bioRxiv2023
NK cells in the peripheral blood of severe COVID-19 patients exhibit a unique profile characterized by activation and dysfunction. Previous studies have identified soluble factors, including type I interferon and TGF{beta}, that underlie this dysregulation. However, the role of cell-cell interactions in mediating changes in NK cell function during COVID-19 remains unclear. To address this question, we combined cell-cell communication analysis on existing single-cell RNA sequencing data with in vitro primary cell co-culture experiments to dissect the mechanisms underlying NK cell dysfunction in COVID-19. We found that NK cells are predicted to interact most strongly with monocytes and that this occurs via both soluble factors and direct interactions. To validate these findings, we performed in vitro co-cultures in which NK cells from healthy donors were incubated with monocytes from COVID-19+ or healthy donors. Co-culture of healthy NK cells with monocytes from COVID-19 patients recapitulated aspects of the NK cell phenotype observed in severe COVID-19, including decreased expression of NKG2D, increased expression of activation markers, and increased proliferation. When these experiments were performed in a transwell setting, we found that only CD56bright CD16- NK cells were activated in the presence of severe COVID-19 patient monocytes. O-link analysis of supernatants from transwell co-cultures revealed that cultures containing severe COVID-19 patient monocytes had significantly elevated levels of proinflammatory cytokines and chemokines as well as TGF{beta}. Collectively, these results demonstrate that interactions between NK cells and monocytes in the peripheral blood of COVID-19 patients contribute to NK cell activation and dysfunction in severe COVID-19. BACKGROUNDNatural killer (NK) cells are innate lymphocytes that are critical antiviral effectors. Because of their role in controlling acute viral infections, multiple studies have evaluated the role of NK cells in SARS-CoV-2 infection. Such studies revealed that NK cell phenotype and function are significantly altered by severe COVID-19; the peripheral NK cells of severe COVID-19 patients are highly activated and proliferative(1-5), with increased expression of cytotoxic molecules, Ki-67, and several surface markers of activation(3, 5-8). However, these NK cells also have dysfunctional cytotoxic responses to both tumor target cells(1, 2, 9, 10) and SARS-CoV-2-infected target cells(9, 10). Given that peripheral NK cells are thought to migrate to the lung during COVID-19(11-13), these results suggest that the NK cells of severe COVID-19 patients may be incapable of mounting a successful antiviral response to SARS-CoV-2 infection. Although the unique phenotype and dysfunctionality of NK cells in severe COVID-19 has been well-characterized, the processes underlying these phenomena have not. Only one study has conducted in vitro mechanistic experiments to identify a possible cause of NK cell dysfunction: Witkowski et al. identified serum-derived TGF{beta} as a suppressor of NK cell functionality in severe COVID-19 patients(9). However, this study did not identify the source of serum TGF{beta}. Additionally, given the high degree of complexity within the immune system, there are likely other causes of NK cell dysfunction in COVID-19 that have thus far remain unexplored. One such mechanism may be the myriad of interactions between NK cells and other peripheral immune cells. NK cells are known to interact with CD4 and CD8 T cells, dendritic cells, neutrophils, and macrophages/monocytes(14), which can prime NK cell cytotoxicity or induce tolerance. Previous work by our lab suggested the potential for NK cell-monocyte crosstalk in severe COVID-19 through the expression of ligands for NK cell activating receptors on the monocytes of these patients(3). Crosstalk between NK cells and monocytes plays a role in regulating the NK cell response to other infections, including HIV-1(15, 16), mouse(17) and human cytomegalovirus(18), and malaria(19) through mechanisms including secretion of NK cell-regulating cytokines by monocytes. In this study, we used a combination of computational and in vitro methods to dissect the interactions between NK cells and monocytes in severe COVID-19. We utilized primary NK cells and monocytes from a large cohort of COVID-19 patients to demonstrate that co-culture of healthy NK cells with monocytes from severe COVID-19 donors can partially recapitulate the activated phenotype observed in the NK cells from COVID-19 patients. We then interrogated the mechanisms by which this activation occurs by performing NK cell-monocyte co-cultures in a transwell setting and using O-link to analyze the cytokines present in this system. Collectively, our work identifies monocytes as a driver of NK cell activation in severe COVID-19 and reveals interactions between NK cells and monocytes that may underlie this process.