Linking HLA class II susceptibility to HSPC transcriptomics and antigen presentation in aplastic anemia

Acquired aplastic anemia (AA) is an immune-mediated bone marrow failure syndrome characterized by destruction of hematopoietic stem and progenitor cells (HSPCs). Although HLA class II alleles are consistently associated with disease susceptibility, the candidate self-peptides presented by these molecules remain largely unknown.

We performed a retrospective case-control study evaluating HLA class II associations in 35 patients with AA and 570 unrelated healthy controls. Bulk transcriptomic analysis of untreated Lin^-CD34⁺ HSPCs (GSE165870) identified differentially expressed genes, which were subsequently prioritized using an independent single-cell RNA sequencing dataset (GSE145668) to identify genes preferentially expressed in HSPCs relative to T cells. Canonical protein sequences were fragmented into overlapping 15-mer peptides and analyzed for binding to disease-associated HLA class II molecules using NetMHCIIpan v4.3j.

HLA-DRB1*15:01, HLA-DQB1*06:02, and HLA-DRB5*01:01 were significantly enriched in patients with AA, and the three-allele co-occurrence pattern showed the strongest association with disease susceptibility. Bulk transcriptomic analysis identified 98 differentially expressed genes, including a prominent interferon-associated transcriptional program. Integration with single-cell transcriptomic data prioritized nine HSPC-enriched genes (SNCA, UHRF1, SH2D3C, SLC25A39, ADIPOR1, BNIP3L, LAT2, MEF2A, and LGALS1) for downstream analyses. NetMHCIIpan identified multiple predicted HLA class II binders across the three disease-associated HLA molecules, with SNCA, UHRF1, SH2D3C, ADIPOR1, and SLC25A39 emerging as the highest-priority candidate proteins based on normalized binding density and predicted binding affinity.

This integrative analysis combines HLA genetics, transcriptomics, and computational peptide-binding prediction to prioritize candidate HSPC-derived self-peptides in acquired AA. These findings provide a hypothesis-generating framework for future immunopeptidomic and functional studies aimed at identifying the antigenic targets underlying immune-mediated bone marrow failure.