Release of natural killer cytotoxic factor (NKCF) from canine natural killer (NK) cells stimulated with cytoplasmic membrane of target cells

Characterization and Potential Applications of Dog Natural Killer Cells in Cancer Immunotherapy

Natural killer (NK) cells of the innate immune system are a key focus of research within the field of immuno-oncology based on their ability to recognize and eliminate malignant cells without prior sensitization or priming. However, barriers have arisen in the effective translation of NK cells to the clinic, in part because of critical species differences between mice and humans. Companion animals, especially dogs, are valuable species for overcoming many of these barriers, as dogs develop spontaneous tumors in the setting of an intact immune system, and the genetic and epigenetic factors that underlie oncogenesis appear to be similar between dogs and humans. Here, we summarize the current state of knowledge for dog NK cells, including cell surface marker phenotype, key NK genes and genetic regulation, similarities and differences of dog NK cells to other mammals, especially human and mouse, expression of canonical inhibitory and activating receptors, ex vivo expansion techniques, and current and future clinical applications. While dog NK cells are not as well described as those in humans and mice, the knowledge of the field is increasing and clinical applications in dogs can potentially advance the field of human NK biology and therapy. Better characterization is needed to truly understand the similarities and differences of dog NK cells with mouse and human. This will allow for the canine model to speed clinical translation of NK immunotherapy studies and overcome key barriers in the optimization of NK cancer immunotherapy, including trafficking, longevity, and maximal in vivo support.

Isolation and characterization of canine natural killer cells

NK cells are non-T, non-B lymphocytes that kill target cells without previous activation. The immunophenotype and function of these cells in humans and mice are well defined, but canine NK cells remain incompletely characterized. Our objectives were to isolate and culture canine peripheral blood NK cells, and to define their immunophenotype and killing capability. PBMC were obtained from healthy dogs and T cells were depleted by immunomagnetic separation. The residual cells were cultured in media supplemented with IL-2, IL-15 or both, or with mouse embryonic liver (EL) feeder cells. Non-T, non-B lymphocytes survived and expanded in these cultures. IL-2 was necessary and sufficient for survival; the addition of IL-15 was necessary for expansion, but IL-15 alone did not support survival. Culture with EL cells and IL-2 also fostered survival and expansion. The non-T, non-B lymphocytes uniformly expressed CD45, MHC I, and showed significant cytotoxic activity against CTAC targets. Expression of MHC II, CD11/18 was restricted to subsets of these cells. The data show that cells meeting the criteria for NK cells in other species, i.e., non-T, non-B lymphocytes with cytotoxic activity, can be expanded from canine PBMC by T-cell depletion and culture with cytokines or feeder cells.

Induction of apoptosis by canine natural killer cells

Besides a secretory pathway of canine natural killer (NK) cells, which results in necrosis of the target cell, a second pathway was demonstrated, which results in apoptosis of the target cell. Comparing the Chromium Release Assay (CRA) and the Rose Bengal Assay (RBA) for quantification of in vitro canine NK cell activity, a constant 10% higher NK cell activity was found in the RBA compared with the CRA. To find out the mechanism responsible for the different results of both tests, morphological studies of in vitro canine NK cell activity against epithelial and mesenchymal allogenic target cell lines were performed. Most target cells were undergoing necrosis as a result of NK cell killing, which was evidenced by transmission electron microscopy. However, besides necrotic target cells, shrunken target cells with dense cytoplasm, fragmented nuclei and disruption into membrane-bound bodies were detected, which are known as signs of apoptosis. Additionally, using the terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling (TUNEL) method, 13-23% of target cells presented a positive staining, indicative of apoptosis. These findings give evidence for the ability of canine NK cells to kill their target cells via two different pathways, which results either in apoptosis or necrosis.

Active-oxygen involvement in canine NK-mediated cytotoxicity

We examined the relationship between natural killer (NK) cell-mediated cytotoxicity, the produced active-oxygen and cytotoxic factor (CF) release in co-culturing canine NK cells with tumor cells (CL-1 target cells). In co-culturing, the adding of n-propyl gallate (active-oxide scavenger) removed the produced active-oxygen, which inhibited NK cell-mediated cytotoxicity and the CF release. Moreover, adding of this agent inhibited the tyrosine phosphorylation of NK intracellular protein which observed in co-culturing. Therefore, the active-oxygen produced from canine NK cells are thought to relate the signal transduction in NK-mediated cytotoxicity.

Morphological observation of canine natural killer cells mediated cytotoxicity

The cytotoxic effects of canine NK cells on CL-1 target cells were examined by scanning electron microscopy (SEM). NK cell mediated cytotoxicity on CL-1 target cells was detected by 51Cr release assay. SEM showed that a canine NK cell extended projections to the CL-1 target cell. Furthermore, the surface of CL-1 target cells changed a mesh-like structure. Therefore, the cytotoxic effects of canine NK cells on CL-1 target cells were morphologically demonstrated.

Characterization of natural killer cytotoxic factor (NKCF) from canine NK cells

We investigated the presence of canine natural killer cytotoxic factor (NKCF). Canine natural killer (NK) cell-mediated cytotoxicity measured by 51chromium (51Cr) release assay was found to be highest in the T-cell population, which was fractionated into the 35-40% Percoll fraction by discontinuous gradient centrifugation. The cytotoxicity of NKCF in the culture supernatant showed a similar tendency to NK activity. Release of NKCF was rapid after contact with target cells, and reached a plateau in 60 min. The cytotoxicity of NKCF could be detected within at least 15 min in coculture with CL-1 target cells, reaching a plateau in 60 min. We also characterized canine NKCF and found it to be a protein, which was stable against both heat and cold treatment. These findings suggest that canine NK cells release NKCF immediately after recognition and binding to the target cell, and that NKCF plays an important role in canine NK-mediated cytotoxicity.