Natural Killer Cells Derived From Human Pluripotent Stem Cells Provide a Novel Method to Treat HIV-1 Infection.

Natural killer (NK) cells are known to be key components of the innate immune system with the ability to kill diverse tumor cells and virally infected cells. Our group has previously demonstrated derivation of CD45⁺CD56⁺ natural killer (NK) cells from human embryonic stem cells (hESCs)-derived hematopoietic CD34⁺CD45⁺ progenitor cells. These hESC-derived NK cells demonstrate potent killing of various tumor cells both in vitro and in vivo. More recently, we have also successfully generated NK cells from similar CD34⁺CD45⁺ hematopoietic progenitor cells derived from human induced pluripotent stem cells (iPSCs). Again, we find that these iPSC-derived NK cells also have effective anti-tumor activity in vitro. Notably, both the hESC and iPSC-derived NK cells are uniformly CD94⁺CD117⁻, corresponding to a more mature and cytotoxic NK cell population. This is in contrast to NK cells derived from umbilical cord blood (UCB) progenitor cells that produce a mixture of CD94⁺CD117⁻ and CD94⁻CD117⁺ NK cells that are more heterogeneous in their cytotoxic activity. Previous studies of NK cells isolated from peripheral blood indicate they have activity against HIV-1-infected cells. Therefore, we hypothesized that both hESC and iPSC-derived NK cells would be able to kill HIV-1-infected targets. We have applied multiple complementary systems to test this hypothesis using both HIV-1-infected cell lines and HIV-1-infected primary T cells. First, we used a chronically infected cell line (H9/HTLV IIIB) to demonstrate specific cytolytic activity of hESC-derived NK cells. Here, we found CD107a expression (a marker of NK cell functional activity) was significantly upregulated on hESC-derived effectors stimulated by the HIV-1-infected targets compared to uninfected targets (13.7% vs. 4.3%). Next, we utilized primary human CD4 T cells infected with primary patient isolate HIV96-480 as targets to demonstrate the same effect to specifically activate both hESC and iPSC-derived NK cells. In both of these studies, control NK cell populations derived from human umbilical cord blood progenitor cells were significantly less active against HIV than the hESC and iPSC-derived NK cells. In addition to the cytolytic function against HIV-1-infected targets, we demonstrate hESC and iPSC-derived NK cells also suppress HIV-1 replication by producing CC-chemokines to competitively inhibit CCR5 co-receptor binding. CCL4 (MIP1b), a CCR5 ligand, is greatly induced in both hESC and iPSC-derived NK cells after incubation with HIV-1-infected targets compared to uninfected targets: 37.2% (HIV-infected) vs. 24.8% (uninfected) for hESC-NK cells and 32.5% (HIV-infected) vs. 21.6% (uninfected) for iPSC-NK cells. Lastly, we have also tested suppression of acute HIV-1 infection by hESC-derived NK cells in vitro. Here, the T-cell line CEM-GFP was infected with HIV-1 NL4-3 and cocultured with hESC-derived NK cells for two weeks. HIV-1-infected targets detected by flow cytometry for GFP expression were strongly decreased in comparison to controls in absence of NK cells (% of GFP⁺ cells: 0.74 vs. 26.4). In these analyses, expression of CD107a and CCL4 on the effectors again correlates with the inhibition of HIV-1 replication. Currently, as hESC and iPSC-derived NK cells express the Fc receptor CD16, we are using anti-envelope protein antibodies against HIV-1 infected primary human CD4 T cells to determine if antibody-dependent cellular cytotoxicity provides another mechanism of lytic activity for these novel NK cells. Collectively, our results demonstrate that NK cells derived from hESCs and iPSCs provide an effective novel cellular immunotherapy for HIV-1 infection.

No relevant conflicts of interest to declare.