作为一种天然免疫细胞，自然杀伤（Natural Killer, NK）细胞在肿瘤免疫治疗中具有其独特的生物学优势。NK细胞以主要组织相容性复合体（major histocompatibility complex，MHC）非限制性的方式直接识别肿瘤相关抗原，迅速活化，对肿瘤细胞具有强大的细胞毒活性。因此同种异体NK细胞不会引起移植物抗宿主反应，而且由于固有免疫属性，同种异体NK细胞在体内也不会产生细胞因子释放综合征和神经毒性等安全性隐患。这些特性为开发基于NK细胞的同种异体通用型肿瘤免疫治疗产品奠定了基础。

当前用于肿瘤免疫治疗的NK细胞疗法主要有四大细胞来源，分别是外周血（peripheral blood，PB）来源的NK细胞、脐带血（umbilical cord blood, UCB）来源的NK细胞，NK-92等永生化NK细胞系，以及诱导多能干细胞（Induced Pluripotent Stem Cells，iPSC）来源的NK细胞。但PB-NK细胞的供体异质性、UCB-NK细胞的低细胞毒性、永生化细胞系输注前需进行辐照导致体内持久性降低，这些问题限制了这几种来源的NK细胞的临床应用。而研究证实，诱导多能干细胞来源的NK（iPSC derived NK，iNK）细胞不仅能完整表达CD56、NKG2D等天然NK细胞的关键效应分子，对多种血液肿瘤和实体瘤具有显著的细胞毒活性，而且能克服原代NK细胞基因编辑效率低的瓶颈，能通过多重基因编辑提升NK细胞的抗肿瘤活性，并在细胞产品均一性、不受供者限制方面展现出其独特优势。这些特性使得基因编辑的iNK细胞成为通用型细胞治疗产品的理想候选。

作为NK细胞表达的抑制性受体，NKG2A可通过与肿瘤细胞表面白细胞抗原E（Human Leukocyte Antigen-E，HLA-E）结合，可显著削弱NK细胞的肿瘤杀伤活性。临床前研究表明，使用单克隆抗体阻断NKG2A-HLA-E信号轴可恢复NK细胞约40-60%的肿瘤杀伤效率，说明靶向NKG2A-HLA-E免疫检查点轴的基因修饰策略能显著增强NK细胞的抗肿瘤活性。

基于上述研究背景，本研究拟建立稳定的体外iNK细胞诱导分化、扩增的体系，比较iNK细胞与原代PB-NK细胞的表面标志和杀伤活性的差异；并在此基础上，利用CRISPR Cas9技术对iPSC细胞中的NKG2A进行敲除，挑选单克隆，将其分化成NKG2A KO-iNK细胞，并在体外验证其抗肿瘤活性。

首先我们建立了体外iNK细胞诱导分化、扩增的3D培养体系。在第一阶段通过添加骨形态发生蛋白4（Bone Morphogenetic Proteins 4, BMP4）、血管内皮生长因子（vascular endothelial growth factor, VEGF)、成纤维细胞生长因子（fibroblast growth factor, FGF)、干细胞因子（stem cell factor, SCF）等细胞因子诱导iPSC向造血祖细胞（Hematopoietic progenitor cells，HPC）分化，在第8天利用流式细胞术检测HPC表面标志物CD34、CD43的表达情况。结果显示CD34⁺细胞在30%以上。第二阶段通过添加白细胞介素（interleukin , IL）-3、IL-7、IL-15、SCF、FMS 样酪氨酸激酶3（Fms like tyrosine kinase 3, FLT3L）等细胞因子，诱导HPC向NK细胞的分化，并在第28天利用流式细胞术检测NK细胞的表面标志物CD56和CD16的表达情况。结果显示得到的iNK细胞CD56⁺比例为84%，CD16⁺细胞比例为8%。接下来在体外用滋养层细胞对iNK细胞进行12天的扩增后，利用乳酸脱氢酶释放法对其进行体外杀伤活性检测，结果显示iNK细胞的体外杀伤K562细胞的效率在80%左右，与PB-NK细胞无显著差异。除此之外，我们还检测了iNK细胞和PB-NK细胞表面NKG2A表达量的变化，结果显示iNK细胞和PB-NK细胞的NKG2A表达在体外扩增阶段均呈现先下降后上升，在杀伤过程中基本无变化。

接下来我们利用CRISPR Cas9技术对iPSC细胞中的NKG2A进行敲除，并通过有限稀释法筛选单克隆iPSC，测序结果表明我们成功得到了单克隆NKG2A KO iPSC。然后我们利用前面建立的iNK分化体系，将NKG2A KO iPSC和WT iPSC在体外诱导分化成iNK细胞，其中两组细胞CD3^-CD56⁺均在90%左右，CD3^-CD16⁺均在20%左右。Western blot结果证明NKG2A KO-iNK细胞中的NKG2A被成功敲除。NKG2A KO-iNK细胞在体外能达到平均5倍的扩增，扩增后CD56和CD16的阳性率均在80%以上，与WT-iNK细胞无显著差异。流式细胞术结果显示NKG2A KO-iNK细胞在NK细胞激活受体NKG2D和天然毒性受体NKp30、NKp44和NKp46的表达上，与WT-iNK细胞没有显著差异。

最后我们对NKG2A KO-iNK细胞的体外杀伤活性进行检测。我们首先检测了多种肿瘤细胞系HLA-E表达情况，然后选取了HLA-E表达量不同的三种肿瘤细胞系作为靶细胞。乳酸脱氢酶释放法结果显示，相较于WT-iNK细胞，NKG2A KO-iNK细胞对高表达HLA-E的Nalm6细胞的杀伤活性显著增强，而对低表达或不表达HLA-E的NCL-H929和HepG2的体外杀伤活性无显著差异。进一步我们用IFN-γ处理Nalm6细胞，然后分别检测NKG2A KO-iNK细胞对IFN-γ处理前后的Nalm6细胞的体外杀伤活性变化。Western blot结果显示经IFN-γ处理的Nalm6细胞HLA-E表达显著上调。NKG2A KO-iNK细胞对HLA-E上调的Nalm6细胞的细胞毒活性显著增强，而WT-iNK细胞对是否经过IFN-γ处理的Nalm6细胞的细胞毒活性未显现出差异。这些结果表明NKG2A KO-iNK对HLA-E高表达的肿瘤细胞具有显著增强的细胞毒活性。

综上所述，本研究建立了一种阻断NKG2A-HLA-E负调控通路的基因工程NKG2A KO-iNK细胞，它能够对高表达HLA-E的肿瘤细胞在体外显示出明显增强的细胞毒活性。这些结果提示其作为一种可行的协同增效的策略，NKG2A KO-iNK细胞为肿瘤的NK细胞过继免疫治疗提供了一种新的思路。

As innate immune cells, Natural Killer (NK) cells exhibit distinct biological advantages in tumor immunotherapy. NK cells directly recognize tumor-associated antigens in a major histocompatibility complex (MHC)-unrestricted manner, rapidly activate, and exert potent cytotoxic activity against tumor cells. Consequently, allogeneic NK cells do not induce graft-versus-host disease (GVHD) and, due to their innate immune properties, carry minimal risks of cytokine release syndrome (CRS) or neurotoxicity. These characteristics establish a foundation for developing allogeneic, universal NK cell-based immunotherapies for tumors.

Current NK cell therapies primarily rely on four sources: peripheral blood (PB)-derived NK cells, umbilical cord blood (UCB)-derived NK cells, immortalized NK cell lines (e.g., NK-92), and induced pluripotent stem cell (iPSC)-derived NK cells. However, clinical applications of these sources are constrained by donor heterogeneity in PB-NK cells, low cytotoxicity of UCB-NK cells, and reduced in vivo persistence of irradiated immortalized cell lines. In contrast, iPSC-derived NK cells (iNK cells) not only fully express key effector molecules of natural NK cells (e.g., CD56, NKG2D) and demonstrate robust cytotoxicity against hematologic and solid tumors but also overcome the low gene-editing efficiency of primary NK cells. Through multiplex genetic modifications, iNK cells exhibit enhanced anti-tumor activity while offering advantages in product homogeneity and donor independence. These properties position genetically engineered iNK cells as ideal candidates for universal "off-the-shelf" cell therapies.

NKG2A, an inhibitory receptor expressed on NK cells, significantly impairs tumor-killing activity by binding to human leukocyte antigen-E (HLA-E) on tumor cells. Preclinical studies show that monoclonal antibody blockade of the NKG2A-HLA-E axis restores approximately 40–60% of NK cell cytotoxicity, highlighting that genetic disruption of this immune checkpoint can markedly enhance NK cell efficacy.

Building on this rationale, this study aimed to establish a stable in vitro system for iNK cell differentiation and expansion, compare surface markers and cytotoxicity between iNK and primary PB-NK cells, and employ CRISPR-Cas9 to generate NKG2A knockout (KO) iPSCs. Single clones were differentiated into NKG2A KO-iNK cells, followed by validation of their anti-tumor activity in vitro.

 A 3D culture system was developed for iNK cell differentiation. In Phase 1, iPSCs were induced into hematopoietic progenitor cells (HPCs) using cytokines including bone morphogenetic protein 4 (BMP4), vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), and stem cell factor (SCF). Flow cytometry on day 8 revealed CD34+ cells exceeding 30%. In Phase 2, HPCs were differentiated into NK cells with interleukin-3 (IL-3), IL-7, IL-15, SCF, and FMS-like tyrosine kinase 3 ligand (FLT3L). By day 28, flow cytometry confirmed iNK cells with 84% CD3-CD56+ and 8% CD3-CD16+ populations. Following 12-day expansion on feeder cells, lactate dehydrogenase (LDH) release assays demonstrated ~80% cytotoxicity against K562 cells, comparable to PB-NK cells. NKG2A expression in both iNK and PB-NK cells decreased transiently during expansion but rebounded afterward, remaining stable during cytotoxicity assays.

CRISPR-Cas9-mediated NKG2A KO iPSC clones were validated by sequencing and differentiated into iNK cells. Both KO and wild-type (WT) iNK cells exhibited ~90% CD3-CD56+ and ~20% CD3-CD16+ populations. Western blot confirmed NKG2A knockout in KO-iNK cells. NKG2A KO-iNK cells achieved 5-fold expansion with >80% CD56+/CD16+ rates, similar to WT-iNK cells. Flow cytometry revealed no differences in activating receptors (NKG2D, NKp30, NKp44, NKp46) between KO and WT iNK cells.

In cytotoxicity assays against HLA-E-high Nalm6, HLA-E-low NCL-H929, and HLA-E-negative HepG2 cells, NKG2A KO-iNK cells showed significantly enhanced killing of Nalm6 cells compared to WT-iNK cells, while activity against NCL-H929 and HepG2 remained unchanged. IFN-γ treatment upregulated HLA-E in Nalm6 cells, further enhancing KO-iNK cytotoxicity, whereas WT-iNK activity was unaffected.
  This study established genetically engineered NKG2A KO-iNK cells that disrupt the NKG2A-HLA-E inhibitory axis, selectively enhancing cytotoxicity against HLA-E-high tumors in vitro. These findings propose NKG2A KO-iNK cells as a novel synergistic strategy for adoptive NK cell immunotherapy.