Enhanced IL-15-mediated NK cell activation and proliferation by an ADAM17 function-blocking antibody involves CD16A, CD137, and accessory cells.

Background: NK cells are being extensively studied as a cell therapy for cancer. Their effector functions are induced by the recognition of ligands on tumor cells and by various cytokines. IL-15 is broadly used to stimulate endogenous and adoptively transferred NK cells in cancer patients. These stimuli activate the membrane protease ADAM17, which then cleaves assorted receptors on the surface of NK cells as a negative feedback loop to limit their activation and function. We have shown that ADAM17 inhibition can enhance IL-15-mediated NK cell proliferation in vitro and in vivo . In this study, we investigated the underlying mechanism of this process. Methods: PBMCs or enriched NK cells from human peripheral blood, either unlabeled or labeled with a cell proliferation dye, were cultured for up to 7 days in the presence of rhIL-15 +/- an ADAM17 function-blocking antibody. Different versions of the antibody were generated; Medi-1 (IgG1), Medi-4 (IgG4), Medi-PGLALA, Medi-F(ab') 2 , and TAB16 (anti-ADAM17 and anti-CD16 bispecific) to modulate CD16A engagement on NK cells. Flow cytometry was used to assess NK cell proliferation and phenotypic markers, immunoblotting to examine CD16A signaling, and IncuCyte-based live cell imaging to measure NK cell anti-tumor activity. Results: The ADAM17 function-blocking mAb Medi-1 markedly increased initial NK cell activation by IL-15. Using different engineered versions of the antibody revealed that the activating Fcγ receptor CD16A, a well-described ADAM17 substrate, was critical for enhancing IL-15 stimulation. Hence, Medi-1 bound to ADAM17 on NK cells can be engaged by CD16A and block its shedding, inducing and prolonging its signaling. This process did not promote evident NK cell fratricide, phagocytosis, or dysfunction. Synergistic activity by Medi-1 and IL-15 enhanced the upregulation of CD137 on CD16A + NK cells and augmented their proliferation in the presence of PBMC accessory cells. Conclusions: Our data reveal for the first time that CD16A and CD137 underpin Medi-1 enhancement of IL-15-driven NK cell activation and proliferation, respectively. The use of Medi-1 represents a novel strategy to enhance IL-15-driven NK cell proliferation, and it may be of therapeutic importance by increasing the anti-tumor activity of NK cells in cancer patients. What is already known on this topic: NK cell therapies are being broadly investigated to treat cancer. NK cell stimulation by IL-15 prolongs their survival in cancer patients. Various stimuli including IL-15 activate ADAM17 in NK cells, a membrane protease that regulates the cell surface density of various receptors as a negative feedback mechanism. What this study adds: Treating NK cells with the ADAM17 function-blocking mAb Medi-1 markedly enhanced their activation and proliferation. Our study reveals that the Fc and Fab regions of Medi-1 function synergistically with IL-15 in NK cell activation. Medi-1 treatment augments the upregulation of CD137 by NK cells, which enhances their proliferation in the presence of PBMC accessory cells. How this study might affect research practice or policy: Our study is of translational importance as Medi-1 treatment in combination with IL-15 could potentially augment the proliferation and function of endogenous or adoptively transferred NK cells in cancer patients.

iPSC-derived natural killer cells expressing the FcγR fusion CD64/16A can be armed with antibodies for multitumor antigen targeting.

BACKGROUND: Antibody therapies can direct natural killer (NK) cells to tumor cells, tumor-associated cells, and suppressive immune cells to mediate antibody-dependent cell-mediated cytotoxicity (ADCC). This antigen-specific effector function of human NK cells is mediated by the IgG Fc receptor CD16A (FcγRIIIA). Preclinical and clinical studies indicate that increasing the binding affinity and avidity of CD16A for antibodies improves the therapeutic potential of ADCC. CD64 (FcγRI), expressed by myeloid cells but not NK cells, is the only high affinity IgG Fc receptor and is uniquely capable of stably binding to free monomeric IgG as a physiological function. We have reported on the generation of the FcγR fusion CD64/16A, consisting of the extracellular region of CD64 and the transmembrane and cytoplasmic regions from CD16A, retaining its signaling and cellular activity. Here, we generated induced pluripotent stem cell (iPSC)-derived NK (iNK) cells expressing CD64/16A as a potential adoptive NK cell therapy for increased ADCC potency. METHODS: iPSCs were engineered to express CD64/16A as well as an interleukin (IL)-15/IL-15Rα fusion (IL-15RF) protein and differentiated into iNK cells. iNK cells and peripheral blood NK cells were expanded using irradiated K562-mbIL21-41BBL feeder cells and examined. NK cells, ovarian tumor cell lines, and therapeutic monoclonal antibodies were used to assess ADCC in vitro, performed by a DELFIA EuTDA assay or in real-time by IncuCyte assays, and in vivo. For the latter, we developed a xenograft mouse model with high circulating levels of human IgG for more physiological relevance. RESULTS: We demonstrate that (1) iNK-CD64/16A cells after expansion or thaw from cryopreservation can be coupled to therapeutic antibodies, creating armed iNK cells; (2) antibody-armed iNK-CD64/16A cells can be redirected by added antibodies to target new tumor antigens, highlighting additional potential of these cells; (3) cytokine-autonomous activity by iNK-CD64/16A cells engineered to express IL-15RF; and that (4) antibody-armed iNK-CD64/16A cells thawed from cryopreservation are capable of sustained and robust ADCC in vitro and in vivo, as determined by using a modified tumor xenograft model with high levels of competing human IgG. CONCLUSIONS: iNK cells expressing CD64/16A provide an off-the-shelf multiantigen targeting platform to address tumor heterogeneity and mitigate antigen escape.

HLA-G1+ Expression in GGTA1KO Pigs Suppresses Human and Monkey Anti-Pig T, B and NK Cell Responses.

The human leukocyte antigen G1 (HLA-G1), a non-classical class I major histocompatibility complex (MHC-I) protein, is a potent immunomodulatory molecule at the maternal/fetal interface and other environments to regulate the cellular immune response. We created GGTA1-/HLAG1+ pigs to explore their use as organ and cell donors that may extend xenograft survival and function in both preclinical nonhuman primate (NHP) models and future clinical trials. In the present study, HLA-G1 was expressed from the porcine ROSA26 locus by homology directed repair (HDR) mediated knock-in (KI) with simultaneous deletion of α-1-3-galactotransferase gene (GGTA1; GTKO) using the clustered regularly interspersed palindromic repeats (CRISPR)/CRISPR associated protein 9 (Cas9) (CRISPR/Cas9) gene-editing system. GTKO/HLAG1+ pigs showing immune inhibitory functions were generated through somatic cell nuclear transfer (SCNT). The presence of HLA-G1 at the ROSA26 locus and the deletion of GGTA1 were confirmed by next generation sequencing (NGS) and Sanger's sequencing. Fibroblasts from piglets, biopsies from transplantable organs, and islets were positive for HLA-G1 expression by confocal microscopy, flow cytometry, or q-PCR. The expression of cell surface HLA-G1 molecule associated with endogenous ß2-microglobulin (ß2m) was confirmed by staining genetically engineered cells with fluorescently labeled recombinant ILT2 protein. Fibroblasts obtained from GTKO/HLAG1+ pigs were shown to modulate the immune response by lowering IFN-γ production by T cells and proliferation of CD4+ and CD8+ T cells, B cells and natural killer (NK) cells, as well as by augmenting phosphorylation of Src homology region 2 domain-containing phosphatase-2 (SHP-2), which plays a central role in immune suppression. Islets isolated from GTKO/HLA-G1+ genetically engineered pigs and transplanted into streptozotocin-diabetic nude mice restored normoglycemia, suggesting that the expression of HLA-G1 did not interfere with their ability to reverse diabetes. The findings presented here suggest that the HLA-G1+ transgene can be stably expressed from the ROSA26 locus of non-fetal maternal tissue at the cell surface. By providing an immunomodulatory signal, expression of HLA-G1+ may extend survival of porcine pancreatic islet and organ xenografts.

3'UTR enhances hCD47 cell surface expression, self-signal function, and reduces ER stress in porcine fibroblasts.

INTRODUCTION: Macrophages contribute to xenograft rejection by direct cytotoxicity and by amplifying T cell-mediated immune responses. It has been shown that transgenic expression of hCD47 protects porcine cells from human macrophages by restoring the CD47-SIRPα self-recognition signal. It has also been reported that the long 3' untranslated region (3'UTR) of the hCD47 gene, which is missing from constructs previously used to make hCD47 transgenic pigs, is critical for efficient cell surface expression in human cells. The aim of this study was to investigate the impact of a modified form of the 3'UTR on the expression, localization, and function of hCD47 in transfected porcine cells. METHODS: hCD47 constructs with and without the modified 3'UTR were knocked into the GGTA1 locus in porcine fetal fibroblasts using CRISPR. Flow cytometry of the transfected cells was used to analyze hCD47 localization. Endoplasmic reticulum (ER), mitochondrial, and oxidative stress were examined by gene expression analysis and confocal microscopy. Phagocytosis of transfected cells by human macrophages was measured by flow cytometry, and stimulation of human/non-human (NHP) primate lymphocytes by the cells was examined using a PBMCs proliferation assay. RESULTS: Cells transfected with the construct lacking the 3'UTR (hCD47(3'UTR-)) exhibited predominantly intracellular expression of hCD47, and showed evidence of ER stress, dysregulated mitochondrial biogenesis, oxidative stress, and autophagy. Inclusion of the 3'UTR (hCD47(3'UTR+)) decreased intracellular expression of hCD47 by 36% and increased cell surface expression by 53%. This was associated with a significant reduction in cellular stress markers and a higher level of protection from phagocytosis by human macrophages. Furthermore, hCD47(3'UTR+) porcine cells stimulated significantly less proliferation of human/NHP T cells than hCD47(3'UTR-) cells. CONCLUSION: Our results suggest the potential benefits of using hCD47 constructs containing the 3'UTR to generate genetically engineered hCD47-expressing donor pigs.

Efficient production of GGTA1 knockout porcine embryos using a modified handmade cloning (HMC) method.

Handmade cloning is a zona-free nuclear transfer approach and an economical, efficient, and simple micromanipulation-free alternative to dolly based traditional cloning (TC). In this study, based on handmade cloning with minor modifications, an optimized bi-oocyte fusion (BOF) cloning method was established to produce GGTA1 KO porcine embryos using the CRISPR/Cas9 gene editing system. The GGTA1 gene is responsible for the generation of Gal epitopes on the surface of porcine cells, triggering hyperacute immune rejection in preclinical porcine-to-human xenotransplantation. The purpose of the present study is to establish an efficient protocol for activation of porcine oocyte cytoplast-fibroblast fused constructs developed to GGTA1 KO blastocysts by the zona-free bi-oocyte fusion cloning method. High percentages of cleavage (90 ±â€¯2.6%) and blastocyst rates (39 ±â€¯4.0%) were achieved upon treatment with demecolcine-assisted oocyte enucleation followed by 6 V alternating current for proper alignment and single-step fusion technique using a single direct current pulse of 1.0 kV/cm for 9 µs duration, compared to the double-step fusion method with combined chemical activation using thimerosal and dithiothreitol. Overall blastocyst rate was higher for oocyte enucleation by demecolcine (0.4 µg/ml) and 45 min incubation (42 ±â€¯1.5%) compared to without demecolcine incubation followed by complete chemical thimerosal/dithiothreitol activation (33 ±â€¯1.1%). The blastocyst rate (39 ±â€¯1.0%) was found to be significantly higher 1 h post-electrofusion, compared to at 0 and 4 h (28 ±â€¯1.5 and 6 ±â€¯1.5%, respectively). Blastocyst development rates for GGTA1 knockout embryos (38 ±â€¯1.76%) were comparable to those obtained with wild-type embryos (41.1 ±â€¯0.67%). In conclusion, we achieved high overall efficiency in production of GGTA1 KO blastocysts by modified HMC protocol.

Optimizing sgRNA length to improve target specificity and efficiency for the GGTA1 gene using the CRISPR/Cas9 gene editing system.

The CRISPR/Cas9 gene editing system has enhanced the development of genetically engineered animals for use in xenotransplantation. Potential limitations to the CRISPR/Cas9 system impacting the development of genetically engineered cells and animals include the creation of off-target mutations. We sought to develop a method to reduce the likelihood of off-target mutation while maintaining a high efficiency rate of desired genetic mutations for the GGTA1 gene. Extension of sgRNA length, responsible for recognition of the target DNA sequence for Cas9 cleavage, resulted in improved specificity for the GGTA1 gene and less off-target DNA cleavage. Three PAM sites were selected within exon 1 of the porcine GGTA1 gene and ten sgRNA of variable lengths were designed across these three sites. The sgRNA was tested against synthetic double stranded DNA templates replicating both the native GGTA1 DNA template and the two most likely off-target binding sites in the porcine genome. Cleavage ability for native and off-target DNA was determined by in vitro cleavage assays. Resulting cleavage products were analyzed to determine the cleavage efficiency of the Cas9/sgRNA complex. Extension of sgRNA length did not have a statistical impact on the specificity of the Cas9/sgRNA complex for PAM1 and PAM2 sites. At the PAM3 site, however, an observed increase in specificity for native versus off-target templates was seen with increased sgRNA length. In addition, distance between PAM site and the start codon had a significant impact on cleavage efficiency and target specificity, regardless of sgRNA length. Although the in vitro assays showed off-target cleavage, Sanger sequencing revealed that no off-target mutations were found in GGTA1 knockout cell lines or piglet. These results demonstrate an optimized method for improvement of the CRIPSR/Cas9 gene editing system by reducing the likelihood of damaging off-target mutations in GGTA1 knocked out cells destined for xenotransplant donor production.