Date of Award

Spring 5-15-2021

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Immunology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Natural killer (NK) cells are innate lymphoid cells (ILCs) that mediate anti-tumor and anti-viral immune responses. They do so via two primary effector functions: cytokine production and direct cytotoxicity. Unlike adaptive T and B lymphocytes, NK cells do not rearrange their DNA to express a predominant antigen-specific receptor, and instead express a variety of stochastically-expressed, germline DNA-encoded activating and inhibitory receptors whose signals integrate to govern their functional responses. What results is a diverse repertoire of NK cells capable of recognizing a variety of malignantly-transformed or virally-infected cells. Studies from several groups have established the anti-tumor potential of NK cells, most notably in the setting of allogeneic hematopoietic cell transfer (HCT) or adoptive NK cell therapy, spurring substantial interest in NK cell-based immunotherapies for cancer. While clinical trial results to date have highlighted the favorable safety profile of NK cells, clinical responses to NK cell therapy have been limited in both number and duration. Mechanisms of enhancing the anti-tumor potential of endogenous or adoptively-transferred NK cells are necessary.

It is well-established that NK cell function is tuned by immunomodulatory cytokines. Our group has previously demonstrated that short-term pre-activation of NK cells with the cytokines interleukin (IL)-12, 15, and 18 results in the generation of long-lived memory-like NK cells with robust anti-leukemic properties in vitro and in vivo. These memory-like NK cells display enhanced cytokine production and cytotoxicity compared to conventional NK cells and have proven to be safe in a phase I clinical trial (NCT01898793). While complete responses were observed in this study, not all patients responded to NK cell therapy. We hypothesized that the clinical efficacy of memory-like NK cells could be further enhanced by combining them with monoclonal antibody therapy. To lend evidence to support this hypothesis, we characterized the response profile of memory-like NK cells to FcgRIIIa (CD16)-based stimulation. We demonstrated that, indeed, both interferon (IFN)-g production and killing in response to CD16 ligation are enhanced in memory-like NK cells compared to controls.

In addition, studies of NK cell immunotherapy to date have focused on CD56dim NK cells; the subset that predominates in the peripheral blood and is classically associated with NK cell cytotoxicity. Thus, the anti-tumor potential of immunomodulatory CD56bright NK cells, which predominate in secondary lymphoid tissues and in number overall, has remained relatively unexplored. To this end, we investigated the impact of priming CD56bright NK cells with IL-15 on their anti-tumor function. We demonstrated that IL-15 priming markedly enhanced CD56bright NK cell degranulation, cytotoxicity, and cytokine production in vitro in response to tumor cell lines as well as primary myeloma and acute myeloid leukemia (AML) targets. Furthermore, IL-15 priming augmented the anti-tumor responses of cancer patient CD56bright NK cells to their autologous tumor cells. IL-15 priming of CD56bright NK cells also occurred in vivo, since treatment of multiple myeloma patients with the IL-15 super-agonist ALT-803 resulted in enhanced ex vivo responses of patient CD56bright NK cells to a myeloma cell line. Finally, IL-15-primed CD56bright NK cells were capable of controlling AML in a NOD-SCID-gc (NSG) mouse model. The mechanisms for IL-15 priming involved increased cytotoxic protein and activating receptor expression, integrin activation, and stimulation of the PI3K and MAPK pathways. These results identified CD56bright NK cells as potent anti-tumor effectors that warrant further investigation as a cancer immunotherapy.

Along with activating NK cells, IL-15 is critical for NK cell development and homeostasis. One result of IL-15 signaling is increased expression of Eomesodermin (Eomes), a transcription factor that defines the NK cell vs. ILC1 lineages. Eomes is required for early NK cell development, but its role in mature NK cell homeostasis and function is unknown. To study this, we utilized a novel NK cell-specific inducible Cre mouse model. An inducible model was required since excision of Eomes early in NK cell development precludes normal NK cell generation. We demonstrated that Eomes expression in mature NK cells is required for their homeostasis and function since inducible Eomes deletion resulted in a significant decrease in NK cells across the NK compartment, with a particularly profound effect on the penultimately-mature Stage III population. Stage III NK cells demonstrated increased apoptosis upon Eomes deletion as well as an impaired ability to differentiate from precursor Stage II cells. Inducible Eomes deletion also negatively impacted NK cell function, impairing NK cell anti-tumor cytotoxicity ex vivo and abrogating rejection of 2m-/- splenocytes in vivo. However, in vitro functional responses to a variety of stimuli were largely preserved. These data reveal a previously unappreciated role for Eomes in mature NK cell homeostasis and function, and suggest that an alternate means of classifying NK cell developmental stage may be based on Eomes dependence. In sum, this thesis highlights two potential mechanisms for improving NK cell-based immunotherapy and utilizes a novel NK cell-specific mouse model to demonstrate a previously unappreciated role for Eomes in mature NK cell function and homeostasis.


English (en)

Chair and Committee

Todd A. Fehniger

Committee Members

Megan A. Cooper, Timothy J. Ley, Daniel C. Link, Wayne M. Yokoyama,