Generation of individualized immunocompatible endothelial cells from HLA-I-matched human pluripotent stem cells

Endothelial cells (ECs) derived from human-induced pluripotent stem cell (iPSC) are a valuable cell resource for cardiovascular regeneration. To avoid time-consuming preparation from primary autologous cells, the allogeneic iPSC-ECs are being expected to become "off-the-shelf" cell products. However, allorejection caused by HLA mismatching is a major barrier for this strategy. Although the "hypoimmunogenic" iPSCs could be simply generated by inhibition of HLA-I expression via β-2 microglobulin knockout (B2M KO), the deletion of HLA-I expression will activate natural killer (NK) cells, which kill the HLA-I negative cells. To inhibit NK activation, we proposed to generate HLA-matched iPSCs based on patient's HLA genotyping by HLA exchanging approach to express the required HLA allele. To establish a prototype of HLA exchanging system, the expression of HLA-I molecules of iPSCs was inhibited by CRISPR/Cas9-mediated B2M KO, and then HLA-A*11:01 allele, as a model molecule, was introduced into B2M KO iPSCs by lentiviral gene transfer. HLA-I-modified iPSCs were tested for their pluripotency and ability to differentiate into ECs. The stimulation of iPSC-EC to allogeneic T and NK cells was detected by respective co-culture of PBMC-EC and NK-EC. Finally, the iPSC-ECs were used as the seeding cells to re-endothelialize the decellularized valves. We generated the iPSCs only expressed one HLA-A allele (HLA-A *11:01) by B2M KO plus HLA gene transfer. These HLA-I-modified iPSCs maintained pluripotency and furthermore were successfully differentiated into functional ECs assessed by tube formation assay. Single HLA-A*11:01-matched iPSC-ECs significantly less induced the allogeneic response of CD8+ T cell and NK cells expressing matched HLA-A*11:01 and other HLA-A,-B and -C alleles. These cells were successfully used to re-endothelialize the decellularized valves. In summary, a simple HLA-I exchanging system has been created by efficient HLA engineering of iPSCs to evade both of the alloresponse of CD8+ T cells and the activation of NK cells. This technology has been applied to generate iPSC-ECs for the engineering of cellular heart valves. Our strategy should be extremely useful if the "off-the-shelf" and "non-immunogenic" allogeneic iPSCs were created for the common HLA alleles.