Specific targeting of HLA genes generates hypoimmunogenic hematopoietic stem cells for allogeneic transplantation Free

Pre-transplant human leukocyte antigen (HLA) matching between donors and recipients is essential for minimizing immunological rejection in allogeneic hematopoietic stem cell transplantation (allo-HSCT). However, stringent HLA requirments severely limit the pool of eligible donors. Although umbilical-cord blood, with its inherently low immunogenicity, is an attractive graft source, many patients still lack a suitable donor. Previous studies have used CRISPR/Cas9 to generate hypoimmunogenic immune cells, while this strategy has not been utilized in HSCT. In this study, we employed CRISPR/Cas9 to selectively disrupt HLA genes in human hematopoietic stem cells (HSCs)，enhancing donor-recipient matching and reconstitution capacity.

Firstly, based on the HLA-A/B/C/DR matching criteria for umbilical cord blood transplantation and the frequency of HLA haplotypes in the Chinese population, we calculated the matching probabilities for complete matching at all 4 loci, as well as deletion of each locus. Compared to complete matching, the matching probability for HLA-B/C/DR can be doubled, while those for HLA-C/DR can be tripled. These results indicated that deleting HLA-A alone and deleting both HLA-A and HLA-B can markedly expand the pool of acceptable donors.

To achieve precise HLA knockout, we constructed 2 CRISPR/Cas9 guide RNA (sgRNA) libraries targeting HLA-A and HLA-B. After systematic screening, we identified more than 20 sgRNAs that exceeded 60% knockout efficiency in 293T cells and covered over 99% of HLA-A/B alleles in the Chinese population. Based on knockout efficiency, we selected the top 3 sgRNAs to target HLA genes in umbilical cord blood-derived HSCs. After 3 days' ex vivo culture, knockout efficiency exceeded 90% for single or double knockout of HLA-A and HLA-B (A-KO, B-KO, A-B KO) in CD34+ hematopoietic stem and progenitor cells (HSPCs). In addition, mobilized peripheral blood-derived HSPCs showed comparable knockout efficiency, underscoring the robustness of our approach across clinically relevant stem-cell sources.

To investigate whether the functions of HLA-A/B knockout HSPCs (KO-HSPCs) were impaired in vitro, colony forming cell and multi-lineage differentiation assays were performed and the results confirmed that the abilities of colony forming, self-renewal and lineage differentiation were well retained. To examine the immunogenicity of KO-HSPCs, we performed CFSE-labeled T cell proliferation and CD107a assays. KO-HSPCs were less recognized and killed by T cells, nor did they activate nature killer (NK) cells, further confirming their low immunogenicity. To assess the engraftment and reconstitution function in vivo, we transplanted edited and non-edited HSPCs into NOG mice via tail vein injection. Transplantation studies showed significantly higher engraftment with A and A-B KO HSPCs compared with non-edited ones in peripheral blood, bone marrow and spleen. Additionally, no significant difference was observed in sub-population differentiation between the edited and unedited HSPCs. Competitive transplantation results showed that KO-HSPCs dominate implantation at around 8 weeks. Collectively, these results demonstrated the engraftment advantage of KO-HSPCs.

Finally, we evaluated the safety of the KO-HSPCs. Sequencing of the 3 KO-HSPCs revealed no off-target sites. Pathological morphological analysis of vital organs, including bone marrow, heart, liver, spleen, lung, and kidney, in NOG mice reconstituted with KO-HSPCs after 16 weeks showed no tumorigenesis.

In summary, our data demonstrate the feasibility of generating hypoimmunogenic HSCs by precisely editing HLA alleles, thereby increasing accessibility and availability for patients in need of HSCT.