Natural killer (NK) cells are a group of innate lymphocytes, comprising 5–20% of human peripheral blood lymphoid cells, and serve as an important constituent in host protection against viral infections and immune malignancy surveillance [1
]. These cells can recognize and kill aberrant cells without preliminary antigen sensitization. NK activation and target recognition are determined by the inputs from a complex signal-integrating network that simultaneously receives and processes signals from two types of NK receptors, classified as activating and inhibitory, after engagement of their putative ligands on the target cell. The activating receptors include the family of natural cytotoxicity receptors (NCRs), DNAX accessory molecule-1 (DNAM-1), NK group 2 member D (NKG2D), and 2B4 [3
]. Inhibitory receptors belong to either the superfamily of immunoglobulin or the c-type lectin receptors that mainly recognize major histocompatibility complex (MHC)-I molecules. NK-mediated immune function relies heavily on the dynamic signal balance from the inhibitory and activating receptors [4
]. Appropriate interactions between NK inhibitory receptors and MHC-I molecules turn NK cells “off” and ensure self-tolerance. However, diminution or absence of cell surface MHC-I molecules, resulting from viral infections or tumor transformations, leads to a reduced overall inhibitory signal, thereby activating NK cells to eliminate those missing-self targets [8
]. Specific killing of infected and transformed cells depends on several discrete steps that lead to polarization and exocytosis of lytic granules towards the target cells [10
]. NK-target interaction, also known as conjugation formation, is the first key step for NK cell cytotoxicity [11
]. Engagement of NK cell surface integrin lymphocyte function-associated antigen 1 (LFA-1), a heterodimer of CD11a and CD18, by its ligand intercellular adhesion molecule 1 (ICAM-1) on target cells establishes a stable adhesion of NK cells to their target cells and is sufficient to induce the polarization of lytic granules in resting NK cells [12
]. Then, NK-cell cytotoxicity is achieved by degranulation of these accumulated granules toward the direction of the engaged target.
Tumor/leukemia cells are known for their ability to escape host immunity. Malignancy can even develop in a host with a fully functional immune system, which is known as the cancer immunoediting concept [15
]. During tumorigenesis and leukemogenesis, mutations and epigenetic changes not only affect the protein composition, but also induce profound changes in glycosylation [16
]. One of the hallmarks of such altered glycosylation is the upregulation of terminal sialic acids on secreted or cell surface glycoconjugates. Upregulation of cell surface sialylation has been implicated in complement-mediated tumor cell killing, enhancement of tumor-related inflammation, and especially cancer progression where upregulation of selectin ligands influences metastasis colonization [19
]. Moreover, recent experimental findings suggest that interactions between hypersialylated ligands on tumor cells and NK cell surface immunomodulatory sialic acid-binding immunoglobulin-like lectins (Siglecs) contribute to cancer immunosurveillance and cancer-associated immune suppression [21
], implying the involvement of the NK cell surface Siglecs in cytotoxicity. Restrictedly expressed only on certain cell types in humans, there are 15 functional members in the Siglec family featuring a N-terminal V-set Ig-like domain, which characteristically mediates carbohydrate recognition for signal transduction, followed by a varying number of C2-set Ig-like domains [25
]. For most Siglecs, recognition of their ligands can initiate cellular signaling through immunoreceptor tyrosine-based inhibitory motifs (ITIMs) at their cytoplasmic tails. By function, Siglecs with ITIMs are able to recruit phosphatase, and are therefore referred to as inhibitory-type Siglecs [27
]. Siglec-7 (also termed p75 adhesion inhibitory receptor molecule 1, p75/AIRM-1, CDw328) is expressed on most human peripheral blood NK cells, monocytes, basophils, and mast cells [29
]. Siglec-7 contains three extracellular immunoglobulin-like domains, a transmembrane region, and a cytoplasmic tail with two tyrosine residues embodied in ITIM-like motifs [30
]. Of the other Siglecs expressed by NK cells, Siglec-9 exists in both T cells and basophil cells, and is expressed by 50% of NK cells [33
]. Revealed by recent studies, role of Siglec-7 is emerging as an important regulator in the immune response. In redirected killing assay, Siglec-7 functions as a NK inhibitory receptor for its ability to inhibit human NK cell cytotoxicity against murine P815 target cells [29
]. Also, sialic acid moieties of tumor cell surfaces can be recognized by Siglec-7 to negatively modulate NK cell killing activity, likely leading to escape from NK cell cytotoxicity [35
]. Interestingly, increased expression of sialylated ligands observed on tumor cell surface inhibits NK cell activation through the engagement with Siglec-7 [21
]. On the other hand, decreased expression of NK Siglec-7 has been consistently observed during cancer treatment and such low levels last until the end of treatment [36
]. These results suggest loss of Siglec-7 signaling favors NK cell activation. With the ability to eliminate transformed and tumor/leukemia cells, the NK cells have been considered as an alternative therapy to the conventional chemo- or radioactive treatment. Among NK cell-based therapies, a developed NK cell line, NK-92, has been applied in clinical trials [38
]. However, the sustainability of the infused NK-92 and intensity of cytotoxicity in the treatment protocol still have room for refinement, such as by manipulating cytotoxic activity through enhancing the activating receptor-mediated signals, or eliminating/reducing the inhibitory signals to prolong or enhance its effect.
In this study, we initially found that the cytotoxicity of NK-92MI cells declined over the course of long-term in vitro culture. We hypothesized that there is an important factor or more involved in regulating such NK function along the culture progress and we discovered a correlation between the presence of cell surface Siglec-7 and NK-92MI cytotoxicity. Hence, we established a Siglec-7neg NK cell model (NK-92MI-S7N), derived from NK-92MI, to investigate Siglec-7 effect on NK cytotoxicity, especially on eliminating leukemia cells. Our findings demonstrated that the NK-92MI cells without the presence of Siglec-7 shared very similar phenotypes with parental NK-92MI in cell morphology, expressions of NK activating and inhibitory receptors, and cytotoxic-related proteins and cytokines. However, they possessed higher ICAM-1 binding avidity and affinity as well as high and sustainable cytotoxic activity that was able to eliminate a NK-92MI-resistant cell. With the development of this NK model, there may be an applicational potential for some NK-92-based immunotherapies.