2 cytokines interferon-γ and interleukin-4 varying between donors. Flow cytometry confirmed that the proliferating cells were of the helper CD3+CD4+ phenotype and depletion experiments demonstrated that they were derived from the previously activated “memory” CD45RO+ population. Interestingly, a high background of proliferative responses was observed in response to the control K2 peptides by Th cells from both alloimmunised and unimmunised control donors. In vivo the K2 antigen is N-glycosylated at aa position 191, whereas the synthetic K2 peptide panel tested in vitro were not, raising the possibility that the presence or absence of the sugar residues affects Th cell recognition. The K1 substitution disrupts the glycosylation motif, and so such an effect would be relevant only for the K2 sequences. To test the possible role of glycosylation in Th responses to K2 peptides, PBMC were stimulated with either the original un-glycosylated version of peptide 1t (K2(179–193)), or a glycosylated version of the same peptide (WRISGKWTSLNF-Asn(AcNH-b-Glc)-RT-amide). Glycosylation resulted in abrogation of the Th cell proliferative responses previously observed to the original peptide 1t (K2(179–193)).

Conclusion: A dominant alloreactive Th cell epitope has been mapped to K1 peptide 1m (K1(179–193)). Sugar residues attached to the gylosylation motif formed by the K2 substitution are important for Th recognition, and thus for self-tolerance to the corresponding peptides. Steric hindrance, preventing T cell receptor contact with the relevant amino acid residues, is the most likely mechanism, and, in the absence of sugar moieties on synthetic K2 peptides, these are effectively recognised as “foreign” by a Th repertoire that has been educated in the context of exposure to the gylosylated versions in vivo. We speculate that the immunogenicity of K1 may be attributed, at least in part, to the failure of peptides derived from the glycosylated K2 sequence to be recognised by Th cells during the establishment of self-tolerance in vivo, thereby leading to the survival of a larger pool of Th cells responsive to epitopes spanning the amino acid backbone at the polymorphic site. The identified dominant epitope is a candidate to be taken forward for inducing therapeutic tolerance, to prevent or switch off anti-K1 responses.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

Sign in via your Institution