Figure provided by Dr. Strong.
NKG2D receptors are expressed on CD8 T cells and NK cells and stimulate anti-tumor immunity upon interaction with their ligands on the surface of cancer cells. However, ligand expression has been associated with negative disease outcome challenging the anti-tumoral role of NKG2D receptor. Additionally, previous studies (Benitez, 2011 and Cai, 2014) from the Spies group at the Fred Hutch (Clinical Research Division) have shown that the NKG2D receptor could also be detected on certain cancer cells where it activates oncogenic signaling pathways thus promoting core components of tumorigenesis. “Sharing of signaling receptors between cancer cells and lymphocytes is not that uncommon, examples include growth factor receptors or chemokine receptors; however, NKG2D is unique in that it is, at least according to current understanding, the only bona fide activating T cell and NK cell receptor that becomes coopted by NKG2D ligand positive cancer cells for autonomous stimulation of oncogenic signaling”, explained Dr. Groh.
Using ovarian cancer as a model setting, the new study by the Spies group, in collaboration with Dr. Roland Strong (Basic Sciences Division) and Dr. Charles Drescher (Public Health Sciences Division), established the contribution of NKG2D to cancer development. The results of this study were published in Neoplasia journal.
No evidence was available regarding the role of NKG2D in tumorigenesis versus tumor evasion. Based on their previous observations, the authors hypothesized that the NKG2D receptor may promote cancer cell plasticity and demonstrated that disease outcomes were worse in ovarian cancer patients with high proportions of NKG2D+ cancer cells, supporting their hypothesis. Looking more specifically into freshly isolated cancer cells ex vivo, NKG2D expression correlated with expression of EpCAM and CD44 markers as well as with aldehyde dehydrogenase isoform 1 (ALDH1) enzymatic activity, indicators of high cancer cell plasticity. Additionally, NKG2D positive but not NKG2D negative cancer cells developed tumor spheres in vitro, an indicator of tumor formation potential. NKG2D inhibition by siRNA abrogated this potential. These data demonstrated the potential of NKG2D to support tumorigenesis in vitro.
Researchers then established the role of NKG2D to support tumor formation in vivo in immunodeficient mice. To this end, NKG2D+ ovarian cancer cells or NKG2D- cells as controls, isolated ex vivo from high-grade serous epithelial ovarian cancer were injected subcutaneously. Mice were monitored for tumor development. Tumor incidence was higher with NKG2D+ cells relative to NKG2D- cells. These observations were confirmed using MDAH-2774 cells, a cancer cell line known as deprived from endogenous NKG2D expression which allowed tumor development when modified to overexpress NKG2D by opposition to parental MDAH-2774 cells, even when fewer cells were injected. Altogether these data demonstrate a role for NKG2D in tumor initiation in this in vivo model.
Eight NKG2D ligands have been identified at the surface of tumors whose expression is heterogeneous and unpredictable, challenging ligands targeting. To address this, Dr. Strong’s group developed a pan-ligand named NKG2Dscd multimer to neutralize the different types of ligands. Indeed, “most previous treatments targeting the NKG2D axis focused on biologics that bound to NKG2D, rather than our reagent, which binds its ligands. The problem with the former approach is that you run the risk of activating through NKG2D, the opposite of the desired effect”, said Dr. Strong. By blocking the binding of these ligands to the NKG2D receptor, the authors observed a decrease in tumor development from NKG2D positive but not NKG2D negative ex vivo isolated cancer cells confirming the implication of NKG2D specifically in cancer development.
Dr. Spies explained that “in the tumors that become clinically ‘visible’ and would thus potentially be treated, lymphocyte NKG2D (both on tumor infiltrating and peripheral T and NK cells) is already impaired due to ligandmediated down regulation; therefore, if anything, in addition to preventing cancer cell NKG2D functions, the heptamer pan-ligand may have further beneficial effects via neutraliziation of ligands followed by rebounding of lymphocyte NKG2D”. As such, “the safety of NKG2Dscd pan-ligand is comparable or better than other biologics at this stage of development potentially better, as we have carefully designed to minimizing antireagent humoral responses. We’re currently at the stage of evaluating preclinical safety, with experiments ongoing. This approach really is the first potentially effective means of specifically downregulating NKG2Dmediated events, with multiple potential applications beyond cancer. From a personal perspective, it is really quite exciting, as a basic scientist, to have the opportunity to move so quickly to working in a clinical context the Hutch is a really amazing place in that regard”, added Dr. Strong. “In addition to pursuing our collaboration with the Strong lab we are most interested in trying to identify the causes and mechanisms that lead to the unusual expression of a lymphocyte receptor in subsets of cancer cells”, concluded Dr. Spies.
This study was funded by a Fred Hutchinson Cancer Research Center VIDD Faculty Initiative Award 2014 and by the National Cancer Institute (NIH).
X Cai, A Caballero-Benitez, MM Gewe, IC Jenkins, CW Drescher, RK Strong, T Spies, V Groh. 2017. Control of Tumor Initiation by NKG2D Naturally Expressed on Ovarian Cancer Cells. Neoplasia. 19 (6), 471-482.
1- A Caballero-Benitez, Z Dai, HH Mann, RS Reeves, DH Margineantu, TA Gooley, V. Groh, T. Spies. 2011. Expression, signaling proficiency, and stimulatory function of the NKG2D lymphocyte receptor in human cancer cells. Proceedings of the National Academy of Sciences. 108 (10), 4081–4086.
2- X Cai, Z Dai, RS Reeves, A Caballero-Benitez, KL Duran, JJ Delrow, PL Porter, T Spies, V Groh. 2014. Autonomous stimulation of cancer cell plasticity by the human NKG2D lymphocyte receptor coexpressed with its ligands on cancer cells. PloS one. 9 (10).